From mwatters at opencsw.org Fri May 1 05:31:57 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 30 Apr 2009 22:31:57 -0500 Subject: [csw-maintainers] mediawiki now in testing Message-ID: <49FA6D2D.9020204@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 version 1.14.0 LaTeX math extension included in mediawiki_math package - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn6bS0ACgkQLrhmsXMSLxecOQCgpGEhzVuMZNsgnxayNN4ZquMn KTgAn1TP8TBa6/5FvD0pguU115YKhXbc =Cteq -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 1 08:13:16 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 30 Apr 2009 23:13:16 -0700 Subject: [csw-maintainers] mediawiki now in testing In-Reply-To: <49FA6D2D.9020204@opencsw.org> References: <49FA6D2D.9020204@opencsw.org> Message-ID: <20090501061316.GA16206@bolthole.com> On Thu, Apr 30, 2009 at 10:31:57PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > version 1.14.0 > LaTeX math extension included in mediawiki_math package errr... that sounds like an odd way to put it. do you mean, "extension for mediawiki included, that takes advantage of LaTeX math extension"? From dam at opencsw.org Fri May 1 15:12:44 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 1 May 2009 15:12:44 +0200 Subject: [csw-maintainers] Change of upstream watch in mGAR Message-ID: Hi, in the current implementation of check-upstream FILES2CHECK are defined with := invoking bin/upstream_watch. This means the definition is made and the binary called on each invocation of make, regardless of the target. With a limited internet connection this causes massive delays on the lftp call. I changed the implementation in r4633 to only calculate the names if necessary: Best regards -- Dago From Darin.Perusich at cognigencorp.com Fri May 1 15:30:35 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Fri, 01 May 2009 09:30:35 -0400 Subject: [csw-maintainers] relocatable on sparc but not x86 In-Reply-To: <20090430213405.GI79653@bolthole.com> References: <49FA0D7E.4060103@cognigencorp.com> <20090430213405.GI79653@bolthole.com> Message-ID: <49FAF97B.8040803@cognigencorp.com> I'm not asking how to fix it, I made the package non-relocatable, I'm curious to know what the cause might be. When I run staring against the sparc build it has various paths to 'gpg' listed and /opt/csw/bin happens to be one of them. These are not listed when I use strings on the i386 binary. Philip Brown wrote: > On Thu, Apr 30, 2009 at 04:43:42PM -0400, Darin Perusich wrote: >> I'm staging a package and on SPARC it's not relocatable but on i386 it's >> not, why would this be? >> > > that's kinda equivalent to saying, "it's broken. how can I fix it?" without saying > anything else :-/ > > > at minimum, you need to find out what bits are non-relocatable, and tell us > what they are... > -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From mwatters at opencsw.org Fri May 1 16:15:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 01 May 2009 09:15:44 -0500 Subject: [csw-maintainers] mediawiki now in testing In-Reply-To: <20090501061316.GA16206@bolthole.com> References: <49FA6D2D.9020204@opencsw.org> <20090501061316.GA16206@bolthole.com> Message-ID: <49FB0410.4030601@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Thu, Apr 30, 2009 at 10:31:57PM -0500, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> version 1.14.0 >> LaTeX math extension included in mediawiki_math package > > errr... that sounds like an odd way to put it. > > > do you mean, "extension for mediawiki included, that takes advantage > of LaTeX math extension"? > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers yes, I guess that is a better way to put it. This is the first iteration of the package, so I am sure I will have a few more pop through testing before release. I am still Tweaking the install (since it is normally a manual thing) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkn7BA8ACgkQLrhmsXMSLxd5pgCgzOg2yzGx6N2fj9bdnkmLIzLM sFwAoMUE0kg823ygoBOipQZ3kq/tkgYD =QjZP -----END PGP SIGNATURE----- From william at wbonnet.net Fri May 1 16:30:57 2009 From: william at wbonnet.net (William Bonnet) Date: Fri, 01 May 2009 16:30:57 +0200 Subject: [csw-maintainers] Change of upstream watch in mGAR In-Reply-To: References: Message-ID: <49FB07A1.2050100@wbonnet.net> Thanks a lot for this fix cheers W. Dagobert Michelsen a ?crit : > Hi, > > in the current implementation of check-upstream FILES2CHECK are defined > with := invoking bin/upstream_watch. This means the definition is made > and the binary called on each invocation of make, regardless of the > target. > With a limited internet connection this causes massive delays on the > lftp call. I changed the implementation in r4633 to only calculate > the names if necessary: > > > Best regards > > -- Dago From mwatters at opencsw.org Sun May 3 15:50:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 03 May 2009 08:50:22 -0500 Subject: [csw-maintainers] Fixing build8xt isaexec packages Message-ID: <49FDA11E.3040100@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I am re-installing the following packages on build8xt: CSWaudiofile CSWbzip2 CSWfconfig CSWflac CSWgcc4ada CSWgcc4adart CSWgcc4core CSWgcc4corert CSWgcc4g++ CSWgcc4g++rt CSWgcc4gfortran CSWgcc4gfortranrt CSWgcc4java CSWgcc4javart CSWgcc4objc CSWgcc4objcrt CSWgnupg CSWgzip CSWlibntlm CSWlibpq CSWlibtasn1 CSWlsof CSWmysql5 CSWpcre CSWpcrert CSWpostgresql CSWrsync CSWsetoolkit CSWtop CSWxmlstarlet I will let you know when complete ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkn9oR4ACgkQLrhmsXMSLxcYagCeIFprYQDzW2vnJJKPGmSsES2L MYQAn3JUHP1uSdIyrAPONCn8Q21bBnqA =B0HQ -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 3 17:25:39 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 03 May 2009 10:25:39 -0500 Subject: [csw-maintainers] Fixing build8xt isaexec packages In-Reply-To: <49FDA11E.3040100@opencsw.org> References: <49FDA11E.3040100@opencsw.org> Message-ID: <49FDB773.3010709@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > I am re-installing the following packages on build8xt: > CSWaudiofile > CSWbzip2 > CSWfconfig > CSWflac > CSWgcc4ada > CSWgcc4adart > CSWgcc4core > CSWgcc4corert > CSWgcc4g++ > CSWgcc4g++rt > CSWgcc4gfortran > CSWgcc4gfortranrt > CSWgcc4java > CSWgcc4javart > CSWgcc4objc > CSWgcc4objcrt > CSWgnupg > CSWgzip > CSWlibntlm > CSWlibpq > CSWlibtasn1 > CSWlsof > CSWmysql5 > CSWpcre > CSWpcrert > CSWpostgresql > CSWrsync > CSWsetoolkit > CSWtop > CSWxmlstarlet > > I will let you know when complete ;) > > _______________________________________________ maintainers mailing list maintainers at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/maintainers All packages re-installed, no version changes on any packages. on a side note. gcc4.3.3 is WORKING CORRECTLY on build8xt. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkn9t3MACgkQLrhmsXMSLxd+ZACgzxy+PiPGn+t90iJ2TvGpLXsl Ts8AnAj0D+Kt2uIJAdCZ9ZH7CEwmnjVr =8h5N -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 3 23:25:50 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 03 May 2009 16:25:50 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing Message-ID: <49FE0BDE.1010809@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 After Weeks of Beating my head against the wall on the custom merge. with many thanks to the build-farm maintainers 'specially Ben and Dago for putting up with my huge package installs one or more times / day. I repackaged to remove la~ backup files from the manual merge. installed without error on build8xt. I have successfully compiled/tested vim, libffi, and binutils on build8xt. I will do a sanity test on sparc tomorrow. Does anyone have a solaris 10 x86 server to test amd64 compiles? ... mine became a centOS5.3 x86_64 server. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn+C94ACgkQLrhmsXMSLxea0QCcCzzCANyK9DX/vZ/49kRbFKvw GIEAn3dF0sZjsvdJSOruwSfEUcqWTu83 =SXOs -----END PGP SIGNATURE----- From bwalton at opencsw.org Mon May 4 22:15:17 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 04 May 2009 16:15:17 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <49FE0BDE.1010809@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> Message-ID: <1241468034-sup-3126@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Sun May 03 17:25:50 -0400 2009: Hi Mike, > Does anyone have a solaris 10 x86 server to test amd64 compiles? > ... mine became a centOS5.3 x86_64 server. I'm updating the gcc and g++ packages on a sol10/x86 box right now. I'll build the same packages you did and let you know. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 03:07:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 04 May 2009 20:07:44 -0500 Subject: [csw-maintainers] squid now in testing Message-ID: <49FF9160.1040108@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: * Updated to version 2.7-STABLE6 * significant changes in postinstall preremove scripts * smf support on solaris10 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn/kWAACgkQLrhmsXMSLxfHGgCgmj5xoHBNsl2PyRk98lq5Ja2Y RSkAn3R9jB/1BYTfS9mex2hto1LR8p12 =IyCD -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue May 5 05:05:32 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 04 May 2009 22:05:32 -0500 Subject: [csw-maintainers] repackage squid in testing Message-ID: <49FFACFC.3050809@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: enabled more compile time options including bug http://www.opencsw.org/mantis/view.php?id=2858 full list of enabled options: - --enable-arp-acl - --enable-auth=basic - --enable-basic-auth-helpers=LDAP,SMB,YP,PAM,SASL,NCSA - --enable-cache-digests - --enable-carp - --enable-delay-pools - --enable-devpoll - --enable-forw-via-db - --enable-htcp - --enable-icmp - --enable-large-cache-files - --enable-referer-log - --enable-removal-policies=heap,lru #address bug 2858 - --enable-select - --enable-snmp - --enable-ssl - --enable-useragent-log - --with-dl - --with-large-files - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn/rPsACgkQLrhmsXMSLxfHxQCfVDFnTGWfGIeHfMuWcJpvXv5z DN0AnjgoEevQNgfolbL8Zk/1GddugHeP =QfT7 -----END PGP SIGNATURE----- From dam at opencsw.org Tue May 5 15:38:06 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 5 May 2009 15:38:06 +0200 Subject: [csw-maintainers] I am running in circles around GNOME Message-ID: <07EC02B3-2504-4902-BD97-89E4B9273C62@opencsw.org> Hi, after the great preparation of William and compiling loads of packages I hit a roadblock here where I need some advice: - pango needs an update cairo - cairo needs an updated pixman - pixman needs an updated pangocairo from pango Is there some magic to break the circle? Best regards -- Dago libtool: link: mv -f ".libs/libpangocairo-1.0.expT" ".libs/ libpangocairo-1.0.exp" libtool: link: echo "{ global:" > .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: cat .libs/libpangocairo-1.0.exp | /opt/csw/bin/gsed -e "s/\(.*\)/\1;/" >> .libs/libpangocairo-1.0.so.0.2400.2.exp libtool: link: echo "local: *; };" >> .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -M .libs/ libpangocairo-1.0.so.0.2400.2.exp -h libpangocairo-1.0.so.0 -o .libs/ libpangocairo-1.0.so.0.2400.2 .libs/libpangocairo_1_0_la-pangocairo- context.o .libs/libpangocairo_1_0_la-pangocairo-font.o .libs/ libpangocairo_1_0_la-pangocairo-fontmap.o .libs/libpangocairo_1_0_la- pangocairo-render.o .libs/libpangocairo_1_0_la-pangocairo- fcfont.o .libs/libpangocairo_1_0_la-pangocairo-fcfontmap.o -R/home/ dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/ pango/.libs -R/opt/csw/lib/64 -L/home/dam/mgar/pkg/pango/trunk/work/ build-isa-sparcv9/pango-1.24.2/pango/.libs -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 ./.libs/libpango-1.0.so -L/opt/csw/lib -lcairo ./.libs/ libpangoft2-1.0.so /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so -lgobject-2.0 - lgmodule-2.0 -lglib-2.0 -lintl -lm -lfreetype -lfontconfig -lc - xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/libcairo.so: wrong ELF class: ELFCLASS32 ld: warning: file /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so: linked to ./.libs/ libpango-1.0.so: attempted multiple inclusion of file ld: fatal: File processing errors. No output written to .libs/ libpangocairo-1.0.so.0.2400.2 gmake[6]: *** [libpangocairo-1.0.la] Error 1 gmake[6]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[5]: *** [all-recursive] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[1]: *** [build-work/build-isa-sparcv9/pango-1.24.2/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pango/trunk' gmake: *** [merge-isa-sparcv9] Error 2 /bin/bash ../libtool --tag=CC --mode=link /opt/studio/SOS11/SUNWspro/ bin/cc -xO3 -xarch=v9 -I/opt/csw/X11/include -I/opt/csw/include - D_POSIX_PTHREAD_SEMANTICS -version-info 10802:6:10800 -no-undefined - xarch=v9 -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 -o libcairo.la - rpath /opt/csw/lib/64 cairo- analysis-surface.lo cairo-arc.lo cairo-array.lo cairo-atomic.lo cairo- base85-stream.lo cairo-bentley-ottmann.lo cairo.lo cairo-cache.lo cairo-clip.lo cairo-color.lo cairo-debug.lo cairo-fixed.lo cairo-font- face.lo cairo-font-face-twin.lo cairo-font-face-twin-data.lo cairo- font-options.lo cairo-freelist.lo cairo-gstate.lo cairo-hash.lo cairo- hull.lo cairo-image-surface.lo cairo-lzw.lo cairo-matrix.lo cairo-meta- surface.lo cairo-misc.lo cairo-mutex.lo cairo-output-stream.lo cairo- paginated-surface.lo cairo-path-bounds.lo cairo-path.lo cairo-path- fill.lo cairo-path-fixed.lo cairo-path-stroke.lo cairo-pattern.lo cairo-pen.lo cairo-polygon.lo cairo-rectangle.lo cairo-region.lo cairo- scaled-font.lo cairo-skiplist.lo cairo-slope.lo cairo-spline.lo cairo- stroke-style.lo cairo-surface.lo cairo-surface-fallback.lo cairo- system.lo cairo-traps.lo cairo-unicode.lo cairo-user-font.lo cairo- version.lo cairo-wideint.lo cairo-cff-subset.lo cairo-scaled-font- subsets.lo cairo-truetype-subset.lo cairo-type1-fallback.lo cairo- type1-subset.lo cairo-type3-glyph-surface.lo cairo-pdf-operators.lo cairo-xlib-display.lo cairo-xlib-screen.lo cairo-xlib-surface.lo cairo- xlib-visual.lo cairo-png.lo cairo-ft-font.lo cairo-ps- surface.lo cairo-pdf-surface.lo cairo-deflate-stream.lo cairo-svg- surface.lo test-fallback-surface.lo test-meta-surface.lo test- paginated-surface.lo -L/opt/csw/lib -lpixman-1 -L/opt/csw/lib - lfreetype -L/opt/csw/lib -lfontconfig -L/opt/csw/lib -lpng12 - lm -lz -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 - lsocket -lnsl -lm -lm libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -z defs -h libcairo.so.2 -o .libs/libcairo.so.2.10800.6 .libs/cairo-analysis- surface.o .libs/cairo-arc.o .libs/cairo-array.o .libs/cairo- atomic.o .libs/cairo-base85-stream.o .libs/cairo-bentley- ottmann.o .libs/cairo.o .libs/cairo-cache.o .libs/cairo-clip.o .libs/ cairo-color.o .libs/cairo-debug.o .libs/cairo-fixed.o .libs/cairo-font- face.o .libs/cairo-font-face-twin.o .libs/cairo-font-face-twin- data.o .libs/cairo-font-options.o .libs/cairo-freelist.o .libs/cairo- gstate.o .libs/cairo-hash.o .libs/cairo-hull.o .libs/cairo-image- surface.o .libs/cairo-lzw.o .libs/cairo-matrix.o .libs/cairo-meta- surface.o .libs/cairo-misc.o .libs/cairo-mutex.o .libs/cairo-output- stream.o .libs/cairo-paginated-surface.o .libs/cairo-path- bounds.o .libs/cairo-path.o .libs/cairo-path-fill.o .libs/cairo-path- fixed.o .libs/cairo-path-stroke.o .libs/cairo-pattern.o .libs/cairo- pen.o .libs/cairo-polygon.o .libs/cairo-rectangle.o .libs/cairo- region.o .libs/cairo-scaled-font.o .libs/cairo-skiplist.o .libs/cairo- slope.o .libs/cairo-spline.o .libs/cairo-stroke-style.o .libs/cairo- surface.o .libs/cairo-surface-fallback.o .libs/cairo-system.o .libs/ cairo-traps.o .libs/cairo-unicode.o .libs/cairo-user-font.o .libs/ cairo-version.o .libs/cairo-wideint.o .libs/cairo-cff-subset.o .libs/ cairo-scaled-font-subsets.o .libs/cairo-truetype-subset.o .libs/cairo- type1-fallback.o .libs/cairo-type1-subset.o .libs/cairo-type3-glyph- surface.o .libs/cairo-pdf-operators.o .libs/cairo-xlib-display.o .libs/ cairo-xlib-screen.o .libs/cairo-xlib-surface.o .libs/cairo-xlib- visual.o .libs/cairo-png.o .libs/cairo-ft-font.o .libs/cairo-ps- surface.o .libs/cairo-pdf-surface.o .libs/cairo-deflate-stream.o .libs/ cairo-svg-surface.o .libs/test-fallback-surface.o .libs/test-meta- surface.o .libs/test-paginated-surface.o -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 -L/opt/csw/lib -lpixman-1 -lfreetype -lfontconfig - lpng12 -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 -lsocket - lnsl -lm -lc -xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/64/libpixman-1.so: wrong ELF class: ELFCLASS32 ld: fatal: File processing errors. No output written to .libs/ libcairo.so.2.10800.6 gmake[5]: *** [libcairo.la] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[1]: *** [build-work/build-isa-sparcv9/cairo-1.8.6/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk' gmake: *** [merge-isa-sparcv9] Error 2 checking dynamic linker characteristics... solaris2.8 ld.so (cached) (cached) checking how to hardcode library paths into programs... immediate checking for getisax... no checking whether byte ordering is bigendian... yes checking for inline... inline checking whether __SUNPRO_C is declared... yes checking whether __amd64 is declared... no checking for perl... /opt/csw/bin/perl checking for -fvisibility... no checking for -xldscope (Sun compilers)... yes checking whether to use MMX intrinsics... no checking whether to use SSE2 intrinsics... no checking whether to use VMX/Altivec intrinsics... no checking whether to use ARM SIMD assembler... no checking for pkg-config... /opt/csw/bin/pkg-config checking pkg-config is at least version 0.9.0... yes checking for GTK... configure: error: Package requirements (gtk+-2.0) were not met: sh: gnome-config: not found Package pangocairo was not found in the pkg-config search path. Perhaps you should add the directory containing `pangocairo.pc' to the PKG_CONFIG_PATH environment variable Package 'pangocairo', required by 'GDK', not found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables GTK_CFLAGS and GTK_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. gmake[1]: *** [configure-work/build-isa-sparcv8/pixman-0.15.2/ configure] Error 1 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pixman/trunk' gmake: *** [merge-isa-sparcv8] Error 2 From dam at opencsw.org Tue May 5 15:45:27 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 5 May 2009 15:45:27 +0200 Subject: [csw-maintainers] I am running in circles around GNOME Message-ID: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> Hi, after the great preparation of William and compiling loads of packages I hit a roadblock here where I need some advice: - pango needs an update cairo - cairo needs an updated pixman - pixman needs an updated pangocairo from pango Is there some magic to break the circle? Best regards -- Dago libtool: link: mv -f ".libs/libpangocairo-1.0.expT" ".libs/ libpangocairo-1.0.exp" libtool: link: echo "{ global:" > .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: cat .libs/libpangocairo-1.0.exp | /opt/csw/bin/gsed -e "s/\(.*\)/\1;/" >> .libs/libpangocairo-1.0.so.0.2400.2.exp libtool: link: echo "local: *; };" >> .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -M .libs/ libpangocairo-1.0.so.0.2400.2.exp -h libpangocairo-1.0.so.0 -o .libs/ libpangocairo-1.0.so.0.2400.2 .libs/libpangocairo_1_0_la-pangocairo- context.o .libs/libpangocairo_1_0_la-pangocairo-font.o .libs/ libpangocairo_1_0_la-pangocairo-fontmap.o .libs/libpangocairo_1_0_la- pangocairo-render.o .libs/libpangocairo_1_0_la-pangocairo- fcfont.o .libs/libpangocairo_1_0_la-pangocairo-fcfontmap.o -R/home/ dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/ pango/.libs -R/opt/csw/lib/64 -L/home/dam/mgar/pkg/pango/trunk/work/ build-isa-sparcv9/pango-1.24.2/pango/.libs -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 ./.libs/libpango-1.0.so -L/opt/csw/lib -lcairo ./.libs/ libpangoft2-1.0.so /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so -lgobject-2.0 - lgmodule-2.0 -lglib-2.0 -lintl -lm -lfreetype -lfontconfig -lc - xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/libcairo.so: wrong ELF class: ELFCLASS32 ld: warning: file /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so: linked to ./.libs/ libpango-1.0.so: attempted multiple inclusion of file ld: fatal: File processing errors. No output written to .libs/ libpangocairo-1.0.so.0.2400.2 gmake[6]: *** [libpangocairo-1.0.la] Error 1 gmake[6]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[5]: *** [all-recursive] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[1]: *** [build-work/build-isa-sparcv9/pango-1.24.2/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pango/trunk' gmake: *** [merge-isa-sparcv9] Error 2 /bin/bash ../libtool --tag=CC --mode=link /opt/studio/SOS11/SUNWspro/ bin/cc -xO3 -xarch=v9 -I/opt/csw/X11/include -I/opt/csw/include - D_POSIX_PTHREAD_SEMANTICS -version-info 10802:6:10800 -no-undefined - xarch=v9 -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 -o libcairo.la - rpath /opt/csw/lib/64 cairo- analysis-surface.lo cairo-arc.lo cairo-array.lo cairo-atomic.lo cairo- base85-stream.lo cairo-bentley-ottmann.lo cairo.lo cairo-cache.lo cairo-clip.lo cairo-color.lo cairo-debug.lo cairo-fixed.lo cairo-font- face.lo cairo-font-face-twin.lo cairo-font-face-twin-data.lo cairo- font-options.lo cairo-freelist.lo cairo-gstate.lo cairo-hash.lo cairo- hull.lo cairo-image-surface.lo cairo-lzw.lo cairo-matrix.lo cairo-meta- surface.lo cairo-misc.lo cairo-mutex.lo cairo-output-stream.lo cairo- paginated-surface.lo cairo-path-bounds.lo cairo-path.lo cairo-path- fill.lo cairo-path-fixed.lo cairo-path-stroke.lo cairo-pattern.lo cairo-pen.lo cairo-polygon.lo cairo-rectangle.lo cairo-region.lo cairo- scaled-font.lo cairo-skiplist.lo cairo-slope.lo cairo-spline.lo cairo- stroke-style.lo cairo-surface.lo cairo-surface-fallback.lo cairo- system.lo cairo-traps.lo cairo-unicode.lo cairo-user-font.lo cairo- version.lo cairo-wideint.lo cairo-cff-subset.lo cairo-scaled-font- subsets.lo cairo-truetype-subset.lo cairo-type1-fallback.lo cairo- type1-subset.lo cairo-type3-glyph-surface.lo cairo-pdf-operators.lo cairo-xlib-display.lo cairo-xlib-screen.lo cairo-xlib-surface.lo cairo- xlib-visual.lo cairo-png.lo cairo-ft-font.lo cairo-ps- surface.lo cairo-pdf-surface.lo cairo-deflate-stream.lo cairo-svg- surface.lo test-fallback-surface.lo test-meta-surface.lo test- paginated-surface.lo -L/opt/csw/lib -lpixman-1 -L/opt/csw/lib - lfreetype -L/opt/csw/lib -lfontconfig -L/opt/csw/lib -lpng12 - lm -lz -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 - lsocket -lnsl -lm -lm libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -z defs -h libcairo.so.2 -o .libs/libcairo.so.2.10800.6 .libs/cairo-analysis- surface.o .libs/cairo-arc.o .libs/cairo-array.o .libs/cairo- atomic.o .libs/cairo-base85-stream.o .libs/cairo-bentley- ottmann.o .libs/cairo.o .libs/cairo-cache.o .libs/cairo-clip.o .libs/ cairo-color.o .libs/cairo-debug.o .libs/cairo-fixed.o .libs/cairo-font- face.o .libs/cairo-font-face-twin.o .libs/cairo-font-face-twin- data.o .libs/cairo-font-options.o .libs/cairo-freelist.o .libs/cairo- gstate.o .libs/cairo-hash.o .libs/cairo-hull.o .libs/cairo-image- surface.o .libs/cairo-lzw.o .libs/cairo-matrix.o .libs/cairo-meta- surface.o .libs/cairo-misc.o .libs/cairo-mutex.o .libs/cairo-output- stream.o .libs/cairo-paginated-surface.o .libs/cairo-path- bounds.o .libs/cairo-path.o .libs/cairo-path-fill.o .libs/cairo-path- fixed.o .libs/cairo-path-stroke.o .libs/cairo-pattern.o .libs/cairo- pen.o .libs/cairo-polygon.o .libs/cairo-rectangle.o .libs/cairo- region.o .libs/cairo-scaled-font.o .libs/cairo-skiplist.o .libs/cairo- slope.o .libs/cairo-spline.o .libs/cairo-stroke-style.o .libs/cairo- surface.o .libs/cairo-surface-fallback.o .libs/cairo-system.o .libs/ cairo-traps.o .libs/cairo-unicode.o .libs/cairo-user-font.o .libs/ cairo-version.o .libs/cairo-wideint.o .libs/cairo-cff-subset.o .libs/ cairo-scaled-font-subsets.o .libs/cairo-truetype-subset.o .libs/cairo- type1-fallback.o .libs/cairo-type1-subset.o .libs/cairo-type3-glyph- surface.o .libs/cairo-pdf-operators.o .libs/cairo-xlib-display.o .libs/ cairo-xlib-screen.o .libs/cairo-xlib-surface.o .libs/cairo-xlib- visual.o .libs/cairo-png.o .libs/cairo-ft-font.o .libs/cairo-ps- surface.o .libs/cairo-pdf-surface.o .libs/cairo-deflate-stream.o .libs/ cairo-svg-surface.o .libs/test-fallback-surface.o .libs/test-meta- surface.o .libs/test-paginated-surface.o -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 -L/opt/csw/lib -lpixman-1 -lfreetype -lfontconfig - lpng12 -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 -lsocket - lnsl -lm -lc -xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/64/libpixman-1.so: wrong ELF class: ELFCLASS32 ld: fatal: File processing errors. No output written to .libs/ libcairo.so.2.10800.6 gmake[5]: *** [libcairo.la] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[1]: *** [build-work/build-isa-sparcv9/cairo-1.8.6/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk' gmake: *** [merge-isa-sparcv9] Error 2 checking dynamic linker characteristics... solaris2.8 ld.so (cached) (cached) checking how to hardcode library paths into programs... immediate checking for getisax... no checking whether byte ordering is bigendian... yes checking for inline... inline checking whether __SUNPRO_C is declared... yes checking whether __amd64 is declared... no checking for perl... /opt/csw/bin/perl checking for -fvisibility... no checking for -xldscope (Sun compilers)... yes checking whether to use MMX intrinsics... no checking whether to use SSE2 intrinsics... no checking whether to use VMX/Altivec intrinsics... no checking whether to use ARM SIMD assembler... no checking for pkg-config... /opt/csw/bin/pkg-config checking pkg-config is at least version 0.9.0... yes checking for GTK... configure: error: Package requirements (gtk+-2.0) were not met: sh: gnome-config: not found Package pangocairo was not found in the pkg-config search path. Perhaps you should add the directory containing `pangocairo.pc' to the PKG_CONFIG_PATH environment variable Package 'pangocairo', required by 'GDK', not found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables GTK_CFLAGS and GTK_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. gmake[1]: *** [configure-work/build-isa-sparcv8/pixman-0.15.2/ configure] Error 1 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pixman/trunk' gmake: *** [merge-isa-sparcv8] Error 2 From bonivart at opencsw.org Tue May 5 18:00:10 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 5 May 2009 18:00:10 +0200 Subject: [csw-maintainers] CSWpymysql listed twice in catalog Message-ID: <625385e30905050900o4b32147fx97a258e5766093@mail.gmail.com> pymysql 1.2.2,REV=2009.04.02 CSWpymysql pymysql-1.2.2,REV=2009.04.02-SunOS5.8-sparc-CSW.pkg.gz 3e46259ed56d5986008192c4a9326436 61565 CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none pymysql5 1.2.2,REV=2009.03.11 CSWpymysql pymysql5-1.2.2,REV=2009.03.11-SunOS5.8-sparc-CSW.pkg.gz f9d234764be352370ac1e83a2edaeb88 64970 CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none I assume the second one should be removed. -- /peter From phil at bolthole.com Tue May 5 18:24:54 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 5 May 2009 09:24:54 -0700 Subject: [csw-maintainers] CSWpymysql listed twice in catalog In-Reply-To: <625385e30905050900o4b32147fx97a258e5766093@mail.gmail.com> References: <625385e30905050900o4b32147fx97a258e5766093@mail.gmail.com> Message-ID: <20090505162454.GA36027@bolthole.com> On Tue, May 05, 2009 at 06:00:10PM +0200, Peter Bonivart wrote: > pymysql 1.2.2,REV=2009.04.02 CSWpymysql > pymysql-1.2.2,REV=2009.04.02-SunOS5.8-sparc-CSW.pkg.gz > 3e46259ed56d5986008192c4a9326436 61565 > CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none > > pymysql5 1.2.2,REV=2009.03.11 CSWpymysql > pymysql5-1.2.2,REV=2009.03.11-SunOS5.8-sparc-CSW.pkg.gz > f9d234764be352370ac1e83a2edaeb88 64970 > CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none > > I assume the second one should be removed. thanks for noticing that. kept pymysql. deleted pymysql5 From bwalton at opencsw.org Tue May 5 19:49:22 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 13:49:22 -0400 Subject: [csw-maintainers] php5 packaging Message-ID: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Hi All, I updated a box that uses php5 today, so I got the new versions that were just released. After bringing the webserver back online, the hosted app wasn't responding with anything but the logs showed the requested being handled. Turns out that since many of the modules that used to be part of the core php5 package (session being the one that got me) are now separate packages, I simply needed to pull them in and I was back up and running. I was just wondering if there is a better way to handle situations like this...a user doing a 1:1 update of packages will have less functionality after the update than they did before. I think the changes are ok overall, but the user interaction may not be. Thoughts? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Tue May 5 20:09:07 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 05 May 2009 20:09:07 +0200 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241545460-sup-965@ntdws12.chass.utoronto.ca> (Ben Walton's message of "Tue\, 05 May 2009 13\:49\:22 -0400") References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: Ben Walton writes: > I was just wondering if there is a better way to handle situations > like this...a user doing a 1:1 update of packages will have less > functionality after the update than they did before. I think the > changes are ok overall, but the user interaction may not be. Using transition packages? A transition package serves as a temporary package, offering 1:1 content, having the same name as the one transitioned. Of course, after the transition, the package disappears or has a lesser surface. -- Peter From bwalton at opencsw.org Tue May 5 20:17:48 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 14:17:48 -0400 Subject: [csw-maintainers] php5 packaging In-Reply-To: References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: <1241547276-sup-5004@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Tue May 05 14:09:07 -0400 2009: > Using transition packages? A transition package serves as a temporary > package, offering 1:1 content, having the same name as the one > transitioned. Of course, after the transition, the package disappears or > has a lesser surface. So have an empty package named php5 that depends on php5core, php5session, etc? Then stop updating php5 at some point in the future? That works, but leaves the dangling (although harmless) php5 empty package. Is there a way to handle it other than this? I like the extra granularity that Mike introduced, but if users break their setups (or waste time sorting it out, anyway), they may not be very happy. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 20:19:01 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 13:19:01 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241545460-sup-965@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: <4A008315.70203@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Hi All, > > I updated a box that uses php5 today, so I got the new versions that > were just released. After bringing the webserver back online, the > hosted app wasn't responding with anything but the logs showed the > requested being handled. > > Turns out that since many of the modules that used to be part of the > core php5 package (session being the one that got me) are now separate > packages, I simply needed to pull them in and I was back up and > running. > > I was just wondering if there is a better way to handle situations > like this...a user doing a 1:1 update of packages will have less > functionality after the update than they did before. I think the > changes are ok overall, but the user interaction may not be. > > Thoughts? > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I didn't think about that part of the upgrade. I wrote admin scripts to gracefully handle the php.ini files. The new packages do not overwrite the existing php.ini. There are a couple ways to handle this I think. examine the existing php.ini file for enabled extensions: 1) kick off a pkg-get -i for each extension that is enabled. 2) display a note to the user to add the extra packages themselves. This does not account for anything that was previously compiled statically I am not sure if there were any. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAgxUACgkQLrhmsXMSLxdrKwCeLejN1qO1q2DjHl900VI1tD66 FgEAnR5LcHsyslNKgzfy8U85Qq7GK8m3 =8HRD -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue May 5 20:30:06 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 13:30:06 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <4A008315.70203@opencsw.org> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> Message-ID: <4A0085AE.2080807@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > Ben Walton wrote: >> Hi All, > >> I updated a box that uses php5 today, so I got the new versions that >> were just released. After bringing the webserver back online, the >> hosted app wasn't responding with anything but the logs showed the >> requested being handled. > >> Turns out that since many of the modules that used to be part of the >> core php5 package (session being the one that got me) are now separate >> packages, I simply needed to pull them in and I was back up and >> running. > >> I was just wondering if there is a better way to handle situations >> like this...a user doing a 1:1 update of packages will have less >> functionality after the update than they did before. I think the >> changes are ok overall, but the user interaction may not be. > >> Thoughts? >> -Ben > > >> ------------------------------------------------------------------------ > >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > I didn't think about that part of the upgrade. > I wrote admin scripts to gracefully handle the php.ini files. > The new packages do not overwrite the existing php.ini. > > There are a couple ways to handle this I think. > > examine the existing php.ini file for enabled extensions: > > 1) kick off a pkg-get -i for each extension that is enabled. > 2) display a note to the user to add the extra packages themselves. > > This does not account for anything that was previously compiled > statically I am not sure if there were any. > I will start by definitely changing the existing php.ini warning to let the user's know of the new granularity. does anyone have a preference on the manual vs automatic? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAha4ACgkQLrhmsXMSLxdH+ACfSqaYjyCIloMHjjefSQ3Vli91 kfkAoJ1mt9aqkU4/pZ+B4wY54ytNTK/I =BdcX -----END PGP SIGNATURE----- From bwalton at opencsw.org Tue May 5 20:52:45 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 14:52:45 -0400 Subject: [csw-maintainers] php5 packaging In-Reply-To: <4A008315.70203@opencsw.org> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> Message-ID: <1241549280-sup-7549@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Tue May 05 14:19:01 -0400 2009: > I didn't think about that part of the upgrade. > I wrote admin scripts to gracefully handle the php.ini files. > The new packages do not overwrite the existing php.ini. > > There are a couple ways to handle this I think. > > examine the existing php.ini file for enabled extensions: > > 1) kick off a pkg-get -i for each extension that is enabled. This gets troublesome in the face of locally added extensions (we use oracle here, for example)....not impossible, but you'd maybe need to whitelist extensions against those available from the catalog or something. > 2) display a note to the user to add the extra packages themselves. What about a combination of these two? Provide a php5-update-extensions script or something that does 1 and note it at package install time? This could still get lost in a sea of output though, so it might not be sufficient. > This does not account for anything that was previously compiled > statically I am not sure if there were any. I'm not sure either. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 21:13:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 14:13:24 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241549280-sup-7549@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> <1241549280-sup-7549@ntdws12.chass.utoronto.ca> Message-ID: <4A008FD4.9040606@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Mike Watters's message of Tue May 05 14:19:01 -0400 2009: >> I didn't think about that part of the upgrade. >> I wrote admin scripts to gracefully handle the php.ini files. >> The new packages do not overwrite the existing php.ini. >> >> There are a couple ways to handle this I think. >> >> examine the existing php.ini file for enabled extensions: >> >> 1) kick off a pkg-get -i for each extension that is enabled. > > This gets troublesome in the face of locally added extensions (we use > oracle here, for example)....not impossible, but you'd maybe need to > whitelist extensions against those available from the catalog or > something. > >> 2) display a note to the user to add the extra packages themselves. > > What about a combination of these two? Provide a > php5-update-extensions script or something that does 1 and note it at > package install time? This could still get lost in a sea of output > though, so it might not be sufficient. > >> This does not account for anything that was previously compiled >> statically I am not sure if there were any. > > I'm not sure either. > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Here is what I have come up with thus far, Please review and let me know what you think. http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/php5/trunk/files/CSWphp5.postinstall?revision=4695&view=markup - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAj9QACgkQLrhmsXMSLxdoDACg5jWyG0LnnLl2F45aFDdaXYBb coAAn2oaZEy/8dKnc9OGAyxVP59w6Tzs =4IPo -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue May 5 21:15:09 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 14:15:09 -0500 Subject: [csw-maintainers] squidclamav 4.0 now in testing Message-ID: <4A00903D.4090900@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: new package Once I get done tweaking the post-install scripts for CSWsquid, I will start Tweaking them for CSWsquidclamav. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAkD0ACgkQLrhmsXMSLxcErACfYxewviGLZmE8LtoOjNhYNj+h JgEAn3gqs8W8fWTCNzYZxpLdL0PnjY89 =8ELu -----END PGP SIGNATURE----- From phil at bolthole.com Tue May 5 21:44:17 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 5 May 2009 12:44:17 -0700 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241545460-sup-965@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: <20090505194417.GE36027@bolthole.com> On Tue, May 05, 2009 at 01:49:22PM -0400, Ben Walton wrote: > I was just wondering if there is a better way to handle situations > like this...a user doing a 1:1 update of packages will have less > functionality after the update than they did before. I think the > changes are ok overall, but the user interaction may not be. > > Thoughts? > -Ben Reguardless of what we decide here: Mike, you need to write (separately sent) emails to both the users, and the announce, list, warning people about this right away. From bwalton at opencsw.org Tue May 5 21:49:17 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 15:49:17 -0400 Subject: [csw-maintainers] php5 packaging In-Reply-To: <4A008FD4.9040606@opencsw.org> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> <1241549280-sup-7549@ntdws12.chass.utoronto.ca> <4A008FD4.9040606@opencsw.org> Message-ID: <1241552758-sup-6316@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Tue May 05 15:13:24 -0400 2009: > Here is what I have come up with thus far, Please review and let me > know what you think. > http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/php5/trunk/files/CSWphp5.postinstall?revision=4695&view=markup Looks good. Still possible to miss in the output of a large update, but hopefully it will suffice. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 21:51:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 14:51:03 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241552758-sup-6316@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> <1241549280-sup-7549@ntdws12.chass.utoronto.ca> <4A008FD4.9040606@opencsw.org> <1241552758-sup-6316@ntdws12.chass.utoronto.ca> Message-ID: <4A0098A7.8030300@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Mike Watters's message of Tue May 05 15:13:24 -0400 2009: > >> Here is what I have come up with thus far, Please review and let me >> know what you think. >> http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/php5/trunk/files/CSWphp5.postinstall?revision=4695&view=markup > > Looks good. Still possible to miss in the output of a large update, > but hopefully it will suffice. > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers OK, I am writing the emails to users and announce, I will package that change and get it out. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAmKYACgkQLrhmsXMSLxeR+ACgpZqA8ikDExeTNmR7fE5MkNQy OS0AoLVp544Cqpd8nAl8kQFdMsZ0oPNK =wZkL -----END PGP SIGNATURE----- From harpchad at opencsw.org Tue May 5 22:37:41 2009 From: harpchad at opencsw.org (Chad Harp) Date: Tue, 05 May 2009 15:37:41 -0500 Subject: [csw-maintainers] I am running in circles around GNOME In-Reply-To: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> References: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> Message-ID: <4A00A395.1030500@opencsw.org> I believe from the searching I did (few months ago) that the pango dependency (via gtk) in pixman is just for some of the make check tests. Try building pixman with --disable-gtk Dagobert Michelsen wrote: > Hi, > > after the great preparation of William and compiling loads > of packages I hit a roadblock here where I need some advice: > > - pango needs an update cairo > - cairo needs an updated pixman > - pixman needs an updated pangocairo from pango > > Is there some magic to break the circle? > > > Best regards > > -- Dago > > > > libtool: link: mv -f ".libs/libpangocairo-1.0.expT" > ".libs/libpangocairo-1.0.exp" > libtool: link: echo "{ global:" > .libs/libpangocairo-1.0.so.0.2400.2.exp > libtool: link: cat .libs/libpangocairo-1.0.exp | /opt/csw/bin/gsed -e > "s/\(.*\)/\1;/" >> .libs/libpangocairo-1.0.so.0.2400.2.exp > libtool: link: echo "local: *; };" >> > .libs/libpangocairo-1.0.so.0.2400.2.exp > libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -M > .libs/libpangocairo-1.0.so.0.2400.2.exp -h libpangocairo-1.0.so.0 -o > .libs/libpangocairo-1.0.so.0.2400.2 > .libs/libpangocairo_1_0_la-pangocairo-context.o > .libs/libpangocairo_1_0_la-pangocairo-font.o > .libs/libpangocairo_1_0_la-pangocairo-fontmap.o > .libs/libpangocairo_1_0_la-pangocairo-render.o > .libs/libpangocairo_1_0_la-pangocairo-fcfont.o > .libs/libpangocairo_1_0_la-pangocairo-fcfontmap.o > -R/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs > -R/opt/csw/lib/64 > -L/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs > -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 ./.libs/libpango-1.0.so > -L/opt/csw/lib -lcairo ./.libs/libpangoft2-1.0.so > /home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so > -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lm -lfreetype > -lfontconfig -lc -xarch=v9 -xarch=v9 > ld: fatal: file /opt/csw/lib/libcairo.so: wrong ELF class: ELFCLASS32 > ld: warning: file > /home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so: > linked to ./.libs/libpango-1.0.so: attempted multiple inclusion of file > ld: fatal: File processing errors. No output written to > .libs/libpangocairo-1.0.so.0.2400.2 > gmake[6]: *** [libpangocairo-1.0.la] Error 1 > gmake[6]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango' > gmake[5]: *** [all-recursive] Error 1 > gmake[5]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango' > gmake[4]: *** [all] Error 2 > gmake[4]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango' > gmake[3]: *** [all-recursive] Error 1 > gmake[3]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2' > gmake[2]: *** [all] Error 2 > gmake[2]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2' > gmake[1]: *** [build-work/build-isa-sparcv9/pango-1.24.2/Makefile] Error 2 > gmake[1]: Leaving directory `/home/dam/mgar/pkg/pango/trunk' > gmake: *** [merge-isa-sparcv9] Error 2 > > > > > > /bin/bash ../libtool --tag=CC --mode=link > /opt/studio/SOS11/SUNWspro/bin/cc -xO3 -xarch=v9 -I/opt/csw/X11/include > -I/opt/csw/include -D_POSIX_PTHREAD_SEMANTICS -version-info > 10802:6:10800 -no-undefined -xarch=v9 -L/opt/csw/lib/64 > -L/opt/csw/X11/lib/64 -o libcairo.la -rpath > /opt/csw/lib/64 > cairo-analysis-surface.lo cairo-arc.lo cairo-array.lo cairo-atomic.lo > cairo-base85-stream.lo cairo-bentley-ottmann.lo cairo.lo cairo-cache.lo > cairo-clip.lo cairo-color.lo cairo-debug.lo cairo-fixed.lo > cairo-font-face.lo cairo-font-face-twin.lo cairo-font-face-twin-data.lo > cairo-font-options.lo cairo-freelist.lo cairo-gstate.lo cairo-hash.lo > cairo-hull.lo cairo-image-surface.lo cairo-lzw.lo cairo-matrix.lo > cairo-meta-surface.lo cairo-misc.lo cairo-mutex.lo > cairo-output-stream.lo cairo-paginated-surface.lo cairo-path-bounds.lo > cairo-path.lo cairo-path-fill.lo cairo-path-fixed.lo > cairo-path-stroke.lo cairo-pattern.lo cairo-pen.lo cairo-polygon.lo > cairo-rectangle.lo cairo-region.lo cairo-scaled-font.lo > cairo-skiplist.lo cairo-slope.lo cairo-spline.lo cairo-stroke-style.lo > cairo-surface.lo cairo-surface-fallback.lo cairo-system.lo > cairo-traps.lo cairo-unicode.lo cairo-user-font.lo cairo-version.lo > cairo-wideint.lo cairo-cff-subset.lo cairo-scaled-font-subsets.lo > cairo-truetype-subset.lo cairo-type1-fallback.lo cairo-type1-subset.lo > cairo-type3-glyph-surface.lo cairo-pdf-operators.lo > cairo-xlib-display.lo cairo-xlib-screen.lo cairo-xlib-surface.lo > cairo-xlib-visual.lo cairo-png.lo cairo-ft-font.lo > cairo-ps-surface.lo cairo-pdf-surface.lo cairo-deflate-stream.lo > cairo-svg-surface.lo test-fallback-surface.lo test-meta-surface.lo > test-paginated-surface.lo -L/opt/csw/lib -lpixman-1 -L/opt/csw/lib > -lfreetype -L/opt/csw/lib -lfontconfig -L/opt/csw/lib -lpng12 > -lm -lz -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 > -lsocket -lnsl -lm -lm > libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -z defs -h > libcairo.so.2 -o .libs/libcairo.so.2.10800.6 > .libs/cairo-analysis-surface.o .libs/cairo-arc.o .libs/cairo-array.o > .libs/cairo-atomic.o .libs/cairo-base85-stream.o > .libs/cairo-bentley-ottmann.o .libs/cairo.o .libs/cairo-cache.o > .libs/cairo-clip.o .libs/cairo-color.o .libs/cairo-debug.o > .libs/cairo-fixed.o .libs/cairo-font-face.o .libs/cairo-font-face-twin.o > .libs/cairo-font-face-twin-data.o .libs/cairo-font-options.o > .libs/cairo-freelist.o .libs/cairo-gstate.o .libs/cairo-hash.o > .libs/cairo-hull.o .libs/cairo-image-surface.o .libs/cairo-lzw.o > .libs/cairo-matrix.o .libs/cairo-meta-surface.o .libs/cairo-misc.o > .libs/cairo-mutex.o .libs/cairo-output-stream.o > .libs/cairo-paginated-surface.o .libs/cairo-path-bounds.o > .libs/cairo-path.o .libs/cairo-path-fill.o .libs/cairo-path-fixed.o > .libs/cairo-path-stroke.o .libs/cairo-pattern.o .libs/cairo-pen.o > .libs/cairo-polygon.o .libs/cairo-rectangle.o .libs/cairo-region.o > .libs/cairo-scaled-font.o .libs/cairo-skiplist.o .libs/cairo-slope.o > .libs/cairo-spline.o .libs/cairo-stroke-style.o .libs/cairo-surface.o > .libs/cairo-surface-fallback.o .libs/cairo-system.o .libs/cairo-traps.o > .libs/cairo-unicode.o .libs/cairo-user-font.o .libs/cairo-version.o > .libs/cairo-wideint.o .libs/cairo-cff-subset.o > .libs/cairo-scaled-font-subsets.o .libs/cairo-truetype-subset.o > .libs/cairo-type1-fallback.o .libs/cairo-type1-subset.o > .libs/cairo-type3-glyph-surface.o .libs/cairo-pdf-operators.o > .libs/cairo-xlib-display.o .libs/cairo-xlib-screen.o > .libs/cairo-xlib-surface.o .libs/cairo-xlib-visual.o .libs/cairo-png.o > .libs/cairo-ft-font.o .libs/cairo-ps-surface.o .libs/cairo-pdf-surface.o > .libs/cairo-deflate-stream.o .libs/cairo-svg-surface.o > .libs/test-fallback-surface.o .libs/test-meta-surface.o > .libs/test-paginated-surface.o -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 > -L/opt/csw/lib -lpixman-1 -lfreetype -lfontconfig -lpng12 -lz -lXrender > -lSM -lICE -L/usr/openwin/lib -lX11 -lsocket -lnsl -lm -lc -xarch=v9 > -xarch=v9 > ld: fatal: file /opt/csw/lib/64/libpixman-1.so: wrong ELF class: ELFCLASS32 > ld: fatal: File processing errors. No output written to > .libs/libcairo.so.2.10800.6 > gmake[5]: *** [libcairo.la] Error 1 > gmake[5]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6/src' > gmake[4]: *** [all] Error 2 > gmake[4]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6/src' > gmake[3]: *** [all-recursive] Error 1 > gmake[3]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6' > gmake[2]: *** [all] Error 2 > gmake[2]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6' > gmake[1]: *** [build-work/build-isa-sparcv9/cairo-1.8.6/Makefile] Error 2 > gmake[1]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk' > gmake: *** [merge-isa-sparcv9] Error 2 > > > > > checking dynamic linker characteristics... solaris2.8 ld.so > (cached) (cached) checking how to hardcode library paths into > programs... immediate > checking for getisax... no > checking whether byte ordering is bigendian... yes > checking for inline... inline > checking whether __SUNPRO_C is declared... yes > checking whether __amd64 is declared... no > checking for perl... /opt/csw/bin/perl > checking for -fvisibility... no > checking for -xldscope (Sun compilers)... yes > checking whether to use MMX intrinsics... no > checking whether to use SSE2 intrinsics... no > checking whether to use VMX/Altivec intrinsics... no > checking whether to use ARM SIMD assembler... no > checking for pkg-config... /opt/csw/bin/pkg-config > checking pkg-config is at least version 0.9.0... yes > checking for GTK... configure: error: Package requirements (gtk+-2.0) > were not met: > > sh: gnome-config: not found > Package pangocairo was not found in the pkg-config search path. > Perhaps you should add the directory containing `pangocairo.pc' > to the PKG_CONFIG_PATH environment variable > Package 'pangocairo', required by 'GDK', not found > > Consider adjusting the PKG_CONFIG_PATH environment variable if you > installed software in a non-standard prefix. > > Alternatively, you may set the environment variables GTK_CFLAGS > and GTK_LIBS to avoid the need to call pkg-config. > See the pkg-config man page for more details. > > gmake[1]: *** [configure-work/build-isa-sparcv8/pixman-0.15.2/configure] > Error 1 > gmake[1]: Leaving directory `/home/dam/mgar/pkg/pixman/trunk' > gmake: *** [merge-isa-sparcv8] Error 2 > > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers From ihsan at opencsw.org Tue May 5 23:07:27 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 05 May 2009 23:07:27 +0200 Subject: [csw-maintainers] [csw-announce] changes to php5 packages In-Reply-To: <4A009BFD.4070507@opencsw.org> References: <4A009BFD.4070507@opencsw.org> Message-ID: <4A00AA8F.5090905@opencsw.org> Thanks Mike. I think it makes sense that we announce major changes on the maintainers list (just for the future). Ihsan Am 5.5.2009 22:05 Uhr, Mike Watters schrieb: > Hello, > There are a few changes to the php5 packages you need to be aware of. > > Going forward, starting with the current release of php5-5.2.9,REV=2009.04.29 > the php modules that are bundled with the php5 source code have been packaged > separately from php5. This will allow better granularity on installs. > > When you run pkg-get -i php5 you will get a "bare-bone" install containing > docs, php, pear and pecl. > > all the modules will be listed and commented out in > /opt/csw/php5/lib/php.ini.CSW on a system that does not contain an existing > php.ini file the php.ini.CSW will be copied to php.ini > > all module packages now contain admin scripts to enable and disable themselves > in the php.ini file. _______________________________________________ announce mailing list announce at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/announce -- ihsan at dogan.ch http://blog.dogan.ch/ From mwatters at opencsw.org Wed May 6 02:44:39 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 19:44:39 -0500 Subject: [csw-maintainers] fixme script renamed to fixlibtool Message-ID: <4A00DD77.6060501@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I looked at the existing fixlibtool script in gar/bin. it looks like this script has been obsoleted. the functionality in fixlibtool exists in replacer ( which is called in gar.mk ) I can not see anywhere fixlibtool is used any longer. I renamed the fixme.sh to fixlibtool, but at the moment, it still lives in mgar/pkgs. if no one has any objection, I would like to: 1) svn mv the existing fixlibtool to orig_fixlibtool to keep it around. 2) move my fixlibtool into gar/bin. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoA3XcACgkQLrhmsXMSLxcCsgCdEVBuhV4/WMcCPlHID0su/cRl BiwAnRR/GQoJRDg8o/JHIuGAgaFlkciO =buCj -----END PGP SIGNATURE----- From dam at opencsw.org Wed May 6 10:53:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 6 May 2009 10:53:01 +0200 Subject: [csw-maintainers] I am running in circles around GNOME In-Reply-To: <4A00A395.1030500@opencsw.org> References: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> <4A00A395.1030500@opencsw.org> Message-ID: <07CBEACE-7FCB-42C9-A0A9-192612234EC3@opencsw.org> Hi Chad, Am 05.05.2009 um 22:37 schrieb Chad Harp: > I believe from the searching I did (few months ago) that the pango > dependency (via gtk) in pixman is just for some of the make check > tests. > > Try building pixman with --disable-gtk Good advice, that worked! Thank you :-) Best regards -- Dago From dam at opencsw.org Wed May 6 10:55:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 6 May 2009 10:55:19 +0200 Subject: [csw-maintainers] fixme script renamed to fixlibtool In-Reply-To: <4A00DD77.6060501@opencsw.org> References: <4A00DD77.6060501@opencsw.org> Message-ID: Hi Mike, Am 06.05.2009 um 02:44 schrieb Mike Watters: > I looked at the existing fixlibtool script in gar/bin. > it looks like this script has been obsoleted. the functionality in > fixlibtool > exists in replacer ( which is called in gar.mk ) I can not see > anywhere > fixlibtool is used any longer. > > I renamed the fixme.sh to fixlibtool, but at the moment, it still > lives in > mgar/pkgs. if no one has any objection, I would like to: > > 1) svn mv the existing fixlibtool to orig_fixlibtool to keep it > around. > 2) move my fixlibtool into gar/bin. Good thing. That would also avoid copying it in to every package. Please make sure to also adjust the wiki. Best regards -- Dago From dam at opencsw.org Wed May 6 11:46:24 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 6 May 2009 11:46:24 +0200 Subject: [csw-maintainers] DONE: pixman, pango, cairo References: Message-ID: <1C70E2B5-A67A-4D20-8B58-395320C69C25@opencsw.org> Hi, I just finished pixman, pango and cairo. The packages for Sparc are available in testing: build-01.May.2009/xproto-7.0.15,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/xextproto-7.0.5,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/xtrans-1.2.3,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/kbproto-1.0.3,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/inputproto-1.5.0,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/libxau-1.0.4,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-01.May.2009/libxau_devel-1.0.4,REV=2009.05.01-SunOS5.8-sparc- CSW.pkg build-01.May.2009/libpthreadstubs-0.1,REV=2009.05.01-SunOS5.8-all- CSW.pkg build-01.May.2009/xcbproto-1.4,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/libxcb-1.2,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-01.May.2009/libxcb_dev-1.2,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-01.May.2009/libxcb_doc-1.2,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-02.May.2009/renderproto-0.9.3,REV=2009.05.02-SunOS5.8-sparc- CSW.pkg build-02.May.2009/libxrender-0.9.4,REV=2009.05.02-SunOS5.8-sparc-CSW.pkg build-02.May.2009/renderdev-0.9.4,REV=2009.05.02-SunOS5.8-sparc-CSW.pkg build-05.May.2009/libxft2-2.1.13,REV=2009.05.05-SunOS5.8-sparc-CSW.pkg build-05.May.2009/libx11-1.2.1,REV=2009.05.05-SunOS5.8-sparc-CSW.pkg build-05.May.2009/libx11_devel-1.2.1,REV=2009.05.05-SunOS5.8-sparc- CSW.pkg build-06.May.2009/pixman-0.15.2,REV=2009.05.06-SunOS5.8-sparc-CSW.pkg build-06.May.2009/libpango-1.24.2,REV=2009.05.06-SunOS5.8-sparc-CSW.pkg build-06.May.2009/libpango_devel-1.24.2,REV=2009.05.06-SunOS5.8-sparc- CSW.pkg build-06.May.2009/libpango_doc-1.24.2,REV=2009.05.06-SunOS5.8-sparc- CSW.pkg build-06.May.2009/libcairo-1.8.6,REV=2009.05.06-SunOS5.8-sparc-CSW.pkg build-06.May.2009/libcairo_devel-1.8.6,REV=2009.05.06-SunOS5.8-sparc- CSW.pkg build-06.May.2009/libcairo_doc-1.8.6,REV=2009.05.06-SunOS5.8-all-CSW.pkg Also, all packages are installed on build8st, just in case you want to compile something against the new libs. If there are no issues I will repackage/release all packages in order on the official build machines for both sparc/i386. Best regards -- Dago From Darin.Perusich at cognigencorp.com Wed May 6 15:27:42 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Wed, 06 May 2009 09:27:42 -0400 Subject: [csw-maintainers] copy package to www:/home/newpkgs Message-ID: <4A01904E.9090908@cognigencorp.com> I'm being prompted for a password when I try and login to www to copy a package to /home/newpkgs. Can someone fix this please? -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From phil at bolthole.com Wed May 6 17:57:33 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 08:57:33 -0700 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <4A01904E.9090908@cognigencorp.com> References: <4A01904E.9090908@cognigencorp.com> Message-ID: <20090506155733.GB4461@bolthole.com> On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: > I'm being prompted for a password when I try and login to www to copy a > package to /home/newpkgs. Can someone fix this please? it's just ssh. since you have access to both boxes, you can fix it yourself by appending the .pub key from login, to your authorized_keys on www. From mwatters at opencsw.org Wed May 6 18:17:18 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 11:17:18 -0500 Subject: [csw-maintainers] fixme script renamed to fixlibtool In-Reply-To: References: <4A00DD77.6060501@opencsw.org> Message-ID: <4A01B80E.5000708@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi Mike, > > Am 06.05.2009 um 02:44 schrieb Mike Watters: > > Good thing. That would also avoid copying it in to every package. > Please make sure to also adjust the wiki. > > > Best regards > > -- Dago > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I have moved it in place and committed. In Addition to copying into place, I modified gar to create a new variable: STRIP_LIBTOOL when set to 1 in your Makefile it will call fixlibtool as the final part of the configure directive. revision: 4720 I will adjust the wiki accordingly - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoBuA4ACgkQLrhmsXMSLxf9JQCdHWfEgdAlrgx845AgMzgF2JUA YdQAnjyJbvoziU0La1stqlHtRDUlWSH6 =4yAT -----END PGP SIGNATURE----- From Darin.Perusich at cognigencorp.com Wed May 6 19:37:25 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Wed, 06 May 2009 13:37:25 -0400 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <20090506155733.GB4461@bolthole.com> References: <4A01904E.9090908@cognigencorp.com> <20090506155733.GB4461@bolthole.com> Message-ID: <4A01CAD5.5080101@cognigencorp.com> Philip Brown wrote: > On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: >> I'm being prompted for a password when I try and login to www to copy a >> package to /home/newpkgs. Can someone fix this please? > > it's just ssh. since you have access to both boxes, you can fix it yourself > by appending the .pub key from login, to your authorized_keys on www. This doesn't make any sense. If I'm unable to get logged into www with ssh how am I suppose to fix this problem? My id_rsa.pub from login is in my authorized_keys file so unless www is mounting my home directory it's not going to be there and I won't be able to login. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From phil at bolthole.com Wed May 6 19:42:49 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 10:42:49 -0700 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <4A01CAD5.5080101@cognigencorp.com> References: <4A01904E.9090908@cognigencorp.com> <20090506155733.GB4461@bolthole.com> <4A01CAD5.5080101@cognigencorp.com> Message-ID: <20090506174249.GH4461@bolthole.com> On Wed, May 06, 2009 at 01:37:25PM -0400, Darin Perusich wrote: > > Philip Brown wrote: > > On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: > >> I'm being prompted for a password when I try and login to www to copy a > >> package to /home/newpkgs. Can someone fix this please? > > > > it's just ssh. since you have access to both boxes, you can fix it yourself > > by appending the .pub key from login, to your authorized_keys on www. > > This doesn't make any sense. If I'm unable to get logged into www with > ssh how am I suppose to fix this problem? you cannot log in *from login.bo.opencsw.org* to www. but you can log into www from your own box, cant you? If so, thencopy it from login, to your home box, and then to www. From Darin.Perusich at cognigencorp.com Wed May 6 19:56:00 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Wed, 06 May 2009 13:56:00 -0400 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <20090506174249.GH4461@bolthole.com> References: <4A01904E.9090908@cognigencorp.com> <20090506155733.GB4461@bolthole.com> <4A01CAD5.5080101@cognigencorp.com> <20090506174249.GH4461@bolthole.com> Message-ID: <4A01CF30.5020403@cognigencorp.com> Philip Brown wrote: > On Wed, May 06, 2009 at 01:37:25PM -0400, Darin Perusich wrote: >> Philip Brown wrote: >>> On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: >>>> I'm being prompted for a password when I try and login to www to copy a >>>> package to /home/newpkgs. Can someone fix this please? >>> it's just ssh. since you have access to both boxes, you can fix it yourself >>> by appending the .pub key from login, to your authorized_keys on www. >> This doesn't make any sense. If I'm unable to get logged into www with >> ssh how am I suppose to fix this problem? > > you cannot log in *from login.bo.opencsw.org* to www. > but you can log into www from your own box, cant you? > > If so, thencopy it from login, to your home box, and then to www. > What an amazing thing clarity is. Thanks -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From pfelecan at opencsw.org Wed May 6 20:08:31 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 06 May 2009 20:08:31 +0200 Subject: [csw-maintainers] "garbage" when installing with pkg-get Message-ID: Before filling a bug report let me know if you have also this kind of behaviour: - SunOS host 5.10 Generic_137138-09 i86pc i386 i86pc - update the catalog pkg-get updatecatalog - upgrade everything pkg-get upgrade - install additional software pkg-get install nrpe I get, systematically the following: INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for o Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for l Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for r Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for t Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for | Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for t Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for p Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for w Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for r Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for a Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for p Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for | Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for w Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for l Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for a Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for u Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for t Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for i Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for l Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for | Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for o Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for m Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for m Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for o Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for n Perhaps your catalog is out of date No existing install of CSWnrpe found. Installing... Trying file:///htdocs/opencsw/unstable/i386/5.10/nrpe-2.12,REV=2009.04.26-SunOS5.8-i386-CSW.pkg.gz - PATH /opt/csw/flex-2.5.33/bin:/opt/csw/bin:/opt/csw/gcc3/bin:/usr/ccs/bin:/opt/csw/gnu:/usr/sbin:/usr/bin:/usr/dt/bin:/usr/openwin/bin:/bin -- Peter From phil at bolthole.com Wed May 6 20:16:42 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 11:16:42 -0700 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: References: Message-ID: <20090506181642.GB17668@bolthole.com> On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: > Before filling a bug report let me know if you have also this kind of > behaviour: > before filing a bug report, you should also either use it with a "normal" mirror, or specify where you are syncing your local copy from, and/or update your local copy again? > No existing install of CSWnrpe found. Installing... > Trying file:///htdocs/opencsw/unstable/i386/5.10/nrpe-2.12,REV=2009.04.26-SunOS5.8-i386-CSW.pkg.gz From pfelecan at opencsw.org Wed May 6 20:27:15 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 06 May 2009 20:27:15 +0200 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: <20090506181642.GB17668@bolthole.com> (Philip Brown's message of "Wed\, 6 May 2009 11\:16\:42 -0700") References: <20090506181642.GB17668@bolthole.com> Message-ID: Philip Brown writes: > On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: >> Before filling a bug report let me know if you have also this kind of >> behaviour: >> > > before filing a bug report, you should also either use it with a "normal" > mirror, or specify where you are syncing your local copy from, > and/or update your local copy again? You're right: rsync://www.ibiblio.org/sun-packages/opencsw -- Peter From phil at bolthole.com Wed May 6 20:39:39 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 11:39:39 -0700 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: References: <20090506181642.GB17668@bolthole.com> Message-ID: <20090506183939.GD17668@bolthole.com> On Wed, May 06, 2009 at 08:27:15PM +0200, Peter FELECAN wrote: > Philip Brown writes: > > > On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: > >> Before filling a bug report let me know if you have also this kind of > >> behaviour: > >> > > > > before filing a bug report, you should also either use it with a "normal" > > mirror, or specify where you are syncing your local copy from, > > and/or update your local copy again? > > You're right: rsync://www.ibiblio.org/sun-packages/opencsw Weeeelll... i just used the following command: pkg-get -s http://www.ibiblio.org/pub/solaris/opencsw/current -U -u (Internal SCCS code revision @(#) pkg-get 4.7@(#)) (on a solaris 10 sparc machine) and it worked ok. so... ? From pfelecan at opencsw.org Wed May 6 20:58:43 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 06 May 2009 20:58:43 +0200 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: <20090506183939.GD17668@bolthole.com> (Philip Brown's message of "Wed\, 6 May 2009 11\:39\:39 -0700") References: <20090506181642.GB17668@bolthole.com> <20090506183939.GD17668@bolthole.com> Message-ID: Philip Brown writes: > On Wed, May 06, 2009 at 08:27:15PM +0200, Peter FELECAN wrote: >> Philip Brown writes: >> >> > On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: >> >> Before filling a bug report let me know if you have also this kind of >> >> behaviour: >> >> >> > >> > before filing a bug report, you should also either use it with a "normal" >> > mirror, or specify where you are syncing your local copy from, >> > and/or update your local copy again? >> >> You're right: rsync://www.ibiblio.org/sun-packages/opencsw > > Weeeelll... > i just used the following command: > > pkg-get -s http://www.ibiblio.org/pub/solaris/opencsw/current -U -u > > (Internal SCCS code revision @(#) pkg-get 4.7@(#)) > (on a solaris 10 sparc machine) > > and it worked ok. > so... Add to your path /opt/csw/gnu, e.g. /opt/csw/flex-2.5.33/bin:/opt/csw/bin:/opt/csw/gcc3/bin:/usr/ccs/bin:/opt/csw/gnu:/usr/sbin:/usr/bin:/usr/dt/bin:/usr/openwin/bin:/bin -- Peter From mwatters at opencsw.org Thu May 7 03:01:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 20:01:22 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241468034-sup-3126@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> Message-ID: <4A0232E2.10707@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ben, Ben Walton wrote: > > I'm updating the gcc and g++ packages on a sol10/x86 box right now. > I'll build the same packages you did and let you know. > > Thanks > -Ben Any Word on gcc4.3.3 testing amd64? The i386, sparcv8 and sparcv9 have all been successful in testing. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoCMuEACgkQLrhmsXMSLxeh3wCeIE1Shboxi8B8uZpe/qIu/Mzw guEAn3/p59jxyFP6Wm9TG6tTxCh/9Ph+ =msMf -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 05:15:14 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 06 May 2009 23:15:14 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A0232E2.10707@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> Message-ID: <1241665912-sup-3788@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Wed May 06 21:01:22 -0400 2009: Hi Mike, > Any Word on gcc4.3.3 testing amd64? > The i386, sparcv8 and sparcv9 have all been successful in testing. I'm sorry, I forgot about this...the vim build with BUILD_64=1 was successful, but I haven't done the other package yet. I'll try to get to that tomorrow. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu May 7 06:30:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 23:30:08 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241665912-sup-3788@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> Message-ID: <4A0263D0.6010001@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > I'm sorry, I forgot about this...the vim build with BUILD_64=1 was > successful, but I haven't done the other package yet. I'll try to get > to that tomorrow. no worries, I know you are busy. you get to it when you get to it. Thank You for testing it for me. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoCY9AACgkQLrhmsXMSLxcQmwCfZ0GplDpxCRifuizimcMWup3G qdkAoK+PaRINeYywbfp+SrGdtccb2WbA =T65I -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu May 7 06:33:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 23:33:13 -0500 Subject: [csw-maintainers] php5 (NO Extensions) in testing Message-ID: <4A026489.1040605@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Changes: Added the elusive php-cgi binary - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoCZIgACgkQLrhmsXMSLxfLFwCgqraWlRR4qm3WcUdHtUAGq/Z6 dOQAoIlqGlbYNK327S2/zLQs0LhlxQWK =q92g -----END PGP SIGNATURE----- From dam at opencsw.org Thu May 7 11:36:03 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 11:36:03 +0200 Subject: [csw-maintainers] python and 64 bit Message-ID: <8F2A3D07-27A9-486C-86FD-87794FB9EB72@opencsw.org> Hi, I am still tweaking X11 stuff and noticed that xcb-proto generates some python files. > root/opt/csw/X11/lib/amd64 > root/opt/csw/X11/lib/amd64/pkgconfig > root/opt/csw/X11/lib/amd64/pkgconfig/xcb-proto.pc > root/opt/csw/X11/lib/pkgconfig > root/opt/csw/X11/lib/pkgconfig/xcb-proto.pc > root/opt/csw/X11/lib/python2.6 > root/opt/csw/X11/lib/python2.6/site-packages > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.py > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.pyc > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.pyo > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.py > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.pyc > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.pyo > ... Usually I try to always provide libs in 64 bit too. Last time we talked about this was in November and Joshua said he had build this some time ago. IMHO it would be great if we would have a 64 bit Python (CSWpython64? CSWpythonx in Sun-terminology?). The new mGAR should make it fairly easy to add 64 bit. Best regards -- Dago From ihsan at opencsw.org Thu May 7 15:43:33 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 15:43:33 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage Message-ID: <4A02E585.5080901@opencsw.org> Hello, For the last 4 years Courier-IMAP served us very well. But with the increased amount of users using IMAP, Courier-IMAP reached it limits. Especially users with huge mailboxes are affected by this issue. Therefore I will migrate this evening at 18:00 UTC to Dovecot IMAP. Dovecot IMAP is a new IMAP server, which is much faster than Courier and it provides backward compatibility to Courier. I'm expecting a downtime of 1-2 hours. During this time, there will be no mail services available (SMTP and IMAP). Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From mwatters at opencsw.org Thu May 7 16:06:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 09:06:17 -0500 Subject: [csw-maintainers] python and 64 bit In-Reply-To: <8F2A3D07-27A9-486C-86FD-87794FB9EB72@opencsw.org> References: <8F2A3D07-27A9-486C-86FD-87794FB9EB72@opencsw.org> Message-ID: <4A02EAD9.7000108@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago, Dagobert Michelsen wrote: > > Usually I try to always provide libs in 64 bit too. Last time we > talked about this was in November and Joshua said he had build > this some time ago. IMHO it would be great if we would have a > 64 bit Python (CSWpython64? CSWpythonx in Sun-terminology?). > The new mGAR should make it fairly easy to add 64 bit. > I have a bug report for this already ;) I have been holding for the GCC4.3.3 release. Python compiles against some gcc dependent libraries that need 64bit support. All the testing thus far on gcc4.3.3 is good, Ben is finishing up testing the amd64 builds. I hope to get gcc released by the end of the week. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoC6sMACgkQLrhmsXMSLxd/MwCfbffh9mV8zBZOJYgjkljunqWK j7AAoLlRxDP/Urjj8KzZk78MNnZYXTXF =JGYM -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 16:31:30 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 10:31:30 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02E585.5080901@opencsw.org> References: <4A02E585.5080901@opencsw.org> Message-ID: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: > For the last 4 years Courier-IMAP served us very well. But with the > increased amount of users using IMAP, Courier-IMAP reached it limits. > Especially users with huge mailboxes are affected by this issue. I made the move from courier and uw-imap a few years back and have never regretted it. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darin.Perusich at cognigencorp.com Thu May 7 16:40:45 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 07 May 2009 10:40:45 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Message-ID: <4A02F2ED.1010306@cognigencorp.com> Ben Walton wrote: > Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >> For the last 4 years Courier-IMAP served us very well. But with the >> increased amount of users using IMAP, Courier-IMAP reached it limits. >> Especially users with huge mailboxes are affected by this issue. > > I made the move from courier and uw-imap a few years back and have > never regretted it. > I've always preferred cyrus personally ;-) -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From ihsan at opencsw.org Thu May 7 16:45:52 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 16:45:52 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Message-ID: <4A02F420.9090301@opencsw.org> Am 7.5.2009 16:31 Uhr, Ben Walton schrieb: >> For the last 4 years Courier-IMAP served us very well. But with the >> increased amount of users using IMAP, Courier-IMAP reached it limits. >> Especially users with huge mailboxes are affected by this issue. > I made the move from courier and uw-imap a few years back and have > never regretted it. I was really thinking for long time about that, but I hesitated because Courier worked very well. I've made a few tests and I was impressed how fast Dovecot is. Even the migration is very easy. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Thu May 7 16:47:25 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 16:47:25 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02F2ED.1010306@cognigencorp.com> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F2ED.1010306@cognigencorp.com> Message-ID: <4A02F47D.8090302@opencsw.org> Am 7.5.2009 16:40 Uhr, Darin Perusich schrieb: >> Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >>> For the last 4 years Courier-IMAP served us very well. But with the >>> increased amount of users using IMAP, Courier-IMAP reached it limits. >>> Especially users with huge mailboxes are affected by this issue. >> I made the move from courier and uw-imap a few years back and have >> never regretted it. > > I've always preferred cyrus personally ;-) I've looked at Courier as well, but the not existing documentation makes very hard to start with Cyrus. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Thu May 7 16:51:06 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 10:51:06 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02F420.9090301@opencsw.org> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F420.9090301@opencsw.org> Message-ID: <1241707760-sup-9083@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Thu May 07 10:45:52 -0400 2009: > I've made a few tests and I was impressed how fast Dovecot is. Even the > migration is very easy. When we switched a machine from uw to dovecot last year, the load drop was simply awesome. The indexing that dovecot does on mbox files really makes a huge difference. It's extended the useful life of this particular machine by quite a bit. ...why we still have mbox in use is another story! :( -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darin.Perusich at cognigencorp.com Thu May 7 16:53:58 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 07 May 2009 10:53:58 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02F47D.8090302@opencsw.org> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F2ED.1010306@cognigencorp.com> <4A02F47D.8090302@opencsw.org> Message-ID: <4A02F606.3000607@cognigencorp.com> Ihsan Dogan wrote: > Am 7.5.2009 16:40 Uhr, Darin Perusich schrieb: > >>> Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >>>> For the last 4 years Courier-IMAP served us very well. But with the >>>> increased amount of users using IMAP, Courier-IMAP reached it limits. >>>> Especially users with huge mailboxes are affected by this issue. >>> I made the move from courier and uw-imap a few years back and have >>> never regretted it. >> I've always preferred cyrus personally ;-) > > I've looked at Courier as well, but the not existing documentation makes > very hard to start with Cyrus. > Yeah, that's one area where they fall short, especially if you're trying to setup murder, but it's not as bad as it use to be. Once it's up and running you can basically forget about it and does it scale! -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From bwalton at opencsw.org Thu May 7 16:56:52 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 10:56:52 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A0263D0.6010001@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> Message-ID: <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: > no worries, I know you are busy. > you get to it when you get to it. Thank You for testing it for me. libffi builds fine, but there were linking issues (shown by ldd and dump), not during the actual build. The following small GAR modification resolved this: $ svn diff Index: gar.conf.mk =================================================================== --- gar.conf.mk (revision 4723) +++ gar.conf.mk (working copy) @@ -477,7 +477,7 @@ OPTFLAGS ?= $(strip $($(GARCOMPILER)_CC_FLAGS) $(_CATEGORY_OPTFLAGS) $(EXTRA_OPTFLAGS)) GCC3_LD_OPTIONS = -R$(GCC3_CC_HOME)/lib $(EXTRA_GCC3_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) -GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib $(EXTRA_GCC4_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) +GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib/\$$ISALIST $(EXTRA_GCC4_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) SOS11_LD_OPTIONS = $(strip $(EXTRA_SOS11_LD_OPTIONS) $(EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) SOS12_LD_OPTIONS = $(strip $(EXTRA_SOS12_LD_OPTIONS) $(EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) I don't know if that's the best way to make this change sticky, so Dago's input here would be helpful. I'm doing binutils now. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at opencsw.org Thu May 7 17:55:59 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 17:55:59 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241707760-sup-9083@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F420.9090301@opencsw.org> <1241707760-sup-9083@ntdws12.chass.utoronto.ca> Message-ID: <4A03048F.3070601@opencsw.org> Am 7.5.2009 16:51 Uhr, Ben Walton schrieb: >> I've made a few tests and I was impressed how fast Dovecot is. Even the >> migration is very easy. > When we switched a machine from uw to dovecot last year, the load drop > was simply awesome. The indexing that dovecot does on mbox files > really makes a huge difference. It's extended the useful life of this > particular machine by quite a bit. Are you using IMAP IDLE as well? I couldn't find any information, how much load IMAP IDLE will generate. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From phil at bolthole.com Thu May 7 18:45:39 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 7 May 2009 09:45:39 -0700 Subject: [csw-maintainers] php5 (NO Extensions) in testing In-Reply-To: <4A026489.1040605@opencsw.org> References: <4A026489.1040605@opencsw.org> Message-ID: <20090507164539.GC84608@bolthole.com> On Wed, May 06, 2009 at 11:33:13PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Changes: > Added the elusive php-cgi binary wierd. there's a command-line "php" standalone binary.. but there's also a SEPARATE php-cgi binary? i guess possibly to support "Fast-CGI"? From bwalton at opencsw.org Thu May 7 18:53:13 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 12:53:13 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A03048F.3070601@opencsw.org> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F420.9090301@opencsw.org> <1241707760-sup-9083@ntdws12.chass.utoronto.ca> <4A03048F.3070601@opencsw.org> Message-ID: <1241714904-sup-8536@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Thu May 07 11:55:59 -0400 2009: > Are you using IMAP IDLE as well? I couldn't find any information, how > much load IMAP IDLE will generate. I've always left that in the default setting state (eg: the workaround for lookout/express is enabled). We have client connections active for (sometimes) days on end. Whether many of the clients in play use the IDLE command, I don't know...we've got a hodge-podge of software out there (Pegasus, Eudora, old Netscape, etc in addition to newer/decent stuff). I haven't tried disabling it to see if performance is better though... Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu May 7 18:56:10 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 18:56:10 +0200 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241708078-sup-5541@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Message-ID: Hi Ben, Am 07.05.2009 um 16:56 schrieb Ben Walton: > Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 > 2009: >> no worries, I know you are busy. >> you get to it when you get to it. Thank You for testing it for me. > > libffi builds fine, but there were linking issues (shown by ldd and > dump), not during the actual build. The following small GAR > modification resolved this: > > > $ svn diff > Index: gar.conf.mk > =================================================================== > --- gar.conf.mk (revision 4723) > +++ gar.conf.mk (working copy) > @@ -477,7 +477,7 @@ > OPTFLAGS ?= $(strip $($(GARCOMPILER)_CC_FLAGS) $(_CATEGORY_OPTFLAGS) > $(EXTRA_OPTFLAGS)) > > GCC3_LD_OPTIONS = -R$(GCC3_CC_HOME)/lib $(EXTRA_GCC3_LD_OPTIONS) $ > (EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) > -GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib $(EXTRA_GCC4_LD_OPTIONS) $ > (EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) > +GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib/\$$ISALIST $ > (EXTRA_GCC4_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) If you want $ISALIST, yes. Otherwise you may also use $(abspath $ (GCC4_CC_HOME)/lib/$(MM_LIBDIR)) for just 64 bits if there are no optimized libs. > SOS11_LD_OPTIONS = $(strip $(EXTRA_SOS11_LD_OPTIONS) $ > (EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) > SOS12_LD_OPTIONS = $(strip $(EXTRA_SOS12_LD_OPTIONS) $ > (EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) > > I don't know if that's the best way to make this change sticky, so > Dago's input here would be helpful. I'm doing binutils now. Please go ahead and commit either the ISALIST or the static /64 version, it was indeed a bug. Best regards -- Dago From dam at opencsw.org Thu May 7 19:38:52 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 19:38:52 +0200 Subject: [csw-maintainers] Updating all build servers Message-ID: Hi, I am updating all build servers to current/ now. Please stand by. Best regards -- Dago From dam at opencsw.org Thu May 7 19:46:10 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 19:46:10 +0200 Subject: [csw-maintainers] Updating all build servers In-Reply-To: References: Message-ID: <7B37DE1C-7314-44D5-BD67-0DF274573165@opencsw.org> Hi, Am 07.05.2009 um 19:38 schrieb Dagobert Michelsen: > I am updating all build servers to current/ now. Please stand by. Please note: this includes Subversion. Your local repository will be updated when you next access it with the new svn 1.6 client. Best regards -- Dago From mwatters at opencsw.org Thu May 7 19:54:27 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 12:54:27 -0500 Subject: [csw-maintainers] php5 (NO Extensions) in testing In-Reply-To: <20090507164539.GC84608@bolthole.com> References: <4A026489.1040605@opencsw.org> <20090507164539.GC84608@bolthole.com> Message-ID: <4A032053.4070007@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Wed, May 06, 2009 at 11:33:13PM -0500, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Changes: >> Added the elusive php-cgi binary > > wierd. there's a command-line "php" standalone binary.. but there's also a > SEPARATE php-cgi binary? > > i guess possibly to support "Fast-CGI"? > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers yes and no the php-cgi is for specifically running php in cgi mode, not necessarily for fast-cgi. The best justification I have found to use php-cgi over php: "Using PHP as a CGI binary is an option for setups that for some reason do not wish to integrate PHP as a module into server software (like Apache), or will use PHP with different kinds of CGI wrappers to create safe chroot and setuid environments for scripts." http://us.php.net/manual/en/security.cgi-bin.attacks.php - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDIFIACgkQLrhmsXMSLxcTJgCdEVnW6W+GBo9HsBilSCgoDAuc eFAAn0CYL+nKNFMx/kHpQXdMPf8oV/xo =k5fj -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 20:57:52 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 14:57:52 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A0263D0.6010001@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> Message-ID: <1241722471-sup-1684@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: The binutils build is successful, but the merge fails with output that pasted below...The binaries that do get merged look ok. I don't know if this is a build issue or a gar description issue at this point. I won't have a chance to look in the next little while, either. /bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/nm ./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/strip pax: link: : No such file or directory /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/strip ./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/c++filt /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/c++filt ./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/as pax: link: : No such file or directory /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/as ./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/gprof /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/gprof ./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/ld pax: link: : No such file or directory /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/ld ./opt/csw/bin/amd64/stpIain1 >> ./opt/csw/bin/amd64/stpIain1 /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/stpIain1 gmake[1]: *** [merge-copy-relocated-only] Error 1 gmake[1]: Leaving directory `/homes/bwalton/packages/binutils/trunk' gmake: *** [merge-isa-amd64] Error 2 Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Thu May 7 21:06:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 15:06:08 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Message-ID: <1241723127-sup-5535@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 -0400 2009: Hi Dago, > If you want $ISALIST, yes. Otherwise you may also use $(abspath $ > (GCC4_CC_HOME)/lib/$(MM_LIBDIR)) > for just 64 bits if there are no optimized libs. I think ISALIST is the best approach here, so I'll commit that shortly. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at opencsw.org Thu May 7 21:17:57 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 21:17:57 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02E585.5080901@opencsw.org> References: <4A02E585.5080901@opencsw.org> Message-ID: <4A0333E5.4060902@opencsw.org> Am 7.5.2009 15:43 Uhr, Ihsan Dogan schrieb: > For the last 4 years Courier-IMAP served us very well. But with the > increased amount of users using IMAP, Courier-IMAP reached it limits. > Especially users with huge mailboxes are affected by this issue. > > Therefore I will migrate this evening at 18:00 UTC to Dovecot IMAP. > Dovecot IMAP is a new IMAP server, which is much faster than Courier and > it provides backward compatibility to Courier. > > I'm expecting a downtime of 1-2 hours. During this time, there will be > no mail services available (SMTP and IMAP). The migration is over and everything is up and running again. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Thu May 7 21:20:11 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 15:20:11 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Message-ID: <1241723902-sup-9306@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 -0400 2009: Before I committed the change, I took a peek in /opt/csw/gcc4/lib. It doesn't have the 64 -> amd64, 32 -> ., i386 -> . symlinks that exist in /opt/csw/lib. These should likely be added to the gcc4 core package before the $ISALIST is added, no? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu May 7 21:33:54 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 21:33:54 +0200 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241723902-sup-9306@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> <1241723902-sup-9306@ntdws12.chass.utoronto.ca> Message-ID: <4D3E9D15-3A6C-45EF-BA62-11E20BC46F22@opencsw.org> Hi Ben, Am 07.05.2009 um 21:20 schrieb Ben Walton: > Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 > -0400 2009: > > Before I committed the change, I took a peek in /opt/csw/gcc4/lib. It > doesn't have the 64 -> amd64, 32 -> ., i386 -> . symlinks that exist > in /opt/csw/lib. These should likely be added to the gcc4 core > package before the $ISALIST is added, no? That would be nice, yes. Otherwise you must add -R$(GCC4_CC_HOME)/lib/\$$ISALIST -R$(abspath $(GCC4_CC_HOME)/lib/$ (MM_LIBDIR)) If you specify both you catch 64 bit with lib/(sparcv9|amd64) and 32 bit with lib/. Best regards -- Dago From mwatters at opencsw.org Thu May 7 21:44:36 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 14:44:36 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241723902-sup-9306@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> <1241723902-sup-9306@ntdws12.chass.utoronto.ca> Message-ID: <4A033A24.1090409@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 -0400 2009: > > Before I committed the change, I took a peek in /opt/csw/gcc4/lib. It > doesn't have the 64 -> amd64, 32 -> ., i386 -> . symlinks that exist > in /opt/csw/lib. These should likely be added to the gcc4 core > package before the $ISALIST is added, no? > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Those links are missing on the sparc package as well. gcc4 does not use gar to merge :( I just committed a change to my recipe to fix the missing links, I will repackage with the links in place. The merge problems you had on binutils, I have the exact same issue on sparc. ( I never tested the merge, just up through install ) The binaries work OK for sparcv9. Looks like a merge problem in gar. just to be sure, I am testing the merge with Sun Studio compiled binaries and will let you know what happens. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDOiQACgkQLrhmsXMSLxcpqACgjUe0gl1GztEliOmR30wxB3TR IqsAoNv6sWYfes0JE5BLNoVfM4w45fET =Ecgx -----END PGP SIGNATURE----- From ihsan at opencsw.org Thu May 7 21:44:47 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 21:44:47 +0200 Subject: [csw-maintainers] IMPORTAN information for Procmail users on mail (IMAP) Message-ID: <4A033A2F.3080504@opencsw.org> Hello, The Courier IMAP on the mail server has been replaced by Dovecot. If you are using Procmail to sort mail, you might consider to deliver your mail with the Dovecot LDA in order to update the index files immediately. There is a document on the Dovecot Wiki: http://wiki.dovecot.org/procmail If you had for example this entry in your ~/.procmailrc to deliver the mails from the announce mailing list to Inbox/opencsw/announce :0H * ^List-Post: .opencsw.announce/ You would have to replace it with: DELIVER="/opt/csw/libexec/dovecot/deliver" :0H * ^List-Post: | $DELIVER -m INBOX.opencsw.announce Please note that we don't use the default namespace. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Thu May 7 22:13:02 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 22:13:02 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Message-ID: <4A0340CE.60109@opencsw.org> Am 7.5.2009 16:31 Uhr, Ben Walton schrieb: > Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >> For the last 4 years Courier-IMAP served us very well. But with the >> increased amount of users using IMAP, Courier-IMAP reached it limits. >> Especially users with huge mailboxes are affected by this issue. > I made the move from courier and uw-imap a few years back and have > never regretted it. I regret now, that I haven changed earlier. I was testing the migration in a VM on my Mac and I've noticed that Dovecot was faster than Courier. After going live with Dovecot today, I'm really suprised that the difference is that big. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From mwatters at opencsw.org Thu May 7 22:31:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 15:31:08 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241722471-sup-1684@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> Message-ID: <4A03450C.2070801@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: > > The binutils build is successful, but the merge fails with output > that pasted below...The binaries that do get merged look ok. I don't > know if this is a build issue or a gar description issue at this > point. I won't have a chance to look in the next little while, > either. > > > /bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/nm > ./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/strip > pax: link: : No such file or directory > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/strip > ./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/c++filt > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/c++filt > ./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/as > pax: link: : No such file or directory > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/as > ./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/gprof > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/gprof > ./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/ld > pax: link: : No such file or directory > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/ld > ./opt/csw/bin/amd64/stpIain1 >> ./opt/csw/bin/amd64/stpIain1 > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/stpIain1 > gmake[1]: *** [merge-copy-relocated-only] Error 1 > gmake[1]: Leaving directory `/homes/bwalton/packages/binutils/trunk' > gmake: *** [merge-isa-amd64] Error 2 > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I got the same merge errors compiling with Sun Studio I would give this a 99.9% probability of a bug in either gar or binutils. I will re-package gcc4.3.3 to fix the missing links in gar and push out for release. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDRQwACgkQLrhmsXMSLxcdpQCgvjaZjja84GBDl58WHmYko8vH 8WAAn1QG7ngL/YatuDpiZ27klKBVS+n6 =8nou -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu May 7 22:32:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 15:32:17 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A03450C.2070801@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> <4A03450C.2070801@opencsw.org> Message-ID: <4A034551.1060904@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > Ben Walton wrote: >> Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: > >> The binutils build is successful, but the merge fails with output >> that pasted below...The binaries that do get merged look ok. I don't >> know if this is a build issue or a gar description issue at this >> point. I won't have a chance to look in the next little while, >> either. > > >> /bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/nm >> ./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/strip >> pax: link: : No such file or directory >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/c++filt >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/as >> pax: link: : No such file or directory >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/gprof >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/ld >> pax: link: : No such file or directory >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/stpIain1 >> ./opt/csw/bin/amd64/stpIain1 >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/stpIain1 >> gmake[1]: *** [merge-copy-relocated-only] Error 1 >> gmake[1]: Leaving directory `/homes/bwalton/packages/binutils/trunk' >> gmake: *** [merge-isa-amd64] Error 2 > >> Thanks >> -Ben > > >> ------------------------------------------------------------------------ > >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > I got the same merge errors compiling with Sun Studio > > I would give this a 99.9% probability of a bug in either gar or binutils. > > I will re-package gcc4.3.3 to fix the missing links in gar and push out for > release. > > fix the missing links in "lib" not gar... sorry - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDRVAACgkQLrhmsXMSLxfiCwCcDCaFBf3y74lCkKPnZfxYgkPp eLMAn1uJUhqyhc+x5gA8jX2+bmr8y8j6 =7fXo -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 23:54:38 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 17:54:38 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A034551.1060904@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> <4A03450C.2070801@opencsw.org> <4A034551.1060904@opencsw.org> Message-ID: <1241733234-sup-1049@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu May 07 16:32:17 -0400 2009: > > I would give this a 99.9% probability of a bug in either gar or > > binutils. I agree with this. I think that once you get those links added to lib that the package looks ready to roll. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Fri May 8 00:23:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 00:23:01 +0200 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241733234-sup-1049@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> <4A03450C.2070801@opencsw.org> <4A034551.1060904@opencsw.org> <1241733234-sup-1049@ntdws12.chass.utoronto.ca> Message-ID: <4EE27F31-0FEB-4D29-8483-8A90613FF97F@opencsw.org> Hi Ben, Am 07.05.2009 um 23:54 schrieb Ben Walton: > Excerpts from Mike Watters's message of Thu May 07 16:32:17 -0400 > 2009: > >>> I would give this a 99.9% probability of a bug in either gar or >>> binutils. > > I agree with this. I think that once you get those links added to lib > that the package looks ready to roll. It is a GAR issue. I'll take care of it. Best regards -- Dago From mwatters at opencsw.org Fri May 8 06:10:02 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 23:10:02 -0500 Subject: [csw-maintainers] bug 3034 separate server and user command line binaries Message-ID: <4A03B09A.4060103@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I have to recompile subversion to fix an issue with ap2_subversion. I want to include this bug in the re-package. which binaries constitute server vs client? Can I get some confirmation on this before I split it up. I believe it should be as follows: Package Name:CSWsvn Catalog Name:svn ~~~~~~~~~~~~~~~~~~~~~~~~~~~ server - bin/svnserve server - bin/svnsync server - bin/svnadmin server - bin/svndumpfilter Package Name:CSWsvn-client Catalog Name:svn_client ~~~~~~~~~~~~~~~~~~~~~~~~~~~ client - bin/svnversion client - bin/svnlook client - bin/svn - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoDsJoACgkQLrhmsXMSLxf8fQCfXpRREcA0/YB7WWIGZmzjaOdY niMAmwaxV+oz32HJ1sr1m1bgzF/QnGMg =psHQ -----END PGP SIGNATURE----- From bonivart at opencsw.org Fri May 8 09:46:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 8 May 2009 09:46:14 +0200 Subject: [csw-maintainers] Fwd: [csw-users] Converting from Blastwave with pkgutil In-Reply-To: <71083D3F-F533-4834-B50E-78E3099671E8@gmail.com> References: <71083D3F-F533-4834-B50E-78E3099671E8@gmail.com> Message-ID: <625385e30905080046u1b9b8c25nf40cef5d517a42e0@mail.gmail.com> User request... -- /peter ---------- Forwarded message ---------- From: Cody Herriges Date: Fri, May 8, 2009 at 12:34 AM Subject: [csw-users] Converting from Blastwave with pkgutil To: users at lists.opencsw.org When pkgutil became the desired package management tool in Blastwave around the time of the split I switched my entire jumpstart configuration to be more tailored towards the use of pkgutil. ?Due to the lack of some key user apps that aren't being updated I have decided to switch over to opencsw. ?Any chance we could get a pkgutil package dropped in the root of the opencsw package repository so I can grab the latest pkgutil during my jumpstarts just as easily as I can grab the pkg-get package? Thanks, ---------------------------------------------------- Cody Herriges - Lead Unix System Administrator MCECS - ?Portland State University From pfelecan at opencsw.org Fri May 8 17:44:46 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 17:44:46 +0200 Subject: [csw-maintainers] OpenCSW default PHP version Message-ID: I'm packaging a product which has a PHP binding, amongst other bindings. I must make a choice to bind for PHP4 or PHP5 and I would like to choose the "default one"... Which is? -- Peter From bwalton at opencsw.org Fri May 8 17:56:03 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 11:56:03 -0400 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: References: Message-ID: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: > I'm packaging a product which has a PHP binding, amongst other > bindings. I must make a choice to bind for PHP4 or PHP5 and I would like > to choose the "default one"... Which is? PHP4 is deprecated and should be available as legacy only. It was eol'd in 2007 with critical security patches only until 2008. Nothing new should be built against it, imo. http://www.php.net/archive/2007.php (search down to 'end of life') -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Fri May 8 18:05:26 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:05:26 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding Message-ID: Again, I'm trying to provide a python binding and I obtain a strange situation, in my opinion incorrect: CFLAGS="-O3 -march=i486 -Wall -DSWF_LITTLE_ENDIAN -fno-strict-aliasing" \ /opt/csw/bin/python setup.py build running build running build_py creating build creating build/lib.solaris-2.10-i86pc-2.6 copying [...]/py_ext/ming.py -> build/lib.solaris-2.10-i86pc-2.6 copying [...]/py_ext/mingc.py -> build/lib.solaris-2.10-i86pc-2.6 running build_ext building '_mingc' extension creating build/temp.solaris-2.10-i86pc-2.6 [...] gcc -DNDEBUG -O -O3 -march=i486 -Wall -DSWF_LITTLE_ENDIAN -fno-strict-aliasing \ -I/usr/local/include [...]/py_ext/ming_wrap.o /opt/studio/SOS11/SUNWspro/bin/cc -G -O3 -march=i486 -Wall \ -DSWF_LITTLE_ENDIAN \ -fno-strict-aliasing [...]/py_ext/ming_wrap.o \ -L../src/.libs -L/usr/local/lib/ -lming -lz -lpng -lungif \ -lpython2.6 -o build/lib.solaris-2.10-i86pc-2.6/_mingc.so unable to execute /opt/studio/SOS11/SUNWspro/bin/cc: No such file or directory error: command '/opt/studio/SOS11/SUNWspro/bin/cc' failed with exit status 1 There are 2 issues: 1. I'm using gcc, so why cc is used? Because Python is generated with Sun Studio? 2. How comes that a specific build machine path is used, i.e. Sun Studio is installed by default in /opt/SUNWspro (I know that it can be installed at the administrator's convenience and on the build machines is installed in a /opt/studio directory but that is another story) The second issue is a show stopper for me... -- Peter From mwatters at opencsw.org Fri May 8 18:24:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:24:03 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: References: Message-ID: <4A045CA3.5070908@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Peter, Peter FELECAN wrote: > There are 2 issues: > > 1. I'm using gcc, so why cc is used? Because Python is generated with > Sun Studio? Yes that is exactly correct, when you build a python extension it will look at the way python was compiled and use those settings ( this is just like PERL ) > 2. How comes that a specific build machine path is used, i.e. Sun > Studio is installed by default in /opt/SUNWspro (I know that it can > be installed at the administrator's convenience and on the build > machines is installed in a /opt/studio directory but that is another > story) Unfortunately, this is just the way it is handled, it is true for PERL and most if not all the other packages that have a pkg-config like structure. you should be able to work around this by setting the environment variables CC={path to cc} CXX={path to c++ (CC in SunStudio)} > The second issue is a show stopper for me... - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEXKIACgkQLrhmsXMSLxfRWgCfWix2FTTnX34uuzsr5AWXDPao zeYAn2gbcXoeX73QwLFKdd73y1NiOBpf =+rV1 -----END PGP SIGNATURE----- From pfelecan at opencsw.org Fri May 8 18:27:34 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:27:34 +0200 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> (Ben Walton's message of "Fri\, 08 May 2009 11\:56\:03 -0400") References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> Message-ID: Ben Walton writes: > Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: >> I'm packaging a product which has a PHP binding, amongst other >> bindings. I must make a choice to bind for PHP4 or PHP5 and I would like >> to choose the "default one"... Which is? > > PHP4 is deprecated and should be available as legacy only. It was > eol'd in 2007 with critical security patches only until 2008. Nothing > new should be built against it, imo. > > http://www.php.net/archive/2007.php (search down to 'end of life') Thank you Ben. This led me to ask: why have we PHP4 in /opt/csw/bin and not PHP5? If I wish to use the current PHP I would expect to have it in /opt/csw/bin and not in some specific path, i.e. /opt/csw/php5/bin Can this be corrected? -- Peter From mwatters at opencsw.org Fri May 8 18:28:38 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:28:38 -0500 Subject: [csw-maintainers] Odd issues with the new IMAP Message-ID: <4A045DB6.90104@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I am using thunderbird/enigmail/imap for my opencsw account. Since the upgrade, I am getting Messages that are showing up at different times. for example: at 11:00 my time I received 4 messages from earlier in the day 08:32 08:48 09:23 and 10:47 "mytime" this is not really a problem, just Odd and only after the imap upgrade. anyone else experiencing issues similar? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEXbYACgkQLrhmsXMSLxdMvgCg5dkJ2BNWX1VPLmxNghcy4v08 X5IAn3W3n5brBZwVJQ/dlk3Nyn70Szwr =qPSt -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri May 8 18:30:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:30:24 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> Message-ID: <4A045E20.30306@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Ben Walton writes: > >> Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: >>> I'm packaging a product which has a PHP binding, amongst other >>> bindings. I must make a choice to bind for PHP4 or PHP5 and I would like >>> to choose the "default one"... Which is? >> PHP4 is deprecated and should be available as legacy only. It was >> eol'd in 2007 with critical security patches only until 2008. Nothing >> new should be built against it, imo. >> >> http://www.php.net/archive/2007.php (search down to 'end of life') > > Thank you Ben. This led me to ask: why have we PHP4 in /opt/csw/bin and > not PHP5? If I wish to use the current PHP I would expect to have it in > /opt/csw/bin and not in some specific path, i.e. /opt/csw/php5/bin > > Can this be corrected? I am just about to re-compile php5 to fix the latest bugs/feature requests. now would be the time to change it. ;) let me know, I will hold on the recompile until someone weighs in on this. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEXh8ACgkQLrhmsXMSLxdp2QCg4pc09Jm0JAPdWBmpJzaKgNE0 FKUAoJaIdMQWa2hVb7vb50uQTgj7K5xG =Fumm -----END PGP SIGNATURE----- From dam at opencsw.org Fri May 8 18:35:55 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 18:35:55 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A045CA3.5070908@opencsw.org> References: <4A045CA3.5070908@opencsw.org> Message-ID: Hi, Am 08.05.2009 um 18:24 schrieb Mike Watters: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi Peter, > Peter FELECAN wrote: > >> There are 2 issues: >> >> 1. I'm using gcc, so why cc is used? Because Python is generated with >> Sun Studio? > Yes that is exactly correct, when you build a python extension it will > look at the way python was compiled and use those settings ( this is > just like > PERL ) > >> 2. How comes that a specific build machine path is used, i.e. Sun >> Studio is installed by default in /opt/SUNWspro (I know that it can >> be installed at the administrator's convenience and on the build >> machines is installed in a /opt/studio directory but that is >> another >> story) > > Unfortunately, this is just the way it is handled, it is true for > PERL and > most if not all the other packages that have a pkg-config like > structure. > > you should be able to work around this by setting the environment > variables > CC={path to cc} CXX={path to c++ (CC in SunStudio)} Is it really this bad? Then we should make Sun Studio 11 the default at /opt/SUNWspro and let GAR switch flags on the detected compiler if there is e. g. SS12 at that location. Best regards -- Dago From pfelecan at opencsw.org Fri May 8 18:36:49 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:36:49 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A045CA3.5070908@opencsw.org> (Mike Watters's message of "Fri\, 08 May 2009 11\:24\:03 -0500") References: <4A045CA3.5070908@opencsw.org> Message-ID: Mike Watters writes: > Hi Peter, Thank you Mike for your explanation. > Peter FELECAN wrote: > >> There are 2 issues: >> >> 1. I'm using gcc, so why cc is used? Because Python is generated with >> Sun Studio? > Yes that is exactly correct, when you build a python extension it will > look at the way python was compiled and use those settings ( this is just like > PERL ) You're right about perl, however it uses the well behaved cc driver and not an absolute path, as for Python (see the second issue) because cc is /usr/bin/cc which is a link to /opt/SUNWspro/bin/cc. This works on my system but will not work without tweaking on the build system... >> 2. How comes that a specific build machine path is used, i.e. Sun >> Studio is installed by default in /opt/SUNWspro (I know that it can >> be installed at the administrator's convenience and on the build >> machines is installed in a /opt/studio directory but that is another >> story) > > Unfortunately, this is just the way it is handled, it is true for PERL and > most if not all the other packages that have a pkg-config like structure. > > you should be able to work around this by setting the environment variables > CC={path to cc} CXX={path to c++ (CC in SunStudio)} But this will work only if I use Sun Studio C and not gcc. I must confess that I don't like this situation of having more than a compiler and doing acrobatics to handle weird issues. Well, I'm opting to not provide some bindings. -- Peter From pfelecan at opencsw.org Fri May 8 18:39:31 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:39:31 +0200 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <4A045E20.30306@opencsw.org> (Mike Watters's message of "Fri\, 08 May 2009 11\:30\:24 -0500") References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> Message-ID: Mike Watters writes: > Peter FELECAN wrote: >> Ben Walton writes: >> >>> Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: >>>> I'm packaging a product which has a PHP binding, amongst other >>>> bindings. I must make a choice to bind for PHP4 or PHP5 and I would like >>>> to choose the "default one"... Which is? >>> PHP4 is deprecated and should be available as legacy only. It was >>> eol'd in 2007 with critical security patches only until 2008. Nothing >>> new should be built against it, imo. >>> >>> http://www.php.net/archive/2007.php (search down to 'end of life') >> >> Thank you Ben. This led me to ask: why have we PHP4 in /opt/csw/bin and >> not PHP5? If I wish to use the current PHP I would expect to have it in >> /opt/csw/bin and not in some specific path, i.e. /opt/csw/php5/bin >> >> Can this be corrected? > > I am just about to re-compile php5 to fix the latest bugs/feature requests. > now would be the time to change it. > > ;) let me know, I will hold on the recompile until someone weighs in on this. I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at least, put it in /opt/csw/php4. -- Peter From pfelecan at opencsw.org Fri May 8 18:41:06 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:41:06 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: (Dagobert Michelsen's message of "Fri\, 8 May 2009 18\:35\:55 +0200") References: <4A045CA3.5070908@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi, > > Am 08.05.2009 um 18:24 schrieb Mike Watters: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi Peter, >> Peter FELECAN wrote: >> >>> There are 2 issues: >>> >>> 1. I'm using gcc, so why cc is used? Because Python is generated with >>> Sun Studio? >> Yes that is exactly correct, when you build a python extension it will >> look at the way python was compiled and use those settings ( this is >> just like >> PERL ) >> >>> 2. How comes that a specific build machine path is used, i.e. Sun >>> Studio is installed by default in /opt/SUNWspro (I know that it can >>> be installed at the administrator's convenience and on the build >>> machines is installed in a /opt/studio directory but that is >>> another >>> story) >> >> Unfortunately, this is just the way it is handled, it is true for >> PERL and >> most if not all the other packages that have a pkg-config like >> structure. >> >> you should be able to work around this by setting the environment >> variables >> CC={path to cc} CXX={path to c++ (CC in SunStudio)} > > Is it really this bad? Then we should make Sun Studio 11 the default at > /opt/SUNWspro and let GAR switch flags on the detected compiler if there > is e. g. SS12 at that location. Please. This will work for those not using gar, for the moment. __ Peter From mwatters at opencsw.org Fri May 8 18:49:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:49:51 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: References: <4A045CA3.5070908@opencsw.org> Message-ID: <4A0462AF.7070508@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi, > > Am 08.05.2009 um 18:24 schrieb Mike Watters: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi Peter, >> Peter FELECAN wrote: >> >>> There are 2 issues: >>> >>> 1. I'm using gcc, so why cc is used? Because Python is generated with >>> Sun Studio? >> Yes that is exactly correct, when you build a python extension it will >> look at the way python was compiled and use those settings ( this is >> just like >> PERL ) >> >>> 2. How comes that a specific build machine path is used, i.e. Sun >>> Studio is installed by default in /opt/SUNWspro (I know that it can >>> be installed at the administrator's convenience and on the build >>> machines is installed in a /opt/studio directory but that is another >>> story) >> >> Unfortunately, this is just the way it is handled, it is true for >> PERL and >> most if not all the other packages that have a pkg-config like structure. >> >> you should be able to work around this by setting the environment >> variables >> CC={path to cc} CXX={path to c++ (CC in SunStudio)} > > Is it really this bad? Then we should make Sun Studio 11 the default at > /opt/SUNWspro and let GAR switch flags on the detected compiler if there > is e. g. SS12 at that location. That would be OK, I guess, but I think SOS12 should be the default. Right now SOS12 is not installed on solaris 8. I have tested it on a solaris 8 zone (from a solaris 10 global) and it seems to work OK. has anyone else tried this out? I installed SOS12 on solaris10 global and "shared" it to the solaris 8 zone. zoneadm -z s8 halt zonecfg -z s8 add fs set dir=/opt/SUNWspro set special=/opt/SUNWspro set type=lofs add options rw add options nodevices end commit exit zoneadm -z s8 boot - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEYq8ACgkQLrhmsXMSLxfysgCgs4l8laApYzNG/9T9ffJKeTFD ygAAn2t/yXpxtf36LC1E1+QHTAm+6eUG =pH7D -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri May 8 18:57:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 12:57:59 -0400 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> Message-ID: <1241801752-sup-8737@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Fri May 08 12:39:31 -0400 2009: > I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at > least, put it in /opt/csw/php4. I definitely agree that php5 should become the default. I argued last time that php4 should be removed from the catalog too, but Phil thought it was best to keep it at that time. If I changed my argument to 'remove it from the catalog, but leave it on the mirrors' would change your opinion Phil? Users would be free to manually download and install it, but it wouldn't be supported directly any more? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Fri May 8 18:59:47 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:59:47 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241801752-sup-8737@ntdws12.chass.utoronto.ca> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> Message-ID: <4A046503.4010702@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Peter FELECAN's message of Fri May 08 12:39:31 -0400 2009: >> I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at >> least, put it in /opt/csw/php4. > > I definitely agree that php5 should become the default. I argued last > time that php4 should be removed from the catalog too, but Phil > thought it was best to keep it at that time. If I changed my argument > to 'remove it from the catalog, but leave it on the mirrors' would > change your opinion Phil? Users would be free to manually download > and install it, but it wouldn't be supported directly any more? > I no longer see php4 core on the package page? I see the php4_* modules but no php or php4. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEZQMACgkQLrhmsXMSLxe0GgCg7Cz0iA8Ho5P4l/Z9IXAaGEDc NGoAn3PTi9tK36PiBx2u1g8iv5zP+aK9 =TuMu -----END PGP SIGNATURE----- From dam at opencsw.org Fri May 8 19:01:47 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 19:01:47 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A0462AF.7070508@opencsw.org> References: <4A045CA3.5070908@opencsw.org> <4A0462AF.7070508@opencsw.org> Message-ID: <499E4F07-9C1C-4193-A4E4-49E924D5DEF9@opencsw.org> Hi Mike, Am 08.05.2009 um 18:49 schrieb Mike Watters: > That would be OK, I guess, but I think SOS12 should be the default. > Right now SOS12 is not installed on solaris 8. SOS12 is not supported on Solaris 8, hence I don't think it is a good idea to use it. Best regards -- Dago From mwatters at opencsw.org Fri May 8 19:05:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 12:05:08 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: References: <4A045CA3.5070908@opencsw.org> Message-ID: <4A046644.8090701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Peter, Peter FELECAN wrote: > But this will work only if I use Sun Studio C and not gcc. > > I must confess that I don't like this situation of having more than a > compiler and doing acrobatics to handle weird issues. Well, I'm opting > to not provide some bindings. I have to agree with you, I use gar to build and it is still a problem. the subversion package, is/was my PITA for this problem. Perl, Python are built using Studio, Ruby is built with GCC I had to Hack up a "fix" to change the makefiles to be GCC computible. this worked without a problem. if you would like to see what I did, here is the link in gar. http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/subversion/trunk/Makefile?revision=4609&view=markup Lines 133 - 150 and Lines 194 - 200 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEZkQACgkQLrhmsXMSLxcmZwCdGII29c6Y6b5XMVCcNi8t0njC uDMAoLfSJ5eRl7bWANJidb5a78imGrii =c9BG -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 8 20:44:33 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 8 May 2009 11:44:33 -0700 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A046644.8090701@opencsw.org> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> Message-ID: <20090508184433.GA96779@bolthole.com> On Fri, May 08, 2009 at 12:05:08PM -0500, Mike Watters wrote: > Perl, Python are built using Studio, Ruby is built with GCC this is a problem. if ruby Can be built with studio, it Should be built with it. btw: make sure that you are using -mt flag for base things like python and ruby when you compile them please From bwalton at opencsw.org Fri May 8 20:51:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 14:51:19 -0400 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <20090508184433.GA96779@bolthole.com> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> Message-ID: <1241808516-sup-4065@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri May 08 14:44:33 -0400 2009: > this is a problem. if ruby Can be built with studio, it Should be built > with it. The ruby build fails with studio but passes with with gcc. It's an issue of the (somewhat nasty) way that Makefiles are generated for some of the extensions (without auto*). These are done on the fly, so it wasn't a simple post-configure step. It ends up passing flags to ld that ld doesn't like. I wasn't able to nail it down, but didn't fight it too hard, either. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Fri May 8 21:16:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 21:16:01 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <1241808516-sup-4065@ntdws12.chass.utoronto.ca> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> <1241808516-sup-4065@ntdws12.chass.utoronto.ca> Message-ID: <64E60A27-3644-4AF1-8A5F-AB13BFC4308A@opencsw.org> Hi Ben, Am 08.05.2009 um 20:51 schrieb Ben Walton: > The ruby build fails with studio but passes with with gcc. It's an > issue of the (somewhat nasty) way that Makefiles are generated for > some of the extensions (without auto*). These are done on the fly, so > it wasn't a simple post-configure step. It ends up passing flags to > ld that ld doesn't like. I wasn't able to nail it down, but didn't > fight it too hard, either. James gave me an incredibly useful advice: Interpose execution and rewrite arguments. For example to rewrite arguments to cc and CC you can write this (taken from http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/flac/trunk/Makefile > include gar/category.mk > > CC := $(abspath ./bin/cc) > CXX := $(abspath ./bin/CC) and then have heavy hacking (here, includes to relative pathes are reordered to be before include to absolute pathes): > :::::::::::::: > bin/CC > :::::::::::::: > #!/opt/csw/bin/perl -w > > use strict; > > my @oargs; > > my @includes = grep { /^-I/ } @ARGV; > @includes = ((grep { /^-I[^\/]/ } @includes),(grep { /^-I[\/]/ } > @includes)); > my @argswoincludes = grep { !/^-I/ } @ARGV; > > foreach (@ARGV) { > if( /^-I/ ) { > push @oargs, @includes, @argswoincludes; > last; > } > push @oargs, shift @argswoincludes; > } > > my $cc = "/opt/studio/SOS11/SUNWspro/bin/" . ($0 =~ /cc/ ? "cc" : > "CC"); > > print "$cc ", join( " ", @oargs ), "\n"; > exec( $cc, @oargs ); You could poke around stuff for "ld" also, of course... Best regards -- Dago From mwatters at opencsw.org Fri May 8 21:33:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 14:33:22 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <1241808516-sup-4065@ntdws12.chass.utoronto.ca> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> <1241808516-sup-4065@ntdws12.chass.utoronto.ca> Message-ID: <4A048902.30400@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ben, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri May 08 14:44:33 -0400 2009: >> this is a problem. if ruby Can be built with studio, it Should be built >> with it. > > The ruby build fails with studio but passes with with gcc. It's an > issue of the (somewhat nasty) way that Makefiles are generated for > some of the extensions (without auto*). These are done on the fly, so > it wasn't a simple post-configure step. It ends up passing flags to > ld that ld doesn't like. I wasn't able to nail it down, but didn't > fight it too hard, either. > > -Ben Ruby compiles fine with Studio 12. we could just skip solaris8 ;) ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEiQIACgkQLrhmsXMSLxdvfACgp8tm++TOFNjGowebZyjVEn+J UAgAn0eGOVvo9imxQjogqAXG85JPsy9j =a2BN -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri May 8 21:39:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 15:39:59 -0400 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A048902.30400@opencsw.org> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> <1241808516-sup-4065@ntdws12.chass.utoronto.ca> <4A048902.30400@opencsw.org> Message-ID: <1241811478-sup-1525@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Fri May 08 15:33:22 -0400 2009: Hi Mike, > Ruby compiles fine with Studio 12. Interesting! :) Thanks for testing that. > we could just skip solaris8 ;) ;) I can't, as that's still my production rails box!! I will do a little happy dance of joy when I no longer have any sol8 boxes around. It's been a workhorse, but it's tired now. If I get a few spare cycles, I'll try the interposing of a wrapper script to see about resolve the Studio 11 issues. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Fri May 8 21:52:29 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 14:52:29 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <4A046503.4010702@opencsw.org> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <4A046503.4010702@opencsw.org> Message-ID: <4A048D7D.2070807@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Phil, Mike Watters wrote: > Ben Walton wrote: >> Excerpts from Peter FELECAN's message of Fri May 08 12:39:31 -0400 2009: >>> I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at >>> least, put it in /opt/csw/php4. >> I definitely agree that php5 should become the default. I argued last >> time that php4 should be removed from the catalog too, but Phil >> thought it was best to keep it at that time. If I changed my argument >> to 'remove it from the catalog, but leave it on the mirrors' would >> change your opinion Phil? Users would be free to manually download >> and install it, but it wouldn't be supported directly any more? > > > I no longer see php4 core on the package page? I see the php4_* modules but no > php or php4. > What is the verdict on php4 vs php5? I found the php binaries in php4_cgi should I move php5 to the default php? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEjX0ACgkQLrhmsXMSLxc8MQCgvBGa36bpLZnlk4/JrIjzl+uB /bwAoJlk/xZ7g6Hv82iVNkEIBtqaFdJe =syKm -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 8 23:23:22 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 8 May 2009 14:23:22 -0700 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241801752-sup-8737@ntdws12.chass.utoronto.ca> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> Message-ID: <20090508212322.GD63097@bolthole.com> On Fri, May 08, 2009 at 12:57:59PM -0400, Ben Walton wrote: > I argued last > time that php4 should be removed from the catalog too, but Phil > thought it was best to keep it at that time. do you remember the specifics of WHY? i have forgotten. > If I changed my argument > to 'remove it from the catalog, but leave it on the mirrors' would > change your opinion Phil? That makes no sense. if we keep it, it stays in the catalog. its worth discussing what should be /opt/csw/bin/php though. I am in favour of it being php5. it doesnt look like there are any dependancies of OURS that use it. However, it would need to be carefully announced ahead of time to users. From bwalton at opencsw.org Sat May 9 02:59:46 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 20:59:46 -0400 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <20090508212322.GD63097@bolthole.com> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> Message-ID: <1241826545-sup-6478@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri May 08 17:23:22 -0400 2009: > do you remember the specifics of WHY? > i have forgotten. You didn't think it was hurting anything to keep it as is. My view is that since it's dead upstream and our release is behind their last security release it should be dropped...I don't even think upping to the last 4.4.9 release is a reasonable time investment at this point. > > If I changed my argument to 'remove it from the catalog, but > > leave it on the mirrors' would change your opinion Phil? > > That makes no sense. > if we keep it, it stays in the catalog. I was just trying to provide a way for people that want it (for whatever reason) to be able to easily grab it without it being a supported package with the csw tool chain. Drop it is my official stance. > its worth discussing what should be /opt/csw/bin/php though. > > I am in favour of it being php5. That's my vote, too. I don't even think this is a real question at this point. > it doesnt look like there are any dependancies of OURS that use it. > However, it would need to be carefully announced ahead of time to users. Agreed. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Sat May 9 03:56:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 20:56:49 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241826545-sup-6478@ntdws12.chass.utoronto.ca> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> <1241826545-sup-6478@ntdws12.chass.utoronto.ca> Message-ID: <4A04E2E1.3020701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Philip Brown's message of Fri May 08 17:23:22 -0400 2009: >> do you remember the specifics of WHY? >> i have forgotten. > > You didn't think it was hurting anything to keep it as is. My view is > that since it's dead upstream and our release is behind their last > security release it should be dropped...I don't even think upping to > the last 4.4.9 release is a reasonable time investment at this point. > >>> If I changed my argument to 'remove it from the catalog, but >>> leave it on the mirrors' would change your opinion Phil? >> That makes no sense. >> if we keep it, it stays in the catalog. > > I was just trying to provide a way for people that want it (for > whatever reason) to be able to easily grab it without it being a > supported package with the csw tool chain. Drop it is my official > stance. > >> its worth discussing what should be /opt/csw/bin/php though. >> >> I am in favour of it being php5. > > That's my vote, too. I don't even think this is a real question at > this point. > >> it doesnt look like there are any dependancies of OURS that use it. >> However, it would need to be carefully announced ahead of time to users. > > Agreed. OK, Phil, can you put out a press release regarding php4 decommission. I will handle php5. Here is my plan. I will keep the 5.2.9 version "as is" ... after the bug-fix and feature request I am currently compiling. I will make an announcement that as of release 5.3 ... road map for Q2 2009 ... we are moving the installation to the "default" location. thoughts? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoE4uEACgkQLrhmsXMSLxeHYwCgkNx1kQ0uXbvHGK4hF12Ivp+G NuMAoIXI2okjW6CYWLoiONrhkQZgOisA =GdnY -----END PGP SIGNATURE----- From ihsan at opencsw.org Sat May 9 12:03:26 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 12:03:26 +0200 Subject: [csw-maintainers] Odd issues with the new IMAP In-Reply-To: <4A045DB6.90104@opencsw.org> References: <4A045DB6.90104@opencsw.org> Message-ID: <4A0554EE.7050001@opencsw.org> Hello Mike, Am 8.5.2009 18:28 Uhr, Mike Watters schrieb: > I am using thunderbird/enigmail/imap for my opencsw account. > Since the upgrade, I am getting Messages that are showing up > at different times. I'm using the same combination. > for example: at 11:00 my time I received 4 messages from earlier in the day > 08:32 08:48 09:23 and 10:47 "mytime" > > this is not really a problem, just Odd and only after the imap upgrade. > > anyone else experiencing issues similar? I haven't experienced something similar. Did you had a look into the mail header? Maybe the mails were stuck somewhere. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From rupert at opencsw.org Sat May 9 13:47:00 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 9 May 2009 13:47:00 +0200 Subject: [csw-maintainers] mod_wsgi, how to get it from testing into release? Message-ID: <6af4270905090447g1841defbj3519fc8f481eb27e@mail.gmail.com> how does one get mod_wsgi from http://mirror.opencsw.org/testing.html into http://opencsw.org/packages ? we tested it successful here by: pkgutil -U pkgutil -u pkgutil -t http://mirror.opencsw.org/opencsw/testing -i ap2_modwsgi. the last command offered to upgrade other packages as well, but i did not says yes to stay on the current productive version. then we did a successful test with: - native installed trac-0.11.4, using csw subversion-1.6.1, displaying svn-1.5 format repositories. - native installed mercurial-1.1 rupert. From rupert at opencsw.org Sat May 9 13:49:25 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 9 May 2009 13:49:25 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> Message-ID: <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> hi, while subversion-1.6.1 works fine with the old repository format, we hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after upgrading to 1.6 repository format. the fix will be in 1.6.2. i'll give it a try. rupert. ------------------------------------------------------------------------------ From: "Hyrum K. Wright" Date: Thu, 7 May 2009 11:49:25 -0500 Local: Thurs, May 7 2009 6:49 pm Subject: 1.6.2 tarballs up for signing/testing I'm pleased to announce that Subversion 1.6.2 is up for testing and signing. ?The magic revision is r37639. ?You can find the tarballs here: http://orac.ece.utexas.edu/pub/svn/1.6.2/ Please be sure to test the bindings. You know the drill: signatures from full committers back to me, and enthusiastic tester feedback is welcome. ?At this point, this candidate is not yet blessed for wide release, so please don't make it available to people not interested in test-driving the new release. Distro package maintainers, please do NOT include any pre-release builds, even blessed, into operating system distros. ?The reasons for not doing so were very eloquently outlined by Karl in a mail, which is summarized at the above address. The quick version is: we don't guarantee compatibility between the pre- releases and the final release, so if people install the release candidate, all their repositories and working copies might break irreparably when they upgrade to 1.6.2 proper. ?We don't want that kind of bad publicity, and neither do you. -Hyrum ------------------------------------------------------ http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessag... From mwatters at opencsw.org Sat May 9 16:19:36 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 09 May 2009 09:19:36 -0500 Subject: [csw-maintainers] Odd issues with the new IMAP In-Reply-To: <4A0554EE.7050001@opencsw.org> References: <4A045DB6.90104@opencsw.org> <4A0554EE.7050001@opencsw.org> Message-ID: <4A0590F8.6030205@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ihsan, Ihsan Dogan wrote: > > I haven't experienced something similar. Did you had a look into the > mail header? Maybe the mails were stuck somewhere. I didn't check the headers, I guess I should have, but it is not a problem I just thought it odd. I have really only seen it once, right after the conversion. if I do "catch" it again, I will double check the headers and gather as much info as possible. just so we can verify it is not an issue with dovecot, I doubt very much it would be a dovecot issue. but, my first thought was a minor bug in the re-indexing dovecot does. possibly holding new mail until the re-index was complete. Just wanted to see if anyone else had seen that. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoFkPcACgkQLrhmsXMSLxd2FQCgt+IXut/1vkkKWHwg45rLSpyG 0H4An2GMu7mpV2eRdxhw6HmTlNs7uXEs =Tq/N -----END PGP SIGNATURE----- From mwatters at opencsw.org Sat May 9 16:30:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 09 May 2009 09:30:03 -0500 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> Message-ID: <4A05936B.3020001@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > hi, > > while subversion-1.6.1 works fine with the old repository format, we > hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after > upgrading to 1.6 repository format. > > the fix will be in 1.6.2. i'll give it a try. This sounds eerily similar to http://www.opencsw.org/mantis/view.php?id=3661 http://subversion.tigris.org/issues/show_bug.cgi?id=3387 I will patch up 1.6.1 with the fix for our bug 3361. if you wouldn't mind testing that when released against your bug? if it does not fix it, please file a new opencsw bug. I hope to have the fix for subversion done next week. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw =Du0U -----END PGP SIGNATURE----- From ihsan at opencsw.org Sat May 9 17:30:40 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 17:30:40 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp Message-ID: <4A05A1A0.3090702@opencsw.org> Hello, Back in December we have been talking about a summer camp in Norway. I think it's now time to talk about the topics for the summer camp. I would like to open the discussion now. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat May 9 17:33:19 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 17:33:19 +0200 Subject: [csw-maintainers] Webmail: Squirrelmail now active again Message-ID: <4A05A23F.40704@opencsw.org> Hello, After having a lot of troubles with Roundcube, I switch back to Squirrelmail. --> https://mail.opencsw.org/ If you still would like to use Roundcube, please point your browser to https://mail.opencsw.org/roundcubemail/ . Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat May 9 17:39:31 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 17:39:31 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <4A05A3B3.3070508@opencsw.org> Am 9.5.2009 17:30 Uhr, Ihsan Dogan schrieb: > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. A possible topic would be the website. There is already some work done and it would be a good chance to move forward. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From william at wbonnet.net Sat May 9 19:24:18 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 09 May 2009 19:24:18 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <4A05BC42.4080201@wbonnet.net> Hi > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. > I would be interested in particpating. It will depend of the date... I should be able to find some time in July (end) or first weeks of August. cheers W. From Murray.Jensen at csiro.au Sun May 10 10:47:20 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Sun, 10 May 2009 18:47:20 +1000 Subject: [csw-maintainers] suncsw domain names expiring Message-ID: <9637.1241945240@gerd> Hi CSW people, all the domain names I registered in anticipation of the CSW split are about to expire - namely sun{,-}csw.{org,net,com}. I was just going to allow them to expire, but I thought I should let everyone know. Cheers! Murray... -- Murray Jensen, CSIRO Materials Science and Engineering Phone: +61 3 9545 2075 Private Bag 33, Clayton South 3169, Australia Fax: +61 3 9544 1128 Internet: Murray.Jensen at csiro.au To the extent permitted by law, CSIRO does not represent, warrant and/or guarantee that the integrity of this communication has been maintained or that the communication is free of errors, virus, interception or interference. The information contained in this e-mail may be confidential or privileged. Any unauthorised use or disclosure is prohibited. If you have received this e-mail in error, please delete it immediately and notify Murray Jensen on +61 3 9545 2075. Thank you. From dam at opencsw.org Sun May 10 15:38:03 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 15:38:03 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website Message-ID: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Hi, I just found out that there is a Java Chatclient at Freenode. How about putting on the webpage to let interested people join immediately? Best regards -- Dago From dam at opencsw.org Sun May 10 15:48:41 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 15:48:41 +0200 Subject: [csw-maintainers] suncsw domain names expiring In-Reply-To: <9637.1241945240@gerd> References: <9637.1241945240@gerd> Message-ID: Hi Murray, Am 10.05.2009 um 10:47 schrieb : > Hi CSW people, all the domain names I registered in anticipation of > the CSW > split are about to expire - namely sun{,-}csw.{org,net,com}. I was > just going > to allow them to expire, but I thought I should let everyone know. > Cheers! That is IMHO okay. I won't extend suncsw.de either. They have done their purpose and opencsw.org is IMHO just right. Best regards -- Dago From bwalton at opencsw.org Sun May 10 16:30:35 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 10 May 2009 10:30:35 -0400 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Message-ID: <1241965727-sup-4496@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Sun May 10 09:38:03 -0400 2009: > I just found out that there is a Java Chatclient at Freenode. How about > putting > > on the webpage to let interested people join immediately? I think it'd be good. It might be a good question forum for some. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Sun May 10 18:03:42 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 10 May 2009 18:03:42 +0200 Subject: [csw-maintainers] Is Hudson still available? Message-ID: <4A06FADE.8060707@opencsw.org> Hi, just wondering: Is Hudson still available / functional? When i go to http://buildfarm.opencsw.org/hudson/ i just get a 503: Service Temporarily Unavailable The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later. Sebastian From trygvis at opencsw.org Sun May 10 18:52:51 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 18:52:51 +0200 Subject: [csw-maintainers] Is Hudson still available? In-Reply-To: <4A06FADE.8060707@opencsw.org> References: <4A06FADE.8060707@opencsw.org> Message-ID: <4A070663.3050807@opencsw.org> Sebastian Kayser wrote: > Hi, > > just wondering: Is Hudson still available / functional? When i go to > http://buildfarm.opencsw.org/hudson/ i just get a 503: > > Service Temporarily Unavailable > > The server is temporarily unable to service your request due to > maintenance downtime or capacity problems. Please try again later. The instance has probably been rebooted, should be working again now. -- Trygve From trygvis at opencsw.org Sun May 10 18:54:28 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 18:54:28 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Message-ID: <4A0706C4.4050105@opencsw.org> Dagobert Michelsen wrote: > Hi, > > I just found out that there is a Java Chatclient at Freenode. How about > putting > > on the webpage to let interested people join immediately? I like the idea, but think mibbit[1] is a better client as it don't require anything other than a browser. The applet integration on Solaris has been a bit shady in my experience. [1]: http://www.mibbit.com/ -- Trygve From trygvis at opencsw.org Sun May 10 18:57:22 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 18:57:22 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <4A070772.2010206@opencsw.org> Ihsan Dogan wrote: > Hello, > > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. I'm interested, but I doubt Norway is the best place to go. It will require many people to travel far and stuff is generally considered expensive in Norway. I can get access to conference rooms and work facilities with connectivity for free though. -- Trygve From ihsan at opencsw.org Sun May 10 19:28:06 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:28:06 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A070772.2010206@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> Message-ID: <4A070EA6.3060307@opencsw.org> Am 10.5.2009 18:57 Uhr, Trygve Laugst?l schrieb: >> Back in December we have been talking about a summer camp in Norway. I >> think it's now time to talk about the topics for the summer camp. I >> would like to open the discussion now. > I'm interested, but I doubt Norway is the best place to go. It will > require many people to travel far and stuff is generally considered > expensive in Norway. I can get access to conference rooms and work > facilities with connectivity for free though. The pricing looks like in Switzerland. For me personally it doesn't matter, but if it should take place somewhere else, if someone would volunteer now. We should know the place and the date 2 months in advance. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 10 19:29:22 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:29:22 +0200 Subject: [csw-maintainers] suncsw domain names expiring In-Reply-To: References: <9637.1241945240@gerd> Message-ID: <4A070EF2.8050402@opencsw.org> Am 10.5.2009 15:48 Uhr, Dagobert Michelsen schrieb: >> Hi CSW people, all the domain names I registered in anticipation of >> the CSW >> split are about to expire - namely sun{,-}csw.{org,net,com}. I was >> just going >> to allow them to expire, but I thought I should let everyone know. >> Cheers! > That is IMHO okay. I won't extend suncsw.de either. They have done their > purpose > and opencsw.org is IMHO just right. I'm going to remove suncsw.org from our name server. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 10 19:45:30 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:45:30 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0706C4.4050105@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A0706C4.4050105@opencsw.org> Message-ID: <4A0712BA.9070504@opencsw.org> Am 10.5.2009 18:54 Uhr, Trygve Laugst?l schrieb: >> I just found out that there is a Java Chatclient at Freenode. How about >> putting >> >> on the webpage to let interested people join immediately? > > I like the idea, but think mibbit[1] is a better client as it don't > require anything other than a browser. The applet integration on Solaris > has been a bit shady in my experience. > > [1]: http://www.mibbit.com/ Does this need anything to be installed on the web server? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 10 19:48:05 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:48:05 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Message-ID: <4A071355.3020204@opencsw.org> Am 10.5.2009 15:38 Uhr, Dagobert Michelsen schrieb: > I just found out that there is a Java Chatclient at Freenode. How about > putting > > on the webpage to let interested people join immediately? For the internal use, it might be handy to have a Instant Messenger contact list. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From trygvis at opencsw.org Sun May 10 20:11:48 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 20:11:48 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0712BA.9070504@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A0706C4.4050105@opencsw.org> <4A0712BA.9070504@opencsw.org> Message-ID: <4A0718E4.9090400@opencsw.org> Ihsan Dogan wrote: > Am 10.5.2009 18:54 Uhr, Trygve Laugst?l schrieb: > >>> I just found out that there is a Java Chatclient at Freenode. How about >>> putting >>> >>> on the webpage to let interested people join immediately? >> I like the idea, but think mibbit[1] is a better client as it don't >> require anything other than a browser. The applet integration on Solaris >> has been a bit shady in my experience. >> >> [1]: http://www.mibbit.com/ > > Does this need anything to be installed on the web server? No, we don't have to do anything. Just explain how to use it to connect to our room, like server name and channel name. I've never used it, but it is supposedly nice. -- Trygve From trygvis at inamo.no Sun May 10 20:11:02 2009 From: trygvis at inamo.no (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 20:11:02 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A071355.3020204@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> Message-ID: <4A0718B6.2000903@inamo.no> Ihsan Dogan wrote: > Am 10.5.2009 15:38 Uhr, Dagobert Michelsen schrieb: > >> I just found out that there is a Java Chatclient at Freenode. How about >> putting >> >> on the webpage to let interested people join immediately? > > For the internal use, it might be handy to have a Instant Messenger > contact list. I don't mind sharing my MSN or Jabber IDs in a private file somwhere. -- Trygve From dam at opencsw.org Sun May 10 20:28:09 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 20:28:09 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <49A5CFAA.2010806@wbonnet.net> References: <49A5CFAA.2010806@wbonnet.net> Message-ID: <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> Hi, I have now rebuild all the X11 packages from William, pixman, cairo, pango, libxft2, libatk and gtk+ in GAR with 32/64 bit. The packages are not in testing as they have been build on build8st instead of the official build machines. The following dependencies are still missing, but I consider them not that relevant, so they may be added later: - openjpeg (uses statical Makefile optimized for Linux) (requires defining loads of variables in the Makefile) - poppler (broken runtimelinkerpath, needs openjpeg) (needs some inspection, shouldn't be too hard) - spectre (needs 64 bit libgs.o, which needs pixman etc., can be done after general package release in a second step) All packages have been installed on build8st, please try compiling and testing packages against them. After this was successfull I am going to package and release them one-by-one as there are massive dependencies between them. I don't want to take over packages from William, how about a technical user "gnometeam" releasing the packages and receiving the mails with all gnome maintainers on it? Am 26.02.2009 um 00:09 schrieb William Bonnet: > I have moved the X11 stuff to /opt/csw/X11 (but licenses which are > under > /opt/csw/share/$(GARNAME). The existing libxrender package puts shared libraries in /opt/csw/lib: I guess if they are relocated to /opt/csw/x11/lib they can no longer be found. It would be however good to bundle X11 libs in /opt/csw/lib/ X11. Make links from lib/X11 to lib/, issue bug reports and remove the links when all bugs have been closed? Best regards -- Dago From skayser at opencsw.org Sun May 10 21:42:01 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 10 May 2009 21:42:01 +0200 Subject: [csw-maintainers] Is Hudson still available? In-Reply-To: <4A070663.3050807@opencsw.org> References: <4A06FADE.8060707@opencsw.org> <4A070663.3050807@opencsw.org> Message-ID: <4A072E09.8010203@opencsw.org> Trygve Laugst?l wrote: > Sebastian Kayser wrote: >> just wondering: Is Hudson still available / functional? When i go to >> http://buildfarm.opencsw.org/hudson/ i just get a 503: >> >> Service Temporarily Unavailable >> >> The server is temporarily unable to service your request due to >> maintenance downtime or capacity problems. Please try again later. > > The instance has probably been rebooted, should be working again now. Web page is accessible again, but both "build executors" are marked as offline (with several jobs pending). Could you have a look at it please? Sebastian From phil at bolthole.com Sun May 10 21:45:29 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 10 May 2009 12:45:29 -0700 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> Message-ID: <20090510194529.GD96486@bolthole.com> On Sun, May 10, 2009 at 08:28:09PM +0200, Dagobert Michelsen wrote: > I don't want to take over packages from William, how about a technical > user "gnometeam" releasing the packages and receiving the mails > with all gnome maintainers on it? trouble with that is, there's a tendancy for one person to always think, "well, the other person will update it... From mwatters at opencsw.org Sun May 10 22:37:06 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 10 May 2009 15:37:06 -0500 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <20090510194529.GD96486@bolthole.com> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> Message-ID: <4A073AF2.2000309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Sun, May 10, 2009 at 08:28:09PM +0200, Dagobert Michelsen wrote: >> I don't want to take over packages from William, how about a technical >> user "gnome team" releasing the packages and receiving the mails >> with all gnome maintainers on it? > > trouble with that is, there's a tendency for one person to always think, > "well, the other person will update it... I have to agree, reluctantly, with Philip. However, I do think a separate mailing list for the Big Projects, like GNOME, KDE... etc. with all the maintainers on them. that way those that would like to follow the progress can, and those that do not, can filter out the incoming mail. This would also allow the other maintainers to follow the progress and make it easier to contribute free cycles if they so choose. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoHOvIACgkQLrhmsXMSLxdEtgCgjm6EP9BJQ8gnYz8+395R3Osl WpgAoJt1BmF4LfHOzXGAtNWhT95s9Uai =HIeO -----END PGP SIGNATURE----- From william at wbonnet.net Sun May 10 22:45:42 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 10 May 2009 22:45:42 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <20090510194529.GD96486@bolthole.com> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> Message-ID: <4A073CF6.6070201@wbonnet.net> Hi >> I don't want to take over packages from William, how about a technical >> user "gnometeam" releasing the packages and receiving the mails >> with all gnome maintainers on it? >> > > trouble with that is, there's a tendancy for one person to always think, > "well, the other person will update it... > I've been silent for a long time. Sorry for that. I proposed the idea of teams several times, and we should really consider to apply it. I don't agree with the argument of "well, the other person will update it...". We are going to create team of twenty persons, and communication will be easy. Even if sometime someone is off for different reasons, like an "office overload", hollidays, or whatever, it is still possible to communicate within the team, or to announce we are off for a few weeks with internet. The gnome team is one of the team we need to set up soon, mozilla should follow i think (and maybe kde). I am right now fixing bugs on firefox and finishing TB package in GAR, but it takes time, a lot of time, and i may be overloaded some time when it comes to release some security fix. I would appreciate someone else than me could do it. Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From william at wbonnet.net Sun May 10 22:47:27 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 10 May 2009 22:47:27 +0200 Subject: [csw-maintainers] Mantis login problem Message-ID: <4A073D5F.4030302@wbonnet.net> Hi It seems like i cannot login into mantis. Any known problem ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Sun May 10 22:49:39 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 10 May 2009 22:49:39 +0200 Subject: [csw-maintainers] Mantis login problem In-Reply-To: <4A073D5F.4030302@wbonnet.net> References: <4A073D5F.4030302@wbonnet.net> Message-ID: <4A073DE3.3080702@opencsw.org> William Bonnet wrote: > It seems like i cannot login into mantis. Any known problem ? Works for me. Any error messages? Sebastian From william at wbonnet.net Sun May 10 23:03:40 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 10 May 2009 23:03:40 +0200 Subject: [csw-maintainers] Mantis login problem In-Reply-To: <4A073DE3.3080702@opencsw.org> References: <4A073D5F.4030302@wbonnet.net> <4A073DE3.3080702@opencsw.org> Message-ID: <4A07412C.7090804@wbonnet.net> Hi Sebastian >> It seems like i cannot login into mantis. Any known problem ? >> > > Works for me. Any error messages? > Nope... login button is inactive I tested with a different browser and it works with seamonkey ! I am going to have a look to my environement cheers -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From phil at bolthole.com Sun May 10 23:20:39 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 10 May 2009 14:20:39 -0700 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A073CF6.6070201@wbonnet.net> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> Message-ID: <20090510212039.GA28401@bolthole.com> On Sun, May 10, 2009 at 10:45:42PM +0200, William Bonnet wrote: > I don't agree with the argument of "well, the other person will update > it...". We are going to create team of twenty persons, and communication > will be easy. communication IS easy. responsability (in an organization consisting of unpaid people) is hard. > follow i think (and maybe kde). I am right now fixing bugs on firefox > and finishing TB package in GAR, but it takes time, a lot of time, and i > may be overloaded some time when it comes to release some security fix. > I would appreciate someone else than me could do it. that is doable. ON the flip side, that does not require having group "ownership" of an individual package. all that is required, is being known as an oficial member of "the gnome team", and then for the release manager (ie: me) to allow automatic takeover of gnome packages, between gnome team members. From dam at opencsw.org Sun May 10 23:27:24 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 23:27:24 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <20090510212039.GA28401@bolthole.com> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> Message-ID: <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> Hi, Am 10.05.2009 um 23:20 schrieb Philip Brown: > On Sun, May 10, 2009 at 10:45:42PM +0200, William Bonnet wrote: >> I don't agree with the argument of "well, the other person will >> update >> it...". We are going to create team of twenty persons, and >> communication >> will be easy. > > communication IS easy. responsability (in an organization consisting > of > unpaid people) is hard. > >> follow i think (and maybe kde). I am right now fixing bugs on firefox >> and finishing TB package in GAR, but it takes time, a lot of time, >> and i >> may be overloaded some time when it comes to release some security >> fix. >> I would appreciate someone else than me could do it. > > that is doable. ON the flip side, that does not require having group > "ownership" of an individual package. > all that is required, is being known as an oficial member of "the > gnome > team", and then for the release manager (ie: me) to allow automatic > takeover of gnome packages, between gnome team members. Ok then. Please someone take a look at build8st and I need a decision on linking X11/lib back to lib/ for the interimtime. Then the packages will be released. Ihsan: Please test rrdtool against the current cairo and see if the error goes away. Bets regards -- Dagp From ihsan at opencsw.org Sun May 10 23:30:38 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 23:30:38 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0718B6.2000903@inamo.no> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0718B6.2000903@inamo.no> Message-ID: <4A07477E.2030607@opencsw.org> Am 10.5.2009 20:11 Uhr, Trygve Laugst?l schrieb: >>> I just found out that there is a Java Chatclient at Freenode. How about >>> putting >>> >>> on the webpage to let interested people join immediately? >> >> For the internal use, it might be handy to have a Instant Messenger >> contact list. > > I don't mind sharing my MSN or Jabber IDs in a private file somwhere. Ok, I will try to create such a list. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From rupert at opencsw.org Sun May 10 23:31:41 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 10 May 2009 23:31:41 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <4A05936B.3020001@opencsw.org> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> <4A05936B.3020001@opencsw.org> Message-ID: <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> you are right, i put a hint also into the mantis issue. i tested with the upcoming 1.6.2 release, and it works. i left it on testing for now, but we should remove it as soon as possible (http://mirror.opencsw.org/testing/subversion-1.6.2,REV=2009.05.09-SunOS5.8-sparc-UNCOMMITTED.pkg.gz ). as 1.6.2 is already tagged we might wait the couple of days for the final 1.6.2 release instead of patching 1.6.1. rupert. On Sat, May 9, 2009 at 16:30, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > rupert THURNER wrote: >> hi, >> >> while subversion-1.6.1 works fine with the old repository format, we >> hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after >> upgrading to 1.6 repository format. >> >> the fix will be in 1.6.2. i'll give it a try. > > This sounds eerily similar to > http://www.opencsw.org/mantis/view.php?id=3661 > http://subversion.tigris.org/issues/show_bug.cgi?id=3387 > > I will patch up 1.6.1 with the fix for our bug 3361. ?if you wouldn't mind > testing that when released against your bug? > > if it does not fix it, please file a new opencsw bug. > I hope to have the fix for subversion done next week. > > - -- > Thanks, > Mike > > "Any intelligent fool can make things bigger, more complex, > and more violent. ?It takes a touch of genius -- and a lot of courage -- > to move in the opposite direction." > > * Albert Einstein 1879 - 1955 > ? ?US German-born Theoretical Physicist > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA > 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw > =Du0U > -----END PGP SIGNATURE----- > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > From rupert at opencsw.org Sun May 10 23:39:31 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 10 May 2009 23:39:31 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> <4A05936B.3020001@opencsw.org> <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> Message-ID: <6af4270905101439j6eee7dc2je4f5fe796515feb9@mail.gmail.com> because we are just talking about svn-1.6.X. one of the main features is that "bad clients" are now filtered out. but, this may lead to difficulties with non-versioned properties containing mixed end of line styles. they could be caused by eclipse subclipse e.g. when changing the svn log message. see http://svn.haxx.se/users/archive-2009-03/0797.shtml for details. rupert. On Sun, May 10, 2009 at 23:31, rupert THURNER wrote: > you are right, i put a hint also into the mantis issue. > > i tested with the upcoming 1.6.2 release, and it works. i left it on > testing for now, but we should remove it as soon as possible > (http://mirror.opencsw.org/testing/subversion-1.6.2,REV=2009.05.09-SunOS5.8-sparc-UNCOMMITTED.pkg.gz > ). > > as 1.6.2 is already tagged we might wait the couple of days for the > final 1.6.2 release instead of patching 1.6.1. > > rupert. > > > On Sat, May 9, 2009 at 16:30, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> rupert THURNER wrote: >>> hi, >>> >>> while subversion-1.6.1 works fine with the old repository format, we >>> hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after >>> upgrading to 1.6 repository format. >>> >>> the fix will be in 1.6.2. i'll give it a try. >> >> This sounds eerily similar to >> http://www.opencsw.org/mantis/view.php?id=3661 >> http://subversion.tigris.org/issues/show_bug.cgi?id=3387 >> >> I will patch up 1.6.1 with the fix for our bug 3361. ?if you wouldn't mind >> testing that when released against your bug? >> >> if it does not fix it, please file a new opencsw bug. >> I hope to have the fix for subversion done next week. >> >> - -- >> Thanks, >> Mike >> >> "Any intelligent fool can make things bigger, more complex, >> and more violent. ?It takes a touch of genius -- and a lot of courage -- >> to move in the opposite direction." >> >> * Albert Einstein 1879 - 1955 >> ? ?US German-born Theoretical Physicist >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.9 (GNU/Linux) >> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org >> >> iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA >> 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw >> =Du0U >> -----END PGP SIGNATURE----- >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers >> > From ihsan at opencsw.org Sun May 10 23:49:48 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 23:49:48 +0200 Subject: [csw-maintainers] IM Contacts Message-ID: <4A074BFC.503@opencsw.org> Hello, I'm trying to make a a list with the Instant Messanging contact of the OpenCSW maintainers. This list will be not public and only accessible to the maintainers. Please fill out this form and send it back to me personally (the list is public). AIM / ICQ / iChat : Jabber / Google Talk : MSN : Yahoo : Skype : Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 03:48:07 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 03:48:07 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: <4A077AE6.8050403@opencsw.org> References: <4A077AE6.8050403@opencsw.org> Message-ID: <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Hi, Am 11.05.2009 um 03:09 schrieb Mike Watters: > Please upgrade the buildfarm All buildservers are upgraded to current/ now. Please stand by. This include GCC :-) Best regards -- Dago From phil at bolthole.com Mon May 11 04:34:13 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 10 May 2009 19:34:13 -0700 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <4A04E2E1.3020701@opencsw.org> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> <1241826545-sup-6478@ntdws12.chass.utoronto.ca> <4A04E2E1.3020701@opencsw.org> Message-ID: <20090511023413.GA83321@bolthole.com> On Fri, May 08, 2009 at 08:56:49PM -0500, Mike Watters wrote: > > OK, Phil, can you put out a press release regarding php4 decommission. I shall do so. and lets be specific: we are "decommissioning" it from being /opt/csw/bin/php. The package will still be around, in legacy mode. This will require a repackage of the thing, to remove the old /opt/csw/bin/php reference. However, to "properly" retire php4, we really should repackage it at the last stable released version as well. Any volunteers? please? in theory, I could do it myself... but realistically speaking, if this is going to be waiting on me... it culd be waiting a loong time. From mwatters at opencsw.org Mon May 11 05:02:19 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 10 May 2009 22:02:19 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <20090511023413.GA83321@bolthole.com> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> <1241826545-sup-6478@ntdws12.chass.utoronto.ca> <4A04E2E1.3020701@opencsw.org> <20090511023413.GA83321@bolthole.com> Message-ID: <4A07953B.9070808@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Fri, May 08, 2009 at 08:56:49PM -0500, Mike Watters wrote: >> OK, Phil, can you put out a press release regarding php4 decommission. > > I shall do so. > > and lets be specific: we are "decommissioning" it from being > /opt/csw/bin/php. > The package will still be around, in legacy mode. > > This will require a repackage of the thing, to remove the old > /opt/csw/bin/php reference. > > However, to "properly" retire php4, we really should repackage it at the > last stable released version as well. > > > Any volunteers? please? I will do it, theoretically I should be able to use the php5 recipe with just a few minor modifications. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoHlTsACgkQLrhmsXMSLxcnLACgsCrvpZVDQLVDRimM+bJdHNvI mh0AoLFbFZITycQQP+EVBOnHezrSVJ3s =eqpt -----END PGP SIGNATURE----- From dam at opencsw.org Mon May 11 05:25:06 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 05:25:06 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: <8E3F475C-2A34-4AE1-906B-652DFCF1E2E7@opencsw.org> Hi, Am 11.05.2009 um 03:48 schrieb Dagobert Michelsen: > Am 11.05.2009 um 03:09 schrieb Mike Watters: >> Please upgrade the buildfarm > > All buildservers are upgraded to current/ now. Please stand by. > This include GCC :-) Done. This excludes build8st and build8xt. Best regards -- Dago From pfelecan at opencsw.org Mon May 11 11:06:57 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 11:06:57 +0200 Subject: [csw-maintainers] IM Contacts In-Reply-To: <4A074BFC.503@opencsw.org> (Ihsan Dogan's message of "Sun\, 10 May 2009 23\:49\:48 +0200") References: <4A074BFC.503@opencsw.org> Message-ID: Ihsan Dogan writes: > Hello, > > I'm trying to make a a list with the Instant Messanging contact of the > OpenCSW maintainers. This list will be not public and only accessible to > the maintainers. How you ensure that the list is not public and accessible only to maintainers? -- Peter From pfelecan at opencsw.org Mon May 11 11:09:36 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 11:09:36 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> (Dagobert Michelsen's message of "Mon\, 11 May 2009 03\:48\:07 +0200") References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: Dagobert Michelsen writes: > Am 11.05.2009 um 03:09 schrieb Mike Watters: >> Please upgrade the buildfarm > > All buildservers are upgraded to current/ now. Please stand by. > This include GCC :-) Is this reasonable when the packages didn't hit yet the mirrors? My only issue is that I cannot reproduce the same environment on my build stack. -- Peter From dam at opencsw.org Mon May 11 12:24:57 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 12:24:57 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: Hi Peter, Am 11.05.2009 um 11:09 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >> Am 11.05.2009 um 03:09 schrieb Mike Watters: >>> Please upgrade the buildfarm >> >> All buildservers are upgraded to current/ now. Please stand by. >> This include GCC :-) > > Is this reasonable when the packages didn't hit yet the mirrors? My > only > issue is that I cannot reproduce the same environment on my build > stack. The packages have been pushed to rsync.opencsw.org, all mirrors synchronize from this one, including mirror.opencsw.org. The farm is only updated with released packages. mirror.opencsw.org is updated every 6 hours and manually by me if something is needed fast. You can always pkg-get from that one to reproduce issues of supercurrent packages. Best regards -- Dago From ihsan at opencsw.org Mon May 11 12:29:39 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 12:29:39 +0200 Subject: [csw-maintainers] IM Contacts In-Reply-To: References: <4A074BFC.503@opencsw.org> Message-ID: <4A07FE13.4090601@opencsw.org> Am 11.5.2009 11:06 Uhr, Peter FELECAN schrieb: >> I'm trying to make a a list with the Instant Messanging contact of the >> OpenCSW maintainers. This list will be not public and only accessible to >> the maintainers. > How you ensure that the list is not public and accessible only to > maintainers? I was thinking to do it with mod_auth_pam. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From pfelecan at opencsw.org Mon May 11 12:33:11 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 12:33:11 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: (Dagobert Michelsen's message of "Mon\, 11 May 2009 12\:24\:57 +0200") References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > Am 11.05.2009 um 11:09 schrieb Peter FELECAN: >> Dagobert Michelsen writes: >> >>> Am 11.05.2009 um 03:09 schrieb Mike Watters: >>>> Please upgrade the buildfarm >>> >>> All buildservers are upgraded to current/ now. Please stand by. >>> This include GCC :-) >> >> Is this reasonable when the packages didn't hit yet the mirrors? My >> only >> issue is that I cannot reproduce the same environment on my build >> stack. > > The packages have been pushed to rsync.opencsw.org, all mirrors > synchronize from this one, including mirror.opencsw.org. The > farm is only updated with released packages. > > mirror.opencsw.org is updated every 6 hours and manually by me > if something is needed fast. You can always pkg-get from that > one to reproduce issues of supercurrent packages. Thank you Dago for the explanation. As I use a local mirror, can I update from mirror.opencsw.org with rsync? -- Peter From ihsan at opencsw.org Mon May 11 13:15:12 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 13:15:12 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> Message-ID: <4A0808C0.4020807@opencsw.org> Am 10.5.2009 23:27 Uhr, Dagobert Michelsen schrieb: > Ihsan: Please test rrdtool against the current cairo and see if the error > goes away. Configure complains, that it can't find pangocairo: * I found a copy of pkgconfig, but there is no pangocairo.pc file around. You may want to set the PKG_CONFIG_PATH variable to point to its location. Interesting is, that pkg-config lists pangocairo: ihsan at build8st:~$ pkg-config --list-all | grep pangocairo pangocairo Pango Cairo - Cairo rendering support for Pango sh: gnome-config: not found Package xrender was not found in the pkg-config search path. Perhaps you should add the directory containing `xrender.pc' to the PKG_CONFIG_PATH environment variable Package 'xrender', required by 'Xft', not found Didn't you had a similar issue? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 13:15:42 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 13:15:42 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: <6BAEA35E-4656-44E6-BD5C-F8F67F25CCED@opencsw.org> Hi Peter, Am 11.05.2009 um 12:33 schrieb Peter FELECAN: > As I use a local mirror, can I update from mirror.opencsw.org with > rsync? Sure: > web at web [web]:/etc > rsync rsync://mirror.opencsw.org/ > svn GAR SVN Tree > opencsw OpenCSW Software Directory > testing OpenCSW Testing Packages However, please keep in mind that we have only 2 MBit up, so please use this only when necessary and prefer the other sites which also have rsync. Best regards -- Dago From dam at opencsw.org Mon May 11 13:17:29 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 13:17:29 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A0808C0.4020807@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A0808C0.4020807@opencsw.org> Message-ID: Hi Ihsan, Am 11.05.2009 um 13:15 schrieb Ihsan Dogan: > Am 10.5.2009 23:27 Uhr, Dagobert Michelsen schrieb: > >> Ihsan: Please test rrdtool against the current cairo and see if the >> error >> goes away. > > Configure complains, that it can't find pangocairo: > > * I found a copy of pkgconfig, but there is no pangocairo.pc file > around. > You may want to set the PKG_CONFIG_PATH variable to point to its > location. > > Interesting is, that pkg-config lists pangocairo: > > ihsan at build8st:~$ pkg-config --list-all | grep pangocairo > pangocairo Pango Cairo - Cairo rendering support > for Pango > sh: gnome-config: not found > Package xrender was not found in the pkg-config search path. > Perhaps you should add the directory containing `xrender.pc' > to the PKG_CONFIG_PATH environment variable > Package 'xrender', required by 'Xft', not found > > Didn't you had a similar issue? Make sure you add the following lines to use X11-libraries: > EXTRA_INC = $(prefix)/X11/include > EXTRA_LIB = $(prefix)/X11/lib > EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib You may also need these if supported: > CONFIGURE_ARGS += --x-include=$(prefix)/X11/include > CONFIGURE_ARGS += --x-libraries=$(abspath $(prefix)/X11/lib/$ > (MM_LIBDIR)) Best regards -- Dago From bonivart at opencsw.org Mon May 11 13:37:23 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 13:37:23 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A070772.2010206@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> Message-ID: <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> 2009/5/10 Trygve Laugst?l : > I'm interested, but I doubt Norway is the best place to go. It will > require many people to travel far and stuff is generally considered > expensive in Norway. I can get access to conference rooms and work > facilities with connectivity for free though. Norway sounds excellent to me. ;-) -- /peter From ihsan at opencsw.org Mon May 11 13:44:28 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 13:44:28 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A0808C0.4020807@opencsw.org> Message-ID: <4A080F9C.3010009@opencsw.org> Hello Dago, Am 11.5.2009 13:17 Uhr, Dagobert Michelsen schrieb: >> Configure complains, that it can't find pangocairo: >> >> * I found a copy of pkgconfig, but there is no pangocairo.pc file around. >> You may want to set the PKG_CONFIG_PATH variable to point to its >> location. >> >> Interesting is, that pkg-config lists pangocairo: >> >> ihsan at build8st:~$ pkg-config --list-all | grep pangocairo >> pangocairo Pango Cairo - Cairo rendering support for >> Pango >> sh: gnome-config: not found >> Package xrender was not found in the pkg-config search path. >> Perhaps you should add the directory containing `xrender.pc' >> to the PKG_CONFIG_PATH environment variable >> Package 'xrender', required by 'Xft', not found >> >> Didn't you had a similar issue? > > Make sure you add the following lines to use X11-libraries: > >> EXTRA_INC = $(prefix)/X11/include >> EXTRA_LIB = $(prefix)/X11/lib >> EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib That did it. Thanks. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Mon May 11 13:44:29 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 13:44:29 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> Message-ID: <4A080F9D.1050504@opencsw.org> Am 11.5.2009 13:37 Uhr, Peter Bonivart schrieb: >> I'm interested, but I doubt Norway is the best place to go. It will >> require many people to travel far and stuff is generally considered >> expensive in Norway. I can get access to conference rooms and work >> facilities with connectivity for free though. > > Norway sounds excellent to me. ;-) You can reach even by walk. ;-) Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Mon May 11 14:18:46 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 14:18:46 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A080F9C.3010009@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A0808C0.4020807@opencsw.org> <4A080F9C.3010009@opencsw.org> Message-ID: <4A0817A6.6090701@opencsw.org> Am 11.5.2009 13:44 Uhr, Ihsan Dogan schrieb: >>> Configure complains, that it can't find pangocairo: >>> >>> * I found a copy of pkgconfig, but there is no pangocairo.pc file around. >>> You may want to set the PKG_CONFIG_PATH variable to point to its >>> location. >>> >>> Interesting is, that pkg-config lists pangocairo: >>> >>> ihsan at build8st:~$ pkg-config --list-all | grep pangocairo >>> pangocairo Pango Cairo - Cairo rendering support for >>> Pango >>> sh: gnome-config: not found >>> Package xrender was not found in the pkg-config search path. >>> Perhaps you should add the directory containing `xrender.pc' >>> to the PKG_CONFIG_PATH environment variable >>> Package 'xrender', required by 'Xft', not found >>> >>> Didn't you had a similar issue? >> Make sure you add the following lines to use X11-libraries: >> >>> EXTRA_INC = $(prefix)/X11/include >>> EXTRA_LIB = $(prefix)/X11/lib >>> EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib > > That did it. Thanks. Could you please install the new rrdtool on the test machines? http://mirror.opencsw.org/testing/rrdtool-1.3.7,REV=2009.05.11-SunOS5.8-i386-CSW.pkg.gz http://mirror.opencsw.org/testing/rrdtool-1.3.7,REV=2009.05.11-SunOS5.8-sparc-CSW.pkg.gz Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bonivart at opencsw.org Mon May 11 14:27:55 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 14:27:55 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A080F9D.1050504@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> <4A080F9D.1050504@opencsw.org> Message-ID: <625385e30905110527g178d7926y4a6aefb6757df692@mail.gmail.com> On Mon, May 11, 2009 at 1:44 PM, Ihsan Dogan wrote: > You can reach even by walk. ;-) I think the Norwegians say "g? p? tur". :-) -- /peter From ihsan at opencsw.org Mon May 11 14:48:23 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 14:48:23 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> Message-ID: <4A081E97.5000804@opencsw.org> Hello Dago, Am 10.5.2009 23:27 Uhr, Dagobert Michelsen schrieb: > Ihsan: Please test rrdtool against the current cairo and see if the error > goes away. It looks already much better; at least it's not core dumping anymore. But I'm getting this error: ---------------------------------------------------------- ihsan at build8st:~$ /opt/csw/share/rrdtool/examples/stripes.pl (process:23307): Pango-CRITICAL **: No modules found: No builtin or dynamically loaded modules were found. PangoFc will not work correctly. This probably means there was an error in the creation of: '/opt/csw/etc/pango/pango.modules' You should create this file by running: pango-querymodules > '/opt/csw/etc/pango/pango.modules' (process:23307): Pango-WARNING **: failed to choose a font, expect ugly output. engine-type='PangoRenderFc', script='common' (process:23307): Pango-WARNING **: failed to choose a font, expect ugly output. engine-type='PangoRenderFc', script='latin' This script has created stripes.png in the current directory This demonstrates the use of the TIME and % RPN operators ---------------------------------------------------------- Could you please run pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's just to see if the error goes away. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 15:05:08 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 15:05:08 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A081E97.5000804@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> Message-ID: Hi Ihsan, Am 11.05.2009 um 14:48 schrieb Ihsan Dogan: >> Ihsan: Please test rrdtool against the current cairo and see if the >> error >> goes away. > > It looks already much better; at least it's not core dumping anymore. > > But I'm getting this error: > > ---------------------------------------------------------- > > ihsan at build8st:~$ /opt/csw/share/rrdtool/examples/stripes.pl > > (process:23307): Pango-CRITICAL **: No modules found: > No builtin or dynamically loaded modules were found. > PangoFc will not work correctly. > This probably means there was an error in the creation of: > '/opt/csw/etc/pango/pango.modules' > You should create this file by running: > pango-querymodules > '/opt/csw/etc/pango/pango.modules' > > (process:23307): Pango-WARNING **: failed to choose a font, expect > ugly > output. engine-type='PangoRenderFc', script='common' > > (process:23307): Pango-WARNING **: failed to choose a font, expect > ugly > output. engine-type='PangoRenderFc', script='latin' > This script has created stripes.png in the current directory > This demonstrates the use of the TIME and % RPN operators > ---------------------------------------------------------- > > Could you please run > pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's > just > to see if the error goes away. Done. stripes.pl now doesn't throw an error any more. Should I put this on postinstall or how is this handled? Best regards -- Dago From ihsan at opencsw.org Mon May 11 15:09:23 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 15:09:23 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> Message-ID: <4A082383.3010904@opencsw.org> Hi Dago, Am 11.5.2009 15:05 Uhr, Dagobert Michelsen schrieb: >> Could you please run >> pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's just >> to see if the error goes away. > > Done. stripes.pl now doesn't throw an error any more. Should I put > this on postinstall or how is this handled? I don't know pango at all, but I think it makes sense to place it in the postinstall script. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 15:17:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 15:17:37 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A082383.3010904@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> <4A082383.3010904@opencsw.org> Message-ID: <87897C5D-4B9B-47B1-A28C-9FC10E14C815@opencsw.org> Hi, Am 11.05.2009 um 15:09 schrieb Ihsan Dogan: > Hi Dago, > > Am 11.5.2009 15:05 Uhr, Dagobert Michelsen schrieb: > >>> Could you please run >>> pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? >>> It's just >>> to see if the error goes away. >> >> Done. stripes.pl now doesn't throw an error any more. Should I put >> this on postinstall or how is this handled? > > I don't know pango at all, but I think it makes sense to place it in > the > postinstall script. Any expert on this onboard? William? Ken? Best regards -- Dago From william at wbonnet.net Mon May 11 15:50:53 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 15:50:53 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <87897C5D-4B9B-47B1-A28C-9FC10E14C815@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> <4A082383.3010904@opencsw.org> <87897C5D-4B9B-47B1-A28C-9FC10E14C815@opencsw.org> Message-ID: <4A082D3D.4070609@wbonnet.net> Hi >>>> Could you please run >>>> pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's >>>> just >>>> to see if the error goes away. >>> >>> Done. stripes.pl now doesn't throw an error any more. Should I put >>> this on postinstall or how is this handled? >> >> I don't know pango at all, but I think it makes sense to place it in the >> postinstall script. > > Any expert on this onboard? William? Ken? I'm not really an expert... but i never commited the postinstall script i made when i updated pango before going off. It is done now ... but package has to be rebuild. Dago please ? or i'll do it tonight cheers W. From bwalton at opencsw.org Mon May 11 16:57:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 11 May 2009 10:57:47 -0400 Subject: [csw-maintainers] anyone 'owning' fastcgi currently? Message-ID: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> Hi All, I updated a box today and realized it was still running my local csw package of FastCGI. I hadn't released it after creating. I see there is a GAR description for it, but it hasn't been touched in a while and I don't see a released package for it. If nobody is working on it, I'll take it over and get a release out. If nobody pipes up in a few days, I'll take that as consent. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Mon May 11 17:07:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 11 May 2009 10:07:35 -0500 Subject: [csw-maintainers] libffi now in testing Message-ID: <4A083F37.8090404@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 New package this should allow python's ctypes module to be built ;) and may help some other packages be built with studio rather then gcc - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoIPzYACgkQLrhmsXMSLxddAACg6UGBOipCCZ0EsDVqW4gKJsTu 50UAn047vNF4pAfVKxizGeUkgQpFlpBe =vnIX -----END PGP SIGNATURE----- From ihsan at opencsw.org Mon May 11 17:10:26 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 17:10:26 +0200 Subject: [csw-maintainers] anyone 'owning' fastcgi currently? In-Reply-To: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> References: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> Message-ID: <4A083FE2.40107@opencsw.org> Hello Ben, Am 11.5.2009 16:57 Uhr, Ben Walton schrieb: > I updated a box today and realized it was still running my local csw > package of FastCGI. I hadn't released it after creating. I see there > is a GAR description for it, but it hasn't been touched in a while and > I don't see a released package for it. Do you mean the ap2_modfastcgi package? I've took it over from Cory and updated it last year. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Mon May 11 17:13:23 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 11 May 2009 11:13:23 -0400 Subject: [csw-maintainers] anyone 'owning' fastcgi currently? In-Reply-To: <4A083FE2.40107@opencsw.org> References: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> <4A083FE2.40107@opencsw.org> Message-ID: <1242054719-sup-5308@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Mon May 11 11:10:26 -0400 2009: Hi Ihsan, > Do you mean the ap2_modfastcgi package? I've took it over from Cory and > updated it last year. No, I'm referring to the library that you build client support against. I built (for example) the ruby gem fcgi using the libfastcgi bits. This allows ruby to speak fcgi with apache through the ap2_modfastcgi module. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Mon May 11 19:05:06 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 19:05:06 +0200 Subject: [csw-maintainers] nginx pacakge References: <8E9AF002-6C8F-4F7C-8555-781C71DFCE9B@opencsw.org> Message-ID: <5841B334-5A6B-48C5-8ED2-7750D0BDF47D@opencsw.org> Hi Andy, Am 11.05.2009 um 18:51 schrieb Andy Igoshin: > after reading message below i was confused. > do i use cswclassutils in my package? > if yes do i use it in a right way? It looks ok to me. The classutils are documented here: Phil, what did you have in mind posting this? Best regards -- Dago > ---------- Forwarded Message ---------- > > Subject: Re: nginx pacakge > Date: Mon of May 11 2009 > From: Andy Igoshin > To: Philip Brown > > On Mon of May 11 2009 19:38:38 you wrote: >> On Mon, May 11, 2009 at 06:41:43PM +0400, Andy Igoshin wrote: >>>>> - nginx package supports both init.d and SMF. >>>> >>>> thats why we have cswclassutils. So that one package, can >>>> transparently >>>> support both, depending on what kind of system it is installed to. >>> >>> it is already done. >> >> what do you mean by that? you already use it, or "the package is >> already >> made"? >> The second, is no excuse to not redo it a better way :-) >> >> I see that you did not use cswclassutils. >> >> Please redo (both) packages, to use cswclassutils smf support, >> instead of >> the way you do it now? >> >> It is beneficial both because it makes your package layout simpler, >> AND >> also because we can add extra feature to the common cswclassutils >> package >> later on, without you having to redo the package every time. >> >>> solaris 8/9 does not have 'event port' at all. >>> >>> nginx built on solaris 10 can use select, poll, devpoll, event port. >>> >>> nginx built on solaris 8/9 can use only select, poll, devpoll. >> >> very well... i hate doing it, but i will accept separate packages >> for sol8 >> and sol10. >> >> please redo to use cswclassutils, though. > > CSWnginx.gspec: > ----------------------------------------------------------------- > %var bitname nginx > %var pkgname CSWnginx > %include url file://%{PKGLIB}/csw_dyndepend.gspec > %include url file://%{PKGLIB}/smf_enabled.gspec > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > %var SERVICE_CONF_FILES /opt/csw/etc/nginx/nginx.conf > /opt/csw/etc/nginx/fastcgi_params /opt/csw/etc/nginx/mime.types > %var RC_INIT_SCRIPT cswnginx > %var INIT_KILL_PRIO 16 > %var INIT_START_PRIO 90 > %var CAN_BE_AUTOENABLED no > %copyright url file://%{WORKSRC_FIRSTMOD}/LICENSE > ----------------------------------------------------------------- > > i thought using of such file CSWnginx.gspec in such form > means using of cswclassutils. if it is not correct could you > please give me an example of using cswclassutils in some package? From phil at bolthole.com Mon May 11 19:40:16 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 10:40:16 -0700 Subject: [csw-maintainers] nginx pacakge In-Reply-To: <5841B334-5A6B-48C5-8ED2-7750D0BDF47D@opencsw.org> References: <8E9AF002-6C8F-4F7C-8555-781C71DFCE9B@opencsw.org> <5841B334-5A6B-48C5-8ED2-7750D0BDF47D@opencsw.org> Message-ID: <20090511174015.GC96177@bolthole.com> On Mon, May 11, 2009 at 07:05:06PM +0200, Dagobert Michelsen wrote: > Hi Andy, > > Am 11.05.2009 um 18:51 schrieb Andy Igoshin: >> after reading message below i was confused. >> do i use cswclassutils in my package? >> if yes do i use it in a right way? > > It looks ok to me. The classutils are documented here: > > > Phil, what did you have in mind [with your email to Andy?] I looked at one of his packages. It was using an internal-to-the-package smf class, rather than cswclassutils. Perhaps you looked at a different one of his, that was using cswclassutils. My point is that all of his packages should be using it. From pfelecan at opencsw.org Mon May 11 19:47:58 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 19:47:58 +0200 Subject: [csw-maintainers] libffi now in testing In-Reply-To: <4A083F37.8090404@opencsw.org> (Mike Watters's message of "Mon\, 11 May 2009 10\:07\:35 -0500") References: <4A083F37.8090404@opencsw.org> Message-ID: Mike Watters writes: > New package > > this should allow python's ctypes module to be built ;) > and may help some other packages be built with studio rather then gcc And we still have the capacity to build the same set of packages with gcc? -- Peter From mwatters at opencsw.org Mon May 11 19:52:00 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 11 May 2009 12:52:00 -0500 Subject: [csw-maintainers] libffi now in testing In-Reply-To: References: <4A083F37.8090404@opencsw.org> Message-ID: <4A0865C0.10805@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Mike Watters writes: > >> New package >> >> this should allow python's ctypes module to be built ;) >> and may help some other packages be built with studio rather then gcc > > And we still have the capacity to build the same set of packages with gcc? yep, this should allow the packages that only require the ffi libraries from gcc to be built using studio. specifically I need it for the ctypes module in python. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoIZcAACgkQLrhmsXMSLxfK4ACfQpeKcvfkBPRLhuLu50ErhT3c ZXoAn3PFv8h1TKMvmnDyiEfvWHaJQ1QH =axHm -----END PGP SIGNATURE----- From bonivart at opencsw.org Mon May 11 22:02:19 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 22:02:19 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> On Sat, May 9, 2009 at 5:30 PM, Ihsan Dogan wrote: > Hello, > > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. Topic suggestions from me: - Stable collection - Release process - Tiered packages - Download stats -- /peter From william at wbonnet.net Mon May 11 22:07:57 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 22:07:57 +0200 Subject: [csw-maintainers] Mantis login problem In-Reply-To: <4A07412C.7090804@wbonnet.net> References: <4A073D5F.4030302@wbonnet.net> <4A073DE3.3080702@opencsw.org> <4A07412C.7090804@wbonnet.net> Message-ID: <4A08859D.4080405@wbonnet.net> Hi Sebastian It seems to work. Sorry for the noise ;) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From william at wbonnet.net Mon May 11 22:13:05 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 22:13:05 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A071355.3020204@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> Message-ID: <4A0886D1.9080800@wbonnet.net> Hi > For the internal use, it might be handy to have a Instant Messenger > contact list. > What about having our own jabber server ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Mon May 11 22:19:38 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 22:19:38 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0886D1.9080800@wbonnet.net> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> Message-ID: Hi William, Am 11.05.2009 um 22:13 schrieb William Bonnet: >> For the internal use, it might be handy to have a Instant Messenger >> contact list. >> > > What about having our own jabber server ? Nice :-) However, I would prefer if someone (Phil?) would insert these two lines I posted onto the active webpage now. Then bring the new website online, then have our own jabber server :-) Best regards -- Dago From william at wbonnet.net Mon May 11 22:26:35 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 22:26:35 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> Message-ID: <4A0889FB.9030205@wbonnet.net> Hi > I have now rebuild all the X11 packages from William, pixman, > cairo, pango, libxft2, libatk and gtk+ in GAR with 32/64 bit. Thank you very for having updated these packages Dago. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From phil at bolthole.com Mon May 11 22:29:55 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 13:29:55 -0700 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> Message-ID: <20090511202955.GE96177@bolthole.com> On Mon, May 11, 2009 at 10:19:38PM +0200, Dagobert Michelsen wrote: > > Nice :-) However, I would prefer if someone (Phil?) would insert > these two lines I posted onto the active webpage now. Then > bring the new website online, then have our own jabber server :-) please remind me of the specific lines and also which page you think it would best belong on. From dam at opencsw.org Mon May 11 22:36:41 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 22:36:41 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <20090511202955.GE96177@bolthole.com> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> Message-ID: Hi Phil, Am 11.05.2009 um 22:29 schrieb Philip Brown: > On Mon, May 11, 2009 at 10:19:38PM +0200, Dagobert Michelsen wrote: >> >> Nice :-) However, I would prefer if someone (Phil?) would insert >> these two lines I posted onto the active webpage now. Then >> bring the new website online, then have our own jabber server :-) > > please remind me of the specific lines and also which page you think > it > would best belong on. Here: Am 10.05.2009 um 18:54 schrieb Trygve Laugst?l: > Dagobert Michelsen wrote: >> Hi, >> >> I just found out that there is a Java Chatclient at Freenode. How >> about >> putting >> >> on the webpage to let interested people join immediately? > > I like the idea, but think mibbit[1] is a better client as it don't > require anything other than a browser. The applet integration on > Solaris > has been a bit shady in my experience. > > [1]: http://www.mibbit.com/ Maybe right on front, just above the packages updated box. Trgve or Peter may help in integrating mibbit. Best regards -- Dago From phil at bolthole.com Mon May 11 22:46:29 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 13:46:29 -0700 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> Message-ID: <20090511204629.GH96177@bolthole.com> On Mon, May 11, 2009 at 10:36:41PM +0200, Dagobert Michelsen wrote: > Hi Phil, *wave* > Am 11.05.2009 um 22:29 schrieb Philip Brown: >> please remind me of the specific lines and also which page you think >> it would best belong on. > Here: You did not give specific html code to include. You also did not give ONE choice, but a partial discussion on TWO alternatives. I can't deploy a quantum cat to our web pages; please give me something to put in that does not involve Heisenburg ;-) > Am 10.05.2009 um 18:54 schrieb Trygve Laugst?l: > >> Dagobert Michelsen wrote: >>> Hi, >>> >>> I just found out that there is a Java Chatclient at Freenode. How >>> about >>> putting >>> >>> on the webpage to let interested people join immediately? >> >> I like the idea, but think mibbit[1] is a better client as it don't >> require anything other than a browser. The applet integration on >> Solaris >> has been a bit shady in my experience. >> >> [1]: http://www.mibbit.com/ > > Maybe right on front, just above the packages updated box. > Trgve or Peter may help in integrating mibbit. From ihsan at opencsw.org Mon May 11 23:03:33 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 23:03:33 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> Message-ID: <4A0892A5.80100@opencsw.org> Am 11.5.2009 22:02 Uhr, Peter Bonivart schrieb: >> Back in December we have been talking about a summer camp in Norway. I >> think it's now time to talk about the topics for the summer camp. I >> would like to open the discussion now. > > Topic suggestions from me: > > - Stable collection > - Release process > - Tiered packages > - Download stats I've created quickly wiki page --> http://wiki.opencsw.org/summercamp-2009 I think it's the best, if we use this place to organize the summercamp as we did with the IRL meeting in Switzerland. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Mon May 11 23:08:23 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 23:08:23 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0886D1.9080800@wbonnet.net> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> Message-ID: <4A0893C7.4070909@opencsw.org> Am 11.5.2009 22:13 Uhr, William Bonnet schrieb: >> For the internal use, it might be handy to have a Instant Messenger >> contact list. >> > What about having our own jabber server ? A few people have sent me already their contacts and all of them had already a Jabber address, but an own Jabber server is something what we could think of. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bonivart at opencsw.org Mon May 11 23:29:22 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 23:29:22 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <20090511204629.GH96177@bolthole.com> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> <20090511204629.GH96177@bolthole.com> Message-ID: <625385e30905111429s565ebf1eh1f0611f8b54c8ae3@mail.gmail.com> On Mon, May 11, 2009 at 10:46 PM, Philip Brown wrote: > You did not give specific html code to include. You also did not > give ONE choice, but a partial discussion on TWO alternatives. > I can't deploy a quantum cat to our web pages; please give me something to > put in that does not involve Heisenburg ;-) Mibbit was suggested since it's more lightweight (Ajax vs Java). This is all the HTML that is needed: IRC -- /peter From bonivart at opencsw.org Mon May 11 23:36:12 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 23:36:12 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A0892A5.80100@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <4A0892A5.80100@opencsw.org> Message-ID: <625385e30905111436y89a5f15pa7d579bd8e3fb863@mail.gmail.com> On Mon, May 11, 2009 at 11:03 PM, Ihsan Dogan wrote: > I've created quickly wiki page --> http://wiki.opencsw.org/summercamp-2009 Great! > I think it's the best, if we use this place to organize the summercamp > as we did with the IRL meeting in Switzerland. When looking at the page for the first meeting it strikes me that we should make somewhat more official notes from the summer meeting. I have no idea what topics were discussed at the first meeting, someone tried to recap the meeting on the mailing list but maybe some notes should be passed around a few of the attendees for review and then be made public. -- /peter From phil at bolthole.com Mon May 11 23:42:24 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 14:42:24 -0700 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <625385e30905111429s565ebf1eh1f0611f8b54c8ae3@mail.gmail.com> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> <20090511204629.GH96177@bolthole.com> <625385e30905111429s565ebf1eh1f0611f8b54c8ae3@mail.gmail.com> Message-ID: <20090511214224.GI96177@bolthole.com> On Mon, May 11, 2009 at 11:29:22PM +0200, Peter Bonivart wrote: > Mibbit was suggested since it's more lightweight (Ajax vs Java). This > is all the HTML that is needed: > > target="_blank">IRC added From mwatters at opencsw.org Mon May 11 23:59:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 11 May 2009 16:59:26 -0500 Subject: [csw-maintainers] libtool: link: cannot find the library `' Message-ID: <4A089FBE.1020105@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I ran across this error when compiling php4. I did some googling and found this bit of very handy information. "... In this case one can make libtool be more verbose by editing the appropriate line to $echo "$modename: cannot find the library \`$lib' (lib${name}.la: $deplib)" 1>&2 Resulting error message becomes something like: libtool: link: cannot find the library `' (whatever-library-is-missed: and-where-it-is-missed) ..." http://naranja.umh.es/~atg/B1098370398/C2036966428/E737470728/index.html - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoIn74ACgkQLrhmsXMSLxdjogCcDLJst6gtcoVuKeRgEZa6ehLV cDoAn2/Rb0Wj8YCMH6QuEW0oIXCt16QF =RKsJ -----END PGP SIGNATURE----- From phil at bolthole.com Tue May 12 00:03:25 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 15:03:25 -0700 Subject: [csw-maintainers] libtool: link: cannot find the library `' In-Reply-To: <4A089FBE.1020105@opencsw.org> References: <4A089FBE.1020105@opencsw.org> Message-ID: <20090511220325.GJ96177@bolthole.com> On Mon, May 11, 2009 at 04:59:26PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I ran across this error when compiling php4. I did some googling and found > this bit of very handy information. sounds lovely.. why dont you just go ahead and file a 'bug' against our libtool package as a feature request? From pfelecan at opencsw.org Tue May 12 10:19:09 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 12 May 2009 10:19:09 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> (Peter Bonivart's message of "Mon\, 11 May 2009 22\:02\:19 +0200") References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> Message-ID: Peter Bonivart writes: > Topic suggestions from me: > > - Stable collection > - Release process > - Tiered packages Can you explain what you mean by "tiered packages"? > - Download stats -- Peter From bonivart at opencsw.org Tue May 12 10:49:01 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 12 May 2009 10:49:01 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> Message-ID: <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> On Tue, May 12, 2009 at 10:19 AM, Peter FELECAN wrote: > Can you explain what you mean by "tiered packages"? Sure, this is from the Suggestions page on the wiki: # Packages in tiers. [bonivart] We should classify each package in a tier according to how common/critical/important it is. Tier 1 should be the packages that most will install and are likely to be critical on those systems, examples can be Apache2, Bind, MySQL, OpenSSL and Sendmail. Tier 2 should be packages similar to tier 1 but maybe not as common and/or critical, one example might be Perl. Tier 3 is the rest. ? ?* I envision tier 1 to be around 25 packages, tier 2 around 100 and tier 3 the rest. ? ?* Tier 1 packages must always be actively maintained, possibly by more than one maintainer for quick responses to security updates and bug reports. All tier 1 packages must be in GAR. ? ?* Tier 2 packages should be actively maintained by maintainers who use their own packages on a daily basis. Packages should be in GAR. ? ?* Tier 3 has no special requirements. -- /peter From phil at bolthole.com Tue May 12 18:53:35 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 12 May 2009 09:53:35 -0700 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> Message-ID: <20090512165334.GA84438@bolthole.com> On Tue, May 12, 2009 at 10:49:01AM +0200, Peter Bonivart wrote: >... > ? ?* Tier 1 packages must always be actively maintained, possibly by > more than one maintainer for quick responses to security updates and > bug reports. All tier 1 packages must be in GAR. What you are suggesting, would be described in US political terms, as an "unfunded mandate". (saying "do this", but providing no resources to actually do it) Even if we declared this somehow to be in effect, that does not magically produce maintainers who are willing to step up and do the job. Things like this best happen from the bottom up. If people want to volunteer to help out [current maintainer], then they are free to do so now. > ? ?* Tier 2 packages should be actively maintained by maintainers who > use their own packages on a daily basis. Packages should be in GAR. Again, an "unfunded mandate". It's all very well to declare, "well, package x is really important: it MUST be maintained by someone who uses it on a daily basis!!" But now, try to magically produce someone who both fits these requirements, AND is willing to put in the effort to maintain it! Your suggestions in this area are remarkably, and ironically, dictatorial :-) If package x, has maintainer X, who rarely uses it, and maintainer Y DOES use it every day... maintainer Y is perfectly allowed to step forward and volunteer to take over the package. This happens already, and it has never been turned down, either by me, or by the "maintainer X' in question. There's no need to make unneccessary extra "rules" about this stuff. Particularly rules that just dont fit into a volunteer organization such as ours. From bonivart at opencsw.org Tue May 12 19:13:53 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 12 May 2009 19:13:53 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <20090512165334.GA84438@bolthole.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> <20090512165334.GA84438@bolthole.com> Message-ID: <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> On Tue, May 12, 2009 at 6:53 PM, Philip Brown wrote: > On Tue, May 12, 2009 at 10:49:01AM +0200, Peter Bonivart wrote: >>... >> ? ?* Tier 1 packages must always be actively maintained, possibly by >> more than one maintainer for quick responses to security updates and >> bug reports. All tier 1 packages must be in GAR. > > What you are suggesting, would be described in US political terms, as > an "unfunded mandate". (saying "do this", but providing no resources > ?to actually do it) Well, to identify which packages should go into which tier we could use another suggested topic of mine - download stats. > Things like this best happen from the bottom up. > If people want to volunteer to help out [current maintainer], then > they are free to do so now. Yes, but that usually only happens when we already have a problem and the current maintainer doesn't respond, that's what I want to avoid. To me, it would be good to identify the most critical packages and allocate resources in the form of active maintainers to them but if the community doesn't agree that's fine with me. Topics were asked for and I provided a few. -- /peter From william at wbonnet.net Tue May 12 21:00:18 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 12 May 2009 21:00:18 +0200 Subject: [csw-maintainers] Firefox 2.0.0.20 is in testing Message-ID: <4A09C742.6050704@wbonnet.net> Hi Firefox 2.0.0.20 is available from testing. You will need to update libcairo, libpango, pixman from testing, and an up to date glib2 from current (2.20) Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bwalton at opencsw.org Wed May 13 04:11:22 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 12 May 2009 22:11:22 -0400 Subject: [csw-maintainers] git 1.6.3 in testing Message-ID: <1242180411-sup-4562@ntdws12.chass.utoronto.ca> Hi All, I've placed packages for the newly released git 1.6.3 in testing. Feedback welcome. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 13 05:06:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 12 May 2009 23:06:59 -0400 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A070772.2010206@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> Message-ID: <1242183959-sup-1057@ntdws12.chass.utoronto.ca> Excerpts from Trygve Laugst?l's message of Sun May 10 12:57:22 -0400 2009: > I'm interested, but I doubt Norway is the best place to go. It will > require many people to travel far and stuff is generally considered > expensive in Norway. I can get access to conference rooms and work > facilities with connectivity for free though. I'd love to pop over to Norway, but it won't work for me this summer. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Wed May 13 12:10:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 12:10:25 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 Message-ID: Hi, I am currently updating glib (version 1.x) to 1.3.15 and get the attached errors on compilation. I don't really understand how this SGML/XML stuff works, so maybe someone who as more expertise may have a look? The package has been committed to mGAR as pkg/glib. Best regards -- Dago ... /opt/csw/bin/openjade:I: maximum number of errors (200) reached; change with -E option /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:73:W: cannot generate system identifier for public text "-//James Clark//DTD DSSSL Style Sheet//EN" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:2:91:W: cannot generate system identifier for public text "-//Norman Walsh// DOCUMENT DocBook HTML Stylesheet//EN" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:3:0:E: reference to entity "STYLE-SHEET" for which no system identifier could be generated /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:0: entity was defined here /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:3:0:E: DTD did not contain element declaration for document type name /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:2:0:E: notation "DSSSL" for entity "dbstyle" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:5:12:E: element "STYLE-SHEET" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:6:25:E: there is no attribute "USE" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:6:34:E: element "STYLE-SPECIFICATION" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:7:25:E: element "STYLE-SPECIFICATION-BODY" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl: 463:27:E: there is no attribute "ID" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl: 463:46:E: there is no attribute "DOCUMENT" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl: 463:55:E: element "EXTERNAL-SPECIFICATION" undefined /opt/csw/bin/openjade:I: (invalid message) GLib Reference Manual GLib Overview GLib is a general-purpose utility library, which provides many useful data types, macros, type conversions, string utilities, file utilities, a main loop abstraction, and so on. It works on many UNIX-like platforms, Windows, OS/2 and BeOS. GLib is released under the GNU Library General Public License (GNU LGPL). GLib depends on the following: iconv() In order to implement conversions between character sets, GLib requires an implementation of the standard iconv() routine. Most modern systems will have a suitable implementation, however many older systems lack an iconv() implementation. On such systems, you must install the libiconv library. a thread implementation The thread support in GLib can be based upon several native thread implementations, e.g. POSIX threads, DCE threads or Solaris threads. Compiling the GLib package 3 GLib Library Compiling the GLib Package How to compile GLib itself Building the Library on UNIX On UNIX, GLib uses the standard GNU build system, using autoconf for package configuration and resolving portability issues, automake for building makefiles that comply with the GNU Coding Standards, and libtool for building shared libraries on multiple platforms. The normal sequence for compiling and installing the GLib library is thus: ./configure make make install The standard options provided by GNU autoconf may be passed to the configure script. Please see the autoconf documentation or run ./configure --help for information about the standard options. Extra Configuration Options In addition to the normal options, the configure script in the GTK+ library supports these additional arguments: configure --enable-debug=[no|minimum|yes] --disable-gc-friendly --enable-gc-friendly --disable-mem-pools --enable-mem-pools --disable-threads --enable-threads --with-threads=[none|posix|dce|solaris|win32] --disable-gtk-doc --enable-gtk-doc --enable-debug Turns on various amounts of debugging support. Setting this to 'no' disables g_assert(), g_return_if_fail(), g_return_val_if_fail() and all cast checks between different object types. Setting it to 'minimum' disables only cast checks. Setting it to 'yes' enables runtime debugging. The default is 'minimum'. Note that 'no' is fast, but dangerous as it tends to destabilize even mostly bug-free software by changing the effect of many bugs from simple warnings into fatal crashes. Thus --enable-debug=no should not be used for stable releases of gtk+. --disable-gc-friendly and --enable-gc-friendly When enabled all memory freed by the application, but retained by GLib for performance reasons is set to zero, thus making deployed garbage collection or memory profiling tools detect unlinked memory correctly. This will make GLib slightly slower and is thus disabled by default. --disable-mem-pools and --enable-mem-pools Many small chunks of memory are often allocated via collective pools in GLib and are cached after release to speed up reallocations. For sparse memory systems this behaviour is often inferior, so memory pools can be disabled to avoid excessive caching and force atomic maintenance of chunks through the g_malloc() and g_free() functions. Code currently affected by this: GList, GSList, GNode allocations GMemChunks become basically non-effective GSignal disables all caching (potentially very slow) GType doesn't honour the GTypeInfo n_preallocs field anymore the GBSearchArray flag G_BSEARCH_ALIGN_POWER2 becomes non-functional --disable-threads and --enable-threads Do not compile GLib to be multi thread safe. GLib will be slightly faster then. This is however not recommended, as many programs rely on GLib being multi thread safe. --with-threads Specify a thread implementation to use. 'posix' and 'dce' can be used interchangeable to mean the different versions of posix threads. configure tries to find out, which one is installed. 'solaris' uses the native Solaris thread implementation. 'none' means that GLib will be thread safe, but does not have a default thread implementation. This has to be supplied to g_thread_init() by the programmer. --disable-gtk-doc and --enable-gtk-doc By default the configure script will try to auto-detect whether the gtk-doc package is installed. If it is, then it will use it to extract and build the documentation for the GLib library. These options can be used to explicitly control whether gtk-doc should be used or not. If it is not used, the distributed, pre-generated HTML files will be installed instead of building them on your machine. Compiling GLib Applications 3 GLib Library Compiling GLib Applications How to compile your GLib application Compiling GLib Applications on UNIX To compile a GLib application, you need to tell the compiler where to find the GLib header files and libraries. This is done with the pkg-config utility. The following interactive shell session demonstrates how pkg-config is used: $ pkg-config --cflags glib-2.0 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include $ pkg-config --libs glib-2.0 -L/usr/lib -lm -lglib-1.3 If your application uses modules, threads or GObject features, it must be compiled and linked with the options returned by the following pkg-config invokations: $ pkg-config --cflags --libs gmodule-2.0 $ pkg-config --cflags --libs gthread-2.0 $ pkg-config --cflags --libs gobject-2.0 The simplest way to compile a program is to use the "backticks" feature of the shell. If you enclose a command in backticks (not single quotes), then its output will be substituted into the command line before execution. So to compile a GLib Hello, World, you would type the following: $ cc `pkg-config --cflags --libs glib-2.0` hello.c -o hello Running GLib Applications 3 GLib Library Running GLib Applications How to run and debug your GLib application Running and debugging GLib Applications Environment variables GLib inspects a few of environment variables in addition to standard variables like LANG, PATH or HOME. G_BROKEN_FILENAMES If this environment variable is set, GLib assumes that filenames are in the locale encoding rather than in UTF-8. G_MESSAGES_PREFIXED A list of log levels for which messages should be prefixed by the program name and PID of the application. The default is to prefix everything except G_LOG_LEVEL_MESSAGE and G_LOG_LEVEL_INFO. G_DEBUG If GLib has been configured with --enable-debug=yes, this variable can be set to a list of debug options, which cause GLib to print out different types of debugging information. fatal_warnings Causes GLib to abort the program at the first call to g_warning(). This option is special in that it doesn't require GLib to be configured with debugging support. Traps and traces gmake[6]: *** [html-build.stamp] Error 1 gmake[6]: Leaving directory `/home/dam/mgar/pkg/glib/trunk/work/build- isa-sparcv8/glib-1.3.15/docs/reference/glib' gmake[5]: *** [all-recursive] Error 1 From schwindt at dfki.uni-kl.de Wed May 13 12:42:30 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Wed, 13 May 2009 12:42:30 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: Your message of "Wed, 13 May 2009 12:10:25 +0200." Message-ID: <200905131042.n4DAgURm020716@dfki.uni-kl.de> > Hi, > > I am currently updating glib (version 1.x) to 1.3.15 and > get the attached errors on compilation. I don't really > understand how this SGML/XML stuff works, so maybe someone > who as more expertise may have a look? > > The package has been committed to mGAR as pkg/glib. > > > Best regards > > -- Dago > > > > ... > /opt/csw/bin/openjade:I: maximum number of errors (200) reached; > change with -E option > /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:73:W: > cannot generate system identifier for public text "-//James Clark//DTD > DSSSL Style Sheet//EN" Long time since I did this, this is what I remember : * openjade docbook docbook-dsssl need to be installed * The variable SGML_CATALOG_FILES needs to be set I can't for sure remember if to /opt/csw/etc/sgml/catalog or may be even to /opt/csw/share/sgml/openjade/catalog * try make in the dir docs/reference/glib Nicolai From dam at opencsw.org Wed May 13 13:46:24 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 13:46:24 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <200905131042.n4DAgURm020716@dfki.uni-kl.de> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> Message-ID: <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> Hi Nicolai, Am 13.05.2009 um 12:42 schrieb Nicolai Schwindt: >> I am currently updating glib (version 1.x) to 1.3.15 and >> get the attached errors on compilation. I don't really >> understand how this SGML/XML stuff works, so maybe someone >> who as more expertise may have a look? >> >> The package has been committed to mGAR as pkg/glib. >> >> >> ... >> /opt/csw/bin/openjade:I: maximum number of errors (200) reached; >> change with -E option >> /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:73:W: >> cannot generate system identifier for public text "-//James Clark// >> DTD >> DSSSL Style Sheet//EN" > > > Long time since I did this, this is what I remember : > > * openjade docbook docbook-dsssl need to be installed I thought I had that done: > build8s% pkginfo | grep -i book > application CSWdocbookdsssl docbookdsssl - Norman Walsh's > modular stylesheets for DocBook > application CSWdocbookdtds docbookdtds - SGML and XML document > type definitions for DocBook. > application CSWdocbookxsl docbookxsl - Norman Walsh's XSL > stylesheets for DocBook XML > application CSWdocbookxsldoc docbookxsldoc - Documentation for > the Docbook XSL stylesheets > * The variable SGML_CATALOG_FILES needs to be set > I can't for sure remember if to /opt/csw/etc/sgml/catalog or > may be even to /opt/csw/share/sgml/openjade/catalog AFAIK this is done automatically on package install. > * try make in the dir docs/reference/glib Same errors: > build8s% cd work/build-isa-sparcv8/glib-1.3.15/docs/reference/glib/ > build8s% gmake > *** Building HTML *** > rm -rf ./html > mkdir ./html > cd ./html && gtkdoc-mkhtml glib ../glib-docs.sgml > /opt/csw/bin/openjade:../glib-docs.sgml:1:59:W: cannot generate > system identifier for public text "-//Davenport//DTD DocBook V3.0//EN" > /opt/csw/bin/openjade:../glib-docs.sgml:61:0:E: reference to entity > "BOOK" for which no system identifier could be generated > /opt/csw/bin/openjade:../glib-docs.sgml:1:0: entity was defined here > /opt/csw/bin/openjade:../glib-docs.sgml:61:0:E: DTD did not contain > element declaration for document type name > /opt/csw/bin/openjade:../glib-docs.sgml:62:9:E: there is no > attribute "ID" > /opt/csw/bin/openjade:../glib-docs.sgml:62:16:E: element "BOOK" > undefined > /opt/csw/bin/openjade:../glib-docs.sgml:63:11:E: element "BOOKINFO" > undefined > /opt/csw/bin/openjade:../glib-docs.sgml:64:10:E: element "TITLE" > undefined > /opt/csw/bin/openjade:../glib-docs.sgml:67:14:E: there is no > attribute "ID" Best regards -- Dago From bwalton at opencsw.org Wed May 13 14:58:49 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 08:58:49 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> Message-ID: <1242219479-sup-5302@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Wed May 13 07:46:24 -0400 2009: Hi Dago, I'll take a look at this. I'm no expert with it either, but it's my package so I'll see if it's a packaging/config/postinstall issue. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Wed May 13 15:12:28 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:12:28 +0200 Subject: [csw-maintainers] Compiler optimization Sun Studio vs. GCC In-Reply-To: References: <20090512113534.8C94F61AD@mail.opencsw.org> <0F37CBDB-EA93-46ED-9189-7BE648D689BD@opencsw.org> <6ED24C28-C9B4-4B5D-AB1A-E7A0A422ED20@opencsw.org> <2D6EF361-CF70-4A6C-B8BE-C118EA177390@opencsw.org> <2F4453A9-6414-46B3-9AE9-77425B7B0B1D@opencsw.org> <3BB667EF-A861-47C1-905C-ECCD02B4D809@opencsw.org> Message-ID: <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> Hi Peter, I'm adding James to the post as he has experience in compiler optimization and used it for some of his packages extensively. Am 13.05.2009 um 15:04 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >>>> You have to install it :-) Apart from that the code generation for >>>> Sparc of gcc is very bad, that's why I also installed gccfss if >>>> the project absolutely demands gcc, but this further compilcates >>>> compilation. >>> >>> Having to install it is not an issue IMHO. >>> >>> When you say "very bad" what do you mean? >> >> The optimized is inferior to the one of SOS. I remember reading a >> comparison of the two, but I can't find it at >> >> > > Didn't find it either. Well, my question was motivated to really > qualify > the differences, lets say on a scale of 1 to 100, how do we position > the > 2 compilers for SPARC generated code. Saying "very bad" is relative > but > to what and in what measure? > > Finally, is that so important for the kind of packages that we > provide? > For 1%, 10% or 66%... Maybe James or some other maintainer has real, first-hand performance data? Best regards -- Dago From schwindt at dfki.uni-kl.de Wed May 13 15:20:35 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Wed, 13 May 2009 15:20:35 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: Your message of "Wed, 13 May 2009 13:46:24 +0200." <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> Message-ID: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> [...] > > * The variable SGML_CATALOG_FILES needs to be set > > I can't for sure remember if to /opt/csw/etc/sgml/catalog or > > may be even to /opt/csw/share/sgml/openjade/catalog > > AFAIK this is done automatically on package install. I tried with your files - sorry - that way I could try the same setup. export SGML_CATALOG_FILES=/opt/csw/etc/sgml/catalog gtkdoc-mkhtml glib /home/dam/mgar/pkg/glib/trunk/work/build-isa-sparcv8/glib-1 .3.15/docs/reference/glib/glib-docs.sgml take a look at build8s : bash-2.03$ pwd /tmp/schwindt Nicolai From pfelecan at acm.org Wed May 13 15:21:29 2009 From: pfelecan at acm.org (Peter FELECAN) Date: Wed, 13 May 2009 15:21:29 +0200 Subject: [csw-maintainers] Compiler optimization Sun Studio vs. GCC In-Reply-To: <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> (Dagobert Michelsen's message of "Wed\, 13 May 2009 15\:12\:28 +0200") References: <20090512113534.8C94F61AD@mail.opencsw.org> <0F37CBDB-EA93-46ED-9189-7BE648D689BD@opencsw.org> <6ED24C28-C9B4-4B5D-AB1A-E7A0A422ED20@opencsw.org> <2D6EF361-CF70-4A6C-B8BE-C118EA177390@opencsw.org> <2F4453A9-6414-46B3-9AE9-77425B7B0B1D@opencsw.org> <3BB667EF-A861-47C1-905C-ECCD02B4D809@opencsw.org> <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > I'm adding James to the post as he has experience in > compiler optimization and used it for some of his packages > extensively. > > Am 13.05.2009 um 15:04 schrieb Peter FELECAN: > >> Dagobert Michelsen writes: >> >>>>> You have to install it :-) Apart from that the code generation for >>>>> Sparc of gcc is very bad, that's why I also installed gccfss if >>>>> the project absolutely demands gcc, but this further compilcates >>>>> compilation. >>>> >>>> Having to install it is not an issue IMHO. >>>> >>>> When you say "very bad" what do you mean? >>> >>> The optimized is inferior to the one of SOS. I remember reading a >>> comparison of the two, but I can't find it at >>> >>> >> >> Didn't find it either. Well, my question was motivated to really >> qualify >> the differences, lets say on a scale of 1 to 100, how do we position >> the >> 2 compilers for SPARC generated code. Saying "very bad" is relative >> but >> to what and in what measure? >> >> Finally, is that so important for the kind of packages that we >> provide? >> For 1%, 10% or 66%... > > Maybe James or some other maintainer has real, first-hand performance > data? Let James wisdom shine upon us! -- Peter FELECAN mailto:pfelecan at acm.org From dam at opencsw.org Wed May 13 15:33:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:33:25 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> Message-ID: <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> Hi, Am 13.05.2009 um 15:20 schrieb Nicolai Schwindt: > [...] >>> * The variable SGML_CATALOG_FILES needs to be set >>> I can't for sure remember if to /opt/csw/etc/sgml/catalog or >>> may be even to /opt/csw/share/sgml/openjade/catalog >> >> AFAIK this is done automatically on package install. > > I tried with your files - sorry - that way I could try the same setup. > > export SGML_CATALOG_FILES=/opt/csw/etc/sgml/catalog > gtkdoc-mkhtml glib /home/dam/mgar/pkg/glib/trunk/work/build-isa- > sparcv8/glib-1 > .3.15/docs/reference/glib/glib-docs.sgml > > take a look at build8s : > > bash-2.03$ pwd > /tmp/schwindt Looks good. Thanks! I guess it would make sence to add this to the GAR defaults? Best regards -- Dago From james at Wed May 13 15:33:41 2009 From: james at (James Lee) Date: Wed, 13 May 2009 13:33:41 GMT Subject: [csw-maintainers] Compiler optimization Sun Studio vs. GCC In-Reply-To: References: <20090512113534.8C94F61AD@mail.opencsw.org> <0F37CBDB-EA93-46ED-9189-7BE648D689BD@opencsw.org> <6ED24C28-C9B4-4B5D-AB1A-E7A0A422ED20@opencsw.org> <2D6EF361-CF70-4A6C-B8BE-C118EA177390@opencsw.org> <2F4453A9-6414-46B3-9AE9-77425B7B0B1D@opencsw.org> <3BB667EF-A861-47C1-905C-ECCD02B4D809@opencsw.org> <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> Message-ID: <20090513.13334100.439557401@gyor.oxdrove.co.uk> On 13/05/09, 14:21:29, Peter FELECAN wrote regarding Re: Compiler optimization Sun Studio vs. GCC: > >> Didn't find it either. Well, my question was motivated to really > >> qualify > >> the differences, lets say on a scale of 1 to 100, how do we position > >> the > >> 2 compilers for SPARC generated code. Saying "very bad" is relative > >> but > >> to what and in what measure? > >> > >> Finally, is that so important for the kind of packages that we > >> provide? > >> For 1%, 10% or 66%... > > > > Maybe James or some other maintainer has real, first-hand performance > > data? > Let James wisdom shine upon us! What type of software, or specifically what, are we compiling? In many cases the "best" optimisation is "-xO3 -xspace" because fastest CPU alone isn't best. Unless a person is waiting or a machine is CPU saturated we don't need to worry. Think too about memory usage and if it has a very short run life load times (code size), be careful that repeated tests don't just test memory and disc cache performance. James. From bwalton at opencsw.org Wed May 13 15:40:13 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 09:40:13 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> Message-ID: <1242221865-sup-2499@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Wed May 13 09:33:25 -0400 2009: > Looks good. Thanks! I guess it would make sence to add this to > the GAR defaults? You could possibly leverage the content in /opt/csw/etc/sgml/sgml.conf so that any future changes are reflected? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Wed May 13 15:45:45 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:45:45 +0200 Subject: [csw-maintainers] Updated glib References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> Message-ID: Hi, Anfang der weitergeleiteten E-Mail: > ====================================================================== > Date Submitted: 2009-05-12 15:32 CEST > Last Modified: 2009-05-12 19:41 CEST > ====================================================================== > Summary: libgdk_pixbuf.la references non- > existent .la files, > breaks graphviz builds > Description: > Building graphviz fails with: > > /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info > 6:0:0 > -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/ > graphviz > gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la > -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 > -lintl > grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory > > > > I believe this error is originating from: /opt/csw/lib/ > libgmodule-2.0.la > which contains a reference to this non-existent file. (Also to ls: > /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) > > > A more general question is: why are .la files being installed at all > by > opencsw? > I don't understand all the rationale, but I note that Fedora does not > install any .la files. > > > ====================================================================== > > ---------------------------------------------------------------------- > (0006105) John Ellson (reporter) - 2009-05-12 16:13 > http://opencsw.org/bugtrack/view.php?id=3666#c6105 > ---------------------------------------------------------------------- > Correction: > > I believe the problem originates in the installed: > /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent > files. > > ---------------------------------------------------------------------- > (0006106) dam (reporter) - 2009-05-12 16:34 > http://opencsw.org/bugtrack/view.php?id=3666#c6106 > ---------------------------------------------------------------------- > .la files were included in the past, but the current policy is to > exclude > them: > > > New packages should take care to EXCLUDE libtool .la files. They are > not > helpful, and often create more problems than they solve. > Unfortunately, > existing packages may need to preserve them, until all dependant > packages > have their own configs adjusted to not use .la files. > > There is also a script which fixes Makefile to use linker directives > instead of .la files. The usage in GAR is documented at > Please ask on > maintainers@ for further advice if needed. > > ---------------------------------------------------------------------- > (0006107) phil (manager) - 2009-05-12 18:58 > http://opencsw.org/bugtrack/view.php?id=3666#c6107 > ---------------------------------------------------------------------- > this is actually a bug in gmodule. > Please file a bug with THAT package, to repackage without .la files. > > ---------------------------------------------------------------------- > (0006108) John Ellson (reporter) - 2009-05-12 19:21 > http://opencsw.org/bugtrack/view.php?id=3666#c6108 > ---------------------------------------------------------------------- > This may *also* be a bug with gmodule, but *this* package is > installing > /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so > minimally it > needs to be repackaged without it. > > I'll file a bug against glib for /opt/csw/lib/libgmodule.la > > (I'm finding the .la files on the build8s host) > > ---------------------------------------------------------------------- > (0006109) phil (manager) - 2009-05-12 19:41 > http://opencsw.org/bugtrack/view.php?id=3666#c6109 > ---------------------------------------------------------------------- > since the offending package is mine, please feel free to repackage > and take > over glib. I have made an updated glib 1.2.10 package with the latest package structure (license/, no static libs, no .la) in testing/: glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz The latest 1.x.y library, 1.3.15, builds a different shared library, whereas all the dependent packages rely on /opt/csw/lib/libglib-1.2.so.0 I guess it wouldn't make sense to recompile them against a newer 1.x version? Best regards -- Dago From dam at opencsw.org Wed May 13 15:46:47 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:46:47 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242221865-sup-2499@ntdws12.chass.utoronto.ca> References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> <1242221865-sup-2499@ntdws12.chass.utoronto.ca> Message-ID: Hi Ben, Am 13.05.2009 um 15:40 schrieb Ben Walton: > Excerpts from Dagobert Michelsen's message of Wed May 13 09:33:25 > -0400 2009: >> Looks good. Thanks! I guess it would make sence to add this to >> the GAR defaults? > > You could possibly leverage the content in /opt/csw/etc/sgml/sgml.conf > so that any future changes are reflected? Is it safe to assume that when the file is there that exactly the variables SGML_BASE_DIR SGML_CATALOGS_DIR are set and need to be exported? Best regards -- Dago From dam at opencsw.org Wed May 13 15:47:17 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:47:17 +0200 Subject: [csw-maintainers] Updated glib References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> Message-ID: <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> Hi, Anfang der weitergeleiteten E-Mail: > ====================================================================== > Date Submitted: 2009-05-12 15:32 CEST > Last Modified: 2009-05-12 19:41 CEST > ====================================================================== > Summary: libgdk_pixbuf.la references non- > existent .la files, > breaks graphviz builds > Description: > Building graphviz fails with: > > /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info > 6:0:0 > -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/ > graphviz > gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la > -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 > -lintl > grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory > > > > I believe this error is originating from: /opt/csw/lib/ > libgmodule-2.0.la > which contains a reference to this non-existent file. (Also to ls: > /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) > > > A more general question is: why are .la files being installed at all > by > opencsw? > I don't understand all the rationale, but I note that Fedora does not > install any .la files. > > > ====================================================================== > > ---------------------------------------------------------------------- > (0006105) John Ellson (reporter) - 2009-05-12 16:13 > http://opencsw.org/bugtrack/view.php?id=3666#c6105 > ---------------------------------------------------------------------- > Correction: > > I believe the problem originates in the installed: > /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent > files. > > ---------------------------------------------------------------------- > (0006106) dam (reporter) - 2009-05-12 16:34 > http://opencsw.org/bugtrack/view.php?id=3666#c6106 > ---------------------------------------------------------------------- > .la files were included in the past, but the current policy is to > exclude > them: > > > New packages should take care to EXCLUDE libtool .la files. They are > not > helpful, and often create more problems than they solve. > Unfortunately, > existing packages may need to preserve them, until all dependant > packages > have their own configs adjusted to not use .la files. > > There is also a script which fixes Makefile to use linker directives > instead of .la files. The usage in GAR is documented at > Please ask on > maintainers@ for further advice if needed. > > ---------------------------------------------------------------------- > (0006107) phil (manager) - 2009-05-12 18:58 > http://opencsw.org/bugtrack/view.php?id=3666#c6107 > ---------------------------------------------------------------------- > this is actually a bug in gmodule. > Please file a bug with THAT package, to repackage without .la files. > > ---------------------------------------------------------------------- > (0006108) John Ellson (reporter) - 2009-05-12 19:21 > http://opencsw.org/bugtrack/view.php?id=3666#c6108 > ---------------------------------------------------------------------- > This may *also* be a bug with gmodule, but *this* package is > installing > /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so > minimally it > needs to be repackaged without it. > > I'll file a bug against glib for /opt/csw/lib/libgmodule.la > > (I'm finding the .la files on the build8s host) > > ---------------------------------------------------------------------- > (0006109) phil (manager) - 2009-05-12 19:41 > http://opencsw.org/bugtrack/view.php?id=3666#c6109 > ---------------------------------------------------------------------- > since the offending package is mine, please feel free to repackage > and take > over glib. I have made an updated glib 1.2.10 package with the latest package structure (license/, no static libs, no .la) in testing/: glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz The latest 1.x.y library, 1.3.15, builds a different shared library, whereas all the dependent packages rely on /opt/csw/lib/libglib-1.2.so.0 I guess it wouldn't make sense to recompile them against a newer 1.x version? Best regards -- Dago From bwalton at opencsw.org Wed May 13 15:54:25 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 09:54:25 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> <1242221865-sup-2499@ntdws12.chass.utoronto.ca> Message-ID: <1242222692-sup-7411@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Wed May 13 09:46:47 -0400 2009: Hi Dago, > Is it safe to assume that when the file is there that exactly the > variables > SGML_BASE_DIR > SGML_CATALOGS_DIR > are set and need to be exported? Those variables set sane defaults for programs to look in for files (stylesheets of various types, catalogs, dtds, etc). So yes, if the file exists, exporting them should be fine. I would also use SGML_CATALOG_DIR to build SGML_CATALOG_FILES (which solved the glib issue for you). There should always be a base catalog when that file exists, but its existence doesn't mean that the required stylesheets or dtds are installed. Hopefully programs can use the catalog to determine this and present reasonable errors if they don't have what they need. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ellson at research.att.com Wed May 13 16:02:20 2009 From: ellson at research.att.com (John Ellson) Date: Wed, 13 May 2009 10:02:20 -0400 Subject: [csw-maintainers] Updated glib In-Reply-To: <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> Message-ID: <4A0AD2EC.105@research.att.com> Dagobert Michelsen wrote: > Hi, > > Anfang der weitergeleiteten E-Mail: >> ====================================================================== >> Date Submitted: 2009-05-12 15:32 CEST >> Last Modified: 2009-05-12 19:41 CEST >> ====================================================================== >> Summary: libgdk_pixbuf.la references non-existent >> .la files, >> breaks graphviz builds >> Description: >> Building graphviz fails with: >> >> /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info >> 6:0:0 >> -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/graphviz >> gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la >> -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 >> -lintl >> grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory >> >> >> >> I believe this error is originating from: >> /opt/csw/lib/libgmodule-2.0.la >> which contains a reference to this non-existent file. (Also to ls: >> /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) >> >> >> A more general question is: why are .la files being installed at all by >> opencsw? >> I don't understand all the rationale, but I note that Fedora does not >> install any .la files. >> >> >> ====================================================================== >> >> ---------------------------------------------------------------------- >> (0006105) John Ellson (reporter) - 2009-05-12 16:13 >> http://opencsw.org/bugtrack/view.php?id=3666#c6105 >> ---------------------------------------------------------------------- >> Correction: >> >> I believe the problem originates in the installed: >> /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent >> files. >> >> ---------------------------------------------------------------------- >> (0006106) dam (reporter) - 2009-05-12 16:34 >> http://opencsw.org/bugtrack/view.php?id=3666#c6106 >> ---------------------------------------------------------------------- >> .la files were included in the past, but the current policy is to >> exclude >> them: >> >> >> New packages should take care to EXCLUDE libtool .la files. They are not >> helpful, and often create more problems than they solve. Unfortunately, >> existing packages may need to preserve them, until all dependant >> packages >> have their own configs adjusted to not use .la files. >> >> There is also a script which fixes Makefile to use linker directives >> instead of .la files. The usage in GAR is documented at >> Please ask on >> maintainers@ for further advice if needed. >> >> ---------------------------------------------------------------------- >> (0006107) phil (manager) - 2009-05-12 18:58 >> http://opencsw.org/bugtrack/view.php?id=3666#c6107 >> ---------------------------------------------------------------------- >> this is actually a bug in gmodule. >> Please file a bug with THAT package, to repackage without .la files. >> >> ---------------------------------------------------------------------- >> (0006108) John Ellson (reporter) - 2009-05-12 19:21 >> http://opencsw.org/bugtrack/view.php?id=3666#c6108 >> ---------------------------------------------------------------------- >> This may *also* be a bug with gmodule, but *this* package is installing >> /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so >> minimally it >> needs to be repackaged without it. >> >> I'll file a bug against glib for /opt/csw/lib/libgmodule.la >> >> (I'm finding the .la files on the build8s host) >> >> ---------------------------------------------------------------------- >> (0006109) phil (manager) - 2009-05-12 19:41 >> http://opencsw.org/bugtrack/view.php?id=3666#c6109 >> ---------------------------------------------------------------------- >> since the offending package is mine, please feel free to repackage >> and take >> over glib. > > I have made an updated glib 1.2.10 package with the latest package > structure (license/, no static libs, no .la) in testing/: > > glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz > glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz > > The latest 1.x.y library, 1.3.15, builds a different shared library, > whereas all the dependent packages rely on > /opt/csw/lib/libglib-1.2.so.0 > > I guess it wouldn't make sense to recompile them against a newer 1.x > version? Thanks. Graphviz does use glib2.0. I'm afraid that I've caused some confusion. I originally reported the libgdk_pixbuf.la problem against the gdk_pixbuf package, and Phil said, rightly as it turns out, that it "wasn't caused by that package, perhaps glib." Using: pkgchk -l -p /opt/csw/lib/libgdk_pixbuf-2.0.la I see that the real culprit is the CSWgtk2 package. Sorry for the confusion. I'll generate a Mantis report against gtk2 requesting removal of its .la files. John From dam at opencsw.org Wed May 13 16:44:34 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 16:44:34 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AD2EC.105@research.att.com> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> Message-ID: <945EC7B4-52F1-46D4-808C-435DF8CBD5BD@opencsw.org> Hi John, Am 13.05.2009 um 16:02 schrieb John Ellson: > Sorry for the confusion. I'll generate a Mantis report against gtk2 > requesting removal of its .la files. You can do so. Please note that there is already a current version of gtk2 is waiting on build8st for testing. Please try compiling graphviz there and help testing the package so it can be released soon. Best regards -- Dago From mwatters at opencsw.org Wed May 13 17:33:19 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 13 May 2009 10:33:19 -0500 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AD2EC.105@research.att.com> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> Message-ID: <4A0AE83F.3090502@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > Dagobert Michelsen wrote: >> Hi, >> >> Anfang der weitergeleiteten E-Mail: >>> ====================================================================== >>> Date Submitted: 2009-05-12 15:32 CEST >>> Last Modified: 2009-05-12 19:41 CEST >>> ====================================================================== >>> Summary: libgdk_pixbuf.la references non-existent >>> .la files, >>> breaks graphviz builds >>> Description: >>> Building graphviz fails with: >>> >>> /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info >>> 6:0:0 >>> -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/graphviz >>> gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la >>> -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 >>> -lintl >>> grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory >>> >>> >>> >>> I believe this error is originating from: >>> /opt/csw/lib/libgmodule-2.0.la >>> which contains a reference to this non-existent file. (Also to ls: >>> /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) >>> >>> >>> A more general question is: why are .la files being installed at all by >>> opencsw? >>> I don't understand all the rationale, but I note that Fedora does not >>> install any .la files. >>> >>> >>> ====================================================================== >>> >>> ---------------------------------------------------------------------- >>> (0006105) John Ellson (reporter) - 2009-05-12 16:13 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6105 >>> ---------------------------------------------------------------------- >>> Correction: >>> >>> I believe the problem originates in the installed: >>> /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent >>> files. >>> >>> ---------------------------------------------------------------------- >>> (0006106) dam (reporter) - 2009-05-12 16:34 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6106 >>> ---------------------------------------------------------------------- >>> .la files were included in the past, but the current policy is to >>> exclude >>> them: >>> >>> >>> New packages should take care to EXCLUDE libtool .la files. They are not >>> helpful, and often create more problems than they solve. Unfortunately, >>> existing packages may need to preserve them, until all dependant >>> packages >>> have their own configs adjusted to not use .la files. >>> >>> There is also a script which fixes Makefile to use linker directives >>> instead of .la files. The usage in GAR is documented at >>> Please ask on >>> maintainers@ for further advice if needed. >>> >>> ---------------------------------------------------------------------- >>> (0006107) phil (manager) - 2009-05-12 18:58 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6107 >>> ---------------------------------------------------------------------- >>> this is actually a bug in gmodule. >>> Please file a bug with THAT package, to repackage without .la files. >>> >>> ---------------------------------------------------------------------- >>> (0006108) John Ellson (reporter) - 2009-05-12 19:21 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6108 >>> ---------------------------------------------------------------------- >>> This may *also* be a bug with gmodule, but *this* package is installing >>> /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so >>> minimally it >>> needs to be repackaged without it. >>> >>> I'll file a bug against glib for /opt/csw/lib/libgmodule.la >>> >>> (I'm finding the .la files on the build8s host) >>> >>> ---------------------------------------------------------------------- >>> (0006109) phil (manager) - 2009-05-12 19:41 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6109 >>> ---------------------------------------------------------------------- >>> since the offending package is mine, please feel free to repackage >>> and take >>> over glib. >> >> I have made an updated glib 1.2.10 package with the latest package >> structure (license/, no static libs, no .la) in testing/: >> >> glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz >> glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz >> >> The latest 1.x.y library, 1.3.15, builds a different shared library, >> whereas all the dependent packages rely on >> /opt/csw/lib/libglib-1.2.so.0 >> >> I guess it wouldn't make sense to recompile them against a newer 1.x >> version? > > Thanks. Graphviz does use glib2.0. > > I'm afraid that I've caused some confusion. I originally reported the > libgdk_pixbuf.la problem > against the gdk_pixbuf package, and Phil said, rightly as it turns out, > that it "wasn't caused by that package, perhaps glib." > > Using: > pkgchk -l -p /opt/csw/lib/libgdk_pixbuf-2.0.la > I see that the real culprit is the CSWgtk2 package. > > Sorry for the confusion. I'll generate a Mantis report against gtk2 > requesting removal of its .la files. Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar Makefile. or if you are building by hand, after you run configure run gar/bin/fixlibtool {Build Directory} - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoK6D8ACgkQLrhmsXMSLxdytgCfXz65a5rUIR6NiE4334yt133i T7UAoMJVO5K3T5zsTxNPvF/unPrj72kj =k9t6 -----END PGP SIGNATURE----- From ellson at opencsw.org Wed May 13 17:49:50 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 11:49:50 -0400 Subject: [csw-maintainers] Updated glib In-Reply-To: <945EC7B4-52F1-46D4-808C-435DF8CBD5BD@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <945EC7B4-52F1-46D4-808C-435DF8CBD5BD@opencsw.org> Message-ID: <4A0AEC1E.4020606@opencsw.org> Dagobert Michelsen wrote: > .... there is already a current version of > gtk2 is waiting on build8st for testing. Please try compiling > graphviz there and help testing the package so it can be released soon. build8st is asking me for a passwd. Can you copy over my .ssh/ ? From pfelecan at opencsw.org Wed May 13 17:57:47 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 13 May 2009 17:57:47 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AE83F.3090502@opencsw.org> (Mike Watters's message of "Wed\, 13 May 2009 10\:33\:19 -0500") References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Mike Watters writes: > John Ellson wrote: >> Sorry for the confusion. I'll generate a Mantis report against gtk2 >> requesting removal of its .la files. > > Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar > Makefile. > > or if you are building by hand, after you run configure run gar/bin/fixlibtool > {Build Directory} This implies that you're on the public build stack and cannot be reproduced on a private build stack, i.e. my private servers. IMHO tools as fixlibtool should be in a package such as cswutils. -- Peter From ellson at opencsw.org Wed May 13 18:05:09 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 12:05:09 -0400 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AE83F.3090502@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: <4A0AEFB5.1000301@opencsw.org> Mike Watters wrote: > >> Using: >> pkgchk -l -p /opt/csw/lib/libgdk_pixbuf-2.0.la >> I see that the real culprit is the CSWgtk2 package. >> >> Sorry for the confusion. I'll generate a Mantis report against gtk2 >> requesting removal of its .la files. >> > > Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar > Makefile. > Ah, well, "Getting Started With GAR" is a whole other issue ;-) > or if you are building by hand, after you run configure run gar/bin/fixlibtool > {Build Directory} > Which, apparently still needs me to get started with gar :-( And anyway, both of these sound like hacks, instead of fixing the real problem which requires removing the broken .la files from the distribution. Removing them will also help regular OpenCSW users, not just the maintainers. John From phil at bolthole.com Wed May 13 18:13:48 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 09:13:48 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <200905131042.n4DAgURm020716@dfki.uni-kl.de> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> Message-ID: <20090513161348.GB41179@bolthole.com> On Wed, May 13, 2009 at 12:42:30PM +0200, Nicolai Schwindt wrote: > .... > * The variable SGML_CATALOG_FILES needs to be set > I can't for sure remember if to /opt/csw/etc/sgml/catalog or > may be even to /opt/csw/share/sgml/openjade/catalog I vaguely remember talking about this with Ben when he made that stuff; My opinion at the time was, "if the software needs the user to set an env var for it to work right, then the software installation is broken". I thought that Ben agreed, and had set it up to work without anything external set. Ben...? From phil at bolthole.com Wed May 13 18:16:39 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 09:16:39 -0700 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AE83F.3090502@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: <20090513161639.GC41179@bolthole.com> On Wed, May 13, 2009 at 10:33:19AM -0500, Mike Watters wrote: > > Sorry for the confusion. I'll generate a Mantis report against gtk2 > > requesting removal of its .la files. > > Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar > Makefile. no, the bug should still be filed either way, since it IS a bug in the gtk2/glib2 package :-) From phil at bolthole.com Wed May 13 18:59:07 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 09:59:07 -0700 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> <20090512165334.GA84438@bolthole.com> <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> Message-ID: <20090513165907.GH41179@bolthole.com> On Tue, May 12, 2009 at 07:13:53PM +0200, Peter Bonivart wrote: > Well, to identify which packages should go into which tier we could > use another suggested topic of mine - download stats. Those stats might be skewed in terms of which packages get frequently updated, not neccessarily in terms of importance. And then there is a relative definition of "importance". Does it mean "popular", or does it mean "critical to other programs"? Or some combination of both? As far as "importance" goes, I think a fairly basic and concrete measure would be, the number of dependants that a package has. eg: gtk2 has around 100. Thus, it is incontrovertibly an "important package" ;-) > Yes, but that usually only happens when we already have a problem and > the current maintainer doesn't respond, that's what I want to avoid. > > To me, it would be good to identify the most critical packages and > allocate resources in the form of active maintainers to them but if > the community doesn't agree that's fine with me. Topics were asked for > and I provided a few. I actually agree that it is important to *identify* critical packages. I just dont think we can go around making statements that they "must" be maintained in a particular manner. btw, if you'd like to spend some time putting together a dynamic page with a dependancy-ranked chart of packages, correlated with which ones are non-maintained, I'd be happy to email you more about how to access our various databases with that information. From bwalton at opencsw.org Wed May 13 19:01:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:01:59 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513161348.GB41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> Message-ID: <1242231941-sup-7634@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 12:13:48 -0400 2009: > I vaguely remember talking about this with Ben when he made that > stuff; This stuff specifically, or the issues you had when building mutt docs using it? > My opinion at the time was, "if the software needs the user to set an env > var for it to work right, then the software installation is broken". Well, things like libxml know where to look for the files and catalogs, and the `sgmlwhich` command provides the mechanism for client software to determine what is needed in a dynamic fashion. [cat `sgmlwhich`] I really don't recall discussing this before outside of the mutt issue which was due to a build-time option setup in opensp (now fixed). > I thought that Ben agreed, and had set it up to work without anything > external set. I only packaged up the stack the way it was done elsewhere...My uses of it, has been functional without environment settings. Nothing I've built uses jade though. Jade is considered (in my understanding) the 'old' way of doing sgml transformation and styling. I did a little poking at this, and I think it's likely an oddity of openjade (a bug likely, but who knows). I ran the command that generates the errors during the Make while is a subdirectory and referencing the sgml input file via ../ (this is the way the Makefile sets it up). This saw the errors Dago reported and an truss shows: open("/opt/csw/etc/sgml/../catalog", O_RDONLY) Err#2 ENOENT I then cd'd up a level and removed the ../ from the referenced sgml file. The file was then processed correctly and truss showed: open("/opt/csw/etc/sgml/catalog", O_RDONLY) = 3 Neither of those runs saw the environment modified. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 13 19:07:17 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 10:07:17 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242231941-sup-7634@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> Message-ID: <20090513170717.GI41179@bolthole.com> 9On Wed, May 13, 2009 at 01:01:59PM -0400, Ben Walton wrote: > I did a little poking at this, and I think it's likely an oddity of > openjade (a bug likely, but who knows). I ran the command that > generates the errors during the Make while is a subdirectory and > referencing the sgml input file via ../ (this is the way the Makefile > sets it up). This saw the errors Dago reported and an truss shows: > > open("/opt/csw/etc/sgml/../catalog", O_RDONLY) Err#2 ENOENT > > I then cd'd up a level and removed the ../ from the referenced sgml > file. The file was then processed correctly and truss showed: > > open("/opt/csw/etc/sgml/catalog", O_RDONLY) = 3 > > Neither of those runs saw the environment modified. Nice detective work. So then, sounds like jade needs to be recompiled, so that it doesnt use the '..' by default? From bwalton at opencsw.org Wed May 13 19:19:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:19:01 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513170717.GI41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> Message-ID: <1242234588-sup-3636@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 13:07:17 -0400 2009: > Nice detective work. So then, sounds like jade needs to be recompiled, so > that it doesnt use the '..' by default? I suspect the issue is deeper than that...in fact, I just confirmed it. I made a second directory nesting that required placing ../../ in front of the input file. The truss output now shows (on a failed run): open("/opt/csw/etc/sgml/../../catalog", O_RDONLY) Err#2 ENOENT With a fully qualified input, I tried to open catalog from the fully qualified directory of the input file. It's using the prefix from the input file for some reason. I'm not 100% positive that it's even a bug, simply because something _this_ silly should have shown up ages ago and been fixed. That, and they do provide a method to directly supply the catalog file path via the environment. If I have time, I'll see if this is a known or intended issue. If it's a known or unknown bug, I _might_ spend time fixing it, but not necessarily. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 13 19:30:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:30:47 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242234588-sup-3636@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> Message-ID: <1242235791-sup-1208@ntdws12.chass.utoronto.ca> Excerpts from Ben Walton's message of Wed May 13 13:19:01 -0400 2009: > It's using the prefix from the input file for some reason. I'm not > 100% positive that it's even a bug, simply because something _this_ > silly should have shown up ages ago and been fixed. That, and they do > provide a method to directly supply the catalog file path via the > environment. ...following up on this, the documentation for openjade explicitly instructs the user to set the variable: http://tldp.org/HOWTO/DocBook-OpenJade-SGML-XML-HOWTO/x456.html Braindead, but that's the way it is. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 13 19:36:27 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 10:36:27 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242234588-sup-3636@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> Message-ID: <20090513173627.GJ41179@bolthole.com> On Wed, May 13, 2009 at 01:19:01PM -0400, Ben Walton wrote: > It's using the prefix from the input file for some reason. I'm not > 100% positive that it's even a bug, simply because something _this_ > silly should have shown up ages ago and been fixed. That, and they do > provide a method to directly supply the catalog file path via the > environment. But I vaguely recall that you tried similar things, with some other "distro", and openjade did not behave in this manner. If so, then its a bug in "our" package....? From phil at bolthole.com Wed May 13 19:41:03 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 10:41:03 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242235791-sup-1208@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> <1242235791-sup-1208@ntdws12.chass.utoronto.ca> Message-ID: <20090513174103.GK41179@bolthole.com> On Wed, May 13, 2009 at 01:30:47PM -0400, Ben Walton wrote: > ...following up on this, the documentation for openjade explicitly > instructs the user to set the variable: > http://tldp.org/HOWTO/DocBook-OpenJade-SGML-XML-HOWTO/x456.html I would disagree with your summary. If you "have to", then openjade would refuse to function at all, if it is not set. But that is not the case. It has some notion of default. the default is not functioning properly. Therefore, the default set in our package, needs fixing somehow. From ellson at opencsw.org Wed May 13 19:46:56 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 13:46:56 -0400 Subject: [csw-maintainers] pango, cairo, ... xcb, ... graphviz Message-ID: <4A0B0790.7040702@opencsw.org> I see there is a rebuild of pango/cairo underway, so I'm guessing you don't need a full bug report for this? On build8st: $ pkg-config --exists --print-errors "pangocairo >= 1.14.9" sh: gnome-config: not found Package xcb was not found in the pkg-config search path. Perhaps you should add the directory containing `xcb.pc' to the PKG_CONFIG_PATH environment variable Package 'xcb', required by 'cairo', not found From bwalton at opencsw.org Wed May 13 19:51:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:51:39 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513173627.GJ41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> <20090513173627.GJ41179@bolthole.com> Message-ID: <1242237020-sup-7245@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 13:36:27 -0400 2009: > But I vaguely recall that you tried similar things, with some other > "distro", and openjade did not behave in this manner. You're going to need to dig up the email on that, as I have no memory of it. > If so, then its a bug in "our" package....? Yes, if it behaves differently in the csw environment than elsewhere, it's a locally introduced bug (or at least environment/system specific). -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 13 19:57:03 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:57:03 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513174103.GK41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> <1242235791-sup-1208@ntdws12.chass.utoronto.ca> <20090513174103.GK41179@bolthole.com> Message-ID: <1242237370-sup-7728@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 13:41:03 -0400 2009: > I would disagree with your summary. > If you "have to", then openjade would refuse to function at all, if it is > not set. , but yes, you're correct. > But that is not the case. It has some notion of default. the default is not > functioning properly. Therefore, the default set in our package, needs > fixing somehow. I just looked and there is a configure option to supply the default catalog. I'll rebuild with that option set explicitly. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at acm.org Wed May 13 20:40:33 2009 From: pfelecan at acm.org (Peter FELECAN) Date: Wed, 13 May 2009 20:40:33 +0200 Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: I'm forwarding this as it's of interest for all the maintainers and lets continue on the list please. -------------- next part -------------- An embedded message was scrubbed... From: James Lee Subject: Re: Compiler optimization Sun Studio vs. GCC Date: Wed, 13 May 2009 18:13:09 GMT Size: 3973 URL: -------------- next part -------------- -- Peter FELECAN mailto:pfelecan at acm.org From ellson at opencsw.org Wed May 13 21:03:23 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 15:03:23 -0400 Subject: [csw-maintainers] gar v. mgar ???? (warning: frustrated newbie questions) Message-ID: <4A0B197B.9090707@opencsw.org> What is "mgar" Is it a newer version of gar? Is mgar the recommended one for opencsw? Does it have a working "getting stated guide" ? I tried: http://apps.sourceforge.net/mediawiki/gar/index.php?title=Getting_Started_with_GAR referenced from: http://www.opencsw.org/standards/pkg-walkthrough but the "Checking out the build tree" instructions don't work. I've tried: http://svn.blastwave.org/trac/wiki/GettingStartedQuickly but "make fetch" doesn't work! John From bonivart at opencsw.org Wed May 13 21:10:05 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 13 May 2009 21:10:05 +0200 Subject: [csw-maintainers] gar v. mgar ???? (warning: frustrated newbie questions) In-Reply-To: <4A0B197B.9090707@opencsw.org> References: <4A0B197B.9090707@opencsw.org> Message-ID: <625385e30905131210n720945d0se1d837a9147cdc65@mail.gmail.com> On Wed, May 13, 2009 at 9:03 PM, John Ellson wrote: > What is "mgar" I think it means Modular GAR and it's the current GAR for OpenCSW. Try this: http://apps.sourceforge.net/trac/gar/wiki/Packaging%20with%20mGAR -- /peter From dam at opencsw.org Wed May 13 21:20:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:20:25 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Hi Peter, Am 13.05.2009 um 17:57 schrieb Peter FELECAN: > Mike Watters writes: >> John Ellson wrote: >>> Sorry for the confusion. I'll generate a Mantis report against gtk2 >>> requesting removal of its .la files. >> >> Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in >> your mGar >> Makefile. >> >> or if you are building by hand, after you run configure run gar/bin/ >> fixlibtool >> {Build Directory} > > This implies that you're on the public build stack and cannot be > reproduced on > a private build stack, i.e. my private servers. Not necessarily. Of course you can check out GAR at home and either build it with GAR (preferred) or use the script and build manually. > IMHO tools as fixlibtool > should be in a package such as cswutils. No, CSWcswclassutils contains runtime tools, fixlibtool is a compiletime tool. Best Regards -- Dago From dam at opencsw.org Wed May 13 21:24:29 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:24:29 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AEFB5.1000301@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> <4A0AEFB5.1000301@opencsw.org> Message-ID: Hi John, Am 13.05.2009 um 18:05 schrieb John Ellson: > Mike Watters wrote: >> or if you are building by hand, after you run configure run gar/bin/ >> fixlibtool >> {Build Directory} >> > > Which, apparently still needs me to get started with gar :-( Yes. If you have any questions about GAR you are welcome to post them here :-) > And anyway, both of these sound like hacks, instead of fixing the > real problem which requires removing the > broken .la files from the distribution. > > Removing them will also help regular OpenCSW users, not just the > maintainers. Absolutely. However, there are not all packages in GAR and rebuilding all packages containing .la-files is a huge amount of work. That's why we decided to remove them one by one. Unfortunately, this means breaking other package builds who rely on a full .la-chain. To solve this, the script fixes these broken chains in the meantime. If you want the .la-files removed on a specific package you are free to file a bug report. Best regards -- Dago From phil at bolthole.com Wed May 13 21:33:25 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 12:33:25 -0700 Subject: [csw-maintainers] Updated glib In-Reply-To: References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: <20090513193324.GA24398@bolthole.com> On Wed, May 13, 2009 at 09:20:25PM +0200, Dagobert Michelsen wrote: > Hi Peter, > > Am 13.05.2009 um 17:57 schrieb Peter FELECAN: >> IMHO tools as fixlibtool >> should be in a package such as cswutils. > > No, CSWcswclassutils contains runtime tools, fixlibtool is a compiletime > tool. > you misread. Peter F is correct; it *would* nicely fit into CSWcswutils cswutils is not currently in gar. you can either give me an "official copy" to include, or you can put it in gar and take over the package. (You already know under what circumstances I will migrate my packages to gar ; unfortunately, you keep getting distracted before implementing them :-) From dam at opencsw.org Wed May 13 21:36:40 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:36:40 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <20090513165907.GH41179@bolthole.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> <20090512165334.GA84438@bolthole.com> <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> <20090513165907.GH41179@bolthole.com> Message-ID: Hi Phil, Am 13.05.2009 um 18:59 schrieb Philip Brown: > On Tue, May 12, 2009 at 07:13:53PM +0200, Peter Bonivart wrote: >> Well, to identify which packages should go into which tier we could >> use another suggested topic of mine - download stats. > > Those stats might be skewed in terms of which packages get frequently > updated, not neccessarily in terms of importance. If you take statistics per-file instead of per-package you would get proper statistics. Nonetheless I would also like to see the download values. Sunfreeware has a policy that all mirrors must provide download statistics upstream for this reason. > And then there is a relative definition of "importance". > Does it mean "popular", or does it mean "critical to other programs"? > Or some combination of both? > > As far as "importance" goes, I think a fairly basic and concrete > measure > would be, the number of dependants that a package has. > eg: gtk2 has around 100. Thus, it is incontrovertibly an > "important package" ;-) Sure. That's why William and I repackaged it. If now someone would test it that would be even better ;-) Best regards -- Dago From dam at opencsw.org Wed May 13 21:47:51 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:47:51 +0200 Subject: [csw-maintainers] pango, cairo, ... xcb, ... graphviz In-Reply-To: <4A0B0790.7040702@opencsw.org> References: <4A0B0790.7040702@opencsw.org> Message-ID: <02A0B863-A125-429B-9651-8AA667ED376A@opencsw.org> Hi John, Am 13.05.2009 um 19:46 schrieb John Ellson: > I see there is a rebuild of pango/cairo underway, so I'm guessing > you don't need a full bug report for this? > > On build8st: > > $ pkg-config --exists --print-errors "pangocairo >= 1.14.9" > sh: gnome-config: not found > Package xcb was not found in the pkg-config search path. > Perhaps you should add the directory containing `xcb.pc' > to the PKG_CONFIG_PATH environment variable > Package 'xcb', required by 'cairo', not found You need to extend the PKG_CONFIG_PATH to /opt/csw/X11/lib/pkgconfig for 32 bit /opt/csw/X11/lib/64/pkgconfig for 64 bit In GAR you can do this by setting the following variables: EXTRA_INC = $(prefix)/X11/include EXTRA_LIB = $(prefix)/X11/lib EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib Best regards -- Dago From dam at opencsw.org Wed May 13 21:54:59 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:54:59 +0200 Subject: [csw-maintainers] gar v. mgar ???? (warning: frustrated newbie questions) In-Reply-To: <4A0B197B.9090707@opencsw.org> References: <4A0B197B.9090707@opencsw.org> Message-ID: Hi John, Am 13.05.2009 um 21:03 schrieb John Ellson: > What is "mgar" > > Is it a newer version of gar? Yes. The GAR versions are described at > Is mgar the recommended one for opencsw? Yes, definitely. > Does it have a working "getting stated guide" ? > > I tried: > http://apps.sourceforge.net/mediawiki/gar/index.php?title=Getting_Started_with_GAR > referenced from: > http://www.opencsw.org/standards/pkg-walkthrough > but the "Checking out the build tree" instructions don't work. > > I've tried: > http://svn.blastwave.org/trac/wiki/GettingStartedQuickly > but "make fetch" doesn't work! The Blastwave GAR is a somewhat "old" as all the advancements like multi-ISA-builds, dynamic prototypes, single package out etc. were all developed after the fork. The documentation at covers the current OpenCSW GAR. Unfortunately, some of the docs have not been updated yet and I apologize for this. I am working on the docs from time to time, but as you know it is a time-consuming process. If you need any help on GAR please mail either me or ask on maintainers at . Best regards -- Dago From skayser at opencsw.org Wed May 13 22:30:57 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 13 May 2009 22:30:57 +0200 (CEST) Subject: [csw-maintainers] dbus-python (CSWpydbus) anyone? Message-ID: <49473.217.227.14.39.1242246657.squirrel@ssl.skayser.de> Hi, would anyone with fancy for python modules (Mike?) mind packaging dbus-python (http://dbus.freedesktop.org/releases/dbus-python/)? CSWpidgin contains purple-remote which can be used to cli-control pidgin. We would like to use this at work to automatically update the IM online status when people insert their smartcard into or pull their smartcard from Sun Ray DTUs. However, purple-remote requires the python dbus module, which i couldn't find in the catalogue. I had filed a bug report against pidgin [1], but as Chad correctly noted it is not really a bug in pidgin, so if someone could put together CSWpydbus this would make some guys here real happy. Sebastian [1] http://opencsw.org/mantis/view.php?id=3673 From mwatters at opencsw.org Wed May 13 22:43:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 13 May 2009 15:43:26 -0500 Subject: [csw-maintainers] libmcal now in testing Message-ID: <4A0B30EE.4040909@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Changes: updated to 0.7 now shared library removed static library 64bit support - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoLMO4ACgkQLrhmsXMSLxdvXACeNuU7OqHXslcB5U8zLUaKVlNg mBAAnjMXoN8i8UNcnYnBfkW/WnVS77YM =LxRv -----END PGP SIGNATURE----- From ellson at opencsw.org Wed May 13 23:10:10 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 17:10:10 -0400 Subject: [csw-maintainers] status report: graphviz packaging - new maintainer Message-ID: <4A0B3732.8010808@opencsw.org> If any maintainers are interested, there is a reasonably complete version of graphviz (today's development snapshot) in /home/ellson/pkgs/graphviz/graphviz-2.23.20090513.0445,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz on the build farm. Should I put this someplace? It built. I haven't tried executing it. After installing, it will need "dot -c" run with installer privileges in order to register its plugins. If it installs ok you could try it with something like: echo "digraph{Hello->World}" | dot -Tpng | xv - I tried building an equivalent i386 version, but it needs the latest gtk2 which is not yet on build8xt ? I don't feel that this package is ready for release yet. I want to do the following first: - Generate a new stable release of graphviz upstream. - Migrate the opencsw maintenance to the mgar mechanism - Find out how to run "dot -c" in a post-install step with installer privileges. To register plugins. - Find out how to split the build products into multiple binary packages with segregated dependencies. - Try to package "gts" and "lasi" first, so that graphviz can use them - Not really critical, but see if I can get some more of the swig'ed language bindings to work currently building with support for: perl, C#, tcl missing are: python, php, ruby, guile, java, lua, ocaml It may just be that these languages are not installed on the build hosts? I haven't investigated yet. John From phil at bolthole.com Wed May 13 23:47:28 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 14:47:28 -0700 Subject: [csw-maintainers] status report: graphviz packaging - new maintainer In-Reply-To: <4A0B3732.8010808@opencsw.org> References: <4A0B3732.8010808@opencsw.org> Message-ID: <20090513214728.GB21796@bolthole.com> On Wed, May 13, 2009 at 05:10:10PM -0400, John Ellson wrote: > > I don't feel that this package is ready for release yet. certainly not. the primary reason being, that it was compiled against packages that have not themselves been released yet! You cannot release a package that has been built on the buildXXXxxxt machines. the final "t" means "test *only*" From mwatters at opencsw.org Wed May 13 23:48:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 13 May 2009 16:48:08 -0500 Subject: [csw-maintainers] dbus-python (CSWpydbus) anyone? In-Reply-To: <49473.217.227.14.39.1242246657.squirrel@ssl.skayser.de> References: <49473.217.227.14.39.1242246657.squirrel@ssl.skayser.de> Message-ID: <4A0B4018.9040904@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Sebastian, Sebastian Kayser wrote: > Hi, > > would anyone with fancy for python modules (Mike?) mind packaging > dbus-python (http://dbus.freedesktop.org/releases/dbus-python/)? > > CSWpidgin contains purple-remote which can be used to cli-control pidgin. > We would like to use this at work to automatically update the IM online > status when people insert their smartcard into or pull their smartcard > from Sun Ray DTUs. > > However, purple-remote requires the python dbus module, which i couldn't > find in the catalogue. I had filed a bug report against pidgin [1], but as > Chad correctly noted it is not really a bug in pidgin, so if someone could > put together CSWpydbus this would make some guys here real happy. > > Sebastian > > [1] http://opencsw.org/mantis/view.php?id=3673 > > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I will add it to my list of things to compile after I get the bug fixes done for the current python. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoLQBcACgkQLrhmsXMSLxdf0gCfcLxkwd1DA/60HeXrbMjSYrKt fu4AnjUz11TwgCj/5dukyYbaxXPJLwaZ =SJmO -----END PGP SIGNATURE----- From maybird1776 at yahoo.com Thu May 14 02:47:29 2009 From: maybird1776 at yahoo.com (ken mays) Date: Wed, 13 May 2009 17:47:29 -0700 (PDT) Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: <962721.43059.qm@web111302.mail.gq1.yahoo.com> GCC 4.4.x (and v4.3.3) beats Sun Studio 11. From pfelecan at opencsw.org Thu May 14 09:54:04 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 14 May 2009 09:54:04 +0200 Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC In-Reply-To: <962721.43059.qm@web111302.mail.gq1.yahoo.com> (ken mays's message of "Wed\, 13 May 2009 17\:47\:29 -0700 \(PDT\)") References: <962721.43059.qm@web111302.mail.gq1.yahoo.com> Message-ID: ken mays writes: > GCC 4.4.x (and v4.3.3) beats Sun Studio 11. Can you develop this affirmation, i.e. what tests have you made to support it -- Peter From pfelecan at opencsw.org Thu May 14 09:57:04 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 14 May 2009 09:57:04 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: (Dagobert Michelsen's message of "Wed\, 13 May 2009 21\:20\:25 +0200") References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Dagobert Michelsen writes: >> IMHO tools as fixlibtool >> should be in a package such as cswutils. > > No, CSWcswclassutils contains runtime tools, fixlibtool is a compiletime > tool. Alright. Maybe this is the time to add a cswutils-devel package -- Peter From dam at opencsw.org Thu May 14 11:25:44 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 14 May 2009 11:25:44 +0200 Subject: [csw-maintainers] status report: graphviz packaging - new maintainer In-Reply-To: <4A0B3732.8010808@opencsw.org> References: <4A0B3732.8010808@opencsw.org> Message-ID: <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> Hi John, Am 13.05.2009 um 23:10 schrieb John Ellson: > If any maintainers are interested, there is a reasonably complete > version of graphviz (today's development snapshot) in > /home/ellson/pkgs/graphviz/ > graphviz-2.23.20090513.0445,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz > on the build farm. Should I put this someplace? Yes. Stuff you want to have tested can be copied to /home/testing from all machines in the BO buildfarm. > It built. I haven't tried executing it. After installing, > it will need "dot -c" run > with installer privileges in order to register its plugins. Before asking others to test it you should of course test it yourself to avoid wasting cycles of helping hands. > I tried building an equivalent i386 version, but it needs the latest > gtk2 which is not yet on build8xt ? Correct. The problem is the libs also include 64 bit versions. For x86 this means they must be build on Solaris 10, as this is the first version supporting 64 bit. Unfortunately we don't have a Solaris 10 x86 machine - yet! > I don't feel that this package is ready for release yet. I want > to do the following first: > > - Generate a new stable release of graphviz upstream. > - Migrate the opencsw maintenance to the mgar mechanism Let me know if you encounter any problems. > - Find out how to run "dot -c" in a post-install step with > installer privileges. To register plugins. This is simple. Just take a look at the many packages doing it: cd mgar/pkg grep postinstall */trunk/Makefile > - Find out how to split the build products into multiple binary > packages with segregated dependencies. This is documented at > - Try to package "gts" and "lasi" first, so that graphviz can use > them > - Not really critical, but see if I can get some more of the > swig'ed language bindings to work > currently building with support for: perl, C#, tcl > missing are: python, php, ruby, guile, java, lua, ocaml > It may just be that these languages are not installed on the > build hosts? I haven't investigated yet. Swig is also unmaintained at the moment and outdated. Feel free to take it over or to ask for somebody else to update it if you need a newer version. Best regards -- Dago From dam at opencsw.org Thu May 14 11:30:48 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 14 May 2009 11:30:48 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Hi Peter, Am 14.05.2009 um 09:57 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >>> IMHO tools as fixlibtool >>> should be in a package such as cswutils. >> >> No, CSWcswclassutils contains runtime tools, fixlibtool is a >> compiletime >> tool. > > Alright. Maybe this is the time to add a cswutils-devel package Phil informed me that there is also a CSWcswutils package which I wasn't aware of. fixlibtool will go into that package at the next opportunity. Best regards -- Dago From maybird1776 at yahoo.com Thu May 14 15:39:59 2009 From: maybird1776 at yahoo.com (ken mays) Date: Thu, 14 May 2009 06:39:59 -0700 (PDT) Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: <500379.89918.qm@web111302.mail.gq1.yahoo.com> --- On Thu, 5/14/09, Peter FELECAN wrote: > From: Peter FELECAN > Subject: Re: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC > To: "internal list for the CSW maintainers" > Date: Thursday, May 14, 2009, 3:54 AM > ken mays > writes: > > > GCC 4.4.x (and v4.3.3) beats Sun Studio 11. > > Can you develop this affirmation, i.e. what tests have you > made to support it. > -- > Peter Hi Peter, Yes, I have been testing both compilers for many years in various situations. I'm not here to bash Sun Studio, though, and it is just my opinion (based on years of factual knowledge). But in saying that, I do use Sun Studio and like it for what it does. Basically, a sedan and a sports car will provide you with decent transportation. Just depends on how you like to get there and the "creature features" those cars provide.... Ken From maybird1776 at yahoo.com Thu May 14 15:59:33 2009 From: maybird1776 at yahoo.com (ken mays) Date: Thu, 14 May 2009 06:59:33 -0700 (PDT) Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: <947301.24137.qm@web111306.mail.gq1.yahoo.com> > > GCC 4.4.x (and v4.3.3) beats Sun Studio 11. > > Can you develop this affirmation, i.e. what tests have you > made to > support it > -- > Peter For the record, performance-wise and feature-wise you'll usually find that Sun Studio 11 beats GCC with proper speed optimizations being set. I'll always chose Sun Studio when I need to get performance and utilize advanced features (aka the Aston Martin One-77 sports car in this case). You take it out on the weekends and let the wind blow through your toes and fingers. GCC is like the reliable car in the garage... you never want to just stick it on the shelf as that sports car is just not the full family and everyday work car just yet. Here, it beats Sun Studio 11 in handling and porting that mangled code (aka bad children) in the wild. I could speak on fine wine versus wine (grape juice). But, I like using cars... (smile) ~ Ken From ellson at opencsw.org Thu May 14 16:56:39 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 10:56:39 -0400 Subject: [csw-maintainers] help with "make fetch" Message-ID: <4A0C3127.7080203@opencsw.org> I'm working on add my packages to mgar, but I can't get "make fetch" to work. So I tried "make fetch" in an existing package, libxml2, and it doesn't work there either! I suspect I must have some broken utility in my PATH. Any idea what? I've check the obvious: sed, gsed, awg, gawk, find, make, gmake, And what happened to the ":" in the ftp URL? ellson at tools:/mgar/pkg/libxml2/trunk> make fetch [===== NOW BUILDING: libxml2-2.7.2 =====] ginstall -d cookies/global ginstall -d download ginstall -d download/partial ==> Verifying for installed package CSWgmake: installed ==> Verifying for installed package CSWgtar: installed ==> Verifying for installed package CSWggrep: installed ==> Verifying for installed package CSWdiffutils: installed ==> Verifying for installed package CSWgfile: installed ==> Verifying for installed package CSWtextutils: installed ==> Verifying for installed package CSWwget: installed ==> Verifying for installed package CSWfindutils: installed ==> Verifying for installed package CSWgsed: installed ==> Verifying for installed package CSWgawk: installed ==> Verifying for installed package CSWbzip2: installed [prerequisite] complete for libxml2. ==> Grabbing download/CSWlibxml2.depend ==> Trying file//files/CSWlibxml2.depend make: Fatal error in reader: Makefile, line 7: Unexpected end of line seen Current working directory /home/ellson/mgar/pkg/libxml2/trunk ==> Trying file///home/src/CSWlibxml2.depend make: Fatal error in reader: Makefile, line 7: Unexpected end of line seen Current working directory /home/ellson/mgar/pkg/libxml2/trunk ==> Trying ftp//xmlsoft.org/libxml2/CSWlibxml2.depend make: Fatal error in reader: Makefile, line 7: Unexpected end of line seen Current working directory /home/ellson/mgar/pkg/libxml2/trunk (!!!) Failed to download download/CSWlibxml2.depend! make: *** [download/CSWlibxml2.depend] Error 255 ellson at tools:/mgar/pkg/libxml2/trunk> From bonivart at opencsw.org Thu May 14 17:05:51 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 14 May 2009 17:05:51 +0200 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C3127.7080203@opencsw.org> References: <4A0C3127.7080203@opencsw.org> Message-ID: <625385e30905140805t95e9469g73a874f69127f71f@mail.gmail.com> On Thu, May 14, 2009 at 4:56 PM, John Ellson wrote: > I'm working on add my packages to mgar, but I can't get "make fetch" to > work. I don't know what make points at in your environment but I always use gmake. This is my path: PATH=/opt/csw/bin:/opt/studio/SOS11/SUNWspro/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/csw/bin -- /peter From bwalton at opencsw.org Thu May 14 17:07:25 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 11:07:25 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C3127.7080203@opencsw.org> References: <4A0C3127.7080203@opencsw.org> Message-ID: <1242313508-sup-7220@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 10:56:39 -0400 2009: Hi John, > So I tried "make fetch" in an existing package, libxml2, and it doesn't > work there either! The errors look as though the file CSWlibxml2.depend is missing from the trunk/files directory where it would be expected to live. Can you see the file in the directory? > I suspect I must have some broken utility in my PATH. Any idea what? > I've check the obvious: > sed, gsed, awg, gawk, find, make, gmake, As long as the proxy settings are proper (for wherever you're doing the fetch), you should be fine. > And what happened to the ":" in the ftp URL? : is a special character in Make, so urls for fetching are stipped of them. The actual wget commands add them back in (gar.lib.mk, if you're interested). HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 14 17:26:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 11:26:06 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <1242313508-sup-7220@ntdws12.chass.utoronto.ca> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> Message-ID: <4A0C380E.60402@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Thu May 14 10:56:39 -0400 2009: > > Hi John, > > >> So I tried "make fetch" in an existing package, libxml2, and it doesn't >> work there either! >> > > The errors look as though the file CSWlibxml2.depend is missing from > the trunk/files directory where it would be expected to live. Can you > see the file in the directory? > Yes. Its there. Should I try another mgar package that you know works for you? > >> I suspect I must have some broken utility in my PATH. Any idea what? >> I've check the obvious: >> sed, gsed, awg, gawk, find, make, gmake, >> > > As long as the proxy settings are proper (for wherever you're doing > the fetch), you should be fine. > No proxy settings needed here. "wget ftp://xmlsoft.org/libxml2/libxml2-2.7.2.tar.gz" works from the command line. > >> And what happened to the ":" in the ftp URL? >> > > : is a special character in Make, so urls for fetching are stipped of > them. The actual wget commands add them back in (gar.lib.mk, if > you're interested). > > HTH. > -Ben > From bwalton at opencsw.org Thu May 14 17:39:28 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 11:39:28 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C380E.60402@opencsw.org> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> Message-ID: <1242315443-sup-7175@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 11:26:06 -0400 2009: > Should I try another mgar package that you know works for you? No, give Peter's gmake suggestion a try. I didn't think of that (since it's part of my habit now). HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 14 17:44:02 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 11:44:02 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <1242315443-sup-7175@ntdws12.chass.utoronto.ca> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> Message-ID: <4A0C3C42.9050205@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Thu May 14 11:26:06 -0400 2009: > >> Should I try another mgar package that you know works for you? >> > > No, give Peter's gmake suggestion a try. I didn't think of that > (since it's part of my habit now). > > HTH. > -Ben > That worked, but why ??? I have: $ which make /home/ellson/bin/make $ ls -l /home/ellson/bin/make lrwxrwxrwx 1 ellson ellson 18 2009-04-28 18:02 /home/ellson/bin/make -> /opt/csw/bin/gmake $ which gmake /opt/csw/bin/gmake $ So why doesn't "make" work? It is "gmake" John From bwalton at opencsw.org Thu May 14 17:51:48 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 11:51:48 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C3C42.9050205@opencsw.org> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> <4A0C3C42.9050205@opencsw.org> Message-ID: <1242316123-sup-6398@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 11:44:02 -0400 2009: > That worked, but why ??? The recursive calls to $(MAKE) will simply call 'make' later on, which will use the non-gnu system version. You could override PATH in your Makefile to prepend /opt/csw/gnu (which has a make -> gmake symlink) to make this go if you wanted, but you'd have to do that in every recipe. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 14 18:01:07 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 12:01:07 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <1242316123-sup-6398@ntdws12.chass.utoronto.ca> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> <4A0C3C42.9050205@opencsw.org> <1242316123-sup-6398@ntdws12.chass.utoronto.ca> Message-ID: <4A0C4043.7070208@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Thu May 14 11:44:02 -0400 2009: > >> That worked, but why ??? >> > > The recursive calls to $(MAKE) will simply call 'make' later on, which > will use the non-gnu system version. You could override PATH in your > Makefile to prepend /opt/csw/gnu (which has a make -> gmake symlink) > to make this go if you wanted, but you'd have to do that in every > recipe. > > Thanks > -Ben > I don't have an /opt/csw/gnu What package does that come from? I have /home/ellson/bin at the beginning of my PATH, and it already contains make->/opt/csw/bin/gmake. Doesn't fix it. Adding: export MAKE=/opt/csw/bin/make seems to help. Didn't someone write a: "Recursive makes considered harmful" ? From bwalton at opencsw.org Thu May 14 18:31:21 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 12:31:21 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C4043.7070208@opencsw.org> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> <4A0C3C42.9050205@opencsw.org> <1242316123-sup-6398@ntdws12.chass.utoronto.ca> <4A0C4043.7070208@opencsw.org> Message-ID: <1242318528-sup-4777@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 12:01:07 -0400 2009: > I don't have an /opt/csw/gnu What package does that come from? gnulinks > I have /home/ellson/bin at the beginning of my PATH, and it already > contains make->/opt/csw/bin/gmake. Doesn't fix it. This is because GAR sets a path explicitly. You can override it (per-recipe) by setting something like: PATH := /opt/csw/gnu:$(PATH) (The := prevents recursive expansion.) > Adding: > export MAKE=/opt/csw/bin/make > seems to help. That would do it too. > Didn't someone write a: "Recursive makes considered harmful" ? I don't know. In situations where the system make is used, it's not really an issue. This is only a problem because the default make isn't capable of running GAR (which leverages many features of the gnu version that aren't available in the solaris one). -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 235 bytes Desc: not available URL: From ellson at opencsw.org Thu May 14 20:29:25 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 14:29:25 -0400 Subject: [csw-maintainers] mgar: how to maintain a patch? Message-ID: <4A0C6305.6020902@opencsw.org> I'm trying to package gts with mgar. Its an optional dependency for graphviz and its a smaller package for me to learn about mgar. gts-0.7.6 doesn't build on Solaris (it uses fabsf() which isn't available). The patch is easy, and I've reported the problem upstream and provided them with a patch. Meanwhile, how do I maintain and apply a patch in mgar? A pointer to an example pkg with patches would be fine... John From skayser at opencsw.org Thu May 14 20:51:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 14 May 2009 20:51:19 +0200 (CEST) Subject: [csw-maintainers] mgar: how to maintain a patch? In-Reply-To: <4A0C6305.6020902@opencsw.org> References: <4A0C6305.6020902@opencsw.org> Message-ID: <59208.194.246.122.22.1242327079.squirrel@ssl.skayser.de> Hi John, John Ellson wrote: > I'm trying to package gts with mgar. Its an optional dependency for > graphviz and its a smaller > package for me to learn about mgar. > > gts-0.7.6 doesn't build on Solaris (it uses fabsf() which isn't > available). The patch is easy, and I've reported the > problem upstream and provided them with a patch. > > Meanwhile, how do I maintain and apply a patch in mgar? A pointer to > an example pkg with patches would be fine... socat [1], or xterm [2], or ncdu [3] would be an example. You simply create a patch file (or multiple patch files) for the extracted source directory, store them in the files/ subdirectory of your GAR build directory and list the patch files in $(PATCHFILES). They would then get applied in the "patch" phase of the build. So if you do a "gmake extract", the patches won't be in there yet, but when you do a "gmake patch" they will be applied. Ah, and you need to do a "gmake makesum" after referencing the patch files, so that the diffs are covered by the checksums file. To initially create the patch you can either create it manually or call "gmake extract", patch the sources and then do a "gmake makepatch". You will end up with one gar-base.diff in the files/ subdirectory, which you can reference in PATCHFILES. I have done so for ncdu. Maintaining a set of separate patch files can be a bit of a hassle, but i have found it easier to keep track of which patch is responsible for what (and updating the PATCHFILES list, once a patch has been committed upstream). Sebastian [1] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/socat/trunk/Makefile [2] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/xterm/trunk/Makefile [3] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/ncdu/trunk/Makefile From william at wbonnet.net Thu May 14 20:52:53 2009 From: william at wbonnet.net (William Bonnet) Date: Thu, 14 May 2009 20:52:53 +0200 Subject: [csw-maintainers] mgar: how to maintain a patch? In-Reply-To: <4A0C6305.6020902@opencsw.org> References: <4A0C6305.6020902@opencsw.org> Message-ID: <4A0C6885.2020502@wbonnet.net> Hi John > I'm trying to package gts with mgar. Its an optional dependency for > graphviz and its a smaller > package for me to learn about mgar. > > gts-0.7.6 doesn't build on Solaris (it uses fabsf() which isn't > available). The patch is easy, and I've reported the > problem upstream and provided them with a patch. > > Meanwhile, how do I maintain and apply a patch in mgar? A pointer > to an example pkg with patches would be fine... Many packages are using patches, You can have a look to x11/libX11 for an example. Basically it is easy to create. 1/ Extract the sources gmake extract 2/ Patch manually the files 3/ Create the patch file gmake makepatch 4/ Add the following line to your Makefile PATCHFILES = gar-base.diff 5/ rebuild your package (at least go until patching) gmake clean && gmake patch This should clean sources, download, extract and apply patch file. The new sources should have the modification you made Here is for the short version :) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Fri May 15 03:13:25 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 14 May 2009 20:13:25 -0500 Subject: [csw-maintainers] php4.4.9 now in testing Message-ID: <4A0CC1B5.3070506@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 this is the last of the PHP4.x releases from upstream. this has had no upstream support since December 31st 2007. If you are still using PHP4, Please consider upgrading to PHP5 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoMwbUACgkQLrhmsXMSLxfiNQCfaILknhDOyzJinhscfk0dGvHT o14AnjJqlzbVrujCj9G27Wg9P20XMSFw =2lGb -----END PGP SIGNATURE----- From rupert at opencsw.org Fri May 15 06:10:42 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 15 May 2009 06:10:42 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> <4A05936B.3020001@opencsw.org> <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> Message-ID: <6af4270905142110o2e3fa9fcy64870fcb87a77b0a@mail.gmail.com> svn-1.6.2 is out now ... checksum is the same. did you always configure without bdb support, because when building it complains that bdb is not compiled in, as it is not installed on the build servers? rupert. On Sun, May 10, 2009 at 23:31, rupert THURNER wrote: > you are right, i put a hint also into the mantis issue. > > i tested with the upcoming 1.6.2 release, and it works. i left it on > testing for now, but we should remove it as soon as possible > (http://mirror.opencsw.org/testing/subversion-1.6.2,REV=2009.05.09-SunOS5.8-sparc-UNCOMMITTED.pkg.gz > ). > > as 1.6.2 is already tagged we might wait the couple of days for the > final 1.6.2 release instead of patching 1.6.1. > > rupert. > > > On Sat, May 9, 2009 at 16:30, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> rupert THURNER wrote: >>> hi, >>> >>> while subversion-1.6.1 works fine with the old repository format, we >>> hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after >>> upgrading to 1.6 repository format. >>> >>> the fix will be in 1.6.2. i'll give it a try. >> >> This sounds eerily similar to >> http://www.opencsw.org/mantis/view.php?id=3661 >> http://subversion.tigris.org/issues/show_bug.cgi?id=3387 >> >> I will patch up 1.6.1 with the fix for our bug 3361. ?if you wouldn't mind >> testing that when released against your bug? >> >> if it does not fix it, please file a new opencsw bug. >> I hope to have the fix for subversion done next week. >> >> - -- >> Thanks, >> Mike >> >> "Any intelligent fool can make things bigger, more complex, >> and more violent. ?It takes a touch of genius -- and a lot of courage -- >> to move in the opposite direction." >> >> * Albert Einstein 1879 - 1955 >> ? ?US German-born Theoretical Physicist >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.9 (GNU/Linux) >> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org >> >> iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA >> 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw >> =Du0U >> -----END PGP SIGNATURE----- >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers >> > From rupert at opencsw.org Fri May 15 07:04:49 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 15 May 2009 07:04:49 +0200 Subject: [csw-maintainers] parallel build on build8x and build8s fails? Message-ID: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> i tried to build svn-1.6.2 at the same time on build8x, and build8s. on build8s the following error(s) came .... /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-contrib/svn-push /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svnversion /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnauthz-validate /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnmucc /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff3 /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svn-populate-node-origins-index /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff4 /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter gmake[1]: *** [merge-copy-all] Error 1 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' gmake: *** [merge-isa-sparcv8] Error 2 ... /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter gmake[1]: *** [merge-copy-all] Error 1 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' gmake: *** [merge-isa-sparcv8] Error 2 is there anythign i could do to avoid that - other than build one after the other? From rupert at opencsw.org Fri May 15 07:13:22 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 15 May 2009 07:13:22 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> Message-ID: <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> as "top" said "96% idle" on build8s, i thought i'd try: ? make -j60 package but .. it gives: ? gmake[1]: warning: jobserver unavailable: using -j1. ?Add `+' to parent make rule. is there a possibility to trigger parallel compiles? rupert. From mwatters at opencsw.org Fri May 15 08:18:34 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 15 May 2009 01:18:34 -0500 Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> Message-ID: <4A0D093A.9040107@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > i tried to build svn-1.6.2 at the same time on build8x, and build8s. > on build8s the following error(s) came > > .... > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-contrib/svn-push > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svnversion > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnauthz-validate > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnmucc > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff3 > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svn-populate-node-origins-index > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff4 > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter > gmake[1]: *** [merge-copy-all] Error 1 > gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' > gmake: *** [merge-isa-sparcv8] Error 2 > ... > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter > gmake[1]: *** [merge-copy-all] Error 1 > gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' > gmake: *** [merge-isa-sparcv8] Error 2 > > > is there anythign i could do to avoid that - other than build one > after the other? > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers for a good example on doing parallel builds see the gcc4 recipe. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoNCToACgkQLrhmsXMSLxeHOwCgmJGqJmOnn4lna4I2PIs3x7tB DAoAn3NmUQ5egClpUuR2fyCbO2dxeZ0t =UbfS -----END PGP SIGNATURE----- From dam at opencsw.org Fri May 15 08:55:10 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 15 May 2009 08:55:10 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Hi Rupert, Am 15.05.2009 um 07:13 schrieb rupert THURNER: > as "top" said "96% idle" on build8s, i thought i'd try: > make -j60 package > > but .. it gives: > gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to > parent make rule. > > is there a possibility to trigger parallel compiles? If you want to do it with GAR please use PARALLELMFLAGS="-l -j 30" gmake package This way the flags are properly propagated to subinvocations. Best regards -- Dago From thurner.rupert at redleo.org Fri May 15 07:09:30 2009 From: thurner.rupert at redleo.org (THURNER rupert) Date: Fri, 15 May 2009 07:09:30 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... Message-ID: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> as "top" said "96% idle" on build8s, i thought i'd try: make -j60 package but .. it gives: gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to parent make rule. is there a possibility to trigger parallel compiles? rupert. From pfelecan at opencsw.org Fri May 15 10:21:00 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 15 May 2009 10:21:00 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: (Dagobert Michelsen's message of "Fri\, 15 May 2009 08\:55\:10 +0200") References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Dagobert Michelsen writes: > Hi Rupert, > > Am 15.05.2009 um 07:13 schrieb rupert THURNER: >> as "top" said "96% idle" on build8s, i thought i'd try: >> make -j60 package >> >> but .. it gives: >> gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to >> parent make rule. >> >> is there a possibility to trigger parallel compiles? > > If you want to do it with GAR please use > PARALLELMFLAGS="-l -j 30" gmake package > This way the flags are properly propagated to subinvocations. -j 30 ? How many CPUs do you have in the machine? Reading of the GNU make manual is recommended. -- Peter From dam at opencsw.org Fri May 15 10:54:56 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 15 May 2009 10:54:56 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Hi Peter, Am 15.05.2009 um 10:21 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >> Hi Rupert, >> >> Am 15.05.2009 um 07:13 schrieb rupert THURNER: >>> as "top" said "96% idle" on build8s, i thought i'd try: >>> make -j60 package >>> >>> but .. it gives: >>> gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to >>> parent make rule. >>> >>> is there a possibility to trigger parallel compiles? >> >> If you want to do it with GAR please use >> PARALLELMFLAGS="-l -j 30" gmake package >> This way the flags are properly propagated to subinvocations. > > -j 30 ? How many CPUs do you have in the machine? Reading of the GNU > make manual is recommended. > build8s% psrinfo > 0 on-line since 04/06/2009 16:33:22 > 1 on-line since 04/06/2009 16:33:26 > 2 on-line since 04/06/2009 16:33:26 > 3 on-line since 04/06/2009 16:33:26 > 4 on-line since 04/06/2009 16:33:26 > 5 on-line since 04/06/2009 16:33:26 > 6 on-line since 04/06/2009 16:33:26 > 7 on-line since 04/06/2009 16:33:26 > 8 on-line since 04/06/2009 16:33:26 > 9 on-line since 04/06/2009 16:33:26 > 10 on-line since 04/06/2009 16:33:26 > 11 on-line since 04/06/2009 16:33:26 > 12 on-line since 04/06/2009 16:33:26 > 13 on-line since 04/06/2009 16:33:26 > 14 on-line since 04/06/2009 16:33:26 > 15 on-line since 04/06/2009 16:33:26 > 16 on-line since 04/06/2009 16:33:26 > 17 on-line since 04/06/2009 16:33:26 > 18 on-line since 04/06/2009 16:33:26 > 19 on-line since 04/06/2009 16:33:26 > 20 on-line since 04/06/2009 16:33:26 > 21 on-line since 04/06/2009 16:33:26 > 22 on-line since 04/06/2009 16:33:26 > 23 on-line since 04/06/2009 16:33:26 > 24 on-line since 04/06/2009 16:33:26 > 25 on-line since 04/06/2009 16:33:26 > 26 on-line since 04/06/2009 16:33:26 > 27 on-line since 04/06/2009 16:33:26 > 28 on-line since 04/06/2009 16:33:26 > 29 on-line since 04/06/2009 16:33:26 > 30 on-line since 04/06/2009 16:33:26 > 31 on-line since 04/06/2009 16:33:26 I guess it is a good idea to save 2 CPUs for other users ;-) (It is a T5220 with a 4 Core T2 w/8 strands each) AND: DO NOT DO THIS UNDER ANY CIRCUMSTANCE ON build8x OR ANY OTHER X86 MACHINE FROM THE FARM OR IT WILL CATCH FIRE!! Best regards -- Dago From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 11:07:59 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 11:07:59 +0200 Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> Message-ID: <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> rupert THURNER wrote: > i tried to build svn-1.6.2 at the same time on build8x, and build8s. > on build8s the following error(s) came > > .... > is there anythign i could do to avoid that - other than build one > after the other? use the schily makefile system ;-) You need a build system that was written to support parallel builds. The schily makefilesystem allows to run concurrent make's on all supported platforms since 1992. All other build systems that I am aware don't. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 11:13:32 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 11:13:32 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <4a0d323c.S75GRQ6tB9nGkuZV%Joerg.Schilling@fokus.fraunhofer.de> Dagobert Michelsen wrote: > I guess it is a good idea to save 2 CPUs for other users ;-) > (It is a T5220 with a 4 Core T2 w/8 strands each) For best performance (and in case of real CPUs - not only CMT), I use twice the number of the available CPUs if I compile Solaris If you have 4 real CPU cores and use dmake -j4 to compile Solaris, you will have CPU's that are idle because they are waiting for I/O. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From james at opencsw.org Fri May 15 11:13:57 2009 From: james at opencsw.org (James Lee) Date: Fri, 15 May 2009 09:13:57 GMT Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <20090515.9135700.3717487274@gyor.oxdrove.co.uk> On 15/05/09, 09:54:56, Dagobert Michelsen wrote regarding Re: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ...: > > -j 30 ? How many CPUs do you have in the machine? Reading of the GNU > > make manual is recommended. > > build8s% psrinfo ... > > 31 on-line since 04/06/2009 16:33:26 > I guess it is a good idea to save 2 CPUs for other users ;-) Assuming another using isn't doing the same, ie, don't bother to assume. It's rare that the full complement of 32 threads is used for long by real makefiles. > AND: DO NOT DO THIS UNDER ANY CIRCUMSTANCE ON build8x OR ANY OTHER > X86 MACHINE FROM THE FARM OR IT WILL CATCH FIRE!! $ gmake -j $(psrinfo | wc -l) ... James. From james at opencsw.org Fri May 15 11:21:29 2009 From: james at opencsw.org (James Lee) Date: Fri, 15 May 2009 09:21:29 GMT Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20090515.9212900.560377280@gyor.oxdrove.co.uk> On 15/05/09, 10:07:59, Joerg Schilling wrote regarding Re: [csw-maintainers] parallel build on build8x and build8s fails?: > > is there anythign i could do to avoid that - other than build one > > after the other? > use the schily makefile system ;-) > You need a build system that was written to support parallel builds. > The schily makefilesystem allows to run concurrent make's on all > supported platforms since 1992. All other build systems that I am aware > don't. The OpenOffice.org build system can perform parallel builds of both its project components and can pass a parallel flag to individual projects (although not all sub projects take advantage of the passed flags). James. From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 11:24:56 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 11:24:56 +0200 Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <20090515.9212900.560377280@gyor.oxdrove.co.uk> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> <20090515.9212900.560377280@gyor.oxdrove.co.uk> Message-ID: <4a0d34e8.4ixWzebDMg5/iKa2%Joerg.Schilling@fokus.fraunhofer.de> James Lee wrote: > On 15/05/09, 10:07:59, Joerg Schilling > wrote regarding Re: [csw-maintainers] > parallel build on build8x and build8s fails?: > > > > is there anythign i could do to avoid that - other than build one > > > after the other? > > > use the schily makefile system ;-) > > > You need a build system that was written to support parallel builds. > > The schily makefilesystem allows to run concurrent make's on all > > supported platforms since 1992. All other build systems that I am aware > > don't. > > The OpenOffice.org build system can perform parallel builds of both its > project components and can pass a parallel flag to individual projects > (although not all sub projects take advantage of the passed flags). Are you talking about parallel builds running on different platforms using the same NFS mounted directory tree? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From james at opencsw.org Fri May 15 11:36:18 2009 From: james at opencsw.org (James Lee) Date: Fri, 15 May 2009 09:36:18 GMT Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <4a0d34e8.4ixWzebDMg5/iKa2%Joerg.Schilling@fokus.fraunhofer.de> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> <20090515.9212900.560377280@gyor.oxdrove.co.uk> <4a0d34e8.4ixWzebDMg5/iKa2%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20090515.9361800.3578221933@gyor.oxdrove.co.uk> On 15/05/09, 10:24:56, Joerg Schilling wrote regarding Re: [csw-maintainers] parallel build on build8x and build8s fails?: > > The OpenOffice.org build system can perform parallel builds of both its > > project components and can pass a parallel flag to individual projects > > (although not all sub projects take advantage of the passed flags). > Are you talking about parallel builds running on different platforms > using the same NFS mounted directory tree? No, although it can share the source tree for different arch builds. I mean it is a amalgamation of several different sources or projects and it can build non-dependent sub projects in parallel. It is also possible to distribute builds across multiple machines. James. From pfelecan at opencsw.org Fri May 15 12:21:32 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 15 May 2009 12:21:32 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: (Dagobert Michelsen's message of "Fri\, 15 May 2009 10\:54\:56 +0200") References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > Am 15.05.2009 um 10:21 schrieb Peter FELECAN: > >> Dagobert Michelsen writes: >> >>> Hi Rupert, >>> >>> Am 15.05.2009 um 07:13 schrieb rupert THURNER: >>>> as "top" said "96% idle" on build8s, i thought i'd try: >>>> make -j60 package >>>> >>>> but .. it gives: >>>> gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to >>>> parent make rule. >>>> >>>> is there a possibility to trigger parallel compiles? >>> >>> If you want to do it with GAR please use >>> PARALLELMFLAGS="-l -j 30" gmake package >>> This way the flags are properly propagated to subinvocations. >> >> -j 30 ? How many CPUs do you have in the machine? Reading of the GNU >> make manual is recommended. > >> build8s% psrinfo >> 0 on-line since 04/06/2009 16:33:22 >> [...] >> 31 on-line since 04/06/2009 16:33:26 > > I guess it is a good idea to save 2 CPUs for other users ;-) > (It is a T5220 with a 4 Core T2 w/8 strands each) Oh, I didn't know that. The fact that the build8s is a zone has any impact on this kind of usage? -- Peter From dam at opencsw.org Fri May 15 12:29:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 15 May 2009 12:29:19 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <759CF688-6020-4F61-859E-F40A65132A7A@opencsw.org> Hi Peter, Am 15.05.2009 um 12:21 schrieb Peter FELECAN: > Oh, I didn't know that. The fact that the build8s is a zone has any > impact on this kind of usage? No, I haven't configured resource constraints on the zones. The machine should be large enough to hold some more builds. Best regards -- Dago From skayser at opencsw.org Fri May 15 15:43:10 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 15 May 2009 15:43:10 +0200 Subject: [csw-maintainers] pkgname limit 20 chars? Message-ID: <4A0D716E.4010700@opencsw.org> Hi, i just wanted to package the perl module DateTime::TimeZone and checkpkg bails out on the pkgname: CSWpmdatetimetimezone ==> Checking compliance: CSWpmdatetimetimezone Examining /home/skayser/pkgs/pm_datetimetimezone-0.90,REV=2009.05.15-SunOS5.8-all-CSW.pkg.gz ... pkgname='CSWpmdatetimetimezone' ... ERROR: /tmp/pm_datetimetimezone-0.90,REV=2009.05.15-SunOS5.8-all-CSW.pkg.gz: pkg name greater than 20 chars gmake: *** [pkgcheck-CSWpmdatetimetimezone] Error 2 I have seen that a couple of packages already condense their package and software names, but IMHO this leads to a rather inconsistent naming. Take for example the following perl modules: Perl Module Package Name Software Name -------------------- ------------------- ---------------------- Class::Inspector CSWpmclassinspector pm_classinspector Class::Accessor CSWpmclsaccessor pm_clsaccessor Class::ReturnValue CSWpmclassretval pm_classretval Ideally (IMHO), all the perl modules would just have their full name spelled out in the package / software name (as with Class::Inspector) or at least follow a consistent naming scheme. Are we restricted by external limitations here? Sebastian From phil at bolthole.com Fri May 15 16:45:30 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 07:45:30 -0700 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <4A0D716E.4010700@opencsw.org> References: <4A0D716E.4010700@opencsw.org> Message-ID: <20090515144530.GB84272@bolthole.com> On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: > > Are we restricted by external limitations here? have to draw the line somewhere. we've drawn the line at 20 chars for [mumble] years. it's a reasonable place to draw it, and it doesnt matter much what the CSWxxxx package name is to users, so not having it exactly match the software name, really isnt important. From pfelecan at opencsw.org Fri May 15 17:46:36 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 15 May 2009 17:46:36 +0200 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <20090515144530.GB84272@bolthole.com> (Philip Brown's message of "Fri\, 15 May 2009 07\:45\:30 -0700") References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: Philip Brown writes: > On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: >> >> Are we restricted by external limitations here? > > have to draw the line somewhere. we've drawn the line at 20 chars for > [mumble] years. it's a reasonable place to draw it, and it doesnt matter > much what the CSWxxxx package name is to users, so not having it exactly > match the software name, really isnt important. Sure but it's a PITA for people using "automatic" tools which derive the package name from the software name. BTW, the only restriction is in Sys V pkg where is defined as 256 ASCII characters (long and possibly ugly but aesthetics are subjective by nature). I'm voting for longer than 20 characters (which is of the kind of the 64Kb of the first PC, i.e. "big enough"... and blue) -- Peter From phil at bolthole.com Fri May 15 18:23:01 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:23:01 -0700 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <20090515162301.GD10154@bolthole.com> btw, if it hasnt been mentioned yet; if the package is non-gnuified you might try dmake -j something instead of gmake From phil at bolthole.com Fri May 15 18:26:32 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:26:32 -0700 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: <20090515162632.GE10154@bolthole.com> On Fri, May 15, 2009 at 05:46:36PM +0200, Peter FELECAN wrote: > Philip Brown writes: > > have to draw the line somewhere. we've drawn the line at 20 chars for > > [mumble] years. it's a reasonable place to draw it, and it doesnt matter > > much what the CSWxxxx package name is to users, so not having it exactly > > match the software name, really isnt important. > > Sure but it's a PITA for people using "automatic" tools which derive the > package name from the software name. conversely, reallystupidlylongpackagenamesthataddnovalue are a "PITA" for people who dont use infinately wide screens, and want columnar output summarizing things. Other than gar "automatic creation of pkginfo" type stuff, any "automatic tools", should be doing a lookup from the catalog, not making bad assumptions. From skayser at opencsw.org Fri May 15 18:34:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 15 May 2009 18:34:19 +0200 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <20090515144530.GB84272@bolthole.com> References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: <4A0D998B.4010807@opencsw.org> Philip Brown wrote: > On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: >> Are we restricted by external limitations here? > > have to draw the line somewhere. we've drawn the line at 20 chars for > [mumble] years. it's a reasonable place to draw it, and it doesnt matter > much what the CSWxxxx package name is to users, so not having it exactly > match the software name, really isnt important. Ok, just so that i get it right (please correct me if am wrong): 1) The software name is the one users should work with and we can/should keep it in a consistent, readable, and "no-surprise" way. This means i can keep pm_datetimetimezone as software name for the DateTime::TimeZone perl module. Shortening of software names like it was done for some Class::xxx packages (pm_clsxxx versus pm_classxxx) should be avoided. 2) The package name is limited (by us) to 20 chars, so that we need to come up with a way to condense package names > 20 chars. pm_datetimetimezone -> CSWpmdatetimetz pm_xmlatomsimplefeed -> CSWpmxmlatomsimplefd Question: Is this self-imposed 20 char limitation worth have discrepancies between software and package names? Does it gain us something? Although i install packages via pkg-get or pkgutil and by their software names, i often find myself dealing with packages by their package names afterwards (pkginfo, pkgrm, /var/sadm/pkg/...). IMHO just a source of possible confusion. Sebastian From phil at bolthole.com Fri May 15 18:38:48 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:38:48 -0700 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <4A0D998B.4010807@opencsw.org> References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> <4A0D998B.4010807@opencsw.org> Message-ID: <20090515163848.GF10154@bolthole.com> On Fri, May 15, 2009 at 06:34:19PM +0200, Sebastian Kayser wrote: > Ok, just so that i get it right (please correct me if am wrong): > > 1) The software name is the one users should work with and we can/should > keep it in a consistent, readable, and "no-surprise" way. yup. > This means i > can keep pm_datetimetimezone as software name for the DateTime::TimeZone > perl module. Yup. There IS also a limit on software name, for sanity purposes. it's just longer. > 2) The package name is limited (by us) to 20 chars, so that we need to > come up with a way to condense package names > 20 chars. > > pm_datetimetimezone -> CSWpmdatetimetz > pm_xmlatomsimplefeed -> CSWpmxmlatomsimplefd I dont think it is beneficial to try to enforce "one true way of shortening". It depends on the software name. Some collections of softwares obviously > Question: Is this self-imposed 20 char limitation worth have > discrepancies between software and package names? yes. we've had it for 6 years. It's reasonable. There's no decent reason to uproot the world to change this,just because a few software authors are insane. > Although i install packages via pkg-get or pkgutil and by > their software names, i often find myself dealing with packages by their > package names afterwards (pkginfo, pkgrm, /var/sadm/pkg/...). IMHO just > a source of possible confusion. as a side topic: what would make you have to do that sort of thing less? [as a regular sysadmin, not a pkg maintainer? ] From phil at bolthole.com Fri May 15 18:40:10 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:40:10 -0700 Subject: [csw-maintainers] php4.4.9 now in testing In-Reply-To: <4A0CC1B5.3070506@opencsw.org> References: <4A0CC1B5.3070506@opencsw.org> Message-ID: <20090515164010.GG10154@bolthole.com> On Thu, May 14, 2009 at 08:13:25PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > this is the last of the PHP4.x releases from upstream. > this has had no upstream support since December 31st 2007. > > If you are still using PHP4, Please consider upgrading to PHP5 > i sugest you post this to the users list. with the url to it, in the message body. From Darin.Perusich at cognigencorp.com Fri May 15 18:42:40 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Fri, 15 May 2009 12:42:40 -0400 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: <4A0D9B80.10407@cognigencorp.com> Peter FELECAN wrote: > Philip Brown writes: > >> On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: >>> Are we restricted by external limitations here? >> have to draw the line somewhere. we've drawn the line at 20 chars for >> [mumble] years. it's a reasonable place to draw it, and it doesnt matter >> much what the CSWxxxx package name is to users, so not having it exactly >> match the software name, really isnt important. > > Sure but it's a PITA for people using "automatic" tools which derive the > package name from the software name. BTW, the only restriction is in Sys > V pkg where is defined as 256 ASCII characters (long and possibly ugly > but aesthetics are subjective by nature). I'm voting for longer than 20 > characters (which is of the kind of the 64Kb of the first PC, i.e. "big > enough"... and blue) If the defined limit 256 characters that should be "our" self-defined limit. Additional this should apply for ALL the pkginfo fields which have limits, which I've detailed below from the man page. ARCH = 16 chars CATEGORY = 16 chars NAME = 256 chars PKG = 32 chars VERSION = 256 chars DESC = 256 chars EMAIL = 256 chars HOTLINE = 256 chars -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From mwatters at opencsw.org Fri May 15 18:45:57 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 15 May 2009 11:45:57 -0500 Subject: [csw-maintainers] php4.4.9 now in testing In-Reply-To: <20090515164010.GG10154@bolthole.com> References: <4A0CC1B5.3070506@opencsw.org> <20090515164010.GG10154@bolthole.com> Message-ID: <4A0D9C45.2010302@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Thu, May 14, 2009 at 08:13:25PM -0500, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> this is the last of the PHP4.x releases from upstream. >> this has had no upstream support since December 31st 2007. >> >> If you are still using PHP4, Please consider upgrading to PHP5 >> > > i sugest you post this to the users list. > > with the url to it, in the message body. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I always CC the user list when putting new packages in testing ;) I did not include a link. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoNnEUACgkQLrhmsXMSLxfWiQCfXEXx0wR3XDT1iZUbp23Dd+Sf w/cAn1E1p5pUpZOnTKt1xmL1MfFs8ndk =vwu0 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri May 15 18:50:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 15 May 2009 11:50:08 -0500 Subject: [csw-maintainers] [csw-users] php4.4.9 now in testing In-Reply-To: <4A0CC1B5.3070506@opencsw.org> References: <4A0CC1B5.3070506@opencsw.org> Message-ID: <4A0D9D40.5020402@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > this is the last of the PHP4.x releases from upstream. > this has had no upstream support since December 31st 2007. > > If you are still using PHP4, Please consider upgrading to PHP5 More info on the Releases of PHP can be found at http://www.php.net/releases/ - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoNnUAACgkQLrhmsXMSLxdmqwCg06HHBCtmCT1pF3zx2lEEj/6O FzIAn0hU94US0dcCR9/h2/RufDkLjp3i =pYp8 -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 15 19:09:55 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:09:55 -0700 Subject: [csw-maintainers] [csw-users] php4.4.9 now in testing In-Reply-To: <4A0D9D40.5020402@opencsw.org> References: <4A0CC1B5.3070506@opencsw.org> <4A0D9D40.5020402@opencsw.org> Message-ID: <20090515170955.GH10154@bolthole.com> Arg. Dont crosspost between the lists please. and you didnt include the url to testing. From phil at bolthole.com Fri May 15 19:14:27 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:14:27 -0700 Subject: [csw-maintainers] bug 3034 separate server and user command line binaries In-Reply-To: <4A03B09A.4060103@opencsw.org> References: <4A03B09A.4060103@opencsw.org> Message-ID: <20090515171427.GI10154@bolthole.com> On Thu, May 07, 2009 at 11:10:02PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi, > > I have to recompile subversion to fix an issue with ap2_subversion. > I want to include this bug in the re-package. > > which binaries constitute server vs client? > > Can I get some confirmation on this before I split it up. > I believe it should be as follows: there's also the issue of: 'when people say "install subversion"', are they most likely looking for the server, or are they most likely looking for the client? Personally, I would think they'd most likely be looking for the client. also, I thought the existing package set already had some kind of command-line local-only-stuff package, that skipped all the server stuff? So, that reinforces what I'm saying, that rather than split off a separate "client" package, you should consider splitting off some kind of "core-server" package, that ap2_subversion would then depend on? From phil at bolthole.com Fri May 15 19:17:38 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:17:38 -0700 Subject: [csw-maintainers] problems with existing gtk libs Message-ID: <20090515171738.GJ10154@bolthole.com> there's something up with our "current" gnome libs. My browsers would crash occasionally before. but now they crash multiple times a day, after upgrading to the "latest" stuffs in current. and now, even /usr/sfw/bin/mozilla crashes seamonkey, firefox, and sun mozilla, all crash more frequently now. sigh. multiple times a day :-( From phil at bolthole.com Fri May 15 19:18:55 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:18:55 -0700 Subject: [csw-maintainers] problems with existing gtk libs(and related libs) In-Reply-To: <20090515171738.GJ10154@bolthole.com> References: <20090515171738.GJ10154@bolthole.com> Message-ID: <20090515171855.GK10154@bolthole.com> On Fri, May 15, 2009 at 10:17:38AM -0700, Philip Brown wrote: > seamonkey, firefox, and sun mozilla, all crash more frequently now. sigh. > multiple times a day :-( I guess I should give the error. it's always, The program 'Gecko' received an X Window System error. This probably reflects a bug in the program. The error was 'BadDrawable (invalid Pixmap or Window parameter)'. (Details: serial 81745 error_code 9 request_code 14 minor_code 0) (Note to programmers: normally, X errors are reported asynchronously; blahblahblah. This is sol 10 sparc. dual-headed. From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 19:37:44 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 19:37:44 +0200 Subject: [csw-maintainers] New schily tools in testing Message-ID: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> Hi, I just compiled a new set of packages and put them into testing. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From phil at bolthole.com Fri May 15 20:06:37 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 11:06:37 -0700 Subject: [csw-maintainers] problems with existing gtk libs Message-ID: <20090515180637.GL10154@bolthole.com> Hmm.. actually... i just noticed that I didnt have a swap file configured. this SHOULDNT be a problem, as i HAVE free ram. but... maybe this is one o them there "stupid design problems", where things die if there isnt disk based "backing store". will report later if adding swap, has improved things. It seems to have fixed one reoccuring url crash, at least. From william at wbonnet.net Fri May 15 21:03:50 2009 From: william at wbonnet.net (William Bonnet) Date: Fri, 15 May 2009 21:03:50 +0200 Subject: [csw-maintainers] problems with existing gtk libs In-Reply-To: <20090515180637.GL10154@bolthole.com> References: <20090515180637.GL10154@bolthole.com> Message-ID: <4A0DBC96.8060101@wbonnet.net> Hi Phil > Hmm.. actually... i just noticed that I didnt have a swap file configured. > > this SHOULDNT be a problem, as i HAVE free ram. > > but... maybe this is one o them there "stupid design problems", where > things die if there isnt disk based "backing store". > > will report later if adding swap, has improved things. > It seems to have fixed one reoccuring url crash, at least I am using FF and TB on a daily basis, and i could even say i have a heavy use of these software, and i experienced no crashes of this kind since update to GLib 2.20 Let us know if adding swap does not solve your problem cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From ihsan at opencsw.org Fri May 15 23:56:36 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Fri, 15 May 2009 23:56:36 +0200 Subject: [csw-maintainers] [csw-users] php4.4.9 now in testing In-Reply-To: <20090515170955.GH10154@bolthole.com> References: <4A0CC1B5.3070506@opencsw.org> <4A0D9D40.5020402@opencsw.org> <20090515170955.GH10154@bolthole.com> Message-ID: <4A0DE514.9010404@opencsw.org> Am 15.5.2009 19:09 Uhr, Philip Brown schrieb: > Dont crosspost between the lists please. It makes actually also my live easier, because I have to review all the post made by non-members. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From rupert at opencsw.org Sat May 16 12:05:43 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 12:05:43 +0200 Subject: [csw-maintainers] git without openssh dependency? Message-ID: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> would it make sense to have git without the openssh dependency? we use a special ssh to accept x509 certs, and installing cswgit then disables this. or does cswopenssh contain any specialities needed for git? rupert. From bwalton at opencsw.org Sat May 16 14:42:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 16 May 2009 08:42:47 -0400 Subject: [csw-maintainers] git without openssh dependency? In-Reply-To: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> References: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> Message-ID: <1242477649-sup-4094@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sat May 16 06:05:43 -0400 2009: > would it make sense to have git without the openssh dependency? we use > a special ssh to accept x509 certs, and installing cswgit then > disables this. or does cswopenssh contain any specialities needed for > git? The upcoming 1.6.3 will depend only on the client package. I depended on the full ssh originally since it wasn't split until recently. Git does make use of ssh for a transport tunnel very commonly, so I do need to depend on it...sshd isn't needed. Does that resolve your issues? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From rupert at opencsw.org Sat May 16 14:51:33 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 14:51:33 +0200 Subject: [csw-maintainers] git without openssh dependency? In-Reply-To: <1242477649-sup-4094@ntdws12.chass.utoronto.ca> References: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> <1242477649-sup-4094@ntdws12.chass.utoronto.ca> Message-ID: <6af4270905160551s59dd803ah8c25a583d3f6f49b@mail.gmail.com> On Sat, May 16, 2009 at 14:42, Ben Walton wrote: > Excerpts from rupert THURNER's message of Sat May 16 06:05:43 -0400 2009: >> would it make sense to have git without the openssh dependency? we use >> a special ssh to accept x509 certs, and installing cswgit then >> disables this. or does cswopenssh contain any specialities needed for >> git? > > The upcoming 1.6.3 will depend only on the client package. ?I depended > on the full ssh originally since it wasn't split until recently. ?Git > does make use of ssh for a transport tunnel very commonly, so I do > need to depend on it...sshd isn't needed. > > Does that resolve your issues? sure. as long one can without problem use the commonly installed ssh :) many thanks! rupert. From rupert at opencsw.org Sat May 16 15:21:48 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:21:48 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" Message-ID: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> rupert at build8x:~/mgar/pkg $ svn up ... A cpan/Time-HiRes A cpan/Time-HiRes/trunk svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read response body: Secure connection truncated (https://gar.svn.sourceforge.net) can it be that the processor is too fast for the solaris timeout? it happened 5 times while checking out half of the directory. the other half went without problem on build8s. or is this a subversion "bug"? it does not really harm, as the workaround is re-run the "svn up" ... rupert. From rupert at opencsw.org Sat May 16 15:34:01 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:34:01 +0200 Subject: [csw-maintainers] subversion-1.6.2 in testing, but where does the "UNCOMMITTED" come from? Message-ID: <6af4270905160634t293107e4q439d54707df5f8bb@mail.gmail.com> i compiled svn-1.6.2 and put it into testing. for build8s, i am wondering where the "UNCOMMITTED" comes from in the filename? last time i did not commit the change to the makefile and i guessed that gar is intelligent and picks this up :) but this time i did commit it before building. see on http://mirror.opencsw.org/testing.html, e.g. subversion-1.6.2,REV=2009.05.16-SunOS5.8-sparc-UNCOMMITTED.pkg.gz rupert. From rupert at opencsw.org Sat May 16 15:36:40 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:36:40 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> Message-ID: <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> it happens on build8s also now. i guess it is not the client then and we can forget about it. On Sat, May 16, 2009 at 15:21, rupert THURNER wrote: > rupert at build8x:~/mgar/pkg > $ svn up > ... > A ? ?cpan/Time-HiRes > A ? ?cpan/Time-HiRes/trunk > svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read > response body: Secure connection truncated > (https://gar.svn.sourceforge.net) > > can it be that the processor is too fast for the solaris timeout? it > happened 5 times while checking out half of the directory. the other > half went without problem on build8s. or is this a subversion "bug"? > > it does not really harm, as the workaround is re-run the "svn up" ... > > rupert. > From mwatters at opencsw.org Sat May 16 15:41:45 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 08:41:45 -0500 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> Message-ID: <4A0EC299.1000806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > it happens on build8s also now. i guess it is not the client then and > we can forget about it. > > On Sat, May 16, 2009 at 15:21, rupert THURNER wrote: >> rupert at build8x:~/mgar/pkg >> $ svn up >> ... >> A ? ? cpan/Time-HiRes >> A ? ? cpan/Time-HiRes/trunk >> svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read >> response body: Secure connection truncated >> (https://gar.svn.sourceforge.net) >> >> can it be that the processor is too fast for the solaris timeout? it >> happened 5 times while checking out half of the directory. the other >> half went without problem on build8s. or is this a subversion "bug"? >> >> it does not really harm, as the workaround is re-run the "svn up" ... >> >> rupert. >> > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers do a svn status anything that is not External will cause it to be UNCOMMITTED also, you are not overwriting my recipe(s) for svn are you? ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoOwpgACgkQLrhmsXMSLxcxRwCfVOFSbcNFMDUQAPF5MmH10up0 ITEAnR6Hp3AhB/AhKLyL7Cf/rfiAtxM/ =NRUW -----END PGP SIGNATURE----- From rupert at opencsw.org Sat May 16 15:48:54 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:48:54 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <4A0EC299.1000806@opencsw.org> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> <4A0EC299.1000806@opencsw.org> Message-ID: <6af4270905160648k718dfd37p33bd993350913f4f@mail.gmail.com> On Sat, May 16, 2009 at 15:41, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > rupert THURNER wrote: >> it happens on build8s also now. i guess it is not the client then and >> we can forget about it. >> >> On Sat, May 16, 2009 at 15:21, rupert THURNER wrote: >>> rupert at build8x:~/mgar/pkg >>> $ svn up >>> ... >>> A ? ?? cpan/Time-HiRes >>> A ? ?? cpan/Time-HiRes/trunk >>> svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read >>> response body: Secure connection truncated >>> (https://gar.svn.sourceforge.net) >>> >>> can it be that the processor is too fast for the solaris timeout? it >>> happened 5 times while checking out half of the directory. the other >>> half went without problem on build8s. or is this a subversion "bug"? >>> >>> it does not really harm, as the workaround is re-run the "svn up" ... >>> >>> rupert. >>> >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > do a svn status anything that is not External will cause it to be UNCOMMITTED > > also, you are not overwriting my recipe(s) for svn are you? ;) no way ... just adjusted the version number and everything worked out of the box. impressing design ! From william at wbonnet.net Sat May 16 16:50:21 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 16 May 2009 16:50:21 +0200 Subject: [csw-maintainers] subversion-1.6.2 in testing, but where does the "UNCOMMITTED" come from? In-Reply-To: <6af4270905160634t293107e4q439d54707df5f8bb@mail.gmail.com> References: <6af4270905160634t293107e4q439d54707df5f8bb@mail.gmail.com> Message-ID: <4A0ED2AD.4050802@wbonnet.net> Hi Rupert > i compiled svn-1.6.2 and put it into testing. for build8s, i am > wondering where the "UNCOMMITTED" comes from in the filename? last > time i did not commit the change to the makefile and i guessed that > gar is intelligent and picks this up :) but this time i did commit it > before building. > You're right, GAR detects that some files are not commited. What is the output of svn status in the trunk dir ? For example here is the output with mdified Makefile and checksum [wbonnet at build8s:~/mgar/pkg/thunderbird/trunk]$ svn status X gar M checksums M Makefile cheers W. From rupert at opencsw.org Sat May 16 20:41:10 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 20:41:10 +0200 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? Message-ID: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> i tried to create a package for libserf, which should replace libneon in future, as it is able to multithread subversion calls to a server. as there is no libapr, i tried to depend on apache2 by: CONFIGURE_ARGS = $(DIRPATHS) CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 #CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config #CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config (see also: https://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/libserf/trunk/Makefile ) but, for some reason, /opt/csw/bdb44/lib makes it into the compile, and not /opt/csw/apache2/lib. gmake[2]: Entering directory `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' /opt/csw/apache2/share/build/libtool --silent --mode=link /opt/studio/SOS11/SUNWspro/bin/cc -xarch=386 -L/opt/csw/lib -L/opt/csw/bdb44/lib -static -o test/serf_get libserf-0.la test/serf_get.lo /opt/csw/apache2/lib/libaprutil-1.la -lldap -llber -ldb-4.4 -lexpat -liconv -lapr-1 -luuid -lsendfile -lrt -lsocket -lpthread -ldl -lm -lz -lssl -lcrypto ld: fatal: library -lapr-1: not found ld: fatal: File processing errors. No output written to test/serf_get gmake[2]: *** [test/serf_get] Error 1 gmake[2]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' gmake[1]: *** [build-work/build-isa-i386/serf-0.3.0/Makefile] Error 2 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk' gmake: *** [merge-isa-i386] Error 2 what can this be? rupert. From phil at bolthole.com Sat May 16 23:31:49 2009 From: phil at bolthole.com (Philip Brown) Date: Sat, 16 May 2009 14:31:49 -0700 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> Message-ID: <20090516213149.GB26650@bolthole.com> On Sat, May 16, 2009 at 08:41:10PM +0200, rupert THURNER wrote: > i tried to create a package for libserf, which should replace libneon > in future, as it is able to multithread subversion calls to a server. > > as there is no libapr, i tried to depend on apache2 by: > > CONFIGURE_ARGS = $(DIRPATHS) > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 >... hrrrm.. we reaaaally should split out a libapr, if we have things needing it. From mwatters at opencsw.org Sun May 17 03:33:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 20:33:22 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <20090516213149.GB26650@bolthole.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <20090516213149.GB26650@bolthole.com> Message-ID: <4A0F6962.1000000@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Sat, May 16, 2009 at 08:41:10PM +0200, rupert THURNER wrote: >> i tried to create a package for libserf, which should replace libneon >> in future, as it is able to multithread subversion calls to a server. >> >> as there is no libapr, i tried to depend on apache2 by: >> >> CONFIGURE_ARGS = $(DIRPATHS) >> CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 >> ... > > > hrrrm.. we reaaaally should split out a libapr, if we have things needing > it. > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers we already have an apache2rt package, Looking at the contents, it has a few development files included. I filed a bug http://www.opencsw.org/mantis/view.php?id=3676 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoPaWIACgkQLrhmsXMSLxfyEQCg2iN+Jp3YYYRRvuQ6hUMZ84sX mmgAoMwkXqnNhwyklSAjRgwd3aU2nK74 =kHBx -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 03:41:11 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 20:41:11 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> Message-ID: <4A0F6B37.9090703@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > i tried to create a package for libserf, which should replace libneon > in future, as it is able to multithread subversion calls to a server. > > as there is no libapr, i tried to depend on apache2 by: > > CONFIGURE_ARGS = $(DIRPATHS) > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 > CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 > #CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config > #CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config > > (see also: https://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/libserf/trunk/Makefile > ) > > but, for some reason, /opt/csw/bdb44/lib makes it into the compile, > and not /opt/csw/apache2/lib. > > what can this be? libapr is part of the apache build. the --with-apr and --with-apr-util configure args typically point to the apr-config and apu-config files. are you building on the build farm or your own server? To build against apr you need to have the following installed: application CSWapache2-devel apache2_devel - Apache 2.2 development support application CSWapache2rt apache2rt - Apache 2.2 runtime libraries and then you will need to create a dependency on apache2rt. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoPazcACgkQLrhmsXMSLxffMwCgoXG0LKtkJcnU7jOCCe+ya0GB 0XcAnjXh6XguwTp4jEYq+agwtnGYfJHE =BKmv -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 06:55:09 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 23:55:09 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> Message-ID: <4A0F98AD.7050508@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > > gmake[2]: Entering directory > `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' > /opt/csw/apache2/share/build/libtool --silent --mode=link > /opt/studio/SOS11/SUNWspro/bin/cc -xarch=386 -L/opt/csw/lib > -L/opt/csw/bdb44/lib -static -o test/serf_get libserf-0.la > test/serf_get.lo /opt/csw/apache2/lib/libaprutil-1.la -lldap -llber > -ldb-4.4 -lexpat -liconv -lapr-1 -luuid -lsendfile -lrt -lsocket > -lpthread -ldl -lm -lz -lssl -lcrypto > ld: fatal: library -lapr-1: not found > ld: fatal: File processing errors. No output written to test/serf_get > gmake[2]: *** [test/serf_get] Error 1 > gmake[2]: Leaving directory > `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' > gmake[1]: *** [build-work/build-isa-i386/serf-0.3.0/Makefile] Error 2 > gmake[1]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk' > gmake: *** [merge-isa-i386] Error 2 > > what can this be? I tried to build it on build8s and found the problem. ;) there was a bug with the fixlibtool script. and a bug in gar that was forcing it to run for every build with mGar. I have corrected both the mGar bug fixed in r4909 and the fixlibtool bug is fixed in r4910 make the changes below to your recipe and update your version of gar then run make clean && make test # UPSTREAM_MASTER_SITES = DEPENDS = server/apache2 STRIP_LIBTOOL = 1 CONFIGURE_ARGS = $(DIRPATHS) CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config TEST_TARGET = check include gar/category.mk - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoPmK0ACgkQLrhmsXMSLxd6bwCfcOi6GHN/6YtiR4VqoFW0oSlK GagAnjRDWg9s6Pxi+QvXxd6yX64rxgER =Dukp -----END PGP SIGNATURE----- From rupert at opencsw.org Sun May 17 09:31:16 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 17 May 2009 09:31:16 +0200 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <4A0F98AD.7050508@opencsw.org> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> Message-ID: <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> On Sun, May 17, 2009 at 06:55, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi rupert, > rupert THURNER wrote: >> > > make the changes below to your recipe and update your version of gar > then run make clean && make test > > # UPSTREAM_MASTER_SITES = > > DEPENDS = server/apache2 > STRIP_LIBTOOL = 1 > > CONFIGURE_ARGS = $(DIRPATHS) > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config > CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config many thanks that works! what is the reason that you recooment the args above contrary to: CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 ? rupert. From rupert at opencsw.org Sun May 17 09:37:00 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 17 May 2009 09:37:00 +0200 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> Message-ID: <6af4270905170037l6b81534wbdcb66557aee972f@mail.gmail.com> On Sun, May 17, 2009 at 09:31, rupert THURNER wrote: > On Sun, May 17, 2009 at 06:55, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi rupert, >> rupert THURNER wrote: >>> >> >> make the changes below to your recipe and update your version of gar >> then run make clean && make test >> >> # UPSTREAM_MASTER_SITES = >> >> DEPENDS = server/apache2 >> STRIP_LIBTOOL = 1 what does https://apps.sourceforge.net/trac/gar/changeset/4909 do, resp what did it before that was wrong? rupert. From dam at opencsw.org Sun May 17 15:32:49 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 17 May 2009 15:32:49 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> Message-ID: <5F5B3419-D28C-468D-B281-F1BDB853F931@opencsw.org> Hi Rupert, Am 16.05.2009 um 15:36 schrieb rupert THURNER: > it happens on build8s also now. i guess it is not the client then and > we can forget about it. You are aware that there are maintenance tasks during this time? If any problems happen to SourceForge or Trac, please see Best regards -- Dago From mwatters at opencsw.org Sun May 17 16:44:14 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 09:44:14 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> Message-ID: <4A1022BE.6050501@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > many thanks that works! what is the reason that you recooment the args > above contrary to: > > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 > CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 > when you specify the --with-apr=$(prefix)/apache2 the first thing configure does is look in $(prefix)/apache2/bin for apr-config same thing for the util. so it is redundent to put both in the configure args. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQIr0ACgkQLrhmsXMSLxc60wCgpyZfFBh/o6C8pfUlpdAm/mMS w5UAoIdrxvfqemIIjj/5gT71/s1sbcsT =Gy7L -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 17:04:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 10:04:30 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905170037l6b81534wbdcb66557aee972f@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> <6af4270905170037l6b81534wbdcb66557aee972f@mail.gmail.com> Message-ID: <4A10277E.3040309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > what does https://apps.sourceforge.net/trac/gar/changeset/4909 do, > resp what did it before that was wrong? Basically, what was wrong: in r4784 Dago put in a fix to simplify my code in gar.mk to use the new STRIP_LIBTOOL variable. to see the diff run - svn diff -r4720:4784 gar.mk this changed the requirement to have STRIP_LIBTOOL ?= 0 to not fix libtool to removing the variable all together. The only thing r4909 does is remove the variable STRIP_LIBTOOL from gar.conf.mk to see the diff run - svn diff -r4765:4909 gar.conf.mk - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQJ34ACgkQLrhmsXMSLxc+DQCfVBQ38SX2g3Dh6ND4d4vcUZR0 QXwAoOAVJpKLJKTazbTsRbY1rHRVRIlo =5+Zo -----END PGP SIGNATURE----- From dam at opencsw.org Sun May 17 17:33:46 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 17 May 2009 17:33:46 +0200 Subject: [csw-maintainers] Main "Outstanding bugs" Message-ID: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> Hi, this great email notifying of open mails would be even more useful if it would include links to the maintainer bugs page: I find it much easier to look at that page to give me an immediate impression of what needs to be done instead of clicking through all links for individual bugs. Could someone (William?) please add this link? Best regards -- Dago From mwatters at opencsw.org Sun May 17 17:53:11 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 10:53:11 -0500 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> Message-ID: <4A1032E7.9080609@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi, > > this great email notifying of open mails would be even more useful if > it would include links to the maintainer bugs page: > > > I find it much easier to look at that page to give me an immediate > impression of what needs to be done instead of clicking through > all links for individual bugs. > > Could someone (William?) please add this link? as long as we are asking for things, can we get the "resolved" but not "closed" bugs added to the maintainer buglist page http://www.opencsw.org/buglist/maintainer.cgi?maintainer=mwatters once the bug is marked as resolved, I do this when the bug fix goes into testing. I then will close it when the package gets released. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQMucACgkQLrhmsXMSLxcbvACffU24/msEkLpdna3eyZFvZ1Na xC8AoMw8DkEWfbjTV8eRjKc6s4RWovXW =UG56 -----END PGP SIGNATURE----- From william at wbonnet.net Sun May 17 19:08:32 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 17 May 2009 19:08:32 +0200 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> Message-ID: <4A104490.60302@wbonnet.net> Hi Dago > I find it much easier to look at that page to give me an immediate > impression of what needs to be done instead of clicking through > all links for individual bugs. > > Could someone (William?) please add this link? It's a good idea, but i'm affraid i can't do it. I don't have access to this script. Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Sun May 17 20:01:38 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 13:01:38 -0500 Subject: [csw-maintainers] python 2.5.5 now in testing Message-ID: <4A105102.9080404@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: Update to version 2.5.5 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQUQIACgkQLrhmsXMSLxc+egCgovfDFfpbyEjpoaUp73dL3LJj 3+UAn01fX0+xDKqaTCPiUlT2OGb25Vs2 =tFoU -----END PGP SIGNATURE----- From phil at bolthole.com Sun May 17 20:08:25 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 17 May 2009 11:08:25 -0700 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <4A1032E7.9080609@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> Message-ID: <20090517180825.GA86583@bolthole.com> On Sun, May 17, 2009 at 10:53:11AM -0500, Mike Watters wrote: > can we get the "resolved" but not "closed" bugs > added to the maintainer buglist page > http://www.opencsw.org/buglist/maintainer.cgi?maintainer=mwatters i bugged him to take it OUT, i think :-) From mwatters at opencsw.org Sun May 17 23:00:47 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:00:47 -0500 Subject: [csw-maintainers] TYPO pysqlite2.5.5 now in testing not python In-Reply-To: <4A105102.9080404@opencsw.org> References: <4A105102.9080404@opencsw.org> Message-ID: <4A107AFF.5050005@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > changes: > Update to version 2.5.5 _______________________________________________ maintainers mailing list maintainers at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQev8ACgkQLrhmsXMSLxd7AwCeK01ETUySwBGYGBjaOgn5mZ2H UoIAnAit5KMjTDOviOQhokUxPC28HGJk =5CJ1 -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 23:02:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:02:22 -0500 Subject: [csw-maintainers] gd 2.0.35 now in testing Message-ID: <4A107B5E.8080801@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Changes: relinked the amd64 build against the correct fontconfig library - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQe14ACgkQLrhmsXMSLxdanACg4Cgcu27YBlhfFimXaxMOZg3g VXIAoJxxLtMyg6UYmtw28UhGvzgpvtWS =vH2u -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 23:06:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:06:17 -0500 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <20090517180825.GA86583@bolthole.com> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> <20090517180825.GA86583@bolthole.com> Message-ID: <4A107C49.2020303@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Sun, May 17, 2009 at 10:53:11AM -0500, Mike Watters wrote: >> can we get the "resolved" but not "closed" bugs >> added to the maintainer buglist page >> http://www.opencsw.org/buglist/maintainer.cgi?maintainer=mwatters > > i bugged him to take it OUT, i think :-) > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers could we get a link to a generated second page that contains the resolved bugs? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQfEkACgkQLrhmsXMSLxeSUwCg7EIt3A/YSSS5Yra81Ustk4GX AXkAmgJ9QJ7bvhxS6ydQ3JE8kXZhNwxx =5tHF -----END PGP SIGNATURE----- From dam at opencsw.org Sun May 17 23:27:08 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 17 May 2009 23:27:08 +0200 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <4A107C49.2020303@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> <20090517180825.GA86583@bolthole.com> <4A107C49.2020303@opencsw.org> Message-ID: <5E106507-F8F9-49DD-A74B-0A1AA0148401@opencsw.org> Hi Mike, Am 17.05.2009 um 23:06 schrieb Mike Watters: > could we get a link to a generated second page > that contains the resolved bugs? The page is already there, and by coincidence it is exactly the one I proposed to link on: Best regards -- Dago From mwatters at opencsw.org Sun May 17 23:37:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:37:51 -0500 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <5E106507-F8F9-49DD-A74B-0A1AA0148401@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> <20090517180825.GA86583@bolthole.com> <4A107C49.2020303@opencsw.org> <5E106507-F8F9-49DD-A74B-0A1AA0148401@opencsw.org> Message-ID: <4A1083AF.5080606@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi Mike, > > Am 17.05.2009 um 23:06 schrieb Mike Watters: >> could we get a link to a generated second page >> that contains the resolved bugs? > > The page is already there, and by coincidence it is exactly the > one I proposed to link on: > > > > Best regards > > -- Dago > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers yes but it does not show the resolved bugs, unless there is another "open" bug for the same project. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQg68ACgkQLrhmsXMSLxeGTgCgx4fxdWGe/0a7OrdEo9eLjf86 LycAoIDoduHSot+XofW8HyafnJNmznPj =D+Gr -----END PGP SIGNATURE----- From dam at opencsw.org Mon May 18 00:37:11 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 00:37:11 +0200 Subject: [csw-maintainers] Package "Trac" in testing Message-ID: <5A6F47B5-81E9-45E6-B8E4-9FFD33A4BF7A@opencsw.org> Hi, I noticed that the package name for Trac 0.11.4 is "Trac" instead of the existing "trac". Is this intended? Best regards -- Dago From mwatters at opencsw.org Mon May 18 04:49:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 21:49:13 -0500 Subject: [csw-maintainers] python 2.6.2 in testing Message-ID: <4A10CCA9.3080608@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: moved the modules to an non-version dependent directory. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQzKkACgkQLrhmsXMSLxfs7gCfa/JWkoQc1PJgqVjKrXrwzOgi YEYAn0wkjnHjaq/rsr2myArx+G7ZCiY4 =j+Ed -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Mon May 18 10:38:32 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Mon, 18 May 2009 10:38:32 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: Your message of "Mon, 18 May 2009 00:37:11 +0200." <5A6F47B5-81E9-45E6-B8E4-9FFD33A4BF7A@opencsw.org> Message-ID: <200905180838.n4I8cWFo010338@dfki.uni-kl.de> > Hi, > > I noticed that the package name for Trac 0.11.4 is "Trac" instead of the > existing "trac". Is this intended? No, sorry this is a typo - which I will fix today. I'll have to repackage again as pysqlite 2.5.5 is now available. Nicolai From mwatters at opencsw.org Mon May 18 15:33:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 18 May 2009 08:33:35 -0500 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: <200905180838.n4I8cWFo010338@dfki.uni-kl.de> References: <200905180838.n4I8cWFo010338@dfki.uni-kl.de> Message-ID: <4A1163AF.3050006@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Nicolai Schwindt wrote: >> Hi, >> >> I noticed that the package name for Trac 0.11.4 is "Trac" instead of the >> existing "trac". Is this intended? > > No, sorry this is a typo - which I will fix today. > I'll have to repackage again as pysqlite 2.5.5 is now available. > > > > Nicolai > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Hold off on the rebuild against pysqlite 2.5.5 I want to re-compile that again after the python in testing is released. the python in testing moves the module location to the non-version dependent directory. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoRY68ACgkQLrhmsXMSLxfcgwCfaG8gp+T92a/QkKGluo1xeg75 Xz8An2x24/uBWvMzBvZefHY57PhUuqcj =Yy28 -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Mon May 18 16:18:46 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Mon, 18 May 2009 16:18:46 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: Your message of "Mon, 18 May 2009 08:33:35 CDT." <4A1163AF.3050006@opencsw.org> Message-ID: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> [...] > Hold off on the rebuild against pysqlite 2.5.5 I want to > re-compile that again after the python in testing is released. > > the python in testing moves the module location to the non-version dependent > directory. Which brings me into trouble - as I am on the 30 day probation. /me looking for another package .) From dam at opencsw.org Mon May 18 16:31:32 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 16:31:32 +0200 Subject: [csw-maintainers] New members welcome! Message-ID: Hi, the association welcomes the newly accepted members - Mike Watters Mike already maintains several large packages including gcc, python and php. Please keep in mind that being a maintainer is not only about fame and glory, but also about tedious work to make the packages as good as possible and remove bugs timely when discovered. Please check regularly at the bottom of your maintainer page if there are any open issues. If you have spare cycles please adopt an orphaned package and help bring the complete software stack to a 100% current state. But enough of morality: A very warm welcome! Your membership is tracked at Please let me know on what are you working or are planning to work like "webpage", "maintainer", etc. If you have applied for membership and don't see your name anywhere above please let me know. There has been a lot of mail traffic for me to be processed in the past weeks and I cannot guarantee that I didn't missed one. Best regards -- Dago From dam at opencsw.org Mon May 18 16:33:28 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 16:33:28 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> References: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> Message-ID: <9BA99441-F4AD-4A7A-918F-51FA422A9BDA@opencsw.org> Hi Nicolai, Am 18.05.2009 um 16:18 schrieb Nicolai Schwindt: > [...] >> Hold off on the rebuild against pysqlite 2.5.5 I want to >> re-compile that again after the python in testing is released. >> >> the python in testing moves the module location to the non-version >> dependent >> directory. > > Which brings me into trouble - as I am on the 30 day probation. > > /me looking for another package .) Feel free to adpopt any from the "most wanted"-list or from maintainers who have retired: Best regards -- Dago From mwatters at opencsw.org Mon May 18 17:00:59 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 18 May 2009 10:00:59 -0500 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> References: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> Message-ID: <4A11782B.1010400@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Nicolai Schwindt wrote: > [...] >> Hold off on the rebuild against pysqlite 2.5.5 I want to >> re-compile that again after the python in testing is released. >> >> the python in testing moves the module location to the non-version dependent >> directory. > > Which brings me into trouble - as I am on the 30 day probation. > > /me looking for another package .) > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I will get python tested today and try to fast track phil into release so I can get it out quickly for you. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoReCoACgkQLrhmsXMSLxcdSgCgq7GE30l1a4nNECWrOLetjsTO JhAAnRdrFOzo4nZeLWONslIq0ZU5YjNR =EQQ+ -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Mon May 18 17:03:41 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Mon, 18 May 2009 17:03:41 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: Your message of "Mon, 18 May 2009 10:00:59 CDT." <4A11782B.1010400@opencsw.org> Message-ID: <200905181503.n4IF3f0o018606@dfki.uni-kl.de> > > I will get python tested today and try to fast track phil into release so I c an > get it out quickly for you. Its ok, I'll find something else .) From ellson at opencsw.org Mon May 18 18:10:17 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 12:10:17 -0400 Subject: [csw-maintainers] howto: get commit privilege to Subversion? Message-ID: <4A118869.8050008@opencsw.org> I have my first package, gts, building in mgar, and I'm ready to commit. How do I get commit privileges into Subversion for this package? I'm getting: Authentication realm: SourceForge Subversion area Password for 'ellson': John From dam at opencsw.org Mon May 18 18:40:31 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 18:40:31 +0200 Subject: [csw-maintainers] howto: get commit privilege to Subversion? In-Reply-To: <4A118869.8050008@opencsw.org> References: <4A118869.8050008@opencsw.org> Message-ID: <63C907AF-6F2D-4FAC-A8AE-DB064A675BB2@opencsw.org> Hi John, Am 18.05.2009 um 18:10 schrieb John Ellson: > I have my first package, gts, building in mgar, and I'm ready to > commit. > > How do I get commit privileges into Subversion for this package? > > I'm getting: > > Authentication realm: > SourceForge Subversion area > Password for 'ellson': Please send me your SourceForge account name and the "SF Team" (that is Sebastian and me) will grant you commit rights. Best regards -- Dago From ihsan at opencsw.org Mon May 18 21:14:55 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 18 May 2009 21:14:55 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 Message-ID: <4A11B3AF.7090709@opencsw.org> Hello, The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure out when it should take place, I've create a Doodle poll. Please fill out the link bellow: http://www.doodle.com/6zr9kmdr8p9vv69p NOTE: If possible, always select a full weekend (Saturday and Sunday). I think it's the best if we will do it as last year. We should meet for socialising on Friday evening, while on Saturday and Sunday we will work. @Trygve: Thanks again for organizing. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ellson at opencsw.org Mon May 18 21:28:28 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 15:28:28 -0400 Subject: [csw-maintainers] Package "gts" in testing Message-ID: <4A11B6DC.6000803@opencsw.org> There is a new package: "gts: GNU Triangulated Surface Library" now available in testing. This is my first package. It was built with mgar, so a review of my mgar/pkg/gts/ tree in svn would also be appreciated. I'm building this package because graphviz can use it, and because it is much smaller than graphviz for my first attempt. Your feedback please. John From dam at opencsw.org Mon May 18 22:45:28 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 22:45:28 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <4A11B6DC.6000803@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> Message-ID: Hi John, Am 18.05.2009 um 21:28 schrieb John Ellson: > There is a new package: > > "gts: GNU Triangulated Surface Library" > > now available in testing. > > This is my first package. It was built with mgar, so a review of > my mgar/pkg/gts/ tree in svn would also be appreciated. Some suggestions: - Please use for SourceForge MASTER_SITES = $(SF_MIRRORS) For several big sites the URLs have been stored centrally. - For SourceForge the upstream watch works slightly different due to the structure of the download page. Please use UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to your id) UPSTREAM_USE_SF = 1 - You are using a static depend CSWgts.depend and dynamic depends with REQUIRED_PKGS. They both serve the same purpose and defining both will ignore REQUIRED_PKGS. Please remove CSWgts.depend and the notion of it in DISTFILES. - You are using a static gspec file CSWgts.gspec. As the bitname and pkgname are the same as the GARNAME the defaults will be fine. You can safely remove the gspec-file and the notion of it in DISTFILES without change in behaviour. Static gspec files are also old-style and the newer variants from should be used wherever possible. - .la-files are already excluded during merge by default. See for details - The NO_ISAEXEC line is only important when you build multiple ISAs (like the commented out 64 bit ISA with BUILD64=1). When building only one ISA it does nothing. These are a lot of comments. Please take them as result of the, umh, imperfect documentation. Which documentation have you missed most on your path? What can I do to improve your getting started experience with GAR? Best regards -- Dago From mwatters at opencsw.org Mon May 18 23:03:20 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 18 May 2009 16:03:20 -0500 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: References: <4A11B6DC.6000803@opencsw.org> Message-ID: <4A11CD18.2020805@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago: Dagobert Michelsen wrote: > What can I do to improve your getting started experience with GAR? provide Round Trip Air Fair and Hotel Accommodation's so we can sit with you and pick your brain for a month. ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoRzRgACgkQLrhmsXMSLxcZrgCfSTulprgmXEFPaogET3QPzsAY c34AoKHFkKv+OW9Bewocpep7lPok8UZX =8iCz -----END PGP SIGNATURE----- From ellson at opencsw.org Mon May 18 23:36:40 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 17:36:40 -0400 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: References: <4A11B6DC.6000803@opencsw.org> Message-ID: <4A11D4E8.6080907@opencsw.org> On 05/18/2009 04:45 PM, Dagobert Michelsen wrote: > > - For SourceForge the upstream watch works slightly different due to the > structure of the download page. Please use > UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to > your id) What ID is this? My sourceforge id for me? Or some id for the GTS project? How do i find what it is? John From ellson at opencsw.org Tue May 19 00:46:35 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 18:46:35 -0400 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: References: <4A11B6DC.6000803@opencsw.org> Message-ID: <4A11E54B.3030406@opencsw.org> On 05/18/2009 04:45 PM, Dagobert Michelsen wrote: > Hi John, > > Am 18.05.2009 um 21:28 schrieb John Ellson: >> There is a new package: >> >> "gts: GNU Triangulated Surface Library" >> >> now available in testing. >> >> This is my first package. It was built with mgar, so a review of >> my mgar/pkg/gts/ tree in svn would also be appreciated. > > Some suggestions: > - Please use for SourceForge > MASTER_SITES = $(SF_MIRRORS) > For several big sites the URLs have been stored centrally. > - For SourceForge the upstream watch works slightly different due to the > structure of the download page. Please use > UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to > your id) > UPSTREAM_USE_SF = 1 > - You are using a static depend CSWgts.depend and dynamic depends with > REQUIRED_PKGS. They both serve the same purpose and defining both > will ignore REQUIRED_PKGS. Please remove CSWgts.depend and the notion > of it in DISTFILES. > - You are using a static gspec file CSWgts.gspec. As the bitname and > pkgname > are the same as the GARNAME the defaults will be fine. You can safely > remove the gspec-file and the notion of it in DISTFILES without change > in behaviour. Static gspec files are also old-style and the newer > variants from > > should be used wherever possible. > - .la-files are already excluded during merge by default. See > > > for details > - The NO_ISAEXEC line is only important when you build multiple ISAs > (like the commented out 64 bit ISA with BUILD64=1). When building only > one ISA it does nothing. > > These are a lot of comments. Please take them as result of the, umh, > imperfect > documentation. Which documentation have you missed most on your path? > What can I do to improve your getting started experience with GAR? OK, Got those. New gts packages posted to /home/testing. I'll work on some notes about GAR for you.... John From dam at opencsw.org Tue May 19 01:15:59 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 01:15:59 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <4A11D4E8.6080907@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11D4E8.6080907@opencsw.org> Message-ID: <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> Hi John, Am 18.05.2009 um 23:36 schrieb John Ellson: > On 05/18/2009 04:45 PM, Dagobert Michelsen wrote: >> >> - For SourceForge the upstream watch works slightly different due >> to the >> structure of the download page. Please use >> UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change >> to your id) > > What ID is this? My sourceforge id for me? Or some id for the GTS > project? How do i find what it is? First you go to the project main page at and then Download -> Browse All Packages This leads to Aha! -> 1343 AFAIK there is no easier way on SF to do this, William did some research when he implemented upsream watch. Best regards -- Dago From dam at opencsw.org Tue May 19 01:16:36 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 01:16:36 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <4A11CD18.2020805@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11CD18.2020805@opencsw.org> Message-ID: <2900082C-0D62-4043-8CC2-F6742A1EFCFE@opencsw.org> Hi Mike, Am 18.05.2009 um 23:03 schrieb Mike Watters: >> What can I do to improve your getting started experience with GAR? > > provide Round Trip Air Fair and Hotel Accommodation's so we can sit > with you > and pick your brain for a month. ;) Good thing we can just do that during the Summercamp :-) Best regards -- Dago From phil at bolthole.com Tue May 19 01:42:10 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 18 May 2009 16:42:10 -0700 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11D4E8.6080907@opencsw.org> <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> Message-ID: <20090518234210.GE13590@bolthole.com> On Tue, May 19, 2009 at 01:15:59AM +0200, Dagobert Michelsen wrote: > Hi John, >... >>> UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to > First you go to the project main page at > > and then Download -> Browse All Packages > This leads to > > Aha! -> 1343 AFAIK there is no easier way on SF to do this, > William did some research when he implemented upsream watch. I would like to suggest that there IS a way. If you have the sourceforge name, ie "gts". you then go to the sourceforge-standardized page, as suggested http://sourceforge.net/projects/$PROJNAME You can then use such fanciness as wget --quiet -O - http://sourceforge.net/projects/gts | sed -n 's/.*showfiles.php[?]group_id=\([0-9][0-9]*\).*/\1/p' | head -1 to automatically derive the sf_project_id. or if you prefer curl, use "curl -s" instead of "wget --quiet -O -" :-) From phil at bolthole.com Tue May 19 01:55:29 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 18 May 2009 16:55:29 -0700 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <20090518234210.GE13590@bolthole.com> References: <4A11B6DC.6000803@opencsw.org> <4A11D4E8.6080907@opencsw.org> <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> <20090518234210.GE13590@bolthole.com> Message-ID: <20090518235529.GG13590@bolthole.com> To follow up on my own email, and "connect the dots"... i guess the overall goal is to define the UPSTREAM_MASTER_SITES variable in gar spec files. I think it would be nice if, once it was defined somewhere in the spec file that [this is the sourceforge project name] that gar would call some kind of gar utility wrapper script to derive the download url using the "latest" sf.net descrambler :-) > wget --quiet -O - http://sourceforge.net/projects/gts | > sed -n 's/.*showfiles.php[?]group_id=\([0-9][0-9]*\).*/\1/p' | > head -1 > > to automatically derive the sf_project_id. [and/or auto-derive the "showfiles url"] From schwindt at dfki.uni-kl.de Tue May 19 09:38:14 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Tue, 19 May 2009 09:38:14 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing Message-ID: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> Wireshark is now in testing, as this is my first package for CSW any feed back is welcome. If non-privileged users shoulkd be able to use this, one would have to set the s bit to /opt/csw/bin/dumpcap. As this is a security breach I have not done that as a default. Is this ok with the opencsw standards ? Nicolai Schwindt. From ihsan at opencsw.org Tue May 19 11:58:48 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 19 May 2009 11:58:48 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <2900082C-0D62-4043-8CC2-F6742A1EFCFE@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11CD18.2020805@opencsw.org> <2900082C-0D62-4043-8CC2-F6742A1EFCFE@opencsw.org> Message-ID: <4A1282D8.3080806@opencsw.org> Am 19.5.2009 1:16 Uhr, Dagobert Michelsen schrieb: >>> What can I do to improve your getting started experience with GAR? >> >> provide Round Trip Air Fair and Hotel Accommodation's so we can sit >> with you >> and pick your brain for a month. ;) > > Good thing we can just do that during the Summercamp :-) Please fill out the poll, so we could fix a date. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Tue May 19 12:53:44 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 12:53:44 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> References: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> Message-ID: <161F75CA-3A78-48F3-8B19-1275D352BDD5@opencsw.org> Hi Nicolai, Am 19.05.2009 um 09:38 schrieb Nicolai Schwindt: > Wireshark is now in testing, as this is my first package for CSW any > feed back > is welcome. > > If non-privileged users shoulkd be able to use this, one would have > to set the > s bit to /opt/csw/bin/dumpcap. As this is a security breach I have > not done > that as a default. > > Is this ok with the opencsw standards ? Sounds reasonable. You can describe additional package use on the wiki at Best regards -- Dago From pfelecan at opencsw.org Tue May 19 13:46:08 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 19 May 2009 13:46:08 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <161F75CA-3A78-48F3-8B19-1275D352BDD5@opencsw.org> (Dagobert Michelsen's message of "Tue\, 19 May 2009 12\:53\:44 +0200") References: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> <161F75CA-3A78-48F3-8B19-1275D352BDD5@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Nicolai, > > Am 19.05.2009 um 09:38 schrieb Nicolai Schwindt: >> Wireshark is now in testing, as this is my first package for CSW any >> feed back >> is welcome. >> >> If non-privileged users shoulkd be able to use this, one would have >> to set the >> s bit to /opt/csw/bin/dumpcap. As this is a security breach I have >> not done >> that as a default. >> >> Is this ok with the opencsw standards ? > > Sounds reasonable. You can describe additional package use on the > wiki at Sure, wikis are a nice thing. Even nicer is to provide a README.CSW in the documentation directory where you explain the idiosyncrasies of your package. -- Peter From dam at opencsw.org Tue May 19 13:54:15 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 13:54:15 +0200 Subject: [csw-maintainers] New maintainers welcome! Message-ID: Hi, the association welcomes the newly accepted members - Peter Felecan Peter has been maintaining packages for a long time now including the previous GCC build and is currently busy with graphic and multimedia libraries. Please keep in mind that being a maintainer is not only about fame and glory, but also about tedious work to make the packages as good as possible and remove bugs timely when discovered. Please check regularly at the bottom of your maintainer page if there are any open issues. If you have spare cycles please adopt an orphaned package and help bring the complete software stack to a 100% current state. But enough of morality: A very warm welcome! Your membership is tracked at Please let me know on what are you working or are planning to work like "webpage", "maintainer", etc. If you have applied for membership and don't see your name anywhere above please let me know. There has been a lot of mail traffic for me to be processed in the past weeks and I cannot guarantee that I didn't missed one. Best regards -- Dago From ellson at opencsw.org Tue May 19 16:53:34 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 10:53:34 -0400 Subject: [csw-maintainers] Offer to maintain swig (was: Re: status report: graphviz packaging - new maintainer) In-Reply-To: <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> Message-ID: <4A12C7EE.2000103@opencsw.org> Dagobert Michelsen wrote: > > Swig is also unmaintained at the moment and outdated. Feel free to > take it > over or to ask for somebody else to update it if you need a newer > version. I'll take it. Do I have svn commit privileges for it now? May I commit? I've updated to swig-1.3.39 in my local tree and tried to update the Makefile for latest GAR practices as I understand them, like trying to use the automatic spec and depend mechanisms. "make build" is ok, and "make package" does a lot of work and seems happy, but doesn't leave me with a package in the staging area? Could someone look at the attached Makefile and tell me what I'm doing wrong? John -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Makefile URL: From mwatters at opencsw.org Tue May 19 17:03:04 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 19 May 2009 10:03:04 -0500 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12C7EE.2000103@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> Message-ID: <4A12CA28.60400@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > spec and depend mechanisms. "make build" is ok, and "make package" > does a lot of work and seems happy, but doesn't leave me with a package in > the staging area? Could someone look at the attached Makefile and tell > me what I'm doing wrong? I don't see a problem with your Makefile ... what does your ~/.garrc file look like? This is mine-> SPKG_PACKAGER = Mike Watters SPKG_EMAIL = mwatters at opencsw.org SPKG_EXPORT = $(HOME)/newpkgs/$(GARNAME) SPKG_SPOOLDIR = $(HOME)/spool/$(GAROSREL)-$(GARCH) GARCHIVEDIR = $(HOME)/src ** SPKG_PACKAGER and SPKG_EMAIL ** needs no explination ** SPKG_EXPORT ** is where gar will put packages when they are done. ** SPKG_SPOOLDIR ** is where gar will "spool" the package before translate to datastream ** GARCHIVEDIR ** is where gar will look for source tarballs before downloading them. helpful if your source tarball is very large. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoSyigACgkQLrhmsXMSLxc6IQCfZb71oUPIKJxQjO8LmBKphl0p Of4AoIEiDRjY0Bl8ra85/1sZaSnnuQRr =1ZTf -----END PGP SIGNATURE----- From ellson at opencsw.org Tue May 19 17:11:12 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 11:11:12 -0400 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12CA28.60400@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> Message-ID: <4A12CC10.7030804@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Ellson wrote: > > >> spec and depend mechanisms. "make build" is ok, and "make package" >> does a lot of work and seems happy, but doesn't leave me with a package in >> the staging area? Could someone look at the attached Makefile and tell >> me what I'm doing wrong? >> > > I don't see a problem with your Makefile ... what does your ~/.garrc file look > like? > This is mine-> > > SPKG_PACKAGER = Mike Watters > SPKG_EMAIL = mwatters at opencsw.org > SPKG_EXPORT = $(HOME)/newpkgs/$(GARNAME) > SPKG_SPOOLDIR = $(HOME)/spool/$(GAROSREL)-$(GARCH) > GARCHIVEDIR = $(HOME)/src > > ** SPKG_PACKAGER and SPKG_EMAIL ** needs no explination > > ** SPKG_EXPORT ** is where gar will put packages when they are done. > ** SPKG_SPOOLDIR ** is where gar will "spool" the package before translate to > datastream > ** GARCHIVEDIR ** is where gar will look for source tarballs before downloading > them. > helpful if your source tarball is very large. > > - -- > > Thanks, > Mike > > > I have this in my .garrc. It works ok for gts packages. John ----- # vim: ft=make # # ~/.garrc - Local settings file for the CSW build system. # # For the most part, any setting found in gar.conf.mk can be overridden by # setting (or unsetting) the variable in this file. The system gar.conf.mk # provides a reasonable set of defaults, but customizations may be required to # build packages outside of the Blastwave build stack. # # Packaging information SPKG_PACKAGER = John Ellson SPKG_EMAIL = ellson at opencsw.org # Where to put finished packages SPKG_EXPORT = /home/ellson/opencsw/staging/build-$(shell date '+%d.%b.%Y') # Proxy settings #http_proxy = #ftp_proxy = From ellson at opencsw.org Tue May 19 17:44:38 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 11:44:38 -0400 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12CA28.60400@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> Message-ID: <4A12D3E6.6080301@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Ellson wrote: > > >> spec and depend mechanisms. "make build" is ok, and "make package" >> does a lot of work and seems happy, but doesn't leave me with a package in >> the staging area? Could someone look at the attached Makefile and tell >> me what I'm doing wrong? >> > > I don't see a problem with your Makefile ... what does your ~/.garrc file look > like? > This is mine-> > > SPKG_PACKAGER = Mike Watters > SPKG_EMAIL = mwatters at opencsw.org > SPKG_EXPORT = $(HOME)/newpkgs/$(GARNAME) > SPKG_SPOOLDIR = $(HOME)/spool/$(GAROSREL)-$(GARCH) > GARCHIVEDIR = $(HOME)/src > > I converted my .garrc to this style, but still no package :( The output from "make package" ends with: ---------------------- Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_wsstream.i Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_wstreambuf.i Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_wstring.i Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_carray.swg test -z "1" || (cd CCache && gmake install) gmake[2]: Entering directory `/home/ellson/mgar/pkg/swig/trunk/work/ellson-build8s.d/swig-1.3.39/CCache' /opt/csw/bin/ginstall -c -d /home/ellson/build.5.8-sparc/opt/csw/bin /opt/csw/bin/ginstall -c -m 755 ccache-swig /home/ellson/build.5.8-sparc/opt/csw/bin/`echo ccache-swig | sed 's,x,x,'` /opt/csw/bin/ginstall -c -d /home/ellson/build.5.8-sparc/opt/csw/share/man/man1 /opt/csw/bin/ginstall -c -m 644 ./ccache-swig.1 /home/ellson/build.5.8-sparc/opt/csw/share/man/man1/`echo ccache-swig | sed 's,x,x,'`.1 gmake[2]: Leaving directory `/home/ellson/mgar/pkg/swig/trunk/work/ellson-build8s.d/swig-1.3.39/CCache' Installation complete gmake[1]: Leaving directory `/home/ellson/mgar/pkg/swig/trunk/work/ellson-build8s.d/swig-1.3.39' ==> fixconfig: /home/ellson/build.5.8-sparc/opt/csw/bin Stripping /home/ellson/build.5.8-sparc/opt/csw/bin/ccache-swig ... Done. Stripping /home/ellson/build.5.8-sparc/opt/csw/bin/swig ... Done. [strip] complete for swig. [install] complete for swig. [package] complete for swig. ---------------------- but there is nothing in $HOME/newpkgs/swig/ John From bonivart at opencsw.org Tue May 19 18:09:57 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 19 May 2009 18:09:57 +0200 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12D3E6.6080301@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> Message-ID: <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> On Tue, May 19, 2009 at 5:44 PM, John Ellson wrote: > ? ? ? [strip] complete for swig. > ? ? ? [install] complete for swig. > ? ? ? [package] complete for swig. > > ---------------------- > > but there is nothing in $HOME/newpkgs/swig/ Try "svn pg svn:externals" and check that you use GAR v2, otherwise all new features don't work. It should be default if this is a new package but if it's an older you're updating you probably have v1. -- /peter From ellson at opencsw.org Tue May 19 19:06:44 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 13:06:44 -0400 Subject: [csw-maintainers] misleading message from "gmake newpkg-liblasi" Message-ID: <4A12E724.3000208@opencsw.org> The message about the gspec-file is misleading with current mgar automatic mechanisms. $ make newpkg-liblasi A liblasi A liblasi/tags A liblasi/branches A liblasi/trunk A liblasi/trunk/files A liblasi/trunk/Makefile A liblasi/trunk/checksums property 'svn:ignore' set on 'liblasi/trunk' property 'svn:externals' set on 'liblasi/trunk' Your package is set up for editing at liblasi/trunk Please don't forget to add the gspec-file! From ellson at opencsw.org Tue May 19 18:56:18 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 12:56:18 -0400 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> Message-ID: <4A12E4B2.5070202@opencsw.org> Peter Bonivart wrote: > On Tue, May 19, 2009 at 5:44 PM, John Ellson wrote: > >> [strip] complete for swig. >> [install] complete for swig. >> [package] complete for swig. >> >> ---------------------- >> >> but there is nothing in $HOME/newpkgs/swig/ >> > > Try "svn pg svn:externals" and check that you use GAR v2, otherwise > all new features don't work. It should be default if this is a new > package but if it's an older you're updating you probably have v1. > > That was it. I tried following the wiki to change to v2: svn propset svn:externals "gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2" . but that wasn't sufficient. I also had to replace the softlink: rm gar ln -s ../../../gar/v2 gar Perhaps this was just because I didn't commit the propset change yet? Anyway, now I have a package: ~/newpkgs/swig/swig-1.3.39,REV=2009.05.19-SunOS5.8-sparc-UNCOMMITTED.pkg.gz Just waiting on an OK to assume this maintainership.... John From dam at opencsw.org Tue May 19 20:26:48 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 20:26:48 +0200 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12E4B2.5070202@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> <4A12E4B2.5070202@opencsw.org> Message-ID: Hi John, Am 19.05.2009 um 18:56 schrieb John Ellson: > Anyway, now I have a package: > > ~/newpkgs/swig/swig-1.3.39,REV=2009.05.19-SunOS5.8-sparc- > UNCOMMITTED.pkg.gz > > > Just waiting on an OK to assume this maintainership.... Don't wait, just go ahead :-) Now commit your changed and "gmake rebuild" so the UNCOMMITTED goes away. Best regards -- Dago From dam at opencsw.org Tue May 19 20:28:59 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 20:28:59 +0200 Subject: [csw-maintainers] misleading message from "gmake newpkg-liblasi" In-Reply-To: <4A12E724.3000208@opencsw.org> References: <4A12E724.3000208@opencsw.org> Message-ID: <73B3895F-EF7E-47D6-A48E-91C933014B49@opencsw.org> Hi John, Am 19.05.2009 um 19:06 schrieb John Ellson: > The message about the gspec-file is misleading with current mgar > automatic mechanisms. > > $ make newpkg-liblasi > A liblasi > A liblasi/tags > A liblasi/branches > A liblasi/trunk > A liblasi/trunk/files > A liblasi/trunk/Makefile > A liblasi/trunk/checksums > property 'svn:ignore' set on 'liblasi/trunk' > property 'svn:externals' set on 'liblasi/trunk' > > Your package is set up for editing at liblasi/trunk > Please don't forget to add the gspec-file! This has been removed in r4958. Thanks for the report! Best regards -- Dago From ellson at opencsw.org Tue May 19 22:31:30 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 16:31:30 -0400 Subject: [csw-maintainers] Package "swig" in testing Message-ID: <4A131722.7050101@opencsw.org> There is an updated package: "swig: A scripting interface generator for C/C++ code." now available in testing. This is an update to swig-1.3.39, and also a conversion to mgar v2. This is my second package, so a review of my mgar/pkg/swig/ tree in svn would be appreciated. Your feedback please. John From mwatters at opencsw.org Tue May 19 23:40:45 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 19 May 2009 16:40:45 -0500 Subject: [csw-maintainers] Need some Feedback Message-ID: <4A13275D.4040204@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I am splitting up the python package into separate core, runtime, and development packages. I am playing around with removing the pyo and pyc compiled files from the core package. Here is my dilemma: to do this properly, I need to compile and install the pyc and pyo files as a post-install script. the compile flies by, but the install using installf takes FOREVER (the core package contains roughly 3000 files that will need to be installed in the package database.) should I trade the size for the speed of install? does anyone have a better way to update the package db without using installf? - --- split package sizes without pyo and pyc files --- 209k idle 5M python 634k python_devel 1.3M python_rt 123k python_tk - --- split package sizes with pyo and pyc files --- 593k idle 12M python 1015k python_devel 1.3M python_rt 395k python_tk - --- Current package sizes --- 589k idle 393k python_tk 14M python - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoTJ10ACgkQLrhmsXMSLxc3rQCguPzUvdZsNZMOwDGSlLdncUUv LCMAn21gqG8lkEMgbNuoKsW62gNODtVY =rMCd -----END PGP SIGNATURE----- From phil at bolthole.com Tue May 19 23:50:22 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 19 May 2009 14:50:22 -0700 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <4A13275D.4040204@opencsw.org> References: <4A13275D.4040204@opencsw.org> Message-ID: <20090519215021.GA56895@bolthole.com> On Tue, May 19, 2009 at 04:40:45PM -0500, Mike Watters wrote: > - --- split package sizes without pyo and pyc files --- > > 209k idle > 5M python > > - --- split package sizes with pyo and pyc files --- > > 593k idle > 12M python Wow. That's some serious savings. Here's a crazy thought: if you can configure it so that the .pyc stuffs go to /var somewheres, then you might not actually "need" to declare it with installf. Although you MIGHT want to declare a special script, and declare it as type "i", such as i cswpurge in your prototype file. Then at such time as pkgutil and pkg-get support a "purge" call, they can call it after doing pkgrm, if the user says "yes, 'purge' anything to do with this package when I remove it". From mwatters at opencsw.org Wed May 20 02:57:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 19 May 2009 19:57:13 -0500 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <20090519215021.GA56895@bolthole.com> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> Message-ID: <4A135569.3000909@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Phil, Philip Brown wrote: > Although you MIGHT want to declare a special script, and declare it as > type "i", such as > > i cswpurge > > in your prototype file. > > Then at such time as pkgutil and pkg-get support a "purge" call, they can > call it after doing pkgrm, if the user says "yes, 'purge' anything to > do with this package when I remove it". I am not sure I want to do a purge script, but that does give me an Idea. I can do a preremove that does a pkgchk -v on the python package and get the .py files and use that list to remove the pyc pyo files. something like pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}[co] - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoTVWkACgkQLrhmsXMSLxfXFQCgsv46HfdcQXHUgo9qmBavw7FG wQ8AnjXPGY3JX6bHTbgKBBAupz62YDvF =Cg9K -----END PGP SIGNATURE----- From phil at bolthole.com Wed May 20 07:59:23 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 19 May 2009 22:59:23 -0700 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <4A135569.3000909@opencsw.org> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> <4A135569.3000909@opencsw.org> Message-ID: <20090520055923.GA96765@bolthole.com> On Tue, May 19, 2009 at 07:57:13PM -0500, Mike Watters wrote: > I am not sure I want to do a purge script, but that does give me an Idea. > I can do a preremove that does a pkgchk -v on the python package and get the > .py files and use that list to remove the pyc pyo files. > > something like > > pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}[co] pkgchk can be very slow. faster to just hardcode a list of files to check for and remove. (a 'hardcoded' list that could actually be autogenerated from the package's prototype list, at package assembly time) If you're gonna hack, may as well make it an EFFICIENT hack ;-) From phil at bolthole.com Wed May 20 08:01:07 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 19 May 2009 23:01:07 -0700 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <20090520055923.GA96765@bolthole.com> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> <4A135569.3000909@opencsw.org> <20090520055923.GA96765@bolthole.com> Message-ID: <20090520060107.GA99105@bolthole.com> PS: also, if you're gonna do the postinstall hack type stuff... you might also consider making it a "class" based hack, or some other sharable mechanism. You could then provide it as a tool to other python-related package maintainers. From dam at opencsw.org Wed May 20 09:09:04 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 09:09:04 +0200 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <4A131722.7050101@opencsw.org> References: <4A131722.7050101@opencsw.org> Message-ID: Hi John, Am 19.05.2009 um 22:31 schrieb John Ellson: > There is an updated package: > > "swig: A scripting interface generator for C/C++ code." > > now available in testing. > This is an update to swig-1.3.39, and also a conversion to mgar v2. > > This is my second package, so a review of my mgar/pkg/swig/ tree in > svn would be appreciated. > > Your feedback please. Fine package :-) However, there is no need leave in commented code unless you want to re-add it very soon as everything is versioned: One little comment: post-install is now called post-install-modulated, as it should be called for every modulation and DESTDIR is not set properly in post-install. Apart from that: Nice work! Best regards -- Dago From dam at opencsw.org Wed May 20 09:56:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 09:56:25 +0200 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <20090520060107.GA99105@bolthole.com> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> <4A135569.3000909@opencsw.org> <20090520055923.GA96765@bolthole.com> <20090520060107.GA99105@bolthole.com> Message-ID: <46D584D5-4B11-463E-A71B-7BE16D81682B@opencsw.org> Hi, Am 20.05.2009 um 08:01 schrieb Philip Brown: > PS: also, if you're gonna do the postinstall hack type stuff... > you might also consider making it a "class" based hack, or some > other sharable mechanism. > You could then provide it as a tool to other python-related package > maintainers. Yes, that also came to my mind. Make a pythonpyc class which compiles on install and deletes on remove. This can be easily put in an automatic recipe. Best regards -- Dago From schwindt at dfki.uni-kl.de Wed May 20 10:23:36 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Wed, 20 May 2009 10:23:36 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: Your message of "Tue, 19 May 2009 13:46:08 +0200." Message-ID: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> [...] > >> Is this ok with the opencsw standards ? > > > > Sounds reasonable. You can describe additional package use on the > > wiki at > > Sure, wikis are a nice thing. Even nicer is to provide a README.CSW in > the documentation directory where you explain the idiosyncrasies of your > package. Well, my question did not target the documentation - this is documented in wireshark ( and if that gets not read - why should I believe mine is being read ?) I was more concerened about the security implications when installing it setuid. Nicolai. From dam at opencsw.org Wed May 20 10:36:39 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 10:36:39 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> References: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> Message-ID: <8D167374-D789-49F1-BB47-91251B62B315@opencsw.org> Hi Nicolai, Am 20.05.2009 um 10:23 schrieb Nicolai Schwindt: > [...] >>>> Is this ok with the opencsw standards ? >>> >>> Sounds reasonable. You can describe additional package use on the >>> wiki at >> >> Sure, wikis are a nice thing. Even nicer is to provide a README.CSW >> in >> the documentation directory where you explain the idiosyncrasies of >> your >> package. > > Well, my question did not target the documentation - this is > documented in > wireshark > ( and if that gets not read - why should I believe mine is being > read ?) Because some users may want to install suid while other won't. > I was more concerened about the security implications when > installing it > setuid. Yes. That's what I meant: install without SUID and put a note in README.CSW like If you want to allow regular users to snoop promiscous, please issue chmod u+s /opt/csw/bin/wireshark Because it is that important you may also want to echo this during postinstall. Best regards -- Dago From ellson at opencsw.org Wed May 20 15:05:41 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 20 May 2009 09:05:41 -0400 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: References: <4A131722.7050101@opencsw.org> Message-ID: <4A140025.9060501@opencsw.org> Dagobert Michelsen wrote: > Hi John, > > Am 19.05.2009 um 22:31 schrieb John Ellson: >> There is an updated package: >> >> "swig: A scripting interface generator for C/C++ code." >> >> now available in testing. >> This is an update to swig-1.3.39, and also a conversion to mgar v2. >> >> This is my second package, so a review of my mgar/pkg/swig/ tree in >> svn would be appreciated. >> >> Your feedback please. > > Fine package :-) However, there is no need leave in commented > code unless you want to re-add it very soon as everything is > versioned: > > > Mostly done. Also removed unused files/* > > One little comment: post-install is now called post-install-modulated, > as it > should be called for every modulation and DESTDIR is not set properly in > post-install. I need a little help here. I don't really understand what this was doing in the old swig package (before I commented it out), or if it is still needed? #post-install: install-swig-doc # #install-swig-doc: # @mkdir -p $(DESTDIR)$(docdir)/$(GARNAME) # @cp -R $(WORKSRC)/Doc $(DESTDIR)$(docdir)/$(GARNAME) # @cp -R $(WORKSRC)/Examples $(DESTDIR)$(docdir)/$(GARNAME) # @$(MAKECOOKIE) # $(DONADA) If its needed, could someone provide me with the basic code? Related to this, how do I find out what files are installed by a package? There is no "man pkg-get" and "pkg-get --help" is just about useless! John From bwalton at opencsw.org Wed May 20 15:19:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 09:19:39 -0400 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <4A140025.9060501@opencsw.org> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> Message-ID: <1242825141-sup-7168@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > I need a little help here. I don't really understand what this was > doing in the old swig package (before I commented it out), or if it > is still needed? > > #post-install: install-swig-doc > # > #install-swig-doc: > # @mkdir -p $(DESTDIR)$(docdir)/$(GARNAME) > # @cp -R $(WORKSRC)/Doc $(DESTDIR)$(docdir)/$(GARNAME) > # @cp -R $(WORKSRC)/Examples $(DESTDIR)$(docdir)/$(GARNAME) > # @$(MAKECOOKIE) > # $(DONADA) > > If its needed, could someone provide me with the basic code? I can't say if it's needed or not. It may be if the `gmake install` for the package doesn't install docs by default. You have a few possible options for this: 1. If there is an alternate Make target built-in to the swig Makefile (install-all, install-doc, etc), you could alter the GAR variable INSTALL_ARGS to the required target. This would be the easiest option if it's supported by the package. 2. If you can't do 1, you'll likely only need to change the line: post-install: ... to post-install-modulated: ... to have that custom code run once per modulation. > Related to this, how do I find out what files are installed by a > package? The GAR view? After the merge step, you can look in .../work/pkgroot/. Before the merge, but after install, you could look in .../work/install-$(MODULATION)/. If the package is installed, you can use the solaris pkg* commands to query for files owned by the package. HTH -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Wed May 20 15:31:32 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 20 May 2009 15:31:32 +0200 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <1242825141-sup-7168@ntdws12.chass.utoronto.ca> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> <1242825141-sup-7168@ntdws12.chass.utoronto.ca> Message-ID: <4A140634.4010100@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > >> I need a little help here. I don't really understand what this was >> doing in the old swig package (before I commented it out), or if it >> is still needed? >> >> #post-install: install-swig-doc >> # >> #install-swig-doc: >> # @mkdir -p $(DESTDIR)$(docdir)/$(GARNAME) >> # @cp -R $(WORKSRC)/Doc $(DESTDIR)$(docdir)/$(GARNAME) >> # @cp -R $(WORKSRC)/Examples $(DESTDIR)$(docdir)/$(GARNAME) >> # @$(MAKECOOKIE) >> # $(DONADA) >> >> If its needed, could someone provide me with the basic code? > > I can't say if it's needed or not. It may be if the `gmake install` > for the package doesn't install docs by default. Btw. (good that this has been brought up), i noticed that we have quite a few packages that use a post-install-modulated part like the above to install additional docs. Could we maybe get something like EXTRA_INCLUDE_DOCS = in GAR that would do this instead? Sebastian From dam at opencsw.org Wed May 20 15:48:46 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 15:48:46 +0200 Subject: [csw-maintainers] Update to the buildfarm In-Reply-To: <1242824555-sup-2752@ntdws12.chass.utoronto.ca> References: <1242824387-sup-8557@ntdws12.chass.utoronto.ca> <1242824555-sup-2752@ntdws12.chass.utoronto.ca> Message-ID: Hi, Am 20.05.2009 um 15:03 schrieb Ben Walton: > Done, except for build10x. Dago, that box is asking for a password. > Has the NFS mount failed so it can't see my ssh keys? I setup DNS for the farm and this was a NFSv4 issue. I switched back to NFSv3 in the meantime. The other hosts shouldn't be affected as they either use lofs or don't speak NFSv4. Additionally, I distributed an updated global ssh_known_hosts on all build machines so ssh'ing between all machines without password should now be possible. The reason for the change is an update to GAR which allows full parallel builds (if you are curious you can take a look at gar/v2-pbuild). For the moment multitail is missing on the farm to watch the different builds, but that should be resolved when Sebastian releases it to current/. Best regards -- Dago From bwalton at opencsw.org Wed May 20 18:05:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 12:05:39 -0400 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request Message-ID: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> Hi Phil and Peter, Would you consider adding a -C option to the package management utilities to at would limit the compare output to only versions behind the catalog (eg: `pkg-get -U -c | grep -v SAME` is what I use to see what will be updated). It's just a creature comfort kind of option. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 20 18:08:51 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 12:08:51 -0400 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <4A140025.9060501@opencsw.org> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> Message-ID: <1242835669-sup-7223@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > There is no "man pkg-get" and "pkg-get --help" is just about > useless! I forgot to comment on this point earlier: There is a `man pkg-get`, but you need to have /opt/csw/share/man in your $MANPATH first. HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 20 18:44:00 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 09:44:00 -0700 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <1242835669-sup-7223@ntdws12.chass.utoronto.ca> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> <1242835669-sup-7223@ntdws12.chass.utoronto.ca> Message-ID: <20090520164400.GA87831@bolthole.com> On Wed, May 20, 2009 at 12:08:51PM -0400, Ben Walton wrote: > Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > > > There is no "man pkg-get" and "pkg-get --help" is just about > > useless! > > I forgot to comment on this point earlier: There is a `man pkg-get`, > but you need to have /opt/csw/share/man in your $MANPATH first. but besides which, that's the wrong place to look for his actual question anyways. pkg-get doesnt technicall install any packages, or konw anything about location of files. pkg-get calls the system level tool pkgadd. documentaion on that is also semi useless to anser the question of "what files are in a package", though :) for one of OUR packages, easiest thing is to use http://www.opencsw.org/search for other installed packages, grep through /var/sadm/install/contents From phil at bolthole.com Wed May 20 18:46:22 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 09:46:22 -0700 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <8D167374-D789-49F1-BB47-91251B62B315@opencsw.org> References: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> <8D167374-D789-49F1-BB47-91251B62B315@opencsw.org> Message-ID: <20090520164622.GB87831@bolthole.com> On Wed, May 20, 2009 at 10:36:39AM +0200, Dagobert Michelsen wrote: > > If you want to allow regular users to snoop promiscous, please issue > chmod u+s /opt/csw/bin/wireshark > > Because it is that important you may also want to echo this during > postinstall. I would disagree. [and personally think that installing it setuid is insane] additionally, wireshark shouldnt need any kind of postinstall. adding one to do this, is rather overkill. From phil at bolthole.com Wed May 20 19:06:21 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 10:06:21 -0700 Subject: [csw-maintainers] debian's python policy Message-ID: <20090520170620.GD87831@bolthole.com> I just discovered that debian has an explicit, separate "python policy". Thought I'd reference it here. http://www.debian.org/doc/packaging-manuals/python-policy/ As usual.. while we are not bound to follow everything debian does... we would be fools not to learn from their amassed experience on the subject. And coincidently, their policy on .pyc files can be found at http://www.debian.org/doc/packaging-manuals/python-policy/ch-module_packages.html#s-bytecompilation "If a package provides any binary-independent modules (foo.py files), the corresponding bytecompiled modules (foo.pyc files) and optimized modules (foo.pyo files) must not ship in the package. Instead, they should be generated in the package's postinst, and removed in the package's prerm. " Also of interest: " The file /etc/python/debian_config allows configuration how modules should be byte-compiled. The postinst scripts should respect these settings." So, us having a class handle this kind of thing, is strongly indicated as a positive thing for future enhancements such as that. From Joerg.Schilling at fokus.fraunhofer.de Wed May 20 19:26:11 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 20 May 2009 19:26:11 +0200 Subject: [csw-maintainers] debian's python policy In-Reply-To: <20090520170620.GD87831@bolthole.com> References: <20090520170620.GD87831@bolthole.com> Message-ID: <4a143d33.X7zRJnvpgxZ0OJ6p%Joerg.Schilling@fokus.fraunhofer.de> Philip Brown wrote: > I just discovered that debian has an explicit, separate > "python policy". > Thought I'd reference it here. > > http://www.debian.org/doc/packaging-manuals/python-policy/ > > As usual.. while we are not bound to follow everything debian does... we > would be fools not to learn from their amassed experience on the subject. > > And coincidently, their policy on .pyc files can be found at > http://www.debian.org/doc/packaging-manuals/python-policy/ch-module_packages.html#s-bytecompilation > > "If a package provides any binary-independent modules (foo.py files), the > corresponding bytecompiled modules (foo.pyc files) and optimized modules > (foo.pyo files) must not ship in the package. Instead, they should be > generated in the package's postinst, and removed in the package's prerm. " What is the reason for such a decision? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From phil at bolthole.com Wed May 20 19:34:59 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 10:34:59 -0700 Subject: [csw-maintainers] debian's python policy In-Reply-To: <4a143d33.X7zRJnvpgxZ0OJ6p%Joerg.Schilling@fokus.fraunhofer.de> References: <20090520170620.GD87831@bolthole.com> <4a143d33.X7zRJnvpgxZ0OJ6p%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20090520173459.GJ87831@bolthole.com> On Wed, May 20, 2009 at 07:26:11PM +0200, Joerg Schilling wrote: > > "If a package provides any binary-independent modules (foo.py files), the > > corresponding bytecompiled modules (foo.pyc files) and optimized modules > > (foo.pyo files) must not ship in the package. Instead, they should be > > generated in the package's postinst, and removed in the package's prerm. " > > What is the reason for such a decision? because it saves a whoooole lot of space/download time. the python package itself, would be reduced to 5MB from 13, I think Mike said? From ellson at opencsw.org Wed May 20 20:34:10 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 20 May 2009 14:34:10 -0400 Subject: [csw-maintainers] packages "gts" and "swig" ready for public release Message-ID: <4A144D22.6030904@opencsw.org> Philip, I have addressed all coments, and so I believe that the "gts" and "swig" packages are now ready for public release. www.opencsw.org:/home/newpkgs/gts-0.7.6,REV=2009.05.20-SunOS5.8-i386-CSW.pkg.gz www.opencsw.org:/home/newpkgs/gts-0.7.6,REV=2009.05.20-SunOS5.8-sparc-CSW.pkg.gz www.opencsw.org:/home/newpkgs/swig-1.3.39,REV=2009.05.20-SunOS5.8-i386-CSW.pkg.gz www.opencsw.org:/home/newpkgs/swig-1.3.39,REV=2009.05.20-SunOS5.8-sparc-CSW.pkg.gz John From mwatters at opencsw.org Wed May 20 20:44:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 20 May 2009 13:44:44 -0500 Subject: [csw-maintainers] debian's python policy In-Reply-To: <20090520170620.GD87831@bolthole.com> References: <20090520170620.GD87831@bolthole.com> Message-ID: <4A144F9C.4000300@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > I just discovered that debian has an explicit, separate > "python policy". > Thought I'd reference it here. > > http://www.debian.org/doc/packaging-manuals/python-policy/ > > As usual.. while we are not bound to follow everything debian does... we > would be fools not to learn from their amassed experience on the subject. > > And coincidently, their policy on .pyc files can be found at > http://www.debian.org/doc/packaging-manuals/python-policy/ch-module_packages.html#s-bytecompilation > > "If a package provides any binary-independent modules (foo.py files), the > corresponding bytecompiled modules (foo.pyc files) and optimized modules > (foo.pyo files) must not ship in the package. Instead, they should be > generated in the package's postinst, and removed in the package's prerm. " This is exactly how it is built now. > > > Also of interest: > > " The file /etc/python/debian_config allows configuration how modules should > be byte-compiled. The postinst scripts should respect these settings." > > So, us having a class handle this kind of thing, is strongly indicated as a > positive thing for future enhancements such as that. I will work on the class scripts for this and release it that way. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoUT5wACgkQLrhmsXMSLxf9PACfaL6aWlyMTYEkAMyps/C7ThGQ LoAAnjuhun1nkUsYpCQ7QScSUJJulq26 =jn88 -----END PGP SIGNATURE----- From skayser at opencsw.org Wed May 20 21:44:28 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 20 May 2009 21:44:28 +0200 Subject: [csw-maintainers] Another GAR RFE, deprecate gar/v1 (was: Re: Offer to maintain swig) In-Reply-To: <4A12E4B2.5070202@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> <4A12E4B2.5070202@opencsw.org> Message-ID: <4A145D9C.7080205@opencsw.org> John Ellson wrote: > Peter Bonivart wrote: >> On Tue, May 19, 2009 at 5:44 PM, John Ellson wrote: >> >>> [strip] complete for swig. >>> [install] complete for swig. >>> [package] complete for swig. >>> >>> ---------------------- >>> >>> but there is nothing in $HOME/newpkgs/swig/ >>> >> Try "svn pg svn:externals" and check that you use GAR v2, otherwise >> all new features don't work. It should be default if this is a new >> package but if it's an older you're updating you probably have v1. >> >> > That was it. As we try to trim our documentation towards mGAR v2 (and forget about the whole different iterations of GAR eventually), could we maybe add a deprecation warning to the gar/v1 Makefile. Something along the lines of "Unless you know what you are doing, please consider switching to the currently maintained version of GAR, by doing ....". Maybe not as harsh, but gently pointing the way :) Sebastian From bonivart at opencsw.org Wed May 20 22:46:23 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 20 May 2009 22:46:23 +0200 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request In-Reply-To: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> References: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> Message-ID: <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> On Wed, May 20, 2009 at 6:05 PM, Ben Walton wrote: > Would you consider adding a -C option to the package management > utilities to at would limit the compare output to only versions behind > the catalog (eg: `pkg-get -U -c | grep -v SAME` is what I use to see > what will be updated). Sure, I can add it to 1.6 which is almost done. I never implemented showing not installed packages but I still get annoyed by using grep -v SAME so it's a good idea. -- /peter From mwatters at opencsw.org Thu May 21 02:46:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 20 May 2009 19:46:30 -0500 Subject: [csw-maintainers] Packages that require either package A or package B Message-ID: <4A14A466.5010602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 There are some packages that require either pkgA or pkgB to be functional. mediawiki for example. it requires Either MySQL OR Postgres. How can we handle this in a package that allows a seamless install. I figure a checkinstall script to determine if one or the other exists and bail gracefully "like a normal requirement not met" if missing both. thoughts? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoUpGUACgkQLrhmsXMSLxcJ2QCfWF6OZg1/p10c5tuir8oNwaFY P+cAn3Okd64mgz+H/7RKaiaFbNnwyvkk =pJXK -----END PGP SIGNATURE----- From phil at bolthole.com Thu May 21 02:53:27 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 17:53:27 -0700 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <4A14A466.5010602@opencsw.org> References: <4A14A466.5010602@opencsw.org> Message-ID: <20090521005327.GA15461@bolthole.com> On Wed, May 20, 2009 at 07:46:30PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > There are some packages that require either pkgA or pkgB to be functional. > > mediawiki for example. it requires Either MySQL OR Postgres. > > How can we handle this in a package that allows a seamless install. > > I figure a checkinstall script to determine if one or the other exists and bail > gracefully "like a normal requirement not met" if missing both. and what if you encounter both installed? In this sort of situation, i think you have a basic choice of either a) just pick one that you the maintainer think is best b) do not auto-install, but provide utility scripts to auto-setup for either of them, and a README.CSW file that indicates, "run the script you prefer, to finish setup". (and of course, some sort of auto-detection for, "this is an UPGRADE, so go with whatever was previously chosen") From bwalton at opencsw.org Thu May 21 03:16:34 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 21:16:34 -0400 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <20090521005327.GA15461@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> Message-ID: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 20 20:53:27 -0400 2009: > a) just pick one that you the maintainer think is best > > b) do not auto-install, but provide utility scripts to auto-setup for > either of them, and a README.CSW file that indicates, > "run the script you prefer, to finish setup". > > (and of course, some sort of auto-detection for, > "this is an UPGRADE, so go with whatever was previously chosen") c) make pkgutil/pkg-get smart enough to handle debian-style dependencies above and beyond what pkgadd can actually handle. Much more work, but a better experience in the end. I suspect that this evolution will need to happen anyway in the face of IPS, which won't (doesn't? I haven't played with it) limit itself to the current rigid dependency system. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 21 14:24:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 21 May 2009 08:24:06 -0400 Subject: [csw-maintainers] Howto package applications that use cmake? Message-ID: <4A1547E6.7050704@opencsw.org> I'd like to package libLASi, mostly because graphviz can use it: http://www.unifont.org/lasi/ I have a couple of problems because it uses cmake. 1. CSWcmake isn't installed on the build hosts 2. How to convince GAR to use cmake? Can anyone point me to another package that uses it? John From mwatters at opencsw.org Thu May 21 15:55:32 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 08:55:32 -0500 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A1547E6.7050704@opencsw.org> References: <4A1547E6.7050704@opencsw.org> Message-ID: <4A155D54.20806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > I'd like to package libLASi, mostly because graphviz can use it: > > http://www.unifont.org/lasi/ > > I have a couple of problems because it uses cmake. > > 1. CSWcmake isn't installed on the build hosts Send a request to buildfarm at lists.opencsw.org requesting the install of cmake on the build farm. > > 2. How to convince GAR to use cmake? Can anyone point me to another > package that uses it? GAR should use whatever configure tells it to use for the actual compile. it will only use gmake for the commands to build the installation environment. > > John > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVXVQACgkQLrhmsXMSLxfEZQCff9GfGeq7WOxP0XyMyoV9Phnd i1EAnR0XUjgW0YfsRws9RXAM97vs0ANz =tGp+ -----END PGP SIGNATURE----- From ellson at opencsw.org Thu May 21 16:10:01 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 21 May 2009 10:10:01 -0400 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A155D54.20806@opencsw.org> References: <4A1547E6.7050704@opencsw.org> <4A155D54.20806@opencsw.org> Message-ID: <4A1560B9.3070201@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Ellson wrote: > >> I'd like to package libLASi, mostly because graphviz can use it: >> >> http://www.unifont.org/lasi/ >> >> I have a couple of problems because it uses cmake. >> >> 1. CSWcmake isn't installed on the build hosts >> > Send a request to buildfarm at lists.opencsw.org requesting the install of cmake > on the build farm. > Will do. > >> 2. How to convince GAR to use cmake? Can anyone point me to another >> package that uses it? >> > > GAR should use whatever configure tells it to use for the actual compile. > it will only use gmake for the commands to build the installation environment. > Thats the problem. There is no "configure". I guess I could patch one in, but it would be better if I could just run "cmake" per the libLASi build instructions: http://www.unifont.org/lasi/#cmake John From mwatters at opencsw.org Thu May 21 16:15:34 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 09:15:34 -0500 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A1560B9.3070201@opencsw.org> References: <4A1547E6.7050704@opencsw.org> <4A155D54.20806@opencsw.org> <4A1560B9.3070201@opencsw.org> Message-ID: <4A156206.3050807@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > > Thats the problem. There is no "configure". I guess I could patch one > in, but it > would be better if I could just run "cmake" per the libLASi build > instructions: > > http://www.unifont.org/lasi/#cmake OK, Sorry I miss read the post. thanks for the clarification. you need to use a custom configure script. if you set the following variable CONFIGURE_SCRIPTS = xxxx then create your configure script configure-xxxx: echo "my config script" ... ... ... @$(MAKECOOKIE) spamassassin/trunk/Makefile has some good examples - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVYgYACgkQLrhmsXMSLxcXRACgoNr73LX/9dYWas2N4FLasJD5 G5MAoM+UGySoKVat2YbBwm3Rphs8Hh7H =f/SQ -----END PGP SIGNATURE----- From phil at bolthole.com Thu May 21 18:08:33 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 09:08:33 -0700 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> Message-ID: <20090521160833.GA89074@bolthole.com> On Wed, May 20, 2009 at 09:16:34PM -0400, Ben Walton wrote: > I suspect that this > evolution will need to happen anyway in the face of IPS, which won't > limit itself to the current rigid > dependency system. funnily enough, there was some semi-related discussion on the ips dev list. And they're running into problems, because they decided to be "more flexible". So I dont think that just copying what IPS does, is a good general strategic move for us. Sometimes, offering more choices, actualy ends up being the wrong thing to do. particularly in a tool designed to simplify and streamline installs. The purpose (for pkg-get, at least), is NOT to offer *everything that everyone could possibly want to do*. It is to simplify installation, for the most common straightforward uses. If people want to do really really fancy non-standard stuff... they should be doing it themselves. Similar to the whole concept of offering binary packages in the first place. We try to offer a set of binary packages, choosing compile flags for the most common good. But if people want to get that 1% extra performance by using one-cpu-specific flags.... they should compile it themselves. That being said... there is a more general issue here about "what do we do when a package wants 'a database' installed?" I shall offer a separate email on that topic, so more people will pay attention :-) From phil at bolthole.com Thu May 21 18:13:07 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 09:13:07 -0700 Subject: [csw-maintainers] "package needs a database" general issue Message-ID: <20090521161307.GB89074@bolthole.com> Hi folks, There is an issue that comes up occasionally, where we want to offer an 'easy install' option for people for a particular package, but that package requires a database to run. Then there is a matter of; WHICH database should we use, for fully automated install?!! usually it comes down to a choice of "postgres or mysql?" I would like to propose that we need to pick an official default database to use. Furthermore, I propose that that database be mysql. It's the simplest to administer, it has the biggest mindshare, and there are lots of apps that will ONLY support mysql, sadly. HOWEVER, I always like to support end-user choice whenever possible. So, in addition to a default assumption of mysql... I would also like to propose a setting in csw.conf. Something like default_dbtype where values can be mysql, postgres, or even oracle. If that value is set, then a package can presume that that database is already installed and running. In those cases though, we also need to agree on some kind of common mechanism to insert values into said database. Comments? From skayser at opencsw.org Thu May 21 18:35:28 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 21 May 2009 18:35:28 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? Message-ID: <4A1582D0.8040709@opencsw.org> Hi, i packaged mailx that was formerly known as nail (CSWnail). Now how would we go about obsoleting CSWnail? Just drop it from the catalog, let CSWnail become an empty package that pulls in CSWmailx and somehow deletes CSWnail, or is there another preferred (proven) way of doing things like this? Sebastian From dam at opencsw.org Thu May 21 18:50:17 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 18:50:17 +0200 Subject: [csw-maintainers] Virtualizing Solaris 8/9 x86 Message-ID: <0DC8DB57-B510-469E-B794-75FCA361AAEF@opencsw.org> Hi, good news on virtualization of Solaris 8 and 9 x86: the new VMware vSphere 4.0 (formerly known as ESX Server) is going to support all versions of Solaris 8 and 9 x86: > Solaris 9, Update 1, Update 2, Update 3, Update 4, Update 5, Update 6, > Update 7, and Update 8, 32-bit. (experimental) See ?Solaris 9 > Operating > System x86 Platform Edition? on page 406. > Solaris 8, 06/00, 10/00, 01/01, 04/01, 07/01, 10/01, and 02/02, 32-bit > (experimental). See ?Solaris 8 Operating System x86 Platform > Edition? on > page 409. This will allow us to make good use of faster hardware and testing servers on demand :-) Best regards -- Dago From dam at opencsw.org Thu May 21 18:58:27 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 18:58:27 +0200 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <20090521160833.GA89074@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> Message-ID: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> Hi, Am 21.05.2009 um 18:08 schrieb Philip Brown: > On Wed, May 20, 2009 at 09:16:34PM -0400, Ben Walton wrote: >> I suspect that this >> evolution will need to happen anyway in the face of IPS, which won't >> limit itself to the current rigid >> dependency system. > > funnily enough, there was some semi-related discussion on the ips > dev list. > And they're running into problems, because they decided to be "more > flexible". So I dont think that just copying what IPS does, is a good > general strategic move for us. Additionally, adding more complexity to System V packaging will bind valuable resources and would make packaging quite more complex. Apart from that the current mechanism has served us well over the past 10+ years and will fade away as the new packaging system is already there (IPS). I would favor to live with the faults and handle dependencies as we have always handled them. In addition, I propose to carefully start working on real IPS packaging instead of putting extra work/complexity in System V packaging. A real IPS backend for GAR with all bells and whistels for dual-personality building of packages is surely a Good Thing. Best regards -- Dago From dam at opencsw.org Thu May 21 18:59:54 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 18:59:54 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A1582D0.8040709@opencsw.org> References: <4A1582D0.8040709@opencsw.org> Message-ID: Hi Sebastian, Am 21.05.2009 um 18:35 schrieb Sebastian Kayser: > i packaged mailx that was formerly known as nail (CSWnail). Now how > would we go about obsoleting CSWnail? Just drop it from the > catalog, let > CSWnail become an empty package that pulls in CSWmailx and somehow > deletes CSWnail, or is there another preferred (proven) way of doing > things like this? I vaguely recall pkg-get has some magic included that does this kind of thing on making new packages incompatible ("I") with prior ones. Phil? Best regards -- Dago From phil at bolthole.com Thu May 21 19:03:34 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 10:03:34 -0700 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A1582D0.8040709@opencsw.org> References: <4A1582D0.8040709@opencsw.org> Message-ID: <20090521170334.GH89074@bolthole.com> On Thu, May 21, 2009 at 06:35:28PM +0200, Sebastian Kayser wrote: > Hi, > > i packaged mailx that was formerly known as nail (CSWnail). Now how > would we go about obsoleting CSWnail? Just drop it from the catalog, let > CSWnail become an empty package that pulls in CSWmailx and somehow > deletes CSWnail, or is there another preferred (proven) way of doing > things like this? ugh. is the binary really called "mailx"? that would be very lame. I think it might actually be nice to keep it as called "nail" :-} From dam at opencsw.org Thu May 21 19:05:05 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 19:05:05 +0200 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A1560B9.3070201@opencsw.org> References: <4A1547E6.7050704@opencsw.org> <4A155D54.20806@opencsw.org> <4A1560B9.3070201@opencsw.org> Message-ID: <88803155-5C85-4DA4-878D-50BB942274C0@opencsw.org> Hi, Am 21.05.2009 um 16:10 schrieb John Ellson: >> Send a request to buildfarm at lists.opencsw.org requesting the >> install of cmake >> on the build farm. > > Will do. cmake is now being installed on all farm servers. Best regards -- Dago From phil at bolthole.com Thu May 21 19:09:44 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 10:09:44 -0700 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: References: <4A1582D0.8040709@opencsw.org> Message-ID: <20090521170944.GA597@bolthole.com> On Thu, May 21, 2009 at 06:59:54PM +0200, Dagobert Michelsen wrote: > Hi Sebastian, > > Am 21.05.2009 um 18:35 schrieb Sebastian Kayser: >> i packaged mailx that was formerly known as nail (CSWnail). Now how >> would we go about obsoleting CSWnail? Just drop it from the catalog, >> let >> CSWnail become an empty package that pulls in CSWmailx and somehow >> deletes CSWnail, or is there another preferred (proven) way of doing >> things like this? > > I vaguely recall pkg-get has some magic included that does this kind > of thing on making new packages incompatible ("I") with prior ones. > Phil? yes. the only magic is that you have to get users to upgrade from "nail" to the new package name. You can either do it in two stages, doin the "empty package that pulls in mailx", then in a few,months, switching it and making mailx conflict. OR, you can just start out with the conflict, and send out an announcement that users of 'nail' need to switch. From Joerg.Schilling at fokus.fraunhofer.de Thu May 21 19:20:36 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 May 2009 19:20:36 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A1582D0.8040709@opencsw.org> References: <4A1582D0.8040709@opencsw.org> Message-ID: <4a158d64.jN+yHRJlXTsKF9bv%Joerg.Schilling@fokus.fraunhofer.de> Sebastian Kayser wrote: > i packaged mailx that was formerly known as nail (CSWnail). Now how > would we go about obsoleting CSWnail? Just drop it from the catalog, let > CSWnail become an empty package that pulls in CSWmailx and somehow > deletes CSWnail, or is there another preferred (proven) way of doing > things like this? Don't omit the name "nail" from the bin directory. It helps to select this right binary. BTW: Was there a bug fix for the old and nasty problem that causes to let "nail" to hang with 50% probability via imaps when the Subject or the mail body is empty? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From bwalton at opencsw.org Thu May 21 19:29:16 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 13:29:16 -0400 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> Message-ID: <1242926828-sup-5777@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu May 21 12:58:27 -0400 2009: > Additionally, adding more complexity to System V packaging will bind > valuable resources and would make packaging quite more complex. > Apart from that the current mechanism has served us well over the > past 10+ years and will fade away as the new packaging system is > already there (IPS). I would favor to live with the faults and Yes, I'm not suggesting it was a great option, just _an_ option. > handle dependencies as we have always handled them. In addition, > I propose to carefully start working on real IPS packaging > instead of putting extra work/complexity in System V packaging. > A real IPS backend for GAR with all bells and whistels for > dual-personality building of packages is surely a Good Thing. Yes, I agree. Will IPS make pkgutil/pkg-get redundant? Eg, will it facilitate simply creating a 'repo' and adding it to the IPS conf (similar to apt/yum)? If so, all the smarts could be punted to the system tools instead, which would be nice. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From phil at bolthole.com Thu May 21 19:36:14 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 10:36:14 -0700 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242926828-sup-5777@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> Message-ID: <20090521173614.GC597@bolthole.com> On Thu, May 21, 2009 at 01:29:16PM -0400, Ben Walton wrote: > Yes, I agree. Will IPS make pkgutil/pkg-get redundant? Eg, will it > facilitate simply creating a 'repo' and adding it to the IPS conf > (similar to apt/yum)? If so, all the smarts could be punted to the > system tools instead, which would be nice. personally, i think they screwed up with the fundamental design of IPS. [in multiple areas] My opinion, is that if we put in enough effort into improving SVR4 for the things that it really lacks (by our use of extending it through class action scripts, for example), that one day, IPS will be the NIS+, compared to pkgadd's NIS. (ie: NIS is still here, even after sun "killed" it, and NIS+ is not :-) Mind you, there's ldap on top of everything :) but still, the bottom line being that nis+ was just too complex for most peoples' needs. From ellson at opencsw.org Thu May 21 18:27:04 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 21 May 2009 12:27:04 -0400 Subject: [csw-maintainers] "package needs a database" general issue In-Reply-To: <20090521161307.GB89074@bolthole.com> References: <20090521161307.GB89074@bolthole.com> Message-ID: <4A1580D8.1040208@opencsw.org> I think many Fedora packages are using SQLite as their default database. No problems with starting daemons or setting up separate userids and passwords. John Philip Brown wrote: > Hi folks, > > There is an issue that comes up occasionally, where we want to offer an > 'easy install' option for people for a particular package, but that package > requires a database to run. > > Then there is a matter of; WHICH database should we use, for fully > automated install?!! > > usually it comes down to a choice of "postgres or mysql?" > > I would like to propose that we need to pick an official default database > to use. > > > Furthermore, I propose that that database be mysql. It's the simplest to > administer, it has the biggest mindshare, and there are lots of apps that > will ONLY support mysql, sadly. > > HOWEVER, I always like to support end-user choice whenever possible. > So, in addition to a default assumption of mysql... I would also like to > propose a setting in csw.conf. > Something like > > default_dbtype > > where values can be mysql, postgres, or even oracle. > If that value is set, then a package can presume that that database is > already installed and running. > In those cases though, we also need to agree on some kind of common > mechanism to insert values into said database. > > Comments? > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > > From bwalton at opencsw.org Thu May 21 19:47:37 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 13:47:37 -0400 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <20090521173614.GC597@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> Message-ID: <1242927821-sup-8734@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 13:36:14 -0400 2009: > personally, i think they screwed up with the fundamental design of > IPS. [in multiple areas] My opinion, is that if we put in enough > effort into improving SVR4 for the things that it really lacks (by > our use of extending it through class action scripts, for example), > that one day, IPS will be the NIS+, compared to pkgadd's NIS. The debian system works well (although I've never packaged anything using it). Good things (imo) are: 1. Depending on or being incompatible with specific versions (or ranges) of package X 2. Depends on mail-server/Provides mail-server 3. Depends on mysql|postgres Where does IPS fit into the landscape in comparison to debian and svr4? (Or tell me to go RTFM!) -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu May 21 19:56:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 12:56:35 -0500 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242927821-sup-8734@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> Message-ID: <4A1595D3.2060404@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Philip Brown's message of Thu May 21 13:36:14 -0400 2009: > >> personally, i think they screwed up with the fundamental design of >> IPS. [in multiple areas] My opinion, is that if we put in enough >> effort into improving SVR4 for the things that it really lacks (by >> our use of extending it through class action scripts, for example), >> that one day, IPS will be the NIS+, compared to pkgadd's NIS. > > The debian system works well (although I've never packaged anything > using it). Good things (imo) are: > > 1. Depending on or being incompatible with specific versions (or > ranges) of package X Sun kinda supports this now in the depend files. man depend > 2. Depends on mail-server/Provides mail-server > 3. Depends on mysql|postgres This is one big thing the current Sun packages are really lacking > > Where does IPS fit into the landscape in comparison to debian and > svr4? (Or tell me to go RTFM!) > > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVldIACgkQLrhmsXMSLxd/VACglxZQErgPXp8nRpofd/qF9o1r RaAAoNHgPx0OfAmqEMnPd41jOh+9RYVL =hq2M -----END PGP SIGNATURE----- From phil at bolthole.com Thu May 21 20:02:57 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 11:02:57 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1595D3.2060404@opencsw.org> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> Message-ID: <20090521180256.GD597@bolthole.com> On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: > > 2. Depends on mail-server/Provides mail-server > > 3. Depends on mysql|postgres > This is one big thing the current Sun packages are really lacking > But how is that even processable? if you apt-get [somepackage], and that package "depends on mysql|postgres", and neither is installed... What happens? From mwatters at opencsw.org Thu May 21 20:05:21 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 13:05:21 -0500 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521180256.GD597@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> Message-ID: <4A1597E1.8060602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: >>> 2. Depends on mail-server/Provides mail-server >>> 3. Depends on mysql|postgres >> This is one big thing the current Sun packages are really lacking >> > > But how is that even processable? > > if you apt-get [somepackage], and that package > "depends on mysql|postgres", and neither is installed... > What happens? > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers If Neither is installed it errors out normally. if both are install it just installs. apt and yum do almost ZERO configuration for users. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVl+EACgkQLrhmsXMSLxcYrACdEZWyJ4ZjcBGWGWX54joHgZlG GCwAnAk3F2XK5WPSy051AQc4S6nai8z2 =9KCY -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 21 20:13:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 14:13:08 -0400 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521180256.GD597@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> Message-ID: <1242929358-sup-207@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 14:02:57 -0400 2009: > if you apt-get [somepackage], and that package > "depends on mysql|postgres", and neither is installed... > What happens? I've never actually interacted with it through command-line apt, but always through aptitude. When I select a package that requires A|B, it will mark that package as having unmet dependencies. I can then resolve the dependency by selecting one of the available options. This satisfies the condition and away we go. Same goes for the general Provides/Requires. Broken until the selected package set meets the criteria. I suspect that command-line apt would simply quit with an error (likely listing the packages that would fulfill the requirements). If you ran apt-get again with firstpkg deppkg, it would run to completion. They also provide a 'suggests' feature that can interact with this (although it can be used independently also to provide additional, optional functionality). A 'suggested' package may be the preferred db backend, for example. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu May 21 20:18:25 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 11:18:25 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1597E1.8060602@opencsw.org> References: <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> Message-ID: <20090521181824.GE597@bolthole.com> On Thu, May 21, 2009 at 01:05:21PM -0500, Mike Watters wrote: > Philip Brown wrote: > > But how is that even processable? > > > > if you apt-get [somepackage], and that package > > "depends on mysql|postgres", and neither is installed... > > What happens? > > If Neither is installed it errors out normally. > if both are install it just installs. pfft. Sounds like there isnt much reason to hack that kind of "multiple|optional dependancy' into the pkg delivery system then. We could accomplish basically the same thing by other means. To take the database dependancy specifically; I'm thinking some kind of generic database wrapper tool package. Kinda like CPAN::DB, but for shellscripts :-) Then packages could depend on CSWdbwrapper, and use the appropriate tools in there as needed. With preference as to WHICH database to use, set in csw.conf locally, as I suggested in my other email today. Not only does it solve the "problem" in that area, but it would solve it BETTER than how debian 'solves' it. How does that sound? From bwalton at opencsw.org Thu May 21 20:27:54 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 14:27:54 -0400 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521181824.GE597@bolthole.com> References: <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> Message-ID: <1242930111-sup-5665@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 14:18:25 -0400 2009: > Sounds like there isnt much reason to hack that kind of > "multiple|optional dependancy' into the pkg delivery system then. > We could accomplish basically the same thing by other means. No, it can be handled above the actual package level by the tools manipulating them (which was the gist of option c)...I do agree that it's likely wasted effort to bolt it on at this point, since Sun will be abandoning svr4 at some point in the not too distant future... > To take the database dependancy specifically; I'm thinking some kind > of generic database wrapper tool package. Kinda like CPAN::DB, but > for shellscripts :-) Then packages could depend on CSWdbwrapper, and > use the appropriate tools in there as needed. To be fair, perl is likely a better tool for this anyway, so the DB module is a good choice. :) > Not only does it solve the "problem" in that area, but it would solve it > BETTER than how debian 'solves' it. ...except for the hole that is: none of those packages depend on the actual db package, so removing it wouldn't generate any warnings (unless some other package marked it explicitly). Would the CSWdbwrapper depend on both? With debian, removing the package providing the 'generic' rdbms would alert you to the dependency breakage. > How does that sound? Given the workings of svr4 packages, about as good as it'll get, most likely. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu May 21 22:02:36 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 22:02:36 +0200 Subject: [csw-maintainers] New Solaris package archive? Message-ID: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> Hi, I noticed today that there is another packagingproject I have never seen before: http://www.sunfreepacks.com They have around 70 packages for Solaris 10 x86 / OpenSolaris and a few for Solaris 10 Sparc. They don't seem to have a catalog or packaging tool. They are building on top of Solaris / SunFreeware packages. Does anyone have experience with it? I am currently mirroring it to /export/mirror/sunfreepacks, just in case someone wants to take a look. Best regards -- Dago From phil at bolthole.com Thu May 21 22:16:30 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 13:16:30 -0700 Subject: [csw-maintainers] New Solaris package archive? In-Reply-To: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> References: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> Message-ID: <20090521201630.GB46162@bolthole.com> On Thu, May 21, 2009 at 10:02:36PM +0200, Dagobert Michelsen wrote: > Hi, > > I noticed today that there is another packagingproject I have > never seen before: > > http://www.sunfreepacks.com kinda odd. no mention of why they are bothering to do this. and it mostly seems to be geared towards opensolaris, and NOT solaris 10. For example, their ASdjvulibre package, is marked as ONLY working on opensolaris. From a.cervellin at acm.org Thu May 21 22:22:14 2009 From: a.cervellin at acm.org (Alessio) Date: Thu, 21 May 2009 22:22:14 +0200 Subject: [csw-maintainers] New Solaris package archive? In-Reply-To: <20090521201630.GB46162@bolthole.com> References: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> <20090521201630.GB46162@bolthole.com> Message-ID: <4A15B7F6.5070904@acm.org> Philip Brown wrote: > On Thu, May 21, 2009 at 10:02:36PM +0200, Dagobert Michelsen wrote: > >> Hi, >> >> I noticed today that there is another packagingproject I have >> never seen before: >> >> http://www.sunfreepacks.com >> > > kinda odd. no mention of why they are bothering to do this. my guess: probably for they own use + advertising... it seems they are sun principal partner and they do web hosting as business From phil at bolthole.com Thu May 21 22:23:49 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 13:23:49 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <1242930111-sup-5665@ntdws12.chass.utoronto.ca> References: <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> <1242930111-sup-5665@ntdws12.chass.utoronto.ca> Message-ID: <20090521202349.GE46162@bolthole.com> On Thu, May 21, 2009 at 02:27:54PM -0400, Ben Walton wrote: > > Not only does it solve the "problem" in that area, but it would solve it > > BETTER than how debian 'solves' it. > > ...except for the hole that is: none of those packages depend on the > actual db package, so removing it wouldn't generate any warnings > (unless some other package marked it explicitly). Would the > CSWdbwrapper depend on both? With debian, removing the package > providing the 'generic' rdbms would alert you to the dependency > breakage. > I doubt that debian does it any better. For example, lets say that you were installing... [foo], which depended on "mysql|postgres", and you had BOTH installed. You chose to install it using mysql. Then later, you choose to remove mysql, but postgres is still there. does apt-get/aptitude somehow register that you chose mysql specifically for package [foo], and complains about the removal? I am scheptical about this. From phil at bolthole.com Thu May 21 22:30:09 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 13:30:09 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521202349.GE46162@bolthole.com> References: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> <1242930111-sup-5665@ntdws12.chass.utoronto.ca> <20090521202349.GE46162@bolthole.com> Message-ID: <20090521203009.GF46162@bolthole.com> On Thu, May 21, 2009 at 01:23:49PM -0700, Philip Brown wrote: > I doubt that debian does it any better. > For example, lets say that you were installing... [foo], which depended on > "mysql|postgres", and you had BOTH installed. > > You chose to install it using mysql. > Then later, you choose to remove mysql, but postgres is still there. > > does apt-get/aptitude somehow register that you chose mysql specifically > for package [foo], > and complains about the removal? > I am scheptical about this. btw; we could do this sort of check for ourselves, via the suggested "dbwrapper" set of tools. If we agree that packages that use the dbwrapper tools, also have to 'register' which database they end up using, then that database can check the registration, and warn about programs using it, at pkgrm time. From dam at opencsw.org Thu May 21 22:55:49 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 22:55:49 +0200 Subject: [csw-maintainers] Commit #5000 today! Message-ID: Hi, today we have to celebrate a jubilee: Subversion commit #5000 was done today from Mike Watters on an updated package for Drupal with commit #4000 just about two month ago. Thanks to all those helping hands who made this possible! Commits are not a virtue on their own, but they reflect the activity on the project. At the current rate over the past month we have well over 300 writes per month with over 1000 updated files: Starting from March the repository reads exploded from a few hundred to several thousands. I guess the number of people rebuilding the OpenCSW packages for fun or tweaks is increasing quickly. So, please everybody grab a piece, I already had mine :-) _____ _..--'''@ @'''--.._ .' @_/-//-\/>/>'/ @ '. ( @ /_ References: <4A1582D0.8040709@opencsw.org> <4a158d64.jN+yHRJlXTsKF9bv%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <4A15C4C4.5000807@opencsw.org> Hi Joerg, Joerg Schilling wrote: > Sebastian Kayser wrote: > >> i packaged mailx that was formerly known as nail (CSWnail). Now how >> would we go about obsoleting CSWnail? Just drop it from the catalog, let >> CSWnail become an empty package that pulls in CSWmailx and somehow >> deletes CSWnail, or is there another preferred (proven) way of doing >> things like this? > > Don't omit the name "nail" from the bin directory. It helps to select this > right binary. Phil also noted it, so i guess i could add a nail symlink (pointing to mailx) to the package. What exactly do you mean by "select this right binary"? > BTW: Was there a bug fix for the old and nasty problem that causes to let "nail" > to hang with 50% probability via imaps when the Subject or the mail body is > empty? I mostly use mailx non-interactive (for scripting) and am neither aware of the issue nor of a fix. However, i put the new package in testing, so if you have been experiencing the issue, feel free to give the new package a spin. http://mirror.opencsw.org/testing.html Sebastian From skayser at opencsw.org Thu May 21 23:26:34 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 21 May 2009 23:26:34 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <20090521170334.GH89074@bolthole.com> References: <4A1582D0.8040709@opencsw.org> <20090521170334.GH89074@bolthole.com> Message-ID: <4A15C70A.1010808@opencsw.org> Philip Brown wrote: > On Thu, May 21, 2009 at 06:35:28PM +0200, Sebastian Kayser wrote: >> Hi, >> >> i packaged mailx that was formerly known as nail (CSWnail). Now how >> would we go about obsoleting CSWnail? Just drop it from the catalog, let >> CSWnail become an empty package that pulls in CSWmailx and somehow >> deletes CSWnail, or is there another preferred (proven) way of doing >> things like this? > > > ugh. is the binary really called "mailx"? that would be very lame. > I think it might actually be nice to keep it as called "nail" :-} Yep, "nail has been integrated into the Heirloom project as mailx" [1] and the binary is called mailx. I could add a symlink called nail that points to mailx. Thoughts? Sebastian [1] http://heirloom.sourceforge.net/mailx.html From phil at bolthole.com Thu May 21 23:31:18 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 14:31:18 -0700 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A15C70A.1010808@opencsw.org> References: <4A1582D0.8040709@opencsw.org> <20090521170334.GH89074@bolthole.com> <4A15C70A.1010808@opencsw.org> Message-ID: <20090521213118.GK46162@bolthole.com> On Thu, May 21, 2009 at 11:26:34PM +0200, Sebastian Kayser wrote: > Yep, "nail has been integrated into the Heirloom project as mailx" [1] > and the binary is called mailx. I could add a symlink called nail that > points to mailx. Thoughts? gets my vote From bwalton at opencsw.org Fri May 22 02:17:02 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 20:17:02 -0400 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521203009.GF46162@bolthole.com> References: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> <1242930111-sup-5665@ntdws12.chass.utoronto.ca> <20090521202349.GE46162@bolthole.com> <20090521203009.GF46162@bolthole.com> Message-ID: <1242951214-sup-5073@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 16:30:09 -0400 2009: > On Thu, May 21, 2009 at 01:23:49PM -0700, Philip Brown wrote: > > I doubt that debian does it any better. > > For example, lets say that you were installing... [foo], which depended on > > "mysql|postgres", and you had BOTH installed. > > > > You chose to install it using mysql. > > Then later, you choose to remove mysql, but postgres is still there. > > > > does apt-get/aptitude somehow register that you chose mysql specifically > > for package [foo], > > and complains about the removal? > > I am scheptical about this. > > btw; we could do this sort of check for ourselves, via the suggested > "dbwrapper" set of tools. > If we agree that packages that use the dbwrapper tools, also have to > 'register' which database they end up using, then that database can > check the registration, and warn about programs using it, at pkgrm time. I just did a quick check on a debian box here. The gallery2 app depends on mysql-client|postgresql-client. With only the mysql version installed, it indicated gallery2 would be removed. Installing postgresql-client and then removing mysql-client triggered other dependency warnings, but not the gallery2 one. I was thinking the same, that a postinstall hook could register which package satisfied the dependency, thus creating something the package manager could use later.... Anyway, as I mentioned, the dbwrapper approach is likely suffiecient for this purpose...I wasn't intended to open a large debate. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Fri May 22 10:32:15 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 10:32:15 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1597E1.8060602@opencsw.org> (Mike Watters's message of "Thu\, 21 May 2009 13\:05\:21 -0500") References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> Message-ID: Mike Watters writes: > Philip Brown wrote: >> On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: >>>> 2. Depends on mail-server/Provides mail-server >>>> 3. Depends on mysql|postgres >>> This is one big thing the current Sun packages are really lacking >>> >> >> But how is that even processable? >> >> if you apt-get [somepackage], and that package >> "depends on mysql|postgres", and neither is installed... >> What happens? >> _______________________________________________ > > If Neither is installed it errors out normally. > if both are install it just installs. > > apt and yum do almost ZERO configuration for users. They doesn't need to. The configuration is the realm of the package based on system defaults/parameters. This is done in pre/post install scripts. Of course, there are helpful mechanisms such as alternatives, standard configurators, &c. Finally, there are policies enforced by various tools. Speaking as someone that have done, professionally Debian and RedHat packages. -- Peter From pfelecan at opencsw.org Fri May 22 10:35:30 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 10:35:30 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521181824.GE597@bolthole.com> (Philip Brown's message of "Thu\, 21 May 2009 11\:18\:25 -0700") References: <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> Message-ID: Philip Brown writes: > On Thu, May 21, 2009 at 01:05:21PM -0500, Mike Watters wrote: >> Philip Brown wrote: >> > But how is that even processable? >> > >> > if you apt-get [somepackage], and that package >> > "depends on mysql|postgres", and neither is installed... >> > What happens? >> >> If Neither is installed it errors out normally. >> if both are install it just installs. > > pfft. > > Sounds like there isnt much reason to hack that kind of > "multiple|optional dependancy' into the pkg delivery system then. > We could accomplish basically the same thing by other means. > > To take the database dependancy specifically; > I'm thinking some kind of generic database wrapper tool package. > Kinda like CPAN::DB, but for shellscripts :-) > Then packages could depend on CSWdbwrapper, and use the appropriate > tools in there as needed. > > With preference as to WHICH database to use, set in csw.conf locally, as I > suggested in my other email today. I'm supporting this idea, even if the idea of a package being a provider of an abstract service is nice but probably harder to implement in our context. > Not only does it solve the "problem" in that area, but it would solve it > BETTER than how debian 'solves' it. Not better, diferent. > How does that sound? Good -- Peter From pfelecan at opencsw.org Fri May 22 10:47:56 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 10:47:56 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: (Dagobert Michelsen's message of "Thu\, 21 May 2009 22\:55\:49 +0200") References: Message-ID: Dagobert Michelsen writes: > Hi, > > > today we have to celebrate a jubilee: Subversion commit #5000 was > done today Well, this show how young the project is... When you are in 5 or more digits it show that you're out of puberty. > So, please everybody grab a piece, I already had mine :-) > > _____ > _..--'''@ @'''--.._ > .' @_/-//-\/>/>'/ @ '. > ( @ /_ |'._ @ //|###########| > |~ ''--..@|',|}}}}}}}}}}}| > | ~ ~ |/ |###########| > | ~~ ~ ~|./|{{{{{{{{{{{| > '._ ~ ~ ~ |,/````````````` > ''--.~.|/ I hope that the next time we'll taste the good German cakes --- have a feeble for the schwarzwald kirche torte and the Bavarian ecosystem. Yum -- Peter From dam at opencsw.org Fri May 22 11:04:07 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 22 May 2009 11:04:07 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: References: Message-ID: Hi Peter, Am 22.05.2009 um 10:47 schrieb Peter FELECAN: >> So, please everybody grab a piece, I already had mine :-) >> >> _____ >> _..--'''@ @'''--.._ >> .' @_/-//-\/>/>'/ @ '. >> ( @ /_> |'._ @ //|###########| >> |~ ''--..@|',|}}}}}}}}}}}| >> | ~ ~ |/ |###########| >> | ~~ ~ ~|./|{{{{{{{{{{{| >> '._ ~ ~ ~ |,/````````````` >> ''--.~.|/ > > I hope that the next time we'll taste the good German cakes --- have a > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. But it is! }}} = cherry, ### = cake, {{{ = cream For clarification: It is "Schwarzw?lder Kirschtorte". "kirche = church" against "kirsche = cherry" :-D Best regards -- Dago From Joerg.Schilling at fokus.fraunhofer.de Fri May 22 11:26:56 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 22 May 2009 11:26:56 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: References: Message-ID: <4a166fe0.D4FEcP60kyIOZmVt%Joerg.Schilling@fokus.fraunhofer.de> Dagobert Michelsen wrote: > > I hope that the next time we'll taste the good German cakes --- have a > > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. > > But it is! }}} = cherry, ### = cake, {{{ = cream > For clarification: It is "Schwarzw?lder Kirschtorte". > "kirche = church" against "kirsche = cherry" :-D See the keyboard manufacturer from "Schwarzwald": Cherry ;-) J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From schwindt at dfki.uni-kl.de Fri May 22 11:47:10 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Fri, 22 May 2009 11:47:10 +0200 Subject: [csw-maintainers] gcal 3.0.1 now in testing Message-ID: <200905220947.n4M9lBcV028915@dfki.uni-kl.de> Very old package - nevertheless very usefull. Nicolai Schwindt. From pfelecan at opencsw.org Fri May 22 14:12:09 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 14:12:09 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: <4a166fe0.D4FEcP60kyIOZmVt%Joerg.Schilling@fokus.fraunhofer.de> (Joerg Schilling's message of "Fri\, 22 May 2009 11\:26\:56 +0200") References: <4a166fe0.D4FEcP60kyIOZmVt%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: > Dagobert Michelsen wrote: > >> > I hope that the next time we'll taste the good German cakes --- have a >> > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. >> >> But it is! }}} = cherry, ### = cake, {{{ = cream >> For clarification: It is "Schwarzw?lder Kirschtorte". >> "kirche = church" against "kirsche = cherry" :-D Sorry about this funny confusion... indeed I prefer the cherry for the torte. > See the keyboard manufacturer from "Schwarzwald": Cherry ;-) For which I have also a feeble: owning 3 of them clunkers, though my office neighbours don't like them... -- Peter From ellson at opencsw.org Fri May 22 20:50:29 2009 From: ellson at opencsw.org (John Ellson) Date: Fri, 22 May 2009 14:50:29 -0400 Subject: [csw-maintainers] Package "liblasi" in testing Message-ID: <4A16F3F5.9020708@opencsw.org> There is a new package: "liblasi: C++ Unicode support for PostScript" now available in testing. The GAR processing of this file was a bit non standard: - libLASi not permitted, so using liblasi for $(GARNAME) - Use of CMAKE - Sun's C++ compiler didn't work, so used g++ so, please see if: mgar/pkg/liblasi/trunk/Makefile can be improved. John From rupert at opencsw.org Fri May 22 21:38:53 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 22 May 2009 21:38:53 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1597E1.8060602@opencsw.org> References: <4A14A466.5010602@opencsw.org> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> Message-ID: <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> On Thu, May 21, 2009 at 20:05, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Philip Brown wrote: >> On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: >>>> 2. Depends on mail-server/Provides mail-server >>>> 3. Depends on mysql|postgres >>> This is one big thing the current Sun packages are really lacking >>> >> >> But how is that even processable? >> >> if you apt-get [somepackage], and that package >> "depends on mysql|postgres", and neither is installed... >> What happens? >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > If Neither is installed it errors out normally. > if both are install it just installs. > > apt and yum do almost ZERO configuration for users. would it be a lot of effort to replace pkgutil with apt? i especially like the remove parts, the dist features, and wajig as quite comfortable command line interface. e.g. wajig purge, wajig purge-orphans, ... rupert. From phil at bolthole.com Fri May 22 21:45:30 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 22 May 2009 12:45:30 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> References: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> Message-ID: <20090522194530.GH82222@bolthole.com> On Fri, May 22, 2009 at 09:38:53PM +0200, rupert THURNER wrote: > > would it be a lot of effort to replace pkgutil with apt? i especially > like the remove parts, the dist features, and wajig as quite > comfortable command line interface. > > e.g. wajig purge, wajig purge-orphans, ... If you want to attempt to port apt to work on top of SVR4 pkgs, please feel free to go right ahead :-) [I think someone actually attempted to do this, so you might be able to pick up on their work. But it's probably fairly out of date] Meanwhile, if you wish to request additional features to pkg-get and/or pkgutil, please also feel free to suggest specifics. You may not have noticed, but in one of my recent emails, I referenced support for a "purge" function. This has been in my plans for pkg-get for a long time. It would be easy for pkgutil to implement it as well. As mentioned in my email, the concept would be that maintainers would provde a "purge" script, as i purge I think that pkg-get could check for it, pre-pkgrm, and run it. From rupert at opencsw.org Fri May 22 22:12:37 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 22 May 2009 22:12:37 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools Message-ID: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> hi, while trying to easy_install python apps, i fell over a problem with the path mike used for compiling python: /opt/studio/.. . this is a problem for filesystems like ours which uses a readonly /opt. one has to set CC, LDSHARED, LDFLAGS, and for some like genshi LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be possible to link/mount sunstudio within /opt/csw especially for python compiles? see http://groups.google.com/group/distutils-sig/browse_thread/thread/d8d2c47eb1d96210 for details. i opened also an extension request, unfortunately in the wrong package as i did not understand it early enough: http://www.opencsw.org/bugtrack/view.php?id=3682. rupert. From rupert at opencsw.org Fri May 22 22:42:33 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 22 May 2009 22:42:33 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090522194530.GH82222@bolthole.com> References: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> <20090522194530.GH82222@bolthole.com> Message-ID: <6af4270905221342u5fb70d43yb5915c52aa4e9e3b@mail.gmail.com> On Fri, May 22, 2009 at 21:45, Philip Brown wrote: > On Fri, May 22, 2009 at 09:38:53PM +0200, rupert THURNER wrote: >> >> would it be a lot of effort to replace pkgutil with apt? i especially >> like the remove parts, the dist features, and wajig as quite >> comfortable command line interface. >> >> e.g. wajig purge, wajig purge-orphans, ... > > If you want to attempt to port apt to work on top of SVR4 pkgs, please feel > free to go right ahead :-) > [I think someone actually attempted to do this, so you might be able to > ?pick up on their work. But it's probably fairly out of date] the only thing up-to-date i know is http://nexenta.org, http://archive.nexenta.org/, who advertise themselves as "power of opensolaris with usability of linux". but i am a too bad packaging person to really judge this, and the impact / benefit it could have on opencsw. rupert. From phil at bolthole.com Fri May 22 23:34:03 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 22 May 2009 14:34:03 -0700 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> Message-ID: <20090522213403.GI82222@bolthole.com> On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: > hi, > > while trying to easy_install python apps, i fell over a problem with > the path mike used for compiling python: /opt/studio/.. . this is a > problem for filesystems like ours which uses a readonly /opt. ?? why is that a problem? hm. it appears from the newsgroup reference, that you are specifically complaining, that python has somehow hardcoded in its configs, CC=/full/path/to/cc instead of a nice simple CC=cc that second one definately sounds better. that being said... > one has to set CC, LDSHARED, LDFLAGS, and for some like genshi > LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be > possible to link/mount sunstudio within /opt/csw especially for python > compiles? > > see http://groups.google.com/group/distutils-sig/browse_thread/thread/d8d2c47eb1d96210 > for details. where did LD_LIBRARY_PATH come in? From phil at bolthole.com Fri May 22 23:38:10 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 22 May 2009 14:38:10 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <6af4270905221342u5fb70d43yb5915c52aa4e9e3b@mail.gmail.com> References: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> <20090522194530.GH82222@bolthole.com> <6af4270905221342u5fb70d43yb5915c52aa4e9e3b@mail.gmail.com> Message-ID: <20090522213810.GJ82222@bolthole.com> On Fri, May 22, 2009 at 10:42:33PM +0200, rupert THURNER wrote: > the only thing up-to-date i know is http://nexenta.org, > http://archive.nexenta.org/, who advertise themselves as "power of > opensolaris with usability of linux". but i am a too bad packaging > person to really judge this, and the impact / benefit it could have on > opencsw. note that they use opensolaris. their own completely custom distribution of solaris. They have solaris-native apt-get, using debian style packaging, NOT SVR4 packaging, I think. Verifying... http://www.nexenta.org/os/DownloadMirrors http://mirror.sourceshare.org/nexenta/apt/dists/elatte-unstable/main/binary-solaris-i386/mail/ has a bunch of .deb files. yup. debian .deb/dpkg format. not svr4. From mwatters at opencsw.org Sat May 23 00:41:25 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 22 May 2009 17:41:25 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> Message-ID: <4A172A15.2030303@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > hi, > > while trying to easy_install python apps, i fell over a problem with > the path mike used for compiling python: /opt/studio/.. . this is a > problem for filesystems like ours which uses a readonly /opt. > > one has to set CC, LDSHARED, LDFLAGS, and for some like genshi > LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be > possible to link/mount sunstudio within /opt/csw especially for python > compiles? > > see http://groups.google.com/group/distutils-sig/browse_thread/thread/d8d2c47eb1d96210 > for details. > > i opened also an extension request, unfortunately in the wrong package > as i did not understand it early enough: > http://www.opencsw.org/bugtrack/view.php?id=3682. > > rupert. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I moved the bug to the right project, The new bug URL is http://www.opencsw.org/mantis/view.php?id=3684 This is a relatively easy fix, I will add it to the next build. the next build will be the one that no longer contains the pyo/pyc files but generates them on install.... still working on the class scripts for this. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoXKhUACgkQLrhmsXMSLxe04gCgmN13tAOsRsT2axVxhKDpInsS UToAoOZ0MNsLO0b7L81m4HALyV+mLbdD =isOE -----END PGP SIGNATURE----- From rupert at opencsw.org Sat May 23 02:56:38 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 23 May 2009 02:56:38 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <20090522213403.GI82222@bolthole.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> Message-ID: <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> On Fri, May 22, 2009 at 23:34, Philip Brown wrote: > On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: >> hi, >> >> while trying to easy_install python apps, i fell over a problem with >> the path mike used for compiling python: /opt/studio/.. . this is a >> problem for filesystems like ours which uses a readonly /opt. > > ?? > > why is that a problem? > > > hm. it appears from the newsgroup reference, that you are specifically > complaining, that python has somehow hardcoded in its configs, > CC=/full/path/to/cc > instead of a nice simple > > CC=cc the "problem" if you want to call it like this is that easy_install uses the settings out of the python compilation makefile as default. it would be nicer if the default, i.e. the way python is compiled, uses a compiler within /opt/csw to make it easy to duplicate it on a opencsw client site, where other paths outside /opt/csw are not necessarily writable. i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set to /opt/csw/lib to be easy_installed at our site though. rupert. From mwatters at opencsw.org Sat May 23 03:03:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 22 May 2009 20:03:17 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> Message-ID: <4A174B55.6000303@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > On Fri, May 22, 2009 at 23:34, Philip Brown wrote: >> On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: >>> hi, >>> >>> while trying to easy_install python apps, i fell over a problem with >>> the path mike used for compiling python: /opt/studio/.. . this is a >>> problem for filesystems like ours which uses a readonly /opt. >> ?? >> >> why is that a problem? >> >> >> hm. it appears from the newsgroup reference, that you are specifically >> complaining, that python has somehow hardcoded in its configs, >> CC=/full/path/to/cc >> instead of a nice simple >> >> CC=cc > > the "problem" if you want to call it like this is that easy_install > uses the settings out of the python compilation makefile as default. > it would be nicer if the default, i.e. the way python is compiled, > uses a compiler within /opt/csw to make it easy to duplicate it on a > opencsw client site, where other paths outside /opt/csw are not > necessarily writable. > > i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set > to /opt/csw/lib to be easy_installed at our site though. > > rupert. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Technically, this could/should be resolved by requesting CSW builds of those extensions. I have already got a genshi package in the works because I have been playing with Trac, all I have to do is a couple tweaks to the build and do a final commit. I am running it on my test Trac server and it works great. ;) I will finish it up next week. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoXS1UACgkQLrhmsXMSLxcERQCfZZN1cH0XSX27+MplX+1PQN1E DD8AniDa2KI0IePNik6P+ghA8PkcpAEJ =A1Cd -----END PGP SIGNATURE----- From rupert at opencsw.org Sat May 23 03:09:34 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 23 May 2009 03:09:34 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <4A174B55.6000303@opencsw.org> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> Message-ID: <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> On Sat, May 23, 2009 at 03:03, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > rupert THURNER wrote: >> On Fri, May 22, 2009 at 23:34, Philip Brown wrote: >>> On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: >>>> hi, >>>> >>>> while trying to easy_install python apps, i fell over a problem with >>>> the path mike used for compiling python: /opt/studio/.. . this is a >>>> problem for filesystems like ours which uses a readonly /opt. >>> ?? >>> >>> why is that a problem? >>> >>> >>> hm. it appears from the newsgroup reference, that you are specifically >>> complaining, that python has somehow hardcoded in its configs, >>> CC=/full/path/to/cc >>> instead of a nice simple >>> >>> CC=cc >> >> the "problem" if you want to call it like this is that easy_install >> uses the settings out of the python compilation makefile as default. >> it would be nicer if the default, i.e. the way python is compiled, >> uses a compiler within /opt/csw to make it easy to duplicate it on a >> opencsw client site, where other paths outside /opt/csw are not >> necessarily writable. >> >> i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set >> to /opt/csw/lib to be easy_installed at our site though. >> >> rupert. >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > Technically, this could/should be resolved by requesting CSW builds of those > extensions. ?I have already got a genshi package in the works because I have > been playing with Trac, ?all I have to do is a couple tweaks to the build and > do a final commit. ?I am running it on my test Trac server and it works great. > ?;) ?I will finish it up next week. that is great to hear! i even would appreciate if the most important plugins make it into trac as well (not enabled by default), e.g. easy_install http://trac-hacks.org/svn/tracwysiwygplugin/0.11/ easy_install http://trac-hacks.org/svn/ticketdeleteplugin/0.11/ easy_install http://trac-hacks.org/svn/ticketchangeplugin/0.11/ easy_install http://trac-hacks.org/svn/customfieldadminplugin/0.11/ easy_install http://trac-hacks.org/svn/tagsplugin/tags/0.6/ easy_install http://svn.ipd.uka.de/repos/javaparty/JP/trac/plugins/tracnav-0.11/ easy_install http://trac-hacks.org/svn/sectioneditplugin/0.11/ easy_install http://trac-hacks.org/svn/extracturlplugin/0.11/ easy_install http://trac-hacks.org/svn/serversideredirectplugin/0.11/ easy_install http://trac-hacks.org/svn/masterticketsplugin/0.11/ easy_install http://trac-hacks.org/svn/autoupgradeplugin/0.11/ but, otoh, one can never package all python apps, so a possibility to quickly compile them with a default setting would be really nice :) rupert. From mwatters at opencsw.org Sat May 23 16:16:56 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 23 May 2009 09:16:56 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> Message-ID: <4A180558.70601@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > > that is great to hear! i even would appreciate if the most important > plugins make it into trac as well (not enabled by default), e.g. > > easy_install http://trac-hacks.org/svn/tracwysiwygplugin/0.11/ > easy_install http://trac-hacks.org/svn/ticketdeleteplugin/0.11/ > easy_install http://trac-hacks.org/svn/ticketchangeplugin/0.11/ > easy_install http://trac-hacks.org/svn/customfieldadminplugin/0.11/ > easy_install http://trac-hacks.org/svn/tagsplugin/tags/0.6/ > easy_install http://svn.ipd.uka.de/repos/javaparty/JP/trac/plugins/tracnav-0.11/ > easy_install http://trac-hacks.org/svn/sectioneditplugin/0.11/ > easy_install http://trac-hacks.org/svn/extracturlplugin/0.11/ > easy_install http://trac-hacks.org/svn/serversideredirectplugin/0.11/ > easy_install http://trac-hacks.org/svn/masterticketsplugin/0.11/ > easy_install http://trac-hacks.org/svn/autoupgradeplugin/0.11/ > I am working on updating Trac in the repository. ( after I get python re-released as separate packages.) I will create a second package for trac, trac_plugin_pack that contains the eggs you listed above. also, your last post on the google group forum link in the ticket. the '-G' is gone because you export LDSHARED=/opt/SUNWspro/bin/cc ''' compiler.linker_so ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] ''' means that then setting for compiler.linker_so is /opt/studio/SOS11/SUNWspro/bin/cc -G os.environ settings override what python finds in its config. export LDSHARED='/opt/SUNWspro/bin/cc -G' would have saved the -G you can easily play with this using the following python script =============================================================== #! /usr/bin/env python from distutils.ccompiler import new_compiler from distutils.sysconfig import customize_compiler compiler = new_compiler(None,None,None,None) customize_compiler(compiler) print """ *** Compiler Settings for Python *** Type: %s Linker: %s Compiler: %s """ % (compiler.compiler_type, compiler.linker_so, compiler.compiler) ====================================================================== $ ./pycompile.py *** Compiler Settings for Python *** Type: unix Linker: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] Compiler: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-DNDEBUG', '-O'] you can set various environment variables and see how that affects those settings. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoYBVgACgkQLrhmsXMSLxdLswCg4qlEUAdZIpxfr6B/0UpQhQHj UW4AoLk0dKQ3J5rsmkEwylnShAaWUtrh =GqGq -----END PGP SIGNATURE----- From phil at bolthole.com Sat May 23 18:36:57 2009 From: phil at bolthole.com (Philip Brown) Date: Sat, 23 May 2009 09:36:57 -0700 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> Message-ID: <20090523163657.GA24550@bolthole.com> On Sat, May 23, 2009 at 02:56:38AM +0200, rupert THURNER wrote: > i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set > to /opt/csw/lib to be easy_installed at our site though. possibly because by default, LD_LIBRARY_PATH is set incorrectly in the user environment to start with. otherwise, how about you post here what breaks when it is not set (with "env - /opt/csw/bin/hg" to clear your environment first) and folks might be able to suggest something From bwalton at opencsw.org Sat May 23 19:12:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 23 May 2009 13:12:01 -0400 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> Message-ID: <1243098505-sup-2495@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Fri May 22 16:12:37 -0400 2009: > one has to set CC, LDSHARED, LDFLAGS, and for some like genshi > LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be > possible to link/mount sunstudio within /opt/csw especially for python > compiles? With regard to the CC vs /opt/studio issue, could we solve this by having GAR prefix the PATH with the appropriate $($(GARCOMPILER)_CC_HOME)/bin and then set CC appropriately without the full path? Doing this would make things (like python, ruby, etc) that store the values used to build them work against an appropriately set path at any site...Am I overlooking a negative impact that would be caused by a change like this? I don't have much input for the other issues and it seems that Mike and Phil have provided some useful starting points. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Sat May 23 19:23:16 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 23 May 2009 12:23:16 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <1243098505-sup-2495@ntdws12.chass.utoronto.ca> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <1243098505-sup-2495@ntdws12.chass.utoronto.ca> Message-ID: <4A183104.8030503@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > With regard to the CC vs /opt/studio issue, could we solve this by > having GAR prefix the PATH with the appropriate > $($(GARCOMPILER)_CC_HOME)/bin and then set CC appropriately without > the full path? Doing this would make things (like python, ruby, etc) > that store the values used to build them work against an appropriately > set path at any site...Am I overlooking a negative impact that would > be caused by a change like this? I can't think of any negative for this. having CC report as cc or gcc is much better IMO then the full path. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoYMQQACgkQLrhmsXMSLxcZOwCgnBmFiDJlzOj98Fl9KSzN4Dvu gv4AoMV6woT9R8A5yFyXtzX0J40xzTLB =swBW -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Sat May 23 20:19:55 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Sat, 23 May 2009 20:19:55 +0200 (CEST) Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <4A174B55.6000303@opencsw.org> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> Message-ID: <32825.88.134.110.210.1243102795.squirrel@www.dfki.uni-kl.de> On Sat, May 23, 2009 03:03, Mike Watters wrote: [...] > Technically, this could/should be resolved by requesting CSW builds of > those > extensions. I have already got a genshi package in the works because I > have > been playing with Trac, all I have to do is a couple tweaks to the build > and > do a final commit. I am running it on my test Trac server and it works > great. > ;) I will finish it up next week. .) That's why I asked how to know whos taking over which packet. I had already done trac 0.11.4 butwas waiting on python, svn, and pysqlite to settle ,) As you have the same Idea of making a trac-plugin-pack pls consider these aswell : TracSubversionLocation-1.0.1 GitPython-0.1.6 accountmanagerplugin_0.11-r5499 downloadsplugin-r5499 httpauthplugin-r5728 iniadminplugin_0.11 privatewikiplugin-r5499 reposearchplugin BTW : ticketdelete and ticketchange have been merged. -- Nicolai ----------------------------------------- This email was sent using SquirrelMail. "Webmail for nuts!" http://squirrelmail.org/ From rupert at opencsw.org Sat May 23 23:31:06 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 23 May 2009 23:31:06 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <32825.88.134.110.210.1243102795.squirrel@www.dfki.uni-kl.de> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <32825.88.134.110.210.1243102795.squirrel@www.dfki.uni-kl.de> Message-ID: <6af4270905231431g34edd432ieb9438d8396b5ff0@mail.gmail.com> On Sat, May 23, 2009 at 20:19, Nicolai Schwindt wrote: > > > > > On Sat, May 23, 2009 03:03, Mike Watters wrote: > [...] >> Technically, this could/should be resolved by requesting CSW builds of >> those >> extensions. ?I have already got a genshi package in the works because I >> have >> been playing with Trac, ?all I have to do is a couple tweaks to the build >> and >> do a final commit. ?I am running it on my test Trac server and it works >> great. >> ?;) ?I will finish it up next week. > > .) That's why I asked how to know whos taking over which packet. > I had already done trac 0.11.4 butwas waiting on python, svn, and pysqlite > to settle ,) > > As you have the same Idea of making a trac-plugin-pack pls consider > these aswell : > > TracSubversionLocation-1.0.1 > GitPython-0.1.6 > accountmanagerplugin_0.11-r5499 > downloadsplugin-r5499 > httpauthplugin-r5728 > iniadminplugin_0.11 > privatewikiplugin-r5499 > reposearchplugin > > BTW : ticketdelete and ticketchange have been merged. did one of you manage to try out http://trac-hacks.org/wiki/TracWikiPrintPlugin which allows to print pdf's? my problem was, that i could not get reportlab to compile correctly with sunstudio. mike, you managed to compile some of the packages with gcc, isn't it? do you have an example so i could try this with reportlab as well? rupert. From ja at opencsw.org Sun May 24 11:44:23 2009 From: ja at opencsw.org (Juergen Arndt) Date: Sun, 24 May 2009 11:44:23 +0200 Subject: [csw-maintainers] Icinga 0.8 in testing Message-ID: Hi, Icinga 0.8 - a fork of Nagios - is now in testing. Still without the database backend IDOutils. Any feedback is welcome. Juergen -- Juergen Arndt From rupert at opencsw.org Sun May 24 21:32:44 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 24 May 2009 21:32:44 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> References: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> Message-ID: <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> i tried to compile, which worked on build8s. but on build8x and build10x it failed. what might be the reason for this? does sun have different compilers? /opt/studio/SOS11/SUNWspro/bin/cc -DNDEBUG -O -xO3 -xarch=386 -I/opt/csw/include -I/opt/csw/include -DRENDERPM_FT -DLIBART_COMPILATION -DLIBART_VERSION="2.3.12" -I/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM -I/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/libart_lgpl -I/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/gt1 -I/opt/csw/include -I/opt/csw/include/freetype2 -I/opt/csw/include/python2.6 -c /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c -o build/temp.solaris-2.8-i86pc-2.6/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.o "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 81: syntax error before or at: 13 "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 96: syntax error before or at: 13 "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 100: syntax error before or at: : "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 102: warning: statement not reached "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 105: cannot recover from previous errors cc: acomp failed for /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c error: command '/opt/studio/SOS11/SUNWspro/bin/cc' failed with exit status 2 gmake[1]: *** [install-custom] Error 1 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/reportlab/trunk' gmake: *** [merge-isa-i386] Error 2 i also checked in the makefile, put the compiled 8s on testing with write access for everybody. rupert. On Sat, Apr 11, 2009 at 12:29, rupert THURNER wrote: > hi, > > for getting http://trac-hacks.org/wiki/TracWikiPrintPlugin we would > need http://reportlab.org/ v2.3 (http://www.reportlab.org/ftp/) . i > saw there is a package http://opencsw.org/packages/reportlab, v2.1. > > as it is not in > http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg, where would > be the source to try it? > > rupert. > From mwatters at opencsw.org Sun May 24 22:08:33 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 24 May 2009 15:08:33 -0500 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> References: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> Message-ID: <4A19A941.6070306@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > i tried to compile, which worked on build8s. but on build8x and > build10x it failed. what might be the reason for this? does sun have > different compilers? > -I/opt/csw/include/python2.6 -c > /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c > -o build/temp.solaris-2.8-i86pc-2.6/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.o > "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", > line 81: syntax error before or at: 13 > "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", > line 96: syntax error before or at: 13 > "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", > line 100: syntax error before or at: : The output of make has clues for you to look at. note the syntax errors. I assume there are header differences or library differences between the 2 systems, this is not uncommon. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoZqUAACgkQLrhmsXMSLxdAOgCgpvHpo9ZMcH1sEiyPaCC0Hl/L fmgAn0zyLu1J99+7uM35f54R38sxNppP =1/GN -----END PGP SIGNATURE----- From ihsan at opencsw.org Sun May 24 22:31:10 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 24 May 2009 22:31:10 +0200 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242926828-sup-5777@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> Message-ID: <4A19AE8E.2030306@opencsw.org> Am 21.5.2009 19:29 Uhr, Ben Walton schrieb: >> handle dependencies as we have always handled them. In addition, >> I propose to carefully start working on real IPS packaging >> instead of putting extra work/complexity in System V packaging. >> A real IPS backend for GAR with all bells and whistels for >> dual-personality building of packages is surely a Good Thing. > Yes, I agree. Will IPS make pkgutil/pkg-get redundant? Eg, will it > facilitate simply creating a 'repo' and adding it to the IPS conf > (similar to apt/yum)? If so, all the smarts could be punted to the > system tools instead, which would be nice. Is it now sure, that IPS is going to be in Solaris 11? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 24 22:40:11 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 24 May 2009 22:40:11 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: References: Message-ID: <4A19B0AB.7030001@opencsw.org> Am 22.5.2009 10:47 Uhr, Peter FELECAN schrieb: > I hope that the next time we'll taste the good German cakes --- have a > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. As far the Schwarzwald Region is just across the border, I could bring a Schwarzw?lder Kirschtorte to Norway. :-) PS: There is also an excellent Beer from the Schwarzwald region: http://www.waldhaus-bier.de/ Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Sun May 24 23:14:08 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 24 May 2009 23:14:08 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <4A19A941.6070306@opencsw.org> References: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> <4A19A941.6070306@opencsw.org> Message-ID: Hi, Am 24.05.2009 um 22:08 schrieb Mike Watters: > rupert THURNER wrote: >> i tried to compile, which worked on build8s. but on build8x and >> build10x it failed. what might be the reason for this? does sun have >> different compilers? > >> -I/opt/csw/include/python2.6 -c >> /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c >> -o build/temp.solaris-2.8-i86pc-2.6/home/rupert/mgar/pkg/reportlab/ >> trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/ >> _renderPM.o >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", >> line 81: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", >> line 96: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", >> line 100: syntax error before or at: : > > The output of make has clues for you to look at. note the syntax > errors. > I assume there are header differences or library differences between > the 2 > systems, this is not uncommon. Or x86 assembler optimizations Sun Studio doesn't understand. You can try disabling these during configure or use gcc instead. Best regards -- Dago From Murray.Jensen at csiro.au Mon May 25 03:04:35 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Mon, 25 May 2009 11:04:35 +1000 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: Your message of "Mon, 25 May 2009 06:08:33 +1000" <4A19A941.6070306@opencsw.org> Message-ID: <10840.1243213475@gerd> On Mon, 25 May 2009 06:08:33 +1000, Mike Watters writes: >rupert THURNER wrote: >> i tried to compile, which worked on build8s. but on build8x and >> build10x it failed. what might be the reason for this? does sun have >> different compilers? Just checked out my stuff and reportlab 2.1 built cleanly out of the box on solaris 8 for both intel and sparc using Sun Studio 8(?). >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >src/rl_addons/renderPM/_renderPM.c", >> line 81: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >src/rl_addons/renderPM/_renderPM.c", >> line 96: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >src/rl_addons/renderPM/_renderPM.c", >> line 100: syntax error before or at: : > >The output of make has clues for you to look at. note the syntax errors. >I assume there are header differences or library differences between the 2 >systems, this is not uncommon. Ahh.. do you have renderPM installed and at the correct version for reportlab 2.3? Also, I used the rl_accel accelerator plugin - seemed like a good idea at the time. Cheers! Murray... From rupert at opencsw.org Mon May 25 15:58:37 2009 From: rupert at opencsw.org (rupert THURNER) Date: Mon, 25 May 2009 15:58:37 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <10840.1243213475@gerd> References: <4A19A941.6070306@opencsw.org> <10840.1243213475@gerd> Message-ID: <6af4270905250658u4e045f47v3e2a763400f14109@mail.gmail.com> On Mon, May 25, 2009 at 03:04, wrote: > On Mon, 25 May 2009 06:08:33 +1000, Mike Watters writes: >>rupert THURNER wrote: >>> i tried to compile, which worked on build8s. but on build8x and >>> build10x it failed. what might be the reason for this? does sun have >>> different compilers? > > Just checked out my stuff and reportlab 2.1 built cleanly out of the box on > solaris 8 for both intel and sparc using Sun Studio 8(?). > >>> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >>src/rl_addons/renderPM/_renderPM.c", >>> line 81: syntax error before or at: 13 >>> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >>src/rl_addons/renderPM/_renderPM.c", >>> line 96: syntax error before or at: 13 >>> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >>src/rl_addons/renderPM/_renderPM.c", >>> line 100: syntax error before or at: : >> >>The output of make has clues for you to look at. note the syntax errors. >>I assume there are header differences or library differences between the 2 >>systems, this is not uncommon. > > Ahh.. do you have renderPM installed and at the correct version for reportlab > 2.3? Also, I used the rl_accel accelerator plugin - seemed like a good idea > at the time. Cheers! i tried to build it on the standard build servers. how did you install it?or could i have access to your build file? rupert. From bonivart at opencsw.org Mon May 25 19:11:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 25 May 2009 19:11:14 +0200 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request In-Reply-To: <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> References: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> Message-ID: <625385e30905251011ta6ae38ar5922911e22412ac7@mail.gmail.com> On Wed, May 20, 2009 at 10:46 PM, Peter Bonivart wrote: > On Wed, May 20, 2009 at 6:05 PM, Ben Walton wrote: >> Would you consider adding a -C option to the package management >> utilities to at would limit the compare output to only versions behind >> the catalog (eg: `pkg-get -U -c | grep -v SAME` is what I use to see >> what will be updated). > > Sure, I can add it to 1.6 which is almost done. I never implemented > showing not installed packages but I still get annoyed by using grep > -v SAME so it's a good idea. There's now a beta in testing: http://mirror.opencsw.org/testing/pkgutil-1.6b1,REV=2009.05.25-SunOS5.8-i386-CSW.pkg.gz http://mirror.opencsw.org/testing/pkgutil-1.6b1,REV=2009.05.25-SunOS5.8-sparc-CSW.pkg.gz Change log: - New option -C, same as -c but only shows different versions, no more piping to grep -v SAME (Ben Walton) - New option --trace to debug pkgadd/pkgrm operations (Yann Rouillard) - Replaced some system commands with native Perl - New option --config for using custom configuration file (Don Jackson) - Fix size formatting - Skip showing pkgadd patch when using -V on 5.11 (Dennis Clarke) - Update credits in perldoc/man page - Perldoc/man page added to both bldcat and chkcat - Chkcat now includes more tests (Derek Whayman) - Chkcat exit values are 0 for OK, 1 for warning and 2 for error - New option -e (erroronly) added to chkcat to supress warnings - New option -q (quiet) added to chkcat to supress all output -- /peter From dam at opencsw.org Mon May 25 22:10:48 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 25 May 2009 22:10:48 +0200 Subject: [csw-maintainers] What to do with Nessus? Message-ID: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> Hi, as I browsed the collection I noticed that we have a very ancient version of Nessus: However, Nessus is no longer be free. As it is a security scanner it is most likely no useful any more, so I propose to drop it and the associated packages nessuslib, nessusplugins and libnasl. Best regards -- Dago From phil at bolthole.com Mon May 25 23:55:06 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 25 May 2009 14:55:06 -0700 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> Message-ID: <20090525215506.GA81042@bolthole.com> On Mon, May 25, 2009 at 10:10:48PM +0200, Dagobert Michelsen wrote: > Hi, > > as I browsed the collection I noticed that we have a very ancient > version of Nessus: > > However, Nessus is no longer be free. errr... its semi-free. but anyways, i think personally, that we should upgrade it to the last truely-free version, if we can From Murray.Jensen at csiro.au Tue May 26 02:32:23 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Tue, 26 May 2009 10:32:23 +1000 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: Your message of "Mon, 25 May 2009 23:58:37 +1000" <6af4270905250658u4e045f47v3e2a763400f14109@mail.gmail.com> Message-ID: <14077.1243297943@gerd> On Mon, 25 May 2009 23:58:37 +1000, rupert THURNER writes: >i tried to build it on the standard build servers. how did you install >it? I have my own "build servers". You will have to request that renderPM be installed on the central "build servers". But first check that the version of renderPM in CSW is late enough for reportlab 2.3 - you may need to build and submit renderPM first, then get it installed on the "build servers", then move on to building reportlab. Also don't forget to check out the rl_accel plugin for reportlab. >or could i have access to your build file? My build file basically is this: python setup.py build mkdir -p ${buildroot} python setup.py install --root=${buildroot} chown -Rh root:bin ${buildroot}/opt i.e. the standard way to build and install python packages (I believe). GAR will probably have mechanisms for this by default(?). Cheers! Murray... From bwalton at opencsw.org Tue May 26 02:37:57 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 25 May 2009 20:37:57 -0400 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request In-Reply-To: <625385e30905251011ta6ae38ar5922911e22412ac7@mail.gmail.com> References: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> <625385e30905251011ta6ae38ar5922911e22412ac7@mail.gmail.com> Message-ID: <1243298243-sup-8147@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Mon May 25 13:11:14 -0400 2009: Hi Peter, > - New option -C, same as -c but only shows different versions, no more > piping to grep -v SAME (Ben Walton) I tried this out just now. It looks great from here! Thanks for adding that so quickly. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darin.Perusich at cognigencorp.com Tue May 26 14:01:21 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Tue, 26 May 2009 08:01:21 -0400 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> Message-ID: <4A1BDA11.4040308@cognigencorp.com> There's always OpenVAS which is a fork of Nessus. Dagobert Michelsen wrote: > Hi, > > as I browsed the collection I noticed that we have a very ancient > version of Nessus: > > However, Nessus is no longer be free. As it is a security scanner it > is most likely no useful any more, so I propose to drop it and the > associated packages nessuslib, nessusplugins and libnasl. > -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From ihsan at opencsw.org Tue May 26 15:29:10 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 26 May 2009 15:29:10 +0200 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <4A1BDA11.4040308@cognigencorp.com> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> Message-ID: <4A1BEEA6.70707@opencsw.org> I think a transition to OpenVAS would make sense then. Am 26.5.2009 14:01 Uhr, Darin Perusich schrieb: > There's always OpenVAS which is a fork of Nessus. > > Dagobert Michelsen wrote: >> Hi, >> >> as I browsed the collection I noticed that we have a very ancient >> version of Nessus: >> >> However, Nessus is no longer be free. As it is a security scanner it >> is most likely no useful any more, so I propose to drop it and the >> associated packages nessuslib, nessusplugins and libnasl. >> > > -- ihsan at dogan.ch http://blog.dogan.ch/ From skayser at opencsw.org Tue May 26 16:07:44 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 26 May 2009 16:07:44 +0200 (CEST) Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? Message-ID: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> Hi, i have some machines that are not connected to the internet, where i would like to install some OpenCSW packages. What would be the best way to download these packages (inluding dependencies) for offline installation? I don't want to pull the whole mirror, just the packages and their deps. I have thought about setting up a full root zone w/o CSW packages and do a pkg-get/pkgutil -d there, but i would prefer if there was an easier way. Sebastian From dam at opencsw.org Tue May 26 17:06:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 17:06:19 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space Message-ID: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> Hi, I am currently setting up build10st and need some more space for it and noticed that Hudson is taking up around 25 GB of space. Looking into /var/tmp is, umh, impressing... > root at ncsw [global]:/zones/hudson/root/var/tmp > time ls -1 | wc -l > 824725 > ls -1 8.70s user 4.23s system 18% cpu 1:08.11 total > wc -l 0.32s user 0.02s system 0% cpu 1:07.86 total Trygve, could you please set up some kind of cleanup job? I'm going to delete the temp stuff for now manually. Best regards -- Dago From phil at bolthole.com Tue May 26 17:27:46 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 26 May 2009 08:27:46 -0700 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <4A1BEEA6.70707@opencsw.org> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> <4A1BEEA6.70707@opencsw.org> Message-ID: <20090526152746.GB17249@bolthole.com> On Tue, May 26, 2009 at 03:29:10PM +0200, Ihsan Dogan wrote: > I think a transition to OpenVAS would make sense then. > can you confirm or deny whether openvas covers all the free-nessus features? From phil at bolthole.com Tue May 26 17:28:38 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 26 May 2009 08:28:38 -0700 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <4A1BDA11.4040308@cognigencorp.com> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> Message-ID: <20090526152838.GC17249@bolthole.com> On Tue, May 26, 2009 at 08:01:21AM -0400, Darin Perusich wrote: > There's always OpenVAS which is a fork of Nessus. > oh, i didnt read this message before sending my other one :-) yes, this makes sense to me. Someone could package that up, and declare conflict with CSWnessus, if they liked, and then i would remove nessus From mwatters at opencsw.org Tue May 26 17:54:06 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 26 May 2009 10:54:06 -0500 Subject: [csw-maintainers] anyone updating openldap? Message-ID: <4A1C109E.10904@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I need amd64 libs for openldap, is anyone working on this package? from the List it is owned by Alex, who is retired. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkocEJ4ACgkQLrhmsXMSLxcFFQCgk2lUfaMIzcjRBW+fsxNvySoD BdAAni1F1ax1p7cTFn6syiMEs4o8MWsk =toQQ -----END PGP SIGNATURE----- From dam at opencsw.org Tue May 26 18:00:38 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 18:00:38 +0200 Subject: [csw-maintainers] anyone updating openldap? In-Reply-To: <4A1C109E.10904@opencsw.org> References: <4A1C109E.10904@opencsw.org> Message-ID: Hi Mike, Am 26.05.2009 um 17:54 schrieb Mike Watters: > I need amd64 libs for openldap, is anyone working on this package? > > from the List it is owned by Alex, who is retired. I made a Makefile for it some time ago but didn't finish the package. The start/stopp mechanism is still missing. If you add them and take a look at the patches I update it to mGAR v2. Best regards -- Dago From phil at bolthole.com Tue May 26 18:14:51 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 26 May 2009 09:14:51 -0700 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> Message-ID: <20090526161451.GE17249@bolthole.com> On Tue, May 26, 2009 at 04:07:44PM +0200, Sebastian Kayser wrote: > Hi, > > i have some machines that are not connected to the internet, where i would > like to install some OpenCSW packages. What would be the best way to > download these packages (inluding dependencies) for offline installation? > I don't want to pull the whole mirror, just the packages and their deps. > > I have thought about setting up a full root zone w/o CSW packages and do a > pkg-get/pkgutil -d there, but i would prefer if there was an easier way. That's about the least annoying method i can think of, unfortunately. slightly improved by the fact that, once you have the ones you want downloaded, you can also download the catalog to that directory, and then use the usual tools to that dir. From mwatters at opencsw.org Tue May 26 18:20:54 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 26 May 2009 11:20:54 -0500 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> Message-ID: <4A1C16E6.2040005@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Sebastian Kayser wrote: > Hi, > > i have some machines that are not connected to the internet, where i would > like to install some OpenCSW packages. What would be the best way to > download these packages (inluding dependencies) for offline installation? > I don't want to pull the whole mirror, just the packages and their deps. > > I have thought about setting up a full root zone w/o CSW packages and do a > pkg-get/pkgutil -d there, but i would prefer if there was an easier way. I assume they have network access internal, if you have a system that can connect to the internet, you could setup a local mirror and you could update all of your internal systems using that mirror. use rsync and update as often as you feel appropriate. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkocFuYACgkQLrhmsXMSLxezfgCfb7YUxfss3+NPx98Cs/t3LtUQ akYAn0g7LNcUHDbVlLxCZF8RRu15niOI =1TM3 -----END PGP SIGNATURE----- From skayser at opencsw.org Tue May 26 21:47:33 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 26 May 2009 21:47:33 +0200 (CEST) Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <4A1C16E6.2040005@opencsw.org> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> Message-ID: <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Mike Watters wrote: > Sebastian Kayser wrote: >> i have some machines that are not connected to the internet, where i >> would >> like to install some OpenCSW packages. What would be the best way to >> download these packages (inluding dependencies) for offline >> installation? >> I don't want to pull the whole mirror, just the packages and their deps. >> >> I have thought about setting up a full root zone w/o CSW packages and do >> a >> pkg-get/pkgutil -d there, but i would prefer if there was an easier way. > > I assume they have network access internal, if you have a system that can > connect to the internet, you could setup a local mirror and you could > update > all of your internal systems using that mirror. use rsync and update as > often > as you feel appropriate. There is no machine on this net with internet access ... restricted environment. While i am still a bit reluctant to sync the whole mirror, there is a workaround using pkgutil. PKG_INSTALL_ROOT=/dev/null pkgutil -d This will make pkgparam which is invoked by pkgutil to check for installed packages return an error and thus pkgutil will think that none of the required/requested packages are installed. I end up with all required pkgs and can transfer them via a USB stick/disk. :D Btw., what's the disk usage of our current/ tree? Sebastian From dam at opencsw.org Tue May 26 22:08:21 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 22:08:21 +0200 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Message-ID: Hi Sebastian, Am 26.05.2009 um 21:47 schrieb Sebastian Kayser: > There is no machine on this net with internet access ... restricted > environment. While i am still a bit reluctant to sync the whole > mirror, > there is a workaround using pkgutil. > > PKG_INSTALL_ROOT=/dev/null pkgutil -d Why don't you use this option? -s, --stream Build a package stream for a certain architecture and OS release. All dependencies will be included in the stream and the needed command to install them in the correct order will be displayed. I requested it some time ago for this exact prupose. > Btw., what's the disk usage of our current/ tree? dam at login [login]:/export/mirror/opencsw/current/sparc/5.8 > du -sh . 4.1G . Feel free to check yourself, it is mounted in @login Best regards -- Dago From william at wbonnet.net Tue May 26 22:31:06 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 26 May 2009 22:31:06 +0200 Subject: [csw-maintainers] pixman 0.15.6 is in testing Message-ID: <4A1C518A.6090503@wbonnet.net> Hi Pixman package has been updated to latest version (0.15.6) and is available from testing Any feedback is welcome kind regards, W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bonivart at opencsw.org Tue May 26 23:09:58 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 26 May 2009 23:09:58 +0200 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Message-ID: <625385e30905261409o62054c33re72cfdf9780513d2@mail.gmail.com> On Tue, May 26, 2009 at 10:08 PM, Dagobert Michelsen wrote: > Why don't you use this option? > > ? ? -s, --stream > ? ? ? ? ?Build a package stream for a certain architecture and > ? ? ? ? ?OS release. All dependencies will be included in the > ? ? ? ? ?stream and the needed command to install them in the > ? ? ? ? ?correct order will be displayed. > > I requested it some time ago for this exact prupose. Thanks for pointing that out Dago. Here's a couple of examples: http://pkgutil.wikidot.com/use-with-examples#toc7 But basically, if you want a stream with all deps for bind and dhcp for example you just do this: # pkgutil -s bind dhcp -- /peter From trygvis at opencsw.org Tue May 26 23:20:19 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Tue, 26 May 2009 23:20:19 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space In-Reply-To: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> References: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> Message-ID: <4A1C5D13.3040500@opencsw.org> Dagobert Michelsen wrote: > Hi, > > I am currently setting up build10st and need some more space for it and > noticed that Hudson is taking up around 25 GB of space. Looking into > /var/tmp is, umh, impressing... > >> root at ncsw [global]:/zones/hudson/root/var/tmp > time ls -1 | wc -l >> 824725 >> ls -1 8.70s user 4.23s system 18% cpu 1:08.11 total >> wc -l 0.32s user 0.02s system 0% cpu 1:07.86 total > > > Trygve, could you please set up some kind of cleanup job? I'm going to > delete the temp stuff for now manually. What kind of files are in /var/tmp? I would be surprised if hudson itself was leaking files there. -- Trygve From dam at opencsw.org Tue May 26 23:24:34 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 23:24:34 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space In-Reply-To: <4A1C5D13.3040500@opencsw.org> References: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> <4A1C5D13.3040500@opencsw.org> Message-ID: <92E1FF20-8F6C-4E70-9D1F-D33C315966AF@opencsw.org> Hi Trygve, Am 26.05.2009 um 23:20 schrieb Trygve Laugst?l: > What kind of files are in /var/tmp? I would be surprised if hudson > itself was leaking files there. Like this: root at ncsw [global]:/zones/hudson/root/var/tmp > find aaazgaqzC aaazgaqzC aaazgaqzC/CSWtomcat4 aaazgaqzC/CSWtomcat4/pkgmap aaazgaqzC/CSWtomcat4/pkginfo Best regards -- Dago From trygvis at opencsw.org Tue May 26 23:26:12 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Tue, 26 May 2009 23:26:12 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space In-Reply-To: <92E1FF20-8F6C-4E70-9D1F-D33C315966AF@opencsw.org> References: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> <4A1C5D13.3040500@opencsw.org> <92E1FF20-8F6C-4E70-9D1F-D33C315966AF@opencsw.org> Message-ID: <4A1C5E74.9040302@opencsw.org> Dagobert Michelsen wrote: > Hi Trygve, > > Am 26.05.2009 um 23:20 schrieb Trygve Laugst?l: >> What kind of files are in /var/tmp? I would be surprised if hudson >> itself was leaking files there. > > Like this: > > root at ncsw [global]:/zones/hudson/root/var/tmp > find aaazgaqzC > aaazgaqzC > aaazgaqzC/CSWtomcat4 > aaazgaqzC/CSWtomcat4/pkgmap > aaazgaqzC/CSWtomcat4/pkginfo Looks like something checkpkg would generate from what I can remember, but I might be wrong. -- Trygve From ihsan at opencsw.org Tue May 26 23:39:12 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 26 May 2009 23:39:12 +0200 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <20090526152746.GB17249@bolthole.com> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> <4A1BEEA6.70707@opencsw.org> <20090526152746.GB17249@bolthole.com> Message-ID: <4A1C6180.8020105@opencsw.org> Am 26.5.2009 17:27 Uhr, Philip Brown schrieb: >> I think a transition to OpenVAS would make sense then. > > can you confirm or deny whether openvas covers all the free-nessus > features? I can't. But if so, I think it's worse to think about. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From skayser at opencsw.org Tue May 26 23:49:17 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 26 May 2009 23:49:17 +0200 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Message-ID: <4A1C63DD.2080100@opencsw.org> Dagobert Michelsen wrote: > Am 26.05.2009 um 21:47 schrieb Sebastian Kayser: >> There is no machine on this net with internet access ... restricted >> environment. While i am still a bit reluctant to sync the whole >> mirror, >> there is a workaround using pkgutil. >> >> PKG_INSTALL_ROOT=/dev/null pkgutil -d > > Why don't you use this option? > > -s, --stream > Build a package stream for a certain architecture and > OS release. All dependencies will be included in the > stream and the needed command to install them in the > correct order will be displayed. > > I requested it some time ago for this exact prupose. Cool, thanks for the pointer. Will give it a try tomorrow. Sebastian From pfelecan at opencsw.org Wed May 27 11:06:26 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 27 May 2009 11:06:26 +0200 Subject: [csw-maintainers] New schily tools in testing In-Reply-To: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> (Joerg Schilling's message of "Fri\, 15 May 2009 19\:37\:44 +0200") References: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: > I just compiled a new set of packages and put them into testing. Thanks. Do you need more testing or are they ready to be released? It would be really nice to have up to date version of your tools in our catalog. -- Peter From pfelecan at opencsw.org Wed May 27 13:55:09 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 27 May 2009 13:55:09 +0200 Subject: [csw-maintainers] libgsf is or isn't orphaned Message-ID: Looking to libgsf package I see that Robin Kay is the maintainer and he's retired. However, libgsf is not marked as orphaned. What's the real status of this package? The packaged release is so old that any new dependant of which I'm aware cannot be linked with. -- Peter From dam at opencsw.org Wed May 27 14:06:18 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 14:06:18 +0200 Subject: [csw-maintainers] libgsf is or isn't orphaned In-Reply-To: References: Message-ID: <2A9DAF13-94AE-4FF7-8579-F2278293E4D2@opencsw.org> Hi Peter, Am 27.05.2009 um 13:55 schrieb Peter FELECAN: > Looking to libgsf package I see that Robin Kay is the maintainer and > he's retired. However, libgsf is not marked as orphaned. What's the > real > status of this package? The packaged release is so old that any new > dependant of which I'm aware cannot be linked with. Roger worked on it, but had troubles: Am 10.04.2009 um 02:19 schrieb Roger H?kansson: > Dagobert Michelsen wrote: >> I just build it from SVN without any problems, maybe the LD_OPTIONS >> change was >> more good than expected? >> Anyway, libgsf is now in testing: >> libgsf-1.14.11,REV=2009.04.09-SunOS5.8-i386-CSW.pkg.gz >> libgsf-1.14.11,REV=2009.04.09-SunOS5.8-sparc-CSW.pkg.gz > > Maybe I was a bit unclear, the package will build just fine, however > libgsf-gnome-1.so isn't linked to the libgsf-1.so that the build > produce, but the old one in /opt/csw/lib. > And I've been trying to fix this in various ways but with no luck. > So far the only working solution I've found is to remove libgsf from > the build machine when packaging. > > hson at build8x :~ > dump -Lv /home/dam/mgar/pkg/libgsf/trunk/work/ > pkgroot/opt/csw/lib/libgsf-gnome-1.so.114.0.11 > ... > [1] NEEDED libgsf-1.so.1 > ... > [16] SONAME libgsf-gnome-1.so.114 > > >>> I've also been trying to build libgda and libgnomedb (which are >>> optional for gnumeric), but there seems to be some incompatibility >>> from upstream so libgnomedb won't built atm. >> Nope, no incompatibility here. Builds absolutely smooth. Packages >> in testing/ now. >> Thanks for the good work! > > Hmm, I get a bunch of "undefined struct/union member" when building > libgnomedb on my build machine, but I've installed the libgda I > packaged, I guess the buildfarm have the older one? > Otherwise its probably some other package that is missing from my > build machine which is installed on the build farm... Feel free to give him a hand and take a look also. Best regards -- Dago From maciej at opencsw.org Wed May 27 16:55:10 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 27 May 2009 15:55:10 +0100 Subject: [csw-maintainers] findutils and its database on the farm Message-ID: Would it be possible to get findutils (glocate) installed on the farm and have the file database up to date? The current state is: build8s -- no findutils build8x -- no findutils build10s -- findutils installed, database missing (need to run gupdatedb?) build10x -- no findutils Maciej From phil at bolthole.com Wed May 27 18:38:08 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 09:38:08 -0700 Subject: [csw-maintainers] libgsf is or isn't orphaned In-Reply-To: References: Message-ID: <20090527163808.GH55163@bolthole.com> On Wed, May 27, 2009 at 01:55:09PM +0200, Peter FELECAN wrote: > Looking to libgsf package I see that Robin Kay is the maintainer and > he's retired. However, libgsf is not marked as orphaned. What's the real > status of this package? The packaged release is so old that any new > dependant of which I'm aware cannot be linked with. Any package whose maintainer is marked as "retired", is open to adoption by another maintainer. From mwatters at opencsw.org Wed May 27 19:01:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 12:01:51 -0500 Subject: [csw-maintainers] new class scripts Message-ID: <4A1D71FF.9080806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I have a new set of class scripts i.pycompile and r.pycompile I was wondering how best to install them. add them to cswclassutils, have the python package add/delete them to/from /usr/sadm/install/scripts ? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodcf8ACgkQLrhmsXMSLxd1twCeMlq/8ayvma1y+3Sje4MOVrfU 9asAoKH4KfEsDbJFWO3uDQahOZcSZ7j0 =xy43 -----END PGP SIGNATURE----- From bwalton at opencsw.org Wed May 27 19:10:22 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 27 May 2009 13:10:22 -0400 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1D71FF.9080806@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> Message-ID: <1243444160-sup-7077@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Wed May 27 13:01:51 -0400 2009: > I have a new set of class scripts i.pycompile and r.pycompile I > was wondering how best to install them. add them to cswclassutils, > have the python package add/delete them to/from > /usr/sadm/install/scripts ? I think that for consistency, they should be added to cswclassutils. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 27 19:16:04 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 10:16:04 -0700 Subject: [csw-maintainers] new class scripts In-Reply-To: <1243444160-sup-7077@ntdws12.chass.utoronto.ca> References: <4A1D71FF.9080806@opencsw.org> <1243444160-sup-7077@ntdws12.chass.utoronto.ca> Message-ID: <20090527171604.GP55163@bolthole.com> On Wed, May 27, 2009 at 01:10:22PM -0400, Ben Walton wrote: > Excerpts from Mike Watters's message of Wed May 27 13:01:51 -0400 2009: > > I have a new set of class scripts i.pycompile and r.pycompile I > > was wondering how best to install them. add them to cswclassutils, > > have the python package add/delete them to/from > > /usr/sadm/install/scripts ? > > I think that for consistency, they should be added to cswclassutils. yup. From bonivart at opencsw.org Wed May 27 19:31:53 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 27 May 2009 19:31:53 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1D71FF.9080806@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> Message-ID: <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> On Wed, May 27, 2009 at 7:01 PM, Mike Watters wrote: > ? ? ?I have a new set of class scripts i.pycompile and r.pycompile > I was wondering how best to install them. ?add them to cswclassutils, have the > python package add/delete them to/from /usr/sadm/install/scripts ? Make them available to me somehow and I will add them to cswclassutils. I will just rename them to i|r.cswpycompile for consistency. -- /peter From mwatters at opencsw.org Wed May 27 19:50:38 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 12:50:38 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> Message-ID: <4A1D7D6E.7000709@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Bonivart wrote: > On Wed, May 27, 2009 at 7:01 PM, Mike Watters wrote: >> I have a new set of class scripts i.pycompile and r.pycompile >> I was wondering how best to install them. add them to cswclassutils, have the >> python package add/delete them to/from /usr/sadm/install/scripts ? > > Make them available to me somehow and I will add them to > cswclassutils. I will just rename them to i|r.cswpycompile for > consistency. > They are checked into subversion under the python package http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile Addition to the README.CSW =========================== - - pycompile This class takes care of creating/removing python's normal and optimized bytecode files. Set all *.py files' class = pycompile to activate. Example: # use prototype filters to set the class PROTOTYPE_FILTER = awk '$$$$3 ~/.*\.py$$$$/ { $$$$2 = "pycompile" } { print }' - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodfW4ACgkQLrhmsXMSLxcK9wCgkC1CufESdEt+S+mhbLohh8vN JIIAnjJ0b4VOlKkR9ymSKdcCQ9EAv+It =S3cZ -----END PGP SIGNATURE----- From bonivart at opencsw.org Wed May 27 20:01:09 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 27 May 2009 20:01:09 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1D7D6E.7000709@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> Message-ID: <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> On Wed, May 27, 2009 at 7:50 PM, Mike Watters wrote: > They are checked into subversion under the python package > > http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile > http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile > > Addition to the README.CSW > =========================== > > - - pycompile > ?This class takes care of creating/removing python's normal and optimized > ?bytecode files. ?Set all *.py files' class = pycompile to activate. > Example: > # use prototype filters to set the class > PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "pycompile" } { print }' I'll add them tomorrow, thanks for the readme text. I updated the wiki page with it to start with, feel free to add/edit. http://wiki.opencsw.org/cswclassutils-package -- /peter From maciej at opencsw.org Wed May 27 20:13:55 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 27 May 2009 19:13:55 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package Message-ID: I'm playing with updating syslog_ng to v2 and getting it to have SMF support via mGAR. Makefile: http://dpaste.com/48317/ SMF manifest: http://dpaste.com/48320/ The changes I made can be summarized as: - added DISTFILES += cswsyslog_ng.xml - installed cswsyslog_ng as svc-cswsyslog_ng in the manifest directory ($(DESTDIR)$(libdir)/svc/method) - added INITSMF = /opt/csw/etc/init.d/cswsyslog_ng The result is that the SMF manifest doesn't get imported: Installation of on zone was successful. root at joy:/var/tmp# svcs -a | grep syslog root at joy:/var/tmp# How does mGAR know which file is the manifest? Just by its name -- cswfoo.xml? Maciej From phil at bolthole.com Wed May 27 20:47:57 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 11:47:57 -0700 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: References: Message-ID: <20090527184757.GU55163@bolthole.com> On Wed, May 27, 2009 at 07:13:55PM +0100, Maciej (Matchek) Blizinski wrote: > I'm playing with updating syslog_ng to v2 and getting it to have SMF > support via mGAR. > .... > The result is that the SMF manifest doesn't get imported: better to use cswclassutils for SMF support anyways, rather than a custom pkg-local manifest, unless you're doing something Really Special. From bonivart at opencsw.org Wed May 27 21:17:03 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 27 May 2009 21:17:03 +0200 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <20090527184757.GU55163@bolthole.com> References: <20090527184757.GU55163@bolthole.com> Message-ID: <625385e30905271217q4a62178aq118e396fbfe34045@mail.gmail.com> On Wed, May 27, 2009 at 8:47 PM, Philip Brown wrote: > better to use cswclassutils for SMF support anyways, rather than a custom > pkg-local manifest, unless you're doing something Really Special. I'm looking into supporting custom manifests so that should be covered soon. -- /peter From phil at bolthole.com Wed May 27 21:24:41 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 12:24:41 -0700 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <625385e30905271217q4a62178aq118e396fbfe34045@mail.gmail.com> References: <20090527184757.GU55163@bolthole.com> <625385e30905271217q4a62178aq118e396fbfe34045@mail.gmail.com> Message-ID: <20090527192441.GV55163@bolthole.com> On Wed, May 27, 2009 at 09:17:03PM +0200, Peter Bonivart wrote: > On Wed, May 27, 2009 at 8:47 PM, Philip Brown wrote: > > better to use cswclassutils for SMF support anyways, rather than a custom > > pkg-local manifest, unless you're doing something Really Special. > > I'm looking into supporting custom manifests so that should be covered soon. coool. i would suggest just another commented-variable in the init.d file. ie: maintainer provides [/etc/opt/csw/]init.d/cswmyservice [/etc/opt/csw/]/cswmyservice.xml and uses # CUSTOMMANIFEST=/path/to/cswmyservice.xml in the first file, which woudl be the only one in class cswinitsmf From Joerg.Schilling at fokus.fraunhofer.de Wed May 27 22:17:13 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 27 May 2009 22:17:13 +0200 Subject: [csw-maintainers] New schily tools in testing In-Reply-To: References: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <4a1d9fc9.TCJ3FSsGMtwXUKLX%Joerg.Schilling@fokus.fraunhofer.de> Peter FELECAN wrote: > Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: > > > I just compiled a new set of packages and put them into testing. > > Thanks. Do you need more testing or are they ready to be released? It > would be really nice to have up to date version of your tools in our > catalog. This is ready to be released. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From ellson at opencsw.org Wed May 27 20:25:50 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 27 May 2009 14:25:50 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker Message-ID: <4A1D85AE.4060006@opencsw.org> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and links without problem. If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" from the Makefile) then I get this error: /opt/studio/SOS11/SUNWspro/bin/CC -KPIC CMakeFiles/example0.dir/MissingGlyphExample.o -o example0 -L/opt/csw/lib -L/home/ellson/mgar/pkg/liblasi/trunk/work/build-isa-sparcv8/libLASi-1.1.0/src -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lpangoft2-1.0 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lfreetype -lz ../src/libLASi.so.0.0.1 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lpangoft2-1.0 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lpangoft2-1.0 -lfreetype -lz -lm -R/opt/csw/lib:/home/ellson/mgar/pkg/liblasi/trunk/work/build-isa-sparcv8/libLASi-1.1.0/src Undefined first referenced symbol in file void LASi::PostscriptDocument::setFont(const char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::FontStretch) CMakeFiles/example0.dir/MissingGlyphExample.o ld: fatal: Symbol referencing errors. No output written to example0 There's something odd about the "const char*const" but I don't know if thats related. Does anyone have any idea what the problem is, and how to fix it, or work around it so as to be able to use Sun's tools? John From dam at opencsw.org Wed May 27 23:12:33 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:12:33 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> Message-ID: <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> Hi, Am 27.05.2009 um 20:01 schrieb Ben Walton: > Excerpts from Maciej (Matchek) Blizinski's message of Wed May 27 > 10:55:10 -0400 2009: >> Would it be possible to get findutils (glocate) installed on the farm >> and have the file database up to date? > > Doing this now. Is this package current in terms of updating cron, etc.? Additionally, it has security issues. Chris: As you already did findutils with glocate, it would be really cool if you could also package up slocate as more secure alternative. Best regards -- Dago From william at wbonnet.net Wed May 27 23:22:57 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 27 May 2009 23:22:57 +0200 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1D85AE.4060006@opencsw.org> References: <4A1D85AE.4060006@opencsw.org> Message-ID: <4A1DAF31.7040109@wbonnet.net> Hi John > If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles > and links without problem. Well... It seems your Makefile description does not access a valid SF url :( I have the following error ==> Grabbing download/libLASi-1.1.0.tar.gz ==> Trying file//files/libLASi-1.1.0.tar.gz gmake[1]: *** [file//files/libLASi-1.1.0.tar.gz] Erreur 255 ==> Trying file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz gmake[1]: *** [file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz] Erreur 255 ==> Trying http//downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz --2009-05-27 23:01:26-- http://downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz Resolving downloads.sourceforge.net... 216.34.181.59 Connecting to downloads.sourceforge.net|216.34.181.59|:80... connected. HTTP request sent, awaiting response... 301 Moved Permanently Location: http://sourceforge.net/projects/liblasi/files [following] --2009-05-27 23:01:26-- http://sourceforge.net/projects/liblasi/files Resolving sourceforge.net... 216.34.181.60 Connecting to sourceforge.net|216.34.181.60|:80... connected. HTTP request sent, awaiting response... 200 OK Length: unspecified [text/html] Saving to: `download/partial/files' 100%[========================================================================================================>] 11,515 43.6K/s in 0.3s 2009-05-27 23:01:27 (43.6 KB/s) - `download/partial/files' saved [11515] mv : impossible d'acc?der ? download/partial/libLASi-1.1.0.tar.gz (!!!) Failed to download download/libLASi-1.1.0.tar.gz! gmake: *** [download/libLASi-1.1.0.tar.gz] Erreur 255 I checked for project number and it is valid, but i cannot download sources, either on build8x nor on my box at home cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Wed May 27 23:25:16 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:25:16 +0200 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? References: <20090527171435.GN55163@bolthole.com> Message-ID: Hi, Anfang der weitergeleiteten E-Mail: > Von: Philip Brown > On Tue, May 26, 2009 at 09:28:58PM +0200, Dagobert Michelsen wrote: >> in Newpkgs is an updated bdb44 with 64 bit libs. I need it for >> OpenLDAP and comparison against bdb47 which comes next: > > why do we need both 44 and 47? > > this sounds kinda bad to me. > I vaguely recall that they DID actually build in some kind of > backwards > compat. > and even our current berkeleydbXX packages may be redundant to some > degree. I don't know. Maybe someone has an advice on how to proceed with this. The GAR recipe for bdb44 has been updated with 64 bit libs, but is this version still needed? There are quite some dependencies listed: This includes subversion and perl, both of which should get updates to 64 bit long term (svn 64 bit is needed for files > 2 GB. I don't want to discuss if this is a good idea, but we have a customer that needed it and where I hand-build it). Or should be officially drop 3.3, 4.2, 4.3 and 4.4 and just build 4.7? Best regards -- Dago From dam at opencsw.org Wed May 27 23:28:07 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:28:07 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> Message-ID: <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> Hi folks, Am 27.05.2009 um 20:01 schrieb Peter Bonivart: > On Wed, May 27, 2009 at 7:50 PM, Mike Watters > wrote: >> They are checked into subversion under the python package >> >> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile >> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile >> >> Addition to the README.CSW >> =========================== >> >> - - pycompile >> This class takes care of creating/removing python's normal and >> optimized >> bytecode files. Set all *.py files' class = pycompile to activate. >> Example: >> # use prototype filters to set the class >> PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "pycompile" } >> { print }' > > I'll add them tomorrow, thanks for the readme text. I updated the wiki > page with it to start with, feel free to add/edit. > > http://wiki.opencsw.org/cswclassutils-package Cool stuff. Just in time when I want to release xcbproto :-) Best regards -- Dago From mwatters at opencsw.org Wed May 27 23:33:23 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 16:33:23 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> Message-ID: <4A1DB1A3.8030205@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi folks, > > Am 27.05.2009 um 20:01 schrieb Peter Bonivart: >> On Wed, May 27, 2009 at 7:50 PM, Mike Watters >> wrote: >>> They are checked into subversion under the python package >>> >>> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile >>> >>> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile >>> >>> >>> Addition to the README.CSW >>> =========================== >>> >>> - - pycompile >>> This class takes care of creating/removing python's normal and >>> optimized >>> bytecode files. Set all *.py files' class = pycompile to activate. >>> Example: >>> # use prototype filters to set the class >>> PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "pycompile" } { print }' >> >> I'll add them tomorrow, thanks for the readme text. I updated the wiki >> page with it to start with, feel free to add/edit. >> >> http://wiki.opencsw.org/cswclassutils-package > > Cool stuff. Just in time when I want to release xcbproto :-) > > > Best regards > > -- Dago > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Dago, if you give it a couple days, I will have the python 2.6.2 version released. I am just waiting on the classutils. I have tested and retested python with this recipe, and the utils. I just want one clean test with the classutils package rather then me just copying the class scripts in place. all the python stuff will have a bug once the new packages are released. I split python into core, runtime, devel, idle and tk. dependencies should be linked to python_rt if linking against libpython.so.X - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodsaIACgkQLrhmsXMSLxc0mACgqbe0k8AN3dQvsDhyGrJeM2zj o9cAoNDFwtHP4TpBw5f0ZnVTdu7fMhyv =nu5A -----END PGP SIGNATURE----- From ellson at opencsw.org Wed May 27 23:46:02 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 27 May 2009 17:46:02 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1DAF31.7040109@wbonnet.net> References: <4A1D85AE.4060006@opencsw.org> <4A1DAF31.7040109@wbonnet.net> Message-ID: <4A1DB49A.8000503@opencsw.org> Fixed. (SF_PROJ != GARNAME != tar flile name != library name ...... grumble mumble ) John On 05/27/2009 05:22 PM, William Bonnet wrote: > Hi John >> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles >> and links without problem. > > Well... It seems your Makefile description does not access a valid SF > url :( > > I have the following error > > ==> Grabbing download/libLASi-1.1.0.tar.gz > ==> Trying file//files/libLASi-1.1.0.tar.gz > gmake[1]: *** [file//files/libLASi-1.1.0.tar.gz] Erreur 255 > ==> Trying > file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz > gmake[1]: *** > [file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz] > Erreur 255 > ==> Trying > http//downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz > --2009-05-27 23:01:26-- > http://downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz > Resolving downloads.sourceforge.net... 216.34.181.59 > Connecting to downloads.sourceforge.net|216.34.181.59|:80... connected. > HTTP request sent, awaiting response... 301 Moved Permanently > Location: http://sourceforge.net/projects/liblasi/files [following] > --2009-05-27 23:01:26-- http://sourceforge.net/projects/liblasi/files > Resolving sourceforge.net... 216.34.181.60 > Connecting to sourceforge.net|216.34.181.60|:80... connected. > HTTP request sent, awaiting response... 200 OK > Length: unspecified [text/html] > Saving to: `download/partial/files' > > 100%[========================================================================================================>] > 11,515 43.6K/s in 0.3s > > 2009-05-27 23:01:27 (43.6 KB/s) - `download/partial/files' saved [11515] > > mv : impossible d'acc?der ? download/partial/libLASi-1.1.0.tar.gz > (!!!) Failed to download download/libLASi-1.1.0.tar.gz! > gmake: *** [download/libLASi-1.1.0.tar.gz] Erreur 255 > > I checked for project number and it is valid, but i cannot download > sources, either on build8x nor on my box at home > > cheers > W. > From phil at bolthole.com Wed May 27 23:47:41 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 14:47:41 -0700 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: References: <20090527171435.GN55163@bolthole.com> Message-ID: <20090527214741.GB55163@bolthole.com> On Wed, May 27, 2009 at 11:25:16PM +0200, Dagobert Michelsen wrote: >... > Or should be officially drop 3.3, 4.2, 4.3 and 4.4 and just build 4.7? (and if so, we should probably rename/create new package, "berkeleydb", and have everything else depend on its rt package) From dam at opencsw.org Wed May 27 23:48:14 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:48:14 +0200 Subject: [csw-maintainers] New schily tools in testing In-Reply-To: <4a1d9fc9.TCJ3FSsGMtwXUKLX%Joerg.Schilling@fokus.fraunhofer.de> References: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> <4a1d9fc9.TCJ3FSsGMtwXUKLX%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <98862867-B256-45AD-B7F6-8860DF9A6296@opencsw.org> Hi Joerg, Am 27.05.2009 um 22:17 schrieb Joerg Schilling: > Peter FELECAN wrote: >> Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: >> >>> I just compiled a new set of packages and put them into testing. >> >> Thanks. Do you need more testing or are they ready to be released? It >> would be really nice to have up to date version of your tools in our >> catalog. > > This is ready to be released. Ok then, just scp them over to www.opencsw.org:/home/newpkgs I cc'ed Phil so they should get pushed ASAP. Best regards -- Dago From mwatters at opencsw.org Wed May 27 23:55:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 16:55:22 -0500 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: <20090527214741.GB55163@bolthole.com> References: <20090527171435.GN55163@bolthole.com> <20090527214741.GB55163@bolthole.com> Message-ID: <4A1DB6CA.6070304@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Wed, May 27, 2009 at 11:25:16PM +0200, Dagobert Michelsen wrote: >> ... >> Or should be officially drop 3.3, 4.2, 4.3 and 4.4 and just build 4.7? > > (and if so, we should probably rename/create new package, > "berkeleydb", and have everything else depend on its rt package) I say just build 4.7 but keep libdb4-4 as deprecated libs in 4.7. rename the package to berkeleydb and have core,runtime and devel just my 2 cents ( or 1/1000th of a euro, the way the US market is going ) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodtsoACgkQLrhmsXMSLxcXbQCgvPYHw6mdNVAGL6yveCMYJGLM JkYAniTTtKmnfQchif6e7DwtLWsVgMua =pIUO -----END PGP SIGNATURE----- From dam at opencsw.org Thu May 28 00:00:42 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 28 May 2009 00:00:42 +0200 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: <4A1DB6CA.6070304@opencsw.org> References: <20090527171435.GN55163@bolthole.com> <20090527214741.GB55163@bolthole.com> <4A1DB6CA.6070304@opencsw.org> Message-ID: Hi, Am 27.05.2009 um 23:55 schrieb Mike Watters: > I say just build 4.7 but keep libdb4-4 as deprecated libs in 4.7. > rename the package to berkeleydb and have core,runtime and devel As CSWbdb would be a new package there is no point in putting in old libraries in there as the existing packages depend on CSWbdb44. I guess we must recompile the dependent packages to make use of the new version. BTW: bdb47 et al. is now in testing/ Best regards -- Dago From car at opencsw.org Thu May 28 00:58:35 2009 From: car at opencsw.org (Chris Reece) Date: Thu, 28 May 2009 10:58:35 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> Message-ID: <50797.131.203.105.13.1243465115.squirrel@webmail.jessies.org> On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: > Chris: As you already did findutils with glocate, it would be > really cool if you could also package up slocate > as more secure alternative. Thus findutils nominates itself as my first GAR package. I'll put aside a few hours this weekend. I've recently moved jobs and my new position involves no Solaris. My intention is to GAR up the few packages I currently own and them throw them back into the pool, as it were, before resigning my maintainer status. From bonivart at opencsw.org Thu May 28 09:39:08 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 09:39:08 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1DB1A3.8030205@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> Message-ID: <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> On Wed, May 27, 2009 at 11:33 PM, Mike Watters wrote: > I am just waiting on the classutils. ?I have tested and retested python with > this recipe, and the utils. ?I just want one clean test with the classutils > package rather then me just copying the class scripts in place. I have dropped a new cswclassutils package into testing that includes cswpycompile: http://mirror.opencsw.org/testing/cswclassutils-1.10,REV=2009.05.28-SunOS5.8-all-CSW.pkg.gz -- /peter From james at opencsw.org Thu May 28 11:09:54 2009 From: james at opencsw.org (James Lee) Date: Thu, 28 May 2009 09:09:54 GMT Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1D85AE.4060006@opencsw.org> References: <4A1D85AE.4060006@opencsw.org> Message-ID: <20090528.9095400.3538465159@gyor.oxdrove.co.uk> On 27/05/09, 19:25:50, John Ellson wrote regarding [csw-maintainers] need help with C++ linker error from Sun's linker: > If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and > links without problem. > If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" > from the Makefile) > then I get this error: ... > Undefined first referenced > symbol in file > void LASi::PostscriptDocument::setFont(const > char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::Font St retch) > CMakeFiles/example0.dir/MissingGlyphExample.o > ld: fatal: Symbol referencing errors. No output written to example0 > There's something odd about the "const char*const" but I don't know if > thats related. I think you've almost answered your own question. gcc does not use "const" type qualifier for types passed by value when making external names. So "const char*const" is the same as "const char*" on the call because the value is copied anyway (the second "const" is of benefit to the compiler in the function only). Nevertheless CC likes consistency so try matching the "const"s in all function declarations. $ cat sub.cc void sub(const char*const arg) { return; } $ g++ -c sub.cc $ nm -C sub.o sub.o: [Index] Value Size Type Bind Other Shndx Name [2] | 0| 0|SECT |LOCL |0 |1 | [3] | 0| 0|SECT |LOCL |0 |2 | [4] | 0| 0|SECT |LOCL |0 |3 | [5] | 0| 0|SECT |LOCL |0 |4 | [6] | 0| 16|FUNC |GLOB |0 |1 |sub(const char*) [_Z3subPKc] [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc $ CC -c sub.cc $ nm -C sub.o sub.o: [Index] Value Size Type Bind Other Shndx Name [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc [2] | 16| 24|FUNC |GLOB |0 |2 |void sub(const char*const) [__1cDsub6Fkpkc_v_] James. From maciej at opencsw.org Thu May 28 11:28:29 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 28 May 2009 10:28:29 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <20090527184757.GU55163@bolthole.com> References: <20090527184757.GU55163@bolthole.com> Message-ID: On Wed, May 27, 2009 at 7:47 PM, Philip Brown wrote: > On Wed, May 27, 2009 at 07:13:55PM +0100, Maciej (Matchek) Blizinski wrote: >> I'm playing with updating syslog_ng to v2 and getting it to have SMF >> support via mGAR. >> .... >> The result is that the SMF manifest doesn't get imported: > > better to use cswclassutils for SMF support anyways, rather than a custom > pkg-local manifest, unless you're doing something Really Special. I'm trying to be as far away from Anything Special as possible. I'm following the common documentation of SMF support: https://apps.sourceforge.net/trac/gar/wiki/SMF/Init%20script%20support%20in%20your%20package A quote: 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - First step, we must have the following files ready in our files/ directory: * the legacy init script, cswfoo.init * the SMF manifest, cswfoo.xml * the SMF service startup script, svc-cswfoo 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - The second bullet list item explicitly tells to create cswfoo.xml. Should I be skeptical about it? For the purposes of testing, I've created 'minimalsmf' package. The idea is to have a minimal amount of code which builds a package with a sleeping shell script. The main Makefile is: http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/minimalsmf/trunk/Makefile Dago, could you please look at this package and remove anything unnecessary? Maciej From phil at bolthole.com Thu May 28 15:03:28 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 28 May 2009 06:03:28 -0700 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: References: <20090527184757.GU55163@bolthole.com> Message-ID: <20090528130328.GA91905@bolthole.com> On Thu, May 28, 2009 at 10:28:29AM +0100, Maciej (Matchek) Blizinski wrote: > I'm trying to be as far away from Anything Special as possible. I'm > following the common documentation of SMF support: > https://apps.sourceforge.net/trac/gar/wiki/SMF/Init%20script%20support%20in%20your%20package > yeah, as with most of the gar wiki documentation, that is out of date :-/ > 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > - - - - - - - - - - - - > First step, we must have the following files ready in our files/ directory: > > * the legacy init script, cswfoo.init > * the SMF manifest, cswfoo.xml > * the SMF service startup script, svc-cswfoo > 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > - - - - - - - - - - - - > From phil at bolthole.com Thu May 28 15:06:39 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 28 May 2009 06:06:39 -0700 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: References: <20090527171435.GN55163@bolthole.com> <20090527214741.GB55163@bolthole.com> <4A1DB6CA.6070304@opencsw.org> Message-ID: <20090528130639.GB91905@bolthole.com> On Thu, May 28, 2009 at 12:00:42AM +0200, Dagobert Michelsen wrote: > Hi, > > Am 27.05.2009 um 23:55 schrieb Mike Watters: >> I say just build 4.7 but keep libdb4-4 as deprecated libs in 4.7. >> rename the package to berkeleydb and have core,runtime and devel > > As CSWbdb would be a new package there is no point in putting > in old libraries in there as the existing packages depend on > CSWbdb44. I guess we must recompile the dependent packages to > make use of the new version. Note: while there would be no point copying over binary libs from the older packages... it would be important to enable compiling "older libs" in the configuration for the compile for bdb47. I think (but could be wrong) that the berkeleydb config stuf has options for building older versions of libs. In which case, i guess we would want to then enable them for "CSWbdb", and conflict with the older packages, and replace those with empty packages or something. Hmm... upgrade path for users might be tricky... From Darin.Perusich at cognigencorp.com Thu May 28 15:21:25 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 28 May 2009 09:21:25 -0400 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup Message-ID: <4A1E8FD5.9000003@cognigencorp.com> I was reviewing the classutils documentation on setting file/dir ownership and I'm not clear on how this is accomplished. The documentation talks about assigning a new class, called 'ugfiles', and assigning it to the files/dirs. Is this 'ugfiles' just an arbitrary name or is it something associated with the cswusergroup class? Maybe I'm just being thickheaded but I am not finding this part of documentation confusing and unclear. Thanks! -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From bonivart at opencsw.org Thu May 28 15:43:17 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 15:43:17 +0200 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup In-Reply-To: <4A1E8FD5.9000003@cognigencorp.com> References: <4A1E8FD5.9000003@cognigencorp.com> Message-ID: <625385e30905280643v36b1b7f4v37396be12864fd5f@mail.gmail.com> On Thu, May 28, 2009 at 3:21 PM, Darin Perusich wrote: > I was reviewing the classutils documentation on setting file/dir > ownership and I'm not clear on how this is accomplished. > > The documentation talks about assigning a new class, called 'ugfiles', > and assigning it to the files/dirs. Is this 'ugfiles' just an arbitrary > ?name or is it something associated with the cswusergroup class? This step is only needed if you want files/dirs owned by the new user. If you only needed a user to run a service you can skip it. An example is my ClamAV package which has a service that runs with the clamav user and the signature databases needs to be owned by this user. Assuming you still want this, the problem is that the "none" class is always executed first so if your files are in this class the new user will not exist yet, ok? That's why we need an extra class for these files so we can tell pkgadd to run this class after cswusergroup, like this: CLASSES = none cswusergroup ugfiles This way the user is created by cswusergroup before files owned by this user are processed. > Maybe I'm just being thickheaded but I am not finding this part of > documentation confusing and unclear. I hope I just explained it better. Any way to improve the documentation? -- /peter From bonivart at opencsw.org Thu May 28 15:45:15 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 15:45:15 +0200 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup In-Reply-To: <4A1E8FD5.9000003@cognigencorp.com> References: <4A1E8FD5.9000003@cognigencorp.com> Message-ID: <625385e30905280645w2ae74aacme89262d5d79943db@mail.gmail.com> On Thu, May 28, 2009 at 3:21 PM, Darin Perusich wrote: > Is this 'ugfiles' just an arbitrary > ?name or is it something associated with the cswusergroup class? Sorry, missed this one, the name doesn't matter. -- /peter From ellson at opencsw.org Thu May 28 15:59:15 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 28 May 2009 09:59:15 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <20090528.9095400.3538465159@gyor.oxdrove.co.uk> References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> Message-ID: <4A1E98B3.7020508@opencsw.org> James, Thanks. I think I understand the issue, but I'm still not sure how to fix it? I'm not that familiar with C++, but I don't see any mismatched definitions in the code. Are you recommending that all "const char* const ... " in the sources be changed to "const char* ... " ? I'm a bit reluctant to make such an extensive change to the API of this library. Is there a linker flag that would tell it to tolerate these extra const's ? John James Lee wrote: > On 27/05/09, 19:25:50, John Ellson wrote regarding > [csw-maintainers] need help with C++ linker error from Sun's linker: > > >> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and >> links without problem. >> > > >> If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" >> from the Makefile) >> then I get this error: >> > > ... > > >> Undefined first referenced >> symbol in file >> void LASi::PostscriptDocument::setFont(const >> >> > char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::Font > St > retch) > >> CMakeFiles/example0.dir/MissingGlyphExample.o >> ld: fatal: Symbol referencing errors. No output written to example0 >> > > > > >> There's something odd about the "const char*const" but I don't know if >> thats related. >> > > > I think you've almost answered your own question. gcc does not use > "const" type qualifier for types passed by value when making external > names. So "const char*const" is the same as "const char*" on the call > because the value is copied anyway (the second "const" is of benefit to > the compiler in the function only). Nevertheless CC likes consistency > so try matching the "const"s in all function declarations. > > > > > $ cat sub.cc > void sub(const char*const arg) > { > return; > } > > $ g++ -c sub.cc > $ nm -C sub.o > > > sub.o: > > [Index] Value Size Type Bind Other Shndx Name > > [2] | 0| 0|SECT |LOCL |0 |1 | > [3] | 0| 0|SECT |LOCL |0 |2 | > [4] | 0| 0|SECT |LOCL |0 |3 | > [5] | 0| 0|SECT |LOCL |0 |4 | > [6] | 0| 16|FUNC |GLOB |0 |1 |sub(const char*) > [_Z3subPKc] > [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc > > $ CC -c sub.cc > $ nm -C sub.o > > > sub.o: > > [Index] Value Size Type Bind Other Shndx Name > > [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc > [2] | 16| 24|FUNC |GLOB |0 |2 |void sub(const > char*const) > > [__1cDsub6Fkpkc_v_] > > > > > James. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > > From ellson at opencsw.org Thu May 28 16:33:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 28 May 2009 10:33:06 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1E98B3.7020508@opencsw.org> References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: <4A1EA0A2.7060105@opencsw.org> Never mind. I've found the problem. It wasn't the "const char*const ...", it was all the other args. John John Ellson wrote: > James, > > Thanks. I think I understand the issue, but I'm still not sure how to > fix it? > > I'm not that familiar with C++, but I don't see any mismatched > definitions in the code. > > Are you recommending that all "const char* const ... " in the sources > be changed to > "const char* ... " ? > I'm a bit reluctant to make such an extensive change to the API of > this library. > Is there a linker flag that would tell it to tolerate these extra > const's ? > > John > > > > James Lee wrote: >> On 27/05/09, 19:25:50, John Ellson wrote regarding >> [csw-maintainers] need help with C++ linker error from Sun's linker: >> >> >>> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and >>> links without problem. >>> >> >> >>> If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" >>> from the Makefile) >>> then I get this error: >>> >> >> ... >> >> >>> Undefined first referenced >>> symbol in file >>> void LASi::PostscriptDocument::setFont(const >>> >>> >> char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::Font >> St >> retch) >> >>> CMakeFiles/example0.dir/MissingGlyphExample.o >>> ld: fatal: Symbol referencing errors. No output written to example0 >>> >> >> >> >> >>> There's something odd about the "const char*const" but I don't >>> know if >>> thats related. >>> >> >> >> I think you've almost answered your own question. gcc does not use >> "const" type qualifier for types passed by value when making external >> names. So "const char*const" is the same as "const char*" on the call >> because the value is copied anyway (the second "const" is of benefit to >> the compiler in the function only). Nevertheless CC likes consistency >> so try matching the "const"s in all function declarations. >> >> >> >> >> $ cat sub.cc >> void sub(const char*const arg) >> { >> return; >> } >> >> $ g++ -c sub.cc >> $ nm -C sub.o >> >> >> sub.o: >> >> [Index] Value Size Type Bind Other Shndx Name >> >> [2] | 0| 0|SECT |LOCL |0 |1 | >> [3] | 0| 0|SECT |LOCL |0 |2 | >> [4] | 0| 0|SECT |LOCL |0 |3 | >> [5] | 0| 0|SECT |LOCL |0 |4 | >> [6] | 0| 16|FUNC |GLOB |0 |1 |sub(const char*) >> [_Z3subPKc] >> [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc >> >> $ CC -c sub.cc >> $ nm -C sub.o >> >> >> sub.o: >> >> [Index] Value Size Type Bind Other Shndx Name >> >> [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc >> [2] | 16| 24|FUNC |GLOB |0 |2 |void sub(const >> char*const) >> >> [__1cDsub6Fkpkc_v_] >> >> >> >> >> James. >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers >> >> > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > From Darin.Perusich at cognigencorp.com Thu May 28 16:39:31 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 28 May 2009 10:39:31 -0400 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup In-Reply-To: <625385e30905280643v36b1b7f4v37396be12864fd5f@mail.gmail.com> References: <4A1E8FD5.9000003@cognigencorp.com> <625385e30905280643v36b1b7f4v37396be12864fd5f@mail.gmail.com> Message-ID: <4A1EA223.4070902@cognigencorp.com> Peter Bonivart wrote: > On Thu, May 28, 2009 at 3:21 PM, Darin Perusich > wrote: >> I was reviewing the classutils documentation on setting file/dir >> ownership and I'm not clear on how this is accomplished. >> >> The documentation talks about assigning a new class, called 'ugfiles', >> and assigning it to the files/dirs. Is this 'ugfiles' just an arbitrary >> name or is it something associated with the cswusergroup class? > > This step is only needed if you want files/dirs owned by the new user. > If you only needed a user to run a service you can skip it. An example > is my ClamAV package which has a service that runs with the clamav > user and the signature databases needs to be owned by this user. This is exactly what I need. The amanda package has a ton of files and need to be owned by the amanda user. > Assuming you still want this, the problem is that the "none" class is > always executed first so if your files are in this class the new user > will not exist yet, ok? That's why we need an extra class for these > files so we can tell pkgadd to run this class after cswusergroup, like > this: > > CLASSES = none cswusergroup ugfiles > > This way the user is created by cswusergroup before files owned by > this user are processed. > >> Maybe I'm just being thickheaded but I am not finding this part of >> documentation confusing and unclear. > > I hope I just explained it better. Any way to improve the documentation? > Your explanation has cleared this up and is what I was thinking but it just wasn't coming across in the documentation. I think the docs are to terse and need a little more explanation so the ideas are properly conveyed. I'm going to think about this for a bit and I'll write some updates out and send them over to you. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From maciej at opencsw.org Thu May 28 17:26:49 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 28 May 2009 16:26:49 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <20090528130328.GA91905@bolthole.com> References: <20090527184757.GU55163@bolthole.com> <20090528130328.GA91905@bolthole.com> Message-ID: On Thu, May 28, 2009 at 2:03 PM, Philip Brown wrote: > On Thu, May 28, 2009 at 10:28:29AM +0100, Maciej (Matchek) Blizinski wrote: >> I'm trying to be as far away from Anything Special as possible. I'm >> following the common documentation of SMF support: >> https://apps.sourceforge.net/trac/gar/wiki/SMF/Init%20script%20support%20in%20your%20package >> > > > yeah, as with most of the gar wiki documentation, that is out of date :-/ Okay, so this page is obsolete. I added a banner with a link to cswclassutils wiki page. I've been pointed at it by Peter. It might be also out of date, hopefully not as much as the first one. I'd be grateful if someone would attack minimalsmf with a saw and make it use the latest version of SMF support (I'm now so confused about naming that I won't use any other keywords than 'SMF' and 'support'.) Maciej From mwatters at opencsw.org Thu May 28 18:05:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 28 May 2009 11:05:42 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> Message-ID: <4A1EB656.8030700@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Peter, Peter Bonivart wrote: > On Wed, May 27, 2009 at 11:33 PM, Mike Watters wrote: >> I am just waiting on the classutils. I have tested and retested python with >> this recipe, and the utils. I just want one clean test with the classutils >> package rather then me just copying the class scripts in place. > > I have dropped a new cswclassutils package into testing that includes > cswpycompile: > > http://mirror.opencsw.org/testing/cswclassutils-1.10,REV=2009.05.28-SunOS5.8-all-CSW.pkg.gz > I forgot to actually copy/remove the original files in the class scripts for cswpycompile. new "working" copies are in my home dir on the build farm. please update cswclassutils - --- /usr/sadm/install/scripts/i.cswpycompile 2009-05-28 08:14:39.000000000 -0500 +++ i.cswpycompile 2009-05-28 10:55:23.320409700 -0500 @@ -20,6 +20,13 @@ if [ "${DEBUG}" ]; then echo PACKAGE: $PKGINST fi +while read src dest; do + if [ "$DEBUG" ]; then + echo SRC: $src DEST: $dest + fi + + /usr/bin/cp -p $src $dest || exit 2 +done if [ ! -f ${PY_BINDIR}/python -a ! -x ${PY_BINDIR}/python ]; then echo "Could not find or execute ${PY_BINDIR}/python" - --- /usr/sadm/install/scripts/r.cswpycompile 2009-05-28 08:14:39.000000000 -0500 +++ r.cswpycompile 2009-05-28 10:55:23.320922600 -0500 @@ -20,4 +20,12 @@ echo "Removing pyo and pyc files for package <$PKGINST> ..." pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}c {}o +while read dest; do + if [ "$DEBUG" ]; then + echo DEST: $dest + fi + /usr/bin/rm -f $dest || exit 2 +done + + exit 0 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoetlYACgkQLrhmsXMSLxeVjwCgv9ggwnVNG8rcpqPsJj4zxh7A pHgAn1nSQkOoBjV3muSqQJORot67IJCa =81R5 -----END PGP SIGNATURE----- From pfelecan at opencsw.org Thu May 28 18:11:18 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 28 May 2009 18:11:18 +0200 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1E98B3.7020508@opencsw.org> (John Ellson's message of "Thu\, 28 May 2009 09\:59\:15 -0400") References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: John Ellson writes: > I'm a bit reluctant to make such an extensive change to the API of > this library. > Is there a linker flag that would tell it to tolerate these extra const's ? Use the Force: gcc used must be -- Peter From ellson at opencsw.org Thu May 28 18:16:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 28 May 2009 12:16:06 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: <4A1EB8C6.108@opencsw.org> Peter FELECAN wrote: > John Ellson writes: > > >> I'm a bit reluctant to make such an extensive change to the API of >> this library. >> Is there a linker flag that would tell it to tolerate these extra const's ? >> > > Use the Force: gcc used must be > Its ok, I've patched the implementation to avoid changing the api. Perhaps Yoda would care to review ? ;-) mgar/pkg/liblasi/trunk John From phil at bolthole.com Thu May 28 18:31:21 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 28 May 2009 09:31:21 -0700 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: <20090528163121.GC55163@bolthole.com> On Thu, May 28, 2009 at 06:11:18PM +0200, Peter FELECAN wrote: > John Ellson writes: > > > I'm a bit reluctant to make such an extensive change to the API of > > this library. > > Is there a linker flag that would tell it to tolerate these extra const's ? > > Use the Force: gcc used must be NOOOOOO!!!! Once you start down the dark path, forever will it dominate your destiny. Consume you it will! :-} From pfelecan at opencsw.org Thu May 28 18:45:02 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 28 May 2009 18:45:02 +0200 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1EB8C6.108@opencsw.org> (John Ellson's message of "Thu\, 28 May 2009 12\:16\:06 -0400") References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> <4A1EB8C6.108@opencsw.org> Message-ID: John Ellson writes: > Peter FELECAN wrote: >> John Ellson writes: >> >> >>> I'm a bit reluctant to make such an extensive change to the API of >>> this library. >>> Is there a linker flag that would tell it to tolerate these extra const's ? >>> >> >> Use the Force: gcc used must be >> > Its ok, I've patched the implementation to avoid changing the api. > > Perhaps Yoda would care to review ? ;-) > > mgar/pkg/liblasi/trunk no light on the mGar path by hand everything I made -- Peter From bonivart at opencsw.org Thu May 28 20:38:20 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 20:38:20 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1EB656.8030700@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> Message-ID: <625385e30905281138p39ea82afx14fabbb0f5c5691a@mail.gmail.com> On Thu, May 28, 2009 at 6:05 PM, Mike Watters wrote: > I forgot to actually copy/remove the original files in the class scripts for > cswpycompile. ?new "working" copies are in my home dir on the build farm. > please update cswclassutils No problem, I will do it tomorrow. -- /peter From william at wbonnet.net Thu May 28 21:28:54 2009 From: william at wbonnet.net (William Bonnet) Date: Thu, 28 May 2009 21:28:54 +0200 Subject: [csw-maintainers] Thunderbird 2.0.0.21 is in testing Message-ID: <4A1EE5F6.9060403@wbonnet.net> Hi Thunderbird 2.0.0.21 is available from testing. You will need to update libcairo, libpango, pixman from testing, and an up to date glib2 from current (at least 2.20) Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Thu May 28 22:20:04 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 28 May 2009 15:20:04 -0500 Subject: [csw-maintainers] mediawiki 1.14.0 now in testing Message-ID: <4A1EF1F4.1090304@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 latest version: Feedback Welcome! - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoe8fQACgkQLrhmsXMSLxdFEgCfbJmPPwtzCje9O6+rSeF8Ha1f rokAnR0MM9wlVaPnSYEswrqlCyZWYqk/ =tIMr -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu May 28 23:22:59 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 28 May 2009 16:22:59 -0500 Subject: [csw-maintainers] squirrelmail 1.4.17 now in testing Message-ID: <4A1F00B3.5090309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Feedback Welcome! - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkofALMACgkQLrhmsXMSLxdfgACdHCDGHdWr93Q/2QXm7gC0vkn3 X64AnRSeK9BR5PHmxsv73NpKwvVfmV9m =wtFx -----END PGP SIGNATURE----- From maciej at opencsw.org Thu May 28 23:39:18 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 28 May 2009 22:39:18 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: References: <20090527184757.GU55163@bolthole.com> <20090528130328.GA91905@bolthole.com> Message-ID: On Thu, May 28, 2009 at 4:26 PM, Maciej (Matchek) Blizinski wrote: > I'd be grateful if someone would attack minimalsmf with a saw and make > it use the latest version of SMF support (I'm now so confused about > naming that I won't use any other keywords than 'SMF' and 'support'.) Latest news: I've slogged away at the code tree, figured out that clamav does SMF via classutils, figured out that it was consistent with the 'cswclassutls' page, and updated the minimalsmf package. It worked! I've created a version of syslog_ng with mgar v2 and cswclassutils-driven SMF support. I thought that committing the change directly to the syslog_ng/trunk directory might be considered rude, so I branched the build and committed my change to the branch. Here's the change: http://apps.sourceforge.net/trac/gar/changeset/5096 Ihsan, can you please look at it? If it looks good to you, you could merge it to the trunk. Maciej From bonivart at opencsw.org Fri May 29 14:43:48 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 29 May 2009 14:43:48 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1EB656.8030700@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> Message-ID: <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> On Thu, May 28, 2009 at 6:05 PM, Mike Watters wrote: > I forgot to actually copy/remove the original files in the class scripts for > cswpycompile. ?new "working" copies are in my home dir on the build farm. > please update cswclassutils I have updated the package in testing: http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz -- /peter From mwatters at opencsw.org Fri May 29 15:58:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 29 May 2009 08:58:28 -0500 Subject: [csw-maintainers] python 2.6.2 in testing Message-ID: <4A1FEA04.5000708@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: update to 2.6.2 now contains no pre-compiled bytecode in the pacakge requires Peter's cswclassutils 1.14,REV=2009.05.29 that is also in testing. various minor bug fixes. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkof6gQACgkQLrhmsXMSLxd29QCgnha4inMeZfA71FzrF4ImKW6a JIMAoMaMGShP/4WtHtlaxQW4z1IgW3WK =LI0e -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri May 29 16:06:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 29 May 2009 09:06:26 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> Message-ID: <4A1FEBE2.3060907@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Bonivart wrote: > On Thu, May 28, 2009 at 6:05 PM, Mike Watters wrote: >> I forgot to actually copy/remove the original files in the class scripts for >> cswpycompile. new "working" copies are in my home dir on the build farm. >> please update cswclassutils > > I have updated the package in testing: > > http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz > Most Excellent, cswpycompile class works like a charm now. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkof6+IACgkQLrhmsXMSLxdwGACZAdHpre9MpneJgMi2pOKUzbEt pPQAniX9CqpWOYB3jjifPXFN+Ggcnb2I =OIMH -----END PGP SIGNATURE----- From bonivart at opencsw.org Fri May 29 17:03:52 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 29 May 2009 17:03:52 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1FEBE2.3060907@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> Message-ID: <625385e30905290803q3ca0c86dlb8bd8cfd24ff3a16@mail.gmail.com> On Fri, May 29, 2009 at 4:06 PM, Mike Watters wrote: > Most Excellent, cswpycompile class works like a charm now. That's good to hear. :-) However, there's quite a few changes to cswinitsmf in 1.14 as well so we need some testing before releasing cswclassutils. I hope other maintainers can help with testing. - fixed bug that always enabled services using init scripts (#3633) - the persistent state code is back (keeps state during upgrades) (#3634) - AUTOENABLE tweak (provides a way of overriding local csw.conf) (#3635) - MANIFEST tweak (custom manifest instead of autogenerated) (#3636) The tweaks are briefly documented here: http://wiki.opencsw.org/cswclassutils-package#toc2 -- /peter From phil at bolthole.com Fri May 29 18:40:34 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 29 May 2009 09:40:34 -0700 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905290803q3ca0c86dlb8bd8cfd24ff3a16@mail.gmail.com> References: <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <625385e30905290803q3ca0c86dlb8bd8cfd24ff3a16@mail.gmail.com> Message-ID: <20090529164032.GR55163@bolthole.com> On Fri, May 29, 2009 at 05:03:52PM +0200, Peter Bonivart wrote: > - AUTOENABLE tweak (provides a way of overriding local csw.conf) (#3635) yikes. that worried me when I just read that summary :-} but fyi to others, this is explicitly to DISABLE the service until properly configured, even if csw.conf says "autoenable_daemons=yes". From dam at opencsw.org Fri May 29 22:54:18 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 29 May 2009 22:54:18 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1FEBE2.3060907@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> Message-ID: Hi Mike, Am 29.05.2009 um 16:06 schrieb Mike Watters: > Peter Bonivart wrote: >> On Thu, May 28, 2009 at 6:05 PM, Mike Watters >> wrote: >>> I forgot to actually copy/remove the original files in the class >>> scripts for >>> cswpycompile. new "working" copies are in my home dir on the >>> build farm. >>> please update cswclassutils >> >> I have updated the package in testing: >> >> http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz > > Most Excellent, cswpycompile class works like a charm now. I have some troubles with cswpycompile: - I added support to xcb-proto like this: > EXTRA_MERGE_EXCLUDE_FILES = .*\.pyo .*\.pyc > PROTOTYPE_FILTER = awk '$$$$3 ~ /.*\.py$$$$/ { $$$$2 = > "cswpycompile" } { print }' > SPKG_CLASSES = none cswpycompile > REQUIRED_PKGS += CSWcswclassutils - On installation something does get compiled: > Compiling py files to normal bytecode ... > Listing /opt/csw/lib/python ... > Listing /opt/csw/lib/python/bsddb ... > Listing /opt/csw/lib/python/bsddb/test ... > Listing /opt/csw/lib/python/compiler ... > Listing /opt/csw/lib/python/ctypes ... > Listing /opt/csw/lib/python/ctypes/macholib ... > Listing /opt/csw/lib/python/ctypes/test ... > Listing /opt/csw/lib/python/curses ... > Listing /opt/csw/lib/python/email ... > Listing /opt/csw/lib/python/email/mime ... > Listing /opt/csw/lib/python/email/test ... > Listing /opt/csw/lib/python/email/test/data ... > Listing /opt/csw/lib/python/encodings ... > Listing /opt/csw/lib/python/hotshot ... > Listing /opt/csw/lib/python/json ... > Listing /opt/csw/lib/python/json/tests ... > Listing /opt/csw/lib/python/lib-dynload ... > Listing /opt/csw/lib/python/lib-old ... > Listing /opt/csw/lib/python/lib2to3 ... > Listing /opt/csw/lib/python/lib2to3/fixes ... > Listing /opt/csw/lib/python/lib2to3/pgen2 ... > Listing /opt/csw/lib/python/lib2to3/tests ... > Listing /opt/csw/lib/python/logging ... > Listing /opt/csw/lib/python/multiprocessing ... > Listing /opt/csw/lib/python/multiprocessing/dummy ... > Listing /opt/csw/lib/python/plat-sunos5 ... > Listing /opt/csw/lib/python/site-packages ... > Listing /opt/csw/lib/python/sqlite3 ... > Listing /opt/csw/lib/python/sqlite3/test ... > Listing /opt/csw/lib/python/test ... > Listing /opt/csw/lib/python/test/decimaltestdata ... > Listing /opt/csw/lib/python/wsgiref ... > Listing /opt/csw/lib/python/xml ... > Listing /opt/csw/lib/python/xml/dom ... > Listing /opt/csw/lib/python/xml/etree ... > Listing /opt/csw/lib/python/xml/parsers ... > Listing /opt/csw/lib/python/xml/sax ... > Compiling py files to optimized bytecode ... > Listing /opt/csw/lib/python ... > Listing /opt/csw/lib/python/bsddb ... > Listing /opt/csw/lib/python/bsddb/test ... > Listing /opt/csw/lib/python/compiler ... > Listing /opt/csw/lib/python/ctypes ... > Listing /opt/csw/lib/python/ctypes/macholib ... > Listing /opt/csw/lib/python/ctypes/test ... > Listing /opt/csw/lib/python/curses ... > Listing /opt/csw/lib/python/email ... > Listing /opt/csw/lib/python/email/mime ... > Listing /opt/csw/lib/python/email/test ... > Listing /opt/csw/lib/python/email/test/data ... > Listing /opt/csw/lib/python/encodings ... > Listing /opt/csw/lib/python/hotshot ... > Listing /opt/csw/lib/python/json ... > Listing /opt/csw/lib/python/json/tests ... > Listing /opt/csw/lib/python/lib-dynload ... > Listing /opt/csw/lib/python/lib-old ... > Listing /opt/csw/lib/python/lib2to3 ... > Listing /opt/csw/lib/python/lib2to3/fixes ... > Listing /opt/csw/lib/python/lib2to3/pgen2 ... > Listing /opt/csw/lib/python/lib2to3/tests ... > Listing /opt/csw/lib/python/logging ... > Listing /opt/csw/lib/python/multiprocessing ... > Listing /opt/csw/lib/python/multiprocessing/dummy ... > Listing /opt/csw/lib/python/plat-sunos5 ... > Listing /opt/csw/lib/python/site-packages ... > Listing /opt/csw/lib/python/sqlite3 ... > Listing /opt/csw/lib/python/sqlite3/test ... > Listing /opt/csw/lib/python/test ... > Listing /opt/csw/lib/python/test/decimaltestdata ... > Listing /opt/csw/lib/python/wsgiref ... > Listing /opt/csw/lib/python/xml ... > Listing /opt/csw/lib/python/xml/dom ... > Listing /opt/csw/lib/python/xml/etree ... > Listing /opt/csw/lib/python/xml/parsers ... > Listing /opt/csw/lib/python/xml/sax ... > [ verifying class ] > > Installation of was successful. - From the remove script I would guess the compiled and optimized files are in the same directory as the .py-files: > echo "Removing pyo and pyc files for package <$PKGINST> ..." > pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}c {}o - However, in the directory where the packages installed the .py-files nothing is added: > build8st# ls -l /opt/csw/X11/lib/python2.6/site-packages/xcbgen > total 64 > -rw-r--r-- 1 root bin 1 May 29 22:24 __init__.py > -rw-r--r-- 1 root bin 116 May 29 22:24 error.py > -rw-r--r-- 1 root bin 2965 May 29 22:24 expr.py > -rw-r--r-- 1 root bin 3534 May 29 22:24 matcher.py > -rw-r--r-- 1 root bin 5141 May 29 22:24 state.py > -rw-r--r-- 1 root bin 16352 May 29 22:24 xtypes.py So, what's going wrong here? After this is settled and CSWcswclassutils and CSWpython get released I'll modify GAR to automatically do it right: - look if there are .py files, if yes: - remove .pyc and .pyo files on merge - modify class of .py to cswpycompile - Add dependency to CSWcswclassutils Best regards -- Dago From mwatters at opencsw.org Sat May 30 01:27:46 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 29 May 2009 18:27:46 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> Message-ID: <4A206F72.7010909@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi Mike, > > Am 29.05.2009 um 16:06 schrieb Mike Watters: >> Peter Bonivart wrote: >>> On Thu, May 28, 2009 at 6:05 PM, Mike Watters >>> wrote: >>>> I forgot to actually copy/remove the original files in the class >>>> scripts for >>>> cswpycompile. new "working" copies are in my home dir on the build >>>> farm. >>>> please update cswclassutils >>> >>> I have updated the package in testing: >>> >>> http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz >>> >> >> Most Excellent, cswpycompile class works like a charm now. > > I have some troubles with cswpycompile: > > - I added support to xcb-proto like this: > >> EXTRA_MERGE_EXCLUDE_FILES = .*\.pyo .*\.pyc >> PROTOTYPE_FILTER = awk '$$$$3 ~ /.*\.py$$$$/ { $$$$2 = "cswpycompile" >> } { print }' >> SPKG_CLASSES = none cswpycompile >> REQUIRED_PKGS += CSWcswclassutils > > - On installation something does get compiled: > >> Compiling py files to normal bytecode ... >> Listing /opt/csw/lib/python ... >> Listing /opt/csw/lib/python/bsddb ... >> Listing /opt/csw/lib/python/bsddb/test ... >> Listing /opt/csw/lib/python/compiler ... >> Listing /opt/csw/lib/python/ctypes ... >> Listing /opt/csw/lib/python/ctypes/macholib ... >> Listing /opt/csw/lib/python/ctypes/test ... >> Listing /opt/csw/lib/python/curses ... >> Listing /opt/csw/lib/python/email ... >> Listing /opt/csw/lib/python/email/mime ... >> Listing /opt/csw/lib/python/email/test ... >> Listing /opt/csw/lib/python/email/test/data ... >> Listing /opt/csw/lib/python/encodings ... >> Listing /opt/csw/lib/python/hotshot ... >> Listing /opt/csw/lib/python/json ... >> Listing /opt/csw/lib/python/json/tests ... >> Listing /opt/csw/lib/python/lib-dynload ... >> Listing /opt/csw/lib/python/lib-old ... >> Listing /opt/csw/lib/python/lib2to3 ... >> Listing /opt/csw/lib/python/lib2to3/fixes ... >> Listing /opt/csw/lib/python/lib2to3/pgen2 ... >> Listing /opt/csw/lib/python/lib2to3/tests ... >> Listing /opt/csw/lib/python/logging ... >> Listing /opt/csw/lib/python/multiprocessing ... >> Listing /opt/csw/lib/python/multiprocessing/dummy ... >> Listing /opt/csw/lib/python/plat-sunos5 ... >> Listing /opt/csw/lib/python/site-packages ... >> Listing /opt/csw/lib/python/sqlite3 ... >> Listing /opt/csw/lib/python/sqlite3/test ... >> Listing /opt/csw/lib/python/test ... >> Listing /opt/csw/lib/python/test/decimaltestdata ... >> Listing /opt/csw/lib/python/wsgiref ... >> Listing /opt/csw/lib/python/xml ... >> Listing /opt/csw/lib/python/xml/dom ... >> Listing /opt/csw/lib/python/xml/etree ... >> Listing /opt/csw/lib/python/xml/parsers ... >> Listing /opt/csw/lib/python/xml/sax ... >> Compiling py files to optimized bytecode ... >> Listing /opt/csw/lib/python ... >> Listing /opt/csw/lib/python/bsddb ... >> Listing /opt/csw/lib/python/bsddb/test ... >> Listing /opt/csw/lib/python/compiler ... >> Listing /opt/csw/lib/python/ctypes ... >> Listing /opt/csw/lib/python/ctypes/macholib ... >> Listing /opt/csw/lib/python/ctypes/test ... >> Listing /opt/csw/lib/python/curses ... >> Listing /opt/csw/lib/python/email ... >> Listing /opt/csw/lib/python/email/mime ... >> Listing /opt/csw/lib/python/email/test ... >> Listing /opt/csw/lib/python/email/test/data ... >> Listing /opt/csw/lib/python/encodings ... >> Listing /opt/csw/lib/python/hotshot ... >> Listing /opt/csw/lib/python/json ... >> Listing /opt/csw/lib/python/json/tests ... >> Listing /opt/csw/lib/python/lib-dynload ... >> Listing /opt/csw/lib/python/lib-old ... >> Listing /opt/csw/lib/python/lib2to3 ... >> Listing /opt/csw/lib/python/lib2to3/fixes ... >> Listing /opt/csw/lib/python/lib2to3/pgen2 ... >> Listing /opt/csw/lib/python/lib2to3/tests ... >> Listing /opt/csw/lib/python/logging ... >> Listing /opt/csw/lib/python/multiprocessing ... >> Listing /opt/csw/lib/python/multiprocessing/dummy ... >> Listing /opt/csw/lib/python/plat-sunos5 ... >> Listing /opt/csw/lib/python/site-packages ... >> Listing /opt/csw/lib/python/sqlite3 ... >> Listing /opt/csw/lib/python/sqlite3/test ... >> Listing /opt/csw/lib/python/test ... >> Listing /opt/csw/lib/python/test/decimaltestdata ... >> Listing /opt/csw/lib/python/wsgiref ... >> Listing /opt/csw/lib/python/xml ... >> Listing /opt/csw/lib/python/xml/dom ... >> Listing /opt/csw/lib/python/xml/etree ... >> Listing /opt/csw/lib/python/xml/parsers ... >> Listing /opt/csw/lib/python/xml/sax ... >> [ verifying class ] >> >> Installation of was successful. > > - From the remove script I would guess the compiled and optimized files are > in the same directory as the .py-files: > >> echo "Removing pyo and pyc files for package <$PKGINST> ..." >> pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}c {}o > > - However, in the directory where the packages installed the .py-files > nothing is added: > >> build8st# ls -l /opt/csw/X11/lib/python2.6/site-packages/xcbgen >> total 64 >> -rw-r--r-- 1 root bin 1 May 29 22:24 __init__.py >> -rw-r--r-- 1 root bin 116 May 29 22:24 error.py >> -rw-r--r-- 1 root bin 2965 May 29 22:24 expr.py >> -rw-r--r-- 1 root bin 3534 May 29 22:24 matcher.py >> -rw-r--r-- 1 root bin 5141 May 29 22:24 state.py >> -rw-r--r-- 1 root bin 16352 May 29 22:24 xtypes.py > > So, what's going wrong here? > > After this is settled and CSWcswclassutils and CSWpython get released > I'll modify GAR to automatically do it right: > > - look if there are .py files, if yes: > - remove .pyc and .pyo files on merge > - modify class of .py to cswpycompile > - Add dependency to CSWcswclassutils > > > Best regards > > -- Dago > > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers which version of classutils are you using? The Listing is normal, it just means it found a directory and is Listing the contents to the script. The way I have the install class script written, is it should try to compile everything under /opt/csw/lib/python that can be compiled. I need to research this weekend on whether or not this will traverse the site-packages directory. I thought it did, but it may not. Look at your install directory, does it contain pyo/pyc files? Thanks for the report. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkogb3IACgkQLrhmsXMSLxc4aQCgs/3Oj/b8nxrv2Nmu4NRnvALS GNMAoK1/WvSSTxlwVe32AexLYtQiY/An =TT+I -----END PGP SIGNATURE----- From rupert at opencsw.org Sun May 31 17:05:28 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 31 May 2009 17:05:28 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <14077.1243297943@gerd> References: <6af4270905250658u4e045f47v3e2a763400f14109@mail.gmail.com> <14077.1243297943@gerd> Message-ID: <6af4270905310805h5187ef6cn4e81ec046005abb5@mail.gmail.com> On Tue, May 26, 2009 at 02:32, wrote: > On Mon, 25 May 2009 23:58:37 +1000, rupert THURNER writes: >>i tried to build it on the standard build servers. how did you install >>it? > > I have my own "build servers". You will have to request that renderPM be > installed on the central "build servers". > > But first check that the version of renderPM in CSW is late enough for > reportlab 2.3 - you may need to build and submit renderPM first, then > get it installed on the "build servers", then move on to building > reportlab. > > Also don't forget to check out the rl_accel plugin for reportlab. > >>or could i have access to your build file? > > My build file basically is this: > > ? ? ? ?python setup.py build > ? ? ? ?mkdir -p ${buildroot} > ? ? ? ?python setup.py install --root=${buildroot} > ? ? ? ?chown -Rh root:bin ${buildroot}/opt > > i.e. the standard way to build and install python packages (I believe). > > GAR will probably have mechanisms for this by default(?). Cheers! > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Murray... hmm .. meanwhile i tried the version built on build8s, and it works great. but on build8x i still did not see whom to ask to correct what, or what i could correct ? From Murray.Jensen at csiro.au Sun May 31 17:10:54 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Mon, 1 Jun 2009 01:10:54 +1000 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: Your message of "Mon, 01 Jun 2009 01:05:28 +1000" <6af4270905310805h5187ef6cn4e81ec046005abb5@mail.gmail.com> Message-ID: <29335.1243782654@gerd> On Mon, 1 Jun 2009 01:05:28 +1000, rupert THURNER writes: >hmm .. meanwhile i tried the version built on build8s, and it works >great. but on build8x i still did not see whom to ask to correct what, >or what i could correct ? Are you sure that renderPM is installed on both build8s and build8x? Cheers! Murray... From dam at opencsw.org Sun May 31 22:53:32 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 31 May 2009 22:53:32 +0200 Subject: [csw-maintainers] Fwd: [csw-users] feedback on pkg-get behaviour change requested References: Message-ID: Hi, this was just on users at . Gary: what do you think? I guess nobody was aware of this issue. Best regards -- Dago Anfang der weitergeleiteten E-Mail: > Von: Chris May > Datum: 31. Mai 2009 22:37:16 MESZ > An: users at lists.opencsw.org > Betreff: Re: [csw-users] feedback on pkg-get behaviour change > requested > Antwort an: questions and discussions > > Philip Brown wrote: > [...] >> I am considering altering the behavior of "pkg-get -c", so that it >> only prints out installed packages, that have DIFFERENT versions in >> the >> catalog. > > Has this change now been made? I notice what looks like this > difference between the > behaviour of 4.1.2 and 4.2, when querying for a package which is > already installed. > Comparing the two... > > # /tmp/pkg-get.4.1.2 -c nrpe > # (From site http://ibiblio.org/pub/packages/solaris/opencsw/current ) > software localrev > remoterev > nrpe > 2.12,REV=2009.05.17 SAME > > # /tmp/pkg-get.4.2 -c nrpe > # (From site http://ibiblio.org/pub/packages/solaris/opencsw/current ) > software localrev > remoterev > # > > This causes a problem for the Puppet configuration management > software(http://reductivelabs.com/trac/puppet), which parses the > output of pkg-get -c {package} to determine whether a package is > installed or not. The absence of the "SAME" line causes it to think > that the package is not present, and hence re-install it. > > Is this intended to be a permanent change? If so, I'll raise an > enhancement request with the Puppet maintainers to use '-c -a > {package}' instead - which looks as if it should work in the same way > with either pkg-get 4.1.2 or 4.2 From dam at opencsw.org Sun May 31 23:00:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 31 May 2009 23:00:37 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A206F72.7010909@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> Message-ID: <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Hi Mike, Am 30.05.2009 um 01:27 schrieb Mike Watters: > which version of classutils are you using? The one from testing, of couse. That is 1.5,REV=2009.03.28 > The Listing is normal, it just means it found a directory and is > Listing the > contents to the script. > > The way I have the install class script written, > is it should try to compile everything under /opt/csw/lib/python > that can be > compiled. I need to research this weekend on whether or not this > will traverse > the site-packages directory. I thought it did, but it may not. > > Look at your install directory, does it contain pyo/pyc files? No. I opensnoop'ed the pkginstall of my package from the globalzone, and there doesn't seem to be any compiled files generated. The package is additionally in testing/: Best regards -- Dago root at ncsw [global]:/root > opensnoop | fgrep py 0 17552 cpio 3 install/copyright 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/xtypes.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/error.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/state.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/__init__.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/expr.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/matcher.py 0 17546 pkginstall 5 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ install/copyright 0 17546 pkginstall 12 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ install/copyright 0 17546 pkginstall 14 /var/sadm/pkg/CSWpython/pkginfo 0 17546 pkginstall 14 /var/sadm/pkg/CSWpython/pkginfo 0 17546 pkginstall 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ install/copyright 0 17546 pkginstall 8 /var/sadm/pkg/CSWxcbproto/install/ copyright 0 17554 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.py 0 17556 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/expr.py 0 17557 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/matcher.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/expr.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/matcher.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/state.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py 0 17555 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.py 0 17553 sh 7 /usr/sadm/install/scripts/i.cswpycompile 0 17558 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/state.py 0 17559 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py 0 17560 python 7 /var/ld/ld.config 0 17560 python 7 /usr/lib/libc.so.1 0 17560 python 7 /usr/lib/libdl.so.1 0 17560 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17560 python 7 /.machine 0 17560 python 7 /.hostid 0 17560 python 7 /var/ld/ld.config 0 17560 python 7 /usr/lib/secure/s8_preload.so.1 0 17560 python 7 /opt/csw/lib/libpython2.6.so.1.0 0 17560 python 7 /usr/lib/libresolv.so.2 0 17560 python 7 /usr/lib/libsocket.so.1 0 17560 python 7 /usr/lib/libnsl.so.1 0 17560 python 7 /usr/lib/librt.so.1 0 17560 python 7 /usr/lib/libdl.so.1 0 17560 python 7 /usr/lib/libpthread.so.1 0 17560 python 7 /usr/lib/libm.so.1 0 17560 python 7 /usr/lib/libc.so.1 0 17560 python 7 /usr/lib/libmp.so.2 0 17560 python 7 /usr/lib/libaio.so.1 0 17560 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17560 python 7 /usr/lib/libthread.so.1 0 17560 python -1 /opt/csw/lib/python/warnings.so 0 17560 python -1 /opt/csw/lib/python/warningsmodule.so 0 17560 python 7 /opt/csw/lib/python/warnings.py 0 17560 python 8 /opt/csw/lib/python/warnings.pyc 0 17560 python -1 /opt/csw/lib/python/linecache.so 0 17560 python -1 /opt/csw/lib/python/linecachemodule.so 0 17560 python 8 /opt/csw/lib/python/linecache.py 0 17560 python 10 /opt/csw/lib/python/linecache.pyc 0 17560 python -1 /opt/csw/lib/python/os.so 0 17560 python -1 /opt/csw/lib/python/osmodule.so 0 17560 python 10 /opt/csw/lib/python/os.py 0 17560 python 12 /opt/csw/lib/python/os.pyc 0 17560 python -1 /opt/csw/lib/python/posixpath.so 0 17560 python -1 /opt/csw/lib/python/posixpathmodule.so 0 17560 python 12 /opt/csw/lib/python/posixpath.py 0 17560 python 13 /opt/csw/lib/python/posixpath.pyc 0 17560 python -1 /opt/csw/lib/python/stat.so 0 17560 python -1 /opt/csw/lib/python/statmodule.so 0 17560 python 13 /opt/csw/lib/python/stat.py 0 17560 python 14 /opt/csw/lib/python/stat.pyc 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/usercustomize.so 0 17560 python -1 /opt/csw/lib/python/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/usercustomize.py 0 17560 python -1 /opt/csw/lib/python/usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/genericpath.so 0 17560 python -1 /opt/csw/lib/python/genericpathmodule.so 0 17560 python 13 /opt/csw/lib/python/genericpath.py 0 17560 python 14 /opt/csw/lib/python/genericpath.pyc 0 17560 python -1 /opt/csw/lib/python/UserDict.so 0 17560 python -1 /opt/csw/lib/python/UserDictmodule.so 0 17560 python 12 /opt/csw/lib/python/UserDict.py 0 17560 python 13 /opt/csw/lib/python/UserDict.pyc 0 17560 python -1 /opt/csw/lib/python/_abcoll.so 0 17560 python -1 /opt/csw/lib/python/_abcollmodule.so 0 17560 python 13 /opt/csw/lib/python/_abcoll.py 0 17560 python 14 /opt/csw/lib/python/_abcoll.pyc 0 17560 python -1 /opt/csw/lib/python/abc.so 0 17560 python -1 /opt/csw/lib/python/abcmodule.so 0 17560 python 14 /opt/csw/lib/python/abc.py 0 17560 python 15 /opt/csw/lib/python/abc.pyc 0 17560 python -1 /opt/csw/lib/python/copy_reg.so 0 17560 python -1 /opt/csw/lib/python/copy_regmodule.so 0 17560 python 12 /opt/csw/lib/python/copy_reg.py 0 17560 python 13 /opt/csw/lib/python/copy_reg.pyc 0 17560 python -1 /opt/csw/lib/python/types.so 0 17560 python -1 /opt/csw/lib/python/typesmodule.so 0 17560 python 13 /opt/csw/lib/python/types.py 0 17560 python 14 /opt/csw/lib/python/types.pyc 0 17560 python -1 /opt/csw/lib/python/re.so 0 17560 python -1 /opt/csw/lib/python/remodule.so 0 17560 python 8 /opt/csw/lib/python/re.py 0 17560 python 10 /opt/csw/lib/python/re.pyc 0 17560 python -1 /opt/csw/lib/python/sre_compile.so 0 17560 python -1 /opt/csw/lib/python/sre_compilemodule.so 0 17560 python 10 /opt/csw/lib/python/sre_compile.py 0 17560 python 12 /opt/csw/lib/python/sre_compile.pyc 0 17560 python -1 /opt/csw/lib/python/sre_parse.so 0 17560 python -1 /opt/csw/lib/python/sre_parsemodule.so 0 17560 python 12 /opt/csw/lib/python/sre_parse.py 0 17560 python 13 /opt/csw/lib/python/sre_parse.pyc 0 17560 python -1 /opt/csw/lib/python/sre_constants.so 0 17560 python -1 /opt/csw/lib/python/ sre_constantsmodule.so 0 17560 python 13 /opt/csw/lib/python/sre_constants.py 0 17560 python 14 /opt/csw/lib/python/sre_constants.pyc 0 17560 python -1 /opt/csw/lib/python/site.so 0 17560 python -1 /opt/csw/lib/python/sitemodule.so 0 17560 python 7 /opt/csw/lib/python/site.py 0 17560 python 8 /opt/csw/lib/python/site.pyc 0 17560 python 8 /etc/.name_service_door 0 17560 python 10 /opt/csw/lib/python/site-packages 0 17560 python -1 /opt/csw/lib/python/sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/__init__.so 0 17560 python -1 /opt/csw/lib/python/encodings/ __init__module.so 0 17560 python 7 /opt/csw/lib/python/encodings/__init__.py 0 17560 python 10 /opt/csw/lib/python/encodings/ __init__.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/codecs.so 0 17560 python -1 /opt/csw/lib/python/encodings/ codecsmodule.so 0 17560 python -1 /opt/csw/lib/python/encodings/codecs.py 0 17560 python -1 /opt/csw/lib/python/encodings/codecs.pyc 0 17560 python -1 /opt/csw/lib/python/codecs.so 0 17560 python -1 /opt/csw/lib/python/codecsmodule.so 0 17560 python 10 /opt/csw/lib/python/codecs.py 0 17560 python 12 /opt/csw/lib/python/codecs.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/ encodings.so 0 17560 python -1 /opt/csw/lib/python/encodings/ encodingsmodule.so 0 17560 python -1 /opt/csw/lib/python/encodings/ encodings.py 0 17560 python -1 /opt/csw/lib/python/encodings/ encodings.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/aliases.so 0 17560 python -1 /opt/csw/lib/python/encodings/ aliasesmodule.so 0 17560 python 10 /opt/csw/lib/python/encodings/aliases.py 0 17560 python 12 /opt/csw/lib/python/encodings/aliases.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__.so 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__module.so 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__.py 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/ascii.so 0 17560 python -1 /opt/csw/lib/python/encodings/ asciimodule.so 0 17560 python 7 /opt/csw/lib/python/encodings/ascii.py 0 17560 python 10 /opt/csw/lib/python/encodings/ascii.pyc 0 17560 python 7 /opt/csw/lib/python/compileall.py 0 17560 python 7 /opt/csw/lib/python/compileall.py 0 17560 python -1 /opt/csw/lib/python/py_compile.so 0 17560 python -1 /opt/csw/lib/python/py_compilemodule.so 0 17560 python 7 /opt/csw/lib/python/py_compile.py 0 17560 python 10 /opt/csw/lib/python/py_compile.pyc 0 17560 python -1 /opt/csw/lib/python/traceback.so 0 17560 python -1 /opt/csw/lib/python/tracebackmodule.so 0 17560 python 10 /opt/csw/lib/python/traceback.py 0 17560 python 12 /opt/csw/lib/python/traceback.pyc 0 17560 python -1 /opt/csw/lib/python/getopt.so 0 17560 python -1 /opt/csw/lib/python/getoptmodule.so 0 17560 python 7 /opt/csw/lib/python/getopt.py 0 17560 python 10 /opt/csw/lib/python/getopt.pyc 0 17560 python 7 /opt/csw/lib/python 0 17560 python 7 /opt/csw/lib/python/bsddb 0 17560 python 7 /opt/csw/lib/python/compiler 0 17560 python 7 /opt/csw/lib/python/ctypes 0 17560 python 7 /opt/csw/lib/python/ctypes/macholib 0 17560 python 7 /opt/csw/lib/python/ctypes/test 0 17560 python 7 /opt/csw/lib/python/bsddb/test 0 17560 python 7 /opt/csw/lib/python/curses 0 17560 python 7 /opt/csw/lib/python/email 0 17560 python 7 /opt/csw/lib/python/email/mime 0 17560 python 7 /opt/csw/lib/python/email/test 0 17560 python 7 /opt/csw/lib/python/email/test/data 0 17560 python 7 /opt/csw/lib/python/encodings 0 17560 python 7 /opt/csw/lib/python/hotshot 0 17560 python 7 /opt/csw/lib/python/json 0 17560 python 7 /opt/csw/lib/python/json/tests 0 17560 python 7 /opt/csw/lib/python/lib-dynload 0 17560 python 7 /opt/csw/lib/python/lib-old 0 17560 python 7 /opt/csw/lib/python/lib2to3 0 17560 python 7 /opt/csw/lib/python/lib2to3/fixes 0 17560 python 7 /opt/csw/lib/python/lib2to3/pgen2 0 17560 python 7 /opt/csw/lib/python/lib2to3/tests 0 17560 python 7 /opt/csw/lib/python/logging 0 17560 python 7 /opt/csw/lib/python/multiprocessing 0 17560 python 7 /opt/csw/lib/python/multiprocessing/dummy 0 17560 python 7 /opt/csw/lib/python/plat-sunos5 0 17560 python 7 /opt/csw/lib/python/site-packages 0 17560 python 7 /opt/csw/lib/python/sqlite3 0 17560 python 7 /opt/csw/lib/python/sqlite3/test 0 17560 python 7 /opt/csw/lib/python/test 0 17560 python 7 /opt/csw/lib/python/test/decimaltestdata 0 17560 python 7 /opt/csw/lib/python/wsgiref 0 17560 python 7 /opt/csw/lib/python/xml 0 17561 python 7 /var/ld/ld.config 0 17561 python 7 /usr/lib/libc.so.1 0 17561 python 7 /usr/lib/libdl.so.1 0 17561 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17561 python 7 /.machine 0 17561 python 7 /.hostid 0 17561 python 7 /var/ld/ld.config 0 17561 python 7 /usr/lib/secure/s8_preload.so.1 0 17561 python 7 /opt/csw/lib/libpython2.6.so.1.0 0 17561 python 7 /usr/lib/libresolv.so.2 0 17561 python 7 /usr/lib/libsocket.so.1 0 17561 python 7 /usr/lib/libnsl.so.1 0 17561 python 7 /usr/lib/librt.so.1 0 17561 python 7 /usr/lib/libdl.so.1 0 17561 python 7 /usr/lib/libpthread.so.1 0 17561 python 7 /usr/lib/libm.so.1 0 17561 python 7 /usr/lib/libc.so.1 0 17561 python 7 /usr/lib/libmp.so.2 0 17561 python 7 /usr/lib/libaio.so.1 0 17561 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17561 python 7 /usr/lib/libthread.so.1 0 17561 python -1 /opt/csw/lib/python/warnings.so 0 17561 python -1 /opt/csw/lib/python/warningsmodule.so 0 17561 python 7 /opt/csw/lib/python/warnings.py 0 17561 python 8 /opt/csw/lib/python/warnings.pyo 0 17561 python -1 /opt/csw/lib/python/linecache.so 0 17561 python -1 /opt/csw/lib/python/linecachemodule.so 0 17561 python 8 /opt/csw/lib/python/linecache.py 0 17561 python 10 /opt/csw/lib/python/linecache.pyo 0 17561 python -1 /opt/csw/lib/python/os.so 0 17561 python -1 /opt/csw/lib/python/osmodule.so 0 17561 python 10 /opt/csw/lib/python/os.py 0 17561 python 12 /opt/csw/lib/python/os.pyo 0 17561 python -1 /opt/csw/lib/python/posixpath.so 0 17561 python -1 /opt/csw/lib/python/posixpathmodule.so 0 17561 python 12 /opt/csw/lib/python/posixpath.py 0 17561 python 13 /opt/csw/lib/python/posixpath.pyo 0 17561 python -1 /opt/csw/lib/python/stat.so 0 17561 python -1 /opt/csw/lib/python/statmodule.so 0 17561 python 13 /opt/csw/lib/python/stat.py 0 17561 python 14 /opt/csw/lib/python/stat.pyo 0 17560 python 7 /opt/csw/lib/python/xml/dom 0 17560 python 7 /opt/csw/lib/python/xml/etree 0 17560 python 7 /opt/csw/lib/python/xml/parsers 0 17560 python 7 /opt/csw/lib/python/xml/sax 0 17561 python -1 /opt/csw/lib/python/genericpath.so 0 17561 python -1 /opt/csw/lib/python/genericpathmodule.so 0 17561 python 13 /opt/csw/lib/python/genericpath.py 0 17561 python 14 /opt/csw/lib/python/genericpath.pyo 0 17561 python -1 /opt/csw/lib/python/UserDict.so 0 17561 python -1 /opt/csw/lib/python/UserDictmodule.so 0 17561 python 12 /opt/csw/lib/python/UserDict.py 0 17561 python 13 /opt/csw/lib/python/UserDict.pyo 0 17561 python -1 /opt/csw/lib/python/_abcoll.so 0 17561 python -1 /opt/csw/lib/python/_abcollmodule.so 0 17561 python 13 /opt/csw/lib/python/_abcoll.py 0 17561 python 14 /opt/csw/lib/python/_abcoll.pyo 0 17561 python -1 /opt/csw/lib/python/abc.so 0 17561 python -1 /opt/csw/lib/python/abcmodule.so 0 17561 python 14 /opt/csw/lib/python/abc.py 0 17561 python 15 /opt/csw/lib/python/abc.pyo 0 17561 python -1 /opt/csw/lib/python/copy_reg.so 0 17561 python -1 /opt/csw/lib/python/copy_regmodule.so 0 17561 python 12 /opt/csw/lib/python/copy_reg.py 0 17561 python 13 /opt/csw/lib/python/copy_reg.pyo 0 17561 python -1 /opt/csw/lib/python/types.so 0 17561 python -1 /opt/csw/lib/python/typesmodule.so 0 17561 python 13 /opt/csw/lib/python/types.py 0 17561 python 14 /opt/csw/lib/python/types.pyo 0 17561 python -1 /opt/csw/lib/python/re.so 0 17561 python -1 /opt/csw/lib/python/remodule.so 0 17561 python 8 /opt/csw/lib/python/re.py 0 17561 python 10 /opt/csw/lib/python/re.pyo 0 17561 python -1 /opt/csw/lib/python/sre_compile.so 0 17561 python -1 /opt/csw/lib/python/sre_compilemodule.so 0 17561 python 10 /opt/csw/lib/python/sre_compile.py 0 17561 python 12 /opt/csw/lib/python/sre_compile.pyo 0 17561 python -1 /opt/csw/lib/python/sre_parse.so 0 17561 python -1 /opt/csw/lib/python/sre_parsemodule.so 0 17561 python 12 /opt/csw/lib/python/sre_parse.py 0 17561 python 13 /opt/csw/lib/python/sre_parse.pyo 0 17561 python -1 /opt/csw/lib/python/sre_constants.so 0 17561 python -1 /opt/csw/lib/python/ sre_constantsmodule.so 0 17561 python 13 /opt/csw/lib/python/sre_constants.py 0 17561 python 14 /opt/csw/lib/python/sre_constants.pyo 0 17561 python -1 /opt/csw/lib/python/site.so 0 17561 python -1 /opt/csw/lib/python/sitemodule.so 0 17561 python 7 /opt/csw/lib/python/site.py 0 17561 python 8 /opt/csw/lib/python/site.pyo 0 17561 python 8 /etc/.name_service_door 0 17561 python 10 /opt/csw/lib/python/site-packages 0 17561 python -1 /opt/csw/lib/python/sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/usercustomize.so 0 17561 python -1 /opt/csw/lib/python/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/usercustomize.py 0 17561 python -1 /opt/csw/lib/python/usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/__init__.so 0 17561 python -1 /opt/csw/lib/python/encodings/ __init__module.so 0 17561 python 7 /opt/csw/lib/python/encodings/__init__.py 0 17561 python 10 /opt/csw/lib/python/encodings/ __init__.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/codecs.so 0 17561 python -1 /opt/csw/lib/python/encodings/ codecsmodule.so 0 17561 python -1 /opt/csw/lib/python/encodings/codecs.py 0 17561 python -1 /opt/csw/lib/python/encodings/codecs.pyo 0 17561 python -1 /opt/csw/lib/python/codecs.so 0 17561 python -1 /opt/csw/lib/python/codecsmodule.so 0 17561 python 10 /opt/csw/lib/python/codecs.py 0 17561 python 12 /opt/csw/lib/python/codecs.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/ encodings.so 0 17561 python -1 /opt/csw/lib/python/encodings/ encodingsmodule.so 0 17561 python -1 /opt/csw/lib/python/encodings/ encodings.py 0 17561 python -1 /opt/csw/lib/python/encodings/ encodings.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/aliases.so 0 17561 python -1 /opt/csw/lib/python/encodings/ aliasesmodule.so 0 17561 python 10 /opt/csw/lib/python/encodings/aliases.py 0 17561 python 12 /opt/csw/lib/python/encodings/aliases.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__.so 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__module.so 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__.py 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/ascii.so 0 17561 python -1 /opt/csw/lib/python/encodings/ asciimodule.so 0 17561 python 7 /opt/csw/lib/python/encodings/ascii.py 0 17561 python 10 /opt/csw/lib/python/encodings/ascii.pyo 0 17561 python 7 /opt/csw/lib/python/compileall.py 0 17561 python 7 /opt/csw/lib/python/compileall.py 0 17561 python -1 /opt/csw/lib/python/py_compile.so 0 17561 python -1 /opt/csw/lib/python/py_compilemodule.so 0 17561 python 7 /opt/csw/lib/python/py_compile.py 0 17561 python 10 /opt/csw/lib/python/py_compile.pyo 0 17561 python -1 /opt/csw/lib/python/traceback.so 0 17561 python -1 /opt/csw/lib/python/tracebackmodule.so 0 17561 python 10 /opt/csw/lib/python/traceback.py 0 17561 python 12 /opt/csw/lib/python/traceback.pyo 0 17561 python -1 /opt/csw/lib/python/getopt.so 0 17561 python -1 /opt/csw/lib/python/getoptmodule.so 0 17561 python 7 /opt/csw/lib/python/getopt.py 0 17561 python 10 /opt/csw/lib/python/getopt.pyo 0 17561 python 7 /opt/csw/lib/python 0 17561 python 7 /opt/csw/lib/python/bsddb 0 17561 python 7 /opt/csw/lib/python/bsddb/test 0 17561 python 7 /opt/csw/lib/python/compiler 0 17561 python 7 /opt/csw/lib/python/ctypes 0 17561 python 7 /opt/csw/lib/python/ctypes/macholib 0 17561 python 7 /opt/csw/lib/python/ctypes/test 0 17561 python 7 /opt/csw/lib/python/curses 0 17561 python 7 /opt/csw/lib/python/email 0 17561 python 7 /opt/csw/lib/python/email/mime 0 17561 python 7 /opt/csw/lib/python/email/test 0 17561 python 7 /opt/csw/lib/python/email/test/data 0 17561 python 7 /opt/csw/lib/python/encodings 0 17561 python 7 /opt/csw/lib/python/hotshot 0 17561 python 7 /opt/csw/lib/python/json 0 17561 python 7 /opt/csw/lib/python/json/tests 0 17561 python 7 /opt/csw/lib/python/lib-dynload 0 17561 python 7 /opt/csw/lib/python/lib-old 0 17561 python 7 /opt/csw/lib/python/lib2to3 0 17561 python 7 /opt/csw/lib/python/lib2to3/fixes 0 17561 python 7 /opt/csw/lib/python/lib2to3/pgen2 0 17561 python 7 /opt/csw/lib/python/lib2to3/tests 0 17561 python 7 /opt/csw/lib/python/logging 0 17561 python 7 /opt/csw/lib/python/multiprocessing 0 17561 python 7 /opt/csw/lib/python/multiprocessing/dummy 0 17561 python 7 /opt/csw/lib/python/plat-sunos5 0 17561 python 7 /opt/csw/lib/python/site-packages 0 17561 python 7 /opt/csw/lib/python/sqlite3 0 17561 python 7 /opt/csw/lib/python/sqlite3/test 0 17561 python 7 /opt/csw/lib/python/test 0 17561 python 7 /opt/csw/lib/python/test/decimaltestdata 0 17561 python 7 /opt/csw/lib/python/wsgiref 0 17561 python 7 /opt/csw/lib/python/xml 0 17561 python 7 /opt/csw/lib/python/xml/dom 0 17561 python 7 /opt/csw/lib/python/xml/etree 0 17561 python 7 /opt/csw/lib/python/xml/parsers 0 17561 python 7 /opt/csw/lib/python/xml/sax 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/__init__.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/error.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/expr.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/matcher.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/state.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/xtypes.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/__init__.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/error.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/expr.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/matcher.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/state.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/xtypes.py 0 17567 rm 11 python2.6 ^CBroken Pipe From phil at bolthole.com Sun May 31 23:07:04 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 31 May 2009 14:07:04 -0700 Subject: [csw-maintainers] Fwd: [csw-users] feedback on pkg-get behaviour change requested In-Reply-To: References: Message-ID: <20090531210704.GA55699@bolthole.com> On Sun, May 31, 2009 at 10:53:32PM +0200, Dagobert Michelsen wrote: > Hi, > > this was just on users at . > Gary: what do you think? I guess nobody was aware of this issue. oh. oops. I will change the behaviour of pkg-get, so that pkg-get -c pkgname acts differently than pkg-get -c >From the perspective of puppet, that will restore the original behaviour. From mwatters at opencsw.org Fri May 1 05:31:57 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 30 Apr 2009 22:31:57 -0500 Subject: [csw-maintainers] mediawiki now in testing Message-ID: <49FA6D2D.9020204@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 version 1.14.0 LaTeX math extension included in mediawiki_math package - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn6bS0ACgkQLrhmsXMSLxecOQCgpGEhzVuMZNsgnxayNN4ZquMn KTgAn1TP8TBa6/5FvD0pguU115YKhXbc =Cteq -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 1 08:13:16 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 30 Apr 2009 23:13:16 -0700 Subject: [csw-maintainers] mediawiki now in testing In-Reply-To: <49FA6D2D.9020204@opencsw.org> References: <49FA6D2D.9020204@opencsw.org> Message-ID: <20090501061316.GA16206@bolthole.com> On Thu, Apr 30, 2009 at 10:31:57PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > version 1.14.0 > LaTeX math extension included in mediawiki_math package errr... that sounds like an odd way to put it. do you mean, "extension for mediawiki included, that takes advantage of LaTeX math extension"? From dam at opencsw.org Fri May 1 15:12:44 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 1 May 2009 15:12:44 +0200 Subject: [csw-maintainers] Change of upstream watch in mGAR Message-ID: Hi, in the current implementation of check-upstream FILES2CHECK are defined with := invoking bin/upstream_watch. This means the definition is made and the binary called on each invocation of make, regardless of the target. With a limited internet connection this causes massive delays on the lftp call. I changed the implementation in r4633 to only calculate the names if necessary: Best regards -- Dago From Darin.Perusich at cognigencorp.com Fri May 1 15:30:35 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Fri, 01 May 2009 09:30:35 -0400 Subject: [csw-maintainers] relocatable on sparc but not x86 In-Reply-To: <20090430213405.GI79653@bolthole.com> References: <49FA0D7E.4060103@cognigencorp.com> <20090430213405.GI79653@bolthole.com> Message-ID: <49FAF97B.8040803@cognigencorp.com> I'm not asking how to fix it, I made the package non-relocatable, I'm curious to know what the cause might be. When I run staring against the sparc build it has various paths to 'gpg' listed and /opt/csw/bin happens to be one of them. These are not listed when I use strings on the i386 binary. Philip Brown wrote: > On Thu, Apr 30, 2009 at 04:43:42PM -0400, Darin Perusich wrote: >> I'm staging a package and on SPARC it's not relocatable but on i386 it's >> not, why would this be? >> > > that's kinda equivalent to saying, "it's broken. how can I fix it?" without saying > anything else :-/ > > > at minimum, you need to find out what bits are non-relocatable, and tell us > what they are... > -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From mwatters at opencsw.org Fri May 1 16:15:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 01 May 2009 09:15:44 -0500 Subject: [csw-maintainers] mediawiki now in testing In-Reply-To: <20090501061316.GA16206@bolthole.com> References: <49FA6D2D.9020204@opencsw.org> <20090501061316.GA16206@bolthole.com> Message-ID: <49FB0410.4030601@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Thu, Apr 30, 2009 at 10:31:57PM -0500, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> version 1.14.0 >> LaTeX math extension included in mediawiki_math package > > errr... that sounds like an odd way to put it. > > > do you mean, "extension for mediawiki included, that takes advantage > of LaTeX math extension"? > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers yes, I guess that is a better way to put it. This is the first iteration of the package, so I am sure I will have a few more pop through testing before release. I am still Tweaking the install (since it is normally a manual thing) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkn7BA8ACgkQLrhmsXMSLxd5pgCgzOg2yzGx6N2fj9bdnkmLIzLM sFwAoMUE0kg823ygoBOipQZ3kq/tkgYD =QjZP -----END PGP SIGNATURE----- From william at wbonnet.net Fri May 1 16:30:57 2009 From: william at wbonnet.net (William Bonnet) Date: Fri, 01 May 2009 16:30:57 +0200 Subject: [csw-maintainers] Change of upstream watch in mGAR In-Reply-To: References: Message-ID: <49FB07A1.2050100@wbonnet.net> Thanks a lot for this fix cheers W. Dagobert Michelsen a ?crit : > Hi, > > in the current implementation of check-upstream FILES2CHECK are defined > with := invoking bin/upstream_watch. This means the definition is made > and the binary called on each invocation of make, regardless of the > target. > With a limited internet connection this causes massive delays on the > lftp call. I changed the implementation in r4633 to only calculate > the names if necessary: > > > Best regards > > -- Dago From mwatters at opencsw.org Sun May 3 15:50:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 03 May 2009 08:50:22 -0500 Subject: [csw-maintainers] Fixing build8xt isaexec packages Message-ID: <49FDA11E.3040100@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I am re-installing the following packages on build8xt: CSWaudiofile CSWbzip2 CSWfconfig CSWflac CSWgcc4ada CSWgcc4adart CSWgcc4core CSWgcc4corert CSWgcc4g++ CSWgcc4g++rt CSWgcc4gfortran CSWgcc4gfortranrt CSWgcc4java CSWgcc4javart CSWgcc4objc CSWgcc4objcrt CSWgnupg CSWgzip CSWlibntlm CSWlibpq CSWlibtasn1 CSWlsof CSWmysql5 CSWpcre CSWpcrert CSWpostgresql CSWrsync CSWsetoolkit CSWtop CSWxmlstarlet I will let you know when complete ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkn9oR4ACgkQLrhmsXMSLxcYagCeIFprYQDzW2vnJJKPGmSsES2L MYQAn3JUHP1uSdIyrAPONCn8Q21bBnqA =B0HQ -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 3 17:25:39 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 03 May 2009 10:25:39 -0500 Subject: [csw-maintainers] Fixing build8xt isaexec packages In-Reply-To: <49FDA11E.3040100@opencsw.org> References: <49FDA11E.3040100@opencsw.org> Message-ID: <49FDB773.3010709@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > I am re-installing the following packages on build8xt: > CSWaudiofile > CSWbzip2 > CSWfconfig > CSWflac > CSWgcc4ada > CSWgcc4adart > CSWgcc4core > CSWgcc4corert > CSWgcc4g++ > CSWgcc4g++rt > CSWgcc4gfortran > CSWgcc4gfortranrt > CSWgcc4java > CSWgcc4javart > CSWgcc4objc > CSWgcc4objcrt > CSWgnupg > CSWgzip > CSWlibntlm > CSWlibpq > CSWlibtasn1 > CSWlsof > CSWmysql5 > CSWpcre > CSWpcrert > CSWpostgresql > CSWrsync > CSWsetoolkit > CSWtop > CSWxmlstarlet > > I will let you know when complete ;) > > _______________________________________________ maintainers mailing list maintainers at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/maintainers All packages re-installed, no version changes on any packages. on a side note. gcc4.3.3 is WORKING CORRECTLY on build8xt. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkn9t3MACgkQLrhmsXMSLxd+ZACgzxy+PiPGn+t90iJ2TvGpLXsl Ts8AnAj0D+Kt2uIJAdCZ9ZH7CEwmnjVr =8h5N -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 3 23:25:50 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 03 May 2009 16:25:50 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing Message-ID: <49FE0BDE.1010809@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 After Weeks of Beating my head against the wall on the custom merge. with many thanks to the build-farm maintainers 'specially Ben and Dago for putting up with my huge package installs one or more times / day. I repackaged to remove la~ backup files from the manual merge. installed without error on build8xt. I have successfully compiled/tested vim, libffi, and binutils on build8xt. I will do a sanity test on sparc tomorrow. Does anyone have a solaris 10 x86 server to test amd64 compiles? ... mine became a centOS5.3 x86_64 server. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn+C94ACgkQLrhmsXMSLxea0QCcCzzCANyK9DX/vZ/49kRbFKvw GIEAn3dF0sZjsvdJSOruwSfEUcqWTu83 =SXOs -----END PGP SIGNATURE----- From bwalton at opencsw.org Mon May 4 22:15:17 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 04 May 2009 16:15:17 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <49FE0BDE.1010809@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> Message-ID: <1241468034-sup-3126@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Sun May 03 17:25:50 -0400 2009: Hi Mike, > Does anyone have a solaris 10 x86 server to test amd64 compiles? > ... mine became a centOS5.3 x86_64 server. I'm updating the gcc and g++ packages on a sol10/x86 box right now. I'll build the same packages you did and let you know. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 03:07:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 04 May 2009 20:07:44 -0500 Subject: [csw-maintainers] squid now in testing Message-ID: <49FF9160.1040108@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: * Updated to version 2.7-STABLE6 * significant changes in postinstall preremove scripts * smf support on solaris10 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn/kWAACgkQLrhmsXMSLxfHGgCgmj5xoHBNsl2PyRk98lq5Ja2Y RSkAn3R9jB/1BYTfS9mex2hto1LR8p12 =IyCD -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue May 5 05:05:32 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 04 May 2009 22:05:32 -0500 Subject: [csw-maintainers] repackage squid in testing Message-ID: <49FFACFC.3050809@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: enabled more compile time options including bug http://www.opencsw.org/mantis/view.php?id=2858 full list of enabled options: - --enable-arp-acl - --enable-auth=basic - --enable-basic-auth-helpers=LDAP,SMB,YP,PAM,SASL,NCSA - --enable-cache-digests - --enable-carp - --enable-delay-pools - --enable-devpoll - --enable-forw-via-db - --enable-htcp - --enable-icmp - --enable-large-cache-files - --enable-referer-log - --enable-removal-policies=heap,lru #address bug 2858 - --enable-select - --enable-snmp - --enable-ssl - --enable-useragent-log - --with-dl - --with-large-files - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkn/rPsACgkQLrhmsXMSLxfHxQCfVDFnTGWfGIeHfMuWcJpvXv5z DN0AnjgoEevQNgfolbL8Zk/1GddugHeP =QfT7 -----END PGP SIGNATURE----- From dam at opencsw.org Tue May 5 15:38:06 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 5 May 2009 15:38:06 +0200 Subject: [csw-maintainers] I am running in circles around GNOME Message-ID: <07EC02B3-2504-4902-BD97-89E4B9273C62@opencsw.org> Hi, after the great preparation of William and compiling loads of packages I hit a roadblock here where I need some advice: - pango needs an update cairo - cairo needs an updated pixman - pixman needs an updated pangocairo from pango Is there some magic to break the circle? Best regards -- Dago libtool: link: mv -f ".libs/libpangocairo-1.0.expT" ".libs/ libpangocairo-1.0.exp" libtool: link: echo "{ global:" > .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: cat .libs/libpangocairo-1.0.exp | /opt/csw/bin/gsed -e "s/\(.*\)/\1;/" >> .libs/libpangocairo-1.0.so.0.2400.2.exp libtool: link: echo "local: *; };" >> .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -M .libs/ libpangocairo-1.0.so.0.2400.2.exp -h libpangocairo-1.0.so.0 -o .libs/ libpangocairo-1.0.so.0.2400.2 .libs/libpangocairo_1_0_la-pangocairo- context.o .libs/libpangocairo_1_0_la-pangocairo-font.o .libs/ libpangocairo_1_0_la-pangocairo-fontmap.o .libs/libpangocairo_1_0_la- pangocairo-render.o .libs/libpangocairo_1_0_la-pangocairo- fcfont.o .libs/libpangocairo_1_0_la-pangocairo-fcfontmap.o -R/home/ dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/ pango/.libs -R/opt/csw/lib/64 -L/home/dam/mgar/pkg/pango/trunk/work/ build-isa-sparcv9/pango-1.24.2/pango/.libs -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 ./.libs/libpango-1.0.so -L/opt/csw/lib -lcairo ./.libs/ libpangoft2-1.0.so /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so -lgobject-2.0 - lgmodule-2.0 -lglib-2.0 -lintl -lm -lfreetype -lfontconfig -lc - xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/libcairo.so: wrong ELF class: ELFCLASS32 ld: warning: file /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so: linked to ./.libs/ libpango-1.0.so: attempted multiple inclusion of file ld: fatal: File processing errors. No output written to .libs/ libpangocairo-1.0.so.0.2400.2 gmake[6]: *** [libpangocairo-1.0.la] Error 1 gmake[6]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[5]: *** [all-recursive] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[1]: *** [build-work/build-isa-sparcv9/pango-1.24.2/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pango/trunk' gmake: *** [merge-isa-sparcv9] Error 2 /bin/bash ../libtool --tag=CC --mode=link /opt/studio/SOS11/SUNWspro/ bin/cc -xO3 -xarch=v9 -I/opt/csw/X11/include -I/opt/csw/include - D_POSIX_PTHREAD_SEMANTICS -version-info 10802:6:10800 -no-undefined - xarch=v9 -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 -o libcairo.la - rpath /opt/csw/lib/64 cairo- analysis-surface.lo cairo-arc.lo cairo-array.lo cairo-atomic.lo cairo- base85-stream.lo cairo-bentley-ottmann.lo cairo.lo cairo-cache.lo cairo-clip.lo cairo-color.lo cairo-debug.lo cairo-fixed.lo cairo-font- face.lo cairo-font-face-twin.lo cairo-font-face-twin-data.lo cairo- font-options.lo cairo-freelist.lo cairo-gstate.lo cairo-hash.lo cairo- hull.lo cairo-image-surface.lo cairo-lzw.lo cairo-matrix.lo cairo-meta- surface.lo cairo-misc.lo cairo-mutex.lo cairo-output-stream.lo cairo- paginated-surface.lo cairo-path-bounds.lo cairo-path.lo cairo-path- fill.lo cairo-path-fixed.lo cairo-path-stroke.lo cairo-pattern.lo cairo-pen.lo cairo-polygon.lo cairo-rectangle.lo cairo-region.lo cairo- scaled-font.lo cairo-skiplist.lo cairo-slope.lo cairo-spline.lo cairo- stroke-style.lo cairo-surface.lo cairo-surface-fallback.lo cairo- system.lo cairo-traps.lo cairo-unicode.lo cairo-user-font.lo cairo- version.lo cairo-wideint.lo cairo-cff-subset.lo cairo-scaled-font- subsets.lo cairo-truetype-subset.lo cairo-type1-fallback.lo cairo- type1-subset.lo cairo-type3-glyph-surface.lo cairo-pdf-operators.lo cairo-xlib-display.lo cairo-xlib-screen.lo cairo-xlib-surface.lo cairo- xlib-visual.lo cairo-png.lo cairo-ft-font.lo cairo-ps- surface.lo cairo-pdf-surface.lo cairo-deflate-stream.lo cairo-svg- surface.lo test-fallback-surface.lo test-meta-surface.lo test- paginated-surface.lo -L/opt/csw/lib -lpixman-1 -L/opt/csw/lib - lfreetype -L/opt/csw/lib -lfontconfig -L/opt/csw/lib -lpng12 - lm -lz -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 - lsocket -lnsl -lm -lm libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -z defs -h libcairo.so.2 -o .libs/libcairo.so.2.10800.6 .libs/cairo-analysis- surface.o .libs/cairo-arc.o .libs/cairo-array.o .libs/cairo- atomic.o .libs/cairo-base85-stream.o .libs/cairo-bentley- ottmann.o .libs/cairo.o .libs/cairo-cache.o .libs/cairo-clip.o .libs/ cairo-color.o .libs/cairo-debug.o .libs/cairo-fixed.o .libs/cairo-font- face.o .libs/cairo-font-face-twin.o .libs/cairo-font-face-twin- data.o .libs/cairo-font-options.o .libs/cairo-freelist.o .libs/cairo- gstate.o .libs/cairo-hash.o .libs/cairo-hull.o .libs/cairo-image- surface.o .libs/cairo-lzw.o .libs/cairo-matrix.o .libs/cairo-meta- surface.o .libs/cairo-misc.o .libs/cairo-mutex.o .libs/cairo-output- stream.o .libs/cairo-paginated-surface.o .libs/cairo-path- bounds.o .libs/cairo-path.o .libs/cairo-path-fill.o .libs/cairo-path- fixed.o .libs/cairo-path-stroke.o .libs/cairo-pattern.o .libs/cairo- pen.o .libs/cairo-polygon.o .libs/cairo-rectangle.o .libs/cairo- region.o .libs/cairo-scaled-font.o .libs/cairo-skiplist.o .libs/cairo- slope.o .libs/cairo-spline.o .libs/cairo-stroke-style.o .libs/cairo- surface.o .libs/cairo-surface-fallback.o .libs/cairo-system.o .libs/ cairo-traps.o .libs/cairo-unicode.o .libs/cairo-user-font.o .libs/ cairo-version.o .libs/cairo-wideint.o .libs/cairo-cff-subset.o .libs/ cairo-scaled-font-subsets.o .libs/cairo-truetype-subset.o .libs/cairo- type1-fallback.o .libs/cairo-type1-subset.o .libs/cairo-type3-glyph- surface.o .libs/cairo-pdf-operators.o .libs/cairo-xlib-display.o .libs/ cairo-xlib-screen.o .libs/cairo-xlib-surface.o .libs/cairo-xlib- visual.o .libs/cairo-png.o .libs/cairo-ft-font.o .libs/cairo-ps- surface.o .libs/cairo-pdf-surface.o .libs/cairo-deflate-stream.o .libs/ cairo-svg-surface.o .libs/test-fallback-surface.o .libs/test-meta- surface.o .libs/test-paginated-surface.o -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 -L/opt/csw/lib -lpixman-1 -lfreetype -lfontconfig - lpng12 -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 -lsocket - lnsl -lm -lc -xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/64/libpixman-1.so: wrong ELF class: ELFCLASS32 ld: fatal: File processing errors. No output written to .libs/ libcairo.so.2.10800.6 gmake[5]: *** [libcairo.la] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[1]: *** [build-work/build-isa-sparcv9/cairo-1.8.6/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk' gmake: *** [merge-isa-sparcv9] Error 2 checking dynamic linker characteristics... solaris2.8 ld.so (cached) (cached) checking how to hardcode library paths into programs... immediate checking for getisax... no checking whether byte ordering is bigendian... yes checking for inline... inline checking whether __SUNPRO_C is declared... yes checking whether __amd64 is declared... no checking for perl... /opt/csw/bin/perl checking for -fvisibility... no checking for -xldscope (Sun compilers)... yes checking whether to use MMX intrinsics... no checking whether to use SSE2 intrinsics... no checking whether to use VMX/Altivec intrinsics... no checking whether to use ARM SIMD assembler... no checking for pkg-config... /opt/csw/bin/pkg-config checking pkg-config is at least version 0.9.0... yes checking for GTK... configure: error: Package requirements (gtk+-2.0) were not met: sh: gnome-config: not found Package pangocairo was not found in the pkg-config search path. Perhaps you should add the directory containing `pangocairo.pc' to the PKG_CONFIG_PATH environment variable Package 'pangocairo', required by 'GDK', not found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables GTK_CFLAGS and GTK_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. gmake[1]: *** [configure-work/build-isa-sparcv8/pixman-0.15.2/ configure] Error 1 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pixman/trunk' gmake: *** [merge-isa-sparcv8] Error 2 From dam at opencsw.org Tue May 5 15:45:27 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 5 May 2009 15:45:27 +0200 Subject: [csw-maintainers] I am running in circles around GNOME Message-ID: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> Hi, after the great preparation of William and compiling loads of packages I hit a roadblock here where I need some advice: - pango needs an update cairo - cairo needs an updated pixman - pixman needs an updated pangocairo from pango Is there some magic to break the circle? Best regards -- Dago libtool: link: mv -f ".libs/libpangocairo-1.0.expT" ".libs/ libpangocairo-1.0.exp" libtool: link: echo "{ global:" > .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: cat .libs/libpangocairo-1.0.exp | /opt/csw/bin/gsed -e "s/\(.*\)/\1;/" >> .libs/libpangocairo-1.0.so.0.2400.2.exp libtool: link: echo "local: *; };" >> .libs/libpangocairo-1.0.so. 0.2400.2.exp libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -M .libs/ libpangocairo-1.0.so.0.2400.2.exp -h libpangocairo-1.0.so.0 -o .libs/ libpangocairo-1.0.so.0.2400.2 .libs/libpangocairo_1_0_la-pangocairo- context.o .libs/libpangocairo_1_0_la-pangocairo-font.o .libs/ libpangocairo_1_0_la-pangocairo-fontmap.o .libs/libpangocairo_1_0_la- pangocairo-render.o .libs/libpangocairo_1_0_la-pangocairo- fcfont.o .libs/libpangocairo_1_0_la-pangocairo-fcfontmap.o -R/home/ dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/ pango/.libs -R/opt/csw/lib/64 -L/home/dam/mgar/pkg/pango/trunk/work/ build-isa-sparcv9/pango-1.24.2/pango/.libs -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 ./.libs/libpango-1.0.so -L/opt/csw/lib -lcairo ./.libs/ libpangoft2-1.0.so /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so -lgobject-2.0 - lgmodule-2.0 -lglib-2.0 -lintl -lm -lfreetype -lfontconfig -lc - xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/libcairo.so: wrong ELF class: ELFCLASS32 ld: warning: file /home/dam/mgar/pkg/pango/trunk/work/build-isa- sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so: linked to ./.libs/ libpango-1.0.so: attempted multiple inclusion of file ld: fatal: File processing errors. No output written to .libs/ libpangocairo-1.0.so.0.2400.2 gmake[6]: *** [libpangocairo-1.0.la] Error 1 gmake[6]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[5]: *** [all-recursive] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2/pango' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/pango/trunk/work/build- isa-sparcv9/pango-1.24.2' gmake[1]: *** [build-work/build-isa-sparcv9/pango-1.24.2/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pango/trunk' gmake: *** [merge-isa-sparcv9] Error 2 /bin/bash ../libtool --tag=CC --mode=link /opt/studio/SOS11/SUNWspro/ bin/cc -xO3 -xarch=v9 -I/opt/csw/X11/include -I/opt/csw/include - D_POSIX_PTHREAD_SEMANTICS -version-info 10802:6:10800 -no-undefined - xarch=v9 -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 -o libcairo.la - rpath /opt/csw/lib/64 cairo- analysis-surface.lo cairo-arc.lo cairo-array.lo cairo-atomic.lo cairo- base85-stream.lo cairo-bentley-ottmann.lo cairo.lo cairo-cache.lo cairo-clip.lo cairo-color.lo cairo-debug.lo cairo-fixed.lo cairo-font- face.lo cairo-font-face-twin.lo cairo-font-face-twin-data.lo cairo- font-options.lo cairo-freelist.lo cairo-gstate.lo cairo-hash.lo cairo- hull.lo cairo-image-surface.lo cairo-lzw.lo cairo-matrix.lo cairo-meta- surface.lo cairo-misc.lo cairo-mutex.lo cairo-output-stream.lo cairo- paginated-surface.lo cairo-path-bounds.lo cairo-path.lo cairo-path- fill.lo cairo-path-fixed.lo cairo-path-stroke.lo cairo-pattern.lo cairo-pen.lo cairo-polygon.lo cairo-rectangle.lo cairo-region.lo cairo- scaled-font.lo cairo-skiplist.lo cairo-slope.lo cairo-spline.lo cairo- stroke-style.lo cairo-surface.lo cairo-surface-fallback.lo cairo- system.lo cairo-traps.lo cairo-unicode.lo cairo-user-font.lo cairo- version.lo cairo-wideint.lo cairo-cff-subset.lo cairo-scaled-font- subsets.lo cairo-truetype-subset.lo cairo-type1-fallback.lo cairo- type1-subset.lo cairo-type3-glyph-surface.lo cairo-pdf-operators.lo cairo-xlib-display.lo cairo-xlib-screen.lo cairo-xlib-surface.lo cairo- xlib-visual.lo cairo-png.lo cairo-ft-font.lo cairo-ps- surface.lo cairo-pdf-surface.lo cairo-deflate-stream.lo cairo-svg- surface.lo test-fallback-surface.lo test-meta-surface.lo test- paginated-surface.lo -L/opt/csw/lib -lpixman-1 -L/opt/csw/lib - lfreetype -L/opt/csw/lib -lfontconfig -L/opt/csw/lib -lpng12 - lm -lz -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 - lsocket -lnsl -lm -lm libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -z defs -h libcairo.so.2 -o .libs/libcairo.so.2.10800.6 .libs/cairo-analysis- surface.o .libs/cairo-arc.o .libs/cairo-array.o .libs/cairo- atomic.o .libs/cairo-base85-stream.o .libs/cairo-bentley- ottmann.o .libs/cairo.o .libs/cairo-cache.o .libs/cairo-clip.o .libs/ cairo-color.o .libs/cairo-debug.o .libs/cairo-fixed.o .libs/cairo-font- face.o .libs/cairo-font-face-twin.o .libs/cairo-font-face-twin- data.o .libs/cairo-font-options.o .libs/cairo-freelist.o .libs/cairo- gstate.o .libs/cairo-hash.o .libs/cairo-hull.o .libs/cairo-image- surface.o .libs/cairo-lzw.o .libs/cairo-matrix.o .libs/cairo-meta- surface.o .libs/cairo-misc.o .libs/cairo-mutex.o .libs/cairo-output- stream.o .libs/cairo-paginated-surface.o .libs/cairo-path- bounds.o .libs/cairo-path.o .libs/cairo-path-fill.o .libs/cairo-path- fixed.o .libs/cairo-path-stroke.o .libs/cairo-pattern.o .libs/cairo- pen.o .libs/cairo-polygon.o .libs/cairo-rectangle.o .libs/cairo- region.o .libs/cairo-scaled-font.o .libs/cairo-skiplist.o .libs/cairo- slope.o .libs/cairo-spline.o .libs/cairo-stroke-style.o .libs/cairo- surface.o .libs/cairo-surface-fallback.o .libs/cairo-system.o .libs/ cairo-traps.o .libs/cairo-unicode.o .libs/cairo-user-font.o .libs/ cairo-version.o .libs/cairo-wideint.o .libs/cairo-cff-subset.o .libs/ cairo-scaled-font-subsets.o .libs/cairo-truetype-subset.o .libs/cairo- type1-fallback.o .libs/cairo-type1-subset.o .libs/cairo-type3-glyph- surface.o .libs/cairo-pdf-operators.o .libs/cairo-xlib-display.o .libs/ cairo-xlib-screen.o .libs/cairo-xlib-surface.o .libs/cairo-xlib- visual.o .libs/cairo-png.o .libs/cairo-ft-font.o .libs/cairo-ps- surface.o .libs/cairo-pdf-surface.o .libs/cairo-deflate-stream.o .libs/ cairo-svg-surface.o .libs/test-fallback-surface.o .libs/test-meta- surface.o .libs/test-paginated-surface.o -L/opt/csw/lib/64 -L/opt/ csw/X11/lib/64 -L/opt/csw/lib -lpixman-1 -lfreetype -lfontconfig - lpng12 -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 -lsocket - lnsl -lm -lc -xarch=v9 -xarch=v9 ld: fatal: file /opt/csw/lib/64/libpixman-1.so: wrong ELF class: ELFCLASS32 ld: fatal: File processing errors. No output written to .libs/ libcairo.so.2.10800.6 gmake[5]: *** [libcairo.la] Error 1 gmake[5]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[4]: *** [all] Error 2 gmake[4]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6/src' gmake[3]: *** [all-recursive] Error 1 gmake[3]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk/work/ build-isa-sparcv9/cairo-1.8.6' gmake[1]: *** [build-work/build-isa-sparcv9/cairo-1.8.6/Makefile] Error 2 gmake[1]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk' gmake: *** [merge-isa-sparcv9] Error 2 checking dynamic linker characteristics... solaris2.8 ld.so (cached) (cached) checking how to hardcode library paths into programs... immediate checking for getisax... no checking whether byte ordering is bigendian... yes checking for inline... inline checking whether __SUNPRO_C is declared... yes checking whether __amd64 is declared... no checking for perl... /opt/csw/bin/perl checking for -fvisibility... no checking for -xldscope (Sun compilers)... yes checking whether to use MMX intrinsics... no checking whether to use SSE2 intrinsics... no checking whether to use VMX/Altivec intrinsics... no checking whether to use ARM SIMD assembler... no checking for pkg-config... /opt/csw/bin/pkg-config checking pkg-config is at least version 0.9.0... yes checking for GTK... configure: error: Package requirements (gtk+-2.0) were not met: sh: gnome-config: not found Package pangocairo was not found in the pkg-config search path. Perhaps you should add the directory containing `pangocairo.pc' to the PKG_CONFIG_PATH environment variable Package 'pangocairo', required by 'GDK', not found Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables GTK_CFLAGS and GTK_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details. gmake[1]: *** [configure-work/build-isa-sparcv8/pixman-0.15.2/ configure] Error 1 gmake[1]: Leaving directory `/home/dam/mgar/pkg/pixman/trunk' gmake: *** [merge-isa-sparcv8] Error 2 From bonivart at opencsw.org Tue May 5 18:00:10 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 5 May 2009 18:00:10 +0200 Subject: [csw-maintainers] CSWpymysql listed twice in catalog Message-ID: <625385e30905050900o4b32147fx97a258e5766093@mail.gmail.com> pymysql 1.2.2,REV=2009.04.02 CSWpymysql pymysql-1.2.2,REV=2009.04.02-SunOS5.8-sparc-CSW.pkg.gz 3e46259ed56d5986008192c4a9326436 61565 CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none pymysql5 1.2.2,REV=2009.03.11 CSWpymysql pymysql5-1.2.2,REV=2009.03.11-SunOS5.8-sparc-CSW.pkg.gz f9d234764be352370ac1e83a2edaeb88 64970 CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none I assume the second one should be removed. -- /peter From phil at bolthole.com Tue May 5 18:24:54 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 5 May 2009 09:24:54 -0700 Subject: [csw-maintainers] CSWpymysql listed twice in catalog In-Reply-To: <625385e30905050900o4b32147fx97a258e5766093@mail.gmail.com> References: <625385e30905050900o4b32147fx97a258e5766093@mail.gmail.com> Message-ID: <20090505162454.GA36027@bolthole.com> On Tue, May 05, 2009 at 06:00:10PM +0200, Peter Bonivart wrote: > pymysql 1.2.2,REV=2009.04.02 CSWpymysql > pymysql-1.2.2,REV=2009.04.02-SunOS5.8-sparc-CSW.pkg.gz > 3e46259ed56d5986008192c4a9326436 61565 > CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none > > pymysql5 1.2.2,REV=2009.03.11 CSWpymysql > pymysql5-1.2.2,REV=2009.03.11-SunOS5.8-sparc-CSW.pkg.gz > f9d234764be352370ac1e83a2edaeb88 64970 > CSWmysql5rt|CSWpython|CSWosslrt|CSWzlib|CSWcommon none > > I assume the second one should be removed. thanks for noticing that. kept pymysql. deleted pymysql5 From bwalton at opencsw.org Tue May 5 19:49:22 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 13:49:22 -0400 Subject: [csw-maintainers] php5 packaging Message-ID: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Hi All, I updated a box that uses php5 today, so I got the new versions that were just released. After bringing the webserver back online, the hosted app wasn't responding with anything but the logs showed the requested being handled. Turns out that since many of the modules that used to be part of the core php5 package (session being the one that got me) are now separate packages, I simply needed to pull them in and I was back up and running. I was just wondering if there is a better way to handle situations like this...a user doing a 1:1 update of packages will have less functionality after the update than they did before. I think the changes are ok overall, but the user interaction may not be. Thoughts? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Tue May 5 20:09:07 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 05 May 2009 20:09:07 +0200 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241545460-sup-965@ntdws12.chass.utoronto.ca> (Ben Walton's message of "Tue\, 05 May 2009 13\:49\:22 -0400") References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: Ben Walton writes: > I was just wondering if there is a better way to handle situations > like this...a user doing a 1:1 update of packages will have less > functionality after the update than they did before. I think the > changes are ok overall, but the user interaction may not be. Using transition packages? A transition package serves as a temporary package, offering 1:1 content, having the same name as the one transitioned. Of course, after the transition, the package disappears or has a lesser surface. -- Peter From bwalton at opencsw.org Tue May 5 20:17:48 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 14:17:48 -0400 Subject: [csw-maintainers] php5 packaging In-Reply-To: References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: <1241547276-sup-5004@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Tue May 05 14:09:07 -0400 2009: > Using transition packages? A transition package serves as a temporary > package, offering 1:1 content, having the same name as the one > transitioned. Of course, after the transition, the package disappears or > has a lesser surface. So have an empty package named php5 that depends on php5core, php5session, etc? Then stop updating php5 at some point in the future? That works, but leaves the dangling (although harmless) php5 empty package. Is there a way to handle it other than this? I like the extra granularity that Mike introduced, but if users break their setups (or waste time sorting it out, anyway), they may not be very happy. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 20:19:01 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 13:19:01 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241545460-sup-965@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: <4A008315.70203@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Hi All, > > I updated a box that uses php5 today, so I got the new versions that > were just released. After bringing the webserver back online, the > hosted app wasn't responding with anything but the logs showed the > requested being handled. > > Turns out that since many of the modules that used to be part of the > core php5 package (session being the one that got me) are now separate > packages, I simply needed to pull them in and I was back up and > running. > > I was just wondering if there is a better way to handle situations > like this...a user doing a 1:1 update of packages will have less > functionality after the update than they did before. I think the > changes are ok overall, but the user interaction may not be. > > Thoughts? > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I didn't think about that part of the upgrade. I wrote admin scripts to gracefully handle the php.ini files. The new packages do not overwrite the existing php.ini. There are a couple ways to handle this I think. examine the existing php.ini file for enabled extensions: 1) kick off a pkg-get -i for each extension that is enabled. 2) display a note to the user to add the extra packages themselves. This does not account for anything that was previously compiled statically I am not sure if there were any. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAgxUACgkQLrhmsXMSLxdrKwCeLejN1qO1q2DjHl900VI1tD66 FgEAnR5LcHsyslNKgzfy8U85Qq7GK8m3 =8HRD -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue May 5 20:30:06 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 13:30:06 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <4A008315.70203@opencsw.org> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> Message-ID: <4A0085AE.2080807@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > Ben Walton wrote: >> Hi All, > >> I updated a box that uses php5 today, so I got the new versions that >> were just released. After bringing the webserver back online, the >> hosted app wasn't responding with anything but the logs showed the >> requested being handled. > >> Turns out that since many of the modules that used to be part of the >> core php5 package (session being the one that got me) are now separate >> packages, I simply needed to pull them in and I was back up and >> running. > >> I was just wondering if there is a better way to handle situations >> like this...a user doing a 1:1 update of packages will have less >> functionality after the update than they did before. I think the >> changes are ok overall, but the user interaction may not be. > >> Thoughts? >> -Ben > > >> ------------------------------------------------------------------------ > >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > I didn't think about that part of the upgrade. > I wrote admin scripts to gracefully handle the php.ini files. > The new packages do not overwrite the existing php.ini. > > There are a couple ways to handle this I think. > > examine the existing php.ini file for enabled extensions: > > 1) kick off a pkg-get -i for each extension that is enabled. > 2) display a note to the user to add the extra packages themselves. > > This does not account for anything that was previously compiled > statically I am not sure if there were any. > I will start by definitely changing the existing php.ini warning to let the user's know of the new granularity. does anyone have a preference on the manual vs automatic? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAha4ACgkQLrhmsXMSLxdH+ACfSqaYjyCIloMHjjefSQ3Vli91 kfkAoJ1mt9aqkU4/pZ+B4wY54ytNTK/I =BdcX -----END PGP SIGNATURE----- From bwalton at opencsw.org Tue May 5 20:52:45 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 14:52:45 -0400 Subject: [csw-maintainers] php5 packaging In-Reply-To: <4A008315.70203@opencsw.org> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> Message-ID: <1241549280-sup-7549@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Tue May 05 14:19:01 -0400 2009: > I didn't think about that part of the upgrade. > I wrote admin scripts to gracefully handle the php.ini files. > The new packages do not overwrite the existing php.ini. > > There are a couple ways to handle this I think. > > examine the existing php.ini file for enabled extensions: > > 1) kick off a pkg-get -i for each extension that is enabled. This gets troublesome in the face of locally added extensions (we use oracle here, for example)....not impossible, but you'd maybe need to whitelist extensions against those available from the catalog or something. > 2) display a note to the user to add the extra packages themselves. What about a combination of these two? Provide a php5-update-extensions script or something that does 1 and note it at package install time? This could still get lost in a sea of output though, so it might not be sufficient. > This does not account for anything that was previously compiled > statically I am not sure if there were any. I'm not sure either. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 21:13:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 14:13:24 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241549280-sup-7549@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> <1241549280-sup-7549@ntdws12.chass.utoronto.ca> Message-ID: <4A008FD4.9040606@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Mike Watters's message of Tue May 05 14:19:01 -0400 2009: >> I didn't think about that part of the upgrade. >> I wrote admin scripts to gracefully handle the php.ini files. >> The new packages do not overwrite the existing php.ini. >> >> There are a couple ways to handle this I think. >> >> examine the existing php.ini file for enabled extensions: >> >> 1) kick off a pkg-get -i for each extension that is enabled. > > This gets troublesome in the face of locally added extensions (we use > oracle here, for example)....not impossible, but you'd maybe need to > whitelist extensions against those available from the catalog or > something. > >> 2) display a note to the user to add the extra packages themselves. > > What about a combination of these two? Provide a > php5-update-extensions script or something that does 1 and note it at > package install time? This could still get lost in a sea of output > though, so it might not be sufficient. > >> This does not account for anything that was previously compiled >> statically I am not sure if there were any. > > I'm not sure either. > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Here is what I have come up with thus far, Please review and let me know what you think. http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/php5/trunk/files/CSWphp5.postinstall?revision=4695&view=markup - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAj9QACgkQLrhmsXMSLxdoDACg5jWyG0LnnLl2F45aFDdaXYBb coAAn2oaZEy/8dKnc9OGAyxVP59w6Tzs =4IPo -----END PGP SIGNATURE----- From mwatters at opencsw.org Tue May 5 21:15:09 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 14:15:09 -0500 Subject: [csw-maintainers] squidclamav 4.0 now in testing Message-ID: <4A00903D.4090900@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: new package Once I get done tweaking the post-install scripts for CSWsquid, I will start Tweaking them for CSWsquidclamav. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAkD0ACgkQLrhmsXMSLxcErACfYxewviGLZmE8LtoOjNhYNj+h JgEAn3gqs8W8fWTCNzYZxpLdL0PnjY89 =8ELu -----END PGP SIGNATURE----- From phil at bolthole.com Tue May 5 21:44:17 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 5 May 2009 12:44:17 -0700 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241545460-sup-965@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> Message-ID: <20090505194417.GE36027@bolthole.com> On Tue, May 05, 2009 at 01:49:22PM -0400, Ben Walton wrote: > I was just wondering if there is a better way to handle situations > like this...a user doing a 1:1 update of packages will have less > functionality after the update than they did before. I think the > changes are ok overall, but the user interaction may not be. > > Thoughts? > -Ben Reguardless of what we decide here: Mike, you need to write (separately sent) emails to both the users, and the announce, list, warning people about this right away. From bwalton at opencsw.org Tue May 5 21:49:17 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 05 May 2009 15:49:17 -0400 Subject: [csw-maintainers] php5 packaging In-Reply-To: <4A008FD4.9040606@opencsw.org> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> <1241549280-sup-7549@ntdws12.chass.utoronto.ca> <4A008FD4.9040606@opencsw.org> Message-ID: <1241552758-sup-6316@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Tue May 05 15:13:24 -0400 2009: > Here is what I have come up with thus far, Please review and let me > know what you think. > http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/php5/trunk/files/CSWphp5.postinstall?revision=4695&view=markup Looks good. Still possible to miss in the output of a large update, but hopefully it will suffice. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Tue May 5 21:51:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 14:51:03 -0500 Subject: [csw-maintainers] php5 packaging In-Reply-To: <1241552758-sup-6316@ntdws12.chass.utoronto.ca> References: <1241545460-sup-965@ntdws12.chass.utoronto.ca> <4A008315.70203@opencsw.org> <1241549280-sup-7549@ntdws12.chass.utoronto.ca> <4A008FD4.9040606@opencsw.org> <1241552758-sup-6316@ntdws12.chass.utoronto.ca> Message-ID: <4A0098A7.8030300@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Mike Watters's message of Tue May 05 15:13:24 -0400 2009: > >> Here is what I have come up with thus far, Please review and let me >> know what you think. >> http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/php5/trunk/files/CSWphp5.postinstall?revision=4695&view=markup > > Looks good. Still possible to miss in the output of a large update, > but hopefully it will suffice. > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers OK, I am writing the emails to users and announce, I will package that change and get it out. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoAmKYACgkQLrhmsXMSLxeR+ACgpZqA8ikDExeTNmR7fE5MkNQy OS0AoLVp544Cqpd8nAl8kQFdMsZ0oPNK =wZkL -----END PGP SIGNATURE----- From harpchad at opencsw.org Tue May 5 22:37:41 2009 From: harpchad at opencsw.org (Chad Harp) Date: Tue, 05 May 2009 15:37:41 -0500 Subject: [csw-maintainers] I am running in circles around GNOME In-Reply-To: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> References: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> Message-ID: <4A00A395.1030500@opencsw.org> I believe from the searching I did (few months ago) that the pango dependency (via gtk) in pixman is just for some of the make check tests. Try building pixman with --disable-gtk Dagobert Michelsen wrote: > Hi, > > after the great preparation of William and compiling loads > of packages I hit a roadblock here where I need some advice: > > - pango needs an update cairo > - cairo needs an updated pixman > - pixman needs an updated pangocairo from pango > > Is there some magic to break the circle? > > > Best regards > > -- Dago > > > > libtool: link: mv -f ".libs/libpangocairo-1.0.expT" > ".libs/libpangocairo-1.0.exp" > libtool: link: echo "{ global:" > .libs/libpangocairo-1.0.so.0.2400.2.exp > libtool: link: cat .libs/libpangocairo-1.0.exp | /opt/csw/bin/gsed -e > "s/\(.*\)/\1;/" >> .libs/libpangocairo-1.0.so.0.2400.2.exp > libtool: link: echo "local: *; };" >> > .libs/libpangocairo-1.0.so.0.2400.2.exp > libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -M > .libs/libpangocairo-1.0.so.0.2400.2.exp -h libpangocairo-1.0.so.0 -o > .libs/libpangocairo-1.0.so.0.2400.2 > .libs/libpangocairo_1_0_la-pangocairo-context.o > .libs/libpangocairo_1_0_la-pangocairo-font.o > .libs/libpangocairo_1_0_la-pangocairo-fontmap.o > .libs/libpangocairo_1_0_la-pangocairo-render.o > .libs/libpangocairo_1_0_la-pangocairo-fcfont.o > .libs/libpangocairo_1_0_la-pangocairo-fcfontmap.o > -R/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs > -R/opt/csw/lib/64 > -L/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs > -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 ./.libs/libpango-1.0.so > -L/opt/csw/lib -lcairo ./.libs/libpangoft2-1.0.so > /home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so > -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lm -lfreetype > -lfontconfig -lc -xarch=v9 -xarch=v9 > ld: fatal: file /opt/csw/lib/libcairo.so: wrong ELF class: ELFCLASS32 > ld: warning: file > /home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango/.libs/libpango-1.0.so: > linked to ./.libs/libpango-1.0.so: attempted multiple inclusion of file > ld: fatal: File processing errors. No output written to > .libs/libpangocairo-1.0.so.0.2400.2 > gmake[6]: *** [libpangocairo-1.0.la] Error 1 > gmake[6]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango' > gmake[5]: *** [all-recursive] Error 1 > gmake[5]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango' > gmake[4]: *** [all] Error 2 > gmake[4]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2/pango' > gmake[3]: *** [all-recursive] Error 1 > gmake[3]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2' > gmake[2]: *** [all] Error 2 > gmake[2]: Leaving directory > `/home/dam/mgar/pkg/pango/trunk/work/build-isa-sparcv9/pango-1.24.2' > gmake[1]: *** [build-work/build-isa-sparcv9/pango-1.24.2/Makefile] Error 2 > gmake[1]: Leaving directory `/home/dam/mgar/pkg/pango/trunk' > gmake: *** [merge-isa-sparcv9] Error 2 > > > > > > /bin/bash ../libtool --tag=CC --mode=link > /opt/studio/SOS11/SUNWspro/bin/cc -xO3 -xarch=v9 -I/opt/csw/X11/include > -I/opt/csw/include -D_POSIX_PTHREAD_SEMANTICS -version-info > 10802:6:10800 -no-undefined -xarch=v9 -L/opt/csw/lib/64 > -L/opt/csw/X11/lib/64 -o libcairo.la -rpath > /opt/csw/lib/64 > cairo-analysis-surface.lo cairo-arc.lo cairo-array.lo cairo-atomic.lo > cairo-base85-stream.lo cairo-bentley-ottmann.lo cairo.lo cairo-cache.lo > cairo-clip.lo cairo-color.lo cairo-debug.lo cairo-fixed.lo > cairo-font-face.lo cairo-font-face-twin.lo cairo-font-face-twin-data.lo > cairo-font-options.lo cairo-freelist.lo cairo-gstate.lo cairo-hash.lo > cairo-hull.lo cairo-image-surface.lo cairo-lzw.lo cairo-matrix.lo > cairo-meta-surface.lo cairo-misc.lo cairo-mutex.lo > cairo-output-stream.lo cairo-paginated-surface.lo cairo-path-bounds.lo > cairo-path.lo cairo-path-fill.lo cairo-path-fixed.lo > cairo-path-stroke.lo cairo-pattern.lo cairo-pen.lo cairo-polygon.lo > cairo-rectangle.lo cairo-region.lo cairo-scaled-font.lo > cairo-skiplist.lo cairo-slope.lo cairo-spline.lo cairo-stroke-style.lo > cairo-surface.lo cairo-surface-fallback.lo cairo-system.lo > cairo-traps.lo cairo-unicode.lo cairo-user-font.lo cairo-version.lo > cairo-wideint.lo cairo-cff-subset.lo cairo-scaled-font-subsets.lo > cairo-truetype-subset.lo cairo-type1-fallback.lo cairo-type1-subset.lo > cairo-type3-glyph-surface.lo cairo-pdf-operators.lo > cairo-xlib-display.lo cairo-xlib-screen.lo cairo-xlib-surface.lo > cairo-xlib-visual.lo cairo-png.lo cairo-ft-font.lo > cairo-ps-surface.lo cairo-pdf-surface.lo cairo-deflate-stream.lo > cairo-svg-surface.lo test-fallback-surface.lo test-meta-surface.lo > test-paginated-surface.lo -L/opt/csw/lib -lpixman-1 -L/opt/csw/lib > -lfreetype -L/opt/csw/lib -lfontconfig -L/opt/csw/lib -lpng12 > -lm -lz -lz -lXrender -lSM -lICE -L/usr/openwin/lib -lX11 > -lsocket -lnsl -lm -lm > libtool: link: /opt/studio/SOS11/SUNWspro/bin/cc -G -z defs -h > libcairo.so.2 -o .libs/libcairo.so.2.10800.6 > .libs/cairo-analysis-surface.o .libs/cairo-arc.o .libs/cairo-array.o > .libs/cairo-atomic.o .libs/cairo-base85-stream.o > .libs/cairo-bentley-ottmann.o .libs/cairo.o .libs/cairo-cache.o > .libs/cairo-clip.o .libs/cairo-color.o .libs/cairo-debug.o > .libs/cairo-fixed.o .libs/cairo-font-face.o .libs/cairo-font-face-twin.o > .libs/cairo-font-face-twin-data.o .libs/cairo-font-options.o > .libs/cairo-freelist.o .libs/cairo-gstate.o .libs/cairo-hash.o > .libs/cairo-hull.o .libs/cairo-image-surface.o .libs/cairo-lzw.o > .libs/cairo-matrix.o .libs/cairo-meta-surface.o .libs/cairo-misc.o > .libs/cairo-mutex.o .libs/cairo-output-stream.o > .libs/cairo-paginated-surface.o .libs/cairo-path-bounds.o > .libs/cairo-path.o .libs/cairo-path-fill.o .libs/cairo-path-fixed.o > .libs/cairo-path-stroke.o .libs/cairo-pattern.o .libs/cairo-pen.o > .libs/cairo-polygon.o .libs/cairo-rectangle.o .libs/cairo-region.o > .libs/cairo-scaled-font.o .libs/cairo-skiplist.o .libs/cairo-slope.o > .libs/cairo-spline.o .libs/cairo-stroke-style.o .libs/cairo-surface.o > .libs/cairo-surface-fallback.o .libs/cairo-system.o .libs/cairo-traps.o > .libs/cairo-unicode.o .libs/cairo-user-font.o .libs/cairo-version.o > .libs/cairo-wideint.o .libs/cairo-cff-subset.o > .libs/cairo-scaled-font-subsets.o .libs/cairo-truetype-subset.o > .libs/cairo-type1-fallback.o .libs/cairo-type1-subset.o > .libs/cairo-type3-glyph-surface.o .libs/cairo-pdf-operators.o > .libs/cairo-xlib-display.o .libs/cairo-xlib-screen.o > .libs/cairo-xlib-surface.o .libs/cairo-xlib-visual.o .libs/cairo-png.o > .libs/cairo-ft-font.o .libs/cairo-ps-surface.o .libs/cairo-pdf-surface.o > .libs/cairo-deflate-stream.o .libs/cairo-svg-surface.o > .libs/test-fallback-surface.o .libs/test-meta-surface.o > .libs/test-paginated-surface.o -L/opt/csw/lib/64 -L/opt/csw/X11/lib/64 > -L/opt/csw/lib -lpixman-1 -lfreetype -lfontconfig -lpng12 -lz -lXrender > -lSM -lICE -L/usr/openwin/lib -lX11 -lsocket -lnsl -lm -lc -xarch=v9 > -xarch=v9 > ld: fatal: file /opt/csw/lib/64/libpixman-1.so: wrong ELF class: ELFCLASS32 > ld: fatal: File processing errors. No output written to > .libs/libcairo.so.2.10800.6 > gmake[5]: *** [libcairo.la] Error 1 > gmake[5]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6/src' > gmake[4]: *** [all] Error 2 > gmake[4]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6/src' > gmake[3]: *** [all-recursive] Error 1 > gmake[3]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6' > gmake[2]: *** [all] Error 2 > gmake[2]: Leaving directory > `/home/dam/mgar/pkg/libcairo/trunk/work/build-isa-sparcv9/cairo-1.8.6' > gmake[1]: *** [build-work/build-isa-sparcv9/cairo-1.8.6/Makefile] Error 2 > gmake[1]: Leaving directory `/home/dam/mgar/pkg/libcairo/trunk' > gmake: *** [merge-isa-sparcv9] Error 2 > > > > > checking dynamic linker characteristics... solaris2.8 ld.so > (cached) (cached) checking how to hardcode library paths into > programs... immediate > checking for getisax... no > checking whether byte ordering is bigendian... yes > checking for inline... inline > checking whether __SUNPRO_C is declared... yes > checking whether __amd64 is declared... no > checking for perl... /opt/csw/bin/perl > checking for -fvisibility... no > checking for -xldscope (Sun compilers)... yes > checking whether to use MMX intrinsics... no > checking whether to use SSE2 intrinsics... no > checking whether to use VMX/Altivec intrinsics... no > checking whether to use ARM SIMD assembler... no > checking for pkg-config... /opt/csw/bin/pkg-config > checking pkg-config is at least version 0.9.0... yes > checking for GTK... configure: error: Package requirements (gtk+-2.0) > were not met: > > sh: gnome-config: not found > Package pangocairo was not found in the pkg-config search path. > Perhaps you should add the directory containing `pangocairo.pc' > to the PKG_CONFIG_PATH environment variable > Package 'pangocairo', required by 'GDK', not found > > Consider adjusting the PKG_CONFIG_PATH environment variable if you > installed software in a non-standard prefix. > > Alternatively, you may set the environment variables GTK_CFLAGS > and GTK_LIBS to avoid the need to call pkg-config. > See the pkg-config man page for more details. > > gmake[1]: *** [configure-work/build-isa-sparcv8/pixman-0.15.2/configure] > Error 1 > gmake[1]: Leaving directory `/home/dam/mgar/pkg/pixman/trunk' > gmake: *** [merge-isa-sparcv8] Error 2 > > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers From ihsan at opencsw.org Tue May 5 23:07:27 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 05 May 2009 23:07:27 +0200 Subject: [csw-maintainers] [csw-announce] changes to php5 packages In-Reply-To: <4A009BFD.4070507@opencsw.org> References: <4A009BFD.4070507@opencsw.org> Message-ID: <4A00AA8F.5090905@opencsw.org> Thanks Mike. I think it makes sense that we announce major changes on the maintainers list (just for the future). Ihsan Am 5.5.2009 22:05 Uhr, Mike Watters schrieb: > Hello, > There are a few changes to the php5 packages you need to be aware of. > > Going forward, starting with the current release of php5-5.2.9,REV=2009.04.29 > the php modules that are bundled with the php5 source code have been packaged > separately from php5. This will allow better granularity on installs. > > When you run pkg-get -i php5 you will get a "bare-bone" install containing > docs, php, pear and pecl. > > all the modules will be listed and commented out in > /opt/csw/php5/lib/php.ini.CSW on a system that does not contain an existing > php.ini file the php.ini.CSW will be copied to php.ini > > all module packages now contain admin scripts to enable and disable themselves > in the php.ini file. _______________________________________________ announce mailing list announce at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/announce -- ihsan at dogan.ch http://blog.dogan.ch/ From mwatters at opencsw.org Wed May 6 02:44:39 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 05 May 2009 19:44:39 -0500 Subject: [csw-maintainers] fixme script renamed to fixlibtool Message-ID: <4A00DD77.6060501@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I looked at the existing fixlibtool script in gar/bin. it looks like this script has been obsoleted. the functionality in fixlibtool exists in replacer ( which is called in gar.mk ) I can not see anywhere fixlibtool is used any longer. I renamed the fixme.sh to fixlibtool, but at the moment, it still lives in mgar/pkgs. if no one has any objection, I would like to: 1) svn mv the existing fixlibtool to orig_fixlibtool to keep it around. 2) move my fixlibtool into gar/bin. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoA3XcACgkQLrhmsXMSLxcCsgCdEVBuhV4/WMcCPlHID0su/cRl BiwAnRR/GQoJRDg8o/JHIuGAgaFlkciO =buCj -----END PGP SIGNATURE----- From dam at opencsw.org Wed May 6 10:53:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 6 May 2009 10:53:01 +0200 Subject: [csw-maintainers] I am running in circles around GNOME In-Reply-To: <4A00A395.1030500@opencsw.org> References: <409D629D-999F-43F8-ACE6-9EA6426D4B1E@opencsw.org> <4A00A395.1030500@opencsw.org> Message-ID: <07CBEACE-7FCB-42C9-A0A9-192612234EC3@opencsw.org> Hi Chad, Am 05.05.2009 um 22:37 schrieb Chad Harp: > I believe from the searching I did (few months ago) that the pango > dependency (via gtk) in pixman is just for some of the make check > tests. > > Try building pixman with --disable-gtk Good advice, that worked! Thank you :-) Best regards -- Dago From dam at opencsw.org Wed May 6 10:55:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 6 May 2009 10:55:19 +0200 Subject: [csw-maintainers] fixme script renamed to fixlibtool In-Reply-To: <4A00DD77.6060501@opencsw.org> References: <4A00DD77.6060501@opencsw.org> Message-ID: Hi Mike, Am 06.05.2009 um 02:44 schrieb Mike Watters: > I looked at the existing fixlibtool script in gar/bin. > it looks like this script has been obsoleted. the functionality in > fixlibtool > exists in replacer ( which is called in gar.mk ) I can not see > anywhere > fixlibtool is used any longer. > > I renamed the fixme.sh to fixlibtool, but at the moment, it still > lives in > mgar/pkgs. if no one has any objection, I would like to: > > 1) svn mv the existing fixlibtool to orig_fixlibtool to keep it > around. > 2) move my fixlibtool into gar/bin. Good thing. That would also avoid copying it in to every package. Please make sure to also adjust the wiki. Best regards -- Dago From dam at opencsw.org Wed May 6 11:46:24 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 6 May 2009 11:46:24 +0200 Subject: [csw-maintainers] DONE: pixman, pango, cairo References: Message-ID: <1C70E2B5-A67A-4D20-8B58-395320C69C25@opencsw.org> Hi, I just finished pixman, pango and cairo. The packages for Sparc are available in testing: build-01.May.2009/xproto-7.0.15,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/xextproto-7.0.5,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/xtrans-1.2.3,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/kbproto-1.0.3,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/inputproto-1.5.0,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/libxau-1.0.4,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-01.May.2009/libxau_devel-1.0.4,REV=2009.05.01-SunOS5.8-sparc- CSW.pkg build-01.May.2009/libpthreadstubs-0.1,REV=2009.05.01-SunOS5.8-all- CSW.pkg build-01.May.2009/xcbproto-1.4,REV=2009.05.01-SunOS5.8-all-CSW.pkg build-01.May.2009/libxcb-1.2,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-01.May.2009/libxcb_dev-1.2,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-01.May.2009/libxcb_doc-1.2,REV=2009.05.01-SunOS5.8-sparc-CSW.pkg build-02.May.2009/renderproto-0.9.3,REV=2009.05.02-SunOS5.8-sparc- CSW.pkg build-02.May.2009/libxrender-0.9.4,REV=2009.05.02-SunOS5.8-sparc-CSW.pkg build-02.May.2009/renderdev-0.9.4,REV=2009.05.02-SunOS5.8-sparc-CSW.pkg build-05.May.2009/libxft2-2.1.13,REV=2009.05.05-SunOS5.8-sparc-CSW.pkg build-05.May.2009/libx11-1.2.1,REV=2009.05.05-SunOS5.8-sparc-CSW.pkg build-05.May.2009/libx11_devel-1.2.1,REV=2009.05.05-SunOS5.8-sparc- CSW.pkg build-06.May.2009/pixman-0.15.2,REV=2009.05.06-SunOS5.8-sparc-CSW.pkg build-06.May.2009/libpango-1.24.2,REV=2009.05.06-SunOS5.8-sparc-CSW.pkg build-06.May.2009/libpango_devel-1.24.2,REV=2009.05.06-SunOS5.8-sparc- CSW.pkg build-06.May.2009/libpango_doc-1.24.2,REV=2009.05.06-SunOS5.8-sparc- CSW.pkg build-06.May.2009/libcairo-1.8.6,REV=2009.05.06-SunOS5.8-sparc-CSW.pkg build-06.May.2009/libcairo_devel-1.8.6,REV=2009.05.06-SunOS5.8-sparc- CSW.pkg build-06.May.2009/libcairo_doc-1.8.6,REV=2009.05.06-SunOS5.8-all-CSW.pkg Also, all packages are installed on build8st, just in case you want to compile something against the new libs. If there are no issues I will repackage/release all packages in order on the official build machines for both sparc/i386. Best regards -- Dago From Darin.Perusich at cognigencorp.com Wed May 6 15:27:42 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Wed, 06 May 2009 09:27:42 -0400 Subject: [csw-maintainers] copy package to www:/home/newpkgs Message-ID: <4A01904E.9090908@cognigencorp.com> I'm being prompted for a password when I try and login to www to copy a package to /home/newpkgs. Can someone fix this please? -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From phil at bolthole.com Wed May 6 17:57:33 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 08:57:33 -0700 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <4A01904E.9090908@cognigencorp.com> References: <4A01904E.9090908@cognigencorp.com> Message-ID: <20090506155733.GB4461@bolthole.com> On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: > I'm being prompted for a password when I try and login to www to copy a > package to /home/newpkgs. Can someone fix this please? it's just ssh. since you have access to both boxes, you can fix it yourself by appending the .pub key from login, to your authorized_keys on www. From mwatters at opencsw.org Wed May 6 18:17:18 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 11:17:18 -0500 Subject: [csw-maintainers] fixme script renamed to fixlibtool In-Reply-To: References: <4A00DD77.6060501@opencsw.org> Message-ID: <4A01B80E.5000708@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi Mike, > > Am 06.05.2009 um 02:44 schrieb Mike Watters: > > Good thing. That would also avoid copying it in to every package. > Please make sure to also adjust the wiki. > > > Best regards > > -- Dago > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I have moved it in place and committed. In Addition to copying into place, I modified gar to create a new variable: STRIP_LIBTOOL when set to 1 in your Makefile it will call fixlibtool as the final part of the configure directive. revision: 4720 I will adjust the wiki accordingly - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoBuA4ACgkQLrhmsXMSLxf9JQCdHWfEgdAlrgx845AgMzgF2JUA YdQAnjyJbvoziU0La1stqlHtRDUlWSH6 =4yAT -----END PGP SIGNATURE----- From Darin.Perusich at cognigencorp.com Wed May 6 19:37:25 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Wed, 06 May 2009 13:37:25 -0400 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <20090506155733.GB4461@bolthole.com> References: <4A01904E.9090908@cognigencorp.com> <20090506155733.GB4461@bolthole.com> Message-ID: <4A01CAD5.5080101@cognigencorp.com> Philip Brown wrote: > On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: >> I'm being prompted for a password when I try and login to www to copy a >> package to /home/newpkgs. Can someone fix this please? > > it's just ssh. since you have access to both boxes, you can fix it yourself > by appending the .pub key from login, to your authorized_keys on www. This doesn't make any sense. If I'm unable to get logged into www with ssh how am I suppose to fix this problem? My id_rsa.pub from login is in my authorized_keys file so unless www is mounting my home directory it's not going to be there and I won't be able to login. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From phil at bolthole.com Wed May 6 19:42:49 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 10:42:49 -0700 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <4A01CAD5.5080101@cognigencorp.com> References: <4A01904E.9090908@cognigencorp.com> <20090506155733.GB4461@bolthole.com> <4A01CAD5.5080101@cognigencorp.com> Message-ID: <20090506174249.GH4461@bolthole.com> On Wed, May 06, 2009 at 01:37:25PM -0400, Darin Perusich wrote: > > Philip Brown wrote: > > On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: > >> I'm being prompted for a password when I try and login to www to copy a > >> package to /home/newpkgs. Can someone fix this please? > > > > it's just ssh. since you have access to both boxes, you can fix it yourself > > by appending the .pub key from login, to your authorized_keys on www. > > This doesn't make any sense. If I'm unable to get logged into www with > ssh how am I suppose to fix this problem? you cannot log in *from login.bo.opencsw.org* to www. but you can log into www from your own box, cant you? If so, thencopy it from login, to your home box, and then to www. From Darin.Perusich at cognigencorp.com Wed May 6 19:56:00 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Wed, 06 May 2009 13:56:00 -0400 Subject: [csw-maintainers] copy package to www:/home/newpkgs In-Reply-To: <20090506174249.GH4461@bolthole.com> References: <4A01904E.9090908@cognigencorp.com> <20090506155733.GB4461@bolthole.com> <4A01CAD5.5080101@cognigencorp.com> <20090506174249.GH4461@bolthole.com> Message-ID: <4A01CF30.5020403@cognigencorp.com> Philip Brown wrote: > On Wed, May 06, 2009 at 01:37:25PM -0400, Darin Perusich wrote: >> Philip Brown wrote: >>> On Wed, May 06, 2009 at 09:27:42AM -0400, Darin Perusich wrote: >>>> I'm being prompted for a password when I try and login to www to copy a >>>> package to /home/newpkgs. Can someone fix this please? >>> it's just ssh. since you have access to both boxes, you can fix it yourself >>> by appending the .pub key from login, to your authorized_keys on www. >> This doesn't make any sense. If I'm unable to get logged into www with >> ssh how am I suppose to fix this problem? > > you cannot log in *from login.bo.opencsw.org* to www. > but you can log into www from your own box, cant you? > > If so, thencopy it from login, to your home box, and then to www. > What an amazing thing clarity is. Thanks -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From pfelecan at opencsw.org Wed May 6 20:08:31 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 06 May 2009 20:08:31 +0200 Subject: [csw-maintainers] "garbage" when installing with pkg-get Message-ID: Before filling a bug report let me know if you have also this kind of behaviour: - SunOS host 5.10 Generic_137138-09 i86pc i386 i86pc - update the catalog pkg-get updatecatalog - upgrade everything pkg-get upgrade - install additional software pkg-get install nrpe I get, systematically the following: INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for o Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for l Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for r Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for t Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for | Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for t Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for p Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for w Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for r Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for a Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for p Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for | Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for w Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for l Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for a Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for u Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for t Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for i Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for l Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for s Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for | Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for C Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for S Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for W Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for c Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for o Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for m Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for m Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for o Perhaps your catalog is out of date INTERNAL ERROR: cannot get remote version for n Perhaps your catalog is out of date No existing install of CSWnrpe found. Installing... Trying file:///htdocs/opencsw/unstable/i386/5.10/nrpe-2.12,REV=2009.04.26-SunOS5.8-i386-CSW.pkg.gz - PATH /opt/csw/flex-2.5.33/bin:/opt/csw/bin:/opt/csw/gcc3/bin:/usr/ccs/bin:/opt/csw/gnu:/usr/sbin:/usr/bin:/usr/dt/bin:/usr/openwin/bin:/bin -- Peter From phil at bolthole.com Wed May 6 20:16:42 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 11:16:42 -0700 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: References: Message-ID: <20090506181642.GB17668@bolthole.com> On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: > Before filling a bug report let me know if you have also this kind of > behaviour: > before filing a bug report, you should also either use it with a "normal" mirror, or specify where you are syncing your local copy from, and/or update your local copy again? > No existing install of CSWnrpe found. Installing... > Trying file:///htdocs/opencsw/unstable/i386/5.10/nrpe-2.12,REV=2009.04.26-SunOS5.8-i386-CSW.pkg.gz From pfelecan at opencsw.org Wed May 6 20:27:15 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 06 May 2009 20:27:15 +0200 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: <20090506181642.GB17668@bolthole.com> (Philip Brown's message of "Wed\, 6 May 2009 11\:16\:42 -0700") References: <20090506181642.GB17668@bolthole.com> Message-ID: Philip Brown writes: > On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: >> Before filling a bug report let me know if you have also this kind of >> behaviour: >> > > before filing a bug report, you should also either use it with a "normal" > mirror, or specify where you are syncing your local copy from, > and/or update your local copy again? You're right: rsync://www.ibiblio.org/sun-packages/opencsw -- Peter From phil at bolthole.com Wed May 6 20:39:39 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 6 May 2009 11:39:39 -0700 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: References: <20090506181642.GB17668@bolthole.com> Message-ID: <20090506183939.GD17668@bolthole.com> On Wed, May 06, 2009 at 08:27:15PM +0200, Peter FELECAN wrote: > Philip Brown writes: > > > On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: > >> Before filling a bug report let me know if you have also this kind of > >> behaviour: > >> > > > > before filing a bug report, you should also either use it with a "normal" > > mirror, or specify where you are syncing your local copy from, > > and/or update your local copy again? > > You're right: rsync://www.ibiblio.org/sun-packages/opencsw Weeeelll... i just used the following command: pkg-get -s http://www.ibiblio.org/pub/solaris/opencsw/current -U -u (Internal SCCS code revision @(#) pkg-get 4.7@(#)) (on a solaris 10 sparc machine) and it worked ok. so... ? From pfelecan at opencsw.org Wed May 6 20:58:43 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 06 May 2009 20:58:43 +0200 Subject: [csw-maintainers] "garbage" when installing with pkg-get In-Reply-To: <20090506183939.GD17668@bolthole.com> (Philip Brown's message of "Wed\, 6 May 2009 11\:39\:39 -0700") References: <20090506181642.GB17668@bolthole.com> <20090506183939.GD17668@bolthole.com> Message-ID: Philip Brown writes: > On Wed, May 06, 2009 at 08:27:15PM +0200, Peter FELECAN wrote: >> Philip Brown writes: >> >> > On Wed, May 06, 2009 at 08:08:31PM +0200, Peter FELECAN wrote: >> >> Before filling a bug report let me know if you have also this kind of >> >> behaviour: >> >> >> > >> > before filing a bug report, you should also either use it with a "normal" >> > mirror, or specify where you are syncing your local copy from, >> > and/or update your local copy again? >> >> You're right: rsync://www.ibiblio.org/sun-packages/opencsw > > Weeeelll... > i just used the following command: > > pkg-get -s http://www.ibiblio.org/pub/solaris/opencsw/current -U -u > > (Internal SCCS code revision @(#) pkg-get 4.7@(#)) > (on a solaris 10 sparc machine) > > and it worked ok. > so... Add to your path /opt/csw/gnu, e.g. /opt/csw/flex-2.5.33/bin:/opt/csw/bin:/opt/csw/gcc3/bin:/usr/ccs/bin:/opt/csw/gnu:/usr/sbin:/usr/bin:/usr/dt/bin:/usr/openwin/bin:/bin -- Peter From mwatters at opencsw.org Thu May 7 03:01:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 20:01:22 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241468034-sup-3126@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> Message-ID: <4A0232E2.10707@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ben, Ben Walton wrote: > > I'm updating the gcc and g++ packages on a sol10/x86 box right now. > I'll build the same packages you did and let you know. > > Thanks > -Ben Any Word on gcc4.3.3 testing amd64? The i386, sparcv8 and sparcv9 have all been successful in testing. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoCMuEACgkQLrhmsXMSLxeh3wCeIE1Shboxi8B8uZpe/qIu/Mzw guEAn3/p59jxyFP6Wm9TG6tTxCh/9Ph+ =msMf -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 05:15:14 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 06 May 2009 23:15:14 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A0232E2.10707@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> Message-ID: <1241665912-sup-3788@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Wed May 06 21:01:22 -0400 2009: Hi Mike, > Any Word on gcc4.3.3 testing amd64? > The i386, sparcv8 and sparcv9 have all been successful in testing. I'm sorry, I forgot about this...the vim build with BUILD_64=1 was successful, but I haven't done the other package yet. I'll try to get to that tomorrow. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu May 7 06:30:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 23:30:08 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241665912-sup-3788@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> Message-ID: <4A0263D0.6010001@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > I'm sorry, I forgot about this...the vim build with BUILD_64=1 was > successful, but I haven't done the other package yet. I'll try to get > to that tomorrow. no worries, I know you are busy. you get to it when you get to it. Thank You for testing it for me. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoCY9AACgkQLrhmsXMSLxcQmwCfZ0GplDpxCRifuizimcMWup3G qdkAoK+PaRINeYywbfp+SrGdtccb2WbA =T65I -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu May 7 06:33:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 06 May 2009 23:33:13 -0500 Subject: [csw-maintainers] php5 (NO Extensions) in testing Message-ID: <4A026489.1040605@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Changes: Added the elusive php-cgi binary - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoCZIgACgkQLrhmsXMSLxfLFwCgqraWlRR4qm3WcUdHtUAGq/Z6 dOQAoIlqGlbYNK327S2/zLQs0LhlxQWK =q92g -----END PGP SIGNATURE----- From dam at opencsw.org Thu May 7 11:36:03 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 11:36:03 +0200 Subject: [csw-maintainers] python and 64 bit Message-ID: <8F2A3D07-27A9-486C-86FD-87794FB9EB72@opencsw.org> Hi, I am still tweaking X11 stuff and noticed that xcb-proto generates some python files. > root/opt/csw/X11/lib/amd64 > root/opt/csw/X11/lib/amd64/pkgconfig > root/opt/csw/X11/lib/amd64/pkgconfig/xcb-proto.pc > root/opt/csw/X11/lib/pkgconfig > root/opt/csw/X11/lib/pkgconfig/xcb-proto.pc > root/opt/csw/X11/lib/python2.6 > root/opt/csw/X11/lib/python2.6/site-packages > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.py > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.pyc > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.pyo > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.py > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.pyc > root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.pyo > ... Usually I try to always provide libs in 64 bit too. Last time we talked about this was in November and Joshua said he had build this some time ago. IMHO it would be great if we would have a 64 bit Python (CSWpython64? CSWpythonx in Sun-terminology?). The new mGAR should make it fairly easy to add 64 bit. Best regards -- Dago From ihsan at opencsw.org Thu May 7 15:43:33 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 15:43:33 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage Message-ID: <4A02E585.5080901@opencsw.org> Hello, For the last 4 years Courier-IMAP served us very well. But with the increased amount of users using IMAP, Courier-IMAP reached it limits. Especially users with huge mailboxes are affected by this issue. Therefore I will migrate this evening at 18:00 UTC to Dovecot IMAP. Dovecot IMAP is a new IMAP server, which is much faster than Courier and it provides backward compatibility to Courier. I'm expecting a downtime of 1-2 hours. During this time, there will be no mail services available (SMTP and IMAP). Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From mwatters at opencsw.org Thu May 7 16:06:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 09:06:17 -0500 Subject: [csw-maintainers] python and 64 bit In-Reply-To: <8F2A3D07-27A9-486C-86FD-87794FB9EB72@opencsw.org> References: <8F2A3D07-27A9-486C-86FD-87794FB9EB72@opencsw.org> Message-ID: <4A02EAD9.7000108@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago, Dagobert Michelsen wrote: > > Usually I try to always provide libs in 64 bit too. Last time we > talked about this was in November and Joshua said he had build > this some time ago. IMHO it would be great if we would have a > 64 bit Python (CSWpython64? CSWpythonx in Sun-terminology?). > The new mGAR should make it fairly easy to add 64 bit. > I have a bug report for this already ;) I have been holding for the GCC4.3.3 release. Python compiles against some gcc dependent libraries that need 64bit support. All the testing thus far on gcc4.3.3 is good, Ben is finishing up testing the amd64 builds. I hope to get gcc released by the end of the week. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoC6sMACgkQLrhmsXMSLxd/MwCfbffh9mV8zBZOJYgjkljunqWK j7AAoLlRxDP/Urjj8KzZk78MNnZYXTXF =JGYM -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 16:31:30 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 10:31:30 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02E585.5080901@opencsw.org> References: <4A02E585.5080901@opencsw.org> Message-ID: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: > For the last 4 years Courier-IMAP served us very well. But with the > increased amount of users using IMAP, Courier-IMAP reached it limits. > Especially users with huge mailboxes are affected by this issue. I made the move from courier and uw-imap a few years back and have never regretted it. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darin.Perusich at cognigencorp.com Thu May 7 16:40:45 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 07 May 2009 10:40:45 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Message-ID: <4A02F2ED.1010306@cognigencorp.com> Ben Walton wrote: > Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >> For the last 4 years Courier-IMAP served us very well. But with the >> increased amount of users using IMAP, Courier-IMAP reached it limits. >> Especially users with huge mailboxes are affected by this issue. > > I made the move from courier and uw-imap a few years back and have > never regretted it. > I've always preferred cyrus personally ;-) -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From ihsan at opencsw.org Thu May 7 16:45:52 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 16:45:52 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Message-ID: <4A02F420.9090301@opencsw.org> Am 7.5.2009 16:31 Uhr, Ben Walton schrieb: >> For the last 4 years Courier-IMAP served us very well. But with the >> increased amount of users using IMAP, Courier-IMAP reached it limits. >> Especially users with huge mailboxes are affected by this issue. > I made the move from courier and uw-imap a few years back and have > never regretted it. I was really thinking for long time about that, but I hesitated because Courier worked very well. I've made a few tests and I was impressed how fast Dovecot is. Even the migration is very easy. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Thu May 7 16:47:25 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 16:47:25 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02F2ED.1010306@cognigencorp.com> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F2ED.1010306@cognigencorp.com> Message-ID: <4A02F47D.8090302@opencsw.org> Am 7.5.2009 16:40 Uhr, Darin Perusich schrieb: >> Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >>> For the last 4 years Courier-IMAP served us very well. But with the >>> increased amount of users using IMAP, Courier-IMAP reached it limits. >>> Especially users with huge mailboxes are affected by this issue. >> I made the move from courier and uw-imap a few years back and have >> never regretted it. > > I've always preferred cyrus personally ;-) I've looked at Courier as well, but the not existing documentation makes very hard to start with Cyrus. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Thu May 7 16:51:06 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 10:51:06 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02F420.9090301@opencsw.org> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F420.9090301@opencsw.org> Message-ID: <1241707760-sup-9083@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Thu May 07 10:45:52 -0400 2009: > I've made a few tests and I was impressed how fast Dovecot is. Even the > migration is very easy. When we switched a machine from uw to dovecot last year, the load drop was simply awesome. The indexing that dovecot does on mbox files really makes a huge difference. It's extended the useful life of this particular machine by quite a bit. ...why we still have mbox in use is another story! :( -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darin.Perusich at cognigencorp.com Thu May 7 16:53:58 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 07 May 2009 10:53:58 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02F47D.8090302@opencsw.org> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F2ED.1010306@cognigencorp.com> <4A02F47D.8090302@opencsw.org> Message-ID: <4A02F606.3000607@cognigencorp.com> Ihsan Dogan wrote: > Am 7.5.2009 16:40 Uhr, Darin Perusich schrieb: > >>> Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >>>> For the last 4 years Courier-IMAP served us very well. But with the >>>> increased amount of users using IMAP, Courier-IMAP reached it limits. >>>> Especially users with huge mailboxes are affected by this issue. >>> I made the move from courier and uw-imap a few years back and have >>> never regretted it. >> I've always preferred cyrus personally ;-) > > I've looked at Courier as well, but the not existing documentation makes > very hard to start with Cyrus. > Yeah, that's one area where they fall short, especially if you're trying to setup murder, but it's not as bad as it use to be. Once it's up and running you can basically forget about it and does it scale! -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From bwalton at opencsw.org Thu May 7 16:56:52 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 10:56:52 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A0263D0.6010001@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> Message-ID: <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: > no worries, I know you are busy. > you get to it when you get to it. Thank You for testing it for me. libffi builds fine, but there were linking issues (shown by ldd and dump), not during the actual build. The following small GAR modification resolved this: $ svn diff Index: gar.conf.mk =================================================================== --- gar.conf.mk (revision 4723) +++ gar.conf.mk (working copy) @@ -477,7 +477,7 @@ OPTFLAGS ?= $(strip $($(GARCOMPILER)_CC_FLAGS) $(_CATEGORY_OPTFLAGS) $(EXTRA_OPTFLAGS)) GCC3_LD_OPTIONS = -R$(GCC3_CC_HOME)/lib $(EXTRA_GCC3_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) -GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib $(EXTRA_GCC4_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) +GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib/\$$ISALIST $(EXTRA_GCC4_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) SOS11_LD_OPTIONS = $(strip $(EXTRA_SOS11_LD_OPTIONS) $(EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) SOS12_LD_OPTIONS = $(strip $(EXTRA_SOS12_LD_OPTIONS) $(EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) I don't know if that's the best way to make this change sticky, so Dago's input here would be helpful. I'm doing binutils now. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at opencsw.org Thu May 7 17:55:59 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 17:55:59 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241707760-sup-9083@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F420.9090301@opencsw.org> <1241707760-sup-9083@ntdws12.chass.utoronto.ca> Message-ID: <4A03048F.3070601@opencsw.org> Am 7.5.2009 16:51 Uhr, Ben Walton schrieb: >> I've made a few tests and I was impressed how fast Dovecot is. Even the >> migration is very easy. > When we switched a machine from uw to dovecot last year, the load drop > was simply awesome. The indexing that dovecot does on mbox files > really makes a huge difference. It's extended the useful life of this > particular machine by quite a bit. Are you using IMAP IDLE as well? I couldn't find any information, how much load IMAP IDLE will generate. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From phil at bolthole.com Thu May 7 18:45:39 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 7 May 2009 09:45:39 -0700 Subject: [csw-maintainers] php5 (NO Extensions) in testing In-Reply-To: <4A026489.1040605@opencsw.org> References: <4A026489.1040605@opencsw.org> Message-ID: <20090507164539.GC84608@bolthole.com> On Wed, May 06, 2009 at 11:33:13PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Changes: > Added the elusive php-cgi binary wierd. there's a command-line "php" standalone binary.. but there's also a SEPARATE php-cgi binary? i guess possibly to support "Fast-CGI"? From bwalton at opencsw.org Thu May 7 18:53:13 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 12:53:13 -0400 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A03048F.3070601@opencsw.org> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> <4A02F420.9090301@opencsw.org> <1241707760-sup-9083@ntdws12.chass.utoronto.ca> <4A03048F.3070601@opencsw.org> Message-ID: <1241714904-sup-8536@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Thu May 07 11:55:59 -0400 2009: > Are you using IMAP IDLE as well? I couldn't find any information, how > much load IMAP IDLE will generate. I've always left that in the default setting state (eg: the workaround for lookout/express is enabled). We have client connections active for (sometimes) days on end. Whether many of the clients in play use the IDLE command, I don't know...we've got a hodge-podge of software out there (Pegasus, Eudora, old Netscape, etc in addition to newer/decent stuff). I haven't tried disabling it to see if performance is better though... Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu May 7 18:56:10 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 18:56:10 +0200 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241708078-sup-5541@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Message-ID: Hi Ben, Am 07.05.2009 um 16:56 schrieb Ben Walton: > Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 > 2009: >> no worries, I know you are busy. >> you get to it when you get to it. Thank You for testing it for me. > > libffi builds fine, but there were linking issues (shown by ldd and > dump), not during the actual build. The following small GAR > modification resolved this: > > > $ svn diff > Index: gar.conf.mk > =================================================================== > --- gar.conf.mk (revision 4723) > +++ gar.conf.mk (working copy) > @@ -477,7 +477,7 @@ > OPTFLAGS ?= $(strip $($(GARCOMPILER)_CC_FLAGS) $(_CATEGORY_OPTFLAGS) > $(EXTRA_OPTFLAGS)) > > GCC3_LD_OPTIONS = -R$(GCC3_CC_HOME)/lib $(EXTRA_GCC3_LD_OPTIONS) $ > (EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) > -GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib $(EXTRA_GCC4_LD_OPTIONS) $ > (EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) > +GCC4_LD_OPTIONS = -R$(GCC4_CC_HOME)/lib/\$$ISALIST $ > (EXTRA_GCC4_LD_OPTIONS) $(EXTRA_GCC_LD_OPTIONS) $(EXTRA_LD_OPTIONS) If you want $ISALIST, yes. Otherwise you may also use $(abspath $ (GCC4_CC_HOME)/lib/$(MM_LIBDIR)) for just 64 bits if there are no optimized libs. > SOS11_LD_OPTIONS = $(strip $(EXTRA_SOS11_LD_OPTIONS) $ > (EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) > SOS12_LD_OPTIONS = $(strip $(EXTRA_SOS12_LD_OPTIONS) $ > (EXTRA_SOS_LD_OPTIONS) $(EXTRA_LD_OPTIONS)) > > I don't know if that's the best way to make this change sticky, so > Dago's input here would be helpful. I'm doing binutils now. Please go ahead and commit either the ISALIST or the static /64 version, it was indeed a bug. Best regards -- Dago From dam at opencsw.org Thu May 7 19:38:52 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 19:38:52 +0200 Subject: [csw-maintainers] Updating all build servers Message-ID: Hi, I am updating all build servers to current/ now. Please stand by. Best regards -- Dago From dam at opencsw.org Thu May 7 19:46:10 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 19:46:10 +0200 Subject: [csw-maintainers] Updating all build servers In-Reply-To: References: Message-ID: <7B37DE1C-7314-44D5-BD67-0DF274573165@opencsw.org> Hi, Am 07.05.2009 um 19:38 schrieb Dagobert Michelsen: > I am updating all build servers to current/ now. Please stand by. Please note: this includes Subversion. Your local repository will be updated when you next access it with the new svn 1.6 client. Best regards -- Dago From mwatters at opencsw.org Thu May 7 19:54:27 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 12:54:27 -0500 Subject: [csw-maintainers] php5 (NO Extensions) in testing In-Reply-To: <20090507164539.GC84608@bolthole.com> References: <4A026489.1040605@opencsw.org> <20090507164539.GC84608@bolthole.com> Message-ID: <4A032053.4070007@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Wed, May 06, 2009 at 11:33:13PM -0500, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Changes: >> Added the elusive php-cgi binary > > wierd. there's a command-line "php" standalone binary.. but there's also a > SEPARATE php-cgi binary? > > i guess possibly to support "Fast-CGI"? > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers yes and no the php-cgi is for specifically running php in cgi mode, not necessarily for fast-cgi. The best justification I have found to use php-cgi over php: "Using PHP as a CGI binary is an option for setups that for some reason do not wish to integrate PHP as a module into server software (like Apache), or will use PHP with different kinds of CGI wrappers to create safe chroot and setuid environments for scripts." http://us.php.net/manual/en/security.cgi-bin.attacks.php - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDIFIACgkQLrhmsXMSLxcTJgCdEVnW6W+GBo9HsBilSCgoDAuc eFAAn0CYL+nKNFMx/kHpQXdMPf8oV/xo =k5fj -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 20:57:52 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 14:57:52 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A0263D0.6010001@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> Message-ID: <1241722471-sup-1684@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: The binutils build is successful, but the merge fails with output that pasted below...The binaries that do get merged look ok. I don't know if this is a build issue or a gar description issue at this point. I won't have a chance to look in the next little while, either. /bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/nm ./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/strip pax: link: : No such file or directory /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/strip ./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/c++filt /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/c++filt ./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/as pax: link: : No such file or directory /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/as ./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/gprof /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/gprof ./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/ld pax: link: : No such file or directory /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/ld ./opt/csw/bin/amd64/stpIain1 >> ./opt/csw/bin/amd64/stpIain1 /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/stpIain1 gmake[1]: *** [merge-copy-relocated-only] Error 1 gmake[1]: Leaving directory `/homes/bwalton/packages/binutils/trunk' gmake: *** [merge-isa-amd64] Error 2 Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Thu May 7 21:06:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 15:06:08 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Message-ID: <1241723127-sup-5535@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 -0400 2009: Hi Dago, > If you want $ISALIST, yes. Otherwise you may also use $(abspath $ > (GCC4_CC_HOME)/lib/$(MM_LIBDIR)) > for just 64 bits if there are no optimized libs. I think ISALIST is the best approach here, so I'll commit that shortly. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ihsan at opencsw.org Thu May 7 21:17:57 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 21:17:57 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <4A02E585.5080901@opencsw.org> References: <4A02E585.5080901@opencsw.org> Message-ID: <4A0333E5.4060902@opencsw.org> Am 7.5.2009 15:43 Uhr, Ihsan Dogan schrieb: > For the last 4 years Courier-IMAP served us very well. But with the > increased amount of users using IMAP, Courier-IMAP reached it limits. > Especially users with huge mailboxes are affected by this issue. > > Therefore I will migrate this evening at 18:00 UTC to Dovecot IMAP. > Dovecot IMAP is a new IMAP server, which is much faster than Courier and > it provides backward compatibility to Courier. > > I'm expecting a downtime of 1-2 hours. During this time, there will be > no mail services available (SMTP and IMAP). The migration is over and everything is up and running again. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Thu May 7 21:20:11 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 15:20:11 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> Message-ID: <1241723902-sup-9306@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 -0400 2009: Before I committed the change, I took a peek in /opt/csw/gcc4/lib. It doesn't have the 64 -> amd64, 32 -> ., i386 -> . symlinks that exist in /opt/csw/lib. These should likely be added to the gcc4 core package before the $ISALIST is added, no? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu May 7 21:33:54 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 7 May 2009 21:33:54 +0200 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241723902-sup-9306@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> <1241723902-sup-9306@ntdws12.chass.utoronto.ca> Message-ID: <4D3E9D15-3A6C-45EF-BA62-11E20BC46F22@opencsw.org> Hi Ben, Am 07.05.2009 um 21:20 schrieb Ben Walton: > Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 > -0400 2009: > > Before I committed the change, I took a peek in /opt/csw/gcc4/lib. It > doesn't have the 64 -> amd64, 32 -> ., i386 -> . symlinks that exist > in /opt/csw/lib. These should likely be added to the gcc4 core > package before the $ISALIST is added, no? That would be nice, yes. Otherwise you must add -R$(GCC4_CC_HOME)/lib/\$$ISALIST -R$(abspath $(GCC4_CC_HOME)/lib/$ (MM_LIBDIR)) If you specify both you catch 64 bit with lib/(sparcv9|amd64) and 32 bit with lib/. Best regards -- Dago From mwatters at opencsw.org Thu May 7 21:44:36 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 14:44:36 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241723902-sup-9306@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241708078-sup-5541@ntdws12.chass.utoronto.ca> <1241723902-sup-9306@ntdws12.chass.utoronto.ca> Message-ID: <4A033A24.1090409@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Dagobert Michelsen's message of Thu May 07 12:56:10 -0400 2009: > > Before I committed the change, I took a peek in /opt/csw/gcc4/lib. It > doesn't have the 64 -> amd64, 32 -> ., i386 -> . symlinks that exist > in /opt/csw/lib. These should likely be added to the gcc4 core > package before the $ISALIST is added, no? > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Those links are missing on the sparc package as well. gcc4 does not use gar to merge :( I just committed a change to my recipe to fix the missing links, I will repackage with the links in place. The merge problems you had on binutils, I have the exact same issue on sparc. ( I never tested the merge, just up through install ) The binaries work OK for sparcv9. Looks like a merge problem in gar. just to be sure, I am testing the merge with Sun Studio compiled binaries and will let you know what happens. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDOiQACgkQLrhmsXMSLxcpqACgjUe0gl1GztEliOmR30wxB3TR IqsAoNv6sWYfes0JE5BLNoVfM4w45fET =Ecgx -----END PGP SIGNATURE----- From ihsan at opencsw.org Thu May 7 21:44:47 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 21:44:47 +0200 Subject: [csw-maintainers] IMPORTAN information for Procmail users on mail (IMAP) Message-ID: <4A033A2F.3080504@opencsw.org> Hello, The Courier IMAP on the mail server has been replaced by Dovecot. If you are using Procmail to sort mail, you might consider to deliver your mail with the Dovecot LDA in order to update the index files immediately. There is a document on the Dovecot Wiki: http://wiki.dovecot.org/procmail If you had for example this entry in your ~/.procmailrc to deliver the mails from the announce mailing list to Inbox/opencsw/announce :0H * ^List-Post: .opencsw.announce/ You would have to replace it with: DELIVER="/opt/csw/libexec/dovecot/deliver" :0H * ^List-Post: | $DELIVER -m INBOX.opencsw.announce Please note that we don't use the default namespace. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Thu May 7 22:13:02 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Thu, 07 May 2009 22:13:02 +0200 Subject: [csw-maintainers] IMPORTANT: Mail Server Outage In-Reply-To: <1241706655-sup-8801@ntdws12.chass.utoronto.ca> References: <4A02E585.5080901@opencsw.org> <1241706655-sup-8801@ntdws12.chass.utoronto.ca> Message-ID: <4A0340CE.60109@opencsw.org> Am 7.5.2009 16:31 Uhr, Ben Walton schrieb: > Excerpts from Ihsan Dogan's message of Thu May 07 09:43:33 -0400 2009: >> For the last 4 years Courier-IMAP served us very well. But with the >> increased amount of users using IMAP, Courier-IMAP reached it limits. >> Especially users with huge mailboxes are affected by this issue. > I made the move from courier and uw-imap a few years back and have > never regretted it. I regret now, that I haven changed earlier. I was testing the migration in a VM on my Mac and I've noticed that Dovecot was faster than Courier. After going live with Dovecot today, I'm really suprised that the difference is that big. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From mwatters at opencsw.org Thu May 7 22:31:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 15:31:08 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241722471-sup-1684@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> Message-ID: <4A03450C.2070801@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: > > The binutils build is successful, but the merge fails with output > that pasted below...The binaries that do get merged look ok. I don't > know if this is a build issue or a gar description issue at this > point. I won't have a chance to look in the next little while, > either. > > > /bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/nm > ./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/strip > pax: link: : No such file or directory > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/strip > ./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/c++filt > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/c++filt > ./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/as > pax: link: : No such file or directory > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/as > ./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/gprof > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/gprof > ./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/ld > pax: link: : No such file or directory > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/ld > ./opt/csw/bin/amd64/stpIain1 >> ./opt/csw/bin/amd64/stpIain1 > /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/stpIain1 > gmake[1]: *** [merge-copy-relocated-only] Error 1 > gmake[1]: Leaving directory `/homes/bwalton/packages/binutils/trunk' > gmake: *** [merge-isa-amd64] Error 2 > > Thanks > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I got the same merge errors compiling with Sun Studio I would give this a 99.9% probability of a bug in either gar or binutils. I will re-package gcc4.3.3 to fix the missing links in gar and push out for release. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDRQwACgkQLrhmsXMSLxcdpQCgvjaZjja84GBDl58WHmYko8vH 8WAAn1QG7ngL/YatuDpiZ27klKBVS+n6 =8nou -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu May 7 22:32:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 15:32:17 -0500 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A03450C.2070801@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> <4A03450C.2070801@opencsw.org> Message-ID: <4A034551.1060904@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > Ben Walton wrote: >> Excerpts from Mike Watters's message of Thu May 07 00:30:08 -0400 2009: > >> The binutils build is successful, but the merge fails with output >> that pasted below...The binaries that do get merged look ok. I don't >> know if this is a build issue or a gar description issue at this >> point. I won't have a chance to look in the next little while, >> either. > > >> /bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/nm >> ./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/strip >> pax: link: : No such file or directory >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/strip >> ./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/c++filt >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/c++filt >> ./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/as >> pax: link: : No such file or directory >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/as >> ./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/gprof >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/gprof >> ./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/ld >> pax: link: : No such file or directory >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/ld >> ./opt/csw/bin/amd64/stpIain1 >> ./opt/csw/bin/amd64/stpIain1 >> /homes/bwalton/packages/binutils/trunk/work/pkgroot/./opt/csw/bin/amd64/stpIain1 >> gmake[1]: *** [merge-copy-relocated-only] Error 1 >> gmake[1]: Leaving directory `/homes/bwalton/packages/binutils/trunk' >> gmake: *** [merge-isa-amd64] Error 2 > >> Thanks >> -Ben > > >> ------------------------------------------------------------------------ > >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > I got the same merge errors compiling with Sun Studio > > I would give this a 99.9% probability of a bug in either gar or binutils. > > I will re-package gcc4.3.3 to fix the missing links in gar and push out for > release. > > fix the missing links in "lib" not gar... sorry - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoDRVAACgkQLrhmsXMSLxfiCwCcDCaFBf3y74lCkKPnZfxYgkPp eLMAn1uJUhqyhc+x5gA8jX2+bmr8y8j6 =7fXo -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 7 23:54:38 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 07 May 2009 17:54:38 -0400 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <4A034551.1060904@opencsw.org> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> <4A03450C.2070801@opencsw.org> <4A034551.1060904@opencsw.org> Message-ID: <1241733234-sup-1049@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Thu May 07 16:32:17 -0400 2009: > > I would give this a 99.9% probability of a bug in either gar or > > binutils. I agree with this. I think that once you get those links added to lib that the package looks ready to roll. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Fri May 8 00:23:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 00:23:01 +0200 Subject: [csw-maintainers] gcc4.3.3 Final version in testing In-Reply-To: <1241733234-sup-1049@ntdws12.chass.utoronto.ca> References: <49FE0BDE.1010809@opencsw.org> <1241468034-sup-3126@ntdws12.chass.utoronto.ca> <4A0232E2.10707@opencsw.org> <1241665912-sup-3788@ntdws12.chass.utoronto.ca> <4A0263D0.6010001@opencsw.org> <1241722471-sup-1684@ntdws12.chass.utoronto.ca> <4A03450C.2070801@opencsw.org> <4A034551.1060904@opencsw.org> <1241733234-sup-1049@ntdws12.chass.utoronto.ca> Message-ID: <4EE27F31-0FEB-4D29-8483-8A90613FF97F@opencsw.org> Hi Ben, Am 07.05.2009 um 23:54 schrieb Ben Walton: > Excerpts from Mike Watters's message of Thu May 07 16:32:17 -0400 > 2009: > >>> I would give this a 99.9% probability of a bug in either gar or >>> binutils. > > I agree with this. I think that once you get those links added to lib > that the package looks ready to roll. It is a GAR issue. I'll take care of it. Best regards -- Dago From mwatters at opencsw.org Fri May 8 06:10:02 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 07 May 2009 23:10:02 -0500 Subject: [csw-maintainers] bug 3034 separate server and user command line binaries Message-ID: <4A03B09A.4060103@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I have to recompile subversion to fix an issue with ap2_subversion. I want to include this bug in the re-package. which binaries constitute server vs client? Can I get some confirmation on this before I split it up. I believe it should be as follows: Package Name:CSWsvn Catalog Name:svn ~~~~~~~~~~~~~~~~~~~~~~~~~~~ server - bin/svnserve server - bin/svnsync server - bin/svnadmin server - bin/svndumpfilter Package Name:CSWsvn-client Catalog Name:svn_client ~~~~~~~~~~~~~~~~~~~~~~~~~~~ client - bin/svnversion client - bin/svnlook client - bin/svn - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoDsJoACgkQLrhmsXMSLxf8fQCfXpRREcA0/YB7WWIGZmzjaOdY niMAmwaxV+oz32HJ1sr1m1bgzF/QnGMg =psHQ -----END PGP SIGNATURE----- From bonivart at opencsw.org Fri May 8 09:46:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 8 May 2009 09:46:14 +0200 Subject: [csw-maintainers] Fwd: [csw-users] Converting from Blastwave with pkgutil In-Reply-To: <71083D3F-F533-4834-B50E-78E3099671E8@gmail.com> References: <71083D3F-F533-4834-B50E-78E3099671E8@gmail.com> Message-ID: <625385e30905080046u1b9b8c25nf40cef5d517a42e0@mail.gmail.com> User request... -- /peter ---------- Forwarded message ---------- From: Cody Herriges Date: Fri, May 8, 2009 at 12:34 AM Subject: [csw-users] Converting from Blastwave with pkgutil To: users at lists.opencsw.org When pkgutil became the desired package management tool in Blastwave around the time of the split I switched my entire jumpstart configuration to be more tailored towards the use of pkgutil. ?Due to the lack of some key user apps that aren't being updated I have decided to switch over to opencsw. ?Any chance we could get a pkgutil package dropped in the root of the opencsw package repository so I can grab the latest pkgutil during my jumpstarts just as easily as I can grab the pkg-get package? Thanks, ---------------------------------------------------- Cody Herriges - Lead Unix System Administrator MCECS - ?Portland State University From pfelecan at opencsw.org Fri May 8 17:44:46 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 17:44:46 +0200 Subject: [csw-maintainers] OpenCSW default PHP version Message-ID: I'm packaging a product which has a PHP binding, amongst other bindings. I must make a choice to bind for PHP4 or PHP5 and I would like to choose the "default one"... Which is? -- Peter From bwalton at opencsw.org Fri May 8 17:56:03 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 11:56:03 -0400 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: References: Message-ID: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: > I'm packaging a product which has a PHP binding, amongst other > bindings. I must make a choice to bind for PHP4 or PHP5 and I would like > to choose the "default one"... Which is? PHP4 is deprecated and should be available as legacy only. It was eol'd in 2007 with critical security patches only until 2008. Nothing new should be built against it, imo. http://www.php.net/archive/2007.php (search down to 'end of life') -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Fri May 8 18:05:26 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:05:26 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding Message-ID: Again, I'm trying to provide a python binding and I obtain a strange situation, in my opinion incorrect: CFLAGS="-O3 -march=i486 -Wall -DSWF_LITTLE_ENDIAN -fno-strict-aliasing" \ /opt/csw/bin/python setup.py build running build running build_py creating build creating build/lib.solaris-2.10-i86pc-2.6 copying [...]/py_ext/ming.py -> build/lib.solaris-2.10-i86pc-2.6 copying [...]/py_ext/mingc.py -> build/lib.solaris-2.10-i86pc-2.6 running build_ext building '_mingc' extension creating build/temp.solaris-2.10-i86pc-2.6 [...] gcc -DNDEBUG -O -O3 -march=i486 -Wall -DSWF_LITTLE_ENDIAN -fno-strict-aliasing \ -I/usr/local/include [...]/py_ext/ming_wrap.o /opt/studio/SOS11/SUNWspro/bin/cc -G -O3 -march=i486 -Wall \ -DSWF_LITTLE_ENDIAN \ -fno-strict-aliasing [...]/py_ext/ming_wrap.o \ -L../src/.libs -L/usr/local/lib/ -lming -lz -lpng -lungif \ -lpython2.6 -o build/lib.solaris-2.10-i86pc-2.6/_mingc.so unable to execute /opt/studio/SOS11/SUNWspro/bin/cc: No such file or directory error: command '/opt/studio/SOS11/SUNWspro/bin/cc' failed with exit status 1 There are 2 issues: 1. I'm using gcc, so why cc is used? Because Python is generated with Sun Studio? 2. How comes that a specific build machine path is used, i.e. Sun Studio is installed by default in /opt/SUNWspro (I know that it can be installed at the administrator's convenience and on the build machines is installed in a /opt/studio directory but that is another story) The second issue is a show stopper for me... -- Peter From mwatters at opencsw.org Fri May 8 18:24:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:24:03 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: References: Message-ID: <4A045CA3.5070908@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Peter, Peter FELECAN wrote: > There are 2 issues: > > 1. I'm using gcc, so why cc is used? Because Python is generated with > Sun Studio? Yes that is exactly correct, when you build a python extension it will look at the way python was compiled and use those settings ( this is just like PERL ) > 2. How comes that a specific build machine path is used, i.e. Sun > Studio is installed by default in /opt/SUNWspro (I know that it can > be installed at the administrator's convenience and on the build > machines is installed in a /opt/studio directory but that is another > story) Unfortunately, this is just the way it is handled, it is true for PERL and most if not all the other packages that have a pkg-config like structure. you should be able to work around this by setting the environment variables CC={path to cc} CXX={path to c++ (CC in SunStudio)} > The second issue is a show stopper for me... - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEXKIACgkQLrhmsXMSLxfRWgCfWix2FTTnX34uuzsr5AWXDPao zeYAn2gbcXoeX73QwLFKdd73y1NiOBpf =+rV1 -----END PGP SIGNATURE----- From pfelecan at opencsw.org Fri May 8 18:27:34 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:27:34 +0200 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> (Ben Walton's message of "Fri\, 08 May 2009 11\:56\:03 -0400") References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> Message-ID: Ben Walton writes: > Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: >> I'm packaging a product which has a PHP binding, amongst other >> bindings. I must make a choice to bind for PHP4 or PHP5 and I would like >> to choose the "default one"... Which is? > > PHP4 is deprecated and should be available as legacy only. It was > eol'd in 2007 with critical security patches only until 2008. Nothing > new should be built against it, imo. > > http://www.php.net/archive/2007.php (search down to 'end of life') Thank you Ben. This led me to ask: why have we PHP4 in /opt/csw/bin and not PHP5? If I wish to use the current PHP I would expect to have it in /opt/csw/bin and not in some specific path, i.e. /opt/csw/php5/bin Can this be corrected? -- Peter From mwatters at opencsw.org Fri May 8 18:28:38 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:28:38 -0500 Subject: [csw-maintainers] Odd issues with the new IMAP Message-ID: <4A045DB6.90104@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I am using thunderbird/enigmail/imap for my opencsw account. Since the upgrade, I am getting Messages that are showing up at different times. for example: at 11:00 my time I received 4 messages from earlier in the day 08:32 08:48 09:23 and 10:47 "mytime" this is not really a problem, just Odd and only after the imap upgrade. anyone else experiencing issues similar? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEXbYACgkQLrhmsXMSLxdMvgCg5dkJ2BNWX1VPLmxNghcy4v08 X5IAn3W3n5brBZwVJQ/dlk3Nyn70Szwr =qPSt -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri May 8 18:30:24 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:30:24 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> Message-ID: <4A045E20.30306@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Ben Walton writes: > >> Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: >>> I'm packaging a product which has a PHP binding, amongst other >>> bindings. I must make a choice to bind for PHP4 or PHP5 and I would like >>> to choose the "default one"... Which is? >> PHP4 is deprecated and should be available as legacy only. It was >> eol'd in 2007 with critical security patches only until 2008. Nothing >> new should be built against it, imo. >> >> http://www.php.net/archive/2007.php (search down to 'end of life') > > Thank you Ben. This led me to ask: why have we PHP4 in /opt/csw/bin and > not PHP5? If I wish to use the current PHP I would expect to have it in > /opt/csw/bin and not in some specific path, i.e. /opt/csw/php5/bin > > Can this be corrected? I am just about to re-compile php5 to fix the latest bugs/feature requests. now would be the time to change it. ;) let me know, I will hold on the recompile until someone weighs in on this. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEXh8ACgkQLrhmsXMSLxdp2QCg4pc09Jm0JAPdWBmpJzaKgNE0 FKUAoJaIdMQWa2hVb7vb50uQTgj7K5xG =Fumm -----END PGP SIGNATURE----- From dam at opencsw.org Fri May 8 18:35:55 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 18:35:55 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A045CA3.5070908@opencsw.org> References: <4A045CA3.5070908@opencsw.org> Message-ID: Hi, Am 08.05.2009 um 18:24 schrieb Mike Watters: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi Peter, > Peter FELECAN wrote: > >> There are 2 issues: >> >> 1. I'm using gcc, so why cc is used? Because Python is generated with >> Sun Studio? > Yes that is exactly correct, when you build a python extension it will > look at the way python was compiled and use those settings ( this is > just like > PERL ) > >> 2. How comes that a specific build machine path is used, i.e. Sun >> Studio is installed by default in /opt/SUNWspro (I know that it can >> be installed at the administrator's convenience and on the build >> machines is installed in a /opt/studio directory but that is >> another >> story) > > Unfortunately, this is just the way it is handled, it is true for > PERL and > most if not all the other packages that have a pkg-config like > structure. > > you should be able to work around this by setting the environment > variables > CC={path to cc} CXX={path to c++ (CC in SunStudio)} Is it really this bad? Then we should make Sun Studio 11 the default at /opt/SUNWspro and let GAR switch flags on the detected compiler if there is e. g. SS12 at that location. Best regards -- Dago From pfelecan at opencsw.org Fri May 8 18:36:49 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:36:49 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A045CA3.5070908@opencsw.org> (Mike Watters's message of "Fri\, 08 May 2009 11\:24\:03 -0500") References: <4A045CA3.5070908@opencsw.org> Message-ID: Mike Watters writes: > Hi Peter, Thank you Mike for your explanation. > Peter FELECAN wrote: > >> There are 2 issues: >> >> 1. I'm using gcc, so why cc is used? Because Python is generated with >> Sun Studio? > Yes that is exactly correct, when you build a python extension it will > look at the way python was compiled and use those settings ( this is just like > PERL ) You're right about perl, however it uses the well behaved cc driver and not an absolute path, as for Python (see the second issue) because cc is /usr/bin/cc which is a link to /opt/SUNWspro/bin/cc. This works on my system but will not work without tweaking on the build system... >> 2. How comes that a specific build machine path is used, i.e. Sun >> Studio is installed by default in /opt/SUNWspro (I know that it can >> be installed at the administrator's convenience and on the build >> machines is installed in a /opt/studio directory but that is another >> story) > > Unfortunately, this is just the way it is handled, it is true for PERL and > most if not all the other packages that have a pkg-config like structure. > > you should be able to work around this by setting the environment variables > CC={path to cc} CXX={path to c++ (CC in SunStudio)} But this will work only if I use Sun Studio C and not gcc. I must confess that I don't like this situation of having more than a compiler and doing acrobatics to handle weird issues. Well, I'm opting to not provide some bindings. -- Peter From pfelecan at opencsw.org Fri May 8 18:39:31 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:39:31 +0200 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <4A045E20.30306@opencsw.org> (Mike Watters's message of "Fri\, 08 May 2009 11\:30\:24 -0500") References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> Message-ID: Mike Watters writes: > Peter FELECAN wrote: >> Ben Walton writes: >> >>> Excerpts from Peter FELECAN's message of Fri May 08 11:44:46 -0400 2009: >>>> I'm packaging a product which has a PHP binding, amongst other >>>> bindings. I must make a choice to bind for PHP4 or PHP5 and I would like >>>> to choose the "default one"... Which is? >>> PHP4 is deprecated and should be available as legacy only. It was >>> eol'd in 2007 with critical security patches only until 2008. Nothing >>> new should be built against it, imo. >>> >>> http://www.php.net/archive/2007.php (search down to 'end of life') >> >> Thank you Ben. This led me to ask: why have we PHP4 in /opt/csw/bin and >> not PHP5? If I wish to use the current PHP I would expect to have it in >> /opt/csw/bin and not in some specific path, i.e. /opt/csw/php5/bin >> >> Can this be corrected? > > I am just about to re-compile php5 to fix the latest bugs/feature requests. > now would be the time to change it. > > ;) let me know, I will hold on the recompile until someone weighs in on this. I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at least, put it in /opt/csw/php4. -- Peter From pfelecan at opencsw.org Fri May 8 18:41:06 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 08 May 2009 18:41:06 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: (Dagobert Michelsen's message of "Fri\, 8 May 2009 18\:35\:55 +0200") References: <4A045CA3.5070908@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi, > > Am 08.05.2009 um 18:24 schrieb Mike Watters: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi Peter, >> Peter FELECAN wrote: >> >>> There are 2 issues: >>> >>> 1. I'm using gcc, so why cc is used? Because Python is generated with >>> Sun Studio? >> Yes that is exactly correct, when you build a python extension it will >> look at the way python was compiled and use those settings ( this is >> just like >> PERL ) >> >>> 2. How comes that a specific build machine path is used, i.e. Sun >>> Studio is installed by default in /opt/SUNWspro (I know that it can >>> be installed at the administrator's convenience and on the build >>> machines is installed in a /opt/studio directory but that is >>> another >>> story) >> >> Unfortunately, this is just the way it is handled, it is true for >> PERL and >> most if not all the other packages that have a pkg-config like >> structure. >> >> you should be able to work around this by setting the environment >> variables >> CC={path to cc} CXX={path to c++ (CC in SunStudio)} > > Is it really this bad? Then we should make Sun Studio 11 the default at > /opt/SUNWspro and let GAR switch flags on the detected compiler if there > is e. g. SS12 at that location. Please. This will work for those not using gar, for the moment. __ Peter From mwatters at opencsw.org Fri May 8 18:49:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:49:51 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: References: <4A045CA3.5070908@opencsw.org> Message-ID: <4A0462AF.7070508@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi, > > Am 08.05.2009 um 18:24 schrieb Mike Watters: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi Peter, >> Peter FELECAN wrote: >> >>> There are 2 issues: >>> >>> 1. I'm using gcc, so why cc is used? Because Python is generated with >>> Sun Studio? >> Yes that is exactly correct, when you build a python extension it will >> look at the way python was compiled and use those settings ( this is >> just like >> PERL ) >> >>> 2. How comes that a specific build machine path is used, i.e. Sun >>> Studio is installed by default in /opt/SUNWspro (I know that it can >>> be installed at the administrator's convenience and on the build >>> machines is installed in a /opt/studio directory but that is another >>> story) >> >> Unfortunately, this is just the way it is handled, it is true for >> PERL and >> most if not all the other packages that have a pkg-config like structure. >> >> you should be able to work around this by setting the environment >> variables >> CC={path to cc} CXX={path to c++ (CC in SunStudio)} > > Is it really this bad? Then we should make Sun Studio 11 the default at > /opt/SUNWspro and let GAR switch flags on the detected compiler if there > is e. g. SS12 at that location. That would be OK, I guess, but I think SOS12 should be the default. Right now SOS12 is not installed on solaris 8. I have tested it on a solaris 8 zone (from a solaris 10 global) and it seems to work OK. has anyone else tried this out? I installed SOS12 on solaris10 global and "shared" it to the solaris 8 zone. zoneadm -z s8 halt zonecfg -z s8 add fs set dir=/opt/SUNWspro set special=/opt/SUNWspro set type=lofs add options rw add options nodevices end commit exit zoneadm -z s8 boot - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEYq8ACgkQLrhmsXMSLxfysgCgs4l8laApYzNG/9T9ffJKeTFD ygAAn2t/yXpxtf36LC1E1+QHTAm+6eUG =pH7D -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri May 8 18:57:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 12:57:59 -0400 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> Message-ID: <1241801752-sup-8737@ntdws12.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Fri May 08 12:39:31 -0400 2009: > I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at > least, put it in /opt/csw/php4. I definitely agree that php5 should become the default. I argued last time that php4 should be removed from the catalog too, but Phil thought it was best to keep it at that time. If I changed my argument to 'remove it from the catalog, but leave it on the mirrors' would change your opinion Phil? Users would be free to manually download and install it, but it wouldn't be supported directly any more? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Fri May 8 18:59:47 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 11:59:47 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241801752-sup-8737@ntdws12.chass.utoronto.ca> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> Message-ID: <4A046503.4010702@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Peter FELECAN's message of Fri May 08 12:39:31 -0400 2009: >> I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at >> least, put it in /opt/csw/php4. > > I definitely agree that php5 should become the default. I argued last > time that php4 should be removed from the catalog too, but Phil > thought it was best to keep it at that time. If I changed my argument > to 'remove it from the catalog, but leave it on the mirrors' would > change your opinion Phil? Users would be free to manually download > and install it, but it wouldn't be supported directly any more? > I no longer see php4 core on the package page? I see the php4_* modules but no php or php4. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEZQMACgkQLrhmsXMSLxe0GgCg7Cz0iA8Ho5P4l/Z9IXAaGEDc NGoAn3PTi9tK36PiBx2u1g8iv5zP+aK9 =TuMu -----END PGP SIGNATURE----- From dam at opencsw.org Fri May 8 19:01:47 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 19:01:47 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A0462AF.7070508@opencsw.org> References: <4A045CA3.5070908@opencsw.org> <4A0462AF.7070508@opencsw.org> Message-ID: <499E4F07-9C1C-4193-A4E4-49E924D5DEF9@opencsw.org> Hi Mike, Am 08.05.2009 um 18:49 schrieb Mike Watters: > That would be OK, I guess, but I think SOS12 should be the default. > Right now SOS12 is not installed on solaris 8. SOS12 is not supported on Solaris 8, hence I don't think it is a good idea to use it. Best regards -- Dago From mwatters at opencsw.org Fri May 8 19:05:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 12:05:08 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: References: <4A045CA3.5070908@opencsw.org> Message-ID: <4A046644.8090701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Peter, Peter FELECAN wrote: > But this will work only if I use Sun Studio C and not gcc. > > I must confess that I don't like this situation of having more than a > compiler and doing acrobatics to handle weird issues. Well, I'm opting > to not provide some bindings. I have to agree with you, I use gar to build and it is still a problem. the subversion package, is/was my PITA for this problem. Perl, Python are built using Studio, Ruby is built with GCC I had to Hack up a "fix" to change the makefiles to be GCC computible. this worked without a problem. if you would like to see what I did, here is the link in gar. http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/subversion/trunk/Makefile?revision=4609&view=markup Lines 133 - 150 and Lines 194 - 200 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEZkQACgkQLrhmsXMSLxcmZwCdGII29c6Y6b5XMVCcNi8t0njC uDMAoLfSJ5eRl7bWANJidb5a78imGrii =c9BG -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 8 20:44:33 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 8 May 2009 11:44:33 -0700 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A046644.8090701@opencsw.org> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> Message-ID: <20090508184433.GA96779@bolthole.com> On Fri, May 08, 2009 at 12:05:08PM -0500, Mike Watters wrote: > Perl, Python are built using Studio, Ruby is built with GCC this is a problem. if ruby Can be built with studio, it Should be built with it. btw: make sure that you are using -mt flag for base things like python and ruby when you compile them please From bwalton at opencsw.org Fri May 8 20:51:19 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 14:51:19 -0400 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <20090508184433.GA96779@bolthole.com> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> Message-ID: <1241808516-sup-4065@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri May 08 14:44:33 -0400 2009: > this is a problem. if ruby Can be built with studio, it Should be built > with it. The ruby build fails with studio but passes with with gcc. It's an issue of the (somewhat nasty) way that Makefiles are generated for some of the extensions (without auto*). These are done on the fly, so it wasn't a simple post-configure step. It ends up passing flags to ld that ld doesn't like. I wasn't able to nail it down, but didn't fight it too hard, either. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Fri May 8 21:16:01 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 8 May 2009 21:16:01 +0200 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <1241808516-sup-4065@ntdws12.chass.utoronto.ca> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> <1241808516-sup-4065@ntdws12.chass.utoronto.ca> Message-ID: <64E60A27-3644-4AF1-8A5F-AB13BFC4308A@opencsw.org> Hi Ben, Am 08.05.2009 um 20:51 schrieb Ben Walton: > The ruby build fails with studio but passes with with gcc. It's an > issue of the (somewhat nasty) way that Makefiles are generated for > some of the extensions (without auto*). These are done on the fly, so > it wasn't a simple post-configure step. It ends up passing flags to > ld that ld doesn't like. I wasn't able to nail it down, but didn't > fight it too hard, either. James gave me an incredibly useful advice: Interpose execution and rewrite arguments. For example to rewrite arguments to cc and CC you can write this (taken from http://gar.svn.sourceforge.net/viewvc/gar/csw/mgar/pkg/flac/trunk/Makefile > include gar/category.mk > > CC := $(abspath ./bin/cc) > CXX := $(abspath ./bin/CC) and then have heavy hacking (here, includes to relative pathes are reordered to be before include to absolute pathes): > :::::::::::::: > bin/CC > :::::::::::::: > #!/opt/csw/bin/perl -w > > use strict; > > my @oargs; > > my @includes = grep { /^-I/ } @ARGV; > @includes = ((grep { /^-I[^\/]/ } @includes),(grep { /^-I[\/]/ } > @includes)); > my @argswoincludes = grep { !/^-I/ } @ARGV; > > foreach (@ARGV) { > if( /^-I/ ) { > push @oargs, @includes, @argswoincludes; > last; > } > push @oargs, shift @argswoincludes; > } > > my $cc = "/opt/studio/SOS11/SUNWspro/bin/" . ($0 =~ /cc/ ? "cc" : > "CC"); > > print "$cc ", join( " ", @oargs ), "\n"; > exec( $cc, @oargs ); You could poke around stuff for "ld" also, of course... Best regards -- Dago From mwatters at opencsw.org Fri May 8 21:33:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 14:33:22 -0500 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <1241808516-sup-4065@ntdws12.chass.utoronto.ca> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> <1241808516-sup-4065@ntdws12.chass.utoronto.ca> Message-ID: <4A048902.30400@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ben, Ben Walton wrote: > Excerpts from Philip Brown's message of Fri May 08 14:44:33 -0400 2009: >> this is a problem. if ruby Can be built with studio, it Should be built >> with it. > > The ruby build fails with studio but passes with with gcc. It's an > issue of the (somewhat nasty) way that Makefiles are generated for > some of the extensions (without auto*). These are done on the fly, so > it wasn't a simple post-configure step. It ends up passing flags to > ld that ld doesn't like. I wasn't able to nail it down, but didn't > fight it too hard, either. > > -Ben Ruby compiles fine with Studio 12. we could just skip solaris8 ;) ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEiQIACgkQLrhmsXMSLxdvfACgp8tm++TOFNjGowebZyjVEn+J UAgAn0eGOVvo9imxQjogqAXG85JPsy9j =a2BN -----END PGP SIGNATURE----- From bwalton at opencsw.org Fri May 8 21:39:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 15:39:59 -0400 Subject: [csw-maintainers] issue when trying to provide a python binding In-Reply-To: <4A048902.30400@opencsw.org> References: <4A045CA3.5070908@opencsw.org> <4A046644.8090701@opencsw.org> <20090508184433.GA96779@bolthole.com> <1241808516-sup-4065@ntdws12.chass.utoronto.ca> <4A048902.30400@opencsw.org> Message-ID: <1241811478-sup-1525@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Fri May 08 15:33:22 -0400 2009: Hi Mike, > Ruby compiles fine with Studio 12. Interesting! :) Thanks for testing that. > we could just skip solaris8 ;) ;) I can't, as that's still my production rails box!! I will do a little happy dance of joy when I no longer have any sol8 boxes around. It's been a workhorse, but it's tired now. If I get a few spare cycles, I'll try the interposing of a wrapper script to see about resolve the Studio 11 issues. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Fri May 8 21:52:29 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 14:52:29 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <4A046503.4010702@opencsw.org> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <4A046503.4010702@opencsw.org> Message-ID: <4A048D7D.2070807@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Phil, Mike Watters wrote: > Ben Walton wrote: >> Excerpts from Peter FELECAN's message of Fri May 08 12:39:31 -0400 2009: >>> I'm voting to move php5 out of /opt/csw/php5 and retire PHP4 or, at >>> least, put it in /opt/csw/php4. >> I definitely agree that php5 should become the default. I argued last >> time that php4 should be removed from the catalog too, but Phil >> thought it was best to keep it at that time. If I changed my argument >> to 'remove it from the catalog, but leave it on the mirrors' would >> change your opinion Phil? Users would be free to manually download >> and install it, but it wouldn't be supported directly any more? > > > I no longer see php4 core on the package page? I see the php4_* modules but no > php or php4. > What is the verdict on php4 vs php5? I found the php binaries in php4_cgi should I move php5 to the default php? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoEjX0ACgkQLrhmsXMSLxc8MQCgvBGa36bpLZnlk4/JrIjzl+uB /bwAoJlk/xZ7g6Hv82iVNkEIBtqaFdJe =syKm -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 8 23:23:22 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 8 May 2009 14:23:22 -0700 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241801752-sup-8737@ntdws12.chass.utoronto.ca> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> Message-ID: <20090508212322.GD63097@bolthole.com> On Fri, May 08, 2009 at 12:57:59PM -0400, Ben Walton wrote: > I argued last > time that php4 should be removed from the catalog too, but Phil > thought it was best to keep it at that time. do you remember the specifics of WHY? i have forgotten. > If I changed my argument > to 'remove it from the catalog, but leave it on the mirrors' would > change your opinion Phil? That makes no sense. if we keep it, it stays in the catalog. its worth discussing what should be /opt/csw/bin/php though. I am in favour of it being php5. it doesnt look like there are any dependancies of OURS that use it. However, it would need to be carefully announced ahead of time to users. From bwalton at opencsw.org Sat May 9 02:59:46 2009 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 08 May 2009 20:59:46 -0400 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <20090508212322.GD63097@bolthole.com> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> Message-ID: <1241826545-sup-6478@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Fri May 08 17:23:22 -0400 2009: > do you remember the specifics of WHY? > i have forgotten. You didn't think it was hurting anything to keep it as is. My view is that since it's dead upstream and our release is behind their last security release it should be dropped...I don't even think upping to the last 4.4.9 release is a reasonable time investment at this point. > > If I changed my argument to 'remove it from the catalog, but > > leave it on the mirrors' would change your opinion Phil? > > That makes no sense. > if we keep it, it stays in the catalog. I was just trying to provide a way for people that want it (for whatever reason) to be able to easily grab it without it being a supported package with the csw tool chain. Drop it is my official stance. > its worth discussing what should be /opt/csw/bin/php though. > > I am in favour of it being php5. That's my vote, too. I don't even think this is a real question at this point. > it doesnt look like there are any dependancies of OURS that use it. > However, it would need to be carefully announced ahead of time to users. Agreed. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Sat May 9 03:56:49 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 08 May 2009 20:56:49 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <1241826545-sup-6478@ntdws12.chass.utoronto.ca> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> <1241826545-sup-6478@ntdws12.chass.utoronto.ca> Message-ID: <4A04E2E1.3020701@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Philip Brown's message of Fri May 08 17:23:22 -0400 2009: >> do you remember the specifics of WHY? >> i have forgotten. > > You didn't think it was hurting anything to keep it as is. My view is > that since it's dead upstream and our release is behind their last > security release it should be dropped...I don't even think upping to > the last 4.4.9 release is a reasonable time investment at this point. > >>> If I changed my argument to 'remove it from the catalog, but >>> leave it on the mirrors' would change your opinion Phil? >> That makes no sense. >> if we keep it, it stays in the catalog. > > I was just trying to provide a way for people that want it (for > whatever reason) to be able to easily grab it without it being a > supported package with the csw tool chain. Drop it is my official > stance. > >> its worth discussing what should be /opt/csw/bin/php though. >> >> I am in favour of it being php5. > > That's my vote, too. I don't even think this is a real question at > this point. > >> it doesnt look like there are any dependancies of OURS that use it. >> However, it would need to be carefully announced ahead of time to users. > > Agreed. OK, Phil, can you put out a press release regarding php4 decommission. I will handle php5. Here is my plan. I will keep the 5.2.9 version "as is" ... after the bug-fix and feature request I am currently compiling. I will make an announcement that as of release 5.3 ... road map for Q2 2009 ... we are moving the installation to the "default" location. thoughts? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoE4uEACgkQLrhmsXMSLxeHYwCgkNx1kQ0uXbvHGK4hF12Ivp+G NuMAoIXI2okjW6CYWLoiONrhkQZgOisA =GdnY -----END PGP SIGNATURE----- From ihsan at opencsw.org Sat May 9 12:03:26 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 12:03:26 +0200 Subject: [csw-maintainers] Odd issues with the new IMAP In-Reply-To: <4A045DB6.90104@opencsw.org> References: <4A045DB6.90104@opencsw.org> Message-ID: <4A0554EE.7050001@opencsw.org> Hello Mike, Am 8.5.2009 18:28 Uhr, Mike Watters schrieb: > I am using thunderbird/enigmail/imap for my opencsw account. > Since the upgrade, I am getting Messages that are showing up > at different times. I'm using the same combination. > for example: at 11:00 my time I received 4 messages from earlier in the day > 08:32 08:48 09:23 and 10:47 "mytime" > > this is not really a problem, just Odd and only after the imap upgrade. > > anyone else experiencing issues similar? I haven't experienced something similar. Did you had a look into the mail header? Maybe the mails were stuck somewhere. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From rupert at opencsw.org Sat May 9 13:47:00 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 9 May 2009 13:47:00 +0200 Subject: [csw-maintainers] mod_wsgi, how to get it from testing into release? Message-ID: <6af4270905090447g1841defbj3519fc8f481eb27e@mail.gmail.com> how does one get mod_wsgi from http://mirror.opencsw.org/testing.html into http://opencsw.org/packages ? we tested it successful here by: pkgutil -U pkgutil -u pkgutil -t http://mirror.opencsw.org/opencsw/testing -i ap2_modwsgi. the last command offered to upgrade other packages as well, but i did not says yes to stay on the current productive version. then we did a successful test with: - native installed trac-0.11.4, using csw subversion-1.6.1, displaying svn-1.5 format repositories. - native installed mercurial-1.1 rupert. From rupert at opencsw.org Sat May 9 13:49:25 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 9 May 2009 13:49:25 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> Message-ID: <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> hi, while subversion-1.6.1 works fine with the old repository format, we hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after upgrading to 1.6 repository format. the fix will be in 1.6.2. i'll give it a try. rupert. ------------------------------------------------------------------------------ From: "Hyrum K. Wright" Date: Thu, 7 May 2009 11:49:25 -0500 Local: Thurs, May 7 2009 6:49 pm Subject: 1.6.2 tarballs up for signing/testing I'm pleased to announce that Subversion 1.6.2 is up for testing and signing. ?The magic revision is r37639. ?You can find the tarballs here: http://orac.ece.utexas.edu/pub/svn/1.6.2/ Please be sure to test the bindings. You know the drill: signatures from full committers back to me, and enthusiastic tester feedback is welcome. ?At this point, this candidate is not yet blessed for wide release, so please don't make it available to people not interested in test-driving the new release. Distro package maintainers, please do NOT include any pre-release builds, even blessed, into operating system distros. ?The reasons for not doing so were very eloquently outlined by Karl in a mail, which is summarized at the above address. The quick version is: we don't guarantee compatibility between the pre- releases and the final release, so if people install the release candidate, all their repositories and working copies might break irreparably when they upgrade to 1.6.2 proper. ?We don't want that kind of bad publicity, and neither do you. -Hyrum ------------------------------------------------------ http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessag... From mwatters at opencsw.org Sat May 9 16:19:36 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 09 May 2009 09:19:36 -0500 Subject: [csw-maintainers] Odd issues with the new IMAP In-Reply-To: <4A0554EE.7050001@opencsw.org> References: <4A045DB6.90104@opencsw.org> <4A0554EE.7050001@opencsw.org> Message-ID: <4A0590F8.6030205@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ihsan, Ihsan Dogan wrote: > > I haven't experienced something similar. Did you had a look into the > mail header? Maybe the mails were stuck somewhere. I didn't check the headers, I guess I should have, but it is not a problem I just thought it odd. I have really only seen it once, right after the conversion. if I do "catch" it again, I will double check the headers and gather as much info as possible. just so we can verify it is not an issue with dovecot, I doubt very much it would be a dovecot issue. but, my first thought was a minor bug in the re-indexing dovecot does. possibly holding new mail until the re-index was complete. Just wanted to see if anyone else had seen that. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoFkPcACgkQLrhmsXMSLxd2FQCgt+IXut/1vkkKWHwg45rLSpyG 0H4An2GMu7mpV2eRdxhw6HmTlNs7uXEs =Tq/N -----END PGP SIGNATURE----- From mwatters at opencsw.org Sat May 9 16:30:03 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 09 May 2009 09:30:03 -0500 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> Message-ID: <4A05936B.3020001@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > hi, > > while subversion-1.6.1 works fine with the old repository format, we > hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after > upgrading to 1.6 repository format. > > the fix will be in 1.6.2. i'll give it a try. This sounds eerily similar to http://www.opencsw.org/mantis/view.php?id=3661 http://subversion.tigris.org/issues/show_bug.cgi?id=3387 I will patch up 1.6.1 with the fix for our bug 3361. if you wouldn't mind testing that when released against your bug? if it does not fix it, please file a new opencsw bug. I hope to have the fix for subversion done next week. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw =Du0U -----END PGP SIGNATURE----- From ihsan at opencsw.org Sat May 9 17:30:40 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 17:30:40 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp Message-ID: <4A05A1A0.3090702@opencsw.org> Hello, Back in December we have been talking about a summer camp in Norway. I think it's now time to talk about the topics for the summer camp. I would like to open the discussion now. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat May 9 17:33:19 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 17:33:19 +0200 Subject: [csw-maintainers] Webmail: Squirrelmail now active again Message-ID: <4A05A23F.40704@opencsw.org> Hello, After having a lot of troubles with Roundcube, I switch back to Squirrelmail. --> https://mail.opencsw.org/ If you still would like to use Roundcube, please point your browser to https://mail.opencsw.org/roundcubemail/ . Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sat May 9 17:39:31 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sat, 09 May 2009 17:39:31 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <4A05A3B3.3070508@opencsw.org> Am 9.5.2009 17:30 Uhr, Ihsan Dogan schrieb: > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. A possible topic would be the website. There is already some work done and it would be a good chance to move forward. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From william at wbonnet.net Sat May 9 19:24:18 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 09 May 2009 19:24:18 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <4A05BC42.4080201@wbonnet.net> Hi > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. > I would be interested in particpating. It will depend of the date... I should be able to find some time in July (end) or first weeks of August. cheers W. From Murray.Jensen at csiro.au Sun May 10 10:47:20 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Sun, 10 May 2009 18:47:20 +1000 Subject: [csw-maintainers] suncsw domain names expiring Message-ID: <9637.1241945240@gerd> Hi CSW people, all the domain names I registered in anticipation of the CSW split are about to expire - namely sun{,-}csw.{org,net,com}. I was just going to allow them to expire, but I thought I should let everyone know. Cheers! Murray... -- Murray Jensen, CSIRO Materials Science and Engineering Phone: +61 3 9545 2075 Private Bag 33, Clayton South 3169, Australia Fax: +61 3 9544 1128 Internet: Murray.Jensen at csiro.au To the extent permitted by law, CSIRO does not represent, warrant and/or guarantee that the integrity of this communication has been maintained or that the communication is free of errors, virus, interception or interference. The information contained in this e-mail may be confidential or privileged. Any unauthorised use or disclosure is prohibited. If you have received this e-mail in error, please delete it immediately and notify Murray Jensen on +61 3 9545 2075. Thank you. From dam at opencsw.org Sun May 10 15:38:03 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 15:38:03 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website Message-ID: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Hi, I just found out that there is a Java Chatclient at Freenode. How about putting on the webpage to let interested people join immediately? Best regards -- Dago From dam at opencsw.org Sun May 10 15:48:41 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 15:48:41 +0200 Subject: [csw-maintainers] suncsw domain names expiring In-Reply-To: <9637.1241945240@gerd> References: <9637.1241945240@gerd> Message-ID: Hi Murray, Am 10.05.2009 um 10:47 schrieb : > Hi CSW people, all the domain names I registered in anticipation of > the CSW > split are about to expire - namely sun{,-}csw.{org,net,com}. I was > just going > to allow them to expire, but I thought I should let everyone know. > Cheers! That is IMHO okay. I won't extend suncsw.de either. They have done their purpose and opencsw.org is IMHO just right. Best regards -- Dago From bwalton at opencsw.org Sun May 10 16:30:35 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 10 May 2009 10:30:35 -0400 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Message-ID: <1241965727-sup-4496@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Sun May 10 09:38:03 -0400 2009: > I just found out that there is a Java Chatclient at Freenode. How about > putting > > on the webpage to let interested people join immediately? I think it'd be good. It might be a good question forum for some. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Sun May 10 18:03:42 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 10 May 2009 18:03:42 +0200 Subject: [csw-maintainers] Is Hudson still available? Message-ID: <4A06FADE.8060707@opencsw.org> Hi, just wondering: Is Hudson still available / functional? When i go to http://buildfarm.opencsw.org/hudson/ i just get a 503: Service Temporarily Unavailable The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later. Sebastian From trygvis at opencsw.org Sun May 10 18:52:51 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 18:52:51 +0200 Subject: [csw-maintainers] Is Hudson still available? In-Reply-To: <4A06FADE.8060707@opencsw.org> References: <4A06FADE.8060707@opencsw.org> Message-ID: <4A070663.3050807@opencsw.org> Sebastian Kayser wrote: > Hi, > > just wondering: Is Hudson still available / functional? When i go to > http://buildfarm.opencsw.org/hudson/ i just get a 503: > > Service Temporarily Unavailable > > The server is temporarily unable to service your request due to > maintenance downtime or capacity problems. Please try again later. The instance has probably been rebooted, should be working again now. -- Trygve From trygvis at opencsw.org Sun May 10 18:54:28 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 18:54:28 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Message-ID: <4A0706C4.4050105@opencsw.org> Dagobert Michelsen wrote: > Hi, > > I just found out that there is a Java Chatclient at Freenode. How about > putting > > on the webpage to let interested people join immediately? I like the idea, but think mibbit[1] is a better client as it don't require anything other than a browser. The applet integration on Solaris has been a bit shady in my experience. [1]: http://www.mibbit.com/ -- Trygve From trygvis at opencsw.org Sun May 10 18:57:22 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 18:57:22 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <4A070772.2010206@opencsw.org> Ihsan Dogan wrote: > Hello, > > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. I'm interested, but I doubt Norway is the best place to go. It will require many people to travel far and stuff is generally considered expensive in Norway. I can get access to conference rooms and work facilities with connectivity for free though. -- Trygve From ihsan at opencsw.org Sun May 10 19:28:06 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:28:06 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A070772.2010206@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> Message-ID: <4A070EA6.3060307@opencsw.org> Am 10.5.2009 18:57 Uhr, Trygve Laugst?l schrieb: >> Back in December we have been talking about a summer camp in Norway. I >> think it's now time to talk about the topics for the summer camp. I >> would like to open the discussion now. > I'm interested, but I doubt Norway is the best place to go. It will > require many people to travel far and stuff is generally considered > expensive in Norway. I can get access to conference rooms and work > facilities with connectivity for free though. The pricing looks like in Switzerland. For me personally it doesn't matter, but if it should take place somewhere else, if someone would volunteer now. We should know the place and the date 2 months in advance. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 10 19:29:22 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:29:22 +0200 Subject: [csw-maintainers] suncsw domain names expiring In-Reply-To: References: <9637.1241945240@gerd> Message-ID: <4A070EF2.8050402@opencsw.org> Am 10.5.2009 15:48 Uhr, Dagobert Michelsen schrieb: >> Hi CSW people, all the domain names I registered in anticipation of >> the CSW >> split are about to expire - namely sun{,-}csw.{org,net,com}. I was >> just going >> to allow them to expire, but I thought I should let everyone know. >> Cheers! > That is IMHO okay. I won't extend suncsw.de either. They have done their > purpose > and opencsw.org is IMHO just right. I'm going to remove suncsw.org from our name server. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 10 19:45:30 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:45:30 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0706C4.4050105@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A0706C4.4050105@opencsw.org> Message-ID: <4A0712BA.9070504@opencsw.org> Am 10.5.2009 18:54 Uhr, Trygve Laugst?l schrieb: >> I just found out that there is a Java Chatclient at Freenode. How about >> putting >> >> on the webpage to let interested people join immediately? > > I like the idea, but think mibbit[1] is a better client as it don't > require anything other than a browser. The applet integration on Solaris > has been a bit shady in my experience. > > [1]: http://www.mibbit.com/ Does this need anything to be installed on the web server? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 10 19:48:05 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 19:48:05 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> Message-ID: <4A071355.3020204@opencsw.org> Am 10.5.2009 15:38 Uhr, Dagobert Michelsen schrieb: > I just found out that there is a Java Chatclient at Freenode. How about > putting > > on the webpage to let interested people join immediately? For the internal use, it might be handy to have a Instant Messenger contact list. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From trygvis at opencsw.org Sun May 10 20:11:48 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 20:11:48 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0712BA.9070504@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A0706C4.4050105@opencsw.org> <4A0712BA.9070504@opencsw.org> Message-ID: <4A0718E4.9090400@opencsw.org> Ihsan Dogan wrote: > Am 10.5.2009 18:54 Uhr, Trygve Laugst?l schrieb: > >>> I just found out that there is a Java Chatclient at Freenode. How about >>> putting >>> >>> on the webpage to let interested people join immediately? >> I like the idea, but think mibbit[1] is a better client as it don't >> require anything other than a browser. The applet integration on Solaris >> has been a bit shady in my experience. >> >> [1]: http://www.mibbit.com/ > > Does this need anything to be installed on the web server? No, we don't have to do anything. Just explain how to use it to connect to our room, like server name and channel name. I've never used it, but it is supposedly nice. -- Trygve From trygvis at inamo.no Sun May 10 20:11:02 2009 From: trygvis at inamo.no (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Sun, 10 May 2009 20:11:02 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A071355.3020204@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> Message-ID: <4A0718B6.2000903@inamo.no> Ihsan Dogan wrote: > Am 10.5.2009 15:38 Uhr, Dagobert Michelsen schrieb: > >> I just found out that there is a Java Chatclient at Freenode. How about >> putting >> >> on the webpage to let interested people join immediately? > > For the internal use, it might be handy to have a Instant Messenger > contact list. I don't mind sharing my MSN or Jabber IDs in a private file somwhere. -- Trygve From dam at opencsw.org Sun May 10 20:28:09 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 20:28:09 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <49A5CFAA.2010806@wbonnet.net> References: <49A5CFAA.2010806@wbonnet.net> Message-ID: <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> Hi, I have now rebuild all the X11 packages from William, pixman, cairo, pango, libxft2, libatk and gtk+ in GAR with 32/64 bit. The packages are not in testing as they have been build on build8st instead of the official build machines. The following dependencies are still missing, but I consider them not that relevant, so they may be added later: - openjpeg (uses statical Makefile optimized for Linux) (requires defining loads of variables in the Makefile) - poppler (broken runtimelinkerpath, needs openjpeg) (needs some inspection, shouldn't be too hard) - spectre (needs 64 bit libgs.o, which needs pixman etc., can be done after general package release in a second step) All packages have been installed on build8st, please try compiling and testing packages against them. After this was successfull I am going to package and release them one-by-one as there are massive dependencies between them. I don't want to take over packages from William, how about a technical user "gnometeam" releasing the packages and receiving the mails with all gnome maintainers on it? Am 26.02.2009 um 00:09 schrieb William Bonnet: > I have moved the X11 stuff to /opt/csw/X11 (but licenses which are > under > /opt/csw/share/$(GARNAME). The existing libxrender package puts shared libraries in /opt/csw/lib: I guess if they are relocated to /opt/csw/x11/lib they can no longer be found. It would be however good to bundle X11 libs in /opt/csw/lib/ X11. Make links from lib/X11 to lib/, issue bug reports and remove the links when all bugs have been closed? Best regards -- Dago From skayser at opencsw.org Sun May 10 21:42:01 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 10 May 2009 21:42:01 +0200 Subject: [csw-maintainers] Is Hudson still available? In-Reply-To: <4A070663.3050807@opencsw.org> References: <4A06FADE.8060707@opencsw.org> <4A070663.3050807@opencsw.org> Message-ID: <4A072E09.8010203@opencsw.org> Trygve Laugst?l wrote: > Sebastian Kayser wrote: >> just wondering: Is Hudson still available / functional? When i go to >> http://buildfarm.opencsw.org/hudson/ i just get a 503: >> >> Service Temporarily Unavailable >> >> The server is temporarily unable to service your request due to >> maintenance downtime or capacity problems. Please try again later. > > The instance has probably been rebooted, should be working again now. Web page is accessible again, but both "build executors" are marked as offline (with several jobs pending). Could you have a look at it please? Sebastian From phil at bolthole.com Sun May 10 21:45:29 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 10 May 2009 12:45:29 -0700 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> Message-ID: <20090510194529.GD96486@bolthole.com> On Sun, May 10, 2009 at 08:28:09PM +0200, Dagobert Michelsen wrote: > I don't want to take over packages from William, how about a technical > user "gnometeam" releasing the packages and receiving the mails > with all gnome maintainers on it? trouble with that is, there's a tendancy for one person to always think, "well, the other person will update it... From mwatters at opencsw.org Sun May 10 22:37:06 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 10 May 2009 15:37:06 -0500 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <20090510194529.GD96486@bolthole.com> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> Message-ID: <4A073AF2.2000309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Sun, May 10, 2009 at 08:28:09PM +0200, Dagobert Michelsen wrote: >> I don't want to take over packages from William, how about a technical >> user "gnome team" releasing the packages and receiving the mails >> with all gnome maintainers on it? > > trouble with that is, there's a tendency for one person to always think, > "well, the other person will update it... I have to agree, reluctantly, with Philip. However, I do think a separate mailing list for the Big Projects, like GNOME, KDE... etc. with all the maintainers on them. that way those that would like to follow the progress can, and those that do not, can filter out the incoming mail. This would also allow the other maintainers to follow the progress and make it easier to contribute free cycles if they so choose. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoHOvIACgkQLrhmsXMSLxdEtgCgjm6EP9BJQ8gnYz8+395R3Osl WpgAoJt1BmF4LfHOzXGAtNWhT95s9Uai =HIeO -----END PGP SIGNATURE----- From william at wbonnet.net Sun May 10 22:45:42 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 10 May 2009 22:45:42 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <20090510194529.GD96486@bolthole.com> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> Message-ID: <4A073CF6.6070201@wbonnet.net> Hi >> I don't want to take over packages from William, how about a technical >> user "gnometeam" releasing the packages and receiving the mails >> with all gnome maintainers on it? >> > > trouble with that is, there's a tendancy for one person to always think, > "well, the other person will update it... > I've been silent for a long time. Sorry for that. I proposed the idea of teams several times, and we should really consider to apply it. I don't agree with the argument of "well, the other person will update it...". We are going to create team of twenty persons, and communication will be easy. Even if sometime someone is off for different reasons, like an "office overload", hollidays, or whatever, it is still possible to communicate within the team, or to announce we are off for a few weeks with internet. The gnome team is one of the team we need to set up soon, mozilla should follow i think (and maybe kde). I am right now fixing bugs on firefox and finishing TB package in GAR, but it takes time, a lot of time, and i may be overloaded some time when it comes to release some security fix. I would appreciate someone else than me could do it. Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From william at wbonnet.net Sun May 10 22:47:27 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 10 May 2009 22:47:27 +0200 Subject: [csw-maintainers] Mantis login problem Message-ID: <4A073D5F.4030302@wbonnet.net> Hi It seems like i cannot login into mantis. Any known problem ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From skayser at opencsw.org Sun May 10 22:49:39 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 10 May 2009 22:49:39 +0200 Subject: [csw-maintainers] Mantis login problem In-Reply-To: <4A073D5F.4030302@wbonnet.net> References: <4A073D5F.4030302@wbonnet.net> Message-ID: <4A073DE3.3080702@opencsw.org> William Bonnet wrote: > It seems like i cannot login into mantis. Any known problem ? Works for me. Any error messages? Sebastian From william at wbonnet.net Sun May 10 23:03:40 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 10 May 2009 23:03:40 +0200 Subject: [csw-maintainers] Mantis login problem In-Reply-To: <4A073DE3.3080702@opencsw.org> References: <4A073D5F.4030302@wbonnet.net> <4A073DE3.3080702@opencsw.org> Message-ID: <4A07412C.7090804@wbonnet.net> Hi Sebastian >> It seems like i cannot login into mantis. Any known problem ? >> > > Works for me. Any error messages? > Nope... login button is inactive I tested with a different browser and it works with seamonkey ! I am going to have a look to my environement cheers -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From phil at bolthole.com Sun May 10 23:20:39 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 10 May 2009 14:20:39 -0700 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A073CF6.6070201@wbonnet.net> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> Message-ID: <20090510212039.GA28401@bolthole.com> On Sun, May 10, 2009 at 10:45:42PM +0200, William Bonnet wrote: > I don't agree with the argument of "well, the other person will update > it...". We are going to create team of twenty persons, and communication > will be easy. communication IS easy. responsability (in an organization consisting of unpaid people) is hard. > follow i think (and maybe kde). I am right now fixing bugs on firefox > and finishing TB package in GAR, but it takes time, a lot of time, and i > may be overloaded some time when it comes to release some security fix. > I would appreciate someone else than me could do it. that is doable. ON the flip side, that does not require having group "ownership" of an individual package. all that is required, is being known as an oficial member of "the gnome team", and then for the release manager (ie: me) to allow automatic takeover of gnome packages, between gnome team members. From dam at opencsw.org Sun May 10 23:27:24 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 10 May 2009 23:27:24 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <20090510212039.GA28401@bolthole.com> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> Message-ID: <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> Hi, Am 10.05.2009 um 23:20 schrieb Philip Brown: > On Sun, May 10, 2009 at 10:45:42PM +0200, William Bonnet wrote: >> I don't agree with the argument of "well, the other person will >> update >> it...". We are going to create team of twenty persons, and >> communication >> will be easy. > > communication IS easy. responsability (in an organization consisting > of > unpaid people) is hard. > >> follow i think (and maybe kde). I am right now fixing bugs on firefox >> and finishing TB package in GAR, but it takes time, a lot of time, >> and i >> may be overloaded some time when it comes to release some security >> fix. >> I would appreciate someone else than me could do it. > > that is doable. ON the flip side, that does not require having group > "ownership" of an individual package. > all that is required, is being known as an oficial member of "the > gnome > team", and then for the release manager (ie: me) to allow automatic > takeover of gnome packages, between gnome team members. Ok then. Please someone take a look at build8st and I need a decision on linking X11/lib back to lib/ for the interimtime. Then the packages will be released. Ihsan: Please test rrdtool against the current cairo and see if the error goes away. Bets regards -- Dagp From ihsan at opencsw.org Sun May 10 23:30:38 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 23:30:38 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0718B6.2000903@inamo.no> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0718B6.2000903@inamo.no> Message-ID: <4A07477E.2030607@opencsw.org> Am 10.5.2009 20:11 Uhr, Trygve Laugst?l schrieb: >>> I just found out that there is a Java Chatclient at Freenode. How about >>> putting >>> >>> on the webpage to let interested people join immediately? >> >> For the internal use, it might be handy to have a Instant Messenger >> contact list. > > I don't mind sharing my MSN or Jabber IDs in a private file somwhere. Ok, I will try to create such a list. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From rupert at opencsw.org Sun May 10 23:31:41 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 10 May 2009 23:31:41 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <4A05936B.3020001@opencsw.org> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> <4A05936B.3020001@opencsw.org> Message-ID: <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> you are right, i put a hint also into the mantis issue. i tested with the upcoming 1.6.2 release, and it works. i left it on testing for now, but we should remove it as soon as possible (http://mirror.opencsw.org/testing/subversion-1.6.2,REV=2009.05.09-SunOS5.8-sparc-UNCOMMITTED.pkg.gz ). as 1.6.2 is already tagged we might wait the couple of days for the final 1.6.2 release instead of patching 1.6.1. rupert. On Sat, May 9, 2009 at 16:30, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > rupert THURNER wrote: >> hi, >> >> while subversion-1.6.1 works fine with the old repository format, we >> hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after >> upgrading to 1.6 repository format. >> >> the fix will be in 1.6.2. i'll give it a try. > > This sounds eerily similar to > http://www.opencsw.org/mantis/view.php?id=3661 > http://subversion.tigris.org/issues/show_bug.cgi?id=3387 > > I will patch up 1.6.1 with the fix for our bug 3361. ?if you wouldn't mind > testing that when released against your bug? > > if it does not fix it, please file a new opencsw bug. > I hope to have the fix for subversion done next week. > > - -- > Thanks, > Mike > > "Any intelligent fool can make things bigger, more complex, > and more violent. ?It takes a touch of genius -- and a lot of courage -- > to move in the opposite direction." > > * Albert Einstein 1879 - 1955 > ? ?US German-born Theoretical Physicist > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA > 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw > =Du0U > -----END PGP SIGNATURE----- > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > From rupert at opencsw.org Sun May 10 23:39:31 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 10 May 2009 23:39:31 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> <4A05936B.3020001@opencsw.org> <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> Message-ID: <6af4270905101439j6eee7dc2je4f5fe796515feb9@mail.gmail.com> because we are just talking about svn-1.6.X. one of the main features is that "bad clients" are now filtered out. but, this may lead to difficulties with non-versioned properties containing mixed end of line styles. they could be caused by eclipse subclipse e.g. when changing the svn log message. see http://svn.haxx.se/users/archive-2009-03/0797.shtml for details. rupert. On Sun, May 10, 2009 at 23:31, rupert THURNER wrote: > you are right, i put a hint also into the mantis issue. > > i tested with the upcoming 1.6.2 release, and it works. i left it on > testing for now, but we should remove it as soon as possible > (http://mirror.opencsw.org/testing/subversion-1.6.2,REV=2009.05.09-SunOS5.8-sparc-UNCOMMITTED.pkg.gz > ). > > as 1.6.2 is already tagged we might wait the couple of days for the > final 1.6.2 release instead of patching 1.6.1. > > rupert. > > > On Sat, May 9, 2009 at 16:30, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> rupert THURNER wrote: >>> hi, >>> >>> while subversion-1.6.1 works fine with the old repository format, we >>> hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after >>> upgrading to 1.6 repository format. >>> >>> the fix will be in 1.6.2. i'll give it a try. >> >> This sounds eerily similar to >> http://www.opencsw.org/mantis/view.php?id=3661 >> http://subversion.tigris.org/issues/show_bug.cgi?id=3387 >> >> I will patch up 1.6.1 with the fix for our bug 3361. ?if you wouldn't mind >> testing that when released against your bug? >> >> if it does not fix it, please file a new opencsw bug. >> I hope to have the fix for subversion done next week. >> >> - -- >> Thanks, >> Mike >> >> "Any intelligent fool can make things bigger, more complex, >> and more violent. ?It takes a touch of genius -- and a lot of courage -- >> to move in the opposite direction." >> >> * Albert Einstein 1879 - 1955 >> ? ?US German-born Theoretical Physicist >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.9 (GNU/Linux) >> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org >> >> iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA >> 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw >> =Du0U >> -----END PGP SIGNATURE----- >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers >> > From ihsan at opencsw.org Sun May 10 23:49:48 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 10 May 2009 23:49:48 +0200 Subject: [csw-maintainers] IM Contacts Message-ID: <4A074BFC.503@opencsw.org> Hello, I'm trying to make a a list with the Instant Messanging contact of the OpenCSW maintainers. This list will be not public and only accessible to the maintainers. Please fill out this form and send it back to me personally (the list is public). AIM / ICQ / iChat : Jabber / Google Talk : MSN : Yahoo : Skype : Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 03:48:07 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 03:48:07 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: <4A077AE6.8050403@opencsw.org> References: <4A077AE6.8050403@opencsw.org> Message-ID: <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Hi, Am 11.05.2009 um 03:09 schrieb Mike Watters: > Please upgrade the buildfarm All buildservers are upgraded to current/ now. Please stand by. This include GCC :-) Best regards -- Dago From phil at bolthole.com Mon May 11 04:34:13 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 10 May 2009 19:34:13 -0700 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <4A04E2E1.3020701@opencsw.org> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> <1241826545-sup-6478@ntdws12.chass.utoronto.ca> <4A04E2E1.3020701@opencsw.org> Message-ID: <20090511023413.GA83321@bolthole.com> On Fri, May 08, 2009 at 08:56:49PM -0500, Mike Watters wrote: > > OK, Phil, can you put out a press release regarding php4 decommission. I shall do so. and lets be specific: we are "decommissioning" it from being /opt/csw/bin/php. The package will still be around, in legacy mode. This will require a repackage of the thing, to remove the old /opt/csw/bin/php reference. However, to "properly" retire php4, we really should repackage it at the last stable released version as well. Any volunteers? please? in theory, I could do it myself... but realistically speaking, if this is going to be waiting on me... it culd be waiting a loong time. From mwatters at opencsw.org Mon May 11 05:02:19 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 10 May 2009 22:02:19 -0500 Subject: [csw-maintainers] OpenCSW default PHP version In-Reply-To: <20090511023413.GA83321@bolthole.com> References: <1241797963-sup-8059@ntdws12.chass.utoronto.ca> <4A045E20.30306@opencsw.org> <1241801752-sup-8737@ntdws12.chass.utoronto.ca> <20090508212322.GD63097@bolthole.com> <1241826545-sup-6478@ntdws12.chass.utoronto.ca> <4A04E2E1.3020701@opencsw.org> <20090511023413.GA83321@bolthole.com> Message-ID: <4A07953B.9070808@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Fri, May 08, 2009 at 08:56:49PM -0500, Mike Watters wrote: >> OK, Phil, can you put out a press release regarding php4 decommission. > > I shall do so. > > and lets be specific: we are "decommissioning" it from being > /opt/csw/bin/php. > The package will still be around, in legacy mode. > > This will require a repackage of the thing, to remove the old > /opt/csw/bin/php reference. > > However, to "properly" retire php4, we really should repackage it at the > last stable released version as well. > > > Any volunteers? please? I will do it, theoretically I should be able to use the php5 recipe with just a few minor modifications. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoHlTsACgkQLrhmsXMSLxcnLACgsCrvpZVDQLVDRimM+bJdHNvI mh0AoLFbFZITycQQP+EVBOnHezrSVJ3s =eqpt -----END PGP SIGNATURE----- From dam at opencsw.org Mon May 11 05:25:06 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 05:25:06 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: <8E3F475C-2A34-4AE1-906B-652DFCF1E2E7@opencsw.org> Hi, Am 11.05.2009 um 03:48 schrieb Dagobert Michelsen: > Am 11.05.2009 um 03:09 schrieb Mike Watters: >> Please upgrade the buildfarm > > All buildservers are upgraded to current/ now. Please stand by. > This include GCC :-) Done. This excludes build8st and build8xt. Best regards -- Dago From pfelecan at opencsw.org Mon May 11 11:06:57 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 11:06:57 +0200 Subject: [csw-maintainers] IM Contacts In-Reply-To: <4A074BFC.503@opencsw.org> (Ihsan Dogan's message of "Sun\, 10 May 2009 23\:49\:48 +0200") References: <4A074BFC.503@opencsw.org> Message-ID: Ihsan Dogan writes: > Hello, > > I'm trying to make a a list with the Instant Messanging contact of the > OpenCSW maintainers. This list will be not public and only accessible to > the maintainers. How you ensure that the list is not public and accessible only to maintainers? -- Peter From pfelecan at opencsw.org Mon May 11 11:09:36 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 11:09:36 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> (Dagobert Michelsen's message of "Mon\, 11 May 2009 03\:48\:07 +0200") References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: Dagobert Michelsen writes: > Am 11.05.2009 um 03:09 schrieb Mike Watters: >> Please upgrade the buildfarm > > All buildservers are upgraded to current/ now. Please stand by. > This include GCC :-) Is this reasonable when the packages didn't hit yet the mirrors? My only issue is that I cannot reproduce the same environment on my build stack. -- Peter From dam at opencsw.org Mon May 11 12:24:57 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 12:24:57 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: Hi Peter, Am 11.05.2009 um 11:09 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >> Am 11.05.2009 um 03:09 schrieb Mike Watters: >>> Please upgrade the buildfarm >> >> All buildservers are upgraded to current/ now. Please stand by. >> This include GCC :-) > > Is this reasonable when the packages didn't hit yet the mirrors? My > only > issue is that I cannot reproduce the same environment on my build > stack. The packages have been pushed to rsync.opencsw.org, all mirrors synchronize from this one, including mirror.opencsw.org. The farm is only updated with released packages. mirror.opencsw.org is updated every 6 hours and manually by me if something is needed fast. You can always pkg-get from that one to reproduce issues of supercurrent packages. Best regards -- Dago From ihsan at opencsw.org Mon May 11 12:29:39 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 12:29:39 +0200 Subject: [csw-maintainers] IM Contacts In-Reply-To: References: <4A074BFC.503@opencsw.org> Message-ID: <4A07FE13.4090601@opencsw.org> Am 11.5.2009 11:06 Uhr, Peter FELECAN schrieb: >> I'm trying to make a a list with the Instant Messanging contact of the >> OpenCSW maintainers. This list will be not public and only accessible to >> the maintainers. > How you ensure that the list is not public and accessible only to > maintainers? I was thinking to do it with mod_auth_pam. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From pfelecan at opencsw.org Mon May 11 12:33:11 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 12:33:11 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: (Dagobert Michelsen's message of "Mon\, 11 May 2009 12\:24\:57 +0200") References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > Am 11.05.2009 um 11:09 schrieb Peter FELECAN: >> Dagobert Michelsen writes: >> >>> Am 11.05.2009 um 03:09 schrieb Mike Watters: >>>> Please upgrade the buildfarm >>> >>> All buildservers are upgraded to current/ now. Please stand by. >>> This include GCC :-) >> >> Is this reasonable when the packages didn't hit yet the mirrors? My >> only >> issue is that I cannot reproduce the same environment on my build >> stack. > > The packages have been pushed to rsync.opencsw.org, all mirrors > synchronize from this one, including mirror.opencsw.org. The > farm is only updated with released packages. > > mirror.opencsw.org is updated every 6 hours and manually by me > if something is needed fast. You can always pkg-get from that > one to reproduce issues of supercurrent packages. Thank you Dago for the explanation. As I use a local mirror, can I update from mirror.opencsw.org with rsync? -- Peter From ihsan at opencsw.org Mon May 11 13:15:12 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 13:15:12 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> Message-ID: <4A0808C0.4020807@opencsw.org> Am 10.5.2009 23:27 Uhr, Dagobert Michelsen schrieb: > Ihsan: Please test rrdtool against the current cairo and see if the error > goes away. Configure complains, that it can't find pangocairo: * I found a copy of pkgconfig, but there is no pangocairo.pc file around. You may want to set the PKG_CONFIG_PATH variable to point to its location. Interesting is, that pkg-config lists pangocairo: ihsan at build8st:~$ pkg-config --list-all | grep pangocairo pangocairo Pango Cairo - Cairo rendering support for Pango sh: gnome-config: not found Package xrender was not found in the pkg-config search path. Perhaps you should add the directory containing `xrender.pc' to the PKG_CONFIG_PATH environment variable Package 'xrender', required by 'Xft', not found Didn't you had a similar issue? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 13:15:42 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 13:15:42 +0200 Subject: [csw-maintainers] [csw-buildfarm] GCC 4.3.3 Released In-Reply-To: References: <4A077AE6.8050403@opencsw.org> <88573DF3-3423-49F9-B3B5-DFE18252B26E@opencsw.org> Message-ID: <6BAEA35E-4656-44E6-BD5C-F8F67F25CCED@opencsw.org> Hi Peter, Am 11.05.2009 um 12:33 schrieb Peter FELECAN: > As I use a local mirror, can I update from mirror.opencsw.org with > rsync? Sure: > web at web [web]:/etc > rsync rsync://mirror.opencsw.org/ > svn GAR SVN Tree > opencsw OpenCSW Software Directory > testing OpenCSW Testing Packages However, please keep in mind that we have only 2 MBit up, so please use this only when necessary and prefer the other sites which also have rsync. Best regards -- Dago From dam at opencsw.org Mon May 11 13:17:29 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 13:17:29 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A0808C0.4020807@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A0808C0.4020807@opencsw.org> Message-ID: Hi Ihsan, Am 11.05.2009 um 13:15 schrieb Ihsan Dogan: > Am 10.5.2009 23:27 Uhr, Dagobert Michelsen schrieb: > >> Ihsan: Please test rrdtool against the current cairo and see if the >> error >> goes away. > > Configure complains, that it can't find pangocairo: > > * I found a copy of pkgconfig, but there is no pangocairo.pc file > around. > You may want to set the PKG_CONFIG_PATH variable to point to its > location. > > Interesting is, that pkg-config lists pangocairo: > > ihsan at build8st:~$ pkg-config --list-all | grep pangocairo > pangocairo Pango Cairo - Cairo rendering support > for Pango > sh: gnome-config: not found > Package xrender was not found in the pkg-config search path. > Perhaps you should add the directory containing `xrender.pc' > to the PKG_CONFIG_PATH environment variable > Package 'xrender', required by 'Xft', not found > > Didn't you had a similar issue? Make sure you add the following lines to use X11-libraries: > EXTRA_INC = $(prefix)/X11/include > EXTRA_LIB = $(prefix)/X11/lib > EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib You may also need these if supported: > CONFIGURE_ARGS += --x-include=$(prefix)/X11/include > CONFIGURE_ARGS += --x-libraries=$(abspath $(prefix)/X11/lib/$ > (MM_LIBDIR)) Best regards -- Dago From bonivart at opencsw.org Mon May 11 13:37:23 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 13:37:23 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A070772.2010206@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> Message-ID: <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> 2009/5/10 Trygve Laugst?l : > I'm interested, but I doubt Norway is the best place to go. It will > require many people to travel far and stuff is generally considered > expensive in Norway. I can get access to conference rooms and work > facilities with connectivity for free though. Norway sounds excellent to me. ;-) -- /peter From ihsan at opencsw.org Mon May 11 13:44:28 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 13:44:28 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A0808C0.4020807@opencsw.org> Message-ID: <4A080F9C.3010009@opencsw.org> Hello Dago, Am 11.5.2009 13:17 Uhr, Dagobert Michelsen schrieb: >> Configure complains, that it can't find pangocairo: >> >> * I found a copy of pkgconfig, but there is no pangocairo.pc file around. >> You may want to set the PKG_CONFIG_PATH variable to point to its >> location. >> >> Interesting is, that pkg-config lists pangocairo: >> >> ihsan at build8st:~$ pkg-config --list-all | grep pangocairo >> pangocairo Pango Cairo - Cairo rendering support for >> Pango >> sh: gnome-config: not found >> Package xrender was not found in the pkg-config search path. >> Perhaps you should add the directory containing `xrender.pc' >> to the PKG_CONFIG_PATH environment variable >> Package 'xrender', required by 'Xft', not found >> >> Didn't you had a similar issue? > > Make sure you add the following lines to use X11-libraries: > >> EXTRA_INC = $(prefix)/X11/include >> EXTRA_LIB = $(prefix)/X11/lib >> EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib That did it. Thanks. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Mon May 11 13:44:29 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 13:44:29 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> Message-ID: <4A080F9D.1050504@opencsw.org> Am 11.5.2009 13:37 Uhr, Peter Bonivart schrieb: >> I'm interested, but I doubt Norway is the best place to go. It will >> require many people to travel far and stuff is generally considered >> expensive in Norway. I can get access to conference rooms and work >> facilities with connectivity for free though. > > Norway sounds excellent to me. ;-) You can reach even by walk. ;-) Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Mon May 11 14:18:46 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 14:18:46 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A080F9C.3010009@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A0808C0.4020807@opencsw.org> <4A080F9C.3010009@opencsw.org> Message-ID: <4A0817A6.6090701@opencsw.org> Am 11.5.2009 13:44 Uhr, Ihsan Dogan schrieb: >>> Configure complains, that it can't find pangocairo: >>> >>> * I found a copy of pkgconfig, but there is no pangocairo.pc file around. >>> You may want to set the PKG_CONFIG_PATH variable to point to its >>> location. >>> >>> Interesting is, that pkg-config lists pangocairo: >>> >>> ihsan at build8st:~$ pkg-config --list-all | grep pangocairo >>> pangocairo Pango Cairo - Cairo rendering support for >>> Pango >>> sh: gnome-config: not found >>> Package xrender was not found in the pkg-config search path. >>> Perhaps you should add the directory containing `xrender.pc' >>> to the PKG_CONFIG_PATH environment variable >>> Package 'xrender', required by 'Xft', not found >>> >>> Didn't you had a similar issue? >> Make sure you add the following lines to use X11-libraries: >> >>> EXTRA_INC = $(prefix)/X11/include >>> EXTRA_LIB = $(prefix)/X11/lib >>> EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib > > That did it. Thanks. Could you please install the new rrdtool on the test machines? http://mirror.opencsw.org/testing/rrdtool-1.3.7,REV=2009.05.11-SunOS5.8-i386-CSW.pkg.gz http://mirror.opencsw.org/testing/rrdtool-1.3.7,REV=2009.05.11-SunOS5.8-sparc-CSW.pkg.gz Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bonivart at opencsw.org Mon May 11 14:27:55 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 14:27:55 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A080F9D.1050504@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> <625385e30905110437i7e1b7aa9l67fb39c4cbbafb07@mail.gmail.com> <4A080F9D.1050504@opencsw.org> Message-ID: <625385e30905110527g178d7926y4a6aefb6757df692@mail.gmail.com> On Mon, May 11, 2009 at 1:44 PM, Ihsan Dogan wrote: > You can reach even by walk. ;-) I think the Norwegians say "g? p? tur". :-) -- /peter From ihsan at opencsw.org Mon May 11 14:48:23 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 14:48:23 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> Message-ID: <4A081E97.5000804@opencsw.org> Hello Dago, Am 10.5.2009 23:27 Uhr, Dagobert Michelsen schrieb: > Ihsan: Please test rrdtool against the current cairo and see if the error > goes away. It looks already much better; at least it's not core dumping anymore. But I'm getting this error: ---------------------------------------------------------- ihsan at build8st:~$ /opt/csw/share/rrdtool/examples/stripes.pl (process:23307): Pango-CRITICAL **: No modules found: No builtin or dynamically loaded modules were found. PangoFc will not work correctly. This probably means there was an error in the creation of: '/opt/csw/etc/pango/pango.modules' You should create this file by running: pango-querymodules > '/opt/csw/etc/pango/pango.modules' (process:23307): Pango-WARNING **: failed to choose a font, expect ugly output. engine-type='PangoRenderFc', script='common' (process:23307): Pango-WARNING **: failed to choose a font, expect ugly output. engine-type='PangoRenderFc', script='latin' This script has created stripes.png in the current directory This demonstrates the use of the TIME and % RPN operators ---------------------------------------------------------- Could you please run pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's just to see if the error goes away. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 15:05:08 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 15:05:08 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A081E97.5000804@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> Message-ID: Hi Ihsan, Am 11.05.2009 um 14:48 schrieb Ihsan Dogan: >> Ihsan: Please test rrdtool against the current cairo and see if the >> error >> goes away. > > It looks already much better; at least it's not core dumping anymore. > > But I'm getting this error: > > ---------------------------------------------------------- > > ihsan at build8st:~$ /opt/csw/share/rrdtool/examples/stripes.pl > > (process:23307): Pango-CRITICAL **: No modules found: > No builtin or dynamically loaded modules were found. > PangoFc will not work correctly. > This probably means there was an error in the creation of: > '/opt/csw/etc/pango/pango.modules' > You should create this file by running: > pango-querymodules > '/opt/csw/etc/pango/pango.modules' > > (process:23307): Pango-WARNING **: failed to choose a font, expect > ugly > output. engine-type='PangoRenderFc', script='common' > > (process:23307): Pango-WARNING **: failed to choose a font, expect > ugly > output. engine-type='PangoRenderFc', script='latin' > This script has created stripes.png in the current directory > This demonstrates the use of the TIME and % RPN operators > ---------------------------------------------------------- > > Could you please run > pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's > just > to see if the error goes away. Done. stripes.pl now doesn't throw an error any more. Should I put this on postinstall or how is this handled? Best regards -- Dago From ihsan at opencsw.org Mon May 11 15:09:23 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 15:09:23 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> Message-ID: <4A082383.3010904@opencsw.org> Hi Dago, Am 11.5.2009 15:05 Uhr, Dagobert Michelsen schrieb: >> Could you please run >> pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's just >> to see if the error goes away. > > Done. stripes.pl now doesn't throw an error any more. Should I put > this on postinstall or how is this handled? I don't know pango at all, but I think it makes sense to place it in the postinstall script. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Mon May 11 15:17:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 15:17:37 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <4A082383.3010904@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> <4A082383.3010904@opencsw.org> Message-ID: <87897C5D-4B9B-47B1-A28C-9FC10E14C815@opencsw.org> Hi, Am 11.05.2009 um 15:09 schrieb Ihsan Dogan: > Hi Dago, > > Am 11.5.2009 15:05 Uhr, Dagobert Michelsen schrieb: > >>> Could you please run >>> pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? >>> It's just >>> to see if the error goes away. >> >> Done. stripes.pl now doesn't throw an error any more. Should I put >> this on postinstall or how is this handled? > > I don't know pango at all, but I think it makes sense to place it in > the > postinstall script. Any expert on this onboard? William? Ken? Best regards -- Dago From william at wbonnet.net Mon May 11 15:50:53 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 15:50:53 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <87897C5D-4B9B-47B1-A28C-9FC10E14C815@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> <20090510194529.GD96486@bolthole.com> <4A073CF6.6070201@wbonnet.net> <20090510212039.GA28401@bolthole.com> <15E86313-D20E-4563-8D9A-B91096D8F707@opencsw.org> <4A081E97.5000804@opencsw.org> <4A082383.3010904@opencsw.org> <87897C5D-4B9B-47B1-A28C-9FC10E14C815@opencsw.org> Message-ID: <4A082D3D.4070609@wbonnet.net> Hi >>>> Could you please run >>>> pango-querymodules > /opt/csw/etc/pango/pango.modules by hand? It's >>>> just >>>> to see if the error goes away. >>> >>> Done. stripes.pl now doesn't throw an error any more. Should I put >>> this on postinstall or how is this handled? >> >> I don't know pango at all, but I think it makes sense to place it in the >> postinstall script. > > Any expert on this onboard? William? Ken? I'm not really an expert... but i never commited the postinstall script i made when i updated pango before going off. It is done now ... but package has to be rebuild. Dago please ? or i'll do it tonight cheers W. From bwalton at opencsw.org Mon May 11 16:57:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 11 May 2009 10:57:47 -0400 Subject: [csw-maintainers] anyone 'owning' fastcgi currently? Message-ID: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> Hi All, I updated a box today and realized it was still running my local csw package of FastCGI. I hadn't released it after creating. I see there is a GAR description for it, but it hasn't been touched in a while and I don't see a released package for it. If nobody is working on it, I'll take it over and get a release out. If nobody pipes up in a few days, I'll take that as consent. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Mon May 11 17:07:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 11 May 2009 10:07:35 -0500 Subject: [csw-maintainers] libffi now in testing Message-ID: <4A083F37.8090404@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 New package this should allow python's ctypes module to be built ;) and may help some other packages be built with studio rather then gcc - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoIPzYACgkQLrhmsXMSLxddAACg6UGBOipCCZ0EsDVqW4gKJsTu 50UAn047vNF4pAfVKxizGeUkgQpFlpBe =vnIX -----END PGP SIGNATURE----- From ihsan at opencsw.org Mon May 11 17:10:26 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 17:10:26 +0200 Subject: [csw-maintainers] anyone 'owning' fastcgi currently? In-Reply-To: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> References: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> Message-ID: <4A083FE2.40107@opencsw.org> Hello Ben, Am 11.5.2009 16:57 Uhr, Ben Walton schrieb: > I updated a box today and realized it was still running my local csw > package of FastCGI. I hadn't released it after creating. I see there > is a GAR description for it, but it hasn't been touched in a while and > I don't see a released package for it. Do you mean the ap2_modfastcgi package? I've took it over from Cory and updated it last year. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bwalton at opencsw.org Mon May 11 17:13:23 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 11 May 2009 11:13:23 -0400 Subject: [csw-maintainers] anyone 'owning' fastcgi currently? In-Reply-To: <4A083FE2.40107@opencsw.org> References: <1242053782-sup-8861@ntdws12.chass.utoronto.ca> <4A083FE2.40107@opencsw.org> Message-ID: <1242054719-sup-5308@ntdws12.chass.utoronto.ca> Excerpts from Ihsan Dogan's message of Mon May 11 11:10:26 -0400 2009: Hi Ihsan, > Do you mean the ap2_modfastcgi package? I've took it over from Cory and > updated it last year. No, I'm referring to the library that you build client support against. I built (for example) the ruby gem fcgi using the libfastcgi bits. This allows ruby to speak fcgi with apache through the ap2_modfastcgi module. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Mon May 11 19:05:06 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 19:05:06 +0200 Subject: [csw-maintainers] nginx pacakge References: <8E9AF002-6C8F-4F7C-8555-781C71DFCE9B@opencsw.org> Message-ID: <5841B334-5A6B-48C5-8ED2-7750D0BDF47D@opencsw.org> Hi Andy, Am 11.05.2009 um 18:51 schrieb Andy Igoshin: > after reading message below i was confused. > do i use cswclassutils in my package? > if yes do i use it in a right way? It looks ok to me. The classutils are documented here: Phil, what did you have in mind posting this? Best regards -- Dago > ---------- Forwarded Message ---------- > > Subject: Re: nginx pacakge > Date: Mon of May 11 2009 > From: Andy Igoshin > To: Philip Brown > > On Mon of May 11 2009 19:38:38 you wrote: >> On Mon, May 11, 2009 at 06:41:43PM +0400, Andy Igoshin wrote: >>>>> - nginx package supports both init.d and SMF. >>>> >>>> thats why we have cswclassutils. So that one package, can >>>> transparently >>>> support both, depending on what kind of system it is installed to. >>> >>> it is already done. >> >> what do you mean by that? you already use it, or "the package is >> already >> made"? >> The second, is no excuse to not redo it a better way :-) >> >> I see that you did not use cswclassutils. >> >> Please redo (both) packages, to use cswclassutils smf support, >> instead of >> the way you do it now? >> >> It is beneficial both because it makes your package layout simpler, >> AND >> also because we can add extra feature to the common cswclassutils >> package >> later on, without you having to redo the package every time. >> >>> solaris 8/9 does not have 'event port' at all. >>> >>> nginx built on solaris 10 can use select, poll, devpoll, event port. >>> >>> nginx built on solaris 8/9 can use only select, poll, devpoll. >> >> very well... i hate doing it, but i will accept separate packages >> for sol8 >> and sol10. >> >> please redo to use cswclassutils, though. > > CSWnginx.gspec: > ----------------------------------------------------------------- > %var bitname nginx > %var pkgname CSWnginx > %include url file://%{PKGLIB}/csw_dyndepend.gspec > %include url file://%{PKGLIB}/smf_enabled.gspec > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > %var SERVICE_CONF_FILES /opt/csw/etc/nginx/nginx.conf > /opt/csw/etc/nginx/fastcgi_params /opt/csw/etc/nginx/mime.types > %var RC_INIT_SCRIPT cswnginx > %var INIT_KILL_PRIO 16 > %var INIT_START_PRIO 90 > %var CAN_BE_AUTOENABLED no > %copyright url file://%{WORKSRC_FIRSTMOD}/LICENSE > ----------------------------------------------------------------- > > i thought using of such file CSWnginx.gspec in such form > means using of cswclassutils. if it is not correct could you > please give me an example of using cswclassutils in some package? From phil at bolthole.com Mon May 11 19:40:16 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 10:40:16 -0700 Subject: [csw-maintainers] nginx pacakge In-Reply-To: <5841B334-5A6B-48C5-8ED2-7750D0BDF47D@opencsw.org> References: <8E9AF002-6C8F-4F7C-8555-781C71DFCE9B@opencsw.org> <5841B334-5A6B-48C5-8ED2-7750D0BDF47D@opencsw.org> Message-ID: <20090511174015.GC96177@bolthole.com> On Mon, May 11, 2009 at 07:05:06PM +0200, Dagobert Michelsen wrote: > Hi Andy, > > Am 11.05.2009 um 18:51 schrieb Andy Igoshin: >> after reading message below i was confused. >> do i use cswclassutils in my package? >> if yes do i use it in a right way? > > It looks ok to me. The classutils are documented here: > > > Phil, what did you have in mind [with your email to Andy?] I looked at one of his packages. It was using an internal-to-the-package smf class, rather than cswclassutils. Perhaps you looked at a different one of his, that was using cswclassutils. My point is that all of his packages should be using it. From pfelecan at opencsw.org Mon May 11 19:47:58 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Mon, 11 May 2009 19:47:58 +0200 Subject: [csw-maintainers] libffi now in testing In-Reply-To: <4A083F37.8090404@opencsw.org> (Mike Watters's message of "Mon\, 11 May 2009 10\:07\:35 -0500") References: <4A083F37.8090404@opencsw.org> Message-ID: Mike Watters writes: > New package > > this should allow python's ctypes module to be built ;) > and may help some other packages be built with studio rather then gcc And we still have the capacity to build the same set of packages with gcc? -- Peter From mwatters at opencsw.org Mon May 11 19:52:00 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 11 May 2009 12:52:00 -0500 Subject: [csw-maintainers] libffi now in testing In-Reply-To: References: <4A083F37.8090404@opencsw.org> Message-ID: <4A0865C0.10805@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter FELECAN wrote: > Mike Watters writes: > >> New package >> >> this should allow python's ctypes module to be built ;) >> and may help some other packages be built with studio rather then gcc > > And we still have the capacity to build the same set of packages with gcc? yep, this should allow the packages that only require the ffi libraries from gcc to be built using studio. specifically I need it for the ctypes module in python. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoIZcAACgkQLrhmsXMSLxfK4ACfQpeKcvfkBPRLhuLu50ErhT3c ZXoAn3PFv8h1TKMvmnDyiEfvWHaJQ1QH =axHm -----END PGP SIGNATURE----- From bonivart at opencsw.org Mon May 11 22:02:19 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 22:02:19 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A05A1A0.3090702@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> Message-ID: <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> On Sat, May 9, 2009 at 5:30 PM, Ihsan Dogan wrote: > Hello, > > Back in December we have been talking about a summer camp in Norway. I > think it's now time to talk about the topics for the summer camp. I > would like to open the discussion now. Topic suggestions from me: - Stable collection - Release process - Tiered packages - Download stats -- /peter From william at wbonnet.net Mon May 11 22:07:57 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 22:07:57 +0200 Subject: [csw-maintainers] Mantis login problem In-Reply-To: <4A07412C.7090804@wbonnet.net> References: <4A073D5F.4030302@wbonnet.net> <4A073DE3.3080702@opencsw.org> <4A07412C.7090804@wbonnet.net> Message-ID: <4A08859D.4080405@wbonnet.net> Hi Sebastian It seems to work. Sorry for the noise ;) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From william at wbonnet.net Mon May 11 22:13:05 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 22:13:05 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A071355.3020204@opencsw.org> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> Message-ID: <4A0886D1.9080800@wbonnet.net> Hi > For the internal use, it might be handy to have a Instant Messenger > contact list. > What about having our own jabber server ? cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Mon May 11 22:19:38 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 22:19:38 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0886D1.9080800@wbonnet.net> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> Message-ID: Hi William, Am 11.05.2009 um 22:13 schrieb William Bonnet: >> For the internal use, it might be handy to have a Instant Messenger >> contact list. >> > > What about having our own jabber server ? Nice :-) However, I would prefer if someone (Phil?) would insert these two lines I posted onto the active webpage now. Then bring the new website online, then have our own jabber server :-) Best regards -- Dago From william at wbonnet.net Mon May 11 22:26:35 2009 From: william at wbonnet.net (William Bonnet) Date: Mon, 11 May 2009 22:26:35 +0200 Subject: [csw-maintainers] X11, Cairo, Pango and GTK+ In-Reply-To: <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> References: <49A5CFAA.2010806@wbonnet.net> <75BCFCDD-AE97-4F64-8982-40FBFD1093C4@opencsw.org> Message-ID: <4A0889FB.9030205@wbonnet.net> Hi > I have now rebuild all the X11 packages from William, pixman, > cairo, pango, libxft2, libatk and gtk+ in GAR with 32/64 bit. Thank you very for having updated these packages Dago. cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From phil at bolthole.com Mon May 11 22:29:55 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 13:29:55 -0700 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> Message-ID: <20090511202955.GE96177@bolthole.com> On Mon, May 11, 2009 at 10:19:38PM +0200, Dagobert Michelsen wrote: > > Nice :-) However, I would prefer if someone (Phil?) would insert > these two lines I posted onto the active webpage now. Then > bring the new website online, then have our own jabber server :-) please remind me of the specific lines and also which page you think it would best belong on. From dam at opencsw.org Mon May 11 22:36:41 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 11 May 2009 22:36:41 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <20090511202955.GE96177@bolthole.com> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> Message-ID: Hi Phil, Am 11.05.2009 um 22:29 schrieb Philip Brown: > On Mon, May 11, 2009 at 10:19:38PM +0200, Dagobert Michelsen wrote: >> >> Nice :-) However, I would prefer if someone (Phil?) would insert >> these two lines I posted onto the active webpage now. Then >> bring the new website online, then have our own jabber server :-) > > please remind me of the specific lines and also which page you think > it > would best belong on. Here: Am 10.05.2009 um 18:54 schrieb Trygve Laugst?l: > Dagobert Michelsen wrote: >> Hi, >> >> I just found out that there is a Java Chatclient at Freenode. How >> about >> putting >> >> on the webpage to let interested people join immediately? > > I like the idea, but think mibbit[1] is a better client as it don't > require anything other than a browser. The applet integration on > Solaris > has been a bit shady in my experience. > > [1]: http://www.mibbit.com/ Maybe right on front, just above the packages updated box. Trgve or Peter may help in integrating mibbit. Best regards -- Dago From phil at bolthole.com Mon May 11 22:46:29 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 13:46:29 -0700 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> Message-ID: <20090511204629.GH96177@bolthole.com> On Mon, May 11, 2009 at 10:36:41PM +0200, Dagobert Michelsen wrote: > Hi Phil, *wave* > Am 11.05.2009 um 22:29 schrieb Philip Brown: >> please remind me of the specific lines and also which page you think >> it would best belong on. > Here: You did not give specific html code to include. You also did not give ONE choice, but a partial discussion on TWO alternatives. I can't deploy a quantum cat to our web pages; please give me something to put in that does not involve Heisenburg ;-) > Am 10.05.2009 um 18:54 schrieb Trygve Laugst?l: > >> Dagobert Michelsen wrote: >>> Hi, >>> >>> I just found out that there is a Java Chatclient at Freenode. How >>> about >>> putting >>> >>> on the webpage to let interested people join immediately? >> >> I like the idea, but think mibbit[1] is a better client as it don't >> require anything other than a browser. The applet integration on >> Solaris >> has been a bit shady in my experience. >> >> [1]: http://www.mibbit.com/ > > Maybe right on front, just above the packages updated box. > Trgve or Peter may help in integrating mibbit. From ihsan at opencsw.org Mon May 11 23:03:33 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 23:03:33 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> Message-ID: <4A0892A5.80100@opencsw.org> Am 11.5.2009 22:02 Uhr, Peter Bonivart schrieb: >> Back in December we have been talking about a summer camp in Norway. I >> think it's now time to talk about the topics for the summer camp. I >> would like to open the discussion now. > > Topic suggestions from me: > > - Stable collection > - Release process > - Tiered packages > - Download stats I've created quickly wiki page --> http://wiki.opencsw.org/summercamp-2009 I think it's the best, if we use this place to organize the summercamp as we did with the IRL meeting in Switzerland. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Mon May 11 23:08:23 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 11 May 2009 23:08:23 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <4A0886D1.9080800@wbonnet.net> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> Message-ID: <4A0893C7.4070909@opencsw.org> Am 11.5.2009 22:13 Uhr, William Bonnet schrieb: >> For the internal use, it might be handy to have a Instant Messenger >> contact list. >> > What about having our own jabber server ? A few people have sent me already their contacts and all of them had already a Jabber address, but an own Jabber server is something what we could think of. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From bonivart at opencsw.org Mon May 11 23:29:22 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 23:29:22 +0200 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <20090511204629.GH96177@bolthole.com> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> <20090511204629.GH96177@bolthole.com> Message-ID: <625385e30905111429s565ebf1eh1f0611f8b54c8ae3@mail.gmail.com> On Mon, May 11, 2009 at 10:46 PM, Philip Brown wrote: > You did not give specific html code to include. You also did not > give ONE choice, but a partial discussion on TWO alternatives. > I can't deploy a quantum cat to our web pages; please give me something to > put in that does not involve Heisenburg ;-) Mibbit was suggested since it's more lightweight (Ajax vs Java). This is all the HTML that is needed: IRC -- /peter From bonivart at opencsw.org Mon May 11 23:36:12 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 11 May 2009 23:36:12 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A0892A5.80100@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <4A0892A5.80100@opencsw.org> Message-ID: <625385e30905111436y89a5f15pa7d579bd8e3fb863@mail.gmail.com> On Mon, May 11, 2009 at 11:03 PM, Ihsan Dogan wrote: > I've created quickly wiki page --> http://wiki.opencsw.org/summercamp-2009 Great! > I think it's the best, if we use this place to organize the summercamp > as we did with the IRL meeting in Switzerland. When looking at the page for the first meeting it strikes me that we should make somewhat more official notes from the summer meeting. I have no idea what topics were discussed at the first meeting, someone tried to recap the meeting on the mailing list but maybe some notes should be passed around a few of the attendees for review and then be made public. -- /peter From phil at bolthole.com Mon May 11 23:42:24 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 14:42:24 -0700 Subject: [csw-maintainers] Add Chat-Applet to Website In-Reply-To: <625385e30905111429s565ebf1eh1f0611f8b54c8ae3@mail.gmail.com> References: <9FEC2BD9-559E-4044-B91B-1599E37D320D@opencsw.org> <4A071355.3020204@opencsw.org> <4A0886D1.9080800@wbonnet.net> <20090511202955.GE96177@bolthole.com> <20090511204629.GH96177@bolthole.com> <625385e30905111429s565ebf1eh1f0611f8b54c8ae3@mail.gmail.com> Message-ID: <20090511214224.GI96177@bolthole.com> On Mon, May 11, 2009 at 11:29:22PM +0200, Peter Bonivart wrote: > Mibbit was suggested since it's more lightweight (Ajax vs Java). This > is all the HTML that is needed: > > target="_blank">IRC added From mwatters at opencsw.org Mon May 11 23:59:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 11 May 2009 16:59:26 -0500 Subject: [csw-maintainers] libtool: link: cannot find the library `' Message-ID: <4A089FBE.1020105@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I ran across this error when compiling php4. I did some googling and found this bit of very handy information. "... In this case one can make libtool be more verbose by editing the appropriate line to $echo "$modename: cannot find the library \`$lib' (lib${name}.la: $deplib)" 1>&2 Resulting error message becomes something like: libtool: link: cannot find the library `' (whatever-library-is-missed: and-where-it-is-missed) ..." http://naranja.umh.es/~atg/B1098370398/C2036966428/E737470728/index.html - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoIn74ACgkQLrhmsXMSLxdjogCcDLJst6gtcoVuKeRgEZa6ehLV cDoAn2/Rb0Wj8YCMH6QuEW0oIXCt16QF =RKsJ -----END PGP SIGNATURE----- From phil at bolthole.com Tue May 12 00:03:25 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 11 May 2009 15:03:25 -0700 Subject: [csw-maintainers] libtool: link: cannot find the library `' In-Reply-To: <4A089FBE.1020105@opencsw.org> References: <4A089FBE.1020105@opencsw.org> Message-ID: <20090511220325.GJ96177@bolthole.com> On Mon, May 11, 2009 at 04:59:26PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I ran across this error when compiling php4. I did some googling and found > this bit of very handy information. sounds lovely.. why dont you just go ahead and file a 'bug' against our libtool package as a feature request? From pfelecan at opencsw.org Tue May 12 10:19:09 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 12 May 2009 10:19:09 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> (Peter Bonivart's message of "Mon\, 11 May 2009 22\:02\:19 +0200") References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> Message-ID: Peter Bonivart writes: > Topic suggestions from me: > > - Stable collection > - Release process > - Tiered packages Can you explain what you mean by "tiered packages"? > - Download stats -- Peter From bonivart at opencsw.org Tue May 12 10:49:01 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 12 May 2009 10:49:01 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> Message-ID: <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> On Tue, May 12, 2009 at 10:19 AM, Peter FELECAN wrote: > Can you explain what you mean by "tiered packages"? Sure, this is from the Suggestions page on the wiki: # Packages in tiers. [bonivart] We should classify each package in a tier according to how common/critical/important it is. Tier 1 should be the packages that most will install and are likely to be critical on those systems, examples can be Apache2, Bind, MySQL, OpenSSL and Sendmail. Tier 2 should be packages similar to tier 1 but maybe not as common and/or critical, one example might be Perl. Tier 3 is the rest. ? ?* I envision tier 1 to be around 25 packages, tier 2 around 100 and tier 3 the rest. ? ?* Tier 1 packages must always be actively maintained, possibly by more than one maintainer for quick responses to security updates and bug reports. All tier 1 packages must be in GAR. ? ?* Tier 2 packages should be actively maintained by maintainers who use their own packages on a daily basis. Packages should be in GAR. ? ?* Tier 3 has no special requirements. -- /peter From phil at bolthole.com Tue May 12 18:53:35 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 12 May 2009 09:53:35 -0700 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> Message-ID: <20090512165334.GA84438@bolthole.com> On Tue, May 12, 2009 at 10:49:01AM +0200, Peter Bonivart wrote: >... > ? ?* Tier 1 packages must always be actively maintained, possibly by > more than one maintainer for quick responses to security updates and > bug reports. All tier 1 packages must be in GAR. What you are suggesting, would be described in US political terms, as an "unfunded mandate". (saying "do this", but providing no resources to actually do it) Even if we declared this somehow to be in effect, that does not magically produce maintainers who are willing to step up and do the job. Things like this best happen from the bottom up. If people want to volunteer to help out [current maintainer], then they are free to do so now. > ? ?* Tier 2 packages should be actively maintained by maintainers who > use their own packages on a daily basis. Packages should be in GAR. Again, an "unfunded mandate". It's all very well to declare, "well, package x is really important: it MUST be maintained by someone who uses it on a daily basis!!" But now, try to magically produce someone who both fits these requirements, AND is willing to put in the effort to maintain it! Your suggestions in this area are remarkably, and ironically, dictatorial :-) If package x, has maintainer X, who rarely uses it, and maintainer Y DOES use it every day... maintainer Y is perfectly allowed to step forward and volunteer to take over the package. This happens already, and it has never been turned down, either by me, or by the "maintainer X' in question. There's no need to make unneccessary extra "rules" about this stuff. Particularly rules that just dont fit into a volunteer organization such as ours. From bonivart at opencsw.org Tue May 12 19:13:53 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 12 May 2009 19:13:53 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <20090512165334.GA84438@bolthole.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> <20090512165334.GA84438@bolthole.com> Message-ID: <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> On Tue, May 12, 2009 at 6:53 PM, Philip Brown wrote: > On Tue, May 12, 2009 at 10:49:01AM +0200, Peter Bonivart wrote: >>... >> ? ?* Tier 1 packages must always be actively maintained, possibly by >> more than one maintainer for quick responses to security updates and >> bug reports. All tier 1 packages must be in GAR. > > What you are suggesting, would be described in US political terms, as > an "unfunded mandate". (saying "do this", but providing no resources > ?to actually do it) Well, to identify which packages should go into which tier we could use another suggested topic of mine - download stats. > Things like this best happen from the bottom up. > If people want to volunteer to help out [current maintainer], then > they are free to do so now. Yes, but that usually only happens when we already have a problem and the current maintainer doesn't respond, that's what I want to avoid. To me, it would be good to identify the most critical packages and allocate resources in the form of active maintainers to them but if the community doesn't agree that's fine with me. Topics were asked for and I provided a few. -- /peter From william at wbonnet.net Tue May 12 21:00:18 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 12 May 2009 21:00:18 +0200 Subject: [csw-maintainers] Firefox 2.0.0.20 is in testing Message-ID: <4A09C742.6050704@wbonnet.net> Hi Firefox 2.0.0.20 is available from testing. You will need to update libcairo, libpango, pixman from testing, and an up to date glib2 from current (2.20) Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bwalton at opencsw.org Wed May 13 04:11:22 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 12 May 2009 22:11:22 -0400 Subject: [csw-maintainers] git 1.6.3 in testing Message-ID: <1242180411-sup-4562@ntdws12.chass.utoronto.ca> Hi All, I've placed packages for the newly released git 1.6.3 in testing. Feedback welcome. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 13 05:06:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 12 May 2009 23:06:59 -0400 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <4A070772.2010206@opencsw.org> References: <4A05A1A0.3090702@opencsw.org> <4A070772.2010206@opencsw.org> Message-ID: <1242183959-sup-1057@ntdws12.chass.utoronto.ca> Excerpts from Trygve Laugst?l's message of Sun May 10 12:57:22 -0400 2009: > I'm interested, but I doubt Norway is the best place to go. It will > require many people to travel far and stuff is generally considered > expensive in Norway. I can get access to conference rooms and work > facilities with connectivity for free though. I'd love to pop over to Norway, but it won't work for me this summer. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Wed May 13 12:10:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 12:10:25 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 Message-ID: Hi, I am currently updating glib (version 1.x) to 1.3.15 and get the attached errors on compilation. I don't really understand how this SGML/XML stuff works, so maybe someone who as more expertise may have a look? The package has been committed to mGAR as pkg/glib. Best regards -- Dago ... /opt/csw/bin/openjade:I: maximum number of errors (200) reached; change with -E option /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:73:W: cannot generate system identifier for public text "-//James Clark//DTD DSSSL Style Sheet//EN" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:2:91:W: cannot generate system identifier for public text "-//Norman Walsh// DOCUMENT DocBook HTML Stylesheet//EN" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:3:0:E: reference to entity "STYLE-SHEET" for which no system identifier could be generated /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:0: entity was defined here /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:3:0:E: DTD did not contain element declaration for document type name /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:2:0:E: notation "DSSSL" for entity "dbstyle" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:5:12:E: element "STYLE-SHEET" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:6:25:E: there is no attribute "USE" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:6:34:E: element "STYLE-SPECIFICATION" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:7:25:E: element "STYLE-SPECIFICATION-BODY" undefined /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl: 463:27:E: there is no attribute "ID" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl: 463:46:E: there is no attribute "DOCUMENT" /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl: 463:55:E: element "EXTERNAL-SPECIFICATION" undefined /opt/csw/bin/openjade:I: (invalid message) GLib Reference Manual GLib Overview GLib is a general-purpose utility library, which provides many useful data types, macros, type conversions, string utilities, file utilities, a main loop abstraction, and so on. It works on many UNIX-like platforms, Windows, OS/2 and BeOS. GLib is released under the GNU Library General Public License (GNU LGPL). GLib depends on the following: iconv() In order to implement conversions between character sets, GLib requires an implementation of the standard iconv() routine. Most modern systems will have a suitable implementation, however many older systems lack an iconv() implementation. On such systems, you must install the libiconv library. a thread implementation The thread support in GLib can be based upon several native thread implementations, e.g. POSIX threads, DCE threads or Solaris threads. Compiling the GLib package 3 GLib Library Compiling the GLib Package How to compile GLib itself Building the Library on UNIX On UNIX, GLib uses the standard GNU build system, using autoconf for package configuration and resolving portability issues, automake for building makefiles that comply with the GNU Coding Standards, and libtool for building shared libraries on multiple platforms. The normal sequence for compiling and installing the GLib library is thus: ./configure make make install The standard options provided by GNU autoconf may be passed to the configure script. Please see the autoconf documentation or run ./configure --help for information about the standard options. Extra Configuration Options In addition to the normal options, the configure script in the GTK+ library supports these additional arguments: configure --enable-debug=[no|minimum|yes] --disable-gc-friendly --enable-gc-friendly --disable-mem-pools --enable-mem-pools --disable-threads --enable-threads --with-threads=[none|posix|dce|solaris|win32] --disable-gtk-doc --enable-gtk-doc --enable-debug Turns on various amounts of debugging support. Setting this to 'no' disables g_assert(), g_return_if_fail(), g_return_val_if_fail() and all cast checks between different object types. Setting it to 'minimum' disables only cast checks. Setting it to 'yes' enables runtime debugging. The default is 'minimum'. Note that 'no' is fast, but dangerous as it tends to destabilize even mostly bug-free software by changing the effect of many bugs from simple warnings into fatal crashes. Thus --enable-debug=no should not be used for stable releases of gtk+. --disable-gc-friendly and --enable-gc-friendly When enabled all memory freed by the application, but retained by GLib for performance reasons is set to zero, thus making deployed garbage collection or memory profiling tools detect unlinked memory correctly. This will make GLib slightly slower and is thus disabled by default. --disable-mem-pools and --enable-mem-pools Many small chunks of memory are often allocated via collective pools in GLib and are cached after release to speed up reallocations. For sparse memory systems this behaviour is often inferior, so memory pools can be disabled to avoid excessive caching and force atomic maintenance of chunks through the g_malloc() and g_free() functions. Code currently affected by this: GList, GSList, GNode allocations GMemChunks become basically non-effective GSignal disables all caching (potentially very slow) GType doesn't honour the GTypeInfo n_preallocs field anymore the GBSearchArray flag G_BSEARCH_ALIGN_POWER2 becomes non-functional --disable-threads and --enable-threads Do not compile GLib to be multi thread safe. GLib will be slightly faster then. This is however not recommended, as many programs rely on GLib being multi thread safe. --with-threads Specify a thread implementation to use. 'posix' and 'dce' can be used interchangeable to mean the different versions of posix threads. configure tries to find out, which one is installed. 'solaris' uses the native Solaris thread implementation. 'none' means that GLib will be thread safe, but does not have a default thread implementation. This has to be supplied to g_thread_init() by the programmer. --disable-gtk-doc and --enable-gtk-doc By default the configure script will try to auto-detect whether the gtk-doc package is installed. If it is, then it will use it to extract and build the documentation for the GLib library. These options can be used to explicitly control whether gtk-doc should be used or not. If it is not used, the distributed, pre-generated HTML files will be installed instead of building them on your machine. Compiling GLib Applications 3 GLib Library Compiling GLib Applications How to compile your GLib application Compiling GLib Applications on UNIX To compile a GLib application, you need to tell the compiler where to find the GLib header files and libraries. This is done with the pkg-config utility. The following interactive shell session demonstrates how pkg-config is used: $ pkg-config --cflags glib-2.0 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include $ pkg-config --libs glib-2.0 -L/usr/lib -lm -lglib-1.3 If your application uses modules, threads or GObject features, it must be compiled and linked with the options returned by the following pkg-config invokations: $ pkg-config --cflags --libs gmodule-2.0 $ pkg-config --cflags --libs gthread-2.0 $ pkg-config --cflags --libs gobject-2.0 The simplest way to compile a program is to use the "backticks" feature of the shell. If you enclose a command in backticks (not single quotes), then its output will be substituted into the command line before execution. So to compile a GLib Hello, World, you would type the following: $ cc `pkg-config --cflags --libs glib-2.0` hello.c -o hello Running GLib Applications 3 GLib Library Running GLib Applications How to run and debug your GLib application Running and debugging GLib Applications Environment variables GLib inspects a few of environment variables in addition to standard variables like LANG, PATH or HOME. G_BROKEN_FILENAMES If this environment variable is set, GLib assumes that filenames are in the locale encoding rather than in UTF-8. G_MESSAGES_PREFIXED A list of log levels for which messages should be prefixed by the program name and PID of the application. The default is to prefix everything except G_LOG_LEVEL_MESSAGE and G_LOG_LEVEL_INFO. G_DEBUG If GLib has been configured with --enable-debug=yes, this variable can be set to a list of debug options, which cause GLib to print out different types of debugging information. fatal_warnings Causes GLib to abort the program at the first call to g_warning(). This option is special in that it doesn't require GLib to be configured with debugging support. Traps and traces gmake[6]: *** [html-build.stamp] Error 1 gmake[6]: Leaving directory `/home/dam/mgar/pkg/glib/trunk/work/build- isa-sparcv8/glib-1.3.15/docs/reference/glib' gmake[5]: *** [all-recursive] Error 1 From schwindt at dfki.uni-kl.de Wed May 13 12:42:30 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Wed, 13 May 2009 12:42:30 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: Your message of "Wed, 13 May 2009 12:10:25 +0200." Message-ID: <200905131042.n4DAgURm020716@dfki.uni-kl.de> > Hi, > > I am currently updating glib (version 1.x) to 1.3.15 and > get the attached errors on compilation. I don't really > understand how this SGML/XML stuff works, so maybe someone > who as more expertise may have a look? > > The package has been committed to mGAR as pkg/glib. > > > Best regards > > -- Dago > > > > ... > /opt/csw/bin/openjade:I: maximum number of errors (200) reached; > change with -E option > /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:73:W: > cannot generate system identifier for public text "-//James Clark//DTD > DSSSL Style Sheet//EN" Long time since I did this, this is what I remember : * openjade docbook docbook-dsssl need to be installed * The variable SGML_CATALOG_FILES needs to be set I can't for sure remember if to /opt/csw/etc/sgml/catalog or may be even to /opt/csw/share/sgml/openjade/catalog * try make in the dir docs/reference/glib Nicolai From dam at opencsw.org Wed May 13 13:46:24 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 13:46:24 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <200905131042.n4DAgURm020716@dfki.uni-kl.de> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> Message-ID: <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> Hi Nicolai, Am 13.05.2009 um 12:42 schrieb Nicolai Schwindt: >> I am currently updating glib (version 1.x) to 1.3.15 and >> get the attached errors on compilation. I don't really >> understand how this SGML/XML stuff works, so maybe someone >> who as more expertise may have a look? >> >> The package has been committed to mGAR as pkg/glib. >> >> >> ... >> /opt/csw/bin/openjade:I: maximum number of errors (200) reached; >> change with -E option >> /opt/csw/bin/openjade:/opt/csw/share/gtk-doc/data/gtk-doc.dsl:1:73:W: >> cannot generate system identifier for public text "-//James Clark// >> DTD >> DSSSL Style Sheet//EN" > > > Long time since I did this, this is what I remember : > > * openjade docbook docbook-dsssl need to be installed I thought I had that done: > build8s% pkginfo | grep -i book > application CSWdocbookdsssl docbookdsssl - Norman Walsh's > modular stylesheets for DocBook > application CSWdocbookdtds docbookdtds - SGML and XML document > type definitions for DocBook. > application CSWdocbookxsl docbookxsl - Norman Walsh's XSL > stylesheets for DocBook XML > application CSWdocbookxsldoc docbookxsldoc - Documentation for > the Docbook XSL stylesheets > * The variable SGML_CATALOG_FILES needs to be set > I can't for sure remember if to /opt/csw/etc/sgml/catalog or > may be even to /opt/csw/share/sgml/openjade/catalog AFAIK this is done automatically on package install. > * try make in the dir docs/reference/glib Same errors: > build8s% cd work/build-isa-sparcv8/glib-1.3.15/docs/reference/glib/ > build8s% gmake > *** Building HTML *** > rm -rf ./html > mkdir ./html > cd ./html && gtkdoc-mkhtml glib ../glib-docs.sgml > /opt/csw/bin/openjade:../glib-docs.sgml:1:59:W: cannot generate > system identifier for public text "-//Davenport//DTD DocBook V3.0//EN" > /opt/csw/bin/openjade:../glib-docs.sgml:61:0:E: reference to entity > "BOOK" for which no system identifier could be generated > /opt/csw/bin/openjade:../glib-docs.sgml:1:0: entity was defined here > /opt/csw/bin/openjade:../glib-docs.sgml:61:0:E: DTD did not contain > element declaration for document type name > /opt/csw/bin/openjade:../glib-docs.sgml:62:9:E: there is no > attribute "ID" > /opt/csw/bin/openjade:../glib-docs.sgml:62:16:E: element "BOOK" > undefined > /opt/csw/bin/openjade:../glib-docs.sgml:63:11:E: element "BOOKINFO" > undefined > /opt/csw/bin/openjade:../glib-docs.sgml:64:10:E: element "TITLE" > undefined > /opt/csw/bin/openjade:../glib-docs.sgml:67:14:E: there is no > attribute "ID" Best regards -- Dago From bwalton at opencsw.org Wed May 13 14:58:49 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 08:58:49 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> Message-ID: <1242219479-sup-5302@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Wed May 13 07:46:24 -0400 2009: Hi Dago, I'll take a look at this. I'm no expert with it either, but it's my package so I'll see if it's a packaging/config/postinstall issue. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Wed May 13 15:12:28 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:12:28 +0200 Subject: [csw-maintainers] Compiler optimization Sun Studio vs. GCC In-Reply-To: References: <20090512113534.8C94F61AD@mail.opencsw.org> <0F37CBDB-EA93-46ED-9189-7BE648D689BD@opencsw.org> <6ED24C28-C9B4-4B5D-AB1A-E7A0A422ED20@opencsw.org> <2D6EF361-CF70-4A6C-B8BE-C118EA177390@opencsw.org> <2F4453A9-6414-46B3-9AE9-77425B7B0B1D@opencsw.org> <3BB667EF-A861-47C1-905C-ECCD02B4D809@opencsw.org> Message-ID: <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> Hi Peter, I'm adding James to the post as he has experience in compiler optimization and used it for some of his packages extensively. Am 13.05.2009 um 15:04 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >>>> You have to install it :-) Apart from that the code generation for >>>> Sparc of gcc is very bad, that's why I also installed gccfss if >>>> the project absolutely demands gcc, but this further compilcates >>>> compilation. >>> >>> Having to install it is not an issue IMHO. >>> >>> When you say "very bad" what do you mean? >> >> The optimized is inferior to the one of SOS. I remember reading a >> comparison of the two, but I can't find it at >> >> > > Didn't find it either. Well, my question was motivated to really > qualify > the differences, lets say on a scale of 1 to 100, how do we position > the > 2 compilers for SPARC generated code. Saying "very bad" is relative > but > to what and in what measure? > > Finally, is that so important for the kind of packages that we > provide? > For 1%, 10% or 66%... Maybe James or some other maintainer has real, first-hand performance data? Best regards -- Dago From schwindt at dfki.uni-kl.de Wed May 13 15:20:35 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Wed, 13 May 2009 15:20:35 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: Your message of "Wed, 13 May 2009 13:46:24 +0200." <0237EBA3-2EB4-4DDB-A93A-A43A97C34BBF@opencsw.org> Message-ID: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> [...] > > * The variable SGML_CATALOG_FILES needs to be set > > I can't for sure remember if to /opt/csw/etc/sgml/catalog or > > may be even to /opt/csw/share/sgml/openjade/catalog > > AFAIK this is done automatically on package install. I tried with your files - sorry - that way I could try the same setup. export SGML_CATALOG_FILES=/opt/csw/etc/sgml/catalog gtkdoc-mkhtml glib /home/dam/mgar/pkg/glib/trunk/work/build-isa-sparcv8/glib-1 .3.15/docs/reference/glib/glib-docs.sgml take a look at build8s : bash-2.03$ pwd /tmp/schwindt Nicolai From pfelecan at acm.org Wed May 13 15:21:29 2009 From: pfelecan at acm.org (Peter FELECAN) Date: Wed, 13 May 2009 15:21:29 +0200 Subject: [csw-maintainers] Compiler optimization Sun Studio vs. GCC In-Reply-To: <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> (Dagobert Michelsen's message of "Wed\, 13 May 2009 15\:12\:28 +0200") References: <20090512113534.8C94F61AD@mail.opencsw.org> <0F37CBDB-EA93-46ED-9189-7BE648D689BD@opencsw.org> <6ED24C28-C9B4-4B5D-AB1A-E7A0A422ED20@opencsw.org> <2D6EF361-CF70-4A6C-B8BE-C118EA177390@opencsw.org> <2F4453A9-6414-46B3-9AE9-77425B7B0B1D@opencsw.org> <3BB667EF-A861-47C1-905C-ECCD02B4D809@opencsw.org> <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > I'm adding James to the post as he has experience in > compiler optimization and used it for some of his packages > extensively. > > Am 13.05.2009 um 15:04 schrieb Peter FELECAN: > >> Dagobert Michelsen writes: >> >>>>> You have to install it :-) Apart from that the code generation for >>>>> Sparc of gcc is very bad, that's why I also installed gccfss if >>>>> the project absolutely demands gcc, but this further compilcates >>>>> compilation. >>>> >>>> Having to install it is not an issue IMHO. >>>> >>>> When you say "very bad" what do you mean? >>> >>> The optimized is inferior to the one of SOS. I remember reading a >>> comparison of the two, but I can't find it at >>> >>> >> >> Didn't find it either. Well, my question was motivated to really >> qualify >> the differences, lets say on a scale of 1 to 100, how do we position >> the >> 2 compilers for SPARC generated code. Saying "very bad" is relative >> but >> to what and in what measure? >> >> Finally, is that so important for the kind of packages that we >> provide? >> For 1%, 10% or 66%... > > Maybe James or some other maintainer has real, first-hand performance > data? Let James wisdom shine upon us! -- Peter FELECAN mailto:pfelecan at acm.org From dam at opencsw.org Wed May 13 15:33:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:33:25 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> Message-ID: <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> Hi, Am 13.05.2009 um 15:20 schrieb Nicolai Schwindt: > [...] >>> * The variable SGML_CATALOG_FILES needs to be set >>> I can't for sure remember if to /opt/csw/etc/sgml/catalog or >>> may be even to /opt/csw/share/sgml/openjade/catalog >> >> AFAIK this is done automatically on package install. > > I tried with your files - sorry - that way I could try the same setup. > > export SGML_CATALOG_FILES=/opt/csw/etc/sgml/catalog > gtkdoc-mkhtml glib /home/dam/mgar/pkg/glib/trunk/work/build-isa- > sparcv8/glib-1 > .3.15/docs/reference/glib/glib-docs.sgml > > take a look at build8s : > > bash-2.03$ pwd > /tmp/schwindt Looks good. Thanks! I guess it would make sence to add this to the GAR defaults? Best regards -- Dago From james at Wed May 13 15:33:41 2009 From: james at (James Lee) Date: Wed, 13 May 2009 13:33:41 GMT Subject: [csw-maintainers] Compiler optimization Sun Studio vs. GCC In-Reply-To: References: <20090512113534.8C94F61AD@mail.opencsw.org> <0F37CBDB-EA93-46ED-9189-7BE648D689BD@opencsw.org> <6ED24C28-C9B4-4B5D-AB1A-E7A0A422ED20@opencsw.org> <2D6EF361-CF70-4A6C-B8BE-C118EA177390@opencsw.org> <2F4453A9-6414-46B3-9AE9-77425B7B0B1D@opencsw.org> <3BB667EF-A861-47C1-905C-ECCD02B4D809@opencsw.org> <31A941E2-F259-46F3-8052-CCE3C168A40A@opencsw.org> Message-ID: <20090513.13334100.439557401@gyor.oxdrove.co.uk> On 13/05/09, 14:21:29, Peter FELECAN wrote regarding Re: Compiler optimization Sun Studio vs. GCC: > >> Didn't find it either. Well, my question was motivated to really > >> qualify > >> the differences, lets say on a scale of 1 to 100, how do we position > >> the > >> 2 compilers for SPARC generated code. Saying "very bad" is relative > >> but > >> to what and in what measure? > >> > >> Finally, is that so important for the kind of packages that we > >> provide? > >> For 1%, 10% or 66%... > > > > Maybe James or some other maintainer has real, first-hand performance > > data? > Let James wisdom shine upon us! What type of software, or specifically what, are we compiling? In many cases the "best" optimisation is "-xO3 -xspace" because fastest CPU alone isn't best. Unless a person is waiting or a machine is CPU saturated we don't need to worry. Think too about memory usage and if it has a very short run life load times (code size), be careful that repeated tests don't just test memory and disc cache performance. James. From bwalton at opencsw.org Wed May 13 15:40:13 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 09:40:13 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> Message-ID: <1242221865-sup-2499@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Wed May 13 09:33:25 -0400 2009: > Looks good. Thanks! I guess it would make sence to add this to > the GAR defaults? You could possibly leverage the content in /opt/csw/etc/sgml/sgml.conf so that any future changes are reflected? -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Wed May 13 15:45:45 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:45:45 +0200 Subject: [csw-maintainers] Updated glib References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> Message-ID: Hi, Anfang der weitergeleiteten E-Mail: > ====================================================================== > Date Submitted: 2009-05-12 15:32 CEST > Last Modified: 2009-05-12 19:41 CEST > ====================================================================== > Summary: libgdk_pixbuf.la references non- > existent .la files, > breaks graphviz builds > Description: > Building graphviz fails with: > > /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info > 6:0:0 > -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/ > graphviz > gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la > -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 > -lintl > grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory > > > > I believe this error is originating from: /opt/csw/lib/ > libgmodule-2.0.la > which contains a reference to this non-existent file. (Also to ls: > /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) > > > A more general question is: why are .la files being installed at all > by > opencsw? > I don't understand all the rationale, but I note that Fedora does not > install any .la files. > > > ====================================================================== > > ---------------------------------------------------------------------- > (0006105) John Ellson (reporter) - 2009-05-12 16:13 > http://opencsw.org/bugtrack/view.php?id=3666#c6105 > ---------------------------------------------------------------------- > Correction: > > I believe the problem originates in the installed: > /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent > files. > > ---------------------------------------------------------------------- > (0006106) dam (reporter) - 2009-05-12 16:34 > http://opencsw.org/bugtrack/view.php?id=3666#c6106 > ---------------------------------------------------------------------- > .la files were included in the past, but the current policy is to > exclude > them: > > > New packages should take care to EXCLUDE libtool .la files. They are > not > helpful, and often create more problems than they solve. > Unfortunately, > existing packages may need to preserve them, until all dependant > packages > have their own configs adjusted to not use .la files. > > There is also a script which fixes Makefile to use linker directives > instead of .la files. The usage in GAR is documented at > Please ask on > maintainers@ for further advice if needed. > > ---------------------------------------------------------------------- > (0006107) phil (manager) - 2009-05-12 18:58 > http://opencsw.org/bugtrack/view.php?id=3666#c6107 > ---------------------------------------------------------------------- > this is actually a bug in gmodule. > Please file a bug with THAT package, to repackage without .la files. > > ---------------------------------------------------------------------- > (0006108) John Ellson (reporter) - 2009-05-12 19:21 > http://opencsw.org/bugtrack/view.php?id=3666#c6108 > ---------------------------------------------------------------------- > This may *also* be a bug with gmodule, but *this* package is > installing > /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so > minimally it > needs to be repackaged without it. > > I'll file a bug against glib for /opt/csw/lib/libgmodule.la > > (I'm finding the .la files on the build8s host) > > ---------------------------------------------------------------------- > (0006109) phil (manager) - 2009-05-12 19:41 > http://opencsw.org/bugtrack/view.php?id=3666#c6109 > ---------------------------------------------------------------------- > since the offending package is mine, please feel free to repackage > and take > over glib. I have made an updated glib 1.2.10 package with the latest package structure (license/, no static libs, no .la) in testing/: glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz The latest 1.x.y library, 1.3.15, builds a different shared library, whereas all the dependent packages rely on /opt/csw/lib/libglib-1.2.so.0 I guess it wouldn't make sense to recompile them against a newer 1.x version? Best regards -- Dago From dam at opencsw.org Wed May 13 15:46:47 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:46:47 +0200 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242221865-sup-2499@ntdws12.chass.utoronto.ca> References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> <1242221865-sup-2499@ntdws12.chass.utoronto.ca> Message-ID: Hi Ben, Am 13.05.2009 um 15:40 schrieb Ben Walton: > Excerpts from Dagobert Michelsen's message of Wed May 13 09:33:25 > -0400 2009: >> Looks good. Thanks! I guess it would make sence to add this to >> the GAR defaults? > > You could possibly leverage the content in /opt/csw/etc/sgml/sgml.conf > so that any future changes are reflected? Is it safe to assume that when the file is there that exactly the variables SGML_BASE_DIR SGML_CATALOGS_DIR are set and need to be exported? Best regards -- Dago From dam at opencsw.org Wed May 13 15:47:17 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 15:47:17 +0200 Subject: [csw-maintainers] Updated glib References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> Message-ID: <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> Hi, Anfang der weitergeleiteten E-Mail: > ====================================================================== > Date Submitted: 2009-05-12 15:32 CEST > Last Modified: 2009-05-12 19:41 CEST > ====================================================================== > Summary: libgdk_pixbuf.la references non- > existent .la files, > breaks graphviz builds > Description: > Building graphviz fails with: > > /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info > 6:0:0 > -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/ > graphviz > gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la > -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 > -lintl > grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory > > > > I believe this error is originating from: /opt/csw/lib/ > libgmodule-2.0.la > which contains a reference to this non-existent file. (Also to ls: > /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) > > > A more general question is: why are .la files being installed at all > by > opencsw? > I don't understand all the rationale, but I note that Fedora does not > install any .la files. > > > ====================================================================== > > ---------------------------------------------------------------------- > (0006105) John Ellson (reporter) - 2009-05-12 16:13 > http://opencsw.org/bugtrack/view.php?id=3666#c6105 > ---------------------------------------------------------------------- > Correction: > > I believe the problem originates in the installed: > /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent > files. > > ---------------------------------------------------------------------- > (0006106) dam (reporter) - 2009-05-12 16:34 > http://opencsw.org/bugtrack/view.php?id=3666#c6106 > ---------------------------------------------------------------------- > .la files were included in the past, but the current policy is to > exclude > them: > > > New packages should take care to EXCLUDE libtool .la files. They are > not > helpful, and often create more problems than they solve. > Unfortunately, > existing packages may need to preserve them, until all dependant > packages > have their own configs adjusted to not use .la files. > > There is also a script which fixes Makefile to use linker directives > instead of .la files. The usage in GAR is documented at > Please ask on > maintainers@ for further advice if needed. > > ---------------------------------------------------------------------- > (0006107) phil (manager) - 2009-05-12 18:58 > http://opencsw.org/bugtrack/view.php?id=3666#c6107 > ---------------------------------------------------------------------- > this is actually a bug in gmodule. > Please file a bug with THAT package, to repackage without .la files. > > ---------------------------------------------------------------------- > (0006108) John Ellson (reporter) - 2009-05-12 19:21 > http://opencsw.org/bugtrack/view.php?id=3666#c6108 > ---------------------------------------------------------------------- > This may *also* be a bug with gmodule, but *this* package is > installing > /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so > minimally it > needs to be repackaged without it. > > I'll file a bug against glib for /opt/csw/lib/libgmodule.la > > (I'm finding the .la files on the build8s host) > > ---------------------------------------------------------------------- > (0006109) phil (manager) - 2009-05-12 19:41 > http://opencsw.org/bugtrack/view.php?id=3666#c6109 > ---------------------------------------------------------------------- > since the offending package is mine, please feel free to repackage > and take > over glib. I have made an updated glib 1.2.10 package with the latest package structure (license/, no static libs, no .la) in testing/: glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz The latest 1.x.y library, 1.3.15, builds a different shared library, whereas all the dependent packages rely on /opt/csw/lib/libglib-1.2.so.0 I guess it wouldn't make sense to recompile them against a newer 1.x version? Best regards -- Dago From bwalton at opencsw.org Wed May 13 15:54:25 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 09:54:25 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: References: <200905131320.n4DDKZiT023319@dfki.uni-kl.de> <18636D8B-8400-4307-AF1E-9F0CF6124846@opencsw.org> <1242221865-sup-2499@ntdws12.chass.utoronto.ca> Message-ID: <1242222692-sup-7411@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Wed May 13 09:46:47 -0400 2009: Hi Dago, > Is it safe to assume that when the file is there that exactly the > variables > SGML_BASE_DIR > SGML_CATALOGS_DIR > are set and need to be exported? Those variables set sane defaults for programs to look in for files (stylesheets of various types, catalogs, dtds, etc). So yes, if the file exists, exporting them should be fine. I would also use SGML_CATALOG_DIR to build SGML_CATALOG_FILES (which solved the glib issue for you). There should always be a base catalog when that file exists, but its existence doesn't mean that the required stylesheets or dtds are installed. Hopefully programs can use the catalog to determine this and present reasonable errors if they don't have what they need. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From ellson at research.att.com Wed May 13 16:02:20 2009 From: ellson at research.att.com (John Ellson) Date: Wed, 13 May 2009 10:02:20 -0400 Subject: [csw-maintainers] Updated glib In-Reply-To: <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> Message-ID: <4A0AD2EC.105@research.att.com> Dagobert Michelsen wrote: > Hi, > > Anfang der weitergeleiteten E-Mail: >> ====================================================================== >> Date Submitted: 2009-05-12 15:32 CEST >> Last Modified: 2009-05-12 19:41 CEST >> ====================================================================== >> Summary: libgdk_pixbuf.la references non-existent >> .la files, >> breaks graphviz builds >> Description: >> Building graphviz fails with: >> >> /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info >> 6:0:0 >> -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/graphviz >> gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la >> -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 >> -lintl >> grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory >> >> >> >> I believe this error is originating from: >> /opt/csw/lib/libgmodule-2.0.la >> which contains a reference to this non-existent file. (Also to ls: >> /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) >> >> >> A more general question is: why are .la files being installed at all by >> opencsw? >> I don't understand all the rationale, but I note that Fedora does not >> install any .la files. >> >> >> ====================================================================== >> >> ---------------------------------------------------------------------- >> (0006105) John Ellson (reporter) - 2009-05-12 16:13 >> http://opencsw.org/bugtrack/view.php?id=3666#c6105 >> ---------------------------------------------------------------------- >> Correction: >> >> I believe the problem originates in the installed: >> /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent >> files. >> >> ---------------------------------------------------------------------- >> (0006106) dam (reporter) - 2009-05-12 16:34 >> http://opencsw.org/bugtrack/view.php?id=3666#c6106 >> ---------------------------------------------------------------------- >> .la files were included in the past, but the current policy is to >> exclude >> them: >> >> >> New packages should take care to EXCLUDE libtool .la files. They are not >> helpful, and often create more problems than they solve. Unfortunately, >> existing packages may need to preserve them, until all dependant >> packages >> have their own configs adjusted to not use .la files. >> >> There is also a script which fixes Makefile to use linker directives >> instead of .la files. The usage in GAR is documented at >> Please ask on >> maintainers@ for further advice if needed. >> >> ---------------------------------------------------------------------- >> (0006107) phil (manager) - 2009-05-12 18:58 >> http://opencsw.org/bugtrack/view.php?id=3666#c6107 >> ---------------------------------------------------------------------- >> this is actually a bug in gmodule. >> Please file a bug with THAT package, to repackage without .la files. >> >> ---------------------------------------------------------------------- >> (0006108) John Ellson (reporter) - 2009-05-12 19:21 >> http://opencsw.org/bugtrack/view.php?id=3666#c6108 >> ---------------------------------------------------------------------- >> This may *also* be a bug with gmodule, but *this* package is installing >> /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so >> minimally it >> needs to be repackaged without it. >> >> I'll file a bug against glib for /opt/csw/lib/libgmodule.la >> >> (I'm finding the .la files on the build8s host) >> >> ---------------------------------------------------------------------- >> (0006109) phil (manager) - 2009-05-12 19:41 >> http://opencsw.org/bugtrack/view.php?id=3666#c6109 >> ---------------------------------------------------------------------- >> since the offending package is mine, please feel free to repackage >> and take >> over glib. > > I have made an updated glib 1.2.10 package with the latest package > structure (license/, no static libs, no .la) in testing/: > > glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz > glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz > > The latest 1.x.y library, 1.3.15, builds a different shared library, > whereas all the dependent packages rely on > /opt/csw/lib/libglib-1.2.so.0 > > I guess it wouldn't make sense to recompile them against a newer 1.x > version? Thanks. Graphviz does use glib2.0. I'm afraid that I've caused some confusion. I originally reported the libgdk_pixbuf.la problem against the gdk_pixbuf package, and Phil said, rightly as it turns out, that it "wasn't caused by that package, perhaps glib." Using: pkgchk -l -p /opt/csw/lib/libgdk_pixbuf-2.0.la I see that the real culprit is the CSWgtk2 package. Sorry for the confusion. I'll generate a Mantis report against gtk2 requesting removal of its .la files. John From dam at opencsw.org Wed May 13 16:44:34 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 16:44:34 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AD2EC.105@research.att.com> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> Message-ID: <945EC7B4-52F1-46D4-808C-435DF8CBD5BD@opencsw.org> Hi John, Am 13.05.2009 um 16:02 schrieb John Ellson: > Sorry for the confusion. I'll generate a Mantis report against gtk2 > requesting removal of its .la files. You can do so. Please note that there is already a current version of gtk2 is waiting on build8st for testing. Please try compiling graphviz there and help testing the package so it can be released soon. Best regards -- Dago From mwatters at opencsw.org Wed May 13 17:33:19 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 13 May 2009 10:33:19 -0500 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AD2EC.105@research.att.com> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> Message-ID: <4A0AE83F.3090502@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > Dagobert Michelsen wrote: >> Hi, >> >> Anfang der weitergeleiteten E-Mail: >>> ====================================================================== >>> Date Submitted: 2009-05-12 15:32 CEST >>> Last Modified: 2009-05-12 19:41 CEST >>> ====================================================================== >>> Summary: libgdk_pixbuf.la references non-existent >>> .la files, >>> breaks graphviz builds >>> Description: >>> Building graphviz fails with: >>> >>> /bin/bash ../../libtool --tag=CC --mode=link cc -g -version-info >>> 6:0:0 >>> -L/opt/csw/lib -o libgvplugin_gdk_pixbuf.la -rpath /opt/csw/lib/graphviz >>> gvplugin_gdk_pixbuf.lo gvdevice_gdk_pixbuf.lo ../../lib/gvc/libgvc.la >>> -L/opt/csw/lib -lgdk_pixbuf-2.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 >>> -lintl >>> grep: /opt/csw/lib/libgmodule-2.0.la: No such file or directory >>> >>> >>> >>> I believe this error is originating from: >>> /opt/csw/lib/libgmodule-2.0.la >>> which contains a reference to this non-existent file. (Also to ls: >>> /opt/csw/lib/libgobject-2.0.la and /opt/csw/lib/libglib-2.0.la) >>> >>> >>> A more general question is: why are .la files being installed at all by >>> opencsw? >>> I don't understand all the rationale, but I note that Fedora does not >>> install any .la files. >>> >>> >>> ====================================================================== >>> >>> ---------------------------------------------------------------------- >>> (0006105) John Ellson (reporter) - 2009-05-12 16:13 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6105 >>> ---------------------------------------------------------------------- >>> Correction: >>> >>> I believe the problem originates in the installed: >>> /opt/csw/lib/libgdk_pixbuf-2.0.la which references the non-existent >>> files. >>> >>> ---------------------------------------------------------------------- >>> (0006106) dam (reporter) - 2009-05-12 16:34 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6106 >>> ---------------------------------------------------------------------- >>> .la files were included in the past, but the current policy is to >>> exclude >>> them: >>> >>> >>> New packages should take care to EXCLUDE libtool .la files. They are not >>> helpful, and often create more problems than they solve. Unfortunately, >>> existing packages may need to preserve them, until all dependant >>> packages >>> have their own configs adjusted to not use .la files. >>> >>> There is also a script which fixes Makefile to use linker directives >>> instead of .la files. The usage in GAR is documented at >>> Please ask on >>> maintainers@ for further advice if needed. >>> >>> ---------------------------------------------------------------------- >>> (0006107) phil (manager) - 2009-05-12 18:58 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6107 >>> ---------------------------------------------------------------------- >>> this is actually a bug in gmodule. >>> Please file a bug with THAT package, to repackage without .la files. >>> >>> ---------------------------------------------------------------------- >>> (0006108) John Ellson (reporter) - 2009-05-12 19:21 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6108 >>> ---------------------------------------------------------------------- >>> This may *also* be a bug with gmodule, but *this* package is installing >>> /opt/csw/lib/libgdk_pixbuf-2.0.la with broken references, so >>> minimally it >>> needs to be repackaged without it. >>> >>> I'll file a bug against glib for /opt/csw/lib/libgmodule.la >>> >>> (I'm finding the .la files on the build8s host) >>> >>> ---------------------------------------------------------------------- >>> (0006109) phil (manager) - 2009-05-12 19:41 >>> http://opencsw.org/bugtrack/view.php?id=3666#c6109 >>> ---------------------------------------------------------------------- >>> since the offending package is mine, please feel free to repackage >>> and take >>> over glib. >> >> I have made an updated glib 1.2.10 package with the latest package >> structure (license/, no static libs, no .la) in testing/: >> >> glib-1.2.10,REV=2009.05.13-SunOS5.8-i386-CSW.pkg.gz >> glib-1.2.10,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz >> >> The latest 1.x.y library, 1.3.15, builds a different shared library, >> whereas all the dependent packages rely on >> /opt/csw/lib/libglib-1.2.so.0 >> >> I guess it wouldn't make sense to recompile them against a newer 1.x >> version? > > Thanks. Graphviz does use glib2.0. > > I'm afraid that I've caused some confusion. I originally reported the > libgdk_pixbuf.la problem > against the gdk_pixbuf package, and Phil said, rightly as it turns out, > that it "wasn't caused by that package, perhaps glib." > > Using: > pkgchk -l -p /opt/csw/lib/libgdk_pixbuf-2.0.la > I see that the real culprit is the CSWgtk2 package. > > Sorry for the confusion. I'll generate a Mantis report against gtk2 > requesting removal of its .la files. Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar Makefile. or if you are building by hand, after you run configure run gar/bin/fixlibtool {Build Directory} - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoK6D8ACgkQLrhmsXMSLxdytgCfXz65a5rUIR6NiE4334yt133i T7UAoMJVO5K3T5zsTxNPvF/unPrj72kj =k9t6 -----END PGP SIGNATURE----- From ellson at opencsw.org Wed May 13 17:49:50 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 11:49:50 -0400 Subject: [csw-maintainers] Updated glib In-Reply-To: <945EC7B4-52F1-46D4-808C-435DF8CBD5BD@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <945EC7B4-52F1-46D4-808C-435DF8CBD5BD@opencsw.org> Message-ID: <4A0AEC1E.4020606@opencsw.org> Dagobert Michelsen wrote: > .... there is already a current version of > gtk2 is waiting on build8st for testing. Please try compiling > graphviz there and help testing the package so it can be released soon. build8st is asking me for a passwd. Can you copy over my .ssh/ ? From pfelecan at opencsw.org Wed May 13 17:57:47 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 13 May 2009 17:57:47 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AE83F.3090502@opencsw.org> (Mike Watters's message of "Wed\, 13 May 2009 10\:33\:19 -0500") References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Mike Watters writes: > John Ellson wrote: >> Sorry for the confusion. I'll generate a Mantis report against gtk2 >> requesting removal of its .la files. > > Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar > Makefile. > > or if you are building by hand, after you run configure run gar/bin/fixlibtool > {Build Directory} This implies that you're on the public build stack and cannot be reproduced on a private build stack, i.e. my private servers. IMHO tools as fixlibtool should be in a package such as cswutils. -- Peter From ellson at opencsw.org Wed May 13 18:05:09 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 12:05:09 -0400 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AE83F.3090502@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: <4A0AEFB5.1000301@opencsw.org> Mike Watters wrote: > >> Using: >> pkgchk -l -p /opt/csw/lib/libgdk_pixbuf-2.0.la >> I see that the real culprit is the CSWgtk2 package. >> >> Sorry for the confusion. I'll generate a Mantis report against gtk2 >> requesting removal of its .la files. >> > > Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar > Makefile. > Ah, well, "Getting Started With GAR" is a whole other issue ;-) > or if you are building by hand, after you run configure run gar/bin/fixlibtool > {Build Directory} > Which, apparently still needs me to get started with gar :-( And anyway, both of these sound like hacks, instead of fixing the real problem which requires removing the broken .la files from the distribution. Removing them will also help regular OpenCSW users, not just the maintainers. John From phil at bolthole.com Wed May 13 18:13:48 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 09:13:48 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <200905131042.n4DAgURm020716@dfki.uni-kl.de> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> Message-ID: <20090513161348.GB41179@bolthole.com> On Wed, May 13, 2009 at 12:42:30PM +0200, Nicolai Schwindt wrote: > .... > * The variable SGML_CATALOG_FILES needs to be set > I can't for sure remember if to /opt/csw/etc/sgml/catalog or > may be even to /opt/csw/share/sgml/openjade/catalog I vaguely remember talking about this with Ben when he made that stuff; My opinion at the time was, "if the software needs the user to set an env var for it to work right, then the software installation is broken". I thought that Ben agreed, and had set it up to work without anything external set. Ben...? From phil at bolthole.com Wed May 13 18:16:39 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 09:16:39 -0700 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AE83F.3090502@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: <20090513161639.GC41179@bolthole.com> On Wed, May 13, 2009 at 10:33:19AM -0500, Mike Watters wrote: > > Sorry for the confusion. I'll generate a Mantis report against gtk2 > > requesting removal of its .la files. > > Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in your mGar > Makefile. no, the bug should still be filed either way, since it IS a bug in the gtk2/glib2 package :-) From phil at bolthole.com Wed May 13 18:59:07 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 09:59:07 -0700 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> <20090512165334.GA84438@bolthole.com> <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> Message-ID: <20090513165907.GH41179@bolthole.com> On Tue, May 12, 2009 at 07:13:53PM +0200, Peter Bonivart wrote: > Well, to identify which packages should go into which tier we could > use another suggested topic of mine - download stats. Those stats might be skewed in terms of which packages get frequently updated, not neccessarily in terms of importance. And then there is a relative definition of "importance". Does it mean "popular", or does it mean "critical to other programs"? Or some combination of both? As far as "importance" goes, I think a fairly basic and concrete measure would be, the number of dependants that a package has. eg: gtk2 has around 100. Thus, it is incontrovertibly an "important package" ;-) > Yes, but that usually only happens when we already have a problem and > the current maintainer doesn't respond, that's what I want to avoid. > > To me, it would be good to identify the most critical packages and > allocate resources in the form of active maintainers to them but if > the community doesn't agree that's fine with me. Topics were asked for > and I provided a few. I actually agree that it is important to *identify* critical packages. I just dont think we can go around making statements that they "must" be maintained in a particular manner. btw, if you'd like to spend some time putting together a dynamic page with a dependancy-ranked chart of packages, correlated with which ones are non-maintained, I'd be happy to email you more about how to access our various databases with that information. From bwalton at opencsw.org Wed May 13 19:01:59 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:01:59 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513161348.GB41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> Message-ID: <1242231941-sup-7634@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 12:13:48 -0400 2009: > I vaguely remember talking about this with Ben when he made that > stuff; This stuff specifically, or the issues you had when building mutt docs using it? > My opinion at the time was, "if the software needs the user to set an env > var for it to work right, then the software installation is broken". Well, things like libxml know where to look for the files and catalogs, and the `sgmlwhich` command provides the mechanism for client software to determine what is needed in a dynamic fashion. [cat `sgmlwhich`] I really don't recall discussing this before outside of the mutt issue which was due to a build-time option setup in opensp (now fixed). > I thought that Ben agreed, and had set it up to work without anything > external set. I only packaged up the stack the way it was done elsewhere...My uses of it, has been functional without environment settings. Nothing I've built uses jade though. Jade is considered (in my understanding) the 'old' way of doing sgml transformation and styling. I did a little poking at this, and I think it's likely an oddity of openjade (a bug likely, but who knows). I ran the command that generates the errors during the Make while is a subdirectory and referencing the sgml input file via ../ (this is the way the Makefile sets it up). This saw the errors Dago reported and an truss shows: open("/opt/csw/etc/sgml/../catalog", O_RDONLY) Err#2 ENOENT I then cd'd up a level and removed the ../ from the referenced sgml file. The file was then processed correctly and truss showed: open("/opt/csw/etc/sgml/catalog", O_RDONLY) = 3 Neither of those runs saw the environment modified. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 13 19:07:17 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 10:07:17 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242231941-sup-7634@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> Message-ID: <20090513170717.GI41179@bolthole.com> 9On Wed, May 13, 2009 at 01:01:59PM -0400, Ben Walton wrote: > I did a little poking at this, and I think it's likely an oddity of > openjade (a bug likely, but who knows). I ran the command that > generates the errors during the Make while is a subdirectory and > referencing the sgml input file via ../ (this is the way the Makefile > sets it up). This saw the errors Dago reported and an truss shows: > > open("/opt/csw/etc/sgml/../catalog", O_RDONLY) Err#2 ENOENT > > I then cd'd up a level and removed the ../ from the referenced sgml > file. The file was then processed correctly and truss showed: > > open("/opt/csw/etc/sgml/catalog", O_RDONLY) = 3 > > Neither of those runs saw the environment modified. Nice detective work. So then, sounds like jade needs to be recompiled, so that it doesnt use the '..' by default? From bwalton at opencsw.org Wed May 13 19:19:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:19:01 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513170717.GI41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> Message-ID: <1242234588-sup-3636@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 13:07:17 -0400 2009: > Nice detective work. So then, sounds like jade needs to be recompiled, so > that it doesnt use the '..' by default? I suspect the issue is deeper than that...in fact, I just confirmed it. I made a second directory nesting that required placing ../../ in front of the input file. The truss output now shows (on a failed run): open("/opt/csw/etc/sgml/../../catalog", O_RDONLY) Err#2 ENOENT With a fully qualified input, I tried to open catalog from the fully qualified directory of the input file. It's using the prefix from the input file for some reason. I'm not 100% positive that it's even a bug, simply because something _this_ silly should have shown up ages ago and been fixed. That, and they do provide a method to directly supply the catalog file path via the environment. If I have time, I'll see if this is a known or intended issue. If it's a known or unknown bug, I _might_ spend time fixing it, but not necessarily. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 13 19:30:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:30:47 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242234588-sup-3636@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> Message-ID: <1242235791-sup-1208@ntdws12.chass.utoronto.ca> Excerpts from Ben Walton's message of Wed May 13 13:19:01 -0400 2009: > It's using the prefix from the input file for some reason. I'm not > 100% positive that it's even a bug, simply because something _this_ > silly should have shown up ages ago and been fixed. That, and they do > provide a method to directly supply the catalog file path via the > environment. ...following up on this, the documentation for openjade explicitly instructs the user to set the variable: http://tldp.org/HOWTO/DocBook-OpenJade-SGML-XML-HOWTO/x456.html Braindead, but that's the way it is. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 13 19:36:27 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 10:36:27 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242234588-sup-3636@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> Message-ID: <20090513173627.GJ41179@bolthole.com> On Wed, May 13, 2009 at 01:19:01PM -0400, Ben Walton wrote: > It's using the prefix from the input file for some reason. I'm not > 100% positive that it's even a bug, simply because something _this_ > silly should have shown up ages ago and been fixed. That, and they do > provide a method to directly supply the catalog file path via the > environment. But I vaguely recall that you tried similar things, with some other "distro", and openjade did not behave in this manner. If so, then its a bug in "our" package....? From phil at bolthole.com Wed May 13 19:41:03 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 10:41:03 -0700 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <1242235791-sup-1208@ntdws12.chass.utoronto.ca> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> <1242235791-sup-1208@ntdws12.chass.utoronto.ca> Message-ID: <20090513174103.GK41179@bolthole.com> On Wed, May 13, 2009 at 01:30:47PM -0400, Ben Walton wrote: > ...following up on this, the documentation for openjade explicitly > instructs the user to set the variable: > http://tldp.org/HOWTO/DocBook-OpenJade-SGML-XML-HOWTO/x456.html I would disagree with your summary. If you "have to", then openjade would refuse to function at all, if it is not set. But that is not the case. It has some notion of default. the default is not functioning properly. Therefore, the default set in our package, needs fixing somehow. From ellson at opencsw.org Wed May 13 19:46:56 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 13:46:56 -0400 Subject: [csw-maintainers] pango, cairo, ... xcb, ... graphviz Message-ID: <4A0B0790.7040702@opencsw.org> I see there is a rebuild of pango/cairo underway, so I'm guessing you don't need a full bug report for this? On build8st: $ pkg-config --exists --print-errors "pangocairo >= 1.14.9" sh: gnome-config: not found Package xcb was not found in the pkg-config search path. Perhaps you should add the directory containing `xcb.pc' to the PKG_CONFIG_PATH environment variable Package 'xcb', required by 'cairo', not found From bwalton at opencsw.org Wed May 13 19:51:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:51:39 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513173627.GJ41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> <20090513173627.GJ41179@bolthole.com> Message-ID: <1242237020-sup-7245@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 13:36:27 -0400 2009: > But I vaguely recall that you tried similar things, with some other > "distro", and openjade did not behave in this manner. You're going to need to dig up the email on that, as I have no memory of it. > If so, then its a bug in "our" package....? Yes, if it behaves differently in the csw environment than elsewhere, it's a locally introduced bug (or at least environment/system specific). -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 13 19:57:03 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 13 May 2009 13:57:03 -0400 Subject: [csw-maintainers] Issues with openjade when compiling glib 1.3.15 In-Reply-To: <20090513174103.GK41179@bolthole.com> References: <200905131042.n4DAgURm020716@dfki.uni-kl.de> <20090513161348.GB41179@bolthole.com> <1242231941-sup-7634@ntdws12.chass.utoronto.ca> <20090513170717.GI41179@bolthole.com> <1242234588-sup-3636@ntdws12.chass.utoronto.ca> <1242235791-sup-1208@ntdws12.chass.utoronto.ca> <20090513174103.GK41179@bolthole.com> Message-ID: <1242237370-sup-7728@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 13 13:41:03 -0400 2009: > I would disagree with your summary. > If you "have to", then openjade would refuse to function at all, if it is > not set. , but yes, you're correct. > But that is not the case. It has some notion of default. the default is not > functioning properly. Therefore, the default set in our package, needs > fixing somehow. I just looked and there is a configure option to supply the default catalog. I'll rebuild with that option set explicitly. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at acm.org Wed May 13 20:40:33 2009 From: pfelecan at acm.org (Peter FELECAN) Date: Wed, 13 May 2009 20:40:33 +0200 Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: I'm forwarding this as it's of interest for all the maintainers and lets continue on the list please. -------------- next part -------------- An embedded message was scrubbed... From: James Lee Subject: Re: Compiler optimization Sun Studio vs. GCC Date: Wed, 13 May 2009 18:13:09 GMT Size: 3973 URL: -------------- next part -------------- -- Peter FELECAN mailto:pfelecan at acm.org From ellson at opencsw.org Wed May 13 21:03:23 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 15:03:23 -0400 Subject: [csw-maintainers] gar v. mgar ???? (warning: frustrated newbie questions) Message-ID: <4A0B197B.9090707@opencsw.org> What is "mgar" Is it a newer version of gar? Is mgar the recommended one for opencsw? Does it have a working "getting stated guide" ? I tried: http://apps.sourceforge.net/mediawiki/gar/index.php?title=Getting_Started_with_GAR referenced from: http://www.opencsw.org/standards/pkg-walkthrough but the "Checking out the build tree" instructions don't work. I've tried: http://svn.blastwave.org/trac/wiki/GettingStartedQuickly but "make fetch" doesn't work! John From bonivart at opencsw.org Wed May 13 21:10:05 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 13 May 2009 21:10:05 +0200 Subject: [csw-maintainers] gar v. mgar ???? (warning: frustrated newbie questions) In-Reply-To: <4A0B197B.9090707@opencsw.org> References: <4A0B197B.9090707@opencsw.org> Message-ID: <625385e30905131210n720945d0se1d837a9147cdc65@mail.gmail.com> On Wed, May 13, 2009 at 9:03 PM, John Ellson wrote: > What is "mgar" I think it means Modular GAR and it's the current GAR for OpenCSW. Try this: http://apps.sourceforge.net/trac/gar/wiki/Packaging%20with%20mGAR -- /peter From dam at opencsw.org Wed May 13 21:20:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:20:25 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Hi Peter, Am 13.05.2009 um 17:57 schrieb Peter FELECAN: > Mike Watters writes: >> John Ellson wrote: >>> Sorry for the confusion. I'll generate a Mantis report against gtk2 >>> requesting removal of its .la files. >> >> Before Generating the bug report, Try setting STRIP_LIBTOOL = 1 in >> your mGar >> Makefile. >> >> or if you are building by hand, after you run configure run gar/bin/ >> fixlibtool >> {Build Directory} > > This implies that you're on the public build stack and cannot be > reproduced on > a private build stack, i.e. my private servers. Not necessarily. Of course you can check out GAR at home and either build it with GAR (preferred) or use the script and build manually. > IMHO tools as fixlibtool > should be in a package such as cswutils. No, CSWcswclassutils contains runtime tools, fixlibtool is a compiletime tool. Best Regards -- Dago From dam at opencsw.org Wed May 13 21:24:29 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:24:29 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: <4A0AEFB5.1000301@opencsw.org> References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> <4A0AEFB5.1000301@opencsw.org> Message-ID: Hi John, Am 13.05.2009 um 18:05 schrieb John Ellson: > Mike Watters wrote: >> or if you are building by hand, after you run configure run gar/bin/ >> fixlibtool >> {Build Directory} >> > > Which, apparently still needs me to get started with gar :-( Yes. If you have any questions about GAR you are welcome to post them here :-) > And anyway, both of these sound like hacks, instead of fixing the > real problem which requires removing the > broken .la files from the distribution. > > Removing them will also help regular OpenCSW users, not just the > maintainers. Absolutely. However, there are not all packages in GAR and rebuilding all packages containing .la-files is a huge amount of work. That's why we decided to remove them one by one. Unfortunately, this means breaking other package builds who rely on a full .la-chain. To solve this, the script fixes these broken chains in the meantime. If you want the .la-files removed on a specific package you are free to file a bug report. Best regards -- Dago From phil at bolthole.com Wed May 13 21:33:25 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 12:33:25 -0700 Subject: [csw-maintainers] Updated glib In-Reply-To: References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: <20090513193324.GA24398@bolthole.com> On Wed, May 13, 2009 at 09:20:25PM +0200, Dagobert Michelsen wrote: > Hi Peter, > > Am 13.05.2009 um 17:57 schrieb Peter FELECAN: >> IMHO tools as fixlibtool >> should be in a package such as cswutils. > > No, CSWcswclassutils contains runtime tools, fixlibtool is a compiletime > tool. > you misread. Peter F is correct; it *would* nicely fit into CSWcswutils cswutils is not currently in gar. you can either give me an "official copy" to include, or you can put it in gar and take over the package. (You already know under what circumstances I will migrate my packages to gar ; unfortunately, you keep getting distracted before implementing them :-) From dam at opencsw.org Wed May 13 21:36:40 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:36:40 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp In-Reply-To: <20090513165907.GH41179@bolthole.com> References: <4A05A1A0.3090702@opencsw.org> <625385e30905111302y4d633e33o3072bfa2f4219c64@mail.gmail.com> <625385e30905120149q31ececf0hdf59e34986ef9935@mail.gmail.com> <20090512165334.GA84438@bolthole.com> <625385e30905121013u30e9c201xaf42566c5dd8d8ef@mail.gmail.com> <20090513165907.GH41179@bolthole.com> Message-ID: Hi Phil, Am 13.05.2009 um 18:59 schrieb Philip Brown: > On Tue, May 12, 2009 at 07:13:53PM +0200, Peter Bonivart wrote: >> Well, to identify which packages should go into which tier we could >> use another suggested topic of mine - download stats. > > Those stats might be skewed in terms of which packages get frequently > updated, not neccessarily in terms of importance. If you take statistics per-file instead of per-package you would get proper statistics. Nonetheless I would also like to see the download values. Sunfreeware has a policy that all mirrors must provide download statistics upstream for this reason. > And then there is a relative definition of "importance". > Does it mean "popular", or does it mean "critical to other programs"? > Or some combination of both? > > As far as "importance" goes, I think a fairly basic and concrete > measure > would be, the number of dependants that a package has. > eg: gtk2 has around 100. Thus, it is incontrovertibly an > "important package" ;-) Sure. That's why William and I repackaged it. If now someone would test it that would be even better ;-) Best regards -- Dago From dam at opencsw.org Wed May 13 21:47:51 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:47:51 +0200 Subject: [csw-maintainers] pango, cairo, ... xcb, ... graphviz In-Reply-To: <4A0B0790.7040702@opencsw.org> References: <4A0B0790.7040702@opencsw.org> Message-ID: <02A0B863-A125-429B-9651-8AA667ED376A@opencsw.org> Hi John, Am 13.05.2009 um 19:46 schrieb John Ellson: > I see there is a rebuild of pango/cairo underway, so I'm guessing > you don't need a full bug report for this? > > On build8st: > > $ pkg-config --exists --print-errors "pangocairo >= 1.14.9" > sh: gnome-config: not found > Package xcb was not found in the pkg-config search path. > Perhaps you should add the directory containing `xcb.pc' > to the PKG_CONFIG_PATH environment variable > Package 'xcb', required by 'cairo', not found You need to extend the PKG_CONFIG_PATH to /opt/csw/X11/lib/pkgconfig for 32 bit /opt/csw/X11/lib/64/pkgconfig for 64 bit In GAR you can do this by setting the following variables: EXTRA_INC = $(prefix)/X11/include EXTRA_LIB = $(prefix)/X11/lib EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib Best regards -- Dago From dam at opencsw.org Wed May 13 21:54:59 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 13 May 2009 21:54:59 +0200 Subject: [csw-maintainers] gar v. mgar ???? (warning: frustrated newbie questions) In-Reply-To: <4A0B197B.9090707@opencsw.org> References: <4A0B197B.9090707@opencsw.org> Message-ID: Hi John, Am 13.05.2009 um 21:03 schrieb John Ellson: > What is "mgar" > > Is it a newer version of gar? Yes. The GAR versions are described at > Is mgar the recommended one for opencsw? Yes, definitely. > Does it have a working "getting stated guide" ? > > I tried: > http://apps.sourceforge.net/mediawiki/gar/index.php?title=Getting_Started_with_GAR > referenced from: > http://www.opencsw.org/standards/pkg-walkthrough > but the "Checking out the build tree" instructions don't work. > > I've tried: > http://svn.blastwave.org/trac/wiki/GettingStartedQuickly > but "make fetch" doesn't work! The Blastwave GAR is a somewhat "old" as all the advancements like multi-ISA-builds, dynamic prototypes, single package out etc. were all developed after the fork. The documentation at covers the current OpenCSW GAR. Unfortunately, some of the docs have not been updated yet and I apologize for this. I am working on the docs from time to time, but as you know it is a time-consuming process. If you need any help on GAR please mail either me or ask on maintainers at . Best regards -- Dago From skayser at opencsw.org Wed May 13 22:30:57 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 13 May 2009 22:30:57 +0200 (CEST) Subject: [csw-maintainers] dbus-python (CSWpydbus) anyone? Message-ID: <49473.217.227.14.39.1242246657.squirrel@ssl.skayser.de> Hi, would anyone with fancy for python modules (Mike?) mind packaging dbus-python (http://dbus.freedesktop.org/releases/dbus-python/)? CSWpidgin contains purple-remote which can be used to cli-control pidgin. We would like to use this at work to automatically update the IM online status when people insert their smartcard into or pull their smartcard from Sun Ray DTUs. However, purple-remote requires the python dbus module, which i couldn't find in the catalogue. I had filed a bug report against pidgin [1], but as Chad correctly noted it is not really a bug in pidgin, so if someone could put together CSWpydbus this would make some guys here real happy. Sebastian [1] http://opencsw.org/mantis/view.php?id=3673 From mwatters at opencsw.org Wed May 13 22:43:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 13 May 2009 15:43:26 -0500 Subject: [csw-maintainers] libmcal now in testing Message-ID: <4A0B30EE.4040909@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Changes: updated to 0.7 now shared library removed static library 64bit support - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoLMO4ACgkQLrhmsXMSLxdvXACeNuU7OqHXslcB5U8zLUaKVlNg mBAAnjMXoN8i8UNcnYnBfkW/WnVS77YM =LxRv -----END PGP SIGNATURE----- From ellson at opencsw.org Wed May 13 23:10:10 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 13 May 2009 17:10:10 -0400 Subject: [csw-maintainers] status report: graphviz packaging - new maintainer Message-ID: <4A0B3732.8010808@opencsw.org> If any maintainers are interested, there is a reasonably complete version of graphviz (today's development snapshot) in /home/ellson/pkgs/graphviz/graphviz-2.23.20090513.0445,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz on the build farm. Should I put this someplace? It built. I haven't tried executing it. After installing, it will need "dot -c" run with installer privileges in order to register its plugins. If it installs ok you could try it with something like: echo "digraph{Hello->World}" | dot -Tpng | xv - I tried building an equivalent i386 version, but it needs the latest gtk2 which is not yet on build8xt ? I don't feel that this package is ready for release yet. I want to do the following first: - Generate a new stable release of graphviz upstream. - Migrate the opencsw maintenance to the mgar mechanism - Find out how to run "dot -c" in a post-install step with installer privileges. To register plugins. - Find out how to split the build products into multiple binary packages with segregated dependencies. - Try to package "gts" and "lasi" first, so that graphviz can use them - Not really critical, but see if I can get some more of the swig'ed language bindings to work currently building with support for: perl, C#, tcl missing are: python, php, ruby, guile, java, lua, ocaml It may just be that these languages are not installed on the build hosts? I haven't investigated yet. John From phil at bolthole.com Wed May 13 23:47:28 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 13 May 2009 14:47:28 -0700 Subject: [csw-maintainers] status report: graphviz packaging - new maintainer In-Reply-To: <4A0B3732.8010808@opencsw.org> References: <4A0B3732.8010808@opencsw.org> Message-ID: <20090513214728.GB21796@bolthole.com> On Wed, May 13, 2009 at 05:10:10PM -0400, John Ellson wrote: > > I don't feel that this package is ready for release yet. certainly not. the primary reason being, that it was compiled against packages that have not themselves been released yet! You cannot release a package that has been built on the buildXXXxxxt machines. the final "t" means "test *only*" From mwatters at opencsw.org Wed May 13 23:48:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 13 May 2009 16:48:08 -0500 Subject: [csw-maintainers] dbus-python (CSWpydbus) anyone? In-Reply-To: <49473.217.227.14.39.1242246657.squirrel@ssl.skayser.de> References: <49473.217.227.14.39.1242246657.squirrel@ssl.skayser.de> Message-ID: <4A0B4018.9040904@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Sebastian, Sebastian Kayser wrote: > Hi, > > would anyone with fancy for python modules (Mike?) mind packaging > dbus-python (http://dbus.freedesktop.org/releases/dbus-python/)? > > CSWpidgin contains purple-remote which can be used to cli-control pidgin. > We would like to use this at work to automatically update the IM online > status when people insert their smartcard into or pull their smartcard > from Sun Ray DTUs. > > However, purple-remote requires the python dbus module, which i couldn't > find in the catalogue. I had filed a bug report against pidgin [1], but as > Chad correctly noted it is not really a bug in pidgin, so if someone could > put together CSWpydbus this would make some guys here real happy. > > Sebastian > > [1] http://opencsw.org/mantis/view.php?id=3673 > > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I will add it to my list of things to compile after I get the bug fixes done for the current python. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoLQBcACgkQLrhmsXMSLxdf0gCfcLxkwd1DA/60HeXrbMjSYrKt fu4AnjUz11TwgCj/5dukyYbaxXPJLwaZ =SJmO -----END PGP SIGNATURE----- From maybird1776 at yahoo.com Thu May 14 02:47:29 2009 From: maybird1776 at yahoo.com (ken mays) Date: Wed, 13 May 2009 17:47:29 -0700 (PDT) Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: <962721.43059.qm@web111302.mail.gq1.yahoo.com> GCC 4.4.x (and v4.3.3) beats Sun Studio 11. From pfelecan at opencsw.org Thu May 14 09:54:04 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 14 May 2009 09:54:04 +0200 Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC In-Reply-To: <962721.43059.qm@web111302.mail.gq1.yahoo.com> (ken mays's message of "Wed\, 13 May 2009 17\:47\:29 -0700 \(PDT\)") References: <962721.43059.qm@web111302.mail.gq1.yahoo.com> Message-ID: ken mays writes: > GCC 4.4.x (and v4.3.3) beats Sun Studio 11. Can you develop this affirmation, i.e. what tests have you made to support it -- Peter From pfelecan at opencsw.org Thu May 14 09:57:04 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 14 May 2009 09:57:04 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: (Dagobert Michelsen's message of "Wed\, 13 May 2009 21\:20\:25 +0200") References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Dagobert Michelsen writes: >> IMHO tools as fixlibtool >> should be in a package such as cswutils. > > No, CSWcswclassutils contains runtime tools, fixlibtool is a compiletime > tool. Alright. Maybe this is the time to add a cswutils-devel package -- Peter From dam at opencsw.org Thu May 14 11:25:44 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 14 May 2009 11:25:44 +0200 Subject: [csw-maintainers] status report: graphviz packaging - new maintainer In-Reply-To: <4A0B3732.8010808@opencsw.org> References: <4A0B3732.8010808@opencsw.org> Message-ID: <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> Hi John, Am 13.05.2009 um 23:10 schrieb John Ellson: > If any maintainers are interested, there is a reasonably complete > version of graphviz (today's development snapshot) in > /home/ellson/pkgs/graphviz/ > graphviz-2.23.20090513.0445,REV=2009.05.13-SunOS5.8-sparc-CSW.pkg.gz > on the build farm. Should I put this someplace? Yes. Stuff you want to have tested can be copied to /home/testing from all machines in the BO buildfarm. > It built. I haven't tried executing it. After installing, > it will need "dot -c" run > with installer privileges in order to register its plugins. Before asking others to test it you should of course test it yourself to avoid wasting cycles of helping hands. > I tried building an equivalent i386 version, but it needs the latest > gtk2 which is not yet on build8xt ? Correct. The problem is the libs also include 64 bit versions. For x86 this means they must be build on Solaris 10, as this is the first version supporting 64 bit. Unfortunately we don't have a Solaris 10 x86 machine - yet! > I don't feel that this package is ready for release yet. I want > to do the following first: > > - Generate a new stable release of graphviz upstream. > - Migrate the opencsw maintenance to the mgar mechanism Let me know if you encounter any problems. > - Find out how to run "dot -c" in a post-install step with > installer privileges. To register plugins. This is simple. Just take a look at the many packages doing it: cd mgar/pkg grep postinstall */trunk/Makefile > - Find out how to split the build products into multiple binary > packages with segregated dependencies. This is documented at > - Try to package "gts" and "lasi" first, so that graphviz can use > them > - Not really critical, but see if I can get some more of the > swig'ed language bindings to work > currently building with support for: perl, C#, tcl > missing are: python, php, ruby, guile, java, lua, ocaml > It may just be that these languages are not installed on the > build hosts? I haven't investigated yet. Swig is also unmaintained at the moment and outdated. Feel free to take it over or to ask for somebody else to update it if you need a newer version. Best regards -- Dago From dam at opencsw.org Thu May 14 11:30:48 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 14 May 2009 11:30:48 +0200 Subject: [csw-maintainers] Updated glib In-Reply-To: References: <06e9229f5c0b834b3bfc92b3a5d592db@opencsw.org> <3B41CD10-1B59-4F6F-8367-9556F4A5DF1C@opencsw.org> <4A0AD2EC.105@research.att.com> <4A0AE83F.3090502@opencsw.org> Message-ID: Hi Peter, Am 14.05.2009 um 09:57 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >>> IMHO tools as fixlibtool >>> should be in a package such as cswutils. >> >> No, CSWcswclassutils contains runtime tools, fixlibtool is a >> compiletime >> tool. > > Alright. Maybe this is the time to add a cswutils-devel package Phil informed me that there is also a CSWcswutils package which I wasn't aware of. fixlibtool will go into that package at the next opportunity. Best regards -- Dago From maybird1776 at yahoo.com Thu May 14 15:39:59 2009 From: maybird1776 at yahoo.com (ken mays) Date: Thu, 14 May 2009 06:39:59 -0700 (PDT) Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: <500379.89918.qm@web111302.mail.gq1.yahoo.com> --- On Thu, 5/14/09, Peter FELECAN wrote: > From: Peter FELECAN > Subject: Re: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC > To: "internal list for the CSW maintainers" > Date: Thursday, May 14, 2009, 3:54 AM > ken mays > writes: > > > GCC 4.4.x (and v4.3.3) beats Sun Studio 11. > > Can you develop this affirmation, i.e. what tests have you > made to support it. > -- > Peter Hi Peter, Yes, I have been testing both compilers for many years in various situations. I'm not here to bash Sun Studio, though, and it is just my opinion (based on years of factual knowledge). But in saying that, I do use Sun Studio and like it for what it does. Basically, a sedan and a sports car will provide you with decent transportation. Just depends on how you like to get there and the "creature features" those cars provide.... Ken From maybird1776 at yahoo.com Thu May 14 15:59:33 2009 From: maybird1776 at yahoo.com (ken mays) Date: Thu, 14 May 2009 06:59:33 -0700 (PDT) Subject: [csw-maintainers] [James Lee] Re: Compiler optimization Sun Studio vs. GCC Message-ID: <947301.24137.qm@web111306.mail.gq1.yahoo.com> > > GCC 4.4.x (and v4.3.3) beats Sun Studio 11. > > Can you develop this affirmation, i.e. what tests have you > made to > support it > -- > Peter For the record, performance-wise and feature-wise you'll usually find that Sun Studio 11 beats GCC with proper speed optimizations being set. I'll always chose Sun Studio when I need to get performance and utilize advanced features (aka the Aston Martin One-77 sports car in this case). You take it out on the weekends and let the wind blow through your toes and fingers. GCC is like the reliable car in the garage... you never want to just stick it on the shelf as that sports car is just not the full family and everyday work car just yet. Here, it beats Sun Studio 11 in handling and porting that mangled code (aka bad children) in the wild. I could speak on fine wine versus wine (grape juice). But, I like using cars... (smile) ~ Ken From ellson at opencsw.org Thu May 14 16:56:39 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 10:56:39 -0400 Subject: [csw-maintainers] help with "make fetch" Message-ID: <4A0C3127.7080203@opencsw.org> I'm working on add my packages to mgar, but I can't get "make fetch" to work. So I tried "make fetch" in an existing package, libxml2, and it doesn't work there either! I suspect I must have some broken utility in my PATH. Any idea what? I've check the obvious: sed, gsed, awg, gawk, find, make, gmake, And what happened to the ":" in the ftp URL? ellson at tools:/mgar/pkg/libxml2/trunk> make fetch [===== NOW BUILDING: libxml2-2.7.2 =====] ginstall -d cookies/global ginstall -d download ginstall -d download/partial ==> Verifying for installed package CSWgmake: installed ==> Verifying for installed package CSWgtar: installed ==> Verifying for installed package CSWggrep: installed ==> Verifying for installed package CSWdiffutils: installed ==> Verifying for installed package CSWgfile: installed ==> Verifying for installed package CSWtextutils: installed ==> Verifying for installed package CSWwget: installed ==> Verifying for installed package CSWfindutils: installed ==> Verifying for installed package CSWgsed: installed ==> Verifying for installed package CSWgawk: installed ==> Verifying for installed package CSWbzip2: installed [prerequisite] complete for libxml2. ==> Grabbing download/CSWlibxml2.depend ==> Trying file//files/CSWlibxml2.depend make: Fatal error in reader: Makefile, line 7: Unexpected end of line seen Current working directory /home/ellson/mgar/pkg/libxml2/trunk ==> Trying file///home/src/CSWlibxml2.depend make: Fatal error in reader: Makefile, line 7: Unexpected end of line seen Current working directory /home/ellson/mgar/pkg/libxml2/trunk ==> Trying ftp//xmlsoft.org/libxml2/CSWlibxml2.depend make: Fatal error in reader: Makefile, line 7: Unexpected end of line seen Current working directory /home/ellson/mgar/pkg/libxml2/trunk (!!!) Failed to download download/CSWlibxml2.depend! make: *** [download/CSWlibxml2.depend] Error 255 ellson at tools:/mgar/pkg/libxml2/trunk> From bonivart at opencsw.org Thu May 14 17:05:51 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 14 May 2009 17:05:51 +0200 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C3127.7080203@opencsw.org> References: <4A0C3127.7080203@opencsw.org> Message-ID: <625385e30905140805t95e9469g73a874f69127f71f@mail.gmail.com> On Thu, May 14, 2009 at 4:56 PM, John Ellson wrote: > I'm working on add my packages to mgar, but I can't get "make fetch" to > work. I don't know what make points at in your environment but I always use gmake. This is my path: PATH=/opt/csw/bin:/opt/studio/SOS11/SUNWspro/bin:/usr/bin:/bin:/usr/sbin:/sbin:/opt/csw/bin -- /peter From bwalton at opencsw.org Thu May 14 17:07:25 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 11:07:25 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C3127.7080203@opencsw.org> References: <4A0C3127.7080203@opencsw.org> Message-ID: <1242313508-sup-7220@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 10:56:39 -0400 2009: Hi John, > So I tried "make fetch" in an existing package, libxml2, and it doesn't > work there either! The errors look as though the file CSWlibxml2.depend is missing from the trunk/files directory where it would be expected to live. Can you see the file in the directory? > I suspect I must have some broken utility in my PATH. Any idea what? > I've check the obvious: > sed, gsed, awg, gawk, find, make, gmake, As long as the proxy settings are proper (for wherever you're doing the fetch), you should be fine. > And what happened to the ":" in the ftp URL? : is a special character in Make, so urls for fetching are stipped of them. The actual wget commands add them back in (gar.lib.mk, if you're interested). HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 14 17:26:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 11:26:06 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <1242313508-sup-7220@ntdws12.chass.utoronto.ca> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> Message-ID: <4A0C380E.60402@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Thu May 14 10:56:39 -0400 2009: > > Hi John, > > >> So I tried "make fetch" in an existing package, libxml2, and it doesn't >> work there either! >> > > The errors look as though the file CSWlibxml2.depend is missing from > the trunk/files directory where it would be expected to live. Can you > see the file in the directory? > Yes. Its there. Should I try another mgar package that you know works for you? > >> I suspect I must have some broken utility in my PATH. Any idea what? >> I've check the obvious: >> sed, gsed, awg, gawk, find, make, gmake, >> > > As long as the proxy settings are proper (for wherever you're doing > the fetch), you should be fine. > No proxy settings needed here. "wget ftp://xmlsoft.org/libxml2/libxml2-2.7.2.tar.gz" works from the command line. > >> And what happened to the ":" in the ftp URL? >> > > : is a special character in Make, so urls for fetching are stipped of > them. The actual wget commands add them back in (gar.lib.mk, if > you're interested). > > HTH. > -Ben > From bwalton at opencsw.org Thu May 14 17:39:28 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 11:39:28 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C380E.60402@opencsw.org> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> Message-ID: <1242315443-sup-7175@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 11:26:06 -0400 2009: > Should I try another mgar package that you know works for you? No, give Peter's gmake suggestion a try. I didn't think of that (since it's part of my habit now). HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 14 17:44:02 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 11:44:02 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <1242315443-sup-7175@ntdws12.chass.utoronto.ca> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> Message-ID: <4A0C3C42.9050205@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Thu May 14 11:26:06 -0400 2009: > >> Should I try another mgar package that you know works for you? >> > > No, give Peter's gmake suggestion a try. I didn't think of that > (since it's part of my habit now). > > HTH. > -Ben > That worked, but why ??? I have: $ which make /home/ellson/bin/make $ ls -l /home/ellson/bin/make lrwxrwxrwx 1 ellson ellson 18 2009-04-28 18:02 /home/ellson/bin/make -> /opt/csw/bin/gmake $ which gmake /opt/csw/bin/gmake $ So why doesn't "make" work? It is "gmake" John From bwalton at opencsw.org Thu May 14 17:51:48 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 11:51:48 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C3C42.9050205@opencsw.org> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> <4A0C3C42.9050205@opencsw.org> Message-ID: <1242316123-sup-6398@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 11:44:02 -0400 2009: > That worked, but why ??? The recursive calls to $(MAKE) will simply call 'make' later on, which will use the non-gnu system version. You could override PATH in your Makefile to prepend /opt/csw/gnu (which has a make -> gmake symlink) to make this go if you wanted, but you'd have to do that in every recipe. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 14 18:01:07 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 12:01:07 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <1242316123-sup-6398@ntdws12.chass.utoronto.ca> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> <4A0C3C42.9050205@opencsw.org> <1242316123-sup-6398@ntdws12.chass.utoronto.ca> Message-ID: <4A0C4043.7070208@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Thu May 14 11:44:02 -0400 2009: > >> That worked, but why ??? >> > > The recursive calls to $(MAKE) will simply call 'make' later on, which > will use the non-gnu system version. You could override PATH in your > Makefile to prepend /opt/csw/gnu (which has a make -> gmake symlink) > to make this go if you wanted, but you'd have to do that in every > recipe. > > Thanks > -Ben > I don't have an /opt/csw/gnu What package does that come from? I have /home/ellson/bin at the beginning of my PATH, and it already contains make->/opt/csw/bin/gmake. Doesn't fix it. Adding: export MAKE=/opt/csw/bin/make seems to help. Didn't someone write a: "Recursive makes considered harmful" ? From bwalton at opencsw.org Thu May 14 18:31:21 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 14 May 2009 12:31:21 -0400 Subject: [csw-maintainers] help with "make fetch" In-Reply-To: <4A0C4043.7070208@opencsw.org> References: <4A0C3127.7080203@opencsw.org> <1242313508-sup-7220@ntdws12.chass.utoronto.ca> <4A0C380E.60402@opencsw.org> <1242315443-sup-7175@ntdws12.chass.utoronto.ca> <4A0C3C42.9050205@opencsw.org> <1242316123-sup-6398@ntdws12.chass.utoronto.ca> <4A0C4043.7070208@opencsw.org> Message-ID: <1242318528-sup-4777@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Thu May 14 12:01:07 -0400 2009: > I don't have an /opt/csw/gnu What package does that come from? gnulinks > I have /home/ellson/bin at the beginning of my PATH, and it already > contains make->/opt/csw/bin/gmake. Doesn't fix it. This is because GAR sets a path explicitly. You can override it (per-recipe) by setting something like: PATH := /opt/csw/gnu:$(PATH) (The := prevents recursive expansion.) > Adding: > export MAKE=/opt/csw/bin/make > seems to help. That would do it too. > Didn't someone write a: "Recursive makes considered harmful" ? I don't know. In situations where the system make is used, it's not really an issue. This is only a problem because the default make isn't capable of running GAR (which leverages many features of the gnu version that aren't available in the solaris one). -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 235 bytes Desc: not available URL: From ellson at opencsw.org Thu May 14 20:29:25 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 14 May 2009 14:29:25 -0400 Subject: [csw-maintainers] mgar: how to maintain a patch? Message-ID: <4A0C6305.6020902@opencsw.org> I'm trying to package gts with mgar. Its an optional dependency for graphviz and its a smaller package for me to learn about mgar. gts-0.7.6 doesn't build on Solaris (it uses fabsf() which isn't available). The patch is easy, and I've reported the problem upstream and provided them with a patch. Meanwhile, how do I maintain and apply a patch in mgar? A pointer to an example pkg with patches would be fine... John From skayser at opencsw.org Thu May 14 20:51:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 14 May 2009 20:51:19 +0200 (CEST) Subject: [csw-maintainers] mgar: how to maintain a patch? In-Reply-To: <4A0C6305.6020902@opencsw.org> References: <4A0C6305.6020902@opencsw.org> Message-ID: <59208.194.246.122.22.1242327079.squirrel@ssl.skayser.de> Hi John, John Ellson wrote: > I'm trying to package gts with mgar. Its an optional dependency for > graphviz and its a smaller > package for me to learn about mgar. > > gts-0.7.6 doesn't build on Solaris (it uses fabsf() which isn't > available). The patch is easy, and I've reported the > problem upstream and provided them with a patch. > > Meanwhile, how do I maintain and apply a patch in mgar? A pointer to > an example pkg with patches would be fine... socat [1], or xterm [2], or ncdu [3] would be an example. You simply create a patch file (or multiple patch files) for the extracted source directory, store them in the files/ subdirectory of your GAR build directory and list the patch files in $(PATCHFILES). They would then get applied in the "patch" phase of the build. So if you do a "gmake extract", the patches won't be in there yet, but when you do a "gmake patch" they will be applied. Ah, and you need to do a "gmake makesum" after referencing the patch files, so that the diffs are covered by the checksums file. To initially create the patch you can either create it manually or call "gmake extract", patch the sources and then do a "gmake makepatch". You will end up with one gar-base.diff in the files/ subdirectory, which you can reference in PATCHFILES. I have done so for ncdu. Maintaining a set of separate patch files can be a bit of a hassle, but i have found it easier to keep track of which patch is responsible for what (and updating the PATCHFILES list, once a patch has been committed upstream). Sebastian [1] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/socat/trunk/Makefile [2] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/xterm/trunk/Makefile [3] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/ncdu/trunk/Makefile From william at wbonnet.net Thu May 14 20:52:53 2009 From: william at wbonnet.net (William Bonnet) Date: Thu, 14 May 2009 20:52:53 +0200 Subject: [csw-maintainers] mgar: how to maintain a patch? In-Reply-To: <4A0C6305.6020902@opencsw.org> References: <4A0C6305.6020902@opencsw.org> Message-ID: <4A0C6885.2020502@wbonnet.net> Hi John > I'm trying to package gts with mgar. Its an optional dependency for > graphviz and its a smaller > package for me to learn about mgar. > > gts-0.7.6 doesn't build on Solaris (it uses fabsf() which isn't > available). The patch is easy, and I've reported the > problem upstream and provided them with a patch. > > Meanwhile, how do I maintain and apply a patch in mgar? A pointer > to an example pkg with patches would be fine... Many packages are using patches, You can have a look to x11/libX11 for an example. Basically it is easy to create. 1/ Extract the sources gmake extract 2/ Patch manually the files 3/ Create the patch file gmake makepatch 4/ Add the following line to your Makefile PATCHFILES = gar-base.diff 5/ rebuild your package (at least go until patching) gmake clean && gmake patch This should clean sources, download, extract and apply patch file. The new sources should have the modification you made Here is for the short version :) cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Fri May 15 03:13:25 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 14 May 2009 20:13:25 -0500 Subject: [csw-maintainers] php4.4.9 now in testing Message-ID: <4A0CC1B5.3070506@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 this is the last of the PHP4.x releases from upstream. this has had no upstream support since December 31st 2007. If you are still using PHP4, Please consider upgrading to PHP5 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoMwbUACgkQLrhmsXMSLxfiNQCfaILknhDOyzJinhscfk0dGvHT o14AnjJqlzbVrujCj9G27Wg9P20XMSFw =2lGb -----END PGP SIGNATURE----- From rupert at opencsw.org Fri May 15 06:10:42 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 15 May 2009 06:10:42 +0200 Subject: [csw-maintainers] pysqlite / trac not yet working with subversion-1.6.1 In-Reply-To: <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> References: <94E2A58978FC324196A142DE8399AB7801E75BA8@chsa1556.share.beluni.net> <6af4270905090449p671eeaf5n4bbc91eff0314e0e@mail.gmail.com> <4A05936B.3020001@opencsw.org> <6af4270905101431w56fffbddm796d4b62bac0ffbc@mail.gmail.com> Message-ID: <6af4270905142110o2e3fa9fcy64870fcb87a77b0a@mail.gmail.com> svn-1.6.2 is out now ... checksum is the same. did you always configure without bdb support, because when building it complains that bdb is not compiled in, as it is not installed on the build servers? rupert. On Sun, May 10, 2009 at 23:31, rupert THURNER wrote: > you are right, i put a hint also into the mantis issue. > > i tested with the upcoming 1.6.2 release, and it works. i left it on > testing for now, but we should remove it as soon as possible > (http://mirror.opencsw.org/testing/subversion-1.6.2,REV=2009.05.09-SunOS5.8-sparc-UNCOMMITTED.pkg.gz > ). > > as 1.6.2 is already tagged we might wait the couple of days for the > final 1.6.2 release instead of patching 1.6.1. > > rupert. > > > On Sat, May 9, 2009 at 16:30, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> rupert THURNER wrote: >>> hi, >>> >>> while subversion-1.6.1 works fine with the old repository format, we >>> hit http://trac.edgewall.org/ticket/8199 with trac-0.11.4 after >>> upgrading to 1.6 repository format. >>> >>> the fix will be in 1.6.2. i'll give it a try. >> >> This sounds eerily similar to >> http://www.opencsw.org/mantis/view.php?id=3661 >> http://subversion.tigris.org/issues/show_bug.cgi?id=3387 >> >> I will patch up 1.6.1 with the fix for our bug 3361. ?if you wouldn't mind >> testing that when released against your bug? >> >> if it does not fix it, please file a new opencsw bug. >> I hope to have the fix for subversion done next week. >> >> - -- >> Thanks, >> Mike >> >> "Any intelligent fool can make things bigger, more complex, >> and more violent. ?It takes a touch of genius -- and a lot of courage -- >> to move in the opposite direction." >> >> * Albert Einstein 1879 - 1955 >> ? ?US German-born Theoretical Physicist >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v1.4.9 (GNU/Linux) >> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org >> >> iEYEARECAAYFAkoFk2sACgkQLrhmsXMSLxcIUgCfRdPMEBIRiyVwEElcTY1brJIA >> 1iIAoMvXfvesGIcvkUH36E9Ub/O0f0iw >> =Du0U >> -----END PGP SIGNATURE----- >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers >> > From rupert at opencsw.org Fri May 15 07:04:49 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 15 May 2009 07:04:49 +0200 Subject: [csw-maintainers] parallel build on build8x and build8s fails? Message-ID: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> i tried to build svn-1.6.2 at the same time on build8x, and build8s. on build8s the following error(s) came .... /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-contrib/svn-push /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svnversion /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnauthz-validate /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnmucc /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff3 /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svn-populate-node-origins-index /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff4 /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter gmake[1]: *** [merge-copy-all] Error 1 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' gmake: *** [merge-isa-sparcv8] Error 2 ... /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter gmake[1]: *** [merge-copy-all] Error 1 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' gmake: *** [merge-isa-sparcv8] Error 2 is there anythign i could do to avoid that - other than build one after the other? From rupert at opencsw.org Fri May 15 07:13:22 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 15 May 2009 07:13:22 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> Message-ID: <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> as "top" said "96% idle" on build8s, i thought i'd try: ? make -j60 package but .. it gives: ? gmake[1]: warning: jobserver unavailable: using -j1. ?Add `+' to parent make rule. is there a possibility to trigger parallel compiles? rupert. From mwatters at opencsw.org Fri May 15 08:18:34 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 15 May 2009 01:18:34 -0500 Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> Message-ID: <4A0D093A.9040107@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > i tried to build svn-1.6.2 at the same time on build8x, and build8s. > on build8s the following error(s) came > > .... > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-contrib/svn-push > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svnversion > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnauthz-validate > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svnmucc > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff3 > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/svn-populate-node-origins-index > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svn-tools/diff4 > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter > gmake[1]: *** [merge-copy-all] Error 1 > gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' > gmake: *** [merge-isa-sparcv8] Error 2 > ... > /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/bin/svndumpfilter > gmake[1]: *** [merge-copy-all] Error 1 > gmake[1]: Leaving directory `/home/rupert/mgar/pkg/subversion/trunk' > gmake: *** [merge-isa-sparcv8] Error 2 > > > is there anythign i could do to avoid that - other than build one > after the other? > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers for a good example on doing parallel builds see the gcc4 recipe. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoNCToACgkQLrhmsXMSLxeHOwCgmJGqJmOnn4lna4I2PIs3x7tB DAoAn3NmUQ5egClpUuR2fyCbO2dxeZ0t =UbfS -----END PGP SIGNATURE----- From dam at opencsw.org Fri May 15 08:55:10 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 15 May 2009 08:55:10 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Hi Rupert, Am 15.05.2009 um 07:13 schrieb rupert THURNER: > as "top" said "96% idle" on build8s, i thought i'd try: > make -j60 package > > but .. it gives: > gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to > parent make rule. > > is there a possibility to trigger parallel compiles? If you want to do it with GAR please use PARALLELMFLAGS="-l -j 30" gmake package This way the flags are properly propagated to subinvocations. Best regards -- Dago From thurner.rupert at redleo.org Fri May 15 07:09:30 2009 From: thurner.rupert at redleo.org (THURNER rupert) Date: Fri, 15 May 2009 07:09:30 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... Message-ID: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> as "top" said "96% idle" on build8s, i thought i'd try: make -j60 package but .. it gives: gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to parent make rule. is there a possibility to trigger parallel compiles? rupert. From pfelecan at opencsw.org Fri May 15 10:21:00 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 15 May 2009 10:21:00 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: (Dagobert Michelsen's message of "Fri\, 15 May 2009 08\:55\:10 +0200") References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Dagobert Michelsen writes: > Hi Rupert, > > Am 15.05.2009 um 07:13 schrieb rupert THURNER: >> as "top" said "96% idle" on build8s, i thought i'd try: >> make -j60 package >> >> but .. it gives: >> gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to >> parent make rule. >> >> is there a possibility to trigger parallel compiles? > > If you want to do it with GAR please use > PARALLELMFLAGS="-l -j 30" gmake package > This way the flags are properly propagated to subinvocations. -j 30 ? How many CPUs do you have in the machine? Reading of the GNU make manual is recommended. -- Peter From dam at opencsw.org Fri May 15 10:54:56 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 15 May 2009 10:54:56 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Hi Peter, Am 15.05.2009 um 10:21 schrieb Peter FELECAN: > Dagobert Michelsen writes: > >> Hi Rupert, >> >> Am 15.05.2009 um 07:13 schrieb rupert THURNER: >>> as "top" said "96% idle" on build8s, i thought i'd try: >>> make -j60 package >>> >>> but .. it gives: >>> gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to >>> parent make rule. >>> >>> is there a possibility to trigger parallel compiles? >> >> If you want to do it with GAR please use >> PARALLELMFLAGS="-l -j 30" gmake package >> This way the flags are properly propagated to subinvocations. > > -j 30 ? How many CPUs do you have in the machine? Reading of the GNU > make manual is recommended. > build8s% psrinfo > 0 on-line since 04/06/2009 16:33:22 > 1 on-line since 04/06/2009 16:33:26 > 2 on-line since 04/06/2009 16:33:26 > 3 on-line since 04/06/2009 16:33:26 > 4 on-line since 04/06/2009 16:33:26 > 5 on-line since 04/06/2009 16:33:26 > 6 on-line since 04/06/2009 16:33:26 > 7 on-line since 04/06/2009 16:33:26 > 8 on-line since 04/06/2009 16:33:26 > 9 on-line since 04/06/2009 16:33:26 > 10 on-line since 04/06/2009 16:33:26 > 11 on-line since 04/06/2009 16:33:26 > 12 on-line since 04/06/2009 16:33:26 > 13 on-line since 04/06/2009 16:33:26 > 14 on-line since 04/06/2009 16:33:26 > 15 on-line since 04/06/2009 16:33:26 > 16 on-line since 04/06/2009 16:33:26 > 17 on-line since 04/06/2009 16:33:26 > 18 on-line since 04/06/2009 16:33:26 > 19 on-line since 04/06/2009 16:33:26 > 20 on-line since 04/06/2009 16:33:26 > 21 on-line since 04/06/2009 16:33:26 > 22 on-line since 04/06/2009 16:33:26 > 23 on-line since 04/06/2009 16:33:26 > 24 on-line since 04/06/2009 16:33:26 > 25 on-line since 04/06/2009 16:33:26 > 26 on-line since 04/06/2009 16:33:26 > 27 on-line since 04/06/2009 16:33:26 > 28 on-line since 04/06/2009 16:33:26 > 29 on-line since 04/06/2009 16:33:26 > 30 on-line since 04/06/2009 16:33:26 > 31 on-line since 04/06/2009 16:33:26 I guess it is a good idea to save 2 CPUs for other users ;-) (It is a T5220 with a 4 Core T2 w/8 strands each) AND: DO NOT DO THIS UNDER ANY CIRCUMSTANCE ON build8x OR ANY OTHER X86 MACHINE FROM THE FARM OR IT WILL CATCH FIRE!! Best regards -- Dago From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 11:07:59 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 11:07:59 +0200 Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> Message-ID: <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> rupert THURNER wrote: > i tried to build svn-1.6.2 at the same time on build8x, and build8s. > on build8s the following error(s) came > > .... > is there anythign i could do to avoid that - other than build one > after the other? use the schily makefile system ;-) You need a build system that was written to support parallel builds. The schily makefilesystem allows to run concurrent make's on all supported platforms since 1992. All other build systems that I am aware don't. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 11:13:32 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 11:13:32 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <4a0d323c.S75GRQ6tB9nGkuZV%Joerg.Schilling@fokus.fraunhofer.de> Dagobert Michelsen wrote: > I guess it is a good idea to save 2 CPUs for other users ;-) > (It is a T5220 with a 4 Core T2 w/8 strands each) For best performance (and in case of real CPUs - not only CMT), I use twice the number of the available CPUs if I compile Solaris If you have 4 real CPU cores and use dmake -j4 to compile Solaris, you will have CPU's that are idle because they are waiting for I/O. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From james at opencsw.org Fri May 15 11:13:57 2009 From: james at opencsw.org (James Lee) Date: Fri, 15 May 2009 09:13:57 GMT Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <20090515.9135700.3717487274@gyor.oxdrove.co.uk> On 15/05/09, 09:54:56, Dagobert Michelsen wrote regarding Re: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ...: > > -j 30 ? How many CPUs do you have in the machine? Reading of the GNU > > make manual is recommended. > > build8s% psrinfo ... > > 31 on-line since 04/06/2009 16:33:26 > I guess it is a good idea to save 2 CPUs for other users ;-) Assuming another using isn't doing the same, ie, don't bother to assume. It's rare that the full complement of 32 threads is used for long by real makefiles. > AND: DO NOT DO THIS UNDER ANY CIRCUMSTANCE ON build8x OR ANY OTHER > X86 MACHINE FROM THE FARM OR IT WILL CATCH FIRE!! $ gmake -j $(psrinfo | wc -l) ... James. From james at opencsw.org Fri May 15 11:21:29 2009 From: james at opencsw.org (James Lee) Date: Fri, 15 May 2009 09:21:29 GMT Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20090515.9212900.560377280@gyor.oxdrove.co.uk> On 15/05/09, 10:07:59, Joerg Schilling wrote regarding Re: [csw-maintainers] parallel build on build8x and build8s fails?: > > is there anythign i could do to avoid that - other than build one > > after the other? > use the schily makefile system ;-) > You need a build system that was written to support parallel builds. > The schily makefilesystem allows to run concurrent make's on all > supported platforms since 1992. All other build systems that I am aware > don't. The OpenOffice.org build system can perform parallel builds of both its project components and can pass a parallel flag to individual projects (although not all sub projects take advantage of the passed flags). James. From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 11:24:56 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 11:24:56 +0200 Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <20090515.9212900.560377280@gyor.oxdrove.co.uk> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> <20090515.9212900.560377280@gyor.oxdrove.co.uk> Message-ID: <4a0d34e8.4ixWzebDMg5/iKa2%Joerg.Schilling@fokus.fraunhofer.de> James Lee wrote: > On 15/05/09, 10:07:59, Joerg Schilling > wrote regarding Re: [csw-maintainers] > parallel build on build8x and build8s fails?: > > > > is there anythign i could do to avoid that - other than build one > > > after the other? > > > use the schily makefile system ;-) > > > You need a build system that was written to support parallel builds. > > The schily makefilesystem allows to run concurrent make's on all > > supported platforms since 1992. All other build systems that I am aware > > don't. > > The OpenOffice.org build system can perform parallel builds of both its > project components and can pass a parallel flag to individual projects > (although not all sub projects take advantage of the passed flags). Are you talking about parallel builds running on different platforms using the same NFS mounted directory tree? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From james at opencsw.org Fri May 15 11:36:18 2009 From: james at opencsw.org (James Lee) Date: Fri, 15 May 2009 09:36:18 GMT Subject: [csw-maintainers] parallel build on build8x and build8s fails? In-Reply-To: <4a0d34e8.4ixWzebDMg5/iKa2%Joerg.Schilling@fokus.fraunhofer.de> References: <6af4270905142204i6141d242pe23e52aea7aa7b1f@mail.gmail.com> <4a0d30ef.MjnVYswCWFHxCgic%Joerg.Schilling@fokus.fraunhofer.de> <20090515.9212900.560377280@gyor.oxdrove.co.uk> <4a0d34e8.4ixWzebDMg5/iKa2%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20090515.9361800.3578221933@gyor.oxdrove.co.uk> On 15/05/09, 10:24:56, Joerg Schilling wrote regarding Re: [csw-maintainers] parallel build on build8x and build8s fails?: > > The OpenOffice.org build system can perform parallel builds of both its > > project components and can pass a parallel flag to individual projects > > (although not all sub projects take advantage of the passed flags). > Are you talking about parallel builds running on different platforms > using the same NFS mounted directory tree? No, although it can share the source tree for different arch builds. I mean it is a amalgamation of several different sources or projects and it can build non-dependent sub projects in parallel. It is also possible to distribute builds across multiple machines. James. From pfelecan at opencsw.org Fri May 15 12:21:32 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 15 May 2009 12:21:32 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: (Dagobert Michelsen's message of "Fri\, 15 May 2009 10\:54\:56 +0200") References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: Dagobert Michelsen writes: > Hi Peter, > > Am 15.05.2009 um 10:21 schrieb Peter FELECAN: > >> Dagobert Michelsen writes: >> >>> Hi Rupert, >>> >>> Am 15.05.2009 um 07:13 schrieb rupert THURNER: >>>> as "top" said "96% idle" on build8s, i thought i'd try: >>>> make -j60 package >>>> >>>> but .. it gives: >>>> gmake[1]: warning: jobserver unavailable: using -j1. Add `+' to >>>> parent make rule. >>>> >>>> is there a possibility to trigger parallel compiles? >>> >>> If you want to do it with GAR please use >>> PARALLELMFLAGS="-l -j 30" gmake package >>> This way the flags are properly propagated to subinvocations. >> >> -j 30 ? How many CPUs do you have in the machine? Reading of the GNU >> make manual is recommended. > >> build8s% psrinfo >> 0 on-line since 04/06/2009 16:33:22 >> [...] >> 31 on-line since 04/06/2009 16:33:26 > > I guess it is a good idea to save 2 CPUs for other users ;-) > (It is a T5220 with a 4 Core T2 w/8 strands each) Oh, I didn't know that. The fact that the build8s is a zone has any impact on this kind of usage? -- Peter From dam at opencsw.org Fri May 15 12:29:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 15 May 2009 12:29:19 +0200 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <759CF688-6020-4F61-859E-F40A65132A7A@opencsw.org> Hi Peter, Am 15.05.2009 um 12:21 schrieb Peter FELECAN: > Oh, I didn't know that. The fact that the build8s is a zone has any > impact on this kind of usage? No, I haven't configured resource constraints on the zones. The machine should be large enough to hold some more builds. Best regards -- Dago From skayser at opencsw.org Fri May 15 15:43:10 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 15 May 2009 15:43:10 +0200 Subject: [csw-maintainers] pkgname limit 20 chars? Message-ID: <4A0D716E.4010700@opencsw.org> Hi, i just wanted to package the perl module DateTime::TimeZone and checkpkg bails out on the pkgname: CSWpmdatetimetimezone ==> Checking compliance: CSWpmdatetimetimezone Examining /home/skayser/pkgs/pm_datetimetimezone-0.90,REV=2009.05.15-SunOS5.8-all-CSW.pkg.gz ... pkgname='CSWpmdatetimetimezone' ... ERROR: /tmp/pm_datetimetimezone-0.90,REV=2009.05.15-SunOS5.8-all-CSW.pkg.gz: pkg name greater than 20 chars gmake: *** [pkgcheck-CSWpmdatetimetimezone] Error 2 I have seen that a couple of packages already condense their package and software names, but IMHO this leads to a rather inconsistent naming. Take for example the following perl modules: Perl Module Package Name Software Name -------------------- ------------------- ---------------------- Class::Inspector CSWpmclassinspector pm_classinspector Class::Accessor CSWpmclsaccessor pm_clsaccessor Class::ReturnValue CSWpmclassretval pm_classretval Ideally (IMHO), all the perl modules would just have their full name spelled out in the package / software name (as with Class::Inspector) or at least follow a consistent naming scheme. Are we restricted by external limitations here? Sebastian From phil at bolthole.com Fri May 15 16:45:30 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 07:45:30 -0700 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <4A0D716E.4010700@opencsw.org> References: <4A0D716E.4010700@opencsw.org> Message-ID: <20090515144530.GB84272@bolthole.com> On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: > > Are we restricted by external limitations here? have to draw the line somewhere. we've drawn the line at 20 chars for [mumble] years. it's a reasonable place to draw it, and it doesnt matter much what the CSWxxxx package name is to users, so not having it exactly match the software name, really isnt important. From pfelecan at opencsw.org Fri May 15 17:46:36 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 15 May 2009 17:46:36 +0200 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <20090515144530.GB84272@bolthole.com> (Philip Brown's message of "Fri\, 15 May 2009 07\:45\:30 -0700") References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: Philip Brown writes: > On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: >> >> Are we restricted by external limitations here? > > have to draw the line somewhere. we've drawn the line at 20 chars for > [mumble] years. it's a reasonable place to draw it, and it doesnt matter > much what the CSWxxxx package name is to users, so not having it exactly > match the software name, really isnt important. Sure but it's a PITA for people using "automatic" tools which derive the package name from the software name. BTW, the only restriction is in Sys V pkg where is defined as 256 ASCII characters (long and possibly ugly but aesthetics are subjective by nature). I'm voting for longer than 20 characters (which is of the kind of the 64Kb of the first PC, i.e. "big enough"... and blue) -- Peter From phil at bolthole.com Fri May 15 18:23:01 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:23:01 -0700 Subject: [csw-maintainers] make -j60 package, parallel build possible? build8s 96% idle while building ... In-Reply-To: <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> References: <6af4270905142209t2326ecfar39c99887eb19dbae@mail.gmail.com> <6af4270905142213h3671fc3o176d6735d668384f@mail.gmail.com> Message-ID: <20090515162301.GD10154@bolthole.com> btw, if it hasnt been mentioned yet; if the package is non-gnuified you might try dmake -j something instead of gmake From phil at bolthole.com Fri May 15 18:26:32 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:26:32 -0700 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: <20090515162632.GE10154@bolthole.com> On Fri, May 15, 2009 at 05:46:36PM +0200, Peter FELECAN wrote: > Philip Brown writes: > > have to draw the line somewhere. we've drawn the line at 20 chars for > > [mumble] years. it's a reasonable place to draw it, and it doesnt matter > > much what the CSWxxxx package name is to users, so not having it exactly > > match the software name, really isnt important. > > Sure but it's a PITA for people using "automatic" tools which derive the > package name from the software name. conversely, reallystupidlylongpackagenamesthataddnovalue are a "PITA" for people who dont use infinately wide screens, and want columnar output summarizing things. Other than gar "automatic creation of pkginfo" type stuff, any "automatic tools", should be doing a lookup from the catalog, not making bad assumptions. From skayser at opencsw.org Fri May 15 18:34:19 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 15 May 2009 18:34:19 +0200 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <20090515144530.GB84272@bolthole.com> References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: <4A0D998B.4010807@opencsw.org> Philip Brown wrote: > On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: >> Are we restricted by external limitations here? > > have to draw the line somewhere. we've drawn the line at 20 chars for > [mumble] years. it's a reasonable place to draw it, and it doesnt matter > much what the CSWxxxx package name is to users, so not having it exactly > match the software name, really isnt important. Ok, just so that i get it right (please correct me if am wrong): 1) The software name is the one users should work with and we can/should keep it in a consistent, readable, and "no-surprise" way. This means i can keep pm_datetimetimezone as software name for the DateTime::TimeZone perl module. Shortening of software names like it was done for some Class::xxx packages (pm_clsxxx versus pm_classxxx) should be avoided. 2) The package name is limited (by us) to 20 chars, so that we need to come up with a way to condense package names > 20 chars. pm_datetimetimezone -> CSWpmdatetimetz pm_xmlatomsimplefeed -> CSWpmxmlatomsimplefd Question: Is this self-imposed 20 char limitation worth have discrepancies between software and package names? Does it gain us something? Although i install packages via pkg-get or pkgutil and by their software names, i often find myself dealing with packages by their package names afterwards (pkginfo, pkgrm, /var/sadm/pkg/...). IMHO just a source of possible confusion. Sebastian From phil at bolthole.com Fri May 15 18:38:48 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:38:48 -0700 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: <4A0D998B.4010807@opencsw.org> References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> <4A0D998B.4010807@opencsw.org> Message-ID: <20090515163848.GF10154@bolthole.com> On Fri, May 15, 2009 at 06:34:19PM +0200, Sebastian Kayser wrote: > Ok, just so that i get it right (please correct me if am wrong): > > 1) The software name is the one users should work with and we can/should > keep it in a consistent, readable, and "no-surprise" way. yup. > This means i > can keep pm_datetimetimezone as software name for the DateTime::TimeZone > perl module. Yup. There IS also a limit on software name, for sanity purposes. it's just longer. > 2) The package name is limited (by us) to 20 chars, so that we need to > come up with a way to condense package names > 20 chars. > > pm_datetimetimezone -> CSWpmdatetimetz > pm_xmlatomsimplefeed -> CSWpmxmlatomsimplefd I dont think it is beneficial to try to enforce "one true way of shortening". It depends on the software name. Some collections of softwares obviously > Question: Is this self-imposed 20 char limitation worth have > discrepancies between software and package names? yes. we've had it for 6 years. It's reasonable. There's no decent reason to uproot the world to change this,just because a few software authors are insane. > Although i install packages via pkg-get or pkgutil and by > their software names, i often find myself dealing with packages by their > package names afterwards (pkginfo, pkgrm, /var/sadm/pkg/...). IMHO just > a source of possible confusion. as a side topic: what would make you have to do that sort of thing less? [as a regular sysadmin, not a pkg maintainer? ] From phil at bolthole.com Fri May 15 18:40:10 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 09:40:10 -0700 Subject: [csw-maintainers] php4.4.9 now in testing In-Reply-To: <4A0CC1B5.3070506@opencsw.org> References: <4A0CC1B5.3070506@opencsw.org> Message-ID: <20090515164010.GG10154@bolthole.com> On Thu, May 14, 2009 at 08:13:25PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > this is the last of the PHP4.x releases from upstream. > this has had no upstream support since December 31st 2007. > > If you are still using PHP4, Please consider upgrading to PHP5 > i sugest you post this to the users list. with the url to it, in the message body. From Darin.Perusich at cognigencorp.com Fri May 15 18:42:40 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Fri, 15 May 2009 12:42:40 -0400 Subject: [csw-maintainers] pkgname limit 20 chars? In-Reply-To: References: <4A0D716E.4010700@opencsw.org> <20090515144530.GB84272@bolthole.com> Message-ID: <4A0D9B80.10407@cognigencorp.com> Peter FELECAN wrote: > Philip Brown writes: > >> On Fri, May 15, 2009 at 03:43:10PM +0200, Sebastian Kayser wrote: >>> Are we restricted by external limitations here? >> have to draw the line somewhere. we've drawn the line at 20 chars for >> [mumble] years. it's a reasonable place to draw it, and it doesnt matter >> much what the CSWxxxx package name is to users, so not having it exactly >> match the software name, really isnt important. > > Sure but it's a PITA for people using "automatic" tools which derive the > package name from the software name. BTW, the only restriction is in Sys > V pkg where is defined as 256 ASCII characters (long and possibly ugly > but aesthetics are subjective by nature). I'm voting for longer than 20 > characters (which is of the kind of the 64Kb of the first PC, i.e. "big > enough"... and blue) If the defined limit 256 characters that should be "our" self-defined limit. Additional this should apply for ALL the pkginfo fields which have limits, which I've detailed below from the man page. ARCH = 16 chars CATEGORY = 16 chars NAME = 256 chars PKG = 32 chars VERSION = 256 chars DESC = 256 chars EMAIL = 256 chars HOTLINE = 256 chars -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From mwatters at opencsw.org Fri May 15 18:45:57 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 15 May 2009 11:45:57 -0500 Subject: [csw-maintainers] php4.4.9 now in testing In-Reply-To: <20090515164010.GG10154@bolthole.com> References: <4A0CC1B5.3070506@opencsw.org> <20090515164010.GG10154@bolthole.com> Message-ID: <4A0D9C45.2010302@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Thu, May 14, 2009 at 08:13:25PM -0500, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> this is the last of the PHP4.x releases from upstream. >> this has had no upstream support since December 31st 2007. >> >> If you are still using PHP4, Please consider upgrading to PHP5 >> > > i sugest you post this to the users list. > > with the url to it, in the message body. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I always CC the user list when putting new packages in testing ;) I did not include a link. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoNnEUACgkQLrhmsXMSLxfWiQCfXEXx0wR3XDT1iZUbp23Dd+Sf w/cAn1E1p5pUpZOnTKt1xmL1MfFs8ndk =vwu0 -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri May 15 18:50:08 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 15 May 2009 11:50:08 -0500 Subject: [csw-maintainers] [csw-users] php4.4.9 now in testing In-Reply-To: <4A0CC1B5.3070506@opencsw.org> References: <4A0CC1B5.3070506@opencsw.org> Message-ID: <4A0D9D40.5020402@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > this is the last of the PHP4.x releases from upstream. > this has had no upstream support since December 31st 2007. > > If you are still using PHP4, Please consider upgrading to PHP5 More info on the Releases of PHP can be found at http://www.php.net/releases/ - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoNnUAACgkQLrhmsXMSLxdmqwCg06HHBCtmCT1pF3zx2lEEj/6O FzIAn0hU94US0dcCR9/h2/RufDkLjp3i =pYp8 -----END PGP SIGNATURE----- From phil at bolthole.com Fri May 15 19:09:55 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:09:55 -0700 Subject: [csw-maintainers] [csw-users] php4.4.9 now in testing In-Reply-To: <4A0D9D40.5020402@opencsw.org> References: <4A0CC1B5.3070506@opencsw.org> <4A0D9D40.5020402@opencsw.org> Message-ID: <20090515170955.GH10154@bolthole.com> Arg. Dont crosspost between the lists please. and you didnt include the url to testing. From phil at bolthole.com Fri May 15 19:14:27 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:14:27 -0700 Subject: [csw-maintainers] bug 3034 separate server and user command line binaries In-Reply-To: <4A03B09A.4060103@opencsw.org> References: <4A03B09A.4060103@opencsw.org> Message-ID: <20090515171427.GI10154@bolthole.com> On Thu, May 07, 2009 at 11:10:02PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi, > > I have to recompile subversion to fix an issue with ap2_subversion. > I want to include this bug in the re-package. > > which binaries constitute server vs client? > > Can I get some confirmation on this before I split it up. > I believe it should be as follows: there's also the issue of: 'when people say "install subversion"', are they most likely looking for the server, or are they most likely looking for the client? Personally, I would think they'd most likely be looking for the client. also, I thought the existing package set already had some kind of command-line local-only-stuff package, that skipped all the server stuff? So, that reinforces what I'm saying, that rather than split off a separate "client" package, you should consider splitting off some kind of "core-server" package, that ap2_subversion would then depend on? From phil at bolthole.com Fri May 15 19:17:38 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:17:38 -0700 Subject: [csw-maintainers] problems with existing gtk libs Message-ID: <20090515171738.GJ10154@bolthole.com> there's something up with our "current" gnome libs. My browsers would crash occasionally before. but now they crash multiple times a day, after upgrading to the "latest" stuffs in current. and now, even /usr/sfw/bin/mozilla crashes seamonkey, firefox, and sun mozilla, all crash more frequently now. sigh. multiple times a day :-( From phil at bolthole.com Fri May 15 19:18:55 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 10:18:55 -0700 Subject: [csw-maintainers] problems with existing gtk libs(and related libs) In-Reply-To: <20090515171738.GJ10154@bolthole.com> References: <20090515171738.GJ10154@bolthole.com> Message-ID: <20090515171855.GK10154@bolthole.com> On Fri, May 15, 2009 at 10:17:38AM -0700, Philip Brown wrote: > seamonkey, firefox, and sun mozilla, all crash more frequently now. sigh. > multiple times a day :-( I guess I should give the error. it's always, The program 'Gecko' received an X Window System error. This probably reflects a bug in the program. The error was 'BadDrawable (invalid Pixmap or Window parameter)'. (Details: serial 81745 error_code 9 request_code 14 minor_code 0) (Note to programmers: normally, X errors are reported asynchronously; blahblahblah. This is sol 10 sparc. dual-headed. From Joerg.Schilling at fokus.fraunhofer.de Fri May 15 19:37:44 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 15 May 2009 19:37:44 +0200 Subject: [csw-maintainers] New schily tools in testing Message-ID: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> Hi, I just compiled a new set of packages and put them into testing. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From phil at bolthole.com Fri May 15 20:06:37 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 15 May 2009 11:06:37 -0700 Subject: [csw-maintainers] problems with existing gtk libs Message-ID: <20090515180637.GL10154@bolthole.com> Hmm.. actually... i just noticed that I didnt have a swap file configured. this SHOULDNT be a problem, as i HAVE free ram. but... maybe this is one o them there "stupid design problems", where things die if there isnt disk based "backing store". will report later if adding swap, has improved things. It seems to have fixed one reoccuring url crash, at least. From william at wbonnet.net Fri May 15 21:03:50 2009 From: william at wbonnet.net (William Bonnet) Date: Fri, 15 May 2009 21:03:50 +0200 Subject: [csw-maintainers] problems with existing gtk libs In-Reply-To: <20090515180637.GL10154@bolthole.com> References: <20090515180637.GL10154@bolthole.com> Message-ID: <4A0DBC96.8060101@wbonnet.net> Hi Phil > Hmm.. actually... i just noticed that I didnt have a swap file configured. > > this SHOULDNT be a problem, as i HAVE free ram. > > but... maybe this is one o them there "stupid design problems", where > things die if there isnt disk based "backing store". > > will report later if adding swap, has improved things. > It seems to have fixed one reoccuring url crash, at least I am using FF and TB on a daily basis, and i could even say i have a heavy use of these software, and i experienced no crashes of this kind since update to GLib 2.20 Let us know if adding swap does not solve your problem cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From ihsan at opencsw.org Fri May 15 23:56:36 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Fri, 15 May 2009 23:56:36 +0200 Subject: [csw-maintainers] [csw-users] php4.4.9 now in testing In-Reply-To: <20090515170955.GH10154@bolthole.com> References: <4A0CC1B5.3070506@opencsw.org> <4A0D9D40.5020402@opencsw.org> <20090515170955.GH10154@bolthole.com> Message-ID: <4A0DE514.9010404@opencsw.org> Am 15.5.2009 19:09 Uhr, Philip Brown schrieb: > Dont crosspost between the lists please. It makes actually also my live easier, because I have to review all the post made by non-members. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From rupert at opencsw.org Sat May 16 12:05:43 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 12:05:43 +0200 Subject: [csw-maintainers] git without openssh dependency? Message-ID: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> would it make sense to have git without the openssh dependency? we use a special ssh to accept x509 certs, and installing cswgit then disables this. or does cswopenssh contain any specialities needed for git? rupert. From bwalton at opencsw.org Sat May 16 14:42:47 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 16 May 2009 08:42:47 -0400 Subject: [csw-maintainers] git without openssh dependency? In-Reply-To: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> References: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> Message-ID: <1242477649-sup-4094@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sat May 16 06:05:43 -0400 2009: > would it make sense to have git without the openssh dependency? we use > a special ssh to accept x509 certs, and installing cswgit then > disables this. or does cswopenssh contain any specialities needed for > git? The upcoming 1.6.3 will depend only on the client package. I depended on the full ssh originally since it wasn't split until recently. Git does make use of ssh for a transport tunnel very commonly, so I do need to depend on it...sshd isn't needed. Does that resolve your issues? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From rupert at opencsw.org Sat May 16 14:51:33 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 14:51:33 +0200 Subject: [csw-maintainers] git without openssh dependency? In-Reply-To: <1242477649-sup-4094@ntdws12.chass.utoronto.ca> References: <6af4270905160305g782dc329r51805433035d040a@mail.gmail.com> <1242477649-sup-4094@ntdws12.chass.utoronto.ca> Message-ID: <6af4270905160551s59dd803ah8c25a583d3f6f49b@mail.gmail.com> On Sat, May 16, 2009 at 14:42, Ben Walton wrote: > Excerpts from rupert THURNER's message of Sat May 16 06:05:43 -0400 2009: >> would it make sense to have git without the openssh dependency? we use >> a special ssh to accept x509 certs, and installing cswgit then >> disables this. or does cswopenssh contain any specialities needed for >> git? > > The upcoming 1.6.3 will depend only on the client package. ?I depended > on the full ssh originally since it wasn't split until recently. ?Git > does make use of ssh for a transport tunnel very commonly, so I do > need to depend on it...sshd isn't needed. > > Does that resolve your issues? sure. as long one can without problem use the commonly installed ssh :) many thanks! rupert. From rupert at opencsw.org Sat May 16 15:21:48 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:21:48 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" Message-ID: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> rupert at build8x:~/mgar/pkg $ svn up ... A cpan/Time-HiRes A cpan/Time-HiRes/trunk svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read response body: Secure connection truncated (https://gar.svn.sourceforge.net) can it be that the processor is too fast for the solaris timeout? it happened 5 times while checking out half of the directory. the other half went without problem on build8s. or is this a subversion "bug"? it does not really harm, as the workaround is re-run the "svn up" ... rupert. From rupert at opencsw.org Sat May 16 15:34:01 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:34:01 +0200 Subject: [csw-maintainers] subversion-1.6.2 in testing, but where does the "UNCOMMITTED" come from? Message-ID: <6af4270905160634t293107e4q439d54707df5f8bb@mail.gmail.com> i compiled svn-1.6.2 and put it into testing. for build8s, i am wondering where the "UNCOMMITTED" comes from in the filename? last time i did not commit the change to the makefile and i guessed that gar is intelligent and picks this up :) but this time i did commit it before building. see on http://mirror.opencsw.org/testing.html, e.g. subversion-1.6.2,REV=2009.05.16-SunOS5.8-sparc-UNCOMMITTED.pkg.gz rupert. From rupert at opencsw.org Sat May 16 15:36:40 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:36:40 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> Message-ID: <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> it happens on build8s also now. i guess it is not the client then and we can forget about it. On Sat, May 16, 2009 at 15:21, rupert THURNER wrote: > rupert at build8x:~/mgar/pkg > $ svn up > ... > A ? ?cpan/Time-HiRes > A ? ?cpan/Time-HiRes/trunk > svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read > response body: Secure connection truncated > (https://gar.svn.sourceforge.net) > > can it be that the processor is too fast for the solaris timeout? it > happened 5 times while checking out half of the directory. the other > half went without problem on build8s. or is this a subversion "bug"? > > it does not really harm, as the workaround is re-run the "svn up" ... > > rupert. > From mwatters at opencsw.org Sat May 16 15:41:45 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 08:41:45 -0500 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> Message-ID: <4A0EC299.1000806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > it happens on build8s also now. i guess it is not the client then and > we can forget about it. > > On Sat, May 16, 2009 at 15:21, rupert THURNER wrote: >> rupert at build8x:~/mgar/pkg >> $ svn up >> ... >> A ? ? cpan/Time-HiRes >> A ? ? cpan/Time-HiRes/trunk >> svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read >> response body: Secure connection truncated >> (https://gar.svn.sourceforge.net) >> >> can it be that the processor is too fast for the solaris timeout? it >> happened 5 times while checking out half of the directory. the other >> half went without problem on build8s. or is this a subversion "bug"? >> >> it does not really harm, as the workaround is re-run the "svn up" ... >> >> rupert. >> > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers do a svn status anything that is not External will cause it to be UNCOMMITTED also, you are not overwriting my recipe(s) for svn are you? ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoOwpgACgkQLrhmsXMSLxcxRwCfVOFSbcNFMDUQAPF5MmH10up0 ITEAnR6Hp3AhB/AhKLyL7Cf/rfiAtxM/ =NRUW -----END PGP SIGNATURE----- From rupert at opencsw.org Sat May 16 15:48:54 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 15:48:54 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <4A0EC299.1000806@opencsw.org> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> <4A0EC299.1000806@opencsw.org> Message-ID: <6af4270905160648k718dfd37p33bd993350913f4f@mail.gmail.com> On Sat, May 16, 2009 at 15:41, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > rupert THURNER wrote: >> it happens on build8s also now. i guess it is not the client then and >> we can forget about it. >> >> On Sat, May 16, 2009 at 15:21, rupert THURNER wrote: >>> rupert at build8x:~/mgar/pkg >>> $ svn up >>> ... >>> A ? ?? cpan/Time-HiRes >>> A ? ?? cpan/Time-HiRes/trunk >>> svn: REPORT of '/svnroot/gar/!svn/vcc/default': Could not read >>> response body: Secure connection truncated >>> (https://gar.svn.sourceforge.net) >>> >>> can it be that the processor is too fast for the solaris timeout? it >>> happened 5 times while checking out half of the directory. the other >>> half went without problem on build8s. or is this a subversion "bug"? >>> >>> it does not really harm, as the workaround is re-run the "svn up" ... >>> >>> rupert. >>> >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > do a svn status anything that is not External will cause it to be UNCOMMITTED > > also, you are not overwriting my recipe(s) for svn are you? ;) no way ... just adjusted the version number and everything worked out of the box. impressing design ! From william at wbonnet.net Sat May 16 16:50:21 2009 From: william at wbonnet.net (William Bonnet) Date: Sat, 16 May 2009 16:50:21 +0200 Subject: [csw-maintainers] subversion-1.6.2 in testing, but where does the "UNCOMMITTED" come from? In-Reply-To: <6af4270905160634t293107e4q439d54707df5f8bb@mail.gmail.com> References: <6af4270905160634t293107e4q439d54707df5f8bb@mail.gmail.com> Message-ID: <4A0ED2AD.4050802@wbonnet.net> Hi Rupert > i compiled svn-1.6.2 and put it into testing. for build8s, i am > wondering where the "UNCOMMITTED" comes from in the filename? last > time i did not commit the change to the makefile and i guessed that > gar is intelligent and picks this up :) but this time i did commit it > before building. > You're right, GAR detects that some files are not commited. What is the output of svn status in the trunk dir ? For example here is the output with mdified Makefile and checksum [wbonnet at build8s:~/mgar/pkg/thunderbird/trunk]$ svn status X gar M checksums M Makefile cheers W. From rupert at opencsw.org Sat May 16 20:41:10 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 16 May 2009 20:41:10 +0200 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? Message-ID: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> i tried to create a package for libserf, which should replace libneon in future, as it is able to multithread subversion calls to a server. as there is no libapr, i tried to depend on apache2 by: CONFIGURE_ARGS = $(DIRPATHS) CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 #CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config #CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config (see also: https://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/libserf/trunk/Makefile ) but, for some reason, /opt/csw/bdb44/lib makes it into the compile, and not /opt/csw/apache2/lib. gmake[2]: Entering directory `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' /opt/csw/apache2/share/build/libtool --silent --mode=link /opt/studio/SOS11/SUNWspro/bin/cc -xarch=386 -L/opt/csw/lib -L/opt/csw/bdb44/lib -static -o test/serf_get libserf-0.la test/serf_get.lo /opt/csw/apache2/lib/libaprutil-1.la -lldap -llber -ldb-4.4 -lexpat -liconv -lapr-1 -luuid -lsendfile -lrt -lsocket -lpthread -ldl -lm -lz -lssl -lcrypto ld: fatal: library -lapr-1: not found ld: fatal: File processing errors. No output written to test/serf_get gmake[2]: *** [test/serf_get] Error 1 gmake[2]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' gmake[1]: *** [build-work/build-isa-i386/serf-0.3.0/Makefile] Error 2 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk' gmake: *** [merge-isa-i386] Error 2 what can this be? rupert. From phil at bolthole.com Sat May 16 23:31:49 2009 From: phil at bolthole.com (Philip Brown) Date: Sat, 16 May 2009 14:31:49 -0700 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> Message-ID: <20090516213149.GB26650@bolthole.com> On Sat, May 16, 2009 at 08:41:10PM +0200, rupert THURNER wrote: > i tried to create a package for libserf, which should replace libneon > in future, as it is able to multithread subversion calls to a server. > > as there is no libapr, i tried to depend on apache2 by: > > CONFIGURE_ARGS = $(DIRPATHS) > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 >... hrrrm.. we reaaaally should split out a libapr, if we have things needing it. From mwatters at opencsw.org Sun May 17 03:33:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 20:33:22 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <20090516213149.GB26650@bolthole.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <20090516213149.GB26650@bolthole.com> Message-ID: <4A0F6962.1000000@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Sat, May 16, 2009 at 08:41:10PM +0200, rupert THURNER wrote: >> i tried to create a package for libserf, which should replace libneon >> in future, as it is able to multithread subversion calls to a server. >> >> as there is no libapr, i tried to depend on apache2 by: >> >> CONFIGURE_ARGS = $(DIRPATHS) >> CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 >> ... > > > hrrrm.. we reaaaally should split out a libapr, if we have things needing > it. > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers we already have an apache2rt package, Looking at the contents, it has a few development files included. I filed a bug http://www.opencsw.org/mantis/view.php?id=3676 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoPaWIACgkQLrhmsXMSLxfyEQCg2iN+Jp3YYYRRvuQ6hUMZ84sX mmgAoMwkXqnNhwyklSAjRgwd3aU2nK74 =kHBx -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 03:41:11 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 20:41:11 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> Message-ID: <4A0F6B37.9090703@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > i tried to create a package for libserf, which should replace libneon > in future, as it is able to multithread subversion calls to a server. > > as there is no libapr, i tried to depend on apache2 by: > > CONFIGURE_ARGS = $(DIRPATHS) > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 > CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 > #CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config > #CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config > > (see also: https://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/libserf/trunk/Makefile > ) > > but, for some reason, /opt/csw/bdb44/lib makes it into the compile, > and not /opt/csw/apache2/lib. > > what can this be? libapr is part of the apache build. the --with-apr and --with-apr-util configure args typically point to the apr-config and apu-config files. are you building on the build farm or your own server? To build against apr you need to have the following installed: application CSWapache2-devel apache2_devel - Apache 2.2 development support application CSWapache2rt apache2rt - Apache 2.2 runtime libraries and then you will need to create a dependency on apache2rt. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoPazcACgkQLrhmsXMSLxffMwCgoXG0LKtkJcnU7jOCCe+ya0GB 0XcAnjXh6XguwTp4jEYq+agwtnGYfJHE =BKmv -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 06:55:09 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 16 May 2009 23:55:09 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> Message-ID: <4A0F98AD.7050508@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > > gmake[2]: Entering directory > `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' > /opt/csw/apache2/share/build/libtool --silent --mode=link > /opt/studio/SOS11/SUNWspro/bin/cc -xarch=386 -L/opt/csw/lib > -L/opt/csw/bdb44/lib -static -o test/serf_get libserf-0.la > test/serf_get.lo /opt/csw/apache2/lib/libaprutil-1.la -lldap -llber > -ldb-4.4 -lexpat -liconv -lapr-1 -luuid -lsendfile -lrt -lsocket > -lpthread -ldl -lm -lz -lssl -lcrypto > ld: fatal: library -lapr-1: not found > ld: fatal: File processing errors. No output written to test/serf_get > gmake[2]: *** [test/serf_get] Error 1 > gmake[2]: Leaving directory > `/home/rupert/mgar/pkg/libserf/trunk/work/build-isa-i386/serf-0.3.0' > gmake[1]: *** [build-work/build-isa-i386/serf-0.3.0/Makefile] Error 2 > gmake[1]: Leaving directory `/home/rupert/mgar/pkg/libserf/trunk' > gmake: *** [merge-isa-i386] Error 2 > > what can this be? I tried to build it on build8s and found the problem. ;) there was a bug with the fixlibtool script. and a bug in gar that was forcing it to run for every build with mGar. I have corrected both the mGar bug fixed in r4909 and the fixlibtool bug is fixed in r4910 make the changes below to your recipe and update your version of gar then run make clean && make test # UPSTREAM_MASTER_SITES = DEPENDS = server/apache2 STRIP_LIBTOOL = 1 CONFIGURE_ARGS = $(DIRPATHS) CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config TEST_TARGET = check include gar/category.mk - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoPmK0ACgkQLrhmsXMSLxd6bwCfcOi6GHN/6YtiR4VqoFW0oSlK GagAnjRDWg9s6Pxi+QvXxd6yX64rxgER =Dukp -----END PGP SIGNATURE----- From rupert at opencsw.org Sun May 17 09:31:16 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 17 May 2009 09:31:16 +0200 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <4A0F98AD.7050508@opencsw.org> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> Message-ID: <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> On Sun, May 17, 2009 at 06:55, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi rupert, > rupert THURNER wrote: >> > > make the changes below to your recipe and update your version of gar > then run make clean && make test > > # UPSTREAM_MASTER_SITES = > > DEPENDS = server/apache2 > STRIP_LIBTOOL = 1 > > CONFIGURE_ARGS = $(DIRPATHS) > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config > CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config many thanks that works! what is the reason that you recooment the args above contrary to: CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 ? rupert. From rupert at opencsw.org Sun May 17 09:37:00 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 17 May 2009 09:37:00 +0200 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> Message-ID: <6af4270905170037l6b81534wbdcb66557aee972f@mail.gmail.com> On Sun, May 17, 2009 at 09:31, rupert THURNER wrote: > On Sun, May 17, 2009 at 06:55, Mike Watters wrote: >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Hi rupert, >> rupert THURNER wrote: >>> >> >> make the changes below to your recipe and update your version of gar >> then run make clean && make test >> >> # UPSTREAM_MASTER_SITES = >> >> DEPENDS = server/apache2 >> STRIP_LIBTOOL = 1 what does https://apps.sourceforge.net/trac/gar/changeset/4909 do, resp what did it before that was wrong? rupert. From dam at opencsw.org Sun May 17 15:32:49 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 17 May 2009 15:32:49 +0200 Subject: [csw-maintainers] build8x: subversion client stops with "Could not read response body: Secure connection truncated" In-Reply-To: <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> References: <6af4270905160621gfb5d1dbu129ab2d5c78f2c6d@mail.gmail.com> <6af4270905160636l24e8365ekb13bb852e7764ce0@mail.gmail.com> Message-ID: <5F5B3419-D28C-468D-B281-F1BDB853F931@opencsw.org> Hi Rupert, Am 16.05.2009 um 15:36 schrieb rupert THURNER: > it happens on build8s also now. i guess it is not the client then and > we can forget about it. You are aware that there are maintenance tasks during this time? If any problems happen to SourceForge or Trac, please see Best regards -- Dago From mwatters at opencsw.org Sun May 17 16:44:14 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 09:44:14 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> Message-ID: <4A1022BE.6050501@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > many thanks that works! what is the reason that you recooment the args > above contrary to: > > CONFIGURE_ARGS += --with-apr=$(prefix)/apache2 > CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2 > when you specify the --with-apr=$(prefix)/apache2 the first thing configure does is look in $(prefix)/apache2/bin for apr-config same thing for the util. so it is redundent to put both in the configure args. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQIr0ACgkQLrhmsXMSLxc60wCgpyZfFBh/o6C8pfUlpdAm/mMS w5UAoIdrxvfqemIIjj/5gT71/s1sbcsT =Gy7L -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 17:04:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 10:04:30 -0500 Subject: [csw-maintainers] serf - how to correctly depend on libapr ? In-Reply-To: <6af4270905170037l6b81534wbdcb66557aee972f@mail.gmail.com> References: <6af4270905161141t16480a31u917bacc3a4c2a303@mail.gmail.com> <4A0F98AD.7050508@opencsw.org> <6af4270905170031h52988c42j6d7f5584d6f61595@mail.gmail.com> <6af4270905170037l6b81534wbdcb66557aee972f@mail.gmail.com> Message-ID: <4A10277E.3040309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > what does https://apps.sourceforge.net/trac/gar/changeset/4909 do, > resp what did it before that was wrong? Basically, what was wrong: in r4784 Dago put in a fix to simplify my code in gar.mk to use the new STRIP_LIBTOOL variable. to see the diff run - svn diff -r4720:4784 gar.mk this changed the requirement to have STRIP_LIBTOOL ?= 0 to not fix libtool to removing the variable all together. The only thing r4909 does is remove the variable STRIP_LIBTOOL from gar.conf.mk to see the diff run - svn diff -r4765:4909 gar.conf.mk - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQJ34ACgkQLrhmsXMSLxc+DQCfVBQ38SX2g3Dh6ND4d4vcUZR0 QXwAoOAVJpKLJKTazbTsRbY1rHRVRIlo =5+Zo -----END PGP SIGNATURE----- From dam at opencsw.org Sun May 17 17:33:46 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 17 May 2009 17:33:46 +0200 Subject: [csw-maintainers] Main "Outstanding bugs" Message-ID: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> Hi, this great email notifying of open mails would be even more useful if it would include links to the maintainer bugs page: I find it much easier to look at that page to give me an immediate impression of what needs to be done instead of clicking through all links for individual bugs. Could someone (William?) please add this link? Best regards -- Dago From mwatters at opencsw.org Sun May 17 17:53:11 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 10:53:11 -0500 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> Message-ID: <4A1032E7.9080609@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi, > > this great email notifying of open mails would be even more useful if > it would include links to the maintainer bugs page: > > > I find it much easier to look at that page to give me an immediate > impression of what needs to be done instead of clicking through > all links for individual bugs. > > Could someone (William?) please add this link? as long as we are asking for things, can we get the "resolved" but not "closed" bugs added to the maintainer buglist page http://www.opencsw.org/buglist/maintainer.cgi?maintainer=mwatters once the bug is marked as resolved, I do this when the bug fix goes into testing. I then will close it when the package gets released. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQMucACgkQLrhmsXMSLxcbvACffU24/msEkLpdna3eyZFvZ1Na xC8AoMw8DkEWfbjTV8eRjKc6s4RWovXW =UG56 -----END PGP SIGNATURE----- From william at wbonnet.net Sun May 17 19:08:32 2009 From: william at wbonnet.net (William Bonnet) Date: Sun, 17 May 2009 19:08:32 +0200 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> Message-ID: <4A104490.60302@wbonnet.net> Hi Dago > I find it much easier to look at that page to give me an immediate > impression of what needs to be done instead of clicking through > all links for individual bugs. > > Could someone (William?) please add this link? It's a good idea, but i'm affraid i can't do it. I don't have access to this script. Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Sun May 17 20:01:38 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 13:01:38 -0500 Subject: [csw-maintainers] python 2.5.5 now in testing Message-ID: <4A105102.9080404@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: Update to version 2.5.5 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQUQIACgkQLrhmsXMSLxc+egCgovfDFfpbyEjpoaUp73dL3LJj 3+UAn01fX0+xDKqaTCPiUlT2OGb25Vs2 =tFoU -----END PGP SIGNATURE----- From phil at bolthole.com Sun May 17 20:08:25 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 17 May 2009 11:08:25 -0700 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <4A1032E7.9080609@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> Message-ID: <20090517180825.GA86583@bolthole.com> On Sun, May 17, 2009 at 10:53:11AM -0500, Mike Watters wrote: > can we get the "resolved" but not "closed" bugs > added to the maintainer buglist page > http://www.opencsw.org/buglist/maintainer.cgi?maintainer=mwatters i bugged him to take it OUT, i think :-) From mwatters at opencsw.org Sun May 17 23:00:47 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:00:47 -0500 Subject: [csw-maintainers] TYPO pysqlite2.5.5 now in testing not python In-Reply-To: <4A105102.9080404@opencsw.org> References: <4A105102.9080404@opencsw.org> Message-ID: <4A107AFF.5050005@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Mike Watters wrote: > changes: > Update to version 2.5.5 _______________________________________________ maintainers mailing list maintainers at lists.opencsw.org https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQev8ACgkQLrhmsXMSLxd7AwCeK01ETUySwBGYGBjaOgn5mZ2H UoIAnAit5KMjTDOviOQhokUxPC28HGJk =5CJ1 -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 23:02:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:02:22 -0500 Subject: [csw-maintainers] gd 2.0.35 now in testing Message-ID: <4A107B5E.8080801@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Changes: relinked the amd64 build against the correct fontconfig library - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQe14ACgkQLrhmsXMSLxdanACg4Cgcu27YBlhfFimXaxMOZg3g VXIAoJxxLtMyg6UYmtw28UhGvzgpvtWS =vH2u -----END PGP SIGNATURE----- From mwatters at opencsw.org Sun May 17 23:06:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:06:17 -0500 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <20090517180825.GA86583@bolthole.com> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> <20090517180825.GA86583@bolthole.com> Message-ID: <4A107C49.2020303@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Sun, May 17, 2009 at 10:53:11AM -0500, Mike Watters wrote: >> can we get the "resolved" but not "closed" bugs >> added to the maintainer buglist page >> http://www.opencsw.org/buglist/maintainer.cgi?maintainer=mwatters > > i bugged him to take it OUT, i think :-) > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers could we get a link to a generated second page that contains the resolved bugs? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQfEkACgkQLrhmsXMSLxeSUwCg7EIt3A/YSSS5Yra81Ustk4GX AXkAmgJ9QJ7bvhxS6ydQ3JE8kXZhNwxx =5tHF -----END PGP SIGNATURE----- From dam at opencsw.org Sun May 17 23:27:08 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 17 May 2009 23:27:08 +0200 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <4A107C49.2020303@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> <20090517180825.GA86583@bolthole.com> <4A107C49.2020303@opencsw.org> Message-ID: <5E106507-F8F9-49DD-A74B-0A1AA0148401@opencsw.org> Hi Mike, Am 17.05.2009 um 23:06 schrieb Mike Watters: > could we get a link to a generated second page > that contains the resolved bugs? The page is already there, and by coincidence it is exactly the one I proposed to link on: Best regards -- Dago From mwatters at opencsw.org Sun May 17 23:37:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 16:37:51 -0500 Subject: [csw-maintainers] Main "Outstanding bugs" In-Reply-To: <5E106507-F8F9-49DD-A74B-0A1AA0148401@opencsw.org> References: <737BE8AB-D69A-4BC7-A7DF-4729ECD2E47F@opencsw.org> <4A1032E7.9080609@opencsw.org> <20090517180825.GA86583@bolthole.com> <4A107C49.2020303@opencsw.org> <5E106507-F8F9-49DD-A74B-0A1AA0148401@opencsw.org> Message-ID: <4A1083AF.5080606@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi Mike, > > Am 17.05.2009 um 23:06 schrieb Mike Watters: >> could we get a link to a generated second page >> that contains the resolved bugs? > > The page is already there, and by coincidence it is exactly the > one I proposed to link on: > > > > Best regards > > -- Dago > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers yes but it does not show the resolved bugs, unless there is another "open" bug for the same project. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQg68ACgkQLrhmsXMSLxeGTgCgx4fxdWGe/0a7OrdEo9eLjf86 LycAoIDoduHSot+XofW8HyafnJNmznPj =D+Gr -----END PGP SIGNATURE----- From dam at opencsw.org Mon May 18 00:37:11 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 00:37:11 +0200 Subject: [csw-maintainers] Package "Trac" in testing Message-ID: <5A6F47B5-81E9-45E6-B8E4-9FFD33A4BF7A@opencsw.org> Hi, I noticed that the package name for Trac 0.11.4 is "Trac" instead of the existing "trac". Is this intended? Best regards -- Dago From mwatters at opencsw.org Mon May 18 04:49:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 17 May 2009 21:49:13 -0500 Subject: [csw-maintainers] python 2.6.2 in testing Message-ID: <4A10CCA9.3080608@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: moved the modules to an non-version dependent directory. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoQzKkACgkQLrhmsXMSLxfs7gCfa/JWkoQc1PJgqVjKrXrwzOgi YEYAn0wkjnHjaq/rsr2myArx+G7ZCiY4 =j+Ed -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Mon May 18 10:38:32 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Mon, 18 May 2009 10:38:32 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: Your message of "Mon, 18 May 2009 00:37:11 +0200." <5A6F47B5-81E9-45E6-B8E4-9FFD33A4BF7A@opencsw.org> Message-ID: <200905180838.n4I8cWFo010338@dfki.uni-kl.de> > Hi, > > I noticed that the package name for Trac 0.11.4 is "Trac" instead of the > existing "trac". Is this intended? No, sorry this is a typo - which I will fix today. I'll have to repackage again as pysqlite 2.5.5 is now available. Nicolai From mwatters at opencsw.org Mon May 18 15:33:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 18 May 2009 08:33:35 -0500 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: <200905180838.n4I8cWFo010338@dfki.uni-kl.de> References: <200905180838.n4I8cWFo010338@dfki.uni-kl.de> Message-ID: <4A1163AF.3050006@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Nicolai Schwindt wrote: >> Hi, >> >> I noticed that the package name for Trac 0.11.4 is "Trac" instead of the >> existing "trac". Is this intended? > > No, sorry this is a typo - which I will fix today. > I'll have to repackage again as pysqlite 2.5.5 is now available. > > > > Nicolai > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Hold off on the rebuild against pysqlite 2.5.5 I want to re-compile that again after the python in testing is released. the python in testing moves the module location to the non-version dependent directory. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoRY68ACgkQLrhmsXMSLxfcgwCfaG8gp+T92a/QkKGluo1xeg75 Xz8An2x24/uBWvMzBvZefHY57PhUuqcj =Yy28 -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Mon May 18 16:18:46 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Mon, 18 May 2009 16:18:46 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: Your message of "Mon, 18 May 2009 08:33:35 CDT." <4A1163AF.3050006@opencsw.org> Message-ID: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> [...] > Hold off on the rebuild against pysqlite 2.5.5 I want to > re-compile that again after the python in testing is released. > > the python in testing moves the module location to the non-version dependent > directory. Which brings me into trouble - as I am on the 30 day probation. /me looking for another package .) From dam at opencsw.org Mon May 18 16:31:32 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 16:31:32 +0200 Subject: [csw-maintainers] New members welcome! Message-ID: Hi, the association welcomes the newly accepted members - Mike Watters Mike already maintains several large packages including gcc, python and php. Please keep in mind that being a maintainer is not only about fame and glory, but also about tedious work to make the packages as good as possible and remove bugs timely when discovered. Please check regularly at the bottom of your maintainer page if there are any open issues. If you have spare cycles please adopt an orphaned package and help bring the complete software stack to a 100% current state. But enough of morality: A very warm welcome! Your membership is tracked at Please let me know on what are you working or are planning to work like "webpage", "maintainer", etc. If you have applied for membership and don't see your name anywhere above please let me know. There has been a lot of mail traffic for me to be processed in the past weeks and I cannot guarantee that I didn't missed one. Best regards -- Dago From dam at opencsw.org Mon May 18 16:33:28 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 16:33:28 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> References: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> Message-ID: <9BA99441-F4AD-4A7A-918F-51FA422A9BDA@opencsw.org> Hi Nicolai, Am 18.05.2009 um 16:18 schrieb Nicolai Schwindt: > [...] >> Hold off on the rebuild against pysqlite 2.5.5 I want to >> re-compile that again after the python in testing is released. >> >> the python in testing moves the module location to the non-version >> dependent >> directory. > > Which brings me into trouble - as I am on the 30 day probation. > > /me looking for another package .) Feel free to adpopt any from the "most wanted"-list or from maintainers who have retired: Best regards -- Dago From mwatters at opencsw.org Mon May 18 17:00:59 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 18 May 2009 10:00:59 -0500 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> References: <200905181418.n4IEIkGG017435@dfki.uni-kl.de> Message-ID: <4A11782B.1010400@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Nicolai Schwindt wrote: > [...] >> Hold off on the rebuild against pysqlite 2.5.5 I want to >> re-compile that again after the python in testing is released. >> >> the python in testing moves the module location to the non-version dependent >> directory. > > Which brings me into trouble - as I am on the 30 day probation. > > /me looking for another package .) > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I will get python tested today and try to fast track phil into release so I can get it out quickly for you. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoReCoACgkQLrhmsXMSLxcdSgCgq7GE30l1a4nNECWrOLetjsTO JhAAnRdrFOzo4nZeLWONslIq0ZU5YjNR =EQQ+ -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Mon May 18 17:03:41 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Mon, 18 May 2009 17:03:41 +0200 Subject: [csw-maintainers] Package "Trac" in testing In-Reply-To: Your message of "Mon, 18 May 2009 10:00:59 CDT." <4A11782B.1010400@opencsw.org> Message-ID: <200905181503.n4IF3f0o018606@dfki.uni-kl.de> > > I will get python tested today and try to fast track phil into release so I c an > get it out quickly for you. Its ok, I'll find something else .) From ellson at opencsw.org Mon May 18 18:10:17 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 12:10:17 -0400 Subject: [csw-maintainers] howto: get commit privilege to Subversion? Message-ID: <4A118869.8050008@opencsw.org> I have my first package, gts, building in mgar, and I'm ready to commit. How do I get commit privileges into Subversion for this package? I'm getting: Authentication realm: SourceForge Subversion area Password for 'ellson': John From dam at opencsw.org Mon May 18 18:40:31 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 18:40:31 +0200 Subject: [csw-maintainers] howto: get commit privilege to Subversion? In-Reply-To: <4A118869.8050008@opencsw.org> References: <4A118869.8050008@opencsw.org> Message-ID: <63C907AF-6F2D-4FAC-A8AE-DB064A675BB2@opencsw.org> Hi John, Am 18.05.2009 um 18:10 schrieb John Ellson: > I have my first package, gts, building in mgar, and I'm ready to > commit. > > How do I get commit privileges into Subversion for this package? > > I'm getting: > > Authentication realm: > SourceForge Subversion area > Password for 'ellson': Please send me your SourceForge account name and the "SF Team" (that is Sebastian and me) will grant you commit rights. Best regards -- Dago From ihsan at opencsw.org Mon May 18 21:14:55 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Mon, 18 May 2009 21:14:55 +0200 Subject: [csw-maintainers] OpenCSW Summer Camp 2009 Message-ID: <4A11B3AF.7090709@opencsw.org> Hello, The OpenCSW Summer Camp 2009 will take place in Oslo/Norway. To figure out when it should take place, I've create a Doodle poll. Please fill out the link bellow: http://www.doodle.com/6zr9kmdr8p9vv69p NOTE: If possible, always select a full weekend (Saturday and Sunday). I think it's the best if we will do it as last year. We should meet for socialising on Friday evening, while on Saturday and Sunday we will work. @Trygve: Thanks again for organizing. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ellson at opencsw.org Mon May 18 21:28:28 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 15:28:28 -0400 Subject: [csw-maintainers] Package "gts" in testing Message-ID: <4A11B6DC.6000803@opencsw.org> There is a new package: "gts: GNU Triangulated Surface Library" now available in testing. This is my first package. It was built with mgar, so a review of my mgar/pkg/gts/ tree in svn would also be appreciated. I'm building this package because graphviz can use it, and because it is much smaller than graphviz for my first attempt. Your feedback please. John From dam at opencsw.org Mon May 18 22:45:28 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 18 May 2009 22:45:28 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <4A11B6DC.6000803@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> Message-ID: Hi John, Am 18.05.2009 um 21:28 schrieb John Ellson: > There is a new package: > > "gts: GNU Triangulated Surface Library" > > now available in testing. > > This is my first package. It was built with mgar, so a review of > my mgar/pkg/gts/ tree in svn would also be appreciated. Some suggestions: - Please use for SourceForge MASTER_SITES = $(SF_MIRRORS) For several big sites the URLs have been stored centrally. - For SourceForge the upstream watch works slightly different due to the structure of the download page. Please use UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to your id) UPSTREAM_USE_SF = 1 - You are using a static depend CSWgts.depend and dynamic depends with REQUIRED_PKGS. They both serve the same purpose and defining both will ignore REQUIRED_PKGS. Please remove CSWgts.depend and the notion of it in DISTFILES. - You are using a static gspec file CSWgts.gspec. As the bitname and pkgname are the same as the GARNAME the defaults will be fine. You can safely remove the gspec-file and the notion of it in DISTFILES without change in behaviour. Static gspec files are also old-style and the newer variants from should be used wherever possible. - .la-files are already excluded during merge by default. See for details - The NO_ISAEXEC line is only important when you build multiple ISAs (like the commented out 64 bit ISA with BUILD64=1). When building only one ISA it does nothing. These are a lot of comments. Please take them as result of the, umh, imperfect documentation. Which documentation have you missed most on your path? What can I do to improve your getting started experience with GAR? Best regards -- Dago From mwatters at opencsw.org Mon May 18 23:03:20 2009 From: mwatters at opencsw.org (Mike Watters) Date: Mon, 18 May 2009 16:03:20 -0500 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: References: <4A11B6DC.6000803@opencsw.org> Message-ID: <4A11CD18.2020805@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Dago: Dagobert Michelsen wrote: > What can I do to improve your getting started experience with GAR? provide Round Trip Air Fair and Hotel Accommodation's so we can sit with you and pick your brain for a month. ;) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoRzRgACgkQLrhmsXMSLxcZrgCfSTulprgmXEFPaogET3QPzsAY c34AoKHFkKv+OW9Bewocpep7lPok8UZX =8iCz -----END PGP SIGNATURE----- From ellson at opencsw.org Mon May 18 23:36:40 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 17:36:40 -0400 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: References: <4A11B6DC.6000803@opencsw.org> Message-ID: <4A11D4E8.6080907@opencsw.org> On 05/18/2009 04:45 PM, Dagobert Michelsen wrote: > > - For SourceForge the upstream watch works slightly different due to the > structure of the download page. Please use > UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to > your id) What ID is this? My sourceforge id for me? Or some id for the GTS project? How do i find what it is? John From ellson at opencsw.org Tue May 19 00:46:35 2009 From: ellson at opencsw.org (John Ellson) Date: Mon, 18 May 2009 18:46:35 -0400 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: References: <4A11B6DC.6000803@opencsw.org> Message-ID: <4A11E54B.3030406@opencsw.org> On 05/18/2009 04:45 PM, Dagobert Michelsen wrote: > Hi John, > > Am 18.05.2009 um 21:28 schrieb John Ellson: >> There is a new package: >> >> "gts: GNU Triangulated Surface Library" >> >> now available in testing. >> >> This is my first package. It was built with mgar, so a review of >> my mgar/pkg/gts/ tree in svn would also be appreciated. > > Some suggestions: > - Please use for SourceForge > MASTER_SITES = $(SF_MIRRORS) > For several big sites the URLs have been stored centrally. > - For SourceForge the upstream watch works slightly different due to the > structure of the download page. Please use > UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to > your id) > UPSTREAM_USE_SF = 1 > - You are using a static depend CSWgts.depend and dynamic depends with > REQUIRED_PKGS. They both serve the same purpose and defining both > will ignore REQUIRED_PKGS. Please remove CSWgts.depend and the notion > of it in DISTFILES. > - You are using a static gspec file CSWgts.gspec. As the bitname and > pkgname > are the same as the GARNAME the defaults will be fine. You can safely > remove the gspec-file and the notion of it in DISTFILES without change > in behaviour. Static gspec files are also old-style and the newer > variants from > > should be used wherever possible. > - .la-files are already excluded during merge by default. See > > > for details > - The NO_ISAEXEC line is only important when you build multiple ISAs > (like the commented out 64 bit ISA with BUILD64=1). When building only > one ISA it does nothing. > > These are a lot of comments. Please take them as result of the, umh, > imperfect > documentation. Which documentation have you missed most on your path? > What can I do to improve your getting started experience with GAR? OK, Got those. New gts packages posted to /home/testing. I'll work on some notes about GAR for you.... John From dam at opencsw.org Tue May 19 01:15:59 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 01:15:59 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <4A11D4E8.6080907@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11D4E8.6080907@opencsw.org> Message-ID: <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> Hi John, Am 18.05.2009 um 23:36 schrieb John Ellson: > On 05/18/2009 04:45 PM, Dagobert Michelsen wrote: >> >> - For SourceForge the upstream watch works slightly different due >> to the >> structure of the download page. Please use >> UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change >> to your id) > > What ID is this? My sourceforge id for me? Or some id for the GTS > project? How do i find what it is? First you go to the project main page at and then Download -> Browse All Packages This leads to Aha! -> 1343 AFAIK there is no easier way on SF to do this, William did some research when he implemented upsream watch. Best regards -- Dago From dam at opencsw.org Tue May 19 01:16:36 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 01:16:36 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <4A11CD18.2020805@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11CD18.2020805@opencsw.org> Message-ID: <2900082C-0D62-4043-8CC2-F6742A1EFCFE@opencsw.org> Hi Mike, Am 18.05.2009 um 23:03 schrieb Mike Watters: >> What can I do to improve your getting started experience with GAR? > > provide Round Trip Air Fair and Hotel Accommodation's so we can sit > with you > and pick your brain for a month. ;) Good thing we can just do that during the Summercamp :-) Best regards -- Dago From phil at bolthole.com Tue May 19 01:42:10 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 18 May 2009 16:42:10 -0700 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11D4E8.6080907@opencsw.org> <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> Message-ID: <20090518234210.GE13590@bolthole.com> On Tue, May 19, 2009 at 01:15:59AM +0200, Dagobert Michelsen wrote: > Hi John, >... >>> UPSTREAM_MASTER_SITES = $(SF_PROJECT_SHOWFILE)=66612 (change to > First you go to the project main page at > > and then Download -> Browse All Packages > This leads to > > Aha! -> 1343 AFAIK there is no easier way on SF to do this, > William did some research when he implemented upsream watch. I would like to suggest that there IS a way. If you have the sourceforge name, ie "gts". you then go to the sourceforge-standardized page, as suggested http://sourceforge.net/projects/$PROJNAME You can then use such fanciness as wget --quiet -O - http://sourceforge.net/projects/gts | sed -n 's/.*showfiles.php[?]group_id=\([0-9][0-9]*\).*/\1/p' | head -1 to automatically derive the sf_project_id. or if you prefer curl, use "curl -s" instead of "wget --quiet -O -" :-) From phil at bolthole.com Tue May 19 01:55:29 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 18 May 2009 16:55:29 -0700 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <20090518234210.GE13590@bolthole.com> References: <4A11B6DC.6000803@opencsw.org> <4A11D4E8.6080907@opencsw.org> <99FCDCE1-E100-44D8-8BF2-1A30A967BD5E@opencsw.org> <20090518234210.GE13590@bolthole.com> Message-ID: <20090518235529.GG13590@bolthole.com> To follow up on my own email, and "connect the dots"... i guess the overall goal is to define the UPSTREAM_MASTER_SITES variable in gar spec files. I think it would be nice if, once it was defined somewhere in the spec file that [this is the sourceforge project name] that gar would call some kind of gar utility wrapper script to derive the download url using the "latest" sf.net descrambler :-) > wget --quiet -O - http://sourceforge.net/projects/gts | > sed -n 's/.*showfiles.php[?]group_id=\([0-9][0-9]*\).*/\1/p' | > head -1 > > to automatically derive the sf_project_id. [and/or auto-derive the "showfiles url"] From schwindt at dfki.uni-kl.de Tue May 19 09:38:14 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Tue, 19 May 2009 09:38:14 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing Message-ID: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> Wireshark is now in testing, as this is my first package for CSW any feed back is welcome. If non-privileged users shoulkd be able to use this, one would have to set the s bit to /opt/csw/bin/dumpcap. As this is a security breach I have not done that as a default. Is this ok with the opencsw standards ? Nicolai Schwindt. From ihsan at opencsw.org Tue May 19 11:58:48 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 19 May 2009 11:58:48 +0200 Subject: [csw-maintainers] Package "gts" in testing In-Reply-To: <2900082C-0D62-4043-8CC2-F6742A1EFCFE@opencsw.org> References: <4A11B6DC.6000803@opencsw.org> <4A11CD18.2020805@opencsw.org> <2900082C-0D62-4043-8CC2-F6742A1EFCFE@opencsw.org> Message-ID: <4A1282D8.3080806@opencsw.org> Am 19.5.2009 1:16 Uhr, Dagobert Michelsen schrieb: >>> What can I do to improve your getting started experience with GAR? >> >> provide Round Trip Air Fair and Hotel Accommodation's so we can sit >> with you >> and pick your brain for a month. ;) > > Good thing we can just do that during the Summercamp :-) Please fill out the poll, so we could fix a date. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Tue May 19 12:53:44 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 12:53:44 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> References: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> Message-ID: <161F75CA-3A78-48F3-8B19-1275D352BDD5@opencsw.org> Hi Nicolai, Am 19.05.2009 um 09:38 schrieb Nicolai Schwindt: > Wireshark is now in testing, as this is my first package for CSW any > feed back > is welcome. > > If non-privileged users shoulkd be able to use this, one would have > to set the > s bit to /opt/csw/bin/dumpcap. As this is a security breach I have > not done > that as a default. > > Is this ok with the opencsw standards ? Sounds reasonable. You can describe additional package use on the wiki at Best regards -- Dago From pfelecan at opencsw.org Tue May 19 13:46:08 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Tue, 19 May 2009 13:46:08 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <161F75CA-3A78-48F3-8B19-1275D352BDD5@opencsw.org> (Dagobert Michelsen's message of "Tue\, 19 May 2009 12\:53\:44 +0200") References: <200905190738.n4J7cEK6029745@dfki.uni-kl.de> <161F75CA-3A78-48F3-8B19-1275D352BDD5@opencsw.org> Message-ID: Dagobert Michelsen writes: > Hi Nicolai, > > Am 19.05.2009 um 09:38 schrieb Nicolai Schwindt: >> Wireshark is now in testing, as this is my first package for CSW any >> feed back >> is welcome. >> >> If non-privileged users shoulkd be able to use this, one would have >> to set the >> s bit to /opt/csw/bin/dumpcap. As this is a security breach I have >> not done >> that as a default. >> >> Is this ok with the opencsw standards ? > > Sounds reasonable. You can describe additional package use on the > wiki at Sure, wikis are a nice thing. Even nicer is to provide a README.CSW in the documentation directory where you explain the idiosyncrasies of your package. -- Peter From dam at opencsw.org Tue May 19 13:54:15 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 13:54:15 +0200 Subject: [csw-maintainers] New maintainers welcome! Message-ID: Hi, the association welcomes the newly accepted members - Peter Felecan Peter has been maintaining packages for a long time now including the previous GCC build and is currently busy with graphic and multimedia libraries. Please keep in mind that being a maintainer is not only about fame and glory, but also about tedious work to make the packages as good as possible and remove bugs timely when discovered. Please check regularly at the bottom of your maintainer page if there are any open issues. If you have spare cycles please adopt an orphaned package and help bring the complete software stack to a 100% current state. But enough of morality: A very warm welcome! Your membership is tracked at Please let me know on what are you working or are planning to work like "webpage", "maintainer", etc. If you have applied for membership and don't see your name anywhere above please let me know. There has been a lot of mail traffic for me to be processed in the past weeks and I cannot guarantee that I didn't missed one. Best regards -- Dago From ellson at opencsw.org Tue May 19 16:53:34 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 10:53:34 -0400 Subject: [csw-maintainers] Offer to maintain swig (was: Re: status report: graphviz packaging - new maintainer) In-Reply-To: <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> Message-ID: <4A12C7EE.2000103@opencsw.org> Dagobert Michelsen wrote: > > Swig is also unmaintained at the moment and outdated. Feel free to > take it > over or to ask for somebody else to update it if you need a newer > version. I'll take it. Do I have svn commit privileges for it now? May I commit? I've updated to swig-1.3.39 in my local tree and tried to update the Makefile for latest GAR practices as I understand them, like trying to use the automatic spec and depend mechanisms. "make build" is ok, and "make package" does a lot of work and seems happy, but doesn't leave me with a package in the staging area? Could someone look at the attached Makefile and tell me what I'm doing wrong? John -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: Makefile URL: From mwatters at opencsw.org Tue May 19 17:03:04 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 19 May 2009 10:03:04 -0500 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12C7EE.2000103@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> Message-ID: <4A12CA28.60400@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > spec and depend mechanisms. "make build" is ok, and "make package" > does a lot of work and seems happy, but doesn't leave me with a package in > the staging area? Could someone look at the attached Makefile and tell > me what I'm doing wrong? I don't see a problem with your Makefile ... what does your ~/.garrc file look like? This is mine-> SPKG_PACKAGER = Mike Watters SPKG_EMAIL = mwatters at opencsw.org SPKG_EXPORT = $(HOME)/newpkgs/$(GARNAME) SPKG_SPOOLDIR = $(HOME)/spool/$(GAROSREL)-$(GARCH) GARCHIVEDIR = $(HOME)/src ** SPKG_PACKAGER and SPKG_EMAIL ** needs no explination ** SPKG_EXPORT ** is where gar will put packages when they are done. ** SPKG_SPOOLDIR ** is where gar will "spool" the package before translate to datastream ** GARCHIVEDIR ** is where gar will look for source tarballs before downloading them. helpful if your source tarball is very large. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoSyigACgkQLrhmsXMSLxc6IQCfZb71oUPIKJxQjO8LmBKphl0p Of4AoIEiDRjY0Bl8ra85/1sZaSnnuQRr =1ZTf -----END PGP SIGNATURE----- From ellson at opencsw.org Tue May 19 17:11:12 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 11:11:12 -0400 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12CA28.60400@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> Message-ID: <4A12CC10.7030804@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Ellson wrote: > > >> spec and depend mechanisms. "make build" is ok, and "make package" >> does a lot of work and seems happy, but doesn't leave me with a package in >> the staging area? Could someone look at the attached Makefile and tell >> me what I'm doing wrong? >> > > I don't see a problem with your Makefile ... what does your ~/.garrc file look > like? > This is mine-> > > SPKG_PACKAGER = Mike Watters > SPKG_EMAIL = mwatters at opencsw.org > SPKG_EXPORT = $(HOME)/newpkgs/$(GARNAME) > SPKG_SPOOLDIR = $(HOME)/spool/$(GAROSREL)-$(GARCH) > GARCHIVEDIR = $(HOME)/src > > ** SPKG_PACKAGER and SPKG_EMAIL ** needs no explination > > ** SPKG_EXPORT ** is where gar will put packages when they are done. > ** SPKG_SPOOLDIR ** is where gar will "spool" the package before translate to > datastream > ** GARCHIVEDIR ** is where gar will look for source tarballs before downloading > them. > helpful if your source tarball is very large. > > - -- > > Thanks, > Mike > > > I have this in my .garrc. It works ok for gts packages. John ----- # vim: ft=make # # ~/.garrc - Local settings file for the CSW build system. # # For the most part, any setting found in gar.conf.mk can be overridden by # setting (or unsetting) the variable in this file. The system gar.conf.mk # provides a reasonable set of defaults, but customizations may be required to # build packages outside of the Blastwave build stack. # # Packaging information SPKG_PACKAGER = John Ellson SPKG_EMAIL = ellson at opencsw.org # Where to put finished packages SPKG_EXPORT = /home/ellson/opencsw/staging/build-$(shell date '+%d.%b.%Y') # Proxy settings #http_proxy = #ftp_proxy = From ellson at opencsw.org Tue May 19 17:44:38 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 11:44:38 -0400 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12CA28.60400@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> Message-ID: <4A12D3E6.6080301@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Ellson wrote: > > >> spec and depend mechanisms. "make build" is ok, and "make package" >> does a lot of work and seems happy, but doesn't leave me with a package in >> the staging area? Could someone look at the attached Makefile and tell >> me what I'm doing wrong? >> > > I don't see a problem with your Makefile ... what does your ~/.garrc file look > like? > This is mine-> > > SPKG_PACKAGER = Mike Watters > SPKG_EMAIL = mwatters at opencsw.org > SPKG_EXPORT = $(HOME)/newpkgs/$(GARNAME) > SPKG_SPOOLDIR = $(HOME)/spool/$(GAROSREL)-$(GARCH) > GARCHIVEDIR = $(HOME)/src > > I converted my .garrc to this style, but still no package :( The output from "make package" ends with: ---------------------- Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_wsstream.i Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_wstreambuf.i Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_wstring.i Installing /home/ellson/build.5.8-sparc/opt/csw/share/swig/1.3.39/std/std_carray.swg test -z "1" || (cd CCache && gmake install) gmake[2]: Entering directory `/home/ellson/mgar/pkg/swig/trunk/work/ellson-build8s.d/swig-1.3.39/CCache' /opt/csw/bin/ginstall -c -d /home/ellson/build.5.8-sparc/opt/csw/bin /opt/csw/bin/ginstall -c -m 755 ccache-swig /home/ellson/build.5.8-sparc/opt/csw/bin/`echo ccache-swig | sed 's,x,x,'` /opt/csw/bin/ginstall -c -d /home/ellson/build.5.8-sparc/opt/csw/share/man/man1 /opt/csw/bin/ginstall -c -m 644 ./ccache-swig.1 /home/ellson/build.5.8-sparc/opt/csw/share/man/man1/`echo ccache-swig | sed 's,x,x,'`.1 gmake[2]: Leaving directory `/home/ellson/mgar/pkg/swig/trunk/work/ellson-build8s.d/swig-1.3.39/CCache' Installation complete gmake[1]: Leaving directory `/home/ellson/mgar/pkg/swig/trunk/work/ellson-build8s.d/swig-1.3.39' ==> fixconfig: /home/ellson/build.5.8-sparc/opt/csw/bin Stripping /home/ellson/build.5.8-sparc/opt/csw/bin/ccache-swig ... Done. Stripping /home/ellson/build.5.8-sparc/opt/csw/bin/swig ... Done. [strip] complete for swig. [install] complete for swig. [package] complete for swig. ---------------------- but there is nothing in $HOME/newpkgs/swig/ John From bonivart at opencsw.org Tue May 19 18:09:57 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 19 May 2009 18:09:57 +0200 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12D3E6.6080301@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> Message-ID: <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> On Tue, May 19, 2009 at 5:44 PM, John Ellson wrote: > ? ? ? [strip] complete for swig. > ? ? ? [install] complete for swig. > ? ? ? [package] complete for swig. > > ---------------------- > > but there is nothing in $HOME/newpkgs/swig/ Try "svn pg svn:externals" and check that you use GAR v2, otherwise all new features don't work. It should be default if this is a new package but if it's an older you're updating you probably have v1. -- /peter From ellson at opencsw.org Tue May 19 19:06:44 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 13:06:44 -0400 Subject: [csw-maintainers] misleading message from "gmake newpkg-liblasi" Message-ID: <4A12E724.3000208@opencsw.org> The message about the gspec-file is misleading with current mgar automatic mechanisms. $ make newpkg-liblasi A liblasi A liblasi/tags A liblasi/branches A liblasi/trunk A liblasi/trunk/files A liblasi/trunk/Makefile A liblasi/trunk/checksums property 'svn:ignore' set on 'liblasi/trunk' property 'svn:externals' set on 'liblasi/trunk' Your package is set up for editing at liblasi/trunk Please don't forget to add the gspec-file! From ellson at opencsw.org Tue May 19 18:56:18 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 12:56:18 -0400 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> Message-ID: <4A12E4B2.5070202@opencsw.org> Peter Bonivart wrote: > On Tue, May 19, 2009 at 5:44 PM, John Ellson wrote: > >> [strip] complete for swig. >> [install] complete for swig. >> [package] complete for swig. >> >> ---------------------- >> >> but there is nothing in $HOME/newpkgs/swig/ >> > > Try "svn pg svn:externals" and check that you use GAR v2, otherwise > all new features don't work. It should be default if this is a new > package but if it's an older you're updating you probably have v1. > > That was it. I tried following the wiki to change to v2: svn propset svn:externals "gar https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/gar/v2" . but that wasn't sufficient. I also had to replace the softlink: rm gar ln -s ../../../gar/v2 gar Perhaps this was just because I didn't commit the propset change yet? Anyway, now I have a package: ~/newpkgs/swig/swig-1.3.39,REV=2009.05.19-SunOS5.8-sparc-UNCOMMITTED.pkg.gz Just waiting on an OK to assume this maintainership.... John From dam at opencsw.org Tue May 19 20:26:48 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 20:26:48 +0200 Subject: [csw-maintainers] Offer to maintain swig In-Reply-To: <4A12E4B2.5070202@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> <4A12E4B2.5070202@opencsw.org> Message-ID: Hi John, Am 19.05.2009 um 18:56 schrieb John Ellson: > Anyway, now I have a package: > > ~/newpkgs/swig/swig-1.3.39,REV=2009.05.19-SunOS5.8-sparc- > UNCOMMITTED.pkg.gz > > > Just waiting on an OK to assume this maintainership.... Don't wait, just go ahead :-) Now commit your changed and "gmake rebuild" so the UNCOMMITTED goes away. Best regards -- Dago From dam at opencsw.org Tue May 19 20:28:59 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 19 May 2009 20:28:59 +0200 Subject: [csw-maintainers] misleading message from "gmake newpkg-liblasi" In-Reply-To: <4A12E724.3000208@opencsw.org> References: <4A12E724.3000208@opencsw.org> Message-ID: <73B3895F-EF7E-47D6-A48E-91C933014B49@opencsw.org> Hi John, Am 19.05.2009 um 19:06 schrieb John Ellson: > The message about the gspec-file is misleading with current mgar > automatic mechanisms. > > $ make newpkg-liblasi > A liblasi > A liblasi/tags > A liblasi/branches > A liblasi/trunk > A liblasi/trunk/files > A liblasi/trunk/Makefile > A liblasi/trunk/checksums > property 'svn:ignore' set on 'liblasi/trunk' > property 'svn:externals' set on 'liblasi/trunk' > > Your package is set up for editing at liblasi/trunk > Please don't forget to add the gspec-file! This has been removed in r4958. Thanks for the report! Best regards -- Dago From ellson at opencsw.org Tue May 19 22:31:30 2009 From: ellson at opencsw.org (John Ellson) Date: Tue, 19 May 2009 16:31:30 -0400 Subject: [csw-maintainers] Package "swig" in testing Message-ID: <4A131722.7050101@opencsw.org> There is an updated package: "swig: A scripting interface generator for C/C++ code." now available in testing. This is an update to swig-1.3.39, and also a conversion to mgar v2. This is my second package, so a review of my mgar/pkg/swig/ tree in svn would be appreciated. Your feedback please. John From mwatters at opencsw.org Tue May 19 23:40:45 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 19 May 2009 16:40:45 -0500 Subject: [csw-maintainers] Need some Feedback Message-ID: <4A13275D.4040204@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I am splitting up the python package into separate core, runtime, and development packages. I am playing around with removing the pyo and pyc compiled files from the core package. Here is my dilemma: to do this properly, I need to compile and install the pyc and pyo files as a post-install script. the compile flies by, but the install using installf takes FOREVER (the core package contains roughly 3000 files that will need to be installed in the package database.) should I trade the size for the speed of install? does anyone have a better way to update the package db without using installf? - --- split package sizes without pyo and pyc files --- 209k idle 5M python 634k python_devel 1.3M python_rt 123k python_tk - --- split package sizes with pyo and pyc files --- 593k idle 12M python 1015k python_devel 1.3M python_rt 395k python_tk - --- Current package sizes --- 589k idle 393k python_tk 14M python - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoTJ10ACgkQLrhmsXMSLxc3rQCguPzUvdZsNZMOwDGSlLdncUUv LCMAn21gqG8lkEMgbNuoKsW62gNODtVY =rMCd -----END PGP SIGNATURE----- From phil at bolthole.com Tue May 19 23:50:22 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 19 May 2009 14:50:22 -0700 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <4A13275D.4040204@opencsw.org> References: <4A13275D.4040204@opencsw.org> Message-ID: <20090519215021.GA56895@bolthole.com> On Tue, May 19, 2009 at 04:40:45PM -0500, Mike Watters wrote: > - --- split package sizes without pyo and pyc files --- > > 209k idle > 5M python > > - --- split package sizes with pyo and pyc files --- > > 593k idle > 12M python Wow. That's some serious savings. Here's a crazy thought: if you can configure it so that the .pyc stuffs go to /var somewheres, then you might not actually "need" to declare it with installf. Although you MIGHT want to declare a special script, and declare it as type "i", such as i cswpurge in your prototype file. Then at such time as pkgutil and pkg-get support a "purge" call, they can call it after doing pkgrm, if the user says "yes, 'purge' anything to do with this package when I remove it". From mwatters at opencsw.org Wed May 20 02:57:13 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 19 May 2009 19:57:13 -0500 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <20090519215021.GA56895@bolthole.com> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> Message-ID: <4A135569.3000909@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Phil, Philip Brown wrote: > Although you MIGHT want to declare a special script, and declare it as > type "i", such as > > i cswpurge > > in your prototype file. > > Then at such time as pkgutil and pkg-get support a "purge" call, they can > call it after doing pkgrm, if the user says "yes, 'purge' anything to > do with this package when I remove it". I am not sure I want to do a purge script, but that does give me an Idea. I can do a preremove that does a pkgchk -v on the python package and get the .py files and use that list to remove the pyc pyo files. something like pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}[co] - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoTVWkACgkQLrhmsXMSLxfXFQCgsv46HfdcQXHUgo9qmBavw7FG wQ8AnjXPGY3JX6bHTbgKBBAupz62YDvF =Cg9K -----END PGP SIGNATURE----- From phil at bolthole.com Wed May 20 07:59:23 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 19 May 2009 22:59:23 -0700 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <4A135569.3000909@opencsw.org> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> <4A135569.3000909@opencsw.org> Message-ID: <20090520055923.GA96765@bolthole.com> On Tue, May 19, 2009 at 07:57:13PM -0500, Mike Watters wrote: > I am not sure I want to do a purge script, but that does give me an Idea. > I can do a preremove that does a pkgchk -v on the python package and get the > .py files and use that list to remove the pyc pyo files. > > something like > > pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}[co] pkgchk can be very slow. faster to just hardcode a list of files to check for and remove. (a 'hardcoded' list that could actually be autogenerated from the package's prototype list, at package assembly time) If you're gonna hack, may as well make it an EFFICIENT hack ;-) From phil at bolthole.com Wed May 20 08:01:07 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 19 May 2009 23:01:07 -0700 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <20090520055923.GA96765@bolthole.com> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> <4A135569.3000909@opencsw.org> <20090520055923.GA96765@bolthole.com> Message-ID: <20090520060107.GA99105@bolthole.com> PS: also, if you're gonna do the postinstall hack type stuff... you might also consider making it a "class" based hack, or some other sharable mechanism. You could then provide it as a tool to other python-related package maintainers. From dam at opencsw.org Wed May 20 09:09:04 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 09:09:04 +0200 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <4A131722.7050101@opencsw.org> References: <4A131722.7050101@opencsw.org> Message-ID: Hi John, Am 19.05.2009 um 22:31 schrieb John Ellson: > There is an updated package: > > "swig: A scripting interface generator for C/C++ code." > > now available in testing. > This is an update to swig-1.3.39, and also a conversion to mgar v2. > > This is my second package, so a review of my mgar/pkg/swig/ tree in > svn would be appreciated. > > Your feedback please. Fine package :-) However, there is no need leave in commented code unless you want to re-add it very soon as everything is versioned: One little comment: post-install is now called post-install-modulated, as it should be called for every modulation and DESTDIR is not set properly in post-install. Apart from that: Nice work! Best regards -- Dago From dam at opencsw.org Wed May 20 09:56:25 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 09:56:25 +0200 Subject: [csw-maintainers] Need some Feedback In-Reply-To: <20090520060107.GA99105@bolthole.com> References: <4A13275D.4040204@opencsw.org> <20090519215021.GA56895@bolthole.com> <4A135569.3000909@opencsw.org> <20090520055923.GA96765@bolthole.com> <20090520060107.GA99105@bolthole.com> Message-ID: <46D584D5-4B11-463E-A71B-7BE16D81682B@opencsw.org> Hi, Am 20.05.2009 um 08:01 schrieb Philip Brown: > PS: also, if you're gonna do the postinstall hack type stuff... > you might also consider making it a "class" based hack, or some > other sharable mechanism. > You could then provide it as a tool to other python-related package > maintainers. Yes, that also came to my mind. Make a pythonpyc class which compiles on install and deletes on remove. This can be easily put in an automatic recipe. Best regards -- Dago From schwindt at dfki.uni-kl.de Wed May 20 10:23:36 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Wed, 20 May 2009 10:23:36 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: Your message of "Tue, 19 May 2009 13:46:08 +0200." Message-ID: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> [...] > >> Is this ok with the opencsw standards ? > > > > Sounds reasonable. You can describe additional package use on the > > wiki at > > Sure, wikis are a nice thing. Even nicer is to provide a README.CSW in > the documentation directory where you explain the idiosyncrasies of your > package. Well, my question did not target the documentation - this is documented in wireshark ( and if that gets not read - why should I believe mine is being read ?) I was more concerened about the security implications when installing it setuid. Nicolai. From dam at opencsw.org Wed May 20 10:36:39 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 10:36:39 +0200 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> References: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> Message-ID: <8D167374-D789-49F1-BB47-91251B62B315@opencsw.org> Hi Nicolai, Am 20.05.2009 um 10:23 schrieb Nicolai Schwindt: > [...] >>>> Is this ok with the opencsw standards ? >>> >>> Sounds reasonable. You can describe additional package use on the >>> wiki at >> >> Sure, wikis are a nice thing. Even nicer is to provide a README.CSW >> in >> the documentation directory where you explain the idiosyncrasies of >> your >> package. > > Well, my question did not target the documentation - this is > documented in > wireshark > ( and if that gets not read - why should I believe mine is being > read ?) Because some users may want to install suid while other won't. > I was more concerened about the security implications when > installing it > setuid. Yes. That's what I meant: install without SUID and put a note in README.CSW like If you want to allow regular users to snoop promiscous, please issue chmod u+s /opt/csw/bin/wireshark Because it is that important you may also want to echo this during postinstall. Best regards -- Dago From ellson at opencsw.org Wed May 20 15:05:41 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 20 May 2009 09:05:41 -0400 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: References: <4A131722.7050101@opencsw.org> Message-ID: <4A140025.9060501@opencsw.org> Dagobert Michelsen wrote: > Hi John, > > Am 19.05.2009 um 22:31 schrieb John Ellson: >> There is an updated package: >> >> "swig: A scripting interface generator for C/C++ code." >> >> now available in testing. >> This is an update to swig-1.3.39, and also a conversion to mgar v2. >> >> This is my second package, so a review of my mgar/pkg/swig/ tree in >> svn would be appreciated. >> >> Your feedback please. > > Fine package :-) However, there is no need leave in commented > code unless you want to re-add it very soon as everything is > versioned: > > > Mostly done. Also removed unused files/* > > One little comment: post-install is now called post-install-modulated, > as it > should be called for every modulation and DESTDIR is not set properly in > post-install. I need a little help here. I don't really understand what this was doing in the old swig package (before I commented it out), or if it is still needed? #post-install: install-swig-doc # #install-swig-doc: # @mkdir -p $(DESTDIR)$(docdir)/$(GARNAME) # @cp -R $(WORKSRC)/Doc $(DESTDIR)$(docdir)/$(GARNAME) # @cp -R $(WORKSRC)/Examples $(DESTDIR)$(docdir)/$(GARNAME) # @$(MAKECOOKIE) # $(DONADA) If its needed, could someone provide me with the basic code? Related to this, how do I find out what files are installed by a package? There is no "man pkg-get" and "pkg-get --help" is just about useless! John From bwalton at opencsw.org Wed May 20 15:19:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 09:19:39 -0400 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <4A140025.9060501@opencsw.org> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> Message-ID: <1242825141-sup-7168@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > I need a little help here. I don't really understand what this was > doing in the old swig package (before I commented it out), or if it > is still needed? > > #post-install: install-swig-doc > # > #install-swig-doc: > # @mkdir -p $(DESTDIR)$(docdir)/$(GARNAME) > # @cp -R $(WORKSRC)/Doc $(DESTDIR)$(docdir)/$(GARNAME) > # @cp -R $(WORKSRC)/Examples $(DESTDIR)$(docdir)/$(GARNAME) > # @$(MAKECOOKIE) > # $(DONADA) > > If its needed, could someone provide me with the basic code? I can't say if it's needed or not. It may be if the `gmake install` for the package doesn't install docs by default. You have a few possible options for this: 1. If there is an alternate Make target built-in to the swig Makefile (install-all, install-doc, etc), you could alter the GAR variable INSTALL_ARGS to the required target. This would be the easiest option if it's supported by the package. 2. If you can't do 1, you'll likely only need to change the line: post-install: ... to post-install-modulated: ... to have that custom code run once per modulation. > Related to this, how do I find out what files are installed by a > package? The GAR view? After the merge step, you can look in .../work/pkgroot/. Before the merge, but after install, you could look in .../work/install-$(MODULATION)/. If the package is installed, you can use the solaris pkg* commands to query for files owned by the package. HTH -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From skayser at opencsw.org Wed May 20 15:31:32 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 20 May 2009 15:31:32 +0200 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <1242825141-sup-7168@ntdws12.chass.utoronto.ca> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> <1242825141-sup-7168@ntdws12.chass.utoronto.ca> Message-ID: <4A140634.4010100@opencsw.org> Ben Walton wrote: > Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > >> I need a little help here. I don't really understand what this was >> doing in the old swig package (before I commented it out), or if it >> is still needed? >> >> #post-install: install-swig-doc >> # >> #install-swig-doc: >> # @mkdir -p $(DESTDIR)$(docdir)/$(GARNAME) >> # @cp -R $(WORKSRC)/Doc $(DESTDIR)$(docdir)/$(GARNAME) >> # @cp -R $(WORKSRC)/Examples $(DESTDIR)$(docdir)/$(GARNAME) >> # @$(MAKECOOKIE) >> # $(DONADA) >> >> If its needed, could someone provide me with the basic code? > > I can't say if it's needed or not. It may be if the `gmake install` > for the package doesn't install docs by default. Btw. (good that this has been brought up), i noticed that we have quite a few packages that use a post-install-modulated part like the above to install additional docs. Could we maybe get something like EXTRA_INCLUDE_DOCS = in GAR that would do this instead? Sebastian From dam at opencsw.org Wed May 20 15:48:46 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 20 May 2009 15:48:46 +0200 Subject: [csw-maintainers] Update to the buildfarm In-Reply-To: <1242824555-sup-2752@ntdws12.chass.utoronto.ca> References: <1242824387-sup-8557@ntdws12.chass.utoronto.ca> <1242824555-sup-2752@ntdws12.chass.utoronto.ca> Message-ID: Hi, Am 20.05.2009 um 15:03 schrieb Ben Walton: > Done, except for build10x. Dago, that box is asking for a password. > Has the NFS mount failed so it can't see my ssh keys? I setup DNS for the farm and this was a NFSv4 issue. I switched back to NFSv3 in the meantime. The other hosts shouldn't be affected as they either use lofs or don't speak NFSv4. Additionally, I distributed an updated global ssh_known_hosts on all build machines so ssh'ing between all machines without password should now be possible. The reason for the change is an update to GAR which allows full parallel builds (if you are curious you can take a look at gar/v2-pbuild). For the moment multitail is missing on the farm to watch the different builds, but that should be resolved when Sebastian releases it to current/. Best regards -- Dago From bwalton at opencsw.org Wed May 20 18:05:39 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 12:05:39 -0400 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request Message-ID: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> Hi Phil and Peter, Would you consider adding a -C option to the package management utilities to at would limit the compare output to only versions behind the catalog (eg: `pkg-get -U -c | grep -v SAME` is what I use to see what will be updated). It's just a creature comfort kind of option. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From bwalton at opencsw.org Wed May 20 18:08:51 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 12:08:51 -0400 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <4A140025.9060501@opencsw.org> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> Message-ID: <1242835669-sup-7223@ntdws12.chass.utoronto.ca> Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > There is no "man pkg-get" and "pkg-get --help" is just about > useless! I forgot to comment on this point earlier: There is a `man pkg-get`, but you need to have /opt/csw/share/man in your $MANPATH first. HTH. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 20 18:44:00 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 09:44:00 -0700 Subject: [csw-maintainers] Package "swig" in testing In-Reply-To: <1242835669-sup-7223@ntdws12.chass.utoronto.ca> References: <4A131722.7050101@opencsw.org> <4A140025.9060501@opencsw.org> <1242835669-sup-7223@ntdws12.chass.utoronto.ca> Message-ID: <20090520164400.GA87831@bolthole.com> On Wed, May 20, 2009 at 12:08:51PM -0400, Ben Walton wrote: > Excerpts from John Ellson's message of Wed May 20 09:05:41 -0400 2009: > > > There is no "man pkg-get" and "pkg-get --help" is just about > > useless! > > I forgot to comment on this point earlier: There is a `man pkg-get`, > but you need to have /opt/csw/share/man in your $MANPATH first. but besides which, that's the wrong place to look for his actual question anyways. pkg-get doesnt technicall install any packages, or konw anything about location of files. pkg-get calls the system level tool pkgadd. documentaion on that is also semi useless to anser the question of "what files are in a package", though :) for one of OUR packages, easiest thing is to use http://www.opencsw.org/search for other installed packages, grep through /var/sadm/install/contents From phil at bolthole.com Wed May 20 18:46:22 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 09:46:22 -0700 Subject: [csw-maintainers] wireshark 1.0.7 in testing In-Reply-To: <8D167374-D789-49F1-BB47-91251B62B315@opencsw.org> References: <200905200823.n4K8NbGA022695@dfki.uni-kl.de> <8D167374-D789-49F1-BB47-91251B62B315@opencsw.org> Message-ID: <20090520164622.GB87831@bolthole.com> On Wed, May 20, 2009 at 10:36:39AM +0200, Dagobert Michelsen wrote: > > If you want to allow regular users to snoop promiscous, please issue > chmod u+s /opt/csw/bin/wireshark > > Because it is that important you may also want to echo this during > postinstall. I would disagree. [and personally think that installing it setuid is insane] additionally, wireshark shouldnt need any kind of postinstall. adding one to do this, is rather overkill. From phil at bolthole.com Wed May 20 19:06:21 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 10:06:21 -0700 Subject: [csw-maintainers] debian's python policy Message-ID: <20090520170620.GD87831@bolthole.com> I just discovered that debian has an explicit, separate "python policy". Thought I'd reference it here. http://www.debian.org/doc/packaging-manuals/python-policy/ As usual.. while we are not bound to follow everything debian does... we would be fools not to learn from their amassed experience on the subject. And coincidently, their policy on .pyc files can be found at http://www.debian.org/doc/packaging-manuals/python-policy/ch-module_packages.html#s-bytecompilation "If a package provides any binary-independent modules (foo.py files), the corresponding bytecompiled modules (foo.pyc files) and optimized modules (foo.pyo files) must not ship in the package. Instead, they should be generated in the package's postinst, and removed in the package's prerm. " Also of interest: " The file /etc/python/debian_config allows configuration how modules should be byte-compiled. The postinst scripts should respect these settings." So, us having a class handle this kind of thing, is strongly indicated as a positive thing for future enhancements such as that. From Joerg.Schilling at fokus.fraunhofer.de Wed May 20 19:26:11 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 20 May 2009 19:26:11 +0200 Subject: [csw-maintainers] debian's python policy In-Reply-To: <20090520170620.GD87831@bolthole.com> References: <20090520170620.GD87831@bolthole.com> Message-ID: <4a143d33.X7zRJnvpgxZ0OJ6p%Joerg.Schilling@fokus.fraunhofer.de> Philip Brown wrote: > I just discovered that debian has an explicit, separate > "python policy". > Thought I'd reference it here. > > http://www.debian.org/doc/packaging-manuals/python-policy/ > > As usual.. while we are not bound to follow everything debian does... we > would be fools not to learn from their amassed experience on the subject. > > And coincidently, their policy on .pyc files can be found at > http://www.debian.org/doc/packaging-manuals/python-policy/ch-module_packages.html#s-bytecompilation > > "If a package provides any binary-independent modules (foo.py files), the > corresponding bytecompiled modules (foo.pyc files) and optimized modules > (foo.pyo files) must not ship in the package. Instead, they should be > generated in the package's postinst, and removed in the package's prerm. " What is the reason for such a decision? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From phil at bolthole.com Wed May 20 19:34:59 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 10:34:59 -0700 Subject: [csw-maintainers] debian's python policy In-Reply-To: <4a143d33.X7zRJnvpgxZ0OJ6p%Joerg.Schilling@fokus.fraunhofer.de> References: <20090520170620.GD87831@bolthole.com> <4a143d33.X7zRJnvpgxZ0OJ6p%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <20090520173459.GJ87831@bolthole.com> On Wed, May 20, 2009 at 07:26:11PM +0200, Joerg Schilling wrote: > > "If a package provides any binary-independent modules (foo.py files), the > > corresponding bytecompiled modules (foo.pyc files) and optimized modules > > (foo.pyo files) must not ship in the package. Instead, they should be > > generated in the package's postinst, and removed in the package's prerm. " > > What is the reason for such a decision? because it saves a whoooole lot of space/download time. the python package itself, would be reduced to 5MB from 13, I think Mike said? From ellson at opencsw.org Wed May 20 20:34:10 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 20 May 2009 14:34:10 -0400 Subject: [csw-maintainers] packages "gts" and "swig" ready for public release Message-ID: <4A144D22.6030904@opencsw.org> Philip, I have addressed all coments, and so I believe that the "gts" and "swig" packages are now ready for public release. www.opencsw.org:/home/newpkgs/gts-0.7.6,REV=2009.05.20-SunOS5.8-i386-CSW.pkg.gz www.opencsw.org:/home/newpkgs/gts-0.7.6,REV=2009.05.20-SunOS5.8-sparc-CSW.pkg.gz www.opencsw.org:/home/newpkgs/swig-1.3.39,REV=2009.05.20-SunOS5.8-i386-CSW.pkg.gz www.opencsw.org:/home/newpkgs/swig-1.3.39,REV=2009.05.20-SunOS5.8-sparc-CSW.pkg.gz John From mwatters at opencsw.org Wed May 20 20:44:44 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 20 May 2009 13:44:44 -0500 Subject: [csw-maintainers] debian's python policy In-Reply-To: <20090520170620.GD87831@bolthole.com> References: <20090520170620.GD87831@bolthole.com> Message-ID: <4A144F9C.4000300@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > I just discovered that debian has an explicit, separate > "python policy". > Thought I'd reference it here. > > http://www.debian.org/doc/packaging-manuals/python-policy/ > > As usual.. while we are not bound to follow everything debian does... we > would be fools not to learn from their amassed experience on the subject. > > And coincidently, their policy on .pyc files can be found at > http://www.debian.org/doc/packaging-manuals/python-policy/ch-module_packages.html#s-bytecompilation > > "If a package provides any binary-independent modules (foo.py files), the > corresponding bytecompiled modules (foo.pyc files) and optimized modules > (foo.pyo files) must not ship in the package. Instead, they should be > generated in the package's postinst, and removed in the package's prerm. " This is exactly how it is built now. > > > Also of interest: > > " The file /etc/python/debian_config allows configuration how modules should > be byte-compiled. The postinst scripts should respect these settings." > > So, us having a class handle this kind of thing, is strongly indicated as a > positive thing for future enhancements such as that. I will work on the class scripts for this and release it that way. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoUT5wACgkQLrhmsXMSLxf9PACfaL6aWlyMTYEkAMyps/C7ThGQ LoAAnjuhun1nkUsYpCQ7QScSUJJulq26 =jn88 -----END PGP SIGNATURE----- From skayser at opencsw.org Wed May 20 21:44:28 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Wed, 20 May 2009 21:44:28 +0200 Subject: [csw-maintainers] Another GAR RFE, deprecate gar/v1 (was: Re: Offer to maintain swig) In-Reply-To: <4A12E4B2.5070202@opencsw.org> References: <4A0B3732.8010808@opencsw.org> <5A1F76E6-AA11-49B4-AB1D-113ADCBF8900@opencsw.org> <4A12C7EE.2000103@opencsw.org> <4A12CA28.60400@opencsw.org> <4A12D3E6.6080301@opencsw.org> <625385e30905190909n1f5f6e7cl34073b610242c00@mail.gmail.com> <4A12E4B2.5070202@opencsw.org> Message-ID: <4A145D9C.7080205@opencsw.org> John Ellson wrote: > Peter Bonivart wrote: >> On Tue, May 19, 2009 at 5:44 PM, John Ellson wrote: >> >>> [strip] complete for swig. >>> [install] complete for swig. >>> [package] complete for swig. >>> >>> ---------------------- >>> >>> but there is nothing in $HOME/newpkgs/swig/ >>> >> Try "svn pg svn:externals" and check that you use GAR v2, otherwise >> all new features don't work. It should be default if this is a new >> package but if it's an older you're updating you probably have v1. >> >> > That was it. As we try to trim our documentation towards mGAR v2 (and forget about the whole different iterations of GAR eventually), could we maybe add a deprecation warning to the gar/v1 Makefile. Something along the lines of "Unless you know what you are doing, please consider switching to the currently maintained version of GAR, by doing ....". Maybe not as harsh, but gently pointing the way :) Sebastian From bonivart at opencsw.org Wed May 20 22:46:23 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 20 May 2009 22:46:23 +0200 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request In-Reply-To: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> References: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> Message-ID: <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> On Wed, May 20, 2009 at 6:05 PM, Ben Walton wrote: > Would you consider adding a -C option to the package management > utilities to at would limit the compare output to only versions behind > the catalog (eg: `pkg-get -U -c | grep -v SAME` is what I use to see > what will be updated). Sure, I can add it to 1.6 which is almost done. I never implemented showing not installed packages but I still get annoyed by using grep -v SAME so it's a good idea. -- /peter From mwatters at opencsw.org Thu May 21 02:46:30 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 20 May 2009 19:46:30 -0500 Subject: [csw-maintainers] Packages that require either package A or package B Message-ID: <4A14A466.5010602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 There are some packages that require either pkgA or pkgB to be functional. mediawiki for example. it requires Either MySQL OR Postgres. How can we handle this in a package that allows a seamless install. I figure a checkinstall script to determine if one or the other exists and bail gracefully "like a normal requirement not met" if missing both. thoughts? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoUpGUACgkQLrhmsXMSLxcJ2QCfWF6OZg1/p10c5tuir8oNwaFY P+cAn3Okd64mgz+H/7RKaiaFbNnwyvkk =pJXK -----END PGP SIGNATURE----- From phil at bolthole.com Thu May 21 02:53:27 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 20 May 2009 17:53:27 -0700 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <4A14A466.5010602@opencsw.org> References: <4A14A466.5010602@opencsw.org> Message-ID: <20090521005327.GA15461@bolthole.com> On Wed, May 20, 2009 at 07:46:30PM -0500, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > There are some packages that require either pkgA or pkgB to be functional. > > mediawiki for example. it requires Either MySQL OR Postgres. > > How can we handle this in a package that allows a seamless install. > > I figure a checkinstall script to determine if one or the other exists and bail > gracefully "like a normal requirement not met" if missing both. and what if you encounter both installed? In this sort of situation, i think you have a basic choice of either a) just pick one that you the maintainer think is best b) do not auto-install, but provide utility scripts to auto-setup for either of them, and a README.CSW file that indicates, "run the script you prefer, to finish setup". (and of course, some sort of auto-detection for, "this is an UPGRADE, so go with whatever was previously chosen") From bwalton at opencsw.org Thu May 21 03:16:34 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 20 May 2009 21:16:34 -0400 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <20090521005327.GA15461@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> Message-ID: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Wed May 20 20:53:27 -0400 2009: > a) just pick one that you the maintainer think is best > > b) do not auto-install, but provide utility scripts to auto-setup for > either of them, and a README.CSW file that indicates, > "run the script you prefer, to finish setup". > > (and of course, some sort of auto-detection for, > "this is an UPGRADE, so go with whatever was previously chosen") c) make pkgutil/pkg-get smart enough to handle debian-style dependencies above and beyond what pkgadd can actually handle. Much more work, but a better experience in the end. I suspect that this evolution will need to happen anyway in the face of IPS, which won't (doesn't? I haven't played with it) limit itself to the current rigid dependency system. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From ellson at opencsw.org Thu May 21 14:24:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 21 May 2009 08:24:06 -0400 Subject: [csw-maintainers] Howto package applications that use cmake? Message-ID: <4A1547E6.7050704@opencsw.org> I'd like to package libLASi, mostly because graphviz can use it: http://www.unifont.org/lasi/ I have a couple of problems because it uses cmake. 1. CSWcmake isn't installed on the build hosts 2. How to convince GAR to use cmake? Can anyone point me to another package that uses it? John From mwatters at opencsw.org Thu May 21 15:55:32 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 08:55:32 -0500 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A1547E6.7050704@opencsw.org> References: <4A1547E6.7050704@opencsw.org> Message-ID: <4A155D54.20806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > I'd like to package libLASi, mostly because graphviz can use it: > > http://www.unifont.org/lasi/ > > I have a couple of problems because it uses cmake. > > 1. CSWcmake isn't installed on the build hosts Send a request to buildfarm at lists.opencsw.org requesting the install of cmake on the build farm. > > 2. How to convince GAR to use cmake? Can anyone point me to another > package that uses it? GAR should use whatever configure tells it to use for the actual compile. it will only use gmake for the commands to build the installation environment. > > John > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVXVQACgkQLrhmsXMSLxfEZQCff9GfGeq7WOxP0XyMyoV9Phnd i1EAnR0XUjgW0YfsRws9RXAM97vs0ANz =tGp+ -----END PGP SIGNATURE----- From ellson at opencsw.org Thu May 21 16:10:01 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 21 May 2009 10:10:01 -0400 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A155D54.20806@opencsw.org> References: <4A1547E6.7050704@opencsw.org> <4A155D54.20806@opencsw.org> Message-ID: <4A1560B9.3070201@opencsw.org> Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > John Ellson wrote: > >> I'd like to package libLASi, mostly because graphviz can use it: >> >> http://www.unifont.org/lasi/ >> >> I have a couple of problems because it uses cmake. >> >> 1. CSWcmake isn't installed on the build hosts >> > Send a request to buildfarm at lists.opencsw.org requesting the install of cmake > on the build farm. > Will do. > >> 2. How to convince GAR to use cmake? Can anyone point me to another >> package that uses it? >> > > GAR should use whatever configure tells it to use for the actual compile. > it will only use gmake for the commands to build the installation environment. > Thats the problem. There is no "configure". I guess I could patch one in, but it would be better if I could just run "cmake" per the libLASi build instructions: http://www.unifont.org/lasi/#cmake John From mwatters at opencsw.org Thu May 21 16:15:34 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 09:15:34 -0500 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A1560B9.3070201@opencsw.org> References: <4A1547E6.7050704@opencsw.org> <4A155D54.20806@opencsw.org> <4A1560B9.3070201@opencsw.org> Message-ID: <4A156206.3050807@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Ellson wrote: > > Thats the problem. There is no "configure". I guess I could patch one > in, but it > would be better if I could just run "cmake" per the libLASi build > instructions: > > http://www.unifont.org/lasi/#cmake OK, Sorry I miss read the post. thanks for the clarification. you need to use a custom configure script. if you set the following variable CONFIGURE_SCRIPTS = xxxx then create your configure script configure-xxxx: echo "my config script" ... ... ... @$(MAKECOOKIE) spamassassin/trunk/Makefile has some good examples - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVYgYACgkQLrhmsXMSLxcXRACgoNr73LX/9dYWas2N4FLasJD5 G5MAoM+UGySoKVat2YbBwm3Rphs8Hh7H =f/SQ -----END PGP SIGNATURE----- From phil at bolthole.com Thu May 21 18:08:33 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 09:08:33 -0700 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> Message-ID: <20090521160833.GA89074@bolthole.com> On Wed, May 20, 2009 at 09:16:34PM -0400, Ben Walton wrote: > I suspect that this > evolution will need to happen anyway in the face of IPS, which won't > limit itself to the current rigid > dependency system. funnily enough, there was some semi-related discussion on the ips dev list. And they're running into problems, because they decided to be "more flexible". So I dont think that just copying what IPS does, is a good general strategic move for us. Sometimes, offering more choices, actualy ends up being the wrong thing to do. particularly in a tool designed to simplify and streamline installs. The purpose (for pkg-get, at least), is NOT to offer *everything that everyone could possibly want to do*. It is to simplify installation, for the most common straightforward uses. If people want to do really really fancy non-standard stuff... they should be doing it themselves. Similar to the whole concept of offering binary packages in the first place. We try to offer a set of binary packages, choosing compile flags for the most common good. But if people want to get that 1% extra performance by using one-cpu-specific flags.... they should compile it themselves. That being said... there is a more general issue here about "what do we do when a package wants 'a database' installed?" I shall offer a separate email on that topic, so more people will pay attention :-) From phil at bolthole.com Thu May 21 18:13:07 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 09:13:07 -0700 Subject: [csw-maintainers] "package needs a database" general issue Message-ID: <20090521161307.GB89074@bolthole.com> Hi folks, There is an issue that comes up occasionally, where we want to offer an 'easy install' option for people for a particular package, but that package requires a database to run. Then there is a matter of; WHICH database should we use, for fully automated install?!! usually it comes down to a choice of "postgres or mysql?" I would like to propose that we need to pick an official default database to use. Furthermore, I propose that that database be mysql. It's the simplest to administer, it has the biggest mindshare, and there are lots of apps that will ONLY support mysql, sadly. HOWEVER, I always like to support end-user choice whenever possible. So, in addition to a default assumption of mysql... I would also like to propose a setting in csw.conf. Something like default_dbtype where values can be mysql, postgres, or even oracle. If that value is set, then a package can presume that that database is already installed and running. In those cases though, we also need to agree on some kind of common mechanism to insert values into said database. Comments? From skayser at opencsw.org Thu May 21 18:35:28 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 21 May 2009 18:35:28 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? Message-ID: <4A1582D0.8040709@opencsw.org> Hi, i packaged mailx that was formerly known as nail (CSWnail). Now how would we go about obsoleting CSWnail? Just drop it from the catalog, let CSWnail become an empty package that pulls in CSWmailx and somehow deletes CSWnail, or is there another preferred (proven) way of doing things like this? Sebastian From dam at opencsw.org Thu May 21 18:50:17 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 18:50:17 +0200 Subject: [csw-maintainers] Virtualizing Solaris 8/9 x86 Message-ID: <0DC8DB57-B510-469E-B794-75FCA361AAEF@opencsw.org> Hi, good news on virtualization of Solaris 8 and 9 x86: the new VMware vSphere 4.0 (formerly known as ESX Server) is going to support all versions of Solaris 8 and 9 x86: > Solaris 9, Update 1, Update 2, Update 3, Update 4, Update 5, Update 6, > Update 7, and Update 8, 32-bit. (experimental) See ?Solaris 9 > Operating > System x86 Platform Edition? on page 406. > Solaris 8, 06/00, 10/00, 01/01, 04/01, 07/01, 10/01, and 02/02, 32-bit > (experimental). See ?Solaris 8 Operating System x86 Platform > Edition? on > page 409. This will allow us to make good use of faster hardware and testing servers on demand :-) Best regards -- Dago From dam at opencsw.org Thu May 21 18:58:27 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 18:58:27 +0200 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <20090521160833.GA89074@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> Message-ID: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> Hi, Am 21.05.2009 um 18:08 schrieb Philip Brown: > On Wed, May 20, 2009 at 09:16:34PM -0400, Ben Walton wrote: >> I suspect that this >> evolution will need to happen anyway in the face of IPS, which won't >> limit itself to the current rigid >> dependency system. > > funnily enough, there was some semi-related discussion on the ips > dev list. > And they're running into problems, because they decided to be "more > flexible". So I dont think that just copying what IPS does, is a good > general strategic move for us. Additionally, adding more complexity to System V packaging will bind valuable resources and would make packaging quite more complex. Apart from that the current mechanism has served us well over the past 10+ years and will fade away as the new packaging system is already there (IPS). I would favor to live with the faults and handle dependencies as we have always handled them. In addition, I propose to carefully start working on real IPS packaging instead of putting extra work/complexity in System V packaging. A real IPS backend for GAR with all bells and whistels for dual-personality building of packages is surely a Good Thing. Best regards -- Dago From dam at opencsw.org Thu May 21 18:59:54 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 18:59:54 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A1582D0.8040709@opencsw.org> References: <4A1582D0.8040709@opencsw.org> Message-ID: Hi Sebastian, Am 21.05.2009 um 18:35 schrieb Sebastian Kayser: > i packaged mailx that was formerly known as nail (CSWnail). Now how > would we go about obsoleting CSWnail? Just drop it from the > catalog, let > CSWnail become an empty package that pulls in CSWmailx and somehow > deletes CSWnail, or is there another preferred (proven) way of doing > things like this? I vaguely recall pkg-get has some magic included that does this kind of thing on making new packages incompatible ("I") with prior ones. Phil? Best regards -- Dago From phil at bolthole.com Thu May 21 19:03:34 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 10:03:34 -0700 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A1582D0.8040709@opencsw.org> References: <4A1582D0.8040709@opencsw.org> Message-ID: <20090521170334.GH89074@bolthole.com> On Thu, May 21, 2009 at 06:35:28PM +0200, Sebastian Kayser wrote: > Hi, > > i packaged mailx that was formerly known as nail (CSWnail). Now how > would we go about obsoleting CSWnail? Just drop it from the catalog, let > CSWnail become an empty package that pulls in CSWmailx and somehow > deletes CSWnail, or is there another preferred (proven) way of doing > things like this? ugh. is the binary really called "mailx"? that would be very lame. I think it might actually be nice to keep it as called "nail" :-} From dam at opencsw.org Thu May 21 19:05:05 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 19:05:05 +0200 Subject: [csw-maintainers] Howto package applications that use cmake? In-Reply-To: <4A1560B9.3070201@opencsw.org> References: <4A1547E6.7050704@opencsw.org> <4A155D54.20806@opencsw.org> <4A1560B9.3070201@opencsw.org> Message-ID: <88803155-5C85-4DA4-878D-50BB942274C0@opencsw.org> Hi, Am 21.05.2009 um 16:10 schrieb John Ellson: >> Send a request to buildfarm at lists.opencsw.org requesting the >> install of cmake >> on the build farm. > > Will do. cmake is now being installed on all farm servers. Best regards -- Dago From phil at bolthole.com Thu May 21 19:09:44 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 10:09:44 -0700 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: References: <4A1582D0.8040709@opencsw.org> Message-ID: <20090521170944.GA597@bolthole.com> On Thu, May 21, 2009 at 06:59:54PM +0200, Dagobert Michelsen wrote: > Hi Sebastian, > > Am 21.05.2009 um 18:35 schrieb Sebastian Kayser: >> i packaged mailx that was formerly known as nail (CSWnail). Now how >> would we go about obsoleting CSWnail? Just drop it from the catalog, >> let >> CSWnail become an empty package that pulls in CSWmailx and somehow >> deletes CSWnail, or is there another preferred (proven) way of doing >> things like this? > > I vaguely recall pkg-get has some magic included that does this kind > of thing on making new packages incompatible ("I") with prior ones. > Phil? yes. the only magic is that you have to get users to upgrade from "nail" to the new package name. You can either do it in two stages, doin the "empty package that pulls in mailx", then in a few,months, switching it and making mailx conflict. OR, you can just start out with the conflict, and send out an announcement that users of 'nail' need to switch. From Joerg.Schilling at fokus.fraunhofer.de Thu May 21 19:20:36 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Thu, 21 May 2009 19:20:36 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A1582D0.8040709@opencsw.org> References: <4A1582D0.8040709@opencsw.org> Message-ID: <4a158d64.jN+yHRJlXTsKF9bv%Joerg.Schilling@fokus.fraunhofer.de> Sebastian Kayser wrote: > i packaged mailx that was formerly known as nail (CSWnail). Now how > would we go about obsoleting CSWnail? Just drop it from the catalog, let > CSWnail become an empty package that pulls in CSWmailx and somehow > deletes CSWnail, or is there another preferred (proven) way of doing > things like this? Don't omit the name "nail" from the bin directory. It helps to select this right binary. BTW: Was there a bug fix for the old and nasty problem that causes to let "nail" to hang with 50% probability via imaps when the Subject or the mail body is empty? J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From bwalton at opencsw.org Thu May 21 19:29:16 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 13:29:16 -0400 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> Message-ID: <1242926828-sup-5777@ntdws12.chass.utoronto.ca> Excerpts from Dagobert Michelsen's message of Thu May 21 12:58:27 -0400 2009: > Additionally, adding more complexity to System V packaging will bind > valuable resources and would make packaging quite more complex. > Apart from that the current mechanism has served us well over the > past 10+ years and will fade away as the new packaging system is > already there (IPS). I would favor to live with the faults and Yes, I'm not suggesting it was a great option, just _an_ option. > handle dependencies as we have always handled them. In addition, > I propose to carefully start working on real IPS packaging > instead of putting extra work/complexity in System V packaging. > A real IPS backend for GAR with all bells and whistels for > dual-personality building of packages is surely a Good Thing. Yes, I agree. Will IPS make pkgutil/pkg-get redundant? Eg, will it facilitate simply creating a 'repo' and adding it to the IPS conf (similar to apt/yum)? If so, all the smarts could be punted to the system tools instead, which would be nice. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. From phil at bolthole.com Thu May 21 19:36:14 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 10:36:14 -0700 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242926828-sup-5777@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> Message-ID: <20090521173614.GC597@bolthole.com> On Thu, May 21, 2009 at 01:29:16PM -0400, Ben Walton wrote: > Yes, I agree. Will IPS make pkgutil/pkg-get redundant? Eg, will it > facilitate simply creating a 'repo' and adding it to the IPS conf > (similar to apt/yum)? If so, all the smarts could be punted to the > system tools instead, which would be nice. personally, i think they screwed up with the fundamental design of IPS. [in multiple areas] My opinion, is that if we put in enough effort into improving SVR4 for the things that it really lacks (by our use of extending it through class action scripts, for example), that one day, IPS will be the NIS+, compared to pkgadd's NIS. (ie: NIS is still here, even after sun "killed" it, and NIS+ is not :-) Mind you, there's ldap on top of everything :) but still, the bottom line being that nis+ was just too complex for most peoples' needs. From ellson at opencsw.org Thu May 21 18:27:04 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 21 May 2009 12:27:04 -0400 Subject: [csw-maintainers] "package needs a database" general issue In-Reply-To: <20090521161307.GB89074@bolthole.com> References: <20090521161307.GB89074@bolthole.com> Message-ID: <4A1580D8.1040208@opencsw.org> I think many Fedora packages are using SQLite as their default database. No problems with starting daemons or setting up separate userids and passwords. John Philip Brown wrote: > Hi folks, > > There is an issue that comes up occasionally, where we want to offer an > 'easy install' option for people for a particular package, but that package > requires a database to run. > > Then there is a matter of; WHICH database should we use, for fully > automated install?!! > > usually it comes down to a choice of "postgres or mysql?" > > I would like to propose that we need to pick an official default database > to use. > > > Furthermore, I propose that that database be mysql. It's the simplest to > administer, it has the biggest mindshare, and there are lots of apps that > will ONLY support mysql, sadly. > > HOWEVER, I always like to support end-user choice whenever possible. > So, in addition to a default assumption of mysql... I would also like to > propose a setting in csw.conf. > Something like > > default_dbtype > > where values can be mysql, postgres, or even oracle. > If that value is set, then a package can presume that that database is > already installed and running. > In those cases though, we also need to agree on some kind of common > mechanism to insert values into said database. > > Comments? > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > > From bwalton at opencsw.org Thu May 21 19:47:37 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 13:47:37 -0400 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <20090521173614.GC597@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> Message-ID: <1242927821-sup-8734@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 13:36:14 -0400 2009: > personally, i think they screwed up with the fundamental design of > IPS. [in multiple areas] My opinion, is that if we put in enough > effort into improving SVR4 for the things that it really lacks (by > our use of extending it through class action scripts, for example), > that one day, IPS will be the NIS+, compared to pkgadd's NIS. The debian system works well (although I've never packaged anything using it). Good things (imo) are: 1. Depending on or being incompatible with specific versions (or ranges) of package X 2. Depends on mail-server/Provides mail-server 3. Depends on mysql|postgres Where does IPS fit into the landscape in comparison to debian and svr4? (Or tell me to go RTFM!) -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Thu May 21 19:56:35 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 12:56:35 -0500 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242927821-sup-8734@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> Message-ID: <4A1595D3.2060404@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > Excerpts from Philip Brown's message of Thu May 21 13:36:14 -0400 2009: > >> personally, i think they screwed up with the fundamental design of >> IPS. [in multiple areas] My opinion, is that if we put in enough >> effort into improving SVR4 for the things that it really lacks (by >> our use of extending it through class action scripts, for example), >> that one day, IPS will be the NIS+, compared to pkgadd's NIS. > > The debian system works well (although I've never packaged anything > using it). Good things (imo) are: > > 1. Depending on or being incompatible with specific versions (or > ranges) of package X Sun kinda supports this now in the depend files. man depend > 2. Depends on mail-server/Provides mail-server > 3. Depends on mysql|postgres This is one big thing the current Sun packages are really lacking > > Where does IPS fit into the landscape in comparison to debian and > svr4? (Or tell me to go RTFM!) > > -Ben > > > ------------------------------------------------------------------------ > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVldIACgkQLrhmsXMSLxd/VACglxZQErgPXp8nRpofd/qF9o1r RaAAoNHgPx0OfAmqEMnPd41jOh+9RYVL =hq2M -----END PGP SIGNATURE----- From phil at bolthole.com Thu May 21 20:02:57 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 11:02:57 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1595D3.2060404@opencsw.org> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> Message-ID: <20090521180256.GD597@bolthole.com> On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: > > 2. Depends on mail-server/Provides mail-server > > 3. Depends on mysql|postgres > This is one big thing the current Sun packages are really lacking > But how is that even processable? if you apt-get [somepackage], and that package "depends on mysql|postgres", and neither is installed... What happens? From mwatters at opencsw.org Thu May 21 20:05:21 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 21 May 2009 13:05:21 -0500 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521180256.GD597@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> Message-ID: <4A1597E1.8060602@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: >>> 2. Depends on mail-server/Provides mail-server >>> 3. Depends on mysql|postgres >> This is one big thing the current Sun packages are really lacking >> > > But how is that even processable? > > if you apt-get [somepackage], and that package > "depends on mysql|postgres", and neither is installed... > What happens? > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers If Neither is installed it errors out normally. if both are install it just installs. apt and yum do almost ZERO configuration for users. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoVl+EACgkQLrhmsXMSLxcYrACdEZWyJ4ZjcBGWGWX54joHgZlG GCwAnAk3F2XK5WPSy051AQc4S6nai8z2 =9KCY -----END PGP SIGNATURE----- From bwalton at opencsw.org Thu May 21 20:13:08 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 14:13:08 -0400 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521180256.GD597@bolthole.com> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> Message-ID: <1242929358-sup-207@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 14:02:57 -0400 2009: > if you apt-get [somepackage], and that package > "depends on mysql|postgres", and neither is installed... > What happens? I've never actually interacted with it through command-line apt, but always through aptitude. When I select a package that requires A|B, it will mark that package as having unmet dependencies. I can then resolve the dependency by selecting one of the available options. This satisfies the condition and away we go. Same goes for the general Provides/Requires. Broken until the selected package set meets the criteria. I suspect that command-line apt would simply quit with an error (likely listing the packages that would fulfill the requirements). If you ran apt-get again with firstpkg deppkg, it would run to completion. They also provide a 'suggests' feature that can interact with this (although it can be used independently also to provide additional, optional functionality). A 'suggested' package may be the preferred db backend, for example. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Thu May 21 20:18:25 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 11:18:25 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1597E1.8060602@opencsw.org> References: <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> Message-ID: <20090521181824.GE597@bolthole.com> On Thu, May 21, 2009 at 01:05:21PM -0500, Mike Watters wrote: > Philip Brown wrote: > > But how is that even processable? > > > > if you apt-get [somepackage], and that package > > "depends on mysql|postgres", and neither is installed... > > What happens? > > If Neither is installed it errors out normally. > if both are install it just installs. pfft. Sounds like there isnt much reason to hack that kind of "multiple|optional dependancy' into the pkg delivery system then. We could accomplish basically the same thing by other means. To take the database dependancy specifically; I'm thinking some kind of generic database wrapper tool package. Kinda like CPAN::DB, but for shellscripts :-) Then packages could depend on CSWdbwrapper, and use the appropriate tools in there as needed. With preference as to WHICH database to use, set in csw.conf locally, as I suggested in my other email today. Not only does it solve the "problem" in that area, but it would solve it BETTER than how debian 'solves' it. How does that sound? From bwalton at opencsw.org Thu May 21 20:27:54 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 14:27:54 -0400 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521181824.GE597@bolthole.com> References: <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> Message-ID: <1242930111-sup-5665@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 14:18:25 -0400 2009: > Sounds like there isnt much reason to hack that kind of > "multiple|optional dependancy' into the pkg delivery system then. > We could accomplish basically the same thing by other means. No, it can be handled above the actual package level by the tools manipulating them (which was the gist of option c)...I do agree that it's likely wasted effort to bolt it on at this point, since Sun will be abandoning svr4 at some point in the not too distant future... > To take the database dependancy specifically; I'm thinking some kind > of generic database wrapper tool package. Kinda like CPAN::DB, but > for shellscripts :-) Then packages could depend on CSWdbwrapper, and > use the appropriate tools in there as needed. To be fair, perl is likely a better tool for this anyway, so the DB module is a good choice. :) > Not only does it solve the "problem" in that area, but it would solve it > BETTER than how debian 'solves' it. ...except for the hole that is: none of those packages depend on the actual db package, so removing it wouldn't generate any warnings (unless some other package marked it explicitly). Would the CSWdbwrapper depend on both? With debian, removing the package providing the 'generic' rdbms would alert you to the dependency breakage. > How does that sound? Given the workings of svr4 packages, about as good as it'll get, most likely. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From dam at opencsw.org Thu May 21 22:02:36 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 22:02:36 +0200 Subject: [csw-maintainers] New Solaris package archive? Message-ID: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> Hi, I noticed today that there is another packagingproject I have never seen before: http://www.sunfreepacks.com They have around 70 packages for Solaris 10 x86 / OpenSolaris and a few for Solaris 10 Sparc. They don't seem to have a catalog or packaging tool. They are building on top of Solaris / SunFreeware packages. Does anyone have experience with it? I am currently mirroring it to /export/mirror/sunfreepacks, just in case someone wants to take a look. Best regards -- Dago From phil at bolthole.com Thu May 21 22:16:30 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 13:16:30 -0700 Subject: [csw-maintainers] New Solaris package archive? In-Reply-To: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> References: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> Message-ID: <20090521201630.GB46162@bolthole.com> On Thu, May 21, 2009 at 10:02:36PM +0200, Dagobert Michelsen wrote: > Hi, > > I noticed today that there is another packagingproject I have > never seen before: > > http://www.sunfreepacks.com kinda odd. no mention of why they are bothering to do this. and it mostly seems to be geared towards opensolaris, and NOT solaris 10. For example, their ASdjvulibre package, is marked as ONLY working on opensolaris. From a.cervellin at acm.org Thu May 21 22:22:14 2009 From: a.cervellin at acm.org (Alessio) Date: Thu, 21 May 2009 22:22:14 +0200 Subject: [csw-maintainers] New Solaris package archive? In-Reply-To: <20090521201630.GB46162@bolthole.com> References: <68287851-FCAD-453C-BAD5-50C291F07669@opencsw.org> <20090521201630.GB46162@bolthole.com> Message-ID: <4A15B7F6.5070904@acm.org> Philip Brown wrote: > On Thu, May 21, 2009 at 10:02:36PM +0200, Dagobert Michelsen wrote: > >> Hi, >> >> I noticed today that there is another packagingproject I have >> never seen before: >> >> http://www.sunfreepacks.com >> > > kinda odd. no mention of why they are bothering to do this. my guess: probably for they own use + advertising... it seems they are sun principal partner and they do web hosting as business From phil at bolthole.com Thu May 21 22:23:49 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 13:23:49 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <1242930111-sup-5665@ntdws12.chass.utoronto.ca> References: <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> <1242930111-sup-5665@ntdws12.chass.utoronto.ca> Message-ID: <20090521202349.GE46162@bolthole.com> On Thu, May 21, 2009 at 02:27:54PM -0400, Ben Walton wrote: > > Not only does it solve the "problem" in that area, but it would solve it > > BETTER than how debian 'solves' it. > > ...except for the hole that is: none of those packages depend on the > actual db package, so removing it wouldn't generate any warnings > (unless some other package marked it explicitly). Would the > CSWdbwrapper depend on both? With debian, removing the package > providing the 'generic' rdbms would alert you to the dependency > breakage. > I doubt that debian does it any better. For example, lets say that you were installing... [foo], which depended on "mysql|postgres", and you had BOTH installed. You chose to install it using mysql. Then later, you choose to remove mysql, but postgres is still there. does apt-get/aptitude somehow register that you chose mysql specifically for package [foo], and complains about the removal? I am scheptical about this. From phil at bolthole.com Thu May 21 22:30:09 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 13:30:09 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521202349.GE46162@bolthole.com> References: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> <1242930111-sup-5665@ntdws12.chass.utoronto.ca> <20090521202349.GE46162@bolthole.com> Message-ID: <20090521203009.GF46162@bolthole.com> On Thu, May 21, 2009 at 01:23:49PM -0700, Philip Brown wrote: > I doubt that debian does it any better. > For example, lets say that you were installing... [foo], which depended on > "mysql|postgres", and you had BOTH installed. > > You chose to install it using mysql. > Then later, you choose to remove mysql, but postgres is still there. > > does apt-get/aptitude somehow register that you chose mysql specifically > for package [foo], > and complains about the removal? > I am scheptical about this. btw; we could do this sort of check for ourselves, via the suggested "dbwrapper" set of tools. If we agree that packages that use the dbwrapper tools, also have to 'register' which database they end up using, then that database can check the registration, and warn about programs using it, at pkgrm time. From dam at opencsw.org Thu May 21 22:55:49 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 21 May 2009 22:55:49 +0200 Subject: [csw-maintainers] Commit #5000 today! Message-ID: Hi, today we have to celebrate a jubilee: Subversion commit #5000 was done today from Mike Watters on an updated package for Drupal with commit #4000 just about two month ago. Thanks to all those helping hands who made this possible! Commits are not a virtue on their own, but they reflect the activity on the project. At the current rate over the past month we have well over 300 writes per month with over 1000 updated files: Starting from March the repository reads exploded from a few hundred to several thousands. I guess the number of people rebuilding the OpenCSW packages for fun or tweaks is increasing quickly. So, please everybody grab a piece, I already had mine :-) _____ _..--'''@ @'''--.._ .' @_/-//-\/>/>'/ @ '. ( @ /_ References: <4A1582D0.8040709@opencsw.org> <4a158d64.jN+yHRJlXTsKF9bv%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <4A15C4C4.5000807@opencsw.org> Hi Joerg, Joerg Schilling wrote: > Sebastian Kayser wrote: > >> i packaged mailx that was formerly known as nail (CSWnail). Now how >> would we go about obsoleting CSWnail? Just drop it from the catalog, let >> CSWnail become an empty package that pulls in CSWmailx and somehow >> deletes CSWnail, or is there another preferred (proven) way of doing >> things like this? > > Don't omit the name "nail" from the bin directory. It helps to select this > right binary. Phil also noted it, so i guess i could add a nail symlink (pointing to mailx) to the package. What exactly do you mean by "select this right binary"? > BTW: Was there a bug fix for the old and nasty problem that causes to let "nail" > to hang with 50% probability via imaps when the Subject or the mail body is > empty? I mostly use mailx non-interactive (for scripting) and am neither aware of the issue nor of a fix. However, i put the new package in testing, so if you have been experiencing the issue, feel free to give the new package a spin. http://mirror.opencsw.org/testing.html Sebastian From skayser at opencsw.org Thu May 21 23:26:34 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Thu, 21 May 2009 23:26:34 +0200 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <20090521170334.GH89074@bolthole.com> References: <4A1582D0.8040709@opencsw.org> <20090521170334.GH89074@bolthole.com> Message-ID: <4A15C70A.1010808@opencsw.org> Philip Brown wrote: > On Thu, May 21, 2009 at 06:35:28PM +0200, Sebastian Kayser wrote: >> Hi, >> >> i packaged mailx that was formerly known as nail (CSWnail). Now how >> would we go about obsoleting CSWnail? Just drop it from the catalog, let >> CSWnail become an empty package that pulls in CSWmailx and somehow >> deletes CSWnail, or is there another preferred (proven) way of doing >> things like this? > > > ugh. is the binary really called "mailx"? that would be very lame. > I think it might actually be nice to keep it as called "nail" :-} Yep, "nail has been integrated into the Heirloom project as mailx" [1] and the binary is called mailx. I could add a symlink called nail that points to mailx. Thoughts? Sebastian [1] http://heirloom.sourceforge.net/mailx.html From phil at bolthole.com Thu May 21 23:31:18 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 21 May 2009 14:31:18 -0700 Subject: [csw-maintainers] Packaged CSWmailx. How to obsolete CSWnail? In-Reply-To: <4A15C70A.1010808@opencsw.org> References: <4A1582D0.8040709@opencsw.org> <20090521170334.GH89074@bolthole.com> <4A15C70A.1010808@opencsw.org> Message-ID: <20090521213118.GK46162@bolthole.com> On Thu, May 21, 2009 at 11:26:34PM +0200, Sebastian Kayser wrote: > Yep, "nail has been integrated into the Heirloom project as mailx" [1] > and the binary is called mailx. I could add a symlink called nail that > points to mailx. Thoughts? gets my vote From bwalton at opencsw.org Fri May 22 02:17:02 2009 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 21 May 2009 20:17:02 -0400 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521203009.GF46162@bolthole.com> References: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> <1242930111-sup-5665@ntdws12.chass.utoronto.ca> <20090521202349.GE46162@bolthole.com> <20090521203009.GF46162@bolthole.com> Message-ID: <1242951214-sup-5073@ntdws12.chass.utoronto.ca> Excerpts from Philip Brown's message of Thu May 21 16:30:09 -0400 2009: > On Thu, May 21, 2009 at 01:23:49PM -0700, Philip Brown wrote: > > I doubt that debian does it any better. > > For example, lets say that you were installing... [foo], which depended on > > "mysql|postgres", and you had BOTH installed. > > > > You chose to install it using mysql. > > Then later, you choose to remove mysql, but postgres is still there. > > > > does apt-get/aptitude somehow register that you chose mysql specifically > > for package [foo], > > and complains about the removal? > > I am scheptical about this. > > btw; we could do this sort of check for ourselves, via the suggested > "dbwrapper" set of tools. > If we agree that packages that use the dbwrapper tools, also have to > 'register' which database they end up using, then that database can > check the registration, and warn about programs using it, at pkgrm time. I just did a quick check on a debian box here. The gallery2 app depends on mysql-client|postgresql-client. With only the mysql version installed, it indicated gallery2 would be removed. Installing postgresql-client and then removing mysql-client triggered other dependency warnings, but not the gallery2 one. I was thinking the same, that a postinstall hook could register which package satisfied the dependency, thus creating something the package manager could use later.... Anyway, as I mentioned, the dbwrapper approach is likely suffiecient for this purpose...I wasn't intended to open a large debate. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From pfelecan at opencsw.org Fri May 22 10:32:15 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 10:32:15 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1597E1.8060602@opencsw.org> (Mike Watters's message of "Thu\, 21 May 2009 13\:05\:21 -0500") References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> Message-ID: Mike Watters writes: > Philip Brown wrote: >> On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: >>>> 2. Depends on mail-server/Provides mail-server >>>> 3. Depends on mysql|postgres >>> This is one big thing the current Sun packages are really lacking >>> >> >> But how is that even processable? >> >> if you apt-get [somepackage], and that package >> "depends on mysql|postgres", and neither is installed... >> What happens? >> _______________________________________________ > > If Neither is installed it errors out normally. > if both are install it just installs. > > apt and yum do almost ZERO configuration for users. They doesn't need to. The configuration is the realm of the package based on system defaults/parameters. This is done in pre/post install scripts. Of course, there are helpful mechanisms such as alternatives, standard configurators, &c. Finally, there are policies enforced by various tools. Speaking as someone that have done, professionally Debian and RedHat packages. -- Peter From pfelecan at opencsw.org Fri May 22 10:35:30 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 10:35:30 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090521181824.GE597@bolthole.com> (Philip Brown's message of "Thu\, 21 May 2009 11\:18\:25 -0700") References: <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <20090521181824.GE597@bolthole.com> Message-ID: Philip Brown writes: > On Thu, May 21, 2009 at 01:05:21PM -0500, Mike Watters wrote: >> Philip Brown wrote: >> > But how is that even processable? >> > >> > if you apt-get [somepackage], and that package >> > "depends on mysql|postgres", and neither is installed... >> > What happens? >> >> If Neither is installed it errors out normally. >> if both are install it just installs. > > pfft. > > Sounds like there isnt much reason to hack that kind of > "multiple|optional dependancy' into the pkg delivery system then. > We could accomplish basically the same thing by other means. > > To take the database dependancy specifically; > I'm thinking some kind of generic database wrapper tool package. > Kinda like CPAN::DB, but for shellscripts :-) > Then packages could depend on CSWdbwrapper, and use the appropriate > tools in there as needed. > > With preference as to WHICH database to use, set in csw.conf locally, as I > suggested in my other email today. I'm supporting this idea, even if the idea of a package being a provider of an abstract service is nice but probably harder to implement in our context. > Not only does it solve the "problem" in that area, but it would solve it > BETTER than how debian 'solves' it. Not better, diferent. > How does that sound? Good -- Peter From pfelecan at opencsw.org Fri May 22 10:47:56 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 10:47:56 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: (Dagobert Michelsen's message of "Thu\, 21 May 2009 22\:55\:49 +0200") References: Message-ID: Dagobert Michelsen writes: > Hi, > > > today we have to celebrate a jubilee: Subversion commit #5000 was > done today Well, this show how young the project is... When you are in 5 or more digits it show that you're out of puberty. > So, please everybody grab a piece, I already had mine :-) > > _____ > _..--'''@ @'''--.._ > .' @_/-//-\/>/>'/ @ '. > ( @ /_ |'._ @ //|###########| > |~ ''--..@|',|}}}}}}}}}}}| > | ~ ~ |/ |###########| > | ~~ ~ ~|./|{{{{{{{{{{{| > '._ ~ ~ ~ |,/````````````` > ''--.~.|/ I hope that the next time we'll taste the good German cakes --- have a feeble for the schwarzwald kirche torte and the Bavarian ecosystem. Yum -- Peter From dam at opencsw.org Fri May 22 11:04:07 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 22 May 2009 11:04:07 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: References: Message-ID: Hi Peter, Am 22.05.2009 um 10:47 schrieb Peter FELECAN: >> So, please everybody grab a piece, I already had mine :-) >> >> _____ >> _..--'''@ @'''--.._ >> .' @_/-//-\/>/>'/ @ '. >> ( @ /_> |'._ @ //|###########| >> |~ ''--..@|',|}}}}}}}}}}}| >> | ~ ~ |/ |###########| >> | ~~ ~ ~|./|{{{{{{{{{{{| >> '._ ~ ~ ~ |,/````````````` >> ''--.~.|/ > > I hope that the next time we'll taste the good German cakes --- have a > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. But it is! }}} = cherry, ### = cake, {{{ = cream For clarification: It is "Schwarzw?lder Kirschtorte". "kirche = church" against "kirsche = cherry" :-D Best regards -- Dago From Joerg.Schilling at fokus.fraunhofer.de Fri May 22 11:26:56 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Fri, 22 May 2009 11:26:56 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: References: Message-ID: <4a166fe0.D4FEcP60kyIOZmVt%Joerg.Schilling@fokus.fraunhofer.de> Dagobert Michelsen wrote: > > I hope that the next time we'll taste the good German cakes --- have a > > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. > > But it is! }}} = cherry, ### = cake, {{{ = cream > For clarification: It is "Schwarzw?lder Kirschtorte". > "kirche = church" against "kirsche = cherry" :-D See the keyboard manufacturer from "Schwarzwald": Cherry ;-) J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From schwindt at dfki.uni-kl.de Fri May 22 11:47:10 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Fri, 22 May 2009 11:47:10 +0200 Subject: [csw-maintainers] gcal 3.0.1 now in testing Message-ID: <200905220947.n4M9lBcV028915@dfki.uni-kl.de> Very old package - nevertheless very usefull. Nicolai Schwindt. From pfelecan at opencsw.org Fri May 22 14:12:09 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 22 May 2009 14:12:09 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: <4a166fe0.D4FEcP60kyIOZmVt%Joerg.Schilling@fokus.fraunhofer.de> (Joerg Schilling's message of "Fri\, 22 May 2009 11\:26\:56 +0200") References: <4a166fe0.D4FEcP60kyIOZmVt%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: > Dagobert Michelsen wrote: > >> > I hope that the next time we'll taste the good German cakes --- have a >> > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. >> >> But it is! }}} = cherry, ### = cake, {{{ = cream >> For clarification: It is "Schwarzw?lder Kirschtorte". >> "kirche = church" against "kirsche = cherry" :-D Sorry about this funny confusion... indeed I prefer the cherry for the torte. > See the keyboard manufacturer from "Schwarzwald": Cherry ;-) For which I have also a feeble: owning 3 of them clunkers, though my office neighbours don't like them... -- Peter From ellson at opencsw.org Fri May 22 20:50:29 2009 From: ellson at opencsw.org (John Ellson) Date: Fri, 22 May 2009 14:50:29 -0400 Subject: [csw-maintainers] Package "liblasi" in testing Message-ID: <4A16F3F5.9020708@opencsw.org> There is a new package: "liblasi: C++ Unicode support for PostScript" now available in testing. The GAR processing of this file was a bit non standard: - libLASi not permitted, so using liblasi for $(GARNAME) - Use of CMAKE - Sun's C++ compiler didn't work, so used g++ so, please see if: mgar/pkg/liblasi/trunk/Makefile can be improved. John From rupert at opencsw.org Fri May 22 21:38:53 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 22 May 2009 21:38:53 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <4A1597E1.8060602@opencsw.org> References: <4A14A466.5010602@opencsw.org> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> Message-ID: <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> On Thu, May 21, 2009 at 20:05, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Philip Brown wrote: >> On Thu, May 21, 2009 at 12:56:35PM -0500, Mike Watters wrote: >>>> 2. Depends on mail-server/Provides mail-server >>>> 3. Depends on mysql|postgres >>> This is one big thing the current Sun packages are really lacking >>> >> >> But how is that even processable? >> >> if you apt-get [somepackage], and that package >> "depends on mysql|postgres", and neither is installed... >> What happens? >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > If Neither is installed it errors out normally. > if both are install it just installs. > > apt and yum do almost ZERO configuration for users. would it be a lot of effort to replace pkgutil with apt? i especially like the remove parts, the dist features, and wajig as quite comfortable command line interface. e.g. wajig purge, wajig purge-orphans, ... rupert. From phil at bolthole.com Fri May 22 21:45:30 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 22 May 2009 12:45:30 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> References: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> Message-ID: <20090522194530.GH82222@bolthole.com> On Fri, May 22, 2009 at 09:38:53PM +0200, rupert THURNER wrote: > > would it be a lot of effort to replace pkgutil with apt? i especially > like the remove parts, the dist features, and wajig as quite > comfortable command line interface. > > e.g. wajig purge, wajig purge-orphans, ... If you want to attempt to port apt to work on top of SVR4 pkgs, please feel free to go right ahead :-) [I think someone actually attempted to do this, so you might be able to pick up on their work. But it's probably fairly out of date] Meanwhile, if you wish to request additional features to pkg-get and/or pkgutil, please also feel free to suggest specifics. You may not have noticed, but in one of my recent emails, I referenced support for a "purge" function. This has been in my plans for pkg-get for a long time. It would be easy for pkgutil to implement it as well. As mentioned in my email, the concept would be that maintainers would provde a "purge" script, as i purge I think that pkg-get could check for it, pre-pkgrm, and run it. From rupert at opencsw.org Fri May 22 22:12:37 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 22 May 2009 22:12:37 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools Message-ID: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> hi, while trying to easy_install python apps, i fell over a problem with the path mike used for compiling python: /opt/studio/.. . this is a problem for filesystems like ours which uses a readonly /opt. one has to set CC, LDSHARED, LDFLAGS, and for some like genshi LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be possible to link/mount sunstudio within /opt/csw especially for python compiles? see http://groups.google.com/group/distutils-sig/browse_thread/thread/d8d2c47eb1d96210 for details. i opened also an extension request, unfortunately in the wrong package as i did not understand it early enough: http://www.opencsw.org/bugtrack/view.php?id=3682. rupert. From rupert at opencsw.org Fri May 22 22:42:33 2009 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 22 May 2009 22:42:33 +0200 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <20090522194530.GH82222@bolthole.com> References: <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> <20090522194530.GH82222@bolthole.com> Message-ID: <6af4270905221342u5fb70d43yb5915c52aa4e9e3b@mail.gmail.com> On Fri, May 22, 2009 at 21:45, Philip Brown wrote: > On Fri, May 22, 2009 at 09:38:53PM +0200, rupert THURNER wrote: >> >> would it be a lot of effort to replace pkgutil with apt? i especially >> like the remove parts, the dist features, and wajig as quite >> comfortable command line interface. >> >> e.g. wajig purge, wajig purge-orphans, ... > > If you want to attempt to port apt to work on top of SVR4 pkgs, please feel > free to go right ahead :-) > [I think someone actually attempted to do this, so you might be able to > ?pick up on their work. But it's probably fairly out of date] the only thing up-to-date i know is http://nexenta.org, http://archive.nexenta.org/, who advertise themselves as "power of opensolaris with usability of linux". but i am a too bad packaging person to really judge this, and the impact / benefit it could have on opencsw. rupert. From phil at bolthole.com Fri May 22 23:34:03 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 22 May 2009 14:34:03 -0700 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> Message-ID: <20090522213403.GI82222@bolthole.com> On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: > hi, > > while trying to easy_install python apps, i fell over a problem with > the path mike used for compiling python: /opt/studio/.. . this is a > problem for filesystems like ours which uses a readonly /opt. ?? why is that a problem? hm. it appears from the newsgroup reference, that you are specifically complaining, that python has somehow hardcoded in its configs, CC=/full/path/to/cc instead of a nice simple CC=cc that second one definately sounds better. that being said... > one has to set CC, LDSHARED, LDFLAGS, and for some like genshi > LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be > possible to link/mount sunstudio within /opt/csw especially for python > compiles? > > see http://groups.google.com/group/distutils-sig/browse_thread/thread/d8d2c47eb1d96210 > for details. where did LD_LIBRARY_PATH come in? From phil at bolthole.com Fri May 22 23:38:10 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 22 May 2009 14:38:10 -0700 Subject: [csw-maintainers] Packages that require either package A or package B' In-Reply-To: <6af4270905221342u5fb70d43yb5915c52aa4e9e3b@mail.gmail.com> References: <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> <20090521173614.GC597@bolthole.com> <1242927821-sup-8734@ntdws12.chass.utoronto.ca> <4A1595D3.2060404@opencsw.org> <20090521180256.GD597@bolthole.com> <4A1597E1.8060602@opencsw.org> <6af4270905221238q432627b8g4885b605fff971d@mail.gmail.com> <20090522194530.GH82222@bolthole.com> <6af4270905221342u5fb70d43yb5915c52aa4e9e3b@mail.gmail.com> Message-ID: <20090522213810.GJ82222@bolthole.com> On Fri, May 22, 2009 at 10:42:33PM +0200, rupert THURNER wrote: > the only thing up-to-date i know is http://nexenta.org, > http://archive.nexenta.org/, who advertise themselves as "power of > opensolaris with usability of linux". but i am a too bad packaging > person to really judge this, and the impact / benefit it could have on > opencsw. note that they use opensolaris. their own completely custom distribution of solaris. They have solaris-native apt-get, using debian style packaging, NOT SVR4 packaging, I think. Verifying... http://www.nexenta.org/os/DownloadMirrors http://mirror.sourceshare.org/nexenta/apt/dists/elatte-unstable/main/binary-solaris-i386/mail/ has a bunch of .deb files. yup. debian .deb/dpkg format. not svr4. From mwatters at opencsw.org Sat May 23 00:41:25 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 22 May 2009 17:41:25 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> Message-ID: <4A172A15.2030303@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > hi, > > while trying to easy_install python apps, i fell over a problem with > the path mike used for compiling python: /opt/studio/.. . this is a > problem for filesystems like ours which uses a readonly /opt. > > one has to set CC, LDSHARED, LDFLAGS, and for some like genshi > LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be > possible to link/mount sunstudio within /opt/csw especially for python > compiles? > > see http://groups.google.com/group/distutils-sig/browse_thread/thread/d8d2c47eb1d96210 > for details. > > i opened also an extension request, unfortunately in the wrong package > as i did not understand it early enough: > http://www.opencsw.org/bugtrack/view.php?id=3682. > > rupert. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers I moved the bug to the right project, The new bug URL is http://www.opencsw.org/mantis/view.php?id=3684 This is a relatively easy fix, I will add it to the next build. the next build will be the one that no longer contains the pyo/pyc files but generates them on install.... still working on the class scripts for this. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoXKhUACgkQLrhmsXMSLxe04gCgmN13tAOsRsT2axVxhKDpInsS UToAoOZ0MNsLO0b7L81m4HALyV+mLbdD =isOE -----END PGP SIGNATURE----- From rupert at opencsw.org Sat May 23 02:56:38 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 23 May 2009 02:56:38 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <20090522213403.GI82222@bolthole.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> Message-ID: <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> On Fri, May 22, 2009 at 23:34, Philip Brown wrote: > On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: >> hi, >> >> while trying to easy_install python apps, i fell over a problem with >> the path mike used for compiling python: /opt/studio/.. . this is a >> problem for filesystems like ours which uses a readonly /opt. > > ?? > > why is that a problem? > > > hm. it appears from the newsgroup reference, that you are specifically > complaining, that python has somehow hardcoded in its configs, > CC=/full/path/to/cc > instead of a nice simple > > CC=cc the "problem" if you want to call it like this is that easy_install uses the settings out of the python compilation makefile as default. it would be nicer if the default, i.e. the way python is compiled, uses a compiler within /opt/csw to make it easy to duplicate it on a opencsw client site, where other paths outside /opt/csw are not necessarily writable. i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set to /opt/csw/lib to be easy_installed at our site though. rupert. From mwatters at opencsw.org Sat May 23 03:03:17 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 22 May 2009 20:03:17 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> Message-ID: <4A174B55.6000303@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > On Fri, May 22, 2009 at 23:34, Philip Brown wrote: >> On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: >>> hi, >>> >>> while trying to easy_install python apps, i fell over a problem with >>> the path mike used for compiling python: /opt/studio/.. . this is a >>> problem for filesystems like ours which uses a readonly /opt. >> ?? >> >> why is that a problem? >> >> >> hm. it appears from the newsgroup reference, that you are specifically >> complaining, that python has somehow hardcoded in its configs, >> CC=/full/path/to/cc >> instead of a nice simple >> >> CC=cc > > the "problem" if you want to call it like this is that easy_install > uses the settings out of the python compilation makefile as default. > it would be nicer if the default, i.e. the way python is compiled, > uses a compiler within /opt/csw to make it easy to duplicate it on a > opencsw client site, where other paths outside /opt/csw are not > necessarily writable. > > i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set > to /opt/csw/lib to be easy_installed at our site though. > > rupert. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Technically, this could/should be resolved by requesting CSW builds of those extensions. I have already got a genshi package in the works because I have been playing with Trac, all I have to do is a couple tweaks to the build and do a final commit. I am running it on my test Trac server and it works great. ;) I will finish it up next week. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoXS1UACgkQLrhmsXMSLxcERQCfZZN1cH0XSX27+MplX+1PQN1E DD8AniDa2KI0IePNik6P+ghA8PkcpAEJ =A1Cd -----END PGP SIGNATURE----- From rupert at opencsw.org Sat May 23 03:09:34 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 23 May 2009 03:09:34 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <4A174B55.6000303@opencsw.org> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> Message-ID: <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> On Sat, May 23, 2009 at 03:03, Mike Watters wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > rupert THURNER wrote: >> On Fri, May 22, 2009 at 23:34, Philip Brown wrote: >>> On Fri, May 22, 2009 at 10:12:37PM +0200, rupert THURNER wrote: >>>> hi, >>>> >>>> while trying to easy_install python apps, i fell over a problem with >>>> the path mike used for compiling python: /opt/studio/.. . this is a >>>> problem for filesystems like ours which uses a readonly /opt. >>> ?? >>> >>> why is that a problem? >>> >>> >>> hm. it appears from the newsgroup reference, that you are specifically >>> complaining, that python has somehow hardcoded in its configs, >>> CC=/full/path/to/cc >>> instead of a nice simple >>> >>> CC=cc >> >> the "problem" if you want to call it like this is that easy_install >> uses the settings out of the python compilation makefile as default. >> it would be nicer if the default, i.e. the way python is compiled, >> uses a compiler within /opt/csw to make it easy to duplicate it on a >> opencsw client site, where other paths outside /opt/csw are not >> necessarily writable. >> >> i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set >> to /opt/csw/lib to be easy_installed at our site though. >> >> rupert. >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers > > Technically, this could/should be resolved by requesting CSW builds of those > extensions. ?I have already got a genshi package in the works because I have > been playing with Trac, ?all I have to do is a couple tweaks to the build and > do a final commit. ?I am running it on my test Trac server and it works great. > ?;) ?I will finish it up next week. that is great to hear! i even would appreciate if the most important plugins make it into trac as well (not enabled by default), e.g. easy_install http://trac-hacks.org/svn/tracwysiwygplugin/0.11/ easy_install http://trac-hacks.org/svn/ticketdeleteplugin/0.11/ easy_install http://trac-hacks.org/svn/ticketchangeplugin/0.11/ easy_install http://trac-hacks.org/svn/customfieldadminplugin/0.11/ easy_install http://trac-hacks.org/svn/tagsplugin/tags/0.6/ easy_install http://svn.ipd.uka.de/repos/javaparty/JP/trac/plugins/tracnav-0.11/ easy_install http://trac-hacks.org/svn/sectioneditplugin/0.11/ easy_install http://trac-hacks.org/svn/extracturlplugin/0.11/ easy_install http://trac-hacks.org/svn/serversideredirectplugin/0.11/ easy_install http://trac-hacks.org/svn/masterticketsplugin/0.11/ easy_install http://trac-hacks.org/svn/autoupgradeplugin/0.11/ but, otoh, one can never package all python apps, so a possibility to quickly compile them with a default setting would be really nice :) rupert. From mwatters at opencsw.org Sat May 23 16:16:56 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 23 May 2009 09:16:56 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <6af4270905221809u232fdfe7l7f55df4705a0434@mail.gmail.com> Message-ID: <4A180558.70601@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi rupert, rupert THURNER wrote: > > that is great to hear! i even would appreciate if the most important > plugins make it into trac as well (not enabled by default), e.g. > > easy_install http://trac-hacks.org/svn/tracwysiwygplugin/0.11/ > easy_install http://trac-hacks.org/svn/ticketdeleteplugin/0.11/ > easy_install http://trac-hacks.org/svn/ticketchangeplugin/0.11/ > easy_install http://trac-hacks.org/svn/customfieldadminplugin/0.11/ > easy_install http://trac-hacks.org/svn/tagsplugin/tags/0.6/ > easy_install http://svn.ipd.uka.de/repos/javaparty/JP/trac/plugins/tracnav-0.11/ > easy_install http://trac-hacks.org/svn/sectioneditplugin/0.11/ > easy_install http://trac-hacks.org/svn/extracturlplugin/0.11/ > easy_install http://trac-hacks.org/svn/serversideredirectplugin/0.11/ > easy_install http://trac-hacks.org/svn/masterticketsplugin/0.11/ > easy_install http://trac-hacks.org/svn/autoupgradeplugin/0.11/ > I am working on updating Trac in the repository. ( after I get python re-released as separate packages.) I will create a second package for trac, trac_plugin_pack that contains the eggs you listed above. also, your last post on the google group forum link in the ticket. the '-G' is gone because you export LDSHARED=/opt/SUNWspro/bin/cc ''' compiler.linker_so ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] ''' means that then setting for compiler.linker_so is /opt/studio/SOS11/SUNWspro/bin/cc -G os.environ settings override what python finds in its config. export LDSHARED='/opt/SUNWspro/bin/cc -G' would have saved the -G you can easily play with this using the following python script =============================================================== #! /usr/bin/env python from distutils.ccompiler import new_compiler from distutils.sysconfig import customize_compiler compiler = new_compiler(None,None,None,None) customize_compiler(compiler) print """ *** Compiler Settings for Python *** Type: %s Linker: %s Compiler: %s """ % (compiler.compiler_type, compiler.linker_so, compiler.compiler) ====================================================================== $ ./pycompile.py *** Compiler Settings for Python *** Type: unix Linker: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-G'] Compiler: ['/opt/studio/SOS11/SUNWspro/bin/cc', '-DNDEBUG', '-O'] you can set various environment variables and see how that affects those settings. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoYBVgACgkQLrhmsXMSLxdLswCg4qlEUAdZIpxfr6B/0UpQhQHj UW4AoLk0dKQ3J5rsmkEwylnShAaWUtrh =GqGq -----END PGP SIGNATURE----- From phil at bolthole.com Sat May 23 18:36:57 2009 From: phil at bolthole.com (Philip Brown) Date: Sat, 23 May 2009 09:36:57 -0700 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> Message-ID: <20090523163657.GA24550@bolthole.com> On Sat, May 23, 2009 at 02:56:38AM +0200, rupert THURNER wrote: > i do not know why genshi, and mercurial need the LD_LIBRARY_PATH set > to /opt/csw/lib to be easy_installed at our site though. possibly because by default, LD_LIBRARY_PATH is set incorrectly in the user environment to start with. otherwise, how about you post here what breaks when it is not set (with "env - /opt/csw/bin/hg" to clear your environment first) and folks might be able to suggest something From bwalton at opencsw.org Sat May 23 19:12:01 2009 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 23 May 2009 13:12:01 -0400 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> Message-ID: <1243098505-sup-2495@ntdws12.chass.utoronto.ca> Excerpts from rupert THURNER's message of Fri May 22 16:12:37 -0400 2009: > one has to set CC, LDSHARED, LDFLAGS, and for some like genshi > LD_LIBRARY_PATH (/otp/csw) to make easy_install work. would it be > possible to link/mount sunstudio within /opt/csw especially for python > compiles? With regard to the CC vs /opt/studio issue, could we solve this by having GAR prefix the PATH with the appropriate $($(GARCOMPILER)_CC_HOME)/bin and then set CC appropriately without the full path? Doing this would make things (like python, ruby, etc) that store the values used to build them work against an appropriately set path at any site...Am I overlooking a negative impact that would be caused by a change like this? I don't have much input for the other issues and it seems that Mike and Phil have provided some useful starting points. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From mwatters at opencsw.org Sat May 23 19:23:16 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sat, 23 May 2009 12:23:16 -0500 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <1243098505-sup-2495@ntdws12.chass.utoronto.ca> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <1243098505-sup-2495@ntdws12.chass.utoronto.ca> Message-ID: <4A183104.8030503@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Ben Walton wrote: > With regard to the CC vs /opt/studio issue, could we solve this by > having GAR prefix the PATH with the appropriate > $($(GARCOMPILER)_CC_HOME)/bin and then set CC appropriately without > the full path? Doing this would make things (like python, ruby, etc) > that store the values used to build them work against an appropriately > set path at any site...Am I overlooking a negative impact that would > be caused by a change like this? I can't think of any negative for this. having CC report as cc or gcc is much better IMO then the full path. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoYMQQACgkQLrhmsXMSLxcZOwCgnBmFiDJlzOj98Fl9KSzN4Dvu gv4AoMV6woT9R8A5yFyXtzX0J40xzTLB =swBW -----END PGP SIGNATURE----- From schwindt at dfki.uni-kl.de Sat May 23 20:19:55 2009 From: schwindt at dfki.uni-kl.de (Nicolai Schwindt) Date: Sat, 23 May 2009 20:19:55 +0200 (CEST) Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <4A174B55.6000303@opencsw.org> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> Message-ID: <32825.88.134.110.210.1243102795.squirrel@www.dfki.uni-kl.de> On Sat, May 23, 2009 03:03, Mike Watters wrote: [...] > Technically, this could/should be resolved by requesting CSW builds of > those > extensions. I have already got a genshi package in the works because I > have > been playing with Trac, all I have to do is a couple tweaks to the build > and > do a final commit. I am running it on my test Trac server and it works > great. > ;) I will finish it up next week. .) That's why I asked how to know whos taking over which packet. I had already done trac 0.11.4 butwas waiting on python, svn, and pysqlite to settle ,) As you have the same Idea of making a trac-plugin-pack pls consider these aswell : TracSubversionLocation-1.0.1 GitPython-0.1.6 accountmanagerplugin_0.11-r5499 downloadsplugin-r5499 httpauthplugin-r5728 iniadminplugin_0.11 privatewikiplugin-r5499 reposearchplugin BTW : ticketdelete and ticketchange have been merged. -- Nicolai ----------------------------------------- This email was sent using SquirrelMail. "Webmail for nuts!" http://squirrelmail.org/ From rupert at opencsw.org Sat May 23 23:31:06 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 23 May 2009 23:31:06 +0200 Subject: [csw-maintainers] sun studio in standard path within /opt/csw, for the sake of python setuptools In-Reply-To: <32825.88.134.110.210.1243102795.squirrel@www.dfki.uni-kl.de> References: <6af4270905221312t7e9b76e5se68ccb2f1fa3e8fe@mail.gmail.com> <20090522213403.GI82222@bolthole.com> <6af4270905221756n51065a62w7f496ee8456dd736@mail.gmail.com> <4A174B55.6000303@opencsw.org> <32825.88.134.110.210.1243102795.squirrel@www.dfki.uni-kl.de> Message-ID: <6af4270905231431g34edd432ieb9438d8396b5ff0@mail.gmail.com> On Sat, May 23, 2009 at 20:19, Nicolai Schwindt wrote: > > > > > On Sat, May 23, 2009 03:03, Mike Watters wrote: > [...] >> Technically, this could/should be resolved by requesting CSW builds of >> those >> extensions. ?I have already got a genshi package in the works because I >> have >> been playing with Trac, ?all I have to do is a couple tweaks to the build >> and >> do a final commit. ?I am running it on my test Trac server and it works >> great. >> ?;) ?I will finish it up next week. > > .) That's why I asked how to know whos taking over which packet. > I had already done trac 0.11.4 butwas waiting on python, svn, and pysqlite > to settle ,) > > As you have the same Idea of making a trac-plugin-pack pls consider > these aswell : > > TracSubversionLocation-1.0.1 > GitPython-0.1.6 > accountmanagerplugin_0.11-r5499 > downloadsplugin-r5499 > httpauthplugin-r5728 > iniadminplugin_0.11 > privatewikiplugin-r5499 > reposearchplugin > > BTW : ticketdelete and ticketchange have been merged. did one of you manage to try out http://trac-hacks.org/wiki/TracWikiPrintPlugin which allows to print pdf's? my problem was, that i could not get reportlab to compile correctly with sunstudio. mike, you managed to compile some of the packages with gcc, isn't it? do you have an example so i could try this with reportlab as well? rupert. From ja at opencsw.org Sun May 24 11:44:23 2009 From: ja at opencsw.org (Juergen Arndt) Date: Sun, 24 May 2009 11:44:23 +0200 Subject: [csw-maintainers] Icinga 0.8 in testing Message-ID: Hi, Icinga 0.8 - a fork of Nagios - is now in testing. Still without the database backend IDOutils. Any feedback is welcome. Juergen -- Juergen Arndt From rupert at opencsw.org Sun May 24 21:32:44 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 24 May 2009 21:32:44 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> References: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> Message-ID: <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> i tried to compile, which worked on build8s. but on build8x and build10x it failed. what might be the reason for this? does sun have different compilers? /opt/studio/SOS11/SUNWspro/bin/cc -DNDEBUG -O -xO3 -xarch=386 -I/opt/csw/include -I/opt/csw/include -DRENDERPM_FT -DLIBART_COMPILATION -DLIBART_VERSION="2.3.12" -I/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM -I/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/libart_lgpl -I/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/gt1 -I/opt/csw/include -I/opt/csw/include/freetype2 -I/opt/csw/include/python2.6 -c /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c -o build/temp.solaris-2.8-i86pc-2.6/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.o "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 81: syntax error before or at: 13 "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 96: syntax error before or at: 13 "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 100: syntax error before or at: : "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 102: warning: statement not reached "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", line 105: cannot recover from previous errors cc: acomp failed for /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c error: command '/opt/studio/SOS11/SUNWspro/bin/cc' failed with exit status 2 gmake[1]: *** [install-custom] Error 1 gmake[1]: Leaving directory `/home/rupert/mgar/pkg/reportlab/trunk' gmake: *** [merge-isa-i386] Error 2 i also checked in the makefile, put the compiled 8s on testing with write access for everybody. rupert. On Sat, Apr 11, 2009 at 12:29, rupert THURNER wrote: > hi, > > for getting http://trac-hacks.org/wiki/TracWikiPrintPlugin we would > need http://reportlab.org/ v2.3 (http://www.reportlab.org/ftp/) . i > saw there is a package http://opencsw.org/packages/reportlab, v2.1. > > as it is not in > http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg, where would > be the source to try it? > > rupert. > From mwatters at opencsw.org Sun May 24 22:08:33 2009 From: mwatters at opencsw.org (Mike Watters) Date: Sun, 24 May 2009 15:08:33 -0500 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> References: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> Message-ID: <4A19A941.6070306@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 rupert THURNER wrote: > i tried to compile, which worked on build8s. but on build8x and > build10x it failed. what might be the reason for this? does sun have > different compilers? > -I/opt/csw/include/python2.6 -c > /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c > -o build/temp.solaris-2.8-i86pc-2.6/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.o > "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", > line 81: syntax error before or at: 13 > "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", > line 96: syntax error before or at: 13 > "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", > line 100: syntax error before or at: : The output of make has clues for you to look at. note the syntax errors. I assume there are header differences or library differences between the 2 systems, this is not uncommon. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkoZqUAACgkQLrhmsXMSLxdAOgCgpvHpo9ZMcH1sEiyPaCC0Hl/L fmgAn0zyLu1J99+7uM35f54R38sxNppP =1/GN -----END PGP SIGNATURE----- From ihsan at opencsw.org Sun May 24 22:31:10 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 24 May 2009 22:31:10 +0200 Subject: [csw-maintainers] Packages that require either package A or package B In-Reply-To: <1242926828-sup-5777@ntdws12.chass.utoronto.ca> References: <4A14A466.5010602@opencsw.org> <20090521005327.GA15461@bolthole.com> <1242868338-sup-2474@ntdws12.chass.utoronto.ca> <20090521160833.GA89074@bolthole.com> <9242740A-F9EE-4921-A031-BE21E37EAF13@opencsw.org> <1242926828-sup-5777@ntdws12.chass.utoronto.ca> Message-ID: <4A19AE8E.2030306@opencsw.org> Am 21.5.2009 19:29 Uhr, Ben Walton schrieb: >> handle dependencies as we have always handled them. In addition, >> I propose to carefully start working on real IPS packaging >> instead of putting extra work/complexity in System V packaging. >> A real IPS backend for GAR with all bells and whistels for >> dual-personality building of packages is surely a Good Thing. > Yes, I agree. Will IPS make pkgutil/pkg-get redundant? Eg, will it > facilitate simply creating a 'repo' and adding it to the IPS conf > (similar to apt/yum)? If so, all the smarts could be punted to the > system tools instead, which would be nice. Is it now sure, that IPS is going to be in Solaris 11? Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From ihsan at opencsw.org Sun May 24 22:40:11 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Sun, 24 May 2009 22:40:11 +0200 Subject: [csw-maintainers] Commit #5000 today! In-Reply-To: References: Message-ID: <4A19B0AB.7030001@opencsw.org> Am 22.5.2009 10:47 Uhr, Peter FELECAN schrieb: > I hope that the next time we'll taste the good German cakes --- have a > feeble for the schwarzwald kirche torte and the Bavarian ecosystem. As far the Schwarzwald Region is just across the border, I could bring a Schwarzw?lder Kirschtorte to Norway. :-) PS: There is also an excellent Beer from the Schwarzwald region: http://www.waldhaus-bier.de/ Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From dam at opencsw.org Sun May 24 23:14:08 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 24 May 2009 23:14:08 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <4A19A941.6070306@opencsw.org> References: <6af4270904110329q3a5f0b6an9bf797ba501d06fb@mail.gmail.com> <6af4270905241232v41ea3e91rfdc50d1a6f36d00e@mail.gmail.com> <4A19A941.6070306@opencsw.org> Message-ID: Hi, Am 24.05.2009 um 22:08 schrieb Mike Watters: > rupert THURNER wrote: >> i tried to compile, which worked on build8s. but on build8x and >> build10x it failed. what might be the reason for this? does sun have >> different compilers? > >> -I/opt/csw/include/python2.6 -c >> /home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c >> -o build/temp.solaris-2.8-i86pc-2.6/home/rupert/mgar/pkg/reportlab/ >> trunk/work/build-isa-i386/ReportLab_2_3/src/rl_addons/renderPM/ >> _renderPM.o >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", >> line 81: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", >> line 96: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ >> ReportLab_2_3/src/rl_addons/renderPM/_renderPM.c", >> line 100: syntax error before or at: : > > The output of make has clues for you to look at. note the syntax > errors. > I assume there are header differences or library differences between > the 2 > systems, this is not uncommon. Or x86 assembler optimizations Sun Studio doesn't understand. You can try disabling these during configure or use gcc instead. Best regards -- Dago From Murray.Jensen at csiro.au Mon May 25 03:04:35 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Mon, 25 May 2009 11:04:35 +1000 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: Your message of "Mon, 25 May 2009 06:08:33 +1000" <4A19A941.6070306@opencsw.org> Message-ID: <10840.1243213475@gerd> On Mon, 25 May 2009 06:08:33 +1000, Mike Watters writes: >rupert THURNER wrote: >> i tried to compile, which worked on build8s. but on build8x and >> build10x it failed. what might be the reason for this? does sun have >> different compilers? Just checked out my stuff and reportlab 2.1 built cleanly out of the box on solaris 8 for both intel and sparc using Sun Studio 8(?). >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >src/rl_addons/renderPM/_renderPM.c", >> line 81: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >src/rl_addons/renderPM/_renderPM.c", >> line 96: syntax error before or at: 13 >> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >src/rl_addons/renderPM/_renderPM.c", >> line 100: syntax error before or at: : > >The output of make has clues for you to look at. note the syntax errors. >I assume there are header differences or library differences between the 2 >systems, this is not uncommon. Ahh.. do you have renderPM installed and at the correct version for reportlab 2.3? Also, I used the rl_accel accelerator plugin - seemed like a good idea at the time. Cheers! Murray... From rupert at opencsw.org Mon May 25 15:58:37 2009 From: rupert at opencsw.org (rupert THURNER) Date: Mon, 25 May 2009 15:58:37 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <10840.1243213475@gerd> References: <4A19A941.6070306@opencsw.org> <10840.1243213475@gerd> Message-ID: <6af4270905250658u4e045f47v3e2a763400f14109@mail.gmail.com> On Mon, May 25, 2009 at 03:04, wrote: > On Mon, 25 May 2009 06:08:33 +1000, Mike Watters writes: >>rupert THURNER wrote: >>> i tried to compile, which worked on build8s. but on build8x and >>> build10x it failed. what might be the reason for this? does sun have >>> different compilers? > > Just checked out my stuff and reportlab 2.1 built cleanly out of the box on > solaris 8 for both intel and sparc using Sun Studio 8(?). > >>> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >>src/rl_addons/renderPM/_renderPM.c", >>> line 81: syntax error before or at: 13 >>> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >>src/rl_addons/renderPM/_renderPM.c", >>> line 96: syntax error before or at: 13 >>> "/home/rupert/mgar/pkg/reportlab/trunk/work/build-isa-i386/ReportLab_2_3/= >>src/rl_addons/renderPM/_renderPM.c", >>> line 100: syntax error before or at: : >> >>The output of make has clues for you to look at. note the syntax errors. >>I assume there are header differences or library differences between the 2 >>systems, this is not uncommon. > > Ahh.. do you have renderPM installed and at the correct version for reportlab > 2.3? Also, I used the rl_accel accelerator plugin - seemed like a good idea > at the time. Cheers! i tried to build it on the standard build servers. how did you install it?or could i have access to your build file? rupert. From bonivart at opencsw.org Mon May 25 19:11:14 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Mon, 25 May 2009 19:11:14 +0200 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request In-Reply-To: <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> References: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> Message-ID: <625385e30905251011ta6ae38ar5922911e22412ac7@mail.gmail.com> On Wed, May 20, 2009 at 10:46 PM, Peter Bonivart wrote: > On Wed, May 20, 2009 at 6:05 PM, Ben Walton wrote: >> Would you consider adding a -C option to the package management >> utilities to at would limit the compare output to only versions behind >> the catalog (eg: `pkg-get -U -c | grep -v SAME` is what I use to see >> what will be updated). > > Sure, I can add it to 1.6 which is almost done. I never implemented > showing not installed packages but I still get annoyed by using grep > -v SAME so it's a good idea. There's now a beta in testing: http://mirror.opencsw.org/testing/pkgutil-1.6b1,REV=2009.05.25-SunOS5.8-i386-CSW.pkg.gz http://mirror.opencsw.org/testing/pkgutil-1.6b1,REV=2009.05.25-SunOS5.8-sparc-CSW.pkg.gz Change log: - New option -C, same as -c but only shows different versions, no more piping to grep -v SAME (Ben Walton) - New option --trace to debug pkgadd/pkgrm operations (Yann Rouillard) - Replaced some system commands with native Perl - New option --config for using custom configuration file (Don Jackson) - Fix size formatting - Skip showing pkgadd patch when using -V on 5.11 (Dennis Clarke) - Update credits in perldoc/man page - Perldoc/man page added to both bldcat and chkcat - Chkcat now includes more tests (Derek Whayman) - Chkcat exit values are 0 for OK, 1 for warning and 2 for error - New option -e (erroronly) added to chkcat to supress warnings - New option -q (quiet) added to chkcat to supress all output -- /peter From dam at opencsw.org Mon May 25 22:10:48 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 25 May 2009 22:10:48 +0200 Subject: [csw-maintainers] What to do with Nessus? Message-ID: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> Hi, as I browsed the collection I noticed that we have a very ancient version of Nessus: However, Nessus is no longer be free. As it is a security scanner it is most likely no useful any more, so I propose to drop it and the associated packages nessuslib, nessusplugins and libnasl. Best regards -- Dago From phil at bolthole.com Mon May 25 23:55:06 2009 From: phil at bolthole.com (Philip Brown) Date: Mon, 25 May 2009 14:55:06 -0700 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> Message-ID: <20090525215506.GA81042@bolthole.com> On Mon, May 25, 2009 at 10:10:48PM +0200, Dagobert Michelsen wrote: > Hi, > > as I browsed the collection I noticed that we have a very ancient > version of Nessus: > > However, Nessus is no longer be free. errr... its semi-free. but anyways, i think personally, that we should upgrade it to the last truely-free version, if we can From Murray.Jensen at csiro.au Tue May 26 02:32:23 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Tue, 26 May 2009 10:32:23 +1000 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: Your message of "Mon, 25 May 2009 23:58:37 +1000" <6af4270905250658u4e045f47v3e2a763400f14109@mail.gmail.com> Message-ID: <14077.1243297943@gerd> On Mon, 25 May 2009 23:58:37 +1000, rupert THURNER writes: >i tried to build it on the standard build servers. how did you install >it? I have my own "build servers". You will have to request that renderPM be installed on the central "build servers". But first check that the version of renderPM in CSW is late enough for reportlab 2.3 - you may need to build and submit renderPM first, then get it installed on the "build servers", then move on to building reportlab. Also don't forget to check out the rl_accel plugin for reportlab. >or could i have access to your build file? My build file basically is this: python setup.py build mkdir -p ${buildroot} python setup.py install --root=${buildroot} chown -Rh root:bin ${buildroot}/opt i.e. the standard way to build and install python packages (I believe). GAR will probably have mechanisms for this by default(?). Cheers! Murray... From bwalton at opencsw.org Tue May 26 02:37:57 2009 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 25 May 2009 20:37:57 -0400 Subject: [csw-maintainers] pkg-get/pkgutil creature comfort request In-Reply-To: <625385e30905251011ta6ae38ar5922911e22412ac7@mail.gmail.com> References: <1242835410-sup-4669@ntdws12.chass.utoronto.ca> <625385e30905201346l685f2617o969bacb1bc031698@mail.gmail.com> <625385e30905251011ta6ae38ar5922911e22412ac7@mail.gmail.com> Message-ID: <1243298243-sup-8147@ntdws12.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Mon May 25 13:11:14 -0400 2009: Hi Peter, > - New option -C, same as -c but only shows different versions, no more > piping to grep -v SAME (Ben Walton) I tried this out just now. It looks great from here! Thanks for adding that so quickly. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From Darin.Perusich at cognigencorp.com Tue May 26 14:01:21 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Tue, 26 May 2009 08:01:21 -0400 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> Message-ID: <4A1BDA11.4040308@cognigencorp.com> There's always OpenVAS which is a fork of Nessus. Dagobert Michelsen wrote: > Hi, > > as I browsed the collection I noticed that we have a very ancient > version of Nessus: > > However, Nessus is no longer be free. As it is a security scanner it > is most likely no useful any more, so I propose to drop it and the > associated packages nessuslib, nessusplugins and libnasl. > -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From ihsan at opencsw.org Tue May 26 15:29:10 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 26 May 2009 15:29:10 +0200 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <4A1BDA11.4040308@cognigencorp.com> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> Message-ID: <4A1BEEA6.70707@opencsw.org> I think a transition to OpenVAS would make sense then. Am 26.5.2009 14:01 Uhr, Darin Perusich schrieb: > There's always OpenVAS which is a fork of Nessus. > > Dagobert Michelsen wrote: >> Hi, >> >> as I browsed the collection I noticed that we have a very ancient >> version of Nessus: >> >> However, Nessus is no longer be free. As it is a security scanner it >> is most likely no useful any more, so I propose to drop it and the >> associated packages nessuslib, nessusplugins and libnasl. >> > > -- ihsan at dogan.ch http://blog.dogan.ch/ From skayser at opencsw.org Tue May 26 16:07:44 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 26 May 2009 16:07:44 +0200 (CEST) Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? Message-ID: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> Hi, i have some machines that are not connected to the internet, where i would like to install some OpenCSW packages. What would be the best way to download these packages (inluding dependencies) for offline installation? I don't want to pull the whole mirror, just the packages and their deps. I have thought about setting up a full root zone w/o CSW packages and do a pkg-get/pkgutil -d there, but i would prefer if there was an easier way. Sebastian From dam at opencsw.org Tue May 26 17:06:19 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 17:06:19 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space Message-ID: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> Hi, I am currently setting up build10st and need some more space for it and noticed that Hudson is taking up around 25 GB of space. Looking into /var/tmp is, umh, impressing... > root at ncsw [global]:/zones/hudson/root/var/tmp > time ls -1 | wc -l > 824725 > ls -1 8.70s user 4.23s system 18% cpu 1:08.11 total > wc -l 0.32s user 0.02s system 0% cpu 1:07.86 total Trygve, could you please set up some kind of cleanup job? I'm going to delete the temp stuff for now manually. Best regards -- Dago From phil at bolthole.com Tue May 26 17:27:46 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 26 May 2009 08:27:46 -0700 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <4A1BEEA6.70707@opencsw.org> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> <4A1BEEA6.70707@opencsw.org> Message-ID: <20090526152746.GB17249@bolthole.com> On Tue, May 26, 2009 at 03:29:10PM +0200, Ihsan Dogan wrote: > I think a transition to OpenVAS would make sense then. > can you confirm or deny whether openvas covers all the free-nessus features? From phil at bolthole.com Tue May 26 17:28:38 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 26 May 2009 08:28:38 -0700 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <4A1BDA11.4040308@cognigencorp.com> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> Message-ID: <20090526152838.GC17249@bolthole.com> On Tue, May 26, 2009 at 08:01:21AM -0400, Darin Perusich wrote: > There's always OpenVAS which is a fork of Nessus. > oh, i didnt read this message before sending my other one :-) yes, this makes sense to me. Someone could package that up, and declare conflict with CSWnessus, if they liked, and then i would remove nessus From mwatters at opencsw.org Tue May 26 17:54:06 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 26 May 2009 10:54:06 -0500 Subject: [csw-maintainers] anyone updating openldap? Message-ID: <4A1C109E.10904@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I need amd64 libs for openldap, is anyone working on this package? from the List it is owned by Alex, who is retired. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkocEJ4ACgkQLrhmsXMSLxcFFQCgk2lUfaMIzcjRBW+fsxNvySoD BdAAni1F1ax1p7cTFn6syiMEs4o8MWsk =toQQ -----END PGP SIGNATURE----- From dam at opencsw.org Tue May 26 18:00:38 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 18:00:38 +0200 Subject: [csw-maintainers] anyone updating openldap? In-Reply-To: <4A1C109E.10904@opencsw.org> References: <4A1C109E.10904@opencsw.org> Message-ID: Hi Mike, Am 26.05.2009 um 17:54 schrieb Mike Watters: > I need amd64 libs for openldap, is anyone working on this package? > > from the List it is owned by Alex, who is retired. I made a Makefile for it some time ago but didn't finish the package. The start/stopp mechanism is still missing. If you add them and take a look at the patches I update it to mGAR v2. Best regards -- Dago From phil at bolthole.com Tue May 26 18:14:51 2009 From: phil at bolthole.com (Philip Brown) Date: Tue, 26 May 2009 09:14:51 -0700 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> Message-ID: <20090526161451.GE17249@bolthole.com> On Tue, May 26, 2009 at 04:07:44PM +0200, Sebastian Kayser wrote: > Hi, > > i have some machines that are not connected to the internet, where i would > like to install some OpenCSW packages. What would be the best way to > download these packages (inluding dependencies) for offline installation? > I don't want to pull the whole mirror, just the packages and their deps. > > I have thought about setting up a full root zone w/o CSW packages and do a > pkg-get/pkgutil -d there, but i would prefer if there was an easier way. That's about the least annoying method i can think of, unfortunately. slightly improved by the fact that, once you have the ones you want downloaded, you can also download the catalog to that directory, and then use the usual tools to that dir. From mwatters at opencsw.org Tue May 26 18:20:54 2009 From: mwatters at opencsw.org (Mike Watters) Date: Tue, 26 May 2009 11:20:54 -0500 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> Message-ID: <4A1C16E6.2040005@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Sebastian Kayser wrote: > Hi, > > i have some machines that are not connected to the internet, where i would > like to install some OpenCSW packages. What would be the best way to > download these packages (inluding dependencies) for offline installation? > I don't want to pull the whole mirror, just the packages and their deps. > > I have thought about setting up a full root zone w/o CSW packages and do a > pkg-get/pkgutil -d there, but i would prefer if there was an easier way. I assume they have network access internal, if you have a system that can connect to the internet, you could setup a local mirror and you could update all of your internal systems using that mirror. use rsync and update as often as you feel appropriate. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkocFuYACgkQLrhmsXMSLxezfgCfb7YUxfss3+NPx98Cs/t3LtUQ akYAn0g7LNcUHDbVlLxCZF8RRu15niOI =1TM3 -----END PGP SIGNATURE----- From skayser at opencsw.org Tue May 26 21:47:33 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 26 May 2009 21:47:33 +0200 (CEST) Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <4A1C16E6.2040005@opencsw.org> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> Message-ID: <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Mike Watters wrote: > Sebastian Kayser wrote: >> i have some machines that are not connected to the internet, where i >> would >> like to install some OpenCSW packages. What would be the best way to >> download these packages (inluding dependencies) for offline >> installation? >> I don't want to pull the whole mirror, just the packages and their deps. >> >> I have thought about setting up a full root zone w/o CSW packages and do >> a >> pkg-get/pkgutil -d there, but i would prefer if there was an easier way. > > I assume they have network access internal, if you have a system that can > connect to the internet, you could setup a local mirror and you could > update > all of your internal systems using that mirror. use rsync and update as > often > as you feel appropriate. There is no machine on this net with internet access ... restricted environment. While i am still a bit reluctant to sync the whole mirror, there is a workaround using pkgutil. PKG_INSTALL_ROOT=/dev/null pkgutil -d This will make pkgparam which is invoked by pkgutil to check for installed packages return an error and thus pkgutil will think that none of the required/requested packages are installed. I end up with all required pkgs and can transfer them via a USB stick/disk. :D Btw., what's the disk usage of our current/ tree? Sebastian From dam at opencsw.org Tue May 26 22:08:21 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 22:08:21 +0200 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Message-ID: Hi Sebastian, Am 26.05.2009 um 21:47 schrieb Sebastian Kayser: > There is no machine on this net with internet access ... restricted > environment. While i am still a bit reluctant to sync the whole > mirror, > there is a workaround using pkgutil. > > PKG_INSTALL_ROOT=/dev/null pkgutil -d Why don't you use this option? -s, --stream Build a package stream for a certain architecture and OS release. All dependencies will be included in the stream and the needed command to install them in the correct order will be displayed. I requested it some time ago for this exact prupose. > Btw., what's the disk usage of our current/ tree? dam at login [login]:/export/mirror/opencsw/current/sparc/5.8 > du -sh . 4.1G . Feel free to check yourself, it is mounted in @login Best regards -- Dago From william at wbonnet.net Tue May 26 22:31:06 2009 From: william at wbonnet.net (William Bonnet) Date: Tue, 26 May 2009 22:31:06 +0200 Subject: [csw-maintainers] pixman 0.15.6 is in testing Message-ID: <4A1C518A.6090503@wbonnet.net> Hi Pixman package has been updated to latest version (0.15.6) and is available from testing Any feedback is welcome kind regards, W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From bonivart at opencsw.org Tue May 26 23:09:58 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 26 May 2009 23:09:58 +0200 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Message-ID: <625385e30905261409o62054c33re72cfdf9780513d2@mail.gmail.com> On Tue, May 26, 2009 at 10:08 PM, Dagobert Michelsen wrote: > Why don't you use this option? > > ? ? -s, --stream > ? ? ? ? ?Build a package stream for a certain architecture and > ? ? ? ? ?OS release. All dependencies will be included in the > ? ? ? ? ?stream and the needed command to install them in the > ? ? ? ? ?correct order will be displayed. > > I requested it some time ago for this exact prupose. Thanks for pointing that out Dago. Here's a couple of examples: http://pkgutil.wikidot.com/use-with-examples#toc7 But basically, if you want a stream with all deps for bind and dhcp for example you just do this: # pkgutil -s bind dhcp -- /peter From trygvis at opencsw.org Tue May 26 23:20:19 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Tue, 26 May 2009 23:20:19 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space In-Reply-To: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> References: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> Message-ID: <4A1C5D13.3040500@opencsw.org> Dagobert Michelsen wrote: > Hi, > > I am currently setting up build10st and need some more space for it and > noticed that Hudson is taking up around 25 GB of space. Looking into > /var/tmp is, umh, impressing... > >> root at ncsw [global]:/zones/hudson/root/var/tmp > time ls -1 | wc -l >> 824725 >> ls -1 8.70s user 4.23s system 18% cpu 1:08.11 total >> wc -l 0.32s user 0.02s system 0% cpu 1:07.86 total > > > Trygve, could you please set up some kind of cleanup job? I'm going to > delete the temp stuff for now manually. What kind of files are in /var/tmp? I would be surprised if hudson itself was leaking files there. -- Trygve From dam at opencsw.org Tue May 26 23:24:34 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 26 May 2009 23:24:34 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space In-Reply-To: <4A1C5D13.3040500@opencsw.org> References: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> <4A1C5D13.3040500@opencsw.org> Message-ID: <92E1FF20-8F6C-4E70-9D1F-D33C315966AF@opencsw.org> Hi Trygve, Am 26.05.2009 um 23:20 schrieb Trygve Laugst?l: > What kind of files are in /var/tmp? I would be surprised if hudson > itself was leaking files there. Like this: root at ncsw [global]:/zones/hudson/root/var/tmp > find aaazgaqzC aaazgaqzC aaazgaqzC/CSWtomcat4 aaazgaqzC/CSWtomcat4/pkgmap aaazgaqzC/CSWtomcat4/pkginfo Best regards -- Dago From trygvis at opencsw.org Tue May 26 23:26:12 2009 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Tue, 26 May 2009 23:26:12 +0200 Subject: [csw-maintainers] Hudson taking up a lot of space In-Reply-To: <92E1FF20-8F6C-4E70-9D1F-D33C315966AF@opencsw.org> References: <7A941ACF-73F1-48D7-BD43-5E97B4772DA4@opencsw.org> <4A1C5D13.3040500@opencsw.org> <92E1FF20-8F6C-4E70-9D1F-D33C315966AF@opencsw.org> Message-ID: <4A1C5E74.9040302@opencsw.org> Dagobert Michelsen wrote: > Hi Trygve, > > Am 26.05.2009 um 23:20 schrieb Trygve Laugst?l: >> What kind of files are in /var/tmp? I would be surprised if hudson >> itself was leaking files there. > > Like this: > > root at ncsw [global]:/zones/hudson/root/var/tmp > find aaazgaqzC > aaazgaqzC > aaazgaqzC/CSWtomcat4 > aaazgaqzC/CSWtomcat4/pkgmap > aaazgaqzC/CSWtomcat4/pkginfo Looks like something checkpkg would generate from what I can remember, but I might be wrong. -- Trygve From ihsan at opencsw.org Tue May 26 23:39:12 2009 From: ihsan at opencsw.org (Ihsan Dogan) Date: Tue, 26 May 2009 23:39:12 +0200 Subject: [csw-maintainers] What to do with Nessus? In-Reply-To: <20090526152746.GB17249@bolthole.com> References: <7B155F0C-620C-4064-849B-126A2E9B5EDF@opencsw.org> <4A1BDA11.4040308@cognigencorp.com> <4A1BEEA6.70707@opencsw.org> <20090526152746.GB17249@bolthole.com> Message-ID: <4A1C6180.8020105@opencsw.org> Am 26.5.2009 17:27 Uhr, Philip Brown schrieb: >> I think a transition to OpenVAS would make sense then. > > can you confirm or deny whether openvas covers all the free-nessus > features? I can't. But if so, I think it's worse to think about. Ihsan -- ihsan at dogan.ch http://blog.dogan.ch/ From skayser at opencsw.org Tue May 26 23:49:17 2009 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 26 May 2009 23:49:17 +0200 Subject: [csw-maintainers] pkg-get/pkgutil: Download full dependency tree for offline installation? In-Reply-To: References: <56205.194.246.122.22.1243346864.squirrel@ssl.skayser.de> <4A1C16E6.2040005@opencsw.org> <63346.194.246.122.22.1243367253.squirrel@ssl.skayser.de> Message-ID: <4A1C63DD.2080100@opencsw.org> Dagobert Michelsen wrote: > Am 26.05.2009 um 21:47 schrieb Sebastian Kayser: >> There is no machine on this net with internet access ... restricted >> environment. While i am still a bit reluctant to sync the whole >> mirror, >> there is a workaround using pkgutil. >> >> PKG_INSTALL_ROOT=/dev/null pkgutil -d > > Why don't you use this option? > > -s, --stream > Build a package stream for a certain architecture and > OS release. All dependencies will be included in the > stream and the needed command to install them in the > correct order will be displayed. > > I requested it some time ago for this exact prupose. Cool, thanks for the pointer. Will give it a try tomorrow. Sebastian From pfelecan at opencsw.org Wed May 27 11:06:26 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 27 May 2009 11:06:26 +0200 Subject: [csw-maintainers] New schily tools in testing In-Reply-To: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> (Joerg Schilling's message of "Fri\, 15 May 2009 19\:37\:44 +0200") References: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: > I just compiled a new set of packages and put them into testing. Thanks. Do you need more testing or are they ready to be released? It would be really nice to have up to date version of your tools in our catalog. -- Peter From pfelecan at opencsw.org Wed May 27 13:55:09 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Wed, 27 May 2009 13:55:09 +0200 Subject: [csw-maintainers] libgsf is or isn't orphaned Message-ID: Looking to libgsf package I see that Robin Kay is the maintainer and he's retired. However, libgsf is not marked as orphaned. What's the real status of this package? The packaged release is so old that any new dependant of which I'm aware cannot be linked with. -- Peter From dam at opencsw.org Wed May 27 14:06:18 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 14:06:18 +0200 Subject: [csw-maintainers] libgsf is or isn't orphaned In-Reply-To: References: Message-ID: <2A9DAF13-94AE-4FF7-8579-F2278293E4D2@opencsw.org> Hi Peter, Am 27.05.2009 um 13:55 schrieb Peter FELECAN: > Looking to libgsf package I see that Robin Kay is the maintainer and > he's retired. However, libgsf is not marked as orphaned. What's the > real > status of this package? The packaged release is so old that any new > dependant of which I'm aware cannot be linked with. Roger worked on it, but had troubles: Am 10.04.2009 um 02:19 schrieb Roger H?kansson: > Dagobert Michelsen wrote: >> I just build it from SVN without any problems, maybe the LD_OPTIONS >> change was >> more good than expected? >> Anyway, libgsf is now in testing: >> libgsf-1.14.11,REV=2009.04.09-SunOS5.8-i386-CSW.pkg.gz >> libgsf-1.14.11,REV=2009.04.09-SunOS5.8-sparc-CSW.pkg.gz > > Maybe I was a bit unclear, the package will build just fine, however > libgsf-gnome-1.so isn't linked to the libgsf-1.so that the build > produce, but the old one in /opt/csw/lib. > And I've been trying to fix this in various ways but with no luck. > So far the only working solution I've found is to remove libgsf from > the build machine when packaging. > > hson at build8x :~ > dump -Lv /home/dam/mgar/pkg/libgsf/trunk/work/ > pkgroot/opt/csw/lib/libgsf-gnome-1.so.114.0.11 > ... > [1] NEEDED libgsf-1.so.1 > ... > [16] SONAME libgsf-gnome-1.so.114 > > >>> I've also been trying to build libgda and libgnomedb (which are >>> optional for gnumeric), but there seems to be some incompatibility >>> from upstream so libgnomedb won't built atm. >> Nope, no incompatibility here. Builds absolutely smooth. Packages >> in testing/ now. >> Thanks for the good work! > > Hmm, I get a bunch of "undefined struct/union member" when building > libgnomedb on my build machine, but I've installed the libgda I > packaged, I guess the buildfarm have the older one? > Otherwise its probably some other package that is missing from my > build machine which is installed on the build farm... Feel free to give him a hand and take a look also. Best regards -- Dago From maciej at opencsw.org Wed May 27 16:55:10 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 27 May 2009 15:55:10 +0100 Subject: [csw-maintainers] findutils and its database on the farm Message-ID: Would it be possible to get findutils (glocate) installed on the farm and have the file database up to date? The current state is: build8s -- no findutils build8x -- no findutils build10s -- findutils installed, database missing (need to run gupdatedb?) build10x -- no findutils Maciej From phil at bolthole.com Wed May 27 18:38:08 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 09:38:08 -0700 Subject: [csw-maintainers] libgsf is or isn't orphaned In-Reply-To: References: Message-ID: <20090527163808.GH55163@bolthole.com> On Wed, May 27, 2009 at 01:55:09PM +0200, Peter FELECAN wrote: > Looking to libgsf package I see that Robin Kay is the maintainer and > he's retired. However, libgsf is not marked as orphaned. What's the real > status of this package? The packaged release is so old that any new > dependant of which I'm aware cannot be linked with. Any package whose maintainer is marked as "retired", is open to adoption by another maintainer. From mwatters at opencsw.org Wed May 27 19:01:51 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 12:01:51 -0500 Subject: [csw-maintainers] new class scripts Message-ID: <4A1D71FF.9080806@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I have a new set of class scripts i.pycompile and r.pycompile I was wondering how best to install them. add them to cswclassutils, have the python package add/delete them to/from /usr/sadm/install/scripts ? - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodcf8ACgkQLrhmsXMSLxd1twCeMlq/8ayvma1y+3Sje4MOVrfU 9asAoKH4KfEsDbJFWO3uDQahOZcSZ7j0 =xy43 -----END PGP SIGNATURE----- From bwalton at opencsw.org Wed May 27 19:10:22 2009 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 27 May 2009 13:10:22 -0400 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1D71FF.9080806@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> Message-ID: <1243444160-sup-7077@ntdws12.chass.utoronto.ca> Excerpts from Mike Watters's message of Wed May 27 13:01:51 -0400 2009: > I have a new set of class scripts i.pycompile and r.pycompile I > was wondering how best to install them. add them to cswclassutils, > have the python package add/delete them to/from > /usr/sadm/install/scripts ? I think that for consistency, they should be added to cswclassutils. -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 GPG Key Id: 8E89F6D2; Key Server: pgp.mit.edu Contact me to arrange for a CAcert assurance meeting. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From phil at bolthole.com Wed May 27 19:16:04 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 10:16:04 -0700 Subject: [csw-maintainers] new class scripts In-Reply-To: <1243444160-sup-7077@ntdws12.chass.utoronto.ca> References: <4A1D71FF.9080806@opencsw.org> <1243444160-sup-7077@ntdws12.chass.utoronto.ca> Message-ID: <20090527171604.GP55163@bolthole.com> On Wed, May 27, 2009 at 01:10:22PM -0400, Ben Walton wrote: > Excerpts from Mike Watters's message of Wed May 27 13:01:51 -0400 2009: > > I have a new set of class scripts i.pycompile and r.pycompile I > > was wondering how best to install them. add them to cswclassutils, > > have the python package add/delete them to/from > > /usr/sadm/install/scripts ? > > I think that for consistency, they should be added to cswclassutils. yup. From bonivart at opencsw.org Wed May 27 19:31:53 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 27 May 2009 19:31:53 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1D71FF.9080806@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> Message-ID: <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> On Wed, May 27, 2009 at 7:01 PM, Mike Watters wrote: > ? ? ?I have a new set of class scripts i.pycompile and r.pycompile > I was wondering how best to install them. ?add them to cswclassutils, have the > python package add/delete them to/from /usr/sadm/install/scripts ? Make them available to me somehow and I will add them to cswclassutils. I will just rename them to i|r.cswpycompile for consistency. -- /peter From mwatters at opencsw.org Wed May 27 19:50:38 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 12:50:38 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> Message-ID: <4A1D7D6E.7000709@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Bonivart wrote: > On Wed, May 27, 2009 at 7:01 PM, Mike Watters wrote: >> I have a new set of class scripts i.pycompile and r.pycompile >> I was wondering how best to install them. add them to cswclassutils, have the >> python package add/delete them to/from /usr/sadm/install/scripts ? > > Make them available to me somehow and I will add them to > cswclassutils. I will just rename them to i|r.cswpycompile for > consistency. > They are checked into subversion under the python package http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile Addition to the README.CSW =========================== - - pycompile This class takes care of creating/removing python's normal and optimized bytecode files. Set all *.py files' class = pycompile to activate. Example: # use prototype filters to set the class PROTOTYPE_FILTER = awk '$$$$3 ~/.*\.py$$$$/ { $$$$2 = "pycompile" } { print }' - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodfW4ACgkQLrhmsXMSLxcK9wCgkC1CufESdEt+S+mhbLohh8vN JIIAnjJ0b4VOlKkR9ymSKdcCQ9EAv+It =S3cZ -----END PGP SIGNATURE----- From bonivart at opencsw.org Wed May 27 20:01:09 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 27 May 2009 20:01:09 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1D7D6E.7000709@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> Message-ID: <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> On Wed, May 27, 2009 at 7:50 PM, Mike Watters wrote: > They are checked into subversion under the python package > > http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile > http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile > > Addition to the README.CSW > =========================== > > - - pycompile > ?This class takes care of creating/removing python's normal and optimized > ?bytecode files. ?Set all *.py files' class = pycompile to activate. > Example: > # use prototype filters to set the class > PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "pycompile" } { print }' I'll add them tomorrow, thanks for the readme text. I updated the wiki page with it to start with, feel free to add/edit. http://wiki.opencsw.org/cswclassutils-package -- /peter From maciej at opencsw.org Wed May 27 20:13:55 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 27 May 2009 19:13:55 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package Message-ID: I'm playing with updating syslog_ng to v2 and getting it to have SMF support via mGAR. Makefile: http://dpaste.com/48317/ SMF manifest: http://dpaste.com/48320/ The changes I made can be summarized as: - added DISTFILES += cswsyslog_ng.xml - installed cswsyslog_ng as svc-cswsyslog_ng in the manifest directory ($(DESTDIR)$(libdir)/svc/method) - added INITSMF = /opt/csw/etc/init.d/cswsyslog_ng The result is that the SMF manifest doesn't get imported: Installation of on zone was successful. root at joy:/var/tmp# svcs -a | grep syslog root at joy:/var/tmp# How does mGAR know which file is the manifest? Just by its name -- cswfoo.xml? Maciej From phil at bolthole.com Wed May 27 20:47:57 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 11:47:57 -0700 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: References: Message-ID: <20090527184757.GU55163@bolthole.com> On Wed, May 27, 2009 at 07:13:55PM +0100, Maciej (Matchek) Blizinski wrote: > I'm playing with updating syslog_ng to v2 and getting it to have SMF > support via mGAR. > .... > The result is that the SMF manifest doesn't get imported: better to use cswclassutils for SMF support anyways, rather than a custom pkg-local manifest, unless you're doing something Really Special. From bonivart at opencsw.org Wed May 27 21:17:03 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 27 May 2009 21:17:03 +0200 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <20090527184757.GU55163@bolthole.com> References: <20090527184757.GU55163@bolthole.com> Message-ID: <625385e30905271217q4a62178aq118e396fbfe34045@mail.gmail.com> On Wed, May 27, 2009 at 8:47 PM, Philip Brown wrote: > better to use cswclassutils for SMF support anyways, rather than a custom > pkg-local manifest, unless you're doing something Really Special. I'm looking into supporting custom manifests so that should be covered soon. -- /peter From phil at bolthole.com Wed May 27 21:24:41 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 12:24:41 -0700 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <625385e30905271217q4a62178aq118e396fbfe34045@mail.gmail.com> References: <20090527184757.GU55163@bolthole.com> <625385e30905271217q4a62178aq118e396fbfe34045@mail.gmail.com> Message-ID: <20090527192441.GV55163@bolthole.com> On Wed, May 27, 2009 at 09:17:03PM +0200, Peter Bonivart wrote: > On Wed, May 27, 2009 at 8:47 PM, Philip Brown wrote: > > better to use cswclassutils for SMF support anyways, rather than a custom > > pkg-local manifest, unless you're doing something Really Special. > > I'm looking into supporting custom manifests so that should be covered soon. coool. i would suggest just another commented-variable in the init.d file. ie: maintainer provides [/etc/opt/csw/]init.d/cswmyservice [/etc/opt/csw/]/cswmyservice.xml and uses # CUSTOMMANIFEST=/path/to/cswmyservice.xml in the first file, which woudl be the only one in class cswinitsmf From Joerg.Schilling at fokus.fraunhofer.de Wed May 27 22:17:13 2009 From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) Date: Wed, 27 May 2009 22:17:13 +0200 Subject: [csw-maintainers] New schily tools in testing In-Reply-To: References: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <4a1d9fc9.TCJ3FSsGMtwXUKLX%Joerg.Schilling@fokus.fraunhofer.de> Peter FELECAN wrote: > Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: > > > I just compiled a new set of packages and put them into testing. > > Thanks. Do you need more testing or are they ready to be released? It > would be really nice to have up to date version of your tools in our > catalog. This is ready to be released. J?rg -- EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin js at cs.tu-berlin.de (uni) joerg.schilling at fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily From ellson at opencsw.org Wed May 27 20:25:50 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 27 May 2009 14:25:50 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker Message-ID: <4A1D85AE.4060006@opencsw.org> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and links without problem. If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" from the Makefile) then I get this error: /opt/studio/SOS11/SUNWspro/bin/CC -KPIC CMakeFiles/example0.dir/MissingGlyphExample.o -o example0 -L/opt/csw/lib -L/home/ellson/mgar/pkg/liblasi/trunk/work/build-isa-sparcv8/libLASi-1.1.0/src -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lpangoft2-1.0 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lfreetype -lz ../src/libLASi.so.0.0.1 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lpangoft2-1.0 -lpango-1.0 -lgobject-2.0 -lgmodule-2.0 -lglib-2.0 -lintl -lpangoft2-1.0 -lfreetype -lz -lm -R/opt/csw/lib:/home/ellson/mgar/pkg/liblasi/trunk/work/build-isa-sparcv8/libLASi-1.1.0/src Undefined first referenced symbol in file void LASi::PostscriptDocument::setFont(const char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::FontStretch) CMakeFiles/example0.dir/MissingGlyphExample.o ld: fatal: Symbol referencing errors. No output written to example0 There's something odd about the "const char*const" but I don't know if thats related. Does anyone have any idea what the problem is, and how to fix it, or work around it so as to be able to use Sun's tools? John From dam at opencsw.org Wed May 27 23:12:33 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:12:33 +0200 Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> Message-ID: <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> Hi, Am 27.05.2009 um 20:01 schrieb Ben Walton: > Excerpts from Maciej (Matchek) Blizinski's message of Wed May 27 > 10:55:10 -0400 2009: >> Would it be possible to get findutils (glocate) installed on the farm >> and have the file database up to date? > > Doing this now. Is this package current in terms of updating cron, etc.? Additionally, it has security issues. Chris: As you already did findutils with glocate, it would be really cool if you could also package up slocate as more secure alternative. Best regards -- Dago From william at wbonnet.net Wed May 27 23:22:57 2009 From: william at wbonnet.net (William Bonnet) Date: Wed, 27 May 2009 23:22:57 +0200 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1D85AE.4060006@opencsw.org> References: <4A1D85AE.4060006@opencsw.org> Message-ID: <4A1DAF31.7040109@wbonnet.net> Hi John > If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles > and links without problem. Well... It seems your Makefile description does not access a valid SF url :( I have the following error ==> Grabbing download/libLASi-1.1.0.tar.gz ==> Trying file//files/libLASi-1.1.0.tar.gz gmake[1]: *** [file//files/libLASi-1.1.0.tar.gz] Erreur 255 ==> Trying file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz gmake[1]: *** [file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz] Erreur 255 ==> Trying http//downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz --2009-05-27 23:01:26-- http://downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz Resolving downloads.sourceforge.net... 216.34.181.59 Connecting to downloads.sourceforge.net|216.34.181.59|:80... connected. HTTP request sent, awaiting response... 301 Moved Permanently Location: http://sourceforge.net/projects/liblasi/files [following] --2009-05-27 23:01:26-- http://sourceforge.net/projects/liblasi/files Resolving sourceforge.net... 216.34.181.60 Connecting to sourceforge.net|216.34.181.60|:80... connected. HTTP request sent, awaiting response... 200 OK Length: unspecified [text/html] Saving to: `download/partial/files' 100%[========================================================================================================>] 11,515 43.6K/s in 0.3s 2009-05-27 23:01:27 (43.6 KB/s) - `download/partial/files' saved [11515] mv : impossible d'acc?der ? download/partial/libLASi-1.1.0.tar.gz (!!!) Failed to download download/libLASi-1.1.0.tar.gz! gmake: *** [download/libLASi-1.1.0.tar.gz] Erreur 255 I checked for project number and it is valid, but i cannot download sources, either on build8x nor on my box at home cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From dam at opencsw.org Wed May 27 23:25:16 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:25:16 +0200 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? References: <20090527171435.GN55163@bolthole.com> Message-ID: Hi, Anfang der weitergeleiteten E-Mail: > Von: Philip Brown > On Tue, May 26, 2009 at 09:28:58PM +0200, Dagobert Michelsen wrote: >> in Newpkgs is an updated bdb44 with 64 bit libs. I need it for >> OpenLDAP and comparison against bdb47 which comes next: > > why do we need both 44 and 47? > > this sounds kinda bad to me. > I vaguely recall that they DID actually build in some kind of > backwards > compat. > and even our current berkeleydbXX packages may be redundant to some > degree. I don't know. Maybe someone has an advice on how to proceed with this. The GAR recipe for bdb44 has been updated with 64 bit libs, but is this version still needed? There are quite some dependencies listed: This includes subversion and perl, both of which should get updates to 64 bit long term (svn 64 bit is needed for files > 2 GB. I don't want to discuss if this is a good idea, but we have a customer that needed it and where I hand-build it). Or should be officially drop 3.3, 4.2, 4.3 and 4.4 and just build 4.7? Best regards -- Dago From dam at opencsw.org Wed May 27 23:28:07 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:28:07 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> Message-ID: <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> Hi folks, Am 27.05.2009 um 20:01 schrieb Peter Bonivart: > On Wed, May 27, 2009 at 7:50 PM, Mike Watters > wrote: >> They are checked into subversion under the python package >> >> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile >> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile >> >> Addition to the README.CSW >> =========================== >> >> - - pycompile >> This class takes care of creating/removing python's normal and >> optimized >> bytecode files. Set all *.py files' class = pycompile to activate. >> Example: >> # use prototype filters to set the class >> PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "pycompile" } >> { print }' > > I'll add them tomorrow, thanks for the readme text. I updated the wiki > page with it to start with, feel free to add/edit. > > http://wiki.opencsw.org/cswclassutils-package Cool stuff. Just in time when I want to release xcbproto :-) Best regards -- Dago From mwatters at opencsw.org Wed May 27 23:33:23 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 16:33:23 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> Message-ID: <4A1DB1A3.8030205@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi folks, > > Am 27.05.2009 um 20:01 schrieb Peter Bonivart: >> On Wed, May 27, 2009 at 7:50 PM, Mike Watters >> wrote: >>> They are checked into subversion under the python package >>> >>> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/i.pycompile >>> >>> http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/python/trunk/files/r.pycompile >>> >>> >>> Addition to the README.CSW >>> =========================== >>> >>> - - pycompile >>> This class takes care of creating/removing python's normal and >>> optimized >>> bytecode files. Set all *.py files' class = pycompile to activate. >>> Example: >>> # use prototype filters to set the class >>> PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "pycompile" } { print }' >> >> I'll add them tomorrow, thanks for the readme text. I updated the wiki >> page with it to start with, feel free to add/edit. >> >> http://wiki.opencsw.org/cswclassutils-package > > Cool stuff. Just in time when I want to release xcbproto :-) > > > Best regards > > -- Dago > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers Dago, if you give it a couple days, I will have the python 2.6.2 version released. I am just waiting on the classutils. I have tested and retested python with this recipe, and the utils. I just want one clean test with the classutils package rather then me just copying the class scripts in place. all the python stuff will have a bug once the new packages are released. I split python into core, runtime, devel, idle and tk. dependencies should be linked to python_rt if linking against libpython.so.X - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodsaIACgkQLrhmsXMSLxc0mACgqbe0k8AN3dQvsDhyGrJeM2zj o9cAoNDFwtHP4TpBw5f0ZnVTdu7fMhyv =nu5A -----END PGP SIGNATURE----- From ellson at opencsw.org Wed May 27 23:46:02 2009 From: ellson at opencsw.org (John Ellson) Date: Wed, 27 May 2009 17:46:02 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1DAF31.7040109@wbonnet.net> References: <4A1D85AE.4060006@opencsw.org> <4A1DAF31.7040109@wbonnet.net> Message-ID: <4A1DB49A.8000503@opencsw.org> Fixed. (SF_PROJ != GARNAME != tar flile name != library name ...... grumble mumble ) John On 05/27/2009 05:22 PM, William Bonnet wrote: > Hi John >> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles >> and links without problem. > > Well... It seems your Makefile description does not access a valid SF > url :( > > I have the following error > > ==> Grabbing download/libLASi-1.1.0.tar.gz > ==> Trying file//files/libLASi-1.1.0.tar.gz > gmake[1]: *** [file//files/libLASi-1.1.0.tar.gz] Erreur 255 > ==> Trying > file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz > gmake[1]: *** > [file///export/home/william/community/opencsw/cache/libLASi-1.1.0.tar.gz] > Erreur 255 > ==> Trying > http//downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz > --2009-05-27 23:01:26-- > http://downloads.sourceforge.net/liblasi/libLASi-1.1.0.tar.gz > Resolving downloads.sourceforge.net... 216.34.181.59 > Connecting to downloads.sourceforge.net|216.34.181.59|:80... connected. > HTTP request sent, awaiting response... 301 Moved Permanently > Location: http://sourceforge.net/projects/liblasi/files [following] > --2009-05-27 23:01:26-- http://sourceforge.net/projects/liblasi/files > Resolving sourceforge.net... 216.34.181.60 > Connecting to sourceforge.net|216.34.181.60|:80... connected. > HTTP request sent, awaiting response... 200 OK > Length: unspecified [text/html] > Saving to: `download/partial/files' > > 100%[========================================================================================================>] > 11,515 43.6K/s in 0.3s > > 2009-05-27 23:01:27 (43.6 KB/s) - `download/partial/files' saved [11515] > > mv : impossible d'acc?der ? download/partial/libLASi-1.1.0.tar.gz > (!!!) Failed to download download/libLASi-1.1.0.tar.gz! > gmake: *** [download/libLASi-1.1.0.tar.gz] Erreur 255 > > I checked for project number and it is valid, but i cannot download > sources, either on build8x nor on my box at home > > cheers > W. > From phil at bolthole.com Wed May 27 23:47:41 2009 From: phil at bolthole.com (Philip Brown) Date: Wed, 27 May 2009 14:47:41 -0700 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: References: <20090527171435.GN55163@bolthole.com> Message-ID: <20090527214741.GB55163@bolthole.com> On Wed, May 27, 2009 at 11:25:16PM +0200, Dagobert Michelsen wrote: >... > Or should be officially drop 3.3, 4.2, 4.3 and 4.4 and just build 4.7? (and if so, we should probably rename/create new package, "berkeleydb", and have everything else depend on its rt package) From dam at opencsw.org Wed May 27 23:48:14 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 27 May 2009 23:48:14 +0200 Subject: [csw-maintainers] New schily tools in testing In-Reply-To: <4a1d9fc9.TCJ3FSsGMtwXUKLX%Joerg.Schilling@fokus.fraunhofer.de> References: <4a0da868.YMDL6tn9CxczHyn7%Joerg.Schilling@fokus.fraunhofer.de> <4a1d9fc9.TCJ3FSsGMtwXUKLX%Joerg.Schilling@fokus.fraunhofer.de> Message-ID: <98862867-B256-45AD-B7F6-8860DF9A6296@opencsw.org> Hi Joerg, Am 27.05.2009 um 22:17 schrieb Joerg Schilling: > Peter FELECAN wrote: >> Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling) writes: >> >>> I just compiled a new set of packages and put them into testing. >> >> Thanks. Do you need more testing or are they ready to be released? It >> would be really nice to have up to date version of your tools in our >> catalog. > > This is ready to be released. Ok then, just scp them over to www.opencsw.org:/home/newpkgs I cc'ed Phil so they should get pushed ASAP. Best regards -- Dago From mwatters at opencsw.org Wed May 27 23:55:22 2009 From: mwatters at opencsw.org (Mike Watters) Date: Wed, 27 May 2009 16:55:22 -0500 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: <20090527214741.GB55163@bolthole.com> References: <20090527171435.GN55163@bolthole.com> <20090527214741.GB55163@bolthole.com> Message-ID: <4A1DB6CA.6070304@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Philip Brown wrote: > On Wed, May 27, 2009 at 11:25:16PM +0200, Dagobert Michelsen wrote: >> ... >> Or should be officially drop 3.3, 4.2, 4.3 and 4.4 and just build 4.7? > > (and if so, we should probably rename/create new package, > "berkeleydb", and have everything else depend on its rt package) I say just build 4.7 but keep libdb4-4 as deprecated libs in 4.7. rename the package to berkeleydb and have core,runtime and devel just my 2 cents ( or 1/1000th of a euro, the way the US market is going ) - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkodtsoACgkQLrhmsXMSLxcXbQCgvPYHw6mdNVAGL6yveCMYJGLM JkYAniTTtKmnfQchif6e7DwtLWsVgMua =pIUO -----END PGP SIGNATURE----- From dam at opencsw.org Thu May 28 00:00:42 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 28 May 2009 00:00:42 +0200 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: <4A1DB6CA.6070304@opencsw.org> References: <20090527171435.GN55163@bolthole.com> <20090527214741.GB55163@bolthole.com> <4A1DB6CA.6070304@opencsw.org> Message-ID: Hi, Am 27.05.2009 um 23:55 schrieb Mike Watters: > I say just build 4.7 but keep libdb4-4 as deprecated libs in 4.7. > rename the package to berkeleydb and have core,runtime and devel As CSWbdb would be a new package there is no point in putting in old libraries in there as the existing packages depend on CSWbdb44. I guess we must recompile the dependent packages to make use of the new version. BTW: bdb47 et al. is now in testing/ Best regards -- Dago From car at opencsw.org Thu May 28 00:58:35 2009 From: car at opencsw.org (Chris Reece) Date: Thu, 28 May 2009 10:58:35 +1200 (NZST) Subject: [csw-maintainers] [csw-buildfarm] findutils and its database on the farm In-Reply-To: <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> References: <1243447269-sup-6062@ntdws12.chass.utoronto.ca> <1F55086E-8BFD-4EEF-AD16-48A17923EE91@opencsw.org> Message-ID: <50797.131.203.105.13.1243465115.squirrel@webmail.jessies.org> On Thu, May 28, 2009 9:12 am, Dagobert Michelsen wrote: > Chris: As you already did findutils with glocate, it would be > really cool if you could also package up slocate > as more secure alternative. Thus findutils nominates itself as my first GAR package. I'll put aside a few hours this weekend. I've recently moved jobs and my new position involves no Solaris. My intention is to GAR up the few packages I currently own and them throw them back into the pool, as it were, before resigning my maintainer status. From bonivart at opencsw.org Thu May 28 09:39:08 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 09:39:08 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1DB1A3.8030205@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> Message-ID: <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> On Wed, May 27, 2009 at 11:33 PM, Mike Watters wrote: > I am just waiting on the classutils. ?I have tested and retested python with > this recipe, and the utils. ?I just want one clean test with the classutils > package rather then me just copying the class scripts in place. I have dropped a new cswclassutils package into testing that includes cswpycompile: http://mirror.opencsw.org/testing/cswclassutils-1.10,REV=2009.05.28-SunOS5.8-all-CSW.pkg.gz -- /peter From james at opencsw.org Thu May 28 11:09:54 2009 From: james at opencsw.org (James Lee) Date: Thu, 28 May 2009 09:09:54 GMT Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1D85AE.4060006@opencsw.org> References: <4A1D85AE.4060006@opencsw.org> Message-ID: <20090528.9095400.3538465159@gyor.oxdrove.co.uk> On 27/05/09, 19:25:50, John Ellson wrote regarding [csw-maintainers] need help with C++ linker error from Sun's linker: > If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and > links without problem. > If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" > from the Makefile) > then I get this error: ... > Undefined first referenced > symbol in file > void LASi::PostscriptDocument::setFont(const > char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::Font St retch) > CMakeFiles/example0.dir/MissingGlyphExample.o > ld: fatal: Symbol referencing errors. No output written to example0 > There's something odd about the "const char*const" but I don't know if > thats related. I think you've almost answered your own question. gcc does not use "const" type qualifier for types passed by value when making external names. So "const char*const" is the same as "const char*" on the call because the value is copied anyway (the second "const" is of benefit to the compiler in the function only). Nevertheless CC likes consistency so try matching the "const"s in all function declarations. $ cat sub.cc void sub(const char*const arg) { return; } $ g++ -c sub.cc $ nm -C sub.o sub.o: [Index] Value Size Type Bind Other Shndx Name [2] | 0| 0|SECT |LOCL |0 |1 | [3] | 0| 0|SECT |LOCL |0 |2 | [4] | 0| 0|SECT |LOCL |0 |3 | [5] | 0| 0|SECT |LOCL |0 |4 | [6] | 0| 16|FUNC |GLOB |0 |1 |sub(const char*) [_Z3subPKc] [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc $ CC -c sub.cc $ nm -C sub.o sub.o: [Index] Value Size Type Bind Other Shndx Name [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc [2] | 16| 24|FUNC |GLOB |0 |2 |void sub(const char*const) [__1cDsub6Fkpkc_v_] James. From maciej at opencsw.org Thu May 28 11:28:29 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 28 May 2009 10:28:29 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <20090527184757.GU55163@bolthole.com> References: <20090527184757.GU55163@bolthole.com> Message-ID: On Wed, May 27, 2009 at 7:47 PM, Philip Brown wrote: > On Wed, May 27, 2009 at 07:13:55PM +0100, Maciej (Matchek) Blizinski wrote: >> I'm playing with updating syslog_ng to v2 and getting it to have SMF >> support via mGAR. >> .... >> The result is that the SMF manifest doesn't get imported: > > better to use cswclassutils for SMF support anyways, rather than a custom > pkg-local manifest, unless you're doing something Really Special. I'm trying to be as far away from Anything Special as possible. I'm following the common documentation of SMF support: https://apps.sourceforge.net/trac/gar/wiki/SMF/Init%20script%20support%20in%20your%20package A quote: 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - First step, we must have the following files ready in our files/ directory: * the legacy init script, cswfoo.init * the SMF manifest, cswfoo.xml * the SMF service startup script, svc-cswfoo 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - The second bullet list item explicitly tells to create cswfoo.xml. Should I be skeptical about it? For the purposes of testing, I've created 'minimalsmf' package. The idea is to have a minimal amount of code which builds a package with a sleeping shell script. The main Makefile is: http://apps.sourceforge.net/trac/gar/browser/csw/mgar/pkg/minimalsmf/trunk/Makefile Dago, could you please look at this package and remove anything unnecessary? Maciej From phil at bolthole.com Thu May 28 15:03:28 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 28 May 2009 06:03:28 -0700 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: References: <20090527184757.GU55163@bolthole.com> Message-ID: <20090528130328.GA91905@bolthole.com> On Thu, May 28, 2009 at 10:28:29AM +0100, Maciej (Matchek) Blizinski wrote: > I'm trying to be as far away from Anything Special as possible. I'm > following the common documentation of SMF support: > https://apps.sourceforge.net/trac/gar/wiki/SMF/Init%20script%20support%20in%20your%20package > yeah, as with most of the gar wiki documentation, that is out of date :-/ > 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > - - - - - - - - - - - - > First step, we must have the following files ready in our files/ directory: > > * the legacy init script, cswfoo.init > * the SMF manifest, cswfoo.xml > * the SMF service startup script, svc-cswfoo > 8< - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > - - - - - - - - - - - - > From phil at bolthole.com Thu May 28 15:06:39 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 28 May 2009 06:06:39 -0700 Subject: [csw-maintainers] Berkeley DB: Keep old versions or concentrate on 4.7? In-Reply-To: References: <20090527171435.GN55163@bolthole.com> <20090527214741.GB55163@bolthole.com> <4A1DB6CA.6070304@opencsw.org> Message-ID: <20090528130639.GB91905@bolthole.com> On Thu, May 28, 2009 at 12:00:42AM +0200, Dagobert Michelsen wrote: > Hi, > > Am 27.05.2009 um 23:55 schrieb Mike Watters: >> I say just build 4.7 but keep libdb4-4 as deprecated libs in 4.7. >> rename the package to berkeleydb and have core,runtime and devel > > As CSWbdb would be a new package there is no point in putting > in old libraries in there as the existing packages depend on > CSWbdb44. I guess we must recompile the dependent packages to > make use of the new version. Note: while there would be no point copying over binary libs from the older packages... it would be important to enable compiling "older libs" in the configuration for the compile for bdb47. I think (but could be wrong) that the berkeleydb config stuf has options for building older versions of libs. In which case, i guess we would want to then enable them for "CSWbdb", and conflict with the older packages, and replace those with empty packages or something. Hmm... upgrade path for users might be tricky... From Darin.Perusich at cognigencorp.com Thu May 28 15:21:25 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 28 May 2009 09:21:25 -0400 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup Message-ID: <4A1E8FD5.9000003@cognigencorp.com> I was reviewing the classutils documentation on setting file/dir ownership and I'm not clear on how this is accomplished. The documentation talks about assigning a new class, called 'ugfiles', and assigning it to the files/dirs. Is this 'ugfiles' just an arbitrary name or is it something associated with the cswusergroup class? Maybe I'm just being thickheaded but I am not finding this part of documentation confusing and unclear. Thanks! -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From bonivart at opencsw.org Thu May 28 15:43:17 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 15:43:17 +0200 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup In-Reply-To: <4A1E8FD5.9000003@cognigencorp.com> References: <4A1E8FD5.9000003@cognigencorp.com> Message-ID: <625385e30905280643v36b1b7f4v37396be12864fd5f@mail.gmail.com> On Thu, May 28, 2009 at 3:21 PM, Darin Perusich wrote: > I was reviewing the classutils documentation on setting file/dir > ownership and I'm not clear on how this is accomplished. > > The documentation talks about assigning a new class, called 'ugfiles', > and assigning it to the files/dirs. Is this 'ugfiles' just an arbitrary > ?name or is it something associated with the cswusergroup class? This step is only needed if you want files/dirs owned by the new user. If you only needed a user to run a service you can skip it. An example is my ClamAV package which has a service that runs with the clamav user and the signature databases needs to be owned by this user. Assuming you still want this, the problem is that the "none" class is always executed first so if your files are in this class the new user will not exist yet, ok? That's why we need an extra class for these files so we can tell pkgadd to run this class after cswusergroup, like this: CLASSES = none cswusergroup ugfiles This way the user is created by cswusergroup before files owned by this user are processed. > Maybe I'm just being thickheaded but I am not finding this part of > documentation confusing and unclear. I hope I just explained it better. Any way to improve the documentation? -- /peter From bonivart at opencsw.org Thu May 28 15:45:15 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 15:45:15 +0200 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup In-Reply-To: <4A1E8FD5.9000003@cognigencorp.com> References: <4A1E8FD5.9000003@cognigencorp.com> Message-ID: <625385e30905280645w2ae74aacme89262d5d79943db@mail.gmail.com> On Thu, May 28, 2009 at 3:21 PM, Darin Perusich wrote: > Is this 'ugfiles' just an arbitrary > ?name or is it something associated with the cswusergroup class? Sorry, missed this one, the name doesn't matter. -- /peter From ellson at opencsw.org Thu May 28 15:59:15 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 28 May 2009 09:59:15 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <20090528.9095400.3538465159@gyor.oxdrove.co.uk> References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> Message-ID: <4A1E98B3.7020508@opencsw.org> James, Thanks. I think I understand the issue, but I'm still not sure how to fix it? I'm not that familiar with C++, but I don't see any mismatched definitions in the code. Are you recommending that all "const char* const ... " in the sources be changed to "const char* ... " ? I'm a bit reluctant to make such an extensive change to the API of this library. Is there a linker flag that would tell it to tolerate these extra const's ? John James Lee wrote: > On 27/05/09, 19:25:50, John Ellson wrote regarding > [csw-maintainers] need help with C++ linker error from Sun's linker: > > >> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and >> links without problem. >> > > >> If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" >> from the Makefile) >> then I get this error: >> > > ... > > >> Undefined first referenced >> symbol in file >> void LASi::PostscriptDocument::setFont(const >> >> > char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::Font > St > retch) > >> CMakeFiles/example0.dir/MissingGlyphExample.o >> ld: fatal: Symbol referencing errors. No output written to example0 >> > > > > >> There's something odd about the "const char*const" but I don't know if >> thats related. >> > > > I think you've almost answered your own question. gcc does not use > "const" type qualifier for types passed by value when making external > names. So "const char*const" is the same as "const char*" on the call > because the value is copied anyway (the second "const" is of benefit to > the compiler in the function only). Nevertheless CC likes consistency > so try matching the "const"s in all function declarations. > > > > > $ cat sub.cc > void sub(const char*const arg) > { > return; > } > > $ g++ -c sub.cc > $ nm -C sub.o > > > sub.o: > > [Index] Value Size Type Bind Other Shndx Name > > [2] | 0| 0|SECT |LOCL |0 |1 | > [3] | 0| 0|SECT |LOCL |0 |2 | > [4] | 0| 0|SECT |LOCL |0 |3 | > [5] | 0| 0|SECT |LOCL |0 |4 | > [6] | 0| 16|FUNC |GLOB |0 |1 |sub(const char*) > [_Z3subPKc] > [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc > > $ CC -c sub.cc > $ nm -C sub.o > > > sub.o: > > [Index] Value Size Type Bind Other Shndx Name > > [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc > [2] | 16| 24|FUNC |GLOB |0 |2 |void sub(const > char*const) > > [__1cDsub6Fkpkc_v_] > > > > > James. > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > > From ellson at opencsw.org Thu May 28 16:33:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 28 May 2009 10:33:06 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1E98B3.7020508@opencsw.org> References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: <4A1EA0A2.7060105@opencsw.org> Never mind. I've found the problem. It wasn't the "const char*const ...", it was all the other args. John John Ellson wrote: > James, > > Thanks. I think I understand the issue, but I'm still not sure how to > fix it? > > I'm not that familiar with C++, but I don't see any mismatched > definitions in the code. > > Are you recommending that all "const char* const ... " in the sources > be changed to > "const char* ... " ? > I'm a bit reluctant to make such an extensive change to the API of > this library. > Is there a linker flag that would tell it to tolerate these extra > const's ? > > John > > > > James Lee wrote: >> On 27/05/09, 19:25:50, John Ellson wrote regarding >> [csw-maintainers] need help with C++ linker error from Sun's linker: >> >> >>> If I compile liblasi (mgar/pkg/liblasi/trunk) using g++ it compiles and >>> links without problem. >>> >> >> >>> If I use Sun's tools (remove the line: "CXX=/opt/csw/gcc4/bin/g++ \" >>> from the Makefile) >>> then I get this error: >>> >> >> ... >> >> >>> Undefined first referenced >>> symbol in file >>> void LASi::PostscriptDocument::setFont(const >>> >>> >> char*const,LASi::FontStyle,LASi::FontWeight,LASi::FontVariant,LASi::Font >> St >> retch) >> >>> CMakeFiles/example0.dir/MissingGlyphExample.o >>> ld: fatal: Symbol referencing errors. No output written to example0 >>> >> >> >> >> >>> There's something odd about the "const char*const" but I don't >>> know if >>> thats related. >>> >> >> >> I think you've almost answered your own question. gcc does not use >> "const" type qualifier for types passed by value when making external >> names. So "const char*const" is the same as "const char*" on the call >> because the value is copied anyway (the second "const" is of benefit to >> the compiler in the function only). Nevertheless CC likes consistency >> so try matching the "const"s in all function declarations. >> >> >> >> >> $ cat sub.cc >> void sub(const char*const arg) >> { >> return; >> } >> >> $ g++ -c sub.cc >> $ nm -C sub.o >> >> >> sub.o: >> >> [Index] Value Size Type Bind Other Shndx Name >> >> [2] | 0| 0|SECT |LOCL |0 |1 | >> [3] | 0| 0|SECT |LOCL |0 |2 | >> [4] | 0| 0|SECT |LOCL |0 |3 | >> [5] | 0| 0|SECT |LOCL |0 |4 | >> [6] | 0| 16|FUNC |GLOB |0 |1 |sub(const char*) >> [_Z3subPKc] >> [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc >> >> $ CC -c sub.cc >> $ nm -C sub.o >> >> >> sub.o: >> >> [Index] Value Size Type Bind Other Shndx Name >> >> [1] | 0| 0|FILE |LOCL |0 |ABS |sub.cc >> [2] | 16| 24|FUNC |GLOB |0 |2 |void sub(const >> char*const) >> >> [__1cDsub6Fkpkc_v_] >> >> >> >> >> James. >> _______________________________________________ >> maintainers mailing list >> maintainers at lists.opencsw.org >> https://lists.opencsw.org/mailman/listinfo/maintainers >> >> > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > From Darin.Perusich at cognigencorp.com Thu May 28 16:39:31 2009 From: Darin.Perusich at cognigencorp.com (Darin Perusich) Date: Thu, 28 May 2009 10:39:31 -0400 Subject: [csw-maintainers] setting file/dir ownership with cswusergroup In-Reply-To: <625385e30905280643v36b1b7f4v37396be12864fd5f@mail.gmail.com> References: <4A1E8FD5.9000003@cognigencorp.com> <625385e30905280643v36b1b7f4v37396be12864fd5f@mail.gmail.com> Message-ID: <4A1EA223.4070902@cognigencorp.com> Peter Bonivart wrote: > On Thu, May 28, 2009 at 3:21 PM, Darin Perusich > wrote: >> I was reviewing the classutils documentation on setting file/dir >> ownership and I'm not clear on how this is accomplished. >> >> The documentation talks about assigning a new class, called 'ugfiles', >> and assigning it to the files/dirs. Is this 'ugfiles' just an arbitrary >> name or is it something associated with the cswusergroup class? > > This step is only needed if you want files/dirs owned by the new user. > If you only needed a user to run a service you can skip it. An example > is my ClamAV package which has a service that runs with the clamav > user and the signature databases needs to be owned by this user. This is exactly what I need. The amanda package has a ton of files and need to be owned by the amanda user. > Assuming you still want this, the problem is that the "none" class is > always executed first so if your files are in this class the new user > will not exist yet, ok? That's why we need an extra class for these > files so we can tell pkgadd to run this class after cswusergroup, like > this: > > CLASSES = none cswusergroup ugfiles > > This way the user is created by cswusergroup before files owned by > this user are processed. > >> Maybe I'm just being thickheaded but I am not finding this part of >> documentation confusing and unclear. > > I hope I just explained it better. Any way to improve the documentation? > Your explanation has cleared this up and is what I was thinking but it just wasn't coming across in the documentation. I think the docs are to terse and need a little more explanation so the ideas are properly conveyed. I'm going to think about this for a bit and I'll write some updates out and send them over to you. -- Darin Perusich Unix Systems Administrator Cognigen Corporation 395 Youngs Rd. Williamsville, NY 14221 Phone: 716-633-3463 Email: darinper at cognigencorp.com From maciej at opencsw.org Thu May 28 17:26:49 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 28 May 2009 16:26:49 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: <20090528130328.GA91905@bolthole.com> References: <20090527184757.GU55163@bolthole.com> <20090528130328.GA91905@bolthole.com> Message-ID: On Thu, May 28, 2009 at 2:03 PM, Philip Brown wrote: > On Thu, May 28, 2009 at 10:28:29AM +0100, Maciej (Matchek) Blizinski wrote: >> I'm trying to be as far away from Anything Special as possible. I'm >> following the common documentation of SMF support: >> https://apps.sourceforge.net/trac/gar/wiki/SMF/Init%20script%20support%20in%20your%20package >> > > > yeah, as with most of the gar wiki documentation, that is out of date :-/ Okay, so this page is obsolete. I added a banner with a link to cswclassutils wiki page. I've been pointed at it by Peter. It might be also out of date, hopefully not as much as the first one. I'd be grateful if someone would attack minimalsmf with a saw and make it use the latest version of SMF support (I'm now so confused about naming that I won't use any other keywords than 'SMF' and 'support'.) Maciej From mwatters at opencsw.org Thu May 28 18:05:42 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 28 May 2009 11:05:42 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> Message-ID: <4A1EB656.8030700@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Peter, Peter Bonivart wrote: > On Wed, May 27, 2009 at 11:33 PM, Mike Watters wrote: >> I am just waiting on the classutils. I have tested and retested python with >> this recipe, and the utils. I just want one clean test with the classutils >> package rather then me just copying the class scripts in place. > > I have dropped a new cswclassutils package into testing that includes > cswpycompile: > > http://mirror.opencsw.org/testing/cswclassutils-1.10,REV=2009.05.28-SunOS5.8-all-CSW.pkg.gz > I forgot to actually copy/remove the original files in the class scripts for cswpycompile. new "working" copies are in my home dir on the build farm. please update cswclassutils - --- /usr/sadm/install/scripts/i.cswpycompile 2009-05-28 08:14:39.000000000 -0500 +++ i.cswpycompile 2009-05-28 10:55:23.320409700 -0500 @@ -20,6 +20,13 @@ if [ "${DEBUG}" ]; then echo PACKAGE: $PKGINST fi +while read src dest; do + if [ "$DEBUG" ]; then + echo SRC: $src DEST: $dest + fi + + /usr/bin/cp -p $src $dest || exit 2 +done if [ ! -f ${PY_BINDIR}/python -a ! -x ${PY_BINDIR}/python ]; then echo "Could not find or execute ${PY_BINDIR}/python" - --- /usr/sadm/install/scripts/r.cswpycompile 2009-05-28 08:14:39.000000000 -0500 +++ r.cswpycompile 2009-05-28 10:55:23.320922600 -0500 @@ -20,4 +20,12 @@ echo "Removing pyo and pyc files for package <$PKGINST> ..." pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}c {}o +while read dest; do + if [ "$DEBUG" ]; then + echo DEST: $dest + fi + /usr/bin/rm -f $dest || exit 2 +done + + exit 0 - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoetlYACgkQLrhmsXMSLxeVjwCgv9ggwnVNG8rcpqPsJj4zxh7A pHgAn1nSQkOoBjV3muSqQJORot67IJCa =81R5 -----END PGP SIGNATURE----- From pfelecan at opencsw.org Thu May 28 18:11:18 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 28 May 2009 18:11:18 +0200 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1E98B3.7020508@opencsw.org> (John Ellson's message of "Thu\, 28 May 2009 09\:59\:15 -0400") References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: John Ellson writes: > I'm a bit reluctant to make such an extensive change to the API of > this library. > Is there a linker flag that would tell it to tolerate these extra const's ? Use the Force: gcc used must be -- Peter From ellson at opencsw.org Thu May 28 18:16:06 2009 From: ellson at opencsw.org (John Ellson) Date: Thu, 28 May 2009 12:16:06 -0400 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: <4A1EB8C6.108@opencsw.org> Peter FELECAN wrote: > John Ellson writes: > > >> I'm a bit reluctant to make such an extensive change to the API of >> this library. >> Is there a linker flag that would tell it to tolerate these extra const's ? >> > > Use the Force: gcc used must be > Its ok, I've patched the implementation to avoid changing the api. Perhaps Yoda would care to review ? ;-) mgar/pkg/liblasi/trunk John From phil at bolthole.com Thu May 28 18:31:21 2009 From: phil at bolthole.com (Philip Brown) Date: Thu, 28 May 2009 09:31:21 -0700 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> Message-ID: <20090528163121.GC55163@bolthole.com> On Thu, May 28, 2009 at 06:11:18PM +0200, Peter FELECAN wrote: > John Ellson writes: > > > I'm a bit reluctant to make such an extensive change to the API of > > this library. > > Is there a linker flag that would tell it to tolerate these extra const's ? > > Use the Force: gcc used must be NOOOOOO!!!! Once you start down the dark path, forever will it dominate your destiny. Consume you it will! :-} From pfelecan at opencsw.org Thu May 28 18:45:02 2009 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 28 May 2009 18:45:02 +0200 Subject: [csw-maintainers] need help with C++ linker error from Sun's linker In-Reply-To: <4A1EB8C6.108@opencsw.org> (John Ellson's message of "Thu\, 28 May 2009 12\:16\:06 -0400") References: <4A1D85AE.4060006@opencsw.org> <20090528.9095400.3538465159@gyor.oxdrove.co.uk> <4A1E98B3.7020508@opencsw.org> <4A1EB8C6.108@opencsw.org> Message-ID: John Ellson writes: > Peter FELECAN wrote: >> John Ellson writes: >> >> >>> I'm a bit reluctant to make such an extensive change to the API of >>> this library. >>> Is there a linker flag that would tell it to tolerate these extra const's ? >>> >> >> Use the Force: gcc used must be >> > Its ok, I've patched the implementation to avoid changing the api. > > Perhaps Yoda would care to review ? ;-) > > mgar/pkg/liblasi/trunk no light on the mGar path by hand everything I made -- Peter From bonivart at opencsw.org Thu May 28 20:38:20 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Thu, 28 May 2009 20:38:20 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1EB656.8030700@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> Message-ID: <625385e30905281138p39ea82afx14fabbb0f5c5691a@mail.gmail.com> On Thu, May 28, 2009 at 6:05 PM, Mike Watters wrote: > I forgot to actually copy/remove the original files in the class scripts for > cswpycompile. ?new "working" copies are in my home dir on the build farm. > please update cswclassutils No problem, I will do it tomorrow. -- /peter From william at wbonnet.net Thu May 28 21:28:54 2009 From: william at wbonnet.net (William Bonnet) Date: Thu, 28 May 2009 21:28:54 +0200 Subject: [csw-maintainers] Thunderbird 2.0.0.21 is in testing Message-ID: <4A1EE5F6.9060403@wbonnet.net> Hi Thunderbird 2.0.0.21 is available from testing. You will need to update libcairo, libpango, pixman from testing, and an up to date glib2 from current (at least 2.20) Cheers W. -- William http://www.wbonnet.net http://www.sunwizard.net Le site fran?ais des amateurs de stations Unix http://www.opencsw.org Community SoftWare for Solaris http://www.guses.org French speaking Solaris User Group From mwatters at opencsw.org Thu May 28 22:20:04 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 28 May 2009 15:20:04 -0500 Subject: [csw-maintainers] mediawiki 1.14.0 now in testing Message-ID: <4A1EF1F4.1090304@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 latest version: Feedback Welcome! - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkoe8fQACgkQLrhmsXMSLxdFEgCfbJmPPwtzCje9O6+rSeF8Ha1f rokAnR0MM9wlVaPnSYEswrqlCyZWYqk/ =tIMr -----END PGP SIGNATURE----- From mwatters at opencsw.org Thu May 28 23:22:59 2009 From: mwatters at opencsw.org (Mike Watters) Date: Thu, 28 May 2009 16:22:59 -0500 Subject: [csw-maintainers] squirrelmail 1.4.17 now in testing Message-ID: <4A1F00B3.5090309@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Feedback Welcome! - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkofALMACgkQLrhmsXMSLxdfgACdHCDGHdWr93Q/2QXm7gC0vkn3 X64AnRSeK9BR5PHmxsv73NpKwvVfmV9m =wtFx -----END PGP SIGNATURE----- From maciej at opencsw.org Thu May 28 23:39:18 2009 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Thu, 28 May 2009 22:39:18 +0100 Subject: [csw-maintainers] SMF support for syslog_ng package In-Reply-To: References: <20090527184757.GU55163@bolthole.com> <20090528130328.GA91905@bolthole.com> Message-ID: On Thu, May 28, 2009 at 4:26 PM, Maciej (Matchek) Blizinski wrote: > I'd be grateful if someone would attack minimalsmf with a saw and make > it use the latest version of SMF support (I'm now so confused about > naming that I won't use any other keywords than 'SMF' and 'support'.) Latest news: I've slogged away at the code tree, figured out that clamav does SMF via classutils, figured out that it was consistent with the 'cswclassutls' page, and updated the minimalsmf package. It worked! I've created a version of syslog_ng with mgar v2 and cswclassutils-driven SMF support. I thought that committing the change directly to the syslog_ng/trunk directory might be considered rude, so I branched the build and committed my change to the branch. Here's the change: http://apps.sourceforge.net/trac/gar/changeset/5096 Ihsan, can you please look at it? If it looks good to you, you could merge it to the trunk. Maciej From bonivart at opencsw.org Fri May 29 14:43:48 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 29 May 2009 14:43:48 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1EB656.8030700@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> Message-ID: <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> On Thu, May 28, 2009 at 6:05 PM, Mike Watters wrote: > I forgot to actually copy/remove the original files in the class scripts for > cswpycompile. ?new "working" copies are in my home dir on the build farm. > please update cswclassutils I have updated the package in testing: http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz -- /peter From mwatters at opencsw.org Fri May 29 15:58:28 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 29 May 2009 08:58:28 -0500 Subject: [csw-maintainers] python 2.6.2 in testing Message-ID: <4A1FEA04.5000708@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 changes: update to 2.6.2 now contains no pre-compiled bytecode in the pacakge requires Peter's cswclassutils 1.14,REV=2009.05.29 that is also in testing. various minor bug fixes. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkof6gQACgkQLrhmsXMSLxd29QCgnha4inMeZfA71FzrF4ImKW6a JIMAoMaMGShP/4WtHtlaxQW4z1IgW3WK =LI0e -----END PGP SIGNATURE----- From mwatters at opencsw.org Fri May 29 16:06:26 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 29 May 2009 09:06:26 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> Message-ID: <4A1FEBE2.3060907@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Peter Bonivart wrote: > On Thu, May 28, 2009 at 6:05 PM, Mike Watters wrote: >> I forgot to actually copy/remove the original files in the class scripts for >> cswpycompile. new "working" copies are in my home dir on the build farm. >> please update cswclassutils > > I have updated the package in testing: > > http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz > Most Excellent, cswpycompile class works like a charm now. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (SunOS) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkof6+IACgkQLrhmsXMSLxdwGACZAdHpre9MpneJgMi2pOKUzbEt pPQAniX9CqpWOYB3jjifPXFN+Ggcnb2I =OIMH -----END PGP SIGNATURE----- From bonivart at opencsw.org Fri May 29 17:03:52 2009 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 29 May 2009 17:03:52 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1FEBE2.3060907@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> Message-ID: <625385e30905290803q3ca0c86dlb8bd8cfd24ff3a16@mail.gmail.com> On Fri, May 29, 2009 at 4:06 PM, Mike Watters wrote: > Most Excellent, cswpycompile class works like a charm now. That's good to hear. :-) However, there's quite a few changes to cswinitsmf in 1.14 as well so we need some testing before releasing cswclassutils. I hope other maintainers can help with testing. - fixed bug that always enabled services using init scripts (#3633) - the persistent state code is back (keeps state during upgrades) (#3634) - AUTOENABLE tweak (provides a way of overriding local csw.conf) (#3635) - MANIFEST tweak (custom manifest instead of autogenerated) (#3636) The tweaks are briefly documented here: http://wiki.opencsw.org/cswclassutils-package#toc2 -- /peter From phil at bolthole.com Fri May 29 18:40:34 2009 From: phil at bolthole.com (Philip Brown) Date: Fri, 29 May 2009 09:40:34 -0700 Subject: [csw-maintainers] new class scripts In-Reply-To: <625385e30905290803q3ca0c86dlb8bd8cfd24ff3a16@mail.gmail.com> References: <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <625385e30905290803q3ca0c86dlb8bd8cfd24ff3a16@mail.gmail.com> Message-ID: <20090529164032.GR55163@bolthole.com> On Fri, May 29, 2009 at 05:03:52PM +0200, Peter Bonivart wrote: > - AUTOENABLE tweak (provides a way of overriding local csw.conf) (#3635) yikes. that worried me when I just read that summary :-} but fyi to others, this is explicitly to DISABLE the service until properly configured, even if csw.conf says "autoenable_daemons=yes". From dam at opencsw.org Fri May 29 22:54:18 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 29 May 2009 22:54:18 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A1FEBE2.3060907@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> Message-ID: Hi Mike, Am 29.05.2009 um 16:06 schrieb Mike Watters: > Peter Bonivart wrote: >> On Thu, May 28, 2009 at 6:05 PM, Mike Watters >> wrote: >>> I forgot to actually copy/remove the original files in the class >>> scripts for >>> cswpycompile. new "working" copies are in my home dir on the >>> build farm. >>> please update cswclassutils >> >> I have updated the package in testing: >> >> http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz > > Most Excellent, cswpycompile class works like a charm now. I have some troubles with cswpycompile: - I added support to xcb-proto like this: > EXTRA_MERGE_EXCLUDE_FILES = .*\.pyo .*\.pyc > PROTOTYPE_FILTER = awk '$$$$3 ~ /.*\.py$$$$/ { $$$$2 = > "cswpycompile" } { print }' > SPKG_CLASSES = none cswpycompile > REQUIRED_PKGS += CSWcswclassutils - On installation something does get compiled: > Compiling py files to normal bytecode ... > Listing /opt/csw/lib/python ... > Listing /opt/csw/lib/python/bsddb ... > Listing /opt/csw/lib/python/bsddb/test ... > Listing /opt/csw/lib/python/compiler ... > Listing /opt/csw/lib/python/ctypes ... > Listing /opt/csw/lib/python/ctypes/macholib ... > Listing /opt/csw/lib/python/ctypes/test ... > Listing /opt/csw/lib/python/curses ... > Listing /opt/csw/lib/python/email ... > Listing /opt/csw/lib/python/email/mime ... > Listing /opt/csw/lib/python/email/test ... > Listing /opt/csw/lib/python/email/test/data ... > Listing /opt/csw/lib/python/encodings ... > Listing /opt/csw/lib/python/hotshot ... > Listing /opt/csw/lib/python/json ... > Listing /opt/csw/lib/python/json/tests ... > Listing /opt/csw/lib/python/lib-dynload ... > Listing /opt/csw/lib/python/lib-old ... > Listing /opt/csw/lib/python/lib2to3 ... > Listing /opt/csw/lib/python/lib2to3/fixes ... > Listing /opt/csw/lib/python/lib2to3/pgen2 ... > Listing /opt/csw/lib/python/lib2to3/tests ... > Listing /opt/csw/lib/python/logging ... > Listing /opt/csw/lib/python/multiprocessing ... > Listing /opt/csw/lib/python/multiprocessing/dummy ... > Listing /opt/csw/lib/python/plat-sunos5 ... > Listing /opt/csw/lib/python/site-packages ... > Listing /opt/csw/lib/python/sqlite3 ... > Listing /opt/csw/lib/python/sqlite3/test ... > Listing /opt/csw/lib/python/test ... > Listing /opt/csw/lib/python/test/decimaltestdata ... > Listing /opt/csw/lib/python/wsgiref ... > Listing /opt/csw/lib/python/xml ... > Listing /opt/csw/lib/python/xml/dom ... > Listing /opt/csw/lib/python/xml/etree ... > Listing /opt/csw/lib/python/xml/parsers ... > Listing /opt/csw/lib/python/xml/sax ... > Compiling py files to optimized bytecode ... > Listing /opt/csw/lib/python ... > Listing /opt/csw/lib/python/bsddb ... > Listing /opt/csw/lib/python/bsddb/test ... > Listing /opt/csw/lib/python/compiler ... > Listing /opt/csw/lib/python/ctypes ... > Listing /opt/csw/lib/python/ctypes/macholib ... > Listing /opt/csw/lib/python/ctypes/test ... > Listing /opt/csw/lib/python/curses ... > Listing /opt/csw/lib/python/email ... > Listing /opt/csw/lib/python/email/mime ... > Listing /opt/csw/lib/python/email/test ... > Listing /opt/csw/lib/python/email/test/data ... > Listing /opt/csw/lib/python/encodings ... > Listing /opt/csw/lib/python/hotshot ... > Listing /opt/csw/lib/python/json ... > Listing /opt/csw/lib/python/json/tests ... > Listing /opt/csw/lib/python/lib-dynload ... > Listing /opt/csw/lib/python/lib-old ... > Listing /opt/csw/lib/python/lib2to3 ... > Listing /opt/csw/lib/python/lib2to3/fixes ... > Listing /opt/csw/lib/python/lib2to3/pgen2 ... > Listing /opt/csw/lib/python/lib2to3/tests ... > Listing /opt/csw/lib/python/logging ... > Listing /opt/csw/lib/python/multiprocessing ... > Listing /opt/csw/lib/python/multiprocessing/dummy ... > Listing /opt/csw/lib/python/plat-sunos5 ... > Listing /opt/csw/lib/python/site-packages ... > Listing /opt/csw/lib/python/sqlite3 ... > Listing /opt/csw/lib/python/sqlite3/test ... > Listing /opt/csw/lib/python/test ... > Listing /opt/csw/lib/python/test/decimaltestdata ... > Listing /opt/csw/lib/python/wsgiref ... > Listing /opt/csw/lib/python/xml ... > Listing /opt/csw/lib/python/xml/dom ... > Listing /opt/csw/lib/python/xml/etree ... > Listing /opt/csw/lib/python/xml/parsers ... > Listing /opt/csw/lib/python/xml/sax ... > [ verifying class ] > > Installation of was successful. - From the remove script I would guess the compiled and optimized files are in the same directory as the .py-files: > echo "Removing pyo and pyc files for package <$PKGINST> ..." > pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}c {}o - However, in the directory where the packages installed the .py-files nothing is added: > build8st# ls -l /opt/csw/X11/lib/python2.6/site-packages/xcbgen > total 64 > -rw-r--r-- 1 root bin 1 May 29 22:24 __init__.py > -rw-r--r-- 1 root bin 116 May 29 22:24 error.py > -rw-r--r-- 1 root bin 2965 May 29 22:24 expr.py > -rw-r--r-- 1 root bin 3534 May 29 22:24 matcher.py > -rw-r--r-- 1 root bin 5141 May 29 22:24 state.py > -rw-r--r-- 1 root bin 16352 May 29 22:24 xtypes.py So, what's going wrong here? After this is settled and CSWcswclassutils and CSWpython get released I'll modify GAR to automatically do it right: - look if there are .py files, if yes: - remove .pyc and .pyo files on merge - modify class of .py to cswpycompile - Add dependency to CSWcswclassutils Best regards -- Dago From mwatters at opencsw.org Sat May 30 01:27:46 2009 From: mwatters at opencsw.org (Mike Watters) Date: Fri, 29 May 2009 18:27:46 -0500 Subject: [csw-maintainers] new class scripts In-Reply-To: References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> Message-ID: <4A206F72.7010909@opencsw.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dagobert Michelsen wrote: > Hi Mike, > > Am 29.05.2009 um 16:06 schrieb Mike Watters: >> Peter Bonivart wrote: >>> On Thu, May 28, 2009 at 6:05 PM, Mike Watters >>> wrote: >>>> I forgot to actually copy/remove the original files in the class >>>> scripts for >>>> cswpycompile. new "working" copies are in my home dir on the build >>>> farm. >>>> please update cswclassutils >>> >>> I have updated the package in testing: >>> >>> http://mirror.opencsw.org/testing/cswclassutils-1.14,REV=2009.05.29-SunOS5.8-all-CSW.pkg.gz >>> >> >> Most Excellent, cswpycompile class works like a charm now. > > I have some troubles with cswpycompile: > > - I added support to xcb-proto like this: > >> EXTRA_MERGE_EXCLUDE_FILES = .*\.pyo .*\.pyc >> PROTOTYPE_FILTER = awk '$$$$3 ~ /.*\.py$$$$/ { $$$$2 = "cswpycompile" >> } { print }' >> SPKG_CLASSES = none cswpycompile >> REQUIRED_PKGS += CSWcswclassutils > > - On installation something does get compiled: > >> Compiling py files to normal bytecode ... >> Listing /opt/csw/lib/python ... >> Listing /opt/csw/lib/python/bsddb ... >> Listing /opt/csw/lib/python/bsddb/test ... >> Listing /opt/csw/lib/python/compiler ... >> Listing /opt/csw/lib/python/ctypes ... >> Listing /opt/csw/lib/python/ctypes/macholib ... >> Listing /opt/csw/lib/python/ctypes/test ... >> Listing /opt/csw/lib/python/curses ... >> Listing /opt/csw/lib/python/email ... >> Listing /opt/csw/lib/python/email/mime ... >> Listing /opt/csw/lib/python/email/test ... >> Listing /opt/csw/lib/python/email/test/data ... >> Listing /opt/csw/lib/python/encodings ... >> Listing /opt/csw/lib/python/hotshot ... >> Listing /opt/csw/lib/python/json ... >> Listing /opt/csw/lib/python/json/tests ... >> Listing /opt/csw/lib/python/lib-dynload ... >> Listing /opt/csw/lib/python/lib-old ... >> Listing /opt/csw/lib/python/lib2to3 ... >> Listing /opt/csw/lib/python/lib2to3/fixes ... >> Listing /opt/csw/lib/python/lib2to3/pgen2 ... >> Listing /opt/csw/lib/python/lib2to3/tests ... >> Listing /opt/csw/lib/python/logging ... >> Listing /opt/csw/lib/python/multiprocessing ... >> Listing /opt/csw/lib/python/multiprocessing/dummy ... >> Listing /opt/csw/lib/python/plat-sunos5 ... >> Listing /opt/csw/lib/python/site-packages ... >> Listing /opt/csw/lib/python/sqlite3 ... >> Listing /opt/csw/lib/python/sqlite3/test ... >> Listing /opt/csw/lib/python/test ... >> Listing /opt/csw/lib/python/test/decimaltestdata ... >> Listing /opt/csw/lib/python/wsgiref ... >> Listing /opt/csw/lib/python/xml ... >> Listing /opt/csw/lib/python/xml/dom ... >> Listing /opt/csw/lib/python/xml/etree ... >> Listing /opt/csw/lib/python/xml/parsers ... >> Listing /opt/csw/lib/python/xml/sax ... >> Compiling py files to optimized bytecode ... >> Listing /opt/csw/lib/python ... >> Listing /opt/csw/lib/python/bsddb ... >> Listing /opt/csw/lib/python/bsddb/test ... >> Listing /opt/csw/lib/python/compiler ... >> Listing /opt/csw/lib/python/ctypes ... >> Listing /opt/csw/lib/python/ctypes/macholib ... >> Listing /opt/csw/lib/python/ctypes/test ... >> Listing /opt/csw/lib/python/curses ... >> Listing /opt/csw/lib/python/email ... >> Listing /opt/csw/lib/python/email/mime ... >> Listing /opt/csw/lib/python/email/test ... >> Listing /opt/csw/lib/python/email/test/data ... >> Listing /opt/csw/lib/python/encodings ... >> Listing /opt/csw/lib/python/hotshot ... >> Listing /opt/csw/lib/python/json ... >> Listing /opt/csw/lib/python/json/tests ... >> Listing /opt/csw/lib/python/lib-dynload ... >> Listing /opt/csw/lib/python/lib-old ... >> Listing /opt/csw/lib/python/lib2to3 ... >> Listing /opt/csw/lib/python/lib2to3/fixes ... >> Listing /opt/csw/lib/python/lib2to3/pgen2 ... >> Listing /opt/csw/lib/python/lib2to3/tests ... >> Listing /opt/csw/lib/python/logging ... >> Listing /opt/csw/lib/python/multiprocessing ... >> Listing /opt/csw/lib/python/multiprocessing/dummy ... >> Listing /opt/csw/lib/python/plat-sunos5 ... >> Listing /opt/csw/lib/python/site-packages ... >> Listing /opt/csw/lib/python/sqlite3 ... >> Listing /opt/csw/lib/python/sqlite3/test ... >> Listing /opt/csw/lib/python/test ... >> Listing /opt/csw/lib/python/test/decimaltestdata ... >> Listing /opt/csw/lib/python/wsgiref ... >> Listing /opt/csw/lib/python/xml ... >> Listing /opt/csw/lib/python/xml/dom ... >> Listing /opt/csw/lib/python/xml/etree ... >> Listing /opt/csw/lib/python/xml/parsers ... >> Listing /opt/csw/lib/python/xml/sax ... >> [ verifying class ] >> >> Installation of was successful. > > - From the remove script I would guess the compiled and optimized files are > in the same directory as the .py-files: > >> echo "Removing pyo and pyc files for package <$PKGINST> ..." >> pkgchk -v $PKGINST 2>&1 |grep \.py$ | xargs -i rm -f {}c {}o > > - However, in the directory where the packages installed the .py-files > nothing is added: > >> build8st# ls -l /opt/csw/X11/lib/python2.6/site-packages/xcbgen >> total 64 >> -rw-r--r-- 1 root bin 1 May 29 22:24 __init__.py >> -rw-r--r-- 1 root bin 116 May 29 22:24 error.py >> -rw-r--r-- 1 root bin 2965 May 29 22:24 expr.py >> -rw-r--r-- 1 root bin 3534 May 29 22:24 matcher.py >> -rw-r--r-- 1 root bin 5141 May 29 22:24 state.py >> -rw-r--r-- 1 root bin 16352 May 29 22:24 xtypes.py > > So, what's going wrong here? > > After this is settled and CSWcswclassutils and CSWpython get released > I'll modify GAR to automatically do it right: > > - look if there are .py files, if yes: > - remove .pyc and .pyo files on merge > - modify class of .py to cswpycompile > - Add dependency to CSWcswclassutils > > > Best regards > > -- Dago > > > > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers which version of classutils are you using? The Listing is normal, it just means it found a directory and is Listing the contents to the script. The way I have the install class script written, is it should try to compile everything under /opt/csw/lib/python that can be compiled. I need to research this weekend on whether or not this will traverse the site-packages directory. I thought it did, but it may not. Look at your install directory, does it contain pyo/pyc files? Thanks for the report. - -- Thanks, Mike "Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius -- and a lot of courage -- to move in the opposite direction." * Albert Einstein 1879 - 1955 US German-born Theoretical Physicist -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkogb3IACgkQLrhmsXMSLxc4aQCgs/3Oj/b8nxrv2Nmu4NRnvALS GNMAoK1/WvSSTxlwVe32AexLYtQiY/An =TT+I -----END PGP SIGNATURE----- From rupert at opencsw.org Sun May 31 17:05:28 2009 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 31 May 2009 17:05:28 +0200 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: <14077.1243297943@gerd> References: <6af4270905250658u4e045f47v3e2a763400f14109@mail.gmail.com> <14077.1243297943@gerd> Message-ID: <6af4270905310805h5187ef6cn4e81ec046005abb5@mail.gmail.com> On Tue, May 26, 2009 at 02:32, wrote: > On Mon, 25 May 2009 23:58:37 +1000, rupert THURNER writes: >>i tried to build it on the standard build servers. how did you install >>it? > > I have my own "build servers". You will have to request that renderPM be > installed on the central "build servers". > > But first check that the version of renderPM in CSW is late enough for > reportlab 2.3 - you may need to build and submit renderPM first, then > get it installed on the "build servers", then move on to building > reportlab. > > Also don't forget to check out the rl_accel plugin for reportlab. > >>or could i have access to your build file? > > My build file basically is this: > > ? ? ? ?python setup.py build > ? ? ? ?mkdir -p ${buildroot} > ? ? ? ?python setup.py install --root=${buildroot} > ? ? ? ?chown -Rh root:bin ${buildroot}/opt > > i.e. the standard way to build and install python packages (I believe). > > GAR will probably have mechanisms for this by default(?). Cheers! > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Murray... hmm .. meanwhile i tried the version built on build8s, and it works great. but on build8x i still did not see whom to ask to correct what, or what i could correct ? From Murray.Jensen at csiro.au Sun May 31 17:10:54 2009 From: Murray.Jensen at csiro.au (Murray.Jensen at csiro.au) Date: Mon, 1 Jun 2009 01:10:54 +1000 Subject: [csw-maintainers] python-reportlab 2.3 In-Reply-To: Your message of "Mon, 01 Jun 2009 01:05:28 +1000" <6af4270905310805h5187ef6cn4e81ec046005abb5@mail.gmail.com> Message-ID: <29335.1243782654@gerd> On Mon, 1 Jun 2009 01:05:28 +1000, rupert THURNER writes: >hmm .. meanwhile i tried the version built on build8s, and it works >great. but on build8x i still did not see whom to ask to correct what, >or what i could correct ? Are you sure that renderPM is installed on both build8s and build8x? Cheers! Murray... From dam at opencsw.org Sun May 31 22:53:32 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 31 May 2009 22:53:32 +0200 Subject: [csw-maintainers] Fwd: [csw-users] feedback on pkg-get behaviour change requested References: Message-ID: Hi, this was just on users at . Gary: what do you think? I guess nobody was aware of this issue. Best regards -- Dago Anfang der weitergeleiteten E-Mail: > Von: Chris May > Datum: 31. Mai 2009 22:37:16 MESZ > An: users at lists.opencsw.org > Betreff: Re: [csw-users] feedback on pkg-get behaviour change > requested > Antwort an: questions and discussions > > Philip Brown wrote: > [...] >> I am considering altering the behavior of "pkg-get -c", so that it >> only prints out installed packages, that have DIFFERENT versions in >> the >> catalog. > > Has this change now been made? I notice what looks like this > difference between the > behaviour of 4.1.2 and 4.2, when querying for a package which is > already installed. > Comparing the two... > > # /tmp/pkg-get.4.1.2 -c nrpe > # (From site http://ibiblio.org/pub/packages/solaris/opencsw/current ) > software localrev > remoterev > nrpe > 2.12,REV=2009.05.17 SAME > > # /tmp/pkg-get.4.2 -c nrpe > # (From site http://ibiblio.org/pub/packages/solaris/opencsw/current ) > software localrev > remoterev > # > > This causes a problem for the Puppet configuration management > software(http://reductivelabs.com/trac/puppet), which parses the > output of pkg-get -c {package} to determine whether a package is > installed or not. The absence of the "SAME" line causes it to think > that the package is not present, and hence re-install it. > > Is this intended to be a permanent change? If so, I'll raise an > enhancement request with the Puppet maintainers to use '-c -a > {package}' instead - which looks as if it should work in the same way > with either pkg-get 4.1.2 or 4.2 From dam at opencsw.org Sun May 31 23:00:37 2009 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 31 May 2009 23:00:37 +0200 Subject: [csw-maintainers] new class scripts In-Reply-To: <4A206F72.7010909@opencsw.org> References: <4A1D71FF.9080806@opencsw.org> <625385e30905271031w562aa607pacd7fac19b9cbcf8@mail.gmail.com> <4A1D7D6E.7000709@opencsw.org> <625385e30905271101i753abedagbb9b4d81c680ec4b@mail.gmail.com> <31207629-77D7-4D43-B481-0AE424B0200D@opencsw.org> <4A1DB1A3.8030205@opencsw.org> <625385e30905280039h2fbac6abnda278c0fbecddb71@mail.gmail.com> <4A1EB656.8030700@opencsw.org> <625385e30905290543v11a38709odcfd931cbfe8fe6f@mail.gmail.com> <4A1FEBE2.3060907@opencsw.org> <4A206F72.7010909@opencsw.org> Message-ID: <14DD2525-EFDC-4980-8870-B7016DE29A91@opencsw.org> Hi Mike, Am 30.05.2009 um 01:27 schrieb Mike Watters: > which version of classutils are you using? The one from testing, of couse. That is 1.5,REV=2009.03.28 > The Listing is normal, it just means it found a directory and is > Listing the > contents to the script. > > The way I have the install class script written, > is it should try to compile everything under /opt/csw/lib/python > that can be > compiled. I need to research this weekend on whether or not this > will traverse > the site-packages directory. I thought it did, but it may not. > > Look at your install directory, does it contain pyo/pyc files? No. I opensnoop'ed the pkginstall of my package from the globalzone, and there doesn't seem to be any compiled files generated. The package is additionally in testing/: Best regards -- Dago root at ncsw [global]:/root > opensnoop | fgrep py 0 17552 cpio 3 install/copyright 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/xtypes.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/error.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/state.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/__init__.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/expr.py 0 17552 cpio 3 root/opt/csw/X11/lib/python2.6/site- packages/xcbgen/matcher.py 0 17546 pkginstall 5 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ install/copyright 0 17546 pkginstall 12 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ install/copyright 0 17546 pkginstall 14 /var/sadm/pkg/CSWpython/pkginfo 0 17546 pkginstall 14 /var/sadm/pkg/CSWpython/pkginfo 0 17546 pkginstall 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ install/copyright 0 17546 pkginstall 8 /var/sadm/pkg/CSWxcbproto/install/ copyright 0 17554 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.py 0 17556 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/expr.py 0 17557 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/matcher.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/__init__.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/expr.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/matcher.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/state.py 0 17546 pkginstall 8 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py 0 17555 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/error.py 0 17553 sh 7 /usr/sadm/install/scripts/i.cswpycompile 0 17558 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/state.py 0 17559 cp 7 /var/tmp/dstreAAAmFa4qI/CSWxcbproto/ root/opt/csw/X11/lib/python2.6/site-packages/xcbgen/xtypes.py 0 17560 python 7 /var/ld/ld.config 0 17560 python 7 /usr/lib/libc.so.1 0 17560 python 7 /usr/lib/libdl.so.1 0 17560 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17560 python 7 /.machine 0 17560 python 7 /.hostid 0 17560 python 7 /var/ld/ld.config 0 17560 python 7 /usr/lib/secure/s8_preload.so.1 0 17560 python 7 /opt/csw/lib/libpython2.6.so.1.0 0 17560 python 7 /usr/lib/libresolv.so.2 0 17560 python 7 /usr/lib/libsocket.so.1 0 17560 python 7 /usr/lib/libnsl.so.1 0 17560 python 7 /usr/lib/librt.so.1 0 17560 python 7 /usr/lib/libdl.so.1 0 17560 python 7 /usr/lib/libpthread.so.1 0 17560 python 7 /usr/lib/libm.so.1 0 17560 python 7 /usr/lib/libc.so.1 0 17560 python 7 /usr/lib/libmp.so.2 0 17560 python 7 /usr/lib/libaio.so.1 0 17560 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17560 python 7 /usr/lib/libthread.so.1 0 17560 python -1 /opt/csw/lib/python/warnings.so 0 17560 python -1 /opt/csw/lib/python/warningsmodule.so 0 17560 python 7 /opt/csw/lib/python/warnings.py 0 17560 python 8 /opt/csw/lib/python/warnings.pyc 0 17560 python -1 /opt/csw/lib/python/linecache.so 0 17560 python -1 /opt/csw/lib/python/linecachemodule.so 0 17560 python 8 /opt/csw/lib/python/linecache.py 0 17560 python 10 /opt/csw/lib/python/linecache.pyc 0 17560 python -1 /opt/csw/lib/python/os.so 0 17560 python -1 /opt/csw/lib/python/osmodule.so 0 17560 python 10 /opt/csw/lib/python/os.py 0 17560 python 12 /opt/csw/lib/python/os.pyc 0 17560 python -1 /opt/csw/lib/python/posixpath.so 0 17560 python -1 /opt/csw/lib/python/posixpathmodule.so 0 17560 python 12 /opt/csw/lib/python/posixpath.py 0 17560 python 13 /opt/csw/lib/python/posixpath.pyc 0 17560 python -1 /opt/csw/lib/python/stat.so 0 17560 python -1 /opt/csw/lib/python/statmodule.so 0 17560 python 13 /opt/csw/lib/python/stat.py 0 17560 python 14 /opt/csw/lib/python/stat.pyc 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/usercustomize.so 0 17560 python -1 /opt/csw/lib/python/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/usercustomize.py 0 17560 python -1 /opt/csw/lib/python/usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-old/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomize.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomize.py 0 17560 python -1 /opt/csw/lib/python/site-packages/ usercustomize.pyc 0 17560 python -1 /opt/csw/lib/python/genericpath.so 0 17560 python -1 /opt/csw/lib/python/genericpathmodule.so 0 17560 python 13 /opt/csw/lib/python/genericpath.py 0 17560 python 14 /opt/csw/lib/python/genericpath.pyc 0 17560 python -1 /opt/csw/lib/python/UserDict.so 0 17560 python -1 /opt/csw/lib/python/UserDictmodule.so 0 17560 python 12 /opt/csw/lib/python/UserDict.py 0 17560 python 13 /opt/csw/lib/python/UserDict.pyc 0 17560 python -1 /opt/csw/lib/python/_abcoll.so 0 17560 python -1 /opt/csw/lib/python/_abcollmodule.so 0 17560 python 13 /opt/csw/lib/python/_abcoll.py 0 17560 python 14 /opt/csw/lib/python/_abcoll.pyc 0 17560 python -1 /opt/csw/lib/python/abc.so 0 17560 python -1 /opt/csw/lib/python/abcmodule.so 0 17560 python 14 /opt/csw/lib/python/abc.py 0 17560 python 15 /opt/csw/lib/python/abc.pyc 0 17560 python -1 /opt/csw/lib/python/copy_reg.so 0 17560 python -1 /opt/csw/lib/python/copy_regmodule.so 0 17560 python 12 /opt/csw/lib/python/copy_reg.py 0 17560 python 13 /opt/csw/lib/python/copy_reg.pyc 0 17560 python -1 /opt/csw/lib/python/types.so 0 17560 python -1 /opt/csw/lib/python/typesmodule.so 0 17560 python 13 /opt/csw/lib/python/types.py 0 17560 python 14 /opt/csw/lib/python/types.pyc 0 17560 python -1 /opt/csw/lib/python/re.so 0 17560 python -1 /opt/csw/lib/python/remodule.so 0 17560 python 8 /opt/csw/lib/python/re.py 0 17560 python 10 /opt/csw/lib/python/re.pyc 0 17560 python -1 /opt/csw/lib/python/sre_compile.so 0 17560 python -1 /opt/csw/lib/python/sre_compilemodule.so 0 17560 python 10 /opt/csw/lib/python/sre_compile.py 0 17560 python 12 /opt/csw/lib/python/sre_compile.pyc 0 17560 python -1 /opt/csw/lib/python/sre_parse.so 0 17560 python -1 /opt/csw/lib/python/sre_parsemodule.so 0 17560 python 12 /opt/csw/lib/python/sre_parse.py 0 17560 python 13 /opt/csw/lib/python/sre_parse.pyc 0 17560 python -1 /opt/csw/lib/python/sre_constants.so 0 17560 python -1 /opt/csw/lib/python/ sre_constantsmodule.so 0 17560 python 13 /opt/csw/lib/python/sre_constants.py 0 17560 python 14 /opt/csw/lib/python/sre_constants.pyc 0 17560 python -1 /opt/csw/lib/python/site.so 0 17560 python -1 /opt/csw/lib/python/sitemodule.so 0 17560 python 7 /opt/csw/lib/python/site.py 0 17560 python 8 /opt/csw/lib/python/site.pyc 0 17560 python 8 /etc/.name_service_door 0 17560 python 10 /opt/csw/lib/python/site-packages 0 17560 python -1 /opt/csw/lib/python/sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomizemodule.so 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.py 0 17560 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/__init__.so 0 17560 python -1 /opt/csw/lib/python/encodings/ __init__module.so 0 17560 python 7 /opt/csw/lib/python/encodings/__init__.py 0 17560 python 10 /opt/csw/lib/python/encodings/ __init__.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/codecs.so 0 17560 python -1 /opt/csw/lib/python/encodings/ codecsmodule.so 0 17560 python -1 /opt/csw/lib/python/encodings/codecs.py 0 17560 python -1 /opt/csw/lib/python/encodings/codecs.pyc 0 17560 python -1 /opt/csw/lib/python/codecs.so 0 17560 python -1 /opt/csw/lib/python/codecsmodule.so 0 17560 python 10 /opt/csw/lib/python/codecs.py 0 17560 python 12 /opt/csw/lib/python/codecs.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/ encodings.so 0 17560 python -1 /opt/csw/lib/python/encodings/ encodingsmodule.so 0 17560 python -1 /opt/csw/lib/python/encodings/ encodings.py 0 17560 python -1 /opt/csw/lib/python/encodings/ encodings.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/aliases.so 0 17560 python -1 /opt/csw/lib/python/encodings/ aliasesmodule.so 0 17560 python 10 /opt/csw/lib/python/encodings/aliases.py 0 17560 python 12 /opt/csw/lib/python/encodings/aliases.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__.so 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__module.so 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__.py 0 17560 python -1 /opt/csw/lib/python/encodings/ __builtin__.pyc 0 17560 python -1 /opt/csw/lib/python/encodings/ascii.so 0 17560 python -1 /opt/csw/lib/python/encodings/ asciimodule.so 0 17560 python 7 /opt/csw/lib/python/encodings/ascii.py 0 17560 python 10 /opt/csw/lib/python/encodings/ascii.pyc 0 17560 python 7 /opt/csw/lib/python/compileall.py 0 17560 python 7 /opt/csw/lib/python/compileall.py 0 17560 python -1 /opt/csw/lib/python/py_compile.so 0 17560 python -1 /opt/csw/lib/python/py_compilemodule.so 0 17560 python 7 /opt/csw/lib/python/py_compile.py 0 17560 python 10 /opt/csw/lib/python/py_compile.pyc 0 17560 python -1 /opt/csw/lib/python/traceback.so 0 17560 python -1 /opt/csw/lib/python/tracebackmodule.so 0 17560 python 10 /opt/csw/lib/python/traceback.py 0 17560 python 12 /opt/csw/lib/python/traceback.pyc 0 17560 python -1 /opt/csw/lib/python/getopt.so 0 17560 python -1 /opt/csw/lib/python/getoptmodule.so 0 17560 python 7 /opt/csw/lib/python/getopt.py 0 17560 python 10 /opt/csw/lib/python/getopt.pyc 0 17560 python 7 /opt/csw/lib/python 0 17560 python 7 /opt/csw/lib/python/bsddb 0 17560 python 7 /opt/csw/lib/python/compiler 0 17560 python 7 /opt/csw/lib/python/ctypes 0 17560 python 7 /opt/csw/lib/python/ctypes/macholib 0 17560 python 7 /opt/csw/lib/python/ctypes/test 0 17560 python 7 /opt/csw/lib/python/bsddb/test 0 17560 python 7 /opt/csw/lib/python/curses 0 17560 python 7 /opt/csw/lib/python/email 0 17560 python 7 /opt/csw/lib/python/email/mime 0 17560 python 7 /opt/csw/lib/python/email/test 0 17560 python 7 /opt/csw/lib/python/email/test/data 0 17560 python 7 /opt/csw/lib/python/encodings 0 17560 python 7 /opt/csw/lib/python/hotshot 0 17560 python 7 /opt/csw/lib/python/json 0 17560 python 7 /opt/csw/lib/python/json/tests 0 17560 python 7 /opt/csw/lib/python/lib-dynload 0 17560 python 7 /opt/csw/lib/python/lib-old 0 17560 python 7 /opt/csw/lib/python/lib2to3 0 17560 python 7 /opt/csw/lib/python/lib2to3/fixes 0 17560 python 7 /opt/csw/lib/python/lib2to3/pgen2 0 17560 python 7 /opt/csw/lib/python/lib2to3/tests 0 17560 python 7 /opt/csw/lib/python/logging 0 17560 python 7 /opt/csw/lib/python/multiprocessing 0 17560 python 7 /opt/csw/lib/python/multiprocessing/dummy 0 17560 python 7 /opt/csw/lib/python/plat-sunos5 0 17560 python 7 /opt/csw/lib/python/site-packages 0 17560 python 7 /opt/csw/lib/python/sqlite3 0 17560 python 7 /opt/csw/lib/python/sqlite3/test 0 17560 python 7 /opt/csw/lib/python/test 0 17560 python 7 /opt/csw/lib/python/test/decimaltestdata 0 17560 python 7 /opt/csw/lib/python/wsgiref 0 17560 python 7 /opt/csw/lib/python/xml 0 17561 python 7 /var/ld/ld.config 0 17561 python 7 /usr/lib/libc.so.1 0 17561 python 7 /usr/lib/libdl.so.1 0 17561 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17561 python 7 /.machine 0 17561 python 7 /.hostid 0 17561 python 7 /var/ld/ld.config 0 17561 python 7 /usr/lib/secure/s8_preload.so.1 0 17561 python 7 /opt/csw/lib/libpython2.6.so.1.0 0 17561 python 7 /usr/lib/libresolv.so.2 0 17561 python 7 /usr/lib/libsocket.so.1 0 17561 python 7 /usr/lib/libnsl.so.1 0 17561 python 7 /usr/lib/librt.so.1 0 17561 python 7 /usr/lib/libdl.so.1 0 17561 python 7 /usr/lib/libpthread.so.1 0 17561 python 7 /usr/lib/libm.so.1 0 17561 python 7 /usr/lib/libc.so.1 0 17561 python 7 /usr/lib/libmp.so.2 0 17561 python 7 /usr/lib/libaio.so.1 0 17561 python 7 /usr/platform/SUNW,SPARC-Enterprise- T5220/lib/libc_psr.so.1 0 17561 python 7 /usr/lib/libthread.so.1 0 17561 python -1 /opt/csw/lib/python/warnings.so 0 17561 python -1 /opt/csw/lib/python/warningsmodule.so 0 17561 python 7 /opt/csw/lib/python/warnings.py 0 17561 python 8 /opt/csw/lib/python/warnings.pyo 0 17561 python -1 /opt/csw/lib/python/linecache.so 0 17561 python -1 /opt/csw/lib/python/linecachemodule.so 0 17561 python 8 /opt/csw/lib/python/linecache.py 0 17561 python 10 /opt/csw/lib/python/linecache.pyo 0 17561 python -1 /opt/csw/lib/python/os.so 0 17561 python -1 /opt/csw/lib/python/osmodule.so 0 17561 python 10 /opt/csw/lib/python/os.py 0 17561 python 12 /opt/csw/lib/python/os.pyo 0 17561 python -1 /opt/csw/lib/python/posixpath.so 0 17561 python -1 /opt/csw/lib/python/posixpathmodule.so 0 17561 python 12 /opt/csw/lib/python/posixpath.py 0 17561 python 13 /opt/csw/lib/python/posixpath.pyo 0 17561 python -1 /opt/csw/lib/python/stat.so 0 17561 python -1 /opt/csw/lib/python/statmodule.so 0 17561 python 13 /opt/csw/lib/python/stat.py 0 17561 python 14 /opt/csw/lib/python/stat.pyo 0 17560 python 7 /opt/csw/lib/python/xml/dom 0 17560 python 7 /opt/csw/lib/python/xml/etree 0 17560 python 7 /opt/csw/lib/python/xml/parsers 0 17560 python 7 /opt/csw/lib/python/xml/sax 0 17561 python -1 /opt/csw/lib/python/genericpath.so 0 17561 python -1 /opt/csw/lib/python/genericpathmodule.so 0 17561 python 13 /opt/csw/lib/python/genericpath.py 0 17561 python 14 /opt/csw/lib/python/genericpath.pyo 0 17561 python -1 /opt/csw/lib/python/UserDict.so 0 17561 python -1 /opt/csw/lib/python/UserDictmodule.so 0 17561 python 12 /opt/csw/lib/python/UserDict.py 0 17561 python 13 /opt/csw/lib/python/UserDict.pyo 0 17561 python -1 /opt/csw/lib/python/_abcoll.so 0 17561 python -1 /opt/csw/lib/python/_abcollmodule.so 0 17561 python 13 /opt/csw/lib/python/_abcoll.py 0 17561 python 14 /opt/csw/lib/python/_abcoll.pyo 0 17561 python -1 /opt/csw/lib/python/abc.so 0 17561 python -1 /opt/csw/lib/python/abcmodule.so 0 17561 python 14 /opt/csw/lib/python/abc.py 0 17561 python 15 /opt/csw/lib/python/abc.pyo 0 17561 python -1 /opt/csw/lib/python/copy_reg.so 0 17561 python -1 /opt/csw/lib/python/copy_regmodule.so 0 17561 python 12 /opt/csw/lib/python/copy_reg.py 0 17561 python 13 /opt/csw/lib/python/copy_reg.pyo 0 17561 python -1 /opt/csw/lib/python/types.so 0 17561 python -1 /opt/csw/lib/python/typesmodule.so 0 17561 python 13 /opt/csw/lib/python/types.py 0 17561 python 14 /opt/csw/lib/python/types.pyo 0 17561 python -1 /opt/csw/lib/python/re.so 0 17561 python -1 /opt/csw/lib/python/remodule.so 0 17561 python 8 /opt/csw/lib/python/re.py 0 17561 python 10 /opt/csw/lib/python/re.pyo 0 17561 python -1 /opt/csw/lib/python/sre_compile.so 0 17561 python -1 /opt/csw/lib/python/sre_compilemodule.so 0 17561 python 10 /opt/csw/lib/python/sre_compile.py 0 17561 python 12 /opt/csw/lib/python/sre_compile.pyo 0 17561 python -1 /opt/csw/lib/python/sre_parse.so 0 17561 python -1 /opt/csw/lib/python/sre_parsemodule.so 0 17561 python 12 /opt/csw/lib/python/sre_parse.py 0 17561 python 13 /opt/csw/lib/python/sre_parse.pyo 0 17561 python -1 /opt/csw/lib/python/sre_constants.so 0 17561 python -1 /opt/csw/lib/python/ sre_constantsmodule.so 0 17561 python 13 /opt/csw/lib/python/sre_constants.py 0 17561 python 14 /opt/csw/lib/python/sre_constants.pyo 0 17561 python -1 /opt/csw/lib/python/site.so 0 17561 python -1 /opt/csw/lib/python/sitemodule.so 0 17561 python 7 /opt/csw/lib/python/site.py 0 17561 python 8 /opt/csw/lib/python/site.pyo 0 17561 python 8 /etc/.name_service_door 0 17561 python 10 /opt/csw/lib/python/site-packages 0 17561 python -1 /opt/csw/lib/python/sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-old/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.py 0 17561 python -1 /opt/csw/lib/python/site-packages/ sitecustomize.pyo 0 17561 python -1 /opt/csw/lib/python/usercustomize.so 0 17561 python -1 /opt/csw/lib/python/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/usercustomize.py 0 17561 python -1 /opt/csw/lib/python/usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/plat-sunos5/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-old/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/lib-dynload/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomize.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomizemodule.so 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomize.py 0 17561 python -1 /opt/csw/lib/python/site-packages/ usercustomize.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/__init__.so 0 17561 python -1 /opt/csw/lib/python/encodings/ __init__module.so 0 17561 python 7 /opt/csw/lib/python/encodings/__init__.py 0 17561 python 10 /opt/csw/lib/python/encodings/ __init__.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/codecs.so 0 17561 python -1 /opt/csw/lib/python/encodings/ codecsmodule.so 0 17561 python -1 /opt/csw/lib/python/encodings/codecs.py 0 17561 python -1 /opt/csw/lib/python/encodings/codecs.pyo 0 17561 python -1 /opt/csw/lib/python/codecs.so 0 17561 python -1 /opt/csw/lib/python/codecsmodule.so 0 17561 python 10 /opt/csw/lib/python/codecs.py 0 17561 python 12 /opt/csw/lib/python/codecs.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/ encodings.so 0 17561 python -1 /opt/csw/lib/python/encodings/ encodingsmodule.so 0 17561 python -1 /opt/csw/lib/python/encodings/ encodings.py 0 17561 python -1 /opt/csw/lib/python/encodings/ encodings.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/aliases.so 0 17561 python -1 /opt/csw/lib/python/encodings/ aliasesmodule.so 0 17561 python 10 /opt/csw/lib/python/encodings/aliases.py 0 17561 python 12 /opt/csw/lib/python/encodings/aliases.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__.so 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__module.so 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__.py 0 17561 python -1 /opt/csw/lib/python/encodings/ __builtin__.pyo 0 17561 python -1 /opt/csw/lib/python/encodings/ascii.so 0 17561 python -1 /opt/csw/lib/python/encodings/ asciimodule.so 0 17561 python 7 /opt/csw/lib/python/encodings/ascii.py 0 17561 python 10 /opt/csw/lib/python/encodings/ascii.pyo 0 17561 python 7 /opt/csw/lib/python/compileall.py 0 17561 python 7 /opt/csw/lib/python/compileall.py 0 17561 python -1 /opt/csw/lib/python/py_compile.so 0 17561 python -1 /opt/csw/lib/python/py_compilemodule.so 0 17561 python 7 /opt/csw/lib/python/py_compile.py 0 17561 python 10 /opt/csw/lib/python/py_compile.pyo 0 17561 python -1 /opt/csw/lib/python/traceback.so 0 17561 python -1 /opt/csw/lib/python/tracebackmodule.so 0 17561 python 10 /opt/csw/lib/python/traceback.py 0 17561 python 12 /opt/csw/lib/python/traceback.pyo 0 17561 python -1 /opt/csw/lib/python/getopt.so 0 17561 python -1 /opt/csw/lib/python/getoptmodule.so 0 17561 python 7 /opt/csw/lib/python/getopt.py 0 17561 python 10 /opt/csw/lib/python/getopt.pyo 0 17561 python 7 /opt/csw/lib/python 0 17561 python 7 /opt/csw/lib/python/bsddb 0 17561 python 7 /opt/csw/lib/python/bsddb/test 0 17561 python 7 /opt/csw/lib/python/compiler 0 17561 python 7 /opt/csw/lib/python/ctypes 0 17561 python 7 /opt/csw/lib/python/ctypes/macholib 0 17561 python 7 /opt/csw/lib/python/ctypes/test 0 17561 python 7 /opt/csw/lib/python/curses 0 17561 python 7 /opt/csw/lib/python/email 0 17561 python 7 /opt/csw/lib/python/email/mime 0 17561 python 7 /opt/csw/lib/python/email/test 0 17561 python 7 /opt/csw/lib/python/email/test/data 0 17561 python 7 /opt/csw/lib/python/encodings 0 17561 python 7 /opt/csw/lib/python/hotshot 0 17561 python 7 /opt/csw/lib/python/json 0 17561 python 7 /opt/csw/lib/python/json/tests 0 17561 python 7 /opt/csw/lib/python/lib-dynload 0 17561 python 7 /opt/csw/lib/python/lib-old 0 17561 python 7 /opt/csw/lib/python/lib2to3 0 17561 python 7 /opt/csw/lib/python/lib2to3/fixes 0 17561 python 7 /opt/csw/lib/python/lib2to3/pgen2 0 17561 python 7 /opt/csw/lib/python/lib2to3/tests 0 17561 python 7 /opt/csw/lib/python/logging 0 17561 python 7 /opt/csw/lib/python/multiprocessing 0 17561 python 7 /opt/csw/lib/python/multiprocessing/dummy 0 17561 python 7 /opt/csw/lib/python/plat-sunos5 0 17561 python 7 /opt/csw/lib/python/site-packages 0 17561 python 7 /opt/csw/lib/python/sqlite3 0 17561 python 7 /opt/csw/lib/python/sqlite3/test 0 17561 python 7 /opt/csw/lib/python/test 0 17561 python 7 /opt/csw/lib/python/test/decimaltestdata 0 17561 python 7 /opt/csw/lib/python/wsgiref 0 17561 python 7 /opt/csw/lib/python/xml 0 17561 python 7 /opt/csw/lib/python/xml/dom 0 17561 python 7 /opt/csw/lib/python/xml/etree 0 17561 python 7 /opt/csw/lib/python/xml/parsers 0 17561 python 7 /opt/csw/lib/python/xml/sax 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/__init__.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/error.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/expr.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/matcher.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/state.py 0 17546 pkginstall 7 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/xtypes.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/__init__.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/error.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/expr.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/matcher.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/state.py 0 17546 pkginstall 13 /opt/csw/X11/lib/python2.6/site- packages/xcbgen/xtypes.py 0 17567 rm 11 python2.6 ^CBroken Pipe From phil at bolthole.com Sun May 31 23:07:04 2009 From: phil at bolthole.com (Philip Brown) Date: Sun, 31 May 2009 14:07:04 -0700 Subject: [csw-maintainers] Fwd: [csw-users] feedback on pkg-get behaviour change requested In-Reply-To: References: Message-ID: <20090531210704.GA55699@bolthole.com> On Sun, May 31, 2009 at 10:53:32PM +0200, Dagobert Michelsen wrote: > Hi, > > this was just on users at . > Gary: what do you think? I guess nobody was aware of this issue. oh. oops. I will change the behaviour of pkg-get, so that pkg-get -c pkgname acts differently than pkg-get -c >From the perspective of puppet, that will restore the original behaviour.