From william at wbonnet.net  Wed Jul  1 00:07:29 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Jul 2009 00:07:29 +0200
Subject: [csw-maintainers] New in testing: gtk2 2.16.1
In-Reply-To: <4A4A595A.3010304@opencsw.org>
References: <A1A966C3-001F-4777-8B3C-0830A5735370@opencsw.org>	<4A4A3B04.2040701@opencsw.org>	<4A4A5204.7070705@opencsw.org>
	<4A4A595A.3010304@opencsw.org>
Message-ID: <4A4A8CA1.7090605@wbonnet.net>

Hi

> While playing with pygtk, I found a "nicety" for the package install.
> have a post install that runs:
>
> /opt/csw/bin/gdk-pixbuf-query-loaders >/opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders
>
> I am going to file a feature bug against the pango package to request the same
> for /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules
>   
The postinstall script is under svn. Package has to be rebuild to 
include 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 pfelecan at opencsw.org  Wed Jul  1 09:48:35 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 01 Jul 2009 09:48:35 +0200
Subject: [csw-maintainers] tcsh 6.16.00 in testing
In-Reply-To: <15954.131.203.105.1.1246399049.squirrel@webmail.jessies.org>
	(Chris Reece's message of "Wed\,
	1 Jul 2009 09\:57\:29 +1200 \(NZST\)")
References: <15954.131.203.105.1.1246399049.squirrel@webmail.jessies.org>
Message-ID: <x663ec3kto.fsf@abhavana.free.fr>

"Chris Reece" <car at opencsw.org> writes:

> On Tue, June 30, 2009 11:36 pm, Peter Bonivart wrote:
>> It was years ago I used tcsh but it installs/removes cleanly and seems
>> to work properly during my short test on a Solaris 9 Sparc and a Solaris 10
>> x86. Man page also works.
>
> Thanks, Peter, much obliged.
>
> My only remaining problem is that I can't log in to www.opencsw.org, which the
> documentation suggests is the location of /home/newpkgs.  None of my keys are
> accepted.  Or, rather, there are "Too many authentication failures" after all
> of my keys have been tried.
>
> I can log in to login.opencsw.org, and from there to build8s and build8x.

>From login.opencsw.org you can scp your packages to
www.opencsw.org:/home/newpkgs
-- 
Peter

From william at wbonnet.net  Wed Jul  1 14:32:00 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Jul 2009 14:32:00 +0200
Subject: [csw-maintainers] Conflicting header file
Message-ID: <4A4B5740.90407@wbonnet.net>

Hi

A conflicting header is installed by qt package. It is the file 
"/opt/csw/include/npapi.h"

When including this file, it generates the following error

"/opt/csw/include/npapi.h", line 1: Error: #error 
"/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
must be provided by Netscape".

A mantis entry exists, but unfortnatly package has no maintainer and was 
not fixed. Would it be possible to nuke this file on the build farm 
please ? (well at least rename the file to .old...). This file is 
blocking for compile FF3.

npapi.h is provided by mozilla packages, especialy firefox, not by qt.

[wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$ grep 
npapi.h /var/sadm/install/contents
/opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908 CSWqt
/opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273 
1110647274 CSWqtgcc

Thanks in advance

cheers
W.



From dam at opencsw.org  Wed Jul  1 14:55:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Jul 2009 14:55:34 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A4B5740.90407@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
Message-ID: <595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>

Hi William,

Am 01.07.2009 um 14:32 schrieb William Bonnet:
> A conflicting header is installed by qt package. It is the file "/ 
> opt/csw/include/npapi.h"
>
> When including this file, it generates the following error
>
> "/opt/csw/include/npapi.h", line 1: Error: #error "/home/espenr/tmp/ 
> qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h must be  
> provided by Netscape".
>
> A mantis entry exists, but unfortnatly package has no maintainer and  
> was not fixed. Would it be possible to nuke this file on the build  
> farm please ? (well at least rename the file to .old...). This file  
> is blocking for compile FF3.
>
> npapi.h is provided by mozilla packages, especialy firefox, not by qt.
>
> [wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$  
> grep npapi.h /var/sadm/install/contents
> /opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908  
> CSWqt
> /opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273  
> 1110647274 CSWqtgcc

The file does not include anything of interest:

dam at build9x :/home/dam > more /opt/csw/include/npapi.h
#error "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/ 
npapi.h must be provided by Netscape"
#error "This file is just a placeholder. Please see the documentation"
#error "to learn how to obtain the real file"

Moved to npapi.h.old on build9x only for now.


Best regards

   -- Dago

From dam at opencsw.org  Wed Jul  1 15:02:58 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Jul 2009 15:02:58 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A4B5740.90407@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
Message-ID: <6798D88D-CFE6-4A1F-872A-3082CD490124@opencsw.org>

Hi William,

Am 01.07.2009 um 14:32 schrieb William Bonnet:
> A conflicting header is installed by qt package. It is the file "/ 
> opt/csw/include/npapi.h"
>
> When including this file, it generates the following error
>
> "/opt/csw/include/npapi.h", line 1: Error: #error "/home/espenr/tmp/ 
> qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h must be  
> provided by Netscape".
>
> A mantis entry exists, but unfortnatly package has no maintainer and  
> was not fixed. Would it be possible to nuke this file on the build  
> farm please ? (well at least rename the file to .old...). This file  
> is blocking for compile FF3.
>
> npapi.h is provided by mozilla packages, especialy firefox, not by qt.
>
> [wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$  
> grep npapi.h /var/sadm/install/contents
> /opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908  
> CSWqt
> /opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273  
> 1110647274 CSWqtgcc

The file does not include anything of interest:

dam at build9x :/home/dam > more /opt/csw/include/npapi.h
#error "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/ 
npapi.h must be provided by Netscape"
#error "This file is just a placeholder. Please see the documentation"
#error "to learn how to obtain the real file"

Moved to npapi.h.old on build9x only for now.


Best regards

  -- Dago

From william at wbonnet.net  Wed Jul  1 16:07:58 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Jul 2009 16:07:58 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
Message-ID: <4A4B6DBE.5030606@wbonnet.net>

Hi
>
> The file does not include anything of interest:
>
> dam at build9x :/home/dam > more /opt/csw/include/npapi.h
> #error 
> "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
> must be provided by Netscape"
> #error "This file is just a placeholder. Please see the documentation"
> #error "to learn how to obtain the real file"
>
> Moved to npapi.h.old on build9x only for now.
>
Thanks. If it works on this box, i may ask for more after ;)

cheers
W.

From skayser at opencsw.org  Wed Jul  1 22:08:56 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 1 Jul 2009 22:08:56 +0200 (CEST)
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
 instead of PROTOTYPE_FILTER
Message-ID: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>

Hi,

while working on postfix 2.6 i have noticed again that it is not quite
straight-forward to assign files to classes. Instead of exposing the inner
workings of the prototype stuff to the Makefile, could we maybe get some
variables to assign files to classes (functions)?

So instead of having:

CONFIG_BASE = \/etc\/opt\/csw\/postfix\/
CONFIG_FILES = access aliases canonical generic header_checks
CONFIG_FILES += main.cf master.cf
CONFIG_FILES += relocated transport virtual

PROTOTYPE_FILTER  = awk '\
	$(foreach C,$(prefix $(CONFIG_BASE),$(CONFIG_FILES)), \
		$$$$3 ~ /^$(C)$$$$/ { $$$$2 = "cswcpsampleconf" }) \
        ....

just have something like:

CONFIG_FILES = /etc/default/cswpostgrey
CONFIG_FILES += /etc/opt/csw/postfix/access
CONFIG_FILES += /etc/opt/csw/postfix/aliases
CONFIG_FILES += /etc/opt/csw/postfix/canonical
CONFIG_FILES += /etc/opt/csw/postfix/generic
CONFIG_FILES += /etc/opt/csw/postfix/header_checks
CONFIG_FILES += /etc/opt/csw/postfix/main.cf
CONFIG_FILES += /etc/opt/csw/postfix/master.cf
CONFIG_FILES += /etc/opt/csw/postfix/relocated
CONFIG_FILES += /etc/opt/csw/postfix/transport
CONFIG_FILES += /etc/opt/csw/postfix/virtual

To me this would feel more declarative instead of the procedural way we do
it with PROTOTYE_FILTER right now (note, it can be shortened with
$(foreach and $(prefix, just like above). It would not only tag the class,
but also wrap the other housekeeping, like moving the config files to .CSW
during the build process.

(Side note: Thinking ahead, it could also be an conceptual abstraction
level WRT to IPS: Do IPS packages still have a prototype file?)

This is only an example for config files, which i encounter most often.
Other classes could be "wrapped" as well. One would just have to come up
with the requirements, a proper naming scheme for the variables, and ....
actually implement it in GAR.

Sebastian


From dam at opencsw.org  Wed Jul  1 22:21:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Jul 2009 22:21:07 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
Message-ID: <3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>

Hi Sebastian,

Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
> while working on postfix 2.6 i have noticed again that it is not quite
> straight-forward to assign files to classes. Instead of exposing the  
> inner
> workings of the prototype stuff to the Makefile, could we maybe get  
> some
> variables to assign files to classes (functions)?
>
> So instead of having:
>
> CONFIG_BASE = \/etc\/opt\/csw\/postfix\/
> CONFIG_FILES = access aliases canonical generic header_checks
> CONFIG_FILES += main.cf master.cf
> CONFIG_FILES += relocated transport virtual
>
> PROTOTYPE_FILTER  = awk '\
> 	$(foreach C,$(prefix $(CONFIG_BASE),$(CONFIG_FILES)), \
> 		$$$$3 ~ /^$(C)$$$$/ { $$$$2 = "cswcpsampleconf" }) \
>        ....
>
> just have something like:
>
> CONFIG_FILES = /etc/default/cswpostgrey
> CONFIG_FILES += /etc/opt/csw/postfix/access
> CONFIG_FILES += /etc/opt/csw/postfix/aliases
> CONFIG_FILES += /etc/opt/csw/postfix/canonical
> CONFIG_FILES += /etc/opt/csw/postfix/generic
> CONFIG_FILES += /etc/opt/csw/postfix/header_checks
> CONFIG_FILES += /etc/opt/csw/postfix/main.cf
> CONFIG_FILES += /etc/opt/csw/postfix/master.cf
> CONFIG_FILES += /etc/opt/csw/postfix/relocated
> CONFIG_FILES += /etc/opt/csw/postfix/transport
> CONFIG_FILES += /etc/opt/csw/postfix/virtual
>
> To me this would feel more declarative instead of the procedural way  
> we do
> it with PROTOTYE_FILTER right now (note, it can be shortened with
> $(foreach and $(prefix, just like above). It would not only tag the  
> class,
> but also wrap the other housekeeping, like moving the config files  
> to .CSW
> during the build process.
>
> (Side note: Thinking ahead, it could also be an conceptual abstraction
> level WRT to IPS: Do IPS packages still have a prototype file?)
>
> This is only an example for config files, which i encounter most  
> often.
> Other classes could be "wrapped" as well. One would just have to  
> come up
> with the requirements, a proper naming scheme for the variables,  
> and ....
> actually implement it in GAR.

For CSWcswclassutils this is already in there :-) Just define

SAMPLECONF
PRESERVECONF
INITSMF

and everything else will be taken cared of.


Bestregards

   -- Dago

From skayser at opencsw.org  Thu Jul  2 00:05:23 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 2 Jul 2009 00:05:23 +0200 (CEST)
Subject: [csw-maintainers] Heads up: GAR documentation main page content
 simplified (gar.sf.net)
Message-ID: <55969.85.88.7.146.1246485923.squirrel@ssl.skayser.de>

Hi,

to better guide GAR novices i have just simplified the main GAR
documentation page [1]. What changed?

- The advanced topics have been moved to a separate page [2] and linked to
  from the main site.

- The "revision needed" and "scratchpad style" pages have been moved to
  a separate page [3] that is accessible via the index (upper right).

This is only an attempt of how it could look like and was born because i
have seen a couple of GAR beginners struggle with the "too many"
documentation pages. The main page now doesn't throw everything in their
way, but tries to get them going step by step.

I assume, advanced GAR users are likely to still find what they are
looking for (and there is still the index to search across all page
topics). Feedback, thoughts, and ideas welcome.

Sebastian

[1] http://gar.sf.net, http://sourceforge.net/apps/trac/gar/
[2] http://sourceforge.net/apps/trac/gar/wiki/Learning%20the%20details
[3] http://sourceforge.net/apps/trac/gar/wiki/AA_Notepad


From trygvis at opencsw.org  Thu Jul  2 10:00:03 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 02 Jul 2009 10:00:03 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
Message-ID: <4A4C6903.2040906@opencsw.org>

If someone feel like getting this mostly python package out, feel free.

I almost got 1.2 working but there where too many test cases that 
failed. Most of the failures was because of missing gnu'isms in the 
tools the test used from what I could tell. Would be nice to patch up 
the test suite to be able to control the path to the tools it uses.

--
Trygve
-------------- next part --------------
An embedded message was scrubbed...
From: Upstream Package Watch <uwatch at mirror.opencsw.org>
Subject: [svn] mercurial upstream update notification
Date: Thu, 2 Jul 2009 02:44:21 +0200 (CEST)
Size: 2306
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/aa41526a/attachment.eml>

From schwindt at dfki.uni-kl.de  Thu Jul  2 12:18:55 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Thu, 02 Jul 2009 12:18:55 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
Message-ID: <200907021018.n62AIucZ003707@dfki.uni-kl.de>


Starting with libpango-1.24.3,REV=2009.06.27 the content of i.e. 
windowtitles get garbled. Using xfce4 as a windowmanager gives
you only empty squares where gtk normaly would put text.

This happens on a system with a plain csw-install, but also 
on a test machine where I upgrade some packages by hand like libconv-1.13,
xfce4.6.

Am I doing something wrong ? Or is this fixed by upgrading to versions from 
testing ?


Nicolai




From william at wbonnet.net  Thu Jul  2 12:24:49 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 02 Jul 2009 12:24:49 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: <200907021018.n62AIucZ003707@dfki.uni-kl.de>
References: <200907021018.n62AIucZ003707@dfki.uni-kl.de>
Message-ID: <4A4C8AF1.6030508@wbonnet.net>

Hi Nicolai

Does the following file exist ?

/opt/csw/etc/pango/pango.modules

if not please run

/opt/csw/bin/pango-querymodules > /opt/csw/etc/pango/pango.modules

This issue is under fixing...

chers
W.
> Starting with libpango-1.24.3,REV=2009.06.27 the content of i.e. 
> windowtitles get garbled. Using xfce4 as a windowmanager gives
> you only empty squares where gtk normaly would put text.
>
> This happens on a system with a plain csw-install, but also 
> on a test machine where I upgrade some packages by hand like libconv-1.13,
> xfce4.6.
>
> Am I doing something wrong ? Or is this fixed by upgrading to versions from 
> testing ?
>
>
> Nicolai
>
>
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>   


From bwalton at opencsw.org  Thu Jul  2 19:05:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 13:05:01 -0400
Subject: [csw-maintainers] package hooks (update)
Message-ID: <1246554093-sup-189@ntdws12.chass.utoronto.ca>


Hi All,

I've tweaked the package hooks rfc slightly and would appreciate any
feedback.  Changes are:

1. Use /var/tmp for file storage of package lists.  I think storing
   temp data in /var/opt/csw/pkg-hooks was wrong, since it's not long
   lived.
2. Renamed and expanded the list of proposed hooks.  I think the new
   scheme adds better clarity of purpose to the names of the hooks.

http://wiki.opencsw.org/package-hooks

If this is acceptable, I'm going to go ahead and begin the
implementation.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/66fb6e85/attachment.asc>

From phil at bolthole.com  Thu Jul  2 19:46:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 10:46:57 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
Message-ID: <20090702174657.GJ84645@bolthole.com>

On Thu, Jul 02, 2009 at 01:05:01PM -0400, Ben Walton wrote:
> 
> Hi All,
> 
> I've tweaked the package hooks rfc slightly and would appreciate any
> feedback.  Changes are:
> 
> 1. Use /var/tmp for file storage of package lists.  I think storing
>    temp data in /var/opt/csw/pkg-hooks was wrong, since it's not long
>    lived.


package lists?


From phil at bolthole.com  Thu Jul  2 19:51:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 10:51:32 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
Message-ID: <20090702175132.GK84645@bolthole.com>

Also...

On Thu, Jul 02, 2009 at 01:05:01PM -0400, Ben Walton wrote:
> http://wiki.opencsw.org/package-hooks

I dont see where it explicitly calls how how a package tool can
differentiate between "batch" and "per-package" hooks.

also, the communication between package tools and hooks, seems to be rather
vague.
An rfc normally deals with fairly concrete directives. and even when it is
vague, it is explicitly vague. that is to say, "must" vs "may" :-)

	
What I mean is, I think it would be beneficial to have the "communication"
section rewritten along the lines of,

[in situation X, the package tool should pass in argument Y]

Right now, it seems like package tools "may" pass in arguments to hooks,
basically on a whim. there is no rhyme or reason to it that i can see.


From mwatters at opencsw.org  Thu Jul  2 20:21:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Jul 2009 13:21:50 -0500
Subject: [csw-maintainers] mGar RFE *** new make target reset-source ***
Message-ID: <4A4CFABE.3040801@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dago,

could we get an option to reset the package back to post-extract status.

it should be fairly easy although, I am not sure how to deal with the
patches...  I think they make a backup of the original.

will do some more research on how the patches are done and how to revert them,
unless you have it figured 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

iEYEARECAAYFAkpM+r4ACgkQLrhmsXMSLxfiGgCgpr6bGcKYY1VvkGcAfHw963bQ
hmQAoIKGEw1kEtgde1Y9a7Hh0pj9OqTT
=aUG2
-----END PGP SIGNATURE-----

From dam at opencsw.org  Thu Jul  2 20:43:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 2 Jul 2009 20:43:46 +0200
Subject: [csw-maintainers] mGar RFE *** new make target reset-source ***
In-Reply-To: <4A4CFABE.3040801@opencsw.org>
References: <4A4CFABE.3040801@opencsw.org>
Message-ID: <38A01277-81E2-4B9E-B546-9C0E1E6B625B@opencsw.org>

Hi Mike,

Am 02.07.2009 um 20:21 schrieb Mike Watters:
> could we get an option to reset the package back to post-extract  
> status.

 From where? patch, configure or build?

> it should be fairly easy although, I am not sure how to deal with the
> patches...  I think they make a backup of the original.
>
> will do some more research on how the patches are done and how to  
> revert them,
> unless you have it figured out.   ;-)

The patching-thing is not solved well currently in GAR. I am thinking
of something like
   gmake preparepatch
which unpacks the sources multiple times like
   work/pristine-<mod>   Pristine, unpatched sources
   work/patched-<mod>    Patched sources
   work/build-<mod>      Patches sources that will be worked on
pristine-* will just be unpacked and always stay that way. As maintainer
you than start editing files in patched-*. gmake makepatch generates
a patch from the diffs between pristine-* and patched-*. After including
the patch in PATCHFILES you can do
   gmake repreparepatch
and pristine-* will contain the sources with the patches from PATCHFILES
applied and patched-* a copy from that to generate more patches.

Feel free to enhance this draft.


Best regards

   -- Dagp

From bwalton at opencsw.org  Thu Jul  2 20:59:28 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 14:59:28 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702175132.GK84645@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
Message-ID: <1246560996-sup-1808@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 13:51:32 -0400 2009:
> > http://wiki.opencsw.org/package-hooks

> I dont see where it explicitly calls how how a package tool can
> differentiate between "batch" and "per-package" hooks.

Well, if you read the description of the two groups of hooks, you
should see that it notes:

<snip>
Batch Hooks:

These hooks are run before or after a batch package operation. In this
sense, batch is taken to mean all package actions, regardless of the
number of packages affected.
</snip>

<snip>
Per-package Hooks:

These hooks are run before and after individual calls to pkgadd or
pkgrm.
</snip>

I've added some examples of the call sequences at the bottom of the
document in a 'Hook Invocation' section to solidify these
descriptions.

> also, the communication between package tools and hooks, seems to be rather
> vague.
> An rfc normally deals with fairly concrete directives. and even when it is
> vague, it is explicitly vague. that is to say, "must" vs "may" :-)

For better or worse.

> What I mean is, I think it would be beneficial to have the "communication"
> section rewritten along the lines of,
> [in situation X, the package tool should pass in argument Y]

Done.  See changes.

> Right now, it seems like package tools "may" pass in arguments to hooks,
> basically on a whim. there is no rhyme or reason to it that i can see.

Well, I thought it was clear, but I wrote it, so I knew what I meant!
:)

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/fae34ddb/attachment.asc>

From phil at bolthole.com  Thu Jul  2 21:05:30 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 12:05:30 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246560996-sup-1808@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
Message-ID: <20090702190530.GE94553@bolthole.com>

On Thu, Jul 02, 2009 at 02:59:28PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Thu Jul 02 13:51:32 -0400 2009:
> > > http://wiki.opencsw.org/package-hooks
> 
> > I dont see where it explicitly calls how how a package tool can
> > differentiate between "batch" and "per-package" hooks.
> ....
> Well, if you read the description of the two groups of hooks, you
> should see that it notes:
> 

Yes, it describes what [a batch hook] should do.
However, it does not explicitly describe how a package installation tool
should RECOGNIZE a batch hook, AS a batch hook, that i could see.

If the tool does not know which scripts are batch hooks, then it wont know
_what_ to call, even if it knows _when_ to call them! :-D

From bwalton at opencsw.org  Thu Jul  2 21:27:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 15:27:15 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702190530.GE94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
Message-ID: <1246562604-sup-5118@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 15:05:30 -0400 2009:
> Yes, it describes what [a batch hook] should do.
> However, it does not explicitly describe how a package installation tool
> should RECOGNIZE a batch hook, AS a batch hook, that i could see.

The tool doesn't need to _recognize_ anything.  The hook names are
defined.  Running a hook consists of testing for existence of a
specific directory (based on the name of the hook) and executing in
lexical order any files in that directory.  If the directory isn't
there, the tool continues.  So, since hooks have names, they're
'addressed' by name.  No need to recognize them.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/99b9fb85/attachment.asc>

From phil at bolthole.com  Thu Jul  2 21:39:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 12:39:50 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246562604-sup-5118@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
Message-ID: <20090702193950.GH94553@bolthole.com>

On Thu, Jul 02, 2009 at 03:27:15PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Thu Jul 02 15:05:30 -0400 2009:
> > Yes, it describes what [a batch hook] should do.
> > However, it does not explicitly describe how a package installation tool
> > should RECOGNIZE a batch hook, AS a batch hook, that i could see.
> 
> The tool doesn't need to _recognize_ anything.  The hook names are
> defined.  Running a hook consists of testing for existence of a
> specific directory (based on the name of the hook) and executing in
> lexical order any files in that directory.  If the directory isn't
> there, the tool continues.  So, since hooks have names, they're
> 'addressed' by name.  No need to recognize them.

Sigh. failure to communicate here.

http://wiki.opencsw.org/package-hooks


The general format of a hook script location, would seem to be

/etc/opt/csw/pkg-hooks/DIRECTORY/script-goes-here

but you have not explicitly defined the acceptible values for DIRECTORY.

You've given a few examples. But you have not explicitly mapped out the
full space.

you have not defined which DIRECTORY names should be used for per-package,
vs "batch" hooks, for example.




From william at wbonnet.net  Thu Jul  2 21:42:05 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 02 Jul 2009 21:42:05 +0200
Subject: [csw-maintainers] mGar RFE *** new make target reset-source ***
In-Reply-To: <38A01277-81E2-4B9E-B546-9C0E1E6B625B@opencsw.org>
References: <4A4CFABE.3040801@opencsw.org>
	<38A01277-81E2-4B9E-B546-9C0E1E6B625B@opencsw.org>
Message-ID: <4A4D0D8D.8050802@wbonnet.net>

Hi
> From where? patch, configure or build?
 From whatever :)

My personal need is the following :

. Let say I do work on a package with a large amount of sources (For 
example Firefox... ;) )
. I run a compile then choose to change some options in the Makefile 
used by GAR
. I would like to do a "make clean and reconfigure" with the lowest 
overhead possible.

Actually this process may take some time to make clean, download the 
sources again, and extract sources. It may be a question of hour for 
Firefox, Thunderbird or Eclipse. So any solution that may help to reduce 
this time would be great.

Most of the time i need this after a build, successful or not.
> which unpacks the sources multiple times like
>   work/pristine-<mod>   Pristine, unpatched sources
>   work/patched-<mod>    Patched sources
>   work/build-<mod>      Patches sources that will be worked on
I do really like this !

Many times i have been making cp foo.c foo.c.orig in my working folder.

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  Thu Jul  2 22:02:56 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 16:02:56 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702193950.GH94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
Message-ID: <1246563839-sup-2157@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 15:39:50 -0400 2009:
> Sigh. failure to communicate here.

I think so.

> http://wiki.opencsw.org/package-hooks

> The general format of a hook script location, would seem to be
> /etc/opt/csw/pkg-hooks/DIRECTORY/script-goes-here
> but you have not explicitly defined the acceptible values for
> DIRECTORY.

> You've given a few examples. But you have not explicitly mapped out the
> full space.

Not in terms of a specific list, no, but I have given the hook names,
noted that there is a pre and post for each hook and specified the
name format for the directories.

...Ok, the wiki now contains a _complete_, _explicit_ list of valid
hooks and the corresponding directories.

> you have not defined which DIRECTORY names should be used for per-package,
> vs "batch" hooks, for example.

Again, the grouping isn't important in any way other than the type of
argument passed.  Batch hooks get a file name, per-package hooks get a
package name.  Other than that, they all live happily in
/etc/opt/csw/pkg-hooks.  For reference, in case it's still not clear,
these directories would be:

/etc/opt/csw/pkg-hooks/prebatchadd.d/
/etc/opt/csw/pkg-hooks/postbatchadd.d/
/etc/opt/csw/pkg-hooks/prebatchremove.d/
/etc/opt/csw/pkg-hooks/postbatchremove.d/
/etc/opt/csw/pkg-hooks/preinstall.d/
/etc/opt/csw/pkg-hooks/postinstall.d/
/etc/opt/csw/pkg-hooks/preupdate .d/
/etc/opt/csw/pkg-hooks/postupdate.d/
/etc/opt/csw/pkg-hooks/preremove.d/
/etc/opt/csw/pkg-hooks/postremove.d/

I have also noted on the wiki that the only difference between batch
and per-package is the argument type.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/5d2c37d4/attachment-0001.asc>

From phil at bolthole.com  Fri Jul  3 00:04:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 15:04:45 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246563839-sup-2157@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
Message-ID: <20090702220445.GI94553@bolthole.com>

On Thu, Jul 02, 2009 at 04:02:56PM -0400, Ben Walton wrote:
> > http://wiki.opencsw.org/package-hooks
> 
> ...Ok, the wiki now contains a _complete_, _explicit_ list of valid
> hooks and the corresponding directories.

thank you.

more comments:

Dont depend on files that are living raw in /var/tmp.
It's a security hole.
I dont think you should be using files to pass around arguments at all.
particularly something as trivial as "a list of affected packages".
I think you shoudl have the list of packages, just be passed as the
argument list.

Additional comment:
you need to explicitly define behaviour of install vs update.
that is to say, you need to specify whether update hooks will be called
IN ADDITION TO, or INSTEAD OF, install and/or remove hooks.
hay waitaminute, you dont have remove hooks.

well actually you mention them in one section, but not in another section.
Missing from the "Proposed Per-package Hooks" section


From bwalton at opencsw.org  Fri Jul  3 02:43:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 20:43:36 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702220445.GI94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
Message-ID: <1246577961-sup-2700@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 18:04:45 -0400 2009:

> Dont depend on files that are living raw in /var/tmp.
> It's a security hole.

Yes, I thought of this on the train ride home.

> I dont think you should be using files to pass around arguments at
> all.  particularly something as trivial as "a list of affected
> packages".  I think you shoudl have the list of packages, just be
> passed as the argument list.

Ok.  I was thinking along the lines of overflowing ARG_MAX, but upon
inspection Solaris (32/64 bit -> 1048320/2096640) uses a huge value
compared to Linux (131072).  This is no longer a concern for me.

I do wonder though whether passing all packages as arguments is the
way to go.  If it was decided to add more information later (version
numbers or something), using the argument list would be more limiting
than a file-based approach.  Would passing lines on stdin be a better
method?

The spec could define: each line of stdin should be word split on
whitespace (to match the catalog delimitter).  The first word is the
package being acted on.  Hook scripts should ignore anything beyond
the first word.

This would be more flexible in terms of future expansion.  The same
could be done with command line arguments, but I think stdin works
better for this.

[No changes to the doc have been made in this area yet.]

> Additional comment:
> you need to explicitly define behaviour of install vs update.
> that is to say, you need to specify whether update hooks will be called
> IN ADDITION TO, or INSTEAD OF, install and/or remove hooks.

<snip>
Proposed Per-package Hooks:

install

This hook should bracket any action that will involve a pkgadd without
a preceeding pkgrm.

update

This hook should bracket any action that will involve a pkgrm followed
immediately by a pkgadd for the same package.
</snip>

That's the existing text (with s/update/upgrade/).  Not clear enough?
I understand what you're asking, but in my mind those definitions are
clear.  If I'm adding a package that doesn't exist (pkgadd without a
preceeding pkgrm), the install hook gets called.  If I'm adding a
package and I already have a current version (meaning I do pkgrm then
pkgadd), I do update (which I've renamed upgrade, since I feel that's
a better term than update).

There is no distinction between batchinstall or batchupgrade if that's
what you were asking.  We could note the difference by adding a
batchupgrade hook if you think that's worthwhile.  I'm not sure it is,
but it wouldn't necessarily hurt either.

> hay waitaminute, you dont have remove hooks.

> well actually you mention them in one section, but not in another section.
> Missing from the "Proposed Per-package Hooks" section

See purge (per-package) and batchremove (batch).  I used purge since
that's the term Dago proposed in his mail the other day.  Now that
remove has become batchremove, purge could be remove...Actually, I've
now made this change, since I think it is better language.  [I had
also incorrectly listed pre/post remove instead of pre/post purge in
one section, so that may be part of this problem/comment.]

Getting closer...

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/2c23d734/attachment.asc>

From phil at bolthole.com  Fri Jul  3 03:03:47 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 18:03:47 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246577961-sup-2700@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
Message-ID: <20090703010347.GK94553@bolthole.com>

On Thu, Jul 02, 2009 at 08:43:36PM -0400, Ben Walton wrote:
[http://wiki.opencsw.org/package-hooks]
> <snip>
> Proposed Per-package Hooks:
> 
> install
> ...

> That's the existing text (with s/update/upgrade/).  Not clear enough?

no it's not clear enough :) you need to be explicit, and state
whether or not a "upgrade" hook overlaps install/remove hooks, or whether
it gets called instead of them.
You also have to explicitly define the order of calling, if they both
get run

[hmm actually, you sort of say this in your examples. but you make
 some mistakes in your examples :) for example, you mix stating
  that something is being "installed" vs being "upgraded", in the same
   sample paragraph]

It's best to state explicitly in the specification, the 'correct'
behaviour.
Examples are suposed to explain the specification, not be instead of
the specification.




> There is no distinction between batchinstall or batchupgrade if that's
> what you were asking.  We could note the difference by adding a
> batchupgrade hook if you think that's worthwhile.  I'm not sure it is,
> but it wouldn't necessarily hurt either.

If you think it's neccessary to have a per-package upgrade, i think you
should provide for a batch upgrade, just for completeness. even if you cant
think of a use for it now, someone later might, if you provide a hook for
it.

> See purge (per-package) and batchremove (batch).  I used purge since
> that's the term Dago proposed in his mail the other day.

Please note: "purge" is DIFFERENT from "remove".

It has a long-understood meaning in the debian world, and possibly others.

"remove" means "run pkgrm".
"purge" should mean "run pkgrm, and REMOVE ANY LEFTOVERS"
  (in other words, remove any capability/saved state for a delayed upgrade via
    pkgadd later)


From schwindt at dfki.uni-kl.de  Fri Jul  3 10:40:37 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 03 Jul 2009 10:40:37 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: Your message of "Thu, 02 Jul 2009 12:24:49 +0200."
	<4A4C8AF1.6030508@wbonnet.net> 
Message-ID: <200907030840.n638eboe019606@dfki.uni-kl.de>

> Hi Nicolai
> 
> Does the following file exist ?
> 
> /opt/csw/etc/pango/pango.modules

Not existsing
 
> if not please run
> 
> /opt/csw/bin/pango-querymodules > /opt/csw/etc/pango/pango.modules

Did that - file was created successfull - did not change anything.
 
> This issue is under fixing...

The problem is : I am running on an U24 which is 64-bit this triggers
pango-querymodules to generate a list of the 64-bit modules.
Starting xfce which is only available as 32-bit loads libpango in 32-bit.
This one falls flat on the face.


It seems there is the need for separate pangorc files containing the correct
definition of ModulesFiles and AliasFiles.

This implies each and every other package that uses pango must have both
32 and 64 bit versions. The other solution was having a wrapper for each
executable setting the correct environment.

I did not look into the source of pango for a while, is there any chance
of patching pango to hardcode PANGO_RC_FILE=/opt/csw/etc/pangorc{32,64} into
it ?

BTW: afaik pixbuf-loader is the same pain in the arse
 


Nicolai



From dam at opencsw.org  Fri Jul  3 13:42:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 3 Jul 2009 13:42:45 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: <200907030840.n638eboe019606@dfki.uni-kl.de>
References: <200907030840.n638eboe019606@dfki.uni-kl.de>
Message-ID: <FAF013CB-5FCF-4559-9891-6025FFF564A2@opencsw.org>

Hi Nicolai,

Am 03.07.2009 um 10:40 schrieb Nicolai Schwindt:
>> if not please run
>>
>> /opt/csw/bin/pango-querymodules > /opt/csw/etc/pango/pango.modules
>
> Did that - file was created successfull - did not change anything.
>
>> This issue is under fixing...
>
> The problem is : I am running on an U24 which is 64-bit this triggers
> pango-querymodules to generate a list of the 64-bit modules.
> Starting xfce which is only available as 32-bit loads libpango in 32- 
> bit.
> This one falls flat on the face.

Ok, there is obiously an error in the package creation:
   /opt/csw/bin/pango-querymodules
is linked to isaexec and chooses the best ISA. This is wrong for this
kind of use. The isaexec should be removed and there should be two
binaries:
   /opt/csw/bin/pango-querymodules for 32 bit
   /opt/csw/bin/(sparcv9|amd64)/pango-querymodules for 64 bit
You can try this right now with
   rm /opt/csw/bin/pango-querymodules
   ln -s /opt/csw/bin/sparcv8/pango-querymodules /opt/csw/bin/pango- 
querymodules
Please tell me if that worked.

> It seems there is the need for separate pangorc files containing the  
> correct
> definition of ModulesFiles and AliasFiles.

We may need /opt/csw/etc/64/pango for this

> This implies each and every other package that uses pango must have  
> both
> 32 and 64 bit versions.

Why is that? A 32 bit app needs to pull in 32 bit libs and 64 bit apps
pull in 64 bit libs, but why should both ISAs be needed?

> The other solution was having a wrapper for each
> executable setting the correct environment.
>
> I did not look into the source of pango for a while, is there any  
> chance
> of patching pango to hardcode PANGO_RC_FILE=/opt/csw/etc/ 
> pangorc{32,64} into
> it ?

Is this really necessary on a per-app basis? Can't this be patched
into pango itself?

> BTW: afaik pixbuf-loader is the same pain in the arse

William, any advice on this?


Best regards

   -- Dago

From bwalton at opencsw.org  Fri Jul  3 15:15:04 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 03 Jul 2009 09:15:04 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090703010347.GK94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
	<20090703010347.GK94553@bolthole.com>
Message-ID: <1246626185-sup-8972@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 21:03:47 -0400 2009:
> On Thu, Jul 02, 2009 at 08:43:36PM -0400, Ben Walton wrote:
> [http://wiki.opencsw.org/package-hooks]

> no it's not clear enough :) you need to be explicit, and state
> whether or not a "upgrade" hook overlaps install/remove hooks, or whether
> it gets called instead of them.
> You also have to explicitly define the order of calling, if they both
> get run

> [hmm actually, you sort of say this in your examples. but you make
>  some mistakes in your examples :) for example, you mix stating
>   that something is being "installed" vs being "upgraded", in the same
>    sample paragraph]

I'll give you the point, but the error is only in perspective.  If a
user requests CSWfoo to be installed when it already exists, the tool
turns it into an upgrade action (layered semantics).  My wording was
from the point of view of the user request.  I've corrected things to
be from the point of view of the tool, since that's more appropriate
to the document.

> > There is no distinction between batchinstall or batchupgrade if that's
> > what you were asking.  We could note the difference by adding a
> > batchupgrade hook if you think that's worthwhile.  I'm not sure it is,
> > but it wouldn't necessarily hurt either.
> 
> If you think it's neccessary to have a per-package upgrade, i think you
> should provide for a batch upgrade, just for completeness. even if you cant
> think of a use for it now, someone later might, if you provide a hook for
> it.

Ok.  We'll add it.  Since this is (again) a synthetic addition, I'll
define batchupgrade such that if _any_ package is to be upgraded from
an older version, the batchupgrade hook will supersede batchadd (which
I've now named batchinstall to be consistent with per-package names).

> Please note: "purge" is DIFFERENT from "remove".
> It has a long-understood meaning in the debian world, and possibly
> others.

I'm aware of the differences.  Purge was used to avoid a name
collision.  That collision no longer exists and we'll forget about
purge.

> "remove" means "run pkgrm".
> "purge" should mean "run pkgrm, and REMOVE ANY LEFTOVERS" (in other
> words, remove any capability/saved state for a delayed upgrade via
> pkgadd later)

Personally, I think even upgrade is a bit of a stretch, since we're
now layering semantics on top of pkgadd/pkgrm.  It's low hanging fruit
though, since the information is readily available in the tools.
While we have the sampleconf classes that can accomplish purge-ish
behaviour, I don't think it would be good to extend too far beyond the
boundaries of pkgadd/pkgrm.

It's also quite difficult for pkg-get/pkgutil to detect the difference
between remove and purge (as handled by the classes), if we wanted to
synthesize it, since it would have to start snooping around to even
get that info.  Unless someone really thinks purge is important, I
think we should walk away from 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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/4ff27dda/attachment-0001.asc>

From schwindt at dfki.uni-kl.de  Fri Jul  3 15:04:33 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 03 Jul 2009 15:04:33 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: Your message of "Fri, 03 Jul 2009 13:42:45 +0200."
	<FAF013CB-5FCF-4559-9891-6025FFF564A2@opencsw.org> 
Message-ID: <200907031304.n63D4Y7k025114@dfki.uni-kl.de>

[...]
> You can try this right now with
>    rm /opt/csw/bin/pango-querymodules
>    ln -s /opt/csw/bin/sparcv8/pango-querymodules /opt/csw/bin/pango- 
> querymodules
> Please tell me if that worked.

I basically did that and it worked.


[...]
> Is this really necessary on a per-app basis? Can't this be patched
> into pango itself?

If pango was done better, there was an option for i.e. 
PANGO_MODULES_PATH_{32,64}
which could point to /opt/csw/lib or /opt/csw/lib/64. There it could read
pango.modules only containing filename like pango-indic-lang.so not the 
absolute
path. But this would leed to the necessity of pango knowing if it is 32/64 bit.


The main problem  is pango is needing its pango.modules and does not honor in 
any kind
the architecture. pango32 and pango64 will basically open the same file.
In my case it meens, to be able to use xcfe which is 32-bit I generated 
pango.modules
by hand to contain only the 32-bit modules. Anything which now tries to use 
pango64 will
get into trouble.


This flaw gets other packagers in trouble also.


Nicolai



From phil at bolthole.com  Fri Jul  3 16:06:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 3 Jul 2009 07:06:32 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246626185-sup-8972@ntdws12.chass.utoronto.ca>
References: <20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
	<20090703010347.GK94553@bolthole.com>
	<1246626185-sup-8972@ntdws12.chass.utoronto.ca>
Message-ID: <20090703140631.GB64022@bolthole.com>

On Fri, Jul 03, 2009 at 09:15:04AM -0400, Ben Walton wrote:
> ...
> It's also quite difficult for pkg-get/pkgutil to detect the difference
> between remove and purge (as handled by the classes), if we wanted to
> synthesize it, since it would have to start snooping around to even
> get that info.  Unless someone really thinks purge is important, I
> think we should walk away from it.

I have been meaning to implement explicit purge in pkg-get, for a long
time.

Its the reason why the cswpreserveconf class exists, for example:
so that there will be a deterministic way of accomplishing purge.
That class makes it possible in theory to run "pkg-get purge", even if
  the original package is no longer installed! 
Too bad I havent written it yet :)


From bwalton at opencsw.org  Fri Jul  3 16:18:09 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 03 Jul 2009 10:18:09 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090703140631.GB64022@bolthole.com>
References: <20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
	<20090703010347.GK94553@bolthole.com>
	<1246626185-sup-8972@ntdws12.chass.utoronto.ca>
	<20090703140631.GB64022@bolthole.com>
Message-ID: <1246630414-sup-2279@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 03 10:06:32 -0400 2009:

> I have been meaning to implement explicit purge in pkg-get, for a long
> time.

Well then, we can either define a purge hook up front or add it later
when the ability exists to actually use it.  Adding hooks later is
much easier than removing them after they're defined (and used).

My preference would be to add it later, but it's not something I feel
strongly about either way.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/164ac626/attachment.asc>

From bwalton at opencsw.org  Sat Jul  4 02:46:53 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 03 Jul 2009 20:46:53 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702220445.GI94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
Message-ID: <1246667705-sup-8854@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 18:04:45 -0400 2009:
> I dont think you should be using files to pass around arguments at all.
> particularly something as trivial as "a list of affected packages".
> I think you shoudl have the list of packages, just be passed as the
> argument list.

Ok, then.  How about the following for input to the scripts:

Each affected package (for batch hooks, the entire list, for
per-package, a single package) is passed as an argument of the form
$svr4name-$version

where $svr4name corresponds to the third field of the catalog and the
$version corresponds to the second field of the catalog.

A hypothetical call to a prebatchupgrade hook might look like:

/etc/opt/csw/pkg-hooks/prebatchupgrade.d/01-CSWfoo-logupgrades CSWbar-1.2,REV=2009.06.05 CSWbaz-1.3.4,REV=2008.12.01

A hyptothetical call to a postremove hook might look like:

/etc/opt/csw/pkg-hooks/postremove.d/02-CSWbar-mailaboutremoval CSWbar-1.2,REV=2009.06.05

This should be all of the useful data we can pass to the hooks without
doing backflips.

If a specific hook wants to be able to differentiate between which
packages are being upgraded and which are installed, it can use
multiple hooks and correlate data between the calls or simply monitor
the per-package hooks instead of the batch hooks.

If a specific hook wants to check currently installed version, it can
do that in a pre* hook, possibly saving state somewhere.

For saved state, I think we should define a directory (maybe the
previously proposed /var/opt/csw/pkg-hooks/) as the place where hooks
are to store any temporary files across different hook
scripts...attempting to keep scripts away from /var/tmp.  Obviously we
won't really be able to force this, but if it's documented, those
reading the documentation to add hooks of their own will hopefully see
and use it.

Reasonable?

Also, when we finish nailing down the details, will the required
directories be added to CSWcommon?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/fceaefb7/attachment.asc>

From rupert at opencsw.org  Sat Jul  4 12:39:45 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 12:39:45 +0200
Subject: [csw-maintainers] include asvn script into cswsvn,
	at least client: how to do this?
Message-ID: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>

hi,

how would one include asvn
(http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn ) best
into the cswsvn package?

rupert.



---------- Forwarded message ----------
From: Thurner Rupert (KIRF 65) <rupert.thurner at credit-suisse.com>
Date: Wed, Jul 1, 2009 at 18:07
Subject: include asvn into cswsvn pls
To: rupert THURNER <rupert at opencsw.org>




______________________________________________
From: ? Lawie Duncan (CS)
Sent:?? Wednesday, July 01, 2009 3:17 PM
To:???? Thurner Rupert (KIRF 65); Kannenberg Ralf (KIRG 1); Molini
Robert R. (KIRG 1)
Subject:??????? SVN and file permissions

Rupert,

it took us a good few drawings, but I think our discussion this
afternoon was very useful.

The "asvn" script I referred to is available at
http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn

It looks to do exactly what we need, in terms of maintaining file
permissions, but I haven't actually been able to test it because it
relies on a -printf option to "find"

PRINTDETAILS="-printf \"file='%p' mode=%m user=%u(%U) group=%g(%G)\n\""
eval "find $PCWD $SKIPSVN -o \( \( -type d ! -name .svn \) -o -type f
\) $PRINTDETAILS"

I understood that we should be able to install? GNU find alongside
svn, which would allow us to test this script.? Would it be best to
get it from http://www.gnu.org/software/findutils/ ?

Assuming that this does what we need, is there a formal mechanism for
requesting that asvn and GNU find are made a part of the standard SVN
product?

Thanks,
Duncan.

Duncan Lawie
Database Component Engineering - Oracle
+44 (0)20 788 3 9229

"If you will take my advice, you will think little of Socrates and a
great deal more for truth." -- Socrates.

From bwalton at opencsw.org  Sat Jul  4 14:35:57 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sat, 04 Jul 2009 08:35:57 -0400
Subject: [csw-maintainers] include asvn script into cswsvn,
	at least client: how to do this?
In-Reply-To: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
References: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
Message-ID: <1246710529-sup-4564@ntdws12.chass.utoronto.ca>

Excerpts from rupert THURNER's message of Sat Jul 04 06:39:45 -0400 2009:
> how would one include asvn
> (http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn ) best
> into the cswsvn package?

You'll need file a bug/feature request, or maybe just ping Mike
directly.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090704/52e99f61/attachment.asc>

From mwatters at opencsw.org  Sat Jul  4 15:07:58 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 04 Jul 2009 08:07:58 -0500
Subject: [csw-maintainers] include asvn script into cswsvn,
 at least client: how to do this?
In-Reply-To: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
References: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
Message-ID: <4A4F542E.4020306@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

rupert THURNER wrote:

> how would one include asvn
> (http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn ) best
> into the cswsvn package?

file a bug 'feature request' so I can track the request.  I will add it to the
next 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 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkpPVC0ACgkQLrhmsXMSLxcRRACdFreriUDFQfJekcl9NRJF0wse
BQQAn2SdnZicVuK1NyEp3DLbhGGeVvnq
=FVU5
-----END PGP SIGNATURE-----

From skayser at opencsw.org  Sat Jul  4 18:29:04 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 04 Jul 2009 18:29:04 +0200
Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka
 "Which packages are in GAR")
In-Reply-To: <4A493506.9050204@wbonnet.net>
References: <4A2403D0.4090109@opencsw.org>
	<50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de>
	<4A493506.9050204@wbonnet.net>
Message-ID: <4A4F8350.9090100@opencsw.org>

William Bonnet wrote:
> Hi Sebastian
>> Sebastian Kayser wrote:
>>   
>>> a couple of RFEs for the "GAR package status" [1] page. Can we get:
>>>
>>> * an "last updated" field at the top of the page
>>> * the numbers / stats right at the top of the page (instead of further
>>>   down)
>>> * a contact address for the maintainer of the page on the page (for
>>>   queries like this one and to get packages that are in GAR but not
>>>   listed as such fixed)
>>>
>>> Ah, and some sort of monthly historical data would be cool also. So that
>>> we can see some progress.
>>>
>>> Sebastian
>>>
>>> [1] http://www.opencsw.org/package-gar-status.html
>>>     
>> Anyone taking care of the GAR status page?
>>   
> I'm currently working on this page and uwatch. Rewriting it in Perl and 
> improving things.
> 
> I will modify the page by the end of the week, and uwatch next week. 
> uwatch have several bugs to fix (problem with garname used to retrieve 
> package instead of catalog name). I would like to merge some information 
> from uwatch to this page.

Just in case you don't have it on your radar, yet: The recent
sourceforge page makeover introduced an easy URL for project files.

  http://sourceforge.net/projects/<project_name>/files/

Might help to simplify uwatch and check-upstream handling.

Sebastian

From rupert at opencsw.org  Sat Jul  4 18:40:14 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 18:40:14 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
In-Reply-To: <4A4C6903.2040906@opencsw.org>
References: <4A4C6903.2040906@opencsw.org>
Message-ID: <6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>

i compiled it with gar v2, upgraded the version number, and it runs
through without any noticable prolems. i also checked this in. did not
yet try to fix the tests though ...

rupert.

2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
> If someone feel like getting this mostly python package out, feel free.
>
> I almost got 1.2 working but there where too many test cases that failed.
> Most of the failures was because of missing gnu'isms in the tools the test
> used from what I could tell. Would be nice to patch up the test suite to be
> able to control the path to the tools it uses.
>
> --
> Trygve
>
>
> Hello dear mercurial maintainer,
>
> The upstream notification job has detected the availability of new files for
> mercurial.
>
> The following upstream file(s):
> ? ?mercurial-1.3.tar.gz
>
> is/are available at the following url(s):
> ? ?http://selenic.com/mercurial/release/
>
> Please consider updating your package.
>
> --
> Kindest regards
> upstream notification job
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>

From trygvis at opencsw.org  Sat Jul  4 18:59:42 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sat, 04 Jul 2009 18:59:42 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream
	update	notification]
In-Reply-To: <6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
Message-ID: <4A4F8A7E.70704@opencsw.org>

rupert THURNER wrote:
> i compiled it with gar v2, upgraded the version number, and it runs
> through without any noticable prolems. i also checked this in. did not
> yet try to fix the tests though ...

Do you have a URl to the version in testing?

--
Trygve

> rupert.
> 
> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>> If someone feel like getting this mostly python package out, feel free.
>>
>> I almost got 1.2 working but there where too many test cases that failed.
>> Most of the failures was because of missing gnu'isms in the tools the test
>> used from what I could tell. Would be nice to patch up the test suite to be
>> able to control the path to the tools it uses.
>>
>> --
>> Trygve
>>
>>
>> Hello dear mercurial maintainer,
>>
>> The upstream notification job has detected the availability of new files for
>> mercurial.
>>
>> The following upstream file(s):
>>    mercurial-1.3.tar.gz
>>
>> is/are available at the following url(s):
>>    http://selenic.com/mercurial/release/
>>
>> Please consider updating your package.
>>
>> --
>> Kindest regards
>> upstream notification job
>>
>> _______________________________________________
>> 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 rupert at opencsw.org  Sat Jul  4 19:01:03 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 19:01:03 +0200
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
Message-ID: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>

hi,

when doing the mercurial tests, there is a warning.

gpg: Warning: using insecure memory!
  --> http://www.gnupg.org/faq.html#q6.1

the faq says that it should be installed with setuid root. is this
deliberately different on opencsw?

rupert.

From skayser at opencsw.org  Sat Jul  4 19:12:05 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 04 Jul 2009 19:12:05 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
 instead of PROTOTYPE_FILTER
In-Reply-To: <3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
Message-ID: <4A4F8D65.9080203@opencsw.org>

Dagobert Michelsen wrote:
> Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
>> while working on postfix 2.6 i have noticed again that it is not quite
>> straight-forward to assign files to classes. Instead of exposing the  
>> inner
>> workings of the prototype stuff to the Makefile, could we maybe get  
>> some
>> variables to assign files to classes (functions)?
> 
> For CSWcswclassutils this is already in there :-) Just define
> 
> SAMPLECONF
> PRESERVECONF
> INITSMF
> 
> and everything else will be taken cared of.

Sweet!!! :D Can we add something for USERGROUP also?

Index: /home/skayser/mgar/gar/v2/gar.pkg.mk
===================================================================
--- /home/skayser/mgar/gar/v2/gar.pkg.mk        (revision 5361)
+++ /home/skayser/mgar/gar/v2/gar.pkg.mk        (working copy)
@@ -155,12 +155,14 @@
 SPKG_CLASSES := $(SPKG_CLASSES) $(if $(SAMPLECONF),$(if $(filter cswsampleconf,$(SPKG_CLASSES)),,cswsampleconf))
 SPKG_CLASSES := $(SPKG_CLASSES) $(if $(PRESERVECONF),$(if $(filter cswpreserveconf,$(SPKG_CLASSES)),,cswpreserveconf))
 SPKG_CLASSES := $(SPKG_CLASSES) $(if $(INITSMF),$(if $(filter cswinitsmf,$(SPKG_CLASSES)),,cswinitsmf))
+SPKG_CLASSES := $(SPKG_CLASSES) $(if $(USERGROUP),$(if $(filter cswusergroup,$(SPKG_CLASSES)),,cswusergroup))
 # - set class for all config files
 ifneq ($(SAMPLECONF)$(PRESERVECONF)$(INITSMF),)
 _CSWCLASS_FILTER = | perl -ane '\
                $(foreach FILE,$(SAMPLECONF),$$F[1] = "cswsampleconf" if ( $$F[2] =~ m(^$(FILE)$$) );)\
                $(foreach FILE,$(PRESERVECONF),$$F[1] = "cswpreserveconf" if( $$F[2] =~ m(^$(FILE)$$) );)\
                $(foreach FILE,$(INITSMF),$$F[1] = "cswinitsmf" if( $$F[2] =~ m(^$(FILE)$$) );)\
+               $(foreach FILE,$(USERGROUP),$$F[1] = "cswusergroup" if( $$F[2] =~ m(^$(FILE)$$) );)\
                print join(" ", at F),"\n";'
 _EXTRA_GAR_PKGS += CSWcswclassutils
 endif

Sebastian

From skayser at opencsw.org  Sat Jul  4 19:21:52 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 04 Jul 2009 19:21:52 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream
	update	notification]
In-Reply-To: <6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
Message-ID: <4A4F8FB0.8040105@opencsw.org>

rupert THURNER wrote:
> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>> If someone feel like getting this mostly python package out, feel free.
>>
>> I almost got 1.2 working but there where too many test cases that failed.
>> Most of the failures was because of missing gnu'isms in the tools the test
>> used from what I could tell. Would be nice to patch up the test suite to be
>> able to control the path to the tools it uses.
>
> i compiled it with gar v2, upgraded the version number, and it runs
> through without any noticable prolems. i also checked this in. did not
> yet try to fix the tests though ...

If tests fail because they rely on GNU features of various tools, you
could try to tweak the PATH used for the tests.

  TEST_ARGS = "PATH=/opt/csw/gnu:$$PATH"

Helped here in a few cases, don't know whether it does for mercurial.
TEST_ENV (instead of TEST_ARGS) should also do the job now that i think
about it.

Sebastian

>> Hello dear mercurial maintainer,
>>
>> The upstream notification job has detected the availability of new files for
>> mercurial.
>>
>> The following upstream file(s):
>>    mercurial-1.3.tar.gz
>>
>> is/are available at the following url(s):
>>    http://selenic.com/mercurial/release/
>>
>> Please consider updating your package.
>>
>> --
>> Kindest regards
>> upstream notification job

From rupert at opencsw.org  Sat Jul  4 19:37:35 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 19:37:35 +0200
Subject: [csw-maintainers] v2-pbuild - test with subversion
Message-ID: <6af4270907041037s4861a955y30e5b94c1f1197c@mail.gmail.com>

hi,

i tried to switch the svn package to v2-pbuild, and noticed the following:


1. parallel start
the file gets downloaded wrongly (parallel wget)


2. start without clean, after the other build is at "unzipping"
the build on build8s is faster, and finishes. the build on build8x
finishes later with:

/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Client
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.so
: Permission denied
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.so
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.bs
: Permission denied
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.bs
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Core
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Core/_Core.so
: Permission denied
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Core/_Core.so
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Core/.packlist
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Core/_Core.bs
: Permission denied


rupert

From rupert at opencsw.org  Sat Jul  4 20:03:30 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 20:03:30 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
In-Reply-To: <4A4F8A7E.70704@opencsw.org>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
	<4A4F8A7E.70704@opencsw.org>
Message-ID: <6af4270907041103s69e51601rb54722e60498f886@mail.gmail.com>

the test result is: # Ran 345 tests, 30 skipped, 8 failed.

Failed test-command-template: output changed
Failed test-convert-cvs-branch: output changed and returned error code 1
Failed test-convert-git: output changed
Failed test-gpg: output changed
Failed test-hook: output changed
Failed test-repair-strip: output changed
Failed test-template-engine: output changed and returned error code 255
Failed test-transplant: output changed


and some tests are skipped which probably should not:

Skipped test-convert-hg-svn: missing feature: subversion python bindings
Skipped test-convert-svn-branches: missing feature: subversion python bindings
Skipped test-convert-svn-encoding: missing feature: subversion python bindings
Skipped test-convert-svn-move: missing feature: subversion python bindings
Skipped test-convert-svn-sink: missing feature: subversion python bindings
Skipped test-convert-svn-source: missing feature: subversion python bindings
Skipped test-convert-svn-startrev: missing feature: subversion python bindings
Skipped test-convert-svn-tags: missing feature: subversion python bindings

is there a possibility to make this working on the build servers?

for some failures i have no idea where the failure is coming from,
will ask the mercurial mailing list. i noticed that also for fedora
the test cases failed sometimes and they were disabled for release
building.


rupert.


2009/7/4 Trygve Laugst?l <trygvis at opencsw.org>:
> rupert THURNER wrote:
>>
>> i compiled it with gar v2, upgraded the version number, and it runs
>> through without any noticable prolems. i also checked this in. did not
>> yet try to fix the tests though ...
>
> Do you have a URl to the version in testing?
>
> --
> Trygve
>
>> rupert.
>>
>> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>>>
>>> If someone feel like getting this mostly python package out, feel free.
>>>
>>> I almost got 1.2 working but there where too many test cases that failed.
>>> Most of the failures was because of missing gnu'isms in the tools the
>>> test
>>> used from what I could tell. Would be nice to patch up the test suite to
>>> be
>>> able to control the path to the tools it uses.
>>>
>>> --
>>> Trygve
>>>
>>>
>>> Hello dear mercurial maintainer,
>>>
>>> The upstream notification job has detected the availability of new files
>>> for
>>> mercurial.
>>>
>>> The following upstream file(s):
>>> ? mercurial-1.3.tar.gz
>>>
>>> is/are available at the following url(s):
>>> ? http://selenic.com/mercurial/release/
>>>
>>> Please consider updating your package.
>>>
>>> --
>>> Kindest regards
>>> upstream notification job
>>>
>>> _______________________________________________
>>> 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
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

From rupert at opencsw.org  Sat Jul  4 20:36:52 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 20:36:52 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
In-Reply-To: <6af4270907041103s69e51601rb54722e60498f886@mail.gmail.com>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
	<4A4F8A7E.70704@opencsw.org>
	<6af4270907041103s69e51601rb54722e60498f886@mail.gmail.com>
Message-ID: <6af4270907041136i1bbba3b8kc97a9c85414197d7@mail.gmail.com>

i put the packages onto http://mirror.opencsw.org/testing.html, and
made them writeable for others as well.

rupert.

On Sat, Jul 4, 2009 at 20:03, rupert THURNER<rupert at opencsw.org> wrote:
> the test result is: # Ran 345 tests, 30 skipped, 8 failed.
>
> Failed test-command-template: output changed
> Failed test-convert-cvs-branch: output changed and returned error code 1
> Failed test-convert-git: output changed
> Failed test-gpg: output changed
> Failed test-hook: output changed
> Failed test-repair-strip: output changed
> Failed test-template-engine: output changed and returned error code 255
> Failed test-transplant: output changed
>
>
> and some tests are skipped which probably should not:
>
> Skipped test-convert-hg-svn: missing feature: subversion python bindings
> Skipped test-convert-svn-branches: missing feature: subversion python bindings
> Skipped test-convert-svn-encoding: missing feature: subversion python bindings
> Skipped test-convert-svn-move: missing feature: subversion python bindings
> Skipped test-convert-svn-sink: missing feature: subversion python bindings
> Skipped test-convert-svn-source: missing feature: subversion python bindings
> Skipped test-convert-svn-startrev: missing feature: subversion python bindings
> Skipped test-convert-svn-tags: missing feature: subversion python bindings
>
> is there a possibility to make this working on the build servers?
>
> for some failures i have no idea where the failure is coming from,
> will ask the mercurial mailing list. i noticed that also for fedora
> the test cases failed sometimes and they were disabled for release
> building.
>
>
> rupert.
>
>
> 2009/7/4 Trygve Laugst?l <trygvis at opencsw.org>:
>> rupert THURNER wrote:
>>>
>>> i compiled it with gar v2, upgraded the version number, and it runs
>>> through without any noticable prolems. i also checked this in. did not
>>> yet try to fix the tests though ...
>>
>> Do you have a URl to the version in testing?
>>
>> --
>> Trygve
>>
>>> rupert.
>>>
>>> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>>>>
>>>> If someone feel like getting this mostly python package out, feel free.
>>>>
>>>> I almost got 1.2 working but there where too many test cases that failed.
>>>> Most of the failures was because of missing gnu'isms in the tools the
>>>> test
>>>> used from what I could tell. Would be nice to patch up the test suite to
>>>> be
>>>> able to control the path to the tools it uses.
>>>>
>>>> --
>>>> Trygve
>>>>
>>>>
>>>> Hello dear mercurial maintainer,
>>>>
>>>> The upstream notification job has detected the availability of new files
>>>> for
>>>> mercurial.
>>>>
>>>> The following upstream file(s):
>>>> ? mercurial-1.3.tar.gz
>>>>
>>>> is/are available at the following url(s):
>>>> ? http://selenic.com/mercurial/release/
>>>>
>>>> Please consider updating your package.
>>>>
>>>> --
>>>> Kindest regards
>>>> upstream notification job
>>>>
>>>> _______________________________________________
>>>> 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
>>
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
>

From rupert at opencsw.org  Sat Jul  4 20:40:26 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 20:40:26 +0200
Subject: [csw-maintainers] libserf-0.3.0 now in testing
Message-ID: <6af4270907041140u74538208md46c69f8b9eaa13d@mail.gmail.com>

many thanks! libserf-0.3.0 is now in testing.

rupert.

On Mon, Jun 29, 2009 at 20:03, Mike Watters<mwatters at opencsw.org> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> rupert THURNER wrote:
>
>>
>> how could one avoid that?
>>
>
> the main issue is the Makefile generated by configre is using "libtool" from
> /opt/csw/apache2/share/build/libtool which does "NOT" get modified by my strip.
>
> Here is a patch to the Makefile currently in svn that will allow you to build
> and test successfully libserf
>
>
> if you cut and paste, please take care to watch for wrapping I have Thunderbird
> configured to wrap at 100 characters.
>
> $ svn diff Makefile
> Index: Makefile
> ===================================================================
> - --- Makefile ? ?(revision 5414)
> +++ Makefile ? ?(working copy)
> @@ -23,13 +23,19 @@
> ?# UPSTREAM_MASTER_SITES =
>
> ?DEPENDS = server/apache2rt
> - -STRIP_LIBTOOL = 1
>
> ?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=$(prefix)/apache2/bin/apu-config
> +CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config
CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config
>
> ?TEST_TARGET = check
> ?include gar/category.mk
> +
> +EXTRA_LDFLAGS = -R/opt/csw/bdb44/lib
> +APR_LIBTOOL = $(shell /opt/csw/apache2/bin/apr-config --apr-libtool)
> +post-configure-modulated:
> + ? ? ? gcp $(APR_LIBTOOL) $(WORKSRC)
> + ? ? ? perl -i -pne 's|$(APR_LIBTOOL)|$(abspath $(WORKSRC)/libtool)|' \
> + ? ? ? ? ? ? ? $(WORKSRC)/Makefile
> + ? ? ? $(GARBIN)/fixlibtool $(WORKSRC)
> + ? ? ? @$(MAKECOOKIE)
>
>
>
> - --
>
> 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
>
> iEYEARECAAYFAkpJAdUACgkQLrhmsXMSLxeBXQCgmTxH6hTrxyY2TrT6eSMj+S4t
> c+4Anj0nU/L6fyoBYYX4VqFjMCGd4iRo
> =qeGL
> -----END PGP SIGNATURE-----
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>

From praba_tuty at yahoo.com  Fri Jul  3 11:47:19 2009
From: praba_tuty at yahoo.com (prabakaran)
Date: Fri, 3 Jul 2009 02:47:19 -0700 (PDT)
Subject: [csw-maintainers] HELP: testing for new package CSQL
Message-ID: <887751.48196.qm@web52903.mail.re2.yahoo.com>



Hi Team,
I am creating csw package for CSQL main memory database management system. This is my first package creation experience.. I have put the package under the below nfs directory

/home/testing/csql-2.5.1,REV=2009.07.02-SunOS5.10-sparc-CSW.pkg.gz

It is created on solaris 10 as it uses atomic compare and swap which is supported only in solaris 10.
Can somebody test this( as i do not have root access) and let me know your comments.

Thanks and Regards
Prabakaran




      


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/bd512ebe/attachment.html>

From bonivart at opencsw.org  Mon Jul  6 11:18:39 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Jul 2009 11:18:39 +0200
Subject: [csw-maintainers] HELP: testing for new package CSQL
In-Reply-To: <887751.48196.qm@web52903.mail.re2.yahoo.com>
References: <887751.48196.qm@web52903.mail.re2.yahoo.com>
Message-ID: <625385e30907060218s5c6c6bb5j539fd58003b06518@mail.gmail.com>

On Fri, Jul 3, 2009 at 11:47 AM, prabakaran<praba_tuty at yahoo.com> wrote:
> Hi Team,
> I am creating csw package for CSQL main memory database management system.
> This is my first package creation experience.. I have put the package under
> the below nfs directory
>
> /home/testing/csql-2.5.1,REV=2009.07.02-SunOS5.10-sparc-CSW.pkg.gz
>
> It is created on solaris 10 as it uses atomic compare and swap which is
> supported only in solaris 10.
> Can somebody test this( as i do not have root access) and let me know your
> comments.

I tried to install it on Solaris 10 x86 but I got a prompt for base
directory, that needs to be fixed.

"Enter path to package base directory [?,q]"

Also, don't display the license during installation. You don't seem to
be using GAR (why not?) so put this in the copyright file:

"Please see /opt/csw/share/doc/csql/license for license information"
and then copy the full license to that file.

-- 
/peter

From ja at opencsw.org  Mon Jul  6 14:25:54 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Mon, 06 Jul 2009 14:25:54 +0200
Subject: [csw-maintainers] Munin 1.3.4 in testing
Message-ID: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>

Hi,

I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW.  
The software consists of the packages munin_master and munin_node.

Feedback is welcome.

Juergen

-- 
Juergen Arndt

From skayser at opencsw.org  Mon Jul  6 17:34:51 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Mon, 06 Jul 2009 17:34:51 +0200
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
Message-ID: <4A52199B.5070808@opencsw.org>

Hi,

i am seeing some weird issues while trying to build libiconv on
build8s. The regular build with GAR works just fine, but when i try
to incorporate the optimization flags that Phil has used so far
(--fast --xnolibmopt), the ./configure on build8s just hangs at

  checking for stdbool.h that conforms to C99...

GAR build description is at [1]. build8x is fine, i can only see
this issue on build8s. The ptree looks something like

        25961 gmake build-isa-sparcv8
          26068 gmake MODULATION=isa-sparcv8 ISA=sparcv8 build-modulated
            26886 /bin/sh -c cd work/build-isa-sparcv8/libiconv-1.13 && prefix="/opt/csw" exec_pr
              26887 /bin/bash ./configure --prefix=/opt/csw --exec_prefix=/opt/csw --bindir=/opt/cs
                29070 /bin/bash ./configure --prefix=/opt/csw --exec_prefix=/opt/csw --bindir=/opt/cs
                  29071 /opt/studio/SOS11/SUNWspro/bin/cc -c -fast -xnolibmopt -xarch=v8 -I/opt/csw/inc
                    29072 /opt/studio/SOS11/SUNWspro/prod/bin/acomp -xintrinsic -xldscope=global -depend

Trussing the acomp shows a myriad of brk()s and the process hogs
memory like there is no tomorrow.

   PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
 29072 skayser   684M  670M cpu13   30    0   0:01.23 3.1% acomp/1

When i take the conftest.c that ./configure tries to compile [2] and
try to compile it manually it simply bails out like it does on
build8x [3] (cannot find stdbool.h), but it does not hang. Anyone
seen something like this before? Or any other idea what could be the
culprit here?

Sebastian

[1] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/libiconv/trunk/Makefile
[2] http://pastebin.com/m6010e35c
[3] http://pastebin.com/m18121654

P.S.: Building with these optimization flags on build8x makes the
libiconv test suite complete in about 1:35 - 1:40 minutes versus 1:55
minutes for the non-optimized version, so it seems worth making the
opimization flags work.

From phil at bolthole.com  Mon Jul  6 18:11:51 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 09:11:51 -0700
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
	./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A52199B.5070808@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
Message-ID: <20090706161151.GA36831@bolthole.com>

On Mon, Jul 06, 2009 at 05:34:51PM +0200, Sebastian Kayser wrote:
> Trussing the acomp shows a myriad of brk()s and the process hogs
> memory like there is no tomorrow.
> 
>    PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
>  29072 skayser   684M  670M cpu13   30    0   0:01.23 3.1% acomp/1

yes. this is known behaviour of cc.
(although I havent seen it with libiconv specifically)

certain obscure optimization cases, explode, so to speak, taking extreme
memory usage and disk space for it to "traverse the problem space", if you
will.

Your choices are:

a) wait for a long time, and it will eventually complete, if the machine
doesnt run out of resources

b) reduce the optimization flags to -xO2 or -xO3


From trygvis at opencsw.org  Mon Jul  6 18:22:45 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Mon, 06 Jul 2009 18:22:45 +0200
Subject: [csw-maintainers] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <4A5224D5.8030803@opencsw.org>

Juergen Arndt wrote:
> Hi,
> 
> I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW. 
> The software consists of the packages munin_master and munin_node.
> 
> Feedback is welcome.

I just tried your packages. The installation look fine and I made some 
small notes on the wiki[1]. However, when I tried to create a graph, all 
and any letters are rendered as a box. See the attached image. I assume 
that this really is an rrdtool issue more than a Munin issue.

[1]: http://wiki.opencsw.org/munin-package

--
Trygve

From trygvis at opencsw.org  Mon Jul  6 18:32:29 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Mon, 06 Jul 2009 18:32:29 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <4A52271D.4000106@opencsw.org>

Juergen Arndt wrote:
> Hi,
> 
> I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW. 
> The software consists of the packages munin_master and munin_node.

Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know to 
be as stable as 1.2.6?

I see that Fedora and Debian both package 1.2.5/.6:

http://dag.wieers.com/rpm/packages/munin/
http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all

--
Trygve

From ja at opencsw.org  Mon Jul  6 18:51:18 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Mon, 06 Jul 2009 18:51:18 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A52271D.4000106@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A52271D.4000106@opencsw.org>
Message-ID: <op.uwnnrrurjbmpvo@imac.lan>

On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l <trygvis at opencsw.org>  
wrote:

> Juergen Arndt wrote:
>> Hi,
>>  I've put Munin 1.3.4 into testing. It's the initial version for  
>> OpenCSW. The software consists of the packages munin_master and  
>> munin_node.
>
> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know to  
> be as stable as 1.2.6?
>
> I see that Fedora and Debian both package 1.2.5/.6:
>
> http://dag.wieers.com/rpm/packages/munin/
> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all

1.2.6 is the last official stable version, but it's already more than a  
year old and I was asked to package a more current version. If this does  
not comply with our policies, I will package 1.2.6 as well, no problem.

> I just tried your packages. The installation look fine and I made some  
> small notes on the wiki[1]. However, when I tried to create a graph, all  
> and any letters are rendered as a box. See the attached image. I assume  
> that this really is an rrdtool issue more than a Munin issue.
>  [1]: http://wiki.opencsw.org/munin-package

I cannot confirm that, on my installation the graphs look as expected.  
Maybe some dependencies are missing, I will try to dig into it.

I will update your wiki page in the next future with some more  
installation details.

Juergen

-- 
Juergen Arndt

From trygvis at opencsw.org  Mon Jul  6 19:04:57 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 06 Jul 2009 19:04:57 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnnrrurjbmpvo@imac.lan>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>
	<op.uwnnrrurjbmpvo@imac.lan>
Message-ID: <4A522EB9.9090400@opencsw.org>

Juergen Arndt wrote:
> On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l 
> <trygvis at opencsw.org> wrote:
> 
>> Juergen Arndt wrote:
>>> Hi,
>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>> OpenCSW. The software consists of the packages munin_master and 
>>> munin_node.
>>
>> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know 
>> to be as stable as 1.2.6?
>>
>> I see that Fedora and Debian both package 1.2.5/.6:
>>
>> http://dag.wieers.com/rpm/packages/munin/
>> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all 
>>
> 
> 1.2.6 is the last official stable version, but it's already more than a 
> year old and I was asked to package a more current version. If this does 
> not comply with our policies, I will package 1.2.6 as well, no problem.

No, I was just wondering really.

>> I just tried your packages. The installation look fine and I made some 
>> small notes on the wiki[1]. However, when I tried to create a graph, 
>> all and any letters are rendered as a box. See the attached image. I 
>> assume that this really is an rrdtool issue more than a Munin issue.
>>  [1]: http://wiki.opencsw.org/munin-package
> 
> I cannot confirm that, on my installation the graphs look as expected. 
> Maybe some dependencies are missing, I will try to dig into it.
> 
> I will update your wiki page in the next future with some more 
> installation details.

I think it would be nice if the package included two complete apache 
configuration files (the basic and the one with CGI) so that the user 
can just Include them from httpd.conf. From what I can tell it is not 
likely that a user will need to adjust the configuration at all, at 
least not to get started.

As for my rrd issue I just did a test from the rrd tutorial and the 
graphs look ok there.

Which encoding does your system use? I'm using en_US.UTF-8.

--
Trygve


From phil at bolthole.com  Mon Jul  6 19:07:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 10:07:52 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnnrrurjbmpvo@imac.lan>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A52271D.4000106@opencsw.org> <op.uwnnrrurjbmpvo@imac.lan>
Message-ID: <20090706170752.GB60655@bolthole.com>

On Mon, Jul 06, 2009 at 06:51:18PM +0200, Juergen Arndt wrote:
> 1.2.6 is the last official stable version, but it's already more than a  
> year old and I was asked to package a more current version. If this does  
> not comply with our policies, I will package 1.2.6 as well, no problem.

well, the "policies" have to be flexible, due to the stupidity of some
software "publisher"s lack of standards and consistency.
The usual questions to be asked are:

1. how long has [specific version proposed] been out in the wild
2. how likely is it that an official new 'stable' version will be released
   soon?

 (and 2a: is the'stable' version so ancient as to be virtually useless)

If the answers are long the lines of

1. a very long time
2. very unlikely

then a newer but non "stable" release of software can be accepted as a
package.


From trygvis at opencsw.org  Mon Jul  6 19:28:18 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 06 Jul 2009 19:28:18 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A522EB9.9090400@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>	<op.uwnnrrurjbmpvo@imac.lan>
	<4A522EB9.9090400@opencsw.org>
Message-ID: <4A523432.3000304@opencsw.org>

Trygve Laugst?l wrote:
> Juergen Arndt wrote:
>> On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l 
>> <trygvis at opencsw.org> wrote:
>>
>>> Juergen Arndt wrote:
>>>> Hi,
>>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>>> OpenCSW. The software consists of the packages munin_master and 
>>>> munin_node.
>>>
>>> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know 
>>> to be as stable as 1.2.6?
>>>
>>> I see that Fedora and Debian both package 1.2.5/.6:
>>>
>>> http://dag.wieers.com/rpm/packages/munin/
>>> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all 
>>>
>>
>> 1.2.6 is the last official stable version, but it's already more than 
>> a year old and I was asked to package a more current version. If this 
>> does not comply with our policies, I will package 1.2.6 as well, no 
>> problem.
> 
> No, I was just wondering really.
> 
>>> I just tried your packages. The installation look fine and I made 
>>> some small notes on the wiki[1]. However, when I tried to create a 
>>> graph, all and any letters are rendered as a box. See the attached 
>>> image. I assume that this really is an rrdtool issue more than a 
>>> Munin issue.
>>>  [1]: http://wiki.opencsw.org/munin-package
>>
>> I cannot confirm that, on my installation the graphs look as expected. 
>> Maybe some dependencies are missing, I will try to dig into it.
>>
>> I will update your wiki page in the next future with some more 
>> installation details.
> 
> I think it would be nice if the package included two complete apache 
> configuration files (the basic and the one with CGI) so that the user 
> can just Include them from httpd.conf. From what I can tell it is not 
> likely that a user will need to adjust the configuration at all, at 
> least not to get started.
> 
> As for my rrd issue I just did a test from the rrd tutorial and the 
> graphs look ok there.

I tried to use the CGI version and got a Pango related error:

[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] (process:6716): 
Pango-CRITICAL **: No modules found:
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] No builtin or 
dynamically loaded modules were found.
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] PangoFc will not 
work correctly.
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] This probably 
means there was an error in the creation of:
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] 
'/opt/csw/etc/pango/pango.modules'
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] You should create 
this file by running:
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] 
pango-querymodules > '/opt/csw/etc/pango/pango.modules'

Which I did and I got this output:

/opt/csw/lib/64/pango/1.6.0/modules/pango-arabic-fc.so 
ArabicScriptEngineFc PangoEngineShape PangoRenderFc arabic:* nko:*
[snip]

Why are there only 64-bit modules there?

Anyway, running the CGI again gave me this:

Pango-WARNING **: ld.so.1: perl: fatal: 
/opt/csw/lib/64/pango/1.6.0/modules/pango-basic-fc.so: wrong ELF class: 
ELFCLASS64

--
Trygve

From trygvis at opencsw.org  Mon Jul  6 19:56:04 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 06 Jul 2009 19:56:04 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A523432.3000304@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>	<op.uwnnrrurjbmpvo@imac.lan>	<4A522EB9.9090400@opencsw.org>
	<4A523432.3000304@opencsw.org>
Message-ID: <4A523AB4.6010308@opencsw.org>

Trygve Laugst?l wrote:
> Trygve Laugst?l wrote:
>> Juergen Arndt wrote:
>>> On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l 
>>> <trygvis at opencsw.org> wrote:
>>>
>>>> Juergen Arndt wrote:
>>>>> Hi,
>>>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>>>> OpenCSW. The software consists of the packages munin_master and 
>>>>> munin_node.
>>>>
>>>> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know 
>>>> to be as stable as 1.2.6?
>>>>
>>>> I see that Fedora and Debian both package 1.2.5/.6:
>>>>
>>>> http://dag.wieers.com/rpm/packages/munin/
>>>> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all 
>>>>
>>>
>>> 1.2.6 is the last official stable version, but it's already more than 
>>> a year old and I was asked to package a more current version. If this 
>>> does not comply with our policies, I will package 1.2.6 as well, no 
>>> problem.
>>
>> No, I was just wondering really.
>>
>>>> I just tried your packages. The installation look fine and I made 
>>>> some small notes on the wiki[1]. However, when I tried to create a 
>>>> graph, all and any letters are rendered as a box. See the attached 
>>>> image. I assume that this really is an rrdtool issue more than a 
>>>> Munin issue.
>>>>  [1]: http://wiki.opencsw.org/munin-package
>>>
>>> I cannot confirm that, on my installation the graphs look as 
>>> expected. Maybe some dependencies are missing, I will try to dig into 
>>> it.
>>>
>>> I will update your wiki page in the next future with some more 
>>> installation details.
>>
>> I think it would be nice if the package included two complete apache 
>> configuration files (the basic and the one with CGI) so that the user 
>> can just Include them from httpd.conf. From what I can tell it is not 
>> likely that a user will need to adjust the configuration at all, at 
>> least not to get started.
>>
>> As for my rrd issue I just did a test from the rrd tutorial and the 
>> graphs look ok there.
> 
> I tried to use the CGI version and got a Pango related error:
> 
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] (process:6716): 
> Pango-CRITICAL **: No modules found:
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] No builtin or 
> dynamically loaded modules were found.
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] PangoFc will not 
> work correctly.
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] This probably 
> means there was an error in the creation of:
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] 
> '/opt/csw/etc/pango/pango.modules'
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] You should create 
> this file by running:
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] pango-querymodules 
>  > '/opt/csw/etc/pango/pango.modules'
> 
> Which I did and I got this output:
> 
> /opt/csw/lib/64/pango/1.6.0/modules/pango-arabic-fc.so 
> ArabicScriptEngineFc PangoEngineShape PangoRenderFc arabic:* nko:*
> [snip]
> 
> Why are there only 64-bit modules there?
> 
> Anyway, running the CGI again gave me this:
> 
> Pango-WARNING **: ld.so.1: perl: fatal: 
> /opt/csw/lib/64/pango/1.6.0/modules/pango-basic-fc.so: wrong ELF class: 
> ELFCLASS64

Ok, I've made more progress. If I ran 
/opt/csw/bin/i386/pango-querymodules I got a more sane configuration.

After that the CGI stuff started to fail, but when I went back to the 
non-CGI way it suddenly worked. Not using the CGI way is not an issue 
for me as it is not recommended at all so I'm happy.

--
Trygve


From trygvis at opencsw.org  Mon Jul  6 20:48:00 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Mon, 06 Jul 2009 20:48:00 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <4A5246E0.80105@opencsw.org>

Juergen Arndt wrote:
> Hi,
> 
> I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW. 
> The software consists of the packages munin_master and munin_node.
> 
> Feedback is welcome.

(I sure hope you're able to collect all my comments :)

Two comments:

1) The files in lib/ should most likely be in libexec/ as they're 
executables and not libraries. Not sure about the font file, probably 
doesn't matter much.

2) The graphs was not rendered with a monospaced font, probably because 
of using RRD 1.3. By hacking the LEGEND field like this it looks much 
better:

--- /opt/csw/lib/munin/munin-graph.orig	2009-07-06 20:43:48.340046000 +0200
+++ /opt/csw/lib/munin/munin-graph	2009-07-06 20:43:41.601520000 +0200
@@ -827,7 +827,7 @@
  	my @complete = ();
  	if ($RRDkludge) {
  	    push (@complete,
-		  '--font' ,'LEGEND:7:/opt/csw/lib/munin/VeraMono.ttf',
+		  '--font' ,'LEGEND:7:monospace',
  		  '--font' ,'UNIT:7:/opt/csw/lib/munin/VeraMono.ttf',
  		  '--font' ,'AXIS:7:/opt/csw/lib/munin/VeraMono.ttf');
  	}

This is the same as Debian does [1] amd has been integrated to the main 
build [2].

[1]: http://munin.projects.linpro.no/changeset/1763
[2]: http://munin.projects.linpro.no/changeset/1797

--
Trygve

From mwatters at opencsw.org  Mon Jul  6 21:44:31 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Mon, 06 Jul 2009 14:44:31 -0500
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A523AB4.6010308@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>	<op.uwnnrrurjbmpvo@imac.lan>	<4A522EB9.9090400@opencsw.org>	<4A523432.3000304@opencsw.org>
	<4A523AB4.6010308@opencsw.org>
Message-ID: <4A52541F.5000203@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Trygve,
Trygve Laugst?l wrote:

>> [snip]
>>
>> Why are there only 64-bit modules there?
>>
>> Anyway, running the CGI again gave me this:
>>
>> Pango-WARNING **: ld.so.1: perl: fatal:
>> /opt/csw/lib/64/pango/1.6.0/modules/pango-basic-fc.so: wrong ELF
>> class: ELFCLASS64
> 
> Ok, I've made more progress. If I ran
> /opt/csw/bin/i386/pango-querymodules I got a more sane configuration.
> 
> After that the CGI stuff started to fail, but when I went back to the
> non-CGI way it suddenly worked. Not using the CGI way is not an issue
> for me as it is not recommended at all so I'm happy.



There is already a thread bouncing around on the pango problem.
subject: [csw-maintainers] libpango 1.24 Problems ?

- --

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

iEYEARECAAYFAkpSVB8ACgkQLrhmsXMSLxddMwCgsENM1ApVj3PVChVK2Myw2wNL
aBkAoIZuA91F0wBy0hN/ZMSUCU9pywGz
=3bd1
-----END PGP SIGNATURE-----

From skayser at opencsw.org  Mon Jul  6 22:12:14 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Mon, 06 Jul 2009 22:12:14 +0200
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <20090706161151.GA36831@bolthole.com>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com>
Message-ID: <4A525A9E.10306@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 06, 2009 at 05:34:51PM +0200, Sebastian Kayser wrote:
>> Trussing the acomp shows a myriad of brk()s and the process hogs
>> memory like there is no tomorrow.
>>
>>    PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
>>  29072 skayser   684M  670M cpu13   30    0   0:01.23 3.1% acomp/1
> 
> yes. this is known behaviour of cc.
> (although I havent seen it with libiconv specifically)
> 
> certain obscure optimization cases, explode, so to speak, taking extreme
> memory usage and disk space for it to "traverse the problem space", if you
> will.
> 
> Your choices are:
> 
> a) wait for a long time, and it will eventually complete, if the machine
> doesnt run out of resources
> 
> b) reduce the optimization flags to -xO2 or -xO3

Thanks for your feedback Phil. Reducing to -xO2 or -xO3 didn't help
though. Now comes the fun part. When i keep the --fast --xnolibmopt
flags as they are and run the configure command that GAR would issue
manually, it completes just fine :/

There is someone over at the OpenSolaris forums who has observed
something similar. cc hanging when invoked from make, but fine on its
own [1]. Unfortunately, neither the forum thread nor the referenced bug
entry [2] help with a working solution.

Sebastian

[1] http://opensolaris.org/jive/thread.jspa?messageID=209219
[2] http://bugs.sun.com/view_bug.do?bug_id=6668199

From phil at bolthole.com  Mon Jul  6 22:29:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 13:29:56 -0700
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
	./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A525A9E.10306@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com> <4A525A9E.10306@opencsw.org>
Message-ID: <20090706202956.GG60655@bolthole.com>

On Mon, Jul 06, 2009 at 10:12:14PM +0200, Sebastian Kayser wrote:
> Philip Brown wrote:
> > Your choices are:
> > 
> > a) wait for a long time, and it will eventually complete, if the machine
> > doesnt run out of resources
> > 
> > b) reduce the optimization flags to -xO2 or -xO3
> 
> Thanks for your feedback Phil. Reducing to -xO2 or -xO3 didn't help
> though. Now comes the fun part. When i keep the --fast --xnolibmopt
> flags as they are and run the configure command that GAR would issue
> manually, it completes just fine :/

Did you try letting it run for half an hour?


From skayser at opencsw.org  Mon Jul  6 22:35:22 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Mon, 06 Jul 2009 22:35:22 +0200
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <20090706202956.GG60655@bolthole.com>
References: <4A52199B.5070808@opencsw.org>	<20090706161151.GA36831@bolthole.com>
	<4A525A9E.10306@opencsw.org> <20090706202956.GG60655@bolthole.com>
Message-ID: <4A52600A.6070803@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 06, 2009 at 10:12:14PM +0200, Sebastian Kayser wrote:
>> Philip Brown wrote:
>>> Your choices are:
>>>
>>> a) wait for a long time, and it will eventually complete, if the machine
>>> doesnt run out of resources
>>>
>>> b) reduce the optimization flags to -xO2 or -xO3
>> Thanks for your feedback Phil. Reducing to -xO2 or -xO3 didn't help
>> though. Now comes the fun part. When i keep the --fast --xnolibmopt
>> flags as they are and run the configure command that GAR would issue
>> manually, it completes just fine :/
> 
> Did you try letting it run for half an hour?

10 minutes maybe (didn't help), not yet half an hour.

build8s is currently already being killed by something else (load
average: 33.58, 33.62, 34.02), will try the half hour thing on some of
the next few days. With the manual workaround it's at least possible now
to complete the build.

Sebastian

From phil at bolthole.com  Mon Jul  6 22:44:19 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 13:44:19 -0700
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
	./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A52600A.6070803@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com> <4A525A9E.10306@opencsw.org>
	<20090706202956.GG60655@bolthole.com>
	<4A52600A.6070803@opencsw.org>
Message-ID: <20090706204418.GH60655@bolthole.com>

On Mon, Jul 06, 2009 at 10:35:22PM +0200, Sebastian Kayser wrote:
> build8s is currently already being killed by something else (load
> average: 33.58, 33.62, 34.02), will try the half hour thing on some of
> the next few days. With the manual workaround it's at least possible now
> to complete the build.
> 

yeah, when in doubt, hand-hack the configure script to just eliminate the
stupid test case :)


From ja at opencsw.org  Mon Jul  6 23:10:42 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Mon, 06 Jul 2009 23:10:42 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A5246E0.80105@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org>
Message-ID: <op.uwnzr4lqjbmpvo@imac.lan>

On Mon, 06 Jul 2009 20:48:00 +0200, Trygve Laugst?l <trygvis at opencsw.org>  
wrote:

> Juergen Arndt wrote:
>> Hi,
>>  I've put Munin 1.3.4 into testing. It's the initial version for  
>> OpenCSW. The software consists of the packages munin_master and  
>> munin_node.
>>  Feedback is welcome.
>
> (I sure hope you're able to collect all my comments :)

I'll try my very best ;) (I use for such purposes a mind map tool -  
http://freemind.sourceforge.net/wiki/index.php/Main_Page, which is really  
capable for such things. Maybe it's interesting for someone too)

So, I have the following points:

1) Use of monospaced font for the graphs
2) Moving executables from /lib to /libexec (thanks for your attention, I  
absolutely didn't see that)
3) Provide Apache config files ready for including - where should the  
files be placed? Already in /opt/csw/apache2/etc/extra ?
4) The problems with pango and the cgi mode. I guess, this will take the  
most time, actually I don't know yet, where the problem is located - in  
pango or in munin or in both.
5) Maybe packages for 1.2.6 are needed.

Did I forget anything? :) I hope, I can solve the first three issues  
tomorrow.

Juergen

-- 
Juergen Arndt

From trygvis at opencsw.org  Tue Jul  7 03:35:05 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Tue, 07 Jul 2009 03:35:05 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnzr4lqjbmpvo@imac.lan>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A5246E0.80105@opencsw.org>
	<op.uwnzr4lqjbmpvo@imac.lan>
Message-ID: <4A52A649.1070606@opencsw.org>

Juergen Arndt wrote:
> On Mon, 06 Jul 2009 20:48:00 +0200, Trygve Laugst?l 
> <trygvis at opencsw.org> wrote:
> 
>> Juergen Arndt wrote:
>>> Hi,
>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>> OpenCSW. The software consists of the packages munin_master and 
>>> munin_node.
>>>  Feedback is welcome.
>>
>> (I sure hope you're able to collect all my comments :)
> 
> I'll try my very best ;) (I use for such purposes a mind map tool - 
> http://freemind.sourceforge.net/wiki/index.php/Main_Page, which is 
> really capable for such things. Maybe it's interesting for someone too)
> 
> So, I have the following points:
> 
> 1) Use of monospaced font for the graphs
> 2) Moving executables from /lib to /libexec (thanks for your attention, 
> I absolutely didn't see that)

You probably want to make sure that your use owner:group consistently 
throughout the file objects.

> 3) Provide Apache config files ready for including - where should the 
> files be placed? Already in /opt/csw/apache2/etc/extra ?

I would suggest /etc/opt/csw/munin/apache.conf and 
/etc/opt/csw/munin/apache-cgi.conf

> 4) The problems with pango and the cgi mode. I guess, this will take the 
> most time, actually I don't know yet, where the problem is located - in 
> pango or in munin or in both.

I would ignore this for now as Munin only calls out to rrdtool and the 
HTML etc was generated perfectly.

> 5) Maybe packages for 1.2.6 are needed.

The package is working just fine for me now, so I think this should be 
ok. I've even made a plugin for it and it work :)

> Did I forget anything? :) I hope, I can solve the first three issues 
> tomorrow.

I'm leaving for vacation tomorrow so I probably won't be able to test it 
much, but I can do it as soon as I get back home.

--
Trygve

From phil at bolthole.com  Tue Jul  7 05:08:13 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 20:08:13 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A52A649.1070606@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
Message-ID: <20090707030813.GA20526@bolthole.com>

On Tue, Jul 07, 2009 at 03:35:05AM +0200, Trygve Laugst??l wrote:
>> 3) Provide Apache config files ready for including - where should the  
>> files be placed? Already in /opt/csw/apache2/etc/extra ?
>
> I would suggest /etc/opt/csw/munin/apache.conf and  
> /etc/opt/csw/munin/apache-cgi.conf

i would disagree.
these files are supposed to be read-only. so they are fine for /opt/csw
somewhere, and actually much better off there.


/opt/csw/(some-munin-path) would be just fine, so long as they are well
documented in /opt/csw/share/doc/munin/README  or something.


From bwalton at opencsw.org  Tue Jul  7 05:54:11 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 06 Jul 2009 23:54:11 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707030813.GA20526@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
Message-ID: <1246937677-sup-8061@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Mon Jul 06 23:08:13 -0400 2009:
> >> files be placed? Already in /opt/csw/apache2/etc/extra ?
> >
> > I would suggest /etc/opt/csw/munin/apache.conf and  
> > /etc/opt/csw/munin/apache-cgi.conf
> 
> i would disagree.
> these files are supposed to be read-only. so they are fine for /opt/csw
> somewhere, and actually much better off there.

...But they are part of the apache configuration, so it would make
sense to group them logically with the rest of the apache
configuration.  Personally though, I like the way debian does this.

An example: I pulled squirrelmail onto a box tonight (as a quick
example, phpmyadmin is another).  It added
/etc/squirrelmail/apache.conf.  The README.Debian.gz states:

<blockquote>
Apache
======

If you want to use squirrelmail with apache, it is recommended to
modify /etc/squirrelmail/apache.conf to your own wishes, and then add
a symlink to that file from /etc/apache{,-ssl,-perl,2}/conf.d, like:

 $ ln -s /etc/squirrelmail/apache.conf
 /etc/apache2/conf.d/squirrelmail.conf

Alternatively, you could also 'Include' the configuration file in the
relevant apache configuration file(s)
</blockquote>

The phpmyadmin package has a similarly worded README.Debian file.

It's configuration, whether commonly edited or not, so it should go in
etc somewhere.  Debian has a really nice apache setup and 'add on'
integration.  We follow their lead in other ways, why not here too?

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090706/bfd994e6/attachment.asc>

From phil at bolthole.com  Tue Jul  7 06:45:35 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 21:45:35 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246937677-sup-8061@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
Message-ID: <20090707044535.GA5845@bolthole.com>

On Mon, Jul 06, 2009 at 11:54:11PM -0400, Ben Walton wrote:
> ...
> ...But they are part of the apache configuration, so it would make
> sense to group them logically with the rest of the apache
> configuration.

it doesnt share well. This is something meant to be potentialy shared.
if there isnt a strong reason to keep something OUT of /opt/csw.. it
belongs under /opt/csw

or to put it another way.. if there isnt a strong reason to keep a file
machine-local, it belongs under/opt/csw

there is no reason to keep the read-only config as machine-local



>  Personally though, I like the way debian does this.

Hnmmm. not sure i like that way we can come up with something better
perhaps :)

	

From jeff at cjsa.com  Tue Jul  7 08:49:09 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Tue, 7 Jul 2009 06:49:09 GMT
Subject: [csw-maintainers] Library problem?
Message-ID: <KMEG9x.Cxz@cjsa.com>

I have just started noticing that something has changed on my system and
now a number of applications are displaying all text as rectangular boxes.
The rapidsvn application is one example.  This occurs even though these
applications have not changed, so the problem must be in a library.  My
best guess is that it might be glib2 or gtk2, but I actually know little
about them, so this is just a guess.  However, I have another app. that did
not come from CSW but is linked to these two libraries, so I think they may
be suspects.  Is anyone else seeing this problem, and if we can identify
the source, can we please get it fixed ASAP?

Regards,
-- 
Jeffery Small


From james at opencsw.org  Tue Jul  7 10:22:02 2009
From: james at opencsw.org (James Lee)
Date: Tue, 07 Jul 2009 08:22:02 GMT
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A52600A.6070803@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com> <4A525A9E.10306@opencsw.org>
	<20090706202956.GG60655@bolthole.com>
	<4A52600A.6070803@opencsw.org>
Message-ID: <20090707.8220200.4140658907@gyor.oxdrove.co.uk>

On 06/07/09, 21:35:22, Sebastian Kayser <skayser at opencsw.org> wrote 
regarding Re: [csw-maintainers] libiconv and --fast --xnolibmopt, 
./configure hangs at "checking for stdbool.h that conforms to C99":

> build8s is currently already being killed by something else (load
> average: 33.58, 33.62, 34.02),

It's build10s, the same physical machine, that needs some signal 9s.

In the meantime use build8s.go.opencsw.org




James.

From ja at opencsw.org  Tue Jul  7 13:10:25 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Tue, 07 Jul 2009 13:10:25 +0200
Subject: [csw-maintainers] csqlserver processes killed on buildfarm
Message-ID: <op.uwo2nnlcjbmpvo@ibm-1344209b5ff.ibm.com>

Hi all,

Dago asked me to kill all csqlserver processes on the buildfarm because  
the system was under heavy load and not very responsive.

Now everything should be ok again. Sorry, if someone's work got  
interrupted because of this.

Juergen

-- 
Juergen Arndt

From trygvis at opencsw.org  Tue Jul  7 13:32:22 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 07 Jul 2009 13:32:22 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707044535.GA5845@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A5246E0.80105@opencsw.org>
	<op.uwnzr4lqjbmpvo@imac.lan>	<4A52A649.1070606@opencsw.org>	<20090707030813.GA20526@bolthole.com>	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
Message-ID: <4A533246.30908@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 06, 2009 at 11:54:11PM -0400, Ben Walton wrote:
>> ...
>> ...But they are part of the apache configuration, so it would make
>> sense to group them logically with the rest of the apache
>> configuration.
> 
> it doesnt share well. This is something meant to be potentialy shared.
> if there isnt a strong reason to keep something OUT of /opt/csw.. it
> belongs under /opt/csw
> 
> or to put it another way.. if there isnt a strong reason to keep a file
> machine-local, it belongs under/opt/csw
> 
> there is no reason to keep the read-only config as machine-local

It's not read only but it will work unmodified in a lot of cases. One 
reason I see to modify it is to add security. A read-only .CSW copy is 
always nice to have so you can diff on upgrades.

>>  Personally though, I like the way debian does this.

I also find the Debian way very easy to work with.

--
Trygve

From bwalton at opencsw.org  Tue Jul  7 15:47:24 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 09:47:24 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707044535.GA5845@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
Message-ID: <1246973415-sup-1078@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 00:45:35 -0400 2009:
> > ...But they are part of the apache configuration, so it would make
> > sense to group them logically with the rest of the apache
> > configuration.
> 
> it doesnt share well. This is something meant to be potentialy shared.
> if there isnt a strong reason to keep something OUT of /opt/csw.. it
> belongs under /opt/csw

The munin package already ships files in /etc/opt/csw/munin though,
which are machine local configuration.  Why not keep things together?
If munin shipped a manifest that didn't touch /etc/ or /var on the
machine, your point would be better received on my end.  [It's still
well recieved, I just don't think it applies well to this particular
situation.]

> there is no reason to keep the read-only config as machine-local

But it's not 100% read only.  As Trygve mentioned, it may need to be
tuned to put security restrictions into it, etc.

> >  Personally though, I like the way debian does this.
> 
> Hnmmm. not sure i like that way we can come up with something better
> perhaps :)

It would be nice to have this extended into something like their
a2enmod/a2ensite setup (for modules and vhosts).  Placing the files in
$apacheconf/addons-available and then symlinking into
$apacheconf/addons-enabled would be nicer, I think.  If it were
decided that all shipped apache config snippets should live with the
rest of the apache config, this would get my vote.

Maybe one the following ground rules could be used?

1. Install all apache config files in the directory apache puts its
   own configuration in (most likely a sub-directory, for clarity).
2. If the package ships other machine local config, put the file with
   those, otherwise, somewhere in /opt/csw/etc?

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 mwatters at opencsw.org  Tue Jul  7 16:06:30 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Jul 2009 09:06:30 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <KMEG9x.Cxz@cjsa.com>
References: <KMEG9x.Cxz@cjsa.com>
Message-ID: <4A535666.4030009@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jeffery Small wrote:
> I have just started noticing that something has changed on my system and
> now a number of applications are displaying all text as rectangular boxes.
> The rapidsvn application is one example.  This occurs even though these
> applications have not changed, so the problem must be in a library.  My
> best guess is that it might be glib2 or gtk2, but I actually know little
> about them, so this is just a guess.  However, I have another app. that did
> not come from CSW but is linked to these two libraries, so I think they may
> be suspects.  Is anyone else seeing this problem, and if we can identify
> the source, can we please get it fixed ASAP?
> 
> Regards,

have you updated anything at all recently?

I can not be sure without more info, but my suspicion is you are missing
/opt/csw/etc/pango/pango.modules or the file contains the 64bit versions.


if you are running 1.24 you need to run
/opt/csw/bin/sparcv8/pango-querymodules >/opt/csw/etc/pango/pango.modules

for an earlier version you should run
/opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules

Please let us know if this does not fix the issue.


- --

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

iEYEARECAAYFAkpTVmUACgkQLrhmsXMSLxf6ZACg2FzCyExdMT/yiPlHXDA56TJS
e24AnibSo+HOwdMRBtraZDPQV7QavfqA
=w3LJ
-----END PGP SIGNATURE-----

From phil at bolthole.com  Tue Jul  7 16:25:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 07:25:17 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246973415-sup-1078@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
Message-ID: <20090707142516.GB11549@bolthole.com>

On Tue, Jul 07, 2009 at 09:47:24AM -0400, Ben Walton wrote:
> The munin package already ships files in /etc/opt/csw/munin though,
> which are machine local configuration.

ah ok then.




From phil at bolthole.com  Tue Jul  7 16:26:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 07:26:33 -0700
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A535666.4030009@opencsw.org>
References: <KMEG9x.Cxz@cjsa.com> <4A535666.4030009@opencsw.org>
Message-ID: <20090707142633.GC11549@bolthole.com>

On Tue, Jul 07, 2009 at 09:06:30AM -0500, Mike Watters wrote:
> if you are running 1.24 you need to run
> /opt/csw/bin/sparcv8/pango-querymodules >/opt/csw/etc/pango/pango.modules
> 
> for an earlier version you should run
> /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules
> 
> Please let us know if this does not fix the issue.

It would seem as though our existing users might be similarly plagued with
this sort of thing if they upgraded recently.

if so, this very nasty situation, warrants an email about it, to the
announce and user lists.

From bwalton at opencsw.org  Tue Jul  7 16:31:41 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 10:31:41 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707142516.GB11549@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
Message-ID: <1246976931-sup-3020@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 10:25:17 -0400 2009:
> > The munin package already ships files in /etc/opt/csw/munin though,
> > which are machine local configuration.
> 
> ah ok then.

But, which is the better approach for us, going forward.  There are a
lot of packages that could ship a similar apache snippet.  Read-only
/opt/csw issues as a separate issue, which do people feel is cleaner?
Lumping apache config snippets with apache, or sticking them with the
config for the package that ships it?  I think I could argue either
way.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/c0978a0e/attachment.asc>

From william at wbonnet.net  Tue Jul  7 17:28:36 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 07 Jul 2009 17:28:36 +0200
Subject: [csw-maintainers] [csw-users] Kicad app broke with upgrade
	pkg-get -Uu
In-Reply-To: <4A536697.2010704@optonline.net>
References: <4A520D78.2070702@optonline.net>
	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>
	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>
	<4A527E0F.7040308@wbonnet.net> <4A536697.2010704@optonline.net>
Message-ID: <4A5369A4.6080002@wbonnet.net>

Hi Paul

>> Can you please send me the full output of the following commands :
>>
>> ldd /opt/csw/bin/kicad
>> pkginfo -l CSWkicad
Thanks, i was thinking at something but now i have the answer :)

> $ pkginfo -l CSWkicad
>   HOTLINE:  http://www.blastwave.org/bugtrack
>     EMAIL:  dlaigle at blastwave.org
You are using the package from Blastwave, not from OpenCSW. And this is 
not an error on your side.

You installed the package before the fork, and this package was not 
updated since. But... the libs it is using have been modified, and 
libxcb linking is no longer compatible between both project.

I suggest your try to remove then reinstall kicad. It should work.

Thanks for raising this problem. We will try to solve it in a "global way".

Cheers
W.

From phil at bolthole.com  Tue Jul  7 17:34:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 08:34:59 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246976931-sup-3020@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
Message-ID: <20090707153459.GL60655@bolthole.com>

On Tue, Jul 07, 2009 at 10:31:41AM -0400, Ben Walton wrote:
> But, which is the better approach for us, going forward.  There are a
> lot of packages that could ship a similar apache snippet.  Read-only
> /opt/csw issues as a separate issue, which do people feel is cleaner?
> Lumping apache config snippets with apache, or sticking them with the
> config for the package that ships it?  I think I could argue either
> way.

I think that, for sysadmins coming late to the scene, having an "include"
that is a long one, such as

include  /some/path/to/other/application/stuff

is actually more informative to have in the apache configs. it provides a
better "trail of breadcrumbs", to where the "other application" really
lives.


From william at wbonnet.net  Tue Jul  7 17:50:59 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 07 Jul 2009 17:50:59 +0200
Subject: [csw-maintainers] [csw-users] Kicad app broke with
 upgrade	pkg-get -Uu
In-Reply-To: <4A5369A4.6080002@wbonnet.net>
References: <4A520D78.2070702@optonline.net>	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>	<4A527E0F.7040308@wbonnet.net>
	<4A536697.2010704@optonline.net> <4A5369A4.6080002@wbonnet.net>
Message-ID: <4A536EE3.4000705@wbonnet.net>

Hi

> You are using the package from Blastwave, not from OpenCSW. And this 
> is not an error on your side.
>
> You installed the package before the fork, and this package was not 
> updated since. But... the libs it is using have been modified, and 
> libxcb linking is no longer compatible between both project.
>
> I suggest your try to remove then reinstall kicad. It should work.
The problem Paul raised may happen again with several package which have 
not been updated since we forked.

I'm affraid we will have several other packages with broken dependencies 
if we don't rebuild it. Maybe we could script the detection of such 
packages ?

cheers
W.

From bwalton at opencsw.org  Tue Jul  7 18:03:51 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:03:51 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707153459.GL60655@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
Message-ID: <1246982619-sup-2337@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 11:34:59 -0400 2009:
> include  /some/path/to/other/application/stuff
> 
> is actually more informative to have in the apache configs. it provides a
> better "trail of breadcrumbs", to where the "other application" really
> lives.

I agree with this.  Shall we adopt 'the debian way' then?

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/3c8ef3f6/attachment.asc>

From bwalton at opencsw.org  Tue Jul  7 18:06:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:06:36 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246982619-sup-2337@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
Message-ID: <1246982722-sup-4775@ntdws12.chass.utoronto.ca>

Excerpts from Ben Walton's message of Tue Jul 07 12:03:51 -0400 2009:
> I agree with this.  Shall we adopt 'the debian way' then?

Sorry, by 'debian way,' I mean: apache config snippets live with the
rest of the package's config files (be it in /etc/opt/csw/ or
/opt/csw/etc).

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/228a8b57/attachment.asc>

From phil at bolthole.com  Tue Jul  7 18:08:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 09:08:06 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246982619-sup-2337@ntdws12.chass.utoronto.ca>
References: <op.uwnzr4lqjbmpvo@imac.lan> <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
Message-ID: <20090707160806.GQ60655@bolthole.com>

On Tue, Jul 07, 2009 at 12:03:51PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Tue Jul 07 11:34:59 -0400 2009:
> > include  /some/path/to/other/application/stuff
> > 
> > is actually more informative to have in the apache configs. it provides a
> > better "trail of breadcrumbs", to where the "other application" really
> > lives.
> 
> I agree with this.  Shall we adopt 'the debian way' then?

Errr... the "debian way", if I understood it directly, was to make
symlinks.
What I was saying, did not involve any symlinks.

From phil at bolthole.com  Tue Jul  7 18:11:43 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 09:11:43 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246982722-sup-4775@ntdws12.chass.utoronto.ca>
References: <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
	<1246982722-sup-4775@ntdws12.chass.utoronto.ca>
Message-ID: <20090707161143.GR60655@bolthole.com>

On Tue, Jul 07, 2009 at 12:06:36PM -0400, Ben Walton wrote:
> Excerpts from Ben Walton's message of Tue Jul 07 12:03:51 -0400 2009:
> > I agree with this.  Shall we adopt 'the debian way' then?
> 
> Sorry, by 'debian way,' I mean: apache config snippets live with the
> rest of the package's config files (be it in /etc/opt/csw/ or
> /opt/csw/etc).

if by "the package", you mean "the package that isnt apache itself", then
yes I agree.


From bwalton at opencsw.org  Tue Jul  7 18:16:20 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:16:20 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707160806.GQ60655@bolthole.com>
References: <op.uwnzr4lqjbmpvo@imac.lan> <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
	<20090707160806.GQ60655@bolthole.com>
Message-ID: <1246983310-sup-582@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 12:08:06 -0400 2009:
> Errr... the "debian way", if I understood it directly, was to make
> symlinks.
> What I was saying, did not involve any symlinks.

That choice would still be up to the maintainer (whether we're
referring to debian or csw).  The key point is where the real file
lives, which I think we agree is with the rest of the config for the
package.  [Apologies for lack of clarity.]

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/a019a102/attachment.asc>

From bwalton at opencsw.org  Tue Jul  7 18:17:19 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:17:19 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246983310-sup-582@ntdws12.chass.utoronto.ca>
References: <op.uwnzr4lqjbmpvo@imac.lan> <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
	<20090707160806.GQ60655@bolthole.com>
	<1246983310-sup-582@ntdws12.chass.utoronto.ca>
Message-ID: <1246983414-sup-8022@ntdws12.chass.utoronto.ca>

Excerpts from Ben Walton's message of Tue Jul 07 12:16:20 -0400 2009:
> That choice would still be up to the maintainer (whether we're

s/maintainer/administrator/

Jeezaloo, I need another coffee.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/a6847c97/attachment.asc>

From mwatters at opencsw.org  Tue Jul  7 20:25:11 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Jul 2009 13:25:11 -0500
Subject: [csw-maintainers] [csw-users] Kicad app broke with
 upgrade	pkg-get -Uu
In-Reply-To: <4A536EE3.4000705@wbonnet.net>
References: <4A520D78.2070702@optonline.net>	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>	<4A527E0F.7040308@wbonnet.net>	<4A536697.2010704@optonline.net>
	<4A5369A4.6080002@wbonnet.net> <4A536EE3.4000705@wbonnet.net>
Message-ID: <4A539307.4010601@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi
> 
>> You are using the package from Blastwave, not from OpenCSW. And this
>> is not an error on your side.
>>
>> You installed the package before the fork, and this package was not
>> updated since. But... the libs it is using have been modified, and
>> libxcb linking is no longer compatible between both project.
>>
>> I suggest your try to remove then reinstall kicad. It should work.
> The problem Paul raised may happen again with several package which have
> not been updated since we forked.
> 
> I'm affraid we will have several other packages with broken dependencies
> if we don't rebuild it. Maybe we could script the detection of such
> packages ?
> 
> cheers
> W.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

if we file a "critical bug" against those packages, then my page to determine
retired maintainers and orphaned packages.[1]
could be modified to weight the packages by # of bugs or bug severity or
importance (somehow).  thoughts?




[1] http://www.opencsw.org/need_maintainer

- --

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

iEYEARECAAYFAkpTkwcACgkQLrhmsXMSLxczyQCffyTatB94U907s+FkY2mhHRCm
C9IAoMLe+QLsIy2/UhKxumdbMISvjyXu
=pbr9
-----END PGP SIGNATURE-----

From phil at bolthole.com  Tue Jul  7 20:31:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 11:31:56 -0700
Subject: [csw-maintainers] [csw-users] Kicad app broke with
	upgrade	pkg-get -Uu
In-Reply-To: <4A539307.4010601@opencsw.org>
References: <4A520D78.2070702@optonline.net> <4A520E2A.4010009@wbonnet.net>
	<4A5210E6.5070802@optonline.net> <4A52133F.5050505@wbonnet.net>
	<4A5224C9.6020503@optonline.net> <4A527E0F.7040308@wbonnet.net>
	<4A536697.2010704@optonline.net> <4A5369A4.6080002@wbonnet.net>
	<4A536EE3.4000705@wbonnet.net> <4A539307.4010601@opencsw.org>
Message-ID: <20090707183156.GS60655@bolthole.com>

On Tue, Jul 07, 2009 at 01:25:11PM -0500, Mike Watters wrote:
> if we file a "critical bug" against those packages, then my page to determine
> retired maintainers and orphaned packages.[1]
> could be modified to weight the packages by # of bugs or bug severity or
> importance (somehow).  thoughts?

There is a pre-existing "show bugs" page somewheres... I think Dago wrote
it.
it could probably be tweaked more eaasily for what you have in mind.

From skayser at opencsw.org  Tue Jul  7 20:49:15 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Jul 2009 20:49:15 +0200
Subject: [csw-maintainers] [csw-users] Kicad app broke
 with	upgrade	pkg-get -Uu
In-Reply-To: <20090707183156.GS60655@bolthole.com>
References: <4A520D78.2070702@optonline.net>
	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>
	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>
	<4A527E0F.7040308@wbonnet.net>	<4A536697.2010704@optonline.net>
	<4A5369A4.6080002@wbonnet.net>	<4A536EE3.4000705@wbonnet.net>
	<4A539307.4010601@opencsw.org>
	<20090707183156.GS60655@bolthole.com>
Message-ID: <4A5398AB.7060801@opencsw.org>

Philip Brown wrote:
> On Tue, Jul 07, 2009 at 01:25:11PM -0500, Mike Watters wrote:
>> if we file a "critical bug" against those packages, then my page to determine
>> retired maintainers and orphaned packages.[1]
>> could be modified to weight the packages by # of bugs or bug severity or
>> importance (somehow).  thoughts?
> 
> There is a pre-existing "show bugs" page somewheres... I think Dago wrote
> it.
> it could probably be tweaked more eaasily for what you have in mind.

Are you referring to this one?

  http://opencsw.org/buglist/buglist.cgi

Sebastian

From phil at bolthole.com  Tue Jul  7 20:58:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 11:58:26 -0700
Subject: [csw-maintainers] [csw-users] Kicad app broke with
	upgrade	pkg-get -Uu
In-Reply-To: <4A5398AB.7060801@opencsw.org>
References: <4A5210E6.5070802@optonline.net> <4A52133F.5050505@wbonnet.net>
	<4A5224C9.6020503@optonline.net> <4A527E0F.7040308@wbonnet.net>
	<4A536697.2010704@optonline.net> <4A5369A4.6080002@wbonnet.net>
	<4A536EE3.4000705@wbonnet.net> <4A539307.4010601@opencsw.org>
	<20090707183156.GS60655@bolthole.com>
	<4A5398AB.7060801@opencsw.org>
Message-ID: <20090707185826.GT60655@bolthole.com>

On Tue, Jul 07, 2009 at 08:49:15PM +0200, Sebastian Kayser wrote:
> Philip Brown wrote:
> > There is a pre-existing "show bugs" page somewheres... I think Dago wrote
> > it.
> > it could probably be tweaked more eaasily for what you have in mind.
> 
> Are you referring to this one?
> 
>   http://opencsw.org/buglist/buglist.cgi

yes.


From james at opencsw.org  Tue Jul  7 21:25:56 2009
From: james at opencsw.org (James Lee)
Date: Tue, 07 Jul 2009 19:25:56 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
Message-ID: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>

Anyone understand pkgconfig?  This looks easy but I'm blind as to why
it's not working:

build8s:/home/james$ export PKG_CONFIG_PATH=/opt/csw/lib/pkgconfig
build8s:/home/james$ if /opt/csw/bin/pkg-config --exists 'gtk+-2.0'
if> then
then>    echo YES
then> else
else>    echo NO
else> fi
sh: gnome-config: not found
NO
build8s:/home/james$ ls -l /opt/csw/lib/pkgconfig/gtk+-2.0.pc
-rw-r--r--   1 root     bin          345 Jun  7  2008
/opt/csw/lib/pkgconfig/gtk+-2.0.pc



There seems to be a dependency problem but I still can't work out what's
gone missing:

$ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
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



Help!  I'm close to editing "if true" in to my good friend the
configure file.




James.


From phil at bolthole.com  Tue Jul  7 21:35:55 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 12:35:55 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
Message-ID: <20090707193555.GU60655@bolthole.com>

On Tue, Jul 07, 2009 at 07:25:56PM +0000, James Lee wrote:
> Anyone understand pkgconfig?  This looks easy but I'm blind as to why
> it's not working:
> ....
> $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
> 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

seems fairly clear: xcb isnt installed on build8s.
or at least the pkgconfig files for it arent installed.

or actually... they are installed in /opt/csw/X11/lib/pkgconfig

Methinks cairo needs to be rebuilt or something, to give the "new and
proper" path to it or something. but otherwise, perhaps you can add that
dir to your pkgconfig path as first.


I hope our cairo/new-x11 maintainres can come up with a more transparent
solution to this stuff though.

From james at opencsw.org  Tue Jul  7 21:51:48 2009
From: james at opencsw.org (James Lee)
Date: Tue, 07 Jul 2009 19:51:48 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707193555.GU60655@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
Message-ID: <20090707.19514800.2647352656@gyor.oxdrove.co.uk>

On 07/07/09, 20:35:55, Philip Brown <phil at bolthole.com> wrote regarding Re: 
[csw-maintainers] "pkgconfig --exist ..." is telling porkies:

> On Tue, Jul 07, 2009 at 07:25:56PM +0000, James Lee wrote:
> > Anyone understand pkgconfig?  This looks easy but I'm blind as to why
> > it's not working:
> > ....
> > $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
> > 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

> seems fairly clear: xcb isnt installed on build8s.

Why does that clearly affect whether gtk+-2.0 exists which clearly
it does.  Even truss doesn't show it looking for xcb.  OK then, too
much time wasted, "if true" in the patch file for configure.




> or at least the pkgconfig files for it arent installed.

> or actually... they are installed in /opt/csw/X11/lib/pkgconfig

> Methinks cairo needs to be rebuilt or something, to give the "new and
> proper" path to it or something. but otherwise, perhaps you can add that
> dir to your pkgconfig path as first.

Which package needs the bug report?  Gtk+-2.0, cairo, xcb or pkgconfig?




James.

From mwatters at opencsw.org  Tue Jul  7 22:01:14 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Jul 2009 15:01:14 -0500
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707.19514800.2647352656@gyor.oxdrove.co.uk>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
Message-ID: <4A53A98A.8060201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James Lee wrote:
> On 07/07/09, 20:35:55, Philip Brown <phil at bolthole.com> wrote regarding Re: 
> [csw-maintainers] "pkgconfig --exist ..." is telling porkies:
> 
>> On Tue, Jul 07, 2009 at 07:25:56PM +0000, James Lee wrote:
>>> Anyone understand pkgconfig?  This looks easy but I'm blind as to why
>>> it's not working:
>>> ....
>>> $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
>>> 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
> 
>> seems fairly clear: xcb isnt installed on build8s.
> 
> Why does that clearly affect whether gtk+-2.0 exists which clearly
> it does.  Even truss doesn't show it looking for xcb.  OK then, too
> much time wasted, "if true" in the patch file for configure.
> 
> 
> 
> 
>> or at least the pkgconfig files for it arent installed.
> 
>> or actually... they are installed in /opt/csw/X11/lib/pkgconfig
> 
>> Methinks cairo needs to be rebuilt or something, to give the "new and
>> proper" path to it or something. but otherwise, perhaps you can add that
>> dir to your pkgconfig path as first.
> 
> Which package needs the bug report?  Gtk+-2.0, cairo, xcb or pkgconfig?

xcb and in fact all the "new" xlib packages have their PKG_CONFIG_PATH as
/opt/csw/X11/lib/pkgconfig

when doing anything that requires CSWx* packages you need to use something
along the lines of: EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib

I think pkg-config is walking down the dependency trail.  cairo is required for
gtk, but xcb is required for cairo. it couldn't find xcb so it quit instead of
spending an hour compiling just to die a horrible death when trying to link to
the missing xcb libraries.


- --

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

iEYEARECAAYFAkpTqYoACgkQLrhmsXMSLxdFZwCdG9doD+tJJ2lqnWdklOWa0dDz
sVMAoIPe9Vp5Fe1w15EFFHOT161OQdHN
=M/Ni
-----END PGP SIGNATURE-----

From william at wbonnet.net  Tue Jul  7 22:03:34 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 07 Jul 2009 22:03:34 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707193555.GU60655@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
Message-ID: <4A53AA16.1000302@wbonnet.net>

Hi

> I hope our cairo/new-x11 maintainres can come up with a more transparent
> solution to this stuff though.
>   
Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH

If you are using GAR add

EXTRA_PKG_CONFIG_PATH += /opt/csw/X11/lib/pkgconfig

to your Makefile

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  Tue Jul  7 22:22:54 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 13:22:54 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4A53A98A.8060201@opencsw.org>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
Message-ID: <20090707202254.GV60655@bolthole.com>

On Tue, Jul 07, 2009 at 03:01:14PM -0500, Mike Watters wrote:
> > Which package needs the bug report?  Gtk+-2.0, cairo, xcb or pkgconfig?
> 
> xcb and in fact all the "new" xlib packages have their PKG_CONFIG_PATH as
> /opt/csw/X11/lib/pkgconfig
> 
> when doing anything that requires CSWx* packages you need to use something
> along the lines of: EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib

idealy though... our cairo package should "do the right thing" somehow, in
my opinion.
after all, that's what pkg-config is for.. to take all the by-hand
messiness out of compiling?

I dont know much about pkg-config, but might there be some kind of,
"and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can put in
/opt/csw/lib/pkgconfig/cairo.pc  or something?

or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
should not exist there in the first place, but be moved
to /opt/csw/lib/X11/lib/pkgconfig/   instead?

I dunno.

From william at wbonnet.net  Wed Jul  8 01:00:10 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 01:00:10 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
Message-ID: <4A53D37A.40502@wbonnet.net>

Hi

I do have the same problem with the /opt/csw/include/jri.h file. It 
comes from the same packages and also says it has to be provided by 
netscape.

Could someone please rename it to .old on build9x ?

thanks in advance

cheers
W.

> Hi William,
>
> Am 01.07.2009 um 14:32 schrieb William Bonnet:
>> A conflicting header is installed by qt package. It is the file 
>> "/opt/csw/include/npapi.h"
>>
>> When including this file, it generates the following error
>>
>> "/opt/csw/include/npapi.h", line 1: Error: #error 
>> "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
>> must be provided by Netscape".
>>
>> A mantis entry exists, but unfortnatly package has no maintainer and 
>> was not fixed. Would it be possible to nuke this file on the build 
>> farm please ? (well at least rename the file to .old...). This file 
>> is blocking for compile FF3.
>>
>> npapi.h is provided by mozilla packages, especialy firefox, not by qt.
>>
>> [wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$ 
>> grep npapi.h /var/sadm/install/contents
>> /opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908 CSWqt
>> /opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273 
>> 1110647274 CSWqtgcc
>
> The file does not include anything of interest:
>
> dam at build9x :/home/dam > more /opt/csw/include/npapi.h
> #error 
> "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
> must be provided by Netscape"
> #error "This file is just a placeholder. Please see the documentation"
> #error "to learn how to obtain the real file"
>
> Moved to npapi.h.old on build9x only for now.
>
>
> Best regards
>
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>
>


-- 
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 Jul  8 02:37:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 20:37:15 -0400
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A53D37A.40502@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
	<4A53D37A.40502@wbonnet.net>
Message-ID: <1247013402-sup-5213@ntdws12.chass.utoronto.ca>

Excerpts from William Bonnet's message of Tue Jul 07 19:00:10 -0400 2009:
> I do have the same problem with the /opt/csw/include/jri.h file. It 
> comes from the same packages and also says it has to be provided by 
> netscape.
> 
> Could someone please rename it to .old on build9x ?

Doing this now.
-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/d57094d6/attachment.asc>

From jeff at cjsa.com  Wed Jul  8 08:49:56 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Wed, 8 Jul 2009 06:49:56 GMT
Subject: [csw-maintainers] Library problem?
References: <KMEG9x.Cxz@cjsa.com> <4A535666.4030009@opencsw.org>
	<20090707142633.GC11549@bolthole.com>
Message-ID: <KMGAz8.2Ir@cjsa.com>

Philip Brown <phil at bolthole.com> writes:

>On Tue, Jul 07, 2009 at 09:06:30AM -0500, Mike Watters wrote:
>> if you are running 1.24 you need to run
>> /opt/csw/bin/sparcv8/pango-querymodules >/opt/csw/etc/pango/pango.modules
>> 
>> for an earlier version you should run
>> /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules
>> 
>> Please let us know if this does not fix the issue.

>It would seem as though our existing users might be similarly plagued with
>this sort of thing if they upgraded recently.

>if so, this very nasty situation, warrants an email about it, to the
>announce and user lists.

Thanks to everyone who responded to my query.  Yes, apparently the problem
was with the pango.modules file and after running the above command (the
binaries under sparc and sparcv8 are the same) my problem was resolved.

I am using the latest 1.24.3,REV=2009.07.01 version of libpango and I
typically update once a week or so.  I do not know exactly when the last
update was to libpango on my system.  Does this look like a general bug in
the post-install script that failed to automatically regenerate the
module list?

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040


From bonivart at opencsw.org  Wed Jul  8 10:33:37 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Jul 2009 10:33:37 +0200
Subject: [csw-maintainers] /testing BIND 9.6.1
Message-ID: <625385e30907080133i54dbf31cn45ba86da04773252@mail.gmail.com>

Latest BIND DNS Server in testing:

http://mirror.opencsw.org/testing/bind-9.6.1,REV=2009.07.07-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind-9.6.1,REV=2009.07.07-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind_utils-9.6.1,REV=2009.07.07-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind_utils-9.6.1,REV=2009.07.07-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libbind-9.6.1,REV=2009.07.07-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/libbind-9.6.1,REV=2009.07.07-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind_devel-9.6.1,REV=2009.07.07-SunOS5.8-all-CSW.pkg.gz

# pkgutil -t http://mirror.opencsw.org/opencsw/testing -i bind

-- 
/peter

From james at opencsw.org  Wed Jul  8 10:45:58 2009
From: james at opencsw.org (James Lee)
Date: Wed, 08 Jul 2009 08:45:58 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4A53AA16.1000302@wbonnet.net>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<4A53AA16.1000302@wbonnet.net>
Message-ID: <20090708.8455800.3441876416@gyor.oxdrove.co.uk>

On 07/07/09, 21:03:34, William Bonnet <william at wbonnet.net> wrote regarding
Re: [csw-maintainers] "pkgconfig --exist ..." is telling porkies:

> Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH

Yes, that helps, thanks.  My blind spot was the .pc files are in odd
places and that it needs things it really doesn't.

The CSW pkgconf should know where to look for the CSW pc files, ie
have this path coded in as the default.  The point of pkgconf is to
tell us paths, if I have to tell it the paths before it works it
defeats its point.  Doing a scan I see the CSW pc files are all over
the place.  It's a mess.



James.

From james at opencsw.org  Wed Jul  8 11:22:48 2009
From: james at opencsw.org (James Lee)
Date: Wed, 08 Jul 2009 09:22:48 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4A53AA16.1000302@wbonnet.net>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<4A53AA16.1000302@wbonnet.net>
Message-ID: <20090708.9224800.775122963@gyor.oxdrove.co.uk>

On 07/07/09, 21:03:34, William Bonnet <william at wbonnet.net> wrote regarding 
Re: [csw-maintainers] "pkgconfig --exist ..." is telling porkies:

> > I hope our cairo/new-x11 maintainres can come up with a more transparent
> > solution to this stuff though.
> >
> Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH

There is another problem with the CSW system, eg:

$ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
sh: gnome-config: not found
Package gobject-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gobject-2.0.pc'
to the PKG_CONFIG_PATH environment variable
Package 'gobject-2.0', required by 'GdkPixbuf', not found



gtk+-2.0.pc is part of CSWgtk2.  gobject-2.0.pc is needed by CSWgtk2's
gtk+-2.0.pc and is part of CSWglib2devel but it's not a depend so I
don't have it installed.  I used to have gobject-2.0.pc as part of
CSWglib2 but it was moved to CSWglib2devel and not added CSWglib2 as
a depend.  In the past when we have split packages we have added the
parts to the main package to ensure continuity.




James.





From william at wbonnet.net  Wed Jul  8 13:31:11 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 13:31:11 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <KMGAz8.2Ir@cjsa.com>
References: <KMEG9x.Cxz@cjsa.com>
	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>
	<KMGAz8.2Ir@cjsa.com>
Message-ID: <4A54837F.9090202@wbonnet.net>

Hi

> Thanks to everyone who responded to my query.  Yes, apparently the problem
> was with the pango.modules file and after running the above command (the
> binaries under sparc and sparcv8 are the same) my problem was resolved.
>
> I am using the latest 1.24.3,REV=2009.07.01 version of libpango and I
> typically update once a week or so.  I do not know exactly when the last
> update was to libpango on my system.  Does this look like a general bug in
> the post-install script that failed to automatically regenerate the
> module list?
>   
I was able to re produce the same problem on a 64 bit VM and thunderbird.

Is it possible that the problem comes from the fact a 32 bit application 
(thunderbird) is running on a 64 bit machine, when loading pango modules 
it retrieves the 64 bits modules from the pango.modules config file and 
this produces a warning / error.

If this is the case i don't really see how having two different version 
of the executable used to generate this file may solve this.

Any clue is welcome :)

cheers
W.

From mwatters at opencsw.org  Wed Jul  8 15:29:27 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 08:29:27 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A54837F.9090202@wbonnet.net>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<KMGAz8.2Ir@cjsa.com>
	<4A54837F.9090202@wbonnet.net>
Message-ID: <4A549F37.50701@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi
> 
>> Thanks to everyone who responded to my query.  Yes, apparently the
>> problem
>> was with the pango.modules file and after running the above command (the
>> binaries under sparc and sparcv8 are the same) my problem was resolved.
>>
>> I am using the latest 1.24.3,REV=2009.07.01 version of libpango and I
>> typically update once a week or so.  I do not know exactly when the last
>> update was to libpango on my system.  Does this look like a general
>> bug in
>> the post-install script that failed to automatically regenerate the
>> module list?
>>   
> I was able to re produce the same problem on a 64 bit VM and thunderbird.
> 
> Is it possible that the problem comes from the fact a 32 bit application
> (thunderbird) is running on a 64 bit machine, when loading pango modules
> it retrieves the 64 bits modules from the pango.modules config file and
> this produces a warning / error.
> 
> If this is the case i don't really see how having two different version
> of the executable used to generate this file may solve this.
> 
> Any clue is welcome :)
> 
> cheers
> W.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

This unfortunately is a known bug in this version
http://www.opencsw.org/mantis/view.php?id=3750

I think I saw some threads bouncing back and forth with regard to the best way
to correct this issue.  the ticket above has a temporary fix in the notes from
Dago.


- --

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

iEYEARECAAYFAkpUnzcACgkQLrhmsXMSLxfBOQCffkTbICPvjRnJnb9Vs4l0VAUh
yyMAn0IT+HSDm/VmcoonfTkBjvtkyCTw
=kE7U
-----END PGP SIGNATURE-----

From maciej at opencsw.org  Wed Jul  8 18:55:21 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 8 Jul 2009 17:55:21 +0100
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
Message-ID: <ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>

On Wed, Jun 24, 2009 at 8:36 PM, Sebastian Kayser<skayser at opencsw.org> wrote:
> I guess it doesn't fully cover your answer, but regarding GAR, you can set
> GARFLAVOR to DBG if you want your binaries built with debugging symbols.
> This will also skip the step where GAR strips all the binaries. See [1].

cups makefiles do honor CFLAGS, so GARFLAVOR = DBG, I can see that the
flag -g is present in compiler invocations.

Using ALL_CFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include -I/usr/include/kerberosv5 -g
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include  -DOPENSSL_DISABLE_OLD_DES_SUPPORT
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
-D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
Using ALL_CXXFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include -g
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include -DOPENSSL_DISABLE_OLD_DES_SUPPORT
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
-D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
Using LDFLAGS=-L../cups -L../filter
-L/export/home/blizinski/build.5.10-i386/opt/csw/lib -L/opt/csw/lib
-R/opt/csw/lib   -w -KPIC -xO2
Using LIBS=-lcups -lssl -lcrypto -lpthread -lresolv -lnsl -lsocket -lm

The binary is in fact not stripped:

$ file cups-polld
cups-polld:     ELF 32-bit LSB executable 80386 Version 1 [FPU],
dynamically linked, not stripped

However, it doesn't seem to be compiled without optimization: when I
run it under the Sun Studio debugger, and display the call stack, I
see:

main(argc = ???, argv = ???) (optimized)

Is there a way to check whether the binary is actually optimized
during compilation?

Maciej

From mwatters at opencsw.org  Wed Jul  8 19:11:04 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 12:11:04 -0500
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
Message-ID: <4A54D328.40709@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Maciej (Matchek) Blizinski wrote:
> On Wed, Jun 24, 2009 at 8:36 PM, Sebastian Kayser<skayser at opencsw.org> wrote:
>> I guess it doesn't fully cover your answer, but regarding GAR, you can set
>> GARFLAVOR to DBG if you want your binaries built with debugging symbols.
>> This will also skip the step where GAR strips all the binaries. See [1].
> 
> cups makefiles do honor CFLAGS, so GARFLAVOR = DBG, I can see that the
> flag -g is present in compiler invocations.
> 
> Using ALL_CFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include -I/usr/include/kerberosv5 -g
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include  -DOPENSSL_DISABLE_OLD_DES_SUPPORT
> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
> Using ALL_CXXFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include -g
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include -DOPENSSL_DISABLE_OLD_DES_SUPPORT
> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
> Using LDFLAGS=-L../cups -L../filter
> -L/export/home/blizinski/build.5.10-i386/opt/csw/lib -L/opt/csw/lib
> -R/opt/csw/lib   -w -KPIC -xO2
                             ^^^
the -xO2 is the optimization flag here... remove it and you should remove the
optimization.

> Using LIBS=-lcups -lssl -lcrypto -lpthread -lresolv -lnsl -lsocket -lm



- --

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

iEYEARECAAYFAkpU0ygACgkQLrhmsXMSLxeTbQCaAyUg6LtZinw//Q2UIJWSE/9M
LLkAn2fQ5razR8Pd9oRCYERfxdAFb/nf
=C1NP
-----END PGP SIGNATURE-----

From skayser at opencsw.org  Wed Jul  8 19:28:56 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 8 Jul 2009 19:28:56 +0200 (CEST)
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <4A54D328.40709@opencsw.org>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
Message-ID: <61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>

Mike Watters wrote:
> Maciej (Matchek) Blizinski wrote:
>> On Wed, Jun 24, 2009 at 8:36 PM, Sebastian Kayser<skayser at opencsw.org>
>> wrote:
>>> I guess it doesn't fully cover your answer, but regarding GAR, you can
>>> set
>>> GARFLAVOR to DBG if you want your binaries built with debugging
>>> symbols.
>>> This will also skip the step where GAR strips all the binaries. See
>>> [1].
>>
>> cups makefiles do honor CFLAGS, so GARFLAVOR = DBG, I can see that the
>> flag -g is present in compiler invocations.
>>
>> Using ALL_CFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include -I/usr/include/kerberosv5 -g
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include  -DOPENSSL_DISABLE_OLD_DES_SUPPORT
>> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
>> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
>> Using ALL_CXXFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include -g
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include -DOPENSSL_DISABLE_OLD_DES_SUPPORT
>> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
>> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
>> Using LDFLAGS=-L../cups -L../filter
>> -L/export/home/blizinski/build.5.10-i386/opt/csw/lib -L/opt/csw/lib
>> -R/opt/csw/lib   -w -KPIC -xO2
>                              ^^^
> the -xO2 is the optimization flag here... remove it and you should remove
> the
> optimization.

And just for the sake of completeness: GAR doesn't add such optimization
flags when you set GARFLAVOR=DBG. Just give

  GARFLAVOR=DBG gmake modenv

a try and GAR will tell you what flags it will use. This doesn't mean that
GAR will filter out such flags for you when you have set them manually
elsewhere, but it won't produce them on its own.

Sebastian


From william at wbonnet.net  Wed Jul  8 19:41:25 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 19:41:25 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <1247013402-sup-5213@ntdws12.chass.utoronto.ca>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>
	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
Message-ID: <4A54DA45.4090902@wbonnet.net>

Hi Ben
> Excerpts from William Bonnet's message of Tue Jul 07 19:00:10 -0400 2009:
>   
>> I do have the same problem with the /opt/csw/include/jri.h file. It 
>> comes from the same packages and also says it has to be provided by 
>> netscape.
>>
>> Could someone please rename it to .old on build9x ?
>>     
>
> Doing this now.
>   
Thanks

It worked, i was able to compile FF3 on solaris 9, and the binaries runs on solaris 10. Now i have to find tests machines running solaris 8 and 9 :)

I also need the same fix on build9x 8x and 9x. I better look to produce a Qt patch :)


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  Wed Jul  8 22:04:08 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 22:04:08 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <20090707142633.GC11549@bolthole.com>
References: <KMEG9x.Cxz@cjsa.com> <4A535666.4030009@opencsw.org>
	<20090707142633.GC11549@bolthole.com>
Message-ID: <4A54FBB8.4010808@wbonnet.net>

Hi

> It would seem as though our existing users might be similarly plagued with
> this sort of thing if they upgraded recently.
>
> if so, this very nasty situation, warrants an email about it, to the
> announce and user lists.
>   
Based on Trygve advices i made a few tests and had a look to the way 
sun's gnome desktop is working.  I run a Ultra 20, thus an amd64 CPU. In 
the Solaris installation there is both 32 and 64 bits versions of pango. 
Here are the files i have :

64 bits version

[william at cyaegha:/usr/lib/amd64]$ ls *pango*
libpango-1.0.so          libpangoft2-1.0.so          
libpangox-1.0.so          libpangoxft-1.0.so
libpango-1.0.so.0        libpangoft2-1.0.so.0        
libpangox-1.0.so.0        libpangoxft-1.0.so.0
libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1  
libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1

pango:
1.4.0

[william at cyaegha:/usr/lib/amd64]$ ls pango/1.4.0/modules/
pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so  
pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so

32 bits version

[william at cyaegha:/usr/lib]$ ls *pango*
libpango-1.0.so          libpangoft2-1.0.so          
libpangox-1.0.so          libpangoxft-1.0.so
libpango-1.0.so.0        libpangoft2-1.0.so.0        
libpangox-1.0.so.0        libpangoxft-1.0.so.0
libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1  
libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1

pango:
1.4.0

[william at cyaegha:/usr/lib]$ ls pango/1.4.0/modules/
pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so  
pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so

By default /etc/pango/pango.modules contains the 32 bits modules, even 
on my amd64 version.

I made the following tests running both 32 bits and 64 bits version of 
gtk-demo (/usr/bin/gtk-demo and /usr/bin/amd64/gtk-demo).

. Running application gtk-demo and pango.modules using same versions 
(both 32 or both 64 bits) work is fine.

. Running 64bits apps with 32 bits modules works also, but running 32 
bits app with 64 bits modules does not work.

Please notice that from Sun's isntallation running both 64bits, or 
64bits apps with 32bits modules produce the same warnings

     (gtk-demo:19058): Gtk-WARNING **: Unable to locate theme engine in 
module_path: "blueprint",

Running a full 32bits stack produce no warnings



So i may suggest to do some more testing, like recompiling a couple of 
our desktop app in 64 bits mode, and test these apps against 32 bits 
modules. If it work fine, we may generate by default a pango.module 
config file containing 32 bits modules.

Because :

1/ it works :)

2/ most of the desktop app are compiled only in 32 bits

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  Wed Jul  8 23:11:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 16:11:50 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A54FBB8.4010808@wbonnet.net>
References: <KMEG9x.Cxz@cjsa.com>
	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>
	<4A54FBB8.4010808@wbonnet.net>
Message-ID: <4A550B96.9050604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi
> 
>> It would seem as though our existing users might be similarly plagued
>> with
>> this sort of thing if they upgraded recently.
>>
>> if so, this very nasty situation, warrants an email about it, to the
>> announce and user lists.
>>   
> Based on Trygve advices i made a few tests and had a look to the way
> sun's gnome desktop is working.  I run a Ultra 20, thus an amd64 CPU. In
> the Solaris installation there is both 32 and 64 bits versions of pango.
> Here are the files i have :
> 
> 64 bits version
> 
> [william at cyaegha:/usr/lib/amd64]$ ls *pango*
> libpango-1.0.so          libpangoft2-1.0.so         
> libpangox-1.0.so          libpangoxft-1.0.so
> libpango-1.0.so.0        libpangoft2-1.0.so.0       
> libpangox-1.0.so.0        libpangoxft-1.0.so.0
> libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1 
> libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1
> 
> pango:
> 1.4.0
> 
> [william at cyaegha:/usr/lib/amd64]$ ls pango/1.4.0/modules/
> pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so 
> pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so
> 
> 32 bits version
> 
> [william at cyaegha:/usr/lib]$ ls *pango*
> libpango-1.0.so          libpangoft2-1.0.so         
> libpangox-1.0.so          libpangoxft-1.0.so
> libpango-1.0.so.0        libpangoft2-1.0.so.0       
> libpangox-1.0.so.0        libpangoxft-1.0.so.0
> libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1 
> libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1
> 
> pango:
> 1.4.0
> 
> [william at cyaegha:/usr/lib]$ ls pango/1.4.0/modules/
> pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so 
> pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so
> 
> By default /etc/pango/pango.modules contains the 32 bits modules, even
> on my amd64 version.
> 
> I made the following tests running both 32 bits and 64 bits version of
> gtk-demo (/usr/bin/gtk-demo and /usr/bin/amd64/gtk-demo).
> 
> . Running application gtk-demo and pango.modules using same versions
> (both 32 or both 64 bits) work is fine.
> 
> . Running 64bits apps with 32 bits modules works also, but running 32
> bits app with 64 bits modules does not work.
> 
> Please notice that from Sun's isntallation running both 64bits, or
> 64bits apps with 32bits modules produce the same warnings
> 
>     (gtk-demo:19058): Gtk-WARNING **: Unable to locate theme engine in
> module_path: "blueprint",
> 
> Running a full 32bits stack produce no warnings
> 
> 
> 
> So i may suggest to do some more testing, like recompiling a couple of
> our desktop app in 64 bits mode, and test these apps against 32 bits
> modules. If it work fine, we may generate by default a pango.module
> config file containing 32 bits modules.
> 
> Because :
> 
> 1/ it works :)
> 
> 2/ most of the desktop app are compiled only in 32 bits
> 
> cheers
> W.
> 
> 

I believe the running proposal is to remove the isaexec from pango, and have
the default pango.modules file containing the 32bit.

there was some discussion on having a /opt/csw/etc/64/pango.modules file
which would be fine with me, AFAIK there is currently nothing in CSW that uses
the 64bit pango libs. ( as they never existed before ) so we can patch the
64bit versions to use the correct modules file.

I believe I saw some messages that Dago is on Holiday?


- --

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

iEYEARECAAYFAkpVC5YACgkQLrhmsXMSLxftaQCfccWs8VOp5ZFYtdz/Y6r6CeSV
xe8An0qHVjDu+T8t1lBKs7q1lES9IfhT
=FEmQ
-----END PGP SIGNATURE-----

From william at wbonnet.net  Wed Jul  8 23:23:57 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 23:23:57 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A550B96.9050604@opencsw.org>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<4A54FBB8.4010808@wbonnet.net>
	<4A550B96.9050604@opencsw.org>
Message-ID: <4A550E6D.70600@wbonnet.net>

Hi Mike

> I believe I saw some messages that Dago is on Holiday?
>   
And without internet access... poor Dago, it should be really difficult ;)

I think he will be back tomorrow or friday.

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  Wed Jul  8 23:28:11 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 16:28:11 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A550E6D.70600@wbonnet.net>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<4A54FBB8.4010808@wbonnet.net>	<4A550B96.9050604@opencsw.org>
	<4A550E6D.70600@wbonnet.net>
Message-ID: <4A550F6B.6040508@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi Mike
> 
>> I believe I saw some messages that Dago is on Holiday?
>>   
> And without internet access... poor Dago, it should be really difficult ;)
> 
> I think he will be back tomorrow or friday.
> 
> cheers
> W.
> 

How can one live without internet access?


- --

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

iEYEARECAAYFAkpVD2sACgkQLrhmsXMSLxemYwCgkVDasQfAdrbZaBk6EvfI/RAX
mXgAoJMo+6f5IgCJLRfhcZa6TYBGWQl1
=r/60
-----END PGP SIGNATURE-----

From maciej at opencsw.org  Thu Jul  9 12:22:18 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 11:22:18 +0100
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
Message-ID: <ce97543b0907090322j3b6463a7ye57f4d9d5165a0ba@mail.gmail.com>

On Sat, Jul 4, 2009 at 6:01 PM, rupert THURNER<rupert at opencsw.org> wrote:
> when doing the mercurial tests, there is a warning.
>
> gpg: Warning: using insecure memory!
>  --> http://www.gnupg.org/faq.html#q6.1
>
> the faq says that it should be installed with setuid root. is this
> deliberately different on opencsw?

Routing to Mike Watters, the gnupg package maintainer.

It would be a relatively simple change:

Index: Makefile
===================================================================
--- Makefile    (revision 5052)
+++ Makefile    (working copy)
@@ -40,4 +40,8 @@

 TEST_TARGET = check

+PROTOTYPE_FILTER  = awk ' \
+    $$$$3 ~ /\/opt\/csw\/bin\/gpg$$$$/ { $$$$4 = "4755" } \
+    { print }'
+
 include gar/category.mk

Maciej

From maciej at opencsw.org  Thu Jul  9 12:23:02 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 11:23:02 +0100
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <ce97543b0907090322j3b6463a7ye57f4d9d5165a0ba@mail.gmail.com>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
	<ce97543b0907090322j3b6463a7ye57f4d9d5165a0ba@mail.gmail.com>
Message-ID: <ce97543b0907090323l64483774o8c229578978af70f@mail.gmail.com>

[+mwatters]

On Thu, Jul 9, 2009 at 11:22 AM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> On Sat, Jul 4, 2009 at 6:01 PM, rupert THURNER<rupert at opencsw.org> wrote:
>> when doing the mercurial tests, there is a warning.
>>
>> gpg: Warning: using insecure memory!
>> ?--> http://www.gnupg.org/faq.html#q6.1
>>
>> the faq says that it should be installed with setuid root. is this
>> deliberately different on opencsw?
>
> Routing to Mike Watters, the gnupg package maintainer.
>
> It would be a relatively simple change:
>
> Index: Makefile
> ===================================================================
> --- Makefile ? ?(revision 5052)
> +++ Makefile ? ?(working copy)
> @@ -40,4 +40,8 @@
>
> ?TEST_TARGET = check
>
> +PROTOTYPE_FILTER ?= awk ' \
> + ? ?$$$$3 ~ /\/opt\/csw\/bin\/gpg$$$$/ { $$$$4 = "4755" } \
> + ? ?{ print }'
> +
> ?include gar/category.mk
>
> Maciej
>

From pfelecan at opencsw.org  Thu Jul  9 14:34:11 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 09 Jul 2009 14:34:11 +0200
Subject: [csw-maintainers] common package in the catalog
Message-ID: <x68wiy2fy4.fsf@abhavana.free.fr>

I'm observing something unusual, at least for me, in the catalog of
SPARC systems: the corresponding entry points toward an uncompressed
file:

on a SPARC system:
common 1.4.6,REV=2008.04.28 CSWcommon
common-1.4.6,REV=2008.04.28-SunOS5.8-sparc-CSW.pkg
d0e61f7542b3a4fbff7de4a7437bb6af 23552 none none

on an Intel system:
common 1.4.6,REV=2008.04.28 CSWcommon
common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
638cac6d1d6d1dba3db8ab76fbd36bb2 3632 none none

This is a new thing as some weeks ago both catalogs referenced the
common package as compressed.
-- 
Peter

From bonivart at opencsw.org  Thu Jul  9 15:00:33 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 9 Jul 2009 15:00:33 +0200
Subject: [csw-maintainers] common package in the catalog
In-Reply-To: <x68wiy2fy4.fsf@abhavana.free.fr>
References: <x68wiy2fy4.fsf@abhavana.free.fr>
Message-ID: <625385e30907090600r72a984ccrfbfa20ec139aa45b@mail.gmail.com>

On Thu, Jul 9, 2009 at 2:34 PM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> I'm observing something unusual, at least for me, in the catalog of
> SPARC systems: the corresponding entry points toward an uncompressed
> file:
>
> on a SPARC system:
> common 1.4.6,REV=2008.04.28 CSWcommon
> common-1.4.6,REV=2008.04.28-SunOS5.8-sparc-CSW.pkg
> d0e61f7542b3a4fbff7de4a7437bb6af 23552 none none
>
> on an Intel system:
> common 1.4.6,REV=2008.04.28 CSWcommon
> common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> 638cac6d1d6d1dba3db8ab76fbd36bb2 3632 none none
>
> This is a new thing as some weeks ago both catalogs referenced the
> common package as compressed.

This is a bug, both should be uncompressed since CSWgzip depends on
common. This should have been taken care of a while back but maybe
only for sparc..?

-- 
/peter

From Darin.Perusich at cognigencorp.com  Thu Jul  9 17:37:03 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 09 Jul 2009 11:37:03 -0400
Subject: [csw-maintainers] setting apache2 cmdline options
Message-ID: <4A560E9F.8000800@cognigencorp.com>

This probably belongs on the users list but I figure I'll pose my
question here instead. How are people going about setting command line
options for apache2 httpd processes? If I specify <IfDefine name>
directives in my httpd.conf how would one go about setting -Dname on the
command line without modifying the manifest? You can't.

I'm asking because I was setting up an SSL server and I'm using the
certificate for multiple apps, stunnel and apache, and I've placed it in
/opt/csw/etc/ssl/certs/ the global location for certificates and when I
started apache it didn't load mod_ssl. After digging around for
documentation on this, which is non-existent, I looked at the manifest
and it only loads -DSSL if $APACHE_HOME/etc/server.crt exists. Well,
what if that isn't were my certificate lives or is named? Have a
configuration file that can be sourced to set command line options would
be ideal and is really the only solution I can see.

Yes I know I can remove the <IfDefine SSL> or touch server.crt but those
are band-aids.

What you peoples thoughts on this?

-- 
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  Thu Jul  9 17:45:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 08:45:45 -0700
Subject: [csw-maintainers] common package in the catalog
In-Reply-To: <625385e30907090600r72a984ccrfbfa20ec139aa45b@mail.gmail.com>
References: <x68wiy2fy4.fsf@abhavana.free.fr>
	<625385e30907090600r72a984ccrfbfa20ec139aa45b@mail.gmail.com>
Message-ID: <20090709154545.GA75736@bolthole.com>

On Thu, Jul 09, 2009 at 03:00:33PM +0200, Peter Bonivart wrote:
> This is a bug, both should be uncompressed since CSWgzip depends on
> common. This should have been taken care of a while back but maybe
> only for sparc..?

oops. guess so.
pushing out unzippped now

From maciej at opencsw.org  Thu Jul  9 17:53:48 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 16:53:48 +0100
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
	<61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
Message-ID: <ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>

One more question: does anyone know how to debug a number of binaries
at once? In my case, there's a suspicion that the problem has
something to do with the multitude of cups-polld daemons. I can't
trigger the issue with just one daemon; I need to run several at the
same time. So far, Sun Studio gives me an impression that it can only
debug one binary at the time. Perhaps there's a way to launch a number
of binaries and attach to one binary at the time?

Maciej

From maciej at opencsw.org  Thu Jul  9 17:55:03 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 16:55:03 +0100
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
	<61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
	<ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
Message-ID: <ce97543b0907090855g22887b29p5b97251c6d1deaa@mail.gmail.com>

On Thu, Jul 9, 2009 at 4:53 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> same time. So far, Sun Studio gives me an impression that it can only
> debug one binary at the time. Perhaps there's a way to launch a number
> of binaries and attach to one binary at the time?

A terminology cleanup: I meant, a number of processes, the same
binary, multiple times, with different options.

Maciej

From phil at bolthole.com  Thu Jul  9 17:59:22 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 08:59:22 -0700
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
	<61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
	<ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
Message-ID: <20090709155922.GA8843@bolthole.com>

On Thu, Jul 09, 2009 at 04:53:48PM +0100, Maciej (Matchek) Blizinski wrote:
> One more question: does anyone know how to debug a number of binaries
> at once? In my case, there's a suspicion that the problem has
> something to do with the multitude of cups-polld daemons. I can't
> trigger the issue with just one daemon; I need to run several at the
> same time. So far, Sun Studio gives me an impression that it can only
> debug one binary at the time. Perhaps there's a way to launch a number
> of binaries and attach to one binary at the time?

use multiple windows, and multiple dbx processes? :-}

From phil at bolthole.com  Thu Jul  9 18:04:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 09:04:50 -0700
Subject: [csw-maintainers] setting apache2 cmdline options
In-Reply-To: <4A560E9F.8000800@cognigencorp.com>
References: <4A560E9F.8000800@cognigencorp.com>
Message-ID: <20090709160450.GB8843@bolthole.com>

On Thu, Jul 09, 2009 at 11:37:03AM -0400, Darin Perusich wrote:
>...
>  After digging around for
> documentation on this, which is non-existent, I looked at the manifest
> and it only loads -DSSL if $APACHE_HOME/etc/server.crt exists. Well,
> what if that isn't were my certificate lives or is named?

then make a symlink from there to the real location?

> Yes I know I can remove the <IfDefine SSL> or touch server.crt but those
> are band-aids.
> 
> What you peoples thoughts on this?

I think that using a package made and put together by someone else,
sometimes means that you need to follow someone else's conventions.
(ie: making the symlink).

and if you think there are better conventions for the maintainer to follow,
you should suggest it to them.

As far as #defines go for apache... I dont think they are meant to be used
like makefile defines, whereeever and however you want. They're mostly just
to be used as a convenience for if a module is defined, in my opinion.
The SSL thing is an oddity, and I've never liked how they have done that.
I think it should be handled more like the others. ie:

<IfModule ssl>

rather than
<IFDefine SSL>

but in the absense of that.. you need some external trigger to let the
startup script know, "hey, i want ssl enabled". and the presense of the
server key file, seems like a reasonable trigger to me.

From dam at opencsw.org  Thu Jul  9 20:56:05 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 20:56:05 +0200
Subject: [csw-maintainers] =?iso-8859-1?q?Additional_Farm_Maintainer=3A_J?=
	=?iso-8859-1?q?=FCrgen_Arndt?=
Message-ID: <06DF27E0-FF15-455A-9040-5A5C5160F8AC@opencsw.org>

Hi,

I'm back from my vacation and noticed that there are obviously not
enough farm managers as Ben wanted to take some time off, too. I have
given J?rgen full admin access to the farm. Ihsan, please add "ja"
to the list of buildfarm maintainers.


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul  9 21:49:38 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 21:49:38 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <4A4F8D65.9080203@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
Message-ID: <EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>

Hi Sebastian,

Am 04.07.2009 um 19:12 schrieb Sebastian Kayser:
> Dagobert Michelsen wrote:
>> Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
>>> while working on postfix 2.6 i have noticed again that it is not  
>>> quite
>>> straight-forward to assign files to classes. Instead of exposing the
>>> inner
>>> workings of the prototype stuff to the Makefile, could we maybe get
>>> some
>>> variables to assign files to classes (functions)?
>>
>> For CSWcswclassutils this is already in there :-) Just define
>>
>> SAMPLECONF
>> PRESERVECONF
>> INITSMF
>>
>> and everything else will be taken cared of.
>
> Sweet!!! :D Can we add something for USERGROUP also?
>
> Index: /home/skayser/mgar/gar/v2/gar.pkg.mk
> ===================================================================
> --- /home/skayser/mgar/gar/v2/gar.pkg.mk        (revision 5361)
> +++ /home/skayser/mgar/gar/v2/gar.pkg.mk        (working copy)
> @@ -155,12 +155,14 @@
> SPKG_CLASSES := $(SPKG_CLASSES) $(if $(SAMPLECONF),$(if $(filter  
> cswsampleconf,$(SPKG_CLASSES)),,cswsampleconf))
> SPKG_CLASSES := $(SPKG_CLASSES) $(if $(PRESERVECONF),$(if $(filter  
> cswpreserveconf,$(SPKG_CLASSES)),,cswpreserveconf))
> SPKG_CLASSES := $(SPKG_CLASSES) $(if $(INITSMF),$(if $(filter  
> cswinitsmf,$(SPKG_CLASSES)),,cswinitsmf))
> +SPKG_CLASSES := $(SPKG_CLASSES) $(if $(USERGROUP),$(if $(filter  
> cswusergroup,$(SPKG_CLASSES)),,cswusergroup))
> # - set class for all config files
> ifneq ($(SAMPLECONF)$(PRESERVECONF)$(INITSMF),)
> _CSWCLASS_FILTER = | perl -ane '\
>                $(foreach FILE,$(SAMPLECONF),$$F[1] = "cswsampleconf"  
> if ( $$F[2] =~ m(^$(FILE)$$) );)\
>                $(foreach FILE,$(PRESERVECONF),$$F[1] =  
> "cswpreserveconf" if( $$F[2] =~ m(^$(FILE)$$) );)\
>                $(foreach FILE,$(INITSMF),$$F[1] = "cswinitsmf" if( $ 
> $F[2] =~ m(^$(FILE)$$) );)\
> +               $(foreach FILE,$(USERGROUP),$$F[1] = "cswusergroup"  
> if( $$F[2] =~ m(^$(FILE)$$) );)\
>                print join(" ", at F),"\n";'
> _EXTRA_GAR_PKGS += CSWcswclassutils
> endif

Looks good. Please go ahead and commit.


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul  9 22:51:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 22:51:15 +0200
Subject: [csw-maintainers] Newpkgs after testing/: pango
Message-ID: <9770FEF1-E44C-4547-A649-AFE85BCDAE30@opencsw.org>

Hi,

there is an updated version of pango in testing/ and newpkgs/:

libpango-1.24.3,REV=2009.07.09-SunOS5.8-i386-CSW.pkg.gz
libpango-1.24.3,REV=2009.07.09-SunOS5.8-sparc-CSW.pkg.gz
libpango_devel-1.24.3,REV=2009.07.09-SunOS5.8-i386-CSW.pkg.gz
libpango_devel-1.24.3,REV=2009.07.09-SunOS5.8-sparc-CSW.pkg.gz
libpango_doc-1.24.3,REV=2009.07.09-SunOS5.8-all-CSW.pkg.gz

The query-bindary is no longer isaexec and there two files
produced now during postinstall:
   /opt/csw/etc/pango/pango.modules
   /opt/csw/etc/64/pango/pango.modules

Please someone other then me verifies that it fully solves the
issue and Phil can release it ASAP.


Best regards

   -- Dago

From mwatters at opencsw.org  Thu Jul  9 23:13:18 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Jul 2009 16:13:18 -0500
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <ce97543b0907090302l17245204k48f8c6e3fa9b09f3@mail.gmail.com>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
	<ce97543b0907090302l17245204k48f8c6e3fa9b09f3@mail.gmail.com>
Message-ID: <4A565D6E.9050201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Maciej (Matchek) Blizinski wrote:
> On Sat, Jul 4, 2009 at 6:01 PM, rupert THURNER<rupert at opencsw.org> wrote:
>> when doing the mercurial tests, there is a warning.
>>
>> gpg: Warning: using insecure memory!
>>  --> http://www.gnupg.org/faq.html#q6.1
>>
>> the faq says that it should be installed with setuid root. is this
>> deliberately different on opencsw?
> 
> Routing to Mike Watters, the gnupg package maintainer.
> 
> It would be a relatively simple change:
> 
> Index: Makefile
> ===================================================================
> --- Makefile    (revision 5052)
> +++ Makefile    (working copy)
> @@ -40,4 +40,8 @@
> 
>  TEST_TARGET = check
> 
> +PROTOTYPE_FILTER  = awk ' \
> +    $$$$3 ~ /\/opt\/csw\/bin\/gpg$$$$/ { $$$$4 = "4755" } \
> +    { print }'
> +
>  include gar/category.mk
> 
> Maciej

I will make the change the next time I release the package.
please file a feature request bug so I don't forget about it ;-}



- --

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

iEYEARECAAYFAkpWXW4ACgkQLrhmsXMSLxfxlgCgpujVgtqAvOHc7xjteyjZoge7
WB8AniFfPWHADQGF1864eRYWOn28QXbK
=+fcb
-----END PGP SIGNATURE-----

From dam at opencsw.org  Thu Jul  9 23:16:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:16:45 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <KMEG9x.Cxz@cjsa.com>
References: <KMEG9x.Cxz@cjsa.com>
Message-ID: <41C24380-8952-4A5E-BFC9-2AC385F9E9A9@opencsw.org>

Hi Jeffery,

Am 07.07.2009 um 08:49 schrieb Jeffery Small:
> I have just started noticing that something has changed on my system  
> and
> now a number of applications are displaying all text as rectangular  
> boxes.
> The rapidsvn application is one example.  This occurs even though  
> these
> applications have not changed, so the problem must be in a library.   
> My
> best guess is that it might be glib2 or gtk2, but I actually know  
> little
> about them, so this is just a guess.  However, I have another app.  
> that did
> not come from CSW but is linked to these two libraries, so I think  
> they may
> be suspects.  Is anyone else seeing this problem, and if we can  
> identify
> the source, can we please get it fixed ASAP?

I guess that was my fault. Unfortunately I was on holidays just these
few days the error occurred :-( Please see if the version in testing/
from pango fixes the issue:
   http://mirror.opencsw.org/testing.html
After the correct fixing has been confirmed the packages can be pushed
out ASAP.


Thanks!

   -- Dago

From dam at opencsw.org  Thu Jul  9 23:27:35 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:27:35 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707202254.GV60655@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
Message-ID: <4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>

Hi Phil,

Am 07.07.2009 um 22:22 schrieb Philip Brown:
> On Tue, Jul 07, 2009 at 03:01:14PM -0500, Mike Watters wrote:
>>> Which package needs the bug report?  Gtk+-2.0, cairo, xcb or  
>>> pkgconfig?
>>
>> xcb and in fact all the "new" xlib packages have their  
>> PKG_CONFIG_PATH as
>> /opt/csw/X11/lib/pkgconfig
>>
>> when doing anything that requires CSWx* packages you need to use  
>> something
>> along the lines of: EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib
>
> idealy though... our cairo package should "do the right thing"  
> somehow, in
> my opinion.
> after all, that's what pkg-config is for.. to take all the by-hand
> messiness out of compiling?

Yes, but you must tell pkg-config which libraries to use. pkg-config
is for setting flags, not finding the libs.

> I dont know much about pkg-config, but might there be some kind of,
> "and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can  
> put in
> /opt/csw/lib/pkgconfig/cairo.pc  or something?

AFAIK this option is not available.

> or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
> should not exist there in the first place, but be moved
> to /opt/csw/lib/X11/lib/pkgconfig/   instead?

It is just for development. Using pkg-config this way allows specific
binding to either the OpenCSW X11 libs or Sun X11 libs with gtk+ on
compilation. This separation was intended when we introduced /opt/csw/ 
X11.
Is this case, however, it is not useful. As it is only relevant in
compilation and integration is well understood and easy in both manual
and GAR builds it shouldn't do too much harm, though.


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul  9 23:29:23 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:29:23 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090708.9224800.775122963@gyor.oxdrove.co.uk>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<4A53AA16.1000302@wbonnet.net>
	<20090708.9224800.775122963@gyor.oxdrove.co.uk>
Message-ID: <2117C253-063E-4E51-BC5B-64915CE239E9@opencsw.org>

Hi James,

Am 08.07.2009 um 11:22 schrieb James Lee:
> On 07/07/09, 21:03:34, William Bonnet <william at wbonnet.net> wrote  
> regarding
> Re: [csw-maintainers] "pkgconfig --exist ..." is telling porkies:
>
>>> I hope our cairo/new-x11 maintainres can come up with a more  
>>> transparent
>>> solution to this stuff though.
>>>
>> Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH
>
> There is another problem with the CSW system, eg:
>
> $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
> sh: gnome-config: not found
> Package gobject-2.0 was not found in the pkg-config search path.
> Perhaps you should add the directory containing `gobject-2.0.pc'
> to the PKG_CONFIG_PATH environment variable
> Package 'gobject-2.0', required by 'GdkPixbuf', not found
>
> gtk+-2.0.pc is part of CSWgtk2.  gobject-2.0.pc is needed by CSWgtk2's
> gtk+-2.0.pc and is part of CSWglib2devel but it's not a depend so I
> don't have it installed.  I used to have gobject-2.0.pc as part of
> CSWglib2 but it was moved to CSWglib2devel and not added CSWglib2 as
> a depend.  In the past when we have split packages we have added the
> parts to the main package to ensure continuity.

This is a very good point. The devel-packages should not only
rely on the base package, but the devel-packages for the base-
packages. Please file a bug report for these so they are not
forgotten.


Best regards

   -- Dago

From phil at bolthole.com  Thu Jul  9 23:35:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 14:35:25 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
	<4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
Message-ID: <20090709213525.GE8843@bolthole.com>

On Thu, Jul 09, 2009 at 11:27:35PM +0200, Dagobert Michelsen wrote:
> Hi Phil,
>
> Am 07.07.2009 um 22:22 schrieb Philip Brown:
>> I dont know much about pkg-config, but might there be some kind of,
>> "and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can  
>> put in
>> /opt/csw/lib/pkgconfig/cairo.pc  or something?
>
> AFAIK this option is not available.
>

Then in my opinion, since libcairo requires libxcb now, then 
I think my suggestion here, would be best:


>> or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
>> should not exist there in the first place, but be moved
>> to /opt/csw/lib/X11/lib/pkgconfig/   instead?


>[Dago writes in reply to that...]
> It is just for development. Using pkg-config this way allows specific
> binding to either the OpenCSW X11 libs or Sun X11 libs with gtk+ on
> compilation.

I do not think that is a good (ie: stable) idea at all.
If people want to use "sun libs", then they should probably use sun cairo!
on the other hand, seeing as how our libcairo now REQUIRES the newer x11
libs... libs that we were "forced" to ship ourselves... it doesnt make
sense to break smooth opencsw development, for the fuzzy benefit of
lalowing pepole to compile brokenly against sun libs as well.


From dam at opencsw.org  Thu Jul  9 23:38:59 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:38:59 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A550F6B.6040508@opencsw.org>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<4A54FBB8.4010808@wbonnet.net>	<4A550B96.9050604@opencsw.org>
	<4A550E6D.70600@wbonnet.net> <4A550F6B.6040508@opencsw.org>
Message-ID: <3E42BEFE-7FAC-4550-94C3-AD5B89DBFA8E@opencsw.org>

Hi Mike,

Am 08.07.2009 um 23:28 schrieb Mike Watters:
> William Bonnet wrote:
>> Hi Mike
>>
>>> I believe I saw some messages that Dago is on Holiday?
>>>
>> And without internet access... poor Dago, it should be really  
>> difficult ;)
>>
>> I think he will be back tomorrow or friday.
>>
>> cheers
>> W.
>>
>
> How can one live without internet access?

It was hard ;-) And when I got home my trusty old Dual-G5 Mac
was dead - mainboard defect :-(( I didn't feel much relaxed
after noticing *that*.


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul  9 23:47:47 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:47:47 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090709213525.GE8843@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
	<4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
	<20090709213525.GE8843@bolthole.com>
Message-ID: <E49405BA-0940-486C-BB8F-F1595C1C51E4@opencsw.org>

Hi Phil,

Am 09.07.2009 um 23:35 schrieb Philip Brown:

> On Thu, Jul 09, 2009 at 11:27:35PM +0200, Dagobert Michelsen wrote:
>> Hi Phil,
>>
>> Am 07.07.2009 um 22:22 schrieb Philip Brown:
>>> I dont know much about pkg-config, but might there be some kind of,
>>> "and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can
>>> put in
>>> /opt/csw/lib/pkgconfig/cairo.pc  or something?
>>
>> AFAIK this option is not available.
>>
>
> Then in my opinion, since libcairo requires libxcb now, then
> I think my suggestion here, would be best:
>
>
>>> or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
>>> should not exist there in the first place, but be moved
>>> to /opt/csw/lib/X11/lib/pkgconfig/   instead?
>
>
>> [Dago writes in reply to that...]
>> It is just for development. Using pkg-config this way allows specific
>> binding to either the OpenCSW X11 libs or Sun X11 libs with gtk+ on
>> compilation.
>
> I do not think that is a good (ie: stable) idea at all.
> If people want to use "sun libs", then they should probably use sun  
> cairo!
> on the other hand, seeing as how our libcairo now REQUIRES the newer  
> x11
> libs... libs that we were "forced" to ship ourselves... it doesnt make
> sense to break smooth opencsw development, for the fuzzy benefit of
> lalowing pepole to compile brokenly against sun libs as well.

To make it nice and clean the cairo shared libs should also go
into /opt/csw/X11/lib. Having pkg-config and libs in separate
places is IMHO even more unelegant than extending PKG_CONFIG_PATH.


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul  9 23:51:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:51:51 +0200
Subject: [csw-maintainers] Please repost open issues
Message-ID: <C7FD443D-3922-447D-87D6-094B7504E3D5@opencsw.org>

Hi,

there was a considerable backlog of messages to process during
my absence. I hope to have adressed all issues now (Rupert:
I stored your report about v2-pbuild malfunction for further
inspection without mail reply yet). However, due to the size
I may have missed a thing.

If you have posted issues I have not addressed yet please
repost them!


Best regards

   -- Dago

From phil at bolthole.com  Thu Jul  9 23:53:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 14:53:31 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <E49405BA-0940-486C-BB8F-F1595C1C51E4@opencsw.org>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
	<4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
	<20090709213525.GE8843@bolthole.com>
	<E49405BA-0940-486C-BB8F-F1595C1C51E4@opencsw.org>
Message-ID: <20090709215331.GF8843@bolthole.com>

On Thu, Jul 09, 2009 at 11:47:47PM +0200, Dagobert Michelsen wrote:
>...
> To make it nice and clean the cairo shared libs should also go
> into /opt/csw/X11/lib. Having pkg-config and libs in separate
> places is IMHO even more unelegant than extending PKG_CONFIG_PATH.

Hmmm....

initially makes sense to me.... except that if you carry that to its
logical extension, that would meant that EVERYTHING that depends on cairo,
etc,  would then also end up in /opt/csw/X11/lib.
That's a lot of libs.

Not sure that's so grand.

From pfelecan at opencsw.org  Fri Jul 10 12:22:26 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 12:22:26 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
Message-ID: <x6zlbc25y5.fsf@abhavana.free.fr>

pkg-get download common
Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
--2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
Resolving proxy... 192.168.1.6
Connecting to proxy|192.168.1.6|:3128... connected.
Proxy request sent, awaiting response... 404 Not Found
2009-07-10 12:21:01 ERROR 404: Not Found.

Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
(Perhaps you need to update your catalog?)
rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz': No such file or directory

Maybe this is a more general issue...
-- 
Peter

From dam at opencsw.org  Fri Jul 10 12:36:10 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 10 Jul 2009 12:36:10 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6zlbc25y5.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
Message-ID: <EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>

Hi Peter,

Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
> pkg-get download common
> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> Resolving proxy... 192.168.1.6
> Connecting to proxy|192.168.1.6|:3128... connected.
> Proxy request sent, awaiting response... 404 Not Found
> 2009-07-10 12:21:01 ERROR 404: Not Found.
>
> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> (Perhaps you need to update your catalog?)
> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386- 
> CSW.pkg.gz': No such file or directory
>
> Maybe this is a more general issue...

No, there was an error that the common package was gzipped and gzip
depended on it, so common is no longer stored gzipped. Just update
your catalog and everything should be fine again.


Best regards

   -- Dago

From pfelecan at opencsw.org  Fri Jul 10 13:29:37 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 13:29:37 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 10 Jul 2009 12\:36\:10 +0200")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>
Message-ID: <x6r5wo22u6.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
>> pkg-get download common
>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> Resolving proxy... 192.168.1.6
>> Connecting to proxy|192.168.1.6|:3128... connected.
>> Proxy request sent, awaiting response... 404 Not Found
>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>>
>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> (Perhaps you need to update your catalog?)
>> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386- 
>> CSW.pkg.gz': No such file or directory
>>
>> Maybe this is a more general issue...
>
> No, there was an error that the common package was gzipped and gzip
> depended on it, so common is no longer stored gzipped. Just update
> your catalog and everything should be fine again.

Read more carefully the subject line and you'll understand that this is
something that I cannot correct.
-- 
Peter

From dam at opencsw.org  Fri Jul 10 15:26:13 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 10 Jul 2009 15:26:13 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6r5wo22u6.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>
	<x6r5wo22u6.fsf@abhavana.free.fr>
Message-ID: <5EF7EE94-D633-4F7D-A68C-E8389FC84991@opencsw.org>

Hi Peter,

Am 10.07.2009 um 13:29 schrieb Peter FELECAN:
> Dagobert Michelsen <dam at opencsw.org> writes:
>> Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
>>> pkg-get download common
>>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>> Resolving proxy... 192.168.1.6
>>> Connecting to proxy|192.168.1.6|:3128... connected.
>>> Proxy request sent, awaiting response... 404 Not Found
>>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>>>
>>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>> (Perhaps you need to update your catalog?)
>>> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386-
>>> CSW.pkg.gz': No such file or directory
>>>
>>> Maybe this is a more general issue...
>>
>> No, there was an error that the common package was gzipped and gzip
>> depended on it, so common is no longer stored gzipped. Just update
>> your catalog and everything should be fine again.
>
> Read more carefully the subject line and you'll understand that this  
> is
> something that I cannot correct.

Aha, the catalog is only updatable by root. Should be working now.
However, the catalogs on the buildfarm are only updated from time
to time. It should IMHO be possible for pkg-get (and pkgutil btw.)
to download without any need for root access. Or have I gotten it
wrong again?


Best regards

   -- Dago

From phil at bolthole.com  Fri Jul 10 16:11:48 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 10 Jul 2009 07:11:48 -0700
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6zlbc25y5.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
Message-ID: <20090710141148.GA99639@bolthole.com>

On Fri, Jul 10, 2009 at 12:22:26PM +0200, Peter FELECAN wrote:
> pkg-get download common
> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> Resolving proxy... 192.168.1.6
> Connecting to proxy|192.168.1.6|:3128... connected.
> Proxy request sent, awaiting response... 404 Not Found
> 2009-07-10 12:21:01 ERROR 404: Not Found.

> 
> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> (Perhaps you need to update your catalog?)

So... "perhaps you need to update your catalog"?

the master catalog points to a .pkg file, not a .pkg.gz file.

if that's what you were referring to?

(and perhaps your proxy has cached the old catalog file or something)

The current live catalog seems ok to me.

From pfelecan at opencsw.org  Fri Jul 10 16:34:43 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 16:34:43 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <5EF7EE94-D633-4F7D-A68C-E8389FC84991@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 10 Jul 2009 15\:26\:13 +0200")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>
	<x6r5wo22u6.fsf@abhavana.free.fr>
	<5EF7EE94-D633-4F7D-A68C-E8389FC84991@opencsw.org>
Message-ID: <x6my7c1u9o.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 10.07.2009 um 13:29 schrieb Peter FELECAN:
>> Dagobert Michelsen <dam at opencsw.org> writes:
>>> Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
>>>> pkg-get download common
>>>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>>> Resolving proxy... 192.168.1.6
>>>> Connecting to proxy|192.168.1.6|:3128... connected.
>>>> Proxy request sent, awaiting response... 404 Not Found
>>>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>>>>
>>>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>>> (Perhaps you need to update your catalog?)
>>>> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386-
>>>> CSW.pkg.gz': No such file or directory
>>>>
>>>> Maybe this is a more general issue...
>>>
>>> No, there was an error that the common package was gzipped and gzip
>>> depended on it, so common is no longer stored gzipped. Just update
>>> your catalog and everything should be fine again.
>>
>> Read more carefully the subject line and you'll understand that this
>> is
>> something that I cannot correct.
>
> Aha, the catalog is only updatable by root. Should be working now.
> However, the catalogs on the buildfarm are only updated from time
> to time. It should IMHO be possible for pkg-get (and pkgutil btw.)
> to download without any need for root access. Or have I gotten it
> wrong again?

The download is possible by a non root user --- this is what I'm doing.

I confirm that you get it now correctly and the catalog on build8x is
now correct.

Thank you for this intervention.
-- 
Peter

From pfelecan at opencsw.org  Fri Jul 10 16:35:51 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 16:35:51 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <20090710141148.GA99639@bolthole.com> (Philip Brown's message of
	"Fri\, 10 Jul 2009 07\:11\:48 -0700")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
Message-ID: <x6iqi01u7s.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Fri, Jul 10, 2009 at 12:22:26PM +0200, Peter FELECAN wrote:
>> pkg-get download common
>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> Resolving proxy... 192.168.1.6
>> Connecting to proxy|192.168.1.6|:3128... connected.
>> Proxy request sent, awaiting response... 404 Not Found
>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>
>> 
>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> (Perhaps you need to update your catalog?)
>
> So... "perhaps you need to update your catalog"?
>
> the master catalog points to a .pkg file, not a .pkg.gz file.
>
> if that's what you were referring to?

No. As stated in the subject, the catalog for build8x, which resides in
/var/pkg-get/catalog-mirror wasn't up to date.

> (and perhaps your proxy has cached the old catalog file or something)
>
> The current live catalog seems ok to me.

It is.
-- 
Peter

From phil at bolthole.com  Fri Jul 10 17:04:46 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 10 Jul 2009 08:04:46 -0700
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6iqi01u7s.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
Message-ID: <20090710150446.GA75987@bolthole.com>

On Fri, Jul 10, 2009 at 04:35:51PM +0200, Peter FELECAN wrote:
>....
> No. As stated in the subject, the catalog for build8x, which resides in
> /var/pkg-get/catalog-mirror wasn't up to date.
> 
> Philip Brown <phil at bolthole.com> writes:
> > (and perhaps your proxy has cached the old catalog file or something)
> >
> > The current live catalog seems ok to me.
> 
> It is.
> -- 

i'm confused what "fixed" this then.

From dam at opencsw.org  Fri Jul 10 17:08:54 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 10 Jul 2009 17:08:54 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <20090710150446.GA75987@bolthole.com>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
Message-ID: <163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>

Hi Phil,

Am 10.07.2009 um 17:04 schrieb Philip Brown:
> On Fri, Jul 10, 2009 at 04:35:51PM +0200, Peter FELECAN wrote:
>> ....
>> No. As stated in the subject, the catalog for build8x, which  
>> resides in
>> /var/pkg-get/catalog-mirror wasn't up to date.
>>
>> Philip Brown <phil at bolthole.com> writes:
>>> (and perhaps your proxy has cached the old catalog file or  
>>> something)
>>>
>>> The current live catalog seems ok to me.
>>
>> It is.
>> -- 
>
> i'm confused what "fixed" this then.

- The catalog on build8x was old
- There was no package available for the "old" information in the  
outdated catalog
   when Peter tried to download the package
- The catalog can only be updated by root
- Peter is not root
- I updated the catalog
- Everything worked


Best regards

   -- Dago

From phil at bolthole.com  Fri Jul 10 18:17:24 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 10 Jul 2009 09:17:24 -0700
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
	<163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
Message-ID: <20090710161724.GA73109@bolthole.com>

On Fri, Jul 10, 2009 at 05:08:54PM +0200, Dagobert Michelsen wrote:
> - The catalog on build8x was old

oh... this was a build-machine-specific problem.

pfft. he should have mailed the buildfarm list not the maintainers list
then! :-P


From maciej at opencsw.org  Fri Jul 10 19:00:31 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 10 Jul 2009 18:00:31 +0100
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <4A565D6E.9050201@opencsw.org>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
	<ce97543b0907090302l17245204k48f8c6e3fa9b09f3@mail.gmail.com>
	<4A565D6E.9050201@opencsw.org>
Message-ID: <ce97543b0907101000i4f44fe42r308f5213edeb224@mail.gmail.com>

On Thu, Jul 9, 2009 at 10:13 PM, Mike Watters<mwatters at opencsw.org> wrote:
> I will make the change the next time I release the package.
> please file a feature request bug so I don't forget about it ;-}

Done.

http://www.opencsw.org/bugtrack/view.php?id=3763

From pfelecan at opencsw.org  Fri Jul 10 19:42:15 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 19:42:15 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <20090710161724.GA73109@bolthole.com> (Philip Brown's message of
	"Fri\, 10 Jul 2009 09\:17\:24 -0700")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com> <x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
	<163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
	<20090710161724.GA73109@bolthole.com>
Message-ID: <x6eiso1ll4.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Fri, Jul 10, 2009 at 05:08:54PM +0200, Dagobert Michelsen wrote:
>> - The catalog on build8x was old
>
> oh... this was a build-machine-specific problem.
>
> pfft. he should have mailed the buildfarm list not the maintainers list
> then! :-P

*he* didn't know that it was build machine specific; it was revealed on
it, following the correction of a global issue, i.e., not build machine
specific...

tat sat
-- 
Peter

From william at wbonnet.net  Sat Jul 11 14:34:14 2009
From: william at wbonnet.net (William Bonnet)
Date: Sat, 11 Jul 2009 14:34:14 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A54DA45.4090902@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
	<4A54DA45.4090902@wbonnet.net>
Message-ID: <4A5886C6.1040405@wbonnet.net>

Hi

> Hi Ben
>> Excerpts from William Bonnet's message of Tue Jul 07 19:00:10 -0400 
>> 2009:
>>  
>>> I do have the same problem with the /opt/csw/include/jri.h file. It 
>>> comes from the same packages and also says it has to be provided by 
>>> netscape.
>>>
>>> Could someone please rename it to .old on build9x ?
>>>     
>>
>> Doing this now.
>>   
> Thanks
>
> It worked, i was able to compile FF3 on solaris 9, and the binaries 
> runs on solaris 10. Now i have to find tests machines running solaris 
> 8 and 9 :)
>
> I also need the same fix on build9x 8x and 9x. I better look to 
> produce a Qt patch :)

I have recompiled FF3.0 on build8x, but i need to have the following 
conflicting file to be removed from build8s

/opt/csw/include/jri.h
/opt/csw/include/npapi.h

Thanks in advance

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 lsw at bender.opencsw.org  Sat Jul 11 18:52:00 2009
From: lsw at bender.opencsw.org (Liu Siwei)
Date: Sat, 11 Jul 2009 18:52:00 +0200
Subject: [csw-maintainers] /testing libUPnP 1.6.6
Message-ID: <20090711165200.GA324@bender.opencsw.org>

http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-sparc-CSW.pkg.gz




# pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libupnp

   or,

# pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libupnp


--
Siwei


From dam at opencsw.org  Sat Jul 11 19:02:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 11 Jul 2009 19:02:07 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A5886C6.1040405@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
	<4A54DA45.4090902@wbonnet.net> <4A5886C6.1040405@wbonnet.net>
Message-ID: <956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>

Hi William,

Am 11.07.2009 um 14:34 schrieb William Bonnet:
> I have recompiled FF3.0 on build8x

Cool :-)

> but i need to have the following conflicting file to be removed from  
> build8s
>
> /opt/csw/include/jri.h
> /opt/csw/include/npapi.h

Done.


Best regards

   -- Dago


From dam at opencsw.org  Sat Jul 11 19:11:21 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 11 Jul 2009 19:11:21 +0200
Subject: [csw-maintainers] Booting login and *s
Message-ID: <AEA7EEC5-83EE-4F2A-B266-0C8F37A4072E@opencsw.org>

Hi,

there is a serious I/O problem on the T5220 and all zones. It seems
that some I/Os are simply stuck, but everything else looks ok. As
I can't figure out what is wrong (no disk errs, no faulted pools,
nothing in messages) I plan to boot the server as last resort in
about 30 minutes.


Sorry for the inconvenience

   -- Dago

From loseweigh at gmail.com  Sat Jul 11 18:37:03 2009
From: loseweigh at gmail.com (Liu Siwei)
Date: Sun, 12 Jul 2009 00:37:03 +0800
Subject: [csw-maintainers] /testing libUPnP 1.6.6
Message-ID: <3dd2fcb80907110937h69c17f81xb3bc00a0c2b5068b@mail.gmail.com>

http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-sparc-CSW.pkg.gz





# pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libupnp

   or,

# pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libupnp



-Siwei
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090712/bd58348b/attachment.html>

From dam at opencsw.org  Sun Jul 12 09:21:33 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 09:21:33 +0200
Subject: [csw-maintainers] New in testing: libX11 1.2.2, libvorbis 1.2.3,
	neon 0.28.5
Message-ID: <4E7AC449-B3B6-468D-B345-3D3F43EC2A09@opencsw.org>

Hi,

new packages in
   http://mirror.opencsw.org/testing.html

neon-0.28.5,REV=2009.07.11-SunOS5.8-sparc-CSW.pkg.gz
neon-0.28.5,REV=2009.07.11-SunOS5.8-i386-CSW.pkg.gz
neon_devel-0.28.5,REV=2009.07.11-SunOS5.8-sparc-CSW.pkg.gz
neon_devel-0.28.5,REV=2009.07.11-SunOS5.8-i386-CSW.pkg.gz

libvorbis-1.2.3,REV=2009.07.11-SunOS5.8-sparc-CSW.pkg.gz
libvorbis-1.2.3,REV=2009.07.11-SunOS5.8-i386-CSW.pkg.gz

libx11-1.2.2,REV=2009.07.12-SunOS5.8-i386-CSW.pkg.gz
libx11-1.2.2,REV=2009.07.12-SunOS5.8-sparc-CSW.pkg.gz
libx11_devel-1.2.2,REV=2009.07.12-SunOS5.8-i386-CSW.pkg.gz
libx11_devel-1.2.2,REV=2009.07.12-SunOS5.8-sparc-CSW.pkg.gz

If you have time please test against your applications.


Best regards

   -- Dago

From ihsan at opencsw.org  Sun Jul 12 12:38:38 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 12 Jul 2009 12:38:38 +0200
Subject: [csw-maintainers] contact maintainer page
In-Reply-To: <4A366723.3050806@opencsw.org>
References: <x6vdmxzjau.fsf@abhavana.free.fr>	<20090615145906.GE34818@bolthole.com>	<55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de>
	<4A366723.3050806@opencsw.org>
Message-ID: <4A59BD2E.7030607@opencsw.org>

Am 15.6.2009 17:22 Uhr, Mike Watters schrieb:

>>>> I'm trying to use the site's mailer:
>>>> http://www.opencsw.org/eml.php?devname=pfelecan
>>>>
>>>> After entering my e-mail address it shows me a page with the text
>>>> "Debug: part 1". Nice but not useful if somebody tries to write me
>>>> something through this...
>>> posix_getpwnam()    [a php call]
>>> seems to be broken somehow on the website now.
>> Might be related to the recent php5 modules split-off? [1]
> 
> Installing the following should fix
> 
> php5_posix 5.2.9,REV=2009.04.29

It did. Thanks & sorry for the delay.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From william at wbonnet.net  Sun Jul 12 13:20:44 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 12 Jul 2009 13:20:44 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>	<4A54DA45.4090902@wbonnet.net>
	<4A5886C6.1040405@wbonnet.net>
	<956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>
Message-ID: <4A59C70C.7000503@wbonnet.net>

Hi

>> I have recompiled FF3.0 on build8x
>
> Cool :-)
Packages are not yet tested on every platform but recompiled. x86 
package works on Solaris 10. I am look for x86 Solaris 8 and 9 VMs (or 
VBox or whatever). I am currently  building Sparc version of FF3

So if some wants to do some testing FF3.0.11 x86 is available from 
testing. I'll announce it on the users list when  sparc version will be ok.

I would like to have a feed back on solaris 8. It seems sol9 will be ok, 
but a  link b-problem exists on solaris 8.
 
>
>> but i need to have the following conflicting file to be removed from 
>> build8s
>>
>> /opt/csw/include/jri.h
>> /opt/csw/include/npapi.h
>
> Done.
Thanks

I'll try to repackage Qt when i'll be done wih FF...

I've able to build FF3.5 on my box (it means no more source to patch or 
link issues to go until the end and poduce package). But binary does not 
work yet (illegal instruction and core dump). So i still have things to 
tweak. I don't know how long it may take.

cheers
W.


From ihsan at opencsw.org  Sun Jul 12 13:26:03 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 12 Jul 2009 13:26:03 +0200
Subject: [csw-maintainers]
	=?iso-8859-1?q?Additional_Farm_Maintainer=3A_J?=
	=?iso-8859-1?q?=FCrgen_Arndt?=
In-Reply-To: <06DF27E0-FF15-455A-9040-5A5C5160F8AC@opencsw.org>
References: <06DF27E0-FF15-455A-9040-5A5C5160F8AC@opencsw.org>
Message-ID: <4A59C84B.7020004@opencsw.org>

Am 9.7.2009 20:56 Uhr, Dagobert Michelsen schrieb:

> I'm back from my vacation and noticed that there are obviously not
> enough farm managers as Ben wanted to take some time off, too. I have
> given J?rgen full admin access to the farm. Ihsan, please add "ja"
> to the list of buildfarm maintainers.

Actually he is already a member of the buildfarm list.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From skayser at opencsw.org  Sun Jul 12 14:54:54 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sun, 12 Jul 2009 14:54:54 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
 instead of PROTOTYPE_FILTER
In-Reply-To: <EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
Message-ID: <4A59DD1E.2060205@opencsw.org>

Dagobert Michelsen wrote:
> Am 04.07.2009 um 19:12 schrieb Sebastian Kayser:
>> Dagobert Michelsen wrote:
>>> Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
>>>> while working on postfix 2.6 i have noticed again that it is not  
>>>> quite
>>>> straight-forward to assign files to classes. Instead of exposing the
>>>> inner
>>>> workings of the prototype stuff to the Makefile, could we maybe get
>>>> some
>>>> variables to assign files to classes (functions)?
>>> For CSWcswclassutils this is already in there :-) Just define
>>>
>>> SAMPLECONF
>>> PRESERVECONF
>>> INITSMF
>>>
>>> and everything else will be taken cared of.
>> Sweet!!! :D Can we add something for USERGROUP also?
> ...
> Looks good. Please go ahead and commit.

Done (r5569). I have also added the variables to the GAR variable
reference [1]. What about cswpycompile, do we want to have that as such
a variable (PYCOMPILE) also?

Peter, would you mind amending the cswclassutils wiki page to reflect
the variables (PRESERVECONF, SAMPLECONF, INITSMF, USERGROUP) that can be
used instead of PROTOTYPE_FILTER?

Sebastian

[1]http://sourceforge.net/apps/trac/gar/wiki/GAR%20Variable%20Reference#cswclassutilsSettings

From dam at opencsw.org  Sun Jul 12 15:23:02 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 15:23:02 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <4A59DD1E.2060205@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
Message-ID: <E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>

Hi Sebastian,

Am 12.07.2009 um 14:54 schrieb Sebastian Kayser:
> Done (r5569). I have also added the variables to the GAR variable
> reference [1]. What about cswpycompile, do we want to have that as  
> such
> a variable (PYCOMPILE) also?

Should be done. Something like this is needed for Python (taken
from x11/xcb-proto):

> PYTHON_PREFIX=$(BUILD_PREFIX)/lib/python/site-packages
>
> EXTRA_PAX_ARGS = -s ',^.*/\(xcbgen/.*\.py\),.$(BUILD_PREFIX)/lib/ 
> python/site-packages/\1,'
> EXTRA_PAX_ARGS += -s ',^\.$(libdir)/python.*,,'
>
> EXTRA_MERGE_EXCLUDE_FILES = .*\.pyo .*\.pyc
> PROTOTYPE_FILTER = awk '$$$$3 ~ /.*\.py$$$$/ { $$$$2 =  
> "cswpycompile" } { print }'
> SPKG_CLASSES = none cswpycompile
> REQUIRED_PKGS += CSWcswclassutils
>
> PKGFILES_CSWpyxcbproto = .*\.py

The detection would however be slightly more complicated as the presence
of .py-files should be automatically detected. If present, the filter
should be put in place and the class changed for the .py-files.
As of the current implementation of the cswpycompile class the .py-files
must be put exactly where they are relocated in this example via
pax-relocation during the merge phase. If they are anywhere else
a warning (error?) should be printed.


Best regards

   -- Dago

From dam at opencsw.org  Sun Jul 12 15:24:30 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 15:24:30 +0200
Subject: [csw-maintainers] contact maintainer page
In-Reply-To: <4A59BD2E.7030607@opencsw.org>
References: <x6vdmxzjau.fsf@abhavana.free.fr>	<20090615145906.GE34818@bolthole.com>	<55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de>
	<4A366723.3050806@opencsw.org> <4A59BD2E.7030607@opencsw.org>
Message-ID: <9E7F8956-A5E6-4110-8DFD-3B70DE9ED1F2@opencsw.org>

Hi,

Am 12.07.2009 um 12:38 schrieb Ihsan Dogan:

> Am 15.6.2009 17:22 Uhr, Mike Watters schrieb:
>
>>>>> I'm trying to use the site's mailer:
>>>>> http://www.opencsw.org/eml.php?devname=pfelecan
>>>>>
>>>>> After entering my e-mail address it shows me a page with the text
>>>>> "Debug: part 1". Nice but not useful if somebody tries to write me
>>>>> something through this...
>>>> posix_getpwnam()    [a php call]
>>>> seems to be broken somehow on the website now.
>>> Might be related to the recent php5 modules split-off? [1]
>>
>> Installing the following should fix
>>
>> php5_posix 5.2.9,REV=2009.04.29
>
> It did. Thanks & sorry for the delay.

Not quite:

> Debug: part 1. Problem with getpwnam Debug part 2. If you see this,  
> problem is fixed Thank you. Your message has been sent to maintainer  
> Dagobert Michelsen .

Please fix the webpage also.


Best regards

   -- Dago

From dam at opencsw.org  Sun Jul 12 15:25:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 15:25:55 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6eiso1ll4.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
	<163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
	<20090710161724.GA73109@bolthole.com>
	<x6eiso1ll4.fsf@abhavana.free.fr>
Message-ID: <9D4229A3-579A-415A-AFFB-484C736B735B@opencsw.org>

Hi Peter,

Am 10.07.2009 um 19:42 schrieb Peter FELECAN:
> Philip Brown <phil at bolthole.com> writes:
>> On Fri, Jul 10, 2009 at 05:08:54PM +0200, Dagobert Michelsen wrote:
>>> - The catalog on build8x was old
>>
>> oh... this was a build-machine-specific problem.
>>
>> pfft. he should have mailed the buildfarm list not the maintainers  
>> list
>> then! :-P
>
> *he* didn't know that it was build machine specific; it was revealed  
> on
> it, following the correction of a global issue, i.e., not build  
> machine
> specific...
>
> tat sat

No problem :-) Apart from that both tools should have an option for
download which does in no case require root access.


Best regards

   -- Dago

From skayser at opencsw.org  Sun Jul 12 16:17:54 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sun, 12 Jul 2009 16:17:54 +0200
Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW
In-Reply-To: <4A294F5B.1070905@cognigencorp.com>
References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de>
	<4A294F5B.1070905@cognigencorp.com>
Message-ID: <4A59F092.6090008@opencsw.org>

Hi Darin,

Darin Perusich wrote:
> Sebastian Kayser wrote:
>> does someone have a working "install me some OpenCSW packages" Jumpstart
>> post-install snippet that he could pass to me? I am in the process of
>> setting up my first Jumpstart server and i would like to have it install a
>> list of CSW pkgs after setting up the machine.
> 
> I attached my CSW package installation script. It's nothing fancy but
> get the job done for me.
>
>> Is this known to work with the various postinstall scripts (or cswinitsmf)
>> that we have in the packages?
>>
>> And out of curiosity: What are the specifics to watch out for so that a
>> postinstall script will work in a Jumpstart environment, just
>> $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up
>> frequently?
> 
> If you chroot the package installation you don't need to worry about
> $PKG_INSTALL_ROOT being set. I use chroot to install everything except
> pkg-get and I've never had a problem. Granted I only install a small
> number of package, sudo, amanda, etc, and build the servers out
> afterwards. If it's a machine I need to make a ton of clones of I set it
> up and create flash images of anyways.

thanks for posting this, I finally had some time to further work on our
Jumpstart setup.

I found it easiest to use JET [1] and custom_scripts_n, so that post
installation scripts like this are run after the final system boot. The
system is then fully functional and there is no need to mess around with
PKG_INSTALL_ROOT or similar. :) Easy.

The adapted version of your script is attached. Thanks again.

Sebastian

[1] http://wikis.sun.com/display/JET/Home
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: install_csw.sh
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090712/ef881f46/attachment.ksh>

From phil at bolthole.com  Sun Jul 12 18:31:00 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 12 Jul 2009 09:31:00 -0700
Subject: [csw-maintainers] contact maintainer page
In-Reply-To: <9E7F8956-A5E6-4110-8DFD-3B70DE9ED1F2@opencsw.org>
References: <x6vdmxzjau.fsf@abhavana.free.fr>
	<20090615145906.GE34818@bolthole.com>
	<55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de>
	<4A366723.3050806@opencsw.org> <4A59BD2E.7030607@opencsw.org>
	<9E7F8956-A5E6-4110-8DFD-3B70DE9ED1F2@opencsw.org>
Message-ID: <20090712163059.GF63696@bolthole.com>

On Sun, Jul 12, 2009 at 03:24:30PM +0200, Dagobert Michelsen wrote:
>> Debug: part 1. Problem with getpwnam Debug part 2. If you see this,  
>> problem is fixed Thank you. Your message has been sent to maintainer  
>> Dagobert Michelsen .
>
> Please fix the webpage also.

this is done

From james at opencsw.org  Sun Jul 12 18:59:01 2009
From: james at opencsw.org (James Lee)
Date: Sun, 12 Jul 2009 16:59:01 GMT
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A59C70C.7000503@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
	<4A53D37A.40502@wbonnet.net>
	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
	<4A54DA45.4090902@wbonnet.net> <4A5886C6.1040405@wbonnet.net>
	<956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>
	<4A59C70C.7000503@wbonnet.net>
Message-ID: <20090712.16590100.1755802829@gyor.oxdrove.co.uk>

On 12/07/09, 12:20:44, William Bonnet <william at wbonnet.net> wrote regarding
Re: [csw-maintainers] Conflicting header file:

> So if some wants to do some testing FF3.0.11 x86 is available from
> testing. I'll announce it on the users list when  sparc version will be
> ok.

> I would like to have a feed back on solaris 8. It seems sol9 will be ok,
> but a  link b-problem exists on solaris 8.



$ uname -a
SunOS bala 5.8 Generic_117351-32 i86pc i386 i86pc
$ firefox
ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
object requires TLS, but TLS failed to initialize
Killed
$ pkgparam CSWfirefox VERSION
3.0.11,REV=2009.07.11




$ uname -a
SunOS gyor 5.8 Generic_117350-38 sun4u sparc SUNW,Ultra-60
$ firefox
ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
object requires TLS, but TLS failed to initialize
Killed
$ pkgparam CSWfirefox VERSION
3.0.11,REV=2009.07.12

From william at wbonnet.net  Sun Jul 12 19:04:48 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 12 Jul 2009 19:04:48 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <20090712.16590100.1755802829@gyor.oxdrove.co.uk>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>	<4A54DA45.4090902@wbonnet.net>
	<4A5886C6.1040405@wbonnet.net>	<956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>	<4A59C70C.7000503@wbonnet.net>
	<20090712.16590100.1755802829@gyor.oxdrove.co.uk>
Message-ID: <4A5A17B0.9040201@wbonnet.net>

Hi James
>> I would like to have a feed back on solaris 8. It seems sol9 will be ok,
>> but a  link b-problem exists on solaris 8.
>>     
>
>
>
> $ uname -a
> SunOS bala 5.8 Generic_117351-32 i86pc i386 i86pc
> $ firefox
> ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
> object requires TLS, but TLS failed to initialize
> Killed
> $ pkgparam CSWfirefox VERSION
> 3.0.11,REV=2009.07.11
>
>
>
>
> $ uname -a
> SunOS gyor 5.8 Generic_117350-38 sun4u sparc SUNW,Ultra-60
> $ firefox
> ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
> object requires TLS, but TLS failed to initialize
> Killed
> $ pkgparam CSWfirefox VERSION
> 3.0.11,REV=2009.07.12
>   

Thanks

That's what i was affraid of :( I will set up a test machine here since 
i'm not in Paris and with a lot of oldies.

Anyone tried on Solaris 9 ?

cheers
W.


From dam at opencsw.org  Mon Jul 13 16:47:20 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 13 Jul 2009 16:47:20 +0200
Subject: [csw-maintainers] Attention: Please clean up /home
Message-ID: <9BA10F46-5ABB-4ED0-8283-F2AAFF91078B@opencsw.org>

Hi,

it looks like the machine hit bug# 6596237:
   <http://sunsolve.sun.com/search/document.do?assetkey=1-1-6596237-1>
I have IDR137886-01 which should bring some relief, however, the IDR
was built against an older kernel version and downgrading requires
removing *a lot* of patches. Nonetheless I will continue working
towards it. Additionally, the problem can be solved by making free
space in the pool.

So: Please delete stuff from your homes you no longer need! The
extra space will speed up processing considerably.


Sorry for the inconvenience

   -- Dago

From maciej at opencsw.org  Mon Jul 13 20:28:28 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 13 Jul 2009 19:28:28 +0100
Subject: [csw-maintainers] stat command
Message-ID: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>

Has anyone built the 'stat' utility on Solaris? In Debian, the
/usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
the same utilities are part of the 'fileutils' package (CSWgfile).
Were there any plans to add 'stat' to fileutils?

Maciej

From bwalton at opencsw.org  Mon Jul 13 20:44:37 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 13 Jul 2009 14:44:37 -0400
Subject: [csw-maintainers] stat command
In-Reply-To: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
Message-ID: <1247510598-sup-7662@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Mon Jul 13 14:28:28 -0400 2009:
> Has anyone built the 'stat' utility on Solaris? In Debian, the
> /usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
> the same utilities are part of the 'fileutils' package (CSWgfile).
> Were there any plans to add 'stat' to fileutils?

Not yet, but I was about to build coreutils for exactly this reason.
I sorely miss the stat command...I was waiting on an update to another
package (forget which without checking) before I could build the
current/stable coreutils though.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090713/44a57cdb/attachment.asc>

From phil at bolthole.com  Mon Jul 13 21:00:36 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 13 Jul 2009 12:00:36 -0700
Subject: [csw-maintainers] stat command
In-Reply-To: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
Message-ID: <20090713190036.GB8171@bolthole.com>

On Mon, Jul 13, 2009 at 07:28:28PM +0100, Maciej (Matchek) Blizinski wrote:
> Has anyone built the 'stat' utility on Solaris? In Debian, the
> /usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
> the same utilities are part of the 'fileutils' package (CSWgfile).


how odd. i can think of no reason why it wouldnt be in the package, if it
really is usually as part of that software distribution.

From Darin.Perusich at cognigencorp.com  Mon Jul 13 21:04:32 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 13 Jul 2009 15:04:32 -0400
Subject: [csw-maintainers] stat command
In-Reply-To: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
Message-ID: <4A5B8540.5020803@cognigencorp.com>

I had a request to updated to coreutils that went unanswered so I
compiled the package. It's in ~darin/build/ if you want to take a look.

Maciej (Matchek) Blizinski wrote:
> Has anyone built the 'stat' utility on Solaris? In Debian, the
> /usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
> the same utilities are part of the 'fileutils' package (CSWgfile).
> Were there any plans to add 'stat' to fileutils?
> 


-- 
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  Mon Jul 13 21:08:39 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 13 Jul 2009 15:08:39 -0400
Subject: [csw-maintainers] stat command
In-Reply-To: <20090713190036.GB8171@bolthole.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
	<20090713190036.GB8171@bolthole.com>
Message-ID: <1247511880-sup-5149@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Mon Jul 13 15:00:36 -0400 2009:
> how odd. i can think of no reason why it wouldnt be in the package, if it
> really is usually as part of that software distribution.

Coreutils is a superset of the fileutils, shellutils and textutils
packages, which are no longer offered as individual packages.  The
actual stat command used to be part of shellutils, I believe.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090713/c429084e/attachment.asc>

From dam at opencsw.org  Mon Jul 13 23:23:48 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 13 Jul 2009 23:23:48 +0200
Subject: [csw-maintainers] ZFS issue again
Message-ID: <13222C2C-B152-47B6-BA8D-00B7D0B5DA46@opencsw.org>

Hi,

as I was not able to install the IDR due to zone patch issues I
replaced the internal 146 GB disks with 300 GB disks. For the
pool resize I need to unmount/remount the pool, so please log
out for a few minutes.


Sorry for the inconvenience

   -- Dago

From dam at opencsw.org  Mon Jul 13 23:38:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 13 Jul 2009 23:38:08 +0200
Subject: [csw-maintainers] ZFS issue again
In-Reply-To: <13222C2C-B152-47B6-BA8D-00B7D0B5DA46@opencsw.org>
References: <13222C2C-B152-47B6-BA8D-00B7D0B5DA46@opencsw.org>
Message-ID: <764C586D-80EE-4CF9-8E88-F203BC5ECA00@opencsw.org>

Hi,

Am 13.07.2009 um 23:23 schrieb Dagobert Michelsen:
> as I was not able to install the IDR due to zone patch issues I
> replaced the internal 146 GB disks with 300 GB disks. For the
> pool resize I need to unmount/remount the pool, so please log
> out for a few minutes.

I hope this is resolved now. At least in my few tests the
performance is quite good :-D

> dam at login [login]:/home/dam > time gtar zxf /home/src/ 
> perl-5.10.0.tar.gz
> gtar zxf /home/src/perl-5.10.0.tar.gz  3.17s user 3.17s system 126%  
> cpu 5.004 total
> dam at login [login]:/home/dam > find perl-5.10.0 | wc -l
>     4680
> dam at login [login]:/home/dam > time mkfile 1g bigfile
> mkfile 1g bigfile  0.36s user 9.52s system 99% cpu 9.892 total



Best regards

   -- Dago

From william at wbonnet.net  Wed Jul 15 12:01:16 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 12:01:16 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
Message-ID: <4A5DA8EC.1080803@wbonnet.net>

Hi

I noticed today that there is an open issue on xpdf since Junuary 2008 
that received several "ping" from the reporter. Current maintaner, 
Alessio is retired. Would someone be interested in taking other this 
package and fix the opened issues ?

Thanks in advance

cheers
W.

From skayser at opencsw.org  Wed Jul 15 12:13:23 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 15 Jul 2009 12:13:23 +0200 (CEST)
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <4A5DA8EC.1080803@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>
Message-ID: <59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>

William Bonnet wrote:
> I noticed today that there is an open issue on xpdf since Junuary 2008
> that received several "ping" from the reporter. Current maintaner,
> Alessio is retired. Would someone be interested in taking other this
> package and fix the opened issues ?

>From what i know, Benny has been working on an updated version. He is at a
customer event right now, but he should be able to tell you whether i was
right or not later on today.

While we are at fixing bugs: Any news on the dbus not properly shutting
down issue that has been open for a while now [1]? We have had two boxes
here hanging on an attempted reboot just the other day.

Sebastian

[1] http://opencsw.org/mantis/view.php?id=3626


From maciej at opencsw.org  Wed Jul 15 13:34:32 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 15 Jul 2009 12:34:32 +0100
Subject: [csw-maintainers] Modifying SMF dependencies
Message-ID: <ce97543b0907150434p31370d05q46eb87e4675e7865@mail.gmail.com>

When working on syslog_ng, I came across a new problem. A system
service such as syslog, usually has some other services depending on
it. I've got into a problem in which I couldn't run rarpd together
with cswsyslog_ng, because rarpd depends on system-log, which was
disabled.

Have any of you gotten around a problem "How do I automate SMF
dependency changes"? If so, do you have any code related to that?

If you haven't, what would you do? I suspect there would be svccfg
export and svccfg import involved, with XML munging along the way.
What would make the minimal dependency collection for XML modification
tools on Solaris? (I only work with Solaris 10, I fear asking about
XML processing on Solaris 8...)

Maciej

From william at wbonnet.net  Wed Jul 15 14:20:22 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 14:20:22 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
Message-ID: <4A5DC986.3020606@wbonnet.net>

Hi
> While we are at fixing bugs: Any news on the dbus not properly shutting
> down issue that has been open for a while now [1]? We have had two boxes
> here hanging on an attempted reboot just the other day.
>   
Not yet :( I'll give a look to this issue tonight. I was focused on 
firefox compilation issues so far...

cheers
W.

From ellson at opencsw.org  Wed Jul 15 18:15:56 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 15 Jul 2009 12:15:56 -0400
Subject: [csw-maintainers] howto control which files go into which part of
	split package
Message-ID: <4A5E00BC.3090800@opencsw.org>

1.   How can I list the file contents of a .pkg.gz to verify what I've done?

2.   How do I specify the set of files to go into a subpackage, and 
exclude those same files from the other
subpackages?


e.g.   if I have:

     PKGFILES_CSWgraphviztcl = $(libexecdir_install)/graphviz/tcl/.*

does that automatically exclude those matched files from the core 
graphviz package?




Current attempt is in:

     /home/ellson/mgar/pkg/graphviz/trunk/Makefile

From skayser at opencsw.org  Wed Jul 15 22:21:46 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 15 Jul 2009 22:21:46 +0200 (CEST)
Subject: [csw-maintainers] howto control which files go into which part
 of split package
In-Reply-To: <4A5E00BC.3090800@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
Message-ID: <50628.217.227.36.102.1247689306.squirrel@ssl.skayser.de>

Hi John,

John Ellson wrote:
> 1.   How can I list the file contents of a .pkg.gz to verify what I've
> done?

  $ gunzip .pkg.gz
  $ pkgtrans .pkg .

Then dig around in CSWpkg. If you just build one package (that is without
splitting files in separate packages), you can also have a look in
work/pkgroot just after the "gmake merge" phase. That is what i usually
do, instead of "gmake package", first do a gmake merge and the find
work/pkgroot |less.

> 2.   How do I specify the set of files to go into a subpackage, and
> exclude those same files from the other
> subpackages?
>
> e.g.   if I have:
>
>      PKGFILES_CSWgraphviztcl = $(libexecdir_install)/graphviz/tcl/.*
>
> does that automatically exclude those matched files from the core
> graphviz package?

Yes, should be the default behaviour. See [1,2].

Sebastian

[1] http://sourceforge.net/apps/trac/gar/wiki/Prototypes
[2] http://sourceforge.net/apps/trac/gar/wiki/DynamicPackageFiles


From dam at opencsw.org  Wed Jul 15 22:22:28 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 15 Jul 2009 22:22:28 +0200
Subject: [csw-maintainers] howto control which files go into which part
	of split package
In-Reply-To: <4A5E00BC.3090800@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
Message-ID: <1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>

Ho John,

Am 15.07.2009 um 18:15 schrieb John Ellson:
> 1.   How can I list the file contents of a .pkg.gz to verify what  
> I've done?

After "gmake package" / "gmake repackage" you can inspect
   work/build-global/*.prototype*

> 2.   How do I specify the set of files to go into a subpackage, and  
> exclude those same files from the other
> subpackages?
>
> e.g.   if I have:
>
>    PKGFILES_CSWgraphviztcl = $(libexecdir_install)/graphviz/tcl/.*
>
> does that automatically exclude those matched files from the core  
> graphviz package?

Yes. This is documented at
   <http://sourceforge.net/apps/trac/gar/wiki/Prototypes>
You must also specify that you want to build multiple packages  
documented at
   <http://sourceforge.net/apps/trac/gar/wiki/DynamicPackageFiles>

You will usually not use *_install, as the installation goes to $ 
(libexecdir).
The use of *_install is mostly for internal things.

J?rgen has also spent some effort on a graphviz package and he'll help  
you
push the package out soon.


Best regards

   -- Dago

From william at wbonnet.net  Wed Jul 15 22:25:59 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 22:25:59 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <4A5DC986.3020606@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
	<4A5DC986.3020606@wbonnet.net>
Message-ID: <4A5E3B57.1030901@wbonnet.net>

Hi

> Not yet :( I'll give a look to this issue tonight. I was focused on 
> firefox compilation issues so far...
1/ Good news is that it seems to be fixed !

2/ Bad news is that i could have done it earlier... many many thanks to 
Sebastian for reminding me to process this important bug. After 
analyzing the bug, it was a real 'showstopper' i did not identified when 
it was reported.

3/ Really bad news is that the CSWdbus package that is available in 
"current" cannot be uninstalled in "normal mode" :( It means, it cannot 
be updated without a manual step... I do apologize for that... (and i do 
feel Phil's feet coming closer and closer from my back).

So what does this means :

I made a simple fix and updated the following packages : CSWdbus, 
CSWlibdbus and CSWlibdbusdev. The only package which have a real problem 
is CSWdbus. The update i made seems to fix the problem on my boxes at 
home. But it would appreciate to have some more feedbacks. Just to be 
sure i have fixed all the issues, in all possible way to make it occur.

The bug Sebastian outlined today is a related to dbus service stop. In 
most of the time it will fail. And it will fail 99.9% of the time during 
an update (either from pkg-get or pkgutil).  Current dbus is generating 
a pid "lock file" in /opt/csw/var/run/dbus/pid . This pid file is not 
processed correctly during service stop (and uninstall), and dbus will 
not stop.

I have updated packages and they are now available from testing. As i 
said, i was able to reproduce the problem before update and not after, 
but i would appreciate some more feedbacks.

If you plan to install update from testing you will have to kill by hand 
dbus process. dbus pid is stored in /opt/csw/var/run/dbus/pid . But 
unfortunately it is ignored by stop (or uninstall) script. So you just 
have to :

kill -9 `cat /opt/csw/var/run/dbus/pid`

before updating.

This has to be announced on thge user  list before any release. Phil, i 
do apologize for this, and please tell me the procedure i have to follow 
to announce it, if i am the one to announce 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 ihsan at opencsw.org  Wed Jul 15 22:34:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 15 Jul 2009 22:34:02 +0200
Subject: [csw-maintainers] syslog_ng 3.0.3 in testing
Message-ID: <4A5E3D3A.2000305@opencsw.org>

Hello,

Thanks to the great support of Maciej, I've updated syslog_ng to 3.0.3.

http://mirror.opencsw.org/testing/syslog_ng-3.0.3,REV=2009.07.15-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/syslog_ng-3.0.3,REV=2009.07.15-SunOS5.8-sparc-CSW.pkg.gz





Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From phil at bolthole.com  Wed Jul 15 22:40:18 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Jul 2009 13:40:18 -0700
Subject: [csw-maintainers] (dbus fixes)
In-Reply-To: <4A5E3B57.1030901@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
	<4A5DC986.3020606@wbonnet.net> <4A5E3B57.1030901@wbonnet.net>
Message-ID: <20090715204018.GA94635@bolthole.com>

On Wed, Jul 15, 2009 at 10:25:59PM +0200, William Bonnet wrote:
> This has to be announced on thge user  list before any release. Phil, i  
> do apologize for this, and please tell me the procedure i have to follow  
> to announce it, if i am the one to announce it.

yes you should be the one to announce it :-)

once it passes through testing, thenI would say as soon as I officialy
clear it through newpkgs, you should send emails to both the user, and
"announce" lists, mentioning what is required.


From phil at bolthole.com  Wed Jul 15 22:41:41 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Jul 2009 13:41:41 -0700
Subject: [csw-maintainers] howto control which files go into which
	part	of split package
In-Reply-To: <50628.217.227.36.102.1247689306.squirrel@ssl.skayser.de>
References: <4A5E00BC.3090800@opencsw.org>
	<50628.217.227.36.102.1247689306.squirrel@ssl.skayser.de>
Message-ID: <20090715204141.GB94635@bolthole.com>

On Wed, Jul 15, 2009 at 10:21:46PM +0200, Sebastian Kayser wrote:
> Hi John,
> 
> John Ellson wrote:
> > 1.   How can I list the file contents of a .pkg.gz to verify what I've
> > done?
> 
>   $ gunzip .pkg.gz
>   $ pkgtrans .pkg .


or if you are masochistic, and short on disk space:

PAGER=less gzmore x.pkg.gz

and look at the equivalent of the pkgmap file that way.


From william at wbonnet.net  Wed Jul 15 22:49:14 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 22:49:14 +0200
Subject: [csw-maintainers] (dbus fixes)
In-Reply-To: <20090715204018.GA94635@bolthole.com>
References: <4A5DA8EC.1080803@wbonnet.net>	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>	<4A5DC986.3020606@wbonnet.net>
	<4A5E3B57.1030901@wbonnet.net>
	<20090715204018.GA94635@bolthole.com>
Message-ID: <4A5E40CA.5070305@wbonnet.net>

Hi

> yes you should be the one to announce it :-)
>   
Master Yoda^W^WPhil, the one to announce it, will I be ;)

> once it passes through testing, thenI would say as soon as I officialy
> clear it through newpkgs, you should send emails to both the user, and
> "announce" lists, mentioning what is required.
>   
So thanks for more feedbacks then i'll request for a release.

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  Thu Jul 16 05:59:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Jul 2009 20:59:06 -0700
Subject: [csw-maintainers] Modifying SMF dependencies
In-Reply-To: <ce97543b0907150434p31370d05q46eb87e4675e7865@mail.gmail.com>
References: <ce97543b0907150434p31370d05q46eb87e4675e7865@mail.gmail.com>
Message-ID: <20090716035906.GA85493@bolthole.com>

On Wed, Jul 15, 2009 at 12:34:32PM +0100, Maciej (Matchek) Blizinski wrote:
 (I only work with Solaris 10, I fear asking about
> XML processing on Solaris 8...)

in solaris 8, you just put your demon "at the end", and cross your fingers
;-)

From bwalton at opencsw.org  Thu Jul 16 17:28:31 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 11:28:31 -0400
Subject: [csw-maintainers] cswclassutils
Message-ID: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>

Hi Peter,

I just added two new i/r scripts to the class utils package.  These
will automatically compile .el to .elc at install time.  If you're ok
with my changes, can you pop out a new release of cswclassutils?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/26c02ea9/attachment.asc>

From pfelecan at opencsw.org  Thu Jul 16 17:51:57 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 16 Jul 2009 17:51:57 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247758050-sup-1395@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 16 Jul 2009 11\:28\:31 -0400")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
Message-ID: <x6ws68wrqa.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

> Hi Peter,

I don't think that you refer to me... however I have some questions

> I just added two new i/r scripts to the class utils package.  These
> will automatically compile .el to .elc at install time.  If you're ok
> with my changes, can you pop out a new release of cswclassutils?

What's the utility of this knowing that .elc files are architectural
neutral?

Packages containing .el and ..elc are usually split such as the .el
files are in an optional package and ..elc in mandatory ones.

Reducing the payload is not, in my opinion, a sufficient reason.
-- 
Peter

From bwalton at opencsw.org  Thu Jul 16 18:21:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 12:21:15 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6ws68wrqa.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
Message-ID: <1247760420-sup-3897@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Thu Jul 16 11:51:57 -0400 2009:

> Reducing the payload is not, in my opinion, a sufficient reason.

I thought that payload reduction in addition to a standard method for
packaging was a good enough win.

Not all packages that provide .el files provide a make target to
transform them to .elc (or even install them at all, for that matter).
With these scripts, the maintainer at most has to ensure that the .el
files are part of the package...I'm planning to extend GAR to do the
prototype and dependency stuff automatically (similar to how
cswpycompile will be integrated).[1]

I prefer to have the .el version around for reading, but that's partly
due to me being a relatively new emacs user and wanting to see how
things are done.  Having to pull in a separate package to get the
source files would work for me, but it wouldn't be as nice.

If your approach is preferred, I'll revert my commit.

Thanks
-Ben

[1] Only if it's decided this is useful.  I assumed it was, but...
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/78bc3d92/attachment.asc>

From phil at bolthole.com  Thu Jul 16 18:28:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 16 Jul 2009 09:28:21 -0700
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247760420-sup-3897@ntdws12.chass.utoronto.ca>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
Message-ID: <20090716162821.GJ80367@bolthole.com>

On Thu, Jul 16, 2009 at 12:21:15PM -0400, Ben Walton wrote:
> Excerpts from Peter FELECAN's message of Thu Jul 16 11:51:57 -0400 2009:
> 
> > Reducing the payload is not, in my opinion, a sufficient reason.
> 
> I thought that payload reduction in addition to a standard method for
> packaging was a good enough win.

It should be mentioned that it can be a significant "payload reduction".
or.. erm.. it was in the case of python. Ben, why dont you give a specific
example of just how much space is saved?


Or, contrariwise, we may indeed want to only distribute the compiled
stuff and have separate src packages, as you say.

> I prefer to have the .el version around for reading, but that's partly
> due to me being a relatively new emacs user and wanting to see how
> things are done.

dont forget to mention/compare how debian and other distros do it.
not that we HAVE to do it the same way, but it's nice to have the
reference.


From pfelecan at opencsw.org  Thu Jul 16 18:44:58 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 16 Jul 2009 18:44:58 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <20090716162821.GJ80367@bolthole.com> (Philip Brown's message of
	"Thu\, 16 Jul 2009 09\:28\:21 -0700")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
Message-ID: <x6skgwwp9x.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Thu, Jul 16, 2009 at 12:21:15PM -0400, Ben Walton wrote:
>> Excerpts from Peter FELECAN's message of Thu Jul 16 11:51:57 -0400 2009:
>> 
>> > Reducing the payload is not, in my opinion, a sufficient reason.
>> 
>> I thought that payload reduction in addition to a standard method for
>> packaging was a good enough win.

I'm not opposed to having this class script for projects don't providing
make targets for .elc generation --- even if it's easy to patch for
that. BTW, can you provide an example of a project didn't providing the
direct or indirect elc target?

> It should be mentioned that it can be a significant "payload reduction".
> or.. erm.. it was in the case of python. Ben, why dont you give a specific
> example of just how much space is saved?

Quick example for abbrev:

.el  13226
.elc 11695

Note, however that I provide the .el files in compressed form

> Or, contrariwise, we may indeed want to only distribute the compiled
> stuff and have separate src packages, as you say.

This is how it's done for my packages containing emacs lisp code.

>> I prefer to have the .el version around for reading, but that's partly
>> due to me being a relatively new emacs user and wanting to see how
>> things are done.

This is why the .el are available in a separate, "source" package.

> dont forget to mention/compare how debian and other distros do it.
> not that we HAVE to do it the same way, but it's nice to have the
> reference.

Debian, and all the other distros that I know are doing it by providing
.elc in the main package and .el in a separate, optional package. What
they are doing in addition is implement a mechanism of recompiling
site-lisp packages on a package by package basis --- they have a emacs.d
directory containing a script by package installing in site-lisp; this
is something that is probably worth the effort to implement.

-- 
Peter

From bwalton at opencsw.org  Thu Jul 16 19:47:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 13:47:36 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6skgwwp9x.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
Message-ID: <1247764689-sup-557@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Thu Jul 16 12:44:58 -0400 2009:
> I'm not opposed to having this class script for projects don't providing
> make targets for .elc generation --- even if it's easy to patch for

Sure, it can be patched around, but that leaves the work to each
maintainer.

> that. BTW, can you provide an example of a project didn't providing the
> direct or indirect elc target?

Ruby is the example currently in my mind.  The misc/ directory
contains the ruby mode files, but nothing else.  As far as I can tell,
there is no make target to even get those files installed.  I don't
have an example where the .el files get installed but not
byte-compiled, so this leaves the maintainer doing work in either
case.

> Quick example for abbrev:
> 
> .el  13226
> .elc 11695

These line up roughly with what I've seen:
14452  inf-ruby.el
10008  inf-ruby.elc
 6747  ruby-electric.el
 6290  ruby-electric.elc
38653  ruby-mode.el
29260  ruby-mode.elc
 1746  ruby-style.el
 1983  ruby-style.elc
 4481  rubydb2x.el
 3952  rubydb2x.elc
 4613  rubydb3x.el
 4177  rubydb3x.elc

...so it's not huge (somewhere south of 50%), but it is something.  If
space were the ultimate goal, shipping .elc primarily and .el as a
source package would be better.  It was only part of my goal though.
Personally, I prefer the ship one, get both approach, but as I said,
I'll go with the preferred decision here.

> This is why the .el are available in a separate, "source" package.

This works, as I said, it's just not my preference.

> Debian, and all the other distros that I know are doing it by providing
> .elc in the main package and .el in a separate, optional package. What
> they are doing in addition is implement a mechanism of recompiling
> site-lisp packages on a package by package basis --- they have a emacs.d
> directory containing a script by package installing in site-lisp; this
> is something that is probably worth the effort to implement.

I don't think this is 'strictly' true.  It seems the preferred files
(for packages I looked at) were the .el ones.  There is then some
'fancy framework' for building/installing these for multiple flavours
of emacs, which can live together peacefully.  The debian policy[1]
doesn't seem to explicitly state which files are to be shipped
though.

Two packages I looked at quickly (from Ubuntu, not Debian, but...)
were css-mode and sepia.  In each case, they placed .el files in
/usr/share/emacs/site-lisp/ and then used a hook script in
/usr/lib/emacsen-common/packages/install/ to do byte compilation.

Thanks
-Ben

[1] http://www.debian.org/doc/packaging-manuals/debian-emacs-policy
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/9f74d3ba/attachment.asc>

From ja at opencsw.org  Thu Jul 16 19:55:57 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Thu, 16 Jul 2009 19:55:57 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
Message-ID: <op.uw59fh0kjbmpvo@imac.lan>

Hi all,

I updated the packages for Munin 1.3.4 in testing.

- sample apache config (ready for include) in /etc/opt/csw/munin/apache/
- munin plugins moved to /opt/csw/libexec
- fix for using monospaced font

Feedback is welcome.

Juergen

-- 
Juergen Arndt

From pfelecan at opencsw.org  Thu Jul 16 20:34:23 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 16 Jul 2009 20:34:23 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247764689-sup-557@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 16 Jul 2009 13\:47\:36 -0400")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com> <x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
Message-ID: <x6ocrkwk7k.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

>> Debian, and all the other distros that I know are doing it by providing
>> .elc in the main package and .el in a separate, optional package. What
>> they are doing in addition is implement a mechanism of recompiling
>> site-lisp packages on a package by package basis --- they have a emacs.d
>> directory containing a script by package installing in site-lisp; this
>> is something that is probably worth the effort to implement.
>
> I don't think this is 'strictly' true.  It seems the preferred files
> (for packages I looked at) were the .el ones.  There is then some
> 'fancy framework' for building/installing these for multiple flavours
> of emacs, which can live together peacefully.  The debian policy[1]
> doesn't seem to explicitly state which files are to be shipped
> though.

It's strictly true for Emacs and that is the reference package. All the
others are just optional components for which an Emacs savvy user doesn't
have issues installing for himself or site wide. Of course, a savvy
packager do better for his users.

As stated in my previous message I'm not opposed to it but wished to
know the rational. As I'm not satisfied with that I'll continue to
supply the .elc in my packages and separately the .el

tat sat
-- 
Peter

From bwalton at opencsw.org  Thu Jul 16 21:35:38 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 15:35:38 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6ocrkwk7k.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
Message-ID: <1247771967-sup-7519@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Thu Jul 16 14:34:23 -0400 2009:
> It's strictly true for Emacs and that is the reference package. All the
> others are just optional components for which an Emacs savvy user doesn't
> have issues installing for himself or site wide. Of course, a savvy
> packager do better for his users.

Ok.  I think we've both been clear on our preferences.  You have your
reasons and I'm ok with them.

> As stated in my previous message I'm not opposed to it but wished to
> know the rational. As I'm not satisfied with that I'll continue to
> supply the .elc in my packages and separately the .el

Ok.  I'm interested in standardizing something for this though.  It
would be better for the users if all of our emacs packaged bits
presented a similar experience.

Shall we agree then that any package providing emacs add-on
modules/functionality should provide .elc files in the primary package
and offer an _el version of the package to deliver the .el source
files?  That's what you've said, but should this become the formal
policy?

A side thought: Does it make sense to have this policy be different
than the python policy?  In theory, python add-ons could ship only the
.pyc/.pyo files and provide _py packages for the .py source...I'm
wondering this aloud for my own curiosity more than anything.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/c1ab3ba7/attachment.asc>

From william at wbonnet.net  Thu Jul 16 22:31:44 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 16 Jul 2009 22:31:44 +0200
Subject: [csw-maintainers] sqlite3 3.6.16 is in testing
Message-ID: <4A5F8E30.1070805@wbonnet.net>

Hi

Sqlite3 package has been updated to latest version 3.6.16 and it 
available from testing

http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3_devel-3.6.16,REV=2009.07.16-SunOS5.8-all-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz

feedbacks are welcomed

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  Fri Jul 17 06:02:13 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 16 Jul 2009 21:02:13 -0700
Subject: [csw-maintainers] patience with releases for a day or two
In-Reply-To: <4A5F8E30.1070805@wbonnet.net>
References: <4A5F8E30.1070805@wbonnet.net>
Message-ID: <20090717040213.GD31978@bolthole.com>

Please note: James and I are trying out some things with releasing
packages. so pending packages will most likely be delayed another day or
two. but should definately go out monday worst case.
possibly a lot  sooner

From dam at opencsw.org  Fri Jul 17 09:25:50 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 17 Jul 2009 09:25:50 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
Message-ID: <A387A12F-F80E-4E3D-A87A-50435EE4C410@opencsw.org>

Hi Ben,

Am 16.07.2009 um 17:28 schrieb Ben Walton:
> I just added two new i/r scripts to the class utils package.  These
> will automatically compile .el to .elc at install time.  If you're ok
> with my changes, can you pop out a new release of cswclassutils?

Should there be automatic GAR support for it? I could imagine the
following:

- Add PKGFILES_EL to strip out files
- Automatically strip .elc files and generate them on install,
   change class of these files


Best regards

   -- Dago

From pfelecan at opencsw.org  Fri Jul 17 11:55:32 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 17 Jul 2009 11:55:32 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247771967-sup-7519@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 16 Jul 2009 15\:35\:38 -0400")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com> <x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
Message-ID: <x6fxcvws4r.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

>> As stated in my previous message I'm not opposed to it but wished to
>> know the rational. As I'm not satisfied with that I'll continue to
>> supply the .elc in my packages and separately the .el
>
> Ok.  I'm interested in standardizing something for this though.  It
> would be better for the users if all of our emacs packaged bits
> presented a similar experience.
>
> Shall we agree then that any package providing emacs add-on
> modules/functionality should provide .elc files in the primary package
> and offer an _el version of the package to deliver the .el source
> files?  That's what you've said, but should this become the formal
> policy?

I think so.

> A side thought: Does it make sense to have this policy be different
> than the python policy?  In theory, python add-ons could ship only the
> .pyc/.pyo files and provide _py packages for the .py source...I'm
> wondering this aloud for my own curiosity more than anything.

I'm leaning toward supplying separate _py packages but not having a
vested interest I remain neutral.
-- 
Peter

From pfelecan at opencsw.org  Fri Jul 17 11:56:22 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 17 Jul 2009 11:56:22 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <A387A12F-F80E-4E3D-A87A-50435EE4C410@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 17 Jul 2009 09\:25\:50 +0200")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<A387A12F-F80E-4E3D-A87A-50435EE4C410@opencsw.org>
Message-ID: <x6bpnjws3d.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Ben,
>
> Am 16.07.2009 um 17:28 schrieb Ben Walton:
>> I just added two new i/r scripts to the class utils package.  These
>> will automatically compile .el to .elc at install time.  If you're ok
>> with my changes, can you pop out a new release of cswclassutils?
>
> Should there be automatic GAR support for it? I could imagine the
> following:
>
> - Add PKGFILES_EL to strip out files
> - Automatically strip .elc files and generate them on install,
>   change class of these files

I don't think so. Read the other messages of the thread.
-- 
Peter

From maciej at opencsw.org  Fri Jul 17 14:05:47 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 17 Jul 2009 13:05:47 +0100
Subject: [csw-maintainers] A single package that needs two source tarballs
Message-ID: <ce97543b0907170505i24ab5eb1y59d016809366bf41@mail.gmail.com>

I'm attempting to write a build for Netscape nss (network security
services). It turns out, nss source file needs nspr sources unpacked
in the same working directory. When attempting to compile nss, I'm
getting:

gmake[3]: Leaving directory
`/export/home/blizinski/opencsw/pkg/nss/trunk/work/build-isa-i386/nss-3.12.3/mozilla/security/coreconf'
gmake: *** No rule to make target ../../nsprpub/configure.  Stop.
gmake[2]: *** [../../nsprpub/configure] Error 1
gmake[2]: Leaving directory
`/export/home/blizinski/opencsw/pkg/nss/trunk/work/build-isa-i386/nss-3.12.3/mozilla/security/nss'
gmake[1]: *** [build-nss] Error 2
gmake[1]: Leaving directory `/export/home/blizinski/opencsw/pkg/nss/trunk'
gmake: *** [merge-isa-i386] Error 2

I've found the missing bit (nsprpub/configure) in the nspr package.
How would you go about it? Have a custom script to unpack nspr stuff
into the nss directory? Any other ideas?

From bwalton at opencsw.org  Fri Jul 17 14:17:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 17 Jul 2009 08:17:59 -0400
Subject: [csw-maintainers] A single package that needs two source
	tarballs
In-Reply-To: <ce97543b0907170505i24ab5eb1y59d016809366bf41@mail.gmail.com>
References: <ce97543b0907170505i24ab5eb1y59d016809366bf41@mail.gmail.com>
Message-ID: <1247832951-sup-209@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 17 08:05:47 -0400 2009:

Hi Maciej,

> I've found the missing bit (nsprpub/configure) in the nspr package.
> How would you go about it? Have a custom script to unpack nspr stuff
> into the nss directory? Any other ideas?

Have a look at the Makefile[1] for docbook-dtds.  It combines multiple
source tarballs into a single build directory.  You'll see how I
overrode the specific extract targets to put things where they need to
be.  I'm not sure if this will cover all of what you need for the nss
package, but it should get you started, anyway.

HTH.
-Ben

[1] http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/docbook-dtds/trunk/Makefile
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090717/e5c6a034/attachment.asc>

From benny at opencsw.org  Fri Jul 17 14:27:36 2009
From: benny at opencsw.org (Benjamin von Mossner)
Date: Fri, 17 Jul 2009 14:27:36 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
Message-ID: <20090717122736.GB63617@vonmossner.de>

Hi,

> >From what i know, Benny has been working on an updated version. He is at a
> customer event right now, but he should be able to tell you whether i was
> right or not later on today.

so finally i had some time to repackage xpdf with GAR.
Updated version can be found in testing, included a fix for 0002724 
and and the latest security patches.

Talking about the bug:
I am definatly not an libXt expert, so i dont know if my fix is
appropriate. In xpdf/XPDFCore.cc the call to XtVaSetValues is
done with the wrong(?) widget and therefor fails to size the
window correctly. My diff therefor (and apparently the fix) is
kind of trivial.
[snip]
-    XtVaSetValues(drawArea, XmNwidth, width, XmNheight, height, NULL);
+    XtVaSetValues(shell, XmNwidth, width, XmNheight, height, NULL);
[snap]
But as mentioned above, i dont know if this is the right place to
fix this issue, nor why the unpatched code works fine on *BSD and
linux.
So if there is anyone familiar with libXt and/or stumbled upon
this or similar issue, i would be eager to know :-)

Cheers,
benny
-- 
/"\ ASCII RIBBON CAMPAIGN                  | Benjamin von Mossner
\ / AGAINST HTML MAIL                      | benny at vonmossner.de
 X  
/ \ multiple exclamation marks are a sure sign of a diseased mind

From ihsan at opencsw.org  Fri Jul 17 17:09:00 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Fri, 17 Jul 2009 17:09:00 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
Message-ID: <4A60940C.5080506@opencsw.org>

Hello,

The current Libtool package is unfortunately not fully functional. That
means, that certain packages do not build anymore.

As far as I could figure out, Libtool seems not to be aware of all the
compilers that we are using on our buildfarm.

Because I'm not an Libtool expert, I would like to ask if anybody with
good Libtool knowledge could have a look into that. Dago placed
everything, including the patch that has been used in the old version,
into Gar.



Thanks & Greetings,
Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From phil at bolthole.com  Fri Jul 17 18:47:41 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 17 Jul 2009 09:47:41 -0700
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6fxcvws4r.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
	<x6fxcvws4r.fsf@abhavana.free.fr>
Message-ID: <20090717164741.GQ80367@bolthole.com>

On Fri, Jul 17, 2009 at 11:55:32AM +0200, Peter FELECAN wrote:
> > A side thought: Does it make sense to have this policy be different
> > than the python policy?  In theory, python add-ons could ship only the
> > .pyc/.pyo files and provide _py packages for the .py source...I'm
> > wondering this aloud for my own curiosity more than anything.


I think that python is a different language, with different mechanisms, so
its ok if it has different policy.

I think that the route being followed by the current python maintainer,
makes sense for python.


From phil at bolthole.com  Fri Jul 17 18:49:28 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 17 Jul 2009 09:49:28 -0700
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247771967-sup-7519@ntdws12.chass.utoronto.ca>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
Message-ID: <20090717164928.GR80367@bolthole.com>

On Thu, Jul 16, 2009 at 03:35:38PM -0400, Ben Walton wrote:
> ...
> Ok.  I'm interested in standardizing something for this though.  It
> would be better for the users if all of our emacs packaged bits
> presented a similar experience.

this sounds like a good thing !



> Shall we agree then that any package providing emacs add-on
> modules/functionality should provide .elc files in the primary package
> and offer an _el version of the package to deliver the .el source
> files?  That's what you've said, but should this become the formal
> policy?

Personally, I think that it could at least be our official
recommendation/guideline.  not sure we have to push it all the way to a
"policy".

From phil at bolthole.com  Fri Jul 17 18:57:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 17 Jul 2009 09:57:57 -0700
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <20090717122736.GB63617@vonmossner.de>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
	<20090717122736.GB63617@vonmossner.de>
Message-ID: <20090717165757.GT80367@bolthole.com>

On Fri, Jul 17, 2009 at 02:27:36PM +0200, Benjamin von Mossner wrote:
> But as mentioned above, i dont know if this is the right place to
> fix this issue, nor why the unpatched code works fine on *BSD and
> linux.
> So if there is anyone familiar with libXt and/or stumbled upon
> this or similar issue, i would be eager to know :-)
> 

well, you could try linking against our "new X11 libs", and see if that
magically fixes things.
(/opt/csw/X11/lib i think)

From pfelecan at opencsw.org  Fri Jul 17 19:29:56 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 17 Jul 2009 19:29:56 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
In-Reply-To: <4A60940C.5080506@opencsw.org> (Ihsan Dogan's message of "Fri\,
	17 Jul 2009 17\:09\:00 +0200")
References: <4A60940C.5080506@opencsw.org>
Message-ID: <x6hbxbusiz.fsf@abhavana.free.fr>

Ihsan Dogan <ihsan at opencsw.org> writes:

> The current Libtool package is unfortunately not fully functional. That
> means, that certain packages do not build anymore.
>
> As far as I could figure out, Libtool seems not to be aware of all the
> compilers that we are using on our buildfarm.
>
> Because I'm not an Libtool expert, I would like to ask if anybody with
> good Libtool knowledge could have a look into that. Dago placed
> everything, including the patch that has been used in the old version,
> into Gar.

Ihsan,

Can you give a concrete example such as we can try to help?

-- 
Peter

From bwalton at opencsw.org  Fri Jul 17 19:43:25 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 17 Jul 2009 13:43:25 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <20090717164928.GR80367@bolthole.com>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
	<20090717164928.GR80367@bolthole.com>
Message-ID: <1247852361-sup-8154@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 17 12:49:28 -0400 2009:
> Personally, I think that it could at least be our official
> recommendation/guideline.  not sure we have to push it all the way to a
> "policy".

Ok, then I'll withdraw (revert) my patch to cswclassutils.  Instead of
modifying GAR to automatically enable the use of the i/r scripts, I'll
look to add a toggle for compiling any .el files without a
corresponding .elc and generating the source packages automatically.
Something along the lines of GAR_EL_SOURCE_PKGS = 1.  Sound good to
everyone?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090717/32fbef08/attachment.asc>

From trygvis at opencsw.org  Fri Jul 17 20:32:13 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Fri, 17 Jul 2009 20:32:13 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <op.uw59fh0kjbmpvo@imac.lan>
References: <op.uw59fh0kjbmpvo@imac.lan>
Message-ID: <4A60C3AD.7050800@opencsw.org>

Juergen Arndt wrote:
> Hi all,
> 
> I updated the packages for Munin 1.3.4 in testing.
> 
> - sample apache config (ready for include) in /etc/opt/csw/munin/apache/
> - munin plugins moved to /opt/csw/libexec
> - fix for using monospaced font
> 
> Feedback is welcome.

I'll try these when I get back home next week.

--
Trygve

From ihsan at opencsw.org  Sat Jul 18 11:13:44 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sat, 18 Jul 2009 11:13:44 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
In-Reply-To: <x6hbxbusiz.fsf@abhavana.free.fr>
References: <4A60940C.5080506@opencsw.org> <x6hbxbusiz.fsf@abhavana.free.fr>
Message-ID: <4A619248.40203@opencsw.org>

Am 17.7.2009 19:29 Uhr, Peter FELECAN schrieb:

>> The current Libtool package is unfortunately not fully functional. That
>> means, that certain packages do not build anymore.
>>
>> As far as I could figure out, Libtool seems not to be aware of all the
>> compilers that we are using on our buildfarm.
>>
>> Because I'm not an Libtool expert, I would like to ask if anybody with
>> good Libtool knowledge could have a look into that. Dago placed
>> everything, including the patch that has been used in the old version,
>> into Gar.
>
> Can you give a concrete example such as we can try to help?

I'm getting this:

/opt/csw/bin/libtool --mode=compile /opt/csw/gcc4/bin/gcc  -O2 -pipe
-mcpu=v8 -I/opt/csw/include -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
-Wall -D_GNU_SOURCE -DNDEBUG -D_LIBRADIUS
-I/home/ihsan/gar/csw/mgar/pkg/freeradius/trunk/work/build-isa-sparcv8/freeradius-server-2.1.6/src
-c dict.c
libtool: compile: unable to infer tagged configuration
libtool: compile: specify a tag with `--tag'
gmake[6]: *** [dict.lo] Error 1



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From pfelecan at opencsw.org  Sat Jul 18 11:36:39 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Sat, 18 Jul 2009 11:36:39 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
In-Reply-To: <4A619248.40203@opencsw.org> (Ihsan Dogan's message of "Sat\,
	18 Jul 2009 11\:13\:44 +0200")
References: <4A60940C.5080506@opencsw.org> <x6hbxbusiz.fsf@abhavana.free.fr>
	<4A619248.40203@opencsw.org>
Message-ID: <x68wimuyc8.fsf@abhavana.free.fr>

Ihsan Dogan <ihsan at opencsw.org> writes:

> Am 17.7.2009 19:29 Uhr, Peter FELECAN schrieb:
>
>>> The current Libtool package is unfortunately not fully functional. That
>>> means, that certain packages do not build anymore.
>>>
>>> As far as I could figure out, Libtool seems not to be aware of all the
>>> compilers that we are using on our buildfarm.
>>>
>>> Because I'm not an Libtool expert, I would like to ask if anybody with
>>> good Libtool knowledge could have a look into that. Dago placed
>>> everything, including the patch that has been used in the old version,
>>> into Gar.
>>
>> Can you give a concrete example such as we can try to help?
>
> I'm getting this:
>
> /opt/csw/bin/libtool --mode=compile /opt/csw/gcc4/bin/gcc  -O2 -pipe
> -mcpu=v8 -I/opt/csw/include -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
> -Wall -D_GNU_SOURCE -DNDEBUG -D_LIBRADIUS
> -I/home/ihsan/gar/csw/mgar/pkg/freeradius/trunk/work/build-isa-sparcv8/freeradius-server-2.1.6/src
> -c dict.c
> libtool: compile: unable to infer tagged configuration
> libtool: compile: specify a tag with `--tag'
> gmake[6]: *** [dict.lo] Error 1

A typical invocation is:

libtool --tag=CC   --mode=compile gcc -std=gnu99 -DHAVE_CONFIG_H
-I. -I.. -I..  -I/opt/csw/include  -I/opt/csw/gcc3/include
-I/opt/csw/gcc3/include/c++/3.4.5  -I/opt/csw/include   -O3 -march=i486
-MT snv.lo -MD -MP -MF .deps/snv.Tpo -c -o snv.lo snv.c

Please note in my example the --tag option which is missing in your
invocation. Can you try with --tag=CC

>From the info manual of libtool we can read:

    6.2 Tags
    ========
    
    Libtool supports multiple languages through the use of tags.
    Technically a tag corresponds to a set of configuration variables
    associated with a language.  These variables tell `libtool' how it
    should create objects and libraries for each language.
    
       Tags are defined at `configure'-time for each language activated in
    the package (see `LT_LANG' in *note LT_INIT::).  Here is the
    correspondence between language names and tags names.
    
    Language name      Tag name
    C                  CC
    C++                CXX
    Java               GCJ
    Fortran 77         F77
    Fortran            FC
    Windows Resource   RC
    
       `libtool' tries to automatically infer which tag to use from the
    compiler command being used to compile or link.  If it can't infer a
    tag, then it defaults to the configuration for the `C' language.
    
       The tag can also be specified using `libtool''s `--tag=TAG' option
    (*note Invoking libtool::).  It is a good idea to do so in `Makefile'
    rules, because that will allow users to substitute the compiler without
    relying on `libtool' inference heuristics.  When no tag is specified,
    `libtool' will default to `CC'; this tag always exists.
    
       Finally, the set of tags available in a particular project can be
    retrieved by tracing for the `LT_SUPPORTED_TAG' macro (*note Trace
    interface::).

What is intriguing is that the default tag is CC and libtool complains
that you are not using it explicitly. For this reason I would try to
use the /opt/csw/gcc4/bin in the PATH environment variable placed as to
be the first component refering toward a C compiler in the
configuration step.
-- 
Peter

From rupert at opencsw.org  Sun Jul 19 15:45:16 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 19 Jul 2009 15:45:16 +0200
Subject: [csw-maintainers] sqlite3 3.6.16 is in testing
In-Reply-To: <4A5F8E30.1070805@wbonnet.net>
References: <4A5F8E30.1070805@wbonnet.net>
Message-ID: <6af4270907190645p44ab966g86e01c57d9207b79@mail.gmail.com>

works for us.

On Thu, Jul 16, 2009 at 22:31, William Bonnet<william at wbonnet.net> wrote:
> Hi
>
> Sqlite3 package has been updated to latest version 3.6.16 and it available
> from testing
>
> http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3_devel-3.6.16,REV=2009.07.16-SunOS5.8-all-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz
>
> feedbacks are welcomed
>
> 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
>
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>

From rupert at opencsw.org  Sun Jul 19 16:11:19 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 19 Jul 2009 16:11:19 +0200
Subject: [csw-maintainers] remove openldap not possible with pkg-get and
	pkgutil
In-Reply-To: <94E2A58978FC324196A142DE8399AB7802315343@chsa1556.share.beluni.net>
References: <94E2A58978FC324196A142DE8399AB7802315343@chsa1556.share.beluni.net>
Message-ID: <6af4270907190711jdc8fe14s4c80318218688272@mail.gmail.com>

i was testing the new source code packages, and i did not pay enough
attention when upgrading. after doing a

pkgutil -t http://mirror.opencsw.org/opencsw/testing -i subversion
pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial

subversion was insisting on finding /opt/csw/lib/libldap-2.3.so.0,
while only /opt/csw/lib/libldap-2.4.so.0 was available. somehow
openldap got upgraded to
openldap_rt-2.4.16,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz from
testing.

i tried to remove it or downgrade it, but it did not work with
pkgutil, and pkg-get.

# pkgutil -r CSWoldaprt
Parsing catalog, may take a while...
Looking at installed CSW packages...
Examining dependencies for:
?CSWoldaprt (in use)
?CSWlibnet (in use)
?CSWsasl (in use)
?CSWossl (remove)
?CSWbdb4 (remove)
?CSWossldevel (remove)
?CSWosslutils (remove)
?CSWosslrt (in use)
?CSWcacertificates (remove)
?CSWcswclassutils (in use)
?CSWcommon (in use)
5 packages to remove. Do you want to continue? [Y,n]


# pkg-get -r CSWoldaprt

WARNING: ?the remove option is not very intelligent.
If there are multiple versions of a package with the same
PKG style name, it will remove the first one it can
(will continue in 5 seconds)
Starting remove operations now...

ERROR: no matching SysV PKG found.
CSWoldaprt CSWoldaprt
(either you mistyped it, or you need to updatecatalog, or
?it isnt available for your OSREV yet)

then i used pkgrm and pkgutil -u to install it again and it worked.


so i wonder:
* why oldap gets upgraded
* how to remove / downgrade packages

rupert.

From rupert at opencsw.org  Sun Jul 19 16:37:33 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 19 Jul 2009 16:37:33 +0200
Subject: [csw-maintainers] how to choose the correct version on testing?
Message-ID: <6af4270907190737gbbf341fx81087eb901ad038@mail.gmail.com>

i tried to test mercurial-1.3, and did
  pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u mercurial
  pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial
like described on http://mirror.opencsw.org/testing.html.

but both insisted installing the verison 1.2.1 which is also there.

is there an option to force it to take the newer one?

rupert.

From dam at opencsw.org  Sun Jul 19 17:35:36 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Jul 2009 17:35:36 +0200
Subject: [csw-maintainers] remove openldap not possible with pkg-get and
	pkgutil
In-Reply-To: <6af4270907190711jdc8fe14s4c80318218688272@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7802315343@chsa1556.share.beluni.net>
	<6af4270907190711jdc8fe14s4c80318218688272@mail.gmail.com>
Message-ID: <ECD4EFD9-BAB8-48F8-8E82-234A1598F4DD@opencsw.org>

Hi Rupert,

Am 19.07.2009 um 16:11 schrieb rupert THURNER:
> i was testing the new source code packages, and i did not pay enough
> attention when upgrading. after doing a
>
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -i subversion
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial
>
> subversion was insisting on finding /opt/csw/lib/libldap-2.3.so.0,
> while only /opt/csw/lib/libldap-2.4.so.0 was available. somehow
> openldap got upgraded to
> openldap_rt-2.4.16,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz from
> testing.

Well, if you install from testing pkgutil thinks that *all* packages  
from testing
within the dependency tree should be updated. However, OpenLDAP is not
finished yet and was put there for some very specific tests and not for
general use.

> i tried to remove it or downgrade it, but it did not work with
> pkgutil, and pkg-get.
>
> # pkgutil -r CSWoldaprt
> Parsing catalog, may take a while...
> Looking at installed CSW packages...
> Examining dependencies for:
>  CSWoldaprt (in use)
>  CSWlibnet (in use)
>  CSWsasl (in use)
>  CSWossl (remove)
>  CSWbdb4 (remove)
>  CSWossldevel (remove)
>  CSWosslutils (remove)
>  CSWosslrt (in use)
>  CSWcacertificates (remove)
>  CSWcswclassutils (in use)
>  CSWcommon (in use)
> 5 packages to remove. Do you want to continue? [Y,n]
>
>
> # pkg-get -r CSWoldaprt
>
> WARNING:  the remove option is not very intelligent.
> If there are multiple versions of a package with the same
> PKG style name, it will remove the first one it can
> (will continue in 5 seconds)
> Starting remove operations now...
>
> ERROR: no matching SysV PKG found.
> CSWoldaprt CSWoldaprt
> (either you mistyped it, or you need to updatecatalog, or
>  it isnt available for your OSREV yet)
>
> then i used pkgrm and pkgutil -u to install it again and it worked.
>
>
> so i wonder:
> * why oldap gets upgraded

Please see my comment above.

> * how to remove / downgrade packages

In this case you can try to force-sync with -S from pkg-get.


Best regards and sorry for the inconvenience

   -- Dago



From dam at opencsw.org  Sun Jul 19 17:37:20 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Jul 2009 17:37:20 +0200
Subject: [csw-maintainers] how to choose the correct version on testing?
In-Reply-To: <6af4270907190737gbbf341fx81087eb901ad038@mail.gmail.com>
References: <6af4270907190737gbbf341fx81087eb901ad038@mail.gmail.com>
Message-ID: <7D15FCF1-6D46-4D6F-821A-6AC75E6735D2@opencsw.org>

Hi Rupert,

Am 19.07.2009 um 16:37 schrieb rupert THURNER:
> i tried to test mercurial-1.3, and did
>  pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u mercurial
>  pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial
> like described on http://mirror.opencsw.org/testing.html.
>
> but both insisted installing the verison 1.2.1 which is also there.
>
> is there an option to force it to take the newer one?

No, because only the latest version is in the catalog. The maintainer  
should
remove the old one. Looks like both versions are from you :-) Do you  
really
need both versions in testing? The usual use is to only put the latest  
in
testing/.


Best regards

   -- Dago

From william at wbonnet.net  Sun Jul 19 18:43:08 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 19 Jul 2009 18:43:08 +0200
Subject: [csw-maintainers] sqlite3 3.6.16 is in testing
In-Reply-To: <6af4270907190645p44ab966g86e01c57d9207b79@mail.gmail.com>
References: <4A5F8E30.1070805@wbonnet.net>
	<6af4270907190645p44ab966g86e01c57d9207b79@mail.gmail.com>
Message-ID: <4A634D1C.5070802@wbonnet.net>

Hi Rupert
> works for us.
>   
Thanks for the feed back. So far it is also working for me. I'll push it 
to current in a couple days if i have no negative feedbacks

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  Sun Jul 19 22:11:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Jul 2009 22:11:07 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <CA5A04C4-192E-4ABF-A62B-CDEEBD41672B@opencsw.org>

Hi J?rgen,

Am 06.07.2009 um 14:25 schrieb Juergen Arndt:
> I've put Munin 1.3.4 into testing. It's the initial version for  
> OpenCSW. The software consists of the packages munin_master and  
> munin_node.

Please make sure the package is ARCHALL=1.


Best regards

   -- Dago

From dam at opencsw.org  Mon Jul 20 22:03:58 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 20 Jul 2009 22:03:58 +0200
Subject: [csw-maintainers] [csw-users] Fwd: Problem with pidgin and X11
	libs
In-Reply-To: <4A64C3E0.7010106@opencsw.org>
References: <4A646BC7.70007@gmail.com>
	<5BF57A34-6C08-46AA-AAB6-6EF7D9B0B48B@opencsw.org>
	<4A64C3E0.7010106@opencsw.org>
Message-ID: <2839A0F1-2EFB-445B-AD92-5D45E527DA41@opencsw.org>

Hi Mike,

Am 20.07.2009 um 21:22 schrieb Mike Watters:
> Chad asked me to build the new version of Pidgin ( the one that  
> fixes yahoo )
> since he does not have the bandwidth at the moment.  I have been  
> waiting on
> gtk+ to be released.  ( I am running the latest version on the gtk+  
> compiled
> from testing with NO problems )

Both gtk+ and devel are now updated on all buildfarm machines.


Best regards

   -- Dago

From rupert at opencsw.org  Tue Jul 21 00:33:01 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 21 Jul 2009 00:33:01 +0200
Subject: [csw-maintainers] mercurial package broken
In-Reply-To: <94E2A58978FC324196A142DE8399AB7802315C23@chsa1556.share.beluni.net>
References: <AcoIiGFjfLdQxjwaRZKhRw38hdYPrA==>
	<94E2A58978FC324196A142DE8399AB7802315C23@chsa1556.share.beluni.net>
Message-ID: <6af4270907201533j44108656u2739deab38d1fc13@mail.gmail.com>

the mercurial-1.3 package seems broken, we get here:

[Sun Jul 19 17:29:03 2009] [error] [client 166.14.3.11] RuntimeError:
No hgweb templates found in []
[Sun Jul 19 17:29:04 2009] [info] mod_wsgi (pid=18239): Create
interpreter 'chvp011vs035.csintra.net|/hg'.
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Target WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi' cannot be loaded as Python
module.
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'.
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] Traceback
(most recent call last):
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] ? File
"/opt/csw/apache2/data/hg09/hgwebdir.wsgi", line 3, in <module>
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] ? ? from
mercurial.hgweb.hgwebdir_mod import hgwebdir
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] ImportError:
No module named mercurial.hgweb.hgwebdir_mod
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Target WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi' cannot be loaded as Python
module., referer: https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'., referer:
https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] Traceback
(most recent call last):, referer: https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] ? File
"/opt/csw/apache2/data/hg09/hgwebdir.wsgi", line 3, in <module>,
referer: https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] ? ? from
mercurial.hgweb.hgwebdir_mod import hgwebdir, referer:
https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] ImportError:
No module named mercurial.hgweb.hgwebdir_mod, referer:
https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] mod_wsgi
(pid=18240): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'.
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] Traceback
(most recent call last):
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/request.py", line
133, in run_wsgi
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 104, in __call__
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 134, in run_wsgi
[Sun Jul 19 17:30:51 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 323, in templater
[Sun Jul 19 17:30:51 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/templater.py", line 245,
in stylemap
[Sun Jul 19 17:30:51 2009] [error] [client 169.57.129.251]
RuntimeError: No hgweb templates found in []
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] mod_wsgi
(pid=25178): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'.
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] Traceback
(most recent call last):
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/request.py", line
133, in run_wsgi
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 104, in __call__
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 134, in run_wsgi
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 323, in templater
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/templater.py", line 245,
in stylemap
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251]
RuntimeError: No hgweb templates found in []

but an easy_install mercurial works.

rupert.

From bwalton at opencsw.org  Tue Jul 21 00:47:44 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 20 Jul 2009 18:47:44 -0400
Subject: [csw-maintainers] mercurial package broken
In-Reply-To: <6af4270907201533j44108656u2739deab38d1fc13@mail.gmail.com>
References: <AcoIiGFjfLdQxjwaRZKhRw38hdYPrA==>
	<94E2A58978FC324196A142DE8399AB7802315C23@chsa1556.share.beluni.net>
	<6af4270907201533j44108656u2739deab38d1fc13@mail.gmail.com>
Message-ID: <1248130010-sup-5486@ntdws12.chass.utoronto.ca>

Excerpts from rupert THURNER's message of Mon Jul 20 18:33:01 -0400 2009:

> '/opt/csw/apache2/data/hg09/hgwebdir.wsgi' cannot be loaded as Python
> module.

This likely needs to be rerolled after the python version update...not
positive, as I don't use it, but that's where I'd look 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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090720/4e98db6a/attachment.asc>

From bwalton at opencsw.org  Tue Jul 21 02:30:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 20 Jul 2009 20:30:01 -0400
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
	<E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
Message-ID: <1248135892-sup-3787@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Jul 12 09:23:02 -0400 2009:
Hi Sebastian and Dago,

I put together a small patch (attached) for GAR to twiddle the
required bits for use of cswpycompile in an automatic fashion.  I
didn't make it happen implicitly, as after thinking more about the
changes I'll make to support automatic elisp handling, I didn't think
implicit was the right way to go here.

The patch will take any value of CSWPYCOMPILE to mean that
cswpycompile is added to SPKG_CLASSES, .pyc and .pyo files are
filtered from the merge by default, .py files are set to cswpycompile
in the prototype and CSWcswclassutils is added as a dependency.

If you're ok with the changes, I'll make the commit and document the
new variable.

> > EXTRA_PAX_ARGS = -s ',^.*/\(xcbgen/.*\.py\),.$(BUILD_PREFIX)/lib/ 
> > python/site-packages/\1,'
> > EXTRA_PAX_ARGS += -s ',^\.$(libdir)/python.*,,'

Dago: I didn't alter any of the PAX stuff for this patch, since it
didn't seem to be necessary...is this part of your recipe due to the
file relocation you're performing here?

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: autopy.patch
Type: application/octet-stream
Size: 2068 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090720/cc89a191/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090720/cc89a191/attachment.asc>

From dam at opencsw.org  Tue Jul 21 09:43:53 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 21 Jul 2009 09:43:53 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <1248135892-sup-3787@ntdws12.chass.utoronto.ca>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
	<E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
	<1248135892-sup-3787@ntdws12.chass.utoronto.ca>
Message-ID: <865A7E17-7BE5-4EBB-AC06-48DD53F48E20@opencsw.org>

Hi Ben,

Am 21.07.2009 um 02:30 schrieb Ben Walton:
> Excerpts from Dagobert Michelsen's message of Sun Jul 12 09:23:02  
> -0400 2009:
> Hi Sebastian and Dago,
>
> I put together a small patch (attached) for GAR to twiddle the
> required bits for use of cswpycompile in an automatic fashion.  I
> didn't make it happen implicitly, as after thinking more about the
> changes I'll make to support automatic elisp handling, I didn't think
> implicit was the right way to go here.
>
> The patch will take any value of CSWPYCOMPILE to mean that
> cswpycompile is added to SPKG_CLASSES, .pyc and .pyo files are
> filtered from the merge by default, .py files are set to cswpycompile
> in the prototype and CSWcswclassutils is added as a dependency.

It should also make sure that the filtered files are all in locations
known to Python compilation. See my comment below.

> If you're ok with the changes, I'll make the commit and document the
> new variable.
>
>>> EXTRA_PAX_ARGS = -s ',^.*/\(xcbgen/.*\.py\),.$(BUILD_PREFIX)/lib/
>>> python/site-packages/\1,'
>>> EXTRA_PAX_ARGS += -s ',^\.$(libdir)/python.*,,'
>
> Dago: I didn't alter any of the PAX stuff for this patch, since it
> didn't seem to be necessary...is this part of your recipe due to the
> file relocation you're performing here?

Yes, because the original location was not checked for compilation. The
pax args relocate from the X11 directory to the standard location for
Python files. It would IMHO be better to read the path in the class
script and compile what it got, but as I understand it that was too
slow.


Best regards

   -- Dago

From maciej at opencsw.org  Tue Jul 21 11:59:16 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 21 Jul 2009 10:59:16 +0100
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
Message-ID: <ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>

Hi Markus,

Do you mind building cups-1.3.11? I've submitted a version bump and a
small bugfix. Can you please build the package?

Here's the mantis issue:
http://www.opencsw.org/bugtrack/view.php?id=3768

Reason for the request: version 1.3.11 contains a bugfix for the
scheduler code. In 1.3.9 cups-polld gets stuck after some time.

To other maintainers: I've filed a bug with mantis, Markus is marked
as active maintainer, but there is a number of unresolved issues in
mantis. When did you last hear from Markus?

Maciej

From bonivart at opencsw.org  Tue Jul 21 12:31:39 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Tue, 21 Jul 2009 12:31:39 +0200
Subject: [csw-maintainers] Just back from vacation (sort of...)
Message-ID: <625385e30907210331n650bb1a2l423d23cdb6acc91d@mail.gmail.com>

Just home after being a mechanic at the MotoGP roadracing event at
Sachsenring, Germany. Will start processing all mail now.

-- 
/peter

From maciej at opencsw.org  Tue Jul 21 14:52:04 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 21 Jul 2009 13:52:04 +0100
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
Message-ID: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>

Suppose we have this build:

PACKAGES = CSWfoo CSWbar
PKGFILES_CSWfoo = /opt/csw/bin/foo
PKGFILES_CSWbar = /opt/csw/bin/.*

There are two files:

/opt/csw/bin/foo
/opt/csw/bin/bar

Is the outcome of such build specified or unspecified? If specified,
what is the order of resolution? First wins, last wins? What's the
order?

Maciej

From dam at opencsw.org  Tue Jul 21 15:19:57 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 21 Jul 2009 15:19:57 +0200
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
In-Reply-To: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
References: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
Message-ID: <04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>

Hi Maciej,

Am 21.07.2009 um 14:52 schrieb Maciej (Matchek) Blizinski:
> Suppose we have this build:
>
> PACKAGES = CSWfoo CSWbar
> PKGFILES_CSWfoo = /opt/csw/bin/foo
> PKGFILES_CSWbar = /opt/csw/bin/.*
>
> There are two files:
>
> /opt/csw/bin/foo
> /opt/csw/bin/bar
>
> Is the outcome of such build specified or unspecified? If specified,
> what is the order of resolution? First wins, last wins? What's the
> order?

In parallel :-) The path must match include and not exclude.
The decision is shortcut. All pathes included in a package are
excluded from all other packages. That means IIRC the package
CSWfoo would contain 'foo' and CSWbar would contain 'foo' and
'bar'. The mechanism was written to simplify saying explicitly
what you want in what package and exclude stuff from other
packages. It was not meant for funky include-but-not-excluded-
if-unless-else :-) These are however Perl-regexps, so it is
allowed to write all advanced REs from Perl like lookahead
or negative matches. Something like this should do what you
want:

> PKGFILES_CSWfoo = /opt/csw/bin/foo
> PKGFILES_CSWbar = /opt/csw/bin/(?!foo)

This should be described here:
   <https://sourceforge.net/apps/trac/gar/wiki/Prototypes>
If there is something unclear please let me know.

See perlre(1) or "Mastering Regular Expressions" for further
reading (or post of course :-)

Here is the interesting expression:
   http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/gar.pkg.mk#L322 
 >


Best regards

   -- Dago

From maciej at opencsw.org  Tue Jul 21 16:19:13 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 21 Jul 2009 15:19:13 +0100
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
In-Reply-To: <04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>
References: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
	<04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>
Message-ID: <ce97543b0907210719u1854f96i8a70399636b589c3@mail.gmail.com>

On Tue, Jul 21, 2009 at 2:19 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
>> Is the outcome of such build specified or unspecified? If specified,
>> what is the order of resolution? First wins, last wins? What's the
>> order?
>
> In parallel :-) The path must match include and not exclude.
> The decision is shortcut. All pathes included in a package are
> excluded from all other packages. That means IIRC the package
> CSWfoo would contain 'foo' and CSWbar would contain 'foo' and
> 'bar'.

If I understand correctly, when CSWbar contains both bin/foo and
bin/bar, paths included in a package aren't excluded from other
packages -- foo is included in both packages.

> The mechanism was written to simplify saying explicitly
> what you want in what package and exclude stuff from other
> packages.

I'm currently working with the cups package. There are two packages:
cupsd and cupsclient. Binaries for those packages are put into the
same paths (/opt/csw/bin, /opt/csw/sbin), so one needs to cherrypick
the right binaries. I hoped to be able to say: take this and that for
cupsd, the rest is cupsclient. If they're processed in parallel, I
need to explicitly list all the binaries then. Is that right? Do you
have any other ideas?

Current prototypes of the mentioned packages:

http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/cups/trunk/files/CSWcupsd.prototype
http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/cups/trunk/files/CSWcupsclient.prototype

Maciej

From dam at opencsw.org  Tue Jul 21 16:28:32 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 21 Jul 2009 16:28:32 +0200
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
In-Reply-To: <ce97543b0907210719u1854f96i8a70399636b589c3@mail.gmail.com>
References: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
	<04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>
	<ce97543b0907210719u1854f96i8a70399636b589c3@mail.gmail.com>
Message-ID: <094D5D6E-9120-4A5F-B67F-2A96B4D0FEE5@opencsw.org>

Hi Maciej,

Am 21.07.2009 um 16:19 schrieb Maciej (Matchek) Blizinski:
> On Tue, Jul 21, 2009 at 2:19 PM, Dagobert Michelsen<dam at opencsw.org>  
> wrote:
>>> Is the outcome of such build specified or unspecified? If specified,
>>> what is the order of resolution? First wins, last wins? What's the
>>> order?
>>
>> In parallel :-) The path must match include and not exclude.
>> The decision is shortcut. All pathes included in a package are
>> excluded from all other packages. That means IIRC the package
>> CSWfoo would contain 'foo' and CSWbar would contain 'foo' and
>> 'bar'.
>
> If I understand correctly, when CSWbar contains both bin/foo and
> bin/bar, paths included in a package aren't excluded from other
> packages -- foo is included in both packages.

All files which match PKGFILES go into that package, inclusion
takes precedence. In your case that is foo and bar for CSWbar.

>> The mechanism was written to simplify saying explicitly
>> what you want in what package and exclude stuff from other
>> packages.
>
> I'm currently working with the cups package. There are two packages:
> cupsd and cupsclient. Binaries for those packages are put into the
> same paths (/opt/csw/bin, /opt/csw/sbin), so one needs to cherrypick
> the right binaries. I hoped to be able to say: take this and that for
> cupsd, the rest is cupsclient. If they're processed in parallel, I
> need to explicitly list all the binaries then. Is that right? Do you
> have any other ideas?

You usually specify PKGFILES for all but one packages which gets all
the rest not included in any package. And you can say
   PKGFILES_CSWcupsd = $(sbindir)/(cupsd|cupsfilter)
and leave out PKGFILES for CSWcupsclient.


Best regards

   -- Dago

From trygvis at opencsw.org  Tue Jul 21 20:47:20 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 21 Jul 2009 20:47:20 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <4A60C3AD.7050800@opencsw.org>
References: <op.uw59fh0kjbmpvo@imac.lan> <4A60C3AD.7050800@opencsw.org>
Message-ID: <4A660D38.7030600@opencsw.org>

Trygve Laugst?l wrote:
> Juergen Arndt wrote:
>> Hi all,
>>
>> I updated the packages for Munin 1.3.4 in testing.
>>
>> - sample apache config (ready for include) in /etc/opt/csw/munin/apache/
>> - munin plugins moved to /opt/csw/libexec
>> - fix for using monospaced font
>>
>> Feedback is welcome.
> 
> I'll try these when I get back home next week.

Can you do a rebuild with ARCH=all so I can test them on my x86 box?

--
Trygve

From ja at opencsw.org  Tue Jul 21 20:56:04 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Tue, 21 Jul 2009 20:56:04 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <4A660D38.7030600@opencsw.org>
References: <op.uw59fh0kjbmpvo@imac.lan> <4A60C3AD.7050800@opencsw.org>
	<4A660D38.7030600@opencsw.org>
Message-ID: <op.uxfljpnjjbmpvo@imac.lan>

On Tue, 21 Jul 2009 20:47:20 +0200, Trygve Laugst?l <trygvis at opencsw.org>  
wrote:

> Trygve Laugst?l wrote:
>> Juergen Arndt wrote:
>>> Hi all,
>>>
>>> I updated the packages for Munin 1.3.4 in testing.
>>>
>>> - sample apache config (ready for include) in  
>>> /etc/opt/csw/munin/apache/
>>> - munin plugins moved to /opt/csw/libexec
>>> - fix for using monospaced font
>>>
>>> Feedback is welcome.
>>  I'll try these when I get back home next week.
>
> Can you do a rebuild with ARCH=all so I can test them on my x86 box?

In testing there are platform independent packages:

munin_master-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz
munin_node-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz

Juergen

-- 
Juergen Arndt

From william at wbonnet.net  Tue Jul 21 22:30:18 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 21 Jul 2009 22:30:18 +0200
Subject: [csw-maintainers] Just back from vacation (sort of...)
In-Reply-To: <625385e30907210331n650bb1a2l423d23cdb6acc91d@mail.gmail.com>
References: <625385e30907210331n650bb1a2l423d23cdb6acc91d@mail.gmail.com>
Message-ID: <4A66255A.2060405@wbonnet.net>

Hi Peter
> Just home after being a mechanic at the MotoGP roadracing event at
> Sachsenring, Germany. 
Mechanic at Sachsenring Moto GP event ... ? Waaaooohhhhh !

Impressing ! Full respect :)

Which team ? pilot ? A french guy is in the Honda team this year. He is 
not doing so well... but at least he takes part :)

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 Jul 22 03:53:46 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 21 Jul 2009 21:53:46 -0400
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <865A7E17-7BE5-4EBB-AC06-48DD53F48E20@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
	<E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
	<1248135892-sup-3787@ntdws12.chass.utoronto.ca>
	<865A7E17-7BE5-4EBB-AC06-48DD53F48E20@opencsw.org>
Message-ID: <1248227258-sup-2329@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Tue Jul 21 03:43:53 -0400 2009:

> Python files. It would IMHO be better to read the path in the class
> script and compile what it got, but as I understand it that was too
> slow.

Ok, having now read (more carefully) i.cswpycompile, I wonder if there
is a compromise that could be reached.  It seems to me that if the
class action script collected a set of directories (using `dirname
$dst`) it could then call compileall.py for each directory.

Mike, since you wrote this, what are your thoughts on that?  I ask
because I don't think writing generic pax rules for relocating .py
files into the standard location is either easy or wise in some cases.
Think of a package with the location of a .py file (or files) hard
coded somewhere after ./configure.  I'm sure that's an edge case
but...

If Mike doesn't think compileall.py per dir is feasible, I'd suggest
that we limit this 'automatic' functionality to things already in the
proper directory.  This would leave the maintainer to handle the
relocation however they'd like.  Alternately we could provide a knob
for users to specify pax rules in a more generic fashion.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090721/421b7fc9/attachment.asc>

From bwalton at opencsw.org  Wed Jul 22 03:56:19 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 21 Jul 2009 21:56:19 -0400
Subject: [csw-maintainers] maintainers near toronto?
Message-ID: <1248227645-sup-7242@ntdws12.chass.utoronto.ca>


Hi All,

Are there any other maintainers in the Toronto area?  I've learned
that Damian Conway is giving a free talk at UofT on Monday.  If you're
out there and interested, ping me and I'll forward you the details.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090721/8ac7e8b5/attachment.asc>

From skayser at opencsw.org  Wed Jul 22 10:45:42 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 22 Jul 2009 10:45:42 +0200
Subject: [csw-maintainers] (dbus fixes)
In-Reply-To: <4A5E40CA.5070305@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>	<4A5DC986.3020606@wbonnet.net>	<4A5E3B57.1030901@wbonnet.net>	<20090715204018.GA94635@bolthole.com>
	<4A5E40CA.5070305@wbonnet.net>
Message-ID: <4A66D1B6.4000302@opencsw.org>

Hi William,

William Bonnet wrote:
>> yes you should be the one to announce it :-)
>>   
> Master Yoda^W^WPhil, the one to announce it, will I be ;)
> 
>> once it passes through testing, thenI would say as soon as I officialy
>> clear it through newpkgs, you should send emails to both the user, and
>> "announce" lists, mentioning what is required.
>>   
> So thanks for more feedbacks then i'll request for a release.

I have just tried to update the dbus package from testing and have seen
two problems.

1) I had to issue "kill -9 `cat /opt/csw/var/run/dbus/pid`" during the
pkgrm of the old dbus package (when it says "Disabling
svc:/system/cswdbus ..."), not before. Doing it before just made SMF
start cswdbus again.

2) After having the new version installed the service doesn't come up
any more. From /var/svc/log/system-cswdbus:default.log

[ Jul 22 10:40:07 Executing start method
("/var/opt/csw/svc/method/svc-cswdbus start") ]
/var/opt/csw/svc/method/svc-cswdbus: smf_is_nonglobalzone: not found
--config-file specified but configuration file
/opt/csw/etc/dbus-1/system.conf already requested
dbus failed to start: error 1
[ Jul 22 10:40:07 Method "start" exited with status 1 ]

# pkginfo -x CSWdbus
CSWdbus  dbus - DBus Message Bus System
         (i386) 1.2.12,REV=2009.07.15

The whole upgrade console log can be found at [1].

Sebastian

[1] http://pastebin.com/m6956c9d3

From Darin.Perusich at cognigencorp.com  Wed Jul 22 13:40:27 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 22 Jul 2009 07:40:27 -0400
Subject: [csw-maintainers] maintainers near toronto?
In-Reply-To: <1248227645-sup-7242@ntdws12.chass.utoronto.ca>
References: <1248227645-sup-7242@ntdws12.chass.utoronto.ca>
Message-ID: <4A66FAAB.60009@cognigencorp.com>

I'm only an hour and in a half away in Buffalo, NY but won't be making
the track north for this. It's probably a little late but there's a
local Perlmonger group I could have forwarded the info on too but they
likely already no.

Ben Walton wrote:
> Hi All,
> 
> Are there any other maintainers in the Toronto area?  I've learned
> that Damian Conway is giving a free talk at UofT on Monday.  If you're
> out there and interested, ping me and I'll forward you the details.
> 
> -Ben

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com

From dam at opencsw.org  Wed Jul 22 14:18:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 22 Jul 2009 14:18:51 +0200
Subject: [csw-maintainers] Updated version of gdb
In-Reply-To: <7B8BE778-3711-4325-A1E5-272F03709513@opencsw.org>
References: <827670CA-A98F-44B4-A09A-DDE7D1E0E038@opencsw.org>
	<c7dddeaa0901241142hf6e3b08m4599b8ce7ed072f1@mail.gmail.com>
	<7B8BE778-3711-4325-A1E5-272F03709513@opencsw.org>
Message-ID: <EF495098-82C9-4418-91F3-10A9FB3BC0D4@opencsw.org>

Hi,

there is now a new version of gdb in testing/:

   gdb-6.8,REV=2009.07.22-SunOS5.8-i386-CSW.pkg.gz
   gdb-6.8,REV=2009.07.22-SunOS5.8-sparc-CSW.pkg.gz

There are some failing tests which may or may not be serious.
There are both 32/64 bit versions for both x86 and sparc.
Please let me know if it works so it can be released soon.


Best regards

   -- Dago

From ellson at opencsw.org  Wed Jul 22 18:05:47 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 12:05:47 -0400
Subject: [csw-maintainers] howto control which files go into which part
 of split package
In-Reply-To: <1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
	<1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>
Message-ID: <4A6738DB.5080105@opencsw.org>

More questions:

1.  How do I get a plugin in one subpackage to recognize that the 
library it depends on comes from the main subpackage?

"make package" fails on "CSWgraphvizgd"  with:

     ERROR: cannot find package for libcdt.so.4
     make: *** [pkgcheck-CSWgraphvizgd] Error 2

even though the main package "CSWgraphviz" was already packaged and 
contains:

     CSWgraphviz/root/opt/csw/lib/libcdt.so.4.0.0

(but no "libcdt.so.4", which would normally be a soflink, is that the 
problem?)





2.  Can I modify the packaging parts of the Makefile, and redo "make 
package"  without rebuilding from
"make clean"  ?



John


From skayser at opencsw.org  Wed Jul 22 20:01:20 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 22 Jul 2009 20:01:20 +0200 (CEST)
Subject: [csw-maintainers] howto control which files go into which part
 of split package
In-Reply-To: <4A6738DB.5080105@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
	<1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>
	<4A6738DB.5080105@opencsw.org>
Message-ID: <53663.194.246.122.22.1248285680.squirrel@ssl.skayser.de>

Hi John,

John Ellson wrote:
> More questions:
>
> 1.  How do I get a plugin in one subpackage to recognize that the
> library it depends on comes from the main subpackage?
>
> "make package" fails on "CSWgraphvizgd"  with:
>
>      ERROR: cannot find package for libcdt.so.4
>      make: *** [pkgcheck-CSWgraphvizgd] Error 2
>
> even though the main package "CSWgraphviz" was already packaged and
> contains:
>
>      CSWgraphviz/root/opt/csw/lib/libcdt.so.4.0.0
>
> (but no "libcdt.so.4", which would normally be a soflink, is that the
> problem?)

that is the checkpkg script complaining and i don't know whether there is
a way to have checkpkg learn about the packages that are part of the
build. I have set ENABLE_CHECK to 0 (disables checkpkg) in such cases and
ran checkpkg on the packages manually (if in doubt).

> 2.  Can I modify the packaging parts of the Makefile, and redo "make
> package"  without rebuilding from
> "make clean"  ?

gmake repackage

Sebastian


From ellson at opencsw.org  Wed Jul 22 23:15:46 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 17:15:46 -0400
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <4A677CAB.8050808@opencsw.org>
References: <4A677CAB.8050808@opencsw.org>
Message-ID: <4A678182.9020005@opencsw.org>

On 07/22/2009 04:55 PM, John Ellson wrote:
> Dago,
>
> Any chance you could rebuild cairo to include SVG support, please?
>
> I can take a shot at it if you like?
>
>
> John


So apparently the reason its not provided is because there is a circular 
depdendency with librsvg ?


Why does a circular dependency matter?    Try to install either and you 
get both - where is the ambiguity?


John

From ellson at opencsw.org  Wed Jul 22 22:55:07 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 16:55:07 -0400
Subject: [csw-maintainers] libcairo svg surface
Message-ID: <4A677CAB.8050808@opencsw.org>

Dago,

Any chance you could rebuild cairo to include SVG support, please?

I can take a shot at it if you like?


John

From ellson at opencsw.org  Thu Jul 23 00:03:39 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 18:03:39 -0400
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <4A678182.9020005@opencsw.org>
References: <4A677CAB.8050808@opencsw.org> <4A678182.9020005@opencsw.org>
Message-ID: <4A678CBB.2030208@opencsw.org>

On 07/22/2009 05:15 PM, John Ellson wrote:
> On 07/22/2009 04:55 PM, John Ellson wrote:
>> Dago,
>>
>> Any chance you could rebuild cairo to include SVG support, please?
>>
>> I can take a shot at it if you like?
>>
>>
>> John
>
>
> So apparently the reason its not provided is because there is a 
> circular depdendency with librsvg ?
>
>
> Why does a circular dependency matter?    Try to install either and 
> you get both - where is the ambiguity?

It builds fine with svg enabled.  There is no library dependency on 
librsvg.  I think it may just be a testing dependency.


John

From ellson at opencsw.org  Thu Jul 23 03:15:18 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 21:15:18 -0400
Subject: [csw-maintainers] New graphviz packages in testing
Message-ID: <4A67B9A6.5030309@opencsw.org>

There are new graphviz packages in testing:

   graphviz        Graph Visualization Tools
   graphvizdevel      headers etc. for development
   graphvizdoc        documentation
   graphvizgraphs     example graphs
   graphvizgd         renderers using gd
   graphvizsharp      language binding for C#
   graphvizguile      language binding for guile
   graphvizperl       language binding for perl
   graphvizpython     language binding for python
   graphviztcl        language binding for tcl

I can't actually install from testing.  I'm not sure why?

     # pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u graphviz
     ...
     Updated description file
     INTERNAL ERROR: cannot get remote version for CSWexpat
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWpango
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWlibcairo
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWgts
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWlibxrender
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWgs
     Perhaps your catalog is out of date
     Error: dependancies for graphviz not up to date.

Can anyone help with this?

Otherwise, perhaps I could ask for just a packaging review?

This graphviz package set is built from the latest stable 
graphviz-2.24.0 release from http://www.graphviz.org/
It uses pango/cairo by default to achieve quality anti-aliased line 
drawings and UTF8 character support,
so the "gd" renderer is mostly obsolete, unless you need GIF support.

"graphviz" obsoletes "graphviz2" and "graphvis"


checkpkg fails on graphvizgd and on the language binding subpackages 
because I don't know how to tell
the subpackages about libraries provided by the main package. Help please?

     $ checkpkg graphvizgd-2.24.0,REV=2009.07.23-SunOS5.8-i386-CSW.pkg.gz
     ...
     Building index from SVR4 installed packages database...
     (May take a while)
     Cross-referencing indexes...
       found SUNWxwplt for libX11.so.4
       found CSWxpm for libXpm.so.4.11
       found SUNWcsl for libc.so.1
     ERROR: cannot find package for libcdt.so.4
     $


Currently there is no support for SVG images in nodes because libcairo 
was built without SVG support.

There are some other language bindings that failed to build:   java, 
ocaml, lua, php
I'll take another look at those after this set is released.


John

From skayser at opencsw.org  Thu Jul 23 13:21:19 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 23 Jul 2009 13:21:19 +0200
Subject: [csw-maintainers] New graphviz packages in testing
In-Reply-To: <4A67B9A6.5030309@opencsw.org>
References: <4A67B9A6.5030309@opencsw.org>
Message-ID: <4A6847AF.6010601@opencsw.org>

John Ellson wrote:
> There are new graphviz packages in testing:
> 
>    graphviz        Graph Visualization Tools
>    graphvizdevel      headers etc. for development
>    graphvizdoc        documentation
>    graphvizgraphs     example graphs
>    graphvizgd         renderers using gd
>    graphvizsharp      language binding for C#
>    graphvizguile      language binding for guile
>    graphvizperl       language binding for perl
>    graphvizpython     language binding for python
>    graphviztcl        language binding for tcl
> 
> I can't actually install from testing.  I'm not sure why?
> 
>      # pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u graphviz
>      ...
>      Updated description file
>      INTERNAL ERROR: cannot get remote version for CSWexpat
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWpango
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWlibcairo
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWgts
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWlibxrender
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWgs
>      Perhaps your catalog is out of date
>      Error: dependancies for graphviz not up to date.
> 
> Can anyone help with this?

Have you tried using pkgutil for the same purpose? pkg-get once couldn't
handle dependencies that spanned different repositories (graphviz from
testing, deps from current in your case). I don't know whether that is
still the case.

If it is indeed the pkg-get issue we should mention this on the testing
page [1] so that people know about it.

Sebastian

[1] http://mirror.opencsw.org/testing.html

From maciej at opencsw.org  Thu Jul 23 17:25:35 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 16:25:35 +0100
Subject: [csw-maintainers] The source code of www.opencsw.org
Message-ID: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>

Hello OpenCSW web site maintainers,

I've been looking at www.opencsw.org website, and was wondering - if I
wanted to suggest a change for it, or work on the website, how would I
do that? I was looking for source code, but I didn't find it. Is the
source code of www.opencsw.org checked into subversion? If not, what
do you think of submitting the code there? It would make it easier to
suggest a change, or submit a patch, and generally make things more
open.

Maciej

From ellson at opencsw.org  Thu Jul 23 17:43:50 2009
From: ellson at opencsw.org (John Ellson)
Date: Thu, 23 Jul 2009 11:43:50 -0400
Subject: [csw-maintainers] New graphviz packages in testing
In-Reply-To: <4A6847AF.6010601@opencsw.org>
References: <4A67B9A6.5030309@opencsw.org> <4A6847AF.6010601@opencsw.org>
Message-ID: <4A688536.7050404@opencsw.org>

On 07/23/2009 07:21 AM, Sebastian Kayser wrote:
> John Ellson wrote:
>    
>> There are new graphviz packages in testing:
>>
>>     graphviz        Graph Visualization Tools
>>     graphvizdevel      headers etc. for development
>>     graphvizdoc        documentation
>>     graphvizgraphs     example graphs
>>     graphvizgd         renderers using gd
>>     graphvizsharp      language binding for C#
>>     graphvizguile      language binding for guile
>>     graphvizperl       language binding for perl
>>     graphvizpython     language binding for python
>>     graphviztcl        language binding for tcl
>>
>> I can't actually install from testing.  I'm not sure why?
>>
>>       # pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u graphviz
>>       ...
>>       Updated description file
>>       INTERNAL ERROR: cannot get remote version for CSWexpat
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWpango
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWlibcairo
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWgts
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWlibxrender
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWgs
>>       Perhaps your catalog is out of date
>>       Error: dependancies for graphviz not up to date.
>>
>> Can anyone help with this?
>>      
> Have you tried using pkgutil for the same purpose? pkg-get once couldn't
> handle dependencies that spanned different repositories (graphviz from
> testing, deps from current in your case). I don't know whether that is
> still the case.
>
> If it is indeed the pkg-get issue we should mention this on the testing
> page [1] so that people know about it.
>
> Sebastian
>
> [1] http://mirror.opencsw.org/testing.html
>    


pkgutil worked for me.

I've fixed some packaging bugs in graphviz this morning, and now the 
installed binaries
seem to be working.

I then removed graphviz and retried the pkg-get method.   I still get 
the INTERNAL ERRORs, but
this time the installation eventually succeeded.

John


From bwalton at opencsw.org  Thu Jul 23 17:45:53 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 11:45:53 -0400
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
Message-ID: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
> source code of www.opencsw.org checked into subversion? If not, what
> do you think of submitting the code there? It would make it easier to
> suggest a change, or submit a patch, and generally make things more
> open.

+1 for making the website code open and public.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/744a10ae/attachment-0001.asc>

From bwalton at opencsw.org  Thu Jul 23 17:46:42 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 11:46:42 -0400
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
Message-ID: <1248363975-sup-2856@ntdws12.chass.utoronto.ca>

Excerpts from Ben Walton's message of Thu Jul 23 11:45:53 -0400 2009:
> +1 for making the website code open and public.

...obviously files containing passwords to databases should be
excepted.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/fe8ca014/attachment.asc>

From trygvis at opencsw.org  Thu Jul 23 17:50:55 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Thu, 23 Jul 2009 17:50:55 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
Message-ID: <4A6886DF.30507@opencsw.org>

Ben Walton wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>> source code of www.opencsw.org checked into subversion? If not, what
>> do you think of submitting the code there? It would make it easier to
>> suggest a change, or submit a patch, and generally make things more
>> open.
> 
> +1 for making the website code open and public.

+1

Would be nice to be able to turn the package database into a separate 
application, I have clients that might see use something like that.

--
Trygve

From phil at bolthole.com  Thu Jul 23 17:53:11 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 08:53:11 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
Message-ID: <20090723155311.GD53578@bolthole.com>

On Thu, Jul 23, 2009 at 11:45:53AM -0400, Ben Walton wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
> > source code of www.opencsw.org checked into subversion? If not, what
> > do you think of submitting the code there? It would make it easier to
> > suggest a change, or submit a patch, and generally make things more
> > open.
> 
> +1 for making the website code open and public.

absolutely not "public".

in contrast, any MAINTAINER who wants to look at it, already can. 
you just need to know where to look :-)

From trygvis at opencsw.org  Thu Jul 23 17:56:49 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 23 Jul 2009 17:56:49 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723155311.GD53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
Message-ID: <4A688841.8020503@opencsw.org>

Philip Brown wrote:
> On Thu, Jul 23, 2009 at 11:45:53AM -0400, Ben Walton wrote:
>> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>>> source code of www.opencsw.org checked into subversion? If not, what
>>> do you think of submitting the code there? It would make it easier to
>>> suggest a change, or submit a patch, and generally make things more
>>> open.
>> +1 for making the website code open and public.
> 
> absolutely not "public".

Why not?

--
Trygve

From bwalton at opencsw.org  Thu Jul 23 17:59:19 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 11:59:19 -0400
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723155311.GD53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
Message-ID: <1248364654-sup-4834@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 23 11:53:11 -0400 2009:

> absolutely not "public".

Aside for the issue of passwords, why not public?

> in contrast, any MAINTAINER who wants to look at it, already can. 
> you just need to know where to look :-)

Care to share this then?  How is it versioned?  Where does it live?
How would a maintainer make edits and commit changes?  I don't
personally care, but it should be knowledge available to all
maintainers.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/2aba8e21/attachment.asc>

From phil at bolthole.com  Thu Jul 23 18:03:07 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:03:07 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A688841.8020503@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
Message-ID: <20090723160307.GF53578@bolthole.com>

On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
>> absolutely not "public".
>
> Why not?

security issues.


From phil at bolthole.com  Thu Jul 23 18:06:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:06:56 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6886DF.30507@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<4A6886DF.30507@opencsw.org>
Message-ID: <20090723160656.GH53578@bolthole.com>

On Thu, Jul 23, 2009 at 05:50:55PM +0200, Trygve Laugst??l wrote:
> Ben Walton wrote:
>> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>>> source code of www.opencsw.org checked into subversion? If not, what
>>> do you think of submitting the code there? It would make it easier to
>>> suggest a change, or submit a patch, and generally make things more
>>> open.
>>
>> +1 for making the website code open and public.
>
> +1
>
> Would be nice to be able to turn the package database into a separate  
> application, I have clients that might see use something like that.

Umm... I doubt your clients would really want to use our code.
Euuu, ick ;-)

besides, there isnt really much "code" for that. What code would they
actually use? the
 "taking apart a package, to analyze and register in a database" bit?

If you're serious about that, I'll consider making my ugly mess something
that is more suitable for publication.

From phil at bolthole.com  Thu Jul 23 18:15:58 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:15:58 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723160656.GH53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<4A6886DF.30507@opencsw.org> <20090723160656.GH53578@bolthole.com>
Message-ID: <20090723161558.GI53578@bolthole.com>

On Thu, Jul 23, 2009 at 09:06:56AM -0700, Philip Brown wrote:
> ....
> besides, there isnt really much "code" for that. What code would they
> actually use? the
>  "taking apart a package, to analyze and register in a database" bit?
> 
> If you're serious about that, I'll consider making my ugly mess something
> that is more suitable for publication.

in mentally reviewing what we do for that.. the code is really, really
specific to opencsw. I doubt it would be useful for anywhere else.
It's very mantis-specific, for example.
But as I said, tell me more about what functionality you think would be
actually useful about your clients, and I'll see if there's any code
sharing that makes sense.


From pfelecan at opencsw.org  Thu Jul 23 18:14:20 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 23 Jul 2009 18:14:20 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248364654-sup-4834@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 23 Jul 2009 11\:59\:19 -0400")
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
Message-ID: <x6ljmfqsv7.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

> Excerpts from Philip Brown's message of Thu Jul 23 11:53:11 -0400 2009:
>
>> absolutely not "public".
>
> Aside for the issue of passwords, why not public?
>
>> in contrast, any MAINTAINER who wants to look at it, already can. 
>> you just need to know where to look :-)
>
> Care to share this then?  How is it versioned?  Where does it live?
> How would a maintainer make edits and commit changes?  I don't
> personally care, but it should be knowledge available to all
> maintainers.

I agree with Ben. And I say that there is too much opacity here en
there... Wasn't communication a theme for a month of this year? Maybe we
should start with internal communication.
-- 
Peter

From phil at bolthole.com  Thu Jul 23 18:24:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:24:45 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <x6ljmfqsv7.fsf@abhavana.free.fr>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
	<x6ljmfqsv7.fsf@abhavana.free.fr>
Message-ID: <20090723162445.GK53578@bolthole.com>

On Thu, Jul 23, 2009 at 06:14:20PM +0200, Peter FELECAN wrote:
> ...
> I agree with Ben. And I say that there is too much opacity here en
> there... Wasn't communication a theme for a month of this year? Maybe we
> should start with internal communication.

ok.. what's lacking in "communication" here? :-)
I have "communicated" that it is already available to look at, by any
maintainer. 

I will even go one step further, by mentioning that for anyone who cares,
it is in a fairly obvious place under /var, on bender.

Beyond that, this isnt a "communication" issue any more.
"internal communication", is not synonymous with "public communication".
In some ways, that is a contradiction: "internal" vs "public".

And I will repeat: anyone who wishes to do actual work on this stuff,
should let me or Ihsan know, and we will be happy to bore you to death with
more details off-list.



From ellson at opencsw.org  Thu Jul 23 18:44:05 2009
From: ellson at opencsw.org (John Ellson)
Date: Thu, 23 Jul 2009 12:44:05 -0400
Subject: [csw-maintainers] howto create catalog files ?
Message-ID: <4A689355.1040405@opencsw.org>

What is the command used to generate a catalog of a set of packages in a 
directory?


Now that I can build graphviz the OpenCSW way,  I'm planning to
use local resources to build nightly development snapshot pkgs
for solaris users, and distribute them like we do with rpms
and debs from our http://www.graphviz.org/ site.

John




From pfelecan at opencsw.org  Thu Jul 23 18:44:27 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 23 Jul 2009 18:44:27 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723162445.GK53578@bolthole.com> (Philip Brown's message of
	"Thu\, 23 Jul 2009 09\:24\:45 -0700")
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
	<x6ljmfqsv7.fsf@abhavana.free.fr> <20090723162445.GK53578@bolthole.com>
Message-ID: <x6hbx3qrh0.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Thu, Jul 23, 2009 at 06:14:20PM +0200, Peter FELECAN wrote:
>> ...
>> I agree with Ben. And I say that there is too much opacity here en
>> there... Wasn't communication a theme for a month of this year? Maybe we
>> should start with internal communication.
>
> ok.. what's lacking in "communication" here? :-)

e.g. speaking, writing, singing, dancing, &c, about methods, procedures
and tools.

> I will even go one step further, by mentioning that for anyone who cares,
> it is in a fairly obvious place under /var, on bender.

sure, the source is the documentation; heard it so often.

> "internal communication", is not synonymous with "public communication".
> In some ways, that is a contradiction: "internal" vs "public".

now that the maintainer's list is "public" we certainly have an issue.
-- 
Peter

From bwalton at opencsw.org  Thu Jul 23 18:48:52 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 12:48:52 -0400
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <4A689355.1040405@opencsw.org>
References: <4A689355.1040405@opencsw.org>
Message-ID: <1248367682-sup-6580@ntdws12.chass.utoronto.ca>

Excerpts from John Ellson's message of Thu Jul 23 12:44:05 -0400 2009:
> What is the command used to generate a catalog of a set of packages in a 
> directory?

You'd want the bldcat tool that Peter B put together.  It's part of
the CSWpkgutil package.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/0f06191d/attachment.asc>

From pfelecan at opencsw.org  Thu Jul 23 18:48:07 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 23 Jul 2009 18:48:07 +0200
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <4A689355.1040405@opencsw.org> (John Ellson's message of "Thu\,
	23 Jul 2009 12\:44\:05 -0400")
References: <4A689355.1040405@opencsw.org>
Message-ID: <x6d47rqraw.fsf@abhavana.free.fr>

John Ellson <ellson at opencsw.org> writes:

> Now that I can build graphviz the OpenCSW way,  I'm planning to
> use local resources to build nightly development snapshot pkgs
> for solaris users, and distribute them like we do with rpms
> and debs from our http://www.graphviz.org/ site.

are you a member of the graphiz project? I read your bio on the
maintainer's page of our site and there wasn't a lot to read...
think: communication.

-- 
Peter

From ellson at opencsw.org  Thu Jul 23 18:52:03 2009
From: ellson at opencsw.org (John Ellson)
Date: Thu, 23 Jul 2009 12:52:03 -0400
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <x6d47rqraw.fsf@abhavana.free.fr>
References: <4A689355.1040405@opencsw.org> <x6d47rqraw.fsf@abhavana.free.fr>
Message-ID: <4A689533.4060400@opencsw.org>

On 07/23/2009 12:48 PM, Peter FELECAN wrote:
> John Ellson<ellson at opencsw.org>  writes:
>
>    
>> Now that I can build graphviz the OpenCSW way,  I'm planning to
>> use local resources to build nightly development snapshot pkgs
>> for solaris users, and distribute them like we do with rpms
>> and debs from our http://www.graphviz.org/ site.
>>      
> are you a member of the graphiz project?

Yes.

> I read your bio on the
> maintainer's page of our site and there wasn't a lot to read...
> think: communication.
>
>    


From phil at bolthole.com  Thu Jul 23 19:01:39 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 10:01:39 -0700
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <4A689533.4060400@opencsw.org>
References: <4A689355.1040405@opencsw.org> <x6d47rqraw.fsf@abhavana.free.fr>
	<4A689533.4060400@opencsw.org>
Message-ID: <20090723170139.GA73684@bolthole.com>

On Thu, Jul 23, 2009 at 12:52:03PM -0400, John Ellson wrote:
> On 07/23/2009 12:48 PM, Peter FELECAN wrote:
>...
>> I read your bio on the
>> maintainer's page of our site and there wasn't a lot to read...
>> think: communication.

A reminder,to John and others:
you can update bender:~/public_html/bio.html (create dir if needed)
to say more about you, and it will then show up in
www.opencsw.org/maintainers/yourname


From maciej at opencsw.org  Thu Jul 23 19:25:25 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 18:25:25 +0100
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723162445.GK53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
	<x6ljmfqsv7.fsf@abhavana.free.fr>
	<20090723162445.GK53578@bolthole.com>
Message-ID: <ce97543b0907231025g508f04eav7c9eb0cb5a69af74@mail.gmail.com>

On Thu, Jul 23, 2009 at 5:24 PM, Philip Brown<phil at bolthole.com> wrote:
> And I will repeat: anyone who wishes to do actual work on this stuff,
> should let me or Ihsan know, and we will be happy to bore you to death with
> more details off-list.

I understand the problem is of a practical nature, not of a principle.
Do you have a rough estimate how much work it would require to get
into a state in which it could live in the repository? Would it be
hard to isolate the dubious bits and source-control the rest?

Maciej

From william at wbonnet.net  Thu Jul 23 22:00:02 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 23 Jul 2009 22:00:02 +0200
Subject: [csw-maintainers] Firefox 3.0.12 is in testing
Message-ID: <4A68C142.6040306@wbonnet.net>

Hi

I have updated the firefox package in testing. It works on Solaris 10, 
and will certainly not work on Solaris 8.

Thanks for the feedbacks

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 maciej at opencsw.org  Thu Jul 23 22:39:33 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 21:39:33 +0100
Subject: [csw-maintainers] Firefox 3.0.12 is in testing
In-Reply-To: <4A68C142.6040306@wbonnet.net>
References: <4A68C142.6040306@wbonnet.net>
Message-ID: <ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>

On Thu, Jul 23, 2009 at 9:00 PM, William Bonnet<william at wbonnet.net> wrote:
> Hi
>
> I have updated the firefox package in testing. It works on Solaris 10, and
> will certainly not work on Solaris 8.
>
> Thanks for the feedbacks

Yey for Firefox 3!

I see a sparc package. Are you going to send out a x86 package as well?

Maciej

From maciej at opencsw.org  Thu Jul 23 22:45:35 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 21:45:35 +0100
Subject: [csw-maintainers] Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
Message-ID: <ce97543b0907231345u2c9962cfj8891f865929bcbaa@mail.gmail.com>

On Thu, Jul 23, 2009 at 9:39 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> I see a sparc package. Are you going to send out a x86 package as well?

Ah, there was an interleave, the x86 version is there as well.
Somehow, pkgutil -t http://mirror.opencsw.org/opencsw/testing -i
firefox doesn't pick it up. Are other people able to install
firefox-3.0.12 in this way?

Maciej

From william at wbonnet.net  Thu Jul 23 22:45:49 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 23 Jul 2009 22:45:49 +0200
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
Message-ID: <4A68CBFD.3000408@wbonnet.net>

Hi Maciej
> Yey for Firefox 3!
>
> I see a sparc package. Are you going to send out a x86 package as well?
>   
There is also the i386 package :)

http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-sparc.pkg.gz
http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-i386.pkg.gz

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 Jul 23 23:23:18 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:23:18 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
Message-ID: <4A68D4C6.3040200@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Hacienda
Hacienda (Matches) Blizinski wrote:
> Hello OpenCSW web site maintainers,
> 
> I've been looking at www.opencsw.org website, and was wondering - if I
> wanted to suggest a change for it, or work on the website, how would I
> do that? I was looking for source code, but I didn't find it. Is the
> source code of www.opencsw.org checked into subversion? If not, what
> do you think of submitting the code there? It would make it easier to
> suggest a change, or submit a patch, and generally make things more
> open.
> 
> Maciej
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

What I did when I had a change was to send an email to Phil with what I wanted
to change. along with a link to the new page from my homedir on www.opencsw.org
http://www.opencsw.org/~mwatters/need_maintainer.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

iEYEARECAAYFAkpo1MYACgkQLrhmsXMSLxfNJgCgqQeJqMlMW6t58e/Q435UieEU
8tQAoKG7gDgDTBntZ2eusA8eLEFjtvCn
=k4sA
-----END PGP SIGNATURE-----

From phil at bolthole.com  Thu Jul 23 23:28:11 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 14:28:11 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68D4C6.3040200@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<4A68D4C6.3040200@opencsw.org>
Message-ID: <20090723212810.GA8541@bolthole.com>

On Thu, Jul 23, 2009 at 04:23:18PM -0500, Mike Watters wrote:
> ...
> What I did when I had a change was to send an email to Phil with what I wanted
> to change. along with a link to the new page from my homedir on www.opencsw.org
> http://www.opencsw.org/~mwatters/need_maintainer.php

and fyi, his script was then moved to the top-level web area, as
http://www.opencsw.org/need_maintainer.php

and he "owns" the script.


From mwatters at opencsw.org  Thu Jul 23 23:29:13 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:29:13 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723160307.GF53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
Message-ID: <4A68D629.9040802@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Phil
Philip Brown wrote:
> On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
>>> absolutely not "public".
>> Why not?
> 
> security issues.
> 

You are correct, as the code lies currently, there would be severe security
issues in publishing the code.

however, with a few minor modifications, it would be quite possible.

What would need to happen is to convert the php from hard coding all values,
php templates.  you can publish the code, but not the back end template (file
containing all values)



- --

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

iEYEARECAAYFAkpo1igACgkQLrhmsXMSLxcKPgCg0Poq0LMzufZXxSI2FhsCHY58
mp4AniwzTfyYjrcwaON44RvGsYOk82up
=f18E
-----END PGP SIGNATURE-----

From mwatters at opencsw.org  Thu Jul 23 23:48:14 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:48:14 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68D629.9040802@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org>
Message-ID: <4A68DA9E.3030405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Ben,

Before anyone starts going down the road on the
existing http://www.opencsw.org/ code, have we had any
progress on http://www-mockup.opencsw.org/

- --

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

iEYEARECAAYFAkpo2p4ACgkQLrhmsXMSLxfvRgCfc96Qr5tMVFX/eq+ntN/dcZQO
DRwAoKlvbBSNRC2XmiRYU+49rQ++QWus
=7X9J
-----END PGP SIGNATURE-----

From mwatters at opencsw.org  Thu Jul 23 23:49:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:49:52 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68DA9E.3030405@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<4A68DA9E.3030405@opencsw.org>
Message-ID: <4A68DB00.7040409@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


> Hi Ben,
> 
> Before anyone starts going down the road on the
> existing http://www.opencsw.org/ code, have we had any
> progress on http://www-mockup.opencsw.org/
> 
sorry Ben, should be directed William I believe.

- --

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

iEYEARECAAYFAkpo2wAACgkQLrhmsXMSLxckhwCgteitjK3l/mjXqDrE8f5y2Dq0
tLIAoK9o/ILNThsKRRlvPe6qe22NATWw
=Ofp5
-----END PGP SIGNATURE-----

From phil at bolthole.com  Fri Jul 24 00:01:49 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 15:01:49 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68D629.9040802@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org>
Message-ID: <20090723220149.GB8541@bolthole.com>

On Thu, Jul 23, 2009 at 04:29:13PM -0500, Mike Watters wrote:
> 
> Hi Phil
> Philip Brown wrote:
> > On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
> >>> absolutely not "public".
> >> Why not?
> > 
> > security issues.
> > 
> 
> You are correct, as the code lies currently, there would be severe security
> issues in publishing the code.
> 
> however, with a few minor modifications, it would be quite possible.

And when changes are made in the future, more holes likely open up.
That is the way of source code.

Better for security to keep it local-visibility-only, if it doesnt need to
be fully public.

It doesnt need to be fully public.


From phil at bolthole.com  Fri Jul 24 01:54:02 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 16:54:02 -0700
Subject: [csw-maintainers] Please resubmit recent packages....
Message-ID: <20090723235402.GE8541@bolthole.com>

Arg....
It would seem there was a bug in the new stuff James and I worked out, to
"help coordination"

If you have recently had your packages released, please put them in newpkgs
again and let me know :-(

sorry for the inconvenience.

From phil at bolthole.com  Fri Jul 24 02:04:07 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 17:04:07 -0700
Subject: [csw-maintainers] Please resubmit recent packages....
In-Reply-To: <20090723235402.GE8541@bolthole.com>
References: <20090723235402.GE8541@bolthole.com>
Message-ID: <20090724000407.GF8541@bolthole.com>

On Thu, Jul 23, 2009 at 04:54:02PM -0700, Philip Brown wrote:
> Arg....
> It would seem there was a bug in the new stuff James and I worked out, to
> "help coordination"
> 
> If you have recently had your packages released, please put them in newpkgs
> again and let me know :-(
> 
> sorry for the inconvenience.

to be clear what happened:  they got "registered" in our database, but did
not make it to our actual mirror tree filesystem. and got deleted :-(
Older versions still remain in the tree.

From mwatters at opencsw.org  Fri Jul 24 04:26:17 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 21:26:17 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723220149.GB8541@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com>
Message-ID: <4A691BC9.2050209@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Jul 23, 2009 at 04:29:13PM -0500, Mike Watters wrote:
>> Hi Phil
>> Philip Brown wrote:
>>> On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
>>>>> absolutely not "public".
>>>> Why not?
>>> security issues.
>>>
>> You are correct, as the code lies currently, there would be severe security
>> issues in publishing the code.
>>
>> however, with a few minor modifications, it would be quite possible.
> 
> And when changes are made in the future, more holes likely open up.
> That is the way of source code.
> 
> Better for security to keep it local-visibility-only, if it doesnt need to
> be fully public.
> 
> It doesnt need to be fully public.

what about a compromise and have either a sourceforge or self-hosted "members
only" subversion,git,cvs,bzr,etc... (pick your favorite version control
software) for the source code.


- --
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

iEYEARECAAYFAkppG8kACgkQLrhmsXMSLxe1OQCfa9vqFwXptAzOb+q9yDqXgLc3
TKwAoMcUsbEVT9ItjdGUbqt2ItVbon0y
=CDdd
-----END PGP SIGNATURE-----

From phil at bolthole.com  Fri Jul 24 06:17:22 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 21:17:22 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A691BC9.2050209@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
Message-ID: <20090724041722.GA50043@bolthole.com>

On Thu, Jul 23, 2009 at 09:26:17PM -0500, Mike Watters wrote:
> 
> what about a compromise and have either a sourceforge or self-hosted "members
> only" subversion,git,cvs,bzr,etc... (pick your favorite version control
> software) for the source code.

that is *exactly* how we do it. :-)

People who write code for the website.. get to "pick their favourite
version control software" for their source code :-)

and sometimes that is SCCS. and sometimes that is cvs. and sometimes that
is "nothing".

Because sometimes, you really  JUST DONT NEED version control.
for a 10 line script, for example.
you just need backups.
which we have.


From phil at bolthole.com  Fri Jul 24 06:27:20 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 21:27:20 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090724041722.GA50043@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
Message-ID: <20090724042720.GB50043@bolthole.com>

On Thu, Jul 23, 2009 at 09:17:22PM -0700, Philip Brown wrote:
> People who write code for the website.. get to "pick their favourite
> version control software" for their source code :-)
> 
> and sometimes that is SCCS. and sometimes that is cvs. and sometimes that
> is "nothing".
> 
> Because sometimes, you really  JUST DONT NEED version control.
> for a 10 line script, for example.
> you just need backups.
> which we have.

To fully connect the dots....

having a big source code repository for packaging type stuff is useful,
becuase it puts things "all in one place", that otherwise, would not have a
home. Otherwise, you'd have some in one persons' home directory, and some
in another person's home directory.

In contrast, for  web work on the www.opencsw.org main webserver,
All the "code" is already "all in one place": in the web server filesystem.

If you are one of the rare people who has chosen to actually do work on the
web coding, then you already know exactly where to look. and you can
already see everyone else's webcode too. its all quite "open" and organized
already.

From maciej at opencsw.org  Fri Jul 24 10:23:48 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 09:23:48 +0100
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090724042720.GB50043@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com> <4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com> <4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
Message-ID: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>

On Fri, Jul 24, 2009 at 5:27 AM, Philip Brown<phil at bolthole.com> wrote:
> having a big source code repository for packaging type stuff is useful,
> becuase it puts things "all in one place", that otherwise, would not have a
> home. Otherwise, you'd have some in one persons' home directory, and some
> in another person's home directory.
>
> In contrast, for ?web work on the www.opencsw.org main webserver,
> All the "code" is already "all in one place": in the web server filesystem.
>
> If you are one of the rare people who has chosen to actually do work on the
> web coding, then you already know exactly where to look. and you can
> already see everyone else's webcode too. its all quite "open" and organized
> already.

Do other people agree with the above? (That it's the web site source
code is open and organized already.)

Maciej

From william at wbonnet.net  Fri Jul 24 10:36:15 2009
From: william at wbonnet.net (William Bonnet)
Date: Fri, 24 Jul 2009 10:36:15 +0200
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907240128ic5b1c73q230ecff15d80d8b9@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>	<4A68CBFD.3000408@wbonnet.net>	<ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>
	<ce97543b0907240128ic5b1c73q230ecff15d80d8b9@mail.gmail.com>
Message-ID: <4A69727F.4080003@wbonnet.net>

Hi

> After I got it installed - it works! I'll keep using it and will come
> back with any comments. Thanks, William!
>   
You're welcome :)

Is someone using it under Solaris 9 ? Can i have some feedbacks please ?

cheers
W.

From pfelecan at opencsw.org  Fri Jul 24 10:47:57 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 24 Jul 2009 10:47:57 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	24 Jul 2009 09\:23\:48 +0100")
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com> <4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com> <4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
Message-ID: <x6ws5ypiv6.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 24, 2009 at 5:27 AM, Philip Brown<phil at bolthole.com> wrote:
>> having a big source code repository for packaging type stuff is useful,
>> becuase it puts things "all in one place", that otherwise, would not have a
>> home. Otherwise, you'd have some in one persons' home directory, and some
>> in another person's home directory.
>>
>> In contrast, for ?web work on the www.opencsw.org main webserver,
>> All the "code" is already "all in one place": in the web server filesystem.
>>
>> If you are one of the rare people who has chosen to actually do work on the
>> web coding, then you already know exactly where to look. and you can
>> already see everyone else's webcode too. its all quite "open" and organized
>> already.
>
> Do other people agree with the above? (That it's the web site source
> code is open and organized already.)

- it's open in the sense that if you have access to the hosting system
  you can read the source

- its organization is intrinsic

- the code is not part of a project manageable by a team, the way that
  the projects that we package are

- there is no documentation for this or other infrastructure of the
  OpenCSW project

Is there opportunity to change this situation? I think that an answer
would give also a measure of the OpenCSW politics.
-- 
Peter

From trygvis at opencsw.org  Fri Jul 24 11:08:04 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Fri, 24 Jul 2009 11:08:04 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org>	<20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
Message-ID: <4A6979F4.5050904@opencsw.org>

Maciej (Matchek) Blizinski wrote:
> On Fri, Jul 24, 2009 at 5:27 AM, Philip Brown<phil at bolthole.com> wrote:
>> having a big source code repository for packaging type stuff is useful,
>> becuase it puts things "all in one place", that otherwise, would not have a
>> home. Otherwise, you'd have some in one persons' home directory, and some
>> in another person's home directory.
>>
>> In contrast, for  web work on the www.opencsw.org main webserver,
>> All the "code" is already "all in one place": in the web server filesystem.
>>
>> If you are one of the rare people who has chosen to actually do work on the
>> web coding, then you already know exactly where to look. and you can
>> already see everyone else's webcode too. its all quite "open" and organized
>> already.
> 
> Do other people agree with the above? (That it's the web site source
> code is open and organized already.)

No, I don't.

Calling the code open and available when it is on a remote file system 
on a *live* web server is not right. If I want to work on source code, I 
need to copy the code locally, test it and probably show it to the rest 
of the group if it contain significant changes. That is not possible as 
it is.

If the code can't be shown to the world without being a security risk, 
and Phil doesn't believe that we can write secure code, I'd like to 
replace the application with something that we all can work on and extend.

--
Trygve

From maciej at opencsw.org  Fri Jul 24 11:45:34 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 10:45:34 +0100
Subject: [csw-maintainers] Subversion on bender
Message-ID: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>

Hello bender admins,

Can you install Subversion client on bender?

Thanks,
Maciej

From maciej at opencsw.org  Fri Jul 24 11:52:52 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 10:52:52 +0100
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
Message-ID: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>

On Fri, Jul 24, 2009 at 10:45 AM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> Can you install Subversion client on bender?

....and pysqlite2. Please. :-)

Maciej

From ihsan at dogan.ch  Fri Jul 24 16:36:34 2009
From: ihsan at dogan.ch (Ihsan Dogan)
Date: Fri, 24 Jul 2009 16:36:34 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
Message-ID: <4A69C6F2.2070003@dogan.ch>

Maciej (Matchek) Blizinski wrote:

>> Can you install Subversion client on bender?
> 
> ....and pysqlite2. Please. :-)

Why do you need Subversion and pysqlite2 on bender?




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From bwalton at opencsw.org  Fri Jul 24 17:07:12 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 11:07:12 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <4A69C6F2.2070003@dogan.ch>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
Message-ID: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>

Excerpts from Ihsan Dogan's message of Fri Jul 24 10:36:34 -0400 2009:

> Why do you need Subversion and pysqlite2 on bender?

<quoting Phil>
People who write code for the website.. get to "pick their favourite
version control software" for their source code :-)
</quoting Phil>

Maciej has written some nice little utilities for displaying a matrix
of installed packages on each machine.  The utilities rely on sqlite
presently.  I guess he prefers svn for his code.

[Every person for themselves version control will eventually lead to
madness...]

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/dfe6aa82/attachment.asc>

From mwatters at opencsw.org  Fri Jul 24 17:45:29 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 10:45:29 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
Message-ID: <4A69D719.6080409@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben Walton wrote:

> [Every person for themselves version control will eventually lead to
> madness...]

I couldn't agree more with that statement.
I propose to standardize on one VCS.

I am partial to subversion (with our without a front end.  eg. trac)
but am open to any of them.

account setup can be setup with our current authentication method to bender.

- --

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

iEYEARECAAYFAkpp1xkACgkQLrhmsXMSLxcn5wCgiRkBg18hVFmGvHQIdM9h/Hy/
x9cAn3I7vSKzkaLszaNu2Kb/8/ZgiwS+
=nBFj
-----END PGP SIGNATURE-----

From phil at bolthole.com  Fri Jul 24 17:57:01 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 08:57:01 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <4A69D719.6080409@opencsw.org>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
Message-ID: <20090724155701.GB91227@bolthole.com>

On Fri, Jul 24, 2009 at 10:45:29AM -0500, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Ben Walton wrote:
> 
> > [Every person for themselves version control will eventually lead to
> > madness...]
> 
> I couldn't agree more with that statement.
> I propose to standardize on one VCS.

"i propose" that only people who are actively doing web source code work,
get to vote on what the VCS for  web server work is.

and for the record, my "vote" is SCCS, with a secondary preference for rcs

From phil at bolthole.com  Fri Jul 24 18:03:43 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 09:03:43 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6979F4.5050904@opencsw.org>
References: <20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
Message-ID: <20090724160343.GC91227@bolthole.com>

On Fri, Jul 24, 2009 at 11:08:04AM +0200, Trygve Laugst??l wrote:
> If the code can't be shown to the world without being a security risk,  
> and Phil doesn't believe that we can write secure code, I'd like to  
> replace the application with something that we all can work on and 
> extend.

There is no "*THE* application".
Additionally pretty much any web application, reguardless of language, has
the potential for security holes. So changing the framework, does not
definitively eliminate the underlying problem here.

This shows a fundamental non-understanding of what is going on in our web
space (and security, for that matter).

Just as people who arent actually maintaining packages, should not get to
tell active maintainers how to package... I think that people who arent
actually DOING WEB WORK, should not be telling those who are, how to do it.

If people want to actually do work, then great! please speak up with what
area you want to work on. Then after that, talk about how you think the 
working environment could be improved.

From maciej at opencsw.org  Fri Jul 24 18:55:29 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 17:55:29 +0100
Subject: [csw-maintainers] Including directory paths in prototypes using
	PKGFILES_CSWfoo
Message-ID: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>

On Tue, Jul 21, 2009 at 3:28 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> You usually specify PKGFILES for all but one packages which gets all
> the rest not included in any package. And you can say
> ?PKGFILES_CSWcupsd = $(sbindir)/(cupsd|cupsfilter)
> and leave out PKGFILES for CSWcupsclient.

More on that matter... There seems to be a problem with including
directories. I'm getting warnings like the following:

## Building pkgmap from package prototype file.
## Processing pkginfo file.
WARNING: missing directory entry for </opt>
WARNING: missing directory entry for </opt/csw>
WARNING: missing directory entry for </opt/csw/lib>

My package's PKGFILES:

PKGFILES_CSWlibcups  = $(prefix)
PKGFILES_CSWlibcups += $(libdir)
PKGFILES_CSWlibcups += $(libdir)/libcups.*

The $(libdir) variable is /opt/csw/lib, after implicitly adding ^ and
$ it becomes ^/opt/csw/lib$, which should match the directory:

$ [[ -d work/install-isa-i386/opt/csw/lib ]] && echo yes
yes

How to convince GAR to include the directories in the prototype?

Maciej

From bwalton at opencsw.org  Fri Jul 24 19:07:34 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 13:07:34 -0400
Subject: [csw-maintainers] Including directory paths in prototypes using
	PKGFILES_CSWfoo
In-Reply-To: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>
References: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>
Message-ID: <1248455214-sup-2440@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 24 12:55:29 -0400 2009:
> How to convince GAR to include the directories in the prototype?

You don't want or need those entries.  They're stripped by GAR on
purpose.  The CSWcommon package includes them.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/e7d8538f/attachment-0001.asc>

From maciej at opencsw.org  Fri Jul 24 19:28:47 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 18:28:47 +0100
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
Message-ID: <ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>

Hi maintainers,

I've commited the change for cups. It migrates the package to mGAR v2:

- dynamic prototypes
- SMF support via cswclassutils
- Zones support
    * /etc/opt/csw instead of /opt/csw/etc
    * /var/opt/csw instead of /opt/csw/var

I'd like to ask the senior maintainers to take a look at the change.
It's relatively large. There are bits I'm unsure about, such as
"ENABLE_CHECK = 0".

http://sourceforge.net/apps/trac/gar/changeset/5642

Maciej

From bwalton at opencsw.org  Fri Jul 24 19:38:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 13:38:36 -0400
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
	<ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
Message-ID: <1248456800-sup-5884@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 24 13:28:47 -0400 2009:

> I'd like to ask the senior maintainers to take a look at the change.
> It's relatively large. There are bits I'm unsure about, such as
> "ENABLE_CHECK = 0".

A while back I made some updates to checkpkg such that it worked on a
set of packages and honoured lib splitting across them such as you're
doing here.  This allowed one to leave ENABLE_CHECK = 1 and still have
checkpkg pass.

After a few revisions of the patch, it was seemingly dropped.  I still
have my local copy of the script I believe if you want to use it to
validate your set of packages _after_ GAR has built them.  I'm not
sure if there are more checks in the current checkpkg though.

Having to disable the checks stinks since they do catch other valuable
things.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/3e8aadc6/attachment.asc>

From bwalton at opencsw.org  Fri Jul 24 20:00:26 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 14:00:26 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724155701.GB91227@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
Message-ID: <1248458300-sup-2415@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 11:57:01 -0400 2009:
> "i propose" that only people who are actively doing web source code work,
> get to vote on what the VCS for  web server work is.

How are you defining 'actively'?  Mike has (as you mentioned) made
previous contributions.  He seems like a good candidate for a 'vote'
to me.  Maciej should also get a vote since he now has code in
'testing' ready for the web.  William as well.

I disagree with this premise for inclusion/exclusion though.  The web
site is a project resource (as opposed to a personal resource like
pkg-get).  It should be a project decision.  Just because someone
isn't personally contributing right this minute doesn't mean they
won't in the future.

Having the consensus of current project members now makes it easier
for more people to feel comfortable contributing at such a time as
they have something valuable to contribute.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/ecad414b/attachment.asc>

From phil at bolthole.com  Fri Jul 24 20:24:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 11:24:21 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248458300-sup-2415@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
Message-ID: <20090724182421.GB16963@bolthole.com>

On Fri, Jul 24, 2009 at 02:00:26PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 11:57:01 -0400 2009:
> > "i propose" that only people who are actively doing web source code work,
> > get to vote on what the VCS for  web server work is.
> 
> How are you defining 'actively'?  Mike has (as you mentioned) made
> previous contributions.  He seems like a good candidate for a 'vote'
> to me.

i agree.  I was just proactively trying to avoid the situation where
everyone decides to put in their oar just because they see the word "vote"
from Mike.


>  Just because someone
> isn't personally contributing right this minute doesn't mean they
> won't in the future.

experience has shown the opposite.

Or do you go by the ideal that EVERYONE in the public at large, should get
"a vote" on ALL opencsw issues?
Because after all, "just because they arent personally contributing right
this minute, doesnt mean they won't in the future".

 To put it another way... it seems a common piece of human nature, that
most people would rather tell other people how to work, than do work
themselves. Those "most people" arent entitled to a vote in how the workers
actually gets things done.

everyone is entitled to an opinion, and to even express that opinion. but
not everyone is entitled to a *vote* on such matters. non-workers have no
right _telling_ workers (especialy volunteer workers) how they are going to 
work.


From phil at bolthole.com  Fri Jul 24 20:53:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 11:53:17 -0700
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <1248456800-sup-5884@ntdws12.chass.utoronto.ca>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
	<ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
	<1248456800-sup-5884@ntdws12.chass.utoronto.ca>
Message-ID: <20090724185317.GA90084@bolthole.com>

On Fri, Jul 24, 2009 at 01:38:36PM -0400, Ben Walton wrote:
> ....
> After a few revisions of the patch, it was seemingly dropped.

My recollection was that you, Dago, and I, discussed a different way of
doing things; you agreed to do it... but then did not finish the method we
discussed.


From bwalton at opencsw.org  Fri Jul 24 21:10:46 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 15:10:46 -0400
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <20090724185317.GA90084@bolthole.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
	<ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
	<1248456800-sup-5884@ntdws12.chass.utoronto.ca>
	<20090724185317.GA90084@bolthole.com>
Message-ID: <1248462369-sup-5261@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 14:53:17 -0400 2009:
> > After a few revisions of the patch, it was seemingly dropped.
> 
> My recollection was that you, Dago, and I, discussed a different way of
> doing things; you agreed to do it... but then did not finish the method we
> discussed.

I knocked out a patch to checkpkg that allowed the calls from GAR to
say -i $pkg,$pkg2, or some such.  It then skipped validating those
dependencies and GAR presumably passed in the set of packages being
built currently.  That was rejected.

I submitted a second patch that let it evaluate multiple packages at
the same time, remembering missing libraries/dependencies and checking
again at the end after checking the rest of the packages.  I'd have to
dig it out, but afaik, there wasn't anything wrong with this patch, it
just wasn't applied.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/fb7b7c74/attachment.asc>

From bwalton at opencsw.org  Fri Jul 24 21:18:40 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 15:18:40 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724182421.GB16963@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
Message-ID: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 14:24:21 -0400 2009:

> everyone is entitled to an opinion, and to even express that opinion. but
> not everyone is entitled to a *vote* on such matters. non-workers have no
> right _telling_ workers (especialy volunteer workers) how they are going to 
> work.

I certainly agree with these points.  I just figured that since effort
was made to form an association with bylaws, voting/non-voting
members[1], etc that this might be something that actually could
warrant a vote from voting members.  The website and how its handled
does have a large impact on the project as a whole, after all.
Maintenace of said domain (which presumably includes the website) is
one of the stated goals of the association.

I'm now requesting access to the web content also.  I'm going to
re-factor all php files such that database connection info is stored
only in a single place, thus facilitating the possibility of storing
this code in a non-local (read: not live) repository in the future[2]
with the db details excluded for security purposes.

William: What is the status of the mock-up?  Are there stumbling
blocks you're facing other than time constraints?

Thanks
-Ben

[1] I couldn't find a list of who gets a vote and who doesn't (or by
    what criteria the board was able to grant or deny this ability).
[2] I'd be happy to check into SCCS for the time being as long as some
    instructions are provided.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/b5f7f1d4/attachment.asc>

From mwatters at opencsw.org  Fri Jul 24 21:32:56 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 14:32:56 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
Message-ID: <4A6A0C68.5040807@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Ben,
Ben Walton wrote:

> I'm now requesting access to the web content also.  I'm going to
> re-factor all php files such that database connection info is stored
> only in a single place, thus facilitating the possibility of storing
> this code in a non-local (read: not live) repository in the future[2]
> with the db details excluded for security purposes.
> 
> William: What is the status of the mock-up?  Are there stumbling
> blocks you're facing other than time constraints?

I am more then happy to help with the re factor or the mock-up if you need an
extra pair of hands.

- --

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

iEYEARECAAYFAkpqDGgACgkQLrhmsXMSLxfBLgCgy+kKGcARLnXXW4oDll/+cMZz
9WYAoMgcuxuS4x7zCCRrSaTYaxyqfLf1
=spuy
-----END PGP SIGNATURE-----

From phil at bolthole.com  Fri Jul 24 21:46:30 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 12:46:30 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
Message-ID: <20090724194630.GA50326@bolthole.com>

On Fri, Jul 24, 2009 at 03:18:40PM -0400, Ben Walton wrote:
> ....
> I'm now requesting access to the web content also.  I'm going to
> re-factor all php files [...]

Wonderful!  welcome to the web team :-)


>  such that database connection info is stored
> only in a single place, thus facilitating the possibility of storing
> this code in a non-local (read: not live) repository in the future[2]
> with the db details excluded for security purposes.

the trouble with that, is that it affects performance.

> [2] I'd be happy to check into SCCS for the time being as long as some
>     instructions are provided.

sccs is about the same as any other local-store version control system.
(or others, even)
usual semmantics of that sort of thing are:

0. initialize a file
1. check out a file
2. check in a file
3. compare a file.


So, 


0. sccs create XYZ
1. sccs edit XYZ
2. sccs delget XYZ  (sends for "delta, then get", if you care ;-)
3. sccs diffs XYZ

there's a few other interesting features of it, but them's the basics.


From bwalton at opencsw.org  Fri Jul 24 21:59:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 15:59:15 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724194630.GA50326@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
Message-ID: <1248465309-sup-92@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:

> > such that database connection info is stored only in a single
> > place, thus facilitating the possibility of storing this code in a
> > non-local (read: not live) repository in the future[2] with the db
> > details excluded for security purposes.
> 
> the trouble with that, is that it affects performance.

No it doesn't.  The live copy on the web would simply be a current
checkout.

The pattern would become:

1. Develop, develop, develop.
2. Commit (which if svn[1] were used, is also a push to central).
3. <optional>Review of changes</optional>
4. Login to bender, proceed to web directory, update local copy.

This could all be done on the live copy to, but with a repo living
outside the code tree, easier testing is facilitated without affecting
the live stuff.

-Ben

[1] I have a personal distaste for svn but it's likely the logical
    choice given that all other project-wide type stuff lives in svn
    already.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/d17a1d03/attachment.asc>

From mwatters at opencsw.org  Fri Jul 24 22:48:48 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 15:48:48 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248465309-sup-92@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
Message-ID: <4A6A1E30.6010405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:
> 
>>> such that database connection info is stored only in a single
>>> place, thus facilitating the possibility of storing this code in a
>>> non-local (read: not live) repository in the future[2] with the db
>>> details excluded for security purposes.
>> the trouble with that, is that it affects performance.
> 
> No it doesn't.  The live copy on the web would simply be a current
> checkout.
> 
> The pattern would become:
> 
> 1. Develop, develop, develop.
> 2. Commit (which if svn[1] were used, is also a push to central).
> 3. <optional>Review of changes</optional>
> 4. Login to bender, proceed to web directory, update local copy.
> 
> This could all be done on the live copy to, but with a repo living
> outside the code tree, easier testing is facilitated without affecting
> the live stuff.
> 
> -Ben
> 
> [1] I have a personal distaste for svn but it's likely the logical
>     choice given that all other project-wide type stuff lives in svn
>     already.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I use Bazzar [1] when checking in/out of Fedora's PackageDB code repo.
not that I do it all that often anymore... but IMO better then svn ( don't take
much ) for those that are stuck in windows it runs nicely on that as well as
linux and solaris...  I can't vouch for mac though... never tried it on my
kid's mac book or the wife's mac book pro.

- --

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

iEYEARECAAYFAkpqHjAACgkQLrhmsXMSLxfbHwCgoKCOt22+MZN/b5F1E5P8May0
mIgAoLuNP+q46x2gs1wDpm/buXSPhHGg
=sZbu
-----END PGP SIGNATURE-----

From skayser at opencsw.org  Fri Jul 24 23:03:28 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 24 Jul 2009 23:03:28 +0200
Subject: [csw-maintainers] Updated libiconv in testing, testing appreciated
Message-ID: <4A6A21A0.3060006@opencsw.org>

Hi,

i have put an updated libiconv 1.13.1 in testing. The most notable
change since version 1.12 is native support for platform-dependent
encoding aliases like the Solaris C locale default of 646 ("for better
compatibility with the platform's own iconv_open function").

This should obsolete related hacks required in other packages. I have it
running on a couple of non-GUI test boxes without hitches, but as it is
a to-be-adopted and fundamental package, I would appreciate a bit more
testing/feedback/review by some of you guys.

I know that the package and software name differ (iconv vs. libiconv),
but that is not something i would want to change now (given that many
dependent packages).

Sebastian

From mwatters at opencsw.org  Fri Jul 24 23:32:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 16:32:50 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <4A6A1E30.6010405@opencsw.org>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>	<20090724194630.GA50326@bolthole.com>	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<4A6A1E30.6010405@opencsw.org>
Message-ID: <4A6A2882.9030300@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Watters wrote:
> Ben Walton wrote:
>> Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:
> 
>>>> such that database connection info is stored only in a single
>>>> place, thus facilitating the possibility of storing this code in a
>>>> non-local (read: not live) repository in the future[2] with the db
>>>> details excluded for security purposes.
>>> the trouble with that, is that it affects performance.
>> No it doesn't.  The live copy on the web would simply be a current
>> checkout.
> 
>> The pattern would become:
> 
>> 1. Develop, develop, develop.
>> 2. Commit (which if svn[1] were used, is also a push to central).
>> 3. <optional>Review of changes</optional>
>> 4. Login to bender, proceed to web directory, update local copy.
> 
>> This could all be done on the live copy to, but with a repo living
>> outside the code tree, easier testing is facilitated without affecting
>> the live stuff.
> 
>> -Ben
> 
>> [1] I have a personal distaste for svn but it's likely the logical
>>     choice given that all other project-wide type stuff lives in svn
>>     already.
> 
> 
>> ------------------------------------------------------------------------
> 
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> I use Bazzar [1] when checking in/out of Fedora's PackageDB code repo.

forgot the link  ;-)

[1] http://bazaar-vcs.org/

- --

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

iEYEARECAAYFAkpqKIIACgkQLrhmsXMSLxdLZwCfbYI1KPAp2X+DNddtqIdOzxOu
RcMAn0kg68ulTQ8j7XjqnbGIVcFSugtD
=UM1m
-----END PGP SIGNATURE-----

From phil at bolthole.com  Fri Jul 24 23:54:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 14:54:52 -0700
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <4A6A21A0.3060006@opencsw.org>
References: <4A6A21A0.3060006@opencsw.org>
Message-ID: <20090724215452.GA73903@bolthole.com>

On Fri, Jul 24, 2009 at 11:03:28PM +0200, Sebastian Kayser wrote:
> 
> I know that the package and software name differ (iconv vs. libiconv),
> but that is not something i would want to change now (given that many
> dependent packages).

Umm... if you're going to be repackaging anyway... then why not repackage
now, and have it tested as such?


From phil at bolthole.com  Fri Jul 24 23:55:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 14:55:50 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248465309-sup-92@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
Message-ID: <20090724215550.GB73903@bolthole.com>

On Fri, Jul 24, 2009 at 03:59:15PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:
> > the trouble with that, is that it affects performance.
> 
> No it doesn't.  The live copy on the web would simply be a current
> checkout.

sorry, misunderstanding. I thougth you meant "the PASSWORDS for database
access, would be in a separate, non-checked-in file".


From Joerg.Schilling at fokus.fraunhofer.de  Sat Jul 25 00:18:05 2009
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sat, 25 Jul 2009 00:18:05 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724194630.GA50326@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org> <20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
Message-ID: <4a6a331d.Q3CFrX1KqEHdCOjZ%Joerg.Schilling@fokus.fraunhofer.de>

Philip Brown <phil at bolthole.com> wrote:

> On Fri, Jul 24, 2009 at 03:18:40PM -0400, Ben Walton wrote:

> > [2] I'd be happy to check into SCCS for the time being as long as some
> >     instructions are provided.

SCCS is OpenSource and I ported to all recent interesting platforms.
Given the fact that my version is 64 bit clean and supports 4 digit year 
numbers, it will even survive Y 2068.

> sccs is about the same as any other local-store version control system.
> (or others, even)

In future, there will be a network aware version ;-)


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 skayser at opencsw.org  Sat Jul 25 01:40:39 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 25 Jul 2009 01:40:39 +0200
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <20090724215452.GA73903@bolthole.com>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com>
Message-ID: <4A6A4677.7000903@opencsw.org>

Philip Brown wrote:
> On Fri, Jul 24, 2009 at 11:03:28PM +0200, Sebastian Kayser wrote:
>> I know that the package and software name differ (iconv vs. libiconv),
>> but that is not something i would want to change now (given that many
>> dependent packages).
> 
> Umm... if you're going to be repackaging anyway... then why not repackage
> now, and have it tested as such?

Ok, updated packages in testing. iconv (CSWiconv) has now be changed to
be an empty package (besides README.CSW and license) that pulls in
libiconv and libiconv_devel as dependencies. libiconv (CSWlibiconv) is
the package that one would want and the naming is consistent.

Is that along the lines of what you had in mind?

Only minor annoyance i see right now is that if I issue an "pkgutil -u
iconv" (pkgutil 1.6) against testing on a current box, the two new
packages are pulled in and libiconv is installed first (before updating
iconv to the empty package). pkgadd complains about files being used by
another package and prompts the user. OTOH, as libiconv is a dependency
of iconv, this installation order makes sense to me. Could this be a
problem somehow?

pkg-get (4.1.2) updates iconv first, so no prompts to the user there
(didn't test with 4.2.1 though).

Sebastian

From phil at bolthole.com  Sat Jul 25 03:23:19 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 18:23:19 -0700
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <4A6A4677.7000903@opencsw.org>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com>
	<4A6A4677.7000903@opencsw.org>
Message-ID: <20090725012319.GC73903@bolthole.com>

On Sat, Jul 25, 2009 at 01:40:39AM +0200, Sebastian Kayser wrote:
> Ok, updated packages in testing. iconv (CSWiconv) has now be changed to
> be an empty package (besides README.CSW and license) that pulls in
> libiconv and libiconv_devel as dependencies. libiconv (CSWlibiconv) is
> the package that one would want and the naming is consistent.
> 
> Is that along the lines of what you had in mind?

Um.... no :-}

when there are only two like this, with pre=exisiting dependancies, I think
it makes more sense to split into CSWiconv and a separate devel package
which depends on that one.
no empty package.


From bwalton at opencsw.org  Sat Jul 25 03:24:04 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 21:24:04 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724215550.GB73903@bolthole.com>
References: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
Message-ID: <1248484642-sup-7791@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 17:55:50 -0400 2009:
> sorry, misunderstanding. I thougth you meant "the PASSWORDS for database
> access, would be in a separate, non-checked-in file".

That's exactly what I meant.  I'll be creating a file called
includes/dbconfig.php.  This file will contain the host, database,
user and password info used for the connection to mysql.  All files
accessing the database will then include this file.  It's extra
overhead, yes, but completely negligible.

This new file won't be checked into the vcs and would need to be
created manually by anyone hacking on this stuff outside of the live
web tree.

Once this info is removed from each file and SQL queries are audited
for simple goofs (made easier by some braindead choices the php devs),
everything should be 'ready to go.'

On a different note, I'd like to propose a group called cswweb (or
similar) on bender with membership reflecting all those who've
expressed interest in modifying web content.  The htdocs (and maybe
cgi-bin?) files for the opencsw vhosts should then be made chgrp
cswweb, chmod g+w.  As it stands now, I'm not able to actually make
any of these changes.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/ea21a8f1/attachment.asc>

From dam at opencsw.org  Sat Jul 25 17:48:33 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 17:48:33 +0200
Subject: [csw-maintainers] Graphviz on OpenCSW
Message-ID: <A0FAA66B-21C8-40AD-A77F-702B561AC5C9@opencsw.org>

Hi,

once again OpenCSW is mentioned for Graphviz-download for a cool
ZFS visualization tool:
   <http://www.c0t0d0s0.org/archives/5589-ZFSviz-0.0001.html>


Best regards

   -- Dago

From dam at opencsw.org  Sat Jul 25 18:07:12 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 18:07:12 +0200
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <4A678CBB.2030208@opencsw.org>
References: <4A677CAB.8050808@opencsw.org> <4A678182.9020005@opencsw.org>
	<4A678CBB.2030208@opencsw.org>
Message-ID: <69092344-647F-4D93-BBDC-9F7A226338CB@opencsw.org>

Hi John,

Am 23.07.2009 um 00:03 schrieb John Ellson:
> On 07/22/2009 05:15 PM, John Ellson wrote:
>> On 07/22/2009 04:55 PM, John Ellson wrote:
>>> Dago,
>>>
>>> Any chance you could rebuild cairo to include SVG support, please?
>>>
>>> I can take a shot at it if you like?
>>
>> So apparently the reason its not provided is because there is a  
>> circular depdendency with librsvg ?
>>
>> Why does a circular dependency matter?    Try to install either and  
>> you get both - where is the ambiguity?

Because neither pkg-get nor pkgutil will handle circular dependencies  
(yet).

> It builds fine with svg enabled.  There is no library dependency on  
> librsvg.  I think it may just be a testing dependency.

Ok then, if it is not a dependency I guess it is ok. What does the  
rest of
the X11 team say?
William?
Phil?


Best regards

   -- Dago

From phil at bolthole.com  Sat Jul 25 18:11:23 2009
From: phil at bolthole.com (Philip Brown)
Date: Sat, 25 Jul 2009 09:11:23 -0700
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <69092344-647F-4D93-BBDC-9F7A226338CB@opencsw.org>
References: <4A677CAB.8050808@opencsw.org> <4A678182.9020005@opencsw.org>
	<4A678CBB.2030208@opencsw.org>
	<69092344-647F-4D93-BBDC-9F7A226338CB@opencsw.org>
Message-ID: <20090725161123.GA3902@bolthole.com>

On Sat, Jul 25, 2009 at 06:07:12PM +0200, Dagobert Michelsen wrote:
> Because neither pkg-get nor pkgutil will handle circular dependencies  
> (yet).
>
>> It builds fine with svg enabled.  There is no library dependency on  
>> librsvg.  I think it may just be a testing dependency.
>
> Ok then, if it is not a dependency I guess it is ok. What does the rest 
> of the X11 team say?
> William?
> Phil?

I said my piece a while back. remove the cycle, and submit it.
Half of that seems done now :-)


From dam at opencsw.org  Sat Jul 25 18:13:21 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 18:13:21 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A691BC9.2050209@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
Message-ID: <40569321-E5A1-46F1-A59E-71E4FE0272A8@opencsw.org>

Hi Mike,

Am 24.07.2009 um 04:26 schrieb Mike Watters:
> what about a compromise and have either a sourceforge or self-hosted  
> "members
> only" subversion,git,cvs,bzr,etc... (pick your favorite version  
> control
> software) for the source code.

For the record: We already have a SourceForge-project for
OpenCSW-specific scripts. You can browse at
   <http://opencsw.svn.sourceforge.net/viewvc/opencsw/>

Best regards

   -- Dago

From dam at opencsw.org  Sat Jul 25 18:18:26 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 18:18:26 +0200
Subject: [csw-maintainers] Including directory paths in prototypes using
	PKGFILES_CSWfoo
In-Reply-To: <1248455214-sup-2440@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>
	<1248455214-sup-2440@ntdws12.chass.utoronto.ca>
Message-ID: <39DFBD0A-74FB-47F0-8F24-DE805AA1F035@opencsw.org>

Hi Maciej,

Am 24.07.2009 um 19:07 schrieb Ben Walton:
> Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 24  
> 12:55:29 -0400 2009:
>> How to convince GAR to include the directories in the prototype?
>
> You don't want or need those entries.  They're stripped by GAR on
> purpose.  The CSWcommon package includes them.

Yes, exactly here:
   <http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/gar.pkg.mk#L305 
 >
The excluded pathes are in
   mgar/gar/v2/etc/commondirs-<arch>
They can be updates if CSWcommon gets updated with
   mgar/gar/v2/bin/update-commondirs
(not nice scripts, but can be hacked by me if necessary)


Best regards

   -- Dago

From trygvis at opencsw.org  Sat Jul 25 20:25:02 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sat, 25 Jul 2009 20:25:02 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090724160343.GC91227@bolthole.com>
References: <20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com>	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
Message-ID: <4A6B4DFE.1050307@opencsw.org>

Philip Brown wrote:
> On Fri, Jul 24, 2009 at 11:08:04AM +0200, Trygve Laugst?l wrote:
>> If the code can't be shown to the world without being a security risk,  
>> and Phil doesn't believe that we can write secure code, I'd like to  
>> replace the application with something that we all can work on and 
>> extend.
> 
> There is no "*THE* application".
> Additionally pretty much any web application, reguardless of language, has
> the potential for security holes. So changing the framework, does not
> definitively eliminate the underlying problem here.
> 
> This shows a fundamental non-understanding of what is going on in our web
> space (and security, for that matter).

It's impossible for us to know what's the current issue with the current 
application when your entire reply is "security".

I don't see how it should be hard to write a secure application to show 
the information that we currently display on the site.

> Just as people who arent actually maintaining packages, should not get to
> tell active maintainers how to package... I think that people who arent
> actually DOING WEB WORK, should not be telling those who are, how to do it.

It's a good thing that I do web development for a living then so I have 
the right to speak.

> If people want to actually do work, then great! please speak up with what
> area you want to work on. Then after that, talk about how you think the 
> working environment could be improved.

I'd like to improve the package info page, the first thing would be to 
remove the "doesn't work yet" comments etc.

--
Trygve

From ellson at opencsw.org  Sat Jul 25 21:38:03 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 15:38:03 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
Message-ID: <4A6B5F1B.6000303@opencsw.org>

I can build libcairo on build8s, but on build8x I'm getting:

     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this 
kernel with the arch 'i386'.  Stop.
     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
     make: *** [merge-isa-amd64] Error 2


Sould I be using a different build host?

John


From phil at bolthole.com  Sat Jul 25 22:30:35 2009
From: phil at bolthole.com (Philip Brown)
Date: Sat, 25 Jul 2009 13:30:35 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6B4DFE.1050307@opencsw.org>
References: <20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
Message-ID: <20090725203035.GA24170@bolthole.com>

On Sat, Jul 25, 2009 at 08:25:02PM +0200, Trygve Laugst?l wrote:
>> If people want to actually do work, then great! please speak up with what
>> area you want to work on. Then after that, talk about how you think the 
>> working environment could be improved.
>
> I'd like to improve the package info page, the first thing would be to  
> remove the "doesn't work yet" comments etc.

great. Please code something better. 
You konw where the code lives now. so please copy it to your own
~/public_html page, and tweak with it to your heart's content. if you get
it working better, wonderful! you can take over maintainnce of the official
page.


From ellson at opencsw.org  Sun Jul 26 01:07:58 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:07:58 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B5F1B.6000303@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org>
Message-ID: <4A6B904E.5050103@opencsw.org>

On 07/25/2009 03:38 PM, John Ellson wrote:
> I can build libcairo on build8s, but on build8x I'm getting:
>
>     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this 
> kernel with the arch 'i386'.  Stop.
>     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>     make: *** [merge-isa-amd64] Error 2
>
>
> Sould I be using a different build host?
>

OK, I see, its from BUILD64=1, and I'm supposed to do:

	build8x $ gmake merge   (builds all 32 bit ISAs and merges them)
	build10x $ gmake merge   (builds the remaining ISAs and merges them)
	 build8x $ gmake package (makes package, so it is installable on Solaris 8)

Painful!   And it doesn't work either due to this other previously reported problem with libm:


	One more thing: What you are trying to do here

	>/  # Solaris 10 on build10x has libm.so which points to libm.so.2.//libm.so.2
/	>/  # is however not available on Solaris 8. So we have to explicitly use
/	>/  # libm.so.1 during linking and get rid of the -lm reference. Otherwise
/	>/  # the package check on build8x when running "gmake package" fails.
/
	is also a flaw of checkpkg: When a package is build on Solaris 8 and Solaris 10
	the complete package may fail when checkpkg'ed on Solaris 8 with the reason above.
	You were able to link to a specific version, but this may not always be possible.
	I have no idea on how this could be fixed in checkpkg apart from not making
	mixed packages or checking only on Solaris 10.

So, Dago, how were you building libcairo before?

John



From ellson at opencsw.org  Sun Jul 26 01:12:13 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:12:13 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B904E.5050103@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B904E.5050103@opencsw.org>
Message-ID: <4A6B914D.3040903@opencsw.org>

On 07/25/2009 07:07 PM, John Ellson wrote:
> On 07/25/2009 03:38 PM, John Ellson wrote:
>> I can build libcairo on build8s, but on build8x I'm getting:
>>
>>     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this 
>> kernel with the arch 'i386'.  Stop.
>>     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>>     make: *** [merge-isa-amd64] Error 2
>>
>>
>> Sould I be using a different build host?
>>
>
> OK, I see, its from BUILD64=1, and I'm supposed to do:
>
>     build8x $ gmake merge   (builds all 32 bit ISAs and merges them)
>     build10x $ gmake merge   (builds the remaining ISAs and merges them)
>      build8x $ gmake package (makes package, so it is installable on 
> Solaris 8)
>
> Painful!   And it doesn't work either due to this other previously 
> reported problem with libm:
>
>
>     One more thing: What you are trying to do here
>
> >/  # Solaris 10 on build10x has libm.so which points to 
> libm.so.2.//libm.so.2
> / >/  # is however not available on Solaris 8. So we have to 
> explicitly use
> / >/  # libm.so.1 during linking and get rid of the -lm reference. 
> Otherwise
> / >/  # the package check on build8x when running "gmake package" fails.
> /
>     is also a flaw of checkpkg: When a package is build on Solaris 8 
> and Solaris 10
>     the complete package may fail when checkpkg'ed on Solaris 8 with 
> the reason above.
>     You were able to link to a specific version, but this may not 
> always be possible.
>     I have no idea on how this could be fixed in checkpkg apart from 
> not making
>     mixed packages or checking only on Solaris 10.
>
> So, Dago, how were you building libcairo before?
>


gmake ENABLE_CHECK=0 package

seems to work.    Doesn't feel like the right solution?


John


From mwatters at opencsw.org  Sun Jul 26 01:14:24 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Jul 2009 18:14:24 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <40569321-E5A1-46F1-A59E-71E4FE0272A8@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>	<20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<40569321-E5A1-46F1-A59E-71E4FE0272A8@opencsw.org>
Message-ID: <4A6B91D0.4070808@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Dago,
Dagobert Michelsen wrote:
> Hi Mike,
> 
> For the record: We already have a SourceForge-project for
> OpenCSW-specific scripts. You can browse at
>   <http://opencsw.svn.sourceforge.net/viewvc/opencsw/>

Yes, I know about that repository, but I believe it is public??
and didn't meet the criteria for "security"

This repo IMHO would be just fine as long as we get the files re factored to
use include files for sensitive configs.


- --
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/

iEYEARECAAYFAkprkc8ACgkQLrhmsXMSLxfRbgCeMhkF9jBsw8rp2Wjx3l0/I3VO
Q3gAmweFS0q45q/SmZMfcTBuXzCvlD4F
=/Z/c
-----END PGP SIGNATURE-----

From mwatters at opencsw.org  Sun Jul 26 01:17:41 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Jul 2009 18:17:41 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6B4DFE.1050307@opencsw.org>
References: <20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>	<20090723220149.GB8541@bolthole.com>	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
Message-ID: <4A6B9295.20506@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Trygve,
Trygve Laugst?l wrote:

> I'd like to improve the package info page, the first thing would be to
> remove the "doesn't work yet" comments etc.
> 

I agree, however, I would recommend putting efforts toward the mock-up format
William has been working on. http://www-mockup.opencsw.org/


- --
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/

iEYEARECAAYFAkprkpQACgkQLrhmsXMSLxf0kACg05D0JyMm9sNJNP5eNSlg2ANf
kv8AoOunOLFCsk8vamgFz390+NRDjZmf
=g9o9
-----END PGP SIGNATURE-----

From mwatters at opencsw.org  Sun Jul 26 01:20:38 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Jul 2009 18:20:38 -0500
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B5F1B.6000303@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org>
Message-ID: <4A6B9346.5060405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi John,
John Ellson wrote:
> I can build libcairo on build8s, but on build8x I'm getting:
> 
>     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this
> kernel with the arch 'i386'.  Stop.
>     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>     make: *** [merge-isa-amd64] Error 2
> 
> 

That error is from gcc
if you plan on using gcc to build 64bit binaries, you *Must* use gcc4

I am not positive, but I believe libcairo will build using Sun Studio.

- --
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/

iEYEARECAAYFAkprk0UACgkQLrhmsXMSLxfLnACgwBUiljv/5Ft7L8ubd/tZHVxJ
caUAoLwpeK0+jSqeQTJx9daCnzmFokfM
=CorV
-----END PGP SIGNATURE-----

From ellson at opencsw.org  Sun Jul 26 01:46:24 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:46:24 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B9346.5060405@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B9346.5060405@opencsw.org>
Message-ID: <4A6B9950.6000507@opencsw.org>

On 07/25/2009 07:20 PM, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi John,
> John Ellson wrote:
>    
>> I can build libcairo on build8s, but on build8x I'm getting:
>>
>>      gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this
>> kernel with the arch 'i386'.  Stop.
>>      gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>>      make: *** [merge-isa-amd64] Error 2
>>
>>
>>      
> That error is from gcc
> if you plan on using gcc to build 64bit binaries, you *Must* use gcc4
>
> I am not positive, but I believe libcairo will build using Sun Studio.
>
>
>    

No, its from the Sun Studio builds on build8x exactly as Dago had it before.
All I'm doing is enabling the SVG backend and rebuilding.


The trick seems to be:

	build8x $ gmake clean
	build8x $ gmake build-isa-i386
	build10x $ gmake merge
	build8x $ gmake ENABLE_CHECK=0 package/
	/build8s $ gmake clean
	build8s $ gmake
	build8s $ gmake package


John

From ellson at opencsw.org  Sun Jul 26 01:47:49 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:47:49 -0400
Subject: [csw-maintainers] libcairo with SVG support now available in testing
Message-ID: <4A6B99A5.6060100@opencsw.org>

The only change in this version of libcairo is to enable SVG support.


John




From dam at opencsw.org  Sun Jul 26 09:30:49 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 09:30:49 +0200
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B5F1B.6000303@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org>
Message-ID: <FED7C188-310D-4DE0-994A-D0EA3EC9A238@opencsw.org>

Hi John,

Am 25.07.2009 um 21:38 schrieb John Ellson:
> I can build libcairo on build8s, but on build8x I'm getting:
>
>    gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this  
> kernel with the arch 'i386'.  Stop.
>    gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>    make: *** [merge-isa-amd64] Error 2
>
> Sould I be using a different build host?

This is normal. For x86 you do

build8x $ gmake merge (builds all 32 bit ISAs and merges them)
build10x $ gmake merge (builds the remaining ISAs and merges them)
build8x $ gmake package (makes package, so it is installable on  
Solaris 8)

For details see
   <http://sourceforge.net/apps/trac/gar/wiki/Building%20for%20multiple%20ISAs 
 >


Best regards

   -- Dago

From dam at opencsw.org  Sun Jul 26 09:32:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 09:32:43 +0200
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B904E.5050103@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B904E.5050103@opencsw.org>
Message-ID: <62019BB1-EA43-4B7B-B1EE-1B86429CA3D4@opencsw.org>

Hi John,

Am 26.07.2009 um 01:07 schrieb John Ellson:
> Painful!   And it doesn't work either due to this other previously  
> reported problem with libm:
>
> 	One more thing: What you are trying to do here
>
> 	>/  # Solaris 10 on build10x has libm.so which points to libm.so. 
> 2.//libm.so.2
> /	>/  # is however not available on Solaris 8. So we have to  
> explicitly use
> /	>/  # libm.so.1 during linking and get rid of the -lm reference.  
> Otherwise
> /	>/  # the package check on build8x when running "gmake package"  
> fails.
> /
> 	is also a flaw of checkpkg: When a package is build on Solaris 8  
> and Solaris 10
> 	the complete package may fail when checkpkg'ed on Solaris 8 with  
> the reason above.
> 	You were able to link to a specific version, but this may not  
> always be possible.
> 	I have no idea on how this could be fixed in checkpkg apart from  
> not making
> 	mixed packages or checking only on Solaris 10.
>
> So, Dago, how were you building libcairo before?

Basically you do
   ENABLE_CHECK=0 gmake repackage
once again to disable this check. checkpkg thinkgs the libm.so.2
should belong to the build host, but it is only for the Solaris 10
binaries, so it is ok here.


Best regards

   -- Dago

From dam at opencsw.org  Sun Jul 26 09:45:03 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 09:45:03 +0200
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B9950.6000507@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B9346.5060405@opencsw.org>
	<4A6B9950.6000507@opencsw.org>
Message-ID: <C05A5D9D-46A0-4170-B4B6-B274820DFA3A@opencsw.org>

Hi John,

Am 26.07.2009 um 01:46 schrieb John Ellson:
>> That error is from gcc

No, it is from GAR and it is normal when you build 64 bit.

> No, its from the Sun Studio builds on build8x exactly as Dago had it  
> before.
> All I'm doing is enabling the SVG backend and rebuilding.
>
> The trick seems to be:
>
> 	build8x $ gmake clean
> 	build8x $ gmake build-isa-i386
> 	build10x $ gmake merge

You should merge the 32 bit build on Solaris 8 or you may
get errors on usage later. I clarified this on
   <http://sourceforge.net/apps/trac/gar/wiki/Building%20for%20multiple%20ISAs 
 >
That is
   build8x $ gmake clean
   build8x $ gmake merge
   build10x $ gmake merge
   build8x $ ENABLE_CHECK=0 gmake package

> 	/build8s $ gmake clean
> 	build8s $ gmake
> 	build8s $ gmake package


You could skip the 'gmake' and do 'gmake package' right away


Best regards

   -- Dago

From trygvis at opencsw.org  Sun Jul 26 11:34:57 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sun, 26 Jul 2009 11:34:57 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090725203035.GA24170@bolthole.com>
References: <20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com>	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
Message-ID: <4A6C2341.8060600@opencsw.org>

Philip Brown wrote:
> On Sat, Jul 25, 2009 at 08:25:02PM +0200, Trygve Laugst?l wrote:
>>> If people want to actually do work, then great! please speak up with what
>>> area you want to work on. Then after that, talk about how you think the 
>>> working environment could be improved.
>> I'd like to improve the package info page, the first thing would be to  
>> remove the "doesn't work yet" comments etc.
> 
> great. Please code something better. 
> You konw where the code lives now. so please copy it to your own
> ~/public_html page, and tweak with it to your heart's content. if you get
> it working better, wonderful! you can take over maintainnce of the official
> page.

Ok, I'll do that.

Can you explain what the security reason for not putting it an SCM is? 
The last reply I got was "security" but that's not enough. I can't think 
of any security risks involved with a read-only application and if I'm 
going to develop on it I'd like to know what to look out for.

--
Trygve

From skayser at opencsw.org  Sun Jul 26 14:28:23 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sun, 26 Jul 2009 14:28:23 +0200 (CEST)
Subject: [csw-maintainers] Updated libiconv in testing,
 testing	appreciated
In-Reply-To: <20090725012319.GC73903@bolthole.com>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com> <4A6A4677.7000903@opencsw.org>
	<20090725012319.GC73903@bolthole.com>
Message-ID: <50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>

Philip Brown wrote:
> On Sat, Jul 25, 2009 at 01:40:39AM +0200, Sebastian Kayser wrote:
>> Ok, updated packages in testing. iconv (CSWiconv) has now be changed to
>> be an empty package (besides README.CSW and license) that pulls in
>> libiconv and libiconv_devel as dependencies. libiconv (CSWlibiconv) is
>> the package that one would want and the naming is consistent.
>>
>> Is that along the lines of what you had in mind?
>
> Um.... no :-}
>
> when there are only two like this, with pre=exisiting dependancies, I
> think
> it makes more sense to split into CSWiconv and a separate devel package
> which depends on that one.
> no empty package.

Before i commence take #3, let me try to understand:

Upstream name is libiconv, right now we have it as CSWiconv (pkgname) and
libiconv (swname). Would be nice to have CSWlibiconv and libiconv instead
to match the upstream name (see my previous attempt), but this introduces
the transitory, empty package CSWiconv for backwards compatibility and
implies long-term pkg dependency updating. Thus it is less hassle to just
leave the package name as CSWiconv and change its software name to iconv
(implies that libiconv as software name is dropped from the catalog)?

Sebastian


From mwatters at opencsw.org  Sun Jul 26 16:28:04 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 26 Jul 2009 09:28:04 -0500
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
References: <4A6A21A0.3060006@opencsw.org>	<20090724215452.GA73903@bolthole.com>
	<4A6A4677.7000903@opencsw.org>	<20090725012319.GC73903@bolthole.com>
	<50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
Message-ID: <4A6C67F4.30900@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Sebastian,
Sebastian Kayser wrote:

> Before i commence take #3, let me try to understand:
> 
> Upstream name is libiconv, right now we have it as CSWiconv (pkgname) and
> libiconv (swname). Would be nice to have CSWlibiconv and libiconv instead
> to match the upstream name (see my previous attempt), but this introduces
> the transitory, empty package CSWiconv for backwards compatibility and
> implies long-term pkg dependency updating. Thus it is less hassle to just
> leave the package name as CSWiconv and change its software name to iconv
> (implies that libiconv as software name is dropped from the catalog)?
> 

The best way (and AFAIK the formal way) to handle such a conversion is as follows:

1) create the package as CSWlibiconv
	1a) set CSWlibiconv incompatible with CSWiconv
2) rebuild CSWiconv as an empty meta-package
	2a) create a dependency on CSWlibiconv
3) push both packages through the normal test/newpkgs/current cycle
4) file a bug against all packages that depend on CSWiconv
   to rebuild against CSWlibiconv.


- --
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/

iEYEARECAAYFAkpsZ/QACgkQLrhmsXMSLxcdHwCeOo/32plYwXZp4wmQK/joDBh1
iKUAn0fgtRPI4Ypvl9VHFlsfmOkfo8ti
=bCRt
-----END PGP SIGNATURE-----

From rupert at opencsw.org  Sun Jul 26 16:55:00 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 26 Jul 2009 16:55:00 +0200
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3 released!
In-Reply-To: <6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
Message-ID: <6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>

hi,

could it be possible that the type of providing the user homedir used
on the csw servers influence the test result of the mercurial package?
see the bottom of belows mail from mads.

i will try in the next weeks on a std filesystem as well with a std checkout.

rupert.

---------- Forwarded message ----------
From: Mads Kiilerich <mads at kiilerich.com>
Date: 2009/7/25
Subject: Re: Solaris package, was Re: Mercurial 1.3 released!
To: "rupert.thurner" <rupert.thurner at gmail.com>
Cc: mercurial at selenic.com, Trygve Laugst?l <trygvis at inamo.no>


rupert.thurner wrote, On 07/25/2009 11:52 AM:
>
> On Jul 5, 1:23 pm, Mads Kiilerich<mads at kiilerich.com> ?wrote:
>
>>
>> rupert.thurner wrote, On 07/04/2009 08:13 PM:
>>
>>
>>>
>>> we tried to package it for solaris,http://opencsw.org. some test
>>> cases fail. some failures are because the test suite did not use /opt/
>>> csw/bin/gdiff, but "diff" from solaris. how do we specify the diff
>>> tool used, or the path to it?
>>>
>>
>> The testsuite uses whatever diff tool is in its path. So you could just
>> set PATH correctly before running the test suite.
>>
>>
>>>
>>> with some other tests i have no idea where the failure is coming from.
>>> what would be the best option to track this down? seehttp://pastebin.com/m32e3576f
>>> for the complete test result.
>>>
>>
>> test-repair-strip shouldn't rely on the cat error message. And the
>> warning in test-gpg should probably just be ignored. Patches has been
>> posted to -devel.
>>
>> Just ignore these tests for now. ?You can temporarily remove their x bit
>> to avoid running them.
>>
>> That will make the failure list shorter and make it easier to spot:
>>
>>
>>>
>>> an excerpt:
>>>
>>
>>
>>>
>>> ERROR: test-transplant output changed
>>> --- Expected output
>>> +++ Test output
>>> @@ -145,20 +145,20 @@
>>> ? % test filter
>>> ? filtering
>>> ? applying 17ab29e464c6
>>> -17ab29e464c6 transplanted to e9ffc54ea104
>>> +17ab29e464c6 transplanted to 88e81fba5b32
>>> ? filtering
>>>
>>
>> This kind of errors looks worse. Do you consistently get exactly the
>> same errors?
>>
>
> i run the tests again with mercurial-1.3.1, see the result here:
> http://pastebin.com/m1a2f283f. especially
>

It seems like pastebin obscures the output slightly. Perhaps it would
be easier to read if it didn't use Python highlighting.

http://www.selenic.com/hg/index.cgi/rev/6e5e548452de (test-gpg:
suppress secure memory warning) should probably have been applied to
stable as well. Anyone?

Regarding test-repair-strip: One of the patches from
http://www.selenic.com/pipermail/mercurial-devel/2009-July/thread.html#13610
should be applied to stable.

You are still running in an environment without GNU diff. And
apparently also an unsupported version of cvs.

The rest of the problems seems to have something to do with something
not being like someone thinks they should be, perhaps forking or
caching or file systems.

Do you get the same failures when you run "make tests" manually from a
source checkout?

From phil at bolthole.com  Sun Jul 26 17:38:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 08:38:56 -0700
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3	released!
In-Reply-To: <6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
	<6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
Message-ID: <20090726153856.GA28096@bolthole.com>

On Sun, Jul 26, 2009 at 04:55:00PM +0200, rupert THURNER wrote:
> hi,
> 
> could it be possible that the type of providing the user homedir used
> on the csw servers influence the test result of the mercurial package?
> see the bottom of belows mail from mads.

I dont think it has much to do with the "homedirs". 
Seems like, according to the bug report, you need to force mercurial to use
"gdiff" instead of "diff".
Makes sense. Some software is like that.



From phil at bolthole.com  Sun Jul 26 17:43:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 08:43:45 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6C2341.8060600@opencsw.org>
References: <20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
	<4A6C2341.8060600@opencsw.org>
Message-ID: <20090726154345.GB28096@bolthole.com>

On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst??l wrote:
> Can you explain what the security reason for not putting it an SCM is?  

"a" source code control mechanism, I dont have much problem with.
putting it in a PUBLIC one, is where my concerns lay.

Beyond that, I dont see the need to force everyone to use one particular
type of scm, just for small web page "code".
If the assorted people working on opencsw web stuff agree to standardize on
one internal-only scheme, i'd go along with it. But frankly, most of the
web related stuff is small enough that the old cycle of

"make a save copy, tweak with the new stuff, publish it, forget about old
 stuff"

is perfectly adequate.

From phil at bolthole.com  Sun Jul 26 17:49:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 08:49:31 -0700
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com>
	<4A6A4677.7000903@opencsw.org>
	<20090725012319.GC73903@bolthole.com>
	<50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
Message-ID: <20090726154931.GC28096@bolthole.com>

On Sun, Jul 26, 2009 at 02:28:23PM +0200, Sebastian Kayser wrote:
> 
> Upstream name is libiconv, right now we have it as CSWiconv (pkgname) and
> libiconv (swname). Would be nice to have CSWlibiconv and libiconv instead
> to match the upstream name (see my previous attempt), [...]
>  Thus it is less hassle to just
> leave the package name as CSWiconv and change its software name to iconv
> (implies that libiconv as software name is dropped from the catalog)?

true, that is the least hassle thing to do.
Howeer, I think that it is nice that the swname matches the upstream name
"libiconv". 

And going the other way, changing the PKG name, of such an oooold, and
crucial package, makes me shudder to think about it.
[by neccessity, it was one of the very first packages built and released, ever]
I'd rather we just leave well enough alone and not change either.

From phil at bolthole.com  Sun Jul 26 18:31:22 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 09:31:22 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248484642-sup-7791@ntdws12.chass.utoronto.ca>
References: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
Message-ID: <20090726163122.GE28096@bolthole.com>

On Fri, Jul 24, 2009 at 09:24:04PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 17:55:50 -0400 2009:
> > sorry, misunderstanding. I thougth you meant "the PASSWORDS for database
> > access, would be in a separate, non-checked-in file".
> 
> That's exactly what I meant.  I'll be creating a file called
> includes/dbconfig.php.  This file will contain the host, database,
> user and password info used for the connection to mysql.  All files
> accessing the database will then include this file.  It's extra
> overhead, yes, but completely negligible.

hrmmm...
well, in extreme cases(high loads), i've seen use of php includes,
dramatically slow down pages. so it's not always negligible.


> This new file won't be checked into the vcs and would need to be
> created manually by anyone hacking on this stuff outside of the live
> web tree.

there's also the concern of someone inevitably doing the equivalent of
"checkin *", and having the file "accidentally" get into the version
control system.


> On a different note, I'd like to propose a group called cswweb (or
> similar) on bender with membership reflecting all those who've
> expressed interest in modifying web content.  The htdocs (and maybe
> cgi-bin?) files for the opencsw vhosts should then be made chgrp
> cswweb, chmod g+w.  As it stands now, I'm not able to actually make
> any of these changes.

and in some ways, lack of thats a good thing ;-) that one person cant just
decide to drastically overhaul everything on a whim. 
(well, ok, people with root-type access 'could' do that. but we just
 dont do that sort of thing)

long term, i think having a web unixgroup makes sense.
How about you start with one page for now though. the "packages" page was
what you expressed interest in, if I recall?

From ja at opencsw.org  Sun Jul 26 20:09:01 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Sun, 26 Jul 2009 20:09:01 +0200
Subject: [csw-maintainers] Cacti 0.8.7e in testing
Message-ID: <op.uxospam6jbmpvo@imac.lan>

Hi all,

I put cacti 0.8.7e into testing.

I'm not sure about the location of the files (especially the php stuff).  
Maybe someone could take a look?

Thank you,

Juergen

-- 
Juergen Arndt

From dam at opencsw.org  Sun Jul 26 20:16:29 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 20:16:29 +0200
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3	released!
In-Reply-To: <20090726153856.GA28096@bolthole.com>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
	<6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
	<20090726153856.GA28096@bolthole.com>
Message-ID: <D92A2BB1-12C2-49A3-9C69-FEE730E3EE21@opencsw.org>

Hi Rupert,

Am 26.07.2009 um 17:38 schrieb Philip Brown:
> On Sun, Jul 26, 2009 at 04:55:00PM +0200, rupert THURNER wrote:
>> could it be possible that the type of providing the user homedir used
>> on the csw servers influence the test result of the mercurial  
>> package?
>> see the bottom of belows mail from mads.
>
> I dont think it has much to do with the "homedirs".
> Seems like, according to the bug report, you need to force mercurial  
> to use
> "gdiff" instead of "diff".
> Makes sense. Some software is like that.

In GAR you can do this with

> include gar/category.mk
>
> # Make sure we are using gnu tools (esp. sed)
> PATH := /opt/csw/gnu:$(PATH)

There are quite a lot of examples for this kind of use. See

> build8s% grep csw/gnu pkg/*/trunk/Makefile


Best regards

   -- Dago

From dam at opencsw.org  Sun Jul 26 20:18:59 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 20:18:59 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090726163122.GE28096@bolthole.com>
References: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
	<20090726163122.GE28096@bolthole.com>
Message-ID: <3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>

Hi Phil,

Am 26.07.2009 um 18:31 schrieb Philip Brown:
>> This new file won't be checked into the vcs and would need to be
>> created manually by anyone hacking on this stuff outside of the live
>> web tree.
>
> there's also the concern of someone inevitably doing the equivalent of
> "checkin *", and having the file "accidentally" get into the version
> control system.

With Subversion you explicitly add files once. When you
commit you automatically commit only added files.
BTW, you could also restrict access to the DB to localhost.


Best regards

   -- Dago

From bwalton at opencsw.org  Sun Jul 26 20:40:57 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Jul 2009 14:40:57 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
References: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
	<20090726163122.GE28096@bolthole.com>
	<3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
Message-ID: <1248633364-sup-220@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Jul 26 14:18:59 -0400 2009:
> With Subversion you explicitly add files once. When you
> commit you automatically commit only added files.

Also, you can explicitly add the file to an ignore list so that you'd
have to intentionally circumvent the vcs package to add/commit the
file.

> BTW, you could also restrict access to the DB to localhost.

...And a user that is limited to SELECT() (read only) would also go a
long way toward securing the access.  Even with access limited to
localhost, SQL injection with a read/write account is still a threat.
Let MySQL remove that threat (barring bugs in it's ACL enforcement)
for you.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090726/af340507/attachment.asc>

From phil at bolthole.com  Sun Jul 26 20:54:05 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 11:54:05 -0700
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3	released!
In-Reply-To: <D92A2BB1-12C2-49A3-9C69-FEE730E3EE21@opencsw.org>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
	<6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
	<20090726153856.GA28096@bolthole.com>
	<D92A2BB1-12C2-49A3-9C69-FEE730E3EE21@opencsw.org>
Message-ID: <20090726185405.GD39975@bolthole.com>

On Sun, Jul 26, 2009 at 08:16:29PM +0200, Dagobert Michelsen wrote:
> In GAR you can do this with
>
>> include gar/category.mk
>>
>> # Make sure we are using gnu tools (esp. sed)
>> PATH := /opt/csw/gnu:$(PATH)
>

please note, though, that will ONLY do it for build time tests. If the
program itself needs gdiff when installed.. you need to be more drastic!

From phil at bolthole.com  Sun Jul 26 20:57:23 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 11:57:23 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
References: <20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
	<20090726163122.GE28096@bolthole.com>
	<3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
Message-ID: <20090726185723.GE39975@bolthole.com>

On Sun, Jul 26, 2009 at 08:18:59PM +0200, Dagobert Michelsen wrote:
>> there's also the concern of someone inevitably doing the equivalent of
>> "checkin *", and having the file "accidentally" get into the version
>> control system.
>
> With Subversion you explicitly add files once.


and someone adding new files, could still do

[add-once-command] *


but its potentially nice that you can exclude, as Ben says. need to hear
more about that.


From dam at opencsw.org  Sun Jul 26 23:19:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 23:19:55 +0200
Subject: [csw-maintainers] General call: stuff in testing/
Message-ID: <4EB6CF3B-265B-489B-BA55-8FA33067B037@opencsw.org>

Hi,

there are quite a lot of packages in testing/ (300 files now). Maybe  
some
cool packages has been forgotten to release? It would be nice if  
everyone
could have a look at his packages and discard/release the package or  
leave
it there if it was for a reason that it is there ;-)

Status for my packages in testing:

- jdk/jre: still no permission from Sun for redistribution
- tcl: still not understood how locations on version upgrades work
- orca: start/stop methods still missing
- libgss: doubts from Phil that it is useful in addition to Suns version
- libgsf/libgda: linker issues as reported on maintainers@
- glib 1.2.10: just a rebuild with GAR, maybe released
- esound: needs more testing. volunteers?
- openldap: waits for berkeleydb general overhaul
- gdb 6.8: needs more testing
- other stuff: released to current, will be removed when arrived at  
mirrors

It would be nice if you all could also give a status, maybe some other
maintainers can help fixing things.


Thanks!

   -- Dago

From bwalton at opencsw.org  Sun Jul 26 23:35:25 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Jul 2009 17:35:25 -0400
Subject: [csw-maintainers] General call: stuff in testing/
In-Reply-To: <4EB6CF3B-265B-489B-BA55-8FA33067B037@opencsw.org>
References: <4EB6CF3B-265B-489B-BA55-8FA33067B037@opencsw.org>
Message-ID: <1248644012-sup-613@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Jul 26 17:19:55 -0400 2009:

Status for my packages in testing:

- ruby*: rolled today to address the recent security vulnerability in
  the previous release.  Will henceforth include a rubymode (emacs major mode)
  and rubymode_el (elisp source) subpackage.

I haven't installed these myself yet, but the standard ruby packages
should be sound...it's the rubymode* ones that may still have
packaging 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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090726/189664e9/attachment.asc>

From trygvis at opencsw.org  Sun Jul 26 23:51:07 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sun, 26 Jul 2009 23:51:07 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090726154345.GB28096@bolthole.com>
References: <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>	<20090725203035.GA24170@bolthole.com>	<4A6C2341.8060600@opencsw.org>
	<20090726154345.GB28096@bolthole.com>
Message-ID: <4A6CCFCB.2040204@opencsw.org>

Philip Brown wrote:
> On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst?l wrote:
>> Can you explain what the security reason for not putting it an SCM is?  
> 
> "a" source code control mechanism, I dont have much problem with.
> putting it in a PUBLIC one, is where my concerns lay.

Ok, that's good to know even if I don't see why it shouldn't be in a 
public one.

I'm still wondering what those "security" concerns was.

> Beyond that, I dont see the need to force everyone to use one particular
> type of scm, just for small web page "code".
> If the assorted people working on opencsw web stuff agree to standardize on
> one internal-only scheme, i'd go along with it. But frankly, most of the
> web related stuff is small enough that the old cycle of
> 
> "make a save copy, tweak with the new stuff, publish it, forget about old
>  stuff"
> 
> is perfectly adequate.

I think that's a point we just have to agree to disagree on.

--
Trygve

From trygvis at opencsw.org  Sun Jul 26 23:53:17 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sun, 26 Jul 2009 23:53:17 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <op.uxfljpnjjbmpvo@imac.lan>
References: <op.uw59fh0kjbmpvo@imac.lan>
	<4A60C3AD.7050800@opencsw.org>	<4A660D38.7030600@opencsw.org>
	<op.uxfljpnjjbmpvo@imac.lan>
Message-ID: <4A6CD04D.3090500@opencsw.org>

Juergen Arndt wrote:
> On Tue, 21 Jul 2009 20:47:20 +0200, Trygve Laugst?l 
> <trygvis at opencsw.org> wrote:
> 
>> Trygve Laugst?l wrote:
>>> Juergen Arndt wrote:
>>>> Hi all,
>>>>
>>>> I updated the packages for Munin 1.3.4 in testing.
>>>>
>>>> - sample apache config (ready for include) in 
>>>> /etc/opt/csw/munin/apache/
>>>> - munin plugins moved to /opt/csw/libexec
>>>> - fix for using monospaced font
>>>>
>>>> Feedback is welcome.
>>>  I'll try these when I get back home next week.
>>
>> Can you do a rebuild with ARCH=all so I can test them on my x86 box?
> 
> In testing there are platform independent packages:
> 
> munin_master-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz
> munin_node-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz

They seem to work perfectly for me, I'm +1 for releasing those to current.

--
Trygve

From maciej at opencsw.org  Mon Jul 27 14:57:01 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 27 Jul 2009 13:57:01 +0100
Subject: [csw-maintainers] Shared library versions on build8st
Message-ID: <ce97543b0907270557q1bb02e79gebb73eb78e6790c5@mail.gmail.com>

The host build8s has CSWoldaprt-2.3. build8st on the other hand, has
CSWoldaprt-2.4. Binaries that depend on ldap, built on build8s, don't
work on build8st. I temporarily downgraded CSWoldaprt on build8st to
2.3, but I'm guessing I might be clashing with Dago who is working on
ldap. Perhaps we should coordinate who is working at the moment.

The current state is - openldap runtime 2.3. I'm testing cups-1.3.11.
I'll send an update when I'm done with it.

From maciej at opencsw.org  Mon Jul 27 18:01:15 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 27 Jul 2009 17:01:15 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
Message-ID: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>

I'm having an issue with the packaging of CUPS. It was all jolly good
until prototypes of some packages started becoming empty. It once was
CSWlibcups. Its prototype was simply empty, and so was the package.
Now libcups has the correct files, but cupsclient became empty. It's
reproducible with my account on both build8{s,x}. The build is at:

https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/pkg/cups/branches/cups-with-classutils

It's weird that the packages have been packaged correctly before, and
I haven't quite caught the moment the problem appeared.

Have you ever had similar issues? Is there a fix or a workaround?

Maciej

From dam at opencsw.org  Mon Jul 27 19:08:12 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 27 Jul 2009 19:08:12 +0200
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
Message-ID: <177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>

Hi Maciej,

Am 27.07.2009 um 18:01 schrieb Maciej (Matchek) Blizinski:
> I'm having an issue with the packaging of CUPS. It was all jolly good
> until prototypes of some packages started becoming empty. It once was
> CSWlibcups. Its prototype was simply empty, and so was the package.
> Now libcups has the correct files, but cupsclient became empty.

You forgot to close a bracket, see r5673.

> It's weird that the packages have been packaged correctly before, and
> I haven't quite caught the moment the problem appeared.

Perl bails out and all packages go into other packages. You can
track this with
   DEBUG_PACKAGING=1 gmake repackage

> Have you ever had similar issues? Is there a fix or a workaround?

Yes, write correct regexps ;-) But I do say that it is hard
sometimes...


Best regards

   -- Dago

From maciej at opencsw.org  Mon Jul 27 19:24:21 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 27 Jul 2009 18:24:21 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
	<177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>
Message-ID: <ce97543b0907271024k1fbc5119xe52c6700c416956f@mail.gmail.com>

On Mon, Jul 27, 2009 at 6:08 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
>> It's weird that the packages have been packaged correctly before, and
>> I haven't quite caught the moment the problem appeared.
>
> Perl bails out and all packages go into other packages. You can
> track this with
> ?DEBUG_PACKAGING=1 gmake repackage

A general idea... make a 'catch all' package, which is supposed to be
empty. If it contains any files, it means that some files haven't been
matched in other packages. Or perhaps mGAR could have such check:
"have all the regexes caught all the files from the install
directory?" What do you think Dago, what would be the best way to go
about it?

Maciej

From dam at opencsw.org  Mon Jul 27 20:23:14 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 27 Jul 2009 20:23:14 +0200
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <ce97543b0907271024k1fbc5119xe52c6700c416956f@mail.gmail.com>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
	<177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>
	<ce97543b0907271024k1fbc5119xe52c6700c416956f@mail.gmail.com>
Message-ID: <C19FB3C7-1DE7-4036-98D9-D9A9CCC788BA@opencsw.org>

Hi Maciej,

Am 27.07.2009 um 19:24 schrieb Maciej (Matchek) Blizinski:
> A general idea... make a 'catch all' package, which is supposed to be
> empty.

There is already a catch-all package: the one for which you don't
specify PKGFILES_*

> If it contains any files, it means that some files haven't been
> matched in other packages.

Usually you split off certain functions into different packages:
documentation, development... and a main package - that main
package is meant to be the catch-all package.

> Or perhaps mGAR could have such check:
> "have all the regexes caught all the files from the install
> directory?" What do you think Dago, what would be the best way to go
> about it?

IMHO for the general case the current implementation is optimal.
In most cases you just say
   PKGFILES_CSWpkgdevel = $(PKGFILES_DEVEL)
and be done with it. I think it is important to make the most common
case easy, while still making hard things possible, and that is what
the current implementation does. I am not opposed to enhancements
but I can't think of a real enhancement right now.


Best regards

   -- Dago

From phil at bolthole.com  Mon Jul 27 23:21:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 27 Jul 2009 14:21:17 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6CCFCB.2040204@opencsw.org>
References: <20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
	<4A6C2341.8060600@opencsw.org>
	<20090726154345.GB28096@bolthole.com>
	<4A6CCFCB.2040204@opencsw.org>
Message-ID: <20090727212117.GG90811@bolthole.com>

On Sun, Jul 26, 2009 at 11:51:07PM +0200, Trygve Laugst?l wrote:
> Philip Brown wrote:
>> On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst?l wrote:
>>> Can you explain what the security reason for not putting it an SCM 
>>> is?  
>>
>> "a" source code control mechanism, I dont have much problem with.
>> putting it in a PUBLIC one, is where my concerns lay.
>
> Ok, that's good to know even if I don't see why it shouldn't be in a  
> public one.

okay then... for the record, where do you do your "daytime work" as a web
developer, and where can we publically find ALL the source code, etc. for
the work you do there?


From william at wbonnet.net  Mon Jul 27 23:47:57 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 27 Jul 2009 23:47:57 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
Message-ID: <4A6E208D.7000605@wbonnet.net>

Hi
> William: What is the status of the mock-up?  Are there stumbling
> blocks you're facing other than time constraints?
So far mostly time problem. It has moved back to top priority since now 
i'm stuck with FF3 for a few days / week ?

The current status of the mockup is the following :

A mockup based on wordpress has been created.  After talking with 
Sebastian, i have set up a first structure of menu. The are divided in 4 
main sections which i would like to be simple and clear. Gathering the 
mainly used actions

Get It
    This menu is about how to get OpenCSW, what is OpenCSW and 
informations about release content (packages etc)
    It includes informations for the first time user, trying to download 
and install opencsw for the first time or on a new box

Use It
    This menu is about "every day" use. How to update packages, 
configure specific things, etc.

Spread It
    This menu is about advance use (rebuilding packagew with gar, 
creating a new package, and contributing to the project

Support
    This menu is about how to get support (bug tracking, wiki, faq, etc.)

The menus are a proposal for web site structure. Of course a main page 
with direct acces link will be here (i don't like the front page i set 
up which is more like a magazine still).

A temporary menu that will disappear is present under the name "Old 
menus". This is group old pages, which a copy paste of the current web 
site (it does not mean i want to delete this pages, but just my "working 
menu").

Any review comment on this menu structure will be welcomed. I will apply 
changes to the structure. Tomorrow and in the next day i'll rewrite some 
css part and main page. contribution to text content will be highly 
appreciated.

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 trygvis at opencsw.org  Mon Jul 27 23:52:48 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 27 Jul 2009 23:52:48 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090727212117.GG90811@bolthole.com>
References: <20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>	<20090725203035.GA24170@bolthole.com>	<4A6C2341.8060600@opencsw.org>	<20090726154345.GB28096@bolthole.com>	<4A6CCFCB.2040204@opencsw.org>
	<20090727212117.GG90811@bolthole.com>
Message-ID: <4A6E21B0.90606@opencsw.org>

Philip Brown wrote:
> On Sun, Jul 26, 2009 at 11:51:07PM +0200, Trygve Laugst?l wrote:
>> Philip Brown wrote:
>>> On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst?l wrote:
>>>> Can you explain what the security reason for not putting it an SCM 
>>>> is?  
>>> "a" source code control mechanism, I dont have much problem with.
>>> putting it in a PUBLIC one, is where my concerns lay.
>> Ok, that's good to know even if I don't see why it shouldn't be in a  
>> public one.
> 
> okay then... for the record, where do you do your "daytime work" as a web
> developer, and where can we publically find ALL the source code, etc. for
> the work you do there?

What's the relevance of that question?

You still skipped my other question on "security".

--
Trygve

From phil at bolthole.com  Mon Jul 27 23:54:42 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 27 Jul 2009 14:54:42 -0700
Subject: [csw-maintainers] berkeleydb packages - changes
Message-ID: <20090727215442.GM90811@bolthole.com>

btw, if anyone is not subscribed to either the users list or announce list
:-}

  ------------------------------------------------------------


This is an important announcement to opencsw users, who use our berkeleydb
packages.

Please be advised that we are making a MAJOR CHANGE to the way we handle
berkeleydb packages.

in the past, we have had many split up versions of berkeleydb4.
specifically, 4.2, 4.3, and 4.4, all separately.

>From here on in, we plan to to consolidate bdb 4.x use, into a single
bdb4.x library. 
Upgrades of our packages, should pull in these changes.

There is also a minor change in the bdb 3.x package, that does not
affect the shared libraries there at all.

In theory, this should be a transparent operation for everyone, but we
wished to advise people of this issue, in case you wished to do additional
testing, before rolling out the updates more widely in your network.

Thank you for your attention, and we hope this makes for a nice improvement
to our "customers" :-)

From phil at bolthole.com  Tue Jul 28 00:04:10 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 27 Jul 2009 15:04:10 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6E21B0.90606@opencsw.org>
References: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
	<4A6C2341.8060600@opencsw.org>
	<20090726154345.GB28096@bolthole.com>
	<4A6CCFCB.2040204@opencsw.org>
	<20090727212117.GG90811@bolthole.com> <4A6E21B0.90606@opencsw.org>
Message-ID: <20090727220410.GN90811@bolthole.com>

On Mon, Jul 27, 2009 at 11:52:48PM +0200, Trygve Laugst??l wrote:
> Philip Brown wrote:
>> okay then... for the record, where do you do your "daytime work" as a web
>> developer, and where can we publically find ALL the source code, etc. for
>> the work you do there?
>
> What's the relevance of that question?

The relevance is, if your employer(s) see security issues, or any other
issues for that matter, from standing in the way of making all their web
source public, and you are ok with that there, then it seems disingenuous
of you to fuss about our web sources not being completely public either.

So lets move on, and focus on what we can do to improve our web source
internally, discussing *internal* source control, 
and drop the needless distraction of making it a public one.


From trygvis at opencsw.org  Tue Jul 28 00:13:04 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 28 Jul 2009 00:13:04 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090727220410.GN90811@bolthole.com>
References: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>	<20090725203035.GA24170@bolthole.com>	<4A6C2341.8060600@opencsw.org>	<20090726154345.GB28096@bolthole.com>	<4A6CCFCB.2040204@opencsw.org>	<20090727212117.GG90811@bolthole.com>
	<4A6E21B0.90606@opencsw.org> <20090727220410.GN90811@bolthole.com>
Message-ID: <4A6E2670.2070104@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 27, 2009 at 11:52:48PM +0200, Trygve Laugst?l wrote:
>> Philip Brown wrote:
>>> okay then... for the record, where do you do your "daytime work" as a web
>>> developer, and where can we publically find ALL the source code, etc. for
>>> the work you do there?
>> What's the relevance of that question?
> 
> The relevance is, if your employer(s) see security issues, or any other
> issues for that matter, from standing in the way of making all their web
> source public, and you are ok with that there, then it seems disingenuous
> of you to fuss about our web sources not being completely public either.

I find the security argument ridiculous given that this is a read only 
application reading stuff from a database. If a person (either because 
of competence or tool chain or any other reason) can't write code that's 
safe enough to be shown to the public, it shouldn't run anywhere.

There are lots of applications which need to be very secure which are 
open source. Why should our source code be different?

> So lets move on, and focus on what we can do to improve our web source
> internally, discussing *internal* source control, 
> and drop the needless distraction of making it a public one.

Looks like another point we have to agree to disagree on, so ok. It's 
dropped.

--
Trygve

From mwatters at opencsw.org  Tue Jul 28 03:24:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Mon, 27 Jul 2009 20:24:33 -0500
Subject: [csw-maintainers] Please update berkeleydb* on the buildfarm please
Message-ID: <4A6E5351.6080209@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


- --
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/

iEYEARECAAYFAkpuU1AACgkQLrhmsXMSLxdJ+gCcDhqkXsdmNr3YMp3oyt9MdmDe
5DgAoJDs+HI0LMpmRB/ny3fESFjVlUKu
=w8Wm
-----END PGP SIGNATURE-----

From dam at opencsw.org  Tue Jul 28 08:58:22 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Jul 2009 08:58:22 +0200
Subject: [csw-maintainers] Updating all buildfarm servers to current now
In-Reply-To: <4A6E5351.6080209@opencsw.org>
References: <4A6E5351.6080209@opencsw.org>
Message-ID: <F57FFFB0-74FD-480D-BE74-E26957F08D00@opencsw.org>

Hi,

due to massive package updates I will be upgrading all buildfarm servers
to current/ now.


Best regards

   -- Dago

From jeff at cjsa.com  Tue Jul 28 09:33:17 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Tue, 28 Jul 2009 07:33:17 GMT
Subject: [csw-maintainers] berkeley DB problems
Message-ID: <KnHEBH.M2E@cjsa.com>


OK, I saw the recent announcement regarding the berkeley DB changes.  The
notuce said a simple upgrade should address the issue.  I did a complete
upgrade for the latestset of released packages and now have the following
installed:

-------------------------------------------------------------------------------
berkeleydb                   4.7.25,REV=2009.07.01                          SAME
berkeleydb3                  3.3.11,REV=2009.07.22                          SAME
berkeleydb4           4.2.52,REV=2008.02.23_rev=p5         4.1.25,REV=2009.06.26
berkeleydb44                 4.4.20,REV=2009.06.26                          SAME
-------------------------------------------------------------------------------

Why the berkeleydb4 is now showing a later rev than the catalog is a mystery.
But now my berkeley 4x libraries are hosed as witness an attempt to run
perl:

-------------------------------------------------------------------------------
1-> perl
ld.so.1: perl: fatal: libdb-4.4.so: open failed: No such file or directory
Killed
-------------------------------------------------------------------------------

Here are the /opt/csw/lib/libdb* libraries now installed on my system:

-------------------------------------------------------------------------------
libdb-3.3.so
libdb-4.7.so*
libdb-4.so@
libdb.so@
libdb_cxx-4.7.so*
libdb_cxx-4.so@
libdb_cxx.so@
libdb_java-4.7.so*
libdb_java-4.7_g.so@
libdb_java-4.so@
libdb_java.so@
-------------------------------------------------------------------------------

So perl, and possibly other programs are looking for libdb-4.4.so which no
longer exists.

Suggestions to get the system back to a sane state are welcome.

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040


From jeff at cjsa.com  Tue Jul 28 09:37:32 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Tue, 28 Jul 2009 07:37:32 GMT
Subject: [csw-maintainers] berkeley DB problems: follows up
Message-ID: <KnHEIK.16x@cjsa.com>

BTW, as a temporary workaround, I created the following link:

    libdb-4.4.so -> libdb-4.7.so

This seems to be addressing the problem for the moment.

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040


From dam at opencsw.org  Tue Jul 28 10:16:54 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Jul 2009 10:16:54 +0200
Subject: [csw-maintainers] Current issue with the buildfarm
Message-ID: <6BC9FC44-B109-4517-BD08-10393C5E83BB@opencsw.org>

Hi,

there is an issue with the updated release of the BerkeleyDB which  
affects
the buildfarm. I am working on it with PRIO 1 as it will also affect
user experience :-((


Best regards

   -- Dago

From dam at opencsw.org  Tue Jul 28 10:31:41 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Jul 2009 10:31:41 +0200
Subject: [csw-maintainers] berkeley DB problems
In-Reply-To: <KnHEBH.M2E@cjsa.com>
References: <KnHEBH.M2E@cjsa.com>
Message-ID: <2C97A401-0BD2-4CB9-BF14-EB8FE95039F4@opencsw.org>

Hi Jeff,

Am 28.07.2009 um 09:33 schrieb Jeffery Small:
> But now my berkeley 4x libraries are hosed as witness an attempt to  
> run
> perl:

I just fixed this. Please use the updated packages from testing/ until
the new version hits the mirrors.


Sorry for the inconvenience

   -- Dago

From bwalton at opencsw.org  Tue Jul 28 16:04:28 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 10:04:28 -0400
Subject: [csw-maintainers] php_mode for emacs
Message-ID: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>


Hi All,

I've put php_mode for emacs in testing (and asked Phil to release it
since it's such a simple package and is working for me).  Just a heads
up for anyone doing php stuff and using emacs.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/340b8bec/attachment.asc>

From bwalton at opencsw.org  Tue Jul 28 16:28:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 10:28:59 -0400
Subject: [csw-maintainers] new git book
Message-ID: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>


For anyone out there using git, you may be interested in the soon to
be released, but already available book titled 'Pro Git.'  It's written
by Scott Chacon (http://jointheconversation.org/about.html) and is
licensed the CC.

http://progit.org/

Enjoy!
-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/1718cf1f/attachment.asc>

From phil at bolthole.com  Tue Jul 28 18:18:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 09:18:06 -0700
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
Message-ID: <20090728161806.GA59031@bolthole.com>

On Tue, Jul 28, 2009 at 10:04:28AM -0400, Ben Walton wrote:
> 
> Hi All,
> 
> I've put php_mode for emacs in testing (and asked Phil to release it
> since it's such a simple package and is working for me).

Errr... please dont do this.
any package should be in EITHER testing, OR newpkgs. never both.


From pfelecan at opencsw.org  Tue Jul 28 18:22:12 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Tue, 28 Jul 2009 18:22:12 +0200
Subject: [csw-maintainers] gcc 3.4.6 available for test
Message-ID: <x6iqhcok0b.fsf@abhavana.free.fr>

This version of the GCC suite is the last one in this branch of the
project. I'm planning to release-it toward the end of this week. The
packages are available at: http://mirror.opencsw.org/testing/

If somebody finds an issue, please send me a description and a mean to
reproduce-it.

TIA
-- 
Peter

From bwalton at opencsw.org  Tue Jul 28 18:42:00 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 12:42:00 -0400
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <20090728161806.GA59031@bolthole.com>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
Message-ID: <1248799206-sup-9413@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 12:18:06 -0400 2009:
> Errr... please dont do this.
> any package should be in EITHER testing, OR newpkgs. never both.

It'll be removed from testing when it hits the mirrors[1]...avoids other
issues and makes it available ahead of the mirrors if anyone is
chomping at the bit for it (not likely in this case).

-Ben

[1] This is my normal workflow, I just don't generally release
    something the same day I roll it.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/ad0236c7/attachment.asc>

From maciej at opencsw.org  Tue Jul 28 19:16:53 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 28 Jul 2009 18:16:53 +0100
Subject: [csw-maintainers] Actual downloads of packages from testing
Message-ID: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>

When we put packages in testing, we are hoping that someone would
download and install them. Of course, we can't force anyone to test
packages, but we could see whether individual packages actually get
downloaded. We could assume that at least some of the downloads result
in testing. The reason why I'm asking is that I have a feeling that
some of the packages that spend time in testing aren't actually tested
before going into newpkgs. Once we know whether it's the case, we can
think of ways to promote the testing of packages; It would be good to
collect some data first.

Thoughts?

Maciej

From bwalton at opencsw.org  Tue Jul 28 19:22:27 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 13:22:27 -0400
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
Message-ID: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28 13:16:53 -0400 2009:

> Thoughts?

There have been requests for stats data on released packages in the
past.  That's tough to get since it would require data from every
mirror.  This should be fairly simple to get though, since testing is
centralized.  +1 for making this info available somehow...would
awstats work for this task?

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/5c29b31f/attachment.asc>

From maciej at opencsw.org  Tue Jul 28 20:08:47 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 28 Jul 2009 19:08:47 +0100
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
Message-ID: <ce97543b0907281108s7f489426maeb7a595b3747c02@mail.gmail.com>

On Tue, Jul 28, 2009 at 6:22 PM, Ben Walton<bwalton at opencsw.org> wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28 13:16:53 -0400 2009:
>
>> Thoughts?
>
> There have been requests for stats data on released packages in the
> past. ?That's tough to get since it would require data from every
> mirror. ?This should be fairly simple to get though, since testing is
> centralized. ?+1 for making this info available somehow...would
> awstats work for this task?

I think so, yes. Perhaps there could be a filter in the Apache config
which would make an additional log file with the package files (as
opposed to HTML or anything else). Awstats would read just that file.

The Apache config could be something along the lines of:

<Location /testing>
   CustomLog [...]
</Location>

Maciej

From phil at bolthole.com  Tue Jul 28 20:13:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:13:21 -0700
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248799206-sup-9413@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
Message-ID: <20090728181321.GA58718@bolthole.com>

On Tue, Jul 28, 2009 at 12:42:00PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Tue Jul 28 12:18:06 -0400 2009:
> > Errr... please dont do this.
> > any package should be in EITHER testing, OR newpkgs. never both.
> 
> It'll be removed from testing when it hits the mirrors[1]...avoids other
> issues and makes it available ahead of the mirrors if anyone is
> chomping at the bit for it (not likely in this case).

let me repeat:

"any package should be in EITHER testing, OR newpkgs. never both."

Once you have decided a package is ready for release, then it doesnt
belong in testing any more. testing, is for pre-release packages.

It should be *removed* from testing, *before* putting it in newpkgs.

testing gets bogged up enough as it is.

From phil at bolthole.com  Tue Jul 28 20:15:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:15:26 -0700
Subject: [csw-maintainers] new git book
In-Reply-To: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
Message-ID: <20090728181526.GC58718@bolthole.com>

On Tue, Jul 28, 2009 at 10:28:59AM -0400, Ben Walton wrote:
> 
> For anyone out there using git, you may be interested in the soon to
> be released, but already available book titled 'Pro Git.'  It's written
> by Scott Chacon (http://jointheconversation.org/about.html) and is
> licensed the CC.

I want to see a book named "Total Git".
With the author's picture on the front.


:-)


From bwalton at opencsw.org  Tue Jul 28 20:19:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 14:19:59 -0400
Subject: [csw-maintainers] new git book
In-Reply-To: <20090728181526.GC58718@bolthole.com>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
	<20090728181526.GC58718@bolthole.com>
Message-ID: <1248805133-sup-7458@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 14:15:26 -0400 2009:
> I want to see a book named "Total Git".
> With the author's picture on the front.

Wouldn't it be even better to be a Pro (paid) Git though? :) Linus
enjoys a good laugh about the name when asked too!

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/2c267be5/attachment.asc>

From bwalton at opencsw.org  Tue Jul 28 20:25:44 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 14:25:44 -0400
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <20090728181321.GA58718@bolthole.com>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
	<20090728181321.GA58718@bolthole.com>
Message-ID: <1248805220-sup-8974@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 14:13:21 -0400 2009:

> let me repeat:
> 
> "any package should be in EITHER testing, OR newpkgs. never both."
> 
> Once you have decided a package is ready for release, then it doesnt
> belong in testing any more. testing, is for pre-release packages.

Well yes, that's all well and good, but sometimes 'bad things happen'
and the copy from newpkgs/ gets rm'd before it is released.  Since I
wouldn't want to store a copy in my local ~/, I leave the copy in
testing/ until it gets released.  This saves me work in the event of
mishaps.

If the release process were such that things like the above didn't
happen, I'd be happy to squish the package from testing after I've
placed it in newpkgs/.

Maybe a cron job to rm files in testing older than 30 days?  (With an
exclusion list?)

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/f3aec094/attachment-0001.asc>

From phil at bolthole.com  Tue Jul 28 20:29:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:29:26 -0700
Subject: [csw-maintainers] new git book
In-Reply-To: <1248805133-sup-7458@ntdws12.chass.utoronto.ca>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
	<20090728181526.GC58718@bolthole.com>
	<1248805133-sup-7458@ntdws12.chass.utoronto.ca>
Message-ID: <20090728182926.GD58718@bolthole.com>

On Tue, Jul 28, 2009 at 02:19:59PM -0400, Ben Walton wrote:
> Wouldn't it be even better to be a Pro (paid) Git though? :) Linus
> enjoys a good laugh about the name when asked too!

Yeah, that's pretty much Linus's job.
Paid Professional Git. glad he likes it :-)

[ as opposed to me, UNpaid professional git. When do I get my 
 opensource handout, Linus?   ha]

From phil at bolthole.com  Tue Jul 28 20:31:37 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:31:37 -0700
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248805220-sup-8974@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
	<20090728181321.GA58718@bolthole.com>
	<1248805220-sup-8974@ntdws12.chass.utoronto.ca>
Message-ID: <20090728183137.GF58718@bolthole.com>

On Tue, Jul 28, 2009 at 02:25:44PM -0400, Ben Walton wrote:
> Well yes, that's all well and good, but sometimes 'bad things happen'
> and the copy from newpkgs/ gets rm'd before it is released.  Since I
> wouldn't want to store a copy in my local ~/, 

why not? at least then it's only cluttering up YOUR space :-)

$ mv testing/xyz ~

instead of

$ rm testing/xyz



From bwalton at opencsw.org  Tue Jul 28 20:52:10 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 14:52:10 -0400
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <20090728183137.GF58718@bolthole.com>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
	<20090728181321.GA58718@bolthole.com>
	<1248805220-sup-8974@ntdws12.chass.utoronto.ca>
	<20090728183137.GF58718@bolthole.com>
Message-ID: <1248806415-sup-2819@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 14:31:37 -0400 2009:

> why not? at least then it's only cluttering up YOUR space :-)

It's the same space (just different eyeballs on it), but
whatever...fine.  You don't delete packages accidentally and I'll
remove them from testing/ after requesting the release.  :)

I've got a few ideas for changes in this process that I'll document
(with Maciek) and propose in a completed form.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/7fdcd342/attachment.asc>

From mwatters at opencsw.org  Wed Jul 29 00:34:54 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Jul 2009 17:34:54 -0500
Subject: [csw-maintainers] new git book
In-Reply-To: <20090728181526.GC58718@bolthole.com>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
	<20090728181526.GC58718@bolthole.com>
Message-ID: <4A6F7D0E.2010104@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
>
> I want to see a book named "Total Git".
> With the author's picture on the front.

no, with picture of Chris Barrie Holding an 'H'

just to see how many people can get the reference.




- --

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

iEYEARECAAYFAkpvfQ4ACgkQLrhmsXMSLxcUMACg4WbLRH92mYxNmo1cGExYJggT
f9IAoOvSHIc39zveRNNNdixOUJwtZNqm
=pt5o
-----END PGP SIGNATURE-----

From mwatters at opencsw.org  Wed Jul 29 00:36:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Jul 2009 17:36:49 -0500
Subject: [csw-maintainers] pidgin 2.5.8 now in testing
Message-ID: <4A6F7D81.6080106@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This version fixes the yahoo instant messenger login protocol problem.


- --

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

iEYEARECAAYFAkpvfYEACgkQLrhmsXMSLxfttACeORnnA1rtcYJEaR5ioywBhFfK
boYAoISjCTdXw2DpVdadSQ1A+t9jBWEJ
=UK67
-----END PGP SIGNATURE-----

From trygvis at opencsw.org  Wed Jul 29 01:10:19 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Wed, 29 Jul 2009 01:10:19 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
Message-ID: <4A6F855B.9020209@opencsw.org>

Ben Walton wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28 13:16:53 -0400 2009:
> 
>> Thoughts?
> 
> There have been requests for stats data on released packages in the
> past.  That's tough to get since it would require data from every
> mirror.  This should be fairly simple to get though, since testing is
> centralized.  +1 for making this info available somehow...would
> awstats work for this task?

Yeah, that would be nice.

It might be hard to estimate the impact on a general basis, but it 
should be possible to measure the impact of individual packages between 
releases.

--
Trygve


From maciej at opencsw.org  Wed Jul 29 01:16:22 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 00:16:22 +0100
Subject: [csw-maintainers] setxkbmap investigation
Message-ID: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>

I've recently built a setxkbmap binary. There's something I can't
quite work out, perhaps somebody will be able to help me out.

Information about the package is here:
http://lists.opencsw.org/pipermail/users/2009-July/004052.html
The Makefile:
http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/setxkbmap/trunk/Makefile

I have a reference binary in /var/tmp, which reads the keymap from a
known location: /usr/X11/lib/X11/xkb/symbols. For instance, when I run
'setxkbmap pl', the file /usr/X11/lib/X11/xkb/symbols/pl gets read.
When I run my setxkbmap, they 'pl' keymap must get loaded from
someplace else, because it's not my (patched) keymap. But I can't
figure out, where it is being read from.

I don't see much choice there:

$ glocate -r 'X11.*/pl$'
/usr/openwin/lib/X11/xkb/keymap/sgi/pl
/usr/openwin/lib/X11/xkb/keymap/sun/pl
/usr/openwin/lib/X11/xkb/symbols/pl
/usr/openwin/lib/X11/xkb/symbols/sun/pl
/usr/X11/lib/X11/xkb/keymap/sgi/pl
/usr/X11/lib/X11/xkb/keymap/sun/pl
/usr/X11/lib/X11/xkb/symbols/pc/pl
/usr/X11/lib/X11/xkb/symbols/pl
/usr/X11/lib/X11/xkb/symbols/sun/pl

I trussed the setxkbmap process (with -f to trace also the child
processes), but it doesn't open any of the above files. Do you have a
good idea where it can read from or how to track that down?

Maciej

From bwalton at opencsw.org  Wed Jul 29 01:30:55 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 19:30:55 -0400
Subject: [csw-maintainers] package hook document
Message-ID: <1248823765-sup-6856@ntdws12.chass.utoronto.ca>


Would anyone interested in the package hook implementation please
have a look at: http://wiki.opencsw.org/package-hooks

I'd like to keep this moving along...

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 dam at opencsw.org  Wed Jul 29 08:29:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 08:29:08 +0200
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248799206-sup-9413@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
Message-ID: <35350311-D392-4B8F-8C8E-E95CD312DC4C@opencsw.org>

Hi Ben,

Am 28.07.2009 um 18:42 schrieb Ben Walton:
> It'll be removed from testing when it hits the mirrors[1]...avoids  
> other
> issues and makes it available ahead of the mirrors if anyone is
> chomping at the bit for it (not likely in this case).

That is ok, my previous mailing about cleaning up testing was about
stuff that was forgotten in testing/. The few extra days for full
mirror sync are no real problem here. Apart from that I did receive
some emails in the past where the package was after I removed it
from testing but before mirrors were synced.


Best regards

   -- Dago

From dam at opencsw.org  Wed Jul 29 08:46:30 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 08:46:30 +0200
Subject: [csw-maintainers] setxkbmap investigation
In-Reply-To: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>
References: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>
Message-ID: <13B9E170-BB9C-405F-A567-919A55AB10D4@opencsw.org>

Hi Maciej,

Am 29.07.2009 um 01:16 schrieb Maciej (Matchek) Blizinski:
> I have a reference binary in /var/tmp, which reads the keymap from a
> known location: /usr/X11/lib/X11/xkb/symbols. For instance, when I run
> 'setxkbmap pl', the file /usr/X11/lib/X11/xkb/symbols/pl gets read.
> When I run my setxkbmap, they 'pl' keymap must get loaded from
> someplace else, because it's not my (patched) keymap. But I can't
> figure out, where it is being read from.

You can try "opensnoop" from the dtracetoolkit (pkg-get -i  
dtracetoolkit):

root at ncsw [global]:/root > opensnoop
   UID    PID COMM          FD PATH
     0  16783 mibiisa        2 /dev/kstat
     0  16783 mibiisa       -1 /dev/e
     0  16783 mibiisa        2 /dev/kstat
     0  18355 sysstatd     551 /proc/5045/psinfo
     0  18355 sysstatd     552 /proc/5045/status
     0  18355 sysstatd     569 /proc/5033/psinfo
     0  18355 sysstatd     570 /proc/5033/status
     0  18355 sysstatd     599 /proc/5031/psinfo
     0  18355 sysstatd     600 /proc/5031/status
     0  18355 sysstatd     899 /proc/5046/psinfo
     0  18355 sysstatd     900 /proc/5046/status
10000   5047 setxkbmap     -1 /var/ld/ld.config
10000   5047 setxkbmap      3 /lib/libc.so.1
10000   5047 setxkbmap      3 /usr/lib/libxkbfile.so.4
10000   5047 setxkbmap      3 /platform/SUNW,SPARC-Enterprise-T5220/ 
lib/libc_psr.so.1
10000   5047 setxkbmap      3 /usr/lib/libX11.so.4
10000   5047 setxkbmap      3 /lib/libsocket.so.1
10000   5047 setxkbmap      3 /lib/libnsl.so.1
10000   5047 setxkbmap      4 /etc/netconfig
10000   5047 setxkbmap      4 /dev/udp
10000   5047 setxkbmap      4 /var/run/name_service_door
10000   5047 setxkbmap      5 /etc/default/nss
10000   5047 setxkbmap      5 /dev/udp6
10000   5047 setxkbmap      5 /home/dam/.Xauthority
10000   5047 setxkbmap     -1 /usr/X11R6/lib/X11/xkb/rules/xorg-C.lst
10000   5047 setxkbmap      5 /usr/X11R6/lib/X11/xkb/rules/xorg.lst
10000   5047 setxkbmap     -1 /usr/X11R6/lib/X11/xkb/rules/xorg-C
10000   5047 setxkbmap      5 /usr/X11R6/lib/X11/xkb/rules/xorg
10000   5047 setxkbmap     -1 ./rules/xorg-C.lst
10000   5047 setxkbmap     -1 ./rules/xorg.lst
     1  17464 nfsmapid       8 /etc/resolv.conf
     0  16448 nscd           4 /etc/inet/ipnodes
^C

However, setxkbmap doesn't work in my environment:

dam at build10st [build10st]:/home/dam > setxkbmap pl
Couldn't interpret _XKB_RULES_NAMES property
Use defaults: rules - 'xorg' model - 'pc101' layout - 'us'
Error loading new keyboard description
zsh: 5047 exit 251   setxkbmap pl



Best regards

   -- Dago

From maciej at opencsw.org  Wed Jul 29 12:16:18 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 11:16:18 +0100
Subject: [csw-maintainers] setxkbmap investigation
In-Reply-To: <13B9E170-BB9C-405F-A567-919A55AB10D4@opencsw.org>
References: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>
	<13B9E170-BB9C-405F-A567-919A55AB10D4@opencsw.org>
Message-ID: <ce97543b0907290316kab18346gef8930f582e91d30@mail.gmail.com>

On Wed, Jul 29, 2009 at 7:46 AM, Dagobert Michelsen<dam at opencsw.org> wrote:
> Hi Maciej,
>
> Am 29.07.2009 um 01:16 schrieb Maciej (Matchek) Blizinski:
>>
>> I have a reference binary in /var/tmp, which reads the keymap from a
>> known location: /usr/X11/lib/X11/xkb/symbols. For instance, when I run
>> 'setxkbmap pl', the file /usr/X11/lib/X11/xkb/symbols/pl gets read.
>> When I run my setxkbmap, they 'pl' keymap must get loaded from
>> someplace else, because it's not my (patched) keymap. But I can't
>> figure out, where it is being read from.
>
> You can try "opensnoop" from the dtracetoolkit (pkg-get -i dtracetoolkit):
>
> root at ncsw [global]:/root > opensnoop
> ?UID ? ?PID COMM ? ? ? ? ?FD PATH
> ? ?0 ?16783 mibiisa ? ? ? ?2 /dev/kstat

I installed the dtracetoolkit, thanks! opensnoop is really useful.

http://dpaste.com/72855/ -- looks like it's the .../pc/pl file that
gets loaded by xkbcomp.

> However, setxkbmap doesn't work in my environment:
>
> dam at build10st [build10st]:/home/dam > setxkbmap pl
> Couldn't interpret _XKB_RULES_NAMES property
> Use defaults: rules - 'xorg' model - 'pc101' layout - 'us'
> Error loading new keyboard description
> zsh: 5047 exit 251 ? setxkbmap pl

Perhaps X11 packages aren't installed.

Thanks for the help. I'll poke it some more to if I can get it to load
the correct keyboard mapping.

Maciej

From maciej at opencsw.org  Wed Jul 29 12:44:49 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 11:44:49 +0100
Subject: [csw-maintainers] A copyright notice in the template
Message-ID: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>

I've committed a change to the template: I added a short copyright
notice (GPLv2) and set the Id keyword so that the Makefile would
contain information about the current file revision:

https://sourceforge.net/apps/trac/gar/changeset/5704

Since it's the template, the impact would be only on new packages; a
package author could always remove the notice if they wanted.

What do you think of this change? Is it okay?

Maciej

From dam at opencsw.org  Wed Jul 29 13:22:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 13:22:08 +0200
Subject: [csw-maintainers] A copyright notice in the template
In-Reply-To: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>
References: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>
Message-ID: <2484D1C9-2EC1-4F69-A9A1-9F0BD33565D2@opencsw.org>

Hi Maciej,

Am 29.07.2009 um 12:44 schrieb Maciej (Matchek) Blizinski:
> I've committed a change to the template: I added a short copyright
> notice (GPLv2) and set the Id keyword so that the Makefile would
> contain information about the current file revision:
>
> https://sourceforge.net/apps/trac/gar/changeset/5704
>
> Since it's the template, the impact would be only on new packages; a
> package author could always remove the notice if they wanted.
>
> What do you think of this change? Is it okay?

I think it is a good idea for a template. There are also programmatical
templates per-category. Do you mind updating these in
   pkg/Makefile
   pkg/cpan/Makefile
   pkg/xfce/Makefile
   pkg/x11/Makefile
also?


Best regards

   -- Dago

From maciej at opencsw.org  Wed Jul 29 14:52:25 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 13:52:25 +0100
Subject: [csw-maintainers] A copyright notice in the template
In-Reply-To: <2484D1C9-2EC1-4F69-A9A1-9F0BD33565D2@opencsw.org>
References: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>
	<2484D1C9-2EC1-4F69-A9A1-9F0BD33565D2@opencsw.org>
Message-ID: <ce97543b0907290552j44ee09c0m995d9afc0aef50a0@mail.gmail.com>

On Wed, Jul 29, 2009 at 12:22 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> I think it is a good idea for a template. There are also programmatical
> templates per-category. Do you mind updating these in
> ?pkg/Makefile
> ?pkg/cpan/Makefile
> ?pkg/xfce/Makefile
> ?pkg/x11/Makefile
> also?

Done, headers added.

Maciej

From maciej at opencsw.org  Wed Jul 29 15:07:09 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 14:07:09 +0100
Subject: [csw-maintainers] HelpLinker
Message-ID: <ce97543b0907290607n766d18a7qc46f15b0d66bf698@mail.gmail.com>

Hello James,

What is the HelpLinker you run off cron on the buildfarm?

Maciej

From maciej at opencsw.org  Wed Jul 29 15:19:03 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 14:19:03 +0100
Subject: [csw-maintainers] HelpLinker
In-Reply-To: <20090729.13132500.3027483693@gyor.oxdrove.co.uk>
References: <ce97543b0907290607n766d18a7qc46f15b0d66bf698@mail.gmail.com>
	<20090729.13132500.3027483693@gyor.oxdrove.co.uk>
Message-ID: <ce97543b0907290619t5b507706nb0591adb22b8ccb7@mail.gmail.com>

On Wed, Jul 29, 2009 at 2:13 PM, James Lee<james at opencsw.org> wrote:
> On 29/07/09, 14:07:09, Maciej "(Matchek)" Blizinski <maciej at opencsw.org>
> wrote regarding HelpLinker:
>
>> What is the HelpLinker you run off cron on the buildfarm?
>
> It is an OpenOffice.org build utility that makes the help pages.
>
> It's run from "at" not "cron" so the build continues after my
> terminal disconnects (damn these ISPs when can't they keep a link
> up longer than 24 hours, actually it allows me to sys-suspend
> too and not stop a build, anything that stops the terminal).

I see. nohup or screen could also work. I thought it was cron because
it was the cron process that was the parent of the HelpLinker
processes.

Thanks for the answer.

Maciej

From phil at bolthole.com  Wed Jul 29 18:23:04 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Jul 2009 09:23:04 -0700
Subject: [csw-maintainers] comment on open source security
Message-ID: <20090729162304.GA45823@bolthole.com>

btw;
https://www.isc.org/node/474

That's the second DNS exploit, in the last 12 months.

[sarcasm on]
Wow, they must have some really non-competant programmers working on that.
They shouldnt be allowed to work on that any more.
[/sarcasm]

From blizinski at google.com  Mon Jul 20 14:29:21 2009
From: blizinski at google.com (Maciej (Matchek) Blizinski)
Date: Mon, 20 Jul 2009 13:29:21 +0100
Subject: [csw-maintainers] Updating cups to 1.3.11
Message-ID: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>

Hi Markus,

Do you mind building cups-1.3.11? I've submitted a version bump and a
small bugfix. Can you please build the package?

Here's the mantis issue:
http://www.opencsw.org/bugtrack/view.php?id=3768

Reason for the request: version 1.3.11 contains a bugfix for the
scheduler code. In 1.3.9 cups-polld gets stuck after some time.

To other maintainers: I've filed a bug with mantis, Markus is marked
as active maintainer, but there is a number of unresolved issues in
mantis. When did you last hear from Markus?

Maciej

From trygvis at inamo.no  Thu Jul 23 18:09:52 2009
From: trygvis at inamo.no (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 23 Jul 2009 18:09:52 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723160656.GH53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<4A6886DF.30507@opencsw.org>
	<20090723160656.GH53578@bolthole.com>
Message-ID: <4A688B50.1070502@inamo.no>

Philip Brown wrote:
> On Thu, Jul 23, 2009 at 05:50:55PM +0200, Trygve Laugst?l wrote:
>> Ben Walton wrote:
>>> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>>>> source code of www.opencsw.org checked into subversion? If not, what
>>>> do you think of submitting the code there? It would make it easier to
>>>> suggest a change, or submit a patch, and generally make things more
>>>> open.
>>> +1 for making the website code open and public.
>> +1
>>
>> Would be nice to be able to turn the package database into a separate  
>> application, I have clients that might see use something like that.
> 
> Umm... I doubt your clients would really want to use our code.
> Euuu, ick ;-)
> 
> besides, there isnt really much "code" for that. What code would they
> actually use? the
>  "taking apart a package, to analyze and register in a database" bit?

That part would definitely be useful.

> If you're serious about that, I'll consider making my ugly mess something
> that is more suitable for publication.

They would like to see the applications that they have available, but 
also the versions available so some extensions would be needed.

I realize it is all quite easy to write, but it is the kind of code that 
take a while to get to work correctly so a starting point would be useful.

--
Trygve

From blizinski at google.com  Thu Jul 23 23:06:39 2009
From: blizinski at google.com (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 22:06:39 +0100
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <4A68CBFD.3000408@wbonnet.net>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
	<4A68CBFD.3000408@wbonnet.net>
Message-ID: <ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>

On Thu, Jul 23, 2009 at 9:45 PM, William Bonnet<william at wbonnet.net> wrote:
> There is also the i386 package :)
>
> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-sparc.pkg.gz
> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-i386.pkg.gz

Your firefox packages should be listed in the catalog file[1], right?
I see my django and pyantlrrt packages, but not firefox. Your package
have been placed around the same time as mine, but mine are there and
yours are not. I smell there's something wrong there.

[1] http://mirror.opencsw.org/opencsw/testing/i386/5.10/catalog

Maciej

From blizinski at google.com  Fri Jul 24 10:28:12 2009
From: blizinski at google.com (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 09:28:12 +0100
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
	<4A68CBFD.3000408@wbonnet.net>
	<ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>
Message-ID: <ce97543b0907240128ic5b1c73q230ecff15d80d8b9@mail.gmail.com>

On Thu, Jul 23, 2009 at 10:06 PM, Maciej (Matchek)
Blizinski<blizinski at google.com> wrote:
> On Thu, Jul 23, 2009 at 9:45 PM, William Bonnet<william at wbonnet.net> wrote:
>> There is also the i386 package :)
>>
>> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-sparc.pkg.gz
>> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-i386.pkg.gz
>
> Your firefox packages should be listed in the catalog file[1], right?

I pulled it manually from the given URL and used pkgadd. I couldn't
install it with pkg-get or pkgutil.

After I got it installed - it works! I'll keep using it and will come
back with any comments. Thanks, William!

Maciej

From shuttlebox at gmail.com  Wed Jul 29 16:57:02 2009
From: shuttlebox at gmail.com (Peter Bonivart)
Date: Wed, 29 Jul 2009 16:57:02 +0200
Subject: [csw-maintainers] Fwd: [csw-users] stable release
References: <20090728.10023200.2665933905@gyor.oxdrove.co.uk>
Message-ID: <1ED128FF-E617-4F2E-99A3-D8B829ED214A@gmail.com>

Is there a new stable release in the works? It will be discussed at  
the august meeting, would be good to have your take on it first.

Sent from my iPhone

Vidarebefordrat brev:

> Fr?n: James Lee <james at opencsw.org>
> Datum: 28 juli 2009 10.02.32 CEST
> Till: questions and discussions <users at lists.opencsw.org>
> ?mne: Re: [csw-users] stable release
> Svara till: questions and discussions <users at lists.opencsw.org>
>

> On 28/07/09, 10:34:35, Klaus Heinz <khfp.blastwave0704 at gmx.de> wrote
> regarding [csw-users] stable release:
>
>> maybe I missed something but it looks to me as if the stable tree has
>> hardly changed since August 2008 (I checked the "stable" directory at
>> ftp.uni-erlangen.de and ftp.df.lth.se).
>
>> Is this really the latest stable version of OpenCSW packages?
>
> In fact the latest is 2008-03.  The August dates are only an OpenCSW
> renaming.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090729/119363dd/attachment.html>

From shuttlebox at gmail.com  Wed Jul 29 17:01:55 2009
From: shuttlebox at gmail.com (Peter Bonivart)
Date: Wed, 29 Jul 2009 17:01:55 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
Message-ID: <9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>

To have download stats from just one mirror would still be a very good  
indication. To get combined stats from all our mirrors will never  
happen.

Sent from my iPhone

28 jul 2009 kl. 19.22 skrev Ben Walton <bwalton at opencsw.org>:

> Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28  
> 13:16:53 -0400 2009:
>
>> Thoughts?
>
> There have been requests for stats data on released packages in the
> past.  That's tough to get since it would require data from every
> mirror.  This should be fairly simple to get though, since testing is
> centralized.  +1 for making this info available somehow...would
> awstats work for this task?

From dam at opencsw.org  Wed Jul 29 19:13:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 19:13:01 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
Message-ID: <31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>

Hi Peter,

Am 29.07.2009 um 17:01 schrieb Peter Bonivart:
> To have download stats from just one mirror would still be a very  
> good indication. To get combined stats from all our mirrors will  
> never happen.

Ok, now everybody calm down. I am working on updating the awstats
package and providing download stats for

   http://mirror.opencsw.org/opencsw/current/
   http://mirror.opencsw.org/opencsw/stable/
   http://mirror.opencsw.org/opencsw/testing/

Michael, can we get stats from your mirror also?


Best regards

   -- Dago

From dam at opencsw.org  Wed Jul 29 20:12:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 20:12:15 +0200
Subject: [csw-maintainers] Statistics on ibiblio
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>
Message-ID: <F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>

Hi there,

there are also statistics available for ibiblio:

1. Log in to <https://stats.ibiblio.org>
   User: guest Password: ibiblio
2. Click on www.ibiblio.org.public
3. Click on Pages & Files
4. Click on Directory Drilldown
5. Click on Directory by Files Drilldown
6. Click on /pub
7. Click on /packages
8. Click on /solaris

...enjoy :-)

(Looks like nobody is using Solaris 8/9 and Blastwave any more)


Best regards

  -- Dago


From phil at bolthole.com  Wed Jul 29 20:27:03 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Jul 2009 11:27:03 -0700
Subject: [csw-maintainers] Statistics on ibiblio
In-Reply-To: <F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>
	<F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
Message-ID: <20090729182703.GG45823@bolthole.com>

On Wed, Jul 29, 2009 at 08:12:15PM +0200, Dagobert Michelsen wrote:
>
> (Looks like nobody is using Solaris 8/9 and Blastwave any more)

wow. interesting.

more precisely, in the last week, there were no downloads in
current/*/5.8 or 5.9

there were a few in
stable/*/5.8 and 5.9


that's a big change.



From maciej at opencsw.org  Thu Jul 30 00:24:35 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 23:24:35 +0100
Subject: [csw-maintainers] unixODBC-2.2.14 in testing
Message-ID: <ce97543b0907291524i365c767ag395076140919221a@mail.gmail.com>

The unixODBC package, version 2.2.14 is now in testing.

http://mirror.opencsw.org/testing/unixodbc-2.2.14,REV=2009.07.29-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/unixodbc-2.2.14,REV=2009.07.29-SunOS5.8-i386-CSW.pkg.gz

It's the first version built with GAR, a remake from scratch. I
haven't got access to the original package. This package doesn't
include GUI (QT) support. It keeps the configuration files in
/etc/opt/csw, as opposed to the earlier /opt/csw/etc. Please feel free
to respond with any feedback on the package.

Maciej

From mwatters at opencsw.org  Thu Jul 30 01:15:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 29 Jul 2009 18:15:50 -0500
Subject: [csw-maintainers] dbus in testing -- smf error
Message-ID: <4A70D826.8070905@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I found an error while testing the version of CSWdbus in testing
  dbus 1.2.12,REV=2009.07.15

while trying to start on solaris 10 using SMF it passes the --config= argument
to the command line. this does not allow the smf service to start.  if I edit
the start/stop SMF discription by hand and s/--config=.*$// ( to make it look
like the version in "current") everything works as expected.

- --

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

iEYEARECAAYFAkpw2CYACgkQLrhmsXMSLxfsUACgxULKBWkB+WeqdgiHMSP4LFWu
wv8An1Gei/KV0zVE/M7cV8dj+qXbqCrq
=zj12
-----END PGP SIGNATURE-----

From pfelecan at opencsw.org  Thu Jul 30 09:27:34 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 30 Jul 2009 09:27:34 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com> (Peter
	Bonivart's message of "Wed\, 29 Jul 2009 17\:01\:55 +0200")
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
Message-ID: <x6tz0uljfd.fsf@abhavana.free.fr>

Peter Bonivart <shuttlebox at gmail.com> writes:

> Sent from my iPhone

a 3, 3G or 3GS?

sent from my hair dryer
-- 
Peter

From michael at opencsw.org  Thu Jul 30 09:55:40 2009
From: michael at opencsw.org (Michael Gernoth)
Date: Thu, 30 Jul 2009 09:55:40 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
	<31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
Message-ID: <20090730075540.GA29118@zerfleddert.de>

Hi Dago,

On Wed, Jul 29, 2009 at 07:13:01PM +0200, Dagobert Michelsen wrote:
> Michael, can we get stats from your mirror also?

I'll try it this weekend. We have logs back to the beginning, so this
should be no problem.

Regards,
  Michael

From skayser at opencsw.org  Thu Jul 30 09:57:38 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 30 Jul 2009 09:57:38 +0200 (CEST)
Subject: [csw-maintainers] dbus in testing -- smf error
In-Reply-To: <4A70D826.8070905@opencsw.org>
References: <4A70D826.8070905@opencsw.org>
Message-ID: <49287.217.227.30.90.1248940658.squirrel@ssl.skayser.de>

Mike Watters wrote:
> I found an error while testing the version of CSWdbus in testing
>   dbus 1.2.12,REV=2009.07.15
>
> while trying to start on solaris 10 using SMF it passes the --config=
> argument
> to the command line. this does not allow the smf service to start.  if I
> edit
> the start/stop SMF discription by hand and s/--config=.*$// ( to make it
> look
> like the version in "current") everything works as expected.

Same here.

  http://lists.opencsw.org/pipermail/maintainers/2009-July/003337.html

Sebastian


From maciej at opencsw.org  Thu Jul 30 10:17:51 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 09:17:51 +0100
Subject: [csw-maintainers] Buildbot and twisted available in testing
Message-ID: <ce97543b0907300117y21e4edbk5856ec73c95abca1@mail.gmail.com>

Buildbot and its dependency, twisted, are available in testing.

>From the homepage: BuildBot is a system to automate the compile/test
cycle required by most software projects to validate code changes. By
automatically rebuilding and testing the tree each time something has
changed, build problems are pinpointed quickly, before other
developers are inconvenienced by the failure.

Direct links to packages:
http://mirror.opencsw.org/testing/buildbot-0.7.11p1,REV=2009.07.29-SunOS5.8-all-CSW.pkg.gz
http://mirror.opencsw.org/testing/twisted-8.2.0,REV=2009.07.29-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/twisted-8.2.0,REV=2009.07.29-SunOS5.8-sparc-CSW.pkg.gz

Buildbot also depends on Zope.

Maciej

From ihsan at opencsw.org  Thu Jul 30 12:12:38 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 30 Jul 2009 12:12:38 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <x6tz0uljfd.fsf@abhavana.free.fr>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
	<x6tz0uljfd.fsf@abhavana.free.fr>
Message-ID: <4A717216.70301@opencsw.org>

Peter FELECAN wrote:

>> Sent from my iPhone
> 
> a 3, 3G or 3GS?
> 
> sent from my hair dryer

An article about that was in the Swiss press, when the first iPhone was
released. Unfortunately only in German.
http://ihsan.dogan.ch/files/iPhone_vs_Blackberry/iPhone_vs_Blackberry.pdf




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From bonivart at opencsw.org  Thu Jul 30 12:25:07 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 12:25:07 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com> 
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com> 
	<31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
Message-ID: <625385e30907300325i63deca6dh143cfa03e6d64364@mail.gmail.com>

On Wed, Jul 29, 2009 at 7:13 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> Hi Peter,
>
> Am 29.07.2009 um 17:01 schrieb Peter Bonivart:
>>
>> To have download stats from just one mirror would still be a very good
>> indication. To get combined stats from all our mirrors will never happen.
>
> Ok, now everybody calm down.

I'm calm. :-) Just meant that we shouldn't deny ourselves any stats
because the perfect combined stats are so complicated to get.

Now it seems to be happening and that's great.

-- 
/peter

From bonivart at opencsw.org  Thu Jul 30 12:26:52 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 12:26:52 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <x6tz0uljfd.fsf@abhavana.free.fr>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com> 
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com> 
	<x6tz0uljfd.fsf@abhavana.free.fr>
Message-ID: <625385e30907300326h6ca5d795g4e2f7b0cc6638773@mail.gmail.com>

On Thu, Jul 30, 2009 at 9:27 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> Peter Bonivart <shuttlebox at gmail.com> writes:
>
>> Sent from my iPhone
>
> a 3, 3G or 3GS?

The 3GS isn't available in Sweden until tomorrow! :-) Didn't notice
that it added a signature, must find that setting, haha.

> sent from my hair dryer

:-)

-- 
/peter

From bonivart at opencsw.org  Thu Jul 30 12:28:59 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 12:28:59 +0200
Subject: [csw-maintainers] Statistics on ibiblio
In-Reply-To: <F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de> 
	<F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
Message-ID: <625385e30907300328u4061083cud8e263a48a9eb4ef@mail.gmail.com>

On Wed, Jul 29, 2009 at 8:12 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> (Looks like nobody is using Solaris 8/9 and Blastwave any more)

Thanks for that link. I don't think BW lists ibiblio as a mirror any
more so their users may have moved to other mirrors.

-- 
/peter

From william at wbonnet.net  Thu Jul 30 12:29:06 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 30 Jul 2009 12:29:06 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <625385e30907300325i63deca6dh143cfa03e6d64364@mail.gmail.com>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
	<31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
	<625385e30907300325i63deca6dh143cfa03e6d64364@mail.gmail.com>
Message-ID: <4A7175F2.1090005@wbonnet.net>

Peter Bonivart a ?crit :
> On Wed, Jul 29, 2009 at 7:13 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
>   
>> Hi Peter,
>>
>> Am 29.07.2009 um 17:01 schrieb Peter Bonivart:
>>     
>>> To have download stats from just one mirror would still be a very good
>>> indication. To get combined stats from all our mirrors will never happen.
>>>       
>> Ok, now everybody calm down.
>>     
>
> I'm calm. :-) Just meant that we shouldn't deny ourselves any stats
> because the perfect combined stats are so complicated to get.
>
> Now it seems to be happening and that's great.
>
>   
What about creating something like debian did with popcon ? That would 
be more accurate than getting stats from only a part of the mirrors.

cheers
W.

From ihsan at dogan.ch  Thu Jul 30 12:48:00 2009
From: ihsan at dogan.ch (Ihsan Dogan)
Date: Thu, 30 Jul 2009 12:48:00 +0200
Subject: [csw-maintainers] comment on open source security
In-Reply-To: <20090729162304.GA45823@bolthole.com>
References: <20090729162304.GA45823@bolthole.com>
Message-ID: <4A717A60.3070305@dogan.ch>

Philip Brown wrote:
> btw;
> https://www.isc.org/node/474
> 
> That's the second DNS exploit, in the last 12 months.
> 
> [sarcasm on]
> Wow, they must have some really non-competant programmers working on that.
> They shouldnt be allowed to work on that any more.
> [/sarcasm]

The fact that Bind 10 is going to be a redesign again, shows that Bind 9
has broken design.

With NSD there is now a good alternative to Bind. ns1.opencsw.org runs
with it.
http://www.opencsw.org/packages/nsd



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From maciej at opencsw.org  Thu Jul 30 15:31:42 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 14:31:42 +0100
Subject: [csw-maintainers] cswpycompile docs
Message-ID: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>

cswclassutils docs[1] say this about cswpycompile:

------8<------
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 }'
------8<------

I think it should not say 'pycompile', but 'cswpycompile':

------8<------
This class takes care of creating/removing python's normal and
optimized bytecode files. Set all *.py files' class = cswpycompile to
activate. Example:

# use prototype filters to set the class
PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "cswpycompile" } { print }'
------8<------

Is that right?

Maciej

[1] http://wiki.opencsw.org/cswclassutils-package

From bwalton at opencsw.org  Thu Jul 30 15:37:10 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 30 Jul 2009 09:37:10 -0400
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
Message-ID: <1248961009-sup-8082@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 30 09:31:42 -0400 2009:

> I think it should not say 'pycompile', but 'cswpycompile':

> Is that right?

I believe that is correct.
-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090730/4815e03c/attachment.asc>

From bonivart at opencsw.org  Thu Jul 30 16:10:13 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 16:10:13 +0200
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
Message-ID: <625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>

On Thu, Jul 30, 2009 at 3:31 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> cswclassutils docs[1] say this about cswpycompile:
>
> I think it should not say 'pycompile', but 'cswpycompile':
>
> Is that right?

Yes, I fixed it this time but please go ahead and fix it yourself next
time. It's a wiki after all. :-)

If you don't have an account on the wiki yet, just apply for one and I
will grant you access.

-- 
/peter

From bonivart at opencsw.org  Thu Jul 30 16:14:35 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 16:14:35 +0200
Subject: [csw-maintainers] /testing BIND 9.6.1-P1 (fix for critical security
	hole)
Message-ID: <625385e30907300714g1b046218m556faf0acc1cebde@mail.gmail.com>

Now in testing...

-- 
/peter

From dam at opencsw.org  Thu Jul 30 16:50:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 30 Jul 2009 16:50:40 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
Message-ID: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>

Hi,

I am currently working on a fix for the gtk2-bug #3775:
   "Package should create /opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders"
   http://www.opencsw.org/mantis/view.php?id=3775

What I have done now is setting it up on postinstall like pango.
However, on Solaris 8 it will try to query 64 bit too, which fails.
If you now share this to Solaris 10 you will end up without pixbuf
loaders. Would you consider it to be ok to skip this step on
Solaris 8?

Or could the pixbuffers be queried during package creation statically?
That would also eliminate the ugly (and slow) postinstall.


Best regards

   -- Dago

From maciej at opencsw.org  Thu Jul 30 17:03:27 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 16:03:27 +0100
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
	<625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>
Message-ID: <ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>

On Thu, Jul 30, 2009 at 3:10 PM, Peter Bonivart<bonivart at opencsw.org> wrote:
> On Thu, Jul 30, 2009 at 3:31 PM, Maciej (Matchek)
> Blizinski<maciej at opencsw.org> wrote:
>> cswclassutils docs[1] say this about cswpycompile:
>>
>> I think it should not say 'pycompile', but 'cswpycompile':
>>
>> Is that right?
>
> Yes, I fixed it this time but please go ahead and fix it yourself next
> time. It's a wiki after all. :-)

Two reasons - I wanted to ask first, and I don't have access.
Otherwise I agree and as a rule I fix stuff myself. (And then people
disagree with me. ;-) )

> If you don't have an account on the wiki yet, just apply for one and I
> will grant you access.

I was about to ask you about it the day before yesterday. I've just
set up an account in wiki.opencsw.org. To keep it simple and
consistent, I used the name 'automatthias'. ;-)

Maciej

From pfelecan at opencsw.org  Thu Jul 30 17:42:32 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 30 Jul 2009 17:42:32 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
In-Reply-To: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org> (Dagobert
	Michelsen's message of "Thu\, 30 Jul 2009 16\:50\:40 +0200")
References: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>
Message-ID: <x6fxcekwif.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> I am currently working on a fix for the gtk2-bug #3775:
>   "Package should create /opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders"
>   http://www.opencsw.org/mantis/view.php?id=3775
>
> What I have done now is setting it up on postinstall like pango.
> However, on Solaris 8 it will try to query 64 bit too, which fails.
> If you now share this to Solaris 10 you will end up without pixbuf
> loaders. Would you consider it to be ok to skip this step on
> Solaris 8?

-1

> Or could the pixbuffers be queried during package creation statically?
> That would also eliminate the ugly (and slow) postinstall.

+1

-- 
Peter

From jeff at cjsa.com  Wed Jul 29 23:11:20 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Wed, 29 Jul 2009 21:11:20 GMT
Subject: [csw-maintainers] New Problems with Berkeley DB
Message-ID: <KnKAuw.28B@cjsa.com>

I had to reboot my Solaris 10 SPARC V250 server and had the following problem
upon restart:

-------------------------------------------------------------------------------
Jul 29 13:58:57 cjsa sm-mta[636]: [ID 801593 mail.crit] NOQUEUE: SYSERR(root):
    Berkeley DB version mismatch: compiled against 4.2.52,
    run-time linked against 4.7.25

[previous message repeated 17 times]

Jul 29 13:59:01 cjsa svc.startd[7]: [ID 748625 daemon.error]
    network/smtp:cswsendmail failed repeatedly: transitioned to
    maintenance (see 'svcs -xv' for details)
-------------------------------------------------------------------------------

1-> svcs -x
svc:/network/smtp:cswsendmail (sendmail SMTP mail transfer agent)
 State: maintenance since Wed Jul 29 13:59:01 2009
Reason: Restarting too quickly.
   See: http://sun.com/msg/SMF-8000-L5
   See: sendmail(1M)
   See: /var/svc/log/network-smtp:cswsendmail.log
   Impact: This service is not running.


So there is still a compatibility problem with the Berkeley DB.  I am
running:

berkeleydb                   4.7.25,REV=2009.07.01
berkeleydb3                  3.3.11,REV=2009.07.22
berkeleydb4                  4.2.52,REV=2009.07.28
berkeleydb44                 4.4.20,REV=2009.07.28

Of cource, this will probably not go out since I do not have sendmail
running. :-(

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040


From phil at bolthole.com  Thu Jul 30 18:02:34 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Jul 2009 09:02:34 -0700
Subject: [csw-maintainers] New Problems with Berkeley DB
In-Reply-To: <KnKAuw.28B@cjsa.com>
References: <KnKAuw.28B@cjsa.com>
Message-ID: <20090730160234.GA6341@bolthole.com>

On Wed, Jul 29, 2009 at 09:11:20PM +0000, Jeffery Small wrote:
> I had to reboot my Solaris 10 SPARC V250 server and had the following problem
> upon restart:
> 
> -------------------------------------------------------------------------------
> Jul 29 13:58:57 cjsa sm-mta[636]: [ID 801593 mail.crit] NOQUEUE: SYSERR(root):
>     Berkeley DB version mismatch: compiled against 4.2.52,
>     run-time linked against 4.7.25

wow. sendmail is really dumb to do this. :-/

> Of cource, this will probably not go out since I do not have sendmail
> running. :-(

it worked though :-)
maybe it's just a whine, not an actual "error"?


From phil at bolthole.com  Thu Jul 30 18:12:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Jul 2009 09:12:21 -0700
Subject: [csw-maintainers] Statistics on ibiblio
In-Reply-To: <625385e30907300328u4061083cud8e263a48a9eb4ef@mail.gmail.com>
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>
	<F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
	<625385e30907300328u4061083cud8e263a48a9eb4ef@mail.gmail.com>
Message-ID: <20090730161221.GB6341@bolthole.com>

On Thu, Jul 30, 2009 at 12:28:59PM +0200, Peter Bonivart wrote:
> On Wed, Jul 29, 2009 at 8:12 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> > (Looks like nobody is using Solaris 8/9 and Blastwave any more)
> 
> Thanks for that link. I don't think BW lists ibiblio as a mirror any
> more so their users may have moved to other mirrors.

they have removed their archives, actually. its just a placeholder now,
seems like.

From bonivart at opencsw.org  Thu Jul 30 19:56:12 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 19:56:12 +0200
Subject: [csw-maintainers] comment on open source security
In-Reply-To: <4A717A60.3070305@dogan.ch>
References: <20090729162304.GA45823@bolthole.com> <4A717A60.3070305@dogan.ch>
Message-ID: <625385e30907301056u5155125bs425a59049cc4d831@mail.gmail.com>

On Thu, Jul 30, 2009 at 12:48 PM, Ihsan Dogan<ihsan at dogan.ch> wrote:
> The fact that Bind 10 is going to be a redesign again, shows that Bind 9
> has broken design.

On the other hand, Bind 9 was designed in 1998. In computer terms
that's a looong time ago.

Nevertheless, making Bind 10 modular will be appreciated. :-)

-- 
/peter

From bonivart at opencsw.org  Thu Jul 30 20:05:16 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 20:05:16 +0200
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com> 
	<625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com> 
	<ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>
Message-ID: <625385e30907301105q73f6c2ecl2cafc5d030bfbb0b@mail.gmail.com>

On Thu, Jul 30, 2009 at 5:03 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> I was about to ask you about it the day before yesterday. I've just
> set up an account in wiki.opencsw.org. To keep it simple and
> consistent, I used the name 'automatthias'. ;-)

I have sent you an invite now, if you accept it you're then a member
of the OpenCSW wiki. :-)

-- 
/peter

From mwatters at opencsw.org  Thu Jul 30 20:40:23 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 30 Jul 2009 13:40:23 -0500
Subject: [csw-maintainers] libcairo in testing RPATH is broken
Message-ID: <4A71E917.1010205@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

here is the excerpt from dump -Lv
this in turn breaks everything that depends on libcairo as it can not find

$ ldd /opt/csw/bin/pidgin # for example
...
        libxcb-render-util.so.0 =>       (file not found)
        libxcb-render.so.0 =>    (file not found)
...

$ dump -Lv /opt/csw/lib/sparcv8/libcairo.so.2

/opt/csw/lib/sparcv8/libcairo.so.2:

  **** DYNAMIC SECTION INFORMATION ****
...
[15]    SONAME          libcairo.so.2
[16]    RUNPATH         /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib
[17]    RPATH           /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib

...

- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpx6RcACgkQLrhmsXMSLxdmLgCfWl89CyMsfnsjQjSDSlhHB4QM
JfMAni48JaZLeYrFAcbovxXCoc92A4rx
=lWnN
-----END PGP SIGNATURE-----

From maciej at opencsw.org  Thu Jul 30 20:58:01 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 19:58:01 +0100
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <625385e30907301105q73f6c2ecl2cafc5d030bfbb0b@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
	<625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>
	<ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>
	<625385e30907301105q73f6c2ecl2cafc5d030bfbb0b@mail.gmail.com>
Message-ID: <ce97543b0907301158p75312624vbe93f49cb48e0131@mail.gmail.com>

On Thu, Jul 30, 2009 at 7:05 PM, Peter Bonivart<bonivart at opencsw.org> wrote:
> On Thu, Jul 30, 2009 at 5:03 PM, Maciej (Matchek)
> Blizinski<maciej at opencsw.org> wrote:
>> I was about to ask you about it the day before yesterday. I've just
>> set up an account in wiki.opencsw.org. To keep it simple and
>> consistent, I used the name 'automatthias'. ;-)
>
> I have sent you an invite now, if you accept it you're then a member
> of the OpenCSW wiki. :-)

Accepted. Thanks!

From dam at opencsw.org  Thu Jul 30 21:53:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 30 Jul 2009 21:53:55 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
In-Reply-To: <x6fxcekwif.fsf@abhavana.free.fr>
References: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>
	<x6fxcekwif.fsf@abhavana.free.fr>
Message-ID: <D547DC8C-5876-459A-A66A-C200894D66C1@opencsw.org>

Hi Peter,

Am 30.07.2009 um 17:42 schrieb Peter FELECAN:
>> Or could the pixbuffers be queried during package creation  
>> statically?
>> That would also eliminate the ugly (and slow) postinstall.
>
> +1

I would prefer this, but it is allowed? Or are there other modules  
that are
not present at build time?


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul 30 22:05:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 30 Jul 2009 22:05:51 +0200
Subject: [csw-maintainers] libcairo in testing RPATH is broken
In-Reply-To: <4A71E917.1010205@opencsw.org>
References: <4A71E917.1010205@opencsw.org>
Message-ID: <EBC16C99-0484-4629-80A0-FFCF7F6CD602@opencsw.org>

Hi John,

Am 30.07.2009 um 20:40 schrieb Mike Watters:
> here is the excerpt from dump -Lv
> this in turn breaks everything that depends on libcairo as it can  
> not find
>
> $ ldd /opt/csw/bin/pidgin # for example
> ...
>        libxcb-render-util.so.0 =>       (file not found)
>        libxcb-render.so.0 =>    (file not found)
> ...
>
> $ dump -Lv /opt/csw/lib/sparcv8/libcairo.so.2
>
> /opt/csw/lib/sparcv8/libcairo.so.2:
>
>  **** DYNAMIC SECTION INFORMATION ****
> ...
> [15]    SONAME          libcairo.so.2
> [16]    RUNPATH         /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib
> [17]    RPATH           /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib

This is very strange.
John: How have you build that package? With GAR? If yes, what revision  
of GAR
itself were you using? I can't remember putting $ORIGIN in anywhere.


Best regards

   -- Dago

From jeff at cjsa.com  Fri Jul 31 01:45:38 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Thu, 30 Jul 2009 23:45:38 GMT
Subject: [csw-maintainers] New Problems with Berkeley DB
References: <KnKAuw.28B@cjsa.com> <20090730160234.GA6341@bolthole.com>
Message-ID: <KnMCo2.4pr@cjsa.com>

Philip Brown <phil at bolthole.com> writes:

>On Wed, Jul 29, 2009 at 09:11:20PM +0000, Jeffery Small wrote:
>> I had to reboot my Solaris 10 SPARC V250 server and had the following problem
>> upon restart:
>> 
>> -------------------------------------------------------------------------------
>> Jul 29 13:58:57 cjsa sm-mta[636]: [ID 801593 mail.crit] NOQUEUE: SYSERR(root):
>>     Berkeley DB version mismatch: compiled against 4.2.52,
>>     run-time linked against 4.7.25

>wow. sendmail is really dumb to do this. :-/

>> Of cource, this will probably not go out since I do not have sendmail
>> running. :-(

>it worked though :-)
>maybe it's just a whine, not an actual "error"?

No, sendmail was out of commission.  I'm back up due to some great help
from Dagobert Michelsen.  This message probably only went out after I got
sendmail back online.

I currently only have the following packages:

berkeleydb                   4.7.25,REV=2009.07.01		Current
berkeleydb3                  3.3.11,REV=2009.07.22		Current
berkeleydb4           4.2.52,REV=2005.04.28_rev=p4		From old archive
berkeleydb44                 4.4.20,REV=2009.07.28		Current

And sendmail became happy once again when the old berkeleydb4 was installed.


Below, I'm now following up to an email I received from Dago.

I wrote:

> And when sendmail complained about the runtime version being 4.7.25,
> how does that relate to anything I could see regarding the installed
> packages?  Upon further examination, I see that the berkeleydb_devel
> package is listed as rev 4.7.25,REV=2009.07.01, but I don't have this
> package installed on my system.

Dago wrote:

> You must have as the new bdb4x-packages depend on it.

Well, berkeleydb44_devel only depends upon berkeleydb44, and
berkeleydb_devel presumably depends upon berkeleydb*, but not the other way
around.  I suspect That Dago meant that the berkeleydb{3,4,44} packages
depend upon berkeleydb and not the *_devel packages.

* [NOTE: From the CSW current package list webpage, if you click on
  berkeleydb or berkeleydb_devel you get and error stating: "Err: cannot find
  maintainer for <package> in mantis", and no details are shown!]

Dago wrote:

> I proposed a versioned current, where you can point your pkg*-tool to a
> specific date and which does not change like opencsw/current-20090730
> However, neither me (as I proposed it) nor another one has picked it up
> yet.  Maybe this is an opportunity to bring this ahead.  This is AFAIK a
> better solution than putting magic in pkg*.

This actually sounds like a good idea.  If you upgrade once a week and
serious new problems develop, you can point pkg-get or pkg-util to a date
for the previous week and bring things back into sync.  This would work
so long as all package dependencies could be updated/rewound reliably;
something to which I have not yet given much thought.

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040


From phil at bolthole.com  Fri Jul 31 02:14:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Jul 2009 17:14:59 -0700
Subject: [csw-maintainers] call for new sendmail maintainer?
Message-ID: <20090731001459.GB47846@bolthole.com>

So.... anyone wanna volunteer to take over sendmail and do a quick
repackaging?

Since its being picky about the berkeleydb upgrade, we need a repackaging
done asap, and we do not have an active maintainer for it any more.


From mwatters at opencsw.org  Fri Jul 31 04:24:17 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 30 Jul 2009 21:24:17 -0500
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <20090731001459.GB47846@bolthole.com>
References: <20090731001459.GB47846@bolthole.com>
Message-ID: <4A7255D1.3020000@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> So.... anyone wanna volunteer to take over sendmail and do a quick
> repackaging?
> 
> Since its being picky about the berkeleydb upgrade, we need a repackaging
> done asap, and we do not have an active maintainer for it any more.
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Benny and I are tag teaming it to get it out quickly...
He got the bulk of it done, I have the last of the install pieces
done.

I was going through the old version and Alex had created a custom class to
handle the fact that sendmail must be compiled separately on each OS.

I have to modify his class to now include at least solaris 10 and write a
custom merge script for gar.  I hope to have this done before I turn in for the
night.



- --
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/

iEYEARECAAYFAkpyVdEACgkQLrhmsXMSLxdkqgCeOP341Mezgf6tsiqH6eC+JAZ2
+VwAn33HPpvDLCO2AFyG6MOS2LR9zppq
=RZGG
-----END PGP SIGNATURE-----

From schwindt at dfki.uni-kl.de  Fri Jul 31 09:16:04 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 31 Jul 2009 09:16:04 +0200
Subject: [csw-maintainers] Updated libiconv in testing,
 testing  appreciated
In-Reply-To: Your message of "Fri, 24 Jul 2009 23:03:28 +0200."
	<4A6A21A0.3060006@opencsw.org> 
Message-ID: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>

[...]
> This should obsolete related hacks required in other packages. I have it
> running on a couple of non-GUI test boxes without hitches, but as it is
> a to-be-adopted and fundamental package, I would appreciate a bit more
> testing/feedback/review by some of you guys.


I will give your version a try, as I always did have a patched version of iconv
around, and am using a self compiled version of 1.13.1 for quite some time now.
I can state that this version generally functions well, troublemaker like 
subversion works perfect with it. 

I'll let you know how things worked out



From skayser at opencsw.org  Fri Jul 31 09:22:43 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 09:22:43 +0200 (CEST)
Subject: [csw-maintainers] Updated libiconv in testing,
 testing  appreciated
In-Reply-To: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>
References: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>
Message-ID: <49274.217.227.42.13.1249024963.squirrel@ssl.skayser.de>

Hi Nicolai,

Nicolai Schwindt wrote:
> [...]
>> This should obsolete related hacks required in other packages. I have it
>> running on a couple of non-GUI test boxes without hitches, but as it is
>> a to-be-adopted and fundamental package, I would appreciate a bit more
>> testing/feedback/review by some of you guys.
>
>
> I will give your version a try, as I always did have a patched version of
> iconv
> around, and am using a self compiled version of 1.13.1 for quite some time
> now.
> I can state that this version generally functions well, troublemaker like
> subversion works perfect with it.
>
> I'll let you know how things worked out

cool, thanks for the support. Note though: I recently pulled the iconv
packages from testing, because they were ones, aimed at the CSWiconv ->
CSWlibiconv package name change. With various bits and pieces of our stack
currently being under heavy change (gtk, pango, bdb) I would like to
postpone such a (purely cosmetical) change.

I will put new iconv packages with the current naming scheme into testing
later this day. Will keep you updated.

Sebastian


From skayser at opencsw.org  Fri Jul 31 09:25:41 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 09:25:41 +0200 (CEST)
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <4A7255D1.3020000@opencsw.org>
References: <20090731001459.GB47846@bolthole.com>
	<4A7255D1.3020000@opencsw.org>
Message-ID: <49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>

Mike Watters wrote:
> Philip Brown wrote:
>> So.... anyone wanna volunteer to take over sendmail and do a quick
>> repackaging?
>>
>> Since its being picky about the berkeleydb upgrade, we need a
>> repackaging
>> done asap, and we do not have an active maintainer for it any more.
>>
> Benny and I are tag teaming it to get it out quickly...
> He got the bulk of it done, I have the last of the install pieces
> done.
>
> I was going through the old version and Alex had created a custom class to
> handle the fact that sendmail must be compiled separately on each OS.
>
> I have to modify his class to now include at least solaris 10 and write a
> custom merge script for gar.  I hope to have this done before I turn in
> for the
> night.

Thinking in terms of simplicity: Wouldn't it also be possible to just wrap
up a package each on build{8,9,10}{s,x} to have a specific package for
each Solaris release, instead of having a potentially complicated GAR
Makefile plus postinstall magic (which might both be subject to errors)?

Just an idea, don't let your progress be slowed down by it if you think
otherwise or if there are reasons against it (no need for discussion from
my side).

Sebastian


From pfelecan at opencsw.org  Fri Jul 31 09:25:49 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 09:25:49 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
In-Reply-To: <D547DC8C-5876-459A-A66A-C200894D66C1@opencsw.org> (Dagobert
	Michelsen's message of "Thu\, 30 Jul 2009 21\:53\:55 +0200")
References: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>
	<x6fxcekwif.fsf@abhavana.free.fr>
	<D547DC8C-5876-459A-A66A-C200894D66C1@opencsw.org>
Message-ID: <x6y6q5joua.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 30.07.2009 um 17:42 schrieb Peter FELECAN:
>>> Or could the pixbuffers be queried during package creation
>>> statically?
>>> That would also eliminate the ugly (and slow) postinstall.
>>
>> +1
>
> I would prefer this, but it is allowed? Or are there other modules
> that are
> not present at build time?

I don't know (never meet the situation). However, I think that, at least
for post addition of modules is the responsibility of the provider of
the module to refresh the pixbuffers; these seems to be similar in
behaviour with fontconfig, don't you think? There are other instances of
it but I don't remember which.

BTW, what is Debian doing for this? It's not such a bad thing to see
what our friends are doing...

HTH
-- 
Peter

From maciej at opencsw.org  Fri Jul 31 10:02:24 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 09:02:24 +0100
Subject: [csw-maintainers] Warnings from GAR
Message-ID: <ce97543b0907310102l5475c4f2kbf70e985428e681e@mail.gmail.com>

When I'm building with buildbot, I'm often getting warnings from GAR:

 	==> Trying file//files/CSWap2prefork.gspec
gmake[2]: Entering directory `/home/buildbot/buildbot-slave/full/build/trunk'
gar/gar.mk:82: warning: overriding commands for target `work/build-global'
gar/gar.pkg.mk:532: warning: ignoring old commands for target
`work/build-global'

I don't see them when building by hand, but I used to see them before.
Dago, do you know why are those warnings there?

Maciej

From maciej at opencsw.org  Fri Jul 31 10:39:53 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 09:39:53 +0100
Subject: [csw-maintainers] Buildbot
Message-ID: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>

Hello maintainers,

I've written a draft of a design to deploy a continuous build system
on the buildfarm, using buildbot. The draft can be found here:
http://wiki.opencsw.org/buildbot

I have a proof of concept running on my workstation at work, with an
IRC interface 'cswbb' present on the #opencsw channel on Freenode. A
couple points based on questions from Dago:

- buildbot can handle parallel builds on the buildfarm, by running
different builds in different subdirectories
- buildbot automatically picks up changes from the repository, there's
no hand-scheduling needed
- buildbot can report the status via http, email and irc
- it runs as a headless daemon

More details, including a deployment plan is in the design draft.

Thoughts?

Maciej

From pfelecan at opencsw.org  Fri Jul 31 11:00:45 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 11:00:45 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 09\:39\:53 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
Message-ID: <x6ljm5jkg2.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> - buildbot can handle parallel builds on the buildfarm, by running
> different builds in different subdirectories

This should be done on a different system than that used for day to day
work. I don't want to be penalised by gcc, OpenOffice and other monsters
being built the same time that I wish to build a new package.

> - buildbot automatically picks up changes from the repository, there's
> no hand-scheduling needed

This is true only for "garified" packages. Is there a plug-in for other
kind of build?

> - buildbot can report the status via http, email and irc

This is good and can be useful for any build. It's a missing feature on
the current build farms.

> Thoughts?

See above.

-- 
Peter

From dam at opencsw.org  Fri Jul 31 11:49:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 31 Jul 2009 11:49:34 +0200
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>
References: <20090731001459.GB47846@bolthole.com>
	<4A7255D1.3020000@opencsw.org>
	<49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>
Message-ID: <BA565FB7-89E7-448F-8112-46C110AF3AC3@opencsw.org>

Hi,

Am 31.07.2009 um 09:25 schrieb Sebastian Kayser:
> Thinking in terms of simplicity: Wouldn't it also be possible to  
> just wrap
> up a package each on build{8,9,10}{s,x} to have a specific package for
> each Solaris release, instead of having a potentially complicated GAR
> Makefile plus postinstall magic (which might both be subject to  
> errors)?

IMHO this is practicable, but is completely against our policy of
central NFS mounted server for all releases. However, running server
services from NFS is also not good design, so I guess it is ok
to make different packages here.


Best regards

   -- Dago

From maciej at opencsw.org  Fri Jul 31 12:25:43 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 11:25:43 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6ljm5jkg2.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>

On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> - buildbot can handle parallel builds on the buildfarm, by running
>> different builds in different subdirectories
>
> This should be done on a different system than that used for day to day
> work. I don't want to be penalised by gcc, OpenOffice and other monsters
> being built the same time that I wish to build a new package.

I see two potential solutions:

1. Provide separate hardware for the buildbot
2. Avoid building certain blacklisted packages. Optional: schedule
periodic or nightly builds for those packages.

>> - buildbot automatically picks up changes from the repository, there's
>> no hand-scheduling needed
>
> This is true only for "garified" packages. Is there a plug-in for other
> kind of build?

Can you come up with an example of a different build? Where would the
source code come from?

There's one more issue: multi-ISA packages. I have a vague idea how to
implement them, but I don't have the proper environment at hand to
actually do that.

Maciej

From pfelecan at opencsw.org  Fri Jul 31 12:31:57 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 12:31:57 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 11\:25\:43 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
Message-ID: <x6hbwtjg82.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>
>>> - buildbot can handle parallel builds on the buildfarm, by running
>>> different builds in different subdirectories
>>
>> This should be done on a different system than that used for day to day
>> work. I don't want to be penalised by gcc, OpenOffice and other monsters
>> being built the same time that I wish to build a new package.
>
> I see two potential solutions:
>
> 1. Provide separate hardware for the buildbot

+1 (I know, I'm not paying the bill) and please, not a zone or a VM on
   the same hardware...

> 2. Avoid building certain blacklisted packages. Optional: schedule
> periodic or nightly builds for those packages.

-1 as it defeats the purpose; in addition, there is no night for an
   international community as ours...

>>> - buildbot automatically picks up changes from the repository, there's
>>> no hand-scheduling needed
>>
>> This is true only for "garified" packages. Is there a plug-in for other
>> kind of build?
>
> Can you come up with an example of a different build? Where would the
> source code come from?

All my packages are outside gar. The source code comes from my home
directory...

-- 
Peter

From maciej at opencsw.org  Fri Jul 31 12:39:06 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 11:39:06 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6hbwtjg82.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>

On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>
>>>> - buildbot can handle parallel builds on the buildfarm, by running
>>>> different builds in different subdirectories
>>>
>>> This should be done on a different system than that used for day to day
>>> work. I don't want to be penalised by gcc, OpenOffice and other monsters
>>> being built the same time that I wish to build a new package.
>>
>> I see two potential solutions:
>>
>> 1. Provide separate hardware for the buildbot
>
> +1 (I know, I'm not paying the bill) and please, not a zone or a VM on
> ? the same hardware...
>
>> 2. Avoid building certain blacklisted packages. Optional: schedule
>> periodic or nightly builds for those packages.
>
> -1 as it defeats the purpose; in addition, there is no night for an
> ? international community as ours...

Automatic builds for some packages are better than automatic builds
for no packages, so it doesn't defeat the purpose completely. Also, we
could do actual measurements: how big is the impact on the
performance?

>>>> - buildbot automatically picks up changes from the repository, there's
>>>> no hand-scheduling needed
>>>
>>> This is true only for "garified" packages. Is there a plug-in for other
>>> kind of build?
>>
>> Can you come up with an example of a different build? Where would the
>> source code come from?
>
> All my packages are outside gar. The source code comes from my home
> directory...

How would you make your source code available to buildbot?

Maciej

From pfelecan at opencsw.org  Fri Jul 31 12:42:44 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 12:42:44 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 11\:39\:06 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
Message-ID: <x6d47hjfq3.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>
>>> On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>>
>>>>> - buildbot can handle parallel builds on the buildfarm, by running
>>>>> different builds in different subdirectories
>>>>
>>>> This should be done on a different system than that used for day to day
>>>> work. I don't want to be penalised by gcc, OpenOffice and other monsters
>>>> being built the same time that I wish to build a new package.
>>>
>>> I see two potential solutions:
>>>
>>> 1. Provide separate hardware for the buildbot
>>
>> +1 (I know, I'm not paying the bill) and please, not a zone or a VM on
>> ? the same hardware...
>>
>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>> periodic or nightly builds for those packages.
>>
>> -1 as it defeats the purpose; in addition, there is no night for an
>> ? international community as ours...
>
> Automatic builds for some packages are better than automatic builds
> for no packages, so it doesn't defeat the purpose completely. Also, we
> could do actual measurements: how big is the impact on the
> performance?

Maybe. But not on the effective build stack, please.

>
>>>>> - buildbot automatically picks up changes from the repository, there's
>>>>> no hand-scheduling needed
>>>>
>>>> This is true only for "garified" packages. Is there a plug-in for other
>>>> kind of build?
>>>
>>> Can you come up with an example of a different build? Where would the
>>> source code come from?
>>
>> All my packages are outside gar. The source code comes from my home
>> directory...
>
> How would you make your source code available to buildbot?

By pointing it to my home directory on the build stack...
-- 
Peter

From skayser at opencsw.org  Fri Jul 31 12:54:47 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 12:54:47 +0200
Subject: [csw-maintainers] Updated libiconv in testing,
	testing  appreciated
In-Reply-To: <49274.217.227.42.13.1249024963.squirrel@ssl.skayser.de>
References: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>
	<49274.217.227.42.13.1249024963.squirrel@ssl.skayser.de>
Message-ID: <4A72CD77.9090209@opencsw.org>

Sebastian Kayser wrote:
> Nicolai Schwindt wrote:
>> [...]
>>> This should obsolete related hacks required in other packages. I have it
>>> running on a couple of non-GUI test boxes without hitches, but as it is
>>> a to-be-adopted and fundamental package, I would appreciate a bit more
>>> testing/feedback/review by some of you guys.
>>
>> I will give your version a try, as I always did have a patched version of
>> iconv
>> around, and am using a self compiled version of 1.13.1 for quite some time
>> now.
>> I can state that this version generally functions well, troublemaker like
>> subversion works perfect with it.
>>
>> I'll let you know how things worked out
> 
> cool, thanks for the support. Note though: I recently pulled the iconv
> packages from testing, because they were ones, aimed at the CSWiconv ->
> CSWlibiconv package name change. With various bits and pieces of our stack
> currently being under heavy change (gtk, pango, bdb) I would like to
> postpone such a (purely cosmetical) change.
> 
> I will put new iconv packages with the current naming scheme into testing
> later this day. Will keep you updated.

Ok, please give the packages in testing a try.

http://mirror.opencsw.org/testing/libiconv-1.13.1,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/libiconv-1.13.1,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz

Modeled after the ones in current, i.e. no spearate devel package, no
pkg/swname changes. Should be least intrusive. File clash on
/opt/csw/lib/{sparcv9,amd64}/charset.alias with gnugpg is known and
already filed [1]. Shouldn't do any harm though, the files are identical.

Sebastian

[1] http://opencsw.org/bugtrack/view.php?id=3755

From maciej at opencsw.org  Fri Jul 31 13:15:42 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 12:15:42 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6d47hjfq3.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	<x6d47hjfq3.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>

On Fri, Jul 31, 2009 at 11:42 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>
>>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>>> periodic or nightly builds for those packages.
>>>
>>> -1 as it defeats the purpose; in addition, there is no night for an
>>> ? international community as ours...
>>
>> Automatic builds for some packages are better than automatic builds
>> for no packages, so it doesn't defeat the purpose completely. Also, we
>> could do actual measurements: how big is the impact on the
>> performance?
>
> Maybe. But not on the effective build stack, please.

Where are large packages currently built? I thought that all packages
must be built on the buildfarm. Next time there's an on-going
compilation of a large package, we can do the measurements.

>>>>>> - buildbot automatically picks up changes from the repository, there's
>>>>>> no hand-scheduling needed
>>>>>
>>>>> This is true only for "garified" packages. Is there a plug-in for other
>>>>> kind of build?
>>>>
>>>> Can you come up with an example of a different build? Where would the
>>>> source code come from?
>>>
>>> All my packages are outside gar. The source code comes from my home
>>> directory...
>>
>> How would you make your source code available to buildbot?
>
> By pointing it to my home directory on the build stack...

Buildbot has plugins for Subversion, Perforce, Arch, Bazaar, Darcs,
Mercurial and Git. Do you use any of these for your source code?

Buildbot needs essentially three things:

1. Build master needs a way to learn about an update to specific
project or branch
2. Build slaves need a way to obtain the source code
3. Build slaves need instructions to build the package

Number 1 could be done by... I don't know... making md5 sums of
specific files in your home directory? Number 2 would be a file copy
operation. Number 3 would be provided by you. Do your builds have a
uniform build procedure?

The GAR packages will be the main focus now, since they comprise about
91% builds (that I know of, that is, ~1083 out of ~1186). Once we're
there, we can extend support to other build sources. Of course, it
would make it easier to implement if a code repository was available
to buildbot.

Maciej

From ihsan at dogan.ch  Fri Jul 31 13:18:58 2009
From: ihsan at dogan.ch (Ihsan Dogan)
Date: Fri, 31 Jul 2009 13:18:58 +0200
Subject: [csw-maintainers] Apache 2.2.12 in testing
Message-ID: <4A72D322.9060906@dogan.ch>

Hello,

I've updated Apache to 2.2.12 and it's available on testing now. Apache
is now linked against Berkeley DB 4.7.

All architectures:
http://mirror.opencsw.org/testing/apache2-2.2.12,REV=2009.07.31-SunOS5.8-all-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2_manual-2.2.12,REV=2009.07.31-SunOS5.8-all-CSW.pkg.gz

sparc:
http://mirror.opencsw.org/testing/ap2_prefork-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_suexec-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_worker-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2_devel-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2c-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2rt-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz

i386:
http://mirror.opencsw.org/testing/ap2_prefork-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_suexec-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_worker-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2_devel-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2c-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2rt-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz

Feedback is always welcome and appreciated.



Ihsan
-- 
ihsan at dogan.ch		http://blog.dogan.ch/

From dam at opencsw.org  Fri Jul 31 13:19:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 31 Jul 2009 13:19:46 +0200
Subject: [csw-maintainers] Warnings from GAR
In-Reply-To: <ce97543b0907310102l5475c4f2kbf70e985428e681e@mail.gmail.com>
References: <ce97543b0907310102l5475c4f2kbf70e985428e681e@mail.gmail.com>
Message-ID: <602A5446-53BC-4DC1-83D4-60171DE413A8@opencsw.org>

Hi Maciej,

Am 31.07.2009 um 10:02 schrieb Maciej (Matchek) Blizinski:
> When I'm building with buildbot, I'm often getting warnings from GAR:
>
> 	==> Trying file//files/CSWap2prefork.gspec
> gmake[2]: Entering directory `/home/buildbot/buildbot-slave/full/ 
> build/trunk'
> gar/gar.mk:82: warning: overriding commands for target `work/build- 
> global'
> gar/gar.pkg.mk:532: warning: ignoring old commands for target
> `work/build-global'
>
> I don't see them when building by hand, but I used to see them before.
> Dago, do you know why are those warnings there?

This is because you didn't specify where to put the packages in
your ~/.garrc. For me this is

SPKG_EXPORT     = /home/dam/staging/build-$(shell date '+%d.%b.%Y')

Best regards

   -- Dago


From dam at opencsw.org  Fri Jul 31 13:25:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 31 Jul 2009 13:25:01 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6hbwtjg82.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
Message-ID: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>

Hi Peter,

Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
> All my packages are outside gar. The source code comes from my home
> directory...

- Is there a specific reason why you are not using GAR? Without GAR it  
is
   hard to do teams, takeover, knowledge sharing, consistent packages,  
etc.
- If you are not using GAR, please also check in the source to
     mgar/pkg/<pkg>/trunk/legacy/
   so it is properly archived. Missing build descriptions on maintainers
   leaving is a hard thing to cope with. IMHO it should be policy that
   no package is published without the recipe committed to the  
repository.


Best regards

   -- Dago

From bwalton at opencsw.org  Fri Jul 31 13:43:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 07:43:59 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
Message-ID: <1249040535-sup-2622@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Fri Jul 31 07:25:01 -0400 2009:

> - If you are not using GAR, please also check in the source to
>   mgar/pkg/<pkg>/trunk/legacy/ so it is properly archived. Missing
>   build descriptions on maintainers leaving is a hard thing to cope
>   with. IMHO it should be policy that no package is published
>   without the recipe committed to the repository.

+1 for this.  If you're still a GAR-o-phobe, at least commit the
manual build scripts, patches, instructions etc so that knowledge
doesn't get lost...[The svn-o-phobe's may not like this. <grin>]

Is this something that could be put on the Oslo agenda?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/0932e918/attachment-0001.asc>

From pfelecan at opencsw.org  Fri Jul 31 16:09:31 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 16:09:31 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 12\:15\:42 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	<x6d47hjfq3.fsf@abhavana.free.fr>
	<ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>
Message-ID: <x68wi5j65g.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 31, 2009 at 11:42 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>
>>> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>>
>>>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>>>> periodic or nightly builds for those packages.
>>>>
>>>> -1 as it defeats the purpose; in addition, there is no night for an
>>>> ? international community as ours...
>>>
>>> Automatic builds for some packages are better than automatic builds
>>> for no packages, so it doesn't defeat the purpose completely. Also, we
>>> could do actual measurements: how big is the impact on the
>>> performance?
>>
>> Maybe. But not on the effective build stack, please.
>
> Where are large packages currently built? I thought that all packages
> must be built on the buildfarm. Next time there's an on-going
> compilation of a large package, we can do the measurements.

Some weeks ago we had I/O issues when gcc3, OpenOffice and Firefox where
built in the same period. What I'm saying, is *automatic build must be
done on a different hardware stack*

>>>>>>> - buildbot automatically picks up changes from the repository, there's
>>>>>>> no hand-scheduling needed
>>>>>>
>>>>>> This is true only for "garified" packages. Is there a plug-in for other
>>>>>> kind of build?
>>>>>
>>>>> Can you come up with an example of a different build? Where would the
>>>>> source code come from?
>>>>
>>>> All my packages are outside gar. The source code comes from my home
>>>> directory...
>>>
>>> How would you make your source code available to buildbot?
>>
>> By pointing it to my home directory on the build stack...
>
> Buildbot has plugins for Subversion, Perforce, Arch, Bazaar, Darcs,
> Mercurial and Git. Do you use any of these for your source code?
>
> Buildbot needs essentially three things:
>
> 1. Build master needs a way to learn about an update to specific
> project or branch
> 2. Build slaves need a way to obtain the source code
> 3. Build slaves need instructions to build the package
>
> Number 1 could be done by... I don't know... making md5 sums of
> specific files in your home directory? Number 2 would be a file copy
> operation. Number 3 would be provided by you. Do your builds have a
> uniform build procedure?

Yes, my packages have an uniform and automatic packaging system.

> The GAR packages will be the main focus now, since they comprise about
> 91% builds (that I know of, that is, ~1083 out of ~1186). Once we're
> there, we can extend support to other build sources. Of course, it
> would make it easier to implement if a code repository was available
> to buildbot.

Are you sure for the figures above?
-- 
Peter

From pfelecan at opencsw.org  Fri Jul 31 16:14:42 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 16:14:42 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 31 Jul 2009 13\:25\:01 +0200")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
Message-ID: <x64ostj5wt.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
>> All my packages are outside gar. The source code comes from my home
>> directory...
>
> - Is there a specific reason why you are not using GAR? Without GAR it
> is
>   hard to do teams, takeover, knowledge sharing, consistent packages,
> etc.

Yes: I prefer to use my own "sausage factory". Known complexity versus
unknown complexity. Besides, my "volunteer-ship" on this project
predating gar's usage gives me the choice to use the system that I
prefer.

When the politics of the foundation change and the usage of gar is
mandatory I reserve my decision.
-- 
Peter

From pfelecan at opencsw.org  Fri Jul 31 16:20:23 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 16:20:23 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249040535-sup-2622@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Fri\, 31 Jul 2009 07\:43\:59 -0400")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<1249040535-sup-2622@ntdws12.chass.utoronto.ca>
Message-ID: <x6zlalhr2w.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

> Excerpts from Dagobert Michelsen's message of Fri Jul 31 07:25:01 -0400 2009:
>
>> - If you are not using GAR, please also check in the source to
>>   mgar/pkg/<pkg>/trunk/legacy/ so it is properly archived. Missing
>>   build descriptions on maintainers leaving is a hard thing to cope
>>   with. IMHO it should be policy that no package is published
>>   without the recipe committed to the repository.
>
> +1 for this.  If you're still a GAR-o-phobe, at least commit the
> manual build scripts, patches, instructions etc so that knowledge
> doesn't get lost...[The svn-o-phobe's may not like this. <grin>]

Yeah, when all the methods, procedures and tools involved in our
processes are documented it'll be reasonable to ask. Until then, you are
free to look in my home directory on the build stack, where everything
is open.

I know recipes for RPM's, DEB's, &c. What's a recipe from the point of
view of gar? Can you show me one?
-- 
Peter

From maciej at opencsw.org  Fri Jul 31 16:25:45 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 15:25:45 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x68wi5j65g.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	<x6d47hjfq3.fsf@abhavana.free.fr>
	<ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>
	<x68wi5j65g.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310725i677d8630u811d6b938659001b@mail.gmail.com>

On Fri, Jul 31, 2009 at 3:09 PM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> On Fri, Jul 31, 2009 at 11:42 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>
>>>> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>>>
>>>>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>>>>> periodic or nightly builds for those packages.
>>>>>
>>>>> -1 as it defeats the purpose; in addition, there is no night for an
>>>>> ? international community as ours...
>>>>
>>>> Automatic builds for some packages are better than automatic builds
>>>> for no packages, so it doesn't defeat the purpose completely. Also, we
>>>> could do actual measurements: how big is the impact on the
>>>> performance?
>>>
>>> Maybe. But not on the effective build stack, please.
>>
>> Where are large packages currently built? I thought that all packages
>> must be built on the buildfarm. Next time there's an on-going
>> compilation of a large package, we can do the measurements.
>
> Some weeks ago we had I/O issues when gcc3, OpenOffice and Firefox where
> built in the same period. What I'm saying, is *automatic build must be
> done on a different hardware stack*

buildbot won't build more than one package on a single system. It
might build more than on the same hardware, because it doesn't know
about the VM layout.

>> The GAR packages will be the main focus now, since they comprise about
>> 91% builds (that I know of, that is, ~1083 out of ~1186). Once we're
>> there, we can extend support to other build sources. Of course, it
>> would make it easier to implement if a code repository was available
>> to buildbot.
>
> Are you sure for the figures above?

Do you want to argue about the figures? :-)

I counted the packages I knew about: the number of subdirectories in
mgar/pkg, plus mgar/pkg/cpan, plus number of packages listed on your
maitainer page. Feel free to offer better figures! :-)

Maciej

From mwatters at opencsw.org  Fri Jul 31 16:50:24 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 09:50:24 -0500
Subject: [csw-maintainers] Freenode?
Message-ID: <4A7304B0.1040201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I can't seem to login to freenode using either xChat or mibbit or anything.  I
keep getting 127.0.0.1 (banned) but I am not passing that as my address?

any thoughts?

- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzBLAACgkQLrhmsXMSLxf/swCglbAaA3C5Eqi5QmORhgUIB2BD
9/IAoLy238q2lNonXy/h5hW1LcoxTyne
=CpGf
-----END PGP SIGNATURE-----

From mwatters at opencsw.org  Fri Jul 31 16:53:32 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 09:53:32 -0500
Subject: [csw-maintainers] in lieu of irc here is my update on sendmail
Message-ID: <4A73056C.3030807@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I have the "install done and working correctly"
I am getting ready to write the custom merge script and should have that done
shortly.


- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzBWwACgkQLrhmsXMSLxdG7gCgnJzYNrmBqnAwD8Vr3zq4YbG4
cxMAoLgNt1apevl9TBT466vrkhXaNJMU
=5NFR
-----END PGP SIGNATURE-----

From maciej at opencsw.org  Fri Jul 31 16:53:22 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 15:53:22 +0100
Subject: [csw-maintainers] Freenode?
In-Reply-To: <4A7304B0.1040201@opencsw.org>
References: <4A7304B0.1040201@opencsw.org>
Message-ID: <ce97543b0907310753j4375c52bv6f485cf06603616a@mail.gmail.com>

On Fri, Jul 31, 2009 at 3:50 PM, Mike Watters<mwatters at opencsw.org> wrote:
> I can't seem to login to freenode using either xChat or mibbit or anything. ?I
> keep getting 127.0.0.1 (banned) but I am not passing that as my address?
>
> any thoughts?

I've seen the same problem today: cswbb couldn't connect to the
channel. Which specific host name and port are you trying to connect
to?

Maciej

From schwindt at dfki.uni-kl.de  Fri Jul 31 16:58:10 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 31 Jul 2009 16:58:10 +0200
Subject: [csw-maintainers] Empty prototype even though
 PKGFILES_CSWcupsclient is set correctly
In-Reply-To: Your message of "Mon, 27 Jul 2009 17:01:15 BST."
	<ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com> 
Message-ID: <200907311458.n6VEwAFp019823@dfki.uni-kl.de>

> I'm having an issue with the packaging of CUPS. It was all jolly good
> until prototypes of some packages started becoming empty. It once was
> CSWlibcups. Its prototype was simply empty, and so was the package.


Talking of cups - I just pulled cups from testing 

like this :

pkgutil -t http://mirror.opencsw.org/opencsw/testing -U -u  CSWcupsclient 
CSWcupsdev



which gives me this : 

tcsh at isg-1516 [/opt/csw/lib] uname -a
SunOS isg-1516 5.10 Generic_141415-05 i86pc i386 i86pc
tcsh at isg-1516 [/opt/csw/lib] file /opt/csw/bin/lpstat
/opt/csw/bin/lpstat:    ELF 32-bit MSB executable SPARC Version 1, dynamically 
linked, stripped


Anyone else ? Or am I doing wrong ?




From skayser at opencsw.org  Fri Jul 31 16:58:32 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 16:58:32 +0200
Subject: [csw-maintainers] Freenode?
In-Reply-To: <4A7304B0.1040201@opencsw.org>
References: <4A7304B0.1040201@opencsw.org>
Message-ID: <4A730698.3050702@opencsw.org>

Mike Watters wrote:
> I can't seem to login to freenode using either xChat or mibbit or anything.  I
> keep getting 127.0.0.1 (banned) but I am not passing that as my address?
> 
> any thoughts?

At least for mibbit.com my last status was that freenode had banned
mibbit (abuse problems) quite a while ago and implemented their own
web-based chat.

  http://webchat.freenode.net/?channels=opencsw

Sebastian

From bonivart at opencsw.org  Fri Jul 31 17:04:56 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Fri, 31 Jul 2009 17:04:56 +0200
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <200907311458.n6VEwAFp019823@dfki.uni-kl.de>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com> 
	<200907311458.n6VEwAFp019823@dfki.uni-kl.de>
Message-ID: <625385e30907310804g647c3634x4a201e21f01a5064@mail.gmail.com>

On Fri, Jul 31, 2009 at 4:58 PM, Nicolai
Schwindt<schwindt at dfki.uni-kl.de> wrote:
> Talking of cups - I just pulled cups from testing
>
> like this :
>
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -U -u ?CSWcupsclient
> CSWcupsdev
>
>
>
> which gives me this :
>
> tcsh at isg-1516 [/opt/csw/lib] uname -a
> SunOS isg-1516 5.10 Generic_141415-05 i86pc i386 i86pc
> tcsh at isg-1516 [/opt/csw/lib] file /opt/csw/bin/lpstat
> /opt/csw/bin/lpstat: ? ?ELF 32-bit MSB executable SPARC Version 1, dynamically
> linked, stripped
>
>
> Anyone else ? Or am I doing wrong ?

I have never seen pkgutil get this wrong. Can you provide the output
of "pkginfo -l CSWcupsclient" and "pkginfo -l CSWcupsdev" to see what
architecture the package is for? I want to know if the package file is
named wrong or what.

-- 
/peter

From maciej at opencsw.org  Fri Jul 31 17:05:41 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 16:05:41 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <200907311458.n6VEwAFp019823@dfki.uni-kl.de>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
	<200907311458.n6VEwAFp019823@dfki.uni-kl.de>
Message-ID: <ce97543b0907310805m6d643d70oe7ae956c5532d221@mail.gmail.com>

On Fri, Jul 31, 2009 at 3:58 PM, Nicolai
Schwindt<schwindt at dfki.uni-kl.de> wrote:
>> I'm having an issue with the packaging of CUPS. It was all jolly good
>> until prototypes of some packages started becoming empty. It once was
>> CSWlibcups. Its prototype was simply empty, and so was the package.
>
>
> Talking of cups - I just pulled cups from testing
>
> like this :
>
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -U -u ?CSWcupsclient
> CSWcupsdev
>
>
>
> which gives me this :
>
> tcsh at isg-1516 [/opt/csw/lib] uname -a
> SunOS isg-1516 5.10 Generic_141415-05 i86pc i386 i86pc
> tcsh at isg-1516 [/opt/csw/lib] file /opt/csw/bin/lpstat
> /opt/csw/bin/lpstat: ? ?ELF 32-bit MSB executable SPARC Version 1, dynamically
> linked, stripped
>
>
> Anyone else ? Or am I doing wrong ?

You're doing it right. The package marked x86 actually contains sparc
binaries. I'm fixing it now.

Maciej

From mwatters at opencsw.org  Fri Jul 31 17:14:54 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 10:14:54 -0500
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <BA565FB7-89E7-448F-8112-46C110AF3AC3@opencsw.org>
References: <20090731001459.GB47846@bolthole.com>	<4A7255D1.3020000@opencsw.org>	<49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>
	<BA565FB7-89E7-448F-8112-46C110AF3AC3@opencsw.org>
Message-ID: <4A730A6E.5060609@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
Dagobert Michelsen wrote:
> Hi,

> IMHO this is practicable, but is completely against our policy of
> central NFS mounted server for all releases. However, running server
> services from NFS is also not good design, so I guess it is ok
> to make different packages here.
> 

I am all for a separate package for each OS. the way Alex's and the current
class script works is it will only install the binaries for the "host OS"
so NFS mounting to different OSs and running sendmail won't work anyway.

I will create the seperate OS packages and create a "common ARCHALL" for the
non-binary stuff.  there are 9 OS specific binaries.

- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzCm4ACgkQLrhmsXMSLxcXvQCgmL5wMnNOedlrNOh0a7DuXQvl
NQgAoJccvFez9D+Qa+f0qqsHTNzi/YLP
=32Nr
-----END PGP SIGNATURE-----

From schwindt at dfki.uni-kl.de  Fri Jul 31 17:16:47 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 31 Jul 2009 17:16:47 +0200
Subject: [csw-maintainers] Empty prototype even though
 PKGFILES_CSWcupsclient is set correctly
In-Reply-To: Your message of "Fri, 31 Jul 2009 17:04:56 +0200."
	<625385e30907310804g647c3634x4a201e21f01a5064@mail.gmail.com> 
Message-ID: <200907311516.n6VFGlIS020054@dfki.uni-kl.de>

[...]
> I have never seen pkgutil get this wrong. Can you provide the output
> of "pkginfo -l CSWcupsclient" and "pkginfo -l CSWcupsdev" to see what
> architecture the package is for? I want to know if the package file is
> named wrong or what.

I reverted both affected machines ( both x86 ) back. 
I'll redo it and get back to you 


Nicolai



From maciej at opencsw.org  Fri Jul 31 18:12:14 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 17:12:14 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <200907311516.n6VFGlIS020054@dfki.uni-kl.de>
References: <625385e30907310804g647c3634x4a201e21f01a5064@mail.gmail.com>
	<200907311516.n6VFGlIS020054@dfki.uni-kl.de>
Message-ID: <ce97543b0907310912v720aab29x23c04cb4f4ecf5cd@mail.gmail.com>

On Fri, Jul 31, 2009 at 4:16 PM, Nicolai
Schwindt<schwindt at dfki.uni-kl.de> wrote:
> [...]
>> I have never seen pkgutil get this wrong. Can you provide the output
>> of "pkginfo -l CSWcupsclient" and "pkginfo -l CSWcupsdev" to see what
>> architecture the package is for? I want to know if the package file is
>> named wrong or what.
>
> I reverted both affected machines ( both x86 ) back.
> I'll redo it and get back to you

Fixed packages are in testing. Sorry for the inconvenience.

Maciej

From phil at bolthole.com  Fri Jul 31 18:47:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 09:47:57 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
Message-ID: <20090731164757.GB30823@bolthole.com>

On Fri, Jul 31, 2009 at 01:25:01PM +0200, Dagobert Michelsen wrote:
> Hi Peter,
>
> Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
>> All my packages are outside gar. The source code comes from my home
>> directory...
>
> - Is there a specific reason why you are not using GAR? Without GAR it  
> is
>   hard to do teams, takeover, knowledge sharing, consistent packages,  
> etc.


I will point out that there are two things, currently intertwined, that
dont neccessarily HAVE to be. that being:

1. our "source tree"
2. our "GAR build system"


If there were a simplified published API, for the source tree, that would
accept and interface build systems other than gar, then perhaps Peter F
(and "others" *cough*) would be more inclined to check their stuff into it.

for example, if there were a flat namespace in the source tree, and if it
were always possible to do

cd mgar/pkg/softwarename
  #and then one or more of
  make
  make package
  (?make clean?)

Define an absolute top-level "API" that doesnt require GAR, and you might
see more stuff in there, even from the "GAR-o-phobes"
Right now, you have sort of a defacto one from GAR usage. But you havent
formally stated, "as long as you follow [this API], its ok even if you're
not using gar]"



From maciej at opencsw.org  Fri Jul 31 19:34:45 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 18:34:45 +0100
Subject: [csw-maintainers] Build systems usage
Message-ID: <ce97543b0907311034q72e3c08aieddf3a534267bc7b@mail.gmail.com>

[splitting the thread]

On Fri, Jul 31, 2009 at 5:47 PM, Philip Brown<phil at bolthole.com> wrote:
> On Fri, Jul 31, 2009 at 01:25:01PM +0200, Dagobert Michelsen wrote:
>> Hi Peter,
>>
>> Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
>>> All my packages are outside gar. The source code comes from my home
>>> directory...
>>
>> - Is there a specific reason why you are not using GAR? Without GAR it
>> is
>> ? hard to do teams, takeover, knowledge sharing, consistent packages,
>> etc.
>
>
> I will point out that there are two things, currently intertwined, that
> dont neccessarily HAVE to be. that being:
>
> 1. our "source tree"
> 2. our "GAR build system"
>
>
> If there were a simplified published API, for the source tree, that would
> accept and interface build systems other than gar, then perhaps Peter F
> (and "others" *cough*) would be more inclined to check their stuff into it.
>
> for example, if there were a flat namespace in the source tree, and if it
> were always possible to do
>
> cd mgar/pkg/softwarename
> ?#and then one or more of
> ?make
> ?make package
> ?(?make clean?)
>
> Define an absolute top-level "API" that doesnt require GAR, and you might
> see more stuff in there, even from the "GAR-o-phobes"
> Right now, you have sort of a defacto one from GAR usage. But you havent
> formally stated, "as long as you follow [this API], its ok even if you're
> not using gar]"

From bwalton at opencsw.org  Fri Jul 31 19:38:50 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 13:38:50 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6zlalhr2w.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<1249040535-sup-2622@ntdws12.chass.utoronto.ca>
	<x6zlalhr2w.fsf@abhavana.free.fr>
Message-ID: <1249061256-sup-6411@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Fri Jul 31 10:20:23 -0400 2009:

> Yeah, when all the methods, procedures and tools involved in our
> processes are documented it'll be reasonable to ask. Until then, you are
> free to look in my home directory on the build stack, where everything
> is open.

Ok, that's a fair point against not using GAR, but you could still pop
your stuff into the legacy part of the svn tree...

Documentation is a weak point, and I've really not contributed much in
that regard.  There should be enough info to get started with 'out
there' but part of the problem right now is that it's scattered around
somewhat.

If you're willing to learn, though, I'll help you move some packages
into GAR...offer valid for as long as you want it to be.  Any
maintainers interested in getting started with GAR can _always_ feel
free to ping me directly.

> I know recipes for RPM's, DEB's, &c. What's a recipe from the point of
> view of gar? Can you show me one?

It's the same really.  You're defining 'special' variables, providing
custom bits of scripts (when necessary) in the appropriate places and
learning the appropriate Make targets.

RPM defined it's own domain specific language for this...GAR is a
language built on top of functionality offered by (GNU) Make.  It's
very flexible, allowing you to interject code at pretty much any place
you need to (and for most builds, some you don't!).

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/b683385e/attachment.asc>

From maciej at opencsw.org  Fri Jul 31 19:43:50 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 18:43:50 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731164757.GB30823@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
Message-ID: <ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>

[...]

I've split the thread about the build systems into another one, and
I'd like to go back to the buildbot topic. Are there any comments
about the design or deployment plan?

Maciej

From bwalton at opencsw.org  Fri Jul 31 19:50:33 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 13:50:33 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6zlalhr2w.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<1249040535-sup-2622@ntdws12.chass.utoronto.ca>
	<x6zlalhr2w.fsf@abhavana.free.fr>
Message-ID: <1249061943-sup-231@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Fri Jul 31 10:20:23 -0400 2009:
> I know recipes for RPM's, DEB's, &c. What's a recipe from the point of
> view of gar? Can you show me one?

Just so that it stands out a bit, I'm linking to a few 'Recipes' that
I've done recently.

gitosis: http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/gitosis/trunk/Makefile

php mode for emacs:
http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/phpmode/trunk/Makefile

git: http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/git/trunk/Makefile

These are of varying complexity and demonstrate a few of the different
things that are commonly done.  Gitosis leverages Mike's new class
action script cswpycompile to generate .py{c,o} files at install.
php_mode hooks into both the post-extract and post-install steps of
the build to add custom script.  Git splits one source package into
several smaller packages, uses a git repo instead of a tarball for the
source and hooks a few more steps in the process.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/331223d3/attachment.asc>

From pfelecan at opencsw.org  Fri Jul 31 19:49:00 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 19:49:00 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 18\:43\:50 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
Message-ID: <x6vdl8ivzn.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> [...]
>
> I've split the thread about the build systems into another one, and
> I'd like to go back to the buildbot topic. Are there any comments
> about the design or deployment plan?

Split or not split, my comments are valid for this
discussion. Consequently, please take into account this design
requirements:

1. The buildbot deployment must be done on a separate hardware from that
   used for the regular, "manual", builds.

2. Other build systems that gar must be able to use the deployed
   "automatic" build infrastructure using buildbot.

TIA
-- 
Peter

From bwalton at opencsw.org  Fri Jul 31 20:07:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 14:07:01 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731164757.GB30823@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
Message-ID: <1249062711-sup-8879@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 31 12:47:57 -0400 2009:

> for example, if there were a flat namespace in the source tree, and if it
> were always possible to do

Part of the reason for the (admittedly) nasty directory structure is
subversion.  While we're following 'best'[1] practices for svn use by
having the {trunk,branches,tags}/ directory structure under each top
level package, it definitely makes the tree structure uglier.

The ability to checkout only a single part of the tree with svn is
it's only useful feature[2], I think.  This feature makes following the
three sub directory convention beneficial as tagging and branching can
be done more easily on a partial checkout[3]

> Define an absolute top-level "API" that doesnt require GAR, and you might
> see more stuff in there, even from the "GAR-o-phobes"
> Right now, you have sort of a defacto one from GAR usage. But you havent
> formally stated, "as long as you follow [this API], its ok even if you're
> not using gar]"

This is a good idea.  While I think GAR is good and it saves me lots
of time, we should promote a set of Make targets that all packages
support, even if they don't use GAR.  What should a Makefile
accomplish so that it's globally usable, say by a build bot system?

I'd say:

1. It must be able to retrieve it's own source tarball/repo/whatever,
   so the 'fetch' target should be honoured.
2. It must be able to apply patches supplied locally, so 'patch' is a
   must.
3. I'd then say both 'build' and 'package' should be individually
   callable steps.  A buildbot may only want to test that the updated
   commit can still successfully build the package while a release
   tool may need to be able to take things right up to the package
   files.
4. A 'clean' target is always a good thing to support.
5. A set of variables should be honoured by any Makefile such that a
   build system knows where to find outputs.

Moving to an API like this should promote more use of the global
repository while still making it usable for a global build/release
tool.  Are there other globally useful make targets I've overlooked?

Thoughts?
-Ben

[1] Putting best in a sentence with svn feels wrong.
[2] When comparing features offered by any modern SCM/VCS tool.
[3] This is my understanding.  Corrections welcome.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/2f0b2af1/attachment.asc>

From james at opencsw.org  Fri Jul 31 20:15:10 2009
From: james at opencsw.org (James Lee)
Date: Fri, 31 Jul 2009 18:15:10 GMT
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731164757.GB30823@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
Message-ID: <20090731.18151000.1268557623@gyor.oxdrove.co.uk>

On 31/07/09, 17:47:57, Philip Brown <phil at bolthole.com> wrote regarding Re:
[csw-maintainers] Buildbot:

> If there were a simplified published API, for the source tree, that would
> accept and interface build systems other than gar, then perhaps Peter F
> (and "others" *cough*) would be more inclined to check their stuff into
> it.

> for example, if there were a flat namespace in the source tree, and if it
> were always possible to do

> cd mgar/pkg/softwarename
>   #and then one or more of
>   make
>   make package
>   (?make clean?)

> Define an absolute top-level "API" that doesnt require GAR

Go further, just have a single entry point per project (not package) per
version that can be run with exec.  i.e., don't use make as the first
step.  You can call make from a script if you want, I suppose so too
could one call a script from make but it's not as pure and simplistic.

A problem with GAR is using excess make syntax which has to one of the
less expressive ways of defining a procedure.




James.

From james at opencsw.org  Fri Jul 31 20:15:08 2009
From: james at opencsw.org (James Lee)
Date: Fri, 31 Jul 2009 18:15:08 GMT
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
Message-ID: <20090731.18150800.1547323678@gyor.oxdrove.co.uk>

On 31/07/09, 18:43:50, Maciej "(Matchek)" Blizinski <maciej at opencsw.org> 
wrote regarding Re: [csw-maintainers] Buildbot:

> I'd like to go back to the buildbot topic. Are there any comments
> about the design or deployment plan?

Why do you think it's needed at all?  Why do you want to rebuild
what presumably has been built already - is the World not hot
enough for you?



James.

From phil at bolthole.com  Fri Jul 31 20:41:08 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 11:41:08 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249062711-sup-8879@ntdws12.chass.utoronto.ca>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
Message-ID: <20090731184108.GB76412@bolthole.com>

On Fri, Jul 31, 2009 at 02:07:01PM -0400, Ben Walton wrote:
> 
> 1. It must be able to retrieve it's own source tarball/repo/whatever,
>    so the 'fetch' target should be honoured.
> 2. It must be able to apply patches supplied locally, so 'patch' is a
>    must.

I think applying the patch should be part of "get source", and there should
not be a mandated separate one.

> 3. I'd then say both 'build' and 'package' should be individually
>    callable steps....

> 4. A 'clean' target is always a good thing to support.


> 5. A set of variables should be honoured by any Makefile such that a
>    build system knows where to find outputs.

I think this is making it too complicated.
It needs to be *simple*,for most widespread use.

*standard* makefile vars, such as CFLAGS, etc, are fine. 
But other things I think should be optional. Otherwise, people tend to get
carried away defining extra things that "must" be supported.





From phil at bolthole.com  Fri Jul 31 20:42:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 11:42:45 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731.18151000.1268557623@gyor.oxdrove.co.uk>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<20090731.18151000.1268557623@gyor.oxdrove.co.uk>
Message-ID: <20090731184245.GC76412@bolthole.com>

On Fri, Jul 31, 2009 at 06:15:10PM +0000, James Lee wrote:
> 
> Go further, just have a single entry point per project (not package) per
> version that can be run with exec.  i.e., don't use make as the first
> step.  You can call make from a script if you want, I suppose so too
> could one call a script from make but it's not as pure and simplistic.

in other words, debian style, have a "build" executable.... even though 99%
of the time, the debian "executable" is a script that starts with

#!/bin/make

:-)

[ie: a makefile!]


From james at opencsw.org  Fri Jul 31 20:56:22 2009
From: james at opencsw.org (James Lee)
Date: Fri, 31 Jul 2009 18:56:22 GMT
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731184245.GC76412@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<20090731.18151000.1268557623@gyor.oxdrove.co.uk>
	<20090731184245.GC76412@bolthole.com>
Message-ID: <20090731.18562200.4124391153@gyor.oxdrove.co.uk>

On 31/07/09, 19:42:45, Philip Brown <phil at bolthole.com> wrote regarding Re:
[csw-maintainers] Buildbot:

> > Go further, just have a single entry point per project (not package) per
> > version that can be run with exec.  i.e., don't use make as the first
> > step.  You can call make from a script if you want, I suppose so too
> > could one call a script from make but it's not as pure and simplistic.

> in other words, debian style, have a "build" executable.... even though
> 99% of the time, the debian "executable" is a script that starts with

> #!/bin/make

> :-)

> [ie: a makefile!]

If that's what floats your boat.  At least you have been able to choose
which make.



James.

From maciej at opencsw.org  Fri Jul 31 21:05:52 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 20:05:52 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249062711-sup-8879@ntdws12.chass.utoronto.ca>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
Message-ID: <ce97543b0907311205q41cb041cj16e091dfdfa19b50@mail.gmail.com>

On Fri, Jul 31, 2009 at 7:07 PM, Ben Walton<bwalton at opencsw.org> wrote:
> 3. I'd then say both 'build' and 'package' should be individually
> ? callable steps. ?A buildbot may only want to test that the updated
> ? commit can still successfully build the package while a release
> ? tool may need to be able to take things right up to the package
> ? files.

It would be nice to have a common interface for building 64-bit and
32-bit ISAs in a single package. How is that handled in non-GAR
builds?

Maciej

From maciej at opencsw.org  Fri Jul 31 21:10:09 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 20:10:09 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6vdl8ivzn.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
	<x6vdl8ivzn.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907311210kf8840e7me6ff41de6313ffda@mail.gmail.com>

On Fri, Jul 31, 2009 at 6:49 PM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> [...]
>>
>> I've split the thread about the build systems into another one, and
>> I'd like to go back to the buildbot topic. Are there any comments
>> about the design or deployment plan?
>
> Split or not split, my comments are valid for this
> discussion. Consequently, please take into account this design
> requirements:
>
> 1. The buildbot deployment must be done on a separate hardware from that
> ? used for the regular, "manual", builds.

This requirement might be a showstopper here. Can we trade it for a
restricted set of packages to build, avoiding the large ones?

> 2. Other build systems that gar must be able to use the deployed
> ? "automatic" build infrastructure using buildbot.

The infrastructure will be actually automatic, not "automatic". ;-)

I've updated the wiki. Buildbot essentially needs to know what shell
command to run in the top package directory. I don't care what it is,
as long as there's a reliable discovery method of what should that
command be.

Maciej

From phil at bolthole.com  Fri Jul 31 22:05:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 13:05:32 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907311210kf8840e7me6ff41de6313ffda@mail.gmail.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
	<x6vdl8ivzn.fsf@abhavana.free.fr>
	<ce97543b0907311210kf8840e7me6ff41de6313ffda@mail.gmail.com>
Message-ID: <20090731200532.GD76412@bolthole.com>

On Fri, Jul 31, 2009 at 08:10:09PM +0100, Maciej (Matchek) Blizinski wrote:
> > 1. The buildbot deployment must be done on a separate hardware from that
> > ?? used for the regular, "manual", builds.
> 
> This requirement might be a showstopper here. Can we trade it for a
> restricted set of packages to build, avoiding the large ones?

what would that accomplish?

it would still have to be a large set, to be useful.
and large sets, take large amounts of time, to compile.


From mwatters at opencsw.org  Fri Jul 31 22:17:00 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 15:17:00 -0500
Subject: [csw-maintainers] HowTo question with mGAR
Message-ID: <4A73513C.5000107@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Can someone tell me how to include "extra" package admin files using mGAR.
I have 3 custom class scripts that I need to include in the package but can't
figure out how to force gar to include them.


- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzUTsACgkQLrhmsXMSLxc9KwCbBQUlThk2IFuP6cQ2ZcCDq9fU
MSgAoLEFtajQAjvCCS9edqOGMAxFfh34
=nizh
-----END PGP SIGNATURE-----

From dam at baltic-online.de  Wed Jul 29 20:05:49 2009
From: dam at baltic-online.de (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 20:05:49 +0200
Subject: [csw-maintainers] Statistics on ibiblio
Message-ID: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>

Hi there,

there are also statistics available for ibiblio:

1. Log in to <https://stats.ibiblio.org>
    User: guest Password: ibiblio
2. Click on www.ibiblio.org.public
3. Click on Pages & Files
4. Click on Directory Drilldown
5. Click on Directory by Files Drilldown
6. Click on /pub
7. Click on /packages
8. Click on /solaris

...enjoy :-)


Best regards

   -- Dago

From bwalton at opencsw.org  Fri Jul 31 23:13:49 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 17:13:49 -0400
Subject: [csw-maintainers] HowTo question with mGAR
In-Reply-To: <4A73513C.5000107@opencsw.org>
References: <4A73513C.5000107@opencsw.org>
Message-ID: <1249074745-sup-6871@ntdws12.chass.utoronto.ca>

Excerpts from Mike Watters's message of Fri Jul 31 16:17:00 -0400 2009:

> Can someone tell me how to include "extra" package admin files using mGAR.
> I have 3 custom class scripts that I need to include in the package but can't
> figure out how to force gar to include them.

Do you mean that you need the package to stuff them into
/usr/sadm/install/scripts from files/?

-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 bwalton at opencsw.org  Fri Jul 31 23:20:33 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 17:20:33 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731184108.GB76412@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
	<20090731184108.GB76412@bolthole.com>
Message-ID: <1249074967-sup-4066@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 31 14:41:08 -0400 2009:

> I think applying the patch should be part of "get source", and there should
> not be a mandated separate one.

Ok.  For a 'broad strokes' approach, that would be acceptable, imo.
(This implies that the 'fetch' step would also encompass 'extract'
which is fine.)

> > 3. I'd then say both 'build' and 'package' should be individually
> >    callable steps....
> > 4. A 'clean' target is always a good thing to support.
> > 5. A set of variables should be honoured by any Makefile such that a
> >    build system knows where to find outputs.

> I think this is making it too complicated.
> It needs to be *simple*,for most widespread use.

Ok, as long as there is some sort of convention where automated files
could expect to find their outputs.  The Variable approach lets the
tool set that as it desires, which makes changing it somewhat easier,
but as long as good choices were made up front...

Are you suggesting that a build step separate from package is also too
much though?  I think that is an important distinction, but arguments
to the contrary are 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.

From phil at bolthole.com  Fri Jul 31 23:55:13 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 14:55:13 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249074967-sup-4066@ntdws12.chass.utoronto.ca>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
	<20090731184108.GB76412@bolthole.com>
	<1249074967-sup-4066@ntdws12.chass.utoronto.ca>
Message-ID: <20090731215513.GF76412@bolthole.com>

On Fri, Jul 31, 2009 at 05:20:33PM -0400, Ben Walton wrote:
> > I think this is making it too complicated.
> > It needs to be *simple*,for most widespread use.
> 
> Ok, as long as there is some sort of convention where automated files
> could expect to find their outputs.

There is no need for that. I think that's overengineering.
There is only one set of "outputs" actually *needed* for the simplified
API,and that is "where do the packages go?"

Similarly, i think that there is no need for variables like "where do I do
builds",because the reasonable assumption would be "in the current
directory".

> Are you suggesting that a build step separate from package is also too
> much though?  I think that is an important distinction, but arguments
> to the contrary are welcome.

I wasnt suggesting that originally. But I think perhaps having two steps
there, is also unneccessary.
Even in the case where there is some need to tweak something silly like the
pkginfo file... a good "build system" should run the (compilephase)build,
but then automatically detect there is nothing that needs recompiling. 
So putting together the new package, with the updated pkginfo file, should
be relatively speedy.


From william at wbonnet.net  Wed Jul  1 00:07:29 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Jul 2009 00:07:29 +0200
Subject: [csw-maintainers] New in testing: gtk2 2.16.1
In-Reply-To: <4A4A595A.3010304@opencsw.org>
References: <A1A966C3-001F-4777-8B3C-0830A5735370@opencsw.org>	<4A4A3B04.2040701@opencsw.org>	<4A4A5204.7070705@opencsw.org>
	<4A4A595A.3010304@opencsw.org>
Message-ID: <4A4A8CA1.7090605@wbonnet.net>

Hi

> While playing with pygtk, I found a "nicety" for the package install.
> have a post install that runs:
>
> /opt/csw/bin/gdk-pixbuf-query-loaders >/opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders
>
> I am going to file a feature bug against the pango package to request the same
> for /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules
>   
The postinstall script is under svn. Package has to be rebuild to 
include 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 pfelecan at opencsw.org  Wed Jul  1 09:48:35 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Wed, 01 Jul 2009 09:48:35 +0200
Subject: [csw-maintainers] tcsh 6.16.00 in testing
In-Reply-To: <15954.131.203.105.1.1246399049.squirrel@webmail.jessies.org>
	(Chris Reece's message of "Wed\,
	1 Jul 2009 09\:57\:29 +1200 \(NZST\)")
References: <15954.131.203.105.1.1246399049.squirrel@webmail.jessies.org>
Message-ID: <x663ec3kto.fsf@abhavana.free.fr>

"Chris Reece" <car at opencsw.org> writes:

> On Tue, June 30, 2009 11:36 pm, Peter Bonivart wrote:
>> It was years ago I used tcsh but it installs/removes cleanly and seems
>> to work properly during my short test on a Solaris 9 Sparc and a Solaris 10
>> x86. Man page also works.
>
> Thanks, Peter, much obliged.
>
> My only remaining problem is that I can't log in to www.opencsw.org, which the
> documentation suggests is the location of /home/newpkgs.  None of my keys are
> accepted.  Or, rather, there are "Too many authentication failures" after all
> of my keys have been tried.
>
> I can log in to login.opencsw.org, and from there to build8s and build8x.

>From login.opencsw.org you can scp your packages to
www.opencsw.org:/home/newpkgs
-- 
Peter


From william at wbonnet.net  Wed Jul  1 14:32:00 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Jul 2009 14:32:00 +0200
Subject: [csw-maintainers] Conflicting header file
Message-ID: <4A4B5740.90407@wbonnet.net>

Hi

A conflicting header is installed by qt package. It is the file 
"/opt/csw/include/npapi.h"

When including this file, it generates the following error

"/opt/csw/include/npapi.h", line 1: Error: #error 
"/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
must be provided by Netscape".

A mantis entry exists, but unfortnatly package has no maintainer and was 
not fixed. Would it be possible to nuke this file on the build farm 
please ? (well at least rename the file to .old...). This file is 
blocking for compile FF3.

npapi.h is provided by mozilla packages, especialy firefox, not by qt.

[wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$ grep 
npapi.h /var/sadm/install/contents
/opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908 CSWqt
/opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273 
1110647274 CSWqtgcc

Thanks in advance

cheers
W.




From dam at opencsw.org  Wed Jul  1 14:55:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Jul 2009 14:55:34 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A4B5740.90407@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
Message-ID: <595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>

Hi William,

Am 01.07.2009 um 14:32 schrieb William Bonnet:
> A conflicting header is installed by qt package. It is the file "/ 
> opt/csw/include/npapi.h"
>
> When including this file, it generates the following error
>
> "/opt/csw/include/npapi.h", line 1: Error: #error "/home/espenr/tmp/ 
> qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h must be  
> provided by Netscape".
>
> A mantis entry exists, but unfortnatly package has no maintainer and  
> was not fixed. Would it be possible to nuke this file on the build  
> farm please ? (well at least rename the file to .old...). This file  
> is blocking for compile FF3.
>
> npapi.h is provided by mozilla packages, especialy firefox, not by qt.
>
> [wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$  
> grep npapi.h /var/sadm/install/contents
> /opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908  
> CSWqt
> /opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273  
> 1110647274 CSWqtgcc

The file does not include anything of interest:

dam at build9x :/home/dam > more /opt/csw/include/npapi.h
#error "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/ 
npapi.h must be provided by Netscape"
#error "This file is just a placeholder. Please see the documentation"
#error "to learn how to obtain the real file"

Moved to npapi.h.old on build9x only for now.


Best regards

   -- Dago


From dam at opencsw.org  Wed Jul  1 15:02:58 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Jul 2009 15:02:58 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A4B5740.90407@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
Message-ID: <6798D88D-CFE6-4A1F-872A-3082CD490124@opencsw.org>

Hi William,

Am 01.07.2009 um 14:32 schrieb William Bonnet:
> A conflicting header is installed by qt package. It is the file "/ 
> opt/csw/include/npapi.h"
>
> When including this file, it generates the following error
>
> "/opt/csw/include/npapi.h", line 1: Error: #error "/home/espenr/tmp/ 
> qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h must be  
> provided by Netscape".
>
> A mantis entry exists, but unfortnatly package has no maintainer and  
> was not fixed. Would it be possible to nuke this file on the build  
> farm please ? (well at least rename the file to .old...). This file  
> is blocking for compile FF3.
>
> npapi.h is provided by mozilla packages, especialy firefox, not by qt.
>
> [wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$  
> grep npapi.h /var/sadm/install/contents
> /opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908  
> CSWqt
> /opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273  
> 1110647274 CSWqtgcc

The file does not include anything of interest:

dam at build9x :/home/dam > more /opt/csw/include/npapi.h
#error "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/ 
npapi.h must be provided by Netscape"
#error "This file is just a placeholder. Please see the documentation"
#error "to learn how to obtain the real file"

Moved to npapi.h.old on build9x only for now.


Best regards

  -- Dago


From william at wbonnet.net  Wed Jul  1 16:07:58 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 01 Jul 2009 16:07:58 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
Message-ID: <4A4B6DBE.5030606@wbonnet.net>

Hi
>
> The file does not include anything of interest:
>
> dam at build9x :/home/dam > more /opt/csw/include/npapi.h
> #error 
> "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
> must be provided by Netscape"
> #error "This file is just a placeholder. Please see the documentation"
> #error "to learn how to obtain the real file"
>
> Moved to npapi.h.old on build9x only for now.
>
Thanks. If it works on this box, i may ask for more after ;)

cheers
W.


From skayser at opencsw.org  Wed Jul  1 22:08:56 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 1 Jul 2009 22:08:56 +0200 (CEST)
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
 instead of PROTOTYPE_FILTER
Message-ID: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>

Hi,

while working on postfix 2.6 i have noticed again that it is not quite
straight-forward to assign files to classes. Instead of exposing the inner
workings of the prototype stuff to the Makefile, could we maybe get some
variables to assign files to classes (functions)?

So instead of having:

CONFIG_BASE = \/etc\/opt\/csw\/postfix\/
CONFIG_FILES = access aliases canonical generic header_checks
CONFIG_FILES += main.cf master.cf
CONFIG_FILES += relocated transport virtual

PROTOTYPE_FILTER  = awk '\
	$(foreach C,$(prefix $(CONFIG_BASE),$(CONFIG_FILES)), \
		$$$$3 ~ /^$(C)$$$$/ { $$$$2 = "cswcpsampleconf" }) \
        ....

just have something like:

CONFIG_FILES = /etc/default/cswpostgrey
CONFIG_FILES += /etc/opt/csw/postfix/access
CONFIG_FILES += /etc/opt/csw/postfix/aliases
CONFIG_FILES += /etc/opt/csw/postfix/canonical
CONFIG_FILES += /etc/opt/csw/postfix/generic
CONFIG_FILES += /etc/opt/csw/postfix/header_checks
CONFIG_FILES += /etc/opt/csw/postfix/main.cf
CONFIG_FILES += /etc/opt/csw/postfix/master.cf
CONFIG_FILES += /etc/opt/csw/postfix/relocated
CONFIG_FILES += /etc/opt/csw/postfix/transport
CONFIG_FILES += /etc/opt/csw/postfix/virtual

To me this would feel more declarative instead of the procedural way we do
it with PROTOTYE_FILTER right now (note, it can be shortened with
$(foreach and $(prefix, just like above). It would not only tag the class,
but also wrap the other housekeeping, like moving the config files to .CSW
during the build process.

(Side note: Thinking ahead, it could also be an conceptual abstraction
level WRT to IPS: Do IPS packages still have a prototype file?)

This is only an example for config files, which i encounter most often.
Other classes could be "wrapped" as well. One would just have to come up
with the requirements, a proper naming scheme for the variables, and ....
actually implement it in GAR.

Sebastian



From dam at opencsw.org  Wed Jul  1 22:21:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 1 Jul 2009 22:21:07 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
Message-ID: <3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>

Hi Sebastian,

Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
> while working on postfix 2.6 i have noticed again that it is not quite
> straight-forward to assign files to classes. Instead of exposing the  
> inner
> workings of the prototype stuff to the Makefile, could we maybe get  
> some
> variables to assign files to classes (functions)?
>
> So instead of having:
>
> CONFIG_BASE = \/etc\/opt\/csw\/postfix\/
> CONFIG_FILES = access aliases canonical generic header_checks
> CONFIG_FILES += main.cf master.cf
> CONFIG_FILES += relocated transport virtual
>
> PROTOTYPE_FILTER  = awk '\
> 	$(foreach C,$(prefix $(CONFIG_BASE),$(CONFIG_FILES)), \
> 		$$$$3 ~ /^$(C)$$$$/ { $$$$2 = "cswcpsampleconf" }) \
>        ....
>
> just have something like:
>
> CONFIG_FILES = /etc/default/cswpostgrey
> CONFIG_FILES += /etc/opt/csw/postfix/access
> CONFIG_FILES += /etc/opt/csw/postfix/aliases
> CONFIG_FILES += /etc/opt/csw/postfix/canonical
> CONFIG_FILES += /etc/opt/csw/postfix/generic
> CONFIG_FILES += /etc/opt/csw/postfix/header_checks
> CONFIG_FILES += /etc/opt/csw/postfix/main.cf
> CONFIG_FILES += /etc/opt/csw/postfix/master.cf
> CONFIG_FILES += /etc/opt/csw/postfix/relocated
> CONFIG_FILES += /etc/opt/csw/postfix/transport
> CONFIG_FILES += /etc/opt/csw/postfix/virtual
>
> To me this would feel more declarative instead of the procedural way  
> we do
> it with PROTOTYE_FILTER right now (note, it can be shortened with
> $(foreach and $(prefix, just like above). It would not only tag the  
> class,
> but also wrap the other housekeeping, like moving the config files  
> to .CSW
> during the build process.
>
> (Side note: Thinking ahead, it could also be an conceptual abstraction
> level WRT to IPS: Do IPS packages still have a prototype file?)
>
> This is only an example for config files, which i encounter most  
> often.
> Other classes could be "wrapped" as well. One would just have to  
> come up
> with the requirements, a proper naming scheme for the variables,  
> and ....
> actually implement it in GAR.

For CSWcswclassutils this is already in there :-) Just define

SAMPLECONF
PRESERVECONF
INITSMF

and everything else will be taken cared of.


Bestregards

   -- Dago


From skayser at opencsw.org  Thu Jul  2 00:05:23 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 2 Jul 2009 00:05:23 +0200 (CEST)
Subject: [csw-maintainers] Heads up: GAR documentation main page content
 simplified (gar.sf.net)
Message-ID: <55969.85.88.7.146.1246485923.squirrel@ssl.skayser.de>

Hi,

to better guide GAR novices i have just simplified the main GAR
documentation page [1]. What changed?

- The advanced topics have been moved to a separate page [2] and linked to
  from the main site.

- The "revision needed" and "scratchpad style" pages have been moved to
  a separate page [3] that is accessible via the index (upper right).

This is only an attempt of how it could look like and was born because i
have seen a couple of GAR beginners struggle with the "too many"
documentation pages. The main page now doesn't throw everything in their
way, but tries to get them going step by step.

I assume, advanced GAR users are likely to still find what they are
looking for (and there is still the index to search across all page
topics). Feedback, thoughts, and ideas welcome.

Sebastian

[1] http://gar.sf.net, http://sourceforge.net/apps/trac/gar/
[2] http://sourceforge.net/apps/trac/gar/wiki/Learning%20the%20details
[3] http://sourceforge.net/apps/trac/gar/wiki/AA_Notepad



From trygvis at opencsw.org  Thu Jul  2 10:00:03 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 02 Jul 2009 10:00:03 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
Message-ID: <4A4C6903.2040906@opencsw.org>

If someone feel like getting this mostly python package out, feel free.

I almost got 1.2 working but there where too many test cases that 
failed. Most of the failures was because of missing gnu'isms in the 
tools the test used from what I could tell. Would be nice to patch up 
the test suite to be able to control the path to the tools it uses.

--
Trygve
-------------- next part --------------
An embedded message was scrubbed...
From: Upstream Package Watch <uwatch at mirror.opencsw.org>
Subject: [svn] mercurial upstream update notification
Date: Thu, 2 Jul 2009 02:44:21 +0200 (CEST)
Size: 2306
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/aa41526a/attachment-0002.eml>

From schwindt at dfki.uni-kl.de  Thu Jul  2 12:18:55 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Thu, 02 Jul 2009 12:18:55 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
Message-ID: <200907021018.n62AIucZ003707@dfki.uni-kl.de>


Starting with libpango-1.24.3,REV=2009.06.27 the content of i.e. 
windowtitles get garbled. Using xfce4 as a windowmanager gives
you only empty squares where gtk normaly would put text.

This happens on a system with a plain csw-install, but also 
on a test machine where I upgrade some packages by hand like libconv-1.13,
xfce4.6.

Am I doing something wrong ? Or is this fixed by upgrading to versions from 
testing ?


Nicolai





From william at wbonnet.net  Thu Jul  2 12:24:49 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 02 Jul 2009 12:24:49 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: <200907021018.n62AIucZ003707@dfki.uni-kl.de>
References: <200907021018.n62AIucZ003707@dfki.uni-kl.de>
Message-ID: <4A4C8AF1.6030508@wbonnet.net>

Hi Nicolai

Does the following file exist ?

/opt/csw/etc/pango/pango.modules

if not please run

/opt/csw/bin/pango-querymodules > /opt/csw/etc/pango/pango.modules

This issue is under fixing...

chers
W.
> Starting with libpango-1.24.3,REV=2009.06.27 the content of i.e. 
> windowtitles get garbled. Using xfce4 as a windowmanager gives
> you only empty squares where gtk normaly would put text.
>
> This happens on a system with a plain csw-install, but also 
> on a test machine where I upgrade some packages by hand like libconv-1.13,
> xfce4.6.
>
> Am I doing something wrong ? Or is this fixed by upgrading to versions from 
> testing ?
>
>
> Nicolai
>
>
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>   



From bwalton at opencsw.org  Thu Jul  2 19:05:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 13:05:01 -0400
Subject: [csw-maintainers] package hooks (update)
Message-ID: <1246554093-sup-189@ntdws12.chass.utoronto.ca>


Hi All,

I've tweaked the package hooks rfc slightly and would appreciate any
feedback.  Changes are:

1. Use /var/tmp for file storage of package lists.  I think storing
   temp data in /var/opt/csw/pkg-hooks was wrong, since it's not long
   lived.
2. Renamed and expanded the list of proposed hooks.  I think the new
   scheme adds better clarity of purpose to the names of the hooks.

http://wiki.opencsw.org/package-hooks

If this is acceptable, I'm going to go ahead and begin the
implementation.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/66fb6e85/attachment-0002.asc>

From phil at bolthole.com  Thu Jul  2 19:46:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 10:46:57 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
Message-ID: <20090702174657.GJ84645@bolthole.com>

On Thu, Jul 02, 2009 at 01:05:01PM -0400, Ben Walton wrote:
> 
> Hi All,
> 
> I've tweaked the package hooks rfc slightly and would appreciate any
> feedback.  Changes are:
> 
> 1. Use /var/tmp for file storage of package lists.  I think storing
>    temp data in /var/opt/csw/pkg-hooks was wrong, since it's not long
>    lived.


package lists?



From phil at bolthole.com  Thu Jul  2 19:51:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 10:51:32 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
Message-ID: <20090702175132.GK84645@bolthole.com>

Also...

On Thu, Jul 02, 2009 at 01:05:01PM -0400, Ben Walton wrote:
> http://wiki.opencsw.org/package-hooks

I dont see where it explicitly calls how how a package tool can
differentiate between "batch" and "per-package" hooks.

also, the communication between package tools and hooks, seems to be rather
vague.
An rfc normally deals with fairly concrete directives. and even when it is
vague, it is explicitly vague. that is to say, "must" vs "may" :-)

	
What I mean is, I think it would be beneficial to have the "communication"
section rewritten along the lines of,

[in situation X, the package tool should pass in argument Y]

Right now, it seems like package tools "may" pass in arguments to hooks,
basically on a whim. there is no rhyme or reason to it that i can see.



From mwatters at opencsw.org  Thu Jul  2 20:21:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 02 Jul 2009 13:21:50 -0500
Subject: [csw-maintainers] mGar RFE *** new make target reset-source ***
Message-ID: <4A4CFABE.3040801@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dago,

could we get an option to reset the package back to post-extract status.

it should be fairly easy although, I am not sure how to deal with the
patches...  I think they make a backup of the original.

will do some more research on how the patches are done and how to revert them,
unless you have it figured 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

iEYEARECAAYFAkpM+r4ACgkQLrhmsXMSLxfiGgCgpr6bGcKYY1VvkGcAfHw963bQ
hmQAoIKGEw1kEtgde1Y9a7Hh0pj9OqTT
=aUG2
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Jul  2 20:43:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 2 Jul 2009 20:43:46 +0200
Subject: [csw-maintainers] mGar RFE *** new make target reset-source ***
In-Reply-To: <4A4CFABE.3040801@opencsw.org>
References: <4A4CFABE.3040801@opencsw.org>
Message-ID: <38A01277-81E2-4B9E-B546-9C0E1E6B625B@opencsw.org>

Hi Mike,

Am 02.07.2009 um 20:21 schrieb Mike Watters:
> could we get an option to reset the package back to post-extract  
> status.

 From where? patch, configure or build?

> it should be fairly easy although, I am not sure how to deal with the
> patches...  I think they make a backup of the original.
>
> will do some more research on how the patches are done and how to  
> revert them,
> unless you have it figured out.   ;-)

The patching-thing is not solved well currently in GAR. I am thinking
of something like
   gmake preparepatch
which unpacks the sources multiple times like
   work/pristine-<mod>   Pristine, unpatched sources
   work/patched-<mod>    Patched sources
   work/build-<mod>      Patches sources that will be worked on
pristine-* will just be unpacked and always stay that way. As maintainer
you than start editing files in patched-*. gmake makepatch generates
a patch from the diffs between pristine-* and patched-*. After including
the patch in PATCHFILES you can do
   gmake repreparepatch
and pristine-* will contain the sources with the patches from PATCHFILES
applied and patched-* a copy from that to generate more patches.

Feel free to enhance this draft.


Best regards

   -- Dagp


From bwalton at opencsw.org  Thu Jul  2 20:59:28 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 14:59:28 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702175132.GK84645@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
Message-ID: <1246560996-sup-1808@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 13:51:32 -0400 2009:
> > http://wiki.opencsw.org/package-hooks

> I dont see where it explicitly calls how how a package tool can
> differentiate between "batch" and "per-package" hooks.

Well, if you read the description of the two groups of hooks, you
should see that it notes:

<snip>
Batch Hooks:

These hooks are run before or after a batch package operation. In this
sense, batch is taken to mean all package actions, regardless of the
number of packages affected.
</snip>

<snip>
Per-package Hooks:

These hooks are run before and after individual calls to pkgadd or
pkgrm.
</snip>

I've added some examples of the call sequences at the bottom of the
document in a 'Hook Invocation' section to solidify these
descriptions.

> also, the communication between package tools and hooks, seems to be rather
> vague.
> An rfc normally deals with fairly concrete directives. and even when it is
> vague, it is explicitly vague. that is to say, "must" vs "may" :-)

For better or worse.

> What I mean is, I think it would be beneficial to have the "communication"
> section rewritten along the lines of,
> [in situation X, the package tool should pass in argument Y]

Done.  See changes.

> Right now, it seems like package tools "may" pass in arguments to hooks,
> basically on a whim. there is no rhyme or reason to it that i can see.

Well, I thought it was clear, but I wrote it, so I knew what I meant!
:)

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/fae34ddb/attachment-0002.asc>

From phil at bolthole.com  Thu Jul  2 21:05:30 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 12:05:30 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246560996-sup-1808@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
Message-ID: <20090702190530.GE94553@bolthole.com>

On Thu, Jul 02, 2009 at 02:59:28PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Thu Jul 02 13:51:32 -0400 2009:
> > > http://wiki.opencsw.org/package-hooks
> 
> > I dont see where it explicitly calls how how a package tool can
> > differentiate between "batch" and "per-package" hooks.
> ....
> Well, if you read the description of the two groups of hooks, you
> should see that it notes:
> 

Yes, it describes what [a batch hook] should do.
However, it does not explicitly describe how a package installation tool
should RECOGNIZE a batch hook, AS a batch hook, that i could see.

If the tool does not know which scripts are batch hooks, then it wont know
_what_ to call, even if it knows _when_ to call them! :-D


From bwalton at opencsw.org  Thu Jul  2 21:27:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 15:27:15 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702190530.GE94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
Message-ID: <1246562604-sup-5118@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 15:05:30 -0400 2009:
> Yes, it describes what [a batch hook] should do.
> However, it does not explicitly describe how a package installation tool
> should RECOGNIZE a batch hook, AS a batch hook, that i could see.

The tool doesn't need to _recognize_ anything.  The hook names are
defined.  Running a hook consists of testing for existence of a
specific directory (based on the name of the hook) and executing in
lexical order any files in that directory.  If the directory isn't
there, the tool continues.  So, since hooks have names, they're
'addressed' by name.  No need to recognize them.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/99b9fb85/attachment-0002.asc>

From phil at bolthole.com  Thu Jul  2 21:39:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 12:39:50 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246562604-sup-5118@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
Message-ID: <20090702193950.GH94553@bolthole.com>

On Thu, Jul 02, 2009 at 03:27:15PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Thu Jul 02 15:05:30 -0400 2009:
> > Yes, it describes what [a batch hook] should do.
> > However, it does not explicitly describe how a package installation tool
> > should RECOGNIZE a batch hook, AS a batch hook, that i could see.
> 
> The tool doesn't need to _recognize_ anything.  The hook names are
> defined.  Running a hook consists of testing for existence of a
> specific directory (based on the name of the hook) and executing in
> lexical order any files in that directory.  If the directory isn't
> there, the tool continues.  So, since hooks have names, they're
> 'addressed' by name.  No need to recognize them.

Sigh. failure to communicate here.

http://wiki.opencsw.org/package-hooks


The general format of a hook script location, would seem to be

/etc/opt/csw/pkg-hooks/DIRECTORY/script-goes-here

but you have not explicitly defined the acceptible values for DIRECTORY.

You've given a few examples. But you have not explicitly mapped out the
full space.

you have not defined which DIRECTORY names should be used for per-package,
vs "batch" hooks, for example.





From william at wbonnet.net  Thu Jul  2 21:42:05 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 02 Jul 2009 21:42:05 +0200
Subject: [csw-maintainers] mGar RFE *** new make target reset-source ***
In-Reply-To: <38A01277-81E2-4B9E-B546-9C0E1E6B625B@opencsw.org>
References: <4A4CFABE.3040801@opencsw.org>
	<38A01277-81E2-4B9E-B546-9C0E1E6B625B@opencsw.org>
Message-ID: <4A4D0D8D.8050802@wbonnet.net>

Hi
> From where? patch, configure or build?
 From whatever :)

My personal need is the following :

. Let say I do work on a package with a large amount of sources (For 
example Firefox... ;) )
. I run a compile then choose to change some options in the Makefile 
used by GAR
. I would like to do a "make clean and reconfigure" with the lowest 
overhead possible.

Actually this process may take some time to make clean, download the 
sources again, and extract sources. It may be a question of hour for 
Firefox, Thunderbird or Eclipse. So any solution that may help to reduce 
this time would be great.

Most of the time i need this after a build, successful or not.
> which unpacks the sources multiple times like
>   work/pristine-<mod>   Pristine, unpatched sources
>   work/patched-<mod>    Patched sources
>   work/build-<mod>      Patches sources that will be worked on
I do really like this !

Many times i have been making cp foo.c foo.c.orig in my working folder.

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  Thu Jul  2 22:02:56 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 16:02:56 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702193950.GH94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
Message-ID: <1246563839-sup-2157@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 15:39:50 -0400 2009:
> Sigh. failure to communicate here.

I think so.

> http://wiki.opencsw.org/package-hooks

> The general format of a hook script location, would seem to be
> /etc/opt/csw/pkg-hooks/DIRECTORY/script-goes-here
> but you have not explicitly defined the acceptible values for
> DIRECTORY.

> You've given a few examples. But you have not explicitly mapped out the
> full space.

Not in terms of a specific list, no, but I have given the hook names,
noted that there is a pre and post for each hook and specified the
name format for the directories.

...Ok, the wiki now contains a _complete_, _explicit_ list of valid
hooks and the corresponding directories.

> you have not defined which DIRECTORY names should be used for per-package,
> vs "batch" hooks, for example.

Again, the grouping isn't important in any way other than the type of
argument passed.  Batch hooks get a file name, per-package hooks get a
package name.  Other than that, they all live happily in
/etc/opt/csw/pkg-hooks.  For reference, in case it's still not clear,
these directories would be:

/etc/opt/csw/pkg-hooks/prebatchadd.d/
/etc/opt/csw/pkg-hooks/postbatchadd.d/
/etc/opt/csw/pkg-hooks/prebatchremove.d/
/etc/opt/csw/pkg-hooks/postbatchremove.d/
/etc/opt/csw/pkg-hooks/preinstall.d/
/etc/opt/csw/pkg-hooks/postinstall.d/
/etc/opt/csw/pkg-hooks/preupdate .d/
/etc/opt/csw/pkg-hooks/postupdate.d/
/etc/opt/csw/pkg-hooks/preremove.d/
/etc/opt/csw/pkg-hooks/postremove.d/

I have also noted on the wiki that the only difference between batch
and per-package is the argument type.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/5d2c37d4/attachment-0002.asc>

From phil at bolthole.com  Fri Jul  3 00:04:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 15:04:45 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246563839-sup-2157@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
Message-ID: <20090702220445.GI94553@bolthole.com>

On Thu, Jul 02, 2009 at 04:02:56PM -0400, Ben Walton wrote:
> > http://wiki.opencsw.org/package-hooks
> 
> ...Ok, the wiki now contains a _complete_, _explicit_ list of valid
> hooks and the corresponding directories.

thank you.

more comments:

Dont depend on files that are living raw in /var/tmp.
It's a security hole.
I dont think you should be using files to pass around arguments at all.
particularly something as trivial as "a list of affected packages".
I think you shoudl have the list of packages, just be passed as the
argument list.

Additional comment:
you need to explicitly define behaviour of install vs update.
that is to say, you need to specify whether update hooks will be called
IN ADDITION TO, or INSTEAD OF, install and/or remove hooks.
hay waitaminute, you dont have remove hooks.

well actually you mention them in one section, but not in another section.
Missing from the "Proposed Per-package Hooks" section



From bwalton at opencsw.org  Fri Jul  3 02:43:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 02 Jul 2009 20:43:36 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702220445.GI94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
Message-ID: <1246577961-sup-2700@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 18:04:45 -0400 2009:

> Dont depend on files that are living raw in /var/tmp.
> It's a security hole.

Yes, I thought of this on the train ride home.

> I dont think you should be using files to pass around arguments at
> all.  particularly something as trivial as "a list of affected
> packages".  I think you shoudl have the list of packages, just be
> passed as the argument list.

Ok.  I was thinking along the lines of overflowing ARG_MAX, but upon
inspection Solaris (32/64 bit -> 1048320/2096640) uses a huge value
compared to Linux (131072).  This is no longer a concern for me.

I do wonder though whether passing all packages as arguments is the
way to go.  If it was decided to add more information later (version
numbers or something), using the argument list would be more limiting
than a file-based approach.  Would passing lines on stdin be a better
method?

The spec could define: each line of stdin should be word split on
whitespace (to match the catalog delimitter).  The first word is the
package being acted on.  Hook scripts should ignore anything beyond
the first word.

This would be more flexible in terms of future expansion.  The same
could be done with command line arguments, but I think stdin works
better for this.

[No changes to the doc have been made in this area yet.]

> Additional comment:
> you need to explicitly define behaviour of install vs update.
> that is to say, you need to specify whether update hooks will be called
> IN ADDITION TO, or INSTEAD OF, install and/or remove hooks.

<snip>
Proposed Per-package Hooks:

install

This hook should bracket any action that will involve a pkgadd without
a preceeding pkgrm.

update

This hook should bracket any action that will involve a pkgrm followed
immediately by a pkgadd for the same package.
</snip>

That's the existing text (with s/update/upgrade/).  Not clear enough?
I understand what you're asking, but in my mind those definitions are
clear.  If I'm adding a package that doesn't exist (pkgadd without a
preceeding pkgrm), the install hook gets called.  If I'm adding a
package and I already have a current version (meaning I do pkgrm then
pkgadd), I do update (which I've renamed upgrade, since I feel that's
a better term than update).

There is no distinction between batchinstall or batchupgrade if that's
what you were asking.  We could note the difference by adding a
batchupgrade hook if you think that's worthwhile.  I'm not sure it is,
but it wouldn't necessarily hurt either.

> hay waitaminute, you dont have remove hooks.

> well actually you mention them in one section, but not in another section.
> Missing from the "Proposed Per-package Hooks" section

See purge (per-package) and batchremove (batch).  I used purge since
that's the term Dago proposed in his mail the other day.  Now that
remove has become batchremove, purge could be remove...Actually, I've
now made this change, since I think it is better language.  [I had
also incorrectly listed pre/post remove instead of pre/post purge in
one section, so that may be part of this problem/comment.]

Getting closer...

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090702/2c23d734/attachment-0002.asc>

From phil at bolthole.com  Fri Jul  3 03:03:47 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 2 Jul 2009 18:03:47 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246577961-sup-2700@ntdws12.chass.utoronto.ca>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
Message-ID: <20090703010347.GK94553@bolthole.com>

On Thu, Jul 02, 2009 at 08:43:36PM -0400, Ben Walton wrote:
[http://wiki.opencsw.org/package-hooks]
> <snip>
> Proposed Per-package Hooks:
> 
> install
> ...

> That's the existing text (with s/update/upgrade/).  Not clear enough?

no it's not clear enough :) you need to be explicit, and state
whether or not a "upgrade" hook overlaps install/remove hooks, or whether
it gets called instead of them.
You also have to explicitly define the order of calling, if they both
get run

[hmm actually, you sort of say this in your examples. but you make
 some mistakes in your examples :) for example, you mix stating
  that something is being "installed" vs being "upgraded", in the same
   sample paragraph]

It's best to state explicitly in the specification, the 'correct'
behaviour.
Examples are suposed to explain the specification, not be instead of
the specification.




> There is no distinction between batchinstall or batchupgrade if that's
> what you were asking.  We could note the difference by adding a
> batchupgrade hook if you think that's worthwhile.  I'm not sure it is,
> but it wouldn't necessarily hurt either.

If you think it's neccessary to have a per-package upgrade, i think you
should provide for a batch upgrade, just for completeness. even if you cant
think of a use for it now, someone later might, if you provide a hook for
it.

> See purge (per-package) and batchremove (batch).  I used purge since
> that's the term Dago proposed in his mail the other day.

Please note: "purge" is DIFFERENT from "remove".

It has a long-understood meaning in the debian world, and possibly others.

"remove" means "run pkgrm".
"purge" should mean "run pkgrm, and REMOVE ANY LEFTOVERS"
  (in other words, remove any capability/saved state for a delayed upgrade via
    pkgadd later)



From schwindt at dfki.uni-kl.de  Fri Jul  3 10:40:37 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 03 Jul 2009 10:40:37 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: Your message of "Thu, 02 Jul 2009 12:24:49 +0200."
	<4A4C8AF1.6030508@wbonnet.net> 
Message-ID: <200907030840.n638eboe019606@dfki.uni-kl.de>

> Hi Nicolai
> 
> Does the following file exist ?
> 
> /opt/csw/etc/pango/pango.modules

Not existsing
 
> if not please run
> 
> /opt/csw/bin/pango-querymodules > /opt/csw/etc/pango/pango.modules

Did that - file was created successfull - did not change anything.
 
> This issue is under fixing...

The problem is : I am running on an U24 which is 64-bit this triggers
pango-querymodules to generate a list of the 64-bit modules.
Starting xfce which is only available as 32-bit loads libpango in 32-bit.
This one falls flat on the face.


It seems there is the need for separate pangorc files containing the correct
definition of ModulesFiles and AliasFiles.

This implies each and every other package that uses pango must have both
32 and 64 bit versions. The other solution was having a wrapper for each
executable setting the correct environment.

I did not look into the source of pango for a while, is there any chance
of patching pango to hardcode PANGO_RC_FILE=/opt/csw/etc/pangorc{32,64} into
it ?

BTW: afaik pixbuf-loader is the same pain in the arse
 


Nicolai




From dam at opencsw.org  Fri Jul  3 13:42:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 3 Jul 2009 13:42:45 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: <200907030840.n638eboe019606@dfki.uni-kl.de>
References: <200907030840.n638eboe019606@dfki.uni-kl.de>
Message-ID: <FAF013CB-5FCF-4559-9891-6025FFF564A2@opencsw.org>

Hi Nicolai,

Am 03.07.2009 um 10:40 schrieb Nicolai Schwindt:
>> if not please run
>>
>> /opt/csw/bin/pango-querymodules > /opt/csw/etc/pango/pango.modules
>
> Did that - file was created successfull - did not change anything.
>
>> This issue is under fixing...
>
> The problem is : I am running on an U24 which is 64-bit this triggers
> pango-querymodules to generate a list of the 64-bit modules.
> Starting xfce which is only available as 32-bit loads libpango in 32- 
> bit.
> This one falls flat on the face.

Ok, there is obiously an error in the package creation:
   /opt/csw/bin/pango-querymodules
is linked to isaexec and chooses the best ISA. This is wrong for this
kind of use. The isaexec should be removed and there should be two
binaries:
   /opt/csw/bin/pango-querymodules for 32 bit
   /opt/csw/bin/(sparcv9|amd64)/pango-querymodules for 64 bit
You can try this right now with
   rm /opt/csw/bin/pango-querymodules
   ln -s /opt/csw/bin/sparcv8/pango-querymodules /opt/csw/bin/pango- 
querymodules
Please tell me if that worked.

> It seems there is the need for separate pangorc files containing the  
> correct
> definition of ModulesFiles and AliasFiles.

We may need /opt/csw/etc/64/pango for this

> This implies each and every other package that uses pango must have  
> both
> 32 and 64 bit versions.

Why is that? A 32 bit app needs to pull in 32 bit libs and 64 bit apps
pull in 64 bit libs, but why should both ISAs be needed?

> The other solution was having a wrapper for each
> executable setting the correct environment.
>
> I did not look into the source of pango for a while, is there any  
> chance
> of patching pango to hardcode PANGO_RC_FILE=/opt/csw/etc/ 
> pangorc{32,64} into
> it ?

Is this really necessary on a per-app basis? Can't this be patched
into pango itself?

> BTW: afaik pixbuf-loader is the same pain in the arse

William, any advice on this?


Best regards

   -- Dago


From bwalton at opencsw.org  Fri Jul  3 15:15:04 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 03 Jul 2009 09:15:04 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090703010347.GK94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
	<20090703010347.GK94553@bolthole.com>
Message-ID: <1246626185-sup-8972@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 21:03:47 -0400 2009:
> On Thu, Jul 02, 2009 at 08:43:36PM -0400, Ben Walton wrote:
> [http://wiki.opencsw.org/package-hooks]

> no it's not clear enough :) you need to be explicit, and state
> whether or not a "upgrade" hook overlaps install/remove hooks, or whether
> it gets called instead of them.
> You also have to explicitly define the order of calling, if they both
> get run

> [hmm actually, you sort of say this in your examples. but you make
>  some mistakes in your examples :) for example, you mix stating
>   that something is being "installed" vs being "upgraded", in the same
>    sample paragraph]

I'll give you the point, but the error is only in perspective.  If a
user requests CSWfoo to be installed when it already exists, the tool
turns it into an upgrade action (layered semantics).  My wording was
from the point of view of the user request.  I've corrected things to
be from the point of view of the tool, since that's more appropriate
to the document.

> > There is no distinction between batchinstall or batchupgrade if that's
> > what you were asking.  We could note the difference by adding a
> > batchupgrade hook if you think that's worthwhile.  I'm not sure it is,
> > but it wouldn't necessarily hurt either.
> 
> If you think it's neccessary to have a per-package upgrade, i think you
> should provide for a batch upgrade, just for completeness. even if you cant
> think of a use for it now, someone later might, if you provide a hook for
> it.

Ok.  We'll add it.  Since this is (again) a synthetic addition, I'll
define batchupgrade such that if _any_ package is to be upgraded from
an older version, the batchupgrade hook will supersede batchadd (which
I've now named batchinstall to be consistent with per-package names).

> Please note: "purge" is DIFFERENT from "remove".
> It has a long-understood meaning in the debian world, and possibly
> others.

I'm aware of the differences.  Purge was used to avoid a name
collision.  That collision no longer exists and we'll forget about
purge.

> "remove" means "run pkgrm".
> "purge" should mean "run pkgrm, and REMOVE ANY LEFTOVERS" (in other
> words, remove any capability/saved state for a delayed upgrade via
> pkgadd later)

Personally, I think even upgrade is a bit of a stretch, since we're
now layering semantics on top of pkgadd/pkgrm.  It's low hanging fruit
though, since the information is readily available in the tools.
While we have the sampleconf classes that can accomplish purge-ish
behaviour, I don't think it would be good to extend too far beyond the
boundaries of pkgadd/pkgrm.

It's also quite difficult for pkg-get/pkgutil to detect the difference
between remove and purge (as handled by the classes), if we wanted to
synthesize it, since it would have to start snooping around to even
get that info.  Unless someone really thinks purge is important, I
think we should walk away from 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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/4ff27dda/attachment-0002.asc>

From schwindt at dfki.uni-kl.de  Fri Jul  3 15:04:33 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 03 Jul 2009 15:04:33 +0200
Subject: [csw-maintainers] libpango 1.24 Problems ?
In-Reply-To: Your message of "Fri, 03 Jul 2009 13:42:45 +0200."
	<FAF013CB-5FCF-4559-9891-6025FFF564A2@opencsw.org> 
Message-ID: <200907031304.n63D4Y7k025114@dfki.uni-kl.de>

[...]
> You can try this right now with
>    rm /opt/csw/bin/pango-querymodules
>    ln -s /opt/csw/bin/sparcv8/pango-querymodules /opt/csw/bin/pango- 
> querymodules
> Please tell me if that worked.

I basically did that and it worked.


[...]
> Is this really necessary on a per-app basis? Can't this be patched
> into pango itself?

If pango was done better, there was an option for i.e. 
PANGO_MODULES_PATH_{32,64}
which could point to /opt/csw/lib or /opt/csw/lib/64. There it could read
pango.modules only containing filename like pango-indic-lang.so not the 
absolute
path. But this would leed to the necessity of pango knowing if it is 32/64 bit.


The main problem  is pango is needing its pango.modules and does not honor in 
any kind
the architecture. pango32 and pango64 will basically open the same file.
In my case it meens, to be able to use xcfe which is 32-bit I generated 
pango.modules
by hand to contain only the 32-bit modules. Anything which now tries to use 
pango64 will
get into trouble.


This flaw gets other packagers in trouble also.


Nicolai




From phil at bolthole.com  Fri Jul  3 16:06:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 3 Jul 2009 07:06:32 -0700
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <1246626185-sup-8972@ntdws12.chass.utoronto.ca>
References: <20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
	<20090703010347.GK94553@bolthole.com>
	<1246626185-sup-8972@ntdws12.chass.utoronto.ca>
Message-ID: <20090703140631.GB64022@bolthole.com>

On Fri, Jul 03, 2009 at 09:15:04AM -0400, Ben Walton wrote:
> ...
> It's also quite difficult for pkg-get/pkgutil to detect the difference
> between remove and purge (as handled by the classes), if we wanted to
> synthesize it, since it would have to start snooping around to even
> get that info.  Unless someone really thinks purge is important, I
> think we should walk away from it.

I have been meaning to implement explicit purge in pkg-get, for a long
time.

Its the reason why the cswpreserveconf class exists, for example:
so that there will be a deterministic way of accomplishing purge.
That class makes it possible in theory to run "pkg-get purge", even if
  the original package is no longer installed! 
Too bad I havent written it yet :)



From bwalton at opencsw.org  Fri Jul  3 16:18:09 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 03 Jul 2009 10:18:09 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090703140631.GB64022@bolthole.com>
References: <20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
	<1246577961-sup-2700@ntdws12.chass.utoronto.ca>
	<20090703010347.GK94553@bolthole.com>
	<1246626185-sup-8972@ntdws12.chass.utoronto.ca>
	<20090703140631.GB64022@bolthole.com>
Message-ID: <1246630414-sup-2279@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 03 10:06:32 -0400 2009:

> I have been meaning to implement explicit purge in pkg-get, for a long
> time.

Well then, we can either define a purge hook up front or add it later
when the ability exists to actually use it.  Adding hooks later is
much easier than removing them after they're defined (and used).

My preference would be to add it later, but it's not something I feel
strongly about either way.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/164ac626/attachment-0002.asc>

From bwalton at opencsw.org  Sat Jul  4 02:46:53 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 03 Jul 2009 20:46:53 -0400
Subject: [csw-maintainers] package hooks (update)
In-Reply-To: <20090702220445.GI94553@bolthole.com>
References: <1246554093-sup-189@ntdws12.chass.utoronto.ca>
	<20090702175132.GK84645@bolthole.com>
	<1246560996-sup-1808@ntdws12.chass.utoronto.ca>
	<20090702190530.GE94553@bolthole.com>
	<1246562604-sup-5118@ntdws12.chass.utoronto.ca>
	<20090702193950.GH94553@bolthole.com>
	<1246563839-sup-2157@ntdws12.chass.utoronto.ca>
	<20090702220445.GI94553@bolthole.com>
Message-ID: <1246667705-sup-8854@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 02 18:04:45 -0400 2009:
> I dont think you should be using files to pass around arguments at all.
> particularly something as trivial as "a list of affected packages".
> I think you shoudl have the list of packages, just be passed as the
> argument list.

Ok, then.  How about the following for input to the scripts:

Each affected package (for batch hooks, the entire list, for
per-package, a single package) is passed as an argument of the form
$svr4name-$version

where $svr4name corresponds to the third field of the catalog and the
$version corresponds to the second field of the catalog.

A hypothetical call to a prebatchupgrade hook might look like:

/etc/opt/csw/pkg-hooks/prebatchupgrade.d/01-CSWfoo-logupgrades CSWbar-1.2,REV=2009.06.05 CSWbaz-1.3.4,REV=2008.12.01

A hyptothetical call to a postremove hook might look like:

/etc/opt/csw/pkg-hooks/postremove.d/02-CSWbar-mailaboutremoval CSWbar-1.2,REV=2009.06.05

This should be all of the useful data we can pass to the hooks without
doing backflips.

If a specific hook wants to be able to differentiate between which
packages are being upgraded and which are installed, it can use
multiple hooks and correlate data between the calls or simply monitor
the per-package hooks instead of the batch hooks.

If a specific hook wants to check currently installed version, it can
do that in a pre* hook, possibly saving state somewhere.

For saved state, I think we should define a directory (maybe the
previously proposed /var/opt/csw/pkg-hooks/) as the place where hooks
are to store any temporary files across different hook
scripts...attempting to keep scripts away from /var/tmp.  Obviously we
won't really be able to force this, but if it's documented, those
reading the documentation to add hooks of their own will hopefully see
and use it.

Reasonable?

Also, when we finish nailing down the details, will the required
directories be added to CSWcommon?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/fceaefb7/attachment-0002.asc>

From rupert at opencsw.org  Sat Jul  4 12:39:45 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 12:39:45 +0200
Subject: [csw-maintainers] include asvn script into cswsvn,
	at least client: how to do this?
Message-ID: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>

hi,

how would one include asvn
(http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn ) best
into the cswsvn package?

rupert.



---------- Forwarded message ----------
From: Thurner Rupert (KIRF 65) <rupert.thurner at credit-suisse.com>
Date: Wed, Jul 1, 2009 at 18:07
Subject: include asvn into cswsvn pls
To: rupert THURNER <rupert at opencsw.org>




______________________________________________
From: ? Lawie Duncan (CS)
Sent:?? Wednesday, July 01, 2009 3:17 PM
To:???? Thurner Rupert (KIRF 65); Kannenberg Ralf (KIRG 1); Molini
Robert R. (KIRG 1)
Subject:??????? SVN and file permissions

Rupert,

it took us a good few drawings, but I think our discussion this
afternoon was very useful.

The "asvn" script I referred to is available at
http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn

It looks to do exactly what we need, in terms of maintaining file
permissions, but I haven't actually been able to test it because it
relies on a -printf option to "find"

PRINTDETAILS="-printf \"file='%p' mode=%m user=%u(%U) group=%g(%G)\n\""
eval "find $PCWD $SKIPSVN -o \( \( -type d ! -name .svn \) -o -type f
\) $PRINTDETAILS"

I understood that we should be able to install? GNU find alongside
svn, which would allow us to test this script.? Would it be best to
get it from http://www.gnu.org/software/findutils/ ?

Assuming that this does what we need, is there a formal mechanism for
requesting that asvn and GNU find are made a part of the standard SVN
product?

Thanks,
Duncan.

Duncan Lawie
Database Component Engineering - Oracle
+44 (0)20 788 3 9229

"If you will take my advice, you will think little of Socrates and a
great deal more for truth." -- Socrates.


From bwalton at opencsw.org  Sat Jul  4 14:35:57 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sat, 04 Jul 2009 08:35:57 -0400
Subject: [csw-maintainers] include asvn script into cswsvn,
	at least client: how to do this?
In-Reply-To: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
References: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
Message-ID: <1246710529-sup-4564@ntdws12.chass.utoronto.ca>

Excerpts from rupert THURNER's message of Sat Jul 04 06:39:45 -0400 2009:
> how would one include asvn
> (http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn ) best
> into the cswsvn package?

You'll need file a bug/feature request, or maybe just ping Mike
directly.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090704/52e99f61/attachment-0002.asc>

From mwatters at opencsw.org  Sat Jul  4 15:07:58 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 04 Jul 2009 08:07:58 -0500
Subject: [csw-maintainers] include asvn script into cswsvn,
 at least client: how to do this?
In-Reply-To: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
References: <6af4270907040339g3ed64bb7p58984d7005ca450@mail.gmail.com>
Message-ID: <4A4F542E.4020306@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

rupert THURNER wrote:

> how would one include asvn
> (http://svn.collab.net/repos/svn/trunk/contrib/client-side/asvn ) best
> into the cswsvn package?

file a bug 'feature request' so I can track the request.  I will add it to the
next 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 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkpPVC0ACgkQLrhmsXMSLxcRRACdFreriUDFQfJekcl9NRJF0wse
BQQAn2SdnZicVuK1NyEp3DLbhGGeVvnq
=FVU5
-----END PGP SIGNATURE-----


From skayser at opencsw.org  Sat Jul  4 18:29:04 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 04 Jul 2009 18:29:04 +0200
Subject: [csw-maintainers] RFEs for "GAR packages status" page (aka
 "Which packages are in GAR")
In-Reply-To: <4A493506.9050204@wbonnet.net>
References: <4A2403D0.4090109@opencsw.org>
	<50341.217.227.50.24.1246310419.squirrel@ssl.skayser.de>
	<4A493506.9050204@wbonnet.net>
Message-ID: <4A4F8350.9090100@opencsw.org>

William Bonnet wrote:
> Hi Sebastian
>> Sebastian Kayser wrote:
>>   
>>> a couple of RFEs for the "GAR package status" [1] page. Can we get:
>>>
>>> * an "last updated" field at the top of the page
>>> * the numbers / stats right at the top of the page (instead of further
>>>   down)
>>> * a contact address for the maintainer of the page on the page (for
>>>   queries like this one and to get packages that are in GAR but not
>>>   listed as such fixed)
>>>
>>> Ah, and some sort of monthly historical data would be cool also. So that
>>> we can see some progress.
>>>
>>> Sebastian
>>>
>>> [1] http://www.opencsw.org/package-gar-status.html
>>>     
>> Anyone taking care of the GAR status page?
>>   
> I'm currently working on this page and uwatch. Rewriting it in Perl and 
> improving things.
> 
> I will modify the page by the end of the week, and uwatch next week. 
> uwatch have several bugs to fix (problem with garname used to retrieve 
> package instead of catalog name). I would like to merge some information 
> from uwatch to this page.

Just in case you don't have it on your radar, yet: The recent
sourceforge page makeover introduced an easy URL for project files.

  http://sourceforge.net/projects/<project_name>/files/

Might help to simplify uwatch and check-upstream handling.

Sebastian


From rupert at opencsw.org  Sat Jul  4 18:40:14 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 18:40:14 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
In-Reply-To: <4A4C6903.2040906@opencsw.org>
References: <4A4C6903.2040906@opencsw.org>
Message-ID: <6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>

i compiled it with gar v2, upgraded the version number, and it runs
through without any noticable prolems. i also checked this in. did not
yet try to fix the tests though ...

rupert.

2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
> If someone feel like getting this mostly python package out, feel free.
>
> I almost got 1.2 working but there where too many test cases that failed.
> Most of the failures was because of missing gnu'isms in the tools the test
> used from what I could tell. Would be nice to patch up the test suite to be
> able to control the path to the tools it uses.
>
> --
> Trygve
>
>
> Hello dear mercurial maintainer,
>
> The upstream notification job has detected the availability of new files for
> mercurial.
>
> The following upstream file(s):
> ? ?mercurial-1.3.tar.gz
>
> is/are available at the following url(s):
> ? ?http://selenic.com/mercurial/release/
>
> Please consider updating your package.
>
> --
> Kindest regards
> upstream notification job
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From trygvis at opencsw.org  Sat Jul  4 18:59:42 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sat, 04 Jul 2009 18:59:42 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream
	update	notification]
In-Reply-To: <6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
Message-ID: <4A4F8A7E.70704@opencsw.org>

rupert THURNER wrote:
> i compiled it with gar v2, upgraded the version number, and it runs
> through without any noticable prolems. i also checked this in. did not
> yet try to fix the tests though ...

Do you have a URl to the version in testing?

--
Trygve

> rupert.
> 
> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>> If someone feel like getting this mostly python package out, feel free.
>>
>> I almost got 1.2 working but there where too many test cases that failed.
>> Most of the failures was because of missing gnu'isms in the tools the test
>> used from what I could tell. Would be nice to patch up the test suite to be
>> able to control the path to the tools it uses.
>>
>> --
>> Trygve
>>
>>
>> Hello dear mercurial maintainer,
>>
>> The upstream notification job has detected the availability of new files for
>> mercurial.
>>
>> The following upstream file(s):
>>    mercurial-1.3.tar.gz
>>
>> is/are available at the following url(s):
>>    http://selenic.com/mercurial/release/
>>
>> Please consider updating your package.
>>
>> --
>> Kindest regards
>> upstream notification job
>>
>> _______________________________________________
>> 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 rupert at opencsw.org  Sat Jul  4 19:01:03 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 19:01:03 +0200
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
Message-ID: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>

hi,

when doing the mercurial tests, there is a warning.

gpg: Warning: using insecure memory!
  --> http://www.gnupg.org/faq.html#q6.1

the faq says that it should be installed with setuid root. is this
deliberately different on opencsw?

rupert.


From skayser at opencsw.org  Sat Jul  4 19:12:05 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 04 Jul 2009 19:12:05 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
 instead of PROTOTYPE_FILTER
In-Reply-To: <3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
Message-ID: <4A4F8D65.9080203@opencsw.org>

Dagobert Michelsen wrote:
> Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
>> while working on postfix 2.6 i have noticed again that it is not quite
>> straight-forward to assign files to classes. Instead of exposing the  
>> inner
>> workings of the prototype stuff to the Makefile, could we maybe get  
>> some
>> variables to assign files to classes (functions)?
> 
> For CSWcswclassutils this is already in there :-) Just define
> 
> SAMPLECONF
> PRESERVECONF
> INITSMF
> 
> and everything else will be taken cared of.

Sweet!!! :D Can we add something for USERGROUP also?

Index: /home/skayser/mgar/gar/v2/gar.pkg.mk
===================================================================
--- /home/skayser/mgar/gar/v2/gar.pkg.mk        (revision 5361)
+++ /home/skayser/mgar/gar/v2/gar.pkg.mk        (working copy)
@@ -155,12 +155,14 @@
 SPKG_CLASSES := $(SPKG_CLASSES) $(if $(SAMPLECONF),$(if $(filter cswsampleconf,$(SPKG_CLASSES)),,cswsampleconf))
 SPKG_CLASSES := $(SPKG_CLASSES) $(if $(PRESERVECONF),$(if $(filter cswpreserveconf,$(SPKG_CLASSES)),,cswpreserveconf))
 SPKG_CLASSES := $(SPKG_CLASSES) $(if $(INITSMF),$(if $(filter cswinitsmf,$(SPKG_CLASSES)),,cswinitsmf))
+SPKG_CLASSES := $(SPKG_CLASSES) $(if $(USERGROUP),$(if $(filter cswusergroup,$(SPKG_CLASSES)),,cswusergroup))
 # - set class for all config files
 ifneq ($(SAMPLECONF)$(PRESERVECONF)$(INITSMF),)
 _CSWCLASS_FILTER = | perl -ane '\
                $(foreach FILE,$(SAMPLECONF),$$F[1] = "cswsampleconf" if ( $$F[2] =~ m(^$(FILE)$$) );)\
                $(foreach FILE,$(PRESERVECONF),$$F[1] = "cswpreserveconf" if( $$F[2] =~ m(^$(FILE)$$) );)\
                $(foreach FILE,$(INITSMF),$$F[1] = "cswinitsmf" if( $$F[2] =~ m(^$(FILE)$$) );)\
+               $(foreach FILE,$(USERGROUP),$$F[1] = "cswusergroup" if( $$F[2] =~ m(^$(FILE)$$) );)\
                print join(" ", at F),"\n";'
 _EXTRA_GAR_PKGS += CSWcswclassutils
 endif

Sebastian


From skayser at opencsw.org  Sat Jul  4 19:21:52 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 04 Jul 2009 19:21:52 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream
	update	notification]
In-Reply-To: <6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
Message-ID: <4A4F8FB0.8040105@opencsw.org>

rupert THURNER wrote:
> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>> If someone feel like getting this mostly python package out, feel free.
>>
>> I almost got 1.2 working but there where too many test cases that failed.
>> Most of the failures was because of missing gnu'isms in the tools the test
>> used from what I could tell. Would be nice to patch up the test suite to be
>> able to control the path to the tools it uses.
>
> i compiled it with gar v2, upgraded the version number, and it runs
> through without any noticable prolems. i also checked this in. did not
> yet try to fix the tests though ...

If tests fail because they rely on GNU features of various tools, you
could try to tweak the PATH used for the tests.

  TEST_ARGS = "PATH=/opt/csw/gnu:$$PATH"

Helped here in a few cases, don't know whether it does for mercurial.
TEST_ENV (instead of TEST_ARGS) should also do the job now that i think
about it.

Sebastian

>> Hello dear mercurial maintainer,
>>
>> The upstream notification job has detected the availability of new files for
>> mercurial.
>>
>> The following upstream file(s):
>>    mercurial-1.3.tar.gz
>>
>> is/are available at the following url(s):
>>    http://selenic.com/mercurial/release/
>>
>> Please consider updating your package.
>>
>> --
>> Kindest regards
>> upstream notification job


From rupert at opencsw.org  Sat Jul  4 19:37:35 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 19:37:35 +0200
Subject: [csw-maintainers] v2-pbuild - test with subversion
Message-ID: <6af4270907041037s4861a955y30e5b94c1f1197c@mail.gmail.com>

hi,

i tried to switch the svn package to v2-pbuild, and noticed the following:


1. parallel start
the file gets downloaded wrongly (parallel wget)


2. start without clean, after the other build is at "unzipping"
the build on build8s is faster, and finishes. the build on build8x
finishes later with:

/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Client
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.so
: Permission denied
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.so
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.bs
: Permission denied
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Client/_Client.bs
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Core
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Core/_Core.so
: Permission denied
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Core/_Core.so
/home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/./opt/csw/lib/perl/site_perl/auto/SVN/_Core/.packlist
pax: /home/rupert/mgar/pkg/subversion/trunk/work/pkgroot/opt/csw/lib/perl/site_perl/auto/SVN/_Core/_Core.bs
: Permission denied


rupert


From rupert at opencsw.org  Sat Jul  4 20:03:30 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 20:03:30 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
In-Reply-To: <4A4F8A7E.70704@opencsw.org>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
	<4A4F8A7E.70704@opencsw.org>
Message-ID: <6af4270907041103s69e51601rb54722e60498f886@mail.gmail.com>

the test result is: # Ran 345 tests, 30 skipped, 8 failed.

Failed test-command-template: output changed
Failed test-convert-cvs-branch: output changed and returned error code 1
Failed test-convert-git: output changed
Failed test-gpg: output changed
Failed test-hook: output changed
Failed test-repair-strip: output changed
Failed test-template-engine: output changed and returned error code 255
Failed test-transplant: output changed


and some tests are skipped which probably should not:

Skipped test-convert-hg-svn: missing feature: subversion python bindings
Skipped test-convert-svn-branches: missing feature: subversion python bindings
Skipped test-convert-svn-encoding: missing feature: subversion python bindings
Skipped test-convert-svn-move: missing feature: subversion python bindings
Skipped test-convert-svn-sink: missing feature: subversion python bindings
Skipped test-convert-svn-source: missing feature: subversion python bindings
Skipped test-convert-svn-startrev: missing feature: subversion python bindings
Skipped test-convert-svn-tags: missing feature: subversion python bindings

is there a possibility to make this working on the build servers?

for some failures i have no idea where the failure is coming from,
will ask the mercurial mailing list. i noticed that also for fedora
the test cases failed sometimes and they were disabled for release
building.


rupert.


2009/7/4 Trygve Laugst?l <trygvis at opencsw.org>:
> rupert THURNER wrote:
>>
>> i compiled it with gar v2, upgraded the version number, and it runs
>> through without any noticable prolems. i also checked this in. did not
>> yet try to fix the tests though ...
>
> Do you have a URl to the version in testing?
>
> --
> Trygve
>
>> rupert.
>>
>> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>>>
>>> If someone feel like getting this mostly python package out, feel free.
>>>
>>> I almost got 1.2 working but there where too many test cases that failed.
>>> Most of the failures was because of missing gnu'isms in the tools the
>>> test
>>> used from what I could tell. Would be nice to patch up the test suite to
>>> be
>>> able to control the path to the tools it uses.
>>>
>>> --
>>> Trygve
>>>
>>>
>>> Hello dear mercurial maintainer,
>>>
>>> The upstream notification job has detected the availability of new files
>>> for
>>> mercurial.
>>>
>>> The following upstream file(s):
>>> ? mercurial-1.3.tar.gz
>>>
>>> is/are available at the following url(s):
>>> ? http://selenic.com/mercurial/release/
>>>
>>> Please consider updating your package.
>>>
>>> --
>>> Kindest regards
>>> upstream notification job
>>>
>>> _______________________________________________
>>> 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
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers


From rupert at opencsw.org  Sat Jul  4 20:36:52 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 20:36:52 +0200
Subject: [csw-maintainers] [Fwd: [svn] mercurial upstream update
	notification]
In-Reply-To: <6af4270907041103s69e51601rb54722e60498f886@mail.gmail.com>
References: <4A4C6903.2040906@opencsw.org>
	<6af4270907040940m11eab87ck8ef6dd335de5fca1@mail.gmail.com>
	<4A4F8A7E.70704@opencsw.org>
	<6af4270907041103s69e51601rb54722e60498f886@mail.gmail.com>
Message-ID: <6af4270907041136i1bbba3b8kc97a9c85414197d7@mail.gmail.com>

i put the packages onto http://mirror.opencsw.org/testing.html, and
made them writeable for others as well.

rupert.

On Sat, Jul 4, 2009 at 20:03, rupert THURNER<rupert at opencsw.org> wrote:
> the test result is: # Ran 345 tests, 30 skipped, 8 failed.
>
> Failed test-command-template: output changed
> Failed test-convert-cvs-branch: output changed and returned error code 1
> Failed test-convert-git: output changed
> Failed test-gpg: output changed
> Failed test-hook: output changed
> Failed test-repair-strip: output changed
> Failed test-template-engine: output changed and returned error code 255
> Failed test-transplant: output changed
>
>
> and some tests are skipped which probably should not:
>
> Skipped test-convert-hg-svn: missing feature: subversion python bindings
> Skipped test-convert-svn-branches: missing feature: subversion python bindings
> Skipped test-convert-svn-encoding: missing feature: subversion python bindings
> Skipped test-convert-svn-move: missing feature: subversion python bindings
> Skipped test-convert-svn-sink: missing feature: subversion python bindings
> Skipped test-convert-svn-source: missing feature: subversion python bindings
> Skipped test-convert-svn-startrev: missing feature: subversion python bindings
> Skipped test-convert-svn-tags: missing feature: subversion python bindings
>
> is there a possibility to make this working on the build servers?
>
> for some failures i have no idea where the failure is coming from,
> will ask the mercurial mailing list. i noticed that also for fedora
> the test cases failed sometimes and they were disabled for release
> building.
>
>
> rupert.
>
>
> 2009/7/4 Trygve Laugst?l <trygvis at opencsw.org>:
>> rupert THURNER wrote:
>>>
>>> i compiled it with gar v2, upgraded the version number, and it runs
>>> through without any noticable prolems. i also checked this in. did not
>>> yet try to fix the tests though ...
>>
>> Do you have a URl to the version in testing?
>>
>> --
>> Trygve
>>
>>> rupert.
>>>
>>> 2009/7/2 Trygve Laugst?l <trygvis at opencsw.org>:
>>>>
>>>> If someone feel like getting this mostly python package out, feel free.
>>>>
>>>> I almost got 1.2 working but there where too many test cases that failed.
>>>> Most of the failures was because of missing gnu'isms in the tools the
>>>> test
>>>> used from what I could tell. Would be nice to patch up the test suite to
>>>> be
>>>> able to control the path to the tools it uses.
>>>>
>>>> --
>>>> Trygve
>>>>
>>>>
>>>> Hello dear mercurial maintainer,
>>>>
>>>> The upstream notification job has detected the availability of new files
>>>> for
>>>> mercurial.
>>>>
>>>> The following upstream file(s):
>>>> ? mercurial-1.3.tar.gz
>>>>
>>>> is/are available at the following url(s):
>>>> ? http://selenic.com/mercurial/release/
>>>>
>>>> Please consider updating your package.
>>>>
>>>> --
>>>> Kindest regards
>>>> upstream notification job
>>>>
>>>> _______________________________________________
>>>> 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
>>
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From rupert at opencsw.org  Sat Jul  4 20:40:26 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sat, 4 Jul 2009 20:40:26 +0200
Subject: [csw-maintainers] libserf-0.3.0 now in testing
Message-ID: <6af4270907041140u74538208md46c69f8b9eaa13d@mail.gmail.com>

many thanks! libserf-0.3.0 is now in testing.

rupert.

On Mon, Jun 29, 2009 at 20:03, Mike Watters<mwatters at opencsw.org> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> rupert THURNER wrote:
>
>>
>> how could one avoid that?
>>
>
> the main issue is the Makefile generated by configre is using "libtool" from
> /opt/csw/apache2/share/build/libtool which does "NOT" get modified by my strip.
>
> Here is a patch to the Makefile currently in svn that will allow you to build
> and test successfully libserf
>
>
> if you cut and paste, please take care to watch for wrapping I have Thunderbird
> configured to wrap at 100 characters.
>
> $ svn diff Makefile
> Index: Makefile
> ===================================================================
> - --- Makefile ? ?(revision 5414)
> +++ Makefile ? ?(working copy)
> @@ -23,13 +23,19 @@
> ?# UPSTREAM_MASTER_SITES =
>
> ?DEPENDS = server/apache2rt
> - -STRIP_LIBTOOL = 1
>
> ?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=$(prefix)/apache2/bin/apu-config
> +CONFIGURE_ARGS += --with-apr=$(prefix)/apache2/bin/apr-config
CONFIGURE_ARGS += --with-apr-util=$(prefix)/apache2/bin/apu-config
>
> ?TEST_TARGET = check
> ?include gar/category.mk
> +
> +EXTRA_LDFLAGS = -R/opt/csw/bdb44/lib
> +APR_LIBTOOL = $(shell /opt/csw/apache2/bin/apr-config --apr-libtool)
> +post-configure-modulated:
> + ? ? ? gcp $(APR_LIBTOOL) $(WORKSRC)
> + ? ? ? perl -i -pne 's|$(APR_LIBTOOL)|$(abspath $(WORKSRC)/libtool)|' \
> + ? ? ? ? ? ? ? $(WORKSRC)/Makefile
> + ? ? ? $(GARBIN)/fixlibtool $(WORKSRC)
> + ? ? ? @$(MAKECOOKIE)
>
>
>
> - --
>
> 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
>
> iEYEARECAAYFAkpJAdUACgkQLrhmsXMSLxeBXQCgmTxH6hTrxyY2TrT6eSMj+S4t
> c+4Anj0nU/L6fyoBYYX4VqFjMCGd4iRo
> =qeGL
> -----END PGP SIGNATURE-----
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From praba_tuty at yahoo.com  Fri Jul  3 11:47:19 2009
From: praba_tuty at yahoo.com (prabakaran)
Date: Fri, 3 Jul 2009 02:47:19 -0700 (PDT)
Subject: [csw-maintainers] HELP: testing for new package CSQL
Message-ID: <887751.48196.qm@web52903.mail.re2.yahoo.com>



Hi Team,
I am creating csw package for CSQL main memory database management system. This is my first package creation experience.. I have put the package under the below nfs directory

/home/testing/csql-2.5.1,REV=2009.07.02-SunOS5.10-sparc-CSW.pkg.gz

It is created on solaris 10 as it uses atomic compare and swap which is supported only in solaris 10.
Can somebody test this( as i do not have root access) and let me know your comments.

Thanks and Regards
Prabakaran




      


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090703/bd512ebe/attachment-0002.html>

From bonivart at opencsw.org  Mon Jul  6 11:18:39 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Mon, 6 Jul 2009 11:18:39 +0200
Subject: [csw-maintainers] HELP: testing for new package CSQL
In-Reply-To: <887751.48196.qm@web52903.mail.re2.yahoo.com>
References: <887751.48196.qm@web52903.mail.re2.yahoo.com>
Message-ID: <625385e30907060218s5c6c6bb5j539fd58003b06518@mail.gmail.com>

On Fri, Jul 3, 2009 at 11:47 AM, prabakaran<praba_tuty at yahoo.com> wrote:
> Hi Team,
> I am creating csw package for CSQL main memory database management system.
> This is my first package creation experience.. I have put the package under
> the below nfs directory
>
> /home/testing/csql-2.5.1,REV=2009.07.02-SunOS5.10-sparc-CSW.pkg.gz
>
> It is created on solaris 10 as it uses atomic compare and swap which is
> supported only in solaris 10.
> Can somebody test this( as i do not have root access) and let me know your
> comments.

I tried to install it on Solaris 10 x86 but I got a prompt for base
directory, that needs to be fixed.

"Enter path to package base directory [?,q]"

Also, don't display the license during installation. You don't seem to
be using GAR (why not?) so put this in the copyright file:

"Please see /opt/csw/share/doc/csql/license for license information"
and then copy the full license to that file.

-- 
/peter


From ja at opencsw.org  Mon Jul  6 14:25:54 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Mon, 06 Jul 2009 14:25:54 +0200
Subject: [csw-maintainers] Munin 1.3.4 in testing
Message-ID: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>

Hi,

I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW.  
The software consists of the packages munin_master and munin_node.

Feedback is welcome.

Juergen

-- 
Juergen Arndt


From skayser at opencsw.org  Mon Jul  6 17:34:51 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Mon, 06 Jul 2009 17:34:51 +0200
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
Message-ID: <4A52199B.5070808@opencsw.org>

Hi,

i am seeing some weird issues while trying to build libiconv on
build8s. The regular build with GAR works just fine, but when i try
to incorporate the optimization flags that Phil has used so far
(--fast --xnolibmopt), the ./configure on build8s just hangs at

  checking for stdbool.h that conforms to C99...

GAR build description is at [1]. build8x is fine, i can only see
this issue on build8s. The ptree looks something like

        25961 gmake build-isa-sparcv8
          26068 gmake MODULATION=isa-sparcv8 ISA=sparcv8 build-modulated
            26886 /bin/sh -c cd work/build-isa-sparcv8/libiconv-1.13 && prefix="/opt/csw" exec_pr
              26887 /bin/bash ./configure --prefix=/opt/csw --exec_prefix=/opt/csw --bindir=/opt/cs
                29070 /bin/bash ./configure --prefix=/opt/csw --exec_prefix=/opt/csw --bindir=/opt/cs
                  29071 /opt/studio/SOS11/SUNWspro/bin/cc -c -fast -xnolibmopt -xarch=v8 -I/opt/csw/inc
                    29072 /opt/studio/SOS11/SUNWspro/prod/bin/acomp -xintrinsic -xldscope=global -depend

Trussing the acomp shows a myriad of brk()s and the process hogs
memory like there is no tomorrow.

   PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
 29072 skayser   684M  670M cpu13   30    0   0:01.23 3.1% acomp/1

When i take the conftest.c that ./configure tries to compile [2] and
try to compile it manually it simply bails out like it does on
build8x [3] (cannot find stdbool.h), but it does not hang. Anyone
seen something like this before? Or any other idea what could be the
culprit here?

Sebastian

[1] https://gar.svn.sf.net/svnroot/gar/csw/mgar/pkg/libiconv/trunk/Makefile
[2] http://pastebin.com/m6010e35c
[3] http://pastebin.com/m18121654

P.S.: Building with these optimization flags on build8x makes the
libiconv test suite complete in about 1:35 - 1:40 minutes versus 1:55
minutes for the non-optimized version, so it seems worth making the
opimization flags work.


From phil at bolthole.com  Mon Jul  6 18:11:51 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 09:11:51 -0700
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
	./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A52199B.5070808@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
Message-ID: <20090706161151.GA36831@bolthole.com>

On Mon, Jul 06, 2009 at 05:34:51PM +0200, Sebastian Kayser wrote:
> Trussing the acomp shows a myriad of brk()s and the process hogs
> memory like there is no tomorrow.
> 
>    PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
>  29072 skayser   684M  670M cpu13   30    0   0:01.23 3.1% acomp/1

yes. this is known behaviour of cc.
(although I havent seen it with libiconv specifically)

certain obscure optimization cases, explode, so to speak, taking extreme
memory usage and disk space for it to "traverse the problem space", if you
will.

Your choices are:

a) wait for a long time, and it will eventually complete, if the machine
doesnt run out of resources

b) reduce the optimization flags to -xO2 or -xO3



From trygvis at opencsw.org  Mon Jul  6 18:22:45 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Mon, 06 Jul 2009 18:22:45 +0200
Subject: [csw-maintainers] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <4A5224D5.8030803@opencsw.org>

Juergen Arndt wrote:
> Hi,
> 
> I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW. 
> The software consists of the packages munin_master and munin_node.
> 
> Feedback is welcome.

I just tried your packages. The installation look fine and I made some 
small notes on the wiki[1]. However, when I tried to create a graph, all 
and any letters are rendered as a box. See the attached image. I assume 
that this really is an rrdtool issue more than a Munin issue.

[1]: http://wiki.opencsw.org/munin-package

--
Trygve


From trygvis at opencsw.org  Mon Jul  6 18:32:29 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Mon, 06 Jul 2009 18:32:29 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <4A52271D.4000106@opencsw.org>

Juergen Arndt wrote:
> Hi,
> 
> I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW. 
> The software consists of the packages munin_master and munin_node.

Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know to 
be as stable as 1.2.6?

I see that Fedora and Debian both package 1.2.5/.6:

http://dag.wieers.com/rpm/packages/munin/
http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all

--
Trygve


From ja at opencsw.org  Mon Jul  6 18:51:18 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Mon, 06 Jul 2009 18:51:18 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A52271D.4000106@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A52271D.4000106@opencsw.org>
Message-ID: <op.uwnnrrurjbmpvo@imac.lan>

On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l <trygvis at opencsw.org>  
wrote:

> Juergen Arndt wrote:
>> Hi,
>>  I've put Munin 1.3.4 into testing. It's the initial version for  
>> OpenCSW. The software consists of the packages munin_master and  
>> munin_node.
>
> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know to  
> be as stable as 1.2.6?
>
> I see that Fedora and Debian both package 1.2.5/.6:
>
> http://dag.wieers.com/rpm/packages/munin/
> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all

1.2.6 is the last official stable version, but it's already more than a  
year old and I was asked to package a more current version. If this does  
not comply with our policies, I will package 1.2.6 as well, no problem.

> I just tried your packages. The installation look fine and I made some  
> small notes on the wiki[1]. However, when I tried to create a graph, all  
> and any letters are rendered as a box. See the attached image. I assume  
> that this really is an rrdtool issue more than a Munin issue.
>  [1]: http://wiki.opencsw.org/munin-package

I cannot confirm that, on my installation the graphs look as expected.  
Maybe some dependencies are missing, I will try to dig into it.

I will update your wiki page in the next future with some more  
installation details.

Juergen

-- 
Juergen Arndt


From trygvis at opencsw.org  Mon Jul  6 19:04:57 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 06 Jul 2009 19:04:57 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnnrrurjbmpvo@imac.lan>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>
	<op.uwnnrrurjbmpvo@imac.lan>
Message-ID: <4A522EB9.9090400@opencsw.org>

Juergen Arndt wrote:
> On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l 
> <trygvis at opencsw.org> wrote:
> 
>> Juergen Arndt wrote:
>>> Hi,
>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>> OpenCSW. The software consists of the packages munin_master and 
>>> munin_node.
>>
>> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know 
>> to be as stable as 1.2.6?
>>
>> I see that Fedora and Debian both package 1.2.5/.6:
>>
>> http://dag.wieers.com/rpm/packages/munin/
>> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all 
>>
> 
> 1.2.6 is the last official stable version, but it's already more than a 
> year old and I was asked to package a more current version. If this does 
> not comply with our policies, I will package 1.2.6 as well, no problem.

No, I was just wondering really.

>> I just tried your packages. The installation look fine and I made some 
>> small notes on the wiki[1]. However, when I tried to create a graph, 
>> all and any letters are rendered as a box. See the attached image. I 
>> assume that this really is an rrdtool issue more than a Munin issue.
>>  [1]: http://wiki.opencsw.org/munin-package
> 
> I cannot confirm that, on my installation the graphs look as expected. 
> Maybe some dependencies are missing, I will try to dig into it.
> 
> I will update your wiki page in the next future with some more 
> installation details.

I think it would be nice if the package included two complete apache 
configuration files (the basic and the one with CGI) so that the user 
can just Include them from httpd.conf. From what I can tell it is not 
likely that a user will need to adjust the configuration at all, at 
least not to get started.

As for my rrd issue I just did a test from the rrd tutorial and the 
graphs look ok there.

Which encoding does your system use? I'm using en_US.UTF-8.

--
Trygve



From phil at bolthole.com  Mon Jul  6 19:07:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 10:07:52 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnnrrurjbmpvo@imac.lan>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A52271D.4000106@opencsw.org> <op.uwnnrrurjbmpvo@imac.lan>
Message-ID: <20090706170752.GB60655@bolthole.com>

On Mon, Jul 06, 2009 at 06:51:18PM +0200, Juergen Arndt wrote:
> 1.2.6 is the last official stable version, but it's already more than a  
> year old and I was asked to package a more current version. If this does  
> not comply with our policies, I will package 1.2.6 as well, no problem.

well, the "policies" have to be flexible, due to the stupidity of some
software "publisher"s lack of standards and consistency.
The usual questions to be asked are:

1. how long has [specific version proposed] been out in the wild
2. how likely is it that an official new 'stable' version will be released
   soon?

 (and 2a: is the'stable' version so ancient as to be virtually useless)

If the answers are long the lines of

1. a very long time
2. very unlikely

then a newer but non "stable" release of software can be accepted as a
package.



From trygvis at opencsw.org  Mon Jul  6 19:28:18 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 06 Jul 2009 19:28:18 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A522EB9.9090400@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>	<op.uwnnrrurjbmpvo@imac.lan>
	<4A522EB9.9090400@opencsw.org>
Message-ID: <4A523432.3000304@opencsw.org>

Trygve Laugst?l wrote:
> Juergen Arndt wrote:
>> On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l 
>> <trygvis at opencsw.org> wrote:
>>
>>> Juergen Arndt wrote:
>>>> Hi,
>>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>>> OpenCSW. The software consists of the packages munin_master and 
>>>> munin_node.
>>>
>>> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know 
>>> to be as stable as 1.2.6?
>>>
>>> I see that Fedora and Debian both package 1.2.5/.6:
>>>
>>> http://dag.wieers.com/rpm/packages/munin/
>>> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all 
>>>
>>
>> 1.2.6 is the last official stable version, but it's already more than 
>> a year old and I was asked to package a more current version. If this 
>> does not comply with our policies, I will package 1.2.6 as well, no 
>> problem.
> 
> No, I was just wondering really.
> 
>>> I just tried your packages. The installation look fine and I made 
>>> some small notes on the wiki[1]. However, when I tried to create a 
>>> graph, all and any letters are rendered as a box. See the attached 
>>> image. I assume that this really is an rrdtool issue more than a 
>>> Munin issue.
>>>  [1]: http://wiki.opencsw.org/munin-package
>>
>> I cannot confirm that, on my installation the graphs look as expected. 
>> Maybe some dependencies are missing, I will try to dig into it.
>>
>> I will update your wiki page in the next future with some more 
>> installation details.
> 
> I think it would be nice if the package included two complete apache 
> configuration files (the basic and the one with CGI) so that the user 
> can just Include them from httpd.conf. From what I can tell it is not 
> likely that a user will need to adjust the configuration at all, at 
> least not to get started.
> 
> As for my rrd issue I just did a test from the rrd tutorial and the 
> graphs look ok there.

I tried to use the CGI version and got a Pango related error:

[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] (process:6716): 
Pango-CRITICAL **: No modules found:
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] No builtin or 
dynamically loaded modules were found.
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] PangoFc will not 
work correctly.
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] This probably 
means there was an error in the creation of:
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] 
'/opt/csw/etc/pango/pango.modules'
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] You should create 
this file by running:
[Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] 
pango-querymodules > '/opt/csw/etc/pango/pango.modules'

Which I did and I got this output:

/opt/csw/lib/64/pango/1.6.0/modules/pango-arabic-fc.so 
ArabicScriptEngineFc PangoEngineShape PangoRenderFc arabic:* nko:*
[snip]

Why are there only 64-bit modules there?

Anyway, running the CGI again gave me this:

Pango-WARNING **: ld.so.1: perl: fatal: 
/opt/csw/lib/64/pango/1.6.0/modules/pango-basic-fc.so: wrong ELF class: 
ELFCLASS64

--
Trygve


From trygvis at opencsw.org  Mon Jul  6 19:56:04 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 06 Jul 2009 19:56:04 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A523432.3000304@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>	<op.uwnnrrurjbmpvo@imac.lan>	<4A522EB9.9090400@opencsw.org>
	<4A523432.3000304@opencsw.org>
Message-ID: <4A523AB4.6010308@opencsw.org>

Trygve Laugst?l wrote:
> Trygve Laugst?l wrote:
>> Juergen Arndt wrote:
>>> On Mon, 06 Jul 2009 18:32:29 +0200, Trygve Laugst?l 
>>> <trygvis at opencsw.org> wrote:
>>>
>>>> Juergen Arndt wrote:
>>>>> Hi,
>>>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>>>> OpenCSW. The software consists of the packages munin_master and 
>>>>> munin_node.
>>>>
>>>> Also, why was 1.3.4 packaged and not the stable 1.2.6? Is 1.3.4 know 
>>>> to be as stable as 1.2.6?
>>>>
>>>> I see that Fedora and Debian both package 1.2.5/.6:
>>>>
>>>> http://dag.wieers.com/rpm/packages/munin/
>>>> http://packages.debian.org/search?keywords=munin&searchon=names&suite=all&section=all 
>>>>
>>>
>>> 1.2.6 is the last official stable version, but it's already more than 
>>> a year old and I was asked to package a more current version. If this 
>>> does not comply with our policies, I will package 1.2.6 as well, no 
>>> problem.
>>
>> No, I was just wondering really.
>>
>>>> I just tried your packages. The installation look fine and I made 
>>>> some small notes on the wiki[1]. However, when I tried to create a 
>>>> graph, all and any letters are rendered as a box. See the attached 
>>>> image. I assume that this really is an rrdtool issue more than a 
>>>> Munin issue.
>>>>  [1]: http://wiki.opencsw.org/munin-package
>>>
>>> I cannot confirm that, on my installation the graphs look as 
>>> expected. Maybe some dependencies are missing, I will try to dig into 
>>> it.
>>>
>>> I will update your wiki page in the next future with some more 
>>> installation details.
>>
>> I think it would be nice if the package included two complete apache 
>> configuration files (the basic and the one with CGI) so that the user 
>> can just Include them from httpd.conf. From what I can tell it is not 
>> likely that a user will need to adjust the configuration at all, at 
>> least not to get started.
>>
>> As for my rrd issue I just did a test from the rrd tutorial and the 
>> graphs look ok there.
> 
> I tried to use the CGI version and got a Pango related error:
> 
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] (process:6716): 
> Pango-CRITICAL **: No modules found:
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] No builtin or 
> dynamically loaded modules were found.
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] PangoFc will not 
> work correctly.
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] This probably 
> means there was an error in the creation of:
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] 
> '/opt/csw/etc/pango/pango.modules'
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] You should create 
> this file by running:
> [Mon Jul 06 19:21:52 2009] [error] [client 127.0.0.1] pango-querymodules 
>  > '/opt/csw/etc/pango/pango.modules'
> 
> Which I did and I got this output:
> 
> /opt/csw/lib/64/pango/1.6.0/modules/pango-arabic-fc.so 
> ArabicScriptEngineFc PangoEngineShape PangoRenderFc arabic:* nko:*
> [snip]
> 
> Why are there only 64-bit modules there?
> 
> Anyway, running the CGI again gave me this:
> 
> Pango-WARNING **: ld.so.1: perl: fatal: 
> /opt/csw/lib/64/pango/1.6.0/modules/pango-basic-fc.so: wrong ELF class: 
> ELFCLASS64

Ok, I've made more progress. If I ran 
/opt/csw/bin/i386/pango-querymodules I got a more sane configuration.

After that the CGI stuff started to fail, but when I went back to the 
non-CGI way it suddenly worked. Not using the CGI way is not an issue 
for me as it is not recommended at all so I'm happy.

--
Trygve



From trygvis at opencsw.org  Mon Jul  6 20:48:00 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Mon, 06 Jul 2009 20:48:00 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <4A5246E0.80105@opencsw.org>

Juergen Arndt wrote:
> Hi,
> 
> I've put Munin 1.3.4 into testing. It's the initial version for OpenCSW. 
> The software consists of the packages munin_master and munin_node.
> 
> Feedback is welcome.

(I sure hope you're able to collect all my comments :)

Two comments:

1) The files in lib/ should most likely be in libexec/ as they're 
executables and not libraries. Not sure about the font file, probably 
doesn't matter much.

2) The graphs was not rendered with a monospaced font, probably because 
of using RRD 1.3. By hacking the LEGEND field like this it looks much 
better:

--- /opt/csw/lib/munin/munin-graph.orig	2009-07-06 20:43:48.340046000 +0200
+++ /opt/csw/lib/munin/munin-graph	2009-07-06 20:43:41.601520000 +0200
@@ -827,7 +827,7 @@
  	my @complete = ();
  	if ($RRDkludge) {
  	    push (@complete,
-		  '--font' ,'LEGEND:7:/opt/csw/lib/munin/VeraMono.ttf',
+		  '--font' ,'LEGEND:7:monospace',
  		  '--font' ,'UNIT:7:/opt/csw/lib/munin/VeraMono.ttf',
  		  '--font' ,'AXIS:7:/opt/csw/lib/munin/VeraMono.ttf');
  	}

This is the same as Debian does [1] amd has been integrated to the main 
build [2].

[1]: http://munin.projects.linpro.no/changeset/1763
[2]: http://munin.projects.linpro.no/changeset/1797

--
Trygve


From mwatters at opencsw.org  Mon Jul  6 21:44:31 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Mon, 06 Jul 2009 14:44:31 -0500
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A523AB4.6010308@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A52271D.4000106@opencsw.org>	<op.uwnnrrurjbmpvo@imac.lan>	<4A522EB9.9090400@opencsw.org>	<4A523432.3000304@opencsw.org>
	<4A523AB4.6010308@opencsw.org>
Message-ID: <4A52541F.5000203@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Trygve,
Trygve Laugst?l wrote:

>> [snip]
>>
>> Why are there only 64-bit modules there?
>>
>> Anyway, running the CGI again gave me this:
>>
>> Pango-WARNING **: ld.so.1: perl: fatal:
>> /opt/csw/lib/64/pango/1.6.0/modules/pango-basic-fc.so: wrong ELF
>> class: ELFCLASS64
> 
> Ok, I've made more progress. If I ran
> /opt/csw/bin/i386/pango-querymodules I got a more sane configuration.
> 
> After that the CGI stuff started to fail, but when I went back to the
> non-CGI way it suddenly worked. Not using the CGI way is not an issue
> for me as it is not recommended at all so I'm happy.



There is already a thread bouncing around on the pango problem.
subject: [csw-maintainers] libpango 1.24 Problems ?

- --

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

iEYEARECAAYFAkpSVB8ACgkQLrhmsXMSLxddMwCgsENM1ApVj3PVChVK2Myw2wNL
aBkAoIZuA91F0wBy0hN/ZMSUCU9pywGz
=3bd1
-----END PGP SIGNATURE-----


From skayser at opencsw.org  Mon Jul  6 22:12:14 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Mon, 06 Jul 2009 22:12:14 +0200
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <20090706161151.GA36831@bolthole.com>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com>
Message-ID: <4A525A9E.10306@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 06, 2009 at 05:34:51PM +0200, Sebastian Kayser wrote:
>> Trussing the acomp shows a myriad of brk()s and the process hogs
>> memory like there is no tomorrow.
>>
>>    PID USERNAME  SIZE   RSS STATE  PRI NICE      TIME  CPU PROCESS/NLWP
>>  29072 skayser   684M  670M cpu13   30    0   0:01.23 3.1% acomp/1
> 
> yes. this is known behaviour of cc.
> (although I havent seen it with libiconv specifically)
> 
> certain obscure optimization cases, explode, so to speak, taking extreme
> memory usage and disk space for it to "traverse the problem space", if you
> will.
> 
> Your choices are:
> 
> a) wait for a long time, and it will eventually complete, if the machine
> doesnt run out of resources
> 
> b) reduce the optimization flags to -xO2 or -xO3

Thanks for your feedback Phil. Reducing to -xO2 or -xO3 didn't help
though. Now comes the fun part. When i keep the --fast --xnolibmopt
flags as they are and run the configure command that GAR would issue
manually, it completes just fine :/

There is someone over at the OpenSolaris forums who has observed
something similar. cc hanging when invoked from make, but fine on its
own [1]. Unfortunately, neither the forum thread nor the referenced bug
entry [2] help with a working solution.

Sebastian

[1] http://opensolaris.org/jive/thread.jspa?messageID=209219
[2] http://bugs.sun.com/view_bug.do?bug_id=6668199


From phil at bolthole.com  Mon Jul  6 22:29:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 13:29:56 -0700
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
	./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A525A9E.10306@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com> <4A525A9E.10306@opencsw.org>
Message-ID: <20090706202956.GG60655@bolthole.com>

On Mon, Jul 06, 2009 at 10:12:14PM +0200, Sebastian Kayser wrote:
> Philip Brown wrote:
> > Your choices are:
> > 
> > a) wait for a long time, and it will eventually complete, if the machine
> > doesnt run out of resources
> > 
> > b) reduce the optimization flags to -xO2 or -xO3
> 
> Thanks for your feedback Phil. Reducing to -xO2 or -xO3 didn't help
> though. Now comes the fun part. When i keep the --fast --xnolibmopt
> flags as they are and run the configure command that GAR would issue
> manually, it completes just fine :/

Did you try letting it run for half an hour?



From skayser at opencsw.org  Mon Jul  6 22:35:22 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Mon, 06 Jul 2009 22:35:22 +0200
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <20090706202956.GG60655@bolthole.com>
References: <4A52199B.5070808@opencsw.org>	<20090706161151.GA36831@bolthole.com>
	<4A525A9E.10306@opencsw.org> <20090706202956.GG60655@bolthole.com>
Message-ID: <4A52600A.6070803@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 06, 2009 at 10:12:14PM +0200, Sebastian Kayser wrote:
>> Philip Brown wrote:
>>> Your choices are:
>>>
>>> a) wait for a long time, and it will eventually complete, if the machine
>>> doesnt run out of resources
>>>
>>> b) reduce the optimization flags to -xO2 or -xO3
>> Thanks for your feedback Phil. Reducing to -xO2 or -xO3 didn't help
>> though. Now comes the fun part. When i keep the --fast --xnolibmopt
>> flags as they are and run the configure command that GAR would issue
>> manually, it completes just fine :/
> 
> Did you try letting it run for half an hour?

10 minutes maybe (didn't help), not yet half an hour.

build8s is currently already being killed by something else (load
average: 33.58, 33.62, 34.02), will try the half hour thing on some of
the next few days. With the manual workaround it's at least possible now
to complete the build.

Sebastian


From phil at bolthole.com  Mon Jul  6 22:44:19 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 13:44:19 -0700
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
	./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A52600A.6070803@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com> <4A525A9E.10306@opencsw.org>
	<20090706202956.GG60655@bolthole.com>
	<4A52600A.6070803@opencsw.org>
Message-ID: <20090706204418.GH60655@bolthole.com>

On Mon, Jul 06, 2009 at 10:35:22PM +0200, Sebastian Kayser wrote:
> build8s is currently already being killed by something else (load
> average: 33.58, 33.62, 34.02), will try the half hour thing on some of
> the next few days. With the manual workaround it's at least possible now
> to complete the build.
> 

yeah, when in doubt, hand-hack the configure script to just eliminate the
stupid test case :)



From ja at opencsw.org  Mon Jul  6 23:10:42 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Mon, 06 Jul 2009 23:10:42 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A5246E0.80105@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org>
Message-ID: <op.uwnzr4lqjbmpvo@imac.lan>

On Mon, 06 Jul 2009 20:48:00 +0200, Trygve Laugst?l <trygvis at opencsw.org>  
wrote:

> Juergen Arndt wrote:
>> Hi,
>>  I've put Munin 1.3.4 into testing. It's the initial version for  
>> OpenCSW. The software consists of the packages munin_master and  
>> munin_node.
>>  Feedback is welcome.
>
> (I sure hope you're able to collect all my comments :)

I'll try my very best ;) (I use for such purposes a mind map tool -  
http://freemind.sourceforge.net/wiki/index.php/Main_Page, which is really  
capable for such things. Maybe it's interesting for someone too)

So, I have the following points:

1) Use of monospaced font for the graphs
2) Moving executables from /lib to /libexec (thanks for your attention, I  
absolutely didn't see that)
3) Provide Apache config files ready for including - where should the  
files be placed? Already in /opt/csw/apache2/etc/extra ?
4) The problems with pango and the cgi mode. I guess, this will take the  
most time, actually I don't know yet, where the problem is located - in  
pango or in munin or in both.
5) Maybe packages for 1.2.6 are needed.

Did I forget anything? :) I hope, I can solve the first three issues  
tomorrow.

Juergen

-- 
Juergen Arndt


From trygvis at opencsw.org  Tue Jul  7 03:35:05 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Tue, 07 Jul 2009 03:35:05 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnzr4lqjbmpvo@imac.lan>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A5246E0.80105@opencsw.org>
	<op.uwnzr4lqjbmpvo@imac.lan>
Message-ID: <4A52A649.1070606@opencsw.org>

Juergen Arndt wrote:
> On Mon, 06 Jul 2009 20:48:00 +0200, Trygve Laugst?l 
> <trygvis at opencsw.org> wrote:
> 
>> Juergen Arndt wrote:
>>> Hi,
>>>  I've put Munin 1.3.4 into testing. It's the initial version for 
>>> OpenCSW. The software consists of the packages munin_master and 
>>> munin_node.
>>>  Feedback is welcome.
>>
>> (I sure hope you're able to collect all my comments :)
> 
> I'll try my very best ;) (I use for such purposes a mind map tool - 
> http://freemind.sourceforge.net/wiki/index.php/Main_Page, which is 
> really capable for such things. Maybe it's interesting for someone too)
> 
> So, I have the following points:
> 
> 1) Use of monospaced font for the graphs
> 2) Moving executables from /lib to /libexec (thanks for your attention, 
> I absolutely didn't see that)

You probably want to make sure that your use owner:group consistently 
throughout the file objects.

> 3) Provide Apache config files ready for including - where should the 
> files be placed? Already in /opt/csw/apache2/etc/extra ?

I would suggest /etc/opt/csw/munin/apache.conf and 
/etc/opt/csw/munin/apache-cgi.conf

> 4) The problems with pango and the cgi mode. I guess, this will take the 
> most time, actually I don't know yet, where the problem is located - in 
> pango or in munin or in both.

I would ignore this for now as Munin only calls out to rrdtool and the 
HTML etc was generated perfectly.

> 5) Maybe packages for 1.2.6 are needed.

The package is working just fine for me now, so I think this should be 
ok. I've even made a plugin for it and it work :)

> Did I forget anything? :) I hope, I can solve the first three issues 
> tomorrow.

I'm leaving for vacation tomorrow so I probably won't be able to test it 
much, but I can do it as soon as I get back home.

--
Trygve


From phil at bolthole.com  Tue Jul  7 05:08:13 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 20:08:13 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <4A52A649.1070606@opencsw.org>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
Message-ID: <20090707030813.GA20526@bolthole.com>

On Tue, Jul 07, 2009 at 03:35:05AM +0200, Trygve Laugst??l wrote:
>> 3) Provide Apache config files ready for including - where should the  
>> files be placed? Already in /opt/csw/apache2/etc/extra ?
>
> I would suggest /etc/opt/csw/munin/apache.conf and  
> /etc/opt/csw/munin/apache-cgi.conf

i would disagree.
these files are supposed to be read-only. so they are fine for /opt/csw
somewhere, and actually much better off there.


/opt/csw/(some-munin-path) would be just fine, so long as they are well
documented in /opt/csw/share/doc/munin/README  or something.



From bwalton at opencsw.org  Tue Jul  7 05:54:11 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 06 Jul 2009 23:54:11 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707030813.GA20526@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
Message-ID: <1246937677-sup-8061@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Mon Jul 06 23:08:13 -0400 2009:
> >> files be placed? Already in /opt/csw/apache2/etc/extra ?
> >
> > I would suggest /etc/opt/csw/munin/apache.conf and  
> > /etc/opt/csw/munin/apache-cgi.conf
> 
> i would disagree.
> these files are supposed to be read-only. so they are fine for /opt/csw
> somewhere, and actually much better off there.

...But they are part of the apache configuration, so it would make
sense to group them logically with the rest of the apache
configuration.  Personally though, I like the way debian does this.

An example: I pulled squirrelmail onto a box tonight (as a quick
example, phpmyadmin is another).  It added
/etc/squirrelmail/apache.conf.  The README.Debian.gz states:

<blockquote>
Apache
======

If you want to use squirrelmail with apache, it is recommended to
modify /etc/squirrelmail/apache.conf to your own wishes, and then add
a symlink to that file from /etc/apache{,-ssl,-perl,2}/conf.d, like:

 $ ln -s /etc/squirrelmail/apache.conf
 /etc/apache2/conf.d/squirrelmail.conf

Alternatively, you could also 'Include' the configuration file in the
relevant apache configuration file(s)
</blockquote>

The phpmyadmin package has a similarly worded README.Debian file.

It's configuration, whether commonly edited or not, so it should go in
etc somewhere.  Debian has a really nice apache setup and 'add on'
integration.  We follow their lead in other ways, why not here too?

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090706/bfd994e6/attachment-0002.asc>

From phil at bolthole.com  Tue Jul  7 06:45:35 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 6 Jul 2009 21:45:35 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246937677-sup-8061@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
Message-ID: <20090707044535.GA5845@bolthole.com>

On Mon, Jul 06, 2009 at 11:54:11PM -0400, Ben Walton wrote:
> ...
> ...But they are part of the apache configuration, so it would make
> sense to group them logically with the rest of the apache
> configuration.

it doesnt share well. This is something meant to be potentialy shared.
if there isnt a strong reason to keep something OUT of /opt/csw.. it
belongs under /opt/csw

or to put it another way.. if there isnt a strong reason to keep a file
machine-local, it belongs under/opt/csw

there is no reason to keep the read-only config as machine-local



>  Personally though, I like the way debian does this.

Hnmmm. not sure i like that way we can come up with something better
perhaps :)

	


From jeff at cjsa.com  Tue Jul  7 08:49:09 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Tue, 7 Jul 2009 06:49:09 GMT
Subject: [csw-maintainers] Library problem?
Message-ID: <KMEG9x.Cxz@cjsa.com>

I have just started noticing that something has changed on my system and
now a number of applications are displaying all text as rectangular boxes.
The rapidsvn application is one example.  This occurs even though these
applications have not changed, so the problem must be in a library.  My
best guess is that it might be glib2 or gtk2, but I actually know little
about them, so this is just a guess.  However, I have another app. that did
not come from CSW but is linked to these two libraries, so I think they may
be suspects.  Is anyone else seeing this problem, and if we can identify
the source, can we please get it fixed ASAP?

Regards,
-- 
Jeffery Small



From james at opencsw.org  Tue Jul  7 10:22:02 2009
From: james at opencsw.org (James Lee)
Date: Tue, 07 Jul 2009 08:22:02 GMT
Subject: [csw-maintainers] libiconv and --fast --xnolibmopt,
 ./configure hangs at "checking for stdbool.h that conforms to C99"
In-Reply-To: <4A52600A.6070803@opencsw.org>
References: <4A52199B.5070808@opencsw.org>
	<20090706161151.GA36831@bolthole.com> <4A525A9E.10306@opencsw.org>
	<20090706202956.GG60655@bolthole.com>
	<4A52600A.6070803@opencsw.org>
Message-ID: <20090707.8220200.4140658907@gyor.oxdrove.co.uk>

On 06/07/09, 21:35:22, Sebastian Kayser <skayser at opencsw.org> wrote 
regarding Re: [csw-maintainers] libiconv and --fast --xnolibmopt, 
./configure hangs at "checking for stdbool.h that conforms to C99":

> build8s is currently already being killed by something else (load
> average: 33.58, 33.62, 34.02),

It's build10s, the same physical machine, that needs some signal 9s.

In the meantime use build8s.go.opencsw.org




James.


From ja at opencsw.org  Tue Jul  7 13:10:25 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Tue, 07 Jul 2009 13:10:25 +0200
Subject: [csw-maintainers] csqlserver processes killed on buildfarm
Message-ID: <op.uwo2nnlcjbmpvo@ibm-1344209b5ff.ibm.com>

Hi all,

Dago asked me to kill all csqlserver processes on the buildfarm because  
the system was under heavy load and not very responsive.

Now everything should be ok again. Sorry, if someone's work got  
interrupted because of this.

Juergen

-- 
Juergen Arndt


From trygvis at opencsw.org  Tue Jul  7 13:32:22 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 07 Jul 2009 13:32:22 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707044535.GA5845@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>	<4A5246E0.80105@opencsw.org>
	<op.uwnzr4lqjbmpvo@imac.lan>	<4A52A649.1070606@opencsw.org>	<20090707030813.GA20526@bolthole.com>	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
Message-ID: <4A533246.30908@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 06, 2009 at 11:54:11PM -0400, Ben Walton wrote:
>> ...
>> ...But they are part of the apache configuration, so it would make
>> sense to group them logically with the rest of the apache
>> configuration.
> 
> it doesnt share well. This is something meant to be potentialy shared.
> if there isnt a strong reason to keep something OUT of /opt/csw.. it
> belongs under /opt/csw
> 
> or to put it another way.. if there isnt a strong reason to keep a file
> machine-local, it belongs under/opt/csw
> 
> there is no reason to keep the read-only config as machine-local

It's not read only but it will work unmodified in a lot of cases. One 
reason I see to modify it is to add security. A read-only .CSW copy is 
always nice to have so you can diff on upgrades.

>>  Personally though, I like the way debian does this.

I also find the Debian way very easy to work with.

--
Trygve


From bwalton at opencsw.org  Tue Jul  7 15:47:24 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 09:47:24 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707044535.GA5845@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
Message-ID: <1246973415-sup-1078@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 00:45:35 -0400 2009:
> > ...But they are part of the apache configuration, so it would make
> > sense to group them logically with the rest of the apache
> > configuration.
> 
> it doesnt share well. This is something meant to be potentialy shared.
> if there isnt a strong reason to keep something OUT of /opt/csw.. it
> belongs under /opt/csw

The munin package already ships files in /etc/opt/csw/munin though,
which are machine local configuration.  Why not keep things together?
If munin shipped a manifest that didn't touch /etc/ or /var on the
machine, your point would be better received on my end.  [It's still
well recieved, I just don't think it applies well to this particular
situation.]

> there is no reason to keep the read-only config as machine-local

But it's not 100% read only.  As Trygve mentioned, it may need to be
tuned to put security restrictions into it, etc.

> >  Personally though, I like the way debian does this.
> 
> Hnmmm. not sure i like that way we can come up with something better
> perhaps :)

It would be nice to have this extended into something like their
a2enmod/a2ensite setup (for modules and vhosts).  Placing the files in
$apacheconf/addons-available and then symlinking into
$apacheconf/addons-enabled would be nicer, I think.  If it were
decided that all shipped apache config snippets should live with the
rest of the apache config, this would get my vote.

Maybe one the following ground rules could be used?

1. Install all apache config files in the directory apache puts its
   own configuration in (most likely a sub-directory, for clarity).
2. If the package ships other machine local config, put the file with
   those, otherwise, somewhere in /opt/csw/etc?

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 mwatters at opencsw.org  Tue Jul  7 16:06:30 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Jul 2009 09:06:30 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <KMEG9x.Cxz@cjsa.com>
References: <KMEG9x.Cxz@cjsa.com>
Message-ID: <4A535666.4030009@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jeffery Small wrote:
> I have just started noticing that something has changed on my system and
> now a number of applications are displaying all text as rectangular boxes.
> The rapidsvn application is one example.  This occurs even though these
> applications have not changed, so the problem must be in a library.  My
> best guess is that it might be glib2 or gtk2, but I actually know little
> about them, so this is just a guess.  However, I have another app. that did
> not come from CSW but is linked to these two libraries, so I think they may
> be suspects.  Is anyone else seeing this problem, and if we can identify
> the source, can we please get it fixed ASAP?
> 
> Regards,

have you updated anything at all recently?

I can not be sure without more info, but my suspicion is you are missing
/opt/csw/etc/pango/pango.modules or the file contains the 64bit versions.


if you are running 1.24 you need to run
/opt/csw/bin/sparcv8/pango-querymodules >/opt/csw/etc/pango/pango.modules

for an earlier version you should run
/opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules

Please let us know if this does not fix the issue.


- --

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

iEYEARECAAYFAkpTVmUACgkQLrhmsXMSLxf6ZACg2FzCyExdMT/yiPlHXDA56TJS
e24AnibSo+HOwdMRBtraZDPQV7QavfqA
=w3LJ
-----END PGP SIGNATURE-----


From phil at bolthole.com  Tue Jul  7 16:25:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 07:25:17 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246973415-sup-1078@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
Message-ID: <20090707142516.GB11549@bolthole.com>

On Tue, Jul 07, 2009 at 09:47:24AM -0400, Ben Walton wrote:
> The munin package already ships files in /etc/opt/csw/munin though,
> which are machine local configuration.

ah ok then.





From phil at bolthole.com  Tue Jul  7 16:26:33 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 07:26:33 -0700
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A535666.4030009@opencsw.org>
References: <KMEG9x.Cxz@cjsa.com> <4A535666.4030009@opencsw.org>
Message-ID: <20090707142633.GC11549@bolthole.com>

On Tue, Jul 07, 2009 at 09:06:30AM -0500, Mike Watters wrote:
> if you are running 1.24 you need to run
> /opt/csw/bin/sparcv8/pango-querymodules >/opt/csw/etc/pango/pango.modules
> 
> for an earlier version you should run
> /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules
> 
> Please let us know if this does not fix the issue.

It would seem as though our existing users might be similarly plagued with
this sort of thing if they upgraded recently.

if so, this very nasty situation, warrants an email about it, to the
announce and user lists.


From bwalton at opencsw.org  Tue Jul  7 16:31:41 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 10:31:41 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707142516.GB11549@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
Message-ID: <1246976931-sup-3020@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 10:25:17 -0400 2009:
> > The munin package already ships files in /etc/opt/csw/munin though,
> > which are machine local configuration.
> 
> ah ok then.

But, which is the better approach for us, going forward.  There are a
lot of packages that could ship a similar apache snippet.  Read-only
/opt/csw issues as a separate issue, which do people feel is cleaner?
Lumping apache config snippets with apache, or sticking them with the
config for the package that ships it?  I think I could argue either
way.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/c0978a0e/attachment-0002.asc>

From william at wbonnet.net  Tue Jul  7 17:28:36 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 07 Jul 2009 17:28:36 +0200
Subject: [csw-maintainers] [csw-users] Kicad app broke with upgrade
	pkg-get -Uu
In-Reply-To: <4A536697.2010704@optonline.net>
References: <4A520D78.2070702@optonline.net>
	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>
	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>
	<4A527E0F.7040308@wbonnet.net> <4A536697.2010704@optonline.net>
Message-ID: <4A5369A4.6080002@wbonnet.net>

Hi Paul

>> Can you please send me the full output of the following commands :
>>
>> ldd /opt/csw/bin/kicad
>> pkginfo -l CSWkicad
Thanks, i was thinking at something but now i have the answer :)

> $ pkginfo -l CSWkicad
>   HOTLINE:  http://www.blastwave.org/bugtrack
>     EMAIL:  dlaigle at blastwave.org
You are using the package from Blastwave, not from OpenCSW. And this is 
not an error on your side.

You installed the package before the fork, and this package was not 
updated since. But... the libs it is using have been modified, and 
libxcb linking is no longer compatible between both project.

I suggest your try to remove then reinstall kicad. It should work.

Thanks for raising this problem. We will try to solve it in a "global way".

Cheers
W.


From phil at bolthole.com  Tue Jul  7 17:34:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 08:34:59 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246976931-sup-3020@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
Message-ID: <20090707153459.GL60655@bolthole.com>

On Tue, Jul 07, 2009 at 10:31:41AM -0400, Ben Walton wrote:
> But, which is the better approach for us, going forward.  There are a
> lot of packages that could ship a similar apache snippet.  Read-only
> /opt/csw issues as a separate issue, which do people feel is cleaner?
> Lumping apache config snippets with apache, or sticking them with the
> config for the package that ships it?  I think I could argue either
> way.

I think that, for sysadmins coming late to the scene, having an "include"
that is a long one, such as

include  /some/path/to/other/application/stuff

is actually more informative to have in the apache configs. it provides a
better "trail of breadcrumbs", to where the "other application" really
lives.



From william at wbonnet.net  Tue Jul  7 17:50:59 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 07 Jul 2009 17:50:59 +0200
Subject: [csw-maintainers] [csw-users] Kicad app broke with
 upgrade	pkg-get -Uu
In-Reply-To: <4A5369A4.6080002@wbonnet.net>
References: <4A520D78.2070702@optonline.net>	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>	<4A527E0F.7040308@wbonnet.net>
	<4A536697.2010704@optonline.net> <4A5369A4.6080002@wbonnet.net>
Message-ID: <4A536EE3.4000705@wbonnet.net>

Hi

> You are using the package from Blastwave, not from OpenCSW. And this 
> is not an error on your side.
>
> You installed the package before the fork, and this package was not 
> updated since. But... the libs it is using have been modified, and 
> libxcb linking is no longer compatible between both project.
>
> I suggest your try to remove then reinstall kicad. It should work.
The problem Paul raised may happen again with several package which have 
not been updated since we forked.

I'm affraid we will have several other packages with broken dependencies 
if we don't rebuild it. Maybe we could script the detection of such 
packages ?

cheers
W.


From bwalton at opencsw.org  Tue Jul  7 18:03:51 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:03:51 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707153459.GL60655@bolthole.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
Message-ID: <1246982619-sup-2337@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 11:34:59 -0400 2009:
> include  /some/path/to/other/application/stuff
> 
> is actually more informative to have in the apache configs. it provides a
> better "trail of breadcrumbs", to where the "other application" really
> lives.

I agree with this.  Shall we adopt 'the debian way' then?

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/3c8ef3f6/attachment-0002.asc>

From bwalton at opencsw.org  Tue Jul  7 18:06:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:06:36 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246982619-sup-2337@ntdws12.chass.utoronto.ca>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
	<4A5246E0.80105@opencsw.org> <op.uwnzr4lqjbmpvo@imac.lan>
	<4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
Message-ID: <1246982722-sup-4775@ntdws12.chass.utoronto.ca>

Excerpts from Ben Walton's message of Tue Jul 07 12:03:51 -0400 2009:
> I agree with this.  Shall we adopt 'the debian way' then?

Sorry, by 'debian way,' I mean: apache config snippets live with the
rest of the package's config files (be it in /etc/opt/csw/ or
/opt/csw/etc).

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/228a8b57/attachment-0002.asc>

From phil at bolthole.com  Tue Jul  7 18:08:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 09:08:06 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246982619-sup-2337@ntdws12.chass.utoronto.ca>
References: <op.uwnzr4lqjbmpvo@imac.lan> <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
Message-ID: <20090707160806.GQ60655@bolthole.com>

On Tue, Jul 07, 2009 at 12:03:51PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Tue Jul 07 11:34:59 -0400 2009:
> > include  /some/path/to/other/application/stuff
> > 
> > is actually more informative to have in the apache configs. it provides a
> > better "trail of breadcrumbs", to where the "other application" really
> > lives.
> 
> I agree with this.  Shall we adopt 'the debian way' then?

Errr... the "debian way", if I understood it directly, was to make
symlinks.
What I was saying, did not involve any symlinks.


From phil at bolthole.com  Tue Jul  7 18:11:43 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 09:11:43 -0700
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246982722-sup-4775@ntdws12.chass.utoronto.ca>
References: <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
	<1246982722-sup-4775@ntdws12.chass.utoronto.ca>
Message-ID: <20090707161143.GR60655@bolthole.com>

On Tue, Jul 07, 2009 at 12:06:36PM -0400, Ben Walton wrote:
> Excerpts from Ben Walton's message of Tue Jul 07 12:03:51 -0400 2009:
> > I agree with this.  Shall we adopt 'the debian way' then?
> 
> Sorry, by 'debian way,' I mean: apache config snippets live with the
> rest of the package's config files (be it in /etc/opt/csw/ or
> /opt/csw/etc).

if by "the package", you mean "the package that isnt apache itself", then
yes I agree.



From bwalton at opencsw.org  Tue Jul  7 18:16:20 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:16:20 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <20090707160806.GQ60655@bolthole.com>
References: <op.uwnzr4lqjbmpvo@imac.lan> <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
	<20090707160806.GQ60655@bolthole.com>
Message-ID: <1246983310-sup-582@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 07 12:08:06 -0400 2009:
> Errr... the "debian way", if I understood it directly, was to make
> symlinks.
> What I was saying, did not involve any symlinks.

That choice would still be up to the maintainer (whether we're
referring to debian or csw).  The key point is where the real file
lives, which I think we agree is with the rest of the config for the
package.  [Apologies for lack of clarity.]

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/a019a102/attachment-0002.asc>

From bwalton at opencsw.org  Tue Jul  7 18:17:19 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 12:17:19 -0400
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <1246983310-sup-582@ntdws12.chass.utoronto.ca>
References: <op.uwnzr4lqjbmpvo@imac.lan> <4A52A649.1070606@opencsw.org>
	<20090707030813.GA20526@bolthole.com>
	<1246937677-sup-8061@ntdws12.chass.utoronto.ca>
	<20090707044535.GA5845@bolthole.com>
	<1246973415-sup-1078@ntdws12.chass.utoronto.ca>
	<20090707142516.GB11549@bolthole.com>
	<1246976931-sup-3020@ntdws12.chass.utoronto.ca>
	<20090707153459.GL60655@bolthole.com>
	<1246982619-sup-2337@ntdws12.chass.utoronto.ca>
	<20090707160806.GQ60655@bolthole.com>
	<1246983310-sup-582@ntdws12.chass.utoronto.ca>
Message-ID: <1246983414-sup-8022@ntdws12.chass.utoronto.ca>

Excerpts from Ben Walton's message of Tue Jul 07 12:16:20 -0400 2009:
> That choice would still be up to the maintainer (whether we're

s/maintainer/administrator/

Jeezaloo, I need another coffee.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/a6847c97/attachment-0002.asc>

From mwatters at opencsw.org  Tue Jul  7 20:25:11 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Jul 2009 13:25:11 -0500
Subject: [csw-maintainers] [csw-users] Kicad app broke with
 upgrade	pkg-get -Uu
In-Reply-To: <4A536EE3.4000705@wbonnet.net>
References: <4A520D78.2070702@optonline.net>	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>	<4A527E0F.7040308@wbonnet.net>	<4A536697.2010704@optonline.net>
	<4A5369A4.6080002@wbonnet.net> <4A536EE3.4000705@wbonnet.net>
Message-ID: <4A539307.4010601@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi
> 
>> You are using the package from Blastwave, not from OpenCSW. And this
>> is not an error on your side.
>>
>> You installed the package before the fork, and this package was not
>> updated since. But... the libs it is using have been modified, and
>> libxcb linking is no longer compatible between both project.
>>
>> I suggest your try to remove then reinstall kicad. It should work.
> The problem Paul raised may happen again with several package which have
> not been updated since we forked.
> 
> I'm affraid we will have several other packages with broken dependencies
> if we don't rebuild it. Maybe we could script the detection of such
> packages ?
> 
> cheers
> W.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

if we file a "critical bug" against those packages, then my page to determine
retired maintainers and orphaned packages.[1]
could be modified to weight the packages by # of bugs or bug severity or
importance (somehow).  thoughts?




[1] http://www.opencsw.org/need_maintainer

- --

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

iEYEARECAAYFAkpTkwcACgkQLrhmsXMSLxczyQCffyTatB94U907s+FkY2mhHRCm
C9IAoMLe+QLsIy2/UhKxumdbMISvjyXu
=pbr9
-----END PGP SIGNATURE-----


From phil at bolthole.com  Tue Jul  7 20:31:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 11:31:56 -0700
Subject: [csw-maintainers] [csw-users] Kicad app broke with
	upgrade	pkg-get -Uu
In-Reply-To: <4A539307.4010601@opencsw.org>
References: <4A520D78.2070702@optonline.net> <4A520E2A.4010009@wbonnet.net>
	<4A5210E6.5070802@optonline.net> <4A52133F.5050505@wbonnet.net>
	<4A5224C9.6020503@optonline.net> <4A527E0F.7040308@wbonnet.net>
	<4A536697.2010704@optonline.net> <4A5369A4.6080002@wbonnet.net>
	<4A536EE3.4000705@wbonnet.net> <4A539307.4010601@opencsw.org>
Message-ID: <20090707183156.GS60655@bolthole.com>

On Tue, Jul 07, 2009 at 01:25:11PM -0500, Mike Watters wrote:
> if we file a "critical bug" against those packages, then my page to determine
> retired maintainers and orphaned packages.[1]
> could be modified to weight the packages by # of bugs or bug severity or
> importance (somehow).  thoughts?

There is a pre-existing "show bugs" page somewheres... I think Dago wrote
it.
it could probably be tweaked more eaasily for what you have in mind.


From skayser at opencsw.org  Tue Jul  7 20:49:15 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Tue, 07 Jul 2009 20:49:15 +0200
Subject: [csw-maintainers] [csw-users] Kicad app broke
 with	upgrade	pkg-get -Uu
In-Reply-To: <20090707183156.GS60655@bolthole.com>
References: <4A520D78.2070702@optonline.net>
	<4A520E2A.4010009@wbonnet.net>	<4A5210E6.5070802@optonline.net>
	<4A52133F.5050505@wbonnet.net>	<4A5224C9.6020503@optonline.net>
	<4A527E0F.7040308@wbonnet.net>	<4A536697.2010704@optonline.net>
	<4A5369A4.6080002@wbonnet.net>	<4A536EE3.4000705@wbonnet.net>
	<4A539307.4010601@opencsw.org>
	<20090707183156.GS60655@bolthole.com>
Message-ID: <4A5398AB.7060801@opencsw.org>

Philip Brown wrote:
> On Tue, Jul 07, 2009 at 01:25:11PM -0500, Mike Watters wrote:
>> if we file a "critical bug" against those packages, then my page to determine
>> retired maintainers and orphaned packages.[1]
>> could be modified to weight the packages by # of bugs or bug severity or
>> importance (somehow).  thoughts?
> 
> There is a pre-existing "show bugs" page somewheres... I think Dago wrote
> it.
> it could probably be tweaked more eaasily for what you have in mind.

Are you referring to this one?

  http://opencsw.org/buglist/buglist.cgi

Sebastian


From phil at bolthole.com  Tue Jul  7 20:58:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 11:58:26 -0700
Subject: [csw-maintainers] [csw-users] Kicad app broke with
	upgrade	pkg-get -Uu
In-Reply-To: <4A5398AB.7060801@opencsw.org>
References: <4A5210E6.5070802@optonline.net> <4A52133F.5050505@wbonnet.net>
	<4A5224C9.6020503@optonline.net> <4A527E0F.7040308@wbonnet.net>
	<4A536697.2010704@optonline.net> <4A5369A4.6080002@wbonnet.net>
	<4A536EE3.4000705@wbonnet.net> <4A539307.4010601@opencsw.org>
	<20090707183156.GS60655@bolthole.com>
	<4A5398AB.7060801@opencsw.org>
Message-ID: <20090707185826.GT60655@bolthole.com>

On Tue, Jul 07, 2009 at 08:49:15PM +0200, Sebastian Kayser wrote:
> Philip Brown wrote:
> > There is a pre-existing "show bugs" page somewheres... I think Dago wrote
> > it.
> > it could probably be tweaked more eaasily for what you have in mind.
> 
> Are you referring to this one?
> 
>   http://opencsw.org/buglist/buglist.cgi

yes.



From james at opencsw.org  Tue Jul  7 21:25:56 2009
From: james at opencsw.org (James Lee)
Date: Tue, 07 Jul 2009 19:25:56 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
Message-ID: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>

Anyone understand pkgconfig?  This looks easy but I'm blind as to why
it's not working:

build8s:/home/james$ export PKG_CONFIG_PATH=/opt/csw/lib/pkgconfig
build8s:/home/james$ if /opt/csw/bin/pkg-config --exists 'gtk+-2.0'
if> then
then>    echo YES
then> else
else>    echo NO
else> fi
sh: gnome-config: not found
NO
build8s:/home/james$ ls -l /opt/csw/lib/pkgconfig/gtk+-2.0.pc
-rw-r--r--   1 root     bin          345 Jun  7  2008
/opt/csw/lib/pkgconfig/gtk+-2.0.pc



There seems to be a dependency problem but I still can't work out what's
gone missing:

$ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
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



Help!  I'm close to editing "if true" in to my good friend the
configure file.




James.



From phil at bolthole.com  Tue Jul  7 21:35:55 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 12:35:55 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
Message-ID: <20090707193555.GU60655@bolthole.com>

On Tue, Jul 07, 2009 at 07:25:56PM +0000, James Lee wrote:
> Anyone understand pkgconfig?  This looks easy but I'm blind as to why
> it's not working:
> ....
> $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
> 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

seems fairly clear: xcb isnt installed on build8s.
or at least the pkgconfig files for it arent installed.

or actually... they are installed in /opt/csw/X11/lib/pkgconfig

Methinks cairo needs to be rebuilt or something, to give the "new and
proper" path to it or something. but otherwise, perhaps you can add that
dir to your pkgconfig path as first.


I hope our cairo/new-x11 maintainres can come up with a more transparent
solution to this stuff though.


From james at opencsw.org  Tue Jul  7 21:51:48 2009
From: james at opencsw.org (James Lee)
Date: Tue, 07 Jul 2009 19:51:48 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707193555.GU60655@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
Message-ID: <20090707.19514800.2647352656@gyor.oxdrove.co.uk>

On 07/07/09, 20:35:55, Philip Brown <phil at bolthole.com> wrote regarding Re: 
[csw-maintainers] "pkgconfig --exist ..." is telling porkies:

> On Tue, Jul 07, 2009 at 07:25:56PM +0000, James Lee wrote:
> > Anyone understand pkgconfig?  This looks easy but I'm blind as to why
> > it's not working:
> > ....
> > $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
> > 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

> seems fairly clear: xcb isnt installed on build8s.

Why does that clearly affect whether gtk+-2.0 exists which clearly
it does.  Even truss doesn't show it looking for xcb.  OK then, too
much time wasted, "if true" in the patch file for configure.




> or at least the pkgconfig files for it arent installed.

> or actually... they are installed in /opt/csw/X11/lib/pkgconfig

> Methinks cairo needs to be rebuilt or something, to give the "new and
> proper" path to it or something. but otherwise, perhaps you can add that
> dir to your pkgconfig path as first.

Which package needs the bug report?  Gtk+-2.0, cairo, xcb or pkgconfig?




James.


From mwatters at opencsw.org  Tue Jul  7 22:01:14 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 07 Jul 2009 15:01:14 -0500
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707.19514800.2647352656@gyor.oxdrove.co.uk>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
Message-ID: <4A53A98A.8060201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

James Lee wrote:
> On 07/07/09, 20:35:55, Philip Brown <phil at bolthole.com> wrote regarding Re: 
> [csw-maintainers] "pkgconfig --exist ..." is telling porkies:
> 
>> On Tue, Jul 07, 2009 at 07:25:56PM +0000, James Lee wrote:
>>> Anyone understand pkgconfig?  This looks easy but I'm blind as to why
>>> it's not working:
>>> ....
>>> $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
>>> 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
> 
>> seems fairly clear: xcb isnt installed on build8s.
> 
> Why does that clearly affect whether gtk+-2.0 exists which clearly
> it does.  Even truss doesn't show it looking for xcb.  OK then, too
> much time wasted, "if true" in the patch file for configure.
> 
> 
> 
> 
>> or at least the pkgconfig files for it arent installed.
> 
>> or actually... they are installed in /opt/csw/X11/lib/pkgconfig
> 
>> Methinks cairo needs to be rebuilt or something, to give the "new and
>> proper" path to it or something. but otherwise, perhaps you can add that
>> dir to your pkgconfig path as first.
> 
> Which package needs the bug report?  Gtk+-2.0, cairo, xcb or pkgconfig?

xcb and in fact all the "new" xlib packages have their PKG_CONFIG_PATH as
/opt/csw/X11/lib/pkgconfig

when doing anything that requires CSWx* packages you need to use something
along the lines of: EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib

I think pkg-config is walking down the dependency trail.  cairo is required for
gtk, but xcb is required for cairo. it couldn't find xcb so it quit instead of
spending an hour compiling just to die a horrible death when trying to link to
the missing xcb libraries.


- --

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

iEYEARECAAYFAkpTqYoACgkQLrhmsXMSLxdFZwCdG9doD+tJJ2lqnWdklOWa0dDz
sVMAoIPe9Vp5Fe1w15EFFHOT161OQdHN
=M/Ni
-----END PGP SIGNATURE-----


From william at wbonnet.net  Tue Jul  7 22:03:34 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 07 Jul 2009 22:03:34 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707193555.GU60655@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
Message-ID: <4A53AA16.1000302@wbonnet.net>

Hi

> I hope our cairo/new-x11 maintainres can come up with a more transparent
> solution to this stuff though.
>   
Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH

If you are using GAR add

EXTRA_PKG_CONFIG_PATH += /opt/csw/X11/lib/pkgconfig

to your Makefile

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  Tue Jul  7 22:22:54 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 7 Jul 2009 13:22:54 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4A53A98A.8060201@opencsw.org>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
Message-ID: <20090707202254.GV60655@bolthole.com>

On Tue, Jul 07, 2009 at 03:01:14PM -0500, Mike Watters wrote:
> > Which package needs the bug report?  Gtk+-2.0, cairo, xcb or pkgconfig?
> 
> xcb and in fact all the "new" xlib packages have their PKG_CONFIG_PATH as
> /opt/csw/X11/lib/pkgconfig
> 
> when doing anything that requires CSWx* packages you need to use something
> along the lines of: EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib

idealy though... our cairo package should "do the right thing" somehow, in
my opinion.
after all, that's what pkg-config is for.. to take all the by-hand
messiness out of compiling?

I dont know much about pkg-config, but might there be some kind of,
"and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can put in
/opt/csw/lib/pkgconfig/cairo.pc  or something?

or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
should not exist there in the first place, but be moved
to /opt/csw/lib/X11/lib/pkgconfig/   instead?

I dunno.


From william at wbonnet.net  Wed Jul  8 01:00:10 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 01:00:10 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
Message-ID: <4A53D37A.40502@wbonnet.net>

Hi

I do have the same problem with the /opt/csw/include/jri.h file. It 
comes from the same packages and also says it has to be provided by 
netscape.

Could someone please rename it to .old on build9x ?

thanks in advance

cheers
W.

> Hi William,
>
> Am 01.07.2009 um 14:32 schrieb William Bonnet:
>> A conflicting header is installed by qt package. It is the file 
>> "/opt/csw/include/npapi.h"
>>
>> When including this file, it generates the following error
>>
>> "/opt/csw/include/npapi.h", line 1: Error: #error 
>> "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
>> must be provided by Netscape".
>>
>> A mantis entry exists, but unfortnatly package has no maintainer and 
>> was not fixed. Would it be possible to nuke this file on the build 
>> farm please ? (well at least rename the file to .old...). This file 
>> is blocking for compile FF3.
>>
>> npapi.h is provided by mozilla packages, especialy firefox, not by qt.
>>
>> [wbonnet at build9x:~/mgar/pkg/firefox/branches/firefox3.5-packaging]$ 
>> grep npapi.h /var/sadm/install/contents
>> /opt/csw/include/npapi.h f none 0644 root bin 225 20260 1091716908 CSWqt
>> /opt/csw/kde-gcc/include/npapi.h f none 0644 root bin 225 20273 
>> 1110647274 CSWqtgcc
>
> The file does not include anything of interest:
>
> dam at build9x :/home/dam > more /opt/csw/include/npapi.h
> #error 
> "/home/espenr/tmp/qt-3-espenr-26122/qt/extensions/nsplugin/src/npapi.h 
> must be provided by Netscape"
> #error "This file is just a placeholder. Please see the documentation"
> #error "to learn how to obtain the real file"
>
> Moved to npapi.h.old on build9x only for now.
>
>
> Best regards
>
>   -- Dago
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>
>


-- 
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 Jul  8 02:37:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 07 Jul 2009 20:37:15 -0400
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A53D37A.40502@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
	<4A53D37A.40502@wbonnet.net>
Message-ID: <1247013402-sup-5213@ntdws12.chass.utoronto.ca>

Excerpts from William Bonnet's message of Tue Jul 07 19:00:10 -0400 2009:
> I do have the same problem with the /opt/csw/include/jri.h file. It 
> comes from the same packages and also says it has to be provided by 
> netscape.
> 
> Could someone please rename it to .old on build9x ?

Doing this now.
-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090707/d57094d6/attachment-0002.asc>

From jeff at cjsa.com  Wed Jul  8 08:49:56 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Wed, 8 Jul 2009 06:49:56 GMT
Subject: [csw-maintainers] Library problem?
References: <KMEG9x.Cxz@cjsa.com> <4A535666.4030009@opencsw.org>
	<20090707142633.GC11549@bolthole.com>
Message-ID: <KMGAz8.2Ir@cjsa.com>

Philip Brown <phil at bolthole.com> writes:

>On Tue, Jul 07, 2009 at 09:06:30AM -0500, Mike Watters wrote:
>> if you are running 1.24 you need to run
>> /opt/csw/bin/sparcv8/pango-querymodules >/opt/csw/etc/pango/pango.modules
>> 
>> for an earlier version you should run
>> /opt/csw/bin/pango-querymodules >/opt/csw/etc/pango/pango.modules
>> 
>> Please let us know if this does not fix the issue.

>It would seem as though our existing users might be similarly plagued with
>this sort of thing if they upgraded recently.

>if so, this very nasty situation, warrants an email about it, to the
>announce and user lists.

Thanks to everyone who responded to my query.  Yes, apparently the problem
was with the pango.modules file and after running the above command (the
binaries under sparc and sparcv8 are the same) my problem was resolved.

I am using the latest 1.24.3,REV=2009.07.01 version of libpango and I
typically update once a week or so.  I do not know exactly when the last
update was to libpango on my system.  Does this look like a general bug in
the post-install script that failed to automatically regenerate the
module list?

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040



From bonivart at opencsw.org  Wed Jul  8 10:33:37 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Wed, 8 Jul 2009 10:33:37 +0200
Subject: [csw-maintainers] /testing BIND 9.6.1
Message-ID: <625385e30907080133i54dbf31cn45ba86da04773252@mail.gmail.com>

Latest BIND DNS Server in testing:

http://mirror.opencsw.org/testing/bind-9.6.1,REV=2009.07.07-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind-9.6.1,REV=2009.07.07-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind_utils-9.6.1,REV=2009.07.07-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind_utils-9.6.1,REV=2009.07.07-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libbind-9.6.1,REV=2009.07.07-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/libbind-9.6.1,REV=2009.07.07-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/bind_devel-9.6.1,REV=2009.07.07-SunOS5.8-all-CSW.pkg.gz

# pkgutil -t http://mirror.opencsw.org/opencsw/testing -i bind

-- 
/peter


From james at opencsw.org  Wed Jul  8 10:45:58 2009
From: james at opencsw.org (James Lee)
Date: Wed, 08 Jul 2009 08:45:58 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4A53AA16.1000302@wbonnet.net>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<4A53AA16.1000302@wbonnet.net>
Message-ID: <20090708.8455800.3441876416@gyor.oxdrove.co.uk>

On 07/07/09, 21:03:34, William Bonnet <william at wbonnet.net> wrote regarding
Re: [csw-maintainers] "pkgconfig --exist ..." is telling porkies:

> Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH

Yes, that helps, thanks.  My blind spot was the .pc files are in odd
places and that it needs things it really doesn't.

The CSW pkgconf should know where to look for the CSW pc files, ie
have this path coded in as the default.  The point of pkgconf is to
tell us paths, if I have to tell it the paths before it works it
defeats its point.  Doing a scan I see the CSW pc files are all over
the place.  It's a mess.



James.


From james at opencsw.org  Wed Jul  8 11:22:48 2009
From: james at opencsw.org (James Lee)
Date: Wed, 08 Jul 2009 09:22:48 GMT
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4A53AA16.1000302@wbonnet.net>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<4A53AA16.1000302@wbonnet.net>
Message-ID: <20090708.9224800.775122963@gyor.oxdrove.co.uk>

On 07/07/09, 21:03:34, William Bonnet <william at wbonnet.net> wrote regarding 
Re: [csw-maintainers] "pkgconfig --exist ..." is telling porkies:

> > I hope our cairo/new-x11 maintainres can come up with a more transparent
> > solution to this stuff though.
> >
> Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH

There is another problem with the CSW system, eg:

$ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
sh: gnome-config: not found
Package gobject-2.0 was not found in the pkg-config search path.
Perhaps you should add the directory containing `gobject-2.0.pc'
to the PKG_CONFIG_PATH environment variable
Package 'gobject-2.0', required by 'GdkPixbuf', not found



gtk+-2.0.pc is part of CSWgtk2.  gobject-2.0.pc is needed by CSWgtk2's
gtk+-2.0.pc and is part of CSWglib2devel but it's not a depend so I
don't have it installed.  I used to have gobject-2.0.pc as part of
CSWglib2 but it was moved to CSWglib2devel and not added CSWglib2 as
a depend.  In the past when we have split packages we have added the
parts to the main package to ensure continuity.




James.






From william at wbonnet.net  Wed Jul  8 13:31:11 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 13:31:11 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <KMGAz8.2Ir@cjsa.com>
References: <KMEG9x.Cxz@cjsa.com>
	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>
	<KMGAz8.2Ir@cjsa.com>
Message-ID: <4A54837F.9090202@wbonnet.net>

Hi

> Thanks to everyone who responded to my query.  Yes, apparently the problem
> was with the pango.modules file and after running the above command (the
> binaries under sparc and sparcv8 are the same) my problem was resolved.
>
> I am using the latest 1.24.3,REV=2009.07.01 version of libpango and I
> typically update once a week or so.  I do not know exactly when the last
> update was to libpango on my system.  Does this look like a general bug in
> the post-install script that failed to automatically regenerate the
> module list?
>   
I was able to re produce the same problem on a 64 bit VM and thunderbird.

Is it possible that the problem comes from the fact a 32 bit application 
(thunderbird) is running on a 64 bit machine, when loading pango modules 
it retrieves the 64 bits modules from the pango.modules config file and 
this produces a warning / error.

If this is the case i don't really see how having two different version 
of the executable used to generate this file may solve this.

Any clue is welcome :)

cheers
W.


From mwatters at opencsw.org  Wed Jul  8 15:29:27 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 08:29:27 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A54837F.9090202@wbonnet.net>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<KMGAz8.2Ir@cjsa.com>
	<4A54837F.9090202@wbonnet.net>
Message-ID: <4A549F37.50701@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi
> 
>> Thanks to everyone who responded to my query.  Yes, apparently the
>> problem
>> was with the pango.modules file and after running the above command (the
>> binaries under sparc and sparcv8 are the same) my problem was resolved.
>>
>> I am using the latest 1.24.3,REV=2009.07.01 version of libpango and I
>> typically update once a week or so.  I do not know exactly when the last
>> update was to libpango on my system.  Does this look like a general
>> bug in
>> the post-install script that failed to automatically regenerate the
>> module list?
>>   
> I was able to re produce the same problem on a 64 bit VM and thunderbird.
> 
> Is it possible that the problem comes from the fact a 32 bit application
> (thunderbird) is running on a 64 bit machine, when loading pango modules
> it retrieves the 64 bits modules from the pango.modules config file and
> this produces a warning / error.
> 
> If this is the case i don't really see how having two different version
> of the executable used to generate this file may solve this.
> 
> Any clue is welcome :)
> 
> cheers
> W.
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

This unfortunately is a known bug in this version
http://www.opencsw.org/mantis/view.php?id=3750

I think I saw some threads bouncing back and forth with regard to the best way
to correct this issue.  the ticket above has a temporary fix in the notes from
Dago.


- --

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

iEYEARECAAYFAkpUnzcACgkQLrhmsXMSLxfBOQCffkTbICPvjRnJnb9Vs4l0VAUh
yyMAn0IT+HSDm/VmcoonfTkBjvtkyCTw
=kE7U
-----END PGP SIGNATURE-----


From maciej at opencsw.org  Wed Jul  8 18:55:21 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 8 Jul 2009 17:55:21 +0100
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
Message-ID: <ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>

On Wed, Jun 24, 2009 at 8:36 PM, Sebastian Kayser<skayser at opencsw.org> wrote:
> I guess it doesn't fully cover your answer, but regarding GAR, you can set
> GARFLAVOR to DBG if you want your binaries built with debugging symbols.
> This will also skip the step where GAR strips all the binaries. See [1].

cups makefiles do honor CFLAGS, so GARFLAVOR = DBG, I can see that the
flag -g is present in compiler invocations.

Using ALL_CFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include -I/usr/include/kerberosv5 -g
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include  -DOPENSSL_DISABLE_OLD_DES_SUPPORT
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
-D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
Using ALL_CXXFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include -g
-I/export/home/blizinski/build.5.10-i386/opt/csw/include
-I/opt/csw/include -DOPENSSL_DISABLE_OLD_DES_SUPPORT
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
-D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
Using LDFLAGS=-L../cups -L../filter
-L/export/home/blizinski/build.5.10-i386/opt/csw/lib -L/opt/csw/lib
-R/opt/csw/lib   -w -KPIC -xO2
Using LIBS=-lcups -lssl -lcrypto -lpthread -lresolv -lnsl -lsocket -lm

The binary is in fact not stripped:

$ file cups-polld
cups-polld:     ELF 32-bit LSB executable 80386 Version 1 [FPU],
dynamically linked, not stripped

However, it doesn't seem to be compiled without optimization: when I
run it under the Sun Studio debugger, and display the call stack, I
see:

main(argc = ???, argv = ???) (optimized)

Is there a way to check whether the binary is actually optimized
during compilation?

Maciej


From mwatters at opencsw.org  Wed Jul  8 19:11:04 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 12:11:04 -0500
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
Message-ID: <4A54D328.40709@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Maciej (Matchek) Blizinski wrote:
> On Wed, Jun 24, 2009 at 8:36 PM, Sebastian Kayser<skayser at opencsw.org> wrote:
>> I guess it doesn't fully cover your answer, but regarding GAR, you can set
>> GARFLAVOR to DBG if you want your binaries built with debugging symbols.
>> This will also skip the step where GAR strips all the binaries. See [1].
> 
> cups makefiles do honor CFLAGS, so GARFLAVOR = DBG, I can see that the
> flag -g is present in compiler invocations.
> 
> Using ALL_CFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include -I/usr/include/kerberosv5 -g
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include  -DOPENSSL_DISABLE_OLD_DES_SUPPORT
> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
> Using ALL_CXXFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include -g
> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
> -I/opt/csw/include -DOPENSSL_DISABLE_OLD_DES_SUPPORT
> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
> Using LDFLAGS=-L../cups -L../filter
> -L/export/home/blizinski/build.5.10-i386/opt/csw/lib -L/opt/csw/lib
> -R/opt/csw/lib   -w -KPIC -xO2
                             ^^^
the -xO2 is the optimization flag here... remove it and you should remove the
optimization.

> Using LIBS=-lcups -lssl -lcrypto -lpthread -lresolv -lnsl -lsocket -lm



- --

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

iEYEARECAAYFAkpU0ygACgkQLrhmsXMSLxeTbQCaAyUg6LtZinw//Q2UIJWSE/9M
LLkAn2fQ5razR8Pd9oRCYERfxdAFb/nf
=C1NP
-----END PGP SIGNATURE-----


From skayser at opencsw.org  Wed Jul  8 19:28:56 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 8 Jul 2009 19:28:56 +0200 (CEST)
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <4A54D328.40709@opencsw.org>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
Message-ID: <61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>

Mike Watters wrote:
> Maciej (Matchek) Blizinski wrote:
>> On Wed, Jun 24, 2009 at 8:36 PM, Sebastian Kayser<skayser at opencsw.org>
>> wrote:
>>> I guess it doesn't fully cover your answer, but regarding GAR, you can
>>> set
>>> GARFLAVOR to DBG if you want your binaries built with debugging
>>> symbols.
>>> This will also skip the step where GAR strips all the binaries. See
>>> [1].
>>
>> cups makefiles do honor CFLAGS, so GARFLAVOR = DBG, I can see that the
>> flag -g is present in compiler invocations.
>>
>> Using ALL_CFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include -I/usr/include/kerberosv5 -g
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include  -DOPENSSL_DISABLE_OLD_DES_SUPPORT
>> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
>> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
>> Using ALL_CXXFLAGS=-I.. -D_CUPS_SOURCE -I/usr/include/kerberosv5
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include -g
>> -I/export/home/blizinski/build.5.10-i386/opt/csw/include
>> -I/opt/csw/include -DOPENSSL_DISABLE_OLD_DES_SUPPORT
>> -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
>> -D_THREAD_SAFE -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
>> Using LDFLAGS=-L../cups -L../filter
>> -L/export/home/blizinski/build.5.10-i386/opt/csw/lib -L/opt/csw/lib
>> -R/opt/csw/lib   -w -KPIC -xO2
>                              ^^^
> the -xO2 is the optimization flag here... remove it and you should remove
> the
> optimization.

And just for the sake of completeness: GAR doesn't add such optimization
flags when you set GARFLAVOR=DBG. Just give

  GARFLAVOR=DBG gmake modenv

a try and GAR will tell you what flags it will use. This doesn't mean that
GAR will filter out such flags for you when you have set them manually
elsewhere, but it won't produce them on its own.

Sebastian



From william at wbonnet.net  Wed Jul  8 19:41:25 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 19:41:25 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <1247013402-sup-5213@ntdws12.chass.utoronto.ca>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>
	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
Message-ID: <4A54DA45.4090902@wbonnet.net>

Hi Ben
> Excerpts from William Bonnet's message of Tue Jul 07 19:00:10 -0400 2009:
>   
>> I do have the same problem with the /opt/csw/include/jri.h file. It 
>> comes from the same packages and also says it has to be provided by 
>> netscape.
>>
>> Could someone please rename it to .old on build9x ?
>>     
>
> Doing this now.
>   
Thanks

It worked, i was able to compile FF3 on solaris 9, and the binaries runs on solaris 10. Now i have to find tests machines running solaris 8 and 9 :)

I also need the same fix on build9x 8x and 9x. I better look to produce a Qt patch :)


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  Wed Jul  8 22:04:08 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 22:04:08 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <20090707142633.GC11549@bolthole.com>
References: <KMEG9x.Cxz@cjsa.com> <4A535666.4030009@opencsw.org>
	<20090707142633.GC11549@bolthole.com>
Message-ID: <4A54FBB8.4010808@wbonnet.net>

Hi

> It would seem as though our existing users might be similarly plagued with
> this sort of thing if they upgraded recently.
>
> if so, this very nasty situation, warrants an email about it, to the
> announce and user lists.
>   
Based on Trygve advices i made a few tests and had a look to the way 
sun's gnome desktop is working.  I run a Ultra 20, thus an amd64 CPU. In 
the Solaris installation there is both 32 and 64 bits versions of pango. 
Here are the files i have :

64 bits version

[william at cyaegha:/usr/lib/amd64]$ ls *pango*
libpango-1.0.so          libpangoft2-1.0.so          
libpangox-1.0.so          libpangoxft-1.0.so
libpango-1.0.so.0        libpangoft2-1.0.so.0        
libpangox-1.0.so.0        libpangoxft-1.0.so.0
libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1  
libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1

pango:
1.4.0

[william at cyaegha:/usr/lib/amd64]$ ls pango/1.4.0/modules/
pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so  
pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so

32 bits version

[william at cyaegha:/usr/lib]$ ls *pango*
libpango-1.0.so          libpangoft2-1.0.so          
libpangox-1.0.so          libpangoxft-1.0.so
libpango-1.0.so.0        libpangoft2-1.0.so.0        
libpangox-1.0.so.0        libpangoxft-1.0.so.0
libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1  
libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1

pango:
1.4.0

[william at cyaegha:/usr/lib]$ ls pango/1.4.0/modules/
pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so  
pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so

By default /etc/pango/pango.modules contains the 32 bits modules, even 
on my amd64 version.

I made the following tests running both 32 bits and 64 bits version of 
gtk-demo (/usr/bin/gtk-demo and /usr/bin/amd64/gtk-demo).

. Running application gtk-demo and pango.modules using same versions 
(both 32 or both 64 bits) work is fine.

. Running 64bits apps with 32 bits modules works also, but running 32 
bits app with 64 bits modules does not work.

Please notice that from Sun's isntallation running both 64bits, or 
64bits apps with 32bits modules produce the same warnings

     (gtk-demo:19058): Gtk-WARNING **: Unable to locate theme engine in 
module_path: "blueprint",

Running a full 32bits stack produce no warnings



So i may suggest to do some more testing, like recompiling a couple of 
our desktop app in 64 bits mode, and test these apps against 32 bits 
modules. If it work fine, we may generate by default a pango.module 
config file containing 32 bits modules.

Because :

1/ it works :)

2/ most of the desktop app are compiled only in 32 bits

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  Wed Jul  8 23:11:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 16:11:50 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A54FBB8.4010808@wbonnet.net>
References: <KMEG9x.Cxz@cjsa.com>
	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>
	<4A54FBB8.4010808@wbonnet.net>
Message-ID: <4A550B96.9050604@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi
> 
>> It would seem as though our existing users might be similarly plagued
>> with
>> this sort of thing if they upgraded recently.
>>
>> if so, this very nasty situation, warrants an email about it, to the
>> announce and user lists.
>>   
> Based on Trygve advices i made a few tests and had a look to the way
> sun's gnome desktop is working.  I run a Ultra 20, thus an amd64 CPU. In
> the Solaris installation there is both 32 and 64 bits versions of pango.
> Here are the files i have :
> 
> 64 bits version
> 
> [william at cyaegha:/usr/lib/amd64]$ ls *pango*
> libpango-1.0.so          libpangoft2-1.0.so         
> libpangox-1.0.so          libpangoxft-1.0.so
> libpango-1.0.so.0        libpangoft2-1.0.so.0       
> libpangox-1.0.so.0        libpangoxft-1.0.so.0
> libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1 
> libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1
> 
> pango:
> 1.4.0
> 
> [william at cyaegha:/usr/lib/amd64]$ ls pango/1.4.0/modules/
> pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so 
> pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so
> 
> 32 bits version
> 
> [william at cyaegha:/usr/lib]$ ls *pango*
> libpango-1.0.so          libpangoft2-1.0.so         
> libpangox-1.0.so          libpangoxft-1.0.so
> libpango-1.0.so.0        libpangoft2-1.0.so.0       
> libpangox-1.0.so.0        libpangoxft-1.0.so.0
> libpango-1.0.so.0.501.1  libpangoft2-1.0.so.0.501.1 
> libpangox-1.0.so.0.501.1  libpangoxft-1.0.so.0.501.1
> 
> pango:
> 1.4.0
> 
> [william at cyaegha:/usr/lib]$ ls pango/1.4.0/modules/
> pango-arabic-fc.so  pango-basic-fc.so  pango-basic-x.so 
> pango-hangul-fc.so  pango-hebrew-fc.so  pango-indic-fc.so  pango-thai-fc.so
> 
> By default /etc/pango/pango.modules contains the 32 bits modules, even
> on my amd64 version.
> 
> I made the following tests running both 32 bits and 64 bits version of
> gtk-demo (/usr/bin/gtk-demo and /usr/bin/amd64/gtk-demo).
> 
> . Running application gtk-demo and pango.modules using same versions
> (both 32 or both 64 bits) work is fine.
> 
> . Running 64bits apps with 32 bits modules works also, but running 32
> bits app with 64 bits modules does not work.
> 
> Please notice that from Sun's isntallation running both 64bits, or
> 64bits apps with 32bits modules produce the same warnings
> 
>     (gtk-demo:19058): Gtk-WARNING **: Unable to locate theme engine in
> module_path: "blueprint",
> 
> Running a full 32bits stack produce no warnings
> 
> 
> 
> So i may suggest to do some more testing, like recompiling a couple of
> our desktop app in 64 bits mode, and test these apps against 32 bits
> modules. If it work fine, we may generate by default a pango.module
> config file containing 32 bits modules.
> 
> Because :
> 
> 1/ it works :)
> 
> 2/ most of the desktop app are compiled only in 32 bits
> 
> cheers
> W.
> 
> 

I believe the running proposal is to remove the isaexec from pango, and have
the default pango.modules file containing the 32bit.

there was some discussion on having a /opt/csw/etc/64/pango.modules file
which would be fine with me, AFAIK there is currently nothing in CSW that uses
the 64bit pango libs. ( as they never existed before ) so we can patch the
64bit versions to use the correct modules file.

I believe I saw some messages that Dago is on Holiday?


- --

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

iEYEARECAAYFAkpVC5YACgkQLrhmsXMSLxftaQCfccWs8VOp5ZFYtdz/Y6r6CeSV
xe8An0qHVjDu+T8t1lBKs7q1lES9IfhT
=FEmQ
-----END PGP SIGNATURE-----


From william at wbonnet.net  Wed Jul  8 23:23:57 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 08 Jul 2009 23:23:57 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A550B96.9050604@opencsw.org>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<4A54FBB8.4010808@wbonnet.net>
	<4A550B96.9050604@opencsw.org>
Message-ID: <4A550E6D.70600@wbonnet.net>

Hi Mike

> I believe I saw some messages that Dago is on Holiday?
>   
And without internet access... poor Dago, it should be really difficult ;)

I think he will be back tomorrow or friday.

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  Wed Jul  8 23:28:11 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 08 Jul 2009 16:28:11 -0500
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A550E6D.70600@wbonnet.net>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<4A54FBB8.4010808@wbonnet.net>	<4A550B96.9050604@opencsw.org>
	<4A550E6D.70600@wbonnet.net>
Message-ID: <4A550F6B.6040508@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

William Bonnet wrote:
> Hi Mike
> 
>> I believe I saw some messages that Dago is on Holiday?
>>   
> And without internet access... poor Dago, it should be really difficult ;)
> 
> I think he will be back tomorrow or friday.
> 
> cheers
> W.
> 

How can one live without internet access?


- --

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

iEYEARECAAYFAkpVD2sACgkQLrhmsXMSLxemYwCgkVDasQfAdrbZaBk6EvfI/RAX
mXgAoJMo+6f5IgCJLRfhcZa6TYBGWQl1
=r/60
-----END PGP SIGNATURE-----


From maciej at opencsw.org  Thu Jul  9 12:22:18 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 11:22:18 +0100
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
Message-ID: <ce97543b0907090322j3b6463a7ye57f4d9d5165a0ba@mail.gmail.com>

On Sat, Jul 4, 2009 at 6:01 PM, rupert THURNER<rupert at opencsw.org> wrote:
> when doing the mercurial tests, there is a warning.
>
> gpg: Warning: using insecure memory!
>  --> http://www.gnupg.org/faq.html#q6.1
>
> the faq says that it should be installed with setuid root. is this
> deliberately different on opencsw?

Routing to Mike Watters, the gnupg package maintainer.

It would be a relatively simple change:

Index: Makefile
===================================================================
--- Makefile    (revision 5052)
+++ Makefile    (working copy)
@@ -40,4 +40,8 @@

 TEST_TARGET = check

+PROTOTYPE_FILTER  = awk ' \
+    $$$$3 ~ /\/opt\/csw\/bin\/gpg$$$$/ { $$$$4 = "4755" } \
+    { print }'
+
 include gar/category.mk

Maciej


From maciej at opencsw.org  Thu Jul  9 12:23:02 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 11:23:02 +0100
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <ce97543b0907090322j3b6463a7ye57f4d9d5165a0ba@mail.gmail.com>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
	<ce97543b0907090322j3b6463a7ye57f4d9d5165a0ba@mail.gmail.com>
Message-ID: <ce97543b0907090323l64483774o8c229578978af70f@mail.gmail.com>

[+mwatters]

On Thu, Jul 9, 2009 at 11:22 AM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> On Sat, Jul 4, 2009 at 6:01 PM, rupert THURNER<rupert at opencsw.org> wrote:
>> when doing the mercurial tests, there is a warning.
>>
>> gpg: Warning: using insecure memory!
>> ?--> http://www.gnupg.org/faq.html#q6.1
>>
>> the faq says that it should be installed with setuid root. is this
>> deliberately different on opencsw?
>
> Routing to Mike Watters, the gnupg package maintainer.
>
> It would be a relatively simple change:
>
> Index: Makefile
> ===================================================================
> --- Makefile ? ?(revision 5052)
> +++ Makefile ? ?(working copy)
> @@ -40,4 +40,8 @@
>
> ?TEST_TARGET = check
>
> +PROTOTYPE_FILTER ?= awk ' \
> + ? ?$$$$3 ~ /\/opt\/csw\/bin\/gpg$$$$/ { $$$$4 = "4755" } \
> + ? ?{ print }'
> +
> ?include gar/category.mk
>
> Maciej
>


From pfelecan at opencsw.org  Thu Jul  9 14:34:11 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 09 Jul 2009 14:34:11 +0200
Subject: [csw-maintainers] common package in the catalog
Message-ID: <x68wiy2fy4.fsf@abhavana.free.fr>

I'm observing something unusual, at least for me, in the catalog of
SPARC systems: the corresponding entry points toward an uncompressed
file:

on a SPARC system:
common 1.4.6,REV=2008.04.28 CSWcommon
common-1.4.6,REV=2008.04.28-SunOS5.8-sparc-CSW.pkg
d0e61f7542b3a4fbff7de4a7437bb6af 23552 none none

on an Intel system:
common 1.4.6,REV=2008.04.28 CSWcommon
common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
638cac6d1d6d1dba3db8ab76fbd36bb2 3632 none none

This is a new thing as some weeks ago both catalogs referenced the
common package as compressed.
-- 
Peter


From bonivart at opencsw.org  Thu Jul  9 15:00:33 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 9 Jul 2009 15:00:33 +0200
Subject: [csw-maintainers] common package in the catalog
In-Reply-To: <x68wiy2fy4.fsf@abhavana.free.fr>
References: <x68wiy2fy4.fsf@abhavana.free.fr>
Message-ID: <625385e30907090600r72a984ccrfbfa20ec139aa45b@mail.gmail.com>

On Thu, Jul 9, 2009 at 2:34 PM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> I'm observing something unusual, at least for me, in the catalog of
> SPARC systems: the corresponding entry points toward an uncompressed
> file:
>
> on a SPARC system:
> common 1.4.6,REV=2008.04.28 CSWcommon
> common-1.4.6,REV=2008.04.28-SunOS5.8-sparc-CSW.pkg
> d0e61f7542b3a4fbff7de4a7437bb6af 23552 none none
>
> on an Intel system:
> common 1.4.6,REV=2008.04.28 CSWcommon
> common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> 638cac6d1d6d1dba3db8ab76fbd36bb2 3632 none none
>
> This is a new thing as some weeks ago both catalogs referenced the
> common package as compressed.

This is a bug, both should be uncompressed since CSWgzip depends on
common. This should have been taken care of a while back but maybe
only for sparc..?

-- 
/peter


From Darin.Perusich at cognigencorp.com  Thu Jul  9 17:37:03 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Thu, 09 Jul 2009 11:37:03 -0400
Subject: [csw-maintainers] setting apache2 cmdline options
Message-ID: <4A560E9F.8000800@cognigencorp.com>

This probably belongs on the users list but I figure I'll pose my
question here instead. How are people going about setting command line
options for apache2 httpd processes? If I specify <IfDefine name>
directives in my httpd.conf how would one go about setting -Dname on the
command line without modifying the manifest? You can't.

I'm asking because I was setting up an SSL server and I'm using the
certificate for multiple apps, stunnel and apache, and I've placed it in
/opt/csw/etc/ssl/certs/ the global location for certificates and when I
started apache it didn't load mod_ssl. After digging around for
documentation on this, which is non-existent, I looked at the manifest
and it only loads -DSSL if $APACHE_HOME/etc/server.crt exists. Well,
what if that isn't were my certificate lives or is named? Have a
configuration file that can be sourced to set command line options would
be ideal and is really the only solution I can see.

Yes I know I can remove the <IfDefine SSL> or touch server.crt but those
are band-aids.

What you peoples thoughts on this?

-- 
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  Thu Jul  9 17:45:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 08:45:45 -0700
Subject: [csw-maintainers] common package in the catalog
In-Reply-To: <625385e30907090600r72a984ccrfbfa20ec139aa45b@mail.gmail.com>
References: <x68wiy2fy4.fsf@abhavana.free.fr>
	<625385e30907090600r72a984ccrfbfa20ec139aa45b@mail.gmail.com>
Message-ID: <20090709154545.GA75736@bolthole.com>

On Thu, Jul 09, 2009 at 03:00:33PM +0200, Peter Bonivart wrote:
> This is a bug, both should be uncompressed since CSWgzip depends on
> common. This should have been taken care of a while back but maybe
> only for sparc..?

oops. guess so.
pushing out unzippped now


From maciej at opencsw.org  Thu Jul  9 17:53:48 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 16:53:48 +0100
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
	<61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
Message-ID: <ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>

One more question: does anyone know how to debug a number of binaries
at once? In my case, there's a suspicion that the problem has
something to do with the multitude of cups-polld daemons. I can't
trigger the issue with just one daemon; I need to run several at the
same time. So far, Sun Studio gives me an impression that it can only
debug one binary at the time. Perhaps there's a way to launch a number
of binaries and attach to one binary at the time?

Maciej


From maciej at opencsw.org  Thu Jul  9 17:55:03 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 9 Jul 2009 16:55:03 +0100
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
	<61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
	<ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
Message-ID: <ce97543b0907090855g22887b29p5b97251c6d1deaa@mail.gmail.com>

On Thu, Jul 9, 2009 at 4:53 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> same time. So far, Sun Studio gives me an impression that it can only
> debug one binary at the time. Perhaps there's a way to launch a number
> of binaries and attach to one binary at the time?

A terminology cleanup: I meant, a number of processes, the same
binary, multiple times, with different options.

Maciej


From phil at bolthole.com  Thu Jul  9 17:59:22 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 08:59:22 -0700
Subject: [csw-maintainers] Debugging with Sun Studio and mGAR
In-Reply-To: <ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
References: <ce97543b0906241007h3357b31dsfbc7bc8ed18dc713@mail.gmail.com>
	<54753.194.246.122.22.1245872203.squirrel@ssl.skayser.de>
	<ce97543b0907080955t57d3784clcf42dcefa54c20bc@mail.gmail.com>
	<4A54D328.40709@opencsw.org>
	<61092.194.246.122.22.1247074136.squirrel@ssl.skayser.de>
	<ce97543b0907090853t5c8ecfb4lee73bf5fc2550711@mail.gmail.com>
Message-ID: <20090709155922.GA8843@bolthole.com>

On Thu, Jul 09, 2009 at 04:53:48PM +0100, Maciej (Matchek) Blizinski wrote:
> One more question: does anyone know how to debug a number of binaries
> at once? In my case, there's a suspicion that the problem has
> something to do with the multitude of cups-polld daemons. I can't
> trigger the issue with just one daemon; I need to run several at the
> same time. So far, Sun Studio gives me an impression that it can only
> debug one binary at the time. Perhaps there's a way to launch a number
> of binaries and attach to one binary at the time?

use multiple windows, and multiple dbx processes? :-}


From phil at bolthole.com  Thu Jul  9 18:04:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 09:04:50 -0700
Subject: [csw-maintainers] setting apache2 cmdline options
In-Reply-To: <4A560E9F.8000800@cognigencorp.com>
References: <4A560E9F.8000800@cognigencorp.com>
Message-ID: <20090709160450.GB8843@bolthole.com>

On Thu, Jul 09, 2009 at 11:37:03AM -0400, Darin Perusich wrote:
>...
>  After digging around for
> documentation on this, which is non-existent, I looked at the manifest
> and it only loads -DSSL if $APACHE_HOME/etc/server.crt exists. Well,
> what if that isn't were my certificate lives or is named?

then make a symlink from there to the real location?

> Yes I know I can remove the <IfDefine SSL> or touch server.crt but those
> are band-aids.
> 
> What you peoples thoughts on this?

I think that using a package made and put together by someone else,
sometimes means that you need to follow someone else's conventions.
(ie: making the symlink).

and if you think there are better conventions for the maintainer to follow,
you should suggest it to them.

As far as #defines go for apache... I dont think they are meant to be used
like makefile defines, whereeever and however you want. They're mostly just
to be used as a convenience for if a module is defined, in my opinion.
The SSL thing is an oddity, and I've never liked how they have done that.
I think it should be handled more like the others. ie:

<IfModule ssl>

rather than
<IFDefine SSL>

but in the absense of that.. you need some external trigger to let the
startup script know, "hey, i want ssl enabled". and the presense of the
server key file, seems like a reasonable trigger to me.


From dam at opencsw.org  Thu Jul  9 20:56:05 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 20:56:05 +0200
Subject: [csw-maintainers] =?iso-8859-1?q?Additional_Farm_Maintainer=3A_J?=
	=?iso-8859-1?q?=FCrgen_Arndt?=
Message-ID: <06DF27E0-FF15-455A-9040-5A5C5160F8AC@opencsw.org>

Hi,

I'm back from my vacation and noticed that there are obviously not
enough farm managers as Ben wanted to take some time off, too. I have
given J?rgen full admin access to the farm. Ihsan, please add "ja"
to the list of buildfarm maintainers.


Best regards

   -- Dago

From dam at opencsw.org  Thu Jul  9 21:49:38 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 21:49:38 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <4A4F8D65.9080203@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
Message-ID: <EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>

Hi Sebastian,

Am 04.07.2009 um 19:12 schrieb Sebastian Kayser:
> Dagobert Michelsen wrote:
>> Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
>>> while working on postfix 2.6 i have noticed again that it is not  
>>> quite
>>> straight-forward to assign files to classes. Instead of exposing the
>>> inner
>>> workings of the prototype stuff to the Makefile, could we maybe get
>>> some
>>> variables to assign files to classes (functions)?
>>
>> For CSWcswclassutils this is already in there :-) Just define
>>
>> SAMPLECONF
>> PRESERVECONF
>> INITSMF
>>
>> and everything else will be taken cared of.
>
> Sweet!!! :D Can we add something for USERGROUP also?
>
> Index: /home/skayser/mgar/gar/v2/gar.pkg.mk
> ===================================================================
> --- /home/skayser/mgar/gar/v2/gar.pkg.mk        (revision 5361)
> +++ /home/skayser/mgar/gar/v2/gar.pkg.mk        (working copy)
> @@ -155,12 +155,14 @@
> SPKG_CLASSES := $(SPKG_CLASSES) $(if $(SAMPLECONF),$(if $(filter  
> cswsampleconf,$(SPKG_CLASSES)),,cswsampleconf))
> SPKG_CLASSES := $(SPKG_CLASSES) $(if $(PRESERVECONF),$(if $(filter  
> cswpreserveconf,$(SPKG_CLASSES)),,cswpreserveconf))
> SPKG_CLASSES := $(SPKG_CLASSES) $(if $(INITSMF),$(if $(filter  
> cswinitsmf,$(SPKG_CLASSES)),,cswinitsmf))
> +SPKG_CLASSES := $(SPKG_CLASSES) $(if $(USERGROUP),$(if $(filter  
> cswusergroup,$(SPKG_CLASSES)),,cswusergroup))
> # - set class for all config files
> ifneq ($(SAMPLECONF)$(PRESERVECONF)$(INITSMF),)
> _CSWCLASS_FILTER = | perl -ane '\
>                $(foreach FILE,$(SAMPLECONF),$$F[1] = "cswsampleconf"  
> if ( $$F[2] =~ m(^$(FILE)$$) );)\
>                $(foreach FILE,$(PRESERVECONF),$$F[1] =  
> "cswpreserveconf" if( $$F[2] =~ m(^$(FILE)$$) );)\
>                $(foreach FILE,$(INITSMF),$$F[1] = "cswinitsmf" if( $ 
> $F[2] =~ m(^$(FILE)$$) );)\
> +               $(foreach FILE,$(USERGROUP),$$F[1] = "cswusergroup"  
> if( $$F[2] =~ m(^$(FILE)$$) );)\
>                print join(" ", at F),"\n";'
> _EXTRA_GAR_PKGS += CSWcswclassutils
> endif

Looks good. Please go ahead and commit.


Best regards

   -- Dago


From dam at opencsw.org  Thu Jul  9 22:51:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 22:51:15 +0200
Subject: [csw-maintainers] Newpkgs after testing/: pango
Message-ID: <9770FEF1-E44C-4547-A649-AFE85BCDAE30@opencsw.org>

Hi,

there is an updated version of pango in testing/ and newpkgs/:

libpango-1.24.3,REV=2009.07.09-SunOS5.8-i386-CSW.pkg.gz
libpango-1.24.3,REV=2009.07.09-SunOS5.8-sparc-CSW.pkg.gz
libpango_devel-1.24.3,REV=2009.07.09-SunOS5.8-i386-CSW.pkg.gz
libpango_devel-1.24.3,REV=2009.07.09-SunOS5.8-sparc-CSW.pkg.gz
libpango_doc-1.24.3,REV=2009.07.09-SunOS5.8-all-CSW.pkg.gz

The query-bindary is no longer isaexec and there two files
produced now during postinstall:
   /opt/csw/etc/pango/pango.modules
   /opt/csw/etc/64/pango/pango.modules

Please someone other then me verifies that it fully solves the
issue and Phil can release it ASAP.


Best regards

   -- Dago


From mwatters at opencsw.org  Thu Jul  9 23:13:18 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 09 Jul 2009 16:13:18 -0500
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <ce97543b0907090302l17245204k48f8c6e3fa9b09f3@mail.gmail.com>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
	<ce97543b0907090302l17245204k48f8c6e3fa9b09f3@mail.gmail.com>
Message-ID: <4A565D6E.9050201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Maciej (Matchek) Blizinski wrote:
> On Sat, Jul 4, 2009 at 6:01 PM, rupert THURNER<rupert at opencsw.org> wrote:
>> when doing the mercurial tests, there is a warning.
>>
>> gpg: Warning: using insecure memory!
>>  --> http://www.gnupg.org/faq.html#q6.1
>>
>> the faq says that it should be installed with setuid root. is this
>> deliberately different on opencsw?
> 
> Routing to Mike Watters, the gnupg package maintainer.
> 
> It would be a relatively simple change:
> 
> Index: Makefile
> ===================================================================
> --- Makefile    (revision 5052)
> +++ Makefile    (working copy)
> @@ -40,4 +40,8 @@
> 
>  TEST_TARGET = check
> 
> +PROTOTYPE_FILTER  = awk ' \
> +    $$$$3 ~ /\/opt\/csw\/bin\/gpg$$$$/ { $$$$4 = "4755" } \
> +    { print }'
> +
>  include gar/category.mk
> 
> Maciej

I will make the change the next time I release the package.
please file a feature request bug so I don't forget about it ;-}



- --

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

iEYEARECAAYFAkpWXW4ACgkQLrhmsXMSLxfxlgCgpujVgtqAvOHc7xjteyjZoge7
WB8AniFfPWHADQGF1864eRYWOn28QXbK
=+fcb
-----END PGP SIGNATURE-----


From dam at opencsw.org  Thu Jul  9 23:16:45 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:16:45 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <KMEG9x.Cxz@cjsa.com>
References: <KMEG9x.Cxz@cjsa.com>
Message-ID: <41C24380-8952-4A5E-BFC9-2AC385F9E9A9@opencsw.org>

Hi Jeffery,

Am 07.07.2009 um 08:49 schrieb Jeffery Small:
> I have just started noticing that something has changed on my system  
> and
> now a number of applications are displaying all text as rectangular  
> boxes.
> The rapidsvn application is one example.  This occurs even though  
> these
> applications have not changed, so the problem must be in a library.   
> My
> best guess is that it might be glib2 or gtk2, but I actually know  
> little
> about them, so this is just a guess.  However, I have another app.  
> that did
> not come from CSW but is linked to these two libraries, so I think  
> they may
> be suspects.  Is anyone else seeing this problem, and if we can  
> identify
> the source, can we please get it fixed ASAP?

I guess that was my fault. Unfortunately I was on holidays just these
few days the error occurred :-( Please see if the version in testing/
from pango fixes the issue:
   http://mirror.opencsw.org/testing.html
After the correct fixing has been confirmed the packages can be pushed
out ASAP.


Thanks!

   -- Dago


From dam at opencsw.org  Thu Jul  9 23:27:35 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:27:35 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090707202254.GV60655@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
Message-ID: <4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>

Hi Phil,

Am 07.07.2009 um 22:22 schrieb Philip Brown:
> On Tue, Jul 07, 2009 at 03:01:14PM -0500, Mike Watters wrote:
>>> Which package needs the bug report?  Gtk+-2.0, cairo, xcb or  
>>> pkgconfig?
>>
>> xcb and in fact all the "new" xlib packages have their  
>> PKG_CONFIG_PATH as
>> /opt/csw/X11/lib/pkgconfig
>>
>> when doing anything that requires CSWx* packages you need to use  
>> something
>> along the lines of: EXTRA_PKG_CONFIG_DIRS = $(prefix)/X11/lib
>
> idealy though... our cairo package should "do the right thing"  
> somehow, in
> my opinion.
> after all, that's what pkg-config is for.. to take all the by-hand
> messiness out of compiling?

Yes, but you must tell pkg-config which libraries to use. pkg-config
is for setting flags, not finding the libs.

> I dont know much about pkg-config, but might there be some kind of,
> "and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can  
> put in
> /opt/csw/lib/pkgconfig/cairo.pc  or something?

AFAIK this option is not available.

> or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
> should not exist there in the first place, but be moved
> to /opt/csw/lib/X11/lib/pkgconfig/   instead?

It is just for development. Using pkg-config this way allows specific
binding to either the OpenCSW X11 libs or Sun X11 libs with gtk+ on
compilation. This separation was intended when we introduced /opt/csw/ 
X11.
Is this case, however, it is not useful. As it is only relevant in
compilation and integration is well understood and easy in both manual
and GAR builds it shouldn't do too much harm, though.


Best regards

   -- Dago


From dam at opencsw.org  Thu Jul  9 23:29:23 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:29:23 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090708.9224800.775122963@gyor.oxdrove.co.uk>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<4A53AA16.1000302@wbonnet.net>
	<20090708.9224800.775122963@gyor.oxdrove.co.uk>
Message-ID: <2117C253-063E-4E51-BC5B-64915CE239E9@opencsw.org>

Hi James,

Am 08.07.2009 um 11:22 schrieb James Lee:
> On 07/07/09, 21:03:34, William Bonnet <william at wbonnet.net> wrote  
> regarding
> Re: [csw-maintainers] "pkgconfig --exist ..." is telling porkies:
>
>>> I hope our cairo/new-x11 maintainres can come up with a more  
>>> transparent
>>> solution to this stuff though.
>>>
>> Please also add  /opt/csw/X11/lib/pkgconfig to your  PKG_CONFIG_PATH
>
> There is another problem with the CSW system, eg:
>
> $ /opt/csw/bin/pkg-config --libs 'gtk+-2.0'
> sh: gnome-config: not found
> Package gobject-2.0 was not found in the pkg-config search path.
> Perhaps you should add the directory containing `gobject-2.0.pc'
> to the PKG_CONFIG_PATH environment variable
> Package 'gobject-2.0', required by 'GdkPixbuf', not found
>
> gtk+-2.0.pc is part of CSWgtk2.  gobject-2.0.pc is needed by CSWgtk2's
> gtk+-2.0.pc and is part of CSWglib2devel but it's not a depend so I
> don't have it installed.  I used to have gobject-2.0.pc as part of
> CSWglib2 but it was moved to CSWglib2devel and not added CSWglib2 as
> a depend.  In the past when we have split packages we have added the
> parts to the main package to ensure continuity.

This is a very good point. The devel-packages should not only
rely on the base package, but the devel-packages for the base-
packages. Please file a bug report for these so they are not
forgotten.


Best regards

   -- Dago


From phil at bolthole.com  Thu Jul  9 23:35:25 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 14:35:25 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
	<4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
Message-ID: <20090709213525.GE8843@bolthole.com>

On Thu, Jul 09, 2009 at 11:27:35PM +0200, Dagobert Michelsen wrote:
> Hi Phil,
>
> Am 07.07.2009 um 22:22 schrieb Philip Brown:
>> I dont know much about pkg-config, but might there be some kind of,
>> "and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can  
>> put in
>> /opt/csw/lib/pkgconfig/cairo.pc  or something?
>
> AFAIK this option is not available.
>

Then in my opinion, since libcairo requires libxcb now, then 
I think my suggestion here, would be best:


>> or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
>> should not exist there in the first place, but be moved
>> to /opt/csw/lib/X11/lib/pkgconfig/   instead?


>[Dago writes in reply to that...]
> It is just for development. Using pkg-config this way allows specific
> binding to either the OpenCSW X11 libs or Sun X11 libs with gtk+ on
> compilation.

I do not think that is a good (ie: stable) idea at all.
If people want to use "sun libs", then they should probably use sun cairo!
on the other hand, seeing as how our libcairo now REQUIRES the newer x11
libs... libs that we were "forced" to ship ourselves... it doesnt make
sense to break smooth opencsw development, for the fuzzy benefit of
lalowing pepole to compile brokenly against sun libs as well.



From dam at opencsw.org  Thu Jul  9 23:38:59 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:38:59 +0200
Subject: [csw-maintainers] Library problem?
In-Reply-To: <4A550F6B.6040508@opencsw.org>
References: <KMEG9x.Cxz@cjsa.com>	<4A535666.4030009@opencsw.org>	<20090707142633.GC11549@bolthole.com>	<4A54FBB8.4010808@wbonnet.net>	<4A550B96.9050604@opencsw.org>
	<4A550E6D.70600@wbonnet.net> <4A550F6B.6040508@opencsw.org>
Message-ID: <3E42BEFE-7FAC-4550-94C3-AD5B89DBFA8E@opencsw.org>

Hi Mike,

Am 08.07.2009 um 23:28 schrieb Mike Watters:
> William Bonnet wrote:
>> Hi Mike
>>
>>> I believe I saw some messages that Dago is on Holiday?
>>>
>> And without internet access... poor Dago, it should be really  
>> difficult ;)
>>
>> I think he will be back tomorrow or friday.
>>
>> cheers
>> W.
>>
>
> How can one live without internet access?

It was hard ;-) And when I got home my trusty old Dual-G5 Mac
was dead - mainboard defect :-(( I didn't feel much relaxed
after noticing *that*.


Best regards

   -- Dago


From dam at opencsw.org  Thu Jul  9 23:47:47 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:47:47 +0200
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <20090709213525.GE8843@bolthole.com>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
	<4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
	<20090709213525.GE8843@bolthole.com>
Message-ID: <E49405BA-0940-486C-BB8F-F1595C1C51E4@opencsw.org>

Hi Phil,

Am 09.07.2009 um 23:35 schrieb Philip Brown:

> On Thu, Jul 09, 2009 at 11:27:35PM +0200, Dagobert Michelsen wrote:
>> Hi Phil,
>>
>> Am 07.07.2009 um 22:22 schrieb Philip Brown:
>>> I dont know much about pkg-config, but might there be some kind of,
>>> "and hey add /opt/csw/X11/lib to PKG_CONFIG_PATH now" magic, you can
>>> put in
>>> /opt/csw/lib/pkgconfig/cairo.pc  or something?
>>
>> AFAIK this option is not available.
>>
>
> Then in my opinion, since libcairo requires libxcb now, then
> I think my suggestion here, would be best:
>
>
>>> or... maybe the issue is that /opt/csw/lib/pkgconfig/cairo.pc
>>> should not exist there in the first place, but be moved
>>> to /opt/csw/lib/X11/lib/pkgconfig/   instead?
>
>
>> [Dago writes in reply to that...]
>> It is just for development. Using pkg-config this way allows specific
>> binding to either the OpenCSW X11 libs or Sun X11 libs with gtk+ on
>> compilation.
>
> I do not think that is a good (ie: stable) idea at all.
> If people want to use "sun libs", then they should probably use sun  
> cairo!
> on the other hand, seeing as how our libcairo now REQUIRES the newer  
> x11
> libs... libs that we were "forced" to ship ourselves... it doesnt make
> sense to break smooth opencsw development, for the fuzzy benefit of
> lalowing pepole to compile brokenly against sun libs as well.

To make it nice and clean the cairo shared libs should also go
into /opt/csw/X11/lib. Having pkg-config and libs in separate
places is IMHO even more unelegant than extending PKG_CONFIG_PATH.


Best regards

   -- Dago


From dam at opencsw.org  Thu Jul  9 23:51:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 9 Jul 2009 23:51:51 +0200
Subject: [csw-maintainers] Please repost open issues
Message-ID: <C7FD443D-3922-447D-87D6-094B7504E3D5@opencsw.org>

Hi,

there was a considerable backlog of messages to process during
my absence. I hope to have adressed all issues now (Rupert:
I stored your report about v2-pbuild malfunction for further
inspection without mail reply yet). However, due to the size
I may have missed a thing.

If you have posted issues I have not addressed yet please
repost them!


Best regards

   -- Dago


From phil at bolthole.com  Thu Jul  9 23:53:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 9 Jul 2009 14:53:31 -0700
Subject: [csw-maintainers] "pkgconfig --exist ..." is telling porkies
In-Reply-To: <E49405BA-0940-486C-BB8F-F1595C1C51E4@opencsw.org>
References: <20090707.19255600.1109300882@gyor.oxdrove.co.uk>
	<20090707193555.GU60655@bolthole.com>
	<20090707.19514800.2647352656@gyor.oxdrove.co.uk>
	<4A53A98A.8060201@opencsw.org>
	<20090707202254.GV60655@bolthole.com>
	<4CC80A59-ADF3-4D74-984C-9BAFA05E2B4E@opencsw.org>
	<20090709213525.GE8843@bolthole.com>
	<E49405BA-0940-486C-BB8F-F1595C1C51E4@opencsw.org>
Message-ID: <20090709215331.GF8843@bolthole.com>

On Thu, Jul 09, 2009 at 11:47:47PM +0200, Dagobert Michelsen wrote:
>...
> To make it nice and clean the cairo shared libs should also go
> into /opt/csw/X11/lib. Having pkg-config and libs in separate
> places is IMHO even more unelegant than extending PKG_CONFIG_PATH.

Hmmm....

initially makes sense to me.... except that if you carry that to its
logical extension, that would meant that EVERYTHING that depends on cairo,
etc,  would then also end up in /opt/csw/X11/lib.
That's a lot of libs.

Not sure that's so grand.


From pfelecan at opencsw.org  Fri Jul 10 12:22:26 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 12:22:26 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
Message-ID: <x6zlbc25y5.fsf@abhavana.free.fr>

pkg-get download common
Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
--2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
Resolving proxy... 192.168.1.6
Connecting to proxy|192.168.1.6|:3128... connected.
Proxy request sent, awaiting response... 404 Not Found
2009-07-10 12:21:01 ERROR 404: Not Found.

Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
(Perhaps you need to update your catalog?)
rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz': No such file or directory

Maybe this is a more general issue...
-- 
Peter


From dam at opencsw.org  Fri Jul 10 12:36:10 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 10 Jul 2009 12:36:10 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6zlbc25y5.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
Message-ID: <EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>

Hi Peter,

Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
> pkg-get download common
> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> Resolving proxy... 192.168.1.6
> Connecting to proxy|192.168.1.6|:3128... connected.
> Proxy request sent, awaiting response... 404 Not Found
> 2009-07-10 12:21:01 ERROR 404: Not Found.
>
> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> (Perhaps you need to update your catalog?)
> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386- 
> CSW.pkg.gz': No such file or directory
>
> Maybe this is a more general issue...

No, there was an error that the common package was gzipped and gzip
depended on it, so common is no longer stored gzipped. Just update
your catalog and everything should be fine again.


Best regards

   -- Dago


From pfelecan at opencsw.org  Fri Jul 10 13:29:37 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 13:29:37 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 10 Jul 2009 12\:36\:10 +0200")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>
Message-ID: <x6r5wo22u6.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
>> pkg-get download common
>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> Resolving proxy... 192.168.1.6
>> Connecting to proxy|192.168.1.6|:3128... connected.
>> Proxy request sent, awaiting response... 404 Not Found
>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>>
>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> (Perhaps you need to update your catalog?)
>> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386- 
>> CSW.pkg.gz': No such file or directory
>>
>> Maybe this is a more general issue...
>
> No, there was an error that the common package was gzipped and gzip
> depended on it, so common is no longer stored gzipped. Just update
> your catalog and everything should be fine again.

Read more carefully the subject line and you'll understand that this is
something that I cannot correct.
-- 
Peter


From dam at opencsw.org  Fri Jul 10 15:26:13 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 10 Jul 2009 15:26:13 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6r5wo22u6.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>
	<x6r5wo22u6.fsf@abhavana.free.fr>
Message-ID: <5EF7EE94-D633-4F7D-A68C-E8389FC84991@opencsw.org>

Hi Peter,

Am 10.07.2009 um 13:29 schrieb Peter FELECAN:
> Dagobert Michelsen <dam at opencsw.org> writes:
>> Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
>>> pkg-get download common
>>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>> Resolving proxy... 192.168.1.6
>>> Connecting to proxy|192.168.1.6|:3128... connected.
>>> Proxy request sent, awaiting response... 404 Not Found
>>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>>>
>>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>> (Perhaps you need to update your catalog?)
>>> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386-
>>> CSW.pkg.gz': No such file or directory
>>>
>>> Maybe this is a more general issue...
>>
>> No, there was an error that the common package was gzipped and gzip
>> depended on it, so common is no longer stored gzipped. Just update
>> your catalog and everything should be fine again.
>
> Read more carefully the subject line and you'll understand that this  
> is
> something that I cannot correct.

Aha, the catalog is only updatable by root. Should be working now.
However, the catalogs on the buildfarm are only updated from time
to time. It should IMHO be possible for pkg-get (and pkgutil btw.)
to download without any need for root access. Or have I gotten it
wrong again?


Best regards

   -- Dago


From phil at bolthole.com  Fri Jul 10 16:11:48 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 10 Jul 2009 07:11:48 -0700
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6zlbc25y5.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
Message-ID: <20090710141148.GA99639@bolthole.com>

On Fri, Jul 10, 2009 at 12:22:26PM +0200, Peter FELECAN wrote:
> pkg-get download common
> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> Resolving proxy... 192.168.1.6
> Connecting to proxy|192.168.1.6|:3128... connected.
> Proxy request sent, awaiting response... 404 Not Found
> 2009-07-10 12:21:01 ERROR 404: Not Found.

> 
> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
> (Perhaps you need to update your catalog?)

So... "perhaps you need to update your catalog"?

the master catalog points to a .pkg file, not a .pkg.gz file.

if that's what you were referring to?

(and perhaps your proxy has cached the old catalog file or something)

The current live catalog seems ok to me.


From pfelecan at opencsw.org  Fri Jul 10 16:34:43 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 16:34:43 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <5EF7EE94-D633-4F7D-A68C-E8389FC84991@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 10 Jul 2009 15\:26\:13 +0200")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<EDB17F58-941F-4C9C-9D38-E15DBEA90647@opencsw.org>
	<x6r5wo22u6.fsf@abhavana.free.fr>
	<5EF7EE94-D633-4F7D-A68C-E8389FC84991@opencsw.org>
Message-ID: <x6my7c1u9o.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 10.07.2009 um 13:29 schrieb Peter FELECAN:
>> Dagobert Michelsen <dam at opencsw.org> writes:
>>> Am 10.07.2009 um 12:22 schrieb Peter FELECAN:
>>>> pkg-get download common
>>>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>>> Resolving proxy... 192.168.1.6
>>>> Connecting to proxy|192.168.1.6|:3128... connected.
>>>> Proxy request sent, awaiting response... 404 Not Found
>>>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>>>>
>>>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>>>> (Perhaps you need to update your catalog?)
>>>> rm: cannot remove `common-1.4.6,REV=2008.04.28-SunOS5.8-i386-
>>>> CSW.pkg.gz': No such file or directory
>>>>
>>>> Maybe this is a more general issue...
>>>
>>> No, there was an error that the common package was gzipped and gzip
>>> depended on it, so common is no longer stored gzipped. Just update
>>> your catalog and everything should be fine again.
>>
>> Read more carefully the subject line and you'll understand that this
>> is
>> something that I cannot correct.
>
> Aha, the catalog is only updatable by root. Should be working now.
> However, the catalogs on the buildfarm are only updated from time
> to time. It should IMHO be possible for pkg-get (and pkgutil btw.)
> to download without any need for root access. Or have I gotten it
> wrong again?

The download is possible by a non root user --- this is what I'm doing.

I confirm that you get it now correctly and the catalog on build8x is
now correct.

Thank you for this intervention.
-- 
Peter


From pfelecan at opencsw.org  Fri Jul 10 16:35:51 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 16:35:51 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <20090710141148.GA99639@bolthole.com> (Philip Brown's message of
	"Fri\, 10 Jul 2009 07\:11\:48 -0700")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
Message-ID: <x6iqi01u7s.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Fri, Jul 10, 2009 at 12:22:26PM +0200, Peter FELECAN wrote:
>> pkg-get download common
>> Trying http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> --2009-07-10 12:21:01--  http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> Resolving proxy... 192.168.1.6
>> Connecting to proxy|192.168.1.6|:3128... connected.
>> Proxy request sent, awaiting response... 404 Not Found
>> 2009-07-10 12:21:01 ERROR 404: Not Found.
>
>> 
>> Error downloading http://mirror/opencsw/current/i386/5.8/common-1.4.6,REV=2008.04.28-SunOS5.8-i386-CSW.pkg.gz
>> (Perhaps you need to update your catalog?)
>
> So... "perhaps you need to update your catalog"?
>
> the master catalog points to a .pkg file, not a .pkg.gz file.
>
> if that's what you were referring to?

No. As stated in the subject, the catalog for build8x, which resides in
/var/pkg-get/catalog-mirror wasn't up to date.

> (and perhaps your proxy has cached the old catalog file or something)
>
> The current live catalog seems ok to me.

It is.
-- 
Peter


From phil at bolthole.com  Fri Jul 10 17:04:46 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 10 Jul 2009 08:04:46 -0700
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6iqi01u7s.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
Message-ID: <20090710150446.GA75987@bolthole.com>

On Fri, Jul 10, 2009 at 04:35:51PM +0200, Peter FELECAN wrote:
>....
> No. As stated in the subject, the catalog for build8x, which resides in
> /var/pkg-get/catalog-mirror wasn't up to date.
> 
> Philip Brown <phil at bolthole.com> writes:
> > (and perhaps your proxy has cached the old catalog file or something)
> >
> > The current live catalog seems ok to me.
> 
> It is.
> -- 

i'm confused what "fixed" this then.


From dam at opencsw.org  Fri Jul 10 17:08:54 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 10 Jul 2009 17:08:54 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <20090710150446.GA75987@bolthole.com>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
Message-ID: <163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>

Hi Phil,

Am 10.07.2009 um 17:04 schrieb Philip Brown:
> On Fri, Jul 10, 2009 at 04:35:51PM +0200, Peter FELECAN wrote:
>> ....
>> No. As stated in the subject, the catalog for build8x, which  
>> resides in
>> /var/pkg-get/catalog-mirror wasn't up to date.
>>
>> Philip Brown <phil at bolthole.com> writes:
>>> (and perhaps your proxy has cached the old catalog file or  
>>> something)
>>>
>>> The current live catalog seems ok to me.
>>
>> It is.
>> -- 
>
> i'm confused what "fixed" this then.

- The catalog on build8x was old
- There was no package available for the "old" information in the  
outdated catalog
   when Peter tried to download the package
- The catalog can only be updated by root
- Peter is not root
- I updated the catalog
- Everything worked


Best regards

   -- Dago


From phil at bolthole.com  Fri Jul 10 18:17:24 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 10 Jul 2009 09:17:24 -0700
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
	<163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
Message-ID: <20090710161724.GA73109@bolthole.com>

On Fri, Jul 10, 2009 at 05:08:54PM +0200, Dagobert Michelsen wrote:
> - The catalog on build8x was old

oh... this was a build-machine-specific problem.

pfft. he should have mailed the buildfarm list not the maintainers list
then! :-P



From maciej at opencsw.org  Fri Jul 10 19:00:31 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 10 Jul 2009 18:00:31 +0100
Subject: [csw-maintainers] gpg: Warning: using insecure memory!
In-Reply-To: <4A565D6E.9050201@opencsw.org>
References: <6af4270907041001i7bba4e72j2ee2834eec793dda@mail.gmail.com>
	<ce97543b0907090302l17245204k48f8c6e3fa9b09f3@mail.gmail.com>
	<4A565D6E.9050201@opencsw.org>
Message-ID: <ce97543b0907101000i4f44fe42r308f5213edeb224@mail.gmail.com>

On Thu, Jul 9, 2009 at 10:13 PM, Mike Watters<mwatters at opencsw.org> wrote:
> I will make the change the next time I release the package.
> please file a feature request bug so I don't forget about it ;-}

Done.

http://www.opencsw.org/bugtrack/view.php?id=3763


From pfelecan at opencsw.org  Fri Jul 10 19:42:15 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 10 Jul 2009 19:42:15 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <20090710161724.GA73109@bolthole.com> (Philip Brown's message of
	"Fri\, 10 Jul 2009 09\:17\:24 -0700")
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com> <x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
	<163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
	<20090710161724.GA73109@bolthole.com>
Message-ID: <x6eiso1ll4.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Fri, Jul 10, 2009 at 05:08:54PM +0200, Dagobert Michelsen wrote:
>> - The catalog on build8x was old
>
> oh... this was a build-machine-specific problem.
>
> pfft. he should have mailed the buildfarm list not the maintainers list
> then! :-P

*he* didn't know that it was build machine specific; it was revealed on
it, following the correction of a global issue, i.e., not build machine
specific...

tat sat
-- 
Peter


From william at wbonnet.net  Sat Jul 11 14:34:14 2009
From: william at wbonnet.net (William Bonnet)
Date: Sat, 11 Jul 2009 14:34:14 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A54DA45.4090902@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
	<4A54DA45.4090902@wbonnet.net>
Message-ID: <4A5886C6.1040405@wbonnet.net>

Hi

> Hi Ben
>> Excerpts from William Bonnet's message of Tue Jul 07 19:00:10 -0400 
>> 2009:
>>  
>>> I do have the same problem with the /opt/csw/include/jri.h file. It 
>>> comes from the same packages and also says it has to be provided by 
>>> netscape.
>>>
>>> Could someone please rename it to .old on build9x ?
>>>     
>>
>> Doing this now.
>>   
> Thanks
>
> It worked, i was able to compile FF3 on solaris 9, and the binaries 
> runs on solaris 10. Now i have to find tests machines running solaris 
> 8 and 9 :)
>
> I also need the same fix on build9x 8x and 9x. I better look to 
> produce a Qt patch :)

I have recompiled FF3.0 on build8x, but i need to have the following 
conflicting file to be removed from build8s

/opt/csw/include/jri.h
/opt/csw/include/npapi.h

Thanks in advance

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 lsw at bender.opencsw.org  Sat Jul 11 18:52:00 2009
From: lsw at bender.opencsw.org (Liu Siwei)
Date: Sat, 11 Jul 2009 18:52:00 +0200
Subject: [csw-maintainers] /testing libUPnP 1.6.6
Message-ID: <20090711165200.GA324@bender.opencsw.org>

http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-sparc-CSW.pkg.gz




# pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libupnp

   or,

# pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libupnp


--
Siwei



From dam at opencsw.org  Sat Jul 11 19:02:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 11 Jul 2009 19:02:07 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A5886C6.1040405@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
	<4A54DA45.4090902@wbonnet.net> <4A5886C6.1040405@wbonnet.net>
Message-ID: <956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>

Hi William,

Am 11.07.2009 um 14:34 schrieb William Bonnet:
> I have recompiled FF3.0 on build8x

Cool :-)

> but i need to have the following conflicting file to be removed from  
> build8s
>
> /opt/csw/include/jri.h
> /opt/csw/include/npapi.h

Done.


Best regards

   -- Dago



From dam at opencsw.org  Sat Jul 11 19:11:21 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 11 Jul 2009 19:11:21 +0200
Subject: [csw-maintainers] Booting login and *s
Message-ID: <AEA7EEC5-83EE-4F2A-B266-0C8F37A4072E@opencsw.org>

Hi,

there is a serious I/O problem on the T5220 and all zones. It seems
that some I/Os are simply stuck, but everything else looks ok. As
I can't figure out what is wrong (no disk errs, no faulted pools,
nothing in messages) I plan to boot the server as last resort in
about 30 minutes.


Sorry for the inconvenience

   -- Dago


From loseweigh at gmail.com  Sat Jul 11 18:37:03 2009
From: loseweigh at gmail.com (Liu Siwei)
Date: Sun, 12 Jul 2009 00:37:03 +0800
Subject: [csw-maintainers] /testing libUPnP 1.6.6
Message-ID: <3dd2fcb80907110937h69c17f81xb3bc00a0c2b5068b@mail.gmail.com>

http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/libupnp-1.6.6,REV=2009.07.10-SunOS5.8-sparc-CSW.pkg.gz





# pkgutil -t http://mirror.opencsw.org/opencsw/testing -i libupnp

   or,

# pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u libupnp



-Siwei
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090712/bd58348b/attachment-0002.html>

From dam at opencsw.org  Sun Jul 12 09:21:33 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 09:21:33 +0200
Subject: [csw-maintainers] New in testing: libX11 1.2.2, libvorbis 1.2.3,
	neon 0.28.5
Message-ID: <4E7AC449-B3B6-468D-B345-3D3F43EC2A09@opencsw.org>

Hi,

new packages in
   http://mirror.opencsw.org/testing.html

neon-0.28.5,REV=2009.07.11-SunOS5.8-sparc-CSW.pkg.gz
neon-0.28.5,REV=2009.07.11-SunOS5.8-i386-CSW.pkg.gz
neon_devel-0.28.5,REV=2009.07.11-SunOS5.8-sparc-CSW.pkg.gz
neon_devel-0.28.5,REV=2009.07.11-SunOS5.8-i386-CSW.pkg.gz

libvorbis-1.2.3,REV=2009.07.11-SunOS5.8-sparc-CSW.pkg.gz
libvorbis-1.2.3,REV=2009.07.11-SunOS5.8-i386-CSW.pkg.gz

libx11-1.2.2,REV=2009.07.12-SunOS5.8-i386-CSW.pkg.gz
libx11-1.2.2,REV=2009.07.12-SunOS5.8-sparc-CSW.pkg.gz
libx11_devel-1.2.2,REV=2009.07.12-SunOS5.8-i386-CSW.pkg.gz
libx11_devel-1.2.2,REV=2009.07.12-SunOS5.8-sparc-CSW.pkg.gz

If you have time please test against your applications.


Best regards

   -- Dago


From ihsan at opencsw.org  Sun Jul 12 12:38:38 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 12 Jul 2009 12:38:38 +0200
Subject: [csw-maintainers] contact maintainer page
In-Reply-To: <4A366723.3050806@opencsw.org>
References: <x6vdmxzjau.fsf@abhavana.free.fr>	<20090615145906.GE34818@bolthole.com>	<55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de>
	<4A366723.3050806@opencsw.org>
Message-ID: <4A59BD2E.7030607@opencsw.org>

Am 15.6.2009 17:22 Uhr, Mike Watters schrieb:

>>>> I'm trying to use the site's mailer:
>>>> http://www.opencsw.org/eml.php?devname=pfelecan
>>>>
>>>> After entering my e-mail address it shows me a page with the text
>>>> "Debug: part 1". Nice but not useful if somebody tries to write me
>>>> something through this...
>>> posix_getpwnam()    [a php call]
>>> seems to be broken somehow on the website now.
>> Might be related to the recent php5 modules split-off? [1]
> 
> Installing the following should fix
> 
> php5_posix 5.2.9,REV=2009.04.29

It did. Thanks & sorry for the delay.




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From william at wbonnet.net  Sun Jul 12 13:20:44 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 12 Jul 2009 13:20:44 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>	<4A54DA45.4090902@wbonnet.net>
	<4A5886C6.1040405@wbonnet.net>
	<956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>
Message-ID: <4A59C70C.7000503@wbonnet.net>

Hi

>> I have recompiled FF3.0 on build8x
>
> Cool :-)
Packages are not yet tested on every platform but recompiled. x86 
package works on Solaris 10. I am look for x86 Solaris 8 and 9 VMs (or 
VBox or whatever). I am currently  building Sparc version of FF3

So if some wants to do some testing FF3.0.11 x86 is available from 
testing. I'll announce it on the users list when  sparc version will be ok.

I would like to have a feed back on solaris 8. It seems sol9 will be ok, 
but a  link b-problem exists on solaris 8.
 
>
>> but i need to have the following conflicting file to be removed from 
>> build8s
>>
>> /opt/csw/include/jri.h
>> /opt/csw/include/npapi.h
>
> Done.
Thanks

I'll try to repackage Qt when i'll be done wih FF...

I've able to build FF3.5 on my box (it means no more source to patch or 
link issues to go until the end and poduce package). But binary does not 
work yet (illegal instruction and core dump). So i still have things to 
tweak. I don't know how long it may take.

cheers
W.



From ihsan at opencsw.org  Sun Jul 12 13:26:03 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sun, 12 Jul 2009 13:26:03 +0200
Subject: [csw-maintainers]
	=?iso-8859-1?q?Additional_Farm_Maintainer=3A_J?=
	=?iso-8859-1?q?=FCrgen_Arndt?=
In-Reply-To: <06DF27E0-FF15-455A-9040-5A5C5160F8AC@opencsw.org>
References: <06DF27E0-FF15-455A-9040-5A5C5160F8AC@opencsw.org>
Message-ID: <4A59C84B.7020004@opencsw.org>

Am 9.7.2009 20:56 Uhr, Dagobert Michelsen schrieb:

> I'm back from my vacation and noticed that there are obviously not
> enough farm managers as Ben wanted to take some time off, too. I have
> given J?rgen full admin access to the farm. Ihsan, please add "ja"
> to the list of buildfarm maintainers.

Actually he is already a member of the buildfarm list.



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From skayser at opencsw.org  Sun Jul 12 14:54:54 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sun, 12 Jul 2009 14:54:54 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
 instead of PROTOTYPE_FILTER
In-Reply-To: <EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
Message-ID: <4A59DD1E.2060205@opencsw.org>

Dagobert Michelsen wrote:
> Am 04.07.2009 um 19:12 schrieb Sebastian Kayser:
>> Dagobert Michelsen wrote:
>>> Am 01.07.2009 um 22:08 schrieb Sebastian Kayser:
>>>> while working on postfix 2.6 i have noticed again that it is not  
>>>> quite
>>>> straight-forward to assign files to classes. Instead of exposing the
>>>> inner
>>>> workings of the prototype stuff to the Makefile, could we maybe get
>>>> some
>>>> variables to assign files to classes (functions)?
>>> For CSWcswclassutils this is already in there :-) Just define
>>>
>>> SAMPLECONF
>>> PRESERVECONF
>>> INITSMF
>>>
>>> and everything else will be taken cared of.
>> Sweet!!! :D Can we add something for USERGROUP also?
> ...
> Looks good. Please go ahead and commit.

Done (r5569). I have also added the variables to the GAR variable
reference [1]. What about cswpycompile, do we want to have that as such
a variable (PYCOMPILE) also?

Peter, would you mind amending the cswclassutils wiki page to reflect
the variables (PRESERVECONF, SAMPLECONF, INITSMF, USERGROUP) that can be
used instead of PROTOTYPE_FILTER?

Sebastian

[1]http://sourceforge.net/apps/trac/gar/wiki/GAR%20Variable%20Reference#cswclassutilsSettings


From dam at opencsw.org  Sun Jul 12 15:23:02 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 15:23:02 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <4A59DD1E.2060205@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
Message-ID: <E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>

Hi Sebastian,

Am 12.07.2009 um 14:54 schrieb Sebastian Kayser:
> Done (r5569). I have also added the variables to the GAR variable
> reference [1]. What about cswpycompile, do we want to have that as  
> such
> a variable (PYCOMPILE) also?

Should be done. Something like this is needed for Python (taken
from x11/xcb-proto):

> PYTHON_PREFIX=$(BUILD_PREFIX)/lib/python/site-packages
>
> EXTRA_PAX_ARGS = -s ',^.*/\(xcbgen/.*\.py\),.$(BUILD_PREFIX)/lib/ 
> python/site-packages/\1,'
> EXTRA_PAX_ARGS += -s ',^\.$(libdir)/python.*,,'
>
> EXTRA_MERGE_EXCLUDE_FILES = .*\.pyo .*\.pyc
> PROTOTYPE_FILTER = awk '$$$$3 ~ /.*\.py$$$$/ { $$$$2 =  
> "cswpycompile" } { print }'
> SPKG_CLASSES = none cswpycompile
> REQUIRED_PKGS += CSWcswclassutils
>
> PKGFILES_CSWpyxcbproto = .*\.py

The detection would however be slightly more complicated as the presence
of .py-files should be automatically detected. If present, the filter
should be put in place and the class changed for the .py-files.
As of the current implementation of the cswpycompile class the .py-files
must be put exactly where they are relocated in this example via
pax-relocation during the merge phase. If they are anywhere else
a warning (error?) should be printed.


Best regards

   -- Dago


From dam at opencsw.org  Sun Jul 12 15:24:30 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 15:24:30 +0200
Subject: [csw-maintainers] contact maintainer page
In-Reply-To: <4A59BD2E.7030607@opencsw.org>
References: <x6vdmxzjau.fsf@abhavana.free.fr>	<20090615145906.GE34818@bolthole.com>	<55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de>
	<4A366723.3050806@opencsw.org> <4A59BD2E.7030607@opencsw.org>
Message-ID: <9E7F8956-A5E6-4110-8DFD-3B70DE9ED1F2@opencsw.org>

Hi,

Am 12.07.2009 um 12:38 schrieb Ihsan Dogan:

> Am 15.6.2009 17:22 Uhr, Mike Watters schrieb:
>
>>>>> I'm trying to use the site's mailer:
>>>>> http://www.opencsw.org/eml.php?devname=pfelecan
>>>>>
>>>>> After entering my e-mail address it shows me a page with the text
>>>>> "Debug: part 1". Nice but not useful if somebody tries to write me
>>>>> something through this...
>>>> posix_getpwnam()    [a php call]
>>>> seems to be broken somehow on the website now.
>>> Might be related to the recent php5 modules split-off? [1]
>>
>> Installing the following should fix
>>
>> php5_posix 5.2.9,REV=2009.04.29
>
> It did. Thanks & sorry for the delay.

Not quite:

> Debug: part 1. Problem with getpwnam Debug part 2. If you see this,  
> problem is fixed Thank you. Your message has been sent to maintainer  
> Dagobert Michelsen .

Please fix the webpage also.


Best regards

   -- Dago


From dam at opencsw.org  Sun Jul 12 15:25:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 12 Jul 2009 15:25:55 +0200
Subject: [csw-maintainers] catalog for build8x is rotten
In-Reply-To: <x6eiso1ll4.fsf@abhavana.free.fr>
References: <x6zlbc25y5.fsf@abhavana.free.fr>
	<20090710141148.GA99639@bolthole.com>
	<x6iqi01u7s.fsf@abhavana.free.fr>
	<20090710150446.GA75987@bolthole.com>
	<163C84FE-DD36-4A96-AEF6-371345E60478@opencsw.org>
	<20090710161724.GA73109@bolthole.com>
	<x6eiso1ll4.fsf@abhavana.free.fr>
Message-ID: <9D4229A3-579A-415A-AFFB-484C736B735B@opencsw.org>

Hi Peter,

Am 10.07.2009 um 19:42 schrieb Peter FELECAN:
> Philip Brown <phil at bolthole.com> writes:
>> On Fri, Jul 10, 2009 at 05:08:54PM +0200, Dagobert Michelsen wrote:
>>> - The catalog on build8x was old
>>
>> oh... this was a build-machine-specific problem.
>>
>> pfft. he should have mailed the buildfarm list not the maintainers  
>> list
>> then! :-P
>
> *he* didn't know that it was build machine specific; it was revealed  
> on
> it, following the correction of a global issue, i.e., not build  
> machine
> specific...
>
> tat sat

No problem :-) Apart from that both tools should have an option for
download which does in no case require root access.


Best regards

   -- Dago


From skayser at opencsw.org  Sun Jul 12 16:17:54 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sun, 12 Jul 2009 16:17:54 +0200
Subject: [csw-maintainers] Jumpstart post-install recipe for OpenCSW
In-Reply-To: <4A294F5B.1070905@cognigencorp.com>
References: <51631.194.246.122.22.1244219260.squirrel@ssl.skayser.de>
	<4A294F5B.1070905@cognigencorp.com>
Message-ID: <4A59F092.6090008@opencsw.org>

Hi Darin,

Darin Perusich wrote:
> Sebastian Kayser wrote:
>> does someone have a working "install me some OpenCSW packages" Jumpstart
>> post-install snippet that he could pass to me? I am in the process of
>> setting up my first Jumpstart server and i would like to have it install a
>> list of CSW pkgs after setting up the machine.
> 
> I attached my CSW package installation script. It's nothing fancy but
> get the job done for me.
>
>> Is this known to work with the various postinstall scripts (or cswinitsmf)
>> that we have in the packages?
>>
>> And out of curiosity: What are the specifics to watch out for so that a
>> postinstall script will work in a Jumpstart environment, just
>> $PKG_INSTALL_ROOT [1] or are there any other caveats that pop up
>> frequently?
> 
> If you chroot the package installation you don't need to worry about
> $PKG_INSTALL_ROOT being set. I use chroot to install everything except
> pkg-get and I've never had a problem. Granted I only install a small
> number of package, sudo, amanda, etc, and build the servers out
> afterwards. If it's a machine I need to make a ton of clones of I set it
> up and create flash images of anyways.

thanks for posting this, I finally had some time to further work on our
Jumpstart setup.

I found it easiest to use JET [1] and custom_scripts_n, so that post
installation scripts like this are run after the final system boot. The
system is then fully functional and there is no need to mess around with
PKG_INSTALL_ROOT or similar. :) Easy.

The adapted version of your script is attached. Thanks again.

Sebastian

[1] http://wikis.sun.com/display/JET/Home
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: install_csw.sh
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090712/ef881f46/attachment-0002.ksh>

From phil at bolthole.com  Sun Jul 12 18:31:00 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 12 Jul 2009 09:31:00 -0700
Subject: [csw-maintainers] contact maintainer page
In-Reply-To: <9E7F8956-A5E6-4110-8DFD-3B70DE9ED1F2@opencsw.org>
References: <x6vdmxzjau.fsf@abhavana.free.fr>
	<20090615145906.GE34818@bolthole.com>
	<55589.78.140.68.9.1245078252.squirrel@ssl.skayser.de>
	<4A366723.3050806@opencsw.org> <4A59BD2E.7030607@opencsw.org>
	<9E7F8956-A5E6-4110-8DFD-3B70DE9ED1F2@opencsw.org>
Message-ID: <20090712163059.GF63696@bolthole.com>

On Sun, Jul 12, 2009 at 03:24:30PM +0200, Dagobert Michelsen wrote:
>> Debug: part 1. Problem with getpwnam Debug part 2. If you see this,  
>> problem is fixed Thank you. Your message has been sent to maintainer  
>> Dagobert Michelsen .
>
> Please fix the webpage also.

this is done


From james at opencsw.org  Sun Jul 12 18:59:01 2009
From: james at opencsw.org (James Lee)
Date: Sun, 12 Jul 2009 16:59:01 GMT
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <4A59C70C.7000503@wbonnet.net>
References: <4A4B5740.90407@wbonnet.net>
	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>
	<4A53D37A.40502@wbonnet.net>
	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>
	<4A54DA45.4090902@wbonnet.net> <4A5886C6.1040405@wbonnet.net>
	<956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>
	<4A59C70C.7000503@wbonnet.net>
Message-ID: <20090712.16590100.1755802829@gyor.oxdrove.co.uk>

On 12/07/09, 12:20:44, William Bonnet <william at wbonnet.net> wrote regarding
Re: [csw-maintainers] Conflicting header file:

> So if some wants to do some testing FF3.0.11 x86 is available from
> testing. I'll announce it on the users list when  sparc version will be
> ok.

> I would like to have a feed back on solaris 8. It seems sol9 will be ok,
> but a  link b-problem exists on solaris 8.



$ uname -a
SunOS bala 5.8 Generic_117351-32 i86pc i386 i86pc
$ firefox
ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
object requires TLS, but TLS failed to initialize
Killed
$ pkgparam CSWfirefox VERSION
3.0.11,REV=2009.07.11




$ uname -a
SunOS gyor 5.8 Generic_117350-38 sun4u sparc SUNW,Ultra-60
$ firefox
ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
object requires TLS, but TLS failed to initialize
Killed
$ pkgparam CSWfirefox VERSION
3.0.11,REV=2009.07.12


From william at wbonnet.net  Sun Jul 12 19:04:48 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 12 Jul 2009 19:04:48 +0200
Subject: [csw-maintainers] Conflicting header file
In-Reply-To: <20090712.16590100.1755802829@gyor.oxdrove.co.uk>
References: <4A4B5740.90407@wbonnet.net>	<595894E3-2042-4FBF-9EC6-6DBFC74147E7@opencsw.org>	<4A53D37A.40502@wbonnet.net>	<1247013402-sup-5213@ntdws12.chass.utoronto.ca>	<4A54DA45.4090902@wbonnet.net>
	<4A5886C6.1040405@wbonnet.net>	<956FD720-FDE5-40DD-8ED4-1A4BE79B9405@opencsw.org>	<4A59C70C.7000503@wbonnet.net>
	<20090712.16590100.1755802829@gyor.oxdrove.co.uk>
Message-ID: <4A5A17B0.9040201@wbonnet.net>

Hi James
>> I would like to have a feed back on solaris 8. It seems sol9 will be ok,
>> but a  link b-problem exists on solaris 8.
>>     
>
>
>
> $ uname -a
> SunOS bala 5.8 Generic_117351-32 i86pc i386 i86pc
> $ firefox
> ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
> object requires TLS, but TLS failed to initialize
> Killed
> $ pkgparam CSWfirefox VERSION
> 3.0.11,REV=2009.07.11
>
>
>
>
> $ uname -a
> SunOS gyor 5.8 Generic_117350-38 sun4u sparc SUNW,Ultra-60
> $ firefox
> ld.so.1: firefox-bin: fatal: /opt/csw/mozilla/firefox/lib/libjemalloc.so:
> object requires TLS, but TLS failed to initialize
> Killed
> $ pkgparam CSWfirefox VERSION
> 3.0.11,REV=2009.07.12
>   

Thanks

That's what i was affraid of :( I will set up a test machine here since 
i'm not in Paris and with a lot of oldies.

Anyone tried on Solaris 9 ?

cheers
W.



From dam at opencsw.org  Mon Jul 13 16:47:20 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 13 Jul 2009 16:47:20 +0200
Subject: [csw-maintainers] Attention: Please clean up /home
Message-ID: <9BA10F46-5ABB-4ED0-8283-F2AAFF91078B@opencsw.org>

Hi,

it looks like the machine hit bug# 6596237:
   <http://sunsolve.sun.com/search/document.do?assetkey=1-1-6596237-1>
I have IDR137886-01 which should bring some relief, however, the IDR
was built against an older kernel version and downgrading requires
removing *a lot* of patches. Nonetheless I will continue working
towards it. Additionally, the problem can be solved by making free
space in the pool.

So: Please delete stuff from your homes you no longer need! The
extra space will speed up processing considerably.


Sorry for the inconvenience

   -- Dago


From maciej at opencsw.org  Mon Jul 13 20:28:28 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 13 Jul 2009 19:28:28 +0100
Subject: [csw-maintainers] stat command
Message-ID: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>

Has anyone built the 'stat' utility on Solaris? In Debian, the
/usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
the same utilities are part of the 'fileutils' package (CSWgfile).
Were there any plans to add 'stat' to fileutils?

Maciej


From bwalton at opencsw.org  Mon Jul 13 20:44:37 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 13 Jul 2009 14:44:37 -0400
Subject: [csw-maintainers] stat command
In-Reply-To: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
Message-ID: <1247510598-sup-7662@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Mon Jul 13 14:28:28 -0400 2009:
> Has anyone built the 'stat' utility on Solaris? In Debian, the
> /usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
> the same utilities are part of the 'fileutils' package (CSWgfile).
> Were there any plans to add 'stat' to fileutils?

Not yet, but I was about to build coreutils for exactly this reason.
I sorely miss the stat command...I was waiting on an update to another
package (forget which without checking) before I could build the
current/stable coreutils though.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090713/44a57cdb/attachment-0002.asc>

From phil at bolthole.com  Mon Jul 13 21:00:36 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 13 Jul 2009 12:00:36 -0700
Subject: [csw-maintainers] stat command
In-Reply-To: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
Message-ID: <20090713190036.GB8171@bolthole.com>

On Mon, Jul 13, 2009 at 07:28:28PM +0100, Maciej (Matchek) Blizinski wrote:
> Has anyone built the 'stat' utility on Solaris? In Debian, the
> /usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
> the same utilities are part of the 'fileutils' package (CSWgfile).


how odd. i can think of no reason why it wouldnt be in the package, if it
really is usually as part of that software distribution.


From Darin.Perusich at cognigencorp.com  Mon Jul 13 21:04:32 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Mon, 13 Jul 2009 15:04:32 -0400
Subject: [csw-maintainers] stat command
In-Reply-To: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
Message-ID: <4A5B8540.5020803@cognigencorp.com>

I had a request to updated to coreutils that went unanswered so I
compiled the package. It's in ~darin/build/ if you want to take a look.

Maciej (Matchek) Blizinski wrote:
> Has anyone built the 'stat' utility on Solaris? In Debian, the
> /usr/bin/stat binary is a part of the coreutils package. In OpenCSW,
> the same utilities are part of the 'fileutils' package (CSWgfile).
> Were there any plans to add 'stat' to fileutils?
> 


-- 
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  Mon Jul 13 21:08:39 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 13 Jul 2009 15:08:39 -0400
Subject: [csw-maintainers] stat command
In-Reply-To: <20090713190036.GB8171@bolthole.com>
References: <ce97543b0907131128i698d066dqaff28dd87ec00dc5@mail.gmail.com>
	<20090713190036.GB8171@bolthole.com>
Message-ID: <1247511880-sup-5149@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Mon Jul 13 15:00:36 -0400 2009:
> how odd. i can think of no reason why it wouldnt be in the package, if it
> really is usually as part of that software distribution.

Coreutils is a superset of the fileutils, shellutils and textutils
packages, which are no longer offered as individual packages.  The
actual stat command used to be part of shellutils, I believe.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090713/c429084e/attachment-0002.asc>

From dam at opencsw.org  Mon Jul 13 23:23:48 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 13 Jul 2009 23:23:48 +0200
Subject: [csw-maintainers] ZFS issue again
Message-ID: <13222C2C-B152-47B6-BA8D-00B7D0B5DA46@opencsw.org>

Hi,

as I was not able to install the IDR due to zone patch issues I
replaced the internal 146 GB disks with 300 GB disks. For the
pool resize I need to unmount/remount the pool, so please log
out for a few minutes.


Sorry for the inconvenience

   -- Dago


From dam at opencsw.org  Mon Jul 13 23:38:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 13 Jul 2009 23:38:08 +0200
Subject: [csw-maintainers] ZFS issue again
In-Reply-To: <13222C2C-B152-47B6-BA8D-00B7D0B5DA46@opencsw.org>
References: <13222C2C-B152-47B6-BA8D-00B7D0B5DA46@opencsw.org>
Message-ID: <764C586D-80EE-4CF9-8E88-F203BC5ECA00@opencsw.org>

Hi,

Am 13.07.2009 um 23:23 schrieb Dagobert Michelsen:
> as I was not able to install the IDR due to zone patch issues I
> replaced the internal 146 GB disks with 300 GB disks. For the
> pool resize I need to unmount/remount the pool, so please log
> out for a few minutes.

I hope this is resolved now. At least in my few tests the
performance is quite good :-D

> dam at login [login]:/home/dam > time gtar zxf /home/src/ 
> perl-5.10.0.tar.gz
> gtar zxf /home/src/perl-5.10.0.tar.gz  3.17s user 3.17s system 126%  
> cpu 5.004 total
> dam at login [login]:/home/dam > find perl-5.10.0 | wc -l
>     4680
> dam at login [login]:/home/dam > time mkfile 1g bigfile
> mkfile 1g bigfile  0.36s user 9.52s system 99% cpu 9.892 total



Best regards

   -- Dago


From william at wbonnet.net  Wed Jul 15 12:01:16 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 12:01:16 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
Message-ID: <4A5DA8EC.1080803@wbonnet.net>

Hi

I noticed today that there is an open issue on xpdf since Junuary 2008 
that received several "ping" from the reporter. Current maintaner, 
Alessio is retired. Would someone be interested in taking other this 
package and fix the opened issues ?

Thanks in advance

cheers
W.


From skayser at opencsw.org  Wed Jul 15 12:13:23 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 15 Jul 2009 12:13:23 +0200 (CEST)
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <4A5DA8EC.1080803@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>
Message-ID: <59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>

William Bonnet wrote:
> I noticed today that there is an open issue on xpdf since Junuary 2008
> that received several "ping" from the reporter. Current maintaner,
> Alessio is retired. Would someone be interested in taking other this
> package and fix the opened issues ?

>From what i know, Benny has been working on an updated version. He is at a
customer event right now, but he should be able to tell you whether i was
right or not later on today.

While we are at fixing bugs: Any news on the dbus not properly shutting
down issue that has been open for a while now [1]? We have had two boxes
here hanging on an attempted reboot just the other day.

Sebastian

[1] http://opencsw.org/mantis/view.php?id=3626



From maciej at opencsw.org  Wed Jul 15 13:34:32 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 15 Jul 2009 12:34:32 +0100
Subject: [csw-maintainers] Modifying SMF dependencies
Message-ID: <ce97543b0907150434p31370d05q46eb87e4675e7865@mail.gmail.com>

When working on syslog_ng, I came across a new problem. A system
service such as syslog, usually has some other services depending on
it. I've got into a problem in which I couldn't run rarpd together
with cswsyslog_ng, because rarpd depends on system-log, which was
disabled.

Have any of you gotten around a problem "How do I automate SMF
dependency changes"? If so, do you have any code related to that?

If you haven't, what would you do? I suspect there would be svccfg
export and svccfg import involved, with XML munging along the way.
What would make the minimal dependency collection for XML modification
tools on Solaris? (I only work with Solaris 10, I fear asking about
XML processing on Solaris 8...)

Maciej


From william at wbonnet.net  Wed Jul 15 14:20:22 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 14:20:22 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
Message-ID: <4A5DC986.3020606@wbonnet.net>

Hi
> While we are at fixing bugs: Any news on the dbus not properly shutting
> down issue that has been open for a while now [1]? We have had two boxes
> here hanging on an attempted reboot just the other day.
>   
Not yet :( I'll give a look to this issue tonight. I was focused on 
firefox compilation issues so far...

cheers
W.


From ellson at opencsw.org  Wed Jul 15 18:15:56 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 15 Jul 2009 12:15:56 -0400
Subject: [csw-maintainers] howto control which files go into which part of
	split package
Message-ID: <4A5E00BC.3090800@opencsw.org>

1.   How can I list the file contents of a .pkg.gz to verify what I've done?

2.   How do I specify the set of files to go into a subpackage, and 
exclude those same files from the other
subpackages?


e.g.   if I have:

     PKGFILES_CSWgraphviztcl = $(libexecdir_install)/graphviz/tcl/.*

does that automatically exclude those matched files from the core 
graphviz package?




Current attempt is in:

     /home/ellson/mgar/pkg/graphviz/trunk/Makefile


From skayser at opencsw.org  Wed Jul 15 22:21:46 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 15 Jul 2009 22:21:46 +0200 (CEST)
Subject: [csw-maintainers] howto control which files go into which part
 of split package
In-Reply-To: <4A5E00BC.3090800@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
Message-ID: <50628.217.227.36.102.1247689306.squirrel@ssl.skayser.de>

Hi John,

John Ellson wrote:
> 1.   How can I list the file contents of a .pkg.gz to verify what I've
> done?

  $ gunzip .pkg.gz
  $ pkgtrans .pkg .

Then dig around in CSWpkg. If you just build one package (that is without
splitting files in separate packages), you can also have a look in
work/pkgroot just after the "gmake merge" phase. That is what i usually
do, instead of "gmake package", first do a gmake merge and the find
work/pkgroot |less.

> 2.   How do I specify the set of files to go into a subpackage, and
> exclude those same files from the other
> subpackages?
>
> e.g.   if I have:
>
>      PKGFILES_CSWgraphviztcl = $(libexecdir_install)/graphviz/tcl/.*
>
> does that automatically exclude those matched files from the core
> graphviz package?

Yes, should be the default behaviour. See [1,2].

Sebastian

[1] http://sourceforge.net/apps/trac/gar/wiki/Prototypes
[2] http://sourceforge.net/apps/trac/gar/wiki/DynamicPackageFiles



From dam at opencsw.org  Wed Jul 15 22:22:28 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 15 Jul 2009 22:22:28 +0200
Subject: [csw-maintainers] howto control which files go into which part
	of split package
In-Reply-To: <4A5E00BC.3090800@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
Message-ID: <1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>

Ho John,

Am 15.07.2009 um 18:15 schrieb John Ellson:
> 1.   How can I list the file contents of a .pkg.gz to verify what  
> I've done?

After "gmake package" / "gmake repackage" you can inspect
   work/build-global/*.prototype*

> 2.   How do I specify the set of files to go into a subpackage, and  
> exclude those same files from the other
> subpackages?
>
> e.g.   if I have:
>
>    PKGFILES_CSWgraphviztcl = $(libexecdir_install)/graphviz/tcl/.*
>
> does that automatically exclude those matched files from the core  
> graphviz package?

Yes. This is documented at
   <http://sourceforge.net/apps/trac/gar/wiki/Prototypes>
You must also specify that you want to build multiple packages  
documented at
   <http://sourceforge.net/apps/trac/gar/wiki/DynamicPackageFiles>

You will usually not use *_install, as the installation goes to $ 
(libexecdir).
The use of *_install is mostly for internal things.

J?rgen has also spent some effort on a graphviz package and he'll help  
you
push the package out soon.


Best regards

   -- Dago

From william at wbonnet.net  Wed Jul 15 22:25:59 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 22:25:59 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <4A5DC986.3020606@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
	<4A5DC986.3020606@wbonnet.net>
Message-ID: <4A5E3B57.1030901@wbonnet.net>

Hi

> Not yet :( I'll give a look to this issue tonight. I was focused on 
> firefox compilation issues so far...
1/ Good news is that it seems to be fixed !

2/ Bad news is that i could have done it earlier... many many thanks to 
Sebastian for reminding me to process this important bug. After 
analyzing the bug, it was a real 'showstopper' i did not identified when 
it was reported.

3/ Really bad news is that the CSWdbus package that is available in 
"current" cannot be uninstalled in "normal mode" :( It means, it cannot 
be updated without a manual step... I do apologize for that... (and i do 
feel Phil's feet coming closer and closer from my back).

So what does this means :

I made a simple fix and updated the following packages : CSWdbus, 
CSWlibdbus and CSWlibdbusdev. The only package which have a real problem 
is CSWdbus. The update i made seems to fix the problem on my boxes at 
home. But it would appreciate to have some more feedbacks. Just to be 
sure i have fixed all the issues, in all possible way to make it occur.

The bug Sebastian outlined today is a related to dbus service stop. In 
most of the time it will fail. And it will fail 99.9% of the time during 
an update (either from pkg-get or pkgutil).  Current dbus is generating 
a pid "lock file" in /opt/csw/var/run/dbus/pid . This pid file is not 
processed correctly during service stop (and uninstall), and dbus will 
not stop.

I have updated packages and they are now available from testing. As i 
said, i was able to reproduce the problem before update and not after, 
but i would appreciate some more feedbacks.

If you plan to install update from testing you will have to kill by hand 
dbus process. dbus pid is stored in /opt/csw/var/run/dbus/pid . But 
unfortunately it is ignored by stop (or uninstall) script. So you just 
have to :

kill -9 `cat /opt/csw/var/run/dbus/pid`

before updating.

This has to be announced on thge user  list before any release. Phil, i 
do apologize for this, and please tell me the procedure i have to follow 
to announce it, if i am the one to announce 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 ihsan at opencsw.org  Wed Jul 15 22:34:02 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Wed, 15 Jul 2009 22:34:02 +0200
Subject: [csw-maintainers] syslog_ng 3.0.3 in testing
Message-ID: <4A5E3D3A.2000305@opencsw.org>

Hello,

Thanks to the great support of Maciej, I've updated syslog_ng to 3.0.3.

http://mirror.opencsw.org/testing/syslog_ng-3.0.3,REV=2009.07.15-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/syslog_ng-3.0.3,REV=2009.07.15-SunOS5.8-sparc-CSW.pkg.gz





Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Wed Jul 15 22:40:18 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Jul 2009 13:40:18 -0700
Subject: [csw-maintainers] (dbus fixes)
In-Reply-To: <4A5E3B57.1030901@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
	<4A5DC986.3020606@wbonnet.net> <4A5E3B57.1030901@wbonnet.net>
Message-ID: <20090715204018.GA94635@bolthole.com>

On Wed, Jul 15, 2009 at 10:25:59PM +0200, William Bonnet wrote:
> This has to be announced on thge user  list before any release. Phil, i  
> do apologize for this, and please tell me the procedure i have to follow  
> to announce it, if i am the one to announce it.

yes you should be the one to announce it :-)

once it passes through testing, thenI would say as soon as I officialy
clear it through newpkgs, you should send emails to both the user, and
"announce" lists, mentioning what is required.



From phil at bolthole.com  Wed Jul 15 22:41:41 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Jul 2009 13:41:41 -0700
Subject: [csw-maintainers] howto control which files go into which
	part	of split package
In-Reply-To: <50628.217.227.36.102.1247689306.squirrel@ssl.skayser.de>
References: <4A5E00BC.3090800@opencsw.org>
	<50628.217.227.36.102.1247689306.squirrel@ssl.skayser.de>
Message-ID: <20090715204141.GB94635@bolthole.com>

On Wed, Jul 15, 2009 at 10:21:46PM +0200, Sebastian Kayser wrote:
> Hi John,
> 
> John Ellson wrote:
> > 1.   How can I list the file contents of a .pkg.gz to verify what I've
> > done?
> 
>   $ gunzip .pkg.gz
>   $ pkgtrans .pkg .


or if you are masochistic, and short on disk space:

PAGER=less gzmore x.pkg.gz

and look at the equivalent of the pkgmap file that way.



From william at wbonnet.net  Wed Jul 15 22:49:14 2009
From: william at wbonnet.net (William Bonnet)
Date: Wed, 15 Jul 2009 22:49:14 +0200
Subject: [csw-maintainers] (dbus fixes)
In-Reply-To: <20090715204018.GA94635@bolthole.com>
References: <4A5DA8EC.1080803@wbonnet.net>	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>	<4A5DC986.3020606@wbonnet.net>
	<4A5E3B57.1030901@wbonnet.net>
	<20090715204018.GA94635@bolthole.com>
Message-ID: <4A5E40CA.5070305@wbonnet.net>

Hi

> yes you should be the one to announce it :-)
>   
Master Yoda^W^WPhil, the one to announce it, will I be ;)

> once it passes through testing, thenI would say as soon as I officialy
> clear it through newpkgs, you should send emails to both the user, and
> "announce" lists, mentioning what is required.
>   
So thanks for more feedbacks then i'll request for a release.

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  Thu Jul 16 05:59:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 15 Jul 2009 20:59:06 -0700
Subject: [csw-maintainers] Modifying SMF dependencies
In-Reply-To: <ce97543b0907150434p31370d05q46eb87e4675e7865@mail.gmail.com>
References: <ce97543b0907150434p31370d05q46eb87e4675e7865@mail.gmail.com>
Message-ID: <20090716035906.GA85493@bolthole.com>

On Wed, Jul 15, 2009 at 12:34:32PM +0100, Maciej (Matchek) Blizinski wrote:
 (I only work with Solaris 10, I fear asking about
> XML processing on Solaris 8...)

in solaris 8, you just put your demon "at the end", and cross your fingers
;-)


From bwalton at opencsw.org  Thu Jul 16 17:28:31 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 11:28:31 -0400
Subject: [csw-maintainers] cswclassutils
Message-ID: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>

Hi Peter,

I just added two new i/r scripts to the class utils package.  These
will automatically compile .el to .elc at install time.  If you're ok
with my changes, can you pop out a new release of cswclassutils?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/26c02ea9/attachment-0002.asc>

From pfelecan at opencsw.org  Thu Jul 16 17:51:57 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 16 Jul 2009 17:51:57 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247758050-sup-1395@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 16 Jul 2009 11\:28\:31 -0400")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
Message-ID: <x6ws68wrqa.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

> Hi Peter,

I don't think that you refer to me... however I have some questions

> I just added two new i/r scripts to the class utils package.  These
> will automatically compile .el to .elc at install time.  If you're ok
> with my changes, can you pop out a new release of cswclassutils?

What's the utility of this knowing that .elc files are architectural
neutral?

Packages containing .el and ..elc are usually split such as the .el
files are in an optional package and ..elc in mandatory ones.

Reducing the payload is not, in my opinion, a sufficient reason.
-- 
Peter


From bwalton at opencsw.org  Thu Jul 16 18:21:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 12:21:15 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6ws68wrqa.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
Message-ID: <1247760420-sup-3897@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Thu Jul 16 11:51:57 -0400 2009:

> Reducing the payload is not, in my opinion, a sufficient reason.

I thought that payload reduction in addition to a standard method for
packaging was a good enough win.

Not all packages that provide .el files provide a make target to
transform them to .elc (or even install them at all, for that matter).
With these scripts, the maintainer at most has to ensure that the .el
files are part of the package...I'm planning to extend GAR to do the
prototype and dependency stuff automatically (similar to how
cswpycompile will be integrated).[1]

I prefer to have the .el version around for reading, but that's partly
due to me being a relatively new emacs user and wanting to see how
things are done.  Having to pull in a separate package to get the
source files would work for me, but it wouldn't be as nice.

If your approach is preferred, I'll revert my commit.

Thanks
-Ben

[1] Only if it's decided this is useful.  I assumed it was, but...
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/78bc3d92/attachment-0002.asc>

From phil at bolthole.com  Thu Jul 16 18:28:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 16 Jul 2009 09:28:21 -0700
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247760420-sup-3897@ntdws12.chass.utoronto.ca>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
Message-ID: <20090716162821.GJ80367@bolthole.com>

On Thu, Jul 16, 2009 at 12:21:15PM -0400, Ben Walton wrote:
> Excerpts from Peter FELECAN's message of Thu Jul 16 11:51:57 -0400 2009:
> 
> > Reducing the payload is not, in my opinion, a sufficient reason.
> 
> I thought that payload reduction in addition to a standard method for
> packaging was a good enough win.

It should be mentioned that it can be a significant "payload reduction".
or.. erm.. it was in the case of python. Ben, why dont you give a specific
example of just how much space is saved?


Or, contrariwise, we may indeed want to only distribute the compiled
stuff and have separate src packages, as you say.

> I prefer to have the .el version around for reading, but that's partly
> due to me being a relatively new emacs user and wanting to see how
> things are done.

dont forget to mention/compare how debian and other distros do it.
not that we HAVE to do it the same way, but it's nice to have the
reference.



From pfelecan at opencsw.org  Thu Jul 16 18:44:58 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 16 Jul 2009 18:44:58 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <20090716162821.GJ80367@bolthole.com> (Philip Brown's message of
	"Thu\, 16 Jul 2009 09\:28\:21 -0700")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
Message-ID: <x6skgwwp9x.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Thu, Jul 16, 2009 at 12:21:15PM -0400, Ben Walton wrote:
>> Excerpts from Peter FELECAN's message of Thu Jul 16 11:51:57 -0400 2009:
>> 
>> > Reducing the payload is not, in my opinion, a sufficient reason.
>> 
>> I thought that payload reduction in addition to a standard method for
>> packaging was a good enough win.

I'm not opposed to having this class script for projects don't providing
make targets for .elc generation --- even if it's easy to patch for
that. BTW, can you provide an example of a project didn't providing the
direct or indirect elc target?

> It should be mentioned that it can be a significant "payload reduction".
> or.. erm.. it was in the case of python. Ben, why dont you give a specific
> example of just how much space is saved?

Quick example for abbrev:

.el  13226
.elc 11695

Note, however that I provide the .el files in compressed form

> Or, contrariwise, we may indeed want to only distribute the compiled
> stuff and have separate src packages, as you say.

This is how it's done for my packages containing emacs lisp code.

>> I prefer to have the .el version around for reading, but that's partly
>> due to me being a relatively new emacs user and wanting to see how
>> things are done.

This is why the .el are available in a separate, "source" package.

> dont forget to mention/compare how debian and other distros do it.
> not that we HAVE to do it the same way, but it's nice to have the
> reference.

Debian, and all the other distros that I know are doing it by providing
.elc in the main package and .el in a separate, optional package. What
they are doing in addition is implement a mechanism of recompiling
site-lisp packages on a package by package basis --- they have a emacs.d
directory containing a script by package installing in site-lisp; this
is something that is probably worth the effort to implement.

-- 
Peter


From bwalton at opencsw.org  Thu Jul 16 19:47:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 13:47:36 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6skgwwp9x.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
Message-ID: <1247764689-sup-557@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Thu Jul 16 12:44:58 -0400 2009:
> I'm not opposed to having this class script for projects don't providing
> make targets for .elc generation --- even if it's easy to patch for

Sure, it can be patched around, but that leaves the work to each
maintainer.

> that. BTW, can you provide an example of a project didn't providing the
> direct or indirect elc target?

Ruby is the example currently in my mind.  The misc/ directory
contains the ruby mode files, but nothing else.  As far as I can tell,
there is no make target to even get those files installed.  I don't
have an example where the .el files get installed but not
byte-compiled, so this leaves the maintainer doing work in either
case.

> Quick example for abbrev:
> 
> .el  13226
> .elc 11695

These line up roughly with what I've seen:
14452  inf-ruby.el
10008  inf-ruby.elc
 6747  ruby-electric.el
 6290  ruby-electric.elc
38653  ruby-mode.el
29260  ruby-mode.elc
 1746  ruby-style.el
 1983  ruby-style.elc
 4481  rubydb2x.el
 3952  rubydb2x.elc
 4613  rubydb3x.el
 4177  rubydb3x.elc

...so it's not huge (somewhere south of 50%), but it is something.  If
space were the ultimate goal, shipping .elc primarily and .el as a
source package would be better.  It was only part of my goal though.
Personally, I prefer the ship one, get both approach, but as I said,
I'll go with the preferred decision here.

> This is why the .el are available in a separate, "source" package.

This works, as I said, it's just not my preference.

> Debian, and all the other distros that I know are doing it by providing
> .elc in the main package and .el in a separate, optional package. What
> they are doing in addition is implement a mechanism of recompiling
> site-lisp packages on a package by package basis --- they have a emacs.d
> directory containing a script by package installing in site-lisp; this
> is something that is probably worth the effort to implement.

I don't think this is 'strictly' true.  It seems the preferred files
(for packages I looked at) were the .el ones.  There is then some
'fancy framework' for building/installing these for multiple flavours
of emacs, which can live together peacefully.  The debian policy[1]
doesn't seem to explicitly state which files are to be shipped
though.

Two packages I looked at quickly (from Ubuntu, not Debian, but...)
were css-mode and sepia.  In each case, they placed .el files in
/usr/share/emacs/site-lisp/ and then used a hook script in
/usr/lib/emacsen-common/packages/install/ to do byte compilation.

Thanks
-Ben

[1] http://www.debian.org/doc/packaging-manuals/debian-emacs-policy
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/9f74d3ba/attachment-0002.asc>

From ja at opencsw.org  Thu Jul 16 19:55:57 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Thu, 16 Jul 2009 19:55:57 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
Message-ID: <op.uw59fh0kjbmpvo@imac.lan>

Hi all,

I updated the packages for Munin 1.3.4 in testing.

- sample apache config (ready for include) in /etc/opt/csw/munin/apache/
- munin plugins moved to /opt/csw/libexec
- fix for using monospaced font

Feedback is welcome.

Juergen

-- 
Juergen Arndt


From pfelecan at opencsw.org  Thu Jul 16 20:34:23 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 16 Jul 2009 20:34:23 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247764689-sup-557@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 16 Jul 2009 13\:47\:36 -0400")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com> <x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
Message-ID: <x6ocrkwk7k.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

>> Debian, and all the other distros that I know are doing it by providing
>> .elc in the main package and .el in a separate, optional package. What
>> they are doing in addition is implement a mechanism of recompiling
>> site-lisp packages on a package by package basis --- they have a emacs.d
>> directory containing a script by package installing in site-lisp; this
>> is something that is probably worth the effort to implement.
>
> I don't think this is 'strictly' true.  It seems the preferred files
> (for packages I looked at) were the .el ones.  There is then some
> 'fancy framework' for building/installing these for multiple flavours
> of emacs, which can live together peacefully.  The debian policy[1]
> doesn't seem to explicitly state which files are to be shipped
> though.

It's strictly true for Emacs and that is the reference package. All the
others are just optional components for which an Emacs savvy user doesn't
have issues installing for himself or site wide. Of course, a savvy
packager do better for his users.

As stated in my previous message I'm not opposed to it but wished to
know the rational. As I'm not satisfied with that I'll continue to
supply the .elc in my packages and separately the .el

tat sat
-- 
Peter


From bwalton at opencsw.org  Thu Jul 16 21:35:38 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 16 Jul 2009 15:35:38 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6ocrkwk7k.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
Message-ID: <1247771967-sup-7519@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Thu Jul 16 14:34:23 -0400 2009:
> It's strictly true for Emacs and that is the reference package. All the
> others are just optional components for which an Emacs savvy user doesn't
> have issues installing for himself or site wide. Of course, a savvy
> packager do better for his users.

Ok.  I think we've both been clear on our preferences.  You have your
reasons and I'm ok with them.

> As stated in my previous message I'm not opposed to it but wished to
> know the rational. As I'm not satisfied with that I'll continue to
> supply the .elc in my packages and separately the .el

Ok.  I'm interested in standardizing something for this though.  It
would be better for the users if all of our emacs packaged bits
presented a similar experience.

Shall we agree then that any package providing emacs add-on
modules/functionality should provide .elc files in the primary package
and offer an _el version of the package to deliver the .el source
files?  That's what you've said, but should this become the formal
policy?

A side thought: Does it make sense to have this policy be different
than the python policy?  In theory, python add-ons could ship only the
.pyc/.pyo files and provide _py packages for the .py source...I'm
wondering this aloud for my own curiosity more than anything.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090716/c1ab3ba7/attachment-0002.asc>

From william at wbonnet.net  Thu Jul 16 22:31:44 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 16 Jul 2009 22:31:44 +0200
Subject: [csw-maintainers] sqlite3 3.6.16 is in testing
Message-ID: <4A5F8E30.1070805@wbonnet.net>

Hi

Sqlite3 package has been updated to latest version 3.6.16 and it 
available from testing

http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3_devel-3.6.16,REV=2009.07.16-SunOS5.8-all-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz

feedbacks are welcomed

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  Fri Jul 17 06:02:13 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 16 Jul 2009 21:02:13 -0700
Subject: [csw-maintainers] patience with releases for a day or two
In-Reply-To: <4A5F8E30.1070805@wbonnet.net>
References: <4A5F8E30.1070805@wbonnet.net>
Message-ID: <20090717040213.GD31978@bolthole.com>

Please note: James and I are trying out some things with releasing
packages. so pending packages will most likely be delayed another day or
two. but should definately go out monday worst case.
possibly a lot  sooner


From dam at opencsw.org  Fri Jul 17 09:25:50 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 17 Jul 2009 09:25:50 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
Message-ID: <A387A12F-F80E-4E3D-A87A-50435EE4C410@opencsw.org>

Hi Ben,

Am 16.07.2009 um 17:28 schrieb Ben Walton:
> I just added two new i/r scripts to the class utils package.  These
> will automatically compile .el to .elc at install time.  If you're ok
> with my changes, can you pop out a new release of cswclassutils?

Should there be automatic GAR support for it? I could imagine the
following:

- Add PKGFILES_EL to strip out files
- Automatically strip .elc files and generate them on install,
   change class of these files


Best regards

   -- Dago


From pfelecan at opencsw.org  Fri Jul 17 11:55:32 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 17 Jul 2009 11:55:32 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247771967-sup-7519@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 16 Jul 2009 15\:35\:38 -0400")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com> <x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
Message-ID: <x6fxcvws4r.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

>> As stated in my previous message I'm not opposed to it but wished to
>> know the rational. As I'm not satisfied with that I'll continue to
>> supply the .elc in my packages and separately the .el
>
> Ok.  I'm interested in standardizing something for this though.  It
> would be better for the users if all of our emacs packaged bits
> presented a similar experience.
>
> Shall we agree then that any package providing emacs add-on
> modules/functionality should provide .elc files in the primary package
> and offer an _el version of the package to deliver the .el source
> files?  That's what you've said, but should this become the formal
> policy?

I think so.

> A side thought: Does it make sense to have this policy be different
> than the python policy?  In theory, python add-ons could ship only the
> .pyc/.pyo files and provide _py packages for the .py source...I'm
> wondering this aloud for my own curiosity more than anything.

I'm leaning toward supplying separate _py packages but not having a
vested interest I remain neutral.
-- 
Peter


From pfelecan at opencsw.org  Fri Jul 17 11:56:22 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 17 Jul 2009 11:56:22 +0200
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <A387A12F-F80E-4E3D-A87A-50435EE4C410@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 17 Jul 2009 09\:25\:50 +0200")
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<A387A12F-F80E-4E3D-A87A-50435EE4C410@opencsw.org>
Message-ID: <x6bpnjws3d.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Ben,
>
> Am 16.07.2009 um 17:28 schrieb Ben Walton:
>> I just added two new i/r scripts to the class utils package.  These
>> will automatically compile .el to .elc at install time.  If you're ok
>> with my changes, can you pop out a new release of cswclassutils?
>
> Should there be automatic GAR support for it? I could imagine the
> following:
>
> - Add PKGFILES_EL to strip out files
> - Automatically strip .elc files and generate them on install,
>   change class of these files

I don't think so. Read the other messages of the thread.
-- 
Peter


From maciej at opencsw.org  Fri Jul 17 14:05:47 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 17 Jul 2009 13:05:47 +0100
Subject: [csw-maintainers] A single package that needs two source tarballs
Message-ID: <ce97543b0907170505i24ab5eb1y59d016809366bf41@mail.gmail.com>

I'm attempting to write a build for Netscape nss (network security
services). It turns out, nss source file needs nspr sources unpacked
in the same working directory. When attempting to compile nss, I'm
getting:

gmake[3]: Leaving directory
`/export/home/blizinski/opencsw/pkg/nss/trunk/work/build-isa-i386/nss-3.12.3/mozilla/security/coreconf'
gmake: *** No rule to make target ../../nsprpub/configure.  Stop.
gmake[2]: *** [../../nsprpub/configure] Error 1
gmake[2]: Leaving directory
`/export/home/blizinski/opencsw/pkg/nss/trunk/work/build-isa-i386/nss-3.12.3/mozilla/security/nss'
gmake[1]: *** [build-nss] Error 2
gmake[1]: Leaving directory `/export/home/blizinski/opencsw/pkg/nss/trunk'
gmake: *** [merge-isa-i386] Error 2

I've found the missing bit (nsprpub/configure) in the nspr package.
How would you go about it? Have a custom script to unpack nspr stuff
into the nss directory? Any other ideas?


From bwalton at opencsw.org  Fri Jul 17 14:17:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 17 Jul 2009 08:17:59 -0400
Subject: [csw-maintainers] A single package that needs two source
	tarballs
In-Reply-To: <ce97543b0907170505i24ab5eb1y59d016809366bf41@mail.gmail.com>
References: <ce97543b0907170505i24ab5eb1y59d016809366bf41@mail.gmail.com>
Message-ID: <1247832951-sup-209@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 17 08:05:47 -0400 2009:

Hi Maciej,

> I've found the missing bit (nsprpub/configure) in the nspr package.
> How would you go about it? Have a custom script to unpack nspr stuff
> into the nss directory? Any other ideas?

Have a look at the Makefile[1] for docbook-dtds.  It combines multiple
source tarballs into a single build directory.  You'll see how I
overrode the specific extract targets to put things where they need to
be.  I'm not sure if this will cover all of what you need for the nss
package, but it should get you started, anyway.

HTH.
-Ben

[1] http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/docbook-dtds/trunk/Makefile
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090717/e5c6a034/attachment-0002.asc>

From benny at opencsw.org  Fri Jul 17 14:27:36 2009
From: benny at opencsw.org (Benjamin von Mossner)
Date: Fri, 17 Jul 2009 14:27:36 +0200
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
Message-ID: <20090717122736.GB63617@vonmossner.de>

Hi,

> >From what i know, Benny has been working on an updated version. He is at a
> customer event right now, but he should be able to tell you whether i was
> right or not later on today.

so finally i had some time to repackage xpdf with GAR.
Updated version can be found in testing, included a fix for 0002724 
and and the latest security patches.

Talking about the bug:
I am definatly not an libXt expert, so i dont know if my fix is
appropriate. In xpdf/XPDFCore.cc the call to XtVaSetValues is
done with the wrong(?) widget and therefor fails to size the
window correctly. My diff therefor (and apparently the fix) is
kind of trivial.
[snip]
-    XtVaSetValues(drawArea, XmNwidth, width, XmNheight, height, NULL);
+    XtVaSetValues(shell, XmNwidth, width, XmNheight, height, NULL);
[snap]
But as mentioned above, i dont know if this is the right place to
fix this issue, nor why the unpatched code works fine on *BSD and
linux.
So if there is anyone familiar with libXt and/or stumbled upon
this or similar issue, i would be eager to know :-)

Cheers,
benny
-- 
/"\ ASCII RIBBON CAMPAIGN                  | Benjamin von Mossner
\ / AGAINST HTML MAIL                      | benny at vonmossner.de
 X  
/ \ multiple exclamation marks are a sure sign of a diseased mind


From ihsan at opencsw.org  Fri Jul 17 17:09:00 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Fri, 17 Jul 2009 17:09:00 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
Message-ID: <4A60940C.5080506@opencsw.org>

Hello,

The current Libtool package is unfortunately not fully functional. That
means, that certain packages do not build anymore.

As far as I could figure out, Libtool seems not to be aware of all the
compilers that we are using on our buildfarm.

Because I'm not an Libtool expert, I would like to ask if anybody with
good Libtool knowledge could have a look into that. Dago placed
everything, including the patch that has been used in the old version,
into Gar.



Thanks & Greetings,
Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From phil at bolthole.com  Fri Jul 17 18:47:41 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 17 Jul 2009 09:47:41 -0700
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <x6fxcvws4r.fsf@abhavana.free.fr>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
	<x6fxcvws4r.fsf@abhavana.free.fr>
Message-ID: <20090717164741.GQ80367@bolthole.com>

On Fri, Jul 17, 2009 at 11:55:32AM +0200, Peter FELECAN wrote:
> > A side thought: Does it make sense to have this policy be different
> > than the python policy?  In theory, python add-ons could ship only the
> > .pyc/.pyo files and provide _py packages for the .py source...I'm
> > wondering this aloud for my own curiosity more than anything.


I think that python is a different language, with different mechanisms, so
its ok if it has different policy.

I think that the route being followed by the current python maintainer,
makes sense for python.



From phil at bolthole.com  Fri Jul 17 18:49:28 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 17 Jul 2009 09:49:28 -0700
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <1247771967-sup-7519@ntdws12.chass.utoronto.ca>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
Message-ID: <20090717164928.GR80367@bolthole.com>

On Thu, Jul 16, 2009 at 03:35:38PM -0400, Ben Walton wrote:
> ...
> Ok.  I'm interested in standardizing something for this though.  It
> would be better for the users if all of our emacs packaged bits
> presented a similar experience.

this sounds like a good thing !



> Shall we agree then that any package providing emacs add-on
> modules/functionality should provide .elc files in the primary package
> and offer an _el version of the package to deliver the .el source
> files?  That's what you've said, but should this become the formal
> policy?

Personally, I think that it could at least be our official
recommendation/guideline.  not sure we have to push it all the way to a
"policy".


From phil at bolthole.com  Fri Jul 17 18:57:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 17 Jul 2009 09:57:57 -0700
Subject: [csw-maintainers] Call for volunteer : xpdf
In-Reply-To: <20090717122736.GB63617@vonmossner.de>
References: <4A5DA8EC.1080803@wbonnet.net>
	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>
	<20090717122736.GB63617@vonmossner.de>
Message-ID: <20090717165757.GT80367@bolthole.com>

On Fri, Jul 17, 2009 at 02:27:36PM +0200, Benjamin von Mossner wrote:
> But as mentioned above, i dont know if this is the right place to
> fix this issue, nor why the unpatched code works fine on *BSD and
> linux.
> So if there is anyone familiar with libXt and/or stumbled upon
> this or similar issue, i would be eager to know :-)
> 

well, you could try linking against our "new X11 libs", and see if that
magically fixes things.
(/opt/csw/X11/lib i think)


From pfelecan at opencsw.org  Fri Jul 17 19:29:56 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 17 Jul 2009 19:29:56 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
In-Reply-To: <4A60940C.5080506@opencsw.org> (Ihsan Dogan's message of "Fri\,
	17 Jul 2009 17\:09\:00 +0200")
References: <4A60940C.5080506@opencsw.org>
Message-ID: <x6hbxbusiz.fsf@abhavana.free.fr>

Ihsan Dogan <ihsan at opencsw.org> writes:

> The current Libtool package is unfortunately not fully functional. That
> means, that certain packages do not build anymore.
>
> As far as I could figure out, Libtool seems not to be aware of all the
> compilers that we are using on our buildfarm.
>
> Because I'm not an Libtool expert, I would like to ask if anybody with
> good Libtool knowledge could have a look into that. Dago placed
> everything, including the patch that has been used in the old version,
> into Gar.

Ihsan,

Can you give a concrete example such as we can try to help?

-- 
Peter


From bwalton at opencsw.org  Fri Jul 17 19:43:25 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 17 Jul 2009 13:43:25 -0400
Subject: [csw-maintainers] cswclassutils
In-Reply-To: <20090717164928.GR80367@bolthole.com>
References: <1247758050-sup-1395@ntdws12.chass.utoronto.ca>
	<x6ws68wrqa.fsf@abhavana.free.fr>
	<1247760420-sup-3897@ntdws12.chass.utoronto.ca>
	<20090716162821.GJ80367@bolthole.com>
	<x6skgwwp9x.fsf@abhavana.free.fr>
	<1247764689-sup-557@ntdws12.chass.utoronto.ca>
	<x6ocrkwk7k.fsf@abhavana.free.fr>
	<1247771967-sup-7519@ntdws12.chass.utoronto.ca>
	<20090717164928.GR80367@bolthole.com>
Message-ID: <1247852361-sup-8154@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 17 12:49:28 -0400 2009:
> Personally, I think that it could at least be our official
> recommendation/guideline.  not sure we have to push it all the way to a
> "policy".

Ok, then I'll withdraw (revert) my patch to cswclassutils.  Instead of
modifying GAR to automatically enable the use of the i/r scripts, I'll
look to add a toggle for compiling any .el files without a
corresponding .elc and generating the source packages automatically.
Something along the lines of GAR_EL_SOURCE_PKGS = 1.  Sound good to
everyone?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090717/32fbef08/attachment-0002.asc>

From trygvis at opencsw.org  Fri Jul 17 20:32:13 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Fri, 17 Jul 2009 20:32:13 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <op.uw59fh0kjbmpvo@imac.lan>
References: <op.uw59fh0kjbmpvo@imac.lan>
Message-ID: <4A60C3AD.7050800@opencsw.org>

Juergen Arndt wrote:
> Hi all,
> 
> I updated the packages for Munin 1.3.4 in testing.
> 
> - sample apache config (ready for include) in /etc/opt/csw/munin/apache/
> - munin plugins moved to /opt/csw/libexec
> - fix for using monospaced font
> 
> Feedback is welcome.

I'll try these when I get back home next week.

--
Trygve


From ihsan at opencsw.org  Sat Jul 18 11:13:44 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Sat, 18 Jul 2009 11:13:44 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
In-Reply-To: <x6hbxbusiz.fsf@abhavana.free.fr>
References: <4A60940C.5080506@opencsw.org> <x6hbxbusiz.fsf@abhavana.free.fr>
Message-ID: <4A619248.40203@opencsw.org>

Am 17.7.2009 19:29 Uhr, Peter FELECAN schrieb:

>> The current Libtool package is unfortunately not fully functional. That
>> means, that certain packages do not build anymore.
>>
>> As far as I could figure out, Libtool seems not to be aware of all the
>> compilers that we are using on our buildfarm.
>>
>> Because I'm not an Libtool expert, I would like to ask if anybody with
>> good Libtool knowledge could have a look into that. Dago placed
>> everything, including the patch that has been used in the old version,
>> into Gar.
>
> Can you give a concrete example such as we can try to help?

I'm getting this:

/opt/csw/bin/libtool --mode=compile /opt/csw/gcc4/bin/gcc  -O2 -pipe
-mcpu=v8 -I/opt/csw/include -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
-Wall -D_GNU_SOURCE -DNDEBUG -D_LIBRADIUS
-I/home/ihsan/gar/csw/mgar/pkg/freeradius/trunk/work/build-isa-sparcv8/freeradius-server-2.1.6/src
-c dict.c
libtool: compile: unable to infer tagged configuration
libtool: compile: specify a tag with `--tag'
gmake[6]: *** [dict.lo] Error 1



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From pfelecan at opencsw.org  Sat Jul 18 11:36:39 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Sat, 18 Jul 2009 11:36:39 +0200
Subject: [csw-maintainers] HELP NEEDED: libtool
In-Reply-To: <4A619248.40203@opencsw.org> (Ihsan Dogan's message of "Sat\,
	18 Jul 2009 11\:13\:44 +0200")
References: <4A60940C.5080506@opencsw.org> <x6hbxbusiz.fsf@abhavana.free.fr>
	<4A619248.40203@opencsw.org>
Message-ID: <x68wimuyc8.fsf@abhavana.free.fr>

Ihsan Dogan <ihsan at opencsw.org> writes:

> Am 17.7.2009 19:29 Uhr, Peter FELECAN schrieb:
>
>>> The current Libtool package is unfortunately not fully functional. That
>>> means, that certain packages do not build anymore.
>>>
>>> As far as I could figure out, Libtool seems not to be aware of all the
>>> compilers that we are using on our buildfarm.
>>>
>>> Because I'm not an Libtool expert, I would like to ask if anybody with
>>> good Libtool knowledge could have a look into that. Dago placed
>>> everything, including the patch that has been used in the old version,
>>> into Gar.
>>
>> Can you give a concrete example such as we can try to help?
>
> I'm getting this:
>
> /opt/csw/bin/libtool --mode=compile /opt/csw/gcc4/bin/gcc  -O2 -pipe
> -mcpu=v8 -I/opt/csw/include -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS
> -Wall -D_GNU_SOURCE -DNDEBUG -D_LIBRADIUS
> -I/home/ihsan/gar/csw/mgar/pkg/freeradius/trunk/work/build-isa-sparcv8/freeradius-server-2.1.6/src
> -c dict.c
> libtool: compile: unable to infer tagged configuration
> libtool: compile: specify a tag with `--tag'
> gmake[6]: *** [dict.lo] Error 1

A typical invocation is:

libtool --tag=CC   --mode=compile gcc -std=gnu99 -DHAVE_CONFIG_H
-I. -I.. -I..  -I/opt/csw/include  -I/opt/csw/gcc3/include
-I/opt/csw/gcc3/include/c++/3.4.5  -I/opt/csw/include   -O3 -march=i486
-MT snv.lo -MD -MP -MF .deps/snv.Tpo -c -o snv.lo snv.c

Please note in my example the --tag option which is missing in your
invocation. Can you try with --tag=CC

>From the info manual of libtool we can read:

    6.2 Tags
    ========
    
    Libtool supports multiple languages through the use of tags.
    Technically a tag corresponds to a set of configuration variables
    associated with a language.  These variables tell `libtool' how it
    should create objects and libraries for each language.
    
       Tags are defined at `configure'-time for each language activated in
    the package (see `LT_LANG' in *note LT_INIT::).  Here is the
    correspondence between language names and tags names.
    
    Language name      Tag name
    C                  CC
    C++                CXX
    Java               GCJ
    Fortran 77         F77
    Fortran            FC
    Windows Resource   RC
    
       `libtool' tries to automatically infer which tag to use from the
    compiler command being used to compile or link.  If it can't infer a
    tag, then it defaults to the configuration for the `C' language.
    
       The tag can also be specified using `libtool''s `--tag=TAG' option
    (*note Invoking libtool::).  It is a good idea to do so in `Makefile'
    rules, because that will allow users to substitute the compiler without
    relying on `libtool' inference heuristics.  When no tag is specified,
    `libtool' will default to `CC'; this tag always exists.
    
       Finally, the set of tags available in a particular project can be
    retrieved by tracing for the `LT_SUPPORTED_TAG' macro (*note Trace
    interface::).

What is intriguing is that the default tag is CC and libtool complains
that you are not using it explicitly. For this reason I would try to
use the /opt/csw/gcc4/bin in the PATH environment variable placed as to
be the first component refering toward a C compiler in the
configuration step.
-- 
Peter


From rupert at opencsw.org  Sun Jul 19 15:45:16 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 19 Jul 2009 15:45:16 +0200
Subject: [csw-maintainers] sqlite3 3.6.16 is in testing
In-Reply-To: <4A5F8E30.1070805@wbonnet.net>
References: <4A5F8E30.1070805@wbonnet.net>
Message-ID: <6af4270907190645p44ab966g86e01c57d9207b79@mail.gmail.com>

works for us.

On Thu, Jul 16, 2009 at 22:31, William Bonnet<william at wbonnet.net> wrote:
> Hi
>
> Sqlite3 package has been updated to latest version 3.6.16 and it available
> from testing
>
> http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3_devel-3.6.16,REV=2009.07.16-SunOS5.8-all-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-i386-CSW.pkg.gz
> http://mirror.opencsw.org/testing/sqlite3_rt-3.6.16,REV=2009.07.16-SunOS5.8-sparc-CSW.pkg.gz
>
> feedbacks are welcomed
>
> 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
>
>
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers
>


From rupert at opencsw.org  Sun Jul 19 16:11:19 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 19 Jul 2009 16:11:19 +0200
Subject: [csw-maintainers] remove openldap not possible with pkg-get and
	pkgutil
In-Reply-To: <94E2A58978FC324196A142DE8399AB7802315343@chsa1556.share.beluni.net>
References: <94E2A58978FC324196A142DE8399AB7802315343@chsa1556.share.beluni.net>
Message-ID: <6af4270907190711jdc8fe14s4c80318218688272@mail.gmail.com>

i was testing the new source code packages, and i did not pay enough
attention when upgrading. after doing a

pkgutil -t http://mirror.opencsw.org/opencsw/testing -i subversion
pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial

subversion was insisting on finding /opt/csw/lib/libldap-2.3.so.0,
while only /opt/csw/lib/libldap-2.4.so.0 was available. somehow
openldap got upgraded to
openldap_rt-2.4.16,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz from
testing.

i tried to remove it or downgrade it, but it did not work with
pkgutil, and pkg-get.

# pkgutil -r CSWoldaprt
Parsing catalog, may take a while...
Looking at installed CSW packages...
Examining dependencies for:
?CSWoldaprt (in use)
?CSWlibnet (in use)
?CSWsasl (in use)
?CSWossl (remove)
?CSWbdb4 (remove)
?CSWossldevel (remove)
?CSWosslutils (remove)
?CSWosslrt (in use)
?CSWcacertificates (remove)
?CSWcswclassutils (in use)
?CSWcommon (in use)
5 packages to remove. Do you want to continue? [Y,n]


# pkg-get -r CSWoldaprt

WARNING: ?the remove option is not very intelligent.
If there are multiple versions of a package with the same
PKG style name, it will remove the first one it can
(will continue in 5 seconds)
Starting remove operations now...

ERROR: no matching SysV PKG found.
CSWoldaprt CSWoldaprt
(either you mistyped it, or you need to updatecatalog, or
?it isnt available for your OSREV yet)

then i used pkgrm and pkgutil -u to install it again and it worked.


so i wonder:
* why oldap gets upgraded
* how to remove / downgrade packages

rupert.


From rupert at opencsw.org  Sun Jul 19 16:37:33 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 19 Jul 2009 16:37:33 +0200
Subject: [csw-maintainers] how to choose the correct version on testing?
Message-ID: <6af4270907190737gbbf341fx81087eb901ad038@mail.gmail.com>

i tried to test mercurial-1.3, and did
  pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u mercurial
  pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial
like described on http://mirror.opencsw.org/testing.html.

but both insisted installing the verison 1.2.1 which is also there.

is there an option to force it to take the newer one?

rupert.


From dam at opencsw.org  Sun Jul 19 17:35:36 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Jul 2009 17:35:36 +0200
Subject: [csw-maintainers] remove openldap not possible with pkg-get and
	pkgutil
In-Reply-To: <6af4270907190711jdc8fe14s4c80318218688272@mail.gmail.com>
References: <94E2A58978FC324196A142DE8399AB7802315343@chsa1556.share.beluni.net>
	<6af4270907190711jdc8fe14s4c80318218688272@mail.gmail.com>
Message-ID: <ECD4EFD9-BAB8-48F8-8E82-234A1598F4DD@opencsw.org>

Hi Rupert,

Am 19.07.2009 um 16:11 schrieb rupert THURNER:
> i was testing the new source code packages, and i did not pay enough
> attention when upgrading. after doing a
>
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -i subversion
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial
>
> subversion was insisting on finding /opt/csw/lib/libldap-2.3.so.0,
> while only /opt/csw/lib/libldap-2.4.so.0 was available. somehow
> openldap got upgraded to
> openldap_rt-2.4.16,REV=2009.06.29-SunOS5.8-sparc-CSW.pkg.gz from
> testing.

Well, if you install from testing pkgutil thinks that *all* packages  
from testing
within the dependency tree should be updated. However, OpenLDAP is not
finished yet and was put there for some very specific tests and not for
general use.

> i tried to remove it or downgrade it, but it did not work with
> pkgutil, and pkg-get.
>
> # pkgutil -r CSWoldaprt
> Parsing catalog, may take a while...
> Looking at installed CSW packages...
> Examining dependencies for:
>  CSWoldaprt (in use)
>  CSWlibnet (in use)
>  CSWsasl (in use)
>  CSWossl (remove)
>  CSWbdb4 (remove)
>  CSWossldevel (remove)
>  CSWosslutils (remove)
>  CSWosslrt (in use)
>  CSWcacertificates (remove)
>  CSWcswclassutils (in use)
>  CSWcommon (in use)
> 5 packages to remove. Do you want to continue? [Y,n]
>
>
> # pkg-get -r CSWoldaprt
>
> WARNING:  the remove option is not very intelligent.
> If there are multiple versions of a package with the same
> PKG style name, it will remove the first one it can
> (will continue in 5 seconds)
> Starting remove operations now...
>
> ERROR: no matching SysV PKG found.
> CSWoldaprt CSWoldaprt
> (either you mistyped it, or you need to updatecatalog, or
>  it isnt available for your OSREV yet)
>
> then i used pkgrm and pkgutil -u to install it again and it worked.
>
>
> so i wonder:
> * why oldap gets upgraded

Please see my comment above.

> * how to remove / downgrade packages

In this case you can try to force-sync with -S from pkg-get.


Best regards and sorry for the inconvenience

   -- Dago




From dam at opencsw.org  Sun Jul 19 17:37:20 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Jul 2009 17:37:20 +0200
Subject: [csw-maintainers] how to choose the correct version on testing?
In-Reply-To: <6af4270907190737gbbf341fx81087eb901ad038@mail.gmail.com>
References: <6af4270907190737gbbf341fx81087eb901ad038@mail.gmail.com>
Message-ID: <7D15FCF1-6D46-4D6F-821A-6AC75E6735D2@opencsw.org>

Hi Rupert,

Am 19.07.2009 um 16:37 schrieb rupert THURNER:
> i tried to test mercurial-1.3, and did
>  pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u mercurial
>  pkgutil -t http://mirror.opencsw.org/opencsw/testing -i mercurial
> like described on http://mirror.opencsw.org/testing.html.
>
> but both insisted installing the verison 1.2.1 which is also there.
>
> is there an option to force it to take the newer one?

No, because only the latest version is in the catalog. The maintainer  
should
remove the old one. Looks like both versions are from you :-) Do you  
really
need both versions in testing? The usual use is to only put the latest  
in
testing/.


Best regards

   -- Dago


From william at wbonnet.net  Sun Jul 19 18:43:08 2009
From: william at wbonnet.net (William Bonnet)
Date: Sun, 19 Jul 2009 18:43:08 +0200
Subject: [csw-maintainers] sqlite3 3.6.16 is in testing
In-Reply-To: <6af4270907190645p44ab966g86e01c57d9207b79@mail.gmail.com>
References: <4A5F8E30.1070805@wbonnet.net>
	<6af4270907190645p44ab966g86e01c57d9207b79@mail.gmail.com>
Message-ID: <4A634D1C.5070802@wbonnet.net>

Hi Rupert
> works for us.
>   
Thanks for the feed back. So far it is also working for me. I'll push it 
to current in a couple days if i have no negative feedbacks

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  Sun Jul 19 22:11:07 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 19 Jul 2009 22:11:07 +0200
Subject: [csw-maintainers] [csw-users] Munin 1.3.4 in testing
In-Reply-To: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
References: <op.uwnbhdamjbmpvo@ibm-1344209b5ff.ibm.com>
Message-ID: <CA5A04C4-192E-4ABF-A62B-CDEEBD41672B@opencsw.org>

Hi J?rgen,

Am 06.07.2009 um 14:25 schrieb Juergen Arndt:
> I've put Munin 1.3.4 into testing. It's the initial version for  
> OpenCSW. The software consists of the packages munin_master and  
> munin_node.

Please make sure the package is ARCHALL=1.


Best regards

   -- Dago

From dam at opencsw.org  Mon Jul 20 22:03:58 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 20 Jul 2009 22:03:58 +0200
Subject: [csw-maintainers] [csw-users] Fwd: Problem with pidgin and X11
	libs
In-Reply-To: <4A64C3E0.7010106@opencsw.org>
References: <4A646BC7.70007@gmail.com>
	<5BF57A34-6C08-46AA-AAB6-6EF7D9B0B48B@opencsw.org>
	<4A64C3E0.7010106@opencsw.org>
Message-ID: <2839A0F1-2EFB-445B-AD92-5D45E527DA41@opencsw.org>

Hi Mike,

Am 20.07.2009 um 21:22 schrieb Mike Watters:
> Chad asked me to build the new version of Pidgin ( the one that  
> fixes yahoo )
> since he does not have the bandwidth at the moment.  I have been  
> waiting on
> gtk+ to be released.  ( I am running the latest version on the gtk+  
> compiled
> from testing with NO problems )

Both gtk+ and devel are now updated on all buildfarm machines.


Best regards

   -- Dago


From rupert at opencsw.org  Tue Jul 21 00:33:01 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Tue, 21 Jul 2009 00:33:01 +0200
Subject: [csw-maintainers] mercurial package broken
In-Reply-To: <94E2A58978FC324196A142DE8399AB7802315C23@chsa1556.share.beluni.net>
References: <AcoIiGFjfLdQxjwaRZKhRw38hdYPrA==>
	<94E2A58978FC324196A142DE8399AB7802315C23@chsa1556.share.beluni.net>
Message-ID: <6af4270907201533j44108656u2739deab38d1fc13@mail.gmail.com>

the mercurial-1.3 package seems broken, we get here:

[Sun Jul 19 17:29:03 2009] [error] [client 166.14.3.11] RuntimeError:
No hgweb templates found in []
[Sun Jul 19 17:29:04 2009] [info] mod_wsgi (pid=18239): Create
interpreter 'chvp011vs035.csintra.net|/hg'.
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Target WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi' cannot be loaded as Python
module.
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'.
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] Traceback
(most recent call last):
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] ? File
"/opt/csw/apache2/data/hg09/hgwebdir.wsgi", line 3, in <module>
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] ? ? from
mercurial.hgweb.hgwebdir_mod import hgwebdir
[Sun Jul 19 17:29:58 2009] [error] [client 166.14.3.11] ImportError:
No module named mercurial.hgweb.hgwebdir_mod
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Target WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi' cannot be loaded as Python
module., referer: https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] mod_wsgi
(pid=26257): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'., referer:
https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] Traceback
(most recent call last):, referer: https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] ? File
"/opt/csw/apache2/data/hg09/hgwebdir.wsgi", line 3, in <module>,
referer: https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] ? ? from
mercurial.hgweb.hgwebdir_mod import hgwebdir, referer:
https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:10 2009] [error] [client 166.14.3.11] ImportError:
No module named mercurial.hgweb.hgwebdir_mod, referer:
https://chvp011vs035.csintra.net/hg
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] mod_wsgi
(pid=18240): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'.
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] Traceback
(most recent call last):
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/request.py", line
133, in run_wsgi
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 104, in __call__
[Sun Jul 19 17:30:50 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 134, in run_wsgi
[Sun Jul 19 17:30:51 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 323, in templater
[Sun Jul 19 17:30:51 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/templater.py", line 245,
in stylemap
[Sun Jul 19 17:30:51 2009] [error] [client 169.57.129.251]
RuntimeError: No hgweb templates found in []
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] mod_wsgi
(pid=25178): Exception occurred processing WSGI script
'/opt/csw/apache2/data/hg09/hgwebdir.wsgi'.
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] Traceback
(most recent call last):
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/request.py", line
133, in run_wsgi
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 104, in __call__
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 134, in run_wsgi
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/hgweb/hgwebdir_mod.py",
line 323, in templater
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251] ? File
"/opt/csw/lib/python/site-packages/mercurial/templater.py", line 245,
in stylemap
[Sun Jul 19 17:30:56 2009] [error] [client 169.57.129.251]
RuntimeError: No hgweb templates found in []

but an easy_install mercurial works.

rupert.


From bwalton at opencsw.org  Tue Jul 21 00:47:44 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 20 Jul 2009 18:47:44 -0400
Subject: [csw-maintainers] mercurial package broken
In-Reply-To: <6af4270907201533j44108656u2739deab38d1fc13@mail.gmail.com>
References: <AcoIiGFjfLdQxjwaRZKhRw38hdYPrA==>
	<94E2A58978FC324196A142DE8399AB7802315C23@chsa1556.share.beluni.net>
	<6af4270907201533j44108656u2739deab38d1fc13@mail.gmail.com>
Message-ID: <1248130010-sup-5486@ntdws12.chass.utoronto.ca>

Excerpts from rupert THURNER's message of Mon Jul 20 18:33:01 -0400 2009:

> '/opt/csw/apache2/data/hg09/hgwebdir.wsgi' cannot be loaded as Python
> module.

This likely needs to be rerolled after the python version update...not
positive, as I don't use it, but that's where I'd look 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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090720/4e98db6a/attachment-0002.asc>

From bwalton at opencsw.org  Tue Jul 21 02:30:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Mon, 20 Jul 2009 20:30:01 -0400
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
	<E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
Message-ID: <1248135892-sup-3787@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Jul 12 09:23:02 -0400 2009:
Hi Sebastian and Dago,

I put together a small patch (attached) for GAR to twiddle the
required bits for use of cswpycompile in an automatic fashion.  I
didn't make it happen implicitly, as after thinking more about the
changes I'll make to support automatic elisp handling, I didn't think
implicit was the right way to go here.

The patch will take any value of CSWPYCOMPILE to mean that
cswpycompile is added to SPKG_CLASSES, .pyc and .pyo files are
filtered from the merge by default, .py files are set to cswpycompile
in the prototype and CSWcswclassutils is added as a dependency.

If you're ok with the changes, I'll make the commit and document the
new variable.

> > EXTRA_PAX_ARGS = -s ',^.*/\(xcbgen/.*\.py\),.$(BUILD_PREFIX)/lib/ 
> > python/site-packages/\1,'
> > EXTRA_PAX_ARGS += -s ',^\.$(libdir)/python.*,,'

Dago: I didn't alter any of the PAX stuff for this patch, since it
didn't seem to be necessary...is this part of your recipe due to the
file relocation you're performing here?

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: autopy.patch
Type: application/octet-stream
Size: 2068 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090720/cc89a191/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090720/cc89a191/attachment-0002.asc>

From dam at opencsw.org  Tue Jul 21 09:43:53 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 21 Jul 2009 09:43:53 +0200
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <1248135892-sup-3787@ntdws12.chass.utoronto.ca>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
	<E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
	<1248135892-sup-3787@ntdws12.chass.utoronto.ca>
Message-ID: <865A7E17-7BE5-4EBB-AC06-48DD53F48E20@opencsw.org>

Hi Ben,

Am 21.07.2009 um 02:30 schrieb Ben Walton:
> Excerpts from Dagobert Michelsen's message of Sun Jul 12 09:23:02  
> -0400 2009:
> Hi Sebastian and Dago,
>
> I put together a small patch (attached) for GAR to twiddle the
> required bits for use of cswpycompile in an automatic fashion.  I
> didn't make it happen implicitly, as after thinking more about the
> changes I'll make to support automatic elisp handling, I didn't think
> implicit was the right way to go here.
>
> The patch will take any value of CSWPYCOMPILE to mean that
> cswpycompile is added to SPKG_CLASSES, .pyc and .pyo files are
> filtered from the merge by default, .py files are set to cswpycompile
> in the prototype and CSWcswclassutils is added as a dependency.

It should also make sure that the filtered files are all in locations
known to Python compilation. See my comment below.

> If you're ok with the changes, I'll make the commit and document the
> new variable.
>
>>> EXTRA_PAX_ARGS = -s ',^.*/\(xcbgen/.*\.py\),.$(BUILD_PREFIX)/lib/
>>> python/site-packages/\1,'
>>> EXTRA_PAX_ARGS += -s ',^\.$(libdir)/python.*,,'
>
> Dago: I didn't alter any of the PAX stuff for this patch, since it
> didn't seem to be necessary...is this part of your recipe due to the
> file relocation you're performing here?

Yes, because the original location was not checked for compilation. The
pax args relocate from the X11 directory to the standard location for
Python files. It would IMHO be better to read the path in the class
script and compile what it got, but as I understand it that was too
slow.


Best regards

   -- Dago


From maciej at opencsw.org  Tue Jul 21 11:59:16 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 21 Jul 2009 10:59:16 +0100
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
Message-ID: <ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>

Hi Markus,

Do you mind building cups-1.3.11? I've submitted a version bump and a
small bugfix. Can you please build the package?

Here's the mantis issue:
http://www.opencsw.org/bugtrack/view.php?id=3768

Reason for the request: version 1.3.11 contains a bugfix for the
scheduler code. In 1.3.9 cups-polld gets stuck after some time.

To other maintainers: I've filed a bug with mantis, Markus is marked
as active maintainer, but there is a number of unresolved issues in
mantis. When did you last hear from Markus?

Maciej


From bonivart at opencsw.org  Tue Jul 21 12:31:39 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Tue, 21 Jul 2009 12:31:39 +0200
Subject: [csw-maintainers] Just back from vacation (sort of...)
Message-ID: <625385e30907210331n650bb1a2l423d23cdb6acc91d@mail.gmail.com>

Just home after being a mechanic at the MotoGP roadracing event at
Sachsenring, Germany. Will start processing all mail now.

-- 
/peter


From maciej at opencsw.org  Tue Jul 21 14:52:04 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 21 Jul 2009 13:52:04 +0100
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
Message-ID: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>

Suppose we have this build:

PACKAGES = CSWfoo CSWbar
PKGFILES_CSWfoo = /opt/csw/bin/foo
PKGFILES_CSWbar = /opt/csw/bin/.*

There are two files:

/opt/csw/bin/foo
/opt/csw/bin/bar

Is the outcome of such build specified or unspecified? If specified,
what is the order of resolution? First wins, last wins? What's the
order?

Maciej


From dam at opencsw.org  Tue Jul 21 15:19:57 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 21 Jul 2009 15:19:57 +0200
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
In-Reply-To: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
References: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
Message-ID: <04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>

Hi Maciej,

Am 21.07.2009 um 14:52 schrieb Maciej (Matchek) Blizinski:
> Suppose we have this build:
>
> PACKAGES = CSWfoo CSWbar
> PKGFILES_CSWfoo = /opt/csw/bin/foo
> PKGFILES_CSWbar = /opt/csw/bin/.*
>
> There are two files:
>
> /opt/csw/bin/foo
> /opt/csw/bin/bar
>
> Is the outcome of such build specified or unspecified? If specified,
> what is the order of resolution? First wins, last wins? What's the
> order?

In parallel :-) The path must match include and not exclude.
The decision is shortcut. All pathes included in a package are
excluded from all other packages. That means IIRC the package
CSWfoo would contain 'foo' and CSWbar would contain 'foo' and
'bar'. The mechanism was written to simplify saying explicitly
what you want in what package and exclude stuff from other
packages. It was not meant for funky include-but-not-excluded-
if-unless-else :-) These are however Perl-regexps, so it is
allowed to write all advanced REs from Perl like lookahead
or negative matches. Something like this should do what you
want:

> PKGFILES_CSWfoo = /opt/csw/bin/foo
> PKGFILES_CSWbar = /opt/csw/bin/(?!foo)

This should be described here:
   <https://sourceforge.net/apps/trac/gar/wiki/Prototypes>
If there is something unclear please let me know.

See perlre(1) or "Mastering Regular Expressions" for further
reading (or post of course :-)

Here is the interesting expression:
   http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/gar.pkg.mk#L322 
 >


Best regards

   -- Dago


From maciej at opencsw.org  Tue Jul 21 16:19:13 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 21 Jul 2009 15:19:13 +0100
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
In-Reply-To: <04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>
References: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
	<04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>
Message-ID: <ce97543b0907210719u1854f96i8a70399636b589c3@mail.gmail.com>

On Tue, Jul 21, 2009 at 2:19 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
>> Is the outcome of such build specified or unspecified? If specified,
>> what is the order of resolution? First wins, last wins? What's the
>> order?
>
> In parallel :-) The path must match include and not exclude.
> The decision is shortcut. All pathes included in a package are
> excluded from all other packages. That means IIRC the package
> CSWfoo would contain 'foo' and CSWbar would contain 'foo' and
> 'bar'.

If I understand correctly, when CSWbar contains both bin/foo and
bin/bar, paths included in a package aren't excluded from other
packages -- foo is included in both packages.

> The mechanism was written to simplify saying explicitly
> what you want in what package and exclude stuff from other
> packages.

I'm currently working with the cups package. There are two packages:
cupsd and cupsclient. Binaries for those packages are put into the
same paths (/opt/csw/bin, /opt/csw/sbin), so one needs to cherrypick
the right binaries. I hoped to be able to say: take this and that for
cupsd, the rest is cupsclient. If they're processed in parallel, I
need to explicitly list all the binaries then. Is that right? Do you
have any other ideas?

Current prototypes of the mentioned packages:

http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/cups/trunk/files/CSWcupsd.prototype
http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/cups/trunk/files/CSWcupsclient.prototype

Maciej


From dam at opencsw.org  Tue Jul 21 16:28:32 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 21 Jul 2009 16:28:32 +0200
Subject: [csw-maintainers] Order of matching of PKGFILES_CSWfoo patterns
In-Reply-To: <ce97543b0907210719u1854f96i8a70399636b589c3@mail.gmail.com>
References: <ce97543b0907210552w701aa651k67e1fb3544671a38@mail.gmail.com>
	<04F01C80-8B52-4178-8508-04B3A27162B0@opencsw.org>
	<ce97543b0907210719u1854f96i8a70399636b589c3@mail.gmail.com>
Message-ID: <094D5D6E-9120-4A5F-B67F-2A96B4D0FEE5@opencsw.org>

Hi Maciej,

Am 21.07.2009 um 16:19 schrieb Maciej (Matchek) Blizinski:
> On Tue, Jul 21, 2009 at 2:19 PM, Dagobert Michelsen<dam at opencsw.org>  
> wrote:
>>> Is the outcome of such build specified or unspecified? If specified,
>>> what is the order of resolution? First wins, last wins? What's the
>>> order?
>>
>> In parallel :-) The path must match include and not exclude.
>> The decision is shortcut. All pathes included in a package are
>> excluded from all other packages. That means IIRC the package
>> CSWfoo would contain 'foo' and CSWbar would contain 'foo' and
>> 'bar'.
>
> If I understand correctly, when CSWbar contains both bin/foo and
> bin/bar, paths included in a package aren't excluded from other
> packages -- foo is included in both packages.

All files which match PKGFILES go into that package, inclusion
takes precedence. In your case that is foo and bar for CSWbar.

>> The mechanism was written to simplify saying explicitly
>> what you want in what package and exclude stuff from other
>> packages.
>
> I'm currently working with the cups package. There are two packages:
> cupsd and cupsclient. Binaries for those packages are put into the
> same paths (/opt/csw/bin, /opt/csw/sbin), so one needs to cherrypick
> the right binaries. I hoped to be able to say: take this and that for
> cupsd, the rest is cupsclient. If they're processed in parallel, I
> need to explicitly list all the binaries then. Is that right? Do you
> have any other ideas?

You usually specify PKGFILES for all but one packages which gets all
the rest not included in any package. And you can say
   PKGFILES_CSWcupsd = $(sbindir)/(cupsd|cupsfilter)
and leave out PKGFILES for CSWcupsclient.


Best regards

   -- Dago


From trygvis at opencsw.org  Tue Jul 21 20:47:20 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 21 Jul 2009 20:47:20 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <4A60C3AD.7050800@opencsw.org>
References: <op.uw59fh0kjbmpvo@imac.lan> <4A60C3AD.7050800@opencsw.org>
Message-ID: <4A660D38.7030600@opencsw.org>

Trygve Laugst?l wrote:
> Juergen Arndt wrote:
>> Hi all,
>>
>> I updated the packages for Munin 1.3.4 in testing.
>>
>> - sample apache config (ready for include) in /etc/opt/csw/munin/apache/
>> - munin plugins moved to /opt/csw/libexec
>> - fix for using monospaced font
>>
>> Feedback is welcome.
> 
> I'll try these when I get back home next week.

Can you do a rebuild with ARCH=all so I can test them on my x86 box?

--
Trygve


From ja at opencsw.org  Tue Jul 21 20:56:04 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Tue, 21 Jul 2009 20:56:04 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <4A660D38.7030600@opencsw.org>
References: <op.uw59fh0kjbmpvo@imac.lan> <4A60C3AD.7050800@opencsw.org>
	<4A660D38.7030600@opencsw.org>
Message-ID: <op.uxfljpnjjbmpvo@imac.lan>

On Tue, 21 Jul 2009 20:47:20 +0200, Trygve Laugst?l <trygvis at opencsw.org>  
wrote:

> Trygve Laugst?l wrote:
>> Juergen Arndt wrote:
>>> Hi all,
>>>
>>> I updated the packages for Munin 1.3.4 in testing.
>>>
>>> - sample apache config (ready for include) in  
>>> /etc/opt/csw/munin/apache/
>>> - munin plugins moved to /opt/csw/libexec
>>> - fix for using monospaced font
>>>
>>> Feedback is welcome.
>>  I'll try these when I get back home next week.
>
> Can you do a rebuild with ARCH=all so I can test them on my x86 box?

In testing there are platform independent packages:

munin_master-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz
munin_node-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz

Juergen

-- 
Juergen Arndt


From william at wbonnet.net  Tue Jul 21 22:30:18 2009
From: william at wbonnet.net (William Bonnet)
Date: Tue, 21 Jul 2009 22:30:18 +0200
Subject: [csw-maintainers] Just back from vacation (sort of...)
In-Reply-To: <625385e30907210331n650bb1a2l423d23cdb6acc91d@mail.gmail.com>
References: <625385e30907210331n650bb1a2l423d23cdb6acc91d@mail.gmail.com>
Message-ID: <4A66255A.2060405@wbonnet.net>

Hi Peter
> Just home after being a mechanic at the MotoGP roadracing event at
> Sachsenring, Germany. 
Mechanic at Sachsenring Moto GP event ... ? Waaaooohhhhh !

Impressing ! Full respect :)

Which team ? pilot ? A french guy is in the Honda team this year. He is 
not doing so well... but at least he takes part :)

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 Jul 22 03:53:46 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 21 Jul 2009 21:53:46 -0400
Subject: [csw-maintainers] GAR RFE: Variable(s) to tag files for classes
	instead of PROTOTYPE_FILTER
In-Reply-To: <865A7E17-7BE5-4EBB-AC06-48DD53F48E20@opencsw.org>
References: <51363.194.246.122.22.1246478936.squirrel@ssl.skayser.de>
	<3E39F7BB-726B-4250-8388-43EEF3D24B3F@opencsw.org>
	<4A4F8D65.9080203@opencsw.org>
	<EC85A63A-D2AB-4684-A4E8-7FF39ED8B1D9@opencsw.org>
	<4A59DD1E.2060205@opencsw.org>
	<E5971410-DC88-46B6-BFDA-AE2A59C09E5D@opencsw.org>
	<1248135892-sup-3787@ntdws12.chass.utoronto.ca>
	<865A7E17-7BE5-4EBB-AC06-48DD53F48E20@opencsw.org>
Message-ID: <1248227258-sup-2329@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Tue Jul 21 03:43:53 -0400 2009:

> Python files. It would IMHO be better to read the path in the class
> script and compile what it got, but as I understand it that was too
> slow.

Ok, having now read (more carefully) i.cswpycompile, I wonder if there
is a compromise that could be reached.  It seems to me that if the
class action script collected a set of directories (using `dirname
$dst`) it could then call compileall.py for each directory.

Mike, since you wrote this, what are your thoughts on that?  I ask
because I don't think writing generic pax rules for relocating .py
files into the standard location is either easy or wise in some cases.
Think of a package with the location of a .py file (or files) hard
coded somewhere after ./configure.  I'm sure that's an edge case
but...

If Mike doesn't think compileall.py per dir is feasible, I'd suggest
that we limit this 'automatic' functionality to things already in the
proper directory.  This would leave the maintainer to handle the
relocation however they'd like.  Alternately we could provide a knob
for users to specify pax rules in a more generic fashion.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090721/421b7fc9/attachment-0002.asc>

From bwalton at opencsw.org  Wed Jul 22 03:56:19 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 21 Jul 2009 21:56:19 -0400
Subject: [csw-maintainers] maintainers near toronto?
Message-ID: <1248227645-sup-7242@ntdws12.chass.utoronto.ca>


Hi All,

Are there any other maintainers in the Toronto area?  I've learned
that Damian Conway is giving a free talk at UofT on Monday.  If you're
out there and interested, ping me and I'll forward you the details.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090721/8ac7e8b5/attachment-0002.asc>

From skayser at opencsw.org  Wed Jul 22 10:45:42 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 22 Jul 2009 10:45:42 +0200
Subject: [csw-maintainers] (dbus fixes)
In-Reply-To: <4A5E40CA.5070305@wbonnet.net>
References: <4A5DA8EC.1080803@wbonnet.net>	<59785.194.246.122.22.1247652803.squirrel@ssl.skayser.de>	<4A5DC986.3020606@wbonnet.net>	<4A5E3B57.1030901@wbonnet.net>	<20090715204018.GA94635@bolthole.com>
	<4A5E40CA.5070305@wbonnet.net>
Message-ID: <4A66D1B6.4000302@opencsw.org>

Hi William,

William Bonnet wrote:
>> yes you should be the one to announce it :-)
>>   
> Master Yoda^W^WPhil, the one to announce it, will I be ;)
> 
>> once it passes through testing, thenI would say as soon as I officialy
>> clear it through newpkgs, you should send emails to both the user, and
>> "announce" lists, mentioning what is required.
>>   
> So thanks for more feedbacks then i'll request for a release.

I have just tried to update the dbus package from testing and have seen
two problems.

1) I had to issue "kill -9 `cat /opt/csw/var/run/dbus/pid`" during the
pkgrm of the old dbus package (when it says "Disabling
svc:/system/cswdbus ..."), not before. Doing it before just made SMF
start cswdbus again.

2) After having the new version installed the service doesn't come up
any more. From /var/svc/log/system-cswdbus:default.log

[ Jul 22 10:40:07 Executing start method
("/var/opt/csw/svc/method/svc-cswdbus start") ]
/var/opt/csw/svc/method/svc-cswdbus: smf_is_nonglobalzone: not found
--config-file specified but configuration file
/opt/csw/etc/dbus-1/system.conf already requested
dbus failed to start: error 1
[ Jul 22 10:40:07 Method "start" exited with status 1 ]

# pkginfo -x CSWdbus
CSWdbus  dbus - DBus Message Bus System
         (i386) 1.2.12,REV=2009.07.15

The whole upgrade console log can be found at [1].

Sebastian

[1] http://pastebin.com/m6956c9d3


From Darin.Perusich at cognigencorp.com  Wed Jul 22 13:40:27 2009
From: Darin.Perusich at cognigencorp.com (Darin Perusich)
Date: Wed, 22 Jul 2009 07:40:27 -0400
Subject: [csw-maintainers] maintainers near toronto?
In-Reply-To: <1248227645-sup-7242@ntdws12.chass.utoronto.ca>
References: <1248227645-sup-7242@ntdws12.chass.utoronto.ca>
Message-ID: <4A66FAAB.60009@cognigencorp.com>

I'm only an hour and in a half away in Buffalo, NY but won't be making
the track north for this. It's probably a little late but there's a
local Perlmonger group I could have forwarded the info on too but they
likely already no.

Ben Walton wrote:
> Hi All,
> 
> Are there any other maintainers in the Toronto area?  I've learned
> that Damian Conway is giving a free talk at UofT on Monday.  If you're
> out there and interested, ping me and I'll forward you the details.
> 
> -Ben

-- 
Darin Perusich
Unix Systems Administrator
Cognigen Corporation
395 Youngs Rd.
Williamsville, NY 14221
Phone: 716-633-3463
Email: darinper at cognigencorp.com


From dam at opencsw.org  Wed Jul 22 14:18:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 22 Jul 2009 14:18:51 +0200
Subject: [csw-maintainers] Updated version of gdb
In-Reply-To: <7B8BE778-3711-4325-A1E5-272F03709513@opencsw.org>
References: <827670CA-A98F-44B4-A09A-DDE7D1E0E038@opencsw.org>
	<c7dddeaa0901241142hf6e3b08m4599b8ce7ed072f1@mail.gmail.com>
	<7B8BE778-3711-4325-A1E5-272F03709513@opencsw.org>
Message-ID: <EF495098-82C9-4418-91F3-10A9FB3BC0D4@opencsw.org>

Hi,

there is now a new version of gdb in testing/:

   gdb-6.8,REV=2009.07.22-SunOS5.8-i386-CSW.pkg.gz
   gdb-6.8,REV=2009.07.22-SunOS5.8-sparc-CSW.pkg.gz

There are some failing tests which may or may not be serious.
There are both 32/64 bit versions for both x86 and sparc.
Please let me know if it works so it can be released soon.


Best regards

   -- Dago


From ellson at opencsw.org  Wed Jul 22 18:05:47 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 12:05:47 -0400
Subject: [csw-maintainers] howto control which files go into which part
 of split package
In-Reply-To: <1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
	<1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>
Message-ID: <4A6738DB.5080105@opencsw.org>

More questions:

1.  How do I get a plugin in one subpackage to recognize that the 
library it depends on comes from the main subpackage?

"make package" fails on "CSWgraphvizgd"  with:

     ERROR: cannot find package for libcdt.so.4
     make: *** [pkgcheck-CSWgraphvizgd] Error 2

even though the main package "CSWgraphviz" was already packaged and 
contains:

     CSWgraphviz/root/opt/csw/lib/libcdt.so.4.0.0

(but no "libcdt.so.4", which would normally be a soflink, is that the 
problem?)





2.  Can I modify the packaging parts of the Makefile, and redo "make 
package"  without rebuilding from
"make clean"  ?



John



From skayser at opencsw.org  Wed Jul 22 20:01:20 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Wed, 22 Jul 2009 20:01:20 +0200 (CEST)
Subject: [csw-maintainers] howto control which files go into which part
 of split package
In-Reply-To: <4A6738DB.5080105@opencsw.org>
References: <4A5E00BC.3090800@opencsw.org>
	<1D2622B6-2886-429B-A5DB-E9097402A1D8@opencsw.org>
	<4A6738DB.5080105@opencsw.org>
Message-ID: <53663.194.246.122.22.1248285680.squirrel@ssl.skayser.de>

Hi John,

John Ellson wrote:
> More questions:
>
> 1.  How do I get a plugin in one subpackage to recognize that the
> library it depends on comes from the main subpackage?
>
> "make package" fails on "CSWgraphvizgd"  with:
>
>      ERROR: cannot find package for libcdt.so.4
>      make: *** [pkgcheck-CSWgraphvizgd] Error 2
>
> even though the main package "CSWgraphviz" was already packaged and
> contains:
>
>      CSWgraphviz/root/opt/csw/lib/libcdt.so.4.0.0
>
> (but no "libcdt.so.4", which would normally be a soflink, is that the
> problem?)

that is the checkpkg script complaining and i don't know whether there is
a way to have checkpkg learn about the packages that are part of the
build. I have set ENABLE_CHECK to 0 (disables checkpkg) in such cases and
ran checkpkg on the packages manually (if in doubt).

> 2.  Can I modify the packaging parts of the Makefile, and redo "make
> package"  without rebuilding from
> "make clean"  ?

gmake repackage

Sebastian



From ellson at opencsw.org  Wed Jul 22 23:15:46 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 17:15:46 -0400
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <4A677CAB.8050808@opencsw.org>
References: <4A677CAB.8050808@opencsw.org>
Message-ID: <4A678182.9020005@opencsw.org>

On 07/22/2009 04:55 PM, John Ellson wrote:
> Dago,
>
> Any chance you could rebuild cairo to include SVG support, please?
>
> I can take a shot at it if you like?
>
>
> John


So apparently the reason its not provided is because there is a circular 
depdendency with librsvg ?


Why does a circular dependency matter?    Try to install either and you 
get both - where is the ambiguity?


John


From ellson at opencsw.org  Wed Jul 22 22:55:07 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 16:55:07 -0400
Subject: [csw-maintainers] libcairo svg surface
Message-ID: <4A677CAB.8050808@opencsw.org>

Dago,

Any chance you could rebuild cairo to include SVG support, please?

I can take a shot at it if you like?


John


From ellson at opencsw.org  Thu Jul 23 00:03:39 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 18:03:39 -0400
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <4A678182.9020005@opencsw.org>
References: <4A677CAB.8050808@opencsw.org> <4A678182.9020005@opencsw.org>
Message-ID: <4A678CBB.2030208@opencsw.org>

On 07/22/2009 05:15 PM, John Ellson wrote:
> On 07/22/2009 04:55 PM, John Ellson wrote:
>> Dago,
>>
>> Any chance you could rebuild cairo to include SVG support, please?
>>
>> I can take a shot at it if you like?
>>
>>
>> John
>
>
> So apparently the reason its not provided is because there is a 
> circular depdendency with librsvg ?
>
>
> Why does a circular dependency matter?    Try to install either and 
> you get both - where is the ambiguity?

It builds fine with svg enabled.  There is no library dependency on 
librsvg.  I think it may just be a testing dependency.


John


From ellson at opencsw.org  Thu Jul 23 03:15:18 2009
From: ellson at opencsw.org (John Ellson)
Date: Wed, 22 Jul 2009 21:15:18 -0400
Subject: [csw-maintainers] New graphviz packages in testing
Message-ID: <4A67B9A6.5030309@opencsw.org>

There are new graphviz packages in testing:

   graphviz        Graph Visualization Tools
   graphvizdevel      headers etc. for development
   graphvizdoc        documentation
   graphvizgraphs     example graphs
   graphvizgd         renderers using gd
   graphvizsharp      language binding for C#
   graphvizguile      language binding for guile
   graphvizperl       language binding for perl
   graphvizpython     language binding for python
   graphviztcl        language binding for tcl

I can't actually install from testing.  I'm not sure why?

     # pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u graphviz
     ...
     Updated description file
     INTERNAL ERROR: cannot get remote version for CSWexpat
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWpango
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWlibcairo
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWgts
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWlibxrender
     Perhaps your catalog is out of date
     INTERNAL ERROR: cannot get remote version for CSWgs
     Perhaps your catalog is out of date
     Error: dependancies for graphviz not up to date.

Can anyone help with this?

Otherwise, perhaps I could ask for just a packaging review?

This graphviz package set is built from the latest stable 
graphviz-2.24.0 release from http://www.graphviz.org/
It uses pango/cairo by default to achieve quality anti-aliased line 
drawings and UTF8 character support,
so the "gd" renderer is mostly obsolete, unless you need GIF support.

"graphviz" obsoletes "graphviz2" and "graphvis"


checkpkg fails on graphvizgd and on the language binding subpackages 
because I don't know how to tell
the subpackages about libraries provided by the main package. Help please?

     $ checkpkg graphvizgd-2.24.0,REV=2009.07.23-SunOS5.8-i386-CSW.pkg.gz
     ...
     Building index from SVR4 installed packages database...
     (May take a while)
     Cross-referencing indexes...
       found SUNWxwplt for libX11.so.4
       found CSWxpm for libXpm.so.4.11
       found SUNWcsl for libc.so.1
     ERROR: cannot find package for libcdt.so.4
     $


Currently there is no support for SVG images in nodes because libcairo 
was built without SVG support.

There are some other language bindings that failed to build:   java, 
ocaml, lua, php
I'll take another look at those after this set is released.


John


From skayser at opencsw.org  Thu Jul 23 13:21:19 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 23 Jul 2009 13:21:19 +0200
Subject: [csw-maintainers] New graphviz packages in testing
In-Reply-To: <4A67B9A6.5030309@opencsw.org>
References: <4A67B9A6.5030309@opencsw.org>
Message-ID: <4A6847AF.6010601@opencsw.org>

John Ellson wrote:
> There are new graphviz packages in testing:
> 
>    graphviz        Graph Visualization Tools
>    graphvizdevel      headers etc. for development
>    graphvizdoc        documentation
>    graphvizgraphs     example graphs
>    graphvizgd         renderers using gd
>    graphvizsharp      language binding for C#
>    graphvizguile      language binding for guile
>    graphvizperl       language binding for perl
>    graphvizpython     language binding for python
>    graphviztcl        language binding for tcl
> 
> I can't actually install from testing.  I'm not sure why?
> 
>      # pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u graphviz
>      ...
>      Updated description file
>      INTERNAL ERROR: cannot get remote version for CSWexpat
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWpango
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWlibcairo
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWgts
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWlibxrender
>      Perhaps your catalog is out of date
>      INTERNAL ERROR: cannot get remote version for CSWgs
>      Perhaps your catalog is out of date
>      Error: dependancies for graphviz not up to date.
> 
> Can anyone help with this?

Have you tried using pkgutil for the same purpose? pkg-get once couldn't
handle dependencies that spanned different repositories (graphviz from
testing, deps from current in your case). I don't know whether that is
still the case.

If it is indeed the pkg-get issue we should mention this on the testing
page [1] so that people know about it.

Sebastian

[1] http://mirror.opencsw.org/testing.html


From maciej at opencsw.org  Thu Jul 23 17:25:35 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 16:25:35 +0100
Subject: [csw-maintainers] The source code of www.opencsw.org
Message-ID: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>

Hello OpenCSW web site maintainers,

I've been looking at www.opencsw.org website, and was wondering - if I
wanted to suggest a change for it, or work on the website, how would I
do that? I was looking for source code, but I didn't find it. Is the
source code of www.opencsw.org checked into subversion? If not, what
do you think of submitting the code there? It would make it easier to
suggest a change, or submit a patch, and generally make things more
open.

Maciej


From ellson at opencsw.org  Thu Jul 23 17:43:50 2009
From: ellson at opencsw.org (John Ellson)
Date: Thu, 23 Jul 2009 11:43:50 -0400
Subject: [csw-maintainers] New graphviz packages in testing
In-Reply-To: <4A6847AF.6010601@opencsw.org>
References: <4A67B9A6.5030309@opencsw.org> <4A6847AF.6010601@opencsw.org>
Message-ID: <4A688536.7050404@opencsw.org>

On 07/23/2009 07:21 AM, Sebastian Kayser wrote:
> John Ellson wrote:
>    
>> There are new graphviz packages in testing:
>>
>>     graphviz        Graph Visualization Tools
>>     graphvizdevel      headers etc. for development
>>     graphvizdoc        documentation
>>     graphvizgraphs     example graphs
>>     graphvizgd         renderers using gd
>>     graphvizsharp      language binding for C#
>>     graphvizguile      language binding for guile
>>     graphvizperl       language binding for perl
>>     graphvizpython     language binding for python
>>     graphviztcl        language binding for tcl
>>
>> I can't actually install from testing.  I'm not sure why?
>>
>>       # pkg-get -s http://mirror.opencsw.org/opencsw/testing -U -u graphviz
>>       ...
>>       Updated description file
>>       INTERNAL ERROR: cannot get remote version for CSWexpat
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWpango
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWlibcairo
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWgts
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWlibxrender
>>       Perhaps your catalog is out of date
>>       INTERNAL ERROR: cannot get remote version for CSWgs
>>       Perhaps your catalog is out of date
>>       Error: dependancies for graphviz not up to date.
>>
>> Can anyone help with this?
>>      
> Have you tried using pkgutil for the same purpose? pkg-get once couldn't
> handle dependencies that spanned different repositories (graphviz from
> testing, deps from current in your case). I don't know whether that is
> still the case.
>
> If it is indeed the pkg-get issue we should mention this on the testing
> page [1] so that people know about it.
>
> Sebastian
>
> [1] http://mirror.opencsw.org/testing.html
>    


pkgutil worked for me.

I've fixed some packaging bugs in graphviz this morning, and now the 
installed binaries
seem to be working.

I then removed graphviz and retried the pkg-get method.   I still get 
the INTERNAL ERRORs, but
this time the installation eventually succeeded.

John



From bwalton at opencsw.org  Thu Jul 23 17:45:53 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 11:45:53 -0400
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
Message-ID: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
> source code of www.opencsw.org checked into subversion? If not, what
> do you think of submitting the code there? It would make it easier to
> suggest a change, or submit a patch, and generally make things more
> open.

+1 for making the website code open and public.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/744a10ae/attachment-0002.asc>

From bwalton at opencsw.org  Thu Jul 23 17:46:42 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 11:46:42 -0400
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
Message-ID: <1248363975-sup-2856@ntdws12.chass.utoronto.ca>

Excerpts from Ben Walton's message of Thu Jul 23 11:45:53 -0400 2009:
> +1 for making the website code open and public.

...obviously files containing passwords to databases should be
excepted.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/fe8ca014/attachment-0002.asc>

From trygvis at opencsw.org  Thu Jul 23 17:50:55 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Thu, 23 Jul 2009 17:50:55 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
Message-ID: <4A6886DF.30507@opencsw.org>

Ben Walton wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>> source code of www.opencsw.org checked into subversion? If not, what
>> do you think of submitting the code there? It would make it easier to
>> suggest a change, or submit a patch, and generally make things more
>> open.
> 
> +1 for making the website code open and public.

+1

Would be nice to be able to turn the package database into a separate 
application, I have clients that might see use something like that.

--
Trygve


From phil at bolthole.com  Thu Jul 23 17:53:11 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 08:53:11 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248363911-sup-6322@ntdws12.chass.utoronto.ca>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
Message-ID: <20090723155311.GD53578@bolthole.com>

On Thu, Jul 23, 2009 at 11:45:53AM -0400, Ben Walton wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
> > source code of www.opencsw.org checked into subversion? If not, what
> > do you think of submitting the code there? It would make it easier to
> > suggest a change, or submit a patch, and generally make things more
> > open.
> 
> +1 for making the website code open and public.

absolutely not "public".

in contrast, any MAINTAINER who wants to look at it, already can. 
you just need to know where to look :-)


From trygvis at opencsw.org  Thu Jul 23 17:56:49 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 23 Jul 2009 17:56:49 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723155311.GD53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
Message-ID: <4A688841.8020503@opencsw.org>

Philip Brown wrote:
> On Thu, Jul 23, 2009 at 11:45:53AM -0400, Ben Walton wrote:
>> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>>> source code of www.opencsw.org checked into subversion? If not, what
>>> do you think of submitting the code there? It would make it easier to
>>> suggest a change, or submit a patch, and generally make things more
>>> open.
>> +1 for making the website code open and public.
> 
> absolutely not "public".

Why not?

--
Trygve


From bwalton at opencsw.org  Thu Jul 23 17:59:19 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 11:59:19 -0400
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723155311.GD53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
Message-ID: <1248364654-sup-4834@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Thu Jul 23 11:53:11 -0400 2009:

> absolutely not "public".

Aside for the issue of passwords, why not public?

> in contrast, any MAINTAINER who wants to look at it, already can. 
> you just need to know where to look :-)

Care to share this then?  How is it versioned?  Where does it live?
How would a maintainer make edits and commit changes?  I don't
personally care, but it should be knowledge available to all
maintainers.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/2aba8e21/attachment-0002.asc>

From phil at bolthole.com  Thu Jul 23 18:03:07 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:03:07 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A688841.8020503@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
Message-ID: <20090723160307.GF53578@bolthole.com>

On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
>> absolutely not "public".
>
> Why not?

security issues.



From phil at bolthole.com  Thu Jul 23 18:06:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:06:56 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6886DF.30507@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<4A6886DF.30507@opencsw.org>
Message-ID: <20090723160656.GH53578@bolthole.com>

On Thu, Jul 23, 2009 at 05:50:55PM +0200, Trygve Laugst??l wrote:
> Ben Walton wrote:
>> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>>> source code of www.opencsw.org checked into subversion? If not, what
>>> do you think of submitting the code there? It would make it easier to
>>> suggest a change, or submit a patch, and generally make things more
>>> open.
>>
>> +1 for making the website code open and public.
>
> +1
>
> Would be nice to be able to turn the package database into a separate  
> application, I have clients that might see use something like that.

Umm... I doubt your clients would really want to use our code.
Euuu, ick ;-)

besides, there isnt really much "code" for that. What code would they
actually use? the
 "taking apart a package, to analyze and register in a database" bit?

If you're serious about that, I'll consider making my ugly mess something
that is more suitable for publication.


From phil at bolthole.com  Thu Jul 23 18:15:58 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:15:58 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723160656.GH53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<4A6886DF.30507@opencsw.org> <20090723160656.GH53578@bolthole.com>
Message-ID: <20090723161558.GI53578@bolthole.com>

On Thu, Jul 23, 2009 at 09:06:56AM -0700, Philip Brown wrote:
> ....
> besides, there isnt really much "code" for that. What code would they
> actually use? the
>  "taking apart a package, to analyze and register in a database" bit?
> 
> If you're serious about that, I'll consider making my ugly mess something
> that is more suitable for publication.

in mentally reviewing what we do for that.. the code is really, really
specific to opencsw. I doubt it would be useful for anywhere else.
It's very mantis-specific, for example.
But as I said, tell me more about what functionality you think would be
actually useful about your clients, and I'll see if there's any code
sharing that makes sense.



From pfelecan at opencsw.org  Thu Jul 23 18:14:20 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 23 Jul 2009 18:14:20 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <1248364654-sup-4834@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Thu\, 23 Jul 2009 11\:59\:19 -0400")
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
Message-ID: <x6ljmfqsv7.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

> Excerpts from Philip Brown's message of Thu Jul 23 11:53:11 -0400 2009:
>
>> absolutely not "public".
>
> Aside for the issue of passwords, why not public?
>
>> in contrast, any MAINTAINER who wants to look at it, already can. 
>> you just need to know where to look :-)
>
> Care to share this then?  How is it versioned?  Where does it live?
> How would a maintainer make edits and commit changes?  I don't
> personally care, but it should be knowledge available to all
> maintainers.

I agree with Ben. And I say that there is too much opacity here en
there... Wasn't communication a theme for a month of this year? Maybe we
should start with internal communication.
-- 
Peter


From phil at bolthole.com  Thu Jul 23 18:24:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 09:24:45 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <x6ljmfqsv7.fsf@abhavana.free.fr>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
	<x6ljmfqsv7.fsf@abhavana.free.fr>
Message-ID: <20090723162445.GK53578@bolthole.com>

On Thu, Jul 23, 2009 at 06:14:20PM +0200, Peter FELECAN wrote:
> ...
> I agree with Ben. And I say that there is too much opacity here en
> there... Wasn't communication a theme for a month of this year? Maybe we
> should start with internal communication.

ok.. what's lacking in "communication" here? :-)
I have "communicated" that it is already available to look at, by any
maintainer. 

I will even go one step further, by mentioning that for anyone who cares,
it is in a fairly obvious place under /var, on bender.

Beyond that, this isnt a "communication" issue any more.
"internal communication", is not synonymous with "public communication".
In some ways, that is a contradiction: "internal" vs "public".

And I will repeat: anyone who wishes to do actual work on this stuff,
should let me or Ihsan know, and we will be happy to bore you to death with
more details off-list.




From ellson at opencsw.org  Thu Jul 23 18:44:05 2009
From: ellson at opencsw.org (John Ellson)
Date: Thu, 23 Jul 2009 12:44:05 -0400
Subject: [csw-maintainers] howto create catalog files ?
Message-ID: <4A689355.1040405@opencsw.org>

What is the command used to generate a catalog of a set of packages in a 
directory?


Now that I can build graphviz the OpenCSW way,  I'm planning to
use local resources to build nightly development snapshot pkgs
for solaris users, and distribute them like we do with rpms
and debs from our http://www.graphviz.org/ site.

John





From pfelecan at opencsw.org  Thu Jul 23 18:44:27 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 23 Jul 2009 18:44:27 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723162445.GK53578@bolthole.com> (Philip Brown's message of
	"Thu\, 23 Jul 2009 09\:24\:45 -0700")
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
	<x6ljmfqsv7.fsf@abhavana.free.fr> <20090723162445.GK53578@bolthole.com>
Message-ID: <x6hbx3qrh0.fsf@abhavana.free.fr>

Philip Brown <phil at bolthole.com> writes:

> On Thu, Jul 23, 2009 at 06:14:20PM +0200, Peter FELECAN wrote:
>> ...
>> I agree with Ben. And I say that there is too much opacity here en
>> there... Wasn't communication a theme for a month of this year? Maybe we
>> should start with internal communication.
>
> ok.. what's lacking in "communication" here? :-)

e.g. speaking, writing, singing, dancing, &c, about methods, procedures
and tools.

> I will even go one step further, by mentioning that for anyone who cares,
> it is in a fairly obvious place under /var, on bender.

sure, the source is the documentation; heard it so often.

> "internal communication", is not synonymous with "public communication".
> In some ways, that is a contradiction: "internal" vs "public".

now that the maintainer's list is "public" we certainly have an issue.
-- 
Peter


From bwalton at opencsw.org  Thu Jul 23 18:48:52 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 23 Jul 2009 12:48:52 -0400
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <4A689355.1040405@opencsw.org>
References: <4A689355.1040405@opencsw.org>
Message-ID: <1248367682-sup-6580@ntdws12.chass.utoronto.ca>

Excerpts from John Ellson's message of Thu Jul 23 12:44:05 -0400 2009:
> What is the command used to generate a catalog of a set of packages in a 
> directory?

You'd want the bldcat tool that Peter B put together.  It's part of
the CSWpkgutil package.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090723/0f06191d/attachment-0002.asc>

From pfelecan at opencsw.org  Thu Jul 23 18:48:07 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 23 Jul 2009 18:48:07 +0200
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <4A689355.1040405@opencsw.org> (John Ellson's message of "Thu\,
	23 Jul 2009 12\:44\:05 -0400")
References: <4A689355.1040405@opencsw.org>
Message-ID: <x6d47rqraw.fsf@abhavana.free.fr>

John Ellson <ellson at opencsw.org> writes:

> Now that I can build graphviz the OpenCSW way,  I'm planning to
> use local resources to build nightly development snapshot pkgs
> for solaris users, and distribute them like we do with rpms
> and debs from our http://www.graphviz.org/ site.

are you a member of the graphiz project? I read your bio on the
maintainer's page of our site and there wasn't a lot to read...
think: communication.

-- 
Peter


From ellson at opencsw.org  Thu Jul 23 18:52:03 2009
From: ellson at opencsw.org (John Ellson)
Date: Thu, 23 Jul 2009 12:52:03 -0400
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <x6d47rqraw.fsf@abhavana.free.fr>
References: <4A689355.1040405@opencsw.org> <x6d47rqraw.fsf@abhavana.free.fr>
Message-ID: <4A689533.4060400@opencsw.org>

On 07/23/2009 12:48 PM, Peter FELECAN wrote:
> John Ellson<ellson at opencsw.org>  writes:
>
>    
>> Now that I can build graphviz the OpenCSW way,  I'm planning to
>> use local resources to build nightly development snapshot pkgs
>> for solaris users, and distribute them like we do with rpms
>> and debs from our http://www.graphviz.org/ site.
>>      
> are you a member of the graphiz project?

Yes.

> I read your bio on the
> maintainer's page of our site and there wasn't a lot to read...
> think: communication.
>
>    



From phil at bolthole.com  Thu Jul 23 19:01:39 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 10:01:39 -0700
Subject: [csw-maintainers] howto create catalog files ?
In-Reply-To: <4A689533.4060400@opencsw.org>
References: <4A689355.1040405@opencsw.org> <x6d47rqraw.fsf@abhavana.free.fr>
	<4A689533.4060400@opencsw.org>
Message-ID: <20090723170139.GA73684@bolthole.com>

On Thu, Jul 23, 2009 at 12:52:03PM -0400, John Ellson wrote:
> On 07/23/2009 12:48 PM, Peter FELECAN wrote:
>...
>> I read your bio on the
>> maintainer's page of our site and there wasn't a lot to read...
>> think: communication.

A reminder,to John and others:
you can update bender:~/public_html/bio.html (create dir if needed)
to say more about you, and it will then show up in
www.opencsw.org/maintainers/yourname



From maciej at opencsw.org  Thu Jul 23 19:25:25 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 18:25:25 +0100
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723162445.GK53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<1248364654-sup-4834@ntdws12.chass.utoronto.ca>
	<x6ljmfqsv7.fsf@abhavana.free.fr>
	<20090723162445.GK53578@bolthole.com>
Message-ID: <ce97543b0907231025g508f04eav7c9eb0cb5a69af74@mail.gmail.com>

On Thu, Jul 23, 2009 at 5:24 PM, Philip Brown<phil at bolthole.com> wrote:
> And I will repeat: anyone who wishes to do actual work on this stuff,
> should let me or Ihsan know, and we will be happy to bore you to death with
> more details off-list.

I understand the problem is of a practical nature, not of a principle.
Do you have a rough estimate how much work it would require to get
into a state in which it could live in the repository? Would it be
hard to isolate the dubious bits and source-control the rest?

Maciej


From william at wbonnet.net  Thu Jul 23 22:00:02 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 23 Jul 2009 22:00:02 +0200
Subject: [csw-maintainers] Firefox 3.0.12 is in testing
Message-ID: <4A68C142.6040306@wbonnet.net>

Hi

I have updated the firefox package in testing. It works on Solaris 10, 
and will certainly not work on Solaris 8.

Thanks for the feedbacks

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 maciej at opencsw.org  Thu Jul 23 22:39:33 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 21:39:33 +0100
Subject: [csw-maintainers] Firefox 3.0.12 is in testing
In-Reply-To: <4A68C142.6040306@wbonnet.net>
References: <4A68C142.6040306@wbonnet.net>
Message-ID: <ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>

On Thu, Jul 23, 2009 at 9:00 PM, William Bonnet<william at wbonnet.net> wrote:
> Hi
>
> I have updated the firefox package in testing. It works on Solaris 10, and
> will certainly not work on Solaris 8.
>
> Thanks for the feedbacks

Yey for Firefox 3!

I see a sparc package. Are you going to send out a x86 package as well?

Maciej


From maciej at opencsw.org  Thu Jul 23 22:45:35 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 21:45:35 +0100
Subject: [csw-maintainers] Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
Message-ID: <ce97543b0907231345u2c9962cfj8891f865929bcbaa@mail.gmail.com>

On Thu, Jul 23, 2009 at 9:39 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> I see a sparc package. Are you going to send out a x86 package as well?

Ah, there was an interleave, the x86 version is there as well.
Somehow, pkgutil -t http://mirror.opencsw.org/opencsw/testing -i
firefox doesn't pick it up. Are other people able to install
firefox-3.0.12 in this way?

Maciej


From william at wbonnet.net  Thu Jul 23 22:45:49 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 23 Jul 2009 22:45:49 +0200
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
Message-ID: <4A68CBFD.3000408@wbonnet.net>

Hi Maciej
> Yey for Firefox 3!
>
> I see a sparc package. Are you going to send out a x86 package as well?
>   
There is also the i386 package :)

http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-sparc.pkg.gz
http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-i386.pkg.gz

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 Jul 23 23:23:18 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:23:18 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
Message-ID: <4A68D4C6.3040200@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Hacienda
Hacienda (Matches) Blizinski wrote:
> Hello OpenCSW web site maintainers,
> 
> I've been looking at www.opencsw.org website, and was wondering - if I
> wanted to suggest a change for it, or work on the website, how would I
> do that? I was looking for source code, but I didn't find it. Is the
> source code of www.opencsw.org checked into subversion? If not, what
> do you think of submitting the code there? It would make it easier to
> suggest a change, or submit a patch, and generally make things more
> open.
> 
> Maciej
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

What I did when I had a change was to send an email to Phil with what I wanted
to change. along with a link to the new page from my homedir on www.opencsw.org
http://www.opencsw.org/~mwatters/need_maintainer.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

iEYEARECAAYFAkpo1MYACgkQLrhmsXMSLxfNJgCgqQeJqMlMW6t58e/Q435UieEU
8tQAoKG7gDgDTBntZ2eusA8eLEFjtvCn
=k4sA
-----END PGP SIGNATURE-----


From phil at bolthole.com  Thu Jul 23 23:28:11 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 14:28:11 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68D4C6.3040200@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<4A68D4C6.3040200@opencsw.org>
Message-ID: <20090723212810.GA8541@bolthole.com>

On Thu, Jul 23, 2009 at 04:23:18PM -0500, Mike Watters wrote:
> ...
> What I did when I had a change was to send an email to Phil with what I wanted
> to change. along with a link to the new page from my homedir on www.opencsw.org
> http://www.opencsw.org/~mwatters/need_maintainer.php

and fyi, his script was then moved to the top-level web area, as
http://www.opencsw.org/need_maintainer.php

and he "owns" the script.



From mwatters at opencsw.org  Thu Jul 23 23:29:13 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:29:13 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723160307.GF53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
Message-ID: <4A68D629.9040802@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Phil
Philip Brown wrote:
> On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
>>> absolutely not "public".
>> Why not?
> 
> security issues.
> 

You are correct, as the code lies currently, there would be severe security
issues in publishing the code.

however, with a few minor modifications, it would be quite possible.

What would need to happen is to convert the php from hard coding all values,
php templates.  you can publish the code, but not the back end template (file
containing all values)



- --

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

iEYEARECAAYFAkpo1igACgkQLrhmsXMSLxcKPgCg0Poq0LMzufZXxSI2FhsCHY58
mp4AniwzTfyYjrcwaON44RvGsYOk82up
=f18E
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Jul 23 23:48:14 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:48:14 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68D629.9040802@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org>
Message-ID: <4A68DA9E.3030405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Ben,

Before anyone starts going down the road on the
existing http://www.opencsw.org/ code, have we had any
progress on http://www-mockup.opencsw.org/

- --

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

iEYEARECAAYFAkpo2p4ACgkQLrhmsXMSLxfvRgCfc96Qr5tMVFX/eq+ntN/dcZQO
DRwAoKlvbBSNRC2XmiRYU+49rQ++QWus
=7X9J
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Thu Jul 23 23:49:52 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 16:49:52 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68DA9E.3030405@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<4A68DA9E.3030405@opencsw.org>
Message-ID: <4A68DB00.7040409@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


> Hi Ben,
> 
> Before anyone starts going down the road on the
> existing http://www.opencsw.org/ code, have we had any
> progress on http://www-mockup.opencsw.org/
> 
sorry Ben, should be directed William I believe.

- --

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

iEYEARECAAYFAkpo2wAACgkQLrhmsXMSLxckhwCgteitjK3l/mjXqDrE8f5y2Dq0
tLIAoK9o/ILNThsKRRlvPe6qe22NATWw
=Ofp5
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Jul 24 00:01:49 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 15:01:49 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A68D629.9040802@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org>
Message-ID: <20090723220149.GB8541@bolthole.com>

On Thu, Jul 23, 2009 at 04:29:13PM -0500, Mike Watters wrote:
> 
> Hi Phil
> Philip Brown wrote:
> > On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
> >>> absolutely not "public".
> >> Why not?
> > 
> > security issues.
> > 
> 
> You are correct, as the code lies currently, there would be severe security
> issues in publishing the code.
> 
> however, with a few minor modifications, it would be quite possible.

And when changes are made in the future, more holes likely open up.
That is the way of source code.

Better for security to keep it local-visibility-only, if it doesnt need to
be fully public.

It doesnt need to be fully public.



From phil at bolthole.com  Fri Jul 24 01:54:02 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 16:54:02 -0700
Subject: [csw-maintainers] Please resubmit recent packages....
Message-ID: <20090723235402.GE8541@bolthole.com>

Arg....
It would seem there was a bug in the new stuff James and I worked out, to
"help coordination"

If you have recently had your packages released, please put them in newpkgs
again and let me know :-(

sorry for the inconvenience.


From phil at bolthole.com  Fri Jul 24 02:04:07 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 17:04:07 -0700
Subject: [csw-maintainers] Please resubmit recent packages....
In-Reply-To: <20090723235402.GE8541@bolthole.com>
References: <20090723235402.GE8541@bolthole.com>
Message-ID: <20090724000407.GF8541@bolthole.com>

On Thu, Jul 23, 2009 at 04:54:02PM -0700, Philip Brown wrote:
> Arg....
> It would seem there was a bug in the new stuff James and I worked out, to
> "help coordination"
> 
> If you have recently had your packages released, please put them in newpkgs
> again and let me know :-(
> 
> sorry for the inconvenience.

to be clear what happened:  they got "registered" in our database, but did
not make it to our actual mirror tree filesystem. and got deleted :-(
Older versions still remain in the tree.


From mwatters at opencsw.org  Fri Jul 24 04:26:17 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 23 Jul 2009 21:26:17 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723220149.GB8541@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com>
Message-ID: <4A691BC9.2050209@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> On Thu, Jul 23, 2009 at 04:29:13PM -0500, Mike Watters wrote:
>> Hi Phil
>> Philip Brown wrote:
>>> On Thu, Jul 23, 2009 at 05:56:49PM +0200, Trygve Laugst?l wrote:
>>>>> absolutely not "public".
>>>> Why not?
>>> security issues.
>>>
>> You are correct, as the code lies currently, there would be severe security
>> issues in publishing the code.
>>
>> however, with a few minor modifications, it would be quite possible.
> 
> And when changes are made in the future, more holes likely open up.
> That is the way of source code.
> 
> Better for security to keep it local-visibility-only, if it doesnt need to
> be fully public.
> 
> It doesnt need to be fully public.

what about a compromise and have either a sourceforge or self-hosted "members
only" subversion,git,cvs,bzr,etc... (pick your favorite version control
software) for the source code.


- --
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

iEYEARECAAYFAkppG8kACgkQLrhmsXMSLxe1OQCfa9vqFwXptAzOb+q9yDqXgLc3
TKwAoMcUsbEVT9ItjdGUbqt2ItVbon0y
=CDdd
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Jul 24 06:17:22 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 21:17:22 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A691BC9.2050209@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
Message-ID: <20090724041722.GA50043@bolthole.com>

On Thu, Jul 23, 2009 at 09:26:17PM -0500, Mike Watters wrote:
> 
> what about a compromise and have either a sourceforge or self-hosted "members
> only" subversion,git,cvs,bzr,etc... (pick your favorite version control
> software) for the source code.

that is *exactly* how we do it. :-)

People who write code for the website.. get to "pick their favourite
version control software" for their source code :-)

and sometimes that is SCCS. and sometimes that is cvs. and sometimes that
is "nothing".

Because sometimes, you really  JUST DONT NEED version control.
for a 10 line script, for example.
you just need backups.
which we have.



From phil at bolthole.com  Fri Jul 24 06:27:20 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 23 Jul 2009 21:27:20 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090724041722.GA50043@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
Message-ID: <20090724042720.GB50043@bolthole.com>

On Thu, Jul 23, 2009 at 09:17:22PM -0700, Philip Brown wrote:
> People who write code for the website.. get to "pick their favourite
> version control software" for their source code :-)
> 
> and sometimes that is SCCS. and sometimes that is cvs. and sometimes that
> is "nothing".
> 
> Because sometimes, you really  JUST DONT NEED version control.
> for a 10 line script, for example.
> you just need backups.
> which we have.

To fully connect the dots....

having a big source code repository for packaging type stuff is useful,
becuase it puts things "all in one place", that otherwise, would not have a
home. Otherwise, you'd have some in one persons' home directory, and some
in another person's home directory.

In contrast, for  web work on the www.opencsw.org main webserver,
All the "code" is already "all in one place": in the web server filesystem.

If you are one of the rare people who has chosen to actually do work on the
web coding, then you already know exactly where to look. and you can
already see everyone else's webcode too. its all quite "open" and organized
already.


From maciej at opencsw.org  Fri Jul 24 10:23:48 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 09:23:48 +0100
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090724042720.GB50043@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com> <4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com> <4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
Message-ID: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>

On Fri, Jul 24, 2009 at 5:27 AM, Philip Brown<phil at bolthole.com> wrote:
> having a big source code repository for packaging type stuff is useful,
> becuase it puts things "all in one place", that otherwise, would not have a
> home. Otherwise, you'd have some in one persons' home directory, and some
> in another person's home directory.
>
> In contrast, for ?web work on the www.opencsw.org main webserver,
> All the "code" is already "all in one place": in the web server filesystem.
>
> If you are one of the rare people who has chosen to actually do work on the
> web coding, then you already know exactly where to look. and you can
> already see everyone else's webcode too. its all quite "open" and organized
> already.

Do other people agree with the above? (That it's the web site source
code is open and organized already.)

Maciej


From william at wbonnet.net  Fri Jul 24 10:36:15 2009
From: william at wbonnet.net (William Bonnet)
Date: Fri, 24 Jul 2009 10:36:15 +0200
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907240128ic5b1c73q230ecff15d80d8b9@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>	<4A68CBFD.3000408@wbonnet.net>	<ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>
	<ce97543b0907240128ic5b1c73q230ecff15d80d8b9@mail.gmail.com>
Message-ID: <4A69727F.4080003@wbonnet.net>

Hi

> After I got it installed - it works! I'll keep using it and will come
> back with any comments. Thanks, William!
>   
You're welcome :)

Is someone using it under Solaris 9 ? Can i have some feedbacks please ?

cheers
W.


From pfelecan at opencsw.org  Fri Jul 24 10:47:57 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 24 Jul 2009 10:47:57 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	24 Jul 2009 09\:23\:48 +0100")
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>
	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>
	<20090723155311.GD53578@bolthole.com> <4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com> <4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
Message-ID: <x6ws5ypiv6.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 24, 2009 at 5:27 AM, Philip Brown<phil at bolthole.com> wrote:
>> having a big source code repository for packaging type stuff is useful,
>> becuase it puts things "all in one place", that otherwise, would not have a
>> home. Otherwise, you'd have some in one persons' home directory, and some
>> in another person's home directory.
>>
>> In contrast, for ?web work on the www.opencsw.org main webserver,
>> All the "code" is already "all in one place": in the web server filesystem.
>>
>> If you are one of the rare people who has chosen to actually do work on the
>> web coding, then you already know exactly where to look. and you can
>> already see everyone else's webcode too. its all quite "open" and organized
>> already.
>
> Do other people agree with the above? (That it's the web site source
> code is open and organized already.)

- it's open in the sense that if you have access to the hosting system
  you can read the source

- its organization is intrinsic

- the code is not part of a project manageable by a team, the way that
  the projects that we package are

- there is no documentation for this or other infrastructure of the
  OpenCSW project

Is there opportunity to change this situation? I think that an answer
would give also a measure of the OpenCSW politics.
-- 
Peter


From trygvis at opencsw.org  Fri Jul 24 11:08:04 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Fri, 24 Jul 2009 11:08:04 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org>	<20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
Message-ID: <4A6979F4.5050904@opencsw.org>

Maciej (Matchek) Blizinski wrote:
> On Fri, Jul 24, 2009 at 5:27 AM, Philip Brown<phil at bolthole.com> wrote:
>> having a big source code repository for packaging type stuff is useful,
>> becuase it puts things "all in one place", that otherwise, would not have a
>> home. Otherwise, you'd have some in one persons' home directory, and some
>> in another person's home directory.
>>
>> In contrast, for  web work on the www.opencsw.org main webserver,
>> All the "code" is already "all in one place": in the web server filesystem.
>>
>> If you are one of the rare people who has chosen to actually do work on the
>> web coding, then you already know exactly where to look. and you can
>> already see everyone else's webcode too. its all quite "open" and organized
>> already.
> 
> Do other people agree with the above? (That it's the web site source
> code is open and organized already.)

No, I don't.

Calling the code open and available when it is on a remote file system 
on a *live* web server is not right. If I want to work on source code, I 
need to copy the code locally, test it and probably show it to the rest 
of the group if it contain significant changes. That is not possible as 
it is.

If the code can't be shown to the world without being a security risk, 
and Phil doesn't believe that we can write secure code, I'd like to 
replace the application with something that we all can work on and extend.

--
Trygve


From maciej at opencsw.org  Fri Jul 24 11:45:34 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 10:45:34 +0100
Subject: [csw-maintainers] Subversion on bender
Message-ID: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>

Hello bender admins,

Can you install Subversion client on bender?

Thanks,
Maciej


From maciej at opencsw.org  Fri Jul 24 11:52:52 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 10:52:52 +0100
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
Message-ID: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>

On Fri, Jul 24, 2009 at 10:45 AM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> Can you install Subversion client on bender?

....and pysqlite2. Please. :-)

Maciej


From ihsan at dogan.ch  Fri Jul 24 16:36:34 2009
From: ihsan at dogan.ch (Ihsan Dogan)
Date: Fri, 24 Jul 2009 16:36:34 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
Message-ID: <4A69C6F2.2070003@dogan.ch>

Maciej (Matchek) Blizinski wrote:

>> Can you install Subversion client on bender?
> 
> ....and pysqlite2. Please. :-)

Why do you need Subversion and pysqlite2 on bender?




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bwalton at opencsw.org  Fri Jul 24 17:07:12 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 11:07:12 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <4A69C6F2.2070003@dogan.ch>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
Message-ID: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>

Excerpts from Ihsan Dogan's message of Fri Jul 24 10:36:34 -0400 2009:

> Why do you need Subversion and pysqlite2 on bender?

<quoting Phil>
People who write code for the website.. get to "pick their favourite
version control software" for their source code :-)
</quoting Phil>

Maciej has written some nice little utilities for displaying a matrix
of installed packages on each machine.  The utilities rely on sqlite
presently.  I guess he prefers svn for his code.

[Every person for themselves version control will eventually lead to
madness...]

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/dfe6aa82/attachment-0002.asc>

From mwatters at opencsw.org  Fri Jul 24 17:45:29 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 10:45:29 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
Message-ID: <4A69D719.6080409@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben Walton wrote:

> [Every person for themselves version control will eventually lead to
> madness...]

I couldn't agree more with that statement.
I propose to standardize on one VCS.

I am partial to subversion (with our without a front end.  eg. trac)
but am open to any of them.

account setup can be setup with our current authentication method to bender.

- --

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

iEYEARECAAYFAkpp1xkACgkQLrhmsXMSLxcn5wCgiRkBg18hVFmGvHQIdM9h/Hy/
x9cAn3I7vSKzkaLszaNu2Kb/8/ZgiwS+
=nBFj
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Jul 24 17:57:01 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 08:57:01 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <4A69D719.6080409@opencsw.org>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
Message-ID: <20090724155701.GB91227@bolthole.com>

On Fri, Jul 24, 2009 at 10:45:29AM -0500, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Ben Walton wrote:
> 
> > [Every person for themselves version control will eventually lead to
> > madness...]
> 
> I couldn't agree more with that statement.
> I propose to standardize on one VCS.

"i propose" that only people who are actively doing web source code work,
get to vote on what the VCS for  web server work is.

and for the record, my "vote" is SCCS, with a secondary preference for rcs


From phil at bolthole.com  Fri Jul 24 18:03:43 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 09:03:43 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6979F4.5050904@opencsw.org>
References: <20090723155311.GD53578@bolthole.com>
	<4A688841.8020503@opencsw.org>
	<20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
Message-ID: <20090724160343.GC91227@bolthole.com>

On Fri, Jul 24, 2009 at 11:08:04AM +0200, Trygve Laugst??l wrote:
> If the code can't be shown to the world without being a security risk,  
> and Phil doesn't believe that we can write secure code, I'd like to  
> replace the application with something that we all can work on and 
> extend.

There is no "*THE* application".
Additionally pretty much any web application, reguardless of language, has
the potential for security holes. So changing the framework, does not
definitively eliminate the underlying problem here.

This shows a fundamental non-understanding of what is going on in our web
space (and security, for that matter).

Just as people who arent actually maintaining packages, should not get to
tell active maintainers how to package... I think that people who arent
actually DOING WEB WORK, should not be telling those who are, how to do it.

If people want to actually do work, then great! please speak up with what
area you want to work on. Then after that, talk about how you think the 
working environment could be improved.


From maciej at opencsw.org  Fri Jul 24 18:55:29 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 17:55:29 +0100
Subject: [csw-maintainers] Including directory paths in prototypes using
	PKGFILES_CSWfoo
Message-ID: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>

On Tue, Jul 21, 2009 at 3:28 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> You usually specify PKGFILES for all but one packages which gets all
> the rest not included in any package. And you can say
> ?PKGFILES_CSWcupsd = $(sbindir)/(cupsd|cupsfilter)
> and leave out PKGFILES for CSWcupsclient.

More on that matter... There seems to be a problem with including
directories. I'm getting warnings like the following:

## Building pkgmap from package prototype file.
## Processing pkginfo file.
WARNING: missing directory entry for </opt>
WARNING: missing directory entry for </opt/csw>
WARNING: missing directory entry for </opt/csw/lib>

My package's PKGFILES:

PKGFILES_CSWlibcups  = $(prefix)
PKGFILES_CSWlibcups += $(libdir)
PKGFILES_CSWlibcups += $(libdir)/libcups.*

The $(libdir) variable is /opt/csw/lib, after implicitly adding ^ and
$ it becomes ^/opt/csw/lib$, which should match the directory:

$ [[ -d work/install-isa-i386/opt/csw/lib ]] && echo yes
yes

How to convince GAR to include the directories in the prototype?

Maciej


From bwalton at opencsw.org  Fri Jul 24 19:07:34 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 13:07:34 -0400
Subject: [csw-maintainers] Including directory paths in prototypes using
	PKGFILES_CSWfoo
In-Reply-To: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>
References: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>
Message-ID: <1248455214-sup-2440@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 24 12:55:29 -0400 2009:
> How to convince GAR to include the directories in the prototype?

You don't want or need those entries.  They're stripped by GAR on
purpose.  The CSWcommon package includes them.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/e7d8538f/attachment-0002.asc>

From maciej at opencsw.org  Fri Jul 24 19:28:47 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 18:28:47 +0100
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
Message-ID: <ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>

Hi maintainers,

I've commited the change for cups. It migrates the package to mGAR v2:

- dynamic prototypes
- SMF support via cswclassutils
- Zones support
    * /etc/opt/csw instead of /opt/csw/etc
    * /var/opt/csw instead of /opt/csw/var

I'd like to ask the senior maintainers to take a look at the change.
It's relatively large. There are bits I'm unsure about, such as
"ENABLE_CHECK = 0".

http://sourceforge.net/apps/trac/gar/changeset/5642

Maciej


From bwalton at opencsw.org  Fri Jul 24 19:38:36 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 13:38:36 -0400
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
	<ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
Message-ID: <1248456800-sup-5884@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 24 13:28:47 -0400 2009:

> I'd like to ask the senior maintainers to take a look at the change.
> It's relatively large. There are bits I'm unsure about, such as
> "ENABLE_CHECK = 0".

A while back I made some updates to checkpkg such that it worked on a
set of packages and honoured lib splitting across them such as you're
doing here.  This allowed one to leave ENABLE_CHECK = 1 and still have
checkpkg pass.

After a few revisions of the patch, it was seemingly dropped.  I still
have my local copy of the script I believe if you want to use it to
validate your set of packages _after_ GAR has built them.  I'm not
sure if there are more checks in the current checkpkg though.

Having to disable the checks stinks since they do catch other valuable
things.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/3e8aadc6/attachment-0002.asc>

From bwalton at opencsw.org  Fri Jul 24 20:00:26 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 14:00:26 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724155701.GB91227@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
Message-ID: <1248458300-sup-2415@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 11:57:01 -0400 2009:
> "i propose" that only people who are actively doing web source code work,
> get to vote on what the VCS for  web server work is.

How are you defining 'actively'?  Mike has (as you mentioned) made
previous contributions.  He seems like a good candidate for a 'vote'
to me.  Maciej should also get a vote since he now has code in
'testing' ready for the web.  William as well.

I disagree with this premise for inclusion/exclusion though.  The web
site is a project resource (as opposed to a personal resource like
pkg-get).  It should be a project decision.  Just because someone
isn't personally contributing right this minute doesn't mean they
won't in the future.

Having the consensus of current project members now makes it easier
for more people to feel comfortable contributing at such a time as
they have something valuable to contribute.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/ecad414b/attachment-0002.asc>

From phil at bolthole.com  Fri Jul 24 20:24:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 11:24:21 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248458300-sup-2415@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
Message-ID: <20090724182421.GB16963@bolthole.com>

On Fri, Jul 24, 2009 at 02:00:26PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 11:57:01 -0400 2009:
> > "i propose" that only people who are actively doing web source code work,
> > get to vote on what the VCS for  web server work is.
> 
> How are you defining 'actively'?  Mike has (as you mentioned) made
> previous contributions.  He seems like a good candidate for a 'vote'
> to me.

i agree.  I was just proactively trying to avoid the situation where
everyone decides to put in their oar just because they see the word "vote"
from Mike.


>  Just because someone
> isn't personally contributing right this minute doesn't mean they
> won't in the future.

experience has shown the opposite.

Or do you go by the ideal that EVERYONE in the public at large, should get
"a vote" on ALL opencsw issues?
Because after all, "just because they arent personally contributing right
this minute, doesnt mean they won't in the future".

 To put it another way... it seems a common piece of human nature, that
most people would rather tell other people how to work, than do work
themselves. Those "most people" arent entitled to a vote in how the workers
actually gets things done.

everyone is entitled to an opinion, and to even express that opinion. but
not everyone is entitled to a *vote* on such matters. non-workers have no
right _telling_ workers (especialy volunteer workers) how they are going to 
work.



From phil at bolthole.com  Fri Jul 24 20:53:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 11:53:17 -0700
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <1248456800-sup-5884@ntdws12.chass.utoronto.ca>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
	<ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
	<1248456800-sup-5884@ntdws12.chass.utoronto.ca>
Message-ID: <20090724185317.GA90084@bolthole.com>

On Fri, Jul 24, 2009 at 01:38:36PM -0400, Ben Walton wrote:
> ....
> After a few revisions of the patch, it was seemingly dropped.

My recollection was that you, Dago, and I, discussed a different way of
doing things; you agreed to do it... but then did not finish the method we
discussed.



From bwalton at opencsw.org  Fri Jul 24 21:10:46 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 15:10:46 -0400
Subject: [csw-maintainers] Updating cups to 1.3.11
In-Reply-To: <20090724185317.GA90084@bolthole.com>
References: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>
	<ce97543b0907210259ob3e03b0qffbcee8718283711@mail.gmail.com>
	<ce97543b0907241028o2dcfdf03i1d8d184bf355e28e@mail.gmail.com>
	<1248456800-sup-5884@ntdws12.chass.utoronto.ca>
	<20090724185317.GA90084@bolthole.com>
Message-ID: <1248462369-sup-5261@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 14:53:17 -0400 2009:
> > After a few revisions of the patch, it was seemingly dropped.
> 
> My recollection was that you, Dago, and I, discussed a different way of
> doing things; you agreed to do it... but then did not finish the method we
> discussed.

I knocked out a patch to checkpkg that allowed the calls from GAR to
say -i $pkg,$pkg2, or some such.  It then skipped validating those
dependencies and GAR presumably passed in the set of packages being
built currently.  That was rejected.

I submitted a second patch that let it evaluate multiple packages at
the same time, remembering missing libraries/dependencies and checking
again at the end after checking the rest of the packages.  I'd have to
dig it out, but afaik, there wasn't anything wrong with this patch, it
just wasn't applied.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/fb7b7c74/attachment-0002.asc>

From bwalton at opencsw.org  Fri Jul 24 21:18:40 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 15:18:40 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724182421.GB16963@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
Message-ID: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 14:24:21 -0400 2009:

> everyone is entitled to an opinion, and to even express that opinion. but
> not everyone is entitled to a *vote* on such matters. non-workers have no
> right _telling_ workers (especialy volunteer workers) how they are going to 
> work.

I certainly agree with these points.  I just figured that since effort
was made to form an association with bylaws, voting/non-voting
members[1], etc that this might be something that actually could
warrant a vote from voting members.  The website and how its handled
does have a large impact on the project as a whole, after all.
Maintenace of said domain (which presumably includes the website) is
one of the stated goals of the association.

I'm now requesting access to the web content also.  I'm going to
re-factor all php files such that database connection info is stored
only in a single place, thus facilitating the possibility of storing
this code in a non-local (read: not live) repository in the future[2]
with the db details excluded for security purposes.

William: What is the status of the mock-up?  Are there stumbling
blocks you're facing other than time constraints?

Thanks
-Ben

[1] I couldn't find a list of who gets a vote and who doesn't (or by
    what criteria the board was able to grant or deny this ability).
[2] I'd be happy to check into SCCS for the time being as long as some
    instructions are provided.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/b5f7f1d4/attachment-0002.asc>

From mwatters at opencsw.org  Fri Jul 24 21:32:56 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 14:32:56 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
Message-ID: <4A6A0C68.5040807@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Ben,
Ben Walton wrote:

> I'm now requesting access to the web content also.  I'm going to
> re-factor all php files such that database connection info is stored
> only in a single place, thus facilitating the possibility of storing
> this code in a non-local (read: not live) repository in the future[2]
> with the db details excluded for security purposes.
> 
> William: What is the status of the mock-up?  Are there stumbling
> blocks you're facing other than time constraints?

I am more then happy to help with the re factor or the mock-up if you need an
extra pair of hands.

- --

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

iEYEARECAAYFAkpqDGgACgkQLrhmsXMSLxfBLgCgy+kKGcARLnXXW4oDll/+cMZz
9WYAoMgcuxuS4x7zCCRrSaTYaxyqfLf1
=spuy
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Jul 24 21:46:30 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 12:46:30 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
Message-ID: <20090724194630.GA50326@bolthole.com>

On Fri, Jul 24, 2009 at 03:18:40PM -0400, Ben Walton wrote:
> ....
> I'm now requesting access to the web content also.  I'm going to
> re-factor all php files [...]

Wonderful!  welcome to the web team :-)


>  such that database connection info is stored
> only in a single place, thus facilitating the possibility of storing
> this code in a non-local (read: not live) repository in the future[2]
> with the db details excluded for security purposes.

the trouble with that, is that it affects performance.

> [2] I'd be happy to check into SCCS for the time being as long as some
>     instructions are provided.

sccs is about the same as any other local-store version control system.
(or others, even)
usual semmantics of that sort of thing are:

0. initialize a file
1. check out a file
2. check in a file
3. compare a file.


So, 


0. sccs create XYZ
1. sccs edit XYZ
2. sccs delget XYZ  (sends for "delta, then get", if you care ;-)
3. sccs diffs XYZ

there's a few other interesting features of it, but them's the basics.



From bwalton at opencsw.org  Fri Jul 24 21:59:15 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 15:59:15 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724194630.GA50326@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
Message-ID: <1248465309-sup-92@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:

> > such that database connection info is stored only in a single
> > place, thus facilitating the possibility of storing this code in a
> > non-local (read: not live) repository in the future[2] with the db
> > details excluded for security purposes.
> 
> the trouble with that, is that it affects performance.

No it doesn't.  The live copy on the web would simply be a current
checkout.

The pattern would become:

1. Develop, develop, develop.
2. Commit (which if svn[1] were used, is also a push to central).
3. <optional>Review of changes</optional>
4. Login to bender, proceed to web directory, update local copy.

This could all be done on the live copy to, but with a repo living
outside the code tree, easier testing is facilitated without affecting
the live stuff.

-Ben

[1] I have a personal distaste for svn but it's likely the logical
    choice given that all other project-wide type stuff lives in svn
    already.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/d17a1d03/attachment-0002.asc>

From mwatters at opencsw.org  Fri Jul 24 22:48:48 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 15:48:48 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248465309-sup-92@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
Message-ID: <4A6A1E30.6010405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:
> 
>>> such that database connection info is stored only in a single
>>> place, thus facilitating the possibility of storing this code in a
>>> non-local (read: not live) repository in the future[2] with the db
>>> details excluded for security purposes.
>> the trouble with that, is that it affects performance.
> 
> No it doesn't.  The live copy on the web would simply be a current
> checkout.
> 
> The pattern would become:
> 
> 1. Develop, develop, develop.
> 2. Commit (which if svn[1] were used, is also a push to central).
> 3. <optional>Review of changes</optional>
> 4. Login to bender, proceed to web directory, update local copy.
> 
> This could all be done on the live copy to, but with a repo living
> outside the code tree, easier testing is facilitated without affecting
> the live stuff.
> 
> -Ben
> 
> [1] I have a personal distaste for svn but it's likely the logical
>     choice given that all other project-wide type stuff lives in svn
>     already.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

I use Bazzar [1] when checking in/out of Fedora's PackageDB code repo.
not that I do it all that often anymore... but IMO better then svn ( don't take
much ) for those that are stuck in windows it runs nicely on that as well as
linux and solaris...  I can't vouch for mac though... never tried it on my
kid's mac book or the wife's mac book pro.

- --

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

iEYEARECAAYFAkpqHjAACgkQLrhmsXMSLxfbHwCgoKCOt22+MZN/b5F1E5P8May0
mIgAoLuNP+q46x2gs1wDpm/buXSPhHGg
=sZbu
-----END PGP SIGNATURE-----


From skayser at opencsw.org  Fri Jul 24 23:03:28 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 24 Jul 2009 23:03:28 +0200
Subject: [csw-maintainers] Updated libiconv in testing, testing appreciated
Message-ID: <4A6A21A0.3060006@opencsw.org>

Hi,

i have put an updated libiconv 1.13.1 in testing. The most notable
change since version 1.12 is native support for platform-dependent
encoding aliases like the Solaris C locale default of 646 ("for better
compatibility with the platform's own iconv_open function").

This should obsolete related hacks required in other packages. I have it
running on a couple of non-GUI test boxes without hitches, but as it is
a to-be-adopted and fundamental package, I would appreciate a bit more
testing/feedback/review by some of you guys.

I know that the package and software name differ (iconv vs. libiconv),
but that is not something i would want to change now (given that many
dependent packages).

Sebastian


From mwatters at opencsw.org  Fri Jul 24 23:32:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 24 Jul 2009 16:32:50 -0500
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <4A6A1E30.6010405@opencsw.org>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>	<20090724194630.GA50326@bolthole.com>	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<4A6A1E30.6010405@opencsw.org>
Message-ID: <4A6A2882.9030300@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mike Watters wrote:
> Ben Walton wrote:
>> Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:
> 
>>>> such that database connection info is stored only in a single
>>>> place, thus facilitating the possibility of storing this code in a
>>>> non-local (read: not live) repository in the future[2] with the db
>>>> details excluded for security purposes.
>>> the trouble with that, is that it affects performance.
>> No it doesn't.  The live copy on the web would simply be a current
>> checkout.
> 
>> The pattern would become:
> 
>> 1. Develop, develop, develop.
>> 2. Commit (which if svn[1] were used, is also a push to central).
>> 3. <optional>Review of changes</optional>
>> 4. Login to bender, proceed to web directory, update local copy.
> 
>> This could all be done on the live copy to, but with a repo living
>> outside the code tree, easier testing is facilitated without affecting
>> the live stuff.
> 
>> -Ben
> 
>> [1] I have a personal distaste for svn but it's likely the logical
>>     choice given that all other project-wide type stuff lives in svn
>>     already.
> 
> 
>> ------------------------------------------------------------------------
> 
>> _______________________________________________
>> maintainers mailing list
>> maintainers at lists.opencsw.org
>> https://lists.opencsw.org/mailman/listinfo/maintainers
> 
> I use Bazzar [1] when checking in/out of Fedora's PackageDB code repo.

forgot the link  ;-)

[1] http://bazaar-vcs.org/

- --

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

iEYEARECAAYFAkpqKIIACgkQLrhmsXMSLxdLZwCfbYI1KPAp2X+DNddtqIdOzxOu
RcMAn0kg68ulTQ8j7XjqnbGIVcFSugtD
=UM1m
-----END PGP SIGNATURE-----


From phil at bolthole.com  Fri Jul 24 23:54:52 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 14:54:52 -0700
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <4A6A21A0.3060006@opencsw.org>
References: <4A6A21A0.3060006@opencsw.org>
Message-ID: <20090724215452.GA73903@bolthole.com>

On Fri, Jul 24, 2009 at 11:03:28PM +0200, Sebastian Kayser wrote:
> 
> I know that the package and software name differ (iconv vs. libiconv),
> but that is not something i would want to change now (given that many
> dependent packages).

Umm... if you're going to be repackaging anyway... then why not repackage
now, and have it tested as such?



From phil at bolthole.com  Fri Jul 24 23:55:50 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 14:55:50 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248465309-sup-92@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
Message-ID: <20090724215550.GB73903@bolthole.com>

On Fri, Jul 24, 2009 at 03:59:15PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 15:46:30 -0400 2009:
> > the trouble with that, is that it affects performance.
> 
> No it doesn't.  The live copy on the web would simply be a current
> checkout.

sorry, misunderstanding. I thougth you meant "the PASSWORDS for database
access, would be in a separate, non-checked-in file".



From Joerg.Schilling at fokus.fraunhofer.de  Sat Jul 25 00:18:05 2009
From: Joerg.Schilling at fokus.fraunhofer.de (Joerg Schilling)
Date: Sat, 25 Jul 2009 00:18:05 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724194630.GA50326@bolthole.com>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>
	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org> <20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
Message-ID: <4a6a331d.Q3CFrX1KqEHdCOjZ%Joerg.Schilling@fokus.fraunhofer.de>

Philip Brown <phil at bolthole.com> wrote:

> On Fri, Jul 24, 2009 at 03:18:40PM -0400, Ben Walton wrote:

> > [2] I'd be happy to check into SCCS for the time being as long as some
> >     instructions are provided.

SCCS is OpenSource and I ported to all recent interesting platforms.
Given the fact that my version is 64 bit clean and supports 4 digit year 
numbers, it will even survive Y 2068.

> sccs is about the same as any other local-store version control system.
> (or others, even)

In future, there will be a network aware version ;-)


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 skayser at opencsw.org  Sat Jul 25 01:40:39 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sat, 25 Jul 2009 01:40:39 +0200
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <20090724215452.GA73903@bolthole.com>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com>
Message-ID: <4A6A4677.7000903@opencsw.org>

Philip Brown wrote:
> On Fri, Jul 24, 2009 at 11:03:28PM +0200, Sebastian Kayser wrote:
>> I know that the package and software name differ (iconv vs. libiconv),
>> but that is not something i would want to change now (given that many
>> dependent packages).
> 
> Umm... if you're going to be repackaging anyway... then why not repackage
> now, and have it tested as such?

Ok, updated packages in testing. iconv (CSWiconv) has now be changed to
be an empty package (besides README.CSW and license) that pulls in
libiconv and libiconv_devel as dependencies. libiconv (CSWlibiconv) is
the package that one would want and the naming is consistent.

Is that along the lines of what you had in mind?

Only minor annoyance i see right now is that if I issue an "pkgutil -u
iconv" (pkgutil 1.6) against testing on a current box, the two new
packages are pulled in and libiconv is installed first (before updating
iconv to the empty package). pkgadd complains about files being used by
another package and prompts the user. OTOH, as libiconv is a dependency
of iconv, this installation order makes sense to me. Could this be a
problem somehow?

pkg-get (4.1.2) updates iconv first, so no prompts to the user there
(didn't test with 4.2.1 though).

Sebastian


From phil at bolthole.com  Sat Jul 25 03:23:19 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 24 Jul 2009 18:23:19 -0700
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <4A6A4677.7000903@opencsw.org>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com>
	<4A6A4677.7000903@opencsw.org>
Message-ID: <20090725012319.GC73903@bolthole.com>

On Sat, Jul 25, 2009 at 01:40:39AM +0200, Sebastian Kayser wrote:
> Ok, updated packages in testing. iconv (CSWiconv) has now be changed to
> be an empty package (besides README.CSW and license) that pulls in
> libiconv and libiconv_devel as dependencies. libiconv (CSWlibiconv) is
> the package that one would want and the naming is consistent.
> 
> Is that along the lines of what you had in mind?

Um.... no :-}

when there are only two like this, with pre=exisiting dependancies, I think
it makes more sense to split into CSWiconv and a separate devel package
which depends on that one.
no empty package.



From bwalton at opencsw.org  Sat Jul 25 03:24:04 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 24 Jul 2009 21:24:04 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090724215550.GB73903@bolthole.com>
References: <ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>
	<4A69C6F2.2070003@dogan.ch>
	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
Message-ID: <1248484642-sup-7791@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 24 17:55:50 -0400 2009:
> sorry, misunderstanding. I thougth you meant "the PASSWORDS for database
> access, would be in a separate, non-checked-in file".

That's exactly what I meant.  I'll be creating a file called
includes/dbconfig.php.  This file will contain the host, database,
user and password info used for the connection to mysql.  All files
accessing the database will then include this file.  It's extra
overhead, yes, but completely negligible.

This new file won't be checked into the vcs and would need to be
created manually by anyone hacking on this stuff outside of the live
web tree.

Once this info is removed from each file and SQL queries are audited
for simple goofs (made easier by some braindead choices the php devs),
everything should be 'ready to go.'

On a different note, I'd like to propose a group called cswweb (or
similar) on bender with membership reflecting all those who've
expressed interest in modifying web content.  The htdocs (and maybe
cgi-bin?) files for the opencsw vhosts should then be made chgrp
cswweb, chmod g+w.  As it stands now, I'm not able to actually make
any of these changes.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090724/ea21a8f1/attachment-0002.asc>

From dam at opencsw.org  Sat Jul 25 17:48:33 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 17:48:33 +0200
Subject: [csw-maintainers] Graphviz on OpenCSW
Message-ID: <A0FAA66B-21C8-40AD-A77F-702B561AC5C9@opencsw.org>

Hi,

once again OpenCSW is mentioned for Graphviz-download for a cool
ZFS visualization tool:
   <http://www.c0t0d0s0.org/archives/5589-ZFSviz-0.0001.html>


Best regards

   -- Dago


From dam at opencsw.org  Sat Jul 25 18:07:12 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 18:07:12 +0200
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <4A678CBB.2030208@opencsw.org>
References: <4A677CAB.8050808@opencsw.org> <4A678182.9020005@opencsw.org>
	<4A678CBB.2030208@opencsw.org>
Message-ID: <69092344-647F-4D93-BBDC-9F7A226338CB@opencsw.org>

Hi John,

Am 23.07.2009 um 00:03 schrieb John Ellson:
> On 07/22/2009 05:15 PM, John Ellson wrote:
>> On 07/22/2009 04:55 PM, John Ellson wrote:
>>> Dago,
>>>
>>> Any chance you could rebuild cairo to include SVG support, please?
>>>
>>> I can take a shot at it if you like?
>>
>> So apparently the reason its not provided is because there is a  
>> circular depdendency with librsvg ?
>>
>> Why does a circular dependency matter?    Try to install either and  
>> you get both - where is the ambiguity?

Because neither pkg-get nor pkgutil will handle circular dependencies  
(yet).

> It builds fine with svg enabled.  There is no library dependency on  
> librsvg.  I think it may just be a testing dependency.

Ok then, if it is not a dependency I guess it is ok. What does the  
rest of
the X11 team say?
William?
Phil?


Best regards

   -- Dago


From phil at bolthole.com  Sat Jul 25 18:11:23 2009
From: phil at bolthole.com (Philip Brown)
Date: Sat, 25 Jul 2009 09:11:23 -0700
Subject: [csw-maintainers] libcairo svg surface
In-Reply-To: <69092344-647F-4D93-BBDC-9F7A226338CB@opencsw.org>
References: <4A677CAB.8050808@opencsw.org> <4A678182.9020005@opencsw.org>
	<4A678CBB.2030208@opencsw.org>
	<69092344-647F-4D93-BBDC-9F7A226338CB@opencsw.org>
Message-ID: <20090725161123.GA3902@bolthole.com>

On Sat, Jul 25, 2009 at 06:07:12PM +0200, Dagobert Michelsen wrote:
> Because neither pkg-get nor pkgutil will handle circular dependencies  
> (yet).
>
>> It builds fine with svg enabled.  There is no library dependency on  
>> librsvg.  I think it may just be a testing dependency.
>
> Ok then, if it is not a dependency I guess it is ok. What does the rest 
> of the X11 team say?
> William?
> Phil?

I said my piece a while back. remove the cycle, and submit it.
Half of that seems done now :-)



From dam at opencsw.org  Sat Jul 25 18:13:21 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 18:13:21 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A691BC9.2050209@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
Message-ID: <40569321-E5A1-46F1-A59E-71E4FE0272A8@opencsw.org>

Hi Mike,

Am 24.07.2009 um 04:26 schrieb Mike Watters:
> what about a compromise and have either a sourceforge or self-hosted  
> "members
> only" subversion,git,cvs,bzr,etc... (pick your favorite version  
> control
> software) for the source code.

For the record: We already have a SourceForge-project for
OpenCSW-specific scripts. You can browse at
   <http://opencsw.svn.sourceforge.net/viewvc/opencsw/>

Best regards

   -- Dago


From dam at opencsw.org  Sat Jul 25 18:18:26 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sat, 25 Jul 2009 18:18:26 +0200
Subject: [csw-maintainers] Including directory paths in prototypes using
	PKGFILES_CSWfoo
In-Reply-To: <1248455214-sup-2440@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240955s35e7118ci18d32a9b700a653d@mail.gmail.com>
	<1248455214-sup-2440@ntdws12.chass.utoronto.ca>
Message-ID: <39DFBD0A-74FB-47F0-8F24-DE805AA1F035@opencsw.org>

Hi Maciej,

Am 24.07.2009 um 19:07 schrieb Ben Walton:
> Excerpts from Maciej (Matchek) Blizinski's message of Fri Jul 24  
> 12:55:29 -0400 2009:
>> How to convince GAR to include the directories in the prototype?
>
> You don't want or need those entries.  They're stripped by GAR on
> purpose.  The CSWcommon package includes them.

Yes, exactly here:
   <http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/gar.pkg.mk#L305 
 >
The excluded pathes are in
   mgar/gar/v2/etc/commondirs-<arch>
They can be updates if CSWcommon gets updated with
   mgar/gar/v2/bin/update-commondirs
(not nice scripts, but can be hacked by me if necessary)


Best regards

   -- Dago


From trygvis at opencsw.org  Sat Jul 25 20:25:02 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sat, 25 Jul 2009 20:25:02 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090724160343.GC91227@bolthole.com>
References: <20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com>	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
Message-ID: <4A6B4DFE.1050307@opencsw.org>

Philip Brown wrote:
> On Fri, Jul 24, 2009 at 11:08:04AM +0200, Trygve Laugst?l wrote:
>> If the code can't be shown to the world without being a security risk,  
>> and Phil doesn't believe that we can write secure code, I'd like to  
>> replace the application with something that we all can work on and 
>> extend.
> 
> There is no "*THE* application".
> Additionally pretty much any web application, reguardless of language, has
> the potential for security holes. So changing the framework, does not
> definitively eliminate the underlying problem here.
> 
> This shows a fundamental non-understanding of what is going on in our web
> space (and security, for that matter).

It's impossible for us to know what's the current issue with the current 
application when your entire reply is "security".

I don't see how it should be hard to write a secure application to show 
the information that we currently display on the site.

> Just as people who arent actually maintaining packages, should not get to
> tell active maintainers how to package... I think that people who arent
> actually DOING WEB WORK, should not be telling those who are, how to do it.

It's a good thing that I do web development for a living then so I have 
the right to speak.

> If people want to actually do work, then great! please speak up with what
> area you want to work on. Then after that, talk about how you think the 
> working environment could be improved.

I'd like to improve the package info page, the first thing would be to 
remove the "doesn't work yet" comments etc.

--
Trygve


From ellson at opencsw.org  Sat Jul 25 21:38:03 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 15:38:03 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
Message-ID: <4A6B5F1B.6000303@opencsw.org>

I can build libcairo on build8s, but on build8x I'm getting:

     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this 
kernel with the arch 'i386'.  Stop.
     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
     make: *** [merge-isa-amd64] Error 2


Sould I be using a different build host?

John



From phil at bolthole.com  Sat Jul 25 22:30:35 2009
From: phil at bolthole.com (Philip Brown)
Date: Sat, 25 Jul 2009 13:30:35 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6B4DFE.1050307@opencsw.org>
References: <20090723160307.GF53578@bolthole.com>
	<4A68D629.9040802@opencsw.org> <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
Message-ID: <20090725203035.GA24170@bolthole.com>

On Sat, Jul 25, 2009 at 08:25:02PM +0200, Trygve Laugst?l wrote:
>> If people want to actually do work, then great! please speak up with what
>> area you want to work on. Then after that, talk about how you think the 
>> working environment could be improved.
>
> I'd like to improve the package info page, the first thing would be to  
> remove the "doesn't work yet" comments etc.

great. Please code something better. 
You konw where the code lives now. so please copy it to your own
~/public_html page, and tweak with it to your heart's content. if you get
it working better, wonderful! you can take over maintainnce of the official
page.



From ellson at opencsw.org  Sun Jul 26 01:07:58 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:07:58 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B5F1B.6000303@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org>
Message-ID: <4A6B904E.5050103@opencsw.org>

On 07/25/2009 03:38 PM, John Ellson wrote:
> I can build libcairo on build8s, but on build8x I'm getting:
>
>     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this 
> kernel with the arch 'i386'.  Stop.
>     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>     make: *** [merge-isa-amd64] Error 2
>
>
> Sould I be using a different build host?
>

OK, I see, its from BUILD64=1, and I'm supposed to do:

	build8x $ gmake merge   (builds all 32 bit ISAs and merges them)
	build10x $ gmake merge   (builds the remaining ISAs and merges them)
	 build8x $ gmake package (makes package, so it is installable on Solaris 8)

Painful!   And it doesn't work either due to this other previously reported problem with libm:


	One more thing: What you are trying to do here

	>/  # Solaris 10 on build10x has libm.so which points to libm.so.2.//libm.so.2
/	>/  # is however not available on Solaris 8. So we have to explicitly use
/	>/  # libm.so.1 during linking and get rid of the -lm reference. Otherwise
/	>/  # the package check on build8x when running "gmake package" fails.
/
	is also a flaw of checkpkg: When a package is build on Solaris 8 and Solaris 10
	the complete package may fail when checkpkg'ed on Solaris 8 with the reason above.
	You were able to link to a specific version, but this may not always be possible.
	I have no idea on how this could be fixed in checkpkg apart from not making
	mixed packages or checking only on Solaris 10.

So, Dago, how were you building libcairo before?

John




From ellson at opencsw.org  Sun Jul 26 01:12:13 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:12:13 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B904E.5050103@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B904E.5050103@opencsw.org>
Message-ID: <4A6B914D.3040903@opencsw.org>

On 07/25/2009 07:07 PM, John Ellson wrote:
> On 07/25/2009 03:38 PM, John Ellson wrote:
>> I can build libcairo on build8s, but on build8x I'm getting:
>>
>>     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this 
>> kernel with the arch 'i386'.  Stop.
>>     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>>     make: *** [merge-isa-amd64] Error 2
>>
>>
>> Sould I be using a different build host?
>>
>
> OK, I see, its from BUILD64=1, and I'm supposed to do:
>
>     build8x $ gmake merge   (builds all 32 bit ISAs and merges them)
>     build10x $ gmake merge   (builds the remaining ISAs and merges them)
>      build8x $ gmake package (makes package, so it is installable on 
> Solaris 8)
>
> Painful!   And it doesn't work either due to this other previously 
> reported problem with libm:
>
>
>     One more thing: What you are trying to do here
>
> >/  # Solaris 10 on build10x has libm.so which points to 
> libm.so.2.//libm.so.2
> / >/  # is however not available on Solaris 8. So we have to 
> explicitly use
> / >/  # libm.so.1 during linking and get rid of the -lm reference. 
> Otherwise
> / >/  # the package check on build8x when running "gmake package" fails.
> /
>     is also a flaw of checkpkg: When a package is build on Solaris 8 
> and Solaris 10
>     the complete package may fail when checkpkg'ed on Solaris 8 with 
> the reason above.
>     You were able to link to a specific version, but this may not 
> always be possible.
>     I have no idea on how this could be fixed in checkpkg apart from 
> not making
>     mixed packages or checking only on Solaris 10.
>
> So, Dago, how were you building libcairo before?
>


gmake ENABLE_CHECK=0 package

seems to work.    Doesn't feel like the right solution?


John



From mwatters at opencsw.org  Sun Jul 26 01:14:24 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Jul 2009 18:14:24 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <40569321-E5A1-46F1-A59E-71E4FE0272A8@opencsw.org>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>	<20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>
	<40569321-E5A1-46F1-A59E-71E4FE0272A8@opencsw.org>
Message-ID: <4A6B91D0.4070808@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Dago,
Dagobert Michelsen wrote:
> Hi Mike,
> 
> For the record: We already have a SourceForge-project for
> OpenCSW-specific scripts. You can browse at
>   <http://opencsw.svn.sourceforge.net/viewvc/opencsw/>

Yes, I know about that repository, but I believe it is public??
and didn't meet the criteria for "security"

This repo IMHO would be just fine as long as we get the files re factored to
use include files for sensitive configs.


- --
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/

iEYEARECAAYFAkprkc8ACgkQLrhmsXMSLxfRbgCeMhkF9jBsw8rp2Wjx3l0/I3VO
Q3gAmweFS0q45q/SmZMfcTBuXzCvlD4F
=/Z/c
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sun Jul 26 01:17:41 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Jul 2009 18:17:41 -0500
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6B4DFE.1050307@opencsw.org>
References: <20090723155311.GD53578@bolthole.com>	<4A688841.8020503@opencsw.org>	<20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>	<20090723220149.GB8541@bolthole.com>	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
Message-ID: <4A6B9295.20506@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Trygve,
Trygve Laugst?l wrote:

> I'd like to improve the package info page, the first thing would be to
> remove the "doesn't work yet" comments etc.
> 

I agree, however, I would recommend putting efforts toward the mock-up format
William has been working on. http://www-mockup.opencsw.org/


- --
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/

iEYEARECAAYFAkprkpQACgkQLrhmsXMSLxf0kACg05D0JyMm9sNJNP5eNSlg2ANf
kv8AoOunOLFCsk8vamgFz390+NRDjZmf
=g9o9
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Sun Jul 26 01:20:38 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sat, 25 Jul 2009 18:20:38 -0500
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B5F1B.6000303@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org>
Message-ID: <4A6B9346.5060405@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi John,
John Ellson wrote:
> I can build libcairo on build8s, but on build8x I'm getting:
> 
>     gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this
> kernel with the arch 'i386'.  Stop.
>     gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>     make: *** [merge-isa-amd64] Error 2
> 
> 

That error is from gcc
if you plan on using gcc to build 64bit binaries, you *Must* use gcc4

I am not positive, but I believe libcairo will build using Sun Studio.

- --
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/

iEYEARECAAYFAkprk0UACgkQLrhmsXMSLxfLnACgwBUiljv/5Ft7L8ubd/tZHVxJ
caUAoLwpeK0+jSqeQTJx9daCnzmFokfM
=CorV
-----END PGP SIGNATURE-----


From ellson at opencsw.org  Sun Jul 26 01:46:24 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:46:24 -0400
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B9346.5060405@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B9346.5060405@opencsw.org>
Message-ID: <4A6B9950.6000507@opencsw.org>

On 07/25/2009 07:20 PM, Mike Watters wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi John,
> John Ellson wrote:
>    
>> I can build libcairo on build8s, but on build8x I'm getting:
>>
>>      gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this
>> kernel with the arch 'i386'.  Stop.
>>      gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>>      make: *** [merge-isa-amd64] Error 2
>>
>>
>>      
> That error is from gcc
> if you plan on using gcc to build 64bit binaries, you *Must* use gcc4
>
> I am not positive, but I believe libcairo will build using Sun Studio.
>
>
>    

No, its from the Sun Studio builds on build8x exactly as Dago had it before.
All I'm doing is enabling the SVG backend and rebuilding.


The trick seems to be:

	build8x $ gmake clean
	build8x $ gmake build-isa-i386
	build10x $ gmake merge
	build8x $ gmake ENABLE_CHECK=0 package/
	/build8s $ gmake clean
	build8s $ gmake
	build8s $ gmake package


John


From ellson at opencsw.org  Sun Jul 26 01:47:49 2009
From: ellson at opencsw.org (John Ellson)
Date: Sat, 25 Jul 2009 19:47:49 -0400
Subject: [csw-maintainers] libcairo with SVG support now available in testing
Message-ID: <4A6B99A5.6060100@opencsw.org>

The only change in this version of libcairo is to enable SVG support.


John





From dam at opencsw.org  Sun Jul 26 09:30:49 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 09:30:49 +0200
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B5F1B.6000303@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org>
Message-ID: <FED7C188-310D-4DE0-994A-D0EA3EC9A238@opencsw.org>

Hi John,

Am 25.07.2009 um 21:38 schrieb John Ellson:
> I can build libcairo on build8s, but on build8x I'm getting:
>
>    gar/gar.conf.mk:352: *** The ISA 'amd64' can not be build on this  
> kernel with the arch 'i386'.  Stop.
>    gmake[1]: Leaving directory `/home/ellson/mgar/pkg/libcairo/trunk'
>    make: *** [merge-isa-amd64] Error 2
>
> Sould I be using a different build host?

This is normal. For x86 you do

build8x $ gmake merge (builds all 32 bit ISAs and merges them)
build10x $ gmake merge (builds the remaining ISAs and merges them)
build8x $ gmake package (makes package, so it is installable on  
Solaris 8)

For details see
   <http://sourceforge.net/apps/trac/gar/wiki/Building%20for%20multiple%20ISAs 
 >


Best regards

   -- Dago


From dam at opencsw.org  Sun Jul 26 09:32:43 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 09:32:43 +0200
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B904E.5050103@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B904E.5050103@opencsw.org>
Message-ID: <62019BB1-EA43-4B7B-B1EE-1B86429CA3D4@opencsw.org>

Hi John,

Am 26.07.2009 um 01:07 schrieb John Ellson:
> Painful!   And it doesn't work either due to this other previously  
> reported problem with libm:
>
> 	One more thing: What you are trying to do here
>
> 	>/  # Solaris 10 on build10x has libm.so which points to libm.so. 
> 2.//libm.so.2
> /	>/  # is however not available on Solaris 8. So we have to  
> explicitly use
> /	>/  # libm.so.1 during linking and get rid of the -lm reference.  
> Otherwise
> /	>/  # the package check on build8x when running "gmake package"  
> fails.
> /
> 	is also a flaw of checkpkg: When a package is build on Solaris 8  
> and Solaris 10
> 	the complete package may fail when checkpkg'ed on Solaris 8 with  
> the reason above.
> 	You were able to link to a specific version, but this may not  
> always be possible.
> 	I have no idea on how this could be fixed in checkpkg apart from  
> not making
> 	mixed packages or checking only on Solaris 10.
>
> So, Dago, how were you building libcairo before?

Basically you do
   ENABLE_CHECK=0 gmake repackage
once again to disable this check. checkpkg thinkgs the libm.so.2
should belong to the build host, but it is only for the Solaris 10
binaries, so it is ok here.


Best regards

   -- Dago


From dam at opencsw.org  Sun Jul 26 09:45:03 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 09:45:03 +0200
Subject: [csw-maintainers] libcairo failing to build on build8x
In-Reply-To: <4A6B9950.6000507@opencsw.org>
References: <4A6B5F1B.6000303@opencsw.org> <4A6B9346.5060405@opencsw.org>
	<4A6B9950.6000507@opencsw.org>
Message-ID: <C05A5D9D-46A0-4170-B4B6-B274820DFA3A@opencsw.org>

Hi John,

Am 26.07.2009 um 01:46 schrieb John Ellson:
>> That error is from gcc

No, it is from GAR and it is normal when you build 64 bit.

> No, its from the Sun Studio builds on build8x exactly as Dago had it  
> before.
> All I'm doing is enabling the SVG backend and rebuilding.
>
> The trick seems to be:
>
> 	build8x $ gmake clean
> 	build8x $ gmake build-isa-i386
> 	build10x $ gmake merge

You should merge the 32 bit build on Solaris 8 or you may
get errors on usage later. I clarified this on
   <http://sourceforge.net/apps/trac/gar/wiki/Building%20for%20multiple%20ISAs 
 >
That is
   build8x $ gmake clean
   build8x $ gmake merge
   build10x $ gmake merge
   build8x $ ENABLE_CHECK=0 gmake package

> 	/build8s $ gmake clean
> 	build8s $ gmake
> 	build8s $ gmake package


You could skip the 'gmake' and do 'gmake package' right away


Best regards

   -- Dago


From trygvis at opencsw.org  Sun Jul 26 11:34:57 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sun, 26 Jul 2009 11:34:57 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090725203035.GA24170@bolthole.com>
References: <20090723160307.GF53578@bolthole.com>	<4A68D629.9040802@opencsw.org>
	<20090723220149.GB8541@bolthole.com>	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
Message-ID: <4A6C2341.8060600@opencsw.org>

Philip Brown wrote:
> On Sat, Jul 25, 2009 at 08:25:02PM +0200, Trygve Laugst?l wrote:
>>> If people want to actually do work, then great! please speak up with what
>>> area you want to work on. Then after that, talk about how you think the 
>>> working environment could be improved.
>> I'd like to improve the package info page, the first thing would be to  
>> remove the "doesn't work yet" comments etc.
> 
> great. Please code something better. 
> You konw where the code lives now. so please copy it to your own
> ~/public_html page, and tweak with it to your heart's content. if you get
> it working better, wonderful! you can take over maintainnce of the official
> page.

Ok, I'll do that.

Can you explain what the security reason for not putting it an SCM is? 
The last reply I got was "security" but that's not enough. I can't think 
of any security risks involved with a read-only application and if I'm 
going to develop on it I'd like to know what to look out for.

--
Trygve


From skayser at opencsw.org  Sun Jul 26 14:28:23 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Sun, 26 Jul 2009 14:28:23 +0200 (CEST)
Subject: [csw-maintainers] Updated libiconv in testing,
 testing	appreciated
In-Reply-To: <20090725012319.GC73903@bolthole.com>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com> <4A6A4677.7000903@opencsw.org>
	<20090725012319.GC73903@bolthole.com>
Message-ID: <50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>

Philip Brown wrote:
> On Sat, Jul 25, 2009 at 01:40:39AM +0200, Sebastian Kayser wrote:
>> Ok, updated packages in testing. iconv (CSWiconv) has now be changed to
>> be an empty package (besides README.CSW and license) that pulls in
>> libiconv and libiconv_devel as dependencies. libiconv (CSWlibiconv) is
>> the package that one would want and the naming is consistent.
>>
>> Is that along the lines of what you had in mind?
>
> Um.... no :-}
>
> when there are only two like this, with pre=exisiting dependancies, I
> think
> it makes more sense to split into CSWiconv and a separate devel package
> which depends on that one.
> no empty package.

Before i commence take #3, let me try to understand:

Upstream name is libiconv, right now we have it as CSWiconv (pkgname) and
libiconv (swname). Would be nice to have CSWlibiconv and libiconv instead
to match the upstream name (see my previous attempt), but this introduces
the transitory, empty package CSWiconv for backwards compatibility and
implies long-term pkg dependency updating. Thus it is less hassle to just
leave the package name as CSWiconv and change its software name to iconv
(implies that libiconv as software name is dropped from the catalog)?

Sebastian



From mwatters at opencsw.org  Sun Jul 26 16:28:04 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Sun, 26 Jul 2009 09:28:04 -0500
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
References: <4A6A21A0.3060006@opencsw.org>	<20090724215452.GA73903@bolthole.com>
	<4A6A4677.7000903@opencsw.org>	<20090725012319.GC73903@bolthole.com>
	<50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
Message-ID: <4A6C67F4.30900@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Sebastian,
Sebastian Kayser wrote:

> Before i commence take #3, let me try to understand:
> 
> Upstream name is libiconv, right now we have it as CSWiconv (pkgname) and
> libiconv (swname). Would be nice to have CSWlibiconv and libiconv instead
> to match the upstream name (see my previous attempt), but this introduces
> the transitory, empty package CSWiconv for backwards compatibility and
> implies long-term pkg dependency updating. Thus it is less hassle to just
> leave the package name as CSWiconv and change its software name to iconv
> (implies that libiconv as software name is dropped from the catalog)?
> 

The best way (and AFAIK the formal way) to handle such a conversion is as follows:

1) create the package as CSWlibiconv
	1a) set CSWlibiconv incompatible with CSWiconv
2) rebuild CSWiconv as an empty meta-package
	2a) create a dependency on CSWlibiconv
3) push both packages through the normal test/newpkgs/current cycle
4) file a bug against all packages that depend on CSWiconv
   to rebuild against CSWlibiconv.


- --
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/

iEYEARECAAYFAkpsZ/QACgkQLrhmsXMSLxcdHwCeOo/32plYwXZp4wmQK/joDBh1
iKUAn0fgtRPI4Ypvl9VHFlsfmOkfo8ti
=bCRt
-----END PGP SIGNATURE-----


From rupert at opencsw.org  Sun Jul 26 16:55:00 2009
From: rupert at opencsw.org (rupert THURNER)
Date: Sun, 26 Jul 2009 16:55:00 +0200
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3 released!
In-Reply-To: <6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
Message-ID: <6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>

hi,

could it be possible that the type of providing the user homedir used
on the csw servers influence the test result of the mercurial package?
see the bottom of belows mail from mads.

i will try in the next weeks on a std filesystem as well with a std checkout.

rupert.

---------- Forwarded message ----------
From: Mads Kiilerich <mads at kiilerich.com>
Date: 2009/7/25
Subject: Re: Solaris package, was Re: Mercurial 1.3 released!
To: "rupert.thurner" <rupert.thurner at gmail.com>
Cc: mercurial at selenic.com, Trygve Laugst?l <trygvis at inamo.no>


rupert.thurner wrote, On 07/25/2009 11:52 AM:
>
> On Jul 5, 1:23 pm, Mads Kiilerich<mads at kiilerich.com> ?wrote:
>
>>
>> rupert.thurner wrote, On 07/04/2009 08:13 PM:
>>
>>
>>>
>>> we tried to package it for solaris,http://opencsw.org. some test
>>> cases fail. some failures are because the test suite did not use /opt/
>>> csw/bin/gdiff, but "diff" from solaris. how do we specify the diff
>>> tool used, or the path to it?
>>>
>>
>> The testsuite uses whatever diff tool is in its path. So you could just
>> set PATH correctly before running the test suite.
>>
>>
>>>
>>> with some other tests i have no idea where the failure is coming from.
>>> what would be the best option to track this down? seehttp://pastebin.com/m32e3576f
>>> for the complete test result.
>>>
>>
>> test-repair-strip shouldn't rely on the cat error message. And the
>> warning in test-gpg should probably just be ignored. Patches has been
>> posted to -devel.
>>
>> Just ignore these tests for now. ?You can temporarily remove their x bit
>> to avoid running them.
>>
>> That will make the failure list shorter and make it easier to spot:
>>
>>
>>>
>>> an excerpt:
>>>
>>
>>
>>>
>>> ERROR: test-transplant output changed
>>> --- Expected output
>>> +++ Test output
>>> @@ -145,20 +145,20 @@
>>> ? % test filter
>>> ? filtering
>>> ? applying 17ab29e464c6
>>> -17ab29e464c6 transplanted to e9ffc54ea104
>>> +17ab29e464c6 transplanted to 88e81fba5b32
>>> ? filtering
>>>
>>
>> This kind of errors looks worse. Do you consistently get exactly the
>> same errors?
>>
>
> i run the tests again with mercurial-1.3.1, see the result here:
> http://pastebin.com/m1a2f283f. especially
>

It seems like pastebin obscures the output slightly. Perhaps it would
be easier to read if it didn't use Python highlighting.

http://www.selenic.com/hg/index.cgi/rev/6e5e548452de (test-gpg:
suppress secure memory warning) should probably have been applied to
stable as well. Anyone?

Regarding test-repair-strip: One of the patches from
http://www.selenic.com/pipermail/mercurial-devel/2009-July/thread.html#13610
should be applied to stable.

You are still running in an environment without GNU diff. And
apparently also an unsupported version of cvs.

The rest of the problems seems to have something to do with something
not being like someone thinks they should be, perhaps forking or
caching or file systems.

Do you get the same failures when you run "make tests" manually from a
source checkout?


From phil at bolthole.com  Sun Jul 26 17:38:56 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 08:38:56 -0700
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3	released!
In-Reply-To: <6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
	<6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
Message-ID: <20090726153856.GA28096@bolthole.com>

On Sun, Jul 26, 2009 at 04:55:00PM +0200, rupert THURNER wrote:
> hi,
> 
> could it be possible that the type of providing the user homedir used
> on the csw servers influence the test result of the mercurial package?
> see the bottom of belows mail from mads.

I dont think it has much to do with the "homedirs". 
Seems like, according to the bug report, you need to force mercurial to use
"gdiff" instead of "diff".
Makes sense. Some software is like that.




From phil at bolthole.com  Sun Jul 26 17:43:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 08:43:45 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6C2341.8060600@opencsw.org>
References: <20090723220149.GB8541@bolthole.com> <4A691BC9.2050209@opencsw.org>
	<20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
	<4A6C2341.8060600@opencsw.org>
Message-ID: <20090726154345.GB28096@bolthole.com>

On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst??l wrote:
> Can you explain what the security reason for not putting it an SCM is?  

"a" source code control mechanism, I dont have much problem with.
putting it in a PUBLIC one, is where my concerns lay.

Beyond that, I dont see the need to force everyone to use one particular
type of scm, just for small web page "code".
If the assorted people working on opencsw web stuff agree to standardize on
one internal-only scheme, i'd go along with it. But frankly, most of the
web related stuff is small enough that the old cycle of

"make a save copy, tweak with the new stuff, publish it, forget about old
 stuff"

is perfectly adequate.


From phil at bolthole.com  Sun Jul 26 17:49:31 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 08:49:31 -0700
Subject: [csw-maintainers] Updated libiconv in testing,
	testing	appreciated
In-Reply-To: <50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
References: <4A6A21A0.3060006@opencsw.org>
	<20090724215452.GA73903@bolthole.com>
	<4A6A4677.7000903@opencsw.org>
	<20090725012319.GC73903@bolthole.com>
	<50303.84.173.63.233.1248611303.squirrel@ssl.skayser.de>
Message-ID: <20090726154931.GC28096@bolthole.com>

On Sun, Jul 26, 2009 at 02:28:23PM +0200, Sebastian Kayser wrote:
> 
> Upstream name is libiconv, right now we have it as CSWiconv (pkgname) and
> libiconv (swname). Would be nice to have CSWlibiconv and libiconv instead
> to match the upstream name (see my previous attempt), [...]
>  Thus it is less hassle to just
> leave the package name as CSWiconv and change its software name to iconv
> (implies that libiconv as software name is dropped from the catalog)?

true, that is the least hassle thing to do.
Howeer, I think that it is nice that the swname matches the upstream name
"libiconv". 

And going the other way, changing the PKG name, of such an oooold, and
crucial package, makes me shudder to think about it.
[by neccessity, it was one of the very first packages built and released, ever]
I'd rather we just leave well enough alone and not change either.


From phil at bolthole.com  Sun Jul 26 18:31:22 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 09:31:22 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248484642-sup-7791@ntdws12.chass.utoronto.ca>
References: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
Message-ID: <20090726163122.GE28096@bolthole.com>

On Fri, Jul 24, 2009 at 09:24:04PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Fri Jul 24 17:55:50 -0400 2009:
> > sorry, misunderstanding. I thougth you meant "the PASSWORDS for database
> > access, would be in a separate, non-checked-in file".
> 
> That's exactly what I meant.  I'll be creating a file called
> includes/dbconfig.php.  This file will contain the host, database,
> user and password info used for the connection to mysql.  All files
> accessing the database will then include this file.  It's extra
> overhead, yes, but completely negligible.

hrmmm...
well, in extreme cases(high loads), i've seen use of php includes,
dramatically slow down pages. so it's not always negligible.


> This new file won't be checked into the vcs and would need to be
> created manually by anyone hacking on this stuff outside of the live
> web tree.

there's also the concern of someone inevitably doing the equivalent of
"checkin *", and having the file "accidentally" get into the version
control system.


> On a different note, I'd like to propose a group called cswweb (or
> similar) on bender with membership reflecting all those who've
> expressed interest in modifying web content.  The htdocs (and maybe
> cgi-bin?) files for the opencsw vhosts should then be made chgrp
> cswweb, chmod g+w.  As it stands now, I'm not able to actually make
> any of these changes.

and in some ways, lack of thats a good thing ;-) that one person cant just
decide to drastically overhaul everything on a whim. 
(well, ok, people with root-type access 'could' do that. but we just
 dont do that sort of thing)

long term, i think having a web unixgroup makes sense.
How about you start with one page for now though. the "packages" page was
what you expressed interest in, if I recall?


From ja at opencsw.org  Sun Jul 26 20:09:01 2009
From: ja at opencsw.org (Juergen Arndt)
Date: Sun, 26 Jul 2009 20:09:01 +0200
Subject: [csw-maintainers] Cacti 0.8.7e in testing
Message-ID: <op.uxospam6jbmpvo@imac.lan>

Hi all,

I put cacti 0.8.7e into testing.

I'm not sure about the location of the files (especially the php stuff).  
Maybe someone could take a look?

Thank you,

Juergen

-- 
Juergen Arndt


From dam at opencsw.org  Sun Jul 26 20:16:29 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 20:16:29 +0200
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3	released!
In-Reply-To: <20090726153856.GA28096@bolthole.com>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
	<6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
	<20090726153856.GA28096@bolthole.com>
Message-ID: <D92A2BB1-12C2-49A3-9C69-FEE730E3EE21@opencsw.org>

Hi Rupert,

Am 26.07.2009 um 17:38 schrieb Philip Brown:
> On Sun, Jul 26, 2009 at 04:55:00PM +0200, rupert THURNER wrote:
>> could it be possible that the type of providing the user homedir used
>> on the csw servers influence the test result of the mercurial  
>> package?
>> see the bottom of belows mail from mads.
>
> I dont think it has much to do with the "homedirs".
> Seems like, according to the bug report, you need to force mercurial  
> to use
> "gdiff" instead of "diff".
> Makes sense. Some software is like that.

In GAR you can do this with

> include gar/category.mk
>
> # Make sure we are using gnu tools (esp. sed)
> PATH := /opt/csw/gnu:$(PATH)

There are quite a lot of examples for this kind of use. See

> build8s% grep csw/gnu pkg/*/trunk/Makefile


Best regards

   -- Dago


From dam at opencsw.org  Sun Jul 26 20:18:59 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 20:18:59 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <20090726163122.GE28096@bolthole.com>
References: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
	<20090726163122.GE28096@bolthole.com>
Message-ID: <3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>

Hi Phil,

Am 26.07.2009 um 18:31 schrieb Philip Brown:
>> This new file won't be checked into the vcs and would need to be
>> created manually by anyone hacking on this stuff outside of the live
>> web tree.
>
> there's also the concern of someone inevitably doing the equivalent of
> "checkin *", and having the file "accidentally" get into the version
> control system.

With Subversion you explicitly add files once. When you
commit you automatically commit only added files.
BTW, you could also restrict access to the DB to localhost.


Best regards

   -- Dago


From bwalton at opencsw.org  Sun Jul 26 20:40:57 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Jul 2009 14:40:57 -0400
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
References: <1248447803-sup-4618@ntdws12.chass.utoronto.ca>
	<4A69D719.6080409@opencsw.org>
	<20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
	<20090726163122.GE28096@bolthole.com>
	<3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
Message-ID: <1248633364-sup-220@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Jul 26 14:18:59 -0400 2009:
> With Subversion you explicitly add files once. When you
> commit you automatically commit only added files.

Also, you can explicitly add the file to an ignore list so that you'd
have to intentionally circumvent the vcs package to add/commit the
file.

> BTW, you could also restrict access to the DB to localhost.

...And a user that is limited to SELECT() (read only) would also go a
long way toward securing the access.  Even with access limited to
localhost, SQL injection with a read/write account is still a threat.
Let MySQL remove that threat (barring bugs in it's ACL enforcement)
for you.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090726/af340507/attachment-0002.asc>

From phil at bolthole.com  Sun Jul 26 20:54:05 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 11:54:05 -0700
Subject: [csw-maintainers] Fwd: Solaris package,
	was Re: Mercurial 1.3	released!
In-Reply-To: <D92A2BB1-12C2-49A3-9C69-FEE730E3EE21@opencsw.org>
References: <1246487895.21295.450.camel@calx>
	<55b02916-16ec-478b-a650-446994bef399@o6g2000yqj.googlegroups.com>
	<4A508D25.8040308@kiilerich.com>
	<9c79bcfc-dca6-41d0-bda9-b0b6fc058333@k6g2000yqn.googlegroups.com>
	<4A6ADF59.5000008@kiilerich.com>
	<6af4270907260748h5460aa33m5a68b95d063eb5d3@mail.gmail.com>
	<6af4270907260755x1928311fw64f9450422dfb8c1@mail.gmail.com>
	<20090726153856.GA28096@bolthole.com>
	<D92A2BB1-12C2-49A3-9C69-FEE730E3EE21@opencsw.org>
Message-ID: <20090726185405.GD39975@bolthole.com>

On Sun, Jul 26, 2009 at 08:16:29PM +0200, Dagobert Michelsen wrote:
> In GAR you can do this with
>
>> include gar/category.mk
>>
>> # Make sure we are using gnu tools (esp. sed)
>> PATH := /opt/csw/gnu:$(PATH)
>

please note, though, that will ONLY do it for build time tests. If the
program itself needs gdiff when installed.. you need to be more drastic!


From phil at bolthole.com  Sun Jul 26 20:57:23 2009
From: phil at bolthole.com (Philip Brown)
Date: Sun, 26 Jul 2009 11:57:23 -0700
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
References: <20090724155701.GB91227@bolthole.com>
	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>
	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
	<20090724194630.GA50326@bolthole.com>
	<1248465309-sup-92@ntdws12.chass.utoronto.ca>
	<20090724215550.GB73903@bolthole.com>
	<1248484642-sup-7791@ntdws12.chass.utoronto.ca>
	<20090726163122.GE28096@bolthole.com>
	<3E86772E-F48D-45EB-AF6D-03886BC6EB65@opencsw.org>
Message-ID: <20090726185723.GE39975@bolthole.com>

On Sun, Jul 26, 2009 at 08:18:59PM +0200, Dagobert Michelsen wrote:
>> there's also the concern of someone inevitably doing the equivalent of
>> "checkin *", and having the file "accidentally" get into the version
>> control system.
>
> With Subversion you explicitly add files once.


and someone adding new files, could still do

[add-once-command] *


but its potentially nice that you can exclude, as Ben says. need to hear
more about that.



From dam at opencsw.org  Sun Jul 26 23:19:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Sun, 26 Jul 2009 23:19:55 +0200
Subject: [csw-maintainers] General call: stuff in testing/
Message-ID: <4EB6CF3B-265B-489B-BA55-8FA33067B037@opencsw.org>

Hi,

there are quite a lot of packages in testing/ (300 files now). Maybe  
some
cool packages has been forgotten to release? It would be nice if  
everyone
could have a look at his packages and discard/release the package or  
leave
it there if it was for a reason that it is there ;-)

Status for my packages in testing:

- jdk/jre: still no permission from Sun for redistribution
- tcl: still not understood how locations on version upgrades work
- orca: start/stop methods still missing
- libgss: doubts from Phil that it is useful in addition to Suns version
- libgsf/libgda: linker issues as reported on maintainers@
- glib 1.2.10: just a rebuild with GAR, maybe released
- esound: needs more testing. volunteers?
- openldap: waits for berkeleydb general overhaul
- gdb 6.8: needs more testing
- other stuff: released to current, will be removed when arrived at  
mirrors

It would be nice if you all could also give a status, maybe some other
maintainers can help fixing things.


Thanks!

   -- Dago


From bwalton at opencsw.org  Sun Jul 26 23:35:25 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Sun, 26 Jul 2009 17:35:25 -0400
Subject: [csw-maintainers] General call: stuff in testing/
In-Reply-To: <4EB6CF3B-265B-489B-BA55-8FA33067B037@opencsw.org>
References: <4EB6CF3B-265B-489B-BA55-8FA33067B037@opencsw.org>
Message-ID: <1248644012-sup-613@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Sun Jul 26 17:19:55 -0400 2009:

Status for my packages in testing:

- ruby*: rolled today to address the recent security vulnerability in
  the previous release.  Will henceforth include a rubymode (emacs major mode)
  and rubymode_el (elisp source) subpackage.

I haven't installed these myself yet, but the standard ruby packages
should be sound...it's the rubymode* ones that may still have
packaging 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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090726/189664e9/attachment-0002.asc>

From trygvis at opencsw.org  Sun Jul 26 23:51:07 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Sun, 26 Jul 2009 23:51:07 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090726154345.GB28096@bolthole.com>
References: <20090723220149.GB8541@bolthole.com>
	<4A691BC9.2050209@opencsw.org>	<20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>	<20090725203035.GA24170@bolthole.com>	<4A6C2341.8060600@opencsw.org>
	<20090726154345.GB28096@bolthole.com>
Message-ID: <4A6CCFCB.2040204@opencsw.org>

Philip Brown wrote:
> On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst?l wrote:
>> Can you explain what the security reason for not putting it an SCM is?  
> 
> "a" source code control mechanism, I dont have much problem with.
> putting it in a PUBLIC one, is where my concerns lay.

Ok, that's good to know even if I don't see why it shouldn't be in a 
public one.

I'm still wondering what those "security" concerns was.

> Beyond that, I dont see the need to force everyone to use one particular
> type of scm, just for small web page "code".
> If the assorted people working on opencsw web stuff agree to standardize on
> one internal-only scheme, i'd go along with it. But frankly, most of the
> web related stuff is small enough that the old cycle of
> 
> "make a save copy, tweak with the new stuff, publish it, forget about old
>  stuff"
> 
> is perfectly adequate.

I think that's a point we just have to agree to disagree on.

--
Trygve


From trygvis at opencsw.org  Sun Jul 26 23:53:17 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Sun, 26 Jul 2009 23:53:17 +0200
Subject: [csw-maintainers] Updated packages for Munin 1.3.4 in testing
In-Reply-To: <op.uxfljpnjjbmpvo@imac.lan>
References: <op.uw59fh0kjbmpvo@imac.lan>
	<4A60C3AD.7050800@opencsw.org>	<4A660D38.7030600@opencsw.org>
	<op.uxfljpnjjbmpvo@imac.lan>
Message-ID: <4A6CD04D.3090500@opencsw.org>

Juergen Arndt wrote:
> On Tue, 21 Jul 2009 20:47:20 +0200, Trygve Laugst?l 
> <trygvis at opencsw.org> wrote:
> 
>> Trygve Laugst?l wrote:
>>> Juergen Arndt wrote:
>>>> Hi all,
>>>>
>>>> I updated the packages for Munin 1.3.4 in testing.
>>>>
>>>> - sample apache config (ready for include) in 
>>>> /etc/opt/csw/munin/apache/
>>>> - munin plugins moved to /opt/csw/libexec
>>>> - fix for using monospaced font
>>>>
>>>> Feedback is welcome.
>>>  I'll try these when I get back home next week.
>>
>> Can you do a rebuild with ARCH=all so I can test them on my x86 box?
> 
> In testing there are platform independent packages:
> 
> munin_master-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz
> munin_node-1.3.4,REV=2009.07.20-SunOS5.8-all-CSW.pkg.gz

They seem to work perfectly for me, I'm +1 for releasing those to current.

--
Trygve


From maciej at opencsw.org  Mon Jul 27 14:57:01 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 27 Jul 2009 13:57:01 +0100
Subject: [csw-maintainers] Shared library versions on build8st
Message-ID: <ce97543b0907270557q1bb02e79gebb73eb78e6790c5@mail.gmail.com>

The host build8s has CSWoldaprt-2.3. build8st on the other hand, has
CSWoldaprt-2.4. Binaries that depend on ldap, built on build8s, don't
work on build8st. I temporarily downgraded CSWoldaprt on build8st to
2.3, but I'm guessing I might be clashing with Dago who is working on
ldap. Perhaps we should coordinate who is working at the moment.

The current state is - openldap runtime 2.3. I'm testing cups-1.3.11.
I'll send an update when I'm done with it.


From maciej at opencsw.org  Mon Jul 27 18:01:15 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 27 Jul 2009 17:01:15 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
Message-ID: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>

I'm having an issue with the packaging of CUPS. It was all jolly good
until prototypes of some packages started becoming empty. It once was
CSWlibcups. Its prototype was simply empty, and so was the package.
Now libcups has the correct files, but cupsclient became empty. It's
reproducible with my account on both build8{s,x}. The build is at:

https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/pkg/cups/branches/cups-with-classutils

It's weird that the packages have been packaged correctly before, and
I haven't quite caught the moment the problem appeared.

Have you ever had similar issues? Is there a fix or a workaround?

Maciej


From dam at opencsw.org  Mon Jul 27 19:08:12 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 27 Jul 2009 19:08:12 +0200
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
Message-ID: <177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>

Hi Maciej,

Am 27.07.2009 um 18:01 schrieb Maciej (Matchek) Blizinski:
> I'm having an issue with the packaging of CUPS. It was all jolly good
> until prototypes of some packages started becoming empty. It once was
> CSWlibcups. Its prototype was simply empty, and so was the package.
> Now libcups has the correct files, but cupsclient became empty.

You forgot to close a bracket, see r5673.

> It's weird that the packages have been packaged correctly before, and
> I haven't quite caught the moment the problem appeared.

Perl bails out and all packages go into other packages. You can
track this with
   DEBUG_PACKAGING=1 gmake repackage

> Have you ever had similar issues? Is there a fix or a workaround?

Yes, write correct regexps ;-) But I do say that it is hard
sometimes...


Best regards

   -- Dago


From maciej at opencsw.org  Mon Jul 27 19:24:21 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Mon, 27 Jul 2009 18:24:21 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
	<177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>
Message-ID: <ce97543b0907271024k1fbc5119xe52c6700c416956f@mail.gmail.com>

On Mon, Jul 27, 2009 at 6:08 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
>> It's weird that the packages have been packaged correctly before, and
>> I haven't quite caught the moment the problem appeared.
>
> Perl bails out and all packages go into other packages. You can
> track this with
> ?DEBUG_PACKAGING=1 gmake repackage

A general idea... make a 'catch all' package, which is supposed to be
empty. If it contains any files, it means that some files haven't been
matched in other packages. Or perhaps mGAR could have such check:
"have all the regexes caught all the files from the install
directory?" What do you think Dago, what would be the best way to go
about it?

Maciej


From dam at opencsw.org  Mon Jul 27 20:23:14 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Mon, 27 Jul 2009 20:23:14 +0200
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <ce97543b0907271024k1fbc5119xe52c6700c416956f@mail.gmail.com>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
	<177F3BE0-2C6B-4CB2-BFC2-C413D3201180@opencsw.org>
	<ce97543b0907271024k1fbc5119xe52c6700c416956f@mail.gmail.com>
Message-ID: <C19FB3C7-1DE7-4036-98D9-D9A9CCC788BA@opencsw.org>

Hi Maciej,

Am 27.07.2009 um 19:24 schrieb Maciej (Matchek) Blizinski:
> A general idea... make a 'catch all' package, which is supposed to be
> empty.

There is already a catch-all package: the one for which you don't
specify PKGFILES_*

> If it contains any files, it means that some files haven't been
> matched in other packages.

Usually you split off certain functions into different packages:
documentation, development... and a main package - that main
package is meant to be the catch-all package.

> Or perhaps mGAR could have such check:
> "have all the regexes caught all the files from the install
> directory?" What do you think Dago, what would be the best way to go
> about it?

IMHO for the general case the current implementation is optimal.
In most cases you just say
   PKGFILES_CSWpkgdevel = $(PKGFILES_DEVEL)
and be done with it. I think it is important to make the most common
case easy, while still making hard things possible, and that is what
the current implementation does. I am not opposed to enhancements
but I can't think of a real enhancement right now.


Best regards

   -- Dago


From phil at bolthole.com  Mon Jul 27 23:21:17 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 27 Jul 2009 14:21:17 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6CCFCB.2040204@opencsw.org>
References: <20090724041722.GA50043@bolthole.com>
	<20090724042720.GB50043@bolthole.com>
	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
	<4A6C2341.8060600@opencsw.org>
	<20090726154345.GB28096@bolthole.com>
	<4A6CCFCB.2040204@opencsw.org>
Message-ID: <20090727212117.GG90811@bolthole.com>

On Sun, Jul 26, 2009 at 11:51:07PM +0200, Trygve Laugst?l wrote:
> Philip Brown wrote:
>> On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst?l wrote:
>>> Can you explain what the security reason for not putting it an SCM 
>>> is?  
>>
>> "a" source code control mechanism, I dont have much problem with.
>> putting it in a PUBLIC one, is where my concerns lay.
>
> Ok, that's good to know even if I don't see why it shouldn't be in a  
> public one.

okay then... for the record, where do you do your "daytime work" as a web
developer, and where can we publically find ALL the source code, etc. for
the work you do there?



From william at wbonnet.net  Mon Jul 27 23:47:57 2009
From: william at wbonnet.net (William Bonnet)
Date: Mon, 27 Jul 2009 23:47:57 +0200
Subject: [csw-maintainers] Subversion on bender
In-Reply-To: <1248462873-sup-4973@ntdws12.chass.utoronto.ca>
References: <ce97543b0907240245s77a2ba61nf885d9c0a9d0411@mail.gmail.com>	<ce97543b0907240252y610fe865sab579207613e37bd@mail.gmail.com>	<4A69C6F2.2070003@dogan.ch>	<1248447803-sup-4618@ntdws12.chass.utoronto.ca>	<4A69D719.6080409@opencsw.org>	<20090724155701.GB91227@bolthole.com>	<1248458300-sup-2415@ntdws12.chass.utoronto.ca>	<20090724182421.GB16963@bolthole.com>
	<1248462873-sup-4973@ntdws12.chass.utoronto.ca>
Message-ID: <4A6E208D.7000605@wbonnet.net>

Hi
> William: What is the status of the mock-up?  Are there stumbling
> blocks you're facing other than time constraints?
So far mostly time problem. It has moved back to top priority since now 
i'm stuck with FF3 for a few days / week ?

The current status of the mockup is the following :

A mockup based on wordpress has been created.  After talking with 
Sebastian, i have set up a first structure of menu. The are divided in 4 
main sections which i would like to be simple and clear. Gathering the 
mainly used actions

Get It
    This menu is about how to get OpenCSW, what is OpenCSW and 
informations about release content (packages etc)
    It includes informations for the first time user, trying to download 
and install opencsw for the first time or on a new box

Use It
    This menu is about "every day" use. How to update packages, 
configure specific things, etc.

Spread It
    This menu is about advance use (rebuilding packagew with gar, 
creating a new package, and contributing to the project

Support
    This menu is about how to get support (bug tracking, wiki, faq, etc.)

The menus are a proposal for web site structure. Of course a main page 
with direct acces link will be here (i don't like the front page i set 
up which is more like a magazine still).

A temporary menu that will disappear is present under the name "Old 
menus". This is group old pages, which a copy paste of the current web 
site (it does not mean i want to delete this pages, but just my "working 
menu").

Any review comment on this menu structure will be welcomed. I will apply 
changes to the structure. Tomorrow and in the next day i'll rewrite some 
css part and main page. contribution to text content will be highly 
appreciated.

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 trygvis at opencsw.org  Mon Jul 27 23:52:48 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Mon, 27 Jul 2009 23:52:48 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090727212117.GG90811@bolthole.com>
References: <20090724041722.GA50043@bolthole.com>	<20090724042720.GB50043@bolthole.com>	<ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>	<20090725203035.GA24170@bolthole.com>	<4A6C2341.8060600@opencsw.org>	<20090726154345.GB28096@bolthole.com>	<4A6CCFCB.2040204@opencsw.org>
	<20090727212117.GG90811@bolthole.com>
Message-ID: <4A6E21B0.90606@opencsw.org>

Philip Brown wrote:
> On Sun, Jul 26, 2009 at 11:51:07PM +0200, Trygve Laugst?l wrote:
>> Philip Brown wrote:
>>> On Sun, Jul 26, 2009 at 11:34:57AM +0200, Trygve Laugst?l wrote:
>>>> Can you explain what the security reason for not putting it an SCM 
>>>> is?  
>>> "a" source code control mechanism, I dont have much problem with.
>>> putting it in a PUBLIC one, is where my concerns lay.
>> Ok, that's good to know even if I don't see why it shouldn't be in a  
>> public one.
> 
> okay then... for the record, where do you do your "daytime work" as a web
> developer, and where can we publically find ALL the source code, etc. for
> the work you do there?

What's the relevance of that question?

You still skipped my other question on "security".

--
Trygve


From phil at bolthole.com  Mon Jul 27 23:54:42 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 27 Jul 2009 14:54:42 -0700
Subject: [csw-maintainers] berkeleydb packages - changes
Message-ID: <20090727215442.GM90811@bolthole.com>

btw, if anyone is not subscribed to either the users list or announce list
:-}

  ------------------------------------------------------------


This is an important announcement to opencsw users, who use our berkeleydb
packages.

Please be advised that we are making a MAJOR CHANGE to the way we handle
berkeleydb packages.

in the past, we have had many split up versions of berkeleydb4.
specifically, 4.2, 4.3, and 4.4, all separately.

>From here on in, we plan to to consolidate bdb 4.x use, into a single
bdb4.x library. 
Upgrades of our packages, should pull in these changes.

There is also a minor change in the bdb 3.x package, that does not
affect the shared libraries there at all.

In theory, this should be a transparent operation for everyone, but we
wished to advise people of this issue, in case you wished to do additional
testing, before rolling out the updates more widely in your network.

Thank you for your attention, and we hope this makes for a nice improvement
to our "customers" :-)


From phil at bolthole.com  Tue Jul 28 00:04:10 2009
From: phil at bolthole.com (Philip Brown)
Date: Mon, 27 Jul 2009 15:04:10 -0700
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <4A6E21B0.90606@opencsw.org>
References: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>
	<4A6979F4.5050904@opencsw.org>
	<20090724160343.GC91227@bolthole.com>
	<4A6B4DFE.1050307@opencsw.org>
	<20090725203035.GA24170@bolthole.com>
	<4A6C2341.8060600@opencsw.org>
	<20090726154345.GB28096@bolthole.com>
	<4A6CCFCB.2040204@opencsw.org>
	<20090727212117.GG90811@bolthole.com> <4A6E21B0.90606@opencsw.org>
Message-ID: <20090727220410.GN90811@bolthole.com>

On Mon, Jul 27, 2009 at 11:52:48PM +0200, Trygve Laugst??l wrote:
> Philip Brown wrote:
>> okay then... for the record, where do you do your "daytime work" as a web
>> developer, and where can we publically find ALL the source code, etc. for
>> the work you do there?
>
> What's the relevance of that question?

The relevance is, if your employer(s) see security issues, or any other
issues for that matter, from standing in the way of making all their web
source public, and you are ok with that there, then it seems disingenuous
of you to fuss about our web sources not being completely public either.

So lets move on, and focus on what we can do to improve our web source
internally, discussing *internal* source control, 
and drop the needless distraction of making it a public one.



From trygvis at opencsw.org  Tue Jul 28 00:13:04 2009
From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Tue, 28 Jul 2009 00:13:04 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090727220410.GN90811@bolthole.com>
References: <ce97543b0907240123s3b1f70b9vc152cd21975d2118@mail.gmail.com>	<4A6979F4.5050904@opencsw.org>	<20090724160343.GC91227@bolthole.com>	<4A6B4DFE.1050307@opencsw.org>	<20090725203035.GA24170@bolthole.com>	<4A6C2341.8060600@opencsw.org>	<20090726154345.GB28096@bolthole.com>	<4A6CCFCB.2040204@opencsw.org>	<20090727212117.GG90811@bolthole.com>
	<4A6E21B0.90606@opencsw.org> <20090727220410.GN90811@bolthole.com>
Message-ID: <4A6E2670.2070104@opencsw.org>

Philip Brown wrote:
> On Mon, Jul 27, 2009 at 11:52:48PM +0200, Trygve Laugst?l wrote:
>> Philip Brown wrote:
>>> okay then... for the record, where do you do your "daytime work" as a web
>>> developer, and where can we publically find ALL the source code, etc. for
>>> the work you do there?
>> What's the relevance of that question?
> 
> The relevance is, if your employer(s) see security issues, or any other
> issues for that matter, from standing in the way of making all their web
> source public, and you are ok with that there, then it seems disingenuous
> of you to fuss about our web sources not being completely public either.

I find the security argument ridiculous given that this is a read only 
application reading stuff from a database. If a person (either because 
of competence or tool chain or any other reason) can't write code that's 
safe enough to be shown to the public, it shouldn't run anywhere.

There are lots of applications which need to be very secure which are 
open source. Why should our source code be different?

> So lets move on, and focus on what we can do to improve our web source
> internally, discussing *internal* source control, 
> and drop the needless distraction of making it a public one.

Looks like another point we have to agree to disagree on, so ok. It's 
dropped.

--
Trygve


From mwatters at opencsw.org  Tue Jul 28 03:24:33 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Mon, 27 Jul 2009 20:24:33 -0500
Subject: [csw-maintainers] Please update berkeleydb* on the buildfarm please
Message-ID: <4A6E5351.6080209@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


- --
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/

iEYEARECAAYFAkpuU1AACgkQLrhmsXMSLxdJ+gCcDhqkXsdmNr3YMp3oyt9MdmDe
5DgAoJDs+HI0LMpmRB/ny3fESFjVlUKu
=w8Wm
-----END PGP SIGNATURE-----


From dam at opencsw.org  Tue Jul 28 08:58:22 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Jul 2009 08:58:22 +0200
Subject: [csw-maintainers] Updating all buildfarm servers to current now
In-Reply-To: <4A6E5351.6080209@opencsw.org>
References: <4A6E5351.6080209@opencsw.org>
Message-ID: <F57FFFB0-74FD-480D-BE74-E26957F08D00@opencsw.org>

Hi,

due to massive package updates I will be upgrading all buildfarm servers
to current/ now.


Best regards

   -- Dago


From jeff at cjsa.com  Tue Jul 28 09:33:17 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Tue, 28 Jul 2009 07:33:17 GMT
Subject: [csw-maintainers] berkeley DB problems
Message-ID: <KnHEBH.M2E@cjsa.com>


OK, I saw the recent announcement regarding the berkeley DB changes.  The
notuce said a simple upgrade should address the issue.  I did a complete
upgrade for the latestset of released packages and now have the following
installed:

-------------------------------------------------------------------------------
berkeleydb                   4.7.25,REV=2009.07.01                          SAME
berkeleydb3                  3.3.11,REV=2009.07.22                          SAME
berkeleydb4           4.2.52,REV=2008.02.23_rev=p5         4.1.25,REV=2009.06.26
berkeleydb44                 4.4.20,REV=2009.06.26                          SAME
-------------------------------------------------------------------------------

Why the berkeleydb4 is now showing a later rev than the catalog is a mystery.
But now my berkeley 4x libraries are hosed as witness an attempt to run
perl:

-------------------------------------------------------------------------------
1-> perl
ld.so.1: perl: fatal: libdb-4.4.so: open failed: No such file or directory
Killed
-------------------------------------------------------------------------------

Here are the /opt/csw/lib/libdb* libraries now installed on my system:

-------------------------------------------------------------------------------
libdb-3.3.so
libdb-4.7.so*
libdb-4.so@
libdb.so@
libdb_cxx-4.7.so*
libdb_cxx-4.so@
libdb_cxx.so@
libdb_java-4.7.so*
libdb_java-4.7_g.so@
libdb_java-4.so@
libdb_java.so@
-------------------------------------------------------------------------------

So perl, and possibly other programs are looking for libdb-4.4.so which no
longer exists.

Suggestions to get the system back to a sane state are welcome.

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040



From jeff at cjsa.com  Tue Jul 28 09:37:32 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Tue, 28 Jul 2009 07:37:32 GMT
Subject: [csw-maintainers] berkeley DB problems: follows up
Message-ID: <KnHEIK.16x@cjsa.com>

BTW, as a temporary workaround, I created the following link:

    libdb-4.4.so -> libdb-4.7.so

This seems to be addressing the problem for the moment.

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040



From dam at opencsw.org  Tue Jul 28 10:16:54 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Jul 2009 10:16:54 +0200
Subject: [csw-maintainers] Current issue with the buildfarm
Message-ID: <6BC9FC44-B109-4517-BD08-10393C5E83BB@opencsw.org>

Hi,

there is an issue with the updated release of the BerkeleyDB which  
affects
the buildfarm. I am working on it with PRIO 1 as it will also affect
user experience :-((


Best regards

   -- Dago


From dam at opencsw.org  Tue Jul 28 10:31:41 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Tue, 28 Jul 2009 10:31:41 +0200
Subject: [csw-maintainers] berkeley DB problems
In-Reply-To: <KnHEBH.M2E@cjsa.com>
References: <KnHEBH.M2E@cjsa.com>
Message-ID: <2C97A401-0BD2-4CB9-BF14-EB8FE95039F4@opencsw.org>

Hi Jeff,

Am 28.07.2009 um 09:33 schrieb Jeffery Small:
> But now my berkeley 4x libraries are hosed as witness an attempt to  
> run
> perl:

I just fixed this. Please use the updated packages from testing/ until
the new version hits the mirrors.


Sorry for the inconvenience

   -- Dago


From bwalton at opencsw.org  Tue Jul 28 16:04:28 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 10:04:28 -0400
Subject: [csw-maintainers] php_mode for emacs
Message-ID: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>


Hi All,

I've put php_mode for emacs in testing (and asked Phil to release it
since it's such a simple package and is working for me).  Just a heads
up for anyone doing php stuff and using emacs.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/340b8bec/attachment-0002.asc>

From bwalton at opencsw.org  Tue Jul 28 16:28:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 10:28:59 -0400
Subject: [csw-maintainers] new git book
Message-ID: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>


For anyone out there using git, you may be interested in the soon to
be released, but already available book titled 'Pro Git.'  It's written
by Scott Chacon (http://jointheconversation.org/about.html) and is
licensed the CC.

http://progit.org/

Enjoy!
-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/1718cf1f/attachment-0002.asc>

From phil at bolthole.com  Tue Jul 28 18:18:06 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 09:18:06 -0700
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
Message-ID: <20090728161806.GA59031@bolthole.com>

On Tue, Jul 28, 2009 at 10:04:28AM -0400, Ben Walton wrote:
> 
> Hi All,
> 
> I've put php_mode for emacs in testing (and asked Phil to release it
> since it's such a simple package and is working for me).

Errr... please dont do this.
any package should be in EITHER testing, OR newpkgs. never both.



From pfelecan at opencsw.org  Tue Jul 28 18:22:12 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Tue, 28 Jul 2009 18:22:12 +0200
Subject: [csw-maintainers] gcc 3.4.6 available for test
Message-ID: <x6iqhcok0b.fsf@abhavana.free.fr>

This version of the GCC suite is the last one in this branch of the
project. I'm planning to release-it toward the end of this week. The
packages are available at: http://mirror.opencsw.org/testing/

If somebody finds an issue, please send me a description and a mean to
reproduce-it.

TIA
-- 
Peter


From bwalton at opencsw.org  Tue Jul 28 18:42:00 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 12:42:00 -0400
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <20090728161806.GA59031@bolthole.com>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
Message-ID: <1248799206-sup-9413@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 12:18:06 -0400 2009:
> Errr... please dont do this.
> any package should be in EITHER testing, OR newpkgs. never both.

It'll be removed from testing when it hits the mirrors[1]...avoids other
issues and makes it available ahead of the mirrors if anyone is
chomping at the bit for it (not likely in this case).

-Ben

[1] This is my normal workflow, I just don't generally release
    something the same day I roll it.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/ad0236c7/attachment-0002.asc>

From maciej at opencsw.org  Tue Jul 28 19:16:53 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 28 Jul 2009 18:16:53 +0100
Subject: [csw-maintainers] Actual downloads of packages from testing
Message-ID: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>

When we put packages in testing, we are hoping that someone would
download and install them. Of course, we can't force anyone to test
packages, but we could see whether individual packages actually get
downloaded. We could assume that at least some of the downloads result
in testing. The reason why I'm asking is that I have a feeling that
some of the packages that spend time in testing aren't actually tested
before going into newpkgs. Once we know whether it's the case, we can
think of ways to promote the testing of packages; It would be good to
collect some data first.

Thoughts?

Maciej


From bwalton at opencsw.org  Tue Jul 28 19:22:27 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 13:22:27 -0400
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
Message-ID: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28 13:16:53 -0400 2009:

> Thoughts?

There have been requests for stats data on released packages in the
past.  That's tough to get since it would require data from every
mirror.  This should be fairly simple to get though, since testing is
centralized.  +1 for making this info available somehow...would
awstats work for this task?

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/5c29b31f/attachment-0002.asc>

From maciej at opencsw.org  Tue Jul 28 20:08:47 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Tue, 28 Jul 2009 19:08:47 +0100
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
Message-ID: <ce97543b0907281108s7f489426maeb7a595b3747c02@mail.gmail.com>

On Tue, Jul 28, 2009 at 6:22 PM, Ben Walton<bwalton at opencsw.org> wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28 13:16:53 -0400 2009:
>
>> Thoughts?
>
> There have been requests for stats data on released packages in the
> past. ?That's tough to get since it would require data from every
> mirror. ?This should be fairly simple to get though, since testing is
> centralized. ?+1 for making this info available somehow...would
> awstats work for this task?

I think so, yes. Perhaps there could be a filter in the Apache config
which would make an additional log file with the package files (as
opposed to HTML or anything else). Awstats would read just that file.

The Apache config could be something along the lines of:

<Location /testing>
   CustomLog [...]
</Location>

Maciej


From phil at bolthole.com  Tue Jul 28 20:13:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:13:21 -0700
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248799206-sup-9413@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
Message-ID: <20090728181321.GA58718@bolthole.com>

On Tue, Jul 28, 2009 at 12:42:00PM -0400, Ben Walton wrote:
> Excerpts from Philip Brown's message of Tue Jul 28 12:18:06 -0400 2009:
> > Errr... please dont do this.
> > any package should be in EITHER testing, OR newpkgs. never both.
> 
> It'll be removed from testing when it hits the mirrors[1]...avoids other
> issues and makes it available ahead of the mirrors if anyone is
> chomping at the bit for it (not likely in this case).

let me repeat:

"any package should be in EITHER testing, OR newpkgs. never both."

Once you have decided a package is ready for release, then it doesnt
belong in testing any more. testing, is for pre-release packages.

It should be *removed* from testing, *before* putting it in newpkgs.

testing gets bogged up enough as it is.


From phil at bolthole.com  Tue Jul 28 20:15:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:15:26 -0700
Subject: [csw-maintainers] new git book
In-Reply-To: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
Message-ID: <20090728181526.GC58718@bolthole.com>

On Tue, Jul 28, 2009 at 10:28:59AM -0400, Ben Walton wrote:
> 
> For anyone out there using git, you may be interested in the soon to
> be released, but already available book titled 'Pro Git.'  It's written
> by Scott Chacon (http://jointheconversation.org/about.html) and is
> licensed the CC.

I want to see a book named "Total Git".
With the author's picture on the front.


:-)



From bwalton at opencsw.org  Tue Jul 28 20:19:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 14:19:59 -0400
Subject: [csw-maintainers] new git book
In-Reply-To: <20090728181526.GC58718@bolthole.com>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
	<20090728181526.GC58718@bolthole.com>
Message-ID: <1248805133-sup-7458@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 14:15:26 -0400 2009:
> I want to see a book named "Total Git".
> With the author's picture on the front.

Wouldn't it be even better to be a Pro (paid) Git though? :) Linus
enjoys a good laugh about the name when asked too!

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/2c267be5/attachment-0002.asc>

From bwalton at opencsw.org  Tue Jul 28 20:25:44 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 14:25:44 -0400
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <20090728181321.GA58718@bolthole.com>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
	<20090728181321.GA58718@bolthole.com>
Message-ID: <1248805220-sup-8974@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 14:13:21 -0400 2009:

> let me repeat:
> 
> "any package should be in EITHER testing, OR newpkgs. never both."
> 
> Once you have decided a package is ready for release, then it doesnt
> belong in testing any more. testing, is for pre-release packages.

Well yes, that's all well and good, but sometimes 'bad things happen'
and the copy from newpkgs/ gets rm'd before it is released.  Since I
wouldn't want to store a copy in my local ~/, I leave the copy in
testing/ until it gets released.  This saves me work in the event of
mishaps.

If the release process were such that things like the above didn't
happen, I'd be happy to squish the package from testing after I've
placed it in newpkgs/.

Maybe a cron job to rm files in testing older than 30 days?  (With an
exclusion list?)

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/f3aec094/attachment-0002.asc>

From phil at bolthole.com  Tue Jul 28 20:29:26 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:29:26 -0700
Subject: [csw-maintainers] new git book
In-Reply-To: <1248805133-sup-7458@ntdws12.chass.utoronto.ca>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
	<20090728181526.GC58718@bolthole.com>
	<1248805133-sup-7458@ntdws12.chass.utoronto.ca>
Message-ID: <20090728182926.GD58718@bolthole.com>

On Tue, Jul 28, 2009 at 02:19:59PM -0400, Ben Walton wrote:
> Wouldn't it be even better to be a Pro (paid) Git though? :) Linus
> enjoys a good laugh about the name when asked too!

Yeah, that's pretty much Linus's job.
Paid Professional Git. glad he likes it :-)

[ as opposed to me, UNpaid professional git. When do I get my 
 opensource handout, Linus?   ha]


From phil at bolthole.com  Tue Jul 28 20:31:37 2009
From: phil at bolthole.com (Philip Brown)
Date: Tue, 28 Jul 2009 11:31:37 -0700
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248805220-sup-8974@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
	<20090728181321.GA58718@bolthole.com>
	<1248805220-sup-8974@ntdws12.chass.utoronto.ca>
Message-ID: <20090728183137.GF58718@bolthole.com>

On Tue, Jul 28, 2009 at 02:25:44PM -0400, Ben Walton wrote:
> Well yes, that's all well and good, but sometimes 'bad things happen'
> and the copy from newpkgs/ gets rm'd before it is released.  Since I
> wouldn't want to store a copy in my local ~/, 

why not? at least then it's only cluttering up YOUR space :-)

$ mv testing/xyz ~

instead of

$ rm testing/xyz




From bwalton at opencsw.org  Tue Jul 28 20:52:10 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 14:52:10 -0400
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <20090728183137.GF58718@bolthole.com>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
	<20090728181321.GA58718@bolthole.com>
	<1248805220-sup-8974@ntdws12.chass.utoronto.ca>
	<20090728183137.GF58718@bolthole.com>
Message-ID: <1248806415-sup-2819@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Tue Jul 28 14:31:37 -0400 2009:

> why not? at least then it's only cluttering up YOUR space :-)

It's the same space (just different eyeballs on it), but
whatever...fine.  You don't delete packages accidentally and I'll
remove them from testing/ after requesting the release.  :)

I've got a few ideas for changes in this process that I'll document
(with Maciek) and propose in a completed form.

-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090728/7fdcd342/attachment-0002.asc>

From mwatters at opencsw.org  Wed Jul 29 00:34:54 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Jul 2009 17:34:54 -0500
Subject: [csw-maintainers] new git book
In-Reply-To: <20090728181526.GC58718@bolthole.com>
References: <1248791157-sup-9285@ntdws12.chass.utoronto.ca>
	<20090728181526.GC58718@bolthole.com>
Message-ID: <4A6F7D0E.2010104@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
>
> I want to see a book named "Total Git".
> With the author's picture on the front.

no, with picture of Chris Barrie Holding an 'H'

just to see how many people can get the reference.




- --

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

iEYEARECAAYFAkpvfQ4ACgkQLrhmsXMSLxcUMACg4WbLRH92mYxNmo1cGExYJggT
f9IAoOvSHIc39zveRNNNdixOUJwtZNqm
=pt5o
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Wed Jul 29 00:36:49 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Tue, 28 Jul 2009 17:36:49 -0500
Subject: [csw-maintainers] pidgin 2.5.8 now in testing
Message-ID: <4A6F7D81.6080106@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This version fixes the yahoo instant messenger login protocol problem.


- --

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

iEYEARECAAYFAkpvfYEACgkQLrhmsXMSLxfttACeORnnA1rtcYJEaR5ioywBhFfK
boYAoISjCTdXw2DpVdadSQ1A+t9jBWEJ
=UK67
-----END PGP SIGNATURE-----


From trygvis at opencsw.org  Wed Jul 29 01:10:19 2009
From: trygvis at opencsw.org (=?UTF-8?B?VHJ5Z3ZlIExhdWdzdMO4bA==?=)
Date: Wed, 29 Jul 2009 01:10:19 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
Message-ID: <4A6F855B.9020209@opencsw.org>

Ben Walton wrote:
> Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28 13:16:53 -0400 2009:
> 
>> Thoughts?
> 
> There have been requests for stats data on released packages in the
> past.  That's tough to get since it would require data from every
> mirror.  This should be fairly simple to get though, since testing is
> centralized.  +1 for making this info available somehow...would
> awstats work for this task?

Yeah, that would be nice.

It might be hard to estimate the impact on a general basis, but it 
should be possible to measure the impact of individual packages between 
releases.

--
Trygve



From maciej at opencsw.org  Wed Jul 29 01:16:22 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 00:16:22 +0100
Subject: [csw-maintainers] setxkbmap investigation
Message-ID: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>

I've recently built a setxkbmap binary. There's something I can't
quite work out, perhaps somebody will be able to help me out.

Information about the package is here:
http://lists.opencsw.org/pipermail/users/2009-July/004052.html
The Makefile:
http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/setxkbmap/trunk/Makefile

I have a reference binary in /var/tmp, which reads the keymap from a
known location: /usr/X11/lib/X11/xkb/symbols. For instance, when I run
'setxkbmap pl', the file /usr/X11/lib/X11/xkb/symbols/pl gets read.
When I run my setxkbmap, they 'pl' keymap must get loaded from
someplace else, because it's not my (patched) keymap. But I can't
figure out, where it is being read from.

I don't see much choice there:

$ glocate -r 'X11.*/pl$'
/usr/openwin/lib/X11/xkb/keymap/sgi/pl
/usr/openwin/lib/X11/xkb/keymap/sun/pl
/usr/openwin/lib/X11/xkb/symbols/pl
/usr/openwin/lib/X11/xkb/symbols/sun/pl
/usr/X11/lib/X11/xkb/keymap/sgi/pl
/usr/X11/lib/X11/xkb/keymap/sun/pl
/usr/X11/lib/X11/xkb/symbols/pc/pl
/usr/X11/lib/X11/xkb/symbols/pl
/usr/X11/lib/X11/xkb/symbols/sun/pl

I trussed the setxkbmap process (with -f to trace also the child
processes), but it doesn't open any of the above files. Do you have a
good idea where it can read from or how to track that down?

Maciej


From bwalton at opencsw.org  Wed Jul 29 01:30:55 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Tue, 28 Jul 2009 19:30:55 -0400
Subject: [csw-maintainers] package hook document
Message-ID: <1248823765-sup-6856@ntdws12.chass.utoronto.ca>


Would anyone interested in the package hook implementation please
have a look at: http://wiki.opencsw.org/package-hooks

I'd like to keep this moving along...

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 dam at opencsw.org  Wed Jul 29 08:29:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 08:29:08 +0200
Subject: [csw-maintainers] php_mode for emacs
In-Reply-To: <1248799206-sup-9413@ntdws12.chass.utoronto.ca>
References: <1248789806-sup-1061@ntdws12.chass.utoronto.ca>
	<20090728161806.GA59031@bolthole.com>
	<1248799206-sup-9413@ntdws12.chass.utoronto.ca>
Message-ID: <35350311-D392-4B8F-8C8E-E95CD312DC4C@opencsw.org>

Hi Ben,

Am 28.07.2009 um 18:42 schrieb Ben Walton:
> It'll be removed from testing when it hits the mirrors[1]...avoids  
> other
> issues and makes it available ahead of the mirrors if anyone is
> chomping at the bit for it (not likely in this case).

That is ok, my previous mailing about cleaning up testing was about
stuff that was forgotten in testing/. The few extra days for full
mirror sync are no real problem here. Apart from that I did receive
some emails in the past where the package was after I removed it
from testing but before mirrors were synced.


Best regards

   -- Dago


From dam at opencsw.org  Wed Jul 29 08:46:30 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 08:46:30 +0200
Subject: [csw-maintainers] setxkbmap investigation
In-Reply-To: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>
References: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>
Message-ID: <13B9E170-BB9C-405F-A567-919A55AB10D4@opencsw.org>

Hi Maciej,

Am 29.07.2009 um 01:16 schrieb Maciej (Matchek) Blizinski:
> I have a reference binary in /var/tmp, which reads the keymap from a
> known location: /usr/X11/lib/X11/xkb/symbols. For instance, when I run
> 'setxkbmap pl', the file /usr/X11/lib/X11/xkb/symbols/pl gets read.
> When I run my setxkbmap, they 'pl' keymap must get loaded from
> someplace else, because it's not my (patched) keymap. But I can't
> figure out, where it is being read from.

You can try "opensnoop" from the dtracetoolkit (pkg-get -i  
dtracetoolkit):

root at ncsw [global]:/root > opensnoop
   UID    PID COMM          FD PATH
     0  16783 mibiisa        2 /dev/kstat
     0  16783 mibiisa       -1 /dev/e
     0  16783 mibiisa        2 /dev/kstat
     0  18355 sysstatd     551 /proc/5045/psinfo
     0  18355 sysstatd     552 /proc/5045/status
     0  18355 sysstatd     569 /proc/5033/psinfo
     0  18355 sysstatd     570 /proc/5033/status
     0  18355 sysstatd     599 /proc/5031/psinfo
     0  18355 sysstatd     600 /proc/5031/status
     0  18355 sysstatd     899 /proc/5046/psinfo
     0  18355 sysstatd     900 /proc/5046/status
10000   5047 setxkbmap     -1 /var/ld/ld.config
10000   5047 setxkbmap      3 /lib/libc.so.1
10000   5047 setxkbmap      3 /usr/lib/libxkbfile.so.4
10000   5047 setxkbmap      3 /platform/SUNW,SPARC-Enterprise-T5220/ 
lib/libc_psr.so.1
10000   5047 setxkbmap      3 /usr/lib/libX11.so.4
10000   5047 setxkbmap      3 /lib/libsocket.so.1
10000   5047 setxkbmap      3 /lib/libnsl.so.1
10000   5047 setxkbmap      4 /etc/netconfig
10000   5047 setxkbmap      4 /dev/udp
10000   5047 setxkbmap      4 /var/run/name_service_door
10000   5047 setxkbmap      5 /etc/default/nss
10000   5047 setxkbmap      5 /dev/udp6
10000   5047 setxkbmap      5 /home/dam/.Xauthority
10000   5047 setxkbmap     -1 /usr/X11R6/lib/X11/xkb/rules/xorg-C.lst
10000   5047 setxkbmap      5 /usr/X11R6/lib/X11/xkb/rules/xorg.lst
10000   5047 setxkbmap     -1 /usr/X11R6/lib/X11/xkb/rules/xorg-C
10000   5047 setxkbmap      5 /usr/X11R6/lib/X11/xkb/rules/xorg
10000   5047 setxkbmap     -1 ./rules/xorg-C.lst
10000   5047 setxkbmap     -1 ./rules/xorg.lst
     1  17464 nfsmapid       8 /etc/resolv.conf
     0  16448 nscd           4 /etc/inet/ipnodes
^C

However, setxkbmap doesn't work in my environment:

dam at build10st [build10st]:/home/dam > setxkbmap pl
Couldn't interpret _XKB_RULES_NAMES property
Use defaults: rules - 'xorg' model - 'pc101' layout - 'us'
Error loading new keyboard description
zsh: 5047 exit 251   setxkbmap pl



Best regards

   -- Dago


From maciej at opencsw.org  Wed Jul 29 12:16:18 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 11:16:18 +0100
Subject: [csw-maintainers] setxkbmap investigation
In-Reply-To: <13B9E170-BB9C-405F-A567-919A55AB10D4@opencsw.org>
References: <ce97543b0907281616h2d1f06d3pe2a964616308a2fc@mail.gmail.com>
	<13B9E170-BB9C-405F-A567-919A55AB10D4@opencsw.org>
Message-ID: <ce97543b0907290316kab18346gef8930f582e91d30@mail.gmail.com>

On Wed, Jul 29, 2009 at 7:46 AM, Dagobert Michelsen<dam at opencsw.org> wrote:
> Hi Maciej,
>
> Am 29.07.2009 um 01:16 schrieb Maciej (Matchek) Blizinski:
>>
>> I have a reference binary in /var/tmp, which reads the keymap from a
>> known location: /usr/X11/lib/X11/xkb/symbols. For instance, when I run
>> 'setxkbmap pl', the file /usr/X11/lib/X11/xkb/symbols/pl gets read.
>> When I run my setxkbmap, they 'pl' keymap must get loaded from
>> someplace else, because it's not my (patched) keymap. But I can't
>> figure out, where it is being read from.
>
> You can try "opensnoop" from the dtracetoolkit (pkg-get -i dtracetoolkit):
>
> root at ncsw [global]:/root > opensnoop
> ?UID ? ?PID COMM ? ? ? ? ?FD PATH
> ? ?0 ?16783 mibiisa ? ? ? ?2 /dev/kstat

I installed the dtracetoolkit, thanks! opensnoop is really useful.

http://dpaste.com/72855/ -- looks like it's the .../pc/pl file that
gets loaded by xkbcomp.

> However, setxkbmap doesn't work in my environment:
>
> dam at build10st [build10st]:/home/dam > setxkbmap pl
> Couldn't interpret _XKB_RULES_NAMES property
> Use defaults: rules - 'xorg' model - 'pc101' layout - 'us'
> Error loading new keyboard description
> zsh: 5047 exit 251 ? setxkbmap pl

Perhaps X11 packages aren't installed.

Thanks for the help. I'll poke it some more to if I can get it to load
the correct keyboard mapping.

Maciej


From maciej at opencsw.org  Wed Jul 29 12:44:49 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 11:44:49 +0100
Subject: [csw-maintainers] A copyright notice in the template
Message-ID: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>

I've committed a change to the template: I added a short copyright
notice (GPLv2) and set the Id keyword so that the Makefile would
contain information about the current file revision:

https://sourceforge.net/apps/trac/gar/changeset/5704

Since it's the template, the impact would be only on new packages; a
package author could always remove the notice if they wanted.

What do you think of this change? Is it okay?

Maciej


From dam at opencsw.org  Wed Jul 29 13:22:08 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 13:22:08 +0200
Subject: [csw-maintainers] A copyright notice in the template
In-Reply-To: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>
References: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>
Message-ID: <2484D1C9-2EC1-4F69-A9A1-9F0BD33565D2@opencsw.org>

Hi Maciej,

Am 29.07.2009 um 12:44 schrieb Maciej (Matchek) Blizinski:
> I've committed a change to the template: I added a short copyright
> notice (GPLv2) and set the Id keyword so that the Makefile would
> contain information about the current file revision:
>
> https://sourceforge.net/apps/trac/gar/changeset/5704
>
> Since it's the template, the impact would be only on new packages; a
> package author could always remove the notice if they wanted.
>
> What do you think of this change? Is it okay?

I think it is a good idea for a template. There are also programmatical
templates per-category. Do you mind updating these in
   pkg/Makefile
   pkg/cpan/Makefile
   pkg/xfce/Makefile
   pkg/x11/Makefile
also?


Best regards

   -- Dago


From maciej at opencsw.org  Wed Jul 29 14:52:25 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 13:52:25 +0100
Subject: [csw-maintainers] A copyright notice in the template
In-Reply-To: <2484D1C9-2EC1-4F69-A9A1-9F0BD33565D2@opencsw.org>
References: <ce97543b0907290344l39bccf2bl6e68929ee500d452@mail.gmail.com>
	<2484D1C9-2EC1-4F69-A9A1-9F0BD33565D2@opencsw.org>
Message-ID: <ce97543b0907290552j44ee09c0m995d9afc0aef50a0@mail.gmail.com>

On Wed, Jul 29, 2009 at 12:22 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> I think it is a good idea for a template. There are also programmatical
> templates per-category. Do you mind updating these in
> ?pkg/Makefile
> ?pkg/cpan/Makefile
> ?pkg/xfce/Makefile
> ?pkg/x11/Makefile
> also?

Done, headers added.

Maciej


From maciej at opencsw.org  Wed Jul 29 15:07:09 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 14:07:09 +0100
Subject: [csw-maintainers] HelpLinker
Message-ID: <ce97543b0907290607n766d18a7qc46f15b0d66bf698@mail.gmail.com>

Hello James,

What is the HelpLinker you run off cron on the buildfarm?

Maciej


From maciej at opencsw.org  Wed Jul 29 15:19:03 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 14:19:03 +0100
Subject: [csw-maintainers] HelpLinker
In-Reply-To: <20090729.13132500.3027483693@gyor.oxdrove.co.uk>
References: <ce97543b0907290607n766d18a7qc46f15b0d66bf698@mail.gmail.com>
	<20090729.13132500.3027483693@gyor.oxdrove.co.uk>
Message-ID: <ce97543b0907290619t5b507706nb0591adb22b8ccb7@mail.gmail.com>

On Wed, Jul 29, 2009 at 2:13 PM, James Lee<james at oxdrove.co.uk> wrote:
> On 29/07/09, 14:07:09, Maciej "(Matchek)" Blizinski <maciej at opencsw.org>
> wrote regarding HelpLinker:
>
>> What is the HelpLinker you run off cron on the buildfarm?
>
> It is an OpenOffice.org build utility that makes the help pages.
>
> It's run from "at" not "cron" so the build continues after my
> terminal disconnects (damn these ISPs when can't they keep a link
> up longer than 24 hours, actually it allows me to sys-suspend
> too and not stop a build, anything that stops the terminal).

I see. nohup or screen could also work. I thought it was cron because
it was the cron process that was the parent of the HelpLinker
processes.

Thanks for the answer.

Maciej


From phil at bolthole.com  Wed Jul 29 18:23:04 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Jul 2009 09:23:04 -0700
Subject: [csw-maintainers] comment on open source security
Message-ID: <20090729162304.GA45823@bolthole.com>

btw;
https://www.isc.org/node/474

That's the second DNS exploit, in the last 12 months.

[sarcasm on]
Wow, they must have some really non-competant programmers working on that.
They shouldnt be allowed to work on that any more.
[/sarcasm]


From blizinski at google.com  Mon Jul 20 14:29:21 2009
From: blizinski at google.com (Maciej (Matchek) Blizinski)
Date: Mon, 20 Jul 2009 13:29:21 +0100
Subject: [csw-maintainers] Updating cups to 1.3.11
Message-ID: <ce97543b0907200529m9c2c72cod4f585ee06752f03@mail.gmail.com>

Hi Markus,

Do you mind building cups-1.3.11? I've submitted a version bump and a
small bugfix. Can you please build the package?

Here's the mantis issue:
http://www.opencsw.org/bugtrack/view.php?id=3768

Reason for the request: version 1.3.11 contains a bugfix for the
scheduler code. In 1.3.9 cups-polld gets stuck after some time.

To other maintainers: I've filed a bug with mantis, Markus is marked
as active maintainer, but there is a number of unresolved issues in
mantis. When did you last hear from Markus?

Maciej


From trygvis at inamo.no  Thu Jul 23 18:09:52 2009
From: trygvis at inamo.no (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=)
Date: Thu, 23 Jul 2009 18:09:52 +0200
Subject: [csw-maintainers] The source code of www.opencsw.org
In-Reply-To: <20090723160656.GH53578@bolthole.com>
References: <ce97543b0907230825s1e873fe5i8cd130caa13ee747@mail.gmail.com>	<1248363911-sup-6322@ntdws12.chass.utoronto.ca>	<4A6886DF.30507@opencsw.org>
	<20090723160656.GH53578@bolthole.com>
Message-ID: <4A688B50.1070502@inamo.no>

Philip Brown wrote:
> On Thu, Jul 23, 2009 at 05:50:55PM +0200, Trygve Laugst?l wrote:
>> Ben Walton wrote:
>>> Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 23 11:25:35 -0400 2009:
>>>> source code of www.opencsw.org checked into subversion? If not, what
>>>> do you think of submitting the code there? It would make it easier to
>>>> suggest a change, or submit a patch, and generally make things more
>>>> open.
>>> +1 for making the website code open and public.
>> +1
>>
>> Would be nice to be able to turn the package database into a separate  
>> application, I have clients that might see use something like that.
> 
> Umm... I doubt your clients would really want to use our code.
> Euuu, ick ;-)
> 
> besides, there isnt really much "code" for that. What code would they
> actually use? the
>  "taking apart a package, to analyze and register in a database" bit?

That part would definitely be useful.

> If you're serious about that, I'll consider making my ugly mess something
> that is more suitable for publication.

They would like to see the applications that they have available, but 
also the versions available so some extensions would be needed.

I realize it is all quite easy to write, but it is the kind of code that 
take a while to get to work correctly so a starting point would be useful.

--
Trygve


From blizinski at google.com  Thu Jul 23 23:06:39 2009
From: blizinski at google.com (Maciej (Matchek) Blizinski)
Date: Thu, 23 Jul 2009 22:06:39 +0100
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <4A68CBFD.3000408@wbonnet.net>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
	<4A68CBFD.3000408@wbonnet.net>
Message-ID: <ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>

On Thu, Jul 23, 2009 at 9:45 PM, William Bonnet<william at wbonnet.net> wrote:
> There is also the i386 package :)
>
> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-sparc.pkg.gz
> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-i386.pkg.gz

Your firefox packages should be listed in the catalog file[1], right?
I see my django and pyantlrrt packages, but not firefox. Your package
have been placed around the same time as mine, but mine are there and
yours are not. I smell there's something wrong there.

[1] http://mirror.opencsw.org/opencsw/testing/i386/5.10/catalog

Maciej


From blizinski at google.com  Fri Jul 24 10:28:12 2009
From: blizinski at google.com (Maciej (Matchek) Blizinski)
Date: Fri, 24 Jul 2009 09:28:12 +0100
Subject: [csw-maintainers] [csw-users]  Firefox 3.0.12 is in testing
In-Reply-To: <ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>
References: <4A68C142.6040306@wbonnet.net>
	<ce97543b0907231339l37a495ccnd465b36dbc80c8a4@mail.gmail.com>
	<4A68CBFD.3000408@wbonnet.net>
	<ce97543b0907231406o30a75df0g843f0688aef74d2a@mail.gmail.com>
Message-ID: <ce97543b0907240128ic5b1c73q230ecff15d80d8b9@mail.gmail.com>

On Thu, Jul 23, 2009 at 10:06 PM, Maciej (Matchek)
Blizinski<blizinski at google.com> wrote:
> On Thu, Jul 23, 2009 at 9:45 PM, William Bonnet<william at wbonnet.net> wrote:
>> There is also the i386 package :)
>>
>> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-sparc.pkg.gz
>> http://mirror.opencsw.org/testing/firefox-3.0.12,REV=2009.07.23-SunOS5.8-i386.pkg.gz
>
> Your firefox packages should be listed in the catalog file[1], right?

I pulled it manually from the given URL and used pkgadd. I couldn't
install it with pkg-get or pkgutil.

After I got it installed - it works! I'll keep using it and will come
back with any comments. Thanks, William!

Maciej


From shuttlebox at gmail.com  Wed Jul 29 16:57:02 2009
From: shuttlebox at gmail.com (Peter Bonivart)
Date: Wed, 29 Jul 2009 16:57:02 +0200
Subject: [csw-maintainers] Fwd: [csw-users] stable release
References: <20090728.10023200.2665933905@gyor.oxdrove.co.uk>
Message-ID: <1ED128FF-E617-4F2E-99A3-D8B829ED214A@gmail.com>

Is there a new stable release in the works? It will be discussed at  
the august meeting, would be good to have your take on it first.

Sent from my iPhone

Vidarebefordrat brev:

> Fr?n: James Lee <james at opencsw.org>
> Datum: 28 juli 2009 10.02.32 CEST
> Till: questions and discussions <users at lists.opencsw.org>
> ?mne: Re: [csw-users] stable release
> Svara till: questions and discussions <users at lists.opencsw.org>
>

> On 28/07/09, 10:34:35, Klaus Heinz <khfp.blastwave0704 at gmx.de> wrote
> regarding [csw-users] stable release:
>
>> maybe I missed something but it looks to me as if the stable tree has
>> hardly changed since August 2008 (I checked the "stable" directory at
>> ftp.uni-erlangen.de and ftp.df.lth.se).
>
>> Is this really the latest stable version of OpenCSW packages?
>
> In fact the latest is 2008-03.  The August dates are only an OpenCSW
> renaming.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090729/119363dd/attachment-0002.html>

From shuttlebox at gmail.com  Wed Jul 29 17:01:55 2009
From: shuttlebox at gmail.com (Peter Bonivart)
Date: Wed, 29 Jul 2009 17:01:55 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <1248801545-sup-9991@ntdws12.chass.utoronto.ca>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
Message-ID: <9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>

To have download stats from just one mirror would still be a very good  
indication. To get combined stats from all our mirrors will never  
happen.

Sent from my iPhone

28 jul 2009 kl. 19.22 skrev Ben Walton <bwalton at opencsw.org>:

> Excerpts from Maciej (Matchek) Blizinski's message of Tue Jul 28  
> 13:16:53 -0400 2009:
>
>> Thoughts?
>
> There have been requests for stats data on released packages in the
> past.  That's tough to get since it would require data from every
> mirror.  This should be fairly simple to get though, since testing is
> centralized.  +1 for making this info available somehow...would
> awstats work for this task?


From dam at opencsw.org  Wed Jul 29 19:13:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 19:13:01 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
Message-ID: <31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>

Hi Peter,

Am 29.07.2009 um 17:01 schrieb Peter Bonivart:
> To have download stats from just one mirror would still be a very  
> good indication. To get combined stats from all our mirrors will  
> never happen.

Ok, now everybody calm down. I am working on updating the awstats
package and providing download stats for

   http://mirror.opencsw.org/opencsw/current/
   http://mirror.opencsw.org/opencsw/stable/
   http://mirror.opencsw.org/opencsw/testing/

Michael, can we get stats from your mirror also?


Best regards

   -- Dago


From dam at opencsw.org  Wed Jul 29 20:12:15 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 20:12:15 +0200
Subject: [csw-maintainers] Statistics on ibiblio
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>
Message-ID: <F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>

Hi there,

there are also statistics available for ibiblio:

1. Log in to <https://stats.ibiblio.org>
   User: guest Password: ibiblio
2. Click on www.ibiblio.org.public
3. Click on Pages & Files
4. Click on Directory Drilldown
5. Click on Directory by Files Drilldown
6. Click on /pub
7. Click on /packages
8. Click on /solaris

...enjoy :-)

(Looks like nobody is using Solaris 8/9 and Blastwave any more)


Best regards

  -- Dago



From phil at bolthole.com  Wed Jul 29 20:27:03 2009
From: phil at bolthole.com (Philip Brown)
Date: Wed, 29 Jul 2009 11:27:03 -0700
Subject: [csw-maintainers] Statistics on ibiblio
In-Reply-To: <F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>
	<F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
Message-ID: <20090729182703.GG45823@bolthole.com>

On Wed, Jul 29, 2009 at 08:12:15PM +0200, Dagobert Michelsen wrote:
>
> (Looks like nobody is using Solaris 8/9 and Blastwave any more)

wow. interesting.

more precisely, in the last week, there were no downloads in
current/*/5.8 or 5.9

there were a few in
stable/*/5.8 and 5.9


that's a big change.




From maciej at opencsw.org  Thu Jul 30 00:24:35 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Wed, 29 Jul 2009 23:24:35 +0100
Subject: [csw-maintainers] unixODBC-2.2.14 in testing
Message-ID: <ce97543b0907291524i365c767ag395076140919221a@mail.gmail.com>

The unixODBC package, version 2.2.14 is now in testing.

http://mirror.opencsw.org/testing/unixodbc-2.2.14,REV=2009.07.29-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/unixodbc-2.2.14,REV=2009.07.29-SunOS5.8-i386-CSW.pkg.gz

It's the first version built with GAR, a remake from scratch. I
haven't got access to the original package. This package doesn't
include GUI (QT) support. It keeps the configuration files in
/etc/opt/csw, as opposed to the earlier /opt/csw/etc. Please feel free
to respond with any feedback on the package.

Maciej


From mwatters at opencsw.org  Thu Jul 30 01:15:50 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Wed, 29 Jul 2009 18:15:50 -0500
Subject: [csw-maintainers] dbus in testing -- smf error
Message-ID: <4A70D826.8070905@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I found an error while testing the version of CSWdbus in testing
  dbus 1.2.12,REV=2009.07.15

while trying to start on solaris 10 using SMF it passes the --config= argument
to the command line. this does not allow the smf service to start.  if I edit
the start/stop SMF discription by hand and s/--config=.*$// ( to make it look
like the version in "current") everything works as expected.

- --

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

iEYEARECAAYFAkpw2CYACgkQLrhmsXMSLxfsUACgxULKBWkB+WeqdgiHMSP4LFWu
wv8An1Gei/KV0zVE/M7cV8dj+qXbqCrq
=zj12
-----END PGP SIGNATURE-----


From pfelecan at opencsw.org  Thu Jul 30 09:27:34 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 30 Jul 2009 09:27:34 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com> (Peter
	Bonivart's message of "Wed\, 29 Jul 2009 17\:01\:55 +0200")
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
Message-ID: <x6tz0uljfd.fsf@abhavana.free.fr>

Peter Bonivart <shuttlebox at gmail.com> writes:

> Sent from my iPhone

a 3, 3G or 3GS?

sent from my hair dryer
-- 
Peter


From michael at opencsw.org  Thu Jul 30 09:55:40 2009
From: michael at opencsw.org (Michael Gernoth)
Date: Thu, 30 Jul 2009 09:55:40 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
	<31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
Message-ID: <20090730075540.GA29118@zerfleddert.de>

Hi Dago,

On Wed, Jul 29, 2009 at 07:13:01PM +0200, Dagobert Michelsen wrote:
> Michael, can we get stats from your mirror also?

I'll try it this weekend. We have logs back to the beginning, so this
should be no problem.

Regards,
  Michael


From skayser at opencsw.org  Thu Jul 30 09:57:38 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Thu, 30 Jul 2009 09:57:38 +0200 (CEST)
Subject: [csw-maintainers] dbus in testing -- smf error
In-Reply-To: <4A70D826.8070905@opencsw.org>
References: <4A70D826.8070905@opencsw.org>
Message-ID: <49287.217.227.30.90.1248940658.squirrel@ssl.skayser.de>

Mike Watters wrote:
> I found an error while testing the version of CSWdbus in testing
>   dbus 1.2.12,REV=2009.07.15
>
> while trying to start on solaris 10 using SMF it passes the --config=
> argument
> to the command line. this does not allow the smf service to start.  if I
> edit
> the start/stop SMF discription by hand and s/--config=.*$// ( to make it
> look
> like the version in "current") everything works as expected.

Same here.

  http://lists.opencsw.org/pipermail/maintainers/2009-July/003337.html

Sebastian



From maciej at opencsw.org  Thu Jul 30 10:17:51 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 09:17:51 +0100
Subject: [csw-maintainers] Buildbot and twisted available in testing
Message-ID: <ce97543b0907300117y21e4edbk5856ec73c95abca1@mail.gmail.com>

Buildbot and its dependency, twisted, are available in testing.

>From the homepage: BuildBot is a system to automate the compile/test
cycle required by most software projects to validate code changes. By
automatically rebuilding and testing the tree each time something has
changed, build problems are pinpointed quickly, before other
developers are inconvenienced by the failure.

Direct links to packages:
http://mirror.opencsw.org/testing/buildbot-0.7.11p1,REV=2009.07.29-SunOS5.8-all-CSW.pkg.gz
http://mirror.opencsw.org/testing/twisted-8.2.0,REV=2009.07.29-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/twisted-8.2.0,REV=2009.07.29-SunOS5.8-sparc-CSW.pkg.gz

Buildbot also depends on Zope.

Maciej


From ihsan at opencsw.org  Thu Jul 30 12:12:38 2009
From: ihsan at opencsw.org (Ihsan Dogan)
Date: Thu, 30 Jul 2009 12:12:38 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <x6tz0uljfd.fsf@abhavana.free.fr>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
	<x6tz0uljfd.fsf@abhavana.free.fr>
Message-ID: <4A717216.70301@opencsw.org>

Peter FELECAN wrote:

>> Sent from my iPhone
> 
> a 3, 3G or 3GS?
> 
> sent from my hair dryer

An article about that was in the Swiss press, when the first iPhone was
released. Unfortunately only in German.
http://ihsan.dogan.ch/files/iPhone_vs_Blackberry/iPhone_vs_Blackberry.pdf




Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From bonivart at opencsw.org  Thu Jul 30 12:25:07 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 12:25:07 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com> 
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com> 
	<31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
Message-ID: <625385e30907300325i63deca6dh143cfa03e6d64364@mail.gmail.com>

On Wed, Jul 29, 2009 at 7:13 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> Hi Peter,
>
> Am 29.07.2009 um 17:01 schrieb Peter Bonivart:
>>
>> To have download stats from just one mirror would still be a very good
>> indication. To get combined stats from all our mirrors will never happen.
>
> Ok, now everybody calm down.

I'm calm. :-) Just meant that we shouldn't deny ourselves any stats
because the perfect combined stats are so complicated to get.

Now it seems to be happening and that's great.

-- 
/peter


From bonivart at opencsw.org  Thu Jul 30 12:26:52 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 12:26:52 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <x6tz0uljfd.fsf@abhavana.free.fr>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com> 
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>
	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com> 
	<x6tz0uljfd.fsf@abhavana.free.fr>
Message-ID: <625385e30907300326h6ca5d795g4e2f7b0cc6638773@mail.gmail.com>

On Thu, Jul 30, 2009 at 9:27 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> Peter Bonivart <shuttlebox at gmail.com> writes:
>
>> Sent from my iPhone
>
> a 3, 3G or 3GS?

The 3GS isn't available in Sweden until tomorrow! :-) Didn't notice
that it added a signature, must find that setting, haha.

> sent from my hair dryer

:-)

-- 
/peter


From bonivart at opencsw.org  Thu Jul 30 12:28:59 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 12:28:59 +0200
Subject: [csw-maintainers] Statistics on ibiblio
In-Reply-To: <F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de> 
	<F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
Message-ID: <625385e30907300328u4061083cud8e263a48a9eb4ef@mail.gmail.com>

On Wed, Jul 29, 2009 at 8:12 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> (Looks like nobody is using Solaris 8/9 and Blastwave any more)

Thanks for that link. I don't think BW lists ibiblio as a mirror any
more so their users may have moved to other mirrors.

-- 
/peter


From william at wbonnet.net  Thu Jul 30 12:29:06 2009
From: william at wbonnet.net (William Bonnet)
Date: Thu, 30 Jul 2009 12:29:06 +0200
Subject: [csw-maintainers] Actual downloads of packages from testing
In-Reply-To: <625385e30907300325i63deca6dh143cfa03e6d64364@mail.gmail.com>
References: <ce97543b0907281016l26b1a3dbkc7bd8fa900a59019@mail.gmail.com>
	<1248801545-sup-9991@ntdws12.chass.utoronto.ca>	<9DFF6F16-C311-4B00-868F-7592E3BFDC7C@gmail.com>
	<31BA5CBC-D667-49FC-9ED3-C05D45D32E14@opencsw.org>
	<625385e30907300325i63deca6dh143cfa03e6d64364@mail.gmail.com>
Message-ID: <4A7175F2.1090005@wbonnet.net>

Peter Bonivart a ?crit :
> On Wed, Jul 29, 2009 at 7:13 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
>   
>> Hi Peter,
>>
>> Am 29.07.2009 um 17:01 schrieb Peter Bonivart:
>>     
>>> To have download stats from just one mirror would still be a very good
>>> indication. To get combined stats from all our mirrors will never happen.
>>>       
>> Ok, now everybody calm down.
>>     
>
> I'm calm. :-) Just meant that we shouldn't deny ourselves any stats
> because the perfect combined stats are so complicated to get.
>
> Now it seems to be happening and that's great.
>
>   
What about creating something like debian did with popcon ? That would 
be more accurate than getting stats from only a part of the mirrors.

cheers
W.


From ihsan at dogan.ch  Thu Jul 30 12:48:00 2009
From: ihsan at dogan.ch (Ihsan Dogan)
Date: Thu, 30 Jul 2009 12:48:00 +0200
Subject: [csw-maintainers] comment on open source security
In-Reply-To: <20090729162304.GA45823@bolthole.com>
References: <20090729162304.GA45823@bolthole.com>
Message-ID: <4A717A60.3070305@dogan.ch>

Philip Brown wrote:
> btw;
> https://www.isc.org/node/474
> 
> That's the second DNS exploit, in the last 12 months.
> 
> [sarcasm on]
> Wow, they must have some really non-competant programmers working on that.
> They shouldnt be allowed to work on that any more.
> [/sarcasm]

The fact that Bind 10 is going to be a redesign again, shows that Bind 9
has broken design.

With NSD there is now a good alternative to Bind. ns1.opencsw.org runs
with it.
http://www.opencsw.org/packages/nsd



Ihsan

-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From maciej at opencsw.org  Thu Jul 30 15:31:42 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 14:31:42 +0100
Subject: [csw-maintainers] cswpycompile docs
Message-ID: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>

cswclassutils docs[1] say this about cswpycompile:

------8<------
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 }'
------8<------

I think it should not say 'pycompile', but 'cswpycompile':

------8<------
This class takes care of creating/removing python's normal and
optimized bytecode files. Set all *.py files' class = cswpycompile to
activate. Example:

# use prototype filters to set the class
PROTOTYPE_FILTER = awk '$$3 ~/.*\.py$$/ { $$2 = "cswpycompile" } { print }'
------8<------

Is that right?

Maciej

[1] http://wiki.opencsw.org/cswclassutils-package


From bwalton at opencsw.org  Thu Jul 30 15:37:10 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Thu, 30 Jul 2009 09:37:10 -0400
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
Message-ID: <1248961009-sup-8082@ntdws12.chass.utoronto.ca>

Excerpts from Maciej (Matchek) Blizinski's message of Thu Jul 30 09:31:42 -0400 2009:

> I think it should not say 'pycompile', but 'cswpycompile':

> Is that right?

I believe that is correct.
-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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090730/4815e03c/attachment-0002.asc>

From bonivart at opencsw.org  Thu Jul 30 16:10:13 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 16:10:13 +0200
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
Message-ID: <625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>

On Thu, Jul 30, 2009 at 3:31 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> cswclassutils docs[1] say this about cswpycompile:
>
> I think it should not say 'pycompile', but 'cswpycompile':
>
> Is that right?

Yes, I fixed it this time but please go ahead and fix it yourself next
time. It's a wiki after all. :-)

If you don't have an account on the wiki yet, just apply for one and I
will grant you access.

-- 
/peter


From bonivart at opencsw.org  Thu Jul 30 16:14:35 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 16:14:35 +0200
Subject: [csw-maintainers] /testing BIND 9.6.1-P1 (fix for critical security
	hole)
Message-ID: <625385e30907300714g1b046218m556faf0acc1cebde@mail.gmail.com>

Now in testing...

-- 
/peter


From dam at opencsw.org  Thu Jul 30 16:50:40 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 30 Jul 2009 16:50:40 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
Message-ID: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>

Hi,

I am currently working on a fix for the gtk2-bug #3775:
   "Package should create /opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders"
   http://www.opencsw.org/mantis/view.php?id=3775

What I have done now is setting it up on postinstall like pango.
However, on Solaris 8 it will try to query 64 bit too, which fails.
If you now share this to Solaris 10 you will end up without pixbuf
loaders. Would you consider it to be ok to skip this step on
Solaris 8?

Or could the pixbuffers be queried during package creation statically?
That would also eliminate the ugly (and slow) postinstall.


Best regards

   -- Dago


From maciej at opencsw.org  Thu Jul 30 17:03:27 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 16:03:27 +0100
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
	<625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>
Message-ID: <ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>

On Thu, Jul 30, 2009 at 3:10 PM, Peter Bonivart<bonivart at opencsw.org> wrote:
> On Thu, Jul 30, 2009 at 3:31 PM, Maciej (Matchek)
> Blizinski<maciej at opencsw.org> wrote:
>> cswclassutils docs[1] say this about cswpycompile:
>>
>> I think it should not say 'pycompile', but 'cswpycompile':
>>
>> Is that right?
>
> Yes, I fixed it this time but please go ahead and fix it yourself next
> time. It's a wiki after all. :-)

Two reasons - I wanted to ask first, and I don't have access.
Otherwise I agree and as a rule I fix stuff myself. (And then people
disagree with me. ;-) )

> If you don't have an account on the wiki yet, just apply for one and I
> will grant you access.

I was about to ask you about it the day before yesterday. I've just
set up an account in wiki.opencsw.org. To keep it simple and
consistent, I used the name 'automatthias'. ;-)

Maciej


From pfelecan at opencsw.org  Thu Jul 30 17:42:32 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Thu, 30 Jul 2009 17:42:32 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
In-Reply-To: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org> (Dagobert
	Michelsen's message of "Thu\, 30 Jul 2009 16\:50\:40 +0200")
References: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>
Message-ID: <x6fxcekwif.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> I am currently working on a fix for the gtk2-bug #3775:
>   "Package should create /opt/csw/etc/gtk-2.0/gdk-pixbuf.loaders"
>   http://www.opencsw.org/mantis/view.php?id=3775
>
> What I have done now is setting it up on postinstall like pango.
> However, on Solaris 8 it will try to query 64 bit too, which fails.
> If you now share this to Solaris 10 you will end up without pixbuf
> loaders. Would you consider it to be ok to skip this step on
> Solaris 8?

-1

> Or could the pixbuffers be queried during package creation statically?
> That would also eliminate the ugly (and slow) postinstall.

+1

-- 
Peter


From jeff at cjsa.com  Wed Jul 29 23:11:20 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Wed, 29 Jul 2009 21:11:20 GMT
Subject: [csw-maintainers] New Problems with Berkeley DB
Message-ID: <KnKAuw.28B@cjsa.com>

I had to reboot my Solaris 10 SPARC V250 server and had the following problem
upon restart:

-------------------------------------------------------------------------------
Jul 29 13:58:57 cjsa sm-mta[636]: [ID 801593 mail.crit] NOQUEUE: SYSERR(root):
    Berkeley DB version mismatch: compiled against 4.2.52,
    run-time linked against 4.7.25

[previous message repeated 17 times]

Jul 29 13:59:01 cjsa svc.startd[7]: [ID 748625 daemon.error]
    network/smtp:cswsendmail failed repeatedly: transitioned to
    maintenance (see 'svcs -xv' for details)
-------------------------------------------------------------------------------

1-> svcs -x
svc:/network/smtp:cswsendmail (sendmail SMTP mail transfer agent)
 State: maintenance since Wed Jul 29 13:59:01 2009
Reason: Restarting too quickly.
   See: http://sun.com/msg/SMF-8000-L5
   See: sendmail(1M)
   See: /var/svc/log/network-smtp:cswsendmail.log
   Impact: This service is not running.


So there is still a compatibility problem with the Berkeley DB.  I am
running:

berkeleydb                   4.7.25,REV=2009.07.01
berkeleydb3                  3.3.11,REV=2009.07.22
berkeleydb4                  4.2.52,REV=2009.07.28
berkeleydb44                 4.4.20,REV=2009.07.28

Of cource, this will probably not go out since I do not have sendmail
running. :-(

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040



From phil at bolthole.com  Thu Jul 30 18:02:34 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Jul 2009 09:02:34 -0700
Subject: [csw-maintainers] New Problems with Berkeley DB
In-Reply-To: <KnKAuw.28B@cjsa.com>
References: <KnKAuw.28B@cjsa.com>
Message-ID: <20090730160234.GA6341@bolthole.com>

On Wed, Jul 29, 2009 at 09:11:20PM +0000, Jeffery Small wrote:
> I had to reboot my Solaris 10 SPARC V250 server and had the following problem
> upon restart:
> 
> -------------------------------------------------------------------------------
> Jul 29 13:58:57 cjsa sm-mta[636]: [ID 801593 mail.crit] NOQUEUE: SYSERR(root):
>     Berkeley DB version mismatch: compiled against 4.2.52,
>     run-time linked against 4.7.25

wow. sendmail is really dumb to do this. :-/

> Of cource, this will probably not go out since I do not have sendmail
> running. :-(

it worked though :-)
maybe it's just a whine, not an actual "error"?



From phil at bolthole.com  Thu Jul 30 18:12:21 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Jul 2009 09:12:21 -0700
Subject: [csw-maintainers] Statistics on ibiblio
In-Reply-To: <625385e30907300328u4061083cud8e263a48a9eb4ef@mail.gmail.com>
References: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>
	<F544C88B-7EF9-4E54-8253-2C93A9C7BF36@opencsw.org>
	<625385e30907300328u4061083cud8e263a48a9eb4ef@mail.gmail.com>
Message-ID: <20090730161221.GB6341@bolthole.com>

On Thu, Jul 30, 2009 at 12:28:59PM +0200, Peter Bonivart wrote:
> On Wed, Jul 29, 2009 at 8:12 PM, Dagobert Michelsen<dam at opencsw.org> wrote:
> > (Looks like nobody is using Solaris 8/9 and Blastwave any more)
> 
> Thanks for that link. I don't think BW lists ibiblio as a mirror any
> more so their users may have moved to other mirrors.

they have removed their archives, actually. its just a placeholder now,
seems like.


From bonivart at opencsw.org  Thu Jul 30 19:56:12 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 19:56:12 +0200
Subject: [csw-maintainers] comment on open source security
In-Reply-To: <4A717A60.3070305@dogan.ch>
References: <20090729162304.GA45823@bolthole.com> <4A717A60.3070305@dogan.ch>
Message-ID: <625385e30907301056u5155125bs425a59049cc4d831@mail.gmail.com>

On Thu, Jul 30, 2009 at 12:48 PM, Ihsan Dogan<ihsan at dogan.ch> wrote:
> The fact that Bind 10 is going to be a redesign again, shows that Bind 9
> has broken design.

On the other hand, Bind 9 was designed in 1998. In computer terms
that's a looong time ago.

Nevertheless, making Bind 10 modular will be appreciated. :-)

-- 
/peter


From bonivart at opencsw.org  Thu Jul 30 20:05:16 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Thu, 30 Jul 2009 20:05:16 +0200
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com> 
	<625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com> 
	<ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>
Message-ID: <625385e30907301105q73f6c2ecl2cafc5d030bfbb0b@mail.gmail.com>

On Thu, Jul 30, 2009 at 5:03 PM, Maciej (Matchek)
Blizinski<maciej at opencsw.org> wrote:
> I was about to ask you about it the day before yesterday. I've just
> set up an account in wiki.opencsw.org. To keep it simple and
> consistent, I used the name 'automatthias'. ;-)

I have sent you an invite now, if you accept it you're then a member
of the OpenCSW wiki. :-)

-- 
/peter


From mwatters at opencsw.org  Thu Jul 30 20:40:23 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 30 Jul 2009 13:40:23 -0500
Subject: [csw-maintainers] libcairo in testing RPATH is broken
Message-ID: <4A71E917.1010205@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

here is the excerpt from dump -Lv
this in turn breaks everything that depends on libcairo as it can not find

$ ldd /opt/csw/bin/pidgin # for example
...
        libxcb-render-util.so.0 =>       (file not found)
        libxcb-render.so.0 =>    (file not found)
...

$ dump -Lv /opt/csw/lib/sparcv8/libcairo.so.2

/opt/csw/lib/sparcv8/libcairo.so.2:

  **** DYNAMIC SECTION INFORMATION ****
...
[15]    SONAME          libcairo.so.2
[16]    RUNPATH         /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib
[17]    RPATH           /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib

...

- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpx6RcACgkQLrhmsXMSLxdmLgCfWl89CyMsfnsjQjSDSlhHB4QM
JfMAni48JaZLeYrFAcbovxXCoc92A4rx
=lWnN
-----END PGP SIGNATURE-----


From maciej at opencsw.org  Thu Jul 30 20:58:01 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Thu, 30 Jul 2009 19:58:01 +0100
Subject: [csw-maintainers] cswpycompile docs
In-Reply-To: <625385e30907301105q73f6c2ecl2cafc5d030bfbb0b@mail.gmail.com>
References: <ce97543b0907300631v6d5b6df4j21b800ec23fe43da@mail.gmail.com>
	<625385e30907300710x1582382bycaa6b841e7601e6b@mail.gmail.com>
	<ce97543b0907300803l27277b85h931cac717c98c784@mail.gmail.com>
	<625385e30907301105q73f6c2ecl2cafc5d030bfbb0b@mail.gmail.com>
Message-ID: <ce97543b0907301158p75312624vbe93f49cb48e0131@mail.gmail.com>

On Thu, Jul 30, 2009 at 7:05 PM, Peter Bonivart<bonivart at opencsw.org> wrote:
> On Thu, Jul 30, 2009 at 5:03 PM, Maciej (Matchek)
> Blizinski<maciej at opencsw.org> wrote:
>> I was about to ask you about it the day before yesterday. I've just
>> set up an account in wiki.opencsw.org. To keep it simple and
>> consistent, I used the name 'automatthias'. ;-)
>
> I have sent you an invite now, if you accept it you're then a member
> of the OpenCSW wiki. :-)

Accepted. Thanks!


From dam at opencsw.org  Thu Jul 30 21:53:55 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 30 Jul 2009 21:53:55 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
In-Reply-To: <x6fxcekwif.fsf@abhavana.free.fr>
References: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>
	<x6fxcekwif.fsf@abhavana.free.fr>
Message-ID: <D547DC8C-5876-459A-A66A-C200894D66C1@opencsw.org>

Hi Peter,

Am 30.07.2009 um 17:42 schrieb Peter FELECAN:
>> Or could the pixbuffers be queried during package creation  
>> statically?
>> That would also eliminate the ugly (and slow) postinstall.
>
> +1

I would prefer this, but it is allowed? Or are there other modules  
that are
not present at build time?


Best regards

   -- Dago


From dam at opencsw.org  Thu Jul 30 22:05:51 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Thu, 30 Jul 2009 22:05:51 +0200
Subject: [csw-maintainers] libcairo in testing RPATH is broken
In-Reply-To: <4A71E917.1010205@opencsw.org>
References: <4A71E917.1010205@opencsw.org>
Message-ID: <EBC16C99-0484-4629-80A0-FFCF7F6CD602@opencsw.org>

Hi John,

Am 30.07.2009 um 20:40 schrieb Mike Watters:
> here is the excerpt from dump -Lv
> this in turn breaks everything that depends on libcairo as it can  
> not find
>
> $ ldd /opt/csw/bin/pidgin # for example
> ...
>        libxcb-render-util.so.0 =>       (file not found)
>        libxcb-render.so.0 =>    (file not found)
> ...
>
> $ dump -Lv /opt/csw/lib/sparcv8/libcairo.so.2
>
> /opt/csw/lib/sparcv8/libcairo.so.2:
>
>  **** DYNAMIC SECTION INFORMATION ****
> ...
> [15]    SONAME          libcairo.so.2
> [16]    RUNPATH         /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib
> [17]    RPATH           /opt/csw/lib/SALIST:RIGIN/../lib:/opt/csw/lib

This is very strange.
John: How have you build that package? With GAR? If yes, what revision  
of GAR
itself were you using? I can't remember putting $ORIGIN in anywhere.


Best regards

   -- Dago


From jeff at cjsa.com  Fri Jul 31 01:45:38 2009
From: jeff at cjsa.com (Jeffery Small)
Date: Thu, 30 Jul 2009 23:45:38 GMT
Subject: [csw-maintainers] New Problems with Berkeley DB
References: <KnKAuw.28B@cjsa.com> <20090730160234.GA6341@bolthole.com>
Message-ID: <KnMCo2.4pr@cjsa.com>

Philip Brown <phil at bolthole.com> writes:

>On Wed, Jul 29, 2009 at 09:11:20PM +0000, Jeffery Small wrote:
>> I had to reboot my Solaris 10 SPARC V250 server and had the following problem
>> upon restart:
>> 
>> -------------------------------------------------------------------------------
>> Jul 29 13:58:57 cjsa sm-mta[636]: [ID 801593 mail.crit] NOQUEUE: SYSERR(root):
>>     Berkeley DB version mismatch: compiled against 4.2.52,
>>     run-time linked against 4.7.25

>wow. sendmail is really dumb to do this. :-/

>> Of cource, this will probably not go out since I do not have sendmail
>> running. :-(

>it worked though :-)
>maybe it's just a whine, not an actual "error"?

No, sendmail was out of commission.  I'm back up due to some great help
from Dagobert Michelsen.  This message probably only went out after I got
sendmail back online.

I currently only have the following packages:

berkeleydb                   4.7.25,REV=2009.07.01		Current
berkeleydb3                  3.3.11,REV=2009.07.22		Current
berkeleydb4           4.2.52,REV=2005.04.28_rev=p4		From old archive
berkeleydb44                 4.4.20,REV=2009.07.28		Current

And sendmail became happy once again when the old berkeleydb4 was installed.


Below, I'm now following up to an email I received from Dago.

I wrote:

> And when sendmail complained about the runtime version being 4.7.25,
> how does that relate to anything I could see regarding the installed
> packages?  Upon further examination, I see that the berkeleydb_devel
> package is listed as rev 4.7.25,REV=2009.07.01, but I don't have this
> package installed on my system.

Dago wrote:

> You must have as the new bdb4x-packages depend on it.

Well, berkeleydb44_devel only depends upon berkeleydb44, and
berkeleydb_devel presumably depends upon berkeleydb*, but not the other way
around.  I suspect That Dago meant that the berkeleydb{3,4,44} packages
depend upon berkeleydb and not the *_devel packages.

* [NOTE: From the CSW current package list webpage, if you click on
  berkeleydb or berkeleydb_devel you get and error stating: "Err: cannot find
  maintainer for <package> in mantis", and no details are shown!]

Dago wrote:

> I proposed a versioned current, where you can point your pkg*-tool to a
> specific date and which does not change like opencsw/current-20090730
> However, neither me (as I proposed it) nor another one has picked it up
> yet.  Maybe this is an opportunity to bring this ahead.  This is AFAIK a
> better solution than putting magic in pkg*.

This actually sounds like a good idea.  If you upgrade once a week and
serious new problems develop, you can point pkg-get or pkg-util to a date
for the previous week and bring things back into sync.  This would work
so long as all package dependencies could be updated/rewound reliably;
something to which I have not yet given much thought.

Regards,
-- 
Jeff

C. Jeffery Small           CJSA LLC                       206-232-3338
jeff at cjsa.com              7000 E Mercer Way, Mercer Island, WA  98040



From phil at bolthole.com  Fri Jul 31 02:14:59 2009
From: phil at bolthole.com (Philip Brown)
Date: Thu, 30 Jul 2009 17:14:59 -0700
Subject: [csw-maintainers] call for new sendmail maintainer?
Message-ID: <20090731001459.GB47846@bolthole.com>

So.... anyone wanna volunteer to take over sendmail and do a quick
repackaging?

Since its being picky about the berkeleydb upgrade, we need a repackaging
done asap, and we do not have an active maintainer for it any more.



From mwatters at opencsw.org  Fri Jul 31 04:24:17 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Thu, 30 Jul 2009 21:24:17 -0500
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <20090731001459.GB47846@bolthole.com>
References: <20090731001459.GB47846@bolthole.com>
Message-ID: <4A7255D1.3020000@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Philip Brown wrote:
> So.... anyone wanna volunteer to take over sendmail and do a quick
> repackaging?
> 
> Since its being picky about the berkeleydb upgrade, we need a repackaging
> done asap, and we do not have an active maintainer for it any more.
> 
> _______________________________________________
> maintainers mailing list
> maintainers at lists.opencsw.org
> https://lists.opencsw.org/mailman/listinfo/maintainers

Benny and I are tag teaming it to get it out quickly...
He got the bulk of it done, I have the last of the install pieces
done.

I was going through the old version and Alex had created a custom class to
handle the fact that sendmail must be compiled separately on each OS.

I have to modify his class to now include at least solaris 10 and write a
custom merge script for gar.  I hope to have this done before I turn in for the
night.



- --
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/

iEYEARECAAYFAkpyVdEACgkQLrhmsXMSLxdkqgCeOP341Mezgf6tsiqH6eC+JAZ2
+VwAn33HPpvDLCO2AFyG6MOS2LR9zppq
=RZGG
-----END PGP SIGNATURE-----


From schwindt at dfki.uni-kl.de  Fri Jul 31 09:16:04 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 31 Jul 2009 09:16:04 +0200
Subject: [csw-maintainers] Updated libiconv in testing,
 testing  appreciated
In-Reply-To: Your message of "Fri, 24 Jul 2009 23:03:28 +0200."
	<4A6A21A0.3060006@opencsw.org> 
Message-ID: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>

[...]
> This should obsolete related hacks required in other packages. I have it
> running on a couple of non-GUI test boxes without hitches, but as it is
> a to-be-adopted and fundamental package, I would appreciate a bit more
> testing/feedback/review by some of you guys.


I will give your version a try, as I always did have a patched version of iconv
around, and am using a self compiled version of 1.13.1 for quite some time now.
I can state that this version generally functions well, troublemaker like 
subversion works perfect with it. 

I'll let you know how things worked out




From skayser at opencsw.org  Fri Jul 31 09:22:43 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 09:22:43 +0200 (CEST)
Subject: [csw-maintainers] Updated libiconv in testing,
 testing  appreciated
In-Reply-To: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>
References: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>
Message-ID: <49274.217.227.42.13.1249024963.squirrel@ssl.skayser.de>

Hi Nicolai,

Nicolai Schwindt wrote:
> [...]
>> This should obsolete related hacks required in other packages. I have it
>> running on a couple of non-GUI test boxes without hitches, but as it is
>> a to-be-adopted and fundamental package, I would appreciate a bit more
>> testing/feedback/review by some of you guys.
>
>
> I will give your version a try, as I always did have a patched version of
> iconv
> around, and am using a self compiled version of 1.13.1 for quite some time
> now.
> I can state that this version generally functions well, troublemaker like
> subversion works perfect with it.
>
> I'll let you know how things worked out

cool, thanks for the support. Note though: I recently pulled the iconv
packages from testing, because they were ones, aimed at the CSWiconv ->
CSWlibiconv package name change. With various bits and pieces of our stack
currently being under heavy change (gtk, pango, bdb) I would like to
postpone such a (purely cosmetical) change.

I will put new iconv packages with the current naming scheme into testing
later this day. Will keep you updated.

Sebastian



From skayser at opencsw.org  Fri Jul 31 09:25:41 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 09:25:41 +0200 (CEST)
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <4A7255D1.3020000@opencsw.org>
References: <20090731001459.GB47846@bolthole.com>
	<4A7255D1.3020000@opencsw.org>
Message-ID: <49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>

Mike Watters wrote:
> Philip Brown wrote:
>> So.... anyone wanna volunteer to take over sendmail and do a quick
>> repackaging?
>>
>> Since its being picky about the berkeleydb upgrade, we need a
>> repackaging
>> done asap, and we do not have an active maintainer for it any more.
>>
> Benny and I are tag teaming it to get it out quickly...
> He got the bulk of it done, I have the last of the install pieces
> done.
>
> I was going through the old version and Alex had created a custom class to
> handle the fact that sendmail must be compiled separately on each OS.
>
> I have to modify his class to now include at least solaris 10 and write a
> custom merge script for gar.  I hope to have this done before I turn in
> for the
> night.

Thinking in terms of simplicity: Wouldn't it also be possible to just wrap
up a package each on build{8,9,10}{s,x} to have a specific package for
each Solaris release, instead of having a potentially complicated GAR
Makefile plus postinstall magic (which might both be subject to errors)?

Just an idea, don't let your progress be slowed down by it if you think
otherwise or if there are reasons against it (no need for discussion from
my side).

Sebastian



From pfelecan at opencsw.org  Fri Jul 31 09:25:49 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 09:25:49 +0200
Subject: [csw-maintainers] gtk2 issues on x86 and 32/64 bit issues
In-Reply-To: <D547DC8C-5876-459A-A66A-C200894D66C1@opencsw.org> (Dagobert
	Michelsen's message of "Thu\, 30 Jul 2009 21\:53\:55 +0200")
References: <C19C89C3-449D-4BE9-A150-FD976BCD0391@opencsw.org>
	<x6fxcekwif.fsf@abhavana.free.fr>
	<D547DC8C-5876-459A-A66A-C200894D66C1@opencsw.org>
Message-ID: <x6y6q5joua.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 30.07.2009 um 17:42 schrieb Peter FELECAN:
>>> Or could the pixbuffers be queried during package creation
>>> statically?
>>> That would also eliminate the ugly (and slow) postinstall.
>>
>> +1
>
> I would prefer this, but it is allowed? Or are there other modules
> that are
> not present at build time?

I don't know (never meet the situation). However, I think that, at least
for post addition of modules is the responsibility of the provider of
the module to refresh the pixbuffers; these seems to be similar in
behaviour with fontconfig, don't you think? There are other instances of
it but I don't remember which.

BTW, what is Debian doing for this? It's not such a bad thing to see
what our friends are doing...

HTH
-- 
Peter


From maciej at opencsw.org  Fri Jul 31 10:02:24 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 09:02:24 +0100
Subject: [csw-maintainers] Warnings from GAR
Message-ID: <ce97543b0907310102l5475c4f2kbf70e985428e681e@mail.gmail.com>

When I'm building with buildbot, I'm often getting warnings from GAR:

 	==> Trying file//files/CSWap2prefork.gspec
gmake[2]: Entering directory `/home/buildbot/buildbot-slave/full/build/trunk'
gar/gar.mk:82: warning: overriding commands for target `work/build-global'
gar/gar.pkg.mk:532: warning: ignoring old commands for target
`work/build-global'

I don't see them when building by hand, but I used to see them before.
Dago, do you know why are those warnings there?

Maciej


From maciej at opencsw.org  Fri Jul 31 10:39:53 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 09:39:53 +0100
Subject: [csw-maintainers] Buildbot
Message-ID: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>

Hello maintainers,

I've written a draft of a design to deploy a continuous build system
on the buildfarm, using buildbot. The draft can be found here:
http://wiki.opencsw.org/buildbot

I have a proof of concept running on my workstation at work, with an
IRC interface 'cswbb' present on the #opencsw channel on Freenode. A
couple points based on questions from Dago:

- buildbot can handle parallel builds on the buildfarm, by running
different builds in different subdirectories
- buildbot automatically picks up changes from the repository, there's
no hand-scheduling needed
- buildbot can report the status via http, email and irc
- it runs as a headless daemon

More details, including a deployment plan is in the design draft.

Thoughts?

Maciej


From pfelecan at opencsw.org  Fri Jul 31 11:00:45 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 11:00:45 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 09\:39\:53 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
Message-ID: <x6ljm5jkg2.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> - buildbot can handle parallel builds on the buildfarm, by running
> different builds in different subdirectories

This should be done on a different system than that used for day to day
work. I don't want to be penalised by gcc, OpenOffice and other monsters
being built the same time that I wish to build a new package.

> - buildbot automatically picks up changes from the repository, there's
> no hand-scheduling needed

This is true only for "garified" packages. Is there a plug-in for other
kind of build?

> - buildbot can report the status via http, email and irc

This is good and can be useful for any build. It's a missing feature on
the current build farms.

> Thoughts?

See above.

-- 
Peter


From dam at opencsw.org  Fri Jul 31 11:49:34 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 31 Jul 2009 11:49:34 +0200
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>
References: <20090731001459.GB47846@bolthole.com>
	<4A7255D1.3020000@opencsw.org>
	<49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>
Message-ID: <BA565FB7-89E7-448F-8112-46C110AF3AC3@opencsw.org>

Hi,

Am 31.07.2009 um 09:25 schrieb Sebastian Kayser:
> Thinking in terms of simplicity: Wouldn't it also be possible to  
> just wrap
> up a package each on build{8,9,10}{s,x} to have a specific package for
> each Solaris release, instead of having a potentially complicated GAR
> Makefile plus postinstall magic (which might both be subject to  
> errors)?

IMHO this is practicable, but is completely against our policy of
central NFS mounted server for all releases. However, running server
services from NFS is also not good design, so I guess it is ok
to make different packages here.


Best regards

   -- Dago


From maciej at opencsw.org  Fri Jul 31 12:25:43 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 11:25:43 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6ljm5jkg2.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>

On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> - buildbot can handle parallel builds on the buildfarm, by running
>> different builds in different subdirectories
>
> This should be done on a different system than that used for day to day
> work. I don't want to be penalised by gcc, OpenOffice and other monsters
> being built the same time that I wish to build a new package.

I see two potential solutions:

1. Provide separate hardware for the buildbot
2. Avoid building certain blacklisted packages. Optional: schedule
periodic or nightly builds for those packages.

>> - buildbot automatically picks up changes from the repository, there's
>> no hand-scheduling needed
>
> This is true only for "garified" packages. Is there a plug-in for other
> kind of build?

Can you come up with an example of a different build? Where would the
source code come from?

There's one more issue: multi-ISA packages. I have a vague idea how to
implement them, but I don't have the proper environment at hand to
actually do that.

Maciej


From pfelecan at opencsw.org  Fri Jul 31 12:31:57 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 12:31:57 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 11\:25\:43 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
Message-ID: <x6hbwtjg82.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>
>>> - buildbot can handle parallel builds on the buildfarm, by running
>>> different builds in different subdirectories
>>
>> This should be done on a different system than that used for day to day
>> work. I don't want to be penalised by gcc, OpenOffice and other monsters
>> being built the same time that I wish to build a new package.
>
> I see two potential solutions:
>
> 1. Provide separate hardware for the buildbot

+1 (I know, I'm not paying the bill) and please, not a zone or a VM on
   the same hardware...

> 2. Avoid building certain blacklisted packages. Optional: schedule
> periodic or nightly builds for those packages.

-1 as it defeats the purpose; in addition, there is no night for an
   international community as ours...

>>> - buildbot automatically picks up changes from the repository, there's
>>> no hand-scheduling needed
>>
>> This is true only for "garified" packages. Is there a plug-in for other
>> kind of build?
>
> Can you come up with an example of a different build? Where would the
> source code come from?

All my packages are outside gar. The source code comes from my home
directory...

-- 
Peter


From maciej at opencsw.org  Fri Jul 31 12:39:06 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 11:39:06 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6hbwtjg82.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>

On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>
>>>> - buildbot can handle parallel builds on the buildfarm, by running
>>>> different builds in different subdirectories
>>>
>>> This should be done on a different system than that used for day to day
>>> work. I don't want to be penalised by gcc, OpenOffice and other monsters
>>> being built the same time that I wish to build a new package.
>>
>> I see two potential solutions:
>>
>> 1. Provide separate hardware for the buildbot
>
> +1 (I know, I'm not paying the bill) and please, not a zone or a VM on
> ? the same hardware...
>
>> 2. Avoid building certain blacklisted packages. Optional: schedule
>> periodic or nightly builds for those packages.
>
> -1 as it defeats the purpose; in addition, there is no night for an
> ? international community as ours...

Automatic builds for some packages are better than automatic builds
for no packages, so it doesn't defeat the purpose completely. Also, we
could do actual measurements: how big is the impact on the
performance?

>>>> - buildbot automatically picks up changes from the repository, there's
>>>> no hand-scheduling needed
>>>
>>> This is true only for "garified" packages. Is there a plug-in for other
>>> kind of build?
>>
>> Can you come up with an example of a different build? Where would the
>> source code come from?
>
> All my packages are outside gar. The source code comes from my home
> directory...

How would you make your source code available to buildbot?

Maciej


From pfelecan at opencsw.org  Fri Jul 31 12:42:44 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 12:42:44 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 11\:39\:06 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
Message-ID: <x6d47hjfq3.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>
>>> On Fri, Jul 31, 2009 at 10:00 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>>
>>>>> - buildbot can handle parallel builds on the buildfarm, by running
>>>>> different builds in different subdirectories
>>>>
>>>> This should be done on a different system than that used for day to day
>>>> work. I don't want to be penalised by gcc, OpenOffice and other monsters
>>>> being built the same time that I wish to build a new package.
>>>
>>> I see two potential solutions:
>>>
>>> 1. Provide separate hardware for the buildbot
>>
>> +1 (I know, I'm not paying the bill) and please, not a zone or a VM on
>> ? the same hardware...
>>
>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>> periodic or nightly builds for those packages.
>>
>> -1 as it defeats the purpose; in addition, there is no night for an
>> ? international community as ours...
>
> Automatic builds for some packages are better than automatic builds
> for no packages, so it doesn't defeat the purpose completely. Also, we
> could do actual measurements: how big is the impact on the
> performance?

Maybe. But not on the effective build stack, please.

>
>>>>> - buildbot automatically picks up changes from the repository, there's
>>>>> no hand-scheduling needed
>>>>
>>>> This is true only for "garified" packages. Is there a plug-in for other
>>>> kind of build?
>>>
>>> Can you come up with an example of a different build? Where would the
>>> source code come from?
>>
>> All my packages are outside gar. The source code comes from my home
>> directory...
>
> How would you make your source code available to buildbot?

By pointing it to my home directory on the build stack...
-- 
Peter


From skayser at opencsw.org  Fri Jul 31 12:54:47 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 12:54:47 +0200
Subject: [csw-maintainers] Updated libiconv in testing,
	testing  appreciated
In-Reply-To: <49274.217.227.42.13.1249024963.squirrel@ssl.skayser.de>
References: <200907310716.n6V7G4rZ012211@dfki.uni-kl.de>
	<49274.217.227.42.13.1249024963.squirrel@ssl.skayser.de>
Message-ID: <4A72CD77.9090209@opencsw.org>

Sebastian Kayser wrote:
> Nicolai Schwindt wrote:
>> [...]
>>> This should obsolete related hacks required in other packages. I have it
>>> running on a couple of non-GUI test boxes without hitches, but as it is
>>> a to-be-adopted and fundamental package, I would appreciate a bit more
>>> testing/feedback/review by some of you guys.
>>
>> I will give your version a try, as I always did have a patched version of
>> iconv
>> around, and am using a self compiled version of 1.13.1 for quite some time
>> now.
>> I can state that this version generally functions well, troublemaker like
>> subversion works perfect with it.
>>
>> I'll let you know how things worked out
> 
> cool, thanks for the support. Note though: I recently pulled the iconv
> packages from testing, because they were ones, aimed at the CSWiconv ->
> CSWlibiconv package name change. With various bits and pieces of our stack
> currently being under heavy change (gtk, pango, bdb) I would like to
> postpone such a (purely cosmetical) change.
> 
> I will put new iconv packages with the current naming scheme into testing
> later this day. Will keep you updated.

Ok, please give the packages in testing a try.

http://mirror.opencsw.org/testing/libiconv-1.13.1,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/libiconv-1.13.1,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz

Modeled after the ones in current, i.e. no spearate devel package, no
pkg/swname changes. Should be least intrusive. File clash on
/opt/csw/lib/{sparcv9,amd64}/charset.alias with gnugpg is known and
already filed [1]. Shouldn't do any harm though, the files are identical.

Sebastian

[1] http://opencsw.org/bugtrack/view.php?id=3755


From maciej at opencsw.org  Fri Jul 31 13:15:42 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 12:15:42 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6d47hjfq3.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	<x6d47hjfq3.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>

On Fri, Jul 31, 2009 at 11:42 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>
>>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>>> periodic or nightly builds for those packages.
>>>
>>> -1 as it defeats the purpose; in addition, there is no night for an
>>> ? international community as ours...
>>
>> Automatic builds for some packages are better than automatic builds
>> for no packages, so it doesn't defeat the purpose completely. Also, we
>> could do actual measurements: how big is the impact on the
>> performance?
>
> Maybe. But not on the effective build stack, please.

Where are large packages currently built? I thought that all packages
must be built on the buildfarm. Next time there's an on-going
compilation of a large package, we can do the measurements.

>>>>>> - buildbot automatically picks up changes from the repository, there's
>>>>>> no hand-scheduling needed
>>>>>
>>>>> This is true only for "garified" packages. Is there a plug-in for other
>>>>> kind of build?
>>>>
>>>> Can you come up with an example of a different build? Where would the
>>>> source code come from?
>>>
>>> All my packages are outside gar. The source code comes from my home
>>> directory...
>>
>> How would you make your source code available to buildbot?
>
> By pointing it to my home directory on the build stack...

Buildbot has plugins for Subversion, Perforce, Arch, Bazaar, Darcs,
Mercurial and Git. Do you use any of these for your source code?

Buildbot needs essentially three things:

1. Build master needs a way to learn about an update to specific
project or branch
2. Build slaves need a way to obtain the source code
3. Build slaves need instructions to build the package

Number 1 could be done by... I don't know... making md5 sums of
specific files in your home directory? Number 2 would be a file copy
operation. Number 3 would be provided by you. Do your builds have a
uniform build procedure?

The GAR packages will be the main focus now, since they comprise about
91% builds (that I know of, that is, ~1083 out of ~1186). Once we're
there, we can extend support to other build sources. Of course, it
would make it easier to implement if a code repository was available
to buildbot.

Maciej


From ihsan at dogan.ch  Fri Jul 31 13:18:58 2009
From: ihsan at dogan.ch (Ihsan Dogan)
Date: Fri, 31 Jul 2009 13:18:58 +0200
Subject: [csw-maintainers] Apache 2.2.12 in testing
Message-ID: <4A72D322.9060906@dogan.ch>

Hello,

I've updated Apache to 2.2.12 and it's available on testing now. Apache
is now linked against Berkeley DB 4.7.

All architectures:
http://mirror.opencsw.org/testing/apache2-2.2.12,REV=2009.07.31-SunOS5.8-all-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2_manual-2.2.12,REV=2009.07.31-SunOS5.8-all-CSW.pkg.gz

sparc:
http://mirror.opencsw.org/testing/ap2_prefork-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_suexec-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_worker-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2_devel-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2c-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2rt-2.2.12,REV=2009.07.31-SunOS5.8-sparc-CSW.pkg.gz

i386:
http://mirror.opencsw.org/testing/ap2_prefork-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_suexec-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/ap2_worker-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2_devel-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2c-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz
http://mirror.opencsw.org/testing/apache2rt-2.2.12,REV=2009.07.31-SunOS5.8-i386-CSW.pkg.gz

Feedback is always welcome and appreciated.



Ihsan
-- 
ihsan at dogan.ch		http://blog.dogan.ch/


From dam at opencsw.org  Fri Jul 31 13:19:46 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 31 Jul 2009 13:19:46 +0200
Subject: [csw-maintainers] Warnings from GAR
In-Reply-To: <ce97543b0907310102l5475c4f2kbf70e985428e681e@mail.gmail.com>
References: <ce97543b0907310102l5475c4f2kbf70e985428e681e@mail.gmail.com>
Message-ID: <602A5446-53BC-4DC1-83D4-60171DE413A8@opencsw.org>

Hi Maciej,

Am 31.07.2009 um 10:02 schrieb Maciej (Matchek) Blizinski:
> When I'm building with buildbot, I'm often getting warnings from GAR:
>
> 	==> Trying file//files/CSWap2prefork.gspec
> gmake[2]: Entering directory `/home/buildbot/buildbot-slave/full/ 
> build/trunk'
> gar/gar.mk:82: warning: overriding commands for target `work/build- 
> global'
> gar/gar.pkg.mk:532: warning: ignoring old commands for target
> `work/build-global'
>
> I don't see them when building by hand, but I used to see them before.
> Dago, do you know why are those warnings there?

This is because you didn't specify where to put the packages in
your ~/.garrc. For me this is

SPKG_EXPORT     = /home/dam/staging/build-$(shell date '+%d.%b.%Y')

Best regards

   -- Dago



From dam at opencsw.org  Fri Jul 31 13:25:01 2009
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 31 Jul 2009 13:25:01 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6hbwtjg82.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
Message-ID: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>

Hi Peter,

Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
> All my packages are outside gar. The source code comes from my home
> directory...

- Is there a specific reason why you are not using GAR? Without GAR it  
is
   hard to do teams, takeover, knowledge sharing, consistent packages,  
etc.
- If you are not using GAR, please also check in the source to
     mgar/pkg/<pkg>/trunk/legacy/
   so it is properly archived. Missing build descriptions on maintainers
   leaving is a hard thing to cope with. IMHO it should be policy that
   no package is published without the recipe committed to the  
repository.


Best regards

   -- Dago


From bwalton at opencsw.org  Fri Jul 31 13:43:59 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 07:43:59 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
Message-ID: <1249040535-sup-2622@ntdws12.chass.utoronto.ca>

Excerpts from Dagobert Michelsen's message of Fri Jul 31 07:25:01 -0400 2009:

> - If you are not using GAR, please also check in the source to
>   mgar/pkg/<pkg>/trunk/legacy/ so it is properly archived. Missing
>   build descriptions on maintainers leaving is a hard thing to cope
>   with. IMHO it should be policy that no package is published
>   without the recipe committed to the repository.

+1 for this.  If you're still a GAR-o-phobe, at least commit the
manual build scripts, patches, instructions etc so that knowledge
doesn't get lost...[The svn-o-phobe's may not like this. <grin>]

Is this something that could be put on the Oslo agenda?

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/0932e918/attachment-0002.asc>

From pfelecan at opencsw.org  Fri Jul 31 16:09:31 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 16:09:31 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 12\:15\:42 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	<x6d47hjfq3.fsf@abhavana.free.fr>
	<ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>
Message-ID: <x68wi5j65g.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> On Fri, Jul 31, 2009 at 11:42 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>
>>> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>>
>>>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>>>> periodic or nightly builds for those packages.
>>>>
>>>> -1 as it defeats the purpose; in addition, there is no night for an
>>>> ? international community as ours...
>>>
>>> Automatic builds for some packages are better than automatic builds
>>> for no packages, so it doesn't defeat the purpose completely. Also, we
>>> could do actual measurements: how big is the impact on the
>>> performance?
>>
>> Maybe. But not on the effective build stack, please.
>
> Where are large packages currently built? I thought that all packages
> must be built on the buildfarm. Next time there's an on-going
> compilation of a large package, we can do the measurements.

Some weeks ago we had I/O issues when gcc3, OpenOffice and Firefox where
built in the same period. What I'm saying, is *automatic build must be
done on a different hardware stack*

>>>>>>> - buildbot automatically picks up changes from the repository, there's
>>>>>>> no hand-scheduling needed
>>>>>>
>>>>>> This is true only for "garified" packages. Is there a plug-in for other
>>>>>> kind of build?
>>>>>
>>>>> Can you come up with an example of a different build? Where would the
>>>>> source code come from?
>>>>
>>>> All my packages are outside gar. The source code comes from my home
>>>> directory...
>>>
>>> How would you make your source code available to buildbot?
>>
>> By pointing it to my home directory on the build stack...
>
> Buildbot has plugins for Subversion, Perforce, Arch, Bazaar, Darcs,
> Mercurial and Git. Do you use any of these for your source code?
>
> Buildbot needs essentially three things:
>
> 1. Build master needs a way to learn about an update to specific
> project or branch
> 2. Build slaves need a way to obtain the source code
> 3. Build slaves need instructions to build the package
>
> Number 1 could be done by... I don't know... making md5 sums of
> specific files in your home directory? Number 2 would be a file copy
> operation. Number 3 would be provided by you. Do your builds have a
> uniform build procedure?

Yes, my packages have an uniform and automatic packaging system.

> The GAR packages will be the main focus now, since they comprise about
> 91% builds (that I know of, that is, ~1083 out of ~1186). Once we're
> there, we can extend support to other build sources. Of course, it
> would make it easier to implement if a code repository was available
> to buildbot.

Are you sure for the figures above?
-- 
Peter


From pfelecan at opencsw.org  Fri Jul 31 16:14:42 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 16:14:42 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org> (Dagobert
	Michelsen's message of "Fri\, 31 Jul 2009 13\:25\:01 +0200")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
Message-ID: <x64ostj5wt.fsf@abhavana.free.fr>

Dagobert Michelsen <dam at opencsw.org> writes:

> Hi Peter,
>
> Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
>> All my packages are outside gar. The source code comes from my home
>> directory...
>
> - Is there a specific reason why you are not using GAR? Without GAR it
> is
>   hard to do teams, takeover, knowledge sharing, consistent packages,
> etc.

Yes: I prefer to use my own "sausage factory". Known complexity versus
unknown complexity. Besides, my "volunteer-ship" on this project
predating gar's usage gives me the choice to use the system that I
prefer.

When the politics of the foundation change and the usage of gar is
mandatory I reserve my decision.
-- 
Peter


From pfelecan at opencsw.org  Fri Jul 31 16:20:23 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 16:20:23 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249040535-sup-2622@ntdws12.chass.utoronto.ca> (Ben Walton's
	message of "Fri\, 31 Jul 2009 07\:43\:59 -0400")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<1249040535-sup-2622@ntdws12.chass.utoronto.ca>
Message-ID: <x6zlalhr2w.fsf@abhavana.free.fr>

Ben Walton <bwalton at opencsw.org> writes:

> Excerpts from Dagobert Michelsen's message of Fri Jul 31 07:25:01 -0400 2009:
>
>> - If you are not using GAR, please also check in the source to
>>   mgar/pkg/<pkg>/trunk/legacy/ so it is properly archived. Missing
>>   build descriptions on maintainers leaving is a hard thing to cope
>>   with. IMHO it should be policy that no package is published
>>   without the recipe committed to the repository.
>
> +1 for this.  If you're still a GAR-o-phobe, at least commit the
> manual build scripts, patches, instructions etc so that knowledge
> doesn't get lost...[The svn-o-phobe's may not like this. <grin>]

Yeah, when all the methods, procedures and tools involved in our
processes are documented it'll be reasonable to ask. Until then, you are
free to look in my home directory on the build stack, where everything
is open.

I know recipes for RPM's, DEB's, &c. What's a recipe from the point of
view of gar? Can you show me one?
-- 
Peter


From maciej at opencsw.org  Fri Jul 31 16:25:45 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 15:25:45 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x68wi5j65g.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<ce97543b0907310339l43ce415by233b39834bf66e5b@mail.gmail.com>
	<x6d47hjfq3.fsf@abhavana.free.fr>
	<ce97543b0907310415j4896312cq6921efba0aa2a4b7@mail.gmail.com>
	<x68wi5j65g.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907310725i677d8630u811d6b938659001b@mail.gmail.com>

On Fri, Jul 31, 2009 at 3:09 PM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> On Fri, Jul 31, 2009 at 11:42 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>
>>>> On Fri, Jul 31, 2009 at 11:31 AM, Peter FELECAN<pfelecan at opencsw.org> wrote:
>>>>> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>>>>>
>>>>>> 2. Avoid building certain blacklisted packages. Optional: schedule
>>>>>> periodic or nightly builds for those packages.
>>>>>
>>>>> -1 as it defeats the purpose; in addition, there is no night for an
>>>>> ? international community as ours...
>>>>
>>>> Automatic builds for some packages are better than automatic builds
>>>> for no packages, so it doesn't defeat the purpose completely. Also, we
>>>> could do actual measurements: how big is the impact on the
>>>> performance?
>>>
>>> Maybe. But not on the effective build stack, please.
>>
>> Where are large packages currently built? I thought that all packages
>> must be built on the buildfarm. Next time there's an on-going
>> compilation of a large package, we can do the measurements.
>
> Some weeks ago we had I/O issues when gcc3, OpenOffice and Firefox where
> built in the same period. What I'm saying, is *automatic build must be
> done on a different hardware stack*

buildbot won't build more than one package on a single system. It
might build more than on the same hardware, because it doesn't know
about the VM layout.

>> The GAR packages will be the main focus now, since they comprise about
>> 91% builds (that I know of, that is, ~1083 out of ~1186). Once we're
>> there, we can extend support to other build sources. Of course, it
>> would make it easier to implement if a code repository was available
>> to buildbot.
>
> Are you sure for the figures above?

Do you want to argue about the figures? :-)

I counted the packages I knew about: the number of subdirectories in
mgar/pkg, plus mgar/pkg/cpan, plus number of packages listed on your
maitainer page. Feel free to offer better figures! :-)

Maciej


From mwatters at opencsw.org  Fri Jul 31 16:50:24 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 09:50:24 -0500
Subject: [csw-maintainers] Freenode?
Message-ID: <4A7304B0.1040201@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I can't seem to login to freenode using either xChat or mibbit or anything.  I
keep getting 127.0.0.1 (banned) but I am not passing that as my address?

any thoughts?

- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzBLAACgkQLrhmsXMSLxf/swCglbAaA3C5Eqi5QmORhgUIB2BD
9/IAoLy238q2lNonXy/h5hW1LcoxTyne
=CpGf
-----END PGP SIGNATURE-----


From mwatters at opencsw.org  Fri Jul 31 16:53:32 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 09:53:32 -0500
Subject: [csw-maintainers] in lieu of irc here is my update on sendmail
Message-ID: <4A73056C.3030807@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I have the "install done and working correctly"
I am getting ready to write the custom merge script and should have that done
shortly.


- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzBWwACgkQLrhmsXMSLxdG7gCgnJzYNrmBqnAwD8Vr3zq4YbG4
cxMAoLgNt1apevl9TBT466vrkhXaNJMU
=5NFR
-----END PGP SIGNATURE-----


From maciej at opencsw.org  Fri Jul 31 16:53:22 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 15:53:22 +0100
Subject: [csw-maintainers] Freenode?
In-Reply-To: <4A7304B0.1040201@opencsw.org>
References: <4A7304B0.1040201@opencsw.org>
Message-ID: <ce97543b0907310753j4375c52bv6f485cf06603616a@mail.gmail.com>

On Fri, Jul 31, 2009 at 3:50 PM, Mike Watters<mwatters at opencsw.org> wrote:
> I can't seem to login to freenode using either xChat or mibbit or anything. ?I
> keep getting 127.0.0.1 (banned) but I am not passing that as my address?
>
> any thoughts?

I've seen the same problem today: cswbb couldn't connect to the
channel. Which specific host name and port are you trying to connect
to?

Maciej


From schwindt at dfki.uni-kl.de  Fri Jul 31 16:58:10 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 31 Jul 2009 16:58:10 +0200
Subject: [csw-maintainers] Empty prototype even though
 PKGFILES_CSWcupsclient is set correctly
In-Reply-To: Your message of "Mon, 27 Jul 2009 17:01:15 BST."
	<ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com> 
Message-ID: <200907311458.n6VEwAFp019823@dfki.uni-kl.de>

> I'm having an issue with the packaging of CUPS. It was all jolly good
> until prototypes of some packages started becoming empty. It once was
> CSWlibcups. Its prototype was simply empty, and so was the package.


Talking of cups - I just pulled cups from testing 

like this :

pkgutil -t http://mirror.opencsw.org/opencsw/testing -U -u  CSWcupsclient 
CSWcupsdev



which gives me this : 

tcsh at isg-1516 [/opt/csw/lib] uname -a
SunOS isg-1516 5.10 Generic_141415-05 i86pc i386 i86pc
tcsh at isg-1516 [/opt/csw/lib] file /opt/csw/bin/lpstat
/opt/csw/bin/lpstat:    ELF 32-bit MSB executable SPARC Version 1, dynamically 
linked, stripped


Anyone else ? Or am I doing wrong ?





From skayser at opencsw.org  Fri Jul 31 16:58:32 2009
From: skayser at opencsw.org (Sebastian Kayser)
Date: Fri, 31 Jul 2009 16:58:32 +0200
Subject: [csw-maintainers] Freenode?
In-Reply-To: <4A7304B0.1040201@opencsw.org>
References: <4A7304B0.1040201@opencsw.org>
Message-ID: <4A730698.3050702@opencsw.org>

Mike Watters wrote:
> I can't seem to login to freenode using either xChat or mibbit or anything.  I
> keep getting 127.0.0.1 (banned) but I am not passing that as my address?
> 
> any thoughts?

At least for mibbit.com my last status was that freenode had banned
mibbit (abuse problems) quite a while ago and implemented their own
web-based chat.

  http://webchat.freenode.net/?channels=opencsw

Sebastian


From bonivart at opencsw.org  Fri Jul 31 17:04:56 2009
From: bonivart at opencsw.org (Peter Bonivart)
Date: Fri, 31 Jul 2009 17:04:56 +0200
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <200907311458.n6VEwAFp019823@dfki.uni-kl.de>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com> 
	<200907311458.n6VEwAFp019823@dfki.uni-kl.de>
Message-ID: <625385e30907310804g647c3634x4a201e21f01a5064@mail.gmail.com>

On Fri, Jul 31, 2009 at 4:58 PM, Nicolai
Schwindt<schwindt at dfki.uni-kl.de> wrote:
> Talking of cups - I just pulled cups from testing
>
> like this :
>
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -U -u ?CSWcupsclient
> CSWcupsdev
>
>
>
> which gives me this :
>
> tcsh at isg-1516 [/opt/csw/lib] uname -a
> SunOS isg-1516 5.10 Generic_141415-05 i86pc i386 i86pc
> tcsh at isg-1516 [/opt/csw/lib] file /opt/csw/bin/lpstat
> /opt/csw/bin/lpstat: ? ?ELF 32-bit MSB executable SPARC Version 1, dynamically
> linked, stripped
>
>
> Anyone else ? Or am I doing wrong ?

I have never seen pkgutil get this wrong. Can you provide the output
of "pkginfo -l CSWcupsclient" and "pkginfo -l CSWcupsdev" to see what
architecture the package is for? I want to know if the package file is
named wrong or what.

-- 
/peter


From maciej at opencsw.org  Fri Jul 31 17:05:41 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 16:05:41 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <200907311458.n6VEwAFp019823@dfki.uni-kl.de>
References: <ce97543b0907270901t38951f0dha6dbcd8558a95e1c@mail.gmail.com>
	<200907311458.n6VEwAFp019823@dfki.uni-kl.de>
Message-ID: <ce97543b0907310805m6d643d70oe7ae956c5532d221@mail.gmail.com>

On Fri, Jul 31, 2009 at 3:58 PM, Nicolai
Schwindt<schwindt at dfki.uni-kl.de> wrote:
>> I'm having an issue with the packaging of CUPS. It was all jolly good
>> until prototypes of some packages started becoming empty. It once was
>> CSWlibcups. Its prototype was simply empty, and so was the package.
>
>
> Talking of cups - I just pulled cups from testing
>
> like this :
>
> pkgutil -t http://mirror.opencsw.org/opencsw/testing -U -u ?CSWcupsclient
> CSWcupsdev
>
>
>
> which gives me this :
>
> tcsh at isg-1516 [/opt/csw/lib] uname -a
> SunOS isg-1516 5.10 Generic_141415-05 i86pc i386 i86pc
> tcsh at isg-1516 [/opt/csw/lib] file /opt/csw/bin/lpstat
> /opt/csw/bin/lpstat: ? ?ELF 32-bit MSB executable SPARC Version 1, dynamically
> linked, stripped
>
>
> Anyone else ? Or am I doing wrong ?

You're doing it right. The package marked x86 actually contains sparc
binaries. I'm fixing it now.

Maciej


From mwatters at opencsw.org  Fri Jul 31 17:14:54 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 10:14:54 -0500
Subject: [csw-maintainers] call for new sendmail maintainer?
In-Reply-To: <BA565FB7-89E7-448F-8112-46C110AF3AC3@opencsw.org>
References: <20090731001459.GB47846@bolthole.com>	<4A7255D1.3020000@opencsw.org>	<49275.217.227.42.13.1249025141.squirrel@ssl.skayser.de>
	<BA565FB7-89E7-448F-8112-46C110AF3AC3@opencsw.org>
Message-ID: <4A730A6E.5060609@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
Dagobert Michelsen wrote:
> Hi,

> IMHO this is practicable, but is completely against our policy of
> central NFS mounted server for all releases. However, running server
> services from NFS is also not good design, so I guess it is ok
> to make different packages here.
> 

I am all for a separate package for each OS. the way Alex's and the current
class script works is it will only install the binaries for the "host OS"
so NFS mounting to different OSs and running sendmail won't work anyway.

I will create the seperate OS packages and create a "common ARCHALL" for the
non-binary stuff.  there are 9 OS specific binaries.

- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzCm4ACgkQLrhmsXMSLxcXvQCgmL5wMnNOedlrNOh0a7DuXQvl
NQgAoJccvFez9D+Qa+f0qqsHTNzi/YLP
=32Nr
-----END PGP SIGNATURE-----


From schwindt at dfki.uni-kl.de  Fri Jul 31 17:16:47 2009
From: schwindt at dfki.uni-kl.de (Nicolai Schwindt)
Date: Fri, 31 Jul 2009 17:16:47 +0200
Subject: [csw-maintainers] Empty prototype even though
 PKGFILES_CSWcupsclient is set correctly
In-Reply-To: Your message of "Fri, 31 Jul 2009 17:04:56 +0200."
	<625385e30907310804g647c3634x4a201e21f01a5064@mail.gmail.com> 
Message-ID: <200907311516.n6VFGlIS020054@dfki.uni-kl.de>

[...]
> I have never seen pkgutil get this wrong. Can you provide the output
> of "pkginfo -l CSWcupsclient" and "pkginfo -l CSWcupsdev" to see what
> architecture the package is for? I want to know if the package file is
> named wrong or what.

I reverted both affected machines ( both x86 ) back. 
I'll redo it and get back to you 


Nicolai




From maciej at opencsw.org  Fri Jul 31 18:12:14 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 17:12:14 +0100
Subject: [csw-maintainers] Empty prototype even though
	PKGFILES_CSWcupsclient is set correctly
In-Reply-To: <200907311516.n6VFGlIS020054@dfki.uni-kl.de>
References: <625385e30907310804g647c3634x4a201e21f01a5064@mail.gmail.com>
	<200907311516.n6VFGlIS020054@dfki.uni-kl.de>
Message-ID: <ce97543b0907310912v720aab29x23c04cb4f4ecf5cd@mail.gmail.com>

On Fri, Jul 31, 2009 at 4:16 PM, Nicolai
Schwindt<schwindt at dfki.uni-kl.de> wrote:
> [...]
>> I have never seen pkgutil get this wrong. Can you provide the output
>> of "pkginfo -l CSWcupsclient" and "pkginfo -l CSWcupsdev" to see what
>> architecture the package is for? I want to know if the package file is
>> named wrong or what.
>
> I reverted both affected machines ( both x86 ) back.
> I'll redo it and get back to you

Fixed packages are in testing. Sorry for the inconvenience.

Maciej


From phil at bolthole.com  Fri Jul 31 18:47:57 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 09:47:57 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
Message-ID: <20090731164757.GB30823@bolthole.com>

On Fri, Jul 31, 2009 at 01:25:01PM +0200, Dagobert Michelsen wrote:
> Hi Peter,
>
> Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
>> All my packages are outside gar. The source code comes from my home
>> directory...
>
> - Is there a specific reason why you are not using GAR? Without GAR it  
> is
>   hard to do teams, takeover, knowledge sharing, consistent packages,  
> etc.


I will point out that there are two things, currently intertwined, that
dont neccessarily HAVE to be. that being:

1. our "source tree"
2. our "GAR build system"


If there were a simplified published API, for the source tree, that would
accept and interface build systems other than gar, then perhaps Peter F
(and "others" *cough*) would be more inclined to check their stuff into it.

for example, if there were a flat namespace in the source tree, and if it
were always possible to do

cd mgar/pkg/softwarename
  #and then one or more of
  make
  make package
  (?make clean?)

Define an absolute top-level "API" that doesnt require GAR, and you might
see more stuff in there, even from the "GAR-o-phobes"
Right now, you have sort of a defacto one from GAR usage. But you havent
formally stated, "as long as you follow [this API], its ok even if you're
not using gar]"




From maciej at opencsw.org  Fri Jul 31 19:34:45 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 18:34:45 +0100
Subject: [csw-maintainers] Build systems usage
Message-ID: <ce97543b0907311034q72e3c08aieddf3a534267bc7b@mail.gmail.com>

[splitting the thread]

On Fri, Jul 31, 2009 at 5:47 PM, Philip Brown<phil at bolthole.com> wrote:
> On Fri, Jul 31, 2009 at 01:25:01PM +0200, Dagobert Michelsen wrote:
>> Hi Peter,
>>
>> Am 31.07.2009 um 12:31 schrieb Peter FELECAN:
>>> All my packages are outside gar. The source code comes from my home
>>> directory...
>>
>> - Is there a specific reason why you are not using GAR? Without GAR it
>> is
>> ? hard to do teams, takeover, knowledge sharing, consistent packages,
>> etc.
>
>
> I will point out that there are two things, currently intertwined, that
> dont neccessarily HAVE to be. that being:
>
> 1. our "source tree"
> 2. our "GAR build system"
>
>
> If there were a simplified published API, for the source tree, that would
> accept and interface build systems other than gar, then perhaps Peter F
> (and "others" *cough*) would be more inclined to check their stuff into it.
>
> for example, if there were a flat namespace in the source tree, and if it
> were always possible to do
>
> cd mgar/pkg/softwarename
> ?#and then one or more of
> ?make
> ?make package
> ?(?make clean?)
>
> Define an absolute top-level "API" that doesnt require GAR, and you might
> see more stuff in there, even from the "GAR-o-phobes"
> Right now, you have sort of a defacto one from GAR usage. But you havent
> formally stated, "as long as you follow [this API], its ok even if you're
> not using gar]"


From bwalton at opencsw.org  Fri Jul 31 19:38:50 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 13:38:50 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6zlalhr2w.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<1249040535-sup-2622@ntdws12.chass.utoronto.ca>
	<x6zlalhr2w.fsf@abhavana.free.fr>
Message-ID: <1249061256-sup-6411@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Fri Jul 31 10:20:23 -0400 2009:

> Yeah, when all the methods, procedures and tools involved in our
> processes are documented it'll be reasonable to ask. Until then, you are
> free to look in my home directory on the build stack, where everything
> is open.

Ok, that's a fair point against not using GAR, but you could still pop
your stuff into the legacy part of the svn tree...

Documentation is a weak point, and I've really not contributed much in
that regard.  There should be enough info to get started with 'out
there' but part of the problem right now is that it's scattered around
somewhat.

If you're willing to learn, though, I'll help you move some packages
into GAR...offer valid for as long as you want it to be.  Any
maintainers interested in getting started with GAR can _always_ feel
free to ping me directly.

> I know recipes for RPM's, DEB's, &c. What's a recipe from the point of
> view of gar? Can you show me one?

It's the same really.  You're defining 'special' variables, providing
custom bits of scripts (when necessary) in the appropriate places and
learning the appropriate Make targets.

RPM defined it's own domain specific language for this...GAR is a
language built on top of functionality offered by (GNU) Make.  It's
very flexible, allowing you to interject code at pretty much any place
you need to (and for most builds, some you don't!).

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/b683385e/attachment-0002.asc>

From maciej at opencsw.org  Fri Jul 31 19:43:50 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 18:43:50 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731164757.GB30823@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
Message-ID: <ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>

[...]

I've split the thread about the build systems into another one, and
I'd like to go back to the buildbot topic. Are there any comments
about the design or deployment plan?

Maciej


From bwalton at opencsw.org  Fri Jul 31 19:50:33 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 13:50:33 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6zlalhr2w.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<1249040535-sup-2622@ntdws12.chass.utoronto.ca>
	<x6zlalhr2w.fsf@abhavana.free.fr>
Message-ID: <1249061943-sup-231@ntdws12.chass.utoronto.ca>

Excerpts from Peter FELECAN's message of Fri Jul 31 10:20:23 -0400 2009:
> I know recipes for RPM's, DEB's, &c. What's a recipe from the point of
> view of gar? Can you show me one?

Just so that it stands out a bit, I'm linking to a few 'Recipes' that
I've done recently.

gitosis: http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/gitosis/trunk/Makefile

php mode for emacs:
http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/phpmode/trunk/Makefile

git: http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/git/trunk/Makefile

These are of varying complexity and demonstrate a few of the different
things that are commonly done.  Gitosis leverages Mike's new class
action script cswpycompile to generate .py{c,o} files at install.
php_mode hooks into both the post-extract and post-install steps of
the build to add custom script.  Git splits one source package into
several smaller packages, uses a git repo instead of a tarball for the
source and hooks a few more steps in the process.

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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/331223d3/attachment-0002.asc>

From pfelecan at opencsw.org  Fri Jul 31 19:49:00 2009
From: pfelecan at opencsw.org (Peter FELECAN)
Date: Fri, 31 Jul 2009 19:49:00 +0200
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
	(Maciej Blizinski's message of "Fri\,
	31 Jul 2009 18\:43\:50 +0100")
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
Message-ID: <x6vdl8ivzn.fsf@abhavana.free.fr>

"Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:

> [...]
>
> I've split the thread about the build systems into another one, and
> I'd like to go back to the buildbot topic. Are there any comments
> about the design or deployment plan?

Split or not split, my comments are valid for this
discussion. Consequently, please take into account this design
requirements:

1. The buildbot deployment must be done on a separate hardware from that
   used for the regular, "manual", builds.

2. Other build systems that gar must be able to use the deployed
   "automatic" build infrastructure using buildbot.

TIA
-- 
Peter


From bwalton at opencsw.org  Fri Jul 31 20:07:01 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 14:07:01 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731164757.GB30823@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
Message-ID: <1249062711-sup-8879@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 31 12:47:57 -0400 2009:

> for example, if there were a flat namespace in the source tree, and if it
> were always possible to do

Part of the reason for the (admittedly) nasty directory structure is
subversion.  While we're following 'best'[1] practices for svn use by
having the {trunk,branches,tags}/ directory structure under each top
level package, it definitely makes the tree structure uglier.

The ability to checkout only a single part of the tree with svn is
it's only useful feature[2], I think.  This feature makes following the
three sub directory convention beneficial as tagging and branching can
be done more easily on a partial checkout[3]

> Define an absolute top-level "API" that doesnt require GAR, and you might
> see more stuff in there, even from the "GAR-o-phobes"
> Right now, you have sort of a defacto one from GAR usage. But you havent
> formally stated, "as long as you follow [this API], its ok even if you're
> not using gar]"

This is a good idea.  While I think GAR is good and it saves me lots
of time, we should promote a set of Make targets that all packages
support, even if they don't use GAR.  What should a Makefile
accomplish so that it's globally usable, say by a build bot system?

I'd say:

1. It must be able to retrieve it's own source tarball/repo/whatever,
   so the 'fetch' target should be honoured.
2. It must be able to apply patches supplied locally, so 'patch' is a
   must.
3. I'd then say both 'build' and 'package' should be individually
   callable steps.  A buildbot may only want to test that the updated
   commit can still successfully build the package while a release
   tool may need to be able to take things right up to the package
   files.
4. A 'clean' target is always a good thing to support.
5. A set of variables should be honoured by any Makefile such that a
   build system knows where to find outputs.

Moving to an API like this should promote more use of the global
repository while still making it usable for a global build/release
tool.  Are there other globally useful make targets I've overlooked?

Thoughts?
-Ben

[1] Putting best in a sentence with svn feels wrong.
[2] When comparing features offered by any modern SCM/VCS tool.
[3] This is my understanding.  Corrections welcome.
-- 
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: <http://lists.opencsw.org/pipermail/maintainers/attachments/20090731/2f0b2af1/attachment-0002.asc>

From james at opencsw.org  Fri Jul 31 20:15:10 2009
From: james at opencsw.org (James Lee)
Date: Fri, 31 Jul 2009 18:15:10 GMT
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731164757.GB30823@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
Message-ID: <20090731.18151000.1268557623@gyor.oxdrove.co.uk>

On 31/07/09, 17:47:57, Philip Brown <phil at bolthole.com> wrote regarding Re:
[csw-maintainers] Buildbot:

> If there were a simplified published API, for the source tree, that would
> accept and interface build systems other than gar, then perhaps Peter F
> (and "others" *cough*) would be more inclined to check their stuff into
> it.

> for example, if there were a flat namespace in the source tree, and if it
> were always possible to do

> cd mgar/pkg/softwarename
>   #and then one or more of
>   make
>   make package
>   (?make clean?)

> Define an absolute top-level "API" that doesnt require GAR

Go further, just have a single entry point per project (not package) per
version that can be run with exec.  i.e., don't use make as the first
step.  You can call make from a script if you want, I suppose so too
could one call a script from make but it's not as pure and simplistic.

A problem with GAR is using excess make syntax which has to one of the
less expressive ways of defining a procedure.




James.


From james at opencsw.org  Fri Jul 31 20:15:08 2009
From: james at opencsw.org (James Lee)
Date: Fri, 31 Jul 2009 18:15:08 GMT
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
Message-ID: <20090731.18150800.1547323678@gyor.oxdrove.co.uk>

On 31/07/09, 18:43:50, Maciej "(Matchek)" Blizinski <maciej at opencsw.org> 
wrote regarding Re: [csw-maintainers] Buildbot:

> I'd like to go back to the buildbot topic. Are there any comments
> about the design or deployment plan?

Why do you think it's needed at all?  Why do you want to rebuild
what presumably has been built already - is the World not hot
enough for you?



James.


From phil at bolthole.com  Fri Jul 31 20:41:08 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 11:41:08 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249062711-sup-8879@ntdws12.chass.utoronto.ca>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
Message-ID: <20090731184108.GB76412@bolthole.com>

On Fri, Jul 31, 2009 at 02:07:01PM -0400, Ben Walton wrote:
> 
> 1. It must be able to retrieve it's own source tarball/repo/whatever,
>    so the 'fetch' target should be honoured.
> 2. It must be able to apply patches supplied locally, so 'patch' is a
>    must.

I think applying the patch should be part of "get source", and there should
not be a mandated separate one.

> 3. I'd then say both 'build' and 'package' should be individually
>    callable steps....

> 4. A 'clean' target is always a good thing to support.


> 5. A set of variables should be honoured by any Makefile such that a
>    build system knows where to find outputs.

I think this is making it too complicated.
It needs to be *simple*,for most widespread use.

*standard* makefile vars, such as CFLAGS, etc, are fine. 
But other things I think should be optional. Otherwise, people tend to get
carried away defining extra things that "must" be supported.






From phil at bolthole.com  Fri Jul 31 20:42:45 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 11:42:45 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731.18151000.1268557623@gyor.oxdrove.co.uk>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<20090731.18151000.1268557623@gyor.oxdrove.co.uk>
Message-ID: <20090731184245.GC76412@bolthole.com>

On Fri, Jul 31, 2009 at 06:15:10PM +0000, James Lee wrote:
> 
> Go further, just have a single entry point per project (not package) per
> version that can be run with exec.  i.e., don't use make as the first
> step.  You can call make from a script if you want, I suppose so too
> could one call a script from make but it's not as pure and simplistic.

in other words, debian style, have a "build" executable.... even though 99%
of the time, the debian "executable" is a script that starts with

#!/bin/make

:-)

[ie: a makefile!]



From james at opencsw.org  Fri Jul 31 20:56:22 2009
From: james at opencsw.org (James Lee)
Date: Fri, 31 Jul 2009 18:56:22 GMT
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731184245.GC76412@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<20090731.18151000.1268557623@gyor.oxdrove.co.uk>
	<20090731184245.GC76412@bolthole.com>
Message-ID: <20090731.18562200.4124391153@gyor.oxdrove.co.uk>

On 31/07/09, 19:42:45, Philip Brown <phil at bolthole.com> wrote regarding Re:
[csw-maintainers] Buildbot:

> > Go further, just have a single entry point per project (not package) per
> > version that can be run with exec.  i.e., don't use make as the first
> > step.  You can call make from a script if you want, I suppose so too
> > could one call a script from make but it's not as pure and simplistic.

> in other words, debian style, have a "build" executable.... even though
> 99% of the time, the debian "executable" is a script that starts with

> #!/bin/make

> :-)

> [ie: a makefile!]

If that's what floats your boat.  At least you have been able to choose
which make.



James.


From maciej at opencsw.org  Fri Jul 31 21:05:52 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 20:05:52 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249062711-sup-8879@ntdws12.chass.utoronto.ca>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
Message-ID: <ce97543b0907311205q41cb041cj16e091dfdfa19b50@mail.gmail.com>

On Fri, Jul 31, 2009 at 7:07 PM, Ben Walton<bwalton at opencsw.org> wrote:
> 3. I'd then say both 'build' and 'package' should be individually
> ? callable steps. ?A buildbot may only want to test that the updated
> ? commit can still successfully build the package while a release
> ? tool may need to be able to take things right up to the package
> ? files.

It would be nice to have a common interface for building 64-bit and
32-bit ISAs in a single package. How is that handled in non-GAR
builds?

Maciej


From maciej at opencsw.org  Fri Jul 31 21:10:09 2009
From: maciej at opencsw.org (Maciej (Matchek) Blizinski)
Date: Fri, 31 Jul 2009 20:10:09 +0100
Subject: [csw-maintainers] Buildbot
In-Reply-To: <x6vdl8ivzn.fsf@abhavana.free.fr>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
	<x6vdl8ivzn.fsf@abhavana.free.fr>
Message-ID: <ce97543b0907311210kf8840e7me6ff41de6313ffda@mail.gmail.com>

On Fri, Jul 31, 2009 at 6:49 PM, Peter FELECAN<pfelecan at opencsw.org> wrote:
> "Maciej (Matchek) Blizinski" <maciej at opencsw.org> writes:
>
>> [...]
>>
>> I've split the thread about the build systems into another one, and
>> I'd like to go back to the buildbot topic. Are there any comments
>> about the design or deployment plan?
>
> Split or not split, my comments are valid for this
> discussion. Consequently, please take into account this design
> requirements:
>
> 1. The buildbot deployment must be done on a separate hardware from that
> ? used for the regular, "manual", builds.

This requirement might be a showstopper here. Can we trade it for a
restricted set of packages to build, avoiding the large ones?

> 2. Other build systems that gar must be able to use the deployed
> ? "automatic" build infrastructure using buildbot.

The infrastructure will be actually automatic, not "automatic". ;-)

I've updated the wiki. Buildbot essentially needs to know what shell
command to run in the top package directory. I don't care what it is,
as long as there's a reliable discovery method of what should that
command be.

Maciej


From phil at bolthole.com  Fri Jul 31 22:05:32 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 13:05:32 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <ce97543b0907311210kf8840e7me6ff41de6313ffda@mail.gmail.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<ce97543b0907311043n1ac06b6eo738c4df927d589a2@mail.gmail.com>
	<x6vdl8ivzn.fsf@abhavana.free.fr>
	<ce97543b0907311210kf8840e7me6ff41de6313ffda@mail.gmail.com>
Message-ID: <20090731200532.GD76412@bolthole.com>

On Fri, Jul 31, 2009 at 08:10:09PM +0100, Maciej (Matchek) Blizinski wrote:
> > 1. The buildbot deployment must be done on a separate hardware from that
> > ?? used for the regular, "manual", builds.
> 
> This requirement might be a showstopper here. Can we trade it for a
> restricted set of packages to build, avoiding the large ones?

what would that accomplish?

it would still have to be a large set, to be useful.
and large sets, take large amounts of time, to compile.



From mwatters at opencsw.org  Fri Jul 31 22:17:00 2009
From: mwatters at opencsw.org (Mike Watters)
Date: Fri, 31 Jul 2009 15:17:00 -0500
Subject: [csw-maintainers] HowTo question with mGAR
Message-ID: <4A73513C.5000107@opencsw.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Can someone tell me how to include "extra" package admin files using mGAR.
I have 3 custom class scripts that I need to include in the package but can't
figure out how to force gar to include them.


- --
~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
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (SunOS)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpzUTsACgkQLrhmsXMSLxc9KwCbBQUlThk2IFuP6cQ2ZcCDq9fU
MSgAoLEFtajQAjvCCS9edqOGMAxFfh34
=nizh
-----END PGP SIGNATURE-----


From dam at baltic-online.de  Wed Jul 29 20:05:49 2009
From: dam at baltic-online.de (Dagobert Michelsen)
Date: Wed, 29 Jul 2009 20:05:49 +0200
Subject: [csw-maintainers] Statistics on ibiblio
Message-ID: <696FF115-EAAF-4987-B8B3-DDB4BBF786EB@baltic-online.de>

Hi there,

there are also statistics available for ibiblio:

1. Log in to <https://stats.ibiblio.org>
    User: guest Password: ibiblio
2. Click on www.ibiblio.org.public
3. Click on Pages & Files
4. Click on Directory Drilldown
5. Click on Directory by Files Drilldown
6. Click on /pub
7. Click on /packages
8. Click on /solaris

...enjoy :-)


Best regards

   -- Dago


From bwalton at opencsw.org  Fri Jul 31 23:13:49 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 17:13:49 -0400
Subject: [csw-maintainers] HowTo question with mGAR
In-Reply-To: <4A73513C.5000107@opencsw.org>
References: <4A73513C.5000107@opencsw.org>
Message-ID: <1249074745-sup-6871@ntdws12.chass.utoronto.ca>

Excerpts from Mike Watters's message of Fri Jul 31 16:17:00 -0400 2009:

> Can someone tell me how to include "extra" package admin files using mGAR.
> I have 3 custom class scripts that I need to include in the package but can't
> figure out how to force gar to include them.

Do you mean that you need the package to stuff them into
/usr/sadm/install/scripts from files/?

-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 bwalton at opencsw.org  Fri Jul 31 23:20:33 2009
From: bwalton at opencsw.org (Ben Walton)
Date: Fri, 31 Jul 2009 17:20:33 -0400
Subject: [csw-maintainers] Buildbot
In-Reply-To: <20090731184108.GB76412@bolthole.com>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
	<20090731184108.GB76412@bolthole.com>
Message-ID: <1249074967-sup-4066@ntdws12.chass.utoronto.ca>

Excerpts from Philip Brown's message of Fri Jul 31 14:41:08 -0400 2009:

> I think applying the patch should be part of "get source", and there should
> not be a mandated separate one.

Ok.  For a 'broad strokes' approach, that would be acceptable, imo.
(This implies that the 'fetch' step would also encompass 'extract'
which is fine.)

> > 3. I'd then say both 'build' and 'package' should be individually
> >    callable steps....
> > 4. A 'clean' target is always a good thing to support.
> > 5. A set of variables should be honoured by any Makefile such that a
> >    build system knows where to find outputs.

> I think this is making it too complicated.
> It needs to be *simple*,for most widespread use.

Ok, as long as there is some sort of convention where automated files
could expect to find their outputs.  The Variable approach lets the
tool set that as it desires, which makes changing it somewhat easier,
but as long as good choices were made up front...

Are you suggesting that a build step separate from package is also too
much though?  I think that is an important distinction, but arguments
to the contrary are 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.


From phil at bolthole.com  Fri Jul 31 23:55:13 2009
From: phil at bolthole.com (Philip Brown)
Date: Fri, 31 Jul 2009 14:55:13 -0700
Subject: [csw-maintainers] Buildbot
In-Reply-To: <1249074967-sup-4066@ntdws12.chass.utoronto.ca>
References: <ce97543b0907310139s22746360tfd9f765ace8411ba@mail.gmail.com>
	<x6ljm5jkg2.fsf@abhavana.free.fr>
	<ce97543b0907310325p5b173214h5c3c7fec735be030@mail.gmail.com>
	<x6hbwtjg82.fsf@abhavana.free.fr>
	<9C6FD617-8E9C-4649-9361-4525D8A68C98@opencsw.org>
	<20090731164757.GB30823@bolthole.com>
	<1249062711-sup-8879@ntdws12.chass.utoronto.ca>
	<20090731184108.GB76412@bolthole.com>
	<1249074967-sup-4066@ntdws12.chass.utoronto.ca>
Message-ID: <20090731215513.GF76412@bolthole.com>

On Fri, Jul 31, 2009 at 05:20:33PM -0400, Ben Walton wrote:
> > I think this is making it too complicated.
> > It needs to be *simple*,for most widespread use.
> 
> Ok, as long as there is some sort of convention where automated files
> could expect to find their outputs.

There is no need for that. I think that's overengineering.
There is only one set of "outputs" actually *needed* for the simplified
API,and that is "where do the packages go?"

Similarly, i think that there is no need for variables like "where do I do
builds",because the reasonable assumption would be "in the current
directory".

> Are you suggesting that a build step separate from package is also too
> much though?  I think that is an important distinction, but arguments
> to the contrary are welcome.

I wasnt suggesting that originally. But I think perhaps having two steps
there, is also unneccessary.
Even in the case where there is some need to tweak something silly like the
pkginfo file... a good "build system" should run the (compilephase)build,
but then automatically detect there is nothing that needs recompiling. 
So putting together the new package, with the updated pkginfo file, should
be relatively speedy.