From maciej at opencsw.org Sun Oct 2 00:52:13 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 1 Oct 2011 23:52:13 +0100 Subject: [csw-maintainers] OpenCSW Summercamp 2011 in Kiel summary Message-ID: Hello maintainers, Here's a quick summary of the Summarcamp held last weekend in Kiel, Germany. The attendees were, in no particular order: Dagobert Michelsen, Peter Bonivart, Maciej Blizi?ski, Ihsan Dogan, Rafael Ostertag, Sebastian Kayser, J?rg Schilling Topics and conclusions are kept in the meeting minutes, links are below. Here's couple of highlights: - we will be dropping Solaris 9 support, meaning that it goes from full support to best effort - we'll build for sparcv8+ and pentium_pro by default on Solaris 10 - catalogs will be marked using their release names; upon an upgrade from one release to another, pkgutil will issue an additional warning and a confirmation dialog. It will play a major role in the next stable release. Photo: http://t.co/dIFXcdP4 The Summercamp was not just about meetings and minutes. We also had fun! We visited the computer museum in Kiel, which covers years 1940 to 2000, and features an extensive and very interesting collection of computer hardware. In the media dome, the screen covered your whole field of vision. We watched a show which mixed U2 music with computer generated graphics. We also watched an animation of the Kiel city hall becoming a rocket, flying off into space, and landing next to the computer museum. Youtube has a 3D version of that clip. Flying Kiel city hall: http://www.youtube.com/watch?v=lPvgjzgINg0 Here are links to the meeting minutes: Friday notes: http://t.co/zHOxWmzT Saturday notes: http://t.co/hkEDVOg1 Sunday notes: http://t.co/AZeElR1L Many thanks go to Dagobert and Baltic Online for hosting the Summercamp! Maciej From rupert at opencsw.org Sun Oct 2 15:08:42 2011 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 2 Oct 2011 15:08:42 +0200 Subject: [csw-maintainers] NONVERSIONED packages using "gmake platforms" Message-ID: hi, "gmake platforms" called on login.opencsw.org:pkg/mercurial/trunk produced "NONVERSIONED" packages, while "gmake package" in the same directory called on unstable9s worked fine. rupert. From skayser at opencsw.org Sun Oct 2 15:19:08 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 2 Oct 2011 15:19:08 +0200 Subject: [csw-maintainers] gar.opencsw.org: spam on our trac wiki Message-ID: <20111002131908.GD26160@sebastiankayser.de> Hi, someone left a steakhouse ad link on our GAR wiki front page the other day (already deleted). In case this re-occurs, can someone keep an eye on it (and if it starts to get pesky, think about tweaking the wiki permissions)? I'll be away for a week without internet access. Sebastian From maciej at opencsw.org Sun Oct 2 15:42:06 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 2 Oct 2011 14:42:06 +0100 Subject: [csw-maintainers] Magic Cookie for catalogs In-Reply-To: <5FF70CA4-6A97-4D98-8B94-B86945F87685@opencsw.org> References: <5FF70CA4-6A97-4D98-8B94-B86945F87685@opencsw.org> Message-ID: 2011/9/30 Dagobert Michelsen : > Hi Maciej, > > Am 30.09.2011 um 11:06 schrieb Maciej (Matchek) Blizi?ski: >> 2011/9/30 Dagobert Michelsen : >>> I am currently fiddling with who mirrors OpenCSW, reaching mirror maintainers and >>> finding "unauthorized" mirrors. For this it would be more precise if we would introduce >>> a magic cookie in the catalog features which would help us identifying OpenCSW catalogs. >>> I could then google for that cookie and find occurrences and match that against what >>> we already have listed. >> >> Cool idea! >> >> echo -n OpenCSW | gmd5sum >> >> Currently does not match any documents. ?(Not posting the actual value >> to prevent list archives from coming up in search results.) > > Ok, I suggest cycling this on a regular (yearly?) basis like the hash for > ?OpenCSW 2011 > for really old catalogs to not show up any more. I tried to think of a coherent scheme and couldn't. When you think about it, there is already a lot of hashes in the catalog, so we could use one of the existing ones. maciej at quince:~$ curl -s http://mirror.opencsw.org/opencsw/unstable/sparc/5.9/catalog | grep ^libz1 | awk '{print $5}' eb5f2244811b5b4778f4bd2b5b34eda0 This search alone already gives you useful results. Maybe we don't need another hash after all? Maciej From maciej at opencsw.org Sun Oct 2 15:46:27 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 2 Oct 2011 14:46:27 +0100 Subject: [csw-maintainers] GAR performance, calls to mkpackage Message-ID: When finishing 'gmake platforms', the build spends quite a bit of time at this step: * Platform solaris9-i386 (built on host 'unstable9x') Use of uninitialized value $REVISION in sprintf at /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. Use of uninitialized value $REVISION in sprintf at /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. Use of uninitialized value $REVISION in sprintf at /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. Use of uninitialized value $REVISION in sprintf at /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. Use of uninitialized value $REVISION in sprintf at /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. Use of uninitialized value $REVISION in sprintf at /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. It looks like repeated mkpackage calls. But gar already knows what packages have been built, doesn't it? Why call mkpackage again? Maciej From bwalton at opencsw.org Sun Oct 2 16:44:25 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 02 Oct 2011 10:44:25 -0400 Subject: [csw-maintainers] Magic Cookie for catalogs In-Reply-To: References: <5FF70CA4-6A97-4D98-8B94-B86945F87685@opencsw.org> Message-ID: <1317566214-sup-954@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Sun Oct 02 09:42:06 -0400 2011: > I tried to think of a coherent scheme and couldn't. The goal is detecting stale catalogs, right? If so, we need two things. 1. The ability to locate catalogs by some unique string. 2. The ability to see if it's stale when we locate it. The first can be accomplished by some embeded hash. We could use the md5sum of a package that is glacial (CSWcommon) or one that we change with some defined frequency (maybe CSWpki if we start rotating keys for every named release?). Once we can locate the catalogs, the embedded time stamp should be enough to detect a mirror that is not pulling frequently enough. Alternately, we could use two md5sums to detect stale catalogs. The first should be an old md5sum (CSWcommon) and the second a newly added/updated package md5sum. The second could be determined dynamically at the time of the check by comparing REV= stamps and picking something that is within the time window of what we'd consider a current mirror. A fresh mirror should have both of these hashes. A stale mirror would have only one. Would this work? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From bwalton at opencsw.org Sun Oct 2 16:56:59 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 02 Oct 2011 10:56:59 -0400 Subject: [csw-maintainers] GAR performance, calls to mkpackage In-Reply-To: References: Message-ID: <1317567320-sup-3104@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Sun Oct 02 09:46:27 -0400 2011: > When finishing 'gmake platforms', the build spends quite a bit of time > at this step: > > * Platform solaris9-i386 (built on host 'unstable9x') > Use of uninitialized value $REVISION in sprintf at > /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. > Use of uninitialized value $REVISION in sprintf at > /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. > Use of uninitialized value $REVISION in sprintf at > /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. > Use of uninitialized value $REVISION in sprintf at > /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. > Use of uninitialized value $REVISION in sprintf at > /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. > Use of uninitialized value $REVISION in sprintf at > /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. > > It looks like repeated mkpackage calls. But gar already knows what > packages have been built, doesn't it? Why call mkpackage again? It's been a long time since I looked at this part of GAR, but iirc, it's using these calls to generate the file names for display. It's relying on some of the gspec magic, I think, and doing it directly in GAR was challenging so this made sense. (This is all from memory.) Dago, is this right? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From bonivart at opencsw.org Sun Oct 2 20:00:49 2011 From: bonivart at opencsw.org (Peter Bonivart) Date: Sun, 2 Oct 2011 20:00:49 +0200 Subject: [csw-maintainers] OpenCSW Summercamp 2011 in Kiel summary In-Reply-To: References: Message-ID: 2011/10/2 Maciej (Matchek) Blizi?ski : > Many thanks go to Dagobert and Baltic Online for hosting the Summercamp! I also want to thank Dago, nice to meet some of his colleagues at Baltic Online as well. And of course the other participants. Loved the museum! /peter From jeff at cjsa.com Mon Oct 3 04:19:46 2011 From: jeff at cjsa.com (Jeffery Small) Date: Mon, 3 Oct 2011 02:19:46 GMT Subject: [csw-maintainers] Catalog public key Message-ID: I went away for six weeks and when I returned the catalog no longer updates with pkg-get -U from any mirror. The catalog downloads and then I get the error: gpg: WARNING: unsafe ownership on configuration file `/u/jeff/.gnupg/gpg.conf' gpg: Signature made Sat Oct 01 21:36:25 2011 PDT using DSA key ID 9306CC77 gpg: Can't check signature: public key not found ERROR: catalog failed signature check (status 2) You need to install the public key, either manually, or automatically through a keyserver. For keyserver use, try one of echo keyserver search.keyserver.net >>/.gnupg/options echo keyserver search.keyserver.net >>/.gnupg/gpg.conf Catalog failed signature verify. Quitting. I've tried the suggestions above with no positive results. My gpg.conf file contained the line: keyserver hkp://subkeys.pgp.net and worked for years before. Any suggestion as to what has changed and how to resolve it? Pgp has always been something of a mystery to me. Thanks. Regards, -- Jeff C. Jeffery Small CJSA LLC 206-232-3338 jeff at cjsa.com 7000 E Mercer Way, Mercer Island, WA 98040 From maciej at opencsw.org Mon Oct 3 09:58:22 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 3 Oct 2011 08:58:22 +0100 Subject: [csw-maintainers] Catalog public key In-Reply-To: References: Message-ID: 2011/10/3 Jeffery Small : > I went away for six weeks and when I returned the catalog no longer updates > with pkg-get -U from any mirror. ?The catalog downloads and then I get the > error: > > gpg: WARNING: unsafe ownership on configuration file `/u/jeff/.gnupg/gpg.conf' > gpg: Signature made Sat Oct 01 21:36:25 2011 PDT using DSA key ID 9306CC77 > gpg: Can't check signature: public key not found > > ERROR: catalog failed signature check (status 2) > > You need to install the public key, either manually, > or automatically through a keyserver. > For keyserver use, try one of > ?echo keyserver search.keyserver.net >>/.gnupg/options > ?echo keyserver search.keyserver.net >>/.gnupg/gpg.conf > Catalog failed signature verify. Quitting. > > I've tried the suggestions above with no positive results. ?My gpg.conf file > contained the line: > > keyserver hkp://subkeys.pgp.net > > and worked for years before. ?Any suggestion as to what has changed and how > to resolve it? ?Pgp has always been something of a mystery to me. ?Thanks. GPG uses user's home directory to store keys. One failure mode would be to import keys to normal user's keyring and then run pkgutil as root. The fix is to import the gpg key for the root user. In the unstable catalog, we already have the cswpki package, which in conjuction with pkgutil solves the problem - the package delivers the key and pkgutil uses it. If you want to do it manually, you can download the key from the mirrors page. http://www.opencsw.org/mirrors/ Maciej From maciej at opencsw.org Mon Oct 3 10:19:10 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 3 Oct 2011 09:19:10 +0100 Subject: [csw-maintainers] mgar 0.94-rc1 - changes and call for testing In-Reply-To: <20110926184531.GH21825@sebastiankayser.de> References: <20110926184531.GH21825@sebastiankayser.de> Message-ID: The newpkg functionality still doesn't work for me: maciej at login [login]:~ > ssh unstable9s Last login: Mon Oct 3 00:33:39 2011 from login.bo.opencs Sun Microsystems Inc. SunOS 5.9 Generic May 2002 maciej at unstable9s :~ > cd src/opencsw maciej at unstable9s :~/src/opencsw > mgar newpkg foo ERROR: Template /home/maciej/src/opencsw/newpkg-Makefile not found Likely fix: (cd /home/maciej/src/opencsw && svn up newpkg-Makefile) maciej at unstable9s :~/src/opencsw > (cd /home/maciej/src/opencsw && svn up newpkg-Makefile) Updating 'newpkg-Makefile': At revision 15838. maciej at unstable9s :~/src/opencsw > mgar newpkg foo ERROR: Template /home/maciej/src/opencsw/newpkg-Makefile not found Likely fix: (cd /home/maciej/src/opencsw && svn up newpkg-Makefile) maciej at unstable9s :~/src/opencsw > svn info Path: . Working Copy Root Path: /home/maciej/src/opencsw URL: https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar Repository Root: https://gar.svn.sourceforge.net/svnroot/gar Repository UUID: d3b55034-1cff-0310-a425-aefe953e1e90 Revision: 15728 Node Kind: directory Schedule: normal Last Changed Author: dmichelsen Last Changed Rev: 15728 Last Changed Date: 2011-09-24 10:01:34 +0200 (Sat, 24 Sep 2011) Do you have any ideas why could it fail? Maciej From dam at opencsw.org Mon Oct 3 18:13:48 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 3 Oct 2011 18:13:48 +0200 Subject: [csw-maintainers] NONVERSIONED packages using "gmake platforms" In-Reply-To: References: Message-ID: Hi Rupert, Am 02.10.2011 um 15:08 schrieb rupert THURNER: > "gmake platforms" called on login.opencsw.org:pkg/mercurial/trunk > produced "NONVERSIONED" packages, while "gmake package" in the same > directory called on unstable9s worked fine. When I try to reproduce the your result from /home/rupert there is no pkg/ and pkgs/ does not contain a mercurial. Does "svn status" work on login on this directory? Best regards -- Dago From dam at opencsw.org Mon Oct 3 18:21:51 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 3 Oct 2011 18:21:51 +0200 Subject: [csw-maintainers] gar.opencsw.org: spam on our trac wiki In-Reply-To: <20111002131908.GD26160@sebastiankayser.de> References: <20111002131908.GD26160@sebastiankayser.de> Message-ID: <1D482DF1-A01E-476B-8A4A-5B92742B3244@opencsw.org> Hi Sebastian, Am 02.10.2011 um 15:19 schrieb Sebastian Kayser: > someone left a steakhouse ad link on our GAR wiki front page the other > day (already deleted). In case this re-occurs, can someone keep an eye > on it (and if it starts to get pesky, think about tweaking the wiki > permissions)? I'll be away for a week without internet access. He did it again. I opened a ticket for abuse for user "jauhis" and kept the history for the support to see. Best regards -- Dago From dam at opencsw.org Mon Oct 3 18:24:36 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 3 Oct 2011 18:24:36 +0200 Subject: [csw-maintainers] Magic Cookie for catalogs In-Reply-To: <1317566214-sup-954@pinkfloyd.chass.utoronto.ca> References: <5FF70CA4-6A97-4D98-8B94-B86945F87685@opencsw.org> <1317566214-sup-954@pinkfloyd.chass.utoronto.ca> Message-ID: Hi, Am 02.10.2011 um 16:44 schrieb Ben Walton: > Excerpts from Maciej (Matchek) Blizi?ski's message of Sun Oct 02 09:42:06 -0400 2011: > >> I tried to think of a coherent scheme and couldn't. > > The goal is detecting stale catalogs, right? If so, we need two > things. > > 1. The ability to locate catalogs by some unique string. > 2. The ability to see if it's stale when we locate it. > > The first can be accomplished by some embeded hash. We could use the > md5sum of a package that is glacial (CSWcommon) or one that we change > with some defined frequency (maybe CSWpki if we start rotating keys > for every named release?). > > Once we can locate the catalogs, the embedded time stamp should be > enough to detect a mirror that is not pulling frequently enough. > > Alternately, we could use two md5sums to detect stale catalogs. The > first should be an old md5sum (CSWcommon) and the second a newly > added/updated package md5sum. The second could be determined > dynamically at the time of the check by comparing REV= stamps and > picking something that is within the time window of what we'd consider > a current mirror. A fresh mirror should have both of these hashes. A > stale mirror would have only one. > > Would this work? Sounds reasonable. What I would like to do it to use some kind of search API like this: http://code.google.com/intl/de-DE/apis/websearch/ and genearate the list of found mirror candidates for me so I don't need to do everything manually. However, the above is deprecated. Has anyone experience with the new search API? Best regards -- Dago From dam at opencsw.org Mon Oct 3 18:26:21 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 3 Oct 2011 18:26:21 +0200 Subject: [csw-maintainers] GAR performance, calls to mkpackage In-Reply-To: <1317567320-sup-3104@pinkfloyd.chass.utoronto.ca> References: <1317567320-sup-3104@pinkfloyd.chass.utoronto.ca> Message-ID: <316EF16B-1D40-41AD-BC52-3D0B89E3835F@opencsw.org> Hi, Am 02.10.2011 um 16:56 schrieb Ben Walton: > Excerpts from Maciej (Matchek) Blizi?ski's message of Sun Oct 02 09:46:27 -0400 2011: >> When finishing 'gmake platforms', the build spends quite a bit of time >> at this step: >> >> * Platform solaris9-i386 (built on host 'unstable9x') >> Use of uninitialized value $REVISION in sprintf at >> /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. >> Use of uninitialized value $REVISION in sprintf at >> /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. >> Use of uninitialized value $REVISION in sprintf at >> /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. >> Use of uninitialized value $REVISION in sprintf at >> /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. >> Use of uninitialized value $REVISION in sprintf at >> /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. >> Use of uninitialized value $REVISION in sprintf at >> /home/maciej/src/opencsw/.buildsys/v2/gar/gar//bin/mkpackage line 31. >> >> It looks like repeated mkpackage calls. But gar already knows what >> packages have been built, doesn't it? Why call mkpackage again? > > It's been a long time since I looked at this part of GAR, but iirc, > it's using these calls to generate the file names for display. It's > relying on some of the gspec magic, I think, and doing it directly in > GAR was challenging so this made sense. (This is all from memory.) > > Dago, is this right? It was mostly due to static gspec files which parsing was a bit difficult and mkpackage knows best anyway. Now that static gspecs are really deprecated we could get rid of those. Would you mind opening a bug report? Best regards -- Dago From dam at opencsw.org Mon Oct 3 18:27:28 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 3 Oct 2011 18:27:28 +0200 Subject: [csw-maintainers] OpenCSW Summercamp 2011 in Kiel summary In-Reply-To: References: Message-ID: <0BA6E723-1C7B-4ADA-937D-F361943B48A0@opencsw.org> Hi guys, Am 02.10.2011 um 20:00 schrieb Peter Bonivart: > 2011/10/2 Maciej (Matchek) Blizi?ski : >> Many thanks go to Dagobert and Baltic Online for hosting the Summercamp! > > I also want to thank Dago, nice to meet some of his colleagues at > Baltic Online as well. And of course the other participants. > > Loved the museum! Glad you liked it :-) I'd be happy to host it anytime again! Best regards -- Dago From maciej at opencsw.org Mon Oct 3 18:31:52 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 3 Oct 2011 17:31:52 +0100 Subject: [csw-maintainers] GAR performance, calls to mkpackage In-Reply-To: <316EF16B-1D40-41AD-BC52-3D0B89E3835F@opencsw.org> References: <1317567320-sup-3104@pinkfloyd.chass.utoronto.ca> <316EF16B-1D40-41AD-BC52-3D0B89E3835F@opencsw.org> Message-ID: https://sourceforge.net/apps/trac/gar/ticket/61 From bwalton at opencsw.org Wed Oct 5 04:33:07 2011 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 04 Oct 2011 22:33:07 -0400 Subject: [csw-maintainers] status on web/mantis integration In-Reply-To: References: <1316956704-sup-5312@pinkfloyd.chass.utoronto.ca> Message-ID: <1317781747-sup-6712@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Mon Sep 26 06:21:02 -0400 2011: > Cool. I still have the REST interface update on my place, but I > want to get the baseline architecture change out the door first. The REST interface is complete enough to let the mantis integration work. I've run a successful update from the last 'current' catalog to the latest 'unstable' one (over 500 changes) and it's working well. I still need to get this a home inside www-mockup[1], but for the time being I'll cron it on a private box and hopefully catch the little things I missed. Is there anyone that would be willing to inspect the captured outputs of the program to analyze the decisions made for correctness? If things go well, it could be made live in a week or so, I think... Thanks -Ben [1] Zone changes required there still. -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From maciej at opencsw.org Wed Oct 5 08:33:50 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 5 Oct 2011 07:33:50 +0100 Subject: [csw-maintainers] MySQL in /opt/csw available for testing Message-ID: I've been working on an overhaul of the MySQL 5.0 package in the last few days. I've come up with a way to support multiple versions installed at the same time, while the package is compiled in the /opt/csw prefix. The multi-version support is achieved by using --program-suffix and alternatives. For example, /opt/csw/bin/mysql is a symlink to mysql-5.0. I will probably introduce CSWmysql-common when I establish which exactly files can be shared between MySQL versions. The packages are available here: http://buildfarm.opencsw.org/experimental.html#mysql-5.0 ...and also in unstable. One bit of backward-compatibility is that localstatedir is still set to /var/opt/csw/mysql5, to preserve the existing installations. The upgrade should Just Work(tm). Maciej From maciej at opencsw.org Wed Oct 5 08:36:49 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 5 Oct 2011 07:36:49 +0100 Subject: [csw-maintainers] status on web/mantis integration In-Reply-To: <1317781747-sup-6712@pinkfloyd.chass.utoronto.ca> References: <1316956704-sup-5312@pinkfloyd.chass.utoronto.ca> <1317781747-sup-6712@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/5 Ben Walton : > Is there anyone that would be willing to inspect the captured outputs > of the program to analyze the decisions made for correctness? > > If things go well, it could be made live in a week or so, I think... I could take a look. Do you think on putting the log up somewhere for people to look up, or by giving people access to look on a live system? From maciej at opencsw.org Thu Oct 6 09:54:23 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 6 Oct 2011 08:54:23 +0100 Subject: [csw-maintainers] MySQL 5.1 progress Message-ID: I've spent some time on MySQL-5.0 lately, fixed problems with the shared libraries, and started looking at 5.1, since it uses an almost identical build recipe. There's one comical problem I can't quite figure out yet. During the install: Setting up the database Creating MySQL core database in /var/opt/csw/mysql/@MYSQL_BASE_VERSION@ When I look into the filesystem, the installer has indeed used literally "/var/opt/csw/mysql/@MYSQL_BASE_VERSION@": netra ~ # ls -l /var/opt/csw/mysql total 3 drwx------ 4 mysql mysql 4 Oct 6 09:33 @MYSQL_BASE_VERSION@ I tried to grep for MYSQL_BASE_VERSION in the whole installdir and pkgroot: maciej at unstable9s :~/src/opencsw/pkg/mysql5/branches/mysql-5.1.x-optcsw > ggrep -r MYSQL_BASE_VERSION work/solaris9-sparc/install-isa-sparcv8 work/solaris9-sparc/install-isa-sparcv8/opt/csw/include/mysql/mysql_version.h:#define MYSQL_BASE_VERSION "mysqld-5.1" maciej at unstable9s :~/src/opencsw/pkg/mysql5/branches/mysql-5.1.x-optcsw > ggrep -r MYSQL_BASE_VERSION work/solaris9-sparc/pkgroot work/solaris9-sparc/pkgroot/opt/csw/include/mysql/mysql_version.h:#define MYSQL_BASE_VERSION "mysqld-5.1" Nothing interesting in there. Do you have any ideas what could have caused this? If you want to test it out, build the pkg/mysql5/branches/mysql-5.1.x-optcsw/ branch. Also, binary packages may (I'm repeatedly deleting and updating them) be available from: http://buildfarm.opencsw.org/experimental.html#mysql-5.1 Maciej From maciej at opencsw.org Thu Oct 6 10:06:42 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 6 Oct 2011 09:06:42 +0100 Subject: [csw-maintainers] MySQL 5.1 progress In-Reply-To: References: Message-ID: 2011/10/6 Maciej (Matchek) Blizi?ski : > Creating MySQL core database in /var/opt/csw/mysql/@MYSQL_BASE_VERSION@ Found it: the postinstall script. From maciej at opencsw.org Thu Oct 6 11:37:43 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 6 Oct 2011 10:37:43 +0100 Subject: [csw-maintainers] The dublin release: remaining tasks Message-ID: During the Summercamp we agreed that what's left before the dublin release, is a couple of core libraries that need repackaging according to the (not so) new (anymore) policy. The libraries are: - openssl - openldap - libjpeg - sasl Let's use this thread to coordinate and track progress. The libraries above are either unmaintained, or their maintainers don't have the time to rework the build recipes. I looked at openssl. The first problem is that it doesn't build. Not sure whether it's something about my build evironment, or a problem with the build recipe itself. "gmake: *** No rule to make target `libssl.a', needed by `libssl.so.0.9.8'. Stop." Could anyone else try to build openssl and see if they get the same error? I know that Dago has made some progress on openldap and libjpeg. What's left is sasl. Could someone volunteer and start working on sasl? Even partial progress: "try to build, see what fails, and report back", would be helpful. Maciej From dam at opencsw.org Thu Oct 6 12:32:27 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 6 Oct 2011 12:32:27 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: References: Message-ID: Hi Maciej, Am 06.10.2011 um 11:37 schrieb Maciej (Matchek) Blizi?ski: > During the Summercamp we agreed that what's left before the dublin > release, is a couple of core libraries that need repackaging according > to the (not so) new (anymore) policy. The libraries are: > > - openssl > - openldap I made new recipes pkg/openldap23 pkg/openldap24 The 2.3 libs should link against 2.3 instead of 2.4 like they do at the moment (especially libldap-2.3 against liblber-2.4), openldap24 needs to be evaluated. > - libjpeg pkg/libjpeg7 is ready pkg/libjpeg62 doesn't build a shared library at the moment. > - sasl pkg/sasl should be almost done, but has gone through substantial reworking and hasn't been tested yet. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From dam at opencsw.org Thu Oct 6 12:34:18 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 6 Oct 2011 12:34:18 +0200 Subject: [csw-maintainers] gar problem In-Reply-To: <201110061350.48327.ai@vsu.ru> References: <201110061350.48327.ai@vsu.ru> Message-ID: <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> Hi Andy, (cc' maintainers@) please post build questions to maintainers@ so more people have a chance to answer :-) Am 06.10.2011 um 11:50 schrieb Andy Igoshin: > Hi! > > ... > ## Validating control scripts. > ## Packaging complete. > mkp: exec( pkgtrans -s /home/ai/spool.5.10-sparc > /tmp/nginx-1.1.5,REV=2011.10.05-SunOS5.10-sparc-CSW.pkg CSWnginx ) > Transferring package instance > mkp: exec( gzip -9 -f /tmp/nginx-1.1.5,REV=2011.10.05-SunOS5.10-sparc-CSW.pkg ) > mkp: exec( mv /tmp/nginx-1.1.5,REV=2011.10.05-SunOS5.10-sparc-CSW.pkg.gz > /home/ai/staging/build-05.Oct.2011 ) > mkp: exec( rm -rf /home/ai/spool.5.10-sparc/CSWnginx ) > INFO:root:Juicing the srv4 package stream files... > 100% | > ############################################################################################################################| > CRITICAL:root:Fetching from the db has failed: catrel='unstable1' > CRITICAL:root:Available catalog releases: > CRITICAL:root: - current > CRITICAL:root: - dublin > CRITICAL:root: - unstable > CRITICAL:root: - stable > Traceback (most recent call last): > File "gar//bin/checkpkg", line 188, in > main() > File "gar//bin/checkpkg", line 127, in main > sqo_catrel = models.CatalogRelease.selectBy(name=options.catrel).getOne() > File "/opt/csw/lib/python/site-packages/sqlobject/sresults.py", line 280, in > getOne > % self.sourceClass.__name__) > sqlobject.main.SQLObjectNotFound: No results matched the query for > CatalogRelease > gmake: *** [pkgcheck] Error 2 > gmake: Leaving directory `/home/ai/mgar/pkg/nginx/trunk' > Connection to unstable10s closed. > gmake: *** [platforms] Error 2 I don't know where unstable1 comes from. Are you building on unstable* ? Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From dam at opencsw.org Thu Oct 6 12:39:01 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 6 Oct 2011 12:39:01 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: References: Message-ID: Hi, Am 06.10.2011 um 12:32 schrieb Dagobert Michelsen: > Am 06.10.2011 um 11:37 schrieb Maciej (Matchek) Blizi?ski: >> During the Summercamp we agreed that what's left before the dublin >> release, is a couple of core libraries that need repackaging according >> to the (not so) new (anymore) policy. The libraries are: >> >> - openssl >> - openldap > > I made new recipes > pkg/openldap23 > pkg/openldap24 > > The 2.3 libs should link against 2.3 instead of 2.4 like they do at the moment > (especially libldap-2.3 against liblber-2.4), openldap24 needs to be evaluated. > >> - libjpeg > > pkg/libjpeg7 is ready > pkg/libjpeg62 doesn't build a shared library at the moment. > >> - sasl > > pkg/sasl should be almost done, but has gone through substantial reworking and hasn't > been tested yet. ATTENTION: The current SASL is in pkg/sasl/branches/sasl-2.1.23-mGARv2 The one in pkg/sasl/trunk is old! Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From pfelecan at opencsw.org Thu Oct 6 14:36:44 2011 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 06 Oct 2011 14:36:44 +0200 Subject: [csw-maintainers] GCC 4.6 progress In-Reply-To: ("Maciej (Matchek) =?utf-8?Q?Blizi=C5=84ski=22's?= message of "Mon, 26 Sep 2011 08:28:09 +0100") References: Message-ID: "Maciej (Matchek) Blizi?ski" writes: > 2011/8/15 Peter FELECAN : >> Are you trying to build with Sun's C compiler? (I never succeeded to build >> gcc with that compiler) In the affirmative you'll have a heavy time >> generating the Ada compiler. > > This time has just begun! > > The story goes like this: I got PPL and Cloog in place and tried > compiling adding Ada to the list of language frontends. The GCC build > system complained that GNAT is missing. I thought I would build a > separate package with GNAT, but it turns out that it's built by GCC as > part of Ada, so there is a circular dependency: you need Ada to build > Ada. The best and easiest method to build gcc is to use gcc, especially if you wish to have an Ada frontend. This is why I asked if you build with Sun Studio or gcc. -- Peter From pfelecan at opencsw.org Thu Oct 6 15:55:16 2011 From: pfelecan at opencsw.org (Peter FELECAN) Date: Thu, 06 Oct 2011 15:55:16 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: ("Maciej (Matchek) =?utf-8?Q?Blizi=C5=84ski=22's?= message of "Thu, 6 Oct 2011 10:37:43 +0100") References: Message-ID: "Maciej (Matchek) Blizi?ski" writes: > During the Summercamp we agreed that what's left before the dublin > release Is not Kiel release now? the next one? -- Peter From maciej at opencsw.org Thu Oct 6 16:07:46 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 6 Oct 2011 15:07:46 +0100 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: References: Message-ID: 2011/10/6 Peter FELECAN : > "Maciej (Matchek) Blizi?ski" writes: > >> During the Summercamp we agreed that what's left before the dublin >> release > > Is not Kiel release now? the next one? We haven't finished dublin yet, but we're close. Kiel will comprise of the next set of goals, e.g. will be the first release with tiers. This is about closing off the current works. Maciej From raos at opencsw.org Thu Oct 6 16:27:45 2011 From: raos at opencsw.org (Rafael Ostertag) Date: Thu, 6 Oct 2011 16:27:45 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: References: Message-ID: <20111006142745.GA24334@bender.opencsw.org> Hi Maciej On Thu, Oct 06, 2011 at 10:37:43AM +0100, Maciej (Matchek) Blizi??ski wrote: > "gmake: *** No rule to make target `libssl.a', needed by > `libssl.so.0.9.8'. Stop." > > Could anyone else try to build openssl and see if they get the same error? I just built openssl 0.9.8r without a glitch on the build farm (unstable10x/s). Hope this helps... So long rafi From maciej at opencsw.org Thu Oct 6 17:09:52 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 6 Oct 2011 16:09:52 +0100 Subject: [csw-maintainers] GCC 4.6 progress In-Reply-To: References: Message-ID: 2011/10/6 Peter FELECAN : > The best and easiest method to build gcc is to use gcc, especially if > you wish to have an Ada frontend. This is why I asked if you build with > Sun Studio or gcc. Ada seems to be a special case, where you need ada to build it. I looked up build instructions, and the way to bootstrap Ada is to crosscompile. I talked with guys on #ada to find out whether there are any people who use both Ada and Solaris, and found two people on that channel. However, they generally do not use the Ada implementation from gcc. They develop at a project called dragonlace[1], which is upstream with relation to gcc. Source code merges from dragonlace to gcc are done at every minor version release, so gcc sources lag behind dragonlace. Problems: - Unstable has been already updated with the ada-less gcc-4.6.1 - Ada from gcc-3.x is not good for bootstrapping ada from gcc-4.x The plan: - use the current* hosts with gcc-4.3.3 with Ada to build an intermediate gcc-4.6.1 with Ada support - use the intermediate gcc-4.6.1 with Ada support to build the gcc-4.6.1 packages with GAR, on the unstable hosts - update unstable with the new gcc-4.6.1 with Ada - rebuild the gcc-4.6.1 packages with gcc-4.6.1. Maciej [1] http://www.dragonlace.net/ From dam at opencsw.org Fri Oct 7 18:25:51 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 7 Oct 2011 18:25:51 +0200 Subject: [csw-maintainers] gar problem In-Reply-To: <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> References: <201110061350.48327.ai@vsu.ru> <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> Message-ID: Hi folks, Am 06.10.2011 um 12:34 schrieb Dagobert Michelsen: > Am 06.10.2011 um 11:50 schrieb Andy Igoshin: >> Hi! >> >> ... >> ## Validating control scripts. >> ## Packaging complete. >> mkp: exec( pkgtrans -s /home/ai/spool.5.10-sparc >> /tmp/nginx-1.1.5,REV=2011.10.05-SunOS5.10-sparc-CSW.pkg CSWnginx ) >> Transferring package instance >> mkp: exec( gzip -9 -f /tmp/nginx-1.1.5,REV=2011.10.05-SunOS5.10-sparc-CSW.pkg ) >> mkp: exec( mv /tmp/nginx-1.1.5,REV=2011.10.05-SunOS5.10-sparc-CSW.pkg.gz >> /home/ai/staging/build-05.Oct.2011 ) >> mkp: exec( rm -rf /home/ai/spool.5.10-sparc/CSWnginx ) >> INFO:root:Juicing the srv4 package stream files... >> 100% | >> ############################################################################################################################| >> CRITICAL:root:Fetching from the db has failed: catrel='unstable1' >> CRITICAL:root:Available catalog releases: >> CRITICAL:root: - current >> CRITICAL:root: - dublin >> CRITICAL:root: - unstable >> CRITICAL:root: - stable >> Traceback (most recent call last): >> File "gar//bin/checkpkg", line 188, in >> main() >> File "gar//bin/checkpkg", line 127, in main >> sqo_catrel = models.CatalogRelease.selectBy(name=options.catrel).getOne() >> File "/opt/csw/lib/python/site-packages/sqlobject/sresults.py", line 280, in >> getOne >> % self.sourceClass.__name__) >> sqlobject.main.SQLObjectNotFound: No results matched the query for >> CatalogRelease >> gmake: *** [pkgcheck] Error 2 >> gmake: Leaving directory `/home/ai/mgar/pkg/nginx/trunk' >> Connection to unstable10s closed. >> gmake: *** [platforms] Error 2 > > I don't know where unstable1 comes from. Are you building on unstable* ? Has this issue been solved? Best regards -- Dago From maciej at opencsw.org Fri Oct 7 18:41:38 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Fri, 7 Oct 2011 17:41:38 +0100 Subject: [csw-maintainers] gar problem In-Reply-To: References: <201110061350.48327.ai@vsu.ru> <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> Message-ID: I don't think it was. The last change this part of the code was by Sebastian. I smell a bug in a regular expression. Should be easy to fix. Can be tested with mgar show-var CATALOG_RELEASE. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bwalton at opencsw.org Mon Oct 10 05:08:46 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 09 Oct 2011 23:08:46 -0400 Subject: [csw-maintainers] AP2_MODS in GAR Message-ID: <1318215396-sup-1414@pinkfloyd.chass.utoronto.ca> Hi Dago, [CC'ing maintainers@ for general interest in a new and likely not widely known GAR capability.] As per our irc conversation the other day, enabling automatic handling of apache2 modules in GAR is done like this: 1. Turn on AP2_MODS: AP2_MODS = 1 2. Set AP2_MODFILES to a pattern suitable for gfind[1]. To make this work for apache2's suexec module, my recipe contains: AP2_MODS = 1 AP2_MODFILES = *suexec*.so This passes *suexec*.so to $(GARBIN)/ap2mod_build_scripts which will take each file returned by gfind and create a new file in /opt/csw/apache2/ap2mod. This new file is a script that will be passed to the 'build' CAS at install time. The script calls apxs with various parameters to install the module, uninstall the module, etc. There are values in csw.conf that are honoured for 'install but leave disabled' giving the site admin some control of this process. There is a similar setup for php5 modules as well. Let me know if the package you were working on doesn't fit into this setup for some reason as I'd like to address any issues in GAR since I think this is a very useful centralized capability. Thanks -Ben [1] There is a default, but I just noticed that I didn't update this after changing the way AP2_MODS worked. The first attempt used a separate CAS and consumed this directly...it's now passed to gfind instead. I'll fix this up shortly. -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From skayser at opencsw.org Mon Oct 10 11:09:26 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 10 Oct 2011 11:09:26 +0200 Subject: [csw-maintainers] gar problem In-Reply-To: <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> References: <201110061350.48327.ai@vsu.ru> <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> Message-ID: <20111010090926.GG26160@sebastiankayser.de> Hi Andy, * Dagobert Michelsen wrote: > (cc' maintainers@) please post build questions to maintainers@ so more > people have a chance to answer :-) > > Am 06.10.2011 um 11:50 schrieb Andy Igoshin: > > CRITICAL:root:Fetching from the db has failed: catrel='unstable1' > > CRITICAL:root:Available catalog releases: > > CRITICAL:root: - current > > CRITICAL:root: - dublin > > CRITICAL:root: - unstable > > CRITICAL:root: - stable should be fixed with gar/v2 r15903. Sorry for the hickup. Sebastian From dam at opencsw.org Mon Oct 10 12:14:21 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 10 Oct 2011 12:14:21 +0200 Subject: [csw-maintainers] AP2_MODS in GAR In-Reply-To: <1318215396-sup-1414@pinkfloyd.chass.utoronto.ca> References: <1318215396-sup-1414@pinkfloyd.chass.utoronto.ca> Message-ID: Hi Ben, Am 10.10.2011 um 05:08 schrieb Ben Walton: > [CC'ing maintainers@ for general interest in a new and likely not > widely known GAR capability.] > > As per our irc conversation the other day, enabling automatic handling > of apache2 modules in GAR is done like this: > > 1. Turn on AP2_MODS: AP2_MODS = 1 > 2. Set AP2_MODFILES to a pattern suitable for gfind[1]. > > To make this work for apache2's suexec module, my recipe contains: > > AP2_MODS = 1 > AP2_MODFILES = *suexec*.so I needed \*auth_kerb\*so but it works now. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From ai at opencsw.org Mon Oct 10 13:32:39 2011 From: ai at opencsw.org (Andy Igoshin) Date: Mon, 10 Oct 2011 15:32:39 +0400 Subject: [csw-maintainers] gar problem In-Reply-To: <20111010090926.GG26160@sebastiankayser.de> References: <201110061350.48327.ai@vsu.ru> <13BAF644-CF9C-4578-AE85-D4E907940D27@opencsw.org> <20111010090926.GG26160@sebastiankayser.de> Message-ID: <201110101532.39269.ai@opencsw.org> Hi! On 10 of October 2011 13:09:26 Sebastian Kayser wrote: > Hi Andy, > > * Dagobert Michelsen wrote: > > (cc' maintainers@) please post build questions to maintainers@ so more > > people have a chance to answer :-) > > > > Am 06.10.2011 um 11:50 schrieb Andy Igoshin: > > > CRITICAL:root:Fetching from the db has failed: catrel='unstable1' > > > CRITICAL:root:Available catalog releases: > > > CRITICAL:root: - current > > > CRITICAL:root: - dublin > > > CRITICAL:root: - unstable > > > CRITICAL:root: - stable > > should be fixed with gar/v2 r15903. Sorry for the hickup. yes, it works. thanks! > Sebastian -- Andy Igoshin Voronezh State University Phone: +7 473 2522406 Network Operation Center Fax: +7 473 2208820 Voronezh, Russia From maciej at opencsw.org Wed Oct 12 13:24:56 2011 From: maciej at opencsw.org (Maciej (Matchek) Blizinski) Date: Wed, 12 Oct 2011 12:24:56 +0100 Subject: [csw-maintainers] FOSDEM 2012 Message-ID: Is anyone going to FOSDEM 2012? I'm thinking about it. We talked about it on #opencsw, there is an idea to get an OpenCSW stand there. Any thoughts, attendees, stand volunteers? Maciej From dam at opencsw.org Wed Oct 12 13:34:33 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 12 Oct 2011 13:34:33 +0200 Subject: [csw-maintainers] FOSDEM 2012 In-Reply-To: References: Message-ID: <39571E7F-9E76-4877-8774-538325834E88@opencsw.org> Hi Maciej, Am 12.10.2011 um 13:24 schrieb Maciej (Matchek) Blizinski: > Is anyone going to FOSDEM 2012? I'm thinking about it. > > We talked about it on #opencsw, there is an idea to get an OpenCSW stand there. > > Any thoughts, attendees, stand volunteers? I you get us a stand I'll bring a bag of Cashew Nuts and shirts :-) Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From maciej at opencsw.org Thu Oct 13 09:56:23 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 13 Oct 2011 08:56:23 +0100 Subject: [csw-maintainers] FOSDEM 2012 In-Reply-To: <39571E7F-9E76-4877-8774-538325834E88@opencsw.org> References: <39571E7F-9E76-4877-8774-538325834E88@opencsw.org> Message-ID: 2011/10/12 Dagobert Michelsen : > I you get us a stand I'll bring a bag of Cashew Nuts and shirts :-) I suppose that all we need is some kind of display stand, something along the lines of: http://www.easydisplay.ie/banner-display-signs-stands.html If I were to bring it, I would need something that I can take on the plane. The rest is us sitting at a table and talking to people. Maciej From raos at opencsw.org Thu Oct 13 11:01:21 2011 From: raos at opencsw.org (Rafael Ostertag) Date: Thu, 13 Oct 2011 11:01:21 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: References: Message-ID: <20111013090121.GA15669@bender.opencsw.org> Hi Maciej On Thu, Oct 06, 2011 at 10:37:43AM +0100, Maciej (Matchek) Blizi??ski wrote: [...] > I know that Dago has made some progress on openldap and libjpeg. > What's left is sasl. Could someone volunteer and start working on > sasl? Even partial progress: "try to build, see what fails, and > report back", would be helpful. Did anybody volunteer for sasl yet? If not, I could help out. cheers rafi From maciej at opencsw.org Thu Oct 13 11:20:50 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 13 Oct 2011 10:20:50 +0100 Subject: [csw-maintainers] [checkpkg] Check unit test interface Message-ID: If you don't care about checks in checkpkg, you can stop reading now. I recently attended PyCon Ireland 2011, and one of the talks inspired me to improve the check unit test interface in checkpkg. It's the "with" statement, which allows to easily surround a block of code with e.g. setup and teardown code. It's commonly used for opening files, but I found a way to use it in unit testing. In checkpkg unit tests, if you want to check for 2 conditions, you need to write two test functions. Currently, to write one test function, you need to create one unit test class. You can define more than one function if you want to, but then you have to explicitly register each new test function. An example of an old check unit test: http://pastie.org/2687765 In the new helper class, you use the "with" statement, which takes care of setup, run, and teardown of each test. An example of a new check unit test: http://pastie.org/2687767 The main advantage I see is that it's easier to group test functions in classes without worrying about registering them - the abstraction is cleaner that way. If you want to comment on any specific line in the code: http://codereview.appspot.com/5275041/ (click "view side-by-side diffs") Any thoughts? Maciej From maciej at opencsw.org Thu Oct 13 11:26:39 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Thu, 13 Oct 2011 10:26:39 +0100 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: <20111013090121.GA15669@bender.opencsw.org> References: <20111013090121.GA15669@bender.opencsw.org> Message-ID: 2011/10/13 Rafael Ostertag : >> I know that Dago has made some progress on openldap and libjpeg. >> What's left is sasl. ?Could someone volunteer and start working on >> sasl? ?Even partial progress: "try to build, see what fails, and >> report back", would be helpful. > Did anybody volunteer for sasl yet? If not, I could help out. Dago did some work, but he only did the first 90%, so there's the second 90% left to go. Dago is generally overloaded, so any help would be greatly appreciated. Everything is committed to the repository, so you can check out the sources and pick up where Dago left off. One thing I noticed is that it still depends on CSWkrb5lib, which is now an empty package, so the dependencies should be shifted to specific kerberos libraries. Maciej From raos at opencsw.org Thu Oct 13 12:17:14 2011 From: raos at opencsw.org (Rafael Ostertag) Date: Thu, 13 Oct 2011 12:17:14 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: References: <20111013090121.GA15669@bender.opencsw.org> Message-ID: <20111013101714.GB15669@bender.opencsw.org> On Thu, Oct 13, 2011 at 10:26:39AM +0100, Maciej (Matchek) Blizi??ski wrote: > 2011/10/13 Rafael Ostertag : > >> I know that Dago has made some progress on openldap and libjpeg. > >> What's left is sasl. ??Could someone volunteer and start working on > >> sasl? ??Even partial progress: "try to build, see what fails, and > >> report back", would be helpful. > > Did anybody volunteer for sasl yet? If not, I could help out. > > Dago did some work, but he only did the first 90%, so there's the > second 90% left to go. Dago is generally overloaded, so any help > would be greatly appreciated. > > Everything is committed to the repository, so you can check out the > sources and pick up where Dago left off. One thing I noticed is that > it still depends on CSWkrb5lib, which is now an empty package, so the > dependencies should be shifted to specific kerberos libraries. > > Maciej Well, then it's settled. I'll take the second 90% ;) Thx for the advise on kerberos. cheers rafi From dam at opencsw.org Thu Oct 13 12:54:30 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 13 Oct 2011 12:54:30 +0200 Subject: [csw-maintainers] The dublin release: remaining tasks In-Reply-To: <20111013101714.GB15669@bender.opencsw.org> References: <20111013090121.GA15669@bender.opencsw.org> <20111013101714.GB15669@bender.opencsw.org> Message-ID: Hi Rafi, Am 13.10.2011 um 12:17 schrieb Rafael Ostertag: > On Thu, Oct 13, 2011 at 10:26:39AM +0100, Maciej (Matchek) Blizi??ski wrote: >> 2011/10/13 Rafael Ostertag : >>>> I know that Dago has made some progress on openldap and libjpeg. >>>> What's left is sasl. ??Could someone volunteer and start working on >>>> sasl? ??Even partial progress: "try to build, see what fails, and >>>> report back", would be helpful. >>> Did anybody volunteer for sasl yet? If not, I could help out. >> >> Dago did some work, but he only did the first 90%, so there's the >> second 90% left to go. Dago is generally overloaded, so any help >> would be greatly appreciated. >> >> Everything is committed to the repository, so you can check out the >> sources and pick up where Dago left off. One thing I noticed is that >> it still depends on CSWkrb5lib, which is now an empty package, so the >> dependencies should be shifted to specific kerberos libraries. > > Well, then it's settled. I'll take the second 90% ;) Thx for the advise on kerberos. Make sure to start in mgar/pkg/sasl/branches/sasl-2.1.23-mGARv2 as it is the most recent version. I guess you can just replace trunk with it. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From dam at opencsw.org Thu Oct 13 12:59:23 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 13 Oct 2011 12:59:23 +0200 Subject: [csw-maintainers] glib2: finally some results In-Reply-To: <20111009172646.GA25605@bender.opencsw.org> References: <20111009172646.GA25605@bender.opencsw.org> Message-ID: <85555CAF-0F97-422A-88FA-D91663ED540C@opencsw.org> Hi Rafi, Am 09.10.2011 um 19:26 schrieb Rafael Ostertag: > I just committed (r15901) the recipe for glib2 which I hope is deployment ready. > > However, I would appreciate if you could have a look at it and give it > a go/no-go. The recipe looks ok, I am fwd'ing to maintainers for a broader test audience. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From rupert at opencsw.org Sat Oct 15 19:13:52 2011 From: rupert at opencsw.org (rupert THURNER) Date: Sat, 15 Oct 2011 19:13:52 +0200 Subject: [csw-maintainers] website does not reflect real package versions ? Message-ID: to put "solaris package ..." on the package description text makes it finally easy to find opencsw packages on the web. unfortunately it seems not easy to find the current versions. e.g mercurial is still listet with a version 1.8.4 from june. there was one new package every month since then. what is missing currently so the versions are displayed correctly? who needs to do what? where could one help out? rupert From maciej at opencsw.org Sat Oct 15 19:35:50 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 15 Oct 2011 18:35:50 +0100 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: References: Message-ID: 2011/10/15 rupert THURNER : > what is missing currently so the versions are > displayed correctly? who needs to do what? where could one help out? The effort is mainly led by Ben, who currently has the best understanding of the infrastructure - all the databases and the data flow between them. Initial thread: http://lists.opencsw.org/pipermail/maintainers/2011-August/015144.html Last update: http://lists.opencsw.org/pipermail/maintainers/2011-October/015487.html The best way to get started would be talking to Ben. Maciej From bwalton at opencsw.org Sat Oct 15 20:53:01 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sat, 15 Oct 2011 14:53:01 -0400 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: References: Message-ID: <1318702007-sup-9282@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Sat Oct 15 13:35:50 -0400 2011: Hi All, > The effort is mainly led by Ben, who currently has the best > understanding of the infrastructure - all the databases and the data > flow between them. I'm hoping to deploy the mantis integration tonight which will close that gap. Then the remaining piece is getting the web display up to date. To complete that, we need extensions to the pkgdb REST interface. The last time we talked about this I was leaning towards using REST calls for all of the info which would let us abandon the front-end web db entirely. While I think that's still an ok approach (assuming the calls can supply data quickly and meet the current functionality), I'm not sure it's the fastest route. It might be, but I think it'll be more work initially...I'd rather get something running again and improve it later. >From where I sit right now, I think the only extension I'll need to the pkgdb REST interface is a clean way to pull a list of files associated with a package. I already have mechanisms to retrieve the other info I need from catalog file, descriptions file and REST interface. Maciej: Is anything other than interface code required to provide the new URL with the backend query to get a file list? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From skayser at opencsw.org Sun Oct 16 12:48:08 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 16 Oct 2011 12:48:08 +0200 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? Message-ID: <20111016104807.GK26160@sebastiankayser.de> Hi, currently we have opencsw.org/bugtrack opencsw.org/mantis as entry points to our bugtracker. Can we also get bugs.opencsw.org opencsw.org/bugs similar to what Debian has? I'd like to make it easy as possible for people to memorize the entry point URLs and land on our bug tracker. Sebastian From skayser at opencsw.org Sun Oct 16 13:19:32 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 16 Oct 2011 13:19:32 +0200 Subject: [csw-maintainers] How do you determine archive URLs for list emails? Message-ID: <20111016111932.GN26160@sebastiankayser.de> Hi, how do you guys determine the list archive URL for emails on this list? I often refer people to our archives and my somehow tedious workflow looks like this: * Find relevant email/thread topic in mutt * Search URL via google: "site:lists.opencsw.org $topic" The second step is what bugs me. Could mailman include the archive URL in the email headers (currently, it doesn't seem to do)? Then, a MUA search would be sufficient. Sebastian From skayser at opencsw.org Sun Oct 16 13:21:21 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 16 Oct 2011 13:21:21 +0200 Subject: [csw-maintainers] gar.opencsw.org: spam on our trac wiki In-Reply-To: <1D482DF1-A01E-476B-8A4A-5B92742B3244@opencsw.org> References: <20111002131908.GD26160@sebastiankayser.de> <1D482DF1-A01E-476B-8A4A-5B92742B3244@opencsw.org> Message-ID: <20111016112121.GO26160@sebastiankayser.de> * Dagobert Michelsen wrote: > Am 02.10.2011 um 15:19 schrieb Sebastian Kayser: > > someone left a steakhouse ad link on our GAR wiki front page the other > > day (already deleted). In case this re-occurs, can someone keep an eye > > on it (and if it starts to get pesky, think about tweaking the wiki > > permissions)? I'll be away for a week without internet access. > > He did it again. I opened a ticket for abuse for user "jauhis" and kept the > history for the support to see. Thanks! Seems there were no more occurences since then. Sebastian From skayser at opencsw.org Sun Oct 16 13:26:36 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 16 Oct 2011 13:26:36 +0200 Subject: [csw-maintainers] mgar 0.94-rc1 - changes and call for testing In-Reply-To: References: <20110926184531.GH21825@sebastiankayser.de> Message-ID: <20111016112636.GP26160@sebastiankayser.de> Hi Maciej, * Maciej (Matchek) Blizi??ski wrote: > The newpkg functionality still doesn't work for me: mgar 0.94 is already out of the door. I'll look into your problem today and work on a fix for 0.95. Will provide an update later on. Sebastian From bwalton at opencsw.org Sun Oct 16 14:52:08 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 16 Oct 2011 08:52:08 -0400 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? In-Reply-To: <20111016104807.GK26160@sebastiankayser.de> References: <20111016104807.GK26160@sebastiankayser.de> Message-ID: <1318769511-sup-2366@pinkfloyd.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Sun Oct 16 06:48:08 -0400 2011: > bugs.opencsw.org > opencsw.org/bugs > > similar to what Debian has? I'd like to make it easy as possible for > people to memorize the entry point URLs and land on our bug tracker. +1 for this. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From rupert at opencsw.org Sun Oct 16 21:01:25 2011 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 16 Oct 2011 21:01:25 +0200 Subject: [csw-maintainers] mgar usage? Message-ID: i tried: rupert @ unstable9s : ~/mgar-sav/pkg/lang-python/python/branches/python-3.2 $ mgar makesum ERROR: No build system found at /.buildsys/v2 what is wrong here? rupert. -------------- next part -------------- An HTML attachment was scrubbed... URL: From skayser at opencsw.org Sun Oct 16 22:14:53 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 16 Oct 2011 22:14:53 +0200 Subject: [csw-maintainers] mgar usage? In-Reply-To: References: Message-ID: <20111016201453.GQ26160@sebastiankayser.de> Hi Rupert, * rupert THURNER wrote: > i tried: > rupert @ unstable9s : > ~/mgar-sav/pkg/lang-python/python/branches/python-3.2 > $ mgar makesum > ERROR: No build system found at /.buildsys/v2 > what is wrong here? a bit of context would help. Are you using mgar for the first time? If yes, then you need to initialize its configuration first. A check to catch this and to print out instructions is on my todo list. In the meantime please see [1]. Does this help? Sebastian [1] http://wiki.opencsw.org/gar-wrapper#toc2 (see option 2) From skayser at opencsw.org Sun Oct 16 23:24:29 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 16 Oct 2011 23:24:29 +0200 Subject: [csw-maintainers] mgar 0.94-rc1 - changes and call for testing In-Reply-To: References: <20110926184531.GH21825@sebastiankayser.de> Message-ID: <20111016212429.GR26160@sebastiankayser.de> Hi Maciej, * Maciej (Matchek) Blizi??ski wrote: > The newpkg functionality still doesn't work for me: > > maciej at login [login]:~ > ssh unstable9s > Last login: Mon Oct 3 00:33:39 2011 from login.bo.opencs > Sun Microsystems Inc. SunOS 5.9 Generic May 2002 > maciej at unstable9s :~ > cd src/opencsw > maciej at unstable9s :~/src/opencsw > mgar newpkg foo > ERROR: Template /home/maciej/src/opencsw/newpkg-Makefile not found > Likely fix: > (cd /home/maciej/src/opencsw && svn up newpkg-Makefile) > maciej at unstable9s :~/src/opencsw > (cd /home/maciej/src/opencsw && svn > up newpkg-Makefile) > Updating 'newpkg-Makefile': > At revision 15838. seems I was blind and svn didn't make it better. The notion of package tree root has changed with mgar. Prior to mgar it was based on our repository layout and the top level contained gar/ and pkg/. With mgar the package tree root now _is_ the pkg/ directory. To make a long story short, your ~/.garrc BUILDTREE currently points to ~/src/opencsw where in fact it needs to point to ~/src/opencsw/pkg. Guess this mis-configuration slipped in during the early days of mgar usage. [+] Just move ~/src/opencsw/.buildsys to ~/src/opencsw/pkg/.buildsys, adjust BUILDTREE, and you should be ready to go. There might be hickups with newpkg, this area is subject to fixes and enhancements. Comments and feature requests welcome. Sebastian [+] Funny thing here: the repo root doesn't contain newpkg-Makefile and svn happily runs svn up on a non-existing file without complaints. From maciej at opencsw.org Mon Oct 17 09:58:47 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 17 Oct 2011 08:58:47 +0100 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: <1318702007-sup-9282@pinkfloyd.chass.utoronto.ca> References: <1318702007-sup-9282@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/15 Ben Walton : > Maciej: Is anything other than interface code required to provide the > ? ? ? ?new URL with the backend query to get a file list? There is one relatively small obstacle - the timestamp serialization. Currently, the pickled version serializes a Python object instead of a portable string representation. When that's done, we can serialize all package metadata to JSON and send them via REST. Here's one starting point: GetMtime(). http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/lib/python/package_stats.py#L191 The fix would be to make sure that the string representation is used when serializing, and conversions to datetime.datetime are done when necessary. Maciej From dam at opencsw.org Mon Oct 17 12:42:48 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 17 Oct 2011 12:42:48 +0200 Subject: [csw-maintainers] How do you determine archive URLs for list emails? In-Reply-To: <20111016111932.GN26160@sebastiankayser.de> References: <20111016111932.GN26160@sebastiankayser.de> Message-ID: <7916F0BC-D17B-470A-A748-A6BFB10158F9@opencsw.org> Hi Sebastian, Am 16.10.2011 um 13:19 schrieb Sebastian Kayser: > how do you guys determine the list archive URL for emails on this list? > I often refer people to our archives and my somehow tedious workflow > looks like this: > > * Find relevant email/thread topic in mutt > * Search URL via google: "site:lists.opencsw.org $topic" I usually go to http://lists.opencsw.org/pipermail/maintainers/ and look for the specific date of the message found in the previous step. > The second step is what bugs me. Could mailman include the archive URL > in the email headers (currently, it doesn't seem to do)? Then, a MUA > search would be sufficient. Or the footer? Is this at all possible? Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From dam at opencsw.org Mon Oct 17 12:47:08 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 17 Oct 2011 12:47:08 +0200 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? In-Reply-To: <20111016104807.GK26160@sebastiankayser.de> References: <20111016104807.GK26160@sebastiankayser.de> Message-ID: <29E99DC4-611C-40CA-A5AD-6F57FD2224ED@opencsw.org> Hi Sebastian, Am 16.10.2011 um 12:48 schrieb Sebastian Kayser: > currently we have > > opencsw.org/bugtrack > opencsw.org/mantis > > as entry points to our bugtracker. Can we also get > > bugs.opencsw.org This used to link to the "new" JIRA bugtracker until I turned the VM off. > opencsw.org/bugs Should this be a redirect? Or opencsw.org/bugs/id/(\d+) -> http://www.opencsw.org/mantis/view.php?id=$1 opencsw.org/bugs/maintainer/(\s+) -> http://www.opencsw.org/buglist/maintainer.cgi?maintainer='$1' opencsw.org/bugs/pkg/(\s+) -> http://www.opencsw.org/mantis/set_project.php?project_id= > similar to what Debian has? I'd like to make it easy as possible for > people to memorize the entry point URLs and land on our bug tracker. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From skayser at opencsw.org Mon Oct 17 13:48:12 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 17 Oct 2011 13:48:12 +0200 Subject: [csw-maintainers] How do you determine archive URLs for list emails? In-Reply-To: <7916F0BC-D17B-470A-A748-A6BFB10158F9@opencsw.org> References: <20111016111932.GN26160@sebastiankayser.de> <7916F0BC-D17B-470A-A748-A6BFB10158F9@opencsw.org> Message-ID: <20111017114812.GV26160@sebastiankayser.de> * Dagobert Michelsen wrote: > Am 16.10.2011 um 13:19 schrieb Sebastian Kayser: > > how do you guys determine the list archive URL for emails on this list? > > I often refer people to our archives and my somehow tedious workflow > > looks like this: > > > > * Find relevant email/thread topic in mutt > > * Search URL via google: "site:lists.opencsw.org $topic" > > I usually go to > http://lists.opencsw.org/pipermail/maintainers/ > and look for the specific date of the message found in the previous step. Okay, that's also still two steps then. One possibility to reduce it to 1 manual and 1 automated step could be to pipe the email in the MUA to a script that determines the URL. That's dependent on the MUA capabilities then, but still better than nothing. > > The second step is what bugs me. Could mailman include the archive URL > > in the email headers (currently, it doesn't seem to do)? Then, a MUA > > search would be sufficient. > > Or the footer? Is this at all possible? Yes, the footer would be even better. Don't think so though, searched the Mailman resources in the meantime and only found a wishlist item with that feature [1]. Sent an email to the mailman-users list to find out about the current state and other possibilites. Sebastian [1] http://www.gnu.org/s/mailman/todo.html (API between MM and archiver) From skayser at opencsw.org Mon Oct 17 14:10:56 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 17 Oct 2011 14:10:56 +0200 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? In-Reply-To: <29E99DC4-611C-40CA-A5AD-6F57FD2224ED@opencsw.org> References: <20111016104807.GK26160@sebastiankayser.de> <29E99DC4-611C-40CA-A5AD-6F57FD2224ED@opencsw.org> Message-ID: <20111017121056.GW26160@sebastiankayser.de> Hey Dago, * Dagobert Michelsen wrote: > Am 16.10.2011 um 12:48 schrieb Sebastian Kayser: > > currently we have > > > > opencsw.org/bugtrack > > opencsw.org/mantis > > > > as entry points to our bugtracker. Can we also get > > > > bugs.opencsw.org > > This used to link to the "new" JIRA bugtracker until I turned the VM off. I'd vote for changing bugs.opencsw.org into a redirect to opencsw.org/bugs/. Otherwise, having two similar URLs point to different systems might be confusing. If further efforts are made on $anotherbugtracker, use opencsw.org/bugs-$anotherbugtracker. > > opencsw.org/bugs > > Should this be a redirect? Or > opencsw.org/bugs/id/(\d+) -> http://www.opencsw.org/mantis/view.php?id=$1 > opencsw.org/bugs/maintainer/(\s+) -> http://www.opencsw.org/buglist/maintainer.cgi?maintainer='$1' > opencsw.org/bugs/pkg/(\s+) -> http://www.opencsw.org/mantis/set_project.php?project_id= Cool, that's a good starting point. I'd say both. opencsw.org/bugs/xxx as redirects for specific resources (like above) plus a temporary redirect from /bugs/ to the main $bugtracker page. With some time and ideas, /bugs/ can be changed from a redirect to a portal or search page. Can you access the pkg -> id mapping required for the third URL type from the apache config or how would you implement the redirection? Sebastian From dam at opencsw.org Mon Oct 17 14:21:28 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Mon, 17 Oct 2011 14:21:28 +0200 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? In-Reply-To: <20111017121056.GW26160@sebastiankayser.de> References: <20111016104807.GK26160@sebastiankayser.de> <29E99DC4-611C-40CA-A5AD-6F57FD2224ED@opencsw.org> <20111017121056.GW26160@sebastiankayser.de> Message-ID: <07E6688E-F467-43C1-B920-D3618E3DCC45@opencsw.org> Hi Sebastian, Am 17.10.2011 um 14:10 schrieb Sebastian Kayser: > * Dagobert Michelsen wrote: >> Am 16.10.2011 um 12:48 schrieb Sebastian Kayser: >>> currently we have >>> >>> opencsw.org/bugtrack >>> opencsw.org/mantis >>> >>> as entry points to our bugtracker. Can we also get >>> >>> bugs.opencsw.org >> >> This used to link to the "new" JIRA bugtracker until I turned the VM off. > > I'd vote for changing bugs.opencsw.org into a redirect to > opencsw.org/bugs/. Otherwise, having two similar URLs point to different > systems might be confusing. If further efforts are made on > $anotherbugtracker, use opencsw.org/bugs-$anotherbugtracker. > >>> opencsw.org/bugs >> >> Should this be a redirect? Or >> opencsw.org/bugs/id/(\d+) -> http://www.opencsw.org/mantis/view.php?id=$1 >> opencsw.org/bugs/maintainer/(\s+) -> http://www.opencsw.org/buglist/maintainer.cgi?maintainer='$1' >> opencsw.org/bugs/pkg/(\s+) -> http://www.opencsw.org/mantis/set_project.php?project_id= > > Cool, that's a good starting point. I'd say both. opencsw.org/bugs/xxx > as redirects for specific resources (like above) plus a temporary > redirect from /bugs/ to the main $bugtracker page. With some time and > ideas, /bugs/ can be changed from a redirect to a portal or search page. > > Can you access the pkg -> id mapping required for the third URL type > from the apache config or how would you implement the redirection? I guess a rewrite to a cgi generating the target would be best However, you need root on the webserver for this. Would you mind adding the above redirects yourself? Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From skayser at opencsw.org Mon Oct 17 14:26:10 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 17 Oct 2011 14:26:10 +0200 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? In-Reply-To: <07E6688E-F467-43C1-B920-D3618E3DCC45@opencsw.org> References: <20111016104807.GK26160@sebastiankayser.de> <29E99DC4-611C-40CA-A5AD-6F57FD2224ED@opencsw.org> <20111017121056.GW26160@sebastiankayser.de> <07E6688E-F467-43C1-B920-D3618E3DCC45@opencsw.org> Message-ID: <20111017122610.GX26160@sebastiankayser.de> * Dagobert Michelsen wrote: > >> Should this be a redirect? Or > >> opencsw.org/bugs/id/(\d+) -> http://www.opencsw.org/mantis/view.php?id=$1 > >> opencsw.org/bugs/maintainer/(\s+) -> http://www.opencsw.org/buglist/maintainer.cgi?maintainer='$1' > >> opencsw.org/bugs/pkg/(\s+) -> http://www.opencsw.org/mantis/set_project.php?project_id= > > > > Cool, that's a good starting point. I'd say both. opencsw.org/bugs/xxx > > as redirects for specific resources (like above) plus a temporary > > redirect from /bugs/ to the main $bugtracker page. With some time and > > ideas, /bugs/ can be changed from a redirect to a portal or search page. > > > > Can you access the pkg -> id mapping required for the third URL type > > from the apache config or how would you implement the redirection? > > I guess a rewrite to a cgi generating the target would be best > > However, you need root on the webserver for this. Would you mind adding the above > redirects yourself? Given root privileges, some time, and local testing, sure. Sebastian From skayser at opencsw.org Mon Oct 17 18:37:52 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Mon, 17 Oct 2011 18:37:52 +0200 Subject: [csw-maintainers] Can we get bugs.opencsw.org & opencsw.org/bugs (redirects to Mantis)? In-Reply-To: <20111017122610.GX26160@sebastiankayser.de> References: <20111016104807.GK26160@sebastiankayser.de> <29E99DC4-611C-40CA-A5AD-6F57FD2224ED@opencsw.org> <20111017121056.GW26160@sebastiankayser.de> <07E6688E-F467-43C1-B920-D3618E3DCC45@opencsw.org> <20111017122610.GX26160@sebastiankayser.de> Message-ID: <20111017163752.GY26160@sebastiankayser.de> * Sebastian Kayser wrote: > * Dagobert Michelsen wrote: > > I guess a rewrite to a cgi generating the target would be best > > > > However, you need root on the webserver for this. Would you mind > > adding the above redirects yourself? > > Given root privileges, some time, and local testing, sure. Forget about the privileges, there's already sudo for me. I'll take care of the redirects. Can you point the DNS record for bugs.opencsw.org to ocswchbiesv01.opencsw.org? Sebastian From bwalton at opencsw.org Mon Oct 17 21:16:04 2011 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 17 Oct 2011 15:16:04 -0400 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: References: <1318702007-sup-9282@pinkfloyd.chass.utoronto.ca> Message-ID: <1318878870-sup-8026@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Mon Oct 17 03:58:47 -0400 2011: Hi Maciej, > There is one relatively small obstacle - the timestamp > serialization. For the web display, all I need is the path names and the package to which they belong. I think I could actually get the bulk of the code running without this and then add the file update feature when it's ready. I'll look at that tonight right after I finalize the mantis stuff[1]. Thanks -Ben [1] The mantis deploy didn't happen the other night as I didn't have the access I thought I had. Ihsan graciously fixed me up this morning so I'm good to go now. -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From maciej at opencsw.org Mon Oct 17 21:23:27 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 17 Oct 2011 20:23:27 +0100 Subject: [csw-maintainers] GCC 4.6 progress In-Reply-To: References: Message-ID: 2011/10/6 Maciej (Matchek) Blizi?ski : > The plan: > - use the current* hosts with gcc-4.3.3 with Ada to build an > intermediate gcc-4.6.1 with Ada support > - use the intermediate gcc-4.6.1 with Ada support to build the > gcc-4.6.1 packages with GAR, on the unstable hosts > - update unstable with the new gcc-4.6.1 with Ada > - rebuild the gcc-4.6.1 packages with gcc-4.6.1. I'm switching from packaging work to checkpkg / pkgdb development. We cannot promote gcc-4.6.1 to dublin before we build ada. Would anyone be willing to build an intermediate gcc-4.6.1 on current* hosts with ada support? What is needed, is a working gcc-4.6.1 with ada, with any prefix (e.g. --prefix=$HOME/usr), without any slicing to packages. The only function of this gcc will be to use it to build the target gcc-4.6.1. This intermediate gcc is a one-off thing, doesn't have to be built with GAR. It just needs to work and to bootstrap gcc-4.6. on unstable* hosts once. Then our GAR build will be self sufficient. Are there any takers? Maciej From maciej at opencsw.org Mon Oct 17 21:25:57 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 17 Oct 2011 20:25:57 +0100 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: <1318878870-sup-8026@pinkfloyd.chass.utoronto.ca> References: <1318702007-sup-9282@pinkfloyd.chass.utoronto.ca> <1318878870-sup-8026@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/17 Ben Walton : > For the web display, all I need is the path names and the package to > which they belong. I'l see if I can get this working. What do you want to make the queries by? I guess you start with an md5 sum and you get the list of files - right? Maciej From bwalton at opencsw.org Mon Oct 17 21:33:28 2011 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 17 Oct 2011 15:33:28 -0400 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: References: <1318702007-sup-9282@pinkfloyd.chass.utoronto.ca> <1318878870-sup-8026@pinkfloyd.chass.utoronto.ca> Message-ID: <1318879975-sup-8543@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Mon Oct 17 15:25:57 -0400 2011: > I'l see if I can get this working. What do you want to make the > queries by? I guess you start with an md5 sum and you get the list of > files - right? That would be perfect for my needs. Everything else I'm querying is keyed that way and it removes the whole catalog vs package issue from the fetch perspective. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From bwalton at opencsw.org Tue Oct 18 04:00:36 2011 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 17 Oct 2011 22:00:36 -0400 Subject: [csw-maintainers] mantis updates are running... Message-ID: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> Hi All, I've run the initial mantis integration run on the production database just now and set up the cron job to run twice daily at 10am and 10pm (EST; the crontab in CEST on the server is for 4am/4pm). I'll be getting the verbose output by mail as well as logging it. Prior to each run, a mysqldump is captured. (I'll disable this after a while.) I'd appreciate it if you'd check your bug list via the cgi script[1] and the list of packages attributed to you in the 'My Account' view of mantis. Let me know if you see any of the following: 1. Missing packages. 2. Extra packages. 3. Missing bugs. 4. Extra bugs. I'm keying the updates on the {unstable,5.9,sparc} catalog, so as you push via csw-upload-pkg, you can expect to see the changes after a) the next catalog update and b) the subsequent run of the new mantis integration too. Let me know if you have any questions about this. I'm now moving on to the web db integration. Thanks -Ben [1] http://www.opencsw.org/buglist/maintainer.cgi?maintainer='$username' -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From dam at opencsw.org Tue Oct 18 09:26:41 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Tue, 18 Oct 2011 09:26:41 +0200 Subject: [csw-maintainers] mantis updates are running... In-Reply-To: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> References: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> Message-ID: Hi Ben, Am 18.10.2011 um 04:00 schrieb Ben Walton: > I've run the initial mantis integration run on the production database > just now and set up the cron job to run twice daily at 10am and 10pm > (EST; the crontab in CEST on the server is for 4am/4pm). > > I'll be getting the verbose output by mail as well as logging it. > Prior to each run, a mysqldump is captured. (I'll disable this after > a while.) > > I'd appreciate it if you'd check your bug list via the cgi script[1] > and the list of packages attributed to you in the 'My Account' view of > mantis. > > Let me know if you see any of the following: > > 1. Missing packages. > 2. Extra packages. Tons of new packages on my buglist, but looks like I submitted them some time ago. Time to close bugs, I'm happy :-) Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From skayser at opencsw.org Tue Oct 18 13:37:45 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 18 Oct 2011 13:37:45 +0200 Subject: [csw-maintainers] mantis updates are running... In-Reply-To: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> References: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> Message-ID: <20111018113745.GA26160@sebastiankayser.de> * Ben Walton wrote: > I've run the initial mantis integration run on the production database > just now and set up the cron job to run twice daily at 10am and 10pm > (EST; the crontab in CEST on the server is for 4am/4pm). Cool! Another piece is of the puzzle getting ready. > I'd appreciate it if you'd check your bug list via the cgi script[1] > and the list of packages attributed to you in the 'My Account' view of > mantis. > > Let me know if you see any of the following: > > 1. Missing packages. > 2. Extra packages. > 3. Missing bugs. > 4. Extra bugs. >From looking at "My Account", there's an extra package: dovecot_sieve. Jake took over Dovecot and merged the _sieve package into the main package. dovecot_sieve shouldn't exist any more. Out of curiousity, what could cause bugs to go missing or extra ones to appear (does the integration script move bugs between projects)? Sebastian From bwalton at opencsw.org Tue Oct 18 15:25:53 2011 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 18 Oct 2011 09:25:53 -0400 Subject: [csw-maintainers] mantis updates are running... In-Reply-To: <20111018113745.GA26160@sebastiankayser.de> References: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> <20111018113745.GA26160@sebastiankayser.de> Message-ID: <1318944176-sup-8248@pinkfloyd.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Tue Oct 18 07:37:45 -0400 2011: Hi Sebastian, > From looking at "My Account", there's an extra package: dovecot_sieve. > Jake took over Dovecot and merged the _sieve package into the main > package. dovecot_sieve shouldn't exist any more. I'll look at this tonight. > Out of curiousity, what could cause bugs to go missing or extra ones to > appear (does the integration script move bugs between projects)? In the case of a rename, if the old project already existed, I move bugs from one project to another. Thus, if there was a rename that should have seen bugs collapsed into a single catalog name but didn't or if it collapsed them when it shouldn't have, that could result in missing or extra bugs. All of the actions taken by the script are intended to do as little harm as possible and I think moving bugs is likely the most invasive change. Removing a project is an UPDATE querty that toggles the state to inactive rather than running a DELETE, for example. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From maciej at opencsw.org Wed Oct 19 00:50:47 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 18 Oct 2011 23:50:47 +0100 Subject: [csw-maintainers] mgar 0.94-rc1 - changes and call for testing In-Reply-To: <20111016212429.GR26160@sebastiankayser.de> References: <20110926184531.GH21825@sebastiankayser.de> <20111016212429.GR26160@sebastiankayser.de> Message-ID: 2011/10/16 Sebastian Kayser : > Just move ~/src/opencsw/.buildsys to ~/src/opencsw/pkg/.buildsys, adjust > BUILDTREE, and you should be ready to go. It worked! Thanks. From bwalton at opencsw.org Wed Oct 19 05:31:44 2011 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 18 Oct 2011 23:31:44 -0400 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: References: Message-ID: <1318994869-sup-3633@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Sat Oct 15 13:35:50 -0400 2011: > > what is missing currently so the versions are > > displayed correctly? who needs to do what? where could one help out? > > The effort is mainly led by Ben, who currently has the best > understanding of the infrastructure - all the databases and the data > flow between them. I've written the web integration code tonight and the full run from the original catalog (last time Phil's scripts were run) is being fed through it right now. It looks pretty sound so far but I'll need to see what the logs look like tomorrow. As previously discussed, I still need a way to get the list of files in a package (we didn't register directories previously) and if anyone still cares for the feature[1] the shared objects the package relies on, but the basics should be ready to go... Thanks -Ben [1] I've never used this and don't really see the point of having it on the website but others may. I'd rather spend the time hacking it out of the web display code than jumping through hoops to keep it. Feel free to chime in with an alternate viewpoint though. -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From rupert at opencsw.org Wed Oct 19 09:16:53 2011 From: rupert at opencsw.org (rupert THURNER) Date: Wed, 19 Oct 2011 09:16:53 +0200 Subject: [csw-maintainers] Fwd: complicated build, subversion client In-Reply-To: References: <20110529181632.GA9828@ted.stsp.name> <20110529223727.GB9828@ted.stsp.name> <1306800816-sup-4668@pinkfloyd.chass.utoronto.ca> Message-ID: as the python bindings did not compile for more than 6 months, I commented them for the first try of svn-1.7. but I d have a bad feeling if we let this package further than unstable. any help to sort out that build error is greatly appreciated. rupert On Jun 2, 2011 10:06 PM, "rupert THURNER" wrote: > On Tue, May 31, 2011 at 02:25, Ben Walton wrote: > >> Excerpts from rupert THURNER's message of Mon May 30 16:08:35 -0400 2011: >> >> Hi Rupert, >> >> > You could separate out svnserve and other repository administration >> > *binaries* (svnadmin, svnlook, svndumpfilter) into a separate >> > package. >> >> This sounds like a good idea. I've seen a few bug requests for it. > > > stefan had the the convincing argument that we won't save any space on this > as we need to deliver all the dependencies anyway as file:// needs to be > supported see "libraries" below. > > my personal favourite as svn client currently is git. no svn necessary at > all. just three commands are necessary: > git svn clone https://whatever > git svn rebase # update > git svn dcommit # commit > > > With the svn libraries it's a different story. You will want all of >> > them on both the client and server. This means you'll need most (and >> > probably all) dependencies on both sides. >> >> This would be a good subversion-common package. Both the server and >> client packages could then depend on it. >> >> > I suppose the LDAP dependecy comes from berkeley DB? Subversion has >> > no direct dependency on LDAP. >> >> Does checkpkg tell you it's a surplus dep? If so, you could just drop >> it. It may be pulled in via something else and not actually be >> required. I haven't looked at it though. >> >> > You could also take a look at how the Debian packages for Subversion >> > split things up. They probably have things split up as much as >> > reasonably possible. >> >> It looks like a fairly monolithic setup there. >> >> > The files that don't compile with SUN's cc are generated by SWIG. >> > Can you figure out if it's possible to make SWIG generate code that >> > compiles with gcc as well as Sun's cc? If so we might be able to fix >> > your problem by changing the way we generate these files. >> >> Might John's updated swig packages alleviate this? > > >> > >> http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/subversion/trunk/files/0001-make-subversion-sysconfigdir-as-it-should-be-for-csw.patch >> > >> > That patch looks fine. If there is a pre-processor macro that >> > identifies a CWS build we could also include this change upstream >> > wrapped in the right #ifdef. >> >> Aside from the fact that this should be handled by autoconf (or >> whatever build system they're using) instead of the way they're doing >> it, I agree that the current patch shouldn't be too onerous to >> support. If they use autoconf and you'd like a hand building a patch >> for that to make this sane, let me know. >> >> > >> http://sourceforge.net/apps/trac/gar/browser/csw/mgar/pkg/subversion/trunk/files/fixme.sh >> > >> > This I don't really understand. >> >> This may be replaceable by the 'STRIP_LIBTOOL = 1' capability that >> Mike introduced to GAR, but I'm not sure. > > > STRIP_LIBTOOL is there as well. > > >> >> > Note that as of Subversion 1.7 the build system will always use the >> > libtool program that was used to compile APR. This was done so we >> > could remove the configure script code for finding the right libtool >> > and let APR sort this out for us instead. >> >> This should be a good change overall, I think. That will likely mean >> that it uses /opt/csw/bin/libtool (or whatever). It may be possible >> that the version of libtool changes between subversion releases though >> since it won't be in lockstep with apr. >> >> > I suppose the above is something with the fixme.sh gone wrong? >> > I've never seen an error like this. >> >> I'd suspect the same thing. >> >> Hopefully some of my feedback is useful too. > > > many thanks, much appreciated :) i tried removing fixme.sh, but it still > goes wrong. let me copy the part out of libtool where it fails: > > if test -n "$relink_command"; then > # Determine the prefix the user has applied to our future dir. > inst_prefix_dir=`$echo "$destdir" | $SED "s%$libdir\$%%"` > > # Don't allow the user to place us outside of our expected > # location b/c this prevents finding dependent libraries that > # are installed to the same prefix. > # At present, this check doesn't affect windows .dll's that > # are installed into $libdir/../bin (currently, that works fine) > # but it's something to keep an eye on. > if test "$inst_prefix_dir" = "$destdir"; then > $echo "$modename: error: cannot install \`$file' to a directory > not ending in $libdir" 1>&2 > exit $EXIT_FAILURE > fi > > if test -n "$inst_prefix_dir"; then > # Stick the inst_prefix_dir data into the link command. > relink_command=`$echo "$relink_command" | $SP2NL | $SED > "s%@inst_prefix_dir@%-inst-prefix-dir $inst_prefix_dir%" | $NL2SP` > else > relink_command=`$echo "$relink_command" | $SP2NL | $SED > "s%@inst_prefix_dir@%%" | $NL2SP` > fi > > > > kr, rupert. > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rupert at opencsw.org Wed Oct 19 09:23:25 2011 From: rupert at opencsw.org (rupert THURNER) Date: Wed, 19 Oct 2011 09:23:25 +0200 Subject: [csw-maintainers] Fwd: Re: [csw-buildfarm] Fwd: [csw-devel] SF.net SVN: gar:[15079] csw/mgar/pkg/libserf/trunk/Makefile In-Reply-To: References: <1310901647-sup-5939@pinkfloyd.chass.utoronto.ca> <71F517CB-9496-498F-9BFD-393E8611536D@opencsw.org> <0F71E7DE-0872-4215-8C23-F0E1002E1720@opencsw.org> Message-ID: here again the persisting error which made me finally comment out the python binding... ---------- Forwarded message ---------- From: "rupert THURNER" Date: Aug 27, 2011 8:41 PM Subject: Re: [csw-buildfarm] Fwd: [csw-maintainers] [csw-devel] SF.net SVN: gar:[15079] csw/mgar/pkg/libserf/trunk/Makefile To: "Dagobert Michelsen" Cc: "list for CSW buildfarm maintainers" i respan the svn-1.7.0-beta3 build on unstable9x, and the error message stays: cd subversion/bindings/swig/python ; /bin/bash /home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/build-isa-i386/subversion-1.7.0-beta3/libtool --mode=install /opt/csw/bin/ginstall -c _core.la/home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/install-isa-i386/opt/csw/lib/python2.3/libsvn/_ core.la libtool: install: error: cannot install `_core.la' to a directory not ending in /opt/csw/lib/svn/svn-python/libsvn gmake[2]: *** [install-swig-py] Error 1 gmake[2]: Leaving directory `/home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/build-isa-i386/subversion-1.7.0-beta3' gmake[1]: *** [svn-python] Error 2 gmake[1]: Leaving directory `/home/rupert/mgar-sav/pkg/subversion/trunk' gmake: *** [merge-isa-i386] Error 2 i triggered the build as well on unstable9s, this time including the tests, so it is still running. rupert On Mon, Jul 25, 2011 at 12:53, Dagobert Michelsen wrote: > Hi Rupert, > > Am 24.07.2011 um 21:17 schrieb rupert THURNER: > > could you please install the new verisons of swig, apr, apr-util, > > libserf on unstable? > > unstable* should be updated now. However, I noticed an issue of 64 bit apr > which > I am already looking into. > > > Best regards > > -- Dago > > > > > kr, rupert > > > > > > > > ---------- Forwarded message ---------- > > From: Maciej Blizi?ski > > Date: 2011/7/17 > > Subject: Re: [csw-maintainers] [csw-devel] SF.net SVN: gar:[15079] > > csw/mgar/pkg/libserf/trunk/Makefile > > To: List for OpenCSW maintainers > > > > > > Em 17/07/2011 21:49, "rupert THURNER" escreveu: > > > >> > >> On Sun, Jul 17, 2011 at 22:40, Dagobert Michelsen > wrote: > >>> > >>> Hi Rupert, > >>> > >>> Am 17.07.2011 um 18:20 schrieb rupert THURNER: > >>>> many thanks, dago! one cannot say no to such a nice offer, could you > please help me get subversion-1.7 and serf running on unstable, or testing? > currently serf does not work, giving: > >>>> > >>>> $ svn up --ignore-externals > >>>> svn: Error running context: Internal error > >>>> > >>>> you can change the svn client to use serf by inserting in > ~/.subversion/servers: > >>>> [global] > >>>> http-library = serf > >>> > >>> An update of libserf won't work as svn is bound to libserf0-0 and the > new one is libserf1-0 which > >>> needs a recompile of subversion. > >>> > >>> New build of 1.7.0alpha3 is under way, contrib/ seems to be missing in > 1.7.0. > >>> I'll try the above test against libserf1-0 when I have a package. > >>> > >>> The build now fails with > >>> > >>>> /opt/csw/bin/ginstall -c -d > /home/dam/mgar/pkg/subversion/trunk/work/solaris9-i386/install-isa-i386/opt/csw/lib/python2.3/libsvn > >>>> cd subversion/bindings/swig/python ; /bin/bash > /home/dam/mgar/pkg/subversion/trunk/work/solaris9-i386/build-isa-i386/subversion-1.7.0-alpha3/libtool > --mode=install /opt/csw/bin/ginstall -c _core.la/home/dam/mgar/pkg/subversion/trunk/work/solaris9-i386/install-isa-i386/opt/csw/lib/python2.3/libsvn/_ > core.la > > > > Why python2.3 while we have 2.4... that's one odd thing here. > > > >>>> libtool: install: error: cannot install `_core.la' to a directory not > ending in /opt/csw/lib/svn/python/site-packages/libsvn > > > > This is our python libraries directory. > > > > Our Python is different from most distributions in that we have a > > patch which removes the python version from the python lib > > subdirectory. Don't ask me why as the decision predates my > > involvement in the project. Some upstream projects (libxml comes to > > mind) assume that there is or there needs to be a version number in > > the python lib path. It could be an instance of this problem. > > > > I would like to reintroduce the version specific python lib dir, but > > it requires time, and hasn't yet bubbled up to the top of my priority > > list. > > > > Maciej > > > > _______________________________________________ > > maintainers mailing list > > maintainers at lists.opencsw.org > > https://lists.opencsw.org/mailman/listinfo/maintainers > > .:: This mailing list's archive is public. ::. > > _______________________________________________ > > buildfarm mailing list > > buildfarm at lists.opencsw.org > > https://lists.opencsw.org/mailman/listinfo/buildfarm > > -- > "You don't become great by trying to be great, you become great by wanting > to do something, > and then doing it so hard that you become great in the process." - xkcd > #896 > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From maciej at opencsw.org Wed Oct 19 09:25:14 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 19 Oct 2011 08:25:14 +0100 Subject: [csw-maintainers] Fwd: complicated build, subversion client In-Reply-To: References: <20110529181632.GA9828@ted.stsp.name> <20110529223727.GB9828@ted.stsp.name> <1306800816-sup-4668@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/19 rupert THURNER : > any help to sort out that build error is greatly appreciated. What's the error? From maciej at opencsw.org Wed Oct 19 09:26:44 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 19 Oct 2011 08:26:44 +0100 Subject: [csw-maintainers] Fwd: Re: [csw-buildfarm] Fwd: [csw-devel] SF.net SVN: gar:[15079] csw/mgar/pkg/libserf/trunk/Makefile In-Reply-To: References: <1310901647-sup-5939@pinkfloyd.chass.utoronto.ca> <71F517CB-9496-498F-9BFD-393E8611536D@opencsw.org> <0F71E7DE-0872-4215-8C23-F0E1002E1720@opencsw.org> Message-ID: 2011/10/19 rupert THURNER : > i respan the svn-1.7.0-beta3 build on unstable9x, and the error message > stays: > cd subversion/bindings/swig/python ; /bin/bash > /home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/build-isa-i386/subversion-1.7.0-beta3/libtool > --mode=install /opt/csw/bin/ginstall -c _core.la > /home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/install-isa-i386/opt/csw/lib/python2.3/libsvn/_core.la > libtool: install: error: cannot install `_core.la' to a directory not ending > in /opt/csw/lib/svn/svn-python/libsvn Could it be this? http://wiki.opencsw.org/porting-faq#toc2 From maciej at opencsw.org Wed Oct 19 09:29:30 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Wed, 19 Oct 2011 08:29:30 +0100 Subject: [csw-maintainers] GCC 4.6 progress In-Reply-To: References: Message-ID: To make the build a bit easier on current*, I promoted gmp, mpfr, ppl and cloog to dublin. The idea was to have the dependencies in place. Unfortunately, these libraries already depend on CSWlibgcc-s1 and CSWlibstdc++6. The dublin catalog is now in broken state. I will fix it today in the evening. Maciej From dam at opencsw.org Wed Oct 19 16:14:09 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 19 Oct 2011 16:14:09 +0200 Subject: [csw-maintainers] Fwd: complicated build, subversion client In-Reply-To: References: <20110529181632.GA9828@ted.stsp.name> <20110529223727.GB9828@ted.stsp.name> <1306800816-sup-4668@pinkfloyd.chass.utoronto.ca> Message-ID: Hi Rupert, Am 19.10.2011 um 09:16 schrieb rupert THURNER: > as the python bindings did not compile for more than 6 months, I commented them for the first try of svn-1.7. but I d have a bad feeling if we let this package further than unstable. any help to sort out that build error is greatly appreciated. I noticed some thing on 1.7.0: - At least one test is failing: Running tests in copy_tests.py [44/87]........................FAILURE Would you mind checking with upstream? - If you comment out specific things please always make a comment in the Makefile why this has been done and what problem exists. - I cannot reproduce the problem with pkg/subversion/trunk, could you please reenable the python module so I can have a look? I am confident that we can fix this in a joint action. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From bwalton at opencsw.org Thu Oct 20 04:34:36 2011 From: bwalton at opencsw.org (Ben Walton) Date: Wed, 19 Oct 2011 22:34:36 -0400 Subject: [csw-maintainers] mantis updates are running... In-Reply-To: <20111018113745.GA26160@sebastiankayser.de> References: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> <20111018113745.GA26160@sebastiankayser.de> Message-ID: <1319077818-sup-1943@pinkfloyd.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Tue Oct 18 07:37:45 -0400 2011: Hi Sebastian, > From looking at "My Account", there's an extra package: > dovecot_sieve. Jake took over Dovecot and merged the _sieve package > into the main package. dovecot_sieve shouldn't exist any more. I've now had a chance to look at this. The assignment (manager of the project) to you existed in the database prior to me running the integration code. The integration code made no changes to this project, which is the correct outcome in this case. The dovecot_sieve project wasn't in the original catalog I started from[1] and no changes were introduced to the package in the meantime. Thus, the integration code ignored it entirely. This means that at some point in the past the project wasn't marked as 'obsolete' with the old tools. I can either write a command line tool to toggle these states for individual projects or we can use the web interface. For one-off stuff, I'm inclined to use the existing interface. Sound good to you? Thanks -Ben [1] This is the last catalog that Phil pushed. -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From dam at opencsw.org Thu Oct 20 17:34:22 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 20 Oct 2011 17:34:22 +0200 Subject: [csw-maintainers] Webpage getting started References: <97341233612811@192.168.2.69> Message-ID: Hi, looks like our webpage still doesn't direct beginners to the right page: Anfang der weitergeleiteten E-Mail: >>> If i remember right, then i only have instructions how to download >>> from SUN's repositories. (And even those i would have to search.) > > Dagobert Michelsen wrote: >> It is as easy as >> pkgadd -d http://get.opencsw.org/now >> /opt/csw/bin/pkgutil -i git > > This will save me much time with googling. Thank you. > >> http://www.opencsw.org/get-it/pkgutil/ > > I looked at www.opencsw.org but took the wrong road to "Use it" > rather than "Get it". Best regards -- Dago From rupert at opencsw.org Fri Oct 21 06:50:13 2011 From: rupert at opencsw.org (rupert THURNER) Date: Fri, 21 Oct 2011 06:50:13 +0200 Subject: [csw-maintainers] Fwd: complicated build, subversion client In-Reply-To: References: <20110529181632.GA9828@ted.stsp.name> <20110529223727.GB9828@ted.stsp.name> <1306800816-sup-4668@pinkfloyd.chass.utoronto.ca> Message-ID: On Wed, Oct 19, 2011 at 16:14, Dagobert Michelsen wrote: > Hi Rupert, > > Am 19.10.2011 um 09:16 schrieb rupert THURNER: >> as the python bindings did not compile for more than 6 months, I commented them for the first try of svn-1.7. but I d have a bad feeling if we let this package further than unstable. any help to sort out that build error is greatly appreciated. > > I noticed some thing on 1.7.0: > > - At least one test is failing: > ? ?Running tests in copy_tests.py [44/87]........................FAILURE > ?Would you mind checking with upstream? > - If you comment out specific things please always make a comment in the Makefile why > ?this has been done and what problem exists. > - I cannot reproduce the problem with pkg/subversion/trunk, could you please > ?reenable the python module so I can have a look? I am confident that we can > ?fix this in a joint action. enabled it again and checked in. as long as the test is failing SKIPTEST=1 mgar package will produce the error. From maciej at opencsw.org Fri Oct 21 09:49:11 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Fri, 21 Oct 2011 08:49:11 +0100 Subject: [csw-maintainers] website does not reflect real package versions ? In-Reply-To: <1318994869-sup-3633@pinkfloyd.chass.utoronto.ca> References: <1318994869-sup-3633@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/19 Ben Walton : > As previously discussed, I still need a way to get the list of files > in a package (we didn't register directories previously) and if anyone > still cares for the feature[1] the shared objects the package relies > on, but the basics should be ready to go... I implemented a GET method for the lists of files, it returns files, directories and symlinks - basically, whatever is stored in the csw_file table in the buildfarm database. The URL is: http://buildfarm.opencsw.org/pkgdb/rest/srv4//files/ Maciej From maciej at opencsw.org Fri Oct 21 10:24:03 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Fri, 21 Oct 2011 09:24:03 +0100 Subject: [csw-maintainers] OpenCSW Summercamp 2011 in Kiel summary In-Reply-To: References: Message-ID: 2011/10/1 Maciej (Matchek) Blizi?ski : > - we'll build for sparcv8+ and pentium_pro by default on Solaris 10 The changes to GAR and checkpkg are now in the repository. Please run "mgar up --all" to get the code updates, and your builds (on Solaris 10) will be now made for sparcv8+ and pentium_pro by default. No more missing __sync_fetch_and_add_4! Maciej From dam at opencsw.org Fri Oct 21 12:12:58 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Fri, 21 Oct 2011 12:12:58 +0200 Subject: [csw-maintainers] Fwd: complicated build, subversion client In-Reply-To: References: <20110529181632.GA9828@ted.stsp.name> <20110529223727.GB9828@ted.stsp.name> <1306800816-sup-4668@pinkfloyd.chass.utoronto.ca> Message-ID: Hi Rupert, Am 21.10.2011 um 06:50 schrieb rupert THURNER: > On Wed, Oct 19, 2011 at 16:14, Dagobert Michelsen wrote: >> Am 19.10.2011 um 09:16 schrieb rupert THURNER: >>> as the python bindings did not compile for more than 6 months, I commented them for the first try of svn-1.7. but I d have a bad feeling if we let this package further than unstable. any help to sort out that build error is greatly appreciated. >> >> I noticed some thing on 1.7.0: >> >> - At least one test is failing: >> Running tests in copy_tests.py [44/87]........................FAILURE >> Would you mind checking with upstream? >> - If you comment out specific things please always make a comment in the Makefile why >> this has been done and what problem exists. >> - I cannot reproduce the problem with pkg/subversion/trunk, could you please >> reenable the python module so I can have a look? I am confident that we can >> fix this in a joint action. > > enabled it again and checked in. as long as the test is failing > SKIPTEST=1 mgar package > will produce the error. Ok, rebuilding know. I'll see what I can do. Have you reported the failing test upstream? Best regards -- Dago From skayser at opencsw.org Sat Oct 22 00:07:56 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sat, 22 Oct 2011 00:07:56 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: References: <97341233612811@192.168.2.69> Message-ID: <20111021220756.GG26160@sebastiankayser.de> * Dagobert Michelsen wrote: > looks like our webpage still doesn't direct beginners to the right page: > > Anfang der weitergeleiteten E-Mail: > >>> If i remember right, then i only have instructions how to download > >>> from SUN's repositories. (And even those i would have to search.) > > > > Dagobert Michelsen wrote: > >> It is as easy as > >> pkgadd -d http://get.opencsw.org/now > >> /opt/csw/bin/pkgutil -i git > > > > This will save me much time with googling. Thank you. > > > >> http://www.opencsw.org/get-it/pkgutil/ > > > > I looked at www.opencsw.org but took the wrong road to "Use it" > > rather than "Get it". I think, William or Maciej once mentioned that they tend to scan a page for the "Download" link. Just checked the sites of the usual NoSQL suspects (redis, couch, mongo, ..) as a sample for current project sites and they all have one as a main navigation item. How about renaming "Get it" to "Download" (and "Use it" to "Documentation")? Sebastian From bonivart at opencsw.org Sat Oct 22 02:14:39 2011 From: bonivart at opencsw.org (Peter Bonivart) Date: Sat, 22 Oct 2011 02:14:39 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: <20111021220756.GG26160@sebastiankayser.de> References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> Message-ID: On Sat, Oct 22, 2011 at 12:07 AM, Sebastian Kayser wrote: > How about renaming "Get > it" to "Download" (and "Use it" to "Documentation")? +1 From bwalton at opencsw.org Sat Oct 22 02:22:47 2011 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 21 Oct 2011 20:22:47 -0400 Subject: [csw-maintainers] Webpage getting started In-Reply-To: <20111021220756.GG26160@sebastiankayser.de> References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> Message-ID: <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Fri Oct 21 18:07:56 -0400 2011: > I think, William or Maciej once mentioned that they tend to scan a > page for the "Download" link. Just checked the sites of the usual > NoSQL suspects (redis, couch, mongo, ..) as a sample for current > project sites and they all have one as a main navigation item. How > about renaming "Get it" to "Download" (and "Use it" to > "Documentation")? I'm good with that. Both of those are keywords that I quickly scan for. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From skayser at opencsw.org Sun Oct 23 12:04:21 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 23 Oct 2011 12:04:21 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> Message-ID: <20111023100421.GI26160@sebastiankayser.de> * Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Fri Oct 21 18:07:56 -0400 2011: > > > I think, William or Maciej once mentioned that they tend to scan a > > page for the "Download" link. Just checked the sites of the usual > > NoSQL suspects (redis, couch, mongo, ..) as a sample for current > > project sites and they all have one as a main navigation item. How > > about renaming "Get it" to "Download" (and "Use it" to > > "Documentation")? > > I'm good with that. Both of those are keywords that I quickly scan > for. Cool, item names have been changed. I've left the permalinks (/get-it/, /use-it/) alone for now. We can think about changing them (and putting in redirects for the old links) in the future. Sebastian From maciej at opencsw.org Sun Oct 23 12:08:28 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 23 Oct 2011 11:08:28 +0100 Subject: [csw-maintainers] Webpage getting started In-Reply-To: <20111023100421.GI26160@sebastiankayser.de> References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> <20111023100421.GI26160@sebastiankayser.de> Message-ID: 2011/10/23 Sebastian Kayser : > Cool, item names have been changed. I've left the permalinks (/get-it/, > /use-it/) alone for now. We can think about changing them (and putting > in redirects for the old links) in the future. That's not a big deal. Do we have access to the Apache configuration to add a redirection? From bonivart at opencsw.org Sun Oct 23 12:09:45 2011 From: bonivart at opencsw.org (Peter Bonivart) Date: Sun, 23 Oct 2011 12:09:45 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: <20111023100421.GI26160@sebastiankayser.de> References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> <20111023100421.GI26160@sebastiankayser.de> Message-ID: On Sun, Oct 23, 2011 at 12:04 PM, Sebastian Kayser wrote: > Cool, item names have been changed. I've left the permalinks (/get-it/, > /use-it/) alone for now. We can think about changing them (and putting > in redirects for the old links) in the future. Can we also change the top level "Extend It" to Contribute? And the "Get It" at the right to "Get Started"? /peter From skayser at opencsw.org Sun Oct 23 12:21:27 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 23 Oct 2011 12:21:27 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> <20111023100421.GI26160@sebastiankayser.de> Message-ID: <20111023102127.GJ26160@sebastiankayser.de> * Peter Bonivart wrote: > On Sun, Oct 23, 2011 at 12:04 PM, Sebastian Kayser wrote: > > Cool, item names have been changed. I've left the permalinks (/get-it/, > > /use-it/) alone for now. We can think about changing them (and putting > > in redirects for the old links) in the future. > > Can we also change the top level "Extend It" to Contribute? Yes, definitely. > And the "Get It" at the right to "Get Started"? Both done. I also moved Get Started to the top as getting started is probably what a beginner (i.e. newcomer to our page) wants to do. Sebastian From skayser at opencsw.org Sun Oct 23 12:24:17 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 23 Oct 2011 12:24:17 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> <20111023100421.GI26160@sebastiankayser.de> Message-ID: <20111023102417.GK26160@sebastiankayser.de> * Maciej (Matchek) Blizi??ski wrote: > 2011/10/23 Sebastian Kayser : > > Cool, item names have been changed. I've left the permalinks (/get-it/, > > /use-it/) alone for now. We can think about changing them (and putting > > in redirects for the old links) in the future. > > That's not a big deal. Do we have access to the Apache configuration > to add a redirection? Yes, we (as in Ihsan, Dago, Ben, William, you, and I) do. Started to document some of our infrastructure setup on the protected opencsw-orga.wikidot.com wiki that Dago created. Path to the relevant Apache config is in there. Sebastian From skayser at opencsw.org Sun Oct 23 12:32:47 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 23 Oct 2011 12:32:47 +0200 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? Message-ID: <20111023103247.GL26160@sebastiankayser.de> Hi, after just renaming those three top-level navigation items [1], I hid "Feeds" from the top-level to not have it explode length-wise. Those who are after feeds, can still easily find it in the right navigation bar [2]. While doing so, I noticed that our released packages feed is empty [3]. Could someone look after it. Also, currently we only show selected pkg updates on our front page (lower right). To the uninitiated this might look like these are our only updates. Would be awesome to have the released pkgs feed in there. Whose Wordpress / PHP foo is strong enough to do this? Sebastian [1] http://lists.opencsw.org/pipermail/maintainers/2011-October/015557.html [2] http://www.opencsw.org/feeds/ [3] http://www.opencsw.org/cgi-feed/releasedpkgs.atom From bwalton at opencsw.org Sun Oct 23 15:43:53 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 23 Oct 2011 09:43:53 -0400 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <20111023103247.GL26160@sebastiankayser.de> References: <20111023103247.GL26160@sebastiankayser.de> Message-ID: <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Sun Oct 23 06:32:47 -0400 2011: Hi Sebastian, > While doing so, I noticed that our released packages feed is empty [3]. > Could someone look after it. Now that I'm deploying stuff on www.opencsw.org, I'll move the feeds that I've been generating from the buildfarm over to www and hook them up under /feeds. If I recall, the old feed was updated by scripts Phil ran during the pushes. I'm open to suggestions on how to either deprecate or replace the old atom feeds. We could potentially just replace the old feed with one of the new ones... Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From rupert at opencsw.org Sun Oct 23 16:39:06 2011 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 23 Oct 2011 16:39:06 +0200 Subject: [csw-maintainers] Fwd: Re: [csw-buildfarm] Fwd: [csw-devel] SF.net SVN: gar:[15079] csw/mgar/pkg/libserf/trunk/Makefile In-Reply-To: References: <1310901647-sup-5939@pinkfloyd.chass.utoronto.ca> <71F517CB-9496-498F-9BFD-393E8611536D@opencsw.org> <0F71E7DE-0872-4215-8C23-F0E1002E1720@opencsw.org> Message-ID: 2011/10/19 Maciej (Matchek) Blizi?ski : > 2011/10/19 rupert THURNER : >> i respan the svn-1.7.0-beta3 build on unstable9x, and the error message >> stays: >> cd subversion/bindings/swig/python ; /bin/bash >> /home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/build-isa-i386/subversion-1.7.0-beta3/libtool >> --mode=install /opt/csw/bin/ginstall -c _core.la >> /home/rupert/mgar-sav/pkg/subversion/trunk/work/solaris9-i386/install-isa-i386/opt/csw/lib/python2.3/libsvn/_core.la >> libtool: install: error: cannot install `_core.la' to a directory not ending >> in /opt/csw/lib/svn/svn-python/libsvn > > Could it be this? > > http://wiki.opencsw.org/porting-faq#toc2 you mean the STRIP_LIBTOOL = 1 ? it is at least in the gar makefile. but the bindings are built separately so i am unsure where this would end up? From bwalton at opencsw.org Sun Oct 23 18:17:10 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 23 Oct 2011 12:17:10 -0400 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <20111023103247.GL26160@sebastiankayser.de> References: <20111023103247.GL26160@sebastiankayser.de> Message-ID: <1319386521-sup-6814@pinkfloyd.chass.utoronto.ca> Excerpts from Sebastian Kayser's message of Sun Oct 23 06:32:47 -0400 2011: > While doing so, I noticed that our released packages feed is empty [3]. > Could someone look after it. So www/cgi-feed/releasedpkgs.atom is currently a redirect to pkgrelease/cgi-feed/releasedpkgs.atom. pkgrelease is bender. Ihsan, would you prefer to retain this setup where the feeds live on bender or are you ok with me running the feeds on www.opencsw.org directly? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From pfelecan at opencsw.org Sun Oct 23 18:44:15 2011 From: pfelecan at opencsw.org (Peter FELECAN) Date: Sun, 23 Oct 2011 18:44:15 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: <20111023102417.GK26160@sebastiankayser.de> (Sebastian Kayser's message of "Sun, 23 Oct 2011 12:24:17 +0200") References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> <20111023100421.GI26160@sebastiankayser.de> <20111023102417.GK26160@sebastiankayser.de> Message-ID: Sebastian Kayser writes: > Started to document some of our infrastructure setup on the protected > opencsw-orga.wikidot.com wiki that Dago created. Path to the relevant > Apache config is in there. Protected as in "maintainers can access using a user/password combination"? Or is limited to administrators? -- Peter From dam at opencsw.org Sun Oct 23 19:22:29 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Sun, 23 Oct 2011 19:22:29 +0200 Subject: [csw-maintainers] Webpage getting started In-Reply-To: References: <97341233612811@192.168.2.69> <20111021220756.GG26160@sebastiankayser.de> <1319242939-sup-1567@pinkfloyd.chass.utoronto.ca> <20111023100421.GI26160@sebastiankayser.de> <20111023102417.GK26160@sebastiankayser.de> Message-ID: <799B7990-412F-4757-A4D2-333296467F6D@opencsw.org> Hi Peter, Am 23.10.2011 um 18:44 schrieb Peter FELECAN: > Sebastian Kayser writes: >> Started to document some of our infrastructure setup on the protected >> opencsw-orga.wikidot.com wiki that Dago created. Path to the relevant >> Apache config is in there. > > Protected as in "maintainers can access using a user/password > combination"? Or is limited to administrators? I would say limited to people who need the information for their work. Not because it is a secret, but because it is not useful if you don't have the proper system access the config is related to. Currently there is information about downstream mirrors, the mantis-setup and the wordpress setup. Want to join in hacking the site? :-) Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From skayser at opencsw.org Sun Oct 23 20:37:04 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 23 Oct 2011 20:37:04 +0200 Subject: [csw-maintainers] mantis updates are running... In-Reply-To: <1319077818-sup-1943@pinkfloyd.chass.utoronto.ca> References: <1318902574-sup-351@pinkfloyd.chass.utoronto.ca> <20111018113745.GA26160@sebastiankayser.de> <1319077818-sup-1943@pinkfloyd.chass.utoronto.ca> Message-ID: <20111023183704.GN26160@sebastiankayser.de> * Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Tue Oct 18 07:37:45 -0400 2011: > > From looking at "My Account", there's an extra package: > > dovecot_sieve. Jake took over Dovecot and merged the _sieve package > > into the main package. dovecot_sieve shouldn't exist any more. > > I've now had a chance to look at this. The assignment (manager of the > project) to you existed in the database prior to me running the > integration code. The integration code made no changes to this > project, which is the correct outcome in this case. > > The dovecot_sieve project wasn't in the original catalog I started > from[1] and no changes were introduced to the package in the meantime. > Thus, the integration code ignored it entirely. This means that at > some point in the past the project wasn't marked as 'obsolete' with > the old tools. > > I can either write a command line tool to toggle these states for > individual projects or we can use the web interface. For one-off > stuff, I'm inclined to use the existing interface. Sound good to you? Sorry for the late reply. Yes, sounds good. Web interface is perfectly fine, if we only need to do this a few times. Sebastian From skayser at opencsw.org Sun Oct 23 21:06:33 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 23 Oct 2011 21:06:33 +0200 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> Message-ID: <20111023190633.GO26160@sebastiankayser.de> * Ben Walton wrote: > Excerpts from Sebastian Kayser's message of Sun Oct 23 06:32:47 -0400 2011: > > While doing so, I noticed that our released packages feed is empty [3]. > > Could someone look after it. > > Now that I'm deploying stuff on www.opencsw.org, I'll move the feeds > that I've been generating from the buildfarm over to www and hook them > up under /feeds. Good. Are these the ones you mentioned a couple of months ago [1]? > If I recall, the old feed was updated by scripts Phil ran during the > pushes. I'm open to suggestions on how to either deprecate or replace > the old atom feeds. We could potentially just replace the old feed > with one of the new ones... Yes, use one of yours. Migration-wise, the old one is currently empty, so I'd say, just replace it. unstable is where we push to, would IMHO lend itself as a good replacement (redirect from the old feed URL to the new one). Those who want specific release branch update information can subscribe to the individual feeds. Content-wise, the title of each feed item currently says "Updated $pkg" or "Added $pkg", could you add the new version information to it so that it e.g. says "Updated $pkg x.y,REV=...."? Then a stream of such titles could go onto our frontpage. Sebastian [1] http://buildfarm.opencsw.org/~bwalton/ From skayser at opencsw.org Tue Oct 25 01:15:42 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 25 Oct 2011 01:15:42 +0200 Subject: [csw-maintainers] Release process graph // release branch info on opencsw.org Message-ID: <20111024231542.GP26160@sebastiankayser.de> Hi, have you struggled to remember how exactly our different package release branches work (symlinks vs. real branches, manual vs. automated updates vs. no updates)? At least I have, more than one time. Now there's help! Maciej added brief explanations to our master mirror README [1] and the following graph should help to get a good picture of what's actually happening: http://buildfarm.opencsw.org/~skayser/opencsw-release-branches-v3.png The release branch information on opencsw.org however is still outdated. Any volunteer to de-duplicate and update the relevant content [2,3]? Then we could add a frontpage news items (with links to the updated information) to bring our users up to speed and spare them hunting for details. Sebastian [1] http://mirror.opencsw.org/opencsw/ [2] http://www.opencsw.org/get-it/releases/ [3] http://www.opencsw.org/use-it/#release From maciej at opencsw.org Tue Oct 25 10:27:23 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 25 Oct 2011 09:27:23 +0100 Subject: [csw-maintainers] Full package statistics via REST Message-ID: During the Summercamp, there was a discussion about plugins for checkpkg. People wanted to write package checks, but couldn't or didn't want to write in Python. I suggested, as an intermediate step, getting package statistics over REST and writing checks in the language of choice. I've just committed r15980, which make the web interface export full package statistics via REST. What does it mean? It means that if you make a GET request at the REST package details address, you will get a JSON-serialized data structure, which represents everything that the buildfarm database knows about the package. You can see the JSON data printed to screen by calling this address: curl -s 'http://buildfarm.opencsw.org/pkgdb/rest/srv4/3a483f192e2e9d788e922895ef1261e9/' Here's an interactive Python example: >>> import urllib2 >>> import json >>> url = "http://buildfarm.opencsw.org/pkgdb/rest/srv4/3a483f192e2e9d788e922895ef1261e9/" >>> p = json.load(urllib2.urlopen(url)) >>> p["pkg_stats"]["binaries_dump_info"][0] {u'base_name': u'9menu', u'RUNPATH RPATH the same': True, u'runpath': [], u'RPATH set': False, u'needed sonames': [u'libXext.so.0', u'libX11.so.4', u'libsocket.so.1', u'libnsl.so.1', u'libc.so.1'], u'path': u'opt/csw/opt/csw/bin/9menu', u'RUNPATH set': False} An example in Perl, which gets the data and pretty prints them to screen: use strict; use JSON; use LWP::Simple qw(get); use Data::Dumper; my $url = "http://buildfarm.opencsw.org/pkgdb/rest/srv4/3a483f192e2e9d788e922895ef1261e9/"; my $json_repr = get $url; my $pkg_data = decode_json($json_repr); print Dumper($pkg_data); I encourage enthusiasts of other languages to post their examples. Once you fetch and deserialize that data, you know about the package everything that checkpkg knows. You can write your own checking function and run it across any packages. If you want to fetch a list of packages from a specific catalog, call an URL like this one... http://buildfarm.opencsw.org/pkgdb/rest/catalogs/unstable/sparc/SunOS5.9/ ...and deserialize JSON. It takes a while to get that data, so be patient. The REST interface is generally slow at the moment. I haven't spent any time on optimization. I decided I'll spend time on performance when there's a need for it. So far, the consumers of the data are mostly automation, cronjobs, etc. If you have any questions, feel free to ask. Enjoy writing your checks and analyzing the package data! Maciej http://sourceforge.net/apps/trac/gar/changeset/15980 From skayser at opencsw.org Tue Oct 25 15:13:48 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Tue, 25 Oct 2011 15:13:48 +0200 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <20111023190633.GO26160@sebastiankayser.de> References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> Message-ID: <20111025131348.GQ26160@sebastiankayser.de> Hey Ben, * Sebastian Kayser wrote: > * Ben Walton wrote: > > Excerpts from Sebastian Kayser's message of Sun Oct 23 06:32:47 -0400 2011: > > > While doing so, I noticed that our released packages feed is empty [3]. > > > Could someone look after it. > > > > Now that I'm deploying stuff on www.opencsw.org, I'll move the feeds > > that I've been generating from the buildfarm over to www and hook them > > up under /feeds. > > Good. Are these the ones you mentioned a couple of months ago [1]? one more question. Just added the unstable/i386/5.10 feed [1] to ceeswi for broadcasting on IRC. Noticed, that the feed file is ~500KB in size and contains items going back to early this year. The bot took a while to download it. Can you restrict the item retention to a (shorter) timespan to decrease the feed size? Sebastian P.S.: Not sure whether that's how feed content is usually handled, but overall limited size or time range and thus transient content seems to make sense size- and transfer-wise. Please yell, if that's off. [1] http://buildfarm.opencsw.org/~bwalton/feed.mirror.opencsw.org.opencsw.current.SunOS5.10.i386.atom.xml From bonivart at opencsw.org Tue Oct 25 15:19:45 2011 From: bonivart at opencsw.org (Peter Bonivart) Date: Tue, 25 Oct 2011 15:19:45 +0200 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <20111025131348.GQ26160@sebastiankayser.de> References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> <20111025131348.GQ26160@sebastiankayser.de> Message-ID: On Tue, Oct 25, 2011 at 3:13 PM, Sebastian Kayser wrote: > one more question. Just added the unstable/i386/5.10 feed [1] to ceeswi > for broadcasting on IRC. Noticed, that the feed file is ~500KB in size > and contains items going back to early this year. The bot took a while > to download it. Can you restrict the item retention to a (shorter) > timespan to decrease the feed size? I have also seen that as a problem that it's up to the client to limit the scope you see. Since there's no interest in, let's say, more than a month back (or even just a week) it would be nice if the process generating the feed truncated it. Simpler clients that can't set the scope would be much fasten then. /peter From bwalton at opencsw.org Tue Oct 25 15:38:50 2011 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 25 Oct 2011 09:38:50 -0400 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> <20111025131348.GQ26160@sebastiankayser.de> Message-ID: <1319549891-sup-2657@pinkfloyd.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Tue Oct 25 09:19:45 -0400 2011: > I have also seen that as a problem that it's up to the client to limit > the scope you see. Since there's no interest in, let's say, more than > a month back (or even just a week) it would be nice if the process > generating the feed truncated it. Simpler clients that can't set the > scope would be much fasten then. Ok, I'll curtail this. Is everyone happy with a 1 week interval or is a month preferable? In the current implementation, there is no culling at all. :) Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From maciej at opencsw.org Tue Oct 25 16:37:51 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 25 Oct 2011 15:37:51 +0100 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <1319549891-sup-2657@pinkfloyd.chass.utoronto.ca> References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> <20111025131348.GQ26160@sebastiankayser.de> <1319549891-sup-2657@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/25 Ben Walton : > Excerpts from Peter Bonivart's message of Tue Oct 25 09:19:45 -0400 2011: > >> I have also seen that as a problem that it's up to the client to limit >> the scope you see. Since there's no interest in, let's say, more than >> a month back ?(or even just a week) it would be nice if the process >> generating the feed truncated it. Simpler clients that can't set the >> scope would be much fasten then. > > Ok, I'll curtail this. ?Is everyone happy with a 1 week interval or is > a month preferable? ?In the current implementation, there is no > culling at all. :) I'd say a month rather than a week. The feed should still be reasonable in size. Ben, I also wanted the feed to be more terse, e.g. don't report foo, foo_dev, foo_bar, foo_baz in separate items. I would be willing to look at it, if you gave me some pointers, where to get the code and which functions/classes to look at. Maciej From bwalton at opencsw.org Tue Oct 25 16:42:00 2011 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 25 Oct 2011 10:42:00 -0400 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> <20111025131348.GQ26160@sebastiankayser.de> <1319549891-sup-2657@pinkfloyd.chass.utoronto.ca> Message-ID: <1319553641-sup-6051@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Tue Oct 25 10:37:51 -0400 2011: > I'd say a month rather than a week. The feed should still be > reasonable in size. I like this window better too. > Ben, I also wanted the feed to be more terse, e.g. don't report foo, > foo_dev, foo_bar, foo_baz in separate items. I would be willing to > look at it, if you gave me some pointers, where to get the code and > which functions/classes to look at. The best way to do this is via bundle info. Are that available via REST (or are we even storing that in the pkginfo file right now?)...If it's available over REST, it should be fairly easy to modify the feeds to use it. I can also filter out _stub add/remove/update operations. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From bwalton at opencsw.org Tue Oct 25 16:59:35 2011 From: bwalton at opencsw.org (Ben Walton) Date: Tue, 25 Oct 2011 10:59:35 -0400 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: References: Message-ID: <1319554680-sup-8126@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Tue Oct 25 04:27:23 -0400 2011: Hi Maciej, > It means that if you make a GET request at the REST package details > address, you will get a JSON-serialized data structure, which > represents everything that the buildfarm database knows about the > package. This is great. I can grab the .so NEEDED info from this to recreate the existing library usage info. The only thing I wonder is if this info should live at it's own REST-ful URL? Right now, I get this data where I need it or not and that will both increase compute time on the server for every request and increase the data transferred. If it were separate, it could be had on-demand only? What do you think. More work than it's worth? Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From maciej at opencsw.org Tue Oct 25 17:21:18 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 25 Oct 2011 16:21:18 +0100 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: <1319554680-sup-8126@pinkfloyd.chass.utoronto.ca> References: <1319554680-sup-8126@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/25 Ben Walton : > This is great. ?I can grab the .so NEEDED info from this to recreate > the existing library usage info. ?The only thing I wonder is if this > info should live at it's own REST-ful URL? ?Right now, I get this data > where I need it or not and that will both increase compute time on the > server for every request and increase the data transferred. ?If it > were separate, it could be had on-demand only? > > What do you think. ?More work than it's worth? Good suggestion. Done in r15988. http://sourceforge.net/apps/trac/gar/changeset/15988 The URL is now: http://buildfarm.opencsw.org/pkgdb/rest/srv4/${md5}/pkg-stats/ Maciej From maciej at opencsw.org Tue Oct 25 18:43:06 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Tue, 25 Oct 2011 17:43:06 +0100 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: <1319553641-sup-6051@pinkfloyd.chass.utoronto.ca> References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> <20111025131348.GQ26160@sebastiankayser.de> <1319549891-sup-2657@pinkfloyd.chass.utoronto.ca> <1319553641-sup-6051@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/25 Ben Walton : > The best way to do this is via bundle info. ?Are that available via > REST (or are we even storing that in the pkginfo file right now?) It's currently available in the newly published pkg-stats data structure. It's a bit heavy, but it doesn't matter for a batch job. When I get to altering the db schema, we can put this info in the lightweight svr4 file info. Maciej From daniel at opencsw.org Wed Oct 26 14:10:23 2011 From: daniel at opencsw.org (Daniel Pocock) Date: Wed, 26 Oct 2011 20:10:23 +0800 Subject: [csw-maintainers] svn error with gar.svn.sourceforge.net Message-ID: <4EA7F8AF.4000903@opencsw.org> Hi, It's been a while since I've done anything with OpenCSW packages. I've been receiving reminder emails asking me to upgrade the libconfuse package from 2.6 to 2.7 (the latest upstream) So I tried to log in, but SVN won't connect to gar.svn.sourceforge.net - is there any known issue? Regards, Daniel bash-2.03$ hostname current8s bash-2.03$ pwd /home/daniel/ws/libconfuse/trunk bash-2.03$ svn update svn: OPTIONS of 'https://gar.svn.sourceforge.net/svnroot/gar/csw/mgar/pkg/libconfuse/trunk': could not connect to server (https://gar.svn.sourceforge.net) From bonivart at opencsw.org Wed Oct 26 14:25:37 2011 From: bonivart at opencsw.org (Peter Bonivart) Date: Wed, 26 Oct 2011 14:25:37 +0200 Subject: [csw-maintainers] svn error with gar.svn.sourceforge.net In-Reply-To: <4EA7F8AF.4000903@opencsw.org> References: <4EA7F8AF.4000903@opencsw.org> Message-ID: On Wed, Oct 26, 2011 at 2:10 PM, Daniel Pocock wrote: > bash-2.03$ hostname > current8s Try unstable9x or unstable9s instead and see if that helps. /peter From daniel at opencsw.org Wed Oct 26 14:59:41 2011 From: daniel at opencsw.org (Daniel Pocock) Date: Wed, 26 Oct 2011 20:59:41 +0800 Subject: [csw-maintainers] svn error with gar.svn.sourceforge.net In-Reply-To: References: <4EA7F8AF.4000903@opencsw.org> Message-ID: <4EA8043D.3030105@opencsw.org> On 26/10/11 20:25, Peter Bonivart wrote: > On Wed, Oct 26, 2011 at 2:10 PM, Daniel Pocock wrote: > >> bash-2.03$ hostname >> current8s >> > Try unstable9x or unstable9s instead and see if that helps. > assertion failed... see below... there were no local changes in this working copy, so I'll try and check it out again, but it seems a bit unusual to have it fail like this daniel at unstable9s :~/ws/libconfuse > svn upgrade Upgraded '.' Upgraded 'trunk' Upgraded 'trunk/files' Upgraded 'branches' Upgraded 'tags' Upgraded 'trunk/gar' Upgraded 'trunk/gar/categories' Upgraded 'trunk/gar/categories/xorg' Upgraded 'trunk/gar/categories/python' Upgraded 'trunk/gar/categories/lang' Upgraded 'trunk/gar/categories/java' Upgraded 'trunk/gar/categories/cpan' Upgraded 'trunk/gar/categories/xfce' Upgraded 'trunk/gar/categories/net' Upgraded 'trunk/gar/categories/meta' Upgraded 'trunk/gar/categories/devel' Upgraded 'trunk/gar/categories/utils' Upgraded 'trunk/gar/categories/kde' Upgraded 'trunk/gar/categories/apps' Upgraded 'trunk/gar/categories/gnome' Upgraded 'trunk/gar/categories/server' Upgraded 'trunk/gar/categories/lib' Upgraded 'trunk/gar/categories/x11' Upgraded 'trunk/gar/categories/xtra' Upgraded 'trunk/gar/pkglib' Upgraded 'trunk/gar/pkglib/smf' Upgraded 'trunk/gar/pkglib/csw' Upgraded 'trunk/gar/bin' Upgraded 'trunk/gar/bin/sos12-wrappers' Upgraded 'trunk/gar/etc' daniel at unstable9s :~/ws/libconfuse > svn update Updating '.': U trunk/Makefile U trunk/checksums A Makefile Fetching external item into 'trunk/gar': U trunk/gar/gar.mk U trunk/gar/category.mk U trunk/gar/gar.lib.mk U trunk/gar/gar.svn.mk A trunk/gar/upload-application A trunk/gar/categories/rbgems A trunk/gar/categories/rbgems/category.mk U trunk/gar/categories/cpan/category.mk U trunk/gar/categories/java/category.mk A trunk/gar/categories/loose ... ... ... A trunk/gar/lib/web/templates/ErrorTagDetail.html A trunk/gar/lib/web/templates/MaintainerDetail.html A trunk/gar/lib/web/templates/Srv4List.html A trunk/gar/lib/web/templates/Catalogname.html D trunk/gar/bin/upstream_watch svn: E235000: In file 'subversion/libsvn_wc/update_editor.c' line 1582: assertion failed (action == svn_wc_conflict_action_edit || action == svn_wc_conflict_action_delete || action == svn_wc_conflict_action_replace) Abort (core dumped) > /peter > _______________________________________________ > maintainers mailing list > maintainers at lists.opencsw.org > https://lists.opencsw.org/mailman/listinfo/maintainers > .:: This mailing list's archive is public. ::. > From wilbury at opencsw.org Wed Oct 26 15:19:35 2011 From: wilbury at opencsw.org (Juraj Lutter) Date: Wed, 26 Oct 2011 15:19:35 +0200 Subject: [csw-maintainers] svn error with gar.svn.sourceforge.net In-Reply-To: <4EA8043D.3030105@opencsw.org> References: <4EA7F8AF.4000903@opencsw.org> <4EA8043D.3030105@opencsw.org> Message-ID: <4EA808E7.8030208@opencsw.org> On 10/26/2011 02:59 PM, Daniel Pocock wrote: > On 26/10/11 20:25, Peter Bonivart wrote: >> On Wed, Oct 26, 2011 at 2:10 PM, Daniel Pocock wrote: >> >>> bash-2.03$ hostname >>> current8s >>> >> Try unstable9x or unstable9s instead and see if that helps. >> Works fine on unstable10x. -- Juraj Lutter From dam at opencsw.org Wed Oct 26 21:49:20 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Wed, 26 Oct 2011 21:49:20 +0200 Subject: [csw-maintainers] Fwd: complicated build, subversion client In-Reply-To: References: <20110529181632.GA9828@ted.stsp.name> <20110529223727.GB9828@ted.stsp.name> <1306800816-sup-4668@pinkfloyd.chass.utoronto.ca> Message-ID: Hi Rupert, Am 19.10.2011 um 09:16 schrieb rupert THURNER: > as the python bindings did not compile for more than 6 months, I commented them for the first try of svn-1.7. but I d have a bad feeling if we let this package further than unstable. any help to sort out that build error is greatly appreciated. I noticed that here @$(INSTALL_ENV) gmake -C $(WORKSRC) install-swig-py \ swig_pydir=$(libdir)/python2.3/libsvn \ swig_pydir_extra=$(libdir)/python2.3/svn the definitions of swig_pydir and swig_pydir_extra have been introduced after r12557 (the last release of CSWpythonsvn) in r14397: https://sourceforge.net/apps/trac/gar/changeset/14397/csw/mgar/pkg/subversion/trunk/Makefile If I take the explicit setting of swig_* out it compiles out of the box. As the current Python files are in /opt/csw/lib/python/site-packages the setting looks wrong anyway. I found some postings with the error you descibed indicating that DESTDIR should be used instead of manually fiddling with it. If the pathes need to be relocated this needs to be done earlier during compile, not just on install. Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From dam at opencsw.org Thu Oct 27 14:45:24 2011 From: dam at opencsw.org (Dagobert Michelsen) Date: Thu, 27 Oct 2011 14:45:24 +0200 Subject: [csw-maintainers] glib2: finally some results In-Reply-To: <20111014053800.GC15669@bender.opencsw.org> References: <20111009172646.GA25605@bender.opencsw.org> <85555CAF-0F97-422A-88FA-D91663ED540C@opencsw.org> <20111014053800.GC15669@bender.opencsw.org> Message-ID: <86C49AB2-92F0-4311-9488-09219E18BBAB@opencsw.org> Hi Rafi, Am 14.10.2011 um 07:38 schrieb Rafael Ostertag: > On Thu, Oct 13, 2011 at 12:59:23PM +0200, Dagobert Michelsen wrote: >> Hi Rafi, >> >> Am 09.10.2011 um 19:26 schrieb Rafael Ostertag: >>> I just committed (r15901) the recipe for glib2 which I hope is deployment ready. >>> >>> However, I would appreciate if you could have a look at it and give it >>> a go/no-go. >> >> The recipe looks ok, I am fwd'ing to maintainers for a broader test audience. > > Sounds good to me. Thanks! I have compiled pygobject against glib2 and it looks pretty good. Did you have some progress on an updated gtk2 ? Best regards -- Dago -- "You don't become great by trying to be great, you become great by wanting to do something, and then doing it so hard that you become great in the process." - xkcd #896 From bwalton at opencsw.org Fri Oct 28 04:14:33 2011 From: bwalton at opencsw.org (Ben Walton) Date: Thu, 27 Oct 2011 22:14:33 -0400 Subject: [csw-maintainers] file conflicts and the future... Message-ID: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> Hi All, In developing the web integration code (almost done), I need to deal with packages that might contain a file that conflicts with another package. Traditionally, this has not been allowed. Do we consider this tradition worth maintaining? There are a few cases where a file might conflict with another package: 1. Packages that provide similar functionality that is accessed through the a common name (imapd or sendmail). 2. Related packages that provide build support files (.so, .h) for various alternate program versions. 3. Unrelated programs that happen to ship a file with the same name. The first case can be handled (now) with alternatives and I think that's a good solution. The second case is not as easy to handle with alternatives and we've previously agreed (if memory serves) that it's not a good idea to try handling that way. That leaves either namespacing via the filesystem or declaring the packages incompatible. The third case could be handled with alternatives at the technical level but that doesn't make sense as the conflicting files don't provide similar/workalike functionality. This would lead to a case where either a renamed file (which might make it deviate from other platforms) or an incompatible declaration are required. My question to you is: Are people comfortable moving to a system where declaring a package incompatible with another package is sufficient to deliver conflicting files or should we stick to the traditional "1 file, 1 package" mechanism? Although this question is mostly driven by "how should I write the code to deal with this" it's also a general issue that can (and does[1]) arise in non-error situations. Packages in the current set of "to register" (changes since the last manual run) with conflicts are: berkeleydb3_dev gcc4objc gcc4g++ gcc4core I think the first is the result of not removing an older package (devel -> dev) and the gcc4 packages require an update of the older gcc3 packages to correct. The first will most likely disappear but the gcc4 packages will retain the conflict until gcc3 is cleaned up. If I exempt those packages from generating exceptions during the unique file registration, I can now successfully register all changes since the last manual run. Thanks -Ben [1] http://lists.opencsw.org/pipermail/maintainers/2011-February/014160.html -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From skayser at opencsw.org Fri Oct 28 08:48:40 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 28 Oct 2011 08:48:40 +0200 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: References: Message-ID: <20111028064840.GU26160@sebastiankayser.de> Hey Maciej, * Maciej (Matchek) Blizi??ski wrote: > If you want to fetch a list of packages from a specific catalog, call > an URL like this one... > > http://buildfarm.opencsw.org/pkgdb/rest/catalogs/unstable/sparc/SunOS5.9/ > > ...and deserialize JSON. It takes a while to get that data, so be patient. If there's just a single package that I am interested in (I am after the catalogname to md5 mapping), is there a more specific URL to grab? The page above takes really long to load. Btw., can you navigate the REST tree to see which URLs are available? Like an entry page that people can easily remember (or which is added to buildfarm.opencsw.org) and from where one can explore features at will? Sebastian From pfelecan at opencsw.org Fri Oct 28 09:51:46 2011 From: pfelecan at opencsw.org (Peter FELECAN) Date: Fri, 28 Oct 2011 09:51:46 +0200 Subject: [csw-maintainers] file conflicts and the future... In-Reply-To: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> (Ben Walton's message of "Thu, 27 Oct 2011 22:14:33 -0400") References: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> Message-ID: Ben Walton writes: > My question to you is: Are people comfortable moving to a system where > declaring a package incompatible with another package is sufficient to > deliver conflicting files or should we stick to the traditional "1 > file, 1 package" mechanism? Incompatibility is a good solution for most cases. > Although this question is mostly driven by "how should I write the > code to deal with this" it's also a general issue that can (and > does[1]) arise in non-error situations. > > Packages in the current set of "to register" (changes since the last > manual run) with conflicts are: > berkeleydb3_dev > gcc4objc > gcc4g++ > gcc4core > > I think the first is the result of not removing an older package > (devel -> dev) and the gcc4 packages require an update of the older > gcc3 packages to correct. The first will most likely disappear but > the gcc4 packages will retain the conflict until gcc3 is cleaned up. >From my point of view, gcc3 and gcc4 are incompatible but I think that their conflict should be managed by alternatives. -- Peter From skayser at opencsw.org Fri Oct 28 10:36:17 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Fri, 28 Oct 2011 10:36:17 +0200 Subject: [csw-maintainers] Release process graph // release branch info on opencsw.org In-Reply-To: <20111024231542.GP26160@sebastiankayser.de> References: <20111024231542.GP26160@sebastiankayser.de> Message-ID: <20111028083617.GW26160@sebastiankayser.de> * Sebastian Kayser wrote: > The release branch information on opencsw.org however is still outdated. > Any volunteer to de-duplicate and update the relevant content [2,3]? > Then we could add a frontpage news items (with links to the updated > information) to bring our users up to speed and spare them hunting for > details. > > [2] http://www.opencsw.org/get-it/releases/ > [3] http://www.opencsw.org/use-it/#release Any volunteers? Doesn't have to be prose (IMHO) just a concise, current, and honest description of what people can currently expect. I'll go ahead and do it on Sunday if no one jumps in. Don't be surprised by the strange sentences then though, coz u noo, ai em noht a native spika. ;) Sebastian From bwalton at opencsw.org Fri Oct 28 15:03:11 2011 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 28 Oct 2011 09:03:11 -0400 Subject: [csw-maintainers] file conflicts and the future... In-Reply-To: References: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> Message-ID: <1319806891-sup-8866@pinkfloyd.chass.utoronto.ca> Excerpts from Peter FELECAN's message of Fri Oct 28 03:51:46 -0400 2011: > Ben Walton writes: > > > My question to you is: Are people comfortable moving to a system where > > declaring a package incompatible with another package is sufficient to > > deliver conflicting files or should we stick to the traditional "1 > > file, 1 package" mechanism? > > Incompatibility is a good solution for most cases. Agreed. Does pkgutil do anything special with the I-deps in the catalog at this point or do we rely on pkgadd to handle these cases? > > Although this question is mostly driven by "how should I write the > > code to deal with this" it's also a general issue that can (and > > does[1]) arise in non-error situations. > > > > Packages in the current set of "to register" (changes since the last > > manual run) with conflicts are: > > berkeleydb3_dev > > gcc4objc > > gcc4g++ > > gcc4core > > > > I think the first is the result of not removing an older package > > (devel -> dev) and the gcc4 packages require an update of the older > > gcc3 packages to correct. The first will most likely disappear but > > the gcc4 packages will retain the conflict until gcc3 is cleaned up. > > From my point of view, gcc3 and gcc4 are incompatible but I think that > their conflict should be managed by alternatives. Ok. You'd know better than me here. For the initial import of updated data to the web database, I'm simply going to ignore this conflict. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From bonivart at opencsw.org Fri Oct 28 15:08:34 2011 From: bonivart at opencsw.org (Peter Bonivart) Date: Fri, 28 Oct 2011 15:08:34 +0200 Subject: [csw-maintainers] file conflicts and the future... In-Reply-To: <1319806891-sup-8866@pinkfloyd.chass.utoronto.ca> References: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> <1319806891-sup-8866@pinkfloyd.chass.utoronto.ca> Message-ID: On Fri, Oct 28, 2011 at 3:03 PM, Ben Walton wrote: > Does pkgutil do anything special with the I-deps in the > catalog at this point or do we rely on pkgadd to handle these cases? Pkgutil is aware of I-deps and will uninstall them to avoid conflicts. Last time I checked, around 18 months ago, we had ~30 I-deps declared. (It only took a year or so to convince Phil to add I-deps to the catalog) /peter From bwalton at opencsw.org Fri Oct 28 15:12:16 2011 From: bwalton at opencsw.org (Ben Walton) Date: Fri, 28 Oct 2011 09:12:16 -0400 Subject: [csw-maintainers] file conflicts and the future... In-Reply-To: References: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> <1319806891-sup-8866@pinkfloyd.chass.utoronto.ca> Message-ID: <1319807482-sup-8757@pinkfloyd.chass.utoronto.ca> Excerpts from Peter Bonivart's message of Fri Oct 28 09:08:34 -0400 2011: > Pkgutil is aware of I-deps and will uninstall them to avoid > conflicts. Last time I checked, around 18 months ago, we had ~30 > I-deps declared. Cool, so as long as the user is prompted about this, there should be no reason to avoid their use then, which would let us leverage this fully. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From maciej at opencsw.org Sat Oct 29 11:31:36 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sat, 29 Oct 2011 10:31:36 +0100 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: <20111028064840.GU26160@sebastiankayser.de> References: <20111028064840.GU26160@sebastiankayser.de> Message-ID: 2011/10/28 Sebastian Kayser : > If there's just a single package that I am interested in (I am after the > catalogname to md5 mapping), is there a more specific URL to grab? The > page above takes really long to load. Look at the URL scheme in the source code (pkgdb_web.py), it should be easy to find out what URLs to call. > Btw., can you navigate the REST tree to see which URLs are available? > Like an entry page that people can easily remember (or which is added to > buildfarm.opencsw.org) and from where one can explore features at will? Overall, there's more md5 sums that you can reasonably fetch. So the idea is that you can find packages by: - knowing the md5 sum in advance, because you have the file itself - knowing the catalog (release name, architecture, os release version) and the catalogname or pkgname - knowing the catalog, and enumerating all packages belonging to it I could also add something to enumerate the catalogs. From skayser at opencsw.org Sat Oct 29 11:46:31 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sat, 29 Oct 2011 11:46:31 +0200 Subject: [csw-maintainers] Current state of pkg feed? Integration into our front page? In-Reply-To: References: <20111023103247.GL26160@sebastiankayser.de> <1319377320-sup-5216@pinkfloyd.chass.utoronto.ca> <20111023190633.GO26160@sebastiankayser.de> <20111025131348.GQ26160@sebastiankayser.de> <1319549891-sup-2657@pinkfloyd.chass.utoronto.ca> Message-ID: <20111029094631.GX26160@sebastiankayser.de> * Maciej (Matchek) Blizi??ski wrote: > 2011/10/25 Ben Walton : > > Excerpts from Peter Bonivart's message of Tue Oct 25 09:19:45 -0400 2011: > > > >> I have also seen that as a problem that it's up to the client to limit > >> the scope you see. Since there's no interest in, let's say, more than > >> a month back ?(or even just a week) it would be nice if the process > >> generating the feed truncated it. Simpler clients that can't set the > >> scope would be much fasten then. > > > > Ok, I'll curtail this. ?Is everyone happy with a 1 week interval or is > > a month preferable? ?In the current implementation, there is no > > culling at all. :) > > I'd say a month rather than a week. The feed should still be reasonable in size. Just for the fun of it. Feed item size for recent items on the unstable feed is mostly in the 400 - 450 Bytes range [1]. Judging from our packages statistics [2] we have roughly 150 packages updates per month on average. Thus: 450 Bytes * 150 = ~70 KBytes. More than 400KB down from the current 500KB. Sounds good to me. Except when something like around early October happens (looks like a backlog of processed packages). Then, curtailing based on feed item count might make more sense to control the feed size. Sebastian [1] http://buildfarm.opencsw.org/~skayser/feeditemsize-feed.mirror.opencsw.org.opencsw.unstable.SunOS5.10.i386.atom.xml.png [2] http://www.opencsw.org/get-it/package-statistics/ From skayser at opencsw.org Sat Oct 29 20:53:08 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sat, 29 Oct 2011 20:53:08 +0200 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: References: <20111028064840.GU26160@sebastiankayser.de> Message-ID: <20111029185308.GD26160@sebastiankayser.de> * Maciej (Matchek) Blizi??ski wrote: > 2011/10/28 Sebastian Kayser : > > If there's just a single package that I am interested in (I am after the > > catalogname to md5 mapping), is there a more specific URL to grab? The > > page above takes really long to load. > > Look at the URL scheme in the source code (pkgdb_web.py), it should be > easy to find out what URLs to call. Thanks, found the list of URLs and also the kind of URL that I was looking for. Retrieve information for a single package: http://buildfarm.opencsw.org/pkgdb/rest/catalogs/unstable/i386/SunOS5.9/catalognames/mbuffer/ Next up, teach this to ceeswi. > > Btw., can you navigate the REST tree to see which URLs are available? > > Like an entry page that people can easily remember (or which is added to > > buildfarm.opencsw.org) and from where one can explore features at will? > > Overall, there's more md5 sums that you can reasonably fetch. So the > idea is that you can find packages by: > > - knowing the md5 sum in advance, because you have the file itself > - knowing the catalog (release name, architecture, os release version) > and the catalogname or pkgname > - knowing the catalog, and enumerating all packages belonging to it > > I could also add something to enumerate the catalogs. That's fully sufficient. Maybe add a link to pkgdb_web.py [1] on the pkgdb/ index page so that people who want to script stuff get a pointer to what's available. Sebastian [1] http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/lib/web/pkgdb_web.py#L16 From maciej at opencsw.org Sun Oct 30 02:00:01 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 30 Oct 2011 01:00:01 +0100 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: <20111029185308.GD26160@sebastiankayser.de> References: <20111028064840.GU26160@sebastiankayser.de> <20111029185308.GD26160@sebastiankayser.de> Message-ID: 2011/10/29 Sebastian Kayser : >> I could also add something to enumerate the catalogs. > > That's fully sufficient. Maybe add a link to pkgdb_web.py [1] on the > pkgdb/ index page so that people who want to script stuff get a pointer > to what's available. Good idea, done. From maciej at opencsw.org Sun Oct 30 09:27:53 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Sun, 30 Oct 2011 08:27:53 +0000 Subject: [csw-maintainers] file conflicts and the future... In-Reply-To: References: <1319766606-sup-4116@pinkfloyd.chass.utoronto.ca> Message-ID: 2011/10/28 Peter FELECAN : >> Although this question is mostly driven by "how should I write the >> code to deal with this" it's also a general issue that can (and >> does[1]) arise in non-error situations. >> >> Packages in the current set of "to register" (changes since the last >> manual run) with conflicts are: >> berkeleydb3_dev >> gcc4objc >> gcc4g++ >> gcc4core >> >> I think the first is the result of not removing an older package >> (devel -> dev) and the gcc4 packages require an update of the older >> gcc3 packages to correct. ?The first will most likely disappear but >> the gcc4 packages will retain the conflict until gcc3 is cleaned up. > > From my point of view, gcc3 and gcc4 are incompatible but I think that > their conflict should be managed by alternatives. The main conflict is currently the *.so files in /opt/csw/lib. Both gcc3 and gcc4 provide for example /opt/csw/lib/libgcj.so. The main question is whether gcc3 can work correctly with /opt/csw/lib/libgcj.so from gcc4. If not, the gcc3 package will need to do a similar thing to the one we did with the libnet package: provide /opt/csw/lib/gcc3/libgcj.so and add -L/opt/csw/lib/gcc3 to linker invocations. Maciej From rupert at opencsw.org Sun Oct 30 10:45:19 2011 From: rupert at opencsw.org (rupert THURNER) Date: Sun, 30 Oct 2011 10:45:19 +0100 Subject: [csw-maintainers] new git version 1.7.7.1 please ... Message-ID: hi, https://raw.github.com/gitster/git/master/Documentation/RelNotes/1.7.7.1.txt says that a quadratic loop on submodules is fixed. some time ago i asked on the git mailing list why git with submodules is so slow ( http://marc.info/?l=git&m=131447206810156&w=2 ) and i'd be glad to try if the fix has an effect on what i noticed using submodules on the opencsw repository. could you please upgrade the git version to 1.7.7.1 ? rupert. From bwalton at opencsw.org Sun Oct 30 13:22:23 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 30 Oct 2011 08:22:23 -0400 Subject: [csw-maintainers] new git version 1.7.7.1 please ... In-Reply-To: References: Message-ID: <1319977100-sup-984@pinkfloyd.chass.utoronto.ca> Excerpts from rupert THURNER's message of Sun Oct 30 05:45:19 -0400 2011: Hi Rupert, > https://raw.github.com/gitster/git/master/Documentation/RelNotes/1.7.7.1.txt > says that a quadratic loop on submodules is fixed. some time ago i > asked on the git mailing list why git with submodules is so slow ( > http://marc.info/?l=git&m=131447206810156&w=2 ) and i'd be glad to > try if the fix has an effect on what i noticed using submodules on > the opencsw repository. could you please upgrade the git version to > 1.7.7.1 ? I'd looked at the 1.7.7 upgrade but at the time, the newer subversion release was seeing tests that broke. I didn't have the time to look then. In the meantime, the new subversion has gone release (it was beta something then) and 1.7.7.1 has been released. I'll get an update out asap. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From skayser at opencsw.org Sun Oct 30 20:17:14 2011 From: skayser at opencsw.org (Sebastian Kayser) Date: Sun, 30 Oct 2011 20:17:14 +0100 Subject: [csw-maintainers] Release process graph // release branch info on opencsw.org In-Reply-To: <20111028083617.GW26160@sebastiankayser.de> References: <20111024231542.GP26160@sebastiankayser.de> <20111028083617.GW26160@sebastiankayser.de> Message-ID: <20111030191714.GE26160@sebastiankayser.de> * Sebastian Kayser wrote: > * Sebastian Kayser wrote: > > The release branch information on opencsw.org however is still outdated. > > Any volunteer to de-duplicate and update the relevant content [2,3]? > > Then we could add a frontpage news items (with links to the updated > > information) to bring our users up to speed and spare them hunting for > > details. > > > > [2] http://www.opencsw.org/get-it/releases/ > > [3] http://www.opencsw.org/use-it/#release > > Any volunteers? Doesn't have to be prose (IMHO) just a concise, current, > and honest description of what people can currently expect. I'll go > ahead and do it on Sunday if no one jumps in. Don't be surprised by the > strange sentences then though, coz u noo, ai em noht a native spika. ;) Done on www-mockup. Barring veto, I'll push it to www tomorrow and we can always refine it from there. http://www-mockup.opencsw.org/get-it/releases/ Release branch page updated, related information from the user guide deleted (as long as there's no one actively taking care of the website, less redundant information makes it easier to maintain). Sebastian From bwalton at opencsw.org Mon Oct 31 01:06:41 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 30 Oct 2011 20:06:41 -0400 Subject: [csw-maintainers] new git version 1.7.7.1 please ... In-Reply-To: <1319977100-sup-984@pinkfloyd.chass.utoronto.ca> References: <1319977100-sup-984@pinkfloyd.chass.utoronto.ca> Message-ID: <1320019393-sup-1368@pinkfloyd.chass.utoronto.ca> Excerpts from Ben Walton's message of Sun Oct 30 08:22:23 -0400 2011: Hi Rupert, > I'd looked at the 1.7.7 upgrade but at the time, the newer > subversion release was seeing tests that broke. I didn't have the > time to look then. In the meantime, the new subversion has gone > release (it was beta something then) and 1.7.7.1 has been released. > I'll get an update out asap. I just filed a bug on CSWpmsvn. The SVN::Ra binding (which Git::SVN::Ra uses) breaks with URL's containing spaces. When run under the git test suite, the file:// url has a space. I did a bit of digging into the subversion code but not a lot. I'm happy to run the tests with the svn ones disabled to ensure everything else passes and then release the update. (I imagine that the current git release would suffer this same fate on the new svn anyway, so I doubt we're farther behind.) Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From bwalton at opencsw.org Mon Oct 31 03:48:05 2011 From: bwalton at opencsw.org (Ben Walton) Date: Sun, 30 Oct 2011 22:48:05 -0400 Subject: [csw-maintainers] potentially useful script in gar Message-ID: <1320029191-sup-2430@pinkfloyd.chass.utoronto.ca> Hi All, I just added find_unpackaged to gar's bin/ directory. If you point it as your build-global directory after prototypes have been built, it will give you a list of files not included in any of the package prototypes. It's a quick hack at this point, but it might be something worth integrating directly into the GAR run later... Comments welcome. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From trygvis at opencsw.org Mon Oct 31 12:07:31 2011 From: trygvis at opencsw.org (=?ISO-8859-1?Q?Trygve_Laugst=F8l?=) Date: Mon, 31 Oct 2011 12:07:31 +0100 Subject: [csw-maintainers] Full package statistics via REST In-Reply-To: <20111028064840.GU26160@sebastiankayser.de> References: <20111028064840.GU26160@sebastiankayser.de> Message-ID: <4EAE8173.5040001@opencsw.org> On 10/28/2011 08:48 AM, Sebastian Kayser wrote: > Hey Maciej, > > * Maciej (Matchek) Blizi??ski wrote: >> If you want to fetch a list of packages from a specific catalog, call >> an URL like this one... >> >> http://buildfarm.opencsw.org/pkgdb/rest/catalogs/unstable/sparc/SunOS5.9/ >> >> ...and deserialize JSON. It takes a while to get that data, so be patient. > > If there's just a single package that I am interested in (I am after the > catalogname to md5 mapping), is there a more specific URL to grab? The > page above takes really long to load. > > Btw., can you navigate the REST tree to see which URLs are available? > Like an entry page that people can easily remember (or which is added to > buildfarm.opencsw.org) and from where one can explore features at will? The RESTful way to do this is to use a hypermedia. What a hypermedia does is to encode how to find the URLs *in* the document. I made one practical example [1]. I've tried to match the names used in [2]. Not that HTML is a quite restriced hypermedia so it can only generate URLs with query parameters and not path segments. If we made our own hypermedia we could use URL templates which would be able to generate more types of URLs. Sebastian: if you want to go this way take a look at Mechanize, it's a nice library for using forms. [1]: http://buildfarm.opencsw.org/~trygvis/pkgdb-queries.html [2]: http://sourceforge.net/apps/trac/gar/browser/csw/mgar/gar/v2/lib/web/pkgdb_web.py#L16 -- Trygve From maciej at opencsw.org Mon Oct 31 16:15:14 2011 From: maciej at opencsw.org (=?UTF-8?Q?Maciej_=28Matchek=29_Blizi=C5=84ski?=) Date: Mon, 31 Oct 2011 15:15:14 +0000 Subject: [csw-maintainers] potentially useful script in gar In-Reply-To: <1320029191-sup-2430@pinkfloyd.chass.utoronto.ca> References: <1320029191-sup-2430@pinkfloyd.chass.utoronto.ca> Message-ID: In GAR, there is a notion of a catch-all package. If it's not present, GAR could emit a warning with the list of files - it already has all the information. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bwalton at opencsw.org Mon Oct 31 16:23:00 2011 From: bwalton at opencsw.org (Ben Walton) Date: Mon, 31 Oct 2011 11:23:00 -0400 Subject: [csw-maintainers] potentially useful script in gar In-Reply-To: References: <1320029191-sup-2430@pinkfloyd.chass.utoronto.ca> Message-ID: <1320074480-sup-5963@pinkfloyd.chass.utoronto.ca> Excerpts from Maciej (Matchek) Blizi?ski's message of Mon Oct 31 11:15:14 -0400 2011: > In GAR, there is a notion of a catch-all package. If it's not present, GAR > could emit a warning with the list of files - it already has all the > information. Ok, I think my particular case is caused by having the catchall package use a specific set of files and become a metapackage of sorts (bacula). CSWbacula pulls in a few other packages that also have specific PKGFILES definitions. Thus, I override the default gar ability. Thanks -Ben -- Ben Walton Systems Programmer - CHASS University of Toronto C:416.407.5610 | W:416.978.4302 From rupert at opencsw.org Mon Oct 31 18:40:18 2011 From: rupert at opencsw.org (rupert THURNER) Date: Mon, 31 Oct 2011 18:40:18 +0100 Subject: [csw-maintainers] Fwd: Re: libsvn_subr/dirent_uri.c dumps core when url contains a space In-Reply-To: <20111031092923.GA14858@ted.stsp.name> References: <20111031092923.GA14858@ted.stsp.name> Message-ID: fyi ---------- Forwarded message ---------- From: "Stefan Sperling" Date: Oct 31, 2011 10:29 AM Subject: Re: libsvn_subr/dirent_uri.c dumps core when url contains a space To: "rupert.thurner" Cc: On Sun, Oct 30, 2011 at 10:09:52PM -0700, rupert.thurner wrote: > on solaris the perl bindings dump a core when an url contains a space, > see https://www.opencsw.org/mantis/view.php?id=4854. an excerpt: > > The SVN::Ra bindings core dump with the following error: > > $ perl ./test.pl > svn: E235000: In file 'subversion/libsvn_subr/dirent_uri.c' line 2291: > assertion failed (svn_uri_is_canonical(url, pool)) > Abort (core dumped) > > > The content of test.pl: > > require SVN::Ra; > > $s = SVN::Ra->new('file:///home/bwalton/opencsw/git/trunk/work/ > solaris9-sparc/build-isa-sparcv8/git-1.7.7.1/t/trash [^] > directory.t9134-git-svn-ignore-paths/svnrepo'); Your code is wrong. You have to canonicalize URLs and paths before passing them into the SVN API. Something like the following might work better (untested): require SVN::Core; require SVN::Ra; my $url = SVN::Core->svn_uri_canonicalize('file:///home/bwalton/opencsw/git/trunk/work/solaris9-sparc/build-isa-sparcv8/git-1.7.7.1/t/trash [^] directory.t9134-git-svn-ignore-paths/svnrepo') $s = SVN::Ra->new($url); -------------- next part -------------- An HTML attachment was scrubbed... URL: