From matti.p.airas at nokia.com Thu Mar 1 14:53:04 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 01 Mar 2012 15:53:04 +0200 Subject: [PySide] Call for help: regression bugs to be fixed In-Reply-To: <20120229204036.GA32400@mancubus.hell> References: <4F4CEA0D.9080406@nokia.com> <4F4CFA88.9040907@nokia.com> <20120228184012.GB11734@mancubus.hell> <4F4DFCAC.2020200@nokia.com> <20120229204036.GA32400@mancubus.hell> Message-ID: <4F4F7F40.4000402@nokia.com> On 29.02.2012 22:40, ext Andy Kittner wrote: > I just gave it a try with #5, and apparently it worked: > http://codereview.qt-project.org/#change,18299 > Issue #4 also sounded vaguely familiar, so I gave it a shot as well ;) > http://codereview.qt-project.org/#change,18300 Hurrah! > > I also looked at issue 3, but assertNotEqual does not appear in the > mentioned file, and the test passes for me as well (I'm using a debug > build of the 2.7 branch from the cpython mercurial repo, about a week > old or so). Oops. The offending syntax was changed later on in the repo history, so the commit wasn't used any longer. I had made a remark in my notes but managed to miss it. I must be getting sloppy. I resolved the bug invalid. Three bugs down, three to go! Anyone? :-) Cheers, ma. From hugo.lima at openbossa.org Thu Mar 1 15:30:58 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 1 Mar 2012 11:30:58 -0300 Subject: [PySide] Call for help: regression bugs to be fixed In-Reply-To: <4F4F7F40.4000402@nokia.com> References: <4F4CEA0D.9080406@nokia.com> <20120229204036.GA32400@mancubus.hell> <4F4F7F40.4000402@nokia.com> Message-ID: <201203011131.08210.hugo.lima@openbossa.org> On Thursday 01 March 2012 10:53:04 Matti Airas wrote: > On 29.02.2012 22:40, ext Andy Kittner wrote: > > I just gave it a try with #5, and apparently it worked: > > http://codereview.qt-project.org/#change,18299 > > Issue #4 also sounded vaguely familiar, so I gave it a shot as well ;) > > http://codereview.qt-project.org/#change,18300 I can't review the patch, got the following error on gerrit: Application Error Server Error Requires Code Review > Hurrah! > > > I also looked at issue 3, but assertNotEqual does not appear in the > > mentioned file, and the test passes for me as well (I'm using a debug > > build of the 2.7 branch from the cpython mercurial repo, about a week > > old or so). > > Oops. The offending syntax was changed later on in the repo history, so > the commit wasn't used any longer. I had made a remark in my notes but > managed to miss it. I must be getting sloppy. > > I resolved the bug invalid. > > Three bugs down, three to go! Anyone? :-) > > Cheers, > > ma. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matti.p.airas at nokia.com Thu Mar 1 18:58:01 2012 From: matti.p.airas at nokia.com (matti.p.airas at nokia.com) Date: Thu, 1 Mar 2012 17:58:01 +0000 Subject: [PySide] Call for help: regression bugs to be fixed In-Reply-To: <201203011131.08210.hugo.lima@openbossa.org> References: <4F4CEA0D.9080406@nokia.com> <20120229204036.GA32400@mancubus.hell> <4F4F7F40.4000402@nokia.com> <201203011131.08210.hugo.lima@openbossa.org> Message-ID: <69E0510F-48E4-40B4-A339-2A820371F4C4@nokia.com> On Mar 1, 2012, at 16:30, ext Hugo Parente Lima wrote: > On Thursday 01 March 2012 10:53:04 Matti Airas wrote: >> On 29.02.2012 22:40, ext Andy Kittner wrote: >>> I just gave it a try with #5, and apparently it worked: >>> http://codereview.qt-project.org/#change,18299 >>> Issue #4 also sounded vaguely familiar, so I gave it a shot as well ;) >>> http://codereview.qt-project.org/#change,18300 > > I can't review the patch, got the following error on gerrit: > > Application Error > Server Error > Requires Code Review This is probably because you are not members of the Approvers group in Gerrit. I sent you and Marcelo instructions yesterday about how to request yourselves to be added there. Please read that mail and act accordingly; that should get the problem fixed. Cheers, ma. From techtonik at gmail.com Fri Mar 2 09:55:13 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Fri, 2 Mar 2012 10:55:13 +0200 Subject: [PySide] Call for help: regression bugs to be fixed In-Reply-To: <69E0510F-48E4-40B4-A339-2A820371F4C4@nokia.com> References: <4F4CEA0D.9080406@nokia.com> <20120229204036.GA32400@mancubus.hell> <4F4F7F40.4000402@nokia.com> <201203011131.08210.hugo.lima@openbossa.org> <69E0510F-48E4-40B4-A339-2A820371F4C4@nokia.com> Message-ID: On Thu, Mar 1, 2012 at 8:58 PM, wrote: > On Mar 1, 2012, at 16:30, ext Hugo Parente Lima wrote: >> On Thursday 01 March 2012 10:53:04 Matti Airas wrote: >>> On 29.02.2012 22:40, ext Andy Kittner wrote: >>>> I just gave it a try with #5, and apparently it worked: >>>> http://codereview.qt-project.org/#change,18299 >>>> Issue #4 also sounded vaguely familiar, so I gave it a shot as well ;) >>>> http://codereview.qt-project.org/#change,18300 >> >> I can't review the patch, got the following error on gerrit: >> >> Application Error >> Server Error >> Requires Code Review > > This is probably because you are not members of the Approvers group in Gerrit. I sent you and Marcelo instructions yesterday about how to request yourselves to be added there. Please read that mail and act accordingly; that should get the problem fixed. Any progress with opening Gerrit to general public? -- anatoly t. From matti.p.airas at nokia.com Fri Mar 2 10:49:35 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Fri, 02 Mar 2012 11:49:35 +0200 Subject: [PySide] Call for help: regression bugs to be fixed In-Reply-To: References: <4F4CEA0D.9080406@nokia.com> <20120229204036.GA32400@mancubus.hell> <4F4F7F40.4000402@nokia.com> <201203011131.08210.hugo.lima@openbossa.org> <69E0510F-48E4-40B4-A339-2A820371F4C4@nokia.com> Message-ID: <4F5097AF.5060200@nokia.com> On 02.03.2012 10:55, ext anatoly techtonik wrote: > Any progress with opening Gerrit to general public? Hi, Gerrit *is* open to the general public. The issue was anonymous access; that is now provided using the by mirroring the Gerrit repo to a read-only repository at Gitorious. This is the same setup as for all other Qt repos (and indeed, the mirroring is done by Qt, so there's little chance of it breaking up without anyone noticing. Cheers, ma. From techtonik at gmail.com Fri Mar 2 10:57:18 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Fri, 2 Mar 2012 11:57:18 +0200 Subject: [PySide] Open Gerrit to public (Was: Call for help: regression bugs to be fixed) Message-ID: On Fri, Mar 2, 2012 at 12:49 PM, Matti Airas wrote: > On 02.03.2012 10:55, ext anatoly techtonik wrote: >> >> Any progress with opening Gerrit to general public? > > Gerrit *is* open to the general public. The issue was anonymous access; that > is now provided using the by mirroring the Gerrit repo to a read-only > repository at Gitorious. This is the same setup as for all other Qt repos > (and indeed, the mirroring is done by Qt, so there's little chance of it > breaking up without anyone noticing. Well, practical question then - how can I view the links below in Gitorious? >>>> I just gave it a try with #5, and apparently it worked: >>>> http://codereview.qt-project.org/#change,18299 >>>> Issue #4 also sounded vaguely familiar, so I gave it a shot as well ;) >>>> http://codereview.qt-project.org/#change,18300 BTW, are GitHub repos mirrored by Qt too? -- anatoly t. From matti.p.airas at nokia.com Fri Mar 2 11:41:57 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Fri, 02 Mar 2012 12:41:57 +0200 Subject: [PySide] Open Gerrit to public (Was: Call for help: regression bugs to be fixed) In-Reply-To: References: Message-ID: <4F50A3F5.9010605@nokia.com> On 02.03.2012 11:57, ext anatoly techtonik wrote: > On Fri, Mar 2, 2012 at 12:49 PM, Matti Airas wrote: >> On 02.03.2012 10:55, ext anatoly techtonik wrote: >>> Any progress with opening Gerrit to general public? >> Gerrit *is* open to the general public. The issue was anonymous access; that >> is now provided using the by mirroring the Gerrit repo to a read-only >> repository at Gitorious. This is the same setup as for all other Qt repos >> (and indeed, the mirroring is done by Qt, so there's little chance of it >> breaking up without anyone noticing. > Well, practical question then - how can I view the links below in Gitorious? Hint: A bit of constructive attitude will bring you far. Gitorious just mirrors the repositories, not the code review functionality. To view the commits, login to Gerrit. As the login prompt states, create an account in bugreports.qt-project.org (I believe you already have one). Then, use those credentials to login to Gerrit. > BTW, are GitHub repos mirrored by Qt too? -- anatoly t. No, the GitHub repos are unofficial and not part of PySide project setup. Cheers, ma. From techtonik at gmail.com Fri Mar 2 19:45:57 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Fri, 2 Mar 2012 21:45:57 +0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> Message-ID: On Wed, Feb 29, 2012 at 10:45 PM, paulo alcantara wrote: > Hi Hugo, > > On Wed, Feb 29, 2012 at 2:58 PM, Hugo Parente Lima > wrote: >> >> Hi, >> >> Just raising a small issue: We have enough bug fixes to do a release but due >> to this long transition to qt-projects plus the guys that didn't accepted the >> CLA for they patches, mostly of the bugfixes are on github repository and the >> github repository already differs a lot from the gitorious and gerrit >> repositories so the question is: >> >> Should I do a release based on which repository? The most recent github with >> new bugs fixed or the gerit without the new fixes plus minor regressions due >> to people that didn't accepted the CLA? >> > > There's some people waiting for the new PySide release based on the recently > bugs fixed available on the github repository - some of them are important and > should be possibly part of the new release, IMO. > >> A hard choice... because if I do the release based on github I'll have a nice >> PySide release with plenty of bugs fixed but what to do next release? i.e. it >> doesn't really solves the problem. >> >> If I go for gerrit, maybe there's not even enough fixes for a release plus I >> don't want to spend my free time at home doing the re-work of applying patches >> and solving conflicts on the new repository instead of fixing new bugs. >> > > I agree with you. >> While there's not a complete answer for that I can't do a release and PySide >> stays in this limbo with active development but without a release :-/, for >> sure PySide can't be that way forever. >> > > I think that a new release is kind of *urgent* currently. > >> To increase the mess yesterday I pushed commits into Shiboken to simplify >> PySide releases and development, ApiExtractor is now a static library and >> GeneratorRunner is no more, so you will only need two repositories to compile >> PySide, the one for the generator (Shiboken) and PySide itself. >> > > This will really make our life easier. Great work, Hugo! +1 on everything above. There is no reason to delay release, because those legal fixes are only matter for Qt Project, not for PySide users. Moreover I see no reasons why Git experts can not replay patches from GitHub back to Gerrit to keep it in sync while we are seeking volunteers for these Qt Project fixes. On Wed, Feb 29, 2012 at 10:07 PM, wrote: > > If the release would be made from Github, the exact history would be lost once the commits are merged back to the main repo. I don't understand what history will be lost? Tags? Yes, the hash will change and the release will be broken, and if I understand correctly, every tag in new Gerrit repository is broken now, because some changesets are absent from history. Am I right? -- anatoly t. From techtonik at gmail.com Fri Mar 2 20:54:25 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Fri, 2 Mar 2012 22:54:25 +0300 Subject: [PySide] Open Gerrit to public (Was: Call for help: regression bugs to be fixed) In-Reply-To: <4F50A3F5.9010605@nokia.com> References: <4F50A3F5.9010605@nokia.com> Message-ID: On Fri, Mar 2, 2012 at 1:41 PM, Matti Airas wrote: > On 02.03.2012 11:57, ext anatoly techtonik wrote: >> >> On Fri, Mar 2, 2012 at 12:49 PM, Matti Airas >>  wrote: >>> >>> On 02.03.2012 10:55, ext anatoly techtonik wrote: >>>> >>>> Any progress with opening Gerrit to general public? >>> >>> Gerrit *is* open to the general public. The issue was anonymous access; >>> that >>> is now provided using the by mirroring the Gerrit repo to a read-only >>> repository at Gitorious. This is the same setup as for all other Qt repos >>> (and indeed, the mirroring is done by Qt, so there's little chance of it >>> breaking up without anyone noticing. >> >> Well, practical question then - how can I view the links below in >> Gitorious? > > Hint: A bit of constructive attitude will bring you far. I am as constructive as possible. The contributor agreement already puts a high entry barrier for the new people, and further restrictions on accessibility of developer information will do no good. The problem with Gerrit links is that they are posted to the mailing list. This list is read by users, who learn PySide. These users are potential contributors. They might not had the need to write to the list, so they probably read it from the web, and are not even subscribed. If we say that 20% of users read PySide from the web, that's a pretty big amount. Let's assume that 80% of readers are subscribed. About 10% of them have Jira account. About 50% of them are even logged in (because Jira doesn't allow to OAuth/OpenID logins - that's a problem). So only 4% can view those Gerrit links. I think that's a pretty low percentage to take this issue seriously. > Gitorious just mirrors the repositories, not the code review functionality. > To view the commits, login to Gerrit. > > As the login prompt states, create an account in bugreports.qt-project.org > (I believe you already have one). Then, use those credentials to login to > Gerrit. Why not just open Gerrit to anonymous read-only? >> BTW, are GitHub repos mirrored by Qt too? -- anatoly t. > > No, the GitHub repos are unofficial and not part of PySide project setup. But if they are more convenient and stable that those provided by Qt Project, why can't Qt Project sync to them? -- anatoly t. From public at francisco-garcia.net Sun Mar 4 14:46:01 2012 From: public at francisco-garcia.net (=?iso-8859-1?Q?Francisco_Garc=EDa?=) Date: Sun, 4 Mar 2012 14:46:01 +0100 Subject: [PySide] Development with MacPorts Message-ID: <65790526-9E9B-4535-97CD-DD62FAC13B58@francisco-garcia.net> Hi! I would like to know if Mac / Pyside developers have problems using MacPorts The thing is that MacPorts build Qt as a library and not as a framework. I noticed that because of that I get lots of problems debugging my apps. Since the official Qt installation for Mac is a framework I suppose that most libraries and related Qt tools are better tested when built/linked that way. If I want to develop/debug with Pyside, iPython... I make my life easier installing as a framework, but if I do that, many Qt tools from macports fail to compile. For example: Qt So what do you do? Dont you have problems at all with Qt default install from Macports? cheers, Francisco From th.perl at gmail.com Sun Mar 4 16:16:31 2012 From: th.perl at gmail.com (Thomas Perl) Date: Sun, 4 Mar 2012 16:16:31 +0100 Subject: [PySide] Development with MacPorts In-Reply-To: <65790526-9E9B-4535-97CD-DD62FAC13B58@francisco-garcia.net> References: <65790526-9E9B-4535-97CD-DD62FAC13B58@francisco-garcia.net> Message-ID: Hi, 2012/3/4 Francisco García : > So what do you do? Dont you have problems at all with Qt default install from Macports? When building PySide on OS X, I usually use Homebrew and/or the Qt from the Qt SDK: http://mxcl.github.com/homebrew/ http://qt.nokia.com/downloads For automating the build itself, I use the PySide buildscripts: https://github.com/PySide/BuildScripts HTH :) Thomas From juhapekka.piiroinen at gmail.com Mon Mar 5 07:09:56 2012 From: juhapekka.piiroinen at gmail.com (Juhapekka Piiroinen) Date: Mon, 5 Mar 2012 08:09:56 +0200 Subject: [PySide] Fix for PYSIDE-7 QDateTime does not support the 6-argument format Message-ID: Hi, I was browsing the bugs for PYSIDE just for kicks and I noticed one which I thought that I could fix, so I tried to fix the PYSIDE-7 and I pushed the changes to gerrit, could somebody take a look and do a review? http://codereview.qt-project.org/#change,18728 https://bugreports.qt-project.org/browse/PYSIDE-7 Cheers, Juhapekka Piiroinen -------------- next part -------------- An HTML attachment was scrubbed... URL: From juhapekka.piiroinen at gmail.com Mon Mar 5 07:23:14 2012 From: juhapekka.piiroinen at gmail.com (Juhapekka Piiroinen) Date: Mon, 5 Mar 2012 08:23:14 +0200 Subject: [PySide] Fixed bug PYSIDE-2 - pkgconfig libdir variable set incorrectly in 64-bit environments Message-ID: Hi, I have fixed the bug PYSIDE-2 "pkgconfig libdir variable set incorrectly in 64-bit environments" which seemed a trivial one for me. Could somebody review the code: http://codereview.qt-project.org/#change,18729 Cheers, Juhapekka Piiroinen -------------- next part -------------- An HTML attachment was scrubbed... URL: From john.veness.pyside at pelago.org.uk Mon Mar 5 10:48:24 2012 From: john.veness.pyside at pelago.org.uk (John Veness) Date: Mon, 05 Mar 2012 09:48:24 +0000 Subject: [PySide] Open Gerrit to public (Was: Call for help: regression bugs to be fixed) In-Reply-To: References: <4F50A3F5.9010605@nokia.com> Message-ID: <4F548BE8.1050307@pelago.org.uk> This Gerrit-requiring-login issue is a Qt-wide problem, not specific to PySide. I notice that there is already a bugtracker issue raised against it - https://bugreports.qt-project.org/browse/QTQAINFRA-357 . It might be more productive to discuss this on one of the Qt mailing lists (see http://lists.qt-project.org/mailman/listinfo). John From techtonik at gmail.com Mon Mar 5 11:02:23 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Mon, 5 Mar 2012 13:02:23 +0300 Subject: [PySide] Open Gerrit to public (Was: Call for help: regression bugs to be fixed) In-Reply-To: <4F548BE8.1050307@pelago.org.uk> References: <4F50A3F5.9010605@nokia.com> <4F548BE8.1050307@pelago.org.uk> Message-ID: On Mon, Mar 5, 2012 at 12:48 PM, John Veness wrote: > This Gerrit-requiring-login issue is a Qt-wide problem, not specific to > PySide. I notice that there is already a bugtracker issue raised against > it - https://bugreports.qt-project.org/browse/QTQAINFRA-357 . It might > be more productive to discuss this on one of the Qt mailing lists (see > http://lists.qt-project.org/mailman/listinfo). Thanks for the heads up. It is nice to see I am not alone to take it seriously. There is a hint that Jira is the reason, but no technical explanation. It is very unusual that open process of open source project with open government is locked be the dependency from proprietary product. At least that's my impression. I'll write to GEG list. It is not too active, so let's hope there are people who can help. -- anatoly t. From anderson.lizardo at openbossa.org Mon Mar 5 13:41:31 2012 From: anderson.lizardo at openbossa.org (Anderson Lizardo) Date: Mon, 5 Mar 2012 08:41:31 -0400 Subject: [PySide] PySide mailing list & other project facilities In-Reply-To: <4F3E0F35.30105@nokia.com> References: <4F3E0F35.30105@nokia.com> Message-ID: Hi Matti, On Fri, Feb 17, 2012 at 4:26 AM, Matti Airas wrote: > All Qt projects use Qt's Jira bugtracker, and so will PySide. The new > bugtracker can be found at: > > https://bugreports.qt-project.org/ > > Unfortunately, it is not yet possible to report new bugs until Marcelo Lira > has added version numbers to the different PySide components. Nevertheless, > I recommend to refrain from opening new bugs in the old Bugzilla from now > on. > > I am unsure about what would be the best way to deal with the existing > Bugzilla bugs. I'd prefer disabling creating new bugs there as soon as the > kinks with Jira have been ironed out, but there are currently 67 open bugs. > One possibility would be to continue tracking them in Bugzilla in parallel > with the new Jira bugtracker, but I don't know if they can be resolved > before the old server goes away, so I'd be hesitant to take that approach. > My proposal is that we simply comment on the open bugs and ask the reporters > to re-submit the bugs to Jira if they still regard the bugs valid. So, now that Jira setup is okay, can we proceed with this approach (I've seen no comments with other ideas)? Regards, -- Anderson Lizardo Instituto Nokia de Tecnologia - INdT Manaus - Brazil From matti.p.airas at nokia.com Mon Mar 5 18:31:11 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Mon, 05 Mar 2012 19:31:11 +0200 Subject: [PySide] Fixed bug PYSIDE-2 - pkgconfig libdir variable set incorrectly in 64-bit environments In-Reply-To: References: Message-ID: <4F54F85F.4000906@nokia.com> On 05.03.2012 08:23, ext Juhapekka Piiroinen wrote: > Hi, > > I have fixed the bug PYSIDE-2 "pkgconfig libdir variable set > incorrectly in 64-bit environments" which seemed a trivial one for me. > > Could somebody review the code: > http://codereview.qt-project.org/#change,18729 This is great! Thanks a lot! Only PYSIDE-6 to go. (Took some time today to verify that it still is an issue.) Cheers, ma. -------------- next part -------------- An HTML attachment was scrubbed... URL: From matti.p.airas at nokia.com Mon Mar 5 18:39:21 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Mon, 05 Mar 2012 19:39:21 +0200 Subject: [PySide] PySide mailing list & other project facilities In-Reply-To: References: <4F3E0F35.30105@nokia.com> Message-ID: <4F54FA49.3080006@nokia.com> On 05.03.2012 14:41, ext Anderson Lizardo wrote: > Hi Matti, > > On Fri, Feb 17, 2012 at 4:26 AM, Matti Airas wrote: >> I am unsure about what would be the best way to deal with the existing >> Bugzilla bugs. I'd prefer disabling creating new bugs there as soon as the >> kinks with Jira have been ironed out, but there are currently 67 open bugs. >> One possibility would be to continue tracking them in Bugzilla in parallel >> with the new Jira bugtracker, but I don't know if they can be resolved >> before the old server goes away, so I'd be hesitant to take that approach. >> My proposal is that we simply comment on the open bugs and ask the reporters >> to re-submit the bugs to Jira if they still regard the bugs valid. > So, now that Jira setup is okay, can we proceed with this approach > (I've seen no comments with other ideas)? I'm OK with that (but would like to get confirmation that this is the generally preferred approach). I've planned to make a blog post and mailing list announcements about the PySide project migration once we have the last migration bug fixed. I could amend the draft post to comment on the bug tracker migration. Then I could add a notification about the new bug tracker location on the Bugzilla front page and disable creation of new bugs in Bugzilla. I could also make a bulk comment on the old Bugzilla, requesting bug reporters to check whether the bug is still valid and to re-submit it in Jira. Does this sound OK to you? Cheers, ma. From juhapekka.piiroinen at gmail.com Tue Mar 6 12:54:41 2012 From: juhapekka.piiroinen at gmail.com (Juhapekka Piiroinen) Date: Tue, 6 Mar 2012 13:54:41 +0200 Subject: [PySide] Fixed bug PYSIDE-2 - pkgconfig libdir variable set incorrectly in 64-bit environments In-Reply-To: <4F54F85F.4000906@nokia.com> References: <4F54F85F.4000906@nokia.com> Message-ID: Hi, I think that I managed to fix the issue PYSIDE-6. http://codereview.qt-project.org/#change,18937 Cheers, Juhapekka Piiroinen On Mon, Mar 5, 2012 at 7:31 PM, Matti Airas wrote: > On 05.03.2012 08:23, ext Juhapekka Piiroinen wrote: > > Hi, > > I have fixed the bug PYSIDE-2 "pkgconfig libdir variable set incorrectly > in 64-bit environments" which seemed a trivial one for me. > > Could somebody review the code: > http://codereview.qt-project.org/#change,18729 > > > This is great! > > Thanks a lot! > > Only PYSIDE-6 to go. (Took some time today to verify that it still is an > issue.) > > Cheers, > > ma. > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From matti.p.airas at nokia.com Tue Mar 6 15:14:49 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Tue, 06 Mar 2012 16:14:49 +0200 Subject: [PySide] [Announcement] PySide Becomes a Qt Add-On Message-ID: <4F561BD9.6000209@nokia.com> PySide [1] has now finished migration from its previous standalone setup to Qt Project [2] infrastructure. Being a Qt Add-on provides PySide a permanent home and perfect alignment with Qt Frameworks. Furthermore, the project gets improved visibility, as well as a simple, carefully thought out meritocratic project structure. In addition to the wiki that is already hosted by Qt, the PySide mailing list [3] and the bug tracker [4] are also now hosted by Qt. More information on the Qt Project can be found on the project web site [2]. The PySide project now follows Qt Project's governance model [5]. The Maintainer for API Extractor, Generatorrunner, and Shiboken is Marcelo Lira. The Maintainer for the PySide component is Hugo Parente Lima. Paulo Alcantara is an Approver for PySide. All other project roles are informal. Srini Kommoori has kindly volunteered to be the webmaster and wikimaster for the project. To developers using PySide the migration is mostly transparent. PySide is still available under the same licensing terms, and the project facilities are still the mostly unchanged. Instead of having a separate Bugzilla instance, the PySide project now utilizes Qt's Jira bug tracker [4]. Also the mailing list address has changed to pyside at qt-project.org [3]. Qt Project uses Gerrit [6] for code reviews. Developers contributing code to PySide should do it using Gerrit from now on. Read-only access to the git source code repositories is still provided via Gitorious [7]. [1] http://www.pyside.org [2] http://qt-project.org [3] http://lists.qt-project.org/mailman/listinfo/pyside [4] https://bugreports.qt-project.org [5] http://wiki.qt-project.org/The_Qt_Governance_Model [6] http://codereview.qt-project.org [7] http://qt.gitorious.org/pyside Best regards, Matti Airas From matti.p.airas at nokia.com Tue Mar 6 15:38:16 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Tue, 06 Mar 2012 16:38:16 +0200 Subject: [PySide] Fixed bug PYSIDE-2 - pkgconfig libdir variable set incorrectly in 64-bit environments In-Reply-To: References: <4F54F85F.4000906@nokia.com> Message-ID: <4F562158.40208@nokia.com> Cool! I owe you a beer. :-) Using this last regression bug as a trigger, I just spewed out announcements for the PySide project migration on a couple of blogs and a few mailing lists... Cheers, ma. On 06.03.2012 13:54, ext Juhapekka Piiroinen wrote: > Hi, > > I think that I managed to fix the issue PYSIDE-6. > http://codereview.qt-project.org/#change,18937 > > > Cheers, > Juhapekka Piiroinen > > On Mon, Mar 5, 2012 at 7:31 PM, Matti Airas > wrote: > > On 05.03.2012 08:23, ext Juhapekka Piiroinen wrote: >> Hi, >> >> I have fixed the bug PYSIDE-2 "pkgconfig libdir variable set >> incorrectly in 64-bit environments" which seemed a trivial one >> for me. >> >> Could somebody review the code: >> http://codereview.qt-project.org/#change,18729 > > This is great! > > Thanks a lot! > > Only PYSIDE-6 to go. (Took some time today to verify that it still > is an issue.) > > Cheers, > > ma. > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From techtonik at gmail.com Tue Mar 6 15:49:01 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Tue, 6 Mar 2012 16:49:01 +0200 Subject: [PySide] Fixed bug PYSIDE-2 - pkgconfig libdir variable set incorrectly in 64-bit environments In-Reply-To: <4F562158.40208@nokia.com> References: <4F54F85F.4000906@nokia.com> <4F562158.40208@nokia.com> Message-ID: Ice. =) The bug with closed Gerrit has been given critical priority and PyCon starts tomorrow. Good news. -- anatoly t. On Tue, Mar 6, 2012 at 5:38 PM, Matti Airas wrote: > Cool! > > I owe you a beer. :-) > > Using this last regression bug as a trigger, I just spewed out announcements > for the PySide project migration on a couple of blogs and a few mailing > lists... > > Cheers, > > ma. > > > On 06.03.2012 13:54, ext Juhapekka Piiroinen wrote: > > Hi, > > I think that I managed to fix the issue PYSIDE-6. > http://codereview.qt-project.org/#change,18937 > > > Cheers, > Juhapekka Piiroinen > > On Mon, Mar 5, 2012 at 7:31 PM, Matti Airas wrote: > > On 05.03.2012 08:23, ext Juhapekka Piiroinen wrote: > > Hi, > > I have fixed the bug PYSIDE-2 "pkgconfig libdir variable set incorrectly in > 64-bit environments" which seemed a trivial one for me. > > Could somebody review the code: > http://codereview.qt-project.org/#change,18729 > > > This is great! > > Thanks a lot! > > Only PYSIDE-6 to go. (Took some time today to verify that it still is an > issue.) > > Cheers, > > ma. > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > From techtonik at gmail.com Wed Mar 7 09:20:18 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 7 Mar 2012 11:20:18 +0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> Message-ID: Hello guys, Are there any prospects on possible release date? Spyder IDE still can not be started with PySide 1.1.0 on Linux http://bugs.pyside.org/show_bug.cgi?id=1105 And Spyder users are suffering from Windows-specific bug with PySide 1.0.9+ that I can not repeat on Linux: http://code.google.com/p/spyderlib/issues/detail?id=897 -- anatoly t. From hugo.lima at openbossa.org Wed Mar 7 18:39:20 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 7 Mar 2012 14:39:20 -0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> Message-ID: <201203071439.25691.hugo.lima@openbossa.org> On Wednesday 07 March 2012 05:20:18 anatoly techtonik wrote: > Hello guys, > > Are there any prospects on possible release date? We need to backport everything on github to the code on gerrit, when this finish we can finally release a new version. I'm occupied with other projects too, so at the moment I can't promise any dead line for a PySide release. > Spyder IDE still can not be started with PySide 1.1.0 on Linux > http://bugs.pyside.org/show_bug.cgi?id=1105 > > And Spyder users are suffering from Windows-specific bug with PySide > 1.0.9+ that I can not repeat on Linux: > http://code.google.com/p/spyderlib/issues/detail?id=897 > > -- > anatoly t. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From techtonik at gmail.com Thu Mar 8 11:08:35 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 8 Mar 2012 13:08:35 +0300 Subject: [PySide] About the next PySide release In-Reply-To: <201203071439.25691.hugo.lima@openbossa.org> References: <201202291458.25547.hugo.lima@openbossa.org> <201203071439.25691.hugo.lima@openbossa.org> Message-ID: On Wed, Mar 7, 2012 at 8:39 PM, Hugo Parente Lima wrote: > On Wednesday 07 March 2012 05:20:18 anatoly techtonik wrote: >> Hello guys, >> >> Are there any prospects on possible release date? > > We need to backport everything on github to the code on gerrit, when this > finish we can finally release a new version. 1. [ ] find the last common ancestor With HG I would do: hg clone git://gitorious.org/pyside/pyside.git cd pyside hg inc --template '{node}' -l 1 git://github.com/PySide/PySide.git 2. [ ] Make new clone and import revisions starting after the hash from the step above into Mercurial Queue cd .. hg clone git://github.com/PySide/PySide.git pyside2 cd pyside2 hg qimport -r HASH:HEAD hg qpop -a cp -r .hg/patches ../pyside/.hg/ cd ../pyside hg qpush -a hg qfinish -a Unfortunately, after Matti flattened the repository history this won't work anymore and we need to find this common revision manually. On the second thought even if the history was there, the hashes would be reindexed after history edition anyway, so we still have to inspect it manually. In any case, I think that a bad move to kill the project history. -- anatoly t. From ludo.visser at gmail.com Thu Mar 8 11:13:58 2012 From: ludo.visser at gmail.com (Ludo Visser) Date: Thu, 8 Mar 2012 11:13:58 +0100 Subject: [PySide] Building Pyside for Python3 on Mac OS X Message-ID: <4B3C5522-194A-40FB-BA3D-C7549BFA8AE6@gmail.com> Hi all, I'm trying to build PySide for Python 3.2 on Mac OS X using Macports. As of yet, there is no official portfile for PySide for Python 3, so I decided to modify the portfile for the 2.x version. As far as I could figure out, the only modification I had to make with respect to buiding for 2.7 is that Shiboken needs to be built with the USE_PYTHON3 flag set. Everything builds and installs fine, but I then run into the following problem: >>>import PySide.QtGui Traceback (most recent call last): File "", line 1, in ImportError: dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/3.2/lib/python3.2/site-packages/PySide/QtGui.so, 2): Symbol not found: _PyCObject_Type Referenced from: /opt/local/lib/libshiboken-python2.7.1.1.dylib Expected in: flat namespace in /opt/local/lib/libshiboken-python2.7.1.1.dylib While it makes sense that the libshiboken library for Python 2.7 is not going to work with Python 3.2, I can't figure out why this library is being used in the first place. After building Shiboken, I have this: $ ls -l /opt/local/lib/libshiboken* -rwxr-xr-x 1 root admin 212152 Jan 3 04:47 /opt/local/lib/libshiboken-python2.7.1.1.0.dylib lrwxr-xr-x 1 root admin 33 Jan 3 04:48 /opt/local/lib/libshiboken-python2.7.1.1.dylib -> libshiboken-python2.7.1.1.0.dylib lrwxr-xr-x 1 root admin 31 Jan 3 04:48 /opt/local/lib/libshiboken-python2.7.dylib -> libshiboken-python2.7.1.1.dylib -rwxr-xr-x 1 root admin 212152 Mar 3 13:25 /opt/local/lib/libshiboken.cpython-32m.1.1.0.dylib lrwxr-xr-x 1 root admin 35 Mar 3 13:25 /opt/local/lib/libshiboken.cpython-32m.1.1.dylib -> libshiboken.cpython-32m.1.1.0.dylib lrwxr-xr-x 1 root admin 33 Mar 3 13:25 /opt/local/lib/libshiboken.cpython-32m.dylib -> libshiboken.cpython-32m.1.1.dylib It seems I have the library file, but it is not used. Did I miss something while trying to build Shiboken, or something else? Did I miss a configuration step? Thanks, Ludo Visser From matti.p.airas at nokia.com Thu Mar 8 16:15:14 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 08 Mar 2012 17:15:14 +0200 Subject: [PySide] Bugzilla closed for new entries Message-ID: <4F58CD02.8080808@nokia.com> Hi, As discussed before, Qt Project's JIRA Bug Tracker should now be used for reporting new bugs against PySide. https://bugreports.qt-project.org/ I have now disabled opening new bugs in the old Bugzilla instance and requested all open bugs to be re-submitted in JIRA. Sorry for all inconvenience caused. :-) Cheers, ma. From techtonik at gmail.com Thu Mar 8 17:04:32 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 8 Mar 2012 19:04:32 +0300 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: <4F58CD02.8080808@nokia.com> References: <4F58CD02.8080808@nokia.com> Message-ID: On Thu, Mar 8, 2012 at 6:15 PM, Matti Airas wrote: > Hi, > > As discussed before, Qt Project's JIRA Bug Tracker should now be used > for reporting new bugs against PySide. > > https://bugreports.qt-project.org/ > > > I have now disabled opening new bugs in the old Bugzilla instance and > requested all open bugs to be re-submitted in JIRA. > > Sorry for all inconvenience caused. :-) There should be a possibility to provide a link to a newly recreated issue. I also afraid that most people will lack the incentive to recreate bugs anew. -- anatoly t. From hugo.lima at openbossa.org Thu Mar 8 17:33:44 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 8 Mar 2012 13:33:44 -0300 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: References: <4F58CD02.8080808@nokia.com> Message-ID: <201203081333.49700.hugo.lima@openbossa.org> On Thursday 08 March 2012 13:04:32 anatoly techtonik wrote: > On Thu, Mar 8, 2012 at 6:15 PM, Matti Airas wrote: > > Hi, > > > > As discussed before, Qt Project's JIRA Bug Tracker should now be used > > for reporting new bugs against PySide. > > > > https://bugreports.qt-project.org/ > > > > > > I have now disabled opening new bugs in the old Bugzilla instance and > > requested all open bugs to be re-submitted in JIRA. > > > > Sorry for all inconvenience caused. :-) > > There should be a possibility to provide a link to a newly recreated issue. > I also afraid that most people will lack the incentive to recreate bugs > anew. -- +1 > anatoly t. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matti.p.airas at nokia.com Thu Mar 8 18:17:52 2012 From: matti.p.airas at nokia.com (matti.p.airas at nokia.com) Date: Thu, 8 Mar 2012 17:17:52 +0000 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: References: <4F58CD02.8080808@nokia.com> Message-ID: <12EC0125-E459-42D6-8230-9230097C4DB5@nokia.com> On Mar 8, 2012, at 18:04, ext anatoly techtonik wrote: > On Thu, Mar 8, 2012 at 6:15 PM, Matti Airas wrote: >> Hi, >> >> As discussed before, Qt Project's JIRA Bug Tracker should now be used >> for reporting new bugs against PySide. >> >> https://bugreports.qt-project.org/ >> >> >> I have now disabled opening new bugs in the old Bugzilla instance and >> requested all open bugs to be re-submitted in JIRA. >> >> Sorry for all inconvenience caused. :-) > > There should be a possibility to provide a link to a newly recreated issue. Just comment on the bug in either bug tracker - it's not that hard! > I also afraid that most people will lack the incentive to recreate bugs anew. Thank you for volunteering to replicate all open bugs from Bugzilla to Jira! Please report back to the mailing list once you've done it! Cheers, ma. From techtonik at gmail.com Thu Mar 8 18:24:29 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 8 Mar 2012 20:24:29 +0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> <201203071439.25691.hugo.lima@openbossa.org> Message-ID: On Thu, Mar 8, 2012 at 1:08 PM, anatoly techtonik wrote: > On Wed, Mar 7, 2012 at 8:39 PM, Hugo Parente Lima > wrote: >> On Wednesday 07 March 2012 05:20:18 anatoly techtonik wrote: >>> Hello guys, >>> >>> Are there any prospects on possible release date? >> >> We need to backport everything on github to the code on gerrit, when this >> finish we can finally release a new version. > > 1. [ ] find the last common ancestor > With HG I would do: >  hg clone git://gitorious.org/pyside/pyside.git >  cd pyside >  hg inc --template '{node}' -l 1 git://github.com/PySide/PySide.git > > 2. [ ] Make new clone and import revisions starting after the hash > from the step above into Mercurial Queue >  cd .. >  hg clone git://github.com/PySide/PySide.git pyside2 >  cd pyside2 >  hg qimport -r HASH:HEAD >  hg qpop -a >  cp -r .hg/patches ../pyside/.hg/ >  cd ../pyside >  hg qpush -a >  hg qfinish -a > > > Unfortunately, after Matti flattened the repository history this won't > work anymore and we need to find this common revision manually. On the > second thought even if the history was there, the hashes would be > reindexed after history edition anyway, so we still have to inspect it > manually. In any case, I think that a bad move to kill the project > history. It appears that neither me, nor Matti knows how to properly strip those offending revisions from the Git history. Rebasing doesn't work, because subsequent merges screw up everything. So, unless somebody possess required know how, I am afraid we will have to stick with the way things currently are. -- anatoly t. From techtonik at gmail.com Thu Mar 8 18:28:54 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 8 Mar 2012 20:28:54 +0300 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: <12EC0125-E459-42D6-8230-9230097C4DB5@nokia.com> References: <4F58CD02.8080808@nokia.com> <12EC0125-E459-42D6-8230-9230097C4DB5@nokia.com> Message-ID: On Thu, Mar 8, 2012 at 8:17 PM, wrote: > On Mar 8, 2012, at 18:04, ext anatoly techtonik wrote: > >> On Thu, Mar 8, 2012 at 6:15 PM, Matti Airas wrote: >>> Hi, >>> >>> As discussed before, Qt Project's JIRA Bug Tracker should now be used >>> for reporting new bugs against PySide. >>> >>> https://bugreports.qt-project.org/ >>> >>> >>> I have now disabled opening new bugs in the old Bugzilla instance and >>> requested all open bugs to be re-submitted in JIRA. >>> >>> Sorry for all inconvenience caused. :-) >> >> There should be a possibility to provide a link to a newly recreated issue. > > Just comment on the bug in either bug tracker - it's not that hard! > >> I also afraid that most people will lack the incentive to recreate bugs anew. > > Thank you for volunteering to replicate all open bugs from Bugzilla to Jira! Please report back to the mailing list once you've done it! Well, somebody needs to execute this stuff - http://confluence.atlassian.com/display/JIRA/Importing+Data+from+Bugzilla I'll try to contact tracker list about this. -- anatoly t. From hugo.lima at openbossa.org Thu Mar 8 18:58:52 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 8 Mar 2012 14:58:52 -0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> <201203071439.25691.hugo.lima@openbossa.org> Message-ID: <201203081458.56754.hugo.lima@openbossa.org> On Thursday 08 March 2012 07:08:35 anatoly techtonik wrote: > On Wed, Mar 7, 2012 at 8:39 PM, Hugo Parente Lima > > wrote: > > On Wednesday 07 March 2012 05:20:18 anatoly techtonik wrote: > >> Hello guys, > >> > >> Are there any prospects on possible release date? > > > > We need to backport everything on github to the code on gerrit, when this > > finish we can finally release a new version. > > 1. [ ] find the last common ancestor > With HG I would do: > hg clone git://gitorious.org/pyside/pyside.git > cd pyside > hg inc --template '{node}' -l 1 git://github.com/PySide/PySide.git > > 2. [ ] Make new clone and import revisions starting after the hash > from the step above into Mercurial Queue > cd .. > hg clone git://github.com/PySide/PySide.git pyside2 > cd pyside2 > hg qimport -r HASH:HEAD > hg qpop -a > cp -r .hg/patches ../pyside/.hg/ > cd ../pyside > hg qpush -a > hg qfinish -a We can use "git format-patch" then "git am" to apply the patches and hope for no conflicts. > Unfortunately, after Matti flattened the repository history this won't > work anymore and we need to find this common revision manually. On the > second thought even if the history was there, the hashes would be > reindexed after history edition anyway, so we still have to inspect it > manually. In any case, I think that a bad move to kill the project > history. I think the flattened history was maybe the only point that make me feel against the move to qt-projects, the history really helps when developing, it's the documentation of the project, when you don't know what a piece of code does you just look at the history of those lines and see the commit comments, reference for bugs in bugzilla, who wrote the changes, etc... and now all this was gone :-/. Matti told me about a magic script that would do that, I didn't see it and a doubt that it would be on pair with the own git commands. I was talking with Renato days ago about it and he have the same feelings, besides looking at git history with the lack of commits last weeks I see that wasn't only me with this feeling. IMO Gerrit isn't the best tool, isn't too user friendly, requires a password for everything, i.e. it have their problems, but nothing that we can't get used to (like we were used to bugzilla), unfortunately due to Nokia that owns the name PySide (Matti, correct if I'm wrong) we can't just stick on git hub, using their own bug tracking system and maybe even web server that github offers, this would be perfect from the developer point of view (at least my point of view), no moves, no changes the project would go on without headaches... but we (pyside developers) can't do that, because doing that will start the bizarre situation where all original developers forked the project and no developers left on the official project. I *really* don't want this to happen. I still have the plan to do a rebase removing the offending commits from each repository, solve the conflicts by hand then ask for a approval to push the changes to gerrit send to Matti for approval, after that I'll finally feel comfortable to continue to hack on PySide, but this takes times and I have not much free time to code, so would be better to send my time fixing bugs rather doing this code-bureaucracy stuff :-(. Commits are the pulse of any project, developers are the heart, no developers means no commits that means no pulse that means a dead project. > -- > anatoly t. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Thu Mar 8 20:27:01 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 8 Mar 2012 16:27:01 -0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> Message-ID: <201203081627.06135.hugo.lima@openbossa.org> On Thursday 08 March 2012 14:24:29 anatoly techtonik wrote: > On Thu, Mar 8, 2012 at 1:08 PM, anatoly techtonik wrote: > > On Wed, Mar 7, 2012 at 8:39 PM, Hugo Parente Lima > > > > wrote: > >> On Wednesday 07 March 2012 05:20:18 anatoly techtonik wrote: > >>> Hello guys, > >>> > >>> Are there any prospects on possible release date? > >> > >> We need to backport everything on github to the code on gerrit, when > >> this finish we can finally release a new version. > > > > 1. [ ] find the last common ancestor > > With HG I would do: > > hg clone git://gitorious.org/pyside/pyside.git > > cd pyside > > hg inc --template '{node}' -l 1 git://github.com/PySide/PySide.git > > > > 2. [ ] Make new clone and import revisions starting after the hash > > from the step above into Mercurial Queue > > cd .. > > hg clone git://github.com/PySide/PySide.git pyside2 > > cd pyside2 > > hg qimport -r HASH:HEAD > > hg qpop -a > > cp -r .hg/patches ../pyside/.hg/ > > cd ../pyside > > hg qpush -a > > hg qfinish -a > > > > > > Unfortunately, after Matti flattened the repository history this won't > > work anymore and we need to find this common revision manually. On the > > second thought even if the history was there, the hashes would be > > reindexed after history edition anyway, so we still have to inspect it > > manually. In any case, I think that a bad move to kill the project > > history. > > It appears that neither me, nor Matti knows how to properly strip > those offending revisions from the Git history. Rebasing doesn't work, > because subsequent merges screw up everything. So, unless somebody > possess required know how, I am afraid we will have to stick with the > way things currently are. > -- > anatoly t. Ok, first step done! I rebased over 632 commits to remove the offending commit on Shiboken and pushed the changes to a personal branch on github. The offending commit was: c3059779d7628fdbb140ed02cdc0cc7ca80e7ad8 Matti can you check if everything is ok? then I can re-apply the tags (because the hashes changed) and apply the fixes on gerrit on top of this branch and finally force-push it to gerrit with all history glory built-in! :-D The rebased tree without the offending commit can be found at branch "clafix" on https://github.com/hugopl/Shiboken I didn't the same on ApiExtractor because I can't figure out what was the offending commit hash. If everyone agree with this I can do the same on PySide and ApiExtractor too. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Thu Mar 8 21:04:24 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 8 Mar 2012 17:04:24 -0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> Message-ID: <201203081704.29832.hugo.lima@openbossa.org> On Thursday 08 March 2012 14:24:29 anatoly techtonik wrote: > On Thu, Mar 8, 2012 at 1:08 PM, anatoly techtonik wrote: > > On Wed, Mar 7, 2012 at 8:39 PM, Hugo Parente Lima > > > > wrote: > >> On Wednesday 07 March 2012 05:20:18 anatoly techtonik wrote: > >>> Hello guys, > >>> > >>> Are there any prospects on possible release date? > >> > >> We need to backport everything on github to the code on gerrit, when > >> this finish we can finally release a new version. > > > > 1. [ ] find the last common ancestor > > With HG I would do: > > hg clone git://gitorious.org/pyside/pyside.git > > cd pyside > > hg inc --template '{node}' -l 1 git://github.com/PySide/PySide.git > > > > 2. [ ] Make new clone and import revisions starting after the hash > > from the step above into Mercurial Queue > > cd .. > > hg clone git://github.com/PySide/PySide.git pyside2 > > cd pyside2 > > hg qimport -r HASH:HEAD > > hg qpop -a > > cp -r .hg/patches ../pyside/.hg/ > > cd ../pyside > > hg qpush -a > > hg qfinish -a > > > > > > Unfortunately, after Matti flattened the repository history this won't > > work anymore and we need to find this common revision manually. On the > > second thought even if the history was there, the hashes would be > > reindexed after history edition anyway, so we still have to inspect it > > manually. In any case, I think that a bad move to kill the project > > history. > > It appears that neither me, nor Matti knows how to properly strip > those offending revisions from the Git history. Rebasing doesn't work, > because subsequent merges screw up everything. So, unless somebody > possess required know how, I am afraid we will have to stick with the > way things currently are. > -- > anatoly t. Second step done! I rebased PySide removing the two offending commits and pushed the tree to the "clafix" branch of my PySide clone on github (https://github.com/hugopl/PySide) The offending commits were: - 0d60daf1e1a6edd95b58c7bdd33da8f4c1629c90 - 32681f7b7039f41bec4c4965370d4b66bee53532 By what I saw GeneratorRunner doesn't had any problems with missing CLA agreement, then the only missing repository is ApiExtractor. Matti, do you still have the git hash annotated somewhere? -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matti.p.airas at nokia.com Fri Mar 9 09:56:33 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Fri, 09 Mar 2012 10:56:33 +0200 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: References: <4F58CD02.8080808@nokia.com> <12EC0125-E459-42D6-8230-9230097C4DB5@nokia.com> Message-ID: <4F59C5C1.7020201@nokia.com> On 08.03.2012 19:28, ext anatoly techtonik wrote: > Well, somebody needs to execute this stuff - > http://confluence.atlassian.com/display/JIRA/Importing+Data+from+Bugzilla > > I'll try to contact tracker list about this. Neat! Thanks. ma. From matti.p.airas at nokia.com Fri Mar 9 13:40:30 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Fri, 09 Mar 2012 14:40:30 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <201203081627.06135.hugo.lima@openbossa.org> References: <201202291458.25547.hugo.lima@openbossa.org> <201203081627.06135.hugo.lima@openbossa.org> Message-ID: <4F59FA3E.8030804@nokia.com> On 08.03.2012 21:27, ext Hugo Parente Lima wrote: > Ok, first step done! > > I rebased over 632 commits to remove the offending commit on Shiboken and > pushed the changes to a personal branch on github. > > The offending commit was: c3059779d7628fdbb140ed02cdc0cc7ca80e7ad8 > > Matti can you check if everything is ok? then I can re-apply the tags (because > the hashes changed) and apply the fixes on gerrit on top of this branch and > finally force-push it to gerrit with all history glory built-in! :-D > > The rebased tree without the offending commit can be found at branch "clafix" > on https://github.com/hugopl/Shiboken > > I didn't the same on ApiExtractor because I can't figure out what was the > offending commit hash. > > If everyone agree with this I can do the same on PySide and ApiExtractor too. Wow, that was a heroic effort! Big kudos to you! To me, both both Shiboken and PySide clafix branches look just fine - go ahead with it! The last problematic commit on Apiextractor is: c2a35449016e3d5873f3e1657df51e2ff854cd2e And yes, GeneratorRunner is good as-is. Since the submitted changes in Gerrit are commits in a special ref, just make sure that when force-pushing the history back to Gerrit, those commits are not lost (I'm not sure how git works wrt force-pushing...). BTW, how did you actually do the removal? Did you just branch the repos before the offending commits and then merge all commits except the offending ones? Cheers, ma. From anderson.lizardo at openbossa.org Fri Mar 9 14:12:39 2012 From: anderson.lizardo at openbossa.org (Anderson Lizardo) Date: Fri, 9 Mar 2012 09:12:39 -0400 Subject: [PySide] About the next PySide release In-Reply-To: <4F59FA3E.8030804@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <201203081627.06135.hugo.lima@openbossa.org> <4F59FA3E.8030804@nokia.com> Message-ID: On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas wrote: > BTW, how did you actually do the removal? Did you just branch the repos > before the offending commits and then merge all commits except the > offending ones? I wonder if a " git rebase -i commit_id^ " (where "commit_id" is the commit to be removed) and removing the line on the interactive rebase editor, would fix this more or less easily :) Anyway, it's done and let's hope development can progress :) Regards, -- Anderson Lizardo Instituto Nokia de Tecnologia - INdT Manaus - Brazil From matti.p.airas at nokia.com Fri Mar 9 14:14:25 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Fri, 09 Mar 2012 15:14:25 +0200 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> <201203081627.06135.hugo.lima@openbossa.org> <4F59FA3E.8030804@nokia.com> Message-ID: <4F5A0231.6000606@nokia.com> On 09.03.2012 15:12, ext Anderson Lizardo wrote: > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas wrote: >> BTW, how did you actually do the removal? Did you just branch the repos >> before the offending commits and then merge all commits except the >> offending ones? > I wonder if a " git rebase -i commit_id^ " (where "commit_id" is the > commit to be removed) and removing the line on the interactive rebase > editor, would fix this more or less easily :) That was the approach I tried, but that breaks down in merges. Couldn't quite figure out what the exact problem was, though. Cheers, ma. From hugo.lima at openbossa.org Fri Mar 9 18:19:41 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Fri, 9 Mar 2012 14:19:41 -0300 Subject: [PySide] About the next PySide release In-Reply-To: <4F5A0231.6000606@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <4F5A0231.6000606@nokia.com> Message-ID: <201203091419.46192.hugo.lima@openbossa.org> On Friday 09 March 2012 10:14:25 Matti Airas wrote: > On 09.03.2012 15:12, ext Anderson Lizardo wrote: > > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas wrote: > >> BTW, how did you actually do the removal? Did you just branch the repos > >> before the offending commits and then merge all commits except the > >> offending ones? > > > > I wonder if a " git rebase -i commit_id^ " (where "commit_id" is the > > commit to be removed) and removing the line on the interactive rebase > > editor, would fix this more or less easily :) > > That was the approach I tried, but that breaks down in merges. Couldn't > quite figure out what the exact problem was, though. I did git rebase -i --preserve-merges commit_id~1 Removed the commit and went on fixing the conflicts taking care to not recreate the fix done by the removed commit by mistake. I guess just one merge of shiboken had problems and went "inlined", the other were preserved as well. > Cheers, > > ma. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Fri Mar 9 23:41:06 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Fri, 9 Mar 2012 19:41:06 -0300 Subject: [PySide] About the next PySide release In-Reply-To: <4F5A0231.6000606@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <4F5A0231.6000606@nokia.com> Message-ID: <201203091941.11140.hugo.lima@openbossa.org> On Friday 09 March 2012 10:14:25 Matti Airas wrote: > On 09.03.2012 15:12, ext Anderson Lizardo wrote: > > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas wrote: > >> BTW, how did you actually do the removal? Did you just branch the repos > >> before the offending commits and then merge all commits except the > >> offending ones? > > > > I wonder if a " git rebase -i commit_id^ " (where "commit_id" is the > > commit to be removed) and removing the line on the interactive rebase > > editor, would fix this more or less easily :) > > That was the approach I tried, but that breaks down in merges. Couldn't > quite figure out what the exact problem was, though. > > Cheers, > > ma. Work done... .. but gerrit, oh gerrit... I can't push the changes because I don't have permissions to push merge commits: Counting objects: 2086, done. Delta compression using up to 2 threads. Compressing objects: 100% (935/935), done. Writing objects: 100% (2086/2086), 715.24 KiB, done. Total 2086 (delta 1459), reused 1476 (delta 1135) remote: Resolving deltas: 100% (1459/1459) To ssh://hugopl at codereview.qt-project.org:29418/pyside/apiextractor.git ! [remote rejected] master -> master (you are not allowed to upload merges) error: failed to push some refs to 'ssh://hugopl at codereview.qt- project.org:29418/pyside/apiextractor.git' -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From techtonik at gmail.com Sat Mar 10 08:19:24 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 10 Mar 2012 10:19:24 +0300 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: <4F59C5C1.7020201@nokia.com> References: <4F58CD02.8080808@nokia.com> <12EC0125-E459-42D6-8230-9230097C4DB5@nokia.com> <4F59C5C1.7020201@nokia.com> Message-ID: On Fri, Mar 9, 2012 at 11:56 AM, Matti Airas wrote: > On 08.03.2012 19:28, ext anatoly techtonik wrote: >> >> Well, somebody needs to execute this stuff - >> http://confluence.atlassian.com/display/JIRA/Importing+Data+from+Bugzilla >> >> I'll try to contact tracker list about this. > > > Neat! Thanks. Done. The letter should appear at http://lists.qt-project.org/pipermail/jeg/ in the March section. -- anatoly t. From blair at orcaware.com Sat Mar 10 09:23:31 2012 From: blair at orcaware.com (Blair Zajac) Date: Sat, 10 Mar 2012 00:23:31 -0800 Subject: [PySide] 1.1.0 core dump with short-circuit signal name Message-ID: <4F5B0F83.3020102@orcaware.com> Hello, I'm new to PySide develop so wanted to open this on the mailing list before opening a ticket, being uncertain about bug policy on this project. The below script core dumps using PySide 1.1.0 on Ubuntu 12.04 Beta 1. Can anyone else confirm this on their system. The same code doesn't core dump on Ubuntu 11.10 with PySide 1.0.6. If it is reproducible, then it seems that it shouldn't core dump, even if the code isn't valid PySide code. Below is a stack trace and the code. Thanks, Blair PS We're moving a large project from PyQt3 to PySide, so that's how I ran into this. Core was generated by `python t3.py'. Program terminated with signal 11, Segmentation fault. #0 0x000000000041aec9 in PyObject_Call () (gdb) bt #0 0x000000000041aec9 in PyObject_Call () #1 0x00000000004b5d96 in PyEval_CallObjectWithKeywords () #2 0x00007f32da0052da in PySide::SignalManager::callPythonMetaMethod(QMetaMethod const&, void**, _object*, bool) () from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 #3 0x00007f32da00575c in PySide::SignalManager::qt_metacall(QObject*, QMetaObject::Call, int, void**) () from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 #4 0x00007f32d958aad9 in QMetaObject::activate(QObject*, QMetaObject const*, int, void**) () from /usr/lib/x86_64-linux-gnu/libQtCore.so.4 #5 0x00007f32da005990 in PySide::SignalManager::qt_metacall(QObject*, QMetaObject::Call, int, void**) () from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 #6 0x00007f32da005fd7 in PySide::SignalManager::emitSignal(QObject*, char const*, _object*) () from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 #7 0x00007f32da32d938 in ?? () from /usr/lib/python2.7/dist-packages/PySide/QtCore.so #8 0x00000000004b65c1 in PyEval_EvalFrameEx () #9 0x00000000004b6b97 in PyEval_EvalFrameEx () #10 0x00000000004bcd0d in PyEval_EvalCodeEx () #11 0x00000000004bd7e2 in PyEval_EvalCode () #12 0x00000000004dcc52 in run_mod () #13 0x00000000004dd814 in PyRun_FileExFlags () #14 0x00000000004de28e in PyRun_SimpleFileExFlags () #15 0x00000000004ee6dd in Py_Main () #16 0x00007f32da9e076d in __libc_start_main () from /lib/x86_64-linux-gnu/libc.so.6 #17 0x000000000041ce39 in _start () #!/usr/bin/python from PySide import QtCore, QtGui SIGNAL_NAME = 'testSignal' class EmitterClass(QtCore.QObject): def __init__(self): super(EmitterClass, self).__init__() def emitter(self): self.emit(QtCore.SIGNAL(SIGNAL_NAME), 100) class ListeningClass(QtCore.QObject): def __init__(self): super(ListeningClass, self).__init__() self.emitter = EmitterClass() QtCore.QObject.connect(self.emitter, QtCore.SIGNAL(SIGNAL_NAME), self.output) def output(self, number): print number if __name__ == '__main__': c = ListeningClass() c.emitter.emitter() From prokher at gmail.com Sat Mar 10 13:08:49 2012 From: prokher at gmail.com (Alexander A. Prokhorov) Date: Sat, 10 Mar 2012 16:08:49 +0400 Subject: [PySide] Status of PySide with Qt4.8. Message-ID: <4F5B4451.8070304@gmail.com> Dear colleagues, is there any info about Qt4.8 support? When will PySide support it? From techtonik at gmail.com Sat Mar 10 13:17:16 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 10 Mar 2012 15:17:16 +0300 Subject: [PySide] Status of PySide with Qt4.8. In-Reply-To: <4F5B4451.8070304@gmail.com> References: <4F5B4451.8070304@gmail.com> Message-ID: On Sat, Mar 10, 2012 at 3:08 PM, Alexander A. Prokhorov wrote: > Dear colleagues, is there any info about Qt4.8 support? > When will PySide support it? What makes you think so? Have you tried to build for Qt 4.8? -- anatoly t. From prokher at gmail.com Sat Mar 10 13:34:11 2012 From: prokher at gmail.com (Alexander A. Prokhorov) Date: Sat, 10 Mar 2012 16:34:11 +0400 Subject: [PySide] Status of PySide with Qt4.8. In-Reply-To: References: <4F5B4451.8070304@gmail.com> Message-ID: <4F5B4A43.6020100@gmail.com> Is it worth trying? I've seen unresolved bug report few weeks ago, but I've lost track during bug tracker migration. Actually, I just curious is there any info on that. If you think it works fine I'll try to build it. On 03/10/2012 04:17 PM, anatoly techtonik wrote: > On Sat, Mar 10, 2012 at 3:08 PM, Alexander A. Prokhorov > wrote: >> Dear colleagues, is there any info about Qt4.8 support? >> When will PySide support it? > What makes you think so? > Have you tried to build for Qt 4.8? > -- > anatoly t. From techtonik at gmail.com Sat Mar 10 13:43:09 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 10 Mar 2012 15:43:09 +0300 Subject: [PySide] About the next PySide release In-Reply-To: <201203091941.11140.hugo.lima@openbossa.org> References: <201202291458.25547.hugo.lima@openbossa.org> <4F5A0231.6000606@nokia.com> <201203091941.11140.hugo.lima@openbossa.org> Message-ID: On Sat, Mar 10, 2012 at 1:41 AM, Hugo Parente Lima wrote: > On Friday 09 March 2012 10:14:25 Matti Airas wrote: >> On 09.03.2012 15:12, ext Anderson Lizardo wrote: >> > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas > wrote: >> >> BTW, how did you actually do the removal? Did you just branch the repos >> >> before the offending commits and then merge all commits except the >> >> offending ones? >> > >> > I wonder if a  " git rebase -i commit_id^ " (where "commit_id" is the >> > commit to be removed) and removing the line on the interactive rebase >> > editor, would fix this more or less easily :) >> >> That was the approach I tried, but that breaks down in merges. Couldn't >> quite figure out what the exact problem was, though. >> >> Cheers, >> >> ma. > > Work done... Just when I thought I found something.. =) http://linux.die.net/man/1/git-filter-branch Take a look at `git filter-branch --commit-filter`. Is it better than rebase? -- anatoly t. From techtonik at gmail.com Sat Mar 10 14:07:39 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 10 Mar 2012 16:07:39 +0300 Subject: [PySide] Status of PySide with Qt4.8. In-Reply-To: <4F5B4A43.6020100@gmail.com> References: <4F5B4451.8070304@gmail.com> <4F5B4A43.6020100@gmail.com> Message-ID: On Sat, Mar 10, 2012 at 3:34 PM, Alexander A. Prokhorov wrote: > Is it worth trying? Of course! You can spend days waiting for reply just because everybody else (i.e. me) is satisfied with 4.7. =) > I've seen unresolved bug report few weeks ago, but I've > lost track during bug tracker migration. Yes, bug tracking is painful at the moment. Hopefully, all the bugs will be migrated soon - my request for that is still in moderation queue for the Jira mailing list, so no idea when it happens. But Jira is fully operational, and you may fill any issues found on the way - even if they are marked duplicates you won't lose the track. > Actually, I just curious is there any info on that. If you think it works > fine I'll try to build it. I don't know. But if you want it to work, trying and reporting about the progress would definitely clarify the situation. And if you need know how - these should help you - https://github.com/PySide/BuildScripts -- anatoly t. From renatox at gmail.com Sat Mar 10 14:45:37 2012 From: renatox at gmail.com (Renato Araujo) Date: Sat, 10 Mar 2012 10:45:37 -0300 Subject: [PySide] 1.1.0 core dump with short-circuit signal name In-Reply-To: <4F5B0F83.3020102@orcaware.com> References: <4F5B0F83.3020102@orcaware.com> Message-ID: Hi Blair, I recommend to you use the new singal/slot API[1], since this is a bug on the deprecated API this have low priority to fix. [1]http://qt-project.org/wiki/Signals_and_Slots_in_PySide Renato Araujo Oliveira Filho On Sat, Mar 10, 2012 at 5:23 AM, Blair Zajac wrote: > Hello, > > I'm new to PySide develop so wanted to open this on the mailing list > before opening a ticket, being uncertain about bug policy on this project. > > The below script core dumps using PySide 1.1.0 on Ubuntu 12.04 Beta 1. > Can anyone else confirm this on their system.  The same code doesn't > core dump on Ubuntu 11.10 with PySide 1.0.6. > > If it is reproducible, then it seems that it shouldn't core dump, even > if the code isn't valid PySide code. > > Below is a stack trace and the code. > > Thanks, > Blair > > PS We're moving a large project from PyQt3 to PySide, so that's how I > ran into this. > > > Core was generated by `python t3.py'. > Program terminated with signal 11, Segmentation fault. > #0  0x000000000041aec9 in PyObject_Call () > (gdb) bt > #0  0x000000000041aec9 in PyObject_Call () > #1  0x00000000004b5d96 in PyEval_CallObjectWithKeywords () > #2  0x00007f32da0052da in > PySide::SignalManager::callPythonMetaMethod(QMetaMethod const&, void**, > _object*, bool) () >    from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 > #3  0x00007f32da00575c in PySide::SignalManager::qt_metacall(QObject*, > QMetaObject::Call, int, void**) () >    from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 > #4  0x00007f32d958aad9 in QMetaObject::activate(QObject*, QMetaObject > const*, int, void**) () from /usr/lib/x86_64-linux-gnu/libQtCore.so.4 > #5  0x00007f32da005990 in PySide::SignalManager::qt_metacall(QObject*, > QMetaObject::Call, int, void**) () >    from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 > #6  0x00007f32da005fd7 in PySide::SignalManager::emitSignal(QObject*, > char const*, _object*) () from > /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 > #7  0x00007f32da32d938 in ?? () >    from /usr/lib/python2.7/dist-packages/PySide/QtCore.so > #8  0x00000000004b65c1 in PyEval_EvalFrameEx () > #9  0x00000000004b6b97 in PyEval_EvalFrameEx () > #10 0x00000000004bcd0d in PyEval_EvalCodeEx () > #11 0x00000000004bd7e2 in PyEval_EvalCode () > #12 0x00000000004dcc52 in run_mod () > #13 0x00000000004dd814 in PyRun_FileExFlags () > #14 0x00000000004de28e in PyRun_SimpleFileExFlags () > #15 0x00000000004ee6dd in Py_Main () > #16 0x00007f32da9e076d in __libc_start_main () >    from /lib/x86_64-linux-gnu/libc.so.6 > #17 0x000000000041ce39 in _start () > > > > #!/usr/bin/python > > from PySide import QtCore, QtGui > > SIGNAL_NAME = 'testSignal' > > class EmitterClass(QtCore.QObject): >     def __init__(self): >         super(EmitterClass, self).__init__() > >     def emitter(self): >         self.emit(QtCore.SIGNAL(SIGNAL_NAME), 100) > > class ListeningClass(QtCore.QObject): >     def __init__(self): >         super(ListeningClass, self).__init__() >         self.emitter = EmitterClass() >         QtCore.QObject.connect(self.emitter, QtCore.SIGNAL(SIGNAL_NAME), >                                self.output) > >     def output(self, number): >         print number > > if __name__ == '__main__': >     c = ListeningClass() >     c.emitter.emitter() > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From blair at orcaware.com Sat Mar 10 17:09:20 2012 From: blair at orcaware.com (Blair Zajac) Date: Sat, 10 Mar 2012 08:09:20 -0800 Subject: [PySide] 1.1.0 core dump with short-circuit signal name In-Reply-To: References: <4F5B0F83.3020102@orcaware.com> Message-ID: <4F5B7CB0.1040501@orcaware.com> Hi Renato, Thanks, however, we're porting an 80k LOC project from PyQt3, so we'll be using the old signal/slot mechanism for now just to get the app running and then we may cut over to the new signal/slots. I'll open a bug report to track this. Blair On 03/10/2012 05:45 AM, Renato Araujo wrote: > Hi Blair, > > I recommend to you use the new singal/slot API[1], since this is a bug > on the deprecated API this have low priority to fix. > > [1]http://qt-project.org/wiki/Signals_and_Slots_in_PySide > > > Renato Araujo Oliveira Filho > > > > On Sat, Mar 10, 2012 at 5:23 AM, Blair Zajac wrote: >> Hello, >> >> I'm new to PySide develop so wanted to open this on the mailing list >> before opening a ticket, being uncertain about bug policy on this project. >> >> The below script core dumps using PySide 1.1.0 on Ubuntu 12.04 Beta 1. >> Can anyone else confirm this on their system. The same code doesn't >> core dump on Ubuntu 11.10 with PySide 1.0.6. >> >> If it is reproducible, then it seems that it shouldn't core dump, even >> if the code isn't valid PySide code. >> >> Below is a stack trace and the code. >> >> Thanks, >> Blair >> >> PS We're moving a large project from PyQt3 to PySide, so that's how I >> ran into this. >> >> >> Core was generated by `python t3.py'. >> Program terminated with signal 11, Segmentation fault. >> #0 0x000000000041aec9 in PyObject_Call () >> (gdb) bt >> #0 0x000000000041aec9 in PyObject_Call () >> #1 0x00000000004b5d96 in PyEval_CallObjectWithKeywords () >> #2 0x00007f32da0052da in >> PySide::SignalManager::callPythonMetaMethod(QMetaMethod const&, void**, >> _object*, bool) () >> from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >> #3 0x00007f32da00575c in PySide::SignalManager::qt_metacall(QObject*, >> QMetaObject::Call, int, void**) () >> from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >> #4 0x00007f32d958aad9 in QMetaObject::activate(QObject*, QMetaObject >> const*, int, void**) () from /usr/lib/x86_64-linux-gnu/libQtCore.so.4 >> #5 0x00007f32da005990 in PySide::SignalManager::qt_metacall(QObject*, >> QMetaObject::Call, int, void**) () >> from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >> #6 0x00007f32da005fd7 in PySide::SignalManager::emitSignal(QObject*, >> char const*, _object*) () from >> /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >> #7 0x00007f32da32d938 in ?? () >> from /usr/lib/python2.7/dist-packages/PySide/QtCore.so >> #8 0x00000000004b65c1 in PyEval_EvalFrameEx () >> #9 0x00000000004b6b97 in PyEval_EvalFrameEx () >> #10 0x00000000004bcd0d in PyEval_EvalCodeEx () >> #11 0x00000000004bd7e2 in PyEval_EvalCode () >> #12 0x00000000004dcc52 in run_mod () >> #13 0x00000000004dd814 in PyRun_FileExFlags () >> #14 0x00000000004de28e in PyRun_SimpleFileExFlags () >> #15 0x00000000004ee6dd in Py_Main () >> #16 0x00007f32da9e076d in __libc_start_main () >> from /lib/x86_64-linux-gnu/libc.so.6 >> #17 0x000000000041ce39 in _start () >> >> >> >> #!/usr/bin/python >> >> from PySide import QtCore, QtGui >> >> SIGNAL_NAME = 'testSignal' >> >> class EmitterClass(QtCore.QObject): >> def __init__(self): >> super(EmitterClass, self).__init__() >> >> def emitter(self): >> self.emit(QtCore.SIGNAL(SIGNAL_NAME), 100) >> >> class ListeningClass(QtCore.QObject): >> def __init__(self): >> super(ListeningClass, self).__init__() >> self.emitter = EmitterClass() >> QtCore.QObject.connect(self.emitter, QtCore.SIGNAL(SIGNAL_NAME), >> self.output) >> >> def output(self, number): >> print number >> >> if __name__ == '__main__': >> c = ListeningClass() >> c.emitter.emitter() >> _______________________________________________ >> PySide mailing list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside > From th.perl at gmail.com Sat Mar 10 19:02:13 2012 From: th.perl at gmail.com (Thomas Perl) Date: Sat, 10 Mar 2012 19:02:13 +0100 Subject: [PySide] Status of PySide with Qt4.8. In-Reply-To: <4F5B4A43.6020100@gmail.com> References: <4F5B4451.8070304@gmail.com> <4F5B4A43.6020100@gmail.com> Message-ID: Hi, 2012/3/10 Alexander A. Prokhorov : > Is it worth trying? I've seen unresolved bug report few weeks ago, but > I've lost track during bug tracker migration. > Actually, I just curious is there any info on that. If you think it > works fine I'll try to build it. PySide (at least when built from Git) supports Qt 4.8 already (I've build it using the PySide Buildscripts on OS X a few weeks ago, if the current Git head doesn't build, it's just a regression): Python 2.7.1 (r271:86832, Jul 31 2011, 19:30:53) [GCC 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2335.15.00)] on darwin Type "help", "copyright", "credits" or "license" for more information. >>> import PySide >>> PySide.__version__ '1.1.1' >>> import PySide.QtCore >>> PySide.QtCore.__version__ '4.8.0' HTH :) Thomas > On 03/10/2012 04:17 PM, anatoly techtonik wrote: >> On Sat, Mar 10, 2012 at 3:08 PM, Alexander A. Prokhorov >>  wrote: >>> Dear colleagues, is there any info about Qt4.8 support? >>> When will PySide support it? >> What makes you think so? >> Have you tried to build for Qt 4.8? >> -- >> anatoly t. > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From paulo.alcantara at openbossa.org Sun Mar 11 00:43:43 2012 From: paulo.alcantara at openbossa.org (paulo alcantara) Date: Sat, 10 Mar 2012 20:43:43 -0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> <4F5A0231.6000606@nokia.com> <201203091941.11140.hugo.lima@openbossa.org> Message-ID: Maybe its better or not. It doesnt matter either. The execellent work that Hugo has been doing (even with no much spare time to do it) is already done. On 3/10/12, anatoly techtonik wrote: > On Sat, Mar 10, 2012 at 1:41 AM, Hugo Parente Lima > wrote: >> On Friday 09 March 2012 10:14:25 Matti Airas wrote: >>> On 09.03.2012 15:12, ext Anderson Lizardo wrote: >>> > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas >> wrote: >>> >> BTW, how did you actually do the removal? Did you just branch the >>> >> repos >>> >> before the offending commits and then merge all commits except the >>> >> offending ones? >>> > >>> > I wonder if a  " git rebase -i commit_id^ " (where "commit_id" is the >>> > commit to be removed) and removing the line on the interactive rebase >>> > editor, would fix this more or less easily :) >>> >>> That was the approach I tried, but that breaks down in merges. Couldn't >>> quite figure out what the exact problem was, though. >>> >>> Cheers, >>> >>> ma. >> >> Work done... > > Just when I thought I found something.. =) > http://linux.die.net/man/1/git-filter-branch > > Take a look at `git filter-branch --commit-filter`. Is it better than > rebase? > > -- > anatoly t. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > From techtonik at gmail.com Mon Mar 12 10:02:49 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Mon, 12 Mar 2012 12:02:49 +0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> <4F5A0231.6000606@nokia.com> <201203091941.11140.hugo.lima@openbossa.org> Message-ID: I can't confirm that - there are no changes in the Gitorious repository. http://qt.gitorious.org/pyside/pyside/commits/master -- anatoly t. On Sun, Mar 11, 2012 at 2:43 AM, paulo alcantara wrote: > Maybe its better or not. It doesnt matter either. > The execellent work that Hugo has been doing (even with no much spare > time to do it) is already done. > > On 3/10/12, anatoly techtonik wrote: >> On Sat, Mar 10, 2012 at 1:41 AM, Hugo Parente Lima >> wrote: >>> On Friday 09 March 2012 10:14:25 Matti Airas wrote: >>>> On 09.03.2012 15:12, ext Anderson Lizardo wrote: >>>> > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas >>> wrote: >>>> >> BTW, how did you actually do the removal? Did you just branch the >>>> >> repos >>>> >> before the offending commits and then merge all commits except the >>>> >> offending ones? >>>> > >>>> > I wonder if a  " git rebase -i commit_id^ " (where "commit_id" is the >>>> > commit to be removed) and removing the line on the interactive rebase >>>> > editor, would fix this more or less easily :) >>>> >>>> That was the approach I tried, but that breaks down in merges. Couldn't >>>> quite figure out what the exact problem was, though. >>>> >>>> Cheers, >>>> >>>> ma. >>> >>> Work done... >> >> Just when I thought I found something.. =) >> http://linux.die.net/man/1/git-filter-branch >> >> Take a look at `git filter-branch --commit-filter`. Is it better than >> rebase? >> >> -- >> anatoly t. >> _______________________________________________ >> PySide mailing list >> PySide at qt-project.org >> http://lists.qt-project.org/mailman/listinfo/pyside >> From techtonik at gmail.com Mon Mar 12 10:10:38 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Mon, 12 Mar 2012 12:10:38 +0300 Subject: [PySide] 1.1.0 core dump with short-circuit signal name In-Reply-To: <4F5B7CB0.1040501@orcaware.com> References: <4F5B0F83.3020102@orcaware.com> <4F5B7CB0.1040501@orcaware.com> Message-ID: Is it possible to implement better exception handling to warn about old signal/slots being used with a redirect to this bug report before exiting? -- anatoly t. On Sat, Mar 10, 2012 at 7:09 PM, Blair Zajac wrote: > Hi Renato, > > Thanks, however, we're porting an 80k LOC project from PyQt3, so we'll > be using the old signal/slot mechanism for now just to get the app > running and then we may cut over to the new signal/slots. > > I'll open a bug report to track this. > > Blair > > On 03/10/2012 05:45 AM, Renato Araujo wrote: >> Hi Blair, >> >> I recommend to you use the new singal/slot API[1], since this is a bug >> on the deprecated API this have low priority to fix. >> >> [1]http://qt-project.org/wiki/Signals_and_Slots_in_PySide >> >> >> Renato Araujo Oliveira Filho >> >> >> >> On Sat, Mar 10, 2012 at 5:23 AM, Blair Zajac  wrote: >>> Hello, >>> >>> I'm new to PySide develop so wanted to open this on the mailing list >>> before opening a ticket, being uncertain about bug policy on this project. >>> >>> The below script core dumps using PySide 1.1.0 on Ubuntu 12.04 Beta 1. >>> Can anyone else confirm this on their system.  The same code doesn't >>> core dump on Ubuntu 11.10 with PySide 1.0.6. >>> >>> If it is reproducible, then it seems that it shouldn't core dump, even >>> if the code isn't valid PySide code. >>> >>> Below is a stack trace and the code. >>> >>> Thanks, >>> Blair >>> >>> PS We're moving a large project from PyQt3 to PySide, so that's how I >>> ran into this. >>> >>> >>> Core was generated by `python t3.py'. >>> Program terminated with signal 11, Segmentation fault. >>> #0  0x000000000041aec9 in PyObject_Call () >>> (gdb) bt >>> #0  0x000000000041aec9 in PyObject_Call () >>> #1  0x00000000004b5d96 in PyEval_CallObjectWithKeywords () >>> #2  0x00007f32da0052da in >>> PySide::SignalManager::callPythonMetaMethod(QMetaMethod const&, void**, >>> _object*, bool) () >>>     from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >>> #3  0x00007f32da00575c in PySide::SignalManager::qt_metacall(QObject*, >>> QMetaObject::Call, int, void**) () >>>     from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >>> #4  0x00007f32d958aad9 in QMetaObject::activate(QObject*, QMetaObject >>> const*, int, void**) () from /usr/lib/x86_64-linux-gnu/libQtCore.so.4 >>> #5  0x00007f32da005990 in PySide::SignalManager::qt_metacall(QObject*, >>> QMetaObject::Call, int, void**) () >>>     from /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >>> #6  0x00007f32da005fd7 in PySide::SignalManager::emitSignal(QObject*, >>> char const*, _object*) () from >>> /usr/lib/x86_64-linux-gnu/libpyside-python2.7.so.1.1 >>> #7  0x00007f32da32d938 in ?? () >>>     from /usr/lib/python2.7/dist-packages/PySide/QtCore.so >>> #8  0x00000000004b65c1 in PyEval_EvalFrameEx () >>> #9  0x00000000004b6b97 in PyEval_EvalFrameEx () >>> #10 0x00000000004bcd0d in PyEval_EvalCodeEx () >>> #11 0x00000000004bd7e2 in PyEval_EvalCode () >>> #12 0x00000000004dcc52 in run_mod () >>> #13 0x00000000004dd814 in PyRun_FileExFlags () >>> #14 0x00000000004de28e in PyRun_SimpleFileExFlags () >>> #15 0x00000000004ee6dd in Py_Main () >>> #16 0x00007f32da9e076d in __libc_start_main () >>>     from /lib/x86_64-linux-gnu/libc.so.6 >>> #17 0x000000000041ce39 in _start () >>> >>> >>> >>> #!/usr/bin/python >>> >>> from PySide import QtCore, QtGui >>> >>> SIGNAL_NAME = 'testSignal' >>> >>> class EmitterClass(QtCore.QObject): >>>      def __init__(self): >>>          super(EmitterClass, self).__init__() >>> >>>      def emitter(self): >>>          self.emit(QtCore.SIGNAL(SIGNAL_NAME), 100) >>> >>> class ListeningClass(QtCore.QObject): >>>      def __init__(self): >>>          super(ListeningClass, self).__init__() >>>          self.emitter = EmitterClass() >>>          QtCore.QObject.connect(self.emitter, QtCore.SIGNAL(SIGNAL_NAME), >>>                                 self.output) >>> >>>      def output(self, number): >>>          print number >>> >>> if __name__ == '__main__': >>>      c = ListeningClass() >>>      c.emitter.emitter() >>> _______________________________________________ >>> PySide mailing list >>> PySide at qt-project.org >>> http://lists.qt-project.org/mailman/listinfo/pyside >> > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside From matti.p.airas at nokia.com Mon Mar 12 13:13:46 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Mon, 12 Mar 2012 14:13:46 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <201203091941.11140.hugo.lima@openbossa.org> References: <201202291458.25547.hugo.lima@openbossa.org> <4F5A0231.6000606@nokia.com> <201203091941.11140.hugo.lima@openbossa.org> Message-ID: <4F5DE87A.9040707@nokia.com> On 10.03.2012 00:41, ext Hugo Parente Lima wrote: > Work done... > > .. but gerrit, oh gerrit... I can't push the changes because I don't have > permissions to push merge commits: > > Counting objects: 2086, done. > Delta compression using up to 2 threads. > Compressing objects: 100% (935/935), done. > Writing objects: 100% (2086/2086), 715.24 KiB, done. > Total 2086 (delta 1459), reused 1476 (delta 1135) > remote: Resolving deltas: 100% (1459/1459) > To ssh://hugopl at codereview.qt-project.org:29418/pyside/apiextractor.git > ! [remote rejected] master -> master (you are not allowed to upload merges) > error: failed to push some refs to 'ssh://hugopl at codereview.qt- > project.org:29418/pyside/apiextractor.git' > I sent you private mail with contact information of the Qt admin who has performed the previous clone/push operations for me. IMHO, I don't think it's a bad policy to disallow all direct pushes to the git repo. That at least enforces that every commit has passed review (and CI testing, if we'd integrate them to the Qt infra, as well), and I regard that as a good thing. It just causes trouble in the initial setup, but the Qt admins have been very fast to help with any issues so far. Cheers, ma. From hugo.lima at openbossa.org Mon Mar 12 14:11:32 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 12 Mar 2012 10:11:32 -0300 Subject: [PySide] About the next PySide release In-Reply-To: References: <201202291458.25547.hugo.lima@openbossa.org> Message-ID: <201203121011.32638.hugo.lima@openbossa.org> On Monday 12 March 2012 06:02:49 anatoly techtonik wrote: > I can't confirm that - there are no changes in the Gitorious repository. > http://qt.gitorious.org/pyside/pyside/commits/master Is done, but locally because I had problems uploading the repository tree to gerrit. > -- > anatoly t. > > > On Sun, Mar 11, 2012 at 2:43 AM, paulo alcantara > > wrote: > > Maybe its better or not. It doesnt matter either. > > The execellent work that Hugo has been doing (even with no much spare > > time to do it) is already done. > > > > On 3/10/12, anatoly techtonik wrote: > >> On Sat, Mar 10, 2012 at 1:41 AM, Hugo Parente Lima > >> > >> wrote: > >>> On Friday 09 March 2012 10:14:25 Matti Airas wrote: > >>>> On 09.03.2012 15:12, ext Anderson Lizardo wrote: > >>>> > On Fri, Mar 9, 2012 at 8:40 AM, Matti Airas > >>> > >>> wrote: > >>>> >> BTW, how did you actually do the removal? Did you just branch the > >>>> >> repos > >>>> >> before the offending commits and then merge all commits except the > >>>> >> offending ones? > >>>> > > >>>> > I wonder if a " git rebase -i commit_id^ " (where "commit_id" is > >>>> > the commit to be removed) and removing the line on the interactive > >>>> > rebase editor, would fix this more or less easily :) > >>>> > >>>> That was the approach I tried, but that breaks down in merges. > >>>> Couldn't quite figure out what the exact problem was, though. > >>>> > >>>> Cheers, > >>>> > >>>> ma. > >>> > >>> Work done... > >> > >> Just when I thought I found something.. =) > >> http://linux.die.net/man/1/git-filter-branch > >> > >> Take a look at `git filter-branch --commit-filter`. Is it better than > >> rebase? > >> > >> -- > >> anatoly t. > >> _______________________________________________ > >> PySide mailing list > >> PySide at qt-project.org > >> http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matias.rand at nokia.com Tue Mar 13 14:28:53 2012 From: matias.rand at nokia.com (Matias Rand) Date: Tue, 13 Mar 2012 14:28:53 +0100 Subject: [PySide] Qt Project Server Maintenance - Thursday March 15, 08:00 - 08:30 CET Message-ID: <4F5F4B95.2090608@nokia.com> Hi, Our hosting provider need to do a quick update on some of our servers related to qt-project.org. The affected services are: * Gerrit / Code Review - http://codereview.qt-project.org/ * Mailing lists - http://lists.qt-project.org/ * Wiki - http://wiki.qt-project.org/ The services will be unavailable for some minutes during the maintenance window, and are expected to be running as normal by 08:30 CET. Regards -- Matias From matti.p.airas at nokia.com Thu Mar 15 12:50:36 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 15 Mar 2012 13:50:36 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <201203121011.32638.hugo.lima@openbossa.org> References: <201202291458.25547.hugo.lima@openbossa.org> <201203121011.32638.hugo.lima@openbossa.org> Message-ID: <4F61D78C.80402@nokia.com> On 12.03.2012 15:11, ext Hugo Parente Lima wrote: > On Monday 12 March 2012 06:02:49 anatoly techtonik wrote: >> I can't confirm that - there are no changes in the Gitorious repository. >> http://qt.gitorious.org/pyside/pyside/commits/master > Is done, but locally because I had problems uploading the repository tree to > gerrit. ... and now they're there, with JP's bugfixes, and even the Gitorious mirror has already been updated! Thanks a million, Hugo, for the effort! Cheers, ma. From hugo.lima at openbossa.org Thu Mar 15 14:42:08 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 15 Mar 2012 10:42:08 -0300 Subject: [PySide] About the next PySide release In-Reply-To: <4F61D78C.80402@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <201203121011.32638.hugo.lima@openbossa.org> <4F61D78C.80402@nokia.com> Message-ID: <201203151042.13863.hugo.lima@openbossa.org> On Thursday 15 March 2012 08:50:36 Matti Airas wrote: > On 12.03.2012 15:11, ext Hugo Parente Lima wrote: > > On Monday 12 March 2012 06:02:49 anatoly techtonik wrote: > >> I can't confirm that - there are no changes in the Gitorious repository. > >> http://qt.gitorious.org/pyside/pyside/commits/master > > > > Is done, but locally because I had problems uploading the repository tree > > to gerrit. > > ... and now they're there, with JP's bugfixes, and even the Gitorious > mirror has already been updated! Thanks a million, Hugo, for the effort! I also did the merge of GeneratorRunner and ApiExtractor into Shiboken, so GeneratorRunner and ApiExtractor repositories can now be considered deprecated and newer Shiboken packages doesn't need to depend on them.. All history of ApiExtractor and GeneratorRunner were merged into Shiboken repository. Now there are just two failing tests blocking the release, both related to QtWebKit, despite of the fact that I need to check if all is working on a Windows and Mac machine. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matti.p.airas at nokia.com Thu Mar 15 14:49:23 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 15 Mar 2012 15:49:23 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <201203151042.13863.hugo.lima@openbossa.org> References: <201202291458.25547.hugo.lima@openbossa.org> <201203121011.32638.hugo.lima@openbossa.org> <4F61D78C.80402@nokia.com> <201203151042.13863.hugo.lima@openbossa.org> Message-ID: <4F61F363.6050008@nokia.com> On 15.03.2012 15:42, ext Hugo Parente Lima wrote: > I also did the merge of GeneratorRunner and ApiExtractor into Shiboken, so > GeneratorRunner and ApiExtractor repositories can now be considered deprecated > and newer Shiboken packages doesn't need to depend on them.. OK, that's neat. :-) I guess we can just let the repos stay around as memoirs... > All history of ApiExtractor and GeneratorRunner were merged into Shiboken > repository. Oh, that's cool! I think the first time you did it, you just copied the files over? > Now there are just two failing tests blocking the release, both related to > QtWebKit, despite of the fact that I need to check if all is working on a > Windows and Mac machine. You might want to mention that on the mailing list and ask some Windows/Mac developers to help with building the head and running the tests - that'd save you some boring work and get others involved. At the moment, I think we need all the help we can from other people... Cheers, ma. From hugo.lima at openbossa.org Thu Mar 15 15:19:16 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 15 Mar 2012 11:19:16 -0300 Subject: [PySide] About the next PySide release In-Reply-To: <4F61F363.6050008@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <201203151042.13863.hugo.lima@openbossa.org> <4F61F363.6050008@nokia.com> Message-ID: <201203151119.21687.hugo.lima@openbossa.org> On Thursday 15 March 2012 10:49:23 Matti Airas wrote: > On 15.03.2012 15:42, ext Hugo Parente Lima wrote: > > I also did the merge of GeneratorRunner and ApiExtractor into Shiboken, > > so GeneratorRunner and ApiExtractor repositories can now be considered > > deprecated and newer Shiboken packages doesn't need to depend on them.. > > OK, that's neat. :-) I guess we can just let the repos stay around as > memoirs... Yes, they must still around. > > All history of ApiExtractor and GeneratorRunner were merged into Shiboken > > repository. > > Oh, that's cool! I think the first time you did it, you just copied the > files over? No, the first time I did I also merged the histories. > > Now there are just two failing tests blocking the release, both related > > to QtWebKit, despite of the fact that I need to check if all is working > > on a Windows and Mac machine. > > You might want to mention that on the mailing list and ask some > Windows/Mac developers to help with building the head and running the > tests - that'd save you some boring work and get others involved. At the > moment, I think we need all the help we can from other people... Not just boring work, but I don't have how to test on a mac or Windows machines right now, the Windows issue is solvable but I can't say the same about mac. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From thoromyr at mac.com Thu Mar 15 16:01:00 2012 From: thoromyr at mac.com (Tim Doty) Date: Thu, 15 Mar 2012 15:01:00 +0000 (GMT) Subject: [PySide] About the next PySide release In-Reply-To: <201203151119.21687.hugo.lima@openbossa.org> Message-ID: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> What is actually involved for testing on the mac? Is there a test suite to run or what? On Mar 15, 2012, at 09:19 AM, Hugo Parente Lima wrote: > On Thursday 15 March 2012 10:49:23 Matti Airas wrote: > > You might want to mention that on the mailing list and ask some > > Windows/Mac developers to help with building the head and running the > > tests - that'd save you some boring work and get others involved. At the > > moment, I think we need all the help we can from other people... > > Not just boring work, but I don't have how to test on a mac or Windows > machines right now, the Windows issue is solvable but I can't say the same > about mac. > > -- > Hugo Parente Lima > INdT - Instituto Nokia de Tecnologia > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- An HTML attachment was scrubbed... URL: From matti.p.airas at nokia.com Thu Mar 15 16:14:40 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 15 Mar 2012 17:14:40 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <4F61F363.6050008@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <201203121011.32638.hugo.lima@openbossa.org> <4F61D78C.80402@nokia.com> <201203151042.13863.hugo.lima@openbossa.org> <4F61F363.6050008@nokia.com> Message-ID: <4F620760.6070302@nokia.com> On 15.03.2012 15:49, Matti Airas wrote: > You might want to mention that on the mailing list and ask some > Windows/Mac developers to help with building the head and running the > tests - that'd save you some boring work and get others involved. At > the moment, I think we need all the help we can from other people... Haha! Didn't realize the mailing list was CCd. Oops. :-) ma. From matti.p.airas at nokia.com Thu Mar 15 16:28:52 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 15 Mar 2012 17:28:52 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> Message-ID: <4F620AB4.3010403@nokia.com> On 15.03.2012 17:01, ext Tim Doty wrote: > What is actually involved for testing on the mac? Is there a test > suite to run or what? Hugo might want to correct me, but just go to the build directory and run "ctest" and report the results together with your configuration details on the mailing list. Cheers, ma. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmccampbell at enthought.com Thu Mar 15 16:34:05 2012 From: jmccampbell at enthought.com (Jason McCampbell) Date: Thu, 15 Mar 2012 10:34:05 -0500 Subject: [PySide] About the next PySide release In-Reply-To: <4F620AB4.3010403@nokia.com> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <4F620AB4.3010403@nokia.com> Message-ID: Well, I need to update my Mac and Windows builds anyway so I'll run the tests. Gitorious master ( http://qt.gitorious.org/pyside/pyside/commits/master) is now the master repo? I have a couple of crash fixes I had submitted to GitHub that I'll verify are or are not in the Gitorious repo as well. On Thu, Mar 15, 2012 at 10:28 AM, Matti Airas wrote: > On 15.03.2012 17:01, ext Tim Doty wrote: > > What is actually involved for testing on the mac? Is there a test suite to > run or what? > > > Hugo might want to correct me, but just go to the build directory and run > "ctest" and report the results together with your configuration details on > the mailing list. > > Cheers, > > ma. > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -- *Jason McCampbell* Enthought, Inc. 512.536.1057 (Office) 512.850.6069 (Mobile) jmccampbell at enthought.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From matti.p.airas at nokia.com Thu Mar 15 16:38:04 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Thu, 15 Mar 2012 17:38:04 +0200 Subject: [PySide] About the next PySide release In-Reply-To: References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <4F620AB4.3010403@nokia.com> Message-ID: <4F620CDC.9010904@nokia.com> On 15.03.2012 17:34, ext Jason McCampbell wrote: > Well, I need to update my Mac and Windows builds anyway so I'll run > the tests. Gitorious master > (http://qt.gitorious.org/pyside/pyside/commits/master) is now the > master repo? I have a couple of crash fixes I had submitted to GitHub > that I'll verify are or are not in the Gitorious repo as well. Thanks, nice! Gitorious is a read-only mirror of the Gerrit repo. I'm not quite sure how often it is mirrored, though (hourly or daily). But it should be recent enough, in any case. Cheers, ma. -------------- next part -------------- An HTML attachment was scrubbed... URL: From hugo.lima at openbossa.org Thu Mar 15 19:03:25 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 15 Mar 2012 15:03:25 -0300 Subject: [PySide] About the next PySide release In-Reply-To: <4F620AB4.3010403@nokia.com> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <4F620AB4.3010403@nokia.com> Message-ID: <201203151503.25456.hugo.lima@openbossa.org> On Thursday 15 March 2012 12:28:52 Matti Airas wrote: > On 15.03.2012 17:01, ext Tim Doty wrote: > > What is actually involved for testing on the mac? Is there a test > > suite to run or what? > > Hugo might want to correct me, but just go to the build directory and > run "ctest" and report the results together with your configuration > details on the mailing list. just ctest works, but "ctest -V" is better because it prints a lot of things useful to know what failed. > Cheers, > > ma. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From jmccampbell at enthought.com Thu Mar 15 21:26:58 2012 From: jmccampbell at enthought.com (Jason McCampbell) Date: Thu, 15 Mar 2012 15:26:58 -0500 Subject: [PySide] About the next PySide release In-Reply-To: <201203151503.25456.hugo.lima@openbossa.org> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <4F620AB4.3010403@nokia.com> <201203151503.25456.hugo.lima@openbossa.org> Message-ID: Running on Mac OS (10.6) using a debug build of PySide against Qt-4.7.4 I get a single test failure in the phonon package. It's hard abort complaining about "QWidget: Cannot create a QWidget when no GUI is being used", see below. I haven't looked into it any further than this thus far as I figure someone is likely a lot more familiar with what this test case should do than I am. Jason Program received signal SIGABRT, Aborted. 0x972f4c5a in __kill () (gdb) where #0 0x972f4c5a in __kill () #1 0x972f4c4c in kill$UNIX2003 () #2 0x973875a5 in raise () #3 0x9739d6e4 in abort () #4 0x0101eb59 in qt_message_output (msgType=QtFatalMsg, buf=0x42185c0 "QWidget: Cannot create a QWidget when no GUI is being used") at global/qglobal.cpp:2291 #5 0x0101ed17 in qt_message (msgType=QtFatalMsg, msg=0x1bf1e1c "QWidget: Cannot create a QWidget when no GUI is being used", ap=0xbfffbd64 "??\a\001") at global/qglobal.cpp:2337 #6 0x0101ed70 in qFatal (msg=0x1bf1e1c "QWidget: Cannot create a QWidget when no GUI is being used") at global/qglobal.cpp:2520 #7 0x0159604e in QWidgetPrivate::init () #8 0x015962c4 in QWidget::QWidget () #9 0x0790b180 in QGLWidget::QGLWidget () #10 0x04733534 in Phonon::QT7::PhononSharedQGLWidget () #11 0x04706dc7 in Phonon::QT7::QuickTimeVideoPlayer::createVisualContext () #12 0x047082f8 in Phonon::QT7::QuickTimeVideoPlayer::QuickTimeVideoPlayer () #13 0x0471bccc in Phonon::QT7::MediaObject::MediaObject () #14 0x04719000 in Phonon::QT7::Backend::createObject () #15 0x00675b16 in Phonon::Factory::createMediaObject (parent=0x45de650) at ../3rdparty/phonon/phonon/factory.cpp:330 #16 0x006849d9 in Phonon::MediaObjectPrivate::createBackendObject (this=0x45de800) at ../3rdparty/phonon/phonon/mediaobject.cpp:47 #17 0x00680d72 in Phonon::MediaNodePrivate::backendObject (this=0x45de800) at ../3rdparty/phonon/phonon/medianode.cpp:63 #18 0x00683916 in Phonon::MediaObject::setCurrentSource (this=0x45de650, newSource=@0x45ccdc0) at ../3rdparty/phonon/phonon/mediaobject.cpp:230 #19 0x00548ea3 in Sbk_Phonon_MediaObjectFunc_setCurrentSource (self=0x45de610, pyArg=0x45ddbd0) at /Users/jmccampbell/packages/PySide/BuildScripts/pyside/build/PySide/phonon/PySide/phonon/phonon_mediaobject_wrapper.cpp:987 #20 0x000cc5a6 in call_function (pp_stack=0xbfffc350, oparg=1) at Python/ceval.c:4001 #21 0x000c9211 in PyEval_EvalFrameEx (f=0x45d4740, throwflag=0) at Python/ceval.c:2666 #22 0x000cca12 in fast_function (func=0x45cb090, pp_stack=0xbfffc580, n=1, na=1, nk=0) at Python/ceval.c:4099 #23 0x000cc856 in call_function (pp_stack=0xbfffc580, oparg=0) at Python/ceval.c:4034 #24 0x000c9211 in PyEval_EvalFrameEx (f=0x45d42e0, throwflag=0) at Python/ceval.c:2666 #25 0x000cac2e in PyEval_EvalCodeEx (co=0x3e9130, globals=0x3ed6c0, locals=0x0, args=0x45cfcdc, argcount=2, kws=0x302bcc, kwcount=0, defs=0x4070fc, defcount=1, closure=0x0) at Python/ceval.c:3253 #26 0x0003b49d in function_call (func=0x406e50, arg=0x45cfcd0, kw=0x45cfad0) at Objects/funcobject.c:526 #27 0x0000d530 in PyObject_Call (func=0x406e50, arg=0x45cfcd0, kw=0x45cfad0) at Objects/abstract.c:2529 #28 0x000cd4b9 in ext_do_call (func=0x406e50, pp_stack=0xbfffc8cc, flags=3, na=1, nk=0) at Python/ceval.c:4326 #29 0x000c9376 in PyEval_EvalFrameEx (f=0x45d4170, throwflag=0) at Python/ceval.c:2705 #30 0x000cac2e in PyEval_EvalCodeEx (co=0x3e9470, globals=0x3ed6c0, locals=0x0, args=0x45d176c, argcount=2, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:3253 #31 0x0003b49d in function_call (func=0x406ed0, arg=0x45d1760, kw=0x0) at Objects/funcobject.c:526 #32 0x0000d530 in PyObject_Call (func=0x406ed0, arg=0x45d1760, kw=0x0) at Objects/abstract.c:2529 #33 0x00020bfb in instancemethod_call (func=0x406ed0, arg=0x45d1760, kw=0x0) at Objects/classobject.c:2578 #34 0x0000d530 in PyObject_Call (func=0x45d19a0, arg=0x45d1740, kw=0x0) at Objects/abstract.c:2529 #35 0x0007cbaf in slot_tp_call (self=0x45ce2c0, args=0x45d1740, kwds=0x0) at Objects/typeobject.c:5397 #36 0x0000d530 in PyObject_Call (func=0x45ce2c0, arg=0x45d1740, kw=0x0) at Objects/abstract.c:2529 #37 0x000cd023 in do_call (func=0x45ce2c0, pp_stack=0xbfffcf90, na=1, nk=0) at Python/ceval.c:4231 #38 0x000cc87b in call_function (pp_stack=0xbfffcf90, oparg=1) at Python/ceval.c:4036 #39 0x000c9211 in PyEval_EvalFrameEx (f=0x45cf8e0, throwflag=0) at Python/ceval.c:2666 #40 0x000cac2e in PyEval_EvalCodeEx (co=0x409960, globals=0x3e2330, locals=0x0, args=0x45c783c, argcount=2, kws=0x302bcc, kwcount=0, defs=0x40c25c, defcount=1, closure=0x0) at Python/ceval.c:3253 #41 0x0003b49d in function_call (func=0x406620, arg=0x45c7830, kw=0x45d1cc0) at Objects/funcobject.c:526 #42 0x0000d530 in PyObject_Call (func=0x406620, arg=0x45c7830, kw=0x45d1cc0) at Objects/abstract.c:2529 #43 0x000cd4b9 in ext_do_call (func=0x406620, pp_stack=0xbfffd2dc, flags=3, na=1, nk=0) at Python/ceval.c:4326 #44 0x000c9376 in PyEval_EvalFrameEx (f=0x45d1ad0, throwflag=0) at Python/ceval.c:2705 #45 0x000cac2e in PyEval_EvalCodeEx (co=0x3f3a50, globals=0x3e2330, locals=0x0, args=0x45c7d6c, argcount=2, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:3253 #46 0x0003b49d in function_call (func=0x3f3ba0, arg=0x45c7d60, kw=0x0) at Objects/funcobject.c:526 #47 0x0000d530 in PyObject_Call (func=0x3f3ba0, arg=0x45c7d60, kw=0x0) at Objects/abstract.c:2529 #48 0x00020bfb in instancemethod_call (func=0x3f3ba0, arg=0x45c7d60, kw=0x0) at Objects/classobject.c:2578 #49 0x0000d530 in PyObject_Call (func=0x45cc910, arg=0x45d3090, kw=0x0) at Objects/abstract.c:2529 #50 0x0007cbaf in slot_tp_call (self=0x45cf810, args=0x45d3090, kwds=0x0) at Objects/typeobject.c:5397 #51 0x0000d530 in PyObject_Call (func=0x45cf810, arg=0x45d3090, kw=0x0) at Objects/abstract.c:2529 #52 0x000cd023 in do_call (func=0x45cf810, pp_stack=0xbfffd9a0, na=1, nk=0) at Python/ceval.c:4231 #53 0x000cc87b in call_function (pp_stack=0xbfffd9a0, oparg=1) at Python/ceval.c:4036 #54 0x000c9211 in PyEval_EvalFrameEx (f=0x45d32b0, throwflag=0) at Python/ceval.c:2666 #55 0x000cac2e in PyEval_EvalCodeEx (co=0x409960, globals=0x3e2330, locals=0x0, args=0x4370bfc, argcount=2, kws=0x302bcc, kwcount=0, defs=0x40c25c, defcount=1, closure=0x0) at Python/ceval.c:3253 #56 0x0003b49d in function_call (func=0x406620, arg=0x4370bf0, kw=0x45d3220) at Objects/funcobject.c:526 #57 0x0000d530 in PyObject_Call (func=0x406620, arg=0x4370bf0, kw=0x45d3220) at Objects/abstract.c:2529 #58 0x000cd4b9 in ext_do_call (func=0x406620, pp_stack=0xbfffdcec, flags=3, na=1, nk=0) at Python/ceval.c:4326 #59 0x000c9376 in PyEval_EvalFrameEx (f=0x45d30b0, throwflag=0) at Python/ceval.c:2705 #60 0x000cac2e in PyEval_EvalCodeEx (co=0x3f3a50, globals=0x3e2330, locals=0x0, args=0x220278c, argcount=2, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:3253 #61 0x0003b49d in function_call (func=0x3f3ba0, arg=0x2202780, kw=0x0) at Objects/funcobject.c:526 #62 0x0000d530 in PyObject_Call (func=0x3f3ba0, arg=0x2202780, kw=0x0) at Objects/abstract.c:2529 #63 0x00020bfb in instancemethod_call (func=0x3f3ba0, arg=0x2202780, kw=0x0) at Objects/classobject.c:2578 #64 0x0000d530 in PyObject_Call (func=0x45cfdd0, arg=0x45d1ab0, kw=0x0) at Objects/abstract.c:2529 #65 0x0007cbaf in slot_tp_call (self=0x45ccbf0, args=0x45d1ab0, kwds=0x0) at Objects/typeobject.c:5397 #66 0x0000d530 in PyObject_Call (func=0x45ccbf0, arg=0x45d1ab0, kw=0x0) at Objects/abstract.c:2529 #67 0x000cd023 in do_call (func=0x45ccbf0, pp_stack=0xbfffe3b0, na=1, nk=0) at Python/ceval.c:4231 #68 0x000cc87b in call_function (pp_stack=0xbfffe3b0, oparg=1) at Python/ceval.c:4036 #69 0x000c9211 in PyEval_EvalFrameEx (f=0x45d24f0, throwflag=0) at Python/ceval.c:2666 #70 0x000cca12 in fast_function (func=0x421a40, pp_stack=0xbfffe5e0, n=2, na=2, nk=0) at Python/ceval.c:4099 #71 0x000cc856 in call_function (pp_stack=0xbfffe5e0, oparg=1) at Python/ceval.c:4034 #72 0x000c9211 in PyEval_EvalFrameEx (f=0x45d1f90, throwflag=0) at Python/ceval.c:2666 #73 0x000cca12 in fast_function (func=0x4159f0, pp_stack=0xbfffe810, n=1, na=1, nk=0) at Python/ceval.c:4099 #74 0x000cc856 in call_function (pp_stack=0xbfffe810, oparg=0) at Python/ceval.c:4034 #75 0x000c9211 in PyEval_EvalFrameEx (f=0x45cc4e0, throwflag=0) at Python/ceval.c:2666 #76 0x000cac2e in PyEval_EvalCodeEx (co=0x4118f0, globals=0x40d850, locals=0x0, args=0x437275c, argcount=1, kws=0x0, kwcount=0, defs=0x41679c, defcount=10, closure=0x0) at Python/ceval.c:3253 #77 0x0003b49d in function_call (func=0x415980, arg=0x4372750, kw=0x0) at Objects/funcobject.c:526 #78 0x0000d530 in PyObject_Call (func=0x415980, arg=0x4372750, kw=0x0) at Objects/abstract.c:2529 #79 0x00020bfb in instancemethod_call (func=0x415980, arg=0x4372750, kw=0x0) at Objects/classobject.c:2578 #80 0x0000d530 in PyObject_Call (func=0x45cb6a0, arg=0x302bc0, kw=0x0) at Objects/abstract.c:2529 #81 0x0007d58c in slot_tp_init (self=0x45c9a50, args=0x302bc0, kwds=0x0) at Objects/typeobject.c:5657 #82 0x00070198 in type_call (type=0x422000, args=0x302bc0, kwds=0x0) at Objects/typeobject.c:735 #83 0x0000d530 in PyObject_Call (func=0x422000, arg=0x302bc0, kw=0x0) at Objects/abstract.c:2529 #84 0x000cd023 in do_call (func=0x422000, pp_stack=0xbfffef10, na=0, nk=0) at Python/ceval.c:4231 #85 0x000cc87b in call_function (pp_stack=0xbfffef10, oparg=0) at Python/ceval.c:4036 #86 0x000c9211 in PyEval_EvalFrameEx (f=0x3c4ae0, throwflag=0) at Python/ceval.c:2666 #87 0x000cac2e in PyEval_EvalCodeEx (co=0x3c5880, globals=0x329f80, locals=0x329f80, args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:3253 #88 0x000c31c8 in PyEval_EvalCode (co=0x3c5880, globals=0x329f80, locals=0x329f80) at Python/ceval.c:667 #89 0x000f2631 in run_mod (mod=0x895b70, filename=0xbffff4ae "/Users/jmccampbell/packages/PySide/BuildScripts/pyside/tests/phonon/basic_playing_test.py", globals=0x329f80, locals=0x329f80, flags=0xbffff2b8, arena=0x3c1370) at Python/pythonrun.c:1346 #90 0x000f25c7 in PyRun_FileExFlags (fp=0xa08f48e0, filename=0xbffff4ae "/Users/jmccampbell/packages/PySide/BuildScripts/pyside/tests/phonon/basic_playing_test.py", start=257, globals=0x329f80, locals=0x329f80, closeit=1, flags=0xbffff2b8) at Python/pythonrun.c:1332 #91 0x000f1467 in PyRun_SimpleFileExFlags (fp=0xa08f48e0, filename=0xbffff4ae "/Users/jmccampbell/packages/PySide/BuildScripts/pyside/tests/phonon/basic_playing_test.py", closeit=1, flags=0xbffff2b8) at Python/pythonrun.c:936 #92 0x000f0d24 in PyRun_AnyFileExFlags (fp=0xa08f48e0, filename=0xbffff4ae "/Users/jmccampbell/packages/PySide/BuildScripts/pyside/tests/phonon/basic_playing_test.py", closeit=1, flags=0xbffff2b8) at Python/pythonrun.c:740 #93 0x0010afcd in Py_Main (argc=2, argv=0xbffff3a4) at Modules/main.c:599 #94 0x00002532 in main (argc=2, argv=0xbffff3a4) at ./Modules/python.c:23 (gdb) pystack Current language: auto; currently c++ Current language: auto; currently c /Users/jmccampbell/packages/PySide/BuildScripts/pyside/tests/phonon/basic_playing_test.py (26): setUp /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/case.py (319): run /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/case.py (391): __call__ /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/suite.py (110): run /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/suite.py (70): __call__ /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/suite.py (110): run /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/suite.py (70): __call__ /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/runner.py (153): run /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/main.py (230): runTests /Users/jmccampbell/Library/EPD/7.2-i386/lib/python2.7/unittest/main.py (95): __init__ /Users/jmccampbell/packages/PySide/BuildScripts/pyside/tests/phonon/basic_playing_test.py (67): On Thu, Mar 15, 2012 at 1:03 PM, Hugo Parente Lima wrote: > On Thursday 15 March 2012 12:28:52 Matti Airas wrote: > > On 15.03.2012 17:01, ext Tim Doty wrote: > > > What is actually involved for testing on the mac? Is there a test > > > suite to run or what? > > > > Hugo might want to correct me, but just go to the build directory and > > run "ctest" and report the results together with your configuration > > details on the mailing list. > > just ctest works, but "ctest -V" is better because it prints a lot of > things > useful to know what failed. > > > Cheers, > > > > ma. > > -- > Hugo Parente Lima > INdT - Instituto Nokia de Tecnologia > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > > -- *Jason McCampbell* Enthought, Inc. 512.536.1057 (Office) 512.850.6069 (Mobile) jmccampbell at enthought.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From jmccampbell at enthought.com Thu Mar 15 23:09:59 2012 From: jmccampbell at enthought.com (Jason McCampbell) Date: Thu, 15 Mar 2012 17:09:59 -0500 Subject: [PySide] Reference for submitting patches? Message-ID: Hi everyone, Is there a reference somewhere for how to submit patches to Gerrit? I have stumbled through far enough for it to complain about a change-id, but I don't see anything about how to generate one specifically for qt-project.org. Or, better yet, has anyone set up a flow yet to migrate pull requests from GitHub into gerrit? I have patches for several aborts (pyside-35, -36) ready but I swear I have spent more time battling with gerrit than fixing the bugs... Jason -- *Jason McCampbell* Enthought, Inc. 512.536.1057 (Office) 512.850.6069 (Mobile) jmccampbell at enthought.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From blair at orcaware.com Thu Mar 15 23:14:00 2012 From: blair at orcaware.com (Blair Zajac) Date: Thu, 15 Mar 2012 15:14:00 -0700 Subject: [PySide] Reference for submitting patches? In-Reply-To: References: Message-ID: <4F6269A8.5030608@orcaware.com> On 03/15/2012 03:09 PM, Jason McCampbell wrote: > Hi everyone, > > Is there a reference somewhere for how to submit patches to Gerrit? I > have stumbled through far enough for it to complain about a change-id, > but I don't see anything about how to generate one specifically for > qt-project.org . Or, better yet, has anyone set > up a flow yet to migrate pull requests from GitHub into gerrit? I have > patches for several aborts (pyside-35, -36) ready but I swear I have > spent more time battling with gerrit than fixing the bugs... For the ChangeId, follow the instructions here with the scp of commit-msg: http://gerrit.googlecode.com/svn/documentation/2.2.1/user-changeid.html Blair From blair at orcaware.com Thu Mar 15 23:19:48 2012 From: blair at orcaware.com (Blair Zajac) Date: Thu, 15 Mar 2012 15:19:48 -0700 Subject: [PySide] --enable-unicode=ucs4 on Mac fails to import shiboken Message-ID: <4F626B04.9010207@orcaware.com> For consistency with our Linux distro's, I've built Python 2.7 in MacPorts using the --enable-unicode=ucs4 flag, but with this flag, the import fails with: >>> from PySide import QtGui Traceback (most recent call last): File "", line 1, in ImportError: dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PySide/QtGui.so, 2): Symbol not found: _PyUnicodeUCS2 Referenced from: /opt/local/lib/libshiboken-python2.7.1.1.dylib Expected in: flat namespace in /opt/local/lib/libshiboken-python2.7.1.1.dylib Any idea why this would be the case? There's no code in PySide that explicitly lists PyUnicodeUCS2. Doing a fresh build of everything without --enable-unicode=ucs4 works fine. Could it be related to this difference when compiling Python: Without --enable-unicode=ucs4: -checking what type to use for unicode... unsigned short With --enable-unicode=ucs4: +checking what type to use for unicode... no type found Thanks, Blair From lunaryorn at googlemail.com Fri Mar 16 10:05:14 2012 From: lunaryorn at googlemail.com (Sebastian Wiesner) Date: Fri, 16 Mar 2012 10:05:14 +0100 Subject: [PySide] --enable-unicode=ucs4 on Mac fails to import shiboken In-Reply-To: <4F626B04.9010207@orcaware.com> References: <4F626B04.9010207@orcaware.com> Message-ID: Am 15. März 2012 23:19 schrieb Blair Zajac : > For consistency with our Linux distro's, I've built Python 2.7 in > MacPorts using the --enable-unicode=ucs4 flag, but with this flag, the > import fails with: > >  >>> from PySide import QtGui > Traceback (most recent call last): >   File "", line 1, in > ImportError: > dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PySide/QtGui.so, > 2): Symbol not found: _PyUnicodeUCS2 >   Referenced from: /opt/local/lib/libshiboken-python2.7.1.1.dylib >   Expected in: flat namespace >  in /opt/local/lib/libshiboken-python2.7.1.1.dylib > > Any idea why this would be the case?  There's no code in PySide that > explicitly lists PyUnicodeUCS2. UCS2 vs. UCS4 changes the ABI of the CPython library, hence extension modules built against an UCS2 build of CPython will not work in an UCS4 build of CPython and vice versa. Re-build shiboken against the UCS 4 build and everything should work as expected. From hugo.lima at openbossa.org Fri Mar 16 12:05:02 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Fri, 16 Mar 2012 08:05:02 -0300 Subject: [PySide] Reference for submitting patches? In-Reply-To: References: Message-ID: <201203160805.03079.hugo.lima@openbossa.org> On Thursday 15 March 2012 19:09:59 Jason McCampbell wrote: > Hi everyone, > > Is there a reference somewhere for how to submit patches to Gerrit? I have > stumbled through far enough for it to complain about a change-id, but I > don't see anything about how to generate one specifically for > qt-project.org. Or, better yet, has anyone set up a flow yet to migrate > pull requests from GitHub into gerrit? I have patches for several aborts > (pyside-35, -36) ready but I swear I have spent more time battling with > gerrit than fixing the bugs... The beginning is a bit painful, but then you get used to it: http://wiki.qt-project.org/Setting_up_Gerrit > Jason -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Fri Mar 16 12:21:04 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Fri, 16 Mar 2012 08:21:04 -0300 Subject: [PySide] RFC: Removing PySide rom git-hub? Message-ID: <201203160821.08361.hugo.lima@openbossa.org> Hi, gerrit only sync PySide repositories with gitorious, so the git-hub repositories will be out dated because nobody will remember to sync them :-/ So, my question is: Should I remove PySide from git-hub? No code will be lost because the helper repositories found on github and not on gerrit can be found on gitorious. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From jmccampbell at enthought.com Fri Mar 16 13:39:38 2012 From: jmccampbell at enthought.com (Jason McCampbell) Date: Fri, 16 Mar 2012 07:39:38 -0500 Subject: [PySide] Reference for submitting patches? In-Reply-To: <201203160805.03079.hugo.lima@openbossa.org> References: <201203160805.03079.hugo.lima@openbossa.org> Message-ID: > > The beginning is a bit painful, but then you get used to it: > > http://wiki.qt-project.org/Setting_up_Gerrit Ah, that was the reference I was looking for. Somehow I missed on the qt-project.org site. Thanks. -------------- next part -------------- An HTML attachment was scrubbed... URL: From techtonik at gmail.com Fri Mar 16 16:00:44 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Fri, 16 Mar 2012 17:00:44 +0200 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: <201203160821.08361.hugo.lima@openbossa.org> References: <201203160821.08361.hugo.lima@openbossa.org> Message-ID: On Fri, Mar 16, 2012 at 2:21 PM, Hugo Parente Lima wrote: > Hi, > > gerrit only sync PySide repositories with gitorious, so the git-hub > repositories will be out dated because nobody will remember to sync them :-/ > > So, my question is: Should I remove PySide from git-hub? No code will be lost > because the helper repositories found on github and not on gerrit can be found > on gitorious. Nooo. Many people can't access Gerrit and can't submit patches therefore. Gitorious suxx as was discussed many times. I'd like BuildScripts to continue using GitHub. Because barrier for patch entry to Gerrit is too high. Why isn't it possible to sync with GitHub? http://gerrit.googlecode.com/svn/documentation/2.1.5/config-replication.html It is theoretically possible to sign in into Gerrit using GitHub account and pull patches from GitHub automatically. After Gerrit will be decoupled from Jira - this will be the next logical thing to do. I use GitHub to monitor PySide development. It is convenient. -- anatoly t. From renatox at gmail.com Fri Mar 16 17:44:11 2012 From: renatox at gmail.com (Renato Araujo) Date: Fri, 16 Mar 2012 13:44:11 -0300 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: <201203160821.08361.hugo.lima@openbossa.org> References: <201203160821.08361.hugo.lima@openbossa.org> Message-ID: Hi list, I do not like both gitorious or gerrit, but since this is not a question of choice (this changes was mandatory). In my opinion github is the best tool to work with git repositories at moment. But I think we need keep all the changes in the same place to avoid redundancy, and try to keep the work easy for the maintainers. +1 Renato Araujo Oliveira Filho On Fri, Mar 16, 2012 at 8:21 AM, Hugo Parente Lima wrote: > Hi, > > gerrit only sync PySide repositories with gitorious, so the git-hub > repositories will be out dated because nobody will remember to sync them :-/ > > So, my question is: Should I remove PySide from git-hub? No code will be lost > because the helper repositories found on github and not on gerrit can be found > on gitorious. > > -- > Hugo Parente Lima > INdT - Instituto Nokia de Tecnologia > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside > From blair at orcaware.com Fri Mar 16 18:19:26 2012 From: blair at orcaware.com (Blair Zajac) Date: Fri, 16 Mar 2012 10:19:26 -0700 Subject: [PySide] --enable-unicode=ucs4 on Mac fails to import shiboken In-Reply-To: References: <4F626B04.9010207@orcaware.com> Message-ID: <4F63761E.4070205@orcaware.com> On 03/16/2012 02:05 AM, Sebastian Wiesner wrote: > Am 15. März 2012 23:19 schrieb Blair Zajac: >> For consistency with our Linux distro's, I've built Python 2.7 in >> MacPorts using the --enable-unicode=ucs4 flag, but with this flag, the >> import fails with: >> >> >>> from PySide import QtGui >> Traceback (most recent call last): >> File "", line 1, in >> ImportError: >> dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PySide/QtGui.so, >> 2): Symbol not found: _PyUnicodeUCS2 >> Referenced from: /opt/local/lib/libshiboken-python2.7.1.1.dylib >> Expected in: flat namespace >> in /opt/local/lib/libshiboken-python2.7.1.1.dylib >> >> Any idea why this would be the case? There's no code in PySide that >> explicitly lists PyUnicodeUCS2. > > UCS2 vs. UCS4 changes the ABI of the CPython library, hence extension > modules built against an UCS2 build of CPython will not work in an > UCS4 build of CPython and vice versa. > > Re-build shiboken against the UCS 4 build and everything should work > as expected. Right, I'm aware that you need to recompile everything. This failure is after a complete recompile. Here are the commands I ran: $ sudo port uninstall -v --follow-dependents python27 $ sudo port install -v py27-pyside And then I get the link error. Blair From hugo.lima at openbossa.org Fri Mar 16 18:35:06 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Fri, 16 Mar 2012 14:35:06 -0300 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: References: <201203160821.08361.hugo.lima@openbossa.org> Message-ID: <201203161435.11681.hugo.lima@openbossa.org> On Friday 16 March 2012 12:00:44 anatoly techtonik wrote: > On Fri, Mar 16, 2012 at 2:21 PM, Hugo Parente Lima > > wrote: > > Hi, > > > > gerrit only sync PySide repositories with gitorious, so the git-hub > > repositories will be out dated because nobody will remember to sync them > > :-/ > > > > So, my question is: Should I remove PySide from git-hub? No code will be > > lost because the helper repositories found on github and not on gerrit > > can be found on gitorious. > > Nooo. Many people can't access Gerrit and can't submit patches therefore. > Gitorious suxx as was discussed many times. I'd like BuildScripts to > continue using GitHub. Because barrier for patch entry to Gerrit is too > high. > > Why isn't it possible to sync with GitHub? > http://gerrit.googlecode.com/svn/documentation/2.1.5/config-replication.htm > l I agree, but this configuration isn't up to me, I need to ask to people on qt- project if they can do this for us. I'm going to ping them after the weekend. > It is theoretically possible to sign in into Gerrit using GitHub > account and pull patches from GitHub automatically. After Gerrit will > be decoupled from Jira - this will be the next logical thing to do. I > use GitHub to monitor PySide development. It is convenient. yes, I know, but if the repositories get out-dated you will monitor nothing :-/. > -- > anatoly t. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From lunaryorn at googlemail.com Sat Mar 17 11:03:31 2012 From: lunaryorn at googlemail.com (Sebastian Wiesner) Date: Sat, 17 Mar 2012 11:03:31 +0100 Subject: [PySide] --enable-unicode=ucs4 on Mac fails to import shiboken In-Reply-To: <4F63761E.4070205@orcaware.com> References: <4F626B04.9010207@orcaware.com> <4F63761E.4070205@orcaware.com> Message-ID: Am 16. März 2012 18:19 schrieb Blair Zajac : > On 03/16/2012 02:05 AM, Sebastian Wiesner wrote: >> >> Am 15. März 2012 23:19 schrieb Blair Zajac: >>> >>> For consistency with our Linux distro's, I've built Python 2.7 in >>> MacPorts using the --enable-unicode=ucs4 flag, but with this flag, the >>> import fails with: >>> >>>  >>>  from PySide import QtGui >>> Traceback (most recent call last): >>>   File "", line 1, in >>> ImportError: >>> >>> dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PySide/QtGui.so, >>> 2): Symbol not found: _PyUnicodeUCS2 >>>   Referenced from: /opt/local/lib/libshiboken-python2.7.1.1.dylib >>>   Expected in: flat namespace >>>  in /opt/local/lib/libshiboken-python2.7.1.1.dylib >>> >>> Any idea why this would be the case?  There's no code in PySide that >>> explicitly lists PyUnicodeUCS2. >> >> >> UCS2 vs. UCS4 changes the ABI of the CPython library, hence extension >> modules built against an UCS2 build of CPython will not work in an >> UCS4 build of CPython and vice versa. >> >> Re-build shiboken against the UCS 4 build and everything should work >> as expected. > > Right, I'm aware that you need to recompile everything.  This failure is > after a complete recompile.  Here are the commands I ran: > > $ sudo port uninstall -v --follow-dependents python27 > $ sudo port install -v py27-pyside > > And then I get the link error. Did you make sure that you really used the ports python, and not accidentally the system python? From techtonik at gmail.com Sat Mar 17 15:51:22 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 17 Mar 2012 16:51:22 +0200 Subject: [PySide] Open Gerrit to public (Was: Call for help: regression bugs to be fixed) In-Reply-To: References: <4F50A3F5.9010605@nokia.com> <4F548BE8.1050307@pelago.org.uk> Message-ID: On Mon, Mar 5, 2012 at 1:02 PM, anatoly techtonik wrote: > On Mon, Mar 5, 2012 at 12:48 PM, John Veness > wrote: >> This Gerrit-requiring-login issue is a Qt-wide problem, not specific to >> PySide. I notice that there is already a bugtracker issue raised against >> it - https://bugreports.qt-project.org/browse/QTQAINFRA-357 . It might >> be more productive to discuss this on one of the Qt mailing lists (see >> http://lists.qt-project.org/mailman/listinfo). > > Thanks for the heads up. It is nice to see I am not alone to take it > seriously. There is a hint that Jira is the reason, but no technical > explanation. It is very unusual that open process of open source > project with open government is locked be the dependency from > proprietary product. At least that's my impression. > > I'll write to GEG list. It is not too active, so let's hope there are > people who can help. I can't see if I wrote about it, but the issue about opening Gerrit to public has been given Critical priority. ;) https://bugreports.qt-project.org/browse/QTQAINFRA-357 If the time permits, I am going to do some research about using repose.who as an authentication server for Gerrit, so if anybody has experience or interested to help - feel free to join http://lists.qt-project.org/mailman/listinfo/geg -- anatoly t. From techtonik at gmail.com Sat Mar 17 16:25:18 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 17 Mar 2012 17:25:18 +0200 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: <201203161435.11681.hugo.lima@openbossa.org> References: <201203160821.08361.hugo.lima@openbossa.org> <201203161435.11681.hugo.lima@openbossa.org> Message-ID: On Fri, Mar 16, 2012 at 8:35 PM, Hugo Parente Lima wrote: > On Friday 16 March 2012 12:00:44 anatoly techtonik wrote: >> On Fri, Mar 16, 2012 at 2:21 PM, Hugo Parente Lima >> >> wrote: >> > Hi, >> > >> > gerrit only sync PySide repositories with gitorious, so the git-hub >> > repositories will be out dated because nobody will remember to sync them >> > :-/ >> > >> > So, my question is: Should I remove PySide from git-hub? No code will be >> > lost because the helper repositories found on github and not on gerrit >> > can be found on gitorious. >> >> Nooo. Many people can't access Gerrit and can't submit patches therefore. >> Gitorious suxx as was discussed many times. I'd like BuildScripts to >> continue using GitHub. Because barrier for patch entry to Gerrit is too >> high. >> >> Why isn't it possible to sync with GitHub? >> http://gerrit.googlecode.com/svn/documentation/2.1.5/config-replication.htm >> l > > I agree, but this configuration isn't up to me, I need to ask to people on qt- > project if they can do this for us. I'm going to ping them after the weekend. I was about to sent a request to geg at qt-project.org but then noticed that PySide and Gitorious repos are in sync: http://qt.gitorious.org/pyside/pyside/commits/master https://github.com/PySide/PySide/commits/master Is GitHub sync still an issue? It may worth to add a note though that the proper place to sent patches is Gerrit. BTW, what was the decision about http://www.pyside.org/ ? It can be hosted on GitHub. -- anatoly t. From techtonik at gmail.com Sat Mar 17 16:27:23 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Sat, 17 Mar 2012 17:27:23 +0200 Subject: [PySide] About the next PySide release In-Reply-To: <4F61D78C.80402@nokia.com> References: <201202291458.25547.hugo.lima@openbossa.org> <201203121011.32638.hugo.lima@openbossa.org> <4F61D78C.80402@nokia.com> Message-ID: On Thu, Mar 15, 2012 at 2:50 PM, Matti Airas wrote: > On 12.03.2012 15:11, ext Hugo Parente Lima wrote: >> On Monday 12 March 2012 06:02:49 anatoly techtonik wrote: >>> I can't confirm that - there are no changes in the Gitorious repository. >>> http://qt.gitorious.org/pyside/pyside/commits/master >> Is done, but locally because I had problems uploading the repository tree to >> gerrit. > > ... and now they're there, with JP's bugfixes, and even the Gitorious > mirror has already been updated! Thanks a million, Hugo, for the effort! +1 And even GitHub mirror has been updated. ;) -- anatoly t. From blair at orcaware.com Sat Mar 17 19:15:58 2012 From: blair at orcaware.com (Blair Zajac) Date: Sat, 17 Mar 2012 11:15:58 -0700 Subject: [PySide] --enable-unicode=ucs4 on Mac fails to import shiboken In-Reply-To: References: <4F626B04.9010207@orcaware.com> <4F63761E.4070205@orcaware.com> Message-ID: <4F64D4DE.9080607@orcaware.com> On 03/17/2012 03:03 AM, Sebastian Wiesner wrote: > Am 16. März 2012 18:19 schrieb Blair Zajac: >> On 03/16/2012 02:05 AM, Sebastian Wiesner wrote: >>> >>> Am 15. März 2012 23:19 schrieb Blair Zajac: >>>> >>>> For consistency with our Linux distro's, I've built Python 2.7 in >>>> MacPorts using the --enable-unicode=ucs4 flag, but with this flag, the >>>> import fails with: >>>> >>>> >>> from PySide import QtGui >>>> Traceback (most recent call last): >>>> File "", line 1, in >>>> ImportError: >>>> >>>> dlopen(/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/PySide/QtGui.so, >>>> 2): Symbol not found: _PyUnicodeUCS2 >>>> Referenced from: /opt/local/lib/libshiboken-python2.7.1.1.dylib >>>> Expected in: flat namespace >>>> in /opt/local/lib/libshiboken-python2.7.1.1.dylib >>>> >>>> Any idea why this would be the case? There's no code in PySide that >>>> explicitly lists PyUnicodeUCS2. >>> >>> >>> UCS2 vs. UCS4 changes the ABI of the CPython library, hence extension >>> modules built against an UCS2 build of CPython will not work in an >>> UCS4 build of CPython and vice versa. >>> >>> Re-build shiboken against the UCS 4 build and everything should work >>> as expected. >> >> Right, I'm aware that you need to recompile everything. This failure is >> after a complete recompile. Here are the commands I ran: >> >> $ sudo port uninstall -v --follow-dependents python27 >> $ sudo port install -v py27-pyside >> >> And then I get the link error. > > Did you make sure that you really used the ports python, and not > accidentally the system python? I did some follow up work at the MacPorts compile and it did pick up the system's Python at compile time, so it used UCS2. I'm working with the port maintainer to fix it. Sorry for the noise. Blair From hugo.lima at openbossa.org Mon Mar 19 12:32:32 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 19 Mar 2012 08:32:32 -0300 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: References: <201203160821.08361.hugo.lima@openbossa.org> <201203161435.11681.hugo.lima@openbossa.org> Message-ID: <201203190832.32244.hugo.lima@openbossa.org> On Saturday 17 March 2012 12:25:18 anatoly techtonik wrote: > On Fri, Mar 16, 2012 at 8:35 PM, Hugo Parente Lima > > wrote: > > On Friday 16 March 2012 12:00:44 anatoly techtonik wrote: > >> On Fri, Mar 16, 2012 at 2:21 PM, Hugo Parente Lima > >> > >> wrote: > >> > Hi, > >> > > >> > gerrit only sync PySide repositories with gitorious, so the git-hub > >> > repositories will be out dated because nobody will remember to sync > >> > them > >> > > >> > :-/ > >> > > >> > So, my question is: Should I remove PySide from git-hub? No code will > >> > be lost because the helper repositories found on github and not on > >> > gerrit can be found on gitorious. > >> > >> Nooo. Many people can't access Gerrit and can't submit patches > >> therefore. Gitorious suxx as was discussed many times. I'd like > >> BuildScripts to continue using GitHub. Because barrier for patch entry > >> to Gerrit is too high. > >> > >> Why isn't it possible to sync with GitHub? > >> http://gerrit.googlecode.com/svn/documentation/2.1.5/config-replication. > >> htm l > > > > I agree, but this configuration isn't up to me, I need to ask to people > > on qt- project if they can do this for us. I'm going to ping them after > > the weekend. > > I was about to sent a request to geg at qt-project.org but then noticed > that PySide and Gitorious repos are in sync: > > http://qt.gitorious.org/pyside/pyside/commits/master > https://github.com/PySide/PySide/commits/master > > Is GitHub sync still an issue? It may worth to add a note though that > the proper place to sent patches is Gerrit. Yes, it still a issue, and a note should be posted on all repositories description besides setting them read-only. > > BTW, what was the decision about http://www.pyside.org/ ? It can be > hosted on GitHub. I really don't know, Matti is taking care of this but I think he's on vacation/traveling/or something else. > -- > anatoly t. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matti.p.airas at nokia.com Mon Mar 19 12:56:33 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Mon, 19 Mar 2012 13:56:33 +0200 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: <201203160821.08361.hugo.lima@openbossa.org> References: <201203160821.08361.hugo.lima@openbossa.org> Message-ID: <4F671EF1.1050203@nokia.com> On 16.03.2012 13:21, ext Hugo Parente Lima wrote: > Hi, > > gerrit only sync PySide repositories with gitorious, so the git-hub > repositories will be out dated because nobody will remember to sync them :-/ > > So, my question is: Should I remove PySide from git-hub? No code will be lost > because the helper repositories found on github and not on gerrit can be found > on gitorious. Hi, I think it's OK to have the repos mirrored there, but in a read-only mode. If the GitHub repos would be read-write, in that case I think the GitHub project name have to be something else than PySide to indicate that the repos are a fork of the official project. Cheers, ma. From hugo.lima at openbossa.org Tue Mar 20 14:15:02 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Tue, 20 Mar 2012 10:15:02 -0300 Subject: [PySide] PySide on GSoc 2012 Message-ID: <201203201015.07519.hugo.lima@openbossa.org> Hi, Last year PySide participated on gsoc under PSF umbrella, we had few candidates but didn't accept any students. This year we probably will be there again under PSF umbrella[1] and hopping for good students to show on, so this is the time to improve our ideas page: http://qt-project.org/wiki/PySide_GSoc_Ideas Anyone is invited to add your wishes there. [1] I already contacted the 2012 PSF GSoC coordinator and sent the required info but still waiting for a confirmation. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From matti.p.airas at nokia.com Tue Mar 20 14:28:42 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Tue, 20 Mar 2012 15:28:42 +0200 Subject: [PySide] PySide on GSoc 2012 In-Reply-To: <201203201015.07519.hugo.lima@openbossa.org> References: <201203201015.07519.hugo.lima@openbossa.org> Message-ID: <4F68860A.3090808@nokia.com> Neat! Thanks for this! Just a comment on the Android part: it's already been more or less done by Thomas Perl, see: http://thp.io/2011/pyside-android/ Maybe more thorough documentation and pre-built packaging etc "productization" could still be provided, but it's a _very_ low-hanging fruit. Cheers, ma. On 20.03.2012 15:15, ext Hugo Parente Lima wrote: > Hi, > > Last year PySide participated on gsoc under PSF umbrella, we had few > candidates but didn't accept any students. This year we probably will be there > again under PSF umbrella[1] and hopping for good students to show on, so this > is the time to improve our ideas page: > > http://qt-project.org/wiki/PySide_GSoc_Ideas > > Anyone is invited to add your wishes there. > > [1] I already contacted the 2012 PSF GSoC coordinator and sent the required > info but still waiting for a confirmation. > > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- An HTML attachment was scrubbed... URL: From matti.p.airas at nokia.com Tue Mar 20 14:39:34 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Tue, 20 Mar 2012 15:39:34 +0200 Subject: [PySide] PySide on GSoc 2012 In-Reply-To: <201203201015.07519.hugo.lima@openbossa.org> References: <201203201015.07519.hugo.lima@openbossa.org> Message-ID: <4F688896.4070503@nokia.com> One pretty big and important thing would be Qt 5 support. The alpha release should arrive today and the final by the end of June, and it'd be very nice if PySide could support it already then. I wonder has anyone on the list taken a look at it yet? What needs to be done? Nevertheless, I added the idea on the page. Cheers, ma. On 20.03.2012 15:15, ext Hugo Parente Lima wrote: > Hi, > > Last year PySide participated on gsoc under PSF umbrella, we had few > candidates but didn't accept any students. This year we probably will be there > again under PSF umbrella[1] and hopping for good students to show on, so this > is the time to improve our ideas page: > > http://qt-project.org/wiki/PySide_GSoc_Ideas > > Anyone is invited to add your wishes there. > > [1] I already contacted the 2012 PSF GSoC coordinator and sent the required > info but still waiting for a confirmation. > > > > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -------------- next part -------------- An HTML attachment was scrubbed... URL: From hugo.lima at openbossa.org Tue Mar 20 14:59:23 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Tue, 20 Mar 2012 10:59:23 -0300 Subject: [PySide] PySide on GSoc 2012 In-Reply-To: <4F688896.4070503@nokia.com> References: <201203201015.07519.hugo.lima@openbossa.org> <4F688896.4070503@nokia.com> Message-ID: <201203201059.28482.hugo.lima@openbossa.org> On Tuesday 20 March 2012 10:39:34 Matti Airas wrote: > One pretty big and important thing would be Qt 5 support. The alpha > release should arrive today and the final by the end of June, and it'd > be very nice if PySide could support it already then. > > I wonder has anyone on the list taken a look at it yet? What needs to be > done? Besides another type system, I guess we will need to review our hacks to make dynamic MetaObject work, Qt5 may ease the things for us, or not, I didn't spend much time looking at the new QMetaObject yet. > Nevertheless, I added the idea on the page. > > Cheers, > > ma. > > On 20.03.2012 15:15, ext Hugo Parente Lima wrote: > > Hi, > > > > Last year PySide participated on gsoc under PSF umbrella, we had few > > candidates but didn't accept any students. This year we probably will be > > there again under PSF umbrella[1] and hopping for good students to show > > on, so this is the time to improve our ideas page: > > > > http://qt-project.org/wiki/PySide_GSoc_Ideas > > > > Anyone is invited to add your wishes there. > > > > [1] I already contacted the 2012 PSF GSoC coordinator and sent the > > required info but still waiting for a confirmation. > > > > > > > > _______________________________________________ > > PySide mailing list > > PySide at qt-project.org > > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From techtonik at gmail.com Tue Mar 20 17:20:32 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Tue, 20 Mar 2012 19:20:32 +0300 Subject: [PySide] PySide on GSoc 2012 In-Reply-To: <4F68860A.3090808@nokia.com> References: <201203201015.07519.hugo.lima@openbossa.org> <4F68860A.3090808@nokia.com> Message-ID: On Tue, Mar 20, 2012 at 4:28 PM, Matti Airas wrote: > Neat! Thanks for this! > > Just a comment on the Android part: it's already been more or less done by > Thomas Perl, see: > > http://thp.io/2011/pyside-android/ > > Maybe more thorough documentation and pre-built packaging etc > "productization" could still be provided, but it's a _very_ low-hanging > fruit. Added two interesting things to do if I applied as a student. The stuff that seems like I'll never finish it myself. 1. Shiboken tutorial / interactive demo Write extensive tutorial on how to provide a Pythonic binding for your favorite C/C++ library with Shiboken. Take any C/C++ game with non-restrictive license and try to provide a Python console scripting API using Shiboken for it. 2. Make Shiboken compile to NaCl Streamline the process of using Python with Shiboken generated bindings from browser. -- anatoly t. From techtonik at gmail.com Wed Mar 21 06:41:09 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 21 Mar 2012 08:41:09 +0300 Subject: [PySide] BuildScripts and archived repositories on GitHub Message-ID: New repositories have hash numbers reindexed and this breaks all versions tagged by BuildScripts. Can anybody tell which revisions now correspond to the released 1.1.0 so that BuildScript can be updated? Is it possible to upload old repositories somewhere for archival purposes? (searching for hash numbers in Google) -- anatoly t. From techtonik at gmail.com Wed Mar 21 10:05:09 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 21 Mar 2012 12:05:09 +0300 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: References: <4F58CD02.8080808@nokia.com> <12EC0125-E459-42D6-8230-9230097C4DB5@nokia.com> <4F59C5C1.7020201@nokia.com> Message-ID: On Sat, Mar 10, 2012 at 10:19 AM, anatoly techtonik wrote: > On Fri, Mar 9, 2012 at 11:56 AM, Matti Airas wrote: >> On 08.03.2012 19:28, ext anatoly techtonik wrote: >>> >>> Well, somebody needs to execute this stuff - >>> http://confluence.atlassian.com/display/JIRA/Importing+Data+from+Bugzilla >>> >>> I'll try to contact tracker list about this. >> >> >> Neat! Thanks. > > Done. The letter should appear at > http://lists.qt-project.org/pipermail/jeg/ in the March section. Good news. JIRA team received the letter and opened an issue (see below). There are still some questions: 1. Should we import all tickets or those that are open right now? 2. If we import all tickets, should we preserve numbering? Please add your concerns to the JIRA ticket. ---------- Forwarded message ---------- Date: Wed, Mar 21, 2012 at 10:53 AM Subject: RE: [JEG] Import PySide bugs To: techtonik at gmail.com, jeg at qt-project.org A JIRA issue has been started to track this request. https://bugreports.qt-project.org/browse/QTJIRA-164 Please raise requests to that JIRA project. -- anatoly t. From techtonik at gmail.com Wed Mar 21 10:16:47 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 21 Mar 2012 12:16:47 +0300 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: <4F671EF1.1050203@nokia.com> References: <201203160821.08361.hugo.lima@openbossa.org> <4F671EF1.1050203@nokia.com> Message-ID: On Mon, Mar 19, 2012 at 2:56 PM, Matti Airas wrote: > On 16.03.2012 13:21, ext Hugo Parente Lima wrote: >> Hi, >> >> gerrit only sync PySide repositories with gitorious, so the git-hub >> repositories will be out dated because nobody will remember to sync them :-/ >> >> So, my question is: Should I remove PySide from git-hub? No code will be lost >> because the helper repositories found on github and not on gerrit can be found >> on gitorious. > > Hi, > > I think it's OK to have the repos mirrored there, but in a read-only > mode. If the GitHub repos would be read-write, in that case I think the > GitHub project name have to be something else than PySide to indicate > that the repos are a fork of the official project. How about killing those Gerrit repositories and leaving them archived on Github? They are not needed for development anymore. pyside/apiextractor ssh://codereview.qt-project.org:29418/pyside/apiextractor.git pyside/generatorrunner ssh://codereview.qt-project.org:29418/pyside/generatorrunner.git -- anatoly t. From th.perl at gmail.com Wed Mar 21 10:38:01 2012 From: th.perl at gmail.com (Thomas Perl) Date: Wed, 21 Mar 2012 10:38:01 +0100 Subject: [PySide] BuildScripts and archived repositories on GitHub In-Reply-To: References: Message-ID: Hi, 2012/3/21 anatoly techtonik : > Is it possible to upload old repositories somewhere for archival purposes? > (searching for hash numbers in Google) Yes, I'd also like to have the old repositories available somehow. I have a checkout from some weeks ago, so if nothing else comes up, I could upload that somewhere (maybe something like github.com/pyside-old-repositories-2012-03/[...] to make it clear that these are the repositories that are considered "old" as of 2012-03). It would be great if someone from the PySide Core Team could upload these repos, though, because I'm not sure I have the latest revisions from all the old repositories (or just tell me the revision numbers and I'll check if I have the latest). Thanks, Thomas From matti.p.airas at nokia.com Wed Mar 21 11:18:38 2012 From: matti.p.airas at nokia.com (Matti Airas) Date: Wed, 21 Mar 2012 12:18:38 +0200 Subject: [PySide] BuildScripts and archived repositories on GitHub In-Reply-To: References: Message-ID: <4F69AAFE.9060501@nokia.com> On 21.03.2012 11:38, ext Thomas Perl wrote: > Yes, I'd also like to have the old repositories available somehow. I > have a checkout from some weeks ago, so if nothing else comes up, I > could upload that somewhere (maybe something like > github.com/pyside-old-repositories-2012-03/[...] to make it clear that > these are the repositories that are considered "old" as of 2012-03). > > It would be great if someone from the PySide Core Team could upload > these repos, though, because I'm not sure I have the latest revisions > from all the old repositories (or just tell me the revision numbers > and I'll check if I have the latest). I don't really see any reason to delete the apiextractor and generatorrunner repositories from Gerrit. In my opinion, they can well reside there, and together with the Gitorious mirrors provide all the archiving functionality you need. It's not really constructive to keep muddling the project layout with unofficial GitHub repos because new commits have to land in Gerrit to fulfill the contribution agreement requirements. Hence, I'd also prefer if the build scripts would point to the official repos - but that's of course up to you. Cheers, ma. From th.perl at gmail.com Wed Mar 21 11:32:03 2012 From: th.perl at gmail.com (Thomas Perl) Date: Wed, 21 Mar 2012 11:32:03 +0100 Subject: [PySide] BuildScripts and archived repositories on GitHub In-Reply-To: <4F69AAFE.9060501@nokia.com> References: <4F69AAFE.9060501@nokia.com> Message-ID: Hi, 2012/3/21 Matti Airas : > On 21.03.2012 11:38, ext Thomas Perl wrote: >> Yes, I'd also like to have the old repositories available somehow. I >> have a checkout from some weeks ago, so if nothing else comes up, I >> could upload that somewhere (maybe something like >> github.com/pyside-old-repositories-2012-03/[...] to make it clear that >> these are the repositories that are considered "old" as of 2012-03). >> >> It would be great if someone from the PySide Core Team could upload >> these repos, though, because I'm not sure I have the latest revisions >> from all the old repositories (or just tell me the revision numbers >> and I'll check if I have the latest). > > I don't really see any reason to delete the apiextractor and > generatorrunner repositories from Gerrit. In my opinion, they can well > reside there, and together with the Gitorious mirrors provide all the > archiving functionality you need. It's not really constructive to keep > muddling the project layout with unofficial GitHub repos because new > commits have to land in Gerrit to fulfill the contribution agreement > requirements. Hence, I'd also prefer if the build scripts would point to > the official repos - but that's of course up to you. I'll definitely update the build scripts to use the new repositories and the new structure and go forward with the new, simplified build structure for PySide. However, for some use cases in the next few months, it might still be useful to be able to access the old pre-Qt Project repositories (with the "offending" commits still included). One example might be to build PySide Mobility (it seems like it is broken with the new structure, have to investigate) until we get it fixed or to "replay" the PySide on Android instructions with the old repos in case there are problems with the new repository layout and the instructions. It might be useful in debugging and finding errors (maybe there are regressions from migrating apiextractor and generatorrunner into shiboken, not that I would expect any). Of course, the old repos should be clearly marked as such, and no new development should happen there. Disk space is cheap/free, and keeping the original repositories around for a few more months might prove to be useful, and should not hurt (if they are clearly marked as old, deprecated and usually dangerous - both in the URL and maybe with a "OLD-REPOSITORY-DO-NOT-USE.txt" file committed to it containing a description where to get the latest version). Thanks, Thomas From hugo.lima at openbossa.org Wed Mar 21 13:06:21 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 21 Mar 2012 09:06:21 -0300 Subject: [PySide] Bugzilla closed for new entries In-Reply-To: References: <4F58CD02.8080808@nokia.com> Message-ID: <201203210906.25990.hugo.lima@openbossa.org> On Wednesday 21 March 2012 06:05:09 anatoly techtonik wrote: > On Sat, Mar 10, 2012 at 10:19 AM, anatoly techtonik wrote: > > On Fri, Mar 9, 2012 at 11:56 AM, Matti Airas wrote: > >> On 08.03.2012 19:28, ext anatoly techtonik wrote: > >>> Well, somebody needs to execute this stuff - > >>> http://confluence.atlassian.com/display/JIRA/Importing+Data+from+Bugzil > >>> la > >>> > >>> I'll try to contact tracker list about this. > >> > >> Neat! Thanks. > > > > Done. The letter should appear at > > http://lists.qt-project.org/pipermail/jeg/ in the March section. > > Good news. JIRA team received the letter and opened an issue (see below). > There are still some questions: > > 1. Should we import all tickets or those that are open right now? > 2. If we import all tickets, should we preserve numbering? Better to import all tickets preserving the numbering, because many of our unit test are named like test_bugXXX.py, then when we face a regression and need to do some archaeology it's a just a matter of going to bugzilla and see the bug comments. They could ignore bugs with numbers already present on jira, they are really too old anyway, probably from boost::python time. > Please add your concerns to the JIRA ticket. > > ---------- Forwarded message ---------- > Date: Wed, Mar 21, 2012 at 10:53 AM > Subject: RE: [JEG] Import PySide bugs > To: techtonik at gmail.com, jeg at qt-project.org > > > A JIRA issue has been started to track this request. > https://bugreports.qt-project.org/browse/QTJIRA-164 > > Please raise requests to that JIRA project. > -- > anatoly t. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Wed Mar 21 13:20:33 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 21 Mar 2012 09:20:33 -0300 Subject: [PySide] RFC: Removing PySide rom git-hub? In-Reply-To: References: <201203160821.08361.hugo.lima@openbossa.org> <4F671EF1.1050203@nokia.com> Message-ID: <201203210920.40476.hugo.lima@openbossa.org> On Wednesday 21 March 2012 06:16:47 anatoly techtonik wrote: > On Mon, Mar 19, 2012 at 2:56 PM, Matti Airas wrote: > > On 16.03.2012 13:21, ext Hugo Parente Lima wrote: > >> Hi, > >> > >> gerrit only sync PySide repositories with gitorious, so the git-hub > >> repositories will be out dated because nobody will remember to sync them > >> :-/ > >> > >> So, my question is: Should I remove PySide from git-hub? No code will be > >> lost because the helper repositories found on github and not on gerrit > >> can be found on gitorious. > > > > Hi, > > > > I think it's OK to have the repos mirrored there, but in a read-only > > mode. If the GitHub repos would be read-write, in that case I think the > > GitHub project name have to be something else than PySide to indicate > > that the repos are a fork of the official project. > > How about killing those Gerrit repositories and leaving them archived on > Github? They are not needed for development anymore. > > pyside/apiextractor > ssh://codereview.qt-project.org:29418/pyside/apiextractor.git > pyside/generatorrunner > ssh://codereview.qt-project.org:29418/pyside/generatorrunner.git They can still archived everywhere, this doesn't hurt because they all are synchronized, the problem with this multi-repository issue we have is that users may get out dated code from a git repository. > -- > anatoly t. > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Wed Mar 21 13:29:51 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 21 Mar 2012 09:29:51 -0300 Subject: [PySide] BuildScripts and archived repositories on GitHub In-Reply-To: References: Message-ID: <201203210929.51243.hugo.lima@openbossa.org> On Wednesday 21 March 2012 02:41:09 anatoly techtonik wrote: > New repositories have hash numbers reindexed and this breaks all > versions tagged by BuildScripts. > Can anybody tell which revisions now correspond to the released 1.1.0 > so that BuildScript can be updated? Hi The hashes were reindexed but the tags still there in the same commits, the BuildScripts used hashes instead of tag names? > Is it possible to upload old repositories somewhere for archival purposes? > (searching for hash numbers in Google) The old repositories with old hashes are on gitorious: http://qt.gitorious.org/pyside/pyside-archive http://qt.gitorious.org/pyside/shiboken-archive http://qt.gitorious.org/pyside/apiextractor-archive []`s -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From techtonik at gmail.com Wed Mar 21 17:29:32 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 21 Mar 2012 19:29:32 +0300 Subject: [PySide] BuildScripts and archived repositories on GitHub In-Reply-To: <201203210929.51243.hugo.lima@openbossa.org> References: <201203210929.51243.hugo.lima@openbossa.org> Message-ID: On Wed, Mar 21, 2012 at 3:29 PM, Hugo Parente Lima wrote: > On Wednesday 21 March 2012 02:41:09 anatoly techtonik wrote: >> New repositories have hash numbers reindexed and this breaks all >> versions tagged by BuildScripts. >> Can anybody tell which revisions now correspond to the released 1.1.0 >> so that BuildScript can be updated? > > The hashes were reindexed but the tags still there in the same commits, the > BuildScripts used hashes instead of tag names? BuildScripts uses git submodules, which refer to revisions, not to tags. But the problem seems even more complex, because it could happen that starting from revisions that was taken out all tagged builds are broken. >> Is it possible to upload old repositories somewhere for archival purposes? >> (searching for hash numbers in Google) > > The old repositories with old hashes are on gitorious: > > http://qt.gitorious.org/pyside/pyside-archive > http://qt.gitorious.org/pyside/shiboken-archive > http://qt.gitorious.org/pyside/apiextractor-archive That's great. Can anybody filter BuildScripts clone to depend on these canonical repositories for previous revisions, and reupload it so that we could have a working bisect there? In the meanwhile I've created a pull request at https://github.com/PySide/BuildScripts/pull/5 to fix BuildScrips so that people can compile current HEAD for testing. -- anatoly t. From techtonik at gmail.com Wed Mar 21 17:54:33 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 21 Mar 2012 19:54:33 +0300 Subject: [PySide] BuildScripts and archived repositories on GitHub In-Reply-To: <4F69AAFE.9060501@nokia.com> References: <4F69AAFE.9060501@nokia.com> Message-ID: On Wed, Mar 21, 2012 at 1:18 PM, Matti Airas wrote: > On 21.03.2012 11:38, ext Thomas Perl wrote: >> Yes, I'd also like to have the old repositories available somehow. I >> have a checkout from some weeks ago, so if nothing else comes up, I >> could upload that somewhere (maybe something like >> github.com/pyside-old-repositories-2012-03/[...] to make it clear that >> these are the repositories that are considered "old" as of 2012-03). >> >> It would be great if someone from the PySide Core Team could upload >> these repos, though, because I'm not sure I have the latest revisions >> from all the old repositories (or just tell me the revision numbers >> and I'll check if I have the latest). > > I don't really see any reason to delete the apiextractor and > generatorrunner repositories from Gerrit. In my opinion, they can well > reside there, and together with the Gitorious mirrors provide all the > archiving functionality you need. With http://qt.gitorious.org/pyside/apiextractor-archive http://qt.gitorious.org/pyside/generatorrunner-archive the Gerrit mirrors don't make any sense. No commits will be submitted for those, but it will confuse people who will inevitably ask about their role. To make it even more transparent for new people, I would alter the description for http://qt.gitorious.org/pyside/apiextractor-archive http://qt.gitorious.org/pyside/generatorrunner-archive to include the sentence: "This project was merged into Shiboken around the time PySide migrated to Qt Project." > It's not really constructive to keep > muddling the project layout with unofficial GitHub repos because new > commits have to land in Gerrit to fulfill the contribution agreement > requirements. Hence, I'd also prefer if the build scripts would point to > the official repos - but that's of course up to you. It is more to Thomas, but IIRC the GitHub proved to have a much better uptime history, so until Gerrit is open to public, I'd rather not point everything to Gitorious, because it is painful for people who study source from browser. GitHub descriptions could be modified to point to patch submission guide so that people avoid submitting pull requests for the mirrors. -- anatoly t. From hugo.lima at openbossa.org Wed Mar 21 23:53:09 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Wed, 21 Mar 2012 19:53:09 -0300 Subject: [PySide] We got accepted :-) Message-ID: <201203211953.17191.hugo.lima@openbossa.org> Good news everyone! We got accept to participate on gsoc 2012 under PSF umbrella: http://wiki.python.org/moin/SummerOfCode/2012 So if you are a student, have a proposal or want to do any proposal listed on our ideas page[1], go ahead! log in on melange and submit you proposal. Just keep in mind that students already involved with the PySide community are preferred instead of "free riders", so if you are new to PySide try to fix some bugs and send some patches beforehand or will be much hard to you to get accepted as a student. PySide mentors: Hugo (me), melange link_id: hugopl Marcelo Lira, melange link_id: setanta Luciano Wolf, melange link_id: luck [1] http://qt-project.org/wiki/PySide_GSoc_Ideas -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From public at francisco-garcia.net Mon Mar 26 20:18:43 2012 From: public at francisco-garcia.net (=?iso-8859-1?Q?Francisco_Garc=EDa?=) Date: Mon, 26 Mar 2012 20:18:43 +0200 Subject: [PySide] Starting with shiboken Message-ID: Hello, I am trying to collect as many examples as possible binding C++ libraries with Python (cython, ctypes, bost, switg...) First I am trying to do a type of "hello world" for binding. Something like binding this C++ library int add_these_numbers(int a, int b); however doing that with Shiboken and following the tutorial is not as piece of cake as the others. There are so many steps and files, that it can be overwhelming when compared with the others. I am even trying to avoid CMake, just to get a clear picture of what is going on. I have the feeling that Shiboken is not really hard. Just a lack of more blogs talking about it and more documentation. Also I could I could not find the sources of the tutorial. At least it is not in the shiboken repo. where could I download those sources? I also saw on the PySide website a comment about the size of the binaries generated by Boost compared against the size generated by Shiboken. It was a good post with a more detailed explanation about the reasons for droping Boost. Does anyone have the link? Can Shiboken map as much of C++ as Boost? (polymorphism, operator/function overloading...) thanks in advance for all the possible help! Francisco From hugo.lima at openbossa.org Mon Mar 26 20:48:52 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 26 Mar 2012 15:48:52 -0300 Subject: [PySide] Starting with shiboken In-Reply-To: References: Message-ID: <201203261548.58841.hugo.lima@openbossa.org> On Monday 26 March 2012 15:18:43 Francisco García wrote: > Hello, I am trying to collect as many examples as possible binding C++ > libraries with Python (cython, ctypes, bost, switg...) A good read about bindings: http://setantas.net/pages/bindingcpp.html > First I am trying to do a type of "hello world" for binding. Something like > binding this C++ library > > int add_these_numbers(int a, int b); > > however doing that with Shiboken and following the tutorial is not as piece > of cake as the others. There are so many steps and files, that it can be > overwhelming when compared with the others. I am even trying to avoid > CMake, just to get a clear picture of what is going on. I have the feeling > that Shiboken is not really hard. Just a lack of more blogs talking about > it and more documentation. Also I could I could not find the sources of > the tutorial. At least it is not in the shiboken repo. Shiboken isn't too easy to start and the lack of documentation just makes the bootstraping harder, so any initiative to create documentation and/or tutorials are welcome. > where could I download those sources? From the tutorial? Which tutorial? > I also saw on the PySide website a comment about the size of the binaries > generated by Boost compared against the size generated by Shiboken. It was > a good post with a more detailed explanation about the reasons for droping > Boost. Does anyone have the link? This must be a really old post, because we dropped boost::python more than a year ago, maybe two. > Can Shiboken map as much of C++ as Boost? (polymorphism, operator/function > overloading...) Yes it can :-). > thanks in advance for all the possible help! > > Francisco > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Mon Mar 26 20:48:52 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 26 Mar 2012 15:48:52 -0300 Subject: [PySide] Starting with shiboken In-Reply-To: References: Message-ID: <201203261548.58841.hugo.lima@openbossa.org> On Monday 26 March 2012 15:18:43 Francisco García wrote: > Hello, I am trying to collect as many examples as possible binding C++ > libraries with Python (cython, ctypes, bost, switg...) A good read about bindings: http://setantas.net/pages/bindingcpp.html > First I am trying to do a type of "hello world" for binding. Something like > binding this C++ library > > int add_these_numbers(int a, int b); > > however doing that with Shiboken and following the tutorial is not as piece > of cake as the others. There are so many steps and files, that it can be > overwhelming when compared with the others. I am even trying to avoid > CMake, just to get a clear picture of what is going on. I have the feeling > that Shiboken is not really hard. Just a lack of more blogs talking about > it and more documentation. Also I could I could not find the sources of > the tutorial. At least it is not in the shiboken repo. Shiboken isn't too easy to start and the lack of documentation just makes the bootstraping harder, so any initiative to create documentation and/or tutorials are welcome. > where could I download those sources? From the tutorial? Which tutorial? > I also saw on the PySide website a comment about the size of the binaries > generated by Boost compared against the size generated by Shiboken. It was > a good post with a more detailed explanation about the reasons for droping > Boost. Does anyone have the link? This must be a really old post, because we dropped boost::python more than a year ago, maybe two. > Can Shiboken map as much of C++ as Boost? (polymorphism, operator/function > overloading...) Yes it can :-). > thanks in advance for all the possible help! > > Francisco > _______________________________________________ > PySide mailing list > PySide at qt-project.org > http://lists.qt-project.org/mailman/listinfo/pyside -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From hugo.lima at openbossa.org Mon Mar 26 20:58:55 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Mon, 26 Mar 2012 15:58:55 -0300 Subject: [PySide] [Soc2012-general] Working with PySide for GSOC In-Reply-To: References: Message-ID: <201203261558.55705.hugo.lima@openbossa.org> On Saturday 24 March 2012 16:55:38 Julian Cromarty wrote: > Hello everyone. > > I would very much like to work with the Python Software Foundation for the > Google Summer of Code especially the PySide group (which is why I'm posting > to the pyside mailing list as well). I'm a huge fan of both Python and Qt, > though I admit I'm not perfect with the more advanced features of Python, > and as such am very keen to work on combining two of my favourite tools. > > Looking through the ideas page for PySide, I would be more than willing to > work on getting the Qt5 bindings done or just fixing loads of bugs (or in > fact any of the things listed), but the one that particularly interested me > is the idea of porting PySide to Android. I know Qt has already been ported > to Android thanks to the Necessitas project (which also has Qt integration > and so the QtCreator/PySide support might be useful to get done first) and > I'm currently looking into what may be required to compile PySide for > Android. The page linked to from the Ideas page is empty. The Android port was done, but I don't know the current state, Qt5 port is a good choice, you can start doing bindings for qtbase module, it's small and will be useful to you to get a idea of the amount of work needed to accomplish the task, you can go to #pyside and ask for help if needed. > Would anyone be interested in being my mentor for one of these ideas? > > Kind regards > Julian Cromarty -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From th.perl at gmail.com Mon Mar 26 21:11:12 2012 From: th.perl at gmail.com (Thomas Perl) Date: Mon, 26 Mar 2012 21:11:12 +0200 Subject: [PySide] [Soc2012-general] Working with PySide for GSOC In-Reply-To: <201203261558.55705.hugo.lima@openbossa.org> References: <201203261558.55705.hugo.lima@openbossa.org> Message-ID: Hi, 2012/3/26 Hugo Parente Lima : > On Saturday 24 March 2012 16:55:38 Julian Cromarty wrote: >> fact any of the things listed), but the one that particularly interested me >> is the idea of porting PySide to Android. I know Qt has already been ported >> to Android thanks to the Necessitas project (which also has Qt integration >> and so the QtCreator/PySide support might be useful to get done first) and >> I'm currently looking into what may be required to compile PySide for >> Android. The page linked to from the Ideas page is empty. > > The Android port was done, but I don't know the current state, Qt5 port is a > good choice, you can start doing bindings for qtbase module, it's small and > will be useful to you to get a idea of the amount of work needed to accomplish > the task, you can go to #pyside and ask for help if needed. The current state of the Android port: http://thp.io/2011/pyside-android/ It's still not perfect (read: polished). It needs to be upgraded to a newer version of the Qt SDK (Necessitas) and it should probably also re-use an existing SL4A/Py4A installation instead of having a custom installation in /data/python/. Alternatively, one could think about bundling the required Python modules + libpython with the .apk or make it a module for Necessitas/Ministro. This probably needs some coordination with the Necessitas (Qt for Android) project, and it should probably also upgrade to the qt-project.org-based PySide (with the new Shiboken that has apiextractor and generatorrunner included). If you have any questions about the Android port, feel free to ask. HTH :) Thomas From DRaturi at teslamotors.com Tue Mar 27 02:23:51 2012 From: DRaturi at teslamotors.com (Dhruv Raturi) Date: Mon, 26 Mar 2012 17:23:51 -0700 Subject: [PySide] QTimer setter method Message-ID: <2466BD6035D4B246BC828C512331F64D1410FF7455@SJC01-MAIL1.teslamotors.com> Hi all, I am trying to bind a class that has an instance of QTimer, which is a protected member. When run shiboken on it the generated wrapper.h file contains a setter method for this. inline void protected_mTimer_setter(QTimer value) { VitalsPage::mTimer = value; } //<- this is the generated line of code Since, QTimer::operator= is private, the above line generates a compiletime error. How to fix this? Thanks, Dhruv -------------- next part -------------- An HTML attachment was scrubbed... URL: From public at francisco-garcia.net Tue Mar 27 07:54:57 2012 From: public at francisco-garcia.net (=?iso-8859-1?Q?Francisco_Garc=EDa?=) Date: Tue, 27 Mar 2012 07:54:57 +0200 Subject: [PySide] Starting with shiboken In-Reply-To: <201203261548.58841.hugo.lima@openbossa.org> References: <201203261548.58841.hugo.lima@openbossa.org> Message-ID: Am 26.03.2012 um 20:48 schrieb Hugo Parente Lima: > On Monday 26 March 2012 15:18:43 Francisco García wrote: >> Hello, I am trying to collect as many examples as possible binding C++ >> libraries with Python (cython, ctypes, bost, switg...) > > A good read about bindings: > > http://setantas.net/pages/bindingcpp.html > OMG! That was complete bindings porno. It really covers the tiny details. I love it! but still, must face the hello world type of problem with Shiboken :( >> First I am trying to do a type of "hello world" for binding. Something like >> binding this C++ library >> >> int add_these_numbers(int a, int b); >> >> however doing that with Shiboken and following the tutorial is not as piece >> of cake as the others. There are so many steps and files, that it can be >> overwhelming when compared with the others. I am even trying to avoid >> CMake, just to get a clear picture of what is going on. I have the feeling >> that Shiboken is not really hard. Just a lack of more blogs talking about >> it and more documentation. Also I could I could not find the sources of >> the tutorial. At least it is not in the shiboken repo. > > Shiboken isn't too easy to start and the lack of documentation just makes the > bootstraping harder, so any initiative to create documentation and/or > tutorials are welcome. Well, I am for it ! > >> where could I download those sources? > > From the tutorial? Which tutorial? I meant this: http://qt-project.org/wiki/Category:LanguageBindings::PySide::Shiboken::PySide_Binding_Generation_Tutorial Although the subcategories are confusing: http://qt-project.org/wiki/Category:LanguageBindings::PySide::Shiboken I could not find the sources of the example within the shiboken repository. Copy/Paste is painful because the numbers cannot be disable. I am trying to start with a much simpler case and a simple Makefile where I could clearly see the build steps > >> I also saw on the PySide website a comment about the size of the binaries >> generated by Boost compared against the size generated by Shiboken. It was >> a good post with a more detailed explanation about the reasons for droping >> Boost. Does anyone have the link? > > This must be a really old post, because we dropped boost::python more than a > year ago, maybe two. I know it was dropped some years ago. But someone of the team wrote a good post about the reasons and giving real numbers. Like the exact difference of size in the generated binaries. I know boost generates bigger binaries, but it was great having a real case story with real numbers. Sadly I cannot find that now is this still the right place to ask about shiboken? is the source code of the tutorial available? is there an intro more "shiboken for dummies" like me? is it too hard just dropping CMake and doing all the steps by hand? From public at francisco-garcia.net Tue Mar 27 07:54:57 2012 From: public at francisco-garcia.net (=?iso-8859-1?Q?Francisco_Garc=EDa?=) Date: Tue, 27 Mar 2012 07:54:57 +0200 Subject: [PySide] Starting with shiboken In-Reply-To: <201203261548.58841.hugo.lima@openbossa.org> References: <201203261548.58841.hugo.lima@openbossa.org> Message-ID: Am 26.03.2012 um 20:48 schrieb Hugo Parente Lima: > On Monday 26 March 2012 15:18:43 Francisco García wrote: >> Hello, I am trying to collect as many examples as possible binding C++ >> libraries with Python (cython, ctypes, bost, switg...) > > A good read about bindings: > > http://setantas.net/pages/bindingcpp.html > OMG! That was complete bindings porno. It really covers the tiny details. I love it! but still, must face the hello world type of problem with Shiboken :( >> First I am trying to do a type of "hello world" for binding. Something like >> binding this C++ library >> >> int add_these_numbers(int a, int b); >> >> however doing that with Shiboken and following the tutorial is not as piece >> of cake as the others. There are so many steps and files, that it can be >> overwhelming when compared with the others. I am even trying to avoid >> CMake, just to get a clear picture of what is going on. I have the feeling >> that Shiboken is not really hard. Just a lack of more blogs talking about >> it and more documentation. Also I could I could not find the sources of >> the tutorial. At least it is not in the shiboken repo. > > Shiboken isn't too easy to start and the lack of documentation just makes the > bootstraping harder, so any initiative to create documentation and/or > tutorials are welcome. Well, I am for it ! > >> where could I download those sources? > > From the tutorial? Which tutorial? I meant this: http://qt-project.org/wiki/Category:LanguageBindings::PySide::Shiboken::PySide_Binding_Generation_Tutorial Although the subcategories are confusing: http://qt-project.org/wiki/Category:LanguageBindings::PySide::Shiboken I could not find the sources of the example within the shiboken repository. Copy/Paste is painful because the numbers cannot be disable. I am trying to start with a much simpler case and a simple Makefile where I could clearly see the build steps > >> I also saw on the PySide website a comment about the size of the binaries >> generated by Boost compared against the size generated by Shiboken. It was >> a good post with a more detailed explanation about the reasons for droping >> Boost. Does anyone have the link? > > This must be a really old post, because we dropped boost::python more than a > year ago, maybe two. I know it was dropped some years ago. But someone of the team wrote a good post about the reasons and giving real numbers. Like the exact difference of size in the generated binaries. I know boost generates bigger binaries, but it was great having a real case story with real numbers. Sadly I cannot find that now is this still the right place to ask about shiboken? is the source code of the tutorial available? is there an intro more "shiboken for dummies" like me? is it too hard just dropping CMake and doing all the steps by hand? From techtonik at gmail.com Tue Mar 27 09:29:07 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Tue, 27 Mar 2012 10:29:07 +0300 Subject: [PySide] Starting with shiboken In-Reply-To: <201203261548.58841.hugo.lima@openbossa.org> References: <201203261548.58841.hugo.lima@openbossa.org> Message-ID: On Mon, Mar 26, 2012 at 9:48 PM, Hugo Parente Lima wrote: > On Monday 26 March 2012 15:18:43 Francisco García wrote: >> Hello, I am trying to collect as many examples as possible binding C++ >> libraries with Python (cython, ctypes, bost, switg...) > > A good read about bindings: > > http://setantas.net/pages/bindingcpp.html > >> First I am trying to do a type of "hello world" for binding. Something like >> binding this C++ library >> >>     int add_these_numbers(int a, int b); >> >> however doing that with Shiboken and following the tutorial is not as piece >> of cake as the others. There are so many steps and files, that it can be >> overwhelming when compared with the others. I am even trying to avoid >> CMake, just to get a clear picture of what is going on. I have the feeling >> that Shiboken is not really hard. Just a lack of more blogs talking about >> it and more documentation. Also I could I could not find the sources of >> the tutorial. At least it is not in the shiboken repo. > > Shiboken isn't too easy to start and the lack of documentation just makes the > bootstraping harder, so any initiative to create documentation and/or > tutorials are welcome. I was interested to try Shiboken to recreate AVbin - simple Python binding for FFmpeg - http://avbin.github.com/AVbin/Home/Home.html So I've started a tutorial a while ago, but run out of free time to complete it. Feel free to fork and update. It should be a good starting point. And the Mercurial history illustrates the steps I took with problems encountered. https://bitbucket.org/techtonik/shiboken-avbin/overview Seems like typesystem.xml syntax and rules was the major showstopper at that time. -- anatoly t. From techtonik at gmail.com Tue Mar 27 09:29:07 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Tue, 27 Mar 2012 10:29:07 +0300 Subject: [PySide] Starting with shiboken In-Reply-To: <201203261548.58841.hugo.lima@openbossa.org> References: <201203261548.58841.hugo.lima@openbossa.org> Message-ID: On Mon, Mar 26, 2012 at 9:48 PM, Hugo Parente Lima wrote: > On Monday 26 March 2012 15:18:43 Francisco García wrote: >> Hello, I am trying to collect as many examples as possible binding C++ >> libraries with Python (cython, ctypes, bost, switg...) > > A good read about bindings: > > http://setantas.net/pages/bindingcpp.html > >> First I am trying to do a type of "hello world" for binding. Something like >> binding this C++ library >> >>     int add_these_numbers(int a, int b); >> >> however doing that with Shiboken and following the tutorial is not as piece >> of cake as the others. There are so many steps and files, that it can be >> overwhelming when compared with the others. I am even trying to avoid >> CMake, just to get a clear picture of what is going on. I have the feeling >> that Shiboken is not really hard. Just a lack of more blogs talking about >> it and more documentation. Also I could I could not find the sources of >> the tutorial. At least it is not in the shiboken repo. > > Shiboken isn't too easy to start and the lack of documentation just makes the > bootstraping harder, so any initiative to create documentation and/or > tutorials are welcome. I was interested to try Shiboken to recreate AVbin - simple Python binding for FFmpeg - http://avbin.github.com/AVbin/Home/Home.html So I've started a tutorial a while ago, but run out of free time to complete it. Feel free to fork and update. It should be a good starting point. And the Mercurial history illustrates the steps I took with problems encountered. https://bitbucket.org/techtonik/shiboken-avbin/overview Seems like typesystem.xml syntax and rules was the major showstopper at that time. -- anatoly t. From shoulderhigher at gmail.com Tue Mar 27 17:24:13 2012 From: shoulderhigher at gmail.com (higery) Date: Tue, 27 Mar 2012 23:24:13 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea Message-ID: The following is my GSOC2012 proposal although in a very preliminary way, I will appreciate anyone can make some suggestions about this proposal. Proposal for "Visual tool to fast edit type system files" Abstract ----------------- Typesystem file is an important XML file used by API Extractor of Pyside to save information of how the API should be exported to the chosen target language. To support different kinds of complex target languages, Pyside typesystem file must support many simple or complicated meta information, thus it's a little difficult for developers to edit these pure typesystem files, so it's very significant to write a visual tool to fast edit typesystem files. We can develop an Eclipse plugin to fast implement such kind tool. Introduction ------------------ Actually, Eclipse already has a very simple XML editor source code, which only support highlight of the specified schema. To fast implement such kind tool, we can reuse this source code and add all type system schemas which the typesystem files support. The API Extractor Type System contains about 8 parts, e.g. specifying types, manipulating Object and Value Types, modifying functions, modifying arguments, etc. , so we can support these features step by step. In the mean while, our tool should allow users insert code in a quick and intelligent way like Eclipse IDE. What's more, in future, if our tool is developed in a good status, we can add WYSIWYG feature to it -- we can import EMF framework to our tool, thus we can drag and drop an element to a panel, at the same time, generating corresponding snippets code of the type system to our typesystem file. Milestones ------------------ The following is my week-by-week plan: Week 1: investigate current API Extractor and collect the detailed requirements about what kind of tool should be implemented, from reading the source code, and the developer's ideas Week 2: investigate the Eclipse XML editor source code and make an initial solution Week 3: easy beginning - support typesystem, load-typesystem,rejection, Week 4: do more - support primitive-type, enum-type, namespace-type, reject-enum-value, value-type,object-type,interface-type,container-type,custom-type,function Week 5: support inject-code, modify-field,modify-function,add-function,conversion-rule Week 6: support modify-argument,remove,access,rename Week 7: support conversion-rule,remove-argument,rename to, remove-default-expression, replace-default-expression,replace-type,define-ownership,reference-count,replace- value,parent Week 8: support suppress-warning, extra-includes,inlude Week 9: support template,insert-template,replace Week 10: support conversion-rule, native-to-target,target-to-native,add-conversion Week 11: support inject-documentation,modify-documentation Week 12-13: enhance this tool, submit source code, write documentations, and so on. About me -------------------- I'm an undergraduate from China and I'm preparing for my computer science and technology bachelor degree in USTB (University of Science and Technology Beijing). Java and Python is the main programming language when I write programs and software. During my coding days, I’m getting to know more and more about the Open-Source/FOSS, especially its spirits-share, collaborate, make friends. Every time I find an excellent solution about some problems occurred to me, I become so excited and learned a lot from the process of problems finding-reporting-solving. I joined gsoc2011 last year and took one Python idea as my summer project, you can take a look at what the project does via python wiki page : http://wiki.python.org/moin/SummerOfCode/2011/SetuptoolsFeatures, and you also can know more through the my blog for gsoc: http://higery.wordpress.com/ Contact Info: # Name: Xu Dehai(You can call me with my English name-higery) # Blog: http://higery.wordpress.com/ # IRC: higery on irc.freenode.net -------- Best wishes, higery -------------- next part -------------- An HTML attachment was scrubbed... URL: From hugo.lima at openbossa.org Tue Mar 27 18:43:34 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Tue, 27 Mar 2012 13:43:34 -0300 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: Message-ID: <201203271343.39281.hugo.lima@openbossa.org> On Tuesday 27 March 2012 12:24:13 higery wrote: > The following is my GSOC2012 proposal although in a very preliminary way, I > will > appreciate anyone can make some suggestions about this proposal. > > > > Proposal for "Visual tool to fast edit type system files" Hi Xu Dehai, IMO a eclipse plugin doesn't really help too much, but I have a suggestion to you that's not too different from your original proposal: People complain about the difficult to create Pyhton bindings using Shiboken very often on our mailing list, so instead of a visual tool, maybe would be better to write a tool to help the bootstrap of new bindings, i.e.: You inform the header files of libraries that you want do bind and it generates basic typesystem files for these libraries plus the CMakeFiles for the project. UI is optimal, this can be a command line tool, in fact I believe that is better to be a command tool rather than a UI. The hard work will be modify ApiExtractor to do what you want, once you do this, the work can be integrated into Shiboken repository instead of a new repository just for this tool. What do you think? Suggestions? Questions? > > Abstract > ----------------- > Typesystem file is an important XML file used by API Extractor of Pyside to > save information of how the API should be exported to the chosen target > language. > To support different kinds of complex target languages, Pyside typesystem > file > must support many simple or complicated meta information, thus it's a > little > difficult for developers to edit these pure typesystem files, so it's very > significant > to write a visual tool to fast edit typesystem files. We can develop an > Eclipse plugin > to fast implement such kind tool. > > Introduction > ------------------ > Actually, Eclipse already has a very simple XML editor source code, which > only support > highlight of the specified schema. To fast implement such kind tool, we can > reuse this > source code and add all type system schemas which the typesystem files > support. The > API Extractor Type System contains about 8 parts, e.g. specifying types, > manipulating > Object and Value Types, modifying functions, modifying arguments, etc. , so > we can > support these features step by step. In the mean while, our tool should > allow users > insert code in a quick and intelligent way like Eclipse IDE. > > What's more, in future, if our tool is developed in a good status, we can > add WYSIWYG > feature to it -- we can import EMF framework to our tool, thus we can drag > and drop an > element to a panel, at the same time, generating corresponding snippets > code of the type > system to our typesystem file. > > Milestones > ------------------ > The following is my week-by-week plan: > Week 1: investigate current API Extractor and collect the detailed > requirements about > what kind of tool should be implemented, from reading the source code, and > the > developer's ideas > Week 2: investigate the Eclipse XML editor source code and make an initial > solution > Week 3: easy beginning - support typesystem, load-typesystem,rejection, > Week 4: do more - support primitive-type, enum-type, namespace-type, > reject-enum-value, > value-type,object-type,interface-type,container-type,custom-type,function > Week 5: support inject-code, > modify-field,modify-function,add-function,conversion-rule > Week 6: support modify-argument,remove,access,rename > Week 7: support conversion-rule,remove-argument,rename to, > remove-default-expression, > replace-default-expression,replace-type,define-ownership,reference-count,re > place- value,parent > Week 8: support suppress-warning, extra-includes,inlude > Week 9: support template,insert-template,replace > Week 10: support conversion-rule, > native-to-target,target-to-native,add-conversion > Week 11: support inject-documentation,modify-documentation > Week 12-13: enhance this tool, submit source code, write documentations, > and so on. > > > About me > -------------------- > I'm an undergraduate from China and I'm preparing for my computer science > and technology > bachelor degree in USTB (University of Science and Technology Beijing). > Java and Python > is the main programming language when I write programs and software. > During my coding days, I’m getting to know more and more about the > Open-Source/FOSS, > especially its spirits-share, collaborate, make friends. Every time I find > an excellent > solution about some problems occurred to me, I become so excited and > learned a lot from > the process of problems finding-reporting-solving. > > > I joined gsoc2011 last year and took one Python idea as my summer project, > you can take a > look at what the project does via python wiki page : > http://wiki.python.org/moin/SummerOfCode/2011/SetuptoolsFeatures, > and you also can know more through the my blog for gsoc: > http://higery.wordpress.com/ > > Contact Info: > # Name: Xu Dehai(You can call me with my English name-higery) > # Blog: http://higery.wordpress.com/ > # IRC: higery on irc.freenode.net > > > > -------- > Best wishes, > higery -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From techtonik at gmail.com Wed Mar 28 11:14:24 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Wed, 28 Mar 2012 12:14:24 +0300 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: Message-ID: On Tue, Mar 27, 2012 at 6:24 PM, higery wrote: > The following is my GSOC2012 proposal although in a very preliminary way, I > will > appreciate anyone can make some suggestions about this proposal. > > > > Proposal for "Visual tool to fast edit type system files" > > > Abstract > ----------------- > Typesystem file is an important XML file used by API Extractor of Pyside to > save information of how the API should be exported to the chosen target > language. > To support different kinds of complex target languages, Pyside typesystem > file > must support many simple or complicated meta information, thus it's a little > difficult for developers to edit these pure typesystem files, so it's very > significant > to write a visual tool to fast edit typesystem files. We can develop an > Eclipse plugin > to fast implement such kind tool. If it will be a visual tool for PySide, it should be in Python/Qt. So I would investigate which Python/Qt components/widgets will allow to build such tool in non-Eclipse environment. Then provide a detailed list of features that this tool should provide: 1. List of exportable symbols from the given C/C++ lib 2. Wrapper suggestions for exportable symbols 2.1 Propose several ways to wrap exportable 2.2 Make heuristics pluggable/extendible in Python 3. Compare list of exportable symbols with current schema 3.1 Investigate possibility of tree compare if imported lib contains nested structures 4. Detect wrapper errors and security issues 4.1 Make heuristics pluggable/extendible 5. Provide report to summarize export completeness, e.g.: "Exported: XX Converted: YY Skipped: ZZ Unhandled: .." These can be implemented as a separate widgets or as an addition to some IDE like Spyder. It is also possible to improve Spyder plugin system on the way. That's my own wishes. > Week 11: support inject-documentation,modify-documentation > Week 12-13: enhance this tool, submit source code, write documentations, and > so on. In the perfect word source code, documentation and test are delivered weekly. =) -- anatoly t. From shoulderhigher at gmail.com Wed Mar 28 14:59:56 2012 From: shoulderhigher at gmail.com (higery) Date: Wed, 28 Mar 2012 20:59:56 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: <201203271343.39281.hugo.lima@openbossa.org> References: <201203271343.39281.hugo.lima@openbossa.org> Message-ID: >>People complain about the difficult to create Pyhton bindings using Shiboken >>very often on our mailing list, so instead of a visual tool, maybe would be >>better to write a tool to help the bootstrap of new bindings, i.e.: I think it's a great idea which may help Shiboken users in a more direct way, so I will enhance my proposal and import this good idea. >>You inform the header files of libraries that you want do bind and it >>generates basic typesystem files for these libraries plus the CMakeFiles for >>the project. It make senses, I think I will spend more time on the implementation details. >>UI is optimal, this can be a command line tool, in fact I believe that is >>better to be a command tool rather than a UI. If we just want a bootstrap tool, so I agree with you that a command line tool is enough. >>The hard work will be modify ApiExtractor to do what you want, once you do >>this, the work can be integrated into Shiboken repository instead of a new >>repository just for this tool. That is what I want - make real contributions to the master repository! Thank you very much for these good suggestion, I will enhance my proposal ASAP after a deeper investigation. Best regards, -------------- next part -------------- An HTML attachment was scrubbed... URL: From shoulderhigher at gmail.com Wed Mar 28 15:41:57 2012 From: shoulderhigher at gmail.com (higery) Date: Wed, 28 Mar 2012 21:41:57 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: Message-ID: > If it will be a visual tool for PySide, it should be in Python/Qt. So > I would investigate which Python/Qt components/widgets will allow to > build such tool in non-Eclipse environment. Then provide a detailed > list of features that this tool should provide: > > 1. List of exportable symbols from the given C/C++ lib > 2. Wrapper suggestions for exportable symbols > 2.1 Propose several ways to wrap exportable > 2.2 Make heuristics pluggable/extendible in Python > 3. Compare list of exportable symbols with current schema > 3.1 Investigate possibility of tree compare if imported lib contains > nested structures > 4. Detect wrapper errors and security issues > 4.1 Make heuristics pluggable/extendible > 5. Provide report to summarize export completeness, e.g.: > "Exported: XX Converted: YY Skipped: ZZ Unhandled: .." > Now I prefer hugo's idea - implement a command line tool first, but at the same time, you are very right, I think it should cover the most of these features you listed above. I plan to use Python to implement such tool first, and after it is implemented in a good status, we can try to use Qt UI wigdets to implement such Visual tool which will encapsulate the command line tool. > > In the perfect word source code, documentation and test are delivered > weekly. =) > OK. It maybe added in my weekly plan. Best regards, -------------- next part -------------- An HTML attachment was scrubbed... URL: From shoulderhigher at gmail.com Thu Mar 29 07:15:47 2012 From: shoulderhigher at gmail.com (higery) Date: Thu, 29 Mar 2012 13:15:47 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: Message-ID: > To give more context in the APIExtractor part, it does the hard work > of parsing C++ headers, type system xml description of the binding, > and merge these two sets of information in something the binding > generator may work with. > At the moment it works only with both sets of information (headers + > xml), but the ideal would be to modify it to take a given set of > headears and from them produce an initial xml description (or the > information needed by another tool to produce such xml file). > This way you can't avoid dealing with C++. If you decide to do part of > the work modifying APIExtractor, and part creating the tool that uses > its information, you're still going to be in the C++ only camp. > One interesting alternative is to, after improving APIExtractor lib to > work as you need, produce Python bindings for it (or at least the > parts useful for your goals) and use them to write your tool in > Python. Agree. So a plan in a sentence should be improving APIExtractor lib first which can help generate the Python bindings as what I want , and write such command tool using these bindings second. Thank you very much! > Such APIExtractor bindings would certainly be very useful in other > contexts, and a good thing in itself. I also hope so. Best regards, From shoulderhigher at gmail.com Thu Mar 29 07:21:32 2012 From: shoulderhigher at gmail.com (higery) Date: Thu, 29 Mar 2012 13:21:32 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: <201203282240.31473.hugo.pl@gmail.com> References: <201203282240.31473.hugo.pl@gmail.com> Message-ID: > Remember that the tool must parse the C++ headers to find classes and if > possible mark some methods that may need code injection or argument > modification, OK. >I don't know if there are python C++ parsers are up to this task. I also don't know; I think I should spend some time on investigating that first. >> > In the perfect word source code, documentation and test are delivered >> > weekly. =) >> >> OK. It maybe added in my weekly plan. > > I would mentor this proposal. > Thanks for your attention, I will (in 8 hours) paste an enhanced proposal basing on discussings on this thread. Best regards to all of you, From techtonik at gmail.com Thu Mar 29 09:12:54 2012 From: techtonik at gmail.com (anatoly techtonik) Date: Thu, 29 Mar 2012 10:12:54 +0300 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: <201203282240.31473.hugo.pl@gmail.com> Message-ID: On Thu, Mar 29, 2012 at 8:21 AM, higery wrote: >> Remember that the tool must parse the C++ headers to find classes and if >> possible mark some methods that may need code injection or argument >> modification, > OK. > >>I don't know if there are python C++ parsers are up to this task. > I also don't know; I think I should spend some time on investigating that first. There is an article about parsing C++ with clang: http://eli.thegreenplace.net/2011/07/03/parsing-c-in-python-with-clang/ And frankly, I expected Shiboken to provide binding for C code as well, so this can be covered to some degree in proposal too. And I agree that having basis functionality first is more important than GUI tools. -- anatoly t. From shoulderhigher at gmail.com Thu Mar 29 16:48:13 2012 From: shoulderhigher at gmail.com (higery) Date: Thu, 29 Mar 2012 22:48:13 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: <201203282240.31473.hugo.pl@gmail.com> Message-ID: Sorry for replying a little late, I have created a newer proposal basing on all you guys' good idea and suggestions, however there still are many details to talk over, any other suggestions about the project or my proposal are very appreciated. Thank you to everyone! ------------------------------------------------------------------------------------------------- Proposal -- "Useful tool to help bootstrap of new bindings" Abstract ----------------- Shiboken is a front-end plugin for Generator Runner which can generate bindings for C++ libraries using CPython source code. But because of complicated language syntax and runtime mechanism of C++, it seems a little difficult for people who use Shiboken very often to create Python bindings, and there are already some complains in the Pyside mailing list [?]. So, it’s very valuable and significant to implement a tool to help bootstrap of new bindings. Introduction ------------------ This tool would be a command line tool, and when using it, people could inform the header files of libraries which they want to bind, after running, it will generate basic typesystem files for these libraries plus the CMakeFiles for the project. To let APIExtractor do what we want, we have to modify APIExtractor, because it does the hard work of parsing C++ headers, typesystem XML description files of the binding, and must use these two sets of information that the binding generator may work with. So, an ideal way would be to modify APIExtractor to take a given set of headers and from them produce an intermediate XML description file, which contains the information needed by our tool to generate final binding typesystem files. However, in the process of modifying work, we can’t avoid dealing with C++, thus it may be not practical for us to do part work of modifying APIExtractor while creating the tool that uses its information at the sametime. So, in this summer project, preliminary work will be modifying APIExtractor first, and writing the command tool second. The following is the concrete weekly plan: Timeline ------------------ Week 1: Get the clone of Pyside and Study the whole framework of APIExtractor through reading the source code. Week 2-Week 8: Modify APIExtractor to do what we want. The detail is: --Week 2: find the classes through parsing C++ headers --Week 3: handle namespace-type, interface-type, object-type --Week 4: extract primitive-type, enum-type,value-type --Week 5: extract other complex typs like Template --Week 6: handle C/C++ function and mark them if they need code injection or argument modification or not --Week 7: can extract all exportable symbols from the given library --Week 8: write these information to an intermediate XML file Week 9-Week 12: Create the command tool to help bootstrap of new bindings and do test work --Week 9: write Python script to offer function of generating new binding typesystem files from the intermediate file created before --Week 10: write tests against such script --Week 11: through times testing, check if the previous modifying work on APIExtractor behaves well or not --Week 12: help improve our command tool in return Week 13-Week14: perfect the code and documentations, and begin to submit patches --Week 13: perfect the code and documentations --Week 14: create patches and try to submit them Week 15….: summarize the whole work and say goodbye to summer About me -------------------- I'm an undergraduate from China and I'm preparing for my computer science and technology bachelor degree in USTB (University of Science and Technology Beijing). Java and Python is the main programming language when I write programs and software. During my coding days, I’m getting to know more and more about the Open-Source/FOSS, especially its spirits-share, collaborate, make friends. Every time I find an excellent solution about some problems occurred to me, I become so excited and learned a lot from the process of problems finding-reporting-solving. I joined gsoc2011 last year and took one Python idea as my summer project, you can take a look at what the project does via python wiki page http://wiki.python.org/moin/SummerOfCode/2011/SetuptoolsFeatures, and you also can know more through my blog for gsoc: http://higery.wordpress.com/ Contact Info: # Name: Xu Dehai(You can call me with my English name-higery) # Blog: http://higery.wordpress.com/ # IRC: higery on irc.freenode.net -------------- next part -------------- An HTML attachment was scrubbed... URL: From mort.yao at gmail.com Thu Mar 29 19:36:48 2012 From: mort.yao at gmail.com (Mort) Date: Thu, 29 Mar 2012 19:36:48 +0200 Subject: [PySide] Proposal of PySide/Python support for Qt Creator Message-ID: Hello everyone! I'm posting here because I would like to add support for Python/PySide on Qt Creator, and hopefully, as a Google Summer of Code project. I don't know if there are many people interested in this idea, but for those people who ever worked in both C++ and Python with Qt / PySide, an all-in-one IDE should be no harmful, at least. And if you find great fun in making GUIs using QtCreator's designer, as well as programming in Python/PySide, you know what I'm talking about:-) I realized that the PySide community once had made some effort to write a plugin to add Python support into Qt Creator (from what I found on the GitHub PySide/QtCreator repository), however, for some reason it discontinued (that's as far as I know). From what I see in discussion forums, people keep wanting Python support in Qt Creator, and Tobias said though it is not a priority right now, he would like to see someone contributing in this area. I'm eager to see (and also eagerly need myself) an IDE with PySide support therefore I'm thinking of taking it as a GSoC project. Abstract ---------- The main goal of this project is to add fully functional Python/PySide support in a "feature branch" of Qt Creator, as well as some extra components such as optional Python/PySide-related plugins, PySide (and Shiboken) tutorials, demos and examples inside Qt Creator's helper system. Code contributes back to both Qt Creator and PySide repo via Gerrit. Blog post of my full proposal (with detailed schedule): http://www.soimort.org/gsoc/ Take a look if you'r interested in, and feel free to comment on, suggestions, criticisms, anything :) I will be updating that page. Planned working repo: https://gitorious.org/~mort/qt-creator/qt-creator-py Many thanks for reading this, and any feedback is needed please! p.s. one question: from what I read in http://wiki.python.org/moin/SummerOfCode/2012 , prospective students are required to "contribute at least one patch to the project(s) they're applying to work with". Although the idea of PySide/Python support for QtCreator exists on the PySide idea page, there does not exist anything related in the official PySide repo (I believe I will be working on a QtCreator fork instead of a PySide repo). So how do I get started? Does a patch for QtCreator count? Best Regards, Mort Yao Email: mort.yao at gmail.com Address: Olastorpsvägen 23:119, 29139 Kristianstad, Sweden -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.richi at bluewin.ch Thu Mar 29 19:39:20 2012 From: a.richi at bluewin.ch (Aaron Richiger) Date: Thu, 29 Mar 2012 19:39:20 +0200 Subject: [PySide] Proposal of PySide/Python support for Qt Creator In-Reply-To: References: Message-ID: <4F749E48.2050907@bluewin.ch> +1!!!! From expo at expobrain.net Thu Mar 29 20:27:20 2012 From: expo at expobrain.net (Daniele Esposti) Date: Thu, 29 Mar 2012 19:27:20 +0100 Subject: [PySide] Proposal of PySide/Python support for Qt Creator In-Reply-To: References: Message-ID: +1 -- Daniele Esposti My Blog http://www.expobrain.net LinkedIn http://www.linkedin.com/in/danieleesposti Twitter http://www.twitter.com/#!/expobrain From paulo.alcantara at openbossa.org Thu Mar 29 20:32:36 2012 From: paulo.alcantara at openbossa.org (paulo alcantara) Date: Thu, 29 Mar 2012 15:32:36 -0300 Subject: [PySide] Proposal of PySide/Python support for Qt Creator In-Reply-To: References: Message-ID: Hi Mort, > I'm posting here because I would like to add support for Python/PySide on > Qt > Creator, and hopefully, as a Google Summer of Code project. This is really great! > I realized that the PySide community once had made some effort to write a > plugin to add Python support into Qt Creator (from what I found on the > GitHub PySide/QtCreator repository), however, for some reason it > discontinued (that's as far as I know). The project was *started* by INdT (Instituto Nokia de Tecnologia) - I and my team (PySide) were the only contributors in it. An yes, it's currently discontinued and that'd be nice to see someone to continue on hacking it. IIRC, I managed to run Python programs on QtCreator, as well as syntax highlighting support working fine. If you become a student of GSoC 2012 for the PySide, for example, ideally you must keep INdT and/or any other author copyright'ed with your work (just a note). BTW, good luck with your proposal! :-) -pcacjr From jpe at wingware.com Thu Mar 29 22:05:18 2012 From: jpe at wingware.com (John Ehresman) Date: Thu, 29 Mar 2012 16:05:18 -0400 Subject: [PySide] About the next PySide release In-Reply-To: <201203151503.25456.hugo.lima@openbossa.org> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <4F620AB4.3010403@nokia.com> <201203151503.25456.hugo.lima@openbossa.org> Message-ID: <4F74C07E.2060900@wingware.com> On 3/15/12 2:03 PM, Hugo Parente Lima wrote: > On Thursday 15 March 2012 12:28:52 Matti Airas wrote: >> Hugo might want to correct me, but just go to the build directory and >> run "ctest" and report the results together with your configuration >> details on the mailing list. > > just ctest works, but "ctest -V" is better because it prints a lot of things > useful to know what failed. I just get a message that no test configuration file is found when I try this on Windows, either in the root of the pyside source or the build directory where cmake is run. Do I need to run ctest from another directory or is there another way to run the tests? Thanks, John From hugo.lima at openbossa.org Thu Mar 29 22:15:38 2012 From: hugo.lima at openbossa.org (Hugo Parente Lima) Date: Thu, 29 Mar 2012 17:15:38 -0300 Subject: [PySide] About the next PySide release In-Reply-To: <4F74C07E.2060900@wingware.com> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <201203151503.25456.hugo.lima@openbossa.org> <4F74C07E.2060900@wingware.com> Message-ID: <201203291715.44357.hugo.lima@openbossa.org> On Thursday 29 March 2012 17:05:18 John Ehresman wrote: > On 3/15/12 2:03 PM, Hugo Parente Lima wrote: > > On Thursday 15 March 2012 12:28:52 Matti Airas wrote: > >> Hugo might want to correct me, but just go to the build directory and > >> run "ctest" and report the results together with your configuration > >> details on the mailing list. > > > > just ctest works, but "ctest -V" is better because it prints a lot of > > things useful to know what failed. > > I just get a message that no test configuration file is found when I try > this on Windows, either in the root of the pyside source or the build > directory where cmake is run. Do I need to run ctest from another > directory or is there another way to run the tests? You need to run on build directory, I don't remember if the tests were disabled by default on windows builds, but in any case you can enabled then passing: -DBUILD_TESTS=ON on cmake (on both, PySide and Shiboken). > Thanks, > > John -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: From jpe at wingware.com Thu Mar 29 22:17:32 2012 From: jpe at wingware.com (John Ehresman) Date: Thu, 29 Mar 2012 16:17:32 -0400 Subject: [PySide] About the next PySide release In-Reply-To: <4F74C07E.2060900@wingware.com> References: <11659099-0671-40b4-c588-9fc2d35e8f69@me.com> <4F620AB4.3010403@nokia.com> <201203151503.25456.hugo.lima@openbossa.org> <4F74C07E.2060900@wingware.com> Message-ID: <4F74C35C.7010204@wingware.com> On 3/29/12 4:05 PM, John Ehresman wrote: > On 3/15/12 2:03 PM, Hugo Parente Lima wrote: >> On Thursday 15 March 2012 12:28:52 Matti Airas wrote: >>> Hugo might want to correct me, but just go to the build directory and >>> run "ctest" and report the results together with your configuration >>> details on the mailing list. >> >> just ctest works, but "ctest -V" is better because it prints a lot of things >> useful to know what failed. > > I just get a message that no test configuration file is found when I try > this on Windows, either in the root of the pyside source or the build > directory where cmake is run. Do I need to run ctest from another > directory or is there another way to run the tests? Oops, I had disabled tests when building to reduce compilation time and am rebuilding now. Sorry for not checking this first, John From shoulderhigher at gmail.com Fri Mar 30 06:01:41 2012 From: shoulderhigher at gmail.com (higery) Date: Fri, 30 Mar 2012 12:01:41 +0800 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: <201203291401.45113.hugo.pl@gmail.com> References: <201203291401.45113.hugo.pl@gmail.com> Message-ID: > Would be good to check if it's possible to parse C++ using Python, if it's > possible it will ease a lot your work and you can forgot about the > ApiExtractor modification. OK. > I think you need to send this proposal via google-melange, then you can > improve it there. Sure, I will submit my proposal ASAP. I paste my proposal here rather than google-melange, because I'm afraid that mentors may still not see students' proposals before the application deadline while maybe I'm wrong. From schampailler at skynet.be Sat Mar 31 17:41:00 2012 From: schampailler at skynet.be (Stefan Champailler) Date: Sat, 31 Mar 2012 17:41:00 +0200 Subject: [PySide] Proposal for "Visual tool to fast edit type system files" -- a Pyside idea In-Reply-To: References: Message-ID: <4F77258C.2050003@skynet.be> This in an interesting proposal. But, for having had to have a look inside Shiboken and all that stuff, I'd say what's missing most is documentation. There's a lot of it, but a walkthrough, from beginning to end would be very nice. For example, the ownership stuff is not crystal clear, there are a lot of corner cases which, I felt, are not documented enough for people like me who : - are pretty fluent in programming Python and C++ - but are not familiar with the way the bindings work between Python and C++ I know documentation is not the funniest part but that would really help people like me, outside the PySide core team, to be able to be autonomous without having to work/read for days before understanding a bit of what is happening... I hope I don't sound too harsh, that's not my point, I use PySide a lot now and as a product it is really good and reliable. It's just that when I meet a bug it's very hard to dive into the Shiboken code and get a grasp of what's happening. Especially when one knows that it is not that complicated... So the tool would be nice, but it has to make the inner workings of the binding tools very clear. So I see that tool has something to help me to more easily understand and fix PySide. (and again I love PySide :) I'm very grateful of having such a tool for free) Stefan On 03/27/2012 05:24 PM, higery wrote: > The following is my GSOC2012 proposal although in a very preliminary way, I > will > appreciate anyone can make some suggestions about this proposal. >