[20:39:28] Qt meeting @ 1800 UTC today, KDE meeting @ 1900 [20:40:06] *** nirbheek changes topic to 'Gentoo meetings | April 15th: Qt meeting @ 1800 UTC, KDE meeting @ 1900 UTC' [20:40:26] * nirbheek wonders if gnome should have a meeting sometime soon [20:40:58] tnx [20:41:05] yw [20:42:31] *** Joins: ayoy (~ayoy@244-133.lodz.mm.pl) [20:42:31] *** Quits: ayoy (~ayoy@244-133.lodz.mm.pl) (Changing host) [20:42:31] *** Joins: ayoy (~ayoy@gentoo/developer/ayoy) [20:47:02] *** Joins: spatz (~spatz@gentoo/developer/spatz) [20:58:14] * hwoarang is going to make a cup of hot hot hot coffee [20:59:37] * yngwin is drinking one already :) [21:01:57] looks like a short one today [21:02:56] i doubt :) [21:03:29] aaaaaaaand ..... [21:03:33] here we go :D [21:03:42] helow [21:03:54] indeed [21:04:02] !herd qt [21:04:03] (qt) abcd, ayoy, hwoarang, spatz, ssuominen, tampakrap, wired, yngwin [21:04:10] OHAI [21:04:34] hi guys [21:04:50] ok, meeting is starting, agenda is at http://gitorious.org/gentoo-qt/pages/Meeting20100415 [21:04:51] elo [21:05:00] rollcall" is everyone here? [21:05:13] here [21:05:27] * pesa here [21:05:31] * yngwin here [21:05:38] * hwoarang here [21:05:42] * ayoy here [21:05:51] abcd absent [21:05:57] waiting for wired and tampakrap [21:06:03] anyone else? [21:06:04] and ssuominen [21:06:18] * nirbheek is absent [21:06:21] ah i wasnt aware that he was in qt project :) [21:07:24] yes he joined a little while ago, as he was helping so much anyway (with qt3 removal for example) [21:09:19] well, let's start [21:09:32] only half the people are here :/ [21:09:32] 1. migrating ebuilds to qt4-r2.eclass [21:09:49] yes, apparently they couldnt make it in time [21:09:56] lets move on [21:10:27] anyway, any ideas on tackling the ebuilds with old qt4.eclass [21:10:51] bug 311481 [21:10:53] hwoarang: https://bugs.gentoo.org/311481 "[tracker] Migrate ebuilds from qt4.eclass to qt4-r2.eclass usage"; Gentoo Linux, Ebuilds; NEW; yngwin@g.o:qt@g.o [21:11:00] http://dev.gentoo.org/~wired/checks/qt4.eclass.html [21:11:27] mmmm [21:11:36] many bugs need filing [21:11:39] how should we deal with this [21:11:40] yeah [21:11:50] what i started to do is work on some ebuild and ask the maintainer for ok to commit or otherwise submitting a patch [21:11:54] why? cant we do it silently? [21:12:08] it depends on maintainers' attitude I guess [21:12:30] you need to ask maintainers, as it often is also migrating to eapi2/3 [21:12:31] errr opening like 80 bugs wont work [21:12:31] cause we don't have policy on touching others' ebuilds, right? [21:12:32] like ABCD does with all his prefix stuff [21:12:49] he does ask, usually [21:12:59] sure, that's what I meant [21:13:12] ayoy: right, council rejected the proposal [21:13:19] what about an email on -dev [21:13:26] what about filing bugs with readymade patches? [21:13:29] asking the to check the list, fix the ebuild withing the next 60days [21:13:50] *them [21:14:02] I like the first approach better [21:14:08] *** Joins: lazevedo (~lucasazev@200.129.46.206) [21:14:19] the first is easier for us, of course :) [21:14:23] yes but how are we gonna prepare 80 patches? [21:14:28] you can do both [21:14:31] aaa, this "first" :P [21:14:41] ask on dev ml for maintainers to look at this [21:15:08] then start doing patches yourself when maintainers take too long [21:15:14] I think that maintainers not necessarily have to realize the changes between qt4.eclass and qt4-r2 [21:15:14] honestly I want to let the maintainers do all the work [21:15:23] has the qt meeting started? :) [21:15:27] yes [21:15:30] so it will take less time for us [21:15:31] hwoarang: that will take years [21:15:33] give them a month, then start fixing yourself [21:15:42] bonsaikitten++ [21:15:47] what bonsaikitten and yngwin said ^^ [21:15:53] 60 days. Then we touch them [21:15:55] you'd need bugs for that, imo [21:15:56] like sping did with migrating packages to dev-vcs [21:15:57] might not be the popular method, but that's how you fix things [21:16:44] i find that often it is ok if you ping maintainers on irc, especially if you can show a patch [21:17:05] yeah, I like the idea of patching by us [21:17:12] if we care about this issue, we;re gonna fix it [21:17:19] if they dont react, you can file a bug, with the notice "to be applied in 7 days unless you object" [21:17:25] maintainers don't have to care about this thing as much as we do :) [21:17:42] +1 [21:17:50] "this thing" being migrating qt4->qt4-r2 [21:17:55] pretty much what was first suggested, so ok [21:18:28] ok, all agreed? [21:18:53] ok [21:18:58] pretty much yes, but can you sum it up? :) [21:20:04] *** Joins: joost_op (~joost@gentoo/contributor/joost-op) [21:20:21] 1) send dev ml a notice about the migration, asking maintainers to cooperate, and 2) start making patches, ping maintainers on irc for ok, if no reaction file bug that you will apply it in 7 days unless there are objections [21:20:40] cool [21:20:45] fine [21:21:00] yngwin, Can I read the logs after this somewhere, I have to leave [21:21:20] yes, we put up logs of every meeting [21:21:25] thanks [21:21:27] cheers all [21:21:32] next subject? [21:21:36] 2. meego ebuilds and status update [21:21:57] I have one small info here [21:21:58] a start has been made on meego packages [21:22:05] yes? [21:22:07] *** Joins: Thev00d00 (~v00d00@unaffiliated/thev00d00) [21:22:14] libdui, that I've recently added [21:22:20] has ben renamed as libmeegotuoch [21:22:28] however it's not yet tagged [21:22:33] *libmeegotouch [21:22:39] ok [21:22:41] so I'm waiting with changes to ebuilds [21:23:01] meego has recently made the repos public [21:23:03] i've also invited some people from sabayon who are interested in packaging these [21:23:17] i've seen more recent tags in libdui git however [21:23:33] and also in other packages [21:23:33] pesa: true, but renaming occured on moday iirc [21:23:42] ah i see [21:23:52] anyway, we're after the API/ABI unfreeze period [21:23:57] and this stuff is mostly broken [21:23:58] :) [21:24:14] there's huge amount of changes between 0.19 and 0.20 [21:24:20] so i take it you guys will keep an eye on this and cooperate on ebuilds? [21:24:37] sure, I guess so [21:24:47] http://meego.gitorious.org/ [21:25:03] there are many projects there. Do you know which of them are useful to us? [21:25:20] not really actually [21:25:22] cause I dont [21:25:25] *** Joins: pontecorvo (~solshark@93-183-241-28-dynamic.retail.datagroup.ua) [21:25:39] I have to take a lok at it [21:25:58] since at least libdui is still at qt.gitorious.org [21:26:00] * hwoarang is confused with all the meego thingie [21:26:12] hwoarang: Nokia is confused about this, believe me :) [21:26:29] it may also be a good idea to follow discussions in #meego, the meego MLs and forums [21:26:38] i am following -dev [21:26:45] but there is too much info there [21:26:47] hi all! [21:26:55] see my note on agenda [21:27:15] dividing qt to work groups [21:27:35] i was planning to look into taking part in meego itself, so i may be more informed about this in the future :) [21:27:38] hi alexxy [21:27:44] i cant follow all the changes in Qt libs + meego + qt apps. I was wondering if it would make sense to create some workgroups [21:27:59] hwoarang: what is it about? to define responsibility areas across qt ebuilds? [21:28:00] i think it is very early days still [21:28:19] ayoy: some of use to focus on Qt libs, others on meego etc [21:28:26] sure [21:28:32] cause it is somehow impossible to work on all those things [21:28:51] those groups will be flexible ofc [21:28:55] we already work on different things [21:29:02] do we? [21:29:09] until recently i never really touched the live ebuilds [21:29:33] i tend to maintain certain apps, you others [21:30:03] well [21:30:08] i'm not sure if it is needed to formalize this in workgroups/subprojects [21:30:26] imo, no need for more bureaucracy, people just work on what they like [21:30:29] ok so who is working on meego? just ayoy atm ? [21:30:39] it seems so [21:30:46] i would like to work on meego, too [21:30:58] but well, I didn't look at meego.gitorious.org at all :P [21:31:00] though i don't have experience on packaging [21:31:04] but yeah, I'm interested [21:31:08] but i do think it is good to document who is working on what in the wiki or somewhere [21:31:18] like we have done with live ebuilds [21:31:24] yngwin: yes [21:31:35] i think it is good to know who is focusing on what [21:31:41] * yngwin agrees [21:31:46] wiki page? [21:31:46] so we know who to ask about specific bugs [21:31:51] wiki on gitorious [21:31:58] also to make sure everything is covered [21:32:12] not like some herds which go inactive and nobody notices [21:32:20] ofc [21:32:48] In my mind, we have to major areas ( Qt libs + Megoo ) + 1 common ( qt4 apps ) [21:32:52] ayoy: like http://gitorious.org/gentoo-qt/pages/Qt4%20live%20ebuilds [21:32:56] *s/to/two [21:34:19] hwoarang: qt4-apps is quite huge [21:34:28] lazevedo: you're welcome to help, there are enough people here who can answer any questions you might have [21:34:34] there's also PyQt and pyside [21:34:47] yep [21:34:55] pyside with its deps is lots of work sometimes [21:34:56] thanks yngwin [21:34:58] ayoy: regarding qt4-apps each one of us maintains certain packages. So we can leave this aside [21:35:05] lazevedo: we can talk about meego later [21:35:07] if you like [21:35:11] ayoy: sure [21:35:12] hwoarang: sure [21:35:17] i have no idea about pyside [21:35:25] hwoarang: I have an idea :P [21:35:32] lazevedo: you're always welcome in #gentoo-qt [21:35:46] PyQt4+sip are in a good shape. Snapshots + tree packages are working fine [21:35:47] yngwin: ty :) [21:36:00] python herd touches tree packages as well [21:36:41] i wouldnt worry about the so much. They dont require much hardware resources or time to maintain them :) [21:36:46] *them [21:36:53] anything else we need to decide about meego ebuilds now? [21:37:01] * hwoarang stupid lag with woodpecker [21:37:15] yngwin: I guess not. [21:37:16] I don't think so [21:37:27] we can pacakge everything :) [21:37:38] ayoy: pleae send us an email on qt@ when you have meego on working state [21:37:44] hwoarang: sure [21:37:56] ok, so to summarize: we'll look into what is interesting for us to package, and we'll document who is working on what [21:38:08] right [21:38:22] sounds good [21:39:02] alright? any other comments on this subject? [21:39:37] i guess not [21:39:38] well, i have never packaged anything, so i don't think i can add anything for now :b [21:39:47] then 3. call for candidates for project lead (elections next meeting) [21:39:52] lazevedo: we'll teach you if you want [21:39:56] *** reavertm_ is now known as reavertm [21:40:04] lazevedo: http://devmanual.gentoo.org [21:40:06] good place to start [21:40:08] :) [21:40:09] ayoy: sure [21:40:15] ty hwoarang [21:40:24] yngwin: ok elections :) [21:40:26] i'll bookmark it and read it when i get home [21:40:27] please do that after the meeting [21:40:28] next month it will be one year since we had our last "elections", so it is time again [21:40:39] ok cool [21:41:11] yngwin: what is your thoughts about this [21:41:16] *s/is/are [21:41:29] !devaway yngwin [21:41:29] hwoarang: yngwin: minimally active, reconsidering options @ 2010/04/13 13:53Z [21:41:31] so anyone who wants to volunteer for that job, can be a candidate and next meeting the qt project members will vote [21:41:49] unless you want to organize it in a different way [21:41:56] devaway? why? [21:42:23] to reconsider my options before i make the big break [21:42:32] big break? [21:42:42] i will be resigning from gentoo [21:42:53] huh? when? [21:42:56] :O [21:43:03] yngwin: i am only interesting in lead this project just in case you end up inactive because of all the China thing or your retirement [21:43:03] ... [21:43:09] within the next few days [21:43:21] you've already made up your mind? [21:43:34] i really really hope this is a bad joke though [21:43:35] is gentoo forbidden in China? hmm [21:43:46] reavertm: busy RL [21:43:50] i basically decided that early this week, i just thought to take some days to think things over [21:43:52] reavertm, why would it be? [21:44:01] reavertm, our best soc application is from a chinese undergrad [21:44:22] yngwin: you've got to be kidding me. please reconsider [21:44:32] tough shit [21:44:51] but, life has to go on, I understand [21:44:56] spatz: so far i have seen no signs of change [21:45:33] anyway, i'm still around, so we can discuss it after the meeting [21:45:45] yngwin, the what [21:45:51] * nirbheek just glanced at the backlog [21:45:56] but as things stand now, you will need a new project lead [21:46:10] yngwin, I have a bit of friendly advice [21:46:10] nirbheek: ? [21:46:34] ok, we'll discuss that later. you won't run for lead anyway? [21:46:42] yngwin, take a break for a few weeks [21:46:57] spatz: at this moment i am not a candidate [21:47:00] yngwin, or more, if you haven't already [21:47:39] nirbheek: we can discuss this later, we need to move on with this meeting [21:47:53] yngwin, oh, sorry, I thought the meeting was over :) [21:47:54] so we need candidates. anyone stepping up? [21:48:06] I am [21:48:14] great [21:48:16] :) [21:48:26] * hwoarang will stay around after the meeting to make yngwin stay [21:48:34] hehe [21:48:40] so let's move on to the next subject [21:48:44] ok [21:49:08] i would suggest to leave some time for others to step up if they want, as some people are not here [21:49:17] ofc [21:49:18] they have all month :) [21:49:23] ok [21:49:33] 4 Bugs [21:49:33] shall we move to the next topic then? [21:49:46] bug 311481 [21:49:48] Forget about the first bullet as we discussed it earlier [21:49:50] https://bugs.gentoo.org/311481 "[tracker] Migrate ebuilds from qt4.eclass to qt4-r2.eclass usage"; Gentoo Linux, Ebuilds; NEW; yngwin@g.o:qt@g.o [21:49:53] ok [21:49:59] bug 312689 [21:50:03] yngwin: https://bugs.gentoo.org/312689 "x11-libs/qt-core-4.6.2-r1 forces additional CFLAGS and CXXFLAGS for dev-python/PyQt4-4.7.2"; Gentoo Linux, Development; NEW; Martin.vGagern@gmx.net:qt@g.o [21:50:19] pesa: ^ [21:50:36] isnt this a PyQt4 problem? [21:50:56] it should be solved by ayoy's work on mkspecs i think [21:51:15] oh [21:51:29] pesa: could you please test with qting-edge branch? [21:51:37] if you use qt-4.6.2 [21:51:45] i do [21:51:46] did somebody test the new eclass? I am waiting ayoy to port the changes on -edge eclass [21:51:50] anyway, I'm in progress of porting my changes ti qt4-build-edge [21:51:53] ok [21:52:06] yeah, I was about to do that yesterday but failed due to lack of time :/ [21:52:15] so this is being worked on [21:52:16] no worries [21:52:29] anyway i think there's a real bug in sip-generated Makefiles [21:53:03] because user-specified flags should override mkspecs ones [21:53:21] indeed [21:53:24] should we contact phil? [21:53:24] currently it's the opposite [21:53:36] yes [21:53:40] via direct email or through the ML [21:53:58] the ml should be fine [21:54:00] any volunteers? [21:54:03] i'd like to have his opinion [21:54:22] i can do that [21:54:28] good [21:54:30] thankx :) [21:54:39] np [21:54:51] we do have some security bugs [21:55:36] we do? [21:55:55] yes [21:55:59] webkit and arora [21:56:01] yngwin fixed the webkit one [21:56:16] from what i understand that is a false one [21:56:25] the arora one [21:56:35] so do i [21:56:47] and webkit should be fixed now [21:57:00] we need to call arches to stabilize qt-webkit [21:57:40] yes, and remove the old ebuild revisions [21:58:22] so who wants to take care of that? [21:58:28] yngwin: can you do it since you took care of it from the very beginning? [21:59:10] ah yes, i got sidetracked [21:59:35] i will file the bug [21:59:39] thx [21:59:54] bug 304115 [21:59:57] pesa: ^ [21:59:57] hwoarang: https://bugs.gentoo.org/304115 "=dev-python/PyQt4-4.7 with USE="-X" tries to link with -lXext"; Gentoo Linux, Applications; NEW; orzel@freehackers.org:qt@g.o [22:00:02] yes? [22:00:06] do we need to patch sip or just wait for next release [22:00:08] it's fixed upstream [22:00:22] i dont think it's that urgent [22:00:33] ok fine [22:00:37] what do you think? [22:00:54] yeah me neitehr [22:00:57] we only had one report in many months [22:01:22] One last thing. Should we get rid of 4.6.1 ( and maybe 4.5.3 ) ? [22:01:32] kde meeting in 3 mins, we need to wrap up [22:01:33] what about bug #307861 ? [22:01:37] pesa: https://bugs.gentoo.org/307861 "x11-libs/qt-webkit-4.6.2: ld crashes at linking libQtWebKit.so.4.6.2"; Gentoo Linux, Library; NEW; urcindalo@gmail.com:qt@g.o [22:02:13] hwoarang: there is still a regression in 4.6.2, so you want to keep 4.6.1, at least till 4.6.3 is released [22:02:24] ok [22:02:34] pesa: i have no idea wrt that bug [22:02:43] altho, i didnt patch webkit for 4.6.1 [22:02:54] so it would need to be masked [22:03:33] the whole 4.6.1 set? [22:03:54] all or nothing [22:04:04] pesa: see comment #4 there [22:04:57] yeah, it seems related to -ggdb [22:05:02] i'll leave it up to you if you want to do more about that issue [22:05:13] fine [22:05:28] question is: should we take some action or just ignore the problem? [22:05:40] what can we do [22:05:56] investigate if there's a memleak in ld [22:06:12] or just add a warning if -ggdb is in CXXFLAGS [22:06:29] hwoarang: btw, I pushed changes to qt4-build-edge.eclass [22:06:36] please test if you can [22:06:40] you could ask someone from toolchain to take a look? [22:06:47] or close as worksforme/invalid/whatever [22:07:08] i dont think its invalid [22:07:12] yngwin: yes, that would be an option [22:07:19] yngwin: thx [22:07:22] sorry [22:07:24] ayoy: thx [22:07:24] :P [22:07:27] :) [22:07:29] ok [22:07:32] is toolchain responsive enough? [22:07:33] we need to wrap up [22:07:38] FWIW, qt-webkit4.6.2 compiles fine for me with -ggdb [22:07:40] pesa: yes [22:07:48] nice [22:07:54] anything else that cant wait? [22:07:58] nope [22:08:13] ok, thank you all [22:08:18] thanks [22:08:18] ===========================================