[01:50] <tramm> reported my problem as https://bugs.launchpad.net/unity/+bug/904040
[01:50] <ubot5> Ubuntu bug 904040 in Screenlets "Changing screen geometry in the middle of startup messes up positions of autostart application windows" [High,Confirmed]
[09:13] <kamstrup> Kaleo: good morning dude - any way I can get you to check https://code.launchpad.net/~kamstrup/bamf-qt/fixes-for-jhbuild/+merge/85614 soonish?
[09:13] <kamstrup> if I get too many branches floating out there trying to get u2d build under jhbuild I'llgo insane :-)
=== mpt_ is now known as mpt
[11:22] <Kaleo> kamstrup: sorry for the delay, let's have a quick look
[11:23] <kamstrup> Kaleo: great, np
[11:24] <Kaleo> kamstrup: hmmm it looks like removing a feature that oSoMoN added
[11:24] <Kaleo> oSoMoN: do you have an opinion on https://code.launchpad.net/~kamstrup/bamf-qt/fixes-for-jhbuild/+merge/85614 ?
[11:25] <oSoMoN> Kaleo, kamstrup: yeah, it undoes something that was contributed last week, so it cannot be merged as is
[11:26] <Kaleo> kamstrup: that stuff breaks in the jhbuild?
[11:26] <oSoMoN> kamstrup: see the rationale at https://code.launchpad.net/~ballogy/bamf-qt/fix-imports-dir-location/+merge/83248
[11:29]  * kamstrup looking
[11:32] <kamstrup> it gives some weird errors... CMakeCache.txt lists the QT_IMPORTS_DIR variable as uninitialized even though it's set to /home/kamstrup/unity/install/lib/qt4/importds
[11:32] <kamstrup> imports
[11:32] <kamstrup> and it tries to install into /usr, which of course is not the game jhbuild plays :-)
[11:34] <kamstrup> it is also not how dee-qt does it, so I'd assume dee-qt needs the same patching to make Arch devs happy?
[11:34] <kamstrup> oSoMoN, Kaleo: ?^^
[11:35] <oSoMoN> kamstrup: yeah, the arch contributor is working on it afaik
[11:35] <oSoMoN> kamstrup: though I agree that ideally the destination dir should be a relative path to make jhbuild (and possibly other build systems) happy, not an absolute path
[11:36] <oSoMoN> kamstrup: maybe we can trim the leading part of QT_IMPORTS_DIR to make it relative?
[11:39] <kamstrup> oSoMoN: it looks like we're simply overriding QT_INSTALL_IMPORTS if it's set on the command line... maybe that's the real issue
[11:39] <kamstrup> I don't have problems setting -DQT_INSTALL_IMPORTS=foobar on the cmake command line
[11:48] <oSoMoN> kamstrup: I’m not sure what a good solution would be, but in any case we must make sure that it doesn’t break the build on arch, otherwise that would make a contributor unhappy
[11:48] <kamstrup> indeed
=== MacSlow is now known as MacSlow|lunch
[12:18] <kamstrup> oSoMoN, Kaleo: pushed a new patch for https://code.launchpad.net/~kamstrup/bamf-qt/fixes-for-jhbuild/+merge/85614 that should respect the old behaviour
[12:18] <Kaleo> kamstrup: makes a lot of sense
[12:21] <oSoMoN> kamstrup: the comment should be updated though, it is misleading
[12:22] <kamstrup> oSoMoN: probably :-) I did not understand the problem it described just assumed that cmake/qt gurus did :-)
[12:22] <kamstrup> so i'll need a ghost writer :-)
[12:22] <oSoMoN> kamstrup: there’s a bug report that describes the issue
[12:23] <oSoMoN> kamstrup: https://bugs.launchpad.net/ubuntu/+source/cmake/+bug/894805
[12:23] <ubot5> Ubuntu bug 894805 in cmake (Ubuntu Oneiric) "QT_IMPORTS_DIR is not defined when no QML plugins are installed" [Medium,New]
[12:24] <oSoMoN> kamstrup: also, a comment to explain that the imports dir can be overridden by setting -DQT_IMPORTS_DIR on the cmake command line would be welcome
[12:25] <kamstrup> oSoMoN: k, i'll put a link to the bug in there
[12:26] <oSoMoN> thinking about it, if it can be overridden on the command line, it kind of renders the other code path useless, unless the arch packager can’t override the parameters passed to cmake?
[12:30] <kamstrup> well, I think it makes sense to try and do the right thing by default
[12:30] <kamstrup> it just so happens that jhbuild is kinda non-standard in the installation paths
=== MacSlow|lunch is now known as MacSlow
[13:31] <andyrock> hi all
=== om26er_ is now known as om26er
=== yofel_ is now known as yofel
[14:42] <kamstrup> kenvandine: hi - there'll be a slight break in the DeeFilterModel API in the upcoming dee-1.0
=== mterry is now known as mterry_sprinting
[14:43] <kamstrup> needed to make GI work properly, and mem management not very weird :-)
[14:43] <kamstrup> but only very slight
[14:43] <kamstrup> good news is that it'll work in Python as well
[14:43] <kenvandine> kamstrup, thx for the heads up :)
[14:43] <kenvandine> that is good
[14:43] <kamstrup> and hopefully adding some minor optimizations
[14:44] <kamstrup> sure, just wanted to warn you before you go "wtf!" :-)
[14:44] <kenvandine> kamstrup, :)
[14:45] <kenvandine> kamstrup, by the way, sorting got faster with the new glib :)
[14:45] <kenvandine> ~30% improvement
[14:45] <kamstrup> wow!
[14:45] <kamstrup> I wonder what happened :-)
[14:46] <kenvandine> i think they removed some hacks related to locale
=== mterry_ is now known as mterry_sprinting
=== lamal666 is now known as lamalex
[18:10] <itism3> hello
[18:13] <itism3> i just installed apache php and mysql on ubuntu 11.04 after i restart i got ubuntu-classic instread of unity. what happen?
[18:13] <drunk_rambler> needed help on getting started with papercuts projects
=== seiflotfy is now known as seifstrup
=== seifstrup is now known as seif
=== bil21al_ is now known as bil21al
[19:35] <htorque> unity build fails: http://paste.ubuntu.com/770413/ → am i doing something wrong? if i replace 'g_thread_init' with 'g_type_init' it seems to work fine.