[00:02] I can somewhat reliably cause unity-window-decorator, which apparently is part of compiz, to crash when running the standhal mmo game. Shouldn't it crashing cause... shoot... what was the little bug reporting app called? to start? [00:04] psusi: apport? [00:06] that's it [00:07] unity-window-decorator makes a few complaints about G_IS_OBJECT() failing, and invalid unclassed pointer in cast to GObject, then about 50% of the time it segfaults... shouldn't that trigger apport? [00:11] psusi: is apport enabled? [00:12] I didn't do anything to disable it... [00:12] psusi: well, it's off in the stable release, and idk if it's on in thedev release yet officially [00:12] ohh, it's stop/waiting [00:12] ohh... I thought it was always there [00:14] how do I start it? sudo start apport says start:job failed to start [00:14] psusi: sudo service apport start [00:15] same thing [00:15] psusi: sudo service apport start force_start=1 [00:15] there we go [00:15] psusi: you can enable in /etc/default/apport if you want it on all the time [00:18] there we go... [00:18] now let's see what the backtrace says... [00:29] hi robert_ancell, how are you? [00:29] micahg, hello [00:30] robert_ancell: I take it you won't be planning on touching webkit this cycle? [00:30] micahg, just normal updates if required... [00:31] micahg, do you know if they have any plans for this cycle? [00:31] robert_ancell: no, haven't seen anything yet [00:31] 1.4.0 is out [00:31] yay! [00:31] is it worth sruing? [00:32] robert_ancell: idk, I haven't checked the diff yet, if it has security fixes, yes, otherwise, we'll wait for the next point release [00:33] I was going to take a look next week [00:35] robert_ancell: actually, if there's an ABI break (which hopefully there isn't), we would go through -proposed first most probably [00:35] All SRUs go though proposed don't they? Or can security fixes bypass that [00:36] robert_ancell: security fixes normally bypass -proposed (except for chromium), but I might start using it more [00:36] but we always build in the security PPA with only -security and the release pocket [00:42] robert_ancell: BTW, tried lightdm for the first time today, worked well, only thing I didn't like was the user list, but I guess there will be a preference for that (and gdm has that by default as well) [00:42] micahg, yeah, the greeter is just an example at the moment, there will be a proper unity designed one, hopefully some work started for A2 [00:43] robert_ancell: I'm using it for xubuntu ATM actually :), we'll probably have our own greeter as well [00:44] oh nice, please let me know how you guys go with the API [00:44] robert_ancell: k, will do [00:45] hey robert_ancell [00:45] cyphermox, hello [00:59] micahg, i've got firefox moving system-installed extensions in to the users profile now :) [00:59] but i had to add support in ubufox for it [01:00] chrisccoulson: k, we have to update that anyways, so I'm fine with it [01:00] chrisccoulson: awesome though, thanks === asac_ is now known as asac [05:45] LibreOffice build = BEAST! [05:47] +/c [06:23] Good morning [06:25] Hey pitti. [06:52] Hey pitti! If I wanted to drop libgl1-mesa-dri-experimental (as it contains only the nouveau 3D driver, which we're planning to ship by default), what does jockey need to be told about? [06:53] RAOF: you mean "install by default"? [06:53] we need to drop jockey's handler for it [06:54] pitti: I mean “drop the -experimental package, and ship nouveau with the regular -dri package”. [06:54] RAOF: Is it really no longer experimental? Or are we just taking a big gamble? [06:54] BTW, both you pitti and you RAOF expired from ubuntu-sponsors... [06:54] TheMuso: Bug reports are no longer considered worthless :) [06:54] RAOF: ah. [06:54] TheMuso: yeah, I just asked in #u-devel, very weird [06:54] there was no advance warning [06:55] Yeah I know. [06:55] I'm fairly sure I recently *renewed* my ubuntu-sponsors membership? [06:55] Thats what I thought. [07:04] good morning [07:07] RAOF: so, want me to drop the handler then? [07:07] hey didrocks [07:07] pitti: guten morgen! Thanks for the approval :) [07:08] I'll upload pkgbinarymangler now in oneiric, I'll need it for unity-2d [07:08] * pitti dies a little inside when looking at http://people.canonical.com/~platform/workitems/oneiric/canonical-desktop-team.html [07:09] didrocks: oh, sorry; I ran the test suite last night and forgot to dput, hang on [07:09] pitti: Yes, please. [07:09] didrocks: (it's already tagged in bzr0 [07:09] didrocks: uploaded, sorry [07:09] nice, version 99 [07:09] what will version 100 bring? [07:10] Is that *480* work items for oneiric? [07:10] pitti: heh, who knows! [07:10] pitti: thanks a lot :-) [07:10] * Strip out the biggest 1/3 of files to save CD space [07:11] Take that, vmlinuz! [07:11] after the oom killer, the oos deleter! :-) [07:11] RAOF: yes :-( [07:11] for the record, in natty we had 300 [07:11] so, everyone cut a spec [07:11] pitti: We've trippled the size of the Desktop team, right? :) [07:12] the sad thing is that the really fat ones are pretty much must-have (firefox maintenance, gnome 3 migration, cd space) [07:13] can I cute the "Qt by default?" then, the size memory, langpack and everything is fixed. Magic \o/ [07:13] we can cut thunderbird, quickly, lightdm, etc. but it starts getting painful there [07:14] the first one who proposed oneconf to be cut… It's too guilty of out of time for the last couple of cycles and I think it's the right time to get it by default (without desktopcouch) :-) [07:17] wow - http://cdimage.ubuntu.com/daily-live/20110525/ [07:17] and it's not oversized, how come? [07:18] I haven't pulled llvm on there yet :) [07:18] yeah, but with all the new gtk3/gnome3 packages and autosyncs I had expected some fallout there [07:19] presumably we lost a couple of old gnome 2 libs [07:19] where "lost" == "got rid of" [07:19] hello people! can i ask questions about gnome shell here? [07:20] the CD failed tonight? [07:20] just ask, and someone might answer [07:20] didrocks: it's not cron'ed yet; cjwatson built it manually [07:20] pitti: oh ok :) [07:24] ok so i installed ubuntu mini iso for natty and installed gnome shell from the official gnome 3 ppa (i didnt break unity) and i got a successful gnome shell build but === tkamppeter_ is now known as tkamppeter [07:24] the network manager looks and behaves awful [07:25] it does not integrate well with the gnome shell [07:26] is there a solution to integrate it well in the gnome shell panel [07:26] Debashish: people working on this ppa aren't around right now. You can maybe ping them a little bit later today? [07:26] ricotz: ^ [07:27] ok thanks i will check it out later [07:27] yw :) [07:27] Debashish, you are missing network-manager 0.9 which isnt available in the ppa yet [07:27] hmm [07:27] so where can i get the source? [07:28] Debashish, it is in my testing ppa, but add this isnt officially suggested [07:28] i am willing to check it for my personal use, so can i get it? [07:29] Debashish, ppa:ricotz/testing, if you got problems there should be several postings which can help you [07:29] ok thanks [07:31] btw the sound applet in the panel also crashes sometimes while changing the backends so what can be done? [07:33] Debashish, without an error message it is hard to tell, but i havent seens such thing, so it might be a pulseaudio problem [07:34] it does not give any error message and the window just vanishes, so is it crashing or something else [07:35] can i not switch to alsa on gnome shell? [07:36] Debashish, you mean the gnome-control-center crashes? [07:39] no but sort of because the sound preferences item is under gnome-control-center and sound preferences from the sound menu in the panel opens that and that is the thing that crashes/vanishes but other items under gnome control center work flawlessly [07:41] Debashish, so when you choose/click on the "sound-icon" it immediately crashes? [07:41] no but when i change backends [07:42] ...after the window is already opened [07:42] Debashish, ok, sounds like a pulseaudio/libcanberra problem [07:43] Debashish, what is the output if you run gnome-control-center from a terminal? [07:43] pitti: Good morning! I have yet another SRU proposal: https://code.launchpad.net/~gunnarhj/ubuntu/natty/language-selector/natty-proposed-imfix/+merge/62416 [07:43] pitti: Noticed that you have set status "Won't Fix" on bug 663776, but considering the discussion on the bug report I ask you to reconsider. [07:43] Launchpad bug 663776 in unity "[dash] Search field in Unity can not support iBus" [High,In progress] https://launchpad.net/bugs/663776 [07:44] hmm not yet checked and now i am running from windows so i will let you know sometime later [07:44] hey GunnarHj [07:45] GunnarHj: oh, that was mostly bug cleanup, I thought for natty the im-switch SRU would suffice; but apparently that causes too much unexpected behaviour, so I guess we shouldn't release that; WDYT? [07:45] GunnarHj: the language-selector fix about auto-updating the input method combobox and settnig the value from the language indeed seems appropriate to me [07:45] Debashish, ok, try to report a bug, against the ppa and check if you have libcanberra-gtk3-modules installed [07:45] GunnarHj: I've seen your MP, I'll upload it asap [07:47] ok sure will. thanks ricotz for the replies and that's all. [07:47] GunnarHj: i. e. for the l-s fix, people will get the right thing when they use it; for im-switch, it changes unexpectedly, i. e. people who use English in China would suddenly lose their ibus configuration, won't they? [07:48] Debashish, np, thanks [07:49] btw ricotz can i not switch to alsa and remove pulseaudio or would that break the system? [07:50] Debashish: pulse uses alsa [07:51] Debashish: why do you want to "switch off" pulseaudio? it's by and large the standard desktop audio API these days [07:51] oh ok my bad :) [07:51] does it cause disturbance etc.? [07:51] because it is a bit buggy? [07:52] Debashish: you can try to switch to direct alsa in gstreamer-properties and see if it makes any difference [07:52] Debashish: right, I'm interesting what particular bugs you have [07:52] "interested" [07:52] pitti: I too thought that the im-switch SRU would be enough. OTOH, not releasing it wouldn't be good either, since there are other effects from the Natty i18n changes. I believe that if both im-switch and l-s are SRUed, there will be no significant adverse effects left. As regards people loosing their ibus config if they use English, please see comment #60 on the bug report. [07:52] yeah i would like to do that [07:53] GunnarHj: I was concerned about comment 59; Tualatrix certainly knows how to get it back, but it's still a major behaviour change in an SRU [07:54] GunnarHj: I agree that it is a correct fix, and we should have it in oneiric, but for SRUs I'm very hesitant [07:57] pitti, just a couple:1 network manager integration in the panel, 1.1 sometimes its icon doesnt show after login and if it shows it doesn't show the monochrome one, 2 sound preference window crash/vanishing without a message while changing backends [07:57] but now i gotta go because the laptop's battery is low and i have no power source. [07:58] Debashish: neither of these seem pulse related [07:58] meh [08:07] good morning everyone [08:08] hey chrisccoulson [08:08] how are you? [08:08] hi didrocks, i'm good thanks, although a bit tired. how are you? [08:08] Morning [08:08] RAOF, you there? [08:08] hi czajkowski [08:09] chrisccoulson: I'm very good, thanks! :-) [08:09] has anyone every heard of a random reboot for no reason on natty. My machine keeps going to reboot while I'm on irc or doing things [08:10] it goes into the boot up screen where it say battery ok etc then goes to login prompt and then back to normal [08:10] sounds like X is crashing ;) [08:11] yeah I don't know how to reproduce or report the bug [08:11] as it doesnt happen every day but is happen more and more ove the last 2 weeks [08:17] pitti: I understand. The behavior change is basically: Previously you enabled ibus from l-s for the LANG locale (2:nd tab). (People probably have thought that they enabled it irrespective of other settings.) Now you control ibus from l-s for the current language. [08:17] pitti: If you e.g. have zh_TW as the first language, "ibus" is default, which is shown on the IM control. If you change it in l-s to e.g. English, you instantly see that the text on the ibus control changes to "none" (the default for English), which should imply that if you want to use English with IM enabled, you need to change it to "ibus". [08:17] pitti: One of my thoughts about justifying this as an SRU is that considering the Unity+ibus problem (not yet solved), this is a tiny detail... [08:17] pitti: Think that's all I can say on this topic. :) [08:22] chrisccoulson, hello, could you take a look at https://bugs.launchpad.net/ubuntu/+source/gjs/+bug/788500 [08:22] Launchpad bug 788500 in gjs "Use libmoz185 instead of xulrunner-2.0" [Undecided,New] [08:24] ricotz, sure [08:24] chrisccoulson, thanks [08:26] ricotz, you tested it and it builds ok? [08:26] it looks fine, so i'll just sponsor it [08:27] chrisccoulson, yes i am running it [08:27] excellent, thanks [08:27] chrisccoulson, ty [08:29] chrisccoulson: any idea on how I report that kinda bug? [08:29] czajkowski, there might be a trace in one of the Xorg log files after it crashes (/var/log/Xorg.*.log) [08:30] chrisccoulson: so ubuntu-bug application? [08:31] czajkowski, i'm not sure if that will pick up the correct log file [08:31] i think the old file is rotated when X starts again [08:31] (so it might be /var/log/Xorg.0.log.old or something) [08:31] no what do I file it against ? just so I know [08:31] sorry [08:32] probably xorg-server for now [08:32] thank you [08:34] ricotz, ok, uploaded [08:34] thanks! [08:36] pitti, i figured out how to kill most of the extensions on https://wiki.ubuntu.com/MozillaTeam/FirefoxUpgradeChecklist/5.0 for lucid and maverick so we don't have to update them every 6 weeks, and without the user even noticing :) [08:36] (ie, they will still get up-to-date and fully functional extensions with upgrades) [08:39] chrisccoulson: thanks for your help [08:45] hey [08:47] salut seb128 [08:47] lut didrocks [08:52] morning [08:52] hey rodrigo_, how are you? [08:53] hi seb128 [08:53] pitti, evolution-common is 10meg installed, 2.8meg of deb [08:53] so the evo = 0.5 isn't right [08:53] (just reading the blueprints emails) [08:56] chrisccoulson: so, if you answer to https://bugzilla.gnome.org/show_bug.cgi?id=608402#c3, vuntz agreed to review the patch again (FYI, I've updated it in the latest gnome-session upload, you should maybe start from that?) [08:56] Gnome bug 608402 in general "Add support for delaying autostart applications" [Enhancement,New] [08:59] bonjour seb128 [08:59] seb128: ah, thanks [08:59] hey pitti [09:00] pitti, you should count evolution-exchange as well and maybe some of e-d-s ui libs [09:00] seb128: do we really have to ship evolution-exchange by default btw? [09:00] it will not compensate the tb but will not be 0.5 either [09:00] seb128: right, updating the spec, thanks for pointing out [09:00] didrocks, "have to" [09:01] didrocks, well it's like everything, some users need it especially in corporate environments and it's small enough [09:01] like, can we ran e-d-s without it? Seems more a corporate environment requirement? [09:01] it's of no use if we don't use evo [09:01] chrisccoulson: oh, awesome! you can switch them over to the firefox plugin updater somehow? [09:01] that's why I mentioned it as something to count on the "to drop list" [09:02] pitti - yeah, that's the plan - http://bazaar.launchpad.net/~mozillateam/ubufox/trunk/revision/233 :) [09:02] although, i need to test all cases to make sure it works still [09:02] pitti, btw don't we have some langpacks still to drop to make up for it? i.e getting tb seems it would be about one langpack [09:02] seb128: but even if we stay with evo, seems like 479k to win, isn't it? [09:02] hey chrisccoulson, you are early today [09:02] hi seb128. i'm not that early am i? [09:03] seb128: that's the part what I meant with "drop something else instead" [09:03] didrocks, well, the number of users who need to connect to exchange servers for their work seem to warrant that [09:04] seb128: 16 MB extra is 2 langpacks, yes; which essentially means that only chinese will be left, and Spanish on i386 (and powerpc will be screwed even more) [09:04] chrisccoulson, ok, maybe it's me who is late today ;-) [09:04] seb128, yes, you're slacking today already ;) [09:04] seb128: ok, I was more thinking as an easy target, but if you feel most our users need it (but IIRC, thunderbird doesn't have this support, so if it's "required", the discussion can stop ;)) [09:04] seb128: I just don't consider langpacks as something which we can throw away without any pain; it makes testing the live system significantly harder for non-English speakers [09:04] chrisccoulson: did you see my message about the gnome-session patch? [09:05] and at least Spanish, Portugese, and Chinese cover a lot of the world.. [09:05] didrocks, yeah, will look in a bit [09:05] chrisccoulson: excellent, thanks :) [09:05] pitti, though we are down to 3 locales anyway so that or none is close to be the same [09:06] seb128: well, if you count a difference of 1.5 Billion people to be negligible :) [09:07] chrisccoulson: nice trick! [09:07] chrisccoulson: that would go along with updating these packages to be empty transitionals then? [09:08] pitti - nearly. we'd still need to ship the xpi in the package, which would then just be picked up by the addon installer [09:09] pitti, btw libevo is 2.5m deb [09:09] pitti, so adding the deb you should get at least 6m rather than 0.5 [09:09] that's not needed by e-d-s? [09:10] ah, indeed [09:10] no, that's build from evo, that's the ui part [09:10] ok, changing again [09:10] that's basically what they splitted to be able to write things like (anjal), i.e mobile uis based on evo [09:10] pitti - we also need to make sure that the xpi's have an updateURL field in their install.rdf too [09:10] (else firefox can't update them) [09:10] seb128: so, down to 14 MB delta; better than 20 indeed [09:11] but doing this once is definitely going to be easier than doing it every 6 weeks :) [09:11] of which 8 MB is growth between the curretn version (12 MB) and the daily builds (20 MB) [09:11] chrisccoulson: do you think that this 8 MB growth can be reduced again a bit? [09:11] we should move all the comments on the whiteboard to a wiki or something [09:11] it's hard to spot the actual diffs in the email ;-) [09:12] pitti - i'm not sure what's caused it to grow yet (firefox grew by a similar size too) [09:12] so if we could find a solution for sharing libxul, and reduce the size growth, we'd be almost in the ballpark [09:14] seb128: we didn't count the size of lightning yet, though [09:14] i'm really concerned about lightning :/ [09:14] I take it we'd need to add that as well, unless we want to drop calendar support [09:14] (just like all of the binary extensions) [09:15] well, or anything which provides calendar support on the desktop [09:15] they don't have a solid calendar story [09:15] as long as it tells you about your appointments and has a reasonable UI [09:15] that's a weak point for the switch imho [09:15] they made it clear that they are not maintaining the calendar [09:15] so it means there is no garanty the calendar keeps working on updates etc [09:16] it's really third party code [09:21] b'ah, my ubufox trick doesn't handle incompatible extensions too well. seems i need to manually trigger an extension update too :/ [09:21] * chrisccoulson looks at more documentation [09:53] pitti, btw if you want to start making some space on the CD we can clean the gnome-system-tools stack now [09:53] the new gnome-control-center does manager user accounts [09:53] which was the remaining bit we needed in gst [09:53] morning everyone [09:54] once the at-spi2 mirs are reviewed we might be able to drop bonobo etc as well [09:54] lut huats [09:54] hello seb128 [09:54] seb128: ah, indeed; seems still a bit buggy, but should by and large do what we need [09:57] pitti, right, and if it lacks something we should fix it rather than to stay back [09:57] seb128: fully agree; I'm looking forward to getting rid of all this Perl :) [09:57] ;-) [09:58] seb128: the only thing that immediately sprang into my eye is the wrong handling of "administrator" accounts [09:58] it apparently doesn't check for the "admin" group, but for something else [09:58] but that seems trivial to fix [09:58] does somebody know where "libgtk3.0-bin" might come from? I've seen two users already in #ubuntu+1 having problems with it (bug 786170) [09:58] Launchpad bug 786170 in gtk+3.0 "libgtk-3-bin failed to install/upgrade: diversion of /usr/sbin/update-icon-caches to /usr/sbin/update-icon-caches.gtk2 by libgtk-3-bin clashes with same diversion by libgtk3.0-bin" [Medium,New] https://launchpad.net/bugs/786170 [09:58] geser, gtk3? [09:59] libgtk3.0-bin is an old package name [09:59] seb128: the current package is named libgtk-3-bin [09:59] presumably from a PPA or so? [09:59] that's a leftover I guess [09:59] Package: libgtk-3-bin [09:59] Conflicts: libgtk3.0-bin [10:03] geser, pitti: the most recent commit in the debian svn is [10:03] "Remove leftover diversions from the libgtk3.0-bin package [10:03] that got replaced by libgtk-3-bin (Closes: #617662 ). [10:03] " [10:03] ah, nice [10:03] so we should backport that to oneiric [10:07] pitti, the admin issue is likely similar to http://bugs.debian.org/618764 [10:08] c'mon bugs.d.o [10:08] pitti, there is a patch in accountsservice which does [10:08] + if (cd->account_type == ACCOUNT_TYPE_ADMINISTRATOR) { [10:08] + add_user_to_group (context, cd->user_name, "sudo"); [10:09] + grp = getgrnam ("sudo"); [10:09] pitti, we might need to tweak that in ubuntu [10:09] right, also for the check if a user is an admin [10:09] s/sudo/admin/ [10:09] so, should be easy [10:09] - if (groups[i] == wheel) { [10:09] + if (groups[i] == admin) { [10:09] g_free (groups); [10:09] return ACCOUNT_TYPE_ADMINISTRATOR; [10:09] hum [10:10] seems it does check for admin [10:10] is admin == getgrpname('admin')? [10:11] chrisccoulson: I'm here now. [10:11] pitti, ok, they have [10:11] RAOF, do you have an account on bugzilla.mozilla.org? [10:11] grp = getgrnam ("sudo"); [10:11] if (grp == NULL) { [10:11] g_debug ("sudo group not found"); [10:11] r [10:11] return ACCOUNT_TYPE_STANDARD; [10:12] chrisccoulson: I don't think so, no. [10:12] woah, removing the entire python protocol support (iteration, subscription) for the treemodel path stuff in GI/gtk3 is just … not nice [10:12] mvo: there are some overrides for that, they don't work? [10:13] RAOF, i got subscribed to a private security bug that might be of interest to you (it's really an X bug) [10:13] mvo: we can add some more overrides if needed [10:13] pitti, so I guess changing that check sudo -> admin should do it, I will try that [10:13] pitti: oh, get stuff that used to work (like len(path), path[0] etc) is no longer working for me [10:13] pitti: do I need to do anything to make this work again, i.e. set some compat option or somesuch? [10:13] chrisccoulson: I'll get me (yet) a(nother) bugzilla account. [10:13] honestly, forcing that C api for this on people is IMO a huge step in the wrong direction [10:14] RAOF, mozilla bug 658219 (i don't think you can view it unless i subscribe you though) [10:14] chrisccoulson: Error: Error getting Mozilla bug #658219: NotPermitted [10:14] mvo: there are no "compat" modes or so; it's by and large doing what GTK does, plus /usr/lib/python2.7/dist-packages/gi/overrides/Gtk.py [10:14] RAOF, there's not much information on there atm, but you might be a better person to extract the right information and figure out what's going on [10:14] mvo: the overrides do define a __getitem__ [10:15] which support both a Gtk.TreeIter and an int as value [10:16] pitti: thanks, that is a good pointer, I check it out [10:16] (the py file) [10:16] where's the page that summarizes all WIs from the blueprints? [10:16] mvo: if you have something which should be added there, please let bz.g.o. know, I can mediate with upstream (or even commit myself) [10:16] rodrigo_, http://people.canonical.com/~platform/workitems/oneiric/canonical-desktop-team.html [10:16] mvo: for apport and jockey these were quite sufficient, though [10:17] rodrigo_, if you only want our team, similar pages exist for other teams [10:17] pitti: ok, thanks! [10:17] seb128, yes, just want our team for now, thanks! [10:17] pitti: glad to hear, for a moment I was really scared :) [10:18] (I thought it was like with the default arguments for pack_start() - stay close to the C etc) [10:19] Yes! gnome-session no longer has a 1ms window where you could perform some input and have the session perpetually marked as idle! [10:19] chrisccoulson: raofubuntu.com is ready for your subscription. [10:20] RAOF, thanks. you should be able to see it now [10:21] Yup. [10:21] Aaah, firefox and XRender. [10:21] heh :) [10:22] Firefox is fond of rendering to insanely sized pixmaps (like 16k x 4k) [10:22] ah, ok :) [10:22] Is that magyar? :) [10:23] RAOF: thanks for the patch, and sorry for not reading about it earlier :-) [10:24] vuntz: No problem. If it had been an easier [10:24] -to-trigger bug I'd have pushed harder :) [10:25] Oh, balls. Lucid. [10:26] RAOF, yeah, that's a pain. i was going to try booting from a live image later to see if i could reproduce it [10:27] I can't, but oneiric has a *slightly* different X stack :) [10:27] yeah, i couldn't reproduce it on oneiric either === kinoucho` is now known as kinouchou [10:38] vuntz: autostart-delay> \o/ [10:38] vuntz: want me to close the bug then? [10:38] (sorry if that came through twice, got a disconnect) [10:42] vuntz, reviewing patches \o/ [10:46] pitti: autostart-delay is not in yet, but will re-read the patch and do the requested changes if chrisccoulson is not faster, I guess [10:46] ah, it was only the libegg one, sorry for misreading [10:47] pitti: don't be sorry for being enthusiastic! [10:47] vuntz, i'll have a look in a bit. i haven't made a gnome contribution in ages, so i probably should do ;) [10:57] pitti: any hints why I get http://paste.ubuntu.com/613131/ ? AttributeError: 'gi.repository.Gtk' object has no attribute 'drag_dest_set' - I do require version 3.0 of Gtk in the file and I can see it in the gir xml (gi.require_version("Gtk", "3.0")) [10:58] * pitti looks, I think I know why [10:58] mvo, hi. http://paste.ubuntu.com/613132/ known? [10:58] mvo: yep, it's a method of Gtk.Widget, not a function in Gtk [10:59] mvo: i. e. use mywidget.drag_dest_set(flags, targets, actions) [11:01] pitti, w00t, this does the job quite nicely now - http://bazaar.launchpad.net/~mozillateam/ubufox/trunk/revision/234 :) [11:01] it triggers an update of the extension after installing it now [11:01] fta: yes, thanks for letting me know. oneiric only, right? [11:01] pitti: thanks! [11:02] chrisccoulson: ah, you can trigger the update check manually with that? \o/ [11:02] mvo: how did the tree iter problem work out? [11:02] pitti - yeah. so, the intention is that users who install the old and incompatible extensions in the archive will get an update to a compatible version without me having to lift a finger :) [11:04] pitti: I use path.get_indicies() instead of subscribing directly, that works fine [11:06] mvo, right [11:06] hm, no longer crashing but ot accepting drops either [11:06] * mvo looks further [11:10] * pitti looks at current langpack-o-matic change [11:10] 575 files changed, 24 insertions(+), 4277 deletions(-) [11:10] c'est la cleanup! [11:12] pitti, :) [11:15] pitti, now the question is how you landed to write that number of lines for nothing ;-) [11:16] seb128: well, most of it was due to moving the language-support-* dependencies to language-selector [11:16] ripping out the code and test cases for language-support-* did the rest [11:16] we don't carry a cairo patch to enable freetype subpixel smoothing any more do we? [11:16] but it makes the thing quite a bit simpler indeed [11:17] chrisccoulson, no we don't [11:17] that went upstream in 1.10 [11:18] pitti, nice ;-) [11:18] seb128 - oh, that's good. i can drop that from firefox and thunderbird now then. they just upgraded to cairo 1.10, and broke all my builds :) [11:18] pitti, btw I fixed the admin thing, will upload in a bit [11:18] * pitti hugs seb128, you rock! [11:18] * seb128 hugs pitti back === chaotic_ is now known as chaotic [11:24]