=== bil21al is now known as s9iper1 [01:02] robert_ancell: Would you mind looking at some code I'm working on to add shortcut keys to some of the menu options in the accessibility menu for unity-greeter? It builds both in vala and C, but throws an assert at runtime. Have I missed anything? https://code.launchpad.net/~themuso/unity-greeter/add-a11y-shortcuts [01:02] TheMuso, ok [01:03] I.e the greeter runs in test mode, but GTK throws an assert about the accel_group variable, which should be a GtkAccelGroup object. [01:04] TheMuso, hmm, I don't know a lot about accelerators but it looks correct to me [01:06] TheMuso, can you pastebin the error [01:07] robert_ancell: Yeah, I also see some other interesting asserts further down in this log: http://paste.ubuntu.com/834683/ [01:09] TheMuso, the first asserts are something broken in a recent GTK+ upload, the later stuff is the indicators being noisy [01:09] Ok. [01:12] it's the [+0.09s] CRITICAL: gtk_widget_add_accelerator: assertion `GTK_IS_ACCEL_GROUP (accel_group)' failed that's suspicious [01:12] TheMuso, can you print out with %p the accel_group and check it was actually created? [01:13] also, as and aside you don't need to make the accel_group a class variable - just use var as it's only used inside that one method [01:14] robert_ancell: Do you mean just after calling gtk_accel_group_new? [01:14] TheMuso, yup [01:14] I see no reason why it wouldn't be valid [01:14] but worth checking [01:14] Yup. [01:14] Ok will do. [01:17] Its also worth noting that whenever I run unity-greeter in test mode, my keyboard layout gets changed for no reason. [01:17] This is trunk of course. [01:22] robert_ancell: ok, its a non-zero value, accel_group = 0x7fd3e8083100 [01:25] TheMuso, hmm, don't know - might be worth making a small test program and confirming that it works in the simple case or checking the generated C code to see if it's doing the right thing [01:26] The C code looks ok, but I may fiddle with it anyway to see if I can get things working. [01:26] But yeah a test program may be worth trying too. [01:48] robert_ancell, quick q: on multi-monitor I am seeing my wallpaper tiled when transitioning from LightDM to Unity - have you seen this issue before? [02:17] jono, is it mirroring then switching to multi-display? [02:17] robert_ancell, yep [02:17] jono, yeah, that's the current behaviour, but we want to sort that out [02:17] cool [02:18] I think I filed a bug about this, I can't quite remember [03:44] robert_ancell: Ok, just wrote up a simple test app with a single menu, and menu item with an accelerator, and it works fine, no asserts... [03:44] So its something to do with the unity-greeter code, and the only thing I can think of, is that its pointer related... [03:45] Or something is not passing through from one object/method to the other. [03:45] But even that doesn't make sense. [03:45] To be clear, the test app is in vala. [03:48] Perhaps some fresh eyes on my code will help. I'll ping Michael when I see him on next, probably tomorrow morning. [03:49] yeah, weird [03:49] Thanks anyway. === JackyAlcine is now known as ThisCode === ThisCode is now known as JackyAlcine [04:36] jasoncwarner_, i just uploaded a fix for the ido bug, the white background in the sound menu [04:37] not sure if it is a real fix or just masking the real problem... but it seems to work [04:37] DX really needs to find a maintainer for libido... and not me! [04:37] :) [04:38] dobey, did you have any more fun with that branch? [04:39] Good morning [04:39] desrt: no, nessita's two fixes came after 3.1.0 [04:40] desrt: the env one didn't seem too important for me, so I didn't bother backporting it; if you need it, I can [04:41] good morning pitti [04:53] kenvandine, rumor has it you may have experience doing async pixbuf loading [04:54] DBO, ha... well i have a working implementation [04:55] kenvandine, using? [04:55] vala [04:55] gdk_pixbuf_loader? [04:55] lp:gwibber/libgwibber-gtk/stream-view-tile.vala [04:55] kenvandine, you're not just using g_idle are you? [04:56] no [04:56] look at the load_avatar_async function [04:57] are we looking at the same code? [04:57] no we are not [04:57] sorry [04:59] kenvandine, isn't new Gdk.Pixbuf.from_stream_at_scale more of a sync call... [04:59] kenvandine, it looks like what you have implemented is a deferred loader [04:59] agateau, awake? [05:05] DBO, it's wrapped inside a async function [05:05] so not what you want [05:06] kenvandine, I can wrap an arbitrary call inside an async function [05:06] I was hoping more to see something loading a stream async and pumping that into a gdk_pixbuf_loader [05:09] You mean like a preloading filestream that you could pass into the loader? [05:15] kenvandine: hm, empathy repeatably FTBFSes, seems to miss x11-xcb; missing build dep? [05:15] looking [05:17] RAOF, im not sure what I want yet [05:17] I am more or less shopping for a solution [05:17] window shopping if you will [05:17] pitti, interesting, the build log does look like a missing build dep, but it builds on the other arches [05:17] Package 'x11-xcb', required by 'egl', not found [05:17] ah [05:17] or bug in egl? [05:18] Do the build logs span a mesa upload? [05:18] uhg, so many worthless things done on startup in unity [05:18] pitti, that is what i am thinking [05:18] I wish I had a way to benchmark the startup times [05:20] the x11-xcb thing also affects gnome-games, cheese, & gnome-documents [05:20] everything clutter? [05:20] looks like something on arm requires that which is missing a dep on x11-xcb [05:21] Yeah, looks like a mesa bug; egl Requires.private x11-xcb [05:21] RAOF, can you fix that? [05:22] i. e. egl-dev should grow a depends: x11-xdb-dev? [05:22] pitti: Yeah, seems like it. [05:22] on libx11-xcb-dev [05:45] * RAOF wonders how hard it'd be to write a lintian check for underspecified -dev dependencies... [05:45] RAOF: running through Requires:/Requires.private, map through pkg-config, and map through dpkg -S? [05:46] that would be highly useful indeed [05:46] it's a very common cause of FTBFS [05:46] Right. [05:47] It seems reasonable to assume that all those packages are installed as a part of the build, so the dpkg -S route should work. [05:49] yes, otherwise it should be very hard to build the package in the first place [05:51] Oh. I don't think that'd work in the case of sbuild's automatic post-build lintian run. :( [05:56] kenvandine: sweet..will test! thanks :) [05:57] jasoncwarner_, just don't let anyone make me the upstream maintainer of ido [05:57] :) [05:58] kenvandine: wait, what? shoot..."UNDO SEND! UNDO SEND!" [05:58] ;) [05:58] :) [05:58] kenvandine: bedtime for you? getting to be 1am? [05:59] soonish :) [05:59] more hacking to do [05:59] kenvandine: me too; this spreadsheet won't make itself! [05:59] * jasoncwarner_ cries at spreadsheet hacking ;) [05:59] haha [06:00] better you than me [06:14] so, who here has a 3G dongle and want to see my crummy C code fail horribly? [06:16] cyphermox: I still have one, although not used in ages [06:16] cyphermox: that needs usb-modeswitch [06:16] cool. [06:17] I'll be rolling out a package to test in a minute, if you could give it a shot [06:18] I just spent 4 hours or so adapting all the last changes in usb_modeswitch_disabler^Wdispatcher to my C code [06:18] also, good morning pitti :) [06:30] good morning [06:31] jbicha, kenvandine: Your egl build failures should be able to be given back once the new mesa's built. [06:31] RAOF, thx [06:40] didrocks: hey Didier. check the first comment https://bugs.launchpad.net/activity-log-manager/+bug/928946 [06:40] Launchpad bug 928946 in activity-log-manager "The .desktop.in files don't match with executable file name" [High,Confirmed] [06:40] hey manish, looking [06:41] hey didrocks, good morning :) [06:41] manish: right, from dash, it's logicial [06:41] logical [06:41] manish: but from gnome-control-center, it does, as I told you [06:41] yes [06:41] hey cyphermox, how are you? [06:41] tired :) [06:41] :) [06:41] but I think the problem should be solved from all sides [06:41] ttyl [06:42] didrocks: I am opening an ubuntu task on that bug [06:42] manish: indeed, but I told that can wait your next release if you are going to release today or tomorrow [06:43] didrocks: might release it on 13th [06:43] since the UI changes are not done completely [06:43] manish: that's good as well :) [06:43] so is it fine that dash launching be broken so long? [06:45] manish: it's not installed by default [06:45] manish: and not really ready for wide consumption [06:46] (seeing the number of debug message and warning on the console) [06:46] so not that important, that can wait Monday [06:47] didrocks: seif fixed that debug issues [06:47] great [06:47] not installed by default? wasn't it supposed to be as per the blueprint? [06:48] manish: it will when: [06:48] - there will be no debug [06:48] - there will be no warning [06:48] - we fixed the desktop thing [06:48] okay. will be so on 13th release [06:49] :) [07:29] * kenvandine heads to bed, bbl [07:55] multiarchified libgnomeui, fixes gnucash [08:15] tjaalton: GNOME 2 is alive!!! [08:16] pitti: :) [08:17] well, gnucash setup wizard window was blank without this [08:21] pitti: that was a joke or seriously? [08:21] Sweetshark: any chance to get bug 917153 fixed soon? it causes upgrade failures all over the place [08:21] Launchpad bug 917153 in libreoffice "failed to upgrade from oneiric to precise: /usr/lib/libreoffice/program/unopkg.bin: error while loading shared libraries: libicule.so.48: cannot open shared object file: No such file or directory" [High,Triaged] https://launchpad.net/bugs/917153 [08:22] manish: joke, of course; I alluded to tjaalton uploading libgnomeui *blowing off the dust* [08:22] pitti: so when is bonobo getting uploaded? :P [08:22] when we finish porting it to vala and dbus [08:23] but i still miss GNOME 1.x, we should totally return back to it and make sure the desktop is like how it should be - unusable (stolen joke) [08:37] so.. gst-plugins-good/bad merges.. -bad split the libs into a separate package, but we've moved them to -good. should -good do a similar split? [08:37] *moved some of them [08:39] kinda silly though that the moves were done with the promise that upstream would move them too, but a year has passed.. [08:48] tjaalton: you moved the libraries to -good too? wtf? [08:49] tjaalton: why are the libraries needed in -good too? [08:49] slomo_: not moved anything [08:49] yet [08:49] slomo_: -bad is in universe, -good is in main [08:50] tjaalton: and upstream would move the stuff if there would be someone to actually work on it and care enough... apparently none of the distributions care enough and the telepathy/empathy people don't care enough either [08:50] slomo_: right.. [08:50] tjaalton: afaik you only moved some farsight/telepathy related plugins from -bad to -good because of this, i don't think the libraries would be needed for these plugins [08:50] oh wait, you want gstreamer-vaapi in main, right? :) [08:51] right now I'm just trying to do the merges [08:51] dunno if -vaapi needs to be in main or not, probably not [08:51] so, if there's someone at ubuntu who would like to help to get the stuff moved to -good, i'd be happy to help :) [08:52] libgstbasecamerabinsrc and libgstphotography are the ones that have been moved to -good.. [08:53] of the libs [08:53] oh, for camerabin2 which is used by cheese? [08:53] no idea [08:54] i think camerabin2 and these two libs are almost ready to be moved to -base/-good [08:54] would be nice [08:55] so, what's needed? [08:55] for that to happen? [08:55] tjaalton: someone to actually push for it, to make sure that all documentation is there is required, that there are unit tests (there are), that the API of the plugins and libraries make sense and can be kept stable for a long time [08:59] hey [08:59] i can move stuff arond, but can't say anything about the api issues [09:00] *around [09:00] hey seb128 [09:00] hey pitti, wie gehts? [09:00] seb128: quite fine, thanks! [09:03] tjaalton: moving the stuff around is actually the smallest problem, that's easy, there are even scripts for that ;) [09:04] slomo_: right, I'm not up for the reset [09:04] rest [09:04] frankly, this is getting way out of hand :) [09:05] :) [09:07] so I'm just inclined to do the merges now and leave it at that for now.. [09:10] tjaalton: ok [09:20] seb128: FYI, the pygobject backported fixes are in now [09:21] pitti, yeah, I noticed, that will make nessita happy, thanks ;-) [09:21] pitti, they just rolled a tarball with the 3 commits you backported ;-) [09:21] seb128: really: I saw 3.0.4 [09:21] ? [09:22] but we'd need a 3.1.1 [09:22] oh right [09:22] I didn't look into details just saw the email on the ftp list [09:22] grrrr, hate webkit hate hate [09:24] "collect2: ld terminated with signal 9 [Killed] [09:24] libtool: install: error: relink `libwebkitgtk-1.0.la' with the above command before installing it [09:24] make[3]: *** [install-libLTLIBRARIES] Error 1" [09:24] that's how the ppa amd64 build failed after 10 hours build [09:24] hi seb128 ;) [09:24] hey chrisccoulson [09:25] chrisccoulson, how are you? [09:25] chrisccoulson, still don't want to take on webkit for me? ;-) [09:25] seb128, yeah, not too bad thanks. how are you? [09:25] I was good until getting that webkit email :p [09:25] no, firefox and thunderbird are enough for me ;) [09:53] pitti: will be fixed with the next release (3.5.0 final) [09:54] Sweetshark: ah, when is that due? [09:55] pitti: upstream? ~this week [09:56] pitti: 3.5.0 rc3 is expected to be final really (and is out), but ssshh, you didnt hear that from me .... [10:07] morning all [10:08] updated, and now chromium-browser throws a segfault, but not when I strace it [10:08] hey [10:08] what did you update? [10:09] ohidunno :) [10:09] did you try running it in gdb to get a stracktrace? [10:09] ah, it's not every time. twice in a row it worked, there. [10:09] going for gdb [10:11] seb128: 20 times it started ok within gdb [10:11] "fun" [10:11] does it print anything on stdout when it errors out without gdb? [10:12] three times it worked ok without gdb, the fourth, segfault [10:12] $ chromium-browser [10:12] Segmentation fault (core dumped) [10:12] nope [10:12] hum [10:12] no apport catching it? [10:12] nope [10:12] :-( [10:12] that's going to be fun to debug [10:13] can you check /var/log/apport.log [10:14] this executable already crashed 2 times, ignoring [10:14] and it looks like i've got something in /var/crash about chromium [10:21] Chipaca, ok, report it? i.e run ubuntu-bug on it [10:21] Chipaca, or if it's too much to upload you can apport-unpack the file and look at Stacktrace in the unpacked directory [10:22] seb128: i'll do the unpack thing anyway, but reporting should be doable [10:28] pitti, \o/ gzip bug fixed! [10:29] seb128: yeah, indeed! I reverted the workaround in gtk+2.0 bzr this morning [10:29] cool [10:29] I figure we can revert a few more [10:29] yeah, I will look at that [10:29] probably no need to upload but undo the workaround in the vcses [10:30] *nod* [10:41] slomo_: btw, debian bug 657011: drop libmusicbrainz4-dev from the build-deps [10:41] Debian bug 657011 in gst-plugins-bad0.10 "gst-plugins-bad0.10: Please drop the build-dependency on libmusicbrainz4-dev" [Important,Open] http://bugs.debian.org/657011 === tkamppeter_ is now known as tkamppeter [11:24] slomo_: so.. scripts for moving stuff around? that might help me with the merge, probably a good idea to update the moved plugins.. [11:25] tjaalton: it's just something to extract all git commits for a number of directories/files and import them into another git repository [11:25] tjaalton: nothing that would help you in your case unfortunately [11:32] slomo_: ok, I'll just copy & create a patch from it === MacSlow is now known as MacSlow|lunch === Ursinha` is now known as Ursinha === greyback is now known as greyback|lunch [13:24] pitti, hello [13:25] hello glatzor, wie gehts? [13:25] pitti, I just looked at your bug from yesterday [13:25] pitti, fine and yourself [13:25] ? [13:25] pitti, http://paste.ubuntu.com/835187/ [13:25] quite well, thanks! [13:25] ooh [13:25] glatzor: I tried to set error_handler in the ctor [13:26] glatzor: brilliant, thank you! [13:26] pitti, the gtk3 api of the dialog isn't very nice. the run() method should really fake a nested main loop and iterate the mainloop itself [13:27] if it is caled synchronously [13:37] seb128: objections if I split the shared libs & dev files from gst-plugins-good like I did for -bad? [13:37] tjaalton, is slomo_,Debian going to follow you on that? [13:38] tjaalton, I would like to stay close from Debian [13:38] seb128: probably, once the libs are moved upstream from -bad to -good [13:38] tjaalton, btw if you work on it could you also update to the current version? ;-) [13:38] seb128: no, there are no libraries in -good (and never will upstream) [13:38] slomo_: uhm? [13:38] so why are we moving these tehn [13:38] then [13:38] tjaalton: the libraries will be in -base [13:38] slomo_: ah gotcha [13:39] in that case I won't split them [13:39] seb128: I've merged the experimental version yes [13:39] well, you need to... where would you put the headers otherwise? [13:39] tjaalton, well, here is the deal, if you add diff over debian you win the maintainship of that source, i.e you will do the merges on debian when new versions come ;-) [13:39] and copied the new versions from -bad [13:40] slomo_: they are not shipped currently [13:40] just the libs with -plugins-good [13:40] seb128: i don't think this rule is that strict ;) otherwise there always would've been someone to merge with debian after you moved the plugins and libraries [13:40] tjaalton: ok, fine then [13:40] libgstbasecamerabinsrc and libgstphotography [13:41] anyway, I'll keep it on my local branch if ever needed [13:43] slomo_: btw, typo in -bad control.in, the lib short description is the same as for -dev :) === MacSlow|lunch is now known as MacSlow [14:03] dear pitti, can https://launchpad.net/~didrocks/+archive/ppa/+build/3198709 get some priority bump? I think that it really deserves it if possible :) === greyback|lunch is now known as greyback [14:04] didrocks: done [14:04] pitti: thanks :) [14:04] didrocks: that was the i386 build, though, ok? [14:04] I'd assume that most developer folks run 64 bit [14:04] pitti: not me though ;) === davmor2_ is now known as davmor2 [14:14] seb128: I'm on holiday tomorrow; would you happen to have time to do the release meeting stuff? [14:15] pitti, yes, can do [14:15] seb128: merci beaucoup [14:15] pitti, did you update the status page? [14:15] seb128: no, not yet [14:15] I usually do it Friday morning, but I can do it now if you want [14:15] pitti, will you do it or should I add that to my todo for tomorrow? [14:16] it's just getting a bit tight [14:16] pitti, if you have time and want to do it you are welcome to do so [14:16] meeting at 15:30, then I need to run out soon after [14:16] otherwise I will do it, or finish what you didn't do tomorrow [14:16] seb128: I'll start on it now [14:16] thanks [14:23] hey seb128, remember you were going to look at why we can't load g-c-c panels that are hidden with the desktop file? [14:24] like in unity, gnome-control-center sound [14:24] should load the old sound panel [14:24] seb128, good morning btw :) [14:24] kenvandine, hey, yes, it's still on my list, was not very high though [14:25] seb128, well i am about to upload the goa change [14:25] :) [14:25] like today [14:26] kenvandine, do we need the goa dialog to be called under unity? [14:26] kenvandine, thinking about it I'm unsure that's a good idea to make that [14:27] kenvandine, it means the "all settings" would take you back to a screen which doesn't list the capplet you were using [14:27] we do [14:27] empathy-accounts calls it [14:28] it will only happen for upgraders that created the account with goa [14:28] seb128: https://wiki.ubuntu.com/DesktopTeam/ReleaseStatus updated [14:29] pitti, thanks [14:29] seb128: perhaps you can give it an once-over tomorrow, in case some more things change [14:29] pitti, will do [14:29] seb128: tomorrow morning, can you please send a mail like https://lists.ubuntu.com/archives/ubuntu-release/2012-January/000654.html ? [14:29] pitti, ok [14:29] seb128: sorry, no, like https://lists.ubuntu.com/archives/ubuntu-release/2012-February/000790.html [14:29] the format changed a bit (release note delta, etc.) [14:30] ok [14:30] seb128: meeting is at 17:00 CET as usual, but now much quicker these days [14:30] (basically, just answering questions, no report pasting, etc.) [14:31] I like that ;-) [14:32] hmm, is it normal that .jpg files automatically are associated with gwibber instead of the image viewer ? [14:34] ogra_, no, is that specific to your user? [14:34] [14:34] warning [14:34] bug 929384 [14:34] Launchpad bug 929384 in nvidia-graphics-drivers "nvidia drivers broken by the recent libc update" [High,Confirmed] https://launchpad.net/bugs/929384 [14:34] havent tried yet, since i was busy filing a nautilus bug (nautilus crashes if i try to change the file association in "preferences") [14:35] don't upgrade your precise libc if you are on nvidia binary drivers [14:35] heh, well, i'm on arm [14:35] ogra_, can you give me the bug number for the nautilus issue when you have it? [14:35] (using nvidia drivers, but thats quite different from the x86 ones) [14:35] one sec [14:35] ogra_, also check your .local/share/applications/mimetypes.list [14:37] kenvandine: didn't futz with it any more yet really. though pondering just switching it to use something else instead [14:37] seb128, bug 929557 [14:38] ogra_, thanks, waiting on the retracers [14:38] dobey, something else? [14:38] not sure there are armel retracers [14:39] seb128, hmm, indeed i have "image/jpeg=gwibber.desktop" in there, i wonder how that happened, i havent used gwibber ever in my life [14:39] kenvandine: http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntu-sso-client/trunk/view/head:/ubuntu_sso/utils/txsecrets.py [14:39] * ogra_ doesnt have a twitter account or anything [14:39] ogra_, there used to be a bug in oneiric which would do that [14:39] ah [14:39] ogra_, when opening the control center system infos [14:39] ogra_, but is it possible that you had the bug since oneiric and didn't notice before? [14:40] likely, i havent opened jpegs in ages [14:40] ok [14:40] * ogra_ removes the line [14:40] kenvandine: though since it's designed for twisted, i guess it might not work as well in gwibber [14:41] hmm, looking at these associations, banshee doesnt clean up after itself either [14:42] oh, i dont even have a nautilus icon in my launcher anymore (nor a desktop) ... [14:42] * ogra_ logs in newly [14:42] dobey, interesting... we could create a simple wrapper to just use dbus though right? [14:43] dobey, you should go to #gwibber [14:45] k, desktop is back, jpegs are opened by image viewer by default again ... [14:45] but there is no home icon in my launcher [14:46] and the "top left corner" setting for unity-2d doesnt switch off "edge" it seems [14:46] isnt it an "either or" thing ? [14:46] seb128, oh, and thanks ! ;) [14:47] ogra_, yw [14:47] hmm, actually "top left corner" doesnt seem to have any effect [14:53] dobey, are you going to be the package maintainer for the video lens when it lands? i assume it'll go in your package set for u1 [14:55] kenvandine: not afaik. i understood it would fall onto dx [14:55] dobey, but your team is writing it :) [14:55] so maybe you will just hand it off [15:09] good morning [15:10] hey jbicha, how are you? [15:10] seb128: doing good [15:11] seb128: could you sync itstool? the new version makes xml warnings non-fatal so ubuntu-docs won't FTBFS [15:15] jbicha, done [15:16] pitti, btw... congrats on GNOME membership! [15:16] kenvandine: thanks! [15:28] good night everyone, have a nice weekend! [15:29] seb128: thanks for covering for me tomorrow [15:29] pitti, yw, have a nice w.e! [15:29] I'll check mail again tomorrow morning, I'm not leaving until ~ 11 am [15:38] humm, so the MIR for dbus-test-runner is fix released, i know it'll go into main when something depends on it in main [15:38] but what if something only needs it as a build depends? [15:39] jasoncwarner_, hi [15:39] tkamppeter, it's 3am for him, I doubt he's online [15:40] kenvandine, that's fine, it will show on component mismatch [15:40] seb128, sorry, I forgot. [15:40] seb128, cool [16:18] hello, launcher autohide seems to be broken. known issue? [16:18]