[00:01] haha [00:02] what is a .udeb? [00:06] You can ignore that; that's for the installer. [00:08] TO be clear, its for the Debian installer system. Whilst Ubiquity uses parts of Debian Installer, it doesn't use udebs. [00:08] ^^ for thumper's benefit. [00:15] thanks [00:22] thumper, so, you're totally keen to be an ubuntu developer now right? Loving our shaky architecture? [00:27] haha, oh he left [00:30] where is package config normally kept? [00:37] thumper: What do you mean by ‘package config’ in this context? [00:37] nm... [00:37] trying to use pkg-config [00:37] and attempting to workout the names of packages [00:42] pkg-config --list-all | grep pango ☺ [00:49] robert_ancell: bad news... it doesn't actually fix the bigger memory leak [00:49] thumper, got a good test case for it? [00:50] hmm... [00:50] looking at the output for the valgrind log of the test case attached to the bugzilla bug [00:51] just comparing valgrind output for the pre and post fix [00:52] thumper, you mean the 500 errors valgrind gives? [00:52] :) [00:52] yeah [00:53] a lot of the possibly lost is just g_type_init [00:53] and I can look past those [00:54] robert_ancell: ok, compared, and it did fix quite a big leak [00:54] but not all the leaks :) [00:54] so, write some test cases and they'll get fixed! [00:55] :) [00:55] I'm getting there [00:59] robert_ancell: in the pango-leak.c file connected to the bugzilla bug [00:59] robert_ancell: when I run it under valgrind with: G_SLICE=always-malloc G_DEBUG=gc-friendly valgrind --tool=memcheck --num-callers=50 --leak-check=full --track-origins=yes --log-file=valgrind.log ./pango_leak [00:59] I get two definite loss records [01:00] yeah, I got that too, but only on varied with the number of loops, the other one seemed to be constant [01:00] robert_ancell: http://pastebin.ubuntu.com/1041714/ [01:00] robert_ancell: yeah, probably a single parse call thingy [01:01] thumper, file a new bug :) [01:01] :) [02:53] How do I find the source for /usr/lib/x86_64-linux-gnu/libpixman-1.so.0.24.4 ? [02:53] 'cause it has a bug :( [03:02] thumper: apt-cache showsrc $(dpkg -S libpixman-1.so | cut -d':' -f1) will do it :) [03:02] But that's showing off; dpkg -S will find you the package(s) that contains libpixman-1.so, apt-cache showsrc will show you the source package for that. [03:03] Or you could just ‘apt-get source libpixman-1-0’ [03:03] ;) === m_conley is now known as m_conley_away [03:16] RAOF: ta [03:17] thumper: if you don't have it installed, apt-file is a nice tool to have installed [03:17] s/it/the package with the file in question/ [04:53] Bonjour mes amis === Amaranthus is now known as Amaranth [05:57] good morning [05:57] hi didrocks [05:57] hey micahg [05:59] bonjour didrocks, hi micahg [06:00] hi rickspencer3 [06:00] micahg, are you not in USA? you must be up late [06:00] rickspencer3: yeah, bad habits and such, hopefully will start having earlier nights soon :) [06:01] haha [06:01] micahg: where early for you means 2AM? [06:01] heh, yeah :) [06:02] bonjour rickspencer3, comment ça va? [06:03] didrocks, ça va bien ... je n'ai pas beacuoup des reunions a'jourdi, donc, je suis content and productif ;) [06:04] didrocks, et tois? [06:04] rickspencer3: ça va bien, je peux prendre le temps de faire les choses biens, rien de trop pressé, donc ça change :) [06:06] très bien :) === tkamppeter__ is now known as tkamppeter [06:39] didrocks: should I propose merges for my changes on top of the ones that sil2000 gave me (just changelog or packaging stuff) [06:40] micahg: packaging stuff? [06:40] didrocks: natty I reverted to source format 1.0, oneiric I converted it into a patch (as I would've had to revert a patch to convert) [06:41] micahg: no need to convert it as a patch, we are using bzr merge-upstream workflow [06:42] that just gives extra headaches as a patch, that's why we never do that when the packaging branch is derived from the parent one [06:42] didrocks: I did it that so as to make a minimum chnge [06:42] Good Morning [06:42] micahg: well, we wanted to use that opportunity to remove the wrong quilt usage in the previous upload [06:42] didrocks: if you SRU to oneiric, feel free to revert, I took the path of least resistance in each [06:42] micahg: the diff was small, so it was a good opportunity [06:42] so no need for us to take the branch :) [06:44] didrocks: well, I didn't want to do that in a -security upload, so I guess you just need my changelog [06:44] if you SRU to oneiric, you can fight it out with the SRU team [06:45] well, when it's moving 2 files from debian/patches/ to inline patch, the diff was small… [06:45] I guess everyone around here is smart enough to understand it :) [06:46] so now it will be more difficult because you changed the second patch to a quilt patch [06:46] but well, we'll get over it :) [06:46] didrocks: nah, just say that change never made it into your VCS :) [07:22] hey didrocks [07:22] guten morgen pitti, wie geht's? [07:23] didrocks: sehr gut, danke! [07:23] didrocks: how about you? [07:24] pitti: I'm fine, thanks :) [07:24] taking time to play with python3 and python-mock :) [07:26] didrocks: ah, nice; p-mock was introduced to apport quite recently, and it's quite neat [07:27] pitti: yeah, I love being able to test assert_called_once_with(args) [07:28] ah, nice; I use something like self.assertEqual(self.app.ui_start_upload_progress.call_count, 1) [07:28] your's looks easier to read, thanks for pointin gout [07:29] pitti: http://www.voidspace.org.uk/python/mock/, look for "assert_called_once_with" [07:29] (also, using decorator "@patch", loving it!) [07:30] yeah, indeed [07:30] before I used it, I had some gems like [07:30] try: [07:31] MyClass.my_method = my_foo_method [07:31] ... [07:31] finally: [07:31] MyClass.my_method = orig_method [07:31] @patch is so much easier [07:32] oh yeah, way better than doing and hoping you will revert successfully to the previous state :) [07:32] finally: will care for this, but it's still a bit bulky [07:34] yeah :) [07:34] now, I would just hope that nosetests3 get coverage support (python-coverage is only for python2) [08:06] good morning everyone [08:06] who sent a pandaboard to my house? [08:06] one just arrived ;) [08:06] hey [08:07] chrisccoulson, jasoncwarner ordered one for everyone in the team [08:07] chrisccoulson, cf weekly summary email from last week [08:07] seb128, ah, thanks. that's good then. it means i have a new toy to play with :) [08:07] Because we care about arm, and it's a pain in the arse not having any hardware. [08:07] ;-) [08:07] Good morning chrisccoulson & seb128! [08:07] hi RAOF, how are you? [08:08] I'm pretty good. [08:08] hey RAOF, chrisccoulson: happy friday! how are you guys on the rainy day? [08:08] It's the weekend now! [08:08] lucky you ;-) [08:08] Also, I think I've pretty much got enough to unblock Robert on the lightdm work for the system compositor, which means I'll be able to go back to actually coding. Huzzah! [08:09] RAOF, nice! are we going to see anything landing in distro to play with soon? [08:09] Although there's probably the need to fix upstream mesa's build system (again!). [08:10] seb128: Landing in the distro is blocked on a couple of transitions; mesa 8.1 and xserver 1.12, specifically. [08:11] RAOF, ok, not next week I guess then ... I mostly asking because I'm about to write the weekly team summary for -release [08:11] xserver 1.12 is about ready and should go in soon; mesa 8.1 will require me (or, god willing, someone else) to wade through upstream's half-autotools, half-hand-rolled-makefiles and get the damn thing to build. [08:11] RAOF, which includes "changes coming in the next week that might impact other teams or users" [08:11] RAOF, looks like the mesa and xserver transitions might be for next week then? [08:12] chrisccoulson, did you get anywhere with g-s-d btw? [08:12] xserver is likely for next week, yeah. [08:12] Mesa, maybe not. [08:13] ppa:raof/system-compositor-scratchpad + ppa:ubuntu-x-swat/x-staging (until 1.12 lands) should get Robert a functioning system to do lightdm integration work on, though. [08:13] RAOF, thanks [08:13] salut seb128 :) [08:13] didrocks, lut [08:13] seb128: FYI, it's *sunny* here [08:13] and the whole week-end will be *sunny* [08:14] didrocks, we will get some of that on sunday apparently [08:14] seb128: you should think bigger and request sun from the friday, not only on sunday! [08:15] didrocks, yeah, I'm learning, will do better next week! ;-) [08:15] seb128, i didn't hear anything back from mterry actually. i think the session in gdb yesterday was taking a long time [08:15] seb128: I hope so, you will get an evaluation soon :) [08:15] chrisccoulson, would have been faster to install a 32bits vm for you :p [08:15] i wonder if i'll be able to reproduce it if i run a i386 ISO on bare metal [08:15] i couldn't reproduce it in a vm yesterday :( [08:17] wth is robert_ancell doing? [08:17] he did upload stuff depending on gtk 3.5 without gtk [08:18] I made the same comment this morning when seeing it :) [08:18] didrocks, was robert_ancell still around? did you ask him? [08:19] seb128: no, he wasn't online when I looked at the user list [08:19] ok [08:19] I dropped him an email [08:19] let's see [08:20] didrocks, btw look at your queries, I've asked you a question there ;-) [08:21] * didrocks didn't receive any query from seb128 [08:21] so you can wait for an anwser :) [08:21] answer* [08:22] didrocks, thanks, it was a good idea to ask because my client was behaving like I was connected where I was not [08:23] seb128: indeed :) [08:37] pitti, guten tag [08:47] ah, /me just saw that as well :P [08:47] hey seb128 [08:47] pitti, hello [09:47] didrocks, ok, for info robert_ancell meant to upload gtk 3.5 but we agreed to not risk that on a friday and let stuff depwait until monday [09:47] seb128: hum, we should have wait for uploading everything once ready then, if we have something else to upload on the depwait, but well, not that important :) [09:49] didrocks, well, everything was ready, he missed the dput [09:49] ok :) [09:49] didrocks, he though he had uploaded it [09:50] ah ok, so the "we agreed to not risk" is the post-reason [09:50] not what was meant on purpose :) [09:51] right [09:51] didrocks, it's "I just discussed with him, he could dput and let us deal with any potential issue" [09:51] didrocks, but I think I will prefer the "let's deal with it on monday" [09:51] seb128: agreed :) [09:54] is anyone on updating glib to 2.33.2 in debian experimental? [09:55] pitti, I was going to ask you if you wanted to do it at some point, I just didn't want to ask too many things at the same time since I already pinged about the udisks,gvfs,loop thing ;-) [09:56] yeah, I don't have time for it today [09:56] I run stuff in jhbuild for the things I need it for, so I'm not immeidately blocked [09:56] I was mostly curious [09:57] I could try to look at that on Monday === chaoticuk_ is now known as chaoticuk === MacSlow is now known as MacSlow|lunch [12:44] wow, it has just got incredibly dark outside [12:45] got my fingers crossed for a thunderstorm [12:46] lol [12:46] you really like those ;-) [12:46] :) [12:51] didrocks, pitti: you guys have thinkpads right? did you get issues with the pad not working or had to do anything special to get it working? [12:51] I'm looking at bug #804109 [12:51] Launchpad bug 804109 in gnome-settings-daemon "can't enable touchpad in Ubuntu (thinkpads)" [Low,Incomplete] https://launchpad.net/bugs/804109 [12:51] seb128: nothing special to do, works out of the box [12:52] no, no such thing here [12:52] ok, thanks [12:57] seb128: the trackpad? no, that has always worked without a hitch for me [12:59] chrisccoulson, so, the gnome-settings-daemon gdb session is still going. (it was sleeping overnight, so it only had the workday to run) It's possible that we're not hitting the bug with this setup? Like, maybe some race condition isn't being met? (seems unlikely, I don't know why a plt matter would be racy...) [13:00] chrisccoulson, the last stack trace I see is still from loading plugins. I'll let you know if I get one that's past that point. [13:00] mterry, if it's still going, then maybe it isn't going to hit it :( [13:00] chrisccoulson, well, the crash was from an idle callback in a plugin, so it would have had to finish loading plugins to hit it [13:01] chrisccoulson, ooh, let me turn off the watch until then. I was thinking we'd hit it during plugin load, but it's an idle callback! [13:05] mterry, if you want to reduce "noise", try to turn off all the plugins but mouse (set their "enable" in dconf to false) [13:06] pitti, ok, thanks, I asked for details on the bug, let's see what config those users have [13:07] seb128, yeah no, just skipping all plugins does it. i forgot the crash happened in an idle callback, and thus after plugins. now let's see what the next major blocker is [13:08] seb128, oh oh btw, is GUADEC awesome? Or more particularly, terribly useful? I'm trying to figure out whether it makes sense for me to go [13:09] mterry, depends what you are looking for, it's less of an hack conference than Boston summit, it's rather a "let's meet people once a year" [13:10] mterry, I would say it's a sort of a distro rally for GNOME, in a conference format (I wouldn't say UDS like before the sessions format is rather plenary like than work oriented) [13:10] before->because [13:11] mterry, but it's good to meet everybody, get face to face discussion, and do some hacking on the side [13:12] hmm [13:14] It happens at an inconvenient time for me, so I am a bit torn [13:17] didrocks, you also go to GUADECs a lot, right? [13:18] mterry: yeah, I try to, why? :) [13:18] oh backlogging ;) [13:18] mterry: yeah, seb128 gave a good description [13:19] mterry: don't expect to be able to ack a lot or get technical discussions, it's more someone gives a talk on something [13:19] So good to keep up with what people are doing in GNOME [13:19] mterry: exactly === MacSlow|lunch is now known as MacSlow === ogra_ is now known as ogra === ogra is now known as ogra_ [16:40] hey didrocks, I'm trying to build unity-panel-service from trunk. cmake doesn't run through because libnux-2.0-dev is missing, which I can't install because I have "held broken packages" [16:41] libboost1.46-dev to be precise [16:41] larsu: on what are you on? precise, quantal? [16:41] quantal [16:41] larsu: if you apt-get install libboost1.46-dev, what does it tell to you? [16:42] didrocks, it wants to remove 70mb of packages [16:42] interesting, I'm sure you want to do that :) [16:42] can you pastebin that? [16:42] yup [16:42] it seems to be boost mostly [16:42] *shudder* [16:43] didrocks, http://paste.ubuntu.com/1042620/ [16:43] ah you do have installed libboost 1.49 [16:44] so yeah, it's one boost stack or the other :) [16:44] unity with boost 1.49 is still no uploaded because of the now famous C++ libsig++ issue [16:44] I don't care which one, don't even know why I have the 1.49 installed [16:44] larsu: so just accept the change, it seems you won't loose anything :) [16:45] unity-panel-service should really be buildable without configuring all of unity [16:45] larsu: right, the source is really not modular [16:45] didrocks, yup, doing that now. Thanks! [16:45] larsu: yw :) [16:46] worked! thanks again didrocks [16:47] larsu: no worry! [17:01] * didrocks waves good evening and good week-end! [17:06] chrisccoulson, while I wasn't looking, the gdb run finished and crashed! [17:06] mterry, oh, that's fantastic :) [17:06] chrisccoulson, the previous two backtraces: http://pastebin.ubuntu.com/1042651/ [17:10] mterry, interesting. i wonder if the reason i couldn't trigger it here is because my VM doesn't have a touchpad [17:11] let me try writing a USB stick and booting on my actual hardware :) [17:11] if i can find USB stick [17:11] interesting, good point [17:19] ok, bbiab [17:38] mterry, ok, bingo! [17:38] you can work around it btw, by turning off "disable touchpad whilst typing" [17:39] * mterry has come to love his unthemed GTK+ [17:39] heh [17:48] mterry, ok, i've updated the bug now [17:53] chrisccoulson, awesome, thanks dude [17:54] mterry, i might continue looking at this on monday. not sure if i want to bury my head in assembler on a friday evening ;) [17:55] :) === m_conley is now known as m_conley_away