=== eeejay is now known as eeejay_is_afk [01:13] desrt: I had [01:14] jbicha: there? [01:15] cyphermox: howdy [01:15] hey :) [01:15] jbicha: so you tried the Autostart condition? what did you use for a condition? [01:16] I just used AutostartCondition=GNOME3 unless-session gnome in /etc/xdg/autostart/nm-applet.desktop [01:31] ok... I'll need to look that up [01:34] ah, I see mterry mentioned it in an email before to u-d [01:34] amazing :D [01:34] * cyphermox logs out to test [01:45] jbicha: yup, seems to be covering all cases [02:18] I've got a heck of a question. [02:20] Is there currently a way to install gnome 3 on Lucid? [02:21] That's unlikely to be a particularly fruitful endevour. Why do you want to? [02:22] I just wanted to try it out. Something different I guess. [02:23] I tried gnome-shell and thought it was pretty cool so I wanted to see if I could install the whole environment [02:23] I'd try a livecd, I think. [02:23] What distro is on the live cd? [02:23] Or, rather, a liveUSB, with persistence. [02:23] ben_: Ubuntu 11.10 is the first release to officially support GNOME 3 and it's a very easy install to get GNOME Shell there [02:24] there's a PPA for 11.04 that might work, but trying on Lucid is probably not worth the effort [02:25] it'll more or less break your Classic desktop since indicators don't work with gnome-panel 3 yet [02:25] Which DE have you guys been working with on your setup? [02:27] Oh I see. It's opensuse [04:17] Good morning [04:23] Morning pitti. [04:31] jbicha: oh, caribou got rejected, did you get an email why? [04:31] pitti: we rejected it because Debian made changes yesterday that they wanted included [04:31] ah, ok [04:32] I updated my branch, but bigon has another bug he plans to look at today before we upload again [04:32] jbicha: and then we'll just sync instead? [04:33] jbicha: I made a small fix to the copyright file [04:33] pitti: no because it looks like caribou won't build with GNOME 3.0 so I don't think it'll be pushed to experimental soon [04:33] we fixed the copyright file too [04:38] pitti: hi [04:38] pitti: I am waiting for a freeze exception to be cleared before merging a patch in Unity 2D [04:38] hey Kaleo, how are you? [04:38] pitti: https://bugs.launchpad.net/unity-2d/+bug/654988 [04:38] Ubuntu bug 654988 in unity-2d "Unity doesn't mirror its interface for RTL locales." [Medium,In progress] [04:38] pitti: very good, thanks [04:38] pitti: I am afraid I might have missed a step in the process of the freeze exception [04:39] pitti: basically I don't know where to look for an answer about it (acceptance or not) [04:39] pitti: it's also really early in the morning so my brain is slow too :) [04:39] Kaleo: hm, given how old this issue already is, it seems a bit risky to introduce that past beta-2? we are in rather deep freeze now [04:40] Kaleo: there hasn't been any recent comment to this bug which would trigger RT attention [04:40] pitti: I understand, I am confident that the patch won't introduce regressions for LTR languages [04:40] it was opened in January or so [04:40] pitti: yeah, I see [04:40] Kaleo: right, I mean for RTL - it's quite a disruption there? [04:40] pitti: yes, though I think a very welcome one [04:41] pitti: code wise it's fairly minimal since we are using Qt/QML's infrastructure for it [04:41] Kaleo: if I understand right, this will mirror RTL on Unity 2d but not for Unity 3D, that might be enough reason to wait... [04:41] jbicha: that's correct [04:42] that', too [04:42] jbicha: on the other hand we have other deltas such as a big a11y delta between 2D and 3D already [04:42] jbicha: meaning that if we always have to wait for the other version before fixing one version it can get tricky [04:42] jbicha: though in this case it is a fairly big difference visually for RTL [04:44] do you think it's possible to merge it tentatively? that is, to merge it and give it a week in the distro and revert it if there are significant issues? [04:44] (it has been tested over a couple of weeks by mardy and myself) [04:44] Kaleo: I'd be more comfortable if this could be put into a PPA with a big call for testing, and some responses from RTL users [04:44] so I guess the question is whether it's enough of a bug for RTL users that they won't mind that Unity 3D works so differently [04:45] and the consistency issue is a rather bad blocker for this as well [04:45] pitti: can totally do a ppa [04:45] Kaleo: perhaps you can add the feature, and turn it off by default, and document a way to enable it? [04:45] pitti: hmmm that's a good idea [04:45] Kaleo: then this could even land in the distro, and the call for testing would be easier [04:45] pitti: that's the best idea! [04:45] pitti: so it is off by default the way it's done [04:45] and for people who want it they would at least have a way to use it [04:45] and for P we then turn it on for both 2d and 3d? [04:45] pitti: it is activated when a certain string is set in the translation [04:46] oh, a single, e. g. Hebrew, string is enough to enable it? [04:46] pitti: yes [04:46] shouldn't it look at your locale rather? [04:46] but well, maybe that's the right thing, I'm not sure [04:46] pitti: I am no expert in the domain, maybe [04:46] pitti: mardy may know better [04:47] pitti: but in our case it's a good thing :) [04:47] jbicha: do you think the idea is good? [04:49] Kaleo: shipping it disabled by default won't cause any problems and it's definitely something that should be fixed for P [04:49] is that RTL translation string thing standard for Qt? [04:51] jbicha: i would need to check with mardy [04:54] Kaleo: ah, found the documentation, looks like you need to add a line for the translators to have something to translate [04:54] http://doc.qt.nokia.com/4.7-snapshot/qml-righttoleft.html#default-layout-direction [04:56] jbicha: right thats what we do [04:57] thanks a lot guys [04:58] pitti: thank you [04:58] nice [04:59] it looked like you just checked whether the value was set but didn't provide the string for translators to use [04:59] it's cool that RTL is built in to QML like that :) [05:00] jbicha: thats correct; lets merge it and do a call for testing with instructions how to do that [05:40] good morning [05:44] RAOF, what sets the default X keyboard layout? [05:45] udev attributes [05:45] is it ubiquity? [05:45] RAOF, but udev can't possibly know the layout right [05:45] console-setup and X.org set it from /etc/default/keybaord [05:46] ubiquity/d-i write that file, yes [05:46] Ah, right. You're after policy, not mechanism :) [05:46] RAOF, thanks [05:46] good morning didrocks, robert_ancell, pitti, RAOF [05:47] pitti, thanks too [05:47] hey rickspencer3 [05:47] beta2 today? What's the word on the street? [05:47] Morning rickspencer3! How's your laptop going? [05:47] hi RAOF except for my zompbie nvidia chip, it works perfectly [05:48] hey rickspencer3 [05:48] taking me a bit to adapt to the next keyboard, which is not helped that it's a French keyboard and I'm using a US mapping [05:48] rickspencer3: you really brought an azerty layout laptop? :-) [05:48] didrocks, yes I did [05:48] :-) [05:49] nice, you are on the right way ;) [05:49] hah [05:49] I wish [05:49] And you look at the keys? :) This is a German keyboard, and it's fine for me :) [05:49] time to break the keyboard layout switcher ;) [05:49] pitti, how does Beta 2 look? [05:49] rickspencer3: after many respins it's now behaving [05:49] didrocks, nah, I don't much look at the keys, but it's still a different size and all [05:50] pitti, were the respins mostly those ubiquity bugs, or were there other things going on? [05:50] like the ca-certs thing [05:50] rickspencer3: mostly ubiquity, but the ca-certificates one was nasty, too [05:51] pitti, do you still have to be connected to the network to run Ubiquity? [05:54] rickspencer3: no, that was fixed (reason for two respins) [05:54] ah good [05:54] that was quite annoying [05:54] pitti, I guess the gsd thing was fixed while ev was in there? [05:54] * rickspencer3 did 2 installs 2 nights ago [06:13] rickspencer3: yeah, I debugged it last morning, and then we got it in quickly [06:14] sweet [06:14] I am *loving* LightDM [06:14] the new greeter looks so good [06:15] and it's so snappy [06:17] GDM 3.2 feels like GNOME Shell so it probably would have been a huge headache to try to Ubuntu-ify it [06:23] bah, package updates become hard to do now; we need to unfreeze [06:26] but versions.html considering UNAPPROVED greatly helps [06:27] hey jbicha! [06:28] pitti: hmm, I thought we're staying frozen until the end [06:28] micahg: until it's 100% sure that we aren't going to respin [06:28] I think that's some point during the day [06:28] pitti: no, I thought until final release [06:29] uh, for three weeks? [06:29] yep [06:29] I thougth we only did that between RC and final [06:29] * micahg too, until I was reminded of last cycle... [06:30] pitti: http://irclogs.ubuntu.com/2011/09/16/%23ubuntu-meeting.html#t15:55 [06:31] micahg: ah, thanks [06:31] micahg: I wasn't in last week's release meeting, so that escaped me [06:31] yeah, the idea was kicked around in -release earlier last week [06:31] that'll require a lot of handholding from the release team [06:32] Hm. That's what I assumed would be happening anyway. [06:33] pitti: we actually don't have an RC this cycle, so kinda makes sense [06:39] jbicha: oh, you can't upload g-shell? [06:40] ah, universe, I guess [06:40] * pitti will sponsor your branches [06:42] pitti: gnome-shell won't build though until we add caribou [06:42] jbicha: did you test the shell build? [06:42] jbicha: the new configure.ac has [06:42] +PKG_CHECK_MODULES(BROWSER_PLUGIN, gio-2.0 >= $GIO_MIN_VERSION json-glib-1.0 >= 0.13.2) [06:42] jbicha: but there is no json-glib-dev build dep [06:44] I wasn't sure how to pbuilder it since I needed new versions of packages [06:45] jbicha: I usually check a configure.ac diff between the old and new versions to see bumps and new build deps [06:47] I think I finally have a reliable reproducer for that annoying z-order stacking bug in Unity [06:47] 2 days after it's been fixed, I think :/ [06:47] anyway, the recovery dialog from LO seems to trigger it [06:47] jbicha: so, I'll just add libjson-glib-dev after the merge, ok? or do you want to do it in your branch? [06:48] jbicha: https://wiki.ubuntu.com/Packaging/Training/Logs/2009-04-16 [06:48] rickspencer3: questions I was asking you before you quit, you have the compiz ppa version isn't it? [06:48] didrocks, no [06:49] this is an fresh install from 2 nights ago [06:49] rickspencer3: there is a compiz uploaded on the ppa which should fix stacking issues [06:49] pitti: it looks like mutter pulls in clutter which pulls in libjson-glib-dev [06:49] didrocks, do you want me to try it? [06:49] rickspencer3: compiz and c-p-m, seems to fix major issues [06:49] rickspencer3: yes please [06:49] jbicha: ah, ok; but I'll add it anyway, safer [06:49] pitti: ok, no problem [06:49] didrocks, what's the url for the ppa? [06:49] rickspencer3: as those are very intrusive changes, we pend the tests between multiples people while we were frozen [06:50] rickspencer3: https://launchpad.net/~ubuntu-desktop/+archive/ppa [06:51] rickspencer3: What z-order bug are you thinking of? The one where the focused window isn't raised to the top of the stacking order? [06:53] RAOF, it's where the alt-tab strip and the dash and the launcher appear under windows on the desktop [06:53] Oh, _that_ one. [06:53] I just found that it starts after the LO recovery dialog first appears [06:54] I think the fix for that might have broken my alt-tab, but no one else seems to be able to reproduce. [06:54] didrocks, after I add the PPA do I have to reinstall unity, or will a dist-upgrade just do it? [06:55] rickspencer3: even just an upgrade should work [06:55] rickspencer3: this will bring you others packages though [06:55] like a speedy g-s-d [06:55] and fixed Qt 4.7.4 for vlc and some other apps [06:55] didrocks, urk, I'm getting a 404 from the ppa :/ [06:55] on apt-get update? [06:55] yes [06:56] I can reach it from there [06:56] I just ran update and got a 404 from main too! [06:56] pitti: can you sync anjuta-extras from sid? the version in oneiric conflicts with our anjuta [06:56] jbicha: doesn't seem to be on any image, sure [06:56] rickspencer3: seems you are more in trouble then, do you have Internet on that box? (is it the one you have IRC?) [06:56] didrocks, yes :) [06:56] it's only main and the desktop PPA that are giving me a 404 [06:57] GRAGHoughheuu splutter splutter. Having focus not return to the previously focused window in a group drives me crazy. [06:57] rickspencer3: for main, french or main archives? [06:57] didrocks, I'm sure there are some temporary server side issues, I'll try again in a bit [06:57] rickspencer3: fr.archive or archive [06:57] rickspencer3: yeah, let's pend that to a later try [06:57] didrocks, no, I switched to archive because the fr were often not working for me [06:58] what, fr always works ;-) [06:58] on strike too often [06:58] on archives strike were scheduled :p [06:58] pitti: that's soooooooo easy! [06:58] maybe I was always trying between noon and 2pm? [06:59] roh, come on! :-) [07:00] rickspencer3: if you go to descent cities like Lyon, most of shops aren't closed between noon and 2pm :-) [07:00] didrocks, I like it here, that they take things easy [07:00] and the grocery store and bakery near us stay opened [07:00] so, not really complaining [07:01] rickspencer3: indeed, that the big advantage of where you live, life is cooler I guess :) [07:01] (same in the very south of France) [07:01] indeed [07:01] most people are relaxed and friendly here [07:01] more so then in Seattle, for sure [07:02] you won't find that in Paris :-) (I would tell Lyon is average, between Paris and Toulouse, but less relaxed than Toulouse) [07:02] except on Sunday morning when there is a line at the bakery and you don't understand what they are asking you when they won't to know what specific type of criossant you want [07:02] in these cases, they are less relaxed [07:04] ahah, you should just point with your finger like "euh… euh… I want that" [07:05] rickspencer3: you should know that a "baguette" has as well different naming depending on which city you live [07:05] jbicha: done [07:06] didrocks, hmmm, fortunately they understand me when I say "baguette" here [07:06] then I don't understand them when they try to clarify which of the 5 types I want [07:06] but then they understand me when I point and say "set baguette" [07:07] rickspencer3: just ask for the regular one, it's the best :) "baguette tradition" is just a trap to pay it at a higher cost for no real additional value ;) [07:07] didrocks, I like the ones covered in salt sometimes :) [07:07] and they are the same .80 euro price [07:07] Bring your own salt? [07:07] with salt? not flour? [07:08] I imagine Rick can tell the difference between salt and flour :) [07:08] pitti: I wonder if we should just upload caribou, it might cause problems if people update to mutter without also upgrading to the corresponding gshell [07:09] RAOF, indeed, as I told seb128, the first thing I sorted out was how to get food [07:09] salt on top of a baguette, never saw that :) sounds healthy :) [07:09] Most correct! [07:10] didrocks, and then it's deep fried. [07:10] bryceh: of course! mostly needed :-) [07:10] Urgh. Could indicator-datetime-service kindly not spam DBus, causing compiz to be terribly unresponsive? [07:11] RAOF: at least, now, you can click on the calendar and change the month! [07:11] OH WOW! [07:11] I hadn't noticed that it now works. Awesome! [07:11] RAOF: yeah, after 5 months! you can have calendar again :-) [07:12] RAOF: more seriously, on the spammy datetime, tedg discussed about it, but he has no solution right now [07:12] didrocks, oh yay! That's been bugging me for ages [07:12] robert_ancell: yeah, I totally rejoice when I saw that being released :) [07:13] didrocks: Does he know what's going wrong? It really really degrades any interaction with unity. [07:13] didrocks: nice! finally I can stop using cal(1) [07:13] mvo: ahah, you were using it as well? was my only way when I needed a calendar ;) [07:14] Who is looking after the network-manager-applet package these days? I have a packaging bug to file, but I need to talk about it with a human first. [07:14] RAOF: yeah, he knows about it, I don't have the detail in memory, but I'll try to reping him later today [07:14] mvo: hey, you're fast! awesome :-D [07:14] didrocks: same here. I have the new oneconf now and everythng is working perfectly. I get diffs and the new machine shows up almost instantly, really really nice [07:14] mvo: so which one of the structs members should I use? [07:14] kamstrup: you did all the hard work, so that was easy :) [07:14] mvo: \o/ what a change compared to the desktopcouch version! :-) [07:15] kamstrup: depends on what you need, ratings average (the first) if you need to do the filtering. but sort the sorting I would recommend the last one [07:15] mvo: I'm really eager to integrate ubiquity next cycle and all what we can do with that, crossing fingers ;) [07:15] mvo: ok, i'll try with the first one and then experiment a little, thanks! [07:15] kamstrup: we basicly use http://www.evanmiller.org/how-not-to-sort-by-average-rating.html for the sorting to avoid that the 1x5 star app floats on top [07:16] kamstrup: that is the last member of the struct [07:16] didrocks: yeah, that is going to be awsome, especially with the "keep-home" feature [07:17] indeed, you just keep home and you can reinstall all your apps in a breeze :) [07:17] yes [07:17] * mvo hugs didrocks [07:17] mvo: btw, thinking about it, do you think it will be hard to add a new renderer for the gtktree (AppStore2) to add in the categories "install all"/"remove all" [07:17] * didrocks hugs mvo [07:18] kamstrup: please tell me if everything works as expected, if so I will merge and upload after the freeze [07:18] mvo: not sure it can still enters oneiric… but at least for p [07:18] good morning everyone [07:18] didrocks: I don't think it will be very hard, but everything is a bit hard with the gtk.TreeView [07:18] hey chrisccoulson [07:18] hi mvo, how are you? [07:19] sleepy, but otherwise good. and you? [07:19] mvo: yeah, I'll try to have a look for that for P. Apart from that and an easy way to remove a computer from the list on don't have anymore, I don't see anything else for usc itself right now [07:19] * mvo nods [07:20] mvo epic quote "I don't think it will be very hard, but everything is a bit hard with the gtk.TreeView" [07:21] pitti, hey, do you know if seb128 is in today? [07:21] kamstrup: well, that's not that untrue with treeview :) [07:22] heh :) [07:22] robert_ancell: he should be [07:22] didrocks: not only "not untrue", but painstakingly true! [07:22] pitti, good morning [07:23] hey ricotz [07:23] mvo: how do I trink it to dump a db? [07:23] kamstrup: I think we should make pain and stacking synonyms now :-) [07:23] lol [07:23] pitti, i noticed some uploads to the queue which might be uploaded after an clutter/cogl upload [07:24] kamstrup: just get the branch and run s-c from it once (via ./software-center) and it should appear a couple of seconds later [07:24] pitti, http://people.ubuntu.com/~ricotz/clutter/ [07:24] pitti, ok, good. I have a unity-greeter which is in the upload queue, can you have a look at that. There's a lightdm release pushed to bzr, but I'd like seb128 to take a look over it today before uploading and we need to resolved bug 845549 before release. Can you also look at it and see if it's ok to upload pending that fix? [07:24] Launchpad bug 845549 in lightdm "Do not ship /etc/lightdm/lightdm-gtk-greeter.conf" [High,Fix committed] https://launchpad.net/bugs/845549 [07:24] mvo: thanks, i see it now :-) I was just running ./run_local.sh thinking I was a l337 haxor [07:26] hey [07:26] salut seb128 [07:27] robert_ancell: summoning powers! [07:27] lut didrocks [07:27] hey pitti robert_ancell [07:27] ricotz: you need these sponsored? [07:27] pitti, how are you? [07:27] seb128: quite fine, thanks! [07:27] kamstrup: heh :) [07:28] what's the beta2 status? [07:28] hey mvo [07:28] arrived late, but seems ok now [07:28] pitti, yes, but they should be built and published before gnome-games, g-s, mutter [07:28] ricotz: why, did these forget to bump their build dependencies? [07:28] hey seb128! [07:28] and cheese [07:28] ricotz: or does the new clutter change ABI? [07:28] pitti, no, cogl bump its soname [07:29] mvo: can you add the libdb version to the db path name? [07:29] pitti, clutter is stable, but cogl is also in their rdepends [07:30] kamstrup: sure, I will do that, I have a call now, I will do it right after that [07:30] mvo: you have db.DB_VERSION_{MAJOR,MINOR} [07:30] mvo: sure [07:33] pitti, was the summoning powers for me joining? robert_ancell looking for me? [07:33] seb128: yes, robert_ancell asked for you [07:33] robert_ancell, hello! [07:33] seb128, hey. I think I've got all the issues you had fixed, it's just bug 845549 blocking the release to B2 now [07:33] Launchpad bug 845549 in lightdm "Do not ship /etc/lightdm/lightdm-gtk-greeter.conf" [High,Fix committed] https://launchpad.net/bugs/845549 [07:34] robert_ancell, well it's too late by a week for beta2, we are frozen since friday [07:35] seb128, I was hoping to ask the release team nicely :) [07:35] robert_ancell, that's a packaging change though, I would just undo your additional binary and not block 0.9.8 on it [07:35] we can fix that bug in -0ubuntu2 [07:35] seb128, ok, I'll do that [07:35] robert_ancell, well, you can but beta2 is supposed to be today and I doubt they will respin isos for lightdm and restart the testing from 0 [07:36] oh, is that already started? I thought that started tonight [07:36] robert_ancell, btw did you get an ecryptfs login issue as well? [07:36] ok, np, it can all be post b2 [07:36] yes [07:36] robert_ancell, no, beta2 is supposed to be out tonight :p [07:36] robert_ancell, i.e validated and published, isos started being rolled on monday [07:37] robert_ancell, you really need an ecryptfs testcase ;-) [07:37] or to convert your main user to it :p [07:37] seb128, yes, but I can't test it in the regression tests [07:37] stupid PAM is a pita [07:37] ok [07:38] robert_ancell, well in any case imho you should roll back the fix for bug #845549 and wait for the patch to use alternatives [07:38] Launchpad bug 845549 in lightdm "Do not ship /etc/lightdm/lightdm-gtk-greeter.conf" [High,Fix committed] https://launchpad.net/bugs/845549 [07:38] then you can upload 0.9.8 [07:38] ricotz: mutter is already accepted, BTW [07:38] ricotz: so we might need another upload for that and bump build deps? [07:38] I will grab it from the queue if you upload today still, otherwise will do a trunk build for testing [07:39] robert_ancell, what bugs are still on your list for oneiric? getting close now ;-) [07:39] I have 8 I'm tracking now [07:39] pitti, a simple rebuild will do, the required versions hasnt change, just to pick up the new lib [07:39] robert_ancell, do you tag those or milestone them or something which makes easy to see the list for me? [07:40] seb128, should I just use the ubuntu 11.10 milestone? [07:40] robert_ancell, works for me [07:40] robert_ancell, btw you probably need a ffe bug for 0.9.8 [07:41] since you added the language selector [07:41] it will probably be easy to get an ack since it's optional on a greeter we don't use by default (but that some derivatives use) [07:42] morning [07:42] seb128, ok, all tagged and assigned to me [07:42] robert_ancell, thanks [07:43] later all [07:44] robert_ancell, 'night [07:46] oh, jbicha doing updates in Debian now ;-) [07:47] jbicha, hey, I noticed that you have a pitivi ffe bug waiting for a comment from you [07:47] jbicha, i.e pitti asked some question a week ago and you didn't reply [07:47] hey rodrigo_, how are you? [07:47] hi seb128 [07:47] hey rodrigo_ [07:48] hi pitti [07:48] ricotz: oh, cogl ABI bump -- that requires an FFE [07:48] hmm, did someone do libgnomekbd update? [07:48] rodrigo_: yes, in unapproved [07:48] ah ok [07:49] I marked it on the pad, didn't I? [07:49] and it's green on versions.html [07:49] yeah [07:49] pitti, not sure cogl broke ABI, they just decide to keep going for the unstable cycle and bump the soname for the stable [07:50] seb128: well, it's a library transition, with some rdepends, so we shoudl have a bug for trackign the transition now [07:50] well, soname bump is an abi break I guess [07:50] pitti, right [07:50] quadrapassel, mutter, cheese, champlain, gnome-shell, AFAICS [07:50] pitti, I just wanted to say that the transition should be easy, they didn't break api [07:51] but agreed, that seem some tracking [07:51] seb128, pitti, they moved some stuff to there experimental api, so nothing was actually removed, but yeah it is an break [07:51] pitti, btw davidz did an api change in gdbus-codegen, we should track it as well [07:52] pitti, seb128, but it doesnt break/ftbfs any of its depends [07:52] ricotz: right [07:53] ricotz: it's fine, but please file an FFE bug anyway, so that we can add the tasks needed for rebuild, etc. [07:53] ok [07:53] ricotz: both uploaded, thanks! [07:55] pitti, thanks [07:57] hey [07:57] pitti, ricotz I've some trouble with how in caribou the gtk module is loaded [08:00] rodrigo_: re: the System Info Checking for Updates bug, line 1880ish doesn't look right to me [08:00] http://git.gnome.org/browse/gnome-control-center/tree/panels/info/cc-info-panel.c#n1856 [08:00] jbicha, looking [08:01] I think the Xsession script is not needed as there is a .desktop file that is installed in gnome-setting-daemon that seems to load it already [08:01] rodrigo_: just a guess, I'm not much of a C coder nor do I understand what parts of PK Ubuntu has by default [08:02] hey seb128, how are you? [08:02] jbicha, it has that interface, yes [08:02] pitti, https://bugs.launchpad.net/ubuntu/+source/cogl/+bug/856179 [08:02] Ubuntu bug 856179 in cogl "[FFe] cogl 1.8.0 with soname bump" [Undecided,New] [08:02] checking anyway [08:02] chrisccoulson, hey, I'm great thanks, how are you? [08:02] rodrigo_, it seems federico agrees with me about removing the refresh in gnome-desktop when applying a display configuration :) [08:02] seb128, yeah, i'm good thanks [08:02] chrisccoulson, cool! [08:03] bigon, in what way? [08:03] so, i'll remove that. that wins us back some time for people who configured display settings :) [08:06] chrisccoulson, \o/ [08:06] chrisccoulson, I need to double check if you update broke something there [08:06] jbicha, hmm, indeed we don't, I guess that was changed, because it was using an interface session-installer implements [08:06] chrisccoulson, I've a "weird","buggy" case [08:06] seb128, oh? [08:07] pitti, please add libchamplain-0.8 too [08:08] mmh? what about champlain-0.8? [08:08] hmm, weird, has session-installer been changed? [08:08] it used to implement the GetUpdates method, iirc [08:08] but d-feet doesn't show it [08:08] bigon, it needs an rebuild for the cogl soname bump [08:08] oh ok :) [08:08] chrisccoulson, so for some reason I can't boot my laptop docked since I updated this week, I see the purple color but lightdm never displays and I can't switch to vts [08:09] chrisccoulson, the screen is on but nothing display [08:09] well anyway that's another issue [08:09] boot works fine undocked though [08:09] hey chrisccoulson [08:09] so I boot, log in, then dock my laptop [08:09] doing that give me a 1440 mirrored resolution [08:09] but when I open the xrandr capplet it switches back to the correct config [08:09] i.e 1920 [08:10] (which is my external monitor resolution, I work with lid closed) [08:10] hmmm :/ [08:10] so I'm wondering if it applies the stored config when I open the capplet and if it should do it when I dock the laptop rather [08:10] but I'm not sure that worked before [08:10] that seems to happen to me too, but it's not a new problem for me [08:10] ok, so maybe it's not yours [08:10] I got the "hang" issue on the same day I installed your updates [08:11] I never booted undocked and docked it this way before [08:11] so I don't know if that ever worked [08:11] ie, docking my laptop always seems to get me a clone config, which either corrects when i open the display settings, or requires me to manually configure the display [08:11] ok [08:11] good, so not a new bug [08:11] i'll have a look, just to make sure. but i don't think it's a new bug [08:11] still seems weird that opening the display settings do the right thing [08:11] it should do whatever it does there when the laptop is docked ;-) [08:11] Sounds like hotplug isn't being triggered properly. [08:12] Because opening display settings triggers a full xrandr probe, it should always pick up anything that you've got plugged in at that moment. [08:12] i guess it would be easy to check if gnome-settings-daemon ever gets a notification about the display config changing [08:13] 10:03 < ricotz> bigon, in what way? << well I think that the one of the desktop file in the caribou bin pkg make gnome-setting-daemon load the gtk module [08:13] That's not what to look for; it's the udev events that you should look for. [08:13] so the hack to add it to the evironoment variable in Xsession.d is not needed [08:13] but that will require to change where that file is installed [08:14] Hm, actually, I'm not exactly sure what X events should trigger on hotplug. Certainly X shouldn't actually *configure* the newly discovered displays, just notify that they've appeared. It's g-s-d's job to actually set a mode on them. [08:14] (also I'm not sure if the caribou gtk modules shouldn't depends against caribou and not the opposite as it's done now) [08:15] RAOF, RRScreenChangeNotify [08:15] Ah, ok. [08:15] that's the event that gnome-settings-daemon is expecting to see [08:16] ricotz: /usr/lib/gnome-settings-daemon-3.0/gtk-modules/caribou-gtk-module.desktop [08:16] seb128, what's in your ~/.config/monitors.xml? [08:17] does it actually have a mirrored configuration in there? [08:23] chrisccoulson, http://pastebin.ubuntu.com/695000/ [08:26] seb128, hmm, so, it doesn't look like it [08:27] no [08:27] well maybe it's not in mirror when I dock it [08:27] it's hard to say since my lid is closed and opening the gui switches it to 1920 before I can see what were the settings [08:28] it might just be that the external monitor stays in 1440 [08:34] interesting [08:34] i'll have a look at this today, and see if there is anything obvious going wrong [08:34] but this hasn't really ever worked reliably for me [08:34] although i would be very happy if it did work reliably :) [08:35] doesn't seem a priority but would be nice to fix [08:35] indeed [08:37] pitti, will lots of people get spam if i make ubuntu-core-dev a member of https://launchpad.net/~mozillateam ? (and can i actually do that) [08:37] i just noticed that nobody (basically except for me, micahg and jdstrand) can actually commit to our branches [08:37] chrisccoulson: you need to set a contact email for mozillateam first [08:38] otherwise everyone will get spammed, yes [08:38] chrisccoulson: yes, you can invite, and the owner of core-dev can then accept [08:38] i guess ubuntu-mozillateam@lists.ubuntu.com would be an appropriate contact? [08:39] that sounds fine [08:40] pitti - oh, that address is already used for https://launchpad.net/~ubuntu-mozillateam [08:40] which doesn't really exist ;) [08:43] mvo, ping [08:43] chrisccoulson: hm, case for #launchpad? [08:43] to clean it up [08:44] pitti - yeah, just asked there now [08:48] rodrigo_: FYI, I reopened bug 853751, as it's not fixed in ubuntu yet; can you close it with the changelog? [08:48] Launchpad bug 853751 in gnome-control-center "http proxy setting blanked when user goes back to proxy settings" [High,Fix committed] https://launchpad.net/bugs/853751 [08:49] pitti, hmm, it's in 3.1.92 [08:49] rodrigo_: ah, ok; the last comment sounded otherwise [08:50] rodrigo_: closed again, thanks! [08:50] sounds like the bug I filed upstream [08:50] https://bugzilla.gnome.org/show_bug.cgi?id=658447 - want to close? [08:50] pitti, yeah, I should have mentioned 3-2 branch, not master [08:50] Gnome bug 658447 in Network "Clears proxy host when opened" [Normal,Unconfirmed] [08:50] kamstrup: added that now too, I will add some cleanup, tests and it should be ready later today (we are in freeze anyway) [08:51] rodrigo_: poing [08:51] rodrigo_: pong even [08:51] mvo, :) [08:51] mvo, you've worked on sessioninstaller, right? [08:51] yes [08:52] mvo, I thought it implemented the GetTid/GetUpdates interface of packagekit, but looking at the code I don't see it [08:52] mvo, so, I guess I have dreamed it? [08:52] and it worked for me because I have packagekit installed? [08:52] rodrigo_: this sounds like its the system api, not the session api? or am I missing something? [08:52] rodrigo_: that sounds likely [08:52] mvo, I thought it was the session API [08:53] ok, we need to patch g-c-c then [08:53] since packagekit is not on the CD [08:53] what is it doing with that info? [08:54] could you double check if its really part of the session api? if so, we could add it to s-i [08:54] mvo, on the system info panel in g-c-c, checks for updates, and enables a button to install those [08:54] mvo, yes, sure [08:54] it would actually be quite nice to have this on the session api [08:54] rodrigo_, mvo The glib client library of PackageKit uses the system bus directly [08:56] mvo, glatzor: yes, they're from the system api [08:56] mvo, rodrigo_ once I find some time to continue the work on the glib library for aptdaemon we could just port the plugin [08:56] glatzor, which plugin you mean? [08:57] the packagekit plugin of gnome-session-daemon [08:57] right [08:57] is there any plan to implement the PK system API? [08:57] in aptdaemon? [08:59] rodrigo_, but why port the plugin if we still have the universal ubuntu sessio daemon, formerly called update-notifier arround :) are there any plans to port functionallity from update-notifier to g-s-d? [09:00] glatzor, well, we already have a way to check and install updates in the system info [09:00] I'll port that to aptdaemon, if it's got a dbus interface, does it? [09:01] it does, yes [09:02] rodrigo_, currently we don't have got any aquivalent method to GetUpdates. The idea of aptdaemon was only to move the operation which require root privilliges to a separate daemon [09:02] right, I see [09:02] chrisccoulson, what's happening with bug #743096 [09:02] Launchpad bug 743096 in evolution-couchdb "contacts only sync with evolution not with thunderbird" [High,Fix released] https://launchpad.net/bugs/743096 [09:02] chrisccoulson, did you ship your u1 addressbook creating code with tb? [09:04] need to reboot, brb [09:04] rodrigo_, but we could think about providing a small subset of PackageKit API that we want to support by aptdaemon. for update checking it doesn't make sense to keep an apt cache in every active user session [09:08] glatzor: I see requests for this from various parties now, the indicator logout applet wants to know info about pending updates too, now g-c-c, it seems like it does make sense to provide info for some common operations [09:08] glatzor: especially since the libapt c++ api is pretty tricky to use from pure c apps [09:13] glatzor: I got approval for the add-license-key call, are you fine with me merging that code into trunk now ? [09:15] mvo, sorry. not yet. We currently don't have got a way to verify the key. see the latest comments on the bug [09:15] mvo, would it be possible to download the key by aptdaemon? [09:15] mvo, so we could rely on ssl and url validation [09:18] chrisccoulson: bug 805136, do you think that has an actual chance for oneiric? [09:18] Launchpad bug 805136 in thunderbird "accounts window doesn't fit on screen and no scrollbar to show hidden fields" [High,Triaged] https://launchpad.net/bugs/805136 [09:18] glatzor: indeed, its a oauth protected url [09:19] glatzor: let me think about this a little bit [09:23] didrocks: desktop-o-oneconf, "Wait for ISD to publish the webcatalog part" -> is that done now? [09:24] pitti: right, it's done, let me update it [09:25] didrocks: I can update it, just wanted to confirm [09:25] pitti: yeah, the spec is done [09:25] (done) [09:25] thanks pitti :) [09:33] mvo: all the pkgname keys in the db are prefixed with a ","... any reason? [09:34] kamstrup: no, I can fix that [09:35] kamstrup: its a relict of the past, it used to be that we stored the application name as well, but that is just too errorprone [09:35] mvo: ah, I see [09:35] mvo: but other than that, it works beautifully here :-) And very fast! [09:35] kamstrup: awsome! [09:36] mvo: looks like you're dumping the db from the main thread, before the UI is shown..? [09:36] or am I missing something? [09:37] kamstrup: yeah, its just proof of concept currently, I will optimize this before the final merge [09:37] mvo: oki [09:41] pitti, is apport collecting the .xsession-errors warning etc when the issue happen or when you run ubuntu-bug? [09:42] seb128: not for all bugs; hookutils has an xsession_errors() function [09:42] kamstrup: the code surrounding it is also a bit clumsy, I will take the opportunity to clean that up too [09:42] seb128: oh, which is called in generic hook, so it should be ther [09:42] pitti, right, but at what time is that called? [09:42] pitti, on the issue time (when the .crash is written) or when ubuntu-bug is run? [09:43] seb128: but it only shows lines which match (\(.*:\d+\): \w+-(WARNING|CRITICAL|ERROR))|(Error: .*No Symbols named) [09:43] to avoid catching passwords and the like [09:43] seb128: when ubuntu-bug is run [09:43] pitti, yeah, I've looked at the apport source and seen that [09:43] pitti, ok, that might explain why it's often useless [09:43] pitti, it collects the infos at the wrong time [09:44] pitti, well, something for next cycle, we have other things to do that improving apport before Oneiric now [09:44] I will keep it on my list [09:44] pitti, thanks ;-) [09:45] nice to test Kubuntu CDs from time to time and see how it looks like these days [09:46] pitti, we should get GNObuntu CDs to test how GNOME looks like these days ;-) [09:46] heh, right; I never built enough in jhbuild to get that far [09:46] only the basic libs (pygobject, gtk, etc.) [09:47] well, installing gnome-shell to Oneiric should be close enough from upstream GNOME [09:47] *nod* [09:47] I think stracciatella-session is pretty much obsolete, I think we should just remove it [09:47] oh, yes [09:48] g-shell isn't using indicators, or unity, or our panel layout, or what not [09:48] indeed [09:48] there is still notify-osd on the panel session, isn't it? [09:49] not sure if it's blackisted in gnome-shell [09:49] I thought shell had a builtin notification service? [09:49] they do [09:49] yeah, not sure if it's using dbus and so our service is spawn first [09:49] debian is dropped service activation from the notification daemons [09:50] letting those being started by autostart in their corresponding session [09:50] GNOME did the same for notification-daemon [09:50] ok, so it's their shell service which is started in the gnome-shell session? [09:50] didrocks, well, our service has a list of sessions where notify-osd should be running and gnome-shell is not in there [09:51] $ lp-remove-package.py -u pitti -m 'gnome-session-fallback and gnome-shell now do not have ubuntu modifications, and stracciatella-session does not work any more' stracciatella-session [09:51] * pitti sheds a tear [09:52] seb128: ineed, it's been updated, as I didn't touch, I wasn't sure there was still "gnome" listed in it [09:53] there is some cleaning still needed [09:53] and use DESTKOP_SESSION has well [09:53] but that can wait [09:55] didrocks, we should just drop the .service as others are doing and add an autostart with OnlyShowIn=Unity [09:56] mvo: w00t.Everything works very well now :-) It's great fun toying around with the ratings in the apps lens :-D [09:56] can work as well [09:57] kamstrup: :) [10:00] kamstrup: the ",pkgname" issue should be fixed now too [10:11] see you mvo I am leaving for work! have a nice day [10:13] thanks glatzor [10:13] glatzor: you too! [10:16] chrisccoulson, how did you profile g-s-d? did you have any clever way or basically markers in the code? [10:17] seb128, there's a disabled patch in the package in the PPA, and you also need to do --enable-profiling [10:17] (as a build option) [10:17] chrisccoulson, ok, well I want to profile nautilus so I was wondering if you used a tool I can re-use, seems not [10:18] guess I'm back to the markers in the code and strace ;-) [10:18] seb128, yeah, g-s-d already has code for profiling, which you can inspect with strace [10:18] i just modified it to print stuff out, with timestamps [10:19] brb. need to restart [10:25] need to change the plugs for my internet router, waiting for new delivery, continuing some oneconf offline work. [10:25] bbl [10:33] bah [10:33] I really wonder how much u1 test their stuff [10:33] the g-s-d integration is broken in oneiric, i.e the .so doesn't load, and nobody noticed [10:42] seb128, oh, this is a bit suspicous. when i dock my laptop, g-s-d doesn't just get a randr event, but it also gets a hotkey event from the display key :/ [10:43] i wonder where that comes from [10:46] yeah, that happens fairly consistently [10:47] chrisccoulson: do you have a log that shows the hotkey event? [10:47] tseliot, i've got the debug output from gsd [10:47] yep, that's what I meant [10:47] 1 second, jsut going to switch off the media-keys plugin, just to see if that keypress actually makes a difference [10:49] bingo! [10:49] stopping that keypress from happening causes gsd to correctly apply my stored config [10:49] seb128 ^^ [10:49] tseliot, http://paste.ubuntu.com/695039/ [10:50] not sure that tells you much though ;) [10:50] chrisccoulson, nice catch [10:51] that's the first time i've ever docked my laptop and it hasn't defaulted to a clone config, but actually applied my working config :) [10:51] chrisccoulson: I'm wondering if this is just because code that handles the hotkey event is also the same whithc handles RandR events [10:51] so, now to track down where this keypress comes from [10:51] tseliot, inside X? [10:52] chrisccoulson: no, in libgnome-desktop [10:52] tseliot, the keypress isn't handled by libgnome-desktop [10:52] it's handled in a separate plugin, and is a real X key event [10:52] chrisccoulson: gnome-settings-daemon then? [10:53] tseliot, yeah [10:53] i wonder if my dock does that? [10:53] perhaps i should just try pulling out the monitor cable [10:53] 1 second [10:55] there we go [10:55] so my dock is stupid [10:55] it's triggering that key event when i dock/undock [10:55] it doesn't if i just pull the monitor cable out [10:55] let me just verify that one more time [10:56] seb128: btw, I'll have another holiday tomorrow [10:56] seb128: I updated https://wiki.ubuntu.com/DesktopTeam/ReleaseStatus already, but I'm not sure whether we are going to have a meeting tomorrow [10:57] seb128: if we have one, would you be able to lurk, in case questions come up? [10:57] (I'll tell skaet and point her to the report) [10:57] tseliot, ok, totally confirmed - my docking station is a complete idiot :) [10:57] chrisccoulson: :) [10:57] i consistently get the display key event every time i dock [10:58] at the same time as the randr event [10:58] which breaks gsd :( [10:58] i wonder if that's the same for seb128 too [10:59] that's really annoying [10:59] tseliot, is there anything we can do about that? :) [10:59] it would be really awesome to not have to reconfigure my display every time i dock my laptop :) [10:59] pitti, can do [11:00] pitti: libreoffice_3.4.3-1ubuntu3 is on chinstrap [11:00] chrisccoulson: so, just to understand things better (as I've never used a dock), when you use the dock your external screen is used and the laptop's switches off or otherwise how does it work? [11:01] tseliot, i use both screens when i dock, with them both side-by-side [11:01] i have my settings in a stored configuration, because gsd defaults to cloning them [11:01] so, when i dock, i expect it to just see a randr event, and for gsd to happily apply my stored config (with the displays side-by-side) [11:02] but the extra hotkey event from my dock causes it to apply one of the default configs, with the first one being "clone" [11:02] chrisccoulson: right, that's what should happen but if you get a hotkey event, that makes it switch to a different configuration [11:02] tseliot, yeah :) [11:05] chrisccoulson: according to the log (if I'm reading it correctly), gsd is handling the hotkey event first and then it tries to apply your stored settings, so this should work, at least in theory [11:05] or maybe those events are triggered by the hotkey event? [11:06] pitti: hold on, there is one commit missing [11:07] yeah, i'm not too sure what happens there [11:07] chrisccoulson: you should probably check that before we can think of a workaround [11:07] * tseliot -> lunch [11:09] * didrocks grrr at UPS telling they arrive in 10 minutes for making the swap and finally arrived 50 minutes later [11:10] i think i might be better off dusting off xtrace here [11:11] pitti: fixed === MacSlow is now known as MacSlow|lunch [11:16] Sweetshark: libreoffice_3.4.3-1ubuntu3_source.changes, right? [11:17] hmm so, what needs to be done to apply for MOTU? [11:17] pitti: right [11:18] Sweetshark: uploaded [11:19] rodrigo_: https://wiki.ubuntu.com/DeveloperMembershipBoard/ApplicationProcess [11:19] didrocks, thanks [11:19] rodrigo_: yw! [11:22] pitti: thanks [11:22] * Sweetshark lunches [11:26] mmm, lunch, me 2 [11:26] * chrisccoulson gets hungry too [11:32] hum, it's lunch time indeed! [11:44] can I get some endorsements at https://wiki.ubuntu.com/RodrigoMoya/MOTUApplication please? [11:44] (I'll pay it back with beer :) [11:45] btw, for MOTU you get per package permissions, or just permissions for all packages? [11:46] rodrigo_, will comment on it, you get access to "universe" [11:47] ok, so every package there? [11:47] yes [11:47] ok cool [11:47] * rodrigo_ prepares to break lots of stuff [11:47] they will like you there :p [11:47] :) [12:19] lunch, bbl === MacSlow|lunch is now known as MacSlow [12:28] * pitti reviews unapproved [12:28] chrisccoulson: did you upload the g-s-d and g-desktop fixes yet? *yearn* [12:28] pitti - not yet, i was waiting on comments from upstream [12:28] ah, ok [12:29] and upstream recommended that i do to gnome-desktop what i wanted to do in the first place :) [12:29] (ie, get rid of the useless reprobe when applying a saved config) [12:29] which is nice :) [12:33] seb128: if you double-click a jpg file, does it open with eog or with shotwell? [12:33] I get shotwell here, which seems a bit odd [12:33] pitti, should be eog [12:34] pitti, you might be hitting https://bugzilla.gnome.org/show_bug.cgi?id=658188 [12:34] Gnome bug 658188 in gio "g_app_info_set_as_last_used_for_type generates a broken mimeapps.list" [Normal,Unconfirmed] [12:35] ah, that again [12:35] pitti, do you have that file in .local/share/applications? [12:35] rodrigo_, did you want to update the gsd branch with your tidied up patch? [12:35] pitti, if yes what does it contain? [12:35] seb128: right, indeed [12:35] http://paste.ubuntu.com/695083/ [12:36] pitti, there you go [12:36] * pitti nukes both files [12:36] pitti, I've been discussing it with desrt and some other GNOME guys [12:36] let's see if it gets fixed for this cycle [12:36] happy again now [12:37] great [12:37] thanks! [12:37] yw ;-) [12:44] waouh [12:44] early unfreeze \o/ [12:45] seb128: no, not thawed [12:45] seb128: apparently last week's release meeting said that oneiric will not thaw any more [12:46] seb128: but the b2 images are as good as they are going to be, so I pushed a bit for reviewing/flusing unapproved [12:46] pitti, sorry, wrong word, "early queue relax" \o/ [12:46] * didrocks will pushes compiz then [12:46] push* [12:46] didrocks, \o/ [12:47] wth [12:47] for now I'm still only accepting packages which I'd trust not to break the images, though [12:47] also, I need pkgbinarymangler and vala-0.14 to publish to accept some more [12:47] nautilus open and read like 10 times the default wallpaper on start [12:47] no wonder it's so slow [12:47] ugh [12:48] investigating... [12:48] that might explain why it balooned from 0.5 to 5 seconds [12:48] (see my wiki page) [12:48] seb128: nice catch! [12:48] yeah, that's where I come from ;-) [12:48] well wiki and charts I did [12:48] hopefully, it reads 10 times from my cache :p [12:49] didrocks, [12:49] 1316692376.273702 access("file:///usr/share/backgrounds/warty-final-ubuntu.png", F_OK) = -1 ENOENT (No such file or directory) [12:49] 1316692376.273765 getcwd("/builddir/nautilus", 4096) = 37 [12:49] 1316692376.273816 open("/builddir/nautilus/file:///usr/share/backgrounds/warty-final-ubuntu.png", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) [12:49] 1316692376.273898 open("/usr/share/backgrounds/warty-final-ubuntu.png", O_RDONLY|O_LARGEFILE) = 24 [12:49] 1316692376.273948 fstat64(24, {st_mode=S_IFREG|0644, st_size=295416, ...}) = 0 [12:49] 1316692376.274009 read(24, "\377\330\377\340\0\20JFIF\0\1\2\0\0d\0d\0\0\377\354\0\21Ducky\0\1\0"..., 8192) = 8192 [12:49] seems not [12:49] it does that every time [12:49] seb128: does it still use gnomedesktop? [12:50] seb128: I tried to tweak my cached background and it's the one which is used [12:50] didrocks, yeah, but that's not even my background [12:50] my user has another image [12:50] it reads the default one... [12:50] ah nice! :-) [12:50] so I think it's just bog [12:51] didrocks, you are still good I think [12:51] well your code rather ;-) you as well of course ... ;-) [12:51] good morning all [12:52] "1316692376.868893 lstat64("/usr/share/backgrounds/Bird_by_Magnus.jpg", {st_mode=S_IFREG|0644, st_size=84958, ...}) = 0 [12:52] 1316692376.869103 lstat64("/home/seb128/.cache/wallpaper/0_5_1920_1080_708d778bd7093a84d30ab7d51ad92436", {st_mode=S_IFREG|0664, st_size=694268, ...}) = 0 [12:52] 1316692376.869323 open("/home/seb128/.cache/wallpaper/0_5_1920_1080_708d778bd7093a84d30ab7d51ad92436", O_RDONLY|O_LARGEFILE) = 23" [12:52] didrocks, ^ [12:52] seb128: waow, double goodness! :-) [12:52] didrocks, that's my background image [12:52] hey kenvandine, how are you? [12:52] good [12:52] so ok, it just stat it [12:52] glad to be home finally :) [12:52] to ensure the cache is up to date [12:52] didrocks, yeah, seems to do what it should [12:52] so, now, just have to ensure why it reads the default which is even not set [12:52] kenvandine, how is your jetlag? [12:53] none... amazingly [12:53] didrocks, right, that's what I'm looking at [12:53] hey kenvandine, goo dmorning [12:53] i was on monday after traveling from europe on sunday [12:53] bah, I really hate the "reveal on the edge" [12:53] but heading back to the east coast had no affect :) [12:53] i really don't like any sort of hiding... [12:53] keep showing the unity launcher when I want to do "back" in firefox and browse dconf-editor left pane [12:54] pitti: that segv with pygobject3.0 and python lenses, is related to gvariant handling I believe... Can you check https://bugs.launchpad.net/unity/+bug/855402 [12:55] Ubuntu bug 855402 in unity "Python lenses segfault with pygobject3" [Critical,In progress] [12:56] good morning! [13:02] DOH [13:02] didrocks, pitti, chrisccoulson: stupid dx! [13:02] user-desktop.svg: [13:02] what have they done? [13:02] it's for the desktop icon [13:02] urgh? [13:02] they load and compute the background in each svg [13:03] "nice" [13:03] 22 48 64 128 [13:03] ok, that explains everything [13:03] nautilus takes half the cpu with the gnome icon theme... [13:03] njpatel, ^ you guys have to be kinding us [13:03] so, I have some python code to do that for oneconf, which just do it when you change your wallpaper :) [13:04] seb128, sorry? [13:06] seb128, I don't think that's a DX thing dude [13:06] njpatel, ignore me, it's a an issue with icon people, don't you ;-) [13:06] njpatel, but I don't know who "icon people" is nowadays ;-) [13:06] seb128, sladen [13:06] is the point of contact [13:06] lord knows who actually creates them :) [13:06] njpatel, yeah, thanks, moving channel === njpatel is now known as njpatel|flu === m_conley_away is now known as m_conley [13:41] chrisccoulson, well, doesn't mind really, the result is the same, so I guess there's no need to update the patch [13:41] chrisccoulson, just reworked your patch to make it upstreamable [13:46] tseliot, i'm not sure why the gsd-randr log showed the keypress before the randr event [13:47] but when i debug it in gdb, i get the randr event before the keypress [13:47] so it does apply the correct config, then immediately changes it again [13:47] why a keypress would trigger an xrandr change? [13:48] seb128, it's the fn+f7 keypress [13:48] oh [13:48] the one that cycles through display configs :) [13:48] it seems the dock triggers that :( [13:48] perhaps to work around a windows bug ;) [13:48] do they try to be smart assuming the software will be dumb? [13:48] yeah, i guess so [13:49] I wonder if that's dell specific... [13:49] just like the remapping of Fn+F7 to Super + P in the BIOS ;) [13:50] chrisccoulson: it makes sense. Maybe we can make the app ignore hotkey events for a small amount of time after randr events? We should consider any potential regression though [13:50] tseliot, i guess that's possible. it sounds a bit hacky though ;) [13:50] chrisccoulson: only because it is ;) [13:51] i'm not sure what to set the time to though. my laptop is fairly fast, so i could pick a time that works for me, but it still breaks for everyone else [13:52] rodrigo_, who normally works on the randr stuff in gsd? [13:53] chrisccoulson, federico [13:53] chrisccoulson: you can also ask RAOF [13:53] rodrigo_, is that his IRC name too? i don't see him online atm [13:54] chrisccoulson, he's doing some house moving, he told me by mail, but his mail is federico@gnome.org [13:54] he's usually 'federico' on irc [13:55] rodrigo_, thanks. i'll drop him an e-mail and see if he has an opinion on this too [13:55] ok [13:55] i have to run out for a bit... my wife's van won't start so I need to go provide some road-side assistance [13:56] kenvandine, will that assistance involve her driving back in your motor, leaving you stranded with her van? ;) [13:56] that's what i would expect to happen if jo was stranded somewhere :) [14:02] seb128: i see a couple of ecryptfs highlights above ... ecryptfs broken? [14:07] kirkland, no, just the lightdm candidate version which had login issues with ecryptfs users again [14:07] kirkland, but we did catch it up before release [14:07] kirkland, robert_ancell said it's difficult to write tests for ecryptfs login in lightdm because pam sucks basically [14:11] dobey, bug #856331 btw [14:11] Launchpad bug 856331 in ubuntuone-client-gnome "the gsd integration doesn't work in Oneiric" [High,Confirmed] https://launchpad.net/bugs/856331 [14:12] seb128: yeah, i saw that e-mail earlier. so we have to install real schemas somewhere now? [14:13] dobey, yes [14:13] boo :( [14:13] dobey, well it's just a .xml to add to your source [14:13] with a few makefile lines [14:14] yeah, and new strings to be translated [14:15] dobey: do you know if there's any progress on fixing the tomboy U1 stuff? [14:16] there are sadly a few bugs affecting tomboy sync atm :( [14:17] Laney: no news :-/ [14:18] dobey, who cares about gsettings key description to be translated [14:18] what do you think? [14:19] if it's not fixed very soon it might be best to use the standard sync auth thing? [14:20] Laney, does it work via the tomboyweb type? [14:20] 848250 affects me too, I raised it with upstream [14:20] rodrigo_: it at least lets you connect [14:20] ^ that bug means I can't sync [14:20] don't know if it affects everyone though [14:20] Laney, can you raise https://bugzilla.gnome.org/show_bug.cgi?id=657180 with them as well? [14:20] Gnome bug 657180 in General "tomboy blocks shutdown" [Normal,Unconfirmed] [14:20] Laney, syncing doesn't work for me neither, not only setting up the syncing service [14:20] there's a problem with the sensitivity of the save button too [14:20] that bug is ridiculous [14:21] rodrigo already raised that one [14:21] ok [14:21] we should have let tomboy out of the CD for this cycle, seems like it will be really buggy for Oneiric :-( [14:21] bug 848250 [14:21] Launchpad bug 848250 in tomboy "Tomboy sync fails on Oneiric, New Note Template already exists" [Undecided,Confirmed] https://launchpad.net/bugs/848250 [14:21] i suspect some small fixes will make syncing work again [14:22] the biggest problems are in an ubuntu patch [14:22] ah [14:22] chrisccoulson: FYI, tkamppeter's g-s-d fix overrode your PPA version [14:22] bug 848250 [14:22] Launchpad bug 848250 in tomboy "Tomboy sync fails on Oneiric, New Note Template already exists" [Undecided,Confirmed] https://launchpad.net/bugs/848250 [14:22] pitti - yeah, going to upload shortly [14:23] just investigating another (but unrelated) display config problem [14:23] pitti, 10v login down to 25s ;-) [14:23] well rather closer from 23s [14:23] the chart line is a bit weird, nautilus and compiz are done at 23s [14:24] are we still frozen? [14:24] seb128: but you still have CPU activity? [14:24] i guess we are :) [14:24] after that for 2s? [14:24] we aren't unfreezing [14:24] didrocks, I've a gap of 1.5 seconds and then zg-datahub that I had delay by 10s [14:24] didrocks, if we delay by 15s or 20s it will be 23s I think [14:24] hey! could one of the gtk guys quickly double check what I do wrong in http://paste.ubuntu.com/695136/ ? when I click the button it always returns a assert error and "None" for combo.get_active_text() [14:25] I'm pretty sure this used to work [14:25] didrocks, it just at the wrong timing and gets picked [14:25] seb128: see, 20 is fine :) [14:25] ;-) [14:27] Laney, hmm, tomboy syncing just worked for me, after restarting it [14:27] I got some conflict message about the new note template when I first synced [14:27] Laney, yes, saw the bug, not sure why it fails there [14:27] but I have no idea why it wants to sync that [14:28] it syncs everything afair [14:28] nor does it show up on U1 so shouldn't be a conflict in any case [14:28] mvo: following up with myself, Gtk.ComboBoxText() - does not work ; Gtk.ComboBoxText.new() - works fine. now I want to weep a bit [14:28] pitti: could you upload yelp-tools for me (it's not part of the desktop set yet)? [14:32] jbicha: sure, where is it? [14:32] mvo, good catch [14:32] mvo, loving gir don't you? ;-) [14:33] mvo: oh i have been weeping all week from gir :) [14:34] mvo: indeed, I took a lot of time with stupid things like that in the OneConf view, so if you see .new() sometimes, it's because the other one doesn't work… [14:34] mvo, didrocks: you guys have it easy :) [14:35] dobey: well, I just loved autocompletion on ipython so much while doing that! :-) [14:35] dobey, seb128: like honestly, this is not even remotely funny, I want to switch to qt^Whtml5^Wvisual basic [14:35] pitti: https://code.launchpad.net/~ubuntu-desktop/yelp-tools/ubuntu [14:35] *ANYTHING ELSE* [14:36] mvo: try polling an fd in a glib main lop with gir, AND having your code be backward compatible to run all the way back to lucid [14:36] dobey: woah, WOAH! [14:36] mvo: visual basic, sounds interesting, the future? (or even better, vba) :-) [14:36] mvo: hint: none of the fd polling APIs in glib/gobject/gio are actually bound in gir :) [14:36] dobey: I had some compat code in s-c to support both old and new glib, that was already a big pita [14:37] didrocks: I believe its more consistent … [14:37] * mvo stops ranting now [14:37] mvo: you aren't using twisted. pita doesn't even begin to describe it ;) [14:37] mvo: does gtk_comboboxtext_new() do anything except init'ing the object and setting properties? that's what the default gobject constructor does [14:37] seems that the metro window 8 apps are using html5 + js [14:38] pitti: I don't know, but that makes it pretty likely [14:38] pitti: I mean, it appears that its doing more [14:38] pitti: what is crazy is that I think that this worked at some point in the past, but I may be wrong [14:38] if you just call Class(), that'll just call the GObject constructor and set any property which you pass [14:39] which should usually work, unless the _new() C method does something on top of that [14:40] mvo: and reading the code it just calls g_object_new() with properties [14:41] mvo: so I'd expect that ComboboxText.new() sets id-column=1 and entry-text-column=0, while ComboboxText() just leaves these properties at their default [14:41] mvo: ah, default of these is -1 [14:42] popey, you're such a meanie [14:42] mvo: or is it breaking with something else? then we need to have a closer look [14:42] :D [14:42] DBO: get fired, then I'll let you in [14:42] pitti: that sounds pretty likely that this is the crux [14:46] pitti, do you think bug #840089 is worth oneiric notes [14:46] Launchpad bug 840089 in gnome-settings-daemon "doesn't convert the power options from gconf to gsettings" [Medium,Invalid] https://launchpad.net/bugs/840089 [14:46] pitti, basically upgrades to natty will have their behaviour of lid close, etc resetted [14:49] seb128: ah, yes I think we should have that [14:49] .. documented [14:49] at least there's a way to configure it now [15:00] pitti, btw, opened bug #856475 about the nautilus issue [15:00] Launchpad bug 856475 in ubuntu-mono "user-desktop and desktop icons slow the desktop a lot" [High,New] https://launchpad.net/bugs/856475 [15:03] is it me or the pad is down? [15:03] seb128: thanks, nice catch! Updated the wiki page and linked to boot speed project [15:03] ah back now [15:03] didrocks: unity stacking issues have been gone since using the ppa, btw [15:03] didrocks, it tends to have issues for some time [15:04] tjaalton: excellent new \o/ [15:04] news* [15:04] pitti, thanks ;-) [15:04] tjaalton: should be in oneiric soon I guess, still frozen [15:05] seb128, cjwatson: wasn't gnome-contacts added to the desktop set? or was it gnome-online-accounts? [15:06] didrocks: seems like the archive was unfrozen already [15:07] desrt, btw, forgot to tell you yesterday -> https://bugzilla.gnome.org/show_bug.cgi?id=659738 [15:07] Gnome bug 659738 in gsettings "Need a way to tell GSettings to delay writes for rapidly changing properties" [Normal,Unconfirmed] [15:07] rodrigo_: showed up on my radar, thanks :) [15:07] desrt, ok [15:07] tjaalton: updates are cherry-picked rather [15:07] tjaalton: but not the compiz/c-p-m ones yet [15:09] didrocks: ah, ok [15:09] tjaalton: see https://launchpad.net/ubuntu/oneiric/+queue?queue_state=1&queue_text= [15:10] rodrigo_, it was g-o-a, g-c is in universe iirc [15:10] seb128, oh, ok [15:10] so lp:~ubuntu-desktop/gnome-contacts/ubuntu up for sponsoring [15:10] didrocks: compiz upload refers to bug 845765; is that a bug # typo, or is it really a compiz bug? I think the former, as the changelog talks about stackign order and that bug is a crash [15:10] Launchpad bug 845765 in indicator-weather "indicator-weather crashed with SIGSEGV in g_array_append_vals()" [Medium,New] https://launchpad.net/bugs/845765 [15:12] pitti: urgh, this is the end of the changelog that smspillaz completed, I didn't double checked, sorry about this [15:12] * didrocks knows now that he couldn't really trust anything from compiz… [15:12] pitti - after linking the ubuntu-boot-speed project to several bugs this week, i noticed that the bug supervisor is the foundations team [15:12] they don't mind do they? ;) [15:13] I just subscribed to all bugs in that project, too [15:13] pitti: can't get a similar one, it wasn't on my initial list [15:13] chrisccoulson: I think that was Robbie; I guess they mainly read it out of interest [15:13] pitti - ah, ok [15:13] i'll subscribe to those too [15:13] didrocks: ok, so just leave it like it is; the bug won't be closed then, which looks right [15:13] pitti: ok, let's leave it like that, thanks! [15:14] didrocks: no problem; I just wanted to ask whether I should fix the task or so [15:20] pitti: did you not add ubuntuone-couch also? [15:21] dobey: uh, no; isn't that the kind of package which should rather be a dependency? [15:21] dobey: I mainly reverted to the beta-1 situation before teh installer [15:21] pitti: it was a recommends from deja-dup before it was changed to suggests for bug #833824 [15:21] Launchpad bug 833824 in ubuntu-meta "Remove Ubuntu One packages from Oneiric CD" [High,Fix released] https://launchpad.net/bugs/833824 [15:22] dobey: we didn't ship it in natty either; but if it's useful for anything, I don't mind having it, it's tiny [15:22] mterry mentionned it's needed IIRC [15:22] ah [15:22] pitti: so it is need for deja-dup integration with u1 [15:22] maybe deja-dup should recommend it rather? [15:22] dobey, mterry: would you mind if we'd restore that recommends instead? it sounds more correct to me, as it's also supposed to work when you install it on e. g. xubuntu [15:23] pitti, I intend to restore the recommends as soon as beta is over. I dropped it because of the whole installer plan [15:23] sounds fine i guess. [15:23] mterry: ah, nice (you can upload, btw) [15:23] pitti, which is no longer the direction, right? [15:24] beta is over :) [15:24] pitti, oh! heh, I hadn't noticed [15:24] mterry: yes, was quite an about-face :) [15:24] mterry: you could have upload all the time [15:24] mterry: oneiric is frozen (and will never thaw again), so all uploads are held in unapproved anyway [15:24] pitti, oh, OK [15:24] mterry: we are now flushing the queue as the b2 images are final now [15:25] pitti: care to sponsor https://code.launchpad.net/~dobey/ubuntu/oneiric/ubuntuone-file-storage-api/fix-packages/+merge/76279 ? i can't upload it anyway, regardless of the freeze. :) [15:25] dobey: sure, doing [15:26] thanks [15:26] dobey, you need to get U1 package upload rights [15:28] mterry: i need to tet a u1 package set made [15:45] chrisccoulson: is the "stable distribution.ini" issue still on your radar? we need this by the release if we don't want to break defaults packages with each and every firefox release [15:49] pitti, I have found what makes CUPS crashing, and a workaround, not a fix. [15:52] pitti, When creating or removing a queue, the printer is (un)registered for DNS-SD broadcasting this queue and CUPS keeps in an array which queues are broadcasting. It adds the DNS-SD service name to the printer's record and then adds the record to the array, with the service name as sorting/searching criteria. [15:53] pitti, when it removes a queue, it searches the record in the array to remove it and afterwards sets the service name in the record to NULL. When creating a queue it searches the array to find the correct place for the new record, so that they keep sorted. [15:57] pitti on this array search (it does a binary search) it chokes on a record in the array where the pointer to the service name is not the start adrres of a string nor 0 (NULL), but 1 (!). I did not find out how this happened, but I can work around it by defining the array without comparison function (set it to NULL), getting an unsorted array which is searched linearly. Array size is the number of locally defined queues. Most users hav [15:57] e less than 100. Would the linear search a problem for a user with ~10000 queues? [15:59] pitti, 10000 queues would be comparing 10000 pointers would this take too long? [15:59] pedro_, ola, thanks for sending that gconf bug upstream btw ;-) [16:02] tkamppeter: hm, maybe report it upstream first with that analysis, perhaps Mike has a better patch quickly? [16:03] tkamppeter: it depends on how much time it takes for each operation; if it's just a second or two in total, we could certainly do that workaround, but if it would actually need to contact the printer, it seems way too expensive [16:09] pitti, it is nothing more than a comparison of pointers, this simple loop: [16:09] for (current = 0; current < a->num_elements; current ++) [16:09] if (a->elements[current] == e) [16:09] { [16:09] diff = 0; [16:09] break; [16:09] } [16:09] [16:10] where a->elements[] is the haystack and e is the needle. [16:10] pitti ^^ [16:10] tkamppeter: ah, that's simple [16:10] tkamppeter: linear search should be fine there [16:10] binary search seems to be over-sophisticated here. === eeejay_is_afk is now known as eeejay [16:35] While testing Ubuntu 11.10 w/ Unity2D, I noticed that when clicking on a global menu option in OpenShot (such as "New Title", or "Preferences"), a new non-modal dialog would open (as expected), but it is hidden behind the existing window. Any thoughts or suggestions? Thanks! [16:37] okay, running Unity2D ... why sometimes when I alt-tab, the icon just wiggles in the left panel? I can't get it to actually come to the foreground with alt-tab .... [16:49] kirkland, that would have something to do with focus stealing prevention in metacity [16:49] not sure if we have a bug filed about it, but i know people have very different opinions on focus stealing [16:50] i do think alt-tab should really raise, so it should get fixed [16:50] kenvandine: yeah, totally [16:50] best to file a bug against unity-2d to make it try to deal with it in metacity, worse case the unity-2d guys will move it to metacity [16:51] yeah, maybe there is a raise on alt + tab [16:55] out for a bit, bbl [16:58] ah, no, we still have TB meeting today [17:03] pitti: if you want to ack Qt just before leaving after the meeting… [17:07] pitti, updated bug 855445 and tested workaround, no perceivable delays [17:07] Launchpad bug 855445 in cups "cupsd crashed with SIGSEGV in _cups_strcasecmp()" [Undecided,Triaged] https://launchpad.net/bugs/855445 [17:13] ok, time for some exercice, have a good evening everyone! [17:26] kenvandine: when you update ubuntu-wallpapers, do you update ubuntu-wallpapers-xml.in manually? or is there some trick to it that i'm missing? [17:33] i did it manually [17:34] pitti, tested adding and removing 10000 queues no perceivable delays. [17:35] seb128, ah no problem ;-) (re gconf bug) [17:46] does somebody want to work on bug #839407 [17:46] Launchpad bug 839407 in gnome-utils "gnome-font-viewer missing in oneiric" [Undecided,Confirmed] https://launchpad.net/bugs/839407 [17:46] ? [17:46] it would basically add a binary for it (or copy what debian did) [17:46] it's built we just don't ship in any binary [17:47] jbicha, kenvandine, somebody else? [17:50] seb128, i can take a look [17:51] FYI, i just uploaded notify-osd [17:51] kenvandine, thanks [17:51] kenvandine, thanks ;-) [17:51] had to fix a ftbfs too :) [17:54] pitti, workaround for bug 855445 committed to BZR, can you upload a new CUPS package? Thanks. [17:54] Launchpad bug 855445 in cups "cupsd crashed with SIGSEGV in _cups_strcasecmp()" [Undecided,Fix committed] https://launchpad.net/bugs/855445 [17:56] pitti, and thanks for fixing cups-dbg. [17:57] tkamppeter: ah, thanks; will do [17:58] rickspencer3, how do i get photobomb? [17:58] i thought it was in repos for O [17:58] lamalex, it's in my PPA [17:58] why isn't it just in the archive? [18:01] lamalex: buy it from Software Center :) [18:01] i'm trying! [18:01] the search doesn't find it [18:01] I think it's only for sale in Natty [18:01] ah [18:02] that's silly [18:10] * Laney wonders how you get source for Free apps distributed in that way [18:10] also, s-c doesn't display prices on the index? [18:18] good night everyone! [18:18] pitti: night [18:21] seb128, thunderbird-couchdb is in source new btw ;) [18:24] chrisccoulson: YAY! [18:24] hrmm [18:24] heh :) [18:25] i wonder if i /have/ to use org.foo.bar.bz.really.long.thing.gschma.xml as filename for gsettings schemas [18:25] makes no sense, that [18:43] pitti, thank you for uploading CUPS. [18:52] tedg, ping about dbusmenu memory [18:53] mterry, Yes? [18:54] tedg, oh, hmm [18:55] tedg, just re-acquainting myself with dbusmenuitem/gtkmenuitem interactions. Was going to ask a question about who owns the dbusmenuitem used for dbusmenu_gtkclient_newitem_base, but I think it's eventually owned by the root member of the gtkclient? [18:56] mterry, Yes [19:03] seb128: good news. fixing that schema bug, i also am fixing the other bug to use gsettings instead of gconf in oneiric. do i need to do anything special to get the values migrated from gconf to gsettings, for existing users, or are they already there? [19:06] dobey, http://developer.gnome.org/gio/stable/ch28s07.html [19:06] dobey, basically you need to ship a .convert which has the gconf to gsettings keys mapping and it will work [19:06] dobey, you can get lot of example in GNOME softwares that switched to gsettings [19:07] ok, great; thanks [19:16] ugh, gsettings doesn't have set/get_list :-/ [19:46] kenvandine, do you use gnome-contacts? there is an update rodrigo_ did that needs sponsoring if you fancy trying it and sponsoring [19:46] cyphermox, hey [19:46] cyphermox, there is a new evolution-data-server tarball fixing some bugs worth rolling a new tarball apparently if you want to do it [19:46] mterry: do you know how xdg-autostart works? does it just run everything in /etc/xdg/... in alphanumeric order? [19:47] achiang, gnome-session drives it. It runs them in "phase" order. desktop files can set a phase they are in [19:47] what is xdg-autostart? [19:47] ok, mterry is around, I will let him handle it ;-) [19:47] :) [19:48] yes, and you should put everything in the application phase [19:48] achiang, there is also another directory, /usr/share/gnome/autostart [19:48] it also checks that they are allowed to run under the environment in question [19:48] don't put anything in the initialization phase, else we will hunt you down :) [19:48] achiang, it also doesn't run things that have Hidden=true, things that say NotShowIn=GNOME, etc. [19:49] chrisccoulson, should we get g-s-d out of there? ;-) [19:49] seb128, that needs to be there [19:49] why? [19:49] just don't add anything else to it ;) [19:49] seb128: sure [19:49] it's already slow enough [19:49] cyphermox, thanks [19:49] this is good too since I just uploaded a new gtkhtml (not that it changes much though) [19:49] chrisccoulson, in lucid we didn't wait for gsd to be done to run the wm at least iirc [19:49] mterry: ok, thanks. as usual, i'm doing something odd and custom [19:49] seb128, it needs to have the theme and font stuff set up before anything else appears on screen [19:49] chrisccoulson: ^^ so don't worry, i won't mess with upstream. :) [19:49] cyphermox, yeah, you own us an evolution update as well ;-) [19:50] yup yup, getting there [19:50] seb128, i actually tried flattening it out this week, just to see what happened, and i ended up with unity with the wong icon theme [19:50] * cyphermox is stabbing the fugly NM 3G mobile connection wizard bug [19:50] and nautilus with the wrong theme too [19:50] achiang, (if you aren't doing something time sensitive, you can also set X-GNOME-Autostart-Delay=60 or so to delay it from slowing boot [19:50] chrisccoulson, oh I need to try that, gtk3 is supposed to pick themes correctly, like nautilus shouldn't stay grey when g-s-d get sdown and respawn during the session [19:51] seb128, yeah, it still happens [19:51] chrisccoulson, did you actually check with gtk3? upstream told me during the natty cycle that it was working with it [19:51] mterry: trying to solve a different problem; i want to know when the desktop is fully up and usable, so i'm looking for a likely process name to snoop for in a bash busy-wait loop [19:51] bah [19:51] mterry: unless you have a better idea. ;) [19:51] chrisccoulson, (never trust what upstream says ;-) [19:51] in any case, changing theme / fonts settings and monitor settings once there is already stuff on screen is really slow [19:51] heh ;) [19:52] chrisccoulson, btw in case we didn't mention it, no holidays for you until all the g-s-d issues are fixed in Oneiric ;-) [19:52] lol [19:52] my last day is tomorrow ;) [19:52] seriously though, i'll be around at the start of next week [19:52] we don't go on holiday until friday [19:53] then it'll be no interweb for a week [19:53] achiang, you mean, when all the autostart files have been processed? looking into gnome-session is your best bet. I don't *think* it sends out a useful dbus event or anything by that time. [19:53] achiang, or just set an autostart delay of 0 or 1? [19:54] mterry: hm. ok [19:54] mterry, we've had this discussion before, when xsplash was conceived [19:54] and there isn't really a reliable way to do that atm [19:54] ie, i think gnome-session will tell you once it has started everything in the Application phase [19:54] but that doesn't really tell you whether the desktop is loaded [19:54] seb128, i can sponsor gnome-contacts [19:55] chrisccoulson: what i'm doing now is this: [19:56] http://pastebin.ubuntu.com/695273/ [19:56] chrisccoulson, heh, joke aside don't work even next week! [19:56] chrisccoulson, you really deserve some time off work ;-) [19:56] we launch gdm in the background, and then look for indicator-datetime startup... once it's up, then we exit the script [19:56] chrisccoulson, enjoy your holidays and be back for UDS ;-) [19:57] heh, thanks :) [19:57] we have gdm set to auto-login [19:57] mterry: ^^ [19:57] achiang, i guess that would work. i wish we had a better way of doing it though :) [19:57] i'm not sure there is really a good way to tell when the desktop is really loaded and ready to use though :( [19:57] you could maybe try to monitor cpu and stop after 3 seconds under 50% or something [19:58] seb128: not a good heuristic, because we're actually inside a chroot [19:58] :) [19:58] kenvandine: are you able to sponsor yelp-tools for me also? [19:58] jbicha, did you email cjwatson about having it added to the desktop set? [19:59] achiang, ok, well the way usplash was working or bootchart is doing its line is to check what is usually the slowest process and to wait for it to be done [19:59] achiang, we patched nautilus by then to send a dbus signal to say it was done with loading [19:59] seb128: hm, but the nautilus patch has been dropped? [20:00] achiang, there are probably also dbus names you could wait for? com.canonical.Unity, com.canonical.Unity.Panel... but not sure that's any better than watching for datetime [20:00] yes because we stopped using xsplash [20:00] seb128: do you know what era that was, so i know about when to go looking? [20:00] jaunty/karmic wasn't it? [20:00] IIRC ;) [20:01] achiang, karmic [20:01] http://launchpadlibrarian.net/30103945/nautilus_1%3A2.27.4-0ubuntu5_1%3A2.27.4-0ubuntu6.diff.gz [20:01] ok, i'll play around with my hacky technique, and if need be, i'll go do some code archaeology. :) [20:01] seb128: awesome, thanks! [20:02] yw [20:02] except... won't nautilus finish loading before the indicators? [20:02] seb128: yes but perhaps he's been busy or it get lost in his inbox [20:02] there were some iteration over the patch so maybe take the karmic stable version to get the "current" version of it [20:02] jbicha, yeah, no worry about the "no reply yet" I just wanted to make sure it's tracked [20:03] jbicha, sure... where is the branch? [20:03] achiang, what will be loaded first...do a bootchart on your config [20:03] seb128: right, sounds good. thanks [20:03] when we did that gnome-panel and nautilus were the slowest [20:03] * achiang gets it now [20:03] we got both to emit a signal and xsplash to wait on those or timeout [20:04] kenvandine: https://code.launchpad.net/~ubuntu-desktop/yelp-tools/ubuntu [20:04] jbicha, great, i'll do it in a bit [20:04] jbicha, btw I need to discuss with ricotz and you and maybe others [20:04] jbicha, it would be nice to see an desktop-extras set for things in universe, i.e gnome-shell [20:05] jbicha, but somebody needs to build the list of what would be in that set ;-) [20:05] jbicha, I figured you and ricotz might be interested [20:08] seb128: yes, it's a bit tricky figuring out what its scope should be [20:09] jbicha, well there is an easy "starting" set [20:09] anything in GNOME not in main? [20:09] i.e gnome-shell, anjuta, epiphany-browser,... [20:10] cheese, gda, gdl [20:10] etc [20:10] but yeah, everything GNOMish not in the desktop set [20:10] then we can discuss random "side softwares" [20:10] ok, I'll work on a proposed list [20:11] jbicha, thanks! [20:12] jbicha, no hurry, it will be for next cycle most likely, but still would be nice to get that rolling ;-) [20:12] jbicha, ricotz seemed interested as well [20:12] the idea is to have a team who can maintain at least the gnome-shell by their own [20:29] jbicha, sponsored [20:30] kenvandine, btw did you see the gwibber bug I assigned you yesterday? [20:30] kenvandine, micahg seemed to thing it was a frequent issue for users [20:30] kenvandine, oh, welcome back btw ;-) [20:31] seb128, no... it isn't in my assigned mailbox... bug number? [20:31] kenvandine, it has to be! checking [20:31] it probably matched some gwibber rule before the assigned run in my procmailrc [20:32] i should make sure i get that ordered right... :) [20:32] * kenvandine hates email :-p [20:33] kenvandine, no, it's private so it's not emailed [20:33] kenvandine, bug #845374 [20:33] ah [20:36] i don't think that will affect upgraders, but clearly I need to handle that better [20:37] kenvandine, ok, I don't understand the issue but the dup count seemed low enough that I didn't consider a frequent issue [20:37] i suspect it is failing to create the path to store the avatar in [20:38] it should handle the exception gracefully [20:38] i'll take care of it, thx! [20:42] * dupondje is really missing an option to delete wireless accounts in Gnome [20:42] kenvandine, thank you [20:51] hrmm, will have to finish this branch up in the morning; doesn't build on oneiric :-/ === m_conley is now known as m_conley_away [21:26] what up everyone...how are things? [21:28] not bad, not bad [21:29] I managed to fix up the retardedness in the mobile broadband wizard, after a few hours prodding at it :) [21:30] cyphermox: nice... [21:31] how about that bug that rick was having with atheros? the one where his wifi dropped every 30 minutes or so? [21:32] hi jasoncwarner_ [21:32] hey bryceh [21:34] bryceh: you see my email on multi-monitor bugs? want to unblock dx soon so we can get those sorted ... when you've taken a look, let me know what you think [21:35] hey jasoncwarner_ [21:35] jasoncwarner_, yep, read the email [21:35] d'oh, fail. I apparently busted my transfer limit here. [21:35] I'm pretty sure seb128 should be sleeping...I hope didrocks is sleeping [21:36] jasoncwarner_, do you remember a couple sprints ago when we first evaluated Unity? [21:36] jasoncwarner_, RAOF and I put it on our laptops and connected up to the projector [21:36] bryceh: jasoncwarner_: multi-monitor; anything I can help testing? [21:36] cyphermox: I'll foward email [21:37] ok [21:37] jasoncwarner_, at that time we noticed some fairly bad breakages with multimonitor under unity which weren't present with gnome [21:38] smspillaz, heyo. with the new compiz in the archive, I just got the "dash behind other windows" issue [21:38] jasoncwarner_, Sam was in the desktop room and RAOF and I (and I think you?) chatted with him about the multimonitor troubles we'd seen [21:39] at that time we learned that compiz (I think) had it's own multimonitor handling code [21:39] jasoncwarner_, raof and I suggested they look at the gnome-desktop multimonitor logic, because the GNOME guys had already gone through painstakingly debugging MM support over the past couple years [21:41] jasoncwarner_, you took us into the DX room and we went through all these problems with them; pretty sure we even wrote them on flip charts :-) [21:41] :) [21:41] bryceh: so, what you are telling me is that dx should already know all this? [21:42] jasoncwarner_, well, trying to say we're not (intentionally) blocking Dx. It's our impression these are not xorg bugs, but rather issues in compiz' multimonitor logic. [21:44] ok...I'm going to reply to the thread and include you guys. sounds like we all need to get in a room really quickly and figure out what game plan to finish 'em is going to be... [21:44] thanks [21:44] Yeah. I'd looked at those bugs, and while it's possible the nvidia one is an driver bug, I'm pretty certain the intel one is not. [21:45] jasoncwarner_, sure. Fwiw, the bug reports you forwarded look like were already looked at from the X side and determined not to be X bugs [21:45] RAOF, ^ am I remembering things properly? You may have had additional contact with the compiz folks about dual head configuration? [21:46] bryceh RAOF ok.... [21:46] bryceh: Other than bitching to them occasionally, not really. [21:46] jasoncwarner_, I can explain what's going on with bug #830949 in more detail [21:46] Launchpad bug 830949 in xserver-xorg-video-intel "[Intel N10 Graphics] Plugging in external monitor to VGA port makes both displays corrupted with thick slanted lines " [Critical,Invalid] https://launchpad.net/bugs/830949 [21:47] that woudl be great...I think more explanation on those would help this along... [21:47] sure [21:47] I no longer use two monitors day-to-day, but before I switched to the standing desk I was, and Unity was playing nicely on my hardware. [21:47] multimonitor support is working pretty nice in 2d :) [21:48] jasoncwarner_, the hardware is capable of running multiple monitors at some large total resolution. the hardware is also able to render 3D to a certain size buffer [21:48] on some hardware, the 3D buffer size is less than the total resolution size it can map. [21:49] That's true of most hardware, in fact, although users will rarely hit it. [21:49] bryceh: if it helps I can test a bunch of laptops in the Montreal lab tomorrow for multi-monitor [21:49] FYI...for those interested, these are the bugs we are discussing [21:49] https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/830949 [21:49] https://bugs.launchpad.net/ubuntu/+source/unity/+bug/813343 [21:49] https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/830949 [21:49] https://bugs.launchpad.net/ubuntu/+source/unity/+bug/813343 [21:49] https://bugs.launchpad.net/unity/+bug/815613 [21:49] Ubuntu bug 830949 in xserver-xorg-video-intel "[Intel N10 Graphics] Plugging in external monitor to VGA port makes both displays corrupted with thick slanted lines " [Critical,Invalid] [21:49] Ubuntu bug 813343 in unity "nvidia drivers, second monitor covered by black" [High,Confirmed] [21:50] Ubuntu bug 815613 in unity "[intel] When I plug in a second monitor, I get a black bar across the top of my screen and the second monitor is black except for the bar at the top." [High,Confirmed] [21:50] not sure how much this will be useful and fixable so late though :) [21:50] jasoncwarner_, btw those are just 3 bugs (you have two that are the same bugs) [21:50] bryceh: awesome, you guys already "fixed" two bugs ;) [21:51] cyphermox, yeah, of course more testing can't hurt but I think the issues are understood it's more of an analysis/debugging problem now; needs someone conversant in compiz internals I think [21:51] alright [21:51] I think that bug 815613 is unity failing to set the output viewport correctly. [21:51] Launchpad bug 815613 in unity "[intel] When I plug in a second monitor, I get a black bar across the top of my screen and the second monitor is black except for the bar at the top." [High,Confirmed] https://launchpad.net/bugs/815613 [21:51] in general I recall it working pretty well on my system in the office, but it's always intell graphics [21:52] That looks like what unity used to consistently do for me, but I fixed some of it and the rest got fixed. [21:52] and like RAOF my hardware tends to be magical [21:52] cyphermox: Some of my hardware seems to have lost it's shine; compiz just crashed when I enabled the second monitor :) [21:53] RAOF: plus don't you have one system that crashes on wifi or somethign? [21:53] jasoncwarner_, I do like that dbarth is estimating that 80% of Dx's troubles are because of xorg. We get the blame for a lot of non-Xorg problems, but we've never been thrown under the bus quite that forcefully before. ;-) [21:54] cyphermox: Yeah, that's this system ;) [21:55] fun. if you tell me what it is I might have the same card somewhere or the same laptop model [21:56] seems like I'm traveling to the office tomorrow :) [21:56] Dell e6420 [21:56] Network controller: Broadcom Corporation BCM4313 802.11b/g/n [21:56] like the one that was broken forever in Maverick and maybe natty too? [21:57] It worked in natty for a while. [21:57] hey cyphermox I didn't see if you responded about rick's wifi bug..the atheros one? [21:57] ie with bad KMS issues? [21:57] jasoncwarner_: I did [21:57] really weird issue, it dies because of "inactivity" [21:57] It would lock up at the drop of a hat in the early oneiric cycle; it's stable now. [21:58] cyphermox, btw for testing multimonitor a couple things that tend to surface bugs - rotate one or both displays 90 degrees, disable/reenable one display, boot with monitor A and hotplug monitor B where B's resolution is much greater than A's. [21:58] bryceh: that's fun, common cases too [21:59] maybe not rotationing displays, but higher resolutions on external [21:59] does the type of output change much? [21:59] cyphermox, if you're doing automated testing or just want some shortcuts, I've put some scripts into xdiagnose/data/workloads that do some of these things [21:59] we can test VGA, DVI, DP, and with some advance notice, HDMI [21:59] ok [21:59] yes DP and HDMI often tend to have unique bugs [21:59] cyphermox: Got some HDMI with audio sinks? [21:59] RAOF: I do [22:00] DVI seems to be pretty well debugged by now [22:00] I bought two screens when I was in QA -- one Dell with HDMI vga and dvi, with speakers, and one pro model also from dell with vga, dvi and dp [22:00] cyphermox, one issue we've been seeing recently is a ~1-sec lag when probing outputs (such as when running xrandr) [22:00] HDMI ? audio should shake out some bugs. [22:00] the pro model rotates and is a really nice screen :) [22:01] bryceh: as in xrandr blocking for a second? [22:01] * RAOF loves his ultrasharp 24" [22:01] cyphermox, right [22:01] Oooh, which rotates! [22:01] I can look up the price. both monitors went for under 500 :) [22:01] (together) [22:02] * RAOF would need to seriously rearrange his desk to fit another monitor on. [22:03] And I don't think any of my decent hardware can reasonably drive two external displays :( [22:04] I had a nice two monitor setup until my gf decided I had to reconnect her tower :) [22:05] bbl [22:06] * RAOF reboots to finish update before reporting more compiz bugs. [22:17] did smspillaz ever pick up the patch to catch RRChangeNotify events? [22:19] broder, good question [22:19] smspillaz? [22:33] broder, what patch? [22:34] Hi all. I have a question about bug status handling: Could somebody have a look at https://bugs.launchpad.net/df-libreoffice/+bug/745836/comments/51 and give me an opinion/judgement before that escalates into a bug-state-war? (asking here because ubuntu-bugs is silent) [22:34] Ubuntu bug 745836 in libreoffice "soffice.bin crashed with SIGSEGV in cppu::throwException()" [High,Confirmed] [22:35] DBO: http://lists.compiz.org/pipermail/dev/2011-March/001443.html [22:35] sam had some comments about not responding to both randr events and configurenotify events, but that's a pretty straightforward change [22:35] (swallow the configurenotify on the root window if priv->randrExtension) [22:38] broder, the patch is not merged [22:38] I'll ask sam about it [22:38] and if he has no good objections [22:38] we'll merge it [23:13] hey bryceh RAOF and cyphermox , I am going to close my stuff for a bit to get through some TODOs. I'll be back online a bit later to catch up on what you think of those bugs...thanks! [23:13] Catch you later!~ [23:22] I've been annoyed by bug 856884 for a long time but I didn't know what was breaking it [23:22] Launchpad bug 856884 in unity "Running unity --reset breaks metacity keyboard shortcut defaults" [Undecided,New] https://launchpad.net/bugs/856884