=== gatox is now known as gatox_away === Laney is now known as Guest22075 === zyga-afk is now known as zyga [08:05] robert_ancell, https://wiki.ubuntu.com/DesktopTeam/DejaDup [08:06] robert_ancell, https://wiki.ubuntu.com/UbuntuEngineering/12.04/UpstreamDevelopment/ProjectTracking [08:07] robert_ancell, https://wiki.ubuntu.com/UbuntuEngineering/12.04/UpstreamDevelopment talks about requirements a bit [08:25] mterry, i just proposed https://code.launchpad.net/~ken-vandine/gwibber/lp_906916/+merge/88455 [08:25] which i would like to include in this mornings upload [08:25] kenvandine, looking [08:25] mterry, thx [08:25] mterry, it's been in the unity-team ppa and several people have tested it [08:27] cyphermox: bug 836250 [08:27] Launchpad bug 836250 in linux "[Oneiric] [Regression] Intel Corporation Centrino Ultimate-N 6300 poor networking, packet loss and very slow Lenovo X201 and T500 laptops" [Medium,In progress] https://launchpad.net/bugs/836250 [08:34] seb128, https://wiki.ubuntu.com/DesktopTeam/LightDM https://wiki.ubuntu.com/DesktopTeam/UnityGreeter [08:37] mterry, watch -n0.1 'ps -e -o pcpu,rss,comm|grep gwibber-service' [08:46] desrt, https://bugs.launchpad.net/ubuntu/+source/gtk+3.0/+bug/911263 [08:46] Launchpad bug 911263 in gtk+3.0 "Update overlay scrollbar patch" [Undecided,New] [08:55] kenvandine: [09:05] mterry, do you have time to do a review of the libdee/libunity port too? i think didrocks is a bit busy atm [09:05] kenvandine, sure [09:05] ok, let me make sure he hasn't started it [09:06] vuntz: bonjour! ca va? [09:06] pitti: mon ami ! [09:06] mterry, https://code.launchpad.net/~ken-vandine/gwibber/libdee_libunity_port/+merge/88454 [09:06] mterry, thx [09:07] vuntz: got a minute to pick your brain about libwnck? [09:08] pitti: sure thing [09:09] vuntz: so, I was debugging why notify-osd wakes up with each key press and focus change event [09:09] vuntz: and eventually found out that calling wnck_screen_get_default () triggers that [09:09] vuntz: this sets up an X event filter (_wnck_event_filter_init()) [09:09] vuntz: but there is no API to shut that down again [09:09] mvo: good morning! [09:10] good morning vuntz :) [09:10] vuntz: and it's a static reference, you can't use _unref() or so to stop it [09:10] vuntz: would you be opposed to something like wnck_screen_uninitialize() which stops the event filter? [09:10] mvo: I uploaded gir1.2-unity-5.0. Do you want to update update-manager yourself or do you want me to do it (for real, this time ;))? [09:10] pitti: https://bugzilla.gnome.org/show_bug.cgi?id=642692 [09:10] vuntz: I'll file an upstream bug in a minute and are happy to work on a patch [09:10] Gnome bug 642692 in general "Provide wnck_shutdown() (?)" [Normal,New] [09:10] mvo: I saw there is some UNRELEASED stuff there, hence the question [09:10] pitti: there you go [09:11] vuntz: hah [09:11] didrocks: blop :-) [09:11] vuntz: so, if that approach is fine with you, I'll get started on that [09:11] "kill all unnecessary wakeups" [09:11] sure [09:11] vuntz: merci Monsieur [09:12] pitti: although I'm unsure if this would fix the thing for notify-osd -- it sounds weird to use libwnck there, to start with [09:12] vuntz: it uses it to check if there is any maximized window [09:13] pitti: hrm, does that mean that you want to call wnck_screen_get_default() each time there's a notification, and then wnck_shutdown()? [09:13] i. e. it iterates over wnck_screen_get_windows() [09:13] vuntz: is there a better way? [09:13] pitti, can you bump the build score of lightdm in https://launchpad.net/~ubuntu-desktop/+archive/ppa/+packages? [09:13] kenvandine, seriously? ;) [09:13] - _model = new Dee.FilterModel (_sort_filter, _streams_model); [09:13] + _model = new Dee.FilterModel (_streams_model, _sort_filter); [09:13] pitti: don't know, but that's pretty ugly :-) [09:14] vuntz: for sure :) [09:14] mterry, yes... [09:14] i know :) [09:15] pitti: but there's probably no good & easy way to do that another way, I guess (without adding some new stuff elsewhere, I mean) [09:16] vuntz: is wnck_screen_get_default () very expensive? [09:16] it didn't seem so from my cursory glance at the code [09:18] pitti: what's expensive is getting all windows and their attributes [09:19] vuntz: ack; but it already does that each time when there is a bubble :/ [09:19] pitti: not really [09:19] pitti: if you don't call wnck_shutdown(), then the data is automatically fetched when there's a change [09:19] (hence all the wakeups) [09:20] pitti: so it comes for free in that case [09:20] wnck_screen_force_update() doesn't update it again? [09:20] you're kidding me [09:20] vuntz: but even then, updating it at each key press sounds even worse to me [09:20] wnck_screen_force_update() is called every time? [09:20] even though there's a main loop? [09:21] so, the current code is mis-using the API [09:21] anyway [09:21] pitti: if you want to avoid the wakeups, what you propose sound right [09:21] vuntz: http://bazaar.launchpad.net/~canonical-dx-team/notify-osd/oneiric/view/head:/src/dnd.c#L201 [09:21] vuntz: it's calling that for each new bubble [09:21] and yes, it's ridiculously expensive :( [09:22] the wnck_screen_force_update() call shouldn't be needed there [09:22] in fact, it calls dnd_dont_disturb_user() each time, which dnd_has_one_fullscreen_window () is part of [09:22] unless you do the shutdown thing, of course [09:22] vuntz: *nod* [09:22] what this tries to do is to not show a notification if there is a fullscreen window [09:23] but at least notification bubbles are relatively rare, especially when you are offline and on battery [09:23] but, to be fair, I guess that, while still expensive, all the X communications in wnck_screen_force_update() are not too expensive either [09:23] so I would at least fix the wakeups with every X event [09:24] yeah [09:40] a friend of mine upgraded to precise, but has no clock/calendar on the top panel [09:40] should "unity --reset" fix that? [09:41] tjaalton: Do they have indicator-datetime installed [09:41] TheMuso: good point, I'll ask [09:43] TheMuso: apparently not.. [09:43] wonder how's that possible [09:43] after apt-get --fix-policy install and all [09:46] ok, who is wanting to test a new lightdm? [09:46] debs in https://launchpad.net/~ubuntu-desktop/+archive/ppa [09:47] i.e ppa:ubuntu-desktop/ppa [09:47] works fine here [09:56] TheMuso: so it got removed post-upgrade [10:02] hey didrocks [10:02] mvo: hey hey! :) [10:02] didrocks: sorry, missed the ping, what channel was it in? [10:02] mvo: this one ;) one sec [10:02] didrocks: no problem, I can upload [10:02] 10:10:40 didrocks | mvo: I uploaded gir1.2-unity-5.0. Do you want to update update-manager yourself or do you want me to do it (for real, this time ;))? [10:02] didrocks: found it now in the backtrace [10:03] backtrace? too used to python-gobject? :p [10:03] haha [10:05] mvo: so, do you want to deal with that yourself? :) [10:07] didrocks: yes, will do [10:08] mvo: awesome! thanks :-) [10:13] dobey: hi, when you're online can you point me to ubuntuone-client packaging branch? [10:16] seb128, nope [10:16] bschaefer, hey [10:16] so bug #880876 [10:16] Launchpad bug 880876 in unity "Unity causes ibus to not work correctly (spaces incorrectly placed)" [High,Confirmed] https://launchpad.net/bugs/880876 [10:17] seb128, I installed that fix today and every working for me [10:17] bschaefer, what fix? the one in precise? [10:17] seb128, the one from the link, pretty much what your patch was [10:18] agateau, lp:ubuntu/ubuntuone-client [10:18] bschaefer, but the patch doesn't work? [10:18] seb128, well this https://bugs.launchpad.net/ubuntu/+source/unity/+bug/880876 [10:18] Launchpad bug 880876 in unity "Unity causes ibus to not work correctly (spaces incorrectly placed)" [High,Confirmed] [10:18] opps [10:18] https://github.com/fujiwarat/ibus [10:18] this branch is what you were going off wasn't it? [10:18] from git [10:19] seb128: ah thanks! [10:20] dobey: unping, seb128 answered [10:21] seb128, the patch was working, or at lease that git fork. It isn't merged with the main one though [10:22] seb128, screen shot of the 2, the first half was that git branch and other is the main ibus https://launchpadlibrarian.net/89804609/ibus_fix.png === lan3y is now known as Laney [10:24] didrocks: uploaded [10:24] mvo: great! thanks a lot :) [10:24] yw [10:26] bschaefer, ok thanks [10:30] seb128, well I am hoping that bug has been resolved. There was also another small bug where the im_context was not getting focus [10:31] seb128, well I am hoping that bug has been resolved. There was also another small bug where the im_context was not getting focus [10:31] bschaefer, I need to have another look why people say the precise version doesn't work where the git version does [10:33] seb128, hmm, yeah because your patch looked exactly like the git version. [10:33] bschaefer: the patch *is* merged, no? [10:34] isn't it that context async stuff? [10:34] bschaefer, cyphermox said some variable need to be set or not set for it to work correctly [10:34] Use ibus_input_context_process_key_event_async in ibus-x11 [10:34] seb128, cyphermox that was changed in that patch wasn't it? [10:35] seb128: could they already have dropped the env in the new inuty? [10:35] cyphermox, no idea [10:35] cyphermox, are you talking about IBUS_ENABLE_SYNC_MODE env var? [10:35] bschaefer: afair the patch was that "Use ibus_input_context_process_key_event_async in ibus-x11" commit, 28abee something something [10:35] bschaefer: yeah [10:36] cyphermox, that is still used because with out it the normal XIM cant handle the ibus like say the shell can [10:37] cyphermox, it is in IMTextEntry, though it should get unset after the search bar is created, so it doesn't effect other programs [10:37] yeah, that's the problem -- the shell isn't dealing with ibus the same way as the other apps :) [10:37] ok [10:37] * cyphermox reloads his session back to korean [10:38] cyphermox, seb128 yeaah njpatel and I had a big problem with this a while ago when we first got the ibus working [10:38] cyphermox, seb128 that env var was the only way to make the ibus handle normal text [10:39] otherwise the ibus would only handle CJK chars and leave the normal text alone, but that would make the dash not handle normal text... [10:40] bschaefer: what seems weird to me is that variable shouldn't have been working without the patch seb128 added, if I read the ibus code right (which I might not, obviously) [10:41] why is my ibus suddenly not starting? [10:41] cyphermox, seb128 well with the patch seb128 added it should have fixed the bug... [10:41] is the ibus-daemon running? [10:42] yeah, everything is running but the indicator doesn't show for some reason. no big deal for now [10:42] cyphermox, yeah, it seems to be something in ibus-table that gets it going again for me (ibus seems to just want to not work!!!) [10:42] bschaefer: I was testing this yesterday -- it kind of works depending on the applications and the value of IBUS_ENABLE_SYNC_MODE [10:43] cyphermox, yes, when the env var is set to 0 or its NULL everything works correctly [10:43] 몸누랑ㄴ러 [10:43] alright, got it :) [10:43] yup [10:44] cyphermox, sweet, and yeah it is a weird bug but at the same time ibus should still work correctly with that env set [10:45] I guess [10:46] it almost seems to me like this might be interaction between sync mode and RTL, since that's what I can think of that would trigger spaces to show before a character... but only for space? [10:46] and return [10:46] https://launchpad.net/~ubuntu-toolchain-r/+archive/glibc/+packages [10:46] pitti, ^ [10:46] oh, return too? [10:46] yup [10:46] it is the preedit that seems to mess up. So what is in the preedit and you hit space or return it will get commited after the space or return [10:48] because preedit text is still not actually text yet; so it doens't hit the commit yet. [10:48] aye [10:49] sorry, slightly tired nearing 3 am here haha. That is why it looks like it's an upstream problem [10:49] and that git branch fixed the problem for me, and it looked like seb128 patch but I havn't compared them both. [10:49] ill do that tomorrow! [10:50] alright [10:50] bschaefer, let us updated [10:52] cyphermox, seb128 cool. Sorry about being so late to helping! Good luck! [10:52] good night [10:52] bschaefer, thanks for helping, 'night [10:52] thanks, have a good day [11:02] Is anyone experiencing the following with libre office in x64 on P: http://paste.ubuntu.com/802809/ [11:03] mandel: bug 915271, fix underway [11:03] Launchpad bug 915271 in libreoffice "package libreoffice-core 1:3.4.4-0ubuntu2 failed to install/upgrade: rmdir: failed to remove `usr/lib/libreoffice/basis3.4/program/': Directory not empty" [Critical,Triaged] https://launchpad.net/bugs/915271 [11:04] pitti, superb, thx! [11:04] well, not superb.. [11:04] but at least "known" [11:05] pitti, well, I'm a positive lad, being know is way better than a new bug :) [11:28] precise testers needed for Qt! https://launchpad.net/~kubuntu-ppa/+archive/experimental/+packages [11:32] mandel, that depends on the bug. ;) === lool- is now known as lool [12:23] Good Afternoon [12:24] does anyone knows how to setup / pan a Desktop at current ubuntu ? === om26er_ is now known as om26er === kancerman__ is now known as kancerman [14:46] chrisccoulson: is there a way to migrate data from one firefox profile to another (private data and all, but not just copying files over)? [14:47] hey, today had to debug my grandpa's laptop, updates were not working anymore (lucid LTS). the problem is that he did every updates since the lucid release, resulting in 4G of useless kernels filling the / [14:48] linux-image-* and linux-headers-* [14:48] those should really be cleaned automatically IMO... [14:51] xclaesse: isn't "apt-get autoremove" enough to remove them ? [14:53] nerochiaro, do you seriously think ma grandpa knows that? but even then, not it's not removing them ;) [14:54] I think more recently version of ubuntu does remove linux-headers though [14:54] but not linux-images [14:54] nerochiaro: no [14:56] ok, my mistake then. but yeah, ideally it should be automatic, i agree [14:56] well, it should only leave the currently running kernel installed (and the newly installed one) [14:57] since having the old one around is useful in case the new one breaks something [14:57] but you don't need more than the currently running one [14:57] yep exactly [15:10] Sweetshark: ssh scheat.canonical.com === m_conley_away is now known as m_conley [15:52] Sweetshark, hi :) [15:52] Sweetshark, is the gtk3 rendering working already? [15:52] (for libreoffice) === Guest95869 is now known as oakdog8 [15:53] working basically but not that largely tested [15:54] ricotz: feel free to give it a ride ;) [15:55] Sweetshark, i guess installing the -gtk3 package and removing the -gtk would be enough to turn it on? [16:41] pitti, help [16:41] seb128: ? [16:41] seb128: I'm in the ballroom [16:41] pitti, how do I make sure the debs are published hard enough that I can upload my rebuilds? [16:41] pitti, I'm in the desktop room ;-) [16:41] pitti, I don't want to make a fool myself yet another time :p [16:42] seb128: safest is to wait until archive.u.c. Packages.gz has them [16:42] seb128: but it's usually enough if cocoplum shows the version [16:42] pitti, thanks [16:42] seb128: rmadison should be the best compromise [16:42] seb128: i. .e rmadison -r precise pkgbinarymangler [16:42] if I'm lucky the run finishing at 6pm will be ready [16:42] it finished 14 mins ago [16:42] the debs finished to build before 17:30 [16:43] and it's being published [16:43] -s, not -r, sorry [16:43] rmadison is faster than archive.u.c. and "slow enough" for buildds [16:43] pitti, thanks [16:43] de rien [16:43] seb128: you will handle the rebuilds? [16:44] pitti, yes, I'm stacking them already [16:44] seb128: I re-opened the tasks BTW, so that people stop wondering [16:44] (and did a followup) [16:44] seb128: nice, thanks [16:44] I just don't want to upload too early and have to do another round :p [16:44] yw [16:44] sorry for the screwing [16:44] seb128: let me know when you upload control-center [16:44] so I can bump it [16:44] RAOF and I are currently sending a gazillion KDE packages buildd-wards [16:44] (hate perl) [16:44] pitti, ok [17:01] seb128: did you want my recently-merged-from-Debian control-center? [17:01] jbicha, since it was not uploaded I figured you were not done with it [17:02] seb128: it works, one change I didn't add to the changelog yet is that I renamed the Appearance panel to User Interface to match that design spec [17:04] I like having the appearance panel being first in the list though so I don't know if that design idea will be final... [17:05] jbicha, ok [17:05] pkgbinarymangler | 113 | precise | source, all [17:05] seb128: looks fine [17:05] pitti, yeah, I just uploaded [17:06] pitti, https://launchpad.net/ubuntu/precise/+source/gnome-settings-daemon/3.2.2-0ubuntu10 [17:06] https://launchpad.net/ubuntu/precise/+source/gnome-control-center/1:3.2.2-0ubuntu9 [17:06] pitti, if you want to bump scores [17:07] seb128: done [17:07] pitti, danke [18:59] desrt: how do i run an app from the source tree, which uses gsettings, but which doesn't yet have the schemas installed in the system (because i want to run it from the source tree, not install it)? [19:00] dobey, I would ask the same for app-indicator icons... :> [19:01] BigWhale: plug-ins for other applications are a different problem. and that's basically what app indicators are :) [19:02] I figured it is a different problem, but the outcome is the same! :> [19:02] you have to install it if you want to run from source tree :/ [19:02] can't you just hand a pixbuf over the indicator API to use for the icon? [19:02] no [19:02] :/ [19:03] app indicators seem wrong anyway [19:03] i would like to make the top panel smaller though [19:03] Bug #657857 [19:03] Launchpad bug 657857 in kazam "Appindicator cannot use a local icon file" [Low,Confirmed] https://launchpad.net/bugs/657857 [19:03] it looks huge [19:06] huge? [19:07] it's like 28 pixels? [19:07] probably less [19:07] yes, and that is huge [19:08] 24 px [19:08] hrmm [19:08] the icons look huge though [19:08] icons are 22px [19:08] stop using 50" TV as your monitor :> [19:08] i'm not, it's 23" [19:09] i thought i made the launcher panel be smaller too [19:09] but apparently it's 48px wide [19:09] I have launcher set on 32 [19:09] 48 is way too much [19:10] Hmm, would many people be pissed off if I make Kazam Ubuntu only? I have to port support for gnome-indicators :/ [19:11] probably the wrong place to ask [19:11] :) [19:11] :> [19:13] hrmm, i have the icon size set to 32 too [19:13] but the whole launcher is 48px [19:13] meh [19:16] that's strange [19:38] BigWhale: not really. the icons are 32px. there's just a lot junk around the icons too (spacing, borders, arrows) [19:39] but man do these gnome icons look so much better than the humanity ones [19:41] hm true it is 50 pixels [19:41] horrific [19:42] well, at least it's not 64 px [19:42] i just need an ibm t221 [19:42] problem solved. [19:43] sigh; and compiz just crashed :( [20:39] meh, i need ken [20:41] Guys, is this a feature or a bug? https://twitter.com/#!/Chris_Wilkinson/status/157903375470563328/photo/1 [20:42] bug report: https://bugs.launchpad.net/unity/+bug/912682 [20:42] Launchpad bug 912682 in compiz "Compiz add transparency to titlebar along with the panel" [Undecided,Incomplete] [20:48] how gwibber-accounts manages to even work in its current state is beyond me :-/ [20:52] OwaisL: i have no idea [21:14] dobey: GSETTINGS_SCHEMA_DIR [21:14] ah, hrmm === StepNjump is now known as ve2hs === ve2hs is now known as stepnjump [22:07] D/c === m_conley is now known as m_conley_away [22:27] In gnome-terminal, when it says "Use colours from system theme", where can I look up what the actual colour codes are? [22:28] Because I want to configure terminator to use the same colours, but if I "Use colours from system theme" in terminator, it picks entirely different colours === stepnjump is now known as ve2hs === ve2hs is now known as stepnjump