=== olli_ is now known as olli === qengho is now known as CardinalFang === CardinalFang is now known as qengho [06:06] good morning desktopers! [06:11] morning! [06:12] hi guys [06:12] seems upload notification email is broken. [06:15] happyaron, hey! you mean the ones you get when you dput something to the archive? [06:16] yup [06:49] larsu, desrt, could one of you look at bug 1381804? [06:49] bug 1381804 in glib2.0 (Ubuntu) "whoopsie test failure since glib2.0 2.41.2-1 uploaded" [High,New] https://launchpad.net/bugs/1381804 [06:51] seb128: i'll take a peek [06:51] desrt, thanks [06:53] desrt, it hits a "g_main_loop_quit: assertion 'g_atomic_int_get (&loop->ref_count) > 0' failed" [06:53] I've an i386 install/can reproduce if you need info [06:55] desrt, http://bazaar.launchpad.net/~daisy-pluckers/whoopsie/trunk/view/head:/src/tests/test_monitor.c#L125 [06:57] desrt: let me know if you need help [07:07] the new indicator headers on the phone are nice looking! === oCrazyLem is now known as CrazyLemon [07:09] seb128: only on i386? [07:10] desrt, bugs states on !amd64 [07:10] desrt, https://launchpad.net/ubuntu/+source/whoopsie/0.2.38 [07:10] seb128: was just getting there :) [07:11] first theory: testcase is doing something evil :) [07:11] desrt, so yeah, arm, ppc, i386 [07:11] yeah, that's possible [07:11] something that used to work with old glib and still work on amd64 for some reason [07:11] but fixing the testcase is a valid fix ;-) [07:11] seb128, oh, I only just noticed the new headers [07:12] chrisccoulson, hey! they feel nice, don't you think? [07:12] chrisccoulson, how are you btw? ;-) [07:12] yeah, they're much better than before [07:12] I'm good thanks, how are you? [07:12] I'm good, thanks [07:13] larsu, would you fancy looking at https://bugs.launchpad.net/bugs/1381897 ? [07:13] Ubuntu bug 1381897 in evince (Ubuntu) "Evince will not zoom in with shortcut key in 14.10 Release Candidate?" [Low,Incomplete] [07:13] seb128: sure [07:13] larsu, I think that's something you fixed in the distro patched version of evince we had but that seems buggy in current upstream [07:14] larsu, thanks [07:14] larsu, ctrl-shift-+ vs ctrl-+ [07:15] seb128: right, we added = or something for those users [07:16] larsu, maybe you can put a patch upstream up for review, that does that? [07:18] seb128: don't see any mention of atomics in the report [07:19] desrt, that's the output from running the test manually on my box [07:19] I guess the make check run eats stderr [07:19] ** (./test_monitor:9220): WARNING **: Expected file /tmp/whoopsie-test-SLNTNX/fake.uploaded doesn't exist. [07:20] seems like this is the original complaint.... i wonder if you're seeing something different again [07:20] no [07:20] https://launchpadlibrarian.net/186966939/buildlog_ubuntu-utopic-armhf.whoopsie_0.2.38_FAILEDTOBUILD.txt.gz [07:20] " /whoopsie/callback-never-triggered: [04:53:09] Could not read directory (2): /fake [07:20] OK [07:20] /whoopsie/callback-triggered-once: [07:20] (./test_monitor:7527): GLib-CRITICAL **: g_main_loop_quit: assertion 'g_atomic_int_get (&loop->ref_count) > 0' failed [07:20] Trace/breakpoint trap" [07:21] desrt, see how the could not read gives an OK [07:21] desrt, that's the buildlog from armhf [07:21] well, maybe it's a different issue, but the one that current fails the build is the one I'm seeing locally [07:23] * desrt grabs i386 iso [07:23] don't have my VM party when i'm not at home :) [07:24] desrt: ask seb128 to let you ssh into his machine :P [07:24] desrt has ssh access to porter box [07:24] Debian ones at least I think [07:24] i do!? [07:24] oh right.... from ages ago [07:24] i bet i have a new ssh key since then :) [07:26] i think it will be faster to download the iso :p [07:26] seb128: I was kidding... [07:26] code in this test is suspicious [07:26] larsu, I'm not, I ssh to porter box quite often for debugging, easier that vms [07:58] morning [08:02] hey willcooke, how are you? [08:02] hi willcooke [08:02] o/ [08:03] yo [08:03] there were no pips at 9am just now on radio 4 [08:03] so my morning routine is RUINED [08:03] WHAT? [08:04] seriously [08:04] we're probably being invaded or something [08:04] Wasn't that some secret signal for WW3? [08:05] glad I'm getting out of this country tomorrow morning. It's going to down the pan [08:05] Laney: morning :) [08:06] hey larsu [08:06] how's it going [08:06] hey Laney === tiheum is now known as tiheum_ [08:07] willcooke, flying to Washington tomorrow? [08:07] seb128, yeah, made the flights about 1/2 the price [08:07] weird [08:07] hey seb128 [08:07] or are you coming back on friday evening or saturday? [08:07] seb128, Friday evening [08:07] k [08:07] that's probably just the saturday rule [08:07] yeah, saturday night rule [08:08] though it should have been cheap on saturday as well [08:08] just not on sunday [08:08] I dont get that rule. Is it some kind of tourist trap, to make you spend more money? [08:08] willcooke, no, it's what makes business people different from tourists [08:09] ahh, kk [08:09] willcooke, http://en.wikipedia.org/wiki/Saturday-night_stay [08:09] ha - I should have known there'd be a Wikipedia page [08:09] :-) [08:10] seb128, can we work through that Gtk stuff this morning? I'll document it as well. Squeeze another blog post out ;) [08:11] willcooke, sure [08:11] let me take my test laptop [08:12] seb128, shall we do it over IRC? If so, perhaps a side-channel? [08:12] * larsu overheard something about gtk stuff [08:12] willcooke, I think it's going to be a 3 lines IRC discussion [08:12] oh sweet [08:12] let's do it here then? [08:12] just need to check again the magic [08:12] yes [08:13] larsu, having gtk apps listed in the unity8 dash on desktop-next [08:13] larsu, so you can run e.g gedit [08:13] we don't? [08:13] no [08:13] because no backend [08:13] doesn't it use the same setting as unity7's dash? [08:13] oh [08:13] so we didn't want to list something not working [08:13] right [08:13] no [08:13] the issue is that they didn't want to list tons of non working stuff [08:13] like x-apps, java, gtk, etc [08:13] so they went with a whitelist approach [08:14] well "whitelist" [08:14] that makes a lot of sense for now [08:14] some new key in the .desktop [08:14] right [08:14] result is that e.g gedit is not listed [08:14] so you need manual hack to use it [08:14] willcooke wants to know the hack [08:15] there's already a "GTK" desktop file category [08:15] FWIW [08:16] Laney, yeah, but Unity8 doesn't use that atm [08:16] I'm sure not [08:17] If that works it's better than modifying tons of desktop files though [08:17] right [08:17] it's at least vaguely upstreamable [08:17] except if they want to make sure to not list buggy softwares [08:18] which they were leaning toward [08:18] good luck with that [08:18] well, the idea was "add the key to the .desktop if it has been tested/confirmed to work" [08:18] sounds scalable and maintainable [08:18] :p [08:19] this is "only" a quick hack so we can demo to Jane [08:19] willcooke, yeah, we are discussing futur/proper solutions there [08:19] not the demo [08:19] ahh [08:19] kk [08:19] in which case... [08:19] carry on [08:19] ;) [08:20] willcooke, so, I think you just need to add "X-Ubuntu-Touch=true" to your .desktop [08:20] willcooke, try adding that to /usr/share/applications/gedit.desktop and restart your unity8 session [08:20] will do... [08:21] * willcooke install gedit first [08:21] ;) [08:22] oh I should unbrick my phone [08:22] would it "work" for xchat-gnone too? [08:22] 285 was not good [08:22] Laney, did you wait for a while? [08:23] like 60 minutes [08:23] I forgot about it :p [08:23] also there's some posts on the ubuntu-phone list [08:23] willcooke, worth trying, but I think that had issues due to some widgets/xorg use [08:23] ack [08:23] Laney, k, complain to ogra then!: [08:24] hey ... that were slangasek and cjwatson who played with utopic :P [08:25] seb128, done - it appears in the list, but just crashes [08:25] willcooke, xchat or gedit? [08:25] seb128, geofft [08:25] oops sorry geoff [08:25] seb128, gedit [08:25] :-( [08:25] works on my machine [08:26] aw man, I wanna be a text editor [08:26] well, let me update before claiming that [08:26] erk [08:26] seb128, stop! [08:26] * seb128 stops [08:27] (hammer time) [08:27] seb128, now I cant even get lightdm to start [08:27] * Laney dances [08:27] shrug [08:27] after a reboot [08:27] seb128, dont want you to break your machine too [08:27] :) [08:27] best time to brick your laptop [08:27] :) [08:27] before travelling! [08:27] seb128, I'll upgrade everything and try again [08:27] well, that's a test machine, I'm happy to brick it [08:27] so I can help debugging [08:27] * seb128 upgrades [08:30] :) [08:36] seb128, dist-upgrade and I'm back with a GUI [08:37] seb128, also Store seems to be working now [08:38] nice! [08:42] seb128, still cant get gedit to load [08:42] seb128, where should I look for logs? [08:43] willcooke, .cache/upstart/unity8.log === tiheum is now known as tiheum_ [08:47] willcooke, sorry, mine is still dist-upgrade, disk is sloooow, I should change it as well ;-) [08:47] seb128, looking at the logs [09:16] seb128, bah - nothing obvious in the log: http://pastebin.com/vnqdh10a [09:18] willcooke, to me it looks like the recent mir updates made the gtk-mir backend stop working [09:18] arse [09:18] oh well, so be it [09:18] willcooke, there is a "application-legacy-geditmir-.log" with a "gdk_mir_event_source_queue_event: code should not be reached" [09:19] willcooke, somebody for robert_ancell to look at, maybe tomorrow [09:19] cool [09:19] I'll drop him a line, thanks seb128 [09:20] willcooke, I was about to reply to the gtk-mir email, I can do that if you want [09:20] or you do it, that works too ;-) [09:20] seb128, oh, if you could that would probably be better than me saying "it won't go - I have no idea what I'm doing" [09:20] :D [09:20] willcooke, ok, doing so [09:20] thanks [09:29] willcooke, how does that look? [09:29] * willcooke reads [09:30] Laney, is the start up disk creator something you'd still like to work on next cycle? [09:30] seb128, excellent, thank you [09:30] willcooke, yw! [09:39] larsu: https://code.launchpad.net/~laney/whoopsie/source/+merge/238539 [09:43] Laney: thanks! approved [09:43] (can't top-approve) [09:45] cheers [09:45] * Laney uploads [11:27] willcooke: oops, forgot to reply, yes I think so [11:31] Laney, great - I'll include it in the plans we produce next week, but if I forget - poke me. [11:50] oops, linked the wrong bug === MacSlow is now known as MacSlow|lunch === alan_g is now known as alan_g|lunch === tkamppeter_ is now known as tkamppeter [12:19] w00t, new unity [12:22] speaking of unity and keybindings [12:22] is ctrl-alt-L still working for others? [12:22] it changed to super-l for me for some reason [12:22] both are supposed to work no? [12:23] are they? [12:23] I though we said it was not acceptable to stop supporting the legacy combo and had both working [12:23] bregma, Trevinho ^? [12:24] Laney, https://launchpad.net/ubuntu/+source/unity/7.1.2+14.04.20140214.1-0ubuntu1 [12:24] seems weird to have a random keybinding not exposed in the ui [12:24] " * Adds Super+L to lock the screen, while keeping the older shortcut [12:24] around in g-s-d (Ctrl+Alt+L). (LP: #830709)" [12:24] Launchpad bug 830709 in gnome-settings-daemon "Keyboard shortcut - Unity should also use Super-L to lock screen by default" [Medium,New] https://launchpad.net/bugs/830709 [12:24] or did g-s-d change the default in the 3.12 update? [12:24] could be [12:25] ctrl+alt+l still works for me [12:25] wasn't a Unity change that removed that key sequence, it was working for me not too long ago [12:25] doesn't work for me now [12:25] - '<Control><Alt>l' [12:25] + '<Super>l' [12:25] indeed [12:25] http://launchpadlibrarian.net/186130738/gnome-settings-daemon_3.8.6.1-0ubuntu15_3.12.2-1ubuntu1.diff.gz [12:25] I guess we should add an override to ubuntu-settings [12:25] * seb128 does that [12:25] seb128: why does it wotrk for me then? [12:25] Laney, ^ sounds ok to you? [12:26] - '<Control><Alt>l' [12:26] + '<Super>l' [12:26] oh yeah [12:26] you already pasted that [12:26] larsu, because you went to g-c-c since you installed and did ctrl-alt-l in there one day [12:26] larsu, and it's in your user settings since* [12:26] that's probably alright I suppose [12:26] can't remember, but I'm getting older, so might be [12:26] larsu, well, it's enough that you might have tested a bug to see if the keybinding change was working or something [12:27] larsu, gsettings get org.gnome.settings-daemon.plugins.media-keys screensaver & reset & get [12:27] seb128: indeed, it's set to primary+alt+l [12:27] * larsu doesn't reset. muahah [12:28] lol [12:28] I like super-l better [12:28] in your FACE ctrl alt [12:29] Laney: I actually like super+l better too. In theory. Good luck explaining that to my fingers [12:30] reset the key :) [12:32] I'll try. let's see how that goes [12:32] probably results in me losing my private key in DC [12:33] yeah, I had a couple of "screen didn't lock" incidents [12:50] Laney, want to cherrypick https://bug738625.bugzilla-attachments.gnome.org/attachment.cgi?id=288667 on top of your evince upload? that's 1381897 === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g [13:05] Laney, seb128 they should both work... [13:06] Trevinho, hey, right, we figured out the issue, it's the g-s-d update, it relies on g-s-d to catch ctrl-alt-l while compiz does super-l [13:06] seb128: yes, I've just read that as well [13:06] seb128: in fact we just rely on that for ctrl+alt+l [13:06] while unity does handle super+l with his key settings [13:36] seb128: after it gets review [13:37] Laney, not trusting larsu?! [13:37] ;-) [13:37] right on, I don't trust myself either [13:37] system("keylogger"); [14:05] can we get a new epiphany next cycle? [14:06] larsu, does it use gtkheaderbar? [14:06] can we easily add back normal decoration without having a weird UI? [14:07] it would be weird in that case I guess [14:07] because they have the title / locationbar switcher thing [14:07] it's not default, don't see why we would block there [14:08] ya, and maybe next cycle we'll be able to actually resize windows with csds [14:09] Laney, do you know why it was not updated this cycle? [14:09] but yeah, I think it's fine updating [14:09] I doubt many people use !firefox|chrom* anyway [14:09] no idea [14:09] those who do can deal with csd weirdnesses [14:09] :) [14:10] I just want to make fastmail a webapp [14:10] and the current one doesn't seem to work at all anymore [14:10] sluggish scrolling [14:10] and indeed we should make csd work properly [14:10] how are webapp handled? [14:10] it's a standalone browser instance without chrome? [14:10] Laney: still waiting on a compiz fix... [14:11] v-cycle for real guys! [14:11] seb128: yes, exactly. It puts a .desktop file into you .local [14:12] Laney: NUDT guys spend most of there energy on a big bid for the gov in earlier this cycle, so quite some updates flows in this week. [14:12] * there -> their [14:19] happyaron: ho hum [14:19] It means that this "new UI" gets almost no testing before the release [14:20] Laney: the people in charge comes: @JackYu [14:20] Hi pitti [14:20] happyaron: It's ok, I just wanted to highlight it [14:20] ic [14:20] thanks for replying [14:20] np [14:21] local smoke test shows no problem, but could be problems hidden there. [14:21] Laney, happyaron, I'm back. [14:23] hi JackYu, I approved the FFe now [14:23] I wanted to draw some attention to the lateness [14:24] Laney, thanks. I answered your comments just now. [14:24] now you only have a week to fix bugs [14:24] scary :) [14:25] I like the new UI more, by the way :P [14:25] yes^ [14:25] thanks, lol [14:26] Laney: and maybe a rush for PPU rights, ubuntu-kylin-sso-client.. (sry for pinging again :P) [14:27] bad man [14:35] happyaron: I want to try an experiment first [14:35] (generating the ubuntukylin packageset from seeds) [14:36] I'm fine, :) [14:50] happyaron: http://paste.ubuntu.com/8574321/ [14:51] I think maybe I should override chromium ... [14:58] Laney, wdyt of http://paste.ubuntu.com/8574355/ ? [14:58] Laney, sorry, http://paste.ubuntu.com/8574357/ [15:00] LIst? [15:00] Laney, typo, fixed ;-) [15:01] I think you shouldn't repeat the logic [15:02] and otherwise? [15:02] and use QDir::absoluteFileName to get the path instead of concatenating [15:04] you can tell entryList to give you sorted lists too, which should be better for the final sort [15:04] Laney, how does that work? [15:04] success - slides are finished [15:04] with the exception of didrocks [15:04] who will have a nice surprise on Monday morning [15:04] soundsDir.absoluteFileName(dirList[i]) [15:04] or so [15:05] hum [15:05] doesn't that take a name as argument? [15:05] can you give it a path? [15:05] I need to prepend the /custom [15:05] not add a filename at the end [15:05] ? [15:05] you list the contents of two directories [15:06] yeah, but "absoluteFileName" is about a filename [15:06] as the name indicate [15:06] not about a dir [15:06] it works if you do "dir.absolutefilename("entry")" [15:06] don't understand [15:06] not sure about "/some/other/dir" [15:06] what does line 17 do? [15:07] oh ok [15:07] I though you were speaking about [15:07] QDir customSoundsDir("/custom"+dirString); [15:07] sorry [15:07] Laney: what about packages like youker-assistant? [15:08] oh that's QDir customSoundsDir("/custom").absoluteFilePath(dirString), no? [15:08] pretty sure that works [15:08] happyaron: already in there [15:10] your list is only additional to the existing right? [15:10] yep [15:10] cool [15:11] Laney, with dirString = /usr/share/sounds ? [15:11] yeah I think you can do that [15:11] that function is weirdly named [15:11] but try it === mhall119 is now known as mhall119|vacatio [15:23] happyaron: Any opinion on bug #1342677? [15:23] bug 1342677 in im-config (Ubuntu) "Does im-config need to be displayed in menus?" [Undecided,New] https://launchpad.net/bugs/1342677 === gatox is now known as gatox_lunch [15:25] GunnarHj: I don't think the menu is needed either [15:25] GunnarHj: at least for all derivatives that is using language-selector [15:26] happyaron: Right. Kubuntu would be a special case. [15:27] happyaron: And debian itself. Not very likely that they would start using l-s. [15:27] I think Osamu is more interesting to something like im-chooser? not sure [15:28] happyaron: What's im-chooser? [15:28] current IM start-up process isn't all ideal. it doesn't play well with event-driven desktops [15:29] GunnarHj: the Fedora approach [15:29] happyaron: Well, that's about the overall design of im-config... [15:30] it's already rewritten once from the old school im-switch [15:30] much better than before [15:31] happyaron: Yeay, you advocated its implementation at the time. :) [15:32] it works as designed at least, the problem for event-driven desktops are because of design. the problem in im-switch are more about the implementation. [15:37] happyaron: Ok. To be honest, the expression "event-driven" is over my head. [15:42] sort of upstart/systemd user session, plus some more dbus activation [15:42] seb128: am I correct? ^^^^^ [15:42] happyaron, about what? sorry, doing several things at the same time [15:44] seb128: event-driven desktop [15:44] I don't know about the im-config specifics [15:44] but you should be able to get whatever you need to work [15:45] it's just a matter of finding the right events [15:46] GunnarHj is trying to see what does event-driven desktop means/works [15:46] ohnoez - didrock's site is down again, and Cloudflair isn't doing its thing [15:47] happyaron: But we have had either upstart or systemd all the time, haven't we? [15:48] GunnarHj: use the same way of booting system services to manage desktop sessions [15:49] GunnarHj: so effectively upstart/systemd is managing it [15:51] happyaron: I know we have /usr/share/upstart/sessions/im-config.conf in Ubuntu. [15:52] but for setting XIM, it's still required to hook into the xsession script sequence [15:53] and another problem is we aren't to change the variables on the fly, while im-chooser provides some degree of support (but it's not using systemd) [15:55] happyaron: Changing env. variables on the fly is a tricky one for all variables, isn't it? [15:55] yes I agree, but unfortunately it's that way for ages [15:56]