=== TheMuso` is now known as TheMuso === alexabreu is now known as alex-abreu|off [06:36] Bonjour tout le monde ! [06:41] hey pitti! [06:41] how are you ? [06:41] didrocks: bonjour ! [06:42] didrocks: quite tired TBH, it was a long night :) [06:42] so many heart attacks [06:42] hi didrocks and pitti! [06:42] larsu: Alter! [06:42] hey larsu :) [06:42] pitti: who won at the end ? [06:42] will we meet you guys ? [06:42] I got up at an outright "normal" time at 8:30 :) [06:42] hey pitti , didrocks larsu [06:43] didrocks: yes, we will; we won 2:1, but don't ask how :) [06:43] hi darkxst! [06:43] morning darkxst [06:43] didrocks: so, the big showdown on Friday [06:43] pitti: ahah, ok, I won't, I'll just say "nice" :) [06:43] didrocks: FRA : GER Friday 18:00 I mean [06:43] yeah ;) [06:43] I'm still sure I won't watch it TBH :p [06:43] not that France's game against Nigeria was that much more convincing, so let's see [06:43] pitti: by putting the ball into the goal of the enemy, twice? [06:44] already had to suffer 10 minutes of to Nigeria-France to order some beers at the bar with seb128 [06:44] larsu: you didn't watch it, did you? [06:44] (it was of course a really packed) [06:44] pitti: lol. no. [06:44] didrocks: poor you [06:44] I donated blood, took a walk, and went to bed fairly early last night [06:44] pitti: first world issue! :) [06:44] didrocks: I'm really glad that I can at least still be friends with *you* after Friday night! [06:44] * didrocks envies you guys being able to donate blood in the evening [06:45] pitti: ahah, yeah :) [06:45] didrocks: it was more like 5pm (they were available in the school in my street from 4 to 7) [06:45] going too late is bad though, because too many people will be there coming from work [06:47] larsu: here, there are 2 buildings, both are closing at 3:30pm [06:47] and they are at 4kms from home [06:47] so quite hard to leave during the day [06:48] pitti: you're still up before seb128 if that can reconfort you :) [06:49] didrocks: oh? They don't come around the neighborhoods? [06:49] larsu: not where I leave, they are doing that close to the business area, but I'm quite far from it [06:50] going* [06:50] understood [06:53] ahah, a new IT shop from a brand is opening, and they display that: https://plus.google.com/photos/+ldlc/albums/6030735454222755633/6030735454625964194?pid=6030735454625964194&oid=104301983842916548085 [06:54] lol :) === dbarth-afk is now known as dbarth [07:37] mvo, can you join #ubuntu-release? [07:37] good morning desktopers! [07:37] hey didrocks pitti mvo jibel [07:37] hey jibel, seb128! [07:37] bonjour seb128, ça va ! Recontrer à vendredi ! [07:37] Good morning seb128 didrocks [07:37] pitti, mvo, larsu, well done, nice game yesterday ;-) [07:38] seb128: well, FSVO "nice" -- the first 30 mins gave all of us a heart attack! [07:38] * larsu is unsure what he did… [07:38] (well, and the other 90 mins as well really, but in a slightly better way) [07:39] I'm not sure, but I think pitti knows of a "slightly better way" to have a heart attack [07:39] pitti, yeah, it was nice to watch if you like suspens ;-) [07:39] seb128: c'était une nuit très longue, je suis fatigué [07:40] pitti, tu t'es levé tôt ? [07:40] seb128: non, j'ai me levé à 8h30 [07:40] seb128: but of course I woke up with my wife at 5, and didn't really sleep tight after that any more [07:41] seb128: oh, "je suis me levé", not "j'ai me levé" ? [07:42] seb128: I suppose I should learn some French swearing by Friday [07:43] pitti, "je me suis levé" ;-) [07:43] lol [07:44] seb128: what does "never ever was this a foul, you idiot!" mean in French? :-) [07:44] hey seb128 [07:44] lol [07:45] mvo, your upload of update-notifier has been rejected, there is no version in the dependency on update-manager. You can talk to RAOF for details. [07:45] jibel: sure, joining now [07:45] jibel: ok, I will do a re-upload right away, but iirc there is no need for a dependency as the code with use os.path.exists() to figure out if it needs it or not [07:46] hm, appearance page in unity-control-center crashes :( [07:46] larsu, bt? missing schemas and gsettings aborts? [07:47] seb128: some assertions about radio button groups not found. I'll have a look later, aftre I finally finish tab theming [07:47] this has been on my todo _forever_ [07:47] k [07:48] seb128: speaking of which, do we have a screenshot on how it's supposed to look? [07:49] * larsu asks google [07:49] I don't think so [07:49] why? want to change things there? [07:50] no, I want to restore how it's supposed to look and don't have a trusty handy [07:50] I've found a screenshot though [07:51] turns out, theming notebooks is rather messy [07:51] who'd have thought... [07:51] why is it looking different? gtk? [07:51] oh [07:51] yes, after the update. For example, there's a too thick border below tabs [07:51] and frames around the tab header [08:00] do you folks get a crash when opening the appearance control panel? [08:02] see what larsu was just saying [08:03] so seems "yes" [08:03] * seb128 is on trusty so can't test right now [08:04] hey hey [08:04] Laney, hey, how are you? [08:05] yeah, doing good! [08:06] none of this futbol business to worry about here :-) [08:06] you? [08:06] morning Laney! [08:09] hey seb128, may i quickly bug you about https://code.launchpad.net/~ochosi/xdg-utils/support_xfce/+merge/224076 again? [08:10] hey hey didrocks [08:11] hi Laney [08:13] hey Laney [08:13] ochosi, I notice that the upstream and branch patches are different [08:14] Noskcaj: well, ubuntu carries a (useless) patch, so ofc they're different [08:14] ok [08:14] i proposed to drop the patch in ubuntu as well [08:15] but what i'm primarily interested in (now) and with this MR is an SRU for trusty [08:15] cause it kinda sucks that the screensaver always kicks in, even if mediaplayers want to inhibit [08:15] so i kept it minimal [08:15] ok. I've had the same issue [08:15] yeah, everyone does ;) [08:16] Laney, doing good, thanks ;-) [08:16] ochosi, yeah, no reply from upstream then, I'm just going to upload your change [08:21] seb128: thanks, i'm willing to work on a different solution with upstream as well, we can always get that into utopic then. but i'd be happy if we can sru this change to trusty and for 14.04.1 [08:21] right [08:26] mvo: pitti: hey, stupid question, but imagine that one function I called GLib.idle_add() on raises an exception, I'm unsure why my try: exception around mainloop.run() doesn't seem to catch it (like a KeyboardInterrupt event), any idea? seems quite common though… [08:29] I'm trying Ctrl+C in a while True loop and it doesn't seem even to raise the Interrupt [08:29] didrocks, idle_add probably doesnt run in the main thread? [08:30] darkxst: it does, let me pastebin a simple example [08:30] http://paste.ubuntu.com/7730311/ [08:31] when I hit Ctrl + C in the foo function, this should raise an exception… [08:31] didrocks: that's known [08:31] didrocks: pygi can't "tunnel" exceptions through the C layer of callbacks [08:32] pitti: ok, I'm not that crazy then ;). Any best practice/workaround then? [08:32] didrocks: so your callbacks themselves have to intercept the exceptions and handle them, or pass them to the main thread via global variables, signals, messages, etc. [08:33] pitti: ok, that makes sense. Thanks Martin! :) [08:33] didrocks: it's not technically multithreading, but behaviourally it's "kind of" [08:33] didrocks: but the main reason is that it's really difficult to propagate exceptions through arbitrary C functions [08:34] yeah, I guess the conversion isn't really easy, I was just puzzled that even a trivial example wasn't working and was first blaming the input() method or my nested calls :) [08:34] thanks pitti, will do as you suggest === vrruiz_ is now known as rvr [08:39] yes pitti's answer is much better than my "guess" at why that would never work! [08:40] well, for synchronous callbacks it could technically work (with lots of effort) [08:40] but with async callbacks, idle handlers etc. the concept itself also makes no sense as they are "kind of" in a separate thread [08:50] pitti: hum, I propably didn't get exactly what you told then. I tried to have the exception handled in my callback itself, but it didn't work. I reproduced it in that trivial example: http://paste.ubuntu.com/7730381/ [08:51] didrocks: I'm not sure about the semantics of KeyboardInterrupt (in fact, first time I hear about it) [08:51] * didrocks tries an except without anything then [08:51] didrocks: but python-gi installs its own SIGINT handler [08:52] didrocks: which is mostly due to being backwards compatible how pygobject used to behave [08:52] didrocks: i. e. SIGINT quits the main loop, but doesn't raise an exception [08:52] pitti: yeah, it's sucked in from what I see [08:52] hum [08:52] it doesn't quits the main loop in that example [08:53] pitti, that was my guess, i.e. separate thread [08:55] interesting; that sounds like a regression then [08:55] pitti: so, my first example should quit when I hit Ctrl + C, right? http://paste.ubuntu.com/7730311/ [08:56] didrocks: no, you'll never get KeyboardInterrupt with pygobject [08:56] didrocks: well, it should quit because pygobject stops the mainloop [08:56] pitti: but SIGINT should be trapped by the python-gi handler? [08:57] yeah [08:57] but not through the exception handler [08:57] it's not the case [08:57] right right [08:57] sorry, should have removed the handler [08:57] pitti: just to be clear, that should quit: http://paste.ubuntu.com/7730404/ [08:57] didrocks: correct [08:57] it doesn't on trusty at least :/ [08:59] yeah, that's what I meant with "regression" [08:59] there's a test case in pygobject for SIGINT, but apparently it behaves differently [08:59] * pitti bbiab [08:59] pitti: I wonder if it's not https://bugzilla.gnome.org/show_bug.cgi?id=622084 [08:59] Gnome bug 622084 in introspection "Ctrl+C does not exit gtk app" [Normal,New] [09:03] didrocks: yeah, there are probably several bugs for that, we already had some a few years ago when I was looking into that [09:03] pitti: ok, FYI, I used the workaround from the bug description: signal.signal(signal.SIGINT, signal.SIG_DFL) [09:04] and it works for an INT [09:04] at least, I would know as well the other cases for the exceptions raised in the callback and adapt my code to it [09:04] I'm sure it's a declaration of war for Friday! [09:04] * didrocks hugs pitti :) [09:04] didrocks: yeah, that'll restore Python's normal SIGINT behaviour with KeyboardInterrupt [09:38] Dir [09:38] ups [09:39] Laney, why did you remove the QDir include in the click changes? [09:40] it's moved to the header [09:41] are you looking at doing a landing soon? [09:41] yes, today or tomorrow [09:41] Saviq has one lined up first to switch to the suru icon theme [09:41] he aims at landing that today [09:41] okay, I think I can fix the rest of the icons based on my earlier investigation [09:42] stupid spec violators [09:42] do you want to do that on the same merge request? [09:43] it might be good to add some comments in there, I'm re-reading the code you "refactored", it seems suboptimal that we need to play with paths and parsing that way, feels like click could be better a providing the info we need [09:43] having the rational of why we need to do that "manual work" could be useful [09:43] k [09:44] thanks [09:45] "Overwrite the icon with the .desktop file's one if we have it. This one is more reliable." [09:46] what's the point of having an icon info if it's not reliable... [09:48] I think the desktop file one is used by the app scope [09:48] so maybe this one can bitrot or something [09:49] although to be fair ... [09:49] * Laney notices something [10:48] Hi. Can somebody help me? My problem is described here: http://askubuntu.com/questions/490292/dell-fans-not-working === MacSlow is now known as MacSlow|lunch [11:55] lool, if you're still having trouble with large text on HiDPI screens, check the Universal Access 'Large Text' setting in the control center -- see bug #1332947 [11:55] bug 1332947 in unity (Ubuntu) "HDPI: unity randomly turns on large text in universal access" [High,Triaged] https://launchpad.net/bugs/1332947 [11:58] bregma: at the moment the settings panel crashes for me [11:59] bregma: but thanks, I've sub-ed there [11:59] bregma: and indeed, large text is turned on === alan_g is now known as alan_g|lunch [12:20] seb128: I pushed a version that finds ALL the icons on my device === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g [13:32] didrocks: as mentioned, I would like to fetch from launchpad API the list of all bugs with a certain tag [13:33] sil2100: yeah, I'm afraid you need a bug_target for that, let me reread the doc quickly [13:33] didrocks: I already asked on #launchpad, but no one knows it seems [13:33] didrocks: yeah... couldn't find anything, and ubuntu?ws.op=searchTasks only returns ubuntu targetted ones [13:34] sil2100: you can on project, distro, series, milestones for sure [13:34] (so not only ubuntu) [13:34] let me continue to check for available targets [13:34] I'm sure there is person [13:34] yeah, person… [13:34] hmmm... [13:35] sil2100: project_group is maybe what you want? (but we don't really use project_group in LP, apart from the indicators) [13:35] Damn, I would have to 'tag' those specific bugs with a certain dummy project then [13:35] project_group? I wonder what those can be [13:35] * sil2100 looks [13:36] sil2100: yeah, the dummy project is what I did for the unity time [13:36] you can as well have a team that is subscribed to the bugs [13:36] and filter on them [13:36] or ensure they have an ubuntu task [13:36] and filter on that [13:37] didrocks: thanks! I'll try to work-around it in that way then ;) [13:37] that's the 3 easiest options I confirm that is possible (after reareading the doc) [13:37] yw sil2100 :) [13:37] I think that it's probably deliberate [13:37] to not let people search across the collection of all bugs [13:37] Laney: well, you can do that for all ubuntu bugs [13:37] whatever series and so on [13:37] Yeah, but you can still do that in LP manually, and not sure if people use the API so much [13:38] which is I guess even bigger that all projects unioned [13:38] sil2100: there are a lot of people using the API, and if you use a bot, you get kind emails regularly with bots owners :p === ogra_` is now known as ogra_ [14:12] Laney, how did you workaround your issues then print calls were not showing up on stdout from qml? [14:13] use warnings :( [14:13] :/ [14:13] I asked in app-devel but nobody replied [14:27] seb128: which bug are you looking at? [14:28] Laney, I just wanted to put a debug print in the click sorting selector, it seems to always be set on "name" even if the gsettings key is false [14:29] did you try to debug that yet? [14:29] no, was going to though [14:29] well, I did some weeks ago [14:29] before I found out about using warnings [14:46] Laney, do you remember what is supposed to set the sort role on start? [14:46] is that the gsettings key being read and a sortRole being written then? [14:47] lemme look [14:48] erm [14:48] when did qtcreator get a windows style file chooser? [14:48] do you have that upstart fix to set the session-id variable? [14:48] Laney: I've noticed the same in qmlscene [14:48] might have happened with the qt 5.3 update?! [14:49] seems likely [14:49] it used GtkFileChooser before [14:49] I have that variable yes [14:51] :-/ === dpm__ is now known as dpm [14:55] what does qt usually look at to find out which platform it's on? XDG_CURRENT_DESKTOP? [14:56] larsu, GNOME_DESKTOP_SESSION_ID [14:56] oh, sorry, that was qt4 [14:57] (https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/1305294) [14:57] Ubuntu bug 1305294 in upstart (Ubuntu Trusty) "QT uses incorrect theme when GNOME_DESKTOP_SESSION_ID is unset" [Undecided,In progress] [15:10] Laney, you are sure that ever worked since we use the backend rather than the js code we had? [15:11] seb128: I just found the fix [15:11] one liner! [15:11] Laney, great ;-) [15:11] looks like the gsettings isn't hooked up though [15:11] was it a missing init something? [15:11] you have to invalidate the model after re-sorting it [15:12] do you have the bug # handy? [15:16] Laney, https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1320160 [15:16] Ubuntu bug 1320160 in ubuntu-system-settings (Ubuntu) "Sorting by size in "storage" doesn't work" [Medium,Triaged] [15:16] ty [15:17] lemme hook up the gsettings key [15:30] oh, it's meeting time! [15:30] hey! :) [15:32] qengho, Laney, tkamppeter_, larsu, didrocks, hey, it's meeting time [15:32] short list today [15:32] Heya, folks! [15:32] morning guys [15:32] Bjoern is on vac, Canada is having a national holiday [15:32] Vacations? [15:33] qengho, hey [15:33] qengho, you can start I guess ;-) [15:33] Okay! [15:33] * Set up chromium-mir build tree. (Current status, upstream bug: linker failures with ozone enabled.) [15:33] * Worked with Legal to satisfy Ad*be that we know how to deliver Fl*sh plugin. [15:33] * Still testing chromium-browser 35.0.1916.153. I delayed to fix a Aura high-dpi geometry problem, but it's taking too long. Just releasing soon. [15:33] EOF [15:34] qengho, thanks [15:34] Laney, hey [15:35] • Continue to fix folks/zeitgeist fallout in Debian, upload new folks upstream. [15:35] • Investigate / cherry-pick fix for glib/gcc-4.9 test failure [15:35] • Fix ubuntu-sso-client adt failure [15:35] • Merges/syncs: vte3 gupnp glade [15:35] • Patch pilot [15:35] • Fix and fwd nautilus segfault upstream (applied, thanks mclasen) [15:35] • Discuss webkigtk/arm64 with Debian [15:35] • Test/discuss gnome-desktop 3.12, give feedback, need to test second round of fixes from Tim [15:35] • u-s-s: show all click icons, make click list sorting work, reviews [15:35] • Chat to fluendo about a segfault in their plugin, get it fixed, upload to Debian & sync [15:35] • Give desktop-next a little warning on its cdimage page directing to the wiki [15:35] • Fight with a flaky VPS [15:35] ☘ [15:36] what was the fluendo segfault? [15:36] is that in that lib (don't remember the name) that didn't change for years? [15:37] I had an ancient mp3 that was a bit corrupted [15:37] and its parser didn't like that [15:37] got a mediascanner segfault on every startup because of this [15:37] oh, k [15:37] metadata encoding problem? [15:38] what's the status-quo on the gnome-desktop discussion? [15:38] * seb128 lost track [15:38] we are still trying the new service approach? [15:38] rather than robert_ancell code copy one? [15:39] https://core.fluendo.com/gstreamer/trac/changeset/2622/trunk/gst-fluendo-mp3/src/bitstream.c [15:39] yeah just trying darkxst's stuff for now [15:39] k [15:39] but he said he'd be fine doing the other approach if necessary [15:39] right [15:39] let's see how that goes [15:39] Laney, thanks [15:39] tkamppeter_, hey [15:40] no tkamppeter_? [15:40] larsu, hey [15:41] hi! [15:41] last week I did some gsettings-qt cleanup and bug fixing [15:41] added a simpler dbus api to indicator-messages for the push notification guys [15:42] (who I've talked to a lot because libmessaging-menu isn't awesome for their use case on the phone) [15:42] nobody's given me any feedback on that yet, though [15:42] I've also talked to Company and mclasen about the Adwaita -> gtk move and how it effects our theme [15:43] and tested it with master - seems to be fine so far [15:43] also, the dialog/csd patches landed after some cleanup [15:43] hm, I think that's about it ... [15:43] great for master and the csd patches ;-) [15:44] larsu, you got feedback on https://code.launchpad.net/~larsu/indicator-messages/add-simple-dbus-api/+merge/224622 [15:44] did you ask push guys for review on the MP? [15:44] but not review feedback [15:44] ah, I've done further work on getting the theme ready for 3.12 (some notebook fixes and getting headerbars to work for us) [15:44] just "user" feedback [15:44] who should be reviewing that? [15:44] seb128: Chipaca is the push notification guy I talked to [15:44] charles? [15:44] fdo notifications sounds nice ;-) [15:44] could we actually have that? [15:45] Laney, don't get larsu started [15:45] seb128: I think ted wanted to do this [15:45] but I want to :( [15:45] Laney: yes, we could, but don't get me started. [15:45] we all would like that here [15:45] but other teams are not on the same line [15:45] ah, some code reviews and the usual bug discussion as well last week. I think that's it then [15:46] ho hum [15:46] seb128: right. There's also some problems with click packaging and org.freedesktop.Application [15:46] and XDG_DATA_DIRS [15:46] I hope those will get sorted sooner rather than later [15:46] right [15:47] let's not go into too much detailed discussion on those topic during the meeting, we could spend an hour on it [15:47] the API I proposed now doesn't have the org.freedesktop.Applicaiton part, otherwise it's very similar to the one GNOME is using now [15:47] so we could consolidate them if we decide that it's worth it [15:47] great [15:47] seb128: ya, sorry. [15:47] larsu, thanks [15:47] nothing to be sorry about ;-) [15:47] * seb128 gets confused by the order with the missing Canadians [15:48] I think it's didrocks' turn? [15:48] yep! [15:48] ;-) [15:48] * added more and more tests (150+ now) in the developer tools center [15:48] * shell completion enablement of udtc === tkamppeter_ is now known as tkamppeter [15:48] * fighting/had fun to define a model where I can send arbitrary UI from framework. A lot of force and back and now, expanding it :) [15:48] * finally, enabling the android studio setup, using that for the base of the future interactions with framework (and will probably refactor + add tests once the base will be done) [15:48] * some MIRing (still tracker on the list) [15:48] EOF [15:48] seb128, IMO this would be better for ted or tvoss to review; this looks like it's a continuation of the discussion from Malta [15:48] charles, ok, thanks [15:48] s/ted/tedg/ [15:49] didrocks, you forgot the "fought with python-gi" ;-) [15:49] I was trying to be politically correct in the notes :p [15:49] charles: I've pinged tedg a couple of times about it - is he on vacation? [15:49] oh cool we got new clutter in release after libinput [15:49] lol [15:49] thanks didrocks! [15:49] ok, let's not get you started about that then :p [15:49] Laney: yw ;) [15:49] seb128: yeah :) [15:49] didrocks, thanks [15:49] Sorry, I missed it. Here is my stuff: [15:49] ok, my turn [15:49] ups [15:49] - Got hint to new driver project for proprietary Brother laser printers, "brlaser": https://github.com/pdewacht/brlaser, added entries to OpenPrinting [15:49] - cups-filters: Fixed bug in cups-filters, generating queues pointing to remote raw queues [15:49] - ghostscript: SRU for a crash [15:49] - Mentoring of GSoC students, midterm evaluations [15:49] - Some organizational stuff for OpenPrinting Summit. [15:49] tkamppeter, sure, go ahead [15:49] - Bugs. [15:50] tkamppeter, thanks [15:50] k, my turn this time ;-) [15:51] • debian merges (evince) [15:51] • SRUs (shotwell) [15:51] • sponsoring (synced cinnamon) [15:51] • fixed an issue with the shotwell/online accounts integration [15:51] • landings (overlay-scrollbars) [15:51] • discussed/reviewed the options for gnome-desktop updates/what to do with u-s-d/u-c-c [15:51] reviewed touch settings current designs, looked at some issues, filed some bugs, triaged bugslist [15:51] • bug fixes (shotwell) [15:51] [15:51] über triaging [15:52] right, did quite some on u-s-s .... there is quite some work to do there :/ [15:52] larsu, ted is on vacation but I don't know for how many days, checking the calendar now [15:52] Laney: 100 points for proper use of umlauts [15:53] thänkß lärßü [15:53] larsu, ah, he's in DC for Independence Day [15:53] ok [15:53] charles: is the the one supposed to work on that api? [15:53] is there any other topic we should discuss? [15:53] that's right, we all chatted about that last week [15:54] charles: or can you/tvoss/thostr give input as well? [15:54] the situation is a bit defused now, but only because the push guys did more work than necessary :-/ [15:56] seems like that's a wrap then [15:56]