=== marcusto_ is now known as marcustomlinson_ === marcustomlinson_ is now known as marcustomlinson [11:01] ltinkl, hey, checklist! https://code.launchpad.net/~josharenson/unity8/integrated_lightdm_refactor/+merge/260588 [11:01] :) [11:01] oy [11:02] mzanetti, to my defense, this was long ago ;) even before I started [11:02] :) [11:02] granted [11:29] cimi, hey, can you fix this one? http://i.imgur.com/OPVOm5P.jpg [11:30] mzanetti, try changing to us i18n [11:30] ? [11:30] mzanetti, shorter words :D [11:32] cimi, this is nothing yet :D. look at this: http://i.imgur.com/Be9AHyV.jpg [11:32] omg [11:33] although I would say the translation isn't really good [11:33] anyhow, the proper one wouldn't be much shorter [11:33] mzanetti: just use shorter words [11:33] another solution would be to not use Google, right? === MacSlow is now known as MacSlow|lunch [11:45] mzanetti, cimi : wrap the text in 2 lines? [11:45] yeah, or elide it at least [11:45] and yes, I've seen that too [11:45] in the Czech translation [11:46] although not as bad :) [11:46] might not be a bad idea to clip the scope while switching between scopes [11:47] greyback, and performance? [11:47] greyback, scope switching is already slow [11:48] greyback, anyway, will be fixed with scopes as apps [11:49] cimi: clipping a single big rect will have tiny impact. Clipping is bad if you do it in list/gridview delegates, where lots of delegates are on screen (breaks batching) [11:49] but yeah [11:50] cimi, well no.. .the label still shouldn't reach out of the shape [11:50] wtf, why is there Scottish Gaelic as a language option, but no Irish Gaelic! === alan_g is now known as alan_g|lunch [11:53] hmm there must be one guy doing the scottish gaelic translation. it's pretty good, just a few errors [12:17] dednick, are you using the MockMirSurfaceItem? [12:17] dandrader: in my work yes [12:17] dednick, I didn't detect this error as I'm only using the FakeMirSurfaceItem [12:17] dednick, I found it much easier to use a fake than a google mock [12:17] dednick, but, I'll fix it then [12:18] s/but/but ok [12:18] dandrader: i need to verify that a call is made on the surface [12:18] dednick, sure [12:20] mzanetti, greyback: i'm getting a very strange crash in the QSGRenderThread, but can't get a stack. any idea how to get a decent stack from that? [12:21] if i use a mousearea, it crashes, if i use a ubuntu button it doesnt. :/ [12:22] odd [12:22] dednick, not sure why you don't get a stack trace, however, one difference that comes to my mind is that the ubuntu button takes activeFocus, while the mouseArea doesn't [12:22] and only on device, not desktop. [12:22] dednick, dumb question: do you have the debug symbols installed? [12:22] some of them. qt5quick5-dbgsym is [12:23] and core & gui [12:23] dednick, you also need glib [12:23] * dandrader looks for his notes [12:24] dandrader: have that as well [12:24] dednick, you need these: libc6-dbg libdbus-glib-1-2-dbg libglib2.0-0-dbg qtbase5-dbg qtdeclarative5-dbg libubuntu-app-launch2-dbgsym [12:24] dednick, I made this list a while ago. might not be up to date anymore [12:24] hm. maybe because i'm using dbgsym [12:25] I only used dednick dbgsym for packages that don't have a -dbg... [12:27] will give it a try [12:31] dednick, the tons of deprecation warnings we are getting comes from the indicators. it seems that unity8 loads the indicators from somewhere else. where's this code? [12:35] dednick: qtdeclarative5-dbg the main one you'd need for a renderer crash, libglib2.0-0-dbg good too for tracing down into the event loop [12:37] dednick: usual reason for a renderer crash would be if a surface buffer has been deleted before it was removed from the scenegraph on the render thread === alan_g|lunch is now known as alan_g === MacSlow|lunch is now known as MacSlow [12:49] greyback, can we get https://code.launchpad.net/~dandrader/qtubuntu/improve-resize/+merge/258912 in? [12:49] dandrader: superceded? [12:49] greyback, I mean https://code.launchpad.net/~dandrader/qtubuntu/improve-resize/+merge/260950 [12:50] dandrader: yeah ok, I'll give it a quick check [12:50] want to land a few things [12:50] dandrader: a lot of the indicator items are in ubuntu-settings-components [12:51] dandrader: src/ubuntumirclient/input.cpp has nativeEventTypeToStr commented out. Should it just be removed? [12:51] greyback, it's useful when you want to debug things [12:52] dandrader: :how about #ifndef QT_NO_DEBUG ? [12:52] greyback, not enough. it will complain about unused function [12:53] greyback, but I don't care much. can remove if you want to [12:53] dandrader: it's used by the DLOG, which is also commented out. Can fix it with the QT_NO_DEBUG ifndef [12:54] it looks useful, let's not lose it [12:54] greyback, I would also have to uncoment the DLOG, which will make debugging very verbose by default [12:54] dandrader: DLOG only prints if built with debug mode [12:55] #if !defined(QT_NO_DEBUG) [12:55] #define DLOG(...) LOG(__VA_ARGS__) [12:55] #else [12:55] #define DLOG(...) qt_noop() [12:55] greyback, yes, but it will still make debug logs very verbose by default [12:55] greyback, debug logs in debug builds [12:55] dandrader: debug mode is not the default [12:56] that's a dev's problem :) [12:56] if you need to go in to uncomment a useful chunk of code, you can also go in to comment out a DLOG line that's useless [12:58] greyback, didn't quite get you. the thing is: I usually don't want those "UbuntuInput::customEvent(type=%s)" messages printed on every single event when I'm debugging qtubuntu [12:59] greyback, and that's why it's commented out. it's too verbose [12:59] dandrader: ok, it'll do, leave it [13:02] greyback, to make things prettier I could surround nativeEventTypeToStr and that DLOG with a "#if VERBOSE_DEBUG" or something... [13:03] dandrader: only if you want to. It's not worth arguing about :) [13:03] dednick: are you using this: https://code.launchpad.net/~gerboland/qtubuntu/window-close-support/+merge/249579 ? If itt's ok, please approve it and I'll try landing now [13:04] greyback: ya. ok [13:04] ta [13:04] greyback: done [13:05] greyback, hmmm there's already a "#define LOG_EVENTS 0" there, which I can use [13:06] dandrader: again, only if you want to. It's too minute a detail to really care about [13:07] doing it alread [13:07] already [13:07] :) [13:08] done [13:09] weird. not crashing now :/ [13:11] oh. that's because the problem is in media-hub/qtubuntu-media which were downgraded on reflash === dandrader is now known as dandrader|bank === dandrader|bank is now known as dandrader [14:43] dednick: you don't get any stacktrace? or the stacktrace is useless - no valid frames? [14:44] greyback: just a bunch of ? [14:44] 2 to be exact [14:44] dednick: ok, gdb unable to figure out the trace [14:44] ya [14:44] that might imply it crashes in the android/hybris side [14:44] which gdb can't figure out, as different ABI [14:50] dednick: https://wiki.ubuntu.com/Touch/Core/UbuntuDebugAndroid if you fancy a go [14:50] greyback: cool. thanks [14:52] mzanetti, do you know about these errors? https://jenkins.qa.ubuntu.com/job/unity8-wily-i386-ci/94/console [14:52] mzanetti, "tests/plugins/Utils/modeltest.h LGPL (v2.1) *No copyright*" etc [14:52] dandrader, yes, on it right now [14:52] mzanetti, is it a bug in the script? [14:52] dandrader, the licensecheck tool was updated and fails when a / is in the copyright line === alan_g is now known as alan_g|AFK === alan_g|AFK is now known as alan_g [16:58] greyback, kgunn, I think I can't work on the shell cursor yet. mir doesn't have a mouse input event type. I has only pointer. [17:02] dandrader: you'd best liaise with racarr/anpok to describe your requirements and get them on that job === dandrader is now known as dandrader|lunch === alan_g is now known as alan_g|EOD === dandrader|lunch is now known as dandrader === cwayne1 is now known as cwayne [19:32] is fullscreen boderless window support coming to unity8? windows in windowed mode that takes up the entire screen and overlap the top panel. [19:33] it's a feature used by video games [19:34] honestly the only thing stopping me from playing games on ubuntu [19:36] a normal fullscreen app will remove this IIRC (when i was testing some SDL stuff on unity8) [19:36] just like how the camera app works [19:36] ie. no border [19:36] yea but a fullscreen app requires me to alt tab in order to get access to my second screen [19:37] so if i'm waiting for friends to join me in a game, i'd have to alt tab just to accept their skype call [19:37] then im not sure :), i've not seen if theres a borderless option atm [19:38] also IIRC, fullscreen in mir will not change the res of the screen it self [19:39] soo that means in a sense they are already fullscreen borderless, the window just have to be the right size [19:39] as far as mouse grabbing there wont be any of that, and its still in the works on how relative mouse support will work [19:40] soo you'll just need to move your mouse over to skype and all will work [19:40] oh that sounds nice [19:40] still some things that need to work out for games :) [19:41] i hope wine will work the same way. if you fullscreen wine it can freak out the display [19:41] no idea if it does it for the beta [19:41] i dont think wine works on unity8 atm sooo i've clue how that'll even work [19:41] ha true [19:42] oh one last question, not sure if this is the place for it. as for snappy desktop, i assume it will eventually merge back into the current branch when it's ready right? [19:43] hmm #snappy might have a better answer [19:43] * bschaefer isn't really sure :) [19:43] ah okay thanks [19:43] i assume it's #ubuntu-snappy [19:43] np! Hopefully they have an answer there, im just on #snappy, idk if #ubuntu-snappy exists