=== jono is now known as Guest60170 [07:36] Saviq, hi [08:08] Saviq: nevermind, found what I was looking for :) [08:08] (Still hi, though ;-) ) [08:15] davidcalle, o/ [08:32] mhr3, bug #1300649 [08:32] bug 1300649 in unity (Ubuntu) "Schemas should not come with the library package" [Undecided,New] https://launchpad.net/bugs/1300649 [08:48] Saviq: hi. [08:48] mzanetti, hey [08:48] Saviq: anything left to fix for silo 15? [08:49] mzanetti, I just kicked the touch session to build v0.107 [08:49] ah ok, cool [08:49] mzanetti, but also am thinking that we should Breaks: in unity8, not sure what, yet :) [08:49] but ubuntu-touch-session sounds like a good candidate [08:49] hmm... thinking... [08:50] * greyback has his pom-poms at this side, is ready to dance the "right-edge landed" dance [08:50] haha :D [08:50] yeah man... I'm in [08:51] otherwise we can upgrade unity8 itself, which will mean e.g. surfaceflinger starting if lxc-android-config isn't upgraded, but unity8 is incompatible with the old app manager, even if it's installed [08:51] didrocks, ↑ unity8 Breaks: ubuntu-touch-session (<< $new) do you think? [08:52] and why the heck is unity8-fake-env on the image!!?? [08:52] I can purge it straight after flashing, so why is it there? [08:52] Saviq: the dist-upgrade pulls it in still? [08:53] Saviq: unity8-autopilot perhaps? [08:54] now that we have emulators for others in there (not sure if someone uses them yet) it might be pulled by someone, and in turn would pull fake-env [08:54] Saviq: yeah, I would say unity8-autopilot [08:54] Saviq: and if you re not comptible with the old app manager, +1 on the breaks [08:54] didrocks, but we don't install unity8-autopilot [08:55] hmm... [08:55] didrocks, the only rdep of unity8-fake-env on device is unity8, and it's not a direct one, only through the virtual (old) qtubuntu-shell [08:57] Saviq: hello! So, we still see some unity8 crashes on the previous image (#270) - could you guys take a look and see if the crash file is big enough now to retrace it? [08:57] Saviq: http://ci.ubuntu.com/smokeng/trusty/touch/mako/270:20140331.1:20140331/7488/unity8/ <- for instance here [08:58] sil2100, we never said they're fixed yet ;) [08:58] bug #1297350 [08:58] bug 1256360 in Mir "duplicate for #1297350 unity8 crashed with SIGSEGV in glDeleteTextures() from mir::scene::GLPixelBuffer::~GLPixelBuffer() from mir::scene::ThreadedSnapshotStrategy::~ThreadedSnapshotStrategy()" [High,New] https://launchpad.net/bugs/1256360 [08:59] Saviq: so, it's basically only a crash on stop? I was away yesterday so I probably missed something ;) [09:00] But it's a really old bug it seems [09:03] sil2100, yeah, it's only on exit indeed [09:03] sil2100, would still be good to fix, though [09:03] hmm [09:03] A version (7.84+14.04.20140327.1-0ubuntu2) is available at the destination archive for that component but is not in the destination branch which is still at 7.85-0ubuntu1. You need to ensure that your version contains the fix in the destination or you can force rebuild to bypass the check. [09:04] ah it's didrocks's direct upload, /me forces for now [09:04] Yeah, you can force or simply cherry-pick the changelog to your merges [09:04] didrocks, unless you have the MP almost ready? [09:07] Saviq: I have it, but the meeting overruns [09:07] didrocks, ok, I'll wait a bit [09:10] * Saviq no gets it :| [09:14] Saviq: should be in your way in 5 minutes, once we wrap up :p [09:15] (blame dave :p) [09:16] ah crap [09:16] mzanetti, pull from ~unity-team/ubuntu-touch-session/drop-sf-support please into your branch and push again [09:16] mzanetti, /me forgot I dropped that branch [09:17] and that's why we don't have the session 0.107 yet [09:17] ah ok [09:18] Saviq: pushed [09:18] mzanetti, thanks [09:20] Saviq: https://code.launchpad.net/~didrocks/unity8/get-needed-dep/+merge/213619 [09:20] phew, done :p [09:21] let me check for unity8-fake-env [09:21] $ reverse-depends unity8-fake-env [09:21] Reverse-Depends [09:21] =============== [09:21] * unity8 [amd64 armhf i386] [09:22] unity8-fake-env | qtubuntu-shell, [09:22] qtubuntu-shell is installed on the iso? [09:23] (it's a provided package) [09:24] seb128, !!! [09:25] Cimi, ? [09:25] seb128, :P [09:25] seb128, was calling you here with saviq [09:25] what? [09:25] seb128, unfortunately I'm not sure diagnostics works [09:26] seb128, I set up a 'watch qdbus --system com.ubuntu.WhoopsiePreferences /com/ubuntu/WhoopsiePreferences org.freedesktop.DBus.Properties.GetAll com.ubuntu.WhoopsiePreference' [09:26] seb128, and system settings seems only to change ReportCrashes [09:26] not AutomaticallyReportCrashes [09:26] can you mention it on a channel where Laney and ev are? [09:26] like #ubuntu-devel [09:27] sure [09:36] didrocks, thanks [09:36] yw! [09:47] didrocks, can you please fill the checklist in description of https://code.launchpad.net/~didrocks/unity8/get-needed-dep/+merge/213619 [09:48] Saviq: done [09:48] Saviq: well, actually… there is an UI change (from blackscreen to "getting unity8" :p) [09:50] didrocks, subscribe a designer, then! [09:50] hehe :) [09:52] xnox, hey, we just tried x-building unity-scopes-api and found that capnp is completely not multi-arch (e.g. libs are not installed in m-a dirs, capnpc tool depends on libs of the same arch), what do we do in a case like this? [09:52] it's synced from debian directly, too, AFAICT [09:52] or maybe not, we have a -0ubuntu1 there [10:41] tsdgeos: hey, about this one: https://code.launchpad.net/~aacid/unity8/new_tabbar/+merge/210453 [10:41] tsdgeos: is this still valid with new-scopes? [10:41] mzanetti: yes [10:41] it needs the sdk change to land [10:41] ah ok [10:41] so still needs to wait a bit [10:42] yes === MacSlow is now known as MacSlow|lunch [11:17] Saviq: all untilities should typically be M-A:foreign, let me fix that one. [11:19] didrocks, unity8 links to libunity-mir1, shouldn't shlibs:Depends result in a dependency for a particular version of unity-mir? [11:19] didrocks, or does a Depends: libunity-mir1 | foo override that? [11:20] Saviq: hum, good question, not sure if shlibs is that smarts [11:21] didrocks, I thought without .symbols files it would depend on direct version... [11:21] Saviq: it would be (>= …) [11:21] yeah [11:21] oh [11:21] override_dh_makeshlibs: [11:21] dh_makeshlibs -Nunity8-private -Nunity8-fake-env [11:22] didrocks, you did that :) [11:22] didrocks, but anyway it's unity8 itself that's linking, not unity8-private [11:23] Saviq: ah ok, so yeah, it should pick it, I think removing the dep and doing a build would be a way to see if shlibs is so smart to ignore in that case [11:23] * Saviq thinks it's the direct dep on libunity-mir1 that's causing that :| [11:23] Saviq: want me to try or you have a build ready off-hand? [11:23] didrocks, I will [11:24] looks like we should split the plugin out of libunity-mir1... [11:24] Saviq: Unity.Application? [11:24] mzanetti, yeah [11:24] uh... why is that? [11:29] mzanetti, see up, looks like shlibs:Depends isn't smart enough to add the >= to that dependency from linking [11:30] Saviq: hmmm... I see... so you want me to split it out now? [11:31] mzanetti, no, I think I'm at the point where I say "fuck it" [11:31] it's all gonna get upgraded together anyway [11:32] * mzanetti is happy Saviq joined "the point" :D [11:32] * Saviq just does the last tests then [11:32] yeah. while I agree in other scenarios it would be problematic indeed, in this particular case I don't think its worth the troubles to make it 100% correct [11:33] as noone will just upgrade unity8 on his phone [11:39] Saviq: so i have fixed up capnp, as long as it is used as generated and /normal/ cmake cross-compiler is used to compile the generated code. [11:39] xnox, awesome, thanks! [11:40] ok wrong, we're *not* linking to libunity-mir1 === alan_g is now known as alan_g|lunch [12:16] someone has got a minute? https://code.launchpad.net/~mzanetti/unity8/fix-make-trySomething-qt-5.2/+merge/213639 === alan_g|lunch is now known as alan_g === _salem is now known as salem_ [12:58] mzanetti, good catch, thought it would've been a bit more complex [13:15] Saviq: fixed capnp is in the release pocket now. [13:36] Saviq, can you retry unity8 in silo 002? I think it got stuck on libunity-dev any/all nonsense [13:37] mterry, been trying since the morning... [13:37] mterry, but it should in theory work now, so kicking a build now... [13:37] Saviq, OK thanks [13:38] mterry, actually it's still running, should be uploaded sooon [13:38] oh huh, the +packages page doesn't appear to be running [13:38] Saviq: unfortunately its not the end of the story. some rare cases still exist when it wants to call QTest::TestLogger and asserts on that. will hunt that down too [13:38] mterry, ? [13:39] mzanetti, okies [13:39] Saviq, https://launchpad.net/~ci-train-ppa-service/+archive/landing-002/+packages doesn't make it look like unity8 is building [13:39] mterry, it's not, the packages are still being prepared in the train [13:39] Saviq, and then get binary copied there? [13:40] mterry, no, source [13:40] Saviq, huh OK [13:40] standard dput upload [13:49] mterry, refresh, everything's building now [13:50] Saviq, sweet [14:04] mterry, shall I install the utils plugin in ubuntu.systemsettings.wizard.utils? [14:06] Cimi, feels better to me yeah [14:06] ok === alan_g is now known as alan_g|tea [14:27] mterry, done [14:28] mterry, I thought the notifications appeared at first boot [14:29] paulliu: https://wiki.ubuntu.com/Touch/Emulator [14:29] mterry, you're right [14:30] MacSlow, what starts the notification service? [14:30] Cimi, within unity8? [14:30] Cimi, the shell itself... via a plugin (that's where the daemon lives) [14:31] mterry, how can we tweak the wizard to make it run them? [14:34] @unity: standup [14:34] Cimi, well aren't we eventually going to have built-in-prompts for that in the list? [14:35] mterry, not now [14:36] Cimi, maybe that's the next thing to work on then before enabling wifi for real [14:41] mterry, Could you look at silo 4 to make sure that works for you as well? === alan_g|tea is now known as alan_g [14:42] tedg, OK give me a sec [14:43] mterry, NP, I'm just setting it up, just wanted to get on your TODO list. [14:43] waiting for a small review here: https://code.launchpad.net/~elopio/unity8/search_updated/+merge/213155 [14:43] elopio, mterry I missed you two guys, you can text me your update and I'll add it [14:44] Cimi, oh just MIR work, updating split branches for mir/devel, and reviewing your branch [14:44] Cimi: working on a UX test for installing a click app through the scope. [14:45] mterry, I don't think it's possible now [14:45] mterry, for the way the authentication works [14:46] mterry, so what we want it snap decisions to appear [14:46] Cimi, well... we could import the same plugin from unity8 [14:46] for snap decisions? [14:46] Cimi, the Notifications plugin. We might need to point to its private import dir [14:47] Cimi, yeh [14:49] mterry, ok then [14:49] mterry, if Saviq is fine [14:49] he's actually not at the pc right now, so we can make decisions :D [14:49] \o/ [14:50] Cimi, I'm not sure how tied into unity8 that plugin is. maybe it's not standalone, maybe it is [14:50] Cimi, If that's all it takes, steal his computer ;-) [14:51] tedg, good you are here :D [14:51] tedg, is the location service actually working? [14:52] Cimi, I think it only works with GPS straight up, so like 15 min to lock. [14:52] Cimi, Though charles or tvoss may know more than me. [14:52] have any of you guys used qtcreator profiler? [14:53] is it really reliable in what it returns? [14:55] * tedg is using Stallboard, his new favorite app, to test indicator-sound [14:55] tedg, does the indicator actually do something? [14:56] location [14:56] tedg, its always off for me [14:56] I can turn it on [14:56] but after a while is off again [14:58] Cimi, Not sure, I know there are a few MRs waiting on that. === dandrader is now known as dandrader|afk [15:05] mterry, so we can use it if we want, just need to update the import path [15:07] Cimi, sounds good === gatox is now known as gatox_lunch === dandrader|afk is now known as dandrader [15:43] hate how easy is to make qml crash http://paste.ubuntu.com/7190231/ :/ [15:48] hmm... I still get this with the latest image "Unable to connect to Upstart: Empty address ''" [16:03] mterry, So I'm happy with silo4, when you're happy feel free to mark it as tested. [16:09] tedg, I'll ping ya yeah. Not sure I have access to mark it as tested? [16:09] tedg, you want me to test with my split stuff in touch I assume? [16:10] mterry, Yes, if you don't I can, just trying to remove sync points :-) [16:10] mzanetti, unset UPSTART_SESSION [16:10] mzanetti, it's fixed in the right edge silo [16:14] mzanetti, there's something wrong on flo still... launch battery settings from the indicator [16:14] the app is positioned like it'd be in the side stage... [16:15] ah... sounds like greyback's fix ^ [16:15] * Saviq wants [16:16] Saviq: no, I mean he fixed that for the phone. sounds like it's either not fixing it good enough for flo or breaking it there. [16:17] mzanetti, also, if you can check... long-swiping from left is really slow here [16:17] I'll check [16:17] mzanetti, on mako [16:24] mzanetti, Saviq: regarding the exposure of the NotificationModel's Role-enums... do they not also need to be added to lp:unity-api/include/unity/shell/notifications/Enums.h ? [16:24] MacSlow: no [16:28] mzanetti, it was temporary, fine now (the slowness) [16:28] Saviq: ah ok [16:29] mzanetti, so the flo issue, is the fix you mentioned merged already or? [16:29] * Saviq purges ppa === gatox_lunch is now known as gatox [16:36] Saviq: no. we're on it [16:36] mzanetti, ok, that seems to be the last issue then [16:36] LET'S FUCKIN' LAND IT [16:36] ack, thanks [16:45] mzanetti, when launching an app, I can see the previous one until the new one starts drawing, that expected? [16:45] mzanetti, seems to be a regression from trunk :| [16:45] Saviq: that is a timer which displays a "splash" image as we don't have any means of knowing when an app painted [16:46] I tried to find a good compromise between not showing it too long and too short [16:46] do you see them really often? [16:46] mzanetti, every time, yeah [16:46] mzanetti, launch an app, go to dash, launch another app [16:47] mzanetti, I see 2s white, 2s old app, new app [16:47] oh... but it shouldn't happen that the app takes 4s to start [16:47] which usually doesn't happen here [16:50] mzanetti, and we can't show the white all the time can we? [16:50] mzanetti, and it takes 5s to launch dialer on Albert's phone... [16:50] dafuq... [16:50] i.e. trunk [16:50] mzanetti, that's the usual AFAICT [16:50] mzanetti, and still I don't see the old app behind it [16:51] Saviq: hmm... I'll try to improve it [16:51] mzanetti, all apps take that long to launch here on mako, landing 15 or not... [16:52] dialer just started in 3 secs here. but anyways. I agree trunk handles the situation a bit better === alan_g is now known as alan_g|EOD [17:19] mterry, finally got papi and usc to build, should get into the ppa soon === dandrader is now known as dandrader|lunch === dandrader|lunch is now known as dandrader [21:09] umm.. [21:10] anyone seen this? [21:10] FAIL! : TimeFormatterTest::testFormatStrF() Compared values are not the same Actual (formatter.timeString()) : 01-04-2014 11:57 Expected (time.toString("dd-MM-yyyy hh:mmAP")): 01-04-2014 11:57IP. Loc: [/home/antti/branches/unity8/build-area/unity8-7.84+14.04.20140327.1/tests/plugins/Utils/timeformattertest.cpp(62)] [21:10] what is that IP.. [21:11] seems my irc client broke the formatting [21:11] http://pastebin.ubuntu.com/7191617/ === jhodapp is now known as jhodapp|afk === salem_ is now known as _salem === popey_ is now known as popey [22:02] filed this. [22:02] https://bugs.launchpad.net/unity8/+bug/1301038 [22:03] Ubuntu bug 1301038 in Unity 8 "TimeFormatterTest does not set environment properly." [Undecided,New] [22:03] Saviq: ^ [22:03] Wellark, read CODING [22:04] Wellark, there were *some* reasons for not forcing a default locale, not sure which atm... [22:06] Saviq: well, unit tests should not "expect" anything from the environment. If they require C locale then that should be set as part of the test setup :) [22:08] also, if the TimeFormatter is locale aware then I would expect it to have tests for explicitly testing that our custom formatter returns correct results for all of our most important locales [22:08] chinese come to mind.. [22:10] http://en.wikipedia.org/wiki/List_of_languages_by_number_of_native_speakers [22:10] the top two seem quite good test cases [22:10] for anything locale aware components [22:12] Saviq: and given this: http://insights.ubuntu.com/news/press-releases/canonical-announces-first-partners-to-ship-ubuntu-phones-around-the-globe/ [22:12] I would say testing with Mandarin and Spanish would be even more relevant than English ;) [22:13] Saviq: anyway. feel free to mark the bug "Wontfix" === Pici` is now known as Pici [22:45] I don't suppose anyone can tell me how I'm supposed to detect minimize events when using unity as a window manager? [22:45] unity masks the iconify event somehow [22:46] https://bugs.launchpad.net/unity/+bug/998073 [22:46] Ubuntu bug 998073 in Unity "unity hides minimize event from application" [Low,Triaged] [22:46] how do I get around this [23:24] Wug: What do you want to do with the minimise event? [23:43] RAOF: minimize a slave window