[05:03] Saviq: I marked the unity8 branch at https://code.launchpad.net/~aacid/unity8/initialize_m_sceneDistance/+merge/208773 now at least, in case it doesn't go in the other route (l28) [05:03] Saviq: Qt is about there, yes, now the 80 other source packages === iahmad is now known as iahmad|afk [07:59] Mirv, yup, ok [08:37] Saviq: dude, is the bluetooth crash fixed? i can't reproduce it anymore :_S [08:37] on the destkpo [08:37] * tsdgeos tries the phone [08:47] tsdgeos, not that I know of, checking [08:48] tsdgeos, crashed here [08:48] garg [08:49] Saviq: you totally dist-upgraded? [08:49] tsdgeos, upping last 6 here [08:50] tsdgeos, yeah, d-up'ed [08:50] tsdgeos, crashed [08:50] desktop too? [08:50] tsdgeos, yes [08:50] tsdgeos, let me try and drop landing 006 ppa [08:51] :( [08:51] still trying to update the phone, for some reason doesn't seem to be getting interwebs [08:53] ah, i ran out of disk space [08:53] tsdgeos, btw, you might want to use landing-006 (ppa:ci-train-ppa-service/landing-006) where the "real" qt 5.2 is now (you need qt5-beta2 still) [08:54] let's see [08:57] ignore me, i had commented one line that makes it crash :D [08:58] bad me not remembering last minute change on friday [08:58] :) [09:47] dednick, hey, can I steal you a bit? [09:47] Saviq: i guess. depends what you mean to do with me ;) [09:48] dednick, I've been prepping ourselves for the 5.2 migration, and there's a bunch of failing tests in tst_Indicators and tst_Panel [09:49] dednick, and I kind of got lost in them [09:49] Saviq: ok. unity trunk have all 5.2 changes? [09:49] latest i mean [09:49] dednick, no, but trunk is fine to look at that [09:50] dednick, you want ppa:ci-train-ppa-service/landing-006 and ppa:canonical-qt5-edgers/qt5-beta2 [09:50] Saviq: ok. i'll take a look. I have a feeling i know what the failing tests are about. [09:50] dednick, also, this should be fine to be applied: http://paste.ubuntu.com/7026599/ [09:50] dednick, no need for the wrapping Item [09:51] Saviq: ok [09:53] dednick, here are the failures I've seen: http://pastebin.ubuntu.com/7026616/ [09:53] dednick, one caveat: try* doesn't work with 5.2 yet, we need to trick Qt into running them with qmlscene [09:54] Saviq: yeah. qml changed the way ListViews are populated or something. order of ListElements no longer well defined or something silly. [09:55] dednick, one visible thing is that there's a big yellow square in the top-right corner, not sure where that comes from [09:55] dednick, it doesn't seem to be any of the indicator widgets [10:13] Saviq: dude, wtf. 5.2.1 totally screwed up with package installs. [10:13] dpkg: error processing archive /var/cache/apt/archives/libqt5core5a_5.2.1+dfsg-1ubuntu2_amd64.deb (--unpack): [10:13] trying to overwrite '/usr/share/qtchooser/qt5-x86_64-linux-gnu.conf', which is also in package libqt5core5:amd64 5.0.2+dfsg1-7ubuntu18 [10:13] dednick, libqt5core5 should be replaced with libqt5core5a [10:14] dednick, so it should be gone before installing the 5.2.1 version... [10:14] dednick, can you pastebin the whole log? [10:18] Saviq: http://pastebin.ubuntu.com/7026709/ [10:19] libqt5core5 is just being upgraded to 5.2.1 [10:19] dednick, so it did that in the wrong order [10:19] Mirv, ↑↑ [10:20] Saviq: know how to fix that? [10:20] without synaptic? since it doesnt seem to want to load anymore [10:20] dednick, force-upgrade libqtcore5 first (dpkg -i /var/cache/apt/...) [10:21] dednick, then, dpkg --configure -a should be enough (might need to follow with apt-get -f install) [10:21] Saviq: libqt5core5:all 5.2.1+dfsg-1ubuntu2 (Multi-Arch: no) is not co-installable with libqt5core5 which has multiple installed instances [10:22] dednick, so if you have libqt5core5:armhf / :amd64, dist-upgrade will fail? [10:23] Mirv, rather ↑ [10:27] Saviq: i know why the crash of bluetooth is happening, i don't just know yet how to fix it, but basically the outer delegate of a list is being deleted while a function of the inner one is being executed, which at some point causes bad things to happen [10:28] ideally one should not delegate outer delegates while the inner one is getting something executed i'd say [10:28] tsdgeos, there a way to enforce something like this? [10:28] no idea :D looking at the code a bit [10:40] Mirv, I just got the same error when upgrading my phone from stock image to qt5-beta2 + landing-006, so I'd say something's wrong with the transition between 5 and 5a [10:43] * greyback back in 20 [10:52] Saviq: I noticed it too, it's because of the transitional package that was temporarily wanted to be able to test ABI breakages but which is going to go away. on my desktop I just tried again and it progressed fine, it should just have better conflicts in it [10:53] i had to remove all of qt5.0.2 to install. [10:57] Saviq: make tryXXX doesnt work anymore either. [11:03] [10:53] dednick, one caveat: try* doesn't work with 5.2 yet, we need to trick Qt into running them with qmlscene [11:04] Mirv, I had to force (dpkg -i) installation of the new libqt5core5 before it continued [11:04] (on my phone) [11:04] Saviq, anyone tried new scopes with 5.2? [11:04] ie should i? [11:04] mhr3, do [11:04] k [11:04] mhr3, use qt5-beta2 and landing-006 ppas [11:05] wondering if i want to break my desktop or rather the phone [11:05] think i'll go for the phone first :) [11:08] xnox, did the py3 transition land whole https://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-trusty/3180/testReport/unittest.loader.ModuleImportFailure.unity8.application_lifecycle/tests/test_url_dispatcher/ ? [11:12] Saviq: it was in silo since last week, bzoltan has asked me to rebase 4 times now.... and still didn't land it as far as I can tell. [11:13] xnox, k === MacSlow is now known as MacSlow|lunch [11:47] Saviq: did you want me to patch the IndicatorItem, or has it been done elsewhere? [11:48] dednick, please do [11:57] Saviq: https://code.launchpad.net/~nick-dedekind/unity8/indicator-tests-qt5.2.1/+merge/209048 [11:57] dednick, thanks! [11:59] dednick, it will work with 5.0, right? [12:00] Saviq: should do... [12:00] dednick, ok, removing the "5.2.1" from the description, since it's not actually required [12:01] Saviq: oh, done already :) [12:02] dednick, yup [12:08] wow, 4K recording on Sony Z2. probably get aout 5 seconds worth on the internal storage.... [12:22] Saviq, ok to push trunk merge to new-scopes? [12:22] there were some conflicts, but nothing major === alan_g is now known as alan_g|lunch === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g [13:43] Saviq, what's the story with uitk? is the demo ppa version still needed to run new-scopes? [13:44] mhr3_, yeah, 'cause new-scopes accrued some mwc stuff that was unrelated... [13:45] :/ [13:45] Saviq, also did you reply to the trunk merge? i might have missed that [13:46] greyback_: step 1: https://code.launchpad.net/~mzanetti/unity8/support-short-appid-in-gsettings/+merge/209071 [13:49] mhr3_, I might have not? you mean -scopes-shell? [13:49] Saviq, no, merge lp:unity8 into new-scopes [13:49] if you're ok with that [13:50] got it conflict-resolved in my tree here [13:51] mzanetti: could there be a danger of one appId having 2+ desktop files available in the search directory? Is the first valid desktop file definitely the one wanted? [13:54] greyback_: no, there shouldn't be. We only support one version per user [13:54] mzanetti: ok [13:55] greyback_: well, UAL's app upgrade/uninstall hook would fail, then we might have 2, but that would be a bug imo [13:55] mzanetti: so not possible for an app to have a system installed version, and a user-specific one too [13:56] greyback_: not for click. in that case, the system one would be "foo" while the click one would be "com.ubuntu.foo" or the like [13:56] greyback_: not exactly sure what happens when we manually make install an application into ~/.local tho. [13:56] greyback_: but that seems to be a general thing our current implementations are suffering too [13:57] mzanetti: probably not. [13:59] mhr3_, ah yeah, go for it [14:30] @unity: standup [14:30] Saviq: I'll mark I don't attend, but please at least temporarily don't remove me from in the calendar event, k? :) [14:31] karni, clinging on, eh? ;D [14:31] aye ;D [14:34] Saviq, pushed trunk merge into lp:~unity-team/ubuntu-ui-toolkit/rotate-n7 [15:03] elopio, ah, on unity8 taking a long time to stop - I saw that sometimes, no real way to reproduce, and it's not crashing, either, so it will need investigation [15:03] elopio, if you find that there's a test that exposes it more often than others, please let us know [15:05] Saviq: on my phone, it seemed to happen for every test. It's like 10 seconds setting things up, and like 30 seconds doing nothing waiting for it to stop. [15:06] I don't know how to collect extra information about it. [15:06] elopio, Ran 46 tests in 1202.536s for me - do you have a number? [15:06] Saviq: no, I ran them one by one. [15:06] Ran 2 tests in 42.404s [15:06] OK [15:09] elopio, can you check http://pastebin.ubuntu.com/7027986/ [15:09] Saviq, seen http://paste.ubuntu.com/7027988/ ? [15:09] mhr3_, https://code.launchpad.net/~unity-team/unity8/fix-5.2-tests/+merge/209058 [15:10] mhr3_, 5.2 is more strict (and thanks to it for that) [15:10] hum, that one stoped it immediately. [15:10] * elopio keeps trying. [15:10] Saviq, very well... anyway, didn't notice anything majorly broken with new-scopes and 5.2 [15:10] mhr3_, anything minorly broken? [15:11] mhr3_, other than with 5.0? [15:11] Saviq, only thing was that playing videos and then switching to dash resulted in everything being super slow and then unity8 crashing [15:12] Saviq, oh and volume down doesn't work :) [15:12] (mako) [15:13] not like those are new-scopes specific [15:14] mhr3_, k [15:18] tsdgeos: what's the plan for the autopilot tests fixes? [15:18] https://code.launchpad.net/~elopio/unity8/new-scopes-cleanup-autopilot/+merge/208954 [15:18] should we merge into your cleanup branch? [15:20] elopio: sorry i'm 100% on trying to figure out one of our Qt 5.2 crashes, but looks good enough i'd say [15:20] i'll merge it to my branch [15:21] tsdgeos: ok, thanks. I'll be working on the camera fix probably tomorrow. [15:21] cool [15:22] oh, but I need a review here first: [15:22] https://code.launchpad.net/~elopio/unity8/fake_app_from_toolkit/+merge/208002 [15:23] mterry: that's from a file you reviewed before, can you take a look please? [15:24] * mterry looks === greyback is now known as greyback|rebooty === robru is now known as robru-at-doctor [18:54] mterry, From the greeter can I assume any user that starts with "*" should be considered a non-user? [18:54] tedg, there is *guest [18:55] tedg, but yeah then some weird ones like *remote... [18:55] tedg, not sure if you consider *guest a real user or not [18:55] mterry, Yeah, I'm trying to block guest and those out. [18:56] mterry, I only want users that would have set up a sound player in their session. [18:57] tedg, guest may have... [18:57] tedg, though not sure how you can get the real account name [18:58] maybe once we log guest in, unity-greeter's api should return its real name [19:00] mterry, I guess. I'm not too worried about ensuring guest has full features on the lock screen. [19:00] tedg, fair enough === robru-at-doctor is now known as robru-sick === jhodapp_ is now known as jhodapp [22:59] * mterry thought it had been too quiet; turns out IRC was off