=== TheMuso` is now known as TheMuso === jzheng_afk is now known as jzheng [09:01] @unity: correct, while PS Jenkins Bot works, trust that one, Unity8 doesn't run all the QML tests yet, just builds packages [09:11] kk [09:30] mzanetti: i set back to needs review your launcher branches [09:30] tests are failing [10:25] Saviq, hey, I’m looking at https://bugs.launchpad.net/qtubuntu/+bug/1433138 , trying to understand the code that forwards key events in src/ubuntumirclient/input.cpp, and more specifically the translateKeysym function, which apparently was written by Gerry but he doesn’t seem to be around, do you know who could help me make sense of it? [10:25] Launchpad bug 1433138 in qtubuntu "Under Mir, physical keyboard’s Return key does not work within web pages" [High,In progress] === jzheng is now known as jzheng_afk [10:48] oSoMoN: yeah gerry is on holiday [10:51] oSoMoN: what question do you have about that function? [10:59] tsdgeos, see my last comment on the bug report: https://bugs.launchpad.net/webbrowser-app/+bug/1433138/comments/12 [10:59] Launchpad bug 1433138 in qtubuntu "Under Mir, physical keyboard’s Return key does not work within web pages" [High,In progress] [11:03] tsdgeos, and a related question: where do I find the logs for qtubuntu? [11:08] oSoMoN: interesting [11:09] oSoMoN: what do you mean the logs? [11:09] in bzr? [11:11] mzanetti: did you see the other launcher branch also needs updating? [11:12] yes, on it atm [11:12] tsdgeos, ^ [11:12] k [11:26] tsdgeos, something is broken with the launcher [11:27] * mzanetti tries trunk [11:27] mzanetti: seems fine here, what do you find broken? [11:28] dismiss by tap somehow behaves really weird [11:29] seems pretty solid here on the latest phone image [11:30] there was a bug fixed lately about that [11:30] are you trying trunk or your branches? [11:30] maybe you somehow reverted it on your branches? [11:31] back in 15 === alan_g is now known as alan_g|afk [12:03] pstolowski: ping [12:05] tsdgeos, pong [12:05] pstolowski: taping on an artist in the my music scope [12:06] takes a long time to show me the artist "page" on the scope [12:06] at first i thought it was hung or something [12:06] long time like 10 seconds or more [12:07] interesting.. it's instant here. but maybe because i've all the images cached already [12:07] i think i saw the bug where we request insanely large image size for artist, but not sure [12:08] lemme try to find it [12:09] i can reproduce it more or less [12:09] "The Gossip" is the artist [12:09] tsdgeos, can you share the mp3 file? [12:09] i guess that depends on the legal situation of your country :D [12:10] i'll remove if after the test, i promise ;) [12:22] tsdgeos, sorry I didn’t see your answer earlier, by logs I mean the logs that are output (somewhere in a file, presumably) when there is a call to LOG("…") in the code [12:22] ah [12:22] :D [12:22] couldn’t find them anywhere in /var/log nor in /home/phablet/.cache/upstart/ [12:22] not sure where else to look for them [12:23] oSoMoN: LOG is going to qDebug [12:23] DLOG is compile time [12:23] so if you're using the packages may go to nowhere [12:23] LOG should be in /home/phablet/.cache/upstart/ i undertand that in the app log itself [12:25] ah indeed [12:26] I’d swear I had looked for them there too [12:26] tsdgeos, thanks [12:45] tsdgeos, ok... there was a criss cross merge which would duplicate functions (like the assertFocusOnIndex one) without showing me any conflicts at all [12:45] should be good now I hope [13:27] we're getting SUCCESS in CI in vivid qmluitests again [13:27] \o/ [13:28] i'm having a look at the xenial ones [13:37] tsdgeos, nice [13:37] tsdgeos, just started a run in Jenkaas, wonder how that goes ;) [13:37] he he [13:39] tsdgeos, do'h! [13:39] sorry :D [13:41] k [13:56] mzanetti, would you have some spare time to review this one? https://code.launchpad.net/~dandrader/unity8/dontRotateDesktop-lp1515977/+merge/282590 [13:57] dandrader, ack [14:11] dandrader, one question there, nothing needed to be done in the DesktopStage? [14:12] dandrader, meh sorry, overlooked it [14:15] * Saviq positively surprised @jenkaas's package install speed [14:16] it's done within 2m30s for full unity8 installation in a clean chroot [14:16] definitely not a priority to reduce, then [14:17] will first have a look at ccache, then === dandrader is now known as dandrader|afk [14:24] ltinkl, this only happens when interacting with the mouse, correct? https://code.launchpad.net/~mzanetti/unity8/launcher-sizing/+merge/280149/comments/717178 [14:27] mzanetti, let me try with touch [14:27] ltinkl, well, I meant pointer device [14:27] ltinkl, question was really: I can't repro it if sticking to kbd interaction [14:28] but it seems to happen when mixing kbd navigation and pointer interaction [14:31] mzanetti, yeah, with touch too; the point is, if the quicklist is open and the kbd navigation is on, launching something or clicking outside the quicklist renders the focus ring unusuable [14:31] ack [14:31] mzanetti, if you don't touch/tap and just use the keyboard, all seems to be fine [14:32] just wanted to understand your issue properly [14:45] Saviq: the one for the launcher tests that sometimes fail to open the launcher https://code.launchpad.net/~aacid/unity8/launcher_test_remove_dda_constraints/+merge/282600 [14:47] tsdgeos, coolz, that should improve autopkgtest results in the cloud then (as we had a slew of launcher fails on xenial) [14:49] Saviq, yup, just tried that on xenial locally, was passing [14:49] nice nice [14:49] I'm seeing something weird locally in all branches: [14:49] file:///home/ltinkl/bzr/unity8/alt-input/qml/Components/NotificationAudio.qml:38: Error: Qt.createQmlObject(): failed to create object: [14:49] file:///home/ltinkl/bzr/unity8/alt-input/qml/Components/inline:1:34: Cannot assign to non-existent property "source" [14:50] tsdgeos, deos it ring a bell? [14:50] that's how it's supposed to be [14:50] i think [14:50] yeah [14:50] it's our way of load this or that [14:50] should it fail really? [14:50] but this doesn't seem to load anything [14:51] before I used to be getting: Upstream audioRole enum not available, falling back to old role name. [14:51] which is imho correct [14:52] ah wait, this is the tests [14:52] yeah the mock is incomplete [14:52] and doesn't have a source property [14:52] tsdgeos, nice one https://unity8-jenkins.ubuntu.com/view/Maintain/job/run-commands/376/consoleFull - only LazyImage failed, that should be fixed by cimi's lp:~cimi/unity8/fix-lazyImage-test-flakyness, if he ever puts it into Needs review ;) [14:52] not sure why xml output isn't there [14:53] and 40mins... definitely need to improve that [14:53] i guess we could re-add it for the sake of less noisy test runs [14:53] Saviq: compile+test run? [14:53] tsdgeos, "just" the test run, which only compiles mocks and such [14:54] Saviq: it's what it takes, no? whenevre i do a full run it's also slow [14:54] on that note, we're building mocks on a normal build, aren't we [14:54] tsdgeos, well, we can parallelize them [14:54] oh sure, if there's more cpus we could [14:55] Saviq: i think we probably are yes [14:55] we probably shouldn't [14:56] Saviq, it is needs review, but I need to address mzanetti comment [14:56] tsdgeos, ah right ofc, it's in tests :) [14:56] cimi, oh, thought it's down in WiP === dandrader|afk is now known as dandrader [17:06] Mirv, are you aware of a cross-build dependency issue somewhere in qt 5.5 packages? I've had lpunity-api cross-build fine on vivid but fail on xenial, can get you a more info if you need [17:14] mzanetti, ping [17:15] mzanetti, you sure https://code.launchpad.net/~cimi/unity8/fix-lazyImage-test-flakyness/+merge/277459/comments/705903 works? [17:20] cimi, it should [17:31] mzanetti, how would you use it, I added the id to the animations inside the transition, but still fails to get it [17:32] mzanetti, I tried looking with findChild and findInvisibleChild both on the root and from the transition that contains the animation === dandrader is now known as dandrader|afk [17:32] still always same error of cannot find obj [17:42] cimi, hmm... odd [17:44] cimi, but you can use tryCompare on any of the properties that are animated in the transition === dandrader|afk is now known as dandrader [17:58] mzanetti, or I can try to add a property instead a signal, but to be honest I might have already tried in november, there must be a reason why I ended up adding a signal [17:58] * cimi forgot [17:59] I quite dislike adding such signals to a component's api if it's really only for testing... In most cases tests can be done nevertheless just as good [17:59] but if you really find no way out with this one... [17:59] mzanetti, do you have the flakyness on your laptop? [17:59] for this test? [18:00] yes [18:00] need to try [18:00] mzanetti, just run make testLazyImage [18:00] if your pc is fast it fails [18:00] in the offie atm and it does not fail here :) [18:00] my pc used to be fast... now its quite regular [18:00] :) [18:26] cimi, no... passes here