[06:32] <vigo> morning!!
[07:32] <tsdgeos> Trevinho: see you have failing tests at https://code.launchpad.net/~3v1n0/unity8/fake-indicators-model/+merge/297509
[07:50] <Trevinho> tsdgeos: Yes I know, I spoke with ltinkl since it happened since merging with the indicator keyboard... But I can't see what's wrong there
[07:51] <tsdgeos> k
[08:01] <vigo> Hi!
[08:03] <vigo> I'm running unity8 on a laptop and I just updated through cli and now the webbrowser is not connecting anymore to any site
[08:03] <vigo> it shows a message
[08:04] <vigo> "The rendering process has been closed for this tab"
[08:04] <duflu> vigo: Yeah we know - https://bugs.launchpad.net/oxide/+bug/1613258
[08:04] <ubot5> Launchpad bug 1613258 in Oxide "Web browser can't browse anything - "The rendering process has been closed for this tab" [signal 4 ILL_ILLOPN]" [Critical,In progress]
[08:05] <vigo> duflu, thanks!
=== pete-woods[fail] is now known as pete-woods
[10:30] <sil2100> Saviq: hey! I tried building a no-change rebuild of qtmir on vivid in a private PPA (with virt builders for armhf and amr64) and it FTBFS on unit tests
[10:30] <sil2100> "ERROR: 1 leaked mock object found at program exit."
[10:30] <sil2100> Saviq: do you guys know if that's a known issue on virtualized builders?
[10:31] <Saviq> sil2100, I think I heard of that some time, could very well be a virt issue
[10:48] <Saviq> greyback, does "ERROR: 1 leaked mock object found at program exit." ring a bell? sil2100 says he gets that in arm* virtualized builders for qtmir
[10:48] <greyback> Saviq: no, never seen that before. I assume coming from googlemock framework
[10:49] <greyback> does it crash at that point?
[10:49] <sil2100> greyback: https://launchpadlibrarian.net/280494874/buildlog_ubuntu-vivid-armhf.qtmir_0.4.8+16.10.20160816.1-0ubuntu1kpi1~test1~ubuntu15.04.1_BUILDING.txt.gz <- here's the example output
[10:50] <Saviq> so a test failure in cleanup, basically
[10:50] <Saviq> actually, maybe that's just a warning
[10:50] <Saviq> the failure seems to be /«BUILDDIR»/qtmir-0.4.8+16.10.20160816.1/tests/modules/SurfaceManager/mirsurface_test.cpp:79: Failure
[10:50] <Saviq> Value of: spyFrameDropped.count() > 0
[10:50] <Saviq>   Actual: false
[10:50] <Saviq> Expected: true
[10:50] <Saviq> unless related
[10:50] <greyback> yeah the test is failing
[10:50] <greyback> let me have a look
=== marcusto_ is now known as marcustomlinson
=== JanC is now known as Guest58470
=== JanC_ is now known as JanC
[14:06] <fdfgdfgdgdf> hey, where to download unity from?
[14:06] <fdfgdfgdgdf> I want to install Ubuntu today, after using 100% Windows 7........... :D
[14:07] <mterry> fdfgdfgdgdf: unity comes with Ubuntu, it's the default user interface
[14:07] <mterry> fdfgdfgdgdf: and welcome  :)
[14:15] <fdfgdfgdgdf> thanks!
[14:15] <fdfgdfgdgdf> :D
[14:27] <mterry> josharenson: heyo!
[14:27] <josharenson> mterry: good morning
[14:27] <mterry> josharenson: session-chooser-gui two things: can you merge from trunk please, and can you add back a waiting alias in WideView.qml into loginList.waiting
[14:27] <josharenson> mterry: sure thing
[14:27] <mterry> josharenson: I think that got accidentally deleted in some refactoring
[14:27] <mterry> josharenson: you can test with a tablet tryShell and moving off user auth-error and back again
[14:28] <mterry> josharenson: without waiting, auth-error ends up having a password prompt
[14:28] <josharenson> ah ok
[15:46] <mterry> @unity: qml-heads, is there a way to connect a widget to the top of the OSK better than a bottomMargin of inputMethod.keyboardRectangle.height?  (that tends to leave a gap as the OSK is animating)
[15:47] <mzanetti> hmm... afaik keyboardRect.height should animate too
[15:47] <mterry> it does animate, just not in lockstep
[15:47] <mzanetti> (I know it doesn't the very first time)
[15:47] <mterry> So it's a little jittery and sometimes there's a gap
[15:48] <mzanetti> are you in an app or the shell?
[15:48] <mterry> shell
[15:48] <mzanetti> mterry, try with SurfaceManager.inputMethodSurface.height
[15:49] <mzanetti> and unless that changed while I was away this doesn't animate the very first time it is invoked, which is a bug
[15:49] <mzanetti> if you're in the greeter I guess you'll be bitten by that... might needs fixing
[15:50] <mterry> hmm yeah I am in greeter.  will test
[16:01] <mterry> mzanetti: gap still there  :(
[16:01] <mzanetti> hmm
[16:01] <mzanetti> implicitHeight maybe?
[16:01] <mzanetti> mterry, ^
[16:04] <mterry> mzanetti: doesn't seem to be defined
[16:04] <mzanetti> hmm odd
[16:04] <mterry> I had to use SurfaceManager.inputMethodSurface.inputBounds.height before, fyi
[16:04] <mzanetti> should be. I'm using that a lot
[16:04] <mterry> hmmm
[16:04] <mzanetti> ah
[16:04] <mzanetti> no
[16:04] <mzanetti> inputMethodSurface is a MirSurface
[16:04] <mzanetti> that should have implicitHeight
[16:05] <greyback> osk surface is fullscreen
[16:05] <mzanetti> but I just realize that it actually is fullscreen and transparent
[16:05] <mzanetti> right
[16:05] <mterry> SurfaceManager.inputMethodSurface.implicitHeight is undefined as well
[16:05] <greyback> input bounds probably the most accurate info
[16:05] <mzanetti> mterry, I use that to animate windows up and down
[16:05] <mzanetti> it seems reasonably accurate to me
[16:05] <greyback> as that's the osk telling us ther area which can get input, which is probably the bit it is drawing into
[16:06] <mterry> mzanetti: in u8?  I don't see it via grep
[16:06] <mterry> greyback: yeah there just seems to be a slight mismatch during animation.  A gap between the u8-side widget and the osk drawn on screen
[16:06] <greyback> but I suspect animating something in the shell to be "stuck" to the top of the osk as it animates will require a bunch of redesign
[16:07] <greyback> right. You're trying to synchronise 2 separate things.
[16:07] <mterry> Well sure.  :)  Usually works in qml, though I realize these are separate processes
[16:07] <greyback> I know of no easy solution currently, fixing this would need a chunk of redesign work
[16:08] <mterry> greyback: that's what I figured.  I have a hackish solution for the short term though
[16:09] <mterry> greyback, mzanetti: thx for tips
[16:11] <mzanetti> mterry, DesktopStage.qml:158
[16:11] <mzanetti> mterry, as I said, seems reasonably in sync to me when it moves the apps up/down
[16:11] <mzanetti> but if you attach a full size item directly to it it might be a bit worse
[16:12] <mterry> mzanetti: oh oh sure, inputBounds works.  I was talking about implicitHeight being undefined.  But using inputBounds still gives the gap
[19:17] <lborda> ping
[19:21] <mterry> dandrader: heyo!  I updated https://code.launchpad.net/~mterry/unity8/greeter-mock-bg/+merge/302695 to fix some tests, but wanted you to look over it again, since the fix wasn't trivial
[19:22] <mterry> dandrader: namely, the dropping of the LIGHTDM test keyword, and just always linking our mock liblightdm
[19:24] <dandrader> mterry, does -DNO_TESTS=true still work?
[19:25] <mterry> dandrader: oh yeah.  This just means that all tests use the mock liblightdm instead of some tests
[19:28] <dandrader> ah ok
[20:53] <mterry> mzanetti: btw Alex M looked at https://code.launchpad.net/~mterry/unity8/greeter-no-lockscreen/+merge/302711 and is still +1 on it.  So I guess we should try to land it
=== bschaefer_ is now known as bschaefer
=== bschaefer_ is now known as bschaefer