[14:37] <kenvandine> rhuddie, you worked on the switch to using the osk for AP tests right?
[14:37] <kenvandine> there's a test in system-settings that uses TextField.write which is now failing in wily
[14:37] <kenvandine> raise RuntimeError("Keyboard is not on screen")
[14:37] <kenvandine>  RuntimeError: Keyboard is not on screen
[14:41] <rhuddie> kenvandine, yes the osk was enabled for the textfield.write() method. sounds like the osk is not being displayed on screen for some reason. do you have some logs?
[14:42] <kenvandine> http://pastebin.ubuntu.com/11878008/
[14:44] <rhuddie> kenvandine, are you running those tests locally? If so, do you see the osk appear?
[14:44] <kenvandine> it isn't being displayed
[14:45] <rhuddie> weird
[14:45] <rhuddie> kenvandine, could you try and run the autopilot test: ubuntu_keyboard.tests.test_keyboard.UbuntuKeyboardTestsAccess.test_keyboard_is_available
[14:45] <kenvandine> sure
[14:46] <kenvandine> rhuddie, failed too
[14:49] <rhuddie> that's very strange... what happens in system settings if you press on the setting you're testing normally? does osk appear, or is it just when testing that its broken?
[14:51] <rhuddie> kenvandine, btw, if you wish to revert to previous behaviour and use simulated keyboard, that can be done by killing maliit-server before running the test
[14:53] <rhuddie> in this case it seems like there's a bug with the osk not appearing. however I know that was working up until very recently, so must be a recent regression
[14:55] <kenvandine> rhuddie, i can't test it manually on this device, it's a UI that is normally only shown with dual sim
[14:56] <rhuddie> kenvandine, oh, well I could check it for you. what setting was it?
[14:56] <kenvandine> the SIM name under settings
[14:56] <kenvandine> where you can rename the SIM
[14:56] <rhuddie> ok, just a sec
[15:00] <rhuddie> kenvandine, yes, I see the osk when pressing those settings. But when I run the ubuntu_keyboard test above, the osk does not appear and it fails
[15:01] <kenvandine> ok
[15:01] <kenvandine> thanks!
[15:01] <kenvandine> so a bug somewhere in the uitk autopilot stuff?
[15:02] <rhuddie> kenvandine, very strange. the fact it fails in ubuntu_keyboard suggests its something to do with that. - that test doesn't use the textfield write() method even
[15:03] <rhuddie> its just tapping the field and expecting the osk to appear
[15:03] <kenvandine> interesting
[15:03] <kenvandine> it should appear
[15:04] <rhuddie> yes, but it seems the problem is only when using autopilot.
[15:48] <kenvandine> rhuddie, i'll file a bug, what project?
[15:49] <teward> balloons: in case you missed it, infinity and others solved the ifnames discrepancy on the server ISO, so I can relax xD
[15:50] <rhuddie> kenvandine, i just raised this one: https://bugs.launchpad.net/ubuntu-keyboard/+bug/1474444
[15:50] <kenvandine> rhuddie, awesome, thanks!
[15:51] <rhuddie> kenvandine, please just add a comment if you think there is some info missing
[15:52] <balloons> teward, awesome work! thanks for finding that, and fixing it!
[15:54] <teward> balloons: infinity and cypermox fixed it i think.  i just found it, and it turns out pitti was supposed to work on it but didn't get to it.  I also guess that just because I found and reported the bug they said "We should probably fix this now" :P
[16:01] <balloons> :p.. yes we all blamed pitti, hehe
[21:36] <josharenson> cachio: some questions about https://bazaar.launchpad.net/~canonical-platform-qa/ubuntu-system-tests/trunk/view/head:/ubuntu_system_tests/helpers/unity8/shell.py if you have a chance...