| === _salem is now known as salem_ | ||
| === salem_ is now known as _salem | ||
| Mirv | Saviq: now that we finally start to have cu2d & co. up and running, is it known that unity8 tests fail on desktop too? should I file a bug? | 06:41 |
|---|---|---|
| Mirv | Saviq: or 10 of then, to be exact | 06:41 |
| Mirv | bug #1244549 | 07:10 |
| ubot5 | bug 1244549 in Unity 8 "unity8 test failures in trusty" [Critical,New] https://launchpad.net/bugs/1244549 | 07:10 |
| om26er | mzanetti, hey! got to write the test today. https://code.launchpad.net/~om26er/unity8/add_searchIndicator_test/+merge/192647 | 07:56 |
| mzanetti | om26er: cool stuff! (there is a typo in the comment sping -> spin) | 07:58 |
| om26er | ooh - fixed that. | 07:59 |
| === iahmad is now known as iahmad|afk | ||
| Cimi | om26er, hey dude | 08:15 |
| om26er | Cimi, hi | 08:16 |
| Cimi | om26er, just had a look at the test :) | 08:16 |
| om26er | needs fixing ? ;) | 08:16 |
| Cimi | it's good, you should add more situations | 08:16 |
| Cimi | om26er, like you know when the search box comes in or not | 08:16 |
| Cimi | om26er, searchVisible: !greeter.shown && !lockscreen.shown && dash.shown | 08:17 |
| om26er | Cimi, something like, close the app and then make sure the indicator is visible ? | 08:17 |
| Cimi | om26er, so we should test those conditions | 08:17 |
| om26er | I was actually planning to change that to: | 08:17 |
| Cimi | om26er, the idea of the tests is "we test the condition, so if something breaks the test can detect" | 08:17 |
| om26er | searchVisible: !greeter.shown && !lockscreen.shown && !applicationRunning | 08:17 |
| om26er | rather !applicationFocused | 08:19 |
| om26er | Cimi, i'll enhance this test to also test if the indicator is shown while greeter is locked | 08:20 |
| Cimi | you learn quickly ;) | 08:20 |
| om26er | there is one more thing. I think the search indicator should also not be visible if a preview is opened, since tapping on the search indicator bring up the OSK but text does not appear anywhere | 08:21 |
| Cimi | om26er, that too | 08:21 |
| Cimi | om26er, have a chat with a designer | 08:21 |
| Cimi | or mark the bug as incomplete | 08:21 |
| Cimi | on Uubuntu UX | 08:21 |
| om26er | Cimi, ok, will do. | 08:22 |
| Cimi | Saviq, mzanetti how do I get the current lens from a test? | 08:27 |
| Cimi | I can add current index, but names seems to be mapped | 08:27 |
| Cimi | otherwise I go again for autopilot | 08:28 |
| mzanetti | duude | 08:28 |
| mzanetti | Cimi: don't pick a test framework for that reason (unless everything else is the same) | 08:28 |
| Cimi | mzanetti, isn't like selecting best tool for my needs? | 08:29 |
| mzanetti | Cimi: we have no more than 500 tests. if people start writing autopilot tests just because some certain task has a ready made example there, we'll end up with a test-duration of 2 days soon | 08:29 |
| Cimi | there is no ready example, just I like autopilot for few things :) | 08:30 |
| mzanetti | Cimi: I need to read the code to answer your question. give me a few minutes | 08:30 |
| Cimi | but can be done in qmltest | 08:30 |
| Cimi | mzanetti, Dash.qml iirc | 08:30 |
| mzanetti | Cimi: tests/qmltests/tst_Shell.qml line 368 | 08:32 |
| Cimi | mzanetti, what? | 08:43 |
| mzanetti | Cimi: can't you use something like this? | 08:43 |
| Cimi | mzanetti, oh yeah, my branch had different line numbers :) | 08:44 |
| Cimi | mzanetti, 368 was pointing to a useless piece of code :) | 08:44 |
| Cimi | #fridaymorning | 08:44 |
| mzanetti | lol | 08:44 |
| mzanetti | let me check again | 08:44 |
| Cimi | mzanetti, might try | 08:46 |
| Cimi | something similar though | 08:46 |
| mzanetti | Cimi: do you have a pointer to the item you need? | 08:46 |
| mzanetti | e.g. with findChild() | 08:46 |
| mzanetti | then you could use something like the itemIsOnScreen() (line 403) | 08:47 |
| Cimi | ok | 08:47 |
| mzanetti | Cimi: but you have a good point. would be cool to get a list of all items at position x,y | 08:51 |
| mzanetti | Cimi: maybe I'll hack that into our utils at some point | 08:51 |
| mzanetti | (unless you want to give it a shot ;) | 08:51 |
| Cimi | mzanetti, mmm | 08:56 |
| Cimi | mzanetti, in reality I want to test if the application scope is already on screen while I swipe | 08:56 |
| Cimi | mzanetti, does itemisonscreen verify if things are on screen with different z index? | 08:57 |
| mzanetti | Cimi: it doesn't care about the z index | 08:59 |
| Cimi | mzanetti, so it detects if underneath the app we have the dash? | 09:00 |
| mzanetti | Cimi: if the item is fully inside the visible part, it returns true, otherwise false | 09:00 |
| Cimi | brilliant | 09:00 |
| mzanetti | Cimi: yes. can even be covered | 09:00 |
| om26er | mzanetti, can I run a single test instead of running the whole TestCase {} ? | 09:27 |
| om26er | right now I am doing make testShell which runs like 16 tests | 09:27 |
| mzanetti | om26er: nope. you can't (except commenting out all the others temporarily) | 09:27 |
| om26er | mzanetti, ok | 09:28 |
| mzanetti | om26er: hint: you could try to use cleanup() to close any open applications (which will be executed after each test function) | 09:28 |
| om26er | mzanetti, yes I was going to do that. I am writing another test for search indicator taking greeter into account. | 09:28 |
| mzanetti | ah cool :) | 09:28 |
| om26er | how do I get back the greeter in test? | 09:28 |
| om26er | will greeter.show = true do ? | 09:29 |
| om26er | *shown | 09:29 |
| mzanetti | om26er: greeter.show() should do I think | 09:29 |
| Cimi | mzanetti, https://code.launchpad.net/~cimi/unity8/fix-1231996/+merge/192372 | 09:35 |
| mzanetti | Cimi: cheers. will just tag the last 5 bugs and review it then | 09:36 |
| nic-doffay | Saviq, I'm using UbuntuShapeForItem, what's the best way to set it's colour? | 09:39 |
| mzanetti | nic-doffay: Saviq is away today | 09:42 |
| mzanetti | nic-doffay: You can use a Rectangle {} in it | 09:42 |
| mzanetti | Cimi: shouldn't you make sure that there is another scope visible before the swipe? | 09:47 |
| Cimi | mzanetti, it does before | 09:47 |
| mzanetti | ok | 09:47 |
| Cimi | mzanetti, when it checks the home lens | 09:47 |
| Cimi | mzanetti, well, that apps is out of the way | 09:47 |
| nic-doffay | mzanetti, you sure? I've tried that already. | 09:47 |
| * mzanetti wishes there would be a feature to expand the diff to the full source file in LP | 09:48 | |
| mzanetti | nic-doffay: yes, I'm sure | 09:48 |
| Cimi | mzanetti, you remember me how to trigger lock screen? | 09:48 |
| mzanetti | nic-doffay: you can only have one item in there. but if you wrap the existing one inside the rectangle you should be fine | 09:49 |
| Cimi | mzanetti, want to work on the wallpaper shift | 09:49 |
| nic-doffay | mzanetti, that's what I've done. | 09:49 |
| nic-doffay | and set the anchors to fill the ubuntushape parent. | 09:49 |
| nic-doffay | The ubuntu shape doesn't mask off the rectangle though. | 09:49 |
| mzanetti | Cimi: http://design.canonical.com/2013/07/unlocking-the-phone/ | 09:49 |
| mzanetti | nic-doffay: well, I'm quite sure it works if done right. would need to see your code to spot the error | 09:50 |
| mzanetti | brb | 09:51 |
| nic-doffay | mzanetti, http://pastebin.ubuntu.com/6299790/ | 09:52 |
| mzanetti | nic-doffay: image: Rectangle {} | 10:06 |
| mzanetti | nic-doffay: I agree that the API is bad and the property shouldn't be named image but rather "contentItem" or the like. | 10:07 |
| mzanetti | Cimi: approved | 10:20 |
| Cimi | mzanetti, you see anything wrong here? http://paste.ubuntu.com/6299936/ | 10:32 |
| Cimi | mzanetti, lockscreen_background is NoneType | 10:33 |
| mzanetti | Cimi: the greeter might not be visible? | 10:34 |
| mzanetti | Cimi: in which case the Loader unloads it and select_single wont find it (as it's deleted) | 10:34 |
| Cimi | mzanetti, it's the same code of the greeter test I wrote basically | 10:34 |
| Cimi | I cannot see why this should fail | 10:34 |
| Cimi | mzanetti, the lock screen is not on a loader | 10:35 |
| === rachelliu_ is now known as rachelliu | ||
| mzanetti | Cimi: don't know then. also to me this looks like a qml test candidate tbh | 10:40 |
| mzanetti | MacSlow: heya | 10:55 |
| MacSlow | mzanetti, yup | 10:55 |
| mzanetti | MacSlow: do you think this is yours? https://bugs.launchpad.net/indicator-network/+bug/1238990 | 10:55 |
| ubot5 | Ubuntu bug 1238990 in Network Menu ""Unlock SIM" gets stuck if snap decision failed" [Low,Triaged] | 10:55 |
| MacSlow | mzanetti, hm... not sure... but I'll look into it... fullscreen-support is almost done (thank god) so I have time for this | 10:56 |
| mzanetti | MacSlow: yay! very nice :) | 10:57 |
| mzanetti | om26er_: heh, I'm afraid you understood the cleanup() thingie wrong :) | 11:13 |
| om26er_ | how about that ;) | 11:14 |
| mzanetti | om26er_: cleanup() is automatically called after each test function. However, you need to implement it yourself to actually do the cleanup | 11:14 |
| om26er_ | mzanetti, a function to make sure all the apps are killed ? | 11:15 |
| om26er_ | mzanetti, one of the tests is calling cleanup(); in it as well. is that doing wrong ? | 11:17 |
| om26er_ | re: test_rightEdgeDrag() | 11:17 |
| mzanetti | om26er_: probably. let me check | 11:17 |
| Cimi | mzanetti, self.app.select_single doesn't work for images | 11:18 |
| mzanetti | om26er_: that seems to be a special case | 11:18 |
| Cimi | mzanetti, works if I change the lock screen Image to CrossFadeImage | 11:18 |
| mzanetti | om26er_: as he does some more checks after the cleanup. but yes, that one calls cleanup() twice in the end | 11:18 |
| mzanetti | om26er_: but: I see the cleanup() already does the killing of apps | 11:19 |
| om26er_ | mzanetti, so no need to do that in my code ? | 11:19 |
| mzanetti | om26er_: so I've been wrong. your test is fine even without the manual cleanup() | 11:19 |
| om26er_ | mzanetti, ack, does the second test look good ? | 11:19 |
| mzanetti | om26er_: one sec, haven't read through it yet | 11:19 |
| mzanetti | Cimi: hmm... afaik it should work for images too | 11:20 |
| Cimi | mzanetti, doesn't seem to | 11:20 |
| Cimi | mzanetti, shall I change to CrossFadeImage on the lock screen? :) | 11:21 |
| Cimi | is there a reason why we have Image here and not CFI ? | 11:22 |
| om26er_ | Cimi, something else might have the same object name ? | 11:22 |
| mzanetti | Cimi: I'd say no. as it's not possible to change the image while the device its locked. so it would be a waste of resources | 11:22 |
| Cimi | om26er_, lockscreenBackground ,.. | 11:22 |
| mzanetti | om26er_: is this change wanted? 8+ readonly property alias previewShown: previewLoader.onScreen | 11:23 |
| === _salem is now known as salem_ | ||
| Cimi | mzanetti, mmm I dunno how to select that Image then | 11:24 |
| om26er_ | mzanetti, no, not really I think I pushed a local change, I was working on a different bug, let me revert that | 11:24 |
| mzanetti | Cimi: imho you still should use qmltest for this. | 11:24 |
| Cimi | mzanetti, globalRect is super handy | 11:25 |
| mzanetti | but it wastes 40 seconds in every CI run for no benefit | 11:25 |
| om26er_ | mzanetti, pushed. | 11:25 |
| === alan_g is now known as alan_g|lunch | ||
| mzanetti | besides the fact that you're working on it for 3 hours now doesn't make it look super handy tbh | 11:25 |
| mzanetti | om26er_: thanks | 11:25 |
| Cimi | mzanetti, 3 hours? | 11:26 |
| mzanetti | more? | 11:26 |
| Cimi | mzanetti, 1? | 11:26 |
| Cimi | mzanetti, btw it's not my fault if an API doesn't work | 11:27 |
| Cimi | mzanetti, if it were working I would have finished in 20 mins | 11:27 |
| Cimi | mzanetti, but it didn't so I wasted time figuring out why | 11:27 |
| * Cimi is doing the qmltest | 11:29 | |
| === MacSlow is now known as MacSlow|lunch | ||
| nic-doffay | mzanetti, is there anything else that needs to be set in the rectangle? | 11:42 |
| mzanetti | nic-doffay: don't think so... color of course :) | 11:43 |
| mzanetti | nic-doffay: still not working? | 11:44 |
| nic-doffay | mzanetti, nope | 11:44 |
| mzanetti | nic-doffay: can you push the branch? I'll check it out then | 11:44 |
| nic-doffay | mzanetti, yeah sure I'll push it to a junk branch. | 11:47 |
| nic-doffay | mzanetti, here it is: lp:~nicolas-doffay/+junk/ubuntu-shape-colour-test | 11:54 |
| nic-doffay | It's in PageHeader.qml | 11:54 |
| mzanetti | nic-doffay: you need to place the item inside the rectangle | 11:58 |
| mzanetti | which doesn't solve your problem still... hmmm. weird | 12:00 |
| MacSlow|lunch | mzanetti, btw... http://ubuntuone.com/4jWA8iX9nJrNn2o7MkQHgh | 12:01 |
| mzanetti | nic-doffay: ah. got it | 12:01 |
| mzanetti | nic-doffay: the component you set to image: is not a direct child of the shape. hence the anchors.fill: parent doesn't work | 12:02 |
| mzanetti | nic-doffay: give the UbuntushapeForItem an id: secondaryItemShape | 12:02 |
| mzanetti | nic-doffay: and make the rectangle width: secondaryItemShape.width and height: secondaryItemShape.height | 12:02 |
| mzanetti | nic-doffay: another hint: anchors { horizontalCenter: parent.horizontalCenter; verticalCenter: parent.verticalCenter } is the same as "anchors.centerIn: parent" | 12:04 |
| mzanetti | MacSlow|lunch: it's getting close :) | 12:05 |
| MacSlow|lunch | mzanetti, yup... top and bottom margins still give me headaches... but everything else works now smoothly | 12:05 |
| mzanetti | MacSlow|lunch: background image also missing still | 12:07 |
| nic-doffay | mzanetti, ah cool. | 12:08 |
| mzanetti | MacSlow|lunch: if it's possible to use shell.background easily it'd be great. otherwise I think it would be about time to have another go on the blurring stuff | 12:08 |
| MacSlow|lunch | mzanetti, well... I added a skipBackground flag to Lockscreen because otherwise I would not be able to get a consistent background up to the edges... | 12:09 |
| MacSlow|lunch | mzanetti, that's up for discussion though... | 12:09 |
| mzanetti | MacSlow|lunch: ok. I'll give it a shot with the blurring later on | 12:09 |
| MacSlow|lunch | mzanetti, fixing the margins is higher prio ... and with the skipBackground flag it looks nicer (with very little effort) | 12:10 |
| mzanetti | yep. works for me | 12:10 |
| MacSlow|lunch | mzanetti, yeah... the blurring will take more time so this flag is a good effort/looks compromise | 12:11 |
| === MacSlow|lunch is now known as MacSlow | ||
| === alan_g|lunch is now known as alan_g | ||
| === alex_abreu is now known as alex-abreu | ||
| fginther | Saviq, morning | 13:03 |
| fginther | Saviq, can you change this blueprint to the 'core' track? https://blueprints.launchpad.net/ubuntu/+spec/client-1311-upstream-merger-20 | 13:04 |
| kgunn | fginther: Saviq is out today i thikn | 13:26 |
| fginther | kgunn, thanks. There's no rush on my request. I'll try to remember next week | 13:27 |
| mzanetti | dednick_: nic-doffay: standup | 13:32 |
| mzanetti | I'll be away for an hour or so. bbl | 13:42 |
| dandrader | mzanetti, any idea who should I talk to about those automatic releases of packages made by "PS Jenkins bot"? | 13:43 |
| dandrader | fginther, ^ | 13:43 |
| fginther | dandrader, packages are automatically released by the daily release processs, PS Jenkins bot is the user | 13:45 |
| dandrader | fginther, yes. the thing is that it's been a long while since the last release of lp:unity-mir. Why a new release of it hasn't been made? | 13:46 |
| qengho | Hi all. I have a bug report assigned to chromium-browser, but I think it might should be assigned to unity. What do you think? https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1244529 | 13:55 |
| ubot5 | Ubuntu bug 1244529 in chromium-browser (Ubuntu) "no repaint of unity at end of fullscreen chromium" [Undecided,New] | 13:55 |
| fginther | dandrader, ah, finally found it. unity-mir is failing the landing pipeline testing so it hasn't been released | 14:00 |
| dandrader | fginther, anything I can do about it? | 14:02 |
| fginther | dandrader, Mirv is the contact for that item (landing #268) Mirv ^ ? | 14:04 |
| === dandrader is now known as dandrader|afk | ||
| === dandrader|afk is now known as dandrader | ||
| Cimi | my favourite qa engineer | 14:35 |
| Cimi | mzanetti, hey | 14:35 |
| Cimi | mzanetti, I removed MathLocal | 14:35 |
| Cimi | mzanetti, but ubuntu ui toolkit is missing tests we had in shell | 14:36 |
| Cimi | mzanetti, https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1244685 | 14:36 |
| ubot5 | Ubuntu bug 1244685 in Ubuntu UI Toolkit "MathUtils does not have tests" [Undecided,New] | 14:36 |
| Cimi | pls have a look at it, it gets fixed :) | 14:36 |
| Cimi | unity 8 branch is here https://code.launchpad.net/~cimi/unity8/remove-mathlocal/+merge/192709 | 14:36 |
| mzanetti | Cimi: can you move the tests over to the SDK? | 14:44 |
| mzanetti | Cimi: would be a waste to just drop them | 14:44 |
| mzanetti | dandrader: didrocks should be able to help there too | 14:45 |
| Cimi | mzanetti, look at the bugreport | 14:52 |
| Cimi | mzanetti, I attached our test file | 14:52 |
| * didrocks backlogs | 14:52 | |
| didrocks | dandrader: ah, so yeah, we are blocked on Mir | 14:52 |
| didrocks | and this is under resolution | 14:52 |
| didrocks | (they broke the ABI, we are going back on track) | 14:52 |
| dandrader | didrocks, ok. thanks for the info | 14:52 |
| didrocks | dandrader: probably on Mondya | 14:53 |
| didrocks | Monday* | 14:53 |
| sil2100 | Saviq: hiii | 15:14 |
| sil2100 | Saviq: did you guys have a minute to take a look at https://bugs.launchpad.net/unity8/+bug/1244549 ? | 15:15 |
| ubot5 | Ubuntu bug 1244549 in Unity 8 "unity8 test failures in trusty" [Critical,New] | 15:15 |
| === salem_ is now known as _salem | ||
| === jalcine_ is now known as jalcine | ||
| Cimi | MacSlow, ^ | 15:44 |
| MacSlow | Cimi, the bug? | 15:48 |
| MacSlow | Cimi, not sure what you're refering to | 15:48 |
| Cimi | MacSlow, that bug looks like notification autopilot issues | 15:49 |
| Cimi | MacSlow, thought you knew more than Saviq | 15:49 |
| Cimi | sil2100, Saviq is off today | 15:49 |
| MacSlow | Cimi, well these tend to be "fragile" | 15:49 |
| MacSlow | Cimi, so if there's something currently moving with autopilot in trusty I expect these to be false positives for the moment until proven wrong | 15:50 |
| Cimi | MacSlow, don't tell that to me, but the bug report :) | 15:51 |
| MacSlow | sure | 15:51 |
| MacSlow | Cimi, but I wanted to let you know too | 15:51 |
| sil2100 | bregma: ping! Hi! | 15:58 |
| bregma | yo | 15:59 |
| sil2100 | bregma: I don't know who to ping related to those, but we have HUD failures in trusty: https://bugs.launchpad.net/hud/+bug/1244704 | 16:01 |
| ubot5 | Ubuntu bug 1244704 in Unity HUD "Some test_hud tests fail on trusty desktop" [Critical,Incomplete] | 16:01 |
| sil2100 | bregma: tedg got informed, but as these are from lp:unity autopilot, do you know who we should assign to that? | 16:02 |
| MacSlow | Cimi, sil2100: commented on 1244549 | 16:05 |
| === _salem is now known as salem_ | ||
| === dandrader is now known as dandrader|lunch | ||
| sil2100 | MacSlow: thanks, let me try a re-run and see if it's still the same failure pack | 16:13 |
| sil2100 | MacSlow: anyway, maybe in some time-critical cases we could bump the 10-second timeout? (as AP allows that) | 16:15 |
| bregma | sil2100, those failures all look to me like unexpected results are being returned by the HUD (5 instead of 0, for example) .... they don;t appear to be UI issues in unity | 16:18 |
| MacSlow | sil2100, oh... didn't know that | 16:18 |
| MacSlow | sil2100, good to know | 16:18 |
| sil2100 | tedg: ^ | 16:19 |
| === jalcine is now known as jalcine_ | ||
| === jalcine_ is now known as jalcine | ||
| kgunn | greyback_: can you share your qtscenegraph doc with me ? | 17:01 |
| kgunn | like on google | 17:01 |
| greyback_ | kgunn: for the whole internets to see? Okie dokey | 17:01 |
| greyback_ | kgunn: done! | 17:02 |
| kgunn | greyback_: actually...i meant just canonical :) | 17:02 |
| kgunn | but then shared with me as an individual | 17:02 |
| greyback_ | kgunn: well you got shared | 17:03 |
| kgunn | greyback_: you did it right :) | 17:03 |
| * greyback_ eow | 17:05 | |
| === alan_g is now known as alan_g|EOW | ||
| === jalcine is now known as jalcine_ | ||
| === dandrader|lunch is now known as dandrader | ||
| === jalcine_ is now known as jalcine | ||
| === racarr is now known as racarr|sick | ||
| === salem_ is now known as _salem | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!