=== _thumper_ is now known as thumper === salem_ is now known as _salem [08:06] sil2100: hey, how are you? [08:11] didrocks: hello! Fine, just doing late breakfast today ;) How about you? [08:11] sil2100: well, still under a pile of things to deal with [08:11] sil2100: and unfortunately, again, some autopilot tests are failing [08:11] sil2100: and that's also blocking the 100 scopes [08:12] sil2100: do you mind having a quick look, quick bisect if needed? [08:12] Uuuu [08:12] Ok, checking that then [08:12] thanks :) [08:12] sil2100: btw, I plan to kill the staging ppa as we either have the daily-build ppa [08:12] sil2100: and if you need something in between as packaging is inline, it's just about bzr branch -r && bzr bd [08:12] thoughts? [08:12] didrocks: fine with me currently ;) [08:13] It was useful when using quantal, but anyway lp:unity doesn't build properly on quantal anymore [08:13] Not sure if Trevinho or andyrock use staging still, but I suppose it's not a big deal to build a revision on your own [08:14] sil2100: they should use raring anyway :) [08:14] and yeah, not a big deal [08:14] thanks sil2100 [08:14] fginther: FYI ^ [08:14] So as far as I am concerned, we can let it RIP ;) [08:14] np [08:15] sil2100: \o/ [08:16] sil2100: keep me posted, I'm a little bit stressed about those regressions [08:16] sil2100: if you need the rev number for the previous successful runs, I can give them to you [08:22] mzanetti: ping-ie [08:25] tsdgeos: hey [08:25] mzanetti: did you enable the running of qmluitests i saw? [08:25] tsdgeos: huh? where? [08:25] maybe i saw wrong :D [08:25] I'm afraid [08:26] ah right [08:26] it's only the regular ones [08:27] seen this failure https://jenkins.qa.ubuntu.com/job/unity-phablet-quantal-i386-autolanding/42/console ? [08:27] mzanetti: ↑ [08:28] tsdgeos: yeah... I've seen this yesterday night once [08:28] tsdgeos: don't know yet whats wrong [08:32] mzanetti: my test for dragging up the hud failed again, think i'll go with the retry pattern you created, that ok? [08:33] tsdgeos: is the dragging up the actual testcase or a helper? [08:33] mzanetti: it is part of the testcase [08:34] tsdgeos: hmm.. yestderday you said it looked like its not dragging up far enough? [08:34] i added another gu and didn't help [08:35] tsdgeos: ok... try that... in case something would be really broken the testcase would still reveal that... [08:36] try that = retry? [08:36] tsdgeos: yes, we just need to make sure not to make them useless. but I guess its fine in this case too [08:37] actually the first thing that fails is almost the first thing we test [08:37] so i agree that if "something is really broken" [08:37] it'll be found === alan_g is now known as alan_g|afk [08:53] hey unity-ers [08:53] sil2100, hey, are you looking at the failing unity autopilot tests? (asking before having a look to not dup work) [08:59] seb128: yes [09:00] seb128: didrocks asked me in the morning already about those ;) [09:00] sil2100, well I was wondering if you were already looking at it or if you needed help [09:02] I wonder if those are still gcalctool issues [09:03] sil2100, didrocks: gar, I think I know what's the issue is [09:04] stupid one [09:04] https://code.launchpad.net/~mathieu-tl/autopilot/gcalctool/+merge/154196 is buggy [09:04] upstream reverted the .desktop name [09:04] not the binary name [09:04] so the process-name is wrong [09:04] should be set back to gnome-calculator [09:04] let me file a MR [09:05] * sil2100 sighs [09:05] ;) [09:06] seb128: looking at unity release tests and indicator tests, well, it's not enough failures to be caused by just that [09:06] it is [09:06] we were around 3 [09:06] we were around 30 last week [09:07] and all icons* are due to that [09:07] Since I see gcalctool being opened [09:07] sil2100, see the icon ones [09:07] http://10.97.0.1:8080/job/ps-unity-autopilot-release-testing/126/label=autopilot-ati/testReport/junit/unity.tests.launcher.test_icon_behavior/LauncherDragIconsBehavior/test_can_drag_icon_below_bfb_Single_Monitor_outside_/ [09:07] MismatchError: After 10.0 seconds test failed: [] != [] [09:07] sil2100, they are opened but not closed for me [09:08] seb128: good spot! [09:08] didrocks, sil2100: https://code.launchpad.net/~seb128/autopilot/correct-calculator-name/+merge/154304 [09:08] sil2100: btw, while we are at it, the indicators on nvidia had 3 failures [09:08] sil2100: want free karma ^ [09:08] sil2100: 1 on the other [09:08] sil2100: so still have the same issue [09:08] would be nice to get them working : [09:08] :) [09:09] didrocks: I'm actually looking mostly at the indicator tests now, since those are failing less obviously ;) [09:09] sil2100: sweet! [09:10] seb128: waiting for the merge diff to generate itself [09:10] sil2100: click on the commit :p [09:10] didrocks: pfff ;p [09:10] ;) [09:11] I hate the test_super_h test, really really [09:12] sil2100: but they are so super :p [09:12] I would hack-fix it by adding a silly sleep in front of it, grrr [09:12] sil2100, diff is there [09:13] seb128: I approved it already ;) === alan_g|afk is now known as alan_g [09:35] Saviq: how do you test stuff with 30px/gu? got a retina display? [09:35] tsdgeos, no [09:35] mzanetti, ^ ? [09:35] i can't see improvements [09:36] the thing just shows partially on screen [09:36] so it's hard to compare [09:36] tsdgeos, make a screenshot ;) [09:37] sure [09:37] i was trying to compare animation [09:37] but maybe that's not what i should be looking at :D [09:40] i think i see the difference now [09:42] The repository is incredibly big (~40 megs) comparing to what we actually have there. culprit is the history. I suppose we were adding and removing big (or a large number of small) graphical assets in the past? [09:42] Saviq: ? [09:43] I wonder if it's feasible to remove that fat (i.e. stripping the graphical assets from bzr history) [09:43] dandrader: it's because we "share" with "regular" unity [09:43] mzanetti: it is about https://code.launchpad.net/~unity-team/unity/phablet.more-flexible-previews/+merge/150860 [09:43] tsdgeos, hmm? [09:43] tsdgeos, no it's not that [09:43] tsdgeos, we just have all the assets that we had there before [09:44] tsdgeos, like hardcoded images, backgrounds etc. [09:44] Saviq: ok, i think dandrader was complaining about the push time, sorry [09:44] mzanetti, tsdgeos had problems with checking the "flexible preview" MR - not enough screen real estate [09:44] Saviq: mzanetti: https://code.launchpad.net/~unity-team/unity/phablet.more-flexible-previews/+merge/150860 changes the layout also in the tablet, is that ok? [09:44] tsdgeos, the big repository size also reflects in push and branch time [09:44] how do we test changes like that, btw? x/y/width/height ? [09:45] makes the preview wider [09:45] i agree it looks better [09:45] but don't want designers coming after me :D [09:45] mzanetti, you have the details ^ :) [09:46] tsdgeos: you can probably set GRID_UNIT_PX to 4 [09:46] that would scale everything down [09:47] mzanetti: sure, as said, it changes the layouting for the nexus10 here [09:47] is that expected by design? [09:48] Could someone with compiz test a branch for me? [09:48] tsdgeos: what do you mean with change? [09:48] * tsdgeos gets some screenshots, it'll take a while [09:51] anyone voluteers to claim that review? https://code.launchpad.net/~dandrader/unity/phablet_tst_ResponsiveGridView/+merge/154310 [09:51] Cimi: ping [09:51] mzanetti, pong [09:52] didrocks: 13.04 branches (cupstream2distro-config) should go to stacks/13.04/ , right? [09:52] Cimi: could you please help nic-doffay with his branch [09:52] mzanetti: yeah, but not that week [09:52] oupss mmrazik ^ [09:52] mmrazik: I really don't have time to do that properly [09:52] mm yes [09:52] hehe [09:52] and we have the 100scopes, payement in dash === dandrader is now known as dandrader|afk [09:52] mzanetti, what do I need to do? [09:52] didrocks: we just need to adjust autolanding. So I'll just create the dir [09:52] Cimi: thanks. nic-doffay: send Cimi the link, he'll check it out [09:52] mzanetti, nic-doffay ah no [09:53] mzanetti: ok, it's not that different now that i see the screenshots, it just happens to wrap a word differently and changes the text "a lot" that is what gave me the tip [09:53] http://imgur.com/eMTIS3s old [09:53] http://imgur.com/A8eYJy0 new [09:53] mzanetti, nic-doffay no raring here, and unity broken [09:53] mzanetti, nic-doffay, MacSlow has raring I think :P (sorry mirco) [09:53] Cimi: no need to be raring [09:54] Cimi: its basic qmltests stuff [09:54] mzanetti: so i'll approve i guess [09:54] mzanetti, but I am not running compiz [09:54] Cimi, what's up? [09:54] tsdgeos: the thing this should fix is more the steps in between [09:55] mmrazik: I told upstream to wait for next week :( [09:55] didrocks: cyphermox was pinging us yesterday [09:55] mmrazik: if indicators doesn't want, they should be able to wait TBH [09:55] didrocks: ok [09:55] for having autolanding set on trunk [09:55] tsdgeos: it should always look good, no matter how big the window is [09:55] mzanetti: yes, i saw that, i was scared at the change at an "official" resolution [09:55] mzanetti, so yes [09:55] MacSlow, don't worry [09:55] nic-doffay, ok branch? [09:55] tsdgeos, mzanetti, there's no tests for the preview is there ;) [09:56] Thanks Cimi: https://code.launchpad.net/~nicolas-doffay/unity/page-header-test [09:56] Saviq: no. not yet... this MP is hanging in the queue since before the feature freeze [09:56] mzanetti, yeah I know [09:57] Saviq: i know, but there weren't before, tbh it doesn't get us worse than we were and it's not really a new feature, so i myself wouldn't block on that, but if you prefer we can wait for it [09:57] tsdgeos, nah, it's ok, let's clear the queue [09:59] * mzanetti can take care about testing the previews [10:00] dandrader|afk: i'll have a look at that MR === dandrader|afk is now known as dandrader [10:08] tsdgeos, thanks [10:25] nic-doffay, so I tested, passes your tests, what's the one failing? [10:26] mzanetti: mmrazik: https://jenkins.qa.ubuntu.com/job/unity-phablet-quantal-armhf-autolanding/45/console ? [10:28] tsdgeos: hmm... no clue right now... persistent or temporary? [10:28] first time i see it [10:28] retry? [10:29] tsdgeos: retry. ps-panda-10 died [10:29] mzanetti: I start to fear these pandas are going to be hell [10:29] mzanetti: ps-panda-7 and -5 died yesterday [10:29] 10 today [10:29] :S [10:29] we will be just restarting pandas all they long :-/ [10:29] Cimi, how did you test it? [10:30] nic-doffay, I edited the cmakelist [10:30] nic-doffay, then I did make testPageHeader [10:31] Can you go to test/qmluitests and run qmlscene tst_PageHeader.qml [10:31] Cimi, ^ [10:31] I can [10:31] nic-doffay, and? [10:34] Can you see the window Cimi [10:34] nic-doffay, no [10:35] You don't see qmlscene pop up at all Cimi ? [10:36] nic-doffay, no [10:38] Cimi: it should be there... this is how it should look: http://wstaw.org/m/2013/03/19/plasma-desktopy16257.png [10:39] mzanetti, well it's a file with a TestCase, obviusly it doesn't show [10:39] Cimi: ok... now I'm confused [10:40] yesterday it was wrapped in an Item [10:40] that might explain why it didn't work for nic [10:41] I probably forgot to push my most recent changes. [10:41] Let me check. [10:42] Or rather I assumed I had. [10:43] Doh! Cimi committed but not pushed soz. [10:44] Can you pull and try again? [10:45] nic-doffay, you also added an helloworld component under Components [10:45] nic-doffay, watch out! :) [10:45] nic-doffay, now shows up [10:46] nic-doffay, you should move the PageHeader outside TestCase [10:46] nic-doffay, you don't need to embed items inside TestCase, it's not requires [10:47] nic-doffay, because all you care are the properties of PageHeader, and you can access *all* of them using the id [10:47] pageHeader.propertyName [10:47] MacSlow, your cpp would go into plugins/Notifications [10:47] MacSlow, look in plugins/* to see how it's handled there [10:48] Saviq, so it has to become a pluging?! [10:49] Saviq, ok... didn't know that [10:49] MacSlow, it doesn't _have_ to, but that's the best way [10:49] Cimi, does the text show? [10:49] MacSlow: yep... make everything a plugin (with small exceptions - but I don't expect the notifications to be one of those) [10:49] nic-doffay, it does once you move pageHeader outside the TestCase {} [10:50] I'll do that then, thanks Cimi . [10:50] MacSlow, you then go "import Ubuntu.Notifications 0.1" in the QML file when you need it [10:52] MacSlow, I'll also have some comments, will reply to your email [10:53] mmrazik: a lot of branches have been approved between few minutes and 2 hours ago and they are not merging: https://code.launchpad.net/ubuntu-scopes/+activereviews [10:55] dandrader: made a new comment on the review, i think there's a row/column mixup in some variable names [10:55] tsdgeos, ok, will check [10:55] Saviq, thx [10:56] didrocks: looking [10:57] tsdgeos, right, it's mixed indeed [11:00] Saviq: I'm just catching up with blueprint changes [11:00] Saviq: transition to googletest? [11:00] can you elaborate? [11:00] didrocks: all were approved 10mins ago max [11:00] didrocks: I've triggered them manually [11:00] didrocks: but there is nothing wrong AFAICS [11:00] mzanetti, yeah, for c++ tests (which we have almost none, so that transition will be easy) [11:00] mmrazik: 10 minutes? [11:01] let me recheck my email [11:01] mzanetti, unless I'm not thinking about something? [11:01] mmrazik: ok, I think that my thunderbird folder is stuck [11:01] mmrazik: no email from the paste 1h30 on those branches [11:01] Saviq: I also don't really expect us to have non-Qt C++ stuff in unity-phablet [11:01] mmrazik: sorry for the false positive, I should have rechecked on the branch itself [11:02] didrocks: no worries [11:02] Saviq: I don't really see the point in moving away from QtTest [11:02] mzanetti, the point is to be consistent, and yes we will have non-Qt C++ stuff [11:03] mzanetti, the "transport layer" of libunity, so all the APIs, will end up in the same repo [11:03] so apart from implementations themselves [11:04] everything that we expose as Unity APIs, will live with the shell in the same repo (obviously we need to reorganize the repo) [11:04] mzanetti: can you reapprove https://code.launchpad.net/~aacid/unity/cleanHudApiWithFindChild/+merge/154067 i had to merge it with new stuff [11:08] mzanetti, DashBar has properties like initialHeight and maxHeight, itemSize bla bla [11:09] mzanetti, that obviusly have to be sane values [11:09] mzanetti, (like setting 0 or negative will fail) [11:09] mzanetti, does it make sense to test those? [11:09] seb128, stay in submarine, we still have a lot more one liner merge proposals for your inbox! ;) [11:10] davidcalle, hehe, I wish there was a way to say "let me in the team so I can approve some MR if needed but stop spamming me" [11:11] tsdgeos mzanetti can you review? https://code.launchpad.net/~schwann/unity/rename-to-gallery-app/+merge/154324 [11:13] gusch: works, want me to top approve already or need to coordinate stuff? [11:14] tsdgeos: you can just approve [11:14] ok, done [11:14] tsdgeos: thx [11:14] gusch: will you want a shell release today too? [11:14] tsdgeos: would be nice, but I have some links, so it works with the old one too (except the HUD) [11:14] ok [11:16] tsdgeos: I put it back to "needs review" as mzanetti found an issue [11:16] sorry :-& [11:16] mzanetti: what is it? [11:17] oh lol [11:17] :D [11:20] tsdgeos, is it ready to go now? https://code.launchpad.net/~dandrader/unity/phablet_tst_ResponsiveGridView/+merge/154310 [11:21] kaleo, ping [11:22] didrocks, regarding https://bugs.launchpad.net/unity/+bug/1157660 is it with this morning merges on scopes? Because a reason for them not to appear can be a categy id mismatch between scopes and their respective master scopes, which has been merged only a moment ago for almost all scopes. [11:22] Launchpad bug 1157660 in Unity "We need to ensure that all scopes are started and search for by the server" [Undecided,New] [11:22] category* [11:23] davidcalle: the server doesn't as well recommends everything [11:23] didrocks, ok [11:23] davidcalle: anyway, it's a sanity check we need to do I guess :) [11:23] so opened it more like "let's keep that on the radar" [11:23] didrocks, ok [11:26] tsdgeos mzanetti pushed the fix https://code.launchpad.net/~schwann/unity/rename-to-gallery-app/+merge/154324 [11:27] aproved again [11:27] tsdgeos: thx [11:27] signed off [11:27] sorry for nitpicking... but it was so obvious out of line I had to mention it :D === mmrazik is now known as mmrazik|lunch [11:39] Saviq, ping [11:39] tvoss, pong [11:45] mzanetti, Saviq: fyi... pushed my branch again... hooked up as a plugin (cmake picks it up now)... also made cleaner separation between frontend and backend [11:45] MacSlow, I just replied === dandrader is now known as dandrader|lunch [11:48] tvoss: pong [11:48] kaleo, cancel that, sent you an invite to an input method meeting next Monday, feel free to dispatch to the most appropriate person on your team [11:49] k [11:50] tvoss: I'm not sure I understand what it is though [11:51] kaleo, integration of input method with the system and toolkits [11:51] kaleo, I wasn't sure if sdk is interested, so I added you :) feel free to mark yourself optional, too [11:52] tvoss: I'll add loicm [11:52] kaleo, ack === MacSlow is now known as MacSlow|lunch === mmrazik|lunch is now known as mmrazik [12:01] who wrote the original DashBar? [12:02] just wondering if it is supposed to scroll if we have many lenses [12:02] Cimi: hmm... I think noone ever thought about having so many lens that it fills up the bar :D [12:02] Cimi: but yeah. very good point and also testcase [12:02] mzanetti, there is sort of code already [12:03] width: Math.min(Math.max(dashBar.width/2, units.gu(40)), count * itemSize) [12:13] Trevinho, hey [12:13] om26er: hi [12:14] Trevinho, in raring it seems when a window is opened maximized its not fully maximized [12:14] Trevinho, the right most part of the screen is occupied by the application that is opened behind [12:14] Trevinho, have you seen that ? [12:15] om26er: I don't... But both didrocks and andyrock did [12:16] Trevinho, then i will guess the bug is in good hands, i need not worry ;) [12:17] yeah ;) [12:18] mzanetti: https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner/142/artifact/qml_phone_shell.tests.testconfigurations.TestNexus10.test_hide_hud_click.ogv [12:18] ?¿?¿?¿? [12:20] tsdgeos: dafuq [12:21] exactly [12:21] 36 seconds of nothing :D [12:21] tsdgeos: ah wait... this could happen if the shell doesn't load the home lens [12:21] mzanetti: but the mouse didn't even try to click on the login field [12:22] tsdgeos: yeah... because it waits for the lens to be loaded and switched to the Home lens [12:22] tsdgeos: if that never happens the test aborts because the shell didn't start up [12:22] i see [12:22] so increase that 30? :D [12:23] tsdgeos: :/ [12:23] tsdgeos: I actually believe that wouldn't change anything [12:24] tsdgeos: I think theres something fishy in the logic that switches to the home screen [12:24] tsdgeos: as I've seen this on my desktop too every once in a while [12:24] oh :/ [12:24] may be === dandrader|lunch is now known as dandrader [12:28] mzanetti, how can I access first argument of an emitted signal through signalSpy? [12:28] mzanetti, there's signalArguments but doesn't say much === _salem is now known as salem_ [12:28] Cimi: in Qt/C++ you can get all the arguments as a QVariantList [12:29] mzanetti, so not an easy way to access them in qml [12:29] Cimi: I guess in javascripts its just an array then [12:29] ah ok [12:30] can anyone quick review https://code.launchpad.net/~aacid/unity/run_on_devive_update_before_get/+merge/154348 ? [12:31] it's an easy one, i promise :D [12:32] mmrazik: ccache: FATAL: Could not create /var/cache/pbuilder/ccache/e/f/3bdb7d27ce3c81355980190e8fc4a4-1612312.o.tmp.stdout.ps-precise-server-amd64-smp.1608 (permission denied?) in https://jenkins.qa.ubuntu.com/job/unity-phablet-quantal-i386-autolanding/50/console [12:32] tsdgeos: looking [12:32] tsdgeos: approved [12:35] Saviq, I am using model: Lenses {} : how do I get the number of lenses? model.count is undefined [12:35] Cimi, did you look at lenses.cpp? [12:35] no [12:35] or lenses.h, even? [12:35] do :) [12:36] Cimi, http://bazaar.launchpad.net/~unity-team/unity/phablet/view/head:/plugins/Unity/lenses.h#L61 [12:36] ok [12:36] rowCount maybe [12:36] tsdgeos: No idea :-/ First time I see something like this. [12:37] lets see if it happens again [12:37] mmrazik: maybe no free space? [12:37] Cimi, note it's a method, not a property [12:37] tsdgeos: checked that. looks ok [12:37] ok [12:37] Saviq, indeed, that's why maybe [12:37] :) [12:38] Saviq, ok works [12:38] Saviq, problem is, in my tests I want to be abstract I guess [12:38] Saviq, so shouldn't be model.count the right usage? [12:39] or I can keep using model.rowCount() in the tests, assuming I have Lenses {} as model? [12:39] Cimi, rowCount() is required on models for a ListView [12:40] ok [12:43] sil2100: any news on the flacky tests? === Pici` is now known as Pici [12:48] didrocks: with the gcalctool fix seb128 prepared, it's much better now - but the super_h super test will still fail from time to time, the proper fix to this issue requires some compiz hacking which I try to achieve, although I'm not sure if it's even worth the time ;p [12:49] mzanetti, I have issues with signalSpy [12:49] didrocks: but I'm re-building indicator ap tests to see the results [12:49] sil2100: ok thanks [12:49] mzanetti, signalSpy detects two signals while there was only one [12:50] detects to emitted signals [12:50] *two [12:50] Cimi: you clear()ed it before? [12:50] didrocks: I already had a loose discussion about the lack of feedback from compiz during setting the compiz options, but it's a pain in the ass [12:50] mzanetti, yes [12:50] sil2100: yeah, should we set the trigger to 3? [12:50] sil2100: as nvidia has 3 failures [12:50] which is what is blocking the landing regularly [12:50] Cimi: hmm.. would need to see the branch then... [12:50] sil2100, btw any news about https://code.launchpad.net/~sil2100/compiz/quickfix_1141079_raring/+merge/153430 ? [12:52] seb128: I fixed it up properly in the main compiz branch, but Sam also wants to have tests written for it - I'll try finishing those later today [12:52] Then we can have it merged [12:52] sil2100, ok, thanks, we should have it fixed one way or another in raring [12:52] 13:50:28 didrocks | sil2100: yeah, should we set the trigger to 3? [12:52] 13:50:32 didrocks | sil2100: as nvidia has 3 failures [12:52] sil2100: in case you missed it ^ :) [12:53] mzanetti, ~unity-team/unity/phablet.dashBar_bottomswipe [12:53] pushed [12:54] didrocks: no, let's maybe leave it as it is, if the proper fix will take too long I can always hack the super_h test, which would basically mean 2 less failing tests [12:54] mzanetti, [12:54] QDEBUG : qmltestrunner::DashBarTest::test_itemSelected() before click 0 [12:54] QDEBUG : qmltestrunner::DashBarTest::test_itemSelected() onItemSelected, within DashBar [12:54] QDEBUG : qmltestrunner::DashBarTest::test_itemSelected() after click 2 [12:54] didrocks: the one that is left was a failure caused by gcalctool, so it'll be fixed now anyway [12:54] console.log("before click", signalSpy.count) [12:54] // (item, x, y, button, modifiers, delay) [12:54] mouseClick(dashBar, x, dashBar.height / 2, Qt.LeftButton, Qt.NoModifier, 0) [12:54] console.log("after click", signalSpy.count) [12:54] sil2100: ah, so we should relaunch? [12:54] Cimi: ckecking out... please avoid pasting more than 3 lines in an irc channel [12:55] even if its 2x3 lines :D [12:55] didrocks: I relaunched, waiting for the result ;p [12:55] sil2100: argh, you only relaunched the testing job [12:55] sil2100: not the full stack [12:56] so we'll need to relaunch it later on again [12:56]