=== _salem is now known as salem_ === duflu_ is now known as duflu === salem_ is now known as _salem === _salem is now known as salem_ === hyperair is now known as minefield === minefield is now known as hyperair [08:14] Trevinho: ciao! LIM work very well here, but I miss the double click to maximize the window. Do you plan to add it? [08:14] mardy: ciao! [08:14] mardy: it's already there :) [08:15] mardy: you only have to tune it [08:15] Trevinho: ah, dconf? [08:15] mardy: yes, see https://bugs.launchpad.net/unity/+bug/1283695/comments/2 [08:15] Ubuntu bug 1283695 in Unity "LIM breaks double-tap title to maximise" [Undecided,Won't fix] [08:16] Trevinho: cool, thanks! :-) [08:17] mardy: np [08:17] Trevinho: works like a charm :-) [08:19] mardy: I did that at the beginning, then designers preferred to avoid the delay it causes (no other way, unfortunately... I can't go into people's mind :P) [09:13] this is very confusing [09:13] i'm having poblems with LVWPH not scrolling to the end [09:13] and i've converted it to a regular LV and it doesn't scroll to the end either (it bounces a bit early) [09:13] but if instead my own QQuickView i use qmlscene [09:14] then it works [09:14] :S [09:14] what is qmlscene doing! [09:14] * tsdgeos digs [09:32] buuuuuuuuuuuuuuuuug [09:42] now why isn't my tablet showing up in adb devices... [09:45] Saviq, could you pls add https://code.launchpad.net/~mzanetti/unity-scopes-shell/activate-appid-2/+merge/205443 if you plan to land right-edge before new-scopes? [09:45] add to unity8 landing ask that is [09:48] Saviq: do you have 5.0 or 5.2 installed? [09:48] on desktop [10:02] tsdgeos, 5.2 [10:02] mhr3, there isn't a landing ask for right edge yet, but will make sure to add it there [10:03] mhr3, but anyway, isn't -scopes blocked due to 5.2 already? [10:03] Saviq: i have a weird behaviour of using a qquickview vs using qmlscene [10:03] tsdgeos, yeah, I read above, want me to downgrade to 5.0? [10:04] wanted to know if it was new in 5.0 [10:04] or somehow compiz had a regression since last time i tried :D [10:04] Saviq: it's ok, i can do it too, just wanted to know if you had it at hand [10:04] * tsdgeos ppa purges [10:05] greyback, usually killing adb and/or rebooting pc solves [10:05] Cimi: did the first one, second didn't make sense as phone connects fine with adb [10:06] Saviq, yea, it is.. any idea when the 5.2 migration might finish? [10:07] mhr3, we're down to two last issues as of yesterday [10:09] Saviq, so i take it you're not leaving today until that row says Landed? ;) [10:10] mhr3, you mean the new scopes one? didn't I tell you it's not gonna happen before Tuesday, probably? :P [10:10] Saviq, no, 5.2 [10:10] mhr3, ah, not mine to say [10:47] * greyback needs more cpuz [10:49] greyback, hey, there's three of your branches that I reverted from new-scopes to be merged into trunk separately - see history of https://code.launchpad.net/~unity-team/unity8/new-scopes-clean-to-trunk/ [10:49] greyback, would be good to get those in (with tests) sometime [10:49] Saviq: sure [10:50] greyback, not hugely pressing, though [10:51] Cimi, hey, please let me know when you have the carousel changes ready [10:52] Saviq, I was still testing the zoomable branch :) [10:52] Saviq, trying to see why wasn't working properly here [10:52] Saviq, I can switch back to carousel now [10:52] Cimi, please do, higher prio [10:53] Saviq, I don't understand why we have to split [10:53] Saviq, I basically have to restart from scratch [10:53] Cimi, because we want to reduce new scopes diff where possible, and CardTool and such are already in trunk [10:53] ok [10:54] Saviq, I'll do from scratch [10:54] Cimi, should be easy to cherry-pick [10:55] Cimi, I can do it if you're not sure how [10:55] Saviq, I'm not good with bzr, but I know well to copy-paste ;) [10:55] Cimi, sure [10:55] Saviq, how shall we write comments? [10:55] Saviq, I started doing /// because you asked me so weeks ago [10:56] Saviq, not we are back to /* [10:56] Cimi, not /* but /*! [10:56] Cimi, just look at style in the file [10:56] Cimi, we want to be consistent within files at the least (until we become consistent across them) ;) [11:16] what was the last time someone executed all our autopilot tests on the phone? successfully that is [11:16] I can't manage to get a successful run [11:17] elopio: hey, you around yet? [11:19] mzanetti: Saviq did the other day i think [11:19] mzanetti, if you get "process not found" that's Qt 5.0 crash [11:20] right. I get 2 of them for the emulator tests elopio did, and one of the notifications tests failing reliably [11:25] mzanetti, which notification one? and apt-cache policy libautopilot-qt ? [11:25] Saviq: unity8.shell.tests.test_notifications.EphemeralNotificationsTests.test_append_hint [11:25] did unity become longer to compile in the last month or is it just my pc? [11:26] 1.4+14.04.20140303.is.1.4+14.04.20131106.1-0ubuntu1 0 [11:26] mzanetti, what's the failure? [11:26] Cimi, apt-get install ccache [11:26] yeah, process not found [11:26] mzanetti, ah then that's the crash [11:26] interesting that its always this one tho... [11:26] Saviq, don't have much storage [11:26] Saviq, how much it takes? [11:26] in .cache ? [11:26] Cimi, 1GB on my home [11:26] Saviq, ok, might be enough [11:27] I mean, fine [11:49] Saviq, https://code.launchpad.net/~unity-team/unity8/carouselTool/+merge/209655 [11:49] Cimi, cheers [11:50] Cimi, no need for type:int, it's only needed for aliases and in case you want to override what the property type actually is [11:50] Saviq, I forgot to take carousel as grid [11:50] Saviq, is it required? [11:50] Cimi, yes [11:51] Cimi, let me get you a diff [11:51] Saviq, i can merge [11:51] no problem [11:51] Cimi, http://paste.ubuntu.com/7043700/ [11:52] Cimi, that branch is against new-scopes I think, so you don't want it whole [11:52] Saviq, I override those changes [11:53] Saviq, so I don't need it [11:53] apart cardfiltergrid [11:53] Cimi, all of them? [11:53] Cimi, ah, in that case is fine [11:53] Cimi, sentence case in docs, please [11:55] Cimi, what's 32 + property var carouselTool used for? [11:55] mzanetti, question on your right-edge-2 behavior (still flashing a second device to try out your branch again, as I'm doing development on the first one) [11:55] Saviq, not used now [11:55] Saviq, it is used with the new scopes [11:55] mzanetti, when there's no application running and I do a right edge drag on the dash, what happens? [11:56] Saviq, but since I added cardtool to dashrenderer [11:56] Saviq, it can be added [11:56] dandrader: it should do that animation on the dash [11:56] mzanetti, same as currently? [11:56] dandrader: where it shrinks a bit and gets black [11:56] yeah [11:56] Saviq, I tried to reduce the diff for new scopes indeed [11:56] mzanetti, ok [11:57] Cimi, I just wonder, why do we want the whole carouselTool as property, and not just the needed props directly? [11:58] Saviq, because carousel in new scopes needs to read cardtool [11:59] carouseltool [11:59] Cimi, yeah, that's fine, but why do we need the QtObject? [11:59] sorry hold on [11:59] the property is cardtool [11:59] Cimi, why can't CardTool just deal with it directly [11:59] 94 /*! [11:59] 95 + \brief Tool component for Carousel. [11:59] 96 + */ [11:59] 97 + property var carouselTool [12:00] Saviq, doesn't seem to be required from outside indeed [12:01] Saviq, I don't know if it was some sort of future-proofing I did when I wrote the branch or a leftover of an old test [12:02] Saviq, I don't think it's necessary to expose carouselTool from outside [12:02] Cimi, ok, please kill it then [12:02] Saviq, I think that I used it to calculate from the cardcarousel the dimensions of the tiles [12:02] Saviq, then you said to use the static values we set in cardtool [12:03] Saviq, so I removed from cardcarousel but I left in cardtool === MacSlow is now known as MacSlow|lunch [12:03] yep that might be it - removing [12:04] Saviq, done [12:04] Cimi, ok, will do a proper review asap [12:30] mzanetti: I'm here. [12:30] elopio: good morning. Again, everything resolved by now :) [12:30] mzanetti: I like that :D [12:31] I guess so [12:32] mzanetti: I'm watching your presentation, because yesterday the stream was broken for me. I need to turn many autopilot tests on the notes app into qml tests. [12:32] elopio: +1 on that one :D [12:32] I'll probably be asking dumb questions [12:32] elopio: no problem... as long as we get more reliable tests, feel free to ask me any time [12:35] elopio: is the whole presentation up by now? [12:36] 'cause the stream broke down somewhere in the middle and cut the video in 2 parts [12:37] mzanetti: I see two videos of ~20 minutes on david's youtube channel. [12:37] ok. that should be everything then [12:44] trying to turn my N10 display on. "powerd-cli display on bright" doesn't seem to be enough... === smb` is now known as smb [12:47] greyback, Saviq: any ideas? ^ [12:47] dandrader: just had same issue here. Rebooting worked. Unsure why however [12:49] greyback, heh, screen lit up just after I issued "reboot" :) [12:49] dandrader: yeah, suspect powerd isn't doing quite the right thing on manta === alan_g is now known as alan_g|lunch === boiko_ is now known as boiko [13:36] dandrader, it won't power it on, it will keep it on [13:36] dandrader, you need mir to actually talk to the hardware to turn the display on [13:37] dandrader, powerd-cli only overrides the policy, but without mir executing it there's nothing that will actually tell the hw to turn it on [13:38] Saviq, I have unity8 manually running... [13:38] Saviq, does that count as "mir"? [13:40] dandrader, yes, should, assuming it listens to powerd's commands to turn the display on... [13:41] fginther, hey, it looks like https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-mako/5673/console isn't using unity8's helpers for unlocking the screen, could it start doing so? [13:42] fginther, we're removing the BottomBar... thing [13:43] mhr3, oh, you're back, just posted on #ferrets: [13:43] pstolowski, we need to do something so that scope-registry is always started when unity8 is [13:43] pstolowski, /me thinks "start on unity8 starting" or something similar [13:43] Saviq, why would that be necessary? [13:43] mhr3, install unity8, no scopes [13:44] Saviq, it currently starts with dbus, what's the issue? [13:44] Saviq, zmq sockets are kinda connectionless [13:44] mhr3, pstolowski, then maybe it just needs to start on installation [13:44] but that's probably difficult [13:44] mhr3, pstolowski, as right now you need to "start scope-registry" the first time you install it, as otherwise without restarting the session there won't be a "dbus started" event [13:44] yea well... log out after installing unity8 :) === MacSlow|lunch is now known as MacSlow [13:45] mhr3, tell that to the test runners [13:45] hmm [13:45] Saviq, make the test runners start the registry then :) [13:45] mhr3, I can make unity8 job start the registry [13:46] well... why not [13:46] mhr3, but then that's what events are supposed to be, aren't they... [13:46] Saviq, then you should probably start smart-scopes-proxy [13:46] Saviq, as registry depends on it [13:46] pstolowski, right [13:47] * Saviq looks for an upstart expert... [13:51] Cimi, we can't merge the new getRenderer into trunk [13:52] Cimi, https://code.launchpad.net/~unity-team/unity8/carouselTool/+merge/209655/comments/493279 === dandrader is now known as dandrader|afk === alan_g|lunch is now known as alan_g [14:22] Saviq, can you point me to what the unlock should be doing? === dandrader|afk is now known as dandrader [14:27] fginther, so, is it decided that it's ok for unity8 to always run under testability to do the unlocking? if so, there's unity8.autopilot.process_helpers.unlock_unity [14:27] Saviq, still ok to to merge trunk to new-scopes? people are still using the demo ppa, and the pkg there is superseded [14:28] mhr3, the recipes are disabled anyway [14:28] Saviq, not a big deal to trigger them :) [14:28] Saviq, so just run that with phablet-test-runner and the magic happens? [14:29] fginther, .restart_unity_with_testability first [14:29] Saviq, ack [14:29] fginther, and yes, those two calls would be it [14:29] mhr3, we'll merge new-scopes soon enough, people could just force the versions after all [14:29] oh, nvm there isn't much to merge [14:29] Saviq, is remaining on the old method going to be a blocker to testing? [14:30] Saviq, i'll just kick the build for now [14:30] fginther, yes, since it relies on a dbus interface that's going away [14:30] Saviq, ok [14:31] Saviq, I'll get the work on the queue and prioritize it [14:33] fginther, thanks [14:34] Saviq, re attributes, i'll expose them to shell as a single role which will hold the classical array of dicts, right? [14:35] mhr3, yup [14:35] mhr3, no need for a model [14:35] Saviq, and in preview header widget? [14:36] i guess you also want it there as a single attribute? [14:36] greyback, standup? [14:36] d'oh [14:36] mhr3, yeah, same exact thing [14:36] mhr3, as it's meant to be the same widget ;) [14:36] which means it will be different two things for scope authors... but you don't care about that :) [14:39] mhr3, not really ;D [14:39] mzanetti, hey [14:39] mzanetti, regarding https://code.launchpad.net/~mzanetti/unity8/disable-hud/+merge/209701 it seems the unlock script depends on an interface that is being removed [14:39] om26er: one sec. in the standup right now [14:39] om26er, read up my conv with fginther [14:39] sure [14:40] Saviq, I just came in ;) [14:40] om26er, right, yeah, we know, fginther will fix :) [14:41] Saviq, fginther if you want I can work on it [14:41] its a great time to clean that script [14:41] dednick, standup? [14:41] dednick, still have a minute or so to join :) [14:42] om26er: so, yeah... known, we're working on it. for now I've proposed another branch that just disables the hud tests which are blocking the apps [14:44] mzanetti, yeah, saw that as well. [14:48] didrocks, question about 0replaceme - when is that going to be actually replaced? [14:48] om26er, if you have some cycles, that would be awesome, I won't be able to get to it for a few hours [14:49] didrocks, and since we might build the binaries at the same time as when the dep pkgs are built, isn't there going to be a problem? [14:49] didrocks, or is it replaced already when going to a landing silo? [14:49] mhr3: when it will be published in distro [14:49] fginther, yeah, I have many ;) [14:50] mhr3: the dep packages requires the new version? [14:50] didrocks, they might if we have them in the same silo, right? [14:50] mhr3: you need to bump the build-dep in that case [14:51] otherwise, everything will start to build at the same time (as you didn't require latest version) [14:51] didrocks, sure, but when the dep pkg will be building it will see that it needs version 0replaceme instead of the real version [14:51] ...i hate this wifi chip so much [14:52] mhr3: nothing to do with that, this is for binary deps, not build-time [14:52] mhr3: so if you need latest one at build-time, you need to bump the build-deps [14:52] then, if you do that [14:52] and let's say 0replaceme is replaced with "42" [14:53] B will depend on A (=< 42) [14:53] if it's using one of the new symbols [14:54] hmm, i guess i don't really understand when is the version from the symbols used [14:54] but if you say it's going to work, i'll trust you [14:54] it's easy [14:54] so you have A and B [14:54] B depends on A [14:55] A provides new symbols: aaaaa 42 [14:55] if B uses "aaaaa" [14:55] it will be B depends on A (=< 42) [14:55] but for that, you need to have B built against the new A [14:55] if you use new symbols [14:56] so, if you want B wanting on A to be built [14:56] you need to have B bumping the build-dep version on A [14:56] and you need to do that manually [14:58] didrocks, hmm, i thought it's also helping with upgrades, when for example soname of A doesn't change, but there are new symbols in A and B is using them, then it would make sure that B doesn't get upgraded unless A is as well [14:58] mhr3: yeah, it does, once the packages are built [14:58] as you will get: [14:58] B depends on A (=< 42) [14:58] as a consequence [14:59] but nothing build-time wise [14:59] right, but at the point when B builds, all the symbols in A are 0replaceme, so how can that work? [15:04] *all the NEW symbols [15:05] no [15:05] before B builds, my script is changing those [15:05] and then uploading to the ppa [15:06] (with the correct version) [15:06] didrocks, or is it replaced already when going to a landing silo? [15:06] mhr3: when it will be published in distro [15:06] ^^ you're confusing :P [15:06] well, not sure why you want implementation detalis [15:06] if it's when being uploaded to ppa then yea, sure everything fine [15:06] so, it's replaced when I prepare the branch [15:07] this branch is used to build a source package [15:07] this source package goes to the silo ppa [15:07] but this branch is only available to you (and trunk) when the package is published :p [15:07] I just talked about the last part as I thought it was what matters to you ^ [15:08] i see... this time i was more interested in the details :) [15:08] didrocks, but ok cool, everything working fine, so awesome :) [15:08] ;) === alan_g is now known as alan_g|tea === alan_g|tea is now known as alan_g [15:37] mzanetti, since you're on fire, https://code.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/+merge/207282 please :) [15:37] Saviq: ack [15:37] mzanetti, I just pushed tests for it [15:37] Saviq: am I talking bullshit here or does it make sense? https://bugs.launchpad.net/ubuntu-weather-app/+bug/1288819 [15:37] Ubuntu bug 1288819 in Ubuntu Weather App "font.pointSize should not be used together with units.gu()" [Undecided,New] [15:37] mzanetti, definitely [15:38] mzanetti, making sense ;) [15:38] I do wonder a bit why it looks ok on the phone [15:38] probably because the phone doesn't properly set the font DPI for that display [15:38] as we're not using it anyways [15:38] mzanetti, or just because they selected the fonts right ;) [15:39] well, it looks good on desktop with 8 px/gu and also looks good on phone with 18 px/gu, but it breaks on desktop with 18px/gu [15:40] mzanetti, ah, and a pretty small one https://code.launchpad.net/~unity-team/unity8/newscopes-design-tweaks/+merge/209492 [15:40] ack [15:44] * greyback in VT land [15:52] Saviq: make testCard segfaulting, known 'cause of 5.0? [15:52] mzanetti, no, probably fails [15:53] mzanetti, might need try* instead of verify [15:53] mzanetti, let me fix [15:53] mzanetti, that's for the font/mascot tests? [15:53] PASS : qmltestrunner::Card::test_mascotShape(Art and summary) [15:53] Segmentation fault (core dumped) [15:53] Saviq: ^ [15:53] mzanetti, yeah, qmltestrunner segfaults on failing tests in there.... [15:53] ok... [15:54] ok, summed it up in here too: https://code.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/+merge/207282/comments/493431 [16:06] mzanetti, http://bazaar.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/revision/742 http://bazaar.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/revision/743 [16:07] mzanetti, greyback: so an app is running and its tile on the dash is clicked. code "ApplicationManager.activateApplication(application.appId)" is called. Anything against having "application.activate()" instead? [16:09] mzanetti, pushed a small change to the tests, too [16:10] dandrader: not really. AppMan API could do with a review tbh [16:10] dandrader, sounds like should be fine (and it should be going away anyway), and even when we start actually killing apps (and keeping them in Recent ones), we'll keep the Application object still, I think [16:10] dandrader, (it going away → running apps in dash going away) [16:10] dandrader: greyback: fine with me... [16:11] dandrader: ah, if you're changing this... [16:11] yep we keep the Application object [16:11] I'd vote for startApplication() to emit requestFocus() and only the shell then to actually grant focus by calling focusApplication(appId) too [16:11] to keep it aligned with activate() [16:12] greyback: ^ jfi [16:14] mzanetti: agreed. Makes sense on desktop too, where if you launch app2, then focus on existing app1, when app2 appears afterwards, shell can know not to remove focus from app1 [16:14] Saviq, my branch breaks things [16:14] Cimi, not any more it doesn't [16:15] Cimi, http://bazaar.launchpad.net/~unity-team/unity8/carouselTool/revision/758 [16:15] the thing is, the current AppMan API doesn't make all that much sense anymore on the "unity8 as mir compositor" work [16:16] Saviq, hah, had same one locally [16:16] e.g., the focus management will live entirely in the unity8 qml scene [16:16] dandrader: well... not entirely [16:16] dandrader: we still will have a model holding all the apps and their info [16:16] dandrader: the activate/focus/etc might go away indeed [16:16] Cimi, we need a branch on top of that to be merged into new-scopes, please [16:17] dandrader: but all the rest (applicationAdded() appRemoved() etc will continue to make sense imo) [16:17] mzanetti, I didn't mean that the AppMan API itself won't make sense, just part of it [16:18] we will still need it for sure [16:18] yeah. and I think most of is actually ok.. everything related to the model [16:18] just the focus() unfocus() calls yeah... but right. I get your point... makes my previous request obsolete indeed === dandrader is now known as dandrader|lunch [16:31] mzanetti, top-ACK https://code.launchpad.net/~unity-team/unity8/newscopes-design-tweaks/+merge/209492 please? [16:33] mzanetti, https://code.launchpad.net/~unity-team/unity8/new-scopes-clean-to-trunk/+merge/209459 would be the next one I'd like you to look at (see description for things that are not ready for it still) [16:33] Saviq: done... as of now I'll do 5 push ups whenever I forget to top approve [16:33] mzanetti, ;D [16:33] mzanetti, in that case: https://code.launchpad.net/~unity-team/unity8/themeing-font-and-mascot/+merge/207282 ;D [16:33] Saviq: commit message missing [16:33] * mzanetti is doing push ups [16:34] mzanetti, it's WiP still [16:36] oh... its the full thing [16:37] meh... /me starts rebasing its right-edge stuff :P [16:38] won't get through this today [16:38] but I'll start tomorrow [16:38] mzanetti, yeah, for sure [16:38] mzanetti, well now ;) you can always make the review take long enough so that right edge gets in first ;D [16:39] mzanetti, and make sure right edge review is ready soon ;) [16:39] I'm sure there's someone on the team in need of a case of beer ;D [16:39] Saviq: haha [16:39] Saviq: did you revert the conflicting stuff? [16:40] Shell.qml looks quite clean [16:40] so probably not even many conflicts after all [16:40] mzanetti, yeah, see the last few commits [16:40] thanks for that :) [16:40] mzanetti, I cleaned it up to just be new-scopes-related === robru-sick is now known as robru [16:42] Cimi, please make sure to recreate carousel-dynamic-fallback on top of carouselTool, MP'd into lp:~unity-team/unity8/new-scopes-clean-to-trunk [16:42] Saviq, sure [16:42] Cimi, with carouselTool as prereq [17:02] elopio: hey there [17:02] i was looking at http://ci.ubuntu.com/smokeng/trusty/touch/mako/223:20140306:20140304/7006/ [17:02] kgunn: hello. [17:02] and trying to square the tests listed there with the wiki [17:02] https://wiki.ubuntu.com/Touch/Testing#Testing_your_Ubuntu_Touch_Code_before_submission [17:02] ....specifically...is keyboard not run as part of smoke testing ? [17:02] or is it hiding under another test somehow ? [17:03] kgunn: I don't know much about keyboard. veebers was the one working on it. [17:03] I know that keyboard has its own autopilot tests, and if they are not listed here is because they are not being run. [17:03] but they should [17:04] also I know that none of our autopilot tests are using the osk yet. [17:04] elopio: ok...this is the one listed in the wiki "phablet-test-run -p ubuntu-keyboard-autopilot ubuntu_keyboard" [17:05] kgunn: yes, that's the suite. [17:06] kgunn: and there's one discussion we need to have in order to let all the autopilot tests on the phone to use the osk. [17:06] it's about starting malit with testability. [17:06] mhall119, ping [17:07] mhr3: in an appdevweek hangout atm [17:07] kgunn: but if on top of new scopes, qt5.2 and android 4.4 we also switch to use the osk, we will be lynched. [17:07] mhall119, k, sending a mail instead [17:23] elopio: ping :) [17:29] Cimi, the rating input and output widgets are merged in unity8 trunk already? [17:29] think so mhr3 [17:29] Cimi, how is the review text and star rating passed back to the scope? [17:29] (need to document that) === dandrader|lunch is now known as dandrader === alan_g is now known as alan_g|EOW [18:07] mhr3, ah, actually I am not sure I finished that [18:08] Saviq, https://code.launchpad.net/~unity-team/unity8/carouselTool-new-dash/+merge/209746 [18:08] mhr3, was merged I think [18:09] mhr3, https://code.launchpad.net/~cimi/unity8/previews.PreviewRatingDisplay/+merge/207625 [18:17] Cimi, ok, found it, think we'll need to adjust it a bit [18:26] mzanetti: I branched ubuntu-sdk-tutorial and the currency converter tests there are not using UbuntuTestCase [18:27] oh, I've found your branch. [18:27] elopio: https://code.launchpad.net/~mzanetti/ubuntu-sdk-tutorials/use-ubuntutestcase/+merge/209455 [18:27] ah [18:37] Saviq: any specific day/time next week that is bad for you for the Unity API docs website discussion? [18:39] Saviq: http://summit.ubuntu.com/uds-1403/meeting/22199/client-1403-api-website-for-unity/ [18:44] mhall119, any place that's free in my calendar is fine [18:45] mhall119, bonus points if it's before 7pm UTC :) [18:51] mzanetti: what's the package that installs UbuntuTestCase? [19:10] elopio: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1288937 [19:10] Ubuntu bug 1288937 in Ubuntu UI Toolkit "UbuntuTestCase.qml missing in installs" [Undecided,New] [19:10] mzanetti: I was looking at the right place then :) [19:10] elopio: not sure when this happened (in unity we use our own UnityTestCase.qml) [19:11] elopio: for now you can copy it yourself from the repo to that place [19:11] mzanetti: why are you duplicating it? Isn't the toolkit a dependency of unity? [19:11] elopio: unity hat this before the toolkit, and unity's has more [19:12] elopio: but as soon as this is fixed in the sdk I'll drop the findChild() stuff from UnityTestCase and inherit that from UbuntuTestCase [19:12] mzanetti: yes, that would be nice. We also need a suite that hits everything from the API, so this doesn't happen again. [19:13] tst_UbuntuTestCase, that's a dangerous hole to go into :) [19:14] elopio: well, we do have a tst_UnityTestCase.qml :) [19:14] in unity-api we even have mocks to test the tests :D [19:14] seems we've fallen into the hole :D [19:15] mzanetti: vila will like you. [19:20] \o/ [19:21] mzanetti: now I have a ListView of NoteItems. How can I populate that ListView for the test to have something to click? [19:21] mzanetti: it's a hole, but also the entry into the real world ;) [19:21] wait, I'm sorry, isn't it way past your EOD? [19:25] elopio: I just EOD, beer time ;) [19:26] vila: get one from me. I'll pay for it on may. [19:26] but I was asking mzanetti :) I don't know if I should stop bothering him now. [19:26] elopio: well, you set a model to that listview :) [19:27] not really sure what you're asking for [19:27] mzanetti: right. It's hard to make a QML test when you don't know QML :D [19:27] I'll start reading about models and give it a try. [19:27] elopio: ok... I see [19:28] elopio: so.. do you want to use the real notes or a fake one for the test? [19:28] ahh, it's easy. All the vars that seem to come out from nowhere on the delegate should be defined on the model. [19:29] elopio: yeah, thats it [19:29] mzanetti: I'll fake them. I just need to expand and collapse them. [19:29] elopio: if you want to define a mocked one in qml, just do ListModel { ListElement {name: "aaa"}, ListElement { name: "bbb"}} [19:30] then you set that model to the listview and inside the delegate there is a variable (aka. attached property) called "name" containing the values for each delegate [19:33] elopio: here's a nice example: http://bazaar.launchpad.net/~unity-team/unity8/trunk/view/head:/tests/qmltests/Dash/Music/tst_MusicPreview.qml [19:34] mzanetti: http://paste.ubuntu.com/7045932/ [19:35] elopio: yeah, I guess you need to set it still [19:35] not sure how NoteList works [19:35] mzanetti: but it's still showing a white window. [19:35] is this the notes app, right? [19:35] mzanetti: yes. [19:36] elopio: ack, not finished the first one but will drink the next for you ;-p [19:36] elopio: yeah, it's a listView [19:37] one sec, I'll change the paste [19:38] elopio: something like this: http://paste.ubuntu.com/7045954/ [19:39] note: you could also leave the Model definition inside the NotesList, but I think its more readable this way [19:39] I can see it now [19:40] it's easy when you know what you are doing :D I should go one step back and learn more QML. [19:40] thanks mzanetti! [19:40] np [19:41] elopio: jfi: this would work too http://paste.ubuntu.com/7045961/ (and can be handy in some cases) [19:42] mzanetti: got it. Here I like your first paste a little better. [21:46] Saviq: http://summit.ubuntu.com/uds-1403/meeting/22199/client-1403-api-website-for-unity/ 1500 UTC on Wednesday [22:27] mhall119, awesome, thanks