[00:01] lol, just noticed that was in the wrong channel... ;-) [07:12] tsdgeos: good morning [07:12] morning [07:12] how did it end up? [07:12] tsdgeos: how do you feel? [07:12] still sick [07:12] tsdgeos: :/ [07:12] but been sick for 3 weeks :D [07:12] tsdgeos: oh [07:12] not feeling as bad as yesterday evening [07:12] just regular [07:13] regular == sick? [07:13] thats now how it should be [07:14] :D [07:14] i know [07:14] tsdgeos: well, anyways, I'haven't been able to reproduce it locally [07:15] still failing on CI? [07:15] hi, [07:16] does anyone knows how to open autocad drawing in linux mint 14 ? [07:16] off topic [07:17] or any .dwg file? [07:23] tsdgeos: it passed, but the test seems to have become a bit more fragile [07:26] :/ [07:37] MacSlow: you aware that testNotifications is coredumping on the CI machine? [07:49] sil2100: hey! small approval, big karma: https://code.launchpad.net/~didrocks/camera-app/bootstrap/+merge/160553 [07:51] didrocks: hi! Approved ;) Thanks! [07:51] thanks ;) [07:51] It seems we missed quite a few, thought we already did camera-app [07:51] sil2100: keep me posted on the hud story ;-) [07:51] sil2100: yep [07:51] sil2100: on autopilot, I just heard that there are two autopilots [07:51] sil2100: one for phablet, one for desktop [07:52] sil2100: maybe something to look at :) [07:52] hm, I'll check that, as I was always rather targetting the desktop [07:54] sil2100: yeah, but the API is different [07:55] sil2100: so I guess touch apps are using the phablet autopilot API [07:55] (yeah /o\) [08:11] didrocks: not for long [08:11] didrocks: the full story is that the phablet autopilot is hacked together to work on the phablet - it doesn't work on the desktop. Autopilot 1.3 works on both, but isn't quite finished yet [08:12] thomi: yeah, I think we need a clear ETA, because right now, there is no way to have daily release in S [08:12] didrocks: but we're working hard to make it very usable by the time we get to oakland, and hopefully released by the time we leave oakland [08:12] thomi: and so, no more certified image in raring [08:12] thomi: ok, let's discuss that during the sprint :) [08:12] didrocks: absolutely [08:12] thomi: then, see my comment about backward compatbility, it's something we really need to acknowledge [08:13] \o [08:13] thomi: so let's say 1.3 will be the first one to ensure in the future backward compatibility :) [08:13] didrocks: can we leave that conversation until we're face to face? [08:13] sure [08:13] didrocks: we certainbly need to think about it, but now's not the time :) [08:13] my brain is fried [08:13] thomi: heh ;) [08:13] hacking boost::spirit [08:13] it's fun, but turns you into a vegetable afterwards [08:20] mzanetti, you around? [08:20] nic-doffay: I'm always :P [08:21] nic-doffay: nah... today I have to leave early [08:21] nic-doffay: how can I help you? [08:21] Regarding the signals, I wasn't able to find any info on how to connect one to an existing signal, eg onClick. [08:21] mzanetti, ^ [08:22] nic-doffay: you mean to fire a signal? [08:22] or to connect to a signal? [08:22] mzanetti, when onClick is called in another component I'd like to trigger the signal. [08:22] onclicked: doIt(); [08:22] nic-doffay: ^ [08:23] nic-doffay: or better: onClicked: someComponent.someSignal(); [08:23] nic-doffay: just call it like it would be a function [08:23] (which is what it is under the hood) [08:27] nic-doffay: signals short and easy: https://pastebin.canonical.com/89842/ [08:27] hope this helps [08:28] nic-doffay: on a public pastebin: http://paste.ubuntu.com/5597782/ [08:30] Ah right mzanetti thought there was something more to it! [08:32] tsdgeos: http://s-jenkins:8080/job/unity-phablet-qmluitests-testing/ [08:32] tsdgeos: 3 parallel qmluitests jobs on raring machines [08:32] :-) [08:43] mzanetti, in this case now, I'm having troubles calling the function on the dot component, check out this pastebin. What's the correct way to do that? [08:43] http://paste.ubuntu.com/5597803/ [08:45] nic-doffay: this does look good... calling "dots.unlockAnim()" should now trigger "functionInDotComponent()" for each dot. doesn't work? [08:45] Saviq: still here? [08:46] tsdgeos: I think he is flying to US today [08:46] mzanetti, nope. [08:47] mzanetti: i'm consfused in https://code.launchpad.net/~saviq/unity/phablet.release-172/+merge/159881 over if we shall have a quantal or raring release [08:48] nic-doffay: hmm... weird... can you add a print() in onUnlockAnim and see if that is called? [08:48] tsdgeos: it should be raring [08:49] so we shouldn't approve that one [08:49] tsdgeos: hi! [08:49] tsdgeos: I'd say we should use this branch and do the release to raring [08:50] sil2100: ho [08:51] tsdgeos: since you work on some things related to HUD... do you know if the new HUD will include results from indicators in the nearest time? [08:51] (since Ted's sleeping still) [08:51] sil2100: it already does but no it won't :D [08:51] worst sentence ever [08:51] * mzanetti agrees [08:51] Wow [08:51] sil2100: the new hud returns results per app [08:52] but the old hud ui doesn't know about "per app" [08:52] so the results are there [08:52] tsdgeos: and that's 'by design decision' [08:52] you just don't see them [08:52] because the UI is "old" [08:52] sil2100: i guess [08:52] didrocks: afair ted said you said it was ok? [08:53] mzanetti, in the output it says the function isn't defined. [08:53] tsdgeos: ok, so when there is no active application is running, there is supposed to be no results, right? Or should it return the desktop-related results? [08:53] tsdgeos: yep, it's an acceptable regression for the desktop [08:53] sil2100: ^ [08:53] nic-doffay: which function? [08:53] sil2100: no clue to be honest [08:53] Ah I think I see what the issue is mzanetti [08:53] Let me just check. [08:54] didrocks: ok, if this is going to stay, I'll also fix those HUD tests that assume that the indicator entries are supposed to be in the HUD result set [08:54] sil2100: yeah, that would be nice! [08:54] mzanetti, I had to reference it by the id of the Dot. [08:54] mzanetti: you do it (retarget to raring) and i test+approve? [08:55] tsdgeos: ok [08:56] grrrr [08:56] the new raring images come with the ppas [08:56] but the deb-src line commented out [08:56] so our run_on_device fails [08:57] or maybe is the add-apt-repository that doesn't default to -s now? [08:58] yay [09:00] its that? [09:01] think so [09:01] trying [09:08] yep that's working [09:08] thing is [09:08] there's no -s in quantal [09:08] there is in raring [09:08] i guess we can just add it and declare quantal dead? [09:09] for run_on_device at least [09:09] since the images are all raring based now [09:09] mzanetti: ↑ ? others: ↑? [09:09] tsdgeos: +1 [09:11] tsdgeos: ah btw... I found a way to reproduce an issue where the HUD button doesn't disappear any more :) [09:11] tsdgeos: by HUD button I mean the ubuntushape with the magnifier glass that appears when swiping from the bottom [09:12] mzanetti: press+hold and move the mouse out of the "window"? [09:12] tsdgeos: nope... it disappears fine there [09:12] ok [09:12] someone fixed that then :D [09:13] how' [09:13] ? [09:13] tsdgeos: its a bit tricky: open an app. then use the launcher to start another app on top of that. but before the second app gets launched, drag up the HUD button and release _after_ the second app appeared => HUD button is stuck [09:14] lol [09:15] tsdgeos: I had that every once in a while... Took me 2 months to figure how to reproduce it :D [09:15] can't repro [09:15] on the phone, right? [09:16] mzanetti: https://code.launchpad.net/~aacid/unity/raring_run_on_device/+merge/160577 [09:16] tsdgeos: yes, on the phone [09:16] release where on the button or outside the button? [09:16] tsdgeos: https://code.launchpad.net/~mzanetti/unity/phablet-release-172/+merge/160578 [09:16] tsdgeos: outside the button [09:17] nope [09:17] maybe i'm too slow bringing the hud up [09:17] tsdgeos: let me find the phone to try again... it was on the weekend when I played around with it for a while [09:17] mzanetti: can you have a better release message? [09:18] i.e. copy the crap from Saviq's one :D [09:18] tsdgeos: done [09:19] tsdgeos: just thinking a last time if we break anything with your change... [09:20] we can only break run_on_device [09:20] tsdgeos: let me send you a video :D [09:20] that's the only thing we touch [09:21] mzanetti: what may happen is that people with old flashed phones have problems [09:21] well they should a newer image [09:21] raring based [09:26] tsdgeos: http://ubuntuone.com/5vhyvosjZ1q8TANOtcPnPM [09:31] I'll switch qmluitests over to raring now. the job seems stable enough [09:32] mzanetti: that video confuses the hell out of me [09:33] i don't get to see the launcher :D [09:33] is it just because you're too fast? [09:33] or the camera too slow? [09:34] tsdgeos: no... no launcher involved [09:34] ok [09:34] tsdgeos: 2 runnign apps, [09:34] then i didn't read you correctly [09:34] :D [09:34] swiping from right to switch [09:35] ah ok [09:35] you previously said "then use the launcher to start another app on top of that" [09:35] let me try that [09:35] tsdgeos: I think the launcher should work too. but this way it might be easier [09:36] yeap can repro [09:36] wonder why that happens [09:38] may have to do with enabling/disabling the revealer while draging because the app in focus changes [09:38] mzanetti: autopilot test for this? :D [09:38] tsdgeos: it would be possible, yes [09:39] tsdgeos: on the phone you don't have to wait for the mouse cursor to move [09:39] given how slow the runners are [09:39] tsdgeos: so I guess in slow jenkins it would be quite easy to reproduce [09:39] but then again, in jenkins we use mouse instead of finger [09:39] otoh the jenkins uses the "fake" qtubuntu thing [09:40] so may not exactly be getting the same signals in the same order than the real thing [09:41] mzanetti: can you open a bug so we don't forget? [09:43] yes [09:44] tsdgeos: how do we report bugs? just report them to unity? [09:45] ye [09:45] this whole setup feels like a mess imho [09:45] and there you can select the pahblet thing soemwhere [09:45] let me see [09:46] mzanetti: something like what this does https://bugs.launchpad.net/unity/phablet/+bug/1160128 [09:46] Launchpad bug 1160128 in Unity phablet "Memory management in lenses.cpp" [Undecided,Fix committed] [09:51] tsdgeos: https://bugs.launchpad.net/unity/+bug/1172212 [09:51] Launchpad bug 1172212 in Unity phablet "HUD button gets stuck if application switches while dragging" [Undecided,New] [09:56] tx [09:57] mzanetti, I'm afraid I'm going to need some more of your help. Now that I've redone everything without using absolute positioning I'm in slightly unknown territory. I'm assuming my issue is regarding anchor points. Mind if you take a look at my branch? [09:58] nic-doffay: hit me [09:58] mzanetti, just going to push. [09:58] You'll see what's obviously wrong upon running qmlscene Infographics.qml [09:58] Great, my system crashed with a GPU hang, good to loose some AP test results [09:59] ... === sil2100_ is now known as sil2100 [10:01] mzanetti, https://code.launchpad.net/~nicolas-doffay/unity/infographics_transitions [10:02] let me know when you've run it. === mmrazik is now known as mmrazik|lunch [10:11] nic-doffay: do a "bzr merge lp:~mzanetti/unity/infographics_transitions" [10:13] nic-doffay: ok... pushed again... using radius is enough instead of defining centerX and centerY [10:13] I'm assuming that's due to the anchoring? [10:14] mzanetti, ^ [10:14] nic-doffay: no... by using x and y you don't use anchors [10:15] nic-doffay: a x,y of 0,0 would be the top left corner of your image [10:15] mzanetti, what's the use of centerX and centerY, I don't see them being used. [10:15] mzanetti: why unstable? http://10.97.2.10:8080/job/generic-mediumtests/1577/ [10:15] nic-doffay: merge again... I removed them again [10:16] I just did, still there, I'll just remove em myself. [10:16] tsdgeos: there is one failed test [10:16] nic-doffay: oh right... yes. I forgot them [10:16] nic-doffay: just throw them away [10:16] mzanetti: oh right, i was tricked by the no_videos thing [10:17] nic-doffay: does that help you? [10:18] nic-doffay: or was the question a different one? [10:18] mzanetti, not really I just see you added the radius again. [10:18] nic-doffay: so what exactly is the question then? [10:19] mzanetti, how to get the ring of dot's centred within the circle. [10:19] nic-doffay: isn't that whats happening now? [10:20] nic-doffay: ok.. its not totally centered, but thats because the image has some empty space around the circle [10:21] mzanetti, I must have missed that. [10:21] mzanetti, I just check the image, I don' see any empty space? [10:21] mzanetti: ooka, so just reapprove? [10:21] nic-doffay: hmm... right... same here [10:21] nic-doffay: let me check [10:22] tsdgeos: what? [10:22] mzanetti, adding the halfHeight of the dot to the x & y calculation does improve the positioning a bit, but it's still slightly off centre. [10:22] mzanetti: this one https://code.launchpad.net/~aacid/unity/raring_run_on_device/+merge/160577 [10:25] nic-doffay: pushed... now its exactly there [10:26] nic-doffay: I added the radius and subtracted the size of the elements [10:26] elements == dots [10:26] tsdgeos: done [10:26] mzanetti, added the radius to what? [10:26] nic-doffay: to the dot position [10:26] mzanetti, it's already being added, so you mean you added it twice? [10:26] nic-doffay: that was already there in my last commit... just saying what I changed in total [10:27] nic-doffay: no... not twice [10:27] nic-doffay: also the radius calculation had some "-units.gu(1.5)" which I removed [10:28] nic-doffay: anyways... just merge my branch again and you'll see [10:29] mzanetti, ah I see the implicit width. [10:29] mzanetti, before it was the infographicsWidth, but Saviq advised I removed it. [10:30] nic-doffay: yeah.. he seems to dislike height/width ever since he discovered implicitWidth/implicitHeight :D [10:30] I'll fix up the rest now, the rotations off again as a result. [10:30] nic-doffay: not totally agree in all cases with him there... anyways... it works [10:31] cheers for the help mzanetti [10:31] np === Guest5030 is now known as larsu === mmrazik|lunch is now known as mmrazik [10:58] mzanetti, one last thing, what's the best method to centre that circle with the anchors? Before I was setting both the horizontalCentre and VerticalCentre, but I'm sure there's a better way... [10:58] nic-doffay: centerIn [10:58] anchors.centerIn: parent [10:58] tsdgeos, mzanetti cheers. === MacSlow is now known as MacSlow|lunch [11:09] mzanetti: hmmm, how did https://code.launchpad.net/~mzanetti/unity/phablet-release-172/+merge/160578 get merging without autolanding job¿?¿ [11:10] fockin hell... my mouse cursor froze [11:10] can't click the link [11:11] narf... /me doesn't want to reboot now [11:11] lol [11:11] restart the wm? [11:12] tried kwin --replace. didn't help [11:12] this happens sometimes when switching between desktops [11:13] nope... no chance to get it back... bbiab [11:14] didrocks: ok, so, I have some HUD fixes in a merge proposal but I want to run one more test run on my guest session, as I lost some of the results due to a GPU hang [11:14] didrocks: https://code.launchpad.net/~sil2100/unity/autopilot_hud_2.0_fixes/+merge/160610 [11:15] sil2100: excellent! Should I let you coordinate that with ken? [11:15] didrocks: I'm waiting for Ted to appear, as I think there is a regression that needs to be fixed, and also consulting one of the strange things that HUD is doing [11:15] sil2100: the ">"? [11:15] didrocks: I'll also ping him once he's up [11:16] didrocks: not that, since that's a know thing that changed - it's about using an unicode EN SPACE instead of a regular ' ' space in the HUD entries [11:16] didrocks: it's \u2002 [11:16] sil2100: ah, interesting… :) [11:16] sil2100: yeah, better to check! [11:16] I go guesting again ;p brb [11:29] didrocks: sil2100: mediumtests are switched to raring [11:29] tsdgeos: ^ [11:29] goodie [11:30] \o/ [11:30] mzanetti: excellent! [11:30] Thanks :) [11:30] np [11:30] it was about time [11:30] but yeah... kept me busy till midnight yesterday [11:31] mzanetti: excellent news :) === alan_g is now known as alan_g|lunch [11:32] sil2100: didrocks: so please enable them again for the apps [11:34] mzanetti: yeah, do you have a ppa with AP for phablet? [11:34] mzanetti: as it's still different from AP for desktop (which is what we have in the -next ppa for now) [11:34] didrocks: ppa:autopilot/ppa [11:35] didrocks: btw... haven't been able to build/run the camera-app any more after your latest changes [11:37] mzanetti: my change? [11:37] * didrocks doubts the bootstrap commit did change anything [11:37] didrocks: let me rephrase. the changes you've been pushing involving the HUD stuff [11:38] mzanetti: something to check with sil2100, mterry or kenvandine who did the changes? [11:40] where do I get HUD 1.0 from? [11:40] lp:hud [11:40] tsdgeos: don't we have packages for that? [11:41] mzanetti: don't know :D [11:41] i guess there are some [11:41] somewhere [11:41] sil2100 ought to know where they are [11:43] mzanetti: ~ubuntu-unity/daily-build-next [11:44] didrocks: is this save to add? [11:45] mzanetti: it would be better to add the ~ubuntu-unity/next ppa, which is where stuff are copied once the packages are "certified" (built and autopilot running) [11:45] mzanetti: but as I understood, in addition to the HUD transition, AP is different on phablet… [11:45] so it will take the sprint to get everyone on the same page :/ [11:45] didrocks: just use autopilot-phablet everywhere [11:45] (it's amazing how PS can put themselves in difficult position forking everywhere their own projects…) [11:45] mzanetti: no [11:46] mzanetti: it doesn't work on desktop [11:46] didrocks: it does [11:46] mzanetti: ? thomi told that unity doesn't work with it [11:46] it does [11:46] didrocks: ah... old unity [11:46] dunno [11:46] mzanetti: well, current unity [11:46] you know, the thing we are shipping :) [11:46] in a real product [11:47] so yeah, making all those transitions, it's just making things painful [11:47] didrocks: anyways... autopilot-phablet does work on desktop... we use it to test all our apps on desktop and unity-next [11:48] mzanetti: right, but if I put that to the daily-build ppa, we won't have anymore tests passing for unity itself [11:48] mzanetti: so we need an interim solution [11:48] or forking even the package name and namespace [11:48] the package name is different [11:48] mzanetti: you mean, the QML HUD 1.0 ? [11:48] python-autopilot vs autopilot-phablet [11:49] sil2100: yes [11:49] mzanetti: but they do install files in the same path? if not, we can work on getting things on more quickly [11:49] mzanetti: the branch name is lp:libhud-qt while the package name is qtdeclarative5-hud1.0 [11:50] didrocks: yes, they conflict. can't install both at the same time. but do you need to do that? [11:52] sil2100: the package name should be qtdeclarative5-hud1.0-plugin I guess [11:52] sil2100: to be consistent with the other plugins [11:52] mzanetti: yeah, if we want to tests that apps can run in current unity and instrument both [11:53] sil2100: qtdeclarative5-hud1.0 depends on libhud-qt5 which is not in the ppa [12:11] MacSlow|lunch: so... it crashes on raring too. no idea why. I suggest adding debug prints between each of those lines to see where exactly it crashes [12:12] mzanetti: looking === MacSlow|lunch is now known as MacSlow [12:14] mzanetti, I'm on raring too... no crashes here [12:14] mzanetti, I'll mine the test print()s === mmrazik is now known as mmrazik|afk === _salem is now known as salem_ [12:57] tsdgeos: go away! === tsdgeos_ is now known as tsdgeos [13:04] tsdgeos: Is it possible to add ../../../../plugins/Unity/peoplepreviewdata into fake_unity_plugin.cpp?? === alan_g|lunch is now known as alan_g [13:09] paulliu: you mean in the CMakeListst.txt? === mmrazik|afk is now known as mmrazik [13:11] tsdgeos: No, in fake_unity_plugin.cpp. I need PeoplePreviewData. [13:11] dandrader: you up for a simple one? https://code.launchpad.net/~mzanetti/unity/phablet-fix-local-autopilot/+merge/160636 [13:11] paulliu: you mean the .cpp / .h file? [13:11] sure, why not? [13:12] tsdgeos: yeah, I mean qmlRegisterType(uri, 0, 1, "PeoplePreviewData"). [13:12] tsdgeos: Any side effects? I'm checking and I think it is ok. [13:13] paulliu: should be fine [13:13] tsdgeos: ok.. [13:14] tsdgeos: I'll modify it and push it.. [13:14] tsdgeos: I mean MR. [13:16] tsdgeos: you have a minute for this? https://code.launchpad.net/~mzanetti/unity/phablet-fix-local-autopilot/+merge/160636 [13:16] mzanetti, yes, in a couple of minutes [13:16] ah ok... [13:16] tsdgeos: don't mind [13:16] oki [13:17] tsdgeos: ah.. wait. It seems I cannot do that. In peoplepreviewdata.h it includes lens.h [13:17] dandrader: not urgend... I just didn't know if you will read the message at all [13:17] tsdgeos: So it needs to include fake_lens.h instead? [13:18] paulliu: remove the include [13:18] you don't need it [13:18] tsdgeos: hmm, in peoplepreviewdata.h it is needed. It uses Lens type. [13:19] it's not needed [13:19] a forward declaration will suffice [13:19] tsdgeos: ok.. [13:19] paulliu: you know what i mean? [13:20] tsdgeos: I don't know what you mean but I'm trying to remove it. [13:20] paulliu: let me do a quick MR [13:20] tsdgeos: ok [13:25] paulliu: https://code.launchpad.net/~aacid/unity/nolensinclude/+merge/160642 [13:27] though now that i see [13:27] that won't really help you [13:27] :D [13:27] silly me [13:28] tsdgeos: hmm. [13:28] still you should probably approve it [13:28] it's good to not include .h in .h unless it's strictly necessary [13:28] tsdgeos: yeah. === olli_ is now known as olli [13:33] tsdgeos: dandrader: standup [13:33] sil2100, do you know where I can find the package "qtdeclarative5-hud1.0"? [13:33] mzanetti: not really, hangout day is it today, no? [13:33] oh... [13:34] i'm there with dandrader and olli waiting for the rest of you [13:35] renato: it's in lp:libhud-qt, should be in the daily-build-next PPA [13:35] Cimi: https://plus.google.com/hangouts/_/5e9eec5b9e0f2e2903f09967025198df1acaf186 [13:36] sil2100, and where I can find: libhud1-dev?? [13:37] renato: also in the same PPA, it's from lp:hud [13:58] mzanetti: hi! Regarding that qtdeclarative5-hud1.0 libhud-qt5 dep you said... (sorry it took so long) - where do you see t his dependency? [13:59] sil2100: apt-get install qtdeclarative5-hud1.0 [13:59] sil2100: qtdeclarative5-hud1.0 : Depends: libhud-qt1 but it is not going to be installed [13:59] Ah, libhud-qt1 [13:59] kgunn: sorry, can you share the link of google doc again for the sprint? [14:00] sil2100: but its different now... [14:00] Strange, since it should be in the daily-build-next PPA [14:00]