=== duflu_ is now known as duflu [09:03] oh man [09:03] can't get the dash test to fail anymore :D [10:19] sil2100, when auto-landing a project to distro, how is the changelog created? [10:20] sil2100, does it take commit msg of each rev *unless* that commit also changed debian/changelog? [10:24] mhr3: yes [10:25] mhr3: if a commit changed/touched the changelog, then the commit message for that revision is not added [10:25] good to know, thx [10:31] tvoss, I vaguely remember hearing that Mir’s application awareness would fix bug 780776. Is that correct? And if so, how? [10:31] bug 780776 in unity (Ubuntu) "Launcher - No feedback when application launched elsewhere" [Medium,Triaged] https://launchpad.net/bugs/780776 [10:31] e.g. opening a document from the file manager, or a music player from the sound menu [10:32] mpt, right, although we haven't implemented user feedback, yet. But in theory, it would solve the issue [10:36] tvoss, is the feedback design specified somewhere? [10:37] mpt, I don't think so, surely not for the desktop use-case [10:38] tvoss, a job for katie then perhaps? [10:38] mpt, yup, I would think that you will be involved, too :) [12:13] Cimi: ping === RoddieKieley1 is now known as RoddieKieley [12:16] Cimi: looks like that test issue with ubuntu-settings-components is a qt bug. I've filed a bug with them, skipped the test for now and added a TODO for when it is fixed. [12:44] Mirv, I'm at a loss on how to proceed with https://bugs.launchpad.net/bugs/1258057 [12:44] Ubuntu bug 1258057 in Unity API "unity-api fails to build against Qt 5.2" [Critical,In progress] [12:44] Mirv, any suggestions? [12:50] dandrader: so qt5-beta2 PPA + git snapshot of qt declarative (and only qt declarative) works? then maybe the snapshot would be worth considering indeed. it just adds to the uncertainty, but if we already know there's a component that's broken in a sense (qt declarative), then I guess it wouldn't do more harm. [12:50] and yes it's a stable branch, but sometimes indeed the minor stable releases aren't that minor either [12:51] Mirv, yes (assuming qt5-beta2 PPA is essentially git tag v5.2.0) [12:52] Mirv, problem is: 1 - there isn't a single commit to cherry-pick that fixes it. we need to bring in a number of commits [12:52] Mirv, 2 - it fixes the testLauncher crash but causes testApplication to fail [12:54] Mirv, so, taking in the latest qtdeclarative stable branch improves our situation (moving from a crash to a failure) but doesn't completely solve it [12:55] Mirv, to a course of action could be taking qtdeclarative stable branch an filing a qtbug on that failure, which looks like a regression regarding qobject enums not visible on the qml side [12:56] s/to a/so a [12:56] but then, we don't know what other problems or regressions might come along that newer qtdeclarative, as you said... [12:59] dandrader: yeah or it's still ~rc since I'm rebuilding all final releases in qt5-daily at the moment. but yes I'd consider failure better than a crasher, so it's at least worth considering, but I'll concentrate on rebuilding all of final releases now first (and rebuilding everything against them, then) [12:59] there's of course a chance that unity-api would behave slightly differently also with the final release vs. the release candidate in rc2, so worth checking first [13:00] hi ho === dandrader is now known as dandrader|afk === dednick is now known as dednick|lunch === dandrader|afk is now known as dandrader [14:31] dednick|lunch: standup? [14:31] mzanetti: ↑↑ [14:32] Saviq, ping [14:32] tedg: sorry... trashed my PC [14:32] will try to join asap [14:33] tsdgeos: ^^ [14:33] ok [14:33] tvoss: he's out today [14:33] tsdgeos, ack and thx [14:36] mzanetti: we're done i guess you can fill in your notes and just shout here if it's something that you need us to act on [14:37] doh. missed it === dednick|lunch is now known as dednick [14:38] dednick: same applies to you :-) [14:38] tsdgeos: yep. wil do [14:40] tsdgeos: yeah... nothing really. I've just been working on the right edge with unfortunately very little progress so far today. [14:40] ok :-) [14:40] have been following the wrong approach and had to go back one step [14:44] Is Cimi not in today? === dandrader is now known as dandrader|lunch === Saviq_ is now known as Saviq === balloons_ is now known as balloons === Saviq_ is now known as Saviq === Daviey_ is now known as Daviey === dandrader_ is now known as dandrader === balloons_ is now known as balloons === jhodapp_ is now known as jhodapp [16:45] anybody? [16:45] how to reinstall unity after installing gnome. [16:45] ? [16:46] tsdgeos, so VerticalJournal.horizontalSpacing is indeed different than Grid.columnSpacing? [16:46] is the horizontal spacing [16:47] both between columns and between the borders of the item [16:47] being a minimum bound at the right side of course [16:47] tsdgeos, that must be documented [16:48] what other thing can it be? [16:48] tsdgeos, I would expect it to be like Grid.columnSpacing [16:48] well, but it's not called columnSpacing :D [16:49] tsdgeos, so your horizontalSpacing is like a margin [16:49] dandrader: so you'd prefer stuff to start x=0 instead of x=horizontalSpacing [16:49] i understand? [16:50] tsdgeos, I prefer to have the meaning of that property documented to avoid ambiguities and misunderstandings [16:50] that's not the answer i want :D [16:50] i am asking API wise [16:51] do you think it makes more sense for it to be what i say or not [16:51] we're trying to make this good [16:51] so i'm asking what you think makes more sense [16:51] tsdgeos, I would prefer s/horizontalSpacing/columnSpacing [16:51] tsdgeos, that margin can be done via other means [16:52] tsdgeos, yield more flexibility to the layout [16:52] tsdgeos, you use Item's native margin properties to achieve that spacing between the edge columns and the item borders [16:54] tsdgeos, and also mimicking an existing API (Qt's Grid) has the added benefit of having the overall API more uniform and fit better together [16:54] tsdgeos, A developer used to work with Grid will have a much lower learning curve if his existing API expectations can be directly transfered when working with VerticalJournal [16:58] dandrader: ok, can you please add that comment to the review? [16:58] will need to change it for vSpacing too [16:59] tsdgeos, ok === ken_ is now known as kenvandine [17:05] thanks :-) === gatis is now known as gatispaeglis [17:08] dandrader: as for the tryFoo [17:08] we dont' support what you envision :D [17:09] you can't remove items from the middle [17:09] this only works with static models [17:09] you can make it grow [17:09] because that's kind of static [17:09] but that's all [17:10] or shrink from the end [17:13] dandrader: that ok? [17:15] tsdgeos, didn't get the "static model" explanation but, anyway, you can still push and pop (like in a stack), right? [17:16] tsdgeos, if so, the tryVerticalJournal still makes sense [17:16] dandrader: basically we don't support removing from the model [17:16] errr [17:16] model/middle [17:16] you can add stuff at the end and remove from the end [17:17] and that's it [17:17] but sure i am not saying it does not make sense [17:17] i'm just setting your expectations [17:17] that said [17:17] EOD-time [17:17] tty tomorro [17:17] w === hggdh_ is now known as hggdh === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === fginther` is now known as fginther [20:49] greyback, Hey, so doing more timing tests. [20:49] greyback, It seems that the observer for the starting signal is taking 63ms inside the unity codebase (dbus overhead stripped) [20:50] greyback, Is that expected? Seems like a while. [21:01] veebers: hey, hope you're well...just wanted to point out...this one is makin' the news [21:01] https://bugs.launchpad.net/unity8/+bug/1260860 [21:01] Ubuntu bug 1260860 in Unity 8 "some autopilot tests don't use process_helpers" [High,Triaged] [21:02] causing random failures on maguro...can you treat with priority ? [21:05] Hi kgunn, I'm currently on vacation so time in front of my laptop is spotty. I should be able to look at this morning if it's priority though. Should be easy enough to fix for any dev though [21:05] veebers: ack, are you off for christmas ? [21:06] kgunn: aye, off until Jan 6th === slangase` is now known as slangasek