[08:08] tsdgeos: FYI I'm running AP:s on the qtbase ubuntu8, and it's looking good/better now [08:08] :) [08:08] meanwhile it's been approved upstream [08:08] still not merged because their CI basically sucks [08:09] well not their CI [08:09] it's just that their tests are unstable as hell [08:10] yeah, enabling the unit tests has been troublesome for our builds too, but at least we nowadays have them mostly enabled [08:10] tsdgeos: so do I already have +1 from you on that or shall you test it today? [08:11] i tested it with my self built thing, i can give it a try with the proper pacakges [08:11] same silo? [08:11] tsdgeos: yes please try the PPA, 027, check that you get ubuntu8 [08:11] oki [08:32] tsdgeos: not necessarily all perfect, I at least need reruns and a moment ago I had security code being asked which sounds like unity8 restart [08:32] UITK passed everything though, while before it barely get through a couple of tests [08:36] i see [08:36] wonder if that restart/reruns is something we already had without this patchset though [08:37] Mirv: i had never run uitk tests before, what's your opinion? [08:45] tsdgeos: not during UITK tests, and like I said I didn't get them there but later (I think during calculator AP). the way I'm running all AP suites in one pass causes all kinds of random trouble, so this might be it. [08:45] i see [08:45] I'm mostly worried about a lot of ubuntu-system-settings AP failures + that security code screen appearing (I think that's uncommon), but I'll see when I get to do reruns of the failed suites in about 2h when this pass finishes [08:46] usually the fact that whole UITK suite passes without any problems is a good indication, so let's hope there's nothing reproducible. [08:53] Mirv: one good thing to do [08:53] is phablet-hsell [08:53] and then [08:53] sudo gdb attach `pidof unity8` [08:54] this way if unity8 crashes it'll get stuck there [08:54] and you can later on bt and all that [09:36] tsdgeos: there was a crash, and I remembered the whoopsie log, so here's the crash https://errors.ubuntu.com/oops/faf1d51c-76d8-11e4-966f-fa163e75317b [09:38] Mirv: not much of a backtrace there :/ doesn't seem dbus related, but you can never tell if it's because of the patches or not [09:40] tsdgeos: well there's _g_dbus_auth_run_client in the trace, but not really anything Qt [09:41] anyhow, since I had the test run stuck in mir_screencast (which is common...) I'll start the rerun now omitting those that had full pass. [09:41] yeah here it gets stuck there too [09:42] i was wondering if it was common or not === vrruiz_ is now known as rvr === karni is now known as karni-otp === karni-otp is now known as karni [11:02] Mirv: been testing it, could not find any anomally, so +1 for the patchset from my side === _salem is now known as salem_ [11:42] tsdgeos: thanks. I'll need to do this AP testing for some while still... and probably need to compare with/without PPA until there's some sort of consistency. [12:17] Saviq, is ./run.sh still being used nowadays? [12:18] dandrader, sure, although it's not been fixed for split dash === alan_g is now known as alan_g|lunch === MacSlow is now known as MacSlow|lunch [12:32] Saviq: have chance for second pass? https://code.launchpad.net/~mir-team/qtmir/cmake/+merge/235239 [12:37] greyback, yeah, although I hate cmake today (been refactoring QmlTest.cmake for a long time now) ;P [12:38] Saviq: I don't have a lot of love for it either [12:38] if you'd heard my cursing when you poked your head in and started commenting... :D [12:39] greyback, it's not like I was picky (yet) :P [12:40] * greyback thinks qmake ain't so bad now === karni is now known as karni-lunch === alan_g|lunch is now known as alan_g [13:19] greyback, none of the inline comments are addressed in there [13:23] Saviq: damn sorry, I pushed to wrong branch [13:23] updates are there now [13:23] greyback, nw [13:24] greyback, I'll move it to Needs Review to get ci feedback [13:24] ok [13:26] greyback, hmm, shouldn't Unity.Application remain in the shell-specific QML import dir? [13:26] Saviq: it never was there [13:27] greyback, ah indeed, we should probably move it there then [13:27] Saviq: it's not necessarily unity specific [13:27] I'm considering renaming it [13:28] greyback, hmm ok, maybe we'd need another custom import path, not sure it's good in the default one [13:29] greyback, ah, and there's a running effort to rename qtdeclarative5-foo [13:30] yep that too === karni-lunch is now known as karni === MacSlow|lunch is now known as MacSlow === dandrader is now known as dandrader|afk [14:07] Cimi, what's the story with the wizard branches? [14:08] mterry, thought you were on holiday :D [14:08] Cimi, was yesterday and was sick the day before [14:08] Cimi, but I'm back! [14:08] ah ok, let me test... thought I could do that later in the day for monday [14:20] Cimi: there? [14:20] tsdgeos, yes [14:20] Cimi: have you shown https://code.launchpad.net/~cimi/unity8/elastic-not-close-spread/+merge/242949 to someone in design? [14:20] tsdgeos, yes, approved [14:20] it seems very similar to what we have now [14:21] tsdgeos, what do we have "now" ? [14:21] well i never drag that much when closing an app [14:21] i just flick [14:21] tsdgeos, well, probably people who tested did... [14:22] and the flick back we have now and this one are the same [14:22] tsdgeos, yes but this limits the movement [14:22] but ok, i guess it's still and improvement [14:22] tsdgeos, vesa tried the branch and approved [14:22] tsdgeos, he even asked me to tweak a value :/ [14:22] he could have written so in the thing, saved us this trouble :D === dandrader_ is now known as dandrader [14:24] greyback, any idea what caused bug #1396244 btw? it's fine on rtm still... [14:24] bug 1396244 in unity8 (Ubuntu) "Unlocking the phone over dialer is very slow" [High,Triaged] https://launchpad.net/bugs/1396244 [14:25] greyback, we should try and make sure to not land what caused this [14:29] Saviq: can get it with camera too. Seems fullscreen + lock causes the slowdown. [14:29] greyback, yeah, any idea of a change that could've introduced this? [14:30] no idea [14:31] ok, I'll have to remember to look out for this when landing in RTM [14:32] would be worth bisecting [14:33] *could* be that when screen locked, panel is brought in, and app is told to resize its surface to non-fullscreen - which would be wrong. And then on unlock, it gets a flood of resize events [14:33] just a theory tho [14:34] * greyback runs to post office [14:44] Cimi: any chance you'll review https://code.launchpad.net/~aacid/unity8/photoscopeimprovements/+merge/239834 ? [14:45] tsdgeos, I think is fine to go [14:45] tsdgeos, as said, only bugs I had we agreed they were unrelated [14:45] tsdgeos, I'll put it in a silo for vivid today, Monday latest [14:48] Saviq: oki [14:50] dednick: you have conflicts in https://code.launchpad.net/~nick-dedekind/unity8/lp1385331.led/+merge/241417 [15:03] mterry, wizard page after wifi seems broken here, where is upstart log? [15:08] mterry, wizard page after wifi seems broken here, where is upstart log? [15:09] Cimi, hmm that would be the location page -- upstart is ~/.cache/upstart/unity8.log [15:09] Cimi, it doesn't show up? Just looks blank? [15:09] mterry, I see just wallapper [15:09] Cimi, yeah hmm. On which device? I'll try to reproduce/fix [15:09] mterry, mako [15:10] Cimi, ok -- it just means the wizard is waiting for an answer to the question "is HERE installed" before showing the location page or not [15:10] mterry, thought about here [15:10] mterry, maybe I can install it and see if it goes after [15:11] Cimi, well we need to work in the mako/no-here case though :) [15:11] mterry, sure, was just to confirm... [15:11] Cimi, you have the latest debs? I changed how we look for HERE to avoid needing the special " " value -- might have broken something [15:12] mterry, the last on -test [15:19] Cimi, hm yeah I see same thing, will let you know when I've fixed [15:19] mterry, thanks [15:31] Saviq: re those cmake errors, I've no idea what's causing them. Have you seen them before? Was suspecting a qt5 module but have zero evidence for it [15:51] Saviq, reasonable suggestion re: cmake-extras, will look into today [15:55] alesage, there's actually a branch already, but it's a bit naïve, just s/2.8/3.0/ [15:55] Hi folks. Who can I talk to about the badge implementation? [15:55] karni, define "the badge" [15:55] greyback, no, have not seen them yet [15:56] Saviq: the number that shows up on the app icon in the launcher at the left edge [15:56] Saviq, ack, I possibly need Satoris' advice on this [15:57] karni, ah, that one, you can talk to me since mzanetti's only pretending to be around [15:57] karni, "count emblem" is what we call it I think ;) [15:57] haha, all right. [15:58] Saviq: let us file a bug, I'll contact you shortly. basically, it seems to be adding the value to the badge, instead of setting the value of it. if we go it wrong and it's documented, then I'll blame myself. [15:59] Saviq: we pass the "unread messages count" in there, but it seems it's adding them app, is this expected? [15:59] karni, you're talking about push notifications here are you? [15:59] Saviq: yessir [15:59] karni, in that case you'd have to talk to Chipaca [15:59] Saviq: okay :) WFM [15:59] karni, we're setting the value as-is [16:00] gotcha, good news for me [16:00] thanks Michał :) [16:04] greyback, yeah, looks like Qt5Core is the culprit [16:05] greyback, also, just found... [16:05] ""The qt5_use_modules macro is obsolete. Use target_link_libraries with IMPORTED targets instead." [16:05] oh [16:06] ok fixing [16:06] greyback, sorries [16:06] np [16:06] greyback, but it looks like you still don't need explicit INCLUDE_DIRS then [16:07] Saviq: you do for private header dirs [16:07] greyback, yeah, that you do of course [16:13] shouldn't http://paste.ubuntu.com/9284813/ make going from one scope to the other slower?¿ [16:14] because it doesn't === dandrader is now known as dandrader|lunch [16:47] tsdgeos, no [16:47] tsdgeos, we don't have a highlight [16:48] greyback, cmake done, was a bit more picky this time [16:48] Saviq: oh gee great thanks [16:49] Saviq: well, hmmm, right, damned code that does nothing and confuses me :D [16:49] greyback, I thought you'll like it on a Fri afternoon [16:49] tsdgeos, trueth [16:50] *you'd [16:50] Saviq: though we kind of have highligh [16:50] highlightRangeMode: ListView.StrictlyEnforceRange is what makes the current index chane [16:50] change [16:51] tsdgeos, right, but we don't have a highlight object, really [16:51] right tight [16:51] tsdgeos, but I get what you mean [16:51] tsdgeos, we're abusing the ListView there a bit... and we have no control over the swipe threshold, which I was asked about recently [16:52] we might need to stop abusing ListViews for PageLists :P [16:52] Saviq: nor about the snap speed for what i can see [16:52] tsdgeos, nope [16:52] meh :/ [16:52] i *think* the snapping animation is borked going back/front in some cases [16:53] but i can't "see" it, thus i wanted to make it go slower [16:53] QUnifiedTimer::instance()->setSlowMode or something like that [16:55] ok I think I've enough cmake for one night and one day... [16:58] he he [16:59] *almost* there [17:00] everything was fine, and then someone added dbus-launch to their test, *really*, I ask you!? [17:00] and then there's dbus-test-runner, even worse [17:01] yep [17:03] have a good weekend, y'all! o/ [17:04] yeah i'm leaving too [17:04] off to a cabin [17:04] the weekend of the year with most predicted rain [17:04] \o/ [17:04] :D [17:04] :/ [17:04] ahah [17:04] tsdgeos, where u heading? [17:04] Besalú [17:05] nice romanic vilalge [17:05] http://upload.wikimedia.org/wikipedia/commons/a/a5/Pont_Medieval_%28Besal%C3%BA%29_-_7.jpg [17:05] tsdgeos, looks lovely on google (with sun) [17:05] yeah ^_^ [17:05] anyway, tty on monday! [17:06] aloha [17:10] I was there! [17:11] ooh funky new search dropdown in Firefox, maybe useful === dandrader|lunch is now known as dandrader === alan_g is now known as alan_g|EOW [19:12] Oh no... seems I lost all my search engines in Firefox? [19:12] hmm... they're are "hidden" === salem_ is now known as _salem