=== alan_g|EOD is now known as alan_g [08:28] is autonalding not running for unity-mir ? === mpt_ is now known as mpt [08:32] anyone for this simple improvement? https://code.launchpad.net/~aacid/unity8/lvwph_culling_off_by_one/+merge/178689 === alan_g is now known as alan_g|tea === alan_g|tea is now known as alan_g [09:05] hi guys! [09:16] Saviq, hey dude, did you fix the connection? [09:24] greyback: is autonalding not running for unity-mir ? [09:25] tsdgeos: not yet no [09:25] ah ok [09:31] greyback: i thought it was and was wondering why https://code.launchpad.net/~aacid/unity-mir/improve_input_area_handling hadn't been merged :D [09:33] tsdgeos: ah damn, I thought I pushed it [09:33] maybe you did [09:33] no I didn't [09:33] and you did not mark it as merged [09:35] LP detects if it is merged. I've never marked a MR as "Merged" myself [09:36] tsdgeos: landed now, sorry for the delay [09:38] ah [09:40] meh [09:40] what's up with the autopilot tests :-/ [09:40] change one line in LVWPH get all the autopilot tests that don't use LVWPH at all to fail [09:40] :'-( [09:43] not good. Maybe some state not being reset? [09:46] not sure [09:46] MacSlow: you were complaining your autopilot stuff regressed too, right? [10:06] tsdgeos, yep... [10:06] tsdgeos, seems like a timing issue [10:06] I guess [10:26] tsdgeos, have a look at my last comments https://code.launchpad.net/~macslow/unity8/notification-autopilot-tests-dbus/+merge/177780 [10:27] oh it's a different issue than with me [10:27] i had all the tests failing [10:27] tsdgeos, I'm not chasing that down as I need to focus on the extended snap-decision with unitymenumodel [10:36] MacSlow, I just restarted CI for that MP [10:39] Saviq, ok.. *fingers.crossed* :) === MacSlow is now known as MacSlow|lunch [11:09] anyone else get crash when running the qml tests? In MockLauncherModel deconstructor? [11:25] qmluitests? [11:26] or? [11:34] tsdgeos: qmluitests yes === alan_g is now known as alan_g|lunch [11:36] greyback: hmmm, any particular make testFoo i should run to see if it works? [11:36] tsdgeos: nothing special, I just ran "make qmltests" from inside the builddir === alan_g|lunch is now known as alan_g [11:38] seem to work here [11:38] lunch, i'll check more carefully later [11:44] back in a bit, need to log out/in === alan_g is now known as alan_g|lunch === tvoss_ is now known as tvoss|lunch === MacSlow|lunch is now known as MacSlow [12:18] MacSlow, looks like the failing test is indeed autopilot's / introspection acting up [12:19] Cimi, just found your ping in my history - yeah, conn is fine now [12:20] Saviq, *sigh* yeah... I'm not chasing that as I'm busy with getting the SimpleAction stuff working in my proof-of-concept "extended snap-decision" [12:20] MacSlow, yeah, I will take care of it [12:20] Saviq, thanks for looking into it === hikiko is now known as hikiko|lunch === ritz_ is now known as ritz|zzz [12:42] * Saviq is melting.. [12:47] Saviq: come to Ireland! It's a far more productive 18 degrees here [12:47] greyback, good, you're at the top of your abilities, then! ;P [12:47] * Saviq is definitely not [12:47] greyback: actually yes, segfaults [12:48] Saviq: after my holiday, you'd expect that, yeah... :) [12:48] make testShell [12:48] boom [12:48] tsdgeos: ok, it's not just me. deleteLater() being called on something that's already deleted? [12:48] tsdgeos, on exit? [12:48] Saviq: yep [12:49] tsdgeos, yeah, got the same [12:50] core is useless [12:50] lol /me got his background in unity8 all of a sudden :D [12:51] ahhh [12:51] millions of Conditional jump or move depends on uninitialised value(s) [12:51] in valgirnd [12:51] tsdgeos, looks like one of the mock plugins, though? as the real unity8 doesn't crash [12:51] the warnings are DDA [12:52] nothing big i guess [12:52] not related to the crash [12:52] :-( [12:52] * tsdgeos can't unlock the greeter when running on valgrind [12:52] something is probably "too slow" === _salem is now known as salem_ [12:55] Saviq: greyback: same bt? http://paste.ubuntu.com/5954939/ [12:55] tsdgeos, mine was ?? [12:55] tsdgeos, all over [12:55] ah :D [12:55] tsdgeos: yep [12:55] i'm trying to get one with my self compiled one [12:55] it's a bit harder since i need to coherce it to get the ubuntu.components [12:56] tsdgeos, http://pastebin.ubuntu.com/5954944/ ;) [12:56] :D [12:56] ah, it's not the components [12:57] it's that LVWWPH is compiled with the other qt and doesn't find the symbols :-/ [12:59] Saviq, https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1208833 [12:59] Ubuntu bug 1208833 in Ubuntu UI Toolkit "popover triangle is misplaced" [Undecided,New] [12:59] Cimi, thanks [12:59] my internet here is super unreliable [13:00] Cimi, confirmed === hikiko|lunch is now known as hikiko [13:10] greyback: could you shoot the link you shared yesterday about the upstart monitoring ? (pseudo code i think) === alan_g|lunch is now known as alan_g [13:23] Saviq: greyback: all: quick one for the uninitialzd values https://code.launchpad.net/~aacid/unity8/fixuninitializedvalues/+merge/178748 [13:26] Saviq, is there any way we could start using real apps scope but have the shell still provide the running apps? [13:26] mhr3, sure, we're doing that already [13:27] since i think it'll take a bit longer while the mir appmanager bits are there? [13:27] Saviq, well... yes, but it's a mess now :) [13:28] mhr3, not sure how we could un-mess it, though [13:28] mhr3, one thing we could do, is add a means to supply an "override" to Categories [13:29] exactly what i was thinking :) [13:29] mhr3, yeah, that'd definitely be cleaner [13:30] mhr3, should be easy - just tell Categories that for index x use y as results model [13:30] mhr3, I'm all for that [13:30] i just really really want us to start using real scopes, there's always something that's missing to take the leap [13:30] and this would be really nice [13:30] mhr3, we could use the same for Home [13:30] mhr3, +1 [13:31] now, who'll do that? my qmlfu isn't on that level :P [13:32] kgunn: this what you mean: http://bazaar.launchpad.net/~ted/upstart-app-launch/libupstart-flesh/view/head:/libupstart-app-launch/upstart-app-launch.h [13:32] greyback: ah...that's the one [13:32] kgunn: that's the proposed API for tedg's upstart library, which will do the hard work for shell in launching, quitting, and monitoring apps [13:33] greyback: got it...we call those from shell...and rest is tedg magic [13:33] ? [13:33] kgunn: exactamundo [13:34] * tedg breaks out his wand [13:34] tedg: any guesses as to when that might land ? [13:34] Really Upstart does the heavy lifting of tracking processes, etc. [13:34] ...workin' on our app life cycle story [13:34] kgunn, Uhm, Colin sent me an e-mail that the click stuff I needed should be in this morning. [13:35] I think the biggest thing we're waiting on is the MIR for Click. [13:35] Though I need to update to a couple changes he made. [13:36] So, I guess the question is more: when do you need it? If it's soon, we can push those things. If not, they'll probably happen naturally this week. [13:36] Saviq, wait, did you mean to have that override in the unity plugin's Categories class itself? [13:37] mhr3, yeah, that's where getResults is, and creates the DeeFilterModel, right? [13:37] indeed [13:37] mhr3, so instead of creating the DeeFilterModel, it'd return a model provided [13:37] hmm... why not [13:37] thx, i'll try that then [13:38] tsdgeos: we've switched to Qt5.1 in CI, right? [13:38] hmmm [13:38] no? [13:39] greyback, no [13:39] Saviq: ok [13:39] greyback, we'll only switch when 5.1.1's ready [13:39] greyback, at least that's the plan now [13:39] Saviq: good to know, thanks === alan_g_ is now known as alan_g === alan_g is now known as alan_g|afk === tvoss|lunch is now known as tvoss|hungry === tvoss|hungry is now known as tvoss_ [13:52] greyback, lp:~unity-team/unity8/unity8-integrate-mir-obsolete - can be deleted? [13:52] Saviq: yep [13:53] Wellark, lp:~kaijanmaki/unity8/launcher-backend - is this branch active? === alan_g|afk is now known as alan_g [13:55] Saviq: so we have in real repos all that was in unity next? [13:55] and all if it is daily released? [13:55] tsdgeos, yes, unity8 is in distro now [13:55] tsdgeos, so are all our dependencies [13:55] mterry, lp:~mterry/unity8/unity8-mir-update - is this branch active? [13:55] Saviq: yeah i know somehow unity8 installed itself here (or maybe i did) [13:56] ;) [13:56] Saviq: cool we can kill the ppa then, i'll approve your branch [13:56] pstolowski, lp:~stolowski/unity8/rating-filter - can you delete this branch if inactive? [13:56] Saviq, let me see... I don't remember that branch name [13:56] mterry, it's ~mterry ;) [13:56] Saviq, I've been framed! [13:57] tedg: sorry, got distracted...actually we would love that support asap if its as easy as asking for it [13:57] tedg: we were kind of shooting for being in touch trunk by next week [13:58] Saviq, oh... that was for testing an update of the unity8-mir branch from trunk. It's not active [13:58] mterry, delete then, please [13:59] kgunn, Okay, and trunk means in distro now, right? [13:59] So we need to solve the MIR issue as well by then. [13:59] tedg: yeah [13:59] Saviq: ok, done [14:00] pstolowski, cheers [14:00] kgunn, I'll ping Colin on that one again, I think that's the long pole here. I know it's on his TODO list, but I'm not sure where exactly. [14:00] tedg: cool....let me know it helps if i whine :) [14:01] kgunn, wine always helps ;-) [14:02] tedg: what's kinda the best guess on all that concluding & the interface being available.... [14:02] kgunn, I'm thinking end of this week. [14:02] * kgunn prefers reality :) [14:03] tedg: thanks and thanks for chasing Colin [14:03] greyback: ^ [14:03] It's mostly ironing out details at this point. There's no significant work left. [14:04] kgunn: noted, thanks [14:05] Saviq: yes, it is [14:05] Wellark, ok thanks, just doing some housekeeping :) [14:05] tedg: hi! I need to talk with you later today :) [14:05] * Saviq lost gtk-window-decorator again [14:07] grr since when restarting the window manager results in apps being killed? :| [14:08] yikes my prompt is colorful [14:08] Wellark, Ah, cool. What's up? [14:09] tedg: hud bug :) [14:09] Wellark, IMPOSSIBLE! [14:09] we are probably missing a couple of source_changed() somewhere :9 [14:10] Hmm, okay. [14:10] bregma, Did you guys ever get to look in switching to libhud-client? What was the conclusion there? [14:10] and also I'm not sure if I fixed (can't remember really..) the problem with queries becoming broken when hud quits [14:11] kgunn, ping [14:11] Saviq: pong-ish === _salem is now known as salem_ [14:20] dednick, Were you able to find the Upstart signal for indicators starting code? I don't remember where we left that. [14:22] tedg: i havent had chance to look into it [14:24] tedg: i don't even know if there ever was a trigger on phablet. none of the indicators used upstart. [14:24] dednick, There's no trigger needed really. You can just emit random events even if no one is listening. [14:25] tedg: sure, but you need to emit it from somewhere. [14:25] dednick, We now have libupstart, so it'd probably make more sense to use that. Let me do a quick patch there. [14:25] tedg: ok [14:30] tedg, we have neither the expertise nor the time to do that work [14:30] we're overloaded as it is [14:31] hmm Cimi, something's wrong with the background - see https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-saucy/1641/?#showFailuresLink [14:31] Cimi, all the videos show black background [14:31] Cimi, and no of the tests pass 'cause the greeter waits for the image to be ready before treating itself as ready [14:32] *none [14:32] Cimi, there's been a bunch like that already [14:34] bregma, okay, do you think it'll be possible for 14.04? I'd really like to drop the legacy interface for the LTS. [14:35] tedg, hard to say, depends on how many more of my guys get stolen for other projects [14:35] we always welcome merge proposals [14:36] Saviq: indicator listview delegates arent getting notified to delete themselves (deleteLater is being lost). :/ [14:36] they are getting their Component.onDestroy called, but no delete. weird. [14:36] bregma, heh, yes. Let's discuss at UDS. I'd really like it to make the list. [14:37] dednick, interesting [14:37] Perhaps I can convince olli_ to not let your guys get stolen :-) [14:40] Cimi, yeah, the background fallback isn't working [14:44] #boredofbuildingqt5source [14:47] Cimi, with http://pastebin.ubuntu.com/5955258/ testShell shouldn't pass, why does it? [14:49] greyback, kgunn, FYI: https://bugs.launchpad.net/ubuntu/+source/click/+bug/1208800 [14:49] Launchpad bug 1208800 in click (Ubuntu) "[MIR] click" [Undecided,New] [14:51] dednick: get moar cores! [14:51] dednick: anything i can try for you? [14:51] i've the stable branch compiled here [14:52] tsdgeos: thanks, but it's something pretty specific i'm trying to debug. [14:52] Cimi, bug #1208868 is Critical for you [14:52] tsdgeos: my laptop melts is i use 4 cores... [14:52] bug 1208868 in Unity 8 "Background fallback isn't working" [Critical,Triaged] https://launchpad.net/bugs/1208868 [14:52] dednick: :-/ [14:52] * tsdgeos copiles with -j10 [14:52] gets a bit hot sometimes but never has rebooted or anything [14:53] * dednick has evny [14:53] dednick: what are you building stable branch or? [14:53] libqt5core5 [14:54] from our packages? or from git? [14:54] tsdgeos: packages [14:54] ah, ok [14:54] i'd suggest you to try 5.1, everything is much better in there :D [14:55] didrocks, our automated C-I merge is not working due to armhf issues, do you think https://code.launchpad.net/~jbicha/ubiquity/rename-gtk-desktop/+merge/178149 is OK for a manual merge? [14:55] tsdgeos: yeah, i was considering that. [14:55] bregma: if it's ubiquity, ask xnox, but the change looks fine [14:56] bregma, how is a manual merge going to help you? [14:56] bregma, the package is going to fail to build in the archive anyway [14:56] bregma, rsalveti is working on fixing libhybris so hopefully we get stuff back in shape by eod [14:57] * xnox agrees with seb128. arhf FTBFS, will get manual merge stuck in saucy-proposed, and CD are only built from release pocket. [14:57] ah, right, I was thinking the armhf fail was because of moving to the new hardware (the timing was the same), my mind is just catching up with my brain [14:58] sorry for the noise, I'll go back to sleep [14:58] xnox, next time ship a compat symlink the time for users to be ported [14:58] xnox, you can still do that today btw [15:00] seb128: meh, it's once in a lifetime change =) [15:11] whee.. I just installed saucy on my brand new Dell XPS13 just to see compiz crash on the first boot ;( [15:13] oh, wait what.. apport tells me that I have old packages installed even though the installer was supposed to do an update.. [15:20] greyback, tsdgeos https://code.launchpad.net/~saviq/unity8/revert-background/+merge/178782 please [15:20] Cimi, reverting background, there's work still to be done there [15:22] Saviq: waiting for launchpad to catch up :D [15:23] tsdgeos, yeah, and upstream merger, I'd imagine [15:23] that too :D [15:23] tsdgeos, but you can easily test locally [15:23] tsdgeos, just set your background in dconf-editor (org.gnome.desktop.background) to something invalid (like an empty string) [15:23] tsdgeos, ap tests will fail [15:24] tsdgeos, there's a few other bugs around that that I'm filing right now [15:24] * seb128 shakes fist at Saviq [15:24] seb128, ;) [15:25] :p [15:26] Saviq: you mean picture-uri in there? [15:26] tsdgeos, yes [15:28] Saviq: big question is, if that's upsetting autopilot how did it get merged? :S [15:28] tsdgeos, indeed [15:32] tsdgeos: Saviq: tests passing here, so I'm ok to revert [15:33] +1 [15:52] Saviq, having problems creating an animation in a list delegate. "cannot apply object to list". Any suggested workarounds? [15:52] mzanetti, ^ === jasoncwarner__ is now known as jasoncwarner [15:53] nic-doffay, mzanetti is away [15:54] nic-doffay, what do you mean "create an animation in a list delegate"? [15:55] Saviq, eg https://pastebin.canonical.com/95503/ [15:58] nic-doffay, nothing wrong with that, I don't think your error is related to the animation [16:00] nic-doffay, your problem relates to the fact that ListItem.Standard can't have children [16:01] Saviq, it can't have children? [16:01] Children existed before this animation. [16:01] And do now? [16:03] nic-doffay, just use "data: [ UbuntuNumberAnimation { } ]" [16:03] nic-doffay, otherwise it tries to assign a non-visual object to children of ListItem.Standard and fails [16:03] nic-doffay, see http://qt-project.org/doc/qt-5.0/qtquick/qml-qtquick2-item.html#data-prop [16:04] nic-doffay, hmm or wait, that might break things [16:05] nic-doffay, it might be a bug in the SDK, though [16:06] nic-doffay, ah! bug #1190509 [16:06] bug 1190509 in Ubuntu UI Toolkit "[ListItems] Default property being "children" brings some inconvenience" [Medium,Confirmed] https://launchpad.net/bugs/1190509 [16:09] Saviq, ah [16:10] Saviq, how do I then access the animation to start it from resources? [16:10] * tsdgeos waves until tomorrow [16:15] nic-doffay, id: whatever, the usual [16:17] Saviq, doesn't seem to be working. [16:18] Saviq, will do more digging thanks for the help with the resources though. === greyback_ is now known as greyback|food [16:31] Saviq, I was on holiday this afternoon but got time to look at it now... [16:34] Cimi, don't, we've reverted - you'll look into it tomorrow [16:34] Saviq, ok tomo morning [16:35] nic-doffay, it might be a side-effect of the fact that "resources" are slightly special [16:35] nic-doffay, another way would be to go "property NumberAnimation: UbuntuNumberAnimation { }" === greyback|food is now known as greyback [18:15] * greyback eod