[04:57] morning [09:32] sil2100: hey! [09:32] sil2100: there is a new failure, but just one and on the 3 architectures! [09:33] unity.tests.test_hud.HudBehaviorTests.test_alt_arrow_keys_not_eaten [09:33] jibel: FYI ^ [09:33] we were almost there, snif! [09:34] \o/ [09:34] sil2100: I was really believing we would release now :p [09:34] didrocks: ok, let me see that then, looks like a more random failure, but maybe we can make it better not to fail [09:34] at least the next/prev are fixed! [09:34] sil2100: the weird thing is that it's failing on the 3 config [09:34] let's hope it's reproduceable [09:34] sil2100: ps-indicators-autopilot-release-testing [09:34] rev 63 [09:35] Ah, ok, sorry, I misunderstood you [09:35] seb128: popey: btw, do not hesitate to upgrade to the daily-build ppa [09:35] So it's failing on all, hm hm [09:35] sil2100: yep [09:35] seb128: popey: I did it, and it looks good, just have this indicator failure ^ [09:36] * sil2100 pulls latest unity [09:36] didrocks, you rebuilt unity with the compiz opengl abi downgraded? [09:36] sil2100: daily-build ppa [09:36] yep [09:36] didrocks: i do every morning [09:36] sweet ;) [09:36] didrocks, k, upgrading [09:39] Ok, so, a check for is_focused is not enough, *sighs* [09:40] sil2100: so, the test is giving a false positive? [09:40] sil2100: do you know what changed from yesterday? [09:41] didrocks: I have no idea, I mean, this failure seems to happen when the Terminal takes a bit longer to start - i.e. the shell on it [09:42] I added a wait for the Terminal window to have focus, but as I can see it doesn't help always, since the window can be focused but not yet ready to fetch input [09:42] I'll try to think of a quick clean solution [09:44] mhr3, hey [09:45] mhr3, with the current daily build my dash has a sort of weird blue "," between the expendable categories and their ">" [09:45] e.g "installed" in the app lens [09:45] sil2100: thanks, counting on you to be able to release unity :) [09:45] is that normal/wanted? [09:45] mhr3: confirmed here [09:45] seb128, yea, i saw that with trunk some time ago [09:45] not sure who did that :) [09:46] you didn't ask? :/ [09:46] iirc the number of results wasn't displayed either in that case [09:46] no [09:46] there is only a "," [09:46] or "." [09:46] thought it's an upcoming design change [09:47] well, it could be to drop the label, the weird char seems wrong and misaligned though, I somewhat doubt it's a design change [09:47] andyrock, would you know something about that ^? [09:47] it's a ':' here seb128 [09:48] you have good eyes! I never noticed that! ☺ [09:48] I just have a . here [09:48] popey, could be, I've an hard time to see exactly what char it is with my background color [09:48] popey: he's good, isn't it? :) [09:49] mhr3, no ask nick :) [09:51] mhr3, btw if Nick is not online (or busy) i can try to fix it in the afternoon [09:51] we will never be able to release unity :( [09:52] huh? Something besides the AP test is broken? :( [09:53] sil2100: yeah, but not linked [09:53] sil2100: you can continue on the test :) [09:53] Oh... [09:53] ;) [09:54] hey dednick :) [09:54] howdy [09:54] seb128 | mhr3, with the current daily build my dash has a sort of weird blue "," between the expendable categories and [09:54] | their ">" [09:54] seb128 | e.g "installed" in the app lens [09:54] dednick: good, this and one test failing that sil2100 is working on are the last step before enabling daily release of unity :-) [09:55] dednick: do you know about it? or maybe have a look at it? [09:55] seb128: Do you have the "See X more results" showing next to the > ? [09:55] no [09:55] that replace those labels [09:55] replaces [09:56] I've like "Installed "," >" [09:56] without the "" [09:56] didrocks and mhr3 and popey confirmed [09:56] ok. i think i know what this is. mandel was doing something with StaticCairoText width calculations that might have caused a regression. [09:56] dednick: I just have one one the file category working, all the others are like seb128 told [09:56] i know it has caused some problems before [09:56] interesting :) [09:57] didrocks: i'll check with mandel about the status of that quick. [09:57] thanks :) [09:58] didrocks: what do we need to do to allow users to file bugs against the compiz package in the daily ppa with "ubuntu-bug compiz"? I am getting the usual "This is not an official Ubuntu package" popup from apport [09:59] popey: let's see if we need to do that, as the daily-build is just a 4 hours step between build and distro normally [09:59] popey: if we have some stack stalled for more days, maybe it will be useful, I'm just unsure for now [10:01] ok === mmrazik is now known as mmrazik|otp [10:18] didrocks, with unity ppa, dash blinks when I type something in the search bar, I'll try to make a video [10:18] thanks jibel, duflu will maybe have some hint about it [10:23] hm, there doesn't seem any 100% safe way, so I'll do it a bit hackish [10:24] sil2100: so adding more timing? [10:26] Or maybe, let me check something on my guest session ;p === mmrazik|otp is now known as mmrazik [10:27] popey, hey, which tool do you use to record a video of a desktop session with the dash? I tried recordmydesktop, but it's all dark and only captures the background [10:29] jibel: kazam [10:30] jibel, gtk-recordmydesktop works for recording the dash, etc on my raring [10:30] jibel, you can always use your phone or camera or something otherwise ;-) [10:32] seb128, it doesn't here with raring, all updates, etc on an intel atom IGC [10:32] i'll try kazam [10:35] dednick: did you find anything relevant? [10:39] seb128, finally recordmydesktop works fine, the problem comes from totem for the playback [10:42] ahah [10:47] bug 1098502 [10:47] bug 1098502 in Unity "Dash blinks when typing text in search bar" [Undecided,New] https://launchpad.net/bugs/1098502 [10:50] Ok, tested all nice solutions, none were giving me complete guarantee, so I'm using the simplest one [10:50] A sleep call, yea, we like those [10:52] sil2100: well, let's do that I guess… [10:52] * sil2100 sighs [10:53] Preparing a merge [10:53] I'll have a talk with thomi next week about doing it nicer, for now let's hope this will solve the issue for releasing unity [10:54] sil2100: sounds good, yeah, let's do what we can for now :) [10:54] but I need to know where dednick is as well, not sure if we should block the release on that issue, specifically if the fix is trivial… [10:55] didrocks: mandel is looking at the issue now. [10:56] dednick: ah sweet, what is his IRC nickname? can he come into that chan? :) [10:56] didrocks: mandel [10:56] didrocks: i've just asked him to join [10:57] thanks dednick :) [10:57] dednick, hello [10:57] hey mandel ;) [10:57] mandel: so, I guess all eyes are on you to release unity now, but no pressure :-) [10:58] mandel: more seriously, do you think the fix will be trivial? [10:58] didrocks, hello, so I'm looking at the StaticCairotText rendering, it seems that the PrelayoutManagement does have the correct size, witch is interesting [10:58] didrocks,he, no pressure :) [10:59] mandel: oh, "funny" did you notice it's a blue dot as well? [10:59] didrocks, that I did not see, let me check [11:00] mandel: it's barely noticeable, but we are more than one having . > [11:00] with . being : for popey [11:00] didrocks, is more like two blue dots, but is probably it trying to render the font and doing something wrong [11:00] (and blue) [11:00] ok :) [11:00] yeah, could be squished text [11:00] and not actually a : [11:00] but why blue? :p [11:01] well that too [11:02] didrocks, popey no to worry, I already have a fix, I need to clean the code a little and make sure it does not break in other cases [11:02] give me 15/30 mins or so to make sure [11:02] mandel: you rock! :-) [11:02] mandel: remember to check the tooltip as well ;) [11:03] dednick, hehe, yes, I remember that ;) === mmrazik is now known as mmrazik|lunch [11:04] \o/ [11:04] sil2100: btw, if you need some approval on your branch [11:04] https://code.launchpad.net/~sil2100/unity/autopilot_add_ugly_sleep/+merge/142869 [11:04] mandel: just to know, the fix will be in nux or unity? [11:04] Please, free karma ;) [11:04] didrocks, unity [11:04] mandel: ok, great, thanks ;) [11:05] not respawing a rebuild immediately then [11:05] * sil2100 always wonders why he spends so much time writing all this rationale in the merge requests [11:05] sil2100: I read them, worth it! [11:05] :) [11:06] sil2100: thanks! approved :) [11:06] Thanks :) [11:07] andyrock: soo, a Qt application for testing bug #1087422 is fine as well? ;) [11:07] bug 1087422 in unity (Ubuntu) "Windows that start minimized cannot be opened" [High,In progress] https://launchpad.net/bugs/1087422 [11:08] andyrock: I'll add the feature you're requesting in a moment then! [11:09] sil2100, yeah it's fine I've already tested :) [11:09] thank you btw :) [11:18] andyrock: updated - the name is fine btw. ;p [11:18] andyrock: test it once again and if it works as it did before, approve and let's get it merged! [11:19] andyrock: sil2100: can you just wait that we do this release before getting it merged? [11:19] andyrock: sil2100: I don't want we add one another variable parameter before this release :) [11:19] andyrock: ^ ;) [11:20] didrocks, sil2100 yeah no problem :) [11:20] thanks :) [11:21] andyrock: let's get the testapp thing merged in the meantime [11:25] didrocks, so nux revno 742 fixed a logical error in the layouts (no bug number, doh!) and before that was fixed I added a workaound for statica cairo text, with the new code in nux that is not longer needed and goes wrong in a diff way, I have fixed it and changed the test to ensure everything works correctly [11:26] mandel: oh interesting side effect of a workaround then :-) do you have a branch handy? [11:26] didrocks, please take a look at https://code.launchpad.net/~mandel/unity/static-cairo-text/+merge/142873 [11:26] sil2100, ok let me wrote the AP test for unity (without merging it for the moment) [11:26] * didrocks opens and builds while reviewing [11:26] \o/ [11:26] didrocks, I will add no comments about nux && layouts ;-) [11:27] heh :) [11:27] mandel: well, I know what you mean, I felt the pain in the past :) [11:28] didrocks, I don't know if I should feel happy because I knew where to find the problem or otherwise [11:29] mandel: ahah, well, you can call that "expertise on the subject" :) [11:29] some kind of niche knowledge ;) === _salem is now known as salem_ [11:32] can you reproduce that nautilus again starts behind other windows in raring+daily? [11:34] mandel: confirming the fix is working! :) [11:34] Mirv: hum, I don't see that [11:34] mandel: nice work, thanks! ;) [11:34] and still having tooltips :p [11:34] didrocks, take a look at the launcher, but works too, just in case :) [11:34] didrocks, you are faster than me ;) [11:34] eg. maximized firefox, super + 1, nautilus starts to the background not front [11:35] didrocks: ok, interesting, seems to me similar to what quantal had before nautilus was patched [11:35] mandel: heh :) [11:35] Mirv: ah indeed, if started with super + something [11:36] I was cliking on the launcher [11:36] even clicking on the launcher it goes unfocussed for me [11:37] hum, not here, maybe something with the state [11:37] we had to drop the nautilus patch, upstream fixed it in a different way for 3.6 which works for gnome-shell but not us [11:37] same as seb128, although I noticed it requires the maximized window [11:39] andyrock was working on that [11:39] he had a patch for gtk, not sure how that is going [11:40] seb128, what? [11:40] andyrock, nautilus not getting focus in raring [11:40] again? [11:40] andyrock, well, "in raring", in 3.6 since we dropped your patch, since cosimoc did a different version which works for gnome-shell but not unity [11:41] andyrock, well, didn't you say it's a gtk bug in the timestamp handling? [11:41] seb128, yeah i'm sure that gtk/gdk had that bug but startup notifications maybe can workaround it :) [11:42] and unity does not support them [11:42] right [11:42] so it's not "again", it's "still" ;-) [11:42] andyrock, did you ever get any feedback on your gtk patch? [11:42] I wonder if we should just distro patch it [11:43] seb128, no Trevinho and me opened a gdk bug with a very simple test program showing the issue [11:43] btw we really need to support startup notification in unity [11:44] Trevinho wants to do it directly in bamf [11:44] but he is too busy [11:44] we need to put that bug in the "polish blueprint" [11:46] right [11:48] filed bug #1098533 [11:48] bug 1098533 in nautilus (Ubuntu) "New nautilus window starts behind fullscreen window" [Undecided,New] https://launchpad.net/bugs/1098533 [11:49] s/fullscreen/maximized/ isn't it? [11:49] didrocks: thanks, yes, title is wrong [11:49] yw ;) [11:50] test case was correct. fixed title. === mmrazik|lunch is now known as mmrazik === MacSlow is now known as MacSlow|lunch [13:06] seb128: I have a workaround that I'll try to land upstream too... [13:07] seb128: this is the issue (and workaround) https://bugzilla.gnome.org/show_bug.cgi?id=688830 [13:07] Gnome bug 688830 in gtk "gtk_window_present uses an invalid time value, and does not always work" [Major,Unconfirmed] === MacSlow|lunch is now known as MacSlow [13:19] Trevinho, hey, thanks, I was looking for the bug earlier but I didn't remember enough about it to find it back [13:19] Trevinho, we should maybe distro patch it [13:20] Trevinho, not sure upstream is going to go anywhere on that anytime soon [13:20] seb128: I was talking of nautilus only.... Even if that change could maybe done at gdk level... [13:21] oh, ok [13:27] Trevinho: another subject: on the one ws by default, do you still have something to do on the unity side? [13:27] Trevinho: on the g-c-c side, I'll just set the hsize and hsize to 1x1 or 2x2 if disabled/enabled [13:33] didrocks: yes, I'm in the process of updating the shortcut-hints now... That's the final step.. [13:33] didrocks: ah, and probably a few lines to remove in launcher controller (to avoid removal), but on launcher side everything is done [13:33] Trevinho: ok, but agreed that on my side, I just have to do that, change hsize/vsize? [13:33] didrocks: yes, we listen to that value [13:33] if one of them is > 1, you enable the "ws mode" [13:34] Trevinho: excellent, thanks :) [13:34] would be a nice hacking during my first flight on sunday [13:34] too early on Sunday, so need something easy :) === dandrader_ is now known as dandrader|afk === dandrader|afk is now known as dandrader [14:01] cyphermox, more yellow paint. While my branch built fine for i386 in a PPA, amd64 apparently died overnight [14:21] didrocks: will you also include an option to show/hide the desktop icon? [14:21] didrocks: it should be explained here btw https://docs.google.com/a/canonical.com/document/d/1cbPd9WSSbFHg4Z7BSOQxDKusMe2aJjj0FEF2AMxNOZM/edit [14:23] Trevinho: yep, I'll do it at the same time [14:23] didrocks: fine, the notes at the bottom should explain everything [14:27] yep ;) [14:31] didrocks, so how can we match a given daily build to a bzr commit number? [14:33] mterry: quick easily, look at debian/changelog [14:33] you have "automatic snapshot from rev <…>" [14:33] didrocks, ah yes! ok [14:34] mterry: btw, hopefully this build will be the first daily release! :) [14:34] running all the tests [14:34] didrocks, yeah :) [14:34] then, running only the indicator tests on the indicator stack + whole ppa [14:34] and if everything is fine, publishing! [14:34] * mterry crosses fingers [14:37] * didrocks too [14:37] I think ~ 20 minutes for all the test running [14:37] then, the indicators one will start (~55 minutes) [14:37] tests* [14:49] didrocks, how can I see a list of packages uploaded for SRU but not yet accepted into -proposed? [14:50] mterry: check the unapproved queue [14:50] like for quantal: [14:50] https://launchpad.net/ubuntu/quantal/+queue?queue_state=1&queue_text= [14:50] didrocks, there's a page on LP, right? [14:50] yep :) [14:50] didrocks, nice, thanks [14:50] yw [14:50] that's what I couldn't remember [15:07] cyphermox, despite the amd64 build failing for libappindicator in my PPA, I have to work on some other stuff today. But I do think my branch makes things better anyway (i386 worked) [15:34] mterry: cyphermox: FYI, I remove dbusmenu from the list of daily release, it's failing on every arch due (weirdly) to a missing file (see ppa failure) [15:34] mterry: sil2100: on another note, all indicator tests passed \o/ [15:34] nice work sil2100 ;) [15:34] yay [15:35] Yaaay! \o/ [15:35] as soon as unity is published on armhf, it will be time for the first daily release :) [15:35] https://launchpadlibrarian.net/128198353/buildlog_ubuntu-raring-armhf.unity_6.12.0daily12.12.05bzr3034pkg0raring0_FAILEDTOBUILD.txt.gz <- but it looks like a random failure [15:36] didrocks: I guess in daily all is OK? [15:36] sil2100: yeah, waiting for publishing [15:41] sil2100, ld crashed!? ick [15:46] ok, done [15:46] time to force publication \o/ [15:47] Excellent! [15:53] ok, readyyyyyyy [15:54] and copy request to distro done \o/ [15:54] sil2100: mterry: first real daily release of unity done to the distro \o/ [15:54] enabling the cronjob now! [15:55] nice work everyone :) [15:55] w00t! [15:55] * mterry opens champagne [15:55] * sil2100 claps his hands [15:56] * mterry uncomfortably watches the Jenkins bot proposing and approving its own branches [15:57] mterry: heh, you know that even that was broken? [15:57] I had to patch bzr lp-propose to have the option working [15:58] :) === dandrader is now known as dandrader|lunch [16:46] fginther: hey, I thought that merging for the latestsnapshot was skipping the build? https://code.launchpad.net/~ps-jenkins/bamf/latestsnapshot/+merge/142925 [16:49] didrocks, no sorry, there has been no progress on that [16:49] fginther: ok, you follow them and reapprove? :) [16:50] didrocks: indicator-session buils fine now (but with start-service disabled), and I got a autopkgtest test almost working; wanna do a first review before I propose a merge to actually merge this? [16:50] didrocks, I thought the approved rev_id issue was fixed. This is the first one of seen w/o an approved revid for many weeks [16:50] https://code.launchpad.net/~mathieu-tl/indicator-session/fix-tests/+merge/142951 [16:51] cyphermox: I would love to! looking :) [16:52] didrocks, looking into what happened here [16:52] fginther: thanks :) [16:54] cyphermox: it sounds good to me, did you run it with some autopkg infra? [16:55] yeah that's what I'm doing [16:55] cyphermox: maybe localcheck should be integrationcheck ? [16:55] for now it's still failing but I was able to manually run it [16:55] could be [16:55] ok :) [16:55] cyphermox: as you wish for the naming, but this makes more sense to me than "local" [16:55] so it's just a matter of getting the autopkgtest script right [16:55] yeah [16:55] sweet! [16:55] :) [16:55] nice work cyphermox, can't wait to see that one down! :) [16:56] yeah me too [16:56] the heavy drinking is starting to take a toll ;D [16:56] cyphermox: ahah, you needed an excuse for that! :-) [16:56] didrocks: always do :P [16:56] heh === davidcalle_ is now known as davidcalle === dandrader|lunch is now known as dandrader [17:27] fginther: bamf and dee are the only 2 remaining ones not merged yet, isn't it? [17:30] didrocks, compiz is still building [17:31] right :) [17:32] didrocks, yes, just those 3 [17:32] fginther: ok, thanks! [17:32] fginther: yeah, maybe in the future, having those bypassing the build will gain a lot of CPU cycle [17:32] fginther: especially as we are doing that everyday now (for all projects having new code) [17:36] didrocks, I'll make some time to work on this today and next week, will keep you posted [17:36] fginther: excellent, thanks! [17:36] fginther: if you need some special tag to recognize those kinds of merge or have any question, do not hesitate to ask! [17:37] didrocks, thanks! === salem_ is now known as _salem === dandrader is now known as dandrader|afk === rsalveti_ is now known as rsalveti === dandrader|afk is now known as dandrader === _salem is now known as salem_ === salem_ is now known as _salem