=== _thumper_ is now known as thumper [09:23] sil2100: still having issues with ps-indicators-check on intel, could you please look at it? two tests failed. [09:24] didrocks: ^ [09:44] heh, having xorg-gtest as part of our CI is awesome \o/ [09:44] mmrazik: feel free to disable the memcheck job [09:44] (if its taking ages) [09:45] smspillaz: I actually don't mind that much. It mostly slows down compiz as the other branches need to wait [09:45] ah okay, I guess we can leave it on [09:45] i.e. compiz-ci ; autolanding is not affected [09:45] mmrazik: it might be a tad more useful actually to run it with -VV, eg [09:45] ctest -D ExperimentalMemCheck -VV [09:46] that way we can see the valgrind output (and then we can disable -VV on the other test run) [09:46] smspillaz: ok. will look at that (not sure if today, though) [09:46] mmrazik: no problem :) [09:57] cyphermox: looking [09:58] cyphermox: hm, is jenkins down? [09:59] no? [09:59] cyphermox: ah, nevermind, finally got response [10:01] cyphermox: build 140, yes? The first failing test looks as if the panel hung up, since there's no "Ubuntu Desktop" title there even [10:01] The second one looks more likely as an autopilot failure, looking closer [10:02] err ok [10:02] I don't know, just see the two tests that failed, hud or something right? [10:04] Yes, one with the HUD, HUD doesn't seem to be appearing even on the tap of 'alt' [10:04] Sadly, no debug from unity :< [10:15] I think the safest way would be to add a short wait in the panel test [10:26] cyphermox: I created a small enhancement that might make that failure less probable [10:26] https://code.launchpad.net/~sil2100/unity/autopilot_panel_sleep_add/+merge/149515 [10:26] But I think it was anyway a strange timing race, so it shouldn't be much reproducible [10:26] As for the HUD failure, still not sure what went wrong there, maybe it's also a single case, looking into it now in detail [10:27] Hi, me again, is unity-team/staging now contains something that does not really want to move past loading compiz core. CCSM crashes, deleting configuration does not help :( [10:33] cyphermox: heh, I think I see what's going on with the HUD test [10:38] cyphermox: I thought I have fixed that... not sure if we can do it any better :< So for now I would say that all we can do is re-launch, it shouldn't happen again [10:41] cyphermox, mterry, didrocks: build 141 of indicator tests has broken ATI :/ [10:42] cyphermox, mterry, didrocks: seems like unity release testing build 92 also has problems === mmrazik is now known as mmrazik|lunch [10:45] sil2100, I canceled it manually because the run was going to be bad anyway [10:49] mterry: ok, thanks for maintaining that :) [11:07] sil2100, smspillaz: if you dist-upgrade from the daily-build PPA, didrocks tells me that things are broken because of the ccp plugin. Can someone look at that? [11:09] mterry: didrocks: "because of" ? [11:09] mterry: didrocks: what exactly is broken ? [11:10] my system is kinda half-working at the moment, its a bit difficult for me to test that [11:10] smspillaz: did you try dist-upgrading? cyphermox's machine just have an issue with the cpp plugin, as this is what changed… [11:10] didrocks: define "issue" :) [11:10] smspillaz: no ui? [11:10] got a copy of the .xsession-errors handy ? [11:10] cyphermox: ^ ? [11:11] yeah, let's see if cyphermox still has it on his tty on a terminal :) [11:11] smspillaz: FYI, we downgraded compiz/unity and then, back on track [11:11] didrocks: I can't really test it properly at the moment, my system is kinda half-screwed [11:11] didrocks, smspillaz: popey just tried the compiz with the copytex-addition fix you proposed yesterday and he says he has no UI as well on the VM [11:11] http://paste.ubuntu.com/1689845/ <- his .xsession-errors [11:11] Not sure if related? [11:12] smspillaz: did the copytex thing land yesterday? [11:12] i.e. the configuration addition [11:12] yes [11:13] Could those problems be *somehow* related to this fix? hmmm [11:19] sil2100: is that the whole .xsession-errors ? [11:20] smspillaz: yes, in case of a VB instance <- popey ^ [11:20] yo [11:21] smspillaz: lemme know if you need any further debug info === mmrazik|lunch is now known as mmrazik [11:30] smspillaz - no unity plugin gets loaded , i have the same thing, no decorations, nothing. Just compiz core and running unity directly does not do anything better than that [11:30] no panels [11:30] empty spaces [11:31] rye: could you try fetching and pastebining .xsession-errors ? [11:31] I'm trying to test build compiz without the latest commit [11:32] sil2100: ccsm crashes too - http://paste.ubuntu.com/1690024/ but here's .xsession-errors from unity session - http://paste.ubuntu.com/1690026/ [11:35] rye: thanks! So it seems to be the same thing [11:35] I'm betting on rev 3617 then [11:35] rye: sil2100: hrm, I wonder if the settings upgrade stuff died [11:35] I'll have a look into it right now [11:35] Compiz almost finished without it, will check [11:35] might want to roll back the package in the distro [11:36] sil2100: you'll need to reset your settings [11:36] smspillaz: by crash i meant it segfaulted [11:39] just building and installing now [11:39] smspillaz: previously I had 1:0.9.9~daily13.02.19-0ubuntu1 installed, now I have a version based on 3616 [11:39] didrocks: sil2100: it'll take me about 30 minutes to look into this, you might want to hold the package [11:39] smspillaz: but my guest session also crashes, not sure why [11:39] smspillaz: since I never had 3617 installed [11:41] smspillaz: something's wrong, only 'core' is being loaded for me when running unity, hmm [11:45] sil2100: yeah, the settings upgrade proably screwed up and reset active_plugins to just "core" [11:46] like I said, I would hold the package until I figure out whats wrong [11:50] sil2100, is this Compiz problem just happening with unity-team/staging PPA ? [11:51] * MCR_ is wondering that he is not affected this time... [12:01] smspillaz: what info do you want? [12:01] cyphermox: don't worry about it. I'm going to have a look into it now === _salem is now known as salem_ [12:18] MCR_: not sure what's wrong as well [12:23] erk. unity doesn't start after updating to Unity 6.12.0daily13.02.20-0ubuntu1 and compiz 1:0.9.9~daily13.02.20-0ubuntu1 [12:25] http://paste.ubuntu.com/1690204/ .xsession-errors from my desktop [12:25] (not a VM, bare metal) [12:26] can you use other desktop environments? [12:27] popey: sil2100: didrocks: like I said, I know its broken. Hold the package before anyone else breaks their systems please [12:27] sorry, missed that [12:29] popey: yep, smspillaz is working on it [12:29] yeah, /ignore popey ;) [12:29] * sil2100 ignores [12:29] ;) [12:29] ;p [12:32] so, has someone held the package yet ? [12:33] smspillaz: from going to distro? [12:33] didrocks: ^ [12:33] But I think it won't go anyway [12:35] yes [12:35] mterry: ^ [12:46] sil2100, smspillaz: I think it won't go in because tests would fail [13:08] mterry: uhh, there are no tests covering this? [13:09] unless you're referring to AP [13:13] smspillaz: AP tests, since I think mterry meant that AP tests will just fail miserably with the new compiz, as there will be no shell to run them [13:14] didrocks: sil2100: okay, hang on, I'm just getting a test in place, will have a branch up soonish [13:15] can you file a bug ? [13:40] sure thing [13:40] smspillaz: a vague one? :) [13:41] smspillaz: do you mind fixing (if not already, the other bug? at the same time) [13:41] smspillaz: like the one where you didn't change the default to include the new value [13:41] the new plugin [13:41] I meant [13:43] didrocks: already fixed [13:43] \o/ [13:43] smspillaz: bug #1130679 [13:43] didrocks: um, file a bug and I'll stick the appropriate stacktrace in [13:43] bug 1130679 in Compiz "[2013/02/20] compiz/unity don't run, just loading cpp" [Undecided,New] https://launchpad.net/bugs/1130679 [13:43] chers [13:43] *cheers [13:48] okay, test passing, let me just run compiz to double check [13:50] sweet ;) [14:06] didrocks: sil2100 https://code.launchpad.net/~compiz-team/compiz/compiz.fix_1130679/+merge/149574 [14:21] sil2100: do you mind testing that one with the latest unity from the ppa? ^ === dandrader is now known as dandrader|afk [14:33] fginther, didrocks: What do you say to adding unity-greeter to the daily build system? [14:34] mterry: wait wait wait, I love that idea! :-) [14:34] fginther: sorry. I mean I want auto-merging. Not auto-release to distro [14:36] thomi, what is the story with autopilot and qt5 quick? [14:41] mterry: https://bugs.launchpad.net/cupstream2distro/+bug/1099607 [14:41] Launchpad bug 1099607 in Canonical Upstream To Distro "Put an option for a stack to always be manually published" [Undecided,New] [14:47] didrocks: aye [14:47] didrocks: test-building [14:55] sil2100: and test-installing then I guess :) [15:00] Hi guys, where should i shoot my bug reports at? [15:01] I've noticed some irritating things in unity, where would i report them? === francisco is now known as Guest92682 [15:01] MrQuist, https://bugs.launchpad.net/unity [15:01] Thanks MCR_ [15:02] smspillaz, I think it is ready for merging now: https://code.launchpad.net/~mc-return/compiz/compiz0.9.9.merge-plugin-freewins/+merge/146291 [15:03] smspillaz, I have already fixed a lot of stuff there, I fear if I do more I'll break it again ;) [15:05] afk [15:09] smspillaz: ok, I test-built it and installed, but it doesn't work - on the other hand, I am not certain my system builds a working compiz [15:09] Since recently I didn't build compiz on my system and install it [15:12] MCR_, Here's the bug report; https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1130713 [15:12] Launchpad bug 1130713 in unity (Ubuntu) "Empathy doesn't open when clicking messages -> chat in system tray" [Undecided,New] [15:12] mind to give it a quick look? [15:13] smspillaz: would have to do a test build of the working packages [15:21] sil2100: reset your settings and rm -rf ~/.config/compiz-1 [15:24] smspillaz: I think I'm simply not able to build a working compiz [15:25] sil2100, just use jhbuild ;) === andyrock is now known as andyrock|afk [15:25] I use bzr-builddeb ;) [15:26] sil2100: the way I tested it was to run a simple program for compizconfig and check if that segfaults === mmrazik is now known as mmrazik|afk [15:27] (compizconfig processes setting upgrades on startup, so you'll hit the right codepath) [15:29] smspillaz: ok, so... [15:29] smspillaz: I did this: I just built the compiz that is in raring and I get a working compiz [15:29] smspillaz: but when I build your compiz source from your branch, I don't get unity, I only see core being loaded [15:29] mterry, I'll get unity-greeter autolanding today [15:30] smspillaz: and it's on a guest session, so there is no config history [15:30] sil2100: full output ? [15:30] and does it crash ? === mmrazik|afk is now known as mmrazik [15:31] smspillaz: one moment, will provide some info in a moment [15:31] mterry, once I can get to jenkins :-( [15:32] fginther, I don't think it has inline debian/ yet [15:32] fginther, by autolanding, we mean just the code, not released packages to distro, right? [15:33] mterry, yes autolanding == auto merger [15:34] fginther, what do you call releasing packages to a PPA and then distro? (just so I can match terminology) [15:34] mterry, but it would be easier to wait if you are working on the inline packaging [15:34] fginther, I'll do that, then ping you [15:36] Lets all escape from this matrix! [15:38] mterry, umm, I don't have any special terminology for releaseing pacakges [15:38] fginther, we should have one :) [15:39] auto-releasing? [15:39] auto-pushing [15:39] :~ [15:39] auto-distro [15:39] auto-didrocks [15:39] tsss :p [15:39] auto-upload [15:39] * didrocks likes auto-upload [15:39] auto-didrocks +1 [15:39] didrocks has been outvoted [15:39] rohhhh [15:41] o/ auto-didrocks [15:43] smspillaz: http://paste.ubuntu.com/1690934/ [15:43] smspillaz: this is what I get in .xsession-errors, and also the same when trying to launch unity from the console [15:43] sil2100: can you check if its crashing ? [15:44] smspillaz: sometimes it's crashing, but sometimes it's hanging and just staying there [15:44] sil2100: gdb ? [15:44] Usually it's hanging when it's unable to find unity-panel-service [15:45] smspillaz: ok, it seems not to be crashing [15:45] sil2100: so its actually running ? [15:46] fginther, oh actually... if we're not doing auto-didrocks for unity-greeter, we don't need inline packaging. auto-landing itself doesn't require it, right? [15:46] can you fetch the active-plugins value? gsettings get org.compiz.core:/org/compiz/profiles/unity/core active-plugins [15:47] mterry, no, autolanding doesn't require inline packaging. It's just that the jenkins job needs to be modification if inline packaging is added later. not a big deal [15:47] fginther, ok. Well, I can inline then [15:49] sil2100: ^ [15:50] smspillaz: ['core', 'composite', 'opengl', 'compiztoolbox', 'decor', 'vpswitch', 'snap', 'mousepoll', 'resize', 'place', 'move', 'wall', 'grid', 'regex', 'imgpng', 'session', 'gnomecompat', 'animation', 'fade', 'unitymtgrabhandles', 'workarounds', 'scale', 'expo', 'ezoom', 'unityshell'] [15:50] that's curious === mmrazik is now known as mmrazik|afk [15:50] Strange thing ;/ [15:50] sil2100: cat ~/.config/compiz/compizconfig/done_upgrades ? [15:51] smspillaz: one moment (switching guest session is a pain) [15:52] :/ [15:57] smspillaz: there's no compiz directory in ~/.config/ ;/ [15:58] sil2100: so it just isn't loading plugins? [15:59] Seems like it, uh [16:02] sil2100: I'll need to recompile everything then [16:03] smspillaz, welcome to the wonderland of Linux [16:04] MrQuist: no, more like "welcome to the wonderland of i-said-i-would-update-this-machine-last-year-but-was-a-stinge-on-money" [16:04] this thing is like 6 years old [16:05] there were people who were born after I started using this machine [16:05] and I can talk to them === dandrader|afk is now known as dandrader [16:06] hahaha yeah that last sentence was important [16:13] fiscal responsibility has its downsides [16:14] lovely, cpu temperature is 96C [16:14] this thing is going to take forever === mmrazik|afk is now known as mmrazik === bambee is now known as rperier [16:47] bschaefer, Hi :) Wanna try something truly awesome ? [16:48] seems like the answer is no, hehe [16:51] sil2100: hmm, curious [16:51] sil2100: it just creates a profile called "Default" [16:54] mterry, didrocks, an updated xpathselect still isn't getting rebuilt by the daily job [16:55] fginther: yeah, as discussed with mmrazik, it's a known issue and I'm trying to get some time to work on the tests, and then a fix [16:55] fginther: I can workaround it easily, but I would prefer ensuring I don't regress anything else first [16:55] fginther: is that urgent, like, needed before EOW? [16:55] didrocks, ah. thank you. I wasn't aware you had discussed this with mmrazik. [16:56] fginther: can you ping thomi to check how urgently he needs that? [16:56] didrocks, thomi is blocked by this and needs a fix soon [16:57] didrocks, mmrazik I'll let thomi know the status and see how long he can wait [16:57] fginther: thanks a lot! [16:58] sil2100: sure the guest session hasn't just been broken for ages ? [16:58] didrocks, thanks for the update [16:58] certainly feels like it [17:02] MrQuist, bug report looks good, but I cannot confirm it because I am no empathy user... [17:02] fginther: yw! keep me in touch :) [17:08] didrocks, seems like the "Ctrl+Super+Down" magic worked, but I still cannot 100% confirm the g-c-c part... [17:08] didrocks, could you check that ? [17:19] didrocks: sil2100: do you know if we do some kind of gsettings magic on the user session? [17:19] default keys or anything like that? [17:19] I thought we couldn't because the schemas were relocatable [17:30] smspillaz: yeah, we don't have that [17:49] didrocks: how does compiz know the right plugins to load then? [17:50] didrocks: I don't think it would be looking at unity.ini [17:50] smspillaz: gsettings config? [17:50] smspillaz: TBH, if we can't figure out what's wrong, I think we should just revert the offending commits and unblock trunk [17:50] smspillaz: I have no time and multiple issues to deal in // right now [17:51] didrocks: well ... I'd like to figure this out :/ [17:52] didrocks: Ill build again the package and work from there I guess [17:53] feel free to revert stuff if you want, I'll keep looking into this tomorrow [17:53] I always hated this settings code [17:53] sil2100: do you mind doing that? ^ [17:53] thanks smspillaz :) [17:53] smspillaz: yeah, bzr bd should do what you need [17:53] (sorry, really dug into too many things to handle right now) [18:11] oh no! What happened here ? http://imagebin.org/247448 :) === andyrock|afk is now known as andyrock [18:14] Hi andyrock :) Can you help me with http://imagebin.org/247448 ? (just joking) [18:17] ;) [18:45] MCR_: ctrl + super + up/down doesn't work here [18:45] didrocks, hmm [18:46] MCR_: just standard upgraded from quantal machine [18:46] Ctrl+Super+Up was not even touched by me [18:46] interesting [18:46] ah [18:47] for Up, it was my Nautilus window not working [18:47] for a terminal, only Down doesn't work [18:47] grmpf [18:48] didrocks, thanks for testing - I am investigating [18:49] MCR_: thanks :) [18:49] didrocks, one last question [18:49] yeah? [18:49] in g-c-c, I'm seeing "Down" [18:49] does it correctly show up in g-c-c [18:49] so yeah :) [18:49] but the label isn't correct [18:49] oh [18:49] * didrocks realize the label [18:49] Restore window Ctrl+Super+Down ? [18:49] MCR_: is label is Restore window [18:49] the* [18:50] which should rather be minimize/restore, right? [18:50] well, that was my first proposition [18:50] but then we decided it would be much easier to not manipulate the g-c-c name [18:51] MCR_: ok, but yeah, it's associated to that label [18:51] didrocks, do you have CCSM installed on that machine ? [18:51] MCR_: yep [18:51] please open CCSM->General->Bindings [18:52] and tell me what you see there [18:52] MCR_: Ctrl + Super + Down is associated to the right label [18:52] unamximize/minimize… [18:53] it's the right one, isn't it? [18:53] "Unmaximize W" and "Unmaximize or Minimize W" -> ok [18:53] yes [18:53] so strange [18:53] this part is correct, but it still does not work ? [18:53] nope :/ [18:53] tried gedit, terminal [18:53] oh [18:53] sorry [18:53] maybe I screwed it up [18:54] I hope so (TBH) [18:54] this shouldn't minimize as well? [18:54] it should [18:54] like, it's not the 3 states that John wanted? [18:54] ok, so I only have maximize/restore [18:54] but not minimize [18:54] first restore if Fullscreen, then minimize if restored already [18:55] !Fullscreen but Maximized [18:55] MCR_: I am only a bot, please don't think I'm intelligent :) [18:55] MCR_: so the minimize part doesn't work here [18:55] at least, it didn't regress :) [18:56] but it is still strange [18:56] as this CCSM function is a new core window function [18:56] and it will just call this function... [18:57] which will restore a maximized or semi-maximized window or minimize all others (restored or normal) [18:58] but maybe the g-c-c part somehow overrides CCSM and still uses the other function, but in this case the information you see in CCSM would not be correct, so I'm a bit baffled right now... [18:59] MCR_: hum, I think soon, it will be time for printf :) [19:02] didrocks, do not worry about it -> it just can be a minor issue that is left (and there is still time until 13.04 release), but I have to finally set up a bare metal test system and investigate there myself to not steal more of your time... [19:02] MCR_: thanks a lot :) anyway, at least, we didn't regress [19:02] which is the #1 point IMHO :) [19:02] but yeah, would be great to have that working [19:02] sure [19:04] didrocks, just FYI -> this is the probable cause: https://code.launchpad.net/~mc-return/compiz/compiz.merge-fix1115128-expose-unmaximize_or_minimize_key-in-g-c-c/+merge/146384/comments/319897 [19:05] didrocks, what do you think ? [19:05] MCR_: I don't think so, the gconf is just about setting the key [19:05] MCR_: sorry, can't go on on that discussion, urgent issues popping up :) [19:05] yes, but your testing shows that it calls /apps/compizconfig-1/profiles/unity/plugins/core/screen0/options/unmaximize_window_key [19:05] instead of /apps/compizconfig-1/profiles/unity/plugins/core/screen0/options/unmaximize_or_minimize_window_key [19:06] but anyway, thanks for testing [19:06] I'll make sure it gets fixed properly [19:07] ;) [19:14] thomi: https://code.launchpad.net/~mathieu-tl/gtester2xunit/rip-python-stuff/+merge/149666 can you please review? [19:14] didrocks: ^ [19:15] cyphermox: that doesn't look right to me - how do you get XML output? [19:15] yeah [19:15] well, you get test output [19:15] tests run, the build completes [19:16] cyphermox: ...the whole point of using pyruntest is so we can get xml output, and thereby hook that up to jenkins. What's the problem with using pyruntest? [19:16] xml output *where* [19:16] you mean in the gtester2xunit tests? [19:16] yes [19:17] nah, it's not xml output there. but at the same time at some point you got to choose the battles. the problem with pyruntests is it needs much work to land to distro in main, and that's blocking other things [19:18] otherwise we´ll sed the heck out of it to massage the output in xml format [19:18] cyphermox: ok... well as the author of pyruntests it'd be nice if peopel told me when things didn't work... how am I supposed to fix them otherwise? [19:18] I had no idea you wrote pyruntests [19:18] there's exactly 1 bug filed against pyruntest, and it's actually fixed already :) [19:18] and honestly, only vaguely interested in whatever else but unbreaking the archive right now [19:19] cyphermox: yes, I wrote it so we had a common way to run tests and get xml & coverage info for python projects [19:20] cyphermox: well, I understand that "un-breaking the archive" is important, but this feels like a step backwards to me. Perhaps you could take 10 minutes to document exactly what isn't working for you, so we can fix it? [19:20] it needs to make it to main [19:20] that's all, forget that merge [19:20] Is gnome-control-center the right package to file a bug against for a "Mouse and Touchpad" system settings problem ? [19:21] ok [20:37] smspillaz: what's up? How's the situation? === francisco is now known as Guest48228 [22:12] hi all, does anyone have any clue why orca won't read unity dash icons? [22:12] or any pointers how I can debug and fix this? [22:17] AlanBell: The dash atk code is buggy and incomplete. [22:17] I'm not sure dash icons are instrumented properly for orca [22:17] I think they;re just bitmaps sent over a DBus interface and blitted to the screen [22:17] the categories work, it speaks as you move between them [22:18] oh :( [22:18] the lens selection icons at the bottom of the dash are equally non functional [22:19] might be worth opening a bug for each of those problems, it'll be more likely to maybe get fixed some day [22:19] bug 1066157 [22:20] bug 1066157 in unity (Ubuntu) "dash +orca does not speak the names of application icons" [High,Triaged] https://launchpad.net/bugs/1066157 [22:21] ah, excellent [22:21] if I knew where to look I would work on it [22:21] but as it isn't gtk I don't really know how to inspect it or where to start [22:22] I guess as it isn't using standard widgets all the atk stuff has to be written the hard way [22:23] yes [22:23] so, the file plugins/unityshell/src/unity-launcher-icon-accessible.cpp would appear to be responsible for it [22:23] launcher icons and dash icons are not the same thing [22:24] the root of the problem is Nux widgets have to a11y support, which is an outstanding wishlist item [22:24] huh, you can drag them from dash to launcher but they are different widgets [22:24] Yeah, the unity atk code needs to be re-architected purely around Nux. [22:25] And only have extra bits in unity where required. [22:25] or QML [22:25] * bregma maintains studious silence [22:26] unity source is big :( [22:27] No kidding. [22:27] And thats only one part of the whole stack. [22:36] hmm, this doesn't look fun. Too much c++, no idea how it slots together, no idea how to debug it [22:37] and it might be a waste of time if it goes QML [22:38] the source contains a guide to c++ but not a guide to unity [22:42] Unity itself is C++, but the atk code is GObject C. [22:42] With pointers to C++ objects and data structures. [22:42] oh joy, pointers :( [22:43] Well pointers as in many of the variables referred to in the atk code are from unity's C++ objects and data structure. [22:46] one thing I don't get is that clearly there has been a *heap* of work on these unity-*-accessible.cpp files and yet the primary function of unity is completely broken [22:47] I could understand if filters were broken or categories or quicklists, but dash launcher icons seem like the only thing that really matters [22:48] as I said, I don't think the dash code is fully implemented, and it is buggy. [22:48] and this is why I keep recommending people use the LTS, and Unity 2D. [22:48] Because Unity 2D is almost 100% accessible. === salem_ is now known as _salem