/srv/irclogs.ubuntu.com/2014/04/14/#ubuntu-desktop.txt

=== m_conley is now known as m_conley_away
=== maclin__ is now known as maclin
pittiGood morning05:04
mlankhorstHello, world!\n07:02
RAOFmlankhorst: Yo! Whatup!07:09
mlankhorstnothing much, kernel is a pain :P07:16
RAOFHurray! :)07:17
RAOFIn other pain points: looks like helgrind doesn't understand C++11 very well.07:18
Ads20000Hello, I installed Ubuntu GNOME on my Trusty installation on my Ubuntu desktop because Unity wasn't working and I noticed that the GNOME 3 PPA that claims to have the 3.12 packages for Trusty doesn't. Any chance I could help (or at the very least join the GNOME 3 team)? Thanks :)07:23
Ads20000(considering that Debian Experimental has the 3.12 packages - all I would need to do would be to move them over to Ubuntu and test them - should be fairly stable considering they've basically already been tested upstream...)07:24
=== jhernand1z is now known as jhernandez
=== maclin_ is now known as maclin
Laneysalut tout le monde08:05
didrocksbonjour Laney, comment vas-tu?08:08
didrocksbon week-end ? :)08:08
didrocks(you are starting to regret, right? :p)08:09
* Laney summons all the knowledge of french08:11
Laneytrès bien!08:11
didrocksheh ;)08:11
didrocks\o/08:11
Laneyet toi?08:11
didrockstrès bien aussi ;)08:11
didrocksa lot of people this week-end at home08:11
didrocks(we were 7 sleeping at my home). And it was kind of geeky08:12
didrocksreally nice to have seb128 around for some days as well08:12
Laneywow08:12
Laneythe others were free software people?08:12
didrocksyeah ;)08:12
didrocksubuntu-fr/gnome08:12
didrockswe were at jdll: http://www.jdll.org/08:13
didrockswhich is a FLOSS "public oriented" conference + booth08:13
Laneyoh cool08:13
Laneydid anyone give a talk?08:13
didrocksnot this year08:16
didrocksnot a lot of people are coming, I became lazy and didn't submit08:16
didrocksit was more an excuse to hang out with people you know well more or less ;)08:16
seb128good morning desktopers!08:18
seb128didrocks, !!! ;-)08:18
didrockshey seb128! long time no see :)08:18
Laneyola seb12808:18
seb128indeed!08:18
seb128hey Laney08:19
seb128how are you?08:19
Laney¿cómo estás?08:19
Laneypretty good thanks08:19
Laneythe sun has come!08:19
Laneydid you have a nice weekend?08:20
seb128excellent08:24
seb128Lyon is great, and stay at didrock's is awesome ;-)08:24
seb128I managed to do some 13.10 -> 14.04 upgrade for people who paid me back with beers at the conf as well08:25
Laneywho needs a salary?08:25
seb128haha08:25
Laney:P08:25
seb128speaking of which08:25
seb128mvo, mmmmmvvvvvooooooo08:26
seb128you have bogs!08:26
mvoseb128: hm?08:26
didrocksmvo: bugs coming from Lyon, they are good bugs! :)08:26
Laneysome of the finest tasting bugs in all of france08:26
mvodidrocks, seb128: *cough* beer bugs?08:26
seb128mvo, one of the 13.10 -> 14.04 had dist-upgrader getting down with an utf error while trying to desactivate the ppas08:26
mvoseb128: aha! what is the bugnumber?08:26
seb128let me find it back08:27
didrocksLaney: that's the spirit! ;)08:27
seb128mvo, the bt matched https://bugs.launchpad.net/ubuntu/+source/ubuntu-release-upgrader/+bug/127828008:27
ubot2Launchpad bug 1278280 in ubuntu-release-upgrader (Ubuntu) "ubuntu-release-upgrader crashes with DistUpgrade/sourceslist.py", line 416, in save : UnicodeEncodeError: 'ascii' codec can't encode characters in position 97-114: ordinal not in range(128)" [High,Confirmed]08:27
seb128mvo, well, apparently bdmurray fixed it in trusty,, but I guess we should SRU the fix to saucy, since people upgrading run the saucy version?08:28
seb128mvo, or is the dist-upgrader updating itself on start?08:28
mvoseb128: thanks, infinity made me work for him, once I finished that I will look at this08:28
seb128mvo, danke08:29
mvoseb128: hm, during a saucy->trusty upgrade the trusty version of the upgrader is used08:29
seb128ok, I was unsure08:29
seb128so the bug is not fixed08:29
seb128well, people said that on the report as well/reopened it08:29
mvoyeah, looks like the fix is either not working or not "enough"08:29
* mvo nods08:29
seb128mvo, the other smaller issue is that g-c-c/g-s-d are not uninstalled on upgrades08:29
seb128not sure if we have a way to say "clean those if nothing depends on them"08:30
mvoseb128: aha, yeah, you mentioned that and I failed to do it, sorry08:30
seb128they are only needed if you have gdm/GNOME installed08:30
seb128no worry08:30
seb128that's really a detail08:30
seb128it's a bit of disk space08:30
didrocksI guess it's just a question of being puzzled on the dash08:30
didrocksbut yeah, as seb128 a detail08:31
mvoseb128: it should be very simple, I put it on my list08:31
seb128mvo, danke08:31
didrockssaid*08:31
mvodidrocks: very simple (hopefully) - u-r-u will just mark it as auto-installed and the cleanup will automatically  clean it08:33
didrocksah nice ;)08:33
jibelseb128, salut, could someone have a look at bug 130351608:48
ubot2Launchpad bug 1303516 in graphite2 (Ubuntu Trusty) "ubiquity crashed with SIGSEGV in _int_malloc()" [High,Triaged] https://launchpad.net/bugs/130351608:48
jibel?08:48
seb128jibel, salut, I'm going to have a look/forward upstream, I don't think we have anyone knowing that stack though08:50
jibelseb128, I understand. It is not nice to see the installer crash on 1rst screen though08:51
seb128indeed not08:51
seb128is there an easier way to reproduce that booting an iso and running the installed in an env that is not set up for debug?08:51
jibelseb128, you can reproduce it directly on your machine. Install ubiquity-frontend-gtk and start ubiquity --greeter08:53
seb128jibel, thanks, that's going to make debugging easier08:53
jibelseb128, it will change the configuration of your locale/keyboard so be ready to revert to something you can use08:54
seb128k, I should be able to manage ;-)08:54
seb128thanks for the notice though08:54
jibel:)08:54
xnoxjibel: yeah, all python people are in montreal at the moment...09:19
tjaaltonunity --reset is apparently deprecated, but manpage still has it?09:24
seb128tjaalton, we welcome patches ;-)09:26
jibelxnox, do you mean they'll do a brainstorm session on this crash ? ;)09:27
tjaaltonseb128: as always.. :)09:28
tjaaltonjust wondering what the replacement might be09:28
seb128tjaalton, not sure why they dropped it, maybe we should just add it back09:29
seb128I had to google for the gsettings reset command this w.e09:30
=== vrruiz_ is now known as rvr
=== tkamppeter__ is now known as tkamppeter
tkamppetermlankhorst, hi09:57
tkamppetermlankhorst, I will report the bug at freedesktop.org now, as product I have chosen "xorg which is the component which I have to choose?09:58
mlankhorstserver10:02
tkamppetermlankhorst, "Server/Input/Core"?10:09
seb128Trevinho, hey, did you see https://code.launchpad.net/~beidl/unity/unity-lockscreen-gestures/+merge/215569 ?10:09
mlankhorstyeah10:09
tkamppetermlankhorst, you tell you have seen something in my log files. Can you tell me which log files? From which bug and which attachment there?10:10
mlankhorst /var/log/Xorg.0.log iirc10:11
tkamppetermlankhorst, thanks, bug submitted as https://bugs.freedesktop.org/show_bug.cgi?id=7742610:16
ubot2Freedesktop bug 77426 in Server/Input/Core "Several touch screen issues, I do not know whether all are caused by X or not" [Normal,New]10:16
mlankhorsttkamppeter: can you look in Xorg.0.log ?10:18
tkamppetermlankhorst, will do, I am simply reproducing the problems, so that appropriate errors will get there.10:22
tkamppetermlankhorst, /var/log/Xorg.0.log attached to the bug.10:28
mlankhorsttkamppeter: odd, I thought I saw mentions of BUG_WARN_MSG(!(event->device_event.flags & TOUCH_POINTER_EMULATED), "Non-emulating touch event\n"); in a touch bug report :/10:31
tkamppetermlankhorst, any other log needed? Or are all my observations bugs in the appropriate apps?10:33
mlankhorstnot sure yet..10:35
=== beidl_ is now known as beidl
didrockshum, I wonder if the notify-osd didn't regress10:59
didrocksI plugged back to 2 monitors10:59
didrocksah no10:59
didrocksthe behavior is just different from what I expected :)10:59
didrocks(it's only on the monitor which has the focused app, like when typing, not where the mouse is)11:00
seb128didrocks, right11:02
didrocksmaking sense :)11:02
seb128didrocks, before this cycle it was always on the primary monitor11:02
didrocksyeah, I was expecting that we went the "where the mouse is" proposal11:03
seb128but yeah, the new behaviour makes more sense11:03
didrocksbut I prefer the real behavior actually11:03
seb128same here, sometime the cursor end up on the other screen while you are just using the keyboard11:03
=== czajkows1i is now known as czajkowski
om26er_question: why are apport dialog' not disabled before the release ?11:17
om26er_they seem to appear all the times, puts a bad impression on the release.11:18
seb128om26er_, wrong channel for that question11:18
seb128try to ask ev on #ubuntu-devel11:18
om26er_hmm, I thought it affecting the desktop experience ;) will try there.11:19
seb128om26er_, what sort of issues do you have? they shouldn't be that frequent, it might because you work in qa and trigger bugs often, that's less of an issue for normal users11:19
om26er_seb128, I have not enabled anything, whenever I log in I am greeted with apport 'report a problem'11:20
om26er_and they are like 3 dialogs one over the other11:20
seb128om26er_, can you ls /var/crash and share the details/bug numbers?11:20
om26er_seb128, ok, lightdm,unity-greeter and gvfsd-mtp look interesting one's11:22
om26er_I'll report that by clearing everything there and reboot .11:22
seb128om26er_, ok11:23
seb128om26er_, the bottom line is that if we don't collect reports, we don't know what to fix11:23
seb128om26er_, the best solution is to fix the bugs so apport stops prompting, not to keep the bugs and just don't tell the user about those11:23
seb128especially for user apps, when they close it's better to tell the user what happened11:23
om26er_seb128, sure, I was expecting the report a problem thing to be disabled because it has haunted people in the past (talking askubuntu etc)11:24
seb128but yeah, the tradeoff is the quality perception11:24
seb128we disable the reports to launchpad11:24
seb128not the errors collecting11:24
seb128not that other OSes do the same11:24
seb128for what it's worth I don't see apport more than 1 a week on my system (out of some issues with dev tools, like system-dbus-image hitting exceptions when I run autopilot tests)11:25
seb128so your 3 prompts at every login means your machine has specific issues11:25
om26er_brb11:26
* seb128 removes ubiquity11:28
seb128Laney, thanks for taking on that bug, best is probably to report it to upstream harfbuzz11:29
seb128I think it's a good starting point11:29
Laneyyeah it's hard11:29
Laneyvalgrind doesn't work properly on it11:29
Laneyis that language Thai?11:29
seb128Laney, not sure, did you figure it out?11:39
Laneynope11:39
Laneyreporting it upstream11:39
=== maclin__ is now known as maclin
=== MacSlow is now known as MacSlow|lunch
seb128Laney, it tries to apply a "mm" keyboard layout, which is "Burmese" in /usr/share/X11/xkb/rules/evdev.xml11:46
seb128Laney, so yeah, looks like Thai could be right11:47
Laneywhy not burmese?11:50
Laneylooks like that11:50
seb128is that a locale?11:51
seb128I didn't find a language-pack-gnome-mm :p11:51
seb128oh, it's "my"11:51
Laneyyeah, myanmar11:52
seb128Gtk:ERROR:/build/buildd/gtk+3.0-3.10.8/./gtk/gtktextsegment.c:195:_gtk_char_segment_new: assertion failed: (gtk_text_byte_begins_utf8_char (text))11:52
seb128gedit doesn't like ubiquity/ubiquity/my.po11:53
seb128I've that by selecting text in there11:53
seb128could be a better/easier testcase than ubiquity11:53
seb128if that's the same issue/underlining bug11:53
Laneycan't make that one happen11:57
=== psivaa is now known as psivaa-lunch
seb128Laney, weird, happens every time for me, I just left click on the top of the file and move a bit the cursor over the text to select/deselect it12:02
Laneynope12:03
seb128k12:03
seb128well, I'm going to report that on gtk since that's where the error is from the backtrace12:04
=== alan_g is now known as alan_g|lunch
TrevinhoLaney: hey, I've just noticed that a fix that was merged some weeks ago (https://code.launchpad.net/~3v1n0/unity/alpha-windows-shadows-fix/+merge/213352) got deleted by the last merge... I guess there's nothing to do but an SRU, right? :(12:25
LaneyProbably best, I guess you're going to do an early SRU12:25
Laneyhow did that happen?12:26
seb128Trevinho, hey, did you see my ping earlier?12:26
TrevinhoLaney: well, one of the branches for fixing the lockscreen (not mine) was probably not well merged with trunk12:26
Trevinhoseb128: ops, no12:26
seb128Trevinho, hey, did you see https://code.launchpad.net/~beidl/unity/unity-lockscreen-gestures/+merge/215569 ?12:26
seb128Trevinho, I want to land that today, it's a security issue/should be fixed for release12:26
seb128Trevinho, can you review that change?12:27
Trevinhoseb128: yes, I will check that today...12:27
Trevinhoseb128: ok12:27
Trevinhoseb128: any other high-prio thing you can land today?12:27
seb128Trevinho, thanks, if you can do now/not to late that would be nice, they might not wait too much12:27
seb128Trevinho, I doubt it, I'm already going to be happy if we manage to get that in, not sure if they plan to respin isos again12:27
* Trevinho if his shoulder strain make him write...12:27
seb128:-(12:28
seb128what did you do?12:28
Trevinhoseb128: no idea... during the night something weird happened... Maybe I just slept in the wrong position, no ieda... this morning I wasn't able to move in bed as well :o12:28
seb128urg12:28
seb128Trevinho, get better!12:28
beidlTrevinho, wouldn't it be better if hiding the launchers parent window would be taken out for now? it causes the icons as well as the hint shadow not to show up in autohide mode.12:29
Trevinhoit's simliar to the stiff neck, but in the shoujlders...12:29
beidlTrevinho, get well soon!12:29
=== MacSlow|lunch is now known as MacSlow
Trevinhobeidl: mh, hint shadow? it was showing when I proposed it... hmmhm let me check12:30
beidlTrevinho when you drag the mouse pointer to the left edge in autohide12:30
Trevinhobeidl: right it is now showing now, but couldn't that after the last change you proposed?12:30
Trevinhobeidl: not point anything, I just recall it was working :)12:31
beidlTrevinho that change was just for allowing the launcher-drag-in-animation to work when using the 4 finger gesture12:31
Trevinhobeidl: let me see12:32
qenghoHi all. I have a new chromium-browser, which is in universe. What are the chances I can get it into 14.04?12:32
qenghoMany CVEs fixed, as usual.12:33
Laneyqengho: It's on mythbuntu and kylin. Go check in #ubuntu-release12:34
seb128qengho, hey, did you hear about issues where chromium was opening a new instance with your session every time you click on an url to open it?12:34
seb128qengho, rather than opening a tab in the current instance12:35
seb128qengho, the new version from your ppa seems also quite slow (well, I didn't upgrade it since I installed it a week ago so maybe that's resolved since)12:35
qenghoseb128: Was this my testing PPA?12:35
seb128yes12:35
qenghoseb128: I had the testing version use a temporary directory, instead of clobbering a profile you might value. That would explain xdg-open starting new windows.12:37
seb128qengho, k, so should not happen with the archive upload? good!12:37
seb128didrocks, ^12:37
didrocksoh phew!12:37
didrocksthanks qengho, seb128 :)12:37
seb128yw!12:38
qenghodidrocks, seb128, I'd be happy if you try the version in my update-staging PPA to verify it's gone.  https://launchpad.net/~canonical-chromium-builds/+archive/stage12:39
seb128qengho, let me try that version12:39
MalsasaHello. My HUD on Ubuntu 12.04 doesn't work anymore. When I tap ALT, Unity opens menu of focused app. Not the HUD anymore. I can't do ALT+TAB switcher at all. Any idea for me? Thank you.12:47
mvo_seb128: no luck so far reproducing the utf8 bug you mentioned earlier12:55
seb128mvo_, we didn't have luck reproducing it either at the install party, we tried to get debug output but couldn't get it again12:56
seb128mvo_, did you try to put non ascii chars in a sources.list.d source that is going to be commented?12:57
seb128mvo_, or maybe rather than "non ascii" using "non utf8 valid"?12:57
mvo_seb128: yeah, I tried a bunch of different things, I will try harder13:00
seb128mvo_, don't spend too much effort on it, e.u.c doesn't have it ranked in the top issues so it might not be that important13:01
rickspencer3seb128, jasoncwarner what's the word on the street? everything getting settled for releasing 14.04 this week?13:03
seb128rickspencer3, hey, things look alright I think, still a few issues/things we would like to land but that's usual13:05
=== alan_g|lunch is now known as alan_g
rickspencer3seb128 coolio13:07
rickspencer3seb128, the desktop feels very solid to me13:07
rickspencer3I've got it on 2 very different computers, and it's working really well13:07
seb128thanks, I'm quite happy with the state of things as well13:08
seb128we still have stuff to fix/polish to make it even better, but that's always the case13:08
seb128we have SRUs for that as well ;-)13:08
seb128rickspencer3, Didier, I and others were at a small opensource event in Lyon this w.e, we did some 13.10 -> 14.04 upgrade for users who had brought their laptops, things went smoothly in most case and we only had happy users ;-)13:09
rickspencer3\o/13:10
* rickspencer3 fumes about not being invited13:10
=== psivaa-lunch is now known as psivaa
seb128sorry about that, next time ;-)13:11
rickspencer3:)13:12
didrocksrickspencer3: come to Lyon! it will be easier :)13:20
rickspencer3didrocks, sounds good, I can live there as did your belle-soeur13:21
seb128lol13:22
didrocksseb128: see, management tracks their employees! Time to go on strike! ;)13:22
seb128didrocks, yeah, just make sure don't give him my room though!13:22
* seb128 was first to claim it13:22
didrocksheh :)13:22
seb128didrocks, is it sunny enough today to use a strike day?13:23
* seb128 looks outside13:23
seb128seems like it could be one of those days :p13:23
didrocksseb128: yeah, but it's a little bit too late now :p13:24
didrocksbetter to start in the morning13:24
didrocksand tomorrow is sunny as well13:24
* seb128 notes that down13:24
didrockslet's do from torromow until… thursday :p13:24
didrockswell, you have your friday off13:24
seb128it let me time to do a banner on how the NSA^Rick spies on us13:24
didrocksMr "I claim to be French but in fact, I'm half-German" :)13:24
seb128oh, that's right ;-)13:24
seb128"good friday" is good ;-)13:25
didrocksyeah yeah…13:26
* didrocks starts doing a second strike banner13:26
seb128:-(13:26
Laneyfour day weekend this week :-)13:26
seb128didrocks, come to Metz and claim it's a day off for you! ;-)13:26
didrocksLaney: you are German as well?13:26
LaneyWenn ich weniger arbeiten müss, JA!13:28
didrocks:)13:28
Trevinhobeidl: that shadow thing is not related to the window not being shown it seems...13:32
LaneySweetshark: libreoffice autopkgtest went boom13:33
beidlTrevinho, how did you test? I removed every _parent->ShowWindow() call and it works.13:33
Trevinhobeidl: or, wait... sorry I didn't install correctly the plugin :(13:34
beidlTrevinho classic :D also, when minimizing a window, the corresponding icon should pop up for a second and hide again. Reverting the launcher window hide fixes that as well.13:36
beidl* parent_->ShowWindow rather13:36
tkamppeterseb128, I have a problem with bug 1306344. One could solve it by installing the hplip-gui binary package. Could we ship Trusty with this package? Or do we still have the problem that dependencies make Trusty too big then?13:37
ubot2Launchpad bug 1306344 in hplip (Ubuntu) "hp-setup crashes when adding new printer because hp-sendfax is not found" [Undecided,Incomplete] https://launchpad.net/bugs/130634413:37
Trevinhobeidl: yeah, the same for urgent icons13:37
Trevinhobeidl: ok, +1 on reverting that13:37
=== alan_g is now known as alan_g|afk
Trevinhobeidl: do you want to do a branch for that or should I do?13:37
beidlTrevinho I could push it to the branch related to showing the switcher on lock screen using gestures13:38
Trevinhobeidl: it actually saves us a lot of redraws, but the fix should probably add more edge-cases than the real gain13:38
Trevinhobeidl: ok13:38
beidlTrevinho: I figured that, but I guess thats SRU material until all edge cases are found13:38
Trevinhoit's weird btw I was pretty sure I tested these cases... :o13:38
tkamppeterseb128, would be great if we could ship hplip-gui as in Trusty + 1 we could lift the awkward binary package splitting then.13:39
seb128tkamppeter, it's late now to do changes for trusty...13:39
tkamppeterseb128, or should I then make an SRU which makes hplip pulling in hplip-gui?13:40
seb128tkamppeter, try asking on #ubuntu-release if they would be ok with that13:40
SweetsharkLaney: thx checking13:44
LaneyLooks like some kind of test problem rather than an actual issue13:45
Laneysadly13:45
seb128bregma, hey13:45
bregmaseb128, yo13:45
seb128bregma, could you get https://code.launchpad.net/~beidl/unity/unity-lockscreen-gestures/+merge/215569 reviewed?13:45
seb128bregma, that seems like a release blocker13:45
bregmaalready working on it13:46
seb128bregma, I tried to put it in a silo to get testing/an head start, but ppc64el build fails in tests :/13:46
Trevinhoseb128: what failure? there are some tests that seems to fail there13:46
bregmappc64el has been misbehaving in tests lately, usually it succeeds when you rerun just that build13:46
bregmaroll the dice on that one13:46
seb128cjwatson retried it for me13:47
seb128same issue13:47
seb128Trevinho, https://launchpad.net/~ci-train-ppa-service/+archive/landing-008/+build/5908000/+files/buildlog_ubuntu-trusty-ppc64el.unity_7.2.0%2B14.04.20140414-0ubuntu1_FAILEDTOBUILD.txt.gz13:47
SweetsharkLaney: sadly? of course, hickups shouldnt happen, but I much prefer this to the tests pointing to an actual problem at this point of the cycle ;)13:47
seb128us13:47
bregmapreviously it was one of the tests that ran a fake D-Bus service, and the service failed to start for some reason (not logged)13:47
Trevinhoseb128: lp lost something?13:47
seb128Trevinho, the build has been restarted, which wiped the log, let's see13:47
bregmaTrevinho, yes, it lost the diff for the MP, beidl had to cancel it and submit a new MP13:48
Trevinhooh, ok13:49
bregma◔_◔13:49
SweetsharkLaney: hmm, even the log says "dsc0t-junit-subsequentcheck PASS" at the end, so putting that in the https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcQsLL60iOaDgX0D1th2iRRsIHBYgs-WUEgVR-ChlEae4ftxU8sweA bucket for now ...13:50
MalsasaHello. My HUD on Ubuntu 12.04 doesn't work anymore. When I tap ALT, Unity opens menu of focused app, not the HUD. Any idea for me? My complete quesion was here http://askubuntu.com/questions/447587/alt-key-for-hud-doesnt-work. Thank you.13:51
LaneySweetshark: are you looking at the same thing as me? https://jenkins.qa.ubuntu.com/view/Trusty/view/AutoPkgTest/job/trusty-adt-libreoffice/336/ARCH=amd64,label=adt/consoleFull13:51
beidlwhen I noticed launchpad not scanning my branch for 24 hours, I was all like "zomg, I accidentally the whole thing"13:51
bregmabeidl, I just assumed my browser was broken13:53
bregmabecause why should it be any different?13:53
=== alan_g|afk is now known as alan_g
SweetsharkLaney: hmmm, no. Did I end up in a 'last successful artifacts' thing or somesuch. Indeed that doesnt look good -- seems like "something" deleted/modified/whatever the copy of the source in /tmp, while the tests where running on it ...13:57
SweetsharkLaney: there is one very weird line 512 in the syslog ...13:59
beidlbregma, I had connection issues with my ISP that day, might be an incomplete push that caused that error14:02
LaneySweetshark: mmm, that's in the passing one too14:03
Laneyseb128: looks like it worked that time, but they are going to slip the other fix in to this upload14:04
seb128Laney, great14:05
SweetsharkLaney: when Im finished with this bug, I will run that in a local VM to see whats going on.14:05
Laneyre-siloing appreciated14:06
bregmaseb128, I'm going to add another (approved) MP to the Unity silo, if you don't mind, now that ppc64el has successfully built14:07
Laneyjust said that ;-)14:07
beidlTrevinho, the remaining nasty bug that I could think of is the launcher not reappearing if a edge-hit causes the launcher to hide.14:09
seb128bregma, sure, the silo is all yours ;-)14:15
bregmaparty on14:15
seb128bregma, I just started it to have an head start14:15
mvo_seb128: it was gnome-settings-daemon and gnome-control-center, correct?14:18
mvo_seb128: that should be autoremoved if possible?14:18
seb128mvo_, yes14:19
mvo_thanks14:23
=== alan_g is now known as alan_g|tea
=== alan_g|tea is now known as alan_g
Trevinhobeidl: yeah, that's indeed a bug we should check... you say the one on dnd, right?15:04
beidlTrevinho: exactly15:04
Trevinhoseb128: but... is it just me or since last few updates the indicator-sound is not showing up?15:05
Trevinhoseb128: and loading it with indicator-tool brings me an empty indicator15:05
seb128Trevinho, just you15:06
Trevinholarsu: ^ any idea how debug that?15:06
seb128Trevinho, what theme do you use?15:06
Trevinhoseb128: ambiance15:06
seb128Trevinho, what do you have in .cache/upstart/indicator-sound-service.log and unity-panel-service.log?15:06
Trevinhoseb128: ok checking... fyi the service runs btw15:06
seb128can you open the menu?15:07
Trevinhoseb128: on indicator-sound.log (process:22071): GLib-CRITICAL **: g_variant_get_uint32: assertion 'g_variant_is_of_type (value, G_VARIANT_TYPE_UINT32)' faile15:07
Trevinhoseb128: nope15:07
seb128like if you click on another indicator and do left/right?15:07
seb128Trevinho, if you do system-settings -> sound, is the "show" checkbox at the bottom checkeD?15:07
Trevinhoseb128: ops... :D15:08
seb128;-)15:08
Trevinhoseb128: how did it gets clicked :P15:08
Trevinhoseb128: maybe just a wrong action I hope15:08
seb128Trevinho, the option is age old but we had a bug that it was not respected in 13.1015:08
Trevinhoseb128: ah... ok. sorry for bothering15:09
seb128so maybe you had it from while back and it started working again when we fixed the bug15:09
Trevinhoand thanks...15:09
Trevinhomh, might be15:09
seb128Trevinho, no worry, with all the pinging I did in your direction recently you can ping me for a while before we are even ;-)15:09
Trevinhoseb128: oh, no worries... I will take few days :D15:10
seb128hehe15:10
Trevinhomlankhorst:15:10
Trevinhohey15:10
Trevinhojust for information... is there any plan for the future of 14.04 to support DRI3 drivers?15:11
larsuTrevinho: your issue seems resolved now?15:22
Trevinhoseb128: yeah, sorry for the ping :/ i've been just too worried, and it was my fault :(15:23
seb128larsu, ^15:23
Trevinhoseb128: that was for a free ping_count-- for you :D (sorry!)15:24
seb128haha15:24
seb128;-)15:24
larsuTrevinho: no worries15:24
larsuseb128: thanks :)15:24
seb128yw!15:25
beidlTrevinho, had to leave my laptop, but a wild guess: maybe the hide animation has to be reversed as soon as DnD is stopped15:38
Trevinhobeidl: mh, it might be the case... I need to look further on it15:50
seb128qengho, new chromium from your stage ppa seems fine to me15:51
qenghoseb128: thank you.15:52
seb128yw!15:52
ali1234tedg: now that we implemented upstart indicators in xfce, we need a way to blacklist them15:52
ali1234specifically appmenu, since it doesn't work, and if the backend is started it steals all your menus whether the indicator is visible or not15:53
ali1234also -session since it also doesn't work15:53
tedgali1234, Well appmenu isn't a service, you can just avoid loading the .so15:53
ali1234apparently you can't15:53
seb128Trevinho, https://launchpadlibrarian.net/172796846/buildlog_ubuntu-trusty-ppc64el.unity_7.2.0%2B14.04.20140414.1-0ubuntu1_FAILEDTOBUILD.txt.gz15:54
* tedg is really sure about this :-)15:54
seb128Trevinho, the pp64el issue15:54
ali1234if it is installed, it steals the menus, regardless of whether you actually want it loaded or not15:54
seb128bregma, that one ^ needs a retry as well15:54
qenghoseb128: I don't know what then next step is.  infinit-y seemed to say "sounds like a good idea". Should I do something else?15:54
Trevinhoseb128: yeah that's the same that shown up also last time...15:54
seb128qengho, get it uploaded15:54
Trevinhoseb128: and really I've not much clue what might be other than a timing issue15:54
Trevinhoseb128: mhr3 fixed these tests some time ago15:54
ali1234tedg: so what steps must we take to "avoid" loading the .so?15:55
tedgali1234, So that's not an upstart thing, that's in /etc/X11/Xsession.d/80appmenu15:55
tedgali1234, You need an attente15:55
* mhr3 hides15:55
qenghoseb128: yeah, but by whom?15:55
tedgali1234, I'm guessing that script needs to be smarter, it's only checking to see if the file exists. Which is clearly not enough.15:56
qenghoI made noise in #ubuntu-release already.15:56
ali1234it used to be that it wouldn't do anything if UBUNTU_MENUPROXY was not set15:56
seb128qengho, your usual sponsors? I can do it if you want15:56
ali1234well, i'm fairly sure we don't set that in xfce...15:56
tedgali1234, It's set by lightdm using that script. All the scripts in that directory run before the user session.15:57
ali1234if it's not a service, how does it get the menus?15:57
qenghoseb128: yes, please. I think CCC is busy with other stuff.15:57
tedgali1234, All GTK programs load a GTK module, that hides and exports the menus.15:58
ali1234exports them where?15:58
tedgali1234, On the other side of things the .so implemented everything needed to recieve the menus. It's a little different from the others in that the process loading it does a lot more.15:58
ali1234via what mechanism is the menu transfered from the application to the thing that runs in the panel?15:58
tedgdbus15:58
ali1234so there is a service that listens on dbus then?15:59
tedgYes, who ever loads the .so15:59
ali1234so in other words it is a service...15:59
tedgYes, but no one of the indicator services15:59
tedgnot15:59
ali1234so it's more like the mpris stuff? where each app exports it's own menus, and then the indicator goes and finds them all?16:01
mhr3seb128, Trevinho, weird: (test-gtest-service:19151): GLib-GIO-WARNING **: Error releasing name com.canonical.Unity.Scope.testscope1.T29883340183229126: Timeout was reached16:01
qenghoseb128: Also, if you want me to not have to bother you for this, you may endorse my PPU application in the next 30 hours.  https://wiki.ubuntu.com/ChadMiller/DeveloperApplication16:01
mhr3seb128, dbus blew up?16:01
seb128mhr3, that could be, not sure16:01
seb128qengho, noted down16:02
seb128qengho, the version to sponsor is from your ppa?16:02
Trevinhomhr3: fyi the tests just seems to be unstable... sometime the go, some others don't16:02
seb128qengho, 34.0.1847.116-0ubuntu1~pkg1006  is the correct version? you keep the ~pkg for archive uploads?16:02
ali1234tedg: the problem here is that by making everything work off an environment variable set by lightdm, you are asking lightdm to predict the future16:02
mhr3Trevinho, everywhere? or a particular arch?16:02
qenghoseb128: yes. It takes hours to rebuild.  I like uploading exactly what people have tested.16:03
ali1234there is no way it can know if the user is going to load indicator-appmenu or not16:03
seb128qengho, k16:03
tedgali1234, Are you underestimating the abilities of LightDM? ;-)16:03
tedgali1234, I think that the script should probably just set the variable based on which desktop is being loaded.16:03
ali1234tedg: it doesn't have permission to access the settings file16:03
=== om26er_ is now known as om26er|food
ali1234so even if it was able to parse xfce settings, it still would not work16:03
tedgSure but you could force it into the upstart environment table if it was enabled.16:04
tedgHeck, we should just do that…16:04
ali1234yeah, or we could add more crap to accounts service...16:04
tedgThen it could use dconf16:05
ali1234a larger problem is we allow the user to toggle on/off indicators at run time16:05
seb128qengho, hum, do you usually ppa source copy?16:05
tedgYeah, and the Upstart job could just set the variable based on a setting.16:06
tedgIt would run on desktop-start, but could be run anytime as needed.16:06
seb128Laney, there? to upload chromium from https://launchpad.net/~canonical-chromium-builds/+archive/stage/+packages to trusty do you think we can just ppa source copy it?16:06
Trevinhomhr3: only on ppc64el16:06
tedgWould be nice to get another thing out of Xsession16:06
ali1234sounds more sensible16:06
tedgWe need it to die sooner rather than later.16:06
Laneyseb128: is that what they normally do?16:06
seb128Laney, no idea16:07
seb128chrisccoulson, there? how do you sponsor chromium uploads usually?16:07
ali1234at the moment appmenu doesn't even work in xfce - it crashes both the app and the desktop16:07
LaneyLooks like it16:07
ali1234so it doesn't really matter how it gets fixed16:07
mhr3Trevinho, then i'll say that dbus on ppc64 is broken :)16:07
ali1234(in the short term)16:07
ali1234some people are trying to make it work though16:07
mhr3Trevinho, or something dbus related, but clearly not my code ;P16:08
Laneynot clearly16:08
Laneyyou should try to reproduce on the porter machine first16:08
seb128Laney, "copy-package --ppa=canonical-chromium-builds --ppa-name=stage -s trusty --to-primary --to-suite trusty-proposed chromium-browser" ?16:08
Laneyseb128: include binaries16:09
seb128Laney, k16:09
qenghoseb128: I see uploads pretty quickly. It has to be something like that.16:09
seb128qengho, it's chrisccoulson who usually sponsor those right?16:09
qenghoseb128: yes.16:10
seb128k16:10
seb128he doesn't seem to be around16:10
seb128let's try that ppa copy then16:10
LaneyI looked at https://launchpad.net/~canonical-chromium-builds/+archive/stage/+build/5815872 which is a copy from the ppa with binaries16:11
seb128k, it makes sense if the ppa is configured for security updates16:11
=== om26er|food is now known as om26er
qenghoseb128: et merci beaucoup.16:14
seb128qengho, de rien ! ;-)16:18
ali1234tedg: i'm looking on a saucy machine and there is no 80appmenu16:19
tedgali1234, Correct, it changed with teh change from appmenu-gtk to unity-gtk-module16:20
ali1234i also don't see the string "MENUPROXY" anywhere in /etc16:21
=== gatox is now known as gatox_lunch
ali1234i just installed it on trusty and that file still doesn't exist16:24
seb128bregma, Trevinho: the unity in the silo looks fine to me (just from basic user testing), I'm calling it a day in a bit but feel free to publish/nag #ubuntu-release when you feel it's ready16:25
Trevinhoseb128: thanks16:25
seb128yw!16:25
ali1234so which package does /etc/X11/Xsession.d/80appmenu belong to?16:27
ali1234unity-gtk-module isn't a package16:32
ali1234unity-gtk(2|3)-module neither contain 80appmenu16:33
tedgWait, I had it backwards :-/16:33
ali1234appmenu-gtk also isn't a package16:33
tedgThat's an appmenu-gtk thing, it is just left over on this machine.16:33
* tedg hates on /etc for a moment16:34
ali1234https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/874943 <- lol16:34
ubot2Launchpad bug 874943 in gnome-shell (Ubuntu) "Appmenu stays visible behind gnome-shell panel" [Undecided,New]16:34
ali1234same bug, except 2.5 years old16:35
tedgali1234, So yes, we already moved it to an Upstart job, unity-gtk-module16:35
ali1234so how do we kick it out of xubuntu then?16:36
tedgali1234, That job should be smarter16:36
dobeyappmenu-gtk - Export GTK menus over DBus16:36
dobeyappmenu-gtk3 - Export GTK menus over DBus16:36
dobeyali1234: ^^ those look like packages to me16:36
ali1234!info apmenu-gtk trusty16:36
ubot2'trusty' is not a valid distribution:16:36
tedgdobey, Those are probably left over if you've upgraded16:36
ali1234whoops, well, they don't exist here in my package manager16:37
ali1234bbiab16:37
tedgali1234, So I'd say check something in the pre-start of that job, what exactly is something you'll need to talk with attente about.16:37
dobeyoh16:38
dobeyugh, something causing extremely slow rendering issues in current trusty :(16:39
dobeyor maybe i just need a reboot16:40
tedgYup, seems to be after about an hour for me.16:40
=== gatox_lunch is now known as gatox
=== alan_g is now known as alan_g|EOD
ali1234i still don;t get it. that job doesn't do anything unless UBUNTU_MENUPROXY is already set17:11
ali1234why is this even in unity-gtk-module anyway?17:15
sarnoldwhen scrolling the list of online accounts in the unity control center, more than one provider is selected at once; any guesses where I should file the bug? unity? gtk? unity-control-center?17:18
sarnold(I suspect it is only a visual glitch)17:18
beidlTrevinho: found the issue and fixed it. where to push? :)17:45
Trevinhobeidl: cool! :) new branch please17:45
beidlTrevinho: based on trunk?17:45
Trevinhobeidl: yes, or your previous branch is fine anyway... as it will be trunk soon17:46
beidlperfect17:46
Trevinhobeidl: or just set it as a prerequisite on the merge-proposal17:46
Trevinhobeidl: so it won't show the diff of that17:46
beidlTrevinho: if the launcher hides, and I drag the icon away *without* dropping it, should the launcher reappear? currently the launcher reappears only after a new drag start.17:51
Trevinhobeidl: yes, the launcher should hide only if you try to pass over it.. (in multi-monitor you want to move on the other screen for example)17:52
xclaesseI'm writing a gtk app, basically a GtkStack and GtkListBox, the background is all black with ubuntu theme17:57
xclaesseis there known ubuntu gtk theme issues?17:58
beidlTrevinho: is there an event that gets emitted whenever the mouse is moved? all dnd related events except resets won't get called as soon as the launcher is hidden18:11
Trevinhobeidl: mh, right... In theory the dnd collection window might get that, but it's also hidden as soon as it gets something18:13
Trevinhobeidl: so... mhmh, a possibility might make that to show up again in that case18:14
Trevinhobeidl: calling collect on it should be enough18:14
beidlTrevinho: what needs to happen is reversing the animation and setting the DND_PUSHED_OFF quirk to false. currently that only happens when dnd is either complete or aborted18:15
beidlTrevinho: but if the launcher should reappear as soon as mouse.x >= geometry.width18:16
beidlTrevinho: I'd need to find out where to do that. I assume RenderArgs isn't going to get called if the launcher is hidden, right?18:17
Trevinhobeidl: it depeneds, it might... it gets called each redraw... so since the basewindow is always there it happens also if not shown18:18
Trevinhobeidl: that's why I added that change that you reverted... As I didn't want to spend cycles when not needed, it should be done more carefully18:19
beidlTrevinho: yeah, those stupid corner cases :)18:19
Trevinhobeidl: indeed... unfortunately there are so many that is quite easy to hit one :o18:23
beidlTrevinho: I'd basically have the UnityScreen inform the launcher about every single mouse movement. I'm not sure if that's worth it.18:40
Trevinhobeidl: mh, it's probably better to avoid that18:40
beidlTrevinho: should I keep it the way it is now? only make the launcher reappear on DND ending?18:41
Trevinhobeidl: ok, that's enough for now... if there aren't cleaner solution it's already an improvement :)18:42
ali1234tedg, attente: i opened bug 1307657 re appmenus.18:53
ubot2Launchpad bug 1307657 in unity-gtk-module (Ubuntu) "UBUNTU_MENUPROXY should not be set in Xfce" [Undecided,Confirmed] https://launchpad.net/bugs/130765718:53
=== FJKong is now known as FJKong_afk
* tedg commented19:02
ali1234thanks19:03
beidlTrevinho: do you think there's still enough time to fix bug 1163041 ? :)19:04
ubot2Launchpad bug 1163041 in unity (Ubuntu) "Inconsistent behaviour with minimized Trash windows" [Low,Triaged] https://launchpad.net/bugs/116304119:04
Trevinhobeidl: I hope so19:04
Trevinhobeidl: it's on my list19:04
Trevinhobeidl: we can make as sru, as the pieces area ll there, we only have to fix unity19:04
beidlTrevinho: how would you go about fixing it? actually make a trash/device window "fly" into its corresponding launcher window on minimize?19:05
beidllauncher icon*19:06
Trevinhobeidl: I'm about to go now, but I can explain this with some more details tomorrow (or later) if you want19:08
beidlTrevinho: great! :)19:08
beidlTrevinho: my quassel is always-on anyways :D19:08
Trevinhobeidl: btw basically we should make nautilus a case of Application that has sub-windows, and these might be detached from the main app if are "special"19:08
beidlTrevinho: ok, makes sense19:09
Trevinhobeidl: this is an old diff I did to begin on that... but just to give the idea http://pastebin.ubuntu.com/7251294/19:10
beidlTrevinho: looking into it19:10
Trevinhobeidl: see if you can get what I meant.. Need to go now19:10
beidlTrevinho: alright, have a nice day!19:11
Trevinhobeidl: thanks ;)19:11
=== Wellark_ is now known as Wellark
=== marrusl is now known as marrusl_afk
=== marrusl_afk is now known as marrusl

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!