jasoncwarner_hey bryceh or RAOF, around?04:26
RAOFjasoncwarner_: Yup, what's up?04:38
jasoncwarner_hey RAOF remember that mouse reverse scrolling bug?04:38
jasoncwarner_(apple mighty mouse thing)?04:38
RAOFWith the horizontal scrolling?04:39
jasoncwarner_it's back...wondering if my custom config is messing up the upstream fix04:39
pittiGood morning05:10
brycehjason__, tried removing your custom config?05:52
rickspencer3pitti, what's the word on the street about beta2?06:10
rickspencer3(also, I was woken up by a bird very early this morning, I kinda blame you guys for this)06:11
pittirickspencer3: it's out :) haven't heard about catastrophes yet06:11
rickspencer3"no catastrophes" is as good as it gets sometimes ;)06:11
pittirickspencer3: pah -- I've been for about two weeks now, by a bird that sounds like an alarm clock06:11
rickspencer3pitti, I'm actually reading good things about beta2 on reddit06:13
rickspencer3comments are quite positive about both robustness and Unity06:13
pittinothing on heise.de yet06:13
rickspencer3pitti, no archive probs?06:14
rickspencer3I thought we were supposed to have a crush of uploads and a mess to untangle today06:14
pittirickspencer3: if you want some, we'll need to reupload LibO06:15
pittirickspencer3: our -proposed builds failed :/06:15
pittibecause of that LP bug again06:15
pittiso we need to build in precise directly06:15
pittirickspencer3: but it's in trunk now, should be rolled out next week06:15
pittifrom then on we'll be able to use -proposed all the time06:15
rickspencer3pitti, can Sweetshark just fix LO once and for all next cycle06:15
pittirickspencer3: not Sweetshark's fault this time06:16
rickspencer3pitti, oh, I thoguht it was never his fault06:16
rickspencer3I thought the problem was it has a monolythic build structure06:16
pittiit's just the 4 h on amd64 vs. 20 h on armhf difference06:16
rickspencer3I was not taking Sweetshark'06:16
rickspencer3s name in vain !06:17
rickspencer3I remember we had discussed years ago refactoring it into a more modular build structure, so we could rebuild only parts that changed, etc...06:17
pittiright, but upstream went the opposite direction06:18
pittithey merged all their per-project git trees into one06:18
robert_ancellpitti, are you planning on updating pygobject and libgnomekbd?06:19
didrocksgood morning06:19
pittirobert_ancell: yes, I wanted to sync it after the freeze (pygobject)06:20
pittirobert_ancell: and I'm happy to look into libgnomekbd, too; I hope we can just sync as well, I might need to make some fixes to Debian06:20
pittibonjour didrocks06:20
rickspencer3hey didrocks06:21
didrocksguten morgen pitti, how are you?06:21
robert_ancellpitti, I had a quick look and there's more than trivial changes.  I think you had your name on it somewhere so I thought I'd leave it to you if you are most familair06:21
didrocksbonjour rickspencer3 ;)06:21
rickspencer3didrocks, I'm reading nice things about unity this morning06:21
didrocksrickspencer3: nice blog post!06:21
rickspencer3good evening robert_ancell, btw06:21
didrocksrickspencer3: ah?06:21
robert_ancellrickspencer3, hey06:21
didrocks(nice btw ;))06:21
rickspencer3didrocks, I think the combination of all around quality on the desktop, and feature completion in Unity has come together very noticably for beta206:22
rickspencer3didrocks, what does the French community think?06:22
didrocksrickspencer3: they really love precise. Quite bored by the stability (they like when things broke in the unstable forum) even ;) So really, there is the feeling that the Quality really raised compared to the feedback 6 months ago06:23
didrocksrickspencer3: they are still some concerns about some features/design, but the overall impression on Unity is really positive06:24
pittinow we just need to fix that effing HUD to not come up all the time again06:24
pittibut yes, by and large it's really good06:24
rickspencer3pitti, I;m starting to agree with you about HUD invocation06:24
didrockspitti: you still have that in 3D? I know it's an issue on 2D06:24
didrocksbut not sure in 3D06:24
rickspencer3this does not happen to me, but I read about it06:24
RAOFI've not been noticing accidentally bringing up the HUD.06:24
pittididrocks: it was fixed in 5.6, then re-broken in 5.806:24
didrocksrickspencer3: it's an executive decision about the shortcut, we tried to make design coming back on this decision one month ago when all the issues arised06:25
pittiat least it's configurable in 3D06:25
didrockspitti: do you have a test case?06:25
pittibut we really need to make it configurable in 2D, too06:25
didrockspitti: there is really no step by step reproducer in 3D for accidental reveal since 5.606:25
rickspencer3didrocks, by executive decision, do you mean it's was sabdfl decided?06:26
didrocksrickspencer3: indeed06:26
pittiI'll try to come up with one; it's erratic, but happens all the time here06:26
robert_ancellrickspencer3, that's how I read that :)06:26
didrockspitti: please do ;)06:26
rickspencer3so is the bug simply that 2d needs to work like 3d?06:26
didrockspitti: I'm using alt heavily in weechat, that didn't happen to me since 5.606:26
* rickspencer3 should use 2d more06:26
rickspencer3I actually used the hud in gimp yesterday, it is very useful there06:27
didrocksrickspencer3: well, not "simple" though, as the hack is in compiz and related to the fact that unity and compiz are in the same process06:27
didrocksso can't do that in 2d as metacity and unity-2d are 2 processes06:27
rickspencer3didrocks, but from a user perspective, it should work the same?06:27
didrocksrickspencer3: right, from a user perspective, apart from pitti's case who seem to still have a bad test case06:28
didrocks(in 3D)06:28
didrocksjust telling that fixing this in compiz took approximately a full week (maybe a little bit more)06:28
didrocksso not trivial at all06:28
didrockswill talk again with the 2D guys today06:29
didrocksto know where we are standing at06:29
rickspencer3didrocks, is there a specific bug report on this?06:29
=== tkamppeter_ is now known as tkamppeter
didrocksrickspencer3: there were quite a lot in 3D, but they are all fixed in 5.6 (that's why I would love to hear from additional wrong cases)06:30
didrocksrickspencer3: for 2d, there is bug #923410 and bug #94761306:31
ubot2Launchpad bug 923410 in unity-2d "HUD - closing a window with <Alt>+<F4> opens the hud" [Undecided,Confirmed] https://launchpad.net/bugs/92341006:31
ubot2Launchpad bug 947613 in unity-2d "Alt key summoning HUD needs to be configurable" [High,Confirmed] https://launchpad.net/bugs/94761306:31
rickspencer3thanks didrocks06:31
didrocksyw :)06:31
didrocks(the first one was used for tracking in 3D btw)06:31
tkamppeterpitti, hi06:35
pittihey tkamppeter06:35
tkamppeterpitti, can you upload cups and cups-filters to Debian and Ubuntu, now where beta freeze is over?06:36
micahgtkamppeter: you hear any reports of HP desktop printing being broke in precise?06:36
pittitkamppeter: can do, yes06:37
tkamppetermicahg, today there came in a bug that Ghostscript falls into an infinite loop, do you mean this (bug 968785)? If not, tell me the bug number.06:42
ubot2Launchpad bug 968785 in ghostscript "ghostscript runs for indefinitely long period of time when called by foomatic-rip" [Undecided,New] https://launchpad.net/bugs/96878506:42
pittitkamppeter: the poppler 0.18 patch fails to unapply, will need to fix this first06:43
micahgtkamppeter: no, I think it has to do with not uploading the firmware, once I did that manually, I was able to print06:43
RAOFdidrocks: Thinking of keyboard shortcuts, have you seen bug #968914 (tapping <super><left> leaves the launcher and keyboard shortcuts overlay up until you press <super> again)?06:53
ubot2Launchpad bug 968914 in unity "Tapping <Super><left> leaves launcher and “keyboard shortcuts” open" [Undecided,New] https://launchpad.net/bugs/96891406:53
pittitkamppeter: all done06:53
didrocksRAOF: yeah, this one is known and fixed in trunk06:54
didrocksRAOF: we will have on Monday a compiz version with this fix and another without my workaround for upgrade users ;)06:54
RAOFYou mean, users' settings won't get reset? :)06:55
didrocksRAOF: right, and we will have a proper fix :)06:57
didrocks(and no more python process starting at boot time)06:57
didrocks(and then and happier pitti)06:57
didrocks(sun will shine, we will enjoy a beer all together…)06:57
pittididrocks: unity-lens-video and unity-scope-video-remot ?06:57
RAOFBunnies will frolick in the dells06:58
didrockspitti: they are not started at login :)06:58
didrocksRAOF: exactly! :)06:58
tkamppeterpitti, thanks.07:49
=== Guest86541 is now known as jalcine
pittibonjour seb128, ca va?08:02
pittiseb128: congrats for breaking the 300 mark!08:02
seb128pitti, hey, wie geht's?08:02
seb128I'm good thanks08:02
seb128got it? waouh! ;-)08:02
pittiseb128: so, nobody got any counts from the proposed copies08:04
seb128pitti, which are least prevented didrocks to go too high in scoring :p08:04
didrocksthat's not fair!08:05
pittiyou win either way :)08:05
didrocksyeah, but it can be so much more classy :)08:07
dpmgood morning desktop people!08:10
pittihey dpm, wie gehts?08:11
dpmmorgen pitti, gut, danke! Und selbst?08:11
seb128hey dpm08:12
dpmmorning seb12808:12
seb128pitti, go pitti go! if I count well with this morning uploads you need one small extra bug to get to 300 ;-)08:13
pittiseb128: I have one udisks fix committed in Debian08:13
pittiseb128: and debugging the race you had yesterday :)08:13
pittiseb128: oh, did you count the cups stuff?08:13
seb128pitti, yes08:13
pittiseb128: most of these are already fixed08:13
seb128oh ok08:13
pittiseb128: Till did -0bzrN uploads08:13
pittithese just have one for me08:13
seb128ok, so a few extra ones still08:14
pittianyway, getting to 300 should be no problem08:14
seb128pitti, let me know if you need extra debug output on your udisks race08:14
* pitti ^5s seb128, we really made precise measurably better08:14
dpmhi mvo, I know Synaptic might be a low priority package, but if you've got the chance, do you think it'd be possible to do an upload including translations only, to fix bug 889351 - it seems lots of translations done in LP are not shipped in the package08:14
ubot2Launchpad bug 889351 in synaptic "Synaptic does not ship all available translations" [Undecided,Confirmed] https://launchpad.net/bugs/88935108:14
* seb128 hugs pitti08:14
micahgdpm: that should be fixed in precise08:17
dpmcool thanks micahg, if it is, would you mind updating the bug status?08:19
micahgdpm: hmm, nope, no en_* translations08:22
pittiSweetshark: guten Morgen08:26
pittiSweetshark: sorry to tell you that all the recent builds failed due to bug 93021708:26
ubot2Launchpad bug 930217 in launchpad "Make proposed pocket useful for staging uploads" [Low,Fix committed] https://launchpad.net/bugs/93021708:26
pittiSweetshark: that fix will be rolled out next week, but could you please prepare a no-change upload to precise today, so that it can build over the weekend?08:27
pittiSweetshark: (we unfroze precise, so we can't use -proposed any more)08:27
=== chaoticuk_ is now known as chaoticuk
mvodpm: oh, sure08:39
seb128pitti, do you have any opinion on taking that change for precise?08:45
=== zyga is now known as zyga-afk
pittiseb128: doesn't quite match Debian/ubuntu, though08:46
pittimost system users have /bin/false as shell08:46
seb128pitti, there is an extra commit to add /bin/false08:46
pittiand many have /bin/sh actually08:46
pittithe static uids08:47
pittisome dynamic system users seem to get it wrong, too08:47
seb128hum ok, seems a bit risky08:47
pittiand we have nologin in a different path08:47
seb128pitti, the context is https://bugzilla.gnome.org/show_bug.cgi?id=67148408:47
Sweetsharkpitti: well, "no-change" still means a version-bump right?08:47
ubot2Gnome bug 671484 in User Accounts "segfault in um_user_set_icon_file() (when uid < UID_MIN?)" [Critical,Resolved: notgnome]08:47
pittiSweetshark: right08:47
Sweetsharkpitti: k, will do08:47
pittiSweetshark: thanks, and sorry for the mess; I wasn't aware that it would trash builds as well08:48
seb128pitti, i.e the user account dialog has issues with users having an uid < UID_MIN, but I guess it's not a frequent case08:48
seb128pitti, thanks, you convinced me it's less risky to stay with what we have :p08:48
pittiseb128: right, real users should stay with >= 500, and with Debian's default config >= 100008:48
pittiif you use something else, then the GUI tools won't show these08:49
seb128pitti, well there is a bit of a bug somewhere but I guess we should fix it from the g-c-c side08:49
seb128pitti, that segfaults has 5 duplicates in launchpad so it happens to some people08:50
seb128pitti, I will look at making g-c-c robust to those rather, playing catchup on what is a valid shell or not seems risky for this cycle08:51
seb128pitti, thanks08:51
seb128doh, 425 updates today08:51
pittiseb128: yes, and as long as sys users have /bin/sh, it won't help us much08:52
seb128updating to beta2 to post beta2 feels like distupgrading from oneiric to precise :p08:52
pittiseb128: I think this upstream commit doesn't really improve matters; there's a reason LSB prescribes UID ranges08:52
pittiseb128: heh, indeed; and to not small degree your fault :)08:52
pittignome 3.4 love08:52
pittibut that staging in proposed worked really well08:52
seb128pitti, I'm pondering trying to argue with upstream over the uid stuff08:53
pittinot a second of archive breakage, and we could use the builders during freeze08:53
seb128pitti, indeed!08:53
* pitti looks forward to next week, when Colin's fix will get rolled out08:53
dpmawesome, thanks mvo. I've added a comment on the bug that the i18n branch needs to be merged before the upload. Also, I've just updated the Catalan translation, but it won't make it until tomorrow's daily export to the branch. Are you planning to do the upload before tomorrow, and if so, may I send you the PO file (or a merge proposal), if it's not too much extra work?08:53
didrockspitti: +1 on proposed, I buy definitively it :)08:55
pittiI just got three HUDs in a row from pressing ctrl+alt+something08:55
pittibut not reliably08:55
didrockspitti: ctrl + alt + which key?08:55
didrockspitti: not a key handled by compiz?08:56
pitticursor mostly08:56
didrocks(like no switching ws)08:56
pittie. g. ws switch08:56
didrocksI'm always using that :/08:56
pittiI also got it all the time yesterday when grabbing/ungrabbing keyboard/mouse in kvm08:56
didrocksdidn't get the HUD once since 5.6 :/08:56
didrockseven now, trying very hard to just type ctrl + alt + left and release in one shot08:57
seb128yeah, I didn't get it either for like 2 weeks08:57
didrocksmy ws switches08:57
didrocksbut didn't see the hUD08:57
pittiyes, if I hammer on them and want to get it, it doesn't appear08:57
mvodpm: I can put the upload on my agenda for monday08:58
dpmmvo, that would be great, thanks!08:59
mvodpm: I put it on my todo09:00
dpmcool :)09:00
=== cking_ is now known as cking
didrocksseb128: do you know on which version of glib bug #926208 was supposed to be fixed?09:11
ubot2Launchpad bug 926208 in bamf "bamfdaemon crashed with SIGABRT in g_assertion_message()" [Critical,Fix released] https://launchpad.net/bugs/92620809:11
seb128didrocks, 2.31.22-0ubuntu109:11
* didrocks sees that the latest dup is old09:12
didrockslooking for new crashers in bamf09:12
didrocksok, last one was on 2012-03-1509:13
didrocksso I guess the commenter just got a something else and thought it's the same09:14
seb128didrocks, or he didn't restart his session for a week and still have the old glib in process or is using an outdated mirror09:16
didrocksseb128: right, at least, with the crash, he will get a new loaded one :)09:17
tkamppeterpitti, can you have a look at bug 968890. Ther CUPS seems not to start because the Dialog utility for Debconf does not find the correct terminal size.09:26
ubot2Launchpad bug 968890 in cups "package cups 1.5.2-8bzr2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/96889009:26
pittitkamppeter: hm, no immediate idea; perhaps the upgrade is ran in a very small terminal?09:40
tkamppeterpitti, should then be some protection against that done, for example apt-get or dpkg aborting with an error right in the beginning? Or should the Readline version of the Debconf interface be checked so that it actually works?09:44
pittidebconf could perhaps fallback to a different backend then09:45
pittitkamppeter: either way, feel free to reassign to debconf; there's nothing cups specific there09:45
tkamppeterpitti, thanks, will do.09:47
xclaesseahah, I can tell beta2 got released, number of user of the MSN app in GOA jumped from 300 daily to over 400 in one day09:50
tkamppeterpitti, done.09:55
pittididrocks: hm, I get it half of the time when using alt+left/right to switch between channels in weechat, but also if I just do that in a terminal (when it has no particular effect)09:55
didrockspitti: oh, maybe you do press an arrows before alt?09:56
pittididrocks: I'm pretty sure I don't; but I press arrow very quickly after alt09:57
pittididrocks: in all the cases when the HUD popped up, the alt+arrow actually did change teh channel09:57
pittiwhen I press alt, wait a bit, press arrow, it works09:57
didrocksmaybe my machine is too slow, but I tried to ping very quickly09:57
pittibut when I press alt and then arrow very quickly (as I usually do), I get it09:57
pittiI can now reproduce it 100% of the time09:58
didrocksseb128: can you try on your machine? ^09:59
pitti$ xev | grep -A2 ^KeyPress09:59
pittiKeyPress event, serial 36, synthetic NO, window 0x3800001,09:59
pitti    root 0xbe, subw 0x0, time 17590219, (112,69), root:(113,867),09:59
pitti    state 0x0, keycode 64 (keysym 0xffe9, Alt_L), same_screen YES,09:59
pittiKeyPress event, serial 36, synthetic NO, window 0x3800001,09:59
pitti    root 0xbe, subw 0x0, time 17590299, (112,69), root:(113,867),09:59
pitti    state 0x8, keycode 113 (keysym 0xff51, Left), same_screen YES,09:59
pittididrocks: ^ this proves that I get the correct order09:59
pittiand I get the HUD09:59
seb128I can confirm yes10:00
pittididrocks: I better stick that into a bug report, I guess?10:00
didrockspitti: yes please10:00
didrockspitti: I can add it to the priority list10:00
pittiit doesn't matter which window (or none at all) is focussed10:00
* didrocks definitively doesn't reproduce, even if I fly over the keyboard10:01
JamesTaitdidrocks: ping10:02
tkamppeterpitti, for all these "CUPS does start on upgrade" bugs I would appreciate if you add a section to https://wiki.ubuntu.com/DebuggingPrintingProblems with instructions to investigate the bug and your debug version of the Upstart script for download. Then I can start as with all the other bugs "follow the instructions of section "XXX" on https://wiki.ubuntu.com/DebuggingPrintingProblems".10:03
didrockshey JamesTait ;)10:03
didrocksJamesTait: so, you crash, you can reproduce, nice!10:03
didrocksJamesTait: can you run please oneconf-query --hosts --direct?10:03
JamesTaitdidrocks: For the public record: https://pastebin.canonical.com/63438/10:03
didrocks(you can use the ubuntu pastebin for it ;))10:04
didrocksJamesTait: in case you didn't notice ^10:07
JamesTaitdidrocks: Good point. ;) http://pastebin.ubuntu.com/906981/10:07
pittididrocks: added everythign I know about it to bug 96903910:07
ubot2Launchpad bug 969039 in unity "HUD appears when pressing Alt+Cursor key" [Undecided,New] https://launchpad.net/bugs/96903910:07
pittiseb128: ^ if you can reproduce10:07
seb128pitti, ok, will comment on it10:08
seb128though I think the only way is to move the hud out of alt, that's being discussed on unity-design10:08
seb128like I can't reproduce those bugs in "normal" use, you need to be brutal in timing10:08
seb128I'm not sure they will ever manager to get it good enough for those10:09
pittididrocks: oh, https://bugs.launchpad.net/ubuntu/+source/unity/+bug/969039/comments/210:10
ubot2Launchpad bug 969039 in unity "HUD appears when pressing Alt+Cursor key" [Undecided,New]10:10
pittididrocks: that might explain why you cannot reproduce10:10
didrocksJamesTait: ok, so you should have a ~/.cache/oneconf directory as well10:11
didrocksJamesTait: can you please tar it and send it to me? :)10:11
didrocks(there is no credential in it, just your package list)10:11
didrockspitti: oh maybe, would be interesting to play with this settings10:11
didrockspitti: it's either that, either my computer is too slow to pick the timing issue with this resolution10:12
didrocksthe only way to show it for me is to press an arrow before alt10:12
didrocksnot in the common case, whatever brutal I am10:12
JamesTaitdidrocks: Sure, gimme 2 secs.10:13
didrocksstill weird that any of those settings can impact on your machine10:13
pittididrocks: hm, I'm not particularly trying hard to be fast; it's my normal keyboard speed, that's how I get it all the time10:14
pittididrocks: but it indeed seems to be due to my settings, I don't get it in a guest session10:14
seb128pitti, it's still quite brutal compared to what "normal" users do ;-)10:15
didrocksseb128: are you infering pitti is brutal? :)10:15
pitti*shrug* my wife doesn't type any slower than I do10:15
seb128well we probably all use keybindings enough to be snappy on them10:15
seb128well typing is different from keybindings10:15
pittibut yes, compared to my mother I do10:15
pittibut if you use keybindings often enough, they get so deeply into your muscle memory that you are just bound to get very fast with them10:16
pittiI use alt+cursor and ctrl+alt+cursor hundreds of times every day10:16
JamesTaitdidrocks: On its way. :)10:16
seb128well as said before I think it's pretty much impossible to fix10:16
didrocksJamesTait: no worry ;)10:16
seb128they should move hud away from the alt key10:16
pittiseb128: well, this particular issue should be absolutely fixable, even more so because it doesn't happen in a fresh profile10:17
pittithere clearly is another X key event between the alt press and release10:17
pittibut independently of that, I agree to not overloading alt10:17
seb128pitti, it feels like they are playing wack a mole on alt tapping for a month10:17
pittictrl+space or so seems quite fine10:17
pittiand 3d is at least configurable10:18
pitti2d sucks because it's hardwired10:18
seb128I'm puzzled that it doesn't happen in a fresh profile10:18
pittiseb128, didrocks: argh, hang on -- just noticed that my guest session is 2d10:19
seb128but yeah, confirmed, doesn't happen in a guest session, weird10:19
didrockspitti: support for changing the key will be added to 2D btw10:20
didrocks(I just discussed it on #ubuntu-unity)10:20
pittibut I now ran a 3D guest session, and still unreproducible10:21
pittididrocks: are all compiz settings in gconf?10:21
pittididrocks: i. e. if I were to try unity --reset again, and it goes away, could I restore them by saving my gconf tree? (for debugging this)10:21
pittior do we have some ~/.config/compiz whatever stuff?10:21
didrockspitti: all compiz settings are in gconf, yeah, but some unity are in gsettings10:22
pittiok, I'll try to bisect this10:22
didrocksand for those I wanted to be shared between 2D and 3D, they put it into gconf to avoid sharing it10:22
* didrocks sighs10:22
pittiI'll see how to make it reproducible from a fresh profile, and what triggers it10:23
didrocksthanks pitti!10:23
pittithis annoys me enough to warrant spending some time on it10:23
pittifinishing up that apport upload first, though10:24
didrocksyeah, especially I wasn't aware about that one10:24
didrocksJamesTait: ah, I think I know what happened and how you ended up in that situation :)10:26
didrocksJamesTait: at least, the crash makes perfect sense to me now :)10:26
JamesTaitdidrocks: That's good. :)10:26
didrocksJamesTait: I know how to unstuck you, but I would appreciate you keep the broken state to confirm the fix10:26
didrocks(in addition to my incoming test case ;))10:26
JamesTaitdidrocks: Not a problem. :)10:27
didrocksJamesTait: ok, will work on it ASAP (when the world will be a little bit quieter ;))10:27
didrocksJamesTait: thanks again for the debugging info :)10:28
JamesTaitdidrocks: You're welcome. Ping me if you need anything else!10:28
didrockswill do!10:28
seb128pitti, random bet but reading your config and looking at mine, it could happen when putting some wm action binded with an alt combinaison10:29
seb128pitti, I use alt-& to go to ws110:30
seb128pitti, well you put alt-b10:30
pittiah, so I'll try to bind alt+b in the guest session and see whether it happens then10:30
pittihm, that's not it alone, unless it depends on restarting unity10:32
pittihm, why does the guest session default to 2D..10:32
didrockspitti: it happens in the past due to apparmor preventing reading /etc/compizconfig/profile10:36
pittisorry, my fault; leftover from some ubuntu-defaults-builder experiments10:36
popeybug 96904410:37
ubot2Launchpad bug 969044 in ubiquity ""try ubuntu" button launches unity 3d when orca is running" [Undecided,New] https://launchpad.net/bugs/96904410:37
popeyif we have orca loaded, we shouldn't be logging people with a11y requirements in to unity, but unity 2d?10:37
popey(on the CD)10:38
pittiI'm not very up to date on that TBH; we used to in the past, but it appears 3D got a lot more accessible10:38
pittiso is that still necessary then?10:38
pittithen again, users of orca won't have much to enjoy from 3D anyway10:39
popey3d is not accessible10:39
popeyAIUI 2D is the focus for a11y work, so we "recommend" people with a11y requirements use 2d10:39
didrockspopey: if you select the accessibility session at startup (before even ubiquity)10:44
didrockspopey: it doesn't start 2D?10:44
popeyapparently not, Alan Bell (who filed the bug) told me this in the pub so i asked him to check it today and file a bug10:44
didrockspitti: as popey told, 3d had no work on accessibility this cycle10:44
popeyhence bug 95904410:44
pittididrocks: just to confirm, launcher staying with the shortcut numbers selected when I press super+something is the one that's already fixed in trunk, right?10:45
didrockspopey: the bug isn't clear enough10:45
didrockspopey: it doesn't say if orca was starting after ubiquity was showing its first page10:45
popeyI'll test and confirm10:46
pittididrocks: (or alternatively, the help staying around indefinitely)10:46
didrockspopey: or if the accessibility was turned even before ubiquity starts (like if you press enter at the very beginning in the boot and choose a11y)10:46
didrockspopey: thanks10:46
didrockspitti: right10:46
didrockspitti: there will be soon a compiz version available in the ppa (I think today won't again be the OneConf day :()10:47
=== MacSlow is now known as MacSlow|lunch
popeydidrocks / pitti updated bug 969044 - it does indeed start Unity 3d, not 2d when you enable the accessibility screen reader, which does seem like a regression11:30
ubot2Launchpad bug 969044 in ubiquity ""try ubuntu" button launches unity 3d when orca is running" [Undecided,Confirmed] https://launchpad.net/bugs/96904411:30
popeyubuntu 11.10 worked.11:30
popey(worked as in, started 2d)11:30
didrockspopey: I worked on that at the time, but it should really be assigned/talked to TheMuso11:32
didrockspopey: thanks for the feedback :)11:32
didrockspitti: maybe something to track on your list? ^ not sure if the settings I did in the natty/oneiric time has been reverted11:33
pittididrocks: sure, tagged; is that actually in ubiquity, or does that need some magic to adjust the default session?11:34
didrockspitti: not ubiquity, it's in casper11:34
pittididrocks: ah, I'll have a look then11:35
ritzseb128, heya, if I understand correctly, we have lp:gtk upstream tree. and lp:~ubuntu-desktop/gtk/ubuntugtk3 with the basic patches/diff to build from this upstream tree for precise . Is this correct ?11:35
pittididrocks: i. e. casper should check if orca is enabled, and adjust the default session to 2d?11:35
didrockspitti: I can maybe have a look, not sure today though11:35
didrockspitti: exactly11:35
didrockslet me dig to my changes11:35
seb128ritz, correct11:35
pittididrocks: thanks, I'll figure it out11:35
pittididrocks: just wanted to confirm how it should work11:35
* pitti toddles off for lunch, bbl11:36
ritzseb128, sweet, thanks :)11:36
didrockspitti: IIRC, I used at the time the gdm-set-default-session that I rewrote as lightm-set-defaults (you have a session switch) in Oneiric11:37
popeyhmm, if I press super + left arrow, my window sticks to the left edge of the screen, and then the popup shortcuts keys thing appears and stays on screen even after i let go of super.. anyone else reproduce this?11:42
popeyI have to press super again to make it go away11:42
seb128popey, -> #ubuntu-unity for unity issues11:43
popeyah sorry11:43
seb128no worry11:43
seb128but upstreams are not on #ubuntu-desktop ;-)11:43
popeyriiiiiight ☺11:43
=== zyga-afk is now known as zyga
pittiseb128, popey: that's already fixed upstreamn12:29
pittiI get that as well, it's what I asked didrocks about a few hours ago12:29
seb128pitti, ok12:29
didrockspitti: btw, a compiz is building in the ppa:unity-team/staging12:29
didrocksthe workaround-less compiz fixing this issue as well ^12:30
didrocks(and it indeed fixed it for me)12:31
seb128didrocks, is there an abi change?12:32
=== MacSlow|lunch is now known as MacSlow
didrocksseb128: not this time12:32
didrocksso we can even party ;)12:33
seb128didrocks, good, maybe put it on in the desktop ppa as well?12:35
seb128didrocks, for those who want to try the stable compiz without jumping on staging unity12:35
smspillazrwar >12:36
desrtsmspillaz: hey.  you're back!12:36
didrocksseb128: why not :)12:36
smspillazdesrt: been back for two weeks now12:37
desrthaven't seen you around much12:37
desrtprobably been too busy working ;)12:37
Sweetsharkpitti: building source package12:37
Sweetsharkpitti: arrgh, should that be -v1:3.5.1-1ubuntu2 or ubuntu3 when ubuntu3 was a failed build in proposed?12:39
desrtseb128: so that issue you hit yesterday was a bug in the hud algorithm12:40
desrtseb128: we didn't notice it before because it was covered by another bug12:40
pittiSweetshark: doesn't matter much; but use -v1:3.5.1-1ubuntu1 to be truly correct12:41
pittiSweetshark: i. e. the current version in precise12:41
Sweetsharkpitti: k12:41
seb128desrt, the dropping the pre scoring to 1 screwing the results?12:42
seb128desrt, hey btw12:42
desrtseb128: ya12:42
desrtseb128: i have a new algorithm mostly-done now12:42
desrtthe old one was too hard to make sense of :p12:44
seb128desrt, not sure if I ;-) or :-( that12:44
desrti think it's mostly a good thing12:45
desrti'm going to spend the morning writing tests12:45
xclaesseis gnome-online-accounts 3.4.0 build with --enable-facebook ?12:49
xclaesseseems it is noto12:52
xclaesseFYI, fedora activated it12:52
xclaesseI think ubuntu should too12:52
jpdsxclaesse: Sounds like a wishlist bug.12:58
seb128xclaesse, didn't kenvandine has reservations about this? especially that you said that GNOME was checking on legal for it?13:05
Sweetsharkpitti: new package is on chinstrap13:05
seb128xclaesse, in any case we don't use gnome online account in Ubuntu by default so it's not going to make a real difference13:05
desrtseb128: initial testing shows the new algorithm to be somewhere between 4 and 10 times faster13:07
desrtie: query that used to take ~20ms on my box is now taking ~3ms13:08
xclaesseseb128, yes GNOME is "checking" legal... waiting 3months for a lawyer to flip a coin to say yes or no with a probability of 50% to be wrong anyway.13:09
pittiSweetshark: and now it's in precise, too; thanks!13:09
jpdsxclaesse: Better safe than sorry.13:09
Sweetsharkpitti: awesome, thanks13:10
xclaessejpds, no need of a lawyer to be sure that if facebook wants to find a flaw they always can. terms are written for that13:10
seb128desrt, without any functionality drop?13:11
desrtseb128: currently i lost the highlighting again13:12
desrtseb128: will bring it back soon :)13:12
desrtthat's not on the 'critical path' in any case, so bringing it back won't add anything13:12
desrtabout functionality losses: adjusted pre-drop penalty is gone (and i think i convinced myself it never worked properly in the first place)13:13
desrtas we discussed yesterday it was probably better to get rid of that anyway13:14
desrtso now top 3 results for "pre" in firefox are "print PREview" "restore PREvious session" and "PREferences"13:15
desrtie: matches at starts of words are now greatly emphasised13:15
desrtit seems to improve the ability to type "f s p a" to mean "file > save page as"13:19
desrtwhich is kinda nice13:19
seb128desrt, can you style typo something? i.e do you still have a distance based calculation?13:24
seb128desrt, i.e prefre matchs preferences?13:24
desrtseb128: yes.  the core algorithm is basically the same13:24
desrtif i removed that capability, i'd like to think i could have done more than 10x improvement :)13:25
pittiseb128: I'm about to upload glib now; any objections, or do you still have stuff?13:25
pittiseb128: I merged with Debian, to also get the multiarch bug fix13:25
pittiseb128: nice to see that all our remaining delta is only for 12.04 stuff :)13:26
seb128pitti, nice! no, the only stuff I had to my list was the multiarch fix that debian added, if you merged we will get it13:26
seb128pitti, thanks for working on that ;-)13:26
pittiseb128: sorry, you already worked on that?13:27
seb128pitti, oh, remaining delta, make sure you don't do the postinst one again13:27
pittihrw mentioned it a few hours ago, and I didn't see you respond13:27
seb128pitti, no, it was on my "tagged emails to look at before precise", ie. my queue13:27
pittiseb128: "the postinst one"?13:27
seb128pitti, I didn't start don't worry, I just had that noted as "needs to be fixed by somebody for precise"13:28
seb128pitti, so good that you got to it ;-)13:28
pittiseb128: I kept the libzeitgeist-io conflicts and the /etc/etc/bash_completion.d/ removal13:28
seb128pitti, wait13:28
pitti(i. e. the -bin maintscript and postinst)13:28
seb128--- glib2.0-2.31.4.tested/debian/libglib2.0-0.postrm.in2011-12-13 22:15:40.000000000 +000013:29
seb128+++ glib2.0-2.31.6/debian/libglib2.0-0.postrm.in2012-01-03 17:28:37.000000000 +000013:29
seb128@@ -14,7 +14,7 @@13:29
seb128     rmdir -p --ignore-fail-on-non-empty /usr/lib/gio/modules13:29
seb128 fi13:29
seb128 13:29
seb128-if [ -d /usr/share/glib-2.0/schemas ]; then13:29
seb128+if [ "$1" = purge ] && [ -d /usr/share/glib-2.0/schemas ]; then13:29
seb128 13:29
seb128pitti, please make sure you don't drop that diff13:29
seb128pitti, you dropped it once before in the previous merge and it breaks the world ;-)13:29
* pitti looks for the corresponding bug, to commit it to Debian13:29
seb128pitti, not sure there is a bug, it's just "if you delete the mmap on upgrade you get no schemas installed and all applications aborting during that timeframe"13:30
pittiah, right13:30
seb128pitti, it's something which leaded to quite some gsettings abort bugs13:30
seb128pitti, the mmap update is safe, no need to delete it, the update will replace it13:30
pitti  * debian/libglib2.0-0.postrm.in: Only remove the compiled schemas on purge,13:33
pitti    not during upgrades. Otherwise we loose the cache if there is no other13:33
pitti    package during the upgrade that installs a schema and thus calls our13:33
pitti    trigger to rebuild it.13:33
pittiseb128: ^ does that sound about right?13:33
pitti(for the Debian commit)13:33
seb128pitti, I'm not sure about the second part the Otherwise...13:34
seb128pitti, but for sure "Otherwise we have no schemas available during the time the postrm from $olversion is run and the postinst of $newversion is run which leads to softwares to abort on missing schemas"13:35
pittiah, right13:35
dobeyanyone seen anything like bug #969209 ?13:36
ubot2Launchpad bug 969209 in gtk+3.0 "Missing expander arrows in some tree views" [Undecided,Confirmed] https://launchpad.net/bugs/96920913:36
seb128dobey, yeah, should be fixed with http://fpaste.org/oHnp/13:36
pittiseb128: cheers13:37
seb128dobey, ups, that expired13:37
seb128pitti, yw, thanks for getting it in Debian ;-) I mentioned it to mbiebl some weeks ago but he said he would ask Josselin and never commented back13:37
seb128dobey, https://bugs.launchpad.net/ubuntu/+source/gtk+3.0/+bug/96324313:37
ubot2Launchpad bug 963243 in gtk+3.0 "Cannot display passwords in keyrings (dup-of: 963862)" [Low,Confirmed]13:37
ubot2Launchpad bug 963862 in gtk+3.0 "The keyring "login" doesn't display any keys or saved passwords." [High,In progress]13:37
dobeyseb128: ah is that uploaded yet?13:38
seb128dobey, http://pastebin.ubuntu.com/907217/13:39
seb128dobey, if you want the diff13:39
seb128dobey, no, I was sort of waiting for Company to commit it to git upstream, but it's on my list for today13:39
dobeyok, i'll wait then.13:40
seb128dobey, Company doesn't seem to be around, I just wanted to check with him before uploading13:40
seb128dobey, yw13:40
seb128dobey, I closed your bug as dup btw ;-)13:40
dobeysure. if evolution is still weird after that fix goes in, i'll file a new bug against evo for that part13:41
seb128dobey, thanks, I confirmed it fixes the seahorse case13:41
dobeyright, evo is behaving a bit differently, so i'm not sure if it's the same bug, but it's close enough for a mention :)13:42
seb128dobey, do you get your expander if you change theme?13:42
dobeynot in evolution13:43
seb128dobey, that's enough to trigger a refresh and fix it in seahorse13:43
seb128ok, so maybe different bugs indeed13:43
dobeyyep. will see after this fix lands :)13:43
pittiseb128: I tried for a while to synthesize the crash in bug 89975713:52
ubot2Launchpad bug 899757 in udisks "udisks-daemon assertion error: HACK: Wanting to register object at path `%s' but there is already an object there." [Medium,Incomplete] https://launchpad.net/bugs/89975713:52
pittiseb128: do you remember what you did exactly when it crashed? plugging in the device and it bounced a few times?13:52
pittiseb128: your log has dozens of adds and remove, I wonder if these were from 20 tries, or a single plugin which bounces like hell13:53
seb128pitti, I played with the ipod connector contacts to make it mount unmount in serie13:53
* pitti curses at udisks log to not have timestamps13:53
seb128pitti, i.e just stress test to triger the bug13:53
pittiseb128: ah, so plugin, pull out, plugin, pull out in rapid succession?13:54
davmor2seb128: just checking before I try did you resolve the music player issue from yesterday?13:54
seb128pitti, the first time I got it I was plugging and unplugging the ipod and a sd card to debug gsd13:54
pittidavmor2: yes, sohuld be fixed in current precise13:54
seb128davmor2, yes13:54
seb128pitti, right, the log is from a purelly stresscase non normal use to get the log13:54
seb128pitti, so yes, unplug,plug several time by seconds13:54
pittiseb128: thanks13:55
davmor2seb128, pitti: Yay works here :)13:55
bcurtiswxbanshee as my default music player, using the music lens, searching 'adele' and clicking on her '21' album should in theory bring me to the ubuntu music store on banshee, correct?13:56
vibhavpIf adele is present on the Ubuntu One Music Store13:57
pittiI'm pretty sure it is, I bought that album a few weeks ago from 7digital13:57
bcurtiswxi verified beforehand that adele and album 21 are in fact in htere13:57
bcurtiswxso I got to the point where banshee was loaded, but there's no music store showing up in banshee13:57
pittiseb128: hm, I inserted and pulled my poor USB stick about 50 times, with varying timings, and didn't get it; bah13:58
pittibcurtiswx: yes, there is no UbuntuOne music store for banshee any more13:58
bcurtiswxpitti, OK, then IMO rhythmbox should open in that use case, correct?13:59
chrisccoulsonw00t, goodbye ugly flicker in firefox history menu13:59
pittibcurtiswx: yes, I guess so13:59
pittibcurtiswx: if it's any consolation, if I try it here I just get {"error": "bad or missing url"} in RB14:01
pittidobey: ^14:01
pittii. e. when I just click on "ubuntu one" in RB14:01
seb128pitti, well playing with the connector is higher timing that plugin and unpluging, I can play with the contact directly14:02
seb128pitti, I might have got the first one when jungling with the 2 devices14:02
pittiseb128: ah, you keep the power on, but not the data lines?14:02
pittiseb128: I so blame Apple for this bug !!!111!!!one14:03
seb128pitti, I don't know, I play with the dock contacts, what happens with the flacky connections...14:03
dobeybcurtiswx: there is no u1 music store in banshee on precise14:03
bcurtiswxpitti, dobey: also if that lens is used and an album is clicked on and the user doesn't have rhythmbox-ubuntuone installed it should request the user to install it (if it's not supposed to be installed by default)14:03
dobeybcurtiswx: it is installed by default14:03
dobeyor it is supposed to be14:04
bcurtiswxdobey, OK. So is there a way to inform the user that they can't check out albums if it's not installed (like i had it occur) ?14:04
kenvandinehey bcurtiswx14:05
dobeyi don't think so14:05
bcurtiswxkenvandine, hey :)14:05
dobeybcurtiswx: i guess whatever provides the store scope needs to Depends on it14:05
bcurtiswxdobey, whatever works best for the time you all have left :)14:06
pittidoes the U1 ms in RB work for anyone in current precise?14:07
bcurtiswxpitti, i just installd rhythmbox-ubuntuone and upon rhythmbox restart and clicking the store it worked fine with no errors14:08
dobeypitti: i don't have any problems14:08
bcurtiswxuse case: closing rhythmbox and clicking on adele's 21 album from the music lens I get: "Error while getting credentials:14:09
bcurtiswxThe '/org/freedesktop/secrets/collection/login/24' object does not exist"14:09
dobeyah crap14:09
dobeythat *is* a bug, which i thought was fixed, but alas :(14:09
dobeybcurtiswx: what veresions of rhythmbox-ubuntuone and libubuntuoneui-3.0-1?14:10
bcurtiswxboth 2.99.91-0ubuntu114:11
dobeybcurtiswx: can you file a bug please?14:11
bcurtiswxdobey, yes with the error from above?14:12
bcurtiswxdobey, would you like me to assign it to you?14:12
bcurtiswxOK,  brb14:12
dobeyunless you want to fix it :)14:12
bcurtiswxdobey, file against rhythmbox?14:13
dobeyalso, a bug about the lens being odd when rhythmbox-ubuntuone isn't installed14:13
dobeybcurtiswx: libubuntuone for that object does not exist error14:13
bcurtiswxand the lens for rhythmbox-ubuntuone?14:14
bcurtiswxdobey, thx14:17
didrocksand a new request bug #96923514:18
ubot2Launchpad bug 969235 in unity "Keyboard shortcuts - Update some window management shortcuts to use "Ctrl + Super" instead of just "Super" " [Undecided,Confirmed] https://launchpad.net/bugs/96923514:18
didrockspitti: not sure if you need a FFe for this one ^ at least, we should ping the documentation team14:19
pittididrocks: yeah, and translations, too14:19
didrocksI asked design to do the paperwork, quite tired to achieve it :)14:20
didrocksand that push to more and more changes14:20
didrocksso let's wait for them doing it14:20
seb128didrocks, with the same rational maybe they can come with ctrl-alt tapping for the hud ;-)14:21
didrocksseb128: ahah, from what I heard… no for now :(14:22
didrocksbut maybe the change to ask for it? ;)14:22
didrocksat least, now we have bug #96925614:23
ubot2Launchpad bug 969256 in unity "Keyboard shortcuts - The HUD keyboard shortcut should be configurable" [Undecided,Confirmed] https://launchpad.net/bugs/96925614:23
didrocksthe translation team has been warned14:23
didrocksI need some compiz work though to expose the key in ccsm14:23
didrocksoupss, I meant, in g-c-c14:23
didrocksand unity-2d needs now to read gconf keys14:23
didrockswhich was removed from it IIRC14:23
rickspencer3didrocks, someone told me that if you change the HUD from using the ALT key, it breaks ALT for other things in Unity, is this the case?14:24
seb128desrt, I wonder if bug #968898 is a side effect of your patch14:24
ubot2Launchpad bug 968898 in gnome-settings-daemon "background picture not cached" [Undecided,Confirmed] https://launchpad.net/bugs/96889814:24
didrocksrickspencer3: was the case pre-5.814:24
rickspencer3didrocks, I like the sound of that!14:24
desrtseb128: yes.  it is.14:24
desrtseb128: directly caused by my changes, in fact14:24
didrocksrickspencer3: heh, how to end better the week, isn't it? :)14:24
desrtseb128: so you have a choice to make, really...14:25
chrisccoulsonkenvandine, oh, i didn't touch my toes ;)14:25
desrtdo you want to make copies of local files or not?14:25
chrisccoulsoni did do this though: http://bazaar.launchpad.net/~extension-hackers/globalmenu-extension/trunk/revision/31414:25
rickspencer3didrocks, well, it was as tough week, but the feedback I am reading about the beta is quite nice14:25
kenvandinechrisccoulson hehe :)14:25
desrtif you don't make copies and someone erases the original then you get this bug14:25
desrtif you do make copies, you cause the UI to be ugly and cause lightdm to break14:25
kenvandinechrisccoulson, your awesome dbusmenu speedup was in yesterday's upload!14:25
chrisccoulsonexcellent :)14:25
desrtGG CC14:26
seb128desrt, I'm still trying to understand the practical issue in that bug14:26
bcurtiswxdobey, bug #96926214:26
ubot2Launchpad bug 969262 in libubuntuone "Error while getting credentials: The '/org/freedesktop/secrets/collection/login/24' object does not exist" [Undecided,New] https://launchpad.net/bugs/96926214:26
seb128desrt, in practice I would say "don't delete an image you use"14:26
desrtseb128: so would i14:26
desrtseb128: but people expect that they can do that because the old behaviour is that a copy was made14:26
seb128desrt, I will comment on the bug thanks14:26
bcurtiswxi tried 'ubuntu-bug libubuntuone' but it said package did not exist what was I doing wrong there?14:26
dobeybcurtiswx: ubuntu-bug only accepts binary packages as package names it seems14:27
dobeybcurtiswx: so would have been libubuntuone-3.0-114:27
bcurtiswxdobey, OK. I'll keep that in mind. Thx :) onto the second bug14:27
mdeslaurseb128: I don't believe a typical user can associate their wallpaper with a particular file. Maybe it should get copied, but to a public location?14:29
seb128mdeslaur, what do you mean "can't associate with a particular file"?14:30
seb128mdeslaur, if you click on a file and select "set as wallpaper" you probably know on what file you clicked?14:30
seb128mdeslaur, give users some credit ;-)14:30
mdeslaurseb128: saying "this file should be my wallpaper". But then later being able to realize that the file you're deleting is the one you've set as your wallpaper14:30
dobeydoesn't gnome make a copy in ~/.local/share/?14:30
mdeslaurseb128: it's a file management association typically understood by power users14:31
seb128dobey, no, it used to copy to .cache by desrt changed that because it breaks stuff, the path of the image is weird in the capplet for example14:31
seb128mdeslaur, well same as if you delete your a lettre you wrote in word and try to open it later... I mean if you delete stuff you are using too bad for you14:32
chrisccoulsonkenvandine, that change i've just done should also speed up the HUD for firefox too, as the menu structure no longer changes unnecessarily :)14:32
dobeyoh, i guess they broke the xml awesomeness for gnome 3.014:32
dobeyso it only reads xml now14:32
bcurtiswxdobey, bug #96927114:36
ubot2Launchpad bug 969271 in unity-lens-music "Clicking an album should bring up only rhythmbox and make sure that rhythmbox-ubuntuone is installed" [Undecided,New] https://launchpad.net/bugs/96927114:36
dobeydidrocks: ^^ want an easy bug for your kill count?14:37
dobeydidrocks: need to make unity-scope-musicstores Depends on rhythmbox-ubuntuone14:39
didrocksdobey: hum, but it still has the banshee logic as well14:40
didrocksisn't it?14:40
dobeydidrocks: no. well there's no u1 store in banshee in precise14:40
didrocksdobey: ok in this case14:41
bcurtiswxI can't reproduce having banshee as default and the lens opening banshee, so <shrugs> ignroe that part of it for now14:42
dobeybcurtiswx: i suspect that probably happens in 11.10 as well, if you remove the u1 extension14:44
bcurtiswxhmm, lemme try removing the extension and see if banshee loads14:44
bcurtiswxdobey, yes banshee loads in that case14:45
dobeybcurtiswx: also, what cpu, and how much RAM does your computer have? and how many songs are in your regular music library, and how many songs are in your u1ms library?14:45
bcurtiswxwant me to add that to the lens bug ?14:47
bcurtiswxdobey, AMD Turion X2, 4GB RAM14:49
dobeyno. asking in reference to your other bug14:49
dobeyit's a race condition14:49
dobeyi thought i fixed it, but apparently only made it harder to hit14:50
pittiseb128: hm, do you still have time to try something with the ipod? I can't make head or tail of it; your udisks debug log is clearly missing a remove event, and I wonder where that's (not) coming from14:52
seb128pitti, sure14:53
bcurtiswxdobey, 559 regular songs and 2 albums in my u1ms (30 songs)14:54
pittiseb128: so in short, I need an udisks output again, plus a parallel run of udevadm monitor -e --udev 2>&1 | tee /tmp/udev.log14:54
pittiseb128: and at the end, a stack trace of that udisks crash14:54
pittiseb128: (doesn't need to be particularly symbolic, if I can see the rough path it crashes in, that's enough)14:54
pittiseb128: so udisks-dbgsym and libglib2.0-dbg ought to suffice14:54
dobeybcurtiswx: ok, thanks14:55
pittiseb128: beacuse the stack trace from the original reporter doesn't look like it's matching your udisks debug log14:55
bcurtiswxdobey, yw. Off to a seminar, bbl :)14:55
pittiseb128: I get the crash if I disable a particular piece of code in udisks, which is meant to prevent that very thing14:56
seb128pitti, http://pastebin.ubuntu.com/907325/14:58
seb128pitti, that's the stracktrace from yesterday14:58
seb128pitti, from the .crash14:58
seb128pitti, will try to get the logs now14:58
pittiseb128: corresponding to https://launchpadlibrarian.net/98880792/udisks.log ?14:59
pittiseb128: cheers14:59
seb128pitti, yes14:59
pittiok, that matches pretty well then after all14:59
dobeythanks didrocks15:00
didrocksdobey: yw, thanks for the head's up :)15:01
seb128pitti, http://people.canonical.com/~seb128/udev.log15:03
seb128pitti, http://people.canonical.com/~seb128/udisks.log15:03
pittiseb128: splendid, thanks!15:04
seb128pitti, yw15:04
seb128pitti, http://pastebin.ubuntu.com/907337/15:06
seb128pitti, that's the corresponding stacktrace15:06
* pitti hugs seb1815:06
* pitti hugs seb128, too15:06
pittiand any other seb!15:06
* seb128 hugs pitti back ;-)15:06
seb128pitti, https://launchpadlibrarian.net/99017695/fix-lp899858.patch ... +1 0 -1?15:15
pittiseb128: LGTM, if it works :)15:16
seb128pitti, seems to work for the oem guys, I don't feel adventurous on a friday evening to upload but I will give it a try on monday and upload if that works there15:16
mterrytremolux, filing all sorts of USC bugs for ya15:25
pittiseb128: that udev.log was tremendously helpful indeed15:30
seb128pitti, oh great, what weird is happening?15:30
pittiUDEV  [26112.087644] change   /sdb (block)15:30
pittithis is utter bogus15:31
seb128pitti, seems weird indeed15:31
pittidevice paths shoudl be something like15:31
seb128pitti, udev bog then?15:31
pittiUDEV  [26101.941079] change   /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.4/2-1.4:1.0/host25/target25:0:0/25:0:0:0/block/sdb (block)15:31
pittikernel bug, yes (udev is just the messenger)15:31
pittithat looks like an event from the in-kernel media polling15:31
pittiseb128: can you please check something while the ipod is plugged in?15:32
seb128pitti, sure15:32
pittips aux|grep udisks.*polling15:32
seb128$ ps aux|grep udisks.*polling15:32
seb128root     25904  0.0  0.0   6552   728 ?        S    17:04   0:00 udisks-daemon: polling /dev/sdb /dev/sr015:32
seb128pitti, ^15:32
pittiseb128: are you running udev 175-0ubuntu8 ?15:33
seb128ii  udev                                         175-0ubuntu6                                   rule-based device node and kernel event manager15:33
pittiseb128: would you be able to update udev, pull out the ipod, plug it in again, and check that ps aux|grep again?15:34
seb128pitti, do I need to restart udev,the box?15:34
pittiseb128: udev restarts itself, no reboot necessary15:34
pittiyou just need to disconnect and reconnect the device15:34
seb128root     25904  0.0  0.0   6552   728 ?        S    17:04   0:00 udisks-daemon: polling /dev/sr015:34
pittithe wonderful world of hotplugging15:34
seb128pitti, ^15:34
pittihm, why the heck is it still polling the CD15:35
pittibut that's a separate bug15:35
pittiseb128: do you still get the crash now?15:35
seb128(udisks-daemon:5424): udisks-daemon-ERROR **: **** HACK: Wanting to register object at path `/org/freedesktop/UDisks/devices/sdb' but there is already an object there. This is an internal error in the daemon. Aborting.15:36
seb128pitti, yes15:36
pittiseb128: thanks; any chance you could get me an udev log for this?15:36
pittiI want to compare the bogus change event15:36
pittiin any case, I have a slightly better idea now what happens15:37
seb128pitti, http://people.canonical.com/~seb128/udevbis.log15:38
pittiseb128: oh, as for polling sr0, I bet/hope that will be gone after a reboot (you didn't dis/reconnect the CD drive)15:38
seb128pitti, in fact I don't have to tweak the cable so much, just doing normal plug,unplug with some seconds between leads to that15:38
pittiUDEV  [28238.049733] change   /sdb (block)15:38
pittiok, same thing15:38
seb128pitti, but the ipod has a few seconds or "connecting" with the logo spinning on screen15:39
pittiseb128: thanks!15:39
seb128so I'm not sure how those devices work15:39
seb128pitti, yw15:39
pittiseb128: so you don't need to be particularly fast, right?15:40
seb128pitti, no, not crazy fast15:41
pittiin fact, it shoudln't even happen when you do it "crazy fast"15:41
seb128pitti, like plug it, wait a second, unplug, wait a second, plug15:41
pittias the polling is only done every two seconds15:41
seb128do that a few times15:41
pittiright, until the polling kicks in15:41
tremoluxmterry: rock on!! thanks MT  :D15:57
pittiseb128: ah, you run i386, right?15:58
pittiseb128: if I toss you a small udisks patch, could you build a package?15:58
seb128pitti, yes, i386, sure15:58
pittiseb128: http://people.canonical.com/~pitti/tmp/udisks.debug.patch15:59
pittiseb128: I bet I know what this will show, but I'd rather confirm15:59
pittiseb128: I just need the udisks log this time, no crash report or udev log15:59
seb128pitti, building16:01
seb128pitti, http://pastebin.ubuntu.com/907438/16:04
seb128pitti, tail of the log16:04
pitti**** UPDATING /sys/sdb16:05
pitti**** registering /org/freedesktop/UDisks/devices/sdb16:05
pittiseb128: merci beaucoup; I can't reproduce it yet, but now at least I know exactly what happens and what goes wrong16:06
pittiI can fix it, but I really want to find a way to synthesize it to add it to the tests, so that udisks2 etc. won't have the same problem all over again16:07
seb128pitti, yw16:07
seb128pitti, what happens?16:07
seb128just curious16:07
davmor2pitti, seb128: in seahorse you can't get to your keyring is this know,  I'm guessing so but thought I'd check16:07
pittiseb128: there are devices which don't tell us when their media changes, such as CD-ROM or SD card drives16:08
seb128davmor2, it's fixed with today's gtk and you can, just expand the line with the keyboard16:08
pittiseb128: either udisks or the kernel continuously needs to poll them, by and large by open()ing the device with O_NONBLOCK16:08
pittithis will trigger a media change event16:08
pittiseb128: it looks like this polling kicks in shortly after the device gets removed, and as the kernel doesn't happen a real device for sdb any more (and no parent for it), it just creates an /sdb fake thingy16:09
pittithat's the area I don't know well any more16:09
pittiseb128: anyway, udisks misinterprets that as "oh, it's a change event, I don't know about this device so treat it as "added"16:09
pittiseb128: and then when the real device comes back, it tries to add it again16:09
G__81hi seb12816:10
seb128G__81, hi16:10
seb128pitti, ok16:10
G__81seb128: remember we spoke on the other day about Ubuntu-Development. I was interested and have installed Ubuntu too16:10
seb128pitti, thanks for the explanation16:10
seb128G__81, yes16:11
davmor2seb128: Thanks I figured it would of been know and almost certainly fixed :)  you guys rock16:11
G__81seb128: i have a doubt about this sponsorship. I fix a bug and for me to commit, you sponsor me is it?16:11
seb128G__81, right, you fix a bug and do a merge proposal on launchpad or file a bug with a patch and subscribe ubuntu-sponsors16:12
seb128we review those regularly16:12
G__81seb128: oh ok so once i get sponsored what happens for the subsequent bugs, i fix a bug and then submit the same for review and once review is done, i commit it is it since i am already sponsored ?16:13
G__81is that how it is?16:13
seb128what do you mean? it works the same way that contributing to upstream project16:13
seb128somebody needs to commit,upload for you16:13
seb128until you did enough work to be granted direct access16:13
seb128then you can do it yourself without sponsoring16:13
G__81oh for every bug you sponsor is it?16:14
pittianyway, need to run now16:14
pittigood night everyone, have a nice weekend!16:14
G__81the upstream projects that i am part of, i do git format-patch and send out the patch and the maintainer applies it. Is it the same here ?16:14
seb128same here, it has been a buggy week ;-)16:15
seb128pitti, 'night, good w.e, see you monday16:15
pittiseb128: s/buggy/busy/?16:15
pitti(buggy as well for sure, though)16:15
seb128pitti, yeah, busy making ubuntu less buggy :p16:15
seb128G__81, same principle out of the fact that we don't use git so it's not git format-patch, it's usually standard gnu diff, debdiff or merge request on launchpad16:16
G__81oh ok seb128 thanks so can you point me to some easy stuff to start off with ?16:17
G__81some upstream package ?16:17
G__81something easy to start off with16:17
seb128G__81, not off hand, pick a package,component you are interested in, look at the bug list and figure something you would like to work on,fix maybe,16:18
G__81hmmm ok will do that thanks seb12816:19
G__81can you give me the link in LP so that i can see the list of packages ?16:20
nessitahello everyone! quick question: I have the launcher set to auto-hide, and very often happens that the launcher appears and never goes "back" again. Any idea if this is reported already? is there a workaround?16:20
didrocksnessita: are you using 2d or 3d?16:32
didrockshey ;)16:32
nessitahola didrocks! 2d16:33
didrocksseb128: so, the quick hack for g-c-c will be a longer hack FYI16:33
nessitadidrocks: I've tried everything I could think of16:33
didrocksnessita: there are some autothide issues in it16:33
didrocksnessita: I suggest pinging upstream on #ubuntu-unity16:33
nessitadidrocks: thanks, will do16:33
didrocksyw :)16:33
didrocksseb128: alt tapping is not supported by g-c-c shortcut, it even doesn't show it if you set a key to only "<Alt>"16:34
didrocksseb128: so I guess fun times on Monday :)16:34
BigWhaleAny Gtk/Gdk experts around here? I was told at #gtk+ that gdk_window_add_filter() was skipped from introspection16:38
BigWhaleanyone know why and if this can be enabled in the (near) future? :)16:39
dobeythat's a question for #gtk+16:42
BigWhaledobey, yeah... they don't know16:42
dobeythey don't know why their own api isn't being introspected by their own introspection device?16:42
BigWhaleyes... well those that are currently present :)16:43
dobey * gdk_window_add_filter: (skip)16:44
dobeyperhaps that is why it is skipped :)16:44
dobeythe docs in the code say to skip it16:44
BigWhaleerr, yes. I was told that. but they don't know why. :)16:46
BigWhalewhy the skip mark16:46
BigWhaleI'll subscribe to a bunch of mailing lists ..16:48
dobeyBigWhale: then dig through git to see who made that change, and ask that person why it's like that :)17:00
dobeyBigWhale: or patch it out and see what breaks17:00
didrocksok, time for week-end here!17:02
didrockshave a good evening everyone :)17:02
BigWhaledobey, I probably will do something like that. Just not right now. :/17:07
raffa50how can i insert a new mime type?17:30
raffa50example: i made an application  (that you can install whit a .deb) that makes a .slyproj file, but the user should open it whit double click17:30
raffa50(whit double click on .slyproj my application should open)17:30
dobeyby following the xdg shared mime info, and xdg desktop file specifications17:31
JanChttp://standards.freedesktop.org/desktop-entry-spec/latest/ & http://standards.freedesktop.org/shared-mime-info-spec/latest/17:34
raffa50my .deb package can't add the mime type (for other users tath download my app)17:35
JanCraffa50: it can, just read those specs (as dobey says, you'll need both)17:38
dobeyraffa50: there are data files your application has to install17:38
dobeyraffa50: you might want to look at the source trees of other projects that use the same build system as you do, and which do what you want, for some better examples of how to do it17:39
raffa50i'm programming in gambas17:40
raffa50(don't laught at me)17:40
raffa50gambas = visual basic17:40
JanCwe might not like to use it ourselves, but everybody can use what they like/prefer, so nobody will be laughed at17:43
JanCraffa50: you might want to look at other applications written in gambas17:44
raffa50no example i searched17:44
dobeyraffa50: what tool is used to build/install the application?17:45
dobeywhat application is it?17:45
raffa50gambas build it's packages17:46
raffa50and generate a .gambas file for my app17:46
JanCraffa50: gambas makes .deb packages?17:46
raffa50in my case17:46
raffa50you said JanC17:47
JanCraffa50: maybe ask the gambas developers then, and if they don't support this yet, file a bug17:47
raffa50i asked on italian forum17:47
raffa50maybe on english17:47
JanCit only needs to copy a couple of files into some locations during install, so it should be easy to add that17:48
JanCand the gambas developers can ask for help if they are not sure about how to do this correctly17:49
raffa50i can't understand that guided that you gave to me17:53
JanCraffa50: I'm sure you could given enough time, but the easiest way is to have the gambas developers implement this, and then it will work for all gambas programs...17:55
bcurtiswxsorry missing / :P17:59
raffa50i thought i could make it whit .deb pakcage18:00
raffa50no english forum18:04
dobeyis anyone else having weird color issues with flash on youtube?18:04
mdeslaurdobey:  bug #967091 and bug #96718218:05
ubot2Launchpad bug 967091 in adobe-flashplugin "Wrong tint with Nvidia after upgrading to 11.2" [Undecided,Confirmed] https://launchpad.net/bugs/96709118:05
ubot2Launchpad bug 967182 in adobe-flashplugin "Flash plugin and libvdpau1 together cause breakage in firefox and chromium in Precise and Oneiric for nvidia users" [Undecided,Confirmed] https://launchpad.net/bugs/96718218:05
mdeslaurdobey: flash and nvidia don't like each other anymore18:05
dobeyit's weird18:05
dobeyvimeo is fine18:06
dobeybut youtube is messed up18:06
Sweetsharkpitti: I nearly died of the 'failed to build' mails for 3.5.1-1ubuntu3 ....18:10
dobeyhmm :-/18:11
JanCraffa50: you could try the mailing list18:15
JanC("Gambas Users" mailing list)18:15
=== DawnLight is now known as mightyiam
maxbCan anyone tell me where GNOME stores the metadata for its Recycle Bin?18:18
maxbI have an "untitled folder" that is stuck there, and doesn't go away when I empty it18:18
maxbI want to track down where this info might be being stored18:18
JanCmaxb: ~/.local/share/Trash/18:21
JanCor in case of external disks, let me check18:22
maxbThere isn't anything in ~/.local/share/Trash except empty {info,files,expunged} directories18:22
maxbI'm not aware of having any external disks attached18:22
JanChttp://standards.freedesktop.org/trash-spec/trashspec-latest.html might be useful18:23
maxbWhat about a network share18:23
JanCthat's external too indeed18:23
maxbHah, right, there is a .Trash-1000 folder in a cifs share I have mounted :-)18:24
maxbAnd I don't have permissions as my user to modify it18:24
maxbProblem solved :-)18:24
JanChttp://standards.freedesktop.org/ --> is a good place to find cross-desktop (GNOME, KDE, XFCE, etc.) specs18:25
raffa50Torvalds help me!18:36
mvoseb128: hi, silly question - isn't this http://paste.ubuntu.com/907697/ supposed to show a label under bar ? I don't seem to see one :/19:00
mvoseb128: ohhhhh, Gtk.Label.new() again :/19:00
* mvo scratches head19:01

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