=== thomj is now known as thomi [03:27] theMuso, hello. Are you there? I've got a question about orca. Is there a way to programmatically tell when it is up, running, and done initializing? It doesn't seem to register on dbus [03:46] mterry: Not that I know of, other than querying the at-spi registry and finding out whether Orca is running. [03:46] mterry: libatspi2.0 shoudl provide an interface to do that I suspect. [03:47] To query currently running clients, adn I dare say they are registered by name, i.e orca. [03:48] TheMuso, I'm looking at bug 944159, which can be worked around by manually triggering a notify on "focus-event" on the currently focused GtkEntry once orca is up. For some reason, orca doesn't do the right thing on startup itself [03:48] Launchpad bug 944159 in unity-greeter "orca reads the password out loud" [High,Confirmed] https://launchpad.net/bugs/944159 [03:48] What do you mean do the right thing? [03:49] TheMuso, on startup, it only says it is in the unity-greeter window. it doesn't read the name of the GtkEntry that is focused nor does it notice that the entry is a password one and that it should just read "asterisk" for characters [03:50] Right. Orca has a splash screen that could be enabled as a work-around which would force a focus change, but thats ugly. [03:51] TheMuso, agreed. :-/ I could simply do "after 3 seconds, fake the focus-event to kick orca", but that is obviously hacky [03:52] Yep. [04:06] TheMuso, is this intentional orca behavior or a bug? (It seems even in my normal session, on startup, orca only reads the current window frame, but does not read the current focused widget) Tested by doing "orca --replace --no-setup --disable splash-window,main-window &" in a Terminal and comparing with what it gives me when I focus into that same Terminal once orca is running [04:11] mterry: You would have to ask upstream if its intensional or not, I am not sure. [04:11] OK [05:50] good morning [07:04] Good morning [07:05] Sweetsha1k: yes, I use debsign -r chinstrap to sign the packages on chinstrap, and then dput from chinstrap [07:07] good morning [07:10] smspillaz: hm, I still get mangled gsettings on compiz startup, but as bug 1042041 is closed now I rather file a new one [07:10] Launchpad bug 1042041 in compiz "1:0.9.8+bzr3319-0ubuntu1 regression: keeps setting gsettings keys to wrong values" [High,Fix released] https://launchpad.net/bugs/1042041 === tkamppeter_ is now known as tkamppeter [08:05] pitti: which keys aren't working ? [08:05] smspillaz: it keeps forgetting about gsettings set org.gnome.desktop.wm.keybindings lower [08:06] smspillaz: and now it resets org.gnome.desktop.wm.preferences focus-mode [08:06] it didn't do that before [08:06] but I have to manually re-enable sloppy focus now [08:06] I'll file the bug once I restart again, so that I can check the gsettings values after start [08:07] smspillaz: my Alt+b key had been broken with the previous version as well, so that part was neither fixed nor regressed [08:07] I need to check whether it fixed my Super+Up/Down keys for max/unmax, I still have that in my "fix-gsettings" script [08:08] it could just be that the values we're setting it to aren't valid ones. I'll double check [08:08] gsettings is pretty picky about this stuff [08:09] does it do that migration on each start? [08:09] pitti: I do remember seeing something the other day about it complaining about the focus mode [08:09] (from gconf, presumably)? [08:09] smspillaz: perhaps from mvo, I think he has the same bug [08:09] pitti: I don't /think/ so. I'm not in charge of the migration script [08:11] smspillaz: fwiw, I have the same bug but did not look into it at all yet [08:13] the latest compiz release remigrate some forgotten keys [08:13] and as the gconf-gsettings script isn't smart enough to see "this is not the default", it's possible that it's been overwritten [08:13] however, if you still have that from now on, it's really a bug [08:14] ok, let me restart right now and check out what it sets them to [08:14] yep :) [08:15] hey [08:15] morning Laney! How are you? [08:15] pitti: mvo: didrocks: changing the ffm setting seems to work fine here [08:16] hey desktopers [08:16] smspillaz: well, you know, we just have 2 ffm users in the distro, and they are in that channel :) [08:16] hey seb128! [08:16] we ought to kill ffm [08:16] heh :) [08:16] yeah good thank you! I spent part of my weekend volunteering in a local café, which was fun [08:16] didrocks: I'm fairly sure we have a lot more :) [08:17] Laney: oh excellent! it's a community cafe? Like managed only by volonteer people? :) [08:17] anyway, so bug 1042041 did _some_ perceptible change [08:17] Launchpad bug 1042041 in compiz "1:0.9.8+bzr3319-0ubuntu1 regression: keeps setting gsettings keys to wrong values" [High,Fix released] https://launchpad.net/bugs/1042041 [08:17] didrocks: yeah, exactly [08:17] pitti: I was jocking :) [08:17] the settings are now consistently wrong, and there are more settings which are wrong [08:17] Laney: interesting :) [08:17] didrocks: two very vocal ones too! [08:17] smspillaz: do you prefer a new bug or reopening this one? [08:17] I got involved a couple of months ago - on the board now [08:18] mvo: heh, yeah, I will start making stats, like "all Germans people are using ffm" :) [08:18] pitti: can you try to remove session-migrations, to ensure it's not that? [08:18] moin. === Sweetsha1k is now known as Sweetshark [08:18] hey Sweetshark [08:18] didrocks: what could go wrong with a sample size of 2 ;) [08:18] mvo: exactly! :-) [08:19] * mvo draws the conclusion that all germans work on ubuntu [08:19] heh [08:19] didrocks: that's a file? [08:19] pitti: package [08:19] and are between 28 - 38 ;) [08:19] and male! [08:19] pitti: "session-migration" [08:19] good morning everyone [08:19] ahah [08:19] hey chrisccoulson! [08:20] chrisccoulson, hey, how are you? [08:20] hi didrocks, how are you? [08:20] didrocks: dependency fail; I take it sudo mv /usr/bin/session-migration{,.disabled} should do? [08:20] bonjour seb128 [08:20] * mvo goes back trying to fix bugs [08:20] didrocks: restarting without that [08:20] pitti: yeah, should do it :) [08:20] hey seb128, i'm not too bad thanks. although, i've definitely got jo's cold now :( [08:20] how are you? [08:20] Sweetshark, hey, I sponsored your libreoffice yesterday (before robert_ancell, sorry I just didn't reply to the email, I though you would get the launchpad upload email) [08:20] chrisccoulson: I'm fine, thanks! Yourself? [08:20] hi pitti :) [08:20] chrisccoulson, I'm good thanks [08:20] * Sweetshark feels guilty for robert_ancell doing all that libreoffice upload dance yesterday night, just to find seb128 doing a silent sleep upload. [08:20] didrocks, other than a cold, not too bad thanks ;) [08:20] chrisccoulson: argh, better to get it before UDS though :) [08:21] seb128: heh, yes. Just saw that, thanks a lot. [08:22] Sweetshark, yw, sorry you had to work on W.E again [08:22] didrocks: still messing up my settings without the script; is that expected or a surprise? [08:22] Laney, hey, thanks for sorting the fonts story ... is it all working? [08:22] bonjour chrisccoulson [08:22] seb128: libreoffice packing is becoming sexy. please are fight about who is first to review and sponsor it. ;) [08:22] pitti: no, it just shows I'm not guilty and can take my sunglasses :) [08:23] ok, I'll file a new bug [08:23] hi pitti. how are you these days? it feels like i've not spoken to you for ages! [08:23] chrisccoulson: well, for less than a week :) [08:23] thanks pitti :) [08:23] chrisccoulson: but anyway, I'm great :) some sore muscles [08:23] seb128: no problem! I'm not aware of any remaining breakage [08:23] I went to a Taekwondo camp on Saturday [08:23] excellent :) [08:23] 500 people, distributed over 5 areas, with 16 national and international grandmasters [08:23] Laney: what have you done? I didn't follow the end of the story? [08:24] chrisccoulson: how are you? [08:24] didrocks: we found that some of the metadata in the font was wrong, so fixed it with fontforge [08:24] pitti - i'm not too bad, other than a bit of a cold now [08:24] this made LO see and select the right fonts [08:24] chrisccoulson: ah, it's that time of the year again :/ [08:24] Laney: excellent! so maybe the Qt fix/workaround wasn't needed as well? [08:24] nah, the Qt issue was separate [08:25] ok :) [08:25] we needed both fixes [08:25] great to see that fixed! awesome :) [08:25] wasn't too bad in the end [08:25] I think we need to get the last fix communicted to DM so that they can do it properly for the next font release though [08:25] not sure how to do that [08:26] yep [08:26] Laney: get in touch with design maybe? I think they are the ones communicating with them? (or Pat?) [08:26] good idea [08:27] well, see my last message in #c internal [08:28] mvo, smspillaz, didrocks: filed bug 1063617 [08:28] Launchpad bug 1063617 in compiz "1:0.9.8+bzr3319-0ubuntu1 regression: keeps setting gsettings keys to wrong values" [Undecided,New] https://launchpad.net/bugs/1063617 [08:29] thanks pitti: smspillaz, sil2100 ^ [08:29] Laney: yes, thank you for fixing this. I had no weekend or allnighters left to care too deeply about that one in addition. sorry. [08:29] pitti: is it the same as last time? (eg, can you force it back by changing the setting manually or is it just on every startup ? [08:29] didrocks: do you think this is a gconf -> gsettings migration issue? Should I check some values in gconf? [08:29] smspillaz: this time, running my fix-gsettings script works fine [08:30] Sweetshark: it's alright. Perhaps you could investigate whether there was actually a bug in LO (i.e. it should have selected the right font anyway) if you get some time [08:30] pitti: I would try renaming all .convert gconf -> gsettings files [08:30] it only worked 2 out of 3 times last time, it disregarded the auto-raise setting back then [08:30] pitti: to ensure it's not that [08:30] didrocks: how exactly does the migration script work? Does it happen whenever someone tries to read those keys or is it per-session or per instance of the program ? [08:30] pitti: the ones in /usr/share/GConf/gsettings/ [08:31] hmmm, what are the chances of finding someone who can a) reproduce https://bugzilla.mozilla.org/show_bug.cgi?id=798157 using fglrx and b) loan me the hardware that it happens on for a few days? ;) [08:31] Mozilla bug 798157 in Widget: Gtk "awesome bar dropdown background color is missing and looks wrong" [Normal,New: ] [08:31] smspillaz: pitti removed the migration script [08:32] smspillaz: so it's not that one which is faulty [08:32] didrocks: ok, all .convert files disabled, trying again [08:32] how often does this script even run? shouldn't it just run once ? [08:34] * smspillaz wonders if its some distro patch [08:35] Laney: I have no deep knowledge of the fonts stuff in LO -- I set popey up to talk with shm_get (LibreOffice community member who did a lot of font work on OSX) and shm_get came back to me with 'its a font metadata' issue when they discussed it ... [08:36] Sweetshark: OK then. I'm a middle man too. sladen is our main man when it comes to knowing about font stuff. [08:36] so perhaps connect the two of them [08:37] oh weird okay I can get it on trunk now for some reason [08:38] didrocks, smspillaz, sil2100: I followed up in the bug with the results of disabling the gconf->gsettings migration [08:39] so if the auto-raise one doesn't belong in that bug, feel free to ignore that part; the resetting of keybindings and focus behaviour seems to be a problem in compiz itself [08:39] mvo: so you get the same bug 1063617? [08:39] Launchpad bug 1063617 in compiz "1:0.9.8+bzr3319-0ubuntu1 regression: keeps setting gsettings keys to wrong values" [Undecided,New] https://launchpad.net/bugs/1063617 [08:39] yeah I think some keys are being unintentionally reset [08:42] pitti: that's interesting [08:42] pitti: can you please open: ~/.local/share/gsettings-data-convert [08:42] and check if wm-schemas is stamped in it? [08:43] didrocks: yes, it is [08:43] oh, wait [08:43] didrocks: it also stamps all my *.disabled scripts now [08:43] so I guess I should move them to another place rather than just renaming them [08:43] pitti: yeah, it's normal, it means they have run [08:43] ah, you rename them old* [08:43] yeah ;) [08:43] pitti: so they were still runned :) [08:43] run* [08:44] pitti: once they are stamped, normally the gsettings -> gconf convert shouldn't run them [08:45] I'll try this with a fresh user [08:48] hm, with a fresh user the result is REALLY confusing [08:48] is compiz actually using gsettings, or does it merely convert gsettings to some other configuration? [08:49] it's using gsettings AFAIK, but better to check with smspillaz :) [08:49] and sil2100 :) [08:51] followed up in the bug [08:51] with a freshly created user, compiz messes up the settings just like for my account [08:52] but it still behaves as if it hadn't -- i. e. gsettings and behaviour don't match [08:52] perhaps it evaluates the settings first, and then messes them up? [08:53] yes, indeed that seems to be the case [08:53] so this bug is perfectly reproducible for a fresh user account [08:56] followed up with that info [08:59] thanks pitti [09:39] Laney: FYI, I pushed a fix for unity regarding translations (getting another translation with ack from design to avoid a string break) [09:40] oh yeah? [09:40] for the release or SRU? [09:40] Laney: I think it should be in the release [09:40] Laney: the diff is really small (just string change + file in the POTFILES.in) [09:41] and I looked at the langpack to ensure the string is translated + local test [09:41] should probably confirm the string isn't in the docs [09:41] Laney: Workspace Switcher -> Workspaces [09:42] Laney: it's also the one showing in the help when you keep Super pressde [09:42] pressed* [09:42] this is an untranslated string otherwise? [09:42] yeah… [09:42] right ok [09:43] so get an ack from jbicha and it's fine by me [09:43] they forgot to put the string in POTFILES.in :/ [09:43] ok :) [09:56] didrocks, was "Workspace Switcher" translated before? [10:00] seb128: yeah, but it's not anymore in the quantal langpacks and TBH, I feel downloading all .po files and sedding them to get them back can be risky, wdyt? [10:02] didrocks, let me see [10:04] didrocks, https://translations.launchpad.net/ubuntu/quantal/+source/unity/+pots/unity/fr/18/+translate [10:04] didrocks, fixed :p [10:04] seb128: what have you done? [10:04] didrocks, that's what I though, launchpad doesn't delete the translation, I've updated the template to add the string and it rescued the old translations [10:05] oh excellent! [10:05] didrocks, just uploaded a pot with the string added [10:05] so ok, doing that [10:05] Laney: FYI ^ [10:05] didrocks, well, nothing to do [10:05] didrocks, it's done [10:05] seb128: the "doing that" was about removing my upload [10:05] oh ok [10:05] but make sure the next upload has a pot with the string [10:05] even if the file is not in the POTFILES.in [10:05] we have the gettext translation picking them [10:06] Anyone know why gnome-screensaver is incapable of locking my laptop's screen? [10:06] because that will overwrite my manually uploaded template [10:06] seb128: yep :) [10:06] there is a langpack export coming, right? [10:06] didrocks, ;-) [10:06] thanks seb128! [10:06] jpds, what does it say if you run "gnome-screensaver-command -l"? [10:07] didrocks, yw ;-) [10:07] seb128: That works. [10:07] seb128: And it locks, if g-s turns off the screen as well. [10:07] jpds, so what doesn't work? [10:07] seb128: If I set: "Lock screen after 30 seconds". [10:08] hate those settings, I think it means "idle time + 30 seconds" [10:08] seb128: Yeah, I just realized that. [10:08] e.g "lock 30 seconds after the screen turns off" [10:08] Yeah, that's kind of... daft. [10:09] Would be better to lock, then after X amount of time, shut off the screen [10:09] Oh well. [10:09] what's the point to keep the screen on once it's locked? [10:10] I like it better the current way [10:10] seb128: Well, on my external monitor, it'd involve pressing the power button again. [10:10] it means I can move the mouse when it's doing the fading effect to stop the idle timeout without having to enter my password [10:17] seb128: OK, so this needs changing: http://people.canonical.com/~jpds/capture.png [10:18] jpds, the UI you mean? [10:18] seb128: "Extinction de l'ecran" should be "Immédiatement". [10:18] is it? [10:19] didn't we just say it was "idle time + that setting" [10:19] seb128: Yeah, and "Extinction de l'ecran" needs to be part of the text label next to the drop-down. [10:19] oh, I see what you mean [10:20] can you open a bug about that? I think mpt is the one who came with the current strings,ui but I also know that the "idle + $setting" is confusing users for a long time [10:20] Will do. [10:20] thanks [10:22] Hey, don't blame me for that chimera of a panel [10:25] mpt, hey ;-) [10:26] mpt, I just blamed you for the string jpds wants to see changed (and I was unsure it was your fault, I just took a guess ;-) [10:28] https://bugs.launchpad.net/ubuntu/+source/gnome-screensaver/+bug/1063682 [10:28] Ubuntu bug 1063682 in gnome-screensaver "Screensaver settings GUI is confusing" [Undecided,New] [10:29] So in French, should it be "Verouiller l'écran [immédiatement :^] après extinction de l'ecran"? Is that where the menu belongs? [10:31] mpt, the issue is that the "turn the screen after" is not an absolute value but a "when the screen is turned off + that time" [10:31] oh, nasty [10:31] sorry "turn the screen" -> "lock the screen" [10:31] so "lock the screen after: 30 seconds" means ... lock the screen 30 seconds after it goes to suspend [10:32] In that case, why doesn't "Lock screen after" become insensitive if "Turn screen off when inactive for" is set to "Never"? [10:32] "bug" or "missing feature" :p [10:33] I know I sketched a partial redesign in a bug attachment somewhere... [10:34] bug #938076 I think [10:34] Launchpad bug 938076 in gsettings-desktop-schemas "[UIFe]Auto-lock on suspend is still needed when encrypting file system" [Low,Fix released] https://launchpad.net/bugs/938076 [10:34] mpt, https://launchpadlibrarian.net/95941560/Brightness%20%26%20Lock.png [10:34] that's the one :-) [10:35] I see it doesn't solve this particular problem [10:36] no, I'm not even sure why the current way is to add times for the lock setting [10:36] we should maybe just make those 2 independent counters and let people lock their screen before it idles if they want [10:39] seb128: Locking before turning off was the old behaviour in the past. === mpt_ is now known as mpt [10:42] * mpt scowls at dbus-daemon using 93% CPU [10:43] mpt: run by the lightdm user? [10:43] didrocks, yes [10:44] seb128: do you know if we have a bug about that? I get it regularly as well ^ [10:44] mpt: you can "just" kill the process and your life gets better, but it's still worrying [10:46] didrocks, in a meeting a.t.m., but when I've finished, any way I can report a useful bug? strace or something? [10:47] didrocks, not that I know about [10:47] mpt: I would say gdb it and bt full [10:47] mpt: to see where you are at in the process [10:47] (--pid=…) [10:48] seb128, didrocks: I was thinking that we should drop libbamf0 (speaking for R)... and go with only libbamf3... Even if we actually don't really directly depend to gtk3 at all, but keeping both is not anymore needed [10:48] the only package depending on that is ginn, but it can work with libbamf3 as well (i..e https://code.launchpad.net/~3v1n0/ginn/libbamf3/+merge/128463 ) [10:48] Trevinho: ginn is depending on it [10:48] ah, would be fine then :) [10:48] didrocks, mpt: ideally it would be nice to have a "dbus-monitor --session" log from the lightdm user, not sure how to run the command for that user though... [10:49] Trevinho, didrocks: looks fine to me, let's do that early next cycle [10:49] yeah, would be greta to have those info, maybe we need to ping robert tomorrow [10:49] didrocks, did you get the issue yourself as well? [10:49] seb128: I got it once until now [10:49] perhaps you could grab DBUS_SESSION_BUS_ADDRESS from /proc//environ? [10:50] and I remember jason complaining about a new install and CPU full [10:50] maybe it was that one [10:50] (I only noticed it because the fun went crazy due to 100% on one core) [10:52] Laney, yeah, or maybe sudo -u lightdm .. mights need to "chsh" first to put a valid login shell rather than /bin/false [10:52] didrocks, up to you whether you think it's a better use of time to give me detailed gdb instructions, or to wait until you next experience it yourself :-) [10:54] mpt: yeah, maybe just kill it and I'll ensure now that I know I'm not the only one to look at it closely next time I hit it :) [10:55] mpt, wait [11:00] mpt, no, I don't have a good wait to get that log .. I guess sudo a "sudo strace -p " and get a bit of that log would be useful [11:00] when is the pid of the dbus process eating cpu [11:02] seb128, "sudo strace -p 1918 > ./dbus-daemon-cpu" produces an empty file. What am I doing wrong? [11:04] mpt: strace outputs to stderr [11:04] mpt: try sudo strace -p 1918 -o ./dbus-daemon-cpu [11:04] or 2>./dbus-daemon-cpu [11:05] but -o is usually easier and better if you run a new program under strace (to avoid mixing that program's and strace's stderr) [11:06] updating empathy [11:07] Laney: on it already :) [11:08] oh sweet [11:08] Laney: upstream pinged seb and I :) [11:08] nice release btw, fixing tons of bugs [11:08] pitti, -o also gives an empty file, while 2> gives a file containing only "Process 1918 attached - interrupt to quit" "Process 1918 detached" :-) [11:08] it is [11:08] one or two I reported, which is why I noticed :P [11:08] build is almost finished, I just need to test then [11:08] ah, ok :-) [11:09] looks like there might be a fix for the arm buildds [11:09] maybe we'll get ffox/tbird/webkit again [11:09] mpt: well, perhaps that process just doesn't actually do anything :) [11:09] Hm, maybe it's waiting for a child process? [11:09] s/perhaps// [11:10] or maybe it's stuck in a loop that doesn't do anything with the kernel [11:10] or just in a select() or sleep or anything [11:10] right [11:10] it can peg the CPU in computation without doing kernel/IO stuff [11:10] * mpt remembers something he learned last week!! === MacSlow is now known as MacSlow|lunch [11:12] Laney: seems to behave finely! Uploaded [11:13] \o/ [11:13] better 'cos now I can queue it through [11:13] indeed ;) === YDdraigGoch is now known as WelshDragon [11:55] hmmm, is https://code.launchpad.net/~ubuntu-desktop/evolution-data-server/ubuntu used? it seems out of date [11:56] ah, robert did the last upload [12:00] chrisccoulson, he probably forgot to commit,push [12:10] lol @ http://www.thepoke.co.uk/2012/10/07/egyptian-parkour-fail/ [12:23] seb128, pitti, after a simple reboot avahi-daemon starts spinning. [12:24] tkamppeter, well, stop cupsd and network [12:24] to see if that stops [12:24] you should have no dbus backlog queued at this point [12:29] seb128, I have done so now: Stopped cupsd, keeps spinning, stopped network, keeps spinning, attached strace to avahi-daemon, log of some seconds (<10 sec): http://pastebin.ubuntu.com/1267347/. [12:30] tkamppeter, well, you have something locally that spams requests for printing queues [12:30] tkamppeter, can you pastebin a ps aux ? [12:55] good morning === MacSlow|lunch is now known as MacSlow [13:16] right, exercise time [13:16] i really don't feel like doing any today :/ [13:32] Laney: I heard that we are using a pretty old empathy release, we should definitively take the latest and greatest that is now in the unapproved key :) [14:34] JohnLea_, ping [14:38] didrocks, 3.6.0.3 already in the works for empathy? It's a fix for a build failure (but since 3.6.0.2 build IDK what happened) [14:39] bcurtiswx: it's in unapproved [14:39] bcurtiswx: it's not a build failure, it's a warning [14:39] didrocks, ah OK [14:39] but I think if we enter this code, as it should return something and doesn't, it has high chance to crash :) [14:39] yup [14:42] I really liked having the online status icon as a part of the messaging menu icon. Is it a bug it's gone again? [14:43] bcurtiswx: it's not a bug, it's the latest design bheavior [14:43] behavior* [14:43] it was confusing people [15:13] psivaa; pong [15:15] JohnLea_, i opened a couple of bugs, bug 1063711 and bug 1063787 against unity, i am not sure if i should assign them to Ayatana [15:15] Launchpad bug 1063711 in unity "Can not unlock password-less guest seessions once the screen locks due to timeout" [Undecided,Confirmed] https://launchpad.net/bugs/1063711 [15:15] Launchpad bug 1063787 in unity "Can not enable timed screen lock without turning off the screen " [Undecided,New] https://launchpad.net/bugs/1063787 [15:16] psivaa, neither of those are unity bugs [15:17] psivaa, the first one is a lightdm known issue (lock screen should be disabled in guest session, that fails sometime) [15:17] psivaa, the second one is a gnome-control-center,gnome-settings-daemon,gnome-screensaver issue [15:18] seb128, ok thanks, ill reassign them [15:19] psivaa, I did already [15:21] seb128, thanks [15:25] seb128, It doesn't see we have a dep on gconf in the uccsconfigure session. [15:26] seb128, Perhaps gotten via dep'ing on Unity? [15:33] didrocks, would be nice if it was toggleable. [15:34] tedg, oh, could be, I tried to uninstall gconf and looked at the list [15:34] tedg, unity was on the list [15:34] seb128, Ah, makes sense. It did dep on gconf, I thought I hadn't fixed it :-) [15:48] didrocks: "Fix build errors"> how come we didn't get them?! [15:48] (empathy) [15:49] or is it one of those C things where you get a warning but actually your stuff is totally fubared [15:49] Laney: it doesn't fix a build error, see bove ^ [15:50] 16:39:16 didrocks | bcurtiswx: it's not a build failure, it's a warning [15:50] 16:39:33 didrocks | but I think if we enter this code, as it should return something and doesn't, it has high chance to crash :) [15:50] ah [15:50] yeah, one of those things indeed [15:50] that's what I specified on debian/changelog [15:50] on purpose [15:50] and didn't mention "build error" here [15:50] i saw the upstream changelog first [15:50] cheers [15:50] my changelog is better! :-) [15:50] clearly :P [15:51] Laney: heh, no worry, I had the same reaction, that's why I pinged upstream about it first :) [16:11] have a good evening everyone! === bjsnider_ is now known as bjsnider [16:27] pitti: hey [16:27] pitti: got quite a fun problem for you :) [16:28] pitti: it would be really great if someone rewrote the type registration in pygobject [16:28] it's pretty crackful at present [16:31] it calls g_type_class_ref() and then adds interfaces and properties [16:31] and it does it in a rather convoluted way to work around the fact that gobject doesn't like you doing that... [16:32] it would be a lot easier if it did the type registration, added all the interfaces, then did the property installs from the class_init function... [17:05] jbicha: Hi Jeremy, hope all is well. [17:05] jbicha: When do you think we can finalize our talk about bug 1035219 in the light of language/locales settings UI for quantal in GNOME Remix? [17:05] Launchpad bug 1035219 in gnome-control-center "In System Settings preference tool/keyboard layouts page automaticaly wrong language selectedGNOME" [Low,In progress] https://launchpad.net/bugs/1035219 [17:10] seb128: heads up: we made the choice to make 3 or 4 incompatible changes to glib under the premise of "we'll see if this breaks anything..." [17:12] some changes mostly around object and class initialisation but also some signal changes... [17:17] GunnarHj: I really don't want to disable gnome's language selector; actually fixing accountsservice or whatever would be best [17:19] there are several usability problems with language-selector-gnome...and it isn't really GNOME so using it isn't really a very good solution for us [17:29] jbicha: I do respect that, but I think that what to do in GNOME Remix is actually a separate question. I'd like to handle the bug for quantal (and possibly precise) as regards the currently available GNOME desktops in Ubuntu. Then I'll be happy to help making it possible to use "Region and Language" in GNOME Remix. Please see comment #14 in the bug report. [17:30] GunnarHj: your solution is to disable one of the tabs in gnome's language panel & have us use language-selector-gnome [17:32] jbicha: My proposed solution _for quantal_ is to disable three of the tabs. ;-) [17:32] GunnarHj: that is already the case [17:33] jbicha: In Ubuntu yes, as long as you don't login selecting one of the GNOME session options. [17:34] Edubuntu isn't really recommending 12.10 for their audience because of LTSP/Unity 3D, bug 1036752, and related issues [17:34] Launchpad bug 1036752 in gnome-session "[quantal] [regression] Gnome Classic has no compiz plugins loaded" [High,Confirmed] https://launchpad.net/bugs/1036752 [17:34] since the Remix uses packages from the Ubuntu archives without patching, I fail to see how you intend to special-case GNOME Classic [17:34] and the Remix also is shipping GNOME Classic for Quantal (but maybe not for 13.04) [17:35] what exactly is accountsservice doing wrong? [17:40] jbicha: Accountsservice is doing just fine for Ubuntu. Language Support is more or less bug free. But before commenting on what accountsservice, g-c-c or any other package does "wrong", are you seriously suggesting that we'd make them compatible in quantal?? [17:43] jbicha: It has been planned for several cycles to make "Region and Language" ready to use in Ubuntu, but nobody seems to have had time to do the job. And please believe me when I say that there are a few things to handle. [17:46] ok, so I think I'm willing to ship with this annoying bug so anything you can do to fix the language panel without cutting parts of it out is welcome [17:47] this isn't the only critical bug we have for the Remix; bug 1045914 is also bad [17:47] Launchpad bug 1045914 in ibus "Keyboard layout doesn't show in GNOME Shell session" [Low,Triaged] https://launchpad.net/bugs/1045914 === Ursinha_ is now known as Ursinha [17:50] heading to lunch... === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away [19:04] jbicha: Hope the lunch was good. [19:04] jbicha: Think I need some clarifications now: [19:04] 1. Exactly which is the annoying bug you talk about? That the current language isn't preselected? [19:04] 2. Are you talking about GNOME Remix only? Or what about those Ubuntu users who select e.g. GNOME Classic and have Language Support as well? [19:04] 3. You said that the Remix "uses packages from the Ubuntu archives without patching". I for one fail to see how that would be possible. I demonstrate in my PPA (https://launchpad.net/~gunnarhj/+archive/misc) a couple of more or less necessary modifications of the Ubuntu versions of g-c-c and g-s-d. Consequently I don't understand your hesitation as regards approving the proposed solution to bug 1035219. Or am I missing anyt [19:04] Launchpad bug 1035219 in gnome-control-center "In System Settings preference tool/keyboard layouts page automaticaly wrong language selectedGNOME" [Low,In progress] https://launchpad.net/bugs/1035219 [19:04] hing? [19:11] GunnarHj: 1. Yes 2. Both, I think the Ubuntu GNOME Remix gets to more or less decide how GNOME Classic works as long as we are shipping it and it doesn't impact other flavors [19:12] 3. the Remix uses this script to build our images https://code.launchpad.net/~ubuntu-gnome-dev/+junk/iso-build-script === m_conley_away is now known as m_conley [19:45] jbicha: IMHO the Ubuntu archive should include packages that work sensibly with standard Ubuntu at first hand, or else they shouldn't be there. That's currently not the case for gnome-session-fallback due to patch 10_keyboard_layout_on_unity.patch in g-c-c. So logically either the bug (showing four tabs instead of the Layouts tab only in a capplet named "Keyboards Layout") should be fixed, or the package removed from the ar [19:45] chive. Maybe shipped exclusively with GNOME Remix? I for one would prefer that the bug is fixed. [19:45] jbicha: The complete upstream "Region and Language" capplet cannot be used in Ubuntu without changes in e.g. g-c-c, g-s-d and accountsservice. The current design of that script does not change that fact. [19:48] Region & Language will be used in 13.04, right? so we can consider this a preview of that [19:49] gnome-session-fallback is only used by Ubuntu GNOME Remix & Edubuntu [19:49] hiding a bug is not the same as fixing it [19:50] I already said I believe I'm ok with shipping with that bug if it's really not possible to fix it sanely [20:15] jbicha: It was planned to use "Region and Language" in Oneiric, Precise and Quantal also. Maybe it will be used in 13.04 - you tell me. [20:15] jbicha: Some users of other Ubuntu flavours are also using gnome-session-fallback of course, even if it's not included by default. [20:15] jbicha: You did not really answer my question #3 above, did you? [20:16] which part are we still confused on? [20:16] by "without patches" I mean we don't use a PPA; it's not that we're removing Ubuntu's patches so that there are two different versions of like GTK floating around depending on which Ubuntu you downloaded [20:20] jbicha: Without modifying some Ubuntu packages, you can't replace "Language Support with "Region and Language". Period. And that's true whether my proposed bug solution is accepted or not." [20:21] jbicha: i18n is a little more than the look of the GUI. [20:21] what do you mean "period"? You ran Ubuntu GNOME Remix Beta right? aren't we doing that already? [20:22] No. I failed to install it - couldn't proceed to the "Install Ubuntu" dialogue. [20:23] ok, how about the Try Ubuntu option? [20:24] But that wouldn't let me change things for testing, right? [20:24] you could stick it on a USB stick with persistant storage [20:24] Btw that didn't work either. The screen became a mess. [20:25] if you do that, you probably want to use precise's startup disk creator as I believe quantal's is still more or less broken [20:27] I don't follow you as regards sticking it on a USB. What's the difference between "with persistant storage" and installing it? [20:52] seb128, still there? [20:59] tkamppeter, yes [21:00] tkamppeter, ok, I see that you found a patch for the avahi issue and doko said he's looking at it, good work! [21:09] jbicha: Did you see my latest question? [21:11] doko, thanks in advance for looking into the patch for the upload. [21:18] GunnarHj: I only suggested the USB thing if the installer were broken [21:19] jbicha: But could you expand on the USB thing? I.e. how do I get all the files onto a USB without running the installer? [21:22] instead of making a live CD, you can make a live USB http://www.ubuntu.com/download/help/create-a-usb-stick-on-ubuntu === soren_ is now known as soren [21:24] jbicha: Thanks, I'll try that, but probably not today (it's 23 pm in Sweden). [21:25] jbicha: Btw, am I the only one who have had problems with those ISOs? [21:26] GunnarHj: I've not heard of your particular problem, do the regular quantal ISOs work for you? [21:27] you're welcome to file a bug, you can use ubuntu-gnome-meta if you can't find a better place [21:27] jbicha: Haven't tried any of them - working with a distupgrade from precise + updates. :-/ [22:32] jbicha, still here? [22:32] robert_ancell: yes, hi! [22:33] jbicha, do you have an opinion on bug 1055359 - any opposition to just removing the attempt to stop GDM when uninstalling it? [22:33] Launchpad bug 1055359 in gdm "package gdm 3.5.92.1-0ubuntu1 failed to install/upgrade: subprocess installed pre-removal script returned error exit status 1" [Undecided,Confirmed] https://launchpad.net/bugs/1055359 [22:33] or know any reason why that would be a good idea? [22:34] robert_ancell: I was just asking darkxst on #ubuntu-gnome about that [22:34] did you know we have an #ubuntu-gnome channel? [22:34] jbicha, nope [22:34] in there now [22:35] I agree that stopping gdm doesn't sound like a good idea === psivaa_ is now known as psivaa === m_conley is now known as m_conley_away === Ursinha_ is now known as Ursinha