=== oubiwann_ is now known as oubiwann [06:31] bigon: is there a reason you added gjs back to natty? === smspillaz|asleep is now known as smspillaz === warp11 is now known as warp10 [07:36] good morning [08:07] Good morning [08:07] morning pitti, good morning all [08:08] hey dpm [08:17] Good morning all [08:20] hey Sweetshark [08:29] * Sweetshark sees his overnight build of LibreOffice-3.3.1.1 got all the way into the smoketest ... [08:30] ... just to fail there (looks like a simple permission problem though). [08:30] didrocks: hey, re your show-techitems-in-never-visible-mode> it feel to me like in NEVER_DISPLAY_NONAPPS it should also hide the "show nonapps" actionbar, what do you thnk? [08:31] mvo: hum, no, otherwise in OneConf, we will never have that bar showing [08:31] mvo: the mode is that you don't want "advanced smartness" like showing technical items if there are only apps [08:31] mvo: that's particular to OneConf because we are checking "two sides" at once [08:31] the additional app/removed app [08:32] so it's not important if one side is at 0, and not the other [08:32] does it make sense? [08:34] didrocks: sure, you are currently the only user of this mode, so that is fine. you use it to avoid that the call is made twice? [08:34] mvo: right and that in one side we are showing only app, and the other tech items + app [08:35] mvo: but in any case, for natty +1, if I can have real time to work on oneconf, I'll do mpt's "one tree" interface so that will be not valid anymore [08:35] having two datamodel in one isn't really nice TBH != [08:35] s/\!\=/:) [08:35] hey didrocks [08:35] good morning pitti [08:35] how are you? [08:36] Sweetshark: congrats! it took me almost 3 days to get the maverick OO.o building on Natty even.. [08:36] didrocks: pretty well, thanks! saw "Black Swan" last night with some friends [08:37] pitti: oh, how was it? [08:37] didrocks: a bit psycho (that was to be expected), but I had feared a lot worse [08:38] nice composition of drama and art [08:38] pretty much like unity :-P [08:39] ohhhh! :-) [08:40] but I liked the main conflict of it ("be perfect" vs. "just live, dammit") [08:41] I think that can be interesting :) [08:41] even if it's not the kind of movie I go to the cinema for [08:42] * bryceh waves [08:42] hey bryceh [08:42] hey bryceh, how are you? [08:42] late for you.. [08:43] yeah a bit late :-) [08:43] * Sweetshark waves back at bryceh. [08:43] been working on getting all my various boxes upgraded to latest natty and reporting resultant bugs ;-) [08:43] heya Sweetshark! [08:45] hey, probably simple question... the date/time indicator now only shows the time. Is there a way to force it to again show the date in panel? [08:45] I no longer know if it's Sunday, next Tuesday, August 13, ... ;-) [08:46] it shows if you open it [08:46] oh, the date/time indicator shows appointments now, also if they are _totally_ wrong [08:46] oh, they are a subset of mine from the entire week, and in 12 hour format [08:49] And the clock's now in 24h format, and doesn't follow the cultural preference. [08:49] if only we had a set of country/language specific description of currency, time, paper, and so on [08:50] we could call those "locales" or so [08:50] I know, we can screenscrape from canonicaladmin.com! [08:50] "to get your time format right, apply for a contract with Canonical" [08:50] ? [08:51] 3 ... [08:51] 4 Profit!! [08:53] dpm: tested en and de for lucid, adding to wiki page [08:53] dpm: do you have some time to check Spanish? [08:54] pitti, yeah, I could do so, give me a few minutes and I'll add it there [08:54] cool [08:55] dpm: then I'll move the lot from that page to -updates, and we can roll the 10.04.2 CDs [08:55] pitti, ok [08:58] pitti, I've requested a full export for Natty. I think, if po2xpi works well, the language packs should build well too. I saw that the fi, oc, Lucid langpacks were not ok because the files exported from LP were incorrect (bug 715854), but I've had a look at the Natty ones and look ok. I think after danilo's changes and yours they should build ok. The export starts today at 14:00 UTC, perhaps we can have a look at them next week, then [08:58] Launchpad bug 715854 in launchpad "Exported Firefox translations contain wrong references to languages other than the current" [Critical,Triaged] https://launchpad.net/bugs/715854 [08:59] dpm: hm, I thought I already tried them in devmode [08:59] dpm: so perhaps I should disable the automatic natty cron job then, and build them manually first [08:59] (the cron jobs don't rebuild -base anyway) [08:59] pitti, did you try the natty ones? [08:59] yes, both [09:00] oh, I see :( [09:00] I might have done something wrong, of course [09:00] the 4.0 xpis seem to have changed in structure somehow [09:00] but I'll try again next Monday then [09:00] this week I was more concerned about fixing them for lucid (and fixing the bugs which broke the builds completely) [09:02] pitti, yeah, the 4.0 ones haven't got a compressed .jar file inside them, the files are now there uncompressed. That's what danilo fixed in po2xpi (adapted it to the 4.0 format) [09:03] pitti, btw is the cron job that uploads the natty ones really running? I see that the latest langpacks are from 2011-01-17 [09:03] -rw-r--r-- 1 langpack langpack 1712 Feb 8 15:23 natty-upload.log [09:03] the date is right, but it's too small [09:04] argh, "unusable secret key" -- I thought I fixed that last week [09:04] need to investigate [09:04] ok, thanks pitti [09:04] oh, hang on [09:04] that was on Tuesday, before I fixed the stuff [09:04] so the next one should work [09:05] pitti, ok, the next one should be a full one, in case that makes a difference (I can still cancel the request the full export if necessary, though) [09:05] dpm: no, that's fine; let's get natty firefox translations working [09:06] yeah [09:22] didrocks: so there was no new bamf release yesterday after all; should I cherrypick the fix into the Ubuntu package? [09:23] pitti: I've done the release [09:23] but seems that I didn't dput as there were // issues with nux [09:23] so side-tracked and forgot it [09:23] sorry, doing it [09:23] didrocks: ah, don't worry [09:24] didrocks: just wondered what to write in today's release report about this bug [09:24] didrocks: you can't :) [09:24] -rw------- 1 didrocks didrocks 1815 2011-02-10 17:59 bamf_0.2.78-0ubuntu1_source.changes [09:24] -ENOLAUNCHPAD [09:24] oh right [09:24] see, it was waiting there ^^ [09:24] pitti: so just after, I'll push it [09:24] * pitti uploads some stuff to Debian then :) [09:24] * pitti hugs didrocks [09:24] didrocks: no hurry, I was really just wondering if there was something bad about it [09:24] * didrocks hugs pitti, sorry ETOOMANYTHINGINPARALLELANDFORGOTBAMF [09:25] pitti: no no, I've made the local release, uploaded the tarball to launchpad, tested it [09:25] * pitti buys didrocks a bigger brain stack [09:25] just missing the final dput :) [09:25] pitti: it's needed! [09:25] we so much need build from bzr! [09:25] * mvo wants one as well! [09:26] mvo: two for the price of one? :) [09:26] pitti: yeah… [09:26] buy one, forget one free! [09:26] heh [09:29] so lp is down for 90min? I wish bzr would still allow me to push my code [09:30] its a bit tricky to allow that [09:30] would be nice it would allow to pull code at least [09:30] the db that controlls access permissions is in readonly mode [09:30] readonly will be doable eventually [09:43] is it expected that I get now also a menubar for the "desktop" nautilus and gnome-terminal where I disabled the menubar by default? (Classic Desktop) [09:46] dpm: thanks for testing Spanish [09:46] dpm: I think all these tests just referred to the GNOME langpacks, so I shouldn't copy the KDE ones yet? [09:51] pitti, hm, tough decision, but yeah, afaik all the testers on that page are Ubuntu (GNOME) people. We have never explicitly tested the KDE langpacks, and I'm not sure what we did in the past, whether we uploaded them or not. [09:52] I should add a section explicitly for KDE and a column for it on the table too [10:15] ah, LP is back up [10:16] ^ mvo, didrocks: FYI [10:16] pitti: thanks [10:17] yeah! [10:31] pitti: I will soon need to get a rundown of the publishing stuff: Should I register a new branch for the 3.3.1 LO at https://code.launchpad.net/~libreoffice ? How do I do the builds and uploads on various platforms etc. [10:34] Sweetshark: not a +junk one, please === evilvish is now known as vish [10:35] Sweetshark: https://code.launchpad.net/openoffice is the layout we had for oo.o, but perhaps we should make that a little more regular [10:35] Sweetshark: our usual approach is to have lp:~ubuntu-desktop/projectname/ubuntu for the "packaging trunk" [10:35] Sweetshark: which is also what should appear in the Vcs-Bzr: line [10:36] Sweetshark: and if we need to do changes for older releases (SRUs, stable release updates), we call them /maverick [10:36] Sweetshark: so I suggest to push your's to lp:~bjoern-michaelsen/libreoffice/ubuntu/ [10:36] Sweetshark: that is, we are talking about the debian/ branch here, right? [10:36] not an upstream import [10:37] Sweetshark: I'm suggesting you as owner because you aren't in ~ubuntu-desktop yet [10:37] Sweetshark: the owner can be changed later on, for now "owned by you" is fine [10:38] Sweetshark: so I don't think we need a new branch for every upstream release [10:39] Sweetshark: uploads> you need sponsoring, you aren't allowed to upload to Ubuntu yet [10:39] Sweetshark: you can upload to a PPA, of course, for a first test (which would actually ease sponsoring, too) === cking is now known as cking-afk [10:41] heh, yes: the debian branch, upstream imports for LO might be rather unfair to the launchpad team. [10:43] Sweetshark: yeah, lp:~vcs-imports/openoffice/trunk doesn't seem very happy either [10:43] Sweetshark: and I don't think they'd be too useful either [10:44] it's too much of a pain to derive the packaging branch from that one anyway [10:44] so debian/ only sounds fine for that [10:44] also, debian is likely to move to git soon with their repo. As we are merging lots of stuff from them: is there a good bzr/git bridge ? [10:45] it kinda works, but not in both directions [10:45] * Sweetshark is constantly botching his vcs commands with debian/ in bzr, LO in git and OOo in hg ... [10:45] Sweetshark: once they do, I suggest to have an ubuntu branch in the Debian packaging git and negotiating with Debian that you can get commit access to it [10:46] that makes it so much easier to merge from each other [10:46] Sweetshark: buy Rene a beer and promise him to not commit to his master branches? :-) [10:47] Sweetshark: if he doesn't want to, then just move to git as well, I'd say [10:47] then you can still merge back and forth, they just live on different servers then [10:47] pitti: commit access to the debian repo was already offered to me, so that should not be a problem. [10:47] ah, so much the better [10:48] Sweetshark: over time, I suppose you would/should make packaging fixes in the debian branch as much as possible, and only keep the unavoidable delta in ubuntu [10:48] ... that is until I break debian for the first time ;) [10:48] sure [10:48] I'd start with committing them to ubuntu branch and asking Rene to review/cherrypick [10:49] after some time he'll tell you to JFDI [10:49] or "JFCI" in that case :) [10:50] We currently have way to many "patching levels" in LO: The main repos, the stuff that gets patched upon it by libreoffice-build and then again, the stuff that gets patched upon that by debian/* .... [10:50] Sweetshark: btw, were there any mysteries you stumbled over when you studied the libo and general packaging? [10:50] Sweetshark: yeah, these are a pain [10:50] Sweetshark: is libo-build what used to be go-oo? [10:50] Sweetshark, oh, and just as you know the debian lo maintainer doesn't like ubuntu much [10:50] i. e. is that meant to gradually be merged into trunk? or stay forever? [10:51] Sweetshark, so if he has weird comments don't take it against you [10:51] seb128: well, so far we didn't really contribute anything back to Debian's packages.. [10:51] he's usually reasonable with technical patches [10:51] pitti, well, he really hates ubuntu though [10:51] yeah he has a a bit of a grudge; but with Sweetshark now rocking the house that will hopefully get better :) [10:51] hi mvo, I can't get software-properties-gtk to run well on unity. It sometimes load, sometimes the window is hidden and I have to end up killing it. Do you know if this is a known bug? [10:52] how get I rid of those "unwanted" menubars I now get in the Classic Desktop in natty? [10:52] geser: I think the classic desktop now installs the menu bar applet by default; just delete it? [10:55] seb128: I helped getting some patches that debian cared about into OOo upstream ... so now it is kind of a reverted position. [10:55] pitti: you mean the "indicator-appmenu" package? [10:56] geser: no, the applet in the panel [10:56] oh, I thought it was a separate applet; I was 90% sure I removed that [10:56] geser: but if it's the package, sure [10:56] but it shouldn't be necessary to remove that package to remove the global menu [10:56] pitti: I've a menubar (from nautilus) on my second screen where I have no panel [10:57] and gnome-terminal shows also it's menubar even as I've no menubar on new terminals selected in the options [10:57] libo-build is ex-go-oo. There are still >600 patches in the dev300 folder alone though .... [10:57] geser: can't you right-click on the menu bar in the panel and "remove"? [10:57] Sweetshark: right, I know it's huge; I was just wondering if that will eventually be merged [10:58] pitti: no, I'll make a screenshot [11:00] pitti: http://www.bienia.de/tmp/Screenshot.png [11:02] geser: hm, no idea about that then; kenvandine ^ ? [11:05] pitti: It has to be merged asap. However, if I will state that too loud I'll have to do it. Since some patches are only applied conditionally (some distros do, others dont) there is also diplomacy involved in that. But it has to be done, because currently users are told dont file issue of your non-vanilla build at freedesktop.org and on launchpad they are told to not file upstream feature requests. This is frustrating for users and thus [11:06] Having too many different patchsets is very bad from a QA point of view. [11:07] yeah, absolutely [11:08] and nobody really knows which patches are applied where, and thus you effectively have 10.000 combinations out there [11:08] dpm: hello, let me try that after lunch, maybe something with the gir stuff? [11:08] mvo: current natty softwarwe-properties doesn't use GI yet, though [11:09] it's pure pygtk [11:09] but FWIW, it works well here (natty, unity) [11:09] pitti: also at LO upstream is not upsteam: developers usually do a "native build", that is without the patches from libo-build. Another thing that can be heal by moving as much as possible directly in the master ... [11:10] pitti: aha, sorry, you are right of course [11:10] pitti: my head connected dpm with language-selector [11:11] :) [11:11] but that works like a charm for sure, as it's GI!!11! [11:11] mvo, no rush, I just thought I'd mention it. I can trigger the problem when I load software-properties-gtk from the command line [11:12] dpm: with sudo or gksu? [11:12] pitti, with sudo and without [11:12] it doesn't run as user, it just brings up an error dialog "plz run me as root" [11:13] but unity has started behaving a bit crazy here, I'm not sure if that's the cause [11:13] * Sweetshark realizes what a luxury it was to build with --disable-binfilter --disable-mozilla as a developer ... [11:13] brb [11:22] with gksu works well, I only get a hidden window when running it with sudo or without, from the command line [11:48] mvo: yay, got s-properties running with GI [11:49] mvo: http://bazaar.launchpad.net/~pitti/software-properties/pygi/revision/650 [11:50] mvo: now the MirrorTestGTK class doesn't actually do anything GTK specific any more [11:50] mvo: I could simplify this a lot now, by adding the changes to the main MirrorTest class, and just using that from both -gtk and -kde, if you want me to? === MacSlow is now known as MacSlow|lunch [12:01] good morning [12:01] * didrocks just receives his 23' monitor \o/ [12:01] hey cyphermox, how are you? [12:02] hey cyphermox [12:02] didrocks: niice! size does matter! [12:03] pitti: heh ;) yeah, I will not anymore break my eyes on a 17' for 1920x1200 [12:03] I'll have the same resolution on a way bigger screen :) [12:07] pitti, hey, doing alright [12:08] didrocks, nice! [12:09] i miss my monitor, i still can't connect it to my laptop without crashing it [12:09] pitti: please [12:09] didrocks: niccce, which one did you get? [12:09] * mvo was considering a 23/24" for some time already [12:10] mvo: it's a samsung SyncMaster F2380, I have no clue about hw, so I relied on some magazine to compare to see what's nice for non-game usage [12:14] * Sweetshark likes his Sun X7236 24" screen ... Though you have to be careful to turn down the brightness -- it might blind you otherwise ... [12:18] * didrocks knows the first thing to do with the monitor for the week-end: reinstalling the server at home and reset up his blog :) [12:20] didrocks: and your IRC proxy? :-) [12:20] pitti: I'm not sure about that one… I feel my heart "lighter" when knowing I can't be pinged on IRC while being away TBH… [12:21] ah, ok === cking-afk is now known as cking [12:22] we'll see ;) [12:22] mvo: ok, resubmitted [12:27] * pitti sends didrocks a ---`--,<@ flower for dput'ing bamf [12:27] * pitti lunch, bbl & [12:28] pitti: done ;) [12:28] enjoy! [12:38] didrocks: yeah ! now you can debug my dual head issues ;) [12:38] stgraber: heh, I think I'll be forced to ;) [12:38] I don't know where cyphermox is with that btw [12:41] didrocks, with what? [12:41] cyphermox: wow, you're up early ;) [12:42] I've been up for an hour [12:42] cyphermox: multimonitor support, I think you started it, isn't it? [12:42] didrocks, ah, yeah [12:43] I don't understand GL enough to continue though, it's still the same issue, it works when unity starts, but if there is a res change, everything is "fine" except the panel and launcher don't get repainted where they should be, and I haven't been able to get them to do it [12:43] I've been busy with NM and other stuff since [12:45] cyphermox: sure, no worry === oubiwann is now known as oubiwann_ === MacSlow|lunch is now known as MacSlow [13:33] seb128: should we drop some of your remaining WIs for cleaning up old libs? this will be significantly easier with gnome 3.0 in the next cycle, I guess? [13:34] pitti, we are like 2-3 source away to drop some so I was keeping those on purpose [13:34] seb128: oh, good to hear [13:34] pitti, like libglade it's only checkbox which ara seems to have a vcs for and pitivi [13:35] I had the feeling that was pretty much "done" now for .32 [13:35] libgnomeui, etc is probably going to be deferred due to tomboy [13:35] seb128: I was actually wondering about pitivi after reading Robert's trip report [13:35] they use gconfpeditor which is using libgnome [13:35] it's probably not worth the effort to port this cycle [13:36] *nod* [13:36] pitti, well it seems not likely we will change pitivi this cycle [13:36] we might want to port it to gtkbuilder still [13:36] but at the same time libglade is small enough that it might not be worth the efforts [13:36] I will drop the summary on the spec and update the work items [13:37] I was a bit concerned to read that even upstream devs say it's not well maintained ATM [13:39] pitti, well it's like it was the only project short on people maintaining it [13:40] heh [13:45] does anybody else dislike the new behaviour of bash-completion? [13:48] * bcurtiswx waves to room [13:48] chrisccoulson: how so? [13:49] pitti - say, if i want to do "ls" in a directory that is several folders deep, i could just tab-complete my way down the hierarchy [13:50] that seems unchanged here [13:51] pitti - it's broken in the latest bash-completion. when you hit tab and it completes a filename, the cursor ends up in a position where you can't descend any further [13:51] (ie, at each level, you have to hit backspace, and then tab again) [13:51] that changed in the latest bash-completion [13:52] i'm finding it difficult to describe the behaviour ;) [13:52] hm, it completes to dir[tab] -> dirname/ for me, as it should [13:52] pitti - it completes to dir[tab] -> dirname | (with '|' being the new cursor position) [13:52] (for me, anyway) [13:53] so, hitting tab no longer lists the contents of "dirname" [13:53] just a wild guess like this, but I think indicator-datetime has some issues with geoclue right now... [13:53] and i have to press backspace instead [13:53] pitti - which bash-completion do you have? [13:53] 1:1.3-1ubuntu1 [13:53] chrisccoulson: but that shouldn't affect file name completion, that's bash itself [13:54] bash-completion does stuff like expand package names for "apt-get" or options for make [13:54] pitti - hmmm, that's strange. when i downgraded bash-completion, i got the old behaviour back again :/ [13:54] chrisccoulson: hm, so maybe some script there mis-matches [13:54] chrisccoulson: it's at least not meant to fiddle with file name completion === gord_ is now known as gord [14:04] hmmm, X just died [14:05] pitti - this is what happens with me - http://people.canonical.com/~chrisccoulson/bash-completion.ogv [14:05] a screencast explains it much better ;) [14:06] could someone please review my merge requests for connman and ofono? [14:06] cyphermox, oh, you can't upload those? [14:07] chrisccoulson, nah :) [14:07] chrisccoulson, I'll run for motu at the end of the month though, I think [14:07] cyphermox, i can review those later on if nobody else beats me to it ;) [14:07] i've got some other things to do first though [14:07] yeah, you should :) [14:08] probably should for desktop team as well, maybe [14:08] chrisccoulson: ugh, that's indeed nasty [14:08] pitti - it goes away if i downgrade bash-completion :/ [14:08] chrisccoulson: but it's not doing that for me, and I think it's a bug [14:08] chrisccoulson: apparently it misidentifies a directory as a file? [14:08] pitti - yeah, it seems like that [14:09] it was driving me crazy yesterday ;) [14:27] kenvandine, hey [14:28] pitti, it is this chunk in /etc/bash_completion that breaks it: http://paste.ubuntu.com/565880/ [14:29] i just reverted it, and it works again :) [14:30] chrisccoulson: hm, that looks wrong indeed -- why wouldn't ls, less, patch, diff, etc, take filenames?? [14:31] seb128: hey, can you help reviewing https://code.launchpad.net/~rsalveti/clutter/gles/+merge/49333? [14:31] for arm we want clutter to use the gles backend [14:31] I ported the old modifications with some other fixes, and it's working fine on arm [14:32] pitti - not sure ;) [14:32] do you think it's ok to revert that change then? [14:32] rsalveti, hey, can you request review from ~ubuntu-desktop for it? [14:32] rsalveti, I will try to review it but not likely today, it's 3pm on a friday and I still have work I want to do before the end of the day [14:33] seb128: it's already, but I'm pinging you directly because you know more about clutter [14:33] seb128: don't worry [14:33] last time my merge proposal was there for a week [14:33] and then we got a package update [14:33] rsalveti, ok, thanks, I will review it when I've some free time, like on monday then [14:33] seb128: sure, np, thanks a lot [14:33] we don't plan clutter updates so it should be ok [14:33] thanks [14:34] seb128: and this only affect arm [14:34] right [14:37] ken was up late last night, i can vouch for that [14:37] as i was too :-\ [14:37] what did you guys work on? [14:38] kenvandine, was working on getting the new launcher API stuff working with xchat-gnome and i was working on empathy [14:40] he got it working, i'm quite close. if i reset unity then the count shows on the empathy icon, but it doesn't update.. and i have both unity and libunity from trunk [14:48] hey tedg [14:48] i'm fixing your dual build problems now [14:48] welcome to my hell :) [14:49] kenvandine, Heh, I think if I look at the commit logs you brought that to my world ;) [14:49] it was building when i committed :) [14:50] grr, not sure what I hate most now, gnome-polkit for crashing, update-manager to hanging when it does or compiz to screw when apport triggers because of update-manager [14:51] seb128, ah, another version of hell there [14:52] chrisccoulson: where does that come from? debian? I don't mind personally if you revert it, but I guess you should file a bug then [14:54] patti: Now for a real stupid question: Were do the human icons come from? Everything is enabled for them in the 3.3.1.1 package I just build, but the actual icons are missing (I used a copy of the tango icon for testing here). [14:54] pitti - yeah, it was merged from debian [14:55] Sweetshark: good question -- I assume they were an extra tarball in the openoffice.org source package? [14:56] seb128, got a problem with aptdaemon and policykit? [14:58] pitti: well, in all the versions I have here the are gone already ... gotta find an old package. [14:59] hmm [14:59] Sweetshark: http://paste.ubuntu.com/565898/ -- can't see them there [14:59] and yet [14:59] Package: openoffice.org-style-human [14:59] Source: openoffice.org [15:00] Sweetshark: so if you unpack the openoffice.org sources and apply the patches, there should be a images_human.zip file somewhere [15:02] glatzor, I think it's rather update-manager being stupid [15:02] tedg, i am doing the libappindicator builds again dance over here :-D [15:02] glatzor, it the gnome-polkit-daemon crashes it just goes in "waiting for a timeout having a blocked ui" [15:03] glatzor, or if you deny the pwd dialog [15:03] kenvandine, woot! [15:03] kenvandine, I was going through it last night and realizing that we really need a way to "distcheck" with GTK3. [15:03] kenvandine, ted: is the calendar indicator known to not display events which are not on the current day or on remote calendars [15:03] ? [15:04] tedg, kenvandine: btw what happened to indicator-appmenu, no tarball this week? [15:04] seb128, it is only for the default local calendar [15:04] seb128, Remote calendars yes, I didn't know about the events. [15:04] tedg, kenvandine: would be nice to have updates so we can keep testing menus with the current stack [15:04] seb128, Still working on them :( Got blocked on libappindicator... fun... yeah, that's it. [15:04] tedg, distcheck with gtk3 worked for me... when i did it :) [15:04] seb128: pfff ;) [15:04] seb128, indeed... tedg was busy messing with multi-builds [15:05] mvo, hey, your software is driving me crazy, it keeps screwing my desktop enough that I need to restart my session every second time I use it [15:05] kenvandine, hum, k, why is libappindicator blocking indicator-appmenu? [15:05] seb128, it really sucks we can't do remove events in indicator-datetime, but apparently that will be way to slow :/ [15:05] seems they are different code [15:05] seb128, not directly, i am sure [15:05] kenvandine, what? can't do? [15:05] just tedg was busy with fixing that instead of releasing appmenu [15:05] mvo, no offense to you ;-) [15:06] indicator-datetime doesn't display any events on remote calendars, apparently eds doesn't cache any of that [15:06] hum [15:06] it doesn't but it will for natty right? [15:06] so when the menu loads we would have to wait for the remote calendar to respond with events [15:07] why can't it be populated in an async way rather? [15:07] i don't think so, apparently it is a wanted feature in eds that is hard to implement [15:07] seb128: what about just fixing polkit ;) ? [15:07] the standard GNOME applet does it... [15:07] seb128: but yeah, let me have a look whats going on when the auth thing crashes [15:07] i think karl sadi eds doesn't have async methods for that [15:07] mvo, funny you speak about that it seems we get lot of crashes due to your session registration patch ;-) [15:07] seb128, karl said that is why the gnome applet takes so long to load [15:07] kenvandine, the old gnome-panel applet just does it [15:08] wth? [15:08] kenvandine, I just tried, the applet loads fine and event are displayed in an asycn way [15:08] dunno, chat with karl about it... that is how he described it to me a couple days ago [15:08] i was disappointed by the same thing [15:10] kenvandine, ok thanks [15:23] kenvandine: Hi, do you have an idea why I get that menubar from nautilus on my desktop? (http://www.bienia.de/tmp/Screenshot.png) (natty, classic desktop) [15:23] that's deliberate isn't it? [15:24] i don't like the nautilus menubar showing, i think it's weird [15:24] geser, yeah, what chrisccoulson said [15:24] can we get rid of it? :) [15:24] gnome-terminal shows me also a menubar even as I select no menubar by default === Artir is now known as afk|Artir [15:25] geser, yeah, that's expected too. what should happen is that the option in gnome-terminal should be hidden when there is a global menubar [15:25] i've done the same in firefox already (hidden the option to hide the menubar) [15:25] the classic desktop also gets a global menubar? [15:26] geser, i think it does now [15:26] seb128, aptdaemon.policykit1 should raise the AuthorizationFailed in this case [15:26] mine has got one anyway ;) [15:37] tedg, libappindicator uploaded and pushed, so you can merge to see how i fixed it === afk|Artir is now known as Artir [15:40] kenvandine, Great, will do! [15:40] kenvandine, appmenu-gtk on it's way. [15:40] woot [15:41] tedg, this is extra fun on a friday, there is no api breakages right? [15:41] kenvandine, Naw, minor updates. [15:41] kenvandine, Mostly keeping cadence really. [15:41] good [15:42] kenvandine, tedg, mterry: great work on the appmenu front, it's really solid [15:43] it really has gotta significantly better in the past week or so [15:43] kenvandine, We introduced some crashers on purpose so when they went away you'd feel like it's "so much better" ;) [15:44] i kind of assumed so [15:44] tedg, kenvandine: there is one libappindicator issue which is annoying [15:45] gnome-policykit-daemon crashes often in libappindicator code [15:45] in app-indicator.c l1279 [15:45] theme_changed_cb() [15:45] if (priv->dbus_registration != 0 && priv->connection != NULL) { [15:46] I didn't manage to get the issue under valgrind yet though [15:46] but it does often crash when I use update-manager [15:46] seb128, Hmm, okay. Probably not disconnecting a signal. === bjf[ack] is now known as bjf [15:47] what is weird is that there is no theme change [15:47] so not sure why this cb is called [16:04] bigon: is there a reason you added gjs back to natty? [16:05] GLib-GIO-ERROR **: Settings schema 'com.canonical.indicators.sound' is not installed [16:05] is that schema in some not-released-yet version of the indicator stuff? [16:05] seb128, I'm guessing that gsd shutting down or some such? [16:05] micahg: well it is needed for some developers, by gnome-shell [16:06] bigon: right, but gnome-shell isn't in the archive, I'd rather not support gjs if we don't have gnome-shell, it's currently staged in the GNOME3 PPA, does anything else need it [16:06] rodrigo_, should be in the current natty version [16:06] seb128, hmm, right, I see an update here [16:07] tedg, well I doubt so, it's like "use update-manager to install some binaries, then use it again to install some extra ones" [16:08] tedg, I tend to pick the few ones I want to try now then select other upgrades and do a second round [16:08] tedg, thinking about it, it could be after an icon cache update on disk [16:08] seb128, Hmm, interesting. [16:10] micahg: if there is ABI breaks from mozjs I will take care of rebuilding the package [16:10] bigon: I'm more worried about xulrunner transitions, can you commit to porting to future xulrunner releases? [16:11] micahg: well the pkg is maintained in debian too [16:11] do you expect such xulrunner transitions soon? [16:11] bigon: possibly, it depends on what can be worked out with Mozilla's new 3 month release schedule plans [16:13] in 3 months natty is release so I guess that GNOME 3 will enter the official achive [16:13] bigon: yes, but we don't need gjs for natty, we can add it back when gnome-shell hits the archive === bcurtiswx_ is now known as bcurtiswx [16:22] well the reason no other package use it is quite thin IMHO [16:31] bigon: when there are no problems supporting it sure === alecu is now known as alecu-lunch [16:44] my computer hates me today :( === vuntz_ is now known as vuntz [16:58] pitti, new pygobject tarball [16:58] just fyi [16:58] you probably want to call it a week after the meeting [16:58] it can wait for next week as well ;-) [17:04] seb128: guess what just finished building here :) [17:04] ;-) [17:04] * pitti is hanging out regularly in the gi channels now [17:04] seb128: I'm in release meeting, but I still want to test/uplaod this [17:10] seb128: syncing g-i from experimental now, FYI (you asked about it yesterday) [17:11] pitti, oh right, forgot to do that, thanks [17:12] tedg, [17:12] ==8284== Invalid read of size 4 [17:12] ==8284== at 0x40654BB: theme_changed_cb (app-indicator.c:1279) [17:12] ==8284== by 0x45C948B: g_cclosure_marshal_VOID__VOID (gmarshal.c:79) [17:12] ==8284== by 0x45AD351: g_closure_invoke (gclosure.c:767) [17:12] ==8284== by 0x45C0047: signal_emit_unlocked_R (gsignal.c:3252) [17:12] ==8284== by 0x45C8B28: g_signal_emit_valist (gsignal.c:2983) [17:12] but it doesn't give extra details about where it was freed before [17:13] seb128, Hmm, okay. I'll have to do an audit of that signal I think. === alecu-lunch is now known as alecu [17:21] kenvandine, Can you add bug 597317 to the indicator-session changelog please? I forgot :( [17:21] Launchpad bug 597317 in indicator-session "Text on shutdown and restart boxes is selected while tabbing" [Low,Fix released] https://launchpad.net/bugs/597317 [17:21] tedg, sure [17:21] ok, i officially hate swt-gtk and java [17:50] good night everyone! have a nice weekend === bcurtiswx_ is now known as bcurtiswx === bjf is now known as bjf[afk] === bcurtiswx_ is now known as bcurtiswx === bjf[afk] is now known as bjf [21:00] hi to all! [21:08] is somebody tracking banshee for natty? [21:09] I tried a backport of the new banshee versions to maverick, and it's totally unusable [21:09] I killed it at 4 GiB memory usage or so [21:09] :-( [22:51] jibel; do you still need that output? [22:54] patrickmw, yep, there is an issue that I'd like to investigate in LDTP which fails to resolve correctly keysyms to keycode. [22:55] jibel: https://pastebin.canonical.com/43253/ [23:01] patrickmw, thanks, that confirms what I thought. I'll fix that next week. Have a nice week end. [23:01] jibel: np, and hasn't your weekend already begun?! [23:04] patrickmw, Oh, you're right, 4 minutes ago ! I need to hurry now === bjf is now known as bjf[afk]