[05:51] Good morning [05:53] pitti, hey, do you understand how the documentation translations works with ubuntu? [05:53] robert_ancell: not differently than in upstream [05:54] robert_ancell: except that we can pull updated po files from LP [05:54] i. e. you can do an export for the package, stuff it in as a patch, and thus get updated translations during biuld [05:54] pitti, we have long standing bug 543282 where simple-scan help is not translatable [05:54] Launchpad bug 543282 in simple-scan "Documentation cannot be translated" [High,Triaged] https://launchpad.net/bugs/543282 [05:54] pitti, so is it just because there are no .po files there currently? [05:55] robert_ancell: how is the documentation build? Doesn't it take some C (locale, not programming language) input file and use some yelp-tools or similar to generate per-language docs? [05:55] pitti, yeah, it's just standard yelp tools, like gcalctool and most other GNOME projects [05:56] robert_ancell: so step 1 would be to ensure that your package builds a current .pot during build [05:56] pitti, so it doesn't do that, and no other GNOME package does afaict [05:56] robert_ancell: I suppose the .pot (i. e. the domain) for the help is different than the runtime domain [05:56] robert_ancell: in fact, most gnome packages do, by way of cdbs gnome.mk [05:56] or dh_translations [05:57] robert_ancell: but yes, the immediate cause of non-translated docs are certainly missing po files, provided that the build system is set up to use them if they are present [05:57] robert_ancell: reviewing your 3.4.1 uploads, thanks [05:57] good morning [05:58] hey didrocks [05:58] pitti, yes, please have a close look at the vte one. It has the most changes, but they seem ok [05:58] guten morgen pitti :) [05:58] robert_ancell: hm, we usually upload to -proposed nwo [05:58] pitti, last I asked it was back to precise! [05:58] robert_ancell: for most packages it's harmless enough, but for vte I'd prefer proposed as it has more potential for archive breakage [05:59] ok [05:59] feel free to kick it back then [05:59] robert_ancell: that only lasted for two days until LP was fixed [06:00] robert_ancell: vte seems reasonable, assuming that you tested with gnome-terminal [06:00] robert_ancell: I'll reject it, and let you reupload to -proposed, and then accept it right away [06:00] pitti, yes, been running it for a few hours [06:00] robert_ancell: (I trust that the diff is going to be the same) [06:01] * robert_ancell looks devious [06:05] robert_ancell: accepted the others [06:05] I updated the pad [06:05] so that leaves brasero, yelp, cheese [06:05] err, brasero is in [06:05] robert_ancell: are you still working on updates? If so, can you please claim them in the pad? [06:05] I'll take some others then [06:06] pitti, nope, I'm done for the day [06:07] pitti: I'm testing a fix for the software-center animation icon issue (just get it few minutes ago). if that fixes it for me, what do you think about sneaking this other fix as well (one line fix, making the workspace section keys of the overlay translated): http://bazaar.launchpad.net/~unity-team/unity/trunk/revision/2283 [06:09] didrocks: LGTM [06:10] pitti: thanks, building and trying on both my laptop and netbook the fix [06:14] hey didrocks [06:14] didrocks, what do you think is less risky, reverting or moving forward with Jay's fix? [06:14] bonjour rickspencer3 ;) [06:14] I'm a bit worried about piling regression on top of regression [06:16] rickspencer3: the fix doesn't look worrying. As I have one hw which can shows the issues, I can test it, and as well on my laptop. Reverting the issue make revert other memleak fixes and is less tested. So I propose to pick the fix, push into -proposed, having some confirmations all is fine, and go with it if so [06:17] hey rickspencer3 [06:17] hey pitti [06:17] didrocks, ok, let me know when it's built in proposed, and I'll update my netbook [06:18] rickspencer3: will do :) building locally right now and ensuring everything is fine first :) [06:28] pitti, oh, did you solve that gtk+ issue? [06:28] robert_ancell: if you mean the "kills session during build", no [06:34] pitti, doing gnome-games update right now [06:50] pitti, hey, do you follow the valac packaging? [06:50] not closely, why? [06:51] pitti, it's just that apt-get install valac installs the 0.14 version, when it really should do the 0.16 version. I don't know how we're supposed to handle the packaging correctly - currently valac is build from the 0.14 source package [07:02] rickspencer3: pitti: ok, I've done 10 installs/uninstalls in a row without any issue. You have to move the mouse on low configuration to trigger a refresh to get the animation, but it's not a new issue. Will publish it to -proposed. [07:03] hey didrocks [07:03] let me know when it's built and I'll install and test [07:03] didrocks, I'll run through the unity tests to make sure I don't hit any other issues [07:04] rickspencer3: ok, the change is just in functions that are called by software-center installing the application FYI [07:04] (dbus triggered) [07:05] didrocks, you don't know how many times in my life I have heard "there is no way this could affect code outside of foo" [07:05] ;) [07:05] fortunately, we have a good set of tests to run through , so regressions should show up quickly [07:05] rickspencer3: I didn't tell there is no way, I just told "the functions that were touched are only called in that case" :p [07:06] lol [07:11] ah pitti beats me at totem-pl-parser while I was looking at unity :) === tkamppeter_ is now known as tkamppeter [07:32] heh [08:05] hey [08:06] hey seb128 [08:07] hey didrocks [08:15] bonjour seb128 [08:15] oh, a pitti! [08:15] pitti, it's meeting reminder day! ;-) [08:15] http://reports.qa.ubuntu.com/reports/bug-fixing/canonical-desktop-team-precise-fixes-report.html [08:15] * pitti ^5s seb128 [08:15] seb128: oh, thanks [08:15] pitti, how are you? I hope the 3am uploads are delayed approval and just not you going to be at 3am to start again at 6am ;-) [08:16] * seb128 hugs pitti [08:16] seb128: nah, I went to bed at 0:30 [08:16] and got up again at 7:15 [08:16] pitti, tie! good, don't touch anything out of proposed now ;-) [08:16] I was just awake [08:16] seb128: hehe, same for you! [08:16] deal! [08:17] seb128: I need to do an apport and kerneloops upload to disable for the final release, as well as langpacks, but none touch bugs, so you are safe [08:17] seb128: I have a better idea [08:17] we merge ourselves into ~ubuntu-desktop-bug-rockers [08:17] and then beat the didrocks team with a whopping 670 bugs!! [08:17] take that, didrocks! [08:17] we would kick didrocks's butt like that :p [08:18] pitti, "left in pool: 0" [08:18] nice the retracers managed to catch up during the night [08:19] seb128: yeah, I happened to look at the retracer this morning at exactly the time when it was down to 5, and saw it grind down to 0 in 10 seconds [08:19] yeah yeah yeah ;) [08:19] I seem to have a knack for good timing today [08:19] ;-) [08:19] /msg didrocks for ten beer we add you to the team and we'll break the 1.000 mark [08:20] pitti: ten beers, are you cheap like that? ;) [08:20] didrocks: when I have a self-build unity, is it enough to just run "./unity --replace" from ./build ? [08:20] hey mvo [08:20] mvo: no, you would need to install it [08:20] :( [08:20] mvo: sudo make install then [08:20] mvo: but! [08:20] how do people test? [08:20] mvo: I did sudo unity --disto [08:20] distro* [08:21] to remove all the locally installed thingy [08:21] hm, nobody came up with a way to run unity from the build tree? [08:21] oh, mvo is back [08:21] mvo, hey, how are you? [08:21] that's like the first thing I always do when working on a source [08:21] pitti: not possible with compiz :/ [08:21] hello, I'm good, thanks! [08:21] it sucks to not be able to do that [08:21] pitti: I meant, some work on compiz refactoring is needed [08:21] didrocks: well, with some $COMPIZ_PLUGIN_PATH magic or what not [08:21] of course it often needs some LD_LIBRARY_PATH and similar effort [08:22] there is a way to get it in ~/compiz-1, but you still need to have your images installed [08:22] (and the gsettings key and…) [08:22] didrocks: its a core chang to the launcher for the s-c integration, still needs a full install? [08:22] mvo: what you are testing? [08:22] didrocks: unity launcher integration issu(es) [08:22] mvo: don't start frightning me with changes in the s-c integration, we just got a fix [08:23] mvo: you mean, the issue on intel, like your session being screwed? [08:23] (all display not working) [08:24] didrocks: no, much simpler, just that the signal handler does not track success vs failure of the install and adds the launcher unconditionally [08:24] didrocks: even on error or cancel [08:24] didrocks: tiny chnage [08:24] mvo: ok, not the same issue then. SRU for that one ;) [08:24] mvo: you need to make install I'm afraid [08:29] didrocks: ok, so could I just copy the one "launcher" binary easily? [08:29] mvo, you can probably sudo cp unityshell.so over the system version and restart compiz [08:29] seb128: mvo: no [08:29] that doesn't work [08:29] didrocks, ? [08:30] mvo: you will need to symlink /usr/local/share/unity so /usr/share/unity if you do that [08:30] didrocks, the code is not in the unity binary? [08:30] as unity will look for assets in /usr/local/ [08:30] didrocks, well when I say "system version" it's the /usr one, the one shipped by the package [08:30] that's why I did sudo unity --distro which is easier to revert to the distro state [08:30] that's how I would do it anyway [08:30] just copy over the distro .so [08:31] and install --reinstall unity when done [08:31] didrocks: can I give it a build prefix? [08:31] seb128: still, if you compile, but default PREFIX=/usr/local and unityshell plugin will look at this one [08:31] seb128: wherever it is [08:31] mvo: -DCMAKE_PREFIX_PATH=/usr should work [08:32] (last time I checked, but compiz is doing some override in its cmake file and it changed a lot in the precise cycle) [08:35] didrocks: thanks a bunch [08:36] mvo: btw, the shell binary is libunityshell.so FYI :) [08:37] didrocks: I really haven't moved with the recent c++ goodness, but seeing that you can (apparently) almost have inline callbacks in SoftwareCenterLancherIcon.cpp (to connect to the finished sigal) is pretty cool [08:37] my patch build! [08:37] * mvo uploads it [08:37] sure sure, upload :p [08:37] mvo: indeed, you can have inline callbacks, this is nice ;) [08:39] didrocks: that is c++0x, right? the lambda stuff? [08:39] mvo: right, lambdas are part of c++0x, look as wel as the Ptr:: template, it's great to have ref that are not weaks ;) [08:41] didrocks: nice, I think I need to do a bit more c++ again (boy, I never thought that I would say this ;) [08:41] mvo: bored with python? ;) [08:42] didrocks: I still love it, but its a bit on the slow side [09:25] didrocks: I have a branch for you guys for the USC stuff [09:26] mvo: please propose against lp:unity, it will be reviewed :) [09:26] (we mandate tests btw) [09:31] moin! [09:32] pitti, I've just noticed that yesterday's language pack export is a full one, as it picked up your "full export" request. I'd still like to request one today, as it's the deadline expected by translators, but we can use the one from yesterday for testing or as a fallback if something goes wrong. [09:32] https://translations.launchpad.net/ubuntu/precise/+language-packs [09:32] (assuming it got exported correctly :) [09:33] brb [09:36] dpm: yes, I agree [09:36] dpm: I ticked "request full export" again [09:36] ok, cool [09:41] kamstrup: hey [09:41] kamstrup: are you still interested in gnome bug 667244? [09:41] Gnome bug 667244 in introspection "Arrays of variants passed from C to Python always marshal to empty arrays" [Normal,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=667244 [09:41] kamstrup: if you are, and you want to update the patches to current git, I can promise you a timely review [09:41] pitti: any hints on bug 919659. oneiric without updates and 3.4.6-0ubuntu1~ppa1 => no bug, oneiric with all updates installed and 3.4.6-0ubuntu1~ppa1 => bug, oneiric reenabling the update repositories in the sources.list one by one from bottom up => no bug [09:41] Launchpad bug 919659 in libreoffice "Can't open/save document or spreadsheet with password" [Unknown,Confirmed] https://launchpad.net/bugs/919659 [09:42] pitti: seems some other package fubared their maintainer script slightly [09:43] pitti: yeah, still interested, i'll see if I can find some time to forward port them [09:43] Sweetshark: with "... and 3.4.6..." you mean "updated from oneiric-proposed to precise plus that PPA? [09:43] pitti: nope, on oneiric only. I dont see this on precise. [09:47] Sweetshark: I did a glimpse on the bug trail, but it's too long to read and understand quickly [09:47] Sweetshark: what does password decryption have to do with maintscripts? [09:48] Sweetshark: could be that some libnss version in oneiric-updates causes this? [09:48] hm, libnss3/libnspr4 are unchanged in oneiric-updates [09:49] you should ask for comparing the package versions of all dependencies [09:49] "dpkg -l" output before/after [09:50] didrocks: unity ready to copy, want to or shall I? [09:51] pitti: yes, libnss or some related security package. or some different package (maybe firefox/thunderbird) doing some registration stuff or somesuch against libnss on update. [09:51] pitti: I have an available slot to do it today :) [09:51] today/now ;) [09:55] hi everyone, i have a quick question wrt gtk-themes not showing up in unity's theme selector, already tried the -design channel, where could i ask? [09:56] *else [09:57] seb128: btw, orca has a lot of changes, I'm not comfortable with uploading 3.4.1 without proper testing; so I think I'll leave that to TheMuso and an SRU [09:57] ah, gnome-user-docs finally syncs at last [09:57] * pitti does [09:59] pitti, ok, works for me [10:00] seb128: do you still see anything urgent? versions.html looks quite fine here [10:01] pitti: would it be possible to SRU libreoffice 3.4.6 as is as a) the bug also shows in the current 3.4.4 in oneiric and b) likely is not caused by libreoffice itself and would need fixing elsewhere? [10:01] pitti, no, I don't think anything this week was "urgent" in any case, just "bonus" [10:01] pitti, I think we are good [10:02] Sweetshark: sure; however, even 3.4.5 didn't make it out of -proposed yet [10:02] pitti: yes, because of that. [10:02] Sweetshark: bug 915223 is v-failed [10:02] Launchpad bug 915223 in libreoffice "Update LibreOffice on oneiric to 3.4.5" [Medium,Fix committed] https://launchpad.net/bugs/915223 [10:04] bug 915223 is not a regression 3.4.4 -> 3.4.5 -- it is a regression injected somewhere else and also affects 3.4.4 [10:04] Launchpad bug 915223 in libreoffice "Update LibreOffice on oneiric to 3.4.5" [Medium,Fix committed] https://launchpad.net/bugs/915223 [10:20] pitti: arggh. no thats not quite true. I see no bug report for 3.4.4. So: 3.4.4 with or without updates -> no bug, 3.4.5/6 without updates -> no bug, 3.4.5/6 with updates -> bug. [10:20] *sigh* [10:21] ok, so it's a regression after all [10:23] yeah. I should just have been quicker with the 3.4.5 update. than it would be the trouble of the guy who did the changes causing it. [10:34] Sweetshark: btw, did you see my ping last week about the libroffice quicklist bug not being translated? [10:36] didrocks: known issue [10:36] Sweetshark: do you see any way to get it fixed? what we did at the time was setting up a wiki page for firefox as it wasn't using some po files [10:38] pitti: you were reminded of the meeting and the report page, do you desperatly want another reminder (and a hug) or I deleted your email without noticing? ;) [10:39] didrocks: argh, forgot; doing now [10:46] oh, my. i've just a second delivery of beer to my house. except, i didn't order this one :/ [10:47] chrisccoulson: did you have to pay for it? :) [10:47] didrocks, i think it's probably a gift from someone [10:48] or firefox submitting twice :p [10:48] didrocks: the stuff is in, but the patch needs a fix [10:48] it has one of these in it: http://www.brewdog.com/product/tactical-nuclear-penguin [10:48] someone obviously knows what i like! :) [10:48] Sweetshark: so, it's under work? :) [10:49] chrisccoulson: heh, great ;) [11:05] chrisccoulson, you maybe got some firefox users thanking you for all the work you do? ;-) [11:05] seb128, they're from my parents ;) [11:05] i just phoned them [11:05] lol [11:05] there go the happy user theory :p [11:14] seb128, https://twitter.com/#!/chrisccoulson/status/192209221775917056/photo/1 ;) [11:16] right, now to go back to looking at scrollbars [11:16] chrisccoulson, now is not beer'o'clock yet! [11:16] heh [11:16] though you might need one if you start looking at the scrollbars [11:16] yes, this could drive me to drink [11:18] oh! [11:18] hum, no, ignore me ;-) === MacSlow is now known as MacSlow|lunch [11:50] didrocks, pitti, others: retracers dups list since yesterday (I ran it again due to the i386 backlog since it was broken between saturday and yesterday evening): [11:50] http://pastebin.ubuntu.com/933854/ [11:50] looking :) [11:50] ups, http://pastebin.ubuntu.com/933856/ [11:50] well same list mostly [11:51] yeah, seems mostly the same [11:51] in fact it's the same, I though I had copied the wrong one first but no :p [11:52] the top is similar to yesterday's as well [11:52] heh :) [11:52] i.e no "surprise" [11:52] good good! [11:52] it's a bit surprising to see openbox in there ;-) [11:54] indeed, it's maybe the new hype, we should tell it to the omgubuntu people :) [11:54] didrocks, FYI c-p-e did build now [11:54] (on arm that is) [11:55] ogra_: oh? didn't notice [11:55] ricardo just disabled all GL plugins in the arm builds :) [11:55] ogra_: ahah, quite radical ;) [11:55] ogra_: ah, it's not uploaded [11:55] it is [11:55] I thought it was :) [11:56] last night [11:56] * didrocks refreshes launchpad [11:56] ah, here we got [11:56] go* [11:56] ogra_: looks fine to me, good move :) [11:56] :) [11:57] i doubt anyone will notice though ... [11:57] ogra_: I doubt as well, but a lot of people noticed on i386 and am64 when the ABI break :) [11:57] ogra_: can you pull the change in the vcs? [11:57] ogra_: lp:~compiz/compizfusion-plugins-extra/ubuntu [11:57] is c-p-e also maintained the same way ? [11:57] k [11:57] it is for simplicity [11:58] as we need to rebuild it at each ABI break, easier for tracking [11:58] yup [11:59] thanks :) === greyback is now known as greyback|lunch === MacSlow|lunch is now known as MacSlow [12:23] seb128: argh at the nautilus xcb_io thingy [12:24] pitti, the _XAllocID one? [12:24] yes, hate it, I get a least one dup a day in my mailbox [12:24] yes; this seems to be a multi-thread race [12:25] not sure [12:25] it happens with i.e apport, update-notifier, gnome-control-center [12:26] looking at my firefox awesome bar list when typing _XAllocID [12:37] some people should really not file bugs when they are drunk: bug #981308 [12:37] Launchpad bug 981308 in oneconf "oneconf-service crashed with OSError in save_json_file_update(): [Errno 2] No such file or directory" [Undecided,New] https://launchpad.net/bugs/981308 [12:39] it's interesting that this can crash though: [12:39] new_file = file_uri + '.new' [12:39] with open(new_file, 'w') as f: [12:39] json.dump(content, f) [12:39] os.rename(new_file, file_uri) [12:39] mvo: pitti: do you have any idea what can happen that json doesn't give any error and os.rename still raise an OSError? ^ [12:51] didrocks: the directory might not be writable? [12:52] pitti: in that case, json.dump will error first, isn't it? [12:52] didrocks: oh, no, "no such file or dir" [12:52] didrocks: perhaps file_uri is an actual URI? [12:52] you can't rename something to file:///foo or http:/// foo [12:53] didrocks: i. e. the name might have a slash in it? what is file_url? [12:53] pitti: it shouldn't, it should be /home/user/.cache/oneconf// [12:53] ok, then the name is misleading [12:53] didrocks: but I'd say theh directory of file_uri does not exist [12:53] yeah, I need to rename it [12:54] it's the "90% likely" explanation [12:54] oh, wait [12:54] pitti: file_uri is what it will be written to, right? [12:54] I disreagarded the first line [12:54] i. e. new_file = [12:54] yeah [12:54] it's just dumping and renaming [12:54] didrocks: could file_uri be a dangling symlink? [12:55] pitti: shouldn't as I created it in the first place the same way [12:55] that or some other process removing the file in between are the only off-hand explanations that I have [12:55] didrocks: then I'm afraid this needs an strace [12:55] yeah [12:56] but no dupes, so most likely hard to find out [12:56] right [12:56] and given the tone of the description the reporter might not be too keen on running strace [12:56] or figuring out what happened [12:56] pitti: that's one of my concern :) [12:56] I'll ask him if he can reproduce and run strace if he can [12:56] then incomplete and we'll see ;) [12:58] good luck [12:58] heh ;) [13:19] Sweetshark, hi, I assigned you bug #926940, it's a trivial 1 liner fix to a .desktop, would be nice to get in the next upload [13:19] Launchpad bug 926940 in libreoffice "Visio files not associated with LibreOffice Draw" [Low,Confirmed] https://launchpad.net/bugs/926940 === greyback|lunch is now known as greyback [13:29] jbicha, pitti, dpm: is the ubuntu desktop guide supposed to be translated through langpacks? or does it need an export and upload? [13:29] i.e an import from the po in a source package because it builds localized xml [13:30] yohoo [13:31] is here the place to wonder how chooses what gets to be default app in the desktop? [13:47] hi seb128, no, it's not supposed to be translatable, but thanks for checking it out! [13:47] we disabled it in Launchpad a while ago [13:47] dpm, how so it's not supposed to be translatable? [13:47] dpm, you mean our desktop documentation is english only? [13:48] dpm, how broken is that?! [13:48] seb128, ah, wait, we're talking of different things, I think :) [13:48] dpm, I'm talking about what you get if you run "help" from the dash [13:49] ah, that's the ubuntu-docs package, which is indeed translated [13:49] dpm, well not here, I was trying to figure how to debug ;-) [13:49] generally the docs team (I think mdke) takes care of the uploads, but I think the final upload hasn't happened yet [13:49] seb128: I think it's more complicated -- in general, the translated help is built during package build [13:49] dpm, like if that was supposed to be in langpacks or we lack an upload with exported translations [13:49] seb128: but we strip out translated help and stuff them into the langpacksc [13:50] pitti, ok, so we need an upload with an export [13:50] ? [13:50] seb128: yes, I guess so [13:50] pitti, thanks ;-) [13:51] jbicha, hey, do you know if that's planned? ^ [13:51] yes, but I would have thought the docs team would have taken care of it last week, it seems it hasn't happened yet: [13:51] https://launchpad.net/ubuntu/+source/ubuntu-docs/+publishinghistory [13:51] which means if language packs are exported today, there is probably no time to get the ubuntu-docs translations imported into LP and exported into the final language packs [13:52] hm :/ [13:52] dpm, thanks, I will follow up with jbicha [13:52] :-( [13:53] I'd like to release language packs very shortly after release, which will contain the translations, but it would have been nice to include them into the final images [13:54] bbiab [13:54] seb128, if you follow up with jbicha, and you guys do the upload, let me know. Perhaps we can get LP to prioritize the translations exports and get them there on time, but it might be quite tricky [13:55] dpm, thanks [13:57] hum [13:57] dpm, pitti: [13:57] $ dpkg -L language-pack-fr-base | grep ubuntu-docs [13:57] /usr/share/locale-langpack/fr/LC_MESSAGES/kubuntu-docs-directory.mo [13:57] /usr/share/locale-langpack/fr/LC_MESSAGES/kubuntu-docs-kubuntu-docs.mo [13:57] $ [13:57] that seems buggy [13:58] I don't have direct upload rights for ubuntu-docs [13:59] jbicha, hey, do you know what's the deal with translations there? are they supposed to come as translated xml in langpacks or...? [13:59] well I don't think my issue is lack of uploads [14:00] they're part of the lang packs [14:00] like the 12.04.3 upload is recent enough [14:00] jbicha, how are they named? [14:00] jbicha, I don't find them in my langpacks [14:01] hum [14:01] they should be in /usr/share/help/fr/ubuntu-help [14:01] so /usr/share/help-langpack/de/ubuntu-help/about-this-guide.page [14:01] why don't we get the same for french [14:01] seb128: oh you're German today? [14:01] jbicha, no, but I've one for de and not one for fr! [14:02] maybe you should be! [14:02] I blame it on pitti [14:02] yeah, I can feel a trap there ;-) [14:02] jbicha, how much did pitti and mvo pay you? ;-) [14:04] dpm: what is involved in doing the ubuntu-docs upload for translations? will LP automatically merge in the latest translations or do they have to be manually injected into the package before upload? [14:04] dpm: hi, I'm sure I asked this before, but https://translations.launchpad.net/ddtp-ubuntu/+imports shows a bunch of files in "needs-review" state, who does the review and when? [14:04] is here the place to wonder who chooses what gets to be default app in the desktop? [14:05] matanya, sort of, we usually use the mailing list for discussion, it's less restricted fir participants than IRC (like you don't need to be online at the time it's discussed), we discuss it at the ubuntu developer submit as well [14:06] matanya, why? [14:06] hey mvo, looking... [14:06] I though to suggest clementine for default music-player [14:07] *thought [14:07] hoho [14:07] seb128: just not sure where or to whom [14:09] matanya, note that for such a change you need to take a ton of things into account (is the toolkit it uses on the CD already, if not, weher does the space for the toolkit come from, does it work with the sound-indicator menu, if not, who will implement that function, is it integrated with the ubuntu musoc store ... etc etc) [14:09] matanya, emails https://lists.ubuntu.com/archives/ubuntu-desktop/ with a rational of why you think we should use it, what pro it has, how active are upstream, if they are willing to support it in Ubuntu, etc [14:10] well, all is answered, as far as I know [14:10] can I share some thought here? [14:10] where? [14:10] so its not a "oh, lets switch to that shiny app because its cool" thing, but a lot of stuff that has to be taken into account for such a switcdh [14:10] sure [14:10] to start clementine is written in Qt right? [14:10] which means look and feel different from our other apps [14:10] that's not a blocker but something to thing about [14:11] clementine <3 [14:11] is it Qt ? i thought it was efl [14:11] mvo, in principle, if there is a template for the translations, LP will import them directly and you shouldn't worry about them. Right now the overall imports queue is quite full, so it might take a while to import them. Did you do a manual upload of translations, or did you commit them to the branch? [14:11] well, it is in the sources already, very mature, supports many music sources, works with the sound indicator [14:12] even has a scope for use in the unity-shell [14:12] right, all of that banshee and rhythmbox have as well ;-) [14:12] and they use our standard toolkit [14:13] i didnt mean to disappoint you with my words above ... there are just many things to think about before doing a switch :) [14:13] and they are much heavier and bloated [14:13] not to mention banshee uses mono [14:13] so you will firstly need a mail discussion (or show up at the developer summit in person or remotely), other people might bring up other issues etc [14:13] matanya, did you try rhythmbox? it might have defaults but bloated is not one [14:14] banshee is out [14:14] matanya, it's written in C and quite efficient [14:14] it took about 3.5 minutes to open on my 11.10, and drank 3.5gig of ram [14:14] matanya, well anyway that's why those discussions are better on list and with a strong argumentation rational on what we would win [14:15] I see your point [14:15] though no dought qt isn't a good Idea [14:16] oh not another default music player argument [14:16] dpm: commited them to a branch [14:16] dpm: how long is long? hours? days? [14:16] how can I make g_debug() messages shown on the terminal or a logfile? G_DEBUG=foo doesn't seem to cut it [14:17] dobey: don't worry, I'm done [14:17] tjaalton, G_DEBUG_MESSAGES=all ... [14:18] seb128: ahh, thanks [14:18] mvo, I'd say a couple of days with the current state of the queue. Let me pm you the same info I gave Riddell earlier on [14:18] tjaalton, or "all" -> "domain" [14:18] dpm, jbicha: I need help to understand that ubuntu-docs langpack magic [14:18] dpm, jbicha: where,how "/usr/share/help-langpack//ubuntu-help/about-this-guide.page" are built? [14:19] seb128, I don't know much about how the package is built, only the overall process: [14:19] 1. Translations need to be exported [14:19] 2. The package gets uploaded and built with the po files [14:20] 3. Launchpad imports the translated xml files [14:20] ok, so they are built from the source and collected on the builders [14:20] so the source needs an export from launchpad [14:20] 4. Language pack export picks up the translated xml files [14:20] * jbicha heading out for 15 minutes [14:20] jbicha, stay here!!! ;-) [14:20] 5. Langpack-o-matick produces the language packs containing the translated xml files [14:21] dpm, ok, I think I got it, it's all based on the source to build with updated translations [14:21] so we need an upload with a current export [14:21] so the updated xml go through the system [14:21] upload = package upload and build, which generates the translated xmls, which are sent through the archive [14:21] seb128: doesn't seem to work either, and google didn't give any hits about it [14:22] tjaalton, what are you trying to do? [14:22] tjaalton, is that g-s-d? [14:22] seb128: see the messages from g-s-d [14:22] but wait [14:22] tjaalton, G_MESSAGES_DEBUG=all gnome-settings-daemon --debug ? [14:22] ahah :) [14:23] there we go [14:23] thanks [14:23] tjaalton, yw [14:26] seb128, also if you are working on the upload and want to bypass waiting for LP to give you the PO files, you can fetch them from the automatic exports branch: https://code.launchpad.net/~ubuntu-core-doc/ubuntu-docs/precise-translations - the only caveat is that you'll need to put them in the right place in the source tree for the package (which you'd also have to do if you were exporting a translations tarball from LP anyway) [14:27] dpm, well, I'm trying to get a reply to one of the ubuntu-docs guys, I get they have an auto export configured in some way? [14:27] dpm, i.e a script or something that do that for them [14:27] that's the one ^, I worked with them to set it up [14:28] dpm, well I would like to avoid the "you need to put them in the right place of the source tree" [14:28] why can't we just bzr merge ;-) [14:28] like lightdm and other project have translation automerged by a bot or launchpad it seems [14:28] but I guess that's a question for next cycle [14:30] seb128, yeah, unfortunately it cannot be easily solved, as it follows the gnome-doc-utils layout (a folder per locale), which is different than standard gettext (flat layout, all locales in one folder). Launchpad only supports the latter. [14:30] dpm, oh ok, that makes sense, I got the picture, thanks ;-) [14:30] np ;-) [14:31] dpm, thanks for all the details, I will try to pick it up from there [14:31] excellent, thanks seb128 [14:42] jbicha, wb! [14:45] seb128: hi, so I'm going to try to export the ubuntu-docs translations into the ubuntu-docs branch, then I'll just need someone to upload the new package [14:46] jbicha, you know how to do that? [14:46] jbicha, I've an export of each locally I was just looking at it [14:48] jbicha, can you merge in https://code.launchpad.net/~giuseppeterrasi/ubuntu-docs/fixlink/+merge/99265 while you are at it? [14:48] seb128: I believe you just copy the fr.po and so on into the ubuntu-help/fr/ folder [14:49] seb128: I can't merge that in at the moment because it would break the string freeze, it is a candidate for the SRU we'll be doing though [14:51] jbicha, ok [14:52] jbicha, so basically [14:52] bzr branch lp:~ubuntu-core-doc/ubuntu-docs/precise-translations [14:52] bzr branch lp:ubuntu-docs [14:53] ls ubuntu-docs/ubuntu-help > list [14:53] for locale in $(cat list); do cp precise-translations/ubuntu-help/$locale.po ubuntu-docs/ubuntu-help/$locale; done [14:53] with list edited to remove makefile.am, etc [14:53] jbicha, ^ that seems to work [14:54] ok, my bash scripting is a bit rusty but that sounds about right [14:54] jbicha, if you want to do that, I don't have commit right [14:54] or do you want me to publish the diff? [14:55] I don't fancy to push an ubuntu-docs vcs copy it's not small but I can put a diff online [14:55] seb128: sure I'll upload your diff [14:56] Hi! [14:56] sil2100, hey [14:58] jbicha, http://people.canonical.com/~seb128/ubuntu-docs.diff.gz [14:58] jbicha, it's to gunzip and patch -0 < ubuntu-docs.diff from the ubuntu-docs dir [15:01] jbicha, of course it would be too easy if translations were not buggy and breaking build after that :-( [15:09] tjaalton, you can probably use i.e G_DEBUG_MESSAGES=wacom to filter to wacom debug messages [15:13] seb128: tried that, but couldn't figure out what to use there [15:14] tjaalton, well = should work for gsd [15:14] output looks the same [15:15] even without the env variable [15:19] tjaalton, --debug do a setenv G_DEBUG_MESSAGES [15:19] iirc [15:20] jbicha, ok, so el it sl are buggy, also Makefile.am had a HELP_LINGUAS = ... en_UK ... which should be en_GB [15:20] jbicha, I did bzr revert the el one to build, doing the same on it and sl didn't work, I had to copy the versions from your previous upload (I just wanted a build done) [15:21] jbicha, you maybe want to complete HELP_LINGUAS as well, it lists only 28 locales and there are about 50 now, I think it might mean the other ones are not built as they should [15:23] the only language in the Makefile.am that didn't build for me were en_UK (fixed now) and zh_TW (which I just fixed) [15:23] jbicha, weird [15:23] el and sl are out of date in your diff then? [15:23] jbicha, I just did bzr bd lp:ubuntu-docs with my diff applied [15:24] jbicha, no, the diff was the full update [15:24] jbicha, with the diff I did hit the issues I listed [15:24] jbicha, they I locally hacked around them [15:25] jbicha, [15:25] $ dpkg -c ubuntu-docs_12.04.4_all.deb | grep help/id [15:25] $ [15:26] so yeah, all the locales not listed in HELP_LINGUAS are not built [15:26] I hate LINGUAS [15:26] or how to maintain a separate list of the locales rather than just using all the .po in the dir [15:29] several of those other languages are very incomplete but I guess a bit of translation is better than none at all? [15:29] maybe it'll encourage people that speak that language to help out [15:29] yes [15:29] jbicha, well, fr lacks 400 strings... [15:30] so it's not like all listed ones were great [15:30] though the french guys were discussing how to "merge" the gnome-user-guide translations in ubuntu-docs since some are common [15:30] dpm, ^ that's probably something we should look at automate next cycle [15:31] seb128, I know, we've been discussing it on the mailing list. I think the ubuntu-docs tree simply needs a script to fetch the gnome-user-guide translations and merge them [15:31] pitti, meeting time in case somebody has a topic, if you want to use your ping alias ;-) [15:34] yep, indeed [15:34] jasoncwarner, Sweetshark, bryceh, chrisccoulson, didrocks, tremolux, Riddell, kenvandine, cyphermox, mterry, seb128, tkamppeter, pedro_, desrt, agateau: anything to discuss at the meeting? [15:35] nothing for me [15:35] https://wiki.ubuntu.com/DesktopTeam/Meeting/2012-04-17 [15:35] nothing here [15:35] pitti, nothing here [15:36] seb128: hmm, you're right, the build does stop at el; I was just doing autogen.sh; make which worked though [15:37] nothing from me either [15:41] jbicha, ok, so I had only those 3 which is a good news :p [15:42] stupid shell question, how do I change '\n' to ' ' in a file? like I can to concatenate lines [15:43] oh, fantastic. my esta doesn't expire until august [15:43] chrisccoulson, lucky you, mine experied in january, need to pay :p [15:43] I wonder if I need another one if it turns out my passport is lost ... [15:44] Laney, yes [15:44] bah [15:44] * Laney needs to look harder [15:44] Laney, you need new one if you change address, passport, or any of the infos "linked" to the esta [15:45] Laney, well I would say that if you lost your passport you have another issue ;-) [15:45] can you get a new one on short delay? [15:45] yes, there's a one week service [15:45] oh, lucky you [15:45] which I will invoke if I do not find it before, say, Friday [15:45] in France you are screwed if that happens [15:45] it takes like over a month and no short line [15:46] unping for the shell question [15:46] tr '\n' ' ' for the win [15:48] nice! i'm not sure you can even do that with sed, can you? [15:48] jbicha, http://pastebin.ubuntu.com/934150/ [15:48] jbicha, that's the new language list I'm testing [15:48] chrisccoulson, I'm not sure, you can do anything with sed if you know enough magic :p [15:49] but I don't [15:49] I tend to | in awk and tr [15:50] wow, https://bugzilla.mozilla.org/show_bug.cgi?id=733892#c28 is crazy [15:50] Mozilla bug 733892 in Widget "crash aswJsFlt.dll (Avast) and Ant extension" [Critical,New: ] [15:50] chrisccoulson, lol [15:51] they patch binaries? [15:51] inline? [15:51] seb128, apparently some anti-virus vendors do [15:51] utch [15:52] jbicha, good news is that none of the new locales create build issues [15:53] seb128: yeah I've added all of those now too, and am trying to fix the remaining build issues [16:06] seb128: any idea about what this means? http://paste.ubuntu.com/934171/ [16:09] jbicha: maybe missing encoding header in the .po file? [16:11] jbicha, no, seems to be an invalid char somewhere in that file? [16:23] jbicha, ok, found it [16:27] jbicha, I think I found both, test building [16:34] jbicha, http://pastebin.ubuntu.com/934217/ [16:34] jbicha, those are the fixes [16:36] micahg, hi, could you look at bug #956124? [16:36] seb128: Error: Could not parse data returned by Launchpad: The read operation timed out (https://launchpad.net/bugs/956124) [16:37] bug #956124 [16:37] Launchpad bug 956124 in blueman "nautilus-sendto crashed with SIGSEGV in g_return_if_fail_warning()" [High,Confirmed] https://launchpad.net/bugs/956124 [16:37] micahg, since you are the one who uploaded the new version [16:37] i.e the closer from a maintainer in Ubuntu for it ;-) [16:38] hmm...blueman seems very overreaching, I'm tempted to just add a NotShowIn: GNOME; Unity to it [16:39] micahg, that will not fix that [16:39] micahg, it installs a buggy .so in nautilus-sendto's directory [16:39] which makes it segfault for all users who have blueman installed [16:39] micahg, I'm tempted to make a Conflicts blueman on nautilus-sendto :p [16:39] oh, wow, didn't notice that, yeah :) [16:40] * micahg checks the new Debian version to see if it still does that [16:41] the version in Debian does not, I"m quite confused, let me look into this [16:42] ah, there's a configure flag for it, I can just disable it [16:53] seb128: ok, I'll should be able to fix this later today, I think the answer will be just to not install those files (Debian doesn't seem to in any event) [16:53] micahg, ok, thanks [16:58] seb128: did you want to do the ubuntu-docs upload? rev. 130 has the latest fixes in it [17:00] jbicha, I can do [17:00] seb128, whoops, missed the meeting. Was there anything discussed? [17:00] mterry, hey, no, there was no agenda so no meeting [17:10] jbicha, you can change UNRELEASED to precise and debcommit -r push (I don't have commit rights to that vcs) [17:17] I hope the extra translated screenshots for ubuntu-docs don't cause a problem [17:43] good night everyone [17:53] good night everyone [17:57] pitti, a repititive night to you too :) [18:13] glatzor: hi, re your python-apt branch [18:41] servus mvo! [18:41] hey glatzor [19:30] mterry, hey [19:31] seb128, saw the crasher you assigned [19:31] mterry, heh, you are quick to pick emails up, I just wanted to ask if that you have time for it ... it's SRU material, no hurry, but feel free to bounce back if you have too much to do already [19:31] seb128, you're French, eh? Do you have any layout issues in precise? Bug 960096 seems to be rearing again for French users [19:31] Launchpad bug 960096 in libxklavier "Live session started with wrong layout" [Medium,Confirmed] https://launchpad.net/bugs/960096 [19:32] mterry, no, you must confuse me with somebody else, I'm german, did you mean didrocks? [19:34] mterry, (reading the bug) [19:34] seb128, wait, don't you live in France? [19:34] mterry, depends when in time [19:34] ;-) [19:34] * mterry didn't mean to open a can of worms. ;) [19:35] mterry, joke aside that bug is confusing and seems to mix several issues, liveCD and installed,upgraded system [19:35] seb128, yup! And it's been fixed several times [19:35] mterry, my user account is fucked keyboard wise, I keep it broken for debugging but I don't know if that's your issue [19:35] seb128, but I do get some odd behavior on latest live cds with french [19:35] mterry, for me it's stucked in french [19:36] like using the indicator or keybindings to change to a us or de layout doesn't work [19:36] but setxkbmap de or us works [19:36] the indicator never "apply" any change [19:36] seb128, interesting, that's not far off from what I'm seeing on livecd [19:36] like the label will change to "us" but they actual layout will still be fr [19:37] seb128, oh goodness, you could throw a stone from your house and hit Germany [19:37] dholbach had the same issue with "no" on her gf account it seems [19:37] mterry, ;-) [19:37] * mterry moves his mental model of seb128 from "Frenchy" to "Very German Frenchy" [19:37] damn you google, showing people where I live on your maps! ;-) [19:39] mterry, so back to this bug, I've this weird layout issue that I intend to try debug at some point, not sure it's the same bug that the one you pointed though [19:39] seb128, not sure either, but description sounds the same (label changes, but layout doesn't) [19:39] mterry, oh, and you won a second g-s-d power segfault bug, same as the other one, would be nice to fix in GNOME 3.4.2 and a SRU but feel free to ignore it [19:39] mterry, let me know if I can help debugging the layout stuff [19:40] seb128, but I'm having a hard time with the layouts because I don't know what they are supposed to behave exactly. :) [19:40] you can click "display layout map" in the indicator [19:40] seb128, yeah, but there are four characters per key. most of them look the same as english, but how do I get third and fourth character? [19:40] but french is basically doing a<->q z<->w [19:41] urg [19:41] in practice you don't need to go further than the second char, but I think the layout displayed in the chart has extra ones [19:43] mterry, so for non number keys you care usually about the 2 on the left and you get the second (bottom left) one using shift+key [19:43] seb128, is it common for caps lock to be a layout switcher for french users? [19:43] seb128, a/q is a good tell though, i see that now [19:44] mterry, aq zw and m being next to "l" are the most useful ones [19:44] seb128, alright, well, I'm going to dig into this layout stuff over the g-s-d ones because this will affect the CD [19:44] if you don't get into , . : [19:45] mterry, right, as said I don't think we would upload the gsd stuff before SRU anyway [19:46] mterry, if zou qre in the [19:46] ups [19:46] mterry, if you are in the same broken state that my user account you can "setxkbmap us" from it [19:46] to have an usable keyboard [19:46] then try to change to french [19:47] and see if "q" does print a "a" or a "q" [19:47] setxkbmap changes the layout but it's still broken [19:47] i.e it wont let you change back to fr [19:47] k [19:47] it might be easier to debug than to fight an azerty layout ;-) [19:48] mterry, it's very strange though, my issue seems to be config specific, like it happens only to my main account on that box and it happen always [19:48] seb128, in livecd, user gets stuck with us layout and can't change out of it [19:49] mterry, does "setxkbmap fr" works? [19:49] seb128, yes [19:49] ok, same as me [19:51] seb128, do you use autologin? [19:52] mterry, no, I use ecryptfs on that account [19:53] seb128, ok, this is the third distinct flaw in this same bug then. ah well [19:54] seb128, what is the output of: gsettings get org.gnome.libgnomekbd.keyboard layouts [19:54] ['fr\toss', 'de', 'gb'] [19:55] * mterry has a guess that the variant is confusing things [19:56] * mterry reboots [19:56] scratch that, middle of long download [19:59] What's a good version for a ppa package of gtk for precise with wayland enabled? 3.4.1-0ubuntu0~wayland1? [20:00] Darxus, you usually want a version newer than the archive one so people using your ppa get the update [20:01] Darxus, i.e 3.4.1-0ubuntu1.0~wayland1 or something like that [20:01] does anyone know where the default gsettings property for brasero plugins is initialized? [20:02] Darxus, dpkg --compare-version 2 gt 1; echo $? [20:02] ayan, hey, I doubt we have any brasero hacker here [20:03] seb128: Ah, I didn't realize 3.4.1 had been packaged, thanks. [20:03] Darxus, yw [20:03] And I just got the freaking thing building with the 3.4.0 packaging :P [20:13] seb128, if you want your system to be temporarily sane, remove French (Alternate) with just French [20:14] seb128, the variant seems to cause it. Affects unity-greeter and the session itself === zyga is now known as zyga-afk [20:29] mterry, thanks for the hint, I didn't realize my second user account fr was "fr" and not "fr oss" [20:52] dpkg-source: error: can't build with source format '3.0 (quilt)': no upstream tarball found at ../gtk+3.0_3.4.1-0ubuntu1.orig.tar.{bz2,gz,lzma,xz} [20:53] Why is it looking for that, instead of ../gtk+3.0_3.4.1.orig.tar.xz, which is there? [20:56] Darxus: you can't use - in the Debian part of the revision [20:56] s/revision/version [20:57] micahg: I don't understand how your statement relates to my problem. I'd like it to ignore the "-" and everything after it when looking for the orig tarball, sounds like you're saying that's what it should do. [20:58] Darxus: what's your version in the changelog? [20:58] 3.4.1-0ubuntu1-wayland1 [20:58] Darxus: right, that second - should be a + [20:58] Ohh, I should take out the second dash? [20:58] Thanks. [23:19] RAOF, you around? [23:20] RAOF, looking at some odd errors with X layout maps [23:20] RAOF, specifically, getting "key types not defined" when trying to have xkbfile extension use the "fr oss" layout [23:20] (or anyone here that knows X well)