=== eeejay is now known as eeejay_is_afk [05:43] Good morning [05:43] desrt: exactly [05:43] Hey there pitti! [05:51] hey RAOF, how are you? [05:53] I'm pretty good. [05:54] Figured out why my first test started crashing once I added a second :) [06:01] RAOF: ah, working on nux test suite? [06:01] Unity, yeah. [07:14] good morning [07:22] bonjour didrocks [07:31] guten morgen pitti, how are you? [07:31] quite fine! [07:31] just learned about the admin -> sudo group migration in the installer, now trying to catch up [07:31] how about you? [07:32] pitti: still some sore throat, I'm eager that to finish :) [07:32] otherwise, all is fine, need to prepare to go to the ubuntu party in Toulouse for this week-end [07:33] (6th "ubuntu weekend" in a row if we count UDS) [07:33] no wonder you get no rest.. [07:33] * pitti hugs didrocks, enjoy! [07:33] * didrocks hugs pitti back, thanks [07:33] but I will surely enjoy a quiet week-end next week :) [07:35] so, you are upadting the polkit rules to have the "sudo" group as well? [07:35] didrocks: yes, that, polkit-desktop-privileges, workign on accountsservice now [07:35] to recongnize both and set "sudo" for admins [07:35] hoping that's the only 2 offended places, good luck :-) [07:36] KDE's control center likely has that as well [07:36] and I'm sure that there are a number of other places which check for "admin" group only [07:37] quite a surprise this morning.. [07:37] more than possible, indeed [07:47] hello [07:58] chrisccoulson: oh, mozjs wants to go to universe; you are happy about this, I take it? [07:59] didrocks: zeitgeist-extensions wants to go to universe now; is that obsolete and should be removed, or will it be updated to current zeitgeist? [08:00] pitti: you mean, zeitgeist-extension-fts or just the source package, zg-extensions? [08:00] didrocks: both [08:00] related to that, I now have a process "python /usr/share/zeitgeist/fts-python/fts.py"; should I have? [08:00] pitti: zg-fts-extensions should still be there [08:00] I was hoping we could get rid of python there [08:01] pitti: yeah, the extensions are still in python [08:01] so it seems nothign pulls in zg-e-fts any more [08:01] let me see if zeitgeist-fts-extension should be directly seeded or be a recommends of anything [08:02] pitti: I would go for seeding it directly, as debian dropped the recommends [08:04] hum, u-l-a should dep on zg-extension-fts [08:04] pitti: ^ [08:06] weird, why is it in component mismatch? [08:07] http://packages.ubuntu.com/precise/unity-lens-applications [08:07] and u-l-a still in main [08:07] hm, weird indeed [08:07] Provides: zeitgeist-extension-fts [08:07] zg-core [08:08] hum, I need to speak to rainct [08:08] ah, it prefers the already installed package then [08:08] it doens't provide it last time I talked to mhr3 [08:08] indeed [08:08] but that seems wrong then, if core zg doesn't yet provide full text search [08:08] I know it was planned, but it wasn't the case last time I checked [08:08] pitti: I'll check and fix one way or another today, thanks! [08:10] /usr/share/zeitgeist/fts-python, seems it made it, nice [08:10] didrocks: merci [08:10] will check with upstream, if this is the case, we can even remove the other package [08:11] pitti: de rien, merci de m'avoir prévenu ;) [08:23] didrocks: I hope the actual meaning of that is a lot less strong than google translate indicates :) [08:23] bbl [08:24] pitti: hum, it's basically, "thanks for notifying me about it" :) [08:24] ah :) [08:24] google says "thank you for defendant", as if I was sueing you :) [08:26] ahah :) [08:26] ah, prévenu [08:27] yeah, prévenu == prévenir (to notify, to warn about) [08:27] but a "prévenu" is someone you are sueing, right :) [08:46] morning [08:48] hey rodrigo_ [08:48] hi didrocks [08:48] pitti: I'm opening a bug to remove fts extensions, after some investigation, yeah, it's the zg-core version which is the latest and greatest [08:49] hum, or maybe just demote, as it's still in debian [08:49] and potentially, other extensions can come back [09:05] didrocks: either is fine with me [09:05] I went the demotion way :) [09:07] Sweetshark: on http://people.canonical.com/~ubuntu-archive/component-mismatches.txt there is a bunch of java OO.o related libs which want to go to universe; is that ok? I didn't see a LibO upload recently, wonder what changed [09:48] good morning everyone [09:49] hey chrisccoulson [09:50] hi pitti, how are you? [09:50] I'm fine, thanks! how about you? flu gone for good? [09:51] it's almost gone now :) [09:53] hey chrisccoulson! [09:53] hi didrocks [10:06] hello mvo, could please have a look at the two merge proposals for aptdaemon: prettyrepos and cached-changelogs? [10:13] mvo, just upgraded my branch, with a test program and several fixes [10:13] and fixing the systemd comments, as per seb's comments [10:46] glatzor: will do [10:50] mvo, thanks [11:00] glatzor: commented inline, awsome work, but two tiny remarks === huats_ is now known as huats [11:12] pitti: hmm, strange. dunno either. [11:12] Sweetshark: do any of these look like you'd miss them? [11:12] well, we can always re-promote them of course [11:13] pitti: they are transitional only anyway, right? [11:15] Sweetshark: not e. g. libformula-java-openoffice.org [11:15] * didrocks knows show the flag for reaching 101 manual test cases for unity & compiz [11:21] didrocks: wow! was that the goal? [11:22] didrocks: OOI, how long do you estimate will it take to run through that once as a tester? [11:22] * ogra_ thoguht they were automated tests running at build time [11:34] pitti: there is no goal, but I think I covered most of launcher/panel/dash/alt-tab and window functionnality [11:34] didrocks: you rock! [11:34] * didrocks hugs pitti [11:35] pitti: I think it will take approx. 3 hours [11:35] and 1h30 once you know the tests well, most of them are short [11:35] ogra_: the manual tests are the distros one [11:35] ah [11:35] ogra_: for accepting to release a new version === ronoc_ is now known as ronoc [12:15] rodrigo_: Hi Rodrigo, wondering if you noticed https://bugs.freedesktop.org/42857 [12:16] GunnarHj, yes, pitti told me about it [12:16] not about the bug, but about you working on it :) [12:17] rodrigo_: Ok. With that bug I try to make them apply the code upstream, which would benefit GNOME just as much, I think. [12:19] GunnarHj, although not sure about it, we have the SetLocale interface in systemd, which this branch implements -> https://code.launchpad.net/~rodrigo-moya/system-service/new-interfaces/+merge/83308 [12:19] that's what GNOME upstream uses to set the system locale settings [12:19] rodrigo_: Do you possibly know how to best get in contact with Ray or Matthias? [12:19] it saves all the LANG and LC_* vars in /etc/default/locale [12:20] GunnarHj, matthias is in #control-center in gimpnet, most of the time (he's in USA) [12:20] Aha, thanks. [12:20] and he's usually here -> mclasen afaik [12:21] rodrigo_: ah, so that's what we should use now, instead of adding /etc/default/locale writing to accountsservice? [12:21] pitti, I think so === MacSlow is now known as MacSlow|lunch [12:21] GunnarHj, what apps would use that interface in accountsservice? [12:23] so far that's language-selector [12:23] g-c-c, I suppose. We are not only talking about the system wide (/etc/default/locale), but the per-user settings. [12:23] but if that goes away, and g-c-c uses systemd, it'll be obsolete [12:23] g-c-c already uses the systemd interface for setting /etc/default/locale [12:23] so yes, I don't think we need it [12:23] GunnarHj: oh, right; the system-wide thing is l-s' own dbus backend, ignore me [12:23] rodrigo_: ^ [12:23] so we only need it for per-user, yes [12:24] it feels a bit weird to split that up between "per user" and "system" between accountsservice and systemd, but *shrug* [12:24] per-user? we already do that in the region panel [12:24] rodrigo_: trough accountsservice, I guess? [12:25] rodrigo_: But that involves g-s-d, if I remember it correctly. [12:25] pitti, yes [12:26] rodrigo_: per-user region settings through accountsservice? How? [12:27] GunnarHj, when you unlock the permissions in user accounts panel, you can set each user's language, and that's via accountsservice's SetLanguage [12:28] rodrigo_: language, yes, but that's something else. I'm talking about regional formats now. [12:29] hmm right, I got confused, this new method in accountsservice is for the specific user interface, not for system-wide, right? [12:30] rodrigo_: yes [12:30] ok then yes, I guess it makes sense [12:30] sorry, I was confused :) [12:31] rodrigo_: no problem. We all get confused once in a while when discussing locales and such. :) [12:31] :) [12:32] so, formats_locale stores the LC_ALL variable? [12:32] * rodrigo_ looks at the patch deeper [12:32] rodrigo_: Not LC_ALL, but LC_TIME, LC_CURRENCY etc. [12:33] But that you don't see in the patch. [12:33] ok, then I guess it would be better as a "as" in the dbus interface [12:33] that is, an array of strings? [12:34] Don't think so. It's just one value, that is used for a bunch of env. variables. [12:34] or even better, we could have SetLanguage renamed to SetLocale and make accept an array of strings, hence no need for 2 method calls from the region panel [12:35] but yes, let's discuss with mclasen later, ping me when you ping him also, please [12:35] will do. [12:35] systemd does that, SetLocale (as), so that it gets all the vars [12:36] we could get it to accept just a string, but I think it makes sense to do the SetLanguage->SetLocale renaming [12:38] Hmm... haven't thought about that. For 'pedagogical' reasons I wonder if it isn't better to have two properties, after all. People tend to mix up language and regional formats all the time... [12:38] Need to stop now. See you later. [12:39] ok, later [12:42] rodrigo_: FWIW, I agree that keeping SetLanguage() and adding SetRegion() is better [12:42] rodrigo_: it's less confusing, how to map those to LC_* is an implementation detail, and it keeps API compat of SetLanguage() [12:43] rodrigo_: how does 'as' work? the individual strings look like 'LC_FOO=bar'? i. e. they include the var? [12:43] pitti, yes [12:43] or do they have to have a particular order? (that would be sick) [12:43] no, no order [13:25] * rodrigo_ lunch === MacSlow|lunch is now known as MacSlow [14:28] taking a break, back for release meeting [15:17] Riddell: hey, not sure you saw my email about the fact that I fixed pyromaths :) [15:19] didrocks: no, where was that? [15:19] oh, it's in my inbox :) [15:21] accepted [15:23] Riddell: thanks :) [15:35] ok, will take the train now, I'll probably enjoy the 4 hours of train to Toulouse to add test cases to oneconf :) [15:35] see you on monday everyone! === tkamppeter_ is now known as tkamppeter [17:13] Sweetshark: would it be possible to do a rebuild of LibO in precise against teh current libraries? [17:13] sanbar: or is that an inordinate amount of work? [17:13] sorry sanbar, I meant Sweetshark [17:16] pitti; I thinkn sanbar likes the attention ;-) [18:36] * bryceh waves [18:52] bryceh: hello === eeejay_is_afk is now known as eeejay [19:41] pitti: what exactly do you mean by rebuild? just retriggering the build, or a new upload with new version number but without changing the package itself? both wouldnt be too much of a hassle ... [19:47] Sweetshark, i would suspect a no change upload [19:49] kenvandine: thats easy. it only gets tricky with changes which is like playing mikado ;) [19:49] indeed [19:50] kenvandine: luckily everthing will be better with 3.5 [19:50] * Sweetshark still uses his illusion as Guns'n Roses once told him to ... [19:51] haha [20:36] Sweetshark: I think he was asking more about whether or not you think it'll build [20:36] dch -R is the easy part :) === eeejay is now known as eeejay_is_afk === eeejay_is_afk is now known as eeejay