=== Pharmasolin1 is now known as Pharmasolin === Pharmasolin1 is now known as Pharmasolin === Pharmasolin1 is now known as Pharmasolin === mhall119_ is now known as mhall119 [04:15] Good morning [04:26] Morning pitti. :) [04:32] hey TheMuso! [04:38] Laney: FYI, I sent an RT for bos01 which burst into flames [05:30] hello [05:34] good morning [05:43] Good morning didrocks [05:50] hey hikiko, how are you? [06:43] hey pitti, didrocks hikiko [06:44] good evening darkxst [06:48] didrocks, not really, spent the afternoon fighting with gdm3, but it is done now ;) [06:48] the migration stuff was a pita, but should work [06:50] darkxst: I guess you will be happy once this is really built and tested on a wider scale :) === fredp is now known as Guest98533 [06:54] didrocks, well I would be more happy if I could find a graceful way to upgrade gdm daemon without breaking the lock screen, but that seems to be a general issue, not related to the merge [06:57] hey pitti darkxst [06:58] did I missed pitti? I didn't see him saying hi here :p [06:58] ah, it was before I joined :) [06:58] darkxst: how are you breaking the lock screen? [06:58] didrocks, pitti is always up early! [06:59] hey hikiko and darkxst! [06:59] ça va didrocks ! [06:59] yeah, 6:30 is already early [06:59] didrocks, the auth channel between gdm and gnome-shell is quite fragile [06:59] I don't want to get up at 5:30 just to beat him :) [06:59] I suspect that is part of the reason fedora forces updates to delay until reboot [06:59] pitti: ça va, et toi ? :) [06:59] oui, je ne pouvais plus dormir, j'ai me léve à 4:30 à nouveau :( [06:59] didrocks: je vais bien, merci ! [06:59] pitti: argh :/ [07:00] darkxst: ah, you meant, breaking for the session until reboot? [07:00] darkxst: or then, for each gdm update? [07:00] didrocks, after a gdm update, you can't unlock it [07:00] and yea just until reboot [07:01] argh, but for each update? [07:01] that's quite ackward, indeed [07:12] didrocks, I don't know if it happens on 100% of updates, but its been happening atleast sometimes for a while now [07:13] and sometimes its gnome-shell loosing the auth channel, which is recoverable, but not by a normal user [07:20] pitti: waow, I didn't know that python gettext was assuming ascii for encoding [07:20] didrocks: ? [07:20] didrocks: neither did I, and it certainly shoudln't? [07:20] >>> gettext.gettext('ça va ♥') [07:21] 'ça va ♥' [07:21] https://docs.python.org/3/library/gettext.html [07:21] "If the charset encoding is specified, then all message ids and message strings read from the catalog are converted to Unicode using this encoding, else ASCII encoding is assumed. [07:21] or do you mean for matching strings in .mo files? [07:21] " [07:21] pitti: I have some _("") returning in a language that returns me some UnicodeEncodeError [07:21] the typical *** UnicodeEncodeError: 'ascii' codec can't encode character '\xe3' in position 5: ordinal not in range(128) [07:23] uh, so this would be lgettext() then or bind_textdomain_codeset()? [07:24] why on earth doesn't that default to $LC_CTYPE [07:24] pitti: yeah, seems that's what I need, I would like to understand exactly what's going on before adding it though [07:24] (I just bind_textdomain for now) [07:28] do you have a small reproducer for this? [07:30] $ python3 -c 'import gettext, locale; print(locale.setlocale(locale.LC_ALL, "")); gettext.textdomain("gtk30"); print(gettext.gettext("Add a class"))' [07:30] de_DE.UTF-8 [07:30] Eine Klasse hinzufügen [07:30] pitti: no, well… I can reproduce it in one command in Ubuntu Make [07:31] but yeah, I should do something wrong I guess… [07:31] if I run this under LC_CTYPE=C it indeed fails with UnicodeEncodeError, but that's to be expected [07:31] yeah [07:31] with LC_CTYPE=C.UTF-8 it works [07:31] can be as well the .mo file which is wrong for that lang [07:32] "Eu não aceito" is what it doesn't like in PT_br [07:32] .po files specify their own content encoding, but if it's wrong it might not match indeed [07:33] but "Content-Type: text/plain; charset=UTF-8\n" [07:33] so sounds good… [07:34] I initialize gettext in https://github.com/ubuntu/ubuntu-make/blob/master/umake/__init__.py#L32 [07:35] and the failure is when I print the "I don't accept" pt_BR translation set in https://github.com/ubuntu/ubuntu-make/blob/master/umake/__init__.py#L32 [07:35] didrocks: your setlocale() call is someplace else? [07:37] pitti: I never called setlocale [07:37] it's using LANGUAGE=, right? [07:37] didrocks: if set, yes [07:37] $ python3 -c 'import gettext; gettext.textdomain("gtk30"); print(gettext.gettext("Add a class"))' [07:37] Eine Klasse hinzufügen [07:37] indeed that works too [07:37] * didrocks reads http://www.wefearchange.org/2012/06/the-right-way-to-internationalize-your.html [07:38] I thought one had to call setlocale(); one must do in C, but perhaps python does it automagically [07:39] yeah, I never needed to in python contrary to C [07:39] (and the doc list the env variables that it's using) [07:40] $ python3 -c 'import locale; print(locale.getlocale())' [07:40] ('de_DE', 'UTF-8') [07:40] indeed [07:40] $ python -c 'import locale; print(locale.getlocale())' [07:40] (None, None) [07:40] hah [07:41] so that's new in py3 [07:41] $ python -c 'import gettext, locale; print(locale.getlocale()); gettext.textdomain("gtk30"); print(gettext.gettext("Add a class"))' [07:41] (None, None) [07:41] Eine Klasse hinzufügen [07:41] so that works only because py2 blissfully ignores all the encoding stuff [07:42] after reading barry's blog post, seems like I'm doing the right thing in python3… [07:42] as gettext is returning unicode there… [07:50] pitti: ok, I have a small reproducer, are you on xenial? [07:50] didrocks: yes; is there anything else? :-) [07:50] ahah :) [07:51] so apt install ubuntu-make [07:51] (to get the .mo file) [07:51] LANGUAGE=pt_BR python3 -c "import gettext; gettext.textdomain('ubuntu-make'); print(gettext.gettext('I don\'t accept'))" [07:52] so I'll install ubuntu-make and language-pack-{gnome-,}pt in a schroot [07:53] pitti: I don't think you need other langpacks, but yeah [07:54] hum [07:54] LANGUAGE=pt_BR python3 -c 'import sys; print (sys.stdout.encoding)' [07:54] ANSI_X3.4-1968 [07:54] $ LANGUAGE=pt_BR python3 -c 'import sys; print (sys.stdout.encoding)' [07:54] UTF-8 [07:54] on my desktop [07:54] in the schroot it says ANSI_X3.4-1968 (i. e. "ascii") [07:54] getting that on my clean shell [07:54] when changing only LANGUAGE [07:55] because my other LC_* are unset in the schroot [07:55] ahah [07:55] didrocks: what does your "locale" say? [07:55] LANG=pt_BR.UTF-8 LANGUAGE=pt_BR python3 -c 'import sys; print (sys.stdout.encoding)' [07:55] for LC_CTYPE in particular [07:55] pitti: ^ [07:55] LANGUAGE does not set any LC* or encoding, it's just for picking a language [07:55] (well, I just use env variables, so locale returns the french LC="fr_FR.UTF-8") [07:55] unlike LANG which is an entire locale [07:56] LC_ALL is empty though [07:56] that's fine [07:56] if you have $LANG it's the defualt for LC_* [07:56] most people have only $LANG or $LANG+$LANGUAGE [07:56] yeah, so the issue happens when I set LANG= in addition to LANGUAGE [07:56] LANG=pt_BR.UTF-8 LANGUAGE=pt_BR python3 -c 'import sys; print (sys.stdout.encoding)' [07:56] ANSI_X3.4-1968 [07:56] $ LANG=pt_BR.UTF-8 LANGUAGE=pt_BR python3 -c 'import sys; print (sys.stdout.encoding)' [07:56] UTF-8 [07:56] in my schroot [07:57] pitti: http://paste.ubuntu.com/13245981/ [07:57] didrocks: oh! may it be that you don't actually have the pt_BR.UTF-8 locale? [07:57] didrocks: locale -a ? [07:57] oh right, I don't have it installed [07:57] there you are :) [07:58] language-pack-{gnome-,}pt [07:58] so, you think that issue happens for people not installing the locale? [07:58] and so, sys.stdout.encoding returns ANSI_X3.4-1968 [07:58] instead of UTF-8 [07:58] (there is a default encoding by locale?) [07:59] right; "LANG=pt_BR.UTF-8 locale" should complain loudly [07:59] locale: Cannot set LC_CTYPE to default locale: No such file or directory [07:59] etc. [07:59] $ LANGUAGE=pt_BR LANG=pt_BR.UTF-8 python3 -c "import gettext; gettext.textdomain('ubuntu-make'); print(gettext.gettext('I don\'t accept'))" [07:59] Eu não aceito [07:59] didrocks: but that shouldn't be possible to do with the GUI -- you can only select locales if you installed the langpack [08:00] pitti: I still don't get why python doesn't take thus the encoding from LANG… [08:00] so perhaps people who manualy set it [08:00] didrocks: because it doesn't know which encoding that is [08:00] if the locale isn't installed [08:00] isn't what the .UTF-8 from LANG= is about? [08:00] the ".UTF8" is just part of a name [08:00] ah [08:00] an identifier for humans [08:01] e. g. aa_ET is also UTF-8 [08:01] ok, so a way to avoid crashing on those condition would be to force my own encoding? [08:01] as I know my .mo files are UTF-8 encoded [08:01] ah no, the issue is in the print() [08:02] correct [08:02] didrocks: I'd suggest calling locale.setlocale() [08:02] $ LANG=pt_BR.UTF-8 python3 -c 'import gettext, locale; print(locale.setlocale(locale.LC_ALL, ""))' [08:02] locale.Error: unsupported locale setting [08:02] and only if that succeeds you set a text domain [08:03] that might be the most elegant way [08:03] if the env specifies an invalid/broken/missing locale, it just uses C [08:03] yeah, so fallbacking to C… [08:03] so: [08:03] try: [08:04] locale.setlocale(locale.LC_ALL, '') [08:04] gettext.textdomain(...) [08:04] except locale.Error: [08:04] # maybe print some warning [08:04] pass [08:04] yeah [08:04] yay for my perfect indentation this morning! [08:04] ahah good enough :) [08:04] let me still confirm that's the issue on the bug report [08:05] but I'll add that anyway, good way of protecting myself [08:05] thanks a lot for looking at it pitti! [08:05] * didrocks hugs you [08:05] didrocks: je t'en prie ! [08:05] * didrocks is quite happy that wasn't so straightforward :) [08:39] Trevinho, unity has compile errors... [08:40] bamf_matcher_get_application_for_xid_matcher [08:44] Trevinho, first error: /home/eleni/canonical/staging/unity/unity-shared/BamfApplicationManager.cpp: In member function ‘void unity::bamf::Application::UpdateWindows()’: [08:44] /home/eleni/canonical/staging/unity/unity-shared/BamfApplicationManager.cpp:427:43: error: ‘bamf_view_’ is not a member of ‘unity::bamf’ [08:44] for (GList* l = bamf_view_peek_children(bamf::bamf_view_); l; l = l->next) :s [08:45] sorry [08:45] wrong paste [08:45] /home/eleni/canonical/staging/unity/unity-shared/BamfApplicationManager.cpp: In member function ‘void unity::bamf::Application::UpdateWindows()’: [08:45] /home/eleni/canonical/staging/unity/unity-shared/BamfApplicationManager.cpp:427:53: error: ‘bamf_view_peek_children’ was not declared in this scope [08:45] for (GList* l = bamf_view_peek_children(bamf_view_); l; l = l->next) [08:45] that's the 1st error [08:46] mmm now I look at it I should use a pastebin :p [08:56] morning all [08:57] morning willcooke [09:01] hey hey [09:02] pitti: bos01> oh dear [09:02] hey willcooke, Laney [09:02] hi darkxst [09:03] oh man, still no phonon fix! [09:06] hey Laney! :) [09:06] Laney is up before seb128, this guy is really back on his French schedule after a week :p [09:06] hey Laney [09:07] Laney: pitti, the breaker of all clouds.. [09:08] qengho, For the last couple of days Cr. has added another icon to the launcher when running and the original just launcher yet another Cr. [09:08] *launches [09:12] howdy didrocks [09:15] good good, had some nice interesting python encodeissue this morning :) [09:16] and fixed bad translations in pt_BR, making Make failure [09:16] failing* [09:16] + my first Atom.io contribution \o/ [09:16] and it's Friday :p [09:16] congrats didrocks [09:17] thx ;) [09:21] Ouch, irc didn't connect... But morning [09:21] qengho, but removing the old icon and readding sorts it [09:21] hey Trevinho [09:21] hi willcooke [09:22] Trevinho, sounds like the can find us some real hardware for U7 testing on Intel, nVidida, AMD [09:22] \o/ [09:22] willcooke: this is cool! [09:22] re desktopers [09:22] willcooke: even just intel would be fine for me... But still [09:22] :) [09:22] the more the better [09:25] https://bugs.launchpad.net/ubuntu/+source/gtk+3.0/+bug/1512290/comments/8 [09:25] Ubuntu bug 1512290 in gtk+3.0 (Ubuntu Wily) "gnome-screenshot produces very low quality blurry screenshots" [High,Fix committed] [09:25] this guy is CORRECT! [09:25] FAIL [09:25] laarrsssuuu [09:27] do we tag bug #1515810 as a gtk 318 issue? [09:27] bug 1515810 in gedit (Ubuntu) "Doesn't merge into unity panel when maximized" [High,Confirmed] https://launchpad.net/bugs/1515810 [09:27] it's not really [09:27] but I think we said we need to teach gtk to have a flag for that? [09:28] that or rls [09:28] should triage the -incoming bugs btw [09:28] does anyone plan to do that? [09:29] Seems like something that needs more that just one person to go through it [09:30] so, hangout on Monday? [09:30] Laney: meh, they fixed bos01 this morning, but seems it's just dying again [09:30] pitti: oh I just assumed it wasn't fixed yet [09:30] well, we need one person to approve nominations [09:30] seems a job for a manager ;-) [09:31] willcooke: are you in the right team to do bug nominations? [09:31] but I'm happy to help triaging the list [09:31] how do I check Laney? [09:31] Laney: at least it held up for some two hours which reduced the queue quite a bit [09:33] willcooke, you try to nominate a bug [09:33] willcooke: https://bugs.launchpad.net/ubuntu/+source/unity-greeter/+bug/1286878 [09:33] do you have "Nominate for series"? [09:33] "Target to series" [09:33] https://bugs.launchpad.net/ubuntu/+source/unity-greeter/+bug/1286878/+nominate [09:33] do you have access to ^ [09:33] Ubuntu bug 1286878 in unity-greeter (Ubuntu) "Doesn't scale well on a hidpi display" [Medium,Triaged] [09:34] * seb128 bets he doesn't [09:34] well that's what you need to triage the list [09:34] I can nominate [09:34] right, I'm not saying he shouldn't [09:34] you can? good ! [09:35] nominate but not approve [09:35] well, can you approve nominations ? or just suggest those ? [09:35] or target but not nominate [09:35] whatever the stupid terminlogy is [09:36] Laney, bug #1512290 should be verification-failed if there is a regression, no? [09:36] bug 1512290 in gtk+3.0 (Ubuntu Wily) "gnome-screenshot produces very low quality blurry screenshots" [High,Fix committed] https://launchpad.net/bugs/1512290 [09:36] probably if it makes the thing crash [09:36] seb128, Laney - Don't think I can approve [09:36] willcooke: are you in bug control? [09:36] can't remember if that lets you actually approve them [09:36] no it doesn't [09:36] what does then? [09:36] you need to be in ubuntu-drivers I think [09:37] https://launchpad.net/~ubuntu-drivers [09:37] or https://launchpad.net/~ubuntu-release-nominators [09:37] that's part of the drivers [09:37] qengho, but now "open link in browser" from xchat opens a new Cr. window and doesn't load the page [09:38] infinity is admin of that group [09:38] willcooke, maybe ask to be added there? [09:38] ~ubuntu-release-nominators that is [09:38] seb128, will do. thx [09:39] * Laney checked with #launchpad, but probably [09:40] oki, so Laney seb128 - you guys (and others) will nominate and then I will approve them? [09:40] +1 [09:40] Guess I can nominate too right? [09:40] yes [09:40] cool [09:41] I usually do approve things as well [09:41] great [09:41] I think I'm going to keep doing that for obvious things [09:41] but let me know if you prefer be doing those [09:41] wait what? [09:42] oki, I will take a pass of the list on Monday, or once I am a member, which ever is later and then we can sync up afterwards [09:42] I can't only nominate, it approves them right away [09:45] yeah, if you have rights (that every ubuntu core devs do), it will approve right away [09:45] didrocks, I don't think it's coredevs [09:46] it's ubuntu-drivers afaik [09:46] seb128: well, I can approve for sure, and I'm not in any of the 2 teams [09:46] nah it's uploading permissions + drivers [09:46] I think [09:46] yeah [09:46] didrocks, you are not in https://launchpad.net/~ubuntu-drivers ? [09:46] nope [09:46] I'm in through UDS organiser [09:46] * Laney not [09:46] k [09:46] that is some magic team [09:46] launchpad permissions are weird [09:46] yeah [09:46] and ~ubuntu-release-nominators is misleading then [09:47] it's an extra group on top of the uploaders [09:47] for this kind of case [09:47] funny that some people are in that team where they are less involved in ubuntu than uploaders btw :p [09:48] yeah, that can makes sense though [09:48] people defining quality and goals and not all doing technical work [09:48] random canonical upstreams? [09:48] https://launchpad.net/~ubuntu-release-nominators/+members#active [09:48] like Pat is driving the touch product and dealing with bugs and targets but he's not packaging [09:49] that list is really weird… [09:49] yeah, agreed on that [09:49] but like having QA might make sense [09:49] if we trust them to define the quality goals [09:49] random individuals less so [09:49] for those yeah, not telling everyone in that list doesn't make sense [09:49] but a good part of individuals is weird [09:50] that "Will" "Cooke" [09:50] who even is he? [09:51] exactly! [09:51] but we don't even know if that team is actually used [09:51] since we are not member of it and have access to the nominations [09:51] so it is a least not a direct mapping [10:02] https://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/view/head:/lib/lp/bugs/model/bugnomination.py#L120 [10:02] uploader + driver [10:02] k [10:02] thanks Laney! [10:03] np! [10:03] seb128: what do you think about demoting phonon-backend-vlc auralquiz? [10:04] does it mean mitya57 and slangasek failed at getting at the bottom for the phonon issues? [10:04] * seb128 reads IRC backlog from the night [10:04] some patches but he didn't upload... [10:25] * Laney tries them [10:25] them? the patches? [10:26] ye [10:26] I was about to do the demotion [10:26] should I still hit enter? ;-) [10:26] we can fix up things later on [10:26] I assumed silence was no :P [10:26] do it if you're happy [10:27] Laney, seb128 - is this the list I should work from? http://reqorts.qa.ubuntu.com/reports/rls-mgr/rls-x-incoming-bug-tasks.html [10:27] yep [10:27] thx [10:27] +1 [10:27] * willcooke blocks out some hours [10:27] probably wait until you're in the team so you don't have to do get us to pres butan [10:27] Laney, sorry I was in another discussions so delayed the demotions a bit [10:27] willcooke: rls-x-notfixing is the "we don't commit to this" tag [10:28] got it [10:28] ta [10:28] so either way they get out of the incoming list [10:28] Laney, ok, demoted [10:28] oh, good news... seems my details weren't amongst those lifted in the TalkTalk hack [10:28] eventually we are going to hit the other side [10:29] Rejected: [10:29] phonon-backend-vlc 0.8.2-1ubuntu1 in xenial (version older than the phonon-backend-vlc 0.8.2-1ubuntu2 in xenial published in xenial) [10:29] shrug [10:29] I don't understand the demote tool [10:29] why does it try to upload non current versions? [10:30] seb128: new package in proposed, just remove from release [10:30] done [10:30] nice [10:30] you (maybe) got to do the last action [10:31] * seb128 takes a bigger hammer in case it's needed [10:31] we have the "force" hint ;-) [10:31] transition weeks are so much fun [10:31] that gets you sacked from the release team though [10:31] haha [10:32] would be nice to get that done today [10:32] indeed [10:32] ah good job gtk is blocked [10:33] seems .3 broke some stuff [10:33] * Laney looks at .4 [10:33] yeah, cf my comment yesterday evening just after you left :p [10:33] binding booog [10:36] looks like it all happened at the same time [10:36] like the upstream commits were after I made the package and -3 was just around the same time I uploaded it [10:36] doh doh doh dohd ohdodhodh [10:39] Laney: happy friday eh :) [10:39] hey czajkowski! how are you? [10:48] Sweet5hark, hey, could you have a look to bug #1515761 ? [10:48] bug 1515761 in libreoffice (Ubuntu) "Libre office writer reports error when opening MS Word files after running recent trusty update." [Medium,New] https://launchpad.net/bugs/1515761 [10:48] just to check if it's a regression or just a local/one user thing? [10:50] seb128: /me reads [10:50] thanks [10:50] Laney: is bug #1512290 the one you mentioned in the tg group? [10:50] bug 1512290 in gtk+3.0 (Ubuntu Wily) "gnome-screenshot produces very low quality blurry screenshots" [High,Triaged] https://launchpad.net/bugs/1512290 [10:50] larsu, hey again ;-) and yes [10:50] larsu: yeah, I assigned it to you [10:50] sorry :( [10:50] and hey! [10:50] good gym? [10:50] how was the gym today ? [10:53] exhausting! [10:53] * larsu feels very good now :) [10:56] sweet [11:03] pitti: ugh, that MR is the worst [11:03] larsu: the unbreak-my-desktop brightness thing? [11:03] yeah -- this is a kernel/driver issue, not a "let's bikeshed with user settings between off and low" [11:04] ya. Thanks for taking a strong stance against it [11:04] userspace should use the brightness range that the driver offers [11:04] * larsu will pile on [11:04] pitti: clearly [11:06] pitti: also, that MR doesn't add the key to gsettings?! [11:06] or is that in deskto-schemas or similar? [11:06] larsu: maybe, I didn't bother to read that far [11:07] this is conceptually wrong, I see little point in discussing the implementation [11:07]