=== maclin_ is now known as maclin [08:06] hey hey, happy friday! [08:07] hey Laney [08:08] hey mvo, how's it going? [08:08] happy friday desktopers! [08:08] Laney: not too bad, thanks! [08:08] hey Laney, mvo [08:14] hey seb128 === maclin__ is now known as maclin === alan_g is now known as alan_g|errands [10:31] seb128: ping? [10:31] GunnarHj, hey [10:32] seb128: Hello! [10:32] seb128: Why do we bother to put PO files into the language-selector package? [10:34] seb128: I think they are installed in /usr/share/locale, but is anybody using l-s who are not using the langpack system? [10:36] seb128, I tested the patch of https://bugs.freedesktop.org/show_bug.cgi?id=76450 now and it is as you have found out, the drawings appear now but not correctly rotated. Should I attach the debdiff to the LP bug 1242678 so that someone with upload rights on libspectre can upload the fix? [10:36] Freedesktop bug 76450 in general "Documents are not rotated correctly" [Normal,New] [10:36] Launchpad bug 1242678 in libspectre (Ubuntu Trusty) "evince cannot render some EPS files" [High,Triaged] https://launchpad.net/bugs/1242678 [10:36] GunnarHj, yes, people who boot the liveCD in a locale which doesn't have its langpacks on the iso [10:36] GunnarHj, it's the same reason why ubiquity also has translations non stripped [10:37] GunnarHj, so anyone can use those UI to install the system/their language, without needing the langpacks to understand the UI [10:37] tkamppeter, can you look at what is wrong/update the patch to have the correct rotation? [10:38] tkamppeter, once you have a fix, yes please add a debdiff and subscribe ubuntu-sponsors [10:38] seb128, I am not so much expert in PostScript, I could ask the developers of Ghostscript. [10:39] tkamppeter, if you know anyone that you could ask for hints that would be nice [10:40] seb128: Trying to understand. ubiquity is obvious, but is the l-s ui really involved in the installation process? [10:41] GunnarHj, well, if you install without internet, the dialog that opens in the session after first boot to let you complete the locales installation is from l-s no? [10:42] GunnarHj: so ubiquity generates a file with "incomplete translations message" and that file is ~ .desktop file syntax with all translations for all languages we know how to say "You need language packs" [10:42] GunnarHj: and l-s is triggered if that file is on disk and shows translated messages from there. [10:43] GunnarHj: ideally l-s would simply be unstripped from translations, similar to ubiquity & oem-config [10:43] oh, wow ... moving my mouse pointer over the firefox search bar gets me quite a gigantic tooltip [10:44] ogra_: high-dpi ? =) [10:44] heh, only fullHD here [10:44] its funny, i have two windows open ... one gets me that huge black square, the other works as expected [10:45] xnox: Right, because now stripping seems to still happen, since there are mo files in both ../locale and ../locale-langpack [10:45] xnox, l-s is unstripped of translations [10:45] GunnarHj: that sounds ok. [10:45] seb128: Are you sure? [10:46] GunnarHj: we still want to be able to update l-s translations via langpack.... ( i know a bit recursive ) [10:46] GunnarHj: also are you sure those are not magic symlinks or some such? [10:47] xnox: They are real files, not symlinks. [10:47] http://people.canonical.com/~ogra/firefox.png [10:48] GunnarHj, yes, it's in the list of things to not strip of pkgbinarymangler [10:48] GunnarHj, http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/utopic/pkgbinarymangler/utopic/view/head:/striptranslations.blacklist [10:48] xnox, you can't have non stripped and updated by langpacks, you need to pick one [10:49] seb128: Ok.. Why do language-selector.mo still end up in /usr/share/locale-langpack? [10:49] GunnarHj, it doesn't for me [10:50] seb128: ? [10:50] GunnarHj, oh, it does [10:50] dunno then [10:50] but they are not stripped [10:50] so the langpack one is not used [10:50] dpm or pitti might know [10:51] seb128, xnox: Ok, I see no real problem, just thought there was room for simplifying it a bit. Might ask pitti. [10:51] * dpm reads backlog [10:52] dpm: It was me who asked why the l-s translations are both in /usr/share/locale and /usr/share/locale-langpack [10:54] oh, I see [10:54] dpm: Not a big problem, really. Just me who is in a cleaning mood. [10:54] GunnarHj, xnox, I'm not sure if that has the desired effect, as IIRC, /usr/share/locale translations have precedence over those in /usr/share/locale-langpack [10:55] which effectively means we're shipping/updating langpack translations for that package that are never going to be used [10:55] seb128, GunnarHj: language-selector is in pkgbinarymangler's blacklist [10:55] that was requested so that language-selector can be shown in other languages for not-yet-installed langpacks [10:55] and yes, that precludes putting its translations into langpacks :) [10:56] pitti: They still end up in /usr/share/locale-langpack [10:56] pitti, hey [10:56] dpm: hm. [10:56] pitti, yeah, the question is whether it should be in langpacks still if it's unstripped [10:58] I can remove the langpack flag in Launchpad for l-s, so that its translations are not exported in the langpacks tarball [11:00] dpm: Aha, so it's a Launchpad setting? [11:01] dpm, pitti: $ dpkg -S /usr/share/locale-langpack/sv/LC_MESSAGES/language-selector.mo [11:01] language-pack-sv-base: /usr/share/locale-langpack/sv/LC_MESSAGES/language-selector.mo [11:01] GunnarHj, in LP you can set which .mo files will be put in the huge tarball that is fed to langpack-o-matic [11:02] dpm: I see. [11:03] dpm: Sounds like that would be the right thing to do, then. But e.g. pitti should confirm first. [11:03] GunnarHj: yes, blacklisting now just means that it doesn't get stripped [11:04] but a tarball is still built, i. e. fed to LP [11:04] so translations from langpacks take precedence, but there's a fallback to the "builtin" translations [11:04] this should give the best of both worlds [11:04] (for the price of having redundant files) [11:04] pitti, are you sure langpacks take precedence? [11:04] I remember having filed a bug about this a while ago with that request [11:05] but the conclusion was that package translations should take precedence [11:05] dpm: well, that was the intent some years ago when we wrote the libc patch; I haven't checked recently [11:05] let me see if I can find the bug... [11:06] pitti: So I take it that it's fine as it is, then, despite of the duplication. [11:06] unless libc behaves the wrong way now, of course [11:06] pitti, GunnarHj: bug 530872 [11:06] Launchpad bug 530872 in Ubuntu Translations "/usr/share/locale translations have precedence over /usr/share/locale-langpack" [Undecided,Invalid] https://launchpad.net/bugs/530872 [11:07] so /usr/share/locale translations are preferred, unfortunately === alan_g|errands is now known as alan_g [11:10] dpm: oh, I remember -- was that so that you can install newer versions from a PPA or other 3rd party sources and its newer translations would work [11:11] yeah, not sure we've made much use of that, though, except perhaps for some OEM packages [11:12] pitti, dpm: isn't what's preferred determined by bindtextdomain in respective app? [11:12] pitti, dpm: And in the case of l-s, bindtextdomain is set to /usr/share/locale. [11:13] yes, but libc always has the /usr/sharea/locale-langpack fallback [11:13] pitti: Aha. [11:22] pitti, hi :), i hope you had some happy easter! [11:22] hey ricotz; I did, thanks! and you? [11:22] pitti, of course i noticed you systemd ppa too ;) [11:23] pitti, had some nice days too :) [11:23] ricotz: hehe -- soon to be landed in utopic after we unfreeze [11:23] pitti, ah that i was about to ask [11:23] pitti, are you going to follow the stable systemd branch? [11:23] ricotz: for now I'd like to follow Debian, i. e. 204 [11:24] which would be mean to get it updated to 208 [11:24] ah i see [11:24] ricotz: but I'm currently en route to Antwerp for the systemd/gnome sprint [11:24] after that I'll know more :) [11:24] mutter wayland already kind of requires 209 [11:24] ricotz: yes, I'd prefer updating to the current upstream version [11:24] pitti, ok, thanks for the into :) [11:25] info [11:25] but at the moment, Debian's 204 is fairly well tested, so I wanted to jump on a known-good version first (and minimize our delta); I hope at the sprint I can push some more changes to Debian [11:25] I thought we weren't going to be able to take >> 204 without switching init too [11:26] Laney: we are if we teach systemd-shim about cgmanager [11:26] that's the plan so far [11:26] unless we switch to systemd init earlier than we thought :) [11:29] Hrm, interesting [11:29] I thought it was about internal (non-shimmable; non-D-Bus) interfaces === alan_g is now known as alan_g|lunch [12:39] pitti: do you mind if I upload a cherry-pick to precise for https://bugs.launchpad.net/ubuntu/+source/apport/+bug/1007826 ? [12:39] Launchpad bug 1007826 in apport (Ubuntu Precise) "crash with AssertionError: file stream must be in binary mode when trying to save report to file" [Undecided,New] === alan_g|lunch is now known as alan_g === mjohnson151 is now known as mjohnson15 [13:07] * mvo is away for some minutes [13:09] mvo, it's IRC, you can go make some tea without telling us ;-) [13:09] seb128: ;) [13:10] seb128, I think he's using IRC has a timer, looks at the timestamp of the message and he knows how long it has been steeping. [13:14] lol [13:17] tedg, knowing mvo he probably rather use it to know when he needs to get the tea bags out of the kettle === alan_g is now known as alan_g|errand === alan_g|errand is now known as alan_g [13:27] but what about the temperature? There is no IC plugin for it [13:27] seb128: seems he can't do tea and IRC as he left :) === alan_g is now known as alan_g|errand === alan_g|errand is now known as alan_g [14:27] pitti: (uploaded it, feel free to reject if you object [ha]) [14:34] * Laney wonders about SRUing webkit [14:34] Laney, point release? do it [14:35] http://webkitgtk.org/2014/04/14/webkitgtk2.4.1-released.html [14:35] I think we should follow the series we are one, especially for the LTS [14:35] webkit is sort of GNOME-like release wise, they are in stable/bugfix mode [14:35] it does make sense I think [14:38] hey mterry! [14:38] mterry: my backups keeps failing :/ (even when I restart from scratch) [14:42] didrocks, oh no! [14:42] didrocks, what's the error? [14:43] mterry: ah, wait, seems different this time, I have just some files he couldn't save [14:43] I would think it's normal, let me retry to do another backup, but incremental this time [14:45] Laney: no, I don't mind; although it shouldn't really break anything, just for cases where a package already fails to install [14:46] Laney: was offline in Belgian trains :) [14:46] happy sprinting! [14:52] Laney: thanks! [14:52] * Laney is going to try systemd in a minute [14:56] Laney, can we use testing/is that something we try to already push forward? or are you just doing by personal interest? [14:57] or asked differently, should I go through the effort of trying it? [14:58] I'm happy to help testing if needed, but otherwise I'm no special interested trading my working init system by another one ;-) [14:58] Well, the switch is going to happen and I'm interested in making it go smoothly if I can help [14:58] and pitti's laying the groundwork for that already [14:58] mterry: ok, incremental backup passes since I ripped the folder [14:58] didrocks, hrm [14:58] mterry: so wrong alarm, sorry :) [14:59] didrocks, I guess... Probably still something to fix there, but whatever, problem gone [14:59] seb128: it's easy to switch between the two currently - did you see the blog post? [14:59] mterry: I have the old folder if needed, but not sure you have time [14:59] Laney, I saw it mentioned, but to be honest it's only an init system to me, as long as my computer boot I couldn't care less [14:59] didrocks, meh, I have a host of other things to do in deja-dup when I do have time for it (like adding google drive and dropbox support now that U1 is dead) [14:59] I still don't get the hype about init systems ;-) [15:00] mterry: yeah, what I envisionned! :) [15:00] well... yeah, I don't really mind so much at that level [15:00] Laney, we are sure it's going to happen this cycle? [15:00] I hope we don't push too early/agressively [15:00] I'm interested in the level of "this is work that we are going to do, so let's make it work well" [15:00] right [15:00] I'm going to be interested as well when we are about to change [15:00] to make sure we don't screw our users [15:01] seems we'll open utopic soon, so I'll upload the PPA fixes to utopic ASAP [15:01] indeed [15:01] then it's rather easy to test [15:01] I don't know if the target for switching init is this cycle or next [15:01] I hope next [15:02] let's see [15:02] Seems ambitious to expect it to be done in 6 (3) months but who knows [15:02] yeah [15:02] especially with the phone milestones this cycle [15:03] yeah, for sure we need all of our upstart jobs to have corresponding systemd equivalents and also adjust the phone hacks to work [15:03] whether that takes one or two cycles doesn't matter that much [15:03] nod [15:12] ah, a fresh new utopic chroot [15:12] luverly [15:32] What handles screen locking these days? "unity-panel-service"? [15:33] qengho, unity [15:33] I'm filing a password-prompt bug report. [15:33] why? [15:33] Thx. [15:33] what issue? [15:33] Username disappears when cursor blinks. [15:33] weird [15:34] not reported yet afaik [15:34] qengho, btw I tried the chromium issue didrocks reported, it's trivial to trigger here as well, weird that you don't see at least the tab ordering being buggy ... what version did you try on, the trusty one? [15:35] seb128: yes. I made a while loop. 150 runs later, all is still normal. [15:35] qengho, did you try the steps I described in the bug? [15:35] open a new chromium [15:35] open 4 tabs [15:35] got the second one, load a site (e.g launchpad) [15:36] middle click on a link (e.g "ubuntu") [15:36] that opens as tab 3 [15:36] restart chromium [15:36] seb128: I didn't try that. I was looking for missing tabs. [15:36] the tabs moves to the end of the list here [15:41] qengho, I can confirm the missing tabs if I open some 20 tabs [15:41] not with lower numbers though [15:41] qengho, I put the url of a similar upstream chrome/win report in the launchpad bug btw [15:44] seb128: thanks. [15:44] qengho, yw, let me know if you need more info on the issue [16:15] slow systemd boot is slow [16:17] Laney, liar ... they told me its faster ! [16:18] * ogra_ actually asked about that two days ago "numbers show a minimal debian system boots significantly faster than with upstart" [16:18] :P [16:18] make your system more minimal ;) [16:19] I think it's actually hung [16:19] this is where it would start lightdm [16:19] ah [16:32] aha [16:32] I had a lightdm from the ubuntu-desktop PPA instead of pitti's [16:36] wee, hello unity! === gatox is now known as gatox_lunch [17:08] alright, happy weekend! [17:11] Laney, thanks, you too [17:11] same here [17:11] have a good w.e everyone === m_conley_away is now known as m_conley === gatox_lunch is now known as gatox === m_conley is now known as m_conley_away === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away === brainwash_ is now known as brainwash === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away