=== oCrazyLemn is now known as CrazyLemon === _thumper_ is now known as thumper === robert-ancell is now known as robert_ancell [06:33] good morning [06:42] hey pitti, how are you? [06:42] bonjour didrocks -- très fatigue :/ comment vas-tu ? [06:43] pitti: oh, pourquoi fatigué ? [06:43] moi ça va === Guest11916 is now known as happyaron [06:43] didrocks: je suis allé dormir tard hier soir, après basketball [06:43] ah… ok, donc peu de sommeil! [07:58] xnox: do you have anything against me merging gnome-keyring 3.14 (you touched it last)? [07:58] I have prepared a merge based on 3.10.1-1ubuntu7 a couple of days ago (in ppa:mitya57/gnome-test), it needs minor rebasing on -1ubuntu9 though. [08:27] oh. day off. oops [09:03] hey hey [09:03] morning Laney [09:03] how goes? [09:03] pretty decent [09:03] we had curry + cake last night [09:04] the cake in volume was 50% icing and 50% cake [09:04] \o/ [09:04] you? [09:04] nice! :) [09:04] everything's fine here, today is eye doctor day and see what happened in the last 2 years of half :/ [09:05] going to get a new prescription? [09:05] desrt: did you figure out how to use them? [09:05] probably, I'm not confortable anymore with my glasses when watching tv [09:06] I went like 5 years ago and was told I would need reading glasses at some point [09:06] still don't have any problems that I can detect though ... [09:06] well, I guess the longer you don't need them, the better [09:06] ya [09:07] the parents got them when they were about 45 [09:07] that's nice! [09:07] well, I see the positive side that at least, I have no issue with my tooth (never got *anything* to them) [09:08] ha, nice [09:08] we can't have advantages everywhere I guess ;) [09:08] guess the parents made you brush well early on ;-) [09:08] not especially, I would say I wasn't eating a lot of candies, but still… ;) [09:10] hehe [09:10] I need to switch back to offlineimap [09:10] syncmaildir is just too slooooooooowwwwww [09:13] I guess I'll switch to mut + offlineimap soon, thunderbird is getting slower and slower [09:13] my filtering are all server-side, so cost to switch is small === oCrazyLemn is now known as CrazyLemon [09:18] yep [09:18] * Laney runs a sync [09:31] moin [09:54] * mitya57 loves our branch/version numbering [09:54] A branch named lp:~indicator-applet-developers/indicator-applet/trunk.14.04 contains version 12.10.2+15.04.20141111. [09:56] bahaha [09:56] mitya57, hi, just wondering what is up with the "gnome-panel" transition? [09:56] I think if you manually fix the upstream part it works [09:56] i.e. the CI train uses that [09:57] ricotz: I patched all packages except one (apt-watch), and for that I filed RM request. [09:58] That is bug 1396690 (TL;DR: upstream is dead, and update-notifier is better) [09:58] bug 1396690 in apt-watch (Ubuntu) "Please remove apt-watch from archive" [Undecided,New] https://launchpad.net/bugs/1396690 [09:58] mitya57, i see, patched as in locally prepared? [09:58] Laney: I know, but then it will be less funny. [09:58] ricotz: no, in -proposed [09:59] I once wrote a UDD query to find the longest versions ever used in Ubuntu [09:59] the results were cool [09:59] * Laney wonders if it's still in psql history [10:00] OHO! [10:00] flashplugin-nonfree has been knocked off the top spot! [10:00] mitya57,ok, e.g. indicator-applet? [10:00] http://paste.ubuntu.com/9265630/ [10:01] Ooh [10:02] ricotz: indicator-applet doesn't need patching, it needs a simple rebuild (I filed a MP in hope that can be done via the CI Train). [10:02] link? [10:02] if it's just a rebuild then JFDI IMHO [10:02] i.e. self approve and CI train it [10:02] ricotz: The patched packages are uim, workrave, sensors-applet, gnubiff and netspeed. [10:03] Laney: https://code.launchpad.net/~mitya57/indicator-applet/no-change-rebuild/+merge/242470 [10:03] mitya57, yeah, although this one is missing ;) [10:03] I don't have rights to approve/merge it. [10:04] there [10:04] Thanks [10:04] train does the merge part [10:04] :) [10:05] Laney: now I should go to #ubuntu-ci-eng and request a silo, right? or just upload it directly to archive? [10:06] You can upload it or try to use the train if you want to figure out how [10:06] OK, let's try the latter =) [10:06] For no-change rebuilds the next uploader of a real change can just tell it to ignore the trunk/archive mismatch [10:09] mitya57: okay so add a line at the bottom of http://wiki.ubuntu.com/citrain (google spreadsheet) and fill in columns A B F H I J [10:11] I only have read access to that [10:11] blerg [10:11] do you have a google account? [10:12] maybe sil2100 can give you access? [10:12] sil2100: ^ please :) I am mitya57 at gmail com [10:12] Laney: if i do branches, would you do landing for me? or should i just dput into the archive...? [10:12] (for vivid) [10:13] what kind of branches? [10:16] Laney: rebuild unity greeter / control center to pick up 15.04 embeded in the image. [10:17] no-change rebuild> upload IMHO [10:17] but also you should be able to use the train yourself in the same way I'm trying to get mitya57 able to right now [10:19] * mitya57 wonders why xnox ignores his questions about gnome-keyring merge [10:19] mitya57: hm? [10:19] xnox: http://irclogs.ubuntu.com/2014/11/27/%23ubuntu-desktop.html#t07:58 [10:19] mitya57: his as in my own, or yours? [10:20] mitya57: sorry, missed in my irc proxy scrollback. [10:20] My merge [10:20] also http://irclogs.ubuntu.com/2014/11/25/%23ubuntu-desktop.html#t11:37 [10:20] No problem, I also miss it sometimes :) [10:21] mitya57: yeah, i don't want it yet. I'd want to have the current sru land into at least utopic for https://bugs.launchpad.net/ubuntu/+source/gnome-keyring/+bug/1387303 [10:21] Launchpad bug 1387303 in gnome-keyring (Ubuntu Utopic) "regression: gnome-keyring components can't be disabled anymore" [Undecided,New] [10:21] mitya57: and then do merge to get new shiny goodness. [10:21] mitya57: as there are about 9 dupes and people are angry. [10:21] Ok, fine, then I'll wait [10:21] mitya57: also don't merge before talking to TIL =) [10:21] mitya57: i'm happy to give up a few things, but not everything at the moment. [10:22] mitya57: access to the CI Train spreadsheet, right? [10:22] sil2100: yes [10:22] o/ Let me add you then, one momen [10:23] xnox: btw there are LOTS of memory leak fixes in upstream git, we should definitely have that in vivid [10:23] mitya57: yeah, yeah. [10:23] we will. [10:24] mitya57: re:apt-watch why not request removal in debian? and then removals are auto-propagated to ubuntu [10:25] Laney, mitya57: added to both spreadsheet and the LP team [10:25] xnox: in Debian we don't yet have a gnome-panel transition, so that is not urgent [10:25] sil2100: ta [10:25] sil2100: thanks a lot [10:25] mitya57: it fails to build from source or something? or needs porting? [10:25] mitya57: remember that we have documentation for our landing process here: https://wiki.ubuntu.com/citrain/LandingProcess [10:25] needs porting first from gconf to gsettings, and then to new gnome-panel [10:26] Laney: ah seb already uploaded unity-control-center version string bump. so it's only the greeter. [10:26] cool [10:26] mitya57: probably best to demote it from release pocket to -proposed, thus it would not block migrations but still in the development archive / in-sync with debian. [10:27] mitya57: is it the last blocker for the transition? [10:28] xnox: yes. indicator-applet is not yet uploaded but will be done soon [10:29] ~ubuntu-archive is still the right team (for demoting), right? [10:30] The docs mention pinging trainguards, but core-devs can do the silo assignment themselves, btw [10:35] mitya57: just saw https://code.launchpad.net/~mitya57/unity-settings-daemon/headerbar-xsettings - overriding via org.gnome.settings-daemon.plugins.xsettings overrides still work with the fixed_false stuff? [10:35] s/overriding/does overriding/ [10:35] Laney: the docs say that the assign silo button is in "Landing team tools" menubar, but where is that? [10:36] Laney: yes, that will still work [10:36] I see that next to "Help" at the top [10:36] ok, I'm going to build/review/include that [10:36] Ah I see [10:36] got it? [10:36] got it [10:36] Laney: did you read my whole message to ubuntu-desktop@? [10:36] (about that window buttons stuff) [10:38] yeah [10:38] I think we (larsu) were going to work on theming these this cycle [10:39] Actually maybe we don' [10:39] *don't need that fixed_false commit as it is a no-op [10:41] I know gsd overrides it to true, so maybe gtk won't change the default [10:42] it would probably be in bad taste for them to do so really [10:42] Laney: can I grab any empty landing ppa? [10:43] Laney: yes, I figured out that gtk default != g-s-d default only after pushing it [10:43] well I don't mind either way, so revert it in the branch if you like [10:44] if you leave the assign dialog at its defaults it auto-assigns one of the free PPAs [10:45] got this error from jenkins: [10:45] You must use POST method to trigger builds. (From scripts you may instead pass a per-project authentication token, or authenticate with your API token.) If you see this page, it may be because a plugin offered a GET link; file a bug report for that plugin. [10:45] usual fail [10:46] click the button [10:46] authenticate then try again [10:46] authenticate to jenkins that is, using SSO [10:47] It says that I am logged in [10:47] what happened when you clicked proceed? [10:47] * mitya57 logs out and then back in [10:47] Laney: indeed (what exactly are you talking about?) [10:48] looks like it worked [10:48] larsu: there's a patch to u-s-d to make headerbar decorations respect the gsetting [10:48] i.e. for Ubuntu they go on the left [10:48] also, hi, how's it going! [10:49] mitya57: so now I normaly switch to http://people.canonical.com/~platform/citrain_dashboard/ [10:49] Laney: hi! Good! Just found out that we needlessy copy .pngs for Ambiance and Radiance into both themes... [10:49] Laney: but why u-s-d? Shouldn't that be a gtk patch? [10:49] Laney: how are you? [10:49] usd has the xsettings manager [10:50] it's where the mapping from gsettings into xsettings goes, then gtk gets it from there [10:50] I am good [10:51] right, I didn't know it didn't do that part yet [10:51] want me to review? [10:51] ya, g-s-d does but ... forks [10:52] \o/ [10:52] if you like, I think it's mainly a cherry-pick [10:52] https://code.launchpad.net/~mitya57/unity-settings-daemon/headerbar-xsettings [10:54] Laney, larsu: pushed the new version with only second commit [10:55] ty [10:57] I was about to complain :) [10:57] looks good to me now [10:58] not the way I'd have done it, but since this is a cherry pick and working, I won't complain [10:58] Laney: I can't approve though, can you? [11:00] ye [11:01] larsu: want to comment approve? [11:02] done [11:02] merci [11:02] neat: "Lars Uebernickel (community)" [11:02] "(scum)" [11:03] ts ts [11:03] I mean "(valued contributor ♥)" [11:03] typo [11:03] it's a bit hostile in here this morning! [11:03] haha - I want to see the keyboard that makes typos like this easy [11:03] or your hands........ [11:03] dem slippery fingers [11:04] ;) [11:04] * Laney tapes chrisccoulson to a lamppost [11:04] i'll give you hostile [11:04] mitya57: welcome to the world of mouse clicking called citrain! [11:04] lol [11:04] right, I'm getting on the CI train [11:05] * Laney toots [11:05] we used to have #ubuntu-ci-choo-choo for bot notifications, I kind of miss it [11:07] The choo-choo bot wouldn't work anymore with all the changes that happened I suppose [11:08] Mirv: :-) [11:10] and the best thing is that all this knowledge will be obsolete soon™ when we get the airline :) === MacSlow is now known as MacSlow|lunch [11:11] While I am here: is there any way to merge something but not land to archive (i.e. when the change is minor)? [11:15] +1 to ”soon™” [11:16] mitya57: kind of, but I doubt anyone has ever done that since it's cheap to publish too. one can make merge&clean job ignore the fact that the package is not published/indestinationarchive. [11:16] and it's cheap to add another (small or big) MP to the same landing [11:16] ok [11:16] blerg [11:17] mitya57: what has been done though is direct pushing to trunk when there's no landing ongoing and eg. changelog needs updating to account for a no-change rebuild pushed to archives or such thing [11:17] ok [11:18] also, how does one build $random-non-cu2d-package in a silo? [11:18] (i.e. like Laney built webkitgtk for me) [11:18] I put them in 'additional source packages' and then assigned a silo [11:18] and I just dputted to it [11:19] core-dev:s can dput to the silo directly, like a manual upload but not to archives [11:19] because I'm in ~ci-train-ppa-service [11:19] oh, maybe not all core-dev:s [11:19] right, only https://launchpad.net/~ci-train-ppa-service/+members [11:19] yes, I am a core dev, but LP says I am not in ~ci-train-ppa-service [11:20] Not sure who can be added to that, talk to the administrators I guess [11:20] most landings are train landings and then non-train landings are pushed directly to archives, but mixed silos (feature needing both train and non-train packages) need some help from someone in that team [11:21] ok, will talk to admins if I need to test something like that again [11:21] oh, nice, indicator-applet FTBFS [11:22] * mitya57 fixes [11:24] pitti: you attached to your gvfs merge? I could borrow it if not [11:25] Laney: please go wild [11:25] ty [11:25] Laney: but I thought I already merged it [11:25] I uploaded the new release to exp [11:25] Laney: oh, exp [11:26] the wild west === alan_g is now known as alan_g|afk [12:00] didrocks or larsu: If you see bug #1394063 or the one seb128 was talking about yesterday, could you check ppa:ci-train-ppa-service/landing-005 please? [12:00] bug 1394063 in unity-settings-daemon (Ubuntu) "[vivid regression] brightness keys aren't recognized" [High,Confirmed] https://launchpad.net/bugs/1394063 [12:01] Laney: the keys are working for me though (I don't have the upower segfault) [12:01] it's not a segfault [12:01] Laney: sorry, keys work fine for me as well [12:01] ok [12:01] Laney: I meant, plugin not loading [12:01] I think some laptops bypass the usd/userspace layer for brightness [12:01] do you get a notification when changing it? [12:01] I do get it === alan_g|afk is now known as alan_g [12:02] WEIRD! [12:02] there was a race in trusty btw, sometimes, not media key (including brightness) were working [12:02] no* [12:02] only way was to reboot the user session [12:04] don't know about that [12:08] I don't get the notification! [12:08] now that I know this, I want it! [12:09] you guys are bad test cases :p [12:09] if offlineimap ever finishes syncing I'll dist-upgrade to vivid and try myself [12:09] (doesn't seem very likely it's going to do so any time soon) [12:13] Laney: missed the opportunity to say something like "do I have to do *everything* by myself" [12:14] larsu: don't want to challenge offlineimap syncing AND apt? :) [12:15] haha [12:16] heh [12:17] and yeah, ETOOMANYLA* again [12:17] needing a patch to sort this mess! ;) [12:24] \o/ [12:24] compiz works [12:24] that's been my \o/ for years, get a new release working ;) [12:24] (I know, not the same context here :p) [12:25] hah [12:28] no seb128 today? :_( [12:29] Sweet5hark: he took the day off === MacSlow|lunch is now known as MacSlow [14:12] Laney: do you know how to have your chroot sessions being dumped in /tmp ? [14:13] dumped how? [14:13] (apart from a symlink from /var/lib/schroot/mount/ of course) [14:13] like I want my schroot sessions to be in /tmp/ [14:16] hmm, don't know, you could try mounting a tmpfs at that directory though if that's what you want to do [14:17] yeah, I was thinking about that as well, you are not making your sbuild on tmpfs then, I guess? [14:21] I use union-overlay-directory [14:23] Laney: and you point it to a tmpfs dir? [14:23] not /var/lib/schroot/union/overlay/ ? [14:23] Yeah, or you could mount that as tmpfs I guess [14:23] interesting [14:24] I also put /var/lib/sbuild/build on tmpfs but don't immediately remember why [14:24] I think that's where it unpacks to [14:24] Laney: ah nice tip, let me have a try [14:28] Laney: yeah, you're right, it's the unpacking dir === m_conley_away is now known as m_conley [16:19] ruh roh [16:20] There's some problem with glib 2.43 and GSettings [16:20] I'm not getting a 'changed' callback [16:21] hmm [16:22] Laney: isn't what desrt was talking about? There is now an order between connecting a signal and (? getting a value?) [16:22] Probably [16:22] I'm just reading that commit [16:22] I think he talked about it during the sprint and last week meeting :) [16:23] larsu knows as well about this [16:24] Laney: https://git.gnome.org/browse/glib/commit/?id=8ff5668a458344da22d30491e3ce726d861b3619 [16:24] We do this in a simple way, using a simple argument: in order for the [16:24] I know, I'm read it [16:24] user to care that a value changed then they must have: [16:24] reading [16:24] 1) watched for a change signal; and then [16:24] 2) actually read a value [16:29] right, the reasoning is that you can't care about that it changed when you've never read the initial value [16:30] turns out that's wrong for a lot of use cases (for example, the gsettings-tool monitor command) [16:30] Laney: what are you using it for? [16:31] u-s-d watching the idle delay [16:31] it should be reading the value at startup, let me see [16:41] Laney: make sure it connects to the signal before reading the value [16:42] it does [16:42] this is weird though [16:42] what is? [16:43] https://git.gnome.org/browse/gnome-settings-daemon/tree/plugins/power/gsd-power-manager.c#n1710 [16:43] "Don't dim when charging" [16:45] why is that a thing? [16:46] I'm basically confused now [16:46] I guess because dimming is a battery-saving feature? [16:46] it means you don't read the delay on a desktop [16:46] so not really needed when not on battery [16:46] that too [16:48] the setting is called "Turn screen off when inactive for" [16:49] "unless on battery" :P [16:49] *not [16:50] maybe I'm looking in the wrong place [16:50] something is making my screen lock [16:50] (and turn off) [16:50] symptom was that $something wasn't noticing when I changed this setting [16:51] u-s-d seems like a good candidate for $something [17:03] ah [17:03] I think it might come in via the screensaver [17:11] bleh, changing it works now [17:11] * Laney puts this to one side === alex_abreu is now known as alex-abreu [17:50] ok I lied and didn't put it to one side [17:50] larsu: if you're still here ... [17:51] With new glib I don't see a g_settings_bind_with_mapping firing [17:51] is this affected by the same change? [17:52] bind should read the value initially [17:52] I see the first read but not subsequent ones [17:52] not *changes* [17:53] let me check [17:54] which flags are you passing? [17:54] * didrocks waves good evening [17:54] didrocks: bye! [17:55] G_SETTINGS_BIND_GET [17:55] good luck for your debugging larsu & Laney :) [17:55] thanks! [17:55] https://git.gnome.org/browse/gnome-session/tree/gnome-session/gsm-manager.c#n2701 ← that [17:55] Laney: code looks correct. It connects to the changed signal and then reads the value [17:58] Laney: you're sure it does the first read but then it doesn't get changed events? [18:03] larsu: Yeah. Try with glib from vivid-proposed if you want... [18:04] If you run gnome-session with --debug (/usr/share/upstart/sessions/gnome-session.conf append --debug there) you can see the watches being changed when the key changes [18:04] Laney: I'm trying with master right now but gotta run soon. No promises ;) [18:04] I'm running too, no worries, it'll still be broken tomorrow. :P [18:04] yay \o [18:05] but I believe this is the essence of the bug, so that's progress [18:05] * Laney goes away, ttyl! [18:05] * Laney remembers *not* to suspend as offlineimap is still going strong [18:06] enjoy your evening! [18:46] ricotz: indicator-applet landed, but I missed command-runner-applet (which was using libpanel-applet via pygi). Will look at it tomorrow. [19:26] Are there any plans to try building liblldb-3.5-dev on amr64 and ppc64le? [19:26] If not, codelite needs patching to not use it [19:26] *arm64 === benonsoftware is now known as \b === \b is now known as benonsoftware [22:14] robert_ancell: hey, is there a reason you don't create a mir window impl when the window class is GDK_INPUT_ONLY? [22:14] attente_, I don't think Mir has a concept of input only windows [22:17] oh, ok, there seems to be some code related to selecting text that depends on an impl existing [22:17] via GtkInvisible === m_conley is now known as m_conley_away [23:06] Laney: https://bugs.launchpad.net/ubuntu/+source/unity-settings-daemon/+bug/1397135 [23:06] Launchpad bug 1397135 in unity-settings-daemon (Ubuntu) "Error in `/usr/lib/unity-settings-daemon/unity-settings-daemon': free(): invalid pointer" [Undecided,New]