[00:38] <RAOF> desrt: Hm, WTF?
[00:39] <RAOF> desrt: Do dies in gtk_main, in GDK_THREADS_LEAVE.
[00:41] <RAOF> desrt: And, amusingly, removing the call to Gdk.Threads.Init() resolves it.
[00:49] <RAOF> desrt: So... I don't think that Do is doing anything wrong, per se. If you've got a threaded GTK+2.0 application, you're expected to call gdk_threads_init() (which is what Gdk.Threads.Init() calls)
[00:52] <desrt> RAOF: is it calling ENTER before it calls gtk_main()?
[00:52] <desrt> because it's supposed to...
[00:53] <RAOF> Hm...
[00:53] <desrt> this is probably the problem: the new mutex implementation does extra error checking vs. the old one
[00:53] <desrt> and it will complain if you try to unlock a non-locked mutex
[00:53] <desrt> read: abort
[00:54] <desrt> with posix mutexes that would be silently ignored
[00:55] <RAOF> Hm. Looks like it doesn't.
[00:55] <RAOF> Aha!
[01:07] <RAOF> desrt: Thanks
[01:08] <desrt> RAOF: sounds like you could use helgrind to verify if you really want....
[01:08] <desrt> that should catch the error even with the old glib version
[01:09] <RAOF> Valgrind and friends have a spotty history with mono. I could give it a go, though.
[01:35] <desrt> good times.  'do' is in LP
=== Mirv__ is now known as Mirv
[04:58] <pitti> Good morning
[04:58] <pitti> desrt: thanks!
=== ricotz_ is now known as ricotz
[08:04] <Laney> morning!
[08:04] <seb128> hey Laney, how are you?
[08:04] <seb128> good morning desktopers!
[08:05] <darkxst> Hey Laney, seb128
[08:05] <Laney> hey seb128
[08:05] <larsu> morning!
[08:05] <Laney> doing okay, looking forward to new monitor arriving
[08:05] <larsu> ooh, which one?
[08:05] <Laney> currently using a single one like some kind of animal
[08:06] <Laney> yo larsu, what up
[08:06]  * larsu only uses his laptop. Like an savage
[08:06] <Laney> http://www.dell.com/ed/business/p/dell-u2414h/pd
[08:06] <Laney> nothing super fancy
[08:06] <Laney> not hi dpi or anything
[08:06] <larsu> Laney: not much. Didn't even have tea yet
[08:06]  * darkxst is looking at 3 monitors and a laptop, like a king
[08:06] <Laney> since, y'know, gtk
[08:07] <larsu> looks nice
[08:07] <larsu> gtk does hidpi, no?
[08:07] <Laney> not per screen
[08:07] <larsu> ah right
[08:07]  * seb128 only use his laptop but misses his dual monitor config
[08:08] <larsu> I don't miss it at all
[08:08] <darkxst> Laney, only on wayland is hidpi per screen possible
[08:09] <willcooke> morning
[08:09] <willcooke> seb128, do you know who is maintaining Nautilius atm?
[08:09] <seb128> willcooke, define maintaining
[08:10] <willcooke> seb128, erm, bug fixes I guess.
[08:10] <seb128> upstream, debian, ubuntu?
[08:10] <willcooke> ubuntu
[08:10] <seb128> that would be our "group of gnome packagers"
[08:10] <seb128> e.g robert_ancell Laney larsu me
[08:10] <willcooke> FJKong, ^^^^^^^^^^
[08:10] <seb128> I tend to look after nautilus usually
[08:11] <seb128> but robert_ancell did fix some bugs there as well this cycle
[08:11] <seb128> do you have a specific question?
[08:11] <FJKong> seb128: great, thanks
[08:11] <seb128> that might be easier than trying to guess "who is the maintainer" in a universe where we don't have fixed maintainer :p
[08:11] <willcooke> seb128, kinda, FJKong is being asked to look at some things for Kylin and we're just working out which is the best option
[08:12] <willcooke> seb128, do you know what the file manager will be in U8?
[08:12] <FJKong> seb128: NUDT guys want me help them with nautilus or nemo
[08:12] <seb128> willcooke, FJKong: I saw that shared documents about nemo/nautilus, didn't really look at it yet
[08:13] <seb128> willcooke, not sure, I guess likely the filemanager app we have on the phone, though it has a long way to go before being a feature parity for a desktop filemanager
[08:13] <FJKong> seb128: got, but I want to get some comments from you on that
[08:15] <FJKong> seb128: if nautilus will be our choice in next version of ubuntu, helping them with Nautilus will be good
[08:16] <willcooke> I think we will all benefit from some Nautilus work for the next year or so
[08:17] <willcooke> oir
[08:17] <willcooke> or
[08:17] <willcooke> maybe we think about replacing Nautilus with Nemo on the desktop
[08:17] <willcooke> in 15.04
[08:18] <seb128> yeah, we pondered that and decided to stick with patching nautilus for now
[08:18] <seb128> if that stops being practical we might revisit
[08:20] <Laney> I don't think having desktop team members doing bug fixes on parts of our desktop can be bad
[08:20] <FJKong> so Nemo is popular in users now?
[08:20] <Laney> definitely work on nautilus imho
[08:22] <seb128> Laney, did anyone suggest it's not a good idea? I think we all agree than fixing nautilus bugs is good?
[08:22] <Laney> ?
[08:23] <seb128> <Laney> I don't think having desktop team members doing bug fixes on parts of our desktop can be bad
[08:23] <seb128>  
[08:23] <seb128> not sure what that was in response to
[08:24] <Laney> discussion about whether people should work on fixing bugs in nautilus
[08:24] <Laney> "yes"
[08:24] <seb128> well, the discussion is rather "do we want to keep tweaking nautilus, or do we want to look at an alternative like nemo"
[08:25] <Laney> is it?
[08:26] <seb128> well, the NUDT team was looking at using nemo
[08:26] <seb128> we sort of did also for trusty
[08:26] <darkxst> speaking of nautilus where is didrocks this week? nautilus build is stuck waiting on tracker promotion
[08:27] <seb128> he's on holidays for 10 days or so
[08:28] <seb128> Laney, the issue is that nautilus is axing features and going in a way we don't especially agree with, so we get a stack of non trivial change stacked on top of it
[08:28] <seb128> which is sort of work atm, but not sure it's going to keep working fine
[08:28] <seb128> so it might be easier to share work on nemo, which is a fork, rather than maintaining our own fork-though-patches-stacked
[08:30] <darkxst> not to mention every app you guys hold back in the dark ages blocks ubuntu gnome from having the current versions
[08:30] <seb128> that's another topic
[08:30] <seb128> but a known issue as well
[08:30] <seb128> though you overstate it, most of those are 1 or 2 cycles behind
[08:31] <darkxst> seb128, gnome-terminal is 3
[08:32] <seb128> that's not blocked because of csd or unity
[08:32] <seb128> that's a lame example
[08:32] <seb128> it was blocked because of bash issue iirc?
[08:32] <seb128> it's disappointing that you take that one to argue though
[08:32] <seb128> k, moving on to get work done, I'm not interested in trolling
[08:33] <darkxst> seb128, I am not trolling
[08:33] <seb128> well, you are taking an example which doesn't apply to the discussion
[08:33] <seb128> that one is blocked not because of Unity or us
[08:33] <seb128> but because of doko/bash/other issues
[08:34] <darkxst> seb128, right, anyway our  main concern atm is gnome-desktop
[08:36] <seb128> did you talk to robert_ancell about the issues in his changes?
[08:37] <darkxst> seb128, no haven't managed to catch him yet
[08:38] <seb128> k
[08:40] <darkxst> then there is the upower transition blocked only really by indicator-power now
[08:41] <seb128> yeah, charles is busy with phone/rtm work and that is higher priority for him than a desktop transition
[08:41] <seb128> not sure how to resolve that
[08:54] <darkxst> seb128, it needs some work since it was never ported to logind
[08:55] <darkxst> it also has a dodgy queue thing but that can probably just be dropped with the new upower
[10:13] <tiheum> Hi guys, could anybody tell me who should I contact to chat about packaging of Ubuntu themes while didrocks is on holiday?
=== Estilanda_ is now known as Estilanda
[11:15] <qengho> seb128: good morning. Will you please upload chromium?  https://launchpad.net/~canonical-chromium-builds/+archive/ubuntu/stage/+packages
=== MacSlow is now known as MacSlow|lunch
[11:41] <seb128> qengho, sure, can that be pocket copied (with binaries or without?) or is that a source upload usually?
[11:41] <seb128> tiheum, hey, just asking on this channel is fine, I can probably help you
[11:45] <tiheum> seb128: great, we want something quite simple: not to include the SVG app icons in the Suru theme package when it's built (but keep them on launchpad, so that they are still backed up and easly available).
[11:45] <seb128> why not?
[11:45] <seb128> scalable icons are usually useful no?
[11:46] <tiheum> seb128: in general, yes. But app icons are a bit complex and are not rendered nicely by Qt.
[11:47] <seb128> tiheum, shouldn't qt prefers the matching variant, if there is one, to the scalable one?
[11:49] <tiheum> seb128: I don't know why Qt does not use the PNG variants instead of the SVGs. Developers use a built-in function (QtIcon) which look up in the theme and find the right file automatically... and it seems to favor SVGs over PNGs
[11:49] <seb128> tiheum, shouldn't we fix qt then, rather than mangle our theme?
[11:51] <seb128> speaking of icons...
[11:51] <tiheum> seb128: to be clear, we still want to provide SVGs for all other icons but app ones
[11:52] <seb128> larsu, the code that look for themed icons is part of our toolkit iirc? would it be possible to not make it display warning when "fallbacking"?
[11:53] <seb128> larsu, running u-s-s display a stack of warnings like
[11:53] <seb128> WARNING - file:///usr/lib/i386-linux-gnu/qt5/qml/Ubuntu/Settings/Components/StatusIcon.qml:46:5: QML Image: Cannot open: file:///usr/share/icons/suru/status/scalable/preferences-system-phone-symbolic.svg
[11:53] <seb128> it fallbacks to load preferences-system-phone.svg just fine, but that's noisy
[11:53] <seb128> do you know if that's an issue in our toolkit?
[11:53] <seb128>  
[11:54] <seb128> tiheum, still feels like a buggy workaround, those scalables icons are useful under e.g unity7 alt-tab switcher
[11:55] <tiheum> in the apps/scalable folder, all the symbolic icons (monochromatic with a -symbolic suffix) are useful, yes. and we don't want to remove theme. Only the colourful app icons.
[11:55] <larsu> seb128: this warning comes from Qt. We could circumvent it by looking up the icon ourselves, but we're currently jst letting qt do that
[11:55] <seb128> hum, k
[11:56] <seb128> I wonder if I should just make u-s-s not use -symbolic icons then
[11:56] <larsu> hm wait, maybe we can look up the icon anyway
[11:56] <larsu> it would mean that we stat it twice, but meh
[11:56] <seb128> or we could fix qt (assuming we think it's a bug to warn in those cases, which I think it is, fallbacking is quite normal, not a warning case imho)
[11:57] <darkxst> svg's are very slow to load
[11:57] <tiheum> seb128, larsu: I think that the warning above is raised because you don't look in the right folder. preferences-system-phone-symbolic.svg is in apps/scalable not status/scalable
=== alan_g is now known as alan_g|lunch
[11:57] <larsu> interesting. I wonder why it even looks there
[11:57] <tiheum> darkxst: you're right, especially complex ones
[11:58] <darkxst> GNOME will fallback to scalable icons, but only as a last resort, not idea what Qt likes to do
[11:59] <tiheum> darkxst: I think that's because we don't provide PNGs that are exactly the size of the app icons in the dash (or elsewhere).
[12:01] <darkxst> tiheum, maybe you should? GNOME icon themes have a set of icons in various sizes
[12:01] <tiheum> app icons are provided in 256 and 512 px but with our grid unit system, we never meet exactly these sizes
[12:01] <larsu> but we always hit the same size, no? Can't we just include that?
=== MacSlow|lunch is now known as MacSlow
[12:04] <tiheum> darkxst: Gnome uses icons in a limited set of sizes (roughly 16, 24, 48, 64, 128 and 512px). On the mobile, we have different screen densities (18px/gu, 27px/gu, 30px/gu, etc) plus different size of icons.
[12:05] <larsu> seb128: preferences-system-phone-symbolic works for me, while non-symbolic doesn't...
[12:06] <seb128> larsu, I wonder why u-s-s displays those warnings then
[12:06] <larsu> me too...
[12:06] <seb128> larsu, it might be because it looks in the wrong category
[12:06] <seb128> not sure why though
[12:06] <tiheum> seb128: non-symbolic? there's no colourful version for this 'app' bevause it's part of the system settings
[12:06] <larsu> but you don't specify the category when saying which icon you want...
[12:06] <seb128> right
[12:06] <larsu> can you point me to the file/line where this icon is used?
[12:07] <tiheum> larsu: no, you don't need. Each icon name is unique.
[12:07] <larsu> right
[12:07] <seb128> larsu, let me try to come with a small example
[12:08] <larsu> seb128: this is what I use: http://paste.debian.net/116534/
[12:15] <darkxst> tiheum, right, no idea what is the right thing to do there
[12:20] <seb128> larsu, in fact we are using a custom component, so maybe a bug on our side, looking at it
[12:20] <seb128> larsu, http://paste.ubuntu.com/8097502/
[12:20] <seb128> larsu, that's an example, for the record
[12:22] <larsu> seb128: hm, I get the warning but see the image
[12:22] <seb128> larsu, yes, same in settings ;-)
[12:22] <larsu> weirdness
[12:22] <larsu> I'll investigate in a bit
[12:22] <seb128> well, that StatusIcon is doing weird things
[12:22] <larsu> I wonder why I don't see the warning in my example
[12:22] <seb128>         property var icons: {
[12:22] <seb128>             if (String(root.source).match(/^image:\/\/theme/)) {
[12:22] <seb128>                 return String(root.source).replace("image://theme/", "").split(",");
[12:22] <seb128> larsu, because you don't use StatusIcon
[12:22] <larsu> holy shit, what is this doing?
[12:22] <seb128> that's from ubuntu-settings-component
[12:23] <larsu> what's ubuntu-settings-component?
[12:23] <seb128> I wonder if that component is a leftover from the past and should be replaced with an Icon
[12:23] <seb128> larsu, https://launchpad.net/ubuntu/+source/ubuntu-settings-components
[12:23] <seb128> larsu, widgets shared between indicators and settings iirc
[12:23] <larsu> ah, thanks
[12:23] <seb128> larsu, https://launchpad.net/ubuntu-settings-components
[12:24] <larsu> description is awesome:
[12:24] <larsu> It's a set of components to be used for settings
[12:25] <seb128> yeah
[12:25] <seb128> Saviq added the use of StatusIcon when he did the switch to suru
[12:25] <seb128> Saviq, what is StatusIcon providing over a standard Icon or Image?
[12:26] <qengho> seb128: can be pocket copied.
[12:26] <seb128> qengho, great, including binaries?
[12:28] <qengho> seb128: #security copies binaries. A reason the PPA exists was to make copying binaries possible, without building on slower arches agaian.
[12:28] <seb128> qengho, k, going to do that then, thanks
[12:30] <ochosi> seb128: quick question, how do you determine the appicons in unity? read the desktop file or retrieve the icon with wnck and then pass that to the alt-tab dialog?
[12:34] <seb128> ochosi, no idea, would be a question for bregma or Trevinho|holiday
[12:34] <seb128> we got it from bamf iirc, which got it from clever match, which includes the desktop files use
[12:34] <seb128> but not only
[12:34] <seb128> why?
[12:34] <ochosi> ah right, bamf then
[12:35] <bregma> bamf it is
[12:35] <seb128> hey bregma ;-)
[12:35] <bregma> hey seb128
[12:35] <seb128> (got->get)
[12:36] <bregma> seb128, did will ask you about collecting a list of missing features for desktop convergence work?
[12:37] <seb128> bregma, no
[12:37] <seb128> willcooke?
[12:37] <bregma> yes
[12:37]  * willcooke reads the backlog
[12:38] <willcooke> sorry seb128 bregma  - havent had time to speak to seb128 yet
[12:38] <willcooke> I've got had a meeting moved, so seb128 we can chat now if you're free?
[12:38] <seb128> willcooke, works for me
[12:39] <GunnarHj> Hi Laney, any chance that you can revisit bug #1351092?
[12:39] <ubot5> bug 1351092 in language-selector (Ubuntu) "Configuration of the Droid Sans Fallback font" [Medium,In progress] https://launchpad.net/bugs/1351092
[12:39] <willcooke> seb128, hangout ok?  I think it would be quicker.  bregma you want in?
[12:39] <bregma> sure
[12:39] <Laney> GunnarHj: yes sorry, I will do soon, been distracted by other things
[12:39] <seb128> willcooke, let try that, my chromium might still be downgraded/work
[12:39] <GunnarHj> Laney: ok
[12:39] <willcooke> seb128, ack
[12:43] <Saviq> seb128, it can do non-square icons
[12:44] <Saviq> seb128, the image://theme provider dies when you try to pass only one of sourceSize.{width,height}
[12:44] <Saviq> seb128, so StatusIcon is a nasty hack iterating over the icon directories to find the matching file
[12:44] <seb128> Saviq, oh ok, do we have a bug to track cleaning that out?
[12:45] <Saviq> seb128, bug #1284235
[12:45] <ubot5> bug 1284235 in qtbase-opensource-src (Ubuntu) "Icon has to be sized explicitly" [Undecided,Opinion] https://launchpad.net/bugs/1284235
[12:45] <seb128> Saviq, thanks
[12:46] <seb128> Saviq, so that code seems to look in wrong categories and display warnings it shouldn't
[12:46] <seb128> Saviq, like http://paste.ubuntu.com/8097502/ displays a warning
[12:46] <Saviq> seb128, it looks in categories that you tell it to look like
[12:46] <Saviq> s/like/in/
[12:46] <seb128> Saviq, ^ ?
[12:46] <seb128> where do we tell it to use a category?
[12:46] <Saviq> seb128, yeah, that's the way it tries to find the image...
[12:47] <Saviq> seb128, http://bazaar.launchpad.net/~ubuntu-settings-components-team/ubuntu-settings-components/trunk/view/head:/Ubuntu/Settings/Components/StatusIcon.qml#L41
[12:47] <seb128> well, there is no category in that example (that I can say)
[12:47] <Saviq> seb128, it defaults to status, apps
[12:47] <seb128> hum
[12:47] <seb128> "but overriding in settings app doesn't work."
[12:47] <seb128> what does that mean?
[12:51] <Laney> RAOF: FYI, I just skipped the colord autopkgtests since they started running with dbus 1.8 (previously they were being skipped) - there are some issues with running in the autopkgtest environment. I made an attempt and fixed some of them but got stuck with the last test: https://github.com/hughsie/colord/pull/6
[12:51] <Laney> I made sure they pass when ran 'normally'
[12:52] <seb128> bregma, did you discuss "input methods handling" for unity8-desktop? Not sure in which category from that gdoc is should go in
[12:53] <bregma> seb128, no, but it's obviously one of the missing bits, not sure if it's a Shell or a Mir thing (or both?)
[13:06] <seb128> pitti, bug #1359191 is due to your changes/require a touch langpacks update I guess?
[13:06] <ubot5> bug 1359191 in ubuntu-system-settings (Ubuntu) "System Settings not localized" [Low,Invalid] https://launchpad.net/bugs/1359191
=== mjohnson151 is now known as mjohnson15
[13:21] <marga> Anybody knows till when Trevinho|holiday is on vacation and/or if he's going to attend DebConf?
[13:31] <pitti> seb128: right, that already got landed yesterday
[13:32] <seb128> marga, bregma should know
[13:40] <willcooke> OT: is Raring still in the archive?
[13:50] <bregma> marga, he should be back in "civilization" next week
=== alan_g|lunch is now known as alan_g
[14:32] <charles> seb128, I updated https://code.launchpad.net/~charlesk/ubuntu-system-settings/startup-wizard-thousands-of-volunteers/+merge/230643 as per your suggestion about the pot file
[14:33] <seb128> charles, thanks
=== m_conley_away is now known as m_conley
[14:45] <willcooke> seb128, on the Inspiron there is a BIOS setting which makes the function keys act like normal function keys
[14:46] <seb128> willcooke, k
[15:14] <marga> bregma, ok, thanks
[15:24] <willcooke> seb128, am or pm tomorrow?
[15:24] <seb128> willcooke, you get emails for those right? afternoon, I was going to ping you to ask if it's ok
[15:25] <willcooke> yeah, no worries, just adding it now
[15:26] <seb128> thanks!
=== gatox is now known as gatox_lunch
[16:30] <willcooke> I'm going EOD to nurse my cold.  I'll be around if some kind of bizarre emergency happens and, I don't know, you need someone to do the washing up#
[16:32] <ogra_> yeah, better nurse your cold than cool your nurse :)
[16:32] <ogra_> (and get well)
=== gatox_lunch is now known as gatox
=== alan_g is now known as alan_g|EOD
[17:15] <Laney> bye!
[20:41] <Noskcaj> Should gucharmap merges still be done in the ubuntu-desktop branch?
=== fginther` is now known as fginther
=== m_conley is now known as m_conley_away