[03:26] <RAOF> Urfh.  SoundMenu integration in Banshee has broken because libindicate-cil bumped assembly version.
[12:47] <MickStep> Does the "custom-time-format" key do anything in indicator-datetime? Because I changed it to be 24hr format using:
[12:47] <MickStep> gsettings set org.ayatana.indicator.datetime custom-time-format "'%k:%M'"
[12:47] <MickStep> to no effect.
[12:48] <MickStep> nevermind I just realised my mistake
[15:03] <lucidfox> What is the "official" way to ship light and dark monochrome icons with an application?
[15:03] <lucidfox> a third-party one, I mean, not in the default Ubuntu install
[15:15] <kklimonda> lucidfox: ship them in ubuntu-mono-dark and ubuntu-light-dark? that's what shutter does
[15:16] <lucidfox> except there's no guarantee that the user will be using one of those themes
[15:16] <kklimonda> lucidfox: indeed
[15:18] <kklimonda> lucidfox: if you ship both icons in one set there is no way of telling which one should be loaded.
[15:19] <lucidfox> hmm, you give me an idea
[15:19] <lucidfox> perhaps the application could dynamically determine if the panel is-- no, that way darkness lies
[15:21] <kklimonda> :)
[15:21] <kklimonda> Gtk+ will have a native support for symbolic icons and then the problem will go away.
[15:22] <kklimonda> Gtk+ 3*
[15:28] <lucidfox> Symbolic icons?
[15:31] <ScottK> tedg: Do we want Inscape 0.48.0 in Maverick?  Please comment in Bug 628048.
[15:31] <tedg> ScottK, Yes, will do.
[15:31] <ScottK> Thanks.
[15:34] <kklimonda> lucidfox: it's mainstream's way of saying that we suck ;)
[15:34] <kklimonda> lucidfox: j/k - it's their approach to use simple icons in panel
[15:35] <lucidfox> so wait
[15:35] <lucidfox> GTK3 will have its own functional equivalent of app indicators?
[15:35] <tedg> ScottK, Sorry about that, I thought someone had been working on getting it in after we released.  I didn't follow up :(
[15:35] <tedg> ScottK, Thanks for pinging me.
[15:36] <kklimonda> no, just a support for symbolic icons - gnome-shell developers also use them or at least planning on doing that
[15:36] <ScottK> tedg: No problem.  I know if you vouch for it, any problems will get addressed.
[15:37] <tedg> lucidfox, And hopefully we'll support the symbolic icons in the next release of the indicators.   I really wanted to this cycle but ran out of time :(
[15:37] <lucidfox> What *are* symbolic icons? -_-
[15:37]  * ScottK thought icons were symbolic enough on their own.
[15:38] <tedg> ScottK, Yes.  Though, I'd love to redo the packaging on it sometime, it's very... old.  If you find someone asking about how they can help (potential MOTU or something) have them ping me.
[15:38] <tedg> lucidfox, Basically it means that some of the colors in the SVG have symbolic names (alert) instead of color values.  So the colors themselves are then defined by the theme, and replaced as the SVG is loaded.
[15:39] <tedg> lucidfox, Hopefully it'll mean you can make one icon for a variety of themes instead of tons with one color different.
[15:39] <lucidfox> Ooh, neat
[16:18] <lucidfox> tedg, since, as I understand, you're the main developer of libappindicator...
[16:19] <lucidfox> What should I do if I want to add an indicator to a Qt application?
[16:19] <tedg> lucidfox, You should be able to use their standard KNSI interface.  That's what libappindicator is based on.
[16:19] <tedg> agateau, ^
[16:20] <lucidfox> That's in kdelibs
[16:20] <tedg> Oops, I think it's KSNI not KNSI.
[16:20] <lucidfox> I'm talking plain Qt -_-
[16:20] <agateau> lucidfox: right now there is no easy solution :/
[16:20] <agateau> lucidfox: but Qt devs want to add native support for appindicator in Qt itself
[16:20] <lucidfox> Ooh, neat
[16:21] <agateau> lucidfox: I know Quassel devs implemented ksni support themselves
[16:21] <lucidfox> :o
[16:21] <agateau> lucidfox: you might find useful code there
[16:21] <lucidfox> sounds like I might want to merge that into my Q-XDG library
[16:21] <lucidfox> along with Quassel's implementation of XDG notifications
[16:22] <agateau> lucidfox: oh you are the one who wrote the icon theme for Qt lib?
[16:22] <bratsche> How would that work, in a cross-platform sense, for Qt to add libappindicator support?
[16:22] <bratsche> I thought all the Qt API is supposed to basically work on the different platforms.
[16:23] <agateau> bratsche: they want their appindicator app to be a platform specific plugin
[16:23] <bratsche> Oh okay.
[16:23] <agateau> bratsche: Qt provides a class named QSystemTrayIcon
[16:23] <bratsche> Cool, was just curious.
[16:23] <agateau> bratsche: and they want the class to use appindicator when available
[16:23] <bratsche> Nice.
[16:23] <lucidfox> What?!
[16:23] <agateau> bratsche: but it's tricky because they don't want qtgui to depend on qtdbus...
[16:23] <lucidfox> How... would... that... work.
[16:24] <lucidfox> the indicator API is more restricted than tray icon APIs
[16:24] <agateau> lucidfox: true, but not that much
[16:24] <agateau> lucidfox: at least, the kde implementation
[16:24] <agateau> lucidfox: is not as limited as the ubuntu gnome one
[16:25] <bratsche> htorque around?
[16:26] <htorque> yes
[16:26] <lucidfox> agateu> Does it allow both left and right clicks?
[16:26] <bratsche> htorque: Were you posting some issues with appmenu-gtk showing erroneous menus?
[16:27] <agateau> lucidfox: it does
[16:27] <lucidfox> :(
[16:27] <lucidfox> defeats the point that Ayatana is trying to achieve, I think
[16:27] <agateau> lucidfox: it predates ayatana trying to achieve a point
[16:27] <lucidfox> hehe
[16:27] <bratsche> Oh wait, it was Omer Akram.
[16:27] <htorque> bratsche, ok :)
[16:27] <bratsche> Sorry.
[16:28] <agateau> lucidfox: but I agree left and right click distinction is not a good idea
[16:29] <lucidfox> Hmm... I wonder if upstream GNOME could just port GtkStatusIcon to this indicator thing
[16:29] <lucidfox> while allowing unrestricted use of the old API... probably not, though
[16:29] <lucidfox> it allows way too much
[16:30] <agateau> I have no idea there
[16:30]  * agateau works on kde
[16:31] <bratsche> One of the main issues is that upstream Gnome doesn't really want to depend upon packages from Launchpad.
[16:31] <bratsche> They tend to prefer things to be hosted on git.gnome.org or git.freedesktop.org
[16:31] <kklimonda> bratsche: libindicator wouldn't be the first canonical project that is being hosted on git.gnome.org
[16:32] <kklimonda> couchdb-glib's main repository is already there.
[16:32] <kklimonda> I'd say that if there is a chance of getting indicator into upstream Gtk+ that would be worth the effort of moving development there :)
[16:34] <lucidfox> I agree fully
[16:34] <bratsche> Another issue that upstream has is that we require copyright assignment on our modules.
[16:34] <bratsche> kklimonda, lucidfox: I'm not disagreeing with you.  I'm just pointing out the reasons why it's difficult to do, and probably won't happen soon.
[16:35] <lucidfox> By the way, I'm really liking the changes to GTK3 so far
[16:35] <bratsche> Although Evolution was originally also copyright-assignment, and that didn't stop it from being in Gnome.
[16:35] <kklimonda> bratsche: copyright assignment also can be worked around - Intel has a similar one for clutter. I'm not saying it's easy to do -especially that gtk+ devs have some objections to code itself but it's worth it.
[16:36] <lucidfox> They removed all GDK drawing primitives in favor of Cairo
[16:36] <bratsche> Yeah, Company has been working on that.  Awesome stuff.
[16:36] <lucidfox> which means, ironically, that GDK no longer means GIMP Drawing Kit
[16:36] <lucidfox> but then it never meant that :)
[16:36] <lucidfox> really, that is
[16:40] <lucidfox> brasche, I take it you're intimately familiar with upstream GTK development? Because I have a few questions I've gathered after observing it from the outside
[16:41] <pavolzetor> Hallo, I need help with libindicate and icons
[16:44] <jcastro> hi klattimer
[16:44] <klattimer> hi jcastro hows it going
[16:44] <jcastro> I hear you've got more hours than I thought
[16:44] <jcastro> do you still have bugs in your queue?
[16:45] <klattimer> only a few
[16:45] <jcastro> ok
[16:45] <klattimer> I've been trawling through the last week of bugmail today
[16:45] <jcastro> ok
[16:45] <jcastro> let me check the list
[16:45] <klattimer> there's a few things which have gone from my bug queue and I think there are still pending issues
[16:45] <klattimer> so I'm looking to poke those a bit
[16:46] <klattimer> you might want to assign me some new high prior stuff if you like
[16:46] <jcastro> https://bugs.edge.launchpad.net/~karl-qdh/+assignedbugs
[16:46] <jcastro> is this it?
[16:48] <jcastro> tedg: kenvandine: what's up with https://bugs.edge.launchpad.net/ubuntu/+source/indicator-application/+bug/558841
[16:48] <jcastro> klattimer: ah, on that one you forgot to assign it to the desktop team so they can ship it
[16:49] <klattimer> jcastro: I haven't been able to assign to the desktop team for some time
[16:49] <klattimer> :/
[16:49] <klattimer> not sure what's gone wrong
[16:49] <jcastro> oh ok, no worries
[16:49] <klattimer> recently I've just poked seb about it
[16:49] <klattimer> that one might have slipped the net
[16:50] <jcastro> ok, manual poking is always better anyway
[16:50] <jcastro> what's the deal with this g-p-m leak? Maybe that would be a good one to finish off
[16:51] <klattimer> yeah was thinking of giving it a poke tomorrow and retesting to see if it still exists
[16:51] <jcastro> k
[16:52] <jcastro> ok let's do this
[16:52] <jcastro> keep pinging seb and kenvandine about the bt bug
[16:52] <jcastro> finish off the keyboard indicators, since those are really visible
[16:52] <jcastro> and then look at g-p-m
[16:52] <klattimer> k
[16:53] <jcastro> the kb indicator looks great btw
[16:53] <jcastro> well, other than the icon bug thing
[16:53] <jcastro> but whatever. :)
[16:54] <tedg> jcastro, I think that ayan was looking at that one.
[16:54] <tedg> jcastro, It seems to be a GTK thing.  He's got a test program that can cause it in GTK.
[16:54] <pavolzetor> could you help ome?
[16:54] <klattimer> icon bug?
[16:54] <klattimer> which icon bug?
[16:55] <jcastro> klattimer: the 2 icons thing?
[16:55] <pavolzetor> I don't know if it is bug
[16:55] <tedg> jcastro, The GPM one (if I wasn't clear)
[16:55] <klattimer> oh, the ibus and gsm showing the keyboards?
[16:55] <pavolzetor> I wanna add icon to message.im
[16:55] <jcastro> yeah
[16:55] <jcastro> https://bugs.edge.launchpad.net/ayatana-ubuntu/+bug/620331
[16:55] <pavolzetor> but program ends with segfault
[16:55] <klattimer> yeah, well fixable with flags
[16:55] <klattimer> bug we need flags which match mpt's style requirements
[16:55] <jcastro> we're waaay past Beta
[16:56] <klattimer> yeah I know
[16:56] <jcastro> I doubt we'll be getting flags any time soon
[16:56] <mpt> I don't have flag style requirements
[16:56] <tedg> pavolzetor, Can you pastebin the code that's setting the icon?
[16:56] <klattimer> well, for next cycle I'll be merging ibus and kb indicator into a single indicator and capplet
[16:56] <mpt> oh cool
[16:56] <klattimer> mpt you do, "two letters in a box"
[16:56] <klattimer> that's teh flags
[16:56] <klattimer> "flags" are just what they're called in the code
[16:56] <mpt> aha
[16:56] <jcastro> ok so what do we do with this bug for Maverick
[16:57] <klattimer> well, not a great deal unfortunately
[16:57] <jcastro> can we turn the icon off?
[16:57] <klattimer> we could drop the icon, but mpt seemed against that
[16:57] <jcastro> ok
[16:57] <klattimer> i.e. drop the icon, show only the label
[16:57] <jcastro> mpt: let's make a decision now, if we're going to defer I'd like to get him working on things we can fix
[16:57] <klattimer> with the showFlags gconf key turned on, only an icon is shown, but we need flags in order to do that
[16:57] <pavolzetor> shit, where can I get support for libindicate?
[16:58] <mpt> I'm not against it UI-wise as a temporary thing, it would just require a temporary API change just for that, which would be weird
[16:59] <klattimer> I suppose it's how tedg feels about the api change
[17:00] <tedg> klattimer, There shouldn't be a change required for icon only...
[17:00] <tedg> klattimer, Or do you mean icon and radio on the menu item?
[17:00] <klattimer> I mean hiding the icon from the indicator and only showing a label
[17:00] <mpt> tedg, they're talking about text-only titles
[17:01] <mpt> ... temporarily
[17:01] <tedg> mpt, klattimer, Ah.  Hmm, not sure where that's enforced.
[17:02] <tedg> I don't care that much, but it just means that on KDE it won't work at all.
[17:02] <tedg> As they don't have the label.
[17:04] <klattimer> hmm, well that puts us in a bit of a catch 22
[17:04] <mpt> I'm surprised you're still considering this 2.5 weeks after UI freeze.
[17:04] <klattimer> yeah, it's a bit late
[17:05] <klattimer> but we'd more or less decided just to leave it icon+label for now anyway
[17:05] <klattimer> there's not so much we can do about it in the short term
[17:05] <mpt> Set it to Natty and move on.
[17:05] <klattimer> if we get flags drawn, switching the gconf key and doing that would solve it
[17:05] <tedg> +1
[17:06] <klattimer> for natty I'll be merging ibus+kb indicator anyway
[17:06] <klattimer> so we'll only have one icon for both
[17:14] <tedg> ronoc, ping
[17:14] <ronoc> tedg, pong
[19:58] <vish> gah! appmenu doesnt notice when the app has frozen :/
[19:59] <vish> banshee just frozen but still the appmenu shows the menu! while it does not have any effect on banshee though :(  cant even effect a force quit by hitting the close buttons..
[19:59] <vish> had to unmaximize and then do a force quit.
[20:58] <htorque> hello, if there are some missing-icons icons in an appmenu, should this be reported (per application)?
[23:05] <bratsche> htorque: Yeah, I would report them per-application for now.
[23:11] <htorque> bratsche, thanks! actually those are icons, that are only shown when having menus_have_icons enabled, so after reading the comment from mpt on bug 608584 i'm not sure what to do