[08:49] <kwwii> pitti: hey man, did both of my packages get into the last alpha? (the ones I gave you before leaving for vacation)
[08:49] <pitti> hey kwwii
[08:49] <pitti> kwwii: which versions? (just use "rmadison foo" to check the version of foo in all Ubuntu releases)
[08:51] <kwwii> ahh, thanks, now I see it
[09:04] <lool> seb128: Hey, welcome back
[09:05] <seb128> hello lool, thanks
[09:05] <lool> seb128: We discussed some desktop bits on Friday
[09:05] <lool> seb128: (I was the fallback contact for desktop stuff last week as everybody was on vac)
[09:05] <seb128> who is we and which ones exactly? ;-)
[09:06] <lool> During release meeting
[09:06] <lool> You might want to read the first part of the log, skipping mobile stuff
[09:06] <lool> Concerning desktop, gvfsd-trasher crasher and gtk+ screen flicker were discussed
[09:07] <lool> mdz brought up an old gpm bug
[09:07] <seb128> I've noticed those bugs in the summary
[09:07] <lool> I mentionned cheese shall be updated
[09:07] <lool> seb128: I bought a webcam last week
[09:07] <seb128> the gvfsd-trash crasher is not really an issue out of the number of duplicates and I've no clue how to get good informations or fix it
[09:08] <seb128> the gtk+ screen flicker, mclasen consider it as an xorg bug
[09:08] <lool> seb128: Yeah, me neither, the valgrind log isn't very helpful
[09:08] <lool> seb128: That said, the valgrind log might be better with more ddebs or something
[09:08] <seb128> no
[09:08] <lool> seb128: The problem with the screen flicker is that it is a lack of API on the xorg side
[09:08] <seb128> I did get valgrind logs for the trash crasher using a debug build, etc
[09:09] <seb128> there is an incorrect jump listed but the top stacktrace has no symbol
[09:09] <lool> seb128: So what happened is that gtk+ moved to a new API only to realize that it was too expensive/disruptive
[09:09] <seb128> I don't think it's in my power to fix the xorg api ;-)
[09:09] <seb128> or gtk
[09:09] <seb128> I would rather prefer let upstream sort that
[09:09] <lool> seb128: Exactly my point, I agree with the analysis that going forward we ought to fix the API
[09:10] <lool> seb128: But this will be in xrandr 1.3, then all drivers will have to be updated to support it, then gtk+ will be able to use it
[09:10] <lool> And the new API wasn't yet accepted
[09:10] <lool> So we're talking of a 3 months minimum cycle here
[09:10] <lool> (AIUI)
[09:10] <seb128> I'm confident that upstream will do something for FC10 which is due around the intrepid schedule too
[09:10] <seb128> gtk upstream
[09:11] <seb128> so I would just let them sort that
[09:11] <lool> Ok; I personally think gtk+ should defer using randr's problematic API until it has what they need
[09:11] <lool> Hey pitti, welcome back!
[09:12] <lool> pitti: You also might want to poke at the release team meeting's log from friday!  :)
[09:12] <lool> pitti: One of the thing which came up was the consolekit config patch to not allow everybody to shutdown or reboot the machine; this prevents shutdown/reboot from the desktop though
[09:13] <pitti> lool: hey, thanks
[09:13] <lool> pitti: I guess we should revert to the gdm socket method
[09:13] <pitti> lool: right, either we enable the functionality in CK like upstream, or use gdm
[09:13] <lool> pitti: Yup; I thought this was your choice to make as you were on top of misc hal/dbus/ck issues and have some security clue
[09:14] <lool> pitti: I am disturbed too with this shutdown/reboot dbus interface
[09:14] <pitti> lool: I have bug 250506 open in a tab to address it ASAP
[09:14] <lool> That's indeed the one we discussed I think
[10:11] <huats> morning everyone
[10:12] <seb128> lut huats
[10:12] <huats> plop seb128
[10:12] <huats> I hope you had nice (and deserved) holidays :)
[10:14] <seb128> yes, excellent, thanks ;-)
[10:14] <huats> :)
[10:33] <wgrant> seb128: Since you seem to be lord of all things g-s-d and g-c-c, what do you think of my proposed fixes for bug #207781?
[10:34] <seb128> wgrant: I've no clue about touchpad so I would prefer let somebody else comment
[10:34] <seb128> wgrant: and apparently there is no patch attached to this bug?
[10:35] <wgrant> seb128: That's true, I apparently forgot to attach debdiffs.
[10:35] <wgrant> seb128: Who is more qualified?
[10:35] <seb128> you could ask bryce maybe or #ubuntu-x
[10:35] <wgrant> OK, will do. Thanks.
[10:36] <seb128> would be nicer to have an xorg api to query that rather than relying on the naming
[10:37] <wgrant> My patch relies on the name of a property, but that property is what is used to set things, so it's safe.
[10:37] <wgrant> I can't see a better way.
[11:25]  * NCommander is away: This creature sleeps beyond the reaches of time itself
[13:45] <seb128> hey pedro_
[13:45] <pedro_> hey seb128! welcome back ;-)
[13:46] <seb128> thanks ;-)
[13:46] <seb128> pedro_: pitti has started fixing the retracers, that's not perfect yet but the intrepid one should be running again now
[13:46] <pedro_> woohoo go pitti ;-)
[13:46] <pedro_> seb128: thanks for let me know dude
[13:52] <vuntz> seb128: hey
[13:52] <seb128> hello vuntz
[13:52] <vuntz> seb128: just wanted to know what were the ubuntu plans for the logout dialog?
[13:52] <seb128> vuntz: I was sort of waiting on you because you told me you were going to land the opensuse dialog upstream but I guess that's not going to happen this cycle now?
[13:53] <vuntz> seb128: I hope that http://bugzilla.gnome.org/show_bug.cgi?id=507101 will get committed today, but I'm not sure it's fine for you
[13:53] <vuntz> seb128: heh. So, it's going to happen really soon now. Just need an icon and a second release team approval
[13:53] <seb128> vuntz: ah excellent
[13:54] <andre__> vuntz: ping those lazy r-t lurkers, ping them to death!
[13:55] <vuntz> andre__: waiting for the icon to be ready (I'm making artists work right now ;-))
[13:56] <vuntz> seb128: you can see the screenshots at the end of the bug
[13:56] <vuntz> seb128: if you want one dialog with everything, it's possible, but you'll likely need to patch gnome-session (I suggest adding an Ubuntu-specific dbus method for the new dialog)
[13:56] <seb128> vuntz: right, I like the dialog but I've the feeling we will be asked to get a variant listing all the options for ubuntu
[13:57] <seb128> right
[14:16] <seb128> lool: how busy are you nowadays? would you have some time to look at desktop packaging issues?
[14:18] <lool> seb128: I'm overcommitted, but depends what you're going to ask
[14:20] <seb128> lool: the pygobject update is tricky because they added a library and that doesn't play nicely with the multi python versions, doko started doing the packaging changes required before his holidays but there is a bug and it doesn't build
[14:20] <seb128> lool: I've planned to look at it but I think I'll be swamped with updates and catching up this week so I was trying to see if you were maybe interested by having a look at the update
[14:21] <lool> seb128: What source version didn't build?
[14:21] <seb128> 2.15.2
[14:21] <lool> Where I can get this?
[14:21] <lool> ("buildd" says 2.15.1-0ubuntu2 and that it built fine)
[14:21] <seb128> lool: http://people.ubuntu.com/~seb128/pygobject/ is the work in progress that doko did
[14:21] <seb128> lool: right, I didn't upload the new one since it's buggy
[14:25] <seb128> lool: you have the version before the changes (which fails to build because there is a conflict on the new lib because python versions) and the doko changes which ftbfs too but for an another reason (not sure which one now, I looked at it before my holidays quickly)
[14:25] <lool> seb128: Do you know what doko intendend to do?
[14:25] <lool> He added autoconf and automake bdeps and changed the way the source is built
[14:26] <lool> Ah I need to rebase on my version it seems
[14:26] <seb128> lool: I think he wanted to version the library name, he said that's similar to what python-svn does
[14:31] <lool> seb128: Do you know why the library is public instead of being e.g. a python module?
[14:31] <lool>         - Add a new installed library libpyglib-2.0, which contains the extension
[14:31] <lool>           API for third-part modules instead of relying on macros which accesses
[14:31] <lool>           struct fields.
[14:31] <lool> I'm not quite sure who's going to use it
[14:31] <lool> There's no .pc file at least
[14:32] <lool> Hmm there's also public API
[14:34] <seb128> lool: no, better to ask to jdahlin on the gnome IRC
[14:34] <seb128> you already did ;-)
[14:40] <kwwii> seb128: the ubuntulooks package was removed from the seed for intrepid, right?
[14:41] <seb128> no clue, but that's public information so you can look as quickly as I can do
[14:41] <kwwii> erm, where?
[14:41] <lool> kwwii: lp:~ubuntu-core-dev/ubuntu-seeds/intrepid or some such
[14:42] <lool> /platform.intrepid and /ubuntu.intrepid rather
[15:48] <mpt> loic, I've been pointed in your direction for implementing improvements to the codec installation interface. Do you have a few minutes to discuss the current implementation?
[15:48] <mpt> lool, ^^
[16:50] <lool> mpt: Hmm
[16:50] <lool> mpt: Sure
[16:50] <lool> mpt: I might not be the best person depending on what you ask, but I'm happy to discuss this topic
[16:52] <mpt> lool, the window that lists the available codecs looks like a subset of the Add/Remove Programs window, right? How difficult is it to change bits in that window?
[16:52] <lool> mpt: It's gnome-app-install
[16:52] <lool> mpt: Indeed, it's built using synaptic IIRC
[16:52] <mpt> which? :-)
[16:53] <mpt> Specifically, the Help button needs removing (currently it does more harm than good in the codec context), and the "Quick Introduction" is a non-sequitur
[16:53] <lool> mpt: What's the problem with the Help button?
[16:54] <lool> mpt: When I press it, I land in the "Install and Remove Applications" manual in yelp
[16:54] <mpt> The help it displays is completely irrelevant
[16:54] <mpt> exactly
[16:54] <mpt> It's nothing whatsoever to do with installing codecs
[16:55] <mpt> and whenever that happens it makes people less likely to use help anywhere else
[16:56] <lool> mpt: Ok, so currently this uses a generic tool, gnome-app-install; you can launch it manually and imagine that it's useful in different scenarii; I fear that if you want to provide help specialied for code installation, you will have to provide some sort of wrapper or derived app which links to that
[16:56] <mpt> It would be nice to have help for it, but we're past FeatureFreeze, so just removing it is probably best for now
[16:56] <mpt> lool, yes, I realize that, that's why I'm asking how difficult it is to change bits in this window
[16:57] <lool> What you want to do is change bits in this window when it's called for codec installs, correct?
[16:57] <mpt> exactly
[16:57] <mpt> I assume the "Canonical Codec Store" link is context-dependent
[16:58] <lool> I agree that solving this "properly" might be too intrusive post FF, but I wouldn't want to remove the Help button unconditionally
[16:58] <lool> Indeed
[16:58] <mpt> Right, I don't want to remove the Help button unconditionally, just in the codec situation
[16:59] <lool> mpt: I guess you probably know some python; check /usr/lib/python2.5/site-packages/AppInstall/distros/Ubuntu.py
[17:01] <lool>         (label, url) = distro.get_codec_information_link()
[17:01] <lool>         if (label is not None) and (url is not None):
[17:01] <lool>             button = gtk.LinkButton(url, label)
[17:01] <lool> mpt: in AppInstall/activation.py
[17:01] <mpt> I'm in 8.04 at the moment
[17:01] <lool> mpt: grab lp:~ubuntu-core-dev/gnome-app-install/main
[17:01] <mpt> ah, thanks
[17:03] <mpt> I was just wanting to know whether it was possible :-)
[17:04] <lool> mpt: Sure, it should be
[17:04] <mpt> ok
[17:04] <lool> mpt: BTW it's the first time I poke at this :)
[17:05] <lool> It seems the UI already supports various mode of operation by subclassing ActivationStyle
[17:05] <lool> class CodecSearchActivationStyle(SearchActivationStyle):
[17:05] <lool> mpt: So it's very trivial
[17:06] <lool> mpt: I think it would be cleaner to provide Help for installing codecs rather than dropping the button completely though
[17:06] <mpt> Sure, but I'm assuming we can't do that past feature freeze
[17:06] <mpt> If we can, I'll happily write some
[17:07] <mpt> This will all likely be revamped for Intrepid+1
[17:09] <lool> mpt: I think it's a fairly trivial UI tweak, but I could be wrong
[17:09] <lool> mpt: You could discuss this with release team to make sure
[18:00] <mpt> Thanks for your info lool
[19:17] <carciofo> hallo alle zusammen
[19:55] <ember_> what can make pkg-config --cflags return something but --exists none?
[19:57] <ember_> seb128: in gtk2-engines i've added to .install the right .pc but it seems debian/ has the old .pc, should i remove the old .pc from debian or revert the .install?
[20:06] <Laney> If Banshee sometimes stops responding to GNOME global hotkeys that I've set up, whose bug is that?
[20:06] <Laney> And if the answer is "it depends", how can I find out?
[20:14] <seb128> ember_: why did you add the .pc to start? it was not installed because the debian maintainer decided it was not required
[20:51] <carciofo> does any1 know about video editing?