[02:21] <m_conley> chrisccoulson: ping
[08:00] <MacSlow> greetings everybody!
[09:28] <nerochiaro> smspillaz: ping
[09:45] <kvalo> kamstrup: hi. I need to use list of strings as a gobject property. any handy way to do that or do I need to create a class for that?
[09:45] <kamstrup> kvalo: G_TYPE_STRV
[09:46] <kamstrup> kvalo: used a boxed paramspec
[09:47] <kvalo> kamstrup: thanks!
[10:26] <rodrigo_> API, ping
[11:19] <smspillaz> nerochiaro: pong
[11:19] <smspillaz> I might need to run out in 30 mins though
[11:20] <nerochiaro> smspillaz: hi. i just sent you an email a few hours ago after pinging you with a couple questions about compiz
[11:20] <smspillaz> nerochiaro: ah ok, which address did you send it to ?
[11:21] <nerochiaro> smspillaz: sam dot spilsbury at canonical dot com
[11:21] <smspillaz> nerochiaro: cool
[11:21] <smspillaz> nerochiaro: do you want me to reply out here ?
[11:22] <nerochiaro> smspillaz: where you are most convenient
[11:22] <smspillaz> ok, I'll just do it here for the benefit of everyone
[11:22] <smspillaz> so for the decorations, the set up we have is a bit weird
[11:22] <smspillaz> basically in 2D mode - we reparent windows like normal, we have a tree like this:
[11:23] <smspillaz> frame -> (wrapper -> (window) | decor)
[11:23] <smspillaz> we do that in 3d mode as well
[11:23] <smspillaz> so in 2D mode and 3D mode, frame is InputOutput, as is wrapper
[11:24] <smspillaz> but in 3D mode, decor is InputOnly, in 2D mode, InputOutput
[11:24] <smspillaz> the decorators as they stand, don't really paint into the decor window directly, they paint into some other backing window and then reparent that into decor
[11:24] <smspillaz> so you really end up with something like this:
[11:24] <nerochiaro> smspillaz: InputOnly means that it can't be read ?
[11:25] <smspillaz> frame -> (wrapper -> (window) | decor -> decorator window)
[11:25] <smspillaz> nerochiaro: exactly
[11:25] <nerochiaro> smspillaz: but decorator_window can ?
[11:25] <smspillaz> nerochiaro: only in 2D mode
[11:26] <smspillaz> in 3D mode, the decorator paints a very small amount of decoration, as well as some text into a backing pixmap
[11:26] <nerochiaro> smspillaz: so essentially in 3d there's no way to access the pixmap where the decorations are painted ?
[11:26] <smspillaz> with an alpha channel
[11:26] <smspillaz> nerochiaro: there is, sort of
[11:26] <smspillaz> nerochiaro: link libdecoration from compiz and then you should be able to read _COMPIZ_WINDOW_DECOR which contains a pixmap
[11:27] <smspillaz> (_COMPIZ_WINDOW_DECOR is a property)
[11:27] <smspillaz> although, what exactly are you trying to do with the window and decoration here ?
[11:28] <smspillaz> (as for "fake" minimization, we have a hack for that, though it only works in 3D mode)
[11:28] <smspillaz> (see workarounds -> Keep minimized windows)
[11:31] <nerochiaro> smspillaz: well, i'm implementing the workspaces switcher for unity-2d, and i need to get "screenshots" of windows that will be displayed for the user to choose. in metacity we grab the pixmap of the parent window and that contains the decorations. we need to get these decorations also for compiz
[11:31] <smspillaz> nerochiaro: so in 2D mode you should get the decorations
[11:32] <smspillaz> (eg no opengl plugin loaded)
[11:32] <nerochiaro> smspillaz: but as far as i know most people that use compiz use it in 3d mode
[11:32] <smspillaz> nerochiaro: right, so which modes do we want to support ?
[11:32] <smspillaz> 3D and 2D ?
[11:33] <nerochiaro> smspillaz: that's a good question, but i think 3D for sure. 2D mode I don't know, i guess it's less important since unity-2d is supposed to be used when only 2D is available
[11:34] <smspillaz> nerochiaro: in that case it might be worth writing a compiz plugin directly to do your work :)
[11:34] <nerochiaro> er, i meant unity-2d+metacity
[11:34] <nerochiaro> smspillaz: can you expand on that please ?
[11:34] <smspillaz> since using your current approach (redirecting windows again) is not going to end very well I don't think
[11:35] <smspillaz> nerochiaro: well, if you want to support unity-qt along with compiz, then I think it will be faster to write the bit that actually does window compositing as a compiz plugin (eg your switcher and scale view)
[11:35] <smspillaz> or to use the compiz plugins for that directly :)
[11:38] <nerochiaro> smspillaz: well, we'd rather not branch the code so much to support compiz specifically. all the code we have right now for the "spread" functionality ("scale" in compiz terms, i guess) is implemented by copying the pixmaps for windows. if at all possible it would be better to do that for compiz too (maybe a plugin can expose these pixmaps and our app can read them ?)
[11:39] <smspillaz> nerochiaro: you can just get the pixmap with XCompositeNameWindowPixmap, it is just a rather unoptimized path to do it like that
[11:39] <smspillaz> and then for the decorations you'll need some magic to get them too
[11:39] <sladen> kamstrup: tedg: wondering about making the global menus a bit more robust on tear-down, to cure the various vague crashes, or make "unity --replace" work
[11:40] <nerochiaro> smspillaz: we are doing that right now and it seems to work quite well in practice, even on ARM machines.
[11:40] <sladen> kamstrup: tedg: so when unity/compiz dies.  Appmenu kills itself, the D-Bus interface goes away and local menus reappear until a new instance comes along
[11:41] <nerochiaro> smspillaz: even if there are probably more optimized ways to do it
[11:42] <smspillaz> nerochiaro: yeah :)
[11:42] <smspillaz> nerochiaro: so for the decorations, you'll probably need to copy some of the logic in the decor plugin to stretch the pixmap the right way
[11:43] <smspillaz> basically, in the decor spec the decorator sends over a pixmap as well as some quads which specify how the pixmap should be stretched
[11:44] <kamstrup> sladen: there some sense to that, but I don't know if it may have some odd side effets, we need tedg's insight on that
[11:46] <ronoc> lots of compiz crashes ...
[11:48] <nerochiaro> smspillaz: ok. i think i have a good idea now of what the options are. I'll talk a bit with the other guys in my team and we'll see which way to go
[11:48] <nerochiaro> smspillaz: thanks a lot for the help
[11:48] <smspillaz> ronoc: backtrace ?
[11:49] <ronoc> smspillaz, afraid not, just trying using a dual monitor
[11:49] <smspillaz> ronoc: weird
[11:49] <smspillaz> nerochiaro: np
[11:50] <ronoc> happens regularly when swapping between laptop screen and monitor
[11:51] <sladen> kamstrup: hopefully less odd-side effects that the present :)  (Eg. a user is never left without any menus)
[12:03] <czajkowski> sladen: you may be interested in http://www.odfplugfest.co.uk/
[12:12] <kamstrup> sladen: point taken :-)
[13:04] <nerochiaro> does anyone knows where unity gets its workspace switcher icon from ? i can't seem to find it anywhere
[14:56] <ogra> Kaleo, any chance that you can get to my merge request today (i would like to upload and file the MIRs)
[14:59] <Kaleo> ogra: I will try
[14:59] <Kaleo> ogra: I mean, today as in Brasilia time :(
[15:25] <ogra> Kaleo, did you get my ping above (seems i had a disconnect right after typing it in)
[15:27] <ogra> (just to make sure you got it, not to be pushy)
[15:40] <Kaleo> 15:59 < Kaleo> ogra: I will try
[15:40] <Kaleo> 15:59 < Kaleo> ogra: I mean, today as in Brasilia time :(
[15:40] <ogra> ah, k,thanks
[15:40] <ogra> indeed i missed that due to the reconnect
[16:29] <Kaleo> ogra: https://code.launchpad.net/~unity-2d-team/unity-2d/natty-release/+merge/47513
[16:29] <Kaleo> ogra: I reviewed
[16:29] <ogra> why is xdg useless ?
[16:30] <ogra> shoudl it be dropped completely ?
[16:30] <ogra> Kaleo, ?
[16:31] <ogra> do we not want these entries hidden anymore ?
[16:31] <Kaleo> ogra: I think it was introduced for no particular reason
[16:31] <ogra> ah, k
[16:31] <Kaleo> ogra: basically we should mimick what Unity does and I don't think it does that
[16:31] <ogra> Kaleo, the rest is ok with you ?
[16:32] <ogra> i'll remove that and the bits in .install and upload then, persia had some request to adjust the description and text for the unity-2d package too, i'll make that change as well before the upload
[16:34] <Kaleo> ogra: hop hop hop
[16:34] <Kaleo> ogra: did you read all the comments?
[16:35] <ogra> heh, yeah, my mailer is slow
[16:35] <ogra> i get them in piecemeal
[16:35] <Kaleo> ogra: please push the changes to the branch and then I can do a final pass and approve
[16:35] <ogra> k
[16:35] <Kaleo> ogra: thank you
[16:36] <ion> ogra: pkill -HUP -x gconfd-2
[16:37] <ogra> ion, ?
[16:38] <ion> 125	+ # Tell all running daemons to reload their databases
[16:38] <ion> 126	+ kill -s HUP `pidof gconfd-2` >/dev/null 2>&1 || true
[16:39] <ogra> ah, well, not my code
[16:39] <ogra> i only merge the packaging, is that the new way of restarting gconfd ?
[16:40] <ogra> i dont think it matters as long as it gets the HUP, but if you insist using pkill i'll do that
[16:40] <ion> Dunno about restarting, but pkill is better than kill $(pidof …). No need to hide its output, for instance, since “no such processes” does not cause an error.
[16:41] <ion> Running ‘kill -HUP’ is an error.
[16:43] <Kaleo> ogra: I think all the gconf handling in our package is very shaky
[16:43] <Kaleo> ogra: and needs cleaning up
[16:43] <Kaleo> ogra: but I suggest we do that in a future MR
[16:43] <Kaleo> (Merge Request)
[16:44] <ogra> yeah
[16:44] <ogra> lets first get the packages in order (and all that stuff to main before alpha2)
[16:46] <Kaleo> right
[17:14] <ogra> Kaleo, updated
[17:16] <Kaleo> ogra: what's going on
[17:16] <Kaleo> ogra: I received an email as in you resubmited the MR
[17:16] <Kaleo> ogra: but you did not right?
[17:16] <ogra> i updated
[17:16] <Kaleo> ogra: you just did bzr push
[17:16] <Kaleo> ogra: great
[17:16] <ogra> bzr push and updated the MR
[17:17] <Kaleo> ogra: what do you mean by updated the MR?
[17:17] <Kaleo> ogra: changed the description?
[17:17] <ogra> (i also quickly fixed bug #703170 alongside)
[17:17] <Kaleo> ogra: great
[17:17] <ogra> no, i requested another review
[17:18] <ogra> after my commits
[17:18] <Kaleo> ogra: you don't have to
[17:18] <Kaleo> ogra: jsut bzr push is enough
[17:18] <Kaleo> just*
[17:18] <ogra> oh, ok
[17:18] <Kaleo> ok, let me see
[17:19] <Kaleo> ogra: is the switch to pkill important?
[17:19] <ogra> it wont break and ion is right that its cleaner
[17:19] <Kaleo> ogra: does it fix things? (I don't mind the change, I just want to know if it's important to backport it to the ppa packages soon)
[17:20] <ogra> it will work the same as the old code in the end
[17:20] <Kaleo> I think so
[17:20] <Kaleo> ok
[17:21] <Kaleo> ogra: I approved it
[17:21] <ogra> \o/
[17:21] <Kaleo> ogra: so now one last thing
[17:22] <ogra> thanks
[17:22] <Kaleo> ogra: if you set a decent commit message in the merge request
[17:22] <Kaleo> ogra: tarmac can merge automatically
[17:22] <Kaleo> ogra: but if you want to merge yourself, you are free to do so
[17:23] <ogra> well, trunk should be largely in sync, so i can easily do myself
[17:23] <Kaleo> ogra: oh, if it's not mergeable tarmac will fail anyway
[17:23] <Kaleo> ogra: it's just good to have tarmac doing it because: 1) you spend less time
[17:23] <ogra> right
[17:23] <Kaleo> 2) we can hookup automated premerge tests
[17:24] <Kaleo> (which we don't yet)
[17:24] <Kaleo> ogra: can we drop lp:~unity-2d-team/unity-2d/natty-release once it's merged?
[17:24] <Kaleo> ;)
[17:24] <Kaleo> at the very least delete lp:~unity-2d-team/unity-2d/unity-2d-default-settings-natty-release
[17:25] <ogra> as i said, i would like to talk to NCommander (preferably with you attending) first
[17:25] <ogra> i dont want him to feel left out by just killing off his branches
[17:26] <Kaleo> ogra: at least deleting lp:~unity-2d-team/unity-2d/unity-2d-default-settings-natty-release
[17:26] <Kaleo> ogra: because it is not used anymore
[17:26] <ogra> oh, yeah, that we can do indeed
[17:27] <Kaleo> good
[17:48] <ogra> Kaleo, hmm, you deleted unity-2d/natty-release already ?
[17:59] <Kaleo> ogra: nope
[17:59] <Kaleo> ogra: did not do anything
[17:59] <ogra> hmm
[17:59] <Kaleo> ogra: but when it's merged it does not appear in the list of active branches
[18:00] <ogra> https://code.launchpad.net/~unity-2d-team/ misees a good bunch
[18:00] <ogra> dee-qt/natty-release is gone too
[18:02] <Kaleo> I did not do anything :)
[18:03] <ogra> weird
[18:04] <ogra> oh, its LP ... it filters on activity
[18:04] <ogra> silly thing
[18:27] <lucazade> Kaleo: ping
[18:40] <ion> smspillaz: “implement improved borders & shadows (phase 1): DONE” – improved in which way? :-) I’m still unable to try Unity out myself due to nux crashing on fglrx.
[18:41] <ion> I do hope that means reducing the borders to zero pixels. That’s what OSX does and it looks great. :-)
[19:07] <artir> Hi world.
[19:08] <artir> I'd like to ask if I'm the only one that finds the shadow dropped by the unity panel on maximised windows weird
[19:13] <artir> http://ubuntuone.com/p/aJY/ VS http://ubuntuone.com/p/aJX/
[19:21] <artir> ( https://bugs.launchpad.net/ubuntu/+source/unity/+bug/705871 )
[19:30] <lamalex> tedg, is https://bugs.launchpad.net/appmenu-gtk/+bug/646794 actually fixed?
[19:31] <tedg> lamalex, Yup
[19:36] <lamalex> ok
[19:36] <lamalex> woo 1 bug status updated.
[19:36] <lamalex> making progress
[19:38] <lamalex> tedg, is there a way to launch an app with menus
[19:38] <lamalex> like an envvar of some kind?
[19:38] <ion> Unset UBUNTU_MENUPROXY if that’s what you mean.
[19:41] <tedg> lamalex, Like menus in the app window?
[19:42] <lamalex> yes
[19:42] <lamalex> to make them in window
[19:42] <lamalex> tedg, ^
[19:42] <tedg> bratsche, What was that env var?  SHOW_BOTH ?
[19:42] <tedg> lamalex, APPMENU_DISPLAY_BOTH
[19:45] <bratsche> Yeah, I think that's it.
[19:46] <lamalex> thanks
[19:46] <lamalex> no
[19:46] <lamalex> not it
[19:47] <bratsche> APPMENU_DISPLAY_BOTH is it
[19:47] <bratsche> APPMENU_DISPLAY_BOTH=1
[19:47] <lamalex> ummmm didn't work when I ran APPMENU_DISPLAY_BOTH=1 nautilus
[19:47] <bratsche> Did you quit nautilus first?
[19:47] <bratsche> Because it won't work if it's already running.
[19:47] <lamalex> oh right nautilus is already running
[19:48] <lamalex> IT WON'T DIE
[19:49] <lamalex> there we go
[20:03] <doctormo> Does anyone know how to get a list of installed applications (not packages, but applications as seen by USC)?
[20:04] <spikeb> aside from USC itself, I don't know of a way.
[20:07] <tedg> doctormo, ls /usr/share/applications/*desktop
[20:08] <spikeb> i feel stupid now heh
[20:09] <doctormo> tedg: OK, so we don't use pyxdg?
[20:10] <tedg> doctormo, I don't.  Pythons are icky and they bite.
[20:10] <tedg> doctormo, Yes, you should look in the XDG applications list, but probably most "installed" would be there.
[20:10] <doctormo> tedg: Nu-uh, pythons are the cuewtist-ity-animuls evar!
[20:10] <tedg> doctormo, I think didrocks did some playing with that for UNE and evolution-express stuff.
[20:11] <doctormo> tedg: Did you know I have commit access to pyxdg? happenstance of history.
[20:11] <doctormo> I'm just looking into the best way of using the Menu class
[20:13] <doctormo> I think the lack of documentation is what's killing it, it's harder to use than pgp.
[20:26] <Kaleo> lucazade: pong
[20:26] <lucazade> hi Kaleo
[20:26] <lucazade> I got a crash of unity-2d-launcher
[20:26] <lucazade> a fatal error
[20:27] <lucazade> I've reported it in the bug report with .xsession-errors
[20:28] <lucazade> https://bugs.launchpad.net/unity-2d/+bug/705336
[20:31] <lucazade> hope this could help... I wasn't able to reproduce it for days
[20:43] <lucazade> Kaleo: about the hang issue during logging out I've seen a small window called "Unknown" with only metacity decoration and no content for some seconds.. It appears after unity-2d-* processes are closed and before session actually finish
[20:45] <lucazade> it is more visible with a intel gma500 card (which is really slow)
[20:46] <Kaleo> lucazade: oh, nice a screenshot would help I suppose
[20:46] <Kaleo> lucazade: also xprop too
[20:46] <Kaleo> lucazade: that's harder though :)
[20:46] <lucazade> don't know xprop
[20:47] <Kaleo> lucazade: you launch it and then click on the window
[20:47] <Kaleo> lucazade: it will output the X window properties
[20:47] <lucazade> ok
[20:47] <lucazade> i'll try screenshot, xprop and gdm for the other issue
[20:47] <lucazade> *gdìb
[20:47] <lucazade> *gdb
[20:48] <Kaleo> lucazade: thank you very much
[20:48] <lucazade> :) thank you for unity!
[20:48] <Kaleo> lucazade: I am glad you like it
[20:49] <lucazade> a lot... also because my gma500 can handle it
[20:49] <lucazade> nicely
[20:49] <lucazade> on nvidia flies
[20:52] <Kaleo> sweet
[22:07] <lamalex> anyone here use avidmux?
[22:21] <m_conley> chrisccoulson:  ping
[22:22] <chrisccoulson> hi m_conley
[22:22] <chrisccoulson> did you have any luck with virtualbox?
[22:22] <m_conley> chrisccoulson: eventually, yes.  :)
[22:23] <m_conley> chrisccoulson: I'm looking into your Lightning bug
[22:23] <chrisccoulson> awesome :)
[22:23] <m_conley> chrisccoulson: I'm hitting a segfault when I boot up TB with Lightning and globalmenu-extension both enabled.  Are you experiencing something similar?
[22:24] <chrisccoulson> yeah, i get the same on startup. it crashes because it runs out of stack space :)
[22:24] <chrisccoulson> it hits an infinite loop on one of the menus provided by lightning
[22:24] <m_conley> ah, I see - so that's the bug you're talking about.
[22:24] <m_conley> gotcha
[22:25] <m_conley> k, that's all I needed to know.  thanks!
[22:25] <chrisccoulson> thank you :)
[22:25] <chrisccoulson> it would be good to get that fixed, and then i will enable a thunderbird build of the extension
[22:25] <chrisccoulson> i'm reluctant to do it now, because i don't want to break lots of peoples tbird installs ;)
[22:25] <m_conley> of course
[22:26] <m_conley> yep, i'll take a peek
[22:26] <chrisccoulson> thanks, that's appreciated