[00:14] <chrisccoulson> mclasen - i don't have to do that for every call though do i? eg, if i make some gtk calls in a callback triggered from some gtk event, i will already have the gdk lock won't i?
[00:14] <mclasen> right
[00:14] <chrisccoulson> in this case though, the gtk calls are coming from a callback registered with g_timeout_add
[00:14] <chrisccoulson> so it's not protected already
[00:14] <mclasen> the threads docs in gdk spells it out
[00:15] <mclasen> idles, timeouts and the like are outside the lock
[00:15] <mclasen> signals are inside
[00:15] <chrisccoulson> mclasen, so, the sanest way would probably be for me to drop the g_timeout_add, and use gdk_threads_add_timeout?
[00:15] <mclasen> thats one way to do it
[00:15] <chrisccoulson> that's probably the way that requires the minimum effort from me right now :)
[00:16] <chrisccoulson> mclasen - thanks anyway
[05:42] <Sarvatt> anyone have a repeatable way to trigger one of these "_XAllocID: Assertion `ret != inval_id' failed" assertions that are happening?
[16:20] <asac> anyone running karmic?
[16:36] <czajkowski> lordie some update broke stuff on  lucid, I've no buttons now!
[20:47] <asac> 21:47 < asac> any apps using online/offline from NM besides empathy and epiphany?
[20:47] <asac> pidgin i guess too ;)
[20:48] <asac> does gwibber use such a feature?
[20:48] <asac> hmm kenvandine not here ;)
[21:19] <desrt> geesh guys
[21:19] <desrt> enough package updates?!
[21:20]  * desrt files a complaint with the appropriate authority
[21:20] <desrt> didrocks: hey?
[21:22] <mclasen> asac: weather applet, clock applet, evo
[21:23] <desrt> didrocks: just a poke to make sure that the valac update doesn't get dropped because of the (now past) beta freeze
[21:31] <asac> mclasen: thanks
[21:49] <jcastro> asac: last I checked gwibber did use nm
[21:50] <jcastro> asac: I'll find out for sure
[21:54] <jcastro> asac: ryan says "sort of", I guess ken disabled it because it wasn't working right
[22:19] <asac> heh. ok
[22:57]  * hyperair wonders whether it's considered a bug if entering too many characters into gnome-screensaver's password prompt can drive the system to oom