2012/01/24/#ubuntu-desktop.txt

=== mchro- is now known as mchro
[00:38] <TheMuso> c
=== ajmitch__ is now known as ajmitch
=== m_conley_away is now known as m_conley
=== m_conley is now known as m_conley_away
[05:07] <pitti> Good morning
[06:42] <didrocks> good morning
[06:46] <BigWhale> morning ...
[06:46] <RAOF> Aloha didrocks!
[06:47] <didrocks> good morning BigWhale, RAOF!
[06:54] <pitti> bonjour didrocks
[06:55] <pitti> desrt: just sent my GNOME membership application :)
[06:55] <didrocks> guten morgen pitti, how are you?
[06:55] <pitti> hey RAOF, how are you?
[06:55] <RAOF> pitti: Pretty good.  Yourself?
[06:55] <pitti> didrocks: quite fine, thanks!
[06:57] <desrt> pitti: nice :)
[07:03] <RAOF> To the cinema!
[07:05]  * didrocks goes from FTBFS to FTBFS with new compiz since yesterday :/
[07:09] <pitti> RAOF: what will you see?
[07:14] <pitti> vuntz: bonjour!
[07:14] <pitti> vuntz: for gnome bug 642692, I wondered how "committed" you are to the wnck_shutdown() API name
[07:14] <ubot2`> Gnome bug 642692 in general "Provide wnck_shutdown() (?)" [Normal,New] http://bugzilla.gnome.org/show_bug.cgi?id=642692
[07:15] <pitti> vuntz: if the name is fixed, and we just need to discuss the implementation, I would apply that patch to our packages already for broader testing and usage in notify-osd
[07:15] <pitti> vuntz: but I wouldn't like to vendor-patch an API change without your consent
[07:17] <pitti> didrocks: I think I already asked, but I forgot, sorry: What is "Provide some user journey in manual test case form" all about? (johnlea's WI on desktop-p-unity-quality)
[07:18] <didrocks> pitti: John as a lot of "user journey" about "how to start an application", "how to reveal the launcher", and such. He was supposed to turn them into test cases
[07:18] <didrocks> pitti: but I covered them in my test cases at the end
[07:19] <pitti> didrocks: right, was just going to ask -- sounds a lot like the manual test cases you have already?
[07:19] <pitti> didrocks: so I guess we can drop this one?
[07:19] <didrocks> pitti: yeah, I think we can safely drop it
[07:20] <pitti> done, thanks
[07:20] <didrocks> pitti: the test cases that are not covered are the multimonitor and multitouch ones
[07:20] <didrocks> multimonitor -> I'm waiting on the new work done upstream
[07:20] <didrocks> multitouch -> I need time… :/
[07:23] <vuntz> pitti: name is going to stay, feel free to push it
[07:24] <vuntz> pitti: I'll try to make time to look at it today/tomorrow
[07:35] <pitti> vuntz: ok, will upload it to Ubuntu then; merci!
=== tkamppeter_ is now known as tkamppeter
[08:53] <pitti> didrocks: I recently noticed that the launcher doesn't appear if there is a window underneath it; I just confirmed in a guest session
[08:53] <pitti> didrocks: do you mind a quick try on your machine, to ensure it's not something I picked up from a PPA?
[08:53] <pitti> go to an empty desktop
[08:53] <pitti> open a window (tried with terminal and devhelp, doesn't matter)
[08:53] <pitti> move it below the launcher
[08:53] <pitti> press Windows key -> launcher doesn'ta ppear
[08:54] <didrocks> pitti: it works here
[08:54] <didrocks> pitti: it doesn't appear at all, it's not a stacking issue?
[08:54] <pitti> right
[08:55] <didrocks> hum, I don't get this here (application maximized or not)
[08:55] <pitti> didrocks: ok, thanks for trying
[08:55]  * pitti downloads today's desktop CD and tries there
[08:56] <didrocks> pitti: but your issue remember of one I had at some point
[08:56] <didrocks> on oneiric
[08:56] <didrocks> so, there is maybe some screwed state around
[08:56] <pitti> happens for me in a freshly started guest session
[08:57] <pitti> start session, ctrl+alt+t, move, press Windows key
[08:57] <pitti> hm, wait
[08:57] <pitti> windows+number also doesn't work any more
[08:57] <pitti> didrocks: argh, silly me; ignore me, please
[08:58] <didrocks> pitti: oh? what was it?
[08:58] <pitti> seems I accidentally switched my keyboard from "windows" into "Mac" mode
[08:58] <didrocks> :)
[08:58] <pitti> there's a weird key combination to do that, and sometimes I hit that
[08:58] <pitti> *blush*, sorry
[08:58] <didrocks> *phew* no worry ;)
[08:58]  * didrocks hugs pitti
[08:58]  * pitti hugs didrocks
[09:08] <seb128> hey
[09:08] <pitti> bonjour seb128
[09:08] <seb128> hey pitti, how are you?
[09:09] <pitti> quite fine, thanks! how about you?
[09:09] <seb128> I'm good thanks
[09:14] <didrocks> saltu seb128
[09:14] <seb128> lut didrocks
[09:23] <chrisccoulson> good morning everyone
[09:26] <seb128> hey chrisccoulson
[09:26] <seb128> how are you?
[09:27] <chrisccoulson> hi seb128
[09:27] <chrisccoulson> yeah, i'm good thanks. how are you?
[09:27] <chrisccoulson> got the tbird and ffox final builds to do today, but that should be easy :)
[09:28] <seb128> chrisccoulson, I'm good thanks
[09:32] <didrocks> hey chrisccoulson
[09:32] <chrisccoulson> hi didrocks
[10:03] <mvo> seb128: remember that we talked about the segfault in gtk image with the new iconhelper stuff? I have a tiny testcase now http://paste.ubuntu.com/815206/ - obviously what the testcase is doing is not correct and yet it would be nice if gtk would not segfault here, especially not when using gtk :/ - essentially the problem is that icon_helper is NULL on destory and that causes the segv
[10:03] <mvo> seb128: fwiw, oneiric did not segv in the testcase
[10:04] <seb128> mvo, hey, right the old code was doing less in destroy
[10:20] <seb128> mvo, can you open a bug with the testcase?
[10:20] <seb128> mvo, btw is there any chance you look at s-c with the new webkit?
[10:20] <seb128> mvo, nothing else is having an issue, I wonder if that's a s-c bug or if you could come with a small testcase
[10:20] <mvo> seb128: where can I download it? desktop ppa?
[10:20] <seb128> mvo, yes
[10:21] <seb128> mvo, ppa:ubuntu-desktop/ppa
[10:21] <mvo> seb128: sure I can file a bug, I don't have much hopes other than "WONTFIX" though, this is why I did not do it already
[10:21] <seb128> mvo, well you seem to care enough to write a testcase, so we can as well have the bug for the record
[10:21] <seb128> even if it's wontfix it still show a problem we might hit again
[10:22]  * mvo nods
[11:45] <pitti> RAOF: FYI, milestone "precise-beta-1" does not exist (it's ubuntu-12.04-*); but as b1 is the default milestone for that blueprint I just merged it with  "work items:"
=== greyback is now known as greyback|lunch
=== MacSlow is now known as MacSlow|lunch
=== Omega- is now known as Omega
[12:30] <BigWhale> A question... Silly but important one! Is ~/Documents in Slovenia (or Germany or whatever) also ~/Documents or it gets translated?
[12:32] <pitti> BigWhale: it does get translated, but not quite at runtime
[12:32] <pitti> BigWhale: see ~/.config/user-dirs.{dirs,locale}
[12:33] <pitti> BigWhale: /etc/X11/Xsession.d/60xdg-user-dirs-update checks for locale changes, or "first login" creation of those
[12:34] <BigWhale> pitti, ah, thanks!
[12:41] <pitti> didrocks: would it be okay for me to upload an unity-2d to drop the invalid "use-strut" from /usr/share/GConf/gsettings/unity-2d.convert ?
[12:42] <didrocks> pitti: sure, you need to push the same change back to lp:unity-2d as well though
[12:42] <pitti> didrocks: hm, can't push to the branches
[12:42] <pitti> Vcs-Bzr: http://bazaar.launchpad.net/~unity-2d-team/unity-2d/oneiric
[12:42] <didrocks> pitti: is it invalid? I didn't follow the story
[12:42] <pitti> that seems out of date, and a /precise doesn't exist
[12:42] <didrocks> pitti: yeah, it's just lp:unity-2d
[12:43] <pitti> didrocks: yes, it makes gsettings-data-convert crash and thus breaks the settings migrations for every upgrader
[12:43] <pitti> didrocks: ah, no separate upstream/packaging branches?
[12:43] <didrocks> pitti: upstream really wanted the packaging to be in upstream trunk
[12:43] <pitti> didrocks: normal MP, or what is the procedure these days?
[12:43] <didrocks> and that was on the armel time (when the armel team maintained it)
[12:43] <didrocks> pitti: yeah, normal MP
[12:44] <pitti> will also fix Vcs-Bzr: while I'm at it
[12:44] <didrocks> pitti: thanks :)
[12:44] <didrocks> IIRC, use-strut was for the autohide mode
[12:44]  * didrocks checks the schema
[12:46] <didrocks> hum, that's weird
[12:46] <didrocks> pitti: wait, pinging greyback
[12:46] <didrocks> ah, not online right now :/
[12:46] <pitti> ah, ok; committed locally for now
[12:47] <didrocks> those keys should still be there, hence my surprise
[12:47] <pitti> grepping for "strut" there is not much
[12:47] <didrocks> yeah
[12:47] <pitti> no querying from gsettings, anyway, so it doesn't look like an omission in the schema
[12:47] <didrocks> pitti: it's not, ther is stil super-key-enable
[12:47] <pitti> didrocks: r804
[12:47] <didrocks> pitti: but hide-mode is not available as well
[12:48] <pitti> This branch removes the use-struts dconf option and enables struts automatically when the user selects hide-mode 0 (never hide).
[12:48] <didrocks> pitti: do you have "hide-mode" in the schema?
[12:48] <pitti> yes
[12:48] <didrocks> urgh, dconf-editor doesn't show it here
[12:48]  * didrocks pokes gsettings directly
[12:49] <didrocks> pitti: ok, got it there :)
[12:49] <pitti> r804 doesn't look like an accident, anyway
[12:49] <didrocks> yeah, makes sense, I was afraid about not seeing "hide-mode" anymore as well
[12:49] <didrocks> but seems a dconf-editor issue
[12:50] <didrocks> so yeah, just MP please :)
[12:51] <pitti> https://code.launchpad.net/~pitti/unity-2d/fix-gsettings-migration/+merge/89882
[12:53] <pitti> didrocks: I suppose I shouldn't upload before it is merged
[12:53] <didrocks> pitti: if this can wait, yes :)
[12:54] <didrocks> if not, please go ahead, I'll erase in next upload
[12:54] <pitti> a day or two is certainly okay, it's just a major regression since december
[12:54] <pitti> fortunately we have test cases for this since yesterday :)
[12:55] <nessita> hello everyone! I saw Chris email re: X updates... and I see all the updates in apt. Is it safe to upgrade?
[12:56] <pitti> nessita: I've been running with those packages for a week or so, no trouble  here
[12:56] <didrocks> pitti: heh, excellent! will be no thursday 2nd february for next release :)
[12:56] <pitti> Thinkpad X201/intel arrandale
[12:56] <pitti> didrocks: ugh, 2nd Febryary?
[12:57] <pitti> that's still quite some time
[12:57] <pitti> it's also the day of soft-freeze for alpha-2
[12:58] <pitti> well, /me bows to the gods of workflows :)
[12:59] <didrocks> pitti: yeah, post alpha2
[12:59] <pitti> didrocks: can we at least get this into alpha-2?
[12:59] <pitti> to avoid breaking the migration for every upgrader?
[12:59] <didrocks> pitti: the 2 february is the alpha2, isn't it?
[12:59] <didrocks> pitti: so please go ahead and upload
[13:00] <didrocks> pitti: next release will only be post alpha2 (freeze on monday)
[13:00] <pitti> ok, doing that then; I don't care much about preserving the changelog
[13:00] <pitti> didrocks: will that cause you lots of trouble?
[13:00]  * pitti is torn back and forth between pissing off you or upgraders
[13:01] <didrocks> pitti: no no, don't worry, maybe I won't have the changelog in the next upload but that's it :)
[13:02] <pitti> yeah, I don't care much
[13:03] <pitti> seb128: ^ once I locally fix the unity-2d conversion schema on the live system, it stops crashing; so shotwell seems to be a corner case only, or already fixed?
=== MacSlow|lunch is now known as MacSlow
[13:08] <pitti> didrocks: hm, how do you build a source package from these branches these days? use --commit?
[13:09] <didrocks> pitti: with source 3 arguing at you for having distro-patch, right?
[13:09] <seb128> pitti, not sure I understand the shotwell issue, the key they list in the .convert is in one of the .schemas, that case might be fixed by the gconf commit I showed earlier
[13:09] <pitti> didrocks: yes
[13:09] <pitti> seb128: oh, is that already 8in?
[13:09] <didrocks> pitti: remove source 3, I didn't for unity-2d yet but as it doesn't play well with merge-upstream…
[13:09] <seb128> pitti, no, still on my list for today
[13:09] <didrocks> pitti: I did for others, especially thinking about you with SRU :)
[13:09] <pitti> seb128: so it's not that
[13:09] <pitti> didrocks: ok, will do that
[13:10] <seb128> pitti, well, it doesn't happen every time for sure, I don't know what conditions trigger it, it's not as obvious as the unity-2d case where an invalid key was listed
=== greyback|lunch is now known as greyback
[13:27] <jbicha> good morning
[13:27] <jbicha> seb128: could we get librsvg updated to 2.35.1 to fix bug 920796?
[13:27] <ubot2`> Launchpad bug 920796 in librsvg "gcompris FTBFS: 'RsvgSizeFunc' is deprecated [-Werror=deprecated-declarations]" [Medium,Fix released] https://launchpad.net/bugs/920796
[13:28] <pitti> hey jbicha
[13:28] <seb128> hey jbicha
[13:29] <seb128> jbicha, guess we could, evfool did the update to .0 since .1 has an abi break which got revert in git after .1 but they didn't roll .2 yet
[13:29] <seb128> jbicha, so if we do we need to backport extra fixes as well, I was going to wait for .2
[13:29] <seb128> see bug #918016
[13:29] <ubot2`> Launchpad bug 918016 in librsvg "Request to update to new upstream version 2.35.0" [Wishlist,Fix released] https://launchpad.net/bugs/918016
[13:31] <seb128> jbicha, do you want to do the update? or should I add it to my list?
[13:31] <jbicha> seb128: oh ok, I don't have upload permissions for it
[13:32] <seb128> jbicha, I can sponsor it if you want ;-)
[13:34] <jbicha> seb128: ok I'll work on it today :)
[13:34] <seb128> jbicha, thanks
=== m_conley_away is now known as m_conley
[14:26] <arand> Which resolutions for icons does Unity use by default? Is there any benefit to installing x64 and x128 versions if they are just resized versions of a x256 image?
[14:27] <pitti> arand: for most icons, it shouldn't matter; our standard icons are SVG anyway (and thus scale well to any size), and the launcher uses someting in the range of 48 to 64
[14:28] <seb128> arand, you can try asking #ubuntu-unity if you want details on what unity uses in the launcher and dash
[14:28] <arand> I already did so and got no responase :/
[14:28] <seb128> desrt, hey
[14:29] <desrt> hey
[14:29] <seb128> arand, when? maybe it was a bad time and you should try again?
[14:29] <seb128> desrt, some bugs,questions for you about the new appmenu show,hide xsettings based
[14:29] <arand> About an hour ago.
[14:29] <seb128> desrt, if you have a few minutes
[14:30] <desrt> seb128: for you, i have _hours_
[14:30] <desrt> :)
[14:30] <seb128> ;-)
[14:30] <seb128> desrt, ok, so first some people uninstall appmenu-gtk to get ride of the appmenu in oneiric
[14:31] <seb128> desrt, interesting side effect is that they get the xsettings set because unity-panel-service is running but the menus not exported...
[14:31] <seb128> desrt, in precise
[14:31] <seb128> where should that be fixed? (trying to figure where to send the bug)
[14:31] <seb128> in the gtk patch?
[14:31] <desrt> it should be fixed by telling them to stop running the appmenu indicator
[14:32] <seb128> ok, so uninstalling indicator-appmenu as well I guess
[14:32] <desrt> well
[14:32] <seb128> or installing back appmenu-gtk rather
[14:32] <arand> pitti: So there's no benefit to installing smaller icons unless they are actually made for that format? the gtk-icon-cache will generate and store appropriate resizes anyways (so there's no performance benefit)?
[14:32] <desrt> isn't the UBUNTU_MENUPROXDY env var only set by the xsession scripts in the case that appmenu-gtk is installed?
[14:32] <desrt> 80appmenu:	export UBUNTU_MENUPROXY="libappmenu.so"
[14:32] <desrt> 80appmenu-gtk3:	export UBUNTU_MENUPROXY="libappmenu.so"
[14:33] <seb128> desrt, right, thing is that if you uninstall the gtk3 version only you still get the variable
[14:33] <desrt> nice house of cards here
[14:33] <seb128> which is sort of a mixmatch broken config
[14:33] <chrisccoulson> seb128, have you also seen the bug where people are running nautilus with sudo and not getting a menu at all? ;)
[14:33] <desrt> FML.
[14:33] <seb128> chrisccoulson, shush, that's my second one
[14:33] <chrisccoulson> lol
[14:33] <seb128> chrisccoulson, you are spoling my question!
[14:33] <kenvandine> chrisccoulson, that makes sense, no env variable set
[14:33] <seb128> ok
[14:34] <kenvandine> :)
[14:34] <chrisccoulson> kenvandine, right :)
[14:34] <kenvandine> annoying though
[14:34] <seb128> desrt, so second is what chrisccoulson mentioned
[14:34] <seb128> not specific to nautilus
[14:34] <seb128> sudo gedit is the same
[14:34] <desrt> can we just kill our stupid users?
[14:34] <chrisccoulson> YES!
[14:34]  * kenvandine thinks we need a better way than setting env
[14:34] <desrt> let's have the same conversation about what happens if you do X forwarding
[14:34] <seb128> gsettings key?
[14:34] <desrt> i bet the answer is sort of the same!
[14:34] <kenvandine> desrt, if it wasn't for users, our job would be AWESOME :)
[14:35] <kenvandine> seb128, that would be better i think
[14:35] <pitti> arand: I don't think gtk-icon-cache caches pre-rendered icons
[14:35] <chrisccoulson> desrt, we need a "if (getuid() == 0) exit(-EDUMBASS)" in gtk
[14:36] <seb128> kenvandine, I don't think the issue is the environment, sudo doesn't strip it, does it?
[14:36] <seb128> $ sudo sh -c "echo $UBUNTU_MENUPROXY"
[14:36] <seb128> libappmenu.so
[14:36] <kenvandine> but we would need to disable them if libappmenu isn't loaded or the setting was disabled
[14:36] <kenvandine> seb128, how about gtksu ?
[14:36] <kenvandine> i think it might
[14:36] <seb128> desrt has a point that Xorg forwarding will hit similar issues
[14:37] <kenvandine> sudo from a terminal i think is ok
[14:37] <seb128> kenvandine, well first, should stuff running with sudo have their menus exported or not?
[14:37] <desrt> i wonder if xsettings was the wrong approach
[14:37] <seb128> kenvandine, no it's not
[14:37] <kenvandine> but "Run as" kind of thing is broken
[14:37] <kenvandine> ok
[14:37] <seb128> kenvandine, sudo gedit leads to the bug since the xsettings is set for the session
[14:37] <seb128> without consideration for the user
[14:38] <seb128> kenvandine, the menu showing in precise is done based on the xsettings
[14:38] <kenvandine> ok
[14:39] <desrt> seb128: for the first complaint my answer is that you need to educate your users (which will probably happen automatically on various forums)
[14:39] <desrt> seb128: for the second complaint my answer is... uhm... i dunno
[14:39] <seb128> desrt, thrid one, is that gnotravex stopped having is gmenu showing in unity appmenu where that was working during the rally ... do you know why?
[14:39] <desrt> no.  i find that most curious.
[14:39] <seb128> :-(
[14:40] <desrt> oh.  yes.
[14:40] <desrt> it makes sense, actually
[14:40] <seb128> oh?
[14:40] <desrt> so gtk checks the xsettings property to determine if it should show the menu locally
[14:40] <desrt> it sees that shell-shows-menu is set and decides the answer is "no"
[14:40] <desrt> which means that the gtk module has nothing to rip out
[14:41] <seb128> ok, makes sense
[14:41] <desrt> i think it may have been a mistake to use the same xsettings for two very similar but not-quite-the-same meanings
[14:41] <seb128> desrt, you still plan to make appmenu consume gmenus this cycle? i.e to fix that?
[14:42] <seb128> desrt, I don't care much about 1, but I'm a bit nervous about the sudo, fwd and gmenu on unity stuff for the lts ;-)
[14:42] <desrt> i wasn't planning on it, no
[14:42] <desrt> but it seems like i may have to
[14:43] <seb128> desrt, so you think the current xsettings are not a good fit after all?
[14:43] <desrt> the trouble with any dbus-based scheme for knowing when to have the menus is that we get into that silly dancing-menu situation
[14:43] <desrt> xsettings are solid and there right from the start.  nothing to wait for.
[14:44] <jbicha> epiphany also uses the new appmenu but I was leaving that in the gnome3 ppa until webkit migrates to precise
[14:44] <jbicha> I suspect several of the games will be using the new menu soon
[14:44] <desrt> jbicha: several already are
[14:44] <seb128> jbicha, right, gnotravex is my testcase in precise
[14:44] <desrt> seb128: for the record, i consider the gnotravex case to be the only truly worrying one
[14:45] <seb128> desrt, doing sudo gedit and having no menu is a bit annoying as well, people do need to edit system files ;-)
[14:45] <desrt> seb128: i'm with chrisccoulson on this point
[14:45] <desrt> g_assert (getuid() != 0);
[14:45] <seb128> what do you recommend?
[14:45] <seb128> how do you tell users to edit system files?
[14:45] <desrt> nano?
[14:45] <chrisccoulson> vim!
[14:45] <seb128> ...
[14:45] <chrisccoulson> :-)
[14:46] <seb128> not very constructive guys :p
[14:46] <seb128> I agree the nautilus case is on crack
[14:46] <seb128> but I disagree the "edit files in etc using gedit" is
[14:46] <desrt> it's equally bad
[14:46] <desrt> that said, i don't think that this particular way of failing is really acceptable
[14:47] <desrt> again -- i'd prefer if it exploded in your face immediately
[14:47] <desrt> but i doubt you'll want to do that
[14:47] <jbicha> we should just crash when we fail! ;)
[14:48] <desrt> jbicha: GSettings-style software engineering practices :)
[14:48] <kenvandine> grrr
[14:49] <kenvandine> i hate that :)
[14:49] <ricotz> i know
[14:49] <desrt> what we need is a reliable way to determine if we are in the proper session
[14:49] <desrt> ie: detect these weird sudo/x11-fwd cases
[14:49] <desrt> interesting story
[14:49] <seb128> yeah
[14:49] <desrt> i was talking to christian hammond (chipx86) over the weekend
[14:49] <desrt> and he said that your gtk patching makes his life absolutely miserable
[14:50] <desrt> it interacts quite badly with vmware's native-window thing
[14:50] <desrt> because it rips the menu out of the window
[14:50] <desrt> but then vmware puts the window on the windows machine
[14:50] <seb128> desrt, I'm not even sure why the sudo case is not working, would it make sense to use the current user session bus for sudo as well?
[14:50] <desrt> and the menu is nowhere to be found
[14:50] <desrt> so this is another related issue...
[14:50] <desrt> seb128: that's a really complicated question that you're not likely to get a satisfactory answer to
[14:51] <seb128> ok, not the way to go then
[14:51] <desrt> seb128: mostly because any person who is qualified to answer that you ask it to will tell you "don't do that"
[14:52] <seb128> desrt, I guess the sudo issue will happen the way under gnome-shell right? like if you run gnotravex it will get no menu right?
[14:52] <desrt> quite odd
[14:52] <desrt> DISPLAY makes it into sudo
[14:52] <desrt> but DBUS_SESSION_BUS_ADDRESS does not
[14:52] <desrt> seb128: sudo gnotravex? :)
[14:52] <seb128> desrt, yes, as a testcase :p
[14:53] <desrt> ya.  i bet it's broken.
[14:53] <desrt> "don't do that"
[14:53] <pitti> desrt: DISPLAY is one of the few whitelisted sudo env vars which are kept
[14:53] <desrt> pitti: can we turn that off? :)
[14:53] <pitti> ./plugins/sudoers/env.c, initial_keepenv_table[]
[14:53] <pitti> desrt: err, no?
[14:53] <desrt> pitti: of all people i know that i don't need to have the "sudo [x application] is bad" conversation with you :)
[14:54] <pitti> desrt: well, you can turn it off for a particular call, or in /etc/sudoers, but we use it for $deity knows many things, and so do users
[14:54] <pitti> desrt: well, it's not that I like them, but people do use them..
[14:54] <desrt> right.  fair enough.
[14:54] <jbicha> oh wow, this hud things brings back memories of Office XP's personalized menus
[14:54] <desrt> so it's pretty simple
[14:55] <desrt> we need a robust mechanism for determining if the connecting client is in the "native context" of the x server
[14:55] <pitti> desrt: there was an entire project and library (libgksu/gksu/gksudo) written around it, so apparently it's useful
[14:55] <desrt> which in this case means "can connect to the same session bus"
[14:55] <pitti> desrt: although in the past it was a lot more important
[14:55] <pitti> desrt: like, "gksu synaptics" in a .desktop file
[14:55] <pitti> these days we have polkit and root-running d-bus backends and all that
[14:55] <pitti> we aren't quite at zero yet, but very close
[14:55] <desrt> pitti: indeed... that's what i was getting at
[14:56] <desrt> all of this will soon be an unpleasant memory
[14:56] <pitti> not that pkexec would be any better, though :)
[14:56] <pitti> (if you use it to launch X-ish stuff)
[14:56] <desrt> so i have a thought
[14:56] <desrt> unix:abstract=/tmp/dbus-JZ8XIKEi1E,guid=80949262084dbaf430eea21d00000011
[14:56] <desrt> instead of storing the shell-shows-app-menu property as a bool
[14:57] <desrt> we could store the dbus guid
[14:57] <desrt> and we only consider it valid if we have the same guid on our own session bus
[14:57] <pitti> desrt: in fact, I think I (well, apport) is one of the last bad guys, aside from checkbox-gtk and update-notifier
[14:57] <desrt> otherwise we ignore it, because obviously it won't be working anyway
[14:57] <desrt> that will catch every broken case
[14:58] <desrt> this may well break in a systemd future, though
[14:59] <pitti> jbicha: do you know whether freerdp will be uploaded to Debian soon?
[14:59] <pitti> jbicha: (you committed 1.0 to git, but wasn't uploaded yet)
[14:59] <seb128> desrt, well let's do it in a way which works in the present and adapt it to what is available in the futur when that happens
[14:59] <desrt> seb128: i want to make sure that it continues to be possible to do this without synchronous dbus calls
[15:00] <seb128> desrt, well I guess at this point it's better to switch to a bug or list discussion
[15:01] <desrt> interesting
[15:01] <desrt> gdbusconnection has a "guid" property
[15:03] <jbicha> pitti: the issue is that someone else proposed different packaging for freerdp and I'm not sure what of that needs to be added to git
[15:03] <jbicha> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=647498
[15:03] <ubot2`> Debian bug 647498 in freerdp "freerdp: Consider packaging git snapshot?" [Wishlist,Open]
[15:04] <pitti> jbicha: ah, thanks; I'll have a look ASAP
[15:04] <jbicha> pitti: thanks
[15:04] <pitti> jbicha: probably tomorrow, though, need to run in about 20 mins
[15:05] <seb128> pitti, will you be back for the meeting if there is one?
[15:05] <seb128> pitti, or do you want me to handle it?
[15:06] <seb128> pitti, well "it" the "ping people and see if we do a meeting" ;-)
[15:06] <seb128> pitti, ignore the question I just read your reminder email after the first line :p
[15:06] <didrocks> seb128: what, there is content in the email reminder?
[15:07] <didrocks> it's just not a link to click on it? :p
[15:07] <seb128> didrocks, :-p
[15:07] <pitti> seb128: would be nice, thanks
[15:07] <seb128> didrocks, I learnt my lesson ;-)
[15:07] <seb128> pitti, will do
[15:07] <pitti> seb128: but I don't think we'll have one, no items until now
[15:08] <seb128> pitti, right, it's likely we will not, still I can do a round of ping and see if any topic comes
[15:08] <seb128> pitti, I know nessita want to make a call for endorsements for her per package upload application wikipage
[15:08] <nessita> yes I do!
[15:08] <seb128> ;-)
[15:08] <nessita> shall I drop the request now then?
[15:09] <seb128> nessita, you can mention it yes ;-)
[15:09] <seb128> doesn't hurt
[15:09] <seb128> we can mention it again at meeting time as well
[15:10] <nessita> hello everybody! If you can/want to invest some minutes, I would appreciate some endorsement for my per package upload application: https://wiki.ubuntu.com/NataliaBidart/PerPackageUploadApplication
[15:10] <nessita> and I will also welcome any suggestion/improvement that you consider necessary
[15:19] <seb128> ok, I'm out for half an hour or so, will be back for the meeting time for sure ;-)
[15:21] <didrocks> hey cyphermox
[15:21] <didrocks> cyphermox: seems I can't activate the bluetooth anymore
[15:21] <cyphermox> hey didrocks
[15:22] <didrocks> are you aware about such issues?
[15:22] <cyphermox> how's it look?
[15:22] <cyphermox> there's tons of issues with bluetooth ;_;
[15:22] <didrocks> like, the menu just shows "bluetooth active/disable settings"
[15:22] <didrocks> not the long menu like before
[15:22] <cyphermox> didrocks: are you using a usb bluetooth dongle or integrated?
[15:22] <didrocks> and if I start gnome-control-center
[15:22] <didrocks> cyphermox: it's integrated
[15:22] <cyphermox> didrocks: problem would be that bluetoothd isn't running for you
[15:23] <didrocks> I can't enabled it
[15:23] <didrocks> ah
[15:23] <didrocks> should I launch it?
[15:23] <cyphermox> you can, and it's going to solve the issue for you
[15:24] <cyphermox> I'm still working on figuring out why the hell it won't start on boot like it was so clearly working so well before
[15:24] <pitti> good night everyone! need to run now
[15:24] <didrocks> have a good night pitti
[15:25] <didrocks> cyphermox: hum, it exits immediatly
[15:25] <cyphermox> it does?
[15:25] <didrocks> there is no verbose mode
[15:25] <didrocks> yeah
[15:25] <cyphermox> no, it daemonizes
[15:25] <didrocks> exit 0!
[15:25] <cyphermox> ps -ef | grep bluetoothd
[15:25] <cyphermox> ?
[15:25] <didrocks> cyphermox: nothing
[15:26] <cyphermox> f you want to start it with verbose, use  /usr/sbin/bluetoothd -d -n
[15:26] <didrocks> cyphermox: oh, I got it
[15:26] <didrocks> disabling it in the indicator
[15:26] <didrocks> and enabling it again
[15:26] <didrocks> and now, I have a daemon
[15:26] <cyphermox> why was it not starting? any idea?
[15:27] <didrocks> reallly not :/
[15:27] <cyphermox> the indicator can't do anything if the daemon isn't there :/
[15:27] <didrocks> seems it thinked it was already there
[15:27] <cyphermox> ok
[15:27] <didrocks> cyphermox: there is three items in the indicator, isn't it?
[15:27] <didrocks> enable
[15:27] <didrocks> disable
[15:27] <didrocks> settings
[15:27] <didrocks> enable was already "on"
[15:28] <didrocks> not sure what I can provide you as debug info next time
[15:28] <seb128> the indicator is buggy though, don't trust it, trust rather the control center
[15:28] <didrocks> seb128: well, control center didn't let me enabling it
[15:28] <seb128> like the indicator indicates "bluetooth: on" on the 10v which has no bluetooth hardware
[15:28] <didrocks> it told just it's dsabled
[15:28] <cyphermox> didrocks: there should be more like 7-8 items in the indicator really
[15:28] <didrocks> cyphermox: now there is more
[15:28] <didrocks> but before, when it wasn't working, just those 3
[15:28] <cyphermox> yeah
[15:29] <cyphermox> can't know what bluetooth you have until the daemon is running
[15:30] <cyphermox> brb, upstart debugging
[16:34] <seb128> re
[16:34] <seb128> sorry took a bit longer that I though (roads are busy at 5pm!)
[16:34] <seb128> didrocks, chrisccoulson, mterry, kenvandine, cyphermox, tkamppeter: it's meeting time
[16:34] <kenvandine> hey
[16:34] <seb128> if you have anything to discuss
[16:34] <kenvandine> not really... posted on the wiki :)
[16:35] <didrocks> hey
[16:35] <seb128> or any announcement
[16:35] <didrocks> nothing here
[16:35] <seb128> nessita, ^ ;-)
[16:35] <mterry> Heyo
[16:35] <nessita> I have a request, is possible
[16:35] <nessita> if*
[16:35] <nessita> If anyone of you can/want to invest some minutes, I would appreciate some endorsement for my per package upload application: https://wiki.ubuntu.com/NataliaBidart/PerPackageUploadApplication
[16:36] <kenvandine> nessita, awesome, sure!
[16:36] <nessita> kenvandine: thanks! :-D
[16:37] <seb128> ok, nothing special to discuss, no meeting ;-)
[16:37] <seb128> thanks everybody
[16:37] <nessita> thank you!
[16:38] <mterry> :)
=== davidcalle_ is now known as davidcalle
=== bdrung_ is now known as bdrung
=== dpm is now known as dpm-afk
[18:02] <asac_> tedg: what do you think about: http://paste.ubuntu.com/815578/ ... its about allowing to preventing set up guessing magic based on seen_db_seen
[18:03]  * asac_ off on a call
[18:03] <desrt> seb128: https://mail.gnome.org/archives/release-team/2012-January/msg00029.html
[18:04] <desrt> (nb: this list is "read only" for non-release-team members, iirc)
[18:05] <seb128> desrt, thanks ;-)
[18:05] <desrt> seb128: we're not *actively* trying to piss you guys off :)
[18:07] <cawel> Can I ask here a bug I am having with Ubuntu 11.10 regarding wireless ?
[18:09] <jp_wanN> cawel: you should go to ubuntu- [Your language code] chatroom, ubuntu-de e.g.
[18:10] <cawel> jp_wanN: thanks, will do.
[18:10] <jp_wanN> cawel: no problem ;)
[18:10] <cawel> seems like #ubuntu-en is invite-only :(
[18:10] <desrt> #ubuntu-en redirects to #ubuntu
[18:10] <desrt> some irc clients have trouble with redirects
[18:10] <desrt> so you can just join #ubuntu
[18:12] <kenvandine> didrocks, mind if i drop the source format for unity-2d to 1.0?
[18:12] <kenvandine> i need to do a rebuild for libindicator abi bump
[18:23] <didrocks> kenvandine: pitti's version already dropped it, isnt it?
[18:24] <kenvandine> oh... maybe not in bzr?
[18:24]  * kenvandine checks branches
[18:24] <didrocks> kenvandine: no, it's not in bzr
[18:24] <didrocks> kenvandine: don't use the bzr branch for unity-2d as it's in source
[18:24] <kenvandine> okie dokie
[18:28] <BigWhale> Python XDG has no calls for retrieving default folders? :(
[18:29] <kenvandine> yes it does
[18:29] <kenvandine> it is just a little odd :)
[18:30] <kenvandine> BigWhale, look in gwibber for an example
[18:30] <asac_> tedg: http://people.linaro.org/~asac/tmp/desktop-webmail-indicator.png ... thats what I mean with manual :)
[18:30] <kenvandine> BigWhale, gwibber/microblog/util/const.py
[18:31]  * kenvandine goes to lunch while pbuilder thrashes :)
[18:33] <BigWhale> kenvandine, oh... no those I found. I need ~/Videos and ~/Documents :)
=== m_conley is now known as m_conley_away
[19:17] <tedg> asac_, I'm game on the feature, I'm just curious how one would set it then.
[19:17] <tedg> asac_, Do you mean by using a new desktop file?
=== m_conley_away is now known as m_conley
[19:48] <dobey> anyone using synergy on precise?
[19:50] <asac_> tedg: yes... my desktop-webmail.desktop looks like this:
[19:50] <asac_> X-Ayatana-Indicator-Setup-Type=Manual
[19:51] <asac_> in [Desktop Entry]
[20:00] <tedg> asac_, I was expecting another line or two... did you get cut off?
[20:12] <asac_> tedg: 20:50 < asac_> tedg: yes... my desktop-webmail.desktop looks like this:
[20:12] <asac_> 20:51 < asac_> X-Ayatana-Indicator-Setup-Type=Manual
[20:12] <asac_> 20:51 < asac_> in [Desktop Entry]
[20:13] <asac_> tedg: thats the complete one: http://paste.ubuntu.com/815725/
[20:15] <tedg> asac_, So then how do you mark it setup?  Or wait, you're just setting it so that it NEVER has the setup?
[20:16] <asac_> tedg: right :)
[20:16] <asac_> tedg: well. manual means: app is always correct or does it itself
[20:16] <asac_> or something
[20:16] <asac_> X-Ayatana-Indicator-Setup-Type=Manual -> manual mode avoids any heuristic and displays the shortcut group right away
[20:33] <jbicha> seb128: I posted the debdiff for librsvg at bug 920796
[20:33] <ubot2`> Launchpad bug 920796 in librsvg "gcompris FTBFS: 'RsvgSizeFunc' is deprecated [-Werror=deprecated-declarations]" [Medium,Fix released] https://launchpad.net/bugs/920796
[20:48] <seb128> jbicha, thanks
[21:12] <kenvandine> grrr
[21:13] <kenvandine> seb128, unity is the last FTBFS i have with the rebuild for libindicator
[21:13] <kenvandine> it is failing because of deprecations
[21:13] <kenvandine> how can i make cmake not eat the CXXFLAGS?
[21:14] <seb128> kenvandine, can't you just drop -Werror from the CMakeList?
[21:15] <kenvandine> i could probably do that... but i would rather work around it in the rules file
[21:15] <seb128> good luck then
[21:15] <kenvandine> i'll do that for now...
[21:15] <seb128> I just patched the CMakeList when I did an upload for unity
[21:15] <kenvandine> this libindicator abi break has been a pita!
[21:16] <kenvandine> a bunch of FTBFS for deprecations
[21:16] <kenvandine> we need to get those fixed
[21:16]  * kenvandine will yell at tedg about that :)
[21:16] <seb128> kenvandine, before uploading make sure you don't break stuff like the hud ppa by jumping versions ahead, or coordinate with ted,gord,didrocks
[21:17] <kenvandine> i can bump the hud version
[21:17] <seb128> kenvandine, if tedg was doing review and rolling tarballs yeah, we are mid-cycle and we didn't get any indicator update yet
[21:18] <tedg> kenvandine, Uhg, I forgot about the GTK3 deprecations.
[21:19] <kenvandine> tedg, don't worry, i'll be getting you a list of the failures :)
[21:19] <kenvandine> libappindicator fails too with the python bindings
[21:19] <kenvandine> i have a branch to fix that
[21:19] <tedg> Oh, didn't expect that one.
[21:19] <tedg> Can we drop those?
[21:20] <tedg> Or do we need them for the GTK2 case?
[21:20] <kenvandine> i say leave them for now
[21:20] <kenvandine> for gtk2
[21:21] <seb128> kenvandine, is there any good fixes coming in the indicators at least? ;-)
[21:22] <kenvandine> nope
[21:22] <seb128> :-(
[21:22] <kenvandine> indicator-appmenu with the hud service
[21:22] <kenvandine> but first just a rebuild for libindicator
[21:35] <seb128> rodrigo_, hey, there?
[21:39] <rodrigo_> seb128, for a minute, but yes, what's up?
[21:39] <seb128> rodrigo_, hey
[21:39] <rodrigo_> seb128, how are you?
[21:39] <seb128> rodrigo_, I'm good thanks, how are you?
[21:39] <seb128> rodrigo_, http://bazaar.launchpad.net/~ubuntu-desktop/gnome-settings-daemon/ubuntu/revision/230
[21:40]  * rodrigo_ looks
[21:40] <seb128> rodrigo_, do you remember why you did the ca_gtk_play_for_widget() change in 16_use_synchronous_notifications.patch
[21:40] <seb128> using "manager->priv->dialog" rather than "window" break again the sound playing in the notifications
[21:40] <rodrigo_> seb128, oh, iirc it's happened several times
[21:40] <seb128> I was wondering if you did it for a reason
[21:41] <rodrigo_> on that patch, window is set at the beginning of the function, iirc
[21:41] <rodrigo_> so was just a mistake, just making it apply
[21:41] <seb128> rodrigo_, ok, great, that's what I though but I wanted to check
[21:41] <seb128> I reverted that line and that makes the sound work again
[21:41] <rodrigo_> I think it should be 'ca_gtk_play_for_widget (window,
[21:41] <seb128> but I wanted to check if you did it for a reason
[21:41] <seb128> rodrigo_, right, that's what I did
[21:42] <rodrigo_> yeah, it's the 2nd or 3rd time that comes up, sorry :)
[21:42] <chrisccoulson> hi rodrigo_
[21:42] <chrisccoulson> how are you?
[21:42] <rodrigo_> hi chrisccoulson
[21:42] <seb128> rodrigo_, the code is http://pastebin.ubuntu.com/815857/
[21:42] <rodrigo_> chrisccoulson, I'm fine, and you?
[21:42] <chrisccoulson> rodrigo_, yeah, i'm not too bad thanks :)
[21:42] <rodrigo_> seb128, yeah, that's it
[21:42] <seb128> rodrigo_, when you have time another day I need to ping you about the region capplet stuff
[21:42] <seb128> rodrigo_, do you still plan to land that upstream this cycle?
[21:42] <rodrigo_> seb128, yes, please do whenever you want
[21:43] <chrisccoulson> hah, oh, i recognise that hacky bit of code there ;)
[21:43] <rodrigo_> I plan to, just a bit busy on the new job, but please ping me whenever you want to discuss it, and I'll try to finish it asap
[21:43] <seb128> chrisccoulson, that's the bug I was about to ping you about when you started ignoring me :p
[21:43] <rodrigo_> :)
[21:43] <seb128> rodrigo_, ok, thanks
[21:43] <RAOF> Morning chrisccoulson!
[21:43] <rodrigo_> seb128, when's the gnome freeze?
[21:43] <chrisccoulson> hi RAOF
[21:44] <seb128> rodrigo_, I didn't want to stop you on IRC late, thanks for the reply
[21:44] <rodrigo_> seb128, you were lucky, just came for a minute to check mail :)
[21:44] <seb128> rodrigo_, let me check, in a month or so
[21:44] <rodrigo_> seb128, ok, I'll try to work on it some this weekend
[21:44] <jbicha> rodrigo_: wiki says 20 Feb
[21:44] <rodrigo_> ok
[21:44]  * rodrigo_ adds it to his calendar
[21:45] <seb128> rodrigo_, feb 20.
[21:45] <seb128> rodrigo_, thanks
[21:51] <seb128> jbicha, btw the webkit s-c issue has a patch on https://bugs.webkit.org/show_bug.cgi?id=76911
[21:51] <ubot2`> bugs.webkit.org bug 76911 in WebKit Gtk "[GTK] WebKitWebView won't work in a GtkOffscreenWindow" [Normal,Unconfirmed]
[21:51] <seb128> jbicha, if you want to update the ppa feel free, if that works fine we might be able to upload to precise this week
[21:52] <seb128> jbicha, I will add it on my list in any case, I'm just mentioning it in case you want to beat me to it ;-)
[22:01] <desrt> seb128: so some good news
[22:01] <desrt> seb128: gee issue is fixed now
[22:02] <desrt> although the patch to vala causes a regression in folks now
[22:02] <seb128> desrt, in libgee git? 0.6 serie as well?
[22:02] <desrt> seb128: well, it was a vala patch
[22:02] <jbicha> seb128: ok, I'm uploading the patched webkit now
[22:02] <desrt> so gee needs to do a new release that was built using the new vala
[22:02] <seb128> ok
[22:02] <seb128> desrt, is there any sign of a vala release? ;-)
[22:02] <desrt> seb128: the folks regression is a bug in folks itself
[22:03] <desrt> uncovered by the new gobject strictness stuff plus the new way the vala compiler deals with property overrides
[22:03] <desrt> they plan to fix it up in folks
[22:03] <desrt> i think we'll see a vala release quite soon
[22:03] <desrt> juergbi has been running around in the forrest lately
[22:03] <desrt> which is why we didn't have one already
[22:03] <seb128> jbicha, thanks
[22:04] <desrt> i think you'll probably see one in the following days
[22:04] <seb128> desrt, ok
[22:04] <seb128> desrt, I'm not in an hurry, I did distroworkaround glib
[22:04] <desrt> seb128: understood
[22:04] <seb128> but good to know it's being properly fixed upstream still ;-)
[22:04] <desrt> seb128: i think that workaround will not save you from the folks regression
[22:04] <seb128> desrt, right
=== m_conley is now known as m_conley_away
[22:04] <seb128> the workaround was to unblock the dx guys
[22:05] <seb128> not to "fix" anything which was buggy
[22:05] <desrt> so what's the deal with gee versions?
[22:05]  * desrt is terminally confused
[22:05] <seb128> what about those?
[22:05] <seb128> we are on 0.6
[22:05] <desrt> 0.6 is 1.0 and 0.8 is 0.8 or something?
[22:06] <seb128> seems GNOME was unsure if this cycle is going to see a 0.8 or stay on 0.6
[22:06] <seb128> desrt, oh, dunno, I assumed that 0.6 is current stable and 0.7 current unstable
[22:06] <seb128> not sure if they have a schedule for 0.8
[22:06] <desrt> there's some additional complication, i understand
[22:06] <desrt> i'm sure we'll see what happens
[22:07] <seb128> right
[22:07] <desrt> in any case, i think there was a new gee release made
[22:07] <desrt> and it was built using my patch to vala before the fixes were actually applied to vala git itself
[22:08] <seb128> oh, good to know
[22:08] <seb128> desrt, they rolled a tarball for each serie
[22:08] <desrt> it was a 0.7 though
[22:08] <seb128> 0.6 and 0.7
[22:08] <desrt> oh.  that's good
[22:08] <desrt> let me see if it was built with the new vala
[22:09] <desrt> yes.  it was.
[22:09] <desrt> if you package that, the warnings should go away
[22:09] <desrt> 0.6.4, i mean
[22:09] <seb128> desrt, excellent, thanks
[22:09] <desrt> would have been nice if he included some note about that in the news :p
[22:09] <seb128> kenvandine, ^ want to do the update some day when you are out of libindicat* transitions? ;-)
[22:09] <seb128> desrt, indeed
[22:09] <czajkowski> evening, running Precise and just ran an upgrade and got an unusal message http://pastebin.ubuntu.com/815871/  anyone see similar today ?
[22:10] <seb128> hey czajkowski
[22:10] <desrt> seb128: the test, btw, is to look at abstractset.c
[22:10] <desrt> and see how the "read-only-view" property is setup
[22:10] <desrt> if it uses g_object_class_override_property -> bad
[22:10] <czajkowski> seb128: evening
[22:10] <desrt> g_object_class_install_property -> good
[22:10] <seb128> desrt, ok, thanks
[22:12] <seb128> czajkowski, it seems minor, dunno if that happened on my box though, I use update-manager and don't watch stdout
[22:12] <seb128> czajkowski, try asking mvo tomorrow
[22:13] <czajkowski> seb128: grand will do, I have to watch the output like to see what yer doing :)
[22:13] <czajkowski> seb128: what tz is mvo on ?
[22:13] <seb128> czajkowski, european
[22:13] <seb128> gmt+1
[22:28] <seb128> 'night