[06:37] good morning [07:53] good morning [08:03] Good morning [08:03] hey pitti, how was your week-end? [08:03] pretty quiet, and relaxing [08:04] I started to watch Big Bang Theory [08:04] awexome! [08:05] pitti: hehe, yes it is! [08:06] pitti: Julie and I watched all the episods of the 3 seasons in less than 2 weeks! === almaisan-away is now known as al-maisan [08:06] heh [08:06] I'm at 1-16 now [08:06] didrocks: how was your's? [08:07] pitti: was nice, thanks. At least a week-end with some sun. So, we walk a little in Annecy (before the snow which should come today or tomorrow) [08:08] so I've heard -- finally winter [08:09] yeah :) [08:09] seb is working today? [08:12] kklimonda: I think so, he'll be thee in some minutes I guess [08:14] thanks :) [08:55] morning [08:55] o/ rodrigo_ [08:56] hi kklimonda [09:04] kklimonda, since you've merged the branch, can you remove it please (git push :wip/query-response) [09:04] git push origin :wip/query-response, sorry [09:05] rodrigo_: I'll use it to finish _execute_query part [09:18] rodrigo_, hey, I still can't install libgnome-media-dev in natty (with gnome3-builds PPA) [09:25] seb128, FYI libgcr is missing a dep on libgck for the gnome-keyring package in gnome3-build [09:26] seb128, hm, the binary has the dep, but not the -dev package [09:28] also, it is unfortunate that gnome-keyring-3 is not // installable with version2 [09:28] seb128: I've prepared an update for glibmm to 2.27.3 (requested a merge: https://code.launchpad.net/~kklimonda/glibmm2.4/packaging/+merge/41428) and also fixed the issues with atkmm1.6 you had and updated it to 2.22.1 which ships the missing license file. [09:28] that makes all apps linked on the -2 fail [09:33] seb128, also, I can't start the daemon: [09:33] $ gnome-keyring-daemon -s [09:33] ** Message: couldn't connect to control socket at: /tmp/keyring-Gtstrz/control: Connexion refusée [09:34] any idea what's going on? [09:37] Zdra, hey [09:37] Zdra, you should ask rodrigo when he joins [09:37] he worked on gnome-keyring3 [09:37] kklimonda, hey [09:37] kklimonda, ok great, I will sponsor that today [09:38] Zdra, theorically the libraries should be installable at the same time and you need only 1 gnome-keyring service runnin [09:38] you might need to restart your session with the new one though [09:38] bonjour seb128, ca va? [09:39] hey pitti [09:39] oui, et toi ? [09:39] seb128, I have no keyring service running :( [09:40] because the one from new package does not start [09:40] seb128, you uploaded that package in ppa, so I though you were the one to ask... I'll ping rodrigo then :) [09:41] (Hm, no, just can't ready ppa page, you did gnome-desktop3, not keyring) [09:41] sorry :) [09:41] no worry ;-) [09:41] I've not tried or reviewed gnome-keyring3 yet [09:41] so I can't really comment [09:43] seb128, hey, do we have a clutter policy? Any opposition to updating to the latest? [09:43] hey robert_ancell [09:43] robert_ancell, : clutter policy? [09:44] "try to stay close to debian"? [09:44] seb128, I guess I'm asking why we're running an older version, perhaps due to some dx stuff? [09:45] I think just because nobody updated it yet [09:45] robert_ancell: hey, I would say we can even be in sync with debian and drop our patch if they were for unity (all are posted upstream anyway) [09:45] robert_ancell, is the new unity world I don't think dx uses clutter [09:45] is -> in [09:46] didrocks, debian is 1.2, but the stable branch is 1.4 [09:46] seb128, yes, that's what I figured [09:46] robert_ancell, debian is frozen for their next stable [09:46] some updates go to experimental but they mostly lag behind [09:47] actually, you might be able to shine some light on this, why do we have packages like clutter-1.0 and gtk2.0 that go outside 1.0.x and 2.0.x? Why aren't they called clutter-1 and gtk2 or we have gtk2.0 and gtk2.2? [09:49] robert_ancell, because gtk 2.0 and 2.2 turned to be a stable serie [09:49] seems like historical reason at the time where the gtk2 was packaged [09:49] gtk2 and gtk3 would have worked the same [09:50] robert_ancell, btw do you plan to land some GNOME3 components in natty? [09:50] robert_ancell, seems the things you work on at the lest likely to depend on other component [09:50] games, calculator, ... ;-) [09:51] I'm packaging the new games now, that's why I have to update clutter [09:51] yeah sure, I'll push gcalctool into natty [09:52] libclutter-1.0-0: /usr/lib/libclutter-glx-1.0.so.0 [09:52] robert_ancell, seems the clutter version just follows the upstream naming [09:54] seb128, right [09:58] seb128: ah, I remember why I didn't upload gdu 2.32 [09:58] seb128: it doesn't add something significant, but it removes all the docs [09:58] I already attempted it a while ago, and it's in Debian's git [09:58] in that a bug or wanted? [09:59] it was explicitly disabled, since it causes some trouble apparently [09:59] i. e. dropping docs wasn't an unintended side effect [09:59] it's not really "wanted" as such, of course [10:03] pitti, was sort of issue do they have? I guess at some point we will need to update anyway [10:03] I don't know [10:03] need to ask David, but he's asleep right now [10:03] no hurry [10:04] pitti, did we backport http://git.gnome.org/browse/gnome-disk-utility/commit/?id=e3ae27c89d66c683b4a37d40df290d0a0cd0aed1? [10:05] I don't think so [10:05] I think it's the only commit interesting in 2.32 [10:05] well out of tons of translation updates and some gtk cleaning [10:05] seb128: I keep this on my todo list [10:05] pitti, thanks [10:05] I'll eventually upload 32 to natty [10:05] but it's in no hurry I think [10:06] no it's not, I was just wondering if there was a reason to stay behind when the update seems only that crash fix, translation updates and build cleanly with recent gtk versions [10:06] we are still on the "clean merges and catch up with versions" [10:18] good morning everyone [10:19] hey chrisccoulson, how are you? had a nice weekend? [10:20] hi pitti - yeah, i'm good thanks. we had a lot of family and friends around on saturday for my daughters birthday, so it was pretty hectic [10:20] did you have a good weekend? [10:20] hey chrisccoulson [10:20] hi seb128, how are you? [10:21] chrisccoulson: yes, pretty calm; my wife needs to learn for her exams, so I spent it between cleaning the flat, visiting my grandma, meeting some friends, and watching Big Bang Theory :) [10:21] chrisccoulson, I'm fine thanks [10:21] chrisccoulson, it seems you had a nice we ;-) [10:21] pitti - what exams does your wife need to sit? [10:22] seb128 - yeah, it wasn't too bad. had a lot of cleaning up to do on sunday though [10:22] chrisccoulson: final exams for her geoinformation/surveying studies [10:22] chrisccoulson: tomorrow is Photogrammetry and Laser Scanning [10:22] cool! it sounds intense ;) [10:27] 9hey chrisccoulson! Nice week-end for you it seems :) [10:27] hey didrocks - yeah, pretty busy though [10:27] how about you? [10:29] chrisccoulson: nice and sunny. Didn't see a real sunny day for a long time before (maybe Florida was the last time? :)) [10:30] didrocks, yeah, that's the same here too! [10:31] seb128 - do you know what we are doing with yelp this cycle? are we going to stick with the current (gecko-based) version, or take the new version instead? [10:31] chrisccoulson, not sure yet [10:31] but seems we lean toward getting the new one [10:31] it's in GNOME3 ppa [10:32] robert_ancell is working on it [10:32] it need webkit gtk3 though [10:32] seb128 - ah, ok. i wasn't sure whether to spend time porting the current version to gecko 2.0 or not (i want to get everything in main ported this week really) [10:32] which is fine out of the fact that I'm not sure pitti will like 2 webkit on the CD [10:32] yeah, that's extreme ;) [10:32] what else uses webkit? [10:33] seb128: well, it's not about me personally -- we are currently 30 MB oversized, and everything we add needs to be compensated [10:33] so, two webkits really sound like a bad idea indeed [10:34] pitti, it was just a way to point the CD space consideration [10:34] well, it's probably either going to be 2 webkits, or 1 webkit + 1 gecko [10:34] (and banshee tomorrow in the seed? + unity on wednesday… urgh…) [10:34] chrisccoulson, s-c I guess [10:35] chrisccoulson, empathy as well [10:35] shotwell [10:35] ah, ok. that's quite a few things then [10:35] rhythmbox [10:35] yes... [10:36] do we know how much space xulrunner takes on the CD compared to webkit? [10:36] pitti^^ [10:36] not sure if we can get to use gtk2 rather than gtk3 for the update [10:36] chrisccoulson, isn't xul need by couchdb? [10:36] seb128, chrisccoulson, why do we need two webkits? [10:36] robert_ancell, gtk2 and gtk3? [10:36] seb128, but what uses -gtk2 on the CD? [10:37] robert_ancell, software-center, rhythmbox, empathy [10:37] seb128 - couchdb is only using spidermonkey. if we used the new yelp, then i'd probably spend some time to split the packaging so we only pull in the bits we need [10:37] robert_ancell, shotwell [10:37] hello? [10:37] rodrigo_, hey [10:37] ok, back online it seems!! [10:37] hi robert_ancell [10:37] hi seb128 [10:37] seb128, damn [10:37] hey rodrigo_ [10:38] chrisccoulson: xulrunner: 10 MB, libwebkit: ~ 6 MB [10:38] chrisccoulson: i. e. if we can actually drop xulrunner with the new webkit, it's a net win [10:38] chrisccoulson, hey [10:38] kklimonda, sorry, not sure if you told me something, my connection dropped [10:38] robert_ancell, do you know if we could build using gtk2 instead of gtk3? [10:38] hi didrocks, robert_ancell [10:39] pitti - nice. so, it we make an assumption that spidermonkey is around ~2.5MB of that 10MB, then we get a small win [10:39] rodrigo_, Zdra has comments on the gnome-keyring update [10:39] (based on the fact we'd need to keep spidermonkey) [10:39] rodrigo_, some missing depends and the daemon not running [10:39] seb128, ok [10:39] Zdra, ping [10:39] chrisccoulson: how does yelp+webkit hold up wrt. a11y? [10:39] seb128, I guess it's possible. It would be a non-trivial patch though [10:39] pitti - i'm not too sure, that's partly why i asked the original question [10:39] rodrigo_, pong [10:39] pitti, you have been Cc-ed onthose emails [10:39] i wasn't sure if a11y was still a blocker for going to the new version [10:39] pitti, I started a discussion specifically about that last week [10:40] rodrigo_, I'm using your gnome-keyring 3 package from gnome3-build package [10:40] my observations: [10:40] 1) libgcr-dev miss a dep on libgck-dev [10:40] seb128: right, and there wasn't an obvious "this is ok now" conclusion there [10:40] chrisccoulson, pitti: we pretty much decided that better having an useful version with limited a11y than a slow and useless version [10:40] ok, good to know [10:40] Zdra, oh, ok [10:41] ups, "is better" [10:41] 2) gnome-keyring-daemon does not start: $ gnome-keyring-daemon -s [10:41] ** Message: couldn't connect to control socket at: /tmp/keyring-s6Zqfv/control: Connection refused [10:41] seb128 - yeah, i'd prefer that too, especially seeing as i'm pretty much on my own with getting the current version working with the latest gecko [10:41] i guess nobody else is going to try and make that work [10:41] the gecko version is slow and the documentation format is suboptimal [10:41] robert_ancell, let's not spend efforts with that then [10:41] 3) all apps linked on libgcr does not start anymore, because it is not // installed. For example: $ seahorse [10:41] seahorse: error while loading shared libraries: libgcr.so.0: cannot open shared object file: No such file or directory [10:42] let's rather get the new version in using gtk3 and see later on what to do if we need space [10:42] sure [10:42] rodrigo_, that's it :) [10:42] Zdra, rodrigo_: that seems a real issue in the way g-k has been packaged [10:42] Zdra, hmm, is the daemon already running when you start it again? it is running ok here [10:42] ok, i'll leave yelp off my list of things to do then, and add splitting the xulunner package to my list of things to do instead [10:42] why did libgcr0 go away? [10:42] chrisccoulson, great [10:42] rodrigo_, ps aux | grep keyring --> that returns nothing [10:43] seb128, it didn't go away [10:43] it's renamed though [10:43] it's not libgcr-3 [10:43] I guess we should rename the package also [10:44] shouldn't it be libgcr3 ? [10:44] yeah, or -3 [10:45] oh, .so file indeed has that -3, so it really seems to only be a matter of renaming the package to get it in // of version2 [10:45] libgcr3 implies that 3's the .so version of the lib, so -3 sounds better, right? [10:45] rodrigo_, but why " seahorse: error while loading shared libraries: libgcr.so.0: cannot open shared object file: No such file or directory" [10:45] although the .so version is the same [10:45] rodrigo_, that's a packaging screwing [10:46] seb128, it's called now libgcr-3.so* [10:46] rodrigo_, libgcr0 whould still be on disk [10:46] rodrigo_, you can't do that [10:46] seb128, that's the lib name from upstream [10:46] rodrigo_, you should have libgcr0 and libgcr3 on disk [10:46] rodrigo_, well the old binary should be installed with the old name [10:46] as long as thing depends on it [10:46] yeah, needs renaming to libgcr0-3 or libgcr3 [10:46] which one? [10:47] libgcr-3-0 [10:47] ok [10:47] what is the .so name? [10:47] libgcr-3.so.0 [10:47] ? [10:47] .0 [10:47] yes [10:47] ok [10:47] so libgcr-3-0 [10:48] libgcr-3-0 for the binary [10:48] but shouldn't g-k build both versions? [10:48] or do you have a new source? [10:48] robert_ancell, chrisccoulson, pitti: some work that could bring space.. [10:48] it's a new source [10:48] convert gnome-system-monitor from gtkmm to vala or other [10:49] it's the only thing keeping gtkmm on the CD [10:49] oh, interesting [10:49] is anybody working on gnome-system-monitor upstream? [10:49] no [10:49] ah, i didn't think so [10:49] or find another software in universe to replace it [10:50] g-s-m is not great in any case [10:52] (on the topic of glibmm, here's the test code for compiz http://git.compiz.org/~dbo/compiz-with-glib-mainloop/log/?h=glibmm-experimental) [10:53] ok, glibmm and pangomm are small [10:53] the main win would be gtkmm [10:53] I guess we can keep glibmm [10:53] smspillaz, ^ [10:53] later all [10:54] later robert_ancell [10:54] robert_ancell, bye [10:54] seb128: cool [10:55] 10:05 kklimonda | rodrigo_: I'll use it (the branch) to finish _execute_query part of the merge - i.e. add _execute_query and port all other code to use it [10:55] later [10:56] kklimonda, ah, ok [10:56] smspillaz, bye [10:56] pitti, do you want to take a wi on your cd spec about figuring a way to replace g-s-m or port it to non gtkmm rather? [10:56] seb128: sounds fine [10:57] ok, great [10:57] * pitti changes [10:57] RAOF: do you plan to upload the dynamically linked mesa soon? that should help a lot with the current 30 MB oversizedness [10:58] I did a few uploads to reduce changelogs, but they should only gain some 1.5 MB [10:59] do we have langpacks to drop for a1 if needed? [11:00] some [11:05] rodrigo_, so about my issue with starting gnome-keyring-daemon, any idea? [11:05] I see in /tmp a looot of dir like keyring-XXXX [11:05] Zdra, only thing that I can think of is some leftover file from a previous keyring daemon, that confuses it [11:06] Zdra, you have tried it with a clean /tmp? [11:06] yep tried with a clean keyring [11:06] hmm, then that can't be [11:07] rodrigo_, note that with clean /tmp the error is slightly different: [11:07] $ gnome-keyring-daemon -s [11:07] ** Message: couldn't access conrol socket: /tmp/keyring-s6Zqfv/control: No such file or directory [11:07] wondering who is supposed to create that file [11:07] and without the clean /tmp, what message is it? [11:07] I think it's the daemon itself, let me check [11:07] Zdra: did you check the permission of /tmp ? [11:07] ls -ald /tmp [11:07] ** Message: couldn't connect to control socket at: /tmp/keyring-s6Zqfv/control: Connection refused [11:08] rodrigo_, ^ [11:08] oh, ok [11:08] bigon, of course my /tmp is writable [11:08] just that "something" created /tmp/keyring-s6Zqfv/control before and that file isn't readable for keyring-daemon [11:09] chrisccoulson: do you remember the slow launchpad bug we discussed extensively last March? still happens with firefox 3 https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/223238 [11:09] Launchpad bug 223238 in linux-restricted-modules-2.6.24 (Ubuntu) (and 1 other project) "Extremely slow Ajax/Javascript/CSS performance in Firefox 3 using non-free nvidia-glx-new (affects: 20) (dups: 4) (heat: 147)" [Undecided,New] [11:09] (it's happening in both vesa and nvidia driver) [11:11] didrocks, yeah, i remember that. i don't remember what the outcome of the discussion was though [11:11] Zdra, http://mandriva.598463.n5.nabble.com/Bug-61599-gnome-keyring-NEW-keyring-does-not-start-td3256029.html [11:12] Zdra, it suggest to use glib-compile-schemas /usr/share/glib-2.0/schemas [11:12] Zdra, can you try? [11:12] rodrigo_, do you install the schemas? [11:13] chrisccoulson: well, launchpad guy tells it's a Firefox bug not handling properly transparency and duplicated icons (looks at the duplicate) [11:13] it could be that you didn't update the binary to ship it? [11:13] maybe [11:13] but it's working for me, so not sure [11:13] checking [11:13] chrisccoulson: can you try to push that upstream? I see a lot of duplicate googling for "launchpad slow firefox nvidia" which aren't related to download page time but rather rendering [11:14] chrisccoulson: it affects some other websites as well [11:14] didrocks, i'll have a look at that in a bit [11:14] chrisccoulson: thanks :) [11:14] oh, right, it's working for me because I have the old version, the version in the PPA doesn't show up on the updates!! [11:15] right, seems it's missing the schemas indeed [11:16] Zdra, fixing and building locally to fully test it, will upload as soon as it works [11:16] rodrigo_, let me restart my session and see if the trick worked :) [11:16] Zdra, it won't I think, we don't have the schemas in the g-k package [11:17] rodrigo_, rename the library while you are at it [11:17] Zdra, you will want to reinstall the official libgcr0 [11:17] seb128, yes, fixing all the issues Zdra raised [11:17] rodrigo_, I will review your ppa uploads today [11:17] I started on gnome-media the other day [11:17] seb128, ok, coolk [11:18] seb128, they're all on the ubuntugtk3 branches [11:18] I think some reviews will be useful [11:18] rodrigo_, ok [11:18] rodrigo_, seb128: ok [11:19] * rodrigo_ build packages [11:19] will try again when new pacakge is in ppa [11:19] thanks :D [12:01] hmmm, https://wiki.ubuntu.com/DesktopTeam/Specs/Natty/Firefox4/XULRunner20Transition is a nice list of things to work through [12:02] Zdra, ok, seems to work, you want to build it locally to make sure or just wait for it to be in the PPA? [12:03] rodrigo_, I don't really have much time atm, so I'll be lazy and wait for ppa :) [12:03] Zdra, ok, uploading then [12:03] rodrigo_, thanks :) [12:12] session restat, brb [12:12] restart* === al-maisan is now known as almaisan-away [12:20] bah, firefox + launchpad is just unusable here… [12:21] * didrocks switches back to chromium === MacSlow is now known as MacSlow|lunch [12:25] didrocks, hey, do you know if FF 4.0 is going to land in natty? or will old 3.6 be used? [12:25] nisshh: FF 4beta7 is already in natty [12:25] oh coolies [12:32] didrocks, get a decent video card ;-) [12:32] seb128: well, nvidia is nice TBH, I didn't get any issue for compiz/mutter unlike other cards :p [12:33] I don't have any issue with compiz on intel [12:33] just that a launchpad page makes 4-6 seconds to refresh… [12:33] "just" [12:33] heh, i dont want to try compiz unity on my intel IGP :) [12:33] seb128: well, working well with chromium [12:33] seb128, really? [12:33] nisshh, yes, why not? [12:34] I've been using compiz on intel for like 5 years [12:34] seb128, my intel graphics are from 2003 :) [12:34] on different cards, without any issue [12:34] oh, compiz is not the issue [12:34] compiz + unity might be :) [12:34] oh right, I didn't try that yet [12:34] the mutter version of compiz ran rather slow on my machine [12:34] yeah [12:42] didrocks, bug #580295 doesn't seem to be a compiz bug [12:42] Launchpad bug 580295 in unity (Ubuntu) (and 1 other project) "Alt-F2 not working (affects: 44) (dups: 2) (heat: 254)" [Wishlist,Triaged] https://launchpad.net/bugs/580295 [12:42] didrocks, you just need to have a running software handling the keybinding [12:43] seb128: that's what I'm telling in my latest comment, where did you see I'm telling it's a compiz bug? [12:43] seb128: we can tell to compiz "run that on alt+F2" [12:43] didrocks, by reading sam's comment ;-) [12:43] so, not mine :) [12:43] I just didn't understand your reply [12:44] just telling that the workaround sam mentionned works today as there is gnome-panel running, but it won't work when gnome-panel won't be there anymore [12:44] didrocks, isn't that just a matter to have unity handle the keybinding the way gnome-panel is doing it? [12:44] didrocks, you want a new application only for that? [12:44] no, it will be compiz which will handle the shortcut, we just have to teach it to not rely only on gnome-panel [12:45] didrocks, wouldn't it make sense to have it in unity? [12:45] seb128: that was what we discussed at UDS [12:45] seb128: not part of unity, apparently [12:45] not sure why, ask mr Patel :) [12:45] didrocks, such bugs would benefit from reference to specs or uds notes [12:45] reference like url [12:46] seb128: sure, but the note wasn't taken IIRC [12:46] hmm, is LP down [12:46] ? [12:46] rodrigo_, works here [12:46] not sure we should discuss that extensively, just do the work :) [12:46] rodrigo_, seems it's not your day for internet ;-) [12:46] no :( [12:46] didrocks, right, I'm just concerned about adding a new binary or source for such a small thing [12:47] seems work over what it's worth [12:47] seb128: well, the binary can be added to unity package… [12:47] or rather seems it would be easier to teach the unity-panel to do it [12:48] didrocks, ok, I will let njpatel handle that and comment when it's done if I've issue they way it's done ;-) [12:48] seb128: right :) [12:48] what's up [12:48] run dialog? [12:48] njpatel, just wondering about the run dialog being "dedicated app will be needed." [12:49] if we use the gnome-panel one [12:49] I might just rip the code out [12:49] the backing code, and stick and unity front-end on it [12:49] should be do-able, just haven't had the time [12:49] njpatel, so it will be in unity source right? [12:49] njpatel, not a new source? [12:50] yeah, i think that's what we'll do. didrocks and I spoke about ripping out the gnome-panel one into it's own source, but I think we can do better than that through unity itself [12:50] njpatel, ok great, I just wanted to mention that I think it's not worth the work of being a different source [12:50] well, there are bigger fish now than that issue TBH :) [12:50] yep [12:51] didrocks, right, doesn't prevent to comment on bugs though [12:51] rodrigo_, Unpacking libgcr-3-0 (from .../libgcr-3-0_2.92.92.is.2.91.2-0ubuntu2~build1_amd64.deb) ... [12:51] dpkg: error processing /var/cache/apt/archives/libgcr-3-0_2.92.92.is.2.91.2-0ubuntu2~build1_amd64.deb (--unpack): [12:51] trying to overwrite '/usr/share/gcr-3/ui/gcr-unlock-options-widget.ui', which is also in package libgcr0 2.92.92.is.2.91.2-0ubuntu1 [12:51] didrocks, but anyway I was just wondering about that, we discussed it enough and I had my reply so let's move on [12:52] agreed :) [12:52] Zdra, I told you that you needed to reinstall the libgcr0 for ubuntu [12:52] or "from ubuntu" rather [12:52] Zdra, yes, you have the previous wrong libgcr0, install the official one [12:52] Zdra, the official one [12:52] Zdra, 2.92.92.is.2.31.x... [12:52] Zdra, sudo apt-get install libgrc0/natty [12:52] or maverick if you use that version [13:03] rodrigo_, yay all working fine \o/ [13:03]