[05:24] Good morning [05:25] cyphermox_: ah, do we still plan to use i-n instead of nm-applet at some point? [05:48] good morning [05:56] bonjour didrocks [05:57] guten morgen pitti [05:57] pitti: reminder about the missing reminder to remind you about the team report email sending [05:57] didrocks: heh, merci [05:57] de rien ;) [06:10] pitti: btw, if you can bump the score again for the qt build in the ubuntu-desktop ppa (should be the last build, with all new stuff included): https://launchpad.net/~ubuntu-desktop/+archive/ppa i386 is enough [06:11] didrocks: done [06:11] pitti: excellent, thanks! :) [06:16] Can anybody in here with latest updates installed try running either orca or accerciser? if you can, do you get this error? "** (orca:4445): CRITICAL **: pyg_enum_add: assertion `typename != NULL' failed" [06:17] I get that both with latest orca I uploaded to Ubuntu and accerciser. Appears to be a pygobject issue, however I do have 2.90.3 installed. [06:21] The traceback from accerciser is more helpful than orca giving nothing thankfully. [06:24] * didrocks reboots and makes some tests, brb [06:28] TheMuso: looks slightly different here [06:28] Hi all, is there a way to disable all or some of unity multitouch gestures? [06:29] pitti: hrm ok. [06:29] TheMuso: http://paste.ubuntu.com/683163/ [06:29] /usr/lib/python2.7/dist-packages/gi/module.py:123: Warning: cannot register existing type `EventType' [06:29] pitti: right thats what I'm getting from accerciser. [06:29] TheMuso: that sounds like it tries to import both static and GI modules at the same time? [06:29] Orca does funky things with its stderr so one doesn't get a traceback from that. [06:30] pitti: hrm ok will check for that. [06:30] TheMuso: hm, doesn't seem to, though [06:30] TheMuso: I usually edit /usr/share/pyshared/gobject/constants.py to immediately raise an ImportError, to see clearly where the static imports are coming from [06:31] pitti: ok [06:31] pitti: although I can't really see what would need to be weaked in that file to throw sed import errors. [06:32] TheMuso: the exception just pretty darn looks like something already imported the static gdk module before [06:33] ok, metacity regression between 2.30.3 and 2.34, ricotz: did you package some intermediate 2.32 version in your ppa? first version uploaded to oneiric (2.34) brings a world of color shift at startup. Some kind of nightclubbing :) [06:33] ah right [06:40] didrocks, hello, no, there are no metacity packages [06:40] ricotz: ok, I guess you saw this regression as well? (a lot of grey/blue/red color when starting metacity) [06:42] didrocks, actually i am not seeing it ;), i am not running metacity [06:42] ricotz: ok, I was asking about it as you made the update :) [06:42] my current setup isnt really a plain oneiric [06:42] ricotz: people report the same in the ppa [06:43] hmm [06:43] can you give me the bug link? [06:44] ricotz: this one has videos, for instance: https://bugs.launchpad.net/unity-2d/+bug/820553 [06:44] Ubuntu bug 820553 in unity-2d "Flicker upon session start" [High,Confirmed] [06:45] didrocks, which ppa do you mean? [06:45] ricotz: people are using the gnome3 ppa in natty (saw that on forums) [06:46] ok, that has 2.34 of course [06:46] pitti: ok, seems that there are no static imports in accerciser, so far as I can find/test for. [07:23] good morning everyone [07:23] hey chrisccoulson [07:24] hi pitti, how are you? [07:24] quite fine, thanks! [07:24] * pitti tries to keep up with the Seb emulator, uploading the fourth gtk+3.0 in two days [07:25] heh [07:25] that's worse than firefox! [07:25] ;) [07:26] * pitti adds a new gvfs, and now goes to update gtkmm, gnome-desktop3, and gedit [07:27] oh, and before, sponsoring jbicha's branches [07:29] hey chrisccoulson! [07:29] hi didrocks [07:29] i'm hoping today goes smoother than yesterday, like, not having GCC crash on all my builds :) [07:30] heh ;) [07:38] morning [07:39] hey rodrigo__, how are you? [07:39] * pitti wonders how many underscores rodrigo__ wants to pile up [07:40] is it a pure coincidence that the OSD volume icons shown by gsd transition at exactly the same volume levels as the sound indicator now, or did ronoc change the thresholds in the indicator-sound? [07:40] they used to be completely out of sync previously [07:40] hey rodrigo__! [07:41] ok, the issue in metacity is due to compositing activation and X damages. I'm afraid it's a little bit too ahead of my knowledge… [07:48] chrisccoulson: do you happen to have your gedit 3.1.4 upload in bzr, perhaps forgot to push? [07:48] chrisccoulson: if not, I can grab the diff from LP and commit it [07:49] pitti - oh, sorry. i've just pushed that now === htorque_ is now known as htorque [07:52] chrisccoulson: cheers [08:06] wow, "bzr mu" is a lot more clever than I thought [08:06] hum, it seems that /usr/share/gconf/defaults isn't taken into account anymore [08:06] without any arguments it uses uscan, downloads the latest orig, merges it, and adds a new changelog [08:06] oh nice, would be great that it does the same for the "upstream branch in bzr" mode ) [08:07] :) [08:08] hum, there is clearly something weird in gconf, it seems to ignore /var/lib/gconf/debian.defaults/%gconf-tree.xml [08:11] i <3 inline reply in tbird conversations :) [08:11] pitti, someone stole rodrigo and rodrigo_, that's why the extra _, so let's hope nobody takes rodrigo__ :) [08:12] rodrigo__: you really should register your nick [08:12] rodrigo__: you should register your nick :) [08:12] then it'll be your's forever [08:13] rodrigo__: is "rodrigo" registered? if not, you could snatch it [08:14] rodrigo is registered by someone else, rodrigo_ by me [08:14] rodrigo__: ah, so use rodrigo_ ? [08:14] /msg NickServ ghost rodrigo_ YOURPWD [08:15] /nick rodrigo_ [08:15] I use that one, but when someone takes it, there's no way to get it back, right? [08:15] /msg NickServ identify rodrigo_ YOURPWD [08:15] rodrigo__: you can ghost it [08:15] pitti, I have xchat do that every time I log in [08:15] not ghost it, trying that [08:15] hey there [08:15] hey seb128 [08:15] oh, seb128! [08:15] seb128: aren't you supposed to be on vac? [08:16] seb128: I thought we could break gnome 3.1.90, but seems you are watching us :) [08:16] * pitti hugs seb128 [08:16] lol [08:16] pitti, hug ;-) === rodrigo__ is now known as rodrigo_ [08:16] rodrigo_: yay [08:16] yeah, "supposed to", but I think I will work from today until thursday and take some days next week rather [08:16] pitti, cool, ghosting it works! [08:17] isn't .91 the current hot stuff? ;-) [08:17] yep [08:17] some upstream projects are still on .90, though [08:17] seb128, enjoyed your vacation? [08:17] seb128: anything which needs urgent attention, or does the weather just suck? [08:21] rodrigo_, the first week yes, we are having some familly issues so the holidays I had plan this week didn't work [08:21] will report to next week rather [08:22] seb128, oh, I hope nothing serious [08:26] rodrigo_, nothing we will not deal with, no worry, but thanks ;-) [08:26] seb128, good to hear :) [08:29] so, how went beta1? and the GNOME .90 updates? [08:29] rodrigo_, did you manage to close all the g-c-c bugs yet? ;-) [08:29] seb128: beta-1 was a bit bumpy, we had several rebuilds (one during Wed->Thu night), but went fine in the end [08:30] ugh, xchat crashed [08:32] seb128, .90 is all in, now working on .91 [08:32] rodrigo_: well, http://people.canonical.com/~platform/desktop/versions.html still has some 3.1.4 versions [08:32] trying to catch up :) [08:32] btw, who maintains the versions page? we are missing gnome-contacts, gnome-online-accounts and maybe others [08:32] pitti, yeah, right [08:33] I'll upload some more as soon as I fix the gnome-keyring build issue [08:33] I'm working on gnome-keyring now (jbicha encountered an FTBFS, sending patch upstream), and gtkmm3.0 building [08:33] rodrigo_, the vcs is in ~ubuntu-desktop, the url is at the bottom of the page (where nobody ever scroll, after all the green) [08:33] rodrigo_: oh, you are? [08:34] rodrigo_: I really was just about to start, so I guess I'll leave that to you and work on some other 3.1.4 updates? [08:34] hey seb128 :) [08:34] rodrigo_, I can add gnome-contacts and gnome-online-accounts [08:34] hey didrocks [08:35] * pitti grabs -menus and -icon-theme [08:35] pitti, yes, I fixed it on the git checkout, now trying building the package with the patch [08:35] rodrigo_: perhaps mention it on the pad? [08:35] seb128, ah, I guess I have permissions there? so I can do it myself if you want [08:35] pitti, yeah, done [08:35] rodrigo_, if you want feel free [08:36] ok [08:36] rodrigo_, they are on http://people.gnome.org/~vuntz/tmp/versions/versions-unstable [08:36] rodrigo_, so you basically just need to add a line in the table with GNOME_FTP as location [08:36] ok [08:40] any other package we are missing there? I just added couchdb-glib, evo-couchdb, gnome-contacts and g-o-a [08:40] rodrigo_: gtkmm3.0 [08:40] pitti, ok, adding it [08:40] thx [08:41] ('gtkmm2.4', GNOME_FTP), [08:41] ('gtkmm3.0', GNOME_FTP), [08:42] the script knows how to map gtkmm?.? to .../gtkmm/?.? ? [08:44] rodrigo_, no [08:44] meh, seems I get all the huge packages today -- gnome-games, icon-theme, gtk, gtkmm [08:44] rodrigo_, looks for gtk+ in the source [08:44] ok [08:44] * pitti sees his upstream bandwidth set fire [08:44] it has hacks for gtk2 and gtk3 [08:45] waouh [08:45] jbicha has been quite active on doing updates [08:46] absolutely [08:46] pitti, thanks for doing sponsoring! [08:46] I've been sponsoring him for some time now, I think he's ready for ~ubuntu-desktop [08:46] yeah, I was thinking around this line as well [08:48] pitti, I'd like to re-enable oneiric langpacks. If that's ok, I'll go ahead with it now, so that tomorrow we've got a new set of delta ones [08:48] dpm: saw your ping yesterday, already enabled [08:48] dpm: good morning! [08:48] pitti, ah, excellent, thanks and good morning to you too! :-) [08:49] pitti, do you know what's going on with lightdm and unity-greeter? [08:49] seb128, rodrigo_: at times like this, versions.html is severely lagging; do you see any way of speeding it up? [08:49] we didn't get any update in 2 weeks [08:49] like, querying the versions on LP instead of archive.u.c.? [08:49] seb128: I don't know, I'm afraid [08:49] ok [08:50] it's behaving here now, are there still major problems with it? [08:50] pitti, hmm, it runs every hour, right? [08:50] pitti, that's why I usually dumped the "free to grab" list on the etherpad [08:50] rodrigo_, every half an hour [08:50] right, but archive.u.c. only updates every hour [08:50] or maybe we switched back to every hour [08:50] i. e. it has a lag between one and two hours [08:50] http://pastebin.com/gGc6AZ7M [08:51] pitti, right [08:51] pitti, lightdm> not especially, but trunk have quite some improvement, including ui tweaks for the unity greeter to match design [08:51] pitti, doing query from launchpad rather than the archive seems like it would be a win indeed [08:51] that's why I wonder whether we could use launchpadlib to query the most recent source instead [08:51] yes, maybe it could be changes to query on the fly [08:51] but anyway, that would mean some nontrivial engineering, I suppose [08:52] yeah, I guess so [08:52] pitti, not really [08:52] but an lplib call to get the most current source is rather easy [08:53] pitti, l1250 [08:53] "# Get the Ubuntu versions from the sources list [08:53] for source in package_names: [08:53] p = packages[source]" [08:53] (p.ubuntu_version, p.section, p.maintainer) = ... [08:53] so, does this look ok -> http://pastebin.com/AdEv4FP6 ? [08:53] you just have to replace the python-apt call there by a lplib equivalent [08:54] pitti, that loop iterate over the source and get the versions from the apt database [08:54] I just don't know how laggy it would be to make some hundred lplib calls (i.e one for each source) [08:55] rodrigo_, seems fine to me [08:55] worth a try locally [08:55] seb128: if you want me, I can have a look after I'm done with my current batch of updates [08:55] pitti, if you want please feel free [08:55] seb128: where's the code again? (in bzr) [08:55] seb128, ok, I'll push that then [08:56] pitti, lp:~ubuntu-desktop/+junk/versions [08:56] pitti, versions.py ~l1245 is where you want to look [08:58] ugh [08:58] $ bzr push --remember lp:~ubuntu-desktop/+junk/versions [08:58] Doing on-the-fly conversion from RepositoryFormat2a() to RemoteRepositoryFormat(_network_name='Bazaar pack repository format 1 (needs bzr 0.92)\n'). [08:58] This may take some time. Upgrade the repositories to the same format for better performance. [08:58] bzr: ERROR: RemoteRepository(bzr+ssh://bazaar.launchpad.net/~ubuntu-desktop/%2Bjunk/versions/.bzr/) [08:58] is not compatible with [08:58] CHKInventoryRepository('file:///opt/extra/src/canonical/.bzr/repository/') [08:58] different rich-root support [08:59] rodrigo_: do you have an old checkout perhaps? [08:59] rodrigo_: try "bzr upgrade" [08:59] pitti, no, I just checked it out for the 1st time [08:59] ok, trying that [08:59] no, still fails [09:01] pitti, versions says that apport is outdated :p [09:01] seb128: I uploaded that some two hours ago :) [09:01] ok, we need to fix the get_version ;-) [09:01] if you upload at e. g. 8:01, versions.html will only pick it up at 10:00 [09:01] rodrigo_, https://code.launchpad.net/~ubuntu-desktop/+junk/versions/+upgrade [09:01] rodrigo_, I'm clicking on it [09:02] seb128, ok, go! :) [09:02] rodrigo_, try again once it's done upgrading [09:02] ah, that ^ means we should refrain from pushing for a bit [09:02] yes, better to wait [09:02] ok, I did the stack of merges/updates that I claimed on the pad [09:02] looking into versions now [09:03] seb128: we need index/oneiric/ only for the python-apt stuff, right? [09:04] pitti, yes [09:04] seb128: OOI, can we ignore/remove setup-launchpadlib-not-root? these days, the DC has current installations of python-launchpadlib and friends [09:04] pitti, we have one for debian as well, we could probably get those infos from launchpad as well though it's a bit trickier since we consider experimental [09:04] ah, right [09:04] we need the indexes for Debian [09:04] pitti, we can ignore,drop it yes [09:04] pitti, well index/oneiric we don't need, we need index/debian [09:05] pitti, though, https://launchpad.net/debian/+source/gedit [09:05] i.e we can probably get the debian versions from lplib as well [09:05] not sure about experimental though [09:06] https://launchpad.net/debian/+source/pygobject [09:06] ok, it does take experimental [09:06] pitti, ^ [09:06] so we can probably use lplib for debian the same way [09:06] seb128: do you have a trick to debug this, i. e. only run it for some three packages instead of all of them? [09:07] https://launchpad.net/debian/experimental/+source/pygobject [09:07] right [09:07] nice [09:07] pitti, yes, rename package_info = [ to package_info2 = [ [09:07] should be faster [09:07] then do a package_info with only a few sources [09:07] seb128: cheers [09:07] i.e overwrite the package_info table with what you need [09:08] seb128, let me know when it's done upgrading, so that I can try pushing my changes [09:08] rodrigo_, watch https://code.launchpad.net/~ubuntu-desktop/+junk/versions when the yellow banners goes away it's good to try ;-) [09:08] uh, this is rather old lplib code [09:09] seb128, ok [09:09] * pitti will replace that with login_anonymously() [09:09] pitti, yeah, that script started as a hack and we never really spent time to clean,modernize code [09:10] did somebody start on vino,vinagre? [09:10] if not I will do those updates [09:10] I didn't [09:11] I kept the pad up to date [09:11] well, it doesn't have the stuff any more that I already uploaded [09:12] seb128, neither did I [09:12] ok, good [09:12] seems the lp upgrade is done? [09:13] rodrigo_, pitti: let's keep things on the etherpad until they show on version to avoid confusion [09:13] ok, next time [09:13] or, how about we fix versions :) [09:13] next time you will have versions fixed :p [09:13] $ bzr info [09:13] Standalone tree (format: 2a) [09:13] ^ just checked out again [09:13] rodrigo_: push! push! push! [09:14] * rodrigo_ pushes!!! [09:14] pitti, pushed [09:15] rodrigo_: cheers [09:15] rodrigo_, that worked? [09:16] seb128, yes [09:16] great [09:16] so now, when does the online version get the changes from the branch? [09:17] rodrigo_, it bzr pull before running the script at every run [09:17] ah, right [09:17] so at the next run [09:17] so we'll need to wait to see if it works [09:17] yeah [09:17] gsettings-desktop-schemas is done, btw [09:18] and gnome-keyring [09:18] oh, and it shows the vala version incorrectly, we already have 0.13.3 === abhinav_ is now known as abhinav- [09:19] * rodrigo_ looks at the script's code [09:20] rodrigo_, I think we just track the wrong source in the package_info table [09:20] we track "vala" which is the 0.12 version [09:23] rodrigo_, I've dumped some "updates to claim" on the etherpad, you can use that rather than version [09:23] I've checked which ones didn't get uploaded or claimed yet [09:23] yeah, using the pad also [09:23] the only problem is when you remove stuff from the pad and the versions page still shows the old version [09:23] but pitti is fixing that, right? [09:24] working on it, yes [09:25] rodrigo_, that's why the "updates to claim" list on the etherpad is things you can pick now, I did the checking to make sure we don't conflict [09:25] ok [09:38] hmm, last upgrade removed the flash plugin package [09:38] any way to get it back? [09:39] pitti: i I can still abuse a little about your bump power, dx wants to test a new compiz release, https://launchpad.net/~unity-team/+archive/compiz-testing (all arch, all packages) [09:40] seb128: ah, the upgrade apparently destroyed the checkout on lillypilly, fixing.. [09:40] pitti, oh, thanks [09:42] didrocks: done [09:43] pitti: thanks a lot :) [09:47] http://people.canonical.com/~platform/desktop/versions.html updated now [09:47] that's with my simpler lp login code [09:47] and rodrigo's changes [09:48] it's got gtkmm3.0 now, but not gtkmm2.4? [09:49] the dual versions hacks are buggy I think [09:49] pitti, hmm, I still don't see my changes [09:49] rodrigo_, gnome-online-account is listed for example [09:49] hmm [09:49] gnome-online-accounts 3.1.1-0ubuntu2 3.1.90 [09:49] in the orange list [09:49] oh, it's couchdb-glib what is not there [09:50] gnome-contacts is neither [09:50] they are [09:50] but you need to click on the "+" next to package [09:51] right, there they are [09:51] you need to add them to package_whitelist = [09:51] l975 [09:51] since they are not seeded they don't show as being part of the default install otherwise [09:51] rodrigo_: please pull before changing anything, or even better, use a checkout/bind (I'm working on it as well) [09:51] pitti, ok [10:02] * didrocks reboots, brb [10:13] seb128: for Debian, versions.html doesn't make a difference between unstable and experimental, right? [10:13] pitti, no, we just take the most recent one [10:13] which is usually what you want for desktop components [10:13] right [10:16] "The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later." [10:16] hum, the etherpad is having issues today, I got disconnected several times before [10:21] rodrigo_, bug #828756 is fixed with today's keyring update right? [10:21] Launchpad bug 828756 in gnome-keyring "getting "connection is untrusted" warnings" [High,Confirmed] https://launchpad.net/bugs/828756 [10:21] seb128, yes [10:21] forgot to close it [10:21] thanks [10:21] closed it [10:21] thanks [10:22] bug #842072 is getting quite some duplicates [10:22] Launchpad bug 842072 in gnome-control-center "package gnome-control-center-data 1:3.1.5-0ubuntu4 failed to install/upgrade: trying to overwrite '/usr/share/icons/hicolor/16x16/apps/gnome-power-manager.png', which is also in package gnome-power-manager 3.1.3-0ubuntu2" [Undecided,Confirmed] https://launchpad.net/bugs/842072 [10:22] rodrigo_, pitti: ^ [10:22] seb128, yes, 3.1.9* of both should fix it [10:23] ok [10:23] rodrigo_, well if g-c-c ships a file g-p-m used to ship you still need a replaces because you don't know which one will be upgraded first in the upgrade run [10:23] right [10:23] i.e users upgrading from natty could get g-c-c unpacked before gpm and hit the conflict [10:24] so, a Replaces: g-p-m (< 3.1.90) ? [10:24] yes [10:24] or rather << [10:24] yeah, right :) [10:24] fixing it [10:24] thanks [10:28] rodrigo_: what's the last package you uploaded? did you upload anything in the last 23 minutes? [10:28] or seb128 ^ [10:28] pitti, g-c-c and file-roller, although f-r was more than 23 mins ago [10:28] pitti, why? [10:28] pitti, nothing since half an hour [10:29] I want an unpublished source for triple-checking my new versions.py [10:29] it's working now :) [10:29] apt-free, pure lplib, checking "pending" versions, too [10:29] ok, g-c-c is published [10:29] pitti, well, publisher run should not be done yet [10:29] is it? [10:29] seb128: that's just for archive.u.c. [10:29] is the publisher taking less than 45min nowadays?§ [10:30] seb128: packages get published from LP's POV a minute after the publisher starts [10:30] pitti, my g-c-c upload was 0ubuntu2, it still shows 0ubuntu1 [10:30] rodrigo_: yes, I haven't pushed yet [10:30] ah, ok [10:30] currently running with the full package list [10:30] pitti, cool then! [10:30] pitti, I will do some other updates in the next 10 minutes [10:30] pitti, will ping you when I upload [10:33] pitti, I'm just uploading g-c-c 0ubuntu3, if you want to check it [10:33] rodrigo_: ah, nice [10:35] hm, the full versions.py run takes quite a while, poking all the upstream locations.. [10:39] yay, done [10:39] http://people.canonical.com/~pitti/tmp/versions.html [10:39] seb128, rodrigo_ ^ output from my local run [10:39] it shows gnome-games as current already [10:40] pitti, the full run slow> yeah, we have been trying to solve that in the past [10:40] and gconf [10:40] \o/ [10:40] pitti, robert_ancell wanted to maybe get the upstream versions from launchpad itself as well [10:40] pitti, cool [10:40] we don't have pending merge proposals [10:40] but I didn't touch that code at all [10:40] since launchpad can track upstream tarballs if you give it the infos for that [10:41] hm, why did gtk+2.0 move from dark red to light red? [10:41] did I break that, or is that due to rodrigo's version table changes? [10:43] pitti, it got the debian testing version and not the unstable one [10:43] weird [10:43] oh, seems I'm missing some Debian versions [10:46] pitti, the launchpad web ui has the right version, so seems a bug in the script? [10:46] seb128: yes, working on it [10:47] great [10:47] seb128: it's due to how Debian is represented in LP [10:47] seb128: as we don't publish it, all package versions ever imported are in state "pending" [10:47] so I need to iterate through them and pick the most current one [10:52] right [10:54] ok, fixed [10:54] I think this is working now, want me to commit? [10:54] (I'll fix any further breakage from this, of course) [10:55] 4 files changed, 34 insertions(+), 62 deletions(-) [10:55] it's a simplification, after all \o/ [10:56] * pitti commits and takes the bullets [10:58] pitti, yeah, please commit [10:58] it has less code that's a good sign ;-) [10:59] * pitti works around stupid lucid lplib bug [11:00] pitti, one other option I was thinking about to get improved refresh time is to have 2 codepaths [11:00] one for GNOME which would use http://people.gnome.org/~vuntz/tmp/versions/versions-unstable [11:01]