[00:00] RAOF: Quick question for you re GNOME stable updates. Do we have a definitive list of what packages fall under the GNOME stable updates umbrella? [00:01] TheMuso: I don't know, actually. That's an excellent question :). I'd guess at everything in core gnome, and I'd guess that there's a list somewhere on gnome.org for that. [00:02] RAOF: Right, I am dealing with bug 773763 and am wondering whether I need to get rodrigo_ to fill out the full SRU paperwork. [00:02] Launchpad bug 773763 in evolution-rss "evolution freeze on startup with the plug-in "evolution-rss"" [High,In progress] https://launchpad.net/bugs/773763 [00:03] The branch itself checks out as ok. [00:06] I suspect evolution-rss is not core gnome. [00:30] TheMuso: Bah, sorry. irc bouncer decided to suspend. [00:32] TheMuso: Anyway - I can't actually find documentation of the gnome exception. I don't think evolution-rss counts, though. A bit more formal SRU work would be wanted anyway - at least to have a natty task and to have the oneiric task marked as fixed. [00:33] RAOF: Right, I asked for such in the bug. [00:49] TheMuso: evo-rss crash? [00:50] cyphermox: Yes. [00:50] Something along those lines. [00:51] heh [00:52] 0.2.2 to 0.2.5 seems steep... especially since IIRC I had already fixed something in evo-rss to keep it from crashing? [00:56] I didn't look that much into it myself, just reviewed the branch and the bug. [01:01] ok [01:02] well, worst case I guess rodrigo can instead backport just commit 536a117, which seems to be the exact issue, but there are a fair amount of other bugs fixed [01:02] Right. [01:33] 5~/c [02:56] * TheMuso -> lunch [04:15] RAOF: thanks for confirming [04:15] good morning [04:20] Morning pitti. [04:26] Morning pitti. [05:56] ok, images are as ready as they will be, so we may as well lift the freeze [06:22] good morning [06:28] bonjour didrocks [06:49] pitti: hey, (didn't see your ping), how is alpha2 coming? :) [06:50] didrocks: pretty well indeed; I lifted the soft freeze already [06:50] now preparing images, announcements, errata, etc. [06:50] excellent! Time to do some uploads then :) [06:55] Hm. I should probably add mesa 7.11 to the release notes. [06:58] RAOF: ah, please do [06:59] good morning everyone [07:00] hey chrisccoulson [07:00] hi pitti, how are you? [07:01] pretty well, thanks! smoothest alpha ever [07:02] Helloooooo chrisccoulson! [07:02] hi RAOF! [07:02] pitti - are we unfrozen yet? [07:02] chrisccoulson: go wild [07:03] chrisccoulson: not released yet, but the images are fixed now [07:03] pitti - excellent, thanks! [07:04] hey chrisccoulson [07:04] hey didrocks [07:04] * didrocks tries to race to upload Qt before chrisccoulson :p [07:04] but 25 minutes to create the source package :/ [07:05] lol [07:24] robert_ancell: I've finally managed to update my radeon box and reproduce that lightdm “nothing draws after log out” bug. Hah! [07:24] RAOF, any ideas? [07:25] didrocks: time for a faster computer [07:25] desrt: I would say, faster hard drive rather in the Qt case :) [07:26] robert_ancell: I'd guess it's something in the accel code being messed up. I'm investigating. [07:29] didrocks: SSD is totally worth it [07:30] desrt: I think I'll choose one soon. But still first to find what's wrong with my server which died during the really (power, motherboard or CPU?) [07:31] didrocks: just buy a new one [07:31] it's a great excuse for the upgrade and you can argue that any potential cost savings need to be offset against the cost of your time to debug the problem [07:32] (for purposes of self-justification) [07:32] desrt: that's quite right, but still, as the motherboard has some leds lightened, I still want first to avoid that if I can just change my power with another machine I don't use [07:33] no fun :) [07:44] didrocks: btw. do you know the proper process for registering 'official' freenode channels? [07:45] desrt: I think http://blog.freenode.net/2008/04/registering-a-channel-on-freenode/ it quite still up to date [07:46] didrocks: i mean specifically ubuntu 'official' channels (those starting with #ubuntu, not ##ubuntu) [07:46] desrt: oh, I don't remember, at the time, we just sent an email to canonical legal presenting our loco, but this was in 2005 :) [07:47] desrt: https://wiki.ubuntu.com/IRC/CreatingChannels [07:47] ahah. that is what i was looking ofr indeed [07:47] thanks [07:58] hey [07:58] salut seb128 [08:01] didrocks, are there indicator .vapi files? [08:01] lut didrocks [08:01] hey robert_ancell! [08:01] seb128, hello [08:01] robert_ancell, libappindicator3-dev: /usr/share/vala/vapi/appindicator3-0.1.vapi [08:01] ? [08:01] robert_ancell: there are for some indicator, like appindicator, soudindicator [08:02] not sure about everything though [08:03] robert_ancell, you should review the xubuntu patches in launchpad for your next upload if you can, seems having the way to use an alternative config is a blocker for them [08:03] seb128, didrocks, I need the ones for libindicator-dev. [08:03] seb128, it's on my list. I have a big list [08:04] ;-) [08:05] robert_ancell: didn't find any for this one, forward your request to ted I would say [08:05] robert_ancell, I will check for you today with the u.s guys and drop you an email [08:05] seb128, k, thanks. It's not a big deal, I can write my own just for the functions I need [08:06] robert_ancell, what do you need? (just curious) [08:06] seb128, copying the code out of ubiquity for the greeter [08:07] ok [08:08] robert_ancell, did you have a nice flight back this time btw? [08:08] robert_ancell, how is your jetlag? ;-) [08:08] seb128, much easier than last time [08:08] all back to normal [08:09] great [08:09] jasoncwarner_ and I got upgraded to premium economy which made it easier too [08:10] seb128: in unity 2.0, we had an indicator.vapi, not sure how uptodate it is, but maybe it can be of interest for you? [08:10] lucky you ;-) [08:10] didrocks, njpatel has a libindicator vcs with a vapi as well [08:10] didrocks, but I prefer to check with ken and ted why they don't build one, they probably have a technical reason or had issues [08:11] I think it's just that you have to ship it manually and update it yourself :) [08:11] https://code.launchpad.net/~njpatel/libindicator/vala [08:11] but that's old [08:11] didrocks, yeah, that sounds likely :) [08:12] but do you need a vapi? [08:12] or do you need a typelib? [08:12] seb128, you just need a vapi for most things [08:13] didrocks, seb128 I can update it if you want? [08:13] njpatel: what, do you have spare time? :-) [08:13] njpatel, robert_ancell needs it for lightdm [08:13] but I think he said he can ship a copy for the function he needs [08:13] I was just curious of why the libindicator has no gir or vapi [08:13] didrocks, what? I mean no, no, I can't do it [08:13] :) [08:13] I only need a few function calls [08:14] robert_ancell, seb128 didrocks can do it sometime today so will be ready tomorrow, sound good? [08:14] works for me [08:14] sounds good to me! [08:14] hopefully ted hasn't completely broken libindicator ;p [08:14] njpatel, thanks [08:14] np :) [08:15] * njpatel tests out if vga/displayport works with his laptop now with edgers [08:15] thanks njpatel ;) [08:16] damns it [08:17] bah, it's annoying that segfaults on exit bugs always get priority raise because they get duplicates [08:17] we need to find a way to filter those out [08:17] njpatel: You'd want to be trying the Oneiric kernel for that. [08:17] RAOF, i'm in oneiric + xorg edgers, but neither work :( [08:18] neither = VGA or displayport [08:18] njpatel: Sadness. [08:18] both get detected with a 1024x768 monitor (though the monitor is way larger than that), X thinks it's brought up the screen, but the monitors don't get a signal [08:18] sorry, brought up the monitor (i.e. it extends the desktop) [08:19] morning [08:19] RAOF, this is x220...any idea? [08:19] RAOF, (sandy bridge) [08:20] RAOF, just got this from the display settings http://pastebin.com/6GryKjPt [08:21] njpatel: I've not seen anything suggesting that it's a known problem upstream; dmesg and /var/log/Xorg.0.log are most likely to contain interesting debugging info. [08:22] RAOF, dmesg has a million of "[ 4412.963511] [drm:intel_dp_i2c_aux_ch] *ERROR* too many retries, giving up" [08:22] Ah. That suggests it's failing while trying to retrieve the EDID, which would be consistent with X thinking it's a 1024x768 monitor. [08:23] RAOF, should I be worried? :/ [08:23] RAOF, this happened with stock O as well as edgers [08:24] pitti, hey, can I upload today? ;-) [08:24] (new glib and gtk available) [08:24] seb128: yes, see #u-devel topic -- freeze is lifted [08:24] Depends on what you mean by ‘worried’. [08:24] https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/791169 [08:24] Ubuntu bug 791169 in xserver-xorg-video-intel "Picture disappearing when Display Port is connected" [Undecided,Incomplete] [08:24] \o/ [08:24] pitti, danke [08:24] RAOF, will it ever work? :) [08:24] RAOF, or, can I fix it? [08:25] njpatel: Looks like it might; there's a patch series on intel-gfx@ with some DP fixes which Keith says makes his x220 work better :) [08:25] aaah, nice [08:26] RAOF, okay, i can wait for whenever that lands, thanks! [08:28] * njpatel updates bug [08:29] ahah, I was just getting blocked on bug #806661, nice timing :) [08:29] Launchpad bug 806661 in python-imaging "[oneiric] python-imaging needs to be multiarch-aware" [Medium,Confirmed] https://launchpad.net/bugs/806661 [09:09] mvo: *hug* for the doc-base update! [09:09] mvo: so we can close the bug now? [09:12] so who want to do some updates today? [09:13] pitti: I think so, but I want to wait until it hits the archive to be sure that my test-case runs fine then. I did not manage to reproduce it manually, only in the full ubuntu->ubuntu release upgrade [09:13] mvo: ah, makes sense; thanks! [09:20] rodrigo_, hey ;-) [09:20] rodrigo_, do you have some slots for updates still today or do you want to get work done on other things? [09:23] seb128, I am working on infrastructure in g-c-c for lots of our bugs, but will have some slots for sure, so what do you want me to upload? [09:23] rodrigo_, don't bother, there is no hurry, better than you get the bugs count down a bit ;-) [09:24] seb128, I can do some updates anyway, no problem [09:24] if you get bored gnome-utils or gedit should be easy [09:24] ok [09:24] I'm doing the glib, gtk3 updates [09:29] seb128: I'll push the multiarch changes of gobject-introspection to Debian, and will then look into the upgrade [09:29] pitti, danke [09:30] pitti, do you want to look at the dbus update as well when you have a free slot? [09:30] yes [09:30] thanks [09:30] added to pad [09:38] ah, Debian doesn't have glib 2.29 yet, can't update in Debian [09:56] need a reboot, bbiab [10:00] hm, dbus is a rather complex merge, that'll take me a bit [10:05] didrocks, do you have any clue about bug #806888 [10:05] Launchpad bug 806888 in gnome-session "can only login "classic"" [Undecided,New] https://launchpad.net/bugs/806888 [10:06] hum bug #788710 [10:06] Launchpad bug 788710 in gnome-settings-daemon "gnome-settings-daemon crashed with SIGSEGV in exit()" [High,In progress] https://launchpad.net/bugs/788710 [10:06] session restart, brb [10:17] seb128: answered on the bugmail [10:18] didrocks, thanks [10:19] didrocks, did we figure what happened to nvidia that required to jockey if off and on again? [10:19] or did everybody workarounded it this way and ignored the bug from then on? [10:19] seb128: I pinged the Xorg guy for it, I didn't get any feedback about it [10:20] ok [10:20] great, gtk build done [10:20] time to reboot with the new glib and then with the new gtk [10:20] be back in a few [10:20] seb128: the gstreamer packaging will be in git soon btw, not sure if that helps anybody in ubuntu but if you use git too it should make merging new debian versions easier ;) [10:26] it would be so good if there were official ubuntu git hosting [10:26] debian merges between git branches are really nice === MacSlow is now known as MacSlow|lunch [11:03] re [11:04] ok, I'm happy with dbus; let's upload and break the world! [11:13] * soren holds off on upgrading his laptop to Oneiric for another couple of days [11:13] pitti: hi, a gtk3 porting question, it appears that for gtk_text_iter_forward_search() I get a tuple as the return but I would expect a triple actually (found, start, end) - and it crashes for me in update-manager, but that might be releated actually I guess. I will write a test case next [11:15] mvo: right, for (return_bool, match_start, match_end) [11:15] mvo: do you happen to have a small reproducer? or that's what you are writing right now? [11:15] pitti: I'm writing it now [11:26] in the last couple of cycles (natty and oneiric), I always have my eth0 connection shown as unmanaged by NM [11:26] pitti: can it be as simple as http://paste.ubuntu.com/639423/ ? testcase is here: http://paste.ubuntu.com/639424/ [11:26] mvo: that looks very plausible indeed [11:26] pitti: I wonder if match_start, match_end also need to be nullified? but none of the other code seems to do that, its a bit puzzling that I get apparently valid types back that then crash when trying to access them [11:27] pygobject should care about passing initialized pointers [11:28] ok [11:28] mvo: hm, the return type in the gir looks correct, though [11:28] rodrigo_: it seems you pushed gnome-destkop3 to lp:ubuntu/gnome-desktop3 (with some diff from the actual tarball). Can you refresh lp:~ubuntu-desktop/gnome-desktop/ubuntu rather? [11:28] [11:29] whether a match was found [11:29] [11:29] pitti: hm, does the test case work for you? or also crashes? [11:29] didrocks, oh, I thought we used the ubuntu branch for that one [11:30] didrocks, updating the u-desktpop branch now [11:30] mvo: yes, it crashes, and only returns two items [11:30] rodrigo_: thanks :) [11:30] mvo: thanks for the test! looking into it now [11:30] thanks pitti [11:30] pitti: you are right, the gir looks good indeed [11:32] meh, can't built gtk head, needs glib 2.29 and pango 1.29 [11:32] * pitti builds older version [11:32] when a match is found it does not crash (unsuprisingly I guess) [11:33] mvo: oh, so the returned textiters ought to be None? [11:33] but the C function might not actually write stuff into the result values [11:33] yeah, that is what I suspect [11:33] didrocks, hmm, I think the ubuntu-branch is for gnome-desktop2 [11:33] still, makes me wonder why match is not returned [11:33] didrocks, sorry, the ubuntu-desktop branch [11:33] mvo: i. e. when the return code would actually work, one could avoid the crash by checking it [11:34] mvo: aah -- it's overridden [11:34] blatant bug there [11:34] rodrigo_: oh, you're right, why not creating a debian/ only for gnome-destkop3? [11:34] so that we have the same workflow everywhere [11:35] mvo: I see the problem, hang on [11:35] pitti: aha, nice. thanks, once you have the fix, please give me the diff so that I can build it locally to continue my porting :) [11:35] didrocks, not sure, let's ask seb when he's back [11:35] pitti: no rush [11:35] rodrigo_: I think he's more found of no full source :) [11:35] (False, , ) [11:37] !!! [11:37] mvo: the override was done to not actually return a success value [11:37] mvo: so in the real fix it'd return (None, None) instead [11:38] apparently that was done to be more pygtk compatible, or more Pythonic [11:38] mvo: would that work for you? [11:38] sure [11:38] that is the old behavior, if no match is found it would return None iirc [11:38] mvo: can you please patch /usr/share/pyshared/gi/overrides/Gtk.py with http://paste.ubuntu.com/639428/ ? [11:41] pitti: works now, but iirc the old behavior was a single None, I don't changing that though [11:41] mvo: oh, can do as well [11:41] haven't committed yet [11:41] pitti: hold n a sec, I will double check with the old code [11:42] mvo: hmm, http://www.pygtk.org/docs/pygtk/class-gtktextiter.html#method-gtktextiter--forward-search doesn't specify [11:43] pitti: yeah, I noticed that as well [11:43] * pitti tries with old gtk in ipython [11:44] pitti: just started the old u-m, there its a single "None" [11:44] confirmed [11:45] >>> iter.forward_search("h", 0, end) [11:45] (, ) [11:45] >>> iter.forward_search("foo", 0, end) [11:45] >>> [11:47] mvo: http://git.gnome.org/browse/pygobject/commit/?id=fc5c869486c7f6929e285ea7a86623ec41ecd9bd [11:50] mvo: cherrypicked into 2-28 branch as well [11:50] mvo: how soon do you need this in oneiric? [11:52] pitti: there is still a bit of work to make u-m go, so not too soon [11:55] FYI: seb128 is having issues connecting to IRC he's coming back shortly [12:10] hey pedro_ [12:11] pedro_, so, what's needed for having a gnome-control-center bug fixing day? [12:11] hello rodrigo_ [12:11] rodrigo_, bug fixing day? or a bug day? [12:11] hmm, hopefully both [12:12] both at the same time :) [12:12] we just need to schedule it [12:13] rodrigo_, what about next thursday? [12:13] just checked at the planning page and there's nothing scheduled for it [12:13] pedro_, sounds perfect [12:13] ok i'll go ahead and schedule it for now [12:13] chrisccoulson, maybe we should include firefox also? [12:13] i'll ping you back later to set up the instructions [12:14] like what would you like to do to bugs with patches etc [12:18] ok [12:20] time for lunch, bbl [12:22] rodrigo_, a firefox bug fixing day? [12:22] you mean that there are bugs in firefox? ;) [12:22] didrocks, I've updated the indicator vala bindings in my branch, so robert can grab them there [12:23] chrisccoulson, well compared to unity there's probably none :-P [12:23] lol [12:23] * pedro_ runs and hide [12:24] pedro_, WHAT! [12:24] njpatel: oh excellent! nice work dude :) [12:24] pitti: my next pygi question, http://paste.ubuntu.com/639447/ does not produce a popup for me when it should and I'm staring at it with no idea what is wrong. I noticed the popup() itself is no longer used, but popup_for_device [12:25] mvo: pygobject has an override to implement popup() [12:25] so both should work [12:25] and c-j uses it too, so it works in general [12:26] yeah, back! [12:28] wohoo! [12:28] seb128: FYI, james_w | pitti, should be up to date in a few minutes [12:28] pitti, great! [12:32] mvo: hmm, boggle -- by and large the same code works for computer-janitor.. [12:32] mvo: except that there are some actual widgets involved [12:34] wb seb128 [12:34] seb128: pedro was telling that you were slacking ;) [12:34] didrocks, re ;-) [12:34] oh, first he kicks in my door and now he calls me a slacker! [12:34] heh :-) [12:35] seb128: about gnome-destkop3, we have no branch up to date in ~ubuntu-desktop. Is it wanted? (only lp:ubuntu/gnome-desktop3, which contains ful source) [12:35] full* [12:36] I would be in favor to create a debian/ only one in ~ubuntu-desktop [12:36] didrocks, we are on sync with debian, it's annoying to mainain a vcs for those [12:36] didrocks, do you need to patch it for something? === MacSlow|lunch is now known as MacSlow [12:37] didrocks, that and the source is small enough that fetching the source is ok [12:37] something that we could patch in Debian, too?/ [12:37] seb128: are we in sync? 3.1.3-0ubuntu1 [12:37] pitti: yeah, I tried to put it on a treeview, same result, I looked at the c-j code and can't spot a difference (that is significant) [12:37] didrocks, that's because we started on 3.1 before them [12:37] didrocks, but we have been in sync until we updated and we plan to sync again when they update [12:37] mvo: the menu items aren't connected to anythign, but that should hardly make a difference [12:38] didrocks, it's annoying to keep a vcs uptodate when we switch between sync and updates [12:38] didrocks, it means we have to manually commit all debian updates while we are on sync [12:38] pitti: yeah, in my real code (in u-m) they are and I get the same behavior [12:38] seb128: well, same story for most of packages we had in ~ubuntu-desktop then, most of the branchs were just because we were ahead debian. If the politic changed, I'm ok with no vcs [12:38] would just prefer it was announced :) [12:38] mvo: hm, and can't test with gtk2 :/ [12:38] didrocks, not true, most packages have a least a lpi diff [12:39] didrocks, well feel free to update lp:~ubuntu-desktop/gnome-desktop3/ubuntu [12:39] seb128: I don't care, just that we should have a coherent story there [12:39] or is it lp:~ubuntu-desktop/gnome-desktop/ubuntugtk3? [12:39] didrocks, we will not have a coherent story until everybody use UDD [12:40] pitti: ha! I think I have it, its a ref count issue, if I assign it to something like self.menu its fine [12:40] until then, maintaining Vcs-Bzr: is the best we can do, I think [12:40] mvo: aah [12:40] mvo: because "menu" goes out of scope right after the handler [12:40] mvo: that's why it works in c-j [12:40] yeah [12:40] exactly [12:40] d'oh [12:40] pitti, well the question there is, should everything we change for desktop be in ~ubuntu-desktop or can we use some UDD sources [12:41] we already use UDD for some bits [12:41] pitti, which didrocks seems to have an issue with because it's not consistant [12:41] I think in general the ~u-desktop ones are still more comfortable, but for the situation you described UDD migth be better [12:41] the ~u-desktop ones suck for merging [12:41] seb128: just that some branches have to be pushed at some place, some not [12:42] I need to leave now, early start/early finish day [12:42] see you tomorrow! [12:42] yeah, we have a mix of UDD and team vcses in the archive [12:42] seb128: and you really told a lot "I don't want to download the full source everytime because of network bandwidth", but if you changed your opinion on it, it's ok, just tell it clearly :) [12:42] pitti, see you! [12:42] see you pitti [12:42] UDD just sucks differently [12:42] see you pitti [12:42] it's better for merging, and worse for anythign else IMHO [12:43] didrocks, right, and that's still the case [12:43] didrocks, but gnome-desktop3 is [12:43] - small [12:43] - on sync with debian most of the time [12:43] - can be downloaded by apt-get source, and uploaded the old way without vcs [12:43] ok, now I know it, that's all, I wasn't aware that we will have to check for those :) [12:44] we don't have good stories for things we sync [12:44] better explicit than implicit [12:44] right [12:44] we have a gnome-desktop3 vcs for it [12:44] I just can't be bother to commit each debian sync there [12:44] if you want to do it by my guess [12:44] guest [12:44] no need to be angry, it was just a question… [12:45] I'm not angry, it just seems a lot of talking and I don't get what you want from the discussion [12:45] the vcs is still in the ubuntu-desktop namespace with debian dir only [12:45] it just got outdated because we went on sync with debian [12:45] * mvo is angry at pygi [12:45] which flushed the control vcs info [12:45] an answer to my question? "about gnome-destkop3, we have no branch up to date in ~ubuntu-desktop. Is it wanted? (only lp:ubuntu/gnome-desktop3, which contains ful source)" [12:45] so I guess I have it now [12:45] which is unfortunated but I've no reply on how to handle those [12:46] that's it, no need to make such a lenghty discussion for it… [12:46] short reply "I don't care either way, I want it back in sync with debian when we can" [12:46] ;-) [12:46] if we want a vcs it should be lp:~ubuntu-desktop/gnome-desktop/ubuntugtk3? [12:46] -? [12:47] but I'm fine updating it with apt-get source, dput the old way while we break the syncs because we update before debian [12:47] which is what we have been doing there [12:48] so ok, just wanted that everyone is on the same page to chase for commits :) [12:48] is it possible to forward from LP to bugzilla now? bug #806949 is a candidate [12:48] Launchpad bug 806949 in gtk+3.0 "Refcount problem with pygi" [Undecided,New] https://launchpad.net/bugs/806949 [12:49] mvo, define "forward"? you have for sure a bugzilla account and file GNOME bugs before? [12:50] mvo, or you mean "click a button and get lp to do that for you"? in which case "no" and "we don't want that, it would mean any user could spam bugzilla easily with low quality bugs" [12:50] ok, fair enough, then I will manually forward :) [12:51] mvo, thanks ;-) [12:52] pitti: ping? === m_conley_away is now known as m_conley [13:03] mvo, how is your example working? [13:03] mvo, i.e what do I do to see the issue? [13:03] mvo, ignore me, got it [13:03] it doesn't display the menu as it should [13:04] seb128: btw, with your super bleeding fast internet, dconf-qt is in new, apart from the qmake -> cmake dep change and the typo you noticed, it should be the same [13:04] seb128: yeah, it should display it, but it only does if I keep a reference to menu [13:04] took me a good while to figure that out [13:04] mvo, right, I can confirm with yesterday's tarballs so feel free to forward it [13:04] didrocks, ok [13:04] didrocks, is there any hurry? [13:04] if I now find out why my gtkbuilder UpdateManager.ui has a odd resize behavior I'm *mostly* done with the port I think [13:05] seb128: not at all, you have at least 2 weeks I would say :-) [13:05] mvo, \o/ [13:05] seb128: yeah, for once, it's not urgent ;) [13:05] didrocks, great, I will just restart my session once now, if everything works ok I upload gtk3, glib, pango [13:05] then I look at it [13:05] brb [13:05] seb128: dbus and gtk/gtlib in the same upgrade? I want that! :-) [13:06] glib* [13:06] ;-) [13:06] didrocks, we need to make those users complaning that oneiric is too stable happy [13:06] ;-) [13:06] brb [13:06] seb128: agreed, time to push compiz as well! [13:06] good reboot :) [13:41] pedro_, I cleaned a bit the desktop-bugs subscriptions list [13:42] Anyone feel like NEWing itstool? [13:43] hey mterry [13:43] mterry, can do [13:43] seb128, thanks! [13:43] seb128, cool, did you added a new one? like telepathy,etc? [13:44] pedro_, not yet [13:44] seb128, just wondering in case i need to update the 'other packages' page [13:44] pedro_, but I can do that next [13:45] didrocks, dconf-qt newed [13:45] seb128: thanks :) [13:49] mterry, NEWed [13:49] mterry, is that used in GNOME3.1? [13:49] seb128, yeah, it's used for yelp-tools, which is replacing gnome-doc-utils [13:49] ok [13:50] I'll file a MIR once I get yelp-tools in. So if you can stick around to NEW that too... ;) [13:50] ;-) [14:11] seb128, yelp-tools ready for NEW [14:11] mterry, ok [14:24] didrocks, ^ do you have time for a NEW review? just did two and I want to finish some of the updates I'm working on [14:25] will take it in a bit otherwise [14:25] seb128: sure [14:25] thanks [14:27] heh, just used brasero for burning an iso and it worked [14:28] tell it to pedro_ he keeps saying it's broken! [14:28] yeah, pedro_ ^^ [14:28] it worked? wow [14:29] rodrigo_, was that like a release iso ? or did you make an iso of music files, etc? [14:29] for the reports i've seen people complain about the making projects thing, like adding normal data instead of selecting an image and burn it [14:31] mterry: how did you determine the copyright? there is no header, and GPL2.1+ and LGPL2.1+ are both shipped [14:31] pedro_, a distro iso [14:31] didrocks, from looking at the .in files in tools/ [14:31] pedro_, in fact, I remember having problems burning a music iso [14:31] rodrigo_, ah that's why it worked :-P [14:31] didrocks, only thing with a header. The COPYING files seem copied from gnome-doc-utils (very obvious if you look at the COPYING file itself) [14:32] rodrigo_, yeah things like that aren't really working 'that' fine [14:32] not smooth as k3b for example [14:32] mterry: waow, interesting that's in a .in file ;) even licencecheck ignore it :) [14:32] mterry: Format: http://svn.debian.org/wsvn/dep/web/deps/dep5.mdwn?op=file&rev=166 [14:32] is that wanted? I remember you chasing me to use the old url :p [14:33] didrocks, that's the spec'd url isn't it? [14:33] (though I do think it's ugly) [14:34] mterry: I think there is a shorter now (and that one point on nothing) [14:34] one sec [14:34] http://dep.debian.net/deps/dep5 ? [14:34] mterry: that's what dh_make is creating [14:35] didrocks, but look at the actual spec (search for svn.debian) [14:35] didrocks, dh_make is creating an unversioned format, which the dep5 authors didn't want (they liked the rev=XXX part of the URL) [14:36] In reality, who cares, but it's odd that there's a disconnect there (especially that the spec'd url is a bogus one) [14:36] mterry: shouldn't it be http://anonscm.debian.org/viewvc/dep?view=revision&revision=174 ? [14:36] as 166 isn't working [14:36] seb128: the goal is to GIify glib and gobject as well, right? I just found that GLib.markup_escape_text() needs the len of the string, this is so un-pythoninc [14:36] mterry: acking your package, I hate those url too, if you can just fix later so that it points on something which works :) [14:37] didrocks, but anything else wouldn't be to-spec! :) [14:37] mterry: heh ;) [14:38] didrocks, well, I'm serious. I don't know as if we have any parsers right now, and they will probably be generous in what they accept, but if I start changing the URL to non-spec'd things, they will have a harder time of it [14:38] didrocks, the format line is more for parsers than humans [14:38] how can I detect from code if the current distro is ubuntu? I only see /etc/debian_version [14:39] (i think) [14:39] mterry: I let it go, but the priority should be extra IMHO [14:39] rodrigo_: lsb_release -i works [14:39] rodrigo_: lsb_release -i should tell [14:39] heh :) [14:39] mterry: right, the thing is that the link it leading to an error on the vcs right now [14:39] ;) [14:39] cyphermox, mvo: ok, thanks [14:40] (or -i -c if you want to get rid of prefix) [14:40] mterry: I'm still puzzled about what the right format/version we should use in the url-spec [14:40] mvo, yes, they add overrides in pygobject when needed [14:40] didrocks, extra is better, thanks. will change next upload [14:40] mterry: great, I locally built the bin package and they are good, I'll ack them once built [14:41] didrocks, me too. looking at the text again, it seems to not say that you must use that format it provides. Just says "url of the spec" [14:41] didrocks, so if you got the spec from mycoolspecs.com, I guess you could put that in... [14:41] mterry: yeah, it can be widely interpreted ;) [14:41] * didrocks tries mycoolspecs.com, maybe it exists! [14:41] mvo, hum, are the lsb tools still writen in python? wouldn't be easier to read /etc/lsb... than spawning a python interpreter? [14:41] GLib.timeout_add_seconds … introspectable="0" [14:42] *sigh* [14:42] mterry: no, time to register it? ;) [14:42] didrocks, :) [14:42] seb128: well, there is "import lsb_release" and get_lsb_information() in it, that should be fine as well [14:43] seb128: reading the file is not ideal as it *may* change in the future (also I doubt it ever will ;) [14:43] mvo, well my point was that from a C file you probably to read the text file rather than running a python interpreter [14:43] mvo, seems way slower to run python just to get the distro [14:44] sure, its slower. the file format is not guaranteed that is why its not encouraged. if the speed/mem is a issue, then it should be fine to read it directly [15:13] didrocks, maybe I asked for NEW too soon? yelp-tools finished building and says it is in NEW queue [15:14] mterry: you are quite quick ;) I checked for binary new like 10 minutes before and it wasn't there [15:14] mterry, what is in NEW? [15:14] didrocks, oh that's a separate NEW check I guess. itstool has same issue [15:14] There is a check for source + a check for binaries? [15:14] NEW? [15:14] yes [15:14] seb128, makes sense [15:14] mterry: rightn you have source NEW, then if acked, the package build [15:14] that's why you hit new on soname changes [15:14] or when adding a binary [15:15] mterry: and we have to ack for every arch [15:15] (binary NEW) [15:15] acked yelp-tools for i386, but still waited for the other archs [15:17] didrocks, don't wait too much it's arch all ;-) [15:17] you might still be there next year waiting otherwise [15:17] seb128: no, I will keep refreshing the page! :) [15:17] ;-) [15:17] we will pick you up for UDS [15:17] don't worry [15:17] need to be rebooted? :-) [15:18] lol [15:18] :) [15:18] tedg, so, how is the indicator-session user switch merge request going? [15:18] tedg, going to land for today's tarballs? ;-) [15:21] bah, another libindicator transition [15:21] stop those! [15:21] ooh, http://tech.slashdot.org/story/11/07/07/140204/Thunderbird-Unseats-Evolution-In-Ubuntu-1110 [15:21] not that i normally read /. ;) [15:22] seb128, Yeah, this is the fallout from avoiding one last cycle and having two scroll functions. A broken one and a correct one. We're finally dropping the broken one. [15:22] chrisccoulson, you clearly don't have enough work if you read slashdor [15:22] t [15:22] chrisccoulson, but I can fix that for you! ;-) [15:23] seb128, i don't read it. m_conley tweeted it ;) [15:23] oh, you read tweets, that's better :p [15:24] tedg, ok, what about the user switching question? ;-) [15:24] didrocks, do you remember what error you got when trying to delete the gnome-desktop3 merge request? [15:24] didrocks, it worked for me [15:25] seb128: no, I didn't delete it on purpose, I couldn't "reject" [15:25] oh ok [15:25] you wrote "Set it as WIP and add a comment about it as couldn't delete it." in your email [15:25] which got me confused [15:25] seb128: I didn't delete the MR, because I think that james_w wants maybe to fix this and that we don't keep unused cruft branch [15:25] well it's deleted now :p [15:25] argh, I'm really tired I guess :/ [15:25] seb128: well, there are plenty of other examples ;) [15:25] ;-) [15:26] so that's fine! [15:26] didrocks, can you binary-NEW itstool too? (ubuntu2 is now arch-all so should just need i386) [15:26] mterry: well, I'll bin new them then, let me check the package [15:27] didrocks, if you love working on yelp-tools/itstool, have I got a MIR for you! [15:28] * didrocks fires up a dconf-qt MIR to protect himself :) [15:29] Nothing like a nice MIR arms race [15:29] mterry: subscribe me to them, I'll have a look at them tomorrow morning (should be good or yelp-tools, need to look at itstool) [15:29] heh, indeed :) [15:29] for* [15:29] didrocks, done. they are both super trivial [15:30] seb128, ronoc is working on the user switching. I don't think he'll be done today, but it's in good shape. Definitely next week. [15:31] tedg, you saw robert_ancell merge request right? [15:31] tedg, i.e conor is not redoing the work right? [15:32] mterry: will do that tomorrow morning before going to the train station :) [15:32] didrocks, sure, no rush [15:32] seb128, Yes, he shouldn't be. [15:32] ok, great [15:32] waow, I can do what I wanted to do in 4 lines of python to merge two images :) [15:34] hum [15:34] is indicator-messages not indicating what is running for others as well? [15:34] is the > [15:34] is->ie [15:35] seb128, yeah it's gone for me [15:35] yup, confirming [15:35] ok, good, not only me ;-) [15:35] ok, enough packaging for today [15:35] I see an indicator in the sound menu, so probably not a gtk3 transition issue... [15:35] let's take an hour of bug triage and filing [15:36] self.setdata() [15:36] File "/usr/lib/python2.7/dist-packages/duplicity/path.py", line 493, in setdata [15:36] self.stat = os.lstat(self.name) [15:36] OSError: [Errno 107] Transport endpoint is not connected: '/home/seb128/.gvfs' [15:36] booubouh deja-dup! [15:37] mterry, deja-dup hates me [15:37] he does that when I click on "backup now" [15:37] seb128, ugh! I've gotten reports of that error and I can't figure it out. Are you connected to internet? [15:37] it does it in a consistent way [15:38] mterry, I guess so, can you read what I type on IRC? [15:38] (it's the box I'm working on) [15:38] seb128, :) [15:38] mterry, well it's doing it every time I click the button [15:38] so I can get you debug infos if you have questions [15:38] ls -l ~/.gvfs? [15:39] ls: cannot access .gvfs: Transport endpoint is not connected [15:39] OK, so it's not just deja-dup being crazy [15:39] is .gvfs supposed to be mounted when you didn't use gvfs? [15:40] (duplicity shouldn't be stat'ing that folder anyway, since we exclude it, but...) [15:40] seb128, I don't know. It's maintained by gvfs-fuse I think... [15:40] hum [15:40] it works on a fresh login on my test box [15:40] so I guess gvfs got screwed during the course of the day [15:41] seb128, yeah, whenever I get a report of this, a reboot fixes things [15:41] so gvfs bug and duplicity,deja-dup not being robust to it [15:41] seb128, but haven't been able to track it down [15:41] mterry: acked itstool -0ubuntu2 btw [15:42] mterry, ok, trivial to get [15:42] didrocks, thanks [15:42] mterry, kill gvfs-fuse and try using deja-dup [15:42] seb128, fascinating [15:42] just got it on my test box [15:44] seb128, sounds a bit like bug 212789 [15:44] Launchpad bug 212789 in gvfs "gvfs fuse mount is not functional after logout and subsequent login" [High,Fix released] https://launchpad.net/bugs/212789 [15:44] mterry, yes [15:45] mterry, right, there is a gvfs issue there, but as you said deja-dup or duplicity should ignore .gvfs anyway [15:46] mterry, do you want a bug report or did you get some about it already? [15:46] seb128, I can re-open an old one [15:46] seb128, which process do you kill? i don't have a fuse process running [15:46] mterry, gvfs-fuse [15:46] seb128, ah, didn't even have it installed... [15:46] ah ;-) [15:48] mterry, you can run /usr/lib/gvfs/gvfs-fuse-daemon .gvfs [15:50] Hrm, not sufficient to trigger. Let me try to connect to a server === JanC_ is now known as JanC [15:52] seb128, nope. even after connecting to something and killing the fuse daemon, ~/.gvfs is clean for me [15:52] * mterry wishes he could break things like seb128 can [15:53] mterry, how do you kill the daemon? [15:53] killall [15:53] let me try -9 [15:53] I did -9 [15:53] not sure if that makes a difference [15:53] seb128, yup, it does :) [15:53] ;-) [16:07] seb128, bug 794576 btw (thanks for your help, now I'll just dive into duplicity and figure out why we're stat'ing it) [16:07] Launchpad bug 794576 in deja-dup "Transport endpoint is not connected" [High,Confirmed] https://launchpad.net/bugs/794576 [16:07] mterry, thanks [16:09] pedro_, can you upstream bug #805311 [16:09] Launchpad bug 805311 in gnome-bluetooth "gnome-control-center crashed with SIGSEGV in g_type_instance_get_private() (device_services_changed)" [Medium,New] https://launchpad.net/bugs/805311 [16:09] I can confirm here [16:09] eh, what is wrong with using setting_name in gsettings instead of settings-name ?!? why was this restriction added? [16:09] mvo, consistency I guess [16:09] mvo, gconf was a mix of _ and - which is confusing [16:12] ok [16:12] at least there is a reaosn :) [16:15] pedro_, don't bother I do it [16:15] seb128, i'm about to click on submit :-P [16:15] pedro_, ok, great ;-) [16:19] pedro_, the bug report was on 3.0.2 btw, not 3.0.3 but I checked it still happens [16:19] pedro_, the stacktrace lines might just be different [16:25] jibel: thanks for your report! I'm sure it will be really useful :) [16:27] didrocks, stop encouraging spammers! [16:27] seb128: I agree it's quite depressing sometimes, but needed :-) [16:28] ;-) [16:28] but see, there are bugs for everyone! ;) [16:28] didrocks, oh, new zg and zg-datahub uploads to debian [16:28] seb128: yeah, I saw that, I'll try to get the merge tomorrow (it has some reduced mem consumption) [16:28] if not, would you like to handle that? :) [16:29] (still feeling a little bit sick, will stop working soon) [16:30] didrocks, thanks for introducing bugs! at least there's something to write in the report ;-) [16:30] didrocks, the merge is a one liner to add one char to a build-depends ;-) [16:31] didrocks, if you don't manage to do it tomorrow I think I can find an hour for it :p [16:31] didrocks, we should ask rainct to add | to the build-depends in debian so we can sync [16:31] seb128: the thing is that with all the incoming changes, it still need some testing [16:31] seb128: already done, but I think he forgot about it, will try to spam him again :) [16:31] it will maybe fix the segfault that apport keeps reporting on loging ;-) [16:32] jibel: it's just to ensure you have some work :) [16:32] seb128: this one is fixed AFAIK, yes [16:36] didrocks, let me try the new zg [16:36] seb128: oh, you want to do it now? nice :) [16:36] didrocks, well I can build and install it if you need testing [16:36] seb128: datahub is just mterry's double patch IIRC, that could be set to debian I guess :) [16:37] didrocks, not the nodisplay one, that is ubuntu-specific for now [16:37] didrocks, (GNOME didn't want our patch to gnome-session to respect the key) [16:38] mterry: hum, we can maybe put the change in debian/patches/ubuntu.series [16:38] mterry: or do you want to update zg-datahub? ;) [16:38] didrocks, sure. and really, they could accept the patch, it wouldn't hurt anything in debian to have nodisplay [16:39] let me talk with rainct if he's there [16:40] ok [16:41] mterry, oh, they didn't want to? why? [16:41] seb128: we can forget about zg-datahub, it's just an upload to make it into unstable with a bug number [16:41] seb128, they felt it was an overloading of the spec, that nodisplay was only for menus in the panel [16:41] so if it's only zg, I can double check, should be quick [16:42] seb128: just to ensure we don't duplicate the work, should I do it now? ^ [16:42] desrt, what was the issue ricotz was having building glib during the rally? one of the gdbus tests hanging? [16:42] mterry, ok [16:42] didrocks, yes, please do [16:42] rainct isn't there, I'll send him an email for the additional dep [16:48] seb128: the zg change isn't the fts one, it's just doko making a build-dep change not reported to debian [16:48] didrocks, right [16:49] ok, I was on the fts naming which diverged :) [16:49] I though we were speaking about the same thing when I said a one line, one char change ;-) [16:49] sorry about the confusion [16:49] no worry, anyway, both should be in debian to enable sync [16:49] both? [16:50] is fts still required? [16:51] yeah, it's used to scan documentation content [16:52] but it's called zeitgeist-extension-fts in ubuntu (where it was first) and zeitgeist-fts-extension in debian [16:52] so, we have a transitional package in ubuntu [16:52] ok [16:56] session restart, brb [17:00] seb128: no crash at logout \o/ [17:00] nautilus does, though ;) [17:14] seb128: why did you assign me the bug about session saving not working? it's disabled [17:15] woot... i have a power indicator in my panel now :) [17:16] * kenvandine is much happier [17:16] heh, my power indicator will only have 1 state [17:16] "EMPTY" [17:16] ;) [17:21] seb128: I added a rationale, unassigned and call for help [17:42] hey tedg [17:44] kenvandine, Howdy [17:44] tedg, now i just need a release tarball for indicator-power [17:45] packaging is done [17:45] tedg, i did just propose a branch which needs to be merged before releasing though [17:45] https://code.launchpad.net/~ken-vandine/indicator-power/packaging_fixes/+merge/67229 [18:29] seb128, indicator-power uploaded, in source NEW [18:47] kenvandine, seems fine from a glance [18:48] kenvandine, but I think I will just keep the indicator stack in new for the night and get those in at the start of the day tomorrow [18:48] seb128, indeed [18:48] seb128, the whole stack is being updated again [18:48] for new libindicator [18:48] right [18:49] btw speak of libindicator [18:49] do you know why it has no gir or vapi? [18:49] seb128, however, the current indicator-power doesn't use the new libindicator [18:49] robert_ancell was asking about it this morning [18:49] nope [18:49] he needs it for lightdm [18:49] humm [18:49] njpatel said he would get one updated today [18:49] njpatel tweeted about updating libindicator vapi this morning [18:49] right, neil picked up on that [18:50] cool [18:50] still I'm wondering why there was none, i.e if there was a reason, I know there was some gir namespacing issues for some sources [18:50] didn't make the release [18:50] i think just that nobody has needed to create a new indicator in vala [18:50] i guess [18:51] tedg, ^^ [18:51] seb128, libindicate-gtk still has the namespace problem [18:51] * kenvandine grumbles at tedg AGAIN :) [18:52] seb128, it's done [18:52] njpatel, \o/ [18:53] there isn't any as it's way difficult to make it part of the build, it needs to link to the installed library to make the vapi....not sure if the gi is enoug [18:53] njpatel, can you drop an email to robert_ancell about it? [18:53] but I'm going to see if we can patch vapigen to be more clever [18:53] seb128, sure! [18:53] njpatel, thanks ;-) [18:54] done :) [18:58] tedg, does today's indicator-session require libindicator6? [19:00] kenvandine, Yes [19:00] ok [19:00] damn... power just flickered here [19:00] damn thunderstorm [19:00] lol [19:00] if i drop off line, it is because my power went out :) [19:00] when you wrote that I though you were speaking about the indicator [19:01] haha [19:01] * kenvandine is very happy to have power displayed in the panel [19:01] We're going to have to get kenvandine to test the UPS features. [19:01] less happy about a power outage :) [19:01] we should put a "i" before the names to make clear it's indicator [19:01] ipower [19:01] my cable modem isn't plugged into a UPS :) [19:01] haha [19:01] kenvandine, I did that to my DSL modem and router. Too many outages here. [19:01] Sweetshark: hello [19:02] tedg, it is really rare here [19:02] Laptop on battery, everything else on UPS. [19:02] pitti, you called it a day, go back to !work ;-) [19:02] tedg, i should though... just nothing else in the same room needing a UPS [19:02] seb128: still in a train :) [19:02] kenvandine, I picked up a cheap one when Circuit City went under. [19:02] pitti, 3g for the win? ;-) [19:02] seb128: or rather, just went into it -- before we were travelling in a car, where a laptop is impractical [19:03] kenvandine, Won't last long, but they don't draw much power. [19:03] (we went with a colleague of my wife to Leipzig, and now doing the remaining bit to Dresden with train) [19:03] seb128: yeah, the builtin antenna of the x201 is marvellous [19:04] pitti, (oh ok, are you at work tomorrow or having a day off?) [19:05] seb128: I'm officially at work [19:05] ok [19:07] seb128: we went a day early as the uni of my wife does a ceremony/party for this year's graduates [19:07] seb128: so I'll work from Dresden tomorrow [19:07] ah ok [19:23] so, see you tomorrow! good night [19:25] 'night pitti === m_conley is now known as m_conley_away [21:32] kenvandine: reading some back scroll...you have power indicator? yay! when do the rest of us poor slubs get one ;) [21:38] jasoncwarner_, it needs an archive admin to let it in :) [21:38] seb said tomorrow [21:38] ah, ok [21:39] anyone here know why I ended up in Unity-2d after my update this morning? I didn't see anything screaming at me like X/Mesa or NVidia, though, to be fair, it was a scan of the list :P [21:39] happy to be in unity 2d though...reporting some bugs now! [21:43] hey jasoncwarner_ [21:43] what up, cyphermox [21:43] ? [21:43] heh [21:43] nm-applet will create menu only when you click it? [21:44] I'm satisfied enough with the code to upload shortly [21:45] say again? I click on it and the menu is created, yes [21:45] yes [21:45] usually it gets recreated every few seconds, or after we get scan results, etc [21:45] now it gets refreshed when you click on it, just before displaying [21:49] oh, ok [21:49] :) [21:49] gotta reboot...removed nvidia drivers to see if that was problem with 3D...be back [21:56] jasoncwarner_, lp #807209 perhaps [21:56] Launchpad bug 807209 in nvidia-graphics-drivers "Lost glx after first upgrade from oneirc alpha 2 install" [High,New] https://launchpad.net/bugs/807209 [22:04] bryceh: yeah, looks like it [22:04] and also, my system seems very laggy now....(debugging)