[00:00] ok [00:00] my fault, sorry guys [00:00] np [00:02] heya [00:02] * bryceh grabs coffee [00:07] * RAOF will need to get milk before coffee is a viable option. [00:08] heh [00:09] TheMuso bryceh RAOF robert_ancell [00:09] back...ready for https://wiki.ubuntu.com/DesktopTeam/Meeting/2011-04-05 [00:09] ? [00:09] :) [00:09] Wooo! My long, long delayed patch to make dh_clideps error out when it can't resolve dependencies has landed! [00:09] ayup [00:09] RAOF: nice. [00:10] TheMuso: Yeah. Now things will fail to build rather than fail to run on systems without the appropriate -dev packages installed :) [00:10] RAOF: nice :) (not that I know what that is, but I'm happy you're happy) [00:10] Its mono related. [00:10] anyway...jumping into meeting [00:10] [TOPIC] X.org [00:11] bugs consuming most of the time these days [00:11] Bugs, bugs, bugs, bugsā€¦ [00:11] I've been focusing heavily on -intel gpu lockup bugs since they're still coming in at a high rate [00:12] mostly been trying to push them upstream quickly, so we can get patches to give to the kernel team [00:12] I'm testing libdrm 2.4.24; it's got a bunch of intel bugfixes which resolve some bugs. [00:12] RAOF, for the one bug we know about, I cherrypicked and uploaded the patch [00:13] 2.4.24 has some other fixes but they look to me less relevant for natty [00:13] I really should hammer my recently acquired intel GPU more... [00:13] and there were some instances of fixes-for-the-fixes ;-) [00:13] But when at home and working, I tend to prefer my desktop... [00:14] RAOF, anyway so I'm wondering if maybe we should just hold off on updating libdrm and just pull in specific fixes if/when needed? but I don't have strong feelings [00:14] bryceh: Yeah. Some of the 2.4.24 fixes are for new userspace on old kernels, which we don't much care about. [00:14] today I also went through all of the -ati bugs and pushed about half upstream, and made a pass through -fglrx but most bugs there are invalid things from weird self-installs or whatnot [00:15] I thought things had to be rather tightly in sync for intel... [00:15] TheMuso, sometimes [00:15] If you want it to *work* :) [00:15] TheMuso, libdrm 2.4.24 is mainly just a small handful of bug fixes [00:15] Ah ok. [00:16] hey bryceh, is the new fglrx driver included for those systems? I have a system running radeon and would like to try latest binary driver [00:17] jasoncwarner, yes you should be able to update to latest and then use jockey to install fglrx [00:17] since fglrx arrived with so little time before beta1 we were able to get it in, but not update jockey [00:17] bryceh: awesome! [00:17] thanks [00:17] so you need to update past beta1 and then can install it normally [00:18] cool...I'm upto date as of this morning so I should be fine...computer is media center upstrairs so I'll give that a go when we are done here [00:18] anyway, the rate of incoming bugs is starting to exceed my capacity to keep up with them [00:18] http://www.bryceharrington.org/Arsenal/Reports/ubuntu-x-swat/totals-natty-workqueue.svg [00:18] help would be appreciated. I figure by beta2 the rate will just be too high to be able to stay on top of the bugs any further [00:20] I'll do some more bug work, then. [00:20] Yikes! Ok...anyone that can give bryceh a hand, that would be awesome...seems that trend line is going crazy [00:20] That's a terribly impressive graph, though. [00:21] RAOF, thanks; in particular I haven't kept up with mesa, nouveau, or input drivers. [00:21] although numbers for those are small [00:22] I figure once we shut off the apport gpu and crash hooks that'll ease a little bit of pressure [00:22] but probably should wait until post-beta2 to do that [00:22] Actually interesting you should mention that, for natty beta there hasn't been the super influx of audio bugs like usual, so thats good news. :) [00:22] jasoncwarner, anyway I'm rambling now. I'm done :-) [00:23] :) [00:23] cool [00:23] next topic [TOPIC] AOB [00:23] anythign else anyone wants to talk about? [00:24] o/ [00:25] Have other people noticed moderately high CPU usage in Banshee? ~10% when playing with the window open? [00:25] RAOF: ;) you know I did yesterday :) [00:25] * TheMuso doesn't use banshee, there are still a11y issues with that, I really ort to dig into that a bit deeper. [00:25] But I was personally happy with rhythmbox. [00:25] s/was/am/ [00:26] There's no smoking gun in the profile, but since it doesn't seem to occur when the window is closed there are some good leads. [00:27] thanks, RAOF....keep digging..seems weird that it would jump to that much CPU just to play music...weird... [00:27] thanks! [00:27] Sounds like we are about done...Anything else before we call the meeting? [00:27] o/ [00:27] Its mono. What else do you expect? :) [00:27] * TheMuso ducks. [00:27] micahg: ? [00:27] I wanted to ask about webkit 1.4 [00:27] micahg: go ahead [00:28] is robert_ancell planning on working on that? [00:28] micahg, yes, there's a problem with the documentation build, working on that today [00:28] robert_ancell: ok, great thanks [00:28] 1.3.13 that is, still waiting for 1.4.0... [00:29] * micahg thought 1.4.0 would come with GNOME3 this week [00:29] Do I read the release schedule incorrectlyk, or is there no beta freeze for beta 2? [00:29] jasoncwarner: that's it, thanks :) [00:29] TheMuso: apr 11 [00:30] Ah makes sense. [00:30] Thats kinda cool, because it means I have some, if not all of Monday to do work. :) [00:30] actually, I think almost everyone has all of monday [00:30] :) [00:31] ah ok. [00:31] So its EOD Monday. [00:31] * micahg thought freeze was 23:00 UTC monday [00:31] Could very well be. [00:31] Oh, so I'll have time to work on my birthday, too. :) [00:31] lol [00:32] robert_ancell: I'm assuming there haven't been any unplanned ABI breaks for webkit? [00:32] micahg, in 1.3.13? [00:32] RAOF: nice...now none of us have excuses for not remember your bday ;) [00:32] robert_ancell: yes [00:33] micahg, I didn't see any [00:33] jasoncwarner: My thoughts exactly. [00:33] ( RAOF : not saying we will...we just don't have excuses) [00:33] jasoncwarner: It's a cunning ploy! [00:33] robert_ancell: also, could 1.2 be installed alongside 1.4? [00:33] Ok...going to call the official meeting...[END MEETING] [00:34] thanks [00:34] hey robert_ancell [00:34] micahg, looking, I think not though [00:34] seb128, hey [00:34] thanks. [00:34] robert_ancell, just a reminder, if you update gvfs it uses merge-upstream so don't just run dch ;-) [00:35] seb128, yeah, I know! I'll probably just do something wrong though... [00:35] lol [00:35] seb128, so the process is to run merge-upstream, then debcommit -r? or just bzr commit? [00:36] robert_ancell, debcheckout gvfs; cd gvfs; bzr merge-upstream --version 1.8.0 tarball_url [00:36] micahg, no, they would conflict [00:36] robert_ancell, then dch -i, do you packaging tweaks [00:36] debcommit -r, push, upload [00:37] seb128, yeah, I think I keep forgetting the -r, then it doesn't tag and it gets confused [00:37] robert_ancell, if you do gvfs maybe include http://bugzilla-attachments.gnome.org/attachment.cgi?id=185180 [00:37] robert_ancell, it's an upstream patch for bug #682850 [00:38] [00:38] otherwise out of gvfs [00:38] bug 682850 [00:38] https://bugs.launchpad.net/ubuntu/+source/gvfs/+bug/682850 [00:38] no bot? [00:38] seb128, you're the bot today? [00:39] lol [00:39] no :p [00:39] well we can as well wait for that on to land in git [00:39] so don't bother maybe for that update [00:39] robert_ancell: ugh, that will make migrating away from xulrunner a little harder in the stable releases [00:39] micahg, didn't upstream just say "no more xulrunner?". So hopefully everyone will be migrating away from it anyway! [00:40] robert_ancell, other topic while I'm still around, did you figure the documentation index ?index... error thing? I saw you fixed it for nautilus but other components have similar bugs [00:40] seb128, which other components? gedit seemed fine, and I haven't seen any others [00:40] robert_ancell, it was reported on gnome-panel at least [00:41] the issue seemed to be anything accessing documentation from gnome-user-guide (which is V3 now) [00:41] robert_ancell: yes, so I was going to try to backport some of the newer versions of apps based on webkit, but most require some version of 1.3 as the minimum, I was thinking to rename 1.4 to like webkit-1.4 or something in the stable releases so these apps can basically be forklifted in [00:41] micahg, renaming causes so many ongoing issues... [00:42] robert_ancell: well, it would just be for lucid and maverick and only for xul rdepends, I wouldn't want to touch the regular webkit aside from upgrading it :) [00:42] upgrading from the same branch 1.2.x [00:42] robert_ancell, urg, side effect of upgrading I guess, but it means we have user documentation describing GNOME3 on a GNOME 2.32? [00:44] seb128, yes it's a bit of a mess. I'm hoping our doc people will modify it to fit. In the case of nautilus some deprecated feature documentation was gone, but that's not a big issue for us [00:44] it's not like the documentation was uptodate or than users read it anyway... === lamlex is now known as lamalex [02:00] Ah. *That's* why the overlay scrollbars are ugly on banshee! [02:03] Incidentally, if we felt like making banshee start up faster we should really look into AOT compilation for some mono stuff; Banshee spends a bit more than 2 seconds JITing code on my machine. [02:06] youch [02:06] RAOF: Something for next cycle. [02:06] Yup. [02:06] I can only imagine how mono performs on atom CPUs. [02:06] Or arm for that matter. [02:07] Not all of that is during startup, though; that's for an entire run. Much of it will be during startup, though. [02:07] Yeah. [02:07] Yeah. They'd *really* want AOT wherever possible. [02:35] Wow, I wonder if the international issues are biting others on here... Appears to be between US/EU, or somewhere in the US. [02:36] TheMuso: What are you seeing? Unable to resolve canonical.com again? [02:36] RAOF: Yeah. From what I've been reading, there are some major issues with a teer 1 provider in the US, level 3 or another of the big providers. [02:37] They were ok overnight it seems, but are falling over again, as evidence by me no longer being able to resolve ubuntu/canonical.com again. [02:37] ...and my several freenode and other IRC network dropouts. [02:37] Eh. Whatever it is, it's not hitting my internode connection. [02:38] No, since Internode takes other paths out of Oz. [02:38] SOmething which I wish Telstra would do... [02:38] But overall I am on a good bundle with telstra atm for phone/mobile/net, and can get cable so, that makes things difficult if I want to move. [02:39] I'd go internode in a heartbeat if the line I had here was deacent quality/speed. === smspillaz|zzzzzz is now known as smspillaz [02:44] I actually find it weird that for Telstra, we go via the US to get to EU. [02:51] I find it a bit weird that there are different routes for different carriers. [02:51] It's not like we've got a huge web of intercontinental links. [02:51] Yeah. A deacent provider should depend on multiple routes. [02:52] Trouble is, Telstra half owns Reach, one of the major teer 1 providers for Aus, so... [02:52] They aren't likely to use anybody else. === asac_ is now known as asac === Amaranth_ is now known as Amaranth === tjaalton_ is now known as tjaalton === ivanka is now known as ivanka-train [08:02] good morning [08:06] morning didrocks! [08:06] good morning everyone! [08:06] morning pitti [08:06] hey jasoncwarner, how are you? [08:06] pretty good.., you ? [08:07] btw...looks like with you springing forward, and me falling back...the 1:1 is now 2 hours earlier ;) what do you think? want to jump on mumble? [08:08] jasoncwarner: sure [08:08] jasoncwarner: I'm in [08:08] just realizsed...never setup mumble after installing SSD...give me a few minutes [08:10] morning jasoncwarner, pitti! [08:34] pitti, good morning [08:35] pitti, could you take a look at https://edge.launchpad.net/ubuntu/+source/libgtop2/2.28.3-0ubuntu1 [08:41] morning [08:41] rodrigo_: hihi [08:42] hey desrt [08:44] rodrigo_, hey [08:46] hi ricotz [08:47] rodrigo_, feel free to update some ppa packages :), i will stop for now, but i will take care of gedit [08:48] moring mvo [08:48] ricotz: what about it? (sorry, was on phone) [08:48] hey glatzor! [08:48] ricotz: you want it NEWed? [08:48] mvo, I am not sure which lintian tags to use by default. [08:49] meh, LP going down? [08:49] ricotz, I have updated a few in some spare moments, but I'm bug fixing, so going slowly [08:49] rodrigo_: hey [08:49] pitti, hi, yes ;) [08:49] mvo, I already added the option to load distro specific tags files [08:49] hi didrocks, pitti [08:49] rodrigo_: any idea why a gobject_new can crash apart from "no more memory"? https://launchpadlibrarian.net/68241093/Stacktrace.txt [08:49] rodrigo_: the line is child = g_object_new (NA_TYPE_TRAY_CHILD, NULL); [08:50] didrocks, looking [08:50] glatzor: oh, I did not see that in the code [08:50] (and yeah, it's not a computing error, we are using that functions hundreds of time) [08:50] glatzor: look at the data/lintian-checks from my branch, its a subset of the ftp-master tags from lintian [08:50] glatzor: but I removed the ones that where obviously not good [08:51] glatzor: like /opt in the package [08:51] thats fine for us [08:51] argh, postinst of the new kernel has a stupid typo [08:51] rodrigo_, no problem, i wont upload to it for some hours [08:51] && die [08:51] instead of || die [08:51] didrocks, hmm [08:51] rodrigo_: puzzling, isn't it? ;) [08:51] glatzor: just fyi, I look at #742935 next [08:51] ricotz: done [08:52] pitti, thanks [08:52] glatzor: looks like a double release of a fd or something [08:52] didrocks, if it's not a coding error, it's either out of memory or memory corruption [08:52] mvo, I would like to make the changes more transparent. so not removing checks but commenting them out. But how to handle overrides? should we call lintian twice? once with --no-override and the fatal tags and once with the non-fatal ones? [08:52] although the object pointer (0x1000000) looks weird indeed [08:52] glatzor: that sounds sensible [08:53] glatzor: did you merge my other stuff as well? like the exitcode check ? currently we report bad-quality even if its lintian itself that is failing :) [08:53] didrocks, can you easily replicate? [08:53] rodrigo_: not at all [08:54] didrocks, memory corruption then? [08:54] rodrigo_: yeah, the new object pointer adress is what tell me it's out of memory [08:54] rodrigo_: I would say so [08:54] rodrigo_: will confirm with Neil as well, thanks :) [08:54] didrocks, no problem :) [08:54] didrocks, if someone can replicate, running it with valgrind will give the answer [08:57] rodrigo_: yeah, but it seems $random, even if I try to add/remove a lot of apps appearing in the systray [08:57] ok [09:02] hey [09:02] hi seb128 [09:02] hey rodrigo_, how are you? [09:04] seb128, I'm fine, and you? [09:04] I'm a bit tired today, I stayed up quite late, but otherwise ok [09:05] getting some coffee so it should be better after that ;-) [09:07] yeah, saw last night you were late working [09:20] pitti, hey [09:21] bonjour seb128 [09:21] pitti, how are you? [09:22] seb128: pretty well, thanks! et toi? [09:22] I'm a bit tired but fine otherwise [09:22] pitti, did you already see the retracers crash on what is current in the amd64 log? [09:22] in get_source_tree [09:22] I didn't, no [09:23] pitti, do you want apport bugs for such cases or should I just declare the bug buggy untag and restart? [09:23] AssertionError: could not determine source tree root directory === Sweetsha1k is now known as Sweetshark [09:23] (well restart when launchpad will be back to rw) [09:23] seb128: I think this is worth fixing [09:24] seb128: in particualr, it should just skip the source stack trace here, as it usually seems to work [09:24] seb128: I'll fix it right now [09:24] pitti, ok, thanks a lot! [09:25] hm, hang on [09:25] this assertion is really justified, I guess it's a special case with a weird naming [09:25] oh? [09:29] PYTHONPATH=. python -c 'import apport; print apport.packaging.get_source_tree ("indicator-weather", "/tmp/x", "11.03.27+repack-0ubuntu1")' [09:29] hm, that doesn't crash here :/ [09:29] * pitti tries in teh chroot [09:30] ImportError: /usr/lib/libstdc++.so.6: version `GLIBCXX_3.4.14' not found (required by /tmp/tmpfREm2I/usr/lib/libapt-pkg.so.4.10) [09:30] le huh? [09:31] apt-get itself is broken now :/ [09:31] ah, ignore me; was in the lucid dchroot [09:33] seb128: meh, NFC I'm afraid; I can't reproduce it in the chroot either :/ [09:34] and now we need to wait for LP to come back anyway, once it does I'll try a manual apport-retrace [09:34] pitti, ok === czajkows1i is now known as czajkowski [09:44] does anyone know if geoclue will activate on demand? [09:45] bug #740484 is a crasher that some users seem to get when it start but the computer is not online yet [09:45] but that's a non issue if it get respawned later I guess [09:46] https://bugs.launchpad.net/ubuntu/+source/geoclue/+bug/740484 [09:46] other polkit-gnome has an assert quite some users are running into: https://bugs.launchpad.net/ubuntu/+source/policykit-1-gnome/+bug/741425 [09:46] could be a crash on logout though [09:50] good morning everyone [09:53] mvo, since lp is down I will merge the lintian branch in the evening. [09:53] glatzor: thnx [09:53] glatzor: yeah, its a bit anoying for me as well, if only bzr would stay up [09:54] mvo, you can now ship a separate data/lintian-nonfatal.tags.Ubuntu and data/lintian-fatal.tags.Ubuntu [09:54] hey chrisccoulson_ [09:54] hey glatzor mvo [09:54] hi seb128, how are you? [09:54] mvo, I recently discovered "import lsb_release; lsb_release.get_distro_information()". Makes life easier. [09:54] chrisccoulson_, I'm fine though a bit tired, what about you? [09:55] hello seb128 ! [09:55] seb128, mvo, see you guys! have to go to work now. [09:55] have fun ;-) [09:56] seb128 - yeah, a bit tired too. bug 663294 is giving me a real headache now [09:57] oh, lp in read-only mode [09:57] glatzor: great work! [09:57] glatzor: thanks and see you [09:59] mvo, I plan to revert 596.1.1 and just switch to a non-interactive session if the controlling terminal cannot be attached [09:59] mvo, it was a stupid idea to keep the tty open all the time. [10:00] have a nice day! [10:00] pitti: Good day, Martin! [10:00] pitti: Even if you seem to be busy, do you have time to sponsor the l-s side of bug 742857? Basically you already read to document last Friday. [10:00] pitti: While I think the document should be uploaded for Natty, I suggest that translations are postponed as explained in comment #7. [10:11] dpm: hello! sorry if I missed your answer yesterday, but when is the next natty langpack upload planned? [10:11] dpm: there is a file overwrite issue (with packagekit.mo) in the current ones [10:12] mvo: do you need one urgently? I was planning to do a new -base one for beta-2 [10:12] not urgently, just would be nice to know when to expect the auto-upgrade-tester to go more "non-red" again [10:12] mvo, sorry, it was me who missed the question, pitti can answer this one best ^ [10:12] mvo: bug 744831 you men? [10:12] mean [10:12] hey GunnarHj [10:13] yes [10:13] Sweetshark: hey, did you get to the static unity Quicklist bug on libreoffice? [10:13] GunnarHj: oh, sure, I'll be happy to [10:14] GunnarHj: let's see how far I get while LP is readonly [10:14] ssh: connect to host bazaar.launchpad.net port 22: Connection refused [10:14] meh [10:14] GunnarHj: I'll do that as soon as it comes back online [10:14] didrocks: bug 720716? [10:14] pitti: Thanks! [10:15] GunnarHj: it's still sitting in my mailbox, was planning to do that anyway today (was busy with a rather big patch/debugging session Mo/Tu, sorry) [10:15] Hi all. Currently having fun with gnome-shell, but missing all the themes (and no apparent application to switch themes either) [10:15] didrocks: I will give it a try today. [10:15] Currently have gnome-shell installed via the gnome3 ppa, on Natty 11.04 Beta. [10:15] launchpad is back [10:15] Sweetshark: yeah, this is important to design and we should give time to translator [10:15] pitti: No problem. I understand that you always are busy with about 100 things. :) [10:15] The following packages have unmet dependencies: [10:15] gnome-themes: Depends: gtk3-engines-pixbuf which is a virtual package. [10:16] didrocks: backports were a bit nasty this time around. [10:16] That's what I get when I try to install gnome-themes from the PPA. [10:16] I think natty > backports :) [10:16] Tommeh, hi, try #ubuntu for user questions but basically GNOME dropped the appearance capplet and there is only one theme available for the new gtk [10:17] Tommeh: and gnome-tweak-tool is your friend [10:17] right, natty > backports, I agree with didrocks [10:17] didrocks: me too. But my inbox disagrees usually. [10:17] Tommeh: you can just check it out of git and run it directly from the checkout. no need to compile. [10:18] desrt/seb128: thanks. [10:18] Tommeh: it lets you change quite a lot of things, including the theme [10:18] Sweetshark: check with chrisccoulson_ and dpm for translation btw, they dealt with that in the same case than yours (meaning: translation not in launchpad) [10:18] pitti, I'm restarting the i386 retracer and let the amd64 down for you to debug [10:18] seb128: oh, LP back up? [10:18] Tommeh: changing the theme of the shell itself is somewhat more complicated -- it involves replacing a directory in /usr... but it is possible [10:18] seb128: I'm currnetly logged into the amd64 one [10:18] Currently I'd be happy with the stnadard theme [10:18] pitti, yes, it's back [10:18] nice [10:19] I'm on some form of horrid 'fallback' gnome theme at the minute :) [10:19] awesome [10:19] Tommeh: oh. install the gnome-themes-standard package [10:19] it has adwaita in it [10:19] Ah, I've done that since removing a conflicting package (gnome-accessibility-themes) [10:19] But gnome-themes won't install from the PPA [10:19] It's still the natty version. [10:19] Due to the error above [10:19] don't worry about that one. gnome-themes-standard is what is necessary [10:20] Depending on a package that isn't in the PPA [10:20] Ahh, OK [10:20] Logout time -- brb :) [10:20] just logout -> back in [10:20] ya :) [10:20] seb128: hm, did you just rm the log file? [10:20] pitti, rotated it sorry [10:20] ah [10:20] pitti, it's log_20110406.gz [10:21] yep, thanks [10:21] Bang, I has a theme. [10:21] Many thanks desrt :) [10:21] pitti, I want a clean log will make easier to run my "get recent bugs which get duplicates" [10:21] *nod* [10:24] Tommeh: no problems :) [10:25] GunnarHj: great job! [10:25] It doesn't seem to like letting me manage two screens .. I have one screen that I can switch between workspaces, and one that's just 'stuck' with one workspace... Almost like it was never intended to be managed. [10:25] GunnarHj: one question, why did you add a no-op on_button_load_help_clicked()? [10:25] GunnarHj: isn't setting the uri button propery enough? [10:25] But that's almost nice, as I can have persistant applications (like xchat) in on screen, whilst switching others. [10:28] GunnarHj: it works fine without; do you mind if I remove the dummy handler while merging? [10:29] pitti: If it works, I don't mind at all. [10:29] pitti: GTK is another area I know very little about. :( [10:30] GunnarHj: uploaded, thanks a lot! [10:30] Tommeh: multi-monitor is a sore spot in the shell. a bunch of changes landed recently (a couple of weeks ago) to vastly improve the situation but it still leaves quite a bit to be desired [10:31] Tommeh: i sense that it's not a priority for them right now [10:38] desrt: I guess so! Well, at least it's not just me :) [10:38] I'm happier to know things are weird/broken for everyone, rather than just on my machine. [10:39] Tommeh: the gnome3 PPA is, frankly, one of the worst ways to try gnome3 at the moment [10:39] it's very deeply broken [10:39] Yep :( [10:39] which is quite understandable, given that the people behind it have their concerns in other places at the moment [10:39] i expect it will get quite a lot better after the ubuntu release [10:40] Finger's crossed :) [10:40] desrt: FWIW, currently writing GNOME30-Live-latest.iso to an USB stick; I'm curious :) [10:41] pitti: i understand the fedora 15 beta is one of the best choices... [10:44] seb128, since I updated to Natty, Evolution took the bad habit to be confused with some of my imap mails: the title of the mail is correct, but when I click on it it displays a unrelated old mail. Is that a known issue ? [10:44] not that I know about [10:45] like the preview doesn't match you select in the list? [10:45] what protocol do you use? [10:45] pitti: let me know what you think, in any case [10:45] desrt: the mini 10 touchpad still sucks even with gnome 3! [10:45] j/k [10:45] seb128, imap+. The preview is correct but the displayed mail is not [10:46] pitti: and the trackpoint on my thinkpad is still red! i thought this new CSS theme would allow me to customise that! [10:46] where is my refund?! [10:46] cassidy, like if you double click to open it? [10:46] cassidy, I never open email I just read them in the preview [10:47] seb128, oh yeah. By preview I meant the title in the treeview [10:47] seb128, the mail is wrong in the pane displaying the full mail and if I double click on it [10:47] cassidy, then no, works correctly there [10:47] check on #evolution on the gnome irc maybe [10:48] seb128, how hard are you patching evolution? ie, should I open an upstream or lp bug ? [10:49] upstream [10:49] k [10:49] or ask on #evolution if you want to start [10:49] we do patch a bit but only for ui tweaks [10:49] we don't do anything which has to do with the mailer [10:49] desrt: hm, I see the initial throbber, then nothing happens any more, got some trouble booting [10:49] pitti, wait a bit I guess [10:50] pitti, how much did you wait? [10:50] a minute or so [10:50] but the first time I tried it it booted in some 15 s [10:50] but then I tried suspend, and it crashed, so I rebooted [10:50] and now it doesn't come back [10:50] I need to try it again as well [10:50] let me fiddle a bit [10:52] but I think I'm going to miss having a launcher on screen with GNOME3 [10:52] ah, coming back now; took a lot longer, though [10:56] desrt: so, works pretty well here! not too much fun without wifi, though (bloody broadcom chipset..) [10:56] desrt: by now I got used to Windows - search term - enter to launch apps, so that's pretty familiar :) [10:57] pitti: did you at least take a peek at the new networkmanager integration? [10:57] the theme is too generous with spacing/sizing for a small netbook screen, but otherwise this looks quite nice :) [10:57] (even without networks) [10:57] pitti: yes. i agree quite a lot on the theme. i just switched to another -- inset. [10:57] desrt: I did, yes; I really like the status message there [10:58] pitti: i mean about the menu [10:58] network-manager with the translucent shell theme for menus [10:58] pretty neat [10:58] desrt: the "wired - cable unplugged" and "wireless - firmware missing", that's quite clean [10:58] pitti: actually, those notifications are still annoying in some cases [10:58] desrt: yeah, very glossy [10:58] desrt: I didn't get any notification yet [10:59] you still often see the "disconnected from network xyz" when opening your laptop after having disconnected from the network a long time ago [10:59] and the "don't show this again" button doesn't actually work :/ [10:59] the black translucent windows look very elegant (I'm just not a big fan of dark themes, but still) [10:59] pitti: i agree quite a lot. the black is just too... black [10:59] but the gloss isn't so bad [10:59] desrt: is there a way to quick launch your favourite apps? such as Windows+1 or so? [11:00] i think the idea was so that it would fade into the background and let you focus on the work [11:00] pitti: no. [11:00] this is a point that unity wins on, i think [11:00] at the same time, it's not so important [11:00] since windows+f+enter is sufficient to get firefox [11:00] right [11:00] and the search gives higher preference to items that you've marked as favourites (ie: on the dash) [11:00] desrt: work flow wise they are really quite similar to me [11:01] so that helps [11:01] the indicator layout, the windows/overlay dash handling, etc. [11:01] ya [11:01] how do I opena nautilus window, though? [11:01] there are quite some similarities [11:01] pitti: add 'files' to your dash [11:02] (should be there by default?) [11:02] apparently I'm too blind to see a 'Home' button or so [11:02] pitti: i think it's not there, actually [11:02] and the apparent lack of meaningful desktop is also somewhat distressing [11:02] win+hom+enter works [11:03] oh, I can't dnd to the background [11:03] no. the desktop is no longer shown as the background [11:03] it's just some folder in ~ now, with no special meaning [11:03] (wtf?) [11:03] you can turn the desktop back on with gnome-tweak-tool but apparently the designers found the idea of a desktop to be distracting [11:03] ah, I saw a notification now, from RB [11:03] so it's off by default [11:04] I like having a desktop ;-) [11:04] seb128: ya... no kidding [11:04] i also like having a minimise button [11:04] that I don't care much about [11:04] fortunate it's still possible to get both of them back again [11:04] well, it tends to make people not care about where they put their files, though [11:04] my mother's desktop looks horrible [11:04] basically everything she ever touched [11:04] pitti: that's sort of the point, actually [11:04] right [11:05] 3.2 is going to bring some tracker/zeitgeist love [11:05] I think the zeitgeist approach and search is a better long-term strategy [11:05] yup [11:05] desrt: I've never been a big fan of these tracker/indexer/etc. stuff, I know where my stuff is [11:05] but admittedly zeitgeist feels pretty good and magic to me [11:05] i almost get the impression that the desktop was disabled to increase the pain-level to push us in that direction faster [11:05] it even digs out some images I recently looked at from the evolution cache (email attachment), etc. [11:06] magic is magic! [11:06] desrt: well, if people need to be taught to get used to a new workflow, why not do it all the way through :) [11:06] pitti: i won't comment on the motiviations of the designers of the shell [11:06] being an old fart in this the concept of files and folders is deeply molten into my brain [11:06] because they're a bit of a mystery to me, to be honest [11:06] but then again, on my Android phone I never think in terms of files and folders [11:07] (where I'm only a user, not a developer) [11:07] and it really feels right [11:07] thus far i've managed to avoid getting a smartphone [11:07] but i guess that will end quite soon -- nexus S is in canada on april 14 [11:08] desrt: that's the problem -- once you have one, you won't ever go back :) [11:08] which means that from april 14 to april 24th or so, i'll get no work done because of angry birds [11:08] that would have been a nice bday present :) [11:08] funny you should mention that because my birthday is quite soon [11:08] but i don't think i can wait :) [11:08] desrt: so I think the biggest thing that I miss from the current live image is a good files/document search/log integration (or maybe I haven't found it yet) [11:09] desrt: when? [11:09] mine is April 14th [11:09] pitti: i think you don't miss it -- it's just not there [11:09] may 3 [11:09] natty beta 2, nexus S [11:09] same as david barth :) [11:10] desrt: hm, I thought pushing people to work more document and less app centric was a major goal? [11:10] pitti: ya. i think that's complete BS to be honest [11:10] gnome-shell SCREAMS app-centric [11:10] there should at least be some links to the standard XDG dirs [11:10] (as does unity) [11:10] documents/photos/etc. [11:10] i'm not a fan [11:10] unity+zeitgeist is sort of a hybrid [11:11] if you search stuff, it comes up with files and apps [11:11] still no xdg dir links, though [11:11] this is not usually what i think of when i think doc-centric [11:11] yes, I agree [11:11] finding docs is one thing, sure [11:11] but the *identity* of what is open already, and what you're working with [11:11] in both unity and g-s, it's very much tied to the application icon [11:12] I mean, I don't need/want it to find all my source checkouts etc., but finding music or LibO files would be great [11:12] which i think is extremely wrong on a very fundamental level [11:12] windows + floyd wall enter [11:12] should kick me right into RB with The Wall playing [11:12] you could pick a better album :) [11:13] (jokes) [11:13] windows + apo should show me my Apocalyptica albums, etc. [11:13] ah. that's good stuff, now! [11:13] or windows + tagname all my photos which are tagged with that [11:13] but I guess that's still a looong way to go [11:13] i think we'll see this happening quite soon [11:13] (or things like this) [11:14] it's a big feature item for 3.2 [11:14] that sounds all very zeitgeist/trackerish indeed [11:14] you guys are keyboard freaks, I'm wondering what normal user want ;-) [11:14] i think i want to kill the dash [11:14] not sure what i'd replace it with, though [11:14] maybe nothing [11:15] the window switcher applet was actually sort of nice :) [11:15] it listed the windows you had open (ignoring tabs, this is equivalent to documents) [11:15] so.. quite good! [11:15] desrt: the alt+tab one looks suspiciously like the XFCE one :) [11:16] ya.. don't get me started on alt+tab [11:16] that's the other thing on my gnome-shell hate list :) [11:16] again -- application centric. lame. [11:16] desrt: I guess I'm enough of a design philistine to not care too much :) [11:16] i fail to understand how gmail is any more related to facebook than it is to my word processor document [11:18] (which to drive the point home could very easily be either google docs or libreoffice) [11:19] pitti, hum, the i386 retracer crashed on the same assertion [11:19] seb128: ok, I'll stop playing with gnome3 now, and turn my attention to this :) [11:19] pitti, thanks [11:19] seb128: I can reproduce it in the chroot, now need to track it down [11:19] pitti: thanks for your opinions, btw [11:20] pitti, what was the nvidia's GL problem you talked about yesterday? [11:20] pitti, http://paste.ubuntu.com/590160/ <- could it be this? [11:21] rodrigo_, what problem? the cairo one? [11:21] seb128, #evolution has been surprisingly helpful. I fixed the issue by dropping my mail cache and opened https://bugzilla.gnome.org/show_bug.cgi?id=646898 [11:21] seb128, yes [11:21] cassidy, they usually are ;-) [11:21] desrt: so in summary, I really like the simplified network/bluetooth/a11y settings and the general design; more netbook friendly theming/window sizes and better file/XDG dir access are the main quirks that I see [11:22] rodrigo_, the nvidia binary driver create quite some ram use for each process using gl [11:22] rodrigo_, which is basically every gtk application when cairo is built with gl [11:22] pitti: nice to hear from positive words from "the other team" :p [11:22] rodrigo_: I'm afraid not; I was talking about bug 725434 [11:23] desrt: well, it's not like we wouldn't have these problems :) [11:24] pitti, but a cairo app (evo in this case) might crash with out-of-memory conditions because of that bug, right? [11:24] rodrigo_: ah, yes [11:25] pitti, ok, thanks === MacSlow is now known as MacSlow|lunch [11:46] rodrigo_: Your work on the PPA looks nice, now that GNOME 3 is close to release. You can take a look at my own packaging of gdm at ppa:bilalakhtar/gnome-builds [11:47] rodrigo_: I don't think that version of gdm is okay, but somehow it builds and runs [11:47] hence I didn't upload it to the PPA [11:47] where the PPA is the GNOME3 ppa [11:51] cdbs, cool, and yes, we need more testing, so keep it there for now [11:51] cdbs, I'll install and test it as soon as I've got a minute free [11:51] rodrigo_: thanks [11:51] rodrigo_: don't add that PPA, just manually download the package and try it [11:54] cdbs, yes [12:25] can someone please merge this -> https://code.launchpad.net/~rodrigo-moya/ubuntu/natty/evolution-couchdb/correct_cache_dir/+merge/56551 (already uploaded) [12:38] pitti, do you still work on the retracers? [12:39] rodrigo_, ok [12:39] seb128, thanks [12:39] rodrigo_, you're welcome [12:40] :) === MacSlow|lunch is now known as MacSlow [13:04] seb128: yes, it's quite a bit more difficult than I thought, I'm afraid [13:05] seb128: was at lunch now, and had some IRC discussions before, too [13:05] pitti, ok, is that specific to a bug? it started before the launchpad update so shouldn't be it [13:05] hello mterry [13:06] seb128, hi! [13:06] mterry, thanks for the geoname update email, nice to see that you managed to get locations filtered ;-) [13:07] seb128, yeah, i was excited [13:07] seb128: if the i386 retracer crashed as well, it doesn't sound too specific [13:08] seb128: did the i386 one crash on indicator-weather as well? [13:08] pitti, it's weird it just started today [13:08] didrocks: Could you give me a hint on how the mapping of application windows to desktop entries works? I removed the libreoffice-javafilter.desktop file here (which is problematic in itself it seems) and ran update-desktop-cache. Unfortunately, my writer windows have now neither a name, nor a icon in the launcher. [13:08] pitti, dunno, I haven't check the bug it crashed on [13:08] it's certainly mvo's fault! [13:08] it's always mvo's fault [13:09] seb128: the fun thing is that it shouldn't even get that far [13:09] we don't currently have deb-src lines in the chroots [13:09] so apt-get source should already fail [13:09] hmm? [13:09] oh, so it is mvo's fault! [13:10] still too early to point blame, let me dig deeper [13:10] * pitti hugs mvo [13:10] mvo, no worry, retracers crashing again, nothing special ;-) [13:11] ohhh [13:11] * mvo hides, just in case [13:11] mvo, but pitti suggested the apt-get source behaviour could have changed if you don't have a deb-src source [13:11] mvo, but well let him confirm it before investigating ;-) [13:11] that might be possible, there were changes in this area. from what version to what version? [13:12] mvo, that started this night so something uploaded yesterday? [13:12] they update daily [13:13] Sweetshark: the mapping is made by bamf, there is a debug interface to know where the mapping is done. Anyway, just edit the default desktop files we set it the launcher (you don't need libreoffice to be running for that) [13:14] seb128: hm, there is a new apt and a new python-apt but it *should* only affect multiarch setups [13:14] seb128: can I see the trace of the crash? [13:14] mvo, well as said it might be something else, it's just what pitti wrote before [13:15] * rodrigo_ lunch [13:15] mvo, it's not "crashing", it's that apt-get source should fail because there is no deb-src source but it doesn't apparently [13:16] mvo, where "fail" might be the return value changed [13:16] ok [13:16] mvo, but wait for pitti to debug it before spending time on that [13:16] ok [13:17] nessita, hola seniorita ;-) [13:17] a mere apt-get source exits with 100, that looks fine [13:17] seb128: bon jour monsieur! [13:17] nessita, how are you? [13:17] pretty good! having some net issues, but good :-) [13:17] you? [13:17] I'm fine thanks ;-) [13:18] meh, as soon as I add some debugging it crashes [13:19] or hangs, or immediately exits with 0 [13:19] didrocks: well, my writer window showed up as "LibreOffice Small Device Import Filter", which it seems to get from /usr/share/applications/libreoffice-javafilter.desktop. So I removed that and ran update-desktop-cache hoping I would now get the entry from /usr/share/applications/libreoffice-writer.desktop. Unfortunately, now I get neither an icon nor a title. [13:20] Sweetshark, how do you start those? [13:20] ok, letting seb128 dealing with it, back to hack! ;) [13:21] didrocks, ;-) [13:21] didrocks, that's the spirit! [13:21] seb128: well, I hate X, it's definitive :) [13:21] seb128: Launcher->Applications->LibreOffice Writer (in "Most Frequently Used") [13:22] hum [13:22] can you try to double click on an odt from nautilus for example [13:22] just to see if it behaves the same? [13:24] seb128: still "LibreOffice Small Device Import Filter" [13:24] did that issue start recently? [13:25] seb128: no, I think its older. [13:26] * Sweetshark builds LO much more than using it. [13:27] both libreoffice-javafilter.desktop and libreoffice-writer.desktop suspiciously have the same "Exec:" line. [13:27] right, it tries to match running binaries to .desktop [13:28] but when started with something gio it should get the information on what .desktop to use from gio directly [13:29] Sweetshark, bug #751025 [13:30] mterry, do you feel like debugging the crash you just reported? it seems like bug #434825 and a downstream crash likely [13:30] mterry, the call is coming from libindicate [13:31] seb128, sure, I can look at it [13:31] mterry, thanks [13:32] mterry, do you have things to do otherwise or do you want extra bug? [13:32] mterry, if I don't ping you with anything specific maybe just tackle indicator bugs if you feel like working on [13:32] seb128, I've got a couple things, but if you have extra bugs just lying around, I can queue them up [13:33] mterry, ok, I've nothing specific right now but I will do another indicator stack bug review today so I might bounce a few your way [13:33] k [13:34] seb128: erk, I ran it locally on bug 751695, which worked, so I now killed the core dump; but the i386 crashed on teh same problem, so I'll now move on to that and will restart the amd64 one in the meantime [13:34] mterry, having the geonames translation thing sorted would be nice but I'm not sure how doable that is for natty [13:34] pitti, ok [13:34] pitti, let me know if I can help there [13:35] Sweetshark, so yeah, seems there is an unity bug there, I will get Jason to join the channel to discuss it when he comes online (he's in the U.S) [13:35] seb128, I feel like it's close [13:36] mterry, btw ev said yesterday he would do a db update and ask I.S to roll it [13:36] mterry, I want to see if that fixes case like san francisco or lima not having any matching result [13:36] seb128, oh ok. hopefully after merging my filter branch? [13:36] mterry, he said your review was on his list for today [13:37] so yeah, that should land as well [13:37] seb128, I think my filter will. I *think* lima for example was just dropping off after too many hits. Sorting by population brings it to the top [13:37] seb128: #751025 is close on it. I also get the window show up as "LibreOffice Small Device Format Importer". If I "apt-get remove libreoffice-filter-mobiledev" I get no title to the window at all. Calc on the other hand just works fine. [13:37] mterry, oh, great! [13:38] Sweetshark, https://code.launchpad.net/~pitti/bamf/libreoffice-705461/+merge/49245 [13:38] Sweetshark, if that's of any use that's the commit pitti did to get libreoffice to work [13:39] I might have missed a few cases, I more or less just adapted teh OO.o bits [13:40] seb128: hm, how can the i386 one have crashed in the log and yet the process is still running? [13:40] ooh [13:40] 1000 25626 0.0 0.0 25632 1940 ? D 00:08 0:00 apt-get --assume-yes source indicator-weather=11.03.27+repack-0ubuntu1 [13:40] that thing has hanged for 14 hours now [13:40] pitti, weird [13:41] * pitti kills all stuff and starts over [13:41] Sweetshark, I've a suspicion something broke broke the gio .desktop matching logic [13:41] Sweetshark, but in any case seems an unity issue, not one on your side [13:44] seb128: well, it seems to work for Calc windows, but fails for writer. The desktop files look the same though. === dobey_ is now known as dobey [13:46] (modulo the issue with the libreoffice-javafilter.desktop file, but I still get no icon after uninstalling libreoffice-filter-mobiledev) === maxb_ is now known as maxb [14:15] hi, where should I look for, [14:15] if I want indicator applet session apis [14:16] ameyJ, hey, what api exactly? what are you trying to do? [14:16] @seb128, I want to set MeMenu to away when I am locking my computer either via menu or via ctrl+alt+L [14:17] there is no indicator specific api for that [14:17] your im client likely has dbus apis to do that though [14:18] ok, so you mean to say. I should look for dbus event [14:18] ok cool [14:18] thansk seb128. will work on this direction now [14:25] seb128: this is really driving me mad.. I run "apport-retrace -uvo /tmp/x.crash --auth /tmp/auth 752174" and get that crash, with a dubious debugging output [14:25] I run it again, and it exits (with 0!) immediately [14:27] kenvandine, didrocks, seb128: do you know what needs to happen to make the new overlay scrollbars the default? [14:28] besides the GTK patch (which we have now), there's liboverlay-scrollbar-0.1-dev, but merely installing the library won't magically enable them, or will it? [14:28] do we need to build something against liboverlay-scrollbar-0.1-dev? [14:28] pitti, nothing, just install overlay-scrollbar [14:28] that is it [14:29] pitti, gtk will open it and use it available [14:30] *scratches head*, which package is that? [14:30] the overlay-scrollbar source only builds the library [14:30] Package: overlay-scrollbar [14:30] Binary: liboverlay-scrollbar-0.1-0, liboverlay-scrollbar-0.1-dev [14:31] liboverlay-scrollbar-0.1-0 is enough for the current gtk [14:31] liboverlay-scrollbar-0.1-dev isn't needed because it's a dlopen [14:32] ah I remember that hack (aside from the fact that I missed the bad packaging (non-SONAME-specific files in the library) [14:32] seb128: hi, would you know anything about a replacement for vinagre in 11.04 or 11.10? [14:33] didrocks: do you have time to fix it, or want me to? [14:34] so anyway, I'll deal with the MIR and seeding [14:34] pitti: go on with it, of course :) what about the soname? [14:34] pitti, talk to kenvandine about those please [14:34] didrocks: /usr/share/apport/general-hooks/ayatana-scrollbar.py [14:34] pitti, didrocks has enough to do with unity he doesn't work on scrollbars [14:34] we can't ship that in a soname package [14:34] seb128: *nod* [14:34] pitti, kenvandine is handling the scrollbars [14:34] kenvandine, ^ [14:34] pitti: I didn't add the apport hook, so yeah, kenvandine :p [14:34] * kenvandine reads back [14:35] pitti, ok, so we want a separate binary with the apport hook? [14:35] kenvandine: so, we could do a proper "overlay-scrollbar" package which ships the apport hook and depends on the library (to avoid having to seed a particular abi of the library), or rename the hook to be ayatana-scrollbar-0.1.py [14:35] i'll do that [14:36] the latter is easier, the former cleaner IMHO [14:36] but like that it won't pass MIR [14:36] i'll do the former [14:36] makes more sense [14:36] kenvandine: awesome; please let me know when you upload, I'll shepherd it through NEW then [14:36] will do [14:36] and I'll do the MIR review in the meantime [14:36] kenvandine: cheers! [14:42] Sweetshark, is the matching bug blocking you for other things? [14:53] root@osageorange:/# apt-get --assume-yes source update-manager=1:0.147.1 [14:53] Reading package lists... 0% [14:53] seb128: ^FYI, I think I'm getting closer.. [14:53] it just stays at 0% without any progress [14:53] oh [14:54] so it's mvo's fault! [14:54] when I downgrade to 0.8.13ubuntu2, it works [14:54] E: You must put some 'source' URIs in your sources.list [14:54] and immediately returns [14:55] root@osageorange:/# apt-get --assume-yes source update-manager=1:0.147.1 [14:55] root@osageorange:/# s... 0% [14:55] ok, this -pie issue with firefox is driving me crazy now, time to look at something else ;) [14:55] that looks even funnier [14:55] chrisccoulson_: doesn't work with gcc 4.4? [14:55] pitti - i'm not sure it's entirely a gcc issue [14:55] chrisccoulson_, how come we don't get the stable version in natty btw? it's still red on versions ;-) [14:55] it might be partly a binutils issue [14:56] seb128 - the version we have is stable already ;) [14:56] chrisccoulson_, I'm surprised you don't get lot of users complaining about the version ;-) [14:56] seb128 - i do ;) [14:56] they usually tend to do that [14:56] pitti: hm, I just tried to reproduce on my natty box but I don't get this error [14:56] even if the code is the same ;-) [14:56] mvo, did you use a non available version? [14:56] mvo: I don't get it either locally, just in the retracer chroots :/ [14:57] seb128 - it's just not worth a new upload to change the version number ;) [14:57] i'll probably do it before release, but it means i have to upload a new tarball too [14:57] and they're big ;) [14:57] mvo, seb128: the version is fine, the chroot jsut doesn't have any deb-src [14:57] but I tried that locally as well [14:57] pitti, which makes it "non available" ;-) [14:57] in fact this is the only apt-get call that I have in apport [14:57] the rest is python-apt [14:58] but when I wrote that, there was no python-apt way of downloading a source package [14:59] oh, I figure installing debug packages somehow causes it to act up [14:59] after logging into the chroot it works [14:59] but not after retracing something (which installs packages) [14:59] mvo: is there a python-apt way of doing apt-get source? [15:00] pitti, overlay-scrollbar uploaded [15:00] mvo: so it's not really your fault, just the new apt seems to have become a lot more sensitive to broken packages [15:00] kenvandine: \o/ thanks [15:01] np [15:03] pitti: sorry, on the phone, I follow up in a minute [15:11] ok, not related to apport-retrace after all, the chroots previously had an older apt, and as soon as I upgrade to 0.8.13.2ubuntu1 things fall over [15:11] pitti, we can pin the old one for now maybe? [15:11] yes, that's what I'll do right now [15:17] mterry, one other thing you can to your buglist if you want that would be nice is the gvfs, libgnome-keyring sibabort issue [15:17] sig [15:18] seb128, bug me? [15:18] bug number me rather :) [15:18] mterry, the one you opened [15:18] mterry, 743497 [15:18] seb128, heh, I'm a fire and forget bug reporter [15:19] mterry, if you don't fix it at least maybe try to open an upstream bug with the debug info upstream mentioned [15:19] sure [15:19] i'll give it a pass [15:19] mterry, upstream is stefw on #gnome-hackers if you want to talk to him about it [15:19] thanks [15:22] seb128: ok, I think I have reasonably clean chroots again with pinned apt; restarting, cross fingers :) [15:25] pitti, \o/ [15:29] seb128: seems to work [15:33] kenvandine: mind to upload a followup which adds the necessary Replaces:Breaks: (<< 0.1.5-0ubuntu1)? [15:33] hey Laney [15:33] pitti, whoops [15:33] yeah, sorr [15:34] kenvandine: (and perhaps fix the overlay-scrollbar desscription while you are at it -- it's not a devel package) [15:42] jcastro, ping [15:43] rodrigo_: hi! [15:43] jcastro, about http://askubuntu.com/questions/22946/how-do-i-install-the-latest-version-of-gnome-3 [15:44] jcastro, it's better if people just run apt-get dist-upgrade [15:44] ok [15:44] jcastro, and maybe a separate 'apt-get install gnome-shell' if the want g-shell [15:44] is there one metapackage or does the dist-upgrade just take care of that? [15:44] jcastro, dist-upgrade should take care of that [15:44] jcastro, that's why we didn't add the metapackage [15:45] ok fixed [15:45] I added a scarier warning up top too [15:45] pitti, uploaded [15:45] since downgrading looks like, horrible [15:45] jcastro, yes, good idea [15:46] jcastro, hmm, on the UI, no way to do a dist-upgrade? [15:48] kenvandine: thanks! [15:49] seb128: Im on bug 720716 . But I will test it with Calc and hope it will work for writer the same way. [15:50] * Sweetshark starts on patch 18 for libreoffice packaging. [15:50] Sweetshark, you can try to downgrade bamf to 0.2.80-0ubuntu1 [15:50] rodrigo_: the button works that out as far as I know. [15:50] Sweetshark, https://launchpad.net/ubuntu/+source/bamf/0.2.80-0ubuntu1 [15:51] jcastro, ok then [15:51] Sweetshark, if you downgrade it and restart bamfdaemon and unity and let me know if that fixes your issue that would be nice [15:51] Sweetshark, well no hurry but when you will have some time [15:52] seb128: heh. [15:52] seb128: but yes, I will try. [15:54] seb128, hi, didn't you say last week that the weird locations dialog of the datetime applet was fixed? or did I dream it? [15:55] fta, what about it? mterry said that the "no listing any result during typing issue" was fixed [15:56] yeah [15:56] seb128, it's not here, "san francisco" shows nothing between "san" and the "o", then i only get unknown (to me) location, nothing in california [15:56] locationS [15:57] fta, san francisco is a known bug, I believe due to us not sorting by population yet [15:58] same for sydney.. it have to add "airport" or "international airport", and i end up with a list of hotels or universities close to those airports :P [15:59] hence, very long names, making the indicator too wide :P [16:00] mterry, is the server limiting the number of returns? [16:00] mterry, because http://geoname-lookup.ubuntu.com/?query=san francisco doesn't list it either [16:00] seb128, yes [16:00] ok [16:00] seb128, that's why the filtering and sorting branch should make this a lot better [16:00] no more hotels, and better results on top [16:00] right, I got those advantage [16:00] fta, ^ [16:01] ok [16:01] but I didn't know we were dropping locations because of the lack of sorting [16:01] fta, basically, a fix is in, we just wait for it to be deployed server side [16:01] seb128, it's a big limit, like 1000 or so, but yeah [16:01] (I think, would have to look at code) [16:01] mterry, "http://geoname-lookup.ubuntu.com/?query=san francisco" returns a small list [16:01] seb128, then maybe the limit is smaller :) [16:01] but anyway I will try when the server update is rolled [16:03] another weirdness of that datetime indicator, the calendar. the highlighted/selected day is the day of the last desktop restart (useless imho), it used to be the current day [16:04] seb128, Can you mark the appmenu-gtk part of this as Invalid for Natty? I can't seem to :-/ https://bugs.launchpad.net/ubuntu/+source/wxwidgets2.8/+bug/722375 [16:05] tedg, done [16:05] seb128, Thanks! [16:05] fta, known bug as well, karl is working on it [16:05] it might be fixed in trunk [16:05] tedg, yw [16:09] good [16:12] mterry, after a restart/reboot, i have to open the locations dialog 2 or 3 times to be able to use it, the 1st call, i get nothing, the 2nd, it disappears immediately, the 3rd time, i see it for a few seconds then pooff, after that it works fine. is that known? [16:13] fta, seems like it crashed? [16:13] fta, huh... [16:13] rodrigo_: does the dist-upgrade include gnome-shell? [16:13] jcastro, no [16:13] or does the user have to install that explicitly? [16:13] fta, if you click elsewhere in the dialog, the locations dialog disappears [16:13] fta, did you try to start it before being online? [16:13] jcastro, so add a 'apt-get install gnome-shell' to the instructions [16:13] ok [16:14] lets see if it crashed... [16:14] pitti: still in a meeting (its meeting day today). but you can get the urls for the source via python-apt [16:15] nope: http://paste.ubuntu.com/590294/ [16:15] pitti: I can write some test code in a bit (please kick^Wremind me if I doN't do it) [16:15] mvo: don't worry for now; apt itself is broken in the chroots, so the python-apt source code retrieval isn't urgent at all [16:16] seb128, i'm always online ;) [16:17] jcastro, ok, I think they look great now [16:17] jcastro, let's see what people report [16:17] I'm asking a guy to test right now [16:17] difficult to see anything in .xsession-errors. i wish everything was timestamped [16:18] rodrigo_: I'll watch this with great attention today to make sure it's kept up to date [16:18] pitti: ok [16:18] jcastro, ok, cool [16:36] rodrigo_, https://bugs.launchpad.net/ubuntu/+source/evolution-data-server/+bug/750660 [16:36] rodrigo_, is that the evolution-couchdb issue you fixed the other day? [16:36] rodrigo_, https://bugs.launchpad.net/ubuntu/+source/evolution-data-server/+bug/744464 [16:36] same [16:37] seb128, no, but seems related to a problem rye (on #ubuntuone) is having [16:37] rodrigo_, can you add those to your buglist? [16:38] seems frequent crashers [16:38] ok [16:38] rodrigo_, https://bugs.launchpad.net/ubuntu/+source/evolution-data-server?field.searchtext=e-addressbook-factory&orderby=-datecreated [16:38] they are duplicates, right? [16:38] if you want to review, clean it a bit as well [16:38] * rodrigo_ looks for the 2nd one [16:38] rodrigo_, they seem similar yes [16:39] rodrigo_, let's use bug #744464 for it, I assign it to you [16:39] ok [16:39] cyphermox, hey [16:39] cyphermox, did you say you were working on e-d-s crashers? did you work on those e-addressbook-factory ones? just to make sure rodrigo doesn't start on something you already work on [16:40] http://people.ubuntu.com/~fta/lazy-unity.ogv [16:41] fta, what about it? [16:41] fta, you need to hit 0,0 for the launcher to stick [16:43] seb128, wow, lots of e-d-s crashes, I'll go over the list [16:45] seb128, I have a merge up for the addressbook ldap bug, looking into the google one now [16:45] rodrigo_, those are only the addressbook ones ;-) [16:45] seb128, oh :) [16:46] cyphermox, ok, I will review and merge that later on, I will probably wait for other fixes before uploading though [16:46] rodrigo_, seb128, not necessarily looking at crashers per se, but I'll look at the evo/eds bugs in general to make sure we're not dropping the ball [16:46] seb128, sure [16:46] cyphermox, thanks [16:46] if you stop couchdb or contact synced issue ping rodrigo_ about those [16:46] I'd like to include the fix for google if possible, but i think it will be in eds, not evo [16:48] cyphermox, ok, leave this list -> https://bugs.launchpad.net/ubuntu/+source/evolution-data-server?field.searchtext=e-addressbook-factory&orderby=-datecreated to me if you want [16:49] seb128, along with the evo-couchdb branch I pasted before, can you also merge this https://code.launchpad.net/~rodrigo-moya/ubuntu/natty/couchdb-glib/enable-debug-messages/+merge/56597 ? (already uploaded also) [16:49] rodrigo_, ok. well, looks like there may be a few overlaps there -- let me know if down the road you see that we could split them up between us to fix it all ;) [16:50] cyphermox, ah, you're looking at those bugs also? [16:50] cyphermox, since seb128 said those were only the addressbook crashes, I assumed there are lots of other e-d-s bugs [16:51] rodrigo_, https://bugs.launchpad.net/ubuntu/+source/evolution-data-server?field.searchtext=crashed&orderby=-datecreated [16:51] rodrigo_, https://bugs.launchpad.net/ubuntu/+source/evolution-data-server?field.searchtext=SIG&orderby=-datecreated [16:51] rodrigo_, ok. I was going to look at them, but I'll leave them to you and look at the other e-d-s crashers [16:51] rodrigo_, try that url [16:51] hum [16:51] seb128, why do you make this to me??? :) [16:51] lots of crashes man! [16:51] cyphermox, yes, unless you want some explicti addressbook ones, leave those to me [16:52] there are plenty for both of us :) [16:52] ok [16:52] yes, there are ;) [16:52] cyphermox, do you want me to leave the google related ones to you? [16:52] ah, up to you [16:52] lot of noise and duplicates likely [16:52] yes [16:52] I was going to ask whether you wanted to look at the issue with google addressbook at the same time [16:52] well don't bother cleaning old ones [16:53] but seems that most of recents crashes are e-addressbook-factory ones [16:53] cyphermox, ok, let's use the last list [16:53] cyphermox, ok, let's use the last 2 lists seb128 posted for both of us [16:53] cyphermox, just assign to you the bugs you're working on and I'll do the same [16:53] those didn't work ;-) [16:54] but just type SIGSEGV in the search entry and sort by newest [16:54] seb128, hmm, it works for me, 182 bugs [16:55] ok great ;-) [16:55] well don't try clean the buglists, just try to spot things worth fixing for natty [16:55] ok [16:55] we have pedro to clean ;-) [16:55] hey pedro_! ;-) [16:55] heh [16:56] I should start running before pedro throw something in my direction ;-) [16:56] hehe [16:58] ugh, we have a lot of patches in the evo package [16:59] what do you think about doing some patch upstreaming/cleaning days when natty is released? [16:59] it makes upgrading packages a lot harder [17:01] rodrigo_, yes, but I think there may be quite a few which are backports [17:01] yes [17:01] those are ok, as they are removed when we move to the next major version, right? [17:01] afaik they should be well identified as such in changelog and even the file name, so when we're at uploading 2.91. whatever or 3.2 we can drop them [17:01] yeah [17:02] yeah, right [17:02] I remember in opensuse we had packages with more than 20 patches, so we upstreamed/cleaned a lot of them, and upgrades were much easier [17:03] evolution had, iirc, over 100 patches! [17:24] rodrigo_: well, LO still has >500 "shared patches" used by all distros. The 18 patches are just our private debian/ubuntu patches on top of that (some patching the patches). === Cimi__ is now known as Cimi === smspillaz is now known as smspilla|zzz [17:30] dobey, bug #724882 is getting some duplicates recently [17:31] rodrigo_, evo doesn't have a lot of distro patches [17:35] Sweetshark, ugh [17:43] does anybody has interest in geoclue? [17:43] bug #738584 [17:44] hum no bot [17:44] that's an assert crash which seens to happen on resume [17:44] it's likely not a real user situation issue if it respawn when needed though [17:44] kenvandine, cyphermox: ^ [17:44] https://bugs.launchpad.net/ubuntu/+source/geoclue/+bug/738584 [17:45] if one of you want to have a go to it that would be nice [17:46] ok, i'm using the small apple aluminum bluetooth keyboard, and it's lacking an insert key, i'm wondering how i can make a shortcut to issue "insert" as if i was hitting the key-- like ctrl+shift+i (random example) [17:47] ubuntu 10.10/2.6.35.x/gnome/pc [17:47] try #ubuntu for user questions [17:47] ok, thanks [17:51] seb128, sure, I'll take that too [17:51] cyphermox, thanks [17:52] isn't it indicator-datetime which uses geoclue? [17:53] cyphermox, eventually empathy will (i would assume) [17:53] bcurtiswx, cool [17:53] I know redshift does not, but not in the natty package ;) [17:53] not, = now, [17:54] right the indicator use it [17:59] mvo, #741370 [17:59] mvo, it's an aptdaemon bug with quite some duplicates [18:00] ok [18:00] tremolux, bug #728713 [18:00] can you check on it? it's a s-c crash with some duplicates [18:03] seb128: yep, we've been watching that one, I will check on it [18:03] seb128: thx [18:03] thanks [18:03] tremolux, can you make sure it's assigned to somebody? [18:03] seb128: yep [18:19] seb128: ok, i'll look at it [18:19] thanks [18:37] pitti: hey there, would you have some minutes? I have a question regarding feature freeze [18:38] nessita: about to run out, but shoot [18:38] nessita, others can reply [18:38] just ask [18:38] so, we're having an internal discussion about what is an API change and what isn't. We need to add a new parameter to an internal method in syncdaemon, we're saying is not an API change but others disagree [18:39] so, from our point of view, syncdaemon API is the dbus api only [18:39] but the whole implementation is somehow available in /usr/lib [18:39] nessita, well it's an internal api chance, not an external one then? [18:39] seb128: right [18:39] is whatever provide the api public? [18:39] you say it's in /usr/lib [18:40] nessita: I think the question here is: can it break anything that both consumers in Ubuntu as well as third-party developers are using? [18:40] seb128: syncdaemon implementation is in ubuntuone/syncdaemon module as a python module, but end users are not supposed to use that [18:40] nessita: i. e. can they call the function you are about to change? [18:40] pitti: they can call it, in the sense that is publicly accessible code [18:41] pitti: but anyone is supposed to use those things outside syncdaemon [18:41] do you know of anybody using it? [18:41] no, I don't [18:41] nessita, "anyone" -> "no one"? [18:41] nessita: i. e. the official API is the d-bus one, and that's also what our ubuntu packages use? [18:41] seb128: yes :-) [18:41] pitti: yes [18:41] nessita: sounds fine to me then [18:41] great, thanks [18:42] Python code is by nature very "open", but that doesn't mean that everything is public API automatically [18:42] anyway, Taekwondo o'clock, see you tomorrow! [18:42] seems like it's a private api not supposed to be used and those who (abu)sed it will get what happens in such cases ;-) [18:42] pitti, see you [18:42] i prefer my solution to that problem; just avoid python [18:42] pitti: thanks!!! [18:42] * nessita hugs seb128 [18:43] * seb128 hugs nessita [18:43] nessita, you trapped me with your u1 btw, even didrocks made fun of me because I share bug screenshots by publishing them on u1 now! [18:43] ;-) [18:44] nessita: you finally won \o/ [18:47] YEAH [18:47] * nessita happily dances [18:59] seb128, I haven't been able to reproduce the geoclue crash, but I guess I have a fix that makes some sense anyway. [19:00] i'll do some more testing and checking first though -- but now, lunch time :) [19:00] cyphermox, ok thanks [19:00] did you personally experience it? [19:28] seb128, any trivial empathy bugs you've run across? [19:33] bcurtiswx, no [19:33] seb128, OK thanks :) [20:03] kenvandine, everytime i bring up the gwibber window it eats CPU, but when its minimized, it doesn't.... [20:03] seen this? [20:03] bcurtiswx, does it keep eating CPU while it is up? [20:03] only when the window is unminimized [20:04] it takes a little bit too, but it does eventually [20:05] yup, now at 100% CPU [20:05] maybe since i have a refresh rate in the quick end... seems to me [20:05] weird [20:05] you mean the gwibber client window right? not the lens? [20:06] i dont' use the lens [20:06] ok [20:06] i can't see why the client would use CPU [20:06] it doesn't do anything, except when it gets the signal to refresh the WebView [20:07] hmm, i have a lot of services it checks with facebook having a TON of friends.. i wonder if its shear quantity causing it.. [20:07] which is my own fault TBH [20:07] i need less friends... [20:08] yup thats it, gwibber-service on a refresh [20:09] kenvandine, ^^ [20:09] oh, it isn't the client then [20:09] rickspencer3, fix for bug 739923 being uploaded now :) [20:09] yes, sorry for the lack of -service :X [20:09] bcurtiswx, how big is your db? [20:10] ls -lah ~/.config/gwibber/gwibber.sqlite [20:10] 175m [20:10] 175M* [20:11] ugh [20:11] that is why :) [20:11] what version of gwibber? [20:11] 3.0.0-0ubuntu1 [20:11] that number should be much lower [20:11] wow [20:11] how many accounts do you have? [20:11] Twitter, Facebook, Identi.ca [20:12] hummm [20:12] it should keep the latest 2000 posts for each account [20:12] and purge the rest [20:12] so it shouldn't be nearly that big [20:12] bcurtiswx, can you post your log somewhere? [20:13] sure, which log? [20:13] ~/.cache/gwibber/gwibber.log [20:15] kenvandine, sent in IM [20:15] got it [20:15] thx [20:17] dude... what is your interval set to? [20:17] * kenvandine should really cap the interval at like 5 or 10m [20:18] i've added protect in so it doesn't overlap anymore, but you still have basically a constant state of refreshing [20:18] s/protect/protection [20:19] bcurtiswx, ^^ [20:19] kenvandine, 1 O:-) [20:19] please bump that up [20:20] :) [20:20] kenvandine, OK [20:20] pitti, do you think i would need a FFE to prevent people from setting their refresh interval so low? [20:20] in gwibber that is [20:21] 1) change the selector in preferences so it bottoms out at 5m and migrate existing settings of less than 5m to 5m [20:21] this intervals i think cause a fair number of problems [20:22] kenvandine, can i play devils advocate for a minute? [20:22] sure [20:23] there's a lot of us that like quick refresh rates, (like the auto-updating twitter).. is saving 2000 really needed? [20:23] yeah... you'd be surprised how few that is [20:23] regardless though... it isn't about how much is being saved [20:23] your are beating on it constantly [20:24] multiple parallel threads download data, doing stuff with it and insert/updating the db [20:24] and as soon as it is done, it does it again [20:24] kenvandine, what causes the high CPU? it has to read the 2000 and do something with it? [20:24] multiple parallel threads... blah blah [20:24] no [20:24] it is the refreshing [20:25] and actually a db that big doesn't cause a problem for the service [20:25] not CPU wise [20:25] well, if i delve any deeper into the cause we'll get into stuff that can be mentioned at UDS (so i won't go any further). :) [20:25] it was causing slowness for the client though [20:27] i think regardless of any future improvements, preventing intervals < 5m would alleviate a fair bit of pain [20:27] bcurtiswx, i have ideas how we can get the twitter realtime feed pretty easily for 3.2 [20:27] we'll talk about that at UDS [20:28] kenvandine, i can't wait to be involved then :) [20:29] i agree, no need to argue <5 mins should be discouraged for natty :) [20:30] * kenvandine prepares a branch with the interval tweak [20:38] bug #739923 [21:03] rickspencer3, the bot isn't very happy now [21:03] rickspencer3, the pango_overlay thing in GwibberPosterVBox :) [21:08] rickspencer3, after i get out of this meeting, i'll post an example [21:14] Are you aware of gnome-icon-theme-symbolic: Depends: gnome-icon-theme (< 2.92) but 3.0.0-0ubuntu1~build1 is installed. [21:15] IN the gnome3 stack? [22:00] hi there, any ideas why a launchpad build is failing with: [22:00] dpkg-deb: error: control directory has bad permissions 700 (must be >=0755 and <=0775) [22:00] ? [22:03] nessita: ask on #launchpad instead [22:03] ok [22:31] nessita, there is something about umask being discussed on #ubuntu-devel, seems like it might be relevant [22:35] tedg, did mterry's changes to dbusmenu to provide a menu closed signal ever land? [22:35] i couldn't find it here, unless i just need to update my machine ;) [22:35] chrisccoulson_, Honestly, I'm loosing track. I think so :-) [22:35] cyphermox: thanks! [22:36] chrisccoulson_, This week the parser side should land, but I dont' think you'll need that. [22:38] tedg - yeah, i don't use the parser [23:22] * nessita -> eods