=== Amaranth_ is now known as Amaranth [07:20] good morning [07:38] Good morning [08:04] BANZAI! [08:04] http://cdimage.ubuntu.com/daily/current/ [08:05] i386/amd64 not oversized any more [08:05] OO.o helped to kill ~ 10 MB [08:05] * pitti dances around happily [08:06] pitti: BTW, libreoffice is in experimental [08:06] micahg: right, but I won't touch it, I'm afraid [08:07] micahg: we'll get a proper OO.o maintainer soon [08:07] it already took me half a week to get the current version building on natty.. [08:07] pitti: wow, yeah, I just read the changelog, I hope the libreoffice build system ends up a little cleaner [08:08] micahg: so am I; doing patches-for-patches hurts a lot [08:08] micahg: well, I hope/think that most of the go-oo patches can make it into upstream libreoffice [08:09] and for the remaining ones we can hopefully just have a single set of patches which are directly applied to the build tree === almaisan-away is now known as al-maisan [08:22] hey pitti! [08:23] pitti: excelent :) [08:23] pitti: oh really, we have an OO.o maintainer now? \o/ [08:24] didrocks: not yet [08:24] didrocks: just a poor pitti who spend his nights sledgehammering on the current package :) [08:24] pitti: hehe, yeah, I can imagine. I played that game for a week in August [08:25] pitti: at least, I hope that me, documenting all changes in debian/changelog helped you [08:25] (because most of the diff between debian and us wasn't documented…) === al-maisan is now known as almaisan-away === almaisan-away is now known as al-maisan [08:42] unfortunately http://cdimage.ubuntu.com/daily-live/20101213/ is still heavily oversized [08:45] :/ [08:45] pitti: how was your week-end, btw? [08:46] * didrocks just finished a tedious 2 hours of week-end unity bug triaging :/ [08:46] didrocks: went to visit my parents and cousins yesterday, for the ceremonial red cabbage cooking :) [08:46] (my father has an awesome recipe) [08:46] otherwise it was fairly quiet; some OO.o hacking, and some idling, I caught a bit of a cold [08:46] oh nice! [08:46] urgh [08:46] didrocks: last week I discovered http://www.dxx-rebirth.com/ [08:46] take care… not very surprising with this weather btw :) [08:47] that's a productivity killer :) [08:47] I loved Descent back in the DOS era [08:47] oh, excellent! Yeah, you will kill my productivity as well :) [08:47] and now it runs amazingly well under Linux, and it looks 10 times better than back then due to full resolution [08:47] I played that game too much :) [08:48] didrocks: I'm in level 9 in Ace mode now -- gee, they really give you a good beating at that difficulty [08:48] pitti: they changed the difficulty level? [08:48] didrocks: no, I just didn't play Ace back then [08:48] didrocks: btw, if you want to build it, give me a ping; there's some tricks you have to do to get it to build [08:49] pitti: oh sure, I'll. Probably after the holidays I guess (so in January) as I think I'll have to focus on housework first :) [08:49] thanks :) [08:50] * pitti discovers 1.9 MB balooning of libgphoto2 and fixes [08:50] 1.9 MB? what was the cause? [08:52] it accidentally ships the API docs in both the lib and the -dev package [08:53] oh ok [08:53] I'm analyzing the iso-deb-size-compare result between maverick and current daily [08:54] and this one stuck out [08:55] oh there is such a tool or did you hack a shell-script ourself (I did that when I tracked the oversize in UNE). I've always wondered why the manifest doesn't report the size [09:00] didrocks: I hacked it a few months ago [09:00] didrocks: it's in lp:ubuntu-archive-tools [09:02] pitti: thanks I'm refreshing my branch then :) [09:04] didrocks, bug 688803 [09:04] Launchpad bug 688803 in unity (Ubuntu) (and 1 other project) "SIGSEGV in g_atomic_int_get (affects: 1) (heat: 6)" [Medium,Triaged] https://launchpad.net/bugs/688803 [09:04] what's incomplete there (except the call in /usr/lib/libdbusmenu-glib.so.2)? i always thought if i get all function names i'm good to report? [09:05] htorque: if you look, for some function, you ony have the prototype, not the values, not sure why, I was blaming missing dbgsym [09:05] htorque: but in any case, it can be easily reproduced, hence the triaged :) [09:07] didrocks, oh, i see. will try again. :) [09:07] htorque: thanks :) [09:07] htorque: you know about the dbgsym and such, I guess? [09:07] htorque: oh you build from trunk most of the time, scratch it :) [09:07] (but not for bamf and glib I guess) [09:17] hey [09:17] hey seb128 [09:17] hey pitti [09:17] how are you? [09:17] had a nice we? [09:18] seb128: have a bit of a sore throat, but weekend was nice, yes [09:18] seb128: I finally got OO.o to build \o/ [09:18] oh, take care of you then [09:18] nice, so you are officially maintain it now? ;-) [09:18] * seb128 runs away from pitti [09:18] seb128: sorry, I didn't understand your last sentence; must be a temporary transmision glitch [09:19] salut seb128, ça va? [09:19] lut didrocks [09:20] ouais et toi ? [09:20] ça va bien :) [09:21] did the unity and indicators update land on friday? [09:21] did they work? ;-) [09:21] and we have current CD images again, finally [09:22] seb128: unity landed and worked for me. Not sure about the indicators update [09:23] seb128: seems that a lot of people got stuck in the dist-upgrade which removed the indicators [09:23] so, I'm just apt-get upgrade now and then will try a dist-upgrade to see the current status [09:25] ok [09:26] didrocks, yeah, when testing trunk i always remove the -dbgsym packages to avoid conflicts. looks like i didn't reinstall them when testing this with the natty packages. :) [09:27] htorque: ok :) === Tm_T_ is now known as Tm_Tr [09:43] good morning everyone [09:43] hey chrisccoulson [09:43] hey chrisccoulson, how are you? [09:43] how are you? [09:44] good thanks, how are you? [09:44] I'm fine thanks [09:46] hmmm, unity doesn't work well when i dock my laptop :/ [09:46] i have one black screen and one screen with half a shell on it ;) [09:47] hey chrisccoulson [09:47] hi didrocks, how are you? [09:47] chrisccoulson: yeah, multi-monitor isn't good for now [09:47] chrisccoulson: I'm tired but fine, thanks :) [09:48] ah, restarting compiz fixes the issues :) [09:48] i now have 2 monitors back [09:51] chrisccoulson: yeah, but the panel isn't at its best from the bug reports I see :) [09:52] didrocks - yeah, it spans both monitors and the menubar on the opposite monitor is a bit weird ;) [09:52] but i need to run unity now to hack on the ffox menubar :) [09:52] chrisccoulson: there are already discussions on bugs about that :) [09:53] chrisccoulson: hehe :) (I think you can also use the gnome-panel appmenu applet, but I won't discourage you using unity of course :p) [09:54] oh, i didn't realise that! [09:57] catching up on planet [09:57] chrisccoulson, nice appmenu firefox work ;-) [09:58] thanks :) [09:58] pitti, I think one reason to not preselect the gdm user as well was that selecting it starts the pam dialog [09:58] there's still quite a lot for me to do, and i need to get some missing features in to libdbusmenu this week too [09:58] but it's getting there! [09:58] there was some issues, timeout bugs, etc in the past about those [09:58] chrisccoulson, ;-) [09:59] seb128: I see; right, it needs to check for the auth method [10:00] well I don't say it's not doable but it would not be as trivial as activating the widget [10:08] didrocks: do you know what the real packaging bzr branch for unity-asset-pool is? [10:08] didrocks: https://code.launchpad.net/unity-asset-pool is not entirely clear, and the package doesn't have a vcs-bzr [10:08] pitti: should be lp:~unity-team/unity-asset-pool/packaging [10:08] meh, -ECAN'TCOMMIT [10:09] didrocks: if I throw you a branch, could you pull it into that one? [10:09] yeah, that's something which should be fixed, not sure that subscribing all core-dev to unity-team and getting spammed but all unity bugs is nice though :/ [10:10] pitti: sure, sorry for the extra work [10:10] didrocks: no problem [10:10] I also need to fix notify-osd, doesn't build in natty [10:10] but I think I still have upstream commit powers there [10:10] hehe :) [10:10] didrocks, why do we need unity in the unity team vcs? [10:10] the packaging should use the standard lp:ubuntu [10:11] pitti, check with mirco, I think they fixed issues in trunk [10:11] pitti, would be nice to get a new tarball as well [10:11] seb128: well, we set up like that at the beginning, but yeah, that should be pushed in lp:ubuntu. can do it [10:11] didrocks: I can do that as well, while I'm at it [10:11] didrocks: (u-a-p to lp:ubuntu/ [10:11] pitti: sure, please do :) [10:11] I'll do the others [10:11] alrighty [10:12] didrocks: perhaps you can delete the branch from LP then [10:12] brb session restart [10:12] pitti: seb128: the only thing is that for daily build, sometimes, I change the packaging in advance [10:12] too late :) [10:15] chrisccoulson: hey, I got a mail from a guy with problems with the latest ff/tbird (from 2010-12-10) are you interessted in reports like tihs? [10:16] chrisccoulson: very little details unfortuantely [10:16] mvo - it depends on what the problem is ;) [10:17] we get quite a lot of different problems from single reporters which generally tend to either not be real bugs or problems with plugins or extensions, so i normally only look at problems that affect more than one person ;) [10:20] chrisccoulson: yeah, I figured that. fair enough [10:20] mvo - what details do you have though? [10:21] very little [10:21] http://paste.ubuntu.com/542986/ [10:21] but I wanted to raise it just in case it affects more people [10:21] don't worry if its a single report [10:22] seb128: the only thing about packaging in lp:ubuntu, is that for daily build, sometimes, I change the packaging in advance. But it's rare enough to be a corner case [10:23] didrocks, well, you can have unreleased commits in lp:ubuntu [10:24] seb128: yeah, but basically, some commits are in advanced like updating the symbol files for symbols in next release [10:24] but again, that's a corner case [10:24] or maybe ~ubuntu-desktop will make more sense for that? [10:24] pitti: ^^ [10:25] didrocks: no, use lp:ubuntu/ [10:25] didrocks: since ~ubuntu-desktop has upload rights for these packages, they can also commit there [10:25] it's all integrated into per-package uploader etc. [10:25] didrocks: and yes, it's totally fine (and encouraged) to commit stuff there without upload, same as with the old branches [10:26] it's still a bzr branch, after all [10:26] ok, and for compiz? it has been ~compiz for some time already [10:26] mvo's and your call, I guess [10:26] I'm all for allowing people to commit when they can upload [10:26] otherwise it's quite a burden [10:27] well it's not an issue when you have a list set for the teams [10:28] like anybody with upload rights can commit to the ubuntudesktop vcs [10:28] seb128: I need to rebuild seahorse; want me to upgrade to 2.32 while I'm at it? [10:28] but the unity team has no mailing list so it would mail spam everybody [10:28] (we are at 2.91.2) [10:28] erm, ignore me [10:29] I read it backwards [10:29] ;-) [10:29] uscan FTL [10:29] we have 2.32 [10:29] pitti, just curious why do you need to rebuild it? is there a list of "need rebuilds" we can help on? [10:29] seb128: optimize SVGs and PNGs [10:30] seb128: http://paste.ubuntu.com/542992/ [10:30] seb128: I spent some time analyzing today's CDs [10:30] and this is my work list [10:30] I can finish SVGs/PNGs today, then this chapter is done :) [10:31] ok [10:31] those changes make me nervous btw [10:31] which? [10:31] having games excluded from the list seems a workaround [10:32] touching the images without knowing if the software rely on them being in a specific format [10:32] is that being done only on things on the CD? or anything using cdbs will get it? [10:32] anything right now (except games) [10:32] didrocks, ubuntu-desktop is a subteam of compiz [10:32] I haven't heard about problems with that yet, aside from this game [10:33] seb128: should we rather do the same with unity + add a ML? [10:33] pitti, ok, well let's see but it makes me nervous a bit [10:33] seb128: I don't really care, just if we can be consistent :) [10:34] didrocks, being consistent would be to move to lp:ubuntu [10:34] imho [10:34] well do we need a special team for unity? [10:34] seems you abuse the unity upstream team for packaging [10:34] didrocks, you could move it to the ubuntu-desktop team if you prefer [10:34] that would be better than the unity one to start [10:35] seb128: I proposed that as well, but pitti is more in favor for lp:ubuntu/ [10:35] so let's do that way [10:35] well then you can as well as james_w to alias the lp:ubuntu location to it [10:35] that's what we do for dx sources [10:35] indicators [10:35] didrocks: ~branch is by and large ~ubuntu-desktop for desktop packages :) [10:35] well you need to ask james_w [10:36] meh, seems I'm not able to commit to notify-osd trunk any more [10:36] the autoimport source is not a derivative of the upstream vcs [10:36] ok, will do that then, setting up a list and pushing under ~ubuntu-desktop [10:36] * pitti will do a MP [10:36] pitti, what error do you get? did you check the merge proposal for it? [10:36] it seems people were speaking about fixes for notify-osd last week [10:36] seb128: I only checked out trunk [10:36] seb128: usr/bin/ld.bfd.real: notify_osd-bubble.o: undefined reference to symbol 'XDeleteProperty' [10:37] it's a missing -lX11 [10:37] seb128: will look at MPs [10:38] seb128: nope, nothing there; I'll do my own [10:38] pitti, https://wiki.ubuntu.com/RobSavoye/GoldFixes/notify-osd [10:39] pitti, Sarvatt_ was working on it last week according to my logs [10:39] ugh, most discoverable page ever :) [10:39] he uploaded a fixed version to a ppa apparently [10:39] Sarvatt_: do you have a local notify-osd branch for that? [10:39] pitti, maybe check with him to no dup work [10:39] seb128: ok, I'll shelve that until later then [10:39] seb128: thanks for pointing out [10:39] np [10:40] xorg-edgers has it [10:41] hum no, the upload there failed to build [10:41] I will let Sarvatt_ comment ;-) [10:41] right [10:42] not that urgent [10:42] in any case would be nice if macslow could roll a tarball [10:42] the trunk has some bug fixes as well [10:42] two or three, yes [10:43] perhaps he can do a MP review before, and get this fixed [10:43] since this requires autoconf changes [10:52] didrocks, speaking of unity spam, my unity box got some since last week ;-) [10:53] seb128: yeah, I'm not eager of Monday morning because of that :-) [10:53] seb128: btw, you restarted with the new version, isn't it? [10:56] didrocks, yes [10:56] why? [10:57] seb128: just to know if everything's all right on upgrade (gnome-session and unity) :-) [10:57] I had to remove the detection module on Friday evening, it was segfaulting for some people and finally they got to unity + gnome-panel [10:59] didrocks, in fact it's broken [11:00] well first I didn't dist-upgrade, I'm avoiding the python transition [11:00] so I upgraded unity and its depends [11:00] not gnome-session [11:00] but compiz crashes when I activate unity it seems now :-( [11:00] seb128: oh ok [11:00] hum? [11:00] (I had until turned off before you asked) [11:00] Initializing staticswitcher options...done [11:01] ** (process:3106): DEBUG: perf-logger.vala:62: shoop de whoop [11:01] and it seems to exit [11:01] oh, you have the same issue than david [11:01] weird, you both have it and with all the week-end spam, no report of an issue with latest unity [11:01] running "unity" works [11:02] hum no [11:02] hum… it shouldn't :) [11:02] it just starts compiz [11:02] (I had unity turned off again in ccsm) [11:02] and when starting unity, it stops? [11:03] oh ok [11:03] so you changed your unity profile [11:03] it's easier to debug with a wm working :p [11:03] don't tell me! :-) [11:03] yes, I'm on my GNOME session [11:03] it's the standard compiz [11:03] I didn't touch ccsm out of activating unity [11:04] ok, should be good then setting-wise [11:04] not sure about the crash/exit [11:04] I can just say that there is a similar report for a week [11:24] pitti - i'm just about to upload a lucid SRU for pidgin (bug 683076), but the patch is absolutely massive. do you want to take a look before i upload? [11:24] Launchpad bug 683076 in pidgin (Ubuntu) "pidgin can no longer log in to ICQ (affects: 27) (heat: 372)" [Undecided,Confirmed] https://launchpad.net/bugs/683076 [11:26] chrisccoulson: massive in size, or in how much code it touches? [11:26] chrisccoulson: weird, I thought the one for maverick was just changing certificates or so (due to the server move) [11:27] chrisccoulson: but if you tested it, please upload; easier to review from teh queue [11:27] pitti - are you thinking about the recent change for MSN? that was quite a small patch [11:27] chrisccoulson: I remember that [11:28] chrisccoulson: but I thought we also fixed ICQ in maverick-updates [11:29] pitti - yeah, i see that now. the maverick patch does look smaller, but it's more than changing a certificate [11:29] i'll take a look at that [11:30] the lucid diff seemed small enough [11:30] out of the guy who did the debdiff included the .orig in it [11:32] yeah, i just realised that ;) [11:32] d'oh! [11:32] hehe [11:32] so it's smaller than it looks ;) [11:32] so it's ok? ;-) [11:32] when you can check the upstream commit [11:32] it seemed mostly ok to me iirc [12:20] hello everyone [12:27] hey nessita [12:27] how are you? [12:27] hey rickspencer3 [12:28] hi seb128 [12:28] hey seb128! pretty good, thanks. How about you? [12:29] I'm fine thanks [12:29] rickspencer3, how was the qa sprint? are you back on your side of the world? [12:31] seb128, sprint was great, very productive, especially from the SRU testing perspsective [12:31] and I'm in Seattle now, got here yesterday "afternoon", though my body clock is a bit screwy ;) [12:32] ok [12:32] rickspencer3, is there any summary of the sprint or its outcome somewhere? [12:32] or will someone write one? [12:33] seb128, that can be done easily [12:33] victorp kept it very well organized [12:33] so he has index cards for each task started and completed [12:34] nice [12:34] I'm mostly curious to know if there anything we can pick up to improve what we do [12:34] like writing automatic testcases for SRU [12:37] seb128, well, there are 3 kinds of testing that we covered [12:38] 1. SRU certification testing [12:38] so, if we are going to do an SRU, ensure that machines that certified to run Ubuntu in the Canonical cert lab stay that way [12:38] 2. SRU regression testing [12:38] try to make sure that a new kernel does not cause regressions, these are deeper tests, but on fewer machines [12:39] 3. Desktop integration testing [12:39] this is a set of daily tests to ensure that bugs and breakages are caught early [12:39] I think we need more tests for #3 [12:39] seb128, was that the kind of thing you were wondering about? [12:40] rickspencer3, if we should encourage sru uploads to come with a test testing the issue being fixed with the update [12:40] so we increase our testing coverages for applications as well and make sure the things we fix have a test [12:40] seb128, is that to verify that the fix works, or to ensure that there are no regressions in that area in the future? [12:41] both [12:41] seb128, that makes a lot of sense [12:41] rather to make sure we test for this bug so it doesn't come back [12:41] seb128, that counts for bugs on desktop apps too, right? [12:42] right, I was rather thinking about applications bugs [12:42] those are easier to test [12:42] ah [12:42] that would be sweet, yes [12:42] if we care enough about a bug to SRU a fix we should be careful it doesn't come back [12:43] true [12:44] seb128, so, I need to write this up ... for testing apps I am seeing there are basically 3 ways to go about it [12:44] 1. unit tests that work through the gtk builder object [12:44] 2. GUI tests that use the accessability layer [12:44] 3. sikuli (sic?) tests that work by taking screenshots [12:44] between the 3, I don't think there is anything that is "untestable" [12:45] #2 has the advantage of being easy to write tests when there are async aspects to the test [12:45] like, it's easy to test if signals and such are working, if threads returns, etc... (well compared to #1) [12:46] so we discussed last week that we may move all of the mago tests to trial and testtools, and create a MagoTest class in testtools [12:46] (to make it easier to write tests with the accessibility layer) [12:46] [12:46] :) [12:48] rickspencer3, I see, I guess having a small example of each cases on a page and pointing that page from the SRU documentation would be a start [12:49] like "you are encouraged to write a test for the bug you fix, see that page if you need examples" [12:55] seb128, right, we'll get there, and soon, lots of good work going on [12:56] great [12:57] seb128, is it safe for me to dist-upgrade today? is that Python 2.6->2.7 transition settled? [12:57] it should yes [13:00] tell us how it works for you ;-) [13:00] haha [13:00] well, if I disappear and don't come back, you'll have your answer [13:00] * rickspencer3 wants new launcher features [13:08] rickspencer3, i have to get new hardware if i want to run Unity in 11.04 :( [13:08] nisshh, :/ [13:08] and im broke, which doesnt exactly help :) [13:08] nisshh, don't worry, seb128 will ensure that you get a great desktop experience with your current hardware [13:09] you'll still be able to run 11.04, and you'll get all the same apps and capabilities, just a different shell [13:09] rickspencer3, heh, well, i need to upgrade anyway, this is 8 year old hardware :) [13:09] wow [13:09] rickspencer3, but i want Unity... *goes and cries in the corner* :) [13:10] hehe [13:10] rickspencer3, yep, i have been running Ubuntu on a single core with 1GB of RAM :) [13:11] once i get new hardware im going to run irssi in screen on this machine so im on irc 24/7 [13:12] rickspencer3, what launcher features do you want? [13:12] rickspencer3, is it unusual that compiz works on this machine with 10.04, but it says i cant run desktop effects in 11.04 [13:12] I want a way to start the same software several times [13:12] I might be weird but I tend to open several firefox, I don't always use tabs in the running instance :p [13:12] seb128, *raises hand* i would like to actually be able to run Unity :) [13:13] nisshh, desktop effect has not been updated for the new compiz [13:13] seb128, I want the "start new instance of an app on the current workspace" feature [13:14] seb128, yeah, i get the "your computer cannot run/use 3d acceleration" or whatever when i try to use Unity in 11.04 [13:14] so you can switch desktops, then use the launcher to create new instances instead of activating current instances [13:14] yet i can run compiz fine on 10.10 [13:15] nisshh, what videocard do you use? [13:16] rickspencer3, can we turn that to "let me start several instance of the same applications" so it fixes my issue as well? ;-) [13:16] rickspencer3, can we turn that to "let me start several instance of the same applications" so it fixes my issue as well? ;-) [13:16] seb128, intel IGP on an 8 year old motherboard *waits for seb128 to wince* :) [13:16] ups [13:16] nisshh, can you use compiz without unity on natty? [13:17] seb128, I think that's what the fix is [13:17] great [13:17] seb128, no, i get that message when logging into Unity on natty and then i try and log into the classic desktop, but it doesnt like it [13:17] rickspencer3: it's not, it was a debugging option that will be turned off next release [13:17] by default, it will activate, but you can middle click or select "launch new instance" in the menu [13:17] didrocks, oh? [13:17] alex had it on in the sprint last week [13:18] didrocks, I *neeeeed* this feature [13:18] is it hard to turn on? [13:19] i *may* be able to get my hands on a dual core with 6GB of RAM early next year though :) [13:19] but it wont really be worth it [13:19] nisshh, weird, normal compiz should still work [13:20] didrocks, what is a debugging option? [13:20] the "start in a new windows"? [13:20] seb128, yeah, it was odd, i couldnt log into the classic desktop, i cant remember why it wouldnt let me [13:20] from memory it just refused to go past the login screen [13:21] that needs debugging I guess [13:23] urg, unity crashed === alecu_ is now known as alecu [13:23] i can't run hamster in unity! :( [13:24] there is a bug about that [13:24] oh, that would be nice to have :) [13:24] i use it all the time normally, and today is the first day i'm using unity during my working day [13:24]