[00:18] ScottK: yay! [00:18] (and thanks) [01:36] yofel: Those are the source that the .png's are made from. [01:36] So we need them in source, but not binary. [01:37] AIUI anyway. [01:37] ^yup === rbelem_ is now known as rbelem [06:30] haii [07:01] needs kubuntu review: Bug #1011937 (tried subscribing kubuntu-dev, but realized that would end up spamming about 85 people (and am hoping that won't happen) [07:01] Launchpad bug 1011937 in libqalculate (Ubuntu) "PLease merge libqalculate 0.9.7-7 from Debian sid" [Undecided,Incomplete] https://launchpad.net/bugs/1011937 [07:12] Riddell: ScottK: would you mind if I set kubuntu-devel@l.u.c as the contact for the kubuntu-dev team? === kubotu_ is now known as kubotu [07:33] Can anybody have a look at https://bugs.launchpad.net/ubuntu/+source/libqalculate/+bug/1011937 ? [07:33] Launchpad bug 1011937 in libqalculate (Ubuntu) "PLease merge libqalculate 0.9.7-7 from Debian sid" [Undecided,Incomplete] === frederik_ is now known as fregl [07:44] beta 2 report is up: http://felix.fobos.de/kubuntu/4.8.90.htm [07:44] lots of fixing to do [08:04] micahg: that seems fine, why is it needed? [08:05] debfx: wow impressive, well done [08:05] vibhav: can do in a bit [08:07] debfx: yay (does that use list-missing?) [08:07] er, I mean not-installed [08:07] yofel: yep [08:07] debfx: so e.g. kate, why is that not green? [08:08] I can't see any missing build deps or --list-missing output [08:08] Riddell: because of the "Could NOT find SharedMimeInfo" [08:08] -- Could NOT find SharedMimeInfo (missing: UPDATE_MIME_DATABASE_EXECUTABLE) (Required is at least version "0.30" [08:08] i believe [08:08] ah hah [08:09] highlighting the problems in the cmake output is on my todo list [08:10] debfx: and you weren't excessively fristrated by launchpad being slow when building things? [08:10] frustrated [08:11] that depends on how fast you need things. IMO it's tolerable, but you won't get the whole KDE SC done within a day if things break [08:12] our private ppa has a high enough priority anyway so we don't need to way that long for things to build [08:13] yeah I think it's good enough for us [08:14] *wait [08:15] that's true, the private one is high priority [08:16] https://help.launchpad.net/Packaging/BuildScores - for builds in staging or so we could upload with medium urgency (I do that sometimes) [08:17] we shouldn't really need that though [08:39] -./usr/share/man/man1/dragon.1 [08:39] time to reimplement list-missing proper I guess [08:39] in make \o/ [09:12] vibhav: you've a couple comments on that qalculate merge on things that can be simplified, are you able to do those or shall I just do them during prep&upload? [10:15] 27/49 votes [10:24] apachelogger: all the candidates are so good, I don't know how to choose between them! [10:24] I know [10:25] next time we should just roll a dice, less of a headache :) [10:57] Hey all === Blizzzek is now known as Blizzz [12:06] vibhav: voila [12:06] thanks for helping ubuntu [12:11] \o [12:11] Quintasan: fix fix! http://felix.fobos.de/kubuntu/4.8.90.htm [12:12] * Quintasan sighs [12:14] Riddell: any ideas why kiten on i386 is FTBFS when it built just fine? [12:15] Quintasan: I think it's red in debfx's script because it has a list-missing file [12:15] * Quintasan looks at kde-workspace [12:15] which is fine [12:17] and that file is in the debian/not-installed file so it's just debfx's script not reading debian/not-installed [12:17] urgh [12:17] uploading workspace will be a hell [12:18] Quintasan: why? bandwidth? [12:18] more like the upload speed [12:19] Quintasan: need an ec2? [12:19] Riddell: If possible [12:19] Riddell: kiten's not-installed has an invalid syntax. comments are only allowed in separate lines. [12:19] debfx: Should I change that? [12:19] Quintasan: I've already uploaded the kde-workspace orig tarball so you don't have to upload it again [12:19] Quintasan: yes, please [12:20] debfx: Thanks and ok. [12:20] Riddell: No need for ec2 then, thanks anyways [12:22] debfx: invalid according to whom? [12:23] Riddell: according to pkg-kde-tools [12:26] debfx: file a bug with debian while you're doing it then [12:26] Quintasan rather not debfx [12:26] [kiten] Michal Zajac * 35 * debian/not-installed Fixed not-installed syntax so pkg-kde-tools doesnt complain [12:28] debfx: I'm not uploading that to quantal, not that important. [12:29] * Quintasan looks at kde-workspace again [12:29] Quintasan: but do put it in bzr [12:30] I think I did that, unless kbzr is bugged [12:30] Then I'll bash apachelogger [12:33] * Riddell looks into kdepim-runtime [12:34] Riddell: It doesn't build? [12:34] Oh wait, Kolab magic [12:34] Riddell: I'm not sure it works with Kolab [12:34] also a libkgapi needed [12:34] !newpackage libkgapi [12:34] Sorry, I don't know anything about newpackage libkgapi [12:34] kubotu: newpackage libkgapi [12:34] incorrect usage, ask for help using 'kubotu: help newpackage' [12:35] kubotu: help newpackage [12:35] newpackage [description]; Adds needs-packaging bug for entirely new package. [12:35] newversion [description]; Adds update bug for existing package. [12:35] oh [12:35] neat [12:35] kubotu: newpackage libkgapi 0.4.0 for kdepim-runtime [12:35] Package libkgapi already exists! [12:35] mm, no it doesn't [12:35] Now if we can get assign fuction [12:36] apachelogger: kubotu is wrong! [12:36] kubotu: you are wrong [12:36] Riddell: tell that to yofel [12:37] bug 1012103 [12:37] Launchpad bug 1012103 in Ubuntu "[needs-packaging] libkgapi" [Undecided,New] https://launchpad.net/bugs/1012103 [12:37] Riddell, yofel: perhaps it does a binary check rather than a source check? [12:37] there isn't any source, I think it's just checking it exists in launchpad at all which isn't enough [12:41] Riddell: If that thing is not using qmake I think I can get it done [12:41] libkgapi is in new queue [12:41] aah [12:42] feel free to review :) [12:42] oh jtechidna [12:42] * Quintasan hugs jtechidna [12:42] :) [12:43] Riddell: One has to have archive admin powers to review stuff in main? [12:44] Quintasan: yes [12:44] next stop for Quintasan -> archive admin [12:45] though that would require great insight into policy [12:45] not sure if I want to know all that stuff [12:45] Quintasan: yes you'd need to show you had wide and deep experience with policy and licencing needs [12:45] and be a core dev [12:46] then I'm sure you'd be snapped up [12:46] I think I'll stick with being core-dev if I ever make it there [12:49] jtechidna: accepted! [12:49] thanks! [12:49] * Quintasan fires up make -j32 and prays he doesnt run oom [12:49] Any word on whether the Kubuntu packageset will mostly be moved to universe? [12:49] -j4 forces me to use swap :s [12:49] lol [12:49] jtechidna: how much memory do you have [12:49] 4 GB [12:50] @_@ [12:50] jtechidna: yes it will, just waiting on cjwatson to do that [12:50] I have 8GB [12:50] Quintasan: there is some overhead, which would be about the same for both of us [12:50] jtechidna: CD limit is now 1GB and are built from universe (although no CDs of any flavour this week for unknown reasons) [12:50] Riddell: great [12:51] Riddell: What will stay in main? [12:51] Qt will be there for sure, are they moving all KDE stuff? [12:51] Quintasan: whatever something else in main build-deps on [12:51] Ah [12:51] I see [12:51] I was thinking libkgapi would probably need a symbols file for a MIR, if required. But even if it doesn't, I'll add it in an upload. [12:51] Quintasan: he posted a list to kubuntu-devel [12:51] * Quintasan is too scared to open that [12:52] I had my share of notawesome on the kde-packagers today [12:52] uhh [12:53] Krap [12:53] I guess we are still double-building kwin [12:53] Quintasan: kde-packager list? I don't see anything from you [12:54] The whole "discussion" fills me with disgust. [12:54] I was tempted to respond but decided not to add any more fuel to that [12:55] Quintasan: kdelibs stuff? I've been mostly ignoring it [12:56] Well, I intended to do so but the thread was getting longer and longer and I thought something might have really happened [12:57] Quintasan: It looks like a soprano patch will sort things. [12:58] http://lists.kde.org/?l=kde-core-devel&m=133950457802942&w=2 [13:00] Quintasan, jtechidna, yofel, debfx: I've edited the pad to be only the packages still with issues, please add your name and the status when you're working on them http://notes.kde.org/kubuntu-ninjas [13:04] ScottK: I think the double bars aren't necessary on the pad, it's just a legacy from the wiki [13:04] in my opinion [13:04] OK. Just following the pattern. [13:44] OK. No can haz soprano until mysql is fixed. [13:48] ScottK: really? [13:48] since when did it use mysql? [13:48] It doesn't. librdf0-dev : Depends: libmysqlclient-dev but it is not going to be installed [13:49] Build-dep is uninstallable on !i386 at the moment. [13:50] fooey [13:59] [kde-workspace] Michal Zajac * 655 * debian/ (changelog kde-window-manager-common.install) Fixed install files [14:00] Riddell: workspace should build fine now [14:01] Quintasan: updates in ninjas and bzr ? [14:02] Riddell: Yeah [14:04] Quintasan: lovelyness [14:04] Riddell: I don't think kate actually needs shared-mime-info, it just wants to run update-mime-database which makes no sense in the package build [14:05] debfx: aah, maybe [14:20] debfx: however it makes your script turn green so worth having just for that maybe, I'm adding qjson to ark for the same reason (only needed for unit tests which we don't compile) [14:21] Riddell: I have an ignore list for that [14:21] debfx: oh, in your script? [14:22] yes, however there seems to be a bug in the ark cmake script [14:23] instead of "could not find qjson" it displays "CMake Warning at /usr/share/kde4/apps/cmake/modules/MacroOptionalFindPackage.cmake ..." [14:23] mm yeah [14:25] debfx: so I can just add the build-dep to stop another yellow warning that would need to be invesigated each time? [14:27] Riddell: you can just tell me to extend https://bazaar.launchpad.net/~debfx/+junk/kubuntu-automation/view/head:/cmake-ignore.json [14:28] debfx: ok, please extend https://bazaar.launchpad.net/~debfx/+junk/kubuntu-automation/view/head:/cmake-ignore.json :) [14:29] ^^ [14:29] I have already added kate and ark [14:29] lovely [14:29] debfx: I'll remove the build-dep from kate then [14:32] debfx: for kross-interpreters it's requesting ruby and falcon, I'm not going to add them cos I don't think anything uses them [14:33] I heard that non-canonical archive admins can do package removals now? [14:33] jtechidna: We can. [14:34] cool [14:34] if one of you could take a look at bug 1012151, it would be appreciated [14:34] * jtechidna slaps ubottu [14:34] Error: Could not gather data from Launchpad for bug #1012151 (https://launchpad.net/bugs/1012151). The error has been logged [14:34] * ScottK blames launchpad. [14:34] fair point :P [14:35] I don't think I can do the sync blacklist though. [14:35] Riddell: Can we do sync blacklist changes? [14:36] ScottK: I've no idea [14:37] ScottK: sync stuff all got rewritten by cjwatson recently [14:37] debfx: I'm also not adding ruby to kdewebdev, not a build-time thing [14:40] Knights is the other 3rd-party package using libkdegames, but its upstream is fairly active so there's still hope [14:42] Riddell: run time checks in KDE build systems seem like common issues since (AIUI) for people who build directly from upstream source the distinction is less important. Such things seem like bugs that ought to be reported/fixed upstream. [14:44] ScottK: it's unclear how to fix them upstream, a build time check is still a useful thing to have for people who build from source and for packagers who need told of runtime deps [14:44] it's why I kept it for printer-applet with a cmake flag to override it [14:44] but I think the cmake flag doesn't stop it complaining which is a bug [14:45] I think that's a useful middle ground since, as you say, there isn't a really good solution at the moment. [14:56] Riddell: kdewebdev, kross-interpreters, ark and kate are green now [14:57] yay [14:57] debfx: you're on a roll [14:59] eveningz [15:00] yo yo yofel === maco2 is now known as maco [15:38] having a dependency prob in 4.8.80, ffmpeg and libav-tools depends are in the vicious cycle = dependency hell :/ [15:39] BluesKaj: installed from where? [15:43] 12.10 repos [15:44] http://paste.ubuntu.com/1037423/ [15:51] /usr/include/boost/spirit/home/qi/parse.hpp:83:9: internal compiler error: Segmentation fault [15:51] when compiling rocs. ouch [15:51] the terminal prompt insists on opening at ~/Documents [15:52] kaj@Test-Drive:~/Documents$ [15:52] BluesKaj: do you know what wants to install ffmpeg and what wants to install libav-tools ? [15:52] I do [15:53] Riddell, ^ [15:53] BluesKaj: and.. ? [15:53] Riddell: shadeslayer_ is our resident gcc bug reporter ;) [15:53] I'm trying to install it manually since it's missing [15:54] shadeslayer_: rocs is needing some gcc bugage! [15:56] debfx: for the precise backport it seems a good idea to build it in -ninjas again [15:56] but I think ninjas has your 4.8.4 packages still [15:59] hm staging already has 4.8.80 :( [15:59] debfx: that can be deleted [15:59] since it's old news now [16:00] ok, let's see how klearppa works [16:01] ImportError: No module named KubuntuDevTools.launchpad :( [16:02] * debfx adds PYTHONPATH [16:03] duemduem [16:05] it's probably bug 1010896 [16:05] Launchpad bug 1010896 in gcc-4.7 (Ubuntu) "Internal compiler error compiling c++ source files using boost. The same file compiles ok with 12.04 and gcc 4.7 from toolchain ppa" [Undecided,New] https://launchpad.net/bugs/1010896 [16:07] Riddell: the address should be set on the team so that if kubuntu-dev is subscribed to something, it doesn't spam 85 devs, but sends it to the mailing list [16:08] why is kubuntu-dev subscribed somewhere? [16:10] * apachelogger goes all fistwave on multiarch [16:12] bug #1010896 has been fixed :) [16:12] Launchpad bug 1010896 in gcc-4.7 (Ubuntu) "Internal compiler error compiling c++ source files using boost. The same file compiles ok with 12.04 and gcc 4.7 from toolchain ppa" [Undecided,Fix released] https://launchpad.net/bugs/1010896 [16:13] micahg: aye go ahead then [16:14] do we really want bugmail on the kubuntu-devel list? [16:15] no [16:15] kubuntu-bugs [16:15] ! [16:15] or some null address, kubuntu-dev ought not be subscribed to anything anywhere [16:16] micahg: ^^ [16:18] debfx: this is if someone specifically needs kubuntu devs to review something, a null address would work as well (it's also nice to see where to send mail to the team) [16:19] apachelogger: I'm speaking of a manual subscription like I did last night [16:19] micahg: someone would then want kubuntu-members to review something, not kubuntu-dev [16:19] elitst-- [16:19] apachelogger: kubuntu members for kubuntu in general, kubuntu-devs for developer questions [16:20] but the general idea is to have an address set so that people can see how to contact the team [16:21] or that if the team gets subscribed to some artifact, it goes to a sane place instead of the inbox of all the members [16:21] micahg: so it would still be kubuntu-devel@uc [16:21] apachelogger: yes, that's what I was suggesting [16:21] same as kubuntu-members [16:21] micahg: all goody then [16:22] people who subscribe the list somewhere pointless get a beatin from debfx then :P [16:22] * apachelogger scuttles off to organize rebuilds [16:22] oh, hrm, that seems a little weird (having kubuntu-devel as the address for kubuntu-members), but if it works for you folk, so be it :) [16:23] I'm guessing I can't set the same contact address twice though [16:23] micahg: I think the list was there first, so that simply got used for all and everything [16:24] ah, ok, historical perspective is nice :) [16:24] micahg: and that setting twice issue is bringing me back to the point that kubuntu-members is the authority and not kubuntu-dev :P [16:24] * apachelogger suspects launchpad gets a mapping problem there as it likes to map maintainers and whatnot to actual teams with that address [16:25] apachelogger: kubuntu members can't tell me if a change needs to be dropped in a package though :) [16:25] *reliably tell me, nor can the membership of kubuntu-dev [16:25] it's really quite frustrating :) [16:26] but then again, all the flavor dev teams have the same issue (but I think they all have ML addresses set) [16:26] depends on whether one looks at it from a political or practical POV [16:27] * micahg tries to stay away from the politics and stick to the practical [16:29] well.... the KC being almighty and whatnot is representing -members, so I suppose at any rate a KC ay/nay is always outweighing anything a kubuntu-dev member says ;) [16:30] (surprisingly enough no one ever bothered to give kubuntu-dev as a collective any authority other than approving new members for kubuntu-dev) [16:30] micahg: maybe just have a launchpad list for kubuntu-dev and subscribe kubuntu-devel to that? [16:31] or have the sysadmins make a mail alias kubuntu-dev -> kubuntu-devel [16:42] Please try again [16:42] Sorry, there was a problem connecting to the Launchpad server. [16:42] * apachelogger sighs === koolhead17|zzZZ is now known as koolhead17 [17:03] kubuntu-devel discards posts from non-subscribers anyway, so I think sending bugmail there ~ sending to /dev/null. [17:06] Riddell: already reported [17:06] Riddell: it's a bug in 4.7.1 specifically [17:07] hmm ... looks like they fixed it [17:08] * shadeslayer_ retries digikam [17:13] herp derp [17:15] question, is it normal for -dev packages to contain so versions? [17:17] they usually only contain the non-versioned symlink to the versioned .so files [17:17] hm .. libexpat1-dev looks like a exception then [17:54] another question, say a package hasn't been officially uploaded to the debian archive, but, debian packaging for it exists, and I want to sync the packaging in ubuntu, do I just copy over the packaging and lose the entire changelog? [17:54] or do I merge by hand preserving the changelog [18:25] apachelogger: in the beta PPA there is no dragon, no kmix and no Juk, but those are needed for kdemultimedia [18:26] cause they aint packaged for .80 [18:26] whcih is why I suggested .80 should not be published... [18:26] OK [18:26] doesn't make sense indeed [19:17] Most of those didn't have proper licensing for .80 either. [19:17] Fixed in .90 AIUI though. [19:31] yofel: does launchpad always take that long to copy packages from a ppa to another? [19:34] I've started it like half an hour ago and it's still not finished [19:36] debfx: depends on the backend load, I've had it taking over half an hour already in the past, so not unusual - for the whole SC at least, single packages shouldn't take that long [19:43] it finished \o/ [20:08] ::runtime-bugs:: [1011961] sftp connection with password fails @ https://bugs.launchpad.net/bugs/1011961 (by Arrigo Marchiori) [20:08] Launchpad bug 1011961 in kde-runtime (Ubuntu) "sftp connection with password fails" [Undecided,New] [20:22] apachelogger: starting to get usefull :P http://wstaw.org/m/2012/06/12/plasma-desktopgk1991.png [20:23] bbl [21:05] just bumping: https://wiki.ubuntu.com/SteveRiley [21:51] claydoh: why are we bumping? [23:17] MySQL is fixed (FSVO fixed), so soprano is building now. should have [23:18] ... .57 within an hour modulo publisher run timing. [23:20] claydoh: Is he applying for membership? [23:36] ScottK: yes, i am :) [23:38] Great. It would be good to get more than one endorsement. [23:39] Also, please show up at the meeting with an opinion about owncloud. Since you're a cloud dude, I'll be asking. [23:40] i have asked the other kfn admins to contribute [23:41] and yes, i do have some opinions about owncloud. been playing with one installation on amazon web services and another one at home