/srv/irclogs.ubuntu.com/2011/08/18/#kubuntu-devel.txt

=== claydoh_ is now known as claydoh
ScottKQuintasan: Can haz diffs for the FFes?00:21
ScottKrbelem: pong00:22
jjessehere's an interesting problem just updated my oneric vm today and after an update, rebooting and trying to login i get a white box that says "Could not start D-Bus, Can you call qdbus? and has an Okay button, after clicking the button i am prompted again with the login screen00:42
yofeljjesse: known issue, should be fixed with the next qt update, for now install qt4-dev-tools00:44
jjessethanks yofel00:45
rbelemScottK, could you take a look in the ffe? :-)00:49
jjessehrmm ok last night i reported a bug i was having w/ kate and it appears to only be happening in an upgraded version of oneric and not a fresh version01:11
jjessehowever i reported it on bugs.kde.org and now can't find it ;(01:11
jjesseah found it: https://bugs.kde.org/show_bug.cgi?id=28023601:12
ubottuKDE bug 280236 in general "problems with second level menu options" [Normal,Needsinfo: waitingforinfo]01:12
jjesseemail about it send to the kubuntu-devel mailing list01:18
ScottKrbelem: Which?02:49
=== ejat- is now known as ejat
=== ejat- is now known as ejat
=== skreech_ is now known as Daskreech
bambeemorning08:03
Quintasanbambee: \o09:14
QuintasanGood morning as well09:14
bambeeyo :)09:14
QuintasanScottK: You can09:15
QuintasanScottK: https://bugs.launchpad.net/ubuntu/+source/kde-workspace/+bug/82815909:16
ubottuUbuntu bug 828159 in kde-workspace (Ubuntu) "kde-workspace needs to be built with kwin-gles" [Undecided,New]09:16
QuintasanScottK: https://bugs.launchpad.net/ubuntu/+source/kde4libs/+bug/82728609:19
ubottuUbuntu bug 827286 in kde4libs (Ubuntu) "kde4libs package needs to be updated" [Undecided,New]09:19
Quintasanulmlogger: Can I merge kde-runtime-active branch?09:20
QuintasanWUT09:21
Quintasanyofel: PINGO09:21
Quintasanhttp://bazaar.launchpad.net/~yofel/kubuntu-packaging/kde-runtime-active-4.7/revision/22309:21
QuintasanY U NO PATCHES?09:22
ulmloggerQuintasan: from where?09:22
* yofel yawns09:22
ulmloggerQuintasan: I acked the stuff rbelem presented09:22
yofel@_@09:22
ulmloggerlike 3 times or more :P09:22
yofelsry09:22
Quintasanulmlogger: https://code.launchpad.net/~yofel/kubuntu-packaging/kde-runtime-active-4.709:23
jussiulmlogger: PM !!! :D09:23
Quintasanulmlogger: I was looking at that branch but I discovered there are no patches :S09:23
CIA-52[kde-runtime] Philip Muškovac * 224 * debian/patches/ (6 files) Actually add the patches09:23
QuintasanNow I can merge and link to branch!09:24
* Quintasan testbuilds09:25
Quintasanhttp://paste.kde.org/11125909:26
Quintasanhmmm09:26
Quintasan--with xine?09:26
QuintasanIs that some oneeyerick change?09:26
yofelsure it's oneiric since it's only there since dhmk is used09:27
* Quintasan bzr-buildpackage -S09:29
Quintasancan't*09:29
QuintasanI see09:30
QuintasanTESTBUILDIN09:31
Quintasanwerks09:48
QuintasanScottK: https://bugs.launchpad.net/ubuntu/+source/kde-runtime/+bug/82728309:53
ubottuUbuntu bug 827283 in kde-runtime (Ubuntu) "kde-runtime needs to be updated" [Undecided,New]09:53
QuintasanScottK: Then we have FFe for https://bugs.launchpad.net/ubuntu/+bug/82811109:54
ubottuUbuntu bug 828111 in Ubuntu "[needs-packaging] share-like-connect" [Wishlist,New]09:54
Quintasanhttps://bugs.launchpad.net/ubuntu/+bug/51231909:54
ubottuUbuntu bug 512319 in Ubuntu "[needs-packaging] plasma-mobile" [Wishlist,Confirmed]09:54
Quintasanand https://bugs.launchpad.net/ubuntu/+source/plasma-mobile/+bug/82809309:54
ubottuUbuntu bug 828093 in plasma-mobile (Ubuntu) "plasma-mobile needs to be updated" [Undecided,New]09:54
Quintasandisregard 51231909:54
Quintasanbambee: file a FFe for touchegg09:54
bambeeok09:55
bambeesure09:55
Quintasanand ping ScottK with it, in bug description link to REVU package which I acke'd09:55
QuintasanScottK: I need you to approve and upload FFe's for kde* packages so I can testbuild the rest and see if it works09:56
QuintasanIf something doesn't work then we blame rbelem09:56
* Quintasan goes back to cleaning 09:56
rbelem:-D10:10
* rbelem pokes Quintasan 10:10
* rbelem goes to work10:14
bambeeQuintasan: bug 82863710:15
ubottuLaunchpad bug 828637 in Ubuntu "Feature Freeze Exception request for Touchégg" [Undecided,New] https://launchpad.net/bugs/82863710:15
QuintasanWhat a beautiful FFe10:16
QuintasanScottK: ^^10:16
bambeeohh... I need a review from another person normally... (It's the MOTU process...)10:16
Quintasanulmlogger: I extend my every right to demand that you spare some time and review bambee's package http://revu.ubuntuwire.com/p/touchegg10:17
QuintasanNot much to review there, and leave an advocating comment10:18
QuintasanUnless he broke something10:18
* Quintasan hands bambee Mighty Stick of ulmlogger Poking10:18
* ulmlogger puts on some pants10:22
* bambee pokes ulmlogger with his Mighty Stick and waits for "pants ack"10:26
ulmloggeryou can already prepare your shell for fixing10:27
ulmlogger2 seconds into the review -> first grave problem10:27
bambeereally? o_O10:27
* jussi steals the stick from bambee10:29
yofelbambee: this looks fooey10:30
yofel  3 Maintainer: Kubuntu Developers <kubuntu-devel@lists.ubuntu.com>10:30
yofel  4 XSBC-Original-Maintainer: Kubuntu Developers <kubuntu-devel@lists.ubuntu.com>10:30
bambee?10:31
yofeleither put yourself as original maintainer or just drop it10:31
bambeelintian displays warnings if I drop it10:31
yofelo.O10:32
bambee(I'am not sure if it's lintian, but something displays warnings)10:32
bambeelet me check10:32
yofelah, dpkg-source10:33
* yofel usually ignores that10:33
bambeeno it's dpkg-source10:34
bambeedpkg-source: warning: Version number suggests Ubuntu changes, but there is no XSBC-Original-Maintainer field10:34
bambeewell, I ignore it10:34
ulmloggerbambee: revud10:35
bambeeulmlogger: thanks10:35
bambee"upstream tar i tar.gz not xz, do not change upstream tars for no good reason" => the original tarball does not contain the COPYING file10:36
bambeethat's why I've repacked it10:36
yofelwell, then you're missing a +repack in the version10:36
bambeeright10:36
yofelalso, run wrap-and-sort10:37
jussiulmlogger: moar PM! :D10:46
Quintasanbambee: I didn't tell you that you need to add +repack?11:40
Quintasansilly Quintasan11:40
Quintasanbambee: Now you got a list of errors from ulmlogger and should fix them12:16
QuintasanThe package should be mostly good afterwards12:16
debfxScottK: do you have an idea why ntrack breaks kde again? bug #82696912:49
ubottuLaunchpad bug 826969 in ntrack (Ubuntu) "Latest ntrack update in Oneiric causes kded4 to spin the CPU" [High,Confirmed] https://launchpad.net/bugs/82696912:49
debfxor rather how we can fix that12:49
ScottKdebfx: I don't.  I tested the fix that's in 014 and still had problems, which is why it wasn't included in the SRU.12:50
ScottKI'd ask asac.12:50
ScottKPresumably someone needs to figure a way to make it consistently do that and strace it or some such.12:51
mfraz74would someone here be able to help me with a compiling problem?12:52
shadeslayermfraz74: is it related to Kubuntu development?13:11
mfraz74shadeslayer: not really I was trying out the http://developer.ubuntu.com/packaging/html/packaging-from-scratch.html and I'm stuck13:12
shadeslayerhmm ... #ubuntu-motu or #ubuntu-packaging i'd say13:13
mfraz74ok thanks13:13
bambeere13:17
bambeeQuintasan: no you just said me to repack it, BUT I know that I need to add +repack... I just forgot it... silly error13:18
shadeslayer!find qrcodec.h13:34
ubottuPackage/file qrcodec.h does not exist in natty13:34
ScottKulmlogger, debfx, Quintasan, rbelem, etc: Looking at the kde4libs diff, I'm not super happy about the idea of adding experimental headers into kdelibs5-dev.13:44
* ulmlogger complained about that13:45
* ScottK is going to (again) suggest that stuff go into a new kdelibs5-experimental-dev package and ask for arguments why that's not the way we should do it.13:45
ScottKulmlogger: ^^^?13:45
ScottKulmlogger: Any idea where the new symbols in debian/libnepomukquery4a.symbols come from?  That seems a bit scary too.13:46
ulmloggerbut everything in an experimental package13:46
bambeeQuintasan: http://revu.ubuntuwire.com/p/touchegg13:46
bambeefixed13:46
ulmloggeractually, for all I care you could just create one experimental package with libs and headers inside13:47
ScottKdebfx: ^^^ re the symbols - do you know?13:47
ulmloggerit is not like this thing is supposed to be used by anyone nor will it be existing for > 11.1013:47
shadeslayerulmlogger: wait a sec, aren't the experimental libs and headers supposed to be internal ?13:47
ulmloggerScottK: what symbols?13:47
ScottKulmlogger: Look in the diff.13:47
ulmloggerdefine internal13:47
ScottKhttps://launchpadlibrarian.net/77504842/371_370.diff13:47
ulmloggerScottK: where?13:47
shadeslayernot public13:47
shadeslayernot shipped in a seprate package to be precise13:48
ScottKshadeslayer: Internal to what?13:48
ScottKCan't very well build plasma-active without them can we?13:48
shadeslayerScottK: Internal to KDE i meant13:48
shadeslayer. . .13:48
ScottKshadeslayer: Internal to KDE doesn't make any sense.13:48
ulmloggerScottK: the new symbols come from new code13:48
debfxScottK: I guess from a patch?13:48
ScottKSigh.13:48
shadeslayeri don't know how to put my thoughts into a proper sentence right now13:49
* shadeslayer needs coffee13:49
ulmloggerScottK: https://git.reviewboard.kde.org/r/102112/13:49
ScottKSo our experimental patch exposes new symbols that we can't guarantee will be supported?13:49
* ScottK looks13:49
* bambee orders a coffee for shadeslayer13:49
ScottKulmlogger: It'd be nice if that'd get committed to the 4.7 branch too since that sounds bug fixish to me.13:50
ScottKOr not.13:52
debfxwait what, we are adding a gazillion of patches that are from the 4.7 branch?13:52
debfxare *not* from13:53
shadeslayerbambee: oh ... now i have 2 coffee cups ...13:53
ScottKThat's what the FFe we are discussing is about.13:54
bambee:D13:54
ScottKSo we can build the latest Active/Mobile stuff.13:54
shadeslayerwhy can't we put it in a PPA?13:54
ScottKThen it's rather not part of Kubuntu.13:54
ulmloggeralso13:55
ulmloggerwe13:55
ulmloggerrather13:55
ulmloggerneed13:55
ulmloggerarm13:55
ulmloggerppa++ for not having arms13:56
shadeslayerso, its a tradeoff between having a new product vs exposing new symbols?13:56
ulmlogger\o/ <-- obs, o <-- lunchpad13:56
shadeslayerheh13:56
ulmloggershadeslayer: new product?13:56
ulmloggerplasma mobile is not new13:56
shadeslayerulmlogger: Kubuntu Active13:56
ulmloggerkubuntu mobile13:56
shadeslayerare we discussing Active or Mobile?13:56
shadeslayerbecause i thought the patches enable you to build Active stuff13:57
ulmloggerACTIVE IS MOBILE13:59
shadeslayer@_@14:02
* Quintasan hits shadeslayer with ACTIVE sign14:03
QuintasanScottK: Well, I'm not really happy about doing this after feature freeze :P14:03
shadeslayerthere are like a bazillion names in KDE now14:03
ScottKQuintasan: Agreed the timing isn't best, but until 4.7.0 was done there really wasn't a way to do it.14:04
shadeslayerKDE Frameworks/KDE SC/Plasma Active ...14:04
QuintasanScottK: So it would be acceptable if we'd ship this experimental stuff in a separate package, right?14:14
ScottKQuintasan: I think so.14:16
Quintasanbambee: The changelog needs to mention the cause of repack14:16
ScottKThe patch that adds the symbols looks reasonably safe/maintainable.14:16
ScottKQuintasan: I would like it though if the nepomukquery patch were updated to correctly reflect the version we added it in our package rather than say 4.8 as the upstream comment does.14:17
Quintasanulmlogger: I believe one can pass --dbg-package=touchegg-dbg directly to %14:18
QuintasanThat way whole override on strip becomes useless14:18
ulmloggeryes you can14:19
ulmloggerthat does not mean it is the best solution :P14:19
QuintasanIn that case?14:19
QuintasanI'm wondering if we are getting any particular advantage in that case14:19
bambeeless characters... inline command... I like it...14:20
bambee(we've not a lot of arguments to pass to dh, we've only two...)14:20
Quintasanbambee: Same here, I did that in other packages for no apparent reason14:20
Quintasan(override I mean)14:20
Quintasanulmlogger: Thoughts?14:20
Quintasanbambee: There is also nice wrap-and-sort script14:21
QuintasanYou can use it :)14:21
bambeewtf?14:21
bambeewoo14:21
Quintasanbambee: add --dbg-package=touchegg-dbg to dh and get rid of the override14:22
bambee+ the changelog14:22
QuintasanI don't think it will have any impact on THIS package14:22
QuintasanYes14:22
QuintasanAnd I will advocate14:23
QuintasanScottK: What about kde-workspace?14:24
debfxjust to be clear, those patches won't be included in kdelibs 4.7 so we will have to carry them on until we remove kde4libs?14:24
ScottKThe nepomukquery patch will be in 4.8.14:24
ScottKThe rest is (after moving the headers and such into an experimental -dev package) in separate binaries.14:25
ScottKWe can drop them if needed later.14:25
debfxI thought there will be no kdelibs 4.814:26
debfxalso some patches touc hplasma.h14:26
ScottKHmm.14:26
ScottKThe activitymanager changes in -runtime look a bit scary too.14:27
ScottKulmlogger: This all seems rather invasive.14:27
QuintasanScottK: Which patch is the nepomukquery one?14:27
ScottKQuintasan: https://git.reviewboard.kde.org/r/102112/14:27
ScottKThat commit is in our diff.14:28
QuintasanSo you want to get "\since 4.8" replaced with "since 4:4.7.0-0ubuntu3" ?14:29
Quintasanin the patch14:29
ScottKYes.14:30
QuintasanThere is only one instance of that, replacing14:30
ScottKBut I don't want any of it uploaded until we've got through the whole set.14:30
ScottKSomeone convince me why I shouldn't freak out about the scale of the activitymanager changes?14:31
bambeeQuintasan: uploaded14:31
Quintasanbambee: Awesome14:31
Quintasanrbelem or ulmlogger may have sufficient explanations for this14:32
QuintasanScottK: So new files from debian/kdelibs5-dev.install should land into other experimental package, is naming it kdelibs5-experimental-dev enough?14:32
ScottKYes.  I think that's good.14:32
ScottKulmlogger: ^^^ ?14:32
* bambee blames "bzr commit" it does not open emacs anymore.... 14:33
* bambee has a look at "man bzr"14:33
bambee(just in case)14:33
shadeslayerbambee: known bug/feature14:34
Quintasanfeature14:34
Quintasansince vim > emacs14:34
Quintasan:P14:34
* Quintasan hides14:34
shadeslayerheh :P14:34
yofelno, bzr feature14:35
ScottKTell Riddell.  He uses emacs, so I'm sure he'd be right on top of fixing it.14:36
bambeeQ_ASSERT(! vim > emacs);14:36
QuintasanScottK: Should Depends on that experimental package be the same as kdelibs5-dev or I can just make it depend of kdelibs5-dev so it pulls everything?14:37
Quintasans/of/on14:37
ScottKThe latter is fine.14:37
QuintasanScottK: http://paste.kde.org/111319 <-- Any other thoughts?14:38
ScottKQuintasan: It needs some kind of DO NOT USE warning in the description.14:39
ScottKTotally unstable API/ABI, may break without warning, etc.14:40
yofelQuintasan: also, IMO i should depend on the experimental lib packages so you can remove them from kdelibs5-dev14:40
yofels/i/it/14:40
kubotuyofel meant: "Quitntasan: also, IMO i should depend on the experimental lib packages so you can remove them from kdelibs5-dev"14:40
yofelsed fail..14:40
shadeslayerlol14:40
ScottKyofel: Good point.14:41
QuintasanAs far as I can see the kdelibs5-dev I merged doesnt depend on them :D14:44
ScottKThen the experimental one should pull them in.14:44
QuintasanScottK: http://paste.kde.org/11132514:46
ScottKI'd have gone with unstable API rather than production environment.14:46
ScottKWhen someone uses this package and later their stuff doesn't build, I want to be able to say "Dude, did you read the package description - what did you think would happen".14:47
Quintasan This package contains headers with unstable API and should not be used in production development ?14:48
* Quintasan has no idea how to put that in words14:48
QuintasanHeaders in this package provide unstable API, use at your own risk?14:49
ulmloggerUSE NOT AT ALL14:49
QuintasanScottK: ^14:50
ScottKQuintasan: As ulmlogger says.14:50
yofeldon't use unless you know what you're doing?14:50
ScottKNo.  Don't use.14:50
yofelk14:50
Quintasan Headers in this package provide unstable API, do not use them.14:51
* yofel would drop 'them' but otherwise ok14:51
QuintasanTrue14:51
* ScottK is with yofel, but either way.14:51
QuintasanScottK: The whole diff -> http://paste.kde.org/11133114:52
ScottKulmlogger: Convince me on the activitymanager in -runtime.14:52
Quintasanulmlogger: DO EET14:52
ScottKCould this stuff get shoved in some experimental package that's not installed normally?14:53
ScottKQuintasan: That looks good for libs.  No someone needs to sell me on runtime.14:53
* Quintasan pushes14:53
QuintasanScottK: Can you look at workspace while ulmlogger is looking for arguments? :P14:55
ScottKWhat bug is that?14:55
Quintasanhttps://bugs.launchpad.net/ubuntu/+source/kde-workspace/+bug/82815914:55
ubottuUbuntu bug 828159 in kde-workspace (Ubuntu) "kde-workspace needs to be built with kwin-gles" [Undecided,New]14:55
ulmloggerScottK: well, yes, we could do a double build14:58
ulmloggerfirst without patches, second with14:58
ScottKulmlogger: Wouldn't that be less risk?14:58
ulmloggersure it would14:58
ulmloggeralso it is a PITA to do :P14:59
ScottKThe libs stuff seemed mostly isolated to new stuff.  -runtime not so much.14:59
ScottKFortunately Quintasan is an expert now that he did the kwin-gles stuff.14:59
QuintasanFFFFFFFF14:59
ulmloggerwin14:59
* ulmlogger needs to shower and stuff14:59
QuintasanHow do you expect the builder to know when to apply some patches and when not to?15:00
Quintasanbuilding gles was easy since I didn't have to do any magic apart from few overrides15:00
yofelyou run quilt between the builds?15:00
yofelwhich would probably be messy though..15:01
Quintasan!15:01
QuintasanCan we have a separate series file and pass that over to quilt during the second build?15:01
yofelyou could redefine QUILT_PATCHES though maybe15:01
Quintasanor15:02
yofelthat would give you a seperate series file too15:02
Quintasanmore dirty15:02
Quintasanadd the patches but remove them from series15:02
Quintasando the first build15:02
Quintasanappend new patch names in rules before second build15:03
Quintasando a second build15:03
Quintasanyofel: I don't want to redefine it since we need all the previous patches too15:03
yofelwell, apply patches -> redefine -> apply again15:04
yofeldpkg-source will apply the default ones anyway15:04
ScottKQuintasan: I don't understand "+Breaks: kdebase-workspace-bin (<< 4:4.5.80), kde-window-manager (= ${binary:Version})"15:04
yofelyou would need to pop again between the configure and build step though I think15:05
ScottKDoesn't it break every version of kde-window-manager?15:05
ScottKAlso short description should mention GLES.15:05
ScottKSame for both packages15:06
ScottKOther than that it seems OK.15:07
ScottKI approved the FFe, but please fix that stuff up.15:08
ScottKQuintasan: ^^^15:09
QuintasanScottK: Look at the second diff15:31
QuintasanScottK: Those packages cannot be installed at the same time15:31
QuintasanBecause they provide the same files15:31
ScottKQuintasan: I get that.  Why is the breaks versioned?15:31
QuintasanIt shouldn't be versioned?15:32
QuintasanAh15:32
QuintasanStupid question, removing15:32
ScottKGreat.15:32
ScottKQuintasan: That applies to both packages (same issue)15:32
QuintasanScottK: kde-window-manager (>= ${source:Version}) | kde-window-manager-gles (>= ${source:Version})15:35
QuintasanThat's Depends on kde-workspace, is this good?15:35
ScottKThat's fine.15:35
QuintasanConflicts: kde-window-manager-gles15:35
QuintasanThat's all15:36
QuintasanI'm not adding Conflicts: kde-window-manager on kde-window-manager-gles15:36
QuintasanThat's pointless unless I'm misunderstanding how dpkg works15:36
QuintasanIf I have installed -gles and try to install non-gles then it will conflict15:36
Quintasanif I have non-gles installed it will conflict too15:36
Quintasanright?15:37
ScottKI don't think conflicts are bi-directional like that, but I'm not sure.15:37
* bambee hugs Quintasan and asks him to advocate touchegg on revu15:37
bambee:P15:37
Quintasanbambee: Urgh right, let me do kwin15:37
QuintasanScottK: I'll add them back just to be safe15:37
bambeesure, there's no hurry15:38
QuintasanIt won't hurt anyone15:38
QuintasanScottK: http://paste.kde.org/11134315:39
QuintasanBefore I commit15:39
Quintasanbambee: Advocated15:40
ScottKQuintasan: Yes.  Similar changes in the other gles binary too.15:41
Quintasanoh libkwineffects15:42
QuintasanScottK: http://paste.kde.org/11134915:42
ScottKLooks good.15:43
QuintasanPushing15:43
ScottKBefore uploading I'd test to make sure those conflicts really do what you want.15:43
bambeeQuintasan: thanks!15:44
Quintasanbambee: Now convince ScottK to approve FFe15:46
Quintasan:P15:46
ScottKWhat testing has the package had?15:47
QuintasanScottK: I testbuilt and installed it. I don't have any hardware to test how it works15:48
ScottKSo who does?15:48
ScottKDidn't ulmlogger score some hardware?15:48
ScottKIf you can test it, I've no problem with FFe.15:49
* Quintasan looks at ulmlogger15:49
jussiHas anyone tried installing from the CD/usb recently? 15:49
Quintasanjussi: One-eye-rick?15:50
bambeeScottK: actually? it's not tested yet, I've not the hardware for that15:50
jussiQuintasan: yeah15:50
Quintasanjussi: If you are talking about broken DBus then we know about that :P15:50
ScottKbambee: So find someone.15:50
jussiQuintasan: My 15th doesnt install - ubiqity doesnt start15:50
ScottKBroken dbus is fixed.15:50
QuintasanI see15:50
ScottKSounds about right for broken dbus (time wise)15:51
ulmloggerScottK, Quintasan: bulldog98 and I have hardware15:51
ulmloggeralso I am going to do some cursing now15:51
ScottKjussi: Install from the Alpha 3 images and then upgrade.  Those work.15:51
jussiand my upgraded version doesnt even boot - hangs with a white blinking cursor15:51
Quintasanjussi: Did you try upgrading today?15:52
jussiI can make it boot via recovery, then resume and startx15:52
jussiQuintasan: no, I upgraded a few days back15:52
QuintasanAs ScottK said, DBus got fix'd15:52
bambeeCould someone test touchegg with the appropriated hardware please?15:52
jussiahh, lets see15:52
jussiooh, thats lots of upgraded stuff... let me see if it hels15:53
Quintasanbulldog98 or ulmlogger: pull-revu-sources touchegg, compile and try using that15:53
Quintasanwe need that done SOON (TM)15:53
QuintasanThat's sure stupid, Eject doesnt do umount before trying to eject15:59
jussiwas it libqt4-dbus  that needed to be upgraded or?16:01
ScottKjussi: Read kubuntu-devel16:03
jussiScottK: hrm... qdbus does not seem to exist (Im using archive.ubuntu.com, fully dist upgraded) ?16:10
ScottKHmmm.16:12
ScottKjussi: What version of Qt do you have?16:12
jussilibqt4-dev is Version: 4:4.7.3-4ubuntu516:13
ScottKYou don't have the fixed one then.16:13
ScottKWhat you want is ubuntu616:13
ScottKSo you're in the "depending on how long your mirror takes to update" situation.16:14
jussiisnt archive.ubuntu.com the main mirror that goes first? 16:14
ScottKIt is, but it's a mirror nontheless.16:14
ScottKI'd have expected it there by now, but stuff happens16:15
jussioh look at that, apt-get updated and dist upgraded and its getting pulled in now :)16:15
jussi  libappindicator1 libboost-dev are still being held though16:16
ScottKUnrelated.16:16
Quintasan94%16:16
QuintasanOh wait, it's the first build16:16
Quintasan:/16:16
jussialso, anyone know what this is: W: libwmf0.2-7 is already removed. It is recommended to run defoma-app purge libwmf0.2-7.16:17
QuintasanSince it 's a warning you can ignore it :P16:18
jussilol16:19
* Quintasan hides16:19
macoive had a simlar thing show up before16:19
macowhen i ran defoma-app purge  bleh      it did some stuff and said the same thing during it16:19
jussiright, lets go see if kubuntu boots now16:19
Quintasanyofel, ScottK: Can I have a review before I merge? https://code.launchpad.net/~quintasan/kubuntu-packaging/kde-workspace/+merge/7206616:25
ScottKQuintasan: Depends/Build-depends in your changelog entry.16:26
ScottKWhat's Conflicts: kde-window-manager-gles (=< ${binary:Version}) ?16:28
QuintasanWhat the...16:28
QuintasanBRR16:28
* Quintasan deletes that16:29
QuintasanI pushed wrong stuff16:29
jussinope, still no boot :(16:30
QuintasanCan we have /var/log/Xorg.0.log?16:31
* Quintasan goes playing Harvest Moon16:31
jussiQuintasan: when it doesnt boot, it doesnt get anywhere, just hangs with a white flashing cursor16:34
QuintasanCan you boot it in recovery mode?16:34
jussiyes16:34
jussiI do that, hit resume, then startx and get into kde like that16:34
Quintasan:O16:34
jussistrange thing16:35
* Quintasan has no idea how to debug boot up16:35
shadeslayerkdeinit4 is consuming a entire core -.-16:37
shadeslayerjussi: yeah that has happened to me a couple of times16:38
shadeslayersome random update solved the issue for me16:39
* jussi sighs. its sucky to live in a quiet world...16:40
yofelwell, you could either disable splash, or set gfxplayload to text to get actual boot messages (and disable 'quiet'). If you want even more verbose boot add '--verbose' to the kernel command line16:43
ulmloggerQuintasan: no worky worky16:48
ulmloggeralready tried that days ago16:48
ulmloggerthere is some problem with touch input16:48
ulmloggeronly once you use two fingers it will actually be detected as input16:48
ulmloggerso it is a bit of a drag16:48
QuintasanIs it a problem with the application or the input magic?16:48
ulmloggerlitterally :P16:48
ScottKulmlogger: Maybe cnd can help you with that?16:48
ulmloggerQuintasan: not sure16:48
ulmloggerprobably input magic16:49
ScottKIf it sort of works like that, I'm OK with FFe.16:49
ulmloggertouchegg is not meant for touch devices16:49
ulmloggerit is meant for touch pads16:49
ulmloggerthe things on a laptop16:49
ulmloggerwhere it works perfectly16:49
ScottKI see.16:50
ScottKSomeone point me at the FFe bug and I'll approve it.16:51
Quintasanbambee: ^16:54
bambeeQuintasan: pong16:56
* bambee is opening thunderbird16:56
bambeebug 82863716:57
ubottuLaunchpad bug 828637 in Ubuntu "Feature Freeze Exception request for Touchégg" [Undecided,New] https://launchpad.net/bugs/82863716:57
bambeeScottK, Quintasan: ^16:57
ScottKbambee: Approved.16:58
bambeewoo! thanks! :D16:58
Quintasanjussi: On my G+ post, I forgot to mention we are doing that after Feature Freeze :DDD17:09
* rgreening wishes he had a G+ invite.17:36
rgreening^ Quintasan :)17:36
jjessei can give you one17:36
Quintasanrgreening: email please :)17:36
rgreeningjjesse: cool. would be sweet17:36
jjessehttps://plus.google.com/_/notifications/ngemlink?path=%2F%3Fgpinv%3DYm4x4zDkd94%3AF3ZUHdJh1fQ17:36
rgreeningroderick DOT greening AT gmail.com17:36
QuintasanOh jjesse beat it to me17:37
jjesseclick on the link17:37
rgreeningYAY!17:37
rgreeningthanks.17:39
ScottKPlease be fixing http://people.ubuntuwire.org/~wgrant/rebuild-ftbfs-test/test-rebuild-20110816-oneiric.html#kubuntu18:25
ScottKI think I got pimlibs and company fixed.18:26
micahg\o/ packagesets get more eyes :)18:40
afiestasrbelem today I explained the plans for the places view to ereslibre18:58
afiestashe loved them :)18:58
rbelemafiestas, awesome :-D19:04
rbelemafiestas, i already stated to hack that19:04
rbelemafiestas, i changed it to threeview19:04
afiestasalready?19:04
afiestasthat was fast19:04
rbelemafiestas, :-D19:04
rbelemafiestas, it is a pretty simple 19:05
afiestasrbelem: if there is not a screenshot, it never existed :p19:07
rbelemafiestas, as soon as i arrive home i will send a screenshot to you :-)19:26
rbelemQuintasan, ping19:26
markeykonnichiwa me friends, we be doing the hangoutery, and if ye shall desire, ye can join the funy19:52
bcooksleyHi20:17
bcooksleyis somebody around who feels responsible for the kde-workspace patch "kubuntu_11_fix_root_only_kcms.diff"?20:18
bcooksleyit is terribly broken20:18
ulmloggerit is a patch imported from fedora actually20:18
ulmloggermight be that they have an unbroken version20:18
bcooksleyI see20:18
ulmloggerbcooksley: is there a builtin way to handle this?20:18
bcooksleykind of20:18
ulmloggerbecause I really think that would be a worthwhile feature20:18
ulmloggeralso in terms of 3rd party integration20:19
bcooksleythrough the external application feature20:19
ulmloggerthink company wants to expand systemsettings for internal stuff they need20:19
bcooksleysee kde-workspace/systemsettings/examples/external-application.txt20:19
ulmloggerbcooksley: cheers20:19
bcooksleythey'll show up just like any other module, and have a replacement fill in provided by system settings when clicked20:19
bcooksleyallowing you to relaunch it if you want20:20
bcooksleyif you wanted to reimplement root mode - rather than adjusting the control modules20:20
bcooksleyline 209 of ModuleView.cpp is where it should be implemented20:20
Quintasanrbelem: pong20:21
bcooksleythe patch causes http://bugs.kde.org/28035620:21
ubottuKDE bug 280356 in general "Impossible to reuse the System Settings window once Partition manager is started" [Normal,New]20:21
ScottKulmlogger: Are you going to fix it then?20:22
ScottKThat might go nicely with Quintasan GLES magic for one big -workspace upload.20:22
ScottKbcooksley: Thanks for pointing it out.20:22
bcooksleyScottK: np20:23
ulmloggerScottK: if only I figureed out how20:23
bcooksleyif you're talking about system settings, I wrote the code in question (moduleview.cpp) so feel free to ask about that area20:25
rbelemQuintasan, is there some something that i have to do? :-)20:26
ScottKulmlogger: Since the amount of C++ I know would fit into a very small thimble, I think it's up to you.20:27
ulmloggerwell, point being, I just tried to whip the external app stuff into an existing rootonly kcm and everything goes kaput20:28
ulmloggerah20:28
ulmloggergetting somewhere20:28
ulmloggeror not20:28
bcooksleyupdated sycoca?20:29
ulmloggerbcooksley: there is some stuff weird with it20:34
bcooksleysuch as..20:34
ulmloggerwell, with partitonmanager anyway20:35
yofelgreat, tomahawk has fooey licensing too ~.~20:35
bcooksleywhat does it do ulmlogger?20:36
bcooksleyit is supposed to launch the module when you click the icon20:36
ulmloggerwell, there is an actual partitionmanager kcm, or rather it is a kcm20:36
ulmloggerno idea why20:36
ulmloggerbcooksley: yeah, I got the desktop file fixed up20:36
ulmloggerworks now20:36
bcooksleyok, cool20:37
ulmloggerit lacks busy indication though, might be my fault I guess20:37
bcooksleythe system settings app itself forks that off20:37
bcooksleyso it is now independent20:37
ulmloggerright20:37
ulmloggergotta talk to fedora about this too20:38
* ulmlogger is a tad drunk right now, so he'll put this on his todo20:38
bcooksleyok, thanks for taking the lead on this20:38
bcooksleyif they still want compat with existing root modules20:38
bcooksleythen the way to do it is to copy ExternalAppModule, modify that 20:38
bcooksleythen tie it in like ExternalAppModule is in ModuleView20:39
ulmloggernoted, thanks for the info :)20:40
ulmloggeryofel: markey will be pleased to know20:41
ulmloggeryofel: send me a mail20:41
ulmloggerI'll meet with muesli tomorrow20:41
yofelI can talk to muesli myself on IRC. tomahawk itself is mostly ok, but the thirdparty stuff they ship has some missing copies20:42
ulmloggeryofel: 3rd party code copies are discouraged in the archive :P20:45
ulmloggerso that is a no go pretty much really20:45
ulmloggerespecially since one can build tomahawk without that shite20:45
ScottKIf it's just library copies, yes, please get rid of them.20:45
yofel$ ls thirdparty/20:46
yofelCMakeLists.txt  jreen  liblastfm2  libportfwd  qtweetlib  qxt  SPMediaKeyTap  winhttp20:46
yofelhaven't checked if anything of that is actually needed20:46
ulmloggernone20:46
yofelhm, getting rid of that dir would need some cmake fixes it seems20:48
=== yofel_ is now known as yofel
CIA-52[kubuntu-tablet-settings] Harald Sitter <apachelogger@ubuntu.com> * apachelogger@ubuntu.com-20110818211305-kxpym7cqh0dx9a5e * (24 files in 7 dirs) Initial release21:13
ulmloggerbulldog98_: https://code.launchpad.net/kubuntu-tablet-settings plz try, mind that you will have to hook it into startkde as of yet21:16
ulmloggeror deploy an xession file21:16
ulmloggerScottK: ^21:16
ulmloggerQuintasan_: ^21:16
yofelulmlogger: well, 2 of those are optional (without archive packages), qxt is in the archive but with different ABI, liblastfm2 would need some cmake find magic and libportfwd is somehow built-in21:20
yofelSPMediaKeyTap and  winhttp are mac/win21:21
ulmloggertoo drunk21:24
* yofel has a headache and postpones till tomorrow21:27
yofelhm, now I have a runaway kded process too22:22
ulmloggeryeah, Mamarok had it too22:24
ulmloggerthough I really do not feel like looking into that shitty code again22:24
yofel3 people in +1 have it too22:24
yofelbut I don't get which of the 6 threads gdb sees is broken...22:25
yofelhm, thread 4 doesn't start with __GI___poll22:25
yofel#0  0x00007f71988c1913 in select () at ../sysdeps/unix/syscall-template.S:8222:26
yofel#1  0x00007f7199cb36d1 in QProcessManager::run (this=0x7f719a0055a0) at io/qprocess_unix.cpp:24522:26
yofel#2  0x00007f7199beaca5 in QThreadPrivate::start (arg=0x7f719a0055a0) at thread/qthread_unix.cpp:32022:26
yofel#3  0x00007f7196389edc in start_thread (arg=0x7f717f49e700) at pthread_create.c:30422:26
yofel#4  0x00007f71988c889d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:11222:26
yofel#5  0x0000000000000000 in ?? ()22:26
ulmloggerdepper22:30
ulmloggerdeeper22:30
yofelhttp://paste.kde.org/11149922:30
yofelthat's all I get22:30
ulmloggerok22:33
ulmloggeryofel: valgrind it22:33
ulmloggerwith --tool=callgrind22:33
ulmloggerlet it run for a bit22:33
yofelhow does one valgrind a running process?22:33
ulmloggerwith a bit of luck it should show what is called too often22:33
ulmloggerie what causes the endless loop eating th ecpu22:33
ulmloggeryofel: google it22:33
yofel(if that's even possible)22:33
ulmloggersorry, I have the command somewhere22:33
ulmloggerbut I am too drunk to look for it22:34
ulmloggeror actually too tired22:34
yofelok, no22:34
ulmloggerif you do not feel like it I can look for it tomorrow22:34
ulmloggeractually I think I pasted the command in the cpu usage bug for natty22:34
ulmloggeror at least I was close to doing so IIRc22:34
yofelah wait, callgrind can do that somehow it seems22:35
yofelhm no22:36
yofelstill needs to be run in valgrind first22:36
yofelulmlogger: uh fun22:38
ulmloggeryofel: callgrind is a tool for valgrind22:38
yofeltrying to run kded by hand gives:22:38
yofelkded4: symbol lookup error: /usr/lib/ntrack/modules/ntrack-libnl3.so: undefined symbol: ntrack_list_insert_sorted22:38
ulmloggeroh yeah, you might need the dbgsym package from ddebs22:38
ulmloggeralso FWIW valgrind is essentially just a shell for varous memory analysis tools22:38
ulmloggervalgrind, callgrind, cachegrind, massif...22:38
yofelI know what valgrind is, used it a few times, it's just that you can't attach it to running processes since it emulates things22:39
yofelgreat, now I have NO kded process anymore22:39
yofelulmlogger: ROFLMAO22:40
yofellast ntrack upload 22:40
yofel  * debian/libntrack0.symbols: add missing ntrack_info_human_banner symbol.22:40
yofelwell, not really22:41
yofellet's see if this UDD thing actually has a use for once22:41
yofelulmlogger: uh... why would a symbol in a library end up as Undefined but still appear in nm?22:55
yofel$ nm -DC /usr/lib/ntrack/modules/ntrack-libnl3.so | grep ntrack_list_insert_sorted22:55
yofel                 U ntrack_list_insert_sorted22:55
* ScottK is almost sure ntrack is to blame and whatever it is will turn out to be a problem on Natty as well.22:59
ScottKGo yofel.  Go.22:59
bambeeRiddell: Riddell: as I am a kubuntu member now, could you subscribe me to kde-packagers please ? (however, there's no hurry)23:02
yofel+123:03
yofelhm, rebuilding ntrack changed nothing23:04
yofelsomeone fix debsums...23:05
yofelScottK: it doesn't really help that ntrack-libnl3.so has no symbols file in the ntrack package...23:07
ScottKyofel: You know about the dbgsym repo, right?23:08
yofelI said symbols file, not debugging symbols23:09
ScottKAh.23:09
yofelI have latter23:09
ScottKcyphermox: ^^^ You should fix that.23:09
ScottKToday that would be a requirement to get in Main.23:09
yofelwell, the other files have one, but this lib seems to be a module23:10
yofelbut then it shouldn't make kded fail...23:10
wgrantScottK: What's broken?23:15
ScottKyofel needs symbols for ntrack-libnl3.so to troubleshoot random ntrack causes other processes to consume 100% of CPU.23:16
Riddellbambee: I think you need to file a bug on bugs.kde.org for sysadmin to subscribe you23:23
bambeeRiddell: oh sure23:23
bambeenp23:23
cyphermoxScottK: reading backlog23:43
* yofel gives up for today23:43
wgrantScottK: That will teach me to read my scrollback while half asleep. Your link to the FTBFS page showed up as a ping, so I thought you were telling me there was something wrong with it.23:44
wgrantAhem.23:45
cyphermoxntrack-libnl3 has no symbols file, yeah, that's possible...23:47

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!