2015/04/02/#kubuntu-devel.txt

[00:20] <mparillo> I do not see a link like that. Do I need to be added / join? In any case, it looks as if the card is moved all the way to done (it was a quick and ugly editing job, I admit).
[00:20] <ahoneybun> mparillo: I moved it
[00:22] <mparillo> Thank you ahoneybun. I suppose that means it is good enough for an old release that will only be supported for about four more months.
[00:23] <ahoneybun> np
[02:30] <cyphermox> are there currently known issues with the OEM install for kubuntu?
[02:31] <cyphermox> I'm trying to fix a problem with the oem user not always being properly removed, that would affect all flavors, but it looks to me like regardless of any fix, SDDM either will not start or will stay with a black screen
[02:32] <cyphermox>  /var/log/sddm.log shows PAM errors -- specifically Authentication failure
[02:35] <cyphermox> yeah, I can't find any useful state that would be left over either; it would be really helpful if someone could take a lot and tell me if it's because my setup is wrong somehow
[02:35] <cyphermox> *look
[05:54] <lordievader> Good morning.
[08:15] <Riddell> ovidiu-florin: http://starsky.19inch.net/~jr/tmp/desktop_thumb.png
[08:16] <Riddell> ovidiu-florin: http://starsky.19inch.net/~jr/tmp/desktop-wee.png
[08:21] <sitter> Riddell: patchy breaky? http://kci.pangea.pub/job/vivid_stable_oxygen-qt4/
[08:27] <Riddell> sitter: fix pushed
[08:28] <Riddell> sitter: what happened to automatic irc /msgs ? they seemed effective on me :)
[08:28] <sitter> Riddell: they only fire on merge fail
[08:28] <sitter> for builds it's somewhat impossible to tell whether upstream broke it or one of us
[08:28]  * sitter packages modemmanager-qt
[08:29] <Riddell> ah hah
[08:40] <micmord> Any project about yakuake and plasma5?
[08:50] <Riddell> micmord: pardon?
[08:55] <micmord> Riddell: I mean, current yakuake version is 2.9.9-1ubuntu1 for KDE4. Is there any project about porting it on KDE5?
[09:01] <Riddell> micmord: looks like there's a frameworks branch which had activity last month https://projects.kde.org/projects/extragear/utils/yakuake/repository/show?rev=frameworks
[09:01] <Riddell> micmord: beyond that you'd need to ask Eike who turns up here as Sho usually in european afternoons
[09:03] <micmord> Riddell: thank you
=== greyback__ is now known as greyback
[10:13] <sitter> Riddell: y u no fix muon repo :'<
[10:15] <Riddell> sitter: I'm on it!
[10:15] <Riddell> pushed
[10:16] <sitter> phew
[10:16] <sitter> <3
[10:17] <Riddell> so now muon does notify you of release upgrades but only in the passive notification not in the slide out box
[10:18] <sitter> weeh
[10:18]  * sitter thought he filed a but about the release notifications not being handled in the new applet
[10:19] <sitter> s/but/bug
[10:20] <Riddell> sitter: do you have an opinion on the doc move question?
[10:26] <Riddell> sitter: added https://bugs.kde.org/show_bug.cgi?id=345796 now
[10:26] <ubottu> KDE bug 345796 in notifier "Distro update notification not in Updates notification plasmoid" [Normal,Unconfirmed]
[10:27] <sitter> ah maybe the notification was broken as well at some point
[10:28] <sitter> Riddell: muon ci still failing btw
[10:28] <sitter> Riddell: https://launchpadlibrarian.net/201852616/buildlog_ubuntu-vivid-amd64.muon_4%3A5.2.2-1%2Bgit20150402.1018%2B15.04-0ubuntu0_BUILDING.txt.gz
[10:28] <sitter> Oo
[10:34] <Riddell> hmm
[10:35]  * Riddell builds
[10:42] <Riddell> ahoneybun: should I put out a call on ubuntu-translators list for docs translations? what should I say?
[10:51] <BluesKaj> Hi folks
[10:54] <sitter> Riddell: any bugs left in knh btw?
[10:56] <Riddell> sitter: 7 here but they're untriaged and none that are release blockers https://bugs.launchpad.net/ubuntu/+source/kubuntu-notification-helper
=== kbroulik is now known as kbroulik-lunch
[10:58] <sitter> untriaged, what a surprise
[10:58] <sitter> oh oh oh
[10:59] <sitter> Riddell: did you patch whatever thing reads desktopfiles nowadays to look for translations via ki18n?
[11:00] <sgclark> morning
[11:01] <sitter> ahoy sgclark
[11:01] <sitter> oh that reminds me I need to look at the phonon ci stuff
[11:02] <Riddell> sitter: um I think I've entirely forgotten, why do you ask?
[11:02] <Riddell> sgclark: welcome back, nice holiday?
[11:02] <sgclark> yes :) ty
[11:02] <sitter> Riddell: because I also forgot  and without it our KCMs as well as all main packages do not have translations for the desktop file
[11:03] <sitter> sgclark: where does dependencies/logical-module-structure live?
[11:04] <sitter> Riddell: do you perhaps know if pitti is on vacation or something?
[11:05] <Riddell> sitter: he doesn't seem to be around this week so I guess he is away
[11:05] <sitter> the one week I happen to have a random question ^^
[11:05] <sgclark> sitter: kde:kde-build-metadata
[11:06] <sitter> shadeslayer, Riddell: btw, I seem to recall having read about various debian teams (python and ruby) having had strives towards a unified autopkgtest provisioning for all their packages. it feels to me like we might want to go for something like that for all kde software
[11:07] <sitter> given that we always want to mimic the upstream CI env as a base and all frameworks for example have a huge copy pasta of the debian/tests dir with the odd difference here and there
[11:07] <sitter> sgclark: thx
[11:08] <sgclark> np
[11:08] <sitter> sgclark: btw, how does one handle the case where one branch is supposed to be built for both qt4 and qt5 but with different cmake arguments respectively?
[11:09] <sgclark> sitter: that is np, we use configs for that, just put the right branch and maybe let me know which ones need that
[11:10] <sgclark> and the cmake command would be helpful :)
[11:10] <sitter> if only I remembered that one ^^
[11:12] <sitter> sgclark: add phonon... the qt5 builds need -DPHONON_BUILD_PHONON4QT5=ON
[11:12] <sitter> *added
[11:12] <sgclark> sitter: simple enough, thanks
[11:17] <shadeslayer> sitter: I smell a proposal coming up
[11:17] <sitter> are you finally getting married?
[11:20] <shadeslayer> Lol
[11:20] <shadeslayer> You only want that because you want to buy a suit 
[11:22] <sitter> Riddell: piiiiiiiing
[11:23] <soee_> update topic, 19 bugs left =)
[11:23] <sitter> shadeslayer: is Riddell gossiping at the coffee robot or does he just not want to talk to me
[11:23] <shadeslayer> Probably the latter
[11:23] <sitter> you not in yet?
[11:24] <sitter> Riddell: I pushed whoopsie, driver, debug-installer. BUT notification-helper is now notificationS-helper (note the s), just wanted to make sure that is intentional before I push that
[11:25] <shadeslayer> sitter: left for luncheroo
[11:25] <sitter> Riddell: also libkubuntu apparently was left out :(
[11:25] <sitter> :@
[11:25] <sitter> always with those lunchens
[11:26] <sitter> shadeslayer: are you running autopkgtest yet?
[11:27]  * Riddell testing upgrades
[11:27] <sitter> Riddell: please be replying
[11:27] <shadeslayer> Simon_B: no
[11:27] <shadeslayer> Er
[11:27] <shadeslayer> sitter: no
[11:28] <shadeslayer> Too much todo
[11:28] <shadeslayer> /o\
[11:28] <sitter> ..........
[11:28] <Riddell> sitter: testing: dunno autopkgtest stuff still confuses me quite a lot
[11:28] <shadeslayer> sitter: I have to do some armhf stuff
[11:28] <Riddell> sitter: notificationS-helper is not intentional, shall I reopen the ticket to get it removed/re-added ?
[11:29] <sitter> wgrant: are there any plans to provide autopkgtest runs for ppas?
[11:29] <Riddell> sitter: I'm the only one around here who doesn't go for 3 hour lunches every day!
[11:29] <sitter> Riddell: talk to ben what the course of action is I guess
[11:30] <wgrant> sitter: autopkgtest just pulls data from Launchpad. It's not a service that we run.
[11:31] <sitter> wgrant: ah, so I suppose it could pull data from a ppa as well?
[11:31] <wgrant> sitter: I don't see why not. It's just pointing at the primary archive today.
[11:32] <sitter> ok, thanks
[11:32]  * sitter puts talk to pitti on todo for next week ^^
[11:33] <Riddell> mparillo: notes for VividUpgrade page when we come to change it for final, there's no need for an update before the upgrade (yay) and the same screenshot can be used for the muon-update app as in utopic
[11:36] <mparillo> Riddell: You mean on the final version of https://help.ubuntu.com/community/VividUpgrades/Kubuntu ? Simply removing the chunk on running muon-upgrade: https://help.ubuntu.com/community/VividUpgrades/Kubuntu#Ensure_14.10_is_updated ?
[11:38] <Riddell> sitter: updated ticket for k-n-h and added a ticket for libkubuntu
[11:38] <Riddell> mparillo: yeah that can all go, yay
[11:39] <mparillo> Now or when we get the final release candidates?
[11:40] <sitter> Riddell: cheers
=== kbroulik-lunch is now known as kbroulik
[11:53] <ahoneybun> Riddell: I'm rebuilding the docs with sitter's script for the new screenshots
[11:54] <soee_> hmm, if we hover some picture in Dolphin we have preview in side panel, if we remove this file (and empty trash), preview is still visible and we can for example add tags -where those tags go if file is gone for good ?
[12:04] <Riddell> mparillo: later, a day before the final release I guess
[12:05] <mparillo> Short window. I will post it on my calendar.
[12:11] <Riddell> thanks :)
[12:56] <sitter> Riddell: ok, the following repos have been pushed and had their hooks enabled -> kubuntu-debug-installer, kubuntu-driver-kcm, whoopsie-kcm
[12:57] <sitter> so I think you can update their tickets to continue with reviewboard setup etc.
[13:07] <Riddell> sitter: hooks?
[13:07] <Riddell> sitter: shall I export translations from launchpad then tell tosky?
[13:19] <Riddell> d_ed: https://wiki.ubuntu.com/LightDM
[13:34] <mparillo> That link has sudo stop lightdm. Is that different from sudo service lightdm stop?
[13:38] <yofel> stop is an upstart thing, it does the same thing though
[13:41] <soee> can we change loading order after ssytem reboot/boot so the plasmahsell will load and than previous session apps ?
[13:42] <mparillo> TY, and I am being trained anyway on the new systemd way.
[14:11] <Riddell> sitter: does sddm in ci use the master branch?
[14:11] <Riddell> it doesn't seem to have a kubuntu_vivid_archive branch, should it?
[14:24] <sitter> Riddell: yes sddm uses master, yes an archive branch would be handy
[14:24] <sitter> Riddell: translations export only if he wants to import them :P
[14:26] <Riddell> sitter: what KCMs don't have translations from .desktop files?
[14:26] <soee> Riddell: with this https://bugs.launchpad.net/ubuntu/+source/nvidia-prime/+bug/1428328/comments/5
[14:27] <ubottu> Launchpad bug 1428328 in nvidia-prime (Ubuntu) "nvidia-prime needs sddm support" [Medium,Confirmed]
[14:27]  * Riddell reports bug 1439720
[14:27] <ubottu> bug 1439720 in kconfig (Ubuntu) "Port kubuntu_langpack_desktop_files.diff for .desktop gettext support" [Undecided,New] https://launchpad.net/bugs/1439720
[14:27] <soee> we weill be able to make nvidia gpu works with sddm ?
[14:28] <sitter> Riddell: all of ours
[14:29] <sitter> or actually
[14:29] <sitter> only if the desktop file snafu wasn't patched
[14:29] <sitter> ah yes, that report there ^^
[14:30] <Riddell> sitter: translations don't get stripped, /usr/share/kservices5/kcm_phonon.desktop still has all its translations
[14:31] <sitter> ours
[14:31] <Riddell> soee: yes hopefully, do you have the hardware to test?
[14:31] <sitter> not upstreams
[14:31] <sitter> whoopsie, driver, notification
[14:31] <Riddell> oh I see
[14:31] <soee> Riddell: yes, on my laptop im using atm. i have intel + nvidia
[14:36] <Riddell> sitter, shadeslayer, sgclark, ScottK: all the S tabs, what to do with docs path?
[14:36] <sitter> ah yes, the mighty stabs
[14:36] <sitter> IIRC I pointed out this veritable flaw way back in december :P
[14:37] <sitter> Riddell: there really only is a rebuild option
[14:37] <sitter> and TBH I think all of kf5 needs to get rebuilt rather than all of kde4
[14:37] <sitter> former we know what is effected, latter is everything and the kitchen sink
[14:38] <shadeslayer> sitter: I'd make something like a script that is installed as a apt trigger
[14:38] <sitter> I will also say that IIRC khelpcenter had the same problem
[14:38] <sitter> in kde3->4
[14:38] <shadeslayer> so that whenever something is installed to the docs dir, we just symlink the new stuff
[14:38] <sitter> that would need lodsa testing
[14:39] <Riddell> I agree with sitter 
[14:39] <Riddell> sitter: but rebuild kf5? what would that do?
[14:39] <sgclark> hmm
[14:39] <sitter> Riddell: did you talk to the debian people about it btw
[14:39] <sitter> there was a reason why we diverged from upstream
[14:40] <shadeslayer> actually I purposefully did not install to /usr/share/HTML/kde since I did not see the point
[14:40] <sitter> did you ask what the point was?
[14:40] <shadeslayer> I did
[14:40] <shadeslayer> and I did alot of searching
[14:40] <shadeslayer> on the ml
[14:40] <sgclark> I tried to talk to talk to debian some time ago and got ignored, seems you guys are making more progress than I did though
[14:40] <sitter> what was the point
[14:40] <shadeslayer> didn't find it
[14:40] <shadeslayer> even poked MoDaX for a answer
[14:40] <shadeslayer> got none
[14:40] <Riddell> sitter: I asking in debian irc channel, no useful reply
[14:40] <sitter> ...
[14:41] <sitter> well
[14:41] <sitter> one could patch khc :P
[14:41] <shadeslayer> so I saw it as pointless and used the upstream path
[14:41] <shadeslayer> sitter: sure
[14:41] <sitter> actually I think one needs to patch the help slave
[14:41] <shadeslayer> that's the other solution
[14:41] <Riddell> sgclark looked into khc a bit
[14:41] <sitter> either that or rebuild all the kf5 builds
[14:41] <sitter> or rather... all the ECM builds
[14:42] <sgclark> khc should really support more than one install path.. but upstream told me to go away and deal with it distro level
[14:43] <sgclark> again I am shocked that no other distros are facing this issue. seems only debian hard codes docs like that
[14:43] <Riddell> so I still feel mass rebuild of kde4 is best solution
[14:43] <sgclark> oof
[14:44] <shadeslayer> I disagree
[14:44] <shadeslayer> I'd go with patching khc or symlinking
[14:44] <sitter> no
[14:44] <sitter> the best solution is to patch kio_help
[14:44] <shadeslayer> quick, someone assign that task to sitter
[14:44] <sitter> the second best solution is returing to the pointless subdir and rebuild kf5 until we can come up with something better
[14:45] <shadeslayer> sitter: no
[14:45] <shadeslayer> that just promotes the perpetual cycle of "Lets fix it properly later"
[14:45] <sitter> the third best is symlinking since it's hard to asses its reliability and will as all bloody hacks like that will hit the fan eventually
[14:45] <shadeslayer> possibly that's leftover from KDE3 /o\
[14:45] <shadeslayer> and no one knows why /o\
[14:47] <sitter> oh
[14:47] <Riddell> patching kio_help.cpp seems do-able
[14:47] <sitter> another option: reintroduce khelpcenter(4) as actual binary khelpcenter4 and hardcode kdelibs to launch khelpcenter4 rather than khelpcenter
[14:47] <sitter> which is slightly dodgy
[14:47] <sitter> Riddell: yeah, just needs that lookup function adjusted a bit
[14:47] <Riddell> HelpProtocol::get()
[14:48] <sitter> Riddell: is the actual lookup apparently HelpProtocol::langLookup
[14:48] <sitter> get just strips it
[14:49] <sitter> actually, you know what, this is just wrong upstream
[14:49] <sitter> you can set a different doc install path via ecm but kio will not give a fuck
[14:49] <sitter> ...
[14:49]  * sitter hates hardcoded paths
[14:52] <sitter> Riddell: http://paste.ubuntu.com/10724470/
[14:52] <sitter> needs header adjustment
[14:52] <sitter> otherwise that should fly
[14:52] <Riddell> sitter: yummy :)
[14:54] <sitter> I don't quite understand the code in ::get though, not sure it even needs adjustment
[14:54] <sitter> ah ah
[14:55] <sitter> http://paste.ubuntu.com/10724482/
[14:55] <sitter> that probably should fly
[14:55] <shadeslayer> make it fly!
[14:55] <sitter> I'd add some qdebug to ::get and see what it actually does in the cache thing though
[14:55] <shadeslayer> Riddell: is it done yet :P
[14:55] <sitter> that code is mental pretty much
[14:56] <Riddell> sitter: not yet
[14:57] <shadeslayer> Riddell: heh
[14:57] <shadeslayer> I'm blocked on IO
[14:57] <shadeslayer> how fun
[15:02] <Riddell> sitter: I'm confused why you need all that extra code, what else does it need but a second call to localDoc.append( QStandardPaths::locateAll("doc/kde/HTML")  ) ?
[15:03] <sitter> dunno I haven't actually read the function :P
[15:04] <sitter> append away
[15:04] <sitter> also fine with less code
[15:17] <Riddell> sitter: yeah this seems to do it https://paste.kde.org/p2p40a6ow
[15:18] <sitter> lovely
[15:29] <Riddell> awooga, good work on that one team, thanks shadeslayer, sgclark :)
[15:30] <sgclark> great work, thanks
=== ximion is now known as ximion-afk
[19:04] <ahoneybun> Riddell: I updated the screenshots for the english installation on the wiki, ran harald's script, and updated my -docs branch
[19:20] <soee_> killall plasmashell
[19:20] <soee_> killall plasmashell
[19:26] <soee_> :/
[19:26] <soee_> hmm 
[19:27] <soee_> Desktop just freezez for me now
[20:01] <ovidiu-florin> Riddell: thank you for the pictures
[20:01] <ovidiu-florin> I guess those were the only one you can find?
=== claydoh_ is now known as claydoh
[20:03] <ovidiu-florin> Riddell: so I guess this: http://www.kubuntu.org/news/9.04-release and this http://www.kubuntu.org/news/9.10-release remain broken
[20:11] <soee> wrr
[20:12] <soee> tried to install nvidai-prime with propriety drivers with current sddm
[20:12] <soee> better dont do this ;)
[20:26] <KDDA> hi folks
[20:27] <soee> hiho
=== soee_ is now known as soee
[21:16] <soee> Riddell: notification are a bit creazy lately
[21:17] <soee> if we run apt update, than than during updateing packages list, notification is shows up avery few seconds about avaialble updates
=== kubotu_ is now known as kubotu
[22:00] <ahoneybun> any word about plasma active?
[22:05] <valorie> I've not heard anything, but I doubt attention will turn to Active until we have full plasma functionality on the desktop restored
[22:05] <ahoneybun> of course
[22:06] <ahoneybun> now that phones are getting bigger active could really be effective
[22:07] <valorie> sure, and the refactoring etc that is going on now is designed to make Active etc. that much easier to do
[22:07] <valorie> building in touch from the get-go, etc.
[22:08] <valorie> really, as more and more laptops have touch screens, it will be a necessity anyway
[22:12] <ahoneybun> yes and will be super cool
=== Neo31` is now known as eo31
=== eo31 is now known as Neo31`
=== Neo31` is now known as Neo31
=== FatBack changed the topic of #kubuntu-devel to: penis
=== cyphermox changed the topic of #kubuntu-devel to: Kubuntu!
[22:52] <cyphermox> FatBack: please behave...
=== soee changed the topic of #kubuntu-devel to: Kubuntu - 10 years of friendlyness | https://trello.com/kubuntu | http://qa.kubuntu.co.uk/ | Crash tracker : http://goo.gl/Cdynad | 20 bugs to fix http://goo.gl/Chh7uh
[22:54] <cyphermox> soee: thanks
=== ahoneybun_ is now known as ahoneybun
[22:55] <cyphermox> soee: I wanted to put the topic back but I was digging in history to try and find what it was.
[22:55] <ahoneybun> he did that in here too?
[22:55] <soee> cyphermox: konversation has pencil icon before the topic text that shows history :)
[22:56] <cyphermox> any trick for quassel? :)
[22:56] <soee> ahoneybun: i think he did somethig similiar yesterday
[22:56] <ahoneybun> soee: http://hire.jobvite.com/CompanyJobs/Careers.aspx?c=qRJ9Vfwo&page=Job%20Description&j=oeJy0fwf :)
[22:56] <soee> cyphermox: not sure im not using quassel
[22:56] <ahoneybun> soee: he did the same thing in ubuntu-touch
[22:56] <cyphermox> yeah, things should be fixed now, not to worry :)