UbuntuIRC / 2012 /07 /20 /#kubuntu-devel.txt
niansa
Initial commit
4aa5fce
[02:14] <JontheEchidna> pykde takes a while to build
=== jalcine is now known as Jacky
[04:18] <ScottK> JontheEchidna: Did you get a chance to take it for a spin?
[09:01] <Riddell> I miss revu
=== Quintasan_ is now known as Quintasan
[09:25] <Quintasan> Hi
[09:28] <Riddell> hi Quintasan
[09:37] <Quintasan> Riddell: I think kimpanel's upstream is Chinese, that will make communication a little bit harder if he doesn't know English, gotta confirm it first
[09:42] <Riddell> what makes you think he doesn't speak english?
[09:47] <Riddell> kubuntu.org/news/calligra-25-rc
[09:50] <Quintasan> Riddell: Just guessing, so far my adventures in the Asian developer community ended up in a failure because I didn't know Chinese and the other party knew little English.
[09:51] <Quintasan> Riddell: We could also try switching to fcitx if we get enough tests.
[09:54] <Quintasan> ETOOMANYINPUTMETHODS
[09:56] <Riddell> Quintasan: does it work?
[09:57] <Quintasan> Riddell: Well it looks like it is working, let me try that on a clean install.
[09:57] * Quintasan does kvm magic
[09:57] <Riddell> Blizzz: does the owncloud-client use gnome-keyring? it's giving me lots of error messages saying so
[10:01] <Blizzz> Riddell: dragotin says it does not
[10:03] <Riddell> something does, I get lots of this when I run it "WARNING: gnome-keyring:: couldn't connect to: /home/jr/.cache/keyring-9Yp480/pkcs11: No such file or directory"
[11:29] <BluesKaj> Hey all
[11:33] <JontheEchidna> ScottK: no, but it is compiled now. I'll give it a try after work today
[12:43] <ScottK> JontheEchidna: Great. Thanks.
=== apachelogger___ is now known as apachelogger____
[13:00] <Riddell> Blizzz: owncloud-client in experimental ppa for the testing https://launchpad.net/~kubuntu-ppa/+archive/experimental/+packages
[13:00] <Riddell> apachelogger____: what is the meaning of the underscores?
[13:00] <apachelogger____> degree of annoyedness
[13:00] <Riddell> uh oh
[13:10] <afiestas> apachelogger____: remove one of them coz the video I sent u.U
[13:11] <apachelogger____> touch debian/stamp-patched
[13:11] <apachelogger____> that is reason to remove one
=== apachelogger____ is now known as apachelogger___
[13:29] <Riddell> happy birthday Nightrose
[13:29] <Nightrose> Riddell: thank you! :)
[13:52] <Blizzz> Riddell: yay!
=== smarter_ is now known as smarter
[14:16] <jjesse> yeah happy birthday Nightrose
[14:17] <Nightrose> jjesse: thx :)
[14:52] <Riddell> jtechidna: was there any conclusion to the gtk kcm question?
[14:53] <Riddell> seems like the main point is it overriding default gtk settings which would affect ubuntu unity users
[14:55] <jtechidna> hmm, I don't think we reached any consensus on what to do.
[14:55] <jtechidna> though I think if the kcm is part of kubuntu-desktop then it's not that big of an issue
[14:56] <Riddell> jtechidna: what's not an issue?
[14:56] <jtechidna> Riddell: globally setting the gtk settings
[14:56] <Riddell> jtechidna: but that would mess up people who also wanted to use unity no?
[14:57] <jtechidna> they generally wouldn't install kubuntu-desktop, right?
[14:57] <Riddell> some people like to use both
[14:58] <Riddell> jtechidna: how do we set it by default for new users who havn't run the gtk config kcm?
[14:58] <jtechidna> I don't know
[14:59] <jtechidna> how do we do it currently?
[14:59] <Riddell> not sure that's what I'm wondering :)
[14:59] <BluesKaj> i don't see the point of unity with kde available ..I tried unity for 2 weeks , that was enough, now I understand why ubuntu users are looking at alternatives
[15:00] <jtechidna> I know we used to have a shell script as part of startkde, dunno if that's still used
[15:00] <Riddell> ah yes it has "# configuration of the gtk2-engines-oxygen if not already set"
[15:00] <jtechidna> I thought xsettings-kde may have changed things, though
[15:01] <jtechidna> I'm not sure how everything fits together now
[15:01] <Riddell> startkde makes ~/.kde/env/gtk2-default-theme.rc.sh
[15:06] <Riddell> the good news is this new version of oxygen-gtk3 fixes it!
[15:06] * Riddell updates oxygen-gtk2/3
[15:10] <Riddell> well I get how GTK2 theme is set, no idea how GTK 3 theme is set
[15:15] <jtechidna> archive admins are a bit slow :( bug 1012151
[15:15] <ubottu> Launchpad bug 1012151 in tagua (Ubuntu) "Request for removal from archive (binary and source, sync blacklist)" [Medium,Confirmed] https://launchpad.net/bugs/1012151
[15:19] <Riddell> voila
[15:20] <jtechidna> thanks :)
[15:20] <jtechidna> I had forgotten about it, it had been so long, lol
[15:21] <jtechidna> !info knights
[15:21] <ubottu> knights (source: knights): chess interface for the KDE Platform. In component universe, is optional. Version 2.3.2-1build1 (precise), package size 2630 kB, installed size 4012 kB
[15:21] <jtechidna> ^knights is also affected, but it's upstream is still around. there's some hope that a new version can be released
[15:22] <ScottK> jtechidna: We could remove just the binaries for now and then wait and see.
[15:37] <tsdgeos> wassup with knights?
[15:39] <ScottK> Apparently not ported to the new kdegames lib API.
[15:39] <ScottK> JontheEchidna: ^^^
[15:43] <jtechidna> It still uses KGameRenderer and friends for rendering: http://techbase.kde.org/Projects/Games/Porting_to_libkdegames_v5#Theming_and_KGameRenderer
[15:44] <jtechidna> *still uses old KGameRenderer
[15:44] <jtechidna> and the old buildsystem bits
[15:44] <ScottK> tsdgeos: ^^^
[15:45] <tsdgeos> ScottK: the buildsystem bits might be a problem
[15:45] <tsdgeos> KGameRenderer is fine
[15:46] <tsdgeos> actaully i don't know why smajesky changed the LibKDEGames.cmake to KDEGames.cmake
[15:46] <tsdgeos> just breaks stuff
[15:46] <tsdgeos> without really any need
[15:46] <tsdgeos> but now it's too late i guess
[15:46] <ScottK> If it's just a name change, maybe a compatibility symlink for 4.9?
[15:48] <Riddell> jtechidna: oh it's actually xsettings-kde that sets the gtk2 and 3 theme
[15:48] <Riddell> as well as GTK2_RC_FILES
[15:49] <tsdgeos> ScottK: haven't digged much
[15:49] <tsdgeos> but i don't think there's much more changes
[15:49] * ScottK neither
[15:51] * tsdgeos will have a look later if he remembers
[16:07] <jtechidna> :D http://bash.org/?16293
[16:10] <apachelogger___> {standard input}: Fatal error: can't write SplitElementTxn.o: No space left on device
[16:10] <apachelogger___> yehaaa
[19:03] <micahg> umm, if I have a kde langpack installed in precise, it wants to install calligra on upgrade to quantal
[19:05] <micahg> language-pack-kde-xx Depends calligra-l10n-xx Recommends calligra, the langpacks should enhance calligra, not recommend it
[19:05] <micahg> is there an existing bug or should I file one?
[19:07] <micahg> is anyone planning a calligra upload soonish?
[19:15] <micahg> bug #1027215
[19:15] <ubottu> Launchpad bug 1027215 in calligra-l10n (Ubuntu Quantal) "calligra-l10n-xx should enhance, not recommend calligra" [High,Triaged] https://launchpad.net/bugs/1027215
[19:16] <micahg> I'm happy to do the fix/upload as long as it's approved
[19:43] <micahg> Riddell: yofel_^^
[20:20] <ScottK> There's some magic script that possbly needs fixing too. I think apachelogger___ or yofel_ know about that.
=== yofel_ is now known as yofel
[20:38] <yofel> no, calligra-l10n is independent of kde-l10n and has it's own setup. Fixing that ^ would be a 2 line change in rules.
[20:38] <yofel> hm
[20:38] <yofel> spoke too soon
[20:39] <yofel> if it has to be "Enhances" then you'll need to add that in the control file as well
[20:40] <ScottK> micahg: Go for it.
[20:44] <micahg> ScottK: ok, will do over the weekend when I have more time to test
[20:44] <ScottK> Thanks.
[21:11] <JontheEchidna> ScottK: which package is the plugin factory supposed to be in?
[21:11] <ScottK> JontheEchidna: python3-pyqt4.
[21:11] <ScottK> I didn't do anything to make it land there though.
[21:11] <JontheEchidna> you mean -kde4?
[21:11] <ScottK> yes
[21:11] <ScottK> pykde4
[21:11] <JontheEchidna> so it should be in the build dir somewhere
[21:12] <JontheEchidna> found it
[21:12] <ScottK> Yes.
[21:12] <ScottK> I know it gets built.
[21:13] <JontheEchidna> kcmshell4 crashes, installing debugging symbols now
[21:18] <bas_brain> hi guys, will ubuntu still be developed by the community or will be developed by the new company?
[21:19] <ScottK> bas_brain: Di you mean Kubuntu?
[21:19] <ScottK> Di/Did
[21:20] <bas_brain> yeah sorry
[21:20] <ScottK> Some people who work on Kubuntu are paid for it and some aren't. That hasn't changed. Who happens to be paying people doesn't much matter.
[21:20] <bas_brain> so nothing changed, only the company who also support it. I understood
[21:21] <bas_brain> I would like to help the community, any advice?
[21:22] <ScottK> Start doing it.
[21:22] <ScottK> If there's something you think can be improved, improve it.
[21:23] <ScottK> I have to go, but if you mention where you'd like to help, I'm sure someone can suggest something.
[21:35] <bas_brain> i would like to help kubuntu team
[21:38] <bas_brain> how can i join any group?
[21:38] <bas_brain> do i have to register to a admnistrator and then he will decide?
[21:47] <JontheEchidna> barry: Hi, I tested out the python3 plugin factory just now.
[21:47] <JontheEchidna> this crash occurs during plugin creation: http://paste.ubuntu.com/1102500/
[21:48] <barry> JontheEchidna: hmm
[21:49] <JontheEchidna> Fatal Python error: take_gil: NULL tstate
[21:51] <JontheEchidna> here's a bt full for the top part of the stacktrace: http://paste.ubuntu.com/1102518/
[21:51] <barry> JontheEchidna: yikes. okay, that one should be easy. PyEval_InitThreads must be called *after* Py_Initialize in >= 3.2. /me updates the branch
[21:53] <barry> JontheEchidna: update pushed
[22:00] <JontheEchidna> barry: ok, that fixed that. It now crashes the kcmshell or systemsettings when the plugin encounters a syntax error (and tries to call apport?) http://paste.ubuntu.com/1102529/
[22:01] <JontheEchidna> I was using the print statement instead of the print function
[22:01] <barry> that's interesting
[22:03] <JontheEchidna> there's a similar crash with proper syntax, too: http://paste.ubuntu.com/1102536/
[22:03] <barry> probably the PyGILState_Ensure() needs to move above the PyErr_Print(). /me looks
[22:07] <barry> JontheEchidna: this might fix the syntax error case, probably not the other case, but please pull the update and try again
[22:08] <JontheEchidna> ok
[22:20] <JontheEchidna> barry: I don't think it helps the syntax case
[22:20] <barry> JontheEchidna: it still crashes when you have a syntax error?
[22:20] <apachelogger___> yofel: calligra-l10n does the same thing no?
[22:20] <JontheEchidna> barry: yep
[22:21] <barry> JontheEchidna: can you pastebin the backtrace?
[22:21] <JontheEchidna> barry: http://paste.ubuntu.com/1102561/
[22:21] <JontheEchidna> oh, wait
[22:21] <JontheEchidna> the patch didn't apply properly
[22:22] <yofel> apachelogger___: not really, they have a small setup script, but that's only to prepare the initial mutiple-tar source. The package relationships are only defined in control and rules
[22:22] <JontheEchidna> ugh, I hope I don't have to do a clean rebuild. that'll take an hour or two D:
[22:23] <barry> JontheEchidna: ah, no, i see where i screwed up. hang on a sec
[22:23] <barry> i wonder why it doesn't apply properly.
[22:23] <JontheEchidna> actually it did apply properly
[22:23] <JontheEchidna> I'm using the Manual Patch System so that I don't have to re-build the whole thing :P
[22:24] <barry> i think the quilt stack is clean tho
[22:24] <JontheEchidna> e.g. me manually updating the patched file after looking at the bzr diff
[22:24] <barry> ah :)
[22:25] <JontheEchidna> then I just rm -rf builddir-3.2/kpython3pluginfactory, then debuild -us -uc -nc
[22:25] <JontheEchidna> saving me 2 hours compile time
[22:25] <barry> ouch.
[22:25] <barry> JontheEchidna: update pushed. please try again
[22:26] <JontheEchidna> k
[22:27] <barry> JontheEchidna: i'm going to get dinner started. brb.
[22:31] <apachelogger___> yofel: spooky
[22:33] <barry> JontheEchidna: let me know how it goes and i'll keep checking back in
[22:35] <JontheEchidna> barry: no luck yet, the relevant part of the function looks like this: http://paste.ubuntu.com/1102582/
[22:37] <barry> JontheEchidna: yeah, that looks right. the problem was that the gil state wasn't acquired when PyErr_Print() was called. but you're saying that with the latest pastebin, the syntax error case still crashes? if so, backtrace please.
[22:39] <JontheEchidna> barry: http://paste.ubuntu.com/1102591/
[22:40] <barry> JontheEchidna: okay, now i have no idea :/
[22:41] <JontheEchidna> :(
[22:41] <JontheEchidna> I can throw my project up in bzr if you'd like to fiddle around with it yourself
[22:41] <barry> JontheEchidna: sure. please tell me how to build and test it tho
[22:41] <JontheEchidna> ok
[22:43] <JontheEchidna> barry: lp:~echidnaman/+junk/kde-config-drivers
[22:43] <JontheEchidna> mkdir build; cd build; cmake ../ -DCMAKE_INSTALL_PREFIX=/usr; make; sudo make install; kbuildsycoca4; kcmshell4 kde-config-drivers
[22:44] <barry> JontheEchidna: on quantal right?
[22:45] <JontheEchidna> barry: yeah, only quantal has the python3-kde stuff
[22:45] <barry> cool
[23:00] <barry> JontheEchidna: i guess i'll see you in 2 hours ;)
[23:01] <JontheEchidna> well, my computer's a bit crap, but it's still a double-build of the package basically
[23:52] <barry> JontheEchidna: something's wrong with the build system. following your commands in a quantal chroot, after doing 'cmake ...', the make command just exits immediately. afaict, CMakeFiles/Makefile2 has an empty all: target
[23:52] <JontheEchidna> oh, right
[23:52] <JontheEchidna> skip the make step, just do make install
[23:52] <JontheEchidna> it's interperted so there's nothing to make :P
[23:54] <barry> ah, cool. well, looks like the kcmshell4 needs the x server so the chroot is unhappy ;/
[23:55] <barry> i'm going to have to install a build vm for this since i don't have a quantal machine that can handle it atm. i'll do that now but i probably won't get to the build until monday. please do ping me then.
[23:55] <JontheEchidna> ok
[23:55] <JontheEchidna> have a good weekend
[23:55] <barry> thanks, you too