| === Tonio_aw is now known as Tonio_ | |
| === Tonio_ is now known as Tonio_aw | |
| [00:47] <shadeslayer> -> sleep | |
| === daniel is now known as Guest83062 | |
| === Guest83062 is now known as dantti | |
| [02:14] <valorie> shadeslayer: http://wstaw.org/m/2012/12/13/UDS-Badge-Kubuntu_shadeslayer_back_1.png worked perfectly for me | |
| [02:15] <valorie> http://wstaw.org/m/2012/12/14/rect3810.png did not scan with my phone | |
| [02:16] <valorie> http://wstaw.org/m/2012/12/14/rect3810.png works too | |
| [06:02] <Tm_T> shadeslayer: that seems to work | |
| === xeros_ is now known as xeros | |
| [07:30] <jussi> good morning all | |
| [07:49] <soee> good morning | |
| === yofel_ is now known as yofel | |
| === Tonio_aw is now known as Tonio_ | |
| [09:01] <apachelogger> Sput: shouldn't kwin fall in the category of 'modern post-ICCCM'? | |
| [09:02] <Sput> apachelogger: I guess, and it seems to work everywhere except maybe sometimes on *buntu | |
| [09:03] <apachelogger> brave assumption considering that everywhere else does not use it by default :P | |
| [09:04] <Sput> hm? | |
| [09:04] <Sput> what is this "default" thingy you keep talking about? | |
| [09:06] <apachelogger> having the quassel as the irc client you get when you installed a given operating system | |
| [09:06] <Sput> ah, I thought you were talking about kwin | |
| [09:06] <apachelogger> no one uses kwin :P | |
| [09:07] <apachelogger> Sput: I do wonder why you have that fullscreen option at all though | |
| [09:07] <Sput> I certainly use kwin and can fullscreen with now problem | |
| [09:07] <Sput> and we have the fullscreen option because users requested it and sent patches | |
| [09:07] <apachelogger> well | |
| [09:07] <apachelogger> http://onscreencars.com/wp-content/uploads/2010/01/TheHomer.jpg :P | |
| [09:08] <Sput> people love it especially on netbooks and tablets | |
| [09:09] <apachelogger> that's beause quassel has so many frames consuming space | |
| [09:09] <apachelogger> Sput: did you also get the same backtrace? | |
| [09:09] <Sput> I can't reproduce | |
| [09:09] <Sput> neither could anybody else not using kubuntu so far | |
| [09:10] <Sput> and even for kubuntu users it tends to happen only sometimes, if at all | |
| [09:10] <apachelogger> Sput: the dude in the bug apparently can reproduce it somewhat reliably | |
| [09:10] <apachelogger> in case you have questions | |
| [09:11] <Sput> well, as all *we* do is calling showFullscreen(), I'd blame Qt or the window manager | |
| [09:13] <Sput> but let me have an actual look at the stacktrace | |
| [09:15] <Sput> I would say KToggleFullScreenAction goes into a loop somehow | |
| [09:16] <apachelogger> yeah, just saw that as well | |
| [09:16] <apachelogger> explains the stack exhaustion | |
| [09:17] <yofel> hm, I think I just reproduced that. Ctrl+Shift+F -> Whole UI except mouse froze -> after half a minute quassel crashed | |
| [09:17] <yofel> and now apport-retrace is being mean to me -.- | |
| [09:20] <Sput> and I can't reproduce that issue on Gentoo or Arch :/ | |
| [09:20] <yofel> well, I'll agree that this is probably something with Qt... | |
| [09:23] <Sput> or KDE | |
| [09:24] <Sput> unless we're misusing the KToggleFullscreenAction, but why would it work elsewhere then? | |
| [09:24] <yofel> now *this* is fun: | |
| [09:24] <yofel> *** glibc detected *** gdb: double free or corruption (!prev): 0x00000000023c6250 *** | |
| [09:24] <yofel> first time I've seen gdb crash o.O | |
| [09:25] <tsdgeos> you're lucky then :D | |
| [09:25] <tsdgeos> it crashes almost reliably here | |
| [09:25] <yofel> yay...... | |
| [09:25] <tsdgeos> just have to set lots of breakpoints, recompile app, restart it | |
| [09:25] <tsdgeos> boom! | |
| [09:26] <tsdgeos> 70% of the time gdb complains about something and i lose the breakpoints when it crashes :-/ | |
| [09:26] <apachelogger> Only checkable actions can be checked. By default, this is false (the action is unchecked). | |
| [09:26] <tsdgeos> and obiously 70% is an invented number :D but happens soon enough | |
| [09:26] <apachelogger> Sput: you are misusing it :P | |
| [09:27] <apachelogger> actually | |
| [09:27] <apachelogger> if ( d->window->isFullScreen() != isChecked() ) | |
| [09:27] <apachelogger> activate( QAction::Trigger ); | |
| [09:27] <apachelogger> ^ event filter | |
| [09:28] <apachelogger> so the way to exhaust the stack would be to have the action's checked state diverge from the actual fullscreenness of the windows | |
| [09:28] <apachelogger> -s | |
| [09:31] <apachelogger> so it keeps loopying in that event filter until at some point for some reason it manages to break out of it and actually apply the toggle | |
| [09:31] <yofel> yeah, crashes somewhere in the event filter http://paste.kde.org/625118/ | |
| [09:31] <apachelogger> at which point it is lucky enough to have exhaust the stack | |
| [09:32] <apachelogger> which in turn makes it very much stack size dependent | |
| [09:32] <Sput> so the action is broken | |
| [09:32] <apachelogger> maybe, maybe not | |
| [09:34] <apachelogger> Sput: Important: If you need to set/change the fullscreen state manually, use KToggleFullScreenAction::setFullScreen() or a similar function, do not call directly the slot connected to the toggled() signal. The slot still needs to explicitly set the window state though. | |
| [09:35] <apachelogger> you are using it wrong I say :P | |
| [09:35] <apachelogger> also | |
| [09:35] <apachelogger> Note: Do NOT use QWidget::showFullScreen() or QWidget::showNormal(). They have several side-effects besides just switching the fullscreen state (for example, showNormal() resets all window states, not just fullscreen). Use the KToggleFullScreenAction::setFullScreen() helper function. | |
| [09:37] <apachelogger> yofel, tsdgeos: can one of you rebuild and test? | |
| [09:37] <yofel> did you fix it in git? | |
| [09:37] <apachelogger> src/qtui/mainwin.cpp | |
| [09:37] <apachelogger> void MainWin::toggleFullscreen() | |
| [09:38] <tsdgeos> apachelogger: i don't have quassel sources, besides i'm busy writing qml somewhere else | |
| [09:38] <apachelogger> instead of showNormal and showFullScreen call KToggleFullScreenAction::setFullScreen(this, false); and KToggleFullScreenAction::setFullScreen(this, true); | |
| [09:38] <apachelogger> yofel: ^ | |
| [09:38] <apachelogger> (yes that function is static) | |
| [09:39] <yofel> will take a bit to build it, but I can try | |
| [09:39] <apachelogger> I was under the impression quassel was a rather fast build | |
| [09:39] <apachelogger> or am I mixing up things in my head again | |
| [09:40] <yofel> well, I can try to build only the client, that shouldn't take too long | |
| [09:40] <yofel> this isn't my fastest PC here | |
| [09:40] <apachelogger> ah | |
| [09:41] * apachelogger is not at home either :P | |
| [09:41] <apachelogger> also I wasn't able to reproduce it anyway | |
| [09:42] <yofel> well, happened 2/2 times here, so I'll try it | |
| [09:42] <apachelogger> probaly the slot should follow whatever the checkedness is rather than what the fullscreeness is | |
| [09:43] * yofel notes that this probably needs wrapping into the kde detection | |
| [09:44] <yofel> I doubt plain Qt has KToggleFullScreenAction | |
| [09:51] <apachelogger> yofel: it's just so we know whether it fixes the problem | |
| [09:52] <apachelogger> proper fix needs a different slot anyway as to obey the action's signal toggled(bool) rather than discard the bool | |
| [09:53] <yofel> hm, didn't crash. But let me build git without that fix to be sure it actually makes a difference | |
| [09:55] <yofel> apachelogger: yep, fix works | |
| [10:00] <yofel> hm... at least I hope it does. Now X is in a loop without quassel *-.- | |
| [10:04] <apachelogger> well, once the loop happened X may be broken :P | |
| [10:04] <yofel> yeah, your fix works, without it hell breaks loose here | |
| [10:04] <apachelogger> xlib is easily confused | |
| === Tonio_ is now known as Tonio_aw | |
| === Tonio_aw is now known as Tonio_ | |
| === Tonio_ is now known as Tonio_aw | |
| [12:24] <shadeslayer> ohai | |
| [12:25] <Riddell> ohai? a new indian word to me | |
| [12:26] <Riddell> shadeslayer: did you do your seed merging? | |
| [12:27] <shadeslayer> not yet | |
| [12:28] <shadeslayer> a bit perplexed on how to add active-ship to supported | |
| [12:28] <Riddell> shadeslayer: just add it in STRUCTURE no? | |
| [12:29] <shadeslayer> hm | |
| [12:29] <shadeslayer> ah okay | |
| [12:29] <shadeslayer> lemme update | |
| [12:34] <shadeslayer> god damnit TCS, stop emailing me | |
| [12:34] <Riddell> who? | |
| [12:35] <shadeslayer> Riddell: Indian IT Company where I got hired but never bothered to join | |
| [12:35] <shadeslayer> they're asking everyone who got hired to come join the company on 18th of this month | |
| [12:36] <shadeslayer> and I keep getting emails about attire guidelines and how to sit in the office : | |
| [12:36] <shadeslayer> :) | |
| [12:36] <shadeslayer> They basically hire anyone who can write hello world in C/C++/Java | |
| [12:47] <yofel> ... and let them work on... what? | |
| [12:55] <shadeslayer> yofel: it's a services company, so it's mostly maintainence work, and once people join, they train them for 3-4 months | |
| [12:55] <shadeslayer> so that they can work on the code | |
| [12:56] <shadeslayer> kids here are falling head over heals to work for companies like TCS/Infosys/HCL | |
| [12:56] <shadeslayer> not sure why .... | |
| [12:59] <shadeslayer> Riddell: http://marcuzzokde.blogspot.in/2012/12/kids-size-kde.html < Kubuntu being used there :D | |
| [13:17] <Riddell> awooga | |
| [13:21] <BluesKaj> Hi all | |
| === Quintasan_ is now known as Quintasan | |
| [13:55] <Quintasan> \o | |
| [14:02] <agateau> shadeslayer: "how to sit in the office"? I want to read that! | |
| [14:04] <Quintasan> totally | |
| [14:04] <Quintasan> shadeslayer: Can I have you create a poll for business card designs or you want me to do that? | |
| === Tonio_aw is now known as Tonio_ | |
| [14:08] <Quintasan> hmm | |
| [14:08] <Quintasan> shadeslayer: We only have three designs? | |
| [14:11] <Quintasan> 4 including the one at your p.u.c | |
| [14:17] <jussi> hrm | |
| [14:18] <jussi> Is the updates ppa going to get an update soon? its only 4.9.4... | |
| [14:20] <yofel> what are you expecting to see? | |
| [14:21] <jussi> couple of points higher? | |
| [14:21] <yofel> 4.9.5 isn't out yet, and 4.9.90 is in beta | |
| [14:21] <jussi> wait... how does that work? | |
| [14:22] <jussi> Im now very confused | |
| [14:22] <yofel> updates PPA is for point releases. i.e. 4.9.3 -> 4.9.4. Backports is for 4.X -> 4.Y updates after release, beta ppa before it's released | |
| [14:22] <yofel> so 4.9.4 is in updates, 4.9.90 in beta, 4.10 will be in backports | |
| [14:23] <shadeslayer> Quintasan: please do | |
| [14:24] <shadeslayer> agateau: sec :P | |
| [14:24] <yofel> great, google doesn't find our ppa docs o.O | |
| [14:24] <shadeslayer> Quintasan: if you do create the poll, use http://people.ubuntu.com/~rohangarg/cards/UDS-Badge-Kubuntu_shadeslayer.svg and http://people.ubuntu.com/~rohangarg/cards/UDS-Badge-Kubuntu_shadeslayer_back_QR.svg from me | |
| [14:25] <yofel> now... | |
| [14:25] <jussi> yofel: Im more confused about the order of kde releases... | |
| [14:25] <yofel> jussi: https://wiki.kubuntu.org/Kubuntu/KubuntuPPAs | |
| [14:25] <yofel> jussi: what do you mean? What the version numbers stand for? | |
| [14:25] <yofel> or that 4.9.5 comes out after 4.9.90? | |
| [14:25] <jussi> yes, the latter | |
| [14:26] <yofel> 4.9.5 is bugfix for 4.9, 4.9.90 is beta2 for 4.10 | |
| [14:26] <yofel> 4.9 will get bugfixes until 4.10 is out | |
| [14:26] <jussi> uhuh | |
| [14:26] <jussi> ok | |
| [14:26] <jussi> see, I kind f thought it was linear... | |
| [14:26] <jussi> :/ | |
| [14:26] <yofel> well, it is, but the betas and RCs don't count | |
| [14:26] <jussi> thank you for the explanation though | |
| [14:27] <yofel> s/but/as/ | |
| [14:27] <kubotu> yofel meant: "well, it is, as the betas and RCs don't count" | |
| [14:27] <BluesKaj> odd they would cal 4.9.9 , 4.10 ..one expects a %.0 not a 4.ten | |
| [14:28] <BluesKaj> 5.0 that is | |
| [14:28] <ScottK> No. | |
| [14:28] <ScottK> The first digit is a major version number that indicates an ABI break in kdelibs | |
| [14:29] <shadeslayer> agateau: http://dl.dropbox.com/u/17090527/HSE-%20Office%20Ergonomics%20Flyer.pdf | |
| [14:29] <BluesKaj> I know it's no , but to me after 4.9 comes 5 | |
| [14:30] <tsimpson> it's all to do with ABI | |
| [14:30] <yofel> it'll be more fun when we have KDE Workspaces and Applications 4.11 using KDE Platform 5.0 - assuming that they'll have it done in half a year | |
| [14:30] <agateau> shadeslayer: it's not that bad actually | |
| [14:30] <Riddell> version numbers != decimal numbers | |
| [14:31] <agateau> BluesKaj: what comes after 4.9.5? | |
| [14:31] <BluesKaj> it should be 4.10.0 | |
| [14:32] <yofel> it will be, but how do you version the 4.10 beta so it's lower than 4.10 -> you use 4.9.90 | |
| [14:32] <BluesKaj> agateau, if you don't know , then whay are you asking me ? | |
| [14:32] <agateau> BluesKaj: let me rephrase: "what comes after 4.9.5 to you?" | |
| [14:33] <BluesKaj> I'm used to laboratory decimal measure convention, it looks odd to my eye | |
| [14:34] <BluesKaj> 4.9.6 | |
| [14:34] <agateau> I agree 4.10.0 is more correct. And this is actually what is used in formal texts. | |
| [14:34] <agateau> just not that much in irc conversations | |
| [14:35] <BluesKaj> 4.9.7, 4.9.8, 4.9.9, 5.0 | |
| [14:35] <BluesKaj> or 5.00 | |
| [14:35] <yofel> BluesKaj: application version numbers usually follow this idea: Major.Minor.Patch.Build - where every position isn't limited to a single digit | |
| [14:35] <BluesKaj> I'm used to metric measure .. | |
| [14:36] <yofel> chromium-browser follows that nicely: 23.0.1271.91 | |
| [14:36] <BluesKaj> makes more sense to my lab oriented mind | |
| [14:37] <tsimpson> the dot is just a separator, not a decimal point | |
| [14:37] <tsimpson> could just as easily been a comma | |
| [15:48] <rdieter> hi, I just worked with dantti this morning to fix an arguably nasty bug in print-manager that fills cups access_log and users' .xsession-errors with excessive logging about printer subscription renewals, see https://bugs.kde.org/show_bug.cgi?id=311528 | |
| [15:48] <ubottu> KDE bug 311528 in general "Renew-Subscription spam in /var/log/cups/access_log" [Normal,Resolved: fixed] | |
| [15:48] <rdieter> and patch against 0.2 : http://pkgs.fedoraproject.org/cgit/kde-print-manager.git/tree/print-manager-0.2.0-cups_renew_spam.patch?h=f18 | |
| [15:49] <rdieter> since you're shipping it by default (so i'm told) like we are, enjoy. | |
| [15:49] <Riddell> lovely, thanks rdieter | |
| [15:51] <rdieter> much nicer to renew every 3500 seconds, rather than milliseconds | |
| [15:58] <shadeslayer> heh | |
| [16:44] <Riddell> ScottK: problems in kblocks nad kbounce? | |
| [16:44] <ScottK> Riddell: Yes. Missing copyrights. I fixed and uploaded kblocks. Doing kbounce right now. | |
| [16:45] <Riddell> thanks | |
| [16:45] <ScottK> For a one or two line fix in debian/copyright, it's easier for me just to fix and reupload. | |
| [17:33] <ScottK> Riddell: Same with kfourinline | |
| [17:44] <ScottK> And kgoldrunner | |
| [18:42] <xnox> why are we shipping 71 empty language-pack-kde-* packages? | |
| [18:46] <yofel> xnox: IIRC they act as meta packages | |
| [18:46] <yofel> currently at least | |
| [18:47] <xnox> yofel: and kde-l10n-CC cannot act as metapackages? | |
| [18:48] <shadeslayer> ooooohhhhhhhhh | |
| [18:48] <shadeslayer> yofel: I found something | |
| [18:48] <xnox> yofel: also language-pack-kde-* pull in gnome translations. | |
| [18:48] <shadeslayer> yofel: pull-lp-source freetype | |
| [18:48] <shadeslayer> and read the rules | |
| [18:49] <yofel> probably, but I think our language-selector stand-in currently needs the language-pack stuff | |
| [18:49] <yofel> xnox: does language-pack-CC contain gnome translations? | |
| [18:49] <shadeslayer> they do 2 builds :D | |
| [18:49] <xnox> yofel: it has a dependency to gnome language packs. | |
| [18:50] <shadeslayer> btw has anyone tried out infinality? | |
| [18:50] <yofel> xnox: but if we're already on that matter, if someone installs nautilus - how would that get translated without the language pack installed? | |
| [18:52] <xnox> yofel: my naive expection for incomplete-language support daemon to notice things like that. | |
| [18:52] <yofel> IIRC that's currenly broken as we don't use language-selector, but I haven't looked at l10n much lately | |
| [19:16] <ScottK> Riddell: No messages.sh in kgoldrunner or kiriki. I'd accepted kgoldrunner before I noticed. Rejecting kiriki (and not fixing it myself). Would you please have a look at both. | |
| [19:16] <ScottK> Riddell: Sorry, kgoldrunner's fine. It was kigo. | |
| [19:22] <ScottK> kjumpingcube was missing a copyright. Fixed and reuploaded. | |
| === ghostcube_ is now known as ghostcube | |
| [23:10] <Noskcaj> why do you guys still have a ppc version? no-one can run kubuntu on ppc (at least old-world macs) so it get essentially no use and hasn't been tested since i joined. | |
| [23:18] <Riddell> Noskcaj: it only gets released if it's tested so someone will have | |
| [23:18] <Riddell> in general I'd be all for dropping it but others have the opinion of keep it while it gets tested | |