/srv/irclogs.ubuntu.com/2012/11/29/#kubuntu-devel.txt

=== clay_doh is now known as claydoh
=== claydoh is now known as Guest5889
=== Guest5889 is now known as claydoh
ScottKJontheEchidna: Any thoughts about what's wrong in http://paste.ubuntu.com/1395372/ - only fails in armhf.01:53
JontheEchidnaScottK: qreal v. double inconsistency02:11
JontheEchidnain the WorksheetEntry interface class, layOutForWidth is defined as: virtual void layOutForWidth(qreal w, bool force = false) = 0;02:12
JontheEchidnabut it is implemented as virtual void layOutForWidth(double w, bool force = false) = 0; in the derived classes02:12
JontheEchidnawhich on !arm is ok because qreal == double02:12
JontheEchidnabut on arm it's a float, which is a different signature, which means layOutForWidth is not defined in the derived classes02:13
JontheEchidna(I know you know the float v. qreal differences, but to clarify... ;-)02:13
JontheEchidnaScottK: changing the line "virtual void layOutForWidth(double w, bool force = false);" to "virtual void layOutForWidth(qreal w, bool force = false)" would fix02:14
JontheEchidna(in the TextEntry, CommandEntry, and LatexEntry headers)02:14
JontheEchidnaseems to be one of the more subtle bugs of this type02:15
ScottKOK.  Thanks.  I'll try that.02:17
=== ronnoc_ is now known as ronnoc
ScottKqreal ALL the things.02:51
shadeslayer_ronnoc: pong 02:51
ScottKIf someone has a raring install, it'd be lovely for them to build kphotoalbum at https://github.com/jzarl/kphotoalbum/tree/kipi and see if it works.02:53
JontheEchidnak02:57
ScottKComments back to KDE bug 30714803:01
ScottKafiestas: There's a proposed patch in https://bugs.kde.org/show_bug.cgi?id=307147 for Kamoso with KDE SC 4.9.80 that's been there sinced september with no comments.  Would you please have a look at it?03:02
JontheEchidnaScottK: builds fine here03:07
ScottKOK. How about working?03:31
ScottKCantor uploaded...03:47
ScottKJontheEchidna or shadeslayer_ or somebody else with KDE commit rights: would you please commit http://bazaar.launchpad.net/~kubuntu-packagers/kubuntu-packaging/cantor/revision/54 upstream?04:04
=== Guest33109 is now known as gaurav
shadeslayer_ScottK: will have a look today :)05:10
shadeslayer_ScottK: best to just upload to review board and get the maintainers to commit it05:12
shadeslayer_that way you can also apply for kde commit rights at some point in the future05:12
=== shadeslayer_ is now known as shadeslayer
=== vHanda_ is now known as vHanda
shadeslayerincase anyone does kbreakout packaging stuff http://paste.kde.org/616532/06:46
shadeslayerI got upstream to do some packaging :P06:47
=== Tonio_ is now known as Tonio_aw
=== markey_ is now known as markey
RiddellScottK: cantor fix committed09:56
Riddellyou're getting quite good at these arm fixes :)09:56
=== Tonio_aw is now known as Tonio_
ScottKRiddell: Thanks.  JontheEchidna helped on that one.11:10
ScottKMy force-hint worked, so all of 4.9.80 is in raring now.11:20
=== Tonio_ is now known as Tonio_aw
RiddellScottK: yay :)12:04
Riddellthanks for taking care of that12:04
Peace-hola :D12:05
Riddellhi Peace- 12:07
soeeraring ?12:07
* Peace- installing raring12:07
Riddellsoee: our development version12:09
soeeah ok12:09
Peace-hi Riddell12:09
* Peace- installed on pavillion with x200 ati12:18
Peace-ok the installed worked fine :)12:19
Peace-installer *12:19
RiddellPeace-: using today's image?12:29
Peace-Riddell: yes12:32
Peace-let me check but usually i use the daily live 12:32
Peace-Riddell: http://cdimage.ubuntu.com/kubuntu/daily-live/current/12:32
Peace-:)12:32
Peace-The only thing that doesn't work it's the wifi 12:33
Peace-addittional drivers says no proprietary drivers are in use and it gives no options 12:34
Peace-but well i have the broadcomn 430612:34
RiddellPeace-: what version of kde is on it?12:39
Peace-4.9.80 Riddell12:39
Riddelllovely12:41
Peace-Riddell: btw installing the firmware now wifi works 13:11
Peace-but on jokey-kde you can't install it why?13:11
Peace-i remember some old release that asked to install the broadcom stuff13:12
dantti_laptopisn't ubuntu droping jokey?13:15
dantti_laptop*wasn't13:16
Peace-yes13:16
Peace-dantti_laptop: are you brazilian guy ?13:16
dantti_laptopyup13:16
Peace-wtf i have read of your problem with jail -ambassy 13:17
Peace-embassy13:18
Peace-what a mad story man13:19
dantti_laptopyup, unfurtunatelly not over yet :/ 13:19
=== dpm_ is now known as dpm
ScottKdantti_laptop: Welcome back.13:40
dantti_laptopScottK: thanks13:40
=== Tonio_aw is now known as Tonio_
=== yofel_ is now known as yofel
yofelam I the only one whose akonadi demands a vacation in raring? http://paste.kde.org/616892 13:55
ScottKyofel: It may be on strike until you upload 4.9.2 to quantal-proposed.13:56
yofelheh13:56
yofelI'll try to look at 4.9.3 when I get home13:56
ScottKThanks.13:57
ScottKYeah, meant 4.9.3.13:57
=== LordOfTime is now known as TheLordOfTime
=== 20WABM4XS is now known as jussi
=== jussi is now known as jussi01
=== jussi01 is now known as jussi
jussiyay14:24
BluesKajHi all14:24
diahanehey guys and girls here is a free open source site http://bitfungus.org/ feel free to contribuite on that, we would like to make your/our life easier when searching for code resources, sorry for my bad english, if you want you can add your own resource link or third parties one14:39
=== Tonio_ is now known as Tonio_aw
ScottKafiestas: ping15:37
afiestasScottK: read the kamoso thing., is in my todo already15:37
ScottKafiestas: OK. Thanks.15:37
* Riddell onto kdegames packaging15:44
ScottKRiddell: Did libkdegames get sorted?15:48
RiddellScottK: seems in good shape to me, do you know of any issues left with it?15:50
ScottKRiddell: Just the note in the pad.15:50
RiddellI thought the copyright was done, shadeslayer's e-mail to kde-games-devel doesn't seem to have come through15:53
ScottKDunno.  15:54
Riddellshadeslayer?16:01
ScottKshadeslayer: ?16:02
ScottKYour turn.16:02
Riddellrohan?16:03
Riddelloh he's at foss.in16:03
Riddellat the woefully underpublicised kde mini summit16:03
ScottKCalligra RC is supposed to release today.  I think I'll go ahead and upload it.16:04
* ScottK already did packages.16:05
RiddellScottK: I never wait with calligra, I just upload it once the packaging is done16:06
ScottKRight.16:06
ScottKWhat's the lcms that's mentioned in conjunction with calligra in the packaging spec?16:23
ScottKCalligra 2.6 rc2 is uploaded.16:23
Riddella colour library?16:25
oyyes a ICC CMM library16:25
ScottKProject page?16:27
oywww.littlecms.com16:27
ScottKThanks.16:27
oyyou are welcome16:27
ScottKPackage looks pretty unmaintained in Debian.  Two NMUs are the last two uploads.16:30
micahglcms2 should be supported16:35
micahg(assuming calligra supports it :))16:35
RiddellI'm sure I remember adding this as a calligra dep at some point16:35
afiestasa friend of mine asks if there is a ppa or something with beta1 for Quantal16:41
afiestasis there any?16:41
Riddellafiestas: I'm afraid not, first update of the cycle always takes longer16:43
agateauafiestas: project-neon ftw!16:43
Riddelljust packaging kdegames is a long process16:43
agateauRiddell: btw, gwenview now optionally depends on lcms2 as well16:43
oyScottK, yes, 2.4 would be good to have, where are the NMUs listed?16:44
Riddellagateau: noted for further investigation16:44
afiestasRiddell: why is that? having neon already building those games shouldn't speed up the process?16:44
Riddellafiestas: I doubt neon has copyright files complete enough for the archive16:46
=== Tonio_aw is now known as Tonio_
afiestascopyright files for debain should specify old copyright holders? or only the ones owning actual lines of code?16:47
Riddellonly the ones owning actual lines of code16:47
afiestascan't git blame be used for the job then?16:48
afiestasno better way of having a complete copyright 16:48
=== Tonio_ is now known as Tonio_aw
yofelneon is helpful to figure out the build-deps, but the copyright files are... rather incomplete...17:00
afiestasyofel: then can't we use git blame?17:06
afiestasor svn blame for that matter17:06
yofelfor what?17:06
afiestasfor getting all the copyright holders of a file/project17:06
yofelshadeslayer dug out some copyright parser I believe that I haven't tried yet. I usually use a combination of licensecheck and grep17:07
afiestaswhy parse the copyright when you can get all teh accurate data from git/svn ?17:08
yofelafiestas: in case you don't know how the copyright files are supposed to look like: http://dep.debian.net/deps/dep5/17:08
afiestaslicence still will have to be greped though17:08
yofelwe need to copyright file in the binary packages17:08
yofelyou can't read the source from there17:09
afiestasI don't know what a bianry package is, the thing that I install_17:09
afiestas?17:09
yofelthe .deb yes17:09
yofeland if I understood ScottK right, then simply having the source available somewhere isn't enough17:09
afiestasbut you create the copyright file used to build the binary don't you?17:10
afiestascan't that file be generated using accurate data from git?17:10
yofelwell, sure, if you write a copyright file parser that returns a formatted copyright file17:11
yofelwe need a) list of files that the license applies to b) short license text and pointer to full text c) copyright holders with name and year17:11
yofeli.e. what the source files usually have as header17:12
afiestaswe can get C with 100% accurary, the rest must be human checked since there is no perfect parser17:12
afiestasthen I wonder, do we have something that checks if there's been a difference in the header from version X to version X+1 ?17:12
afiestasor do we check all files one by one?17:12
yofelthere's 'licensecheck' which tries to parse the license out of the files, I then usually grep for license and feed the file list into grep Copyright to get the holders and then filter them out17:13
yofelthat's my workflow at least17:14
Riddellcopyright files are rarely updated with new versions (which is a bug)17:14
afiestasI'm a little bit lost still17:14
afiestasso we have shadeslayer neon with master packages that in theory work17:15
yofelRiddell: do you have a workflow for that other than re-creating the file from scratch each time?17:15
afiestaswe can't use them as a base for beta1 release because we need to update copyright17:15
Riddellyofel: nope17:15
afiestasand according to what yofel says checking the copyright seems like a fairly quick thing to do, where is the trick?17:15
yofelafiestas: neon packaging is quite different from archive packaging for various reasons17:16
afiestasamount of packages? amount of manpower?17:16
yofellatter mostly17:16
yofelboth though17:16
Riddellno paticular problem, I'm doing it now and I expect to be finished sometime tomorrow morning17:16
afiestasRiddell: 10 days after release, we should do better ideally17:17
afiestasis KDE not doing a good work preapring releases?17:17
afiestasbeta releases I mean, I know that for final releases there is a period of time where distros can create packages17:18
afiestasdo we do that for beta?17:18
yofelit would be good if someone from the KDE side would do a copyright check with debian-strictness before release. But otherwise the release management is fine17:18
yofelit's mostly lack of manpower for the amount of packages on our side, and our automation scripts are still new and have lot of improvement potential17:19
Riddellafiestas: first update of the cycle always takes extra time17:19
afiestasI know that we can't change this but, is there any legal requirement to do have all this copyright info in the pacakge?17:19
afiestasRiddell: we should start to package earlier then (ofc if possible and if that won't mean wasting effort)17:20
agateauafiestas: yes, most licenses require that packages should be shipped with copyright notices17:22
agateauScottK explained this a few days ago iirc17:22
afiestasbtw I'm talking about this from a upstream PoV, it is stupid to release a beta if nobody can test it17:23
afiestasmaybe we should call it "Package time" or something like that17:23
afiestasagateau: so I guess archlinux is ilegal (they do a simple tarball afaik)17:23
agateauafiestas: most likely, yes17:24
agateauafiestas: I assume no free software developer has yet bothered to attack them for not respecting licenses :)17:24
agateaubut Debian is a lot more picky about licensing17:25
agateauafiestas: have you read heard about the json license problem?17:25
agateaus/read//17:25
* yofel wonders what RPM does here17:26
yofelall I can find is a one-line copyright field17:26
afiestasagateau: nope17:27
agateauafiestas: http://tanguy.ortolo.eu/blog/article46/json-license17:27
yofelfun17:28
shadeslayerRiddell: ScottK I've managed to rope in viranch, he  works on some of the KDE Games and he went through a couple of sources and generated copyrights for us17:29
shadeslayerWe also discussed if we should ship these in the sources by default since then we can just copy the file over to debian/copyright17:29
afiestasagateau: so muon is ilegal as well :33? (shadeslayer :p)17:30
agateauafiestas: why would it be illegal?17:30
shadeslayerafiestas: distrubuting binary packages without the debian/copyright would be illegal, yes :P17:30
afiestasagateau: because yofel said that neon packages are missing some ocpyright stuff17:31
yofelrdieter: just to make sure I understand this right: In a RPM spec you have one single field that documents the License of the whole package? What do you do if licenses are mixed? Put all licenses into that field?17:31
agateauafiestas: ah neon, not muon17:31
afiestasaaaah, sorry17:31
afiestasneon17:31
yofelneon is as grey as it gets I fear...17:31
afiestasI always mess both of them xD17:31
rdieteryofel: yes, e.g. License: LGPLv2+ and GPLv217:31
yofelrdieter: you guys sure have it easy...17:32
agateauI have a great idea to solve that copyright mess!17:32
agateaulet's request copyright assignment!17:32
yofelrdieter: thanks17:32
agateauscnr17:32
rdieteryofel: though we often simplify it if possible at packagers' discretion to only list the "effective" license (ie, in the above example, License: GPLv2)17:32
afiestasso redhat that is super picky with patents and  licences only requrie one line17:33
afiestasinteresting17:33
yofelrdieter: ok, that at least explains why we and debian seem to be the only ones that are picky about the copyright contents17:33
agateauafiestas: unfortunately we inherit Debian pickiness here, not sure we can avoid this17:33
afiestasagateau: first thing I said "I know we can't change this"17:34
yofelnot reall, we work from the same policy17:34
agateauafiestas: the only workaround iirc is ppa17:34
yofel*really17:34
* shadeslayer notes that copyright is probably the most time consuming part17:34
agateauhave a beta ppa which can be used to quickly push packages, then get proper copyrighted packages in the archive17:35
yofelPPA's have some copyright rules as well, it's not like you can put _anything_ in there17:35
shadeslayer^17:35
Riddellshadeslayer: got links to those copyright files?  your kbreakout you posted earlier has timed out17:35
agateauyofel: sure, but it doesn't seem to be much of a problem if we look at neon17:35
shadeslayerRiddell: lemme check17:35
yofelit's not a problem as long nobody cares17:36
shadeslayerdrat17:36
shadeslayerRiddell: all links timeout, I'll just ask him to paste them 'forever'17:36
agateauyofel: do you think people would start to care if we put kde beta and rc in a ppa?17:36
yofelprobably not... as long as we fix it later. Riddell ?17:37
Riddellagateau: not from a legal view no17:37
RiddellI just find it easier to get it sorted in development archive before doing backports to a ppa17:38
afiestaswe need scripts to make this more automatic imho17:40
afiestaswe can easily check modified files in the header area, new created files, and deleted files17:41
afiestasat least give to the packager a hint of what to look at17:41
yofelwe do, they just don't really work for beta117:41
afiestasyofel: why?17:41
yofellots of new dependencies etc. so it took a while until the packages built17:42
yofellater on will be faster17:42
Riddellnew dependencies, new files, deleted files, new apis, new abis, broken abis17:43
* yofel makes his way home - bbl17:43
Riddella large variety of changes needed17:43
RiddellScottK: ninjas build in new if you like to keep apol_ happy (and I do)17:44
apol_:D17:44
* shadeslayer goes off to fix bug in kdiamond17:47
* Riddell is squatting in a closed library and wonders when the polis will show up17:48
* agateau is off as well17:48
ScottKRiddell: I'm away from my computer now.  I'll look at it later.  Calligra-l10n in New tooif you could review.17:49
RiddellScottK: accepted!17:50
ScottKThanks.17:54
Riddellhmm I just looked at the debian packaging to see how they're managing with kdegames but the recent uploads are 4.8.4..17:55
hrwhej18:07
hrwdoes someon know why /usr/lib/kde4/libexec/kscreenlocker_greet takes 12% of cpu all the time?18:08
hrwit is huge amount of power...18:09
afiestashrw: it shouldn't18:09
hrwok, got down to 4% but still 18:10
afiestashrw: which version of kde are you using?18:10
hrwraring-proposed one18:10
afiestasno idea of what is18:11
hrw4:4.9.80-0ubuntu718:11
afiestasoks, then is the new screen locker18:11
afiestascan you install debug symbols for it, and then do some debugging?18:11
hrwyes, the one which requires me to enter password twice.18:11
afiestastwice_18:11
hrwnot today I'm afraid18:11
afiestas?18:11
afiestas:s18:11
afiestasonly once here18:12
hrwon previous version when I pressed Backspace + password it worked. now I wake up screen with Backspace, enter password, get 'wrong password' and enter it again18:12
hrwok, will try to find some time tomorrow to check what is wrong18:13
afiestasbackspace + password?18:13
hrwto remove any chars I entered when screen was off18:14
afiestasI'm quite lost  :/ but report a bug pls18:14
afiestasgetting the new screen locker to the quality of the old one is important18:14
hrwsure18:15
hrwtoday I am out of home so not possible to check ;D18:15
hrwhave a nice day afiestas - have to go18:16
afiestashrw: you too18:19
shadeslayerany ideas why k-d-s depends onlibnewt0.5218:21
Riddell  * Use the new default virtual terminal color palette instead of forcing the18:23
Riddell    original one. (LP: #739943)18:23
ubottuLaunchpad bug 739943 in kubuntu-default-settings (Ubuntu) "package kubuntu-default-settings 1:11.04ubuntu7 failed to install/upgrade: subprocess installed post-installation script returned error exit status 2" [Medium,Fix released] https://launchpad.net/bugs/73994318:23
Riddell  * Depend on libnewt0.52 to make sure that the newt-palette alternatives are18:23
Riddell    registered.18:23
Riddell -- Felix Geyer <debfx-pkg@fobos.de>  Tue, 22 Mar 2011 20:13:21 +010018:23
Riddellsays changelog18:23
shadeslayeraha18:24
shadeslayerokay18:24
yofelre18:38
Riddellwhat does re actually mean?18:39
yofelRiddell: re-turned18:41
yofeli.e. back18:41
Riddellwelcome back :)18:42
tsimpsonif we frown on public away messages, should not we frown on public back messages? ;)18:42
yofelthanks ;)18:43
ronnocDarkwing: Does Blue Systems have any type of formal support services in place as of yet? If not, is it still a goal?20:07
Tm_Tronnoc: wy specifically Blue Systems?20:12
ScottKronnoc: There's work in progress on making support contracts available.21:31
ScottKFor packaging lcms2, the symbol _cmsGetStockOutputFormatter@Base has gone missing.  Any idea if we care.21:32
ronnocOK thanks ScottK. If you remember, can you ping me when there is something concrete? 21:32
ScottKI'm almost always glad to agree to things on the condition I remember to do it.21:33
ScottKCalligra doesn't use it.21:35
yofellcms2 has a lot of rdepends, so i would say we care21:36
yofeland it's used by krita from what I see21:36
ScottKI'm downloading all the sources and I'll grep to see if it's used.21:39
ScottKkrita uses lcms, but not that symbol.21:40
ScottKRiddell: You have mail on the ubuntu-release list.21:40
yofelok, now let's give this sru script a try...22:17
yofelhm, bug 108101822:23
ubottuLaunchpad bug 1081018 in Kubuntu PPA "lose oxygen style with 4.9.3" [Undecided,New] https://launchpad.net/bugs/108101822:23
yofelsomeone kill Trolltech.conf with fire22:24
ScottKRiddell: Accepted ninja.  Watch file doesn't work though. Ought to be fixed.22:28
yofelhm, gnome-keyring killed the script, next try22:32
ScottKlcms2 done.22:44
yofelScottK: is it fine if I mass-upload 4.9.3 to quantal-proposed?22:51
=== Mamarok is now known as Guest88910
=== Guest88910 is now known as Mamarok
* yofel uploads and hopes the release team won't strangle him23:49
yofelI'll do l10n tomorrow, forgot about it23:58

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