[00:00] * rbelem pokes apachelogger === yofel_ is now known as yofel === DarkwingDuck is now known as DWonderly === DWonderly is now known as DarkwingDuck === santiago-ve is now known as foursixnine [06:42] shadeslayer: ping [08:07] apachelogger: ok [08:08] apachelogger: would it be ok to put series as first argument to kgetsource, since you can do better completion than? [08:09] bulldog98_: quite honestly, if you do completion, just drop the stable/unstable altogether [08:09] simply query stable/ and unstable/ build a superset of results that match ^\d\.\d\.\d$ and complete with those [08:10] since you know out of which set a complete version comes you can auto decide what path to use [08:10] incidentially enough that also makes the command more accessible, even without completion [08:12] apachelogger: the problem is completion is an other process than kgetsource, so we can’t notify kgetsource which set to use [08:12] sure, write a helper script that can be used by both kgetsource and the autocompleter [08:13] apachelogger: the helper script should give out something like version series and in the completion you can prase out series [08:13] autocompleter asks that thing for set of valid versions and autocompletes -> kgetsource kdelibs 4.5.0 -> kgetsource in turn uses the helper script to get a full path for kdelibs version 4.5.0 [08:13] (that said helper script might be a bit of a bogus word ... a file with shell functions ... might be better) [08:14] that way you simply need to source it and it will work like a library [08:14] It would be better to have version first, since we than can complete also the names, if there should be new names [08:15] so kgetsource 4.5.0 kdelibs would be used [08:15] bulldog98_: no, two functions get_versions() which returns a list of versions in both stable and unstable, get_full_path(name, version) which returns the ftpmaster url for that version (be it stable or unstable) [08:15] bulldog98_: no goody [08:15] that makes it less usable when not using autocompletion [08:16] and 90000% of the time one will not autocomplete a name one knows anyway and can type faster than the helper could get the information from ftpmaster [08:16] apachelogger: you can cache that information [08:17] even so [08:17] . [08:17] that’s what I’m doing with kbzr [08:17] app thing version is the most logical thing [08:17] then you can also at some point drop the version altogether and app thing will fetch thing with the latest versio available [08:18] apachelogger: that would be best I think [08:18] still explicit versioning must be support [08:19] plus mind you, doing a comparision on 3-component strings in bash is likely a bit of a pain in the behind [08:19] like the only save way to do it that comes to mind would be regexing the string and then comparing the 3 components individually [08:21] apachelogger: grep ^\d\.\d\.\d$ works, doesn’t it? [08:21] grep? [08:21] wtf [08:22] bulldog98_: shell surely has builtin regex support [08:22] I know bash has [08:22] so, Oneiric Beta, all good to upgrade to? [08:23] Riddell: mostly [08:23] depends on your graphics card, as always ^^ [08:23] from the KDE POV it should be good though [08:23] * apachelogger wonders how to get to the airport [08:35] * bulldog98_ found a way to get the version === bulldog98_ is now known as bulldog98 [08:38] apachelogger: how do I make an array out of an command output in bash? [08:41] it is by default [08:41] you can iter on a space or tab or newline separated list without doing anything [08:45] apachelogger: so stable_version=$(command) gives you an array with that? [08:47] yup [08:47] apachelogger: the array is already sorted, because ls -c sorts [08:49] can I use versions+=$unstable_versions in bash? [08:50] well [08:50] you need to program shell I recon [08:50] so that no issues between zsh and bash arise [08:50] hey guys [08:50] so a string concat certainly would be savest [08:50] whenever I tend to find myself going "can I do X in bash" I usually decide it's time to switch to a real programming language [08:50] version = "${stable_versions} ${unstable_versions}" [08:51] Riddell: bash is a real programming language :P [08:51] Oneiric Beta is looking good [08:51] apachelogger: I also could write that stuff in ruby [08:51] Riddell: it even can do pointers :P [08:51] even if the eval expression is like a line long or so ^^ [08:51] It's running well, I've been using it since A2 and I like where it's going. [08:51] apachelogger: so is sed, but I don't write anything over 1 line in it [08:51] bulldog98: whatever you deem most sensible [08:52] * bulldog98 is now learning ruby :) [08:52] * apachelogger likes to keep simple things bash [08:52] actually I like to keep most scripts bash [08:52] like kde-l10n-common's build ^^ [08:52] apachelogger, why bash when you can perl? [08:52] Riddell: well, sed is vastly more cumbersome than sh though IMHO :) [08:52] DarkwingDuck: cause perl aint on the cd no moar :P [08:53] About time! :P [08:53] it is [08:53] woohooo, I have a way to get to the train station [08:53] I've actually given up on it. [08:54] on perlz? [08:54] Aye [08:54] because you failed to hack the kdesrc-build? :P [08:54] perl seems to be mostly around due to KDE bits [08:54] perl is most universal [08:54] Well, every time I tried to hack it, it failed out. [08:55] the meta-repository management script for Qt5 is also perl IIRC [08:55] Python seems to be a good replacement. [08:55] nono [08:55] php [08:55] we should write some php scripts for kubuntu-dev-tools really [08:55] Yeah, learning that. [08:55] amarok-common and kde-runtime both depend on perl [08:55] * apachelogger hasn't done php in ages anyway [08:55] Riddell: amarok-common? :O [08:56] oh, perhaps because of kconfig_update [08:56] those scripts are mostly perl [08:56] What do we need to write/re-write for kubuntu-dev-tools and what is it currently in? [08:56] then again I could not remember amarok ever transiting configs properly so that dep might be bogus [08:56] DarkwingDuck: bzr branch lp:kubuntu-dev-tools :P [08:57] I'd like to have a script that playz 8bit music encoded in plaintext inline in php [08:57] for in-development entertainment [08:57] Why not write it in HTML5? Simplier. [08:58] s/Simplier/Simpler/ [08:58] DarkwingDuck meant: "Why not write it in HTML5? Simpler." [08:58] cauz html5 is a markup language [08:58] that aint suitable for scripts [08:58] True... [08:58] hmm, search doesn't work in Muon [08:59] * apachelogger looks for a bag to throw stuff in for paris [08:59] * DarkwingDuck raises an eyebrow [08:59] Riddell: which muon? [08:59] Riddell, worked for me yesterday... [08:59] software centre [08:59] WFM [08:59] fresh install, can't find anything when doing a search, although it lists everything [08:59] oh, it might be waiting for xapian update [08:59] Riddell, worked for me... [08:59] apachelogger: http://paste.ubuntu.com/680387/ [09:00] apachelogger: better idea [09:00] Riddell: better report a bug so that jon the taco can look at it [09:00] but I am reasonable certain it might simply be that the xapian index is not yet built [09:00] which should probably be reflected in the UI somehow [09:00] * apachelogger points out that silly ubuntu software center also fails to display waiting for xapian [09:01] lol [09:02] * apachelogger considers software that cant play prn a failure anyway [09:02] blimey [09:02] the mooing also does not work [09:02] that is weird [09:07] morning [09:08] guten morgen bambee [09:08] bambee: tu est francais oui? [09:10] Riddell: yes I am [09:11] bambee: si je loue un cololocation qu'est-ce que c'est un "F3"? [09:15] pour example http://www.leboncoin.fr/colocations/230081894.htm?ca=23_s [09:23] Riddell: F3 veut dire , 3 pièces (ou 3 salles) dont 2 chambres [09:23] F3 = usually means, 3 rooms and 2 bedrooms [09:24] this one : area => 61 m² [09:26] bambee: is there a key to that? I see other terms like T3 too [09:26] morning [09:27] * rbelem pokes Quintasan_ [09:31] Riddell: T3 and F3 means the same thing => 3 rooms, in these 3 rooms you can have 2 bedrooms and 1 sitting room for example [09:31] or 1 bedrooms and two other rooms [09:32] TN = N means the number of rooms , usually everything is explained in the description , for this one it's a F3/T3 with 2 bedrooms and 1 sitting room (the sitting room, the kitchen are in the same room) [09:33] warning: no internet === apachelogger is now known as transitlogger [09:39] Riddell: http://paste.ubuntu.com/680421/ [09:39] (I translated it for you) [09:45] bambee: so what's the difference between F3 and T3? [09:47] Riddell: it's the same thing [09:47] très déroutant :) [09:48] * bambee wonders if his translation is understandable :\\ [09:48] Riddell: do you learn french ? :) [09:49] bambee: I hope to move to guadeloupe to learn it [09:49] woo! [11:05] nigelb: pong [11:05] i be packaging :D [11:08] !ninjas [11:08] Ninja Time! apachelogger, bulldog98, debfx, JontheEchidna, Lex79, maco, neversfelde, nhandler, Quintasan, rgreening, Riddell, ScottK, stalcup, txwikinger, yofel [11:08] 4.7.1 tarballs wanting packaging .... [11:11] yus [11:19] * bulldog98 will push my new helper script and then I’m till round 9 UTC [11:21] * bulldog98 will do some packaging this evening, tomorrow I won’t have time the garden needs some love and my parents kill me if I don’t help them === bambee changed the topic of #kubuntu-devel to: Kubuntu: Friendly Computing | Feature Freeze in effect - 11.10 Beta 1 Released | TODO: http://status.ubuntu.com/ubuntu-oneiric/group/topic-oneiric-kubuntu.html | TODO: https://wiki.kubuntu.org/Kubuntu/Ninjas/Packaging [11:51] huh weird [11:51] my system doesn't start swapping after it runs out of memory on tempfs [11:55] shadeslayer: if i remember right by default tmpfs mounts with a size of half your max ram ^^ [11:56] ah .. [11:56] James147: can i override it with a max size in fstab? [11:56] ssh [11:56] shadeslayer: with size=xG option i think [11:56] whoops [11:56] right [11:57] shadeslayer: you can also use size=x% :) [11:58] what does that do? [11:58] ah [11:58] % of physical ram [11:58] James147: percentage of my RAM? [11:58] yeah [11:58] i gave it 2.5 gigs now [11:58] should be enough [11:59] steveire: Could you have a look at Bug #707878, it seems to be piling up duplicates, so something must be up ... [11:59] Launchpad bug 707878 in akonadi (Ubuntu) "akonadi_agent_launcher crashed with SIGSEGV in QThreadStorage::deleteData()" [High,Confirmed] https://launchpad.net/bugs/707878 [12:00] tmpfs + icecc = build output flying off the screen [12:01] ScottK: Ok [12:03] James147: it should start swapping once it reaches a critical limit right? [12:03] * James147 dosnt know... he has never tested it to its limit [12:04] ^^ but I dont see why not [12:04] well : tmpfs 2.6G 1.8G 793M 70% /var/cache/pbuilder/build [12:32] first day in school == three tests next week [12:32] wtf === Quintasan_ is now known as Quintasan [12:34] Quintasan: i hear ya [12:34] Quintasan: seriously? teachers suck... o.O [12:35] school sucks :P [12:35] +1 [12:35] ScottK: Sorry, never checked IRC yesterday. Thanks to whoever took care of the Beta1 announcement though [12:36] bambee: i can haz meta kde? [12:36] How do I link https://bugs.launchpad.net/ubuntu/+source/akonadi/+bug/707878 to its upstream bug? [12:36] Launchpad bug 707878 in akonadi (Ubuntu) "akonadi_agent_launcher crashed with SIGSEGV in QThreadStorage::deleteData()" [High,Confirmed] [12:37] steveire_: https://bugs.launchpad.net/ubuntu/+source/akonadi/+bug/707878/+choose-affected-product [12:37] shadeslayer: already pushed to ninjas [12:37] alright [12:37] building [12:39] does anyone have random mouse freezes ? [12:40] Quintasan: 'Launchpad does not recognize the bug tracker at this URL.' [12:40] http://bugs.kde.org/261788 [12:40] KDE bug 261788 in general "akonadi_agent_launcher segfaults when stopping the Akonadi server" [Crash,New] [12:41] hmm [12:41] wtf [12:41] any ideas if i can install ubuntu core on a system [12:41] like .. is there a ISO somewhere? [12:43] all i need is a installer and a base system for a offline install [12:43] ryanakca: No problem (it was me - I've got the passwords stored more than one place now, so I can find them.) [12:44] Hehe :) [12:44] ScottK: Can you enlighten me and steveire_ how to link to bug reports in foreign bug trackers in LP? [12:45] * ScottK looks [12:45] !find xine.pm [12:45] File xine.pm found in libxine-dev [12:47] Quintasan: Open the > by Akonadi (not akonadi(Ubuntu)) and paste the b.k.o URL into the URL box under remote watch. [12:47] steveire: ^^^ [12:47] There is 1 error in the data you entered. Please fix it and try again. [12:47] ^_^ [12:47] * ScottK tries. [12:47] lunchpad does not want to hold hands with b.k.o :P [12:48] It's the other way around [12:49] No. Apparently apachelogger needs to configure something. [12:49] Quintasan: Ask in #launchpad. I'm not sure why that doesn't work. [12:59] Quintasan: steveire_ ScottK fixed [12:59] you need the XML link to the bug [12:59] ? [12:59] XML [12:59] http://bugs.kde.org/show_bug.cgi?ctype=xml&id=261788 << this one [12:59] the one that displays XML info [12:59] I see. [12:59] thats what LP parses iirc [12:59] LP should know how to get that from the regular URL. [13:00] shadeslayer: Would you file a bug against LP about that? [13:00] yes, quite weird that it didn't know how to do that [13:00] ScottK: sure [13:00] Thanks. [13:01] bbl [13:02] i'll just try and fix it as well [13:04] ScottK: ok, I'm pushing the new Qt to a ppa, I'll open the FFe and assign it to you rather than the release team? (also a question on bug #602913), it's the new appmenu which fixes it, isn't it? (that's what the plasma is using?) [13:04] Launchpad bug 602913 in plasma-widget-menubar (Ubuntu) "Doesn't work with applications started from a saved session" [High,New] https://launchpad.net/bugs/602913 [13:05] didrocks: Go ahead and subscribe the release team, but I'll look at it. [13:05] ScottK: ok, will do [13:05] The issue you pointed me to yesterday sounded like the same issue p-w-menubar was having in that bug. I'd have to ask agateau which package the fix would actually be in. [13:06] didrocks: Will you also take care of the FFe for the new appmenu-qt? [13:06] bug 839543 [13:06] Launchpad bug 839543 in Launchpad itself "Launchpad cannot track bugs.kde.org reports anymore" [Undecided,New] https://launchpad.net/bugs/839543 [13:06] ScottK: ok, I won't upload before Monday anyway, so I'll chat with him. [13:07] ScottK: is a FFe needed for appmenu-qt? It's only bug fixing [13:07] ah, the alt [13:07] yeah, indeed, will take care of that === jjesse_ is now known as jjesse [13:14] didrocks: I was thinking of libdbusmenu-qt. 0.9.0 is out and in Debian and we only have 0.8.3. [13:15] I guess ask agateau if we want that too or it that's intended for KDE 4.8? [13:15] ScottK: I pushed it already, he asked me to push it yesterday [13:16] it's needed for the new appmenu anyway (2 additional symbols) [13:16] Oh. [13:16] Where was the FFe for this? [13:17] Also, why not take from Debian? Is their package different? [13:17] ScottK: there was no changelog, I didn't see the feature scanning it (anyway, the symbol isn't called until the appmenu lands) [13:17] OK. [13:17] ScottK: the package is different, seems that debian didn't take Riddell's work [13:18] (it has never been merged) [13:18] Seems like he wouldn't have done 0.8.3 -> 0.9.0 without adding features ... [13:18] OK. Then we should look at merging next cycle. Thanks for checking. [13:18] ScottK: well, the 0.1.2 -> 0.2 bump for appmenu was licence fix only, so I didn't rely on it, sorry for the oversight [13:19] yeah, merging would be nice [13:19] (was just a little bit too late on the cycle to start with that, changing packaging and such) [13:20] Agreed. [13:20] huh wat [13:20] * shadeslayer sees new symbols [13:23] didrocks: A good clue for it needed an FFe would have been (from the 0.8.3 -> 0.9.0 diff): [13:23] -set(dbusmenu_qt_lib_API_VERSION 5) [13:23] +set(dbusmenu_qt_lib_API_VERSION 6) [13:25] ScottK: yeah, but I'm so used to the API addition for unity/nux (and some GNOME components) to fix actual bugs that I didn't stopped on that, sorry about it, will take more care next time (the new API isn't used until the appmenu release go in anyway) [13:25] will try to take more care for Qt thing which are more logical, seems with the time, my habits are twisted :) [13:30] didrocks: Qt FFe is approved. [13:30] ScottK: oh great, thanks! if you want to take a look at the appmenu one: bug #737419 [13:31] Launchpad bug 737419 in Unity Foundations "FFe: Add support for the show-now-changed signal" [Medium,Triaged] https://launchpad.net/bugs/737419 [13:31] ScottK: I still prefer waiting on Monday for Qt (buildling on the ubuntu-desktop ppa right now), as I won't be online this week-end, I prefer to avoid breaking the world before leaving :) [13:31] didrocks: I figured Monday would be the P in ASAP. [13:32] ScottK: heh, ok :-) [13:32] didrocks: There's no Ubuntu bug there. Those are all upstream tasks. [13:33] ScottK: argh, seemed launchpad timeouted :/ fixed now [13:37] didrocks: Done. [13:37] ScottK: great, thanks [13:42] uh [13:42] what do i do with these new symbols in kdelibs [13:42] shadeslayer: check if they are ABI breaks [13:42] is rekonq not working with forms known? [13:43] Riddell: but its a bug fix release, so shouldn't that be bad? [13:43] Riddell: try out newer webkit from my ppa [13:43] https://launchpad.net/~rohangarg/+archive/experimental << this one [13:43] shadeslayer: yes if they are ABI breaks that's a release critical bug which you should report to KDE. if not it's all good [13:44] Riddell: and how does one check that if its a ABI break [13:44] ah, that's fiddly [13:44] you have to look at the source code that made the symbols and work it out based on the (often weird) c++ binary compatibility rules [13:45] usually adding symbols is ok, but not always [13:45] usually removing symbols is bad, but not always [13:47] or just ask Thiago :) [13:48] haha :D [13:59] why the hell doesn't pbuilder use /etc/pbuilderrc ? o.O [14:04] shadeslayer: yay, your qtwebkit sorts it [14:04] \o/ [14:04] we need to get that into the archive after the freeze is over [14:10] shadeslayer: Freeze is over. [14:10] oh [14:10] Read /topic. [14:10] right, didn't read that [14:11] Quintasan: yofel can either of you upload qtwebkit? [14:11] i'm heading out for a couple of minutes, bbl === rickspencer3_ is now known as rickspencer3 [14:45] ScottK: Do you have any problems with rbelem's new branches? [14:49] heya Quintasan [14:49] rbelem: \o [14:49] Quintasan, did you see my latest dent? [14:50] rbelem: not really, I didn't have time to look at dents today [14:51] Quintasan, i made really nice command in vim to fix the -data-active.install :-) [14:52] Oh, my, really? That's interesting [14:52] Quintasan, s:$debian/tmp-kde-runtime-active/$$.*/$$.*$$:\1\2\3 \2: [14:52] :-D [14:53] ohshi- [14:54] rbelem: Actually, did you testbuild those packages before commiting? [14:56] ScottK: https://code.launchpad.net/~rbelem/kubuntu-packaging/kde-runtime-active-4.7/+merge/73738 [14:56] rbelem: https://code.launchpad.net/~rbelem/kubuntu-packaging/kde4libs-active-4.7/+merge/73739 [14:56] I'm not entirely sure what you did there [14:56] ooh [14:56] there's this thing called c++filt [14:56] it de mangles all these symbols [14:59] rbelem: Please recheckout kdelibs branch and commit those changes you wanted since something strange happened there [15:00] parallel build with pbuilder is totally borked o.O [15:03] ok ... new symbols look alright [15:05] oh oh [15:05] maco: belated happy birthday :) [15:06] thanks :) [15:13] Quintasan, i tested build kde-runtime. i did manage to finish the test build for kdelibs, power failure [15:14] *did not [15:17] uh [15:17] question [15:17] why do have kde-runtime as one binary package but kdebase-runtime-dbg as the debug package? [15:18] ah nvm [15:18] btw kdebase-runtime-dbg does not depend on kdebase-runtime for some reason [15:18] is that intentional? [15:19] aha ..transitional package [15:27] [lp:~kubuntu-packagers/kubuntu-packaging/kde-runtime] Rohan Garg * 228 * debian/control Bump kde-sc-dev-latest to 4:4.7.1 [15:36] Quintasan: around ? could you check and confirm something for me ? [15:38] build a package with "debuild -j$(getconf _NPROCESSORS_ONLN 2>/dev/null || echo 1)" and looks how many jobs it uses, even debuild is borked here [15:38] it's just impossible to build kde-workspace o.O [15:40] bambee: push the packaging, i'll build it for you [15:41] right after runtime [15:44] s/runtime/baseapps [15:45] actually nothing has really changed... I cannot even know if symbols have changed :'( [15:46] jussi: btw it appears you sent a test sms to moi ealier today [15:47] also I have an algorithm for you spell checking bug [15:47] in case you moved that report to kde yet :P [15:49] transitlogger: the one where krunner crashes when you press the backspace key? [15:49] no [15:49] a feature reuqest [15:49] ah ok [15:49] bambee: looking at it [15:49] * transitlogger aint caring about plasmaware [15:49] new file : -./usr/share/doc/kde/HTML/en/dolphin/toolbar.png [15:51] bambee: Check buildlog for "jobserver" [15:51] shadeslayer: did you card also get locked? [15:51] lol [15:51] shadeslayer: did you ask for a replacement card? [15:52] I hear you need a replacement card if your card was locked [15:52] xD [15:52] but beware, it takes up to half a year to get the new card [15:52] What the hell is transitlogger talking about? [15:52] by that time someone will have hacked your account [15:52] transitlogger: i did lock my card at DS [15:52] Quintasan: questions on the GSoC Mailing List [15:52] shadeslayer: did you request a replacement card then? [15:53] about locked cards and allegations that unicorns don't exsist [15:53] transitlogger: nope, i was sane enough to call them and have the issue fixed [15:53] Quintasan: I didn't look yet. Looking now. [15:53] shadeslayer: what? no! s'impossible!!!! [15:53] haha :P [15:54] ScottK: The libs merge is somehow broken, I presume rbelem will fix it today [15:54] I did not read that on the list [15:54] shadeslayer: others deserve to know that you can call numbas to have locking fixed, my oh my [15:54] had I know, I would not have sued google for withholding monies :S [15:54] Quintasan: The runtime one seems odd too. Like it starts from an older version of our packages. [15:55] ^_^ [15:55] rbelem: PING [15:55] Quintasan, pong [15:56] aw ffffffuuuuuu [15:56] so, there is this really stylish person, silver headphones with an orange cord, birght green tshirt and super light blue jeans sort of pants [15:57] transitlogger: Would you have a look at Bug #832864 - It seems likely there is some confusion between phonon headers and the cmake checks in pyside that make the configure test find phonon in Debian and not in Ubuntu. [15:57] Launchpad bug 832864 in pyside (Ubuntu Oneiric) "pyside version 1.0.4-1 failed to build in oneiric" [High,Confirmed] https://launchpad.net/bugs/832864 [15:57] to complete the picture the orange cord of course builds the nicest color combo with the green shirt [15:57] looks like a unicorn barfed all over him [15:57] ScottK, should i bump the pkg version in -runtime? [15:57] * transitlogger likes aiports :D [15:58] Quintasan: I don't see "jobserver" [15:58] rbelem: Start with the version that we already uploaded and diff from that with a new version. [15:58] :\ [15:58] ScottK: I am about to board, so it better be quick [15:58] Quintasan: I am talking about a local parallel job, not through icecc [15:58] transitlogger: I think the relevant infos are in the bug. [15:58] bambee: >implying I am using icecc [15:59] ScottK, i thought that the version that i appended was not uploaded, thats why i just append [15:59] [lp:~kubuntu-packagers/kubuntu-packaging/kde-baseapps] Rohan Garg * 187 * debian/ (6 files) New upstream release [15:59] ScottK: supposedly the cmake part is broken [16:00] rbelem: You need to base on kde-runtime branch for example [16:00] people always write bogus cmake stuff [16:00] it is like qmake really [16:00] kbzr branch kde-runtime [16:00] make changes [16:00] Or something changed in phonon that shouldn't have. [16:00] commit and push somewhere [16:00] oki [16:01] transitlogger: btw lunchpad does not want to talk to b.k.o it seems [16:01] transitlogger: how does one fix that in lunchpad [16:01] ScottK: nope [16:01] unless you lot patched my software [16:01] shadeslayer++ [16:01] which you have [16:01] and you did not send patches upstream [16:01] I should give you all a spanking for that at some point [16:02] shadeslayer: it better not [16:02] shadeslayer: why would it? [16:02] transitlogger: remote bug tracking in ad [16:02] shadeslayer: lunchpad would only break bko [16:02]