[00:16] man, this whole SSO bit is turning into a large project [00:19] apachelogger: trololo, the api has changed entirely: http://paste.ubuntu.com/577661/ [00:19] at least there exists documentation pertaining to this api [00:28] * JontheEchidna gives up for today and moves on to recreational things [01:08] tsimpson: heh, I've never seen somebody file a bug that the modifications he made to a package break it :P [01:09] people are strange, general observation [05:41] How do you make a group with no name? [06:24] JontheEchidna: told ya :P [06:32] Yeah! [06:32] You got told! [06:33] oi [06:33] * apachelogger should be on campus in 25 minutes :O [06:33] and I need to buy foods [06:33] omg omg omg omg [06:34] * apachelogger throws all mobile devices he can find in a bag and looks for a hair dryer [08:49] morning [08:50] hi === hunger_ is now known as hunger [12:36] ScottK: pingo [13:31] apachelogger: Pong [13:31] ScottK: what do I do about the copyright situation with the device integration? [13:32] Is there no hint anywhere in the code of copyright/license? [13:32] nothing [13:32] neither in their rpm packaging [13:32] On a website? [13:34] ScottK: no [13:34] well [13:34] the gitorious project page lists other/multiple or something like that as license [13:34] insanely helpful [13:35] I guess I'd email whoever owns the project and ask. [13:36] * apachelogger feels hold ups creeping up behind him again [14:25] [kdelibs] Harald Sitter * apachelogger@ubuntu.com-20110309142535-12o1zhfcxfft2y4o * debian/ (5 files in 2 dirs) * Import kubuntu_02_no_direct_gl_usage.diff from upstream KDE - removes direct GL usage from glapplet, eliminates need to link against libGL [14:38] ScottK: do you happen to know if our cmake still does recursive linking? [14:39] I believe my earlier gles trials with plasma-mobile failed because kdelibs linked against libgl and rescursive linking dragged this into plasma-mobile which then had weird runtime symbol clash stuff going on (with libgl and libgles being in memory and stuff) [14:39] if indeed recursive linking is at fault we probably need to rebuild workspace, plasma-addons and mobile to get anywhere :S [14:40] Dunno [14:40] also apparently our kernel does not like usb charging ... also our kernel is still stuck in NEW :S [14:44] I can't New kernels. [14:44] Needs Riddell or some other Canonical archive admin. [15:28] for some reason I just pictured Riddell in a bollywood film [15:28] * apachelogger is going mad clearly [15:30] ScottK: don't you have friends to help get it out of new? ;) [17:08] shadeslayer: around ? how was the first day ? [17:12] except anything related to project neon, is there any work left for KDE? [17:13] bambee: tired like hell [17:14] i spent the day running around [17:14] c2tarun: but there are many things related to kde and qt :) [17:14] bambee: like what? [17:14] also [17:14] Knut Yrvin is AWESOME [17:15] c2tarun: http://kde.in/conf/schedule/ [17:15] * c2tarun shadeslayer making me jealous ;) enjoying there alone [17:16] like "pykde4", "how to start contributing to kde", "playing with pyqt" and so on [17:16] we saw him dance today [17:17] (I did read the schedule quickly , I am probably not the best one to talk about it) [17:17] bambee: you sure you gave me the right link? its schedule for kde.in/conf [17:20] bambee: ping [17:21] why I did not ? [17:22] bambee: not getting...... :/ what are you saying? [17:22] we're talking about conf.kde.in nop ? === cmagina is now known as cmagina-lunch [17:23] * c2tarun confused :/ [17:24] apachelogger: qdbusviewer to the rescue: http://paste.ubuntu.com/577936/ [17:25] c2tarun: I asked shadeslayer about the first day of conf.kde.in, to know if it was interesting... then I said "there are few things related to kde, qt, and so on" (mostly qt today). [17:26] that all :) [17:26] agateau: around? [17:26] then you asked me "are you sure?" so I pasted the schedule link [17:26] bambee: actually I missed that event :( I was asking is there any other packaging work left for kde. :/ [17:31] c2tarun: so it was a mis-understanding, sorry [17:31] agateau: we have a bug session tomorrow, can you point me to a list of JJ's [17:31] bambee: no prob :) [17:36] apachelogger: I'll see if I can arrange something. === cmagina-lunch is now known as cmagina-afk === cmagina-afk is now known as cmagina [18:02] apachelogger: jdstrand is going to look into it. [18:03] seiflotfy: good words about you in the nepomuk talk today at conf.kde.in [18:05] Riddell, oh really [18:05] hahaha [18:05] i am writing good things baout KDE [18:06] Riddell: ping [18:41] Riddell, show me show me [18:41] :P [18:45] Riddell: did you update the seeds? :D [18:54] apachelogger: linux-n900 is out of New. [18:55] ScottK: can we push mobile-broadband-provider-info into the Kubuntu seeds? [18:56] shadeslayer: Why? [18:56] ScottK: because without it you cannot tether your phone to your PC [18:56] me and Riddell spent a fair amount of time copying debs today to make it work :P [18:57] It's small enough, I don't see why not. [18:57] I [18:57] I'll add it. [18:57] thanks :) [18:58] basically it serves the default info for connections [19:00] shadeslayer: Seeded. It'll be in the metapackage after the next kubuntu-meta upload. [19:00] kool [19:03] apachelogger: OBS is cool man [19:03] i talked to wstephenson, and i think i'm in love with OBS [19:06] shadeslayer: Last I looked it built pretty crappy .debs, but it's been a while since I looked. [19:06] One of my favorite bits being where they use different package names so they make .debs that conflict with things in the archive. [19:07] ScottK: they have this cool thing where if a build dep gets rebuilt, the package is automatically rebuilt [19:08] s/package/package with that build dep [19:08] and they have spec files which convert debian build deps into rpm named build deps .... [19:13] shadeslayer: You should blog about it and make sure it goes on Planet Ubuntu. [19:17] ScottK: wooohooo, thank you [19:17] shadeslayer: yes it is [19:17] apachelogger: You're welcome. [19:18] if launchpad used OBS and soyuz work actually went into that, launchpad would be a lot more lovely [19:18] only today I filed a bug for soyuz to get buildlog-while-building stuff improved [19:18] it is a sad thing really [19:19] ScottK: crappy debs from OBS are hardly the fault of OBS I must say [19:19] though IIRC it permits a bit of fishy package building WRT deb [19:19] apachelogger: They are when OBS pretends to be able to make a proper .deb from a srcrpm and doesn't manage it. [19:20] then again this seems natural as rpm is surely of more importance to them and deb and rpm are just too diffierent on a packaging level [19:20] It's been awhile since I looked, so it may be better now. [19:20] ScottK: you can upload your debian dir [19:20] just that most people don't do it [19:21] bleh [19:21] kde4libs still building [19:21] at least I did not break it :D [19:21] ScottK: I am happy to report with 4.6.1 plasma-mobile starts on the n900 again [19:21] apachelogger: Excellent. [19:21] but it appears to me our kernel still doesnt do usb charging [19:21] that definitely needs looking into [19:22] since one can only charge via the micro usb port ^^ [20:02] http://blog.canonical.com/?p=528 Ah.. no Kubuntu anymore ;) [20:07] Nice. You must be logged in to comment. [20:07] txwikinger: That's not really a change. [20:07] ScottK: Just teasing === Tscheesy_ is now known as Tscheesy [21:06] https://bugs.launchpad.net/ubuntu/+source/qt4-x11/+bug/731504 <- someone, but not me any more, should respond [21:06] Ubuntu bug 731504 in qt4-x11 (Ubuntu) "Qt will not static compile SQLIte" [Undecided,Confirmed] [22:00] ScottK: I love the remix concept [22:00] as if anyone implemented that [22:05] tsimpson: what would be the hack this person talks about in his last comment? [22:06] apachelogger: basically, they want us to make debian/rules work dynamically with any of the possible configuration options you can give to Qt [22:07] so if they change what files are produced, for instance making a plugin static, it won't complain when it doesn't find the plugin to put in a .deb [22:07] yeah yeah I got that [22:07] basically, just wrong things [22:07] but in his last comment he says that we maintain a hack [22:08] oh, no [22:08] the "unreliable hack" is modifying the debian/control or debian/*.install [22:08] whereas his solution (which actually violates policy by deleting .install and thus the package will not be in prestine state after clean anymore) [22:08] ie: what you're supposed to do [22:08] oh [22:08] well [22:08] easy enough [22:09] if mister president wants something he will have to pay either us or canonical to maintain it [22:09] easy as that [22:09] apachelogger: I, and JontheEchidna, have already marked the bug as invalid several times now [22:10] ^ and explained why to them [22:10] they just refuse to accept that we should not make debian/* files work with any possible modification anyone may see fit to do [22:10] well yes [22:10] so we outline the rules for having this happen [22:11] if they donate 50 bucks per month per flexible option to kubuntu I am all for maintaing it [22:12] I've run into seasonedgeek before. This is typical of him. [22:12] go ahead and reply then ;) [22:12] yeah [22:13] just got distracted by important business of money being thrown at me :P [22:13] as long as it's not coins, they really sting :) [22:41] tsimpson: ping [22:41] apachelogger: pong [22:47] tsimpson: see query [22:53] debfx, JontheEchidna: should the static sqlite bug report get opened again, please close it without comment from now on [22:53] https://bugs.launchpad.net/ubuntu/+source/qt4-x11/+bug/731504/comments/15 [22:53] Ubuntu bug 731504 in qt4-x11 (Ubuntu) "Qt will not static compile SQLIte" [Undecided,Invalid] [22:54] kk. I almost did when I closed it again last night, but he told me to RTFM. :P [22:55] yeah [22:55] generally I'd suggest to put a price stamp on things if people come up with such things [22:55] it becomes a lot less interesting if one has to pay for it [22:56] [22:56] I am going to have to marshall that myself, aren't I? [22:57] [ 40%] Generating SSOInterface.cpp, SSOInterface.h [22:57] Got unknown type `a{ss}' [22:58] *demarshall [23:08] JontheEchidna: yes [23:08] there is code fo rthat in u1-kde though [23:08] I think [23:08] might be shitty code though [23:08] apachelogger: I saw an Error class in the ubuntu sso stuff you did that seems to do similar to what I want [23:08] I only go the hang of sensible demarshalling like after having written 90% of the marshalling code ^^ [23:09] ^^ [23:09] demarshalling a{ss} should be easy though, just iter it and insert the values in a qhash [23:10] I think for QApt2 it would be better use rigidly-defined objects that I can marshal into DBus args rather than typedef'ing a QStringHash [23:10] easier to doc, less prone to incompatibilities, but still expandable [23:11] JontheEchidna: http://bazaar.launchpad.net/~apachelogger/ubuntuone-client-kde/trunk/view/head:/src/libs/MarshalArgs.h [23:11] QStringHashHash is awesome :D [23:12] I was about to comment to that effect :D [23:14] QHash > [23:14] in C++0x you can condense the last two '>' [23:14] best new feature :P [23:14] well [23:15] arguable :P [23:15] no the best is initializer lists [23:17] I take that back, the best is lambda [23:17] how many times have you written a tiny static/inline function just so you can use it in one other function, such a pain [23:18] never :P [23:18] I have written lessThan functions for use in Qt's qSort plenty of times [23:19] and now you can use templates defined inside functions for other template functions [23:19] neat [23:20] JontheEchidna: you dont talkabout software development please [23:20] I just hope people don't abuse the auto keyword [23:21] you do understand very little about it I just read [23:21] :P [23:21] https://bugs.launchpad.net/ubuntu/+source/kubuntu-meta/+bug/414562 [23:21] Ubuntu bug 414562 in kubuntu-meta (Ubuntu) "[Karmic] No longer have any icon method of deleting trash" [Undecided,Invalid] [23:22] lol [23:22] auto is just a short-cut, the compiler still needs to be able to completely deduce the type at compile-time [23:22] screw auto [23:22] screw C++0x [23:22] C ftw! [23:22] no [23:22] LIES [23:22] actually [23:22] JAHAVAHAHA [23:22] <3 [23:23] apachelogger: I bet you are now drinking the Objective C kool-aid [23:23] who doesn't? [23:23] everybody except Apple :P [23:23] objective-c is like writing markup [23:23] with loads of [][][][][[[[[[[]]]]]]]] [23:23] OOP + C, wasn't that what C++ was made for? ;) [23:23] ;) [23:24] srsly [23:24] did you ever try reading objective c? [23:24] serious head twisting is necessary [23:24] "you don't call methods, you send messages" [23:24] apachelogger: oh, and I promise to not talk about software development in 2009 :D [23:24] but I just want to call a method :| [23:25] JontheEchidna: are you claiming this changed? [23:25] so no time travelling back then for me [23:25] tsimpson: forget it [23:25] I actually am very much in favor of only using invokeMethod in Qt [23:26] actually, knowing how that works helps with reading objective c code ^^ [23:26] you love the Q_SLOTS [23:26] but you can't, invokeMethod would involve invoking a method... [23:26] https://bugs.launchpad.net/ubuntu/+source/mysql-dfsg-5.1/+bug/419400 [23:26] lulz [23:26] Ubuntu bug 419400 in mysql-dfsg-5.1 (Ubuntu) "[Karmic] MySQL security problem" [Undecided,Won't fix] [23:26] yeah, I read that one too ^ [23:26] JontheEchidna: Q_INVOKABLE [23:27] slots are for pufftas, real man use a callback and then invokeMethod [23:27] oh [23:27] there is flaw in that logic somewhere [23:27] JontheEchidna: that is how the better part of phonon gstreamer works btw ;) [23:28] though I think invokemethod is mostly gone now [23:28] amazing work on eventlooping has been done, so we only have very few callbacks left [23:32] the reason that qtmultimediakit is now officially inferior to the empire of phonon: http://qt.gitorious.org/qt-mobility/qt-mobility/blobs/master/plugins/multimedia/gstreamer/mediaplayer/qgstreamerplayersession.cpp#line848