[00:00] Did anyone try to contact someone who has the @kubuntu account on Twitter? [00:01] IIRC Darkwing did. [00:02] sleep time, let's see what news tomorrow brings [00:02] Riddell: Good night. [00:06] Niiiiice [00:07] lets see how much the googlesphere cares https://plus.google.com/100951374600950862471/posts/VVxNPRCbSyQ [00:07] wacomtablet 2.0 in saucy [00:08] We should have better support for those in saucy now === yofel changed the topic of #kubuntu-devel to: Kubuntu - Friendly Computing | https://trello.com/kubuntu | https://notes.kde.org/p/kubuntu-ninjas 4.10.90 WIP, most of 4.10.80 in saucy, moving ones stuck in -proposed to -release, raring: kubuntu-ppa/beta | 4.10.4 in raring-proposed === hggdh_ is now known as hggdh [00:10] harr [00:10] 4.10.90 is up already? [00:10] I'll try to do the initial upload before going to bed [00:10] yofel: Can I try doing that? [00:10] Quintasan: go ahead, do you have the bandwidth for it? [00:10] Because I have no idea how to use that script and that needs fixing [00:11] I believe I do [00:11] certainly [00:11] if you never ran it yet, see README [00:12] you need at least the ssh config, and bzr-buildpackage-ppa has to be in PATH [00:12] * yofel needs to move that to kubuntu-dev-tools one day... [00:12] Host key verification failed. [00:12] huh [00:13] for what part? [00:13] ╰─$ ./kdesc-package-names 4.10.90 > kdesc-packages-saucy.txt [00:14] wait [00:15] you shouldn't need to do that [00:15] actually, if you do that you need to remove qyoto and kimono from the result [00:16] Well, doing initial upload then [00:16] 4.10.90 is beta or alpha? [00:16] beta2 [00:17] yofel: I need to grab all the tarballs myself, right? [00:17] no, the script does that [00:18] * Quintasan does ssh-add [00:19] I'm probably going to pull an allnighter. [00:20] do that, if the script causes problems, I'll be awake for another half an hour I think [00:21] shadeslayer: You there? [00:26] manchicken: He is probably asleep, what do you need help with? [00:27] yofel: pls add libdistro-info-perl is welcome to have in README [00:27] Quintasan: He was going to get me into some of the automated tests. [00:27] oh [00:27] I can't help you with that unfortunately [00:27] * Quintasan hits shadeslayer with hammer [00:28] why don't you have devscripts installed o.O? [00:28] I do [00:28] It's not pulled it, at least not in raring [00:28] Quintasan: Yeah, it sounded like a neat thing to do. [00:29] Quintasan: I haven't helped with the project for a long time, and I thought it'd be a useful way for me to get back into it. [00:29] yofel: So generally this script does all the debuild magic [00:29] and then I just do dput *.changes? [00:29] it does recommend it [00:29] yofel: Maybe I used --no-install-recomends then [00:29] Quintasan: debsign and dput, right [00:30] you get things finished to dput in upload/ [00:30] and stuff that didn't build in manual/ [00:30] k [00:37] Quintasan: Is there something small that we need help with? I'm pretty awful at packaging, but I'm excellent with code. [00:37] I'm looking at the list that Riddell gave me, I don't see any low hanging fruit. [00:37] manchicken: I'm afraid not, at least now [00:37] I suspect that this is on account of it all having been grabbed :) [00:38] well, the 13.10 todo list is on https://trello.com/board/13-10/515d717b729ef48449001328 [00:39] I submitted a bug for CUnit a short while ago, but it appears to be a problem with the package. [00:39] yofel: Yeah, that's where I was looking. [00:39] manchicken: You might investigate out knetattach patches and ask upstream about them [00:40] Quintasan: Where are the patches, and is there a bug that they were intended to fix? [00:40] plus if we are talking about code stuff [00:40] I would like to see the PolicyKit KCM coming back [00:40] manchicken: We want to send those patches upstream [00:41] you mean you want to see a rewrite of said KCM. What I saw last was rubbish [00:41] I see [00:41] Is that the settings screen? [00:42] yeah [00:42] http://quickgit.kde.org/?p=polkit-kde-kcmodules-1.git was it I think [00:42] The Mac-style one that we re-did in 2007-2008? [00:42] yes [00:42] Heh [00:42] this is rubbish and we need something that works [00:42] even better [00:43] KDE needs something that works [00:43] Well, does it need a rewrite or just a decent refactoring? [00:43] I dropped off the project right as they were starting motion on that. [00:44] manchicken: as yofel said it's rubbish [00:44] I think it needs a rewrite [00:44] That was a big part of the UDS-Sevilla development effort at the summit. [00:44] it might be usable, not sure [00:44] Well, if memory serves me correctly, the whole infrastructure for those plugins is problematic. [00:45] That said there is a simplier task at hand if you want manchicken [00:45] bzr branch lp:~kubuntu-packagers/kubuntu-packaging/kde-runtime [00:45] grab this [00:45] my last try didn't really work (I think it just crashed), and it installs stuff where it shouldn't [00:45] go to debian/patches and submit knetattach patches upstream [00:46] yofel: Isn't that the same thing that we use for settings now? [00:46] I haven't dug too deep into it, and it wouldn't surprise me if we were using something custom. [00:46] we don't - in fact we ship nothing [00:47] yofel: What is System Settings then? [00:47] Or have I misunderstood entirely what you're talking about? [00:47] manchicken: system settings is a general UI for managing the configuration. [00:47] manchicken: ah lol, System Setting is the whole UI [00:47] KCM is like a part of it [00:47] what we talk about is one of the KDE-Configuration-Modules [00:47] Say when you go to Nepomuk or Phonon configuration [00:48] those are KCM's [00:48] ah, found what I was looking for [00:48] http://lists.kde.org/?l=kde-release-team&m=135109304619893&w=2 [00:48] that was the last people were saying about that thing [00:50] manchicken managed to keep adept working, so he's probably up for whatever when it comes to awful code. [00:50] Heh [00:50] ScottK: Is that what passes for a vote of confidence these days? [00:51] More of an introduction into the pain you've already survived for the new people. [00:52] * yofel never looked at the adept code, and suddenly feels like he should be happy about that... [00:52] Okay, that's better. I can live with that. [00:52] yofel: You have no idea :) [00:52] what what manchicken was the one doing that? [00:52] kubotu: order beer for manchicken [00:52] * kubotu gives manchicken a nice frosty mug of beer. [00:52] He didn't write it, he kept it alive. [00:53] yes, that's why he gets a beer [00:53] :P [00:53] yofel: It was like a twisted mess of scatterbrained genius, like someone started off doing C++, then switched to C, then went back and forth a bunch, then a tornado went through and mixed it all up. [00:53] Uhh [00:53] that sounds apacheloggerish [00:53] nah, just like proper enterprise design [00:54] I don't remember who wrote it originally, but it was brilliant in its madness. [00:54] Once you figured out the code it wasn't that hard to work with... but whenever you had a crash, good effin' luck finding it. [00:55] That was my first introduction to C++ and Qt... so yeah. [00:55] enterprise design you say? [00:55] Now adays I maintain billing code for telecoms which was originally written in COBOL but was programmatically converted to C... so I suppose I've made a career of swimming in filth. [00:55] * Quintasan looks for the abomination [00:56] yofel: https://github.com/Mikkeren/FizzBuzzEnterpriseEdition [00:56] urgh [00:56] Have fun [00:56] though I don't really have an accurate opinion on COBOL. All of the COBOL I know comes from TheDailyWTF :D [00:57] yofel: I've never worked with COBOL. [00:57] yofel: I have only worked with the code that was converted to C. [00:57] COBOL was fine in it's day (although the time I dropped a deck of punch cards sucked mightily) [00:57] * ScottK took it in college, but has never used it in anger. [00:58] COBOL isn't something you use in anger, it's something you use with anger. [00:58] Quintasan: <3 [01:00] lol [01:05] Quintasan: So where are these changes? Or are you saying I need to make a patch based on the changes you gave me and the current latest KDE version? [01:08] manchicken: Nah, actually its a trivial task, basically you have to talk to upstream (KDE) people, preferably the person working on knetattach and asking him if he can merge the patches in the code so we can drop them in our pacakaging [01:09] Shouldn't I have some basic understanding of what the changes are? [01:09] Seems like it might be of at least moderate importance. [01:12] manchicken: I'm assuming you will at least look at them :P [01:14] I don't know what the changes are :) [01:14] I guess I could look at the logs. [01:15] manchicken: The patches are in our packaging [01:15] bzr branch lp:~kubuntu-packagers/kubuntu-packaging/kde-runtime [01:19] Are you guys going to be on for like another hour? [01:19] I need to get 4mi in tonight. [01:19] At least. [01:21] I don't have to leave the house until 8:30 tomorrow, so I was gonna hang on for a bit tonight to see if I could actually accomplish something. [01:21] Work won't let me get anything done, so I might as well get something done here. [01:22] I'm going to bed soon [01:22] 3 in the morning here [01:22] Nice [01:22] Do you think you could email me any details you have about this - if you have any? [01:23] and is this just the kubuntu_nodisplay_knetattach.diff patch? [01:32] manchicken: kubuntu_nodisplay_knetattach.diff kubuntu_oxygenify_knetattach_icon.diff [01:33] So those are pretty tiny patches. [01:33] I'll see if I can figure out how to submit those. [01:33] It looks like I made it to the Disabled Developers group on kde.org. [01:33] Why folder-new and not knetattach for the icon? [01:33] manchicken: Yes, this task is pretty trivial. Honestly I thought you are one of the new people so I didn't want to throw major stuff right off the bat :P [01:34] Just an icon change? [01:34] Quintasan: Well, I need to get all of my accounts fired back up. [01:34] This will help :) [01:35] I'd rather have you do the policykit kcm, I would do that but I suck at coding and uni having me attend java lectures does not help Qt development at all [01:37] aw damn [01:37] I really need sleep [01:37] yofel: I threw the upload, there are 4 broken packages [01:37] will sort that out in the morning [01:44] Quintasan: I can do this patch stuff and then I will look at the KCM stuff. [01:45] Does anybody else have the rationale for this change in their brain? [01:50] argh [01:51] not brain [01:51] * Quintasan hits the bed [01:51] good night [02:22] manchicken: Still around? [02:23] http://qa.ubuntuwire.com/ftbfs/#kubuntu presents some interesting opportunities. [02:23] kdeartwork is just packaging. I'm fixing that. [02:26] I suspect https://launchpadlibrarian.net/143047036/buildlog_ubuntu-saucy-armhf.calligra_1%3A2.6.92-0ubuntu3_FAILEDTOBUILD.txt.gz is somehow related to qreal not being the same as double on arm, but no idea beyond that. [02:37] ScottK: I am now. [02:37] Just finished my run [02:37] K [02:38] If you want something to roll up your sleeves with, have a look at that build failure log. It, of course, works on everything but arm. [02:39] That and the things that fail are called double something. [02:39] Makes me think something is really a qreal and the whole thing needs to be qreal. [02:40] But I know zip at Qt/C++ so once it's more complicated than qreal(2.0), I'm done. [02:40] I have arm boxes we can test build it if you get an idea. [02:42] Be back later. [02:49] I'll look in a minute. [02:51] I just spent ten minutes trying to figure out why a machine wasn't responding on the network just to discover it had gone to sleep. [03:12] ScottK: Howdy [03:14] ScottK: You got the code for /build/buildd/calligra-2.6.92/krita/image/kis_filter_weights_applicator.h ? [03:15] pull-lp-source calligra [03:15] install ubuntu-dev-tools if you don't have it [03:23] ScottK, Riddell - link with outline to be filled in with release notes is: https://wiki.ubuntu.com/SaucySalamander/Alpha1/Kubuntu [03:26] valorie: I hardly think making one package with the help of like 2 people makes me a packager lol [03:26] Riddell: I see error message [03:41] Back in a minute... I have to boot into Windows to upload my run data. I'll be back in a minute. [05:04] This is weird... if I have fr in my language variable, but en_US sorted first, some programs - particularly on the command-line - pick fr first. [05:09] ScottK: Your problem is that you're implicitly coercing a double to a qint32. [05:15] ScottK: I'm surprised that this only complains on arm. [05:16] It might have given a warning elsewhere. [05:16] Shoot me a diff and I'll try to build it. [05:17] manchicken: ^^^ [05:17] Well, there are two problems... [05:17] OK [05:18] 1, you've got potentially 8-bytes you're trying to coerce into a fixed 4-bytes. 2, you've potentially got precision in your 8 bytes that you'll lose in the process. [05:20] Scratch that, for #2 you have known precision that you're losing. [05:20] You will lose precision. [05:20] This is the line it's complaining about: return pixel_l + qreal(0.5); [05:21] OK [05:21] You totally lose precision there, picking it implicitly. [05:21] I think you could fix this like this: return KisFixedPoint(pixel_l + qreal(0.5)); [05:22] OK. [05:22] Shall we try it? [05:23] Err, sorry [05:23] return pixel_l + KisFixedPoint(qreal(0.5)); [05:23] Should I try to build it with that line changed? [05:23] That should do it. [05:23] OK [05:23] Try that. [05:24] Will do. [05:24] I'd do it, but you've already got it building :) [05:24] You don't have arm hardware, do you? [05:24] Something similar on lines 301 and 305. [05:24] Remember the part where I don't do C++ or Qt? [05:25] Naw, I don't. [05:25] I need copy/paste, not similar ... [05:25] Heh [05:25] Not kidding. [05:25] I can package the shit out of stuff, but coding in anything but python is a no go. [05:26] Do you have any experience with OpenGL programming? [05:26] Pastebining. [05:26] Thanks. [05:26] No, I haven't done much (any) of that. [05:26] Interested in learning? [05:27] Most, if not all, of the remaining armhf build failures are due to the fact that we have only GLES on arm and not GL. [05:27] And those packages make direct GL calls, which then fail. [05:28] They could use some kind of porting so they could work with either, but that exhausts my knowledge. [05:28] I could, but I'd need someone to bug a lot. [05:29] which packages are you talking about? [05:29] Not tonight, anyway, it's getting late for me. I'm going to kick off this build and then sleep. [05:29] mgraesslin: I'll get a list. [05:30] * mgraesslin thinks that most devs using OpenGL have no idea what OpenGL ES doesn't allow [05:30] http://pastebin.ubuntu.com/5803733/ [05:31] ScottK: That's my best stab at it... it's been a while since I've played in C++ land... I do mostly C these days at work, so I'm a bit rusty. [05:31] You can find build logs linked on http://qa.ubuntuwire.com/ftbfs/#kubuntu for digikam, kalzium, ksudoko and kubrik [05:31] manchicken: We'll try it. Welcome back. [05:31] mgraesslin: I may be off base in my diagnosis of the problem too. I'm out of my depth with this. [05:32] ScottK: Well it's not like I get to do neat stuff at work... [05:32] ScottK: When I was consulting I got to do neat stuff... this job I've had for a year just doesn't let me do squat. [05:32] Even more reason to hang around with us. [05:33] just looked into it for Kubrik [05:33] ScottK: Well, that, and the free beer. [05:33] seems like there are conflicting typedefs [05:33] ScottK: Free as in beer means free beer, right? [05:33] sure. [05:33] digikam is easy: they use OpenGL 1 [05:34] glBegin... [05:34] Easy for you ... [05:34] mgraesslin: Are the arms little endian like Intel? [05:34] easy to recognize [05:34] not easy to fix [05:34] I haven't programmed for arm outside of Obj-C for iOS or Java for Android. [05:34] in KWin we have that code in an ifdef KWIN_BUILD_OPENGL1 [05:34] (and Perl for android) [05:35] manchicken: AFAIK they are little [05:35] I wish they were big, as we have changes in KWin which might broke on big endian [05:35] Do you know what the word size is? [05:35] mgraesslin: You can always get an Itanium box, right? heh [05:36] well I tried to setup a VM with powerpc [05:36] Even Debian is about to drop ia64 support. [05:36] but so far failed to get non-x86 VMs running [05:36] ScottK: Really? Don't they still support Amiga? [05:37] need to invest more time into it - yesterdays commit by ScottK highlighted that it's not a solution to not build on arm [05:37] mgraesslin: No joke, especially with cheap chromebooks. [05:37] manchicken: google says arm has a word size of 4 byte [05:37] mgraesslin: Maybe I'll sell this Galaxy S3 I have and buy a chromebook. [05:37] Those qreal/double ones are about as complex as I can handle. [05:37] mgraesslin: Well that's good, at least it's not that different as far as nuts and bolts go then. [05:38] ScottK: You're lucky, I had a piece of crap bug like that at work today so it's fresh in my brain. [05:38] ;-) [05:38] Though the problem at work had to do with some moron casting a 16-byte long double to a 4-byte integer, losing precision, and flipping the sign bit in some cases to boot. [05:39] Fun. [05:39] We were using that for sales tax calculation for California... so yeah... I get to re-calculate three years worth of sales taxes I'm pretty sure. [05:39] I just enabled C++11 yesterday and I know we still have some areas which won't build on arm now [05:39] Nice. [05:39] For mobile phone purchases for the #3 mobile carrier in the United States... [05:40] That'll be a fun one to explain to management. [05:40] mgraesslin: I wonder if I could get this stuff to build on a N7. [05:40] mgraesslin: Do the chromebooks play nice like most normal intels, (U)EFI boots and all that? [05:41] I haven't seen one yet [05:41] Which arm do you work with? [05:41] Pandaboard === jalcine_ is now known as jalcine [05:42] * ScottK has Efika MX with ancient crap vendor kernels that now require a specialized eglibc build to work. [05:42] I've never heard of that one. [05:42] oh and a raspberrypi [05:42] * mgraesslin got KWin rendering one frame on it [05:42] NICE! [05:43] then it crashes [05:43] manchicken: Don't worry about them, you can't get them anymore. [05:43] should try again some time - with Wayland it should work [05:43] Speaking of which, did you see Riddell's blog post? [05:43] Me? I didn't know he had a blog. [05:44] yes [05:44] It's on both planet Ubuntu and planet KDE. [05:44] That was directed at Martin though. [05:45] ScottK: will you be at Akademy/Kubuntu developer summit? [05:45] No. [05:45] I live on the wrong side of the Atlantic/can't afford the time off. [05:45] yeah - always a problem [05:46] Nice [05:46] I could probably get work to consent to letting me go... but I doubt I could afford it. [05:47] When I did UDS in 2007 it was on Shuttleworth's dime. [05:47] And I did well to thank him for it by having kids and disappearing :) [05:47] good morning [05:47] Good morning. [05:48] I think it's time for me to say good night. [05:48] I've gotta do the whole job thing tomorrow. [05:48] Hopefully something interesting happens. [05:48] Good night. Thanks for the suggestions. [05:49] No problem. I'll see you around. [05:49] I'll crash too as soon as I get this build kicked off. [05:49] it's close to 11 in your part of the world? [05:49] No, 2am. [05:49] I've got 0049. [05:49] that's worse [05:49] I'm UTC-0500 [05:50] Later all. [05:50] * ScottK should have gone to sleep two hours ago, but I suck at going to sleep when I should. [05:50] I'm -0400 [05:50] good night [05:50] good night [06:09] Not sure how connected I was a minute ago. [06:09] Quintasan: Please update the kdeartwork packaging from bzr. [06:09] Sorry if that's a dupe. [06:13] You just made my day thank you thank you thank you! http://blogs.kde.org/2013/06/26/kubuntu-wont-be-switching-mir-or-xmir [06:15] Riddell: ^^^ [06:22] Hey, I noticed a bug on Kubuntu.... Ubuntu WONTFIX it because Mark Shuttleworth thinks it's fixed... Maybe it should be assigned to Kubuntu :D https://bugs.launchpad.net/ubuntu/+bug/1 [06:22] Launchpad bug 1 in Ubuntu Malaysia LoCo Team "Microsoft has a majority market share" [Critical,In progress] [06:29] palasso: It's all how you define the market. If you include smartphones and tablets, they don't. [06:30] which means the bug is nowadays invalid but not fixed as the market was clearly defined [06:31] Yes that's true and basically that was the explanation Mark gave but I personally believe that there were different motivations for the answer [06:31] but well, nothing I should care about... [06:32] 1. Unity definitely is alienating to ppl who are used to classic windows, it definitely doesn't have the purpose to convert windows users to linux 2. MSFT is now a big customer of Canonical in the Cloud market [06:33] I understand what they want - one interface on everything [06:33] and it sort of makes sense [06:34] it's not fully free, though [06:34] and they shouldn't pretend that it is [06:35] ^ +1 [06:35] I feel the same way I feel about android -- somewhat free is better than totally locked down, but I'm not interested in helping [06:36] Which is all rather off topic for Kubuntu devel anyway. [06:37] sorry, ScottK [06:38] Riddell: kdeartwork built on all archs now. [06:38] calligra test build in progress, but it won't complete for hours, so time for sleep. [06:39] sweet dreams [06:47] Hey guys, how are we at Alpha 1 testing? [06:55] smartboyhw: I'm going to test the 64bit image this afternoon/evening. [06:55] lordievader, great. === smartboyhw changed the topic of #kubuntu-devel to: Kubuntu - Friendly Computing | https://trello.com/kubuntu | https://notes.kde.org/p/kubuntu-ninjas 4.10.90 WIP, most of 4.10.80 in saucy, moving ones stuck in -proposed to -release, raring: kubuntu-ppa/beta | 4.10.4 in raring-proposed | Please test the Alpha 1 images in http://iso.qa.ubuntu.com/qatracker/milestones/297/builds [06:56] 4.10.90 already? Oh no it's packaging series again. [06:59] Fixing kdepim-runtime now. [07:07] Actually laters, it's difficult to build without nepomuk built. [07:09] Riddell, ScottK shouldn't http://iso.qa.ubuntu.com/qatracker/milestones/297/builds/47470/testcases/1450/results be an optional testcase instead of run-once? [07:11] smartboyhw, should now be less work with beta2 ? [07:12] as lot was fixed in beta1 ? [07:12] soee, well the .install files are still a mess:) [08:21] bah [08:22] shadeslayer: good morning ;) [08:22] morning [08:22] more like afternoon here :) [08:41] good $TIME_OF_DAY [08:47] yofel, working on beta2 for raring ? :) [08:48] it's not even done for saucy yet, though much less work than beta1 it seems [08:48] still, earliest I see beta2 for raring would be tomorrrow evening [08:50] thank you [09:24] \o [09:46] morning [09:48] Hello Riddell. [09:48] Getting on with i386 testing of Alpha 1. [09:50] thanks [09:51] Riddell, well, I'm expecting lordievader to do the amd64 work. [09:52] smartboyhw has a minon! [09:52] Riddell, he said he WOULD do it. [09:54] he doesn't usually appear for another few hours [09:54] Riddell, http://iso.qa.ubuntu.com/qatracker/milestones/297/builds/47470/testcases/1450/results should it be removed or marked optional!? [09:55] smartboyhw: that should be kept [09:55] why would we remove netbook? [09:55] Riddell, anyone have a netbook to test [09:55] ? [09:55] I do [09:56] Riddell, OK then, please test:P [09:56] oy! don't you got minioning me! [09:56] oh ok, I'll test it for you [09:56] Riddell, I REALLY DON"T HAVE A NETBOOK. I sincerely thank you:) [10:04] Riddell, please update qa.kubuntu.org.uk for 4.10.90.:P [10:05] smartboyhw: Try bossing apachelogger around [10:05] Doesn't work [10:05] HOLY CRAP [10:05] Quintasan, what doesn't work? [10:06] Riddell: do lines virtual KDE_DEPRECATED QList *customContextMenuActions( ChatSession *manager ); prevent API breakage? [10:06] smartboyhw: More tests is better right? So go a head test amd64 ;) [10:07] oh [10:07] Quintasan: :( [10:07] apachelogger: ^^ [10:07] Quintasan: yes, rather than just removing that method [10:08] Well, somebody competent needs to look at this patch [10:08] Hiyas all [10:08] smartboyhw: appearing now in http://qa.kubuntu.co.uk/ninjas-status/ [10:08] Some of those methods are declared as deprecated and some are not and they are not even there as far as I can tell with kopete [10:08] Riddell, yay. [10:09]