[00:00] and I'm too bleary eyed to sort [00:00] sorry jbicha I modified the existing changelog instead of making a new entry [00:00] I think thats good now [00:01] its gonna get very messy in here [00:02] why is CI noticing me with sarcastic messages? [00:03] time for bed [00:03] Ive cleared the queue [00:03] should shut it up [00:04] if you have worked on the package in question you get a lovely PM [00:04] I saw [00:04] I get tons of them [00:04] Oh no! You're suspected of having broken abc [00:06] ok. night then. I'll come back to those akonadi packages in the morning unless you/simon sorts them in the meantime [00:06] Im heading to bed myself [00:07] I'll wake up to 50+ merge proposals then [00:07] LOL [00:08] yofel: will need to fix KCI [00:08] something wrong with ruby [01:06] @Sick_Rimmit I reverted your change also branched out 16.10 and 16.04 out [02:28] ahoneybun: awesome! [02:30] they have kinda different readme's [02:30] mostly about LTS don't get much new features just fixes [02:30] 16.10 is current master so all changes go to master [02:31] after 16.10 is out I'll merge master to it [02:33] never seen ABORTED before lol [02:36] you rock, m'dear [02:36] mm trying to cherry pick ricks change about activites [02:36] might have to do it byhand [02:36] oh well I'll do it [02:38] sometimes valorie [02:46] and we have an Activities section thanks to Rick [07:06] as kdepimlibs-data is not going to exist an more, then need to sort out what did depend on it http://paste.ubuntu.com/21863977/ [07:10] there is a reverse-depends command somewhere [07:10] or maybe that was kubottu [07:10] rip [07:22] valorie: see the paste above [07:23] oh. on the channel? not a clue [07:26] oh, you know so much more than me [07:26] lol [07:38] urgh, the ruby version bump got pulled and deployed by accident [07:38] I'll fix that over the day [07:42] !testers | 14.04.5 images are up http://iso.qa.ubuntu.com/qatracker/milestones/365/builds [07:48] yofel: I'll see what I can do after I'm sure Lubuntu is good to go. :) === pavlushka_ is now known as Guest82875 [07:50] ummm. so many other packages and build pulled in kdelibs-data, that this is going to be huge fun to unpick [07:51] *kdepimlibs-data [07:51] For now I'm going to try to go back to bed === Guest82875 is now known as pavlushka [07:51] I can help later acheronuk [07:51] thanks. :) [08:07] for example, libkf5akonadiagentbase5 Depends: kdepimlibs-data (>= 4:15.12~) [08:08] but which parts of the old kdepimlibs-data did that actually need? [08:09] and where are we sending those parts now? [08:10] /o/ [08:12] acheronuk: looks to me like it's mostly runtime parts from akonadi-*, so... good questino [08:12] *question [08:13] you could check against what binaries the other akonadi libs link against, and do the appropriate data dep from that [08:17] need to have that corrected, as now kdepimlibs package was deleted from the CI, even some of the things were trying to get build yesterday fail to build as they are yanking in kdepimlibs 15.12 from the archive via akonadi, which then conflict with files etc [08:17] acheronuk: also: is that a direct dep in control or an shlibs dep from a symbol file? [08:18] Package: libkf5akonadiagentbase5 [08:18] Architecture: any [08:18] Multi-Arch: same [08:18] Depends: kdepimlibs-data (>= 4:15.12~), ${misc:Depends}, ${shlibs:Depends} [08:18] Description: Akonadi agent base library [08:18] so direct in control from that [08:18] was that lib in kdepimlibs before? [08:19] that's from akonadi [08:20] :/ [08:22] I know. It's one of those things that goes round in circles with the deps, and the key is probably working out the point where you can put a critical change in to break that chain [08:33] for example, akonadi-contacts now fails due to [08:34] Unpacking libkf5akonadimime5:amd64 (4:16.04.3+p16.10+git20160801.2343-0) ... [08:34] dpkg: error processing archive /var/cache/apt/archives/libkf5akonadimime5_4%3a16.04.3+p16.10+git20160801.2343-0_amd64.deb (--unpack): [08:34] trying to overwrite '/usr/share/config.kcfg/specialmailcollections.kcfg', which is also in package kdepimlibs-data 4:15.12.3-0ubuntu1 [08:35] where kdepimlibs-data 15.12.3-0ubuntu1 is getting pulled in by some of the other deps via akonaki itself [08:37] 2 things [08:37] a) /usr/share should not be in libkf5akonadimime5 - please make a data package for that [08:37] b) that data package needs to break/replace kdepimlibs5-data << 4:16.08~ [08:38] or some version like that [08:40] yes, I was coming to the conclusion that even if it's just one file or 2 to go in them, each of those packages from the split of kdepim-libs will need a data package [08:41] which will then ease sorting out the replacement depends on kdepimlibs-data that is found elsewhere [08:43] going to take a bit of sorting out [08:43] ok. back in a bit [09:16] any cmake experts online? [09:37] doko: not a expert but can maybe help? [09:38] shadeslayer, ta, already found it, just a bad regexp [09:38] regexes are evil [09:38] #getridoftheregexes [10:06] slightly OT: qtwebengine not yet packaged for Ubuntu? [10:07] kfunk: there are builds in kubuntu-ci/unstable [10:08] still WIP and soon pending for debian [10:08] hummm is telegram bridge down? [10:11] yofel: got it. needs 5.6.1. too bad. [10:11] One sided maybe.. [10:11] acheronuk: are the reverse dependancies direct or 2nd + generation? [10:12] kfunk: what release? [10:12] yofel: 5.5 (distro Qt) [10:13] yakkety has it in proposed, xenial.. we'll take a bit longer [10:13] (5.6.1) [10:13] no worries; will have to compile it myself for my dev setup [10:13] k [10:13] clivejo: not sure. only got as far as the akonadi one so far, which is direct [10:18] back for now [10:18] * tsimonq2 reads backlog [10:24] WWYD [10:26] have you made a start on creating a akonadi-mime-data package? [10:26] no, should I? [10:26] clivejo: yes [10:26] sorry was talking to acheronuk [10:27] make is break kdepimlibs [10:27] I did [10:27] and make a depends on it in libkf5akonadimime5 [10:27] I did [10:28] LOL just rubber ducking [10:29] yofel: regarding the name of data file, should it be libkf5akonadimime-data or akonadi-mime-data? [10:30] I did libkf5akonadimime-data [10:30] I noticed, just checking with yofel [10:32] ok [10:32] I still have no idea how PIM works [10:33] confuses the hell outta me everytime [10:33] Not so much here either... [10:33] hopefully by disecting it, I might actually learn now it goes together! [10:34] in my head the package name akonadi-mime-data makes more sense [10:35] but dont ask me why [10:35] * tsimonq2 pokes his head in and waves [10:35] o/ [10:35] how can I help? :) [10:35] a package name, should it be libkf5akonadimime-data or akonadi-mime-data? [10:36] I am torn between them now... [10:39] CI just built it with libkf5akonadimime-data, for now anyway.. [10:40] imho akonadi-mime-data [10:43] would certainly scan easier in all the depends that need changing [10:43] * clivejo wonders has yofel become trapped under a pile of ruby [10:46] I think so lol [10:48] artikulate 4:16.04.2-1 is marked for autoremoval from testing on 2016-09-01 [10:48] It (build-)depends on packages with these RC bugs: [10:48] 806102: qt-gstreamer: FTBFS when built with dpkg-buildpackage -A (No such file or directory) [10:48] ruh roh [10:48] yofel, clivejo ^ [10:59] If there is only one lib. Libfoo-data, otherwise source-data [11:00] ooh he's on Telegram, he IS trapped under a pile of ruby :P [11:00] tsimonq2: that is debian, not ubuntu [11:00] maybe hes practive rugby [11:00] stays as it is then [11:01] practising [11:01] @yo [11:01] argh [11:01] Yes that's why I had enouh rubies to throw at rohan yesterday [11:01] @yofel well doesn't removal in Debian sometimes mean removal in Ubuntu? [11:02] Yes, if auto-symced. Artikulate isn't [11:02] !info artikulate yakkety [11:02] artikulate (source: artikulate): Language learning application. In component universe, is extra. Version 4:15.12.3-0ubuntu1 (yakkety), package size 2397 kB, installed size 3842 kB [11:02] !info artikulate unstable [11:02] artikulate (source: artikulate): Language learning application. In component main, is extra. Version 4:16.04.2-1 (unstable), package size 2501 kB, installed size 4027 kB [11:03] oh ha that makes sense [11:04] yay! new profile pic! [11:05] re [11:11] * clivejo giggles [11:11] https://www.youtube.com/watch?v=mlaVHxUSiNk [11:13] clivejo: cool! :D [11:19] I vaguely remember that [11:25] akonadi-contacts will need a data package. tsimonq2 ? [11:34] acheronuk: need me to do it? [11:35] I just thought you might like to [11:35] acheronuk: and I've never made a data package before, so can you walk me through? (if you haven't done it already) [11:37] I used the old kdepimlibs-data one as a starting point https://git.launchpad.net/~kubuntu-packagers/kubuntu-packaging/+git/kdepimlibs/tree/debian/control?h=kubuntu_unstable [11:40] acheronuk: so this still needs to be done? [11:41] no-one else has AFAIK [11:41] k [11:46] it built for xenial last night, and log should show the left over files [11:46] I think I figured it out :P [11:47] YY build probably needs a retry later [11:49] acheronuk: both are failing... [11:50] in your sbuild? [11:50] acheronuk: Yakkety is failing but KCI shows Xenial as failing [11:51] (although it's passing, which is weirf) [11:51] *weird [11:51] https://launchpad.net/~kubuntu-ci/+archive/ubuntu/unstable/+packages?field.name_filter=akonadi-cont&field.status_filter=published&field.series_filter= [11:51] * tsimonq2 nods [11:52] akonadi-contacts - 4:16.04.3+p16.10+git20160801.2207-0 (changes file) kubuntu-ci-bot 12 hours ago Published Yakkety Libs There were build failures. amd64 [11:53] KCI's criteria for a failure and LP's don't have to be the same. [11:53] ok [11:54] Hiyas all [11:54] o/ BluesKaj, how are you? [11:55] even though LP built it, all those left over files are a fail for KCI [11:56] Hi tsimonq2, fine here, and you? ...hey acheronuk [11:57] great BluesKaj :) [11:57] Hi BluesKaj :) [11:58] argh now it can't find libkf5akonadimime-dev [11:58] how goes the battle with xenial ? I had to revert to 14.04 for stability reasons [12:00] one unstable OS is bad enough and I expect that since I'm testing Yakkety, but Xenial seems very slow to stablize [12:02] tsimonq2: you may need to wait for the latest akonaki-mime to publish before you build again [12:03] argh ffs fine [12:05] actually, akonadi itself may need fixing [12:05] you sort out one thing, and it just exposes another layer of problems! [12:07] tsimonq2: you may have to wait after all on doing that [12:16] acheronuk: wait for what? [12:23] acheronuk: what am I waiting for? [12:24] (1) akonadi-mime to publish (2) maybe a fix on main akonadi package itself [12:28] some build deps still result in the old kdepimlibs-data package being wanted, which is gone. or at least they did earlier this morning === ghostcube__ is now known as ghostcube [13:00] I can see why people steer clear of PIM a bit [13:02] Tis very true [13:02] acheronuk, home users like me usually have no need for the PIM apps [13:02] Its a can of worms [13:03] It's a can of piranha [13:03] BlueKaj there are a lot of people who do use it and that's why we package it. [13:04] ahhh Mr slhk [13:04] hi clivejo [13:05] who needs db for mail and contacts unless you must save 10K emails etc [13:05] slhk: any input on this thread - https://www.kubuntuforums.net/showthread.php?70569-Should-I-be-concerned [13:05] unless you're on the job clivejo [13:06] I know a lot of home users who use kmail, but your entitled to your opinion [13:07] there's finally an option to dump the PIM apps if one doesn't need them without removing the desktop [13:07] I loved kmail until KDE4 arrived [13:10] kmail was a terrific email client before that [13:18] clivejo: that's a different issue from the one I fixed, and as pointed out in the thread it has been fixed before. Anyway, I confirm that update-a-x-i-dbus still does not work, and I'm fixing that right now === mgolden is now known as mgolden_ [13:27] * acheronuk wonders if LP can be bribed to speed up === mhall119_ is now known as mhall119 [13:29] no ^^ [13:30] well yes, but that's impolite [13:30] acheronuk: na we just rant in #launchpad :P [13:30] :O [13:37] tsimonq2: not too often [13:38] :P [13:41] ok. if rebuilding akonadi doesn't break this circular dependency, I'm going to be sad.... [14:27] You aren't past the sad stage? [14:27] Akonadi tends to make me angry [15:03] clivejo: There's still plenty of time [15:22] got bored of waiting and built akonadi in my testing ppa. seems to allow akonadi-contacts to build again. now wait for CI.... [16:05] yofel / others: do you think you could ask on #ubuntu-release to ignore failures for: kwin marble kde-cli-tools libkscreen kdepim-runtime kxmlgui extra-cmake-modules kconfigwidgets okteta akonadi-search kidentitymanagement kdelibs4support kwayland libkscreen ? we're landing the unity8 fix soon after which the remaining blockers to release pocket migration would be those listed autopkgtest [16:05] or if there are some you are worried about showing a real blocker and want to fix instead [16:08] 100% tests passed, 0 tests failed out of 5 [16:08] Total Test time (real) = 2.36 sec [16:08] XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":99" [16:08] after 9428 requests (9428 known processed) with 0 events remaining. [16:08] wtf?!? [16:09] oh wait, acc is what failed, urgh [16:10] I'm somewhat scared the GCC6 will come and suddenly block everything for half eternity :) [16:16] I'm wondering whether I should make a script that removes the autopkgtest setups from all kde packages [16:17] Mirv: is there anything else stuck except qtbase? [16:19] Mirv: and are those all that you need? There are more kde packages in proposed than those [16:26] yofel: most autopkgtests pass also for kde packages, from my manual viewing through http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html I came up with that list [16:26] yofel: well ~everything is stuck because ~everything has reverse dependencies that are in that list ^ and depend on each other. [16:27] generally http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#qtbase-opensource-src and #qtdeclarative-opensource-src show most problems [16:28] Mirv: yes, I'm looking at that, but I'm also seeing nonsense build combinations like: [16:28] autopkgtest for kwin/4:5.6.4-0ubuntu1: amd64: Pass, armhf: Pass, s390x: Always failed [16:28] autopkgtest for kwin/4:5.6.4-0ubuntu2~1: i386: Pass [16:28] autopkgtest for kwin/4:5.7.2-0ubuntu1: ppc64el: Regression ♻ [16:29] we don't migrate packages on an architecture basis, so what's the point in that? [16:33] yofel: it's probably some autopkgtest timing issue that would require stronger hinting from release team if wanted to be executed with the correct version for all archs [16:34] right, but I honestly have no motivation to run after the release team to fix autopkgtests in up to 200 packages. so lets go with force-badtest for everything's broken so that you're not stuck on us [16:37] yofel: yes I agree that if you think functionally everything is ok, ask release team to force-badtest that list of packages. plus: kde-baseapps (I just found by going through the page with grep + sed instead of manually) [16:37] yofel: http://pastebin.ubuntu.com/21910811/ [16:38] well that's manually stripped down of non-Qt/KDE packages [16:38] ah, nice one [16:38] yeah, looks about right from what I can see [16:43] interesting that "proposed migration" thing. it's like a mixture of uploading packages to debian experimental and a debian transition [16:43] santa_: the proposed migration is literally britney [16:44] you mean it's done with britney [16:44] righ [16:44] t [16:44] interesting [16:44] regarding the tests [16:44] well, back in the days we were actively disabling them for kde software [16:45] today we have some effort towards making them work, but I have the impression it's too much effort for little benefit [16:46] i.e. how many real bugs have people fixed thanks to the autotests being executed in packaging? [16:48] vs how many hours you have to spend ... not to fix the software, but to fix the test itself? (i.e. fixing the environment where the test is executed and things like that) [16:49] that being said, I would clearly ignore them wrt blocking other stuff [17:03] bbl [18:24] thanks everyone! :D [18:24] * tsimonq2 is a ninja now [18:24] right, welcome ;) [18:24] now you can do a proper ninja dance [18:25] lol [18:25] tsimonq2: so now you can: commit directly, upload to kubuntu-ppa/* (please don't directly upload to /ppa and /backports, they have -landing equivalents for that), and you can trigger CI builds [18:26] awesome, thank you [18:26] just be careful and ask if in any doubt [18:26] keep using merge requests for changes that you're doubtful about though please [18:26] there are no silly questions here [18:26] yofel: I was just about to comment on the fact that I will do that :) [18:26] :) [18:29] thanks yofel for bringing up the topic on release channel, let's hope it'll be agreed upon! infinity may still ask about this and that failing and what's going to be done about it.. [18:30] well, he always does that ^^ [18:30] yofel: would it be fine with you if I rebuild a package in KCI just so I know how it's done? or would you rather I wait until there's a real need for it? I mean, afair, it won't hurt, right? [18:32] congratulations tsimonq2 :D [18:32] thanks acheronuk :D [18:32] theres still a queue on KCI [18:32] yofel: if he asks about marble symbol updates, do note we could handle it but there's some other FTBFS issue bug #1609083 [18:32] bug 1609083 in marble (Ubuntu) "marble needs updated symbols for Qt 5.6" [Undecided,New] https://launchpad.net/bugs/1609083 [18:33] still waiting for it to build akonadi-contacts [18:33] clivejo: needs akonadi itself to build first [18:34] it has helped before though to at least have a bug filed about an issue [18:34] anyone work on akonadi-notes ? [18:35] * tsimonq2 checks if I did anything [18:35] now I can push so I don't need to poke anyone if I make a trivial change :) === ghostcube__ is now known as ghostcube [18:37] clivejo: no clue if my changes are still relevant but here: https://git.launchpad.net/~tsimonq2/kubuntu-packaging/+git/kubuntu-akonadi-notes?h=kubuntu_unstable [18:37] clivejo: akonadi-notes looked that it at least build ok, though can't say I inspected the result overly much as a was trying to sort contacts [18:37] * tsimonq2 checks if they are still relevant [18:38] just be careful [18:38] no actually, I'll push my changes if you don't mind [18:39] wait no I was looking at master [18:39] nvm [18:39] I imagine those 3 packages still needs some 'housekeeping' to get them up to scratch, even it they do technically build [18:39] trivial changes might not appear so trivial in the grand scheme of things [18:39] clivejo: I understand :) [18:41] I'm always very hesitant at 'git push' still, which is no bad thing really [18:42] and remember the Kubuntu team moto - WWYD [18:43] lol [18:45] hehehehehe [18:46] tsimonq2: I did that data package in the end, as it was tied up with akonadi deps, and wanted to sort them both together [18:46] * yofel throws even more rubies at clivejo [18:46] acheronuk: good idea [18:46] seriously, ever since you made that anology, I don't know if people here mean the original acronym or yours whenever I read it >.< [18:48] I think 90%+ of the time, Y=Yofel now [18:51] tsimonq2: if you just trigger builds, no, I don't mind. But maybe take something that doesn't have many downstream projects (otherwise it'll block those if they're in the queue) [18:52] it means What Would Yofel Do :P [18:52] yofel: have an example I can do a build of? [18:52] tsimonq2: http://kci.pangea.pub/job/yakkety_unstable_sweeper/ [18:53] make sure the checkbox for kubuntu-ninjas is checked when you log in [18:54] I already logged in, but thanks anyways :) [18:54] woah this is cool [18:54] hm [18:54] why are the daily promotion jobs still a thing [18:54] * yofel goes killing [18:57] I think we need to fire soee [18:57] https://www.kde.org/announcements/plasma-5.7.3.php [18:58] oh sorry i spammed Jonathan some hours ago :D [18:59] but please lets make first 5.7 release for Xenial :) [19:00] that ^ [19:00] and yakkety_archive is frozen until the packages make it out of -proposed [19:00] we really need to get that done first [19:04] ahoneybun: Hiya you about [19:05] o/ sick_rimmit [19:05] \o/ tsimonq2 [19:05] Yep [19:05] Perfect [19:06] I've made those changes from you review [19:06] Would you like me to push them to your Master, a seperate Branch or something else ? [19:09] sick_rimmit: I have the master as the current release cycle ie 16.10 [19:09] 16.04 has it's own branch [19:09] kubuntu-16.04-LTS [19:09] Oh OK, I didn't see those branches [19:10] I added the activies part to master [19:10] I made that yesterday [19:10] I'll check you branches again, and see what's available [19:10] sorry about that [19:10] just messing with branches now [19:10] need to figure out how to deal with docs.k.org now [19:11] I think it'll take me a little time get all the work done, and probably good for me to target 16.10 for Oct [19:11] since it points to the LTS currently [19:11] I'm thinking of adding activies to LTS as well [19:12] Oh yes that's a good idea, I think it needs new images for Activities in 16.04 [19:12] well your images are of 16.04 currently [19:12] but you might have the backport ppa as well [19:12] OK here's my suggestion [19:13] I know sphinx can handle dirs so make lts and release dirs [19:13] I'll push what I have to my Master, and issue a pull request. If you review and are happy with the work so far, Merge and then I will pull [19:13] you current Master, [19:13] you'll fetch and sync [19:13] if your working on 16.10 then yes [19:14] Yep, that's good [19:14] my plan is that master will always be the current dev cycle [19:14] then when it is release make a branch for it [19:15] Tags ? [19:15] I've not messed with tags yet [19:15] I know we use branches on our git [19:15] No worries, however you want to do it, I'll work with that [19:15] :-D [19:16] the LTS needs a different branch [19:16] at least [19:16] what are you guys doing/talking about? [19:16] we can tag release [19:16] *releases [19:16] clivejo: the kubuntu manual [19:16] https://github.com/ahoneybun/kubuntu-manual [19:16] Yep agreed, you create that I'll pull it, I can work on those images for it then [19:16] http://docs.kubuntu.org/ [19:18] sick_rimmit: I'd rather have the packaging stuff in a dir under contribute somehow [19:18] not outside of docs [19:18] makes it easier to find everything [19:19] clivejo: is anything now making libkf5akonadisocialutils-dev ? [19:19] not that Ive found [19:19] not that I can either :/ [19:20] so presumably things that did need it don't now? [19:20] you would hope [19:24] wow, lots of missing symbols in okteta [19:26] yes, kdepim-runtime for example has had that social plugin removed from CMakeLists.txt [19:26] can you remove the build dep from the control file too [19:26] clivejo: how can I help? :) [19:26] I was just about to [19:27] sick_rimmit: can you get me some install screenshots from 16.04? all the slides and sets [19:27]