[02:14] phonon-backend-vlc 0.8.2 build fails. pbuilder can't seem to find libvlc-dev and libvlccore-dev. Logged into pbuilder and tried to install the dependencies, but they're not present. Any suggestions? [03:22] rbetzen: make sure universe in apt sources.list [03:25] sgclark: hello again! I'll check. I had it in my head that it was set up when pbuilder created the environment. That's what I get for assuming. Duh. ;) [03:26] rbetzen: yeah I had to add http://paste.ubuntu.com/9510369/ to my .pbuilderrc file to get dpes to install properly, keep in mind mine is for vivid so adjust accordingly [03:33] Ok. I'm going to try again using your settings. I'll let you know how it goes. Thanks a bunch! === rx_t is now known as rx [03:52] sgclark: that took care of my dependency issues. thanks for the help. wading through another issue now... [03:57] rbetzen: np :) good luck! === wgrant_ is now known as wgrant [13:02] Howdy folks === rdieter_work is now known as rdieter === alket_ is now known as alket [13:50] hm, some error in the Calligra update I got today: [13:50] /var/cache/apt/archives/kexi_1%3a2.8.7-0ubuntu1~ubuntu14.10.1~ppa1_amd64.deb [13:51] trying to overwrite '/usr/share/kde4/services/calligra/koreport_chartplugin.desktop', which is also in package calligra-libs 1:2.8.7-0ubuntu1~ubuntu14.10.1~ppa1 [16:24] Mamarok: a code in student did that package, for how you can dpkg --force-overwrite --install /var/cache/apt/archives/kexi_1%3a2.8.7-0ubuntu1~ubuntu14.10.1~ppa1_amd64.deb [17:22] dear phonon, why don't you see any sound devices if I boot with systemd o.O? [17:23] alsa seems to be working.... [17:24] ok, it's pulse that sees no devices, fun [17:31] no, it's alsa o.O [17:35] yofel, sometimes alsa automute is enabled after upgrades [17:36] nah, my problem is: [17:36] $ aplay -l [17:36] aplay: device_list:268: no soundcards found... [17:37] alsa is very meh [17:37] * yofel is just reading init stuff as it works fine with upstart [17:38] seems weird though, our pulse is session bound, so udev and thus alsa should be up and running at that point [17:38] unless you have configured your pulse to be global [17:39] nah, pulse is session bound, and pulseaudio -k doesn't help either [17:39] driver bug maybe? [17:40] how would that break under systemd? [17:40] libalsa has fun issues with reentrancy [17:40] ... [17:41] I keep being told that they are all fixed though :P [17:41] *getting [17:41] ofc ^^ [17:42] * sitter gets a headache from looking through things to do for 15.04 [17:45] er... why does aplay see the devices if I run it with sudo O.O [17:45] how I love permission problems [17:46] yofel: loginctl lists your session? [17:46] 1 148 sddm seat0 [17:46] 2 1000 yofel seat0 [17:49] does F6 in alsamixer show any soundcards? [17:49] if I run it as root yes [17:49] crw-rw----+ 1 root audio 116, 3 Dec 14 18:21 controlC0 [17:49] hm... [17:50] do note that permission control ought to be handled through logind regardless of what the file system permissions say [17:50] ok [17:50] no clue how that works, but I know ubiquity had a similar situation [17:50] that being said, I think the problem there was sddm [17:50] so it's entirely possible that our sddm profile is still busted [17:50] yofel: try lightdm for good measure maybe [17:51] we don't have a sddm systemd profile, it's starting init.d/sddm [17:51] oh actually [17:51] sddm has tight sytemd integration, since our sddm is not built with sytemd integration for reasons I do not know that could potentially cause problems like these [17:51] so yeah, try lightdm :P [17:51] yeah, good idea. brb [17:51] if it works there try to build a sddm with systemd enabled [17:56] yeah, works with lightdm [18:05] https://www.kubuntuforums.net/showthread.php?67037-Website-quot-stealing-quot-KFN-threads [18:06] lulz [18:06] internet be a weird place [18:09] XD [18:11] also why is there no way to view hottest topics of the past week, web forums are well terrible [18:11] claydoh: has anyone ever thought about using a different forum software? [18:14] sgclark: thanks, I already did so :) [18:20] sitter: it indeed was the missing systemd integration in sddm [18:20] works fine now [18:20] * sitter blames Riddell [18:20] * yofel is getting tired of hplip telling him that he has no system tray -.- [18:21] where's that autostart file.. [18:22] what what [18:22] sitter: why *did* we disable that? The git log doesn't give any hint [18:22] I fixed that [18:22] well, not fixed here [18:22] https://bugs.launchpad.net/bugs/1350038 [18:22] Launchpad bug 1350038 in HPLIP "systray needs compatibility adjustments to work with plasma5" [Undecided,New] [18:22] of course upstream is dead or something [18:23] though I think I uploaded a distro patch for that [18:23] yofel: https://launchpad.net/ubuntu/+source/hplip/3.14.6-1ubuntu1 [18:23] yeah, just read the changelog [18:24] hm [18:24] yofel: do you have sni-qt? [18:24] no [18:24] that'd be the problem then [18:24] * sitter ponders that maybe plasma-desktop should recommend it [18:24] then again that is a qt thing [18:25] yup, works once that's installed [18:25] thanks [18:25] Mirv: would ubuntu have any problems with promoting sni-qt to recommends/depends on qt4? [18:25] unity that is [18:28] yofel: I added it as recommends to desktop for the time being, for 15.04 this should not be a problem as we need breeze-qt4 for theme convergence anyway [18:29] k [18:33] *sigh* [18:34] Riddell: it's great if you leave packaging tasks to gci students if they're interested, but please a) do some Q/A *before* uploading to the archive and backporting and b) import the packaging into VCS once you're done if you don't use it during the packaging [18:35] .... [18:35] I even stopped complaining about unwritten changelogs -.- [18:36] changelogs are silly [18:36] like anyone cares :P [18:36] indeed [18:36] but yeah... technically when merging unstable people should compile the change sinto debian/changelog [18:36] that was the original proposal anyway, no one does it though, so... [18:36] I personally couldn't care less ^^ [18:37] I'm talking about calligra, but yeah, that too [18:37] sitter: why not write the changelog as usual in unstable? [18:38] incredibly hard to get to automerge [18:38] ok, I need to remove something from parley then ^^ [18:39] oh, good thing I have not looked at the diff or you'd have gotten a bit of bully me as Riddell put it :P [18:39] there was a way to fill the changelog using the git log right? [18:40] it's a bit meh though, I was actually thinking about making the merger a bit smarter in that changelog conflicts would always yield to 'theirs' which however also can have funny side effects :/ [18:40] yofel: probably [18:40] would require people to write proper commit logs though :/ [18:41] not really, the thing is... changes are supposed to be so atomic that even with proper commit messages the autogenerated output would be largely pointless noise [18:41] which is why I argued that a human should compile the changelog on merge [18:42] an autogenerated changelog with lots of pointless noise is about as useful as no changelog [18:42] that doesn't really scale with kde though... [18:42] which is probably why no one does it :P [18:42] unless the script actually marks packages that had changes in unstable [18:42] what I would find more useful anyway is documenting patches + the rest of the changelog should be upstream changes [18:43] i.e. post-release you'd document your patches for SRU, as that happens anyway. up until then and for PPA you'd document what changed upstream (under the assumption that our packaging did only adopt what upstream did or improved our structuring) [18:45] packaging changes are not useful to the user and right now muon-updater presents the changelog to the user which is mostly useless because the changelog except for post-release only contains what changed packaging-wise which is (or perhaps ought to) not be relevant for the user [19:21] FWIW other option would be to have a CI-exclusive changelog in debian/changelog.ci or debian/meta/changelog which would only hold the changes relative to $release branch and gets cleared upon merge of CI branch into $release branch [19:21] knowing the team I am reasonable certain that this sort of approach would be too much overhead for people so they wouldn't do it :P [19:25] I'll agree with that [19:35] *sigh* ~130 package copy errors in the updates ppa and nobody seems to care [19:35] I'll admit that cleaning those up is a PITA though [19:36] needs IRC bot [19:36] yofel: how do those happen? [19:36] Copied from: Private PPA for Kubuntu Ninjas. Copied by: Scarlett Clark Target series: Utopic [19:36] konsole 4:4.14.2-0ubuntu1~ubuntu14.10~ppa1 in utopic (same version has unpublished binaries in the destination archive for Utopic, please wait for them to be published before copying) [19:37] anyone fancy wiring my lp client to rbot xD http://anonscm.debian.org/cgit/pkg-kde/ci-tooling.git/tree/lib/lp.rb [19:37] yofel: shouldn't kopypackages do that? [19:37] that's why I prefer to use the deprecated syncSource() instead of copyPackage() as the former is synchronous [19:38] well, IMO there should just be a preliminary check for is-everything-published [19:38] well, with --sync it does, because of ^ [19:38] yeah, except that API is supposed to go away at some point so one really shouldn't use it [19:38] plus async is awesomest anyway [19:39] compile the changelog on merge ? what does that mean? [19:39] so IMO kopy should simply make sure everything is published before doing anything and crap out if everything isn't [19:40] in addition to that one could then possibly have a bot notifiction on failed copies but that shouldn't be needed all that often (i.e. only case where it would kick in is if a new package version is uploaded during copy, which is a bit of an edge case) [19:41] http://anonscm.debian.org/cgit/pkg-kde/ci-tooling.git/tree/ppa-wait.rb#n146 [19:41] right, checking whether the version is already there before copying should cover most usually [19:42] sgclark: right now kubuntu_unstable has no debian/changelog, so one would need to look at the changes one merges from kubuntu_unstable and manually write suitable debian/changelog entries [19:43] ahhh [19:43] np here [19:44] for you maybe not, for others yes :P [19:44] as I was saying, I don't particularly care, changelogs outside an SRU context mostly just document what would be documented in the git log anyway, so to me it seems like pointless information duplication [19:44] yeah [19:45] yofel, what is that about copy errors, what did I not do> or do wrong? [19:45] I have np cleaning up my stuff if I know what I need to do [19:46] nah, I just did the cleanup. What I think happened is that you did a mass package copy twice without delay in between (script failed in the middle or so?) [19:47] that's pretty much the only thing that would cause over a hundred not-yet-published errors [19:48] for some reason launchpadlib likes to crap out for weird reasons ^^ [19:48] ok [19:48] so package wise, nothing bad happened, but it fills up the launchpad copy package error list - and as the UI for it sucks, it's annoying [19:49] well, package-wise bad things can happen on copy errors ... if A requires B but B is not published and thus fails to copy A becomes uninstallable [19:50] worse yet if A technically requires B but the packaging relationships is screwed and then a maintainerscript assumes B is present and things explode at install-time [19:50] I don't think launchpad does dependency checks on copying? [19:50] yofel: which is why it can explode [19:50] A gets copied, B doesn't, user gets error [19:51] well yeah, but this case here is about copying A, getting an error, then copying A and B - resulting in a copy error for A [19:51] although, NONE of our scripts actually make sure that everything did get copied in the end [19:51] doh [19:52] I am just saying *technically* this sort of thing can cause errors, they don't necesasrily need to :P [20:58] wow i did noticed the undo function in frameworks it has been added 5.5.0 ? [20:59] *added with [21:07] undo function? [21:13] sgclark: yes this one http://notmart.org/blog/tag/plasma-5-2/ === soee_ is now known as soee [21:15] oooh neat