[00:04] Does anyone have a git tree that they're having Jenkins turn into a dpkg? I've got a few questions if so relating to versioning. :-) [00:20] You can't use the hash of the git commit for the version. You need something montonically increasing (people usually us date based). [01:58] Corey, have you looked at git-buildpackage? [02:03] Kiall: I have. [02:04] Kiall: Unfortunately it doesn't seem to play nicely with the script I dug out for making Jenkins build the package. [02:04] There actually isn't much to building the package in Jenkins, gpb handles most of it.. [02:05] http://review.managedit.ie/graphite/carbon.git is a example git-buildpackage repo i build from jenkins [02:06] http://review.managedit.ie/managedit/ci.git are the most recent (read: still being worked on) scripts I use to build from jenkins.. [02:06] How do I verify the DIST pdebuild is using? [02:06] Kiall: Ooh. Thanks. :-) [02:06] Wait, first one 404'd. [02:06] As does the second. [02:07] Hmm. Kiall, I think you're taunting me. :-) [02:07] git clone.. not your web browser ;) [02:07] Oh jeez. [02:08] Kiall: Which script do you call from the Jenkins project for carbon? [02:10] The jenkins job is basically 1 build step - http://paste.ubuntu.com/861391/ [02:11] whj [02:11] whoops - cut the end off that http://paste.ubuntu.com/861392/ [02:12] Kiall: Ooh. Are build-deb.sh and publish-deb.sh publically available? [02:13] I've spent some time digging, and can't find this type of thing anywhere. [02:13] yea - in that second git repo I linked to above [02:13] http://review.managedit.ie/managedit/ci.git are the most recent (read: still being worked on) scripts I use to build from jenkins.. [02:15] Oh, durh. :-) [02:15] Kiall: You have plans to turn this into a blog post at some point? [02:15] It's hugely useful. [02:15] I'm still working on them, but the bones of building packages has worked for a while.. just adding sbuild / schroot support for building in a clean environment... [02:15] If I had a blog ;) [02:16] Kiall: My current (badly written) script successfully uses pbuilder. [02:21] I've used pbuilder before, but kept hearing good things about sbuild so figured I'd give it a go ;) [02:46] Kiall: I'll have to keep track of your progress on that. :-) [02:46] Sure.. I'm usually here ;) === almaisan-away is now known as al-maisan [07:39] grr pytables still fails on arm! [07:40] at least now 2000 test passed instead of 500 [07:50] good morning === al-maisan is now known as almaisan-away [09:17] Bah! Looks like we missed feature freeze - when did that occur? [09:18] danboid, https://wiki.ubuntu.com/PrecisePangolin/ReleaseSchedule [09:18] I guess you might be also interested in https://wiki.ubuntu.com/FreezeExceptionProcess (if applicable) [09:23] dholbach, Thanks! [09:24] anytime [11:50] danboid: two weeks ago :) But as dholbach notes, if it's important, then a FFe would be the way to go [11:53] sladen, Its important for users of the app in question yes and updating it wouldn't cause probs for any other apps [11:55] danboid: sounds like the advantages outweigh the disadvantages === yofel_ is now known as yofel [15:20] hey guys... is there a way to generate a .dsc file without building the package? [15:23] ESphynx: you want to build the source package without building the binary package(s), don't you? [15:25] debuild -S -sa (or debuild -S -sd if you don't want to include *orig.tar.gz) [15:26] Adri2000: i want to build with pbuilder... [15:26] but I want to drop to a shell so I modified the package to add a failure [15:26] I really don't want to build anything, I just want to update the .dsc that I already have with the matching SHA stuff [15:28] maybe --preserve-buildplace is what I want [15:31] thanks [16:50] Hi all! === DrKranz is now known as DktrKranz === tubadaz_ is now known as tubadaz === lifeless_ is now known as lifeless === almaisan-away is now known as al-maisan [18:44] when merging a new source into ubuntu is there something I need to take care of? [18:45] e.g. -1ubuntu1 is ok as first version number? [18:48] if it's based on -1 from Debian then yes [18:48] jtaylor: No, -0ubuntu1 if it's a new upstream not yet in Debian. [18:49] it is in debian [18:49] just the version in debian cannot be synced directly as it will ftbs [18:49] Then -Xubuntu1, where X = the debian revision [18:52] Hello, I have three points where I'd like to help: 1. Getting my headphone-jacks working correctly 2. getting my multimedia-keys working 3. Getting the package spice-xpi in the ubuntu repository [18:52] Who can help me achieve those? [18:52] phiphi: for the first two please ask in #ubuntu, this is no user support channel [18:52] on 3. we can help [18:53] phiphi: 1+2 #ubuntu, #3, the packaging guide and you can ask here for help [18:53] phiphi: what is spice-xpi, is that a Firefox extension? [18:53] 1. and 2. is not meant to be user-support but hardware-support improvement [18:54] spice-xpi is a firefox-extension needed for accessing vms in RHEV / oVirt (SPICE-Protocol) [18:55] there's a channel for ubuntu mozilla stuff isn't there? [18:55] there is already a ppa with a package of it [18:55] we're generally not taking new mozilla extensions in the archive without a good reason and it would have to be a binary extension [18:56] http://blog.jebpages.com/archives/spice-spice-baby/ [18:56] is it an NPAPI only extension? [18:56] i think it has binary components [18:56] the problem with mozilla extension packages is they'll break horribly when firefox 11 ships. then again with 12, etc. firefox sucks at abi compatibility - which is a problem from a long-term distro support perspective [18:57] right, so unless it's NPAPI only, we're not likely to take it [18:57] arg, I forgot to update-maintainer in planner upload [18:57] is that reason enough to ask for a rejection? [18:58] jtaylor: do you not have your ubuntu E-Mail in DEBEMAIL? [18:58] I do [18:58] but I must have overlooked the warning [18:58] it should error out then [18:58] it didn't [18:58] spice-xpi is the glue to the spice-client which is packaged in the repos. In fedora / redhat it's a package in the repo [18:58] hm probably because my debemail is still the debian mail, I changed it to ubuntu by hand in the changelog [18:59] jtaylor: right :) [18:59] who reviews the packages? [18:59] phiphi: since we're updating to a new Firefox release every 6 weeks, we don't have the resources to keep updating extensions [19:01] micahg: Canonical is strategic supporter of oVirt. Is there noone that should care about it? http://www.ovirt.org/about/sponsors-and-supporters/ [19:02] phiphi: idk, that might be a good reason :) [19:03] phiphi: can you file a needs packaging bug (if there isn't one already) and subscribe me? [19:14] ...is canonical actually supporting oVirt strategically in practice? i thought we still don't even have all the pieces to run FreeIPA [19:15] jtaylor: I forget update-maintainer fairly regularly, and don't lose sleep over it :) [19:16] broder: no idea, but I can ask about it :) [19:17] some DD here you wants to sign a request for a porterbox in debian? [19:17] guest account on one I mean [19:20] jtaylor: I'll happily do that, what for? [19:20] I want to look at pytables [19:20] I'm guessing the thing is just plain broken on !x86 [19:20] (and always was, it previsouly did just not execute its testsuite) [19:21] I want to check how many tests fail to decide if I remove it from arm or keep it reduced functionality [19:22] though its maybe not worth the trouble [19:22] micahg: done https://bugs.launchpad.net/ubuntu/+bug/943510 [19:22] there are probably 0 arm users of that package ._. [19:22] Launchpad bug 943510 in Ubuntu "[needs-packaging] Package spice-xpi" [Undecided,New] [19:22] jtaylor: send me a mail with this data, and I'll sponsor it http://dsa.debian.org/doc/guest-account/ [19:24] phiphi: thanks [19:35] tumbleweed: http://paste.debian.net/158092/ === al-maisan is now known as almaisan-away === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away === medberry is now known as med_ [22:51] AlanBell: all those hi-res icons bugs need to be forwarded upstream, we're not going to carry a permanent diff for these things [22:52] same hould apply to most of the quicklist bugs :/ [22:53] jtaylor: those were already addressed separately [22:53] ah good [22:56] it would be nice if people asked for feedback on the ubuntu-devel list before running "submit patch" programs [22:58] micahg: yes, I think mhall119 is tracking upstreaming of them too [22:59] it was relating to a message on the unity-design list https://lists.launchpad.net/unity-design/msg08430.html [23:00] yes, sorry that the quicklist and keyword submissions are not package-patches, I didn't know enough of that to walk people through, I'm learning it currently [23:00] I personally don't think it's worth filing in launchpad unless it's linked to an upstream bug [23:01] even then, it would only be worth it if we need to pull it in separately or we're worrying people will report it on their own and want to show them it's already been reported [23:06] mhall119: are you on the ubuntu-devel mailing list? there was discussion about the workload created by the quicklist patches, including mentioning how the patches weren't quilted and lacked changelog entries. i was a little disappointed that you didn't see that and address it before starting another patch project [23:06] micahg: they should all be filing with upstream too, I'm checking in to make sure they're doing it [23:07] broder: I am, and I've spoken to both didrocks and seb about it [23:07] mhall119: thanks, I'd actually suggest having people file the upstream bugs first so they can be linked right away and same developers time thinking/worrying about them [23:08] s/same/save/ [23:08] micahg: is there a way to manually pull the upstream bug into launchpad? [23:09] mhall119: you link it by clicking also affect upstream, some upstream bugzilla installations can have their comments imported into launchpad once linked [23:09] sorry, also affects project [23:09] broder: as soon as I learn how to make package patches, I will be blogging a tutorial on it for the people who have been following my others [23:10] dholbach was kind enough to give me a walk-through set of steps [23:30] hello, I have a little packaging issue with a custom software. building it requires a custom library which lives in my ppa so pbuilder fails with pbuilder-satisfydepends-dummy: Depends: libcppinstrospection-3.0-dev which is a virtual package. [23:31] is the ppa added to the sources.list of the ppa? [23:31] however, I added OTHERMIRROR="deb http://ppa.launchpad.net/sat-metalab/metalab/ubuntu lucid main" to my .pbuilder, run pbuilder update and I am still getting the dependency error [23:32] have you checked if the othermirror setting succeeded? [23:32] yes, I have the ppa added to my sources but pbuilder seems to ignore ppa listings... [23:32] how do I check that? [23:32] just look at the sources.list [23:32] can you install libcppinstrospection-3.0-dev in the pbuilder? [23:32] manually [23:34] you mean in the global /etc/apt/sources.list ? [23:34] the one inside the pbuilder chroot [23:34] oh, perhaps I should RTFM on setting pbuilder manually... [23:34] pbuilder --login [23:34] ^ [23:35] and then within that chroot login, you go to /etc/apt/sources.list... [23:35] and add the source in that [23:36] well to change it permanently you would have to use --login --save-after-login [23:36] but for the test installation use only --login to keep the chroot clean [23:37] ah ok, thanks. So that means that my OTHERMIRROR is not taking effect (I checked and the ppa is not in sources.list) [23:38] use --login --save-after-login and add it by hand, getting OTHERMIRROR use right will take much longer :) [23:39] :) [23:42] hmm... but even now, after adding my ppa to the sources.list, and even installing it by hand in pbuilder, it fails with the same error... [23:44] oh, nevermind. there are some other dependecy problems. Will investigate