[00:07] AlanBell: you rock, thanks! (for the tarball) === pdtpatrick_ is now known as pdtpatrick === Quintasan_ is now known as Quintasan [01:05] stgraber: haha. way to grab bug #1e6 [01:05] Launchpad bug 1 in Ubuntu "Microsoft has a majority market share" [Critical,In progress] https://launchpad.net/bugs/1 [01:05] no, not that one :-P [01:05] silly bot [01:06] :) === Amaranthus is now known as Amaranth === Amaranthus is now known as Amaranth === EvilResistance is now known as Resistance [06:58] hah, again self-fixed problem, love those. :) === menesis1 is now known as menesis === almaisan-away is now known as al-maisan === StevenK_ is now known as StevenK === bigon_ is now known as bigon [11:24] I have an ubuntu package which uses dh-translations and thus calls dh -Spython_distutils --with python2,translations [11:24] dh-translations does not exist in debian [11:25] the rules file can easily be fixed up with dpkg-vendor [11:25] but I see no clean way to deal with the difference in build-deps. [11:25] Do lots of ubuntu packages have a delta adding dh-translations as a build-dep? [11:26] Is there a better way to deal with this? (package is usb-creator) [11:29] Also there are xml .ui files for gtk and kde. where it would be useful to insert either Ubuntu or Debian as approporiate. Is there a neat way to do this (nicer than sedding ui.in files in rules)? === Adri2000_ is now known as Adri2000 === al-maisan is now known as almaisan-away [12:07] wookey_: I guess you need a source package for Debian and one for Ubuntu then [12:08] but if you manage this package in a VCS you could create an Ubuntu branch with all the needed Ubuntu delta (labels, build-deps, etc.) and rebase it on every new package revision [12:15] wookey_: conditional symlink based on lsb distro? [12:16] you can have the build system substitute in based on a configure flag [12:16] but I don't know of a solution for the build-deps. === ryanakca_ is now known as ryanakca [12:43] isn't it possible to have a template control file debian/control.in which regenerates debian/control at build? [12:46] it is discouraged I believe [12:48] I don't see how that could work for build-deps [12:57] according to http://raphaelhertzog.com/2010/09/27/different-dependencies-between-debian-and-ubuntu-but-common-source-package/#comment-3779 it should work, even though that's not really clean I admit [13:01] clean isn't run until after b-ds are installed [13:01] have a look at a build log to verify [13:05] moo [13:05] * Laney zaps highvoltage [13:06] *BZZZT* === slangase` is now known as slangasek [13:10] Laney: don't produce a short circuit when playing with highvoltage [13:28] * bregma sips his coffee [18:32] Zhenech: ping. [18:32] Zhenech: why isn't libappindicator in debian? =\ [18:34] hyperair, because you didnt package it [18:34] lol... [18:34] Zhenech: and the reason for that is because pkg-ayatana insists on using bzr.. ¬_¬" [18:34] Zhenech: i thought you were maintaining the indicator stack in debian though. isn't it pretty useless without appindicator as well? [18:36] no, works w/o quite good [18:36] and yes, bzr sucks [18:36] but makes "stealing" from ubunto so mucheasier [18:46] hm [18:46] how so? [18:46] you would have to merge from ubuntu into debian, and then back again, wouldn't you? [18:46] doesn't it get really weird there? [18:49] or upload it as 1.2.3-0debian1 :) [18:52] hyperair, i always wait till ubuntu hass foo-0ubuntu1 and merge that [18:53] hmmmm. [18:53] i see [18:53] super easy [21:46] micahg: https://wiki.ubuntu.com/PackagingGuide/Recipes/PackageUpdate <-- apparently needs some love [21:48] bkerensa: we'd love to get to thepoint where we can remove those pages and just point people at the packaging guide [21:48] :D [21:48] bkerensa: well, we have http://developer.ubuntu.com/packaging/html/udd-merging.html#merging-a-new-upstream-version [21:49] which means either cover non_UDD workflows in the packaging-guide or get everyone usin gUDD (which requires it to work :P ) [21:49] it works, just not in all cases :) [21:50] fair enough [21:50] and I doubt we could call the traditional way of doing things "working" when there's a multi-k-line diff in a 1.0 source package [21:52] * ajmitch would like for a good solution to bzr+quilt that works for every package, but that requires some work [21:52] it sounds like we know what it might be, but nobody is putting the work in [21:54] sounds like a great opportunity for someone who likes pain :) [21:55] ;p [21:55] * tumbleweed bets the git people get there first. there's certainly way more experementation on that side of the fence [21:56] then we can move all ubuntu branches to git? [21:56] launchpad has to support it first :) [21:56] mere details [21:56] not that I'm a bzr hater. But I quite like git too [21:57] and it appears to have won [21:57] there was a "talk to" item about the .pc ignoring stopgap [21:57] that certainly seems more likely than making looms work [21:57] Laney: right, I can't recall if the importer was going to unapply patches [21:57] definitly sounds like there are stopgaps that'll make things a lot better [21:57] looms seem like a nice-to-have [21:57] ajmitch: I think we decided not to do that, it would be fairly insane [21:57] right [21:58] patches would be applied, and that's what you want really [21:58] but rather to get as little of .pc in the branch as possible [21:58] but you can construct .pc on checkout [21:58] so it doesn't need to hit the vcs [21:58] oh, right, that [21:58] http://summit.ubuntu.com/uds-q/meeting/20342/foundations-q-udd/ wasn't entirely clear on which way was chosen & my memory is foggy :) [21:58] aha [21:59] * tumbleweed wasn't a great note-taker this UDS :/ [22:02] tumbleweed: I'm sure you did a better job than I did :) [22:04] * tumbleweed also should have raised the debian stuff with debian-derivatives before the sessions, calling for input [22:04] I wish I'd taken more notes but I was completely information-overloaded [22:04] we're bad DDs [22:05] heh [22:05] highvoltage: there were some sessions where I volunteered to take notes and then found myself staring into space or diving into arguments and losing track of the notes [22:05] good thing you grabbed the audio then [22:06] well, there were /some/ sessions where you couldn't be blamed for staring into space :) [22:06] ah, btw, to get things to show up on status.ubuntu.com, all you need to do is mark the session as approved for quantal [22:06] oh good [22:06] * tumbleweed must go back and check that all his sessions are... [22:07] 15% done already [22:07] not bad [22:07] * ajmitch is 0% done, blame the ubuflu :) [22:07] * tumbleweed is just getting over that [22:08] what's the methodology of enabling multiverse and universe inside a chroot, would that be adding `multiverse universe` to the apt sources.list lines in the chroots? [22:09] yes [22:09] indeed, just wanted to confirm [22:11] tumbleweed: so should i just set the state on my blueprint to approved? [22:11] the...definition state, i guess? [22:11] broder: no, set a "series goal" [22:12] aha [22:12] at least that worked for the debian healthcheck fo rme [22:16] * Laney wibbles [22:16] EVAN BRODER [22:16] DO WE NEED TO UPDATE THE BACKPORTS DOCS TO MENTION THAT WE UPLOAD OURSELVES NOW [22:16] ?????????????????????????????????????????????? [22:16] also, hello! [22:17] >.< [22:17] * ajmitch takes Laney's shift & capslock keys away [22:18] MR LANEY HELLO STOP [22:18] Laney: I want to be a backporter pls show me how [22:18] * Resistance raises an eyebrow [22:18] haha [22:18] oh... [22:19] for a minute i thought there was a spammer in here [22:19] there was [22:19] i can spam if you want! [22:19] * Resistance has autonotify scripts for multiple full-caps lines [22:19] Laney: UNDER NO CIRCUMSTANCES [22:19] you want a ghc upload yes yes? then i get to spam you with uploads [22:19] * tumbleweed disables capslock. it serves NO PURPOSE [22:19] * Laney has it as compose [22:20] there's something satisfying about holding down shift [22:20] * Resistance lights his capslock key on fire [22:20] Laney: oh we're so looking forward to another ghc transition [22:20] HELLO SIR I FROM SUDAN AND MY RICH UNCLE HAS RECENTLY DIED LEAVING ME WITH 6 GAJILLION DOLLARS PLEASE PROVIDE ME WITH YOUR BANK ACCOUNT NUMBER AND I WILL TRANSFER MONEY TO YOU [22:20] Laney: docs should probably be updated, but i'm holding off on doing so until i get a tool for AAs to use to verify things in-queue [22:20] broder: now that is spam, but i got that three hundred times yesterday... [22:22] ok then [22:23] would you consider making backportpackage not remove the temporary directory if dput errors out? [22:23] * micahg was about to ask what the preferred way to upload a backport is now :) [22:23] * Laney had to download the widelands >100 meg orig a few times due to various -u fails [22:23] ouch [22:23] Laney: seems fair [22:23] Laney: use backportpackage with the .dsc :) [22:23] micahg: my preferred way is to ping Laney [22:24] ajmitch: I'm a backporter :) [22:24] I'm not that special [22:24] and micahg and broder are my contacts for backport questions :P [22:24] yeah, but it has this handly ability to download the package for me [22:24] -w . works too I suppose [22:24] want to process python-leveldb and see how you get on with uploading it? [22:24] Laney: that's good the first time around :) [22:25] i.e. if backportpackage does what you want [22:25] Laney: was about to, yes, just waiting for i386 to build in quadrispro's PPA [22:25] cool [22:26] * micahg likes the new bug in changelog feature [22:26] helps with the audit trail [22:27] * ajmitch sees there's only one admin of the backporters team that's seen around here regularly [22:27] micahg: any thoughts on this http://paste.ubuntu.com/991459/ [22:27] we call him The Boss [22:28] bkerensa: a bug in setup.y? [22:28] *setup.py [22:29] it seems to be defined though [22:30] bkerensa: no idea, maybe someone more versed in python can help [22:30] * micahg wonders if the other 2 people in the backporters team should be removed [22:30] oh, nm, they're MOTUs still :) [22:31] lol [22:31] well, we don't want non-MOTUs/core-devs doing backports [22:31] I thought that wasn't possible? [22:32] what's not possible [22:32] micahg: there are a few people pending as backporters, what's the process for helping out? [22:32] non-MOTU / core-devs doing backports [22:32] triage some and get scott to review your work [22:33] tumbleweed: yeah, was worried about lapsed upload rights and what not [22:43] hey motus, I'm watching a critical security bug for the pidgin-otr package -- I'm not sure what the protocol here is, but there's a debdiff on the bug report and an upload about to go into debian -- how could I help the fix get into Ubuntu? [22:43] lp bug is https://bugs.launchpad.net/ubuntu/+source/pidgin-otr/+bug/1000363 [22:43] Launchpad bug 1000363 in pidgin-otr (Ubuntu Quantal) "CVE-2012-2369: Format string security vulnerability" [High,Confirmed] [22:43] eridu: folks in #ubuntu-hardened should be able to guide you [22:43] kees: thanks! [22:44] eridu: looks like its already taken care of, its now in the hand of the security sponsors [22:44] oh, alright then [22:44] same version in all releases makes it easy :) [22:45] how were you able to tell that? [22:46] I remember putting my name on a wiki page as the bug monitor or something for the pidgin-otr package, so I felt obligated to do something [22:46] debfxs has posted a patch [22:47] and forwarded to debian and ack'ed thare [22:49] anyone able to help me figure out why the heck when i pbuilder --login to the chroot, and change the apt sources.list file, the changes dont take? [22:49] Resistance: forgot --save-after-login ? [22:49] ahhh [22:49] use --save-after-login [22:49] probably [22:49] thanks [22:50] any way to tell pbuilder to automatically use that when i use the login command? [22:51] you shouldn't [22:51] okay then [22:51] i wont :) [22:51] pbuild is supposed to provide the same (clean) environment each time [22:51] jtaylor: i had to enable universe / multiverse [22:51] so... [22:51] :P [22:51] you can do that during creation with COMPONENTS [22:52] e.g. in pbuilderrc: COMPONENTS="main restricted universe multiverse" [22:53] huh [22:53] that's in my pbuilderrc [22:53] and its not listening each time i create the environment :/ [22:53] * Resistance smells a bug that's specific to his system [23:29] have the guidelines for requesting a backport changed much since Precise was released? [23:29] Resistance: use requestbackport :) [23:29] micahg: ugh, why the heck do i keep forgetting requestbackport [23:29] oh right, because it segv's on this system [23:30] * Resistance has a slighitly-busted system [23:30] * Resistance goes off to reinstall Precise [23:31] actually... [23:31] * Resistance just reinstalls the program that was causing problems [23:32] anyways, micahg, i guess that if i'm waiting for a build to finish on a PPA (so I can test the backport), then the request will just sit there until i confirm it works? [23:33] Resistance: yes, just note in the description what you're planning [23:34] the heck...? [23:34] micahg: requestbackport doesnt run as a user that isnt the user running it, right? [23:34] AFAIK, runs as the current user in the shell [23:34] hmmmmm [23:34] that's interesting [23:34] because... [23:35] the system isnt unlocking my keyring (even though i've already unlocked it today) [23:35] and this system isnt auto-login [23:35] :/ [23:35] its as if its running as some other user [23:35] (passcode isnt working) [23:35] oh, and apport is complaining because i cancelled [23:35] how wonderful [23:35] it's a terminal in X? No ssh or screen or anything like that? [23:36] tumbleweed: gnome-terminal [23:36] within Ubuntu 12.04 [23:36] * Resistance is confused [23:37] okay, lemme log out and log back in [23:37] sounds sensible [23:37] FWIW python-keyring (or something below it) does seem to be a bit buggy [23:38] that might be where the error is occuring [23:38] perhaps this needs a bug filed against python-keyring? [23:39] we know about the issue, just don't know what causes it [23:39] something throws IOErrors when accessing the keyring every know and then [23:39] well then this sucks, because the error i'm having is authing LP API access from my system [23:39] you still having trouble? [23:40] i cleared all the old auths, trying one last time before i shoot the auth system [23:40] yep, still having trouble [23:40] want the scrollbacks? [23:41] s/scrollbacks./ [23:41] bleh [23:41] i meant the trace [23:41] yes please [23:41] i could generate an apport if necessary, do yo uneed that, or no? [23:41] let's see the traceback first [23:42] wait... [23:42] what the...? [23:43] oh [23:43] right [23:43] * Resistance ignores that error, and removes the package being the problem [23:43] anyways, that trace [23:47] tumbleweed: https://pastebin.com/Y1Sh0LEE [23:47] bleh [23:47] without the https://, but with http:// [23:48] tumbleweed: the "error" is a GUI one: it pops up a box asking for me to put the passcode in for the keyring, but... [23:48] Resistance: right, that traceback doesn't help much, but your description does shed some light [23:48] is the keyring working for other things? [23:48] such as network-manager [23:48] its error says the passcode for the keyring unlock is invalid, and needs to be reentered [23:48] and that point it does not worik [23:48] and yes, it works with network-manager, and packaging signing (GPG) [23:49] s/GPG/PGP/ [23:49] (since my PGP key is in my keyring) [23:49] package signing doesn't use the gnome-keyring [23:49] *shrugs* [23:49] network-manager's working [23:49] and its unlocked and works at-login [23:49] can you browse the keys in seahorse (parsswords tab) [23:49] what keys? [23:50] :/ [23:50] oh wait a minute [23:50] *that* isnt unlocking... [23:50] what the hell? [23:50] bleh [23:51] how do i fix that then [23:51] delete the keyring? [23:51] it's in ~/.gnome2/keyrings/ [23:51] (back it up first, of course) [23:51] back it up, even though its empty? [23:51] this is a new installation of precise [23:51] :/ [23:51] then just delete it [23:52] when you create a new one, use your login password and it should unlock automatically [23:53] ah that worked [23:53] yeah, the passcode for the network i'm on right now in network-manager is available system-wide, so its pulling from the shared keyrings, not my local one [23:54] s/local/personal/. [23:58] tumbleweed: are you a backporter, or should I stab micahg? [23:59] * micahg puts on a suit of armor [23:59] * tumbleweed is not a backporter. I just hang out with them :) [23:59] :P [23:59] okay then [23:59] micahg: https://bugs.launchpad.net/precise-backports/+bug/1000492 i can't test certain packages [23:59] Launchpad bug 1000492 in Precise Backports "Please backport boinc 7.0.27+dfsg-3 (universe) from quantal" [Undecided,New] [23:59] as the dependency for certain packages there is to have certain card architectures i dont have [23:59] Resistance: looks like you still have lots of testing to do [23:59] namely the nvidia-cuda one [23:59] tumbleweed: read the description / testing area