[00:20] * mwhudson lunch [01:30] * thumper is actually enjoying the javascript stuff :) === ursula is now known as Ursinha [02:45] man you can tell the sort of day i've had [02:45] rebooted last night; hadn't started emacs until just now [03:03] mwhudson: oh no [03:03] mwhudson: why is that? [03:03] thumper: because i've been writing email/on the phone/yakking on irc all day [03:03] heh [03:45] man i hate our vcs test fixture thingies we use in the code import testsd [03:46] you could use testresources [04:09] lifeless: OK, now I agree with you that creating SPNs on push is sane. [04:09] I think. [04:09] :) [04:10] wgrant: actually trying to do something can be very educational [04:19] code that deletes the cwd can seem a little odd [05:17] * mwhudson eods === spm changed the topic of #launchpad-dev to: LP Codehosting offline for H/W update 0800-1200 UTC | This is Launchpad Development Channel | Week 2 of 3.1.11 | PQM is open | I am Zero OOPS and So Can You! http://is.gd/4fkLl | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes | This channel is logged: http://irclogs.ubuntu.com/ [08:05] oh stabby [08:05] my ec2 test of bzr-svn-imports failed because the relevant goo isn't installed on the image yet [08:14] mwhudson: :( [08:14] mwhudson: hey, I have a stabby alias now :) [08:14] ✁☹ [08:15] ☕ [08:18] hmm, just tried to push :( [08:19] thumper: you're aware that codehosting is offline right? (see topic) [08:19] noodles775: yeah, I remembered after I tried to push [08:19] noodles775: I have a cool branch to land though :) [08:19] :) [08:19] * wgrant wonders how it could possibly take four hours, unless the disks are being entirely replaced offline. [08:20] wgrant: the resize forces a fschk [08:21] thumper: Ah. [08:21] backup; resize; fsck [08:22] spm: so what will this do for our space? [08:22] thumper: an extra 2-3 weeks breathing space :-P [08:25] I wonder how much space would be saved by regularly pruning unnecessary revisions from stacked branches. [08:30] good morning [08:31] Hi adeuring [08:31] hi noodles775! [09:15] I'd think quite a lot, since for merged branches that amounts to deleting almost all their data [09:16] maxb: Exactly. [09:19] step 1) implement 'bzr gc' [09:19] step 2) implement 'bzr gc --keep-dead-heads' [09:53] wgrant: i wonder that too [09:53] wgrant: want a write a script that will try it? [09:53] lifeless: you could go for a more inaccurate hammer: if the tip of the branch is in the repo of the stacked on branch, delete the repository from the stacked branch [09:53] (i realize this isn't quite the same thing) [09:56] mwhudson: I wouldn't do that. [09:56] mwhudson: as tags would get fucked over. [09:56] tags are stored in the repo? [09:56] i thought it was the branch [09:56] but never mind [09:56] tags refer to revisions [09:57] if you zap the repo, and there is a reference to a revision not in the stacked on repo... boom [09:57] you've just deleted something important === mthaddon changed the topic of #launchpad-dev to: This is Launchpad Development Channel | Week 2 of 3.1.11 | PQM is open | I am Zero OOPS and So Can You! http://is.gd/4fkLl | https://dev.launchpad.net/ | Get the code: https://dev.launchpad.net/Getting | On-call review in #launchpad-reviews | Use http://paste.ubuntu.com/ for pastes | This channel is logged: http://irclogs.ubuntu.com/ === matsubara-afk is now known as matsubara [11:40] codehosting is still down for maintenance, I guess? [12:01] thumper, the yui 3.0. [12:01] thumper, the yui 3.0.0 branch has landed on db-devel === ursula is now known as Ursinha [14:27] morning launchpadders [14:31] barry: bac: stand-up in 2 minutes [14:31] morning flacoste [14:32] stub, Chex, gary_poster, rockstar, Ursinha, sinzui, allenap: LP production meeting in 28 min @ #launchpad-meeting [14:34] flacoste: Hi. Would it be possible to reassign the official meta-lp-deps branches to ~launchpad-committers at this point? [14:34] maxb: sure thing [14:34] matsubara, I'm afraid I won't be able to attend [14:35] Thanks :-) [14:35] Ursinha, that's ok [14:37] maxb: i'm going to mark the dapper and gutsy branch obsolete, any objections? [14:38] Makes sense [14:40] maxb: all done [14:41] flacoste: Could you reassign the dapper and gutsy too? To keep them all together and so that I can patch up their stacked-on urls. [14:44] maxb: doing... [14:45] and done [14:50] suitable surgery performed on the stacked_on_urls [14:50] via horrid assortment of sftp and bzrlib in interactive python :-) [14:51] hi bigjools [14:51] are you going to be around for the lp production meeting? [14:51] in 9 min :-) [14:55] matsubara: no :) [14:55] bigjools, anyone from soyuz will be around? [14:56] noodles775: would you be able to fill in please? [14:56] and good morning/afternoon BTW :) [14:56] bigjools, matsubara: yep. [14:56] thanks noodles775 [14:56] and bigjools === salgado_ is now known as salgado [15:07] hello launchpadders === salgado is now known as salgado-lunch [15:19] hello jml! [15:20] :) [15:20] I would like my own chorus [15:20] heh [15:21] you know what one of the hidden dangers of two weeks solid non-hacking sprinting is? [15:21] rogue ec2 instances [15:21] gary_poster, it's difficult to chorus without timestamps. I see jml, I see barry. Could have joined 6 hours apart for all I know! [15:22] well, except that jml is probably at UDS right now, but I often forget that [15:22] * barry sings d'oh, ray, me [15:22] lol [15:22] jml, eewww [15:23] gary_poster: hi! how screwed are my 2 week old launchpad branches now? :) [15:23] jml, 'ec2 check-for-zombies', put it in a crontab? [15:23] barry: pretty screwed! :-) you have Python 2.5 and YUI 3 final landings now. ;-) [15:23] yeah, I should do something like that. [15:24] I have to put it on my server, obviously, because laptop-powered crontabs are next to useless. [15:24] we all should [15:24] * mars runs a desktop system [15:24] gary_poster: yay! hopefully they're all review branches and landed branches [15:26] gary_poster: I am trying to use a new library with launchpad (ampoule). I've added the tarball to download-cache/dist and I've edited versions.config. But it doesn't get installed when I run make, and make doesn't error either. What am I missing? [15:27] * jml has "Oh What A Night!" stuck in his head [15:27] * barry stabs jml [15:27] at last, sweet mercy [15:29] abentley, I just did that yesterday. check doc/buildout.txt, there is a section with exactly the instructions you need. [15:29] abentley: add it to setup.py. (we need to have indirect versions in versions.cfg, so it is nice to have all the versions in one place. direct dependencies should be declared in setup.py as usual) [15:29] or see mars' reply :-) [15:29] gary_poster, mars: thanks. [15:30] * jml goes to look for deryck & intellectronica [15:30] jml, if you find them or beuno, tell them the bugs subscriber portlet has been broken for a week now :( [15:31] mars, do you have a bug number for that? [15:32] jml, 484848 [15:32] bug 484848 [15:32] Bug #484848: Subscriber icons in the subscribers portlet appear on a separate line from the subscribers' names [15:32] (I successfully demonstrated the bug subscriber portlet last night) [15:32] jml, just a style issue, but when you see it, it makes you go "oops!" [15:32] mars, I'm sure they would gratefully accept a patch :) [15:33] jml, of course. I did some work on it yesterday. I'm not actually committing to fixing it though. [15:33] flacoste, hello sir. === EdwinGrubbs_ is now known as EdwinGrubbs [15:52] hi rockstar, on the phone [15:52] flacoste, no sweat. [15:52] flacoste, I wanted to talk to you about fixing canonical_url in windmill. === matsubara is now known as matsubara-lunch [16:02] mars, I know of the broken prtlet, so does deryck === bac is now known as bac-lunch [16:42] gary_poster, hi [17:04] rockstar: hi. lunching. ping you when I return? [17:04] gary_poster, I got what I needed from flacoste. Thanks though. [17:05] cool === henninge_ is now known as henninge === matsubara-lunch is now known as matsubara [18:48] sinzui: is anyone looking at bug 485237? you want me to? [18:48] Bug #485237: OOPS - KeyError: 'link' - in person-index.pt [18:49] bac no, it was reported a few hours afo [18:49] sinzui: so is that a "no, yes"? === bac-lunch is now known as bac [18:50] bac: no, that is a no. it is not assigned and it is not inprogress. [18:50] sinzui: righto. but would you like to assign it to me? i'm looking for work. [18:50] bac: take it [18:50] okey doke [20:04] morning [20:21] maxb: hi === salgado is now known as salgado-afk === mwhudson_ is now known as mwhudson [20:24] or anyone else who understand how the launchpad ppa works currently... [20:25] * wgrant knows a bit. [20:34] 'moin mwhudson [20:34] mwhudson, did you see my comment about _make_silent_logger imports in your bzr-svn-imports branch? [20:35] jelmer: yes, i think i replied? [20:35] sorry, I must've missed that [20:37] jelmer: i need to build launchpad-dependencies 0.57 for hardy and make an ami for ec2test before i can land it anyway... [20:38] which is why i was asking about the ~launchpad ppa [20:38] wgrant: i built launchpad-dependencies 0.57 for karmic last night [20:38] wgrant: what do i need to do for hardy? [20:39] mwhudson: ah, ok [20:40] mwhudson, what about the import error? === matsubara is now known as matsubara-afk [20:42] mwhudson: You'll need to merge your Karmic branch into the Hardy branch. [20:42] mwhudson: And probably give it a version number like 0.57~hardy1, rather than the traditional 0.57hardy1 which isn't right. [20:42] Actually, the hardy changes might not be necessary now that 2.4 support is not required. [20:42] jelmer: s/_make_silent_logger/QuietFakeLogger/ [20:42] i guess [20:42] wgrant: okay [20:44] mwhudson: ah, great [20:46] wgrant: oh that's true\ [20:46] i forget exactly why hardy is different even though i was responsible for the difference [20:46] (and the bogus version number) [20:47] i have this memory of being very angry though [20:47] Heh. [20:47] What's the diff? [20:48] i think it's just not depending on python2.4-support on hardy [20:51] trunk launchpad-dependencies still depends on all kinds of python 2.4 stuff [20:51] is it too soon to drop all that? [20:51] gary_poster: hi [20:54] wgrant: can you see the landscape overview page now? [20:54] mwhudson, OT question: would bzr-explorer allow me to quickly jump between 'bzr blame', the revision number, the diff for said revision, etc.? [20:54] mars: no idea! [20:54] thumper: It's lying, but yes. [20:54] darn :( [20:54] i don't have any truck with this gui nonsense [20:54] wgrant: :) [20:54] heh [20:54] well not really [20:54] but i don't know [20:55] mars: bzr gannotate can do that [20:55] so i expect explorer can too [20:55] (or soon will) [20:55] if it builds against bzr nightlies... [20:56] there we go: sudo add-apt-repository ppa:bzr-explorer-dev/ppa [20:57] this is one of the application usage models where the command line sucks: random browsing [21:00] mwhudson: hi [21:01] mwhudson: I think we can drop all that stuff. We should convert the -2.4 to the -2.5 [21:01] gary_poster: ok [21:01] mwhudson: does that mean I should say thank you? :-) [21:03] yeah, i'll do that [21:03] gary_poster: will you review the branch? [21:03] mwhudson: absolutely. thank you! [21:03] gary_poster: around for another hour or so? [21:04] mwhudson: yes, for just under two hours [21:04] ish [21:04] cool [21:04] * mwhudson prefers summer in some ways [21:14] bigjools: Can you please review https://code.edge.launchpad.net/~wgrant/launchpad/distroseries-source-format-selection/+merge/14977 when you have time? [21:15] wgrant: is that the one where you already requested a review from me? [21:16] bigjools: It is. [21:16] wgrant: ok, sorry I've not had time to get to it yet, blame UDS [21:16] UDS does that. [21:17] And I'll have a hopefully small branch tonight to fix gina. [21:17] wgrant: I'll do my best to look at it this week, but it's most likely to be Monday [21:17] And I think that should be it. [21:26] gary_poster: do you think sed -i -e 's/python2.4/python2.5/g' debian/control is basically speaking right? [21:27] mwhudson, yeah, removing the resulting dupe python2.5-dev [21:27] mwhudson we might be specifying more than we need, but that's a diff problem for a diff day, IMO [21:27] bigjools, nice sticker [21:29] gary_poster: ok [21:29] Ursinha: yes ma'am! [21:29] hahahaha [21:34] * mwhudson is confused [21:34] seems the launchpad-dependencies that built yesterday are broken? [21:35] What's broken about them? [21:36] A saw a bug about something last night, and it looked like lots of deps might have been removed, but I didn't look at it in detail. [21:36] The following packages have unmet dependencies: [21:36] launchpad-developer-dependencies: Depends: launchpad-dependencies (= 0.57) but 0.56 is installed [21:36] Depends: launchpad-database-dependencies (= 0.57) but 0.56 is installed [21:36] Hm. That's odd. [21:36] yes [21:37] the 0.58 ones i just built installed fine though... [21:37] it is a little odd, since 0.57 seems to install fine on my karmic system [21:37] assuming it's still karmic you're talking about :) [21:38] yes [21:38] i probably buggered up something locally then, good [21:45] gary_poster: https://code.edge.launchpad.net/~mwhudson/meta-lp-deps/bye-2.4-hello-2.5/+merge/15058 [21:45] mwhudson: awesome. on call, will lock after [21:45] hooray for proposing at ??:?4:55 === abentley1 is now known as abentley [21:58] sinzui: ping [21:58] Hi thumper [21:59] sinzui: are we going to talk today? [21:59] sinzui: I have 30 minutes spare now [21:59] mwhudson: we can now use this on hardy because python2.5-support is available there, where python2.4-support was not? (ISTR python2.4-support was the issue with Hardy) [22:00] thumper: now is good [22:02] * gary_poster is confused, sigh. http://packages.ubuntu.com/karmic/python2.5-support doesn't exist. Maybe in PPA? [22:03] gary_poster: It's a virtual package [22:04] wgrant: ah ok, thank you. that makes sense [22:05] gary_poster: eg. if you apt-get install python2.5-support it will tell you that it's choosing python-support instead. [22:05] Because python-support Provides python2.5-support. [22:06] ah, so. only barely aware of virtual packages, thanks [22:33] mwhudson: approved, with note about my question [22:36] gary_poster: it's an interesting question [22:37] mwhudson, we could ask losa to verify practically, or try it on a ec2 instance [22:37] gary_poster: i think python2.5-support is actually part of hardy [22:37] so it would be a bit exciting if it couldn't be installed [22:38] heh [22:38] hm, maybe not [22:38] mwhudson: as I said in the review, if you are comfortable then I am. Otherwise, verifying somehow seems like a good idea before copying over to hardy [22:40] * mwhudson is a little confuses [22:40] http://packages.ubuntu.com/hardy/python-support is part of hardy [22:41] but not python2.x-support [22:41] :-( [22:41] so maybe we just need to depend on python-support [22:43] mwhudson, maybe so, If it were I, I must admit, I would do what had been done before, maintaining the separate Hardy branch. Who added the split? [22:43] * gary_poster looks at PPA [22:44] I think it is gone [22:44] My memory is that Francis had made the split in 0.50 [22:44] He would probably know [22:45] i made the split :) [22:45] i forget why [22:46] heh [22:46] it seems nothing on my system depends on python2.x-support [22:46] just python-support [22:46] i'll see if my package installs on hardy [22:47] (on ec2) [22:47] ok cool [22:50] am i glutton enough to try to build a new ec2 image based on the new canonical-provided hardy ami i wonder... [22:51] ec2 test ate my test run [22:51] what should I do? [22:51] philosophise about how it was meant to be [22:54] mwhudson, gary_poster: Ah, it seems that Hardy's python-support doesn't Provide python2.x-support. Since we're no longer using some awful horribly obsolete removed Python version, it's OK to just depend on python-support. [22:55] (until Lucid removes python2.5, but we'll see...) [22:55] jml: send rude mail to jeff bezos [22:55] :( [22:56] wgrant, ok cool. Sounds good to me. Thank you. Lucid: the hope is to move to 2.6 then, but as you said, we'll see :-) [22:57] gary_poster: You can't really move to 2.6 until after Lucid is released. [22:58] wgrant, yeah I know. :-) in some theoretical world we could move to Lucid and Py 2.6 at once. probably a horrible idea, in addition to being potentially difficult to coordinate [22:58] wgrant: well, actually... [22:59] supposedly we arranged to have resources to have internally supported 2.6 in Hardy if we had gotten there [22:59] Parts already exist [22:59] But almost certainly not everything we need [23:01] how can i install a .deb file with apt? [23:01] (i think that's what i want, i want to install a .deb but meet it's dependencies from the configured archive) [23:02] mwhudson: You'd have to create a repository. [23:02] mwhudson: Otherwise, dpkg -i it. [23:02] Then apt-get -f install. [23:02] The latter will resolve unmet dependencies. [23:02] ah ok [23:02] Although that doesn't help if it depends on other .debs that you don't have in a repo. [23:02] dpkg --fsck-me-harder ? [23:03] well yes, that's something i'd like to find out :) [23:03] gdebi will install a single .deb & resolve dependencies, I believe [23:03] True. [23:03] --force-depends [23:03] ? [23:03] No. That will install it unthinkingly. [23:03] --force-depends is a bit messy, you have to clean up with apt-get -f install afterwards [23:04] You want to dpkg -i, which will get it half-intsalled. [23:04] Then apt-get -f install will resolve dependencies, and finish installing the half-installed package. [23:04] wgrant: aa [23:04] Setting up bzr (1.3.1-1ubuntu0.1) ... [23:04] hmm [23:04] if you can bear a GUI, use gdebi for it I think [23:05] ajmitch: this is ec2 [23:05] that's what I was worried about [23:06] looks like it might have a cli-only mode as well, but if you have it installed now, it doesn't matter :) [23:06] hm, launchpad-dependencies nearly installs without the ~launchpad ppa now [23:07] (after changing to depend on python-support) [23:07] It should. [23:07] What is missing? geoip stuff? [23:09] yeah [23:09] specific version of psycopg2 [23:09] python-profiler [23:09] it's a shame we can't just ram that into karmic-updates (or hardy-updates) [23:09] postgresql-autodoc [23:10] are there bits there that are missing from lucid? [23:10] i don't know [23:11] ajmitch: well python 2.5 is/will be missing from lucid >:) [23:11] woo [23:11] so my packages install on hardy [23:11] from main, anyway, though I understand that you only want to use supported packages :) [23:12] just get python2.6 support in LP done quickly :) [23:13] It had more issues than we had hoped, but it certainly could be done. We will have to do some kind of careful dance for this, with the different supports for Hardy and Lucid. Not having overlapping Python support in concurrent LTS releases is hard. [23:14] * mwhudson is so happy to just have 2.5... [23:14] because you get new bzr-svn toys? [23:15] yes [23:15] and well [23:15] just feeling less like i'm living in the past [23:16] where you live, you'll always be in the future [23:17] gary_poster: new diff on https://code.edge.launchpad.net/~mwhudson/meta-lp-deps/bye-2.4-hello-2.5/+merge/15058 [23:17] mwhudson: cool, +1 :-) [23:17] gary_poster: but i think i'm going to apply "if it builds and installs it's ok" [23:18] +1 [23:19] wow, something set the commit message automatically from the changelog [23:24] * mwhudson wonders what the ppa build queue is like currently... [23:25] empty [23:25] awesome [23:31] hey [23:31] quick doctest help [23:31] say I had a string "Ran 4 tests in 0.123s" [23:31] and I don't care about the exact time [23:31] how would I write a doctest example for that ? [23:31] Ran 4 tests in ...s [23:31] thanks [23:32] there's an option you might need to turn on, depending on your test infrastructrue [23:33] jml: testtools [23:34] jml: I'm taking an early lunch to finish last nights itch [23:35] * jml files https://bugs.edge.launchpad.net/soyuz/+bug/485558 [23:35] Bug #485558: 'SourcePackage' name inconsistent with rest of Soyuz objects

[23:36] bigjools, bug 485558 [23:36] Bug #485558: 'SourcePackage' name inconsistent with rest of Soyuz objects

[23:37] finally, a bug :) [23:38] * lifeless takes a bootstrap shortcut [23:38] bigjools, it's a step forward. [23:39] jml: from the school of JFDI [23:39] jml: btw it should be a bug on registry [23:40] bigjools, we shouldn't have subprojects :P [23:40] jml: do you need to \\ escape a . that you want to be literally matches ? [23:40] s/matches/matched/ [23:40] :) [23:40] lifeless, don't know, sorry. [23:41] lifeless: it's not a regexp at all [23:41] lifeless: '... [23:41] lifeless: '...' matches an arbitrary amount of stuff [23:42] i doubt there's a way to match ... explicitly [23:45] you know [23:46] I think I'm actually going to use blueprints agian. [23:46] jml: don't you mean specifications? [23:46] bigjools, ha! [23:47] Features! [23:47] * jml has a note that says we should consider how daily builds will interact w/ RSS feeds [23:47] but doesn't really know where to put it. [23:47] oo that's an interesting idea [23:48] jml: I was going to say to add that to https://blueprints.edge.launchpad.net/soyuz/+spec/build-from-branch [23:48] but I think we need another specprint for daily builds [23:48] yeah, I was thinking that we could actually benefit a lot from splitting things up [23:49] bigjools, would you mind too much if I gardened the spec a bit on Monday? [23:49] what do you have in mind? [23:49] bigjools, looking at it, maybe splitting thing up and making subspecs maybe. [23:50] yeah sure [23:50] it does seem to have grown [23:50] I want to start dumping thoughts on the UI, and there'll be quite a few flows. [23:50] ui needs a separate blueification I reckon [23:51] yeah [23:51] I'll faceprint that on Monday [23:51] better than facepalming it I suppose