[00:42] <wgrant> al-maisan: Why do you need SPRBJ.gettitle now?
[00:43] <wgrant> Ther's an implementation in the integration branch, IIRC.
[00:43] <al-maisan> because the db-devel branch as it stands fails a test
[00:43] <wgrant> Huh.
[00:43] <al-maisan> Just doing a testfix
[00:44] <wgrant> (wireless laaaaaaaaaag)
[00:44] <al-maisan> the failing test is test_makeJob
[00:44] <al-maisan> SourcePackageRecipeBuildJob does not implement getTitle()
[00:44] <wgrant> Ahh.
[00:45] <wgrant> We need to get the other couple of unreviewed integration pieces landed.
[00:46] <wgrant> But Internet here sort of sucks.
[00:47]  * al-maisan could not agree more
[01:28] <jml> jelmer, al-maisan: my conference bag is in the hotel lobby
[01:28] <jml> al-maisan, jelmer: are either of you there?
[01:29] <thumper> oops
[01:29] <thumper> I'm not there
[01:29] <jml> ...
[01:29] <jml> thumper, hmm.
[01:29] <jml> maybe I can go there & return before mwh gets to the talk.
[01:30] <thumper> jml: do you remember monty's nick?
[01:46] <jml> problem solved
[04:17] <jtv> al-maisan, that branch I promised to refer you to: lp:~jtv/launchpad/bug-499405-translationtemplates-buildmanager
[04:18] <al-maisan> jtv: thanks, will take a look.
[09:26] <mrevell> hi
[09:52] <thumper> hey
[09:53] <beuno> hello
[13:09] <vila> hey guys, is loggerhead down or is it just me ?
[13:09] <vila> see http://bazaar.launchpad.net/~spiv/bzr/per-file-merge-hook-491711/revision/4905
[13:09] <beuno> vila, seems its down, a user just reported ut
[13:10] <beuno> in #launchpad
[13:10] <vila> rhaa, one more channel to track :-}
[13:12] <vila> beuno: thks, will wait
[14:49] <deryck> Morning, all.
[14:54] <magcius> Woah, found this: http://bitbucket.org/RonnyPfannschmidt/anyvc/
[14:55] <magcius> It's another Leaky Abstraction®, but whatever.
[15:52] <kfogel> gmb or anyone: 'make build' failing in a recently-updated branch of db-devel for us -- see http://paste.ubuntu.com/358557/.  Any ideas?  "ImportError: No module named builder.recipe"
[16:04] <kfogel> maxb: random build issue, you might have seen before?  'make build' failing in a recently-updated branch of db-devel for us -- see http://paste.ubuntu.com/358557/.  Any ideas?  "ImportError: No module named builder.recipe"
[16:04]  * kfogel some how associates "maxb" with "understands all sorts of config & build problems"
[16:21] <gmb> intellectronica, bdmurray: Here's the top 100 hot bugs that the abortive staging run produced: https://pastebin.canonical.com/26558/
[16:23] <maxb> kfogel: hah :-)   /me is grinning.  Not seen that one though
[16:23] <kfogel> maxb: we never did figure it out -- we just routed around :-).  Sigh.
[16:23] <maxb> my vague thought is: relative import which shouldn't be?
[17:02] <kfogel> jml: https://code.edge.launchpad.net/~kfogel/launchpad/509194-506514-bug-heat-db-fixes/+merge/17597   easy review, if you feel like, you know, a break from whatever you're doing :-)
[17:24] <intellectronica> looks like there's a new dependency in db-devel, on the bzr builder plugin. because of that you can no longer build a fresh branch from db-devel. anyone knows anything about this?
[17:27] <intellectronica> this is probably related to the recent build-from-branch work, b.t.w
[17:29] <bdmurray> make schema is failing for me due to sourcecode/mailman should that be there?
[17:31] <bdmurray> I think I found it
[17:45] <allenap> gmb: https://code.edge.launchpad.net/~allenap/launchpad/stop-checkwatches-hammering-products-bug-506158/+merge/17601
[17:52] <allenap> gmb: https://code.edge.launchpad.net/~allenap/launchpad/stop-checkwatches-hammering-products-bug-506158/+merge/17602
[18:55] <thumper> intellectronica: update your sourcecode dir and download cache
[19:00]  * thumper waves at the devs
[19:19] <bigjools> hello thumper
[19:25] <thumper> bigjools: hey
[19:25] <thumper> bigjools: it will be 22 degrees C here today and sunny :)
[19:25] <bigjools> bastards
[19:25] <thumper> now the conf has tarted
[19:26] <thumper> started that is
[19:26] <bigjools> thumper: I am not sure how you keep doing 30+ hour flights and remain sane
[19:27] <thumper> bigjools: you assume I'm still sane
[19:27] <bigjools> it's all relative
[19:27] <thumper> :)
[19:41] <goundy> Hi guys
[19:41] <goundy> Something new about wiki support in LP guys ?
[19:46] <jamalta> Hi, what's the proper way to reload the server so browser code gets re-executed?
[19:46] <jamalta> Is closing and re-running make run the only way?
[19:55] <thumper> jamalta: I think so
[19:55] <thumper> it is what I do
[19:56] <jamalta> thumper: alright thanks.. annoying but at least it works
[19:56] <thumper> yeah
[19:56] <thumper> could be worse
[19:56] <jamalta> heh yeah
[19:56] <jamalta> thanks for the input
[20:11] <wgrant> bigjools: Hm, publisher is broken? Did we delete something that we shouldn't have?
[20:11] <bigjools> wgrant: tell you in a sec
[20:11] <bigjools> fixing it
[20:13] <bigjools> wgrant: someone managed to copy from a private to a non-private PPA
[20:14] <bigjools> and p-a thought it was a delayed copy
[20:14] <wgrant> bigjools: Ah, good. So it's not something terrible.
[20:15] <bigjools> I'm guessing the API was used since he couldn't see it in the ui
[20:15] <wgrant> So the PPA was actually still private?
[20:17] <bigjools> in what terms?
[20:18] <bigjools> the target ppa was not private and it had a packageupload with no signing key
[20:22] <jamalta> is there a list of user logins for dev?
[20:22] <jamalta> can't find it in the wiki
[20:31] <wgrant> Hm, private source builds are sadly broken in trunk.
[20:31] <wgrant> I wonder why there are no tests for that.
[20:38] <allenap> rockstar: Hello, are you versed in the ways of the IJob?
[20:38] <rockstar> allenap, I believe I can be classified as "brown belt"
[20:38] <allenap> rockstar: I wondered why the job.reason column is unused?
[20:39] <allenap> rockstar: gmb and I are wondering if we can bend it to our will.
[20:40] <rockstar> allenap, so I don't know the reason (no pun intended)
[20:40] <wgrant> jml: Why was _sendFileToSlave moved from Builder to BuilderSlave?
[20:41] <allenap> rockstar: :) Do you think abentley will know? I'd like to check before I either use it or create a new table.
[20:41] <bigjools> wgrant: I feared that a lot of stuff would be broken after last week
[20:42] <bigjools> what's bust?
[20:43] <rockstar> allenap, abentley is the architect of the job stuff, and I remember him telling me the purpose, but I don't remember the specifics.
[20:43] <allenap> rockstar: Okay, I'll email him about it. Thank you for your help :)
[20:43] <mwhudson> allenap: is there anything in commets.sql about it?
[20:44] <allenap> mwhudson: Tip top idea. "The reason the job was created, if applicable" (thanks to gmb).
 the target ppa was not private and it had a packageupload with no signing key
[20:53] <maxb> Isn't that what happens if you copy a synced-from-Debian upload from the ubuntu primary archive to a public PPA? And that works fine.
[20:54] <bigjools> maxb: well in this case the problem is that in combination with the private archive it came from, it triggered the bug
[20:54] <maxb> ok - just saying that public->public with no signing key works just fine
[20:55] <bigjools> ok thanks
[20:56] <jml> wgrant, because AIUI, BuilderSlave is the LP-side representation of the slave
[20:56] <jamalta> Well, since I haven't received a reply I'm assuming it doesn't exist..
[20:57] <jml> wgrant, sendFile seems like a meaningful operation on that object
[20:57] <jamalta> mind if I make a page on dev.launchpad.net that lists login for dev accounts?
[20:57] <jml> jamalta, sounds like a good idea to me
[21:01] <wgrant> jml: Somwhat, but it's marked private, accessed from other places, and there is another implementation of that interface.
[21:01] <jml> wgrant, it was already accessed from places other than Builder
[21:01] <wgrant> I think there's a README somewhere in the tree that lists dev redentials.
[21:02] <jml> wgrant, multiple implementations of the BuilderSlave interface?
[21:03] <wgrant> lib/canonical/launchpad/pagetests/README.txt
[21:03] <wgrant> jml: Yes. There's RecordingSlave, which is actually used by builddmanager.
[21:03] <jamalta> wgrant: ah ok
[21:04] <jml> wgrant, then it seems to me that the interface needs to be tweaked a bit
[21:04] <jamalta> wgrant: is it ok if i post that ont he wiki? i guess that's where i should have looked, but it will make it easier for people not to come here asking like idiots (aka, me ;)
[21:05] <wgrant> jamalta: I dont like duplicatng information. Perhaps reference the file alongside the existing provided credentials on the wiki.
[21:05] <wgrant> jml: Probably.
[21:06] <wgrant> A lot of interfaces need tweaking :(
[21:06] <jamalta> wgrant: understood
[21:06] <gmb> jml: Could you take a squizz at https://code.edge.launchpad.net/~gmb/launchpad/add-bugheatjob-table/+merge/17620 when you have  a second please? It's a branch to add a BugHeatJob table to the DB.
[21:06] <jamalta> should i still make a table of the current users and reference the file as well?
[21:07] <jml> gmb, the diff is empty
[21:07] <wgrant> jamalta: I think just reference the file.
[21:07] <jamalta> wgrant: ok
[21:07] <gmb> jml: That's because I'm an idiot.
[21:08] <gmb> jml: I've re-pushed with the change actually committed... don't know when the diff will update.
[21:08] <jml> gmb, "soon"
[21:08] <jml> gmb, thanks.
[21:08] <gmb> Kewl
[21:08] <jamalta> wgrant: thanks
[21:11] <jml> gmb, got it
[21:11] <jml> gmb, the approach that the branch jobs take is to have one table for BranchJobs and have an enum type & json metadata
[21:12] <jml> gmb, rather than a new table per type of job
[21:12] <jml> gmb, what do you think about renaming the table to BugJob?
[21:15] <gmb> jml: wfm, sure.
[21:17] <gmb> jml: Thus: https://pastebin.canonical.com/26732/
[21:17] <gmb> ?
[21:19] <jamalta> running a test, i got a "2 import policy violations" notice.. but this isn't from anything i changed. Is it safe to ignore this, or could it be something I broke?
[21:19] <jamalta> the test in question is lib/lp/soyuz/stories/ppa/xx-copy-packages.txt
[21:25] <gmb> jml: Fixed in the branch and pushed.
[21:30] <maxb> https://edge.launchpad.net/~launchpad/+archive/ppa/+build/1452706 https://edge.launchpad.net/~launchpad/+archive/ppa/+build/1452707 ... Could a Canonical person hit the retry button on those? Unfortunately my "uploader" access apparently doesn't imply "retry" access :-/
[21:31] <bigjools> maxb: done
[21:31] <maxb> thanks
[21:31] <bigjools> fwiw there's a bug open about that
[21:32] <bigjools> right, bed time, g'night
[21:32] <gmb> jml: I've assigned the review to you for posterity, FWIW.
[21:32] <maxb> Another interesting edge case is that uploader doesn't implied "can delete" - though that one's more up for debate
[21:35] <kfogel> jml: (we redid this trivial db patch to be from db-devel instead of devel; this supersedes the earlier merge proposal which you were asleep during anyway :-) ):
[21:35] <kfogel> https://code.edge.launchpad.net/~kfogel/launchpad/509194-506514-bug-heat-db-fixes/+merge/17625
[21:47] <wgrant> We beokr more than I thought :(
[21:48] <wgrant> (that was 'broke' with about 20s lag)
[22:09] <EdwinGrubbs> salgado-afk: ping
[22:17] <jml> hello? what?
[22:17] <jml> kfogel, gmb: which reviews do I need to look at?
[22:17] <kfogel> jml: one sec
[22:17] <kfogel> https://code.edge.launchpad.net/~kfogel/launchpad/509194-506514-bug-heat-db-fixes/+merge/17625
[22:17] <kfogel> jml: there you go
[22:17] <kfogel> jml:  there will be another Really Easy one in about five minutes
[22:18] <jml> I've just done my LCA talk
[22:18] <gmb> jml: The one you were looking at before: https://code.edge.launchpad.net/~gmb/launchpad/add-bugheatjob-table/+merge/17620
[22:18] <gmb> If you pleases.
[22:18] <jml> kfogel, Gabriella Coleman's talk was great. thanks for the recommendation.
[22:19] <wgrant> It was really good, yeah.
[22:19] <jml> BjornT is doing his talk right now.
[22:19] <kfogel> jml: glad you liked it!  I'll tell her she pleased at least one person :-).
[22:19] <jml> BjornT, sorry for reviewing db patches while you are talking -- these guys are sprinting
[22:19] <kfogel> wgrant: make that two people
[22:19] <kfogel> wgrant: she has such a different perspective from the coders, it's really enlightening
[22:19] <kfogel> heh
[22:19] <wgrant> Yup.
[22:22] <jml> gmb, approved. You need an approval from stub.
[22:23] <jml> die sample date die die
[22:23] <jml> data.
[22:23] <gmb> jml: Thankee.
[22:24] <jml> kfogel, "The heat bin in which this bugtask appears" -- that doesn't seem correct any more, given the current design
[22:24] <jml> kfogel, it's more like a score than a bin, right?
[22:25] <jml> hmm. maybe I misunderstand.
[22:25] <kfogel> jml: interesting point.  let me ask others here, who know more about the bin terminology
[22:28] <kfogel> jml: I think it's still correct -- that is, it seems there's this other thing going on whereby heat scores may be grouped into various bins, and that's what this is about.  That may be obsolete now (gmb says, as allenap looks slightly sick next to him), but in any case their possible obsolescence is not related to this patch.
[22:28] <jml> kfogel, ok.
[22:29] <jml> kfogel, approved.
[22:30] <jml> kfogel, gmb: stub's pretty prompt w/ db reviews, but I'll poke him later today to have a look at these.
[22:30] <gmb> jml: Thanks
[22:31] <kfogel> jml: thanks.  We'll send the next one to stub (btw, I'm working with adeuring not gmb -- dunno this gmb guy, never heard of him; his name is suspiciously short).
[22:31] <jml> kfogel, there are two db patches
[22:31] <jml> kfogel, he has a middle name, that makes him extra trustworthy in alabama
[22:32] <kfogel> jml: we're about to submit the next one, actually
[22:32] <jml> kfogel, although judging by his accent, I suspect he's from north of the Mason-Dixon line.
[22:32] <kfogel> jml: gmb said the "Welcome to Alabama" sign on the highway coming in from Atlanta airport (in Georgia, the next state over) had bullet holes in it.  Yee-haw!
[22:39] <kfogel> jml: https://code.edge.launchpad.net/~kfogel/launchpad/505845-affects-me-too-from-dups/+merge/17633
[22:39] <jml> kfogel, looking
[22:40] <kfogel> jml: thanks
[22:41] <kfogel> jml: afk for a bit, will be back to answer questions after
[22:53] <Ursinha> ae
[22:53] <jml> ei
[22:53] <Ursinha> this rain won't let me down!!
[22:57] <jml> down from where?
[22:57] <jml> kfogel, reviewed. happy to talk further.
[22:58] <kfogel> jml: we're reading now
[22:58] <jml> kfogel, \o/
[22:59] <maxb> I've just been trying to update-sourcecode, and I've noticed a couple of things:
[22:59] <maxb> (1) it breaks if you have bzr-git installed
[22:59] <maxb> (2) to make it work on lucid we'll need to have bzr in the launchpad PPA _or_ make it run with python 2.6
[23:00] <jml> maxb, that first one surprises me.
[23:01] <maxb> jml: It's because bzr sees the test git repositories within dulwich as branches, and update-sourcecode tries to delete them as obsolete :-)
[23:01] <mwhudson> aaraaarsadhjk
[23:01] <mwhudson> let's not use find_branches
[23:01] <jml> maxb, look, you broke mwhudson
[23:02] <jml> maxb, I hope your sorry.
[23:02] <jml> you're. dammit.
[23:02] <Ursinha> jml: it's raining like it's the end of the world here
[23:02] <maxb> Yeah, bzr-<other-vcs> is cool, but I wish it could read your mind to figure out when to run :-)
[23:02] <jml> Ursinha, :(
[23:03] <maxb> You can imagine the interesting edge cases I run into with bzr-svn installed, where my homedir is a svn wc.....
[23:06] <maxb> So.... soliciting thoughts on which is the lesser evil? Add bzr to the set of things we have to rebuild to work with the Launchpad blessed Python version, or run devscripts stuff under a possibly different python to launchpad?
[23:07] <elmo> *blink* devscripts has python things in it?
[23:07] <mwhudson> maxb: why does bzr need to be rebuilt?
[23:07] <maxb> Because update-sourcecode imports bzrlib
[23:08] <elmo> maxb: I don't see any python in devscripts?
[23:08] <maxb> I mean lib/lp/devscripts/
[23:08] <mwhudson> maxb: i think update-sourcecode being run with the system /usr/bin/python
[23:08] <mwhudson> would be fine
[23:08] <maxb> uh, lib/devscripts/
[23:09] <mwhudson> possibly not the best choice of package name i guess
[23:10] <maxb> Oh, this is going to be pulling on a thread which unravels ridiculously, isn't it :-/
[23:10] <jml> if only we had a vcs that supports renaming
[23:10] <wgrant> Wasn't devscripts stuff running on 2.5 long before LP was ported?
[23:10] <jml> maxb, everything on LP worth doing is like that. stick with it.
[23:10] <wgrant> So it shouldn't be too hard.
[23:11] <elmo> jml: are you crazy?  eventually the whole cloth will come apart if you do that!
[23:11] <maxb> I'm just worrying about any deps wanted from eggs for devsscripts
[23:11] <wgrant> (disclaimer: Launchpad difficulty ratings may differ wildly from generally accepted standards.)
[23:12] <jml> elmo, but think of the long, useful (dare I say light-weight & flexible) yarn
[23:14] <maxb> mm.... do nothing in devscripts looks to be using any launchpad custom code, but some of the launcher scripts in utilities do use it after an "import _pythonpath"
[23:15] <kfogel> jml: responded in merge proposal.  we're about to enter sprint wrapup, so I may not be able to respond till later
[23:17] <maxb> But update-sourcecode can't do that because it is run from rocketfuel-get, which is run from rocketfuel-setup
[23:23] <jml> hi ho
[23:23] <jml> I'm back
[23:23] <jml> kfogel, I'll take a look.