[01:10] <sabdfl> stubarooney!
[01:10] <sabdfl> it's release day
[01:11] <stub> Morning
[01:11] <stub> What are we releaseing?
[01:11] <stub> This is #launchpad ;)
[01:14] <jordi> hello
[01:15] <ddaa> anybody with sufficient privs, please take a large axe and kill whatever is currently blocking pqm
[01:15] <ddaa> lifeless: ^
[01:28] <stub> ddaa: done
[01:29] <ddaa> aaaaargh
[01:30] <ddaa> CVS IS THE DEVIL'S DOING
[01:30] <ddaa> lifeless: bouncing cvs spawn-connect delay, again
[01:33] <ddaa> thanks stub, fine-grained kills like that are good for diagnostic
[01:37] <stub> I'm only allowed to follow my script ;)
[01:41] <stub> spiv, lifeless, jamesh: Can someone confirm that AU passport holders don't need visas for Canada? 
[01:58] <stub> ddaa: If hangs are triggered by us running /usr/bin/cvs, can we invoke a wrapper instead that kills the subprocess either on a timeout or inactivity? Might be a nice simple way of integrating this into the test suite and would be useful for other tests that spawn subprocesses
[01:58] <ddaa> mh
[01:59] <ddaa> it's would fix part of the problem
[01:59] <ddaa> that would prevent the test suite from occasionally blocking indefinitely, but it would still fail occasionally.
[02:00] <stub> Could a simple wrapper do the trick, or would it need to pass signals through too?
[02:00] <ddaa> well, it would need to be able to pass sigterm and sigkill :)
[02:01] <ddaa> i'm not too sure how process reaping work, tbh
[02:02] <ddaa> OMG it's 2am
[02:03] <ddaa> need to go to bed, stub, talk about that tomorrow
[02:03] <stub> Hmm... extra points to extract traces and other debugging info from the subprocess before you kill it ;)
[02:03] <stub> ddaa: Night ;)
[02:06] <lifeless> guys, PQMRobustness
[02:06] <lifeless> its a spec
[02:06] <lifeless> done before and in brazil
[02:06] <lifeless> no need to invent.
[02:07] <lifeless> (as in.. wrappers etc) - the focus would be best served in finishing that spec
[02:07] <lifeless> as for the specific hang, giving cscvs's pserver implementation a timeout + retry on timeout facility would probably fix the hangs completely.
[02:22] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  fixing style and robustness in getWorkedBuildRecords methods (patch-2637: celso.providelo@canonical.com)
[02:55] <dilys> Merge to rocketfuel@canonical.com/launchpad--production--1.36: Cherry pick shipit searching into production 1.36 (patch-6: guilherme.salgado@canonical.com)
[03:11] <dilys> Merge to rocketfuel@canonical.com/cscvs--devel--1.0: [trivial]  increase PServerHelper.startServer delay to 200ms (patch-115: david.allouche@canonical.com)
[03:26] <stub> Any reasons why I would be getting 'Timeout, server not responding' messages from baz archive-mirror when I can access chinstrap just fine?
[03:30] <lifeless> SteveA: lp meeting tonight, is 'at release party getting happy' a good excuse ?
[03:31] <lifeless> stub: archive-location --all-locations
[03:34] <stub> baz: unrecognized command (archive-location)
[03:35] <stub> # This ini file controls the configuration for one archive -
[03:35] <stub> # The archive is named by the same name as the name of this file
[03:35] <stub> url=/home/stub/.arch-archives/stuart.bishop@canonical.com master
[03:35] <stub> when_unsigned=error
[03:35] <stub> url=sftp://chinstrap.warthogs.hbd.com/home/warthogs/archives/stuart.bishop@canonical.com priority=50
[03:35] <stub> allowed_fingerprints=C7ECC365AB6F255E1EB9BA1701FA998FBAC6374A
[03:35] <stub> url=sftp://chinstrap/home/warthogs/archives/stuart.bishop@canonical.com
[03:35] <stub> Hmm.... managed to get two chinstraps in there. Which one is better to nuke - priority=50 or the other one?
[03:43] <stub> lifeless: nope...
[03:45] <lifeless> stub: can you ssh to both 'chinstra' and 'chinstrap.warthogs.hbd.com'
[03:45] <stub> lifeless: yes
[03:46] <lifeless> funkey monkey
[03:46] <stub> I also know baz is actually connecting, because I need to clear up lock files after the aborted runs
[03:46] <lifeless> erm, then it sounds like a networking issue
[03:47] <lifeless> i.e. adsl packet blackholine
[03:49] <stub> I'm not seeing any other network issues, including with other ssh connections to the datacentre
[03:49] <stub> It *did* mirror earlier today, so I know this version of baz can do it
[03:49] <aredhelrim> hi.
[03:54] <stub> $ sftp chinstrap
[03:54] <stub> Connecting to chinstrap...
[03:54] <stub> sftp> put debian-keyring.gpg.bz2
[03:54] <stub> Uploading debian-keyring.gpg.bz2 to /home/stub/debian-keyring.gpg.bz2
[03:54] <stub> debian-keyring.gpg.bz2                          0%    0     0.0KB/s - stalled -
[03:54] <stub> lifeless: But scp works happily.
[03:55] <stub> ooh... no. it stalled too
[03:55] <stub> I might go reboot some stuff
[04:12] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Make gina cope with database exceptions and simplify locking (patch-2638: stuart.bishop@canonical.com)
[04:17] <stub> Seems to be working fine now. Don't know if my system needed the reboot due to the last kernal upgrade or my router was throwing a wobbly.
[04:58] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=lifeless]  importd-archivelocation, use branch urls instead of registered archive name, massive importd cleanup (patch-2639: david.allouche@canonical.com)
[05:05] <jamesh> stub: could you tell me what product(s) bug 1457 and 1458 are assigned to? (they are private)
[05:05] <Ubugtu> Error: I cannot access this bug
[05:08] <stub> jamesh: They are both launchpad bugs
[05:09] <jamesh> stub: is "Launchpad Developers" subscribed to either?
[05:10] <jamesh> I get a "you don't have permission to access this page" error
[05:29] <stub> jamesh: nope. Just kiko & cprov.
[05:30] <jamesh> that's weird
[05:30] <stub> jamesh: This is desirable behavior for launchpad to be blocked. There might be a bug in that administrators can't fix it.
[05:31] <jamesh> okay.  I was just asking because cprov asked me about those two bugs in another one
[05:31] <stub> I opened bug 3097 on it
[05:31] <Ubugtu> Error: I cannot access this bug
[05:54] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Gina love (patch-2640: stuart.bishop@canonical.com)
[06:32] <Keybuk> guys, why is the launchpad authentication stuff down?
[06:32] <Keybuk> /!\The authentication database is temporarily unavailable. Anonymous access only.
[06:32] <stub> because I need to rebuild the full text indexes to support a shipit update
[06:32] <Keybuk> on the ubuntu wiki
[06:33] <Keybuk> you know we're kinda trying to release, right?
[06:33] <stub> Yup. And all those lovely CDs need to be mailed out too.
[06:33] <stub> Anyway... just about done.
[06:51] <jamesh> stub: is staging still having trouble?
[06:52] <stub> thats weird...
[07:18] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Gina love (patch-2641: stuart.bishop@canonical.com)
[07:24] <stub> jamesh: staging is back up. I've got an odd exception in the logs I need to look into further though.
[07:29] <jamesh> looks like the RTL text entry in rosetta is working nicely in staging
[07:52] <srfrog> what is this
[07:52] <srfrog> all bots?
[08:12] <LetterRip> Hi - I'd like to add Blender to the launchpad but I'm apparently getting something wrong
[08:12] <LetterRip> https://launchpad.net/products/blender/+series/2.3/+source
[08:12] <LetterRip> is the page I'm trying to fill out
[08:13] <LetterRip> here is the breezy ubuntu page
[08:13] <LetterRip> http://packages.ubuntu.com/breezy/graphics/blender
[08:13] <LetterRip> here is the page on blender cvs
[08:13] <LetterRip> http://projects.blender.org/cvsx/?group_id=9
[08:16] <spiv> LetterRip: And you're getting an error?
[08:17] <LetterRip> the current released version is blender 2.37a (2.40-alpha1 tommorrow)
[08:17] <LetterRip> yes,
[08:17] <LetterRip> I'm not sure what information it want
[08:18] <LetterRip> ie Source Package - 
[08:18] <LetterRip> blender (2.37a-1ubuntu1)
[08:18] <LetterRip> or blender
[08:18] <LetterRip> or blender 2.37a ?
[08:19] <jamesh> LetterRip: information about Ubuntu packages will be automatically handled by Launchpad (very soon)
[08:19] <LetterRip> ok so just leave it?
[08:19] <jamesh> LetterRip: the /products/blender page is for tracking the upstream product
[08:19] <jamesh> (i.e. before it gets packaged)
[08:20] <LetterRip> i'm sorry could you post the complete path?
[08:22] <LetterRip> er nevermind
[08:38] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Gina love (patch-2642: stuart.bishop@canonical.com)
[09:36] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: fix fmt:text-to-html formatting issues, r=salgado (patch-2643: james.henstridge@canonical.com)
[10:09] <Kinnison> Morning
[10:10] <Kinnison> Congratulations
[10:37] <silbs> hey guys
[10:38] <Kinnison> Good morning miss
[10:38] <silbs> shipit is down. Is there something going on w/ LP db?
[10:40] <silbs> well, some of it works, some of it doesn't.  I guess not a db problem, but a code problem.  Is there anyone here that can help or do I need to wait till Australia or Brazil wakes up?
[10:41] <Kinnison> Probably brazil
[10:41] <Kinnison> Although SteveA might be able to help
[10:44] <silbs> I guess having the system non-operational on release day is one way to avoid those pesky CD requests :(
[10:48] <jordi> worked for me right now
[10:48] <jordi> I wonder what % of requests are not working
[10:52] <Kinnison> Can someone please explain to me what the fuck a 'Schema' is and how I populate one?
[10:53] <silbs> jordi: you were able to place an order (or view your existing one)?
[10:54] <jamesh> Kinnison: a schema is an interface
[10:54] <Kinnison> jamesh: right, so what the hell is distrorelease.components meant to be and how do I populate it with what I want?
[10:55] <jamesh> Kinnison: the form generation machinary can take use an interface as a schema, generating the correct HTML elements, and applying the validators on the input
[10:55] <Kinnison> urgh
[10:55] <Kinnison> so why is it in the database?
[10:55] <jamesh> it isn't (if you are talking about what I'm talking about)
[10:56] <Kinnison> I'm talking about     components = ForeignKey( dbName='components', foreignKey='Schema', notNull=True)
[10:56] <Kinnison> from distrorelease.py
[10:57] <jamesh> that definition says that it is a pointer into a table called schema
[10:58] <jamesh> see database/schema.py
[11:00] <Kinnison> Right
[11:01] <jordi> silbs: I was able to place an order, yes
[11:01] <jamesh> there only seems to be an interface for getting existing schemas
[11:01] <silbs> jordi: okay, thanks
[11:01] <jordi> I can see it now
[11:02] <SteveA> morning
[11:05] <SteveA> silbs: still got problems with shipit?
[11:06] <Kinnison> Okay, can someone explain if I want a schema, or if I want to pretend distrorelease.components doesn't exist for now and just get on with the uploader?
[11:07] <SteveA> Kinnison: want to talk about it on the phone?
[11:07] <silbs> SteveA: yes.  Orders can be placed, marilize can't do her job though. The page she needs to access to process orders says "Programming error. A server error occurred."
[11:07] <Kinnison> SteveA: please
[11:07] <Kinnison> SteveA: on my mobile
[11:07] <SteveA> silbs: please /msg me the urls i need to look at to see the problem, and i'll look into it
[11:35] <SteveA> Kinnison: are there any cronjobs / scripts running in production that are essential for your current work?
[11:35] <Kinnison> Gina must have completed before we can open dapper
[11:35] <Kinnison> Other than that, everything I do will be on drescher
[11:35] <Kinnison> (which needs full db access)
[11:35] <SteveA> do you know if gina does anything to karma
[11:36] <Kinnison> I don't believe she touches karma at all
[11:36] <SteveA> ... full DB access counts as production
[11:36] <SteveA> do you know if gina does anything to Person ?
[11:36] <Kinnison> She creates people if they don't already exist
[11:36] <SteveA> hmm
[11:36] <SteveA> i'm seeing various timeouts
[11:36] <SteveA> don't know why just at the moment
[11:36] <SteveA> i won't kill gina.
[11:36] <Kinnison> stub set her commit rate to be 1000 I think
[11:37] <Kinnison> (IMO that's far far far too high)
[11:37] <SteveA> high being which way?
[11:37] <SteveA> fine grained?  coarse?
[11:37] <Kinnison> too large
[11:37] <Kinnison> to coarse
[11:37] <SteveA> aha
[11:37] <SteveA> yeah, that could be it.
[11:37] <SteveA> i have no idea how to change it, though
[11:39] <Kinnison> stub or elmo would know
[11:39] <SteveA> is it safe to kill gina, and then restart?  is there a lot of state that would be lost?
[11:39] <Kinnison> At a commit rate of 1000, up to 1000 imports may be lost
[11:40] <Kinnison> but tbh if gina is causing transaction problems, I'd kill her and set her commit rate way down
[11:40] <SteveA> okay, thanks
[11:40] <Kinnison> She's slower and spews more at smaller commit sizes
[11:40] <Kinnison> which is probably why stub did it
[12:03] <lbm> do it need to submit my gpg key to ubuntu's keyserver?
[12:04] <lbm> or can i use any other?
[12:04] <SteveA> you need to submit a gpg key only to sign the code of conduct
[12:04] <Kinnison> Any keyserver should do. Allow time for your key to propogate
[12:05] <Kinnison> SteveA: Or if you're expecting to upload to dapper
[12:05] <SteveA> aha... another use at last!
[12:05] <Kinnison> SteveA: see the launchpad list
[12:05] <lbm> Kinnison: thanks, just sent my key to keyserver.net, i'll wait a couple of hours
[12:06] <lbm> maybe the error message (if key wasn't found) should be more precise
[12:13] <ajmitch> hopefully not too many would use sign-only keys
[12:13] <ajmitch> upload rights will be controlled by group membership on launchpad?
[12:16] <Kinnison> ajmitch: yes
[12:17] <Kinnison> ajmitch: there are two teams
[12:17] <Kinnison> ajmitch: ubuntu-core-dev and ubuntu-dev IIRC
[12:17] <ajmitch> great, no more sending keys to elmo
[12:17] <ajmitch> yes, I'm in both now
[12:17] <Kinnison> You're a 'main' uploader?
[12:17] <ajmitch> ye
[12:17] <ajmitch> s
[12:17] <Kinnison> right
[12:17] <ajmitch> was approved a few weeks ago
[12:17] <Kinnison> right
[12:18] <ajmitch> a few MOTUs can upload to main now
[12:20] <jamesh> I've got a branch that should give a proper error message in the sign-only key case
[12:30] <sabdfl> "just a MOTU"? thin, thin ice there, methinks
[12:30] <ajmitch> heh
[12:31] <sabdfl> SteveA: is it possible to do a traverse-and-redirect? in FooNavigation?
[12:32] <sabdfl> easy, even?
[12:32] <SteveA> what exactly is that?  can you give some examples perhaps?
[12:32] <sabdfl> so, i want to be able to traverse foo/bar and i know what bar is
[12:33] <sabdfl> but usually bar lives at /bars/bar/
[12:33] <Kinnison> sabdfl: I'm not scared, I'm not afraid, I'm writing the code which can accept/reject his packages
[12:33] <sabdfl> so i want to allow a link to foo/bar and that page will redirect to /bars/bar
[12:33] <SteveA> does saying redirection('bar', '/bars/bar') work?
[12:33] <SteveA> if it doesn't, i think it should and it is a bug.
[12:33] <sabdfl> "just a MOTU" suggests a pecking order, and i don't think thats accurate
[12:34] <sabdfl> small nitpick
[12:34] <sabdfl> no offence intended, by either of us, i'm sure ;-)
[12:34] <sabdfl> SteveA: but bar is not a string, its a name
[12:34] <sabdfl> i want to get the object, then to a request.redirect(canonical_url(bar))
[12:35] <SteveA> is 'foo' a stepthrough thing, like '+bug' ?
[12:35] <sabdfl> no
[12:35] <sabdfl> foo is a build
[12:35] <sabdfl> and a build produces binaries
[12:35] <SteveA> i see.  you have foo, a Foo, and you have FooNavigation.
[12:35] <sabdfl> so foo/bar should traverse to the binary called bar from build foo
[12:35] <SteveA> got it
[12:35] <sabdfl> BUT
[12:36] <sabdfl> i really want the person to view that page at /distros/ubuntu/hoary/i386/bar/3.2-1
[12:36] <sabdfl> so, i currently can view the page at EITHER location
[12:36] <sabdfl> same page
[12:36] <sabdfl> same object
[12:36] <sabdfl> bar/+index
[12:36] <sabdfl> but i rather want to redirect
[12:36] <SteveA> so, the result of $foo/$bar is a redirection to /bars/$bar
[12:37] <SteveA> and here, i'm using $ to indicate something that varies
[12:37] <jamesh> SteveA: btw, I think there is a bug in Navigation._combined_class_info() -- it looks like it walks __mro__ in the wrong direction
[12:37] <SteveA> jamesh: nice ;-)  i'll look into it
[12:37] <sabdfl> jamesh: can you give us a schedul-o-matic update please?
[12:37] <sabdfl> SteveA: yes
[12:37] <SteveA> sabdfl: 1 min...
[12:40] <SteveA>   from canonical.launchpad.webapp.publisher import RedirectionView ;   (in class FooNavigation) def traverse(self, name): return RedirectionView('/bars/%s' % name, self.request)
[12:40] <SteveA> try that
[12:40] <SteveA> it is using RedirectionView, which is currently internal to how the Navigation works, but is fine to use outside of that.
[12:42] <SteveA> jamesh: i see... it walks __mro__ from most specific to least specific, but destructively updates the dict
[12:42] <SteveA> so that the lest specific things overwrite the most specific
[12:42] <SteveA> thanks, i'll write a test and fix it
[12:42] <jamesh> sabdfl: I've been working on it a bit slowly.  I've got the basics in place but need to do more testing and finish off some of the details
[12:43] <SteveA> jamesh: did you speak with lifeless about it recently?
[12:43] <jamesh> SteveA: no
[12:43] <SteveA> i'd like to talk through the design and issues with you sometime
[12:44] <SteveA> i have a bzr meeting in 15 mins though, but maybe we can talk in 55 minutes, 20 mins before the launchpad meeting?
[12:44] <sabdfl> jamesh: ok. in what format are you storing a "meeting"? i figure we will need to pre-seed it with a bunch of them
[12:45] <sabdfl> for example, on the days when LP guys are there during the distro week, we want to make sure each distro guy meets formally for an hour with each LP guy
[12:45] <sabdfl> so kinnison can show off Soyuz, Bradb can show off Malone, Carlos Rosetta etc
[12:45] <sabdfl> like a matrix of meetings
[12:45] <sabdfl> easiest to pre-gen those and pre-seed the system
[12:46] <jamesh> I was working off a list of Specification records, but I suppose we'd need to handle spec-less meetings
[12:47] <SteveA> there should be *some* output from such meetings
[12:47] <SteveA> if only a paragraph saying "foo and bar met, and we discussed baz"
[12:48] <SteveA> otherwise there's no way of understanding what's gone on later
[12:50] <ajmitch> sabdfl: by distro guy, you mean canonical people on the distro team? or will the rest of us also get some time?
[12:51] <sabdfl> jamesh: the spec list is just one input to the "what meetings can we have today" algorithm
[12:51] <sabdfl> the other input should be "what meetings are already scheduled"?  and that is not in LP
[12:52] <sabdfl> the output should be "what meetings we could have"
[12:52] <sabdfl> we need to be able to combine the "already scheduled" and the "now recommended" meetings to see how the day would look
[12:52] <sabdfl> then add a few manual meetings, perhaps "confirm" some recommended meetings by moving them to the "scheduled" list
[12:53] <Kinnison> sabdfl: I'd like for my meetings with the distro guys to be slightly shaped
[12:53] <sabdfl> and thus iterate towards a final scheduled list for the day
[12:53] <Kinnison> sabdfl: In particular, my meetings with colin, elmo, infinity, lamont and mdz should clearly be longer
[12:53] <sabdfl> Kinnison: if james stores them as xml files, you could shape them yourself
[12:53] <sabdfl> jamesh: unless you want to record all of this in the db. but i suspect you could work more efficiently with your own local xml store
[12:54] <Kinnison> sabdfl: FYI I've almost finished page 5 of 10 of these upload checks
[12:54] <Kinnison> sabdfl: will you be coming in for show+tell today?
[12:54] <sabdfl> cool. i now have nice initial pages for all of the source package things, starting on binaries
[12:54] <sabdfl> and i have a plan for fast effective SEARCH
[12:55] <SteveA> sabdfl: does the navigation stuff i pasted in do what you need?
[12:55] <sabdfl> Kinnison: yes
[12:55] <Kinnison> sabdfl: rock on!
[12:56] <sabdfl> SteveA: sort of
[12:56] <sabdfl> but i don't see why it needs to be that complex
[12:56] <sabdfl> cant it just be:
[12:56] <Kinnison> SteveA: Are the librarian interfaces available via getUtility() ?
[12:56] <sabdfl>  traverse(name):
[12:57] <sabdfl>     obj = self.context.findName(name)
[12:57] <sabdfl>      self.request.redirect(canonical_url(name))
[12:57] <sabdfl>     return None
[12:57] <SteveA> sabdfl: it needs to be that complex, because it isn't something supported by the Navigation stuff right now.  Now that you have identified a need for it, i can add a simple way to redirect from there.
[12:57] <sabdfl> ok, its not so urgent that i will kludge it now
[12:57] <sabdfl> thanks
[12:57] <sabdfl> stepping afk
[12:58] <SteveA> my question was, does its result do what you need
[12:58] <SteveA> i need to know that, in order to write the correct behaviour into the navigation
[01:00] <jblack> jjj
[01:00] <Kinnison> spivtastic
[01:01] <Kinnison> spiv: are the librarian interfaces available as zope utilities?
[01:02] <ddaa> ticket tracker sucks
[01:02] <ddaa> https://launchpad.net/products/launchpad/+ticket/33/+editpriority
[01:02] <ddaa> 404
[01:02] <Kinnison> arsepint, now I need an ISourcePackageFilePublishingSet
[01:02] <Kinnison> sabdfl: will you come in for lunch?
[01:03] <sabdfl> Kinnison: yes
[01:03] <sabdfl> Kinnison: try to avoid Sets
[01:03] <Kinnison> sabdfl: Right, well I'm gonna be going to tesco soon, so when you reach a convenient stopping place I suggest you head over
[01:03] <sabdfl> usually you can just have a SourcePackage.publishedfiles
[01:04] <Kinnison> sabdfl: I need something that gives me a librarian file alias when given a distribution and a filename
[01:04] <sabdfl> distribution.getFile(filname)
[01:04] <Kinnison> sabdfl: I guess
[01:04] <sabdfl> much better
[01:04] <Kinnison> back later
[01:04] <sabdfl> Kinnison: even better is distribution.getFileByName(filename)
[01:05] <SteveA> right... the ContentSets are entry points into the object model.  But, you need only so many entry points.
[01:05] <SteveA> Don't go creating a new entry-point where there's a perfectly good way to get where you need to be, already there.
[01:15] <ddaa> bug 3103
[01:15] <Ubugtu> Malone bug #3103: Cannot edit the assignee and priority of a ticket Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New http://launchpad.net/malone/bugs/3103
[01:15] <ddaa> Somebody please take ownership of that.
[01:17] <SteveA> david: done
[01:52] <salgado> stub?
[01:53] <carlos> hi
[01:55] <zyga> hello carlos
[01:55] <zyga> carlos: would you have 5 minutes after the meeting?
[01:56] <jblack> spiv: around? 
[01:56] <carlos> zyga sure
[01:57] <jamesh> cprov: in a bug comment, you suggested I read bugs 1458 and 1457.  It turns out that only you and kiko can read them.
[01:57] <SteveA> jamesh: the bzr meeting went on for longer than i thought it would
[01:57] <jblack> jamesd: Did you get an email from me a few days ago about your google maps hack?
[01:57] <mpt> Gooooooooooooooooood morning
[01:57] <ajmitch> morning mpt 
[01:57] <jblack> jamesh: Did you get an email from me a few days ago about your google maps hack?
[01:58] <jamesh> jblack: yeah.  I haven't gotten round to updating it yet.  Any particular license you'd like?
[01:58] <jblack> mpt: I'd like your comments on the BzrTargetPage that I hackedu p
[01:58] <matsubara> good morning all
[01:58] <cprov> jamesh: if want I can CC you too pub publish it depends on kiko ( he is the triage man )
[01:58] <cprov> jamesh:  if you want ...
[01:58] <jblack> jamesh: Anything should work, buddy.
[01:59] <kiko> morning hackers
[01:59] <kiko> how is everyone?
[01:59] <jblack> I'm awake!
[01:59] <ddaa> why is everyone?
[01:59] <carlos> kiko: morning!
[01:59] <zyga> cold but fine here
[01:59] <jblack> ddaa: Awake? Probably some sort of picnic or something
[02:00] <jamesh> cprov: okay.  It just seemed weird that "Launchpad Developers" didn't have permission to view the bugs
[02:00] <jblack> mpt: Did you see my question above? 
[02:00] <jblack> Whoops. 8
[02:00] <SteveA> jamesh: launchpad admins have permission to view stuff
[02:00] <kiko> jamesh, which bugs are those? 1xx7 and 1xx8?
[02:00] <mpt> jblack: oh, not until now
[02:00] <cprov> jamesh: indeed
[02:01] <SteveA> MEETING TIME
[02:01] <cprov> kiko: 145[78] 
[02:01] <SteveA> who's here today?
[02:01] <mpt> BjornT: Had you started reviewing my translation form branch yet?
[02:01] <jamesh> kiko: 1457 and 1458
[02:01] <mpt> me
[02:01] <jamesh> me
[02:01] <jblack> me
[02:01] <BjornT> me
[02:01] <niemeyer> Here
[02:01] <carlos> SteveA: Are you launchpad admin, last time you were not able to see a private bug...
[02:01] <cprov> me
[02:01] <kiko> jamesh, yeah, it's a bug, I filed it and it's assigned to bradb 
[02:01] <kiko> ME
[02:01] <BjornT> mpt: no, not yet
[02:01] <spiv> me
[02:01] <jamesh> me
[02:01] <SteveA> kiko, carlos: i reviewed it for him last night.  probably merged by now.
[02:01] <salgado> me
[02:01] <matsubara> me
[02:02] <SteveA> stub?
[02:02] <SteveA> lifeless: ?
[02:02] <bradb> SteveA: the merge failed though. gotta fix today
[02:02] <SteveA> mpool sends apologies.  i have his three sentences.
[02:02] <jblack> mpool and lifeless must be off hacking together and forgot.
[02:02] <mpt> BjornT: good, because I'm just mirroring my latest changes that I committed on Tuesday :-)
[02:02] <SteveA> Kinnison sends apologies -- working on getting dapper open.  i have his three sentences. 
[02:02] <carlos> oh, ok
[02:02] <BjornT> mpt: ok
[02:02] <SteveA> == Agenda ==
[02:02] <SteveA>  - roll call
[02:02] <SteveA>  - agenda
[02:02] <SteveA>  - next meeting (DST changes?)
[02:02] <SteveA>  - production / staging (timeouts?)
[02:02] <SteveA>  - shipit timeouts
[02:02] <SteveA>  - language packs
[02:03] <SteveA>  - three sentences
[02:03] <SteveA> 
[02:03] <SteveA> any other agenda items for today?
[02:03] <SteveA> spiv: can you phone stub please?
[02:03] <jblack> I have a short one
[02:03] <jblack> Check Bzr.
[02:03] <spiv> SteveA: Ok.
[02:03] <kiko> SteveA, rock n roll
[02:03] <SteveA> jblack: okay, "check bzr" added
[02:04] <ddaa> jblack: When there is bzr switch.
[02:04] <mpt> jblack: I have now
[02:04] <SteveA> okay, so we have DST changes coming soon
[02:04] <jblack> mpt: private?
[02:04] <SteveA> when does 1200 UTC get awkward for australians?
[02:04] <jamesh> I don't have daylight saving, so it doesn't affect me
[02:05] <spiv> iirc in NSW it's at the end of the month
[02:05] <spiv> http://www.bom.gov.au/climate/averages/tables/dst_times.shtml
[02:05] <kiko> that's fine for us because we go into dst as well
[02:05] <SteveA> that's when we'll be in montreal or so
[02:05] <spiv> 31/10/2004
[02:05] <mpt> In Sao Carlos it's this weekend
[02:05] <SteveA> so, let's arrange new times when in montreal
[02:05] <spiv> Sounds good.
[02:05] <SteveA> okay, so same time next week?
[02:05] <kiko> surio
[02:05] <cprov> fine
[02:06] <SteveA> i'm going to switch the agenda around a bit, because stu isn't around.  maybe he'll turn up later.
[02:06] <SteveA> so, language packs, carlos
[02:07] <carlos> yep
[02:07] <carlos> SteveA what do you want to talk about them?
[02:07] <SteveA> how are they doing?  any remaining issues from the launchpad end?
[02:08] <kiko> s/any/many
[02:08] <lifeless> here
[02:08] <lifeless> sorry i'm later
[02:08] <carlos> hmm, I think most export issues are fixed, at least the ones known 
[02:08] <carlos> the only remaining one I know is related to comments 
[02:08] <carlos> but it's not too important
[02:08] <SteveA> okay, that's great.
[02:08] <carlos> the main problem are the remaining review-breezy* templates
[02:09] <zyga> carlos: I'd like to ask one question if that's okay - about exports exactly
[02:09] <kiko> carlos, is pitti happy?
[02:09] <spiv> SteveA: stub's home phone is going to answering machine, and I can't find a mobile number for him.
[02:09] <SteveA> spiv: i called his mobile earlier, and got the same.
[02:09] <carlos> there are many of them, but jordi should be able to help here now that I did some fixes to the permissions, will give more rights now that breezy is out (or near to be released)
[02:10] <SteveA> zyga: please, ask
[02:10] <zyga> carlos: I've been working on something that uses exports from rosetta and last available export (3 days ago or something like that) does not validate in several cases. I have a log file available
[02:10] <carlos> kiko: well, we were a bit late so pitti is not 100% happy, but we have a month now to finish any issue we get and to fix all the review-breezy* templates
[02:10] <zyga> carlos: it seems that apart for some obscure charset problems there is at leas one serious problem (/me looks)
[02:11] <zyga> http://www.suxx.pl/ubuntu/language-packs/process.log
[02:11] <carlos> zyga like?, Martin didn't told me about any validation issue
[02:11] <kiko> carlos, a month?
[02:11] <zyga> check the tail - there is a parse error in exported .po
[02:11] <carlos> kiko: next language pack update
[02:12] <SteveA> hi stub 
[02:12] <carlos> zyga: hmm, kiko told me about that bug, I didn't know the language pack had it too and It should not happen... I will take a look 
[02:12] <stub> Sorry I'm late. I'm up to date on my time sheets.
[02:12] <carlos> it's really bad that it still appears as we have many tests for that specific problem
[02:12] <zyga> carlos: okay that's all from me - I'll monitor the export issues closely since I'm totally dependant on them
[02:13] <carlos> zyga: please, notify any problem you detect, your work is a good way to check them
[02:13] <kiko> carlos, it's really weird but I'm seeing it too -- in exports
[02:13] <carlos> SteveA: I'm working on a new way to handle imports that mark asked
[02:13] <zyga> carlos: I'll need further exports though - currently they seem to have been stopped
[02:13] <carlos> to solve the big, big problem we have now (all the review-* templates=
[02:14] <carlos> I want to have it ready before breezy + 1, but seems like it will be imported soon into launchpad... could I know when is that planned?
[02:14] <carlos> zyga: Yeah, I think I have problems with the db link, need to check with stub later
[02:14] <SteveA> i think that is proceding now
[02:14] <carlos> hmmm
[02:15] <carlos> in that case...
[02:15] <ddaa> carlos: breezy+1 = dapper
[02:15] <carlos> stub: Could you comment the cron entry for the rosetta attach script?
[02:15] <carlos> ddaa thank you I'm not able to remember the name :-(
[02:16] <SteveA> okay
[02:16] <SteveA> moving on 
[02:16] <SteveA> jblack: check bzr
[02:17] <jblack> Ok. Quick announcement. Bzr is shaping up pretty well. 
[02:17] <jblack> Lifeless is in a race to get us dogfooding before UBZ.
[02:17] <jblack> Its a good time to start poking with bzr.
[02:18] <jblack> So when 0.10 gets uploaded to ubuntu, give it a try, and see how it feels.
[02:18] <carlos> any chance to get something like pqm for bzr? I'm starting with bzr for a personal project and I need a way to use a shared repository to work with other people
[02:18] <jblack> (all done)
[02:18] <jblack> carlos: There is a pqm for bzr. :)
[02:18] <carlos> ok
[02:18] <carlos> will ask after the meeting
[02:18] <lifeless> 0.1 is in breezy
[02:18] <carlos> thanks
[02:18] <niemeyer> jblack: Isn't it 0.1?
[02:18] <ajmitch> 0.1.1 was uploaded
[02:18] <jblack> Pardon, 0.1
[02:18] <jamesh> 0.1.1
[02:18] <jblack> Yes, its already out. My packages list was out of date. So give it a try
[02:19] <lifeless> carlos: jblack has a tarball, and pqm is in my baz archive robert.collins@canonical.com--public/arch-pqm--main--0
[02:19] <carlos> lifeless: thanks
[02:19] <jblack> SteveA: all done.
[02:19] <lifeless> if I may 
[02:20] <SteveA> you may
[02:20] <lifeless> bzr dogfooding is coming along well
[02:20] <lifeless> the 0.1 format change was major, and with it behind us, the only critical things not yet ready are gpg signing - which is about 50% done, and 'switch', which while not trivial depends on the gpg configuration for a lot of its work.
[02:21] <jblack> and sftp
[02:21] <lifeless> so, mid next week, I plan to move 'dists' to bzr, which will involve a new checkout process for launchpad, and a new release of pqm
[02:21] <lifeless> no
[02:21] <SteveA> jblack: sftp isn't essential
[02:21] <SteveA> jblack: having a way to get things to and from chinstrap is
[02:21] <lifeless> sftp is not on the critical path anymore : pull and push to chinstrap are though, and we have that.
[02:21] <lifeless> ddaa: per branch config data
[02:21] <kiko> lifeless, why dists?
[02:21] <lifeless> anyway..
[02:21] <spiv> ddaa: I was wondering the same thing :)
[02:22] <kiko> why not something like cscvs?
[02:22] <lifeless> folk that commit to dists, I will be looking for you midweek to test and give feedback.
[02:22] <jamesh> kiko: to assemble the launchpad tree from the various branches
[02:22] <kiko> jamesh?
[02:22] <SteveA> kiko: we have two kinds of tree in launchpad
[02:23] <lifeless> in order of questions: product details are by sab mandate not in the bzr branch, so they are an external config item, -> needs the same infrastructure gpg policy recording does.
[02:23] <SteveA> kiko: we have the 'dists' tree, which contains the information to build the other trees
[02:23] <kiko> cprov, why didn't you subscribe launchpad-devel to 1457 and 1458?
[02:23] <jamesh> kiko: our trees are made of launchpad, sqlobject, zope, etc
[02:23] <kiko> I know
[02:23] <SteveA> so, lifeless will convert the 'dists' tree to bzr format
[02:23] <kiko> I understood that
[02:23] <lifeless> secondly, dists because its the least critical tree - a fuckup there is harmless to everyone except stub, because there are trivial workaround
[02:23] <SteveA> he will use whatever tools exist to do so
[02:23] <kiko> my question is why the dists tree and not some other module.
[02:23] <kiko> ah
[02:23] <kiko> I see
[02:23] <lifeless> if I convert cscvs and there are major teething problems, many people suffer.
[02:24] <cprov> kiko: do I need to do it ? I never did, tought it was implicity
[02:24] <lifeless> if I convert dists and there are unexpected issues, stub and I and -maybe- a couple others suffer.
[02:24] <SteveA> also, the dists tree is used at one time, and all the rest of the trees at another time
[02:24] <kiko> cprov, it's not done implicitly, you need to do it.
[02:24] <lifeless> the pqm support for dists-in-bzr is well tested and ready since august
[02:24] <cprov> kiko: ok, doing it now 
[02:24] <kiko> cprov, it's a private bug -- only subscribers can see it
[02:24] <SteveA> it would be a special case to convert just one tree for any other tree
[02:25] <SteveA> okay, to move on
[02:25] <lifeless> so - summary is : TODOs are shrinking, Conversion begins, committers to dists be ready to test.
[02:25] <lifeless> -that is all-
[02:25] <SteveA>  - production / staging
[02:25] <SteveA> there seem to be quite a few timeouts happening in production
[02:25] <SteveA> despite the 15 second timeout
[02:26] <SteveA> some of these are not on the whole request, but for specific queries
[02:26] <SteveA> the error the user sees is a plain page with just ProgrammingError on it
[02:26] <SteveA> i'll be fixing this to display the usual "timed out" error page
[02:26] <kiko> ah cool
[02:26] <kiko> SteveA, did you see my listing
[02:26] <jamesh> that'd probably require a bit of string matching
[02:26] <SteveA> one of the page affected is the shipit requests processing page that marilize uses
[02:27] <SteveA> jamesh: yes, it will
[02:27] <SteveA> kiko: probably...
[02:27] <kiko> SteveA, of error pages?
[02:28] <SteveA> yeah, i think so... things more than 1 day ago have become a blur ;-)
[02:28] <SteveA> stub: so, want to talk about production and staging a bit?
[02:28] <kiko> ok
[02:28] <SteveA> what scripts are currently running?
[02:28] <SteveA> gina, i guess
[02:28] <stub> And it turned out staging hadn't been updating properly for a couple of days, but I fixed that earlier today.
[02:28] <SteveA> pofile stuff?
[02:29] <SteveA> carlos just asked for some rosetta script to be disabled
[02:29] <stub> I killed gina a few minutes ago. There didn't appear to be any rosetta scripts running atm.
[02:29] <carlos> stub oh!, I think that explains the missing translations with language packs without changing any source code...
[02:29] <jamesh> stub: staging seems borked again, btw
[02:29] <jamesh> (just checked now)
[02:29] <carlos> stub: Is staging being updated now?
[02:29] <stub> no
[02:30] <kiko> stub, did you see the email I just forwarded to you?
[02:30] <stub> kiko: Nope, and I won't for a while
[02:30] <kiko> it's about an error in lp/__init__.py
[02:31] <SteveA> kiko:  mail it to me
[02:32] <SteveA> stub: gina does need to run, in general, at the moment, of course
[02:32] <stub> It will kick in 20mins time
[02:32] <stub> kiko: fixed earlier
[02:32] <SteveA> salgado: so, the shipit request page that marilize uses is timing out.  i've sent the information about the problem to stub.
[02:33] <salgado> SteveA, yes, I talked to him. we're going to look how to fix that after the meeting
[02:33] <SteveA> okay, great
[02:33] <salgado> I'm trying to see what's wrong with the broken query, but can't tell much
[02:33] <SteveA> we need to have a fix in production by tomorrow, so that marilize can do her job
[02:34] <salgado> SteveA, we can also revert the cherry-pick until this is fixed
[02:34] <kiko> thanks stub 
[02:34] <SteveA>  - sysadmin requests
[02:34] <salgado> this breakage was caused by a cherry-pick I requested to stub. the query works fine with sampledata, but it breaks in the production database
[02:34] <SteveA> anything blocked on the sysadmins doing things?
[02:35] <ddaa> me, but you know about it
[02:35] <SteveA> ddaa: right
[02:35] <jamesh> not blocked (I have other stuff to do), but there is the bugzilla dump
[02:35] <stub> salgado:  AND
[02:35] <stub>                ((Person.id = ShippingRequest.recipient AND
[02:35] <stub>                  Person.fti @@ ftq('')) OR
[02:35] <stub>                 (EmailAddress.person = ShippingRequest.recipient AND
[02:35] <stub>                  lower(EmailAddress.email) LIKE '%%') means you will get num_people_rows * num_email_address_rows rows being returned, which is doable on sample data but will never complete with production data
[02:35] <ddaa> well, I _do_ have other things to do :)
[02:35] <SteveA> jamesh: you have that now, i think
[02:36] <bradb> jamesh: Are you writing the conversion script?
[02:36] <jamesh> bradb: yeah
[02:36] <bradb> sweet
[02:36] <SteveA> jamesh: i'll look up the email from elmo after this meeting
[02:36] <SteveA> - activity reports
[02:37] <ddaa> I am.
[02:37] <jblack> I am
[02:37] <bradb>  jour, mo
[02:37] <salgado> stub, I see. how can I fix that and still query the emailaddress table?
[02:37] <niemeyer> I'm ok as well
[02:37] <salgado> btw, I'm up to date
[02:37] <mpt> up to date
[02:37] <carlos> As soon as I get my laptop connected to Internet will send all pending reports
[02:37] <spiv> Not up to date: I'm missing a few from the last week
[02:37] <jamesh> sent reports for up til wednesday.  will send today's later
[02:37] <lifeless> up to date
[02:38] <ddaa> bradb: I think that would be "g suis zajour, mo" :)
[02:38] <kiko> I've got one day to send but it already written
[02:38] <SteveA> so, there's going to be a round of appraisal / feedback stuff happening in a month or two.  being timely with activity reports is one of the things that is considered in that.  do yourself service, and keep them up to date.
[02:38] <bradb> or just "j't' jour, mo" :P
[02:39] <stub> salgado: it should be fine if it becomes AND person.id = shippingrequest.recipient and emailaddress.person = shippingrequest.recipient and (person.fi@@ftq('thequery') OR lower(emailaddress.email) like '%the query') and you make sure we don't match the empty string (which will return all records in a very inneficient manner)
[02:39] <SteveA> so, looks like everyone but me and bjorn are more or less up to date this time around
[02:39] <SteveA> well done
[02:39] <carlos> SteveA you, bjorn and me
[02:40] <BjornT> i only lack two days, though, will send them soon
[02:40] <SteveA> BjornT: okay, that's not too bad
[02:40] <SteveA> just me then ;-)
[02:40] <jblack> ~.
[02:40] <SteveA> anything else critical to discuss?
[02:41] <mpt> When does main start using Malone?
[02:41] <mpt> mdz suggested in the release announcement that all bug reports should go in Malone
[02:41] <SteveA> once we've converted the data from bugzilla
[02:41] <SteveA> interesting... but i think we need to have gina done first
[02:41] <spiv> More generally, what does the end of breezy/start of dapper mean for Launchpad? :)
[02:41] <SteveA> we should answer that when Kinnison is around
[02:42] <SteveA> let's go for the three sentences.  i have Kinnison's cprov's and mpool's.
[02:42] <lifeless> DONE: diff between branches, helped release 0.1, performance tuning, general fixes, mgmt stuff
[02:42] <SteveA> DOIT
[02:42] <ddaa> DONE: landed importd-archivelocation, not rolled out, revived BranchDataStorage
[02:42] <ddaa> TODO: python import, BranchDataStorage with niemeyer
[02:42] <ddaa> BLOCKED: XFS for python on production systems
[02:42] <lifeless> TODO: finish gpg configs and the sign/check ui, integrate with testaments.
[02:42] <kiko> DONE: reviews, QA, ensuring critical fixes land, user support
[02:42] <lifeless> BLOCKED: Not.
[02:42] <mpt> DONE: Finished bug listings, reworked translation form, various bugfixes
[02:42] <mpt> TODO: style sheet rationalization, /people/whoever design, more bugfixes
[02:42] <mpt> HINDRANCES: four branches waiting for review (kiko, BjornT)
[02:42] <salgado> DONE: Improved searching in shipit, lots of bug fixes, some code review, help out matsubara and gneuman
[02:42] <salgado> TODO: ShipIt reports, code review, help out matsubara and gneuman
[02:42] <salgado> BLOCKED: No
[02:42] <kiko> TODO: clear out my INBOX and start working on UBZ organization
[02:42] <jblack> DONE: coordinating third party packages
[02:42] <BjornT> DONE: vacation. coming back from vacation. resolved conflicts. fixed some bugs
[02:42] <BjornT> TODO: reviews. cron-job for support tracker. outgoing e-mail for support tracker.
[02:42] <kiko> BLOCKED: no
[02:42] <bradb> DONE: Landed a bugtarget search portlet. Wrote a sortorder widget. Make bug privacy admin-aware. Bought loads more memory!
[02:42] <BjornT> BLOCKED: no
[02:42] <spiv> DONE: reviews, added more debugging to librarian
[02:42] <spiv> TODO: AuthServerCaching
[02:42] <spiv> BLOCKED: no.
[02:42] <bradb> TODO: Get the admin-awareness patch through pqm. Talk to kiko to finish up the sortorder widget review. More user experience optimization and bugfixing.
[02:42] <carlos> DONE: Language packs, debug, debug, debug, fix, fix, fix, bug triage
[02:42] <bradb> BLOCKED: Non.
[02:42] <kiko> mpt, that's what BLOCKED is for
[02:42] <jblack> TODO: more third party tools, more documentation
[02:42] <jblack> BLOCKED: None new.
[02:42] <mpt> kiko: They're not blocking me
[02:43] <kiko> bradb, I've got second thoughts on the admin-awareness patch -- have you consulted with others
[02:43] <niemeyer> DONE: Boolean executable flag in bzr, tags in bzr, chats with David, started working on launchpad
[02:43] <niemeyer> TODO: Fix importd tests
[02:43] <niemeyer> BLOCKED: Nope
 DONE: Sprinting for the Uploader, buildd etc.
 TODO: Uploader needs finishing, Open Dapper on Launchpad
 BLOCKED: Gina run completion on Production
 DONE: BuildUI partially, fine-tune buildd-scoring, NominatedArchIndep
 TODO: Release BuilddUI and other tweaks on buildd
 BLOCKED: None
 DONE: bzr 0.1 release for breezy and including weaves, various bugs
[02:43] <carlos> TODO: PO imports/attachment, language packs, bug triage, give jordi more permissions
[02:43] <kiko> mpt, they are blocking that part of your work, and HINDRANCES is non-standard
 TODO: bzr 0.1.1 with small fixes; gpg signing; revision properties
[02:43] <jamesh> DONE: merged some of my pending branches, some GPG issues, some initial work on
 BLOCKED: no
[02:43] <jamesh> bugzilla -> Malone migration
[02:43] <jamesh> TODO: more schedulomatic work, code reviews
[02:43] <jamesh> BLOCKED: no
[02:43] <SteveA> DONE: more navigation hacking, reviews, management
[02:43] <mpt> ok, ok
[02:43] <SteveA> TODO: breadcrumbs landed, reviews, management
[02:43] <SteveA> BLOCKED: no
[02:43] <carlos> BLOCKED: No (Rosetta needs gina run on production to improve its usability)
[02:44] <bradb> kiko: Yes, previously. I intentionally made admins not allowed to see private bugs, but have heard a few complaints about it from admins since. :)
[02:44] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  fix bug 2769, stop converting distribution supports tickets from causing a system error. (patch-2644: bjorn.tillenius@canonical.com)
[02:44] <kiko> bradb, was I the only one to complain?
[02:44] <bradb> kiko: "a few", so no
[02:44] <bradb> SteveA, jamesh, etc 
[02:44] <SteveA> BjornT: can you do review for mpt today?
[02:44] <kiko> is jamesh a launchpad admin?
[02:45] <kiko> bradb, the issue there was that launchpad wasn't subscribed to that bug
[02:45] <jamesh> kiko: nope.  I'm just on Launchpad Developers
[02:45] <BjornT> SteveA: yeah, i'll try to do it today
[02:45] <kiko> bradb, I think we may be on the wrong track
[02:45] <SteveA> okay
[02:45] <kiko> SteveA, I can do mpt's review too
[02:45] <SteveA> any other blocked issues?
[02:45] <bradb> kiko: if something goes wrong with a private bug, who fixes it?
[02:45] <kiko> the DBA?
[02:45] <bradb> :)
[02:46] <SteveA> kiko, bradb: let's talk about this in 2 mins ;-)
[02:46] <SteveA> MEETING ENDS
[02:46] <ddaa> niemeyer: Good morning
[02:46] <zyga> great :)
[02:46] <niemeyer> ddaa: Morning!
[02:46] <carlos> SteveA thanks
[02:46] <SteveA> thanks everyone. 
[02:46] <zyga> carlos: is there any estimated ETA for further exports? are they usefull to anyone besides me?
[02:46] <ddaa> niemeyer: I'm heading for lunch now, we'll have our daily sync when I'm back in 30-60mins
[02:46] <carlos> zyga: I'm all yours ;-)
[02:46] <SteveA> bradb: i'm going to be implementing "admin hats" in the nearish future.
[02:47] <ddaa> niemeyer: okay with you?
[02:47] <niemeyer> ddaa: Ok
[02:47] <bradb> hats are nice
[02:47] <niemeyer> ddaa: Sure
[02:47] <kiko> SteveA, I remember that now
[02:47] <sabdfl> stub: ping
[02:47] <niemeyer> ddaa: Have a nice lunch
[02:47] <carlos> further exports?
[02:47] <SteveA> bradb: this is a feature whereby admin team members only get admin privs when they specifically elect to do so.
[02:47] <carlos> zyga what do you mean?
[02:47] <carlos> zyga daily exports?
[02:47] <stub> sabdfl: pong
[02:47] <SteveA> bradb: usually, they use the site as a normal user.  without the "admin hat" on, they would not see private bugs.
[02:47] <zyga> carlos: I was using tarballs from pittys ~ directory but now they seem to have been stopped
[02:47] <zyga> carlos: yes, daily exports
[02:48] <bradb> SteveA: nice
[02:48] <SteveA> jamesh: i'm going to /msg you where the bugzilla db is
[02:48] <carlos> zyga well, we were planning to stop doing them and move it to monthly exports like language packs
[02:48] <dreifinger> hey peeps
[02:49] <carlos> zyga I will fix them to resume the daily exports until all know issues are gon
[02:49] <carlos> gone
[02:49] <zyga> carlos: cool, monthly exports are reasonable
[02:49] <SteveA> bradb: then, we'll also log when an admin uses this privilege to view a private bug
[02:49] <dreifinger> new ubunto installe
[02:49] <SteveA> bradb: so there will be some kind of audit trail available
[02:49] <bradb> kiko: if going through the DBA is the best way to resolve problems with private bugs (which may not necessarily be trivial issues to fix in SQL), I don't mind. It means just leaving things the way I have them documented/implemented currently.
[02:49] <carlos> zyga: the idea is that we export the tarball so pitti can generate updated language packs
[02:49] <lifeless> SteveA: it should not log if they would have access as their normal self
[02:50] <zyga> carlos: so you are aware of export issues, I don't need to tell anyone else about them - right?
[02:50] <SteveA> lifeless: correct
[02:50] <SteveA> lifeless: only if the escalation is used
[02:50] <kiko> bradb, I think SteveA's fix would make your change acceptable
[02:50] <zyga> carlos: great - I use it for the exactly... similar purpose :-)
[02:50] <bradb> kiko: I'm inclined to lock private bugs up as tightly as possible, but a true "admin" has no such restrictions, IMHO
[02:50] <SteveA> kiko: this isn't top of my priority list right now, but it isn't bottom either
[02:50] <kiko> SteveA, should bradb wait for this?
[02:51] <carlos> zyga no, don't need to complain to anyone else, I get an email remind me about the problem, but yesterday I was on holidays and today I had network problems, that's why it's still broken, I hope will be fixed later today
[02:51] <SteveA> kiko: i think, no.  but we should have fewer launchpad admins.  i need to add a launchpad developers group to the sample data, and then we can narrow down the launchpad admins a bit.
[02:51] <bradb> indeed
[02:52] <kiko> okay, agreed then
[02:52] <zyga> carlos: great, thanks, that's all from me :)
[02:52] <carlos> zyga ok, cool.
[02:52] <bradb> SteveA, kiko: so, i'll go ahead and baz smash the admin patch then?
[02:52] <SteveA> kiko: so, to recap: i add the launchpad developers group to the sampledata, and allow this group to see tracebacks at all times.  then, we thin out the admins group.  then brad can land this change.  later on, i add 'changing hats'
[02:53] <carlos> see you later
[02:53] <SteveA> bradb: actually, how about you do the sampledata change, and the tracebacks change?
[02:53] <SteveA> that way it is all onemerge
[02:53] <carlos> I need to disconnect while I'm on lunch, will be back in an hour or so
[02:53] <SteveA> carlos: okay, see you later
[02:53] <kiko> bradb, I would appreciate a better message for permission denied though
[02:54] <bradb> kiko: indeed
[02:54] <SteveA> bradb: what page does kiko mean for "permission denied" ?
[02:54] <SteveA> what does that message occur?
[02:54] <bradb> SteveA: I really want to land code. Is there a need to hold back landing to add these other things?
[02:55] <kiko> SteveA, viewing the bug page if you don't have permission to see it
[02:55] <SteveA> bradb: yes.  however, it is about 3 lines of code.  you should make this change, and then land it.
[02:55] <SteveA> kiko: and what information do you want to see on that page?
[02:57] <SteveA> kiko: if this is a custom error page, i'd like to be involved in doing that.
[02:57] <SteveA> as this ties into navigation changes.
[02:57] <kiko> SteveA, Unauthorized: this bug is private: only explicit subscribers are allowed to view it, and you are not a subscriber.
[02:57] <kiko> something like that
[02:58] <kiko> mpt can cook up proper wording
[02:58] <SteveA> kiko: okay.  i have a more general solution to this, but i think we can do a special error page as an interim measure.
[02:59] <ajmitch> kiko: bug submitters are still subscribed, though?
[03:00] <kiko> ajmitch, they are automatically subscribed when the bug is made private IIRC
[03:01] <ajmitch> right, I was just surprised when a launchpad bug I reported was set private & I could still see it
[03:02] <SteveA> mpt: salgado reviewed my initial breadcrumbs code.  i need to make some corrections to it (thanks salgado), and then i'll land it.
[03:02] <SteveA> mpt: i have implemented all of the examples mentioned in the spec.
[03:02] <mpt> SteveA: great
[03:02] <sabdfl> stub: ProgrammingError: ERROR:  could not open file "/usr/share/postgresql/contrib/english.stop
[03:02] <sabdfl> ?
[03:03] <SteveA> mpt: the two things it doesn't do at the moment: handle trailing ">" markers.  Make the crumb not a link if it is the current page
[03:03] <sabdfl> also, stub, when do you plan to branch?
[03:03] <stub> sabdfl: You running PostgreSQL 7.4 ? That looks like it would have to be a tsearch2 issue
[03:03] <SteveA> mpt: i need to do a bit of refactoring before i do those things, because the requred code will end up in 3 places otherwise.
[03:03] <SteveA> mpt: so, that will land tomorrow or so.
[03:03] <mpt> ok
[03:03] <stub> sabdfl: I havn't looked at the recent commits yet. You have something to land?
[03:04] <sabdfl> i'm working on a branch that needs to be in Tuesday
[03:04] <sabdfl> same trick - let me know what RF revision you will tag from, I will merge that and no more
[03:04] <sabdfl> and yes, pg7-4
[03:05] <stub> sabdfl: ok
[03:05] <sabdfl> cool, thanks
[03:05] <SteveA> sabdfl: i'll land a navigation change to allow redirection as a consequence of traversal when i land breadcrumbs today.
[03:05] <sabdfl> i will merge now, then wait for mail from you
[03:05] <SteveA> sabdfl: will you want to merge from my branch for this, to avoid going later into RF?
[03:06] <sabdfl> SteveA: if you make it before the cutoff, i'll get it
[03:06] <SteveA> okay.  i need lunch now.  it'll be submitted this afternoon sometime.
[03:06] <stub> sabdfl: I havn't seen that english.stop error before.
[03:07] <stub> sabdfl: Hmm... that is the hoary location of english.stop. Breezy puts it in /usr/share/postgresql/7.4/contrib/english.stop. Has there been a PostgreSQL update recently?
[03:07] <kunta> hello everibody
[03:08] <kunta> i want to add a new language to ubuntu translation tree
[03:08] <kunta> how can i do that please
[03:08] <sabdfl> stub: no, i suspect a bug in the breezy package that we just discovered :-/
[03:08] <sabdfl> kunta: we need to know the plural form
[03:09] <sabdfl> have you translated PO files before?
[03:09] <stub> sabdfl: I don't know why it isn't affecting me though unless there was an update in the last 6 hours or so...
[03:09] <sabdfl> stub: it seems to be only this query
[03:09] <sabdfl> others work
[03:09] <kunta> once before for mozilla
[03:09] <stub> sabdfl: Now that is freaky...
[03:09] <kunta> but it was for my convenience
[03:10] <kunta> i'm leaving in guadeloupe and my own language isn't in the translation tree
[03:12] <kunta> someone can help me
[03:12] <kiko> kunta, what language is that?
[03:12] <kunta> kreyol
[03:12] <kiko> aha
[03:12] <kiko> jordi, ping?
[03:13] <kunta> se lang a manman mwen e i pa adan biten a zot
[03:13] <kunta> an example 
[03:13] <sabdfl> kunta: ok, you need to do some research. find a PO file in kreyol, and find the Plural-Form: headers
[03:13] <sabdfl> and bring those here
[03:14] <sabdfl> give them to Jordi or Carlos
[03:14] <sabdfl> and they will set it up
[03:14] <sabdfl> then you are OK
[03:14] <sabdfl> google will help
[03:14] <mpt> arggg
[03:14] <kunta> i'm the first to propose that
[03:14] <mpt> bradb?
[03:14] <bradb> mpt: ?
[03:15] <kunta> i'm must create plural form:header and po file ?
[03:16] <bradb> kiko: btw, can you please have a look at my reply to your review of the sortorder widget? i'm hoping to land that this morning.
[03:16] <kiko> yes, bradb 
[03:16] <bradb> great, thakns
[03:16] <bradb> er, thanks
[03:16] <mpt> bradb: If you're not busy, fixing https://launchpad.net/products/malone/+bugs/3091 would make my triage-while-waiting-for-baz much easier :-)
[03:16] <Ubugtu> Malone bug #3091: +viewstatus for bug fails with NotFound error for context/priority/title Fix req. for: malone (upstream), Severity: Normal, Assigned to: Nobody, Status: New http://launchpad.net/malone/bugs/3091
[03:16] <bradb> mpt: the search portlet already does that
[03:16] <kunta> wait a minute please
[03:17] <mpt> bradb: great
[03:18] <sabdfl> mpt: did you do the new ticket icon? nice
[03:18] <bradb> mpt: to be clearer: you can jump to any bug id or do a search from within the bug pages
[03:18] <sabdfl> celso: good work on the build system. the code is generally well structured and easy to grok
[03:18] <mpt> sabdfl: yeah, a few weeks ago, I need to tweak the colors to match the lpintegration icon
[03:18] <Kinnison> sabdfl: FYI, if you need my attention for anything, prod me here 'cos my headphones cancel office noise to reduce distraction
[03:18] <sabdfl> will do
[03:19] <Kinnison> cool
[03:20] <mpt> bradb: I don't see what your clarification has to do with that bug
[03:20] <Kinnison> SteveA: selectOneBy -- does that return zero-or-one, or is it one-or-raise ?
[03:21] <SteveA> returns a thing, or None, or raises if >1
[03:21] <bradb> mpt: it makes hacking the url unnecessary,  until i fix the URL, adding a useful feature at the same time
[03:21] <SteveA> and i'm having lunch, so not here ;-)
[03:21] <Kinnison> SteveA: thanks
[03:22] <mpt> bradb: The bug has nothing to do with URL-hacking that I can see. It's about a missing title for NULL priorities.
[03:22] <bradb> mpt: oh, then you gave me the wrong URL :)
[03:22] <bradb> mpt: I thought you were implying that +bugs/NN should work
[03:23] <mpt> bradb: No, by "fixing <url to bug>" I meant fixing the bug that's described at that URL
[03:23] <bradb> but yes, i'm aware of the priority title problem, have created the branch to start it, just landing other stuff before continuing on that one
[03:23] <mpt> oh!
[03:23] <mpt> I mistyped the URL
[03:23] <mpt> sorry
[03:23] <mpt> but Ubugtu got it right anyway :-)
[03:26] <cprov> SteveA: please, ping me when you get back .
[03:28] <kiko> stub, are you going to bed nowish?
[03:28] <mpt> bradb: other than search, the big annoyance factor seems to be the lack of package/product/project subscriptions
[03:29] <sabdfl> stub: do you have an example of creating an fti?
[03:29] <stub> kiko: nowish I'm helping get shipit operational again
[03:29] <kiko> stub, okay, I'm going to provide a quick fix for +edithomepage
[03:29] <stub> sabdfl: Just edit database/schema/fti.py , adding the stuff you need to the data structure at the top of the script
[03:29] <kiko> (bug 3046)
[03:29] <Ubugtu> Malone bug #3046: Unable to edit my home page Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: neuman, Status: New http://launchpad.net/malone/bugs/3046
[03:30] <bradb> mpt: really? why's that? I would have thought have a way to change the maintainer to accurately reflect who is actually responsible for making sure fixes in that package are uploaded would be somewhat more important.
[03:30] <bradb> s/have a way/that a way/
[03:31] <sabdfl> stub: i have a few text fields on the row, and want the fti to include all of them, prioritised, is that easy?
[03:31] <bradb> (which would also make it easy for a user to ask for a list of specifically "bugs on things i maintain" and get back precisely the list of bugs filed on things for which they are responsible.)
[03:32] <stub> sabdfl: yes. You can assign a priority for each from A to D (A being highest priority)
[03:33] <Kinnison> sabdfl: check out fti.py
[03:33] <Kinnison> sabdfl: it's really easy
[03:33] <kiko> matsubara, ping?
[03:34] <kiko> I am asking myself about your patch-23
[03:34] <ajmitch> bradb: and when different people maintain it in different distros?
[03:34] <kiko> BjornT, ping?
[03:34] <mpt> bradb: well, accurate maintainership works for the maintainer of a package, but subscriptions works for the maintainer *and* anyone else who does bugfixes, *and* for projects like MOTU that cover hundreds of packages
[03:35] <matsubara> kiko, pong
[03:35] <Kinnison> oh yes
[03:35] <kiko> matsubara, I'm curious if it would be possible to avoid generating the null bugdelta alltogether
[03:36] <bradb> ajmitch: our data model handles that, AFAIK
[03:36] <ajmitch> bradb: how would it handle subscriptions of MOTU to > 10K packages?
[03:37] <bradb> ajmitch: we'd need a list of those packages and then we'd write some code, I'd imagine
[03:37] <bradb> (i.e. to make sure the maintainer data is set correctly)
[03:38] <matsubara> checking it, just a min
[03:38] <bradb> ajmitch: for a newcomer to tell LP "I want to subscribe to all MOTU packages" would probably require something like tags/keywords, in my estimation
[03:40] <ajmitch> bradb: I'd want to be able to subscribe to a package or group of packages
[03:40] <ajmitch> usually just those that are related to the teams I'm in though
[03:41] <mpt> bradb: No, just subscribe to the MOTU project
[03:41] <bradb> sure. tags are well-suited to that kind of thing, IMHO. i.e. arbitrary grouping of things.
[03:41] <kiko> mpt, that's an interesting concept
[03:42] <kiko> and correct too
[03:42] <bradb> mpt: that doesn't sound the same, semantically
[03:42] <ajmitch> mpt: then you'd receive all the bugs that the motu team receives?
[03:42] <bradb> and various other things, presumably
[03:42] <mpt> yep
[03:43] <mpt> those on the MOTU mailing list already do, in an awkward fashion
[03:43] <ajmitch> yes
[03:43] <ajmitch> we're trying to get the moderation sorted
[03:44] <mpt> seb128 may be right, people sending comments to Malone shouldn't get random responses from other people/mailing lists/bots
[03:44] <mpt> Does debbugs have bug subscriptions?
[03:44] <ajmitch> yes, it does
[03:45] <mpt> so how does it solve this problem?
[03:45] <ajmitch> at least I think it was recently added
[03:45] <ajmitch> you can subscribe to packages as well, which is separate
[03:45] <mpt> right, but what if you subscribe a moderated mailing list to a package or to a bug
[03:46] <ajmitch> which is often the case - I don't think there's a problem there, other people don't get moderation requests afaik
[03:46] <kiko> stub, can you cherry-pick christian.reis@canonical.com--lozenge/launchpad--trivialities--1--patch-46
[03:46] <mpt> ajmitch: how? :-)
[03:47] <kiko> stub, I've sent to PQM if you can wait, but otherwise, it fixes bug 3046, which is a major embarassment for us
[03:47] <Ubugtu> Malone bug #3046: Unable to edit my home page Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Christian Reis, Status: PendingUpload http://launchpad.net/malone/bugs/3046
[03:47] <mpt> hooray for kiko
[03:47] <Nafallo> yay!
[03:47] <Nafallo> nice fresh bot? :-)
[03:47] <kiko> friggin portlet-actions
[03:48] <ajmitch> mpt: not sure yet :) debbugs subscription was implemented back in july
[03:48] <kiko> BjornT?
[03:48] <ajmitch> mpt: I could be mistaken there, that was from a quick google search..
[03:48] <kiko> bradb?
[03:49] <bradb> kiko: ?
[03:49] <BjornT> hi kiko 
[03:50] <stub> kiko: mirroring can help (or giving me the correct patch#)
[03:50] <kiko> stub, doh
[03:51] <stub> BjornT: your cherry pick should be live
[03:51] <kiko> BjornT, matsubara has a fix for bug 759 that essentially short-circuits when the bug delta is None. 
[03:51] <Ubugtu> Malone bug #759: re-duping an already duped bug causes system error. Fix req. for: malone (upstream), Severity: Normal, Assigned to: Diogo Matsubara, Status: Accepted http://launchpad.net/malone/bugs/759
[03:51] <BjornT> stub: cool, thanks
[03:51] <kiko> BjornT, I'm wondering if there is a way of avoiding even triggering the bug notification
[03:52] <kiko> stub, mirrorred, paper bag on head 
[03:53] <Kinnison> salgado: fancy a teeny tiny review?
[03:54] <mpt> SteveA: When you say "Should the breadcrumbs show on an error page", what kind of breadcrumbs *could* you show safely?
[03:55] <Kinnison> or kiko, what about you? fancy a teeny tiny review?
[03:55] <salgado> Kinnison, does it has to be now?
[03:56] <BjornT> kiko: yes, there should be a way, i'll take a quick look at it. (btw, wouldn't a better solution be to turn 'Mark as duplicate' into 'Unmark as duplicate', or something like that?)
[03:56] <mpt> SteveA: e.g. on an Oops page, what if the error is in one of the breadcrumb titles/links itself?
[03:56] <Kinnison> salgado: I'm on a really tight time budget wrt. dapper, it can wait a bit, but so long that I end up committing more to the branch
[03:56] <Kinnison> salgado: I can nopaste the diff
[03:56] <Kinnison> salgado: It's really very small
[03:56] <salgado> Kinnison, okay
[03:57] <Kinnison> salgado: https://chinstrap.ubuntu.com/~dsilvers/paste/fileiClBw3.html
[03:59] <mpt> stub, how is the "in six minutes" calculated?
[03:59] <ajmitch> mpt: I got familiar with that when bugfixing for breezy :)
[03:59] <mpt> is that using fmt:approximateduration?
[04:00] <stub> mpt: Dunno. Steve wrote that. (If I reviewed it, I've forgotten)
[04:00] <sabdfl> stub: interesting python ./fti.py gives me that same english.stop error
[04:00] <kiko> BjornT, I don't think that catches a double-submit or reload bug
[04:00] <stub> sabdfl: Paste your change to fti.py and I'll test it on my machine
[04:01] <Kinnison> sabdfl: python fti.py -U postgres -d launchpad_dev
[04:02] <sabdfl> Kinnison: same error
[04:02] <Kinnison> sabdfl: want me to come look?
[04:02] <sabdfl> sure
[04:04] <stub> Kinnison: I havn't seen that before. It would be good to know if sabdfl is triggering a lurking bug in our code or if his PostgreSQL installation is screwed somehow
[04:05] <stub> kiko: picked
[04:06] <sabdfl> stub: the db dump came from a hoary machine
[04:06] <sabdfl> if the path is inside the db dump, then that would explain it
[04:06] <sabdfl> how can i update the tsearch2 etc inside an existing db?
[04:07] <sabdfl> stub: are you taged yet?
[04:07] <stub> not tagged yet
[04:07] <sabdfl> how do i reinstall tsearch2 without munging the database?
[04:08] <stub> Sounds like we need to extract the tsearch2 stuff
[04:08] <stub> hmm...
[04:08] <sabdfl> --setup-only worked
[04:08] <sabdfl> oh
[04:09] <sabdfl> no it didn't
[04:09] <kiko> stub, you ROCK
[04:09] <stub> sabdfl: ok. The table the dud data is stored in is ts2.pg_ts_dict
[04:10] <kiko> bradb, how's bug 3077 going?
[04:10] <Ubugtu> Malone bug #3077: Oops at +assignedbugs and +viewstatus from missing NULL priority title Fix req. for: malone (upstream), Severity: Normal, Assigned to: Brad Bollenbach, Status: New http://launchpad.net/malone/bugs/3077
[04:10] <BjornT> kiko: that's true, but it would greatly reduce the risk triggering the bug. anyway, here's one way of fixing the bug:
[04:11] <lifeless> gnight
[04:11] <bradb> kiko: blocked on my current branch, which is making its own changes to the sampledata (due to SteveA's suggestion from earlier). i'm just fixing some broken tests now, which takes especially long without being able to execute individual stories.
[04:11] <stub> sabdfl:  update pg_ts_dict set dict_initoption = '/usr/share/postgresql/7.4/contrib/english.stop' where dict_name=en_stem; should fix it I think
[04:12] <BjornT> kiko: 1) change IBug.duplicateof to be an Object field, with schema=IBug. 2) add BugWidget._toFieldValue that converts the text input to a bug.
[04:13] <kiko> bradb, okay, thanks.
[04:13] <kiko> BjornT, okay, and then?
[04:13] <BjornT> kiko: for the first step you'd have to add IBug['duplicateof'] .schema = IBug after IBug's definition (might be a nicer way)
[04:13] <kiko> BjornT, you can't do that in the constructor?
[04:14] <BjornT> kiko: then the mail notification won't be triggered
[04:14] <BjornT> kiko: you can't reference IBug from inside IBug's definition
[04:15] <kiko> stub, can you also cherrypick rocketfuel@canonical.com/launchpad--devel--0--patch-2604 so I can see what's causing our externalsystem failure?
[04:15] <kiko> stub, hmm, actually, don't.
[04:15] <kiko> stub, use christian.reis@canonical.com--lozenge/launchpad--devel--0--patch-184 instead
[04:15] <kiko> the launchpad patch is too big
[04:19] <kiko> salgado, can you translate "Fix a bug when displaying the vote of the logged in user in a public poll" to me?
[04:20] <kiko> bradb, for the record, you could never have landed the change to bugpriority without testing it in the sample data, it was a major flaw.
[04:21] <BjornT> kiko: actually, it might be nicer to create a Bug field instead, it could import IBug from its validation method. this would avoid having to change IBug after it has been defined.
[04:22] <kiko> BjornT, that sounds better indeed
[04:22] <mpt> BjornT: Do you have any idea when bugzilla.ubuntu.com is being imported into Malone?
[04:22] <kiko> mpt, jamesh is working on it, but not soonish
[04:23] <mpt> ok, ta
[04:30] <ajmitch> bradb: 3077 is a little way off then? it's causing some pain & anguish for me :)
[04:30] <bradb> ajmitch: it should be rolled out tomorrow
[04:30] <ajmitch> great
[04:31] <bradb> unless stub was planning to stay up all night :)
[04:31] <ajmitch> it broke viewing the list of MOTU bugs a couple of days before release
[04:31] <kiko> ajmitch, null priorities :-(
[04:31] <ajmitch> kiko: yeah I heard, I filed a bug as well
[04:37] <bradb> stub: to make sure a patch is cherry-pickable, i can 1. branch off prod, 2. fix, 3. request that branch to be cherry-picked, 4. make another branch off rocketfuel,, 5. merge the cherry-pickable branch into the branch-of-rf branch and sort out the conflicts, right?
[04:37] <kiko> mpt, bradb, note that bug 3091 isn't exactly a dupe of 3077, though they have the same root origin -- be sure to fix in both places
[04:37] <Ubugtu> Malone bug #3091: +viewstatus for bug fails with NotFound error for context/priority/title Fix req. for: malone (upstream), Severity: Normal, Assigned to: Nobody, Status: New http://launchpad.net/malone/bugs/3091
[04:38] <bradb> kiko: ok
[04:38] <mpt> kiko: I put all the URLs in the updated description
[04:39] <mpt> and updated the title too
[04:39] <bradb> SteveA: FTR, adding this sample data change with my other, unrelated changes, is making life much more complicated ;)
[04:39] <stub> bradb: I think lifeless wrote instructions on the wiki somewhere. I'm not a baz expert despite my reputation of omniscience ;)
[04:39] <salgado> kiko, when you go to a public poll in which you already voted, your vote should be displayed. there was a problem that the code was trying to get your vote even when you haven't voted. I fixed that and added a test
[04:39] <stub> bradb: What you describe sounds sane. 
[04:39] <kiko> thanks salgado 
[04:40] <kiko> mpt, are you seeing floating tabs in launchpad yet again? :-(
[04:40] <bradb> stub: when you roll out production, do you roll out a completely fresh tree, or do you merge in the latest changes?
[04:41] <bradb> if you roll out a completely fresh tree, this cherry-pick idea makes sense, according to my mental model anyway
[04:41] <mpt> kiko: yep
[04:41] <stub> bradb: For cherry picks I generally 'baz update' or 'baz switch' to the head of the production branch
[04:41] <kiko> mpt, FFS. this time I can't really see who did it :-(
[04:42] <stub> bradb: once the relevant patches have been 'baz replay'ed onto the production branch of course
[04:42] <mpt> kiko: I'm about to attack the style sheets viciously, so a fix for that should fall out
[04:42] <kiko> thanks mpt 
[04:44] <bradb> stub: so you replay and commit cherrypicked patches to the prod branch?
[04:44] <Kinnison> Do we run the tests in optimised mode or something odd?
[04:44] <stub> bradb: Yes.
[04:45] <bradb> ok
[04:45] <stub> Kinnison: no. 
[04:45] <Kinnison> assert expr, msg
[04:45] <Kinnison> yes?
[04:45] <kiko> correct
[04:46] <stub> Kinnison, kiko: did you get this email -> http://lists.canonical.com/mailman/private/launchpad-error-reports/Week-of-Mon-20051010/014997.html
[04:47] <Kinnison> I've gotta workrave, back in a sec
[04:48] <sabdfl> stub: you gem
[04:49] <kiko> looking stub 
[04:52] <Kinnison> stub: Hmm
[04:52] <Kinnison> stub: Not a clue what that's on about
[04:52] <kiko> Kinnison, I just emailed you about that error, ftr
[04:52] <stub> the errors are just the usual gina borkage. I'm currently trying to work out if Mailman is silently dropping long messages or if my mail provider is
[04:54] <Kinnison> stub: Seems very odd that the constraint is being violated
[04:54] <Kinnison> stub: What's your commit size?
[04:54] <stub> I patched gina earlier today to commit after every one. It is no longer tied to the --countdown option
[04:55] <stub> Because otherwise it is impossible to recover from database exceptions (and we need commit lots  on production anyway)
[04:55] <Kinnison> Right
[04:56] <Kinnison> So she imported the sources okay?
[04:56] <kiko> stub, I don't know -- I suspect I didn't get it but don't know
[04:56] <stub> I guess so ;)
[04:57] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Fix bug 3046: Unable to edit my home page. Add a test for it. (patch-2645: christian.reis@canonical.com)
[04:57] <stub> Kinnison: I don't know if that is warty, hoary, breezy or what though.
[04:57] <Kinnison> https://launchpad.net/distros/ubuntu/warty/+sources/mozilla-firefox
[04:57] <Kinnison> that works
[04:57] <Kinnison> (apart from obviously not being published yet)
[04:57] <Kinnison> that's gonna be fun
[04:57] <sabdfl> stub: do you know if the gpg-keyring import went ahead?
[04:57] <Kinnison> At least the publisher doesn't hold a transaction open while it runs apt-ftparchive any more :-)
[04:58] <stub> sabdfl: No. I opened a bug on it. We have code but it was very unclear what keyrings I should be using or where to find them.
[04:59] <sabdfl> hmm.. bummer, the gina run will create masses of dup people then
[04:59] <Kinnison> sabdfl: bit late, it'll have done it by now :-)
[04:59] <stub> sabdfl: Yes. Unfortunately we couldn't afford to wait any more
[04:59] <Kinnison> Someone needs to do the following to the db.
[05:00] <Kinnison> 1. Mark breezy as the current supported
[05:00] <bradb> stub: what branch do i want to branch from to be sure my patch is cherry-pickable for you tomorrow?
[05:00] <Kinnison> 2. Mark hoary as supported
[05:00] <Kinnison> 3. Create a breezy distrorelease
[05:00] <Kinnison> 4. add sparc and ia64 to each of hoary breezy and dapper
[05:00] <stub> I'm still waiting for someone to own that bug - we spent the time writing the code but someone who was involved in the detailed discussions needs to push it over the finish line
[05:00] <Kinnison> 3 should be 'create a dapper distrorelease'
[05:01] <stub> is that a me job or can someone do it through the ui (with real descriptions rather than lorem ipsum)
[05:01] <Kinnison> sabdfl: what's the UI status for all that?
[05:01] <Kinnison> stub: remember hoary/breezy imports have five archs, not three
[05:01] <stub> bradb: rocketfuel@canonical.com/launchpad--production--1.36
[05:02] <sabdfl> Kinnison: doable, i think
[05:02] <bradb> thanks
[05:02] <kiko> stub, what bug are you referring to?
[05:02] <sabdfl> Kinnison: test on your laptop?
[05:02] <stub> bradb: erm... actually.... stuart.bishop@canonical.com/launchpad--production--1.36 (it hasn't gotten through pqm yet)
[05:02] <Kinnison> sabdfl: My laptop is *way* busy right now
[05:02] <sabdfl> ok
[05:03] <stub> kiko: I would tell you except my 'my bugs' page is giving me an oops :-(
[05:04] <kiko> :-(
[05:04] <Kinnison> stub: oh yeah, and hoary has hppa too
[05:04] <Kinnison> s/hoary/breezy/
[05:04] <stub> Kinnison: they do?
[05:05] <Kinnison> stub: warty has i386, powerpc, amd64
[05:05] <stub> I'll need to do that tomorrow - I havn't a clue if those architectures exist in the production db
[05:05] <Kinnison> stub: hoary has i386, powerpc, amd64, ia64, sparc
[05:05] <Kinnison> stub: breezy has i386, powerpc, amd64, ia64, sparc, hppa
[05:05] <Kinnison> elmo: that's right isn't it?
[05:05] <bradb> stub: so i should wait until that lands to branch, i take it
[05:06] <stub> bradb: You can branch directly from there if you want. 
[05:06] <elmo> Kinnison: no
[05:06] <elmo> hoary + breezy have the same architectures
[05:06] <elmo> all 6
[05:06] <Kinnison> elmo: oh right
[05:07] <Kinnison> stub: so, warty == i386,amd64,powerpc
[05:07] <bradb> stub: ok, branching directly from there then
[05:07] <Kinnison> stub: hoary and breezy == i386,amd64,powerpc,hppa,ia64,sparc
[05:09] <Kinnison> keyb.
[05:09] <kiko> uk
[05:10] <Keybuk> ello
[05:13] <sabdfl> stub: do you have any examples of places where we do non-SQLobject-style access to the db?
[05:13] <sabdfl> like "SELECT DISTINCT foo FROM bar INNER JOIN.... etc"?
[05:13] <stub> sabdfl: people merge
[05:13] <sabdfl> 'k thanks
[05:13] <stub> gina
[05:13] <moyogo> hi
[05:15] <stub> Kinnison: Unless gina magically fills out distroarchrelease, processor, processorfamily the relevant records need to be created for the 3 new hoary and breezy architectures. I would like someone more familiar with the distribution to do this.
[05:16] <stub> Or I could guess, which might be amusing
[05:17] <zyga> since when ubuntu builds for ia64?
[05:17] <Nafallo> zyga: atleast since hoary started.
[05:17] <Kinnison> stub: Okay, so it needs me or sabdfl to do that
[05:18] <zyga> Nafallo: but there aren't any iso images built for it
[05:18] <Nafallo> zyga: it's not an official, supported architecture.
[05:18] <stub> Kinnison: If you have time. It should be about 12 insert statements.
[05:19] <sabdfl> Kinnison: part of my current branch does the official/unofficial port flag for distroarchrelease
[05:19] <SteveA> bradb: really?  because of sample data conflicts?
[05:19] <Kinnison> sabdfl: Cool
[05:19] <stub> Should the extra architectures wait then until that lands?
[05:19] <Kinnison> stub: we can run gina to import the additional archs later
[05:19] <stub> Cool
[05:19] <Kinnison> so yes, we can wait for sabdfl's brach
[05:19] <bradb> SteveA: because of the somewhat increased likelihood of them happening, yeah
[05:19] <SteveA> mpt: can we talk about breadcrumbs and errors a bit?
[05:20] <sabdfl> well, this first drop is skeletal, but thanks Kinnison :-)
[05:21] <zyga> soyuz?
[05:21] <kiko> zyga, the package and distro-management side of launchpad
[05:22] <zyga> something an average person will ever see?
[05:22] <mpt> SteveA: sure
[05:22] <Kinnison> zyga: launchpad.net/distros
[05:22] <SteveA> mpt: so, here's what happens currently...
[05:22] <SteveA> as traversal happens, the request accumulates a set of breadcrumb links, starting at the root and working down the url
[05:23] <SteveA> traversal, and thus new breadcrumbs, will stop being added when an error is encountered
[05:23] <zyga> https://launchpad.net/distros/ubuntu
[05:23] <zyga> outdated -> 5.04 marked as current stable and 5.10 as devel ;] 
[05:23] <mpt> SteveA: ok, so https://launchpad.net/products/launchpad/is/teh/r0xor would show "Launchpad > Products > Launchpad"?
[05:23] <SteveA> so, for example a permissions error on some page will get the full set of breadcrumbs for that page (but no other menus or anything, because the page isn't actually rendered as such)
[05:24] <SteveA> mpt: yes
[05:24] <mpt> great
[05:26] <SteveA> with the latter "Launchpad" made strong
[05:26] <mpt> yup
[05:27] <SteveA> if you're happy with that, then great
[05:27] <kiko-fud> rock rock rock
[05:27] <SteveA> i don't need to do anything special
[05:29] <dilys> Merge to rocketfuel@canonical.com/launchpad--production--1.36: Cherry picks (patch-7: stuart.bishop@canonical.com, bjorn.tillenius@canonical.com, rocketfuel@canonical.com ...)
[05:30] <stub> I've updated the release statuses and dates in the db
[05:31] <Kinnison> stub: thanks
[05:43] <sabdfl> stub: can i trouble you for another db patch review?
[05:44] <sabdfl> stub: https://chinstrap.ubuntu.com/~dsilvers/paste/fileweR7pT.html
[05:44] <sabdfl> with comments of course
[05:45] <Kinnison> s/_spn_/_sourcepackagename_/ you cheat
[05:45] <stub> sabdfl: That another one for the statistician?
[05:46] <sabdfl> stub: yes
[05:46] <sabdfl> Kinnison: fixed
[05:47] <stub> sabdfl: What format will the data in binpkgnames, summaries & descriptions be in?
[05:47] <sabdfl> text, concatenated with spaces
[05:47] <sabdfl> it's just a cache for the fti for very fast searching
[05:48] <sabdfl> no joins required
[05:48] <stub> I see
[05:48] <sabdfl> we'll update it daily, and for any NEW package on acceptance
[05:48] <sabdfl> so it will be 99.99% correct
[05:48] <sabdfl> or otherwise, correct 99.99% of the time :-)
[05:49] <sabdfl> and even then, only very slightly wrong
[05:49] <sabdfl> you would have to be looking for a brand new package at just the wrong moment
[05:52] <stub> ok. This might be premature optimization (I havn't seen what the statistician needs to to do generate the cache), but the data model and concept is fine. patch-25-41-0
[05:53] <sabdfl> stub: i will test performance. the current searching is slow, i think this will make it fast, even on my laptop (which currently has dogfood db installed)
[05:54] <sabdfl> Kinnison's dogfood, with warty / hoary / breezy imported via gina
[05:54] <Kinnison> New "Kinnison dogfood" -- eight out of ten cats prefer it
[05:54] <Kinnison> or something
[05:59] <stub> Bed!
[06:01] <Lathiat> eek, https://launchpad.net/people/motu/+assignedbugs -> "Oops"
[06:02] <Lathiat> doesn't happen on everyone
[06:02] <Lathiat> so might be something specific to teams or the motu set 
[06:06] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: Fix a very inefficient query that breaks in production. r=stub (patch-2646: guilherme.salgado@canonical.com)
[06:12] <bradb> Lathiat: i'm working on it
[06:14] <Lathiat> bradb: cheers
[06:15] <Lathiat> https://launchpad.net/products/launchpad/+bug/3107
[06:15] <Ubugtu> Malone bug #3107: MOTU's '+assignedbugs' -&gt; System Error Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Nobody, Status: New http://launchpad.net/malone/bugs/3107
[06:15] <Lathiat> cute
[06:25] <Kinnison> flippin 'eck: https://chinstrap.ubuntu.com/~dsilvers/paste/fileH7TZsf.html
[06:27] <SteveA> both passed?
[06:27] <SteveA> yay
[06:28] <Kinnison> SteveA: that's 7 of 10 pages of uploader checks passing
[06:30] <SteveA> way cool
[06:40] <bradb> SteveA: in addition, it looks like changing that sample data broke a FOAF test, as i just found out from a failure email from pqm :/
[06:41] <SteveA> you didn't run the page and doctests ?
[06:42] <bradb> nope. they already passed for me. when you made that suggestion, i didn't think to wait another half hour for them to run again, because i didn't see what this would break (and, indeed, it's all of about two lines of one test that end up breaking)
[06:43] <Kinnison> carlos: ping?
[06:44] <bradb> salgado-univ: around?
[06:44] <carlos> Kinnison pong
[06:44] <Kinnison> carlos: Righty, I need a function I call, passing in a full path to a translations tarball, that imports it
[06:45] <SteveA> bradb: doesn't take half an hour
[06:45] <carlos> Kinnison I will need some extra info
[06:45] <SteveA> it takes a few minutes:  python test.py -f canonical.launchpad 
[06:45] <Kinnison> carlos: tell me what contract you want and I'll fulfil it
[06:45] <carlos> Kinnison like distribution release name + sourcepackagename
[06:46] <Kinnison> carlos: define and provide the API and I'll cope
[06:46] <carlos> Kinnison and also the sourcepackage version
[06:46] <carlos> ok
[06:46] <Kinnison> If you can write the interface and docstring and nopaste it I'll say if I can fill it
[06:46] <bradb> SteveA: what does that do differently from "make check"?
[06:46] <carlos> Kinnison email?
[06:47] <Kinnison> carlos: nopaste
[06:47] <SteveA> bradb: it runs only the functional tests under 'launchpad'
[06:47] <carlos> Kinnison ok
[06:47] <SteveA> bradb: a much better alternative to running no tests
[06:47] <SteveA> bradb: everyone should know basically how to operate the test runner. it isn't hard.
[06:47] <bradb> SteveA: isn't that where the "make check" bottleneck is?
[06:47] <SteveA> bradb: what are you talking about?
[06:48] <bradb> running the functional tests under 'launchpad'; isn't that what the real "make check" bottleneck is anyway?
[06:48] <SteveA> most of the stuff that is about the web application is under canonical.launchpad
[06:48] <Lathiat> hey guys, just curious, what framework does launchpad use? (if at all?)
[06:48] <SteveA> i don't know what "real make check bottleneck" means
[06:48] <SteveA> Lathiat: zope3, sqlobject, twisted
[06:49] <Lathiat> SteveA: cool thanks, whats the twisted stuff used for?
[06:49] <SteveA> bradb: many of the slow 'make check' tests are to do with baz imports... cscvs etc.
[06:50] <SteveA> Lathiat: xmlrpc servers, general async programming.
[06:51] <carlos> Kinnison could we assume that Launchpad will have always the info about which packages are published on a given distro release?
[06:51] <Lathiat> SteveA: is there an xmlrpc interface to malone yet/planned?
[06:51] <Kinnison> carlos: how do you mean?
[06:51] <Kinnison> carlos: You're being called as part of accepting an upload
[06:51] <SteveA> Lathiat: it's been discussed, but nothing implemented.  We might write a spec for it at UBZ.
[06:51] <SteveA> Lathiat: you know that malone has an email interface?
[06:51] <Lathiat> SteveA: nope?
[06:52] <carlos> Kinnison if the answer is yes I think the best thing is adding a method to SourcePackageRelease that attachs that tarball
[06:52] <Lathiat> xmlrpc would rock for a desktop bug filing app, or an app to sort through bugs and stuff for me
[06:52] <SteveA> BjornT: where can Lathiat read about malone's email interface?
[06:52] <carlos> Kinnison ok, then the answer is yes
[06:52] <Kinnison> carlos: yep, that sounds great
[06:52] <carlos> hmmm
[06:53] <carlos> ok
[06:54] <BjornT> SteveA, Lathiat: https://wiki.launchpad.canonical.com/MaloneEmailInterfaceUserDoc
[06:55] <SteveA> BjornT: it would be great to have a link there on the "file a bug" page of malone
[06:55] <SteveA> maybe in the info box, top left of the screen
[06:55] <SteveA> saying "Did you know that you can also _file bugs by email_?"
[06:56] <Lathiat> BjornT: thanks
[06:56] <SteveA> mpt: what do you think about raising awareness of the email interface in this way?
[06:57] <carlos> Kinnison It's going to take some extra minutes, I don't have the SSL certificate here and I need to find it and setup this browser to use "nopaste"
[06:59] <Lathiat> BjornT: nifty
[06:59] <Kinnison> carlos: mail me if you're stuck
[07:00] <Kinnison> carlos: but I can wait to verify your interface :-)
[07:00] <Kinnison> carlos: s'not like I don't have enough to do :-)
[07:00] <Lathiat> whos stuart bishop?
[07:01] <carlos> Kinnison then I will send you it later today or tomorrow morning, as soon as I'm able to connect my laptop to Internet, ok?
[07:01] <SteveA> Lathiat: stuart is the database administrator on the launchpad team
[07:01] <Lathiat> i more meant, does he have an irc nick :)
[07:01] <mpt> SteveA: good idea
[07:02] <carlos> Kinnison using that object the API will be really easy, just the tarball as argument and that's all (assuming the object has already the new version string stored)
[07:02] <SteveA> Lathiat: yes, he is stub
[07:02] <Lathiat> SteveA: thanks
[07:03] <SteveA> mpt: breadcrumbs are with pqm
[07:04] <Kinnison> carlos: okay, do you want a libraryfilealias, or a path on disk?
[07:05] <carlos> Kinnison I don't mind, whatever is easier for you
[07:05] <Kinnison> carlos: I'll give you wantever is easiest for you
[07:05] <Kinnison> carlos: I'll have it in the librarian by the time I call you
[07:05] <Kinnison> carlos: so If you want that, I'll give you that
[07:05] <Kinnison> carlos: in fact, best to assume you'll get a LibraryFileAlias object
[07:05] <carlos> Kinnison a librarian link is good for me
[07:06] <Kinnison> carlos: cool, assume you get a lfa instance then
[07:06] <Kinnison> well, a securityproxied one
[07:06] <Kinnison> but hey :-)
[07:06] <carlos> ok
[07:06] <carlos> ;-)
[07:07] <carlos> Kinnison when will be this on production?
[07:07] <carlos> this Tuesday?
[07:07] <carlos> is a bit early, isn't it?
[07:07] <Kinnison> carlos: this code will be running on the ftpmaster box
[07:07] <Kinnison> which won't be a pure production checkout
[07:07] <Kinnison> at least not to begin with
[07:08] <carlos> Kinnison ok, I need to land a patch for that, will try to get it ready tomorrow to get it merged on production next Monday
[07:08] <carlos> Kinnison is it ok for you?
[07:09] <Kinnison> yep that'd be grand
[07:09] <Kinnison> thanks dude
[07:09] <carlos> the patch is not just the new API you asked, we need a new way to handle the imports before dapper's translations are imported into Rosetta
[07:09] <carlos> ok
[07:11] <SteveA> vim has a split view option
[07:11] <mpt> ah, look at that, emacs does too
[07:11] <SteveA> of course
[07:13] <BjornT> mpt: can you resolve the conflicts in your translation-form branch? ping me when you're done.
[07:13] <mpt> ok
[07:18] <mpt> carlos: Was that you changing pofile-translate.pt?
[07:19] <carlos> mpt no idea, could be, I have my mind in so many places that I don't remember exactly all changes I have done recently
[07:25] <dilys> Merge to rocketfuel@canonical.com/launchpad--production--1.36: [trivial]  Cherry picks (patch-8: stuart.bishop@canonical.com, christian.reis@canonical.com)
[07:26] <mpt> this is no good, the slowness of emacs' keyboard shortcuts is greater than the quickness of its split view
[07:27] <carlos> Kinnison are we going to have a gina run with Hoary and Breezy archives on production?
[07:27] <Kinnison> carlos: yes
[07:27] <carlos> when?
[07:28] <SteveA> mpt: use vim
[07:30] <mpt> yeah, learn vim
[07:32] <carlos> Kinnison I want to move POTemplate table to use SourcePackageRelease instead of DistroRelease and SourcePackageName but I cannot do that unless all Ubuntu releases with translations (Hoary, Breezy and Dapper) have information inside the SourcePackageRelease table
[07:32] <Kinnison> carlos: right
[07:32] <Kinnison> well stub will probably get on with it when he wakes up
[07:34] <jordi> kiko-fud: sorry, I had just left office
[07:35] <carlos> ok, I will assume then that it will be done on time but without removing any field to do the migration later if needed
[07:35] <bradb> cprov: where is GPG_ERR_NO_ERROR defined? i'm getting an AttributeError running a test because it appears to be missing in my setup.
[07:36] <cprov> bradb: do you have lib/sourcecode/pyme ? 
[07:36] <mpt> BjornT: I'm not going to fix that conflict today, otherwise I'll be waiting too long for baz
[07:36] <mpt> I'll do the merge overnight
[07:36] <mpt> but thanks for your other review
[07:36] <cprov> bradb: check out you dists
[07:37] <BjornT> mpt: ok
[07:37] <bradb> cprov: 
[07:37] <niemeyer> Given a fully qualified arch/baz patch id, what's the fastest way in baz to see what was done in that patch?
[07:38] <bradb> bradb@oxygen:~/launchpad/sourcecode/pyme $ baz tree-id
[07:38] <bradb> rocketfuel@canonical.com/pyme--devel--0.6.1--patch-3
[07:39] <cprov> bradb: it's correct ! check for the gpgme packages installed
[07:39] <carlos> jordi: did you try to reach the potemplate's +admin page? are you allowed now?
[07:40] <cprov> bradb: libgpgme-error-dev and libgpg-error0
[07:41] <jordi> carlos: I can now
[07:41] <carlos> cool
[07:41] <bradb> cprov: I have the latter installed but not the former
[07:42] <jordi> carlos: yeah
[07:42] <niemeyer> SteveA: ping
[07:42] <SteveA> hi niemeyer 
[07:42] <jordi> Should we do one review-breezy now so I can get a clue?
[07:42] <bradb> cprov: in fact, there is no installation candidate for the former
[07:42] <niemeyer> SteveA: Hi!
[07:42] <jordi> carlos: ther'es a typo in that page
[07:42] <jordi> "soits" should be "so its"
[07:42] <niemeyer> SteveA: I have something like changelog david.allouche@canonical.com--2004/launchpad--branchdatastorage--0--patch-2
[07:42] <bradb> cprov: is this a new dependency that only exists in breezy, or am i on crack?
[07:43] <niemeyer> SteveA: What's the fastest way to turn that into a diff?
[07:43] <carlos> jordi, please, file a bug
[07:43] <carlos> I cannot fix it atm
[07:43] <SteveA> niemeyer: i don't know much about baz
[07:43] <carlos> busy with other changes...
[07:43] <SteveA> i just use it
[07:43] <cprov> bradb: not sure about breezy, didn't have time myself to upgrade 
[07:43] <niemeyer> Ouch
[07:43] <sivang> SteveA: Hi
[07:43] <SteveA> hi sivang 
[07:43] <SteveA> let's get that agreement done today ;-)
[07:44] <bradb> cprov: i can find anything for "libgpgme-error-dev"
[07:44] <bradb> can't, even
[07:44] <SteveA> niemeyer: i think you can look inside that and look at the patches.
[07:44] <SteveA> niemeyer: that's all i ever do.
[07:44] <SteveA> Keybuk would know
[07:44] <cprov> bradb: install the dev packages related to gpgme, they might solve the dependency for a while 
[07:44] <jordi> carlos: done
[07:44] <niemeyer> SteveA: Will try that, thanks
[07:44] <bradb> oh, it's libgpg-error-dev
[07:45] <cprov> bradb: later on I can write down which one is really needed
[07:45] <bradb> but i have that one installed too, hm
[07:46] <carlos> jordi thank you
[07:46] <jordi> np
[07:46] <Keybuk> niemeyer: do you have a tarball?
[07:46] <Keybuk> or a .patches style directoy?
[07:46] <Keybuk> or just that string?
[07:47] <Keybuk> baz get-changeset david.allouche@canonical.com--2004/launchpad--branchdatastorage--0--patch-2 ,,cset
[07:47] <cprov> bradb: cd sourcecode/pyme/; make clean; make
[07:47] <Keybuk> baz show-changeset --diffs ,,cset
[07:47] <Keybuk> kind of thing ...
[07:48] <bradb> cprov: that did it, thanks
[07:48] <niemeyer> Keybuk: I have the up-to-date tree
[07:48] <niemeyer> Keybuk: And that string
[07:49] <niemeyer> Keybuk: Which is from a past revision
[07:49] <cprov> bradb: enjoy 
[07:49] <Keybuk> ok, the two commands above give you the diff for that revision
[07:49] <niemeyer> Keybuk: Thank you very much
[07:50] <Keybuk> (you'll almost certainly need to register the archive too)
[07:50] <niemeyer> Keybuk: I have that tree checked out already
[07:50] <Keybuk> fair enough
[07:53] <niemeyer> Ahhh.. let there be diffs!
[07:53] <niemeyer> Keybuk: You saved my day. Thanks :)
[08:02] <Kinnison> It's not open currently
[08:02] <neilc> hmm, fair enough
[08:02] <neilc> why not?
[08:03] <neilc> someone suggested that malone would be a nice bug tracker for a project i'm working on to use
[08:03] <neilc> but i'm not especially interested in using launchpad.net et al., just malone
[08:03] <sabdfl> so with  @@ ftq(), is the sort order automatically in order of matching quality?
[08:04] <niemeyer> % grep '^+++ mod/lib/importd/archivemanager.py' test.diff | wc -l
[08:04] <niemeyer> 47
[08:04] <sivang> neilc: you can use malone for your project, you do not need the source AFAIK
[08:04] <sivang> neilc: you can just use malone, and be relieved of the enourmous technical details relating to its implementation :)
[08:04] <niemeyer> Can we please move to bzr now? :)
[08:05] <Kinnison> sabdfl: I can't find it in the readme :-(
[08:05] <neilc> sivang: i don't *need* the source, i guess, but i'd like it :)
[08:05] <sabdfl> niemeyer: +1
[08:06] <sabdfl> neilc: malone only makes sense as part of lp
[08:06] <sabdfl> bugzilla or roundup is more efficient if you just want to track bugs internally
[08:06] <sivang> neilc: what sabdfl also just said :)
[08:06] <neilc> sabdfl: hmm, perhaps i'm confused, then (someone had suggested malone as a nicer implementation of the debbugs concept)
[08:06] <sabdfl> if you want to link to distros, then you need to be part of the distro bugtracker, i.e. lp
[08:07] <neilc> sabdfl: this is just for email-centric bug tracking for a particular oss project (i don't care about integrating with distros etc.)
[08:07] <sabdfl> neilc: it is that, but its real killer feature is that you can pass a bug to the ubuntu guys, or another upstream, trivially
[08:07] <sabdfl> for that to work, they need to be using malone as well
[08:07] <neilc> sabdfl: oh, meh :) i just want debbugs++, to be honest
[08:07] <Kinnison> sabdfl: looks like it's a bit more complex than that
[08:07] <sabdfl> ok. well you don't need to use the rest of LP
[08:07] <Kinnison> sabdfl: there's a rank() function
[08:07] <Kinnison> it's all a bit odd
[08:08] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=salgado, trivial]  various improvements to navigation, and also breadcrumbs, and also redirection as a result of traversal. (patch-2647: steve.alexander@canonical.com)
[08:09] <sivang> neilc: however, if you use launchpad to register your application, a multitude of downstream and upstream "clients" would be able to use it, and thuse you will get more testing and exposure. IMHO that's a big benefit compared to the "price" of resgierting somethign in the lp.
[08:10] <moyogo> hi, i'd like to add translation projects for lingala for gnome and ubuntu, how do i go about this?
[08:12] <moyogo> btw when will we be able to translate launchpad ?
[08:12] <sabdfl> did stub announce a tag point for the next production update?
[08:13] <sabdfl> moyogo: good question. we have the basic infrastructure. will probably make rosetta itself i18N'd for rosetta 2.0
[08:15] <moyogo> is there a structure in place to put glossaries of terms on launchpad, or is each translation team supposed to keep that on their own site?
[08:15] <carlos> Kinnison do you have some time to talk about the translation imports with the new buildd? I have some questions about SourcePackageRelease
[08:15] <Kinnison> sure, you can ask
[08:15] <carlos> moyogo not yet
[08:15] <moyogo> this is really awesome btw
[08:15] <Kinnison> carlos: take it to /query so I don't lose your questions
[08:16] <carlos> moyogo about the import of your project, look at https://wiki.ubuntu.com/RosettaFAQ we have there the instructions to follow
[08:16] <carlos> moyogo the Ubuntu version will be available as soon as your project is added to Ubuntu's archive
[08:16] <carlos> Kinnison ok
[08:17] <moyogo> carlos: sorry I wasn't clear, i want to translate into lingala language
[08:18] <moyogo> carlos: thanks for the link
[08:19] <bradb> jblack: ping
[08:22] <SteveA> sabdfl: so, now you can say:  return redirection('/foo/bar')   from within a traversal function of any kind.  Import 'redirection' from webapp, just like Navigation and stepthrough etc.
[08:28] <carlos> moyogo ok, then follow the procedure on that FAQ, look at the section "Translating with Rosetta"
[08:29] <moyogo> carlos: just sent an email
[08:31] <SteveA> cprov: be careful how you put branches into PendingReviews -- when you make mistakes, the diffs aren't produced
[08:32] <cprov> SteveA: what did I do wrong ? 
[08:32] <SteveA> celso.providelo@c.c/launchapad--builddUI--0
[08:32] <SteveA> getting all italian
[08:32] <cprov> SteveA: correcting
[08:33] <cprov> SteveA: hope it does not brake the entire system
[08:34] <cprov> SteveA: fixed
[08:34] <kiko> break even :)
[08:35] <SteveA> kiko: i think that was right... it's slowing me down ;-)
[08:36] <cprov> dudes, whatever, fixed ;) 
[08:38] <kiko> it's slowing is all down dammit
[08:43] <SteveA> cprov: first review done.
[08:43] <cprov> SteveA: thank you
[09:07] <mpt> woohaa
[09:08] <kiko> rock rock rock mpt 
[09:08] <mpt> and there's more where that came from
[09:09] <mpt> Should speed up Launchpad a bit
[09:13] <SteveA> matsubara / gneuman: seen this?  https://launchpad.net/products/launchpad/+bug/3103
[09:13] <Ubugtu> Malone bug #3103: Cannot edit the assignee and priority of a ticket Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Steve Alexander, Status: Accepted http://launchpad.net/malone/bugs/3103
[09:13] <salgado> bradb, pong
[09:14] <bradb> salgado: n/m, problem fixed thanks
[09:15] <kiko> Kinnison, what does DCU do?
[09:15] <kiko> cprov, this change makes main higher-priority, right?
[09:15] <kiko>       [trivial]  Increase the weight of component when scoring build jobs
[09:15] <gneuman> SteveA, i am stuck with a problem, gotta go back to bug 2705 urgent
[09:15] <Ubugtu> Malone bug #2705: A number of pages are untested and currently broken Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Diogo Matsubara, Status: Fixed http://launchpad.net/malone/bugs/2705
[09:16] <SteveA> no worries, i'll take a look at it
[09:16] <kiko> gneuman, there are a number of pages still left untested and broken you know
[09:16] <cprov> kiko: yes, it does, uses higher index than urgency, they are more relevant now
[09:16] <gneuman> i know
[09:17] <matsubara> SteveA: got that Steve. I will make a pagetest for it
[09:17] <kiko> cprov, rock on
[09:17] <SteveA> matsubara: really?  cool.  looks to me like it's just the menu item that is wrong -- should point at +priority.
[09:18] <cprov> kiko: it does ;)
[09:18] <moyogo> err.... how do the plural items work? Were can I read docs about that?
[09:18] <SteveA> moyogo: what language are you interested in?
[09:18] <bradb> SteveA: got a sec for a quick q about one last hickup with the null priority fix?
[09:18] <SteveA> bradb: okay
[09:19] <bradb> SteveA: accessing a URL path like: /products/evolution/+bug/7/+viewstatus/wobbly
[09:19] <bradb> i get:
[09:19] <moyogo> SteveA: lingala and possible kiswahili
[09:19] <bradb>     *  Module zope.app.publication.publicationtraverse, line 51, in traverseName
[09:19] <bradb>       ob2 = ob.publishTraverse(request, nm)
[09:19] <SteveA> moyogo: do you have any po files for this already?
[09:19] <bradb> ForbiddenAttribute: ('publishTraverse', <zope.app.pagetemplate.simpleviewclass.SimpleViewClass from /home/bradb/malone-priority-fix/lib/canonical/launchpad/templates/bugtask-view.pt object at 0x329ad1b0>) 
[09:19] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=SteveA]  allow admins to edit any bug, public or private. also add a Launchpad Developers team, make them a celebrity and allow them to see error messages. (patch-2648: brad.bollenbach@canonical.com)
[09:19] <moyogo> SteveA: the problem is that there are noun classes
[09:19] <bradb> but +bug/6/... returns a 404
[09:19] <moyogo> SteveA: no, not yet
[09:19] <SteveA> moyogo: maybe you can find some, and copy the plural form expression from there?
[09:20] <bradb> SteveA: any idea on why the URL above would also not return a 404?
[09:20] <matsubara> SteveA: well, that probably should be fixed too, doesn't it? 
[09:20] <moyogo> SteveA: I'm in contact with people who've already translated abiword in lingala tho'
[09:20] <SteveA> matsubara: i think that all that needs fixing is the menu item.  the page that has the menu on it can have a test that it points to +priority
[09:21] <SteveA> moyogo: okay, maybe they can give you the right plural form thing for that
[09:21] <moyogo> SteveA: I don't think gettext can handle it
[09:22] <moyogo> SteveA: it's stuff like moninga -> baningo, ese -> bise
[09:22] <bradb> SteveA: note: the key difference between the .../+bug/6/... and .../+bug/7/... of those URLs above is that bug 7 actually exists on that product, bug 6 does not
[09:22] <moyogo> etc.
[09:22] <Ubugtu> Malone bug #7: Need newbie documentation Fix req. for: rosetta (upstream), Severity: Normal, Assigned to: Nobody, Status: New http://launchpad.net/malone/bugs/7
[09:22] <Ubugtu> Malone bug #6: "next 10 entries" at bottom of page Fix req. for: rosetta (upstream), Severity: Normal, Assigned to: Carlos Perello Marin, Status: Rejected http://launchpad.net/malone/bugs/6
[09:22] <SteveA> bradb: http://localhost:8086/products/evolution/+bug/7/+viewstatus/wobbly  <-- i get a normal not found
[09:22] <moyogo> SteveA: sorry, meant moninga -> baninga
[09:22] <matsubara> ok, just assign it to me. I'll try to fix it later.
[09:22] <SteveA> matsubara: okay, cool
[09:22] <bradb> SteveA: you would. the evo bug only exists in my sampledata.
[09:22] <bradb> i.e. bug 7 really does exist here
[09:22] <Ubugtu> Malone bug #7: Need newbie documentation Fix req. for: rosetta (upstream), Severity: Normal, Assigned to: Nobody, Status: New http://launchpad.net/malone/bugs/7
[09:22] <moyogo> SteveA: I'll look how the kiswahili translators have handled it, it's pretty similar
[09:23] <bradb> SteveA: try, say /products/firefox/+bug/1/+viewstatus/wobbly
[09:23] <carlos> Seveas you are the ubugtu author, right?
[09:23] <cprov> kiko: btw, don't we have special marker in fmt:text-to-html for code, something like {{{XXX}}} ?
[09:23] <kiko> cprov, to handle preformatted text? alas, no. complain to mpt.
[09:24] <carlos> Seveas Could I suggest you that the report include who did the change?
[09:24] <SteveA> bradb: okay, so i see the forbidden attr error there.
[09:24] <cprov> kiko: no blame is required but that feature should be nice
[09:25] <kiko> yeah, it would rock
[09:25] <kiko> mpt, are you against a marker that indicated <pre> text in our formatters?
[09:25] <Kinnison> kiko: Hmm?
[09:25] <kiko> o great Kinnison 
[09:25] <kiko> share thy wisdom
[09:26] <Kinnison> one sec
[09:26] <Kinnison>  /quit
[09:26] <kiko> what does the venerable DCU do?
[09:26] <Kinnison> Right
[09:26] <Kinnison> sorry, just quit/reloaded irssi
[09:26] <Kinnison> DCU == DistroComponentUploader yes?
[09:27] <kiko> correctamundo
[09:27] <Kinnison> A record of who can upload what to where. Distributions are permitted to have multiple components. Those components are often subject to different uploader constraints. This table represents those variable constraints by linking a team to a distribution,component tuple.
[09:27] <kiko> okay
[09:27] <kiko> understood completely
[09:27] <kiko> thanks
[09:27] <Kinnison> next time, try \dd <thing>
[09:27] <Kinnison> :-)
[09:27] <kiko> I hadn't merged yet, sorry
[09:27] <Kinnison> s'okay
[09:28] <kiko> I often will go for a while without merging and I like writing up the report as commits come in
[09:28] <kiko> otherwise it's terrible
[09:28] <Kinnison> btw, if I haven't already, thanks for the reports
[09:28] <Kinnison> they're really useful.
[09:28] <bradb> SteveA: there's a bit of special handling here in BugTaskNavigation, like:
[09:28] <bradb>         if name in ("+viewstatus", "+editstatus"):
[09:28] <bradb>             if INullBugTask.providedBy(self.context):
[09:28] <bradb>                 # The bug has not been reported in this context.
[09:28] <bradb>                 return None
[09:28] <bradb>             else:
[09:28] <bradb>                 # The bug has been reported in this context.
[09:28] <bradb>                 return getView(self.context, name + "-page", self.request)
[09:28] <kiko> ah, thanks Kinnison, nice that it's appreciated
[09:28] <bradb> so that +viewstatus and +editstatus are 404s for nullbugtasks
[09:29] <SteveA> i want to simplify that, btw
[09:29] <SteveA> i saw that when i was doing the navigation refactor
[09:29] <kiko> cprov, can you explain NominatedArchIndep -- ?
[09:30] <Kinnison> kiko: Very simply, each distrorelease nominates one distroarchrelease on which arch-independant packages are built
[09:30] <cprov> kiko: sure, is a specific distroarchrelease used to build architecture independent packages
[09:30] <kiko> ah!
[09:30] <Kinnison> kiko: because they should only be built once for the whole distrorelease
[09:30] <kiko> yes
[09:30] <kiko> understood
[09:30] <kiko> thanks
[09:30] <bradb> SteveA: Sure, if it can be written more simply and still clearly communicate the 404 vs. not-404, I'm all for it.
[09:30] <SteveA> i'm looking into it though
[09:30] <mpt> kiko: No, I'm not against it
[09:31] <kiko> mpt, thanks
[09:31] <mpt> (except that it might make people think we take wiki markup when we don't)
[09:31] <kiko> mpt, we could use an alternative marker.. 
[09:31] <mpt> and that page-widening miscreants could use it
[09:31] <kiko> ;)
[09:33] <bradb> SteveA: On the face of it, it looks as though Z3 may behave differently when you return an object from getView in a traverser, as opposed to when that same view is instead a "normal" view simply registered in ZCML, IYKWIM
[09:34] <elmo> Kinnison: does gina still need the projectb dump?
[09:34] <Kinnison> elmo: No, she works entirely from the archive
[09:35] <elmo> thanks
[09:41] <Kinnison> bradb: Les Cowboys Fringants are amusing
[09:41] <Kinnison> bradb: thanks for the recommendation
[09:41] <bradb> de rien
[09:42] <kiko> bradb, can you fix bug 3096's description so I understand the problem?
[09:42] <Ubugtu> Malone bug #3096: SQLObject.select appears to generates bogus SQL if an empty string is passed as the first arg Fix req. for: launchpad (upstream), Severity: Normal, Assigned to: Andrew Bennetts, Status: New http://launchpad.net/malone/bugs/3096
[09:42] <Kinnison> bradb: tu sais que cvd peut bien parler en franais aussi?
[09:42] <kiko> you seem to have bogused your example
[09:42] <bradb> Kinnison: ouais, amusant
[09:43] <kiko> elle parle trs bien, aussi
[09:43] <bradb> kiko: not sure that i can write it much more clearly than giving the word-for-word recipe + code for how I reproduced it.
[09:43] <kiko> bradb, the code is bogus
[09:43] <bradb> tr
[09:43] <bradb> er
[09:43] <bradb> trs, that is
[09:43] <kiko> return Bug.select("", **other_params)
[09:43] <kiko>         if where_clauses:
[09:43] <kiko> that's bogus :)
[09:43] <Kinnison> bradb: ouais
[09:43] <kiko> aiee
[09:43] <bradb> kiko: run it and you'll see :)
[09:44] <bradb> yes, there's a premature return in there, but that's exactly the code i used to reproduce the problem.
[09:44] <kiko> bradb, dude
[09:44] <kiko> the premature return means the rest of the code is unreachable
[09:44] <kiko> so nuke the rest of the code out
[09:44] <kiko> there is no way the python interp can reach the post-return code
[09:45] <kiko> it's just noise
[09:45] <Kinnison> lua wouldn't let you do that :-)
[09:45] <kiko> and it confuses me as to what needs testing :)
[09:45] <bradb> kiko: i'm aware of that. i just left it there to be clear about exactly what i changed.
[09:45] <kiko> Kinnison, make lint would also slap his wrist
[09:45] <Kinnison> kiko: *g*
[09:45] <Kinnison> kiko: It's actually annoying in lua :-)
[09:45] <kiko> bradb, that's anything but clear 
[09:45] <bradb> i'll edit it if it makes you feel better though ;)
[09:45] <Kinnison> kiko: "return" has to be the last statement in a block
[09:45] <Kinnison> kiko: so you find people doing: "do return end"
[09:45] <kiko> heh
[09:45] <Kinnison> kiko: to insert an early return for debugging
[09:46] <kiko> ah
[09:46] <kiko> how interesting
[09:46] <kiko> because of nested blocks it works, wow
[09:46] <Kinnison> Lua 5.0.2  Copyright (C) 1994-2004 Tecgraf, PUC-Rio
[09:46] <Kinnison> > do bar() ; return 3 ; foo() end
[09:46] <Kinnison> stdin:1: `end' expected near `foo'
[09:46] <bradb> kiko: is that clearer?
[09:47] <kiko> I've heard coke's going to be outlawed in brazil
[09:47] <kiko> truth or myth? :)
[09:48] <Kinnison> the girls in the office didn't like my coffee+coke
[09:48] <Kinnison> and they didn't even try it
[09:48] <Kinnison> and they didn't believe me that espress+coke is very common
[09:48] <SteveA> "common" in what sense?
[09:48] <SteveA> (peasant!)
[09:48] <kiko> in the sense "only rabble do it"
[09:49] <bradb> SteveA: any idea about that view problem then? it smells like something deep down in the CA.
[09:49] <SteveA> Kinnison: "double fisting diet coke"?  is that like the really hard red bull test?
[09:49] <Kinnison> SteveA: no, but I was wondering who would suggest it might be
[09:50] <Kinnison> kiko: If you ever need a serious pick-me-up, take a pint of coke and depth-charge a double-espresso
[09:50] <SteveA> --- orig/src/zope/app/form/browser/schemadisplay.py
[09:50] <SteveA> +++ mod/src/zope/app/form/browser/schemadisplay.py
[09:50] <SteveA> @@ -86,7 +86,8 @@
[09:50] <SteveA>          menu and globalBrowserMenuService.getMenuUsage(menu)
[09:50] <SteveA>          )
[09:50] <SteveA>      defineChecker(class_,
[09:50] <SteveA> -                  NamesChecker(("__call__", "__getitem__", "browserDefault"),
[09:50] <SteveA> +                  NamesChecker(("__call__", "__getitem__",
[09:50] <Kinnison> kiko: just don't do it the other way round or it'll explode messily
[09:50] <SteveA> +                                "browserDefault", "publishTraverse"),
[09:50] <SteveA>                                 permission))
[09:50] <SteveA> 
[09:50] <SteveA> brad
[09:51] <SteveA> that is the change required
[09:51] <SteveA> there is nothing wrong with how you are using getView()
[09:51] <SteveA> other than i find it a bit ugly ;-)
[09:51] <bradb> er, oh, it is in zope, hm
[09:51] <SteveA> yes, it is
[09:51] <SteveA> it's fix for upstream
[09:51] <SteveA> if it isn't already there
[09:51] <bradb> in the meantime, i can test this screen by not testing the ../wobbly URL, if you want
[09:52] <bradb> i.e. only test the page itself when priority is null, but not an extra bogus path element hanging off of it
[09:52] <SteveA> i want nothing but your happiness and satisfaction with a bug free malone
[09:52] <bradb> cool. rm -rf templates/*portlet*
[09:52] <bradb> :P
[09:53] <bradb> testing just the page itself can work for now
[09:54] <Kinnison> sabdfl: buy me more RAM
[09:54] <Kinnison> sabdfl: please?
[09:54] <Kinnison> sabdfl: kthxbye
[09:54] <Kinnison> ye gods
[09:55] <Kinnison> how much RAM is that?
[09:55] <sabdfl> guys, i tested rocketfuel on bzr this morning. SECONDS.
[09:55] <bradb> upgrading two powerbooks is a serious thing
[09:55] <Kinnison> and how little is $ca worth?
[09:55] <bradb> Kinnison: about 400 GBP
[09:55] <Kinnison> bradb: FFS
[09:55] <Kinnison> bradb: I'm expecting to spend ca. 80 quid to add a gig to my laptop
[09:55] <Kinnison> maybe a bit more
[09:55] <bradb> Kinnison: added 1G to each machine
[09:56] <Kinnison> bradb: 1G for my laptop will be 97 quid apparently
[09:57] <kiko> Kinnison, that's damned cheap
[09:57] <kiko> I need to buy 1gig of ECC mem
[09:57] <kiko> how much does that cost?
[09:57] <kiko> (for the server)
[09:57] <Kinnison> kiko: what laptop/model?
[09:57] <Kinnison> oh right
[09:57] <Kinnison> ECC DDR?
[09:57] <kiko> it's a server, DDR333 or something
[09:57] <kiko> yeah
[09:58] <Kinnison> buffered or unbuffered?
[09:59] <Kinnison> kiko: and do you want it in one stick or two?
[09:59] <Kinnison> Kiko: DDR PC2700  CL=2.5  REGISTERED  ECC  DDR333  2.5V  128Meg x 72 -- one 1 gig dimm -- 120 quid all-in
[10:00] <carlos> wow, finally, back to Linux
[10:02] <SteveA> cprov: you know that branch you mistyped on PendingReviews?
[10:03] <SteveA> cprov: do you need it reviewed today?
[10:03] <SteveA> cprov: i'm trying to manually get a diff, but even on chinstrap it is taking forever
[10:03] <cprov> SteveA: yes, it'd very nice to have it reviewed today
[10:04] <SteveA> but i mean, are there *consequences* if it waits until tomorrow morning?
[10:04] <cprov> SteveA: but if you can't no problem,  sorting out gzip-buildlog is a step anyway 
[10:05] <cprov> SteveA: that's what I'm saying, nobody will die 
[10:05] <SteveA> okay.  i'll leave it merging here.  i'll go home and eat and sleep, and i'll review it tomorrow morning.
[10:05] <cprov> SteveA: fair enough, take your time ;)
[10:06] <SteveA> and until we have bzr... double check you wrote things correctly on PendingReviews
[10:06] <cprov> SteveA: my BIG fault, sorry 
[10:07] <cprov> SteveA: see the gzip example code I wrote in bug # 3111, it probably help you to understand the open alias 
[10:08] <SteveA> ok
[10:08] <Kinnison> gzipped build logs is a dapper-open requirement
[10:08] <Kinnison> (just FYI)
[10:09] <Kinnison> SteveA: I have a python list which is a set of filenames
[10:09] <SteveA> the diff should have finished by the time i return tomorrow
[10:09] <Kinnison> SteveA: I want to sort it asciibetically by basename
[10:09] <Kinnison> SteveA: is there a neat way
[10:10] <Kinnison> ?
[10:10] <SteveA> L.sort(key=os.path.basename) perhaps
[10:11] <SteveA> or, if you don't want to be destructive, the same with sorted(L, key=...)
[10:11] <kiko> Kinnison, 120 quid doesn;'t sound bad!
[10:11] <Kinnison> SteveA: aha, coolio
[10:13] <elmo> Kinnison: macquarie has all 6 architectures now
[10:13] <Kinnison> elmo: cool
[10:14] <Kinnison> elmo: what's the upstream bw here?
[10:14] <elmo> 768kbit or so
[10:14] <elmo> and I can tell you're using all of it :(
[10:14] <Kinnison> urgh
[10:14] <Kinnison> sclag
[10:14] <Kinnison> s/cl/chl/
[10:14] <Kinnison> well, I'm using enough to ACK my email
[10:15] <kiko> Kinnison, that's a great price, actually. Do you think I can find memory for this price in Montreal?
[10:15] <Kinnison> kiko: dunno, but I'm sure if you asked someone in .uk nicely enough they could buy it for you and bring it over
[10:16] <Kinnison> kiko: Or you could see if www.cruicial.com can post to you in brazil
[10:16] <Kinnison> 120 quid is less than the maximum personal import limit isn't it?
[10:16] <kiko> it's about 3x the limit actually 
[10:16] <Kinnison> oh :-(
[10:16] <kiko> (via mail)
[10:17] <Kinnison> aah
[10:20] <sabdfl> Kinnison: it's show and tell time
[10:20] <Kinnison> sabdfl: rock on, one sec
[10:27] <bradb> http://www.pignonsurroues.com/2boutique.html -- would renting hybrids be lame or what?
[10:28] <kiko> bradb, oh COME ON
[10:28] <kiko> bradb, I'd accept renting a road bike as a second
[10:28] <bradb> ah, right, that would probably be much easier
[10:28] <kiko> are there not lots of mountain bikers in montreal?
[10:28] <bradb> hell yeah. my 'hood is like the amsterdam of NA.
[10:29] <kiko> wonder why it's difficult to rent then
[10:29] <bradb> bikes everywhere. and bike shops, bike cafes, bike everything.
[10:29] <bradb> just not easy to find good mountain bikes to rent
[10:30] <bradb> (like in amsterdam, most people ride crap bikes in my 'hood :P)
[10:31] <bradb> kiko: what size road bike do you wear?
[10:32] <kiko> bradb, 52 or 54cm
[10:32] <bradb> ok
[10:35] <salgado> kiko-afk, can you tell me what you see at https://launchpad.net/people/meanroy/+editwikinames
[10:37] <kiko-afk> looking
[10:37] <salgado> bradb, would you do that (^) for me?
[10:37] <salgado> ooops, nm, bradb 
[10:37] <kiko-afk> rent a bike
[10:37] <kiko-afk> ?
[10:37] <salgado> no, tell me what's in that page
[10:38] <mpt> 'night all
[10:38] <kiko-afk> salgado, I've pasted it. is that enough?
[10:38] <Kinnison> who on earth thought it was a good idea to add a schema called enema?
[10:38] <salgado> kiko-afk, yes, ta!
[10:38] <kiko-afk> stub, probably, Kinnison 
[10:38] <jbailey> Kinnison: Is that for storing data that you want flushed?
[10:38] <Kinnison> jbailey: I dread to think
[10:40] <bradb> lifeless: ping
[10:43] <bradb> alas, email seems a more reliable option
[10:47] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  re-sign the dsc, fix and re-sign the changes. add getFileByName to Distribution. r=salgado (patch-2649: daniel.silverstone@canonical.com)
[10:47] <Kinnison> yay!
[10:47] <Kinnison> sabdfl: also it means you get the satisfaction of regular landings
[10:48] <Kinnison> sabdfl: that merge was for patch-22 on my upload-and-queue branch :-)
[10:48] <sabdfl> has stub branched yet?
[10:52] <Kinnison> What does this mean:
[10:52] <Kinnison> You should not import __doc__ from _strptime: canonical.archivepublisher.nascentupload
[10:52] <Kinnison> ?
[11:08] <sabdfl> SteveA: is there a way to have a menu item in the overview menu of object A take you to a page for object B?
[11:08] <sabdfl> specifically
[11:08] <sabdfl> i want the "Search Packages" menu of the distrorelease to take you to the *distribution* search page
[11:11] <Kinnison> sabdfl: stevea has gone home for the night I think
[11:11] <Kinnison> morning spivveroonie
[11:17] <sabdfl> hey spiv
[11:27] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Sort the publisher file lists by basename (patch-2650: daniel.silverstone@canonical.com)
[11:27] <Kinnison> yay
[11:28] <Kinnison> Exception verifying version is newer: Unknown SQL builtin type: <type 'zope.security._proxy._Proxy'> for <Item RELEASE (0) from <class 'canonical.lp.dbschema.PackagePublishingPocket'>>
[11:29] <Kinnison> FRAZZLE WAZZLE WAZZLE MUMBLE GROAN COMPLAIN COMPLAIN
[11:29] <kiko-afk> heh
[11:29] <lifeless> moin
[11:30] <lifeless> fuzzywuzzywasawarriorwayayayix
[11:30] <Nafallo> hmm, fabbione was right... you guys are scary :-P.
[11:31] <Kinnison> :-)
[11:31] <Kinnison> Nafallo: we're lovely really
[11:31] <Nafallo> Kinnison: I know ;-)
[11:32] <lifeless> I'm not!
[11:32] <Kinnison> lifeless: sodding individual
[11:32] <Nafallo> just have to met you guys so that I can be fully convinced ;-)
[11:32] <Kinnison> Nafallo: coming to UBZ.
[11:32] <Kinnison> ?
[11:33] <ajmitch> Nafallo: oh they're very pleasant chaps
[11:33] <Kinnison> although by the time UBZ comes around, mdz may have chopped off some of my extremeties :-)
[11:33] <Nafallo> Kinnison: nope, girlfriend has her 18 year anniversary and then moves in that week :-/
[11:33] <ajmitch> as long as we can upload at UBZ it'll be fine :)
[11:34] <Kinnison> Nafallo: :-(
[11:34] <Kinnison> ajmitch: If you can't upload well before UBZ I'll be dead, let alone missing extremeties
[11:34] <Nafallo> I'll hope I can make it for dapper+1 :-P
[11:34] <ajmitch> Kinnison: yes, I think the queue of pending uploads is starting to grow already
[12:03] <Kinnison> ARGH, sodding securityproxy