[03:37] <mpt> jamesh / spiv, ping
[03:37] <jamesh> mpt: pong
[03:38] <mpt> jamesh, remember how I asked you the other day if it was ok to delete a standalone branch once I'd imported it into my repository?
[03:39] <jamesh> yes.  I said if you had a copy of the branch in the repo it'd be okay to delete the original standalone branch.
[03:40] <mpt> well, today, I had a disk error
[03:40] <mpt> Almost everything that got corrupted was checkouts, which doesn't matter
[03:41] <mpt> but, one other thing that fsck deleted was myrepo/trivial/.bzr/branch/revision-history
[03:41] <jamesh> okay.
[03:41] <mpt> and of course I don't have the standalone copy any more
[03:41] <jamesh> if the branch had been pushed to devpad, you should be able to copy the file from there
[03:41] <mpt> that's what I was wondering
[03:42] <spiv> For that matter, you could simply branch back off devpad.  It should be fairly quick if you already have all the revisions in the local repo.
[03:43] <mpt> Meanwhile, is it ok to keep importing other branches into my repository?
[03:44] <mpt> or will bzr get all confused that some of the revisions it was counting on are in a branch that doesn't seem to exist any more?
[03:44] <spiv> If you're not sure, do a "bzr check" in the repository.
[03:44] <mpt> ah, of course
[03:45] <spiv> But if nothing has been touched in the .bzr/repository directory, then I expect the repository itself is fine.
[03:45] <spiv> Corruption of branches below the repository doesn't affect the repository itself.
[03:46] <mpt> bzr check says "No WorkingTree exists for file:///home/mpt/hacking/lprepo/.bzr/checkout"
[03:46] <mpt> I don't understand that, the repo itself isn't a branch
[03:47] <spiv> I think you need to run it inside a branch :/
[03:48] <mpt> ok, every branch gives me the same error
[03:48] <mpt> customized to the name of that branch
[03:49] <mpt> Time for a new repository? :-] 
[03:49] <jamesh> mpt: try doing a lightweight checkout of one of the branches, and run "bzr check" there
[03:49] <jamesh> mpt: I think that'll check all the revisions in the repository (which is what you want)
[03:51] <mpt> hmm, would that be a bug that "bzr check --help" doesn't mention it works only on checkouts or standalone branches?
[03:55] <jamesh> possibly.  May as well report it
[03:55] <jamesh> (the bug might be that it doesn't work without a working tree)
[03:56] <mpt> ok
[03:56] <mpt> (at this rate, it looks like bzr check will take about an hour)
[03:58] <jamesh> it is probably worth doing if you are unsure about the state
[03:59] <mpt> yeah
[03:59] <mpt> thanks jamesh, thanks spiv
[04:10] <mpt> (reported bug 64783)
[04:10] <Ubugtu> Malone bug 64783 in bzr ""bzr check" returns unhelpful "No WorkingTree" error" [Undecided,Unconfirmed]  http://launchpad.net/bugs/64783
[05:01] <sicmipa> hallo
[05:02] <mpt> hi
[05:04] <sicmipa> is this room for ubuntu user?
[05:06] <jamesh> sicmipa: you might want to try #ubuntu
[05:06] <jamesh> sicmipa: this channel is for discussion/development of Launchpad (https://launchpad.net)
[05:09] <sicmipa> oh sorry
[05:09] <sicmipa> bye
[06:56] <jamesh> lifeless: could you subscribe launchpad-security to https://launchpad.net/products/launchpad/+bug/31287 ?
[06:56] <jamesh> (or anyone else on the launchpad admins team)
[07:03] <lifeless> jamesh: I'm on the team ;)
[07:03] <lifeless> and, its assigned to me
[07:03] <lifeless> interesting.
[07:03] <lifeless> anyhow, yes, subscribing
[07:04] <lifeless> done
[07:04] <lifeless> jamesh: ^
[07:04] <jamesh> lifeless: thanks.
[07:08] <jamesh> I wonder how many more private bugs we have that the launchpad developers don't have access to?
[07:08] <lifeless> well spiv ahd access
[07:08] <lifeless> and I
[07:08] <lifeless> but yeah, lets do a query
[07:27] <stu1> Off for a few hours for dental stuff
[08:00] <mpt> spiv, "want to people to know" -> "want people to know" :-)
[08:00] <mpt> otherwise, nice article
[08:01] <spiv> mpt: thanks
[08:01] <mpt> also, The renderer named 'menuitems' was not found in <backwards.consecutive.page.EntryPage object at 0x40b3b5ac>.
[08:01] <spiv> Yeah, I know.
[08:01] <spiv> This is my first blog entry in over a year, I had to spend a while remembering how to use the software :)
[08:01] <spiv> (just dropping a file in a directory, but still...)
[08:48] <jamesh> spiv: maybe we should just add weblogging support to Launchpad ...
[08:50] <spiv> jamesh: nah, I should just use my blog more than once every 14 months.
[08:51] <jamesh> spiv: my blogs.gnome.org one uses NewsBruiser -- it looked pretty good when they chose to use it, but stopped getting updated ...
[08:52] <jamesh> we fixed a number of the more annoying bugs but it definitely has some bad scalability problems
[09:52] <SteveA> jamesh, bjornt: http://www.zope.org/Collectors/Zope3-dev/721  <-- talking about some similar things to our formlib stuff
[10:00] <jamesh> SteveA: looks interesting
[10:02] <BjornT> SteveA: yeah, i saw it too, but didn't have time to take a closer look at it.
[10:03] <jamesh> the patch looks fairly trivial
[10:06] <BjornT> yeah, and it looks non-intrusive as well (i.e, it wouldn't break anything).
[10:47] <sivang> morning folks
[11:58] <poolie> hello
[12:11] <lifeless> reviewer meeting in 50 minutes
[01:00] <fabbione> malcc: ping?
[01:00] <malcc> fabbione: Pong
[01:00] <fabbione> malcc: do you have any idea why Contents files are no longer generated for edgy?
[01:00] <fabbione> malcc: this was supposed to be fixed eons ago
[01:00] <malcc> fabbione: I don't have a specific idea, but it probably has to do with the apt-ftparchive upgrade on drescher on Friday evening
[01:01] <malcc> fabbione: Thanks for letting us know it's broken; we'll get right on it
[01:01] <fabbione> malcc: last generation was done in Jul ?
[01:01] <malcc> fabbione: Ah, in that case it sounds like it was never fixed
[01:01] <fabbione> it was supposed to
[01:01] <fabbione> and i am pretty sure it has been running for a while
[01:02] <lifeless> review meeting time
[01:02] <fabbione> + (once again) we have tools that use the Contents files to generate some info for distro
[01:02] <fabbione> that means these tools aren't working
[01:02] <lifeless> == Agenda ==
[01:02] <lifeless>  * Roll call
[01:02] <lifeless>  * Queue status.
[01:02] <lifeless> - Roll Call -
[01:02] <lifeless> I'm here
[01:02] <spiv> I'm here
[01:03] <BjornT> here
[01:03] <malcc> fabbione: I've never looked at this area, let me come up to speed on what's been done and find out what the issues are
[01:03] <lifeless> jamesh: ?
[01:03] <lifeless> SteveA: ?
[01:03] <fabbione> malcc: thanks
[01:04] <fabbione> malcc: check LP.. there was about about it
[01:04] <jamesh> lifeless: here.
[01:04] <lifeless> ok
[01:04] <lifeless> queue status:
[01:04] <lifeless> 13 reviews
[01:04] <lifeless> > 4 days - 5 reviews
[01:04] <lifeless> thats a backslide :(
[01:05] <spiv> Two of them are mine.
[01:05] <jamesh> the two old ones are mine.
[01:05] <lifeless> and one bjornts, who got the 8K line branch - sorry dude.
[01:05] <spiv> They were going to be done on the train to and from Martin's today, but my laptop chose to fail to resume both times!
[01:05] <lifeless> I've tried to balance the review load, but its a little hard :(
[01:05] <BjornT> lifeless: yeah, i'm currently reviewing it, but it's taking a bit longer than usual since it's so big.
[01:06] <lifeless> BjornT: be very sure to include the time spent reviewing in your review message then !
[01:06] <spiv> (or rather, X failed to resume(
[01:06] <lifeless> BjornT: I'm esspecially interested in if it non-linear you see
[01:06] <spiv> I will get the two 5 day ones assigned to me done tonight (i.e. the next couple of hours)
[01:07] <lifeless> jamesh: whats up with yours ?
[01:07] <jamesh> I got a bit sidetracked.  I'll finish off kiko's one tomorrow
[01:08] <lifeless> and stubs ?
[01:08] <jamesh> his one looks a bit easier, so I might do that one first (kiko is away at the moment too)
[01:09] <lifeless> sounds good.
[01:09] <lifeless> any extra business ?
[01:09] <lifeless> 5
[01:09] <lifeless> 2
[01:09] <lifeless> 4
[01:09] <lifeless> 1
[01:09] <lifeless> 3
[01:09] <lifeless> -- meeting over --
[01:10] <jamesh> be nice if people didn't do whitespace cleanup of code they aren't changing
[01:10] <lifeless> if you can chat for a second, I've something to float
[01:10] <lifeless> whoops, sorry there
[01:10] <jamesh> do we have a policy on that?
[01:10] <lifeless> -- meeting onclosed --
[01:10] <lifeless> I'm not aware of a policy
[01:10] <lifeless> but I certainly think it is nicer to us, but on the other hand, cleanups are cleanups. 
[01:11] <jamesh> While in general whitespace cleanup is good, it would be nice to localise it to the code being changed -- otherwise you are more likely to conflict with other people's work unnecessarily
[01:11] <lifeless> as SteveA is not here (sprinting in London I believe with thumper and ddaa) perhaps you can raise this on thursday
[01:12] <jamesh> okay
[01:12] <BjornT> i think it's nice not to include unrelated cleanups in the patches, but on the other hand, it can be quite hard to separate. ofthen you do one related cleanup in a file, and discover that the rest of the file needs a cleanup as well.
[01:12] <lifeless> I would suggest a good policy is 'do whitespace cleanups where you change code, or otherwise in a dedicated whitespace branch that you can merge as trivial immediately.'
[01:12] <malcc> I've always preferred to either change functionality or cleanup code, but not both at once, in any one change, but here I've ended up moving away from that policy, partly because reviews often suggest adding existing-code-cleanup to very small fix-only branches
[01:13] <lifeless> malcc: you seem to generate nice small branches, so  I would not complain :)
[01:13] <malcc> I find that having a change with a clear "this only tidies, everything still works the same way" assertion is very useful
[01:14] <lifeless> malcc: I think we all agree with that. part of review work though is saying 'is this thing in front of us as good as it can be'
[01:14] <lifeless> BjornT: yup, agreed.
[01:14] <lifeless> so if jamesh can present the impact on reviewers at the general meeting, I'm all for a policy that makes it clearer when its a good idea to push it into a separate branch.
[01:15] <jamesh> lifeless: it is more impact on developers, actually.
[01:15] <lifeless> any other business ?
[01:15] <lifeless> jamesh: because of their own branch conflicting a lot ?
[01:15] <jamesh> just wondering if it is something we should pick up on in review
[01:15] <jamesh> lifeless: or causing other people's branches to conflict once they merge
[01:15] <malcc> lifeless: Perhaps some of this new bzr cleverness I hear about will help - a change could be presented as "here's how it was after I added the tests, here's how it was after I changed the functionality to work, and here it is post-cleanup"
[01:15] <lifeless> well for the former I think they get negative reinforcement when they do it
[01:16] <lifeless> for the latter its no different to doing it in a trivial merge-now branch
[01:17] <lifeless> so I think the question is 'is whitespacecleanup on lines you dont have to alter ok' ? And I think its a definite yes
[01:17] <lifeless> jamesh: please do float it at the main meeting though. I dont think we want to discourage cleanup, just get it to work well with peoples workflow
[01:18] <lifeless> any other other business ?
[01:19] <malcc> fabbione: The contents file generation apparently has never made it beyond something which can be run now and then by hand, it's not automatic
[01:19] <lifeless> ok, meeting closed...
[01:19] <lifeless> now
[01:19] <malcc> fabbione: Can we solve your urgent need just by running it once soon, or do we need to urgently solve the underlying problem and make it automatic?
[01:20] <lifeless> one thing I'd like you all to think about
[01:20] <fabbione> malcc: sorry but we were told that it was running once a day already.. clearly somebody told me a lie
[01:20] <fabbione> malcc: well i would like to have it done properly as it was supposed to be
[01:20] <lifeless> is how to encourage folk to do small targeted branches like malcc is doing
[01:20] <malcc> lifeless: I'm afraid I'm about to shatter your confidence in me with a 5000+ line branch for the ArchiveRework spec
[01:20] <lifeless> where each one is focused, does one thing, and is dead easy to tell that it does what it claims to do and does it right
[01:21] <lifeless> malcc: then you will not be the POSTER CHILD anymore.
[01:21] <lifeless> I'd like to get together with the review team at allhands
[01:22] <lifeless> if we can find some tools / methods / mental processes / whatever to encourage this, I think it would make a significant difference ;)
[01:40] <Ubugtu> New bug: #64536 in update-manager "[edgy]  software-properties crash during start" [Undecided,Unconfirmed]  http://launchpad.net/bugs/64536
[02:45] <kristog> hello
[03:05] <malcc> Anyone got a minute for a quick one-line review?
[03:06] <BjornT> malcc: sure
[03:07] <malcc> BjornT: https://devpad.canonical.com/~andrew/paste/fileD1eoLh.html
[03:08] <kristog> do you know who i should ask for an ubuntu.com mail alias?
[03:10] <BjornT> malcc: hmm. would it be hard to write a test for that function?
[03:11] <malcc> BjornT: No, that should be quite simple
[03:12] <BjornT> malcc: ok, then it'd be good to add a test to go along with the fix.
[03:12] <malcc> BjornT: Will do
[03:30] <Ubugtu> New bug: #64840 in soyuz "Override files published empty" [Critical,In progress]  http://launchpad.net/bugs/64840
[04:15] <Ubugtu> New bug: #64849 in rosetta "Deactivated members cannot re-apply to join the team" [Undecided,Unconfirmed]  http://launchpad.net/bugs/64849
[04:19] <malcc> BjornT: Here it is with a unit test just for that method: https://devpad.canonical.com/~andrew/paste/file07laYv.html
[04:19] <malcc> I tried for a test for the underlying bug, but that did turn out to be hard, so it'll have to wait for another day
[04:20] <malcc> BjornT: Oh hang on, I've forgotten to delete my test files
[04:22] <malcc> BjornT: https://devpad.canonical.com/~andrew/paste/filealSnHe.html
[04:26] <BjornT> malcc: ok. r=me. it could be worth splitting the test into two tests, though. it's good to keep the tests as small and narrow as possible.
[04:27] <malcc> BjornT: If you were making it two tests, would you make it its own test class?
[04:29] <BjornT> malcc: i think so. it might be a bit overkill for such a small function, but it makes the tests easier to understand since you can move some code to setUp and tearDown
[04:29] <malcc> BjornT: Ok, makes sense, I'll do it
[04:37] <seb128> hi
[04:37] <seb128> epiphany's upstream is asking why https://launchpad.net/products/epiphany-browser/main has a "Bazaar Status:   Test Failed"
[04:37] <seb128> should a bug be filed or something?
[04:40] <malcc> BjornT: Yes, that came out better: https://devpad.canonical.com/~andrew/paste/filek5zIN2.html
[04:40] <malcc> Small focused test classes ftw! :)
[04:40] <salgado> BjornT, SteveA, is there an easy way to get an rfc822-formatted date from a datetime object?
[04:41] <BjornT> malcc: yeah, looks good! r=me
[04:43] <BjornT> salgado: i remember having that problem some time ago. i'll see if i can find the relevant code...
[04:45] <BjornT> salgado: look at line 156 (msg['Date']  = formatdate(...)) in canonical/launchpad/scripts/bugnotification.py, maybe that lines does what you want to do?
[04:45] <salgado> BjornT, formatdate(rfc822.mktime_tz(email_date.utctimetuple() + (0,))
[04:45] <salgado> yay
[04:46] <salgado> yeah, that should do it. thanks BjornT!
[04:53] <Kamion> Hi. Could somebody trigger an import of the master installer translation files at http://people.ubuntu.com/~cjwatson/installer-po/ into Rosetta, please? Carlos has done it before.
[04:54] <salgado> danilos, can you do that ^^?
[04:55] <danilos> salgado: I think it requires a bit more privileges, maybe try with stub?
[04:55] <Kamion> Is carlos away today?
[04:55] <danilos> or maybe I can create a tarball with this and upload them all at once
[04:55] <danilos> Kamion: he's away for entire week (vacation)
[04:55] <salgado> Kamion, yeah, carlos is on holiday, afair
[04:55] <Kamion> ah
[04:55] <Kamion> I'm not sure if carlos normally uploads everything there or just the new .pot
[04:56] <Kamion> it's mostly the new .pot that I care about
[04:56] <danilos> ok, pot should be simple enough, and admin for debian-installer project should be able to upload that one as well
[04:56] <Kamion> oh, interesting
[04:57] <Kamion> that would be me
[04:58] <Kamion> https://launchpad.net/distros/ubuntu/edgy/+source/debian-installer/+pots/debian-installer/+upload gives Forbidden
[04:58] <Kamion> this is done by means of a magic sourcepackage in Ubuntu, not as a product
[04:58] <danilos> Kamion: ok, then maybe it should still be me, I can do that ;)
[04:59] <Kamion> cool, thanks
[04:59] <danilos> yeah, it would also get automagically imported once the package is rolled out, afaik
[04:59] <Kamion> debian-installer doesn't get uploaded that often
[05:00] <Kamion> and the translations aren't actually part of that package
[05:00] <danilos> ah, ok
[05:00] <Kamion> at least, not the ones from .../installer-po/
[05:00] <Kamion> it's a system to avoid having to point translations to ~50/60 separate source packages when they ask how to "translate the installer"
[05:01] <Kamion> s/translations/translators/
[05:01] <danilos> ok, so do you want me to upload translations as well?
[05:02] <danilos> or are they simply export from rosetta without any other updates?
[05:02] <Kamion> they're automatically generated (just msgmerge) from the component source packages, so probably best not
[05:02] <Kamion> I don't know what the semantics of an upload of msgmerged translations is
[05:03] <danilos> right, no need to upload them then
[05:03] <danilos> ok, I think it will be imported in a couple of minutes (the queue doesn't seem busy)
[05:04] <Kamion> great, thanks
[05:04] <danilos> np
[05:04] <danilos> ping me if I broke anything ;)
[05:06] <malcc> Is pqm healthy? It hasn't noticed my request from 20 minutes ago.
[05:08] <malcc> And, is somebody else running publish-distro on mawson right now?
[05:24] <malcc> Hmm, is it the pqm web page? I just got a failure response from my pqm merge, but the web page claimed a zero queue the entire time
[05:25] <salgado> malcc, pqm.launchpad.net?
[05:25] <malcc> salgado: Yes
[05:37] <salgado> anybody willing to do review an almost-trivial diff which fixes bug 64029?
[05:37] <Ubugtu> Malone bug 64029 in launchpad "XML/RSS of cdimage mirror status" [High,Confirmed]  http://launchpad.net/bugs/64029
[05:40] <salgado> BjornT, jamesh, ^^?
[05:46] <Ubugtu> New bug: #64866 in ubiquity "xubuntu Installation on a P2 400 512mbram & 2,1 GB hard drive" [Undecided,Unconfirmed]  http://launchpad.net/bugs/64866
[05:51] <BjornT> salgado: sorry, i don't have time for that today.
[05:52] <salgado> BjornT, okay, no worries
[05:52] <salgado> SteveA, maybe you can review that for me? it's around 200 lines, mostly template and tests
[05:58] <SteveA> salgado: not for a while... i bzr meetings
[05:58] <SteveA> in bzr meetings
[07:11] <salgado> spiv, since you're doing some reviews now, maybe you'd like to review another one? (almost trivial, around 200 lines, mostly template and tests) ;)
[07:12] <spiv> salgado: Sorry, it's my bedtime!
[07:46] <seb128> epiphany's upstream is asking why https://launchpad.net/products/epiphany-browser/main has a "Bazaar Status:   Test Failed"
[07:46] <seb128> should a bug be filed or something?
[08:14] <salgado> seb128, ddaa would be the right person to ask, but since he's not here I think it's better to file a bug on launchpad-bazaar 
[08:39] <seb128> salgado: ok, thank you. Do you know if he's on holidays or something?
[08:40] <flacoste> seb128: i think there is a launchpad-bazaar sprint going on in London this week
[08:40] <salgado> seb128, he's on a sprint in london, I think
[08:40] <seb128> ok
[09:05] <Ubugtu> New bug: #64893 in launchpad-bazaar ""Test Failed" for the epiphany-browser main serie" [Undecided,Unconfirmed]  http://launchpad.net/bugs/64893
[09:34] <sabdfl> spiv: nice blog!
[10:02] <flacoste> salgado: i'm factoring the ticket related attributes/methods in IPerson to a separate interface, do you have a suggestion for the name of such interface? i'm thinking of  ITicketInvolved
[10:05] <salgado> flacoste, I thought of ITicketRelated, but I think I prefer your suggestion
[10:06] <flacoste> salgado: or maybe ITicketPerson or ITicketActor
[10:07] <flacoste> salgado: yeah, i though also of ITicketRelated first but found that it was too generic because a bug could be also ticket related
[10:08] <salgado> ITicketActor sounds good too. I don't quite like ITicketPerson, though
[10:08] <flacoste> ok, i'll go with ITicketActor then
[11:19] <AlinuxOS> Hello people, It would be great if you put temporary ubiquity as second package to translate after debian-installer.