[06:25]  * mwhudson is in that fun "run tests. edit security.cfg, rinse later repeat" cycle
[06:44] <lifeless> <
[07:27] <lifeless> jml: ping
[08:15] <wgrant> noodles775: Won't your anti-buildd-monopolisation branch pretty effectively break start time and queue length estimations?
[08:15] <wgrant> (it's also completely over-restrictive except for a few PPAs)
[08:26] <adeuring> good morning
[08:26] <noodles775> wgrant: yes - as per the comment on that bug, we'll need to update the estimations etc.
[08:27] <noodles775> wgrant: and I've created bug 450124 for improvements...
[08:27] <mup> Bug #450124: Improvements to IBuilder.findBuildCandidate() <Soyuz:New> <https://launchpad.net/bugs/450124>
[08:27] <ubot3`> Malone bug 450124 in soyuz "Improvements to IBuilder.findBuildCandidate()" [Undecided,New] https://launchpad.net/bugs/450124
[08:27] <mup> Bug #450124: Improvements to IBuilder.findBuildCandidate() <Soyuz:New> <https://launchpad.net/bugs/450124>
[08:27] <mup> Bug #450124: Improvements to IBuilder.findBuildCandidate() <Soyuz:New> <https://launchpad.net/bugs/450124>
[08:28] <noodles775> wgrant: so pls add any issues/improvements you can think of there (I've put one easy solution, but you might have better ideas :) ).
[08:33] <stub> eek - double bots. Danger of feedback loops.
[08:34]  * thumper shoots mup
[08:47] <al-maisan> Good morning
[08:53] <bigjools> thumper: yo
[08:58] <thumper> bigjools: hey
[08:58] <bigjools> thumper: call?
[08:58] <thumper> bigjools: yo
[08:58]  * thumper fires up skype
[08:59] <bigjools> FAIL
[08:59] <thumper> bigjools: first time I hadn't plugged in the head phones
[09:00] <bigjools> thumper: ah yeah, skype doesn't like it when you plug them in when it's already running
[09:00] <bigjools> bites me all the time
[09:03] <mrevell> Morning!
[10:51] <bigjools> thumper: so I can't find a blueprint for build from branch, I'll register one on Soyuz as a start since most of the work is there
[14:07] <gary_poster> salgado: I'm hoping you'll join us on working on Python 2.5/2.6 on a sprint-like thing.  We're hanging out in #launchpad-sprint .  (Others are welcome too!)
[14:11] <gary_poster> BjornT: hey.  You want to call?  I just got on Skype
[14:11] <BjornT> gary_poster: yes, i'm starting skype now
[14:11] <gary_poster> BjornT: cool
[14:13] <BjornT> gary_poster: https://code.launchpad.net/~bjornt/lazr-js/buildoutification/+merge/13003
[14:22] <BjornT> gary_poster: http://code.google.com/p/js-test-driver/
[14:43] <barry> sinzui: network fail
[14:43] <sinzui> I see
[14:43] <barry> sinzui: try me again
[14:44] <sinzui> barry:I cannot connected you
[14:44] <barry> sinzui: restarting :(
[14:47] <barry> sinzui: i guess we're done, eh? ;)
[14:47] <sinzui> barry: are indeed done
[14:47] <barry> sinzui: okie dokie
[15:09] <barry> danilos: ping
[15:09] <danilos> barry: hi
[15:10] <barry> danilos: hi!  i think you're the only translations guy online atm.  do you have a few minutes to help me with something?
[15:10] <danilos> barry: sure, I can give it a shot, though I might be way behind things after two weeks of sprinting
[15:11] <danilos> (though, just imagine the wonders it did for my line :)
[15:12] <barry> danilos: :).  what i'm looking for is the code in poexports that actually creates the tarfiles.  tests such as poexport-request-productseries.txt are failing in the python2.5 branch because the tar format changed.  i want to take a look at how export creates those tarfiles, but i'm having some trouble locating the exact code
[15:12] <barry> danilos: actually looking for where files are being added to the tarfile
[15:12] <barry> danilos: do we use python's tarfile module or do we shell out to tar to do that?
[15:13] <danilos> barry: somewhere in lp.translations.utilities, let me check
[15:13] <danilos> barry: we use tarfile module
[15:13] <barry> danilos: cool, thanks
[15:13] <danilos> barry: translation_export in there
[15:14] <barry> danilos: ExportFileStorage?
[15:14] <barry> TarballFileStorageStrategy?
[15:14] <danilos> barry: LaunchpadWriteTarFile I think
[15:15] <danilos> barry: add_file method if that's what you are looking for
[15:15] <barry> danilos: got it.  thanks!  i think that's what i'm looking for
[15:16] <barry> danilos: let me poke around in there.  i may return with more questions.  in the meantime.... thanks!
[15:16] <danilos> barry: sure, you are welcome
[15:36] <leonardr> adeuring, i have some more questions about the problem publishing IHWDBApplication through the web service, which i asked you about yesterday
[15:37] <adeuring> leonardr: yes?
[15:37] <leonardr> who has permission to see that object?
[15:37] <adeuring> leonardr: a special team called hwdb-team
[15:38] <leonardr> ok
[15:38] <leonardr> is this internal to canonical?
[15:38] <leonardr> i'm trying to figure out whether it makes sense to publish information about the hwdb object in the public api documentation
[15:40] <adeuring> leonardr: it is internal to Canonical
[15:41] <leonardr> adeuring: is this bug stopping those people from doing useful work?
[15:42] <adeuring> leonardr: No; the point is that you need to know how to create the HWDB object "manually", via lp.load().
[15:43] <leonardr> adeuring: i see
[15:43] <adeuring> leonardr: so, it is more an inconvenience, since lp.hwdb does not work
[15:43] <adeuring> (i mean, in scripts that use launchpadlib)
[15:43] <leonardr> right
[15:49] <leonardr> adeuring: is the same true of objects like h_w_device and h_w_device_class? is there no way for normal users to get to them?
[15:49] <leonardr> because if so, they are cluttering up the api doc
[15:50] <adeuring> leonardr: yes. Though I hink that we should open at least parts of the HWDB at some time. But the details are quite difficult...
[15:52] <leonardr> i see
[15:52] <leonardr> i've been telling people about the hwdb publishing as though it was something they could access. that's pretty emberassing
[15:53] <adeuring> well, it was open for some time, but the was, erm, some demand to close it...
[15:55] <leonardr> ok
[18:02] <bigjools> anyone got any ideas why this might happen? http://pastebin.ubuntu.com/292507/
[18:02] <mrevell> Night all
[18:04] <jml> bigjools, no. however I will gladly rewrite said software in Twisted, if you can somehow wrangle me the time to do so :)
[18:05] <bigjools> jml: oh *please*.... with lots of icing and cherries
[18:05] <bigjools> actually I *think* Celso did some of that before he left, I need to poke through his unlanded branches
[18:06] <jml> bigjools, time-wrangling is the trick.
[18:06] <bigjools> jml: I'm sure it's only 30m work for a man of your calibre
[18:08] <jml> bigjools, nothing that involves running tests in Launchpad takes less than 30m.
[18:09] <bigjools> jml: it's ok poppy doesn't have tests :)
[18:10]  * bigjools EODs
[18:10] <bigjools> g'night
[18:10] <jml> bigjools, g'night.
[20:42] <mwhudson> good morning
[20:46] <dobey> hey mwhudson
[20:48] <lifeless> hey
[21:05] <jml> hello
[21:09] <thumper> hi jml
[21:15] <mwhudson> hi jml
[21:15] <mwhudson> jml: thanks for the review
[21:16] <jml> np
[21:17] <thumper> rockstar: skype ping
[21:17] <rockstar> thumper, pong
[21:22] <beuno> flacoste, intellectronica, http://www.456bereastreet.com/archive/200910/remove_the_outline_from_links_on_active_only/
[21:23] <lifeless> hi jml
[21:23] <lifeless> jml: whats your opinion about patches adding the extended test protocol I've been designing to testtools?
[21:25] <jml> lifeless, unformed, generally favourable.
[22:01] <mwhudson> jml: is there a bug for the branchingubuntu work?
[22:01] <jml> mwhudson, I don't think so.
[22:01] <mwhudson> or, gosh, maybe a blueprint!?!
[22:04]  * mwhudson acquires ridiculous amounts of karma by using blueprints
[22:06] <mwhudson> thumper, jml: does https://blueprints.edge.launchpad.net/launchpad-code/+spec/branching-ubuntu/ look ok?
[22:08] <jml> mwhudson, what exactly am I looking for?
[22:09] <mwhudson> jml: sanity
[22:09] <jml> mwhudson, the blueprint metadata looks fine
[22:09] <mwhudson> jml: conspicuously wrong things
[22:09] <mwhudson> jml: ok thanks
[22:09] <jml> mwhudson, I'm pretty sure I wrote the spec, so I'm happy to assume that it's relatively sane :)
[22:10] <mwhudson> jml: heh
[22:13] <rockstar> thumper, did you land the branch index windmill test?
[22:22] <jml> mwhudson, actually, do you have a moment to talk about the code import stuff?
[22:22] <thumper> rockstar: yes
[22:22] <mwhudson> jml: sure
[22:23] <jml> mwhudson, can you call my landline?
[22:23] <mwhudson> jml: sec
[22:23] <rockstar> thumper, it's failing on my machine looking for xpath='//div[contains(@class, "yui-ichoicelist-content")]'
[22:24] <thumper> ??
[22:25]  * thumper tries to run the test
[22:25] <rockstar> thumper, http://pastebin.ubuntu.com/292652/
[22:44] <thumper> flacoste: did we have a call?
[22:46] <flacoste> thumper: we should
[22:47] <simon-o> Hi, I asked this in launchpad-sprint before, but they are busy at the moment (the problem is related to the sprint). Does anyone know, when and where output elements are escaped?
[22:47] <thumper> simon-o: which elements?
[22:48] <thumper> flacoste: I just get your skype voicemail
[22:49] <simon-o> thumper: links to feeds (<link rel="alternate" type="application/atom+xml" ...>)
[22:49] <flacoste> thumper: that's because i needed to hang up with Gary first!
[22:49] <flacoste> thumper: i'm free now!
[22:49] <simon-o> this is the test case feeds/xx-links
[22:49] <thumper> simon-o: what is failing exactly?
[22:50] <wgrant> (note that this only fails with Python 2.5)
[22:50] <simon-o> thumper: http://paste.ubuntu.com/292664/
[22:51] <simon-o> wgrant: yes, it's only py2.5 but I trying to fix this, but don't know where to search
[22:51] <wgrant> The escaping is still fine.
[22:51] <wgrant> It just does it differently.
[22:52] <wgrant> I'm not sure changing it just to match the tests is a good idea.
[22:54] <simon-o> wgrant: ok I saw this while debugging. So this isn't related to py2.5
[22:54] <wgrant> simon-o: Doesn't it only happen in 2.5?
[22:58] <simon-o> wgrant: Sure, what I wanted to say was that this isn't a bug somewhere in python 2.5 it's just a changed behavior
[22:58] <simon-o> :)
[22:59] <wgrant> simon-o: Oh, right.
[23:00] <mwhudson> hooray doctests
[23:00] <mwhudson> simon-o: you could test that the unescaped output is the same?
[23:00] <mwhudson> maybe
[23:00] <simon-o> mwhudson: the test is just about correct escaping
[23:01] <mwhudson> i see
[23:01] <simon-o> and I think title='Announcements for Bad displayname"&gt;&lt;script&gt;alert("h4x0r")&lt;/script&gt;' is escaped correctly
[23:02] <wgrant> Right - it just prefers to use different quotes rather than escape them.
[23:03] <simon-o> wgrant: yes. now I would just be interested where exactly this happens (i.e. which method does the escaping)
[23:03] <wgrant> simon-o: Whatever XML library TAL uses, I would guess.
[23:06] <simon-o> wgrant, mwhudson, thumper: thanks, you helped me a lot