[12:03] <jordi> I assume it didn't happen yet then?
[12:03] <jordi> hmm. no. grumble.
[01:28] <mdz> kiko-zzz: landing page looks excellent
[03:13] <jblack> mdz: wasn't sure you cared or not, but I found a laptop that failed the livecd tets. 
[03:13] <mdz> jblack: please file a bug report with casper.log, xorg.conf and Xorg.0.log
[03:13] <mdz> er, casper/post.log
[03:14] <mdz> Package: xserver-xorg
[08:08] <spiv> stub: ping?
[08:24] <Burgundavia> where is the canonical wiki again?
[08:24] <bob2> it's all private
[08:24] <Burgundavia> not anymore
[08:25] <Burgundavia> sorry, the lp wiki
[08:25] <bob2> yes it is
[08:25] <bob2> right
[08:25] <bob2> it was merged into th ubuntu wiki
[08:25] <Burgundavia> nope, that was the UDU wiki
[08:26] <robitaille> this one? https://wiki.launchpad.canonical.com/
[08:26] <Burgundavia> robitaille, cheers
[08:37] <Burgundavia> feedback? https://wiki.launchpad.canonical.com/RequestingPackages
[08:43] <stub> spiv: pong
[08:44] <spiv> stub: I was wondering if there was any particular reason that sqlbase.quote (and by extension, sqlbase.sqlvalues) doesn't encode unicode objects to utf-8?
[08:44] <spiv> Because it would be convenient for me if it did, and it's a trivial change.
[08:44] <stub> spiv: Because the entire query needs to be encoded into UTF8, not just the arguments
[08:45] <stub> The query is a Unicode string and you can't interpolate UTF8 sequences into the middle of it
[08:45] <spiv> Hmm.
[08:46] <spiv> Ok.  Thanks.
[08:47] <jamesh> canonical.database.sqlbase.quote() isn't really necessary -- sqlobject.sqlrepr() should work fine in all cases
[08:49] <stub> canonical.database.sqlbase.sqlvalues you mean?
[08:49] <jamesh> no
[08:50] <jamesh> the one for quoting a single ar
[08:50] <jamesh> g
[08:51] <stub> Just seems like a more complicated to use quote, since you seem to need to pass in the database
[08:54] <stub> (And we should insulate ourselves from sqlobject where possible to make it easier to cope with upstream changes, such as if they break Unicode support more than they already have)
[08:56] <jamesh> hmm
[08:57] <spiv> Hmm, I can't use sqlvalues for the authserver, which is a shame.  Well, I can manually encode arguments that need it, but that sort of manipulation is what sqlvalues is supposed to avoid.  I guess I could do transaction.execute(("SQL ..." % sqlvalues(...)).encode('utf-8')), but that's ugly and awkward too.
[08:57] <jamesh> it might make sense to use SQLBase._connection.sqlrepr() (which takes one arg) instead of the function version (which takes a second arg)
[08:57] <stub> spiv: But that is the correct way of doing it
[08:58] <spiv> Yep.  Doesn't stop it from being ugly ;)
[08:58] <stub> You could also fix twisted ;)
[08:58] <stub> I don't think modelling Unicode support after PHP is a particularly brilliant idea ;)
[08:59] <spiv> The current way I have it (not using sqlvalues, and doing transaction.execute("SQL ..." % (quote(x), utf8quote(y))) ) works ok, but it would be nice to use sqlvalues to be consisten.
[08:59] <spiv> consistent, rather.
[09:00] <spiv> stub: Or you could fix psycopg to deal with unicode query strings ;)
[09:01] <stub> spiv: You are overestimating my C skills I'm afraid
[09:01] <stub> (psycopg2 does I believe, but no idea if there is a reason why it hasn't been fixed in psycopg1)
[09:03] <stub> Oh... they did fix it, for some values of 'fixed'. It now reports a type error instead of dumping core.
[09:04] <stub> It worth wrapping our cursor objects with something Unicode aware? Or just put up with ugly code for a couple more months.
[09:05] <stub> Garh.... my mock objects are mocking me!
[09:05] <spiv> Just put up with ugly code, I think.
[09:06] <spiv> On the scale of "annoying" to "ugly", it's merely at the "annoying" end :)
[09:08] <jamesh> if we edited site.py to set the default encoding to UTF-8 rather than ASCII, I'm sure some of this would start working magically
[09:08] <jamesh> then C functions expecting a string would get the UTF-8 version of a unicode string
[09:09] <jamesh> rather than the ASCII version (which might not be possible)
[09:10] <spiv> Editing site.py and changing the default encoding are both magical things I'm highly reluctant to do.
[10:49] <Kinnison> Morning
[10:55] <SteveA> morning
[10:59] <Kinnison> SteveA: FYI, I'm rotating back toward brazillian time a little for the next few weeks to help with my interactions with celso
[10:59] <SteveA> okay.  any idea when celso gets back to brazil?
[10:59] <Kinnison> He should have landed yesterday
[11:00] <SteveA> so, we might see him today
[11:00] <Kinnison> and I told him to make sure he rested and relaxed so that this morning he could be up and going on time
[11:00] <Kinnison> so yes, we should see him, I'd guess ca. 12:00 UTC
[11:00] <SteveA> or earlier, if his clock hasn't adjusted :-/
[11:19] <Kinnison> Any bzr peeps around?
[11:19] <bob2> lifeless added support the other day
[11:21] <Kinnison> Will bzr refuse to do much on a tree with "broken" symlinks?
[11:21] <bob2> I don't think it cares
[11:21] <Kinnison> Or rather will whatever bzr calls a 'tree lint' mind broken links?
[11:21] <Kinnison> cool, not caring is exactly the behaviour I want
[11:40] <mdke> jordi, hi?
[12:00] <mdke> elmo, i changed my LP email address on sat, but the ubuntu.com redirect hasn't changed yet, will it do so automagically or is there something I need to do?
[12:08] <bob2> everyone gets @ubuntu.com forwarding now?
[12:14] <jordi> mdke: yup
[12:17] <mdke> jordi, where are those pots to be found? I couldn't see em in the list of translations for breezy
[12:26] <mdke> bob2, i think everyone approved in the Ubuntu Members LP group
[12:36] <mdke> jordi, i'm off for a bit, email is best for now :)
[01:22] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  improved warning output for warnings that occur in doctests.  the now show the offending line of code. (patch-2389: steve.alexander@canonical.com)
[01:46] <Kinnison> SteveA: http://www.theregister.co.uk/2005/09/12/nad_grabbing_granny/
[01:46] <Kinnison> SteveA: anyone you know?
[01:47] <SteveA> no, but i'll certainly ask around
[01:49] <Kinnison> SteveA: let me know what other areas you'd like covered on that page
[01:57] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  made spec and ticket pages appear on the appropriate facet. (patch-2390: steve.alexander@canonical.com)
[02:29] <sumit> hi all
[02:58] <salgado> SteveA, around?
[03:22] <bradb> morning
[03:26] <cprov> lifeless: ping 
[03:36] <kiko> hey ho
[03:36] <kiko> hey gabriel 
[03:55] <gabriel> oi
[03:55] <gabriel> kiko, me chamou
[03:55] <gabriel> ?
[03:57] <SteveA> salgado: hello
[03:57] <lifeless> cprov: pong
[03:58] <salgado> hi SteveA. I was going to ask you about the virtual host for staging's shipit. but I found it
[03:58] <SteveA> ok
[03:58] <SteveA> is shipit working properly on staging?
[03:59] <salgado> yes, it is
[04:05] <segfault> is launchpad.net ok?
[04:06] <SteveA> hmm, seems to be hung
[04:06] <SteveA> lifeless: ping
[04:08] <SteveA> segfault: thanks. i'm talking to the sysadmins to find out what is wrong.
[04:08] <segfault> np
[04:11] <_elmo> sorry, steve I'm not ignoring you, I just can't msg
[04:16] <SteveA> ah, freenode messaging policy
[04:16] <SteveA> there's always jabber
[04:16] <_elmo> I've (temporarily) lost access to my home machine which has my ~/.gaim :(
[04:17] <SteveA> darn
[04:19] <Kinnison> SteveA: you can tell freenode to accept msgs from unregistered users
[04:31] <lifeless> SteveA: pong
[04:48] <cprov> lifeless: I've "ping"ed you too, how is the review of my builddUI branch, any ETA ?
[04:48] <sabdfl> bradb: ping
[04:50] <lifeless> cprov: I ponged you before :).
[04:50] <lifeless> cprov: I'll look now
[04:51] <cprov> lifeless: sorry for the noise  (hard day for me), thank you ...
[04:52] <Kamion> sabdfl: yes?
[04:52] <sabdfl> kamion, bradb: i'm touching the debbugs import stuff again, and conscious that we don't have good tests of the code i wrote. would it be possible to create a small (4 or so bugs, with 20 or so messages in total) debbugs archive for testing purposes?
[04:53] <Kamion> any particular focus?
[04:53] <sabdfl> should use names related to our current sampledata
[04:53] <sabdfl> so...
[04:53] <sabdfl> firefox
[04:53] <sabdfl> gnomebaker
[04:53] <sabdfl> mozilla
[04:53] <sabdfl> pmount
[04:54] <sabdfl> so, say 3 bugs each on pmount, firefox, evolution
[04:54] <sabdfl> that's probably best
[04:54] <sabdfl> should include a few URL's in the emails
[04:54] <sabdfl> especially a few CVE url's
[04:54] <sabdfl> also a bugzilla, and a roundup reference if possible
[04:55] <sabdfl> perhaps a dup or two
[04:55] <sabdfl> so, a small subset of interesting debbugs functionality that you'd like to see mapped into malone
[04:55] <sabdfl> if you could let me have it, i'll do the mapping and then we can compare notes and see what the code missed :-)
[04:55] <sabdfl> thanks Kamion
[04:55] <Kamion> I can't get you 'firefox' because it's still called mozilla-firefox in Debian
[04:56] <sabdfl> this will then become a set of formal tests for the code
[04:56] <Kamion> maybe just mozilla-browser?
[04:56] <sabdfl> oh, we have that one too
[04:56] <Kamion> ok
[04:56] <sabdfl> mozilla-firefox is fine
[04:56] <sabdfl> we also have "uberfrob" in the sampledata, fwiw, but i don't think it's published anywhere :-)
[04:56] <Kamion> I'll tar up some bug data and send it off; where should I send it to?
[04:57] <sabdfl> me's fine. would it be a directory structure like the one we rsync from bugs.debian.org?
[04:57] <sabdfl> rsync -avzH bugs-mirror.debian.org::bts-spool-db db-h
[04:57] <sabdfl> rsync -avzH bugs-mirror.debian.org::bts-spool-archive archive
[04:57] <sabdfl> rsync -avzH bugs-mirror.debian.org::bts-spool-index index
[04:57] <sabdfl> that's what the script expects to have handy
[04:58] <Kamion> I can certainly massage it into that structure
[04:58] <sabdfl> Kamion: over time, it would be handy to add tests to it, so send me everything needed to add future bugs and messages so we can test new features in the sync
[04:59] <Kamion> I wasn't planning on doing any fancy scripting, just sucking bug data directly out of the spool
[04:59] <sabdfl> Kamion: that's probably fine
[05:00] <Kamion> there's not much more fancy scripting that can be done anyway :-)
[05:00] <sabdfl> is it easy to add messages to existing bugs in the test set later?
[05:00] <sabdfl> i just want to be able to say "oh, we should be spotting this or that in the sync", then add a message as a test case
[05:00] <Kamion> if you know the format, yes
[05:00] <Kamion> Debbugs/Log.pm in debbugs has documentation of it
[05:01] <Kamion> it'll just be appending to one of the .log files
[05:01] <sabdfl> ok, then i'm likely to call on you for that in future :-). anyhow, bradb will own this code, and the handover is part of the reason i'd like a good set of tests
[05:01] <sabdfl> ok
[05:02] <Kamion> sabdfl: which bits of the index module do you actually use?
[05:02] <sabdfl> dunno
[05:02] <sabdfl> i just use the python interface, remember the one?
[05:02] <sabdfl> it in turn uses the perl bits, i think
[05:03] <Kamion> just index.db then
[05:04] <sabdfl> Kamion: i do actually parse the emails too
[05:06] <Kamion> the python module expects index.db to be at the root, and db-h to be a subdirectory below that
[05:07] <Kamion> debzilla sets up symlinks into the rsynced trees for that
[05:07] <Kamion> I figure you do too, if you already have an rsync setup going
[05:10] <sabdfl> yes
[05:20] <salgado> SteveA, I have a 20line patch to make sure team owners can reassign their teams. can you review it for me?
[05:20] <SteveA> salgado: yes
[05:22] <salgado> SteveA, https://chinstrap.ubuntu.com/~dsilvers/paste/filelaW75v.html
[05:23] <Kamion> sabdfl: a lot of bugs seem to have CAN-* references but not actual cve.mitre.org URLs, FWIW
[05:24] <mdke> [15:09:22]  < Treenaks> hm, mdke's announcement broke the launchpad?
[05:24] <mdke> what did I do?
[05:25] <SteveA> for some reason, the launchpad servers were having problems.  we're looking into what happened and why.
[05:25] <SteveA> they're running again now
[05:25] <mdke> ok
[05:25] <kiko> salgado, can you create a shipit product in Launchpad?
[05:25] <mdke> as long as I didn't break anything
[05:25] <kiko> https://launchpad.net/products/shipit <-salgado
[05:26] <SteveA> i don't think so.  were you doing anything out of the ordinary?
[05:26] <salgado> kiko, you already created?
[05:26] <mdke> SteveA, nope, I posted an email asking for translation of products/ubuntu-doc
[05:26] <kiko> nope
[05:26] <kiko> salgado, that's a 404?
[05:27] <salgado> kiko, yes, that is
[05:27] <SteveA> mdke: no harm in that then!  that's what launchpad is for.
[05:27] <kiko> salgado, so do it :)
[05:27] <mdke> SteveA, on a lot of mailing lists :) maybe there was a fast response?
[05:27] <bradb> sabdfl: pong (/me reads scrollback)
[05:27] <SteveA> i don't think this was a problem caused by having lots of users suddenly appearing
[05:27] <mdke> k
[05:27] <mdke> phew
[05:28] <sabdfl> bradb: getting the debbugs sync stuff ready to hand over, i need a mini debbugs archive for a test suite
[05:28] <kiko> mpt, shipit.staging.canonical.com -- can you make a list of UI issues you find so we can fix these before wednesday?
[05:28] <bradb> sabdfl: right, cool
[05:31] <jbailey> kiko: When looking at the shippit.staging, I mistook 'PC' for 'CD' as was about to ask you what arch they were for.
[05:33] <jbailey> kiko: Also, there are non-mac PowerPC systems out there.
[05:33] <kiko> jbailey, would "5 PC CDs" be more readable?
[05:33] <kiko> salgado: jbailey, I win!  :-)
[05:33] <jbailey> kiko: I guess so.  Hard to repeat a first glance like I had before.
[05:33] <jbailey> Or rather, can't be worse. =)
[05:34] <jbailey> Or maybe have the traditional arch name in () beside it or something, I don't know.
[05:35] <kiko> jbailey, yeah, I told salgado the same. I'll list it as an issue to be solved.
[05:35] <jbailey> Tx. =)
[05:36] <jbailey> I'm curious ongoing about the ratio of amd64 to i386 ones, too.  Looking in the newspapers, it looks like amd64 laptops are coming down lower than pentium4 ones.
[05:36] <kiko> the nice thing there is that that can be changed entirely through the UI.
[05:36] <jbailey> Might be just a crazy back-to-school rush, though.
[05:37] <jbailey> The shipping information lists organisation but not my name.
[05:37] <kiko> we know your name, but it's a good point.
[05:37] <jbailey> I'm guessing this is because it's pulled from LP, but it would be nice if it displayed it.
[05:37] <jbailey> Does any country in the world other than the US have "States"?  Province or Region might make a better default name.
[05:38] <kiko> in portuguese it's state too
[05:38] <kiko> the original text was State/Province/Region, I thought it was a bit too long
[05:39] <kiko> we could revert
[05:39] <jbailey> Woohoo, failed.
[05:39] <kiko> heh
[05:39] <jbailey> It pukes on my city name, Montrall
[05:39] <jbailey> Err.. Montral
[05:39] <jbailey> It says clearly what do to, but why?
[05:40] <kiko> because the shipping companies don't allow us to provide anything but ascii text.
[05:40] <salgado> jbailey, doesn't it says anything about the shipment companies restrictions?
[05:40] <jbailey> salgado: Sure., but 'not understood' doesn't make sense to me.
[05:41] <jbailey> It's easy enough, just a strange request.
[05:41] <kiko> can fix
[05:41] <jbailey> Will people know what an ASCII is?
[05:41] <kiko> we give examples
[05:41] <kiko> it's hard to define it in any other way, isn't it?
[05:41] <salgado> I guess most people won't. that's why we give the example
[05:42] <kiko> salgado, I'm making a list of issues so we can fix all at once
[05:42] <jbailey> Instead maybe say "Accents and other characters not found in the common English alphabet"
[05:42] <kiko> okidok
[05:43] <salgado> kiko, thanks. :)
[05:44] <jbailey> Why do I hear Kinnison's voice in my head saying "cards not found in a standard 52 card deck..."
[05:44] <jbailey> =)
[05:44] <kiko> heh
[05:44] <jbailey> Oh, hey, I clearly don't read web pages.
[05:45] <jbailey> There's a lovely explanation up top as to what PC, 64-bit PC and Mac mean.
[05:45] <kiko> nobody reads :)
[05:46] <bradb> The first rule of how people read on the web: they /don't/.
[05:47] <jbailey> bradb: It's because it's usually repetitive and unintersting.
[05:47] <bradb> Yep, and due to the feeling of wasting time, when considering how much else out there is waiting to be read as well.
[05:48] <jbailey> Hmm,I wonder if it's that or that I just don't care aout most of the information I read.
[05:48] <kiko> possibly both
[05:48] <bradb> mpt: Have you already fixed the "floating tabs" bug in one of your branches?
[05:49] <Nafallo> hmm, stra is Ostra? ;-)
[05:49] <mpt> bradb: No, I had a look at it while cleaning up the spec tracker, but couldn't figure out the problem
[05:49] <jbailey> FWIW, what brought that to mind was that I got to the "Your current shipit request" and was scrolling down to see if there was anything else I needed to do.
[05:49] <bradb> mpt: ah, hm
[05:49] <mpt> bradb: I'll need to diff the patch levels mentioned in the bug report to see exactly what changed
[05:49] <jbailey> It might be nice if it pointed that out somehow.
[05:49] <bradb> mpt: ok
[05:50] <jbailey> It feels like this is a one last "are yousure>?" page.
[05:50] <jbailey> (like I expect to find an "Ok" button next to the "cancel request" one)
[05:54] <jbailey> kiko: Anything else you need done?
[05:54] <kiko> jbailey, no, that's great feedback. so this last comment is perhaps because we don't provide any "special" feedback that your order was successfully processed?
[05:55] <jbailey> That could be it.
[05:55] <jbailey> I'm thinking of the Avis screen where I have to confirm my registration details that I've entered.
[05:55] <kiko> will look into it; thanks.
[05:55] <kiko> yeah
[05:55] <jbailey> It looks lovely aside from that.  Cheers to no more yellow. =)
[05:56] <Belutz> hello
[05:59] <kiko> thanks jb
[05:59] <kiko> ailey
[06:04] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: Make sure team owners can reassign their own teams. r=SteveA (patch-2391: guilherme.salgado@canonical.com)
[06:17] <BjornT> kiko, salgado: re shipit, if you submit an empty form, the error messages seems a bit misplaced. also, it says that post code is required, even though it's not.
[06:17] <Kamion> sabdfl: ok, you have mail
[06:20] <kiko> BjornT, yeah, ideally the error messages showed up next to the fields, but it's a bit of work
[06:20] <kiko> BjornT, postcode is a bit delicate to get right
[06:21] <kiko-fud> easier to say it's required
[06:21] <BjornT> kiko: yeah, but why one message above the explanation of PC/64bit/Mac, and the rest below?
[06:21] <kiko-fud> the one above just says that "something is wrong", right?
[06:22] <salgado> BjornT, if we put all messages together, close to the form information, there's the risk of the user not seeing that an error happened
[06:24] <BjornT> kiko-fud, salgado: true, a general one on top is ok. but it seems more logical to put the other error message below "Shipping Adress"
[06:25] <sabdfl> kiko, mpt: can we grey out the "Custom" textboxes when "Standard" is selected?
[06:26] <salgado> BjornT, yes, that's something I discussed with kiko. the problem is that right now I have a single bucket with all error messages. I need to differentiate the errors in the address form and the one in the request
[06:26] <sabdfl> who made the person-index.pt changes? i got a ton of conflicts and want to know what got changed so i don't drop anything on the floor
[06:28] <salgado> sabdfl, I think I did it on saturday. the changes are to not display inactive memberships. but I also fixed the indentation through the whole file
[06:28] <sabdfl> salgado: ok, it's the indentation that naile dme i think
[06:28] <sabdfl> i'll find the inactive memberships change and make sure that is preserved
[06:28] <sabdfl> then i'll execute baz smash ...
[06:38] <SteveA> BjornT: can you do a driveby review of a few 100 line of diff for me?
[06:39] <BjornT> SteveA: sure
[06:39] <SteveA> thanks
[06:39] <SteveA> i'll mail it
[06:45] <cprov> SteveA: ping
[06:45] <SteveA> hi cprov 
[07:09] <cprov> SteveA:  I wonder if you received my last email, I had email auth problems 
[07:11] <SteveA> cprov: how can i tell?  i received the last email from you that i received.
[07:12] <cprov> SteveA: ehe, it was the answer of you review for test warning.  it's easy, just say if you have any pending answer from me . 
[07:13] <SteveA> nope, don't have your answer
[07:13] <sabdfl> what happened to lp.encoding?
[07:14] <SteveA> i see a lib/canonical/encoding.py
[07:14] <cprov> SteveA: Message-ID: <432594D0.4000500@gwyddion.com>, ok I'll resend
[07:16] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Fix https://launchpad.net/malone/bugs/2154 (patch-2392: guilherme.salgado@canonical.com)
[07:17] <sabdfl> SteveA: thanks. perfect :-)
[07:17] <SteveA> baz should have some option to answer that question
[07:17] <SteveA> i don't know if it does, but it should do
[07:17] <SteveA> after all, it knows the answer
[07:18] <kiko-afk> SteveA, btw, bjorn't threading patch is going stale pretty quickly..
[07:19] <BjornT> ah, that's right, we should have reviewed that one on friday.... SteveA, do you think you'll have time to review it today or tomorrow?
[07:20] <SteveA> sure
[07:20] <kiko-afk> lifeless, btw, do you know why your pending-reviews patch doesn't appear in pending-reviews?
[07:20] <kiko-afk> ah, it's not against RF
[07:21] <BjornT> SteveA: cool, i'll resolve the conflicts tonight
[07:22] <SteveA> ok
[07:25] <sabdfl> Kamion: SummaryMissing: test/db-h/94/308994.summary
[07:26] <Belutz> it's not easy to translate..... *sigh*
[07:30] <lifeless> kiko-afk: :)
[07:30] <lifeless> SteveA: can you please check my question for you in the celso-review I just did ?
[07:34] <Kamion> sabdfl: oh, sorry, I'm a complete moron and only sent the logs. will fix
[07:34] <sabdfl> Kamion: np
[07:37] <Kamion> done
[07:38] <SteveA> BjornT: how's the driveby review going?
[07:40] <BjornT> SteveA: doing it now
[07:40] <SteveA> thanks
[07:43] <lifeless> SteveA: thanks
[07:43] <lifeless> SteveA: missed the bit on the docstring though :[
[07:51] <lamont> grumble.  baz 1.4.2-1ubuntu1 has unaligned load/stores during its tests.
[07:51] <lamont> --> ftbfs on hppa
[07:57] <sabdfl> how does a cronscript know which user it is running as?
[07:58] <sabdfl> bradb: why does BugTaskSearchParams require a user? Surely it should be None by default? and just not see private bugtasks?
[07:58] <bradb> sabdfl: You'd have to ask kiko. :P But I'm not so sure that making it None by default is a good idea.
[07:58] <SteveA> sabdfl: as in, which database user?
[07:59] <sabdfl> SteveA: yes
[07:59] <sabdfl> bradb: why not?
[07:59] <bradb> sabdfl: Because it might make it easy to forget to pass the user, and end up with strange, hard to explain bugs.
[07:59] <sabdfl> bradb: it would also not bite people who don't know what user their code is running as
[07:59] <kiko-afk> sabdfl, isn't that a feature, though?
[07:59] <bradb> it could be made into a "marker" object by default, and raise an error saying that you need to pass the user, etc.
[07:59] <SteveA> sabdfl: launchpad.conf has a section for each script
[07:59] <sabdfl> kiko-afk: no.
[07:59] <sabdfl> no user is no user
[08:00] <SteveA> there is a 'dbuser' setting for each script
[08:00] <sabdfl> SteveA: that's fine, but how does the script itself know?
[08:00] <kiko> supply user=None if you don't have a valid user, sabdfl.
[08:00] <kiko> hmmm
[08:00] <sabdfl> so, for example, it can pass its user to fascist classes like BugTaskSearchParams
[08:00] <sabdfl> kiko: please make that a keyword with a default
[08:01] <kiko> well, tbh, I think that the reason it doesn't have a default is more because of the backend implementation (where you need to know what user it is) than because of BTSP itself.
[08:01] <kiko> we could default user=None and then assume you're running without being logged in.
[08:02] <bradb> I think that would make it easy to make unfortunate mistakes.
[08:02] <kiko> can't quite remember why I made it mandatory -- BjornT reviewed it and thought it was correct when he did.
[08:02] <kiko> well, apart from making it  obvious to the callsite that the user specified has a lot of importance when using BTSP.
[08:03] <SteveA> sabdfl: argument to initZopeless
[08:03] <SteveA> ./cronscripts/buildd-queue-builder.py:    tm = initZopeless(dbuser='fiera')
[08:03] <SteveA> for example
[08:06] <bradb> sabdfl: I don't understand: if you're writing code against the task search API, but with no idea what user you're supposed to be, how would you expect a sane result to be returned from the search? (e.g. one that searches all the bugs that it should for that user, no more and no less.)
[08:06] <kiko> sabdfl is gone, though.
[08:07] <bradb> oog
[08:29] <kiko> failure
[09:28] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: r=bjornt plus some [trivial]  extra refactorings.  more consistent facet menus across the whole of launchpad.  improvements in the API. (patch-2393: steve.alexander@canonical.com)
[09:42] <mpt> kiko: If "This source package is not published in The Breezy Badger Release", *should* it be translatable? If so, why?
[09:43] <kiko> the issue is that publishing and translation are completely separate workflows, right now
[09:43] <kiko> publishing depends on soyuz working
[09:43] <kiko> rosetta works today
[09:43] <kiko> uploads and downloads are done out of band
[09:44] <kiko> so... that's the reason why it should be translatable.
[09:48] <mpt> Really?
[09:49] <mpt> That sounds like the reason it is, not the reason it should be :-)
[09:49] <mpt> anyway
[09:51] <mpt> kiko: So will that change eventually? If so, I'll just tweak the explanation on the source package page to say "But you can still translate, report bugs", etc
[09:55] <kiko> it will change eventually -- when ubuntu runs off soyuz.
[10:02] <mpt> ok, reported bug 2234
[11:02] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial, text and links r=kiko]  (3rd time lucky?) LaunchpadIntegration pages cleanup (patch-2394: mpt@canonical.com)
[11:02] <kiko> finally
[11:27] <mpt> indeed
[11:39] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Fix for bug 1986:  No way to get back to the list of Foo bugs from the task page; also adds a few traversal tests and other minor fixups (patch-2395: christian.reis@canonical.com)
[11:57] <salgado> is it possible to access the /errors page on staging?
[11:58] <kiko> I thought so