[12:22] <jordaoPT> anyone know if xubuntu-docs will be updated in the next monthly update of the language-packs ???
[12:25] <jordaoPT> can anyone help me???
[12:32] <sivang> jordaoPT: you need carlos or jani, they're are likely to be around during the EU day hours
[12:33] <carlos> jordaoPT: No, documentation is not part of language packs
[12:35] <jordaoPT> but documentation will be updated in dapper right?
[12:35] <carlos> jordaoPT: you will need to check it with mdke or janimo
[12:35] <carlos> jordaoPT: I don't know, sorry
[12:36] <jordaoPT> ok, thanks anyway
[12:51] <sivang> night all
[01:34] <carlos> night
[06:29] <mpt_> Gooooooooooooooooood afternoon Launchpadders!
[06:44] <ruffneck> good morning
[06:44] <ruffneck> moi
[08:51] <lifeless> ight
[08:51] <lifeless> meh
[08:53] <jamesh> mpt: I'll fill in a bit more info on that wiki page
[08:55] <jamesh> SteveA: for when you're around: https://staging.launchpad.net/products/zope/+specs?show=all
[08:55] <lifeless> nice
[09:21] <mpt> jamesh, that question I asked you yesterday about browser.addHeader has had its sequel today in a test failure
[09:22] <mpt> In the second test I have browser.addHeader('Authorization', 'Basic mark@hbd.com:...')
[09:22] <mpt> and the third test, browser.addHeader('Authorization', 'Basic no-priv@canonical.com:...')
[09:22] <mpt> but the third test fails because it thinks it's logged in as mark@hbd.com
[09:23] <mpt> and so does the fourth, where I use anon_browser
[09:27] <jamesh> mpt: you might need to specifically create extra Browser objects to handle the multiple identities
[09:28] <mpt> so browser is acting like a global variable?
[09:28] <jamesh> mpt: as for anon_browser thinking it is mark@hbd.com, are you sure that you weren't mixing up anon_browser and browser for something?
[09:28] <jamesh> yes
[09:28] <spiv> jamesh: It would still be good for the browser objects to be remade between tests.
[09:29] <mpt> spiv, these are tests in the same file, testing how the same page works for different people
[09:29] <jamesh> spiv: they are.  I assume mpt is talking about multiple uses of browser inside a single pagetest
[09:29] <spiv> Oh, right.
[09:29] <mpt> ok, so how do I create a new browser
[09:29] <spiv> In that case, do what jamesh says :)
[09:29] <mpt> owner_browser = ...
[09:30] <jamesh> mpt: from zope.testbrowser.testing import Browser
[09:30] <jamesh> mpt: somebrowser = Browser()
[09:30] <mpt> cool
[09:33] <spiv> ...
[09:33] <mpt> ...
[09:35] <mpt> woohoo
[09:43] <jamesh> mpt: success?
[09:44] <mpt> yes, jamesh, thanks
[09:44] <mpt> once I stopped just using "browser." everywhere :-)
[09:46] <mpt> carlos, ping
[09:47] <carlos> mpt: pong
[09:47] <lifeless> spiv: can you review carlos branch?
[09:47] <lifeless> spiv: should be easy - there are no tests.
[09:48] <mpt> carlos, what is the relationship between rosetta_official and primary_translatable?
[09:48] <carlos> lifeless: there are tests....
[09:48] <carlos> lifeless: pagetests, it's a navigation issue
[09:48] <lifeless> oh sorry, I missed it.
[09:48] <lifeless> :)
[09:48] <carlos> np ;-)
[09:49] <carlos> mpt: rosetta_official means that upstream maintainers are using rosetta officially
[09:51] <carlos> mpt: I don't remember now what's primary_translatable...
[09:51] <carlos> mpt: where is it?
[09:51] <mpt> it's in browser/product.py
[09:51] <spiv> lifeless: ok
[09:51] <mpt> It seems to say "if there's a translatable package, return that, otherwise if the product is translatable, return that"
[09:51] <mpt> carlos, that seems backwards to me, is it?
[09:51] <lifeless> thanks
[09:52] <mpt> If you're on the product page, should Rosetta instead be seeing if the product is translatable before looking for any packages?
[09:52] <carlos> from our interface: 
[09:52] <carlos>     primary_translatable = Attribute(
[09:52] <carlos>         "The best guess we have for what new translators will want to "
[09:52] <carlos>         "translate for a given product. First, tries the current development "
[09:52] <carlos>         "Ubuntu package. Then tries the latest series for which we have "
[09:52] <carlos>         "potemplates.")
[09:53] <mpt> ah, I didn't look in the interface
[09:53] <carlos> mpt: yeah, we should change it
[09:53] <mpt> I was going by the docstring, which was a bit unhelpful
[09:53] <carlos> now that we only import upstream translations that are actually using Rosetta
[09:53] <carlos> we should stop pointing to Ubuntu
[09:54] <carlos> that's confusing a lot of people
[09:54] <carlos> but Mark wanted it that way when we started
[09:54] <mpt> carlos, could I fix that by swapping the if and elif blocks in browser/product.py?
[09:54] <mpt> (lines 374 ~ 395)
[09:56] <carlos> mpt: that will not change anything
[09:57] <carlos> mpt: primary_translatable is either an IProductSeries or ISourcePackage
[09:57] <mpt> yep
[09:57] <carlos> so if you swap the if block, the output will be the same
[09:57] <mpt> why?
[09:57] <carlos> you need to change primary_translatable
[09:57] <carlos> mpt: because if you have 'a'
[09:57] <carlos> foo = 'a'
[09:57] <carlos> if foo == 'a':
[09:57] <carlos> ..
[09:58] <carlos> if foo == 'b':
[09:58] <carlos> ...
[09:58] <mpt> sorry, I was unclear
[09:58] <mpt> here's what I'm suggesting:
[09:58] <carlos> and swap them ... is the same
[09:58] <mpt> if IProductSeries.providedBy(translatable):
[09:58] <mpt>     ... # productSeries stuff
[09:58] <mpt> elif ISourcePackage.providedBy(translatable):
[09:59] <carlos> I got it the first time
[09:59] <mpt>     ... # sourcepackage stuff
[09:59] <carlos> but that's not going to change that translatable cannot be at the same time an IProductSeries and an ISourcePackage
[09:59] <carlos> is one or the other
[09:59] <mpt> I don't want it to be at the same time, right now
[10:00] <mpt> (though that might be nifty, for "You can also translate these packages...", eventually)
[10:00] <mpt> I just want it to look at the product first
[10:00] <mpt> Would that be a bad thing to do?
[10:00] <carlos> dude, we have an understanding problem here....
[10:01] <carlos> let me prepare an easy example...
[10:01] <mpt> so let's say the Foopy upstream uses Rosetta
[10:01] <mpt> and Foopy is also packaged in Baltix
[10:01] <mpt> Currently if you go to /products/foopy/+translations, it will suggest you translate Baltix Foopy, instead of upstream Foopy
[10:02] <mpt> with the links to upstream Foopy down the side
[10:02] <mpt> correct?
[10:03] <carlos> mpt: I see your point and understand what you want to do
[10:03] <carlos> but your change will not change that
[10:03] <carlos> will have the same behaviour
[10:03] <carlos> mpt: https://chinstrap.ubuntu.com/~dsilvers/paste/fileawZ2V1.html
[10:03] <carlos> that's more or less what you want to do and what would happen
[10:04] <carlos> mpt: do you see it now?
[10:05] <carlos> mpt: what you need to do is to change IProduct.primary_translatable to pick first the productseries
[10:06] <carlos> kiko: good morning dude
[10:06] <mpt> So the code that sets IProduct.primary_translatable is not the "def primary_translatable(self):" code in browser/product.py?
[10:07] <mpt> ah, it's in database/ ...
[10:07] <carlos> mpt: right
[10:07] <carlos> the one in browser uses the database one
[10:07] <carlos> translatable = self.context.primary_translatable
[10:08] <mpt> oh!
[10:08] <mpt> providedBy is like "are you one of these"
[10:08] <spiv> Right.
[10:08] <mpt> foo.providedBy(bar) is "does bar have a foo"
[10:09] <mpt> thanks for your patience carlos :-)
[10:09] <spiv> If a class implements an interface, then instances of that class will provide that interface.
[10:09] <mpt> getUtility(ILaunchpadCelebrities).ubuntu, eh
[10:09] <carlos> mpt: don't worry. Thanks for your good UI work ;-)
[10:10] <mpt> heh
[10:10] <mpt> well, this is just four lines to move, rather than 20ish
[10:21] <mpt> carlos, do you have time now to check my rewrite of product-translations.pt for sanity?
[10:21] <carlos> sure
[10:22] <mpt> https://chinstrap.canonical.com/~dsilvers/paste/filecQkEgn.html
[10:22] <sivang> morning
[10:23] <mpt> hi sivang 
[10:23] <mpt> hi bradb
[10:23] <mpt> group hug!
[10:24] <carlos> mpt: could you provide me a diff output?
[10:24] <sivang> can anybody please check out what went wrong with https://launchpad.net/people/ubuntu-dev/+branch/hubackup/ubuntu ?
[10:24] <mpt> sure
[10:24] <carlos> mpt: thanks
[10:24] <mpt> but, like I said, it's a rewrite ...
[10:24] <sivang> I aborted the push since I wanted to provide a cleaner branch with less commits , but now when I try to push the freshened branch I get:
[10:25] <bradb> mpt: hey
[10:25] <bradb> hey sivang 
[10:25] <sivang> bzr: ERROR: File exists: '/~ubuntu-dev/hubackup/ubuntu': mkdir failed: unable to mkdir
[10:26] <mpt> carlos, https://chinstrap.canonical.com/~dsilvers/paste/file3Vok59.html
[10:26] <lifeless> sivang: what went wrong is that you abortewde the push
[10:26] <mpt> (I warn you, it's not useful)
[10:27] <lifeless> sivang: rather than letting it finish and then doing a push --override
[10:27] <sivang> lifeless: I would have thought it would know how to handle this :-) sorry
[10:27] <elmo> *.ubuntu.com, *.launchpad.net and anything else in the Canonical Data Centre is going away for 5 mintes
[10:27] <lifeless> sivang: you'll need to sftp in and remove teh directory '.bzr/branch' 
[10:27] <carlos> mpt: yeah, but I can see the old page in just one window.... I'm too lazy :-
[10:27] <carlos> :-P
[10:27] <mpt> heh
[10:28] <sivang> lifeless: okay, thanks, I'll try
[10:29] <mpt> http://launchpad.ishavingamassage.com/
[10:31] <carlos> mpt: You have my approval for that change. Good work!
[10:31] <sivang> mpt: hehe
[10:31] <sivang> mpt: good one
[10:31] <mpt> thanks carlos
[10:31] <carlos> mpt: but, we should fix the permissions if we are going to leave it that way
[10:32] <mpt> in what way?
[10:32] <carlos> mpt: because atm anyone can request uploads for productseries, not just the owner of a product
[10:32] <carlos> we use launchpad.AnyPerson
[10:32] <mpt> hmm
[10:33] <carlos> mpt: I think is ok to allow only owners to do that
[10:33] <mpt> so the new page is correct
[10:33] <mpt> but the underlying code is too permissive?
[10:34] <carlos> yeah
[10:34] <mpt> I have *no idea* where I'd fix that
[10:34] <carlos> it's easy
[10:34] <mpt> Should I report a bug, or is it easy enough for you to give me instructions? :-)
[10:35] <carlos> mpt: do you want to learn how to do that or should I do it myself?
[10:35] <carlos> I think it's easy enough
[10:35] <mpt> I'm assuming it's a one-liner somewhere
[10:35] <carlos> mpt: lib/canonical/launchpad/zcml/productseries.zcml
[10:35] <carlos> change launchpad.AnyPerson to launchpad.Edit
[10:36] <mpt> on <browser:page name="+translations-upload"...>?
[10:36] <carlos> right
[10:36] <mpt> or just above it, at least
[10:36] <carlos> also, lib/canonical/launchpad/security.py should be updated
[10:36] <carlos> do you know how to do that?
[10:36] <mpt> ok, that blocks access to the page, now to block access to the underlying function...
[10:37] <mpt> oh!
[10:37] <carlos> mpt: I can provide you with the new function we should add there
[10:37] <mpt> This is the file we were chuckling at a couple of weeks ago in Montr?al
[10:37] <SteveA> jamesh: thanks for the specs page
[10:38] <mpt> carlos, great
[10:38] <carlos> I don't know, I wasn't in Montreal ;-)
[10:38] <SteveA> stub: how long will those specs stay on staging for?  I mean, when is the next staging db update?
[10:38] <jamesh> SteveA: many of the specs have no email address, and it seems most of the remainder don't have LP accounts
[10:38] <stub> Staging is rebuilt daily unless someone asks me to switch it off.
[10:38] <stub> So until about 01:00 UTC
[10:39] <mpt> carlos, it was "My god, look at the length of that class name" :-) (particularly EditDistributionMirrorByOwnerOrDistroOwnerOrMirrorAdminsOrAdmins(AuthorizationBase))
[10:39] <jamesh> I can rerun the import tomorrow
[10:39] <carlos> ;-)
[10:39] <SteveA> jamesh: do any that don't have email addresses have WikiName assignees?
[10:39] <SteveA> stub: any problem with turning it off for a few days?
[10:40] <jamesh> SteveA: some of them do, yes.
[10:40] <stub> We should switch to using the MD5 sums of long python identifiers.
[10:40] <stub> SteveA: Not really.
[10:40] <mpt> carlos, so how is this different from launchpad.Edit?
[10:40] <SteveA> stub: then please do so
[10:41] <SteveA> mpt: right now, our security system is made up of a mixture of permissions (launchpad.Edit) and policy components (class name describes the policy it embodies)
[10:41] <carlos> mpt: that file is what controls who has the Edit permission for a ProductSeries
[10:41] <SteveA> I want to move to just policy components, and do some changes in the underlying architecture to make them apply at a more useful granularity
[10:42] <mpt> ah, ok
[10:42] <carlos> mpt: launchpad.Edit is not always for the same kind of people
[10:42] <mpt> I understand
[10:42] <mpt> so this will be something like EditByProductOwnersOrAdmins?
[10:44] <carlos> mpt: https://chinstrap.ubuntu.com/~dsilvers/paste/fileKypL56.html
[10:44] <carlos> mpt: yes, kind of
[10:44] <SteveA> the name of these security components is just a description of what it does
[10:44] <SteveA> they don't *need* to get sillilly long
[10:44] <mpt> thanks, carlos!
[10:44] <carlos> mpt: but giving also access to Rosetta Experts
[10:44] <mpt> SteveA, so it could just be "EditDistributionMirror"
[10:45] <mpt> ah, of course
[10:45] <carlos> mpt: hmmm, please, rename the function I just gave you
[10:45] <carlos> and put 'Edit' as the prefix instead of the suffix
[10:47] <mpt> ok
[10:47] <salgado> mpt, SteveA, well, the classes on security.py are not meant to be used anywhere, so I usually think it's a good idea to have their names describing what they do, even if they're too long
[10:48] <SteveA> salgado: yes, I agree.  This will change in the future, though.  But we can worry about that then.
[10:48] <carlos> salgado: I think that's what a doctest does....
[10:48] <carlos> well, not a doctest, but a doc string ;-)
[10:49] <SteveA> I started off the convention to name the classes long-style
[10:49] <SteveA> We can re-examine the convention if it has ceased to be useful
[10:49] <SteveA> or people have better ideas
[10:49] <mpt> eek
[10:49] <mpt> carlos, I left a "deliberate" bug in product-translations.pt that you didn't spot :-)
[10:50] <carlos> that 'eek' sounds like it was not deliberate ;-)
[10:50] <mpt> true, true
[10:50] <mpt> I left off the condition from <tal:untranslatable>, which would have resulted in two headings
[10:51] <jamesh> SteveA: if we're happy with that spec import, I could run it on production.  I modified the script to be able to update existing registered specs as you suggested, so if we make it pick up more authors later we can rerun it
[10:54] <SteveA> jamesh: cool.  I'm just chatting to Phillip von Weitershausen about this...
[10:54] <SteveA> He's agreed to discuss this with the upstream developers
[10:54] <SteveA> and we can talk with them at EuroPython
[10:55] <jamesh> okay.  I guess making the import available on staging might make things easier politically
[10:56] <lifeless> SteveA: david is just finishing a tea, then I think we will pop over
[10:56] <SteveA> okay, great
[10:56] <SteveA> i'm interested in lunch soon
[10:56] <SteveA> jamesh: let's leave it on staging until I've talked to a few people at EuroPython
[10:56] <lifeless> where are you thinking of ?
[10:59] <SteveA> I hadn't given it any thought.  How about skonis ir kvapas, the tea place.
[11:00] <lifeless> sure ( I dont remember it ;))
[11:09] <SteveA> spiv / stub: was there any issue with twisted libraries on balleny?
[11:10] <SteveA> I have a PQM error report from Mark that has some odd twisted failures in it.
[11:10] <stub> Not that I'm aware of. I thought we only used twisted included in our tree?
[11:12] <salgado> SteveA, flacoste had a twisted failure too. he submitted the branch again and it passed
[11:12] <SteveA> flacoste: if this happens, please always mail the launchpad list.  others may have the same problem, and we should get it sorted out.
[11:12] <flacoste> SteveA: ok
[11:13] <SteveA> https://chinstrap.ubuntu.com/~dsilvers/paste/fileseIe1R.html
[11:13] <SteveA> spiv, stub: https://chinstrap.ubuntu.com/~dsilvers/paste/fileseIe1R.html
[11:13] <SteveA> that's the failure message
[11:15] <stub> Why are we running twisted tests on Launchpad commit? It is a one way dependancy, not two way.
[11:15] <stub> I've never seen something like that anyway
[11:16] <SteveA> stub: I had a long discussion with lifeless about the fact that we run twisted (and other) tests on a Launchpad commit
[11:17] <SteveA> My interpretation of the outcome is that we agree that ideally we should have some kind of dependency representation to say what tests should be run for what commits
[11:17] <SteveA> and pqm should use that
[11:17] <SteveA> but that it does no particular harm to run them all right now
[11:19] <lifeless> agreed
[11:19] <lifeless> we're coming over, meet you outside in 10 ?
[11:19] <SteveA> well
[11:19] <SteveA> this is complicated
[11:20] <SteveA> apparently I have a conf call sometime
[11:20] <SteveA> but kiko would not tell me when!
[11:20] <lifeless> well
[11:20] <lifeless> KIKO!
[11:22] <kiko> oh ffs
[11:23] <kiko> can it be made clear that I'm BUSY
[11:24] <lifeless> possibly, but I'm not sure how
[11:24] <SteveA> lifeless: come round
[11:24] <SteveA> the call will be later thus afternoon
[11:24] <lifeless> ok, see you in < 10
[11:24] <SteveA> ok
[11:29] <elmo> sorry, but once again *.ubuntu.com, *.launchpad.net and anything else in the Canonical Data Centre is going away for 5 mintes
[11:29] <jsgotangco> :/
[11:32] <spiv> SteveA: That test seems to be unreliable.  I'll just disable that one entirely, not I'm convinced it's sane.  I'll also file a bug upstream.
[11:32] <SteveA> okay, thanks
[12:22] <mpt> spiv, how's bug 39814 going? :-)
[12:22] <Ubugtu> Malone bug 39814 in launchpad "Misleading login hint" [Medium,Confirmed]  http://launchpad.net/bugs/39814
 mpt, how do you feel about s/Not for us/Wont Fix Here/ in bug 36059?
[12:30] <salgado> stub, around?
[12:56] <mpt> kiko, I originally thought of "Won't Fix Here", but after we had a big thread across launchpad@ and ubuntu-devel@ on the best wording, I was eventually persuaded to change it to "Not For Us"
[01:00] <mpt> rats
[01:01] <kiko_> mpt, yeah, but could we roll back on that decision? :)
[01:02] <mpt> kiko, I can't remember exactly what the reasons were, but I have a vague memory of them being good ones
[01:02] <kiko> mpt, can you dig that up and update the bug so I can consider? It affects us directly when doing distro release management :-/
[01:03] <mpt> https://lists.ubuntu.com/archives/launchpad-users/2006-March/thread.html#104 ...
[01:07] <mpt> hmm, that's not it
[01:07] <mpt> the main thread had a whole bunch of synonyms, plus mdz
[01:10] <BjornT> mpt, kiko: it was discussed on the launchpad list, in the "Bugging questions" thread
[01:11] <sabdfl> stub: ping
[01:11] <stub> sabdfl: pong
[01:11] <mpt> BjornT, snap, just found it :-)
[01:11] <sabdfl> stub: https://launchpad.net/products/launchpad/+spec/canonical-names
[01:12] <sabdfl> i've just put you down as the assignee for that, could you take a look and come back to me if you think that's inappropriate?
[01:12] <sabdfl> also, today's blueprint landing added some bits to fti.py (just full text searching for Specification table, nothing fancy), not sure if that affects your rollout plans for next week at all
[01:13] <stub> Ok.
[01:13] <sabdfl> thanks
[01:13] <stub> Thanks for the heads up - it will affect the rollout as I either need to schedule more downtime or try rebuilding the indexes live.
[01:13] <stub> Although specs should be fast
[01:13] <stub> Staging will let me know
[01:14] <SteveA> the vilnius posse return from lunch
[01:15] <sabdfl> specs should be very fast, it's just title and summary
[01:15] <sabdfl> hmm... perhaps name should go in there as well?
[01:15] <stub> sabdfl: I'm already familiar with the ideas in the spec and think the idea is fine. Do I need to do anything else with it?
[01:15] <sabdfl> stub: the canonical names stuff will have a fantastic ripple impact on all sorts of things, from URL's to email interfaces to xml-rpc API's, so i'd really like to see that landing asap
[01:16] <sabdfl> stub: implement it :-)
[01:16] <sabdfl> i think the uniqueness should be project/product/distro
[01:16] <sabdfl> people are separate
[01:16] <sabdfl> we need to provide for some "blacklisted names"
[01:17] <sabdfl> much as elmo blacklists potential email addresses members can get @ubuntu.com
[01:17] <stub> There is a spec for blacklisted person.name.
[01:17] <sabdfl> generalise to canonical-names
[01:24] <mpt> kiko, done
[01:26] <sabdfl> mpt: we have a bit of a problem with LP on IE, are you tracking that?
[01:26] <sabdfl> i would see it as one of the biggest blockers for LP adoption by Python.org
[01:27] <SteveA> sabdfl: I asked mpt to do a walk-through.  What specific issues do you have for IE?
[01:28] <mpt> sabdfl, I improved the display in IE a couple of weeks ago, and haven't found any noticable problems since
[01:29] <sabdfl> i tried to show the intel folks in russia bits of LP and page rendering was b0rked all over the place
[01:29] <sabdfl> could it be they had an older version of ie?
[01:29] <mpt> Maybe they had IE5...
[01:30] <mpt> CrossOver just has v6
[01:30] <lifeless> 5 was godawful
[01:31] <OgMaciel> sorry to budge in, but could you guys also look into this bug:  https://launchpad.net/distros/ubuntu/+source/xorg/+bug/49789
[01:31] <Ubugtu> Malone bug 49789 in xorg "Clicking on Codes of Conduct link in Launchpad crashes X" [Untriaged,Confirmed]  
[01:31] <OgMaciel> I just received an email from someone who stated that X is crashing when he clicks on the download link for the PPC ISO
[01:32] <OgMaciel> and good morning everyone by the way  ;)
[01:32] <lifeless> OgMaciel: this is really an ubuntu problem - #ubuntu - not a launchpad problem. According to that bug its been assigned to xorg, and rejected as being a launchpad issue.
[01:32] <mpt> OgMaciel: And I fixed the Code of Conduct page bug anyway
[01:33] <stub> Which will be live on Tuesday
[01:33] <OgMaciel> mpt: lemme try it once again... If I get disconnected it's b/c something else could be doing it
[01:33] <OgMaciel> oh
[01:33] <OgMaciel> not implemented yet stub?
[01:33] <mpt> Does the page work for you now?
[01:33] <OgMaciel> mpt: am afraid of clicking on it..  ;)
[01:33] <stub> Not rolled out.
[01:34] <OgMaciel> mpt: I have a lot of things open
[01:34] <OgMaciel> lifeless: so you're saying I should bring this up at #ubuntu?
[01:35] <stub> A random web page shouldn't be able to crash the browser, let alone X. It is a X bug, and possibly a security vulnerability.
[01:36] <SteveA> start up a new X server
[01:36] <SteveA> startx -- :1   kinda thing
[01:36] <OgMaciel> stub: is there a public calendar for when things take place in LP, like the deployment of the new page?
[01:36] <OgMaciel> SteveA: I could do that
[01:36] <OgMaciel> lemme give it a shot using my Tablet
[01:36] <stub> No - that is maintained on a private wiki I'm afraid.
[01:37] <kiko_> mpt, I know you can run 5 5.5 and 6 side-by-side in a windows box
[01:37] <kiko_> and we have that in the offices in brazil
[01:37] <OgMaciel> stub: gotcha... any ideia when I can expect to see the rollout and could you also check the download links for the PPC ISO for me, please?
[01:37] <stub> Tuesday around 05:00 UTC
[01:37] <mpt> kiko_, ok, I'm packing my bags
[01:38] <OgMaciel> stub: thanks  ;)
[01:38] <OgMaciel> kiko_: so am I  ;)
[01:39] <kiko_> mpt, huh? where are you going?
[01:39] <kiko_> oh
[01:39] <kiko_> right.
[01:39] <mpt> :-)
[01:41] <OgMaciel> mpt: it still crashes
[01:41] <mpt> OgMaciel, that's probably because, as stub says, the fix isn't rolled out yet
[01:41] <lifeless> mpt: do you think the bug is in launchpad ?
[01:41] <OgMaciel> mpt: right... will give it a shot next Tuesday then
[01:42] <lifeless> mpt: surely nothing we do can cause an application/X crash without the app/X being faulty
[01:42] <mpt> OgMaciel, and mention it in the Launchpad bug report, that would be great
[01:42] <OgMaciel> mpt: sure thing...  I first reported it...  ;)
[01:42] <mpt> lifeless, a Launchpad bug (ridiculously long <label>) was revealing a bug in Xorg. I'm well aware that the Xorg bug will still need to be fixed after the Launchpad fix is rolled out.
[01:43] <OgMaciel> mpt: I really needed it fixed so people can sign the CoC and start contributing
[01:43] <salgado> stub, why do we have a specia tearDown for karmaupdater.txt on test_system_documentation.py ?
[01:43] <kiko_> mpt, how long?
[01:43] <mpt> kiko_, the entire text of a signed code of conduct long
[01:44] <mpt> on one line
[01:44] <stub> salgado: I've redone most of that
[01:44] <kiko_> that's pretty doh
[01:44] <mpt> lifeless, the Launchpad bug is bug 39995, and the Xorg bug is bug 49789
[01:44] <Ubugtu> Malone bug 39995 in launchpad "No Line break in CoC active signatures" [Low,Fix committed]  http://launchpad.net/bugs/39995
[01:44] <Ubugtu> Malone bug 49789 in xorg "Clicking on Codes of Conduct link in Launchpad crashes X" [Untriaged,Confirmed]  http://launchpad.net/bugs/49789
[01:44] <salgado> stub, in that branch that I need to review?
[01:44] <lifeless> mpt: ah, the xotrg one probably wants a better title, 'ridiculously long label crashes xorg' ;)
[01:45] <OgMaciel> well, thanks mpt, lifeless, stub...  got to go to work
[01:45] <stub> salgado: It would have been because karmaupdater needs to explicity flag the database as being dirty, as our detect database changes code is not smart enough to detect database changes made by subprocesses.
[01:45] <stub> salgado: Nope.
[01:46] <mpt> lifeless, yes, some fearless person needs to produce a minimal test case for 49789 (it's probable, but not certain, that it's the <label>)
[01:46] <salgado> stub, so, all tests that run a script as a subprocess need to do something like that?
[01:47] <stub> salgado: If they make database changes, yes. I know how to fix this if it becomes a problem.
[01:51] <salgado> stub, so, you have a branch that changes that special tearDown thing? are you going to land it soon?
[01:51] <salgado> I'm doing some work on karma and I moved the test for the karma updater script together with some other tests on another file
[01:52] <stub> salgado: I've got a branch that will hopefully land soon - it should be up for review tonight - that reworks test_system_documentation quite a lot
[01:53] <mpt> spiv, around?
[01:54] <salgado> stub, ah, okay. thanks
[01:56] <SteveA> mpt: I just wrote a quick and dirty script to give numbers for each different user agent from the launchpad web logs
[01:58] <mpt> cool
[02:20] <SteveA> mpt: 19 May to 30 June:   https://chinstrap.canonical.com/~stevea/user-agents.txt
[02:21] <SteveA> format is number of requests, user agent string
[02:22] <SteveA> this is done by unique user agent string
[02:22] <SteveA> so obviously the data can be combined in more intelligent ways
[02:25] <sabdfl> mpt: i think i dropped off previously
[02:25] <sabdfl> can you list out your current roadmap, please?
[02:37] <ruffneck> does the live cd have antivirus?
[02:37] <ruffneck> I should try to fix my sisters windows ;P
[02:41] <SteveA> ask on #ubuntu
[02:41] <SteveA> this channel is for discussing the launchpad website
[02:52] <sabdfl> SteveA: do i mark bugs fixed when the fix goes into production, or when it lands in mainline?
[02:54] <SteveA> when it is in RF, you can mark it "fix committed"
[02:54] <SteveA> when it is in production, it is marked "fix released"
[02:54] <sabdfl> stub: do you want to do an more with https://launchpad.net/products/launchpad/+spec/database-locale ?
[02:54] <sabdfl> thanks SteveA
[03:06] <stub> sabdfl: Eventually, yes. I need to run some more tests.
[03:07] <sabdfl> cool i will leave it there
[03:07] <sabdfl> is https://launchpad.net/products/launchpad/+spec/validating-sign-only-gpg-keys already implemented?
[03:13] <SteveA> yes
[03:39] <kiko> hey
[03:39] <kiko> where's launchpad?
[03:39] <kiko> okay, I see 
[03:42] <sabdfl> SteveA: can we re-assign https://launchpad.net/products/launchpad/+spec/form-layout to someone on the infrastructure team?
[03:44] <sabdfl> also https://launchpad.net/products/launchpad/+spec/person-name-presentation ?
[03:44] <SteveA> please assign it to launchpad-infrastructure
[03:44] <sabdfl> SteveA: i don't want team assignee's
[03:44] <sabdfl> nobody really watches it
[03:44] <sabdfl> i also do not want team approvers
[03:44] <sabdfl> or drafters
[03:44] <SteveA> i can understand it for approvers and drafters
[03:44] <sabdfl> it's more important for assignee
[03:45] <sabdfl> each person should know what's on their plate
[03:45] <kiko> SteveA, what mark is looking for is to make a plan for the next 6-month cycle for infrastructure
[03:45] <sabdfl> if it doesn't matter who, then pick someone who has the least on their plate
[03:45] <sabdfl> https://launchpad.net/products/launchpad/1.0/+specs
[03:45] <SteveA> if we're okay with it being reassigned when the spec reaches the top of the priority queue
[03:45] <sabdfl> i'd like to agree a set of 1.0 goals for October
[03:45] <kiko> so we're looking for an idea of what we agree that people should be doing
[03:45] <kiko> SteveA, sabdfl: I suggest #cm2
[03:46] <sabdfl> sure, i'm happy for it to be reassigned later
[03:46] <SteveA> then that's fine
[03:46] <SteveA> i'll skim the specs and propose people
[03:47] <sabdfl> SteveA: we're covering it here, will ping you for specific ones that look awkward
[03:47] <sabdfl> can run through the whole list in a call later
[03:47] <SteveA> right
[03:47] <sabdfl> #cm2?
[03:47] <SteveA> I'm talking about the specs you just mentioned
[03:47] <SteveA> i'm mostly working on the talks with ddaa and lifeless
[03:48] <sabdfl> right, so we don't want to disturb you for the whole discussion, just where things come up that need a decision
[04:00] <kiko> flacoste, BjornT: the ticket tracker email interface is deployed and works now. can I mark the spec as such?
[04:03] <kiko> hello
[04:03] <kiko> bradb, salgado: poke them :-P
[04:03] <BjornT> kiko: TicketTrackerEmailInterface and TicketTrackerOutgoingEmail are implemented, TicketTrackerIncomingEmail is probably not needed.
[04:03] <lifeless> would it make sense for the per-person specs pages to show the specs from my teams?
[04:03] <kiko> okay!
[04:03] <kiko> BjornT, obsoleted.
[04:04] <lifeless> I mean, if a spec might be reassigned to me from another teammate later, its not something I would have seen coming. But if it was assigned to the team in the first place, and when I look at my list of specs it shows up there automatically because I am in the team, then I know its in the pipeline and that I *might* have it be assigned to me
[04:05] <sabdfl> stub: can you do regex substitutions in SQL?
[04:05] <sabdfl> s/wiki.launchpad.canonical.com/launchpad.canonical.com/ in specification.specurl?
[04:06] <jsgotangco> good evening
[04:08] <SteveA> lifeless: I think a similar feature would be very helpful in Malone.
[04:08] <SteveA> we already use assignment to teams in Malone
[04:08] <SteveA> but the assignments are not particularly visible
[04:09] <lifeless> theres probably a gnereal pattern here
[04:09] <lifeless> that anything assigned to a 'team' should show up in the 'assigned to me' pages for all members of that team
[04:10] <sabdfl> https://launchpad.net/products/launchpad-bazaar/1.0/+specs
[04:11] <sabdfl> lifeless: please pass that on to DDAA
[04:11] <lifeless> done
[04:28] <kiko> SteveA, spiv: authservercaching is already done, isn't it?
[04:29] <SteveA> no
[04:29] <kiko> ok.
[04:29] <SteveA> it isn't particularly needed
[04:30] <SteveA> database downtime is small now
[04:30] <kiko> can we obsolete it?
[04:30] <SteveA> and we'll be moving to replicated databases before we really need authsesrver cacheing
[04:30] <SteveA> i guess
[04:30] <SteveA> it should not be on a roadmap
[04:31] <SteveA> so however the spec system allows you to say that
[04:31] <kiko> ok
[04:53] <sabdfl> stub, SteveA: can we nuke https://launchpad.net/products/launchpad/+spec/special-teams
[04:53] <sabdfl> i'm ok with requiring some additional security precautions for certain things, but i think special-teams is overkill
[04:55] <SteveA> this is old thinking
[04:55] <SteveA> so, yes, please nuke it
[04:56] <sabdfl> it's *gone* :-)
[04:57] <SteveA> cool
[04:57] <SteveA> ta
[04:58] <kiko> so SteveA tell me about description-markup. it's being worked on, is it not?
[04:58] <SteveA> it is
[04:58] <SteveA> mpt is polishing the spec, then jamesh will implement once we have the spec approved
[04:59] <kiko> ok.
[05:24] <kiko> salgado_, teamvoting, implemented, right?
[05:26] <stub> sabdfl: If the functions documented at file:///usr/share/doc/postgresql-doc-8.1/html/functions-string.html aren't enough, it isn't much work to create a Python stored procedure to do the text manipulation you need.
[05:27] <sabdfl> stub: we moved a wiki and have a ton of specs that need to be updated
[05:27] <kiko> bradb, BjornT, flacoste: can you nudge salgado?
[05:28] <sabdfl> could you do the update directly? perhaps on staging so we can test it first?
[05:28] <stub> sabdfl: Also file:///usr/share/doc/postgresql-doc-8.1/html/functions-matching.html#FUNCTIONS-POSIX-REGEXP - regexp_replace looks like it is what you asked for
[05:28] <sabdfl> need to update Specification.specurl
[05:28] <stub> sabdfl: Move what to what? wiki.launchpad.net to launchpad.canonical.com?
[05:29] <flacoste> flacoste: salgado has internet connection problem he says that teamvoting is not implemented yet
[05:29] <flacoste> kiko: salgado has internet connection problem he says that teamvoting is not implemented yet
[05:29] <kiko> stub, wiki.launchpad.canonical.com -> launchpad.canonical.com
[05:29] <SteveA> stub: we shouldn't have ever been using wiki.launchpad.net
[05:29] <kiko> flacoste, really? what's missing? what it does is link to basicvoting, mostly
[05:29] <sabdfl> wiki.launchpad.canonical.com to launchpad.canonical.com
[05:29] <sabdfl> stub: ^
[05:30] <sabdfl> doh, sorry, slow
[05:30] <kiko> salgado!
[05:30] <flacoste> kiko: then it is implemented, salgado thought it was something else
[05:30] <kiko> my man!
[05:30] <kiko> ok.
[05:35] <SteveA> carlos: ping
[05:39] <kiko> matsubara, wakey wakey
[05:39] <stub> sabdfl, kiko: staging updated. let me know if that is fine for production.
[05:39] <kiko> thanks
[05:53] <sabdfl> stub: looks perfect
[05:53] <sabdfl> thanks!
[05:54] <stub> Run on production
[06:00] <sabdfl> bradb: ping
[06:00] <sabdfl> https://launchpad.net/products/malone/1.0/+specs
[06:00] <sabdfl> thanks stubarooney
[06:02] <bradb> sabdfl: pong
[06:03] <bradb> short and sweet! we're about halfway through coding up a guided filebug form too
[06:03] <kiko> nice
[06:04] <sabdfl> bradb: excellent
[06:04] <kiko> screenshots!
[06:04] <sabdfl> bradb: please can you use "Proper specification titles" and not "WikIWordSpecTitles"?
[06:04] <sabdfl> we just got spec searching and that will not help
[06:04] <sabdfl> could you fix all the specs on malone this w/e please?
[06:04] <bradb> sure
[06:05] <sabdfl> also, please use spec-name-format not specnameformat
[06:05] <sabdfl> thanks :-)
[06:05] <sabdfl> we will get better search fu that way
[06:05] <bradb> sounds good
[06:13] <kiko> carlos, ping?
[06:18] <bradb> sabdfl: BTW, maybe the +specs listing should show spec titles instead of names?
[06:21] <kiko> I don't know, they are rather verbose
[06:24] <bradb> I thought that at first too, but I think they would mostly fit in 1024 x 768
[06:26] <sabdfl> carlos: which is the spec for "language" pages?
[06:26] <sabdfl> pages for the actual language objects?
[06:26] <sabdfl> where we could see translators, stuff which needs translation etc
[06:30] <kiko> thanks bradb 
[06:30] <bradb> no prob
[06:33] <LaserJock> anybody know how often bazaar.launchpad.net updates the mirror bzr branches?
[06:49] <sabdfl> Kinnison: https://launchpad.net/products/soyuz/+spec/upload-privileges is that all done now?
[06:51] <carlos> SteveA, kiko: pong
[06:51] <lifeless> carlos: stevea mailed you a partial reply
[06:52] <lifeless> s/reply/review
[06:52] <carlos> sabdfl: https://launchpad.net/products/rosetta/+spec/language-administration ?
[06:52] <carlos> ok
[06:52] <Kinnison> sabdfl: one sec, let me check
[06:53] <Kinnison> sabdfl: The backend ues, the UI is incomplete, as per cprov's comment in the status whiteboard
[08:22] <sivang> lifeless: I tried sftp'ing but it would "Connection reset by peer" (to try fix the partially pushed archive of home-user-backup product) Is there a spcial way to sftp there?
[08:29] <lifeless> i use lftp
[08:29] <lifeless> lftp sftp://bazaar.launhcpad.net/....
[08:34] <sivang> lifeless: if I pushed it to be under the ~ubuntu-dev team, am I still allowed to sftp there and remove stuff?
[08:34] <sivang> (I think maybe the fact it's under the team's control, it does not let me sftp there)
[08:34] <lifeless> yes you are
[08:34] <lifeless> pupshing uses sftp
[08:43] <sivang> lifeless: is there a real difference to using sftp or lftp ? 
[09:01] <lifeless> sivang: lftp takes urls, and is much nicer to use - like has recursive delte
[09:04] <sivang> lifeless: I see nice, still it asks me for password and I am using SSH keys ... any idea?
[09:04] <lifeless> hit enter ?
[09:04] <lifeless> perhaps check your username is right ?
[09:05] <sivang> lifeless: username is as lp username?
[09:05] <sivang> (e.g. $NAME@ubuntu.com)
[09:06] <lifeless> sftp://$NAME@bazaar.launchpad.net/...
[09:06] <lifeless> if you dont have your ssh config setup to know, and it iss different to your unix account
[09:07] <lifeless> just use the exact same url you gave bzr ;)
[09:08] <sivang> I'll retry, IIRC that's exactly what I did before :)
[09:18] <sivang> lifeless: yay, worked
[09:19] <sivang> lifeless: cleaning up the dead branch now
[09:23] <sivang> lifeless: I see I can navigate quite some there, hmm, actually it has the homes of all the teams I am member of
[09:25] <sivang> crap, still getting the same error
[09:25] <sivang> bzr: ERROR: File exists: '/~ubuntu-dev/hubackup/ubuntu': mkdir failed: unable to mkdir
[09:26] <lifeless> sivang: did you leave the .bzr dir, and remove .bzr/branch ?
[09:26] <sivang> lifeless: oops, I removed the whole .bzr dir
[09:27] <lifeless> sivang: you will need to use bzrlib then
[09:27] <lifeless> python
[09:27] <lifeless> import bzrlib.bzrdir
[09:27] <lifeless> bzrlid.bzrdir.BzrDir.create_branch_convenience(URL)
[09:29] <sivang> Out[3] : BzrBranch5('sftp://sivan@bazaar.launchpad.net/%7Eubuntu-dev/hubackup/ubuntu/')
[09:29] <sivang> should I retry pushing ?
[09:31] <lifeless> ues
[09:32] <sivang> lifeless: cool, it's working. Can we get that convinience option to have a cmd line arg in bzr?
[09:33] <sivang> lifeless: I'd reckon it's quite useful for lots of partially pushed branches
[09:33] <lifeless> IIRC there is a bug open already
[09:33] <sivang> ah, cool
[09:33] <lifeless> to allow pushing into an empty dir
[09:33] <lifeless> (and I think a .bzr dir with no branch is already supported)
[09:34] <sivang> lifeless: oh, nice