[04:12] <jamesh> lifeless: did you see the report on the product-release-finder run?
[04:13] <lifeless> yes, have skimmed part of it so far
[04:13] <lifeless> in sprint mode for 0.11 at the moment
[04:13] <jamesh> okay.
[04:14] <jamesh> I'm pretty happy with the results myself.
[04:15] <lifeless> jamesh: excellent. You know about the plan to map via the file alias table to find distro spr's, and thus spn's to link to ?
[04:16] <jamesh> you mentioned it yesterday, yeah.
[04:16] <lifeless> theres a spec somewhere
[04:19] <jamesh> https://features.launchpad.net/products/launchpad-bazaar/+spec/tracking-versions
[04:20] <lifeless> jamesh: nope
[04:20] <lifeless> theres a much older spec
[04:20] <lifeless> before blueprint
[04:20] <jamesh> should register it then ...
[04:21] <lifeless> need to find it first :(
[04:22] <lifeless> https://launchpad.canonical.com/PackagingTablePopulation
[04:23] <jamesh> so is the plan to match up source packages with tarballs aliased to the same content as product-release-finder found for a series?
[04:23] <lifeless> yes
[04:29] <jamesh> now its in blueprint ...
[04:29] <lifeless> thank you
[04:31] <jamesh> By the way, I think it would be good to simplify the PRF related data to a single field in productseries rather than releaseroot and releasefileglob
[04:32] <jamesh> the current PRF code just concatenates them anyway ...
[04:44] <lifeless> jamesh: I agree
[04:44] <lifeless> jamesh: have you looked at the uscan manpage ?
[04:44] <jamesh> yep
[04:45] <lifeless> might be nice to have uscan support, so that folk can copy n paste
[04:49] <jamesh> it sounds like our syntax can handle a few cases uscan can't
[04:49] <jamesh> (or at least our implementation)
[05:01] <mpt> jamesh, got time for a quick pre-implementation discussion? I'm thinking of fixing bug 46
[05:01] <Ubugtu> Malone bug 46 in rosetta ""special symbols" when people copy-paste text from original to translation" [High,Confirmed]  http://launchpad.net/bugs/46
[05:01] <jamesh> mpt: okay
[05:02] <mpt> jamesh, this won't need a voice call, it's pretty simple I think
[05:02] <mpt> In interfaces/rosetta.py, I'll change SPACE_CHAR to <img alt=" " src="something" />, and NEWLINE_CHAR to <img alt="" src="something" /><br />
[05:03] <jamesh> what are they at the moment?
[05:04] <mpt> <span class="po-message-special">?</span> and <span class="po-message-special">somefunnycharacter</span>, respectively
[05:04] <mpt> With an image, when people copy and paste they'll get the alt text, which will Just Work
[05:04] <mpt> because the alt text for the space will be " ", and the alt text for the newline will be nothing at all (with the image itself being followed by a newline)
[05:05] <jamesh> sounds good.  Does that work in all the common browsers?
[05:06] <mpt> It works in Gecko, I haven't checked the others, but I think it can't really be worse than the current behavior
[05:06] <mpt> s/really/easily/
[05:06] <jamesh> well, you could end up with no white space ...
[05:10] <mpt> works in Konqueror...
[05:10] <mpt> Doesn't work in Opera
[05:11] <mpt> or Internet Explorer
[05:11] <jamesh> I wonder what happens if you have "foo<img src="space.png" alt=" "> </img>bar"?
[05:12] <mpt> A validation error, that's what :-)
[05:12] <jamesh> maybe we need "foo<img src="space.png" alt=""><div style="display: none"> </div>bar"
[05:12] <mpt> By "doesn't work" I mean "doesn't copy the alt text", but that will still be an improvement for the newline case
[05:12] <mpt> because nothing is the correct thing to copy
[05:13] <mpt> Where by "an improvement" I mean "perfect behavior" :-)
[05:13] <jamesh> that should be a <span> rather than <div>
[05:14] <mpt> That's a whole 'nother bag of worms, jamesh 
[05:14] <jamesh> so do they do or don't copy text from "display: none" elements?
[05:15] <mpt> I remember there's something in bugzilla.mozilla.org about that
[05:15] <mpt> testing now
[05:15] <mpt> Gecko does
[05:15] <mpt> include the text
[05:16] <mpt> but that's regarded as a bug
[05:16] <mpt> Opera does not include the text
[05:16] <jamesh> okay
[05:17] <mpt> neither does Konqueror
[05:17] <jamesh> I guess having something that works in Mozilla is good enough
[05:17] <mpt> so that's a bit worse
[05:17] <jamesh> the number of people using Launchpad to translate things into english is fairly small ...
[05:17] <jamesh> so they're probably mostly using mozilla or konqueror
[05:17] <mpt> So to summarize, the only thing wrong with the new approach is that leading spaces will be lost if blindly copying-and-pasting using Opera or Internet Explorer.
[05:18] <stub> We can provide a link to the raw text/plain version if that is actually a problem
[05:18] <mpt> Whereas the old approach generates bad characters, for either spaces *or* newlines, if copying-and-pasting using *any* browser.
[05:19] <jamesh> a "copy orginal" button/link is probably more useful overall
[05:19] <jamesh> rather than have people make copy/paste errors
[05:20] <mpt> sure, but it would be appropriate to fix this even if that button was already implemented
[05:20] <mpt> because a few people won't use the button for whatever reason
[05:21] <jamesh> the other way would be something like <span style="background-image: url(space.png)"> </span>
[05:21] <jamesh> (with appropriate height/width settings)
[05:22] <jamesh> that sould copy/paste fine in any browser, right?
[05:22] <mpt> nifty!
[05:25] <mpt> Ok, I'll do that for the spaces
[05:25] <mpt> thanks jamesh 
[05:26] <jamesh> you could probably do an empty span for the newlines, right?
[05:26] <jamesh> <span style="background-image: url(newline.png); height: x; width: x"></span><br />
[05:26] <mpt> I could, but an <img alt="" would work fine for newlines
[05:28] <mpt> and would work even if CSS was turned off
[05:28] <mpt> so I'm trying to think of a reason it would be bad to DTRT for newlines but not spaces
[05:29] <mpt> oh, spaces should be ...>&nbsp;<... rather than ...> <...
[06:21] <mpt> (actually, they should be ... style="white-space: pre;"> <...)
[07:04] <mpt> jamesh, ping
[07:06] <mpt> jamesh, unping
[08:33] <jamesh> mpt: btw, I put together a few helpers for page tests in my jamesh/launchpad/bug-55649 branch
[08:33] <jamesh> mpt: allowing you to do e.g. find_tag_by_id(browser.contents, "foo"), find_portlet(browser.contents, "The title of the portlet"), etc
[08:34] <jamesh> should make it easier to test particular parts of the page
[08:35] <BjornT> lifeless: ping
[09:01] <SteveA> morning
[09:03] <SteveA> lifeless: ping?
[09:06] <carlos> morning
[09:06] <carlos> mpt: hi, around?
[09:46] <lifeless> BjornT: pong
[09:46] <lifeless> SteveA: pong
[09:48] <BjornT> lifeless: so, this zope thing. what's the issue there?
[09:48] <lifeless> right
[09:48] <lifeless> let me see
[09:51] <lifeless> its misconfigured in pqm
[09:51] <lifeless> stub: did you change the zope commit configuraiton for pqnm ?
[09:52] <stub> Nope
[09:52] <stub> At least I don't think so....
[09:53] <lifeless> SteveA: are commits to zope currently meant to run the launchpad tests ?
[09:53] <jamesh> they probably should -- zope API changes could break Launchpad
[09:53] <BjornT> lifeless: yes, they should.
[09:54] <lifeless> thats my opinion too
[09:54] <lifeless> but the setup is bust
[09:54] <lifeless> BjornT: you can commit to zope 3.2 now
[09:54] <BjornT> cool, thanks.
[09:56] <SteveA> lifeless: I'd like commits to zope to run the launchpad tests, and the zope tesst
[09:57] <lifeless> as long as make check_merge does that still, that is what will happen
[09:58] <lifeless> bjorn could not commit because the rule was set to run check_merge, but not to build the full tree with launchpad etc
[10:25] <mpt> carlos!
[10:26] <mpt> jamesh, cool
[10:26] <mpt> carlos, if you do me a small favor, I'll do you a couple of big favors
[10:31] <carlos> mpt: not sure... :-P
[10:31] <carlos> mpt: are they related to bug #46?
[10:31] <Ubugtu> Malone bug 46 in rosetta ""special symbols" when people copy-paste text from original to translation" [High,In progress]  http://launchpad.net/bugs/46
[10:32] <mpt> carlos, yes
[10:32] <mpt> I've fixed it, but I don't know how to test it
[10:32] <carlos> mpt: I wanted to talk about them
[10:32] <carlos> oh, so you did it already....
[10:32] <carlos> conflicts :-(
[10:32] <carlos> ok, I guess the small favor is to write the tests, right?
[10:32] <mpt> carlos, so what I would like is for you to write me a single simple test for msgid_has_tab
[10:33] <mpt> and I'll write the rest.
[10:33] <carlos> ok
[10:33] <carlos> don't we have already a test for the tab case?
[10:33] <mpt> Not that I could find
[10:33] <mpt> I couldn't find "[tab] " in either launchpad/doc/ or launchpad/pagetests/
[10:34] <carlos> ok, I will write such test and send it to you
[10:35] <jamesh> mpt: for your change, don't you just need to do a few msgid_html() calls in the test?
[10:36] <mpt> jamesh, the function returns True or False
[10:36] <mpt> for a msgid containing particular text
[10:36] <mpt> Getting HTML involved would make it unnecessarily brittle, I think
[10:37] <mpt> carlos, by "conflicts" did you mean you were fixing it already?
[10:38] <jamesh> mpt: aren't you trying to test that the generated HTML is correct?
[10:38] <carlos> mpt: no, that I changed a lot that form to implement TranslationReview spec
[10:38] <carlos> mpt: but I will get a diff with your changes and apply them by hand
[10:39] <mpt> jamesh, I wasn't thinking of that, but you're right, I should test that separately.
[10:41] <mpt> jamesh, because msgid_html and pomsgset.py detect things like "contains a tab" in different ways. (That smells a bit -- I put an XXX in pomsgset.py complaining about it.)
[10:45] <mpt> carlos, so I'm asking you to help me make a conflict in one of your branches? hmm, that isn't very nice of me
[10:46] <mpt> I'll pull your branch and see if I'm smart enough to merge the changes in myself
[10:46] <carlos> mpt: well, I was planning to fix that bug while working on TranslationReview
[10:46] <carlos> mpt: don't worry, really, it's not a big problem, unless you rewrote a big part of that form... :-P
[10:47] <jamesh> carlos: he was just working on alternative HTML that copy/pastes correctly
[10:47] <mpt> carlos, I added a couple of extra conditional rows to the table, like the one that's currently there for [tab] 
[10:48] <mpt> "<blahblah> represents a space. Enter a space in the equivalent position in the translation."
[10:48] <carlos> oh, so you added copy/paste buttons?
[10:48] <mpt> No, I didn't.
[10:48] <carlos> just notices about what to do with such symbols?
[10:48] <carlos> that's fine
[10:49] <mpt> I (1) changed leading/trailing space presentation and newline presentation from characters to <img> elements, and (2) replaced the "Special characters:" portlet with context-sensitive help like we have for [tab] .
[10:50] <mpt> So what I'm asking you for is a test of msgid_has_tab()
[10:50] <mpt> then I'll know how to write tests for msgid_has_newline() and msgid_has_leading_or_trailing_spaces().
[10:50] <Ubugtu> New bug: #60533 in soyuz "cron.germinate doesn't need hppa anymore for edgy" [Untriaged,In progress]  http://launchpad.net/bugs/60533
[10:57] <carlos> mpt: ok
[10:58] <carlos> so the old portlet is gone now, right?
[10:58] <mpt> yep
[11:03] <carlos> cool
[11:09] <mpt> carlos, sorry, I didn't realize it would take this long, maybe you should just give me a hint instead, or point me to a similar test of another browser-class function
[11:10] <carlos> mpt: sorry, I was with other things. I'm just updating my trivial branch to write the test
[11:10] <mpt> oh, ok
[11:10] <mpt> in that case, sorry for pestering :-)
[11:13] <carlos> np
[11:19] <SteveA> bug 55992
[11:19] <Ubugtu> Malone bug 55992 in malone "bad parsing of attachment names ? (ref OOPS-223B168)" [Medium,Fix committed]  http://launchpad.net/bugs/55992
[11:25] <carlos> mpt: so, we need some sampledata addition to do that test
[11:25] <carlos> seems like we don't have any msgid in our database with the '\t' char
[11:27] <mpt> carlos, I need to test many different variations of text, so I don't think sampledata is appropriate
[11:27] <mpt> Is there some way of creating a msgid within the test?
[11:28] <mpt> potmsgset.txt seems to do something similar, but I couldn't figure out quite how it worked
[11:28] <carlos> Well, in other tests, what I have done is to update a string as part of the test, and later revert that change
[11:29] <carlos> I could do it
[11:29] <carlos> so we adapt it to our needs
[11:29] <carlos> mpt: yeah, kind of
[11:29] <carlos> I will do it once for \t and you can just change it for the others
[11:29] <mpt> thank you
[11:30] <carlos> what I don't understand is how's possible that we didn't add a check for it before....
[11:32] <mpt> Yeah, I was all happy at fixing the bugs, and then realized that it was in untested code, argh :-)
[11:33] <doko_> Unexpected response from server
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> Firefox doesn't know how to communicate with the server.
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_> 
[11:33] <doko_>     *   Check to make sure your system has the Personal Security Manager
[11:33] <doko_>           installed.
[11:33] <doko_>     *   This might be due to a non-standard configuration on the server.
[11:33] <doko_> oops ...
[11:33] <doko_> I get this browser message when trying to access launchpad ...
[11:34] <mpt> doko_, do you get the same message visiting other HTTPS sites, like https://wiki.ubuntu.com/ ?
[11:36] <jamesh> doko_: maybe you are using an unsupported development distribution?
[11:36] <jamesh> I hear they occasionally break
[11:36] <doko_> jamesh: :-P
[11:37] <doko_> mpt: yes, seems so
[11:38] <doko_> mpt: hmm, works after restarting firefox ...
[11:39] <imbrandon> any word if https://launchpad.net/bugs/58144 got put into production ( celso said it could /POSSIBLY/ go in tuesday , meaning last tuesday since I asked last sunday )
[11:39] <Ubugtu> Malone bug 58144 in soyuz "Backport is rejected if an older backport is already there" [Critical,In progress]  
[11:40] <malcc> imbrandon: No, that didn't go in yet
[11:40] <imbrandon> hum , ok
[11:41] <imbrandon> next tuesday ?
[11:41] <imbrandon> or do code pushes get pushed on other days also 
[11:42] <malcc> imbrandon: For soyuz internals like that, which don't require schema changes, we can cherrypick them on non-tuesdays
[11:42] <imbrandon> ahh could you /please/ see about that one, its holding up a fix to a now broke amarok in backports ;)
[11:42] <malcc> imbrandon: The delay is getting the branch containing the fix through the review process, which while it's excellent for code quality, does have the potential to produce potentially unlimited delays to any one possible fix
[11:42] <imbrandon> right, i totaly understand
[11:43] <malcc> imbrandon: Celso is on vacation today, but I'll pass on your priority bump when he returns on Monday
[11:43] <imbrandon> thats why i only ask every few days
[11:43] <imbrandon> ok sounds good, thanks
[11:47] <SteveA> ddaa: ping
[11:48] <ddaa> SteveA: pouet
[11:48] <carlos> mpt: https://devpad.canonical.com/~andrew/paste/fileFYpIKf.html
[11:48] <carlos> mpt: is that what you want?
[11:49] <mpt> carlos, that looks great
[11:50] <mpt> Do I need to worry about resetting the value at the end?
[11:50] <carlos> mpt: no, it's at the end of the test, testing infrastructure should reset the database....
[11:50] <mpt> cool, thank you
[11:50] <carlos> mpt: btw, instead of 'Foo\tBar' use u'Foo\tBar'
[12:12] <jamesh> SteveA: was there a decision about short URLs (e.g. https://launchpad.net/python) for Launchpad 1.0?
[12:12] <SteveA> yes
[12:13] <SteveA> stu is doing it
[12:16] <stub> But not today cause I feel crap and am going to bed.
[12:16] <stub> Think I'm taking a half day sicky - see what I end up doing on the weekend
[12:17] <mpt> bah
[12:18] <mpt> What are the python regexps for "a space at the start of a string" and "a space at the end of a string"? I thought they'd be "^ " and " $", but apparently not
[12:18] <SteveA> that's basically it
[12:19] <SteveA> >>> import re
[12:19] <SteveA> >>> re.compile('^ .*').match('foo')
[12:19] <SteveA> >>> re.compile('^ .*').match(' foo')
[12:19] <SteveA> <_sre.SRE_Match object at 0xb7dc13a0>
[12:19] <SteveA> >>>
[12:20] <mpt> if (... or ('^ ' in msgid.msgid) or ...):
[12:20] <mpt> I guess I'm misinterpreting "in"
[12:21] <SteveA> "in" is to say whether a string is a substring of another
[12:21] <mpt> ho ho, not a re. in sight
[12:21] <mpt> ok
[12:21] <SteveA> it doesn't understand regular expressions
[12:21] <mpt> spot the newbie
[12:21] <SteveA> if you want to look at the start
[12:21] <SteveA> use " foo".startswith(' ')
[12:21] <stub> msgid.msgid.startswith(' ') or msgid.msgid.endswith(' ') 
[12:21] <mpt> ah, beautiful
[12:34] <carlos> jordi: hi, around?
[12:34] <jordi> yeah
[12:34] <jordi> sort of
[12:34] <mpt> jamesh, there are already tests of msgid_html() in test_helpers.py. I think there probably wouldn't be any benefit from copying the exact HTML I used and pasting it over the dummy "XXXA" and "YYYA" used in the tests.
[12:34] <jordi> mpt: for the record, hell has been reduced to *YAIRD SUCKS*
[12:34] <mpt> jordi, YAIRD?
[12:35] <mpt> Did you have to pay your tax this week too?
[12:35] <carlos> mpt: btw, while you fix the symbols thing... what about changing all white spaces with the middle dot instead of just the ones at the start and the end?
[12:35] <mpt> (Yet Another Inland Revenue Department?)
[12:35] <mpt> carlos, because the dot is very distracting :-)
[12:35] <jordi> yaird - Yet Another mkInitRD
[12:35] <jordi> ;)
[12:35] <AlinuxOS> Hello all, is it possible to search a specific English string into rosetta ?
[12:35] <mpt> AlinuxOS, ask danilos ;-)
[12:35] <carlos> mpt: there were some requests to do it that way to be consistent all time
[12:36] <carlos> AlinuxOS: not yet, but will be possible soon
[12:36] <jamesh> carlos: you really only want to draw attention to important spaces
[12:36] <AlinuxOS> mdz, danilos seganos ? :)
[12:36] <jamesh> carlos: maybe runs of more than one space should be highlighted, but not all of them
[12:36] <jordi> carlos?
[12:36] <mpt> Leave mdz out of this
[12:36] <carlos> jordi: I don't see which email do you want I care about...
[12:36] <jordi> carlos: wait
[12:36] <AlinuxOS> mpt, I'm really happy that Danilo is involved with rosetta, he is great person.
[12:36] <mpt> jamesh, I was thinking that too, but then we'd pick up silly style issues like people using two spaces at the end of a sentence
[12:37] <mpt> (or more precisely, using two spaces between sentences)
[12:37] <danilos> AlinuxOS: thanks, as for the search, it should be available "soon" ;)
[12:37] <jordi> AlinuxOS: don't trust danilo. Under that appearence of cool guy hides pure evil :)
[12:38] <carlos> jordi++
[12:38] <danilos> jordi: hey, don't reveal all about me in a public channel :)
[12:38] <AlinuxOS> :)
[12:38] <danilos> this one is even logged on www! #$%@# ;)
[12:39] <AlinuxOS> jordi, I have some friends from Ugoslavia :) I love that people, and I love their music :) So Danilo must be great :)
[12:40] <AlinuxOS> jordi, it will be great to have some Entrans features in rosetta. Search, mutli edit, and .url file support.
[12:40] <danilos> AlinuxOS: if good music ain't enough, nothing is, so I admit that I am great :P
[12:41] <AlinuxOS> for a moment we (Georgians[ka] ) use Entrans to translate official GNOME.
[12:42] <danilos> carlos: just fyi, gnome-applets seems to include es_ES, es_MX, es_AR... probably for the default location for weather applet, but need to figure out the effect on rosetta and ubuntu ;)
[12:43] <jordi> AlinuxOS: ah, well. do what you want. But really, always keep an eye on danilo if he's around, you never know :)
[12:43] <AlinuxOS> danilos, I'm Vladimer, if you remmeber, some times ago there was an issue about Georgian team coordinator. :)
[12:43] <jordi> danilos: ugh, that's right, and sucks :|
[12:43] <danilos> AlinuxOS: yeah, I do, hi Vladimer :)
[12:43] <AlinuxOS> jordi, thanks :) I know that. He is dangerous :)
[12:43] <AlinuxOS> hehe
[12:43] <jordi> at least in the past, gnumeric also had es_ES, etc.
[12:44] <danilos> yeah, I wear my trousers upside down! beware: my pee marks are out in the open!
[12:44] <AlinuxOS> danilos, http://www.gia.ge/dormidon/main.php?root=476
[12:44] <AlinuxOS> jordi, I'm not sayng that rosetta sucks :)
[12:45] <jordi> AlinuxOS: no, rosetta is ok. Only danilo sucks here.
[12:45] <danilos> AlinuxOS: ah, nice, lets see if there are any ideas to make use of (i.e. steal :)
[12:45] <AlinuxOS> just because of some reasons and some features, Entrans is more fast (because it's on georgian server),has some features that Rosetta don't have.
[12:45] <danilos> jordi: thanks, it's always nice to have a friend back you up :P
[12:46] <AlinuxOS> danilos, yes I agree :)
[12:46] <danilos> AlinuxOS: you should combine that with http://progress.gnome.org/teams/ka (it needs more polish, so it's not announced yet)
[12:46] <AlinuxOS> you can see that Entrans is transated complitely into Georgian... it will be great to have Rosetta translatable too.
[12:46] <jordi> progress!
[12:46] <jordi> I diodn't know this
[12:46] <jordi> what other stats is it going to host?
[12:47] <danilos> jordi: well, it should actually rock really hard (docs, l10n, module blah-blah)
[12:48] <danilos> jordi: the best part is that it's machine hosted by canonical ;)
[12:48] <jordi> danilos: really? cool :D
[12:48] <danilos> AlinuxOS: those entrans guys don't know how to create CSS layers fill 100% of a graph, thus they must suck :P
[12:49] <danilos> jordi: yeah, I found about that from gnome sysadmins, and it was like: hey, we'll be getting a machine from canonical for l10n stuff
[12:49] <danilos> jordi: btw, library.g.o might end up on this one as well
[12:49] <jordi> cool
[12:49] <AlinuxOS> danilos, yes I agree, I need something fused Rosetta + Entrans :D
[12:49] <jordi> is library going to host the aPI docs, or more?
[12:50] <danilos> AlinuxOS: oh, you want carlos' pretty graphs? :)
[12:50] <AlinuxOS> I knew that Rosetta is very slow from georgian internet...is it possible to have a mirror in Georgia ?
[12:50] <danilos> jordi: both user guides and API, afaik
[12:50] <AlinuxOS> I live in Italy so for me everything is ok...but for others :/
[12:50] <danilos> AlinuxOS: that would be expensive
[12:50] <AlinuxOS> danilos, expensive for whom ?
[12:50] <jordi> danilos: hmm, the pages need timestamps
[12:50] <danilos> AlinuxOS: but what is exactly that is slow? we can and should work on improving performance if that's the problem
[12:50] <jordi> so you know when it was last generated?
[12:51] <jordi> http://progress.gnome.org/people/jordim
[12:51] <jordi> this is cool
[12:51] <jordi> I mean, it'll be cool :)
[12:51] <danilos> AlinuxOS: for whoever is hosting the mirror (our database is huge)
[12:52] <danilos> jordi: add your data to people.xml.in inside damned-lies module ;)
[12:52] <danilos> jordi: I even have a hackergotchi in there: http://progress.gnome.org/teams/sr ;)
[12:52] <jordi> oh really :D
[12:52] <AlinuxOS> danilos, refresh, and submitting translations :)
[12:53] <carlos> danilos: well, it's just a matter of approve those files manually....
[12:53] <danilos> jordi: the most important bit of info; btw, what pages need timestamps? updates are (will be) done as cvs commits are done, so they will be per-module, and there are timestamps there
[12:53] <jordi> http://progress.gnome.org/languages/ca/gnome-2-16
[12:53] <danilos> AlinuxOS: ok, so that's something we are working on, we want to improve that as well
[12:54] <jordi> this appears to be outdated, but I can't tell when it was done
[12:54] <AlinuxOS> danilos, some times ago I've requested vsichi@gnome.org :) but I've had no replies... :/
[12:54] <danilos> AlinuxOS: are you a member of Gnome Foundation?
[12:54] <AlinuxOS> danilos, do you use Edgy Eft too ? I mean do you use Ubuntu ?
[12:55] <danilos> jordi: yes, it was regenerated two months ago for the last time
[12:55] <AlinuxOS> danilos, Dear Vladimer Sichinava,
[12:55] <AlinuxOS> We are pleased to inform you that you are now part of the GNOME
[12:55] <AlinuxOS> Foundation Membership.
[12:55] <AlinuxOS> yes I've a mail from GNOME :P
[12:55] <danilos> AlinuxOS: then it should be quick, just annoy gnome-sysadmins with that ;)
[12:56] <AlinuxOS> http://progress.gnome.org/languages/ka it's great :)
[12:56] <jordi> danilos: er
[12:56] <danilos> btw, yes, I do use Ubuntu for some time already :)
[12:56] <jordi> danilos: is this sorted in any way?
[12:56] <danilos> jordi: not sure, maybe not ;)
[12:57] <AlinuxOS> danilos, annoy :) hehe
[12:58] <AlinuxOS> that mail is dated Sat,  8 Jul 2006 13:57:55 -0400 (EDT)  (19:57 CEST)
[12:58] <AlinuxOS> maybe I must wait :)
[12:58] <danilos> AlinuxOS: you don't get your @gnome.org alias automatically, you need to ask for it using policy described on live.gnome.org/NewAccounts
[12:59] <danilos> jordi: btw, another great thing is per-module overview, like http://progress.gnome.org/module/epiphany
[01:00] <AlinuxOS> danilos, http://gnome.inet.ge/ is dead, as my friend has no more time to collaborate. As I'm not web master and ths sort of things...is there some templates for gnome that I can translate and make a good Georgian GNOME community site ?
[01:01] <danilos> AlinuxOS: I don't know, best to stop by GimpNet (irc.gnome.org) and ask in web and marketing channels ;)
[01:01] <AlinuxOS> danilos, I see ;)
[01:01] <AlinuxOS> danilos, is not simple to be a coordinator. :)
[01:02] <danilos> AlinuxOS: yeah, so many hard things to do :)
[01:02] <AlinuxOS> the maximum was installing Entrans and translate it into georgian :)
[01:03] <AlinuxOS> danilos, we (Georgians) are only 4 million, so you can understand ...that there is very small community...
[01:03] <AlinuxOS> we still have too many windows xp pirate copies.
[01:03] <AlinuxOS> :/
[01:04] <danilos> AlinuxOS: Serbs are 10 million and there is almost no community at all (Linux Users Group mailing list is dead for two years already); Macedonians, on the other hand, are probably not even 4 million, yet their community rocks ;)
[01:05] <danilos> so, it's not that simple: keeping a community is a job in itself
[01:05] <AlinuxOS> danilos, are you ortodoxs or muslims in your location ?
[01:05] <AlinuxOS> danilos, community serves a lot for bug hunting.
[01:06] <danilos> AlinuxOS: orthodox, though I consider myself as someone believing in the Ancient Greek mythology ;)
[01:06] <danilos> AlinuxOS: well, in Serbian "community", I am happy if I see a person I am not familiar with posting Ubuntu screenshots in Serbian language :)
[01:09] <AlinuxOS> :)
[01:10] <AlinuxOS> I'm ortodox too, and I believe in peace and love & in open and free source !
[01:11] <mpt> Peace and love? Keep those heretical thoughts out of here
[01:12] <AlinuxOS> I hope that rosetta in near future become complete of everything(Some Entrans features that I mentioned). It will be great to have localised rosetta too :)
[01:12] <AlinuxOS> mpt, yes love love :)
[01:13] <mpt> AlinuxOS, what is ".url file support"?
[01:15] <AlinuxOS> for esample if I would like to import entire GNOME project (.po files) I can update a .url file into entrans and it stars to import files automatically.
[01:15] <AlinuxOS> .url file contains exact urs of GNOME .po files.
[01:16] <carlos> AlinuxOS: we are not going to support that feature, at least in the near future
[01:16] <carlos> AlinuxOS: the first way we will support automatic imports from upstream is using Bazaar trees
[01:16] <danilos> AlinuxOS: well, we do that even better: you don't have to use any URL file, it's done automatically from software packages
[01:16] <danilos> carlos: btw, it's simple to do with Rosetta: use that .url file to download all PO files using "wget", tar them up, and upload a tar file containing translations ;)
[01:18] <AlinuxOS> carlos, danilos, yes but in GNOME project's case, as GNOME package's collection is quite different that ubuntu-desktop, I need something to translate official gnome's po files.
[01:19] <AlinuxOS> http://www.gia.ge/dormidon/main.php?root=476 as I have listed here.
[01:19] <danilos> AlinuxOS: ah well ;)
[01:19] <carlos> AlinuxOS: we plan to do automatic imports for the whole GNOME
[01:19] <danilos> AlinuxOS: you can't have it all right now :)
[01:19] <carlos> so that would solve your problem, but we don't have a date yet
[01:20] <AlinuxOS> so there will be Offical GNOME translation project teams ?
[01:20] <AlinuxOS> I've one, it was created by previous coordinator ("coordinator") but I see no needs :/
[01:21] <carlos> AlinuxOS: yeah
[01:21] <carlos> with closed permissions so only official translators can do changes there
[01:23] <AlinuxOS> so you are tellng me that rosetta becomes something official for Mother GNOME translation...and there will be official package lists http://l10n-status.gnome.org/gnome-2.16/ka/index.html like here and that .po files will be translatable...right ?
[01:24] <AlinuxOS> carlos, will Ubuntu translator, Official GNOME translator to ? 
[01:24] <AlinuxOS> too..
[01:26] <carlos> AlinuxOS: well, it will not be official for GNOME
[01:26] <danilos> AlinuxOS: no, Gnome won't be officially translated using Rosetta, but Rosetta will allow someone to have a team to translate official (rather, upstream) Gnome
[01:26] <carlos> we will set up it so any official team is able to use Rosetta as another tool to translate
[01:26] <carlos> final work should be committed manually by translators
[01:26] <carlos> we will 'suck' .po and .pot files from GNOME
[01:26] <AlinuxOS> ah
[01:27] <carlos> like from KDE and others
[01:27] <AlinuxOS> so it will constantly merged with official gnome right ? :)
[01:27] <carlos> right
[01:27] <carlos> the other way should be handled still by translators
[01:27] <carlos> and to prevent forks for that /products/ URL, we will require you to be an official member in GNOME
[01:28] <AlinuxOS> carlos, just anoher small question...Entrans have a great feature that permits automatic suggestion/translation for standat most common strings.
[01:28] <carlos> that's a glossary/translation memory
[01:28] <AlinuxOS> example edit,file,Do you really want... etc ecc...
[01:28] <AlinuxOS> carlos, right.
[01:28] <AlinuxOS> that one...
[01:28] <carlos> we have something like that, but we never will set a new translation based on what we have
[01:28] <carlos> automatic translations suck
[01:29] <carlos> glossary support is still pending to be implemented
[01:29] <carlos> so we require someone to accept such 'automatic' translation
[01:29] <AlinuxOS> carlos, yes I know but Entrans only gives you an automatic solution...you only need to correct it if it's not 100% good translated..and then commit.
[01:29] <carlos> then it's the same thing here
[01:30] <carlos> you need someone to 'commit'/approve it
[01:30] <carlos> we show you it, you need to accept it or 'fix' it
[01:30] <carlos> but it will never be used unless someone approves it first
[01:31] <AlinuxOS> carlos, and how about, if I'm translating GNOME 2.16 branch... and I have _File string in 7-8 different .po files ? Can I submit that string in more that one .po file ?
[01:31] <carlos> planned feature
[01:32] <carlos> if you have the needed rights, you will be able to do that
[01:32] <AlinuxOS> :) oh...
[01:32] <carlos> we want to finish the review UI
[01:32] <carlos> and once that's done, revisit the spec about that and implement it
[01:33] <AlinuxOS> carlos, what's the difference between GNOME team coordinator and Ubuntu Team Owner ? I'm coordinator for GNOME but Admin in Ubuntu Team...it's ok ?
[01:34] <carlos> AlinuxOS: this is a draft we wrote about the feature: https://wiki.launchpad.canonical.com/MulticastTranslations
[01:34] <AlinuxOS> I've noticed that Owners commits have more power then admins commits.
[01:34] <carlos> AlinuxOS: hmm being an owner or an admin is more or less the same for Rosetta, in fact is the same as being member of the team
[01:35] <AlinuxOS> I mean that some old owners translations appeared in Edgy...and it was double work for me.
[01:35] <carlos> AlinuxOS: well, when we open Edgy, we copied whatever was in Dapper
[01:35] <carlos> to reuse that work
[01:36] <AlinuxOS> carlos, yes of course... but why in Edgy Eft case rosetta prefered owners old translations then my new reviewed strings ?
[01:37] <carlos> AlinuxOS: where and how did you review your strings?
[01:37] <AlinuxOS> just sec
[01:40] <AlinuxOS> https://launchpad.net/distros/ubuntu/edgy/+source/gnome-desktop/+pots/gnome-desktop-2.0/ka/+translate , here for gnome-desktop I was able to re-confirm my translations ..against owners old translations.
[01:41] <AlinuxOS> carlos, I've changed them yesterday.
[01:41] <AlinuxOS> hm..today :)
[01:42] <carlos> AlinuxOS: hmmm
[01:42] <carlos> dapper has the ones from Alexander as the used ones
[01:42] <carlos> AlinuxOS: so we copied them as they were there
[01:43] <carlos> AlinuxOS: did you use the web forms to do that change? or the .po upload form?
[01:43] <AlinuxOS> most frequently .po but sometimes web.
[01:44] <AlinuxOS> bah,...mistery...
[01:44] <carlos> AlinuxOS: did you select 'Published upload' ?
[01:44] <AlinuxOS> hm...what's this ? :D
[01:44] <carlos> https://launchpad.net/distros/ubuntu/dapper/+source/gnome-desktop/+pots/gnome-desktop-2.0/ka/+upload
[01:45] <carlos> There are two options: 'User upload' or 'Published upload'
[01:45] <AlinuxOS> looolz
[01:45] <AlinuxOS> I use User upload sooo :D
[01:45] <carlos> hmm
[01:45] <carlos> then
[01:45] <carlos> it's either a bug or a 'fight' between you and Alexander...
[01:46] <carlos> AlinuxOS: did you manage to see your changes applied?
[01:46] <AlinuxOS> I haven't noticed that 2 options :)
[01:46] <AlinuxOS> poor me :D
[01:46] <AlinuxOS> carlos, don't understand you ? manage too see ?
[01:47] <carlos> AlinuxOS: after you did the upload
[01:47] <AlinuxOS> sorry :) I'm not very strong !
[01:47] <AlinuxOS> ah
[01:47] <carlos> did you go to that form and saw your strings as the ones by default?
[01:47] <AlinuxOS> no normally(for Dapper everything worked great)
[01:47] <carlos> AlinuxOS: in fact, you should get a confirmation email
[01:47] <carlos> did you got it?
[01:48] <carlos> AlinuxOS: it should work the same for Edgy, we didn't change anything related to that code
[01:48] <AlinuxOS> carlos, I know that I must recive a confermation mail...but I've upgrated to Edgy yesterday night...so there was immediatly some strange strings in interface
[01:49] <AlinuxOS> and I've found out that thoose stringes belong to Alexander....that has no more time to translate...he is out for the moment.
[01:49] <carlos> AlinuxOS: well, at least for the link you gave me
[01:50] <carlos> dapper seems to have the same strings: https://launchpad.net/distros/ubuntu/dapper/+source/gnome-desktop/+pots/gnome-desktop-2.0/ka/+translate
[01:50] <carlos> so I'm confused, if you say that those are correct in dapper...
[01:50] <AlinuxOS> carlos, ok...it's not a tragedy...I was simply interested whats happened :)
[01:51] <carlos> well, me too, if we have a bug there... I want to know it :-)
[01:51] <AlinuxOS> ah
[01:51] <AlinuxOS> carlos, so ... my strins that apear there...are from GNOME project...
[01:52] <carlos> I guess
[01:52] <AlinuxOS> but system gives more importance to Alexander-s translations.
[01:52] <carlos> but if they come from upstream, they don't take preference over the ones in Ubuntu
[01:52] <carlos> right
[01:52] <AlinuxOS> ah
[01:52] <carlos> I understand now the problem
[01:52] <AlinuxOS> ok :D
[01:52] <AlinuxOS> me too :)
[01:52] <carlos> Ubuntu has always preference
[01:52] <AlinuxOS> hehe
[01:52] <carlos> :-)
[01:53] <AlinuxOS> and I was thinking that GNOME has preference :)
[01:53] <AlinuxOS> so I was thinking about it entire morning :)
[01:53] <AlinuxOS> carlos, thank you for your time :) And sorry... I'm not so professional as you guys here :)
[01:54] <AlinuxOS> danilos, thank you for your time too :)
[01:54] <danilos> AlinuxOS: no problem ;)
[01:55] <AlinuxOS> danilos, carlos maybe sometime somewhere I'll meet you all :D
[01:55] <danilos> AlinuxOS: plan for next GUADEC, I'll be there ;)
[01:55] <AlinuxOS> So you will have a opportunity to know some Egsotic People from georgia :)
[01:56] <AlinuxOS> people = me :)
[01:56] <danilos> sure :)
[01:56] <carlos> AlinuxOS: you are welcome, and thanks for your time too ;-)
[01:57] <AlinuxOS> I hope it will be near..somewhere in europe :) As student, I'm not rich :)
[01:57] <AlinuxOS> so if it's in Burkina Faso...or me it's a big problem :)
[01:59] <AlinuxOS> hehe carlos copy copy from Entrans (good features of course)Rosetta must be perfect!! :D...it's open source :D
[02:00] <AlinuxOS> danilos, I'm really happy that you are working with this great people ;) I hope you enjoy Canonical :)
[02:00] <AlinuxOS> Danilovo Blogchetanje :)
[02:00] <danilos> AlinuxOS: yeah, I am, thanks ;) btw, GUADEC is in Birmingham next year, July I think
[02:00] <danilos> AlinuxOS: :)
[02:01] <AlinuxOS> danilos, is there some sponsorship for students ?
[02:02] <AlinuxOS> I can sponsor my air tickets... but I don't know nothing about English hotels and prices :)
[02:03] <danilos> AlinuxOS: there is some sponsorship for anyone involved with Gnome, not sure how much will there be for next year
[02:03] <danilos> AlinuxOS: as for stay, we usually get some cheap hostels or something
[02:03] <AlinuxOS> and It will be great to stay with you people! ;) I hope that I'll meet you all there :)
[02:04] <AlinuxOS> danilos, I feel myself involved into Gnome :D
[02:05] <danilos> AlinuxOS: of course you are :)
[02:05] <AlinuxOS> danilos, hehe if you like wine :) I can get for you some great Tuscanian wines! :)
[02:05] <AlinuxOS> carlos, for you too pal! :)
[02:06] <danilos> AlinuxOS: sure, we'll see about that ;)
[02:06] <AlinuxOS> hehe
[02:06] <AlinuxOS> danilos, ok...
[02:06] <AlinuxOS> so In my plans there will be GUADEC in Birmingham too for next year! :)
[02:08] <AlinuxOS> carlos, not hopes but you must be there ...If won't..so no Red Wine ! :D
[02:09] <AlinuxOS> If I'll be in GUADEC, I should meet some Mythologic people/developers from GNOME/Ubuntu/Canonical :)
[02:09] <carlos> AlinuxOS: Well, this time I will not have any university exam, so I should be able to attend...
[02:09] <AlinuxOS> it's root of my happynes :D
[02:10] <AlinuxOS> I hope for me too...
[02:11] <danilos> AlinuxOS: yeah, it was moved even later for 2007 I think :)
[02:13] <AlinuxOS> hehe
[02:13] <AlinuxOS> ;)
[02:13] <AlinuxOS> ok see you people! ;) have a nice day.
[02:14] <carlos> see you later
[02:15] <Ubugtu> New bug: #60556 in rosetta "Missing ubuntu-l10n-ms group in ubuntu-translators" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/60556
[03:39] <neutrinomass> Out of curiosity: Why does bug 2 not exist ?
[03:40] <Ubugtu> New bug: #60566 in launchpad "Register new account form needs better input validation for invalid syntax emails." [Low,Confirmed]  http://launchpad.net/bugs/60566
[03:48] <SteveA> neutrinomass: no idea.  there is no record of bug number 2 in the database.
[03:49] <SteveA> you can try asking on the launchpad-users mailing list.  maybe someone else would know
[03:54] <neutrinomass> SteveA: Thanks. It's not blocking my work or anything, so I'd rather not disturb as much. It just surprised me :)
[03:54] <BjornT> neutrinomass: it's because we use a transactional database; every new bug get a unique id from a database sequence. If however a transaction (that already asked for a bug id) get aborted for some reason, that id will remain unused. the sequence can't be reset since another transaction might already have increased its value.
[03:55] <Ubugtu> New bug: #60568 in launchpad "Cannot import OpenPGP key." [Untriaged,Unconfirmed]  http://launchpad.net/bugs/60568
[03:56] <SteveA> BjornT: that's a reason why a given bug number might not be in the database.  but I guess bug 2 was excluded for some other reason, as it is #2
[03:59] <neutrinomass> BjornT: That sort of explains it :) Thanks
[04:12] <kiko> HELLO THERE CANADA
[04:13] <kiko> I am two steps behind you francis
[04:15] <flacoste> aah, that explains that eery feeling of presence if have been feeling all this time
[04:16] <kiko> you know it
[04:22] <kiko> flacoste, many thanks for researching malcc's problem yesterday
[04:22] <kiko> malcc: good news eh?
[04:22] <flacoste> kiko: my pleasure
[04:23] <malcc> kiko, flacoste: Yes, excellent. Unfortunately my first attempt to land it since hit other errors, I'm trying again in the hope that they're random
[04:23] <kiko> malcc, can you forward those errors as well?
[04:23] <kiko> random errors should not happen.
[04:23] <malcc> kiko: Sure
[04:30] <Ubugtu> New bug: #60574 in malone "Comments/Audit trail does not show multiple attachments" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/60574
[04:37] <kiko> carlos, do you know why xx-rosetta-potemplate-export is disabled?
[04:38] <carlos> hmm, I don't remember any reason to have it disabled
[04:38] <carlos> any XXX comment?
[04:38] <kiko> BjornT, is there a way, using the new pagetest system, to say exactly what form elements you want to send off?
[04:38] <kiko> carlos, nope.
[04:38] <carlos> try to enable it 
[04:38] <kiko> I did
[04:38] <carlos> if it fails tell me it and I will fix it
[04:38] <kiko> it fails but I believe I can fix it
[04:39] <carlos> ok
[04:39] <kiko> no need, i"m already fixing it and that oops we saw in yesterday's report
[04:39] <kiko> related to an export with no "format" message provided
[04:39] <kiko> https://devpad.canonical.com/~jamesh/oops.cgi/2006-09-14/D569
[04:40] <carlos> kiko: I think that error came from mvo playing with curl to do uploads/exports
[04:40] <kiko> carlos, they do. but I think it's nice to fix that anyway, since it's a page which more people may want to automate.
[04:40] <j-a-meinel> Is the sftp => http mirroring script for launchpad stalled again?
[04:40] <carlos> at least, I know he did it to upload tarballs. From that error, I guess he tried to do it also for the exports
[04:41] <kiko> malcc, so make_pidfile and test_gettarget failed?
[04:41] <salgado> the make_pidfile problem I reported some time ago
[04:41] <salgado> and spiv suggested a fix, but I didn't have time to try it out
[04:41] <carlos> kiko: well... we send an email, we don't give any URL to download anything, so I don't think it's so usual...
[04:41] <malcc> kiko: Yes
[04:41] <carlos> kiko: but if it's easy to give a good error, that's also fine
[04:42] <BjornT> kiko: no. if you want test such things you probably should do it as a normal doctest, not as a pagetest.
[04:42] <salgado> malcc, kiko, there's a thread with subject "PQM failure on canonical.pidfile" on the launchpad@ list which has a fix for the make_pidfile problem
[04:42] <malcc> salgado: Thanks, looking
[04:44] <malcc> Ok, I'm guessing I got randomly hit by that, but that the other one, which is a test layer isolation failure, is probably down to my branch
[04:44] <kiko> carlos, it's 2 lines
[04:44] <kiko> j-a-meinel, I don't know how to check :-/
[04:45] <kiko> matsubara, I'm finishing off the fix for yesterday's topcrasher, btw.
[04:45] <j-a-meinel> kiko: Thanks for responding. Do you know who I could ping? I would usually ping someone like spiv/lifeless, but they should be sleeping right now.
[04:45] <kiko> j-a-meinel, ddaa may be able to at least guide us in the right direction. I'd like to be able to help you but I'm a bit clueless, will need some pointing
[04:46] <j-a-meinel> Well, all I really know is that there is a script which does a 'bzr pull' from the sftp upload area to the area where branches are served by http.
[04:46] <j-a-meinel> In the past, it also was the one responsible for mirroring from external http sites.
[04:47] <j-a-meinel> Which meant that a bad remote site would slow everything down.
[04:47] <matsubara> kiko: thanks
[04:48] <malcc> elmo: Ping
[04:51] <kiko> malcc, I'll try fixing the make_pidfile crash meanwhile, since nobody seems to have taken it on.
[04:51] <malcc> kiko: Cool
[05:00] <iwj> Maybe I should ask my bzr/LP question here.  #bzr seemed to think that there was a problem with my branch registration.
[05:00] <iwj> So, I was wondering if anyone would be able to help me with this symptom:
[05:00] <iwj> -anarres:autodebtest> bzr push --create-prefix sftp://ijackson@bazaar.launchpad.net/~ubuntu-core-dev/autopkgtest/ubuntu
[05:00] <iwj> bzr: ERROR: exceptions.IOError: ubuntu
[05:00] <iwj>   at /usr/lib/python2.4/site-packages/bzrlib/transport/sftp.py line 540
[05:00] <iwj>   in _translate_io_exception
[05:02] <LarstiQ> iwj: I haven't looked at it closely enough, but does it work if you first `bzr init sftp://..` and then `bzr push sftp://..`?
[05:03] <iwj> bzr init fails with ENOENT too.
[05:03] <iwj> Strangely quickly.
[05:04] <LarstiQ> what version of bzr is that?
[05:04] <j-a-meinel> LarstiQ: No paramiko?
[05:04] <iwj> I tried bzr init in the wrong window, so I got Debian testing's.
[05:04] <iwj> The edgy one worked.
[05:04] <LarstiQ> j-a-meinel: the bzr.log mentioned sftp transport being used
[05:04] <j-a-meinel> K
[05:05] <LarstiQ> testing is presumably at 0.8.2 still then
[05:05] <LarstiQ> iwj: did just the init work, or the push also?
[05:06] <iwj> I'm trying the push again now.  It's a bit slow.
[05:06] <iwj> Yay!
[05:06] <LarstiQ> No further questions, your honor.
[05:06] <iwj> Thanks.  I'll update the wiki instructions.
[05:06] <iwj> There was no .bzr at that location to start with.
[05:07] <LarstiQ> oh right
[05:07] <iwj> I was following the instructions at https://wiki.ubuntu.com/BzrMaintainerHowto
[05:07] <LarstiQ> but push --create-prefix also didn't want to work? *boggle*
[05:07] <Keybuk> we need a duckie to unfreeze edgy ;)
[05:07] <LarstiQ> iwj: mind you, doing the explicit init is a workaround
[05:08] <j-a-meinel> The explicit init should only work for bzr >= 0.9
[05:08] <j-a-meinel> I'm pretty sure it never worked for 0.8.2.
[05:08] <LarstiQ> correct
[05:08] <jamesh> --create-prefix shouldn't ever be needed: the ~user/product directories magically exist
[05:08] <iwj> Yes, we discovered that.
[05:08] <LarstiQ> that change only made it into 0.9
[05:08] <iwj> jamesh: Without --create-prefix it failed earlier.
[05:09] <LarstiQ> jamesh: any idea what caused iwjs problem?
[05:09] <jamesh> nope
[05:09] <iwj> Is anyone on the case to get 0.9 into etch, jooi ?
[05:09] <iwj> Or should I ask that in #bzr ? :-)
[05:10] <LarstiQ> iwj: just a matter of testing propagation
[05:11] <iwj> LarstiQ: Ah, excellent.
[05:11] <LarstiQ> iwj: then again, with 0.10 recently uploaded and 0.11 close, it needs active effort not to reset the delay
[05:11] <flacoste> malcc: yeah, the IsolationError is not random
[05:13] <malcc> flacoste: Can you shed any light? My brain has decided it's done with thinking for this week...
[05:13] <flacoste> malcc: i'm on it
[05:22] <ddaa> j-a-meinel: sorry, I was deep in paperstuff, what's the status of your branch puller problem?
[05:25] <ddaa> j-a-meinel: previously, the cause of the problem was a network outage in the DC
[05:28] <kiko> matsubara, does a new bzr push take like 6 minutes for you?
[05:29] <ddaa> j-a-meinel: it does not appear to be the same problem now
[05:30] <matsubara> kiko: didn't test it yet
[05:30] <kiko> matsubara, I mean, for a new branch
[05:31] <kiko> salgado, matsubara: OOPS-257A1251 is known, right? is it hard to fix?
[05:31] <Ubugtu> https://devpad.canonical.com/~jamesh/oops.cgi/257A1251
[05:32] <matsubara> kiko: yes, it's know and it's assigned to BjornT 
[05:32] <kiko> ah, okay.
[05:32] <matsubara> s/know/known/
[05:33] <ddaa> j-a-meinel: woohoo, I got supermirror admin privs!
[05:36] <kiko> carlos, can you check out https://sodium.ubuntu.com/~andrew/paste/fileGVh13k.html -- ?
[05:39] <carlos> kiko: well... my only comment on that would be to stop using our own solution to show error messages and use the standard notification system
[05:40] <kiko> carlos, I'm not going to do that as part of this fix. kthxbye :-P
[05:40] <carlos> kiko: other than that, it looks fine
[05:40] <carlos> kiko: ok ;-)
[05:40] <kiko> thanks
[05:41] <ddaa> j-a-meinel: siginted the branch pullers that were running, and increased the verbosity of the cronscript, that might help figure out what's going on
[05:41] <kiko> salgado, matsubara: do any of you have a clue about OOPS-257A20 -- ?
[05:41] <Ubugtu> https://devpad.canonical.com/~jamesh/oops.cgi/257A20
[05:42] <flacoste> stub: ping
[05:43] <salgado> another known bug, kiko 
[05:43] <matsubara> kiko: bug 57759
[05:43] <kiko> salgado, what's it about?
[05:43] <Ubugtu> Malone bug 57759 in launchpad "Accessing +mergerequest-sent page after successfully completing the merging process crashes." [Low,Confirmed]  http://launchpad.net/bugs/57759
[05:43] <salgado> people hitting the back button, probably
[05:43] <kiko> matsubara, that bug title could be updated -- it doesn't say why :)
[05:44] <kiko> oh
[05:44] <kiko> now I read it again
[05:48] <kiko> salgado, can that count() not return more than 1 ever? if not, why not?
[05:48] <salgado> it may return 0
[05:48] <salgado> when the account is merged
[05:48] <kiko> agreed
[05:48] <kiko> but
[05:48] <kiko> what if the account that is being merged 
[05:48] <kiko> has more than one email address associated to it?
[05:49] <kiko> is that impossible?
[05:49] <salgado> yes, but that takes you to a separate page
[05:50] <kiko> oh?
[05:50] <Ubugtu> New bug: #60592 in malone "Sept 15 Dapper updates wiped audio drivers" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/60592
[05:52] <kiko> matsubara, you're too fast!
[05:53] <matsubara> for some strange reason I can't add attachments to bugs in launchpad.dev
[05:53] <matsubara>     *  Module canonical.librarian.client, line 138, in addFile
[05:53] <matsubara>       raise UploadFailed, 'Server said: ' + response
[05:53] <matsubara> UploadFailed: Server said: 500 Internal server error] 
[05:56] <flacoste> malcc: i have found the cause of the IsolationError, but I'm now tripping over another Unauthorized one in LaunchpadZopelessLayer
[05:57] <malcc> flacoste: Looks like I've pulled out a vital twig with my new test, and the whole tree of launchpad testing is collapsing! :)
[05:57] <BjornT> matsubara: try removing /var/tmp/fatsam
[05:59] <kiko> malcc, poetic injustice
[06:00] <matsubara> BjornT: danilos told me that on pm. it worked! thanks
[06:01] <flacoste> boy, those supermirror test are slow...
[06:04] <matsubara> kiko: have you seen this bug 60574?
[06:04] <Ubugtu> Malone bug 60574 in malone "Comments/Audit trail does not show multiple attachments" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/60574
[06:04] <matsubara> kiko: I just confirmed it. isn't it related to the code you landed to hide comments?
[06:04] <kiko> matsubara, no. it sounds scary. :)
[06:06] <j-a-meinel> ddaa: thanks
[06:07] <ddaa> mh, it's not clear that the increased verbosity is going to help...
[06:08] <ddaa> anyway, now I'm in a position to actually handle this sort of problem
[06:10] <kiko> matsubara, a few of the 404s are trivial image renames that need to be fixed
[06:13] <matsubara> kiko: I think they're fixed already
[06:14] <kiko> matsubara, the /padlock reference in bugtask-edit-form?
[06:15] <kiko> matsubara, hmmmm. yes, it is related, good catch.
[06:15] <matsubara> hmm where did you find that padlock one, kiko? 
[06:16] <kiko> lib/canonical/launchpad/templates/bugtask-edit-form.pt
[06:16] <kiko> matsubara, I was looking at my zope output when running
[06:18] <matsubara> kiko: hmm it appears on staging oops report. I'll fix it today.
[06:18] <kiko> matsubara, I've already fixed that one
[06:18] <kiko> there may be others though
[06:18] <j-a-meinel> ddaa: The branch ~bzr/bzr/bzr.urllib.keepalive is the one that I'm aware of, which shows that sftp is not replicating to http. But I don't know a whole lot more than that.
[06:21] <ddaa> j-a-meinel: is that still out of date?
[06:21] <j-a-meinel> I just checked 10s ago.
[06:22] <ddaa> weird
[06:22] <ddaa> it looks like -v or -vv on that script have no effect whatsoever
[06:25] <matsubara> kiko: does your patch in pqm's queue fix OOPS-256D759?
[06:25] <Ubugtu> https://devpad.canonical.com/~jamesh/oops.cgi/256D759
[06:25] <Ubugtu> New bug: #60601 in launchpad "Add an ssh key should allow adding a key without comments" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/60601
[06:26] <kiko> matsubara, hmmm, nope.
[06:27] <ddaa> why is it that everytime I try to do the paperwork, an emergency of some sort arises?
[06:28] <kiko> matsubara, I'll sort it out too
[06:28] <matsubara> kiko: thank you
[06:28] <kiko> matsubara, is there a bug for that? in what oops report did it appear?
[06:29] <j-a-meinel> ddaa: I don't know that this is a pure emergency, more than just an oddity. I can still get access to the branch over sftp, which is enough for me right now.
[06:29] <matsubara> kiko: 2006-09-13
[06:30] <ddaa> it is a service failure of the sort that seriously harms the confidence of users into launchpad hosting
[06:30] <matsubara> kiko: there's no bug for it but I can report if you need it.
[06:30] <ddaa> j-a-meinel: it is an emergency
[06:30] <kiko> matsubara, no, that's fine, it will go in today as well. it's a very easy fix.
[06:31] <j-a-meinel> ddaa: Well, let me know if there is anything I can do. I probably don't have access rights to those machines, but if I do, I'm willing to help out.
[06:31] <ddaa> j-a-meinel: the last time that happened, I had users complaining of "launchpad not mirroring again" for one week after the service was fixed, because that became their initial assumption
[06:31] <j-a-meinel> ddaa: good point.
[06:31] <j-a-meinel> Once the system starts failing, people assume it is its fault, rather than their own.
[06:31] <kiko> j-a-meinel, you can probably get access if you talk it over with rob and david
[06:32] <j-a-meinel> At this point, I don't know much about the Launchpad internals. So I'm not sure I'd be a lot of help. But I am willing to help with bazaar related stuff.
[06:32] <j-a-meinel> (I'm *willing* to help with other stuff, but I probably shouldn't get swamped with it :)
[06:33] <ddaa> j-a-meinel: if you could hit the launchpad code for that script (the entry point is cronscripts/supermirror-pull.py) and sprinkle it with warning, progress and debug logging, that would help me.
[06:33] <j-a-meinel> ddaa: Do I even have access to Launchpad code? I'd certainly be willing to do so.
[06:34] <kiko> j-a-meinel, you should have -- do you have an account on sodium?
[06:34] <ddaa> j-a-meinel: also, if you could get robey to fix paramiko so it does not give EOFError for just any sort of connection problem (timeout, key verification failed, could not route to host, etc.) that would help.
[06:34] <j-a-meinel> I don't think it is an issue of me being specifically forbidden to access it. (I have signed the NDA, etc)
[06:34] <kiko> j-a-meinel, no, on the contrary, it should be absolutely fine
[06:34] <j-a-meinel> ddaa: Are you wanting paramiko to give specific errors. Or just not raise EOF?
[06:35] <ddaa> j-a-meinel: I want some more specific error (a general paramiko.ConnectionError would be fine) so the puller script can catch it.
[06:36] <ddaa> EOF is just not specific enough for that
[06:36] <j-a-meinel> kiko: I don't seem to have a 'sodium' account. At least it isn't letting me in using ssh keys to login.
[06:36] <ddaa> That would also help the bzr UI too...
[06:36] <j-a-meinel> AFAIK, I only have chinstrap, escudero, and orcadas accounts.
[06:36] <kiko> j-a-meinel, so ssh devpad.canonical.com doesn't work?
[06:37] <kiko> j-a-meinel, I'll file an RT request for you, one moment. what's your chinstrap username?
[06:37] <j-a-meinel> ddaa: I agree. I can look into it a little (I've at least paged through the paramiko code in the past)
[06:37] <j-a-meinel> kiko: 'jameinel'
[06:38] <ddaa> wow!
[06:38] <ddaa> got it!
[06:38] <kiko> j-a-meinel, filed request, I'll ping you when it's done.
[06:38] <ddaa> mh, maybe not, but accepting the rsa key for bazaar.launchpad.net cannot hurt
[06:38] <j-a-meinel> thanks
[06:39] <ddaa> j-a-meinel: your sftp branch _should_ be mirrored now
[06:40] <j-a-meinel> ddaa: nope
[06:40] <j-a-meinel> at least not yet
[06:41] <j-a-meinel> bzr log http://bazaar.launchpad.net/~bzr/bzr/bzr.urllib.keepalive/ should be at revno 2010
[06:41] <ddaa> thanks
[06:41] <j-a-meinel> This branch was recently renamed.
[06:41] <j-a-meinel> If that is an issue.
[06:41] <ddaa> haha
[06:41] <ddaa> yes, it is pertinent, but should not be an issue
[06:41] <j-a-meinel> But it was renamed *before* revno 2007
[06:41] <j-a-meinel> So an earlier push has already worked with a renamed branch.
[06:42] <ddaa> So, the renaming was handled fin.
[06:43] <ddaa> New branches and renamed branches need to wait for the supermirror_rewritemap.py script to run before they are accessible
[06:43] <matsubara> kiko: bug 2497 is currently assigned to Stuart, should I assign it to you?
[06:43] <Ubugtu> Malone bug 2497 in rosetta "/people/*/+translations times out for prolific translators" [Critical,Confirmed]  http://launchpad.net/bugs/2497
[06:43] <kiko> matsubara, hmmm, not yet, wait for him to finish the tests.
[06:43] <flacoste> malcc: i got it working!
[06:43] <ddaa> one issues is that I do not even know which user runs this script :(
[06:43] <ddaa> j-a-meinel: in that case, the branch is accessible, so it's not a rewritemap issue
[06:43] <kiko> flacoste!!!!11
[06:44] <malcc> flacoste: You da man. What was wrong?
[06:44] <flacoste> well, the problem was that ImportdTestCase assumed (rightly) that it had an ANONYMOUS interaction setup but some tests before it removed it
[06:45] <flacoste> the Unauthorized error after that was just a bad fix on my part in the definition of support-tracker-emailinterface (i had used LaunchpadZopelessLayer instead of LaunchpadFunctionalLayer)
[06:45] <flacoste> i'm sending a message to the list with my diff against your tree yesterday
[06:45] <ddaa> flacoste: importd test cases are a little hell of their own, they should become real launchpad tests one day
[06:46] <kiko> flacoste, can we avoid this sort of problem occurring in the future?
[06:46] <kiko> flacoste, we shouldn't let one test fuck us all over
[06:46] <flacoste> kiko: i left a check in the ZopelessLayer
[06:46] <kiko> you are a master
[06:46] <flacoste> kiko: but it raises an LayerIsolationError - so aborts the test run
[06:46] <flacoste> kiko: i could instead change it into a warning and recover
[06:47] <flacoste> kiko: but we are probably going to ignore that warning since nobody runs the whole test suite on his desk it seems
[06:47] <kiko> flacoste, I think a crash is great. 
[06:47] <flacoste> kiko: actually, this is for the original error Malcolm got
[06:48] <kiko> I've lost track by now you realize
[06:48] <flacoste> kiko: you'll read my mail then :-)
[06:48] <mez> er - why can I not link a bug to an external tracker?
[06:48] <kiko-fud> mez, you certainly can. what's up?
[06:49] <mez>  kiko-fud - theres a bug on katapult, that is also filed in KDE's bugtreakcer - but I cant link it to that bug
[06:49] <kiko-fud> mez, what happens?
[06:49] <mez> kiko - firt of all I cant find the option to link it
[06:50] <kiko-fud> mez, Also Affects: +Upstream
[06:50] <kiko-fud> mdz, and I don't blame you.
[06:50] <mez> kiko-fud: it is FILED in upstream
[06:50] <mez> https://launchpad.net/products/katapult/+bug/60136
[06:50] <Ubugtu> Malone bug 60136 in katapult "Katapult doesn't work with Amarok >= 1.4.2" [Critical,Confirmed]  
[06:51] <mez> We use Malone as the main bugtracker - but people also post bugs elsewhere
[06:51] <kiko-fud> mez, I see!
[06:51] <mez> kde bug 133030
[06:51] <Ubugtu> KDE bug 133030 in general "Katapult exporting malformed url to Amarok" [Normal,New]  http://bugs.kde.org/show_bug.cgi?id=133030
[06:51] <kiko-fud> that's a very interesting case :)
[06:51] <kiko-fud> wow
[06:52] <mez> I guess it's just me being awkward again and refusing to use KDE's stuff cause I like LP better and me just being a PITA as usual
[06:52] <kiko-fud> mez, you have left me gasping for air now because I have never considered your use case
[06:52] <kiko-fud> mez, the way watches work is 
[06:53] <kiko-fud> if you don't use launchpad officially, you can link to an upstream bug watch and get status syncing.
[06:53] <kiko-fud> if you /do/ use launchpad officially we never took into account the fact that you'd want to sync status from abroad
[06:53] <kiko-fud> mez, here's a question: when you do fix the bug, will you close it in the KDE bugzilla, or in launchpad first?
[06:54] <mez> Lp
[06:54] <kiko-fud> heh
[06:54] <kiko-fud> so I /can/ propose an ugly workaround
[06:54] <kiko-fud> two even
[06:54] <kiko-fud> workaround 1 would be add just a URL to the remote watch; grabbing its status wouldn't be very useful anyway given the above.
[06:55] <kiko-fud> err remote bug
[06:55] <kiko-fud> in a comment
[06:55] <kiko-fud> workaround 2 would be add a task for another hopefully related upstream that does /not/ use LP and add a watch to /that/
[06:55] <kiko-fud> that's a big hack though and you didn't hear it from me
[06:55] <kiko-fud> la la la
[06:55] <mez> ;)
[06:56] <mez> It'd be nice just to be able to say "bug reported "here" too
[06:56] <mdz> kiko-fud: I should hope you don't blame me; I wasn't even part of the conversation
[06:56] <mez> i thought that was waht malone was meant to be ;)
[06:56] <kiko-fud> mez, it is, it is. just that it's tricky to allow everybody to do everything
[06:56] <kiko-fud> mdz, when did I last blame you for anything?
[06:56] <mez>  kiko-fud> mdz, and I don't blame you.
[06:57] <kiko-fud> I have that sequence on my fingers
[06:57] <mdz> kiko-fud: as mez points out, I was having fun with you
[06:57] <mdz> kiko-fud: but to answer your question, "yesterday"
[06:57] <kiko-fud> I didn't do that /that/ explicitly 
[06:57] <kiko-fud> but I GIVE UP :-P
[06:58] <mez> kiko: would it not be possible to set up the thing so it's this bug is reportyed here rather than "this instance of this bug is in this bugtracker"
[06:58] <mez> or to add URLS to it
[07:01] <ddaa> j-a-meinel: sorry, I will not have time to deal with that problem today
[07:01] <j-a-meinel> np
[07:01] <j-a-meinel> We're all busy.
[07:02] <j-a-meinel> You were just the only person I could track down who could work on it at all :)
[07:02] <j-a-meinel> Are you probably the best person right now to look into this sort of thing?
[07:02] <j-a-meinel> Or should I be trying to ping someone else?
[07:02] <ddaa> I'm supposedly where the buck of supermirror things stops.
[07:02] <flacoste> malcc, kiko: my fix is in the patch sent to the list Cc: malcolm
[07:03] <ddaa> Though there are a lot of things which are effectively maintained by spiv and jamesh.
[07:03] <j-a-meinel> ddaa: Is there a page like 'pqm.ubuntu.com' that shows the pqm status for the "other project(s)"?
[07:04] <ddaa> pqm.launchpad.net shows the launchpad-related stuff
[07:04] <j-a-meinel> It would be nice to check it from time to time, when it looks like things have stalled.
[07:04] <j-a-meinel> ddaa: that's what I was looking for. Thanks.
[07:05] <ddaa> I've landed a bunch of arch and other cruft removal branches today, so it can explain pqm being busier than usual
[07:06] <j-a-meinel> np
[07:06] <j-a-meinel> Its more just that the LP test suite takes about 1hr to finish. So it is hard to tell if it is stalled, or just halfway through.
[07:08] <jordi> danilos: won't have time to finish up that KDE email.
[07:08] <jordi> danilos: I've been working through the import queue, there's a lot of cruft in there that needs to be investigated
[07:08] <jordi> it's down to just 40 entries now, tho
[07:08] <danilos> jordi: can you forward what you've got to me, and I'll fill in what needs filling, and ask you for a review before sending it, is that ok with you?
[07:09] <jordi> do you plan to do it now?
[07:09] <jordi> because I could attempt again tomorrow
[07:10] <jordi> danilos: btw, can you see if https://launchpad.net/bugs/32471 and https://launchpad.net/bugs/60029 merit some severity/priority triaging?
[07:10] <Ubugtu> Malone bug 32471 in rosetta "display differences from upstream" [Wishlist,Confirmed]  
[07:10] <jordi> people in the list are quite interested in these
[07:12] <jordi> danilos: also, input from you in rosetta-email's 20060913163134.GA30387@nubol.oskuro.net would be cool
[07:12] <danilos> jordi: I know about them, but what doesn't happen with translationreview carlos is doing is not really high priority
[07:12] <danilos> jordi: we've got more important things right now; I'll just confirm the unconfirmed one
[07:12] <jordi> danilos: ok
[07:14] <jordi> I'll be back
[09:31] <Ubugtu> New bug: #60617 in launchpad "PersonVocabulary crashes when it receives a list" [Low,Confirmed]  http://launchpad.net/bugs/60617
[10:43] <kiko-fud> BjornT, around?
[10:47] <matsubara> kiko-fud: looks like it's on gangotri
[10:52] <kiko-fud> matsubara, yeah, foundit
[10:54] <flacoste> kiko-fud: What do you think of my 'Solved' proposition to replace 'Answered and confirmed'?
[10:54] <kiko-fud> it's great, flacoste 
[10:54] <flacoste> ok, changing this
[10:55] <elmo> oh oh, I know, how about 'Asked and answered'? ;)
[10:58] <flacoste> lol
[10:58] <flacoste> yeah, i'm sure mpt would love that
[10:59] <kiko-fud> I keep reading that as "Asked and Skewered"