=== dholbach [n=daniel@ubuntu/member/dholbach] has left #launchpad ["Ex-Chat"] [12:25] lifeless: you around? [12:28] no [12:28] ovoid [12:28] heh === raptoid [n=raptoid@85.101.150.108] has joined #launchpad === rbelem [n=rodrigo@ubuntu/member/rbelem] has joined #launchpad [01:33] I stayed logged into Launchpad over night [01:33] nice [01:37] the wonders of db-persistent sessions [01:37] tomorrow [01:47] actually I think its shared sessions [01:48] more than the persistence ;0 [01:51] in theory, you should remain logged in between production upgrades too ... [02:09] jamesh: as long as session formats dont change === mgalvin [n=mgalvin@cpe-69-205-47-165.nycap.res.rr.com] has joined #launchpad === mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #launchpad === mpt [n=mpt@203-167-187-77.dsl.clear.net.nz] has joined #launchpad [02:46] Gooooooooooooooood afternoon Launchpadders! [02:48] hi mpt [02:48] mpt: I got brad to test that javascript patch, and have committed it [02:49] ok, great === raptoid [n=raptoid@85.101.150.108] has joined #launchpad === njs [i=njs@adsl-66-159-194-130.dslextreme.com] has joined #launchpad [02:57] a question: as a project maintainer, why should I use rosetta over one of the other centralized translation project thingies? [02:57] (there seem to be a lot of these thingies, and little information to choose between them on) [03:03] njs, what other ones have you seen? [03:04] The only other one I know of is Irma, which has the problem that it's Linspire-only [03:09] Pootle, the Translation Project, ... [03:09] there is, of course, https://wiki.launchpad.canonical.com/RosettaCompetitors [03:15] njs: because it rocks [03:15] njs: because its not bound to one distro [03:16] njs: and because it will be bzr integrated in the future [03:16] making it non centralized [03:18] bzr integration is somewhat less interesting to me than some things, for reasons you might be able to imagine :-) [03:18] though I suppose it is nice for language teams to have such a thing! [03:18] neither pootle nor the translation project are distro-associated [03:19] though TP is clearly a less modern architecture === mpt didn't know about that RosettaCompetitors page :-) === lamont [n=lamont@mix.mmjgroup.com] has joined #launchpad [03:43] mpt: :-) === lamont [n=lamont@mix.mmjgroup.com] has joined #launchpad === stub [n=stub@ppp-58.10.73.202.revip2.asianet.co.th] has joined #launchpad === darkStar [n=raptoid@81.213.134.195] has joined #launchpad === lamont [n=lamont@mix.mmjgroup.com] has joined #launchpad [04:45] jamesh, can I bother you some more about backslashreplace()? [04:45] sure. [04:45] First I tried putting it in __init__.py, under smartquote() [04:45] and then in test.py itself [04:46] but either way I get NameError: name 'backslashreplace' is not defined [04:46] if you put it in webapp/__init__.py, you'll need to do "from canonical.launchpad.webapp import backslashreplace" in the page test .txt file [04:47] the alternative is to put the function definition in the page test file itself, where needed [04:47] I'd suggest either putting the function in the page test, or canonical/launchpad/helpers.py [04:47] well, it'll be needed in multiple pagetests [04:47] so I'll use helpers.py [04:58] AttributeError: 'LaunchpadBrowserResponse' object has no attribute 'decode' [04:59] at least it's importing now :-) === jinty [n=jinty@205.134.224.215] has joined #launchpad [05:01] jamesh, how do I turn a LaunchpadBrowserResponse into a str? [05:02] mpt: str()? [05:02] ha, should have guessed [05:03] hmm, str() gives me a whole pile of "\n"s [05:03] but there's something else in helpers.py for that [05:04] really? [05:05] at least, I thought there was [05:05] hmm, no there isn't [05:05] the repr() of a string with newlines will show '\n' [05:05] but if you use "print" to display it, they'll come up as newlines (as the page tests do at the moment) [05:05] I have >>> backslashreplace(str(http(r"""... === mpt tries re-adding the print [05:06] ah, now we're cooking with gas [05:09] maybe it'd make sense to put the str() call inside backslashreplace() [05:09] yes, if I'm going to be doing that every time anyway === mpt does so [05:09] yahoo accesses https://launchpad.net/robots.txt a lot [05:10] yahoo are just about criminal in the bw use [05:10] "bw"? [05:10] > 3000 times in the same period as google accessed the file ~ 200 times [05:11] bw == bandwidth === mpt wonders if changing robots.txt's caching headers would improve the situation [05:11] nope [05:11] yahoo are just silly [05:12] i have a unmodified for months robots and they did not reduce their hit rate last time I checked. [05:13] googlebot is using two user-agents: [05:13] "Googlebot/2.1 (+http://www.googlebot.com/bot.html)" and "Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)" [05:13] probably different dc's [05:13] http://help.yahoo.com/help/us/ysearch/slurp/slurp-03.html [05:14] hmm, that's not quite relevant [05:14] indeed [05:14] http://help.yahoo.com/help/us/ysearch/slurp/slurp-06.html [05:15] the google page says they access robots.txt at most once a day [05:15] (usually) [05:16] so we could add stuff to robots.txt to tell Yahoo "there is nothing for sale here" and "there are no interesting images here" [05:16] mpt: or we could blacklist them [05:16] or maybe actually, that might be once per day per googlebot instance [05:16] or give them a single page that says 'please use google.' [05:17] What is the number of visitors we get with yahoo.com referers, divided by the number of visitors we get from google.com referers? [05:17] I'm not saying it is necessarily a problem; I was just surprised at the numbers === Mez [n=Mez@ubuntu/member/mez] has joined #launchpad [06:29] Merge to devel/launchpad: [trivial] configs for production launchpad instance on gandwana (r2972: Stuart Bishop) [06:32] stub: we need to get the librarian onto mizuho at some point - I assume there's no further blockers for that? [06:34] elmo: Nope. Whenever your ready really. We can take the librarian down and it will only minimally affect Launchpad. [06:36] stub: hmm, what timezone are you on? [06:37] stub: anyway, I don't suppose now is a good time, but I've setup mizuho as a mirror of macaroni, including your account etc. so if you could login in, check it out, see if anything is missing [06:37] UTC+7 now [06:38] Now if fine if you have the slot. I don't think the migration is urgent through if you have other things to look at or are just having a bad hair day. [06:38] crap your core hours are basically my sleeping hours - that's so unuseful [06:39] stub: nah, I really need to go to bed, and karl won't be up for another of couple of hours in case it all goes Pete Tong [06:39] I can help [06:39] my hours are before stubs [06:39] stub: and it's reasonably urgent, as macaroni is down to 20Gb :P [06:39] lifeless: yeah, I know I just mean in general [06:40] I've split my core hours, with a big break in the middle, so I'm available on UK time. [06:40] StaffCalendar says 04:00-12:00 UTC? [06:40] if that's got a big break in a middle, someone's getting short changed ;-P [06:41] Bah [06:42] anyway, gone, night === stub wonders whos core hours he changed ;) [06:42] night elmo === LetterRip [n=LetterRi@udp038291uds.hawaiiantel.net] has joined #launchpad [06:44] hello all [06:44] I have two questions related to translation and documentation [06:45] Blender has an out of tree translation for the Chinese language [06:45] and would like to know if the section on Chinese translation for Blender can be pointed to their site [06:46] The other question, is that we have a many hundered page manual that is online in wiki format [06:46] that has translations in various states of completion [06:46] for about 6 languages in addition to english [06:47] is there anyway to get that into the rosetta system as well? === mpt [n=mpt@203-167-187-191.dsl.clear.net.nz] has joined #launchpad === LetterRip [n=LetterRi@udp038291uds.hawaiiantel.net] has left #launchpad ["Leaving"] === stu1 [n=stub@gb.ja.99.51.revip.asianet.co.th] has joined #launchpad === Mez [n=Mez@ubuntu/member/mez] has joined #launchpad === mpt [n=mpt@210-246-37-239.paradise.net.nz] has joined #launchpad === mick_home [n=clamwin@adsl-149-3-182.mia.bellsouth.net] has joined #launchpad [07:55] hi everyone [07:55] there are a few things that I'd like to fix in launchpad (a few things annoy me) [07:56] is it possible to get the source (via cvs/svn)? [07:56] or is it closed source? [07:57] mick_home, it's not open source yet [07:57] though the plan is for it to be released over time [07:57] What things annoy you in particular? [07:57] missing ical support [07:57] a calendar does very little for me ;-) [07:57] but ical support does a lot [07:58] mpt: is there a particular reason why it isn't OSS right now? === robitaille [n=robitail@ubuntu/member/robitaille] has joined #launchpad [08:01] i'd also like to see how the bts is tied into it (and see if it is possible to replace it with bugzilla) [08:02] a local install would also be nice to have [08:03] well, much of the benefit of Launchpad comes from people using a single instance [08:03] well, yes [08:04] but many people won't touch it if it isn't OSS [08:04] for example, the Big Idea of the bug tracker is to let multiple distributors, and upstream developers, collaborate on fixing the same bug [08:04] similarly, the advantage of Rosetta is that multiple distributors and upstreams can share each others' translations [08:05] this seems like a logically inconsistent argument. if there's such a benefit to having a single install, then people will continue to use that install, because they get benefits; even if the source code is open. [08:05] but on the other hand, many people will _not_ use that central install on philosophical grounds, if the source code is not open. [08:05] well gforge didn't kill sf.net - i doubt that releasing it as oss would harm launchpad in any way (it would only benefit) [08:06] njs, but then it would take *years* for the economies of scale to bring people into a single installation === robitaille [n=robitail@ubuntu/member/robitaille] has joined #launchpad [08:07] to be honest... i see only benefits [08:07] mick_home, what does sourceforge do to let distributors cooperate with each other? [08:07] for instance, you have a huge amount of people that won't touch launchpad atm (the DDs for instance) [08:08] mpt: i just compared opening up the sf.net code (gforge) and the success of sf.net in general [08:08] I understand that argument [08:08] (though I have less respect for it if such people also use Google:-) [08:08] heh [08:09] mick_home: you can get an ICalendar export at webcal://launchpad.net/people/username/+calendar/+icalendar [08:09] (replace "webcal" with http or https, if you want) [08:10] one lesson I have learned as a maintainer is that whether I respect my users or not they still are important :-) [08:10] mick_home: but the management have made the decision that the benefit from forcing the economies of scale early on is greater than the harm from however many developers won't use it for as long as it's closed [08:10] and I think they're probably right [08:10] blog would also be nice [08:10] since homepage != blog === robitaille [n=robitail@ubuntu/member/robitaille] has joined #launchpad [08:11] mick_home, you mean a Weblog field for a registered person? [08:11] blog link [08:11] blogging possibilities (via launchpad) [08:11] yes [08:11] we've just been considering that in the past week [08:11] and the possibility to parse an rss feed of a blog [08:11] https://wiki.launchpad.canonical.com/BlogIntegration [08:11] mick_home: allowing people to give a link to their web log / syndication feed is being considered [08:12] mick_home: to make it easy to generate a planet config from a LP team, for instance [08:12] so that Launchpad can Planetize teams [08:12] snap ;-) [08:12] well, i'm just adding stuff that i'd be interested in (using and developing) [08:14] mpt - would you mind shooting me an email if it ever does go open source === sd-tux [i=sd@2001:4ca0:0:fe00:0:0:a96:3f18] has joined #launchpad === JanC [n=janc@lugwv/member/JanC] has joined #launchpad [08:17] mpt: are you a dev of launchpad? [08:17] mick_home, I am a developer [08:18] the plan as far as I know is to open source it parts at a time [08:18] Are you interested in particular parts, or just the whole shebang? :-) [08:18] well, i can't test in parts now can i ;-) [08:20] mick_home, if you just want to play around with various features, you can use https://staging.ubuntu.com/ === Keybuk [n=scott@descent.netsplit.com] has joined #launchpad [08:20] (which really should be staging.launchpad.net, but anyway) [08:21] i'm interested in coding features (not testing new features) [08:22] ah [08:22] mpt: staging.launchpad.net does go to the same server as staging.ubuntu.com, but seems to show the shipit interface [08:22] jamesh, yes, there's something in my Inbox about rationalizing the various domains [08:23] actually, for me it redirects [08:23] mick_home, have you done much with Zope before? [08:23] mpt: will there be many differences from production? or is staging what is made production each week? [08:23] ajmitch, the latter [08:23] zope 3 is fun, really ;) [08:24] python yes... zope no [08:24] so it's somewhat newer than production about 5/7 of the time [08:24] ok [08:24] mpt: like i said, mind sending me an email when/if the source ever becomes available? [08:26] i understand if you say "no" btw (since I'm probably not the first person to ask) [08:26] mick_home: when things are released, we won't keep it a secret [08:26] you can be sure of that [08:27] mick_home, you're not the first person to ask for the source, but you're the first to ask for e-mail notificaiton :-) [08:27] well ya i get that [08:27] I'd probably forget [08:27] but i don't check the site 24/7 ;-) [08:27] mick_home, do you read Slashdot? [08:27] mpt: have parts been freed already? [08:27] mpt: oh god no... i used to [08:28] it will probably make it on /. though [08:28] mpt: my experience is with python/wxPython apps [08:28] yes, that's why I asked [08:28] not really python web apps [08:28] mick_home, ok, e-mail mpt@canonical.com and I'll reply when it's all open-sourced :-) [08:29] thanks :-) [08:29] ajmitch, not as far as I know [08:29] ajmitch: no Launchpad specific bits -- we have fed back improvements to open source projects we make use of though [08:29] e.g. psycopg, pyme, etc === lbm [n=lbm@cpe.atm4-0-1301006.0x50a0824e.vgnxx6.customer.tele.dk] has joined #launchpad [08:30] jamesh: yes, I've seen that, which is great === fabbione [i=fabbione@gordian.fabbione.net] has joined #launchpad === Nafallo_away is now known as Nafallo [08:31] mpt: thanks (email sent) === fabbione [i=fabbione@gordian.fabbione.net] has joined #launchpad === fabbione [i=fabbione@gordian.fabbione.net] has joined #launchpad === carlos [n=carlos@198.Red-83-50-62.dynamicIP.rima-tde.net] has joined #launchpad [08:41] morning [08:41] stub, hi [08:41] stub, when are you going to do the production update? [08:46] carlos: yesterday === Nafallo is now known as Nafallo_away [08:46] oh, fuck [08:46] hmm, I need a cherrypick then === Nafallo_away is now known as Nafallo === Nafallo is now known as Nafallo_away === Nafallo_away is now known as Nafallo [09:01] daf, ping === mick_home [n=clamwin@adsl-149-3-182.mia.bellsouth.net] has left #launchpad [] === mick_home [n=clamwin@adsl-149-3-182.mia.bellsouth.net] has joined #launchpad [09:30] hi guys (i'm back) [09:31] i am curious to know if launchpad supports only gettext or other means [09:31] like RC files [09:35] mick_home, only gettext [09:35] mick_home, but there are many tools that gives you a .po file from many other resources to translate [09:35] so i would convert from RC -> .po then have that go from .po -> RC ? [09:36] right [09:36] carlos: have a link? [09:41] mick_home, is the RC file like 'key=value' content? [09:43] standard Windows .rc [09:45] mick_home, sorry, I don't remember the details of that format... could you show me an example, please? [09:53] nm i found what i'm looking for [09:54] carlos: http://icu.sourceforge.net/userguide/localizing.html#Sample_rc [09:54] if you want to see a real example, check out the ReactOS bugzilla setup [09:55] all the localizations are .rc files [09:55] hmmm [09:56] I'm not sure how to get that into a .po file... [09:56] It looks like openoffice's files [09:56] i'll be back once launchpad goes open source... there are too many features missing atm that i obviously can't impliment w/o the source [09:56] mick_home, but you can request us to implement them and we can try to do it... [09:57] yes, but I don't like asking when I can do it myself at a later date ;-) [09:57] my list of wants: [09:57] - blog integration [09:57] - ical support [09:58] - rss parsing of blogs [09:58] - rss /rdf support in general [09:58] - support for other formats than gettext [09:59] - IM clients should be listed (not irc / jabber specific imo) [09:59] carlos: pong [10:00] mick_home, most of those features are planned or being discussed atm [10:01] daf, I'm with #1681. Could you confirm that the final decission to fix it was to allow '\r\n' stored into our database if the msgid had them? [10:01] mick_home, if you file bugs about them I'm sure that the ones that are not planned would be implemented [10:02] Ubugtu, you are sleeping.... [10:03] carlos: I can't remember off the top of my head [10:03] daf, ok [10:04] I think that was the last thing we talked but I'm not sure if with the triage you are doing with SteveA you decided something else [10:04] if we decided something else, we would have said so in the bug :) [10:04] I'll look at the LP mail [10:05] morning all! [10:07] sivang, morning!! === carlos -> breakfast [10:08] carlos: morning carlos, good to be back. === sivang is back of 2 days (apart from weekend) of being sick [10:08] sivang, Welcome back then [10:08] I hope you are at 100% ;-) [10:09] carlos: yes, back to myself , thanks you. [10:09] hi Sivan [10:09] glad you're better [10:09] carlos: hmm, have we ever seen a msgid with '\r' in it [10:09] carlos: I'm wondering whether this is a YAGNI [10:10] daf: thanks :-) [10:11] launchpad_staging=> SELECT count(*) from pomsgid where msgid like '%\r%'; [10:11] count [10:11] ------- [10:11] 39 [10:11] (1 row) [10:11] ok [10:12] my next question is: which templates do these belong to? [10:12] daf, I suppose they will be related to Mono applications [10:12] let me check [10:13] it kind of sucks to expose translators to \r [10:13] you can always convert from logical to physical newlines in the code [10:13] daf, any developer that uses Windows to write the application is adding this "problem" [10:14] 39 / X million is not much [10:14] obviously it's > 0 [10:15] just to check: [10:15] you plan is to strip newlines from translations unless the msgid contains \r? [10:16] 14 templates have the \r [10:16] ok [10:16] maybe nopaste the PTNs? [10:16] daf, if the msgid only has '\n', we strip the translation [10:17] if it has \r\n, we leave it (or fix it if is missing them) [10:17] or if it has no '\n' :) [10:17] I need to check if we have just '\r' [10:17] good idea [10:17] but that's from the old MacOS <= 9 systems... [10:17] or if somebody is doing something weird [10:18] APT uses \r to do progress bars [10:18] but I don't think it exposes that to translators [10:18] exposing that to translators seems like an unambiguous bug :-) [10:19] One of the templates with \r chars belongs to "The Ubuntu package descriptions for the releases." product... [10:19] njs: to sensible people like you and me, sure :) [10:20] carlos: see, I think that should be fixed [10:20] name [10:20] ------------- [10:20] blender [10:20] gnue-forms [10:20] hercules [10:20] kvirc [10:20] libchipcard [10:20] ncpfs [10:20] shadow [10:20] simplecdrx [10:20] slrn [10:21] somaplayer [10:21] zssh [10:21] (11 rows) [10:21] that's the list of the other sourcepackages that use the \r char [10:21] hmm [10:21] daf, so we can fix it on import/submit time without asking the user to pay attention to it [10:21] but only when the '\r' is used as '\r\n' [10:22] I'm not sure I understand what you mean [10:22] by changing the msgids? [10:22] if \r is alone... as you suggests with apt.. that would be a bit more difficult to fix [10:22] daf, no, the translations [10:22] right [10:22] changing the msgids would be dodgy [10:22] daf, changing msgids is broken ;-) [10:22] yes [10:23] so, I say we go with your plan [10:23] but it would be nice to generate a warning if somebody imports a template with '\r' [10:23] It makes no sense to add a visual way to represent the \r [10:23] ok [10:23] how difficult do you think that would be? [10:23] we can use lpnotification for web uploads [10:23] not really.... [10:24] no? [10:24] the web upload is not doing the import [10:24] so we cannot know that at that point [10:24] sure, it's not doing the import [10:24] daf, but we need to add notification mails when the .pot file is imported [10:24] so we can add there the warning === raptoid [n=raptoid@81.213.133.116] has joined #launchpad [10:24] but I think parsing the PO file should be quick enough that it's not a problem to check there [10:25] ah, good point [10:25] is there a bug open on that? [10:26] daf, About if it's difficult or not... I need to finish POMsgSetPage implementation first or I will get a ton of conflicts, but is not too complicate the fix of that bug [10:26] for .pot notifications? [10:26] yep [10:26] I don't think so [10:26] at least I don't remember it [10:26] ok [10:26] I propose: [10:26] do the fix you suggested [10:26] file a bug on notificatoins for POT imports [10:26] mick_home: you can use gettext's msgfmt utility to generate a .Net resource file. That might be similar to a Windows resource file [10:26] mention that we should warn about \r [10:27] did I miss anything out? [10:27] ok, sounds like a plan [10:27] good :) [10:27] daf, thanks for your input [10:27] you're welcome === carlos -> breakfast [10:27] later [10:27] carlos: oh, maybe say in #1681 what the plan we agreed is [10:37] jamesh: ty [10:37] .resx file? [10:38] mick_home: no idea [10:38] hi folks [10:38] hi steve [10:38] mick_home: I'm just looking at the "msgfmt --help" output [10:38] i'm getting ready to catch a plane to london [10:38] is there anything urgent that i should do here? [10:38] test [10:38] mick_home: the xgettext utility also says it supports pulling strings from C# code [10:38] i don't think i need to do any testing right now, lifeless [10:39] SteveA: test that you have everything needed for the sprint [10:39] jamesh: i saw mixed docs on that [10:40] mick_home: anyway, it would probably be worth testing. [10:40] lifeless: you mean, construct a miniature stub sprint, and briefly attend that? [10:41] SteveA: no, just mock it up and walk through the responses [10:43] SteveA: have a good flight === jamesh wonders what stub would think about being miniaturised [10:45] lol [10:46] Honey, I Shrunk the DBA === koke [n=koke@ubuntu/member/koke] has joined #launchpad [10:47] jamesh: he'd probably cope so long as you didn't inject him into elmo's arse or anything === sivang rotfls [10:49] Kinnison: hi daniel! 'sup? [10:51] hi sivang [10:51] I heard you were ill [10:51] Kinnison: yep, I'm better now === Nafallo looks at the new, improved sivang :-) [10:52] sivang: excellent. [10:52] Nafallo: re-energated for sure :) [10:52] Kinnison: Have you had any chance to practice some herbew during the holidays? [10:53] sivang: when I was unwell, I lay in bed reading the book [10:53] sivang: I still don't recognise all the letters :-( === Kinnison thinks his friend nattie wants to learn too, so will talk with her about setting aside regular time to sit together and learn [10:53] everyone has been ill except me? :-P [10:53] Nafallo: possibly :-) [10:54] Kinnison: ill again? during holiday time? bad :-/ [10:54] sivang: aye, just my luck [10:54] I broke my damn toe instead though :-/ === Kinnison returns from taking rob to a funeral [11:36] lifeless: any chance of getting PQM's duplicate newline bug fixed -- it's bloody annoying [11:37] daf: I gave lifeless a patch for that almost a year ago [11:37] daf: patches gratefully accepted [11:37] Kinnison: I have no idea where that is [11:37] just put up a branch to include logs of executed statements in OOPS reports [11:37] lifeless: you suck === Kinnison grins [11:37] Kinnison: yes I do, for the right person [11:37] lifeless: baddumtish [11:37] lifeless: where is pqm's bzr branch? [11:37] daf: http://people.ubuntu.com/~robertc/pqm/trunk [11:37] jamesh: ^^ [11:37] TDD is mandatory for patches [11:38] jamesh: re OOPSes: that's great === Nafallo is now known as Nafallo_away [11:39] we'll also be able to start generating OOPS reports for requests that take a while, but don't hit the hard timeout [11:39] that's also great :) === matsubara [n=matsubar@200-171-140-32.dsl.telesp.net.br] has joined #launchpad [11:40] good morning! [11:40] good morning Brasil! [11:40] morning matsubara ! === Nafallo_away is now known as Nafallo [12:01] lifeless: are items in the "output" list of CommandRunner objects meant to be newline-terminated or not? [12:01] (in pqm) [12:07] I think the problem is its never been defined properly [12:08]