[01:59] <mpt> Goooooooooooooood afternoon Launchpadders!
[02:30] <mpt> jamesh, I have another annoying question about using a repository
[02:40] <mpt> jamesh, I primed the repository as you suggested, by branching from rocketfuel-built. How do you then keep that repository branch of rocketfuel up to date? bzr pull?
[02:55] <mpt> Perhaps spiv knows the answer :-)
[02:56] <jamesh> mpt: you can use pull, or you can ignore it -- as you merge rocketfuel into your other branches in the repo, you'll get the revisions anyway
[02:56] <jamesh> mpt: branching current rocketfuel into the repo is mainly a way to get all the history up front
[02:56] <mpt> jamesh, so I should maintain my rsynced copy of rocketfuel-built, and merge from that into other branches?
[02:57] <mpt> rather than merging from the repo's rocketfuel
[02:57] <jamesh> mpt: that's what I do
[02:57] <mpt> ok
[02:57] <mpt> thank you
[02:57] <jamesh> mpt: on the first branch you merge rocketfuel-built into, it will have to copy the revisions into your repo, but for subsequent ones the revisions will already be there
[02:58] <jamesh> so the second branch you merge into should be quicker
[05:31] <jamesh> spiv: ping?
[05:33] <spiv> jamesh: pong
[05:33] <jamesh> spiv: is there any particular reason we'd want to favour the zopeless database adapter in scripts over the one we use in the webapp?
[05:35] <jamesh> spiv: I noticed that after running execute_zcml_for_scripts(), enough setup is done to access the db without running initZopeless()
[05:35] <spiv> Probably not much reason, then.
[05:35] <spiv> I can hardly rememeber what original problem initZopeless was meant to solve ;)
[05:36] <jamesh> I don't think we had execute_zcml_for_scripts() back then
[05:36] <jamesh> so it was probably to use the SQLObjects without the Zope component framework
[05:36] <jamesh> but most of our scripts use the zope component framework these days
[05:36] <spiv> Yeah, I think that's basically it.
[05:37] <spiv> Certainly, if you're doing execute_zcml_for_scripts, then you have the Zope CA initialised, so "zopeless" doesn't really seem to make much sense for that case.
[05:39] <jamesh> okay.  So if we want to do the error reporting for scripts stuff, we can probably ignore the initZopeless() stuff all together
[05:41] <stub> If we can make initZopeless be almost a noop, perhaps a thin shim to wrap ztm.commit() and ztm.abort(), that would be great.
[05:42] <jamesh> stub: without running initZopeless() you can just use the transaction module
[05:42] <jamesh> mpt: get a faster hard disk or more memory
[05:42] <stub> jamesh: I meant for backwards compatibility
[05:42] <jamesh> stub: "ztm = transaction" ?
[05:42] <mpt> yeah
[05:42] <stub> jamesh: That might do, yeah.
[05:43] <jamesh> mpt: western digital raptor disks are good, but might not fit in a laptop :)
[05:43] <stub> Might need to teach the PsycopgDA to connect as a user other than 'launchpad'
[05:43] <jamesh> yep
[05:44] <jamesh> and be able to turn off timeouts, etc
[05:45] <mpt> jamesh, I'm grepping the same (very large) files for different strings each time (piping the results to wc -l), and it takes about ten minutes each time. I don't suppose there's an easy way of running through all the files once, looking for all the strings simultaneously? Maybe grep isn't the right tool
[05:45] <jamesh> mpt: grep -c "some string" filename1 filename2 ...
[05:46] <jamesh> the above will give you a count of the lines matching the string from each file
[05:46] <jamesh> oh.  you want to search multiple strings in one file
[05:46] <mpt> yeah
[05:47] <jamesh> how big a file is it?
[05:47] <mpt> I suppose -c would be a bit quicker than "| wc -l", but then I'll have to total the counts in each file myself
[05:48] <jamesh> if you want a total, piping to "wc -l" should be about as fast
[05:48] <mpt> jamesh, ~1 GB each
[05:49] <jamesh> mpt: you could write an awk or python script to do the regexp matching and generate totals
[05:50] <spiv> Or egrep?
[05:50] <jamesh> or try and run the different greps at the same time in different terminals, hoping that they'll progress at roughly the same speed
[05:50] <spiv> egrep "one|two|three" files...
[05:50] <spiv> I guess that doesn't give a count per-string.
[05:50] <mpt> I need separate totals, spiv
[05:50] <mpt> yeah
[05:51] <jamesh> but spiv's idea might reduce the amount of data you need to work with
[05:51] <jamesh> save the egrep output to a file and then grep for each pattern individually
[05:51] <spiv> jamesh: I was just about to suggest that :)
[05:52] <mpt> Wouldn't that be slower?
[05:52] <mpt> I'd end up searching for each thing twice instead of once
[05:52] <jamesh> mpt: what proportion of lines do you expect to match any pattern?
[05:53] <mpt> on the order of 1% (but a Zipf distribution)
[05:53] <jamesh> mpt: so you do "egrep 'pattern1|pattern2|...' files ... > output" and you've gone from 1GB down to 10MB
[05:54] <jamesh> you can then do "grep pattern1 output" and you are only processing 10MB
[05:54] <mpt> hmmm, ok
[06:02] <lifeless> I was wondering if it had group stats 
[06:02] <lifeless> but I cant see anything
[06:02] <lifeless> trivial to do in python though
[06:20] <jamesh> I'm getting a bizarre test failure in canonical.base -- it seems that string.letters has been reversed to have uppercase first which causes the test to fail
[06:25] <jamesh> ah.  looks to be locale dependent
[08:23] <yaniv> I have a question: a lot of strings in Rosetta have a "_" within them - what does it mean, and how should I handle it?
[08:49] <jamesh> yaniv: are they for GTK programs?
[08:49] <yaniv> jamesh: yes
[08:50] <jamesh> yaniv: the underscores are used to generate the mnemonics for menu items, buttons, etc
[08:50] <jamesh> yaniv: so if I set the label for a menu to "_File", it will display as "File" with an underscore under the "F"
[08:50] <jamesh> and I can type "alt+f" to open the menu
[08:51] <jamesh> If I translate the string to "blah _blah", then I'd need to type "alt+b" to open the menu
[08:51] <jamesh> (and the second b would be underlined
[08:51] <yaniv> oh. so what should I do when I translate into Hebrew? they  can't mean anything in Hebrew
[08:52] <yaniv> can they?
[08:52] <jamesh> well, you'd want to map one of the hebrew letters to a keyboard accelerator right?
[08:53] <jamesh> so just precede the letter with an underscore
[08:53] <jamesh> (that is precede the letter in the logical flow of the text, rather than how the text is displayed)
[08:54] <yaniv> but what about menu items? they have underscores too, but alt-x doesn't work for them
[08:54] <jamesh> e.g. in the gedit translations there is:
[08:54] <jamesh> msgid "_File"
[08:54] <jamesh> msgstr "_"
[08:55] <jamesh> the alt modifier is only needed to pop up the menu
[08:56] <jamesh> you just type the mnemonic directly to select one of the menu items
[08:56] <yaniv> I see
[08:57] <jamesh> also, if you are using multiple keyboard groups (e.g. hebrew and latin), GTK is smart enough to activate mnemonics/accelerators for the other group
[08:58] <Fujitsu> Is there a way to do mass tagging of bugs?
[08:58] <jamesh> Fujitsu: not at the moment.
[08:59] <Fujitsu> Grrreat. Thanks anyway.
[09:03] <SteveA> morning
[09:03] <Fujitsu> Evening, SteveA.
[09:05] <SteveA> Fujitsu: hi.  what place are you in that it's evening already?
[09:06] <Fujitsu> Melbourne, Australia.
[09:07] <SteveA> awesome!  I'd figured Japan from your nick, but what do I know? ;-)
[09:09] <lifeless> jamesh: how easy would it be to get a summary of the daily oops summary sent to another email address ?
[09:09] <SteveA> I'm in London today... working on the bzr-launchpad connection with ddaa and _thumper_
[09:09] <jamesh> lifeless: pretty easy.  At the moment matsubara is running the script so you'd have to ask him
[09:10] <Fujitsu> SteveA, aha. The nick came from a hard disk, several years ago. A friend decided that wgrant as a username was too boring, and he had a Fujitsu HDD next to him. It's sorta stuck.
[09:11] <jamesh> Fujitsu: so in another life you might have been Maxtor?
[09:11] <Fujitsu> That's what people suggest, yes :P
[09:11] <Fujitsu> Or Mitsubishi has been suggested, as Fujitsu makes air conditioners as well.
[09:12] <jamesh> lifeless: btw, the fixes for the last few product-release-finder problems were included in my latest merge
[09:12] <jamesh> (moving the release pattern fields over to the product edit form, and stripping ".orig" from version numbers)
[09:12] <jamesh> so once that is rolled out, I think we can put the code into production use
[09:14] <jamesh> SteveA: I was talking with spiv earlier: it seems that if you run execute_zcml_for_scripts() in a script, you can access the database using the webapp database adapter
[09:14] <jamesh> without calling initZopeless()
[09:14] <jamesh> which also gives you SQL statement logging, which would be useful for error reporting in scripts
[09:14] <sivang> morning happy people :)
[09:15] <stu1> I know someone called Mitsubishi. She was named after a neon sign.
[09:17] <lifeless> jamesh: AWESOME
[09:17] <lifeless> jamesh: ok, I'll ask matsubara - thanks
[09:17] <SteveA> jamesh: nice
[09:17] <SteveA> jamesh: it makes sense.
[09:19] <sivang> SteveA: how's london today? rainy?
[09:20] <SteveA> sivang: the very instant you asked, it started raining.
[09:21] <sivang> My luck ...tough one ;-)
[09:21] <sivang> SteveA: Are you in South Kensington ?
[09:22] <sivang> SteveA: But to be frank, I don't mind if its pouring. I had a 2 hours walk through Covent Garden, picaddily and oxford circus in heavy rain. Was great fund.
[09:22] <sivang> SteveA: (Visited the UK 2 weeks ago)
[09:36] <SteveA> I'm glad you liked it.  I like the rain when it's warm.  Don't like it when it's cold and windy too.
[09:38] <sivang> agreed. Cold is fine when it's not humid, and without wind. I was lucky such that when I visited, people said it was the hottets weather for that period of year since about 30-40 years ago. It was mostly sunny.
[09:50] <Ubugtu> New bug: #65007 in launchpad "managing bugfix branches is more work than necessary" [Undecided,Unconfirmed]  http://launchpad.net/bugs/65007
[11:35] <SteveA> stu1: ping
[11:36] <SteveA> jamesh: ping
[11:36] <jamesh> SteveA: pong
[11:36] <SteveA> jamesh: https://features.launchpad.net/products/launchpad/+spec/form-layout
[11:36] <stu1> SteveA: pong
[11:37] <SteveA> jamesh: is there anything left for you, or anyone else, to do on that spec?  Please check the status is updated to reflect the current state of things.
[11:37] <SteveA> stub: edge-launchpad, name-blacklist, canonical-names
[11:37] <SteveA> stub: please check that the spec statuses are up to date for these
[11:38] <SteveA> and tell me if there is anything about those specs that you yourself cannot do
[11:40] <jamesh> SteveA: none of the layout improvements from form-layout have been done, iirc
[11:40] <jamesh> we've got the initial form focus and button label customisation though
[11:41] <SteveA> please update the status and the status whiteboard with this information
[11:42] <SteveA> you are listed as the assignee for that spec.  is there anything left to do in order to complete that spec that you yourself cannot do?
[11:43] <stub> SteveA: The NameBlacklist branch is up for review. I haven't done the UI for maintaining the list though, as a) it can be maintained using SQL and b) I will be slow throwing up the forms since I haven't done any UI work for ages (nor used the newer form machinery at all).
[11:44] <SteveA> is the UI for it part of the spec?
[11:45] <stub> SteveA: Yes, including the workflow for renaming dud accounts and sending notifications when a blacklist entry is added or modified.
[11:45] <stub> SteveA: Although I'm not sure if that is actually desirable
[11:45] <SteveA> who would get notified when a blacklist entry is added or changed?
[11:46] <stub> SteveA: (eg. we really don't want to rename the 'admins' team if someone alters the blacklist from '^admin\d+' to '^admin\w*'
[11:46] <stub> SteveA: The owner of the product or person or project or distro or whatever that conflicts with the blacklist entry.
[11:46] <SteveA> there should be a page that lists dud account names
[11:46] <SteveA> and I don't think we should do automatic emails
[11:46] <SteveA> I think we should have a page for lp admins
[11:47] <SteveA> that shows the names that clash
[11:47] <SteveA> and that's all
[11:47] <SteveA> there should be a UI for listing blacklist entries and adding/removing/changing them
[11:47] <SteveA> this can take the form of a single text area
[11:47] <SteveA> parsed line by line
[11:47] <SteveA> no fancy forms stuff required
[11:48] <SteveA> I think changing dud names is something we should do as a human process
[11:48] <SteveA> not automated
[11:48] <SteveA> so, no need for emails
[11:48] <SteveA> so... basically one page required
[11:48] <stub> Sounds fine provided external systems like the @ubuntu.com email address sync understand this
[11:48] <SteveA> text area + list of dud names with links
[11:49] <stub> Anway - the minimum we need is up for review, so I will defer the rest of that until the other 1.0 tasks are done
[11:49] <SteveA> if you can do that one page, then we can sign off the spec
[11:49] <SteveA> that's worth doing I think
[11:50] <SteveA> but please also update the spec to reflect this conversation
[11:50] <SteveA> so that the expectations for the spec are clear
[11:54] <SteveA> jamesh: cc-ed you on a schooltool email about importing their bugs into launchpad.
[11:55] <jamesh> got it
[11:57] <jamesh> SteveA: we've got an importer for the format output by the sourceforge screen scraper.  I am not sure if that counts as a well known format though
[12:02] <SteveA> jamesh: right.  but, you could ask someone to produce bugs in that format, for importing
[12:02] <SteveA> and perhaps be flexible about the statuses used
[01:47] <_thumper_> ping jamesh
[01:48] <jamesh> _thumper_: pong
[01:48] <_thumper_> jamesh: looking at some timeout problems that are occurring with some of the branch stuff
[01:48] <_thumper_> jamesh: SteveA suggested we (ddaa and myself) talk to you about a specific exception to raise
[01:48] <_thumper_> if there are too many (for some definition of many) items
[01:49] <jamesh> which page is this?
[01:49] <jamesh> a branch list for a person or something?
[01:49] <_thumper_> yeah, specifically the import person
[01:50] <_thumper_> particularly with the detailed list of branches
[01:50] <_thumper_> where it does more joins than the default view (which is the list)
[01:50] <jamesh> _thumper_: I don't think we've got a particular exception for that right now, but would batching help here?
[01:51] <_thumper_> jamesh: the idea is to have a generic exception that can be raised in different places
[01:51] <_thumper_> jamesh: the idea is that we don't want to support batching right now on this (don't ask)
[01:52] <_thumper_> I think the idea is to come up with a name and place for this specific type of exception
[01:52] <jamesh> _thumper_: okay.  Possibly lib/canonical/launchpad/interfaces/launchpad.py would be a good place to define the exception
[01:53] <jamesh> you'd also want to register a custom error page for it in the ZCML
[01:53] <jamesh> as for a name, "TooMuchData" or "TooManyItems" maybe?
[01:54] <_thumper_> where are exceptions defined atm?
[01:54] <_thumper_> there don't seem to be any in that file right now
[01:54] <_thumper_> I like TooManyItems
[01:55] <jamesh> usually we define exceptions close to the code that uses them, but this is a pretty generic one.
[01:55] <jamesh> the other place would be somewhere in webapp/
[01:55] <_thumper_> yeah
[01:55] <jamesh> but I am not sure exactly where in webapp/ it would belong
[02:34] <salgado> BjornT, I just saw that my mirror-management-tweaks branch has been assigned to you. we need to get that on staging ASAP, so I was wondering if you'll have some time to review it today (it's around 200 lines, mostly templates and tests)
[02:43] <BjornT> salgado: if it's really urgent, then it's probably best to find another reviewer. my queue is quite overloaded at the moment. i'd be happy to review it if you took one of my reviews, though ;)
[02:46] <BjornT> salgado: i'll try to get around reviewing it tomorrow, though.
[02:55] <salgado> BjornT, thanks dude. I'll see if I can get somebody to review it today, then you'll have one less
[02:56] <salgado> I wish I could take one from your queue, but I just got some extra shipit stuff to do, so I don't think I'll have time for reviews. :/
[03:02] <salgado> spiv, maybe you have a few minutes to review my mirror-management-tweaks branch?
[03:16] <stub> BjornT: Was my nospam branch r=bjornt? Or were there other issues?
[03:20] <BjornT> stub: yeah, it was r=me.
[03:27] <salgado> jamesh, maybe you can review my mirror-management-tweaks branch? (around 200 lines, mostly template and tests)
[03:52] <stub> He has a review for me first, which needs to land before the production systems run out of disk space and Launchpad dies a horrible death.
[04:50] <salgado> SteveA, I guess you don't have a few minutes for a quick review either?
[05:06] <SteveA> salgado: nope
[05:24] <BjornT> salgado: i'll review your branch now.
[05:26] <salgado> BjornT, great dude, can you wait just a second? the diff on pending-reviews is not up to date
[05:26] <BjornT> salgado: sure
[05:26] <salgado> I'm generating one
[05:31] <salgado> BjornT, I have a merge in between the commits, but the changes are not related, so the full diff is what's in pending-reviews plus https://devpad.canonical.com/~andrew/paste/fileLDn6hb.html
[05:36] <kiko> morning
[05:51] <SteveA> salgado: direct person registration work
[05:51] <SteveA> that's registered as "started" in the spec system
[05:51] <SteveA> any further progress?
[05:51] <kiko> it's started
[05:51] <kiko> salgado and I discussed how to move forward
[05:51] <kiko> and it's sorted
[05:51] <kiko> I did that thursday.
[05:52] <SteveA> ok, thanks
[05:58] <Seveas> One can subscribe to bugmail for packages, is this also possible for bugmail against 'Ubuntu' (no package specified)?
[05:59] <bradb> Seveas: not specifically, sorry
[05:59] <Seveas> damn
[06:00] <bradb> Seveas: you could make it work with mail filters though
[06:00] <Seveas> yes, that's what I'm going to do
[06:54] <salgado> flacoste, around?
[06:54] <flacoste> salgado: was about to go for lunch, but go ahead
[06:56] <salgado> flacoste, just a quick question... do you think we should always send a small note to a support contact which doesn't speak the language of that ticket or should we do that only when not a single support contact speaks that language and then, in this case, send the small note to everybody?
[06:56] <salgado> by speak I mean, have that language as preferred
[06:57] <flacoste> salgado: the latter, i.e. only send a note to everybody when none of the support contacts have that language registered
[06:58] <flacoste> salgado: btw, did you saw the changes I made in tt-workflow on the ticket notification implementation?
[06:58] <salgado> in all other cases, only send notifications to people who speak that language?
[06:58] <flacoste> exactly
[06:58] <salgado> oh no, I didn't. I'll check that before moving forward
[06:58] <salgado> it's also a good time to check how many conflicts we already have
[06:59] <salgado> flacoste, thanks for reminding me again. :)
[06:59] <flacoste> salgado: let me know if anything is hard to resolve
[07:00] <salgado> sure thing
[07:00] <flacoste> salgado: my new branch tt-views will also likely conflict with yours since we're both making changes to the searchTickets implementation (and its view for that matter)
[07:00] <flacoste> I'll try to merge yours and see how it goes
[07:01] <flacoste> food time now
[07:15] <salgado> flacoste-lunch, only the conflict on doc/support-tracker-pages.txt is non-trivial, but I'll have to apply your changes to mailnotification.py to my branch before doing the changes I need on that file. it'd be great if you could tell me if/when you do any other changes on it.
[07:41] <SteveA> salgado: ping
[07:41] <salgado> SteveA, pong
[08:21] <flacoste> salgado-afk: hmm, support-tracker-pages.txt, ok, if I remember correctly I mostly removed stuff from that file and simplify the notification testing. Shouldn't be too hard to merge
[08:21] <flacoste> salgado-afk: and I'll keep you posted about future changes to the branch, future changes will be prompted by BjornT's review
[09:10] <kiko> cprov, can you update the status for the build-failure-process?
[09:11] <cprov> kiko: sure
[09:11] <kiko> thanks
[09:13] <cprov> kiko: done
[09:17] <sivang> what the best way to search for translations of a specific product/package?
[09:22] <kiko> there is none currently.
[09:22] <kiko> download the translation file, sivang?
[09:23] <sivang> you mean, Rosetta is out of order?
[09:23] <qbrix> zomg
[09:23] <qbrix> hello!
[09:23] <sivang> kiko: I just got a request from someone to approve his translations for edgy, for network-manager to hebrew. He urged to approve them before release so they wil get included.
[09:24] <kiko> sivang, and?
[09:24] <blixa> hello
[09:24] <qbrix> I got my stickers today!!!
[09:24] <qbrix> THANKS UBUNTU!!!
[09:24] <blixa> Please note that requests usually take from 4 to 6 weeks to deliver, depending on the country of shipping.
[09:24] <sivang> kiko: I'm trying to find the proposed trnaslatoins and can't :)
[09:25] <sivang> translations can only be made against upstream products right? not packages as well...
[09:26] <kiko> sivang, they can be made against both.
[09:26] <qbrix> Are these stickers open source?
[09:26] <qbrix> What license do these stickers work with?
[09:26] <qbrix> GPL?
[09:26] <qbrix> Commons?
[09:26] <kiko> qbrix, you must be asking if the designs on the stickers are open source
[09:26] <qbrix> Can I place these in public?
[09:26] <blixa> the stickers make the OS
[09:26] <sivang> kiko: okay, thanks.
[09:26] <qbrix> Do I have to put on a cd if I burn it with an ubuntu ISO?
[09:27] <kiko> the stickers themselves can go wherever you like, though if you do something illegal with them you may be liable!
[09:27] <qbrix> uh oh, I don't want to get sued
[09:27] <qbrix> can I return these stickers?
[09:27] <qbrix> they should print the GPL on the back of these stickers
[09:28] <qbrix> is kamran v. here?
[09:29] <qbrix> oh god, my hands hurt
[09:29] <qbrix> i think i have aids
[09:30] <kiko> either that or too much LSD
[09:30] <blixa> hehe you shouldn't have fisted that bison
[09:30] <blixa> stick to goatse
[09:30] <qbrix> i'm done with LSD
[09:30] <_Shade_> hi there
[09:30] <qbrix> _Shade_: be careful with the stickers
[09:30] <blixa> don't underestimate their potency
[09:30] <_Shade_> huh?
[09:31] <qbrix> _Shade_:  <qbrix> i think i have aids
[09:31] <qbrix> kekekekeke
[09:32] <qbrix> i can't do this anymore
[09:32] <sivang> hmm
[09:32] <sivang> I think I've just experienced culture shock ;-p
[09:35] <_Shade_> hey what are these karma points for?
[09:36] <_Shade_> talking about culture shock :P
[09:37] <sivang> _Shade_: you receive karma points proportional to your conrtribution.
[09:43] <ryanakca> Keybuk: /query please?
[09:44] <Keybuk> ryanakca: hello
[09:47] <salgado> kiko, look what I get when I try to do a "GET -m HEAD" on a url that redirects me to a ftp url and then what I get doing the same GET on the ftp url that the first one redirects me too:
[09:47] <salgado> https://devpad.canonical.com/~andrew/paste/fileqtdkfH.html
[09:47] <kiko> salgado, is GET -m HEAD the same as HEAD?
[09:48] <salgado> wow, there /is/ a HEAD. I didn't know
[09:49] <kiko> so squid is mangling the redirects being returned?
[09:49] <salgado> yes, the method is derived from the program name
[09:49] <kiko> salgado, can you run with a -S?
[09:49] <salgado> https://devpad.canonical.com/~andrew/paste/fileZsxuUF.html is the relevant squid part
[09:49] <salgado> sure
[09:50] <kiko> salgado, you may have to use nc instead of HEAD to see what's really wrong
[09:50] <Znarl> salgado : The squid log didn't show any signs of the error 500 you were getting.  I wonder if that's the problem.
[09:51] <kiko> Znarl, it appears that the redirect line being returned isn't helping
[09:51] <salgado> this is with -S
[09:51] <salgado> https://devpad.canonical.com/~andrew/paste/fileXSJEhw.html
[09:52] <salgado> in the case of the mirror prober, what I actually get is a timeout instead of a connection refused
[09:52] <kiko> salgado, so what is happening is that the proxy is not being used for redirects
[09:52] <kiko> instead HEAD is trying to connect directly
[09:52] <kiko> to the ftp:// server
[09:53] <salgado> aha
[09:53] <kiko> that appears to be a bug in HEAD.
[09:55] <kiko> salgado, you may have to test using some other tool -- NC?
[09:55] <kiko> or just something that doesn't try to be smart about redirects.
[09:55] <salgado> yeah, will try it
[09:57] <salgado> no nc on mawson. :/
[09:58] <kiko> salgado, and telnet?
[09:58] <salgado> how do I set the proxy for telnet to use?
[09:58] <kiko> salgado, you don't -- you do it manually, connecting to the proxy.
[09:58] <kiko> salgado, but perhaps you want to talk to me about the problem first.
[09:59] <salgado> yeah, I don't feel like talking to proxies today
[09:59] <salgado> better talking to you
[09:59] <kiko> so what's up
[09:59] <salgado> basically, the prober failed to probe https://staging.launchpad.net/distros/ubuntu/+mirror/uk-mirror-service-cd/
[10:00] <salgado> but it works just fine for an ftp-only mirror
[10:00] <kiko> is that the only failing mirror right now?
[10:00] <salgado> failing by problems on our side, yes
[10:00] <kiko> cool.
[10:01] <kiko> salgado, can I see the log for that mirror?
[10:01] <kiko> and also, do we do any special handling of redirects in our code?
[10:01] <salgado> yes, we do
[10:04] <Znarl> salgado : UK mirror services do have a http URL for downloading.  
[10:04] <Znarl> I wonder if it's easier just to use that instead of solving the redirection problem.
[10:04] <salgado> really?
[10:04] <salgado> one which doesn't redirect?
[10:05] <Znarl> Yes
[10:06] <Znarl> I thought they did, let me try to find it again.  :/
[10:12] <Znarl> Oh, only their archives they're not redirecting.
[10:13] <Znarl> salgado : I suspect they're always redirecting for releases because of the >2gig issue.  I could request they not redirect within releases.ubuntu.com too as there are no >2gig files on releases.
[10:14] <Znarl> salgado : Would that help you at all?
[10:15] <kiko> Znarl, yes.
[10:15] <salgado> Znarl, that would workaround the problem, but I'd rather find what the real problem is and fix it because I can remember there are other mirrors which do the same
[10:15] <kiko> salgado, it could help deprioritize the issue though
[10:16] <Keybuk> kiko: random question; is there any way of committing to a Launchpad mirror of a bzr archive?
[10:16] <kiko> Keybuk, hmmm, not that I'm aware of. that would be rather weird. though perhaps if you are the bazaar importer...
[10:16] <Keybuk> https://launchpad.net/people/keybuk/+branch/grepmap/main
[10:16] <Keybuk> is what I'm thinking of
[10:16] <Keybuk> it's a mirror of a branch on rookery
[10:17] <Keybuk> but I want to remove that, and just commit to the bzr repository in launchpad instead
[10:17] <Znarl> salgado, kiko : Ok, I'll email them and ask they stop the redirections.
[10:17] <salgado> Znarl, I'll file a bug about the issue, but for now, if you could convice them to not redirect anything under their /releases.ubuntu.com it'd be great
[10:17] <kiko> Znarl, rock on.
[10:17] <salgado> great
[10:18] <kiko> Keybuk, and use local bound branches?
[10:18] <Keybuk> kiko: right, or just pull/push as usual
[10:18] <kiko> Keybuk, it's a good question. let me check one thing.
[10:19] <Keybuk> the only reason it was a mirror, btw, was that I did that before LP supported hosting bzr branches natively
[10:20] <kiko> yeah
[10:23] <kiko> Keybuk, I think you can actually
[10:23] <Keybuk> and now, obviously, I'd like to not push the archive to rookery -- but straight to LP
[10:24] <kiko> Keybuk, have you tried just changing the branch URL?
[10:24] <kiko> oh hmmm
[10:25] <salgado> dammit
[10:25] <kiko> salgado?
[10:25] <salgado> it doesn't seem to be caused by the redirect
[10:25] <Keybuk> quest grepmap% bzr push sftp://keybuk@bazaar.launchpad.net/~keybuk/grepmap/main
[10:25] <Keybuk> bzr: ERROR: File exists: u'/~keybuk/grepmap/main': mkdir failed: unable to mkdir
[10:25] <Keybuk> is what I get when I try to just push over the top
[10:25] <Keybuk> kiko: I tried removing it, I got "Required input is missing"
[10:29] <kiko> Keybuk, why would bzr be trying to mkdir that though?
[10:29] <Keybuk> kiko: who knows?
[10:33] <gkirk> hi, i've set up a product on launchpad, pointed to svn url for trunk, now not sure how to access .po files for translation
[10:33] <gkirk> is it because Bazaar status: Testing - i need to wait for a review or something?
[10:33] <kiko> Znarl, could it be that they are throttling our IP?
[10:34] <gkirk> i did try and find the answer to this question on the site
[10:34] <kiko> gkirk, okay. you need to upload your translations manually, for now.
[10:35] <kiko> we are not doing the automatic collection of translations from source control
[10:35] <kiko> we do do it for packaged versions of software in Ubuntu
[10:35] <gkirk> kiko: ok, will be glad to - how do i do that?
[10:36] <kiko> gkirk, what product is this for?
[10:37] <gkirk> i click on the translations link, and it just says there are no translations available, but no link to upload
[10:37] <gkirk> https://launchpad.net/products/plonecomments
[10:37] <kiko> ah.
[10:37] <kiko> gkirk, just click on "trunk" there.
[10:40] <gkirk> kiko: i am not seeing an option there
[10:41] <kiko> gkirk, really? are you logged in? can I see a screenshot?
[10:41] <gkirk> i have already filled in svn info
[10:41] <gkirk> what should i see?
[10:42] <kiko> To set up Plone Comments for translation in Rosetta, you need to upload the PO template for one of its product series.
[10:42] <kiko>     * trunk
[10:42] <kiko> Or you may want to link the product to a distribution package set up for translation, if there is one, so their translations can be shared.
[10:42] <gkirk> i saw that before i clicked on Edit Source and filled in svn info
[10:43] <kiko> and now what do you see on https://launchpad.net/products/plonecomments/+translations ?
[10:45] <gkirk> Translation templates
[10:45] <gkirk> No translatable templates available
[10:45] <gkirk> trunk does not have any strings to be translated through Rosetta. If you think this is incorrect, please send an email to the Rosetta mailing list or file a bug in our bug tracking system so we can look into the problem. Thanks.
[10:46] <gkirk> kiko: ah, finally noticed the Upload Translations link
[10:47] <gkirk> it wasn't obvious to me 
[10:47] <gkirk> my eyes are drawn to the middle of the page
[10:52] <kiko> gkirk, I'm not quite sure what page you were looking at
[11:37] <lucasvo> is RDF metadata from launchpad something like doap?
[11:38] <kiko> yes
[11:54] <gkirk> kiko https://launchpad.net/products/plonecomments/trunk/+translations
[11:55] <kiko> gkirk, ah, for the trunk series.
[11:55] <kiko> gkirk, I was talking about
[11:55] <kiko> https://launchpad.net/products/plonecomments/+translations
[11:55] <kiko> note the trunk in there
[11:55] <gkirk> kiko: k
[11:56] <gkirk> i do find the UI a bit confusing to navigate
[11:56] <gkirk> tx for your help