[12:12] <lbrown> I just created a test product and project and would now like to delete them. How to?
[12:52] <hugelmopf> the question above ^^ arose from a thread on kubuntu-devel, so if you want, you can answer there.
[03:47] <mpt> Burgwork, when I'm "away (zzz)", I'm usually not around :-) What can I do for you?
[05:18] <lifeless> staging is back up
[05:31] <mpt> Gooooooooooooooooooooood afternoon Launchpadders!
[05:31] <mpt> It's a lovely afternoon to be inside working
[05:38] <lifeless> stub: ping
[06:08] <stub> lifeless: pong
[06:13] <lifeless> spivs requested update to the authserver - is it done ?
[06:13] <lifeless> that is, are we running the correct .tac now ?
[06:39] <mpt> That is truly, truly messed up
[06:43] <mpt> spiv, got a few minutes?
[06:47] <spiv> mpt: sure
[06:51] <mpt> spiv, go to http://localhost:8086/products?exact_name=1&text=firefox
[06:51] <mpt> See how it redirects you to /products/firefox
[06:52] <mpt> Now in templates/products-index.pt, delete the "@@+portlet-details" line near the beginning
[06:52] <mpt> then try that URL again
[06:53] <mpt> The way the search behaves is dependent on the presence or absence of a portlet!
[06:53] <mpt> I don't understand how that's even possible.
[06:54] <mpt> Any ideas?
[06:54] <stub> lifeless: no
[06:54] <stub> (unless spiv did it)
[06:54] <lifeless> spiv: ^^
[06:55] <spiv> No, I haven't done it.
[06:56] <stub> I'll cherry pic r3403 now
[06:57] <spiv> mpt: I get the same behaviour with or without the portlet.
[06:57] <spiv> mpt: (except for the presence or absence of the portlet, of course...)
[06:58] <jamesh> mpt: with the portlet, the redirect reads "Location: http://localhost:8086/products/firefox"
[06:59] <jamesh> mpt: without, it just reads "Location: firefox"
[06:59] <mpt> exactly
[06:59] <jamesh> mpt: I'd fix ProductSetView to do redirect(canonical_url(product)) rather than redirect(product.name)
[07:00] <mpt> that looks promising
[07:02] <mpt> NoCanonicalUrl: No url for u'firefox' because u'firefox' broke the chain.
[07:02] <mpt> oh, my mistake
[07:02] <mpt> yes! thanks, jamesh 
[07:02] <mpt> and thanks for your time spiv
[07:09] <stub> authserver is now broken
[07:10] <stub>           File "/srv/authserver.ubuntu.com/launchpad/lib/twi sted/internet/threads.py", line 25, in _putResultInDeferred
[07:10] <stub>             result = f(*args, **kwargs)
[07:10] <stub>           File "/srv/authserver.ubuntu.com/launchpad/lib/twi sted/enterprise/adbapi.py", line 372, in _runInteraction
[07:10] <stub>             conn = Connection(self)
[07:10] <stub>           File "/srv/authserver.ubuntu.com/launchpad/lib/twi sted/enterprise/adbapi.py", line 32, in __init__
[07:10] <stub>             self.reconnect()
[07:10] <stub>           File "/srv/authserver.ubuntu.com/launchpad/lib/twi sted/enterprise/adbapi.py", line 69, in reconnect
[07:10] <stub>             self._connection = self._pool.connect()
[07:10] <stub>           File "/srv/authserver.ubuntu.com/launchpad/lib/twi sted/enterprise/adbapi.py", line 342, in connect
[07:10] <stub>             conn = self.dbapi.connect(*self.connargs, **self .connkw)
[07:10] <stub>         psycopg.OperationalError: could not connect to serve r: No such file or directory
[07:10] <stub>                 Is the server running locally and accepting
[07:10] <stub>                 connections on Unix domain socket "/var/run/ postgresql/.s.PGSQL.5432"?
[07:10] <stub> Seems to want to connect to localhost
[07:11] <stub> Ahh bugger
[07:11] <lifeless> I'm ringing spiv now
[07:11] <stub> lifeless: don't worry
[07:11] <spiv> I'm online
[07:11] <jamesh> daemons/authserver.tac doesn't seem to look at config.dbhost
[07:11] <lifeless> spiv: yes, but its urgent :). 
[07:12] <lifeless> spiv: I'm now not ringing you
[07:12] <stub> I'm too used to pulling configs and reversed my rsync line
[07:12] <lifeless> muhhahaha
[07:12] <spiv> jamesh: It does
[07:12] <jamesh> spiv: so it does.  I was looking at an older tree
[07:13] <spiv> jamesh: Yeah, I fixed it yesterday :)
[07:17] <stub> authserver back with new code
[07:20] <lifeless> danke
[07:20] <lifeless> I'll rollout the new sftp and branch puller now
[07:21] <lifeless> wheres the production branch at ? pqm /production/launchpad ?
[07:22] <stub> production/1.57
[07:23] <lifeless> thats the built config ?
[07:23] <stub> Oh... ~pqm/production/launchpad
[07:23] <lifeless> thanks ;)
[07:24] <stub> I just fixed a config bug in the built tree
[07:24] <stub> (which I will now commit - looks like both me and spiv fixed a config bug and both landed breaking the config)
[07:25] <lifeless> btw, this rsyc recipe might be nicer.
[07:25] <lifeless> -rltD --delete-after
[07:25] <lifeless> that will avoid owner, group and permission bits
[07:26] <lifeless> can add p to set permission
[07:26] <lifeless> want to avoid owner and group because we have different users and groups between these machines
[07:29] <stub> We want permission for the executable flag
[07:29] <lifeless> stub: ok so -rltpD
[07:40] <lifeless> spiv: can you tell me if the authservers logs look like its getting good branch status feedback ?
[07:41] <spiv> lifeless: Looks good.
[07:41] <spiv> lifeless: There are POSTs to /branch/ getting 200 responses.
[07:41] <lifeless> woo
[07:42] <spiv> So I'd say it's working.
[07:42] <lifeless> sweet
[07:42] <lifeless> the 'update in last day' filter is now active.
[07:43] <lifeless> 89 branches to pull on the next run
[07:43] <lifeless> hmm, mirrorFailed call failed.
[07:44] <spiv> lifeless: I see a traceback for that
[07:44] <lifeless> yup
[07:44] <lifeless> https://chinstrap.ubuntu.com/~dsilvers/paste/file8Tc0TH.html
[07:44] <spiv> lifeless: https://chinstrap.ubuntu.com/~dsilvers/paste/filek05M3u.html
[07:45] <lifeless> meh
[07:46] <spiv> I'm looking
[07:46] <lifeless> danke
[07:51] <jamesh> lifeless: looks like the reason needs to be converted to an 8-bit string before being passed to sqlvalues
[07:52] <lifeless> fair enough. I think thats the authservers problem - agreed ?
[07:52] <jamesh> the webapp pyscopgda converts unicode query strings to UTF-8, iirc
[07:52] <jamesh> and the authserver isn't using that
[07:52] <lifeless> yeah
[07:54] <spiv> Yeah, the authserver has it's own "utf8" function, which is a bit nasty.
[07:59] <stub> I want to move to psycopg2 sometime soonish, which should accept Unicode queries natively. So a quick fix might be best.
[08:01] <spiv> I have a fix: https://chinstrap.ubuntu.com/~dsilvers/paste/fileA4xKak.html
[08:02] <lifeless> +1
[08:02] <lifeless> erm, r=lifeless
[08:02] <spiv> lifeless: :)
[08:04] <spiv> stub: That would be very nice.
[08:12] <mpt> According to <http://www.zope.org/Control_Panel/Products/PageTemplates/Help/tal-repeat.stx>, "python:repeat['bugtask'] .number" should be the same as "repeat/bugtask/number"
[08:13] <mpt> The latter works, but the former produces "&lt;bound method Iterator.number of &lt;zope.tales.tales.Iterator object at.."
[08:14] <lifeless> the latter is much nicer and we should always use it
[08:15] <spiv> lifeless: ?
[08:15] <spiv> lifeless: in general python: expressions in tal should be discouraged.
[08:15] <mpt> lifeless, but I'm concatenating it to a string
[08:15] <lifeless> spiv: tal - using python: should be discouraged
[08:15] <mpt> because (X)HTML IDs can't start with numbers
[08:15] <jamesh> mpt: maybe you want repeat['bugtask'] .number()?
[08:15] <spiv> lifeless: Oh, my brain forgot what latter and former meant, ignore me :)
[08:16] <lifeless>  /ignore spiv
[08:16] <lifeless> oh thats not what you meant ? :)
[08:16] <spiv> lifeless: In the meantime, I've quick-and-dirty fixed the production authserver code with that fix.
[08:16] <spiv> (While waiting for my mail to escape my ISP and reach PQM)
[08:16] <lifeless> thank you
[08:16] <mpt> jamesh, that works
[08:17] <mpt> thanks
[08:17] <lifeless> nice, we're down to 31 to mirror
[08:18] <jamesh> mpt: alternatively, sometimes using a tal:define can simplify things
[08:20] <lifeless> jamesh: spiv: its live
[08:20] <lifeless> https://launchpad.net/people/mrchoke/+branch/thunderbird-thai-l10n/release-1.5
[08:21] <lifeless> great work guys
[08:21] <jamesh> cool
[08:21] <lifeless> we should probably hook up branch notifications on this next
[08:22] <lifeless> so that when your branch starts failing you get an email
[08:22] <lifeless> perhaps with debounce
[08:22] <lifeless> after 3 failures you get an eamil
[08:23] <lifeless> and 
[08:23] <lifeless> https://launchpad.net/people/salgado/+branch/launchpad/proper-signup-workflow
[08:25] <spiv> lifeless: Sweet
[09:35] <carlos> morning
[09:36] <lifeless> moin
[09:40] <mpt> BjornT, ping
[09:40] <BjornT> pong
[09:41] <mpt> BjornT, I thought my MaloneSimplifications branch would end up in your queue, but lifeless reviewed it first ... Would you be able to fix those mysterious test failures sometime soon? :-)
[09:41] <lifeless> mpt: BjornT already has stuff queued, I didn't. So in the interests of latency I reviewed it
[09:42] <mpt> that's fine
[09:42] <mpt> thanks for the review, anyway
[09:42] <BjornT> mpt: if you want me to review something, you can put it in my queue directly. i'll take a look at the test failures today.
[09:43] <mpt> thanks!
[09:44] <lifeless> mpt: note that unless there is a specific reason to give it to a named reviewer, the general queue is the place to put it
[09:44] <ddaa> dudes
[09:44] <ddaa> branch puller is _still_ offline
[09:44] <lifeless> tis not
[09:44] <ddaa> (or looks like it is)
[09:44] <lifeless> ddaa: look here for instance
[09:44] <lifeless> https://launchpad.net/people/salgado/+branch/launchpad/proper-signup-workflow
[09:44] <lifeless> or
[09:44] <ddaa> lifeless: I have not received any error mail from it
[09:44] <lifeless> https://launchpad.net/people/mrchoke/+branch/thunderbird-thai-l10n/release-1.5
[09:44] <lifeless> ddaa: GOOD
[09:45] <lifeless> you should never recieve error mail from it
[09:46] <ddaa> Mh... okay look like 'tis working indeed
[09:46] <ddaa> good work, thanks
[09:46] <ddaa> ddaa.fork()
[09:46] <ddaa> ENOTGOD
[09:46] <mpt> lifeless, the only reason was that Bjorn wrote the Malone mail code so would most easily work out why my test wasn't working, but that wasn't a particularly strong reason if another reviewer could do the same
[09:46] <ddaa> Grah
[09:47] <ddaa> need another myself to put the new timestamps in the UI
[09:47] <lifeless> mpt: so there are two things here - one is 'help me fix my branch'. Another is 'please review'.
[09:47] <lifeless> mpt: dont conflate them :)
[09:48] <mpt> dude, that's why I brought up this issue yesterday :-P - "should I conflate them or not"
[09:48] <mpt> anyway
[09:49] <ddaa> lifeless: will it send errormail for unexpected problems? That is anything but BranchNotFound and network failures?
[09:53] <lifeless> mpt: you asked if there was an issue with the person that helps you being the reviewer
[09:53] <lifeless> mpt: very different question :)
[09:53] <lifeless> ddaa: ues, if it fails to handle a problem it will quit and cron will maili
[09:54] <ddaa> nice
[09:56] <ddaa> not being able to delete stuff like https://launchpad.net/products/0.38
[09:56] <ddaa> or https://launchpad.net/products/socnetv/+series/zohaib
[09:56] <ddaa> is utter nonsense
[10:05] <dilys> Merge to devel/launchpad/: [Trivial]  Production discovered corrections to the branch puller. (r3411: Robert Collins)
[10:11] <lifeless> ddaa: this is what we had planned all the way back in sao carlos
[10:11] <lifeless> its nice to have it finally live
[10:17] <mpt> ddaa, shall I assign bug 38349 to you, then?
[10:17] <Ubugtu> Malone bug 38349 in launchpad "Can't delete a product you created" [Normal,Unconfirmed]  http://launchpad.net/bugs/38349
[10:18] <lifeless> well, deletion != approved in theory last I heard ;)
[10:19] <mpt> ddaa, how does https://launchpad.net/products/0.38 exist anyway? I thought IDs had to start with a letter, not a number 
[10:20] <lifeless> mpt: thats a series, series can start with numbers because they are version numbers
[10:22] <mpt> But that's just the thing, lifeless, it's not a series
[10:22] <mpt> It was registered as a product by mistake
[10:23] <lifeless> mpt: oh. my confusion is supreme. sorry
[10:23] <mpt> And /products/+new says "At least one lowercase letter or number, followed by letters, dots, hyphens or plusses..."
[10:23] <mpt> oh, "or number"
[10:24] <mpt> fine, I can't read
[10:26] <ddaa> mpt: please don't
[10:26] <ddaa> I'm really not supposed to do that, and I have plenty of important things on my plate
[10:26] <ddaa> I just cannot bear the daily churn of people asking for deletion of dud objects anymore
[10:27] <ddaa> so I'll do it for bitching rights
[10:29] <ddaa> BTW, who is in charge of moving dud objects out of the way nowadays?
[10:30] <ddaa> It looks like you need admin privs
[10:42] <stub> lifeless: do you know why 'make build' fails on vostok but works everywhere else? I don't see why it would be any different.
[10:43] <lifeless> stub: nope
[10:43] <lifeless> stub: but its the tar buildings stuff
[10:43] <lifeless> stub: which I see no need for during 'build' - thats a 'dist' responsibility
[10:44] <stub> Yer - we shouldn't be using that Makefile or any of the generation code. But it sounds like a missing dependency or something.
[10:45] <lifeless> why not ?
[10:45] <lifeless> Make is a good interface for these things
[10:45] <lifeless> well understood, simple for simple tasks - and this is.
[10:47] <stub> Because the Makefile serves a single perpose - to generate releases that people are supposed to use. It changes at my whim, and there are no guarantees that the code works anywhere except my laptop (indeed - I think the tests don't pass if you build it on a 64 bit architecture due to issues with the upstream code)
[10:47] <stub> We should be using the generated python code, not generating the python code.
[10:47] <lifeless> stub: well the rocketfuel policy is that we need:
[10:47] <lifeless> make build
[10:48] <lifeless> make check
[10:48] <lifeless> to 'work'
[10:48] <lifeless> for everything in sourcecode/
[10:48] <stub> Yup. pytz shouldn't have been landed the way it was.
[10:48] <stub> Its a bug
[10:48] <lifeless> how should it be? can we fix that ?
[10:49] <lifeless> should it just be a package ?
[10:49] <stub> It will be fixed when we land Zope 3.2, as pytz is included in that and we can drop the external dependency entirely.
[10:49] <lifeless> ok
[10:50] <lifeless> in which case, just dont run 'make build' on vostok - and dont use --delete when rsyncing to it ;)
[11:22] <dilys> Merge to devel/launchpad/: [r=lifeless]  Ensure that the authserver's /branch/ XML-RPC methods don't blow up on unicode. (r3412: Andrew Bennetts)
[11:25] <ddaa> mpt: what is this "aqua" sort of sitemap them on rocketfuel?
[11:26] <ddaa> mpt: the horizontal gray lines really scream "the web designer thought it would be cool to make it look like MacOS"
[11:28] <lifeless> stub: oh, r3412 is also important - its in production now on the authserver
[11:29] <ddaa> How comes registry is not a celebrity????
[11:31] <ddaa> mpt_: what I said to mpt before ^^
[11:48] <mpt_> ddaa, Mark's choice
[12:00] <mpt> ddaa, it's Mac OS X 10.2 specifically
[12:00] <mpt> In 10.1 the stripes are darker, in 10.3 they're lighter, and in 10.4 the menu bar doesn't have stripes
[12:00] <ddaa> okay, then I guess it qualifies as vintage ;)
[12:03] <ddaa> mh, what's the passwd of mark@hdb.com in sampledata, "test" does not work...
[12:04] <ddaa> works when I spell hbd properly
[12:13] <ddaa> duh: https://staging.ubuntu.com/people/logicalfaith
[12:16] <ddaa> mpt: it looks like neither the title nor the summary of a team is shown on the team's page anymore
[01:51] <ddaa> heya, who's up for a quasi-trivial review?
[01:52] <ddaa> https://chinstrap.ubuntu.com/~dsilvers/paste/fileMJv5Hf.html
[01:56] <BjornT> ddaa: doesn't celebs.registry == registry work?
[01:57] <ddaa> BjornT: maybe it does, I just conformed to the existent code
[01:58] <ddaa> obviously it should also check that the appropriate interface is provided or something, but I think that is out of scope
[01:58] <BjornT> yeah, i think that it didn't work before. it should work now though, if it doesn't it's a bug.
[01:58] <ddaa> you tell me whether you want me to try and change that before merging
[01:59] <BjornT> ddaa: since you conform to the surrounding code, r=bjornt without the change. it's better to change all at once later.
[02:02] <ddaa> thank you
[02:21] <salgado> stub, around?
[02:21] <carlos> stub: hi, around?
[02:21] <carlos> :-P
[02:26] <dilys> Merge to devel/launchpad/: [trivial]  add missing security declaration for the bugnotification db user. (r3413: Bjorn Tillenius)
[02:29] <matsubara> ddaa: ping
[02:30] <ddaa> matsubara: pong
[02:30] <niemeyer> Mornings!
[02:30] <ddaa> hey niemeyer
[02:30] <matsubara> ddaa: in interfaces/branch.py the url field is required=True but the description says otherwise. Which one is correct?
[02:32] <ddaa> Both
[02:32] <ddaa> When the URL field is present in the form, it is required
[02:32] <ddaa> but the database value may be NULL
[02:32] <seb128> hi
[02:32] <seb128> carlos: around?
[02:32] <carlos> seb128: hi
[02:32] <seb128> hey carlos :)
[02:32] <niemeyer> Hey ddaa, what's up?
[02:32] <seb128> https://launchpad.net/distros/ubuntu/+source/gaim/+bug/38330/+index
[02:32] <Ubugtu> Malone bug 38330 in gaim "Rosetta Translations" [Normal,Rejected]  
[02:33] <lifeless> ddaa: *cough* actually it currently works with no url to setup hosted branches
[02:33] <ddaa> matsubara: there is some logic in the edit form to remove the url widget if it's null in the database
[02:33] <seb128> carlos: gaim upstream have some interrogations on how rosetta works on the upstream collaboration plan, etc ... is there an FAQ or some page we can point about that?
[02:33] <lifeless> ddaa: +newbranch -> shows the field but lets it be left blank
[02:33] <ddaa> lifeless: it should not... the interface says it's required!
[02:35] <matsubara> that's my point.
[02:35] <lifeless> ddaa: its how I created a hosted branch
[02:35] <lifeless> ddaa: and the web form says 'leave this blank to create a hosted branch'
[02:35] <carlos> seb128: Hmm, we should have that in our FAQ page, but I don't see anything there
[02:35] <carlos> seb128: could you redirect them to jordi?
[02:35] <carlos> jordi: and please, after answering their questions, update the FAQ, ok?
[02:35] <matsubara> ddaa: in http://localhost:8086/products/firefox/+addbranch it says the branch url is required but you can leave it blank
[02:36] <lifeless> night all
[02:36] <ddaa> I think at some point we decided not to allow registering hosted branch in that way, for some reason.
[02:36] <seb128> carlos: sorry, thank you
[02:36] <ddaa> But honestly I do not remember why.
[02:36] <carlos> seb128: why 'sorry'?
[02:36] <carlos> :-)
[02:36] <seb128> jordi: https://launchpad.net/distros/ubuntu/+source/gaim/+bug/38330/+index ... if you could comment would be nice :)
[02:36] <Ubugtu> Malone bug 38330 in gaim "Rosetta Translations" [Normal,Rejected]  
[02:36] <seb128> carlos: hum, dunno, I was probably thinking to something else :p
[02:36] <seb128> carlos: thank you so :)
[02:37] <carlos> seb128: ;-)
[02:37] <ddaa> matsubara: okay, let's consider one can created a hosted branch in that way, I see no reason to prevent that.
[02:37] <ddaa> matsubara: then the field must be optional in +addbranch
[02:37] <ddaa> but required in +editbranch
[02:38] <ddaa> because otherwise it would allow turning a pull branch into a hosted branch
[02:39] <ddaa> which we decided not to allow because it involves a lot of extra complications
[02:39] <ddaa> matsubara: good luck unraveling that :)
[02:39] <lifeless> ddaa: does it ?
[02:39] <matsubara> ddaa: ok
[02:40] <lifeless> bah, got sucked in .
[02:40] <lifeless> Good Night!
[02:40] <ddaa> lifeless: good night, you can ask on the mailing list if you care. I think that would probably deserve a spec.
[02:41] <ddaa> matsubara: conversely, if you prevent registering hosted branches from the web UI, then it all becomes much simpler: when the URL widget is present, it's required, and when URL is NULL, the widget is not present.
[02:42] <ddaa> I really do not care on way or another. Allowing registering push branches through the web is marginally simpler, but also give more rope to hang oneself with.
[02:43] <ddaa> (simpler in terms of user experience)
[02:44] <salgado> hey Kinnison, around?
[02:59] <stub> carlos: pong
[03:00] <stub> salgado: pong
[03:01] <carlos> stub: could you execute this on production?
[03:01] <carlos> UPDATE translationimportqueueentry set status=6 WHERE path like 'debian/tmp/usr/share/doc/gettext/examples/hello%' OR path like 'gettext-tools/examples/hello%';
[03:01] <carlos> stub: that will set as blocked more than 500 entries
[03:01] <carlos> and will save me some time
[03:03] <stub> carlos: 1134 items - sound right?
[03:03] <carlos> stub: yes
[03:03] <stub> carlos: Done
[03:03] <carlos> stub: thanks
[03:04] <salgado> stub, I got https://chinstrap.ubuntu.com/~dsilvers/paste/fileikNqb5.html when trying to do a make schema on dogfood
[03:06] <stub> salgado: You need to set up an empty database?
[03:07] <stub> export PGUSER=postgres
[03:07] <salgado> stub, well, I need the session database and all the users specified in security.cfg
[03:07] <stub> That should set the default user to one with the rights you need
[03:07] <salgado> but the launchpad user is a superuser. shouldn't that be enough?
[03:08] <stub> launchpad user is not a superuser
[03:09] <stub> its just a standard user with rights on lots of tables
[03:09] <salgado> oh, right. I created it as a superuser but the make schema drops it and creates it again?
[03:09] <stub> Hmm... no idea. Creating it as a superuser isn't good though as you won't catch permission issues.
[03:10] <stub> It won't drop users so not sure what is going on
[03:28] <salgado> stub, can you have a look at https://chinstrap.ubuntu.com/~dsilvers/paste/fileYIBWyZ.html and see if it's okay?
[03:30] <bradb> BjornT: Should a browser view class that is used in the browser:generalform directive also inherit from GFV? Seems redundant, but just thought I'd confirm.
[03:30] <stub> salgado: Looks good. Add a comment to comments.sql and it is approved as patch-40-47-0.sql
[03:32] <salgado> stub, adding the comment right now. thanks!
[03:33] <salgado> stub, what do you think of renaming it to date_consumed?
[03:33] <salgado> (and make it a timestamp col, obviously)
[03:37] <BjornT> bradb: i think it should, it makes things clearer.
[03:37] <bradb> BjornT: ok, thanks
[03:43] <dilys> Merge to devel/launchpad/: [r=BjornT]  create registry celebrity (r3414: David Allouche)
[03:55] <stub> salgado: Sounds good. Means we can clear them out occasionally
[03:56] <salgado> exactly
[03:57] <bradb> jamesh: ping
[05:18] <jcole> (08:00:13) jcole: sladen: simple question (not), at my company, we use http://linuxcoe.sourceforge.net/ to net-install various distros... right now, we're trying to figure out a way so people can contribute their own packages...
[05:18] <jcole> (08:00:16) jcole: sladen: we were thinking about something along the lines of a user only contributing a source package, and then have an auto-build system that creates binary packages for the various distros... any suggestions?
[05:18] <jcole> (08:07:19) sladen: jcole: in theory you can create deriviatives very easily under launchpad.  You could ask on #launchpad what the current status of that is
[05:19] <jcole> (08:08:48) jcole: sladen: so, launchpad is the closest tool for the type of thing we are looking for?
[05:23] <jcole> are there any open source launchpad equivs out there?
[05:44] <carlos> jcole: don't think so, well, outside what Debian uses atm but I don't think it's easy to setup
[05:44] <elmo> hey, if people could not send anything to pqm for a bit, that'd be nice
[05:45] <carlos> elmo: there is a way to disable pqm, the requests are queued but will not be handled until you enable it again
[05:46] <carlos> elmo: but don't know how to do it
[05:46] <elmo> carlos: yeah, but I'm lazy and don't know that way
[05:46] <carlos> elmo: ;-)
[05:46] <carlos> elmo: I guess the easier way is just disable its cron job :-P
[05:46] <elmo> oh, that's a good idea
[05:47] <elmo> done
[05:47] <carlos> elmo: please, remember to enable it again when you are done...
[06:14] <elmo> done - pqm is back up
[06:21] <VobiscuM> the launchpad is property software ?!
[06:23] <carlos> VobiscuM: If you mean if it's closed source, yes, it is
[06:24] <carlos> VobiscuM: Take a look at https://launchpad.net/faq (Is Launchpad open source? Will it be?)
[06:24] <carlos> you have more information there
[06:25] <VobiscuM> carlos: thank you
[06:26] <carlos> you are welcome
[06:58] <carlos> jordi: hi, around?
[07:23] <carlos> see you later
[07:25] <mdz> how can I reject or otherwise get rid of this task?  
[07:25] <mdz> https://launchpad.net/products/linux/+bug/34939/+editstatus
[07:25] <Ubugtu> Malone bug 34939 in linux "Dapper Install Hangs on "Configuring Apt"" [Unknown,Unknown]  
[07:26] <mdz> the bug has nothing to do with linux
[07:28] <dilys> Merge to devel/launchpad/: [r=jamesh]  implement security teams (r3415: Brad Bollenbach)
[07:49] <mdz> bradb: any suggestion?
[07:50] <kiko> mdz?
 how can I reject or otherwise get rid of this task?  
 https://launchpad.net/products/linux/+bug/34939/+editstatus
[07:51] <Ubugtu> Malone bug 34939 in linux "Dapper Install Hangs on "Configuring Apt"" [Unknown,Unknown]  
 the bug has nothing to do with linux
[08:07] <bradb> mdz: Just set the status to rejected
[08:08] <bradb> Deleting of tasks is currently not possible
[08:08] <kiko> I think that bug's task doesn't allow changing of status
[08:08] <kiko> argh
[08:08] <kiko> I said it
[08:09] <bradb> kiko: You're thinking too much like a DBA, and not enough like a user :)
[08:09] <kiko> well, the concept is leaky
[08:10] <bradb> but, since this bug is locked, eh, yeah, hm
[08:10] <bradb> s/bug/bug's status/
[08:12] <bradb> +editstatus is a thoroughly weird page for that bug. Comment on Change widget and all.
[08:15] <bradb> mdz: For now I'll file a bug about it. It shouldn't affect ubuntu devs looking at Ubuntu bug listings
[08:20] <bradb> bug 38460
[08:20] <Ubugtu> Malone bug 38460 in malone "Can't "reject" a bug watch" [Normal,Unconfirmed]  http://launchpad.net/bugs/38460
[08:20] <kiko> that's dupish
[08:23] <bradb> I couldn't find anything matching it
[08:26] <bradb> kiko: What's the story with bug dates then?
[08:29] <kiko> bradb, I wanted to look into callsites further, to see what /I/ thought of the issue, but it seems that our code makes properties a better choice
[08:31] <kiko> bradb, congratulations on landing security teams, btw
[08:31] <bradb> thanks. had to fight a heisenfailure for a while, but nothing beats raw willpower.
[08:32] <bradb> kiko: Can I start on the dates change then?
[08:32] <kiko> I'm adding some sampledata and having the same sort of problem :-(
[08:32] <kiko> bradb, do you have a next best item to work on today? I'd be able to confirm tomorrow on what I think we should do
[08:33] <kiko> early tomorrow for you even
[08:34] <bradb> I can think of some silly bugs to fix, but the other major priority, I think, was the subscriptions stuff.
[08:35] <bradb> I was replying to mpt's email about the portlet, but my brain is hurting trying to do better than his proposal.
[08:37] <bradb> (One of the silly bugs I was referring to was bug 34224)
[08:37] <Ubugtu> Malone bug 34224 in malone "wrong bug number counts" [Normal,Confirmed]  http://launchpad.net/bugs/34224
[08:39] <kiko> I don't even like that portlet with counts, agreed it's silly
[08:39] <kiko> why don t you drop binarypackagename today?
[08:39] <bradb> hey, that's an idea
[08:39] <bradb> sure
[08:39] <kiko> salgado, I need some help with a mirror management fix?
[08:39] <kiko> cool!
[08:39] <kiko> can you spend 5 minutes with me salgado?
[08:40] <salgado> sure
[08:42] <salgado> whant me to go down?
[08:42] <kiko> well
[08:42] <kiko> I'll paste the error to you
[08:42] <kiko> salgado, I did some changes to sampledata, ftr
[08:44] <kiko> salgado, pasted.
[08:45] <salgado> did you add a new entry in securebinarypackagepublishinghistory?
[08:45] <kiko> yes. 2 or 3 entries IIRC.
[08:46] <salgado> do any of them for the same distro release in which that pmount_0.1-1_all.deb is published?
[08:47] <salgado> if yes, and if any of them has a more recent datepublished than the pmount's one, then the failure is expected
[08:47] <salgado> s/distro release/distroarchrelease/
[08:48] <kiko> salgado, yes. but what confuses me is why pmount has disappeared.
[08:48] <kiko> why don't we end up with a 3-item list?
[08:48] <salgado> because there's another one published more recently
[08:49] <kiko> okay. so I can just accept what changed? that test is kinda confusing..
[08:49] <salgado> yes, you can accept that
[08:50] <salgado> why is it confusing?
[08:50] <kiko> sank you
[08:51] <kiko> salgado, well, perhaps because I don't understand how it works.
[09:04] <kiko> salgado, didn't work (didn't actually change anything)
[09:05] <kiko> oh, nm
[09:05] <kiko> found what I was doing wrong
[09:05] <kiko> Kinnison, why is there no constraint datecreated < datepublished? :)
[09:15] <kiko> salgado, it worked, thanks.
[09:15] <kiko> salgado, now, what I'd like to talk about is that TeamParticipation caching thing we discussed previously.
[09:17] <salgado> we didn't actually discuss it, though
[09:17] <kiko> SteveA had said he had talked to you about how to fix it
[09:17] <kiko> was he on crack?
[09:18] <salgado> I don't remember talking with anybody about this. and during the last sprint I wasn't as fucked up as I was in Matar, so I think I'd remember it
[09:18] <kiko> LOL
[09:18] <kiko> dude that was timeless
[09:18] <kiko> TIMELESS
[09:22] <LarstiQ> why is autotools-dev listed in distro packages for bzr>
[09:23] <LarstiQ> and the breezecom bounty...
[09:31] <bradb> kiko: BTW, are we talking about removing bpn from the db too? Or just the UI at first?
[09:31] <kiko> bradb, both
[09:31] <bradb> ok
[09:32] <kiko> hmmm
[09:32] <kiko> actually, bradb, tell me: when the user enters that information, would we lose it?
[09:32] <bradb> yeah. we'd have only the spn.
[09:32] <kiko> and the bpn would only serve to find the bpn. well, let me think for 2 minutes
[09:33] <bradb> the devs that i've heard from seem to want bpn to go away
[09:34] <bradb> At worst, maybe removing it only from the UI could be a test.
[09:34] <kiko> me too
[09:34] <kiko> removing from the UI only isn't very good because we'd forget to remove the field later and the data would be, well, inconsistent
[09:35] <kiko> bradb, okay, mdz and I think dropping it is fine -- mdz said:
 I have looked at thousands of bugs
 and so far there were approximately 2 where that information was even useful
 there were more cases where it was invalid
 losing the data would be an overall win for bug correctness
[09:35] <bradb> i can imagine
[09:36] <bradb> bpn is mostly dust
[09:36] <kiko> dust in mdz's eyes? 
[09:36] <bradb> heh
[09:36] <mdz> do we have any historical data for the open bug count?
[09:36] <mdz> I am interested to see the effect that our efforts are having
[09:37] <mdz> if not, can we start collecting basic stats somehow?  I suppose we could write a screen scraper if not
[09:37] <kiko> not currently, but we will start working on this as soon as we've got the remaining ubuntu requirements nailed
[09:37] <kiko> we mmmm can
[09:37] <kiko> I'll email stuart, mdz
[09:37] <kiko> and actually
[09:38] <bradb> mdz: We will start collecting dates of interesting things that happening on bugs in the very near future.
[09:38] <kiko> mdz, can you think up a set of counts that you'd like to see historically?
[09:38] <kiko> I can ask stub to plot some sql results on a graph
[09:38] <mdz> kiko: that would make me very happy
[09:38] <mdz> kiko: can I give you some advanced search query strings?
[09:39] <kiko> do you just want counts, mdz?
[09:39] <mdz> kiko: yes
[09:39] <kiko> perhaps just listing what you want is best, because stub is going to do raw SQL anyway
[09:40] <mdz> kiko: ok
[09:41] <mdz> kiko: all confirmed+inprogress, severity>=normal in main
[09:42] <kiko> mdz, email?
[09:42] <mdz> kiko: ok
[09:42] <kiko> thanks.
[09:42] <kiko> salgado, driveby https://chinstrap.ubuntu.com/~dsilvers/paste/fileU5jceJ.html
[09:43] <kiko> salgado, fixes searching for binary packages in a distrorelease to match on binarypackagename as well.
[09:44] <mdz> kiko: sent
[09:56] <salgado> kiko, I've already reviewed that patch, haven't I?
[09:57] <kiko> salgado, no, this is a similar patch, but it fixes searching for binary packages.
[09:59] <carlos> kiko: hi, I have a question for Steve that perhaps you could answer...
[09:59] <salgado> kiko, but that one I reviewed previously contained the same changes in database/distrorelease.py, no?
[09:59] <kiko> salgado, yes, I asked you to ignore it :)
[09:59] <kiko> carlos, yeah?
[09:59] <salgado> kiko, no, you asked me to ignore some part related to arch releases
[10:00] <kiko> salgado, I should have told you to ignore parts related to distroreleases as well then.
[10:00] <carlos> kiko: Could I change SourcePackageName.queryByName to be SourcePackageName.getByName to follow the same API other objects are using? (this is a concrete case, but, in general, could I do that while developing code that use those methods?)
[10:01] <kiko> I was going to ask bradb but he's gone
[10:01] <kiko> mmm
[10:02] <carlos> is not Rosetta code and it's used by others, that's why I'm asking. I usually do that with Rosetta specific objects
[10:02] <kiko> I don't know carlos. are you sure getByName is the standard?
[10:03] <kiko> salgado, do you know how the standard works?
[10:03] <carlos> kiko: well, I don't know if it's the standard, I know it's the most common API we have. I'm not sure if we have such standard...
[10:03] <salgado> I don't know of such a standard either
[10:04] <kiko> carlos, don't change it now, I think, but raise the issue on the ML and we can nail it
[10:06] <carlos> ok
[10:06] <carlos> thanks
[10:06] <kiko> thanks for spotting it though
[10:18] <salgado> kiko, that patch looks fine, btw
[10:18] <kiko> thanks salgado 
[10:27] <dilys> Merge to devel/launchpad/: [trivial]  Add for convenience a newsampledata proxy target to the main Makefile. (r3416: kiko)
[11:13] <kiko> elmo, ping?
[11:16] <dilys> Merge to devel/launchpad/: r=salgado Fix for bug related to bug 30500: searching for binary packages under a distribution release would ignore packages whose names were not good fti fodder -- 'at' for instance. Adds sampledata to verify and test, and fixes the query. (r3417: kiko)
[11:16] <kiko> yay
[11:24] <elmo> kiko: ?
[11:24] <kiko> elmo, I'm reviewing your patches, I was going to ask you something but I think email will be better
[11:25] <elmo> kiko: ok
[11:33] <kiko> o/~ milkshake milkshake o/~
[11:34] <kiko> elmo, it is long. will you have time for a proper reply?
[11:35] <elmo> kiko: sure?
[11:35] <kiko> okay, cool.
[11:39] <lifeless> moin moin
[11:39] <kiko> nihau
[11:40] <lifeless> hola
[11:40] <kiko> oi
[11:40] <lifeless> howzitgoingmate
[11:41] <lifeless> kiko: will you have time to do the two reviews currently on your plate ? I can find another reviewer if needed
[11:41] <kiko> next week yeah
[11:42] <kiko> is that okay?
[11:42] <lifeless> cprov is still away right
[11:42] <lifeless> and they are both related to his work, so it sounds fine to me
[11:42] <kiko> he will be back on monday
[11:42] <kiko> great
[11:43] <lifeless> what do you think of the pending reviews tweaks ?
[11:43] <kiko> very cool
[11:43] <kiko> I was going to commend jamesh but I saw you did already
[11:44] <kiko> elmo, are you okay with me CC:ing launchpad-reviews on my reply?
[11:45] <lifeless> kiko: theres no problem with multiple people saying 'thats cool
[11:45] <lifeless> :)
[11:45] <kiko> true
[11:45] <kiko> I have just been too busy listening to Louis XIV
[11:46] <lifeless> I'm louis the XIV I am ?
[11:46] <kiko> not today
[11:46] <kiko> but in a previous incarnation, who knows?
[11:47] <lifeless> theres a song
[11:47] <lifeless> I'm henry the fifth I am
[11:47] <lifeless> very strange
[11:47] <elmo> kiko: if you like, sure
[11:48] <elmo> the code's going to be in RF one way or the other, no matter how bad it is, I can't hide it
[11:48] <kiko> elmo, sent then
[11:48] <kiko> thanks
[12:01] <kiko> salgado, could you review matsubara's patch for 37885? 
[12:02] <kiko> https://chinstrap.ubuntu.com/~dsilvers/paste/fileU5jceJ.html
[12:03] <matsubara> hey it's not ready yet
[12:03] <kiko> oh
[12:03] <kiko> ok
[12:03] <matsubara> and the link you sent surely isn't my patch