[12:04] <johnl> you read everybody loves eric raymond?
[12:04] <kiko> that's what they all say
[12:05] <kiko> I wrote code with ESR once
[12:05] <johnl> I meant the comic strip
[12:05] <johnl> nevermind.
[12:05] <kiko> I don't think that qualifies as loving him though
[12:05] <kiko> I hear his guns love him
[12:05] <kiko> oh, sorry, I got that backwards
[12:09] <kiko> wtf is bradb
[12:11] <johnl> ?
[12:12] <kiko> johnl?
[12:13] <johnl> 2mins
[12:17] <kiko> Kinnison, sorry for appearing to single you out, I was more concerned that others weren't aware, I hadn't noticed it was a dot-tac apologies etc etc
[12:28] <johnl> ok kiko
[12:29] <Kinnison> kiko: s'okay, it's a good rant to have
[12:29] <Kinnison> kiko: I hate tabs too
[12:29] <johnl> heh
[12:30] <johnl> The key I was registering is a sign only key yes.  I have a sub key for encrypting.
[12:30] <Kinnison> johnl: launchpad won't like sign-only keys
[12:31] <Kinnison> (or it might now, I forget)
[12:31] <kiko> johnl, right, so your bug is a dupe. thanks for helping me out :)
[12:33] <mpt> Kinnison: Am I right in thinking that soyuz-index.pt isn't used any more?
[12:35] <mpt> i.e. we don't have a page in Launchpad any more that says "Soyuz is a Linux Distribution Management System ..."
[12:36] <johnl> hehe, that was easy
[12:36] <johnl> I take it you'll fix to accept sign only keys then?
[12:36] <kiko> yes, we will
[12:36] <kiko> but not today :)
[12:37] <johnl> :)
[12:38] <johnl> launchpad looks great btw.  well done.
[12:38] <kiko> thanks, johnl -- kind words are very much appreciated (we get few of them)
[12:39] <Kinnison> mpt: not a clue
[12:40] <johnl> I marked my bug a dupe of #1972
[12:40] <kiko> johnl, and then I second-stepped you, wow, good work
[12:40] <johnl> I might have seen the dupe myself but the bug #1972 is marked as secret :)
[12:41] <johnl> ok, I just updated the crazy button jumping thing too.  need any more on that while I'm here?
[12:43] <kiko> nope, that's more than enough
[12:43] <kiko> thanks
[12:43] <johnl> sleep for me then.  bye all.
[12:46] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Ensure /etc/source-dependencies, rev launchpad-buildd to v10 (patch-2529: daniel.silverstone@canonical.com)
[12:49] <Kinnison> thanks dilys 
[01:03] <Epix> What is Launchpad written in? Im impressed.
[01:04] <kiko> Epix, in python and zope3
[01:05] <Epix> cool. i was thinking RoR...
[01:11] <Kinnison> Which word is better... controller, coordinator, director or sequencer ?
[01:13] <kiko> it depends on what it is!
[01:17] <Kinnison> Well, its something which controlls, coordinates, directs, or indeed simply sequences, the build tasks
[01:17] <Kinnison> Its job is to run the slave scanner, the queue builder, etc in the right order, making sure they don't contend
[01:17] <Kinnison> because cron isn't good enough
[01:17] <kiko> sequencer maybe?
[01:18] <Kinnison> okay?
[01:18] <kiko> yep
[01:18] <Kinnison> * creating version daniel.silverstone@canonical.com--desktop/launchpad--buildd-task-sequencer--0
[01:20] <Kinnison> twistd .tac files are allowed to read canonical.config aren't they?
[01:21] <kiko> that I don't know
[01:22] <Kinnison> I'm gonna assume I can, since reading the config doesn't imply the db at all
[01:26] <lifeless> mpt: pong
[01:26] <lifeless> ddaa dunno
[01:29] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=BjornT]  fix a regression in the Malone menus where the menus had magically disappeared from many pages, likely during the conversion to the menu framework (patch-2530: brad.bollenbach@canonical.com)
[01:32] <lifeless> Kinnison: buildd-master ?
[01:32] <lifeless> ;)
[01:32] <lifeless> look, you're writing buildbot
[01:34] <Kinnison> pardon?
[01:39] <Kinnison> lifeless: if buildbot can do exactly what I want then I'll use it
[01:39] <Kinnison> lifeless: the four requirements are:
[01:39] <Kinnison> lifeless: 1. every N minutes, run the queue builder
[01:39] <Kinnison> lifeless: 2. as often as possible run the slave scanner
[01:40] <Kinnison> lifeless: enforce 10 second gap between each invocation
[01:40] <Kinnison> lifeless: enforce single invocation at a time
[01:40] <Kinnison> lifeless: 4. if the subprocess exits non-zero, mail the output to the nominated person
[01:40] <Kinnison> sorry, the two enforce's were (together) requirement 3
[01:41] <lifeless> Kinnison: I'm trolling, well 99% trolling
[01:42] <lifeless> Kinnison: its just that the more work you do, the more the architecture is resembling buildbots
[01:43] <Kinnison> It may one day collapse
[01:43] <Kinnison> for now we have too many bizarre requirements
[01:43] <Kinnison> poor thing has an avg 90m build job
[01:46] <bob2> eep
[01:48] <Kinnison> bob2: ?
[01:55] <segfault> that's why thunderbird is better.
[01:55] <segfault> oops
[01:58] <mpt> gneuman: <div class="discreet">Total: <tal:total here /></div> <table ...
[01:58] <bob2> 90m
[01:59] <Kinnison> bob2: oh, acl2
[01:59] <gneuman> mpt thx
[02:04] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Fix slight indentation bug which could cause absorbtion of successful builds to fail with obscure unrelated error (patch-2531: daniel.silverstone@canonical.com)
[02:06] <Kinnison> thanks dilys 
[02:10] <kiko_> hey stub 
[02:10] <stub> yo
[02:10] <kiko_> how's it going on your end?
[02:10] <stub> ok
[02:11] <kiko_> it's harsh on mine
[02:11] <kiko_> miles of reviews before the sun shines
[02:12] <kiko_> why is baz merge so flakey I wonder
[02:12] <stub> One day at a time here. I might have a chance to get back on with LibrarianGarbageCollection, although I havn't finished gardening my inbox yet so that might be wishful thinking
[02:13] <kiko_> my inbox is a disaster
[02:13] <kiko_> it says 515 
[02:13] <kiko_> can you believe it?
[02:13] <stub> What particular problem do you have? Works fine here except for occasional memory issues, and I can work around them (baz library-add foo before baz whatever foo)
[02:13] <kiko_> it gives me spurious conflicts when doing side-ways merges
[02:13] <stub> kiko_: Can I interest you in some mail filters ;)
[02:13] <kiko_> I am considering filtering
[02:13] <kiko_> but I am so much more resposive without it
[02:13] <kiko_> responsive too
[02:14] <kiko_> man
[02:14] <kiko_> I just became a massive fan of baz replay
[02:14] <kiko_> wooo!
[02:14] <kiko_> rock rock rock!
[02:16] <kiko_> stub, any chance of cherry-picking the "return of bug titles" patch?
[02:17] <stub> Which patch?
[02:17] <kiko_> omg
[02:17] <kiko_> is it not fixed yet?
[02:20] <kiko_> stub, can you assign both bugs to matsubara?
[02:20] <kiko_> 2668 and 2669?
[02:21] <kiko_> for some reason I can't edit it.
[02:24] <stub> That would be a bug
[02:25] <kiko_> yeah
[02:25] <kiko_> I am already discussing it with brad
[02:25] <kiko_> stub, do you have access to a tree that still has bug-index.pt?
[02:26] <stub> No, not even in my production-1.34 branch. 
[02:26] <kiko_> darn
[02:28] <stub> The poimport pending queue isn't shrinking. Looks like because the script is dying with that SQL violation later jobs are never getting processed.
[02:28] <stub> Should I fix that, or has someone else already handled it?
[02:29] <kiko_> wel
[02:29] <kiko_> I asked carlos about it
[02:29] <kiko_> he told me that your constraint is correct
[02:29] <kiko_> and that the code is broken
[02:29] <kiko_> so...
[02:31] <stub> I mean make the script log SQL exceptions and continue rather than die
[02:31] <stub> like it does with other non-dangerous exceptions (I think)
[02:31] <kiko_> hmmm
[02:31] <kiko_> maybe
[02:31] <kiko_> maybe
[02:31] <stub> Or I could just randomize the queue order ;)
[02:32] <kiko_> that might be more effective :)
[02:36] <stub> kiko_: Why arn't we setting the rawimportstatus to FAILED if it fails?
[02:37] <kiko_> that I don't know -- perhaps carlos would know better. I'm a bit unfamiliar with the database side of things :-(
[02:44] <mpt> EBEDTIME
[02:45] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Call flush database updates before redirecting to the next pending shipit order, when approving/denying an existing order. (patch-2532: guilherme.salgado@canonical.com)
[02:45] <bob2> mpt has been in brazil for a long time now
[02:47] <stub> He is there until december
[02:47] <kiko_> stub, can you please cherry-pick rocketfuel@canonical.com/launchpad--devel--0--patch-2532
[02:48] <bob2> wow
[02:54] <Kinnison> night all
[02:54] <Kinnison> only 14h day today
[02:56] <kiko_> only 11 for me
[02:56] <kiko_> I'm slacking
[02:58] <kiko_> oh ffs
[02:58] <kiko_> it's ridiculous
[03:01] <kiko_> stub, can you baz replay christian.reis@canonical.com--lozenge/launchpad--devel--0--patch-157 -- ?
[03:01] <kiko_> it's a one-liner that gets us titles back in bugs
[03:02] <kiko_> maybe I should add a test for it
[03:11] <kiko_> stub, do you know if pqm will accept merges from me that contain patchlogs from archives it doesn't know about?
[03:11] <kiko_> or lifeless 
[03:12] <stub> kiko_: Both of those?
[03:13] <stub> ok
[03:16] <kiko_> yeah, both
[03:16] <kiko_> the latter might need to be a replay
[03:16] <kiko_> the former a direct merge
[03:19] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Link bugs more aggressively -- allow bug#XXX and bugno.XXX (patch-2533: christian.reis@canonical.com)
[03:20] <spiv> kiko_: pqm won't mind the patchlogs, so long as star-merging that branch into rocketfuel works ok.
[03:20] <kiko> spiv, that's killer -- I have about 7 patches lined up tonight
[03:20] <spiv> AIUI, anyway :)
[03:30] <kiko> friggin system errors right n left
[03:40] <kiko> stub, btw, is staging still hosed?
[03:40] <kiko> I'm still getting cve errormail
[03:42] <stub> If that was from yesterday, it was because the cve update script was scheduled too early, so the staging updates hadn't completed when it was run
[03:59] <kiko> Date: Wed, 28 Sep 2005 02:45:58 +0100 (BST)
[03:59] <kiko> yeah, ok
[03:59] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: r=kiko Fix bug 2539: The Wiki Base URL showed at Your Wiki Name's page is old. Changed UBUNTU_WIKI_URL from http://www.ubuntulinux.com/wiki/ to https://wiki.ubuntu.com/, and updated sampledata and tests. Patch by Diogo Matsubara <matsubara@async.com.br> (patch-2534: matsubara@async.com.br, christian.reis@canonical.com)
[04:01] <spiv> UBUNTU_WIKI_URL has changed?  We'll need to make sure we update the authserver for the next rollout, then.
[04:02] <spiv> (or else wiki logins will be slightly broken)
[04:03] <kiko> spiv, it's changed to the correct value -- so yes.
[04:03] <kiko> are there no tests for the authserver? :-)
[04:03] <spiv> kiko: There are.  It uses the same constant.
[04:03] <spiv> kiko: But I don't think it's usually updated with the weekly rollouts.
[04:04] <kiko> oh, I see
[04:39] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Commit publisher transaction before apt-ftparchive to reduce blockages (patch-2535: daniel.silverstone@canonical.com)
[04:44] <stub> spiv: ok. 
[04:47] <kiko> spiv, I'm already reviewing bjorn's bug 1733 branch
[04:47] <kiko> okay?
[04:47] <spiv> kiko: Oh, sure.  Fine with me.
[04:47] <kiko> cool
[04:47] <kiko> that way you have more time to FIX THE LIBRARIAN BUG
[04:47] <kiko> :)
[04:48] <spiv> Heh.
[04:49] <kiko> lifeless, can you please look at jamesh' patch that is sitting in your queue since the 13th? kthxbye
[04:49] <kiko> I removed all others, you should be safe to look at it
[04:49] <lifeless> kiko: I am looking at it
[04:49] <kiko> good
[04:49] <lifeless> kiko: You've missed me spending nearly a week chasing jasmesh long enough to talk with him about it
[04:50] <lifeless> kiko: which happened yesterday
[04:50] <stub> That librarian bug still open? Let me know if I can help with it.
[04:50] <kiko> there are telephones in australia last I heard
[04:50] <lifeless> couldn't prove it by me
[04:50] <kiko> also
[04:50] <kiko> I didn't see any cacherevs appear in rocketfuel/launchpad--devel--0 today
[04:50] <kiko> we're all still merging against 2422
[04:55] <kiko> lifeless, and /if/ you had started reviewing, you forgot to update the tag to needs-reply
[04:55] <kiko> which confuses everybody including james' magical chinstrap script
[04:57] <lifeless> kiko-zzz: 2472 is cachedreved on chinstrap
[04:58] <lifeless> kiko-zzz: I mirrored it across on tuesday
[05:01] <kiko-zzz> why am I not getting it when merging?
[05:01] <kiko-zzz> hmmm
[05:12] <dilys> Merge to rocketfuel@canonical.com/launchpad--production--1.34: [trivial]  Cherry picks (patch-15: stuart.bishop@canonical.com)
[05:24] <stub> kiko-zzz: Argh - the wiki shouldn't mention Ubuntu. There is a bug report open that says it should be 'Launchpad wikis' or somesuch for all wikis that we run using the auth server.
[05:25] <stub> kiko-zzz: Because derivative distros will use it too, and it is rude to tell people to log into the Guadalinux wiki using their Ubuntu wiki username/password.
[05:25] <stub> Or mpt will cry
[05:32] <stub> lifeless: The commits mailing list reply-to is set to warthogs instead of launchpad. Is that your list?
[05:34] <kiko-zzz> hmmm
[05:34] <kiko-zzz> stub, yeah, I think you're right -- will be easy for matsubara to refix as soon as we give him a target, though
[05:35] <kiko-zzz> stub, can you help find reference so I know what to point him to tomorrow morning?
[05:36] <stub> I can't find the conversation
[05:37] <stub> We just need to pick a unique key - I'd suggest using 'launchpad.net'
[05:37] <kiko-zzz> it's tricky because that's not a real wiki URL 
[05:37] <kiko-zzz> or not?
[05:38] <stub> Who cares?
[05:38] <kiko-zzz> god!
[05:38] <kiko-zzz> he's looking from up there
[05:38] <kiko-zzz> lightning bolts and all
[05:39] <stub> Use http://wiki.launchpad.net then, since that *should* be the URL of the launchpad wiki. But it is just a key - it is hidden from the user in the UI anyway.
[05:39] <kiko-zzz> there's UI however
[05:39] <kiko-zzz> that needs to be updated as well (again)
[05:40] <stub> You could keep using wiki.ubuntu.com as the key in the DB - only the wording on the UI needs to change (to something that is correct - it needs to tell people that this wikiname is constant across all the wikis we run)
[05:41] <kiko-zzz> okay
[05:41] <kiko-zzz> stub, can you at least comment on the bug and reopen it? I really need to go to bed
[05:41] <stub> I'll open a bug
[05:41] <kiko-zzz> thanks
[05:59] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Add a new index to TeamParticipation which the planner wants to use, already in production (patch-2536: stuart.bishop@canonical.com)
[07:01] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: Add support for shipit admins to add orders in behalf of other people. Lots of new tests. Add columns to store the shipping address in the ShippingRequest table. r=kiko (patch-2537: guilherme.salgado@canonical.com)
[08:32] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Fix for bug title not appearing in bug pages any longer -- add a missing interface attribute (patch-2538: christian.reis@canonical.com)
[09:06] <sivang> Morning all!
[09:06] <SteveA> hi
[09:06] <SteveA> anyone seen http://synchroedit.com/ ?   looks interesting for wikis / bofs
[09:08] <spiv> SteveA: sounds a bit like http://jotlive.com/
[09:09] <SteveA> is it free?
[09:09] <spiv> It seems not.  Well, unless you count the 5-page-a-month demo.
[09:11] <SteveA> the synchroedit thing is rumoured to be going to be under the lgpl
[09:11] <sivang> SteveA: are they open sourcing it ?
[09:11] <sivang> SteveA: seems nice, is it a wiki engine of some sort?
[09:12] <SteveA> not really
[09:12] <SteveA> a collaborative editor
[09:12] <lifeless> looked at gobby ?
[09:12] <lifeless> ddaa seems quite pro it
[09:13] <sivang> yuck, they use java to achive waht ajax can give them, I think
[09:13] <spiv> jotlive.com uses ajax.  I think it uses Twisted on the server side.
[09:13] <spiv> It looks slick, but I think they want people to pay for the service :/
[09:17] <sivang> spiv: livejot works on fifie ?
[09:17] <spiv> sivang: fifie?
[09:18] <dilys> Merge to rocketfuel@canonical.com/launchpad--production--1.34: [trivial]  Cherry picks into production--1.34 (patch-16: guilherme.salgado@canonical.com, stuart.bishop@canonical.com, rocketfuel@canonical.com)
[09:19] <sivang> spiv: ah sorry, my pet name for firefox :)
[09:19] <sivang> spiv: you see, I've seen ajax powered web sites that work badly under firefox
[09:19] <sivang> spiv: Twisted = the Python framework ?
[09:21] <bob2> yes
[09:21] <spiv> sivang: Yeah, I think they claim to work in at least firefox and IE.  And yeah, I mean the Twisted framework for Python.
[09:25] <sivang> spiv: myabe we can bring this functionality to Launchpad, I have a couple ideas that will utilitze this, and I'd like to see in launchpad.
[09:26] <spiv> sivang: Sadly, jotlive doesn't appear to be open source :(
[09:26] <sivang> spiv: my point exactly :)
[09:26] <spiv> Ah.
[09:26] <sivang> well, at least if not opensource, then free to use 
[09:27] <sivang> (although launchpad is not currently open, I think it's multitude of free offered services cover nicely)
[09:58] <dilys> Merge to rocketfuel@canonical.com/launchpad--production--1.34: Cherry pick christian.reis@canonical.com--lozenge/launchpad--devel--0--patch-157 into production 1.34 (patch-17: christian.reis@canonical.com)
[10:11] <jordi> mdke, Burgundavia: are you guys sorted out?
[10:19] <SteveA> BjornT: ping
[10:22] <carlos> morning
[10:22] <SteveA> hi carlos 
[10:52] <SteveA> lifeless: is pqm jamed?
[10:52] <SteveA> there's still a merge from 03:23 at the number one spot
[10:54] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  improve check-templates slightly -- add missing zcml bits (patch-2539: christian.reis@canonical.com)
[10:54] <SteveA> aha... guess not jammed any more
[11:03] <foxiness> hi am new member on launchpad and i want to send bug report about breezy but i can not find away from my launchpad 
[11:04] <foxiness> home page
[11:07] <mloskot> Q: If I've found a bug in one of Universe packaged in Breezey, where should I report it: Bugzilla or Malone on Launchpad?
[11:09] <SteveA> https://wiki.ubuntu.com/BugTracking
[11:09] <foxiness> i think i have the same Q mloskot have but may bug on main
[11:09] <SteveA> right now, bugs on things in "main" should be put into the ubuntu bugzilla
[11:09] <SteveA> bugs on other things can go in launchpad
[11:11] <foxiness> clear,but is there way on future to be on my homepage on launchpad to track it on easy way
[11:11] <SteveA> well, let's think about this
[11:11] <SteveA> your web browser says what OS you're using
[11:11] <SteveA> so, your home page on launchpad could see this
[11:12] <SteveA> and have a link to "report a bug in your own OS (breezy)"
[11:12] <mloskot> SteveA & foxiness: Thanks for the explanation.
[11:12] <SteveA> if you're apparently using some other OS, then it would depend if that OS is known by launchpad
[11:13] <SteveA> foxiness: if you like this idea, please file a bug on launchpad about it
[11:13] <foxiness> oh nice but becasue i can not use my modem from breezy im working from windows :)
[11:13] <SteveA> i see
[11:13] <mloskot> I've reported bug on Bugzilla for package from Univers (alexandria) so I think I should report it to Malone.
[11:13] <SteveA> mloskot: right
[11:13] <mloskot> https://bugzilla.ubuntu.com/show_bug.cgi?id=1660
[11:14] <mloskot> https://bugzilla.ubuntu.com/show_bug.cgi?id=16602
[11:14] <mloskot> So, should i move it to Malone? Report it again on Malone?
[11:14] <SteveA> mloskot: are you on any of the ubuntu irc channels?
[11:14] <mloskot> not at this mo
[11:14] <foxiness> SteveA, sure i like it and i will create a bug about it
[11:14] <SteveA> thanks foxiness 
[11:15] <SteveA> mloskot: okay.  try #ubuntu or #ubuntu-devel, because that's where the people who manage and read these bug reports will be.
[11:15] <mloskot> ok
[11:15] <mloskot> I will do so. thanks
[11:15] <SteveA> they'll be able to offer you better advice than i can about where to file bugs to ensure they reach the right people
[11:15] <SteveA> this channel is best for when you have problems with the launchpad website, or if you have ideas on how to make it work better for you
[11:17] <mloskot> I see, thanks again. so I'm going to the right channels. Cheers
[11:20] <sivang> SteveA: hmm, that sounsd fairly easy to implement, and while make things easier for bug reporters , is it already implemented?
[11:21] <SteveA> no.  i don't think it has been discussed before.
[11:29] <carlos> stub, hi, did you see my answer to your email?
[11:30] <stub> I replied to you (or do mean you replied to my reply to your reply?)
[11:32] <stub> carlos: I don't think losing that information is a problem, but you are more familiar with the code that relies on that field. 
[11:32] <carlos> stub, I just got your reply, sorry
[11:33] <carlos> stub, it's just that we are using that field to send the notification to hte user when the import is done
[11:33] <carlos> so that's the only problem I can think on
[11:35] <stub> Actually.... I don't see how we are losing any information
[11:36] <stub> daterawimport is currently only valid for imported objects
[11:36] <stub> I'm just giving it meaning if rawimportstatus == PENDING
[11:36] <carlos> stub, we use it to say... 'The file  foo (uploaded XXXX days/minutes ago) has been imported
[11:36] <ddaa> Mh, there's a meeting today?
[11:36] <carlos> stub, daterawimport tell us when was the file uploaded
[11:37] <stub> carlos: I thought it was set after the import had succeeded?
[11:37] <carlos> stub, no
[11:37] <stub> ohh...
[11:37] <stub> bugger
[11:37] <stub> So it should be renamed dateuploaded?
[11:38] <stub> ahh... date it was attached. 
[11:38] <stub> I guess that is a form of import ;)
[11:41] <carlos> yeah
[11:41] <carlos> that's the idea
[11:41] <carlos> but I'm happy to rename it if you think it's better
[11:41] <carlos> daterawfileattached
[11:41] <carlos> ?
[11:42] <BjornT> SteveA: pong
[11:42] <stub> Other things to worry about first I think
[11:43] <carlos> ok
[11:45] <carlos> stub, so, will we use the suggestion I did?
[11:47] <stub> carlos: We can do it without an extra column if we want (just store last seen locally). Is the dateimported useful to us do you think? If so, we can create the extra column. Otherwise I can fix it to just use pickle.
[11:47] <carlos> stub, I need to rethink a bit the import/attach process so I think it's ok to use that field now and think a better solution later
[11:48] <stub> It sounded like it would screw up the UI somewhat, though (?)
[11:49] <carlos> stub, well, we will lie a bit to the users if we have problems with an import
[11:50] <carlos> not a big deal in exchange to get the imports working
[11:51] <stub> carlos: I have another solution that should only take a few minutes. I'll post another diff.
[11:51] <carlos> stub, ok
[12:03] <Kinnison> 2h to meeting, yes?
[12:03] <SteveA> yes
[12:04] <Kinnison> coolio
[12:08] <ddaa> lifeless: merge your pybaz patch, pretty please
[12:15] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Adding CSS that renders disabled form fields differently from enabled ones, stolen from shipit (patch-2540: christian.reis@canonical.com)
[12:46] <BjornT> SteveA: pong again
[12:46] <SteveA> hi BjornT 
[12:46] <BjornT> hi
[12:47] <SteveA> i had a question about the menus changes brad made last night, but i checked the changeset myself
[12:48] <toresbe> I've got a problem with launchpad
[12:48] <toresbe> I'm signing up for advance orders of the Ubuntu goodness that is Breezy CDs 
[12:48] <toresbe> oops, phone, brb
[12:51] <nitwoh> sometimes this link doen't work : https://launchpad.net/distros/ubuntu/breezy/+lang/fr
[12:51] <nitwoh> it's down about 30% of time
[01:00] <toresbe> back
[01:00] <toresbe> anyway, I got a token link in the mail to sign up for shipments
[01:00] <toresbe> and it 404's
[01:01] <toresbe> https://shipit.ubuntu.com/token/z3rJHqTGHRHWzlrt0NX1
[01:01] <SteveA> toresbe: salgado works on shipit.  he'll be around in 1 hour, for the weekly developers meeting.
[01:01] <SteveA> will you be around in 1 hour?
[01:04] <Nafallo> SteveA: shipit works now right? so that my order is really there? :-)
[01:04] <SteveA> Nafallo: do you see your order when you log in?
[01:08] <Nafallo> SteveA: I'll check :-)
[01:09] <Nafallo> SteveA: yay. guess it works then :-).
[01:09] <toresbe> SteveA: thanks, yeah, I will :)
[01:09] <Nafallo> didn't see the order before that, so I increased the cd amount for this round ;-).
[01:15] <stub> carlos: https://chinstrap.ubuntu.com/~dsilvers/paste/fileB0jzPl.html
[01:18] <dholbach> hi
[01:18] <dholbach> does anybody know how to make mailman happy with the new malone mails change? (From: -> Reply-To:)
[01:18] <dholbach> or can this only be changed by making the list not moderated
[01:19] <stub> dholbach: Use the spam filters. Set one rule to 'accept' email based on the Sender: header (or similar).
[01:19] <dholbach> stub: ok thanks, i'll try that
[01:20] <Kinnison> Where does python wrap the time() syscall?
[01:20] <SteveA> time.time perhaps
[01:21] <Kinnison> that's the doobry, ta
[01:22] <carlos> stub, it looks ok for me
[01:24] <nitwoh> on Shipping Address form, the Name input is fixed as the viewed name on launchpad. so  I can't set a nickname for launchpad :(
[01:36] <asmodai> remind me again, warty is a version before 5?
[01:37] <dholbach> warty was 4.10 :)
[01:37] <dholbach> 4.10  = october 2004
[01:38] <asmodai> ah ok
[01:38] <asmodai> thanks
[01:38] <asmodai> I am not so into Linux versions/names and couldn't find it quickly on the ubuntu site
[01:40] <SteveA> launchpad meeting in 20 mins.  /msg me any items for the meeting
[01:48] <jamesh> carlos: I took a stab at fixing rosetta bug 74
[01:49] <jamesh> carlos: would you like to look at it before I put it up for review?
[01:49] <carlos> jamesh, sure
[01:49] <carlos> jamesh, thanks
[01:50] <jamesh> carlos: I'm mirroring it now.  I'll produce a diff
[01:50] <carlos> ok
[01:51] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  removal of dead templates (patch-2541: steve.alexander@canonical.com)
[01:52] <SteveA> hi salgado 
[01:52] <salgado> hi SteveA 
[01:52] <SteveA> salgado: can you talk with toresbe about shipit please?
[01:52] <salgado> sure
[01:52] <salgado> yo toresbe 
[01:53] <jamesh> carlos: my branch fixes both the RTL text issue, and should provide better rendering for CJK languages (since the correct glyphs will be used for the shared code points)
[01:53] <carlos> jamesh, cool, thank you!
[01:56] <jamesh> carlos: https://chinstrap.ubuntu.com/~jamesh/lp-bug74.patch
[01:57] <niemeyer> Morning!
[01:59] <mpt> Good morning
[01:59] <niemeyer> mpt: Good morning
[01:59] <jblack> weird. have you guys ever seen www.aypwip.org/webnote/ubuntu before ? 
[01:59] <jblack> Its javascript postit notes. Enter a keyword on the upper dir, and get a pile of notes
[02:00] <SteveA> it's meeting time
[02:00] <niemeyer> jbailey: Nice :)
[02:00] <niemeyer> Erm
[02:00] <niemeyer> jblack: Nice :)
[02:00] <SteveA> MEETING BEGINS
[02:00] <SteveA> who's present?
[02:00] <BjornT> me
[02:00] <jblack> me
[02:00] <salgado> me
[02:01] <mpt> meee
[02:01] <jamesh> me
[02:01] <spiv> me
[02:01] <bradb> mo
[02:01] <niemeyer> me
[02:01] <stub> yo
[02:01] <SteveA> ddaa?
[02:01] <nitwoh> me
[02:01] <SteveA> hello nitwoh 
[02:01] <nitwoh> hello SteveA 
[02:02] <mpool> SteveA: here?
[02:02] <ddaa> *chomp* *chomp* mhhh noodles :)
[02:02] <SteveA> hi mpool 
[02:02] <niemeyer> mpool: Just in time :)
[02:02] <gneuman> hi
[02:02] <SteveA> hi
[02:02] <SteveA> kiko-zzz: ?
[02:02] <SteveA> cprov?
[02:02] <SteveA> hi matsubara 
[02:02] <matsubara> hi steve
[02:02] <SteveA> anyone seen kiko or cprov?
[02:03] <matsubara> nope, i just got here
[02:03] <jblack> I saw kiko ~ 6 hours ago
[02:03] <stub> kiko was up pretty late. Don't know about cprov
[02:03] <SteveA> okay, let's start anyway.
[02:03] <SteveA> == Agenda ==
[02:03] <SteveA>  - roll call
[02:03] <SteveA>  - agenda
[02:03] <SteveA>  - activity reports
[02:03] <SteveA>  - production / staging
[02:03] <SteveA>  - problem with /errors on staging
[02:03] <SteveA>  - new menus stuff
[02:03] <SteveA>  - slow pages
[02:03] <SteveA>  - three sentences
[02:03] <SteveA> 
[02:03] <SteveA> any other items to add to the agenda?
[02:03] <SteveA>  - time of next meeting
[02:04] <SteveA> same time next week?
[02:04] <toresbe> salgado: hey, pardon my delay
[02:04] <SteveA> jordi: ping?
[02:04] <Kinnison> SteveA: fine by me
[02:04] <jblack> aye
[02:04] <kiko-zzz> hey ho
[02:04] <kiko> sorry
[02:04] <kiko> was up late
[02:04] <SteveA> hi kiko
[02:05] <salgado> hi toresbe, we're in a meeting now, can we talk later?
[02:05] <toresbe> salgado: of course, pardon my interruption :)
[02:05] <salgado> no worries. :)
[02:05] <SteveA> activity reports: who's up to date, and who, with regret and other such things, isn't 
[02:05] <SteveA> ?
[02:05] <mpt> up to date
[02:06] <kiko> I'm the hacker
[02:06] <jblack> up to date
[02:06] <bradb> My Sent Items folder is up to date...if I could only figure out where in the ether my last two days' activity reports are...
[02:06] <SteveA> bradb: mail them directly to me if you like
[02:06] <bradb> ok
[02:06] <carlos> jamesh, it looks great
[02:07] <SteveA> carlos: activity reports?
[02:07] <carlos> SteveA, sorry, I was distracted with the review
[02:07] <SteveA> gneuman, matsubara, jordi ?
[02:07] <gneuman> hi
[02:07] <carlos> I'm a week behind
[02:07] <mpool> up to date
[02:08] <SteveA> hi gneuman.  how up to date are you with your daily activity reports?
[02:08] <kiko> not at all :)
[02:08] <matsubara> same with me
[02:08] <kiko> I haven't asked him or matsubara send them in -- is it time?
[02:08] <bradb> SteveA: just sent my last two days to you
[02:08] <gneuman> i am not sendind it yet
[02:08] <SteveA> okay.  gneuman and matsubara, please talk with kiko about what to do
[02:08] <SteveA> thanks bradb 
[02:08] <BjornT> bradb: your last two days' activity reports seem to have arrived to the activity list
[02:08] <SteveA> they arrived
[02:08] <gneuman> ok
[02:09] <bradb> really? hrm. maybe some weird gmail thing. it doesn't seem to fetch messages that I sent.
[02:09] <SteveA> next, production / staging.  stub?
[02:09] <lifeless> present
[02:09] <jamesh> bradb: probably a mailman thing (see if you have the not-me-too option set)
[02:09] <lifeless> sorry for lateness
[02:09] <stub> lots of cherry picks have been going on into production - lots of people are in bug-fix mode which is good
[02:10] <mpool> irc is messed up, back in a bit
[02:10] <lifeless> up to date
[02:10] <stub> staging is business as usual.
[02:10] <kiko> crack, damn
[02:10] <kiko> stub, wait
[02:10] <SteveA> there's still a problem with /errors on staging
[02:10] <SteveA> i need to hassle the admins
[02:10] <stub> Load issues with Malone don't seem to be an issue
[02:10] <kiko> staging wasn't refreshing till yesterday, right? :)
[02:10] <stub> Load issues with some Rosetta pages are still being reported
[02:11] <kiko> > wikinames doesn't work, or did he lose it again afterwards?
[02:11] <kiko> she did'nt lose it rather it was'nt working back then :-)
[02:11] <kiko> ouch
[02:11] <stub> kiko: It has been refreshing daily for at least a week
[02:11] <kiko> stub, ah -- the permissions errors we saw were setup issues then?
[02:11] <stub> kiko: Sometimes I have to poke it, but that is what it is for (finding issues before production rollout)
[02:12] <kiko> yeah, ok
[02:13] <SteveA> anything else about production / staging (other than /errors and slow pages?)
[02:13] <SteveA> oh, stub: is pound working correctly?
[02:13] <stub> yes
[02:13] <SteveA> i'm going to phone the sysadmins today to get some things sorted / committed to
[02:13] <SteveA> okay
[02:14] <SteveA> so, that would explain why links to /errors don't work properly when people paste them into irc
[02:14] <SteveA> as they are per-server
[02:14] <SteveA> and people will have affinity with different servers
[02:14] <kiko> yes, I've heard this before
[02:14] <SteveA> this could be fixed by apache redirects directly to /errors1 and /errors2
[02:14] <SteveA> or by a proper error reporting system
[02:15] <spiv> It'd be really nice to get error reports stored somewhere persistent.  Like a mailing list.
[02:15] <SteveA> i'll ask the admins about doing /errors1 and /errors2 as an immediate thing
[02:15] <SteveA> and we need to decide who will implement the error reporting spec
[02:15] <jamesh> I could take a look at it
[02:15] <SteveA> it should be one of me, stub, spiv or jamesh
[02:16] <stub> I can look at it, either before or after I look at LibrarianGarbageCollection (unless spiv has time to take that over)
[02:17] <SteveA> i'll ask the admins about librarian usage
[02:17] <SteveA> then we'll decide.
[02:17] <SteveA>  - slow pages
[02:18] <lifeless> you skipped new menus
[02:18] <SteveA> so, some pages are timing out.  on the next roll-out, there will be a separate error page for requests that timed out, so users of launchpad will at least know why they timed out
[02:18] <SteveA> lifeless: doing it later, as this topic is more to do with production/staging
[02:18] <lifeless> k
[02:19] <SteveA> stub: do you know which pages are still timing out?
[02:19] <SteveA> are cronscripts still causing contention problems that cause timeouts?
[02:19] <stub> I've been seeing occasional reports on various +translate pages 
[02:19] <zyga> hi
[02:19] <SteveA> hi zyga 
[02:20] <zyga> is there any feature planned that will allow the reviewer to select any of the suggested translations with one click?
[02:20] <carlos> From rosetta all language reports pages like launchpad.net/distros/ubuntu/breezy/+lang/fr are slow
[02:20] <stub> Could be, but we need the cronscripts (at least the ones that might affect the +translate pages)
[02:20] <SteveA> or stuffed into a file
[02:20] <kiko> I know what bugs they are
[02:20] <kiko> if you need to know
[02:20] <carlos> SteveA, stub kiko noticied that the karma issue is still there, I need to do a second review
[02:20] <kiko> zyga, yes.
[02:21] <zyga> kiko: great
[02:21] <SteveA> okay... having the karma issue still there is a problem
[02:21] <BjornT> bradb: +packagebugs times out for me. can you talk with stub and try to make the page render faster?
[02:21] <SteveA> both for the slowness of the page, and for having inappropriate karma
[02:21] <stub> I'm not seeing any blockages caused by karma
[02:21] <bradb> BjornT: Yeah, I'll look into optimizing it for good.
[02:21] <SteveA> stub: can you just make the cronscript's dbuser not allowed to write karma?
[02:21] <stub> If karma is still being added innapropriately, I don't think it is part of the timeout problem
[02:22] <SteveA> stub: what's the current timeout in production?
[02:22] <stub> SteveA: Sure.
[02:22] <stub> 12 seconds. We could get away with increasing that again with the second backend online
[02:22] <cprov> SteveA: sorry, I' m late ... 2 days back in activity -report, next meeting Oct 6th is fine
[02:22] <SteveA> hi cprov 
[02:23] <SteveA> stub: okay, let's increase it a bit
[02:23] <cprov> SteveA: hi
[02:23] <SteveA> kiko: do you have a suggestion of what to increase it to?
[02:23] <kiko> 15, perhaps
[02:23] <kiko> I get the feeling now that the pages that are looking bad are /really/ looking bad
[02:24] <SteveA> are the problems to do with specific portlets, or the main functionality of the page?
[02:24] <kiko> it's hard to say from the bug reports
[02:24] <kiko> (because they are just crashes to the end-user)
[02:24] <SteveA> right
[02:25] <kiko> and I haven't seen an error report on it yet
[02:25] <spiv> I wonder if it would be possible to have a shorter timeout when rendering a portlet, and have a way for a single portlet to fail but not break the rest of the page?
[02:25] <SteveA> anything else on slow pages / timeouts ?
[02:25] <SteveA> spiv: kind of
[02:26] <lifeless> spiv: ESI
[02:26] <SteveA> spiv: i've given this some thought
[02:26] <kiko> interesting idea, spiv 
[02:26] <SteveA> one issue with that is that often portlets use the same cached DB objects and information that the main page uses
[02:27] <SteveA> so it is not really clear what has taken the time
[02:27] <SteveA> so, i think making actual programming errors in portlets not make the whole page fail, would be feasible
[02:27] <SteveA> but having different timeouts for portlets is not
[02:28] <SteveA> ESI would work, but that has other problems, such as the lack of database object cacheing, and having portlets rendered in separate transactions to the main page.
[02:28] <SteveA>  - new menus stuff
[02:29] <SteveA> a bunch of new menus stuff has landed.  old actions-portlet templates have been removed.
[02:29] <SteveA> bradb: i saw there was some issue last night about some stuff from app menus that should have been in context menus
[02:29] <SteveA> glad you and mpt got it sorted
[02:29] <bradb> yup
[02:29] <SteveA> the menus docs are lagging behind where we're actually at with them
[02:29] <mpt> well, it was stuff in bug menus that should have been inherited into bugtask menus
[02:29] <SteveA> i'll work on updating them today and tomorrow.
[02:30] <SteveA> any menus questions / comments?
[02:31] <SteveA> okay
[02:31] <kiko> SteveA, from what I've seen the work has been superb, so congratulations
[02:31] <SteveA> thanks kiko.
[02:31] <SteveA> there are still some bugs / issues
[02:31] <SteveA> but for the most part, they are working properly
[02:32] <SteveA> any other points before we look at the sets of three sentences?
[02:32] <Kinnison> if that's okay
[02:32] <kiko> yes
[02:32] <kiko> I also wanted to talk about make lintmerge and check-templates
[02:32] <SteveA> kiko: okay, go for it
[02:33] <SteveA> Kinnison: after kiko please
[02:33] <Kinnison> SteveA: thanks
[02:33] <kiko> so enough people whined that I added a make lintmerge target
[02:33] <kiko> I really recommend you install pyflakes 2.0 to use it
[02:33] <kiko> (it supports py2.4 features)
[02:33] <kiko> I still need to silence some of the more annoying messages
[02:33] <SteveA> kiko: where do you install pyflakes 2.0 from?
[02:34] <kiko> it's mostly pylint's fault because it doesn't understand tuple unpacking for legibility and the callback syntax is awkward
[02:34] <kiko> callbacks need to be named cb_foo to have unused arguments ignored
[02:34] <kiko> I find that hateful (but don't have a better solution)
[02:34] <kiko> so use it and let me know what you think
[02:34] <kiko> http://divmod.org/projects/pyflakes
[02:34] <SteveA> no ubuntu package?
[02:34] <kiko> I'm trying to get it in breezy, but that's a loosing battle
[02:35] <ddaa> kiko: mark unused arguments with "unused = argument" and ignore the "unused" name for unused warnings.
[02:35] <kiko> it is in breezy, but not 0.2.0
[02:35] <SteveA> have you talked to doko ?
[02:35] <kiko> SteveA, I have, will talk again
[02:35] <kiko> anyway
[02:35] <kiko> check-templates
[02:35] <ddaa> I also like this idiom because it's explicit
[02:35] <kiko> this is a nice little script that checks for stale templates
[02:35] <kiko> ddaa, yeah, let's talk about this further
[02:35] <kiko> so when you do refactors or whacks
[02:35] <kiko> run it
[02:35] <kiko> and see if you left any stale templates around
[02:36] <SteveA> okay, cool
[02:36] <SteveA> kiko: done/
[02:36] <SteveA> ?
[02:36] <kiko> I've got one pending change to it to remove 5 false-positives, but then I think it's good to go
[02:36] <kiko> use it!
[02:36] <kiko> thanks
[02:36] <SteveA> Kinnison: 
[02:37] <Kinnison> but it isn't even zopeless, it's utterly non-launchpaddy
[02:37] <Kinnison> (build daemon sequencer)
[02:37] <BjornT> kiko: for check-templates to be useful, someone has to remove unused templates. atm it produces too much output. also template files should be ignored
[02:37] <Kinnison> --that is all--
[02:37] <SteveA> Kinnison: i recommend the standard library.
[02:37] <niemeyer> Kinnison: smtplib?
[02:37] <kiko> stub, the bugs you assigned yourself to are dupes..
[02:37] <ajmitch> kiko: if pyflakes is universe, talk to me after the meeting
[02:37] <kiko> ajmitch, rock!
[02:38] <SteveA> thanks ajmitch 
[02:38] <stub> kiko: Just karma whoring
[02:38] <SteveA> okay, three sentences.  time is short.  go for it!
[02:38] <bradb> Kinnison: Gotta be careful about using the same headers that we use too, etc.
[02:38] <Kinnison> DONE: buildd work, publisher work, SP refcounting in domaintor, gina work. Laptop continues to update from dogfood, dogfood is building. Dogfood updating from cron. Laptop suspends okay now. (ish)
[02:38] <spiv> Kinnison: There's twisted.mail.smtp.sendmail for Twisted-friendly smtplib-like stuff.
[02:38] <ddaa> DONE: samba import, yay! caution: cscvs does not support svn renames yet
[02:38] <ddaa> TODO: merge samba-related patches, importd tests/changes for python, finish importd-archivelocation, finish sprint work
[02:38] <ddaa> BLOCKED: lifeless merging his pybaz patch
[02:38] <Kinnison> TODO: finish build scheduler, bin-nmu support in gina, uploader
[02:38] <Kinnison> BLOCKED: Nothing currently
[02:38] <SteveA> DONE: menus
[02:38] <SteveA> TODO: breadcrumbs and traversal refactor, plans for error reporting, hassle sysadmins
[02:38] <SteveA> BLOCKED: no
[02:38] <mpt> DONE: Support tracker cleanup, menus work, CoC tidyup, started major cleanup
[02:38] <mpt> TODO: Finish major template cleanup, Konqueror and Opera love, bugfixes
[02:38] <mpt> HINDRANCES: Konqueror and Opera not installed on Async system yet (and my laptop power cable is broken)
[02:38] <bradb> DONE: Patches landed: Malone admin awareness, "Bugs on Maintained Software" report, wiki-mode task edit permission
[02:38] <BjornT> DONE: first go at pre-defined bug reports. bug fixing, email wrapping problem and other smaller bugs. reviews.
[02:38] <kiko> DONE: reviews, fixes, bug triage, assistance to shipit and langpacks and anyone else who needed it (hopefully)
[02:38] <spiv> DONE: reviews, chasing librarian bug leading to a bug fix in zopeless, some bzr/Twisted stuff with Robert.
[02:38] <spiv> TODO: finally kill the librarian/rosetta bug. reviews.
[02:38] <spiv> BLOCKED: no.
[02:38] <bradb> s, Malone menus regression fix.
[02:38] <carlos> DONE: Language packs, poimports, user support
[02:38] <niemeyer> DONE: bzr support on meld, svn2bzr, found/fixed difflib misbehavior, SmartPM maintenance
[02:38] <niemeyer> TODO: Continue work on bzr, continue maintenance of other projects.
[02:38] <niemeyer> BLOCKED: Nope
[02:38] <BjornT> TODO: vacation
[02:38] <lifeless> DONE: merge weave up with HEAD, baz2bzr updates to latest apis, various mgmt things
[02:38] <bradb> TODO: Inherit one of BjornT's branches while he's on holiday. Some bugfixes/things to diagnose: kiko's +editstatus
[02:38] <BjornT> BLOCKED: no
[02:38] <lifeless> TODO: finish weave symlink support, merge x bit,
[02:38] <bradb>  problem, envelope from should be set to bounce address, etc. Other stuff.
[02:38] <lifeless> BLOCKED: Not
[02:38] <bradb> BLOCKED: Non.
[02:38] <jblack> DONE: bzr traffic, bzrpqm, others
[02:38] <kiko> TODO: damned spec work I never seem to get around to
[02:39] <kiko> BLOCKED: no
[02:39] <jblack> TODO: bzr traffic, more bzrpqm, others
[02:39] <carlos> TODO: poimports/attachment improvements
[02:39] <carlos> BLOCKED no
[02:39] <stub> DONE: BrowserNotificationMessages
[02:39] <mpool> DONE: merge bzr weave format, few bugs
[02:39] <jblack> BLOCKED: planet.bazaar.canonical.com
[02:39] <salgado> DONE: Lots of shipit, helping matsubara and gneuman, some small fixes.
[02:39] <salgado> TODO: ShipIt exports, keep helping matsubara and gneuman
[02:39] <salgado> BLOCKED: No
[02:39] <stub> TODO: LibrarianGarbageCollection
[02:39] <stub> Blocked: No
[02:39] <mpool> TODO: switch weave to mainline, faster branch
[02:39] <mpool> BLOCKED: no
[02:39] <jamesh> DONE: timing and logging fixes for the test suite.  Fix bug 74 (RTL text in rosetta), some scheduler work
[02:39] <jamesh> TODO: more stuff on scheduler, code reviews
[02:39] <jamesh> BLOCKED: no
[02:40] <SteveA> okay
[02:40] <SteveA> ddaa: well done getting that samba import done
[02:41] <SteveA> lifeless: what's happening about merging your pybaz branch? 
[02:41] <stub> Anyone need to land anything before the production tagging?
[02:41] <kiko> stub, not that I know of, BUT
[02:41] <SteveA> spiv: do you know what is causing the librarian / rosetta bug now?
 BLOCKED: planet.bazaar.canonical.com
[02:41] <SteveA> is this something i can help with, jblack ?
[02:41] <carlos> stub, If we could merge my language pack branch would be really good, but still waiting for review....
[02:41] <kiko> stub, salgado and I need to work on exports for shipit, which will probably need to be cherry-picked mid-week next week
[02:41] <jblack> Ask elmo to read rt@admin.canonical.com? 
[02:41] <lifeless> SteveA: Its pending a reply from scott from an hct bug it uncocvers
[02:42] <kiko> stub, what's the best strategy for it?
[02:42] <spiv> SteveA: No, but I have ruled some stuff out and figured out how to put more diagnostics in.
[02:42] <spiv> I tripped over and fixed a minor bug in zopeless connection handling along the wawy :)
[02:42] <stub> kiko: Avoid merging in from rocketfuel and it should be easily cherry pickable. I prefer to not cherry pick database patches, but can if necessary
[02:42] <kiko> stub, it will require db patches :-(
[02:43] <SteveA> jblack: what is the RT issue number?
[02:43] <kiko> sorry, we've been racing for this one but it's hard to get it in time
[02:43] <lifeless> kiko: then its not cherrypickable
[02:43] <kiko> lifeless, explain that to your CEO
[02:43] <lifeless> kiko: well, not easy to
[02:43] <kiko> :)
[02:43] <SteveA> ddaa, lifeless: i'll talk with scott
[02:43] <lifeless> SteveA: no need
[02:43] <SteveA> lifeless: why?
[02:43] <jblack> 23
[02:43] <lifeless> SteveA: I tried it yesterday when spiv did the review, got the failure overnight, mailed him today
[02:44] <SteveA> lifeless: okay
[02:44] <lifeless> SteveA: so its moving along just fine.
[02:44] <SteveA> jblack: okay, i'll talk with the admins
[02:44] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Removed obsolete code and renamed a method to have a better name (patch-2542: carlos.perello@canonical.com)
[02:44] <lifeless> SteveA: the only urgency on it is because ddaa is choosing not to merge things to public before rocketfuel - a policy choice at most
[02:45] <ddaa> lifeless: SteveA: it's too keep my life simple with pybaz patchflow.
[02:45] <kiko> guys
[02:45] <kiko> I am VERY busy this morning
[02:45] <ddaa> it's not a critical problem, but community feedback raised the priority of the issue a bit.
[02:45] <SteveA> it's time to end the meeting...
[02:45] <kiko> so please rely on someone else if you have anything critical (beyond responding to reviews which I can do)
[02:45] <SteveA> mail / talk to me, not kiko
[02:46] <SteveA> okay, let's wrap this up
[02:46] <SteveA> 3
[02:46] <SteveA> 2
[02:46] <SteveA> 1
[02:46] <lifeless> 0
[02:46] <SteveA> MEETING ENDS
[02:46] <Kinnison> spiv: talk to me about twisted smtp
[02:46] <mpt> carlos: ping
[02:46] <Kinnison> bradb: what headers/envelope stuff are you on about?
[02:46] <kiko> thanks SteveA, rocks as always
[02:46] <SteveA> thanks everyone 
[02:46] <kiko> jordi, ping?
[02:46] <Kinnison> thanks SteveA 
[02:46] <SteveA> bradb: an interesting idea popped up in discussion earlier
[02:47] <spiv> Kinnison: Ok.
[02:47] <bradb> Kinnison: envelope from should be the bounce address
[02:47] <bradb> Sender should be too, and Errors-To, etc.
[02:47] <SteveA> that is, if the user's browser user agent indicates they're using an ubuntu release (or maybe any release launchpad knows about), offer them the opportunity to file a bug directly on that.
[02:47] <Kinnison> bradb: is that true for mails we're sending to LP devs only?
[02:47] <carlos> SteveA, kiko I want to note you that tomorrow I will not be online all day, I have a speech about Rosetta in the afternoon and I need to prepare it and go there (it's about 1hour and a half from my home)
[02:47] <carlos> mpt, pong
[02:47] <SteveA> carlos: okay, thanks.
[02:47] <kiko> carlos, okay
[02:47] <mpt> carlos: in pofile-translate.pt
[02:47] <mpt> carlos: How do I get the name of the thing being translated?
[02:47] <mpt> not the template
[02:47] <bradb> Kinnison: LP users.
[02:47] <Kinnison> bradb: these are mails equivalent to what cron would send if a job wrote some output
[02:47] <kiko> SteveA, do you think you could review carlos' diff this morning? if you can't I can do it but I am swamped today..
[02:47] <mpt> but the thing like "Firefox" or "firefox in Ubuntu 5.04"
[02:47] <spiv> Kinnison: twisted.mail.smtp.sendmail's docstring says that it claims to be a fairly direct replacement for smtplib.sendmail.
[02:48] <SteveA> kiko: i need to get lunch now.  i can review stuff once i get back from lunch.
[02:48] <bradb> SteveA: ah, right. that would be interesting.
[02:48] <carlos> mpt, the name?
[02:48] <mpt> carlos: yes
[02:48] <kiko> SteveA, great, thanks
[02:48] <kiko> it's not TOO big
[02:48] <ajmitch> kiko: you said pyflakes is in breezy already? what package name?
[02:48] <SteveA> you're scaring me, dude
[02:48] <carlos> mpt, context.title
[02:49] <Kinnison> spiv: 
[02:49] <mpt> carlos: No, that's the name of the template
[02:49] <Kinnison> >>> from twisted.mail import smtp
[02:49] <kiko> ajmitch, hmmm, it appears to not be. odd. see:
[02:49] <Kinnison> ...
[02:49] <Kinnison> ImportError: cannot import name smtp
[02:49] <mpt> carlos: I want the name of the product/package
[02:49] <bradb> Kinnison: oh, n/m then, probably
[02:49] <Kinnison> bradb: okies, ta
[02:49] <salgado> toresbe, around?
[02:49] <Kinnison> bradb: next week I'll want to send mails to users and will come to you for advice then :-)
[02:49] <bradb> ok :)
[02:49] <Kinnison> bradb: but I'll be at least zopeless by then
[02:50] <toresbe> salgado: yep
[02:50] <carlos> mpt, context.potemplate.productrelease and context.potemplate.sourcepackagename ?
[02:50] <Kinnison> spiv: is twisted.mail in a different package?
[02:50] <ajmitch> kiko: looks like a 15 min packaging job, I'll try & get it in universe asap
[02:50] <mpt> carlos: So either one of those will exist, or the other, but not both at the same time?
[02:50] <toresbe> salgado: my problem is that I logged in to order some Breezy CDs
[02:50] <spiv> Kinnison: Oh, Twisted 1.3?
[02:50] <spiv> Kinnison: twisted.protocols.smtp then.
[02:50] <toresbe> salgado: and it told me I wasn't confirmed
[02:50] <spiv> (which will still work with 2.0, just with deprecation warnings)
[02:51] <toresbe> so it sent me an email with a token 
[02:51] <kiko> ajmitch, you rock!
[02:51] <toresbe> and the token URL 404s
[02:51] <bradb> anyway, at gf's right now. /me heads home. bbiab.
[02:51] <carlos> mpt, right
[02:51] <kiko> ajmitch, it's a very simple tool -- if you can do it, you get a CD of your choice in Montreal
[02:51] <Kinnison> spiv: will that have a nice simple send_mail_from_string() type method?
[02:51] <salgado> toresbe, dude, I've been looking for this bug for quite a while now
[02:51] <ajmitch> kiko: a good thing I'm going to be at UBZ then :)
[02:51] <salgado> toresbe, did you manage to order the CDs already?
[02:52] <spiv> Kinnison: See the docstring for the 'sendmail' function.
[02:52] <toresbe> salgado: no
[02:52] <mpt> salgado: awesome, thanks
[02:52] <spiv> Kinnison:     @param msg: The message, including headers, either as a file or a string.
[02:52] <salgado> mpt, for what?
[02:52] <spiv> Kinnison: and as it goes on to say, an email.Message should work too.
[02:53] <mpt> salgado: for your help :-)
[02:53] <salgado> toresbe, so, there's a possible workaround. all you have to do is to login on https://launchpad.net/+login
[02:53] <Kinnison>     sendEmail(smtphost, fromEmail, toEmail, content, headers=None, attachments=None, multipartbody='mixed')
[02:53] <Kinnison> aha!
[02:53] <salgado> toresbe, that will send you the token with the "right" url, and then you'll be able to validate your email address
[02:54] <kiko> ajmitch, remind me of my debt to you! CC: me on the upload mail :)
[02:54] <toresbe> ok..
[02:54] <spiv> Kinnison: sendmail won't give you DeprecationWarnings :)
[02:54] <kiko> SteveA, it's all in ajmitch's hands now :)
[02:54] <spiv> Kinnison: sendEmail will.
[02:54] <salgado> kiko, are you following this? this is what's causing that error messages of people accessing shipit.ubuntu.com/token/<whatever>
[02:54] <Kinnison> spiv: heh
[02:54] <Kinnison> spiv: fair enough
[02:55] <kiko> salgado, I'm following it, but I don't see the cause
[02:55] <salgado> toresbe, then, after validating your email address you should be able to login on shipit.ubuntu.com and order the CDs
[02:55] <toresbe> Excellent! thanks :)
[02:55] <salgado> toresbe, sorry for the trouble, and thanks for reporting the problem. I'll look into fixing it today
[02:55] <kiko> toresbe, salgado: wait
[02:56] <kiko> salgado, did you understand why toresbe's request failed?
[02:56] <kiko> this is sort of a precious moment
[02:56] <kiko> a week trying to debug this!
[02:56] <salgado> kiko, his request didn't fail. he's not able to login
[02:56] <kiko> his token
[02:57] <spiv> Kinnison: smtplib would probably work ok as well, except for it blocking the Twisted event loop.
[02:57] <Kinnison>                 sendmail(config.buildsequencer.smtphost,
[02:57] <Kinnison>                          config.buildsequencer.fromaddress,
[02:57] <Kinnison>                          recipient, this_msg)
[02:57] <Kinnison> looks good to me ;-)
[02:57] <kiko> salgado, his token :)
[02:57] <salgado> toresbe, can you confirm to us that the link you received (the 404 one) is something like https://shipit.ubuntu.com/token/<someweirdstring>
[02:57] <toresbe> yeah, want the full one
[02:58] <salgado> no need to
[02:58] <salgado> btw
[02:58] <salgado> you don't even need to login on launchpad
[02:58] <salgado> you can just replace shipit.ubuntu.com with launchpad.net
[02:58] <toresbe> ok
[02:58] <kiko> toresbe, can you tell me how you generated your request?
[02:58] <toresbe> so it's just the prefix url that was wrong?
[02:58] <kiko> what URL you were at?
[02:58] <salgado> toresbe, yes
[02:58] <kiko> toresbe, yes, but we don't understand why or how
[02:58] <toresbe> kiko: I went toi shipit
[02:58] <kiko> okay
[02:59] <kiko> and then?
[02:59] <kiko> what page did you create your new account in?
[02:59] <salgado> kiko, I do understand why and how!
[02:59] <toresbe> and put in my email address, and the password I usually use into the password field, thinking that I might have signed up already
[02:59] <kiko> I'll explain, toresbe: the issue is that the lin..
[02:59] <kiko> toresbe, AH!
[02:59] <kiko> of course
[02:59] <kiko> heh
[02:59] <kiko> salgado, heh :)
[02:59] <toresbe> and then it told me login unsuccessful and sent me an email
[02:59] <salgado> he has an account with no validated addresses
[02:59] <kiko> salgado, you and I are losers :)
[03:00] <salgado> yes, we are
[03:00] <toresbe> not at all
[03:00] <kiko> this is so obvious
[03:00] <kiko> sorry dude
[03:00] <kiko> we owe you one
[03:00] <kiko> argh
[03:00] <ajmitch> kiko: good news is the packaging is done, it'll just need uploaded & cleared by elmo
[03:00] <kiko> SteveA, can you ensure that elmo approves this update?
[03:01] <toresbe> I like the preset amount of cd's thing
[03:01] <kiko> he usually doesn't say yes to stuff I ask so :)
[03:01] <kiko> toresbe, yeah, new feature in shipit 2.0 -- thank salgado and jane silber for the idea
[03:01] <toresbe> but, er, how about a higher ratio of 64-bit CDs?
[03:01] <toresbe> Pretty much all the puters my friends have are AMD64s
[03:01] <kiko> toresbe, the numbers we have are based on the average shipments we did for shipit 1.0
[03:01] <kiko> so hmmm
[03:02] <toresbe> kiko: Yeah, but the 64-bit machines have become far more common in that time
[03:02] <Kinnison> spiv: You rock, that's perfect, thanks dude
[03:02] <kiko> I'll talk to jane, toresbe 
[03:02] <toresbe> kiko: cheers.
[03:02] <kiko> thanks for the hint
[03:02] <spiv> Kinnison: Hmm, for that code I think a dude in Boston is the guy that rocks :)
[03:03] <mpt> carlos: small problem
[03:03] <Kinnison> spiv: Well, you rock for telling me about it.
[03:03] <spiv> :)
[03:03] <Kinnison> spiv: to rock even more, you can help me work out how to properly test my program
[03:04] <spiv> Ah.  Just as I was going to go to bed, too.
[03:04] <mpt> carlos: http://localhost:8086/products/netapplet/+series/releases/+pots/netapplet/xh/+translate has neither context/potemplate/productrelease *nor* context/potemplate/sourcepackagename
[03:04] <spiv> Then you tempt me like this!
[03:04] <Kinnison> spiv: Well, I will put it in your review queue and you can look at it tomorrow morning
[03:04] <Kinnison> :-)
[03:04] <spiv> Kinnison: Which program?  A twisted daemon?
[03:04] <Kinnison> it's surprisingly simple
[03:05] <Kinnison> < 150 lines all-in really
[03:05] <carlos> mpt, sorry, productseries, not productrelease....
[03:05] <carlos> mpt, without either of those you are not able to traverse ....
[03:05] <Kinnison> spiv: and a bunch of that is the email template and interpolation dict for it
[03:05] <spiv> Ok.  I'll look at the patch and see what I think :)
[03:05] <Kinnison> coolio, thanks dude
[03:05] <Kinnison> sleep well
[03:06] <spiv> Ta
[03:06] <Kinnison> stub: When my build sequencer lands, it'll need config updates
[03:06] <Kinnison> stub: What do we do about that?
[03:06] <mpt> carlos: ok, to save you time, where do I look to find the various things that belong to productseries?
[03:06] <ajmitch> kiko: I'll let you know when it gets through elmo to the archives
[03:07] <kiko> thanks
[03:07] <stub> Kinnison: Make the changes to default/launchpad.conf and let me know. Extra points if you want to update staging/launchpad.conf and production/launchpad.conf (in which case definitely let me know)
[03:08] <Kinnison> stub: updating those would be a touch fraught. This is config for the appserver which we designate as the build master
[03:08] <Kinnison> Do we keep a separate config for the librarian machine?
[03:08] <stub> Kinnison: if we don't, we should.
[03:08] <carlos> mpt, I don't understand the question...
[03:09] <Kinnison> stub: I'll just make a note to prod you when this is gonna be merged
[03:09] <mpt> carlos: Where can I find the list of things that belong to a productseries, like "displayname", "version", whatever
[03:09] <stub> Kinnison: If it is just for one particular production box, then the only thing that needs adding to the staging and production configs are any required items (so the config loads)
[03:09] <carlos> mpt, lib/canonical/launchpad/interfaces/productseries.py ?
[03:09] <kiko> BjornT, got the two reviews okay?
[03:10] <Kinnison> stub: right
[03:10] <Kinnison> stub: is there an easy way to check if the config loads?
[03:10] <mpt> carlos: ah, thanks
[03:10] <stub> cd lib; env LPCONFIG=production python -c 'from canonical.config import config'
[03:10] <stub> oh... production1 and production2 are more important than production/launchpad.conf
[03:10] <BjornT> kiko: yeah, thanks. i've sent one reply already, will send the other reply soon.
[03:11] <Kinnison> stub: well they seem to pass
[03:11] <kiko> BjornT, cool
[03:11] <Kinnison>         <section name="*" type="buildsequencer" attribute="buildsequencer" />
[03:11] <carlos> see you later
[03:11] <stub> Cool.
[03:12] <kiko> stub, the bug matt reported is a dupe, but you assigned yourself to it
[03:12] <kiko> what should I do?
[03:12] <stub> Flag it as a dupe?
[03:13] <kiko> okay
[03:13] <kiko> very funny man
[03:14] <Kinnison> stub: fancy casting an eye over my config changes (and runlaunchpad.py and makefile changes) ?
[03:15] <segfault> #2622 can be closed, i guess?
[03:15] <stub> Kinnison: tomorrow ok? 
[03:16] <Kinnison> stub: sure
[03:16] <Kinnison> there's no db changes
[03:17] <Kinnison> So, talking of dogfood, anyone object to me upgrading it?
[03:18] <kiko> nope
[03:18] <kiko> do it
[03:18] <kiko> stub, salgado: is shipit available on staging?
[03:18] <kiko> or SteveA, elmo?
[03:19] <kiko> segfault, no, that's still relevant until we manage to get the strings in
[03:19] <segfault> kiko: but it's there, i'm updating it right now
[03:19] <kiko> segfault, ah, rock!
[03:20] <kiko> cool
[03:20] <segfault> kiko: it was broken in three parts: libapt-pkg3.10, libapt-inst1.1 and apt.
[03:20] <kiko> interesting
[03:21] <stub> kiko: https://shipit.staging.canonical.com
[03:22] <segfault> how about some error dumps when we get system errors?
[03:24] <kiko> thanks stub
[03:24] <segfault> this way the error reports should be more detailed
[03:25] <kiko> segfault, I've thought about this before -- let me talk to SteveA 
[03:26] <segfault> like now, i was just translating some strings... and then, when i try to access this url: https://launchpad.net/distros/ubuntu/breezy/+lang/pt_BR, i get a system error.
[03:26] <asmodai> What would be the most actively used distros currently?  Need to set some up on VMWare for testing, got Fedora Core 4 and Ubuntu 5.04 now (I know 5.10 is out)
[03:27] <asmodai> (or at least preview)
[03:27] <kiko> mandriva maybe?
[03:27] <asmodai> kiko: Do they have downloadable ISOs as well?
[03:27] <kiko> hopefully!
[03:28] <asmodai> Well, you can never be sure with some :)
[03:30] <asmodai> kiko: at least linux iso has them ;)
[03:34] <kiko> BjornT, SteveA: so I can ignore any templates that start with template-* >
[03:34] <kiko> ?
[03:35] <BjornT> kiko: i would assume that
[03:39] <Kinnison> anyone still awake with twisted-fu ?
[03:39] <zakame> hello all! =)
[03:40] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Fixed some comments (patch-2543: carlos.perello@canonical.com)
[03:41] <zakame> hmmm... if there's a package that isn't in launchpad that has some bugs I'd like to report, what do I do? register that package to launchpad?
[03:42] <kiko> zakame, talk to me and I'll add it -- what is it?
[03:44] <zakame> kiko: tuxtype... it doesn't have a .desktop file
[03:44] <kiko> ok
[03:45] <zakame> kiko: thanks :)
[03:46] <kiko> zakame, it appears to already exist. tell me where you are trying to file the bug?
[03:48] <zakame> wait, I'm checking
[03:49] <Kinnison> kiko: build queue sequencer is on dogfood and running perfectly
[03:49] <kiko> hah!
[03:49] <kiko> roxorin
[03:52] <zakame> kiko: to launchpad
[03:52] <kiko> zakame, can you give me a precise URL?
[04:00] <zakame> kiko: http://launchpad.net/products/tuxtype?
[04:01] <Kinnison> does twisted have its own logging infrastructure or does it use the python logging stuff?
[04:03] <lifeless> ddaa: so you got samba live 
[04:03] <lifeless> ?
[04:03] <lifeless> ddaa: and python ?
[04:03] <ddaa> samba yes, python still running
[04:03] <lifeless> sweet
[04:03] <lifeless> wheres python up to ?
[04:03] <lifeless> 25K ?
[04:03] <ddaa> I fully expect to have to patch importd to be able to move python to hoover though
[04:04] <ddaa> 30231
[04:04] <kiko> zakame, okay, common mistake. it's not a product -- it's a package
[04:04] <Crackstar> after installung ubuntu my system is freezing on loading "starting hotplug subsystem" what can i do?! :/
[04:04] <kiko> zakame, do you want to report a bug on a package or an upstream product?
[04:04] <kiko> Crackstar, visit #ubuntu
[04:05] <Crackstar> oh thanx i see :)
[04:06] <zakame> kiko: to the ubuntu package, as i think it is ubuntu-specific
[04:07] <kiko> zakame, okay. use:
[04:08] <zakame> ???
[04:08] <mpt> salgado: Any objection if I nuke https://launchpad.net/foaf ?
[04:08] <mpt> (do we link to it from anywhere?)
[04:08] <salgado> mpt, no objection
[04:08] <salgado> and I don't think we link to it
[04:09] <kiko> BjornT, replied
[04:10] <kiko> zakame, https://launchpad.net/malone/bugs/+package
[04:10] <BjornT> cool
[04:10] <kiko> BjornT, two issues I want us to sort out before merging
[04:10] <BjornT> kiko: ok
[04:10] <Crackstar> ehm when i order ubuntu cd's is it free?
[04:11] <mpt> salgado: There are a bunch of /foaf/people/* links in pagetests, they can all have the "/foaf" stripped and still work, I assume
[04:11] <zakame> kiko: ok, thanks! :)
[04:11] <salgado> mpt, yes, that should work
[04:11] <kiko> zakame, note that if you forget to enter a package name, launchpad blows up :-(
[04:12] <kiko> bug being worked on by matsubara 
[04:14] <Crackstar> ?
[04:14] <kiko> Crackstar, yes, shipit.ubuntu.com
[04:14] <kiko> Crackstar, note that this channel is for launchpad development, you may try your luck at #ubuntu 
[04:14] <Crackstar> kiko hm nice :) and how long does it take?
[04:14] <kiko> BjornT, I just mailed you an odd failure
[04:15] <Crackstar> hm oke oke :)
[04:15] <zakame> kiko: i haven't forgotten :)
[04:15] <kiko> Crackstar, read the site
[04:15] <kiko> zakame, heh
[04:24] <kiko> SteveA, BjornT, what about default-addform?
[04:26] <BjornT> kiko: afaict it should be registered, but it shouldn't have a page title (it should be used only as a macro)
[04:27] <kiko> it's not registered afaict
[04:27] <kiko> btw
[04:27] <kiko> thanks for your help here
[04:29] <zakame> kiko: salamat anyway! :D
[04:30] <kiko> zakame, salamaleikum
[04:30] <kiko> BjornT, weird, ain't it?
[04:33] <mpt> lifeless / jblack: ping
[04:36] <BjornT> kiko: it seems like it isn't used. we also have launchpad-addform.pt which defines the same macro.
[04:40] <BjornT> kiko: reply sent
[04:41] <kiko> thanks
[04:45] <bradb> So, when can we start using bzr? Pie day is a very, very, very short time away.
[04:49] <kiko> so it is, so it is
[04:51] <bradb> hmph
[05:12] <sivang> hmm, I missed today's meeting. bad
[05:18] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=kiko]  fix bug 1733, stop bugmail notifications from being incorrectly wrapped. (patch-2544: bjorn.tillenius@canonical.com)
[05:20] <kiko> rock!
[05:28] <mdke> what is the best way to filter LP bugmail now?
[05:28] <kiko> using the D key
[05:28] <kiko> I get too much of it
[05:28] <mdke> ;)
[05:29] <mdke> i used to filter on sender iirc, but it seems now things have changed
[05:30] <mdke> it looks cool from the point of view that it seems that you can comment on bugs just by replying to the bug emails, but I can't figure out the best way to filter it using my client
[05:32] <mdke> kiko, you know the answer?
[05:34] <BjornT> mdke: at the moment it's best to filter on the Reply-To header
[05:34] <mdke> BjornT, yeah that is what I figured, but I can't see anyway that my client (evolution) does that
[05:36] <mdke> BjornT, sorry scratch that, I've found it, thanks for your help
[05:36] <kiko> mdke, BjornT and bradb are still working on the headers, we may have news next week
[05:36] <kiko> me neither
[05:36] <kiko> I'm afraid of it
[05:36] <mdke> kiko, well the headers have certainly improved because the messages are threading properly now
[05:37] <mdke> since some time tuesday
[05:37] <kiko> that's BjornT's fault
[05:37] <mdke> thanks BjornT 
[05:38] <mdke> and is my assumption that you can now post to bugs by email correct?
[05:38] <mdke> debian-style
[05:38] <BjornT> np, i'm really happy about it as well :)
[05:38] <kiko> mdke, yep
[05:38] <mdke> rock
[05:38] <mdke> nice work you guys
[05:38] <kiko> enjoy it
[05:39] <mdke> :D
[05:49] <SteveA>      * carlos.perello@canonical.com--2004/launchpad--cleanups--1--patch-5
[05:49] <SteveA>        Allow again the use of text. It makes no sense to require Unicode text always for a msgid
[05:49] <SteveA> i'm confused by this commit message, carlos
[05:49] <kiko> we used to require only unicode msgids
[05:49] <kiko> now we allow both
[05:49] <kiko> not sure why the codepath needs this
[05:50] <kiko> we should unicode() it before hitting the check no?
[05:50] <SteveA> unicode and str data are different things
[05:50] <SteveA> a given API should be clear about which it expects
[05:50] <carlos> SteveA, we had a hard requirement for Unicode strings instead of allowing both Unicode and non unicode strings
[05:50] <SteveA> or whether it accepts either and converts
[05:50] <SteveA> and, why relax that requirement?
[05:52] <carlos> SteveA, because when the hard requirement was added there, all calling points were not adapted to that hard requirement and is easier to do that from one single point than change it in all places that can reach that method with a normal string
[05:53] <carlos> SteveA, usually, the argument is just ascii
[05:53] <SteveA> where does the ascii come from?
[05:54] <SteveA> saying "ascii string" in python is saying "i want to point a gun at my foot, and pull the trigger"
[05:54] <SteveA> or, like jesus from the big lebowski, even
[05:54] <kiko> fofl
[05:55] <carlos> SteveA, from the poparser
[05:55] <SteveA> the poparser takes in some po file data
[05:55] <SteveA> which is 8 bit data, encoded in a way that is indicated inside the data
[05:55] <SteveA> so, it should always give out unicode
[05:56] <SteveA> that's its job, no?
[05:56] <carlos> SteveA, well.... poparser sucks....
[05:56] <carlos> SteveA, we are talking about msgids
[05:56] <kiko> the poparsers job is to poo on the brain of anyone reading that code
[05:56] <carlos> usually, that's ASCII
[05:56] <SteveA> "usually"
[05:56] <kiko> I mean
[05:56] <SteveA> make poparser always give out unicode
[05:56] <carlos> and the poparser follows the encoding set by the pofile
[05:57] <carlos> to encode the file. With real data, I have seen a path from the poparser that gives you ascii strings instead of unicode
[05:57] <SteveA> so
[05:57] <carlos> but I was not able to find that path to fix it
[05:57] <SteveA> at the end of the poparser
[05:57] <carlos> I mean, I found we have that path but I was not able to find it
[05:57] <SteveA> force it to be unicode
[05:57] <SteveA> then you make the "pollution" be in just one place
[05:57] <SteveA> in the poparser
[05:59] <carlos> SteveA, kiko could I schedule sometime next week to kill the poparser? daf wrote one from the scratch using python that has an API quite similar to our current parser but that has many tests and it's understandable... I'm starting to be really tired to patch a code that is full of holes and that is very hard to understand...
[06:00] <SteveA> carlos: can you estimate how long it will take?
[06:00] <carlos> I will try to give you it between tonight and tomorrow
[06:01] <Kinnison> carlos: what is msgmerge and why is it chewwing CPU on mawson?
[06:01] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=kiko]  Workaround bug # 2547, probably related to missed public key integrity, tests in doc/gpg-import.txt (patch-2545: celso.providelo@canonical.com)
[06:01] <carlos> Kinnison, I'm preparing a new language pack and doing some testings before sending it to pitti
[06:01] <carlos> Kinnison, it's merging .po files with .pot files to be able to do a diff
[06:01] <kiko> Kinnison, hold on for us, this is important :)
[06:01] <carlos> Kinnison, is that a problem?
[06:16] <SteveA> carlos: i think you should add an XXX comment in that method that it should really only accept unicode.
[06:17] <carlos> SteveA, ok
[06:17] <SteveA> Kinnison:       [trivial]  Fix slight indentation bug which could cause absorbtion of successful builds to fail with obscure unrelated error
[06:17] <SteveA> when you come across a bug, write a TEST first, then fix it
[06:18] <SteveA> i might accidentally cause a similar indentation bug in the future
[06:18] <SteveA> and without a test, it'll get merged
[06:18] <Kinnison> carlos: No problem, just wondered
[06:18] <carlos> Kinnison, ok
[06:19] <Kinnison> SteveA: #include UBZ-How-the-fuck-to-test-the-buildds.bof
[06:19] <SteveA> just do a unit test of that one method
[06:19] <SteveA> the point of unit testing is that you test small units
[06:19] <SteveA> with minimal dependencies
[06:21] <SteveA> Kinnison: where in buildmaster.py was the error?
[06:22] <Kinnison> getFileFromSlave
[06:22] <bob2> evertime you say "poparser", I see an extra "o" and snigger
[06:22] <Kinnison> inside the try: clause
[06:23] <SteveA> Kinnison: two things i'd do thre
[06:23] <SteveA> there
[06:24] <SteveA> 1. have two more methods called getFileFromSlave1 and getFileFromSlave2
[06:24] <SteveA> and delegate the different protocol version handling to each one
[06:25] <SteveA> then, note that the method doesn't use 'self' at all
[06:25] <SteveA> so, what it does is based on the module globals, and stuff passed in
[06:25] <SteveA> it doesn't use any database classes directly
[06:25] <SteveA> so it can be totally unit tested by providing appropriate stub objects as arguments
[06:26] <SteveA> and examining their state after the call
[06:26] <Kinnison> that's the bit I'm concerned about
[06:26] <SteveA> easy
[06:26] <SteveA> add an API in the class to open URLs
[06:27] <SteveA> and provide a way to initialize it with a custom url opener
[06:27] <SteveA> write your code so that it *can* be tested.  talk with lifeless about this.  he is an expert.
[06:27] <Kinnison> that'd be easiest
[06:28] <SteveA> sure, that's another possibility
[06:28] <SteveA> the point is, you're cutting out the bit of code you want to test
[06:28] <SteveA> and providing it with a minimal life-support system
[06:28] <SteveA> that you control
[06:28] <SteveA> entirely
[06:28] <Kinnison> right
[06:28] <SteveA> you'll need one set of stubs for all the build related tests
[06:30] <Kinnison> I think cprov did a bunch of mock objects for prodding at the build master
[06:31] <Kinnison> I'll do my best on the test front
[06:32] <SteveA> you'll like it a lot when you have tests you can rely on
[06:32] <kiko>     + NotFoundError: (<Person at 0x-4e32d154>, '+portlet-actions')
[06:32] <kiko> hmmm
[06:33] <SteveA> sounds like a page that wasn't tested
[06:33] <SteveA> grepping templates for +portlet-actions will show what pages probably aren't tested
[06:33] <kiko> hmmm
[06:33] <kiko> mkay
[06:34] <kiko> but odd because this is PQM returning a failure to me
[06:34] <SteveA> maybe a bad merge
[06:34] <SteveA> and it came back...
[06:35] <kiko> damn
[07:02] <SteveA> mpt: check out the maintenance message on launchpad.net
[07:03] <SteveA> and https://shipit.ubuntu.com/
[07:11] <Nafallo> ehm. _when_ will launchpad go offline and for how long?
[07:11] <Nafallo> can't you wait to tomorrow? (after NonLanguagePack-Translations-Freeze)
[07:11] <SteveA> Nafallo: 1 minute ago, for about 30 seconds
[07:12] <Nafallo> ah :-)
[07:12] <Nafallo> that's _very_ short time indeed ;-)
[07:15] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [r=kiko]  convert bug search listings to use new style listing format. (patch-2546: bjorn.tillenius@canonical.com)
[07:28] <SteveA> mpt: ping?
[07:36] <kiko-fud> SteveA took launchpad down just to show off his message
[07:36] <SteveA>     +                 title="Sample Persons scheduled events"><strong>
[07:36] <SteveA> another example of why fancy apostrophes are bad
[07:37] <SteveA> kiko-fud: still having problems with that page test?
[07:37] <kiko-fud> yes SteveA 
[07:38] <kiko-fud> it still fails
[07:38] <SteveA> i can't tell what's up from what you sent
[07:38] <kiko-fud> and I'm at the limit of my patience with it
[07:38] <SteveA> can i switch to some branch?
[07:38] <kiko-fud> but you should be able to, shouldn't you?
[07:38] <SteveA> no
[07:38] <SteveA> all i have is the output
[07:38] <kiko-fud> but the output should be enough for any competent developer, right?
[07:38] <SteveA> i also need the test, and if it's really tricky, to actually run it for myself
[07:38] <SteveA> no, i need the test
[07:38] <kiko-fud> I mean, this is a bug in pagetest, right?
[07:38] <SteveA> no
[07:39] <kiko-fud> no!
[07:39] <SteveA> you need to be able to read the test
[07:39] <SteveA> to see what it is doing
[07:39] <kiko-fud> hmmm
[07:39] <SteveA> otherwise, how can you tell something is wrong
[07:39] <SteveA> ?
[07:39] <SteveA> this is the same for any test
[07:39] <SteveA> you need to read the test case
[07:39] <kiko-fud> the error output should make it clear
[07:39] <kiko-fud> yes, you can read the test case to debug the test
[07:39] <SteveA> no
[07:39] <kiko-fud> but the error output should make it clear where your code is wrong
[07:39] <SteveA> you need to read the test case to see the conditions set up to make it wrong
[07:40] <SteveA> the error output shows you what the error is
[07:40] <SteveA> and what was unexpected
[07:40] <kiko-fud> in this case there is no error output
[07:40] <SteveA> it doesn't tell you why
[07:40] <kiko-fud> or rather
[07:40] <kiko-fud> it tells you that simply the whole page fails with no mismatches
[07:40] <kiko-fud> anyway
[07:40] <kiko-fud> let me commit for you
[07:41] <SteveA> ok
[07:41] <kiko-fud> it's going to be on christian.reis@canonical.com--lozenge/launchpad--trivialities--1--patch-37 shortly
[07:43] <SteveA> do all other tests pass?
[07:43] <kiko-fud> I was testing it individually
[07:44] <kiko-fud> so maybe
[07:44] <SteveA> eh? https://launchpad.net/bounties/chquite
[07:44] <SteveA> let me know when i can switch
[07:47] <kiko-fud> mirrorred
[07:47] <kiko-fud> SteveA, enjoy
[07:47] <Kinnison> see you tomorrow guys
[07:47] <kiko-fud> night Kinnison 
[07:59] <SteveA> kiko-fud: the test passes for me
[08:00] <SteveA>  python test.py -f --test=xx-calendar-views.txt
[08:00] <SteveA> run with that command
[08:01] <kiko-fud> trying, SteveA 
[08:02] <SteveA> wow, that's a lot of templates
[08:04] <kiko-fud> this test now works
[08:04] <kiko-fud> who can understand this?
[08:04] <SteveA> unless you can get it to break again...
[08:04] <SteveA> it is a lost cause
[08:05] <SteveA> so, those templates
[08:05] <kiko-fud> I'm running make check anyway
[08:05] <kiko-fud> yeah
[08:05] <kiko-fud> all tests pass of course
[08:05] <SteveA> there's a large commented out section in pagetitles
[08:05] <SteveA> i say lose it
[08:05] <SteveA> D  lib/canonical/launchpad/templates/cvereference-index.pt
[08:05] <SteveA> D  lib/canonical/launchpad/templates/cvereferences-index.pt
[08:05] <SteveA> i'm surprised at those two
[08:06] <kiko-fud> let's leave the pagetitles there, just for this week
[08:06] <SteveA> why?
[08:06] <SteveA> they're in rcs
[08:06] <kiko-fud> I suspect I'll need to revert and add tests for templates that are actually used
[08:06] <kiko-fud> or should have been
[08:06] <kiko-fud> hmm
[08:06] <kiko-fud> maybe not I guess
[08:06] <SteveA> or, keep it on your hdd
[08:06] <kiko-fud> yeah, but baz is not really an rcs
[08:06] <SteveA> but remove it from baz
[08:06] <kiko-fud> it's more of a code-grinder
[08:07] <kiko-fud> SteveA, patch looking okay?
[08:08] <SteveA> yep
[08:08] <Burgundavia> jordi, now listed on RosettaPendingImports. for some reason, it got renamed .bin, simply rename it back to .pot
[08:09] <kiko-fud> jordi?
[08:22] <kiko> SteveA, please run make check on that branch you have of mine
[08:22] <kiko> it will fail in that test
[08:24] <SteveA> ok
[08:28] <ondrej> mm all
[08:29] <ondrej> I uploaded partial (100+ strings) czech .po translation to rosetta (yesterday and this morning again), but still no changes in upload.
[08:29] <SteveA> kiko: i get a bunch of errors
[08:29] <ondrej> s/upload./translations/
[08:30] <ondrej> looks like it crashed somewhere in between
[08:30] <kiko> SteveA, me too, but that file errors out -- is that okay?
[08:30] <kiko> ondrej, what package/template?
[08:30] <ondrej> faqguide
[08:30] <SteveA> it's not really okay, but it is going to be tough to debug something when there are so many other errors
[08:31] <kiko> okay.
[08:31] <SteveA> i gotta go buy food then cook it and eat it
[08:32] <SteveA> mail me if it's still a problem later, and i'll look at it in the morning
[08:32] <ondrej> I am using po-mode for editing, so there should not be some _big_ errors in .po file
[08:42] <kiko> uhm
[08:42] <kiko> SteveA, how can it be that we still have portlet-actions on certain templates?
[08:48] <kiko> bradb, ping?
[08:48] <bradb> yo
[08:48] <kiko> bradb, do you have a fully-merged tree?
[08:48] <bradb> yeah
[08:49] <kiko> bradb, can you check person-images and see if it contains a portlet-actions line still?
[08:49] <kiko> person-images.pt
[08:49] <bradb> the reason we still have portlet-actions is because SteveA did some "THIS FILE IS DEAD" thing instead of baz rm, so it's hard to see what still references it.
[08:49] <bradb> i'll check that file
[08:49] <bradb> yup, still in there
[08:49] <kiko> sheesh
[08:50] <kiko> it's not registered in zcml any longer
[08:50] <kiko> so it causes blowups here and there
[08:50] <kiko> okay, will attempt to fix
[08:50] <bradb> ah, ouch
[08:50] <bradb> in that particular case, sounds like it's just plain not tested then
[08:51] <kiko> bradb, it is in my tree
[08:51] <kiko> bradb, can I just delete the line?
[08:51] <bradb> in theory, yes
[08:51] <bradb> that "should work"
[08:52] <bradb> kiko: what do you think reads better, Reply-To: Malone Bug 42 <42@...> or ReplyTo: Foo Bar via Malone <42@...>
[08:52] <kiko> I don't care too much, bradb 
[08:52] <bradb> ok
[08:52] <kiko> I think the former is easier
[08:52] <kiko> and the person's name is already there, right?
[08:53] <bradb> yeah
[08:57] <jordi> hello
[09:43] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  fix bugmail Reply-To encoding bug (#2592) (patch-2547: brad.bollenbach@canonical.com)
[09:58] <kiko> rock bradb 
[09:58] <kiko> rock!
[09:58] <bradb> i cheated though, by sidestepping the issue
[09:58] <kiko> :)
[09:58] <kiko> that's sometimes the best way
[09:59] <bradb> indeed
[10:21] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: r=kiko, salgado. Multiple fixes foiled by odd PQM failure. Fixes for bug 2166: System error if trying to subscribe to a calendar while not  logged in and bug 2328: Calendar should display this month, this week, this year instead of today for month, week and year views. Fixes bug 2501: edit hackergotchi link disabled. Adds hackergotchi pagetests and rewrites the validator code. Fixes bug 2580: Declined m
[10:37] <ajmitch> kiko-afk: looks like pyflakes built & is in the archive
[10:54] <BjornT> bradb: i finally put up my branch for review now. i probably should tell you about some design decisions, but i'm too tired atm... if you have any questions you can mail me.
[10:57] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: r=stevea Removing unused templates, and adding missing pagetitles. Commented out dead pagetitles, will clean out next week if no cries occur. (patch-2549: christian.reis@canonical.com)
[11:02] <mpt> SteveA: pong
[11:10] <bradb> BjornT: ok, sounds good
[11:17] <mpt> kiko-afk: Did those include *dashboard.pt?