[12:14] <kiko> LarstiQ, you know you can use ?batch=300 and then search using firefox?
[12:14] <matsubara> kalosaurusrex: did you receive any error message for the email you've sent to test? 
[12:21] <kalosaurusrex> no I didn't.
[12:29] <LarstiQ> kiko: ah, that does help
[05:16] <pulaski> hello l
[05:16] <pulaski> hello launchpad
[05:20] <jamesh> hi stub
[05:20] <stub> hi
[05:21] <jamesh> I tried using sudo on carbon.ubuntu.com, but it says I don't have permission to change users to launchpad
[05:38] <stub> Hmm... my rt job explicitly requested it along with setting up the user. I can setup a workaround using ssh keys. Can you paste your public ssh key somewhere?
[05:38] <stub> Unless elmo is alive
[05:45] <spiv> stub: https://launchpad.net/people/jamesh/+sshkeys ;)
[05:47] <jamesh> stub: the one spiv mentioned above will do (it is the one I have in LDAP)
[05:55] <stub> jamesh: Try ssh launchpad@carbon
[05:55] <jamesh> stub: that worked, thanks.
[05:56] <stub> Anyone else need me? I need to bugger off for a few hours (friends of the wife in town)
[06:51] <jamesh> well, I can parse all the Python tracker data, which is a start ...
[07:31] <mpt> Gooooooooooooooooood evening Launchpadders!
[09:01] <jamesh> hi danilos 
[09:04] <SteveA> morning
[09:04] <danilos> jamesh: hi jamesh 
[09:13] <SteveA> jamesh: I'm going to change that specs display branch to use dot.  With that and the replies in my review, do you want to see the branch again before I merge it?
[09:14] <jamesh> SteveA: that would be fine to merge r=jamesh
[09:14] <SteveA> thanks
[09:15] <jamesh> SteveA: as for the possible performance side, the graphviz bindings probably expose the dot file reader
[09:16] <sivang> morning
[09:16] <jamesh> SteveA: although I wonder if it would be better to work out when the PNG and image map can be cached
[09:17] <jamesh> if the graphs get too big, the layout process starts to cost a bit
[09:29] <danilos> if "make check" hangs on pagetests/standTests for 600 seconds, what does that mean? (I can't find referenced "standTests" file at all)
[09:30] <danilos> (and all the standalone tests seem to have passed)
[09:33] <jamesh> SF tracker importer seems to be working with the Python dataset
[09:41] <sivang> SteveA: is there anything to see of the python specs import on staging already ?
[09:42] <SteveA> sivang: no.  we're not importing PEPs yet
[09:43] <sivang> SteveA: ah, okay.
[09:49] <carlos> morning
[09:50] <sivang> morning carlos 
[09:55] <danilos> morning carlos
[09:57] <jamesh> bug comment ordering appears to be screwed up in rocketfuel head
[09:58] <carlos> danilos: so, are your changes working?
[09:58] <danilos> carlos: yeah, but I am having some problems with "make check"
[09:59] <carlos> which kind of problems?
 if "make check" hangs on pagetests/standTests for 600 seconds, what does that mean? (I can't find referenced "standTests" file at all)
 (and all the standalone tests seem to have passed)
[09:59] <carlos> danilos: execute ./test.py -vvv
[10:00] <carlos> that should increase the verbosity of the tests
[10:01] <danilos> btw, I've pushed my branch to sftp://chinstrap.ubuntu.com/home/warthogs/archives/danilo/bug-2237/ as well
[10:03] <jamesh> danilos: if your changes are fairly self contained, you can restrict which tests get run by using test.py directly
[10:03] <spiv> danilos: you mean sftp://chinstrap.ubuntu.com/home/warthogs/archives/danilo/launchpad/bug-2237/, I presume
[10:04] <danilos> spiv: yeah :)
[10:04] <spiv> danilos: Also, you ought to add it as a work-in-progress branch to the PendingReviews page.
[10:04] <danilos> spiv: I will, but I wanted all the tests to pass first
[10:05] <carlos> danilos: work-in-progress is just a way to point to others to your work, no one will review it until you change it to needs-review
[10:05] <danilos> I mean, it's "complete" if it works :)
[10:05] <danilos> carlos: yeah, that's ok, I didn't expect this many problems with tests, so I'll add it right away
[10:31] <SteveA> jamesh: got a minute to help me understand some DOT issue?
[10:32] <jamesh> SteveA: sure.
[11:08] <elmo> lifeless: ping?
[12:04] <lifeless> pong
[12:06] <lifeless> elmo: pong
[12:07] <elmo> lifeless: sent mail - just asking about postgres on chinstrap, not urgent
[12:07] <lifeless> theres a postgres on chinstrap ?
[12:07] <Kinnison> I imagine it used to be there for the test suite
[12:11] <carlos> wow SQL sucks sooo much...
[12:11] <lifeless> elmo: PURGE away
[12:11] <carlos> so foo1.bar = foo2.bar is false if both are NULL...
[12:12] <Kinnison> indeed
[12:12] <Kinnison> NULL <=> ANYTHING == False
[12:12] <carlos> that sucks
[12:12] <Kinnison> carlos: otherwise a UNIQUE column which is NULLable wouldn't work
[12:12] <Kinnison> and those are *really* useful
[12:13] <elmo> lifeless: thanks
[12:14] <spiv> carlos: it's by design.  Operations involving nulls generally produce more nulls.
[12:14] <malcc> If you like concepts, you can think of SQL null meaning "unknown"
[12:14] <spiv> carlos: they aren't like None in Python, tempting though it is to pretend they are.
[12:15] <carlos> spiv: yeah, I was confused by C and Python programming
[12:15] <carlos> It took me a while until I figured that....
[12:23] <carlos> test pass now ;-)
[12:24] <lifeless> spiv: well, if None meant undefined, rather than pointer to 0x0, then they would be the same, its really just the NULL != NULL that makes them different
[12:28] <spiv> lifeless: sure, but behaving differently when used in comparisons is a huge practical difference.
[12:30] <Kinnison> carlos: I got my assertion wrong earlier
[12:30] <Kinnison> NULL <=> NULL yields NULL, not false
[12:31] <lifeless> Kinnison: which is 'is one unknown equal to another?' well the answer myfriend, is unknown.
[12:31] <Kinnison> indeed
[12:31] <Kinnison> If I wasn't so used to undef() in perl, I'd be damned confused by NULL
[12:33] <carlos> Kinnison: ok
[01:22] <lifeless> salgado: ping, your review queue is getting kinda long
[01:25] <salgado> hey lifeless.  it wasn't long until this morning, when you added two new ones. ;)
[01:26] <lifeless> salgado: well, it had one entry, but that entry was 3 days old at my morning :)
[01:26] <salgado> anyway, I have to admin I didn't do very good on managing it this week... I'll do some code review now!
[01:26] <lifeless> np, thanks.
[01:26] <lifeless> see y'all monday
[01:26] <salgado> lifeless, have you seen stub today?
[01:26] <lifeless> salgado: nag kiko about his outstanding review too ;)
[01:27] <lifeless> stub? nmope
[02:02] <carlos> Hmm, I'm getting an error and I don't understand why....:
[02:02] <carlos>         assert output == u'', output
[02:02] <carlos>     AssertionError: <unprintable instance object>
[02:03] <carlos> but
[02:03] <carlos>         assert output == u'', type(output)
[02:03] <carlos>     AssertionError: <type 'unicode'>
[02:03] <carlos> if it's a unicode object, shouldn't it be printable?
[02:03] <jamesh> carlos: the unicode string can not be printed
[02:04] <jamesh> e.g. if it is not ascii
[02:04] <carlos> jamesh: that concrete unicode string ?
[02:04] <carlos> hmmm, should I recode it as UTF-8 ?
[02:04] <jamesh> carlos: repr(output) might give more useful output
[02:05] <carlos> let's try...
[02:05] <jamesh> if it is that important, then perhaps an assertion isn't appropriate
[02:07] <SteveA> mpt: ping
[02:07] <carlos> jamesh: well, it's part of a test
[02:08] <carlos> I'm checking that a diff output is empty
[02:08] <jamesh> you can't do self.assertEqual(output, u'')?
[02:08] <mpt> SteveA, yes?
[02:09] <carlos> jamesh: it's a doctest
[02:09] <jamesh> carlos: in that case, just do:
[02:09] <jamesh> >>> output
[02:09] <jamesh> u''
[02:09] <carlos> It's part of a loop
[02:10] <jamesh> ah.
[02:10] <carlos> so I would need to put some u'' after it
[02:10] <carlos> is that ok?
[02:10] <jamesh> if it is in a loop, then it would be good to give some context about which iteration of the loop the test failed in
[02:11] <jamesh> e.g. putting "print i, repr(output)" in the loop to show the output from each iteration
[02:11] <carlos> yeah, I had it, but I was not able to show the 'output' content so I was playing with output directly
[02:12] <carlos> so instead of using an assert, I should use an 'if'
[02:27] <salgado> stub!
[02:28] <stub> yes?
[02:28] <salgado> yesterday, after you left, I realized there was a problem with using that table you suggested, that could possibly make it as slow as the view
[02:29] <salgado> stub, https://chinstrap.ubuntu.com/~dsilvers/paste/fileFAD0lJ.html
[02:31] <stub> I doubt it will be as slow as the view using the OUTER JOIN, and we can use a trigger to populate a value in the table if we want to avoid it altogether.
[02:31] <stub> Good catch.
[02:32] <salgado> after that I discussed with kiko and we decided to do something that took longer to implement, but allowed me to simplify the code in lots of places
[02:32] <salgado> actually, it took longer because I had to change lots of places
[02:33] <salgado> basically, what we decided to do was to drop the cancelled and approved columns and create a new status column
[02:35] <salgado> this way we can have a new status to flag the requests that we'd be flagging by counting the previous requests
[02:36] <stub> Ok. This sounds like a good plan.
[02:36] <salgado> cool. the only drawback is that it now has to inlcude lots of db changes
[02:37] <salgado> stub, https://chinstrap.ubuntu.com/~dsilvers/paste/filecw2bUh.html is the whole patch. it'd be great if you could review the db patch
[02:38] <salgado> (assuming I'll add a "drop cancelled", which I just realized I forgot)
[02:42] <stub> salgado: We can simplify the constraints too if you have a status for 'admin request'
[03:07] <stub> carlos: have you landed your rosetta duplicate fixup patch yet?
[03:18] <ondrej> hi, there seems to be an error in adding attachments to bugs, it ends with OOPS-195B313
[03:18] <Ubugtu> https://chinstrap.ubuntu.com/~jamesh/oops.cgi/195B313
[03:20] <matsubara> ondrej: that oops isn't available yet. In about 10 min I'll check that
[03:31] <carlos> stub: no, I'm waiting for your approval
[03:37] <stub> approved
[03:44] <carlos> stub: ok, thanks
[03:48] <salgado> stub, did you have a look at the new shipit db patch?
[03:50] <stub> salgado: reworking it now with data migration and improved indexes
[03:50] <salgado> cool! did I miss too many things?
[03:52] <stub> salgado: Why don't we have a 'SHIPPED' status? It seems silly to have requests stay as PENDING after they are shipped
[03:52] <salgado> s/PENDING/APPROVED?
[03:52] <stub> Yer
[03:52] <salgado> I guess you're right. I haven't thought about that, but it sounds like a good idea
[03:54] <ondrej> matsubara: ping? did you find anything interesting?
[03:56] <matsubara> bradb_: the +addattachment is 404ing
[03:57] <matsubara> ondrej: you can try to add and attachment while adding a new comment.
[03:57] <bradb_> matsubara: yup
[03:57] <bradb_> it's +addcomment now
[03:57] <bradb_> if there's a link hanging around to +addattachment, it should be fixed
[03:57] <matsubara> bradb_: hmm there's one on the portlet.
[03:58] <salgado> stub, if you want to use this new status on a constraint, please assume it's the dbschema item with id 4. I'm adding it now
[03:58] <kiko> bradb_, next time you change a page, USE GREP.
[03:58] <kiko> launchpad/templates/bug-portlet-attachments.pt
[03:58] <matsubara> ondrej: try this page https://launchpad.net/distros/ubuntu/+source/linux-source-2.6.15/+bug/52546/+addcomment
[03:58] <Ubugtu> Malone bug 52546 in linux-source-2.6.15 "original HP keybord bundled with HP dc7600 doesn't work" [Untriaged,Confirmed]  
[03:58] <bradb_> hrmph, i guess i didn't change the attachments portlet
[03:58] <bradb_> my bad
[03:59] <matsubara> ondrej: if that is the bug you'd like to add an attachment on. :)
[04:03] <ondrej> matsubara: ah, ok...  thanks, I would forgot about that if I left it to some other time... you know :-)
[04:04] <bradb_> flacoste: did you get my reply? j't'au bureau en ce moment.
[04:04] <flacoste> bradb_: non, j'ai pas eu ton reply, mais whois me l'a dit ;-)
[04:04] <bradb_> heh
[04:05] <flacoste> bradb_: I'm leaving for the office about now, should be there in about 30mins
[04:05] <bradb_> sounds good
[04:10] <salgado> danilos, around?
[04:27] <stub> salgado: https://chinstrap.ubuntu.com/~dsilvers/paste/fileocA571.html
[04:28] <salgado> stub, great, thanks!
[04:55] <jamesh> bradb_: https://demo.launchpad.net/projects/python/+bugs
[04:55] <bradb_> jamesh: cool! i'll have a closer look later today.
[04:55] <riccardomurri> hello
[04:57] <riccardomurri> i'm having trouble pushing my first branch in launchpad bzr space
[04:58] <riccardomurri> bzr push sftp://.... does apparently nothing
[04:58] <jamesh> riccardomurri: do you have paramiko installed?
[04:58] <riccardomurri> jamesh: yes
[04:58] <riccardomurri> jamesh: paramiko-1.5
[04:59] <jamesh> riccardomurri: why do you think it did nothing?  Was there an error message?
[04:59] <riccardomurri> jamesh: "0 revision(s) pushed." where the branch actually has 2...
[05:00] <jamesh> riccardomurri: what branch name did you use?
[05:00] <matsubara> jamesh: more details about what riccardomurri has tried here: https://launchpad.net/products/launchpad/+ticket/1197
[05:00] <riccardomurri> matsubara: thanks
[05:02] <jamesh> riccardomurri: and there is no directory called "sftp:" created after your push?
[05:02] <jamesh> locally, that is
[05:03] <riccardomurri> yes, there is one
[05:03] <jamesh> riccardomurri: with bzr 0.8, if it can't find paramiko, sftp paths get treated like local filesystem paths
[05:04] <jamesh> riccardomurri: so for some reason bzr can't initialise paramiko
[05:04] <riccardomurri> jamesh: oh, thanks, i'll look into that, then
[05:05] <jamesh> riccardomurri: the "0 revisions pushed" message would be because the branch you were pushing to (which turns out to be another one on your local disk) already contained the two revisions
[05:06] <riccardomurri> jamesh: found - the python-crypto is installed for python-2.3 and not for python-2.4
[05:06] <riccardomurri> jamesh: so paramiko does'nt work in py2.4
[05:06] <riccardomurri> jamesh: thank you very much!
[05:06] <jamesh> no problem
[05:08] <matsubara> riccardomurri: would you summarize the answer (so others can find the solution later) and mark the ticket 1197 as anwsered?
[05:08] <riccardomurri> jamesh: yes, but...
[05:08] <riccardomurri> jamesh: ...now i get another error:
[05:09] <riccardomurri> jamesh: bzr: ERROR: exceptions.IOError: .bzr \n at /usr/lib/python2.4/site-packages/bzrlib/transport/sftp.py line 540 \n in _translate_io_exception
[05:12] <jamesh> riccardomurri: did you happen to create the ~riccardo-murri/vc-bzr/current directory on the sftp server before?
[05:13] <riccardomurri> yes,
[05:13] <riccardomurri> but i;ve tried changing to bzr push sftp://..../devel and the error is the same
[05:13] <jamesh> riccardomurri: okay.  That is probably the problem
[05:13] <jamesh> riccardomurri: can you try running the following Python commands:
[05:13] <jamesh> import bzrlib.branch
[05:14] <jamesh> bzrlib.branch.Branch.initialize('sftp://riccardo-murri@bazaar.launchpad.net/~riccardo-murri/vc-bzr/current')
[05:15] <riccardomurri> jamesh: a long stack trace...
[05:15] <jamesh> oops.  Did that give a NotLocalUrl error?
[05:16] <jamesh> try this instead:
[05:16] <jamesh> import bzrlib.bzrdir
[05:16] <riccardomurri> jamesh: oops... i changed my username from that reported in ticket1197
[05:16] <riccardomurri> jamesh: i'll retry the python code with the new username
[05:17] <jamesh> bzrlib.bzrdir.BzrDir.create_branch_and_repo('sftp://...')
[05:17] <riccardomurri> jamesh: yes, raise errors.NotLocalUrl(base)
[05:17] <jamesh> try the create_branch_and_repo() one instead
[05:18] <riccardomurri> jamesh: full stack trace at http://no-paste.it/?217
[05:19] <jamesh> riccardomurri: yeah.  I was wrong about that command.  I think the BzrDir one above should work.
[05:20] <riccardomurri> jamesh: with create_branch_and_repo() did another stack trace, see http://no-paste.it/?218
[05:21] <jamesh> riccardomurri: this is a weird one.  What is in that directory, if you just look at it with sftp?
[05:23] <riccardomurri> jamesh: now there's a .bzr directory with some content... but not all, apparently (see http://no-paste.it/?219)
[05:23] <kalosaurusrex> any dba's on?
[05:24] <jamesh> riccardomurri: okay.  Could you try deleting the files in .bzr/ and the .bzr/ directory itself with sftp
[05:24] <jamesh> and then run bzrlib.bzrdir.BzrDir.create_branch_and_repo('sftp://rmurri@bazaar.launchpad.net/~riccardo-murri/vc-bzr/current', True)
[05:25] <riccardomurri> jamesh: i get an error trying to delete .bzr/branch-lock
[05:25] <jamesh> riccardomurri: branch-lock is a directory
[05:27] <riccardomurri> jamesh: IOError: /~rmurri/vc-bzr/current/.bzr/repository/inventory.kndx
[05:27] <riccardomurri> jamesh: full backtrace at http://no-paste.it/?220
[05:29] <jamesh> riccardomurri: okay.  This is getting a bit beyond the scope of my knowledge.  I'm not sure how to fix this
[05:30] <riccardomurri> jamesh: thank you very much for your help, anyway
[05:40] <danilos> salgado: ping
[05:44] <salgado> hi danilos. I've been assigned to review your first branch (you can see review assignments on https://chinstrap.ubuntu.com/~jamesh/pending-reviews/), and I was thinking that maybe it'd be better to review that together with you, on IRC
[05:44] <salgado> what do you think?
[05:47] <danilos> salgado: sure
[05:49] <salgado> danilos, is now a good time to do it?
[05:49] <danilos> salgado: yeah, if it suits you
[05:50] <glatzor> carlos: ping
[06:34] <SteveA> matsubara: do you know who maintains the launchpad-dependencies package?
[06:34] <matsubara> SteveA: lifeless I think
[06:34] <SteveA> okay, thanks
[06:39] <matsubara> fwiw, https://lists.ubuntu.com/mailman/private/launchpad/2006-April/008621.html SteveA 
[06:44] <SteveA> thanks matsubara-lunch 
[06:46] <Laibsch> OK, you asked for feedback.
[06:46] <Laibsch> I would like to see a console app to access or at least report bugs for launchpad.
[06:46] <Laibsch> I always enjoyed reportbug (Except that mails were not masked on the web interface)
[06:47] <kiko> Laibsch, planned and will be delivered for edgy.
[06:47] <SteveA> you can report bugs by email too, btw
[06:47] <Laibsch> kiko: glad to hear that.
[06:48] <Laibsch> SteveA: Where?  But I guess not with the sleekness of reportbug.
[06:49] <Laibsch> bug 7839 is not so good-looking as far as a schedule for improving bug reporting in ubuntu goes.
[06:49] <Ubugtu> Malone bug 7839 in Ubuntu "Ubuntu bug reporting tools need to point to Ubuntu bug systems" [High,Confirmed]  http://launchpad.net/bugs/7839
[06:49] <Laibsch> Ubugtu: I like bots ;-)
[06:52] <carlos> glatzor: pong
[06:52] <flacoste> kiko: got any new opinions on the single/multiple best answers in SupportTrackerWorkflowSpec?
[06:53] <glatzor> Hi carlos, have you read the mail on launchpad-users from the KDE guy?
[06:53] <glatzor> There is really a lack of communication.
[06:53] <carlos> yeah, I talked with Riddell about it
[06:53] <carlos> I will send an answer next Monday
[06:53] <carlos> I think so, yes
[06:53] <glatzor> carlos: I also answered on some issues
[06:54] <carlos> ok, thanks
[06:54] <glatzor> I hope that I haven't written something wrong :)
[06:55] <carlos> don't worry, If I detect something wrong I will clarify it ;-)
[06:55] <Laibsch> bye.
[06:58] <glatzor> carlos: by the way I can confirm that koffice, especially klexi, the database tool that was largely rewritten, isn't 100% translated in Ubuntu dapper only using language-pack-de-kde
[06:59] <glatzor> although the upstream translation is 100%
[07:00] <carlos> glatzor: Riddell told me yesterday that it could be a bug on koffice, he has a patch that should fix that
[07:02] <glatzor> carlos: fine.
[07:03] <glatzor> thanks a lot.
[07:04] <carlos> np
[07:08] <salgado> kiko, ready for another review? ;)
[07:21] <bradb_> kiko: fwiw, i updated my landscape hack now that the branch it depends on landed
[07:26] <kiko> bradb_, thanks.
[07:26] <kiko> salgado, yes, shortly, need to finish fixing my regression.
[09:02] <mdz> Kinnison,cprov: the display at https://launchpad.net/distros/ubuntu/+source/xine-lib is a bit weird; can you explain it?
[09:02] <mdz> it lists the same version twice, once as Published and once as PendingRemoval
[09:02] <Kinnison> override change?
[09:03] <elmo> no, it was uploaded at the same time by two people and accepted twice
[09:03] <Kinnison> urgh
[09:03] <elmo> O  [  48: Tollef Fog Heen     ]  Accepted xine-lib 1.1.2-2ubuntu2 (source)
[09:03] <elmo> O      <  52: Reinhard Tartler    >
[09:03] <cprov> uhhhh
[09:04] <cprov> we can't be sure about which one survived ...
[09:07] <elmo> there use to be a bug about this, but I can't find it anymore
[09:08] <mdz> that bug is still unfixed? argh
[09:09] <elmo> yes, definitely still there, there was several this week
[09:09] <elmo> (I see them so easily because the way gnus sorts my mail makes them jump out)
[09:09] <elmo> O  [ 138: Scott James Remnant ]  Accepted dhcdbd 1.14-2ubuntu1 (source)
[09:09] <elmo> O      < 138: Scott James Remnant >
[09:10] <elmo> was another recent one
[09:10] <elmo> (tho that may have been a crash and dupe mail since it was the same uploader, so err, ignore that)
[09:11] <mdz> I think it's https://launchpad.net/products/qprocd/+bug/34210
[09:11] <Ubugtu> Malone bug 34210 in qprocd "should reject uploads whose versions already exist in other distroreleases" [High,Confirmed]  
[09:11] <mdz> however there is also https://launchpad.net/products/qprocd/+bug/31038 which is supposed to have been fixed
[09:11] <Ubugtu> Malone bug 31038 in qprocd "two accept messages for different udev 079-0ubuntu9 uploads" [Critical,Fix released]  
[09:12] <mdz> cprov: is the fix for 31038 rolled out to drescher?
[09:13] <cprov> mdz: yes, it is 
[09:15] <kiko> hmmm
[09:15] <kiko> pqm don't love me
[09:15] <salgado> but I do
[09:15] <salgado> how about some code review?
[09:16] <kiko> ["PQM Cannot merge between different VCSsystems.
[09:16] <kiko> +'chinstrap.ubuntu.com:/home/warthogs/archives/kiko/launchpad/trivialities-new'(pqm.Baz1_1Hand
[09:16] <kiko> +ler) and '/home/pqm/archives/rocketfuel/launchpad/devel'(pqm.Bazaar2Handler) are different."] 
[09:16] <kiko> has anyone ever seen that?
[09:16] <salgado> '/home/pqm/archives/rocketfuel/launchpad/devel'
[09:16] <salgado> you lack an chinstrap.ubuntu.com: there, no?
[09:16] <salgado> maybe not
[09:16] <salgado> where does that /home/pqm comes from?
[09:17] <kiko> that's where pqm commits to.
[09:17] <kiko> I think I'm going to start a new branch and @#!@$! this
[09:25] <mdz> cprov: then it would appear that the fix hasn't corrected the problem
[09:27] <cprov> mdz: yes, I will investigate why, can you reopen it ?
[09:31] <mdz> cprov: already done
[09:31] <cprov> mdz: okay, thanks
[09:32] <mdz> cprov: perhaps in this case the uploads were processed in the same batch?
[09:32] <cprov> mdz: indeed
[09:38] <kiko> salgado, where's the patch?
[09:39] <salgado> kiko, https://chinstrap.ubuntu.com/~dsilvers/paste/fileE32IrK.html
[09:39] <salgado> kiko, stub suggested adding a SHIPPED status, and I did that because it seemed to be a good idea
[09:39] <kiko> ok.
[09:39] <salgado> but it has one caveat, marked with an XXX that you'll see
[09:42] <kiko> sure thing
[09:46] <Keybuk> elmo: no that was just me being naughty
[09:50] <carlos> enjoy your weekend!!
[09:55] <danilos> out guys, enjoy it :)
[09:56] <kiko> bradb_, review coming up.
[10:36] <bradb_> kiko: cool, looking now. (i'd been pairing with flacoste)
[11:13] <bradb_> BjornT: around?
[11:28] <bradb_> kiko: tbh, that notify() thing has me friggin' confused
[11:30] <bradb_> kiko: I can't figure out how newbug mails are even being sent right now, because it looks like the necessary event is not getting published.
[11:31] <bradb_> but, it doesn't look like that bit has changed since the last rollout
[11:34] <bradb_> ugh, i think i know why
[11:49] <kiko> bradb_, oy?
[11:49] <bradb_> kiko: dude, I think i understand what's going on
[11:49] <bradb_> it is teh suck
[11:50] <kiko> ai ai
[11:50] <kiko> on a friday evening
[11:50] <bradb_> kiko: basically, i think that if you file a bug at /distros/ubuntu/+filebug and specify a packagename, no bugmail gets sent out
[11:50] <kiko> !
[11:50] <kiko> how did the distro guys not detect this?!
[11:50] <kiko> that can't be 
[11:50] <bradb_> yes, it can :/
[11:51] <bradb_> it's not too too hard to overlook, because it'll often work
[11:51] <kiko> bradb_, is this underlining my argument that the database classes should trigger events, not the browser?
[11:51] <kiko> the browser /code/ sorry
[11:51] <bradb_> i.e. +package, the package-specific +filebug, and +filebug /without/ a packagename should all be sending mail
[11:51] <bradb_> kiko: possibly. either way, the point was that the event publishing was not tested.
[11:52] <kiko> well
[11:52] <kiko> if it was done in the database code
[11:52] <kiko> it would have been well tested. :)
[11:54] <bradb_> kiko: i think the quickest way to fix this right now is to fix it in the browser code. because all the tests manually publish the events.
[11:55] <kiko> is the quick fix the correct one in this case?
[11:56] <bradb_> kiko: i think so.
[11:57] <bradb_> afterall, browser:addform and browser:editform published events for us, so publishing events from browser code is somewhat ubiquitous
[11:57] <kiko> I think that's just bad design
[11:58] <bradb_> it might be, but i don't feel like this is the right moment to consider an alternative.
[11:58] <bradb_> it will be a huge change to move it into the db code
[11:59] <kiko> I guess.
[11:59] <kiko> hmm
[11:59] <kiko> you need to understand my reluctance to go with band-aiding this
[11:59] <kiko> because it's a band-aid that will bite us again
[11:59] <kiko> as soon as we add a different interface
[11:59] <bradb_> i understand that the db code might be the right place for this
[12:00] <bradb_> i'm not disputing that
[12:00] <bradb_> but it's not "band-aiding", i'm just fixing something that somehow got indented when it shouldn't have.
[12:01] <bradb_> a "band-aid" implies coming in with some gross hack to make something work
[12:01] <kiko> yeah, yeah.
[12:01] <kiko> just unhappy at this turn of events.
[12:04] <bradb_> me too
[12:04] <kiko> anyway
[12:05] <bradb_> kiko: I can have a patch ready in about 15 mins if you want to review
[12:05] <kiko> with the test you mean?
[12:06] <bradb_> kiko: yeah, that's what i need the other 14m59s for :)
[12:06] <bradb_> I'm writing it now