[07:08] <lifeless> bob2: I'm on leave.
[07:08] <bob2> lifeless: oh, have fun :-)
[07:08] <bob2> and say hi to lynne.
[07:10] <lifeless> roger wilco
[12:40] <Kinnison> Anyone using dogfood? I want to upgrade it
[12:46] <Kinnison> Okay, upgrading...
[01:32] <Kinnison> Upgrade completed, dogfood back
[03:54] <mpt> Hey morgs, did you get lots of hacking and caregiving done while the rest of us were sunning ourselves in Brazil?
[03:56] <morgs> mpt: I got blocked and lonely... bring the wiki back!
[03:56] <mpt> Wiki's coming back today
[03:56] <mpt> if elmo has time to do it
[03:56] <mpt> and/or kiko
[03:59] <morgs> hacking and caregiving are not very compatible... but /me is getting more productive again...
[03:59] <morgs> Get some new specs done in between mosquitos and sunburn?
[04:00] <mpt> oh! actually, I think http://lpwiki.async.com.br/ is accessible from outside Async now
[04:00] <mpt> at least, until the wiki is transferred back
[04:00] <mpt> so if you want to do any wikiing before then, you can
[04:01] <mpt> So yeah, http://lpwiki.async.com.br/RecentChanges
[04:01] <morgs> Hmm, maybe accessible from specific places outside async that do not include Cape Town though...
[04:02] <morgs> "The hostname was not found during the DNS lookup."
[04:07] <mpt> morgs: Well, in the meantime, there are 9 bugs assigned to you :-)
[04:07] <SteveA> hi
[04:08] <SteveA> morgs: i know exactly how to fix that error you were getting with rdf stuff
[04:08] <mpt> hi SteveA, https://launchpad.net/malone/bugs/1244 is an example of docstrings not being suitable for help
[04:10] <SteveA> mpt: right.  which is kinda what docstrings are for, for a python programmer
[04:10] <mpt> yes
[04:10] <SteveA> i'll need to think about how to fix this for us
[04:10] <SteveA> one way would be to make a new schema, that extends the old one
[04:10] <SteveA> with corrected "presentation type" docstrings
[04:10] <mpt> A schema for auto-generated forms?
[04:11] <SteveA> another way would be to see this as a problem that should be fixed upstream
[04:11] <SteveA> by having different programmer_description and user_description fields
[04:11] <SteveA> (with better names, i should think)
[04:11] <SteveA> mpt: yes
[04:11] <SteveA> we're going to be moving all the schemas that are used only for autogenerated forms into browser/
[04:12] <mpt> I was sure I'd started writing a spec on how auto-generated forms should be presented, but I can't find it
[04:12] <SteveA> so, the simplest solution, at least for now, is to do that, and to extend "programmer schemas" with ones in launchpad/browser when we need to
[04:13] <morgs> SteveA, hi, great!
[04:13] <SteveA> morgs: here is what you need to do:
[04:14] <SteveA> 1. change how you wrote the page to include a ViewPageTemplateFile as an attribute of the class, rather than saying 'template="..."' in the zcml
[04:14] <SteveA> 2. make your zcml say that the page is rendered by calling your view class, using 'attribute="__call__"'
[04:16] <SteveA> 3. in your view class' '__call__' method, set the various http headers you need to, say 'unicodedata = self.template()' and then say 'encodeddata = unicodedata.encode('utf-8')' and then 'return encodeddata'
[04:16] <SteveA> of course, you can make it more succinct than that
[04:16] <morgs> SteveA, this would be in my ProductRdfView class, right
[04:17] <SteveA> morgs: um, whereever you are currently setting the content type stuff
[04:17] <SteveA> i'm at an airport -- not very convenient to look it up just now
[04:18] <morgs> Right, yes. ATM I'm setting it in __init__ of that class, so I'll put it in __call__ instead...
[04:18] <SteveA> ye
[04:18] <SteveA> yes
[04:18] <morgs> Great, I'll give this a shot.
[04:39] <morgs> SteveA, AttributeError: 'ProductRdfView' object has no attribute 'template'
[04:40] <morgs> SteveA, how do I set up self.template before calling it?
[04:40] <mpt> Is staging down?
[04:41] <SteveA> morgs: put your template in the class under the name 'template'
[04:42] <SteveA>  template = ViewPageTemplateFile('../templates/whatever.pt')
[04:44] <morgs> duh, right
[04:45] <morgs> SteveA, thanks, that fixes it.
[04:45] <SteveA> great
[04:48] <SteveA> jamesh: ping
[05:34] <elmo> ddaa/jblack?
[05:34] <ddaa> elmo: elmo?
[05:35] <elmo> ddaa: if we had to relocate arch.u.c to a new machine before lifeless came back from holiday,would you be in a position to help with that?
[05:35] <ddaa> Yes, I can handle that.
[05:35] <morgs> elmo: any ETA on wiki.launchpad being back?
[05:36] <elmo> morgs: not until someone gets me a tar ball of the wiki, nope
[05:36] <elmo> ddaa: ok, cool, thanks
[05:36] <ddaa> elmo: some help from stub might be needed though.
[05:37] <ddaa> elmo: note that I took some days off this week.
[05:37] <ddaa> though I will be there for emergencies.
[05:38] <elmo> ok - just out of interest, what happens to stuff if ssh connections were being refused from escudero?
[05:39] <elmo> [I'm not threatening to do that necessarily ;), just exploring migration options] 
[05:49] <bradb> morning
[05:52] <bradb> the pilot had to choose between making the airport curfew, or loading the last 20-30 bags. he choose the former.
[05:53] <Kinnison> oh dear
[06:26] <Kinnison> still no stub :-(
[06:32] <ddaa> elmo: the task fails, it will be retried the next day
[06:33] <ddaa> that's actually the main cause of transient failures at the moment, and I see no way out
[06:33] <ddaa> since we are somehow triggering a bug in ssl muxing...
[06:33] <elmo> ok
[06:44] <bradb> hm, very interesting that i can only recall my social insurance number in english, and my postal code only in french
[06:45] <Kinnison> bradb: s'an occupational hazard of being multilingual
[06:46] <Kinnison> bradb: for a long time I could only remember my two times table in french
[06:46] <bradb> heh
[06:46] <Kinnison> Which made for amusing mumblings under my breath as I worked out the right shifts and adds to do a multiply quickly in assembler
[06:47] <bradb> in other news, i'm very glad that the letter i received from Revenue Qubec saying "You owe us $huge_amount_im_sure_i_already_paid" was only a side-effect of the strike, and that, in fact, i don't really owe them any (more) money
[06:48] <Kinnison> heh
[06:49] <Kinnison> Libr nous les libros!
[06:49] <Kinnison> erm, Librez ?
[06:50] <bradb> librez-nous des libraux
[06:50] <Kinnison> I could have sworn you said 'les' not 'des'
[06:51] <Kinnison> must be your funny blocked-up-nose^W^W^Wqbeqois accent
[06:51] <bradb> hh
[06:51] <Kinnison> hh
[06:55] <bradb> Kinnison: you should try and get ahold of some songs from les cowboys fringants. i'm not sure what kind of stuff you like, but if you can do folkish, you might really like them.
[06:55] <Kinnison> Hey, it'll be fun to try
[06:58] <bradb> la grand-messe is their latest album. les toiles filantes, 8 secondes, plus rien, lettre  lvesque, camping ste-germaine, symphonie pour caza...get your torrent on!
[06:59] <Kinnison> Heh
[06:59] <Kinnison> mail me a list
[07:02] <bradb> of good songs, or of torrents? :)
[07:03] <Kinnison> sonds
[07:03] <Kinnison> songs even
[07:04] <Kinnison> if there's a strong torrent of an album I'll take that too
[07:06] <bradb> i bought the cd; not sure where to download them from. as soon as i'm done catching-up with Real Life this afternoon, i'll send you a list of good stuff.
[07:06] <Kinnison> thanks
[08:08] <cprov> anyone who needs to talk with me, please send email or call, bye 
[08:46] <JanC> is it possible that malone doesn't know all packages in breezy ?
[08:47] <mpt> JanC: It only knows about Universe packages at the moment
[08:48] <JanC> this is a universe package
[08:48] <JanC> and it doesn't know it (yet?)
[09:05] <mpt> that's odd
[09:06] <mpt> I don't know who would fix that, JanC
[09:06] <mpt> but my first guess would be morgs
[09:08] <JanC> the uploader of the package already knew about the bug & the fixed package will be available soon, so there is no hurry  :)
[09:09] <JanC> I found another bug in malone about a package not being available
[09:10] <mpt> ok, so assign your bug to the same person
[09:11] <JanC> it's not assigned  :)
[09:12] <mpt> aha
[09:12] <JanC> are packages automaticly added when they are added ?
[09:12] <JanC> added t omalone when they are added to the archive I mean
[09:12] <mpt> I don't know
[09:14] <salgado> JanC, not yet. soon they'll be. although with a small delay
[09:15] <JanC> ah, then I don't have to file a bug on "missing packages" ?
[09:17] <JanC> the package I couldn't find was xfce4-terminal (bug is already fixed now)
[09:17] <JanC> the other bug report about missing packages is: https://launchpad.net/malone/bugs/1413
[09:17] <salgado> JanC, I'm not 100% sure, but I think they're not yet automatically created in launchpad, so it's better to file the bug
[09:18] <JanC> a generic "missing packages" bug or for the specific package ?
[09:19] <salgado> a generic one, in the launchpad product
[09:20] <JanC> not the "malone" product ?
[09:20] <salgado> no, the launchpad one
[09:20] <JanC> okay, will do
[09:39] <dilys> Merge to rocketfuel@canonical.com/launchpad--devel--0: [trivial]  Add milestone icon, bug priority icons, and remote bug icons (patch-2240: mpt@canonical.com)
[10:17] <philiKON> hi
[10:17] <philiKON> forgive me if i'm blind, but is there a way to send messages to all members of my team?
[10:20] <salgado> philiKON, no, there's no way to do that. although, if launchpad has to send a notification for your team (e.g. a bug was filed in a package maintained by your team), you can have this notification to each of the team's members
[10:20] <philiKON> dang
[10:20] <philiKON> such a functionality would be very useful
[10:20] <salgado> but I guess what you really want is a way for you to send a messate to all members, right?
[10:20] <philiKON> right
[10:20] <philiKON> soemthing like, "finish up translations, release is coming"
[10:22] <ddaa> Sound like something that would fit in the bug tracker.
[10:22] <philiKON> hmm
[10:22] <ddaa> "There might be unfinished translations" milestone="next release"
[10:23] <ddaa> close it when people have acked they've finished the translations
[10:50] <nakeee> Any idea when the rosetta bidi bug is going to be fixed?
[11:01] <bradb> mpt: nice work on the bug icons! where can i see the milestone icon in action? it didn't seem to appear for either /distros/debian or /products/firefox
[11:02] <mpt> bradb: object-portlet-milestones.pt was already expecting it
[11:02] <mpt> If you see anywhere else it should go, let me know
[11:03] <mpt> Milestones were implemented in rather a hurry, so there doesn't appear to be sampledata for them, which made things a bit difficult
[11:03] <bradb> i believe there's sample data for firefox
[11:04] <bradb> mpt: i expected to see it here: http://localhost:8086/distros/debian and here: http://localhost:8086/products/firefox in the milestones portlet
[11:05] <mpt> oo, that's odd
[11:05] <bradb> mpt: also, might you be able to make an icon to be shown in place of an input checkbox, which will mean "this distrorelease is already targetted for the fix"?
[11:06] <mpt> ah, a kludge in launchpad.css
[11:13] <mpt> bradb: <input type="checkbox" disabled="true">
[11:14] <bradb> we planned to use an icon in favour of that
[11:14] <mpt> bradb: For me, http://localhost:8086/distros/debian gives "No milestones defined"
[11:14] <bradb> mpt: you have to add one (as foo bar) to see what it looks like
[11:14] <mpt> bradb: I was going to say "Did you know checkboxes can be disabled", but y'all were discussing more important issues at the time
[11:15] <bradb> yeah, we thought it might be slightly confusing to show a disabled checkbox, as people might wonder why they aren't allowed to uncheck it
[11:15] <bradb> what do you think? is it confusing?
[11:16] <mpt> a checkmark graphic is appropriate for something that wasn't turned on in the first place
[11:16] <mpt> but this thing here (a release target) was already turned on, so a disabled checkbox is entirely appropriate
[11:16] <bradb> ok, i'll go with that instead, thanks
[11:39] <bradb> Kinnison: how often does a package change name between releases? also, how often does it happen that something i used to find in package "foo" is now instead found in package "bar" in a new release?
[11:42] <elmo> bradb: often enough
[11:42] <elmo> to be something you really have to deal with sooner or later; but relative to the number of packages in the distro, it's not a huge percentage
[11:42] <elmo> IANAKIJPOOI
[11:42] <bradb> hm. this would make the bug release targeting/backport UI somewhat more complex.
[11:43] <mpt> bradb: The greatest thing about Malone is that it will mean I no longer have to put up with the text-shadow CSS on bugzilla.ubuntu.com
[11:47] <bradb> Kinnison: a more soyuz-specific question then: will it/is it fairly straightforward to figure out what Hoary package "foo" was called in Warty?
[11:50] <elmo> it should be
[11:50] <elmo> the DB has been designed around that from the beginning, but I haven't seen it's current evolution
[11:50] <elmo> (sorry, I'll stop playing kinnison now)
[11:52] <elmo> AFAICR, the DB has a higher level concept of a package than source or binaries in the archive
[11:52] <elmo> so you can always  refer back to that, at least internally
[11:52] <elmo> if that makes sense?
[11:55] <bradb> yeah, i know about the packaging table, but i'm not sure if it would be able to tell me that the binary program "foo" used to be installed with the binary package "blah-utils", but now it's part of the binary package "fnorb-scripts"
[11:56] <bradb> maybe the best backporting/release targeting UI is a simple dropdown of release names and a package name widget (to avoid all this complexity.) thoughts? mpt?
[11:57] <bradb> the package name widget could at least default to the package that the "current" task is targeting (i.e. you can only get to this targeting/backport screen from the bugtask screen anyway, so we can probably figure out a reasonable default package name to put in there.)
[12:09] <Kinnison> bradb: You can't know about specific binary files anyway
[12:11] <bradb> Kinnison: from the soyuz perspective then, is it at least easy to follow he same source package as it changes names between releases?
[12:11] <bradb> s/he/the/
[12:12] <Kinnison> We certainly looked into that
[12:12] <Kinnison> assuming we can find a sane way to cope with source package renames wrt. superseding I'm sure we'll be able to do something sane
[12:12] <Kinnison> The publishing history retains what superseded what
[12:13] <bradb> if i can fairly easily figure that out, the checkbox UI for backporting/targeting can probably make sense, as long as I give the user the option to manually add a task on a completely different package for the release target, if for some reason that's required.
[12:13] <bradb> i.e. you'd get checkboxes like:
[12:13] <bradb> Target Fix to:
[12:13] <bradb> [ ]  warty mozilla-firefox