[13:03] <facundobatista> Hola a todos
[13:07] <thisfred> morning facundobatista!
[13:07] <verterok> facundobatista: hola!
[13:07] <facundobatista> Hola thisfred, verterok
[13:13] <thisfred> http://vimeo.com/5313148
[13:13] <thisfred> very interesting (physical) buttons designs
[14:59] <CardinalFang> Hi all.  I have a desktopcouch branch that needs review.  https://code.edge.launchpad.net/~cmiller/desktopcouch/kill-asyncore-use-twisted/+merge/8597
[15:00] <jblount> MEETING BEGINS
[15:00] <jblount> Hello, welcome to the Ubuntu One desktop(+) developers meeting. If you are here for the meeting please say, "me". The format for this meeting is TODO/DONE/BLCKED:
[15:00] <dobey> CardinalFang: me too, but i think aquarius is on holiday :)
[15:00] <urbanape> me
[15:00] <jblount> me
[15:00] <jan____> CardinalFang: twsted FTW :_
[15:00] <urbanape> CardinalFang, dobey: I'll take a look
[15:00] <dobey> me
[15:00] <CardinalFang> me
[15:00] <teknico> me
[15:01] <jan____> meh :D
[15:01] <dobey> meh pretty well describe smy enthusiasm toward twisted :)
[15:01] <jblount> rodrigo_ ? statik ? vds ?
[15:01] <rodrigo_> me
[15:01] <vds> me
[15:02] <urbanape> DONE: Not a whole lot, getting reacquainted with the new files UI javascript code. Tried to tame it into some structure.
[15:02] <urbanape> TODO: Keep at it, maybe start by writing up a design doc for just what the heck is going on.
[15:02] <urbanape> BLOCK: Nada
[15:02] <urbanape> jblount: your turn
[15:02] <jblount> DONE: javascript hacking with urbanape, code reviews
[15:02] <jblount> TODO: Get branches for upgrade process and fixing the purchase process
[15:02] <jblount> BLCKED: Kind of waiting for urbanape (who is refactoring the js stuff for /files/new/ but not blocked
[15:02] <jblount> dobey: tag!
[15:02] <statik> me
[15:02] <dobey> DONE: Tarmac mini-sprint, Made HACKING files for client and protocol, moved u1fsfsm build from setup.py to make, fixed one last lint issue in protocol
[15:02] <dobey> TODO: Get rid of setup.py usage in ubuntuone-client, Look at gnome^H^Hxdg-keyring spec and such
[15:02] <dobey> BLCK: None.
[15:02] <dobey> CardinalFang: rock n roll
[15:03] <CardinalFang> DONE: pushed twistedification of desktopcouch pairing
[15:03] <CardinalFang> TODO: stop trying to type "desktopcouth". Help with 'paste' replacement or fixin
[15:03] <CardinalFang> g.
[15:03] <CardinalFang> BLOCKING: nil
[15:03] <CardinalFang> teknico: tag
[15:03] <teknico> DONE: finished configuring the old notebook for developing (the new one broke down :-/), landed the two web views testing branches
[15:03] <teknico> TODO: refactor the createCouchContacts.py script, add more contacts web tests
[15:03] <teknico> BLOCKED: nothing
[15:03] <teknico> NEXT: rodrigo_
[15:03] <rodrigo_> DONE: attended GCDS, lots of discussions about couchdb integration into the desktop
[15:04] <rodrigo_> TODO: lots :) like finish up the become-tomboy-friend branch (tomboy syncing), more evo-couchdb work
[15:04] <rodrigo_> BLOCKED: exhausted by last week
[15:04] <rodrigo_> vds: your turn
[15:04] <vds> DONE: Code review, code review, code review, started a branch with mark to add the funambol ds-server to the sourcedeps of our code base, started one more branch to add the configuration bits to run the ds server in a working env.
[15:04] <vds> TODO: I've suspended the previous branch I was working on (db contact snapshot) and I will resume it once I'm done with the funambol server
[15:04] <vds> BLOCKED: no
[15:04] <vds> I haven't seen any other "me"
[15:04] <statik> me
[15:05] <statik> DONE: tarmac sprint
[15:05] <statik> TODO: fix the wsgi servers with chad and pfibiger
[15:05] <statik> BLCK: none
[15:05] <jblount> Thanks everyone!
[15:05] <jblount> MEETING ENDS
[15:06] <statik> thisfred, teknico, vds: i'm desperate to understand what additional python libraries need to be released and packaged for karmic besides desktopcouch.
[15:06] <statik> for the contacts stuff
[15:07] <jan____> wow, that was a tight meeting
[15:08] <statik> jan____, nice to see you! the daily standup is status only, then we follow up on any issues raised separately so everyone else can get back to work
[15:08] <jan____> statik: hi :) I might wanna steal that concept
[15:09] <vds> thisfred teknico statik: what about a quick skype call? we should involve aquarius but I don't see him
[15:11] <statik> vds: aquarius is not working this week
[15:11] <thisfred> vds aquarius is on holiday
[15:11] <vds> ok
[15:11] <statik> and it's ok to say that in a public channel because he's already announced it on twitter :)
[15:11] <thisfred> the piranha pit *is* activated ;)
[15:12] <thisfred> anyhow: python libraries: contacts (perhaps renamed to desktopcouch-contacts) and that's it AFAIK.
[15:13] <thisfred> statik: vds ^^
[15:13]  * thisfred is thinking if he's missing anything
[15:13] <thisfred> statik:  couchdb-python itself obviously, but that's done I think?
[15:13] <thisfred> skype call is also fine btw
[15:14] <teknico> statik, none that I know of, off the top of my head
[15:14] <statik> thisfred, so lib/ubuntuone/contacts needs to be released as a separate project and packaged?
[15:14] <teknico> vds, I'm available for a call
[15:15] <vds> let's statik decide if a call is needed
[15:15] <statik> i can't do skype right now unfortunately :(
[15:15] <vds> decided :)
[15:16] <teknico> vds, oh, sorry, I hadn't got that it was about statik's question :-)
[15:16] <statik> so if it's just lib/ubuntuone/contacts, then i need someone to do a branch adding a proper setup.py to it, and then i can help with setting up the project and getting the license approved
[15:16] <thisfred> statik: yes, although it's pitfully empty yet. teknico is adding something right now I believe
[15:16] <statik> and i'd really like to get it published to the world this week
[15:17] <thisfred> statik: does it need to have the distutils.extra stuff, or is a plain distutils setup.py ok?
[15:18] <thisfred> statik: vds, teknico markgsaye_: I'd like for the initial release to also include the first version of the contact schema, in text. We can just take what Stuart wrote and dump it in a text file in there.
[15:19] <vds> thisfred: +1
[15:19] <statik> thisfred, plain distutils is ok
[15:20] <thisfred> statik: I can do those things today
[15:20] <statik> thisfred, awesome!
[15:20] <thisfred> statik vds teknico markgsaye_ what do we think of naming it desktopcouch-contacts?
[15:20] <mattgriffin> any tips on how to get the menu icon to stop spinning? it says it's still 'working'.
[15:21] <statik> thisfred, fine with me
[15:21] <teknico> thisfred, if it manages contacts-specific records in desktopcouch, it seems an appropriate name :-)
[15:21] <statik> python-desktopcouch-contacts will be the package name in debian then
[15:22] <thisfred> ok, will do that then
[15:25] <CardinalFang> mattgriffin:  "kill -TSTP ...", har har.  What does the  .cache/ubuntuone/log/syncdaemon.log  file say?
[15:30] <rmcbride> mattgriffin: the end of syncdaemon.log would help (as CardinalFang mentioned) We do seem to have occasional status issues with the applet
[15:30] <mattgriffin> cool. i'll check it out.
[15:30] <mattgriffin> thanks
[15:30] <dpm> hey rodrigo_, thanks for the review on the enable-translations branch!
[15:30]  * CardinalFang looks for a bug about status in the UI.
[15:31] <rodrigo_> dpm: you're welcome, and sorry for the delay, I was flying back from Gran Canaria
[15:31] <rmcbride> CardinalFang: they're definitely there. We should add the status thing to the "frequently reported bugs" master list for linking dupes to (someone mentioned doing this last week)
[15:31] <dpm> rodrigo_: np
[15:35] <CardinalFang> Ah yeah, bug 330769
[15:38] <CardinalFang> jblount, pfibiger: I'm about to ask shipit for a bunch of 9.04 CDs, unless you have some already.
[15:42] <jblount> CardinalFang: No la tengo, all I have left is Kubuntu CDs (which are still valuable and worthwhile)
[15:42] <dobey> i guess we need to create a "faq" about the keyring too
[15:43]  * jblount notices his bad spanish influnced by pfibiger 's taste in music.
[15:43] <CardinalFang> (says you!)
[15:44] <jblount> heh
[15:46] <jblount> jdobrien: Jacob Kaplan-Moss just posted this on his blog about authentication / session / other stuff in Django http://jacobian.org/writing/django-internals-authen/
[15:46] <jdobrien> jblount: uh? ok
[15:47] <dobey> grr
[15:48] <jblount> jdobrien: Sorry, I just thought it might be interesting considering how much you seem to think about account related stuff :)
[16:02] <thisfred> statik: dobey: the tests in desktopcouch depend on testtools, which is probably not a good idea. Should I rewrite them to use plain unittest.TestCase? I think they don't use much of testtools anyway.
[16:05] <statik> thisfred, testtools is packaged for karmic already
[16:05] <statik> i believe it is in universe already, let me check
[16:06] <statik> thisfred, yep python-testtools is in karmic universe so it's fine to build-depend on it (so the tests can use it). http://packages.ubuntu.com/search?keywords=python-testtools
[16:08] <dobey> statik, thisfred: we should get it backported to jaunty and hardy
[16:08] <statik> yeah
[16:09] <statik> i think there is even a backports team
[16:09] <dobey> there is
[16:12] <thisfred> statik: dobey oh ok, that's even better, I can leave them in contacts as well then :)
[16:15] <statik> dobey, can you review the MIRs for ubuntuone-client for kenvandine? there are a couple of things that need to be tweaked on the wiki pages and then I think the packages can go into main
[16:15] <kenvandine> dobey, i will send them to you in a sec
[16:33] <statik> thisfred, dobey: just in case you haven't seen it yet, here is how you would request backports of python-testtools for hardy and jaunty https://help.ubuntu.com/community/UbuntuBackports (of course it goes faster if we do the work, but this is the process to follow)
[16:34] <dobey> statik: yep. i've already requested backports for python-oauth :)
[16:34] <statik> great!
[16:34] <dobey> which reminds me that i need to test james_w's changes to it, on hardy
[16:37] <thisfred> statik: thanks
[17:14] <thisfred> mail sent to Benoît Chesnau, creator of couchdbkit to see if it makes sense to make that speak couchdesktop
[17:25] <jan____> thisfred: he's benoitc on freenode
[17:26] <thisfred> jan____: I know: see #couchdb a few minutes ago ;_
[17:26] <thisfred> statik: I'm a little stuck between stations: I've created the setup.py and renamed contacts, but now of course the imports don't work, and I'm not sure how to fix them, should I go the distance, and bzr mv desktopcouch-contacts into packages, and symlink from lib into its package directory?
[17:29] <jan____> thisfred: hehe
[17:48] <jdobrien> 2009-07-13 12:46:15,754:754.17804718 UbuntuOne.Client.Applet Failed to execute program /usr/bin/ubuntuone-syncdaemon: Success
[17:48] <jdobrien> I got that error ^^^ .... but isn't it supposed to be /usr/lib/ubuntuone-client/ubuntuone-syncdaemon
[17:50] <rmcbride> jdobrien: hmm
[17:50] <rmcbride> jdobrien: you don't have some kind of frankensteined mixed-bag install again, do you?
[17:50] <jdobrien> rmcbride: 'again'?
[17:51] <jdobrien> rmcbride: how can I tell?
[17:51] <rmcbride> jdobrien: last week, when you were having similar issues, and it had to do with having some things pointing to old paths
[17:51] <rmcbride> jdobrien: first thing would be "which ubuntuone-client-applet"
[17:52] <rmcbride> jdobrien: it might be you have an old client in the old path and its being exec'd in place of the installed client
[17:52] <jdobrien>  apt-cache policy ubuntuone-client-applet
[17:52] <jdobrien> W: Unable to locate package ubuntuone-client-applet
[17:52] <rmcbride> no
[17:52] <rmcbride> it's not a package. We want to know which one is execing . 'which ubuntuone-client-applet' at a terminal
[17:52] <rmcbride> jdobrien: is what we need
[17:53] <jdobrien> rmcbride: tell me what to type and I can give you an answer
[17:53] <rmcbride> jdobrien: what I typed in ' quotes is exactly what you need to type
[17:53] <jdobrien> which ubuntuone-client-applet
[17:53] <jdobrien> /usr/bin/ubuntuone-client-applet
[17:53] <rmcbride> jdobrien: that's what I needed. One sec.
[17:54] <rmcbride> jdobrien: that's the right path. Now do 'ls -l /usr/bin/ubuntuone-client-applet'
[17:54] <rmcbride> so we can look at the date on that file
[17:54] <jdobrien> ls -l /usr/bin/ubuntuone-client-applet
[17:54] <jdobrien> -rwxr-xr-x 1 root root 23K 2009-07-09 10:25 /usr/bin/ubuntuone-client-applet*
[17:55] <rmcbride> jdobrien: that's very strange
[17:55] <rmcbride> jdobrien: you're correct that it looks like it's going to the wrong path for the syncdaemon... but that does appear to be the right date stamp
[17:55] <jdobrien> rmcbride: this is why we keep getting people reporting that it is not starting
[17:56] <rmcbride> jdobrien: actually in those cases I think it's to do with the dep stuff.
[17:56] <rmcbride> jdobrien: I'm looking at a couple things on a test box to make sure we're headed down the right path. one sec (or so)
[17:58] <verterok> rmcbride, jdobrien: syndaemon is started via DBus, take a look to the dates in: /usr/share/dbus-1/services/com.ubuntuone.SyncDaemon.service
[17:58] <rmcbride> verterok: you're a step ahead of me. I was almost there :)
[17:58] <verterok> rmcbride: :)
[17:58] <jdobrien> -rw-r--r-- 1 root root 99 2009-07-09 10:23 /usr/share/dbus-1/services/com.ubuntuone.SyncDaemon.service
[17:59] <rmcbride> jdobrien: that's it. cat that file
[17:59] <jdobrien> [D-BUS Service]
[17:59] <jdobrien> Name=com.ubuntuone.SyncDaemon
[17:59] <jdobrien> Exec=/usr/lib/ubuntuone-client/ubuntuone-syncdaemon
[17:59] <rmcbride> OK that's just bizzare
[17:59] <rmcbride> it should not be trying to start it from /usr/bin with that entry there
[17:59] <jdobrien> rmcbride: it doesn't exist in usr/bin
[18:00] <rmcbride> jdobrien: I know
[18:00] <rmcbride> jdobrien: hence the error message you got
[18:00] <rmcbride> jdobrien: the question is WHY did client-applet think that's where it was? too strange.
[18:01] <thisfred> statik: solved,  I think. Running tests now, then I will propose and assign a specific review to you.
[18:02] <verterok> rmcbride, jdobrien: locate com.ubuntuone.SyncDaemon.service
[18:02] <jdobrien> verterok: how?
[18:02] <verterok> jdobrien: execute: 'locate com.ubuntuone.SyncDaemon.service'
[18:02] <rmcbride> jdobrien: type exactly what he put there
[18:02] <verterok> :)
[18:03] <rmcbride> jdobrien: and pastebin the result. SHould be a wall of text
[18:03] <jdobrien> rmcbride: it's 3 lines
[18:03] <rmcbride> jdobrien: OK it's a wall of text on my machine
[18:03] <rmcbride> jdobrien: probably because I have 16 tabs of bugs open
[18:03] <jdobrien> /home/john/wireitup2/configs/dbus-session/com.ubuntuone.SyncDaemon.service
[18:03] <jdobrien> /home/john/wireitup2/packages/ubuntuone-client/data/com.ubuntuone.SyncDaemon.service
[18:03] <jdobrien> /usr/local/share/dbus-1/services/com.ubuntuone.SyncDaemon.service
[18:03] <jdobrien> /usr/share/dbus-1/services/com.ubuntuone.SyncDaemon.service
[18:03] <rmcbride> ahah
[18:04] <rmcbride> jdobrien: cat the copy in /usr/local/share
[18:04] <rmcbride> jdobrien: I'm betting that's the problem
[18:04] <jdobrien> rmcbride:  yes
[18:04] <rmcbride> I have no idea how it would GET there, but that's almost certainly it
[18:04] <jdobrien> should that exist...and where did it come from
[18:04] <rmcbride> No it should not exist.
[18:05] <jdobrien> rmcbride: how do i get rid of it without breaking something else
[18:05] <rmcbride> jdobrien: it won't break anything else if you just remove  /usr/local/share/dbus-1/services/com.ubuntuone.SyncDaemon.service
[18:05] <rmcbride> Very very interesting. I wonder if that was the path we used at one point
[18:06] <rmcbride> and somehow with the renaming and such it didn't get cleaned up durign upgrade. None of my upgrade machines are having that issue...
[18:06] <jdobrien> rmcbride: hard to tell
[18:06] <rmcbride> but it could be something that was cleaned up by a build that didn't get installed on your machine for whatever reason
[18:07] <jdobrien> rmcbride: I wonder what I need to restart in order for that to change
[18:07] <rmcbride> jdobrien: and then subsiquent builds didn't work...
[18:07] <rmcbride> jdobrien: once you've removed that, it should work (if I understand dbus correctly)
[18:07] <rmcbride> jdobrien: otherwise a relog is probably necessary
[18:07] <rmcbride> or at least a restart of the client
[18:08] <jdobrien> ha! I had a renamed folder come back!
[18:08] <rmcbride> wow I really need to clean up my local branches, That's why I saw so many entries.
[18:08] <rmcbride> jdobrien: heh
[18:09] <rmcbride> well now we know to check /usr/local/share in some of these "won't start" cases, should the current packages not resolve
[18:10] <jdobrien> We should create some kind of clean up script or something
[18:11] <jdobrien> rmcbride: that seems to have corrected my problem
[18:11] <rmcbride> jdobrien: it should be handled when the package is upgrade if it moves. In all cases on my test machines it has been. Why it happened is a puzzle
[18:12] <rmcbride> jdobrien: cool. Once we saw teh copy in /usr/local it was clear what was happening.
[18:19] <jdobrien> Could I get some people to test this bug #375011? I can't reproduce this
[18:22] <dobey> we have an e-mail newsletter?
[18:22] <jdobrien> dobey: not yet, but I was told to add that to the site
[18:22] <dobey> it works fine here
[18:24] <rmcbride> meh, I had join/part turned off and missed that he left.
[18:24] <rmcbride> dobey: it worked fine here, or so I thought. Once I moused over the checked box, it cleared teh checkbox. Then selecting a second time seemed to work
[18:25] <rmcbride> can't make it do it again though
[18:26] <dobey> i made it turn red
[18:27] <dobey> but it still enables/disables properly
[18:27] <dobey> at least in the ui
[18:27] <dobey> the rapid succession of clicking might have broken something else
[18:28] <statik> re
[18:28] <rmcbride> dobey: I can reproduce it by deselecting it, clearing my cache, and going back to the page again. Same sequence of events. Adding that to the bug report and moving on :)
[18:32] <thisfred> restatik: prepackaging branch proposed, and review specifically requested of you
[18:40] <statik> thisfred, ok looking now
[18:41] <thisfred> statik: no hurry, *unless* I did something wrong and have to fix it today ;)
[18:42] <statik> https://launchpad.net/people/+me/+activereviews is handy when dealing with reviews across a bunch of projects
[18:42] <statik> thisfred, we normally don't put any code under bzr control in sourcecode
[18:42] <thisfred> statik: oh nice. That doesn't show the reviews you *could* do because of groups you're in though, right?
[18:42] <statik> that directory is for external branches
[18:43] <statik> thisfred: yeah, +activereviews needs to also show reviews I could do
[18:43] <thisfred> statik: yeah I realize, it was a temporary hack to get the fixed imports working ahead of packaging...
[18:43] <thisfred> statik: if it's too dirty,  I'll undo it
[18:44] <statik> thisfred: i think it will bite us when people try to update to the next version. could you just move the stuff from sourcecode to packages/, then once it's published separately we can move the symlink to point into sourcecode/ at the external branch?
[18:45] <statik> thisfred: also, do you think this needs to be a totally separate project, or should we just stick it into the desktopcouch project as a separate module somehow?
[18:46] <statik> so python-desktopcouch-contacts would still be a separate package, and still have a separate setup.py...
[18:46] <thisfred> statik: sure, and yes, that sounds like a good suggestion
[18:47] <statik> i guess the module namespace would be desktopcouch.contacts ?
[18:47] <thisfred> statik: yeah, that would be an improvement even: I though plain 'contacts' was a little presumtuous
[18:47] <thisfred> +p
[18:48] <statik> cool. i'll write all this on the code review. thanks for working on it!
[18:48] <thisfred> statik: ok, I'll start with moving it to packages
[18:49] <thisfred> statik: done
[18:50] <statik> thisfred, if you feel like adding the LGPLv3 headers (you can copy from desktopcouch) that would make things even easier at the next step :)
[18:51] <thisfred> statik: sure
[18:51] <thisfred> statik: that's just those files with all caps names, right?
[18:51] <statik> thisfred: no, it's just looking at the comments at the top of any of the python source files in desktopcouch and putting the same header/copyright on the stuff in desktopcouch.contacts
[18:52] <thisfred> statik: ah ok
[18:55] <thisfred> statik: also pushed
[18:56] <thisfred> statik: for the moving into desktop couch is there anything I can do?
[18:56] <statik> thisfred, awesome! can you resubmit the merge proposal to get a new diff generated?
[18:56] <thisfred> statik: will do
[18:57] <thisfred> statik: 'request another review' will do that, right?
[18:58] <statik> thisfred: i think it's changing the top-level merge proposal status to 'resubmit'
[18:58] <statik> thisfred: don't let me talk you into working too late tonight. i think the next steps for getting this into desktopcouch is simply to propose a branch adding the contacts dir and code to the desktopcouch project
[18:59] <statik> thisfred: since desktopcouch is already a sourcedep...the only other thing to do would be change the imports in ubunet as you have already
[19:00] <thisfred> statik: nah, my wife's away, so I have time ;) (although if you have no further need of me, I'll switch to desktopcouchification of my autoqueue plugin)
[19:00] <thisfred> statik: resubmitted
[19:14] <statik> thisfred, contacts-prepackaging is approved if you want to land it
[19:14] <thisfred> statik: cool, pqm submitting now!
[19:31] <thisfred> statik: and re-submitting: trunk was conflicting
[19:35] <statik> thisfred: you don't need a re-review just for fixing conflicts unless you feel like you want a review to be sure you solved the conflicts the right way
[19:36] <thisfred> statik: ah, no, I meant re-pqm submitted ;)
[19:36] <thisfred> I have confidence in me :)
[19:37] <statik> cool
[19:45] <BUGabundo> bom fim de tarde :)
[19:50] <rmcbride> hiya BUGabundo
[19:50] <BUGabundo> hey rmcbride
[20:12] <BUGabundo> v
[20:36] <thisfred> I have *too much* confidence in me. The tests in contacts itself still imported from the wrong place... :( re-re-pqm-submitted
[20:40] <dobey> heh
[21:16] <sheepeatingtaz> Good evening all
[21:17] <sheepeatingtaz> is there a way of seeing how a sync is prgressing?
[21:17] <sheepeatingtaz> I can see the applet-icon spinning, but would like to know how it is getting on :)
[21:18] <jblount> sheepeatingtaz:  u1sdtool --current-transfers
[21:19] <jblount> sheepeatingtaz: If you throw that into a terminal, it'll give you a bit of data on what's transferring :)
[21:19] <sheepeatingtaz> jblount: that'll do for now, cheers!
[21:23] <thisfred> statik: it finally landed