[08:39] <JamesTait> Morning all! :)
[09:37] <mandel> morning all!
[11:14] <gatox> good morning!
[11:17] <gatox> mandel, ping
[11:17] <mandel> gatox, pong!!!!
[11:17] <mandel> po po po pong
[11:18] <gatox> mandel, hi..... are you turning japanese?
[11:18] <gatox> jej
[11:18] <mandel> a little :)
[11:18] <gatox> now a real questoin
[11:18] <gatox> we were talking with alecu yesterday
[11:19] <mandel> yes
[11:19] <gatox> and we want to use a structure to send data via ipc, like a Struct of a Dict of {String, variant}....... for dbus..... but we need to know if this is valid in windows ipc also.... and you were the guy to ask
[11:21] <mandel> gatox, so the deal is the following, as it is right now you can send with no problem nearly any python object as long as we extend pb.Copyable
[11:22] <mandel> gatox, http://twistedmatrix.com/documents/current/core/howto/pb-copyable.html#auto2
[11:22] <gatox> mandel, ok, sound good
[11:22] <mandel> gatox, now the but..
[11:22] <gatox> mandel, bloody hell!
[11:23] <mandel> gatox, but I'd like to stop using pb at some point for something we can use from other langs.. so keep the objects simple :)
[11:24] <gatox> mandel, the object is "simple", but it's a dict with String keys, and one of the values will be: (sii), and the other one (s)
[11:24] <mandel> no problem then :)
[11:25] <mandel> gatox, ^
[11:27] <gatox> need to restart..... partial upgrade
[11:30] <gatox> back
[12:09] <mandel> gatox, I'm off to have lunch :)
[12:09] <gatox> mandel, please don't forget to review my branch
[12:09] <gatox> i really need you to do that
[12:10] <mandel> gatox, yes!
[12:28] <alecu> mandel: the thing we discussed with gatox was regarding "nested structures" being sent thru ipc
[12:28] <alecu> mandel: so, right now what is being sent is thru pb, and it's using the serialization mechanism for pb, right?
[12:29] <alecu> (good mornin all, btw!)
[12:30] <gatox> alecu, hi
[12:39] <gatox> alecu, i'm having some problems with variant :S
[12:39] <alecu> gatox: tell me
[12:41] <gatox> alecu, one of the variant for example is a list, when the list is empty, d-feet shows: ValueError: Unable to guess signature from an empty list
[12:42] <gatox> do i need to create the variant in some special way?
[12:42] <gatox> i'm just assigning the python list
[12:45] <gatox> ah..... i could do that, when the list is empty, don't send that key :P
[12:45] <gatox> that work
[12:46] <alecu> uh!
[12:46] <gatox> alecu, you don't like it?
[12:46] <alecu> I've just found out that the family we have in bariloche has just arrived for a surprise visit.
[12:46] <alecu> gatox: ^
[12:46] <alecu> gatox: that means that I'll probably be taking a few vacation days next week.
[12:46] <gatox> alecu, ohhhh!! well, you were going to take half of the day off today! so enjoy :P
[12:47] <alecu> gatox: how did you end up doing the variant stuff?
[12:48] <gatox> alecu, if the value that is going to be a variant, is empty..... as an empty list (we are using only list at this moment), don't add that key to the dict we are returning..... we just need to explore the dict with "get" (i think we are always doing this), so not problem there (from my point of view)
[12:54] <ralsina> gatox: there is probably a dbus.List(signature) or something, we had that for dicts
[12:54]  * ralsina goes away again
[12:54] <gatox> ralsina, :P
[12:56] <gatox> brb
[12:56] <alecu> gatox: here's how I would do it:
[12:57] <alecu> http://pastebin.ubuntu.com/1099994/
[12:58] <alecu> gatox: being explicit with the signature of composite dbus types (dict, array, struct) is the best way to avoid nasty surprises.
[12:58] <alecu> gatox: well, in fact, it's nice to use the same for basic types too, because we are defining an api.
[12:59] <alecu> gatox: so, let me expand on this example to include other values...
[13:01] <dobey> hmm
[13:03] <gatox> back
[13:08] <alecu> gatox: take 2: http://pastebin.ubuntu.com/1100006/
[13:08]  * gatox looking..
[13:08] <gatox> i see
[13:34] <dobey> doh
[13:52] <dobey> anyone can review https://code.launchpad.net/~dobey/ubuntuone-control-panel/update-4-0/+merge/115607 ?
[13:55] <dobey> alecu: that example looks like it is duplicating API we already have :-/
[13:56] <dobey> alecu, gatox: also, why are we defining new API here? if this is for integrating in unity, we don't need to define any new APIs afaik
[13:57] <gatox> dobey, it's not only unity
[13:58] <dobey> gatox: right, but maybe we should move the tray icon stuff for windows, into syncdaemon
[14:00] <dobey> or if we need to do stuff outside syncdaemon, we should peruse the APIs we already have, and not define a new one which just lumps all the data in a single signal
[14:01] <mandel> alecu, yes, that is correct, changing it is not a problem
[14:02] <gatox> dobey, we are doing it inside syncdaemon, and we are not using signal, as alecu explain yesterday, we are defying a new method that we are going to check from the menu, only when it's needed
[14:04] <dobey> gatox: still, i think it should result in all the current signals we have, which would be relevant, getting called, rather than defining a new way to get the same data. people will abuse this to get the data they want anyway
[14:05]  * briancurtin coffee run, back in a few mins
[14:14] <progre55> hi guys. I've got ubuntu-one installed on my computer(ubuntu) and got ubuntu-one-mobile on my phone (galaxy s3). I'm sharing a couple of files from my phone, but the computer ubuntu-one doesnt even see the device. When I open one.ubuntu.com from my comp, I only see the computer under "devices", but from the mobile browser, I can see all my devices. Yes, I have cleared the browser cache. Any suggestions, please?
[14:14] <alecu> dobey: moving tray icon stuff inside syncdaemon on windows means using the qtreactor for it, and I don't like the qtreactor that much.
[14:15] <alecu> progre55: what does the web show?
[14:15] <progre55> alecu, it doesnt show my mobile under devices when opened on the computer browser (firefox), but does show it when opened on a mobile browser (firefox)
[14:16] <alecu> progre55: it seems like a bug we've already encountered... let me check.
[14:17] <progre55> alecu, the ubuntu client doesnt have the mobile under "devices" either
[14:19] <alecu> progre55: it looks like two user accounts were created for you...
[14:19] <alecu> progre55: can you check if the email in both accounts is the same?
[14:20] <progre55> hmm.. let me see
[14:20] <progre55> oh, apparently different emails
[14:20] <progre55> how can I join them?
[14:25] <mandel> gatox, review has been done.. by the way, there is some weird problem in the mp, some code we removed in clean-fsevents is back :(
[14:26] <progre55> well I guess I'll just pick one and use it :)
[14:26] <dobey> progre55: file a support request about it at https://one.ubuntu.com/help/contact/
[14:26] <progre55> alecu, nah no problem, I'll just use one, and keep the other one in case I exceed my 5Gb limit :)
[14:27] <alecu> progre55: I'm not sure if support can join both accounts; it might be easier to just pick one, right.
[14:28] <alecu> progre55: I would suggest keeping the one already syncd with your pc
[14:28] <progre55> yes, I also thought so, as I've got my backup data in that one..
[14:28] <progre55> alecu, thanks a lot, appreciate
[14:30] <mandel> dobey, any idea about what happened here: https://code.launchpad.net/~mandel/ubuntuone-client/fsevents-daemon/+merge/114836 ?
[14:31] <dobey> mandel: hrmm
[14:33] <dobey> mandel: looks like maybe a package was missing or something
[14:35] <mandel> weird..
[14:45] <mandel> weekly call is happening, right?
[14:45] <dobey> is it?
[14:45] <briancurtin> aren't we short a bunch of people?
[14:45] <gatox> mandel, branch updated: https://code.launchpad.net/~diegosarmentero/ubuntuone-client/darwin-tests-refactoring/+merge/113289
[14:45] <mandel> gatox, superb!
[14:46] <mandel> briancurtin, dobey, I was not told otherwise
[14:46] <dobey> briancurtin: just ralsina no?
[14:46] <gatox> mandel, weekly call?? ralsiina is on vacation
[14:46] <briancurtin> i thought mmcc and thisfred as well
[14:46] <dobey> oh
[14:46] <dobey> thisfred is here at least
[14:47] <thisfred> are we not skipping it|
[14:47] <briancurtin> ah. mmcc is on the way back to san diego the next few days or something
[14:47] <mandel> so we have all but ralsina and mmcc, is that right?
[14:47] <thisfred> he only said not to party, not specifically to not slack off ;)
[14:47] <mandel> so I guess we can vote to skip it :)
[14:47] <dobey> heh
[14:49]  * thisfred votes skip
[14:49]  * mandel votes skip
[14:49]  * dobey skips voting to skip
[14:49]  * briancurtin skips
[14:49]  * gatox votes skip
[14:51] <mandel> dobey, got the same exact error, can you take a look
[14:51] <mandel> dobey, also, it seems that the config test is giving an error
[14:51] <mandel> and complains about setlocale ?
[14:51] <mandel> wtf
[14:51] <dobey> mandel: yes, because the locale isn't getting set correctly
[14:51] <dobey> mandel: because it fails if the locale encoding isn't in UTF-8
[14:51] <mandel> dobey, what changed in tarmac to be getting this things?
[14:52] <dobey> mandel: setting the locale to C.UTF-8 (now en_US.UTF-8 because lucid/natty don't have C.UTF-8 avaialble at all), so that we can have the encoding set to UTF-8
[14:54] <mandel> dobey, but there, right now it seems that no branch will land, is that correct?
[14:54] <dobey> mandel: yes. i'll look into it
[14:55] <mandel> dobey, thx!
[14:56] <ralsina> Don't slack off!
[14:56]  * ralsina goes back into hibernation
[14:56] <ralsina> just kidding, it's ok
[14:56] <thisfred> :)
[14:57] <ralsina> Also, don't sell the cow formagic beans.
[15:04] <mandel> ralsina, I got a good deal for gatox :)
[15:04] <mandel> 2 bananas and a pony
[15:04] <gatox> mandel, shhhhh
[15:08] <mandel> gatox, I think we should remove that in_modify thing and import the IN_MODIFY from the right spot, which module is importing it?
[15:08] <mandel> gatox, it was remove in trunk AFAIK and everything is ok
[15:08] <gatox> mandel, read the comment... there it explains why i don't tihnnk that is the best solution, and alecu agree with me when i explain tihs
[15:11] <mandel> gatox, but there are no IN_ consumed from that module afaik, not in ubuntuone/platform/filesystem_notifications/windows or darwin
[15:12] <mandel> gatox, right now they all use the pyinotify agnostic module, so I don't understand the comment
[15:12] <gatox> mmmmmm in that case yes it's not really necessary
[15:13] <gatox> mandel, changed pushed
[15:13] <mandel> gatox, super :)
[15:20] <dobey> mandel: ah  i see what happened
[15:22] <mandel> dobey, fixable?
[15:27] <dobey> mandel: yes
[15:28] <mandel> dobey, lovely
[15:38] <dobey> ok, need to get lunch. bbiab
[15:48] <mandel> and I eod, see you all tom
[15:48] <mandel> gatox, branch has my +1
[15:48] <mandel> dobey, can you approve the failing branch whenever tarmac is fixed
[15:57]  * gatox lunch!
[16:00] <briancurtin> heading to the doctor + lunch combo, back in a bit
[16:24] <alecu> dobey: while trying to run trunk on mac, I get this error regarding dirspec: http://pastebin.ubuntu.com/1100349/
[16:24] <alecu> dobey: any ideas?
[16:28] <alecu> and gatox_lunch: the tests won't run on windows nor mac on https://code.launchpad.net/~diegosarmentero/ubuntuone-client/darwin-tests-refactoring/+merge/113289
[16:28] <alecu> ok, early eod for me. See you guys tomorrow.
[16:40] <gatox_lunch> alecu, fixed..... it was a typo
[16:49] <dobey> alecu: you need newer dirspec
[17:37]  * briancurtin back
[18:50] <equinto> anyone using javascript to connect ubuntu one?
[19:39] <dobey> alright, gotta run for a few. brb
[20:02] <gatox> eod here! see you tomorrow people! :D
[20:51] <dobey> ugh, it's about to storm here :(
[20:51] <briancurtin> we had it last night. something like 40k without power
[20:55] <dobey> guess my new truck will get dirty :-/
[20:56] <briancurtin> what'd you get?
[20:56] <dobey> 09 Chevy Avalanche
[20:57] <briancurtin> nice. i kind of want a tahoe in that same body style whenever i'm done with this trailblazer
[20:59] <dobey> yeah, it's the LTZ, so fully loaded with folding running boards and mirrors, and bose dvd+nav, and all kinds of buttons to play with
[22:48] <dobey> later all, gotta run
[22:52] <equinto> anyone issues with https://one.ubuntu.com/oauth/sso-finished-so-get-tokens/ ?