[08:34] <JamesTait> Good morning all!
[08:41] <j0nr> folks, morning
[08:42] <j0nr> I asked in here a week or so ago about how to backup and restore my synced U1 folders...........I forgot
[08:43] <j0nr> I have a backup of my ~/Music folder..... I have a fresh install of U1.... should I put the ~/Music folder onto my fresh install before or after I sign into U1?
[08:43] <mandel> j0nr, you have to do it before, the idea is that you have to amke sure that the metadata is not there before your files, but let me confirm this beofre you do anything
[08:44] <mandel> karni, is that ^ right?
[08:44] <karni> Good morning everyone!
[08:44]  * karni reads
[08:44] <mandel> karni, I remember we once said the wrong thing.. so I want to make sure :P
[08:45] <karni> mandel: j0nr: I believe the user should not copy the metadata over. Just the files. Syncdaemon will first do local rescan, and only then start down/uploading stuff.
[08:45] <karni> mandel: Yeah, I think I told the user to copy metadata as well, which was a mistake
[08:46] <j0nr> So restroe my files as they were i.e. copy my ~/Music onto new system. Then sign into U1
[08:46] <j0nr> just trying to avoid the need to download all my Music
[08:47] <j0nr> sidenote... are you guys in charge of the Android Music App?
[08:47] <karni> j0nr: I believe that is correct. Yes, understood. It' be best to ask our support guys, duanedesign or rye. The former is probably still sleeping.
[08:47] <karni> j0nr: I'm the developer on it.
[08:47] <karni> j0nr: I believe you have something to say :)
[08:48] <rye> karni, you are right, don't copy ~/.local/share/syncdaemon, copy Music only then launch syncdaemon
[08:48] <j0nr> Just wondering if it is still alive and kicking... I put it on my phone ages ago (paid) and haven't seen any updates since...
[08:49] <karni> j0nr: We just made a few improvements and fixed a playlist related server bug. Should be released soon to the Market.
[08:49] <j0nr> karni: Cool! i was going to say, that one really 'bugs' me...
[08:50] <j0nr> the playlists that get saved, are they specific to U1 Android music client or can they be accessed anywhere esle, or by any other programs? e.g. other music apps on my device? Can I access them on my desktop pc?
[08:50] <karni> j0nr: We had some unrelated issues, and this bug hid behind them. This is the only reason it was there for so long, sorry for the inconvenience :<
[08:51] <karni> j0nr: To be honest, I'm not sure.. We save them in couch, so theoretically they should be available. rye, do you know anything about that ↑ ?
[08:52] <karni> j0nr: (I'm gradually taking over the U1 Music app, sorry that I can't answer that particular question yet)
[08:53] <j0nr> Are all playlists the same, e.g. M3U or is it specific to all different applications?
[08:53] <j0nr> if you are storing as M3U then any app should be able to make use of them right? or is there problems with each app knowing where to find the list of music?
[08:53] <j0nr> Sorry I don't know much, just thinking aloud
[08:55] <karni> j0nr: Playlists are stored in CouchDB. It's a database on your computer, that syncs with U1. It's not M3U files.
[08:59] <j0nr> Ok, so I guess that rules out using those playlists in other music apps on my phone
[08:59] <j0nr> fair enough.
[09:00] <karni> j0nr: unless they talk to our Couch db on U1 servers (which is possible for 3rd party developers)
[09:00] <j0nr> ok
[09:54]  * mandel coffee
[09:57] <matti> +`
[09:58] <matti> Ops
[09:58] <matti> +1
[09:58] <matti> Coffee!
[10:12]  * mandel back
[10:12] <mandel> matti,  lol
[10:15] <matti> :)
[10:21] <gatox> hi
[10:22] <matti> Hi
[10:41] <mandel> gatox, buenos dias!
[10:41] <mandel> gatox, truns out it was not deferToThread, but the bloody ubuntuone.platform.tools
[10:42] <gatox> mandel, jejje lot of debugging?
[10:46] <mandel> gatox, more like, lots of swearing :)
[10:46] <gatox> jeje
[11:33]  * mandel walk dog
[11:36] <gatox> mandel, did you already propose the branch that fix the problem in the installer?? because i propose another branch yesterday that  requires the branch you submit yesterday... to know, because if the fix change i have to change my branch
[12:17] <gatox> brb
[12:53] <ralsina> good morning!
[12:53] <ralsina> alecu, nessita, dobey, gatox, mandel: standup in 7'
[12:53] <gatox> ralsina, hi!
[12:54] <ralsina> hola nessita
[12:54] <nessita> hello everyone!
[12:55] <gatox> nessita, hi
[13:00] <ralsina> me
[13:00] <mandel> me
[13:00] <dobey> meh
[13:01] <ralsina> nessita, alecu, gatox: say me!
[13:01] <gatox> me
[13:02] <nessita> me
[13:02] <mandel> @ping
[13:02] <ubot4> pong
[13:02] <nessita> mandel: go
[13:02] <mandel> we do have a stand up, right?
[13:02] <ralsina> mandel: yes
[13:02] <nessita> yeah, as far as I know
[13:02] <gatox> mandel, yes
[13:02] <nessita> mandel: go!
[13:03] <gatox> ralsina, was first
[13:03] <ralsina> DONE: team call, 1-1 with alecu, started resourcing, moved canonicaladmin queue, doctor's appointment. TODO: finish resourcing, work on hiring, nessita's 1-1, 1-1 with the boss, and so on and so forth. BLOCKED: I am feeling rather sick to my stomach, but hey, I feel just as bad in bed
[13:03] <ralsina> mandel, now
[13:03] <nessita> ralsina: sorry, I did not notice you said me first
[13:04] <ralsina> nessita: no problem!
[13:04] <ralsina> ok, mandel, if you pasted it, it did not work :-)
[13:05] <ralsina> dobey, go
[13:05] <dobey> λ DONE: planning call, discussed release/qa procedure with elopio, collected some notes for process e-mail
[13:05] <dobey> λ TODO: send out process e-mail, banshee/libu1 SRUs, bug #467397?
[13:05] <ubot4> Launchpad bug 467397 in python2.6 (Ubuntu) (and 6 other projects) "KeyError: 'ROUND_CEiLiNG' when using turkish locale (affects: 16) (dups: 9) (heat: 108)" [Undecided,Confirmed] https://launchpad.net/bugs/467397
[13:05] <dobey> λ BLCK: No.
[13:05] <dobey> gatox: go
[13:05] <gatox> DONE:
[13:05] <gatox> Fixed some branches, Fixed Congratulations page, working on expanduser.
[13:05] <gatox> TODO:
[13:05] <gatox> Finish expanduser, search & destroy ui bugs.
[13:05] <gatox> BLOCKED:
[13:05] <gatox> No
[13:05] <gatox> nessita, go
[13:06] <nessita> DONE: roadmap call, reviews, discovered a couple of IPC bugs in u1client, more on the epic branch for u1sdtool multiplatfofm
[13:06] <nessita> TODO: 1-1 with ralsina, reviews, stable branch for installer, more epic branch
[13:06] <nessita> BLOCKED: no
[13:06] <nessita> NEXT: alecu
[13:06] <nessita> NOTE: I will be attending the design sprint next week, and UDS the nweek after that, so email will be the best way of contact me if I'm not in IRC
[13:06] <ralsina> Maybe there was a confusion about standups in yesterday's call. We'll keep on having standups except the day of the team call. I am estimating you will not waste more than 2 hours a week on administrivia. If you think that's too much, I can give you *my* list ;-)
[13:07] <mobile_> me
[13:07] <ralsina> Also, in the email I sent I mentioned weekly email reports. Those will not be needed, I will do them based on the 1-1s
[13:07] <ralsina> mobile_: ?
[13:07] <dobey> es mandel
[13:08] <mandel1> yes, stupid irc client
[13:08] <ralsina> ok
[13:08] <dobey> rima-tde.net == spain :)
[13:08] <ralsina> aleeeeeeccccccuuuuu alecu!
[13:08] <mandel1> DONE: worked on fixing bug https://bugs.launchpad.net/ubuntuone-windows-installer/+bug/878259 added extra tests for the moved code. TODO: propose merges for autoupdate after running tests on N. debug issue on vista. BLOCKED: no
[13:08] <ubot4> Launchpad bug 878259 in ubuntuone-windows-installer "Fails to start the installer (affects: 1) (heat: 6)" [High,Confirmed]
[13:09] <mandel1> I don't know why I have issues with the pad irc"..
[13:09] <ralsina> mandel1: because it's evil and proprietary
[13:10] <gatox> mandel, because is mac
[13:10] <gatox> :P
[13:10] <gatox> jeje
[13:10] <gatox> not mac...... but..... you know jeje
[13:10] <mandel1> probably
[13:11] <ralsina> ok, alecu seems to be missing in action, so EOM
[13:11] <mandel1> like our servers, hehehe
[13:11] <ralsina> and if I may say so, the worst standup ever. I think most of you were not even standing.
[13:11] <gatox> jeje
[13:12] <mandel1> ralsina, I was no :P
[13:12] <ralsina> our servers have different freedom capabilities, you insensitive clod!
[13:14] <nessita> ralsina: wasn't alecu taking the morning off?
[13:14] <mandel1> I go back to lunch, I'll be back I'm 10/15 min
[13:15] <ralsina> nessita: right!
[13:15]  * ralsina wishes canonicaladmin would send daily staff rolls
[13:15] <ralsina> of course that would be like, useful, so it's not there
[13:15] <gatox> lol
[13:16] <gatox> nessita, i have a couple of pending reviews
[13:16] <ralsina> gatox: I can do one, probably, if you need it
[13:16] <nessita> gatox: yeah, I'm already with network-detect
[13:16] <nessita> gatox: then I have the 2 you gave me yesterday
[13:16] <gatox> nessita, i fix network-detect..... but i was getting a error with a test for oauth..... not related to my branch :S
[13:17] <gatox> ralsina, yep!!1 i need reviewers..... let me give you one :P
[13:17] <nessita> gatox: need some help?
[13:18] <gatox> ralsina, here is a really small one: https://code.launchpad.net/~diegosarmentero/ubuntuone-windows-installer/connect-files/+merge/79611
[13:19] <gatox> nessita, oops..... i commit the test_after_timeout_cache_expires commented.... let me revert that...... i don't know why that test was failing.... i was looking at that
[13:19] <nessita> gatox: ok, let me know
[13:20] <gatox> nessita, if you are free..... and you are not going to review network-detect right now..... can you please review the one for congratulations page?? is really small (sso and installer)
[13:20] <ralsina> gatox, nessita: shouldn't we yield on connect_files so we actually know if it connects correctly?
[13:22] <nessita> ralsina: I think there is no need, there's much we can do at that stage... the controlpanel will try to connect again if not connected
[13:22] <nessita> ralsina: if we'd like to handle not connecting at that point we would need a design for that... may be to much
[13:23] <ralsina> nessita: ok
[13:23] <nessita> ralsina: nevertheless, I can talk this with liseeeete next week
[13:23] <nessita> gatox: connect-files is giving conflict
[13:23] <nessita> Text conflict in ubuntuone_installer/gui/qt/gui.py
[13:23] <nessita> 1 conflicts encountered.
[13:24] <gatox> fixing.....
[13:24] <ralsina> nessita: I think it may be a good idea, because it will prevent weird errors like "None has no get_rootdir attribute" (wouldn't it?)
[13:25] <nessita> ralsina: as far as I see it, it will not prevent those, they will just appear in the installer
[13:25] <ralsina> or is conenct_files to make syncdaemon connect, instead of connecting to syncdaemon>
[13:25] <nessita> ralsina: right, the former
[13:25] <ralsina> oh
[13:25] <ralsina> then no point
[13:25] <nessita> connect == connect a running syncademon to the server
[13:27] <nessita> ralsina: I just read carefully the email "The plan". Seems like it was chopped off at the end, and Chipaca played along :-)
[13:28] <ralsina> Yes, it seems that way :-)
[13:28] <ralsina> I will post a fixed version with added bits
[13:28] <gatox> nessita, ralsina conflict resolved in connect-files
[13:28] <nessita> gatox: ack
[13:29] <ralsina> ack
[13:31] <nessita> gatox: the successfull page sso branch will need fixing.... sso is a project independent from Ubuntu One, so we can't hardcode the name
[13:31] <nessita> gatox: please note that sso is used from software center also, and ideally any other app that wants sso credentials
[13:32] <nessita> gatox: so, the SUCCESS message, should not change, or at least change to:
[13:32] <gatox> nessita, ahhhhh ok
[13:32] <gatox> i'll add a new one
[13:32] <nessita> SUCCESS = _('You are now logged in into %{app_name}s. Congratulations!')
[13:33] <nessita> gatox: you can't hardcode Ubuntu One, not sure if you got that message :-)
[13:33] <gatox> nessita, ahhhhh ok..... sorry..... i took the text as it was in the bug description
[13:33] <nessita> gatox: right, but we need to adapt it a bit according to the project's specifics :-)
[13:34] <nessita> gatox: so, always remember: sso is generic, knows nothing about ubuntu one itself
[13:34] <gatox> nessita, roger that! :P
[13:37] <nessita> ralsina: shall we 1-1?
[13:37] <ralsina> nessita: in 1'
[13:39] <ralsina> nessita: ok, whenever you want, I am in mumble now
[13:39] <nessita> ralsina: I'm trying to connect but is not letting me
[13:40] <nessita> ralsina: the server is closing the connection on me
[13:40] <ralsina> nessita: it's ok, whenever it works. Or I can just... call you
[13:40] <nessita> ralsina: shall we skype?
[13:40] <ralsina> let's skype
[13:41] <ralsina> nessita: you seem offline in skype
[13:41] <nessita> ralsina: I'm nidart.work
[13:41] <nessita> nbidart.work
[13:56] <nessita> gatox: ubuntuone_installer/gui/qt/tests/__init__.py:
[13:56] <nessita>     62:  [C0111, FakeBackend] Missing docstring
[13:56] <gatox> :(
[14:08] <nessita> gatox: I will be doing another review round after lunch, I will switch to monkey coder role now :-)
[14:09] <gatox> nessita, :P ok
[14:41] <gatox> nessita, the test for oauth is working fine now.... maybe was something blocked in my machine and it needed to be restarted
[14:41] <nessita> gatox: what error were you having?
[14:42] <gatox> nessita, also all the branches should be ready now..... i hope without lint issues :S (this weekend i'm going to try to have u1lint working)
[14:42] <gatox> nessita, the test i mentioned earlier was throwing an exception and blocking the process
[14:45] <dobey> meh, i suck at writing e-mail
[14:46] <nessita> dobey: if it's of any consolation, we all do
[14:46] <nessita> gatox: we will get together and get your env properly working
[14:46] <nessita> gatox: this week has been crazy for me with the incoming travel and all
[14:47] <gatox> nessita, yes, i know, no problem..... let me check if i can fix u1lint this weekend, and if i can't i'll let you know next week to see if you can give a  hand
[14:48] <nessita> gatox: well, next week I will be a little far away from your home... but sure ;-)
[14:49] <gatox> nessita, uds?
[14:49] <nessita> gatox: design sprint and UDS, so I will be out for 2 weeks
[14:49] <nessita> gatox: did you read my standup? :-D
[14:49] <gatox> nessita, :P ok...... emmmmm when you get back jeje
[14:50] <mandel> nessita, this guy is ready for a re-review whenever you have the time: https://code.launchpad.net/~mandel/ubuntuone-windows-installer/fix-reactor-import/+merge/79842
[14:50] <nessita> mandel: ack!
[14:50] <gatox> mandel, you never answer my questionnnnnnnn
[14:50] <gatox> mandel, and here you are :P
[14:51] <gatox> mandel, this is the same branch as yesterday?? i ask because i have another branch that depends on this one.... to keep it up to date
[14:53] <mandel> gatox, oh, wich question? I propbably missed it
[14:53] <mandel> gatox, and yes, is the same branch with some code that has been moved and more tests
[14:54] <gatox> mandel, ahhhhhh ok
[14:55] <dobey> i think i will have to get this e-mail finished up after lunch. my brain is still all over the place after yesterday :-/
[15:13] <dobey> ok, lunch time; bbiab
[15:48] <Reaper> Is there an upload limit in ubuntu 1?
[15:57] <gatox> lunch..... brb
[16:00] <mandel> nessita, ping
[16:01] <nessita> mandel: pong
[16:02] <mandel> nessita, I have a very quick question, in control panel, we are using gi AND pygtk, right?
[16:03] <nessita> mandel: yes, for now
[16:04] <nessita> mandel: I will be migrating that, and very likely removing the GTK UI
[16:04] <mandel> nessita, ok, understood
[16:05] <nessita> mandel: can I help somehow?
[16:05] <mandel> nessita, I had a present for you: http://paste.ubuntu.com/714312/ but then I realized that it would not solve anything because we use pygtk in the widgets :(
[16:06] <nessita> mandel: you made that?
[16:06] <mandel> nessita, I was going to try and propose that for twisted when some changes so that there is not code repetition, but well, other day
[16:06] <mandel> nessita, yes, but based on the gtk2reactor, I'm not THAT smart hehe
[16:07] <mandel> and itamar is not in #twisted so that I could merge it nicely
[16:08] <nessita> mandel: dobey is already working on this, so I would advice talking to him, if you're really interested. On the other hand, I would also advice not to put time ont his, for now...
[16:08] <nessita> mandel: or are you blocked?
[16:08] <mandel> nessita, no, I took some time while watching tv, that is all :)
[16:08] <nessita> mandel: I think you can safely run the tests for the non-ui stuff in isolation?
[16:08] <mandel> nessita, but wanted to try before I EOD, that is all
[16:14] <alecu> hello #ubuntuone!
[16:14] <mandel> alecu, hello!!!
[16:15] <alecu> hi there, mandel!
[16:15] <nessita> hi alecu!
[16:17]  * alecu has sweated more today than in the whole previous year.
[16:17] <dobey> mandel: that pastebin also doesn't even import successfully :)
[16:17] <nessita> alecu: that's goo news! (?)
[16:17] <dobey> mandel: you basically did the exact same thing as someone else did, who proposed it on twisted's trac :)
[16:18] <alecu> on the plus side, I'm almost moved, and starting tomorrow, my office will officially be the back room in my own house.
[16:21] <alecu> gatox, mandel, ralsina: if any of you is having a happy day, doing this review will keep your spirits up: https://code.launchpad.net/~alecu/ubuntu-sso-client/fix-get-token-name/+merge/79594
[16:21] <alecu> if it's not a happy day, it will have no side-effects.
[16:21] <mandel> dobey, it does import correctly, I have tried it :P
[16:22] <mandel> dobey, anyways, it was a 10 min thing while watching tv hehehe
[16:23] <dobey> mandel: well, you got lucky then, because it's point at things that don't exist :)
[16:23] <nessita> alecu: and what about amelia's bedroom? you share? :-)
[16:23] <mandel> dobey, I don't want good generals, I want lucky ones jeje
[16:24] <dobey> heh
[16:24] <alecu> nessita, no, it's Luli's deskroom I'm taking over.
[16:24] <nessita> ah, heh
[16:25] <alecu> nessita, amelia's bedroom is usually a battlefield of plush toys, so it's not easy to work there.
[16:26] <dobey> alecu: more than 3 toys? it must be really boring in there :P
[16:26] <alecu> hahahaha
[16:28] <gatox> alecu, jeje ok on it
[16:34] <dobey> ok, back to e-mail
[16:40] <gatox> alecu, +1
[16:41] <dobey> ugh, this weather sucks
[16:43] <alecu> gatox,   |o|
[16:43] <alecu> (tie fighter)
[16:46] <nessita> alecu: so, when you have some spare time... I'd need your wisdom about twisted.pb. Turns out I found another bug, where the exceptions from the pb server are not being 'forwarded' to the clients
[16:46] <alecu> nessita, now would be a great time
[16:46] <nessita> alecu: I read the doc, and tried some approaches, but I had no luck. Spent quite some time yesterday fighting with this... so I skipped the test and moved on, until you came back
[16:46] <nessita> alecu: nice!
[16:47] <nessita> alecu: can you please branch:
[16:47] <alecu> sure
[16:47] <nessita> lp:~nataliabidart/ubuntuone-client/u1sdtool-multiplatform
[16:47] <nessita> alecu: you can check all is green for you by running, in linux:
[16:47] <nessita> u1trial tests/platform/test_tools.py
[16:48] <nessita> alecu: once that's green on linux, confirm is all green on windows, but with a SKIPPED test
[16:48] <nessita> (JIC, latest revno is 1176)
[16:51] <alecu> nessita, all green in linux. testing windows now
[16:52] <nessita> alecu: so, the SKIPPEd test is  test_rescan_from_scratch_missing_volume. If you un-skip it, you'll get a trace where the exception received is not the expected, but a TypeError. That TypeError is raised from within twisted internals, complaining that the thing raised from the server is a string, which is "a lie"
[16:53] <nessita> alecu: since the exception is raised from: ubuntuone/syncdaemon/interaction_interfaces.py:1282
[16:56] <nessita> alecu: oh, I'll be back in a few minutes, I'm at my mother in law's and she's screaming the lunch is ready :-D
[16:56] <nessita> brb!
[17:02] <alecu> nessita, on windows I'm getting a lot of these: http://pastebin.ubuntu.com/714368/
[17:02] <alecu> on the "interaction_ifaces" module
[17:07] <Reaper> I am trying to send a iso into ubuntuone and its not working.. any ideas.. its it too big?
[17:07] <Reaper> the iso is under 5 gigs.
[17:08] <dobey> phew, finally "done" with that mail
[17:08] <dobey> Reaper: i have uploaded ISOs just fine.
[17:09] <Reaper> hmm i think its user error then
[17:09] <Reaper> lol
[17:09] <Reaper> I tested a txt file and got that just fine
[17:10] <dobey> Reaper: if it's not uploading, it could be network, server, or client issues. but i can't really say without knowing the logs. i can tell you that our servers are *very* busy right now though, with oneiric having just been released :)
[17:10] <Reaper> yeah, thats what ive heard
[17:10] <dobey> and there is a bug or two in the client, for which fixes which are being worked on at the moment, that we hope to get out to oneiric in the next couple weeks (SRUs take at least 7 days to get out)
[17:11] <Reaper> ahh
[17:11] <dobey> do you have a ~/.cache/ubuntuone/log/syncdaemon-exceptions.log ?
[17:13] <dobey> hrmm, i need to get my laptop fixed
[17:17] <Reaper> not sure, ill have to take a look at it
[17:17] <dobey> wtf; now network magically works again on my laptop
[17:17] <Reaper> I love when that happens
[17:18] <nessita> alecu: back
[17:18] <dobey> yeah, i upgraded to oneiric on it, and it stopped working
[17:18] <nessita> alecu: can you just run the test_tools tests?
[17:18] <dobey> spent 3 days trying to figure out what it was and how i could get it updated as there were pending updates still
[17:18] <dobey> shut it off for 2 weeks, and it suddenly works again
[17:18] <Reaper> lol and now it works. that awesome lol
[17:18] <beuno> Reaper, we are having some issues with our servers
[17:19] <beuno> so that's probably the cause
[17:19] <Reaper> thats what I was thinking. I tested a text file earlier today and that went through so ill just hang tight till the servers calm down.
[17:19] <dobey> Reaper: just to check, run this in a terminal please: "grep ValueError ~/.cache/ubuntuone/log/syncdaemon*"
[17:19] <Reaper> k 1 sec
[17:20] <dobey> Reaper: if you see "ValueError: I/O error..." type message, you're also hitting one of the known bugs we're working on a fix for :)
[17:20] <Reaper> ahh got cha
[17:22] <Reaper> yep, getting alot of those errors
[17:22] <dobey> now if i could just get the UI back to how i had it, on my laptop (which i guess isn't going to happen with gnome3) :-/
[17:22] <dobey> Reaper: restarting syncdaemon might help with that: u1sdtool -q; u1sdtool -c
[17:23] <dobey> Reaper: but if you put a big iso in there again, it will likely crop up again
[17:24] <Reaper> says its being uploaded now
[17:24] <Reaper> but its said that before.. its about 991mb
[17:26] <dobey> yeah, the notification will pop up again if you get reconnected, too
[17:27] <Reaper> got cha, its nothing big. was just stoked about trying that svc and was freakin because I couldnt get it to work. Just wanted to make sure I wasnt completely retarded lol
[17:28] <dobey> unfortunately it does seem to be our fault. but hopefully we can have a fix soon
[17:29] <nessita> alecu: let me know when you're back
[17:31] <Reaper> eh, it happens lol
[18:43] <nessita> alecu: ping?
[18:43] <alecu> nessita, pong
[18:43] <nessita> alecu: hey there, sorry for leaving in the middle of the debugging, not sure if you receive my latest messages
[18:44] <alecu> nessita, no problem, I disappeared for a bit too.
[18:44] <alecu> nessita, I'm still trying to follow the code
[18:44] <nessita> alecu: i was saying, could you just run the tests/platform/test_tools.py in windows? that should be all green
[18:44] <alecu> nessita, and I'm not sure at what point the error is being transformed into string.
[18:44] <alecu> nessita, I've run that, and now I'
[18:45] <alecu> I've commented the skip
[18:45] <alecu> nessita, and it shows an error where the exception comes as a string
[18:45] <alecu> nessita, so I'm trying to understand at what point the exception is translated into a string
[18:46] <nessita> alecu: me neither. I read the doc and it said that the exceptions being raised from the pb server should inherit from pb.Error. I did that, I created a new error inheriting from that class, and added a decorator (not in the branch) so every remote method catches everything and raises the specific error, but no luck
[18:51] <alecu> nessita, I'm reading the PB tutorial, and it says exactly what you say:
[18:51] <alecu> http://twistedmatrix.com/documents/current/core/howto/pb-usage.html#auto4
[18:51] <nessita> alecu: yeah, I may have done something wrong... but I couldn't find what
[18:52] <nessita> alecu: anyways, the doc gives an example where even if you raise a non pb.Error exception, the exception is indeed forwarded to the client, but ugly messages appear server side
[18:52] <nessita> alecu: so that puzzles me a lot
[18:52] <alecu> yup, some message should be shown in the log in that case
[18:53] <nessita> alecu: maybe one of the decorators that mandel added are stringyfing that?
[18:54] <nessita> alecu: I looked, and my eyes found nothing, but yours may?
[18:57] <alecu> nessita, I can't see anything wrong neither. I'm looking at both RemoteMeta and signal, the two places with meta-magic, and it's not being done there.
[18:58] <nessita> alecu: I'm thinking skipping this test and filing a bug, for now, if you don't see anything popping up. This has been "broken" the same until now, can remain like this a bit longer
[19:00] <alecu> nessita, I'm fine with that, since it's not something that your branch breaks
[19:00] <nessita> right
[19:00] <nessita> alecu: thanks for your time/eyes!
[19:01] <alecu> nessita, I'll keep looking a bit more
[19:01] <alecu> nessita, no problem. Sorry if I was a little more sleepy than usual :-)
[19:01] <nessita> alecu: -)
[19:01] <nessita> :-)
[19:04] <gatox> ok, EOD for me... see you tomorrow... or later :P
[19:10] <alecu> nessita, I'm pdbing the code, and it seems that it *gets* into """log.msg("Peer will receive following PB traceback:"...."""
[19:10] <nessita> alecu: wtf?
[19:11] <nessita> alecu: how's that possible?
[19:14] <alecu> nessita, perhaps the twisted logs are not enabled on this tests. Or something like that.
[19:29] <alecu> nessita, I've added "from twisted.python import log;import sys;log.startLogging(sys.stderr)" before that test starts. Here's the result: http://pastebin.ubuntu.com/714507/
[19:33] <nessita> alecu: I don't see a str cast there, do you?
[19:35] <alecu> i do not see it either.
[19:35] <alecu> but it looks like what the pb docs say
[19:38] <alecu> nessita, the thing that I don't understand...
[19:38] <alecu> nessita, why do we have some errors as signals (ie, RootMismatch, QuotaExceeded), and some other errors as exceptions? (VolumeDoesNotExist, etc)
[19:40] <nessita> alecu: the signals you mention, in particular, are not 'errors' but 'states' (por llamarlo de alguna forma)
[19:40] <nessita> alecu: despite that, we can't have signal for every error
[19:40] <alecu> nessita, InvalidName, BrokenNode are signals too
[19:40] <alecu> (on a different object, though)
[19:41] <nessita> alecu: yeah, but my point is what if we have an IOError, an outofmem error, whatever unxpected signal
[19:41] <nessita> alecu: I would like to know that would be propagated to the caller, if unhandled
[19:41] <nessita> alecu: so, we can certainly handle this specific case in the server side
[19:41] <nessita> but worries me that the exceptions are not 'forwarded'
[19:42] <alecu> nessita, right.
[19:42] <dobey> ugh, i can't remove multiple packages at once in software-center
[19:42] <alecu> nessita, but the thing with signals is that they can be "async"
[19:42] <alecu> nessita, if I'm returning the error as an exception, it needs to be fast
[19:43] <alecu> (kinda)
[19:43] <dobey> nessita: isn't there a "GenericError" signal on the main object or something, that just gives you all errors on a dbus signal?
[19:43] <alecu> so, that's really the main difference.
[19:43] <nessita> alecu: not sure what that difference is
[19:43] <nessita> dobey: this is windows IPC side...
[19:44] <dobey> nessita: oh, does it not have the same interface as the dbus service?
[19:44] <alecu> nessita, the difference is related to the timeouts we discussed the other day.
[19:45] <alecu> dobey, it should be pretty similar
[19:45] <nessita> dobey: I'm refactoring it so interfaces matches. But ATM, exceptions are not being sent from server to client
[19:45] <nessita> alecu: right
[19:47] <dobey> nessita: ah. well on linux there is a genericerror signal or something, right?
[19:48] <nessita> dobey: yes and no... this specific case was throwing a DBusException
[19:48] <alecu> nessita, so, we should either: a) subclass all exceptions as pb requires. Or b) catch every exception and match it against a list of known exceptions in each function's decorator, and wrap those exceptions in a pb exception.
[19:49] <alecu> nessita, b) sounds way over the top, so we should find a cleanish way to do a)
[19:49] <dobey> nessita: the server was getting a dbusexception from something else? or what?
[19:50] <nessita> alecu: I tried the (b) approach, and even when raising a pb.Error, the TypeError still remains, that's the puzzle
[19:50] <alecu> nessita, I'd love to see *that* branch
[19:50] <alecu> nessita, but first I need to pick up amelia.
[19:50] <nessita> alecu: you can try it yourself: inside the ubuntuone/platform/windows/ipc.py, in rescan_from_scratch, raise a pb.Error('foo')
[19:50] <nessita> alecu: ah, ok, later then
[19:51] <alecu> nessita, good point, I'll try that. thanks!
[19:52] <nessita> dobey: not sure what you ask. When I say server I was referring to the dbus server side... and yes, any unhandled exception in our dbus (and lower) layers will send a dbusexception
[19:52] <dobey> ok, some cake for me, and then poking at tcole's branch
[21:13] <dobey> have a good evening all!
[21:50]  * nessita ->eod
[21:50] <nessita> see ya tomorrow crowd!