[00:15] <karni> verterok: heheh. I noted to myself "u1-java-sp Request should support cancellation". I opened Request.java and see "TODO: add cancel support." written by you :D
[00:28] <lalejand> Hi everybody, is there a way to do some ignore list for u1 sync ? I have a lot of svn files and I don't want them to be synced :/
[00:43] <lalejand> Is there a way to do some ignore list for u1 sync ? I have a lot of svn files and I don't want them to be synced
[01:01] <lalejand> Is there a way to do some ignore list for u1 sync ? I have a lot of svn files and I don't want them to be synced (sorry for repeating, it's just in case someone connects and nows the answer)
[02:33] <kenvandine> ugh
[02:34] <dobey> ugh
[02:42] <kenvandine> dobey, u1 import errors
[02:42] <kenvandine> ImportError: No module named platform.linux.tools
[02:42] <kenvandine> from u1sdtool
[02:43] <kenvandine> and syncdaemon isn't starting
[02:43] <dobey> kenvandine: sponsor my libubuntuone and ubuntuone-dev-tools uploads (merge proposals), and i think nessita did one for ubuntuone-control-panel too
[02:43] <kenvandine> TypeError: signal_handler() takes exactly 1 argument (2 given)
[02:43] <dobey> hmm
[02:44] <kenvandine> looks like u1 is busted all over the place :/
[02:44] <kenvandine> and i needed to share a file!
[02:44] <dobey> we switched the way the python modules install is done; and it's incompatible with the old way :( (yay python)
[02:44] <kenvandine> i can sponsor
[02:44] <kenvandine> ugh
[02:44] <dobey> yeah
[02:44] <kenvandine> incompatible... sounds like time to revert
[02:45] <kenvandine> i can sponsor uploads
[02:45] <dobey> kenvandine: if you sponsor the uploads or install from ppa:ubuntuone/nightlies it should work
[02:46] <kenvandine> i don't see the control panel one
[02:46] <dobey> maybe she didn't do it yet :(
[02:46] <dobey> but i asked her to because i knew it would break
[02:48]  * kenvandine sponsors libu1
[02:49] <dobey> thanks
[02:50] <dobey> maybe i'll have to do one for u1cp too, if nessita didn't
[02:52] <kenvandine> so what should get syncdaemon working?
[02:53] <kenvandine> libu1 doesn't get u1sdtool working
[02:53] <kenvandine> dobey, ^^
[02:54] <dobey> kenvandine: what all directories do you have immediately under /usr/share/pyshared/ubuntuone ?
[02:54] <kenvandine> ah, controlpanel
[02:55] <dobey> if you apt-get remove --purge python-ubuntuone-control-panel it should work
[02:55] <dobey> if it still doesn't, then there's another new bug in syncdaemon i guess :(
[02:55] <kenvandine> yeah, works
[02:55] <dobey> yay
[02:55] <kenvandine> syncdaemon is working now
[02:56]  * kenvandine uploads
[02:56] <kenvandine> can you do controlpanel?
[02:56] <dobey> ok, so we just need a controlpanel release
[02:56] <dobey> let me see where it stands
[02:57] <dobey> kenvandine: did you sponsor ubuntuone-dev-tools also?
[02:58] <kenvandine> working on it
[02:58] <kenvandine> well, still working on libu1
[02:58] <kenvandine> in pbuilder now
[02:58] <dobey> ok, i'll do a cp release
[02:59] <dobey> kenvandine: are you very familiar with vapigen?
[03:00] <kenvandine> yeah
[03:00] <dobey> any idea why it would generate a totally empty vapi save for the comment on the first line saying it was generated, when the .gir seems valid?
[03:01] <kenvandine> no
[03:01] <dobey> damn :(
[03:01] <kenvandine> that is strange
[03:01] <kenvandine> never seen it empty
[03:01] <dobey> yeah, tell me
[03:01] <dobey> i blame webkit
[03:01] <kenvandine> everything is webkit's fault
[03:02] <dobey> pretty much
[03:03] <dobey> it took me a while to get a .gir built for it that vapigen wouldn't just die out on, and now that i got it built, the vapi is empty and vapigen just acts like everything was fine and says 0 errors :(
[03:12] <kenvandine> dobey, ok both of those are sponsored
[03:16] <dobey> great, doing u1cp right now
[03:17] <kenvandine> thx, ping me when it is ready
[03:21] <dobey> kenvandine: https://code.launchpad.net/~dobey/ubuntu/natty/ubuntuone-control-panel/release-0-8-3/+merge/49491
[03:21] <kenvandine> thx
[03:27] <kenvandine> dobey, sponsored, thx
[03:30] <dobey> no problem
[03:30] <dobey> wish i knew what was going on with vapigen on this webkit gir though
[03:35] <dobey> works fine for couchdb-glib and libsyncdaemon though :(
[03:36] <dobey> which sucks, because i am almost at a point where encompass is usable, and i need more of the webkit api than is in the vapi that comes with vala :(
[03:43] <dobey> oh well, 'night
[10:53] <duanedesign> morning all
[10:56] <simontol> Hi I'm trying to sync a folder using ubuntuone u1sync tool but  i receive this message : u1sync: error: --oauth is required
[10:56] <simontol> python-oauth is installed and U1 works via GUI
[11:10] <simontol> is there anybody out there?
[11:52] <karni> morning!
[11:53] <karni> notice simontol trying to use u1sync -- I think it's deprecated and, as just seen, can be misleading.
[11:53] <karni> I personally have never used nor even try using it.
=== AlanBell is now known as world
=== world is now known as AlanBell
[12:37] <kklimonda> hey, I've been getting an indicator "Ubuntu One uploading 26/26 files, 98% complete" since I've restarted my computer after an update. It seems stuck, u1sdtool --waiting-c returns nothing but --waiting-m "crashes" with String parameters to be sent over D-Bus must be valid UTF-8
[12:37] <kklimonda> what can I do to make it stop showing up from time to time? :)
[12:39] <karni> kklimonda: aim that at alecu when he's online
[12:44] <duanedesign> simontol hello
[12:44] <duanedesign> oops, he is gone
[12:44] <duanedesign> hey kklimonda !
[12:44] <duanedesign> hello friend
[12:44] <kklimonda> hey duanedesign :)
[12:44] <duanedesign> hello karni
[12:44] <kklimonda> how are you?
[12:44] <duanedesign> kklimonda: I am well. trying to thaw out from all the snow
[12:44] <karni> hi duanedesign :)
[12:45] <karni> duanedesign: still under snow?
[12:45] <duanedesign> kklimonda: I saw you in #ubuntu-desktop the other day but got busy and did not get a chance to say hi
[12:45] <kklimonda> duanedesign: I've already forgotten what the snow is and it's only half of the winter :/
[12:45]  * karni codes Upload command
[12:45] <duanedesign> karni: yes finally after two weeks we had the first day where it got above freezing yestersday
[12:45] <kklimonda> duanedesign: meh, no problem - I'm also pretty busy lately :)
[12:45] <karni> kklimonda: talk to me. I bought a snowboard and used it once ;d
[12:45] <kklimonda> karni: ahaha :D
[12:46] <duanedesign> yeah my bicycle has not gotten much use lately
[12:47] <duanedesign> karni: karni is u1sync depreaceated across all versions?
[12:47] <duanedesign> my spelling skills are still asleep
[12:47] <duanedesign> I noticed it uninstalled with the last updatte
[12:47] <karni> duanedesign: I have no clue, but I saw guys some time talking it's misleading it's still there. all in all, I don't think it should be used.
[12:48] <karni> aha! :)
[12:48] <duanedesign> i agree. That is what I have always been told. It was for testing purposes only
[12:48] <karni> right.
[12:48] <duanedesign> lots of new stuff in U1 I have a lot to catch up on
[12:48] <duanedesign> the logs are starting to look foriegn :)
[12:49] <karni> duanedesign: :D
[12:57] <duanedesign> karni: from reading the scrollback the last couople weeks it sounds like you are making great progress on your client
[12:58] <karni> duanedesign: it proved to be very challenging, but indeed I have ported quite a few components of syncdaemon for queueing purposes
[12:58] <karni> duanedesign: it's been the time when there's much work and little visual feedback, since it's all internal flesh.
[12:59] <karni> once I've got Down/Upload commands, I want to integrate the whole puzzle to make sure it was all worth it :)
[13:03] <karni> duanedesign: hahahah it'd be fun one day to use this work and have U1 run on any OS supporting Java runtime ;D
[13:15] <karni> I just hate how gracefully Python classes declare typelessly new object fields (__slots__) ;x
[13:44] <duanedesign> karni: you have been doing something in  Python?
[13:44] <karni> duanedesign: I have been intensely studying syfinal Deferred putDeferred = mActionQueue.mClient.putContent(mVolumeId, req.getNewId(), "",	hashInfo.hash, hashInfo.crc32, hashInfo.size, hashInfo.deflatedSize,
[13:44] <duanedesign> ohh,probably mean reading the source
[13:44] <karni> oops sorry
[13:45] <karni> exactly. studying syncdaemon source.
[13:46] <duanedesign> I wish I had more time to code. I am hoping to take a class in April. The college next to my house has some programming classes
[13:46] <karni> duanedesign: actually, reading both ubuntuone-client _and_ ubuntuone-storage-protocol. ubuntuone-java-storageprotocol is not yet complete :<
[13:46] <duanedesign> :)
[13:47] <karni> duanedesign: what makes you busy? no time for code?
[13:47] <duanedesign> looking for a job right now
[13:47] <karni> I see. hopefully you'll find one where you code :D!
[13:48] <duanedesign> hacking has been a hobby for a couple years. I like it so much I would really like to transision into a career doing programming
[13:48] <karni> duanedesign: very cool :) coding has always been what makes me tick!
[13:49] <karni> *tic.. not sure which one ;D
[13:49] <duanedesign> i need to learn C better
[13:50] <karni> once I read comparisons of different languages. I liked the description of C - "so powerful you can shoot yourself in the foot"
[13:50] <duanedesign> I mostly do Python because that is what I learned first. not a lot of jobs around here for Python. Except some Django jobs
[13:50] <karni> uhm
[13:50] <duanedesign> karni: i wish I would of started with C. It helped me understand Python better
[13:51] <karni> you're such a great support here. never thought of applying for a position at canonical for python programmer? perhaps after some courses?
[13:51] <karni> you very well comprehend what's goind on with U1, current bugs, workarounds, solutions. you're a treasure of tips and hints.
[13:52] <duanedesign> thank you :)
[13:52] <karni> you are very welcome
[13:53] <duanedesign> i started working around this project because I thought the application was interesting. I stayed around because the team is so friendly and welcoming.
[13:54] <karni> duanedesign: I seriously you where 'one of them' for quite a long time ;)
[13:55] <karni> *thought
[13:55] <karni> I'm missing words, not good.
[13:55] <duanedesign> ok, my spelling and typing skills are still asleep
[13:56]  * duanedesign gets a cup of coffee
[13:57] <karni> verterok: you were right. I'm now myself seeing how much duplication there is between the protocol and syncdaemon itself. it it was me, I would use the requests as ActionCommandQueue object themselve. I might do that one day.
[14:00] <duanedesign> wow! someone really cleaned up the Ubuntu One bug queue
[14:00] <karni> :)
[14:22] <duanedesign> hmm, ubuntuone-launch would be...foundatiions+  ?
[14:46] <kklimonda> wow, U1 did that again
[14:46] <kklimonda> it started uploading all my music again
[14:49] <kklimonda> ah, not all of it.. just some..
[14:49] <nessita> kklimonda: did you or your your music player changed the file in any way?
[14:50] <nessita> metatada perhaps?
[14:50] <kklimonda> nessita: no
[14:50] <nessita> ah
[14:50] <kklimonda> nessita: one day U1 has decided that it shall reupload all my music to server on my phone data plan ;)
[14:51] <nessita> kklimonda: is this syncdaemon or your phone?
[14:51] <kklimonda> nessita: no, on Ubuntu
[14:51] <nessita> hum
[14:51] <kklimonda> nessita: I was just connected using my phone when I was away from home
[14:51] <nessita> is there any FS_FILE_CLOSE_WRITE or similar in the log file, for that file
[14:52] <nessita> ?
[14:52] <kklimonda> nessita: yes, but it's lyingh ;)
[14:52] <kklimonda> lying even
[14:52] <kklimonda> files were not touched for weeks
[14:54] <kklimonda> maybe something has opened file in the write mode, and then closed it without actually doing any writes?
[14:58] <kklimonda> no, it doesn't generate any events..
[14:59] <kklimonda> so something would actually have to change the content of the file, but leave the old mtime.. I don't think I use any nefarious application like that :)
[15:07] <duanedesign> hello nessita
[15:09] <kklimonda> I've scared her off ;)
[15:37] <duanedesign> :D
[15:44] <mongy> U1 is 'playing up' today.  on/off/on/off/off some more, still off.
[15:46] <kklimonda> indeed
[15:47] <kklimonda> and I get a new notification at every status change.. a bit irritating ;)
[15:47] <nessita> kklimonda: sorry, got distracted :-)
[15:47] <nessita> hi duanedesign!
[15:47] <mongy> im still 10.10, using rye's indicator.
[15:47] <nessita> kklimonda: can you please paste the fragment of the log file where the FS_* event is logged for that file?
[15:48] <mongy> its sufficient enough, and not too ott
[15:50] <kklimonda> nessita: some random lines around LS_ event http://paste.ubuntu.com/566345/
[15:55] <duanedesign> there is a bug in the latest update of Natty that is preventing it from starting on my box
[15:58] <nessita> duanedesign: what bug?
[15:58] <nessita> kklimonda: is "Everything You Can Think.mp3"?
[16:00] <nessita> kklimonda: so, at file system level, if pyinotify reports FS_FILE_CLOSE_WRITE it means that the file changed. If you only open that file without saving it, you get a FS_FILE_CLOSE_NOWRITE
[16:02] <kklimonda> nessita: but I don't see any application modyfing a file, and not updating the mtime for it. I know it's possible, but that would require someone to actually do this and I can't think of a reason for that.
[16:03] <kklimonda> what I can do is download the file from server, and see if it's the same as the one on my disk
[16:03] <nessita> kklimonda: random ideas: maybe you have a music server that scans all your music files to update a database?
[16:04] <kklimonda> or I would, if the web interface worked o.O
[16:04] <kklimonda> I can browse directories, but not touch files.. ugh..
[16:04] <kklimonda> at least not the More menu
[16:05] <duanedesign> nessita: I noticed it yesterday. Looks like there is a new version available. Let me see if it fixes it
[16:06] <duanedesign> nessita: it was an Import Error for the module clientDefs
[16:06] <kklimonda> nessita: the files are exactly the same files - even if notify has fired modified events a quick checksum would save some bandwidth :/
[16:06] <duanedesign> that spelling may be wrong
[16:07] <nessita> kklimonda: yeah, FF4 is not working with our web ui. We're waiting for FF4 beta to settle before fix
[16:07] <nessita> duanedesign: ah, haven't seen that in a while
[16:07] <nessita> ok, I gotta run to buy some groseries
[16:08] <nessita> otherwise I'll starve :-)
[16:08] <nessita> see ya later guys!
[16:08] <kklimonda> nessita: also, how is the progress of delta uploads? ;)
[16:08] <kklimonda> cya
[16:10] <duanedesign> bye nessita
[16:11] <karni> great.. looks like no testing today :S
[16:12] <karni> indeed U1 is 'playing up', and I don't like it.
[16:12] <beuno> karni, hey
[16:12] <beuno> it's fixed
[16:12] <karni> beuno: let me check..
[16:12] <beuno> should be stabalised in a few minutes
[16:13] <karni> beuno: ah.. good, very good
[16:13] <karni> SD was giving me handshake timeouts, and it didn't make me happy ;_;
[16:13] <beuno> yeah, not sure what happened, I just dropped in and saw __lucio__ dealing with it
[16:14] <karni> ack
[16:19] <duanedesign> hello beuno
[16:22] <beuno> heya duanedesign
[16:24] <duanedesign> ok, looks like the import error was fixed
[16:24] <manixate> Hi!
[16:24] <manixate> I want to use ubuntuone with Mint 10
[16:24] <manixate> any help
[16:24] <duanedesign> hello manixate
[16:24] <manixate> hi
[16:25] <manixate> i installed ubuntuone-client
[16:25] <manixate> so i needed to ask that it is only in the mintmenu
[16:25] <manixate> cant it be like its in Me Menu in Ubuntu ?
[16:28] <duanedesign> In the newest version (in Natty) the Ubuntu One link moved to the Messaging Menu
[16:28] <duanedesign> but that is version 1.5.4
[16:30] <verterok> karni: hi, don't know what's the best approach to avoid duplication...but I know it's there :)
[16:30] <karni> verterok: true ;)
[16:30] <verterok> karni: also the Volume hierarchy (defined in volume_manager.py) is duplicated :(
[16:31] <karni> verterok: that's one of the things I haven't looked into (perhaps it's substituted with something else I'm using)
[16:31] <verterok> karni: yes, ti should be possible
[16:31] <verterok> *itt
[16:32] <karni> verterok: it was easy to implement progress callback for the download request (just wrapped the output stream), but I don't think it'll be that easy for upload request, it was somewhat more complicated I think.
[16:39] <verterok> karni: I don't remember the code...but it should be another stream wrapping, right? :)
[16:40] <karni> verterok: it was submitting chunked data to RequestHandler and only there writing to a stream, so the cleanest way would be for me to add callback support on Request level directly to your branch.
[16:41] <verterok> karni: and where is the chuncked data comming from?
[16:42] <karni> in practice, what I'm missing the most: request build-in down/upload progress callbacks, request cancellation, and more verbose exception throwing (too little exceptions gives little or no info on what's going on)
[16:42] <karni> let me check
[16:44] <karni> verterok: what you mean is wrap the input stream I'm reading from in case of Upload request instead of wrapping the stream we're sending the data, correct?
[16:45] <karni> that's doable. it's using is.read(buffer, this.offset, RequestHandler.MAX_PAYLOAD_SIZE); -- so I would wrap the input stream
[16:45] <verterok> karni: got lost, come again? :)
[16:45] <karni> verterok: anyway, exactly. the thing is
[16:45] <karni> u1-s-p has it on the Request level
[16:46]  * karni collects thoughts
[16:46]  * karni would like to have everything ready on plate ;d
[16:46] <karni> verterok: the Upload/Download request of ubuntuone-storage-protocol takes a callback as a parameter
[16:47] <karni> verterok: so I thought perhaps I should commit some changes to your branch. that's it.
[16:47] <karni> instead of doing it on the client side.
[16:47] <verterok> karni: PutContent request doesn't accept a callback :)
[16:47] <karni> 'my' side, that is. the application side. just give the application possibility to provide the callback as a parameter, and be happy with the result.
[16:48] <karni> it doesn't ? wiat.. I must have seen it. I'm sure.
[16:48] <verterok> karni: the GetContent request accepts 2 callback: "callback" and node_attr_callback
[16:49] <verterok> callback is used to process the incomming bytes and node_attr_callback is to process the node_attr messages from the server
[16:49] <karni> ok, now it's changed (newest facundos free-queues branch)         f = UploadProgressWrapper(self.tempfile, uploading, self.progress_hook)
[16:50] <verterok> karni: but that isn't the protocol. that's the syncdaemon upload stuff
[16:50] <karni> one sec ;)
[16:51] <karni> you're right.. damn it, I would sware I saw that callback as a Request parameter
[16:51] <karni> ah! that's it, that's what you're talking about             callback=self.cb, node_attr_callback=self.nacb
[16:52] <karni> verterok: right. I'll just wrap it up, to..
[16:52] <verterok> :)
[16:52] <karni> verterok: do you see how easy it is to get lost using source-as-the-docs ;(
[16:52] <karni> (not to mention I don't know python in practice, because thats _not_ your fault)
[16:53] <verterok> karni: this are just python client implementation details
[16:53] <karni> it's just quite huge chunk of code.
[16:53] <karni> true..
[16:53] <verterok> karni: the "java" protocol client design could have been completely different
[16:55] <karni> I kinda liked the idea to pass a callback object to the request hahah. Anyway, I'll do the same for upload what I did for download.
[17:10] <verterok> k
[17:10] <verterok> lunch time! later!
[17:19] <karni> __lucio__: could you briefly explain what are 'fs', 'keys', 'mdid', '_changes' from the class FSKey under sync.py ?
[17:20] <karni> I'll be right back.
[18:37] <mongy> u1 is still down for me. been that for a few hours now
[18:41] <dobey> nessita: hey. i made a release of u1cp last night, so that everything would work again on narwhal.
[18:41] <dobey> mongy: can you elaborate on what is wrong for you exactly?
[18:51] <nessita> dobey: hey there. what was not working?
[18:51] <dobey> nessita: stuff broke due to the change in how we install python modules in u1 packages
[18:51] <nessita> ah
[18:52] <nessita> dobey: did you uploaded a new package or did you do the whole tarball release?
[18:52] <dobey> nessita: that's why i asked you for the release yesterday, but you hadn't made on when kenvandine came in irc last night to complain about u1 failing with importerrors :)
[18:52] <dobey> nessita: tarball
[18:52] <nessita> dobey: yeah, I was doing that today. But I didn't know it will break!
[18:53] <nessita> dobey: anyways, thanks for fixing that
[18:53] <dobey> no problem
[18:54] <nessita> dobey: you didn't write any release info...
[18:55] <dobey> eh?
[18:55] <dobey> what do you mean?
[18:56] <nessita> dobey: when doing the release of the tarball, as far as I know we have to write the Changelog as the release info
[18:56] <nessita> anyways, is done. Thanks
[18:56] <dobey> what release info?
[18:58] <nessita> dobey: you don't use lp-project-upload?
[18:58] <dobey> no
[18:59] <nessita> dobey: look at every tarball release on https://launchpad.net/ubuntuone-control-panel/+download (except for yours)
[18:59] <nessita> dobey: there is a expandable 'Release information'
[18:59] <nessita> is not mandatory but is very good to add that, to give the release info for each tarball. Is basically the same as the changelog notes you add in the packaging branch
[19:00] <dobey> weird
[19:00] <dobey> the other releases don't realy have a full ChangeLog entry
[19:00] <dobey> they just have one commit
[19:00] <dobey> *shrug*
[19:01] <nessita> dobey: the releases with one commit entry are releases with only one change
[19:01] <nessita> see for example 0.8.0
[19:20] <mongy> dobey, sorry, back... I tried restarting it and didnt work, and it gave up, I just noticed it was disconnected without trying again so I hit connect and its ok now..
[19:20] <dobey> ok
[21:47] <karni> I'll call it a day. Gotta triage that little bug next time I code.
[22:29] <__lucio__> karni, please ping me on monday :)
[22:29] <karni> __lucio__: sure