[00:26] verterok: packages for karmic are indeed broked. We hope to have a fix for tomorrow's release. Finishing up packaging for Karmic proper has taken priority. [00:26] rmcbride: thanks! great to know [00:26] also a good package for configglue should be in karmic universe soon if it is not already. Passed REVU last week [00:27] nice! [00:30] I'm out for the evening. Looking up at backscroll it looks like the good configglue package is recently published, but may not be in all indexes yet? anyhow have a good one all! [00:34] rmcbride: archive admins have to accept the new binary packages === afk is now known as mthaddon === mthaddon is now known as afk [08:48] top of the morning, ubuntuoneros! === afk is now known as mthaddon === thekorn is now known as _thekorn === _thekorn is now known as thekorn [09:41] are .conflict files created by U1? [09:42] I have a GNUCash folder in U1 and the main accounts file is now zero bytes long and there is a .conflict file which is a gzip which won't extract for some reason [09:53] I managed to recover most of it from the most recent .xac file, however a breakage on gnucash isn't nice :-( [10:03] alanbell: I'm sorry you are experiencing this problem [10:03] alanbell: have time to investigate a bit what's happened? [10:07] vds: sure [10:07] alanbell: have you looked at the ubuntuone log files already? [10:08] not yet, I was concentrating on recovery :-) [10:09] alanbell: of course :) [10:09] doing some grepping now [10:12] ok, first looking at the activity of the file I am interested in with grep "Bell Lord Ltd'" * [10:12] http://pastebin.com/f36e8f50d [10:12] now seeing when the .conflicts turn up with grep "Bell Lord Ltd.conf" * [10:13] http://pastebin.com/f6fa3a9cd [10:14] I have several machines which are mine, so it is in my files, plus I have shared it with another account. I don't think I had gnucash working on the file on multiple computers at the same time, but it is possible [10:14] * vds looking [10:17] I have been having a bit of bother with the U1 icon spinning forever, just working and idling and not achieving much [10:17] alanbell: do you have the latest ubuntuone-client installed? [10:18] I think so [10:19] alanbell: i have had the smae issues, especially when adding many files [10:19] it seemed to create no traffic [10:19] what it was doing was creating all files first, then slowly uploading them to "the cloud" after a LONG time [10:20] yes we are working to speed up the upload/download [10:20] vds: yes, it is the latest version, there were some other updates waiting for me in landscape, but not U1client [10:21] alanbell: ok, thanks [10:22] Chipaca: ping [10:22] I tend to keep things uptodate and I am doing a trial of landscape which is quite fun [10:22] vds: i was waiting for chad to return, but i may as well just state it now [10:23] a nice feature would be an actual indication of what the daemon is doing, i.e. a queue indicator or something [10:23] when you hover over the symbol, ya know [10:24] danage: you mean something to get a feedback about what the syncdaemon is doing? that would be nice! [10:25] yeah like that. appears like it wouldn't be too difficult to implement [10:26] danage: right [10:48] alanbell: I'll make sure some of the sync daemon will look at the logs, in the meanwhile would it be possible to check whether more than one instance of gnucash had that file open? [10:50] vds: looking at the dates it is highly likely that only one computer had that file open in GNUcash [10:50] i also think opening the ubuntuone folder with a left-click is unintuitive. to me it would make more sense having it connect/disconnect that way. usually, i open folders with the "places" menu item gnome has [10:52] danage: agreed I think we'll get rid of the applet as it is now [10:52] oh really? [10:52] what's the plan then? [10:53] danage: not sure yet [10:53] not to me at least [10:53] we can ask dobey [10:54] as soon as he gets online [11:01] vds: found the relevant couchdb commit: http://github.com/halorgium/couchdb/commit/b1bfb1481e54474ce302e134a614134cc7f5fcb0 [11:02] vds: so it's doc._local_seq in views [11:07] since the vote for 0.9.1 has just started, this seems likely to land soon [11:07] vds: so we'll need to see if it's too late to get 0.9.1 in Karmic [11:11] Happy birthday Alan Turing! [11:12] thisfred, is he back from the dead? [11:13] teknico: unfortunately no, but we can still party like it's 1949 [11:15] thisfred, did Big Brother start just last year? it seems like it's been going on for ages... [11:30] Hello [11:33] thisfred: should not be too late if we manage to package it soon, but [11:33] thisfred: has it been released yet? [11:33] vds: no, the vote has just started [11:33] FLOZz: hi [11:34] thisfred: can we vote? :) [11:34] FLOZz: hi [11:34] vds, we can, but our votes won't count :) [11:34] I suspected it :) [11:34] vds: anyway, I have seen only +1s, so it's all good [11:35] what I could do maybe tomorrow is to try to repackage it so at least we are ready when it is releasd [11:35] I have a little problem with ubuntu one [11:35] or maybe we can alert kenvandine [11:35] FLOZz: what's the problem [11:35] ? [11:35] :) [11:35] I have created a sub-folder with the web interface... and I can't suppress it [11:37] When I click on [Delete], nothing happen ^^' [11:37] vds, that sounds like a good idea. Also we'll need to start experimenting with it soon, to see what we can and can't make work with the _changes feeds because the update triggers may be gone in that release. Not sure, [11:37] doing more research [11:37] and if I delete it in nautilus, it's the same problem [11:37] FLOZz: delete on the web interface? [11:38] yes [11:38] and also deleting it from nautilus doesn't work [11:38] FLOZz: have you already checked the logs? [11:38] no...where are the logs ? [11:39] FLOZz: .ubuntuone_log/ [11:39] ok thx [11:41] I haven't any .ubuntuone_log directory in my home folder ^^' [11:43] I have found this : ~/.cache/ubuntuone [11:44] FLOZz: right [11:44] I had those logs from a previous version [11:53] So.... [11:53] When I suppress it in nautilus, it come back : [11:53] http://nopaste.com/p/abQOsXiRq [11:53] :s [12:01] * vds looking FLOZz log [12:02] FLOZz: you shared that folder with someone else? [12:02] I have try to share it but he don't accept... so i have remove the share [12:03] and remove the directory after [12:03] ... but i can't remove it [12:03] FLOZz: ah interesting [12:03] FLOZz: let me do some tests and I'll be back at you [12:03] thanks so far [12:04] ^^ [12:50] Hi all [12:51] hi facundobatista [12:57] hi facundobatista [12:58] facundobatista: it looks like once you have shared a folder you cannot delete it anymore even if you remove all the shares [12:59] vds, you're talking desktop or web? what actions you're performing? [12:59] facundobatista: both [13:00] facundobatista: go to a shared folder, try to delete, nothing happens, remove all the shares, try to delete again, nothing happens [13:00] facundobatista: both on web ui and nautilus [13:01] facundobatista: FLOZz has this problem, I've tried myself and I can replicate it [13:01] ok, let me try it [13:01] thanks a lot facundobatista! [13:05] vds, I'm having problems right now with *sharing* a folder... the web ui never comes back [13:05] facundobatista: I had this problem too, after a while it works again, maybe we should take a look at the load? [13:14] vds, it never came back, :( [13:15] facundobatista: uhm [13:15] interesting [13:15] facundobatista: have you tried with nautilus? [13:21] vds, who does the share-through-nautilus work? I right click on a "for_vds" share, put your mail address, click on "share", but in the syncdaemon logs I don't see anything [13:21] dobey, ^? [13:22] facundobatista: dobey might help here [13:22] facundobatista: I just received you share proposal [13:22] facundobatista: should I accept it? [13:24] vds, for *which* share? [13:24] facundobatista: the email doesn't say which share, it says just that you are sharing something [13:25] ah [13:25] sorry [13:25] facundobatista: test [13:25] vds, I did that, wasn't sure that about the logs, also tried with "for_vds" [13:25] please wait for that [13:26] facundobatista: sure [13:26] vds, "for_vds" is easier to search in all the logs, "test" is not good as a searchable keyword, :p [13:29] :) [13:49] vds, did you receive this last share? [13:49] facundobatista: not yey [13:49] I'm checking the mail very often [13:49] nothing so far [13:50] vds: also check websense, maybe, since it caught your share to me at some point [13:51] facundobatista: at the moment, using the desktop client, we can share only one folder at time, but in the future we plan to be able to share more than just one folder right? [13:51] thisfred: right! thanks! [13:51] vds, afaik, we can have several shares... [13:52] facundobatista: sorry I didn't state it correctly [13:53] facundobatista: we can share only things inside the Ubuntu One folder right now [13:53] vds, yes [13:54] facundobatista: we'd like to be able to sync more folder than just Ubuntu One in the future, right? [13:56] a reiteration of some comments i made earlier: 1. a nice feature would be an actual indication of what the daemon is doing, i.e. a queue indicator or something [13:56] also, opening the ubuntuone folder with a left-click is unintuitive. to me it would make more sense having it connect/disconnect that way. usually, i open folders with the "places" menu item gnome has [13:57] danage: right [13:58] danage: would you please file a bug for each one of the two features? [13:59] sure [13:59] danage: thanks [14:03] should i make it wishlist? [14:10] hm i'm reading similar bug reports [14:11] i think adding it to network manager could make sense [14:11] although it's not really a network connection, it's more like a network share [14:15] hello world [14:15] hi statik [14:16] hey jblount, you know about blueman for bluetooth tethering? [14:16] facundobatista: *currently* there is no way to share folders via nautilus (that code was disabled for the moment, during the port to C) [14:17] danage: FUSA is undergoing a complete redesign at the moment and the sketches I saw at UDS had it including ways to be connected/disconnected to various online services, continuing with the idea that is in FUSA currently where you can set your instant messenger connection state from FUSA [14:17] facundobatista: once i get the code that handled doing the sharing via the http api, into syncdaemon, it will log some stuff, and be re-enabled in nautilus [14:17] so the current plan is to put connect/disconnect into fusa rather than network manager [14:17] cool, i just read through the whole bug report and found some of that too. sorry for the redundancy [14:18] no worries [14:18] makes waaaay sense all that === urbanape_ is now known as urbanape [14:18] it's working nicely for me now too. so if the issues in the bug reports are being taken seriously, i think this will be a really really nice service when it's finished [14:20] thanks! we are definitely taking the bug reports seriously. we're working on things that aren't very user-visible at the moment, like packaging, porting, and some security fixes [14:23] i could see myself paying for this [14:24] vds, so, we fall back to the situation where it didn't work through the web? [14:25] facundobatista: it does work for me [14:25] vds, but you can not remove it? [14:25] facundobatista: exactly, I can share, I can remove the shares, but I cannot delete the folder once I've shared it [14:26] vds, ok.. I tried to share a folder called "foo" to you, and it didn't work [14:27] vds, you received an invitation for the "test" directory, which is strange because dobey says he disabled the code [14:27] (I shared "test" from the desktop, using Nautilus) [14:28] facundobatista: you're using old code [14:28] dobey, I'm not using trunk, and have the system fully updated [14:29] difference between beta and nightlies maybe? [14:29] mmm... I update it yesterday, I'm seeing new packages now [14:29] could be [14:29] the new stuff isn't in beta yet [14:29] and even so, updating the packages doesn't restart nautilus [14:29] so the old code could still be running in memory [14:29] (hence the need for changeup) [14:30] dobey, I had to reboot yesterday, kernel changes [14:31] yeah, i had to rebuild psb-kernel-source because of that :-/ [14:33] vds: the is a bug in the model API, a branch that fix that is waiting review [14:42] verterok: what's the bug exactly? [14:42] vds: I don't know if there is a bug [14:48] FLOZz: https://launchpad.net/bugs/383619 [14:48] Ubuntu bug 383619 in ubuntuone-client "Cannot delete a folder from the web interface (folder is shared)" [High,Triaged] [14:49] FLOZz: someone is already taking care of this bug [14:49] vds, ok [14:50] Thanks you vds ^^ [14:51] FLOZz: np [15:00] MEETING BEGINS [15:00] OH HAI HACKERZ! [15:00] Welcome to the Ubuntu One Developers meeting. Please say "me" if you are here to attend! [15:00] me [15:00] me [15:01] me [15:01] me [15:01] me [15:01] me [15:01] aquarius, vds, teknico? [15:01] me [15:01] me [15:01] me [15:02] DONE: Got those two branches up (need a few reviews on the last one) [15:02] TODO: Fix bugs in new /files/ inteface css stuff [15:02] BLOCKED: Nossir! [15:02] rodrigo_: Your turn! [15:02] done: more evo-couchdb work, almost complete, just a few backend methods missing implementation [15:02] todo: finish packaging of all the new couchdb-desktop packages [15:02] blocked: none [15:03] dobey: your turn [15:03] DONE: fix packages for REVU feedback, resolved versioning confusion [15:03] TODO: fix karmic missing icons issue, finish setting up source pkg branches, finish moving share creation code, research and implement ChangeUp API [15:03] BLCK: None. [15:03] urbanape: your roll [15:03] DONE: (well, did) some hacking on Firefox extension. [15:03] TODO: I'm on-call reviewer, I believe (I took over last week when someone was sprinting?). Also tests, and I need to make *some* progress on new files UI today. [15:03] BLOCK: None, just me, really. [15:03] statik, you're up. [15:03] DONE: Landed 4 branches, reviewed 2 branches. Fixed python-testtools after archive admin spotted a problem and resubmitted. Worked on REVU feedback for python-simplesettings, python-mozrunner, and python-spawning. Schemed with jamesh and mwhudson regarding upgrading to bzr 2a branch formats. Wrote a super tiny couchdb patch for aquarius. [15:03] TODO: keep working on packaging windmill and spawning, scheme with cardinalfang regarding new WSGI servers, work on OOPs with pfibiger, figure out whether south could be made to work with storm. [15:03] BLOCKED: Nope. [15:04] CardinalFang, your go [15:04] DONE: face. some smarter timeouts in mail tests. [15:04] TODO: finish mail work. commit today. [15:04] BLOCKED: nyet! [15:04] aquarius: poke. [15:04] DONE: working on starting up a local CouchDB, talking to Till Adam @ kdab about their progress on Akonadi CouchDB back end (they will have screenshots today, hopefully!); argue about naming of Desktop Couch project [15:04] TODO: continue working on it; work out the best way to find the port of a running Couch (trying a statik patch right now); refactor D-Bus port exporter so it starts Couch rather than just inspecting it [15:04] BLOCKED: none [15:04] vds: hit me, big boy [15:04] DONE:some research about couchdb _local docs and docs seqno for the new timestamps implementation, discussed about that with thisfred, still some research and some discussion is needed, face duty [15:04] TODO: face duty and maybe some skype with eric [15:04] your turn teknico [15:04] DONE: better defined the requirements and implementation of the contacts db reset thing with markgsaye, thisfred and vds [15:04] TODO: completing the implementation of the contacts reset db thing [15:04] BLOCKED: nothing [15:04] NEXT: noone [15:05] end of line [15:05] and nobody is blocked, amazing! [15:05] also, I have to go to the dentist in an hour. Not looking forward to that. [15:05] END OF MEETING [15:05] aquarius: ugh, dentist :( [15:05] jblount: thanks for running it! [15:06] rodrigo_: I am seriously debating pulling all of my teeth out with pliers and just eating soup for the next forty years, just so I don't have to go [15:06] statik, what's the rotation of running the standup, if any? [15:06] statik: Sure thing! It provides lots of opportunities for exclamation points :) [15:06] teknico: If it's 10, and you notice, you should start it (this is my personal rule, have yet to be told otherwise) [15:06] s/10/10EST [15:07] teknico: someone runs it at 1400utc [15:07] statik, oh, I get it, first come first serving ;-) [15:07] aquarius, you wouldn't have any problems with beer either ;-) [15:08] aquarius: false denture!! [15:08] aquarius: so you can eat other stuff than soups :) [15:08] is it safe? [15:09] :) [15:09] thisfred, aquarius: how about *gasp* couchdb-desktop ? [15:09] dobey: that was aquarius' suggestion I believe [15:09] dobey: this is why I think it should be Desktop CouchDB [15:10] couchdb-freedesktop [15:10] couchdb-zedd [15:11] but I don't know how the trademark situation is, if any (asked on #couchdb already, no conclusive answer) and whether we'd be presuming too much in the eyes of the couchdb community [15:11] dobey, urbanape, rmcbride: you guys are each running trunk branches on ubuntuone sub-projects. jamesh is going to coordinate upgrading the main ubunet branch to bzr 2a format next week while i'm gone, can you guys take care of upgrading your projects? [15:11] rodrigo_: it isn't a freedesktop thing, though, not yet [15:12] statik: sure [15:12] statik: yup [15:12] dobey, urbanape, rmcbride: i just realized you probably need to coordinate with jamesh in order to get the stacked branches taken care of [15:13] is there going to be a freeze or anything on the current trunk before he upgrades it? If not, why not just pull new branches after he's done? [15:13] or am I missing something? [15:13] (quite likely) [15:14] are we going to back up trunk somewhere? [15:15] ie, what happens if the upgrade eats our babies? [15:16] dobey: then we tell Jonathan Swift to give them back? [15:16] heh [15:17] ask it if it wants some A1 with it? [15:18] yeah, i believe the general approach is to tell everyone to stop pushing branches, make a copy of trunk in the old format, do sysadmin magic to retarget all the stacked branches to the copies of trunk, upgrade the real trunk, unfreeze, and have everyone reconcile and convert their existing unlanded branches into a 2a format repository [15:19] obviously you can do test upgrades to the 2a format on the side to ensure that the upgrade will go smoothly before any of this happens [15:21] i'm not exactly sure how to upgrade the branch on the server itself [15:23] bzr upgrade --2a lp:~foo/bar/baz usually works [15:23] if an upgrade had been done in the past then you use lp:hitchhiker to go in and delete the old bzr.backup directory first [15:23] but stacked branches are what make this complicated [15:24] which is what I was sorta asking. [15:24] cause you don't want to upgrade trunk out from underneath branches that are stacked on it [15:24] how many branches are stacked on it now? (how many in play, really) [15:24] i think launchpad automatically stacks all branches on trunk [15:25] then, can we not jut let the current in-play branches quiesce before upgrading, and then have everyone branch off the upgraded trunk? [15:25] urbanape: yes but we should still fix up all the old stacked branches - we don't normally delete them [15:25] statik: i think it's stacked on whatever you branched from... which is normally trunk [15:25] statik: ah, okay [15:26] it's stacked on the default branch for the default series i think, bzr doesn't really keep track of where you branched from [15:26] urbanape: for yours it should be super easy, since you have control of every single branch [15:28] ah crap [15:28] bzr 1.16 broke bzr-gtk [15:31] bye [15:36] statik: 1:1 call now? [15:37] rodrigo_1, yep, I'll call you on skype [15:37] ok === rodrigo_1 is now known as rodrigo_ [15:57] Has everyone seen the Tomboy integration in Gwibber trunk? DONE: Got those two branches up (need a few reviews on the last one) [15:57] TODO: Fix bugs in new /files/ inteface css stuff [15:57] grr [15:58] haha [15:58] Just call me "Paste-Mc-FAIL pants" [15:58] http://etc.joshuablount.com/save-to-tomboy.ogv is the proper thing that I wanted to paste [15:58] you are pretty pasty, for living in florida and all [15:58] paste "baby-octopus devourer" mc-fail pants, surely [15:59] thisfred: I just threw up a little in my mouth. [15:59] hehe [16:07] jblount: did it taste like baby octopus? === aquarius is now known as aquarius_dentist === yofel_ is now known as yofel [16:45] Hi, I've got an email "You have been invited to Ubuntu One" but it didn't contain an invitation code and ubuntuone.com says I'm still on the waiting list. [16:47] jdobrien: ^^ [16:47] (Also: Hi msch! I think jdobrien was just looking at something like this) [16:48] well, good to know. i just wanted to give you a heads up. === aquarius_dentist is now known as aquarius [17:21] hmm...he left what the heck [17:33] just thinking out loud, is there any overlap between U1 and the OLPC sugar.org sharing of files and objects? [17:34] which is mostly jabber/XMPP based [17:34] it would be rather cool to have an Ubuntu XMPP infrastructure in the cloud [17:43] judging by the number of new bug reports for the "without an invitation code"... something is broken [17:44] jdobrien: it might be worth stopping the spam-o-matic, until it can be fixed [17:45] sladen: are people getting multiple emails? [17:45] sladen: it was something that happened in last nights set of invitation approvals...there was an exception which sent out the emails, but didn't save the approval [17:46] sladen: the good thing is...it's not automatic [17:46] jdobrien: I don't multiple emails, but multiple people are noting on the bug tracker that they receieved one [17:47] sladen: yes...we know of the problem [17:47] jdobrien: four of the last five emails in my launchpad-bugmail folder of four separate people all confirming the same issue [17:48] sladen: I understand that...it was all one batch...and I just fixed the record that caused the problem...it was a problem we were aware of [17:49] the fix is still in the pipeline [17:52] jblount: ping [18:14] urbanape: yo! [18:14] urbanape: Sorry, just back from lunch. [18:15] heya. I've installed fix-da-buttons locally and am not seeing the buttons at all. [18:15] urbanape: this is a problem, I think I probably missed a bzr add somewhere. [18:17] urbanape: Yeah, pushing now. Sorry about that! If you pull inside the branch it should grab those changes. [18:17] coolio [18:18] yay, now that's a branch I can approve [18:19] urbanape: heh [18:20] huh. Wonder why it doesn't go to "approved" after two approvals. Still in needs review. [18:20] urbanape: There isn't a way in LP (yet) to say "my project requires to approvals, then it's approved to merge" [18:20] gotcha [18:21] Hi everyone. I require a little help. I got invited to ubuntuone, and installed the client, but the "control applet" does not want to start. Any clues? === mthaddon is now known as afk [18:23] mrtoftrash: After installing the app, can you start it from Applications > Internet > Ubuntu One? [18:24] jblount: Nope. No error messages either when I type "ubuntuone-client-applet" in a terminal. [18:24] Oh wait... [18:24] jblount: For some reason, it worked this time. IRC always does the trick :) [18:25] mrtoftrash: :D [18:25] mrtoftrash: We've seen other people mention it not launching, you are not alone. It's still undiagnosed AFAIK [18:26] jblount: Very strange. [18:26] mrtoftrash: Agreed. Particularly because I can't repeat it inside of a clean virtual machine. [18:27] jblount: yeap, works fine now. Thanks. Ciao [18:30] * jblount notices his use of "to" instead of "two" and hides his face in shame. [18:35] its oh kay. Eye reed eye arr sea out lowd too maiself, and it sownded phine. [18:56] urbanape: the review policy stuff in LP to automatically mark a proposal approved after two positive reviews should be showing up in the next launchpad releas [18:56] gotcha [18:56] hey, statik [18:56] statik: i suspect that means "next month"? [18:56] yeah [18:57] cool [18:57] I'm putting in some tests for the media revno url injection in the test_index module. [18:57] sladen, btw, regarding the email thing, CardinalFang is putting the final touch on a transactional email sending branch so that if an exception happens when processing a batch of invitations then no emails are sent, which should prevent this kind of thing from happening in the future [18:58] urbanape: great! that wouldn't have caught the edge config problem, but it's a great test to have [18:58] I'm not sure how the browser URLs are meant to match up to reality, though [18:58] urbanape: foo/media/1234/hugs.css i think [18:59] alanbell, yeah we're going to be doing a lot with XMPP I think [18:59] alanbell, i was talking to robot101 about the OLPC stuff last week - i don't think we'll be doing things the way they did, but definitely some shared ideas [19:01] * dobey wonders wtf is up with his car [19:07] hmm. Using the browser makes it harder to test the straight HTTP status codes returned by a valid and invalid media URL... So far, the best I've been able to come up with is sniffing the browser.headers or browser.contents. [19:27] urbanape: it doesn't raise an exception when it 404s? [19:27] it being zope.testbrowser [19:27] it does, but there's no convenient way to trap a 200 in the successful case. [19:28] easier to test the failing case [19:28] if (!fail): success [19:29] raise Exception("Success") ? :) [19:31] def test_dobey_is_actually_rube_goldberg(self):... [19:33] those up and down arrows in nautilus are known to be broken? [19:37] broken? [19:38] Bom S. João para tds, e cuidado com as espinhas :)) [19:39] no habla [19:40] danage: elaborate please [19:40] it shows green download arrows even though the wheel isn't spinning and all files have been downloaded [19:41] danage: on folders, or on files? [19:41] only folders, files seem to be ok [19:42] ok, yeah it's a known issue that they stick around on folders... i have no idea why they do though :-/ [19:45] dobey: Do you know if "labels sticking on folderz" has a bug? [19:45] it does [19:46] but i don't recall the number [19:46] jblount: i think you filed it though [19:46] dobey: RESEARCH! brb [19:46] (meaning, I'll go look for it) [19:47] #374890 [19:47] bug 37489- [19:47] Launchpad bug 37489 in ubuntu-artwork "newly added user doesn't see default theme" [Medium,Fix released] https://launchpad.net/bugs/37489 [19:47] bug 374890 [19:47] Bug 374890 on http://launchpad.net/bugs/374890 is private [19:48] Today is one big typo [19:48] lol [20:35] verterok: ping, are you back? :) [20:35] dobey: yeap [20:36] oh, pong :) [20:36] verterok: i needs-fixing/commented on your client branch [20:36] verterok: and i'm poking in syncdaemon, and had a couple questions for you [20:37] verterok: as i suspect, i shouldn't be calling a syncdaemon dbus method from within syncdaemon, to get the metadata for a path, huh? [20:37] verterok: or i suppose it's ok, if i call it from a Thread()? [20:38] dobey: here I get all the files I added to .bzrignore as unknown [20:38] verterok: really? that's odd [20:38] yes [20:38] verterok: what version of bzr are you using? [20:39] dobey: https://pastebin.canonical.com/18899/ [20:39] dobey: 1.16rc1 [20:40] weird [20:40] dobey: and this is the output of bzr ignored: https://pastebin.canonical.com/18900/ [20:40] dobey: I'd use the twisted thread pool rather than spinning up your own Thread()s [20:40] verterok: i wonder if there is a bug in that [20:41] dobey: in my desktop with bzr 1.16 (final) I get the same result [20:41] tcole: i think this might actually have to be separate, as it's using urllib2 [20:41] verterok: hrmm.. that's very odd indeed [20:41] dobey: maybe you have some global ignores defined [20:42] dobey: hm, howso? [20:42] dobey: as tcole said, it's better to use twisted thread pool, possibly using deferToThread [20:43] verterok: ah, i do have a ~/bazaar/ignore [20:43] :) [20:43] verterok: but i certainly didn't create it [20:43] i wonder where it came from [20:43] also I do have to ask why you're calling a method on the dbus interface from within syncdaemon itself [20:43] that's weird [20:43] it would seem that there should be a less roundabout way to get at what you need [20:44] tcole: well, i'm not. i'm moving code that was in the nautilus extension, to syncdaemon, and wondering if there is a better way [20:45] ah [20:45] verterok: ok. but we should use globs instead of listing individual files, i think :) [20:45] broadly, yes [20:45] verterok: and we don't need to list full paths in the .bzrignore either === BUGabundo1 is now known as BUGabundo [20:52] verterok: hrmm. how should i deferToThread() or callFromThread() inside the VolumeManager.create_share()? [20:59] * verterok_ stabs his ISP [20:59] dobey: sorry, bad day for my ISP [20:59] * dobey lends his stabbin' knife to verterok_ [20:59] verterok: hrmm. how should i deferToThread() or callFromThread() inside the VolumeManager.create_share()? [20:59] dobey: https://pastebin.canonical.com/18901/ [21:00] :) [21:01] well, and you will probably need to do something with the returned deferred [21:01] incidentally, we should use public pastebins out here :) [21:01] pfibiger: you have one review on https://code.edge.launchpad.net/~pfibiger/ubuntuone-storage-protocol/add-ssl-context/+merge/7817 , feel free to ignore my suggestion in the interest of getting this code landed though [21:01] tcole: right... [21:02] verterok: pastebin.ubuntu.com is public [21:02] thanks! [21:02] tcole: thanks for pointing out :) [21:02] tcole: hrmm. but if i jsut ignore it, it should be fine too, yes? [21:03] dobey: http://pastebin.ubuntu.com/202394/ [21:03] dobey: we should handle at least the error case, to give some feedback to the user maybe? [21:04] bye bye verterok! === verterok_ is now known as verterok [21:05] dobey: it depends [21:05] dobey: if you don't care whether or when it succeeds or fails at all, ever, then you can just ignore it [21:05] the *or when* is important too [21:05] if there are things you would like to happen after but not before it [21:06] no, it's just that i need to shove the code in a thread [21:06] it handles errors within the call [21:07] so it's just a background task then? [21:07] is it okay to have more than one of these calls going at once? [21:08] i suppose if you can make use of the nautilus ui that fast, yes [21:08] it just posts an oauth-signed request to a URI [21:23] i wonder if there's a sane way to test this code [21:24] dobey: it's about the share http request?...a mock http server? [21:26] verterok: fwiw. python-configglue is new'd [21:26] Nafallo: it's working now? great!! [21:27] verterok: need to wait for the mirrors to get synced :-) [21:27] :) [21:27] verterok: yeah, i don't want to write a mock http server for that :) [21:28] dobey: maybe you don't have to write it [21:34] Hello [21:34] dobey: what do you need to test [21:34] SiDi: hi! [21:34] Im using xubuntu and the ubuntuone panel doesnt appear when i launch it :| [21:35] dobey: do you know if the applet works on xubuntu? ^ [21:35] it does so long as the deps are installed, and none of the bugs are being hit (particularly the karmic one) [21:36] verterok: need to test creating a share via the http api [21:36] dobey: it doesnt look like it'll work here [21:37] SiDi: what doesn't work exactly? [21:37] dobey: the applet doesnt popup [21:38] no error message when launching from console [21:38] SiDi: what happens? it waits 30 seconds and exits? [21:39] telling you in 30 secs [21:40] i usually kill after 5 :p [21:43] it doesnt quit [21:46] it keeps running after 30 seconds? [21:47] and there's no new icon at all in your panel? [21:47] verterok: where is deferToThread() defined? [21:48] twisted.internet? [21:48] dobey: no, it's in reactor, but the glib2reactor don't have one...so you should use the one we used in the old storagefs [21:54] dobey: from twisted.internet import threads [21:54] dobey: threads.deferToThread [21:54] ok [22:08] dobey: no new icon, keeps running [22:08] jaunty amd64 [22:10] SiDi: are you on karmic? [22:10] dobey: nopee [22:11] verterok: http://pastebin.ubuntu.com/202434/ <- does that look bad to you? [22:12] * verterok looks [22:12] CardinalFang: to set review status by email you need to gpg sign your message, and indent the review voting command by one space. something like: [22:12] review: approve [22:12] no colon though, i think [22:13] dobey: if you already have the mdobj in the metadata it should have a path, there is no need to set it again [22:13] although easier to remember is just " +1" [22:13] verterok: i'm not setting it again. pylint was complaining "unused variable: mdobj" and clearly enough, it's unused [22:14] dobey: just remove the assignment [22:14] verterok: but with the change i pastebinned there, i am getting test failures with odd KeyErrors [22:14] verterok: that's what that patch does! [22:14] verterok: or you mean remove the set_by_path() also? [22:14] dobey: no, the patch removes the except! [22:14] dobey: I misread the patch, sorry [22:14] verterok: the patch removes the try and the except :) [22:14] dobey: you can't remove the except clause [22:15] dobey: btw, the ubuntuone-client pylintrc should be updated [22:15] verterok: why? the mdobj isn't actually used [22:15] updated for what? [22:15] dobey: to be like the other ones we use [22:15] dobey: in the except the mdobj is created! [22:16] but only if isn't there [22:16] it is the exact same one as in ubuntuone-storage-protocol [22:16] verterok: oh, so i should just remove the lvalue [22:16] ? [22:16] dobey: yes [22:17] ok [22:17] thanks [22:18] np [22:18] dobey: pylint is showing warnings for a lot of things we ignore in other pylintrc [22:18] 's [22:18] eh? [22:37] alright, uploads to REVU done for python-mozrunner, python-eventlet, and python-spawning. talk to you all tomorrow! [22:38] aren't you on vacation tomorrow? [22:38] (but good work!) [22:41] SiDi: hrmm. all i can suggest for the moment is to file a bug and attach the logs with "apport-collect -p ubuntuone-client $bugnumber" [22:41] tcole: thursday [22:41] oh [22:42] * tcole has difficulty with this "calendar" concept [22:44] verterok: ubuntuone works again in karmic. still hates pretty little pictures of animated clouds, but syncs and such more important stuff :-) [22:46] Nafallo: great news! [22:47] dobey: willdo [23:44] later [23:46] dobey: will do the apport part tomorrow though. Can't now, got an update running and a damn low bandwidth, so it'll take time