[02:23] $ sudo echo snoozzz | /proc/cpuinfo === rmcbride_ is now known as rmcbride [13:04] Hi all [13:08] facundobatista: Good morning to you! [13:08] jblount, :) [13:20] hello world [14:52] happy wednesday [14:52] It is! [14:54] so, following up on #387772 [14:55] the initial bug report said he couldn't access his "Shared with Me" files. Unless he deleted them locally and that synced upstream, he should still be able to get to them, yeah? [14:58] urbanape: even then, if he didn't have write access granted to those files, they should just get re-downloaded i think (unless there's a bug where they don't), and would certainly see them in the web ui [14:59] I'll follow up with him in email and get some more info [15:01] MEETING BEGINS [15:02] Hi, everyone who is here for the ubuntu one developer standup, please say 'me'. The format for this meeting is to paste 3 lines: TODO, DONE, and BLOCKED. Then say the person that should go next - we can try going in 'me' order. [15:02] me [15:02] me [15:03] dobey, statik, CardinalFang, rodrigo_: ping [15:03] jblount: pong [15:03] me [15:03] me [15:04] Better to get started, and let folks jump in when they get a chance... [15:04] DONE: Landed the file storage stats branch (and helped fix a rollout related bug with mthaddon) [15:04] sorry [15:04] TODO: Pair on /files/new with urbanape, small branch with a few minorish text and css fixes, asking about about making squid rollout proof [15:04] me [15:04] BLOCKED: Nope. [15:04] * jblount passes to urbanape [15:04] DONE: Face duty, blueprint, specs, and skeletal firefox extension for ubuntuone-firefox-bookmark-sync, handoff-meeting with jblount and pfibiger on /files/new UI. [15:04] TODO: Flesh out FF extension, get blueprint into review state, pair with jblount on /files/new UI [15:04] BLOCK: Nada [15:04] rodrigo_: your turn [15:05] done: added missing blueprints for couchdb-in-the-desktop stuff, some evo-couchdb fixes and working on packaging them [15:05] todo: face duty this afternoon (swapped with aquarius), more packaging in my PPA [15:05] blocked: none [15:05] CardinalFang: yours [15:05] First is statik's message, since he's not here. === CardinalFang is now known as statik-puppet [15:05] DONE: python-testtools passed REVU and has been uploaded. lots of phone calls, a couple of branch reviews, design review for firefox plugin with urbanape. [15:05] TODO: lots of phone calls and scrambling for Karmic Feature Definition Freeze today. Screensharing design review, work on OOPs with pfibiger, look at glib-couchdb with rodrigo. === statik-puppet is now known as CardinalFang [15:05] BLOCKED: None [15:05] DONE: reviewed branches. more work on mail queueing -- just need tests now, and to figure out config file for zope.sendmail . [15:05] TODO: get tests to run, damnit. [15:05] BLOCKED: None. [15:05] dobey, you! [15:06] DONE: Lots of client build system hacking (almost done) [15:06] TODO: Finish build system and package changes, move share creation via web api code to syncdaemon, split ubuntuone-client packaging, backport python-oauth [15:06] BLCK: None. [15:06] nobody else? [15:06] dobey: I think that's it. [15:06] END OF MEETING [15:07] so statik is not around? [15:07] statik is in a call [15:10] ok [15:16] Hmm. pidgin is back to hardlocking my system. I wonder why it hates me. [15:16] * jblount hugs empathy [15:17] rodrigo_: you're face just for the afternoon? [15:18] urbanape: yeah, aquarius swapped it this morning with me [15:18] he is sprinting, so really needed someone to take it over [15:18] so we are just swapping half day :) [15:18] k, I cc'd my notes (sparse as they were) to the discussion list [15:19] urbanape: yeah, reading them now [15:21] ok, now #2 task: change the topic, who can do it? [15:24] rodrigo_: rmcbride can, but I think we decided it wasn't a huge deal. [15:25] ah ok [15:25] it's nice, if there are difficult questions, I can just pretend to not read them :) === yofel_ is now known as yofel [15:27] rodrigo_: ;) [15:28] rodrigo_: i would like to point out that everyone can see you now :) [15:28] ugh! :D [15:29] you have been logged :P [15:33] damn! [16:03] hmm, we use GNOME's proxy settings in the u1-client programs, right? [16:03] no [16:03] well, I guess we don't [16:03] right, ok [17:07] * jblount goes looking for food [17:11] so, yeah, too quiet today in here. It was a joke what I said before, you can ask questions :) [17:18] heh [17:25] ok, keep your questions for later then, I'll be back :) [17:28] How popular is the beta testing - is there a long queue for invitations, anyone knows? [17:31] JayMP: We do have a pretty large queue, I don't think we've published a specific numbe. [17:32] OK, it is a great idea. I use Dropbox for work and using several machines it is very useful! Can't wait for an "indigenous" Ubuntu solution.. [17:32] JayMP: :) [17:37] let's just say our queue is very large :) [17:51] rodrigo_: hey. you shouldn't set the bugs to confirmed unless you've reproduced them. generally bugs that are unassigned should stay in the "New" state [19:03] * statik code reviews https://code.edge.launchpad.net/~dobey/ubuntuone-client/nautilus-no-python/+merge/7565 and is excited about seeing source tarballs up soon [19:05] i wish launchpad required you to set the commit message when you proposed a branch for merging [19:09] pfibiger: you aboot? [19:10] i think he's a person, not a boot [19:11] Having never met him, I can neither confirm nor deny [19:12] The voice I spoke with yesterday could easily have been computer generated [19:12] we're poking around in the new files ui. [19:14] hey dobey, i'm having failures in https://code.edge.launchpad.net/~dobey/ubuntuone-client/nautilus-no-python/+merge/7565 I just wrote the details in the merge proposal. configure is unhappy with my versions of libsoup2.4-dev and nautilus-extension-dev [19:14] are you specifying certain versions of those libs, or do they get generated from somewhere else? I'm wondering if I've got a karmic system that has some updates but is missing others [19:15] statik: i think that's a bug in gnome 2.27 stuff, and thus karmic :) [19:15] also, is libsoup for parsing html/json? if so, is it still needed as a dep? [19:16] statik: there shouldn't be any @FOO@ text in any pkg-config files in an installed system. that should have been replaced at build time by whatever package (presumably nautilus or one of its deps) [19:16] libsoup is for HTTP [19:17] i can probably remove it, when i move the create share code over to syncdaemon, that is currently disabled in the C version of the extension [19:17] cool [19:17] that makes sense [19:17] but there is some starter code that still depends on it [19:18] i'll make another quick branch to remove libsoup, after i get this in :) [19:18] great. i'm trying to update my packages now to see if I can get it to build [19:25] urbanape: i am. what's up? [19:27] heya, jblount and I are just poking around the new files UI stuff. As we stumble through, we'll probably poke you with some questions [19:33] great, now inkscape is segfaulting during the build [19:34] statik: i'm going to poke on karmic as well [19:34] dobey, after updating my packages again (i had updated last night) i was able to get through configure. trying the rest of it now [19:34] statik: ok, cool [19:36] dobey, crazy question: when this new package is installed, will it remove the old .pyc file from the nautilus python extension directory? [19:36] for people who had older versions of the package installed? [19:36] i've been wondering whether something like that was the cause of use getting those bug reports about stuff crashing in code that we had deleted long ago [19:37] statik: i think the debian python magic will do that when the old package gets removed, before the new one gets installed [19:37] statik: ie, the pyc don't get left around now when you just uninstall, do they? [19:38] i hopes not [19:38] statik: though i don't know if it necessarily works the same with upgrades... hrmm, let me see [19:39] i'll see what happens as soon as soon as my karmic install finishes upgrading [19:41] dobey, thanks! say, do you get lots of warnings about mismatched quotes at line 209 in ../canonical/ubuntuone/storage/syncdaemon/fsm/fsm_parser.py (and many other files) when running make check? [19:42] yes [19:42] beautiful [19:42] i'm happy with this branch [19:43] it's from intltool-update, because perl doens't handle """ """ in python very well [19:43] there's a long standing bug on it [19:43] dobey, should I wait to mark it as approved until kenvandine reviews it? [19:43] i don't know if he has time to review it [19:44] dobey, i can in like 5 minutes :) [19:44] or 10 [19:44] :) [19:44] ok [19:45] kenvandine, you'll love this branch, it spits out tarballs so stuff can be packaged for foresight :) [19:45] woot [19:45] dobey, is there a similar branch for ubuntuone-storage-protocol? [19:47] statik: no, we're not porting anything to C in there (yet anyway) :) [19:48] i already fixed storage-protocol by adding the MANIFEST.in, so we can make tarballs with setup.py sdist there [19:49] dobey, that makes perfect sense. you know, i really should have remembered that considering that I was examining the tarballs coming from setup.py sdist in ubuntuone-storage-protocol yesterday [19:49] heh [19:49] it's been a busy 2.5 days :) [19:51] boas noites [19:54] dobey, sorry... few more minutes :/ [19:56] statik: dobey: once the 0.90.2 tarballs are in place I will change the package build scripts for the PPAs to use diff.tar.gz magic etc, now that I understand the process. [19:57] cool [19:57] rmcbride, cool. I think the package also needs to be uploaded to REVU, dunno whether you or dobey want to do that [19:58] then once it's accepted we'll have to figure out how to ask for new uploads [19:58] i was going to upload some 0.90.2 release packages to revu [19:58] perfect [19:58] statik: we'll want to put the tarballs up for REVU (dobey should probably do this) and if we do new uploads we'll do the diff.tar.gz stuff [19:58] i think the process is roughly the same for new uploads [19:58] statik: basically 0.90.2 will be the "orig.tar.gz" for that workflow [19:59] well new uploads into karmic we will roll new tarballs for [19:59] rmcbride: whenever we build new source tarballs we'll need new orig.tar.gz though, yes? [19:59] right [19:59] statik: right [20:00] dobey: yea the new tarballs would be generated similarly to how I did the upload for python-configglue, workflow wise. [20:02] ah, i don't know what you did there :) [20:04] dobey: I've got a handle on it. Once I figure out how to explain it I'll have a doc up somewhere. The fact that you added the bits for generating the source tarball makes the whole thing much easier to keep consistent [20:04] right [20:05] dobey: basically we're thinking along the same lines and my technical language ciruits are a bit fuzzed today so it probably sounds like I'm talking gibberish(er tahn usual) :) [20:05] it still needs some fixes (because as a rule, 'make distcheck' should not fail, but am using some hacks at the moment to integrate with distutils, which break that) [20:05] speaking of gibberish [20:06] * dobey goes to pop in Brazil for some entertainment while working [20:06] Yea caught that in the review submission. [20:09] hi there, i want to talk to you about ducts [20:11] dobey, ok... ready [20:11] * kenvandine looks [20:13] dobey, that's a big diff :) [20:13] * kenvandine is reviewing [20:14] yes, yes it is [20:18] hrmm, i wonder what new stuff Ian Holm is in [20:18] kenvandine, if you ignore the deletes its a lot easier to digest [20:18] dobey: so what do you have against gwibber? [20:19] BUGabundo: i don't have anything against gwibber [20:20] oh [20:20] outside of it being yet another app where i have to enter all the same information i entered in some other app [20:20] ahah [20:20] and the ui doesn't really fit my ideal workflow [20:20] but that first one will be fixed in due time [20:20] with new ducts [20:21] yeah [20:21] I know [20:21] I've been following the devel since the beginning [20:26] katherine helmond hasn't been in anything recently either [20:31] dobey: he was Bilbo [20:31] Polonius [20:31] i know he was bilbo [20:32] but that was in 2003 :) [20:32] well, you were talking about Brazil. [20:32] it's newer than that [20:32] well, i'm watching brazil now [20:32] the movie? [20:32] yes [20:32] lve that movie [20:32] indeed [20:33] tax time is non-stop brazil references in my apt [20:33] ah crap [20:33] hrmm [20:33] statik: make check passed for you? [20:34] dobey, it bombed on not finding ubuntuone-storage-protocol because my system is currently a pile of unassembled branches [20:35] dobey, i was happy with the contents of the tarball i got from make dist though [20:35] statik: haha, well, it bombed on me because i made a mistake :) [20:35] but i'll fix that and push real quick [20:35] bad reviewer [20:36] alright, just pushed my first mercurial changeset to bitbucket, fixing up the licensing for the spawning wsgi server [20:36] maybe tomorrow i can figure out git and help rodrigo with glib-couchdb [20:37] i'm an upstream gnome maintainer, and i still haven't figured out git [20:40] pfibiger: did you do the initial migration for the new files UI? I'm stumped on how individual record rows decide how to render their individual columns. Just based on the columnsDef? [20:42] dobey, git is really making my head hurt [20:42] kenvandine: anything in particular? [20:42] kenvandine: indeed. notice how i've moved stuff off gnome.org, and how i haven't made any 2.27.x releases of gnome-icon-theme or evolution-webcal yet :) [20:42] CardinalFang, btw ^ I just sent donovan the licensing patch for spawning, so as soon as he uploads a tarball with those changes to pypi I will be able to get a package up onto REVU, which will make it much easier for us to use spawning to run the web servers. [20:43] i hope the answers are on live.gnome.org, but i haven't had time to search for crap on there either [20:43] urbanape, just a pain to use compared to mercurial and bzr [20:43] urbanape: yeah, based on fileColumnDefs [20:43] urbanape, basically it is a bit harder to remember how to do stuff [20:43] statik: Roger. I pushed it back onto my queue. Someone triaged it as "later". [20:43] hg and bzr are just very intuitive... and git seems like it is designed by people that hate me :) [20:44] pfibiger: so, how does the checkbox column get disassociated with the label in the columnDefs? [20:44] dobey, did you see my comment on the merge request? [20:44] CardinalFang, oh right. that was the big ol' script removing all the bugs from beta2 when we switched to more granular milestones. it should probably go into w09 or the next one [20:45] kenvandine: meh. ok :) [20:47] kenvandine: I guess I was just getting used to it when I left Slate, but yeah, it seemed to capriciously rename defacto standard commands to be intentionally obtuse. [20:47] although I don't have bitkeeper experience, maybe it was designed to mimic it? [20:47] nah, git was designed to not look like bitkeeper. bk was actually pretty nice to use [20:48] git was designed (or rather, not designed so much) to not be like anything else [20:48] it succeeded [20:49] yeah [20:49] git is like a perfect example of -ESUCCESS [21:01] dobey, ImportError: No module named protocol.oauth [21:01] in make check [21:02] kenvandine: ls -ld canonical/ubuntuone/storage/protocol ? [21:02] notta [21:10] kenvandine: added some updates per your comments, went ahead and removed the libsoup dep, and fixed that :) [21:10] rodrigo_, btw, your specs look good to me [21:10] dobey, ok [21:12] ln: creating symbolic link `canonical/ubuntuone/storage/protocol': File exists [21:12] dobey, ^^ [21:12] perhaps that is something make clean should remove? [21:12] make clean does remove it [21:13] i get that on make check [21:13] rm -f canonical/ubuntuone/storage/protocol [21:13] indeed [21:13] and the rule to create it has an if to check that it doesn't exist [21:13] hmmm [21:13] if [ ! -e canonical/ubuntuone/storage/protocol ]; ... [21:13] blowing up here [21:14] weird [21:14] do you have conflicts or something? [21:15] nope [21:17] i just ran make check twice and it works fine [21:17] hrmm [21:19] :/ [21:20] make by itself fails too [21:20] same error [21:23] does that file actually exist? [21:26] yes [21:26] it is getting linked [21:26] no idea where though [21:27] if you remove it, and run make twice, does it still fail? [21:27] yes [21:27] make clean removes it too [21:28] and it still fails? [21:29] your filesystem is weird! [21:31] dobey, yup :) [21:31] i have to run though... kids swim meet [21:31] will look more when i get back [21:32] oh... it fails on make check... if i remove it and run make [21:32] it works [21:32] running make a second time fails [21:32] ok [21:32] that if statement isn't working :) [21:32] weird [21:32] * kenvandine runs [21:32] your filesystem is lying :) [21:33] you must be using xfs or something weird === pfibiger is now known as pfibiger_ === pfibiger_ is now known as pfibiger [21:56] sigh [22:07] hello all, is there any documentation or developer examples for the ubuntu one API, or will that appear closer to production release? [22:11] there are some examples in ubuntuone-storage-protocol, and there is the main client itself [22:13] dobey: from https://wiki.canonical.com/UbuntuOne/BugTriage, I understood confirmed is not only for bugs I can replicate [22:13] kenvandine: cool, have you approved them? [22:15] rodrigo_: "can confirm the bug" means "you see it too" and not "this makes sense that it would be a bug" [22:15] rodrigo_: perhaps that could use a little clarification though [22:15] dobey: yeah, because triaged is not a good default value [22:16] dobey: if we have to keep then as 'New', we should have that on the wiki page [22:18] default value? "new" is the default [22:18] dobey: thanks! === midknihtt is now known as midkniht