[06:29] I'll just leave this question here for tomorrow: does anyone test the sso client on windows from the command line? I made changes to get_activation_cmdline to use the buildout python to run subprocesses during development, and was looking to see if similar changes were necessary for windows, but it looks like it'll always look in the registry for the path to the subprocess, which means that I can't run eg. ubuntu-sso-login-qt from [06:29] command line on windows. Am I missing something? === mthaddon` is now known as mthaddon [08:53] Happy Friday, folks! :-D [08:59] yay Friday \o/ [11:16] good morning! [12:13] holas! [12:17] alecu, buenas [12:22] ralsina: do you happen to know who is our contact in Tomboy project? [12:26] aloha [12:33] rye: I think it is Sandy Armstrong [12:50] rye: sandy is the only person i know that really works on it; though not sure how much anyone hacks on it any more [12:51] czajkowski: can't say aloha. it's too early for me to drink my lovely hawaiian beer [12:51] dobey: never too early! [12:52] dobey: oh but as you're alive! how do I snyc my tomboy notes up now ? [12:52] czajkowski: what do you mean? tomboy should sync the notes just fine, the same way it always has [12:54] hmmm [12:54] rye: nope [12:55] rye: we have not had one in a long time, I think. As in "before I joined" [12:56] well, we never really had a specific "official" contact in tomboy afaik [13:31] dobey: just checked nope my notes don't go to U1 any more :S any idea what I need to do ? [13:33] czajkowski: i have no idea. check your configuration? run it in a terminal with "tomboy --debug" perhaps? [13:35] dobey: http://pastebin.ubuntu.com/1066018/ [13:36] c [13:37] that looks ok i guess. is the configuration correct? [13:38] ralsina, all: it seems my isp has left me marooned again today. [13:38] ralsina: is it working fine for you? [13:39] alecu: internet? Yes, but I am not in telecentro :-) [13:39] and just as I said so, my modem has upstream again. [13:42] czajkowski: rye is probably a better person to help with that though :) [13:50] czajkowski: reading [13:50] czajkowski: could you please try syncing and see whether that prints anything? [13:53] rye: dobey sorry was helping someone else. [13:53] rye: won't let me sync it http://twitpic.com/a1unrn/full [13:54] czajkowski: uhm... Local Folder? [13:54] czajkowski: that's not Ubuntu One service, are you sure? [13:54] rye: notice it's disabled [13:54] rye: she can't select any service [13:55] seems like possibly a tomboy issue though, and nothing to do with u1 [13:55] :( [13:55] dobey: dont say it's so I've more of a chance of getting it fixed if it's U1 [13:58] dobey: yes, disabled, but to switch to another method one needs to clear [13:59] czajkowski: what happens when you synchronize with the current settings? [14:00] rye: http://twitpic.com/a1uree/full [14:05] czajkowski: uhm, you are syncing with local folder, not Ubuntu One [14:05] it wont let me select U1 [14:06] czajkowski: what do you mean "it won't let me" - disabled menu? Then you will need to "Clear" the sync state before switching [14:06] czajkowski: also, bug https://bugs.launchpad.net/ubuntu/+source/tomboy/+bug/1019256 [14:06] Ubuntu bug 1019256 in tomboy (Ubuntu) "Tomboy should trim trailing "/" in WebSyncPreferencesWidget" [Undecided,New] [14:07] czajkowski: so in order to switch you will need to use https://one.ubuntu.com/notes as the URL, without the trailing slash (or install updated tomboy with the patch which I haven't yet published anywhere) [14:08] awwwww [14:08] * czajkowski hugs rye [14:08] beers for you === zyga is now known as zyga-food [14:29] ralsina: cmake2 is merged! :) [14:31] dobey: yay! [14:31] dobey: cmake3 needs reviews then ;-) [14:32] yeah i know :) [14:32] dobey: and then I can wash my eyeballs from that kind of thing for a while [14:37] Good morning folks, let's have a random Alan Perlis quote: "Prolonged contact with the computer turns mathematicians into clerks and vice versa." === mmcc` is now known as mmcc [14:41] mmcc: good morning. And good quote :-) [14:42] * ralsina is now very far on the clerk side of the axis [14:43] stop trolling! [14:50] dobey: I stopped! [14:51] Why do I have Iain Lane's objectives up for review???? [14:52] lol [14:52] Oh, I love allhand's things that look like buttons, but where you have to click on the letters [14:52] eric, rejected, let's see what happened. Are you fired now? [14:53] also the channel mismatches [14:53] ralsina: yay, early weekend! [14:53] Oh, he's on "desktop" and allhands has him filed on the wrong team. Fun. [14:54] ralsina: i guess you should give him some objectives. maybe he can do python3! :) [14:54] twisted needs lubbins [14:55] that'd be a fantastic idea. If you let me port twisted to python 3, I can guarantee I'll run it into the ground for good. [14:56] ralsina: I'm on a team? :( [14:56] :) [14:56] thisfred: you are in two! [14:56] w00t [14:56] twice the meetings! [14:57] thisfred: but we are selling you to a Belgrade team for rights to the churro stand and a ball to be named later. [14:57] so does anyone actually run the windows SSO client from the command line, eg. for testing? [14:57] mmcc: good question. I am guessing no. [14:57] ralsina: so am I, because it might have never worked [14:57] ralsina: at least I'll get to play [14:58] mmcc: but u1cp starts it and it works then ;-) [14:58] mmcc: only when i really need to, which isn't often. otherwise i just let the SSO from the latest release run [14:58] ralsina: yes it should - the only way it knows how to find the path to the SSO backend is using the registry [14:58] whic is why I asked [14:59] briancurtin: that makes sense. have you needed to since I joined and mandel did the tcpactivation refactoring? [14:59] mmcc: there should be a last resort which is "starting the binary from the same place where the sso binary is" [14:59] mmcc: not sure [14:59] or rather, where the u1cp binary is [14:59] pretty likely since you joined, no idea about tcpactivation [14:59] ralsina: agreed. that's what it does for finding the other binaries, but tcpactivation has a different code path [15:00] mmcc: oh, free inconsistencies. Fun. [15:00] me [15:00] thisfred: it's always about you, isn't it [15:00] heh [15:00] solipsism is the only reasonable stance [15:00] me [15:01] me [15:01] ah crap, typing now [15:02] me [15:02] me [15:02] me [15:03] me [15:03] thisfred: go [15:03] DONE: Bug #999029 TODO: objectives | max document size implementation (no bug yet) BLOCKED: no NEXT: mmcc [15:03] DONE: dev path fixes, packaging fixes, fixed bug 1018125 [15:03] TODO: more path fixes, land said fixes, send a .app to ralsina [15:03] BLOCK: no [15:03] NEXT: DOBEY [15:03] Launchpad bug 999029 in U1DB "make test suite cleaner for no c_backend_wrapper" [Low,In progress] https://launchpad.net/bugs/999029 [15:03] Launchpad bug 1018125 in Ubuntu One Client "u1sdtool and ubuntuone-syncdaemon import gireactor,dbus when platform != win32" [Medium,Confirmed] https://launchpad.net/bugs/1018125 [15:03] DONE: team call, chat with mvo about gtk sso move, tarmac tweaks, cmake2 review, started 3.0.2 SRUs [15:03] TODO: objectives, reviews, finish 3.0.2 SRUs, poke at some bugs [15:03] BLCK: None. [15:03] gatox: go [15:03] DONE: [15:03] Reviews and Refactor (see the pattern? :P) [15:03] TODO: [15:03] Finish with tests refactoring TODAY. objectives [15:03] BLOCKED: [15:03] No [15:03] briancurtin, go [15:03] DONE: python 3, read up on unicode, started the unicode changes branch [15:03] TODO: fix up a few branches, push on with unicode [15:03] BLOCKED: no [15:03] NOTE: doctors appt this afternoon so i will need to shift some time around. i'll probably be gone for a while in my afternoon but stick around through early evening [15:03] NEXT: ralsina [15:04] DONE: merged cmake2, proposed cmake3, carefully avoiding doing a cmake4, team call, 1-1s TODO: fix a bug, lots of emails tell you all to DO YOUR OBJECTIVES (done now) BLOCKED: no NEXT alecu [15:04] DONE: team hangout with hats, reviews, played with the t3k branch [15:04] TODO: objectives, more t3k [15:04] BLOCKED: no [15:04] comments? [15:04] DO YOUR OBJECTIVES? [15:04] Talk to me if you don't know what to do about them [15:05] that would be me [15:05] * ralsina opens the guinness book of world records and starts assigning at random [15:05] jjajaja [15:05] thisfred: "walk more than 2.3 miles holding a raw egg in a small spoon" [15:05] 1) be taller. 2) be a shot caller. 3) be a baller. [15:05] done! [15:05] * alecu picks "grow the longest beard" [15:05] I'm pretty good at chugging beer, and eating biscuits and whistling [15:06] thisfred: at the same time? impressive! [15:06] how and where do we do these objectives? "all hands" something? [15:06] briancurtin: allhands.canonical.com [15:06] objective: c [15:06] hehe [15:07] i guess nobody got my joke [15:07] mmcc: lolz [15:07] dobey: sorry, no :-( [15:07] i got that one [15:07] dobey: well, i got the song quote [15:08] dobey: the ship has sailed on be taller for all of us [15:08] ralsina: not me I sleep upside-down [15:08] we can make it a stretch goal [15:08] dobey: and I have not seen mike, but only thisfred and brian can claim to have succeeded [15:08] instantdrumroll.com [15:08] thisfred: BOOM [15:08] why on earth are objectives due on a saturday? [15:08] oh man we have to give target dates? [15:09] I think now I do: "Lil Troy---Wanna be a baller" [15:09] briancurtin: target date is april 2013 [15:09] briancurtin: april 2013 is good for that, unless you really expect to achieve something earlier [15:09] ralsina, so everyone who forgot to do it during the week, can do it :P [15:09] alecu: no that's not it, I think [15:09] (like me) [15:09] i was going to put them all on christmas day, as presents to the team [15:09] You guys *can* set things to earlier dates if you feel ambitious [15:09] alecu: skee-lo [15:09] but I have to *approve* them by tomorrow. I am not approving on saturday, people [15:10] why oh why am i not getting any audio out of flash/html5 :-/ [15:10] ok, lunch for me! === gatox is now known as gatox_lunch [15:10] "survive the return of quetzlcoatl" [15:11] really, that's my only goal for this year. well that, and go see The Hobbit. they happen at roughly the same time [15:11] learn to pronounce quetzlcoatl [15:12] thisfred: it sounds just like it's written [15:12] create new meme: lol-axolotls [15:13] oh, too late http://cheezburger.com/1563258112 [15:14] climb popocatepetl [15:14] thisfred: climbing active volcanos is bad. Specially if you carry an axolotl [15:15] Well I would have it carry me, but it'd probably walk right into the volcano mouth, with its lack of eyesight [15:17] axolotl, the other white meat [15:24] crap, that's an ugly animal. [15:25] why do credit card companies always make it so hard to close an account [15:25] dobey: should be obvious [15:25] s/obvious/illegal/ [15:28] well, time to get lunch [15:28] bbiab [15:30] alecu: it's like working from [15:30] home: when you spend all your life in the dark, you can afford to be ;) === salgado is now known as salgado-lunch [15:39] at least if you don't have a webcam === teknico_ is now known as teknico [15:48] * alecu has lunch and runs errands === yofel_ is now known as yofel [16:12] windows question - on my machine, when activation queries the registry for the path to sso-client, it gets back 'e:\ubuntu-sso-client\use-tcp-activation\dist\windows-ubuntu-sso-client.exe', and I don't have an e: drive, so that fails. that's OK - I've never installed U1 on this VM and maybe that path is something strange from my test setup? [16:12] mmcc: that's from the test setup [16:13] ok, so my question is: is it OK to use os.path.exists to test what I get from the registry in order to fall back to the exe in the source tree? [16:13] the nonexistant e: drive threw me [16:13] mmcc: sure [16:14] or should I only use the registry one if sys.frozen isn't defined? [16:15] because now I'm not sure how you'd test changes to the sso backend on windows without editing the registry, if you've already installed it at some point ... [16:15] mmcc: i think that would work [16:15] i guess only ever try the registry on sys.frozen, and always have the fallback to source tree [16:15] briancurtin: ok, I think it'd work too - it's what we do in the other function that looks for exes [16:15] sounds good! === gatox_lunch is now known as gatox [16:24] is it just me or does the 'next' button on the objective page not work? [16:24] (on chrome) [16:24] oh nevermind, i had a blank objective added === teknico_ is now known as teknico === zyga-food is now known as zyga [16:56] dobey, gatox - I have a few merges from last night and this morning for you guys to look at: [16:57] mmcc, do you need reviews? [16:57] https://code.launchpad.net/~mikemc/ubuntuone-client/fix-1018125-darwin-no-gireactor/+merge/112702 [16:57] mmcc, i was expecting you to let me know when you update one of the reviews with mandel comments [16:57] https://code.launchpad.net/~mikemc/ubuntu-sso-client/fix-992593-backend-path-darwin-pkgd/+merge/112709 and https://code.launchpad.net/~mikemc/ubuntu-sso-client/fix-1018924-use-buildout-python/+merge/112813 [16:58] gatox, I'm not aware of anything pending with mandel's comments... which branch is that about? [16:58] mmcc, i'll review them right now [17:01] mmcc, question...... why tihs: if sys.platform not in ('win32', 'darwin'): [17:01] and not: if sys.platform == 'linux2': [17:01] just asking.... [17:03] hey, a lunch! [17:08] gatox, no great reason. didn't actually know the sys.platform constant for linux, I would've guessed 'linux', so I'm glad I did it the other way :) [17:10] mmcc, i think is more clear to say "do this just for linux", what do you think? [17:10] mmcc: ok, i'll look at them as soon as i can [17:11] gatox sure, I'm happy to change it [17:14] ok gatox, pushed that change. [17:14] mmcc, thx [17:14] btw, for that no-gireactor branch, note that (at least with trunk), those scripts still don't actually work - this fix is just for their imports... [17:15] mmcc, ok, roger that...... so i shouldn't run the tests for that? [17:16] gatox, do the unit tests even cover these scripts? [17:16] mmcc, there is no test for that :P [17:16] mmcc, jeje was just the standard reaction jeje [17:17] mmcc, +1 [17:17] gatox: right :) if you wanted to be super thorough you could merge this branch with your fsevents branch and test them IRL that way, but we'll be doing that soon anyway [17:17] thx === salgado-lunch is now known as salgado [17:26] mmcc, let me know if i'm being annoying :P i added a need fixing here: https://code.launchpad.net/~mikemc/ubuntu-sso-client/fix-992593-backend-path-darwin-pkgd/+merge/112709 [17:26] but should be easy to do [17:28] gatox, not annoying, good idea. thanks :) [17:30] mmcc, and this one: https://code.launchpad.net/~mikemc/ubuntu-sso-client/fix-1018924-use-buildout-python/+merge/112813 seems to be in the same situation, i don't think there are pre-existing tests for that [17:34] isn't there a good reason to avoid doing " " + foo? [17:35] dobey, it think " %s" % foo is better according some comments [17:35] thisfred, ^ [17:35] right, that's whata i thought. but i was hoping someone had a link pointing to something that documented why [17:36] dobey, you mean "python " + foo right? I'm not actually doing " " + am i? [17:36] dobey: it's slower than doing ''.join(list), but it only matters for large numbers of strings [17:36] mmcc: i mean "string literal" + foo [17:37] so for a single plus outside of a loop, it's fine [17:37] in my book [17:37] there may be other books [17:37] it's also more forgiving about types [17:37] where do i buy a copy of the thisfred book [17:37] beuno: % or +? [17:37] dobey, % [17:37] it casts it to a string [17:37] right [17:37] so you don't worry about what foo is [17:38] briancurtin: it's currently out of print [17:38] beuno: I would argue that that's a downside of % [17:38] ;) [17:39] well of %s specifically [17:39] that's why your book is out of print :p [17:39] hehe [17:42] ok so for this use case, we have two things that are definitely short and definitelt strings, and it's done once, so ok to use "" + var? [18:06] ok, time for lunch. will push some new tests when I get back === daker is now known as daker_ === daker_ is now known as daker__ [18:17] lunch/doctor [19:20] ralsina: ok, if we have some fix upstream, whom do we contact to get it packaged in Ubuntu? [19:20] rye: I suppose their mailing list [19:23] rye: fix for what? [19:26] dobey: tomboy https://bugs.launchpad.net/ubuntu/+source/tomboy/+bug/1019256 [19:26] Ubuntu bug 1019256 in tomboy (Ubuntu) "Tomboy should trim trailing "/" in WebSyncPreferencesWidget" [Undecided,New] [19:29] rye: is it fixed in quantal? upstream bug says fix released, but is it in a new version that's already in quantal? [19:29] oh i guess not [19:29] it was merged upstream today [19:30] rye: you can do it yourself. but it needs to be done in quantal first, then an sru for precise could be done [19:55] dobey: quantal.... hmmm [20:10] ok, EOD for me..... [20:10] * gatox goes away cursing twisted.... [20:17] bye gatox! sorry about that :-) [20:17] ralsina: I've sent the objectives. Please take a look when you can. [20:18] * alecu needs to take Amelia to swiming lessons. [20:18] alecu: will do tonight [20:18] ralsina: great. [20:33] bah can't think of a third :-/ [20:41] * briancurtin back [20:42] question. Anyone have a problem where ubuntu-one moves a file to the trash and doesn't update it? For me I have this fun problem where Ubuntu one will just "eat" some files and i have to restore from backup. [20:45] rhys: files that are deleted from the server either via the web site, or other connected machines, will get moved to trash in all the other clients [20:46] ok.. but this is my keepass file. No machine I have is deleting this. [20:46] is there a log I can look at to see whats getting deleted where? [20:46] in ~/.cache/ubuntuone/log/syncdaemon.log [20:47] there should be several similar files, as the logs get rotated [20:47] rhys, does that happen when you use keepass and save changes to the file? [20:48] salgado: it would have to. nothing else touches that file. [20:49] rhys, I ask because it sounds similar to what happens with gnucash files: bug 953621 [20:49] Launchpad bug 809370 in Ubuntu One Client "duplicate for #953621 hardlinks dont make PyInotify say CLOSE_WRITE and content never gets uploaded" [High,Confirmed] https://launchpad.net/bugs/809370 [20:49] salgado: you've seen it before, or have the same issue? [20:49] ah [20:53] ah. [20:54] Yeah, because it only seems to happen with my keepass file. It has happened exactly once with cryptkeeper, which opens/closes encryptfs files via fuse. [21:23] dobey: cmake 3 was missing a push [21:25] oh weird. the lp merge proposals ui changed again [21:26] weird [21:26] make dist made a tar.bz2 file, though the cpack generator is set to TGZ in the CMakelists.txt [21:27] dobey: the bz2 is done by the dist target [21:27] the tgz you would get with make package but that would be a tarball of the installed files [21:27] you can laugh at the deb it does by setting the cpck generator to DEB ;-) [21:28] ugh :) [21:29] hrmm, seeming some issues [21:29] not surprised [21:31] dobey: salgado: does this mean its a client problem and not a Ubuntu One problem? [21:33] Have to go pick my son from a birthday party. Will be back later checking objectives and such. Have a nice weekend! [21:34] dobey: just put all the needsfixing you see, that branch surely needs some extra loving, it's just that I don't quite know the expected results [21:37] ralsina: yeah, understand. i'm doing. have a good weekend :) [21:38] rhys: i think that bug means it's possibly a kernel or pyinotify problem. [21:39] dobey: if you never do a make check, the .so is not built and is not installed. Is that enough? [21:40] * ralsina guesses no ;-) [21:40] no [21:40] dobey: is there anything I can contribute to the fixing of this bug? [21:41] rhys: i'm sorry, but i don't know. best would probably be to keep up with the bug report that salgado linked [21:56] ok, i need to get out of here [21:56] have a good weekend everyone! [21:56] see ya dobey [21:56] bye dobey have a good one [22:08] is there any guideline we have for when merge proposals get too big? i'm at 510 lines right now for Python 3 unicode work, and I'm maybe 25% done...thinking i should just submit this one now [22:09] i just dont want to have some 4000 line difft that no one will look at [22:09] thought we said to keep it below 1k lines? it came up a while back, around when diego's first fsevents branch loomed, darkening the sky [22:10] ah yeah i kind of remember that [22:10] might even have been 500 lines... 500 soundslike a good chunk size if that makes sense to stop there [22:31] briancurtin: you can just do half first and another half later [22:31] 1K lines is the usual threshold [22:31] 500 is better, but over 1K is unbearable [22:31] agreed. i'll see if it makes sense to push on past 510 organizationally, otherwise i'll start a new one [22:32] do we use the 'resubmit' review comment type? what for? [22:33] eg, if gatox suggested I add tests and I add tests, do I note that with a 'resubmit' comment or is that not what that means [23:02] and i'm out of here. enjoy the weekend all [23:38] ok, didn't quite get as far as I'd hoped but it's time to head out. Just a couple of executable path bugs to squash and we should be in the same good shape mandel saw earlier [23:38] bye