| akgraner | alecu - thanks - I try to catch all those things but often miss a few | 01:17 |
|---|---|---|
| akgraner | I'll get that fixed | 01:17 |
| JamesTait | Good morning all! | 09:06 |
| gord | hi all, anyone know how i can get u1 to share a folder without using nautilus? its nautilus integration broke on this machine for some reason (on 11.10) so i don't get any options to do it | 11:00 |
| gatox | hi everyone | 11:15 |
| nessita | good morning everyone! | 11:39 |
| gatox | nessita, good morning | 11:39 |
| nessita | gatox: hi there. I owe you reviews, I'll do them now :-) | 11:39 |
| gatox | nessita, ok, i'm finishing with a branch for a high bug too | 11:42 |
| nessita | gatox: did you ever finished/submitted the branch that replaces some of the mock tests in sso? | 11:44 |
| gatox | nessita, not yet... i couldn't because i was working in some high bugs | 11:45 |
| nessita | gatox: is ok, you can tackle that next, since we're not targeting a date now | 11:45 |
| gatox | nessita, ok, let me finish with this bug which is important and then i will finish that | 11:46 |
| nessita | of course | 11:46 |
| nessita | gatox: no hurries at all | 11:46 |
| mandel | morning all! | 12:03 |
| gatox | mandel, hi! | 12:03 |
| mandel | gatox, hello, I own you a diff, but I ccould not buidl yesterday the package cause I had to reinstall xcode for that :( | 12:07 |
| mandel | gatox, I'll have it in a few mins :) | 12:07 |
| gatox | mandel, no problem!!! i red your twit about xcode :P | 12:08 |
| mandel | gatox, yeah, is terrible, the problem I had is that my pyqt on the mac did not have webkit (must be and old version) so I had to compile sip and pyqt and make was not there, WTF?!?! | 12:09 |
| gatox | mandel, yacks!!!! it seems that pyqt is always a few versions below in mac | 12:09 |
| mandel | gatox, well, they don't have a package so you have to keep track of it and compile it when needed, is not as easy as apt-get :P | 12:11 |
| gatox | :P | 12:11 |
| nessita | mandel: good morning! | 12:18 |
| nessita | gatox: did you talk to mandel re: editor and blank spaces? | 12:18 |
| ralsina | good morning everyone! | 12:18 |
| gatox | nessita, nop | 12:18 |
| gatox | nessita, but i'm ninja-evangelizando anyway | 12:19 |
| ralsina | mandel: USE A REAL EDITOR PLEASE? | 12:19 |
| ralsina | nessita: there :-) | 12:19 |
| nessita | mandel: hey, seems like (some of) your editor(s) are not configured to trim trialing spaces... could you please check that? | 12:19 |
| nessita | ralsina: or that :-) | 12:19 |
| gatox | like......... i don't know...... ninja | 12:20 |
| gatox | jeje | 12:20 |
| * ralsina looks forward to the first file committed with \\r EOLs | 12:20 | |
| * ralsina switched to ninja on windows today | 12:20 | |
| mandel | ralsina, I use vim with the settings from the python core developers, so what is the problem? | 12:21 |
| mandel | nessita, ^ | 12:21 |
| ralsina | mandel: doesn't trim spaces | 12:21 |
| ralsina | I forgot to pep8-check your merge into -installer on friday and it had a bunch of those | 12:21 |
| nessita | mandel: your last branch of autoupdate added lines longer than 79 chars and trailing spaces | 12:21 |
| gatox | ralsina, cool :P | 12:21 |
| ralsina | I wonder if we should just have a set of pre-commit hooks | 12:22 |
| nessita | mandel: these were added by your branch http://pastebin.ubuntu.com/697852/ | 12:22 |
| mandel | ralsina, autocmd BufWritePre *.py normal m`:%s/\s\+$//e | 12:24 |
| mandel | `` does the trick for me | 12:24 |
| ralsina | mandel: in my editor, just clicking "save" does the trick ;-) | 12:24 |
| nessita | mandel, ralsina: can you please review https://code.launchpad.net/~nataliabidart/ubuntu-sso-client/no-more-deprecated-service/+merge/77241 ? | 12:24 |
| ralsina | nessita: on it! | 12:25 |
| mandel | ralsina, that is what that line does :) | 12:26 |
| ralsina | mandel: it clicks save in my editor???? Vim is almighty! ;-) | 12:26 |
| ralsina | nessita: I suppose you checked nothing uses that, right? If yes, the branch is trivial and I will approve it as soon as the tests are done | 12:27 |
| nessita | ralsina: I made a grep over all our projects, and found that somce utilities scripts from u1-servers were using it, but I have a branch in progress to fix that (will not breake until sso version in ubunet source tree is bumped) | 12:28 |
| ralsina | nessita: cool | 12:28 |
| mandel | nessita, sure | 12:35 |
| mandel | nessita, can you let me know the branch that added the pep8 issues so that I can fix it | 12:36 |
| nessita | mandel: gatox fixed all the issues already. The branch was your last installer branch. | 12:36 |
| dobey | ralsina: pre-commit hooks in bzr are pain | 12:41 |
| dobey | ralsina: it would be nice if we could tell pep8 to exit with a non-zero return code though, then problem would be solved :) | 12:41 |
| ralsina | dobey: it's kinda easy, since it's never supposed to print anything unless there is an error | 12:42 |
| ralsina | like [ -z `pep8 whatever`] | 12:42 |
| dobey | true, but having to do that is kind of lame | 12:43 |
| dobey | it should just exit 1 or something if it prints things | 12:44 |
| dobey | like everything else on the planet does | 12:44 |
| ralsina | dobey: yeah | 12:44 |
| ralsina | I wonder if pep8 has an issue tracker somewhere | 12:45 |
| dobey | i don't know. i want to get all our stuff on pyflakes + pep8, and get rid of pylint; it was also mentioned before that there's a flake8 or something that's basically pyflakes+pep8 already | 12:46 |
| nessita | dobey: hey there! I saw an email this morning saying "1140 revisions were removed from the branch. " where "the branch" is u1client, and I freaked out. Then I saw that nothign was missing in u1client... any idea what happened? | 12:47 |
| dobey | nessita: yes; i screwed up yesterday. then promply fixed it when i saw what i did | 12:48 |
| nessita | dobey: oh, ok. Everything is fixed now then? | 12:48 |
| dobey | yep | 12:48 |
| nessita | :-) | 12:49 |
| dobey | side effect of urgency driven development ;) | 12:50 |
| ralsina | pyflakes is easy to use because it misses way too much stuff | 12:51 |
| dobey | pyflakes is easy to use because it's extensible and not compltely batshit insane like pylint | 12:51 |
| ralsina | OTOH, pylint finds way too much stuff and is semi-broken on windows (reads PYTHONOATH wrong) | 12:51 |
| dobey | anyway | 12:52 |
| dobey | hrmm, alecu no aqui | 12:54 |
| nessita | dobey: he should be entering any time now | 12:57 |
| === m_conley_away is now known as m_conley | ||
| nessita | me | 13:00 |
| dobey | meh | 13:00 |
| nessita | dobey: hey, I see no client nightlies build since 2011-09-08, are we still havng the "reached the limit of builds" error? | 13:01 |
| gatox | me | 13:02 |
| nessita | ralsina, mandel, alecu: standup? | 13:02 |
| ralsina | me | 13:02 |
| dobey | nessita: yes, i haven't had time to file a bug against LP for that yet; it's been a lower priority than dealing with oneiric | 13:02 |
| nessita | dobey: of course, can I help somehow? | 13:03 |
| nessita | mandel: say me | 13:03 |
| nessita | DONE: Email catch up, bosses catch up, open-heart meeting, proposed branch for bug #859635 | 13:04 |
| nessita | TODO: finish bug #859635, start ussoc cleanup! :-D | 13:04 |
| nessita | BLOCKED: nopes | 13:04 |
| nessita | NEXT: dobey | 13:04 |
| ubot4 | Launchpad bug 859635 in ubuntuone-servers (and 1 other project) "Remove the deprecated DBus iface (affects: 1) (heat: 6)" [Medium,In progress] https://launchpad.net/bugs/859635 | 13:04 |
| dobey | λ DONE: releases/uploads/stable branches | 13:04 |
| dobey | λ TODO: tickets, debugging | 13:04 |
| dobey | λ BLCK: None. | 13:04 |
| dobey | gatox: go | 13:04 |
| gatox | DONE: | 13:04 |
| gatox | Reset Password and some work on the flow related to forgotten and current user sign in too. | 13:04 |
| gatox | TODO: | 13:04 |
| gatox | Start migrating test to not mocker. | 13:04 |
| gatox | BLOCKED: | 13:04 |
| gatox | No | 13:04 |
| gatox | ralsina, go | 13:04 |
| ralsina | DONE: release stuff, administrivia, reviews, organizing my head, call, mgmt call | 13:04 |
| ralsina | TODO: finish thinking what we should work on | 13:04 |
| ralsina | BLOCKED: no | 13:04 |
| mandel | me | 13:05 |
| nessita | mandel: go | 13:05 |
| mandel | DONE: Implemented code to uninstall msi (I need to clean the pep8 issues) Look at migrating data from old beta to new (we need to make sure we do nto brake things when we generate lots of uplaods." | 13:06 |
| mandel | TODO: as said, clean code and junt people for reviews | 13:06 |
| mandel | BLOCKED no | 13:06 |
| nessita | mandel: you with alecu? | 13:07 |
| mandel | nessita, I'm at his office but not with him | 13:12 |
| mandel | nessita, maybe he is taking amelia to the kinder or something | 13:13 |
| alecu | mandel, or something | 13:13 |
| nessita | alecu: hi there! dobey was looking for you, and I was eager to read your standup | 13:13 |
| alecu | nessita, were you looking for me? | 13:13 |
| alecu | nessita, I pasted my standup a few lines above | 13:14 |
| nessita | alecu: you did? I don't see them... have a timestamp? | 13:14 |
| duanedesign | mandel: trying to help a user with a stuck windows download. So far the only thing i can find is this exception. https://pastebin.canonical.com/53489/ | 13:15 |
| alecu | nessita, 10:06 | 13:15 |
| dobey | hrmm | 13:15 |
| alecu | dobey, you were looking for me? | 13:15 |
| nessita | alecu: I don't have that, actually I see you coming in at 10:11 | 13:15 |
| nessita | (10:11:33 AM) alecu [~alecu@186.19.234.137] entered the room. | 13:15 |
| nessita | alecu: can you please re-paste? | 13:15 |
| dobey | alecu: you didn't paste your standup after joining the channel at least :) | 13:16 |
| alecu | nessita, above that! | 13:16 |
| mandel | duanedesign, let me take a look | 13:16 |
| dobey | alecu: i was wondering if you were going to fix your ubuntuone-client branch so it will land :) | 13:16 |
| alecu | DONE: read a lot on proxies, discussed proxy issues with mandel, started a design document for the proxy features | 13:16 |
| alecu | TODO: more proxies | 13:16 |
| alecu | BLOCKED: no | 13:16 |
| nessita | alecu: this is all I have above! http://pastebin.ubuntu.com/698486/ | 13:16 |
| alecu | dobey, which branch are you referring to? | 13:17 |
| ralsina | alecu: I don't see it either | 13:17 |
| mandel | duanedesign, can you get me a longer log please | 13:17 |
| ralsina | duanedesign: is that on windows? | 13:17 |
| dobey | alecu: the unicode one | 13:17 |
| ralsina | nessita: +1 on deprecated_interface | 13:18 |
| dobey | alecu: https://code.launchpad.net/~alecu/ubuntuone-client/unicode-usernames/+merge/76145 | 13:18 |
| alecu | ok, probably my irc client thought it was connected, but it was not | 13:18 |
| alecu | nessita, ralsina: I've repasted my standup (10:16) did it show up now? | 13:18 |
| nessita | alecu: yes, thanks | 13:18 |
| alecu | cool | 13:18 |
| ralsina | alecu: yes, thanks | 13:19 |
| duanedesign | ralsina: yes mandel, i can | 13:20 |
| alecu | dobey, oh, I see that it's failing with a UnicodeEncodeError now. I'll take a look in an hour or so. | 13:20 |
| alecu | nessita, ralsina: I still have the server-timestamp branches to finish, but as we discussed with ralsina I'm using the mini sprint with mandel to plan the proxy work. | 13:21 |
| nessita | alecu: the main issue is that oneiric final freeze is tomorrow | 13:22 |
| nessita | alecu: so if there is something to ship to Ubuntu, it has to be today | 13:22 |
| alecu | nessita, this is not for oneiric, it's for windows | 13:22 |
| nessita | alecu: right, and the hash queue delay is server only? | 13:23 |
| alecu | nessita, no, the hash queue delay is a client side branch, and it was merged a few weeks ago. | 13:24 |
| alecu | nessita, the server fixes were just for the failing server tests. | 13:25 |
| alecu | also, regarding the other branch, the server-timestamp feature I'm still implementing, we need to test it a lot, and mrPresi already pointed out a few server caching problems with it. | 13:25 |
| alecu | (mrPresi == pindonga) | 13:26 |
| nessita | alecu: ack, thanks for that info | 13:27 |
| alecu | no prob. | 13:28 |
| ralsina | yes, the timestamp branch is at least a week off | 13:28 |
| ralsina | and in linux, we do have a "simple" solution (install ntp!) | 13:28 |
| alecu | also, the timestamp branch may be critical on windows, but it's "medium" on windows. | 13:28 |
| alecu | ralsina, we should suggest the GPLd ntp client on windows too. | 13:29 |
| ralsina | alecu: yes | 13:29 |
| nessita | gatox: ping | 13:30 |
| alecu | ralsina, since I installed it on my devel VMs, I never had that issue anymore. And I used to have it a lot. | 13:30 |
| gatox | nessita, pong | 13:30 |
| nessita | gatox: I'm IRL testing the captcha loading branch in installer | 13:30 |
| gatox | nessita, ajap... | 13:30 |
| nessita | gatox: and what I see is that as soon as I open the installer, the overlay is shown, even if I did not click on register | 13:30 |
| nessita | gatox: that should not be there: the loading overlay shoul only be shown in the register screen (when loading captcha) | 13:31 |
| ralsina | that's because we are loading the captcha in __init__ | 13:31 |
| gatox | nessita, can you send me an screenshot?? i don't quite follow | 13:31 |
| ralsina | and we should load it in initializePage | 13:31 |
| ralsina | that's a sso fix | 13:31 |
| nessita | gatox: open the installer with your captcha branch, having the sso captcha branch in the PYTHONPATH | 13:32 |
| nessita | gatox: you will get the overlay in the first screen of the installer | 13:32 |
| * gatox testing... | 13:32 | |
| nessita | gatox: and that's not good, we need that overlay only in the registration screen | 13:33 |
| ralsina | dobey, nessita, gatox, mandel, alecu (and eric, if you come back to the team at the time): december sprint is official | 13:33 |
| gatox | ralsina, cool | 13:33 |
| nessita | ralsina: meaning book tickets? | 13:33 |
| ralsina | so, start booking | 13:33 |
| thisfred | ralsina: ah, eh when? where? | 13:33 |
| dobey | we have hotel then? is it same as everyone else? | 13:33 |
| thisfred | was there mail about this? | 13:34 |
| ralsina | gatox: you need to do a branch in SSO that delays initialization of the captcha until the page is initialized | 13:34 |
| ralsina | thisfred: sorry, we discussed it on mumble :-/ | 13:34 |
| thisfred | np | 13:34 |
| dobey | thisfred: and by we, he means "not us" | 13:34 |
| ralsina | thisfred: since you are on loan, you are not coming unless the loan ends | 13:34 |
| ralsina | dobey: SORRY OK ;-) | 13:34 |
| ralsina | it's official since 2 minutes ago. Consider this "discussion" ;-) | 13:35 |
| thisfred | ralsina: ok, then I don't know yet, I guess. I think the idea is I come back after UDs | 13:35 |
| ralsina | the hotel is not booked yet, but will surely be either the same as the other sprints or 1 block away | 13:35 |
| ralsina | thisfred: cool, then I'll add you! | 13:35 |
| dobey | well even if he was still on loan, i think he'd still be going under a different guise to a different sprint :) | 13:35 |
| ralsina | dobey: yeah, there is that :-) | 13:35 |
| thisfred | ralsina: let's confirm this with Chipaca and jdo | 13:36 |
| duanedesign | mandel: tried to innclude what i thought was relevant. I got a whole archive of logs from the user. It has two exception logs and a status.log at the end. https://pastebin.canonical.com/53512/ | 13:36 |
| dobey | so either way, he should be booking tickets :P | 13:36 |
| gatox | nessita, ahhhhhh now i see | 13:36 |
| thisfred | so, question remains, where and when? | 13:36 |
| mandel | duanedesign, ok, thx | 13:36 |
| * mandel looking | 13:36 | |
| ralsina | thisfred: buenos aires | 13:37 |
| ralsina | thisfred: december 5-9 | 13:37 |
| mandel | ralsina, is that 100% sure, I'd like to book my tickets so I dont have to go around the globe :) | 13:37 |
| ralsina | mandel: it is | 13:37 |
| mandel | ralsina, cool | 13:37 |
| thisfred | uh oh, that's my anniversary... This is not gonna sit well :) | 13:37 |
| ralsina | thisfred: romantic tango vacation? | 13:38 |
| thisfred | I doubt my wife can get time off, but I'll certainly propose it | 13:38 |
| dobey | thisfred: i hear your wife liked buenos aires :P | 13:38 |
| thisfred | hell yeah | 13:38 |
| ralsina | december is nice. Hot bot not horribly hot | 13:38 |
| ralsina | but* | 13:39 |
| Chipaca | confirm what with who? | 13:39 |
| ralsina | Chipaca: thisfred and sprint | 13:39 |
| mandel | duanedesign, that looks terrible :( | 13:39 |
| thisfred | Chipaca: when the loan to accounts expires, basically | 13:39 |
| ralsina | sprint is confirmed, I am deciding if I want to piss eric's wife by making him travel on his anniversary | 13:40 |
| Chipaca | ralsina: well, he'll be there either with desktop or with accounts. Here's hoping it's accounts. | 13:40 |
| thisfred | dates are the same exactly? | 13:40 |
| ralsina | you are hoping it's accounts? | 13:41 |
| dobey | thisfred: yeah, dates are same | 13:41 |
| mandel | ralsina, everyone is runing away from the desktop :P | 13:42 |
| thisfred | Not me. | 13:42 |
| dobey | the "desktop" is dead; i don't know why people can't see that | 13:42 |
| duanedesign | mandel: oh no :\ terrible is not good. | 13:42 |
| mandel | duanedesign, is there a bug for this? | 13:43 |
| dobey | ralsina: we need to rename our team to awesome+, because it's really what we do. | 13:43 |
| duanedesign | mandel: no this was a user that contacted us through support | 13:44 |
| ralsina | duanedesign, mandel: looks like screwed metadata to me | 13:44 |
| duanedesign | hmm | 13:44 |
| alecu | guys and gals, I need to make the run for the kinder, see you in a half hour or so. | 13:45 |
| alecu | mandel, I'll be there soon | 13:45 |
| ralsina | bye alecu! | 13:45 |
| nessita | gatox: let me know when is fixed | 13:45 |
| gatox | nessita, ok | 13:46 |
| duanedesign | oh, ralsina. Thanks for contributing to the facebook page. Any comments are greatly appreciated. | 13:47 |
| ralsina | duanedesign: np, I just take a look every now and then to see if I can help :-) | 13:48 |
| duanedesign | :) | 13:48 |
| mandel | ralsina, it does look like that except for the fact that we got this 'failure: I/O operation on closed file' | 13:50 |
| ralsina | mandel: that could be just a logging problem | 13:50 |
| ralsina | or that a file got locked while syncdaemon was writing (much scarier) | 13:50 |
| ralsina | or rather, reading | 13:50 |
| mandel | ralsina, that is why I'm thinking | 13:50 |
| ralsina | I say reading because it was an upload | 13:51 |
| dobey | duanedesign: speaking of help; why is the link to the interview with chipaca posted 3 times from facebook (on twitter)? :P | 13:52 |
| Chipaca | because I'm 3x awesome, clearly | 13:52 |
| nessita | gatox: added Needs Info to ubuntu-sso-client/network-detect | 13:53 |
| nessita | gatox: also, I think the commit message for the sso branch is not accurate, since what that branch is fixing is provide a new method in the linux side | 13:54 |
| gatox | nessita, ok, i'll review that in a while, i'm fixing some test in pass recover | 13:54 |
| nessita | gatox: ack | 13:54 |
| duanedesign | dobey: ugh, i will look at that. | 13:55 |
| mandel | ralsina, duanedesign I'm moving this conversation to #chicharra so that verterok can give us a hand | 13:55 |
| ralsina | mandel: cool, good luck! | 13:56 |
| verterok | mandel: I'm here | 13:56 |
| nessita | ralsina: have any ideas for https://bugs.launchpad.net/ubuntuone-windows-installer/+bug/860700 ? | 14:01 |
| ubot4 | Launchpad bug 860700 in ubuntuone-windows-installer "Error message when starting PYTHON27.dll not found (affects: 2) (heat: 10)" [Undecided,New] | 14:01 |
| ralsina | nessita: not really | 14:01 |
| ralsina | nessita: that seems to indicate that a file got deleted but he says he reinstalled | 14:02 |
| ralsina | I'll ask for some info | 14:02 |
| nessita | ralsina: thanks | 14:02 |
| ralsina | np | 14:02 |
| ralsina | who needs a rview from me before I start a non-interruptible task? | 14:18 |
| ralsina | gatox, nessita, dobey, mandel: ^ | 14:18 |
| nessita | ralsina: not here | 14:18 |
| gatox | ralsina, me! | 14:21 |
| ralsina | gatox: where? | 14:21 |
| gatox | ralsina, let me give you the links | 14:22 |
| dobey | not yet | 14:22 |
| gatox | ralsina, https://code.launchpad.net/~diegosarmentero/ubuntuone-windows-installer/network-detect - https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/captcha-refresh/+merge/76621 | 14:22 |
| mandel | ralsina, I do | 14:23 |
| mandel | ralsina, give me a sec I'll get the mp | 14:23 |
| ralsina | mandel: shoot | 14:24 |
| mandel | ralsina, https://code.launchpad.net/~mandel/ubuntuone-windows-installer/check-old-app | 14:24 |
| mandel | nessita, if you can give me an extra one it would be greatly appreciated ^ | 14:25 |
| ralsina | mandel: ack | 14:25 |
| nessita | mandel: I will! | 14:25 |
| mandel | all pep8 issues should be fixed | 14:25 |
| ralsina | gatox: isn't captcha-refresh the branch we were talking about a while ago that causes the installer to show the overlay on startup? | 14:33 |
| nessita | mandel: OldBetaUpdateTestCase should also test are_updates_present and perform_update, no? | 14:34 |
| gatox | ralsina, yes, there are two captcha refresh, one for sso and one for the installer..... don't review that yet, i have to look where i should apply the change | 14:35 |
| ralsina | ok, queuing then | 14:35 |
| gatox | ralsina, yes, there are two captcha refresh, one for sso and one for the installer..... don't review that yet, i have to look where i should apply the change | 14:35 |
| mandel | nessita, let me take a look | 14:36 |
| ralsina | gatox: sso is doing the refresh_captcha a bit early, in setupUi | 14:37 |
| gatox | ralsina, ok, i'll check there then | 14:37 |
| ralsina | gatox: OTOH, moving it to later means that the UI is a bit less responsive, since the user will have to wait for the captcha | 14:37 |
| mandel | nessita, the tests for that are in a previous branch, this is for the migration from the old to the new while the autoupdater is for everyone | 14:37 |
| ralsina | gatox: ideally, we should do it in setupUi but not show the overlay unless somehow the user is already on that page | 14:38 |
| ralsina | gatox: sot of checking for a flag on initializePage, and show the overlay if the captcha is not ready yet | 14:38 |
| ralsina | s/sot/sort/ | 14:38 |
| nessita | mandel: ah, then I'll start with the other one | 14:39 |
| gatox | ralsina, yep | 14:39 |
| nessita | mandel: no, wait | 14:39 |
| mandel | nessita, AutoupdaterTestCase is the test case for auto-updating, the idea is that the migration code will be deprecated rather quickly | 14:39 |
| nessita | mandel: I started with the first branch, the one that does not depend on any other branch | 14:39 |
| mandel | nessita, :) | 14:39 |
| nessita | mandel: so, these methods: | 14:39 |
| nessita | 92+ are_updates_present = lambda *args, **kwargs: False | 14:39 |
| nessita | 93+ perform_update = lambda *args, **kwargs: None | 14:39 |
| nessita | mandel: are added but have no tests | 14:40 |
| ralsina | gatox: on network-detect, why only when using --installer? You need network on the other case too! :-) | 14:40 |
| nessita | mandel: you're saying those are meant to be used in another branch? | 14:40 |
| gatox | ralsina, in every next??...... i based that in your branch where that was only being checked on the next from License Page | 14:41 |
| mandel | nessita, no, are_updates_present and perform update are in trunk, and have their own tests (on linux are_updates_present is tested to return always false) | 14:41 |
| nessita | mandel: then why the diff is showing those as new addings? | 14:41 |
| mandel | nessita, I removed a trailing white space | 14:41 |
| nessita | ah! | 14:41 |
| mandel | :P | 14:41 |
| nessita | mandel: added NF so we do a better error handing in is_old_beta_installed. Defaulting to True for any return code other than ERROR_UNKNOWN_PRODUCT is too error prone | 14:43 |
| nessita | mandel: from http://msdn.microsoft.com/en-us/library/windows/desktop/aa370130%28v=vs.85%29.aspx, I see we should be checking against ERROR_SUCCESS | 14:44 |
| mandel | nessita, for that I'd say is better to do a if error in (ERROR_SUCCESS, ERROR_MORE_DATA) since in our case the mean more or less the same | 14:47 |
| mandel | nessita, ERROR_MORE_DATA is returned when the buffer allocated is to small but the property is indeed there (there is no way to know the size upfront) | 14:48 |
| mandel | and we remove the if | 14:48 |
| nessita | mandel: if we get ERROR_MORE_DATA we need to call again with more buffer space | 14:48 |
| nessita | mandel: do not assume the old beta is installed if we get more data | 14:49 |
| mandel | nessita, we do not need to call it again, the thing works in the following way, we query for the information of a uid, the information is stored in a COM property (que es una mierda) y the uid is present in the system, that is the msi was installed, we either get a sucess (yeas, the property is small) or a need more that | 14:51 |
| mandel | nessita, both errors state that the uid is indeed know by the system ergo we do have the app installed | 14:52 |
| nessita | mandel: ok, please add a big and clear comment stating that :-) | 14:52 |
| mandel | nessita, sure :) | 14:52 |
| nessita | mandel: so when we read a couple of months from now we don't think is a bug | 14:52 |
| mandel | nessita, changing the code now then :) | 14:52 |
| nessita | mandel: and I think you also meant "adding tests"! :-) | 14:53 |
| ralsina | gatox: sorry for the delay: not on every next() but on the beginning even if called without --installer | 14:54 |
| ralsina | gatox: when I did my branch there was no --installer option :-) | 14:54 |
| gatox | ralsina, ah ok! | 14:54 |
| mandel | yes, and that hehe | 14:57 |
| dobey | mandel: i'm glad you approved my branch that landed 2 days ago, today :P | 14:58 |
| mandel | dobey, yeah, I did not see it was merged hhehe | 15:06 |
| mandel | dobey, I finished the tests and then went to approve without refreshing the page :P | 15:06 |
| alecu | mandel, http://blog.robbowley.net/2011/09/21/estimation-is-at-the-root-of-most-software-project-failures/ | 15:09 |
| czajkowski | mandel: http://twitpic.com/6rwkgi/full# | 15:23 |
| gatox | alecu, interesting | 15:25 |
| gatox | nessita, ralsina review please: https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/pass-recover/+merge/77346 | 15:31 |
| nessita | gatox: I will queue that up after the current ones | 15:32 |
| gatox | nessita, ok | 15:32 |
| dobey | ok, must get lunch; bbiab | 15:32 |
| mandel | nessita, I pushed the changes already (sorry I forgot to ping you) | 15:40 |
| mandel | czajkowski, hahahaha | 15:48 |
| === m_conley is now known as m_conley_away | ||
| czajkowski | mandel: if I have to ue windows it's gonna be my way! | 15:56 |
| ralsina | gatox, mandel: just in case you want a new android phone, I can get this one for you when I go to the UK sprint: http://bit.ly/nqhjJD | 15:58 |
| gatox | ralsina, nop...... but if you want to get something like this: http://www.arroba.com.mx/byte/blog/wp-content/uploads/dell_m17x_alienware_front.jpg PLEASE LET ME KNOW! :P | 16:00 |
| mandel | ralsina, I might ping you from spain about it, I dont know if I'm getting a new one from vodafone | 16:00 |
| ralsina | gatox: missed the name of the place? ;-) | 16:00 |
| ralsina | mandel: orange phones are unlocked. That's nice | 16:00 |
| gatox | ralsina, ah right..... i don't know where to get it :P | 16:01 |
| mandel | ralsina, hahaha I missed it, nice one! | 16:01 |
| ralsina | gatox: an alienware? Who are you, sheldor the magnificent? | 16:02 |
| gatox | ralsina, jejejee i want super hardware! :P | 16:02 |
| ralsina | gatox: that's just too hard to bring back | 16:02 |
| gatox | ralsina, i know.... odio los problemas de importaciones!! :P | 16:04 |
| ralsina | I could ddo it, but I am already brining one for my wife | 16:04 |
| gatox | ralsina, no problem!! i can wait, there's no rush | 16:05 |
| ralsina | nessita is out of power | 16:10 |
| ralsina | meaning her lights are out, not that someone deposed her government | 16:10 |
| dobey | i thought Buenos Aires was the seat of power for AR anyway :) | 16:14 |
| dobey | a coup in cordoba probably wouldn't accomplish much? | 16:14 |
| ralsina | well, actually... there is historical precedent | 16:14 |
| ralsina | http://en.wikipedia.org/wiki/Cordobazo | 16:14 |
| dobey | ah | 16:15 |
| ralsina | I mean, not *successful* precedent ;-) | 16:15 |
| dobey | given all the smoke in the pictures on that page, it doesn't seem like it was a *civil* uprising :P | 16:16 |
| ralsina | cordoba is pretty flammable | 16:17 |
| dobey | so it would seem | 16:17 |
| dobey | anyway | 16:21 |
| dobey | webkit is such a pain | 16:22 |
| === beuno is now known as beuno-lunch | ||
| dobey | meh, gdb | 16:26 |
| mandel | ralsina, when you have time, this is also for review: https://code.launchpad.net/~mandel/ubuntuone-windows-installer/uninstall-old-app | 16:41 |
| ralsina | mandel: will get to it, but I have a queue :-) | 16:42 |
| mandel | ralsina, sure, no problem :) | 16:44 |
| ralsina | gatox: lint problems in pass-recover, adding them to the MP | 16:44 |
| gatox | ralsina, ok, i'm finishing with captcha-refresh.......... and that involve 2 branches | 16:44 |
| mandel | ralsina, ping | 16:51 |
| ralsina | mandel: pong | 16:51 |
| mandel | ralsina, when getting the special folders, why do you use SHGetSpecialFolderPathA insteand of SHGetSpecialFolderPathW ? | 16:51 |
| ralsina | mandel: ignorance! | 16:52 |
| mandel | ralsina, ok, I'll fix that in my branch, since you are not getting the unicode one | 16:52 |
| ralsina | I was young and naïve! | 16:52 |
| ralsina | the dog told me to! | 16:52 |
| ralsina | ok, thanks. | 16:52 |
| mandel | ralsina, and here: http://msdn.microsoft.com/en-us/library/windows/desktop/bb762204(v=vs.85).aspx | 16:53 |
| mandel | ralsina, read the remarks, are we using the right function? | 16:53 |
| ralsina | mandel: including a redistributable is a pain | 16:53 |
| mandel | ralsina, ok | 16:53 |
| ralsina | so, I would say "yes we are" :) | 16:54 |
| dobey | i wish it wasn't so hard to build code written in a real language, on windows | 16:54 |
| ralsina | at least until we deprecate XP. Maybe you can add a comment to that effect | 16:54 |
| mandel | ralsina, will do | 16:54 |
| mandel | dobey, you mean c? | 16:54 |
| ralsina | C++ is pretty easy to build, it's just annoying to write! | 16:55 |
| dobey | mandel: well, vala | 16:55 |
| ralsina | Oh, "real" ;-) | 16:55 |
| dobey | well it outputs C, so only the C needs to be built on windows | 16:55 |
| dobey | but we can easily write code in vala and immediately have bindings for pretty much every language anyone actually uses, on linux | 16:56 |
| dobey | but on windows, well… :( | 16:56 |
| === beuno-lunch is now known as beuno | ||
| mandel | dobey, I have been told that shotwell can just compile with certain versions of vala, is that correct? | 17:00 |
| mandel | and if it is, do we want that problem? | 17:00 |
| dobey | mandel: i don't know | 17:01 |
| dobey | mandel: shotwell is insane. i don't know what versions it must have to compile; but requiring vala >= 0.12.0 or whatever seems like a fine problem | 17:01 |
| dobey | it's not like any modern C code compiles on a system using gcc 2.x with libc5 still :) | 17:02 |
| * nessita is back | 17:02 | |
| dobey | i guess the coup is over | 17:03 |
| nessita | dobey: world coup? | 17:03 |
| nessita | :-D | 17:03 |
| dobey | 12:10 < ralsina> nessita is out of power | 17:03 |
| dobey | 12:10 < ralsina> meaning her lights are out, not that someone deposed her government | 17:03 |
| nessita | lol | 17:03 |
| nessita | so do I still get to boss around? :-D | 17:04 |
| dobey | you can try | 17:04 |
| mandel | ralsina, what are the special folder used in the installer, I want to give them var names, reading this dll.SHGetSpecialFolderPathW(None, buf, 5, False) fucks up my brain | 17:06 |
| mandel | what is 5?!?! | 17:06 |
| ralsina | see what they are assigned to | 17:07 |
| === yofel_ is now known as yofel | ||
| ralsina | mandel: ^ | 17:07 |
| mandel | ralsina, cool thx hehe | 17:07 |
| mandel | did not see that :) | 17:07 |
| ralsina | np :-) | 17:07 |
| * ralsina is not *THAT* evil | 17:07 | |
| mandel | ralsina, if you were a perl programmer you would have been for sure | 17:08 |
| * mandel hears ralsinas voice: with 2 lines of code I have executed the script, stolen your tv and your wife is going to have my baby | 17:09 | |
| duanedesign | mandel: on the windows client file names will be limited to utf8, similar to the Ubuntu client? | 17:14 |
| dobey | sort of but more so | 17:16 |
| duanedesign | i have a user who is having issues with chinese characters. Which I understand are currently not supported in the beta but will in the full release | 17:16 |
| dobey | are they not utf-8? | 17:18 |
| dobey | or unicode at all? | 17:18 |
| dobey | i have a file in my ubuntu one account with chinese characters in the name, just fine | 17:18 |
| dobey | though i haven't used the windows client | 17:19 |
| ralsina | duanedesign: the problem is only with the username being unicode or the "Documents and settings" folder being unicode (in XP) | 17:19 |
| ralsina | duanedesign: if the whole path up to "Ubuntu One" is non-unicode and the filename is unicode, that should work in the beta. | 17:20 |
| dobey | speaking of unicode | 17:20 |
| dobey | alecu: fix your branch! :) | 17:20 |
| dobey | ralsina: by non-unicode, you mean US-ASCII? | 17:20 |
| duanedesign | thank you dobey ralsina | 17:21 |
| ralsina | dobey: probably extended ascii works ;-) | 17:21 |
| * mandel lunch | 17:21 | |
| dobey | ralsina: i am guessing KOI8-R doesn't work :) | 17:21 |
| ralsina | dobey: "things for which str(x) doesn't crash in python" is the right term ;-) | 17:21 |
| dobey | or ISO8859-15 | 17:21 |
| dobey | heh | 17:21 |
| dobey | so ASCII :) | 17:21 |
| alecu | dobey, make tarmac use en_US.UTF-8! | 17:21 |
| dobey | alecu: no | 17:22 |
| dobey | alecu: having it catch such bugs is quite nice :) | 17:22 |
| alecu | dobey, I'm working on fixing it, but it sucks. All ubuntus we support have en_US.UTF-8 already set on LANG | 17:22 |
| alecu | dobey, and we don't have that issue on windows, because it has a sane handling of unicode paths | 17:22 |
| * ralsina reads what alecu wrote and walks away slowly | 17:23 | |
| alecu | ralsina, I said "unicode", not "long" | 17:23 |
| ralsina | hehehe | 17:23 |
| dobey | alecu: unicode or utf-8? | 17:24 |
| dobey | Windows-CP1251 isn't quite either :) | 17:24 |
| nessita | dobey: can I please have (sometime in the future) a trivialish review for https://code.launchpad.net/~nataliabidart/ubuntuone-client/update-sso/+merge/77370 ? | 17:26 |
| nessita | mandel: ping | 17:27 |
| dscassel | Is this the right place to ask questions about the music store? | 17:27 |
| mandel | nessita, pong | 17:27 |
| nessita | mandel: any news on https://code.launchpad.net/~nataliabidart/ubuntu-sso-client/no-more-deprecated-service/+merge/77241 ? | 17:27 |
| nessita | dscassel: yes, what issue are you having? | 17:28 |
| * alecu will have lunch | 17:28 | |
| nessita | mandel: also, did you push to https://code.launchpad.net/~mandel/ubuntuone-windows-installer/check-old-app/+merge/76867? I don't see any chances after my review | 17:28 |
| dscassel | nessita: I'm in Canada. 7digital has a Canadian store, but the Ubuntu One music store sends me to 7digital's international store, which is way more expensive with much less selection. | 17:29 |
| dscassel | Is this a bug? Or a contract thing? | 17:29 |
| nessita | dscassel: let me ask the experts | 17:30 |
| nessita | rye or duanedesign, ping | 17:30 |
| dscassel | nessita: Thanks. | 17:30 |
| mandel | nessita, I ran the tests in your branch, I need to set it to approve as soons as I've read the changes, but it seems is mostly deletes and some import changes | 17:30 |
| nessita | mandel: it nukes the deprecared SSOCredentials class | 17:30 |
| mandel | nessita, and yes, I forgot to push, going it now | 17:31 |
| nessita | mandel: so yes, mostly removes | 17:31 |
| mandel | nessita, I'm noticed, which is really really good | 17:31 |
| mandel | nessita, I just want to make sure is not used anywhere :) | 17:31 |
| mandel | nessita, revno of the branch should be 76 now | 17:31 |
| nessita | mandel: ack | 17:32 |
| dobey | dscassel: it's not quit either, really. | 17:32 |
| dscassel | dobey: How so? | 17:33 |
| dobey | dscassel: though we are planning on a way to get a canadian store up. | 17:33 |
| mandel | nessita, I'll give the final +1 to your branch after lunch, ok? | 17:33 |
| nessita | mandel: ok | 17:34 |
| dscassel | dobey: Glad to hear it. :) | 17:34 |
| dobey | dscassel: i'n not really sure how best to describe it, but there are some changes to the store structure we want to make, and it would make it much easier for us to add more regional stores. it's also unfortunate that amazon doesn't have a canadian mp3 store :-/ | 17:35 |
| dobey | dscassel: but hopefully in the next few months we can have that working with a .ca store :) | 17:37 |
| gatox | nessita, ralsina branch fixed: https://code.launchpad.net/~diegosarmentero/ubuntuone-windows-installer/captcha-refresh/+merge/76622 | 17:37 |
| nessita | gatox: did you need to change something on sso? | 17:38 |
| nessita | mandel: your check-old-app branch has several lint issues | 17:39 |
| gatox | nessita, you have also the branch for captcha refresh in sso: https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/captcha-refresh but no changes were applied there | 17:39 |
| nessita | gatox: right, I was wondering since ralsina mentioned the issue was in sso. Looking now! | 17:39 |
| ralsina | nessita: I changed my mind slightly ;-) | 17:39 |
| nessita | ah | 17:40 |
| gatox | nessita, no..... it was fixed in the installer..... it was easy | 17:40 |
| nessita | great news ;-) | 17:40 |
| ralsina | I wanted the captcha to be initiaized when showing the page, but that would make it slower, so we found a better solution | 17:40 |
| duanedesign | nessita: pong | 17:41 |
| ralsina | gatox: approving the sso captcha branch, starting on the -installer one | 17:41 |
| gatox | ralsina, cool | 17:41 |
| duanedesign | nessita: i am reading scrollback now :) | 17:42 |
| nessita | duanedesign: hi there, I was hoping you could give some info to dscassel about the music store redirecting him to the international store... dobey answered, but maybe you have more info? | 17:42 |
| nessita | duanedesign: ok :-) | 17:42 |
| dobey | nessita: the issue is we don't have a canadian store :) | 17:43 |
| duanedesign | yeah our stores dont match all 7digitals stores which causes a little confusion | 17:44 |
| nessita | oh, I see | 17:45 |
| duanedesign | hmm, coffee getting low. Do not know what will come first EOD or EOC :( | 17:45 |
| nessita | mandel: added some needs fixing to take care of | 17:58 |
| gatox | ralsina, lint issues fixed: https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/pass-recover/+merge/77346 | 17:59 |
| ralsina | gatox: ack | 18:04 |
| ralsina | gatox: +1 on ubuntuone-windows-installer/captcha-refresh | 18:04 |
| gatox | ralsina, cool | 18:07 |
| nessita | gatox: remember I added a needs info to https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/network-detect/+merge/77048 | 18:08 |
| gatox | nessita, yep, i'm looking at that right now | 18:14 |
| mandel | nessita, okis | 18:24 |
| mandel | nessita, nearly done with your branch | 18:24 |
| nessita | mandel: great, let me know since is a dep for me | 18:26 |
| mandel | nessita, I have a question, qhen you cleaned this code, did yu take a look to see if we where using the old creds class outised sso? | 18:30 |
| nessita | mandel: yes sir | 18:30 |
| mandel | nessita, ah, cool, you rock :) | 18:31 |
| mandel | nessita, +1 | 18:34 |
| nessita | mandel: thanks! | 18:35 |
| nessita | dobey: hey, when you have a moment (no hurry), you can re-review https://code.launchpad.net/~nataliabidart/ubuntuone-client/update-sso/+merge/77370? I had to make some changes because I found that AuthorizationDenied and CredentialsError weren't handled when using the new iface... | 19:05 |
| nessita | dobey: also, would you know how can I test that IRL? | 19:06 |
| Atlantic777 | is ubuntuone down? status says that it's ok, but can't access it | 19:06 |
| Atlantic777 | apache error | 19:06 |
| nessita | Atlantic777: what URL? | 19:06 |
| Atlantic777 | oh, sorry, fake alarm | 19:07 |
| Atlantic777 | other urls seems to work | 19:07 |
| Atlantic777 | just few aren't | 19:07 |
| nessita | Atlantic777: which few? :-) (so I can report to our admins) | 19:09 |
| dobey | nessita: make, and do "LD_PRELOAD=`pwd`/libsyncdaemon/.libs/libsyncdaemon-1.0.so.1.0.0 tomboy" | 19:10 |
| dobey | nessita: probably without having a token already | 19:10 |
| nessita | dobey: thanks | 19:10 |
| Atlantic777 | nessita: this one http://ubuntuone.com/62dxagNGKYEc3AoG1bGylA | 19:10 |
| Atlantic777 | but it's posted again | 19:10 |
| Atlantic777 | now it works | 19:10 |
| dobey | nessita: how did the oneiric upgrade go btw? :) | 19:10 |
| nessita | dobey: you can guess I chickened out... but I will do it, will start with my laptop | 19:11 |
| === m_conley_away is now known as m_conley | ||
| nessita | dobey: what should I see? nothing changes in the output, no matter if I have or not have credentials | 19:13 |
| nessita | all I get is [INFO 16:12:42.888] Initializing Mono.Addins | 19:13 |
| mandel | alecu, http://msdn.microsoft.com/en-us/library/bb761273(v=vs.85).aspx | 19:13 |
| alecu | mandel, http://json-rpc.org/ | 19:13 |
| mandel | alecu, https://code.launchpad.net/txsymmetricjsonrpc | 19:14 |
| dobey | nessita: you need to configure sync in the prefs i think | 19:14 |
| nessita | ok looking | 19:15 |
| nessita | brb | 19:15 |
| dobey | nessita: or open the prefs, and then authorize in the control panel... | 19:15 |
| dobey | impatient people | 19:15 |
| ralsina | gotta go pickup the kid and then I babysit, so I will do some reviews late at night, mail me what you need! | 19:15 |
| jo-erlend | I was thinking that it would be cool for our Ubuntu loco to have a shared resource for things we make. It would be nice if I could simply share a folder with ubuntu-no and have it stored on a central server that we have. Is this acceptable use of the service? | 19:18 |
| barry | i just dist-upgraded my oneiric box and now it seems tomboy lost the ability to sync to u1. is this a known problem? | 19:21 |
| dobey | is it interview a bunch of u1 week or something? | 19:23 |
| jo-erlend | barry, no: https://wiki.ubuntu.com/UbuntuOne/Status | 19:23 |
| jo-erlend | dobey? | 19:24 |
| dobey | jo-erlend: we don't support sharing to launchpad teams | 19:24 |
| jo-erlend | oh, ok. | 19:24 |
| dobey | jo-erlend: you'd have to share the folder to everyone individually | 19:24 |
| dobey | if you sent a share invite to a mailing list, the first person to claim it would get the share, not everyone | 19:25 |
| jo-erlend | it's not a big deal. It'll probably make more sense to use ssh, but with Ubuntu One it'd be synced with the server automatically, making it available to others from the server. | 19:25 |
| dobey | barry: it is :( | 19:25 |
| dobey | barry: i think configuring any sync service is broken right now :-/ | 19:26 |
| barry | dobey: okay, i'll wait then | 19:26 |
| dobey | jo-erlend: i think your "?" was in response to my interview week comment; an interview with chipaca went up yesterday, and today an interview with aquarius just went up | 19:28 |
| barry | dobey: btw, the problem is that tomboy->preferences->sync has no u1 option | 19:29 |
| nessita | dobey: when you can, would you help me finding out if there is an easy way to create an empty dict in C? other than calling g_hash_table_new (which requires some funcs as args). I need an empty dict to pass in a DBus call, it will not be modified by anyone. | 19:29 |
| mandel | nessita, I just pushed the branch with all the changes you recomended. There was on of the tests you asked for missing, so it will appear in the diff and you will be able to find the other one easily | 19:29 |
| jo-erlend | dobey, oh, ok :) | 19:30 |
| dobey | nessita: C doesn't have dicts. and for glib-based code, you would need to create a new hash table, and destroy it when done using it | 19:30 |
| dobey | nessita: doing dbus stuff in C is quite atrocious really | 19:30 |
| dobey | meh, my laptop is so slow with oneiric :( | 19:31 |
| mandel | nessita, I did no manage to run u1lint in my o vm, so apoligies if there are any lint issues | 19:32 |
| * mandel considers porting u1dev tools to all the platforms he uses | 19:32 | |
| dobey | barry: oh i guess the main issue got "fixed" then; choose "Tomboy Web" as it is set up to sync to u1 by default | 19:32 |
| dobey | mandel: ? | 19:33 |
| barry | dobey: ah, okay. it's a little confusing since i had no idea tomboy web meant u1! | 19:33 |
| dobey | mandel: it works on linux and windows no? | 19:33 |
| mandel | dobey, yes, is lint the one that has problems | 19:34 |
| alecu | nessita, g_hash_table_new (g_str_hash, g_str_equal) | 19:34 |
| dobey | barry: yeah, we had to drop the previous nicer integration bits, because it just wasn't working well with the new tomboy and we didn't have resources to fix it for oneiric :( | 19:34 |
| dobey | mandel: oh, pylint is the problem you mean? | 19:34 |
| mandel | dobey, yes, that is the guy | 19:34 |
| nessita | mandel: ack | 19:34 |
| nessita | dobey, alecu: thanks | 19:35 |
| dobey | mandel: so u1lint is fine, but pylint is just cazzo? | 19:35 |
| alecu | nessita, or (NULL, NULL) would do for your case too. | 19:35 |
| mandel | dobey, u1lint works, and call pylint and the pylint is screwed up | 19:36 |
| barry | dobey: i'm not so sure this is working though. when it hits the browser, it asks for confirmation of the computer access with the computer name "None" (sans quotes). After changing that and clicking "Add this computer", i still can't save in the preferences panel | 19:36 |
| mandel | dobey, specially when it looks for paths etc.. | 19:36 |
| dobey | mandel: that's why i want to move everything to pyflakes | 19:36 |
| dobey | barry: hrmm, oh, so i guess that isn't fixed. :( | 19:37 |
| barry | dobey: is there a bug # i can subscribe to? | 19:37 |
| dobey | yes, but i don't recall what it is right now | 19:37 |
| barry | okay, no worries, thanks | 19:37 |
| dobey | barry: bug #845321 seems to be it | 19:38 |
| ubot4 | Launchpad bug 845321 in tomboy (Ubuntu) (and 1 other project) "Unable to setup Ubuntu One sync with Tomboy (affects: 11) (dups: 1) (heat: 60)" [Undecided,Confirmed] https://launchpad.net/bugs/845321 | 19:38 |
| barry | dobey: thanks. subscribed | 19:39 |
| mandel | dobey, we will get there, we have not been multiplatform for a long time, so we need to start getting things nicer for this new era :P | 19:41 |
| nessita | alecu: would a hash_table by G_TYPE_OBJECT when passing it thru C dbus lib? | 19:41 |
| alecu | nessita, would it what? | 19:42 |
| dobey | nessita: no | 19:42 |
| dobey | nessita: i don't think so anyway | 19:42 |
| nessita | dobey: any idea what type is? (or where to look) | 19:43 |
| dobey | nessita: there is some nasty stuff you have to do | 19:43 |
| nessita | alecu: when doing a dbus call from the C code, I need to pass an empty hash table ({}) to find_credentials | 19:43 |
| dobey | nessita: i'm looking for an example | 19:43 |
| nessita | alecu: so, from the rest of the code, I see that we're passing the app name using: | 19:43 |
| nessita | G_TYPE_STRING, SSO_APP_NAME, | 19:43 |
| nessita | so, I built my empty hash table, but I'm not sure what G_TYPE_FOO to use | 19:44 |
| alecu | no idea about that. | 19:45 |
| dobey | nessita: see line 237 in syncdaemon-authentication.c | 19:49 |
| dobey | nessita: line 237 in trunk that is; not sure what it is in your branch :) | 19:50 |
| dobey | syncdaemon_authentication_find_credentials (SyncdaemonAuthentication *auth) | 19:50 |
| nessita | dobey: that's the method I'm changing | 19:50 |
| dobey | in that function | 19:50 |
| nessita | dobey: we need to pass, besides the app_name, an empty "dict" | 19:50 |
| dobey | dbus_g_type_get_map(...), &hashtable | 19:50 |
| gatox | EOD here!! see you tomorrow! ( nessita hope with the defer complete :P) | 19:50 |
| nessita | gatox: nice! | 19:50 |
| dobey | err, i guess not &hashtable | 19:50 |
| nessita | dobey: but I think I understand what you're saying | 19:51 |
| * nessita tries | 19:51 | |
| dobey | nessita: you're changing the call to find_credentials? | 19:51 |
| nessita | dobey: we have to in order to use the new iface | 19:51 |
| dobey | nessita: ah ok; i see what you need to do now | 19:51 |
| nessita | I think I know | 19:52 |
| nessita | thanks to your pointer | 19:52 |
| nessita | dobey: testing now... | 19:52 |
| dobey | nessita: after SSO_APP_NAME, but before G_TYPE_INVALID, insert a new line, with dbus_g_type_map(...), emptyhash, | 19:52 |
| nessita | yeap | 19:52 |
| nessita | I had that but using G_TYPE_OBJECT, and that segfaulted :-) | 19:52 |
| dobey | yep | 19:52 |
| dobey | nessita: and at the end of the block where you created the empty hash table, you need to g_hash_table_destroy(emptyhash); also :) | 19:53 |
| dobey | (just so you know) | 19:53 |
| nessita | dobey: added already! :-) | 19:53 |
| nessita | and thanks, I could forgot that very easily | 19:53 |
| nessita | writing C is fun! | 19:55 |
| nessita | *not* | 19:55 |
| nessita | dobey: would you know which other app uses syncdaemon-authentication? | 20:00 |
| nessita | dobey: I'm looking and the api for find_credentials is so that app expects the credentials when that method returns | 20:01 |
| dobey | nessita: the music store maybe? | 20:04 |
| nessita | dobey: and tomboy, apparently? | 20:05 |
| dobey | nessita: you can run tests/test-music-store in libubuntuone trunk after building it to test the music store. you'll probably need the same LD_PRELOAD to point to the new libsyncdaemon though, unless you want to install it into /usr/local or something | 20:05 |
| nessita | dobey: ack, thanks | 20:05 |
| dobey | nessita: and tomboy, but i think we've pulled that bit out of tomboy for oneiric now anyway, for other reasons | 20:06 |
| Chipaca | dobey: interview with aquarius ? where? | 20:06 |
| nessita | dobey: right, I wasn't planning on moving this to O anyways... I was tackling this so we have the whole P cycle to confirm nothings get broken | 20:06 |
| nessita | dobey: we do have released everything for O, right? | 20:07 |
| dobey | Chipaca: i just saw it on planet ubuntu | 20:09 |
| dobey | nessita: yes; though there is a nasty crasher introduced by the new webkit that seems to only happen in our music store :( | 20:10 |
| Chipaca | dobey: ah, that's from last week :) | 20:10 |
| nessita | dobey: but we will be merging a "solution" to that, if any, in the stable branch, no? | 20:11 |
| dobey | Chipaca: oh, well, whatever. i saw it today | 20:11 |
| Chipaca | aquarius: ping, btw | 20:11 |
| nessita | dobey: I mean, I was counting on not getting any of these changes in O | 20:11 |
| dobey | nessita: well, i don't know if we have to do anything for libu1 | 20:11 |
| dobey | nessita: we probably won't be putting the changes you're working on right now, into oneiric, no | 20:11 |
| nessita | dobey: no please no | 20:12 |
| dobey | nessita: i don't see any reason to. | 20:12 |
| nessita | exactly | 20:12 |
| dobey | the libu1/webkit issue is totally unrelated | 20:12 |
| nessita | good (in a way) | 20:12 |
| dobey | nessita: and, i hope, we will be making some more organizational "drastic" changes for P anyway | 20:13 |
| nessita | dobey: not sure what you mean | 20:14 |
| dobey | nessita: moving libsyncdaemon into libu1 source tree and out of ubuntuone-client; making the tomboy stuff an actual plug-in that we own, pulling the banshee extension out into our own tree so that we can properly own it | 20:15 |
| nessita | dobey: ah, great | 20:15 |
| dobey | and we need to do some major refactoring on libsyncdaemon/libu1 as well, to make it asynchronous and fast | 20:15 |
| dobey | and move the sso bits out of libsyncdaemon | 20:15 |
| nessita | dobey: just wanted to confirm we're un sync with "the removal of the deprecated DBus service is not supposed to go to oneiric at all" | 20:16 |
| dobey | yep, would break the world | 20:16 |
| dobey | i wonder what day we will have the mumble meeting next week | 20:18 |
| nessita | dobey: you can propose one in the email thread | 20:18 |
| dobey | i don't want to propose one; i want to know when i need to collect my thoughts by :) | 20:19 |
| nessita | dobey: one last question to wrap this sso+libu1 thingy: everything is working, except that now syncdaemon_authentication_find_credentials should always return NULL, instead of an instance of SyncdaemonCredentials. If I would try to maintain the old API, where the credentials were returned, can I somehow connect to the CREDENTIALS_FOUND_SIGNAL in a blocking way? so syncdaemon_authentication_find_credentials does not return until that signal is received | 20:36 |
| nessita | (so I can grab the credentials from the signal and return that to the caller) | 20:37 |
| dobey | nessita: why would it return NULL? | 20:40 |
| dobey | nessita: it was calling dbus synchronously before. does the new interface on the ubuntu-sso-login side break the sync/async magic stuff in python-dbus? | 20:42 |
| nessita | dobey: becasue now find_credentials uses the new Dbus iface, which is fully async, so a call to sso dbus find_credentials returns None, and the creds are returned in the CRedentialsFound signals | 20:42 |
| nessita | signal* | 20:42 |
| dobey | nessita: maybe we should fix the new iface in sso to be compatible first then, and hold off on changing libsyncdaemon for now? | 20:43 |
| nessita | dobey: answer to your question is "yes", the non-deprecated iface (is not new... sadly we did not migrate sooner) does not return anything in any call | 20:43 |
| nessita | dobey: the new iface async, we can't make it wait for the credentials | 20:43 |
| nessita | again, new is the wrong word, is there since early natty | 20:44 |
| dobey | new as opposed to old | 20:44 |
| dobey | it's like buying a new car, that is used | 20:44 |
| nessita | right | 20:44 |
| dobey | still new to you :) | 20:44 |
| nessita | dobey: so, I can confirm that tomboy does connect to that signal | 20:45 |
| nessita | so tomboy is getting the credentials the same | 20:45 |
| nessita | but we should check the music store, I think... the test you mentioned is ins the RB tree? | 20:45 |
| dobey | no, in libubuntuone | 20:46 |
| dobey | but it sounds to me like we can't 'fix' libsyncdaemon yet | 20:46 |
| dobey | because the 'new' iface is totally incompatible | 20:46 |
| nessita | dobey: only for find_credentials... but if I could make the C code block waiting on the signal, we'll have the same result as before | 20:46 |
| nessita | (before, find_credentials was blocking the caller, and the dbus main loop until the credentials were retrieved) | 20:47 |
| dobey | nessita: the C cod was already blocking. the problem is the server changed to break the proper sync APIs | 20:47 |
| nessita | dobey: and that's why we waited 2 cycles to remove the deprecated API | 20:48 |
| dobey | nessita: and you can't use the async api, synchronously | 20:48 |
| nessita | dobey: well, yes, if I block waiting on a signal | 20:48 |
| dobey | you can't block waiting on a signal | 20:48 |
| nessita | dobey: why not? | 20:48 |
| dobey | because the language doesn't work that way | 20:48 |
| nessita | I mean, I can think ways of doing it ugly, with a loop | 20:49 |
| dobey | there is synchronous api, and there is asynchronous api, if you want blocking, you use the synchronous api | 20:49 |
| nessita | dobey: sso does not provide a sync api | 20:49 |
| dobey | nessita: sso making an explicit decision to break the way dbus works, does not change how the dbus C api works | 20:50 |
| nessita | ok, I'll try to build a sync find_credentials for libu1 to use | 20:53 |
| dobey | nessita: how about we just stop for today, and evaluate this a bit better tomorrow, and make a plan that tries to avoid breaking everything :) | 20:54 |
| nessita | dobey: well, since I was sure that we have this code migrated already (which happens to be False), sso deprecated iface is already gone | 20:55 |
| dobey | nessita: i don't know what the sso code is like exctly right now; but i know we can't just up and totally break the libsyncdaemon API/ABI and be done with it | 20:55 |
| dobey | nessita: ok, so nightlies are already broken | 20:55 |
| nessita | I remember we worked on migrated this code to the new iface, I don't understand how come we still are using the old iface... | 20:55 |
| nessita | anyways | 20:56 |
| nessita | yes, tomorrow | 20:56 |
| dobey | nessita: we really, as a team, need to make a lot of hard decisions soon i think; and i think how we best handle this is probably a result of those decisions | 20:56 |
| dobey | nessita: i'm guessing it wasn't migrated, to avoid breakig the API/ABI :) | 20:57 |
| nessita | dobey: makes sense, though I'm not sure what you're adding to the 'hard decisions' bag | 20:57 |
| dobey | nessita: we have been very bad at making decisioins on what we will and will not support with the latest versions of our code | 20:58 |
| dobey | nessita: an extension of my comments in that mail thread; but i didn't go there yet, because i decided i need to collect some thoughts still, to be able to state those concerns properly | 20:59 |
| dobey | but i guess i need to do that asap now | 20:59 |
| dobey | alright, well i am off | 21:07 |
| dobey | have a good evening all! | 21:07 |
| nessita | I'm off, byw all | 21:17 |
| === m_conley is now known as m_conley_away | ||
| alecu | mandel, https://code.launchpad.net/~alecu/ubuntuone-client/winshell-extension | 22:01 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
