=== m_conley is now known as m_conley_away === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away === eu is now known as Guest58987 [09:43] good morning everyone [09:48] uh, I forgot to say good morning... hola! [09:48] dpm, hello! [09:49] hey mandel [09:49] dpm, how is it going? [09:51] mandel, going well, recovering from some network issues... I was about to ask a question about u1, but then I thought better of it [09:51] dpm, lol [09:53] so I wanted to have a public share for anyone to write to, without having to know their e-mail and I wanted to ask if it's possible. But then I thought it might not be a very good idea if I don't want just anyone to delete all files in there... [09:55] Good day, all. :) [10:00] dpm, exactly, you need to know the email, there is no way around, mainly due to that reason [10:01] next time I'll try to think before asking a question, but it doesn't always work out... [10:01] JamesTait, can you access lp? [10:01] dpm, hehehe at least is between us and a public log, who is going to know ;) [10:02] mandel: No, looks like a maintenance page. [10:02] mandel: DB offline. [10:02] mandel, not that it'd be the first time I've made a fool of myself in public :) [10:04] dpm, yeah, tell me about it, and the internet has a great memory [10:04] JamesTait, is amazing how dependent I'm on lp for my daily work, it scares me [10:04] :) [10:09] yay! lp is back \o/ [10:29] Dia duit ar maidin #ubuntuone! [10:34] mandel: lp back for you now? [10:34] mandel: Oh, right. Ignore me. :) [10:50] alecu, early morning! [11:02] hola mandel! [11:31] hola gatox, buenos días! [11:31] alecu, buenas buenas [11:31] alecu, how are you?? do you need any review? [11:32] gatox, buenas! [11:32] mainerror, hi [11:32] gatox, not yet, thanks! [11:32] mandel, hi [11:32] grrrr autocomplete [11:33] gatox, hehehehe [11:33] gatox, I wonder what people think when we talk to them :P [11:33] jejjeej [11:33] I also have a very annoying issue with Tomboy notes [11:33] mandel, "if not path[-1] == os.path.sep:" -> "if path[-1] != os.path.sep:" ? [11:33] I really have a lot of notes there and synced with Ubuntu One [11:33] But for some reason I now get sync errors on two of my machiens, so I don't have the notes available [11:34] Is there something I can do? [11:34] d3ngar, hello [11:34] alecu, true, I moved from endswith to using [] so I stupidly left the not, will make that nicer [11:34] d3ngar, let's ask the right people. rye ping ^ [11:34] rye pong [11:34] reading [11:35] mandel, also perhaps endswith is the right thing... in some weird case where os.path.sep is more than a char long [11:35] d3ngar, what's the version of Ubuntu you are running? [11:36] alecu, hm.. well I did that following brians review.. where he pointed out that doing that is much more faster.. [11:38] d3ngar, in any case, try running the script from http://people.canonical.com/~roman.yepishev/us/tomboy-sync-validator.py - it should show whether all the notes are intact on the server-side [11:41] mandel, yes, it's probably faster because there are no dot lookups, and practically there's no OS where path separators are longer than a char. [11:42] mandel, but I don't think speed is critical in that bit of code, since we are doing .startswith and .endswith elsewhere. [11:43] alecu, indeed, I think we do that in os_helper at least, right? [11:43] mandel, we do .startswith two lines after the one above [11:45] (in _get_partial_child_path_dir ) [11:48] alecu, let me see [11:48] alecu, jest, but there we are nto looking for a single char, but the watch path [11:49] good morning! [11:49] ralsina, hi [11:50] hello gatox. We have the 1-1 pending, we'll do it this afternoon [11:50] ralsina, ok [11:51] ralsina, morning... how's tato doing? [11:51] mandel, self._subdirs[current_child + os.path.sep] = False [11:51] alecu: still fever [11:51] ralsina, hope he gets better :-P [11:51] alecu: but he's keeping food inside, which is a big improvement for everyoneinvolved [11:52] mandel, in the line above, is the list of subdirs being kept with just the subdir name or with the full path? [11:53] ralsina, i'm trying to create a bundle.... can i skip fetch (because i don't have bzr in the windows machine) and just do: python setup.py prepare py2exe? [11:53] do i need to do something manually for that? [11:53] gatox: no, because then you have no sources :-) [11:53] gatox: you could copy the checkouts you have on Linux into sources, and then use that [11:53] gatox: or get the tarballs and expand them there manually [11:54] mandel, or perhaps I'm not understanding the new Watches completely [11:54] gatox: but really? get bzr, even if you don't setup keys, then configure the right brancheson conf.py and live with it ;-) [11:54] ralsina, thanks [11:54] ralsina, ok..... i'll try with bzr :P [11:55] alecu, I'm sure you are understanding them, here is how it works, lets say we have a watch looking at C:\Uses\Mandel [11:55] gatox: I could do it myself, but I really want someone else to be able to do releases in case I am not available someday [11:56] ralsina, of course.... i want to learn too! [11:56] gatox: thanks [11:56] gatox: if it makes you feel better, it used to be *harder* ;-) [11:56] ralsina, i know i know jeje [11:56] alecu, there our watch will have self._path pointing to that dir and a var called _subdirs that contains a dict with {'path_to_child_1': True, 'path_to_child_2':False} [11:56] gatox: remember to set the conf.py to use the stable branches though [11:56] rye: how do I run this script? [11:57] alecu, the boolean value states if we are watching the child events or not, and also keeps track of the subdirs in case we have a delete so that we can add de IS_DIR mask [11:58] alecu, when we add a new child_watch, we add new child paths,. for example path_to_child_1\other_child: False [11:58] alecu, so that we filter on the next level of the path [11:58] d3ngar, are you using ubuntu? [11:58] mandel, ok, I kinda got that. But... [11:59] mandel, in your example above, 'path_to_child_1' is a full path or just a folder name? (a relative path) [11:59] yes! [12:01] alecu, always relative, since there is no need to keep the full path info and ReadDirectoryChangesW returns the paths as relative ones [12:01] mandel, great [12:02] rye: http://pastebin.com/eKhgRKjv [12:02] rye: I can also see the notes in the web-interface [12:03] d3ngar, ugh, could you please try running it as "python tomboy-sync-validator.py", not as a shell script [12:04] rye: http://pastebin.com/saNuapyx [12:05] rye: I see that a lot of this seems to be fine [12:05] Good morning! [12:05] rye: Sadly I only have 16 notes in Tomboy [12:05] d3ngar, yep, no errors. Are you using Ubuntu 11.10 ? [12:07] rye: that is correct [12:07] d3ngar, could you please quit tomboy now and run it as "tomboy" from the terminal? [12:07] d3ngar, wait [12:08] d3ngar, tomboy --debug [12:08] d3ngar, then please try synchronizing and see whether there is any message related to New Note Template [12:08] mandel, any([file_name.startswith(path) and not watched for path, watched in self._subdirs.iteritems()]) [12:08] hello everyone! [12:09] nessita, hi [12:09] mandel, that list comprehension inside the any can be surely replaced by a generator expression [12:09] mandel, just by dropping the [ and ] [12:09] hola nessita! [12:10] hola alecu! [12:10] alecu, you recon, cool [12:10] alecu, can you write those in the review so that I do not forget [12:10] mandel, and that means that a new list won't have to be constructed, and the any will be able to finish faster if the first item matches the expression. [12:10] mandel, sure [12:10] anyone needs a review? [12:10] rye: http://pastebin.com/zs7XkGap [12:11] bug #848250 [12:11] Launchpad bug 848250 in tomboy (Ubuntu) (and 1 other project) "Tomboy sync fails on Oneiric, New Note Template already exists (affects: 35) (dups: 1) (heat: 186)" [Undecided,Confirmed] https://launchpad.net/bugs/848250 [12:15] gatox: will start the reviews now [12:16] rye: was the bug related to me? [12:17] d3ngar, yes, i am now checking what can be done [12:18] d3ngar, in theory, if you open tomboy and edit the title of template called "New Note Template" to be something else, it will allow the sync to proceed. This is Tomboy bug though, so it manifests on all sync backends (files/ssh/webdav) [12:20] nessita, ok..... i checked this morning that there weren't any conflicts [12:20] gatox: did you try them all IRL? (today or yesterday) [12:20] rye: that's a good idea!~ thnks [12:21] nessita, yesterday [12:21] all of them [12:21] gatox: great [12:24] rye: sadly it seems Tomboy is hanging / crashing [12:24] d3ngar, does it print anything to the terminal? [12:24] rye: it just did [12:25] rye:weird [12:25] But stopped again [12:25] I think I kill it? [12:27] rye: now it doesn't do anything anymore. Last debug message is :Autosync pref changed...restarting sync timer [12:28] d3ngar, and it did not sync, right? [12:29] StickyNoteImporter: Sticky Notes XML file does not exist or is invalid [DEBUG 12:28:15.099] Unable to locate 'wdfs' in your PATH [12:29] wdfs? [12:29] d3ngar, that's ok, it is searching for webdav fs connector which you are not using [12:30] rye: but why doesn't it respond at all? [12:30] hm [12:30] It's still open, but I can't interact with the GUI at all9 [12:30] d3ngar, then it locked up, you can terminate it with ctrl+c in the terminal [12:31] Did three times already [12:33] should I remove it, purge it and re-sintall? [12:33] d3ngar, then "killall tomboy" in another terminal should kill it [12:34] d3ngar, installing and removing applications in ubuntu are system-wide, most likely this will not change anything, all the settings that can be purged are in user's home directory but this bug is not yet fixed in tomboy so this renaming may or may not work [12:35] rye: why does ubuntu store so many settings in this couch-db nowadays? [12:35] used to be that tomboy had it's own settings folder [12:35] but this is gone, afaik [12:35] d3ngar, actually ubuntu does not store any settings in couchdb now [12:35] oh [12:36] d3ngar, tomboy store the data in ~/.local/share/tomboy folder [12:36] alright [12:36] So if I wanted to reset, I could delete that [12:36] and settings are most likely in gconf that can be accessed via gconf-editor [12:36] But it looks like after killing the process ID, it worked [12:37] d3ngar, well, yes, but in this case you need to reset a template only, not all other notes [12:37] Done already [12:37] I restarted and it looks as if notes are now working as expected! [12:37] Thank you so much! [12:37] d3ngar, oh, so the sync completed successfully? [12:37] gatox: are the linux tests passing for you with the fix-home branch? [12:38] nessita, yes.... let me check again [12:38] gatox: I'm getting this: http://pastebin.ubuntu.com/809586/ would my env have something wrong? [12:38] rye: yes [12:39] great! [12:39] Yes! [12:40] nessita, let me check if i can reproduce that here [12:42] gatox, ralsina can i have a super simple review, I needed to get my head out of proxy a little and did this: https://code.launchpad.net/~mandel/ubuntuone-client/skip-lint/+merge/89232 [12:42] mandel: sure [12:42] mandel, on it [12:43] ralsina, that should solve the garbage we have in jenkins and will make your life easier when running tests on windows since the buffer wont be used to throw at you the lint errors [12:43] mandel: cool [12:43] ralsina, I needed some context switch away proxy and it was easy enough :P [12:44] * gatox sees mandel branch and cries..... good one [12:44] jeje [12:44] gatox, yeah, that %* is an asshole :P [12:45] mandel: line 27, shouldn't that be set PARAMS=%* ? [12:46] mandel: as in "shift removed %1 from %* so we reset PARAMS to the remaining options"? [12:47] ralsina, %* is an asshole and will not be affected by the SHIFT, so you SHIFT and remember the next [12:47] ralsina, if you do PARAMS=%* it will have all params.. [12:47] mandel: then I don't see how this works at all :-/ [12:48] ralsina, ok, first we shift, so that we forget /skip-lint [12:48] mandel: OTOH, we have exactly one possible parameter for the script so maybe it just works ;-) [12:48] yes [12:49] so now %1 is what was the %2 before [12:49] ralsina, exactly [12:49] then you put the new %1 in PARAMS [12:49] and shift again [12:49] ralsina, then we ask, hey, is %1 == [] that is, do we have more params or not, if we don't lets move on [12:50] ralsina, if we do, that is %1 != [] we add the next one [12:50] mandel: SET PARAMS=%PARAMS% %1 will only keep 2parameters [12:50] %PARAMS% contains theoriginal %2, and%1 has what was %3 [12:51] if there was a %4 it's discarded [12:51] and if /skip-lint is not the first parameter it's ignored [12:52] mandel: unless I really don't get how this works (which wouldn't hurt my feelings a single bit) [12:52] ralsina, so, %PARAMS% gets incremented by %1 in every loop iteration [12:53] mandel: theright way todo this is do a FOR and process %1 on each iteration [12:53] nessita, yep..... i reproduce it..... :S fixing that..... weird! [12:53] mandel: like this: https://pastebin.canonical.com/58399/ [12:53] gatox: ack [12:54] mandel: and in the loop, if %1 is empty, jump to CONTINUE [12:54] mandel: it's actually simpler than your code, too [12:54] mandel: I mean not a FOR loop, of course, but "manually emulating a for loop" ;-) [12:54] ralsina, then, its the same :) [12:54] mandel: no, this one is readable;-) [12:55] mandel: you are only processing /skip-lint if it's %1 [12:55] ralsina, yes, I'm not going to let people do evil things like run-tests -t /skip-lint test_name [12:55] mandel: then why bother looping? [12:56] ralsina, 'cause I need to copy all the params, otherwise, how do i do it? [12:56] mandel: %* [12:56] ralsina, that is wrong [12:57] ralsina, shift does not affect %* [12:57] mandel: if you want to copy allparameters, why shift them? [12:57] mandel: just todiscard /skip-lint ? [12:57] ralsina, yes, because if we pass /skip-lint to u1trial is going to say WTF?!!?! [12:58] mandel: one "good" thing about BAT files, there is only %0 ... %9 ;-) [12:58] ralsina, that is why they invented SHIFT :P [12:58] So, check if %1 is /skip-lint. If it is, pass %2 .... %9 if it isn't, pass %* [12:58] mandel: yes, but this code deserves at least a comment ;-) [12:59] Something like "/skip-lint can only be used as %1" and "copying all parameters to pass to u1trial" [12:59] ralsina, yes, although the fact that you only have up to %9 does not mean that you can only pass 8 params, %0 is the batch name itself, it means you can only access 9.. [12:59] ralsina, I can add that :) [12:59] mandel: ok, that's enough for me ;-) [12:59] cool [13:00] ralsina: any reason why you use relative imports in the unique app module in the control panel? [13:00] nessita: no reason. Habit? [13:01] nessita: Iwould evensay "bad habit" [13:01] ralsina: ack, will propose a branch that also fixes lots of lint issues [13:01] nessita: lots? That got merged, didn't pylint catch those? [13:02] ralsina: well, we're not running the suite in tarmac using precise [13:02] nessita: ok [13:03] ralsina, I will also add some indentation to that file to see if I make it readable... [13:03] mandel: thanks. [13:08] mandel, added some comments to the proposal [13:09] * alecu needs to reboot, ttyl [13:12] nessita, could you run the tests for my u1-cp branch?? i'm able to run the tests for u1-cp in trunk... but i get a failure of u1-dev-tools when i try to run the tests in my branch [13:14] gatox: I can run them using ./run-tests -qt [13:14] gatox: what error is devtools giving you? [13:14] nessita, weird :S [13:14] nessita, http://paste.ubuntu.com/809617/ [13:14] gatox: your directory name is too long [13:15] quack..... [13:15] changing that [13:15] gatox: that means, the directory you're running the test into is too long, you need to make it shorther [13:16] nessita, thanks! [13:16] :-) [13:16] nessita, grosa! :P..... i've just check, and all the test are ok there [13:17] fixing u1-client-user-home [13:17] gatox: ack [13:23] alecu thx I'll fix them right now [13:27] ralsina, I pushed a branch which makes the entire thing more readable [13:27] mandel: cool, checking [13:27] ralsina, did you try the batch IRL ? [13:28] mandel: not yet [13:29] ralsina, can you please, it seems that gatox cannot get it to work while I have no problems [13:29] mandel: ack [13:30] thx [13:30] mandel, ok..... running from C: ignores the -t too..... and in a while i'll tell you if the /skip-lint too [13:31] gatox, in this the vm with the funny username? [13:31] mandel, no..... the username is gatox [13:31] gatox, then I don't understand, I've run it several times, which os and cmd are you using? [13:32] will do some network testing in linux, brb! [13:33] mandel, win7 Ultime SP1...... and the command: http://paste.ubuntu.com/809628/ [13:33] gatox, I mean, you use cmd.exe nothing diff, right? [13:33] mandel, ah yes..... [13:34] gatox, lets wait for ralsina to run it, maybe is my system.. [13:34] mandel: with /skip-lint I do get "skipping style chekcs" [13:34] gatox: what is theproblem exactly? [13:35] grrr why it fails here?! [13:35] ralsina, with the typo, or without it? [13:35] mandel: no typo [13:35] gatox, problem is, is not the first time it happens this in your system.. [13:35] mandel, ok.... i know where the problem is [13:35] ralsina, \o/ I though I did it :) [13:36] it should work now.... [13:37] gatox, what was the issue? [13:37] yep.... working [13:37] the issue was: me [13:37] forgot to merge with your branch [13:38] gatox, lol [13:38] gatox: chem... if I disable/enable the network connection in linux, syncdaemon will no get those changes and will not update its status [13:38] nessita, u1-client-network-detect? [13:38] gatox: yes [13:38] gatox: does it work for you? [13:39] nessita, yes.... that's why i change the code in network_manager (the linux version) to don't disconnect the signal... and that was working [13:40] gatox: you can tell by running u1sdtool -s, the connection status should go from With User With Network to With User Not Network. does that happens for you? perhaps I need to point syncdaemon to the sso trunk? [13:41] nessita, the change was made in sso to support that..... i don't know if you have the last version of sso installed in your system [13:41] gatox: would you know what revno has that change? [13:41] nessita, let me check [13:42] ralsina, thx for the +1 once is merge we can let sidnei know, it should make things better on the jenkins side [13:42] mandel: sure [13:42] gatox: I'm only missing one revno and is alecu's and-post branch:nessita@dali:~/canonical/ussoc$ bzr revno trunk/ [13:42] 835 [13:42] nessita@dali:~/canonical/ussoc$ apt-cache policy ubuntu-sso-client [13:42] ubuntu-sso-client: [13:42] Installed: 3.1+r834-29~precise1 [13:42] Candidate: 3.1+r834-29~precise1 [13:43] nessita, ok..... let me check with u1sdtool..... i test it looking at the logs [13:44] gatox: no rush [13:44] nessita, in the logs was very easy to see when the network state change [13:44] gatox: would you first confirm that sso trunk has the change you mention? because I'm seeing this: [13:44] in the console itself [13:44] def call_result_cb(self, state): [13:44] """Return the state thru the result callback.""" [13:44] if self.state_signal: [13:44] self.state_signal.remove() [13:44] mmmmm that shouldn't be there [13:44] gatox: that's in trunk [13:46] ok, lunch time, cu! [13:48] nessita, i forgot to propose that branch because it was a really small change.... proposing now [13:49] gatox: ack (Please make sure it has tests! :-)) [14:03] alecu: mumble? [14:04] ralsina, joining mumble === m_conley_away is now known as m_conley === facundobatista is now known as x-ipv6 [14:33] I need to go off line a few mins [14:47] * mandel back [14:49] ralsina, gatox can you review the following: https://code.launchpad.net/~mandel/ubuntu-sso-client/skip-lint/+merge/89257 is similar to the last review you did so that we can ignore lint issues in jenkins or when working on linux [14:49] at least until I fix pylint on windows [14:50] mandel, ok [14:50] gatox, thx, I want to make working on windows lesss of a PITA if possible :) [14:50] mandel :) [14:51] gatox, ralsina same here: https://code.launchpad.net/~mandel/ubuntuone-control-panel/skip-lint/+merge/89260 yes in the control panel we do not foward %* to u1trial so is much more simple [14:51] briancurtin, morning! how is the the first week with us going? [14:51] mandel, what happend with the other branch? [14:51] the previous one [14:51] gatox, which other bracnh? the already reviewwd? it has been merged [14:52] gatox, I want to talk with sidnei to set jenkins to use that, and maybe point be to how we can get a jenkins irc bot telling us we broke windows :) [14:52] mandel, ah.... i saw the same description and i thought that maybe you delete the previous one or something [14:52] mandel: its good. getting started on a new project is always a little slow at first, but it has been picking up speed as more and more things start working [14:53] briancurtin, nice! it is indeed a terrible amount of work, at lest when I started we just had linux, which made it easier hehe [14:53] briancurtin, if I can help with anything let me know [14:54] briancurtin, do you have mumble setup, AFAIK we have a spoken standup today [14:54] mandel: yep, i have it all configured and i'm logged in now [14:57] mandel: +2 [14:58] ralsina, thx [14:58] mandel, +1 on https://code.launchpad.net/~mandel/ubuntu-sso-client/skip-lint/+merge/89257 [14:58] reviewing the other one [14:58] ralsina, I should be able to mumble, but I have to be on 3g because the eng is at home fixing some crap about the line... [14:59] ralsina, are you there so I can test it works? [14:59] mandel: sure [14:59] mandel, +1 on this one: https://code.launchpad.net/~mandel/ubuntuone-control-panel/skip-lint/+merge/89260 too [15:00] gatox, ralsina thx! [15:00] call will start 3 minutes late or so because I have to kick a kid out of the office ;-) [15:00] ralsina, I tried to contact sidnei with no luch, if you have the time, can you let him know about the new commands? [15:00] * mandel kicks the dog to the terrace [15:01] mandel: I would prefer to wait until briancurtin has the new buildout thing working well, so we can have a repeatable environment for testing [15:01] mandel: so we don'tmakehim setup things twice [15:01] ralsina, mandel: you in a private call? [15:01] nessita: no, we tested his mumble [15:01] ah :-) [15:02] alecu: mumble? (weekly team meeting) [15:02] dobey: hello there! mumble? [15:03] joshuahoover: you wanted to be on this call :) [15:03] ralsina: i'm on [15:03] :) [15:03] oh yes [15:03] joshuahoover, booh [15:05] mandel: booh? is that spanish for boo? ;) [15:05] boo-urns [15:06] joshuahoover, I don't pronouce the h :P [15:07] alecu: mumble [15:07] uhhh [15:07] This channel remain this quite? [15:07] sak3t: eh? you joined like 30 seconds ago [15:08] LOUD NOISES [15:08] dobey, thanks for reminding me [15:08] it's really loud here, you just can't hear the music [15:08] yup..... not habitual to such silent chat rooms!! [15:08] actually am new to IRC [15:10] clc [15:19] briancurtin, FYI https://launchpad.net/qt4reactor [15:30] joshuahoover, u1-proxy [15:30] thanks guys! [16:01] cold [16:02] alecu, so, regarding the dialog, what did I say wrong or did not understand, I mean, we lauch the dialgo from sso but is a diff process, right? === JanC_ is now known as JanC [16:03] ralsina: so, where is this crash with ntfs bug? [16:03] dobey: let me find it, I amin another call :-/ [16:03] ok [16:12] nessita, ralsina so, I'm talking with briancurtin and we are wondering in which project we should be adding the bootstrap work he is doing, I'm between ubuntuone-client and ubuntuone-windows-installer? [16:12] briancurtin and I think is better on the installer, but is not bad to ask [16:13] I would say in -installer, in scripts [16:13] nessita, briancurtin: ^ [16:13] works for me [16:14] mandel: I would also say installer [16:14] ack [16:16] https://code.launchpad.net/~dobey/rhythmbox-ubuntuone/fix-setup/+merge/89274 [16:17] much smaller branch to review. pretty trivial really :) [16:18] alright, i'm off to get some lunch. bbiab [16:20] mandel, "we launch the dialog from sso, but it's a different process"... that's fine if we are talking about sso trying to get the proxy credentials. [16:22] mandel, and it will be done in a similar way to how the sso "backend" starts a new process asking for the UI stuff [16:22] mandel, but... I'm not sure this all makes sense when we are using the webclient module from the other applications. [16:23] gatox: mumble? #catchingupon1-1s [16:23] ralsina, ok [16:24] mandel, in that case we should think if doing something via ipc is a better solution [16:25] alecu, it does indeed sound bad [16:26] alecu, the deal is, having all that work in sso sounds a little strange [16:27] alecu, nevertheless I think it makes more sense to talk about it with some more code in it [16:28] mandel, if do "mv ubuntu-sso-client ubuntuone-credentials-manager" does it sound better? [16:28] mandel, because it will end up storing and providing proxy credentials for all of our other parts. [16:29]