[03:28] <Brewster> hey I'm looking to sync a folder on ubuntu one via the command line
[03:28] <Brewster> would anyone happen to know the command off the top of their head?
[03:45] <mattgriffin> Brewster: u1sdtool -h
[03:45] <mattgriffin> Brewster: --create-folder=PATH  Create user defined folder in the specified path
[03:45] <mattgriffin> i think
[03:45] <Brewster> cool
[03:45] <Brewster> thanks
[03:46] <mattgriffin> np
[03:58] <Brewster> cya
[04:22] <MonkeyDust> not sure if it is important to report: when clicking Explore in the ubuntu one gui, pcmanfm opens a new tab, but nautilus opens a new window
[04:45] <MonkeyDust> also, I cannot view files I published using U1
[09:16] <JamesTait> Good morning all! :)
[11:05] <gatox> good morning
[11:06] <mandel> gatox, morning!
[11:06] <gatox> mandel, hi
[11:07] <gatox> mandel, how are you?
[11:07] <mandel> gatox, fine, making u1 on darwin work nicer and nicer :)
[11:07] <gatox> mandel, :D
[11:08] <mandel> gatox, I've noticed that we might have a problem with the inhibitor, which is a PITA
[11:08] <mandel> gatox, all the rest works and I already merged the work with current trunk
[11:08] <mandel> gatox, we should land your branches asap
[11:08] <gatox> mandel, yes please! i fix everything you and alecu say yesterday!
[11:09] <gatox> mandel, great review yours
[11:09] <mandel> gatox, I'll take a look now to darwin3, are 3-4 the last ones?
[11:10] <gatox> mandel, i couldn't sleep very well last night..... a lot of drunk people shouting..... ii hope you don't do the same in your partys
[11:10]  * gatox is old
[11:10] <gatox> mandel, darwin4..... is the last one.....
[11:10] <gatox> i'll have 2 more..... but they are for really specific bugs, not part of the whole port
[11:11] <mandel> gatox, ok, lets try to land those 2 today
[11:12] <mandel> so that we can merge mine asap and talk with mmcc about the packaging
[11:12] <mandel> if we have something for this week it means we are one week early to the expected deadline :)
[11:12] <gatox> ohhh darwin2 landed! yey!
[11:14] <mandel> gatox, by the way, can you take a look at https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/add-path-no-user/+merge/112324
[11:14] <gatox> mandel, also.... let me know if you need any reviews..... now i'm working on bugs for fsevents...... not being crazy refactoring a lot of lines :P
[11:14] <gatox> thattttttt
[11:14] <gatox> jeje
[11:14] <gatox> ok
[11:14] <mandel> gatox, there was a problem connecting to the daemon, disconnecting and connectig
[11:14] <mandel> but is solved in that branch :)
[11:18] <gatox> mandel, +1........ also (if you want) take a look at this: http://youtu.be/tA_BJBLdduQ :D
[11:18] <gatox> mandel, should i mark it as globally approve? or do you want more reviews for that branch?
[11:22] <mandel> gatox, lets get a +1 from mmcc
[11:23] <mandel> gatox, believe it or not I read your blog, I already watched that video :)
[11:23] <gatox> mandel, ack
[11:23] <gatox> mandel, jeje imposible! i added this video 2 mins ago jejeje
[11:23] <mandel> gatox, is there another about code completion the?
[11:23] <gatox> the previous one was aabout functions returns..... this is code completion working with inheritance
[11:23] <mandel> s/the/then
[11:23] <mandel> ah!
[11:24] <gatox> there are 3 code completion videos :P
[11:24] <mandel> I'll take a look then
[11:24] <gatox> mandel, jejeje no pressure!!..... i'm just really excited about it :P
[11:25] <mandel> gatox, don't hate me, but there are some needs fixing
[11:25] <mandel> gatox, most of them are just stupid things, but the one about del_watch is not
[11:25] <mandel> sorry..
[11:27] <gatox> mandel, no problem!
[11:27] <gatox> i'll take a look
[11:28] <mandel> gatox, I have the fact that we don't have some pylint error when there is a @defer.inlineCallbacks and no yield
[11:29] <mandel> gatox, having said that, if the yield is just for super which is the last line, you can not use the decorator and do return super().blah() because it will return a deferred like the inlineCallbacks will do
[11:29] <mandel> less lines of code :)
[11:33] <mandel> gatox, does the comment make sense?
[11:34] <gatox> mandel, yappppp
[12:02] <alecu> und... hallo!
[12:03] <gatox> alecu, hi!
[12:03] <mandel> alecu, hello
[12:04] <mandel> alecu, I'd like to talk about the inhibitor implementation on darwin
[12:04] <mandel> alecu, I have seen that when the machine logs out u1 stops and seems not to resume the uploads, does that sound possible
[12:04] <mandel> ?
[12:05] <alecu> mandel: it sounds possible, yes.
[12:05] <alecu> mandel: logout or suspend?
[12:06] <mandel> alecu, I think is suspend, I left the machine over night
[12:06] <mandel> alecu, but I can test with both
[12:06] <mandel> alecu, if that is the case we might need to implement the inhibitor, is that correct?
[12:06] <alecu> mandel: "suspend" is more likely to affect us, yes.
[12:06] <alecu> mandel: no: we should not inhibit suspend
[12:07] <alecu> mandel: the inhibitor is for log out: we warn the user that some files are still being synchronized
[12:07] <mandel> alecu, hm.. ok I'll test that then
[12:07] <alecu> mandel: perhaps we can inhibit "idle" suspend... but we would be chewing a lot of battery...
[12:08] <alecu> mandel: "idle" suspend as opposed to "closing the lid" suspend.
[12:08] <mandel> alecu, anyways, atm I'm a little block on gatox branches landing and talking with mmcc about packaging (need to start the daemon)
[12:08] <mandel> alecu, the rest seems to all work ok with no problem in all the IRL tests I've done
[12:11] <alecu> mandel: those are greaaaat news!
[12:12] <mandel> alecu, and the mem consumption is not bad, I guess I don't not what the virtual mem means in the process monitor from mac 'cause firefox also uses 2 gb, so we look fine syncing 15 gb of music
[12:12] <mandel> alecu, including moving from one wireless to another, changing to 3g etc.. lo puetee todo lo que supe
[12:13] <gatox> back again (internet issues)
[12:32] <mandel> I'm off to have lunch
[12:32]  * mandel lunch
[12:32] <ralsina> good morning!
[12:32] <gatox> ralsina, hi
[12:34] <mandel> ralsina, he, just when i was leaving :)
[12:36] <ralsina> mandel: we are in sync :-)
[12:36] <mandel> ralsina, apparently hehe
[12:36] <mandel> gatox, please let me know when you are done fixing the last needs fixing and I'll take a second look
[12:36] <mandel> now, I'm really off to eat
[12:37] <gatox> mandel, of course
[12:52] <gatox> mandel, this one has been updated: https://code.launchpad.net/~diegosarmentero/ubuntuone-client/darwin3-fsevents/+merge/111666 let me know when you get this message
[13:47] <mandel> gatox, ok, looking
[13:48] <mandel> gatox, there is a merge conflic
[13:48] <mandel> t
[13:49] <gatox> mandel, with darwin3?
[13:49] <mandel> gatox, apparently: Text conflict in ubuntuone/platform/os_helper/darwin.py
[13:49] <mandel> gatox, or so it believes launchpad
[13:49] <gatox> mandel, ok, merging with u1-client
[13:51] <gatox> mandel, resolved
[13:58] <mandel> gatox, ping
[13:58] <gatox> mandel, pong
[13:58] <mandel> gatox, so, you dont think is a good idea to move the def is_valid_syncdaemon_path(path_indexes=None): to darwin?
[13:59] <mandel> gatox, line 1146 of the diff
[14:00] <gatox> mandel, that is in darwin
[14:00] <gatox> mandel, this is what i see: http://paste.ubuntu.com/1062569/
[14:01] <mandel> gatox, sorry I mean 1196
[14:01] <mandel> gatox, we never use that in unix, do we?
[14:01] <gatox> mandel, mmmmm that should be there..... let me check
[14:02] <gatox> mandel, sorry..... copy, not cut
[14:02] <mandel> gatox, lol
[14:02] <mandel> :)
[14:02] <gatox> mandel, done
[14:02] <mandel> gatox, 26dd and then p
[14:03]  * mandel is going to make gatox learn vim 
[14:03] <gatox> ¬¬
[14:07] <mandel> gatox, I would have done return  super(WatchManager, self).del_watch(wd) instead of the decorator.. but yours works the same way  I suppose :P
[14:10] <gatox> mandel, i can change it if you want
[14:10] <gatox> no problem
[14:10] <mandel> gatox, do as you wish
[14:10] <gatox> mandel, i'll do it.. i know you want me to change that jeje
[14:10] <mandel> ;)
[14:11] <gatox> mandel, done
[14:11] <gatox> mandel, now let me merge with darwin4 to see if there isn't any conflict
[14:14] <gatox> mandel, ok, darwin4 is ready too
[14:30] <urbanape> ralsina: quit feeding the trolls
[14:31] <ralsina> urbanape: ha
[14:31] <mandel> gatox, running tests on win and linux, i might have some comments about 4 due to some fears I have about this: self._path = os.path.abspath(path)
[14:31] <ralsina> urbanape: I get coffee, I throw a troll a nugget
[14:31] <mandel> ralsina, he, i was trolling a little too :P
[14:31] <mandel> is waaaaay too much fun hehehe
[14:33] <mandel> gatox, why sis you remove the abspath ?
[14:33] <gatox> mandel, diff-line?
[14:34] <mandel> gatox, 733
[14:35] <gatox> mandel, ah yes
[14:36] <gatox> mandel, that is being applied in darwin and windows..... because of the way paths are created with abspath, in windows i needed to apply that before calling the function, and on darwin after
[14:37] <mandel> gatox, hm.. funny, why do you have to call it after on darwin?
[14:37] <mandel> gatox, I just want to know the reasoning to understand the change :)
[14:37] <gatox> mandel, it was causing some problems with the handling of the paths as key because of the last /
[14:38] <mmcc> hi guys, catching up...
[14:38] <gatox> mmcc, hi
[14:38] <mandel> gatox, what do you mean?
[14:38] <mandel> mmcc, I'd love to chat with you later about the daemon and how to make it in the bundle to start when the machine boots
[14:39] <mandel> mmcc, that is one of the last pieces + gatox branches :)
[14:40] <gatox> mandel, sorry..... it is like this:
[14:40] <gatox> on darwin i need to call it first....... BECAUSE:
[14:40] <mmcc> mandel, ack. after standup?
[14:41] <gatox> mandel, we are adding in common.py the / at the end..... but abspath in darwin, removes that..... so, i want the abspath, but i don't want the separator at the end that common.py adds because it needed to compare it with ignored paths, and so get removed
[14:41] <gatox> so in darwin i get the abspath first
[14:41] <mandel> mmcc, super, I'd like to add it to the bundle with no launchd support for the fd and later add that, but indeed after the standup :)
[14:41] <dobey> wait
[14:42] <dobey> there's something on reddit that *isn't* spam?
[14:42] <mandel> gatox, uh uh uh, removing the / might be an issue
[14:42] <dobey> i want to see it
[14:42] <mandel> gatox, how do you compare it with an ignored path?
[14:42] <gatox> mandel, that's why i get the abspath first
[14:42] <gatox> so we don't have that issue
[14:42] <mandel> dobey, no, there is not, is a great troll competition hehehe
[14:43] <mandel> gatox, os.path.abspath removes the os.path.sep in all platforms
[14:44] <gatox> mandel, so we are using it in another way on windows?
[14:44] <mandel> gatox, so, no matter the os, the path wont have os.path.sep in path[-1]
[14:45] <mandel> gatox, only thing I know is that there we had the abspath with no os.path.sep in the end and you changed it, so I worry :)
[14:45] <gatox> mandel, i don't change it..... in window is doing the same as always
[14:45] <gatox> mandel, or you mean for darwin?
[14:46] <mandel> gatox, darwin
[14:46] <mandel> gatox, that paths is used to see if it is ignored, right?
[14:46] <gatox> mandel, ok..... let me check adding that in darwin again
[14:49] <mandel> gatox, but does it make sense that I'm worried?
[14:50] <gatox> mandel, yes...... but i want to check again why the test were requiring that...... so, let me try, and i'll give you a fresh answer
[14:50] <mandel> gatox, the behavior of os.path.abspath is the exact same in both platforms, is the abspath with no os.path.sep at the end, os it is a little fishy to say I just have to do it that one in a single case
[14:50] <mandel> unless fsevents is stupid, which is possible :)
[14:51] <gatox> mandel, that could be...... but i don't want to guess..... i don't remember exactly, so i'll check
[14:54] <mandel> gatox, I hate you for making me memorize your surname when branching
[14:54] <gatox> mandel, jejejeje
[14:54] <mandel> gatox, I just wanted to let you know.. I really really hate you for that
[14:55] <gatox> mandel, jejeje
[14:55] <dobey> mmcc: can you please change your formatting of commit messages on your merge proposals to use sentence form (start with capital, end with period), and if you're going to include bug numbers in the commit message please use (LP: #$BUGNUM) so launchpad parses it properly? thanks
[14:56] <mmcc> dobey, noted.
[15:00] <briancurtin> me
[15:00] <dobey> meh
[15:00] <thisfred> me
[15:01] <gatox> me
[15:01] <mmcc> me
[15:01] <thisfred> DONE: Bug #999562 TODO: still wrap up Bug #999562  BLOCKED: a tad, by libcurl peculiarities (it hangs when doing  t) NEXT:
[15:01] <thisfred> ouch
[15:01] <thisfred> plz ignore that :)
[15:02] <dobey> bad thisfred
[15:02] <thisfred> very
[15:02] <dobey> alecu, ralsina, mandel: stand-up
[15:02] <alecu> me (no notes)
[15:03] <mandel> me
[15:03] <ralsina> me (no notes)
[15:04] <briancurtin> ready?
[15:04] <dobey> aye
[15:04] <briancurtin> DONE: installer debugging, got XP working after manually installing VS2008 runtime in a clean VM
[15:04] <briancurtin> TODO: test a few more cases, probably put vcredist back into an installer step rather than (or in addition to) side-by-side installation
[15:04] <briancurtin> BLOCKED: none
[15:04] <briancurtin> NEXT: dobey
[15:05] <thisfred> dobey: define aye?
[15:06] <thisfred> DONE: Bug #999562 TODO: still wrap up Bug #999562  BLOCKED: a tad, by libcurl peculiarities (it hangs when doing  the same request twice even though that should work) NEXT: gatox
[15:06] <dobey> DONE: releases/uploads
[15:06] <dobey> TODO: SSO release/upload, u1db package?, check on u1-gnome bugs, check on getting stable PPA up to par
[15:06] <dobey> BLCK: None.
[15:06] <gatox> DONE:
[15:06] <gatox> Changes in my branches regarding the comments, reviews, working on moved from partial bug.
[15:06] <gatox> TODO:
[15:06] <gatox> Make sure that my branches land today :P. Keep working in the reamining fsevents bugs, start sleeping like a normal person.
[15:06] <gatox> BLOCKED:
[15:06] <gatox> No
[15:06] <gatox> mmcc, go
[15:06] <mmcc>  DONE: messy reactor fix on windows and darwin
[15:06] <mmcc>  TODO: discuss packaging fsdaemon, polish up CP reactor fix, setup new macbook-air
[15:06] <mmcc> BLOCK: none
[15:06] <mmcc>  NEXT: alecu
[15:06] <alecu> DONE: some mac reviews, misc debugging, setup of a Q partition, played a bit with py3k
[15:06] <alecu> TODO: more py3k
[15:06] <alecu> BLOCKED: no
[15:06] <alecu> NEXT: mandel
[15:06] <mandel> DONE: Reviews, reviews.. Fixed bug 1018319.
[15:06] <mandel> TODO: more gatox reviews. Adapt my code to gatox code. Talk with mmcc about adding the daemon to the packages.
[15:06] <mandel> BLOCKED: no
[15:06] <mandel> next, ralsina
[15:08] <ralsina> DONE: XP and Q setup (finally), reviews, cmake tweaking, mgmt call, canonicaladmin, minor stuff TODO: finish cmake for u1db, check windows progress, 1-1s, lots of minor stuff BLOCKED: no
[15:10] <ralsina> comments?
[15:11] <ralsina> EOM then
[15:13] <mandel> I have a comment, our manager is a troll :P
[15:13] <dobey> stop your blogspam!
[15:13] <mandel> a good one, but a troll..
[15:14] <dobey> also, someone really needs to turn off the voice synthesizer, if they want to stop hearing you
[15:14] <mandel> gatox, I think I'm ready to approve fsevents 3 at least tests pass and I cannot see any other problem, alecu  can you take a look asap
[15:15] <mandel> alecu, gatox branches do block me from landing my stuff and tom is my last day before pycon
[15:15] <mandel> dobey, lol
[15:15] <mandel> dobey, you are a master troll.. what a bloody team..
[15:16] <dobey> heh
[15:16] <mmcc> btw, I know of a looming problem with the fsevents branches, but we can fix it after they land - need to import reactor inside each function that uses it, importing it at the top installs the wrong reactor
[15:17] <mmcc> that'll be in platform/filesystem_notifications/darwin.py
[15:19] <mandel> mmcc, it does not, here we are lucky :)
[15:19] <mmcc> mandel, ?
[15:21] <mandel> mmcc, the deal is that only bin/ubuntuone-syncdaemon imports main after it imported the reactor
[15:21] <mandel> mmcc, so, when you import the reactor for the first time in the script, it will import the correct one
[15:21] <mmcc> mandel: the problem I'm seeing is with controlpanel's bin/ubuntuone-control-panel-qt
[15:22] <mandel> mmcc, yes, control-panel and sso are a diff worls, anyways, lets mumble talk about packaging and I explain better what I mean :)
[15:22] <mmcc> but syncdaemon probably won't have the same problem if it's a QCoreApplication - if you import qt4reactor before you create a QApplication, qt4reactor will create a QCoreApplication instead for you
[15:23] <mandel> mmcc, exactly, on win and darwin we use the select reactor and on linux the gi one
[15:23] <mmcc> mandel: for syncdaemon, you mean?
[15:23] <mandel> mmcc, yes
[15:26] <gatox> mandel, you were right about the abspath thing..... it seems that i get confuse with and old bug
[15:26]  * mandel high fives gatox!
[15:26] <gatox> mandel, o/
[15:26] <mandel> gatox, cool, one thing less to worry about :)
[15:26] <gatox> mandel, yep
[15:30] <dobey> alecu, ralsina: https://code.launchpad.net/~dobey/ubuntu-sso-client/update-4-0/+merge/112380
[15:30] <ralsina> dobey: got it
[15:32] <gatox> mandel, let me push the changes, because some tests were wrong
[15:33] <ralsina> dobey: +1
[15:33] <dobey> thanks
[15:34] <mandel> gatox, is that for 3 or 4?
[15:34] <gatox> mandel, 4
[15:43]  * briancurtin downloading vista right now, running for coffee
[15:45] <gatox> mandel, i'll let you know when 4 is ready..... i need to fix something because the / thing is breaking the ignored paths checks
[15:46] <mandel> gatox, so, it was a good catch the abspath thing or I made you miserable?
[15:46] <gatox> mandel, the second one
[15:46] <gatox> jejeje
[15:46] <gatox> payback for memorize my surname
[15:47] <mandel> lol
[15:54] <mandel> gatox, can you look at 3, I'm getting an error on windows
[15:54] <mandel> gatox, not related to your code, can you try and reproduce it?
[15:55] <gatox> mandel, running the tests......
[15:56] <mandel> gatox, ok, it might be my machine, just double check
[16:00] <gatox> mandel, yes, i have an issue with test_get_config......   and it's in trunk too
[16:00] <gatox> mandel, if that is what you see
[16:00] <mandel> gatox, yes... fuuuu
[16:00] <gatox> x2
[16:01] <mandel> gatox, I wonder when this happened
[16:01]  * mandel looks at jenkins
[16:03] <mandel> gatox, jenkins seems happy O_o
[16:03] <gatox> mandel, so...... you machina has became to the evil side like mine
[16:03] <gatox> machine
[16:03] <mandel> gatox, I'm forcing a build, lets see what happens
[16:05] <dobey> ok, off to lunch. bbiab
[16:07] <gatox> mandel, ok, darwin4 is ready too... just run the tests to see that i didn't brake anything.... all green
[16:07] <gatox> now lunch for me!
[16:23] <mmcc> heading out to lunch
[16:23] <alecu> gatox_lunch: "TODO: Implement this decorators to fix some encoding issues in darwin"
[16:24] <alecu> gatox_lunch: how are those supposed to be work?
[16:24] <alecu> ok, let's discuss this after lunch.
[16:24] <gatox_lunch> alecu, maybe i'm wrong, i thought mandel was working on that
[16:24] <gatox_lunch> mandel, am i right? or i understand something else
[16:25] <alecu> gatox_lunch: oh, are they coming in a later branch?
[16:25] <alecu> gatox_lunch: no problem then.
[16:25] <gatox_lunch> alecu, we should check with mandel.... i undeerstood that
[16:25] <mandel> gatox_lunch, I worked on listdir etc.. we have to see in which ecoding you get the things from the fsevents lib
[16:25] <mandel> gatox_lunch, we can write a simple test :)
[16:26] <gatox_lunch> mandel, ack, i'll take a look after lunch
[16:27] <mandel> I'm eod, but will be back later :)
[16:35] <alecu> gatox_lunch: I really don't like filesystem_notifications/test_darwin.py
[16:37] <alecu> gatox_lunch: there's so much stuff copied from test_windows that I think we should consider working on refactoring this while mandel goes to pycon.
[16:37] <mandel> alecu, I already mentioned it :)
[16:37] <alecu> mandel: awesome.
[16:37] <ralsina> alecu, mandel: can we do a refactoring *after* it's merged?
[16:37] <mandel> alecu, is a must TODO all tests are the same
[16:38] <mandel> ralsina, alecu, my point is, merge soon, propose mine, clean up while I eat pasta in italy
[16:38] <alecu> gatox_lunch: and I still don't like the time.sleep's around the tests...
[16:38] <ralsina> mandel: +1 on thatplan
[16:39]  * mandel smokes cigar
[16:39] <ralsina> "I +1 when a plan comes together" doesn't sound right.
[16:41] <alecu> mandel: what are your branches pending merge?
[16:42] <mandel> alecu, this needs a review: https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/add-path-no-user
[16:43] <mandel> alecu, and I'll propose lp:~mandel/ubuntuone-client/fsevents-daemon asap
[16:43] <mandel> alecu, which is when diego lands his
[16:43] <mandel> alecu, will be here late night (EU time) to try and propose
[16:51] <gatox_lunch> alecu, IIRC, we talk in mumble that we were going to do this way, and then refactor the tests
[16:52] <alecu> gatox_lunch: yes, I remember. But I'm still not keen on the idea of landing so much copypasted code.
[16:53] <gatox_lunch> alecu, so..... do you want me to change darwin4.... or propose a darwin5 for that?? or??
[16:54] <gatox_lunch> i'll finish lunch.... brb
[16:54] <alecu> gatox_lunch: and since mandel is going to pycon I wondered if we were rushing too much just to land his stuff
[16:54] <alecu> gatox_lunch: go, finish lunch, let's discuss laters.
[16:54]  * alecu will have lunch too.
[16:54] <alecu> sorry for interrupting your lunch so many times :P
[16:54] <mandel> alecu, I can always propose mine and make it land before :)
[16:55] <mandel> alecu, I have no problem with that at all
[16:55] <ralsina> alecu: gatox's fault for having lunch on his computer ;-)
[16:55]  * ralsina sips his soup
[16:55] <alecu> mandel: I branched lp:~mandel/ubuntuone-client/fsevents-daemon and I see that it has a lot of empty tests...
[16:55] <mandel> ralsina, is the only reason he fails the turing test..
[16:56] <mandel> alecu, uh, I did not push the tests, give me a sec
[16:56] <alecu> mandel: perhaps a push is missing
[16:56] <alecu> ditto
[16:56] <alecu> mandel: does it depend on any of the branches that gatox is working on?
[16:56] <mandel> alecu, none
[16:57] <mandel> alecu, and it is merged with current trunk
[16:57] <alecu> mandel: great. But does it touch the same files as gatox's?
[16:57] <mandel> alecu, no
[16:57] <alecu> mandel: then propose it! we can surely land it before the other one.
[16:58] <mandel> alecu, ok
[16:58] <alecu> mandel: one more thing: are you leaving tomorrow for pycon, or working tomorrow?
[16:58] <mandel> alecu, working, I'll be off on friday
[16:59] <alecu> mandel: then propose it, and let's discuss the order in which we land them tomorrow in our AM.
[16:59] <mandel> alecu, ok :)
[17:03] <gatox> back
[17:03] <gatox> so...... now i don't understand why mandel was waiting for my branches jejee
[17:04] <gatox> alecu, now when you finish your lunch.... what should i do with the remaining branches then?
[17:04] <mandel> gatox, because there is an overlap of code
[17:04] <mandel> gatox, alecu, I wanted to make things smaller once you added the code, if we don't we will have some duplicated code
[17:04] <gatox> mandel, you said it doesn't to alecu
[17:04] <mandel> gatox, it does not touch the same files.. which is diff
[17:04] <gatox> ahhh ok
[17:05] <mandel> gatox, I worked assuming the work you would do so that we did not step on each other :)
[17:05] <mandel> alecu, gatox, ralsina, mmcc: here it is: https://code.launchpad.net/~mandel/ubuntuone-client/fsevents-daemon/+merge/112393
[17:05] <mandel> I though it would be smaller, sorry
[17:06] <mandel> although, 1000 lines is normal for gatox  ;)
[17:06] <gatox> mandel, i'll review that
[17:09] <mandel> gatox, so, what I wanted to reuse form your code is the NotifyProcessor, which is one overlap in the code
[17:09] <alecu> gatox: I still accept the idea of refactoring the testing code after it lands on trunk.
[17:09] <ralsina> briancurtin: looks (thanks to rye) like the problem with the 7 build is that it's including two system DLLs in dist/ even though they are listed in dll_excludes in setup.py :-/
[17:09] <mandel> gatox, the monitor and the factory are not shared, and the tests are just for the factory
[17:09] <ralsina> briancurtin: can you check in 7 that in a clean build they are included and maybe delete them on build?
[17:09] <alecu> gatox: what I'm not really convinced is the "time.sleep" in the tests.
[17:09] <briancurtin> ralsina: that's wonderful
[17:09] <ralsina> briancurtin: good news are we only need resigning the installer
[17:09] <alecu> gatox: we should build some other mechanism, because that won't scale.
[17:10] <mandel> alecu, there must be a way to remove them, I know you helped me to remove it on windows and we used a deferred instead by waiting for a number fo events
[17:10] <briancurtin> ralsina: i'll look
[17:10] <alecu> mandel: exactly. gatox ^
[17:11] <gatox> alecu, the thing is.... we are getting those events...... but in some time..... for syncdaemon it will works as always, but for the tests, where they want to access to a specific value in a specific time, that tend to fail.... that's why the sleep
[17:11] <gatox> alecu, mandel can you point where to look at?
[17:11] <gatox> pretty please
[17:11] <mandel> gatox, so, if you say to the test, wait for 4 events, when done return them in a deferred callback
[17:12] <mandel> gatox, look at _perform_actions on windows
[17:12] <gatox> mandel, ack
[17:12] <alecu> gatox: we should be hooking for the signals of those events, and yielding on that, instead of time.sleep, because sleep always ends up giving malfunctioning tests.
[17:13] <mandel> gatox, here: http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntuone-client/trunk/view/head:/tests/platform/filesystem_notifications/test_windows.py#L167
[17:13] <alecu> ok, let's lunch now.
[17:13] <gatox> alecu, nono!! jejee
[17:13] <gatox> mandel, cool! thx..... i'll adapt darwin4 to that........
[17:13] <gatox> the neverending refactor......
[17:13] <mandel> gatox, does the function make sense?
[17:14] <alecu> gatox: can you do that in darwin3 instead? that's where the sleeps are
[17:14] <gatox> alecu, ack
[17:14] <alecu> gatox: (I'm assuming you are using bzr pipeline for all this)
[17:14] <dobey> welcome to darwinia
[17:14] <alecu> gatox: (if not, shame on you :-) )
[17:14] <gatox> alecu, nop
[17:15] <mandel> gatox, the TestHandler waits to receive a number of events, lets say 5, we we do get to 5 we yield, else we timeout and have a failing test
[17:15] <gatox> mandel, honestly...... i'll need to try and see.... i'm a little worried about macfsevents threads and stuff
[17:15] <mandel> gatox, you are not making me confident with that sentence hehehe
[17:15] <gatox> mandel, i'm not confident :P
[17:16] <mandel> gatox, che, just lie to me and say you are ;)
[17:16] <gatox> mandel, i'm not confident..... but i'm honest :P
[17:17] <gatox> mandel, ok..... stop distracting me :P i think i understand the function, i'll try that
[17:17] <mandel> mmcc, may I have a review for: https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/add-path-no-user/+merge/112324
[17:21] <briancurtin> ralsina: oh wtf, removing those does make it work
[17:21] <ralsina> briancurtin: yep
[17:21] <ralsina> briancurtin: but they should not be there, I excluded them manually! :-)
[17:23] <briancurtin> ralsina: i'll check on vista right now just to be 100% sure it's the same issue
[17:23] <ralsina> briancurtin: awesome
[17:23] <briancurtin> (i dont know why it'd be something else, but may as well look now that i have the VM)
[17:24] <gatox> ralsina, 1-1?
[17:25] <gatox> ralsina, i set the alarm to never forget again :P
[17:25] <ralsina> gatox: haha, sure
[17:26] <ralsina> gatox: mumble
[17:26] <gatox> ralsina, ok
[17:32] <briancurtin> ralsina: vista is ok. i will repackage the 3.0.2 installer and send it for re-signing and try to figure out why those DLLs went in
[17:32] <ralsina> briancurtin: you'll have to create a new RT for it
[17:32] <briancurtin> will do
[17:33] <ralsina> briancurtin: you can copy from the last one
[17:42] <mandel> ok, EOD for me, laters!
[17:42]  * mandel walks dog and gets ready for football
[17:45] <gatox> mandel, bye
[18:45]  * briancurtin2 i need a break, will be back shortly.
[19:04]  * briancurtin2 back
[19:23] <mmcc> Apple's strange new universe where you download the IDE from the App store and the command line tools are *optional*.
[19:23] <mmcc> and a separate download...
[19:34] <alecu> gatox: so, did the comment from mandel to wait for the right number of events made sense?
[19:34] <gatox> alecu, yes.... i'm trying to adapt the code to that
[19:34] <alecu> great
[19:35] <gatox> it wil require some........ CHA CHAN CHA CHAN! .......... REFACTOR jejeje but seems possible..... even with the macfsevents things
[19:36] <alecu> gatox: doh
[19:43] <dobey> wait_for_planetary_alignment()
[19:52] <rye> dobey: UniverseError: None
[20:08] <gatox> ok people..... eod for me.... and i need to get some sleep...... see you tomorrow!  :D
[20:08] <ralsina> bye gatox!
[20:08] <mmcc> alecu, ralsina, anyone: quick review of u1-client mac test running script tweak: https://code.launchpad.net/~mikemc/ubuntuone-client/fix-1013268/+merge/112422
[20:08] <ralsina> go to sleep for real
[20:08] <ralsina> mmcc:  looking
[20:09] <alecu> mmcc: looking
[20:09] <gatox> ralsina, bye..... stop trolling crazy people :P
[20:09] <gatox> they don'y diserve your time
[20:09] <ralsina> mmcc: global +1
[20:09] <mmcc> ralsina: thx
[20:09] <ralsina> gatox: I am generous that way
[20:09] <gatox> jejeje
[20:10] <ralsina> gatox: also, the u1db tests take 5 minutes here
[20:10] <ralsina> lots of dead time
[20:10] <gatox> ralsina, jejeje
[20:10] <gatox> ralsina, well, so..... enjoy the trolling :P
[20:19] <dobey> alecu, ralsina: https://code.launchpad.net/~dobey/ubuntuone-client/shebangs/+merge/112425
[20:20] <dobey> alecu, ralsina: https://code.launchpad.net/~dobey/ubuntu-sso-client/shebangs/+merge/112424
[20:20] <ralsina> dobey: got it
[20:21] <ralsina> dobey: +1 trivial
[20:22] <ralsina> and the same on the other one
[20:22] <dobey> thanks
[20:36] <ralsina> dobey: can I trade you with https://code.launchpad.net/~ralsina/u1db/cmake2/+merge/112392 ?
[20:36] <dobey> about to look at it
[20:36] <ralsina> dobey: landing that would require a tarmac reconfiguration, or a top-level Makefile that calls cmake which seems kinda ugly.
[20:37] <ralsina> dobey: cool, thanks
[20:37] <dobey> doesn't the toplevel already call cmake?
[20:39] <ralsina> dobey: the toplevel has no Makefile after this branch
[20:39] <ralsina> dobey: it's roughly in the same state as autotools would be before autogen.sh
[20:40] <dobey> hmm
[20:40] <mmcc> idle question - is there a 'less' or 'more' or 'cat' or 'open' or anything convenient in windows for me to view a file from cmd.exe?
[20:41] <ralsina> mmcc: yes, there is more
[20:41] <ralsina> mmcc: but I would not call it convenient
[20:41] <dobey> heh
[20:41] <ralsina> mmcc: let me find you a set of unixy tools
[20:41] <dobey> mmcc: notepad.exe
[20:41] <mmcc> ah, more, good enough.
[20:41] <dobey> or install mingw
[20:41] <mmcc> I really just wanted to dump the file. I also just noticed Notepad++.exe can be cmdlined
[20:41] <ralsina> yes, I would rather use notepad than more
[20:42] <ralsina> Notepad++ even has an option to be installed over notepad :-)
[20:42] <ralsina> also, http://unxutils.sourceforge.net/
[20:42] <briancurtin2> mmcc: i have gnuwin32 installed so anything i tell you might come from there...but i think on vanilla win32 you can type "edit foo" and it'll open some minimal terminal editor
[20:42] <ralsina> much smaller than cygwin
[20:43] <briancurtin2> mmcc: ah, "type foo"
[20:43] <briancurtin2> type is similar to cat
[20:44] <mmcc> :) thanks guys! more and type will do for now. edit didn't work, told me the system cannot execute the specified program, but I don't need to solve this now
[20:48] <dobey> ralsina: is there a rule to get rid of all the cmake-generated stuff?
[20:48] <ralsina> dobey: delete the build folder
[20:48] <dobey> ralsina: and if i did cmake .?
[20:48] <ralsina> dobey: he
[20:48] <ralsina> dobey: make clean should delete most of it
[20:49] <ralsina> dobey: but probably not all
[20:49] <dobey> but not enough. because if i make a directory and do cmake .. in it, nothing useful happens
[20:49] <dobey> make hulk-angry
[20:50] <ralsina> dobey: probably need to delete CMkeFiles too
[20:50] <ralsina> dobey: but not really sure
[20:55] <dobey> it fails if you don't do the cmake in a subdir :-(
[20:58] <dobey> it won't find u1db.h
[21:01] <ralsina> I may be able to fix that if you want
[21:02] <ralsina> so put a needsfixing and I'll look at it tonight
[21:03] <dobey> wow
[21:04] <dobey> i am glad i generally ignore reddit
[21:07] <mmcc> dobey: you looking at ralsina's thread there, or something else?
[21:07] <mmcc> (I also ignore reddit, and these days HN too... haven't really missed it)
[21:07] <ralsina> mmcc, dobey: feel free to ignore that one too, I just got there because they linked to a post of mine, and have a trolling issue.
[21:08] <mmcc> alecu, ralsina: I am interested in your comments on this fix for the qt4reactor vs. QApplication ordering issue. I don't love how this turned out and had at least one other way of doing it, but this changes the least code...
[21:09] <mmcc> https://code.launchpad.net/~mikemc/ubuntuone-control-panel/fix-1015825-reorder-reactor/+merge/112432
[21:09] <ralsina> mmcc: I'll check it but I have to EOD just about now
[21:09] <ralsina> have my son asking me to play power rangers tower defense ;-)
[21:09] <dobey> people on reddit who purport to be trolls by spouting off nonsense, give proper trolls a bad name
[21:10] <mmcc> ralsina: ok, no prob. it's quick to read, so maybe later. go play
[21:10] <alecu> mmcc: looking
[21:10] <ralsina> dobey: trolls these days, they have no skills!
[21:10] <dobey> they are just goblins
[21:10] <mmcc> I liked that troll about Tolkein ripping off Harry Potter the other day, that was good stuff
[21:11] <mmcc> I like a troll with some depth
[21:15] <dobey> i need to head off too
[21:19] <dobey> have a good evening!
[21:19] <mmcc> bye dobey
[21:28] <mmcc> back - irc died after dobey left. I think it was sad.
[22:06] <alecu> mmcc: sorry, I opened your branch and got distracted.
[22:06] <alecu> mmcc: I think that it looks fine
[22:07] <alecu> mmcc: I think it just needs some unittests for the stuff you've added in ubuntuone/controlpanel/gui/qt/main/__init__.py
[22:08] <mmcc> alecu, hmmm. all that's added is conditionally installing the reactor, unles I'm missing something, right?
[22:08] <alecu> mmcc: right
[22:08] <mmcc> right, so that conflicts with the reactor used in the tests...
[22:09] <alecu> mmcc: I think we have unittests for main() too.
[22:09] <mmcc> which is why I made it conditional
[22:09] <mmcc> an earlier version didn't have the install_reactor_darwin flag, and blew up the tests
[22:09] <mmcc> specifically, the tests die because we already have an installed reactor
[22:10] <mmcc> so I think there's no way to test this cleanly, but I'm certainly happy to be corrected
[22:10] <alecu> mmcc: I think we could move the import plus the install to a new function or method, and unit test the "if"
[22:11] <alecu> mmcc: and the unit test would patch this new method just to check if it's being called.
[22:11] <mmcc> oh wait, I'm being dense. can we patch qt4reactor? or does that not work since I'm importing it in there?
[22:12] <mmcc> maybe I don't need to import it there. I could import it at toplevel, and patch install to check that it's called
[22:12] <alecu> mmcc: I'm not sure if there's a simple way to patch the import...
[22:12] <mmcc> I don't think there are side effects to importing it at toplevel. I'll see...
[22:12] <alecu> mmcc: yes, that might work better
[22:12] <alecu> mmcc: wait
[22:12] <alecu> mmcc: I think that it would matter
[22:13] <alecu> mmcc: because on linux we don't use the qtreactor
[22:13] <mmcc> right, of course. well, I can only import it at toplevel for darwin...
[22:13] <alecu> mmcc: in fact, we are trying to get rid of the reactor on the gui modules
[22:13] <alecu> sorry, the gui apps.
[22:14] <mmcc> yes, I know, but unfortunately this ordering seems to matter more on darwin than on windows (which has the reactor installed in the control-panel-qt main script
[22:14] <mmcc> )
[22:15] <alecu> right.
[22:15] <mmcc> pretty ugly
[22:16] <alecu> we are pushing all this code to run on places where it was not thought it would run... windows, darwin.... so yes it's getting ugly fast :-)
[22:16] <mmcc> yep - well, that's life :)
[22:19] <mmcc> so I've changed it to import qt4reactor only for darwin at toplevel in gui/qt/main/__init__ and am adding tests to be sure we handle the reactor flag correctly. what's a good way to test that the reactor isn't imported (or installed) on other platforms?
[22:28] <alecu> mmcc: what about patching "sys.platform" ?
[22:30] <mmcc> alecu: if I patch sys.platform to = 'darwin', then the test will try to import qt4reactor on linux, right?
[22:30] <mmcc> I guess I could skip the test for linux...
[23:16] <mmcc> gotta go, will come back later tonight to clean this up
[23:32]  * alecu waves, too