=== StaffRingedSeal is now known as Myrtti === teknico_ is now known as teknico [10:35] facundobatista, rye, can I have a very simple review for a branch that fixes a couple of bugs: https://code.launchpad.net/~mandel/ubuntuone-client/reactor-new-line/+merge/112535 [10:36] mandel, yeap [10:38] facundobatista, thx, is a very simple branch [11:09] good morning [11:25] gatox, morning! [11:25] mandel, hi [11:25] gatox, and 'tiron de orejas' for not reading my descriptions ;) [11:26] mandel, i read it..... but that seems to be IRL, and i wanted to know if i can run the tests..... the u1-client only without installing anything [11:26] gatox, then point to the python lib from the other project [11:28] mandel, also..... the code seems REALLY similar to mine.... it needs to be different things.... or should we merge them? [11:28] gatox, I wanted it to merge once you landed yours.. but is taking to long [11:28] gatox, the code is only similar in the notifier, the rest is diff [11:28] gatox, take into account that is using a twisted factory to talk with the daemon [11:29] gatox, tests are diff etc.. [11:29] ahh ok ok [11:31] gatox, take a look at the code and you'll see is a lot simplre [11:31] s/simplre/simpler [11:34] mandel, AHHHHHHH jejjejeej sorry, you are right.... i didn't read the last part of the description [11:35] :P i read everything else, except the test part [11:35] gatox, no problem :) [11:35] gatox, I know I do the same hehehe [11:36] mandel, the code review i did it yesterday (when i thought: OMG they are going to make refactor everything again), and it seems fine [11:36] s/make/make me [11:37] this stop being fun...... 2 refactors ago :P jeje [11:37] gatox, only problem is the notificator, the rest is diff :) [11:37] gatox, which can be shared, but that is a very small part, maybe 300 lines? [11:38] mandel, the code is a +1 for me (already there)...... now going back to try to finish this refactor [11:39] gatox, superb, did you manage to run the tests? [11:39] gatox, by setting the python path I mean [11:39] gatox, also, can I have a super simple review: https://code.launchpad.net/~mandel/ubuntuone-client/reactor-new-line/+merge/112535 [11:40] mandel, no, i did the code review part [11:42] mandel, about lines: [11:42] 30 -FilesystemMonitor = filesystem_notifications.FilesystemMonitor [11:42] 31 -_GeneralINotifyProcessor = filesystem_notifications._GeneralINotifyProcessor [11:42] weren't we using that from somewhere else? [11:42] or that definition was unnecesary? [11:43] mandel, just curious...... [11:46] gatox, it was diff line 9 [11:47] yes, but this one is missing: FilesystemMonitor [11:47] * gatox downloading and using meld [11:50] ah i see [11:51] mandel, +1 [11:51] gatox, thx :) [12:25] gatox, I managed to get the daemon to be launched as root on the mac when you start the machine :) === yofel_ is now known as yofel [12:25] gatox, not sd but the fsevents daemon :) [12:25] mandel, COOL! [12:27] gatox, so right now, the only thing we have to do in the packaing is copy a file to a location and tell launchd to start it, should not be too hard :) [12:29] good morning! [12:29] ralsina, morning! [12:30] ralsina, hi [12:30] ralsina, so before I go, I have a fist launchd integration where we pass the domain-socket as a parameter to the daemon (via xml) and have fsevents at boot time :) [12:30] whoohoo! [12:30] mandel: nice :-) [12:30] ralsina, it means that the installer just has to copy the xml to the correct location and tell launchd to start using it [12:31] ralsina, was probably the last piece we needed to have a first alpha release working [12:35] ralsina, gatox, super simple MP that explains how to set you mac to have the daemon running at boot time: https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/start-boot/+merge/112554 [12:35] mainly xml and fixing a small bug in the command line parsing [12:35] mandel: will try it after my call that starts in a few minutes [12:35] ralsina, ok, no problem :) [12:36] ralsina, also, the reactor issue in control panel and sso should be fixed by https://code.launchpad.net/~mandel/ubuntuone-client/reactor-new-line/+merge/112535 [12:36] ralsina, so we might be able to see things running very soon :) [12:36] mandel: wasn't mmcc doing that? [12:38] dobey: did we ever release anything with versions 3.1 or 4.1? I am guessing no, right? [12:39] ralsina, yes/no the issue was that the reactor was being dragged by the filesystem_notifications that was imported in platform [12:39] mandel: ok [12:39] mandel: that thing has been a pain in the ass since we started the qt migration, so good to finally see it fixed [12:39] ralsina, probably an error when refactoring things in ubuntuone-client, no filesystem_notifications is not in the init and therefore the reactor is not installed [12:40] mandel: instead of the monotonic pile of workarounds we had [12:40] ralsina, well, it was 'fixed' and then we moved things and broke it hehe [12:40] I'm off to have lunch [13:09] alecu, ping [13:15] argh, I am tempted to replace invalid characters in windows with "{{{NAMEOFINVALIDCHARACTER}}}" [13:19] ralsina: no, that's just the version we used in trunk at the time, so that nightlies/trunk are always newer [13:19] dobey: we did put a 4.1 in trunk? [13:19] ralsina: 4.1 is currently the trunk version number, yes [13:19] dobey: I thought trunk was 3.99.x also [13:19] nope [13:19] dobey: oh, ok, good to know we don have time-traveling users :-) [13:21] man, being able to use only one monitor sucks :( [13:40] alecu, are you here?? [13:44] Gatox: my isp seems to be fully broken. [13:44] I'm using my phone, but I hate typing here [13:45] alecu, :S i just wanted to let you know that i fix the darwin3 branch....... and i added a comment regarding the EVENT_CODES [13:45] (SwiftKey makes it a bit more bearable, but still...) [13:46] gatox: ack [13:46] alecu, phones with actually qwerty keyboard FTW :P [13:46] for me is really difficult to type on screens [13:55] alecu, if you are going to use eng you can always dictate, which is a little better :) [13:59] Mandel, do you use that to write code comments? [14:00] So... Google works, YouTube works, the isp website works. The rest is dead. [14:00] alecu, hehehe no, I use my phone ;) [14:01] Net neutrality....? [14:02] mandel, alecu so.... i've updated darwin3 and darwin4.... i review the comments and docstrings inn darwin4 to avoid the same problems as darwin3.... the only thing missing is the thing about EVENT_CODES that i answered to alecu in darwin3 (and reviews) [14:06] alecu, at least you can watch google IO :P [14:06] gatox, ok, if you are not too busy, can we give reviews to mmcc? [14:07] mandel, yes..... my main concern right now is to have reviews :P so i can do reviews too [14:08] mandel, ok, i have mmcc branches here..... i'll start reviewing them [14:08] gatox, thx :) [14:09] gatox, I'll take a look at yours once I'm done with his [14:09] mandel, can you review darwin4 before you leave to pycon please? [14:09] mandel, thx! [14:10] mandel: did you see the latest comment in bug #1017916 ? [14:10] Launchpad bug 1017916 in Ubuntu One Client "[windows] Does not download files with newlines (\n)" [Medium,Fix committed] https://launchpad.net/bugs/1017916 [14:11] ralsina, no, looking [14:12] alecu: you are in telecentro, and it's dead, right? [14:13] alecu: want to push the team call for tomorrw? I'm ok with that. [14:13] ralsina: +1 [14:14] ralsina: you on telecentro too? [14:14] ralsina, people must think I'm stupid and don't test code before I propose it.. the code uses {LINE SEPARATOR} and not {VERTICAL TAB} [14:15] dobey: can I get a rereview of https://code.launchpad.net/~ralsina/u1db/cmake2/+merge/112392 before it starts getting conflicts? :-) [14:15] ralsina: Google and YouTube work fine... perhaps we can hangout instead of mumbling. [14:16] alecu: let me comb myself first then! [14:16] alecu: and get on the plane with my fancy glasses ;-) [14:17] ralsina: sorry, new bzr breaks tarmac. made a branch to fix that real quick [14:17] dobey: sure, that looks much more important indeed :-) [14:18] thisfred, dobey, gatox, alecu, mmcc, briancurtin, mandel: anyone against doing a team hangout? [14:18] also: anyone knows how to *do* a team hangout? [14:18] ralsina: i am against [14:18] ralsina, nop [14:18] no idea how, but i'm up for a try [14:18] ralsina, that I don't know how to use the bloody thing.. but I don't mind [14:18] no, I have no idea if it'll work with my firefox, but I'm willing to try [14:18] dobey: any particular reason, and, are you more or less against it than against a team mumble? [14:19] I will try to start one and invite you guys [14:19] is the plug-in packaged in ubuntu? [14:19] dobey: no idea [14:19] ralsina, oh.... i'll need the other computer..... thx again alienware [14:21] well, not like i have video anyway, so whatever [14:23] ralsina: please invite my gmail address, not my canonical one [14:23] I don't intend to use that on g+ [14:23] thisfred: which one is it? [14:23] ralsina: it's thisfred :) [14:23] thisfred: ack! [14:23] ralsina, I need to get 3g for that [14:24] ralsina, give me 5 mins and then feel free to invite me [14:24] * mandel runs for the 3g modem [14:24] anyway, it's not call time yet :-) [14:25] thisfred: seems like I can't invite you for some reason [14:25] Ahhhh can't invite you twice [14:26] anyway, this seems to work [14:26] ralsina: alecu invited me [14:26] ahhh ok [14:26] anyway, let's do this again at call time ;-) [14:27] * briancurtin -> quick coffee run [14:31] are we using yield from everywhere already? It's been out for more than a day! [14:32] huh? [14:32] dobey: new in python 3.3 [14:32] it removes yield? [14:32] it allows subgenerators [14:33] eh [14:33] dobey: no it introduces 'yield from' [14:33] so you can yield from something else that is a generator. [14:34] yield from range(10) [14:34] for instance [14:38] gatox, ralsina any idea why would the qt ui get blocked after calling load? [14:38] mandel, that use to happend when we had issues with the backend [14:38] like being None or something [14:39] mandel: blocked as in showing the overlay? Usually you have an exception to go with it [14:40] ralsina, gatox, shows the overlay, does nothing after [14:40] man people ask for the craziest things in software [14:40] mandel: start it with --debug ande U1_DEBUG=1 and see if you get an exception in stderr [14:40] ralsina, ok [14:41] mandel, i've started to debug that.... it wasn't really clear where the problem was... but then i needed to keep working on the fsevents branches and i couldn't see what was going on [14:41] it's like if they went to a chevy dealer and asked if they can have a mustang engine in a new corvette [14:41] s/i've started/i started [14:41] gatox, no worries, is just that I want to see everything working in my machine [14:41] ralsina, --debug where? for the bin/ubuntuone-control-panel-qt? [14:42] mandel: yes [14:42] ralsina, seems not to have that option [14:42] mandel: then just setting U1_DEBUG [14:42] mandel, i was doing it with U1_DEBUG [14:43] ack [14:44] i miss my other monitor [14:47] gatox, ralsina, so, it can be that the signal CredentialsNotFound is not received because the sso process is killed before is done something that does no happen on windows because it is never killed [14:47] gatox, ralsina nothing to do with the reactor etc.. [14:48] mandel: maybe we want it not to die on darwin either? [14:49] ralsina, well as a quick fix yes, as the proper way to fix it is to keep track of the signals to send and wait 'til they are done [14:50] catching up, just remembered to say hi. good morning folks [14:50] mandel: yes [14:54] mmcc, +1 to this one: v [14:54] https://code.launchpad.net/~mikemc/ubuntuone-control-panel/fix-1015825-reorder-reactor/+merge/112432 [14:54] gatox, I have concerns about that one [14:54] I don't think that is needed [14:54] mandel, what? [14:54] gatox, the reactor branch from mmc [14:54] mmcc [14:54] sorry [14:54] mandel, why is not needed? [14:55] gatox, testing right now with contorl panel trunk but AFAIK the ui is shown with no problems [14:55] gatox, if it is not needed on windows, why is it on darwin? [14:56] mandel: if what is not needed? [14:56] mmcc, create the Qapp first [14:57] mandel, oh right - I have no idea. I'd love to figure it out, but in irc a couple of days ago alecu suggested that maybe we shouldn't poke the tiger by changing working code on windows [14:58] mmcc, what was the issue? [14:58] I believe that if we just did the same thing on windows as I do on darwin in that branch, it'll work fine. the tests with it that way work fine on windows... [14:59] the issue was that, if you install the qt4reactor first thing, it will create a QCoreApp for you. this breaks UI on darwin, but not on windows, for some reason [14:59] ralsina: so to prep for hangout, what should i be doing right now? do we have canonical plus accounts now? (i dont remember seeing one) [14:59] mmcc, but how did it break the ui on darwin? [14:59] mmcc, mandel, gatox, dobey, thisfred, alecu, let's hangout [15:00] mandel: shows the overlay then hangs, showing no animation. no gui events are sent [15:00] how many nerds does it take to make a g+ hangout work [15:00] dobey, too many.. [15:00] oh yeah, hangout, what do i do? please invite michael.mccracken at gmail..? [15:00] * mmcc puts on a shirt [15:00] ralsina: also, gmail address for me too, don't use canonical address on g+ [15:00] mmcc, os, that is not due to that, is because there is an error from sso to control panel and gets frozen like that [15:01] mmcc, I'm looking on how to do a quick hack to fix that [15:01] mandel I'm pretty sure we're seeing different things [15:01] in my testing I do get the credentials_not_found back from sso, it hangs after that [15:02] hmm [15:03] when is this hangout? on the 9th green at 9? [15:03] mmcc, yo do get an message saying that, but you don't see the stderr from twisted telling you it did not manage to send the signal [15:03] lol [15:03] mmcc: we're in there now [15:03] mmcc: didn't you get the invite? [15:03] ralsina, or me [15:03] for varying definitions of "we" [15:03] * mmcc not one of the cool kids [15:04] ralsina, did you see my private msg? [15:04] ralsina: no, I got no hangout [15:04] mandel: yep, invited you [15:04] mmcc: so need an invite? [15:04] er, no hangout invite [15:04] mmcc: what's your preferred gmail? [15:04] ralsina: yes plz, to my michael.mccracken@gmail [15:05] ralsina, ok, installing plugin, few sec [15:06] la la la [15:06] * mmcc refreshes... [15:06] where do these invite things show up? [15:06] dobey: google + I suppose [15:06] i thought g+ had circles [15:06] and etc [15:06] dobey, on other people's computers, apparently [15:07] dobey: yep, that too [15:07] * dobey is too cool for hangouts [15:08] hangout is hilarious [15:08] I expect it to get boring soon [15:08] * dobey already bored [15:08] dobey, mmcc: are you having trouble? [15:09] define trouble [15:09] there is no invite, and i don't see anyone else hanging out in my stream [15:09] ralsina: I was, apparently there's some plugin that I am now downloading, but there was no message about that [15:10] mmcc: ok [15:10] I just randomly clicked on the "Hang out" button and then it told me [15:10] dobey: I invited you on the canonical address. Should I use another one? [15:10] yes, gmail [15:10] dobey: which one? [15:10] see /msg [15:12] okay, well I have the plugin now but still no invite [15:12] 9th green at 9 [15:12] * mmcc thinks he hears sprinklers [15:12] mmcc: I'll try again [15:13] mmcc: invited again [15:13] mmcc, let me know when you add the test that mandel request here: https://code.launchpad.net/~mikemc/ubuntuone-control-panel/fix-1018614-darwin-raise/+merge/112493 [15:14] where should I be looking again? it shows up as an email? or in google plus? right now, neither one has anything new [15:19] the big problem is dev-tools [15:37] nice mandel [15:38] +1 to hangout from now on. better audio quality, optional video with monocles and scuba goggles, and mandel puppet show [15:38] :) [15:42] ok, need to get lunch; bbiab [15:42] briancurtin: agreed, it seemed to work pretty well. Not sure why it felt necessary to put "Michael McCracken hung out" on my timeline, shared only with myself... it's like, I was there, so... [15:43] anyway, mandel - I'll add tests for calling raise_ only on darwin. I agree with ralsina that it shouldn't hurt on other platforms but I don't know how to prove that :) [15:43] also mandel, can you paste a U1_DEBUG listing from the credentials not found issue you're having? [15:43] mmcc: get it running on ubuntu and windows with raise_ and see it does the same thing as before [15:43] mmcc: it's for the record in case you need an alibi, just share yourself with the police and you're covered [15:44] mmcc: I can do ubuntu for you :-) [15:44] mmcc: just to avoid platform-specific behaviour if it's unnecessary [15:44] ralsina: ok I'll test it all out. I have the code on both running vms [15:44] oh and FYI - it's my review day, i'll keep an eye on my inbox but feel free to directly ping me for reviews [15:44] ralsina: I was mostly concerned with unintended other consequences [15:45] aaaaaand lunch! [15:46] brb, garage door broke apparently?! === ubuntu_guru is now known as Daveonearth [15:52] mmcc, I'm looking closer to see if I get the bloody thing to run :) [15:53] mmcc, gatox, turns out that because we do set #!/usr/bin/python the stupid python path was wrong.. [15:58] <__lucio__> mandel, ralsina, in win32, files come as bytes or unicode or what? [15:58] <__lucio__> verterok, ^^ [15:58] kk [15:59] __lucio__, withint sd everything is bytes, that is, we convert to unicode to talk with windows, get unicode, return bytes to the state machine [15:59] does it make sense? [15:59] <__lucio__> mandel, so windows always returns unicode? [16:00] <__lucio__> we make it bytes in utf-8, but for we always get unicode from the platform [16:00] __lucio__, we allways pass bytes in utf-8 outside ubuntuone/platform [16:01] __lucio__, in the windows modules that interact with windows we use unicode (since listdir(unicode) returns unicode and listdir(bytes) returns bytes) [16:01] <__lucio__> mandel, so, we interact with windows using unicode? or bytes? [16:02] __lucio__, unicode [16:02] <__lucio__> verterok, ^^ [16:02] verterok, don't be shy, you can ask me ;-) [16:02] hehehe [16:05] hehe [16:05] mandel, where was the shebang hurting us? [16:05] mmcc, in ubuntu-sso-login [16:06] mmcc, it gets launched by the ui so it uses the wrong python [16:06] mmcc, that solves some but not all the problems [16:06] mandel: and we algo get unicode from the win-inotify (don't remember the name of the module)? [16:07] verterok, yes, we get the path from the fs in unicode and we pass them to bytes utf-8 [16:07] mandel I'm still not sure what problems you're seeing. perhaps it's the different environment we use? I have been using the buildout python to test controlpanel, and the mac-env script from the buildout [16:07] mandel: kk [16:07] mmcc, have you tested sso without the --login parameter? [16:08] verterok, FYI is under ubuntuone/platform/filesystem_notifications/windows.py [16:11] mandel, I think so, let me check my notes [16:11] kk, thanks [16:14] mmcc, ok, just tested the thing, you are 120% right with the reactor.. [16:14] mmcc, I think we should do it both on win and darwin and not just on darwin [16:15] on win there should be no problem [16:15] mandel, this is about installing qt4reactor after the uniqueapp init? [16:15] mmcc, yes [16:16] ok, yeah I agree. in fact if anything it'll avoid creating an unused qcoreapp on windows... [16:17] I'd like alecu's opinion here, since he was the one who suggested leaving windows code alone [16:19] mmcc, AFAIK there is not diff and the correct way per docs of the qtreactor is the way you did it in darwin [16:20] ok mandel, for the other issue, can you tell me exactly what you're doing? I don't understand why not having --login with sso makes a difference [16:21] mmcc, it calls the backend, just remove the --login param and try to run the thing to create a new account [16:21] * gatox lunch === gatox is now known as gatox_lunch [16:21] oh right, I have done that directly in the past, but doing that from control panel wasn't working for me last night [16:22] mmcc, is because of the python path [16:23] at least in my system [16:25] ok, I think dobey's change broke this for us. I can't run sso directly anymore either [16:27] what I see: when I run bin/ubuntu-sso-login-qt, it gets to the point of trying to run ubuntu-sso-login and that dies, unable to import pyqt [16:27] it's the change in r978 [16:27] mmcc, exactly, that was what I got, simple way to fix it is changing the path [16:27] mmcc, but is not the right way to do it :( [16:27] changing the shebang line in teh script? [16:28] mmcc, yes [16:28] mmcc, if you do that it works + your reactor branch [16:28] mmcc, I'm even going to report a ui bug \o/ [16:29] mandel, great. [16:29] ok so alecu - any opinion on moving the qt4reactor after uniqueapp on windows too? it'll be a quick change, the tests you asked for yesterday are already there [16:34] mmcc, gatox_lunch first funny ui bug 1018918 [16:34] Launchpad bug 1018918 in Ubuntu One Control Panel "Scrolling shares on darwin does not work" [High,Triaged] https://launchpad.net/bugs/1018918 [16:35] oh and ralsina ^ [16:35] mandel, juazzzzzzzzzzz [16:35] gatox_lunch, have you seen the video, wtf? [16:35] mandel, yes! [16:36] mandel, it seems that they were attached to a diff parent or something [16:39] gatox_lunch, also added bug 1018919 [16:39] Launchpad bug 1018919 in Ubuntu One Control Panel "There are blue highligts in the lists" [Medium,Triaged] https://launchpad.net/bugs/1018919 [16:40] mandel, yes, that's because we are not overriding the mac default for that [16:41] ohhh issues that i enjoy more!! [16:41] * gatox_lunch is happy to see these things [16:41] gatox_lunch, and this bug 1018921 [16:41] Launchpad bug 1018921 in Ubuntu One Control Panel "Selected checkboxes are blue in darwin" [Medium,Triaged] https://launchpad.net/bugs/1018921 [16:41] hello, all!!!! [16:42] welcome back alecu [16:42] mmcc, I have tested the reactor branch from mmcc I think we have to do the same on windows [16:42] alecu, ^ [16:42] ups [16:42] mandel: why so? [16:43] alecu, because is what the qtreactor does say [16:43] alecu, it was wrong that we did it this way, it just works on windows, the other way QApp before reactor is better and works in all platforms [16:43] mandel: sorry, I'm catching up with the backlog. Where does it say so? [16:44] mandel: if it does work fine on windows, let's move ahead. [16:44] alecu, let me get you the docs [16:44] mandel: I had some issues with the qt4reactor initialization previously, so I was not convinced about touching those bits on windows. [16:45] alecu, well, we can leave it like that, I don't think is needed [16:45] mandel, mmcc: but if we test all that IRL under windows and it works fine, then let's move ahead. [16:45] anyway, if you look at the backlog, is working :) [16:45] mandel: awesome then. [16:45] alecu, other more important issue is the problem with the python path, but is just hurting use because we use the brew one [16:46] mandel: what's the issue with the windows path? [16:46] mandel: I saw some relative imports with ../ [16:47] alecu, is in shebangs paths that are broken, at least for use with the python buildout [16:48] alecu, points to /usr/bin/python and we use /usr/local/python, I think is fixable if we execute the scripts using python path and not just path [16:48] alecu, but is a minor PITA for developers, the packaged version should work ok [16:49] mandel: oh, I see. I don't mind using "python bin/script" at all. [16:49] so yeah: very minor pita. A very small sandwich you'll make with that pita. [16:49] alecu, hehe [16:50] alecu, so, all the rest works :) [16:50] all, shares, details, sso, sd work [16:50] we might have some bugs when we start sd (I cought an already running error) but seems that we just have packaging left! [16:51] mandel, how are you editing the path to the backend scripts now? where are you adding 'python' exactly? [16:51] mandel is such an optimist :) [16:52] mmcc, I just changed the shebang, the smart way would be to tell the runner to use python + path and not just path and you are sorted [16:52] mmcc, is latet here, so I won't be able to fix that yet I'll add a bug :) [16:57] mmcc, alecu bug 1018924 [16:57] Launchpad bug 1018924 in Ubuntu One Control Panel "Scripts are not ran because the shebang points to the wrong path" [Critical,Triaged] https://launchpad.net/bugs/1018924 [16:57] mandel: ok, add the bug and assign me, I'll do it [16:57] hrmm [16:57] uhm [16:57] please do not change those back to /usr/bin/env [16:58] they cannot be that [16:58] dobey we're not going to do that. [16:58] the idea was to change the code that invokes them to say 'python backend-bin' [16:59] dobey, that wont be touched, is the way we launch them :) [16:59] although it needs some thought, since of course it's different when packaged and I want to be sure it does the right thing on windows [17:00] mmcc, indeed, you are the one that will have to deal with it :) [17:00] mandel: what launches them? [17:01] dobey, they are subprocess from control panel and sso, lets make the call as python script and not script and is fixed [17:01] dobey, the a rnner code somewhere in sso-utils [17:02] ugh [17:02] mandel, and that runner code is only used on windows and darwin right? [17:02] calling "python script" from subprocess is nasty [17:02] and it will break ubuntu [17:02] it's the protocol buffer stuff [17:02] mmcc, yes [17:02] so won't break linux [17:03] dobey, it wont break ubuntu it will just be done in darwin [17:03] perspective broker [17:03] thanks dobey, d'oh [17:03] mmcc, in windows we call the .exe so we have to do something similar [17:03] which reminds me [17:04] briancurtin: we can't just simply switch to nose or something else, for running the tests. we're not just using plain trial, but u1trial which does a bunch of extra stuff we need for various tests [17:05] so we port u1trial or something as well [17:07] i should break up the dirspec stuff i was doing into smaller branches. [17:07] well, it's small as-is, but doesn't work exactly on both 2 and 3, and is being really annoying [17:10] mandel - ok to set status to approved on that reactor reordering mp? you +1ed but left it [17:10] mmcc, before I go, I have this for review: https://code.launchpad.net/~mandel/ubuntuone-fsevents-daemon/start-boot/+merge/112554 [17:12] mmcc, let me check === gatox_lunch is now known as gatox [17:12] mmcc, yes, I just approved in case you wanted to make more changes [17:12] mandel, ah ok. I have no other changes [17:13] mmcc, usually the global approve is done by the person that proposes the mp, just in case :) [17:13] mmcc, feel free to land it [17:13] mandel, oh i didn't know that. someone told me it was usually the 2nd approver... [17:13] mandel your branch looks fine but i can't tell what changed in main.m. are my eyes failing me? [17:13] whitespace? [17:14] mmcc, optional to required argument [17:14] mmcc, and no argument for help [17:14] yeah i just noticed that but what about line 80 of the diff? [17:14] * mmcc needs to get better sleep [17:15] mmcc, typo! [17:15]