[00:18] duanedesign: if you are still around: cleanup the user's metadata, and that should fix itself [08:31] Good morning, everyone! :D [08:44] pff I leave and a lot people talk about me.. :P [11:19] good morning! [11:28] gatox, morning! [11:29] mandel, hi, i was reading your email [11:29] mandel, thx for the review [11:31] gatox, no proble, though I forgo to add the vote: https://code.launchpad.net/~diegosarmentero/ubuntuone-client/refactor-remove-windows-linux/+merge/105705 [11:31] gatox, is a tiny needs fixing [11:31] mandel, on it! :D [11:32] man did you read my comment in the mp? what do you think? [11:33] the code is looking ugly cause of the lint checker [11:34] gatox, I would add it to the lint exceptions, but we should ask alecu for a +1 [11:34] mandel, ack [11:41] mandel, i don't undeerstand you need fixing :P [11:41] mandel, what part of the code you are referring to? [11:44] gatox, the line numbers in the comment are from the diff :) [11:44] gatox, in line 335 you do an from very.long.module import windows [11:44] source = windows [11:44] mandel, yes, but if i do: from ubuntuone.platform.logger import windows as source.... the lint checker complain about a double import [11:45] if i do the same for linux [11:45] gatox, puto lint... [11:45] really? [11:45] * gatox hates the lint checker [11:45] gatox, then add that as a comment and I approve.. bloody hell we need to get a lint that works [11:46] mandel, yes, that's why we were doing that with mmcc ...... i'm going to test it again, just to be 100% sure!!.... but we were having that problem with mmcc [11:46] mandel, well...... i'm fixing those problems in the fork of pyflake we are using in ninja..... maybe i can propose those changes to be merge with pyflake later [11:47] gatox, would be very nice.. but we don't use flake but lint :( [11:47] mandel, i thought we were using flake in u1-client [11:47] mandel, why are we not using the ignore rules then? [11:48] #pylint disable bla bla [11:48] gatox, oh, true is u1-client, we are using flakes there.. sorry [11:50] mandel, yes...... i did it just to be sure..... and with "as source" it fails with: 38: redefinition of unused 'source' from line 35 [11:50] gatox, cool, then you tell me to f off in the MP and I'll approve :) [11:51] mandel, jeje ok, but please don't approve it globally..... so i can ask alecu if i can add the __init__ to the ignored files and prettify the code [11:52] mandel, comment added.... [11:54] good morning! [11:54] ralsina, hi [11:54] gatox, mandel: I think I sent mail to both of you yesterday. No rush, ust a gentle reminder! ;-) [11:55] ralsina, i just answer it [11:55] ralsina, got nothing.. [11:55] gatox: thanks! [11:55] mandel: checking... [11:56] mandel: sent you one about bug#998853 [11:56] ralsina, I got gatox email and replied in my morning and also got one from apple to join the devel team.. [11:56] bug 998853 [11:56] Launchpad bug 998853 in Ubuntu One Windows Installer "Ubuntu One Windows version crash running with Wine" [Undecided,Won't fix] https://launchpad.net/bugs/998853 [11:56] jejeje running with wine..... WHY?! [11:57] gatox: it's explained in the bug! [11:57] ralsina, yes..... just read it..... i see why now :P [11:57] ralsina, and you did not reply with my inception sentence... lol [11:58] ralsina, mandel why the people would do that?? when porting u1 to another platform is so much fun and no frustration!! jejeje [11:58] gatox, haha [11:58] * gatox had a bad day yesterday jejee [11:59] ralsina, gatox, I'm off to have lunch, if you are interested there is some nice work in progress in lp:~mandel/+junk/fsevents include an small example of a twisted protocol I'm using to echo the fs event to the terminal [11:59] mandel, cool [12:00] ralsina, gatox, I'm working on allowing the twisted code to sent a configuration to the paths it wants to listen which is nearly done (security is there I need to write the python code) [12:00] mandel, i'm trying to fix everything around fsevents.... in order to get that working :P [12:00] gatox, I'm going as fast as possible with the objective-c code, I still need to do launchd integration and send an email I promissed about it [12:01] ok, I lunching ;) [12:01] * mandel lunch [12:03] gatox: also, about spinboxes, I think for now I will unstyle them and file a bug about styling them [12:03] ralsina, ok..... with unstyle..... you mean to add an empty style when they don't have focus.... or remove the focus style for the spinbox? [12:04] gatox: removing the focus style for the spinbox [12:04] ralsina, ah ok... because adding the empty style would do the buttons to look always the same.... but they will be kind of ugly buttons [12:05] yes, square-3d like they are now when focused [12:05] yap [12:07] the bug level is maageable in all our projects except ubuntuone-client (600+ bugs) so I may spent a few days attacking that with a machete [12:08] wow... 600 bugs :S [12:08] alecu, just the person i wanted to see :P [12:09] hello! [12:09] gatox: all the others are < 100 [12:09] alecu, hi [12:09] gatox, here am I. Take a good look! [12:09] alecu looks very ASCII today [12:09] alecu, jejeje i have a question for you about this branch..... https://code.launchpad.net/~diegosarmentero/ubuntuone-client/refactor-remove-windows-linux/+merge/105705 [12:10] if you read my comment: https://code.launchpad.net/~diegosarmentero/ubuntuone-client/refactor-remove-windows-linux/+merge/105705/comments/227372 [12:10] * alecu looks [12:10] alecu, i'm asking for your permission to add that file to the ignored ones for the lint checker [12:10] gatox: we have been doing crazy contortions to make pylint not complain about those imports, I would disable [12:11] ralsina, yes..... the code looks really weird just to let the lint checker happy [12:11] gatox: why not do a #pylint disable ? [12:11] ralsina, because in u1-client we are using pyflakes, not pylint [12:11] gatox: who is complaining? pylint? pyflakes? [12:11] we don't have the disable flags in u1-client [12:11] gatox: really? [12:11] alecu, pyflakes [12:11] right. [12:12] ralsina, yes [12:12] we are surrounded by suck [12:12] we have a lot of: from bla.bla import windows [12:12] source = windows [12:12] because we can't do: from bla.bla import windows as source [12:13] because if we do the same for linux..... linux as source in the else part..... pyflakes complains too [12:13] gatox: right. I remember nessita did a lot of source = ... [12:13] gatox: and it probably all was to work around py(flakes|lint) [12:13] alecu, and we did it a lot again with the refactor [12:14] gatox: so, I see that some files are added to pylintrc... [12:14] gatox: is that file used too by flakes? [12:14] alecu, yes... i wanted to add this file there [12:14] alecu, yes [12:17] alecu, so...... it makes sense for you to add the __init__ to the pylintrc to be ignored? or i leave just as it is? [12:17] gatox: I see that in ./ubuntuone/platform/filesystem_notifications/__init__.py there are two definitions: FilesystemMonitor and _GeneralINotifyProcessor... [12:18] aja [12:18] gatox: I don't see what's ugly about that... [12:18] gatox: so, I don't understand why we want to ignore it. [12:18] alecu, no, that...... the code above that.... the part with pyinotify [12:19] alecu, wait.... i can't find it in the diff [12:19] checking if i push it [12:20] gatox: oh, you mean lines 166..194 in the diff... [12:20] gatox: "path_exists = os_helper.path_exists" [12:20] and etc. [12:21] gatox: if it's those lines, they look incredibly ugly and error prone to me. [12:21] gatox, alecu, mandel: today is eric's review day, but he's on sprint, so reviews are by me and someone else at random, I will try not to hit the same guy all the time [12:21] alecu, no..... ahhhhh..... now i see that i ended up removing that code, cause with the lint errors it was producing a lot of problem..... so instead of doing something generic, i did more specific imports, and problem solve [12:21] OTOH, I need reviews on https://code.launchpad.net/~ralsina/ubuntuone-control-panel/fix-966283/+merge/105807 -- I approve but I want second and thir opinions ;-) [12:21] * alecu looks [12:21] ralsina, i can revieeew that [12:22] gatox: cool, it has pictures! [12:22] alecu, so...... forget all my questions..... i fix it in another way..... but it seems that i keep thinking about this :P [12:23] alecu, we were having a pretty nassty import yesterday..... [12:24] mandel, so, when you can approve that branch.... is ready to be merged [12:24] ralsina, +1! [12:24] gatox: thanks! [12:27] ralsina, +1 [12:27] alecu: thanks! [12:27] Another one with pics: https://code.launchpad.net/~ralsina/ubuntuone-control-panel/fix-983665/+merge/105737 [12:28] ralsina, ack [12:31] ralsina, +1 [12:31] gatox: thanks again! [12:35] ralsina, "ubuntuone-control-panel-qt: error: unrecognized arguments: -reversed" [12:37] plap.... didn't test it IRL... same here [12:38] (the tests pass though) [12:41] alecu, et al: review appreciated: https://code.launchpad.net/~facundo/ubuntuone-client/fix-getdelta-repetition/+merge/105220 [12:42] facundobatista, looking. [12:46] alecu: -reverse [12:47] sorry [12:48] ack! [12:51] ralsina, awesome, it works as expected now. [12:51] alecu: ack [12:51] ralsina, I just see a weird bug with one of the links... [12:51] ralsina, with "Go to the web for public and provate sharing options" [12:52] looking [12:52] ralsina, the label alternates between stepping on the icon and not stepping on the icon. [12:52] ralsina, it seems like a qt bug. [12:52] alecu: so it works "sometimes" or it moves? [12:53] ralsina, it moves depending on the focus [12:53] right [12:53] ralsina, only if that label is focused, it looks fine. [12:53] we have some tweaks for that on the LTR mode, probably need to reverse them for RTL [12:54] ralsina, as soon as it looses focus, it looks bad. [12:54] alecu: will fix [13:08] alecu: pushed a smaller change to the RTL branch that breaks less stuff :-) [13:09] ralsina, awesome. [13:23] rye: bug #985484 if I change the label in the UI to "KiB per second" it's fixed? [13:23] Launchpad bug 985484 in ubuntuone-control-panel (Ubuntu) "Qt control panel says kilobits per second, means kibibytes per second" [High,Triaged] https://launchpad.net/bugs/985484 [13:27] ralsina: the gtk control panel had KiB/s, so yes [13:27] rye: cool, thanks! [13:33] gatox, ok, will approve right now! [13:33] mandel, thx [13:37] gatox, you've got my +1 [13:37] mandel, great [13:49] ralsina, the LTR branch works awesomely. May I bother you with yet another detail? [13:49] alecu: of course! [13:50] ralsina, the control panel has a "rounded corner" at the right of the box that's just below the tabs. [13:50] the white box with black 1px border [13:50] ralsina, I think that in RTL that rounded corner should be the left one. [13:51] ralsina, perhaps a screenshot may explain it better... [13:51] * alecu screenshots. [13:54] sure [13:54] gatox, alecu: may I get an opinion on befre/after pics here: https://bugs.launchpad.net/ubuntuone-control-panel/+bug/985484 [13:54] Ubuntu bug 985484 in ubuntuone-control-panel (Ubuntu) "Qt control panel says kilobits per second, means kibibytes per second" [High,Triaged] [13:55] * gatox looking... [13:55] ralsina, http://ubuntuone.com/0cmy5UUvr6mzBbwUW19vxe [13:56] alecu: right [13:56] alecu: picky [13:56] ralsina, me like it [13:56] ralsina, I am, yes :-) [13:56] alecu: will do, is easy [13:56] ralsina, i totally prefer the after one [13:56] gatox: plus it's MUCH more compact [13:57] yep [13:57] ralsina, gatox: +10 [13:57] alecu: I am getting a +1 from design and then I push [13:57] ralsina, it looks *MUCH* nicer. [13:57] ralsina, great! [13:57] * ralsina is doing fun stuff for a change [13:57] * alecu realizes how much he hates the whole text "Kilobits per seconds" [13:58] ralsina, \o/ [13:58] ralsina, question: do we have freaky fridays?? or are they suspend? [13:59] gatox: with caution [13:59] gatox: nothing is on fire, noone is late (yet), so yes [14:00] ralsina, ok..... just asking, because i wanted to finish the indicator app..... but when i try to do it after work, usually i have a lot of stuff to do :S [14:00] gatox: use the friday [14:00] ralsina, thanks! [14:08] alecu, ping [14:12] briancurtin: ping, no rush, I sent you a couple of mails yesterday, please take a look when you can, thanks! [14:16] gatox, alecu: review for the kibblebits; https://code.launchpad.net/~ralsina/ubuntuone-control-panel/kibblebits/+merge/105821 [14:16] ralsina, on it [14:18] ralsina: yep, i got them and will be looking today. about the runas one, i might need to setup a domain in order to test it...nothing really put in words that it's required, it just showed it in all examples as what you might want to do [14:18] also, not sure if you intended it but they went to my gmail [14:19] briancurtin: autocomplete and carelessness ;-) [14:19] alecu: flipped the rounded corner in https://code.launchpad.net/~ralsina/ubuntuone-control-panel/fix-983665/+merge/105737 [14:21] morning folks, reading the backlog... [14:21] ralsina, +1 [14:21] mmcc, hi! [14:22] good morning mmcc! === zyga_ is now known as zyga === yofel_ is now known as yofel [14:37] ralsina, was just looking at the dialog box changes you made. Was that in response to a bug? I'm curious how many people change those settings [14:37] mmcc: the bandwidth settings? [14:38] ralsina: yes, the one where you moved the units into the text box (in cocoa, IIRC that's a "combo box", not sure what qt calls it) === m_conley_away is now known as m_conley [14:43] It's not important, but I noticed that dropbox on osx has a similar dialog (but more hidden) and a similar lack of advice about why you'd want to limit it, and what bandwidth limits make sense... [14:44] I had a couple ideas for how we can improve this, but maybe most users just leave it unlimited? [14:44] anyway, back to my actual bugs. [14:45] does anyone know why mandel's mp here: https://code.launchpad.net/~mandel/ubuntu-sso-client/use-unix-domain/+merge/104244 hasn't been merged? I have a branch that depends on it [14:46] mmcc, well, I need a review for that to land, lets bully people :) [14:47] alecu, since thisfred is ina psrint can you look at https://code.launchpad.net/~mandel/ubuntu-sso-client/use-unix-domain/+merge/104244 ??? [14:47] or maybe ralsina.. [14:47] mandel, sure, I'll take a look. [14:49] alecu, thx! [14:49] alecu, also, whenever you have time, we can take a look at the daemon, I need to clean something a little to make it easirt for testing, I'd like to know how it goes in your mac mini [14:50] mandel, I've got an air too now. [14:50] mandel, the mini is going to ralsina. [14:50] alecu, oh, even better! [14:50] mandel, and... do you have a link for the daemon? [14:50] alecu, yes, lp:~mandel/+junk/fsevents [14:51] orsom [14:51] alecu, I'm removing the very last hardcoded path (the one of the pipe) which will later will be remove to start the server from a fd passed by launchd [14:52] mmcc, sorry, I am on a phone call [14:52] mmcc, you could setup a call with lisette about ideas on how to make that dialog useful [14:53] mmcc: and yes, we kow users use it because they complain about it all the time ;-) [14:53] alecu, what octal is a+rw ? [14:53] mandel: a? it's u, g or o [14:53] mandel: ah, right, 666 [14:54] ralsina, no prob. I'll write an email to lisette and call if necessary [14:54] ralsina, a = all so 666, is correct? [14:54] mandel: right, 0666 is rw for everyone [14:54] ralsina, is there a way I can get pickier? The "dotted lines" that separate the top box are on the wrong side. [14:54] alecu: well, you could but it would involve a monocle [14:55] alecu: looking [14:55] alecu, ralsina - are you guys talking about the settings dialog? [14:55] mmcc: no, about RTL locales [14:56] mmcc: try ubuntuone-control-panel-qt -reverse for a taste [14:56] mmcc, we are discussing the control panel when running in RTL [14:56] ralsina, mmcc: is this about that bug? [14:56] ralsina: ah ok. RTL=? [14:56] right-to-let [14:56] ralsina: oh [14:56] left [14:57] lisettte: yes, bug 985484 [14:57] Launchpad bug 985484 in ubuntuone-control-panel (Ubuntu) "Qt control panel says kilobits per second, means kibibytes per second" [High,Triaged] https://launchpad.net/bugs/985484 [14:57] ralsina, mmcc: could we not just use text? Automatic would be great for most users imo if we can provide such an option? [14:57] ralsina, mmcc: or 666 of course ;) [14:58] lisettte: agreed. dropbox has an "automatic" setting, but only for upload speed, interestingly [14:58] mmcc: i wonder why…. [14:58] alecu: flipped dotted lines [14:58] ralsina, great! [14:59] lisettte: sure, we can do text... I think :-) [15:00] me [15:00] me [15:01] me [15:02] me [15:02] me [15:02] eh???? [15:02] me [15:03] lisette it's ur standup time [15:03] our [15:03] thisfred, ? [15:03] thisfred is on sprint [15:03] oops [15:03] so go gatox go! [15:03] DONE: [15:03] Some errands yesterday, landed some branches to fix a sync status issue in u1-control-panel, and make use of the network detectation page on sso. Landed a branch for u1-client, refactor and remove the remaining linux and windows folder inside platform. Working on getting os-helper and tests working on mac os. [15:03] TODO: [15:03] Finish with os-helper and keep working on fsevents. [15:03] BLOCKED: [15:03] No [15:03] ralsina, go [15:03] DONE: fixed a bunch of bugs, triaged a ton more, mailed you guys a lot, sent a bunch to design, mgmt call, reviews TODO: more of the same, reviews BLOCKED: no NEXT: mandel [15:03] DONE: remove the hardcoded config for the root daemon. Got a python script that sends the configs for the user, security is check and events start to be sent. [15:03] TODO: Make the call really async, right now it assumes that we are sync when sending the config. Provide a param to provide the path to domain socket for testing. [15:03] BLOCKED: no [15:03] mmcc, please [15:04] lifeless, hide! [15:04] DONE: branch for darwin network done [15:04] TODO: propose branch, write up packaging plan [15:04] BLCK: none [15:04] next alecu [15:04] DONE: Attended UDS. Catching up with mail. A few reviews [15:04] TODO: finish security patches [15:04] BLOCKED: no [15:04] NEXT: briancurtin [15:04] DONE: poking around in debugger trying to reproduce the get_rootdir error, tested a few branches, setup a Windows 8 machine [15:04] TODO: look into roberto's emails about a runas and autoupdater issue, get windows jenkins machine back up so tests will run because they should be green! [15:04] BLOCKED: None [15:04] NEXT: no one, i think [15:05] * mmcc brb - emergency cleanup after dog [15:05] * ralsina rechecks dobey's status for the day [15:05] so, comments? [15:05] nop [15:05] cool, so let's kick bug ass as if it were starship troopers! [15:05] mmcc, mandel: how are you guys approaching the network detection on darwin? [15:05] the 1st, not the second one. [15:06] alecu, mmcc has been looking at that more than I have [15:06] mmcc, I'm mostly interested in how the network detection apis work: either sync or async. [15:06] mandel, right; I thought you might have reviewed that. [15:07] * alecu waits for mmcc to get back from the "dog emergency" [15:07] * gatox lunch! === gatox is now known as gatox_lunch [15:08] alecu, not yet propsed AFAIK [15:09] ralsina, approved RTL branch [15:09] * mmcc is back from the dog emergency. it was pee! [15:10] alecu, there are both, and we're using both. it's very similar to how our code works with windows [15:10] alecu: awesome! [15:10] alecu, I'm about to propose a merge for that code, just need to test it on windows [15:11] mmcc, please add me as a reviewer then :-) [15:11] alecu, will do [15:11] mmcc, great :-) [15:11] mmcc, me too! [15:12] :) [15:12] mandel, sure [15:12] so, 'cd' on windows takes either forward slashes or backslashes? wow [15:13] mmcc: sometimes [15:13] mmcc: not if you use quotes [15:13] ralsina: sometimes! hooray, I love computers [15:13] mmcc: or if you start your path with //!/ (or was it \\!\ ? [15:18] mmcc: mandel and I once gave a 1-hour conference on cross-platorm dev, and 20 minutes or so was both of us cursing path handling. [15:19] ralsina: good times :) [15:19] the worst is when applications, such as bzr and hg, do different things with differently slashed paths [15:20] bzr st outputs paths with slashes that bzr diff doesn't recognize :/ [15:20] so you have to manually flip them, or just type the path again [15:20] if things just BROKE when you use the wrong path? Easy. Having it work 85% of the time is pure evil. [15:23] alecu: if I can get the 2nd +1 for https://code.launchpad.net/~ralsina/ubuntuone-control-panel/kibblebits I will be happy and go have lunch :-) [15:24] ok, I'll give it a look now. [15:30] ralsina, approved, go have lunch. [15:30] alecu: yay! === zyga_ is now known as zyga [15:36] alecu, do you have time for a quick mumble? [15:39] mmcc, gatox_mac, are you guys using macs? (got to ask) [15:42] mandel, yes. I only have one computer in my house! A mac mini - I run ubuntu and windows in virtualbox vms. [15:42] mmcc, gatox_lunch, ralsina, alecu: if you are in mac station you can download http://ubuntuone.com/1qEMSM4auG14bRPnRiuKDj (change it to be executable) and branch lp:~mandel/+junk/fsevents there you find a README that tells you how to launch the daemon, launch a python script that will stdout the events [15:42] mandel: not me [15:43] ralsina, well, then you have to trust me :P [15:44] ralsina, but right now, we have both ends talking via json and the domain socket, the python script sends the config (lets say udfs you are interested) and sends the events to the python client, for every event we print it to the terminal [15:44] mandel, that does not run on 10.6 :| [15:44] mmcc, what is the output? [15:44] mandel, linker errors [15:44] mmcc, I guess I have to compile it for 10.6.. [15:44] mandel - we'll need to build it against the 10.6 SDK [15:44] yep [15:44] I can do that too, as soon as I get the new XCode [15:45] mandel btw, you said you got an email about the apple dev program? I didn't... [15:45] I was waiting for that to get the latest xcode [15:45] mmcc, ok, I'm downloading the 10.6 core lib (500mg wtf!) and will try to compile it too.. [15:45] mmcc, yes, I got it this CET morning.. weird [15:46] I'll ping urbanape : "ping" [15:46] pong [15:46] reading backlog [15:46] hey urbanape, did you add me to the dev group for apple? I didn't get an email yet [15:46] urbanape, uh, I'd love for you to test this: [15:46] if you are in mac station you can download http://ubuntuone.com/1qEMSM4auG14bRPnRiuKDj (change it to be executable) and branch lp:~mandel/+junk/fsevents there you find a README that tells you how to launch the daemon, launch a python script that will stdout the events [15:46] I did (or I thought I did) [15:47] urbanape, I got it [15:47] urbanape: do you know what email you used? maybe I need to check some spam folders [15:48] ah, I sent it to the apparently nonexistent michael.mccracken@canonical.com [15:48] just a moment [15:48] urbanape: ah yes - mike.mccracken, oddly enough [15:49] re-done [15:49] urbanape: and... BOOM [15:49] urbanape: thanks [15:49] np [16:00] * mmcc is downloading xcode 4.2 for 10.6 -- 1.6 GB, zorch [16:01] mmcc, is just stupid the size of these things.. === gatox_lunch is now known as gatox [16:02] mandel: yep it has *so* much stuff in it. every dev tool and SDK you might need. [16:03] mandel it's a symptom of Apple's severe allergy to package managers [16:04] mmcc, I don't know why MS and Apple did not copy apt is a so much better approach [16:04] mandel, do you still need me to test that? [16:04] gatox, yea! [16:04] mandel, ok...... i'll try now [16:04] gatox, would be great to see it it works, is understandable and you don't get a grey screen of death :) [16:05] mandel -- the usual argument is that most people don't care and it's just details they don't want to handle. It is a tough design problem! [16:07] but - IMO, Apple's recent approach to tough design problems like this seems to be "punt" and err on radical simplicity. And then sell millions of the simple thing that doesn't do everything I want :\ [16:08] * mmcc is suddenly "talking about apple on the internet". Considered harmful, back to work [16:09] mmcc, lol [16:09] mmcc, yeah, I like it more when I used the old system in my old PPC emac.. but well, is their choice [16:13] mandel, i get this: http://paste.ubuntu.com/989155/ [16:13] gatox_mac, did you call the daemon first? [16:13] gatox_mac, the fsevents one? [16:14] mandel, yes…. i follow the steps as the read me says… and the daemon is runing [16:14] gatox, remove the trailing / in /Users/gatox/Desktop/watch/ [16:14] gatox_mac, ^ [16:14] ack [16:14] mandel, same problem [16:15] the error seems trying to connect to the socket [16:15] gatox_mac, may I see how you launched the daemon please [16:15] mandel, i didn't compile the file…. but it's running, so i assume it's okç [16:15] gatox_mac, yes, is more too see which domain socket path you gave [16:16] mandel, http://paste.ubuntu.com/989164/ [16:17] gatox_mac, read the path you gave: /Users/gatox/Deskt/watch [16:17] gatox_mac, stop typing like Manuel!!!! [16:17] :) [16:17] mandel, mono…. that is a problem in mac console… the path is right [16:18] mmcc: looks like I can [16:18] mandel, Mac-mini-de-Diego-Sarmentero:Desktop gatoxsudo ./FsEvents /Users/gatox/Desktop/watch [16:19] mmcc: Looks like I can't have two separate people be agents. [16:19] so, we'll get to the point where we share the certs and profiles. [16:19] mandel, sometimes when i try to see the console history it shows weird things but execute properly [16:19] gatox_mac, weird.. you break any system hehehe [16:19] mandel, this is how i'm executing the daemon: Mac-mini-de-Diego-Sarmentero:Desktop gatox$ sudo ./FsEvents /Users/gatox/Desktop/watch [16:19] mandel, that's my power [16:20] urbanape: ah, ok. well, some day in the future I may have a favor to ask of you [16:20] gatox, ok, once you have that you see lots of things being printed out to the screen right? [16:20] okay, Godfather [16:20] mandel, yes [16:20] urbanape: :) [16:20] gatox_mac, then, according to what you pasted to passed /Usergatox/Desktop/watch to the python script [16:21] mandel, do you want to see the output? [16:21] gatox_mac, when it should have been "/Users/gatox/Desktop/watch" [16:21] gatox_mac, or is that another terminal fail? [16:21] gatox_mac, I'm no interested in the output is probably lots of things like FS_STAT etc.. [16:21] mandel, terminal problem….. but… the watch thing…. i don't know if it's something that has to exists or not [16:22] gatox_mac, no, the fsevents once running is should be left running [16:22] mandel, problem fix [16:23] gatox_mac, what was it? [16:23] mandel, i was pointing to an existing file [16:23] folder [16:23] gatox_mac, oh, mierda, I should add a check for that.. [16:23] mandel, it's working [16:23] gatox_mac, hurray! [16:24] gatox_mac, play with some files and you should see the events going to the other side :) [16:24] mandel, yes, i did that to see it working [16:24] gatox_mac, ok, so we have a working proof of concept with Events + Security + IPC :) [16:25] mandel, awesome work! [16:25] gatox_mac, thx! now, I have to add support for launchd, better ipc (I need to make calls to be async in the python side) etc.. [16:26] gatox_mac, once that is there, we can hook this to sd and start sending sd events to the state machine! [16:26] mandel, great…. i'm fixing all the modules and imports inside platform, to be able to run the tests, so i can start testing my fsevents... [16:26] gatox_mac, superb! because testing sd with this code is going to be 'interesting' [16:27] jeje agree [16:27] for the daemon part [16:28] gatox_mac, exactly.. I wonder how can we do that, because asking people to run the tests as root is a terrible idea [16:28] mandel - \o/. If you want to try building it for 10.6, I'm happy to test it. Or just wait for my xcode to download - about an hour [16:28] mmcc, let me see, I should have the libs now [16:29] * mandel fights with xcode [16:29] mandel, I was thinking about that - might need to change the daemon design so sometimes it can read "events" from a file for tests [16:29] mandel for testing you want the events to be predictable anyway, so I'd say there's no need to be root... [16:30] mmcc, yep, currently all the sd tests are more 'integration tests' the tests do some file system operation and then we check what happened, which is not a great idea for unit tests [16:31] mmcc, doing a mock daemon for our cases might be the best approach [16:31] mandel, yes a mock daemon for testing the client code is good, but we also need to test the daemon, yes? with mock client? :) [16:31] mandel, alecu - what's the status on that sso client / unix sockets branch? [16:32] I ask because I think I want to do two mps - one that fixes the sso client test scripts on windows and linux to ignore 'test_darwin.py' , since that's simple and unrelated to the network state stuff, and then a second for the network state stuff [16:32] and I was thinking about how to arrange that [16:35] mmcc, mandel i think that Xcode now is divided in two parts… so you don't need to download the whole thing in order to compile stuff [16:36] IIRC [16:46] mmcc, compiled, can you test with the one found here: http://ubuntuone.com/1qEMSM4auG14bRPnRiuKDj [16:46] gatox_mac, I have everything except the old libs.. [16:46] mandel: maybe doing a static build is a better idea [16:47] mandel: larger, but universal, nice for testing [16:48] ralsina, true, I'll need to play with the build settings and maybe move away from ARC so that we do not have issues, the project is small so move way from ARC is doable [16:48] mandel: right [16:50] ralsina, but I'd like to have a chat with you guys and decide if we move this already to a project or not, as a proof of concept, it works, the python code gets the events (I might need to tweak how) and the rest, integration with launchd, tests, better async support in twisted are things I won't do unless we are sure about this [16:50] mandel: why *not* move it to a project? [16:51] mandel: because if there is no reason against, I don't have to decide, do I? ;-) [16:51] ralsina, oh, I'm happing with moving it to a project and start filing bugs :) [16:51] mandel: then, by all means go ahead [16:51] ralsina, well, you opinion + mmcc and alecu is what I need [16:51] mandel: try not to call it "putomac" or something like that please [16:52] ralsina, putisimo-steve? [16:52] mandel: che, que está muerto [16:52] en fin [16:52] ralsina, put-to-mac almost works [16:52] hehe [16:52] ralsina, hehehe [16:52] mmcc: yes, agreed that we should disable the numbers when the checkbox is off and hide the text, I will do that [16:52] +100000 for put-to-mac [16:53] * ralsina kills that while it's still small [16:53] mandel, I'll take a look in a few minutes... I've got lunch served. [16:53] no mandel. Bad mandel1 [16:53] alecu, sure, no problem! [16:53] ralsina hey, 'git' has an obscene name and look how successful it is. We should all want to be like git! [16:53] JUST LIKE GIT [16:53] mmcc: only against (which I don't care about) is the convenience of clicking "up" on the numbers and having the checkbox check itself [16:54] mmcc: and dn't forget "the GIMP" === zyga_ is now known as zyga-afk [16:54] ralsina, agreed. I see the argument against but prefer the tradeoff of being clearer what the state is... And not needing '-1' or '0' or 'inf' with corresponding explanation [16:55] * ralsina goes research why git is obscene [16:55] mmcc: right [16:55] Git: "3. A pubescent kid who thinks it's totally cool to act like a moron on the internet, only because no one can actually reach through the screen and punch their lights out." [16:56] I would say that is accurate description of git, the software [16:56] anyway, +1 on making mandels' daemon a project. I don't forsee any major reworkings - ie, ObjC should not be a problem, and any other adjustments are OK to make in situ [16:56] ralsina: git is a britishism [16:57] ralsina: there is a few bytes of information buried in all the wikipedia noise here: http://en.wikipedia.org/wiki/Git_(British_slang) [16:57] Oh, but it's only obscene for the british and they are too polite to mention it. [16:58] "git" is the sort of word that Harry Potter can use about Draco Malfoy. It's an insult, but it's alarmingly mild. [16:59] thanks aquarius - I knew there had to be someone around who could explain that better than me [16:59] aquarius: still sounds like people talking about git, the VCS [16:59] * aquarius laughs [16:59] * ralsina has a mild dislike for it [16:59] at some point I shall create a vcs called "bellend". [16:59] ralsina: the vcs is insulting and mildly alarming? [17:00] mmcc: I now will always imagine git as having Draco Malfoy's face from the 1st movie [17:00] mmcc: and indeed, git's error messages are an insult [17:00] ralsina: you're welcome [17:01] mmcc: ok, let me burrow back under a bug [17:02] heh, git [17:02] I think puto for a vcs is great [17:02] puto pull branch!! [17:02] puto diff! [17:02] jejeje [17:02] puto blame [17:02] commit puto commit :P [17:02] heh [17:03] mandel: of course a spaniard created the "puto amo" wm [17:03] ralsina, we arr like that ;) and with the economical situation you can imaging the amount of swear words in the pubs [17:04] mandel: do like we did... go burn a few banks, you feel better in the morning. [17:04] ralsina, if we are lucky critina will nationalize them hehehe [17:04] * ralsina makes a sworn statement to the effect that he has nver, in fact, burned a bank or any other kind of building [17:05] ralsina, no worries, we understand, you're just going through some Mercurial Phases ;-) [17:05] teknico: hehe [17:06] teknico: I would revert my statements but git says I should do a checkout instead [17:06] ralsina, don't listen to the silly git [17:07] ralsina: fwiw, if you're making the change I suggested to the bw limit panel UI, bug 847227 might be relevant to your interests [17:07] Launchpad bug 847227 in Ubuntu One Control Panel "Disabling upload/download limits shows '-1' " [Undecided,Incomplete] https://launchpad.net/bugs/847227 [17:07] but git's father gave us Nimbus 2001 broomsticks! (ok, this is getting too convoluted) [17:07] mmcc: yeah [17:08] mmcc: I will use that as my "fix bandwidth limits" bug [17:08] ralsina: also relevant - bug 847233 [17:08] Launchpad bug 847233 in Ubuntu One Control Panel "Give users up/down sync speed suggestions" [Undecided,Incomplete] https://launchpad.net/bugs/847233 [17:09] ok, EOD for me tom create a project add tests to the daemon, add a python lib to talk with it :) [17:09] all, laters! [17:09] eww [17:09] ralsina, embrace the git [17:09] mmcc: that one needs an amount of magic we don't have [17:10] mandel, bye! [17:10] also, 1024 kbps is pretty slow [17:10] dobey: it's actually KiB/s [17:10] dobey: that was another bug ;-) [17:10] dobey, you ewing at the google music screenshot in that bug? [17:10] ralsina: i know, and it's still slow [17:11] mmcc: bandwidth limiting in general [17:11] dobey: depends. It's 3 times my bandwidth [17:11] gatox, a2! [17:11] dobey, amen [17:11] ralsina: it's 1/50th of my bandwidth (and soon, probably will be 1/150th) [17:11] dobey: he, good for you, I guess [17:12] ralsina, only good thing to say to an american with faster internet: or food is better and our women better looking, also, we do no have you crazy creationists.. [17:12] has mandel had Texas BBQ? [17:12] mandel: well you guys have a king who hunts elephants [17:13] i've been to buenos aires a few times. they do have those crazy creationists [17:13] dobey: we have like, 5 of them [17:13] ralsina, and gets hurt trying hehe [17:13] mmcc, no, but I can try next time they make me travel [17:13] ok, really going, is getting to warm in the office, laters! [17:13] mmcc: have you had alligator, llama and sort-of-hamster and ostrich in the same dinner? We did last sprint! [17:14] heh [17:14] i have had an alligator sausage [17:14] also rattlesnake sausage [17:14] ralsina: next sprint, we'll have to have some quetzal [17:14] briancurtin: that one I have not had. Of course who's going to *import* rattlesnakes for sausage-making purposes [17:15] dobey: oh, we missed the chance to eat quirquincho (closest thing to a pangolin around here ;) [17:15] heh [17:15] * gatox heard sprint? [17:15] :P [17:15] gatox: no, no sprints yet ;-) [17:16] oh man i should go to Hot Doug's this week - Smoked Yak Sausage with Roasted Garlic Dijonnaise and Horseradish Havarti Cheese [17:16] ohhhhhh :( jeje [17:16] ralsina: *import* ? they let you do that in argentina now? [17:16] dobey: only things you shouldn't. Rattlesnakes should qualify. [17:16] heh [17:16] I expect "rattlesnakes for everyone" to happen any minute now. [17:17] "Let them eat snake." ? [17:21] briancurtin: I've heard of hot doug's. I'm jealous. [17:26] dobey, re: bw limiting - I agree it's gross, and I doubt people ever know what value to set it at. So do you know of a way to 'nice' network traffic? Seems like it ought to be possible. [17:28] mmcc: not in a cross-platform way really, or with twisted (which is causing us problems with how current stuff is implemented anyway) [17:29] yes, this is at best half a feature nowadays [17:29] there's no nice way to know how much bandwidth is available, or how much of that you should be using, exactly [17:30] dobey, yeah it looks like bsd can have priority queueing at the network interface, but maybe that's not portable [17:30] right, everyone has some sort of QoS, and they're all different [17:30] mmcc: on linux you have tc which lets you do it per-host or per-protocol [17:31] mmcc: on windows you have nothing AFAIK [17:32] ok, so I'll make some notes then forget about this for now... :) [17:32] and trying to implement some kind of detection-and-backoff is a nightmare if two apps do it at once -- you get the "you go ahead", "no, you go ahead", "no, YOU go ahead" thing [17:32] omg....... this is so very broken...... [17:33] deadlocking through overpoliteness. How very British :) [17:33] aquarius: oh yeah, the right place to do it is definitely using the system traffic shaping, IMO [17:33] mmcc: ya, but fiddling with system settings from a user app is nightmarish in itself. :) [17:34] writing code for users is nightmarish in itself ;) [17:34] aquarius: so true! I'm hoping there's API [17:34] aquarius: or preset queues so we can just pick a lower priority and try to flood it, letting the system figure it out [17:34] we can just fix syncdaemon so that you *must* run it as root :P [17:35] transmission has a "tortoise" button which just slows things down -- what it actually does is fill in low numbers into their equivalent of our bandwidth-limit dialog, of course, but maybe that'd be a better approach -- just have "normal mode" and "be slow mode", and don't necessarily say what exactly "be slow mode" does [17:36] aquarius: hm. not a bad idea. probably an improvement over asking users to guess a number [17:36] it would obviously be better to just Do The Right Thing always without asking, but that sounds like it's not possible, and having an explicit be-slow button gives the sense of control back to the user rather than having to just trust U1 to be Right [17:37] i thought all you had to do in transmission to be slow was try to grab an unpopular torrent [17:37] you know, something legal [17:37] heh. be-slow-even-if-you-have-the-option-of-being-fast mode, if you prefer :) [17:37] heh [17:38] I mean, I have no idea what number I should fill into the box; I have no clue what my bandwidth actually is, nor do I know a good way of finding out :) (Note: speedtest.net is not a good way of finding out.) [17:38] 42 [17:39] That dialog is FULL OF LIES. It says specifically "if you set the speed to 0 syncing will be stopped" but what it does is set it to 1 because you can't set it to 0! [17:39] * ralsina sets the lower limit to 1 and if the ser wants to stop it, he can click "STOP" [17:40] there is no stop, is there? there's "completely disconnect the syncdaemon", but that's not the same thing as "allow downloads but not uploads" or vice versa [17:40] aquarius: you can't allow downloads but not uploads, or reverse [17:40] aquarius: I wonder what kind of control people actually want. Not sure I'd ever use "download but not upload" [17:41] ack, late for lunch. back in a while [17:41] dobey: I thought that was the point of having two separate boxes [17:41] aquarius: you cannot stop either one from here wither [17:41] aquarius: since you cannot actually set them to 0 [17:41] aquarius: ideally, yes. practically, no. [17:42] aquarius: because it's for *all* protocol traffic, not only file contents [17:42] aquarius: so blocking download of data would mean the server can't tell the client ot send the data for an upload :) [17:42] mmcc: I'm going here on hearsay -- there seem to be, from hearing people ask about the bandwidth limiter, two reasons it's wanted. The first is "u1 eats all my bandwidth; make it not do that", which is the go-slow-mode button, IMO. The second is "it's really important to me that these few files sync up to U1 before I shut my laptop down; I do not care about it downloading all my music which has been going on now for [17:42] five days" [17:43] dobey, ralsina: aaah. Didn't realise it limited the protocol too :) [17:43] aquarius: right, it shouldn't. but it does. becuase twisted. [17:44] also, i wonder who is going to port protobuf to python3 [17:44] not it! [17:45] * dobey also wishes forms on google docs were more useful [18:02] mmcc, lisettte, alecu, gatox: tell me this is not a ton prettier: http://www.flickr.com/photos/ralsina/7204467152/in/photostream [18:04] ralsina, +1 [18:12] ralsina, yes, grayed out is much better than "-1" [18:46] back. [18:46] ralsina: nice, looks good [18:47] mmcc: about to propose [18:47] anyone, please review https://code.launchpad.net/~ralsina/ubuntuone-control-panel/minus-one-is-the-loneliest-number/+merge/105869 [18:48] * gatox reviewing.... [18:48] I am interested specially on missing tests and such, please be harsh [18:48] while we're talking control panel ui, I just had to reset my password and had a couple of notes regarding that. UI tweaks like the password "lint" speech bubble thing is not big enough, etc. . should I file bugs? [18:48] mmcc: yes please, with pictures :-) [18:48] ralsina: k, good. [18:48] mmcc: on what OS? [18:49] ralsina, Ubuntu 12.04 in virtualbox [18:49] gatox: also interested on knowing i the spinbox styling in that branch looks good on windows [18:49] ralsina, ack [18:50] ralsina, this is unbelievable..... fixing a bug...... take me to another one..... and so on since 3 hours ago....... [18:50] :S [18:50] ralsina, not bug actually...... just making things work in mac [18:50] gatox: heh [18:51] gatox: happens on immature code, deep breaths [18:51] dobey, cylon! jeje [18:51] o/~ one bug, leads to another o/~ [18:51] gatox: destroy all humans! [18:51] hehe [18:51] gatox: A wise puppet once said "Fear leads to anger. Anger leads to hate. Hate leads to suffering." [18:52] You are on step 2 already :-) [18:52] Suffering leads to the pub. [18:52] the pub leads to alcohol poisoning [18:53] gatox, what problems are you hitting? talk it out, it'll make you feel better :) [18:54] hooray me [18:54] mmcc, hehe na, it's just a chain of imports problems, we have in a lot of places if platform == "blabla" do something or another thing..... but in mac the else part is always wrong now.... we have to add those implementations, and a lot of things like that [18:55] so i have poauth ported to python3 [18:55] in so much as the tests path on both 2.x and 3.x [18:55] gatox: oh yeah, I saw that in lots of places. lots of small changes [18:55] mmcc, so, i'm implementing some.... only declaring the classes or methods in other cases.... to try to have things working.... and then start filling the empty implementations. [18:55] gatox: sounds good [18:56] dobey: poauth? [18:56] dobey: nice work [18:56] Chipaca: ^^ [18:56] ralsina: the fork of python-oauth i made a few years ago, that i never got around to finishing [18:57] ralsina: but python-oauth is basically unmaintained, and has no tests [18:57] dobey: well,there's python-oauth2 [18:57] ralsina: which doesn't do oauth 1.0 [18:57] or 1.0a [18:57] dobey: which 1.0a? Ok, old stuff ;-) [18:57] and i think porting the entire world of ubuntu to oauth2 is probably more work than porting u1 client bits to python3 :) [18:58] oauth2 does support oauth 1.0 [18:58] it's in their example code, even :-) [18:58] and it claims to have 100% coverage [18:58] so it lies :) [18:59] anyway [18:59] dobey: it has a 'oauth_version' parameter to make it do oauth1 (but yes, it lies about coverge, surely) [19:00] ralsina, i'm testing your branch IRL..... and the 0 is still there...... also, when it's in -1 and you click the checkbox to limit the speed..... shouldn't set the minor value by default or something?? because you will have limit with -1 [19:00] and it's probably also not ported to python3 [19:00] ralsina, i don't know if 0 is a valid limit value...... just saying [19:00] saying [19:02] dobey: do you know "nose" ? Cool test runner. [19:02] gatox: it should not let you set it to 0 [19:02] ralsina, ahhhhhhh wait..... my bad...... forgot the pythonpat [19:02] gatox: or -1 for that mtter [19:02] gatox: hehe [19:03] dobey: has a plugin so that if it fails more than 8 times it looks like FFFFFFFFUUUUUUUUUUUUUU [19:03] ralsina, ahhhhhh now it's nice! :D [19:03] gatox: hehe [19:04] ralsina: there's also a plugin that if your test is too many lines, it raises CoolStoryBroException [19:04] hello stranger thisfred from the land of sprints! [19:05] briancurtin: ha! [19:05] dobey: yes it is (kinda) https://github.com/simplegeo/python-oauth2/pull/51 [19:06] ralsina: ohai [19:06] thisfred: how's cucumber sandwiches and u1db today? [19:07] we had a productive day today, at least if measured by number of landed branches [19:07] ralsina: off for dinner with lucio now [19:07] I just tried joining ubuntu one music streaming but the payment failed without any way to report this, does anyone know who to talk to? [19:09] My bank/mastercard requires "mastercard securecode" for internet payments and Im guessing this is where things went wrong [19:12] ralsina, your branch from my POV is +1..... but i couldn't test it IRL on windows..... i need to check what is happening with my vm..... sso is failing [19:12] alecu: bug #978903 is bad, scary, ours and needs to get fixed. What do you think? [19:12] Launchpad bug 978903 in ubuntuone-client (Ubuntu Precise) "[precise] Client is stuck due to Upload executing before MakeFile" [High,Triaged] https://launchpad.net/bugs/978903 [19:13] gatox: ok! [19:13] briancurtin: could you take a look at this branch in windows and check everything works as it should? https://code.launchpad.net/~ralsina/ubuntuone-control-panel/minus-one-is-the-loneliest-number [19:13] ralsina: that bug is fixed already [19:13] ralsina, I think we should discuss it with facundobatista. [19:13] dobey: really? [19:13] ralsina: see the 2 branches listed as merged on that bug [19:13] ralsina: yep, will look in a min [19:14] ralsina, also: I need to run to kinder; I'll take a deeper look afterwards [19:14] ralsina: it's not released yet though. was blocking doing release on figuring out the massived windows and ssl issue [19:14] dobey: ack then [19:14] dobey: well, that seems to be cleared now [19:15] dobey: as in, it happens to a very limited set of users [19:15] ok [19:15] so we can probably do a release/sru tomorrow then [19:15] dobey: you maybe missed it, but there is at least one report of itbeing locale-dependent (no, makes no sense) [19:16] dobey: yes, let's do it tomorrow [19:16] well maybe openssl has round ceilings [19:16] dobey: ha [19:17] dobey: well, the round ceiling bug is gone in python 2.7 also [19:17] dobey: have not bothered chcking in older pythons though [19:21] I sent a support question on the payment failure [19:22] duanedes1gn: can you help arune? ^ [19:22] ralsina: +1 on your branch [19:22] briancurtin: cool :-) [19:23] marked as approved [19:23] briancurtin: don't approve globally please [19:23] oops... [19:23] oh, well, nevermind ;-) [19:23] i saw gatox approved and i was the second so i figure it was good. should i undo? [19:23] * ralsina wanted to give design a chance to see it, but now they will have a chance to see it in nightlies instead, no prblem :-) [19:23] duanedes1gn: my supportticket got #15330 [19:25] while I'm here I might as well ask for help restarting my ubuntu one client/daemon/service [19:25] one of my computers does not sync [19:27] someone? [19:28] arune: if you want to restat it all open a terminal and run command u1sdtool -q [19:29] then start again with : u1sdtool -c [19:29] ah, I need a display for that, sorry, have to try when I get home then [19:30] thanks duanedes1gn, will try [19:30] who wants a quick QT UI bug? bug 999885 [19:30] Launchpad bug 999885 in Ubuntu Single Sign On Client "Password assistance QLabel is too small for its text in the reset password dialog " [Undecided,New] https://launchpad.net/bugs/999885 [19:31] arune: you could do "DISPLAY=:0 u1sdtol -c" [19:31] mmcc: dibs! [19:31] mmcc: a tiff????? [19:31] ralsina: is that a problem? [19:31] mmcc: did you scan the screen or something? ;-) [19:32] mmcc: not a problem, just... old fashioned? [19:32] ralsina: thanks! [19:32] OS X default screenshot utility "Grab" writes TIFF. I think it dates from NeXT, maybe [19:32] that's right, it's a screenshot of a VM [19:33] I guess I can check if it syncs now by watching a folder I know should be filled with pictures from my phone? [19:33] duanedes1gn: thanks, please also take a look at my payment issue, I want to support ubuntu but also have a great service [19:33] arune, did you tick the box to sync the new folder with pictures to your desktop? [19:35] beuno: yes, now after restarting the service it syncs [19:35] thanks for help with that duanedes1gn and ralsina [19:37] duanedes1gn: thanks for fast email reply! (but please fix fast, I want to PAY :)) [19:37] :) [19:38] arune: actually, our web/server hackers are doing a lot of work to improve payments right now. :) [19:41] great work guys! [19:42] gatox: you get bug #999885 when you want to rest from refactoring [19:42] Launchpad bug 999885 in Ubuntu Single Sign On Client "Password assistance QLabel is too small for its text in the reset password dialog " [Medium,Triaged] https://launchpad.net/bugs/999885 [19:42] ralsina, ack... probably tomorrow.... now i'm close to get the tests working :D [19:43] gatox: sure, it's not urgent [19:44] and [19:44] twisted 12.0 is landing in quantal now [19:44] as a heads up [19:45] can I get a review on a very trivial merge proposal for control panel? No TIFFs this time: https://code.launchpad.net/~mikemc/ubuntuone-control-panel/fix-999893/+merge/105886 [19:45] mmcc, ack [19:46] gatox: thanks [19:47] mmcc, +1 :P [19:47] gatox: do we need two reviews for that? who else gets to read through all the one line diff? [19:47] mmcc, i don't think so [19:48] gatox: if trivial, one is enough [19:48] ralsina, really trivial..... and we are not in any kind of freeze right now [19:48] also, damn, that is going to force translators to re-translate that string :-( [19:48] a string change [19:48] ralsina, yap [19:48] ralsina: it is? it only removes a space :< [19:48] :( [19:49] mmcc: yes, those are keys for the translation file [19:49] mmcc: so now it loses the old key with all its translations and has a new one [19:49] ralsina: oh yargh. the keys shouldn't be the content of one of the translations! [19:49] mmcc: well, it's how it works since time immemorial: "C" locale is US english [19:50] and immemorial means, like, 1990 [19:51] ralsina: urgh. well, there are ups and downs - Cocoa makes you have a separate string for the key and english is "just one translation" (although it's the only one created by default) [19:51] mmcc: that is the right thing to do. But it's not the way of gettext (although you *can* create a en_US.po I suppose) [19:51] so then you get things like (i forget the actual call) string = NSLocalizedString(@"Edit account label string", @"Edit account label") [19:51] which is ugly [19:52] ralsina: oh. yeah, sounds like the same deal [19:52] just the cocoa default new app templates create an english localization for you [19:53] well, bummer about all that extra work I just created :\ [19:56] ralsina: well, gettext default is not en_US exactly. :) [19:56] mmcc: wel, it happens, it was wrong, so better now than on deadline [19:57] dobey: yes, it's C, I know. I will someday do an app where C is dutch. [19:58] * dobey writes some python that doesn't follow pep8 [19:58] * ralsina mails dobey a bobcat [20:00] cool. the panther could use a friend [20:07] mmcc, are you able to build ubuntuone-storage-protocol on mac? [20:07] (python setup.py build) [20:07] gatox: don't think I've tried, let me go see [20:07] mmcc, please! bzr branch lp:ubuntuone-storage-protocol [20:08] gatox, no - I don't have the protobufs compiler installed anywhere [20:09] mmcc, good, it's not only me...... so i need to fix that now [20:09] mmcc, thanks [20:09] gatox: sure. I guess that should be part of the buildout? [20:09] mmcc, actually....... if it is what i think...... it's in the eggs folder...... so..... [20:09] weird [20:10] gatox: yes the buildout puts protoc in the path, but maybe it needs a tweak for mac [20:11] ralsina, yap..... my next task! little by little this is getting better :P [20:11] ralsina: gatox: % find . -name protoc -print ----> *crickets* [20:12] mmcc, inside the eggs folder in the buildout: protobuf-2.4.1...... [20:13] gatox: I know, I was just pointing out that I have no 'protoc' executable. [20:13] ahhhhhh [20:13] gatox: if I 'find -name "proto*" -print I get the stuff in the eggs and the protocol def in our code... :) [20:14] so I was calling ralsina a liar ;p [20:14] * mmcc ducks [20:14] ok.... that's all for today, eod.... time to rest and keep watching the harry potter marathon :P [20:14] see you tomorrow people! [20:14] gatox, enjoy! bye [20:15] mmcc, bye! [20:20] ralsina, can I just approve that one-line string fix myself? gatox left it as needs-review [20:22] mmcc: yes [20:22] dobey: ok, thanks [20:29] * dobey wonders how to skip tests cleanly on plain python unittest [20:30] dobey: unittest.skip [20:30] http://docs.python.org/library/unittest#skipping-tests-and-expected-failures [20:31] dobey: http://pybites.blogspot.co.uk/2009/03/unittest-now-with-test-skipping-finally.html [20:32] brought to you by "minibar, now with japanese beer" [20:33] thisfred, Asahi? [20:33] hai [20:33] which is also what we had at dinner at the noodle bar [20:34] thisfred: doesn't work on python 2.6 [20:34] added in 2.7 [20:34] so, don't develop on red hat? [20:34] thisfred: and it seems to raise a SkipTest, which results in an ERROR rather than SKIP [20:34] thisfred: lucid [20:34] EOL [20:34] thisfred: not until next april [20:34] close enough for me [20:35] and that's only for non-server [20:35] hardy is even still supported for server :( [20:35] dobey: luckily precise will only be supported for 75 years [20:35] heh [20:36] lifetime guarantee! [20:36] a propos of nothing: wtf has happened to BBC2? It's 100% reality TV now? [20:37] heh [20:44] ok, I have to head out for a couple of hours but I'll be back again tonight for a bit, when alecu is done with his review of https://code.launchpad.net/~mandel/ubuntu-sso-client/use-unix-domain/+merge/104244 - I have one small and one medium MP to put in. [20:51] mmcc: if you have them ready you can just propose with prerequisite [20:51] mmcc: and then we see them in order [20:57] EOD for me [20:57] see you guys tomorrow! === zyga_ is now known as zyga === zyga__ is now known as zyga === zyga is now known as _zyga === m_conley is now known as m_conley_away