[00:02] Night everyone! [00:05] ralsina, cool! [00:05] alecu: then the "watch in [00:05] ralsina, I've proposed it for merging, I'll send an email asking for reviews tomorrow. [00:05] ~" kicked in and everything stands still [00:05] alecu: cool [00:05] re: "watch-in" great! [00:06] If mandel has the watches fix, I think it's a good point for a release of some sort [00:08] alecu, doesn't your branch mean we now need to do a branch on u1cp to connect only to the signals we need? === yofel_ is now known as yofel [00:22] ralsina, no, we don't need that. The ipc client is all contained on the ipc_client.py inside syncdaemon, and that's what's used inside u1cp [07:06] morning all! === zyga-afk is now known as zyga [10:52] Good morning guys [10:54] hi [10:55] who works on the ubuntuone-installer? [10:57] mandel ↑ ? [10:59] I should have guessed it was mandel, as it's always him who breaks stuff :D [11:00] rodrigo_: I meant "mandel, do you know?", I'm sure ralsina or nessita will know, but I don't think they're around yet. [11:00] ah :) [11:00] :) [11:00] ok, I'll ping them later [11:01] rodrigo_, why? [11:01] rodrigo_, I'm sure I can give you a hand :) [11:01] aunque a un desertor? ni agua ;) [11:02] mandel, the .desktop file is wrong, it has the same categories as the ubuntuone-control-panel-gtk.desktop file, so they both show up on the control center [11:02] also, the X-GNOME-Settings-Panel should be removed from ubuntuone-installer.desktop [11:02] mandel, :D [11:04] rodrigo_, in which project is that? in the ubuntuone-control-panel or in the unbutuone-windows-installer? [11:04] hmm, no idea, let me check [11:05] $ dpkg -S /usr/share/applications/ubuntuone-installer.desktop [11:05] ubuntuone-installer: /usr/share/applications/ubuntuone-installer.desktop [11:07] good morning! [11:08] dobey: worked on it, but he's on holiday [11:08] and no I am not working yet ;-) [11:08] rodrigo_, yep, that is the new installer being added for linux so that we do not ship in the cd... [11:09] rodrigo_, we haves sooooo many projects? and I've been told you are porting it to meego, right? [11:09] :P [11:09] no, to Windows 3.1 [11:09] the installer shouldn't be on the control center [11:10] ralsina, this is for you: https://code.launchpad.net/~mandel/ubuntuone-client/eq-add-ancestors-watches/+merge/73775 [11:10] ralsina, can you do a IRL to see if its ok :P [11:10] ralsina, that should fix the watches issues [11:10] rodrigo_, do not say to high :) [11:10] :) [11:14] ralsina, ping [11:16] Hola! [11:17] facundobatista, buenas! [11:17] facundobatista, do you need reviews? [11:18] mandel, not for client, thanks [11:20] facundobatista, cool [11:27] ralsina, ping [11:27] * nessita starts the review run [11:30] gatox: ping [11:32] nessita, pong [11:32] gatox: can you please attach the bug # to https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/reset-password-page/+merge/73522? [11:33] nessita, ok [11:37] gatox: and one ssmall fix to reset-password in sso: I know this is correct because is alphabetical ordered: [11:37] from PyQt4 import QtCore, QtGui [11:37] but this causes some times ugly errors like: [11:37] basenames = [base.__name__ for base in member.__bases__] [11:37] TypeError: C++ type 'QWidget*' is not supported as type [11:37] gatox: so, I'm approving, but before landing, change the Qt import to: [11:38] from PyQt4 import QtGui, QtCore [11:38] nessita, weird....... ok....... changing [11:39] nessita, done [11:42] mandel: ping [11:42] mandel: are you reviewing alecu's branch? [11:44] nessita, yes, just added a needs fixing [11:45] nessita, I think that adding all the remote_calls is dangerous since the signals are a subset of all the remote calls [11:46] nessita, I pushed changes to https://code.launchpad.net/~mandel/ubuntuone-client/add-method-name/+merge/73400 and added the new MP I mentioned in the email, please feel free to hurt my feelings ;) [11:48] ok [11:48] nessita, I need to walk the dog, any comments can you add them to the MP :) [11:48] sure [11:48] I'll me 30 min or so [11:48] * mandel walking beast [12:14] ralsina: ping [12:22] nessita, we talk yesterday that the size of the wizard should be: width=736.... and, do you have in mind which should be the proper height? [12:31] mandel: did you test IRL the ancestors branch? [12:39] gatox: ping [12:39] gatox: sorry, I just saw your question [12:39] nessita, pong [12:40] gatox: ideally, 736x525 [12:40] nessita: pong [12:40] gatox: can you please propose a quick branch for ussoc to solve this? [12:40] == Python Lint Notices == [12:40] ubuntu_sso/qt/tests/test_reset_password.py: [12:40] 24: [F0401] Unable to import 'ubuntu_sso.qt.reset_password_ui' [12:40] 24: [E0611] No name 'reset_password_ui' in module 'ubuntu_sso.qt' [12:40] gatox: nightlies are not building because of that [12:40] nessita, ok [12:40] nessita, nice!!!!! now the size is: 740x538 [12:41] gatox: GREAT! [12:41] that's GREAT [12:41] gatox: I think there is no need to put more time in this for now [12:41] nessita, ok...... working in the branch you mention now [12:41] gatox: thanks (just needs pylint disables) [12:45] alecu, mandel, gatox: Who needs a review? I am selling! [12:45] and nessita too, of curse [12:45] of COURSE [12:45] * ralsina is not cursing [12:45] ralsina, not yet...... almost done!!! [12:46] ralsina: we should talk about this https://code.launchpad.net/~sidnei/ubuntuone-windows-installer/sourcedeps/+merge/73728 in the mumble, before reviewing. I'm not sure if the right solution. [12:50] ralsina, nessita review (if you want, just 2 lines to disable and enable lint to fix some lint issues): https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/reset-lint-fix/+merge/73800 [12:50] gatox: yes! [12:50] gatox: I got it [12:51] * ralsina suggests gatox that he doesn't ask for reviews until after the diff is visible ;-) [12:51] ralsina, ok...... understood [12:52] looks like my cat is having kittens uder my desktop. Anyone wants one? [12:52] ralsina, I'm sure gatox would be interested [12:53] gatox, want some gatitoxs? [12:53] alecu: you're on fire today! [12:53] k-ching! [12:53] ralsina, sorry..... i already have one! [12:53] alecu: question, did you run the windows suite for facundobatista's branch from yesterday? [12:53] (client branch) [12:53] ralsina, and don't play well with others :P [12:53] nessita, very good question. I think I did not. [12:54] shame on me [12:54] boy launchpad is slow to parse branches today [12:54] alecu: yes, because that branch added 7 failures! already debugged and fixed :-) [12:54] nessita, sorry about that :/ [12:55] nessita, anyway, I'm still not getting all tests passing around here [12:56] alecu: I know... [12:56] alecu: we need to debug that (timing issues evdently) [12:58] alecu: "good news" is that we can use the same box as parameter, seems pretty reproducible [12:58] both the non-failures and the failures [13:00] gatox: I am flipping to approved your reset-lint-fix branch [13:00] ralsina, okkk [13:00] ralsina, i have my branches to fix the installer size ready...... just waiting for launchpad to scan them [13:01] standup? [13:01] me [13:01] me [13:02] alecu, nessita mandel standup? [13:02] me [13:02] alecu, mandel? [13:03] me [13:04] gatox: go! [13:04] DONE: [13:04] Branch to fix Installer sizes, now the size is: w: 740, h: 538. [13:04] TODO: [13:04] Keep killing UI bugs. [13:04] BLOCKED: [13:04] No. [13:04] ralsina, go [13:04] DONE: implemented proof-of-concept of the systray icon, worked on stderr fix, lots of IRL testing, reviews, call. TODO: finish stderr, finish systray, reviews, testing, package a release candidate. BLOCKED: waiting for all the needed branches to land before packaging of course [13:04] nessita! [13:04] DONE: landed branch for bug #814113, meetings (TOO MANY), reviews [13:04] TODO: reviews using the new schema (already did 2 hours of review, now switching to coding!). Work on bug #807021 and bug #834766 (I need to solve both to have a clean solution). [13:04] BLOCKED: nopes [13:04] NEXT: alecu [13:04] DONE: proposed a branch to fix the multiple ipc connections issue [13:04] TODO: fix some issues that mandel has regarding this branch [13:04] BLOCKED: no [13:04] NEXT! mandel [13:05] * nessita adds TODO: fix nightlies builds [13:06] nessita, ralsina this branches are ready for review: https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/wizard-sizes/+merge/73803 - https://code.launchpad.net/~diegosarmentero/ubuntuone-windows-installer/wizard-sizes/+merge/73804 [13:06] gatox: queueing those up for after lunch [13:06] nessita, ok [13:06] orry [13:06] gatox, nessita: I'll do a first review of those now [13:07] irc issues [13:07] ralsina, nessita the one from sso is really trivial [13:07] ralsina, gatox: sso one was trivial, so I just approved it [13:07] gatox: yeah, I am setting that one to approved as soon as I see how it looks [13:07] nessita: beat me to it :-) [13:07] DONE: fised issues with add-method-name branch. Proposed fix for the watcher issue. Reviews alecus branch. [13:08] TODO: get fixes to possible review. Look at bug assigned to me [13:08] BLOCKED: no [13:08] mandel: did you test IRL the ancestors branch? [13:08] gatox: what the heck is the difference at line 59 about? [13:08] * gatox looking.... [13:09] ralsina, thats the content of license page...... [13:09] gatox: yes, but what changed? [13:09] I can't parse it :-) [13:09] ralsina, ........ nothing....... [13:09] qt designer.... [13:10] ralsina, do you want me to revert that and try to apply the change of the margin only? [13:10] gatox: can we have that in a variable in the code? [13:10] gatox: nah, I don't mind, I just assumed you had changed the text [13:10] ralsina, nono [13:10] good idea, and with reasonable HTML? [13:10] gatox: so we can avoid stuff like this [13:10] as in not specify font sizes and MS Shell Dlg font [13:11] ralsina, nessita, ok...... moving it to a html format [13:11] gatox: cool [13:11] nessita, no, I did not IRL, did you find any issues? [13:11] nessita, if you did, the tests are wrong then :( [13:11] ralsina: can I have a trivial for https://code.launchpad.net/~nataliabidart/ubuntu-sso-client/build-lucid/+merge/73805 ? (fixes a lucid nightlies build failure) [13:11] nessita, I can do it though [13:11] nessita: sure! [13:11] mandel: I did not, but I would like you to do an extensive IRL test [13:12] mandel: I want you to be confident about the fix :-) Ideally, please first reproduce the bug with trunk, and then use your branch to test [13:12] nessita, sure, I'll test with my test accound with different UDFs created to ensure that no watches are added [13:12] mandel: thanks! (be sure to first reproduce with trunk) [13:12] nessita: +1 and set it to approved [13:12] nessita, sure? although I'm confident hehehe [13:12] great [13:12] alecu, ping [13:13] mandel: I'm not :-D [13:13] ralsina, where should i put that variable?? gui.py? [13:13] mandel: but I trust no one (not even me) [13:13] gatox: yes [13:13] ralsina, ok [13:13] that's where we have internationalized stuff, IIRC [13:13] gatox, ralsina: or the gui/qt/__init__.py? [13:13] nessita, hahaha I told you you were free to hurt my feelings ;) [13:13] where all the other _() are [13:13] where they are :-) [13:13] nessita, yep.... maybe that is better [13:13] mandel: is not personal, I think we all mae mistakes, always [13:14] made* [13:18] gatox: +1 on wizard-sizes, it looks great! [13:18] ralsina, thanks! [13:18] I mean +1 after I finish running the tests of course [13:24] mandel, pong [13:24] mandel, I'm working on fixing the issue you commented. === m_conley_away is now known as m_conley [13:32] gatox, everything passes, +1 [13:32] ralsina, nice [13:38] ralsina, nessita i've moved the license content into a variable as we speak...... now i have to run to the bank!!!!!! [13:38] gatox: ok, I'll take a look now [13:38] I approved it before that. Boy, I miss coffee [13:41] ralsina: question [13:41] nessita: shoot [13:41] ralsina: is there any way to pass custom params to widgets when creating them building them from ui files? [13:41] (not sure if the question is clear) [13:42] not sure what you mean [13:42] ralsina: I need to make all widgets have the same backend instance, can I create it in the top level widget and pass that to all the children? [13:43] you can add a parameter to the __init__ of the class that calls setupUi [13:43] but then when you add them with designer that's not used, so no [13:43] you would have to make the widget use it from parent() [13:44] like self.backend = parent().backend [13:44] That way it propagates down the widget tree [13:47] ralsina: ok, I'll play with that [13:47] nessita: OR you can use app as a "global singleton" [13:47] Since it's guaranteed that QApplication.instance() will always return the same, you can even do a subclass that has the backend [13:48] ralsina: hum... [13:48] ralsina: I think I prefer the first [13:49] I think the second one is a bit cleaner, but the first one is easier === gatox is now known as gatox_away [13:52] mandel, alecu: need any reviews? [13:54] ralsina, besides the watches one no? and I'm doing an IRL sets of tests of it [13:54] mandel: ack [13:55] alecu, I just wanted to ask you if it made sense? [13:55] alecu, the issue I mentioned I mean :) [13:56] mandel, yes, it makes sense. I'm working on it already. [13:56] ralsina, not right now, thanks. [13:56] alecu, cool :) [14:01] ralsina, i can't go to the bank until lunch..... do you have any bug that you prefer i take?? or i just pick one from the list? === gatox_away is now known as gatox [14:01] gatox_away: the page right after login in the installer [14:02] gatox: it should show a "skip setup" button [14:02] gatox: lisette told me about it but I am not sure it has a bug number [14:02] ralsina, alecu, nessita, Chipaca, gatox mumble? [14:02] ralsina, ok..... i'll take a look at that [14:02] mandel, fine by me [14:02] yes! [14:04] Chipaca: mumble? [14:05] going [14:07] aleeeeeecuuuu alecu [14:07] uh, uh, uh [14:07] mumble [14:28] alecu, ralsina, nessita I've done the IRL AFAIK the bug is fixed, please review whenever you can :) [14:43] ok, I have played around and I have my UDFs working!!! [14:43] * mandel dances [14:44] nessita, alecu please let me know if you have any issues with that branch :) [14:45] ralsina, if there is nothing very urgent I'll continued looking at the bugs that have been asigned to me [14:47] mandel: what happened with the bug rick had? [14:48] mandel: could you try a share, too? [14:50] nessita, I have not been able to reproduce it, I wanted to talk with rick so that we tried again? I'll try a share and will get to it too see if I can reproduce again? maybe with a bigger file [14:50] ralsina, waht do you want me to try? [14:50] mandel: subscribe to a share and see if it's replicated correctly [14:50] ralsina, ok [14:50] mandel: can you please try to coordinate that with rick? [14:50] usually I got things locked before the share could start to download [14:50] nessita, yes [14:50] mandel, can you please re-review my branch??? [14:51] mandel: he should ne starting his work day now [14:51] alecu, will do asap [14:51] mandel, that should be urgenter than the bugs [14:51] mandel: could you do my laundry? [14:51] ;-) [14:51] ralsina, I can do you mu? wont say that to you ;) [14:51] alecu, ok, will look at your branh first, let me try shares first :) [14:52] ralsina, I have the aranduka books you shared with me working! [14:52] \o/ [14:52] cool! [14:53] ralsina, if you want to see if it work in your system here is the mp: https://code.launchpad.net/~mandel/ubuntuone-client/eq-add-ancestors-watches/+merge/73775 [14:53] alecu, I'm doing your review now [14:53] mandel, please, wait. I've just seen some broken tests. [14:54] sorry :P [14:56] alecu, ok, I was just doing the code review, so feel free to change it :) [14:58] ralsina, nessita another branch for review (help button): https://code.launchpad.net/~diegosarmentero/ubuntuone-windows-installer/help-button-removed/+merge/73834 [14:59] gatox: you already went to the bank? [14:59] (polite version of stop flooding with reviews :-P) [15:00] nessita, nop....... [15:00] jejeje [15:00] nessita, i have to go at lunch :( [15:00] ack [15:01] mandel, r1125, with the fixed test [15:01] ok [15:04] gatox: I don't understand the test for that branch [15:04] mandel, ping [15:04] ralsina, just checking that the wizard doesn't have the window flag for the help button [15:05] mandel, can it possibly be the case that the code in u.p.w.network_manager is butt ugly? [15:05] gatox: why not just flags & QtCore.Qt.WindowContextHelpButtonHint == 0 [15:05] ralsina, ............ shame on me [15:05] I believe the code in NetworkManager is the thread that's preventing twisted from stopping. [15:05] gatox: there are 10 kinds of programmers. Those who think about doing bitwise operations, and those who don't. [15:06] ralsina, right! jeje [15:07] alecu, is ugly, we could make it a lot nicer by hooking it to twisted like we did with the fs [15:07] alecu, but that is the way to get that the network is up or down, yet ugly [15:07] I was planing to clean that [15:08] mandel, I believe the pythoncom.PumpMessage() is never stopping [15:08] mandel, because it waits for a WM_QUIT that it's never receiving [15:08] alecu, that is possible, that code is ugly ugly [15:09] mandel, also: we are sending the "SYS_NET_CONNECTED" event at start, all the time. Even when the network is not connected. [15:09] ralsina, fixed :P [15:10] ralsina, launchpad not updated yey [15:10] yet [15:10] alecu, always? well that is not a problem due to the sd that will try to connect and twisted would say no no [15:10] gatox: also, line 10 looks wrong [15:10] alecu, which is the technique used when there is no NM [15:10] I mean, it works, but if by some miracle the original flags didn; t have that bit set it would break stuff [15:10] alecu, but that code certainly needs a second iteration [15:11] mandel, right. It's not such a big problem, but that takes some time [15:11] On 11.10, I'm still unable to use Desktopcouch in any way. I've been told here, that it's a known issue. Does anyone have a bug number that I can subscribe to? [15:11] alecu, nevertheless can you take a look at the watches branch first, there are few hours before my EOD [15:11] mandel, sure, sorry. [15:11] gatox: I think the right way is flags & ~QtCore.Qt.WindowContextHelpButtonHint [15:11] ralsina, yep..... i was about to do that! [15:11] gatox: cool [15:12] mandel, it's just that I saw the bit in ipc where the NM thread was started, and I guessed that was the issue with the dangling threads and the qt reactor not stopping. [15:12] mandel, so, I'm running the tests with IPCInterface.test=True, and my guess is that the reactor will stop just fine. [15:13] alecu, yeah? we initially had an issue with the looping call of the fs too, we need to hook that to twisted like the fs and listen to the stop event to kill the thread [15:13] alecu, no, it won't I dont think I was that smart when I wrote that [15:13] mandel, right. We should find a way to stop the thread [15:15] alecu, lets file a bug and do this after the release in our 'sprint' in BA [15:15] mandel, +! [15:15] mandel, +1 [15:15] :D [15:16] mandel, I think I'll try to fix this sooner. If I don't start the NM thread, I can stop syncdaemon with Ctrl-C [15:16] ralsina, done [15:17] gatox: that works without the int() calls? [15:17] yep [15:17] mandel, I think that the NM thread not stopping takes a toll on the reliability of SD [15:17] then you can remove it in line 23 :-) [15:18] alecu, true? it is bad when trying to stop it certainly? I'm nearly done with your review by the way [15:18] alecu, but please, review my branch asap, I wont be here next week :( [15:18] I'm on holidays [15:18] mandel, !!!! [15:18] mandel, I did not know [15:19] mandel, I'll review it asap, and I'll take a look at this thread next week [15:19] alecu, cool thx a lot!!! [15:21] gatox: "then you can remove it in line 23 :-)" just in case you missed it because I didn't say your nick [15:21] ralsina, i read that..... i'm checking if it is ok [15:21] ralsina, if i do that...... i can't compare it with 0 [15:21] ralsina, ....... maybe with QtCore.Qt.QWidget [15:22] Nah, it's ok, use the int() then [15:22] ralsina, ok [15:22] gatox: +1 [15:23] ralsina, thanks [15:23] now is time to go to the bank [15:27] alecu, +1 on your branch, I have approved the MP too [15:29] hmmm we are having a serious memory consumption problem in syncdaemon [15:33] ein? [15:33] mandel: up to 300MB and rasing about 500K/sec [15:34] mandel: probably not your branch's fault [15:34] mandel, revisate ésta: https://code.launchpad.net/~alecu/ubuntuone-client/make-it-stop [15:35] mandel, that is, try the branch by running the tests and seeing that they end up properly, and by running syncdaemon, and being able to stop it with Ctrl-C [15:35] mandel, if it works for you too we can make a proper branch out of it. [15:36] mandel, and don't worry: I'm reviewing your branch, and made that while the tests were running :-) [15:39] alecu, hehe I'm not worried, I know you are a man of your word... [15:39] ralsina, is this when testin my branch? [15:39] ralsina, is that a memleak? [15:39] mandel: yes, but I have seen it on trunk too [15:39] mandel: Iwould guess it is [15:39] mandel, and what word would that be? [15:40] alecu, lego? [15:40] :P [15:40] juas [15:48] hm? ubuntuone.logger is broken on windows? puto! [15:51] ralsina, this is ashot in the dark, but could this be the reason to the memory problems: http://paste.ubuntu.com/680660/ [15:52] mandel, I've added a few comments on the style while I finish the review: https://code.launchpad.net/~mandel/ubuntuone-client/eq-add-ancestors-watches/+merge/73775 [15:52] alecu, ok, on them [15:53] mandel: I was not getting that [15:54] mandel: I will take a better look once your branch and alecu's are landed [15:54] ralsina, I'm getting it on stderr [15:54] but I just killed it and it was using 1.3GB of ram :-( [15:54] ralsina, neverthless the handler is broken [15:54] mandel: you think it was keeping the backlog in memory? [15:55] mandel: that doesn't explain 1.3GB of usage :-) === beuno is now known as beuno-lunch [15:57] ralsina, it could? although 1.3 is waaaaay too much [16:09] ok, lunch for me === ralsina is now known as ralsina_lunch [16:10] alecu, pushed the style changes [16:12] alecu, I need to go for some time since otherwise the stores will be closed, I'll be back on time to make any change required to the review [16:13] mandel, I'm approving... but LP is so slow today. [16:14] alecu, hurray! [16:14] I mean, for the +1 :P [16:37] rye: was this https://bugs.launchpad.net/ubuntuone-android-files/+bug/806002 related to auto-upload or just manual upload of any file? [16:38] manual upload, don't think that anybody will want to blindly upload */* from the storage [16:39] rye: I mean, U1F supports uploading a file of any type. Just touch "plus/add" > File [16:39] rye: Is that what you meant? [16:39] rye: We didn't have that before, but I've added that some time ago, and probably forgot to update the bug status. [16:39] karni, let me check [16:42] karni, tested on Adobe reader - opened pdf, clicked share - no U1 [16:43] rye: ack [16:43] mandel: I'm doing your review now (ancestors) [16:43] karni, it works the other way, like you described, but it does not look like it is able to accept arbitrary mimetypes [16:44] rye: I just uploaded a ".dropsync" file (of unknown to me MIME type) from AndExplorer using Send option (they call it send, it's just share) [16:44] rye: I'll verify how we register the share option. [16:46]