[06:40] i cant connect to the Ubuntu One service, help? [07:55] morning all [09:45] ralsina: ping? [10:45] Is it holidays somewhere? [10:45] dammed… this is boring :P [10:46] mandel: want me to juggle for you? [10:47] Chipaca: aha, so there is someone ;) [10:47] mandel: about to go for coffee. want some? [10:47] hahaha [10:47] Chipaca: if I can download it ;) [10:48] mandel: i can probably get you some coffee if you're not on a small island somewhere :) [10:54] Chipaca: hahaha [10:55] Chipaca: I'm not in a small island, went o see my parents in Madrid, I needed to get out of Barcelona for some days [10:55] mandel: "small island" also includes great britain, btw [10:55] mandel: i know :) [10:56] Chipaca: oh, that is a small, don't tell the britons, they think they are big [10:57] mandel: you say "madrid", is it really "madrid", or is it "madrid" like uds-m was in "brussels"? [10:57] Chipaca: Madrid as in the place where people call it Madriz [10:58] I must say that is a very complicated question... [10:58] mandel: madrid empieza con "M" y termina con "T" [10:59] mandel: think about it [10:59] Chipaca: y ciudad? [10:59] * Chipaca goes for coffee [11:00] Chipaca: I'm lost... [11:02] mandel: ciudad con "C" [11:03] * Chipaca picks mandel off the floor [11:04] Chipaca: ok, you have been in England for far too long [11:04] Chipaca: by the way did they like the netbook? [11:04] mandel: yes, very much [11:04] * mandel is glad [11:04] mandel: it's been complaining about disk problems since day one, but doesn't seem to be broken yet [11:05] mandel: meanwhile i'm looking up what to replace the disc with, for when it dies :) [11:05] mandel: they're loving it; it works flawlessly (even if it's not fast, but it's so obviously puny they don't seem to expect it to voom much) [11:06] mandel: so, all-round win. thank you again. [11:06] Chipaca: no worries, I also have found another small nicer one that I can take there if you want to… only problem AZERTY :( [11:10] mandel: i'm a hardware junkie, and i had to give it all away before coming over, so I can't say "no". But, I don't really need it :) [11:11] Chipaca: same here… but I think you will take more advantage of it, at least if the other onew dies you have a decent one to use [11:11] lets say I'm going through a life style change ;) [11:13] mandel: as you wish, then. I can't complain either way :) [11:13] done then :) [11:16] woo, hardware! [11:39] morning ll [11:39] i was promised juggling ;) [11:40] <.< >,> [11:40] good evening! [11:40] hello ralsina [11:40] * ralsina is pretending to be in argentina today [11:41] Chipaca: I have the 701 with me here, will get it to you on June [11:46] hi duanedesign [12:01] * fagan did the same [12:01] (pretended to be on ar time today [12:01] ) [12:08] fagan: it's 8 AM dude ;-) [12:09] ralsina: yeah I didnt sleep till 7AM my time this morning [12:09] Chipaca: you are a hardware junkie? I have a couple of really special fixes for you ;-) [12:09] * ralsina has a HP Jornada 720 that can run linux from a CF card [12:09] * ralsina has a true first-gen corel netwinder [12:09] a netwinder! [12:09] o man [12:09] i wanted one of those so bad [12:10] not now, tough -- i'm going to be building something with a pandaboard soonish [12:10] a pandaboard, a pixel qi display, and a lenovo usb keyboard, a kilogram of batteries, and i should be golden [12:11] * ralsina has a libretto 50ct with Debian in it [12:13] i need to remember i have less space now than i used to [12:14] hahaha 1kg of batteries [12:14] that should keep it running for a week :) [12:15] Chipaca: how much is a Wii in London? My son has discovered them and will kill me if I don't get one [12:15] ralsina: its 100 pound now [12:15] ralsina: amazon.co.uk, deliver to millbank [12:15] but dont get one now [12:16] since the wii is going to be replaced very soon [12:17] fagan: MY KID IS 4 [12:17] oops, sorry for screaming :-) [12:17] fagan: I would get him a PS1 if I could get away with it ;-) [12:17] ralsina: hahah [12:17] * fagan has a PS2 you can use [12:18] nah, he wants to play those motion-control sport games [12:18] he beat me at fencing last night [12:19] ralsina: the PS2 was very advanced it has the eye toy if you ever saw that :) [12:19] I would get a PS2 for myself. But at least with the Wii you don't sit like a zombie all day [12:19] specially if you are4 and actually believe you have a sword :-) [12:20] ok, onto reviewing nessita's code... [12:20] yeah actually the price is after dropping so your in luck anyway [12:36] 100 pound, wii? [12:37] rye: im guessing because its 120 euro in ireland atm [12:38] what forgotten country am i living in? $240 or more in Ukraine... [12:39] rye: wow [12:44] rye: like U$S 300 in argentina, so... [12:44] mandel: ping, need reviews? [12:45] ralsina, ah, that's... weird? [12:45] ralsina: I had a python question, but I managed to fix it :) [12:46] rye: nah, it's the usual state of electronics in .ar ... it's cheaper to order SD cards from china than buying them in a store, for example. [12:46] mandel: ok, cool [12:47] ralsina, well, if there is a guaranteed delivery of stuff, then yeah, which is not the case for postal service and neither FedEx nor DHL can send anything electronic here [12:48] * fagan feels lucky [12:48] rye: I have ordered dozens of things from dealextreme.com (even little notepads), never lost anything :-) [12:49] ralsina, yeah, vds has done the same too. Were you ordering via regular post delivery? [12:49] yep. Takes a month. [12:49] and the sole idea of buying my pens in china makes me giggle like a little girl. [12:49] takes 5 days max in ireland [12:50] fagan: from china? [12:50] ralsina: never tried [12:50] fagan: go browse dealextreme.com for 10 minutes :-) [12:50] im just saying in general from the eu or us [12:51] buy a solar cockroach :-) [12:51] ha [12:57] Hey urbanape. |:) [12:57] howdy [12:57] You in the lobby, or not quite yet? [13:05] fagan: ping [13:07] ralsina: pong [13:08] fagan: are you a firefox user? Are you in the mood for some adventure? [13:08] ralsina: well im also on ff5 [13:08] ralsina: im on 11.10 already [13:08] fagan: cool, please try bindwood ;-) [13:09] and then try with firefox from https://edge.launchpad.net/~ubuntu-mozilla-daily/+archive/ppa/+packages [13:09] I expect a report later today [13:09] cool will do [13:09] * ralsina feels soooo managerial [13:10] ill be able to check if it syncs on my desktop [13:10] fagan: cool, try it both ways [13:10] sync up and down, so to speak [13:10] yep cool [13:11] ralsina: just checked and the one in 11.10 is up to date more or less with that ppa so ill just use the repo version [13:12] fagan: ok [13:14] ralsina: broken for ff5 [13:14] fagan: crapola [13:14] ralsina: says not compatable in the addons thing [13:14] fagan: broken how? [13:14] ok, may be just a metadata thing. [13:15] I will take a look maybe tomorrow [13:15] sure im on the dailies I can check when ever you have a new version merged [13:15] ok [13:16] oh and while you are at the metadata you link canonical ltd to the bindwood page on lp [13:16] instead of canonical.com [13:18] fagan: file a bug? [13:19] ralsina: will do [13:43] Bug #788594 [13:43] Launchpad bug 788594 in bindwood "Wrong link in metadata (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/788594 === kenvandine_ is now known as kenvandine [13:49] hello #ubuntuone! [13:49] merhaba alecu! [13:50] ralsina, hey there, boss. How's turkey going for you? [13:51] I am sitting under a tree, 50m away from the bosphorus, drinking lipton mango ice tea, it's 22C with a sea breeze. [13:51] so... pretty nicely ;-) [13:51] ralsina, cool! [13:52] hola nessita [13:52] ralsina, I might be late for the standup, there's a bell ringing right now. [13:52] hola ralsina [13:52] fagan: thx for the bug report. Chris Coulson has a patch to make it work with FF 6 so it should work for 5 too [13:52] fagan: I will merge it soonish [13:54] nice ralsina [13:54] ff6 is already in dev thats pretty interesting [13:54] ff20 coming soon [13:55] well... coming in 6*15 weeks or so [13:55] ralsina: sooner than it would have been in their old release system [13:55] fagan: sure. This should convince people that release numbers are not important [13:56] * ralsina has never released a 1.0 :-) [13:56] Highest I ever got was 3.x but that software was 6 years old [13:57] ralsina: well its always good to release a 1.0 when you have like a 0.30..etc [13:57] software years > dog years [13:57] ralsina: answering your review question: "assertEquals is deprecated, so we try to change them to assertEqual every time we modify a test file (we are not doing it all at once because that will be a huge branch)." [13:58] alecu: did you have any chance to review my SD branch? [13:58] nessita: makes sense, thx [13:58] :-) [13:58] thisfred: My highest version number is like 0.16.2 [13:59] I like conservatism and modesty in versioning ;) [13:59] I will someday re-release the same version as 2.0 ;-) [13:59] or 2.16.0 ... [13:59] standup in 1' [14:00] in 5"! :-) [14:00] me === m_conley_away is now known as m_conley [14:00] me [14:00] I wonder what version Silva is at now... [14:00] five or 1? [14:00] me [14:00] mandel: 5 seconds [14:00] me [14:00] ralsina, thisfred, alecu, dobey? [14:00] back! [14:00] me [14:00] * alecu is writing notes. [14:00] ralsina: ha! I lied, it was 2.x, and they're only at 2.3.1 now :) [14:01] nessita: ok [14:01] dobey: say me please? [14:01] mi [14:02] fagan: go! [14:02] DONE [14:02] * Tried out bindwood in ff5 [14:02] * Tried to figure a way of sharing a windows image or install with people so they can test out stuff [14:02] TODO [14:02] * Ask ralsina for something good to do [14:02] Blocked [14:02] * nope [14:02] nessita: go [14:02] DONE: QT windows control panel port, reviews, emails [14:02] TODO: QT windows control panel port, reviews, emails (you're not seeing double ;-)) [14:02] BLOCKED: nopes [14:02] NEXT: mandel [14:02] me [14:02] DONE: Finished reactor plugin for ubuntuone-dev-tools. Fixed CredentialsManagementTool issue in my branches and ensured that all of the work. [14:02] TODO: Ask dobey for a review, propose creds fix. Work on json-rpc [14:02] BLOCKED: no [14:02] ralsina, go go go [14:02] nessita: groundhog day? [14:02] DONE: natl holiday, couple of reviews, mgmt call, administrivia TODO: 1-on-1s with almost all of you (tmorrow!) more reviews, installer coding BLOCKED: nope. Thisfred, your turn! [14:03] DONE: reviews | fixed bug #781119 | filed and investigated bug #788206 TODO: add tests to ubuntuone-couch BLOCKED: no NEXT: dobey [14:03] Launchpad bug 781119 in ubuntuone-couch (Ubuntu) (and 1 other project) "Crashes if not logged into Ubuntu One (affects: 1) (heat: 6)" [Medium,In progress] https://launchpad.net/bugs/781119 [14:03] Launchpad bug 788206 in ubuntuone-servers "Oauth authentication against couchdb.one.ubuntu.com fails (affects: 1) (heat: 8)" [High,Confirmed] https://launchpad.net/bugs/788206 [14:03] λ DONE: working nightlies, reviews [14:03] λ TODO: magic [14:03] λ BLCK: None. [14:03] add "update bindwood with Chris Coulson's patch" to my TODO, I guess [14:04] ralsina: how do I install logmein? I cant find it :/ [14:04] you said install it a while back [14:04] dobey: can you take a look at https://code.launchpad.net/~mandel/ubuntuone-dev-tools/allow_txnamedpipes_reactor/+merge/61515 to see if that is what you meant ? [14:04] alecu: go! [14:04] DONE: national holiday. Dined with the web&mobile guys [14:04] TODO: review nessita's branch, work on qt-net [14:04] BLOCKED: no [14:04] fagan: go to logmein.com [14:04] dobey: it aint perfect, but does make it nicer… [14:04] ralsina: ahh ok [14:05] mandel: ok; can you also fix the conflict in your ubuntuone-client branch? :) [14:05] I am on 7% battery so I will take a shortish break in a random number of minutes [14:05] dobey: yes, on it next, I wanted to fix first the dev-tool sone :) [14:06] s/sone/one [14:06] but I will work late today (I am on almost ART timezone) [14:06] ok, cool, i'll look at it === karni_ is now known as karni [14:13] ralsina: ok installed [14:14] fagan: ok, now share it with alecu privately ;-) [14:14] ralsina: sec will do [14:14] mandel, ping [14:15] alecu: pong [14:15] mandel, were you able to upload the VM we discussed on Tue? [14:16] alecu: I believe that fagan cloned his [14:16] fagan: where is that vm? [14:17] mandel: on my desktop [14:17] in ireland [14:17] well I couldnt find a way to share the image correctly so im trying logmein [14:18] alecu: working? [14:18] ? [14:18] still loading [14:18] alecu: good? [14:20] brb, just dropped a cup of matecocido on the desk [14:20] lol [14:20] :) [14:20] mandel: I tried to put it up on u1 but it would have tied up my net connection for a week [14:21] mandel: and for a .torrent id have to use a tracker and it would be available to the public so that was a no go [14:21] mandel: so allowing remote access is the last step really [14:21] ok [14:22] but I can share my vm with anyone so its cool alecu controlled it there for a sec before he spilled his tea [14:25] mate != tea [14:28] back. [14:28] luckily my laptop got only slightly wet (?) [14:28] and my nexus s missed the action by about 2 cm [14:29] alecu: is it usable [14:29] fagan, yes, it seems to be usable so far. [14:29] nice [14:30] everything is installed fine since I did all my testing on it [14:30] I should probably set the paths but I got too lazy [14:35] fagan, ok. [14:35] fagan, can you show me how you run the tests? I have not worked on the windows codebase much. [14:35] alecu: sure [14:41] fagan, cool. And do you have any u1 branch around? [14:41] alecu: there probably is something in trash [14:48] fagan, my fingers keep forgetting that Alt-F4 closes my local browser, not your remote windows :P [14:49] alecu: hah [14:49] you can still get in with the old link [14:52] * fagan break [14:54] nessita: ping [14:54] mandel: pong [14:55] nessita: I'm nearly done with the credentials issue, but there is something that we did not consider. the exception class... [14:55] alecu: thats copied right click copies it to clipboard [14:55] nessita: we can make a subclass of the ubuntuone.credentials one, same as the credential management tool [14:56] mandel: let me do the context swicth on my brain [14:56] alecu: you cant run the tests on everything in that folder [14:56] fagan, thanks, but I know how to use the windows console :-) [14:56] nessita: sure :) [14:56] * nessita needs to stop singing out loud Avril Lavinge so she can hear her thoughts [14:56] alecu: I forgot myself the first time thats why I was saying :) [14:57] that seems like something that should go on the quotes page [14:57] fagan, the thing is that I'm having a hard time getting it out of the "LogMeIn" flash applet. [14:57] nessita: why'd ya have to and things so complicated I see the way your acting...something something frustrated :D [14:57] fagan: I'm lost inside!!! [14:57] alecu: you can pastebin in the vm [14:58] nessita: hahaha [14:58] mandel: the error is CredentialsError? [14:58] mandel: let me browse code to check when we raise that [14:58] nessita: sure [15:00] mandel, I'm running the u1-client tests on trunk, on fagan's VM, and I'm getting this: "ImportError: no module named credentials" [15:00] mandel: can't we define the same exception name inheriting from CredentialsError? [15:00] mandel, is this what you are discussing with nessita? [15:00] alecu: yeah, that was caused by our revert, mandel is working on landing that fixed [15:01] nessita: yes we can, is ugly but would work [15:01] mandel: so, under linux/credentials, we can have class CredentialsError(CredentialsError). And we get the price to the ugliest code ever [15:01] cool [15:01] mandel: weird; you made a plug-in that loads plug-ins; seems like too much abstraction. and i'm not exactly sure they need to be plug-ins in that way [15:02] dobey: we can merge all together in a single one, but adding new reactors this way would be very easy [15:03] dobey: but feel free to give a diff idea, merging it in a single plugin is not hard [15:04] mandel: i am actually thinking of not using plug-ins at all, at least, in this sense of the word plug-in [15:04] let me think a minute and i'll make a branch [15:04] dobey: sure [15:05] dobey: there is a funny error importing the reactors were you have to use __import__ which is ubber lame [15:05] ? [15:06] nessita, perhaps that can be better written as "from somemodule import CredentialsError as OriginalCredentialsError;class CredentialsError(OriginalCredentialsError)@ [15:06] nessita, perhaps that can be better written as "from somemodule import CredentialsError as OriginalCredentialsError;class CredentialsError(OriginalCredentialsError)" [15:06] alecu: agreed, 100% [15:09] dobey: alecu had the same issue, if you try to import and install the reactor from a function it brakes… at least the qtreactor and the txnamedpipes... [15:10] alecu: was that your issue? [15:10] mandel: what is the error exactly? [15:10] alecu: I did the import as the root exception ;) [15:10] i want to see traceback :) [15:11] dobey: let me write a small example [15:14] mandel, dobey: u1trial installs its own reactor, so if I moved the "from twisted.internet import reactor" at the top of the file I would get a "can't install a reactor twice" kind of error. [15:14] alecu: oh, I found a diff issue, the reactor thing is easy, we are working on u1trial to allow choosing the reactor in an easy way [15:14] alecu: that is the opposite of whatn mandel is saying the problem is though. he's saying you can't do the import *inside* a function [15:15] alecu: and this is before the reactor is installed (since it is the function that installs the reactor) [15:16] dobey: if you grab by branch and go to ubuntuone/devtool/plugins/reactors/txnamedpipes.py and do not use __import__ but from blah import it brakes [15:16] dobey: and I did not have a clue why… [15:19] ok, i'll make a branch with what i'm thinking and see. i think my idea is much simpler [15:22] dobey: sure, I tend to over eng [15:26] alecu: any news on the review? [15:27] nessita, I'm not working on it yet. Do you need me to? [15:27] alecu: kinda, I need to land that u1client branch since is a depends for my u1cp branch (I would like nightlies to be built by the time I request the review for u1cp) [15:28] ok, will switch to it in 5' [15:28] thanks [15:58] nessita, ping. [15:59] nessita, why enable_show_all_notifications in tools calls two different dbus methods (enable_... and disable_...)? [16:00] nessita, why not make two different methods in tools? I remember having this discussion with you earlier, and you supported having two different methods. [16:01] alecu: I know, but i was trying to be consistent with the whoe SyncDaemonTool implementation [16:01] whole* [16:01] nessita, yup, I'm seeing that. Ok, no prob. [16:02] alecu: for file_sync_enabled, and throttling, there is only one method at Sdtool level, (but 2 at DBus layer).... [16:02] * mandel eod [16:05] ralsina: ping re QT [16:10] nessita: pong [16:10] ralsina: if I have a QLabel, how can know what kind of markup I can set to it? in particular I need to set font color. [16:10] how can I* know? [16:11] ralsina: I'm using a similar markup than what I use in GTK and is not working, and my search-for-qt-doc fu is -1 [16:11] nessita: basic HTML +CSS2 should work [16:11] you can probably use the font tag [16:12] but let me find a reference [16:12] ralsina: I'm a web developer but I know nothing about ccs. A reference will be highly appreciated. [16:12] I don't even know how to type css :-P [16:12] nessita: http://doc.qt.nokia.com/latest/richtext-html-subset.html [16:12] doh, mandel is gone [16:13] blah [16:13] that should cover most of what you need ;-) [16:13] ralsina: seems like it, thanks! [16:14] well my reactor stuff mostly works [16:14] ralsina: do I have to set a property/setting to a QLabel to show that formatting? I'm setting: [16:14] File Sync is up-to-date. [16:14] ralsina: and the UI does not show the at all [16:16] hum, seems like a color string will no work? [16:16] nessita: may be a font character coverage problem? [16:16] for example if you put a X instead of a check? [16:16] ralsina: will "green" work? [16:16] nessita: it should [16:17] or you can try #00ff00 [16:17] ralsina: nopes, an X will not work [16:17] trying html color now... [16:17] nessita: are you doing this on designer or in code? [16:17] ralsina: in code [16:17] nopes [16:18] ralsina: is there any setting I need to tweak on a QLabel to show rick text? [16:18] nessita: give me 30"... [16:18] ralsina: no prob, I'll keep digging [16:19] textFormat property [16:19] ralsina: where are you looking? (I would like to learn which are your sources) [16:19] nessita: http://doc.qt.nokia.com/4.7/qlabel.html [16:19] what comes when you google qlabel :-) [16:20] ralsina: right, thing is I tend to google qt label [16:20] close enough! [16:20] nessita: nowadays I use google instead of looking at the help in assistant, actually :-) [16:21] ralsina: is there a help assistance? ;-) [16:21] nessita: start "assistant" (maybe assistant-qt4 [16:22] ralsina: is working now!!! [16:22] yay [16:22] what was the trick? [16:24] ralsina: you really wanna know? [16:24] you may fire me after you find out :-) [16:25] nessita: no I won't. I admire your persistence with Qt ;-) [16:26] ralsina: I was setting the text to the label before enriching it, because I receive the text from syncdaemon, and then on some cases I enrich it. So the setText was in between :-D [16:26] yeah, another shamefully case of PICNIC [16:26] nessita: not as bad an error as you think, since there is an "auto" setting that usually figures out you are using rich text ;-) [16:26] oh wait, yes pretty bad ;-) [16:27] hehehe [16:27] I'll get better [16:27] * nessita promises [16:27] I know you will :-D [16:29] nessita: ping [16:29] joshuahoover: pong [16:30] nessita: user on 11.04 getting "bad email token" after entering the verification code, any ideas? [16:30] joshuahoover: he's entering the wrong token... is he using the latest email he got? (he gets an email with a different code every time he tries) [16:31] joshuahoover: when a new token is issues, the former gets invalidated [16:32] nessita: ok, makes sense...i'll have him check to make sure he's using the latest email [16:32] joshuahoover: the email may take a few minutes to arrive [16:32] make him remove all his previous emails [16:33] (or something similar) [16:38] nessita, approved. [16:38] alecu: thanks! [16:42] nessita, ralsina: I think it should be better not to put html and colors in the labels, and adjust all those properties thru Qt css. [16:42] alecu: well, if you need like one word in bold, you can't do that via CSS [16:42] alecu: how can you do that for the file sync status label? [16:42] nessita, so, I'm thinking it should be better to have two qlabels, one next to the other [16:43] alecu: that label morphs into different strings with different texts and different icons and different colors [16:43] ralsina, not with the bolding, not. But if we start including font changes and color codes in translations strings, the translators will be angry. [16:43] alecu: we have at least 3 combinations [16:43] alecu: the two labels thing is even worse or translators [16:43] alecu: this is just like u1cp in GTK, and no, no markup for translators [16:43] ralsina: ^ [16:44] if it's the same as gtk and the translators coped so far... they won't notice much of a change [16:44] alecu: this is being made programatically, and the markup is prefixed according to SD state [16:44] ralsina, the left label is only used as an icon "./" or "X" or "flechita arriba y abajo" (transfer) [16:44] ralsina: but the translatable strings have no markup in this case (nor in GTK nor in QT) [16:45] nessita: even better [16:45] nessita, right, but then the colors will "atornillados" in our code. Using css makes it easier to write a suitable theme for it. [16:45] alecu: those particular colors we want them atornillados, we discussed this for the GTK control panel... not sure if you were involved [16:46] alecu: I see your point, and I +1 for most of the cases (mainly backgrounds that are theme dependent) [16:46] this particular case we want red to be red in every theme for error [16:46] and green for success [16:46] (at least that is what I talked with the design team) [16:47] nessita, no. We want the theme writer to choose the specific shade of red he wants. [16:47] alecu: honestly? Cpuld we worry about this in, say, three weeks? ;-) [16:47] given the time constraints i will not change this ATM [16:47] ah, what ralsina said [16:47] ok. [16:48] nessita: do you have constants for the colors? [16:48] I will bother you guys with this later. [16:48] alecu: cool [17:01] ralsina: off course [17:01] nessita: then this is a non-issue, we can load those off somewhere later [17:01] ok, i need a branch tested on windows [17:03] dobey: shoot [17:03] https://code.launchpad.net/~dobey/ubuntuone-dev-tools/fukishima/+merge/62506 [17:05] dobey: so to test I just need to do a run_tests with ubuntuone-client? [17:06] no [17:07] you would just run run-tests.bat in that branch, i think; and let me know the results [17:08] ooh ok [17:11] dobey: http://paste.ubuntu.com/613330 [17:11] although sso-client and u1cp will definitely have to be updated for this, but i can make those branches when this lands [17:12] ok [17:13] hrmm, the pylint warnings there annoy me a bit, but i suspect they were there already [17:14] dobey: making pylint shutup on windows means adding like 500 comments in the linux bits [17:14] ralsina: well on devtools it's not that bad, but yes it's annoying [17:15] fagan: just pushed a new revision that switches to using the standard twisted reactor on windows there; can you pull and try again? [17:15] sure do [17:16] tab fail [17:16] dobey: passed [17:16] still pylint warnings [17:16] huzzah [17:16] yes well, pylint warnings were there before [17:17] well I looked down through the branch anyway and everything seems to be in order [17:17] and i'm not sure what the best way to deal with that is [17:17] dobey: just disable them its cool [17:17] (cool by me) [17:19] well i don't want to disable them, because then if that case happens on linux, we won't get the warnings when we should [17:19] i think the right solution will be another branch to add support for ignoring files in u1lint [17:20] dobey: cool anyway I +1ed the merge since it passes the tests and my code review [17:21] ok, well i'm off to get some lunch [17:21] bbiab [17:21] dobey: yes, havinf "file profiles" or somesuch should do the trick [17:27] * fagan goes to get some food and stuff be back later [17:32] verterok: have you seen this before? bug 746640 [17:32] Launchpad bug 746640 in ubuntuone-client "ubuntuone-syncdaemon crashed with IOError in _open(): [Errno 5] Eingabe-/Ausgabefehler: '/home/phil/.local/share/ubuntuone/syncdaemon/tritcask/130090482210061.live.tritcask-v1.data' (affects: 1) (heat: 17)" [Undecided,New] https://launchpad.net/bugs/746640 [17:35] duanedesign: hi, yes [17:35] duanedesign: it might be a broken hdd (or a permissions problem, but not sure about this) [17:36] duanedesign: bug 776386 has a similar traceback [17:36] Launchpad bug 776386 in ubuntuone-client "File Sync error (IOError with tritcask metadata) (affects: 5) (heat: 45)" [High,Incomplete] https://launchpad.net/bugs/776386 [17:41] ok, I'm off to get some lunch! [17:48] verterok: aha, ok [17:48] thank you === rodrigo__ is now known as rodrigo_ [18:34] can i get a second review for https://code.launchpad.net/~dobey/ubuntuone-dev-tools/fukishima/+merge/62506 ? [18:41] on it [18:42] whoot [19:36] EOD [20:30] hrmm [20:31] i wonder if we should always unset these env vars, or only when running under appropriate reactors [20:41] dobey: I just updated dev-tools from nightlies, and I'm getting: [20:41] u1trial: error: no such option: --qt-reactor [20:41] dobey: did that changed? [20:42] nessita: you're running the tests on windows? or manually running u1trial? [20:42] dobey: manually running u1trial under linux for control panel [20:42] till a while ago I used to run: u1trial -t FileSyncStatusTestCase.test_init_loads_ui ubuntuone --qt-reactor=ui [20:42] nessita: it did change. --reactor=qt4 and --gui if you need the gui mode [20:42] dobey: what's the gui mode? ie, how can I know if I need it? :-) [20:43] nessita: if you were passing --qt-reactor=ui before, then you need it. if you were doing --qt-reactor=no-ui, then you don't [20:43] dobey: makes sense, thanks [20:49] dobey: with the latest changes in dev-tools I'm getting Segmentation fault when creating a UI instance in the tests [20:50] nessita: getting it before any tests run, or in the middle of tests? [20:50] in the middle, as soon as the UI is created by calling: [20:50] QtGui.QWidget.__init__(self, parent) [20:51] > /home/nessita/canonical/u1/controlpanel/more-file-sync-status/ubuntuone/controlpanel/gui/qt/filesyncstatus.py(84)__init__() [20:51] -> QtGui.QWidget.__init__(self, parent) [20:51] (Pdb) [20:51] Segmentation fault [20:51] I will try with the previous version of dev tools to confirm [20:51] hmm [20:53] trying with 0.1.3-0ubuntu1 [20:53] i don't think that has any qtreactor support [20:53] oh that option does not have --qt-reactor [20:53] you'd need the previous nightlies [20:53] dobey: have the version # handy? [20:54] 0.1.3+r30-10 i think [20:55] hum, it installed the r31... I better try using the source tree directly [20:56] dobey: confirmed, PYTHONPATH=/home/nessita/canonical/u1/devtools/trunk/ PATH=/home/nessita/canonical/u1/devtools/trunk/bin/:$PATH u1trial -t ControlPanelTestCase ubuntuone --qt-reactor=ui worked as expected [20:56] (revno 30) [20:57] weird [20:57] you're testing sso? [20:57] dobey: nopes, control panel [20:57] ok [20:57] let me try [20:57] you can try the latest trunk... [20:57] sure [20:58] well, you need to tweak the run-tests script, but the point remains :-) [20:58] wth [20:59] i already made a branch to tweak the run-tests script :) [20:59] was just about to push/propose it [20:59] dobey: need a review? [21:00] i get interrupted system call when starting the dbus-daemon [21:00] grr [21:00] dobey: try again [21:00] yeah, it happens 1 every 3 times approxc [21:02] is happening EVERY TIME for me :( [21:03] why oh why is that happening?! [21:04] os i can't even run these tests because of the system call issue [21:04] dobey: every time? weird [21:04] yep, i just did it like 20 times, and never once worked [21:04] dobey: try removing the _trial_temp [21:04] but i did have like 20 copies of dbus-daemon running [21:06] alecu: ping [21:06] no luck [21:07] dobey: do you what the error is about? [21:08] how am i supposed to fix this interrupted system call [21:08] no idea [21:08] it's about something interrupting a system call [21:08] that's clever [21:08] when trying to read stdout from the subprocess.Peopn() [21:09] nessita, pong [21:09] alecu: can you please check if using the latest u1devtools the tests run under linux for QT? [21:10] sure, a sec. [21:10] alecu: you'll need to tweak run-tests a bit, instead of --qt-reactor=ui please use: [21:10] --reactor=qt4 --gui [21:12] ok, weird [21:12] wonder why qt breaks stuff [21:12] i have a fix [21:13] also for the system call error [21:13] hum [21:13] and now none QT code is running for me... for the generated ui files! [21:13] File "/home/nessita/canonical/u1/controlpanel/more-file-sync-status/ubuntuone/controlpanel/gui/qt/ui/controlpanel_ui.py", line 47, in setupUi [21:13] self.file_sync_status = FileSyncStatus(Form) [21:13] TypeError: __init__() should return None, not 'instance' [21:15] ah, I know [21:15] I added a yield and inlineCallbacks... [21:16] dobey: what is the fix? [21:18] nessita: pushing branch right now [21:25] nessita: https://code.launchpad.net/~dobey/ubuntuone-dev-tools/lilbugs/+merge/62554 [21:25] alecu: does it work for you? [21:26] dobey: is less surprising than what I expected. It almost makes sense! [21:27] nessita: and then there is also https://code.launchpad.net/~dobey/ubuntuone-control-panel/new-u1trial-args/+merge/62557 :) [21:31] dobey is that variable really enough? I think if it [21:31] s not used anywhere it gets gc'ed anyway [21:32] at least potentially [21:32] thisfred: well it made it not crash here, and without it, it crashed [21:32] thisfred: and it's the only real difference between what we had before and what we have now [21:32] oh well, then in this case it works I guess :) [21:32] magic is sometimes really just magic [21:33] no science needed [21:33] dobey: I usually set things like that as a property on a long lived object [21:34] setting app = None once that is safe to do would also work I think, unless the gc is much smarter than I think [21:34] eh [21:35] the fact that we even have to instantiate QApplication is a mockery of the system [21:35] well, if we were able to mock it... [21:36] dobey: I was looking to set up an lp recipe for ubuntuone-couch, but I don't think I have the rights to do it [21:38] thisfred: you can make the recipe itself, but i probably need to put the packaging-dailies branch in place [21:38] dobey: the project page does not give me the link to add recipes [21:38] hmm, i should probably turn my a/c on, starting to get hot in here [21:39] thisfred: it's on the branch page [21:39] thisfred: anyone can make a recipe of any branch they can see [21:39] dobey, doh! gotcha [21:45] nessita: with everything updated, the first test crashes u1trial. With dobey's fixes manually applied it worked ok. [21:46] dobey, approved the u1-dev-tools branch. [21:46] * nessita aqpproves too [21:48] whoot [21:55] ubuntuone-couch now has daily builds for Natty and Oneiric in our nightly PPA [21:56] recipes ROCK [21:56] eh? [21:56] hrmm [21:57] why is it older than what's in 11.04? [21:57] ah [21:57] err you did it wrong [21:58] I need to up the version in the branch [21:58] i fixed it [21:59] although i don't know if we want to use the ubuntu packaging branch for nightlies [21:59] i think we want a packaging-only branch like the rest of our stuff [22:00] where do those live? [22:00] lp:~ubuntuone-control-tower/$project/packaging-dailies [22:01] ok, I'll start a branch === m_conley is now known as m_conley_away [22:16] dobey I have a branch that should go into lp:~ubuntuone-control-tower/ubuntuone-couch/packaging-dailies, but as that does not exist, I can't propose it for merging into that (launchpad OOPses when I try) [22:16] lp:~thisfred/ubuntuone-couch/packaging-dailies [22:23] thisfred: ok [22:25] will do that in a bit; need to run off for now [22:26] might get to it tonight; but likely in the am [22:26] have a good evening! [22:31] you too [22:31] gonna walk the dog, later peeps [23:26] * nessita -> eods [23:27] alecu: you still around? [23:27] nessita, I am. [23:29] alecu: I've burnt all my brain-books trying to debug a NAME_ALREADY_EXISTS error in syncdemon client (control panel). Yes, this error is old, but I never really understood what was the problem nor how to solve it (last time it was a bug on devtools). If you feel like it, would you help me debug why running QT tests on this branch [23:29] https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/use-sdtool/+merge/62571 [23:29] raises the error? (but not in the non-qt suite)\ [23:29] the oddest thing is that in trunk both non-qt and qt suites passes all OK [23:29] NAME_ALREADY_EXISTS sounds like a DBus error, right? [23:29] and in the changes I'm adding, the test that fails is the *only* suite I did not change! [23:30] alecu: it is, yes. [23:30] juas! [23:30] alecu: my branch moves all sd_client to SyncDaemonTool (except for the StatusChanged signal that I don't know how to replace), so moving to windows is trivial [23:31] nessita, probably a DBus test you did is not properly releasing the dbus connection, or leaving dangling objects published in the bus. [23:31] nessita, so when this other test tries to open the bus, it fails. [23:31] alecu: is a single test :-/ [23:31] in a suite, a single test [23:32] nessita, yes, but these are not unit tests, these are integration tests. The DBus-daemon instance is not cleared up between tests. [23:32] anyways, if from windows we can provide a SyncDaemonTool that returns deferreds, we're great (I came to this conclusion after seeing how mandel is replacing CredentialsManagementTool) [23:32] nessita, so if some test forgets to remove objects then this error can happen. [23:33] alecu: I understand, but the same suite runs OK in non-qt suite. And the QT suite does not add any dbus stuff (That I know of)\ [23:33] nessita, anyway, my brain is not in shape to debug a 1600 lines patch... I may try tomorrow. [23:34] alecu: is ok. I'll keep trying tomorrow [23:34] ok, I'm off [23:34] nessita, probably because of a different test order... [23:34] bye! [23:34] ok, bye! [23:55] beuno: ping [23:55] um