[07:33] I seem to have a problem with some stuff in my One-folder being synched, and some not. For example moved a couple of folder into another, those were not synched. Renamed a folder at the same location, that gets synched. [07:34] renamed one of the moved folders, that one got synched properly, but still not the other one. [08:06] hi all, just did a reinstall of my netbook. Trying to sync files from the cloud., My local Documents folder is empty, it says sync is up to date but won't let me sync [08:07] "this folder cannot be syncronized because it contains one or more folders that are already syncronised" [08:19] sorry got disconnected. Any solutions? [08:36] Buenos dias! [08:39] tried removing my PC, removing the password from the Passwords & Encryption Keys section. Still no luck. Says I'm already synced u1sdtool --status shows everything as fine [09:07] somethinginteres: hello [09:08] duanedesign: hi [09:14] somethinginteres: can you run the command : u1sdtool --list-folders [09:17] duanedesign: http://paste.ubuntu.com/658537/ [09:20] somethinginteres: ok. Your Documents folder is set to sync with Ubuntu One. However Subscribed is set to False. So it will not sync [09:21] you are wanting to make this Documents folder a cloud folder, crrect? [09:21] duanedesign: how do I subscribe it? I've tried right click "Ubuntu One --> Sync" that doesn't fix it. [09:21] ok [09:21] duanedesign:: I am wanting to d/l my files from the cloud to my PC [09:22] duanedesign: from my cloud "Documents" folder to my local [09:23] did you copy any of the contents of your old computers Home directory to this new computer? [09:23] duanedesign: nope, this is a fresh install [09:23] ok, ust making sure [09:24] just* [09:24] u1sdtool --subscribe-folder=24637050-d981-49f7-bdfd-00ec310c24f4 [09:25] can you run that command [09:25] duanedesign: run the cmd. No error or other output [09:25] somethinginteres: what version of Ubuntu? [09:25] duanedesign: 11.04 [09:25] good [09:26] duanedesign: the error I spoke of before has disappeared. The folder is empty but the sync tickbox on the ribbon is now not ghosted out [09:26] I believe a sync is taking place [09:26] we could of also acheived that by opening the Ubuntu One control Panel, selecting Cloud Folders [09:27] and ticking the box next to Documents [09:27] can you runt he command: u1sdtool --waiting | wc -l [09:27] duanedesign: Output is "1014" [09:28] ok, that is how many items are now queued to sync [09:28] duanedesign: OK may take some time. :) Any insight into why it thought the folder contained synced files? [09:29] duanedesign: Also I need to disconect very shortly, will disrupting this proccess cause a problem? [09:30] somethinginteres: it is a poorly worded error. It just means this folder is set to sync [09:30] duanedesign: Right [09:31] your are definetly not the first confused by it :) I need to make sure their is a bug filed on that [09:31] duanedesign: : safe to disconnect at this point? [09:31] duanedesign: ah, I see :) [09:32] somethinginteres: yes you should be fine now. If you have any other problems let me know [09:32] duanedesign: Will do. Thanks very much. [09:32] o/ [09:33] rogerk_: just saw you needed help. I am going to step out for 45 minutes. I can help when I get back. [09:38] duanedesign: Oh, nice, thanks! I'm at work so might not be available at that time, but I'll try to sync up at some time after that. :-) [10:18] hi everyone [10:18] hey DiegoSarmentero [10:32] rogerk_: no problem just ping me [10:32] hello fagan DiegoSarmentero [10:32] hi duanedesign fagan [10:33] how are you this morning? (morning for me :P) [10:33] hey duanedesign [10:34] DiegoSarmentero: doing well. Trying to decide on a new laptop. too many choices :/ [10:35] duanedesign, alienware! :P [10:35] duanedesign: id suggest lenovo [10:36] fagan: i think i narrowed it down to lenovo or system76 [10:36] duanedesign: I cant buy system76 so never was an option [10:36] I have a acer desktop, msi wind netbook and a macbook [10:36] fagan: the first lenovo model I looked at did not have usb3.0, though I do not know if i would even miss it [10:37] duanedesign: yeah you dont really need usb 3 yet [10:37] nearly everything is still usb 2 [10:37] * duanedesign nods [10:37] the model w/ out 3.0 usb is about $300-400 cheaper [10:38] then its an easy choice [10:38] and the more expensive model has a graphics card that uses Optimus [10:38] though I hear lenovo has a setting in the bios to deal with this. [10:40] duanedesign: interesting do you play games or do anything graphically difficult? [10:40] * fagan has a nvidia card that cost an arm and a leg but I play lots of games so was an investment [10:40] occasionaly...but not often. [10:41] duanedesign: well then the optimus thing isnt really a huge priority if you dont :) [10:41] Just enough that i am interested in a graphics card. But not often enough to really make it essential [10:41] :P [10:41] fagan: I think you are right [10:41] duanedesign: well if it was me id be cheaping out :) [10:43] fagan: considering technology becomes old so fast it is hard to justify spending too much money [10:43] knowing you wil have to but a new computer in 3 years [10:44] fagan: do you run Ubuntu on your macbook? [10:44] duanedesign: well im a funny person when it comes to moores law, I say get an ok computer and just replace it often it costs about the same in the long run [10:44] duanedesign: yeah I do but im not using it at the moment [10:44] im on Mac OS [10:44] But its pretty easy to dual boot [10:45] I have to say the keyboard is bloody nice to use [10:45] :) [10:45] the macbooks are down right sexy [10:46] duanedesign: have you thought about getting a desktop? [10:46] and then getting a netbook [10:46] so you can have something to travel with [10:46] that worked out so well for me [10:46] Although my netbook was a little trash [10:47] fagan: yeah, not a terrible idea. Would be nice to have something more portable then a 6 pound laptop [10:48] duanedesign: well desktops seem to keep their value longer because they have faster processors and better graphics cards [10:48] and they are cheaper [10:48] * duanedesign nods [10:49] so I paid 500 euro for my desktop it has 4 processors and a great graphics card [10:49] and a netbook that cost 250 and it makes up for the fact I cant move my desktop [10:50] and 750 is cheaper than most netbooks with similar specs [10:51] fagan: ahhh, more choices :) [10:52] fagan: i will look at that...sounds reasonable [10:52] duanedesign: well I love complicating things for other people :D [11:00] * fagan errands [11:27] * fagan back [11:35] hello all! [11:35] ralsina: ping [11:36] nessita, hello [11:38] hi DiegoSarmentero, how is it going? yes, I'm doing your reviews now :-) [11:40] nessita, jejejee i'm fighting with page 18 of the installer :@..... [11:40] DiegoSarmentero: question, why did you add the #pylint: disable=C0103 and the enable comment for set_title and set_subtitle? [11:41] nessita, ahhhhhhhhh my bad...... that methods were first setTitle and setSubTitle... but then I realize that setTitle and setSubTitle only were neeeded on SSOWizardPage [11:41] changing... [11:42] DiegoSarmentero: thanks! [11:46] nessita, done [11:46] DiegoSarmentero: another question, why do you call [11:46] self.controller._title = JOIN_HEADER_LABEL % {'app_name': self.app_name} [11:46] self.controller._subtitle = self.help [11:46] at the beginning of the test_titles if you later call set_title and set_subtitle in the same test? [11:47] same question for ErrorControllerTestCase.test_set_ui [11:48] let me check [11:48] (and same question for SuccessControllerTestCase.test_set_ui) [11:50] nessita, :S yes... self.controller_title, etc is no longer needed... that was from before the second change to user a reimplementation of setTitle... removing..... [11:50] DiegoSarmentero: thanks! [11:50] DiegoSarmentero: the rest looks great [11:54] nessita, i removed some of them, in Error and Success they are needed because the method setupUI actually use _title and _subtitle so i have to set that values in the controller [11:54] so i need to set that values, and call then set_title and set_subtitle to record the execution [11:56] nessita, done [11:56] DiegoSarmentero: looking [12:00] DiegoSarmentero: hum, is there any chance you push any extra changes? I think I'm seeing something I did not see before, but maybe I was too sleepy 5 minutes ago :-P. What I see that I don't like is the duplication of the def set_title(self, title='') and def set_subtitle(self, subtitle='') [12:00] DiegoSarmentero: what we talked with ralsina was not to duplicate that code anywhere, but defining it in a single class [12:01] DiegoSarmentero: I personally would like to have that as another widget, but he mentioned it may be overkill doing it that way. Anyway, the requirement is that the logic is not duplicated anywhere, so if you find yourself writing the same piece of code twice, we need to refactor [12:03] nessita, ok... i didn't in that way... because i wasn't sure if EVERY page is will have a title and subtitle... but it is a easy fix [12:03] is going to* [12:03] DiegoSarmentero: well, you can make some pages inheriting this "new title'd class" and some other no, right? [12:05] nessita, yes, we could have two different SSOWizardPage... let me make the change and test if it doesn't broke anything and i will be pushing it in a few minutes [12:05] nessita, meanwhile... is syncdaemon working? i keep receiving an error when i try to sync the folders in the installer [12:06] DiegoSarmentero: no two different SSOWizardPage... but 2 different base class and having mulitple inheritance? [12:06] DiegoSarmentero: what error? [12:06] good morning [12:07] ralsina, good morning [12:07] hi ralsina [12:07] nessita, http://paste.ubuntu.com/658606/ [12:07] ralsina: any news regarding the review I sent to you yesteday? [12:08] DiegoSarmentero: is syncdaemon running in that machine? [12:08] nessita: finally I followed your advice and took the night off, so starting it now [12:10] nessita, i'm running the sso-client without any problem, but when syncdaemon keeps giving me this error: http://paste.ubuntu.com/658608/ [12:11] DiegoSarmentero: the sso you're pointing to syncdaemon is ols? [12:11] old*? [12:11] nessita, i updated it yesterday [12:11] both [12:11] DiegoSarmentero: try this: [12:12] set PYTHONPATH=..\ubuntu-sso-client [12:12] python [12:12] from ubuntu_sso.main.windows import UbuntuSSOClient [12:12] and let me know how it goes [12:12] ok [12:15] nessita, ImportError: No module named ubuntu_sso.main.windows (I'm inside ubuntuone-client) [12:17] nessita, wait!.... [12:17] nessita, nevermind....... [12:18] found the problem [12:20] DiegoSarmentero: what was it? [12:20] nessita, it was pointing to another branch of sso-client :( [12:35] hello. i have a folder in ubuntu one folder which is showing the refreshing icon, but u1sdtool thinks it is up to date. website doesn't show the file - how can i fix? [12:44] duanedesign: ^ [12:44] nessita: +1 on the epic branch! [12:45] ralsina: great! thanks! [12:45] hello oimon [12:45] i also have other folders that i've copied into u1 folder that do sync with the website but still show refreshing icon on the folder, even when it's synced [12:45] I mean +1 as soon as I finish reading the test changes, but the main stuff looks great [12:46] oimon: what do you get when you run the Terminal command: u1sdtool --waiting | wc -l [12:46] ralsina: is ok, I still need to wait for alecu [12:46] duanedesign: i'm on lucid [12:46] waiting-content and metadata are both 0 lines [12:46] oimon: ok [12:46] nessita: could you take a look at https://code.launchpad.net/~ralsina/ubuntu-sso-client/validate-harder/+merge/70320 ? It's very boring ;-) [12:47] ralsina: sure [12:48] duanedesign: my home dir is mounted over NFS, not sure if that is relevant at all [12:48] ralsina: in _validate, shouldn't you call is_correct_password? [12:48] nessita: in which one? [12:48] the first one [12:48] oimon: could you check if this file has anything in it: ~/.cache/ubuntuone/log/syncdaemon-exceptions.log [12:49] nessita: I am not calling intentionally, because suppose someday we change the password requirements, that would mean you can't login with the old password [12:49] duanedesign: nope, the exceptions log files are all 0 butes [12:50] oh, wait, that one is the forgtten password. Yes, on that one I should [12:50] ralsina: is that validation for login or forgotten password? [12:50] nessita, ralsina, that is already the case. My existing SSO password doesn't conform to the nre guidelines [12:50] ah, that [12:50] *new [12:50] we did that in Android and had to revert it [12:50] oimon: ok. can you check what you get with the command: u1sdtool -s [12:50] beuno: right, for login it makes no sense, but for password change it does [12:50] nessita: tke it back, the first one is "Sign In", on that one we should not call it. [12:50] oimon: and pastebin the contents at http://pastebin.ubuntu.com/ [12:51] http://pastebin.ubuntu.com/658624/ [12:51] nessita and that is there, in line 124 [12:52] ralsina: ok. Another thing, suffixing number in variable/method/test cases is not recommended since when reading, is confusing what does it mean. Can you please change CurrentUserControllerTest2 to something more explicit? probably CurrentUserControllerValidationTestCase [12:54] oimon: hmm. ok, could you please post the syncdaemon log ~/.cache/ubuntuone/log/syncdaemon.log [12:54] duanedesign: here's a summary of the situation. copied 1 folder (.liferea_1.6) from another pc with the same home dir mounted via NFS. this doesn't appear on website. then on my main PC , copied "building" and ".liferea_1.2" which do appear on website but still show refresh icon as per http://i.imgur.com/zkJ0c.png [12:55] nessita: Now it's about validation but I want to do any new tests there instead of with mocker. I'll figure out a generic name. [12:55] ralsina: ideally you should have a dedicated testcase for each "grouping" of functionality being tested. Is common to group into a validation tests vase [12:55] test case* [12:55] ralsina: so I would advice a validation test case and then add any other needed test case [12:56] with a accurate name [12:56] Cool, will do that then [12:56] great [12:57] oimon: yeah the nautilus integration in Ubuntu One is less then ideal in Lucid. So the emblems are sometimes not very accurate [12:58] yeah i've also had nautilus crashes with u1 too [12:59] duanedesign: so back to the original issue, could it be because i copied the folder while on another machine? u1 should still poll the directory though? [13:00] .2 [13:00] ralsina, alecu, dobey, DiegoSarmentero, mandel, fagan: standup! [13:00] me [13:00] me [13:00] me [13:00] me [13:01] me [13:01] ralsina: I know no mandel today, but we do have an alecu today, right? [13:01] DiegoSarmentero: have you tried to register "gato" as your irc nickname? [13:02] nessita, AFAIK yes [13:02] DiegoSarmentero: you current one is too long to type it in the daily stand up! [13:02] nessita, no... i will try! :P [13:02] DONE: finished and proposed epic branch [13:02] TODO: hunt reviews for epic branch, do reviews for Diego and ralsina, propose old branch with controlpanel ui improvements [13:02] BLOCKED: nopes [13:02] NEXT: DiegoSarmentero [13:02] DONE: [13:02] Some refactoring in the use of Title and Subtitle [13:02] TODO: [13:02] Finish Pages: 18 and 19 from Wizard. Fix UI bugs from installer [13:02] BLOCKED: [13:02] no [13:02] dobey, go [13:03] λ DONE: yet more shim work [13:03] λ TODO: finish initial shim [13:03] λ BLCK: None. [13:03] ralsina: go [13:04] DONE: half day off, windows call, reviews, proposed a couple of branches. TODO: IRL testing and bug reporting, BLOCKED: no [13:04] fagan go [13:04] DONE [13:04] * more report for college [13:04] TODO [13:04] * Finish the report [13:04] * If anyone needs anything do that [13:04] Blocked [13:04] * nope [13:04] comments? [13:04] is cooold! [13:04] Its hot in ireland [13:05] or maybe its the flu im coming down with === DiegoSarmentero is now known as gato [13:05] gato: new nick? [13:06] fagan, yep... shorter... and it is an usual nickname for me :P [13:06] gato: well everything is just letter tab so it doesnt really matter about length [13:07] gato: congrats on the nickname! (?) [13:07] nessita, jejeje [13:07] lol [13:07] gato gordo? :) [13:07] Ok thats what she said (in honor of mandel) [13:07] about my last statement [13:07] jejeje [13:08] hello #ubuntuone! [13:08] dobey, trying not to be fat!! :P [13:08] hello alecu [13:08] sorry I'm late; me (no notes) [13:08] * fagan break [13:08] gato: so, can you please refresh my memory regarding what we talked in the plane about setting properties to widgets? [13:08] alecu: go!!! [13:08] hola gato! got new nick? [13:08] hehe [13:08] alecu, yep :P [13:09] nessita, yes, you can do it in the designer in the property editor.... or just through code: setProperty("stringId", value) [13:10] brb [13:11] DONE: worked on bug #820350, perhaps found a simple solution [13:11] TODO: discuss a better solution for the above bug, review new issues with bug #803672 [13:11] BLOCKED: no [13:11] Launchpad bug 820350 in ubuntuone-client "WIndows: either set_dir_readwrite and/or set_dir_readonly are not doing what they should (affects: 1) (heat: 6)" [High,Triaged] https://launchpad.net/bugs/820350 [13:11] Launchpad bug 803672 in ubuntuone-client (Ubuntu) (and 2 other projects) "Do not start two instances of this process (affects: 2) (heat: 22)" [Undecided,Invalid] https://launchpad.net/bugs/803672 [13:11] nessita, still needing the epic review? [13:12] alecu: yes please, a lot [13:12] * alecu rolls up his sleeves [13:12] alecu: thanks!!! === m_conley_away is now known as m_conley === DiegoSarmentero is now known as gatox [13:20] * gatox sad... gato was already in use [13:22] gatox: chem, help. How can I set in a stylesheet something equivalent to self.setLayoutDirection(QtCore.Qt.RightToLeft)? [13:23] * gatox looking... [13:24] nessita, why you need to do that in qss? [13:24] gatox: sounds sharper [13:24] fagan, :P [13:25] nessita, i think you can't do that with stylesheet :S [13:25] gatox: no problem then [13:34] duanedesign: I was dragged into a rather lengthy meeting, but now I'm available. Are you? [13:35] ralsina, do yu have a minute? [13:36] gatox: sure [13:36] ralsina, is there any reason why "SuccessController" in gui.py (sso-client) doesn't extend from "BackendController" as the others? [13:37] just to know.... because it will be better for something that i'm testing to make all of them extend from BackendController [13:37] gatox: probably just a mistake [13:38] ralsina, great...... i will modify it! :D [13:43] rogerk_: yes [13:43] rogerk_: what version of Ubuntu are you on? [14:02] duanedesign: 11.04 [14:03] ralsina, alecu, Chipaca, gatox: shall we mumble? [14:04] nessita, ack [14:05] alecu: che, I can't make syncdaemon start, I'm getting a CannotListenError (http://pastebin.ubuntu.com/658660/). I don't have any other instance running, and I never installed a installer in this vm [14:05] alecu: any ideas? [14:05] weirddddd [14:05] nessita, trunk_ [14:05] nessita, trunk? [14:06] alecu: my branch, let me confirm trunk [14:06] nessita, this looks suspicious: self.dbus_iface.shutdown(with_restart) [14:06] nessita, that's from the paste. [14:07] nessita, very likely that shutdown is not stopping the ipc listener before restarting [14:07] nessita, do you know why is syncdaemon trying to restart? [14:08] alecu: the dbus_iface is code located in ubuntuone/platform/windows/__init__.py [14:08] nessita: could you take a look at this https://pastebin.canonical.com/50744/ [14:08] nessita, right. But did you tell syncdaemon to restart? [14:09] nessita: do you know what package is missing [14:09] nessita, look at this, also from your paste: self.main.restart() [14:09] alecu: not at all, I'm looking at the logs to see what happened [14:10] alecu: there was a SYS_UNKNOWN_ERROR pushed to the state machine, which triggers a restart [14:10] rogerk_: can you run the command: u1sdtool --waiting | wc -l [14:11] becasue there was a transition Transition SET_CAPABILITIES --[SYS_SET_CAPABILITIES_OK]--> AUTHENTICATE, which makes no sense! [14:11] alecu: I'll debug further, sorry for pinging you so early [14:11] duanedesign: looking [14:11] nessita, I don't understand that, but do you know how we can "simulate" this to reproduce this error? [14:12] alecu: the restart is pretty easy to simulate in a test [14:12] alecu: this particular transition makes no sense [14:12] nessita, anyway, it's very likely that restarts are broken at the ipc level. [14:12] duanedesign: what system is that? [14:12] alecu: Oh. Syncdaemon uses restart when in despair [14:14] alecu: in trunk, syncdaemon will not move from READY, even if the SSO service returns the credentials :-/ [14:22] nessita: they are running 10.10, but i think they might be using a PPA version of U1 [14:27] duanedesign: looks like they are missing the gir-1.-unity- package [14:27] duanedesign: result i 0. [14:30] ralsina: so, are we mumbling today? [14:32] rogerk_: have you tried restarting the syncdaemon to see if that helps. the command would be" u1sdtool -q; u1sdtool -c [14:34] ralsina: ping? [14:34] alecu: in trunk syncdaemon is also transitioning from [SYS_SET_CAPABILITIES_OK]--> AUTHENTICATE :-/ [14:34] alecu: is it working for you/ [14:34] ? [14:35] sorry, was afk [14:35] I don't think we have much news since yesterday, do we/ [14:35] nessita, let me check [14:36] ralsina: we can skip the mumble, but I want to be sure you and Chipaca know that we're getting behind schedule [14:36] ralsina: and that ATM, syncdaemon is not working for me, still debugging why [14:37] alecu: actually, the problem is not that transition but the SYS_UNKNOWN_ERROR taht is pushed after that [14:37] nessita: I know we are behind, I will make sure Chipaca knows too [14:37] BTW: on linux: ['/home/ralsina/.config/ubuntuone', '/etc/xdg/ubuntuone'], on windows: ['/.config\\ubuntuone'] yikes [14:38] ralsina: one of the things that worry me most, today, is the thing I mentioned about events out of order/missing/etc in the email from eysterday [14:39] Yes, that is really bad, and new :-( [14:43] ralsina: well, is new becasue before the fsm tests were not run because all the issues with threads that alecu and mandel fixed last week [14:43] nessita, found the issue in trunk: [14:43] nessita, the config file location problems for syncdaemon are because xdg is (as usual) broken for windows. I am going to reimplement some pieces of it like you did for xdg_cache_home [14:43] 2011-08-04 11:38:39,907 - ubuntuone.SyncDaemon.ActionQueue - INFO - The request [14:43] 'oauth_authenticate' failed with the error: oauth_authenticate() takes exactly 3 [14:43] arguments (4 given) and was handled with the event: SYS_UNKNOWN_ERROR [14:43] alecu: yeah, I just saw that too :-/ [14:43] I think there was an update i storage protocol [14:43] that I don't have [14:44] indeed: Now on revision 136. [14:44] nessita, yup, it looks like that. [14:45] ralsina: there was a new update to storage protocol, and is mandatory for u1client to work, FYI (for the installer) [14:45] alecu: thanks a lot :-) [14:45] this sucks [14:45] alecu: how's the review going? [14:45] ralsina: same question ^ [14:45] nessita, almost there [14:45] So we all need to update ubuntuone-storage-protocol? [14:45] ralsina: and re build and re install [14:45] nessita: I gave you the +1 a while ago, forgot to mention it [14:45] gatox: ^ [14:45] ralsina: thanks! [14:46] nessita, why re-build? [14:46] nessita, update storage? [14:46] or something else? [14:46] ahhhh ok..... [14:47] alecu: python setup.py build install? [14:47] updating...... [14:47] nessita, I think that just "python setup.py install" in ubuntuone-storage-protocol should be enough [14:47] alecu: maybe, I never knew exactly what's build for :-D [14:47] nessita, I think that "install" depends on "build", and does it automatically if needed. [14:50] alecu: is working now!!! :-) [14:54] nessita, alecu: basically every line in xdg.BaseDirectory is broken. Should I just rewrite it? [14:55] nessita, really? got a pointer to the code? [14:55] I mean, ralsina. [14:55] I can't copy it into our tree because of copyright assignments, right? [14:55] ralsina, I think we can't, right. [14:55] ralsina: we can't [14:55] ralsina: why every line is broken? [14:56] alecu: let me pastebin it [14:56] ralsina: syncdaemon only uses [14:56] 30 load_config_paths, [14:56] 31 save_config_path, [14:56] 32 xdg_data_home, [14:56] 33 xdg_cache_home, [14:56] ralsina: so we need only that [14:56] alecu, nessita: http://pastebin.ubuntu.com/658675/ [14:56] because everything assumes it's linux [14:56] and it's not that big beyond what syncdaemon uses. I would reimplement just those bits [14:57] it uses the wrong separators, home by default is "/" (literally) [14:57] ralsina: it does not look sooo-linux specific, it uses os.path.join [14:57] this is home: os.environ.get('HOME', '/') [14:58] nessita, it [14:58] it's pretty much linux specific: [14:58] split(':') [14:58] nessita: all the paths by default have "/" in them [14:58] yes, on windows there is no $HOME [14:58] it splits paths using ":" instead of os.PATHSEP [14:59] and also it's getting everything from the XDG_* env vars, that will not be set on a windows. [14:59] alecu, ralsina: all that is already fixed in our implementation, we only need load_config_paths to work [14:59] alecu: we have all that fixed already [15:00] in windows I have to do like get_special_folders. So I don't think any of the parts we use is actually correct for windows. It's not big though [15:00] nessita, ok. where is that? already in our tree? [15:00] and perhaps save_config_path [15:00] nessita, or in mandels +junk? [15:00] alecu: yes, platform/xdg_base_directory [15:00] load_config_paths uses everything else [15:01] nessita: so yes, I will add those two there [15:01] cool [15:02] ralsina: I forgot about you review after I mentioned the name change for the testcase, restarting it now. Can you please add an ending dot to """Tests for CurrentUserController, but without Mocker""" [15:09] sure [15:10] Tracking Oneiric has been the most painful dogfood-eating release cycle I've seen yet. :( [15:11] nessita: pushed the "." [15:11] ralsina: all the '.'? I see that several tests cases have the same docstring [15:12] je, no, just the one you mentioned, will fix that in 1' [15:13] ralsina: I added a couple of needs fixing and a question [15:14] brb [15:14] nessita: cool, thx [15:17] nessita, why did you merge test_deleting_open_files_is_no_cause_for_despair and test_stomp_deleted_file_is_no_cause_for_despair_either ? [15:18] nessita, nevermind, it's the same test :P [15:20] alecu: is the same test! :-) [15:20] alecu: I confirmed it with facu [15:21] great [15:21] gah. found a nice bag, but it's apparently only available in australia [15:21] alecu: At first I tried to make it work on windows, but I couldn't, so I just made it more readable and skipped it [15:21] dobey: bug or bag? :-D [15:21] nessita: camera/laptop bag :) [15:21] ah! :-P [15:25] nessita: about the needsfixing, unicode and QStrings compare correctly [15:26] ralsina: then why are we coercing to unicode everywhere? [15:26] ralsina: either we do it everywhere, or nowhere, otherwise confuses the reader [15:27] QStrings are not unicodes in the sense that you can't use them interchangeably, but you can say if QString == unicode and it does the right thing [15:27] But sure, I will add it, just mentioning it [15:36] ok, must get lunch. bbiba [15:36] bbiab [15:39] ralsina, nessita i've finished with title and subtitle... i've applied several improves in the code i've just submitted... i'm waiting to launchpad to rescan my branches to give you the url... i'll keep working on pages 18 and 19 now [15:39] gatox: ack! [15:40] gatox: cool! [15:42] gatox: I just noticed we have a SSOWizardEnhancedEditPage and a SSOWizardPage [15:42] I think we should merge them [15:42] ralsina, maybe... it seems that SSOWizardEnhancedEditPage is only being used by SetupAccounnt [15:43] ralsina, let me check.... and i will apply that change in this branch too if you want [15:43] gatox: it should be used by every page that has EnhancedLineEdits, and we should use those for all line edits that need validation [15:43] I think that's for another branch not this one :-( [15:44] ralsina, ok! better [15:44] nessita, a couple of very small need fixings in https://code.launchpad.net/~nataliabidart/ubuntuone-client/fix-tests/+merge/70373 [15:44] * alecu breathes again [15:47] alecu: looking! [15:49] nessita, ralsina both branches are ready for review: https://code.launchpad.net/~diegosarmentero/ubuntuone-windows-installer/titles-on-pages/+merge/70169 --- https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/titles-on-pages/+merge/70157 [15:49] nessita: responded to your needfixings and fixed them, too. [15:50] gatox: cool! [15:50] nessita, at the end of installer changes test_gui.py it appears to have some merge problems... but that changes are not in the file if you download the branch... it might be some merge problem with trunk [15:51] i'm looking if i can fix it locally and upload [15:55] nessita, ralsina merge problems fixed... the branches are complete [15:56] gatox: the sso branch is much nicer now [15:57] ralsina, :D [15:57] gatox: I think we should make BaseController take the optional messagebox, title, subtitle arguments and handle them. They are always used the same, anyway [15:59] ralsina, ok....... i will remove it from the specialized controller and pass this arguments to the parent in the super statement [16:01] alecu: I used it the or returning a lambda for consistency sake, is how facundobatista does it in other tests. I agree is aweful. [16:02] alecu: not sure if preserve consistency or improve some of those only [16:02] gatox: and of course if the __init__ is now just a super, remove it ;-) [16:03] gatox: Then, since the controller always will have a title/subtitle, maybe the SSOWizardPage can just use it and the controller doesn't have to call set_title and whatever. [16:05] do you mean moving the initialization of title and subtitle from controller to SSOWizardPage??...... because BackendController is a controller... so it will work as usual in that way [16:05] ralsina, ^ [16:07] ralsina, but some controllers are using the set_title, because they have another title based on validation or something like that === beuno is now known as beuno-lunch [16:08] gato: ok, then how about moving it into the BaseController's __init__? If a specific controller wants to set them different, it does it after super() [16:09] gatox: ^ [16:10] ralsina, ok... it almost ready... testing... [16:10] it's* [16:14] nessita: use a lambda that return a tuple instead. Slightly cleaner... [16:15] ralsina: we don't need two results, but an append and a defer returned. I know how to fix it, I'm not sure I want to break consistency with other tests, but I guess we have to [16:16] nessita: oooook, missed that. If the other tests are the same I say leave it as is. [16:16] ralsina, I can't move set_title and set_subtitle from setupUI in controllers to __init__ in BackendControllers, because the header is created after the controller [16:17] gatox it's ok then [16:17] ralsina, i could do it maybe in initializePage [16:17] ralsina, what do you think? [16:17] ralsina, ah..... no, same problem..... [16:18] because the controllers are overwriting that method [16:18] nah, we have 3 "sort-of-line-init" methods, leave it as it is [16:18] the controllers are overwriting initializePage????? [16:20] SSOWizardPage is overwriting initializePage..... and calling pageInitialize in the controllers... [16:20] ralsina, we made this last week [16:20] ralsina, do you remember? to handle default buttons [16:21] oh yeah, that [16:21] I just had a code cache overflow :-) [16:21] ralsina, i'll try to put set_title there.... just one controller is overwriting this method... do you agree? [16:23] no, let's stop fiddling with this branch, I take that objection back ;-) [16:23] ok [16:24] i will submit the first modification.... that it is really small..... and its working properly...... [16:25] alecu: re: lambdas with ors, would you prefer maintain consistency or change some of them? [16:26] nessita, +1 to fix the new ones, and fix the others later. [16:26] ack [16:28] ralsina, done... both branches are updated. [16:30] gatox: looking... [16:31] gatox: +1 with a little star on top [16:32] ralsina, :D [16:35] alecu: everything fixed and pushed [16:36] (and re-scanned by LP) [16:37] nessita, cool === beuno-lunch is now known as beuno [17:11] nessita, alecu, gatox: I am going to change our windows-xdg thing so that it considers these as xdg_config_dirs: ['C:\\Users\\ROBERTO\\AppData\\Local', 'C:\\ProgramData'] [17:11] ralsina: you sure about the last one? [17:11] ralsina, cool. How are you going to build those paths? [17:12] ralsina, are you using some win32 apis? [17:12] ralsina: didn't gatox mentioned something about c:\\allusers\\ for config? [17:12] or getting them from the registry, or something? [17:13] That seems to me like the common practice, according to the ton of apps I have installed [17:16] nessita: it seems all apps put stuff there (including ubuntuone-protocol [17:16] alecu: I got them using win32com, just like we were getting them before [17:17] ralsina: ok then :-) [17:17] great [17:17] c:\users\allusers\appdata seems to be a ""'link" to c:\programdata [17:18] ralsina, I saw that. Seven uses much saner directory names than previous versions, but it puts links around so old paths can still be used. [17:19] alecu: I am getting "CSIDL_COMMON_APPDATA" which should point to logical places on all versions [17:20] anyone has a XP handy to run 3 lines of python? [17:21] ralsina, perhaps fagan does? [17:24] alecu: yes, but we don't have a fagan until tomorrow :-) [17:24] Anyway, it's supposed to work n Windows millenium and higher ;-) [17:41] nessita, alecu, gatox: small branch for review: https://code.launchpad.net/~ralsina/ubuntuone-client/find_config_files/+merge/70472 [17:49] ralsina, on it [17:50] gatox, thanks! [17:55] gatox: can please refresh my memory: how can I load the qss in the designer? [17:57] nessita, not load.... you press right click in the dialog/widget (whatever) and choose "change styleSheet" option [17:57] nessita, and you can play around with that [17:58] gatox: but can I load the stylesheet from the qss in disk? [17:58] nessita, honestly... i don't know... i've never done that..... but you can copy/paste all the content :P [17:59] nessita, i assume this is just for testing [17:59] ack, thanks [17:59] nessita, it's kind of ugly (from my point of view) leave qss styles in the ui [17:59] gatox: of course, I wanted the qss to load visually, not to modify any xml [18:03] ralsina, +1 (review) [18:04] gatox: thanks! [18:04] ralsina: I can run some Python on WinXP SP3 if you need. :) [18:05] nhaines: thanks for the offer but it would probably involve a couple of hours of setup before you can run them ;-) [18:06] ralsina: oh, best not then... they expect me to get *some* reports out today at work. ;) [18:09] gatox: have handy the doc for the button "build" guide? [18:10] nessita, ?? from qt assistant for style sheet or you are talking about something else? [18:10] gatox: I'm talking about the spec for styling our buttons :-) (sorry for being so cryptic) [18:10] nessita, jeje..... let me check [18:10] ralsina: whats up, I can give you a little time if its important I dont mind [18:10] * gatox searching... [18:11] fagan: cool! [18:11] fagan: let me do a pastebin of the python code I need running on XP [18:11] ralsina: Oh crap ill have to do it tomorrow I broke my ubuntu install that that VM is on and forgot to fix it [18:12] ralsina: paste it and ill say if its working for when you guys get up tomorrow [18:12] * fagan fixed one ubuntu but not the other [18:14] fagan: http://pastebin.ubuntu.com/658780/ [18:14] ralsina: cool whats the expected output? [18:14] ok people... i'll take a break (i'm here since 7 :P)... i'll be back around 17/18 to try to close some installer stuff if anyone needs anything [18:15] gatox: enjoy :) [18:15] fagan: surprise me ;-) (should be a list of paths) [18:15] ralsina: cool [18:29] nessita, alecu: if I could get a second review of https://code.launchpad.net/~ralsina/ubuntuone-client/find_config_files/+merge/70472 -- I checked it IRL, syncdaemon can now start without passing config files in the command line with that patch [18:31] ralsina: the branch needs tests... [18:33] trivialish: https://code.launchpad.net/~dobey/ubuntuone-installer/launch-cp/+merge/70479 [18:35] nessita: I will do them. I note that I am replacing something that had no tests and was brokn with something that has no tests but works, though ;-) [18:35] fagan: got the results yet? [18:36] ralsina: the thing that had no tests is a 3rd party lib. And I thought we agreed that every single piece of code *we* add, will have tests [18:36] nessita: yes, we agreed, and I slipped. [18:37] I will add tests for the kazillion other things in that module that don't have it while I am at it [18:38] ralsina: *thanks* [18:43] nessita, your branch is making tarmac unhappy. see: https://code.launchpad.net/~nataliabidart/ubuntuone-client/fix-tests/+merge/70373 [18:44] alecu: looking [18:44] nessita, it looks to me that tarmac is not set to a utf8 locale [18:44] dobey: help? ^ [18:45] nessita, try "unset LANG; unset LANGUAGE; u1trial tests/syncdaemon/test_vm.py" and you get the same error that tarmac throws [18:46] nessita, the thing seems like a bug in the branch [18:46] are you using translated strings in tests? [18:46] dobey: nopes [18:46] nessita, the branch is sending unicode to the os functions on linux [18:46] dobey: but I'm using non ascii chars [18:46] alecu: right, and that is intended [18:46] nessita, why? [18:46] why are you doing that? [18:46] alecu: did you read the comment in the MP? [18:46] nessita, it should be sending bytes on linux [18:47] why? [18:47] nessita, because if not, then we get this kind of errors [18:47] 78+ # path is a byte sequence encoded with utf8. If we pass this to os.walk, [18:47] 79+ # in windows, we'll get results encoded with mbcs [18:47] alecu: we need unicode becasue of that ^ [18:47] nessita, on windows, it makes sense, but on linux we can't [18:48] alecu: if we pass bytes to os.walk we'll get other encoded bytes as result [18:48] alecu: that makes no sense... how come we can't pass unicode to os.walk? [18:49] nessita, the native way to use os.* on linux is by passing bytes [18:49] alecu: what if we have a latin1 linux FS? syncdaemon will transform that to utf, and if we pass that to os.walk... [18:49] nessita, if we pass unicode, python will *try* to convert it to bytes before calling the operating system [18:49] alecu: either should work [18:49] nessita: latin1 probably won't work [18:49] dobey, no: it depends on the os encoding [18:50] the problem is that you can't convert utf8 to ascii, unless you're only using ascii characters [18:50] nessita, and when python tries to convert unicode into "ascii" bytes, we get that error. [18:50] dobey, wrong, that is not *this* problem. [18:51] dobey, this is "unicode" to ascii [18:51] exceptions.UnicodeEncodeError: 'ascii' codec can't encode character u'\xf1' in position 161: ordinal not in range(128) [18:51] alecu: that is what i just said :) [18:51] no [18:51] alecu: so you're saying that python is trying to encode to ascii and not to utf8? [18:51] dobey, you said "utf8" [18:51] nessita, right. It's trying to convert unicode to the default encoding [18:52] nessita, since LANG is not set, the default encoding is ascii [18:52] nessita, and that is *right* [18:52] alecu: then we need our defaults encoding to be utf8, included tarmac [18:52] nessita, no [18:52] alecu: why is right? :-) [18:52] dobey: unicode != utf8 [18:53] yes i know that [18:53] nessita, because we'll get the same error if the user has other encoding [18:53] nessita, it's ok to have another encoding [18:53] dobey: ah, from your last sentence I thought you were mixing them, sorry [18:53] alecu: but this is a tearDown... [18:54] eh, it's not my fault windows is the red headed step child that it is [18:54] alecu: you propose is better to have an ifplatform there instead of us, devs, have default encoding to utf8? [18:54] alecu: I will trust you on this one, of course [18:54] nessita, right. I propose we stick to the convention [18:55] nessita, the convention being "os.* gets unicode on windows, bytes on linux" [18:55] why would os.* get unicode on windows? is that a convention we're enforcing ourselves? i thought python2.x used bytes by default everywhere, and python3 uses unicode by default [18:56] dobey, os.* needs unicode on windows, otherwise it starts doing funky stuff. Like os.listdir returning a list where some items are str and some are unicode. [18:58] well, if you know how to fix it, i'll let you fix it then :) [18:58] alecu: fixing [18:58] alecu: we may consider running the whole tests suite with lang and language unset [19:01] nessita, it seems that tarmac already does that. Or it uses a LANG with no ".UTF-8" appended. [19:01] it doesn't use a lang [19:01] it's run from cron, which basically has no env [19:04] great [19:04] the problem is developers expecting everyone else's env to be exactly like theirs ;) [19:14] gotta go put a contract in the mail, will be back in about one hour [19:19] you shouldn't use mail for that. it's traceable [19:32] nessita, do you need a re-review on the "bytes on linux" change? [19:32] alecu: is pushed, I was waiting for LP to rescan to approve [19:32] alecu: but yes, a re-review is welcomed! [19:35] nessita, alecu, thisfred: quick, pretty trivial review on https://code.launchpad.net/~dobey/ubuntuone-installer/launch-cp/+merge/70479 please? :) [19:37] dobey: I can't say no again ;-) [19:38] dobey: looks good! Question, does ubuntuone.installer.gui import Window exists? [19:39] dobey: also, how are we handling the fact that we're moving to the -qt control panel? I see the gtk version 'hardcoded' [19:40] nessita, tested with LANG unset, and approved again. [19:40] nessita: yes; define "moving to" more explicitly. is there a timeline for making those changes on Ubuntu? because the qt stuff isn't even packaged yet [19:41] nessita: and i was thinking we'd do some stuff to make it work with alternatives or something, so that either would work [19:42] dobey: right, it won't happen in ubuntu for oneriric, but the plan is to happen soonish. So, I was thinking, if this is the entry point for every ubuntu version to have the latest, are we needing a SRU to have older ubuntu running the qt version? [19:42] nessita: it is hardcoded right now, to make stuff work. there will be many changes in the next couple of weeks leading up to UI freeze, but for now, making it work is priority :) [19:42] dobey: ah, that explains it [19:55] dobey: ok, approving [19:56] alecu: thanks! [20:01] nessita: thanks [20:02] I can see the omgubuntu headline now. "Ubuntu One CP moving to QT... in time for oneiric?" And then the copy says "so far there are no plans to have this included in oneiric, but who can tell what the future will bring?" [20:08] irk3z: osom [20:08] :x [20:28] nessita, got 2' for a quick mumble? [20:28] sure [20:39] nessita: lost you in "pongamos que el control panel" [20:47] dobey: seems like tarmac may be in that state were it stops landing sutff [20:47] dobey: my branch is approved 49 minutes ago [20:48] oh, right, because it failed before. ok [20:51] ralsina: question re validate-harder: [20:52] nessita: shoot! [20:52] nessita: should get picked up in ~8.5min [20:52] pylint complains with several lines like this: [20:52] 1161: [E1103, ResetPasswordControllerValidationTest.test_invalid_confirm] Instance of 'FakeResetPasswordPage' has no 'confirm_password_line_edit' member (but some types could not be inferred) [20:52] so I went to look to FakeResetPasswordPage and I see that the errors is "real" since FakeResetPasswordPage doesn't not have that attr but a: [20:52] self.ui.confirm_password_line_edit [20:52] ralsina: so, shouldn't the test call [20:52] self.controller.view.ui.confirm_password_line_edit [20:53] instead of [20:53] self.controller.view.confirm_password_line_edit? [20:53] nessita: indeed that's a bug [20:54] ralsina: is repeated a lot for several classes [20:54] ralsina: I can paste the whole output if you don't have a linux handy [20:55] yes, copy& paste mistake :-) [20:55] I mean :-( [20:55] It's ok, let me change it and push it [20:55] ack [20:57] Actually the tests work because I am using the same object for view and view.ui, anyway [20:57] so lint is actually wrong... [20:58] don't look at me, i told you to use pyflakes [20:58] I get "has no 'email_edit' member (but some types could not be inferred)" [21:00] nessita: ^ [21:03] ralsina: and does it have an email_edit attr? [21:04] nessita: indeed [21:05] ralsina: can you please ignore globally E1103 in that test file? [21:05] Check FakeCurrentUserPage, for example. self.ui = self and then self.ui.email_edit = FakeLineEdit() [21:06] Maybe if I do self.fake_line_edit = FakeLineEdit() pylint will see it [21:06] ralsina: nah, just ignore globally the E1103. there are a couple of valid C0103 though [21:07] nessita: ok, will look for those === yofel_ is now known as yofel [21:14] nessita: pushed with the C0103s fixed [21:23] ralsina: running tests in windows and after that approving === m_conley is now known as m_conley_away [21:23] nessita: cool, thx! [21:25] ralsina: are the tests in windows passing for you? [21:25] I got 35 errors :-/ [21:25] 35???? [21:25] I get 2, about the captcha [21:26] ralsina: let me use a fresh branch JIC [21:27] ralsina: is your trunk up to date? [21:27] nessita: it was a while ago, let me pull again [21:28] nessita: no revisions to pull [21:29] ralsina: I really gotta go, I'll paste the errors and maybe you can help me see what's wrong with my env tomorrow morning? [21:29] nessita: sure thing [21:29] nessita: if you have any reviews left, I can do them in a bit [21:29] http://pastebin.ubuntu.com/658890/ [21:30] ralsina: those are the failing tests ^ [21:30] weeeeeird [21:31] nessita: besides, they are in places unrelated to the changes, too [21:32] right [21:33] ralsina: I'll try to debug this tomorrow first thing. In the mean time, would you try running the tests in a clean branch of your branch, without setting any env var? just running make tests [21:33] run-tests.bat, sorry :-P [21:33] sure! [21:36] nessita: clean cmd, clean branch merged to trunk from launchpad, I get two or 3 failing tests, as usual. Here is a run with 3 failures: https://pastebin.canonical.com/50774/ [21:38] ralsina: any ideas what can be run with my env? [21:39] nessita: not really, never seen that :-( [21:39] (FYI, this failure is not usual: ubuntu_sso.main.tests.test_windows.SSOLoginTestCase.test_emit_login_error, only the generate captcha ones) [21:39] nessita: what's your PYTHONPATH? [21:39] ralsina: nothing [21:39] ralsina: the tets should take care of that [21:40] the bat file set it, right? [21:40] yes, sets it to "." [21:40] right [21:40] weird [21:40] ok, I'm out of here, I'll continue tomorrow [21:40] bye all! [21:40] but it should not even need that [21:40] just in case, are you sure it's unset? [21:40] bye! [21:40] ralsina: yeap, I confirmed [21:40] I'll re-check tom [21:41] reboot! ;-) [21:41] 3 times! [21:41] lol [22:08] more triviality: https://code.launchpad.net/~dobey/ubuntuone-installer/le-appertif/+merge/70499 [22:12] dobey: got it! [22:17] cheers === gato is now known as gatox