=== m_conley_away is now known as m_conley [01:49] wooohooo!!! [01:49] I've got SD requesting credentials to SSO on my windows VM!!!! [01:49] success! === m_conley is now known as m_conley_away === m_conley_away is now known as m_conley [03:12] I have had lots of problems with ubuntuone. Specially upload and with bigger files (50+ MB). Just now it seems to work but it takes long time from that is says 100% to I get a notice that sync is complete [03:13] spacebug-: their was a bug with resuming downloads over 25mb getting corrupted [03:13] hum ok [03:16] spacebug-: are the files getting synced now? Or all you still emperiencing difficulty? [03:16] I am trying it out as we speak [03:20] seems kind of stalled. syncdaemon.log -> http://pastebin.com/G6siGKpW u1sdtool --current-transfer says bytes written = deflated size. It said that like five minutes ago === m_conley is now known as m_conley_away [03:26] spacebug-: looks like their is one still in the queue. [03:26] spacebug-: what does this command return: u1sdtool --waiting | wc -l [03:27] 1 [03:27] ok [03:28] local queue or is that server side? [03:28] try it without the '| wc -l'. That will telll us which file it is [03:28] want just the filename or the whole line from that comamnd? [03:29] Upload(running=True, share_id='', node_id='a8355075-aa40-4d3c-8df8-7bbf9264f9e4', path='/home/spacebug/Ubuntu One/docs.2011-06-24.tar.gz.gpg', crc32='2574283533', hash='sha1:33d589ea0f8566bb87d7a87963d18838919879bf', previous_hash='', size='99420465', upload_id='02672fd1-6165-5e3b-9790-a948998f9eaf') [03:30] spacebug-: if it feels stuck on that file what you might try is removing that file from your Ubuntu One folder. Wait till it is deleted from the server one.ubuntu.com/files. Then readd the file to your Ubuntu One folder [03:31] is does not show on one.ubuntu.com/files [03:34] the file was created, but not uploaded [03:34] the "node" exists in the server, but with no content [03:34] hum [03:34] if I understand the bug correctly :) [03:34] moving the file out of u1 will trigger a unlink of the node, so the stalled upload will be cleanedup [03:34] well it could have the same node I guess since it is an old file I want to upload again to test [03:34] ok I will try that [03:37] now I got an unlink (not running) and still that upload (running) but syncdaemon.log says [03:37] 2011-06-24 04:35:06,486 - ubuntuone.SyncDaemon.sync - INFO - T:LOCAL:F 810eff59-4b2a-4a6a-bde8-b4fdeaa7de7a ['root'::'a8355075-aa40-4d3c-8df8-7bbf9264f9e4'] ''Ubuntu One/docs.2011-06-24.tar.gz.gpg'' | Calling cancel_upload_and_delete_on_server (got FS_FILE_DELETE:{}) [03:37] 2011-06-24 04:35:06,489 - ubuntuone.SyncDaemon.sync - INFO - -:-:- - ['-'::'-'] ''/home/spacebug/Ubuntu One/docs.2011-06-24.tar.gz.gpg'' | Called cancel_upload_and_delete_on_server (In: T:LOCAL:F) [03:52] spacebug-: maybe you need to restart the syncdaemon. u1sdtool -q; u1sdtool -c [03:53] now it looks more corrent [03:54] should I try to add the file back to Ubuntu One folder now? [03:54] ok [03:54] yes [03:54] got a message that file is being uploaded [03:56] u1sdtool --current-transfer now says written bytes = total butes again [03:57] but u1sdtool --waiting still says upload (running) [04:11] creating a small textfile in Ubuntu One folder gets it uploaded at once, no problem [04:18] spacebug-: might see what you get from: u1sdtool --info='/home/spacebug/Ubuntu One/docs.2011-06-24.tar.gz.gpg' [04:19] this http://pastebin.com/5ED0tUYd [04:28] all looks ok...other then the fact that it is not uploading, but you already knew that. [04:28] hehe yeah [04:29] if you want you can restart the syncdaemon in DEBUG mode to get more info in your logs [04:29] ok [04:29] echo -e "[logging]\nlevel = DEBUG" > ~/.config/ubuntuone/logging.conf; u1sdtool -q; u1sdtool -c [04:30] that command should take care of that [04:30] ok will try that [04:36] output http://pastebin.com/mKsv4cve [04:45] now it completed [04:49] last lines when it finally completed. http://pastebin.com/ApheY3d8 [04:54] 6 am here in sweden, so bed time. [06:16] I'm having trouble syncing playlists with the Android UbuntuOne app. I heard there was an outage before, but so far everything else works. The only error I get is "A network error has occured". Is there a more extensive log stored somewhere on my phone? [08:23] hello tntc [08:24] i think karni is who you want to talk with about that...He was in pretty late last night so i do not know how early to expect him today. :) [08:25] tntc: if you can hang around for a little bit. More people will start showing up in the next few hours [09:30] morning! [09:38] ralsina: morning, wow you are up early! [09:38] mandel: decided to sleep early last night instead of midnight hacking :-) [09:39] ralsina: makes sense for an old man ;) [09:39] mandel: y acompañarte un rato para que no te sientas solito ;-) [09:39] ralsina: jajaja muchas gracias :) [09:39] mandel: can't find the problem with running u1trial on windows :-( [09:40] ralsina: may I know what are you executing? [09:40] mandel: does run-tests.bat work for you on https://code.launchpad.net/~ralsina/ubuntuone-windows-installer/fix_800359/+merge/65703 ? [09:41] mandel: for example, u1trial ubutuone_installer --gui --reactor=qt4 [09:42] ralsina: let me try it on my vm, give me some mins [09:42] mandel: great thanks! [09:47] ralsina: le haces un review a esto de alecu: https://code.launchpad.net/~alecu/ubuntuone-client/fix-cmt/+merge/65763 [09:47] mandel: como no... [09:51] mandel: forget about the tests, I fixed it :-) [09:51] ok, no I didn't [09:51] ralsina: I'm looking at it, I'm getting an error from credentials, but that is because I did not follow the isntrcutions :P [09:52] mandel: well, the instructions are way too complicated, but that's where we are right now :-) [09:53] +1 on alecu's... should we set it to approve? [09:53] ralsina: yes, lets do it [09:53] mandel: done [09:54] ralsina: ubuntuone.credentials is in ubuntuone-client, right? [09:54] mandel: yes [09:55] there is an error in the instructions, PYTHONPATH should be .;../ubuntuone-client [09:55] assuming ../ubuntuone-client is trunk of it with the copied clientdefs [10:00] ralsina: I'm getting the following when running the tests: http://paste.ubuntu.com/631717/ [10:02] ralsina: found the error, the reactor you are using is the txnp and not qt [10:03] ralsina: change txnp to qt4 in the run-tests.bat [10:07] mandel: I get reactor not supported with that to [10:07] with reactor=qt4 [10:07] and I have it right there in the same folder [10:07] ralsina: can you edit your u1trial to print the actal exception rather than that lame message [10:08] mandel: sure [10:08] ralsina: I think the issue is when you set the path [10:08] u1trial may not be picking it? [10:11] no module named reactors.qt4 [10:12] it's trying to import it manually using __import__ :-P [10:12] I need to move qtreactor into twisted for it to work, I suppose [10:13] reactor_name = 'ubuntuone.devtools.reactors.%s' % options.reactor [10:13] reactor = __import__(reactor_name, None, None, ['']) [10:13] WTF [10:14] ralsina: that is correct, is trying to find the plugins that install the reactors before it runs the tests [10:15] ralsina: so, the problem here is that ubuntuone.devtool.reactors is not in your path [10:15] take a look to see if according to the way you are setting the path you ca import ubuntuone.devtools.rectors [10:15] ubuntuone.devtools is [10:16] ubuntuone.devtools.reactors fails. Weird, but it's just python now :) [10:16] there is no reactors module there. I just installed this from trunk yesterday! [10:17] ralsina: well , reinstall ;) [10:18] it doesn't copy the reactors module to site-packages [10:18] sigh [10:19] oh, good, setup.py install doesn't copy *anything* to site-packages. Ok, I can fix this. [10:19] oh, it does the modern pth thing now [10:21] ralsina: wht modern pth thing? [10:22] ok, turns out devtools changed how it installs and managed to keep using the old version that had no reactors support [10:22] time to write last night's report then ;-) [10:23] mandel: instead of site-packages/ubuntuone/devtools devtools is now in site-packages/ubuntuone-dev-tools [10:23] and it creates a .pth file in site-packages to make it look like it is in the old path [10:24] but if you *have* a ubuntuone/devtools folder, that has priority and you have the new u1trial with the old modules [10:24] honestly, it's a bug in u1trial that should check versions matching but whatever [10:24] oh… that I hate that, it does not work with py2exe :( [10:27] ok, tests run and fail in amazing ways. I can probably fix that too. Thanks for the help with the reactor mandel! [10:30] ralsina: no worries :) [10:31] mandel: for when I finish with fixing/adding tests... how's the sdtool/syncdaemon status now? [10:33] ralsina: I'm working on using defer.inlinecallback to make it saver… but I'm not going fast, I've got a migraine… [10:33] I might take 30 min to take some drugs and see if it gets better [10:34] mandel: ok. [10:34] mandel: migraines suck. Drugs help :-) [10:34] ralsina: right now I'm in a closed room with all the lights off, lets hope it does not get much worse.. [10:47] mandel: this ring any bells? https://pastebin.canonical.com/48982/ [10:50] * mandel looks [10:50] ralsina: yes, the usso client code is trying to start the process of sso because it is not there but fails because it looks for it path in the registry which you did not set [10:51] Ran 5 tests in 1.867s [10:51] FAILED (errors=20, successes=1) [10:51] QWaitCondition: Destroyed while threads are still waiting [10:51] ralsina: make sure you have sso running or you mock that part, this is becuase all the software we write assumes that things like starting and stopping services on windows work like dbus, and it is not the case... [10:52] mandel: cool, thx [10:56] I get "sso is not running" even when SSO is running. Obviously something is very different when I use the tests compared to the real thing. Ok, mocking it... [11:02] ralsina: well, here you have a n insteresting situation, you are using the qt reactor in order to run the qt tests, which is ok, but as soon as you try to use sso you need txnamedpipes reactor [11:02] I should switch to txnamedpipes? [11:07] ralsina: so, the way to fix this would be to change the way txnamedpipes is installed in ubuntuone-dev tools to understand —gui so that is uses txnamedpipes and qt otherwise you wont have both [11:08] mandel: sounds like a "not now" kind of thing :-( [11:08] ralsina: well, you can do it, is not much work... [11:08] mandel: ok, I'll try :) [11:10] ralsina: just look at how the integration is done in sso, in main.windows.py [11:10] ralsina: and copy it... [11:10] mandel: ok [11:54] morning all [11:54] morning duanedesign [12:03] Hola duanedesign, ralsina [12:09] hola facundobatista [12:15] yes, hola facundobatista. Hello karni, You were up late last night ;) [12:16] duanedesign: Up indeed, but not working late this night. The night before, though, till 7AM ;d [12:16] hi duanedesign :) [12:18] karni: is their a more extensive log for the android client that someone could look at? [12:18] karni: someone was in early this morning asking about it... [12:18] duanedesign: you mean music or files? [12:19] it was music [12:19] duanedesign: I don't think so, as long as the app is released in non-debug mode. I do know, however, that we'll be looking into fixing the music app [12:19] syncing playlists and was getting 'a network error has occured' [12:19] duanedesign: as I heard it behaves badly (reconnects, etc) [12:20] duanedesign: That is CouchDB issue. [12:20] * duanedesign nods [12:20] duanedesign: From what I know, Couch (or some component) has been shut down, and that's why playlists are not working. [12:20] urbanape: What was the conclusion about playlists not working? [12:24] ugh. my screen instance crashed and i had to restart irssi. lost all my scrollback :\ [12:28] duanedesign: You have logs, no? [12:28] rye: you mentioned an upload hanging bug yesterday. Where the file keeps being uploaded 100% and hangs. Had a user that came in last night with this issue. Does restarting the client work to 'unstick' the queue? [12:28] duanedesign, no, this is happening due to the U1-server<->S3 interaction, i will poke verterok today about the progress on that [12:28] karni: I think i do... [12:29] rye: ahh ok [12:44] okay, vcard exporter from couchdb.one.ubuntu.com is ready to be tested [12:45] anybody in a mood for that (and having replication working too, since first shard is currently disabled until the software upgrade) [12:57] * Chipaca looks around for a mandel [13:11] hello all! [13:15] o/ [13:15] hola alecu! [13:16] hola ralsina! [13:16] Chipaca: mandel had a migraine, was taking a little time off to let the drugs work [13:16] alecu: I still have some test problems, do you have 15 minutes to lend a hand? [13:17] ralsina, I forgot that I have to take Amelia to the doctor today 10:30, about the time of our mumble. [13:17] ralsina, so I'll be not around for it [13:17] we can have the mumble a little later, I suppose [13:18] ralsina, cool [13:18] ralsina, do you want to tell me about the test issues? [13:21] sorry, was afk for a minute [13:21] hello everyone! [13:22] alecu: it's with this branch https://code.launchpad.net/~ralsina/ubuntuone-windows-installer/fix_800359/+merge/65703 [13:23] alecu: I managed to make it work but when I run the tests it never can connect to sso. So there is something different when it runs the tests compared to running standalone. [13:23] nessita: ralsina: alecu: any of you know if PIL is used with that name by us, or if we're importing Image? [13:23] alecu: and then I get 20 failures and 1 success for 5 tests :-) [13:23] (we should be using 'Image', but PIL used to work) [13:23] Chipaca: on windows? [13:23] ralsina: yes [13:23] Chipaca: do you have an importerrot? [13:24] no; the wiki says something about renaming the egg [13:24] * mandel back though sleepy due to medication [13:24] ralsina, looking [13:24] Chipaca: from PIL import Image [13:24] guh [13:24] oh ok, i'll rename the egg [13:24] we should be just doing 'import Image' [13:24] Chipaca, mandel, nessita, ralsina: I forgot that I have to take Amelia to the doctor today 10:30, about the time of our mumble, so I'll be not around for it [13:24] Chipaca: that is because easy_install add it to the lib as 'pil' instead of 'PIL' [13:25] Chipaca: import Image fails here [13:25] ralsina: what is 'here'? [13:25] Chipaca: windows 7 [13:26] ralsina: you renamed the egg? [13:26] Chipaca: nope [13:26] ralsina: how did you install PIL? [13:26] Chipaca: then again I installed it so long ago I may have forgotten. [13:26] Chipaca: let me check... [13:27] because what's probably happened is that you installed it via easy_install [13:27] and renamed the egg [13:27] so now python can't find the egg as pointed to from the .pth file [13:27] mandel: what medication? [13:27] but it finds the egg and goes with that [13:27] nessita: migraine [13:27] Chipaca: I have no egg. [13:27] ralsina: that's too eggcentric for my tastes [13:28] Chipaca: I may have used an installer from PIL's website [13:28] nessita: migraines… I have severe ones and the side effects of the drugs are being sleepy and throwing up, not pleasent but better than the headache [13:28] http://effbot.org/downloads/PIL-1.1.7.win32-py2.7.exe [13:29] Chipaca: I had that IMportError, it was fixed by renaming the folder under LIb/site-packages from pil to PIL [13:29] mandel: I was 3 years on a med where one of the side effects was "may cause constant itching, severe coughing and depression" [13:30] dobey: any idea why this branch https://code.launchpad.net/~alecu/ubuntuone-client/fix-cmt/+merge/65763 is not being landed by tarmac? (is approved, has commit message) [13:30] gah [13:30] fail on the part of PIL [13:31] there should be just one way to use the library :( [13:32] Chipaca: known issue for years, really. I will do a branch for it. [13:34] mandel: any news the fixes pending for on https://code.launchpad.net/~mandel/ubuntuone-client/windows_sdtool/+merge/65493? [13:34] nessita: working on them sorry, I woke up terrible this morning and could not work much… I should be done in 30 min - 1hour [13:34] ralsina: I'm seeint 14+ bug mails concerning _features_ you'd like to "Add" in my inbox - this is a feature envy for having a Kanban Board ;P [13:35] mandel: ok, let us know [13:35] karni? Uh? [13:35] karni: no idea what you are talking about [13:36] alecu: can you please approve mandel's branch? I changed my vote to needs fixing to track the needed change, but since you leave at 10:30, I would like to have your approve to land as soon as the changes lands [13:36] nessita, sure [13:36] ralsina: Public bug reported: "Add "update your storage options" page to the installer" "Add "add storage" page to the wizard" "Add "syncing your computer to the cloud" page to the installer" "add "credit card problems" page to the installer" "add "successful payment" page to the installer" and so on. [13:36] thanks! [13:37] karni: those are in the ubuntuone-windows-installer project [13:37] done [13:38] karni: Unless I did something very very wrong two days ago ;-) [13:38] ralsina: "You received this bug notification because you are a member of Ubuntu [13:38] One hackers, which is subscribed to Ubuntu One Windows Installer." [13:39] karni: well, yeah. Sorry, I understood you said I wanted *you* to add features :-) [13:39] ralsina: And the bad part is that, what I've recently heard from someone trying to unsubscribe - hackers can't unsubscribe from lists that they were subscribed by a joining a 'dependant' team [13:39] ralsina: Not really ;) [13:40] karni: you can silence bugs in launchpad. Maybe you can silence projects? [13:40] Oh, but you are in u1-hackers. Ok, yes, you are going to get those. [13:40] ralsina: I'll look into that :) (too bad gmail doesn't have filtering on headers ;< ) [13:40] Just like I get the mobile new bugs and all that [13:45] Chipaca: trivial branch for your PIL problem: https://code.launchpad.net/~ralsina/ubuntu-sso-client/fix_801550/+merge/65782 [13:47] mandel: did you see the reply you got re your bug in win32? [13:47] pywin32 that is [13:47] Chipaca: no, let me take a look [13:48] ralsina: True, but we're no longer using bugs to track features, etc. [13:49] karni: who's we? ;-) We are doing it for the windows client, because we are in a hurry. I could phrase them "there is no successful payment page in the installer" I suppose. [13:49] karni: we (desktop) are [13:49] ralsina: hahah :D good point [13:49] nessita: I know, that's why I started the disscussion in the first place :) [13:49] nessita: I said these bug reports are 'feature envy' for a kanban ^ ^ [13:50] *board [13:51] ah! :-) [13:54] standup in 7' [13:57] Chipaca: answering the bug with example code that should not fail, lets see what is the answer... [14:00] me [14:00] me [14:00] me [14:00] me [14:00] mandel, dobey, thisfred? [14:00] mandel, thisfred, dobey, standup! [14:00] me [14:01] mandel, dobey? [14:02] me [14:02] alecu: go! [14:02] me [14:02] DONE: got SD requesting credentials using CredMgmtTool, bug #800669 [14:02] TODO: make control panel get the credentials right too; fix the rest of CredMgmtTool that was not needed for SD [14:02] BLOCKED: no thanks [14:02] NOTE: will be afk for most of the morning, taking Amelia to the doctor [14:02] * alecu goes to the doctor right now. ttyal [14:02] Launchpad bug 800669 in ubuntuone-client (and 1 other project) "InsecureJelly exception thrown when SD requests credentials (affects: 1) (heat: 6)" [High,In progress] https://launchpad.net/bugs/800669 [14:02] NEXT: fagan [14:02] ALMOST DONE port to rst blocked nope nessita go [14:02] DONE: bug #801254, bug #801256, started settuping the win7 VM [14:02] TODO: integration testing [14:02] BLOCKED: nopes [14:02] NEXT: ralsina [14:02] Launchpad bug 801254 in ubuntuone-control-panel "QT UI: Remove storage info from folder list (affects: 1) (heat: 6)" [Medium,Fix committed] https://launchpad.net/bugs/801254 [14:02] Launchpad bug 801256 in ubuntuone-control-panel "QT UI: Add "Explore" button to each folder row (affects: 1) (heat: 6)" [Medium,Fix committed] https://launchpad.net/bugs/801256 [14:02] DONE [14:02] * Windows catchup call [14:02] * reviews [14:02] * finished initial integration of SSO into wizard [14:02] * Tried to get the tests running in the wizard on windows (failure) [14:02] * bug #801550 [14:02] TODO: fix the tests / start sd/u1cp integration into the wizard [14:02] Launchpad bug 801550 in ubuntu-sso-client "The PIL import sometimes fails depending on how PIL was installed (affects: 1) (heat: 6)" [Undecided,Fix committed] https://launchpad.net/bugs/801550 [14:02] NEXT: thisfred [14:02] I've installed 11.04 on my desktop and kept my home. I used to have 10.04. I can't log onto ubuntu one. I suppose that's because of old settings. How do I fix this? [14:03] DONE: reviews | launcher integration tinkering TODO: wrap launcher integration improvements BLOCKED: no NEXT: dobey [14:03] λ DONE: finished bug #800294, bug #801263, killed 'latest' milestones [14:03] λ TODO: test on oneiric, bug triage, releases [14:03] λ BLCK: None. [14:03] Launchpad bug 800294 in ubuntu-sso-client (Ubuntu) (and 2 other projects) "Support new fdo secrets API property names (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/800294 [14:03] Launchpad bug 801263 in ubuntuone-client (Ubuntu Oneiric) (and 7 other projects) "libsyncdaemon header files don't compile under g++ (affects: 1) (heat: 6)" [Medium,In progress] https://launchpad.net/bugs/801263 [14:03] mandel: vamos [14:03] DONE: Not much, tried to work this morning but got a severe migraine, took drugs and have been throwing up since then. [14:03] TODO: finish sdtool fixes so that nessita can use it in control panel. survive. [14:03] BLOCK: health [14:03] next? [14:03] dobey: any idea what's with tarmac? [14:04] * fagan break [14:04] nessita: I just got a branch merged... [14:04] hum, alecu's branch was approved for at least 3 hours [14:05] dobey: this branch seems to be not landing: https://code.launchpad.net/~alecu/ubuntuone-client/fix-cmt/+merge/65763 [14:07] mandel, Chipaca, ralsina: mumble at half past this hour? [14:07] nessita: fine by me [14:07] sure [14:08] Chipaca: want me to continue what you're doing in the VM? (not sure how far you got) [14:08] nessita: looks like a previous branch failed and left a badly named file around in the tree [14:08] dobey: ah. Which filename is that? === m_conley_away is now known as m_conley [14:11] nessita: something created by a test, not sure which one, but it causes a UnicodeDecodeError [14:14] dobey: care to say the path of the file? [14:15] mandel: if i knew which one exactly, i would [14:15] but the traceback doesn't include it [14:15] oh, ok I though you knew [14:16] hrmm [14:17] i am getting "Value could not be retrieved. (ServerError: (400, ('invalid_consumter_token_pair',…)))" in u1cp on services page, on oneiric [14:18] dobey: do you have the defoult.ini conf set? (custom for u1hackers) [14:20] ah that was it [14:21] right, ran into that yesterday too [14:21] thank you couch for asking before you nuke my settings [14:27] oi [14:28] I've installed 11.04 on my desktop and kept my home. I used to have 10.04. I can't log onto ubuntu one. I suppose that's because of old settings. How do I fix this? [14:28] mandel, ralsina, mumble? [14:28] nessita: ack! [14:29] going [14:37] ah [14:37] thisfred's branch failed i guess [14:38] * thisfred looks [14:38] two of the tests failed [14:38] yep, fixing [14:42] fixed [14:42] waiting for lp to catch up, then reapproving [14:42] Shouldn't I have gotten mail about this? [14:43] you did [14:43] or well, all of ~ubuntuone-hackers did [14:43] and that includes you [14:43] must have misplaced it [14:44] you probably just ignore those mails in notmuch :) [14:50] nope, I have almost 0 automatic filters set anymore: just spam, and I can silence individual threads. This works great, most of the time. I must have archived it without reading it, because I misinterpreted the subject line [14:51] ah, no, I archived the thread without reading the message count, thinking it was just my merge proposal still [14:51] mystery solved, lesson learned [14:53] ping rye [15:00] ralsina: this needs to go in for oneiric too: https://code.launchpad.net/~dobey/ubuntuone-client/ext-install/+merge/65793 [15:01] ^ and a second review would be cool :) [15:01] dobey: will review in a little while [15:03] jeroen-, pong [15:03] hello rye [15:03] remember me? [15:03] I'm still trying to get my files back. [15:03] jeroen-, hello, having attempted to recover the files we faced the issue with the database usage [15:04] jeroen-, so we are now changing the way it works [15:04] ok, but what tells this about my files? Do I get them back? [15:05] rye: ok, but what tells this about my files? Do I get them back? [15:06] jeroen-, yes, i am currently testing the branch that will recover the most recently changed files first [15:06] rye: OK, that would be great. Thanks for that! Again: for me it's only about "Documenten/Studie (archief)/Jaar 1" [15:07] and all the subdirs. it should be a couple of hunders of MB [15:22] nessita, I get the reactor installed when importing ubuntu_sso.qt.controllers or ubuntu_sso.qt.gui, but I can't yet see why. Anyway, here's the trace for the tests right now: https://pastebin.canonical.com/48996/ [15:23] ralsina: how are you running the code that triggers the reactor already imported? [15:24] nessita: I have the imports inside a method so they don't run before my main() [15:24] nessita: which is awful, so I would rather fix that first, since it makes the code much harder to test [15:25] ralsina: right. HAve the code do "what it should" and let's fix the errors, I help [15:25] * fagan back [15:25] nessita: ok, moving things to the right places now and pushing [15:26] ralsina: regarding the test errors, yes, you need to fake the "root" service from the SSO widgets [15:27] nessita: yes. The problem is that doing the imports inside the function is very tricky to replace it. Will be easy once the imports work correctly. [15:27] ack [15:28] This is the (not very interesting) trace about the reactor being installed: https://pastebin.canonical.com/48997/ [15:29] It happens when I import ubuntu_sso.qt.controllers [15:31] ralsina: what branch? [15:31] my branch is lp:~ralsina/ubuntuone-windows-installer/fix_800359 [15:35] ralsina: this import [15:35] is there a functin I can call to know if twisted has a reactor installed? [15:35] 25 from txnamedpipes.threadedreactor import install [15:35] ralsina: you need to move that to move() [15:35] main() [15:35] in ubuntuone_installer/gui/qt/main/windows.py [15:35] hmmm ok [15:36] ralsina: also, the close_callback should call reactor.stop [15:36] (and also app.quit? not sure) [15:36] probably not both [15:36] ralsina: from your source tree, see the linux.py file, it says: [15:37] 39 # The following cannot be imported outside this function [15:37] 40 # because u1trial already provides a reactor. [15:37] 41 # pylint: disable=W0404, F0401 [15:37] 42 from qtreactor import qt4reactor [15:37] 43 qt4reactor.install() [15:37] ralsina: so to not overlap reactors, the imports has to be done inside main() [15:37] Ok, let me try that [15:37] ralsina: check that ussoc is doing the same, please [15:38] ralsina: can i steal you for a quick mumble? [15:39] Chipaca: sure [15:39] nessita: the reactor isimported module-level in ussoc [15:40] ralsina: in which file? I'm grepping and I don't see that [15:40] ubuntu_sso/main/windows.py [15:41] ralsina: but that is ok, is not an specific reactor. What I mean is: that import should be called after you install your own reactor [15:42] ralsina: so, if in your import chain, that code is being executed before you install your reactor, you get the issue you're getting [15:42] ralsina: so, before executing that import (ubuntu_sso.main.windows) you should have your reactor installed [15:42] ralsina: yes? [15:42] aha! [15:43] ralsina: probably from ubuntuone_installer.gui.qt.gui import MainWindow needs to go after the reactor is installed [15:43] inside main() [15:44] but ubuntu_sso.main.windows is imported by controllers :-( [15:48] so, if I import controllers from module level, then main is imported. Therefore, I think I should move the import into main() so it doesn't break [15:49] ralsina: which import to which main? [15:49] the import reactor from ubuntu_sso.main.windows [15:50] ralsina: if that unblock you, move it. But the base issue is in the installer code... [15:50] * nessita -> mate [15:51] I don't get it, sorry [15:51] let me rethink 5 minutes and let's talk after mate [15:55] 2 easy reviews please? :) [15:55] https://code.launchpad.net/~dobey/ubuntuone-client/ext-install/+merge/65793 [15:56] dobey: ill do it [15:56] done +1 [15:56] dobey: I had reviewed it but not +1d it. Done [15:57] thanks [15:59] Arghhhh! If I have "from txnamedpipes.threadedreactor import install" anywhere after "from ubuntuone_installer.gui.qt.gui import MainWindow" I get this: https://pastebin.canonical.com/49004/ [16:00] shen me niao! [16:00] And if I have them the other way around, I get "reactor already installed" [16:00] dobey: your chinese is improving! :-D [16:04] ralsina: you should be doing from txnamedpipes.threadedreactor import install *before* from ubuntuone_installer.gui.qt.gui import MainWindow, inside main() [16:04] nessita: that gives the backtrace above [16:04] the glass house that twisted built [16:05] oops, sorry, that gves the reactor already installed error [16:06] unless I do the import after the install() in which case it works! yay! [16:07] ralsina: of cource after the install! :-) [16:07] thanks nessita, you rock! \oo/ [16:07] the thing is to install a reactor, yey [16:07] ralsina: now, to the mockers! [16:07] So, I moved from 15 ugly imports to two. Progress! ;-) [16:08] :-) [16:09] nessita: first I need to make the app stop when the window closes. Should not be terribly hard. Used to do it ;-) [16:11] ralsina, after the changed mandel suggested for sso (shutting down by using Qapp.quit in addition to reactor.stop), sso "started stopping". But it takes exactly 50 seconds for sso to completely finish since the quit+stop, so something weird remains. [16:12] 50 seconds? whoa [16:13] alecu: hey! [16:13] this one doesn't stop even in 50 seconds, even after quit and stop [16:13] how's amelia? [16:13] eso, how's amelia? [16:14] nessita, amelia is just fine, but I forgot we had an appointment with the doctor to get the paperwork so she can return to kinder on monday. [16:14] ah [16:14] it's Luli and me that have caught the sore throat now :P [16:15] heh [16:15] alecu: question re making sso works on w7: when trying to build pyopenssl, it complains that "could not find openssl.exe". Did you have this? were you able to resolve it? [16:16] alecu: and then you give it back to her, and so on until spring. Been there, done that ;-) [16:16] ralsina, lol [16:16] nessita, try building the smallest amount of windows python modules as possible. In this case, get the precompiled .exe. [16:16] nessita, let me fetch the link. 2.7 on 32 bits, right? [16:17] alecu: yes :-) [16:18] nessita, easy_install http://launchpad.net/pyopenssl/main/0.11/+download/pyOpenSSL-0.11-py2.7-win32.egg [16:19] alecu: I already did easy_installl pyopenssl, is not the same? [16:20] * ralsina goes for coffee, will mock in 5' [16:20] nessita, what version was installed? [16:20] nessita, or did you get the above error? [16:21] alecu: when using easy_install, I got 0.12, and when running show_gui, there was an error 'errtype': 'SSLSOmethingError' [16:22] alecu: I branched pyopenssl from lp and was trying to build it [16:22] alecu: that failed with 'can't find openssl.exe' [16:22] nessita: building random mdules from source on windows usually ends badly [16:22] ecolecuá [16:22] oh you're trying to build pyopenssl [16:22] ralsina: well, I see your point, and I thank the advice. [16:23] nessita, try easy_installing from the egg as I pasted above. [16:23] ralsina: but what am I suppose to do is ussoc does not work by following the wiki procedures? I will try another options, I need this running [16:23] alecu: I'll try [16:24] nessita, hmmmm [16:24] nessita, I have this installed: pyOpenSSL-0.12-py2.7-win32.egg [16:25] so I'll probably used some other package. [16:25] but I surely did not build it from scratch. [16:25] alecu: easyinstall pyopenssl brings that [16:25] (exactly that) [16:26] nessita, then I have that same setup. [16:26] lunch time, bbiab [16:26] alecu: if you run show_gui, when the UI dialog is shown, do you get an error saying "errtype: SSLSOmethingError" where Something is "verification" I think? [16:30] nessita, I don't get that error, but currently I'm not getting the credentials with show_gui either. I'll try updating [16:30] brb [16:31] mandel, alecu: I'm uploading the screenshot with the error. The exact string is: 'SSLHandshakeError' [16:34] mandel, alecu: http://ubuntuone.com/p/10yZ/ [16:35] dobey: ping [16:35] * mandel looks [16:35] mandel, alecu: as you can see, in the back, there are traces that are being raised by the txn reactor (I guess). This screenshot shows the backtraces better: http://ubuntuone.com/p/10ya/ [16:36] nessita: can you pastebin the full trace? [16:37] mandel: I ll try, but there is nothing more than what the screenshot shows [17:04] nessita: dobey alecu ralsina: I want to move all the launcher integration code to u1cp, because there is no good way to make it work with two different processes talking to the launcher. Any objections? (And yes this means extra dbus calls from u1client to u1cp which will then make dbus calls to the launcher :( ) [17:04] No way around that that I see [17:04] thisfred: objection! :-) [17:04] thisfred: won; tthis mean u1cp will run all the time? [17:04] quelle surprise! ;) [17:05] thisfred: I think several stuff that you use is not provided in the dbus iface, and some of those I *think* can't be provided [17:05] ralsina: only when we start talking to it, by dbus activation [17:05] thisfred: and she didn't even call you "your honor" :-) [17:05] thisfred: to solve your specific issue, didn t we discussed this in UDS, a good solution for it? [17:05] nessita: I don't understand? I would be adding new stuff to the dbus interface [17:05] thisfred: for example, if there is need to show a progressbar, and u1cp is the one making unity show it? [17:06] yes [17:06] thisfred: didn't jason said that unity lib will callback the caller? [17:06] unity-lib sucks because it's blocking [17:06] We have no idea if/when that will be available though.., [17:07] and I'm also no longer sure that solves all our issues [17:08] thisfred: well, from my POV, we don't have the time/resources to do so in time for O [17:09] thisfred: despite if it's a good idea or not (not arguing that), I'm not sure we can squeeze that in our roadmap... :-/ [17:09] nessita: which means we can't fix the launcher: clicking it to open the panel breaks all the launcher features [17:09] thisfred: for that particular issue I will remain Jason we need the callback he mentioned in UDS [17:09] remind* sorry [17:10] Ok, I'll ask him [17:11] thisfred: thanks [17:11] thisfred: and sorry, but we're very tight with time vs resources [17:12] I know :) I'll be happy if we can solve it in this way, but if that becomes available just before feature freeze, it's not much good to us until Pretty Pony [17:13] mandel: ping [17:18] ok I have about an hour left of work for this porting thing [17:18] hmm, maybe it's already possible with the low level dbus API: https://wiki.ubuntu.com/Unity/LauncherAPI [17:18] ill take a break because my brain is starting to melt [17:19] joshuahoover: ill ping you when im finished not too much left [17:19] fagan: excellent! thank you :) [17:22] thisfred: i object [17:22] joshuahoover: what's up? [17:23] "dobjection", we calls it [17:28] dobey: any advices to workaround http://code.google.com/p/httplib2/issues/detail?id=154 ? [17:28] dobey: is driving nuts [17:31] Ok, jason refreshed my memory, and it's actually fine: we can leave most everything as it is, and it will start working when libunity fixes multi process communication. All communication with the launcher will be from the client, and then we'll be fine [17:31] nessita: it claims it is fixed? [17:32] nessita: is this the problem you were having with lazr? [17:32] dobey: is not lazr (one sec, mumble) [17:33] oh [17:33] mumble uses python httplib2? [17:33] dobey: cjwatson asked @ the ubuntu release meeting today: could you move ubuntuone-client over from python-support to dh_python2 while you're at it? http://wiki.debian.org/Python/PythonSupportToDHPython2 [17:33] i thought mumble was a c++ qt app [17:34] joshuahoover: maybe. :) [17:34] dobey: k...i told cjwatson i'd pass that along to you, so there you go :) [17:35] ralsina: pong [17:35] duanedesign: playlists are working again (referring to the user with network problem in Ubuntu One Music app){ [17:35] mandel: tengo una duda muy tonta de como usar SSO [17:36] ralsina: here in eng hehe [17:36] * ralsina is getting channel-dizzy [17:37] Basically, if I create a Credentials object, then the right values are used, and everything is dne correctly [17:37] joshuahoover: yep, he's not the only one bugging about it :) [17:37] I change the app name => the new app name is used. That's good [17:37] dobey: heh...got ya [17:38] mandel: however, I am not passing that credentials object to anyone! Is that class some sort of singleton? Or it works by accident? :-) [17:38] ralsina: may I see the code to understand what you mean [17:38] ralsina: pastebin would be good enough [17:39] mandel: here https://pastebin.canonical.com/49015/ [17:39] line 31, I am creating a ubuntu_sso.credentials,Credentials [17:40] Oh, wait, I *am* stupid [17:40] embedded_sso gets it [17:40] * ralsina should stop understanding things AFTER he looks like a moron [17:41] go bleed in peace, mandel ;-) [17:41] ralsina: yeah I should do that too [17:41] :D [17:41] fagan: you bleeding for some reason? [17:42] ralsina: na the understanding things after looking like a moron [17:42] Oh, the moron part ;-) [17:42] * fagan moron :D [17:42] ralsina: my brain is not working fast enough to understand the problem… I should go back to the bat cave [17:42] mandel: it's ok, the problem was in my head [17:45] thisfred: anyway, let's actually discuss that next week perhaps. today is not a good day for discussing big things like that :) [17:45] dobey: sure, though it looks like it's a much smaller thing than I thought [17:45] and i already owe some big change thoughts to nessita [17:46] thisfred: well even if the code is small, the complexitiy increase would be pretty large [17:47] dobey: eh no: we won't be changing anything much: one new dbus method in the client to unset the urgency is all [17:47] and then it works, or will work when libunity is fixed [17:48] and it will not break worse than it does now [17:48] but yeah, shutting up until next week [17:48] oh, i was talking about not doing the "move everything to u1cp" [17:48] though I can't promise I won't build it :) [17:48] if nessita already convinced you, then good :) [17:49] dobey: yeah, that turns out to be an especially stupid idea [17:49] Whenever you two agree on something, I tend to listen ;) [17:51] it's rare, but it happens ;) [17:57] oh, ffs. g-s-d on lucid doesn't have plugindir variable in it's pkgconfig file :( [17:57] alecu, nessita, ralsina: I feel like crap… I need to go, it seems that my body has decided to go against me big time, I might catch up what I did not do today over the weekend [17:57] laters [17:59] bye mandel. Send an email so I don't worry, ok? [18:00] mandel: one question [18:01] mandel: do you have something else for the branch from sdtool? [18:03] alecu: we may need to land the sdtool branch as is and make the improvements later [18:04] dobey: I know you mentioned me several times but I did not follow the conversation. Can you please resume what you need my input in? [18:05] dobey: the problem with httlplib2 is triggered by our code using urllib to fetch the capctha image, so no lazr involved at all [18:06] nessita: good news, I found two bugs on how I was using SSO thanks to the tests (which are working ;-) and am fixing them [18:06] nessita, ok, let's file a bug and land this. [18:06] ralsina: see? "women and tests are always right" [18:06] ralsina: did you attend to my "mythbusters: tests" talk? [18:07] I'll give you a pass on tests being right ;-) [18:07] nessita: don't need your input right now. i was saying i need to get you some input from me :) [18:07] nessita: I did. I believe in tests. It's just that I got to them very late in my programming life and it's hard to form habit. [18:07] nessita: oh, i misread and thought you were saying that mumble had the issue [18:08] "when I was your age we tested by running the program a thousand times. In the snow barefoot! And our computers were made of ROCK!" [18:08] nessita: i thought u1cp was using soup to do HTTP? [18:08] nessita: oh, but sso doesn't? [18:08] ralsina: good news is that the testing habit can be learn, and is not too late for you :-) [18:08] huh [18:08] dobey: right :-/ [18:08] too damn many http libs. [18:08] dobey: anyways, installing httplib2 0.6.0 now [18:09] nessita: yeah. Believe it or not, this week has been the most fun I have had in front of a computer in months ;-) [18:09] nessita: i haven't had time to look at the issue more in depth unfortunately [18:09] ralsina: I'm glad :-) [18:09] dobey: do not worry, thanks [18:09] * alecu is having lunch. [18:10] ralsina: that's what she said ;) [18:11] dobey: I give you a C for effort on that one ;-) [18:13] * alecu is having lunch now fur realz [18:30] nessita, alecu: ready for re-review -> https://code.launchpad.net/~ralsina/ubuntuone-windows-installer/fix_800359/+merge/65703 [18:35] ralsina, on it. [19:06] ralsina: looking [19:08] ralsina, I'm reviewing your branch. When choosing either register or login, the following screen has no text. [19:09] ralsina, all the buttons are there, and all the entry boxes are there as well [19:09] ralsina, but no text [19:09] alecu: do you have the SSO client running on another terminal? [19:09] ralsina, yes [19:09] and it's not giving errors? [19:09] ralsina, the only text is the label next to the captcha [19:10] yes, that is the behaviour when it can't connect to the sso client [19:10] * ralsina doesn't quite understand it either ;-) [19:10] ralsina, ok, checking again [19:11] also, you need to start the sso client a second or two before the wizard, or it doesn't work [19:11] ralsina: is there any chance that you start using separated .py files for the widgets, instead of having all inside gui.py? the file is growing a lot [19:12] nessita: sure, trivial branch after this one. Or you want it for this one? [19:12] ralsina: no, no for this one. Maybe in the next adding (so the next thing you add is another file, no changes what is currently there) [19:12] nessita: sure [19:14] ralsina, http://pastebin.ubuntu.com/631933/ [19:15] ralsina, I start the sso client, wait a few seconds, start the installer [19:15] ralsina, but that does not look at all like the code you've been working on... [19:15] looks like something coming from txnamedpipes [19:17] * ralsina looks [19:17] alecu: no idea, never seen the like of that [19:21] hum [19:21] alecu, ralsina: maybe is not thaaaaaat straighforward to embed the widgets? [19:21] ralsina: I added some needs fixing after the code review [19:22] once those are fixed, I'll test on w7 [19:24] ralsina, "exception thrown when nothing is connected to the signal" -> perhaps you should catch only the relevant Exception type. [19:26] ralsina, (from a previous branch) why is this called twice? setButtonText(QtGui.QWizard.CustomButton1, _("&Print")) [19:27] (on different objects) [19:29] ralsina, the comment on CongratulationsPage.initializePage that goes "Layout without..." is wrong. [19:30] alecu: looking [19:31] alecu: do you have any pending branch for ussoc? [19:31] alecu: in particular to "fix" the bug where it answers on request and then nothing else [19:33] nessita, with this branch, it stops properly: https://code.launchpad.net/~alecu/ubuntu-sso-client/stop-properly [19:33] nessita, well, for a very vague definition of "properly" [19:33] alecu: yey! wanna propose for merge? [19:33] nessita, since it takes 50 seconds to shut down [19:34] guh [19:34] nessita, even after reactor.stop and qapp.quit are called. [19:34] in which order ? (still branching) [19:34] nessita, there surely is some issue in the txnamedpipes reactor that's making it all fail [19:35] nessita, qapp+quit, then reactor.stop. [19:35] nessita, if I only do qapp.quit, then it does not stop. [19:36] right [19:36] alecu: the Print is just a forgotten line, thanks for the "without finish button one" :-) [19:36] that sounds correct (first app, then reactor) [19:37] nessita, well, it sounds weird to me. Every other reactor integration with mainloops ends with just one call. And they end on the spot. Or pretty fast at least. [19:37] nessita, this "threadedselect" reactor is an ugly bicho. [19:37] indeed it is [19:38] did you see the traces I pasted from the txnamedpipes reactor when having the SSLHandshake failure? [19:39] alecu: look in the left: http://ubuntuone.com/p/10ya/ [19:40] nessita, the "generator already executing" ones? I did. [19:40] nessita, I don't like fiddling with threads, but this namedpipes stuff is pushing us that way. [19:42] Ok, got school run in 20 minutes. Will do some non-windows-related stuff and will put a few more hours when I come back [19:42] last night I started wondering again why we are not using dbus on windows, like kde for windows does. [19:42] it *is* very green. But our txnamedpipes is green too. [19:43] amazon rainforest green [19:44] alecu: good questions, not sure we're in position of changing paths at this point :-( [19:44] ralsina: what do you mean with non-windows-stuff? :-) [19:44] alecu: any reason why you don't submit for merge the ussoc branch? [19:44] nessita: canonicaladmin, expenses, etc [19:45] ralsina: ah. Any ETA on having the branch fixed? [19:45] nessita: well, I can't fix the problem alecu has because I don't have it :-( The things about comments and such are fixed already [19:46] ralsina: what about the things I added as needs fixing, are those fixed as well? [19:46] if so, I can start testing that on w7 [19:46] nessita, I don't like how it functions. I wanted to find out what's with the 50 seconds limit, or increase the timeout to +∞ [19:46] nessita: let me check... [19:46] nessita, but I guess I can push it like that. [19:47] alecu: the branch looks reasonable [19:47] alecu: and even if it's not the solution that makes thing work, it makes sense [19:47] nessita: had not seen it. Will reply right away [19:47] ralsina: thanks>! [19:59] nessita: all checked except the test case (will write it after I come back from school) and the import organization (same timeframe) [19:59] ralsina: thanks*2! [20:00] I've installed 11.04 but kept my home from 10.04. I can't use Ubuntu One. What do I do? [20:01] * ralsina needs to read the diffs carefully before proposing. Tattoing that on my belly memento-style tonight. [20:01] ubuntu_sso/qt/tests/show_gui.py:26:1: E302 expected 2 blank lines, found 1 [20:01] ubuntu_sso/qt/tests/show_gui.py:31:1: E302 expected 2 blank lines, found 1 [20:01] hmm [20:02] hmmmm u1lint ignores pylintrc on windows. That's kinda annoying. [20:03] ignores? [20:03] shouldn't [20:03] though devtools might not be fully win32-happy with path strings, if you know what i mean [20:04] dobey: basically gives wildly different results than in Linux. I will take a look later. [20:04] dobey: since the compiled ui files are very much not to pylint's taste, and the ignore is set on the pylintrc, it floods me with a bazillion useless warnings. [20:06] ralsina: what project is this? installer? [20:07] dobey: yes [20:07] ralsina: does setting env USE_PYFLAKES=1 work better? [20:08] dobey: don't have pyflakes right now. Was going to try that eventually [20:08] ralsina: ok, because i would rally like to get all our stuff to just use pyflakes, and switch that to the default, and drop pylint like a bad cup of mate [20:09] dobey: ha, ok, will make a few minutes to try that later today [20:09] great [20:10] dobey: what about using the tool that combines flakes and pep8? /me looks the name [20:10] dobey: flake8 [20:10] isd uses it [20:11] ok, off to school [20:11] I have been using it in emacs === ralsina is now known as ralsina_afk [20:11] works quite well [20:11] thisfred: nice [20:11] I've replaced pep8 and pyflakes with this. Still running pylint as well [20:13] nessita: i haven't looked at it at all [20:15] dobey: apparently is very good [20:16] well, it's pyflakes + pep8, not much more than that :) [20:19] also it [20:20] s not in ubuntu yet [20:26] alecu: is your syncdaemon generating logs under w7? [20:27] nessita, let me point you at the right folder... [20:27] alecu: I searched for 'syncdaemon.log' and no search results [20:30] nessita, C:\Users\user7\AppData\Local\Temporary Internet Files\UbuntuOne\log [20:30] (replace user7 with your username) [20:30] right [20:31] nessita, I think mandel got the xdg stuff confused, because it should be "Temp", not "Temporary Internet Files" [20:31] alecu: probably. Even more, I don't have that path [20:31] I don; t have the Temporary Internet Files dir [20:31] nessita, you wont find it in the file explorer === ralsina_afk is now known as ralsina [20:31] alecu: why not? I'm showing hiddens [20:31] nessita, try changing to that folder manually in the prompt [20:32] alecu: I see AppDara [20:32] nessita, cd "C:\Users\user7\AppData\Local\Temporary Internet Files\UbuntuOne\log" [20:32] ok [20:32] alecu: you're right :-) [20:33] nessita, it took me a looong time to find it :-( [20:34] alecu: aggregator is trying to run on windows and is failing... [20:34] thisfred: ^ [20:34] ouch [20:35] Well the aggregator is not in platform, so I'd expect it to be running [20:35] perhaps we'll need a "windows.py" module with empty methods or something like that. [20:35] nessita, it looks like it's time for thisfred to setup the Windows VM as well ;-) [20:36] we do for the platform specific bits I thought' [20:36] next in line: dobeeh! [20:36] nessita: what's the traceback? [20:36] thisfred: trying to copy the damned log file to this side of the world :-D [20:37] ah kk :) [20:37] I'm failing misserably [20:38] can't pastebinit in the vm? [20:38] aja! [20:38] dobey: yes, but I need the whole log file to also report bazillions of errors from syncdaemon [20:39] ah [20:40] the best transport layer ever: gmail drafts :-D [20:41] heh [20:45] if i disappear, it was the storm [20:46] "is not me, is the storm" :-P [20:46] oh i probably need to do a protocol release too :-/ [20:46] thisfred: https://pastebin.canonical.com/49022/, line 1000 [20:47] nessita: on it [20:48] thisfred: no rush, I would like you to indentify the issue and, please, an estimation of fixing time/effort [20:48] nessita: it looks like a bug I introduced with my latest branch should not be hard to fix [20:49] ok, there aren't any more client branches coming in the next 30 minutes or so, are ther? [20:50] dobey: I don't think so, at least on this end. alecu? [20:51] thisfred: that was the progressbar thingie for sure [20:51] nessita, dobey: no branches coming this way either, go ahead. [20:51] yeah, not sure why it's failing yet, but when I do I suspect it'll be a one or two liner to fix [20:52] whoot [20:52] uh, it just got dark here really fast [20:52] thisfred: well, if you can, would you please queue that to your next available slot (no need to interrupt what you're doing right now)? [20:52] dobey: breatheeee [20:52] i wonder if cthulu came up out of the chesapeake bay [20:53] dobey: uh oh. Not dark green is it? [20:55] http://www.southparkstudios.com/clips/360449/cthulhu#tag=Cthulhu [21:05] so it looks like some events work differently on windows? We're getting None from either bytes_written or deflated size. I'm putting in guards [21:06] thisfred: please file a bug so we can fix the windows implementation [21:06] bug #801736 [21:06] Launchpad bug 801736 in ubuntuone-client "Error when running tests on windows: TypeError: unsupported operand type(s) for +: 'int' and 'NoneType' (affects: 1) (heat: 6)" [Low,In progress] https://launchpad.net/bugs/801736 [21:06] ah, a bug for the None itself [21:06] kk [21:11] The following files contain translations and are currently not in use. Please [21:11] consider adding these to the POTFILES.in file, located in the po/ directory. [21:11] tests/platform/linux/test_tools.py [21:11] wtf [21:13] hmm, intltool bug :( [21:14] how did that even land [21:15] nessita: https://code.launchpad.net/~thisfred/ubuntuone-client/fix-progressbar-tests-on-windows should work around the issue, but more importantly, it will log when the None value is passed as an error level log entry, so it should give us some debugging hints. I'll change the bug to the underlying issue. When you're not busy (HAHA) please run the tests with that branch and see what happens. [21:17] Bug updated [21:19] bug #801736 [21:19] Launchpad bug 801736 in ubuntuone-client "None reported as 'deflated_size' or 'n_bytes_written' (affects: 1) (heat: 6)" [High,In progress] https://launchpad.net/bugs/801736 [21:20] thisfred: :-) [21:20] thisfred: you mean run the tests where? or you mean run the branch on w7? [21:20] (I can do the second for sure) [21:21] nessita: yes, that [21:21] ok, I will do that now [21:21] genial! [21:21] lol [21:29] joshuahoover: done [21:29] not tested though so I presume that there are going to be mistakes [21:29] fagan: great! thank you! [21:29] so ill give it a once over on monday morning [21:30] but all of the pages are there and its pretty trivial to fix any problems [21:31] fagan: ok, on monday i'd like you to do the same for all our tutorial pages (minus the index page): https://wiki.ubuntu.com/UbuntuOne/Tutorials [21:31] joshuahoover: sure [21:31] fagan: you can put tutorials in the same branch if you want...just put them in their own folder (something clever like, "tutorials") [21:32] joshuahoover: cool thats fine [21:32] fagan: did you at least test that the syntax is valid? [21:32] fagan: and BTW, it's 10 PM on a friday there, dude [21:32] ralsina: nope and I know some of it isnt right so ill be doing the rst2html and fixing what ever is wrong [21:32] fagan: ok [21:33] dobey: hi [21:33] ralsina: oh and its only 9:30 :D [21:33] adorilson: hi [21:33] fagan: there are some tools that let you see the HTML live while you edit [21:33] ralsina: oh thats interesting ill give it a google on monday [21:34] fagan: http://rst.ninjs.org/ [21:35] ralsina: nice [21:35] ralsina: nice ill use that to check then [21:35] dobey: can you see this: http://paste.ubuntu.com/632001/ ? [21:35] adorilson: yes, you are missing those things [21:37] adorilson: are you building from a tarball or from a checkout of trunk? [21:37] dobey: from trunk [21:38] dobey: following this https://wiki.ubuntu.com/UbuntuOne/RunClientFromTrunk [21:38] adorilson: you need to install the devel packages for those things that are not found [21:39] adorilson: you also need icontool [21:39] EOW [21:45] thisfred: a run this time will not generate any tracebacks for aggregator, but I also did not see any SYS_QUEUE_ADDED no REMOVED event [21:47] brb [21:57] eh, you're saying the events no longer fire? That seems unlikely to have been caused by my changes, since I only touched event handlers. Anything in the logs? [21:57] thisfred: nonono, I mean syncdaemon is not queueing up stuff [21:57] which is very bad since there are tons of things to update [21:57] thisfred: I'm pretty sure is other stuff, from volume manager or similar [22:03] nessita: sorry about the delay with the branch but bzr is telling me the branch diverged and I am trying to reconcile it === m_conley is now known as m_conley_away [22:09] ralsina: that sounds wei... [22:09] ralsina: why don't you try pushing to a new location? [22:10] nessita: it is. There is nothing to pull, it's up to date, but doesn't let me push [22:10] nessita: I don't want to lose the comments on the merge [22:10] it's almost done [22:10] ralsina: ok [22:14] thisfred: ping [22:14] nessita: pong [22:14] thisfred: I'm "reviewing" your changes, and I would like a function to unify the checking for a given variable value to be not none [22:14] thisfred: before you propose for merge, can you please change that? [22:14] nessita: I don't really want to land this in trunk, actually [22:15] nessita: but if we can't find the underlying problem, yes I will [22:15] thisfred: well, it makes sense to land it, I think those commands can have that value to be NOne [22:15] facundobatista: ping [22:15] thisfred: I'll ask facundobatista, he has been dealing with that for magicicada [22:15] nessita, pong [22:15] nessita: ok, then I'll clean it up and propose [22:16] and then I also don't want to log the None values as errors :) [22:16] nessita, what I can do with you? [22:16] facundobatista: you resolved this for magicicada, can the download-started/finished, etc have the deflated_size to be None? [22:16] *for* you, I mean [22:16] facundobatista: anything you want [22:16] haha [22:16] ah, sorry [22:16] :) [22:16] nessita, thisfred: yes it can [22:16] nessita: https://code.launchpad.net/~ralsina/ubuntuone-windows-installer/fix_800359/+merge/65703 should be ready, unless I forgot to re-merge some change. If there are only cosmetic problems, could you set it to approved? I promise to fix them later. [22:17] facundobatista: why? :-) [22:17] facundobatista: same question for bytes_n_read/writen [22:17] nessita: later as in "before merging" [22:17] thisfred, nessita: when download starts, the server informs the client the deflated size [22:17] ralsina: sure! [22:17] and then BYTES come [22:17] claro [22:17] so, there's a small window of time when the download as an operation started, but the server still didn't inform the client the deflated size [22:18] ah, ok [22:18] nessita, thisfred: for the upload this not happens, as the client is the one who knows the deflated size from beginning [22:18] that sort of breaks the idea of the progress bar as it stands now [22:18] So I need to fix that then [22:18] np, I know how to do it [22:18] thisfred: you can assume that a download with deflated_size None is a download that has not started [22:18] cool. EOD for me then. Will work a bit more late tonight or maybe early tomorrow to make sure that branch is good for monday very early, and will try to have a control panel widget in by monday morning at .ar [22:19] ralsina: thanks! [22:19] nessita: well we get the event from download started, but yeah ;) [22:19] thisfred: I know... :-) [22:19] thisfred: ok, you fix, I review (very likely to be next Monday) [22:19] The code will work with the fix (that I will now clean up) [22:20] I'll make a separate branch to make the progress bar not do weird (but non-breaking) things in this case [22:21] thisfred: great [22:22] nessita: Since this is not a very interesting occurence, I don't think I will log it, which means we just have the if foo is not None left, so I don't think there's much room for refactoring there anymore [22:22] thisfred: True [22:23] AND ubuntuone-client 1.7.0 uploaded to ubuntu [22:24] dobey: I suggest you start drinking right away! [22:24] i am certainly not going to sit at the computer any longer :) [22:26] today's suggestion: cucumber infused vodka + tonic [22:26] ice to taste [22:27] i was thinking rum [22:27] the good ole' black seal [22:34] or Malibu :P [22:35] nah [22:35] i could just go straight for the bourbon [22:35] anyway, have a good weekend all! [22:38] you too! [22:40] nessita: https://code.launchpad.net/~thisfred/ubuntuone-client/fix-progressbar-tests-on-windows/+merge/65838 monday is fine [22:41] Have a good weekend all! [22:41] thisfred: bye! [23:21] hi [23:22] anybody help with this errors: http://paste.ubuntu.com/632056/ ? [23:22] I'm try U1 from trunk.