=== wgrant_ is now known as wgrant [01:35] hey folks, I've been trying to use Ubuntu One for some time now, I have an account set up, I can log into it, but I can't get Ubuntu 10.10 to give me synchronize options when I right click on any folder... [01:37] Geoffrey2: hi pal. that's weird.. so you're saying you are running 10.10, right? [01:38] Geoffrey2: it's quite late (as in: the developers have left few hours ago). perhaps try tomorrow few hours earlier [01:38] Geoffrey2: I can't come up with a solution to your problem :( [01:38] yes, 10.10.... [01:40] Geoffrey2: you're runining Gnome right? (not Kubuntu?) [01:40] I tried the routine Joshua Hoover posted quite a while back on how to completely remove and freshly install ubuntu one, when I followed that, it didn't even create the ubuntu one folder... [01:40] oh.. [01:40] yep, Ubuntu 10.10, Gnome edition [01:41] That may be the case. Looks like something's missing. Things could have changed, when were those instrucitons updated? [01:41] I mean, do they apply to 10.10 as well? [01:42] dunno...posted end of 2009, I was referred to them in the #ubuntu channel.... [01:43] Oh. There are quite a few new things in Ubuntu One on 10.10, Geoffrey2 , it'd be best if you tried to catch the devs here on #ubuntuone tomorrow. nessita or rye perhaps. [01:44] They should be able to help you. [01:44] ok, I'll try then..... [01:44] Joshua also hangs out here from tim eto time. [01:44] *time to time [01:47] * karni gotta go, bye bye [09:11] HELO canonical.com [09:15] morning all [09:24] duanedesign, hi [09:24] hello zyga [09:52] aquarius: hello and god mmorning [09:52] err, good [09:52] seems my typing skills have not woke up yet [09:53] aquarius: is their any docs yet for the U1 api? [09:53] and god morning to you too :) [09:53] s/is/are [09:53] :P [09:53] duanedesign, not yet, unless you count the ones on my laptop :) [09:53] :) [09:54] cool, no worries [09:54] (that would be "U1 APIs" you mean, there; there are many :)) [09:54] lol [09:54] hehehe. sporo ludzi [09:54] ups [09:55] i should quit before I embarrass myself more with my fine grammar and spelling :) [09:56] hello karni [09:56] the documentation should be released as part of the developer programme, along with the natty release [09:56] sounds good [09:56] hi duanedesign, hi aq. im on mobile so please exuse me :) [09:57] karni: i know that feeling. I have yet to find a decent IRC client for my phone [09:57] darairc is not too bad [09:57] * duanedesign looking [09:57] duan: irssi connect bot (on Android) [09:59] i got my invitation to IRCcloud yesterday [10:00] so i have been trying it out. Though I have been, by habit, going to irsssi/screen [10:00] so I have not spent much time using it [10:37] hello all! [10:37] mandel, ping [10:38] alecu: pong [10:38] mandel, would you be able to do a review? [10:39] (see how early I got up... I really want it!!!) [10:39] alecu: do i need unity for the review you requested? [10:39] it's 7.30 am here! [10:39] mandel, no unity needed this time, just gnome. [10:40] alecu: ah, then I'll do it right now! [10:40] https://code.launchpad.net/~alecu/ubuntuone-client/inhibit-logout/+merge/50834 [10:40] coooool! [10:45] alecu: do i hae to try to log out? what does that TestSession do then? [10:50] mandel, the tests just verify that the dbus call was made with the right parameters [10:50] ok [10:50] mandel, to see if it work IRL, try logging out after running the syncdaemon in that branch., [10:57] alecu: do you know qhen did we started to print to stdout the logging from the credentials? [10:58] mandel, no idea... [10:58] mandel, is that happening inside syncdaemon or ubuntu-sso-client? [10:59] alecu: if I trust the logging name, it is in ubuntuone [10:59] alecu: nevethesless it is something that is in trunk, not only your branch [10:59] so +1 [11:00] let's add a bug for that issue... sounds serious. [11:07] ok, this should hit trunk any minute now... [11:07] mandel, thanks for the prompt review! [11:07] alecu: np :) [11:28] \ /4 [11:50] duanedesign, \o/ [11:50] :) [11:51] rye: i got my invite to IRCcloud yesterday. Been trying it out [11:51] but it is hard to break the habit of using irssi/screen [11:52] duanedesign, hm... like bip but on somebody else's server and in the browser... [11:53] * rye needs to be reminded to configure bip [11:57] hello everyone! [11:58] hello [11:59] rye: yeah. I could see IRCcloud being useful if someone did noy have a server or VPS to run irssi/screen on [12:00] duanedesign, well, once I was at a location with nice fast wifi which prohibited ssh on default port, and I was not yet ready to start listening on other ones, since I needed ssh to reconfigure my router... it was sad [12:01] ahh [12:19] ajaxterm ftw! [13:19] Aloha [13:21] I will be unavailable (off-line) for the rest of the day due to travel. Please feel free to leave me messages on IRC, I will get back to them asap. Have a great day! [13:35] karni, enjoy! [13:47] alecu, dobey, mandel: ralsina is delayed, he will be here in about an hour. Stand up in 13'. [14:00] me [14:00] me [14:00] me [14:00] dobey, thisfred? [14:00] me [14:01] ok, alecu, go! [14:01] DONE: did some fixes and got pending branches landed before FF, rushed a branch to inhibit logout, got that landed too (Bug #723359). Yay! [14:01] TODO: today I really get to work on DroidCouch! [14:01] BLOCKED: a blinking and hurting left eye is telling me I should get more rest [14:01] HATE: forgetting to make DBus calls asynchronous and getting the testrunner stuck with no feedback [14:01] LOVE: DBus anyways [14:01] Launchpad bug 723359 in ubuntuone-client "Syncdaemon should inhibit session logout when busy (affects: 1) (heat: 6)" [High,Fix committed] https://launchpad.net/bugs/723359 [14:01] NEXT: mandel-brot [14:02] DONE: Reviews. More IPC code for windows to land on ubuntu-one client. POC of how to get the status of the msi installation. A couple of reviews for alecu [14:02] TODO: more ipc [14:02] BLOCKED: no [14:02] nessita: please [14:02] DONE: finished setup of dev env. Booked tickets for Budapest. Restored replication 'Services' tab in the control panel. Packaged it. [14:02] TODO: look for sponsors for new file storage API package and new u1cp upload. More file storage API work. Follow up on several cp bug reports. [14:02] BLOCKED: nopes [14:02] NEXT: thisfred [14:02] DONE: out sick [14:02] TODO: get u1couch/oauth packaged | do what else needs doing before FF [14:02] BLOCKED: no [14:02] NEXT: dobey [14:02] any comments? [14:03] nothing? ok [14:03] eom! [14:03] thanks to mandel for prompt reviews this early morning! [14:03] * nessita claps [14:03] * thisfred is catching up with mail [14:03] if there's anything urgent in there feel free to ping me [14:04] thisfred: nothing urgent, I assigned bug #723327 to you [14:04] Launchpad bug 723327 in ubuntuone-client (Ubuntu) "Singular form is incorrect in notification (affects: 1) (heat: 6)" [Medium,Confirmed] https://launchpad.net/bugs/723327 [14:04] thisfred: no need to do it this week, but surely before string freeze [14:04] alecu: it was not that early here and I should have done it earlier, but I though I needed unite ;) [14:04] unity* [14:04] nessita: yeah saw that, should be easy enough [14:05] good [14:05] nessita, that bug is already fixed on nightlies. [14:05] it is? awesome [14:05] nessita, now it says "1 file(s) uploading" [14:06] alecu: can you please update the bug report? extra points for linking the branch and/or setting the dup [14:06] sure [14:08] hmm [14:09] by the way, if anyone is available for reviews, I'd really appreciate if you take a look at this https://code.launchpad.net/~mandel/ubuntuone-client/add_syncdaemon_remote_object_tests/+merge/50912 and this https://code.launchpad.net/~mandel/ubuntuone-client/add_syncdaemon_remote_object_tests [14:09] λ DONE: bug #723297 (trunk), banshee testing, music store bug triage [14:09] λ TODO: banshee music store fixes [14:09] λ BLCK: on-call for jury duty [14:09] Launchpad bug 723297 in libubuntuone (Ubuntu) (and 3 other projects) "Logs oauth tokens to console (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/723297 [14:13] * mandel -> small coffee break [14:19] mandel, should I do your reviews on windows? [14:20] alecu: do you have a bug for the work that is blocked by the libunity bug (#709240)? i'd like to file that freeze exception now if we haven't already [14:21] joshuahoover, we already landed that! I added a check so if the libunity python bindings are not installed syncdaemon runs fine anyway. [14:21] joshuahoover, libunity trunk works fine, so it just needs packaging. [14:22] joshuahoover, I was told it should be packaged by tomorrow [14:22] and then the only thing we need to do is add the libunity bindings for python as a dependency. [14:22] (well, the libunity gir bindings, that is) [14:22] alecu: ah, ok...as long as we can package it up tomorrow then we should be ok [14:23] nessita: has the sprint been confirmed yet? [14:23] thisfred: as far as I know, yes. But ralsina said he will be sending email about booking tickets [14:23] thisfred, yes! ralsina confirmed it [14:24] thisfred: ralsina will send e-mails i think [14:24] thisfred: so, as far as I know, no booking tickets yet [14:24] ah ok, I will wait for that then :) [14:24] thisfred: you can ceratinly ask for options though [14:24] to try to get the better ones [14:25] dobey, regarding packaging... our code in trunk uses the unity-gir if installed. It's not installed by default, and the current package is broken, but they say it should be fixed by tomorrow. [14:25] ok [14:25] nessita: Have to coordinate with my wife as we will be adding on a week of vacation in B.A. :) [14:26] alecu: i guess we need to recommends it then [14:26] thisfred: ok, butthe date is confirmed: from april 4 to april 8 [14:26] dobey, that means it will be installed in a default installation? [14:26] alecu: yes [14:27] great. [14:27] alecu: it probably will be anyway from other stuff dx team is doing, but we should have it too [14:28] nessita: well that was confirmed more than a week ago [14:28] btw: april 6 to 17 is BAFICI, the buenos aires film festival. [14:31] mandel, btw: you pasted the same branch twice! [14:32] alecu: ooh! [14:34] it's an independent film festival, and it's usually super-overcrowded, but you can go early in the morning and get tickets for the day. [14:35] alecu: awesome, will they have subtitles for the hispanically challenged? :) [14:36] alecu: oh, did I? dammed, let me see [14:37] thisfred, surely! most films are from non-hispanic countries anyway, and they usually have english subtitles in the film itself, (or french sometimes) plus there's a led prompter with spanish subtitles. [14:37] super! [14:37] alecu: this is the other branch: https://code.launchpad.net/~mandel/ubuntuone-client/add_public_files_remote_object_tests [14:37] and I think the prompter shows english for spanish speaking fims, but I have to check. [14:37] mierda, better the merge link [14:37] *films [14:37] mandel, do I have to try those reviews on windows? [14:38] alecu: no, they are just tests with no implementation, TDD for the win [14:38] alecu: tests should be ok on linux, and fail on windows :) [14:39] mandel, you crazy spaniard. Ok, will run those tests on linux anyways :-) [14:47] mandel, in the second branch, around line 69 in the merge proposal: ¿why is "self.shares.syncdaemon_shares.shares" being called twice? [14:48] I guess emit_free_space checks it twice. [14:51] alecu: yes, it will call it twice, onc to check if present, then to use it [14:51] alecu: I wish mocker had a more expressive suyntax for that... [14:52] mandel, I used to like mocker a lot more before. Now I'm not using it so much, and I think the tests are nicer. [14:53] mandel, I specially dislike the error messages it shows. [14:53] mandel, I believe assertions show much more the intent. [14:54] alecu: ping [14:54] nessita, pong [14:54] alecu: can you please follow #ubuntu-desktop? [14:54] alecu: the syntax in some cases is bad, and the error indeed do not show intent,yet is very nice for whenever you need to do alot of basictests [14:55] mocking in general is a last resort. It ties your tests too closely to implementation details for one thing. It [14:55] s very nice to have it though, when all else fails :) [14:58] I like what alecu's been doing with fake implementations of the various backends though, that seems a lot cleaner [15:01] in my case mocking is way easier for assertions etc… Is making sure that one obj calls the other one correctly since everything else is done by twisted :P [15:01] yeah, when you throw async in the mix, often all bets are off again [15:06] well, I found that when using @inlineCallbacks in trial tests, things do look nice. [15:06] trial ftw. [15:11] mandel, approved the first one. Now waiting for the tests to finish to approve the second. [15:11] alecu: question next door, I'll restart my X session now [15:12] mandel, I'll approve it right now anyway, since it's all in platform/windows. [15:13] thisfred, kenvandine is having issues with u1 not registering in the messaging menu [15:14] thisfred, we are not logging any debug info there, right? [15:14] no I don't think so [15:14] strange, I still have 2 entries on maverick [15:14] thisfred, yeah, i am trying to help debug that [15:14] although I think one may be a message actually [15:15] that shouldn't be there [15:15] but can't reproduce it... since it doesn't even try to register with the menu for me [15:15] kenvandine: registering comes from the symlink in /usr/share etc... right? [15:16] no [15:16] that is just adding the launcher [15:16] which i have [15:16] but it has to connect to it when syncdaemon runs [15:17] i am watching the output from indicator-messages-service and it never gets called to add the running process [15:18] let me pinpoint where this *should* happen [15:19] kenvandine: ubuntuone/platform/linux/messaging.py has everything related to the msgingmenu [15:20] kenvandine, it's very likely we are not registering the process (and that is a bug) but only adding items to the menu when new udfs/shared folders arrive. [15:20] kenvandine: trunk version of that http://paste.ubuntu.com/571174/ [15:21] alecu: kenvandine: I *think* we do the right thing when we initialize the Messaging object, but we may not do that on startup [15:22] kenvandine: after a reboot, I'm getting this http://ubuntuone.com/p/eeh/ when syncdaemon is not running, and this http://ubuntuone.com/p/eeg/ after I started it [15:22] kenvandine: and in the second screenshot, when I click on the control panel entry nothing happens [15:23] oh, you don't show the server until there is a message [15:23] you should show the server at startup [15:24] nessita, so you get two entries when syncdaemon is not running? [15:24] yes :-/ [15:26] nessita, do you have anything in ~/.local/share/indicators/messages/applications [15:27] No such file or directory [15:28] kenvandine: I don't even have the indicators dir [15:28] ok [15:28] whoops [15:28] nessita, how about ~/.config/indicators/messages/applications/ [15:29] nessita@dali:~$ ls ~/.config/indicators/messages/applications-blacklist/ [15:29] nessita@dali:~$ [15:29] ok [15:31] kenvandine: any other stuff I can try? [15:31] kenvandine: do you know why when clicking on the control panel entry nothing happens? [15:32] ok, i added self.server.show() after you setup the server and it shows now [15:33] and i am only getting one entry :/ [15:33] nessita, that is part of the problem i think [15:33] it does for me [15:33] but the entry i have is for the server (launcher) [15:34] i think one of your's is for a message [15:34] how can i get it to add a message for me? [15:34] does someone need to share something to me? [15:34] nessita, did you get that? [15:34] pidgin: STOP CRASHING [15:35] * kenvandine PMsit to nessita just in case [15:37] nessita, this is why it isn't doing anything when you click on it [15:37] if server_callback is not None: [15:37] self.server.connect("server-display", server_callback) [15:37] thisfred, alecu: ^ [15:37] kenvandine: I got that (do we know why pidgin crashes so much?) [15:38] nessita, no... i didn't think anyone still used pidgin :) [15:38] -.- [15:38] :-p [15:38] thisfred: did you get that about the server-display [15:39] kenvandine: so the server_callback being None is (this part of) the problem? [15:39] it is why clicking on it, when it is running doesn't open the control panel [15:39] when it isn't running, libindicate handles launching the app the launcher specified [15:39] but once it is running, clicking that uses that callback [15:40] kenvandine: right, I shall set the default to a sensible, uh, default [15:40] so like in the gwibber case, i use dbus to start the client [15:40] since gwibber also has a separate process that is actually using the indicator [15:40] same as u1 [15:40] thisfred, you should also show the server when you create it [15:41] until you do that, it won't get the triangle indicator thing to show it is running [15:41] right now it looks like that doesn't happen until there is a message to display [15:41] which explains why i wasn't getting it, no messages [15:41] nessita, did you share anything? [15:41] kenvandine: I'm still working on that share [15:42] kenvandine, to get a message in that menu, you should either 1) get a new folder shared to you by some other u1 user, and accept it in the mail it's sent to you, or 2) create a new udf in some other device. [15:42] kenvandine: ok, will fix that too [15:42] thisfred, that might be why there are dupes too [15:43] alecu: I'm trying to share a folder but the nautilus U1 plugin won't work [15:43] do you create instantiate multiple Messaging objects? [15:43] and the web UI does not work with FF4 [15:43] dobey: I installed ubuntuone-client-gnome, why I don't get the U1 context menu in nautilus? [15:44] nessita, nautilus -q [15:44] nessita: did you restart nautilus after installing it? [15:44] yes [15:44] nessita, alecu: was there a bug for the dupes/no indicator problem already? [15:44] but I'll retry [15:44] then run u1sdtool -c and see if it shows up after that [15:45] ok, done [15:45] kenvandine: share email should be in your inbox [15:45] thx [15:45] so i have to accept it before i'll see the indicator? [15:45] yes [15:45] * kenvandine would have thought he would get a message in the indicator that someone shared something to me [15:45] and let me accept it [15:45] :) [15:46] kenvandine: that makes sense, but the sharing mechanism is not good, so syncdaemon will not know about the share until you accept it server side [15:46] I just shared a folder with nessita, kenvandine and thisfred. [15:47] ok, so what should happen when i get it in the messaging menu and click on it? [15:47] open it? [15:47] open the control panel [15:47] thisfred: right? ^ [15:48] right, on the shares/udfs tab [15:48] humm [15:48] i got a notify-osd notification of the share after i accepted it [15:48] but not in the messaging menu [15:48] i bet it is all related to how you are creating that server [15:50] essentially there should be one indicate.IndicateServer created for the service, so at startup [15:50] and individual indicate.IndicateIndicator objects created for each message you display [15:51] thisfred, nessita: make sense? [15:51] yes, it does [15:51] yep [15:51] thisfred: can you work on that, please? [15:52] I am [15:52] kenvandine: so, the messaging issue is not related to the control panel itself. I proposed https://code.launchpad.net/~nataliabidart/ubuntu/natty/ubuntuone-control-panel/ubuntuone-control-panel-0.8.5/+merge/50933, would you be able to sponsor it? [15:52] alecu: so, I'll create a Messaging instance at startup, and attach it to something that lives as long as u1-client does, I guess? [15:53] nessita, sure [15:53] kenvandine: thanks! [15:53] thisfred, or move the server setup out of the Messaging class? [15:53] kenvandine: one last question, what signature does the server callback have? no arguments? [15:54] kenvandine: sure, as long as we can be sure it will still be in the same process it should work right? [15:54] the indicator and a timestamp [15:54] yeah [15:54] kenvandine: so same as the message callback [15:54] yup [15:54] cool [15:55] thisfred, this made me remember: rye and dobey suggested reusing the same notification and update-show it (not only update). There's a bug for that #723227 [15:55] alecu: I have that on my radar [15:57] thisfred, I'll have to change the way notifications are being used inside the aggregator too, so let's discuss about it when you get to it. [15:57] nessita, does this version need those theme changes from the design team? [15:57] last i looked, those weren't merged [15:57] alecu: will get back to you when I do [15:58] kenvandine: speaking of that, daniel fore mentioned I should beg for those to be packaged [15:58] hehe [15:58] in particular, I should beg you [15:58] i've been watching it [15:58] :) [15:58] sladen was going to merge it [15:58] * kenvandine checks on that [15:58] kenvandine: feel yourself begged! :-P [15:58] yup [15:58] :) [16:00] kenvandine: would this trace make any sense to you? http://pastebin.ubuntu.com/571198/ (do not bother much, just a quick look) [16:02] dobey: maybe you can see something there? ^ [16:02] i see invalid free [16:02] would need more info to say where [16:03] nessita, not off hand, but let apport report the bug and see what the retracers come up with [16:03] it definitely needs fixing [16:03] kenvandine: apport was not opened, and this is clean natty install. Shall I install something? [16:03] what is liborc? [16:03] *no idea* [16:04] oh [16:04] probably something streamer uses [16:04] err, gstreamer [16:05] a better traceback with debug packages from gdb or valgrind would probably show exactly what the issue is though === beuno is now known as beuno-lunch [16:17] alecu: that's a mean folder to share! [16:17] thanks god that shares are not subscribed by default [16:17] nessita, on a call [16:22] alecu: nessita: fo u1 client, is there a good place to put "run at startup" code? [16:23] lunch time bbiab [16:27] thisfred: what kind of code? [16:27] thisfred: there is the syncdaemon binary and the Main() __init__ method [16:28] i suspect he is talking about the indicator server [16:28] nessita: right, I was wondering what the best place was, it's for showing the messaging menu (the server item) [16:29] thisfred: I would say add a handler for SYS_INIT_DONE (or similar) [16:29] let me confirm the event name [16:30] nessita: thanks. And I can add that in my own code right? So I don't have to modify anything else? Sweet [16:30] thisfred, we should do it in the main aggregator [16:30] thisfred: SYS_INIT_DONE [16:30] alecu: is that one *always* present though? [16:30] thisfred, we already have code that's executed when syncdaemon starts [16:30] alecu: that can work too, but I'm not very familiar with the code [16:30] oh ok, I'll add it there [16:30] thanks guys [16:30] let me find the right place... [16:31] thisfred, StatusFrontend already sets self.messaging = Messaging() [16:32] thisfred, I would suggest doing that initialization inside Messaging.__init__ [16:32] alecu: right, but apparently that's causing problems, so I want to take it out of the __init__ [16:32] oh, didn't realize that. [16:33] thisfred, what problem is it causing? [16:33] apparently the multiple entries in the menu [16:33] thisfred, we have a strange problem with updated couchdb-bin from backports. During my tests 2 instances of couchdb is created on lucid. This is working well though if e.g. evolution-couchdb pre-starts desktopcouch and somehow gets port first and then ubuntuone-preferences just ask desktopcouch about the port which may succeed. [16:33] perhaps multiple Messaging instances are created, not sure [16:34] rye, we may need to put a different desktopcouch in backports that picks the right binary [16:35] though it's strange it starts both [16:35] If anything I would expect it to only start the old one [16:35] thisfred, that sounds strange... only one instance should be created. [16:35] thisfred, let's trace this: [16:35] thisfred, well, it starts two equal instances [16:36] rye: I don't think I have time to work on this right now [16:36] ubuntuone.syncdaemon.main.Main.__init__ calls start_status_listener, that creates the listener and subscribes to it. [16:36] alecu: what seems not to be created in time is the Server instance [16:37] perhaps it is *all* just due to the fact that the server.show() is not called in the right place. [16:38] that's the point where the initialization starts, by creating the StatusListener, the StatusFrontend and the Aggregator. [16:38] perhaps I'll fix that in a branch, and see what problems remain [16:38] nessita, messaging.Server ? [16:38] magical code is magical [16:38] alecu: I think so, yes [16:41] running tests now on fix branch 1 [16:55] ok, lunchtime [16:55] brbs === beuno-lunch is now known as beuno [17:17] jcastro_: ping [17:36] alecu: https://code.launchpad.net/~thisfred/ubuntuone-client/fix-server-show/+merge/50962, but I have a hard time testing it. kenvandine: could (part of) the problem be that the indicator and the executable it opens are in a different package than the one we actually use the messaging from? Even when moving the server.show() up to be called only once and at application start, I don't see the triangle, and still see the duplicat [17:36] e entries. [17:37] thisfred, that worked for me... [17:37] the server-callback that I also attached does work, so something changed [17:38] let's see what happens when this lands in nightlies, perhaps having a different version of the package installed interferes [17:38] thisfred, looking [17:39] alecu, is there a good way to kill *everything* related to u1-client, and even remove the entry from the msging menu? [17:39] uninstalling I suppose, let me try that [17:42] thisfred: and restarting the messaging service [17:43] right, just found that out :) [17:43] ralsina: hey there! can you make a second review for a contributor's branch? https://code.launchpad.net/~chris-ascentsoftware/ubuntuone-control-panel/fixes_715820/+merge/50417 [17:43] then adding everything to PYTHONPATH I gyess [17:43] nessita: sure [17:43] thanks! [17:47] thisfred, looks fine, I'm testing it right now. [17:48] alecu: I know the server callback worked, but it looks like either the Messaging instance is not created at startup, *or* we're still not doing everything right to make it show up as running in the menu. Or something else is wrong :) [17:50] thisfred, it's working for me as intended. [17:50] thisfred, are you still on maverick? [17:50] really? Ah yes [17:50] that may be it [17:50] let me try trunk [17:50] Coool [17:51] thisfred, and trunk does not work, so this branch seems to fix this for me. [17:51] thisfred, I'm approving! [17:51] I had natty somewhat working on friday with a downgraded xorg, only unity crashed after a minute or so [17:52] alecu: awesome, thanks! [17:52] nessita: https://code.launchpad.net/~thisfred/ubuntuone-client/fix-server-show/+merge/50962 fixes the bugs ken reported [17:53] thisfred: awesome, I'll review [17:54] thisfred, I've approved, and now I'm running the unit tests :-) [17:54] haha [17:54] well, at least I ran those ;) [17:54] thisfred, also: make sure to add the bug# to the branch. [17:55] yeah, I asked but got no answer: nessita is there a bug for this? [17:56] can i get a second review on https://code.launchpad.net/~dobey/libubuntuone/print-no-tokens/+merge/50802 please? [17:56] dobey: on it [17:57] dobey: approved [17:57] cheers [17:57] thisfred: I don't think so, can you please open one? ken is busy right now I think [17:57] sure [18:01] * alecu is going to have some lunch and take a walk. [18:02] thisfred: "only unity crashed after a minute or so" <- that used to happen to me while alt-tabbing [18:02] thisfred, but on an ati chipset [18:03] it happens when I open the first application I think [18:03] I had to configure the compiz window switcher to show icons instead of thumbnails, and now it -mostly- works. [18:03] upgrading again, who knows [18:03] ah, perhaps that's a clue [18:04] best of lucks! [18:04] I tried unity --reset, but that never seems to work entirely either [18:04] thisfred: can you please have a single # pylint: disable=W0613 in messaging.py? [18:04] thisfred, i am building your branch, see how it works for me [18:04] nessita: sure [18:04] thisfred: so you can have the disable, the 2 functions, and the enable [18:06] nessita: I like to isolate them as much as possible, so if the need for them disappears it is easier to remove them locally, but it's probably an illusion that anyone will ever look at removing them ;) [18:06] :-) [18:10] thisfred, ok, your branch adds the server fine for me [18:10] nessita: alecu I'm also changing the server callback (but not the messages themselves) to not open the volumes tab, but just the control panel, I think that makes more sense [18:10] but i am still not getting any individual indicators under it [18:10] kenvandine: awesome, thanks for testing [18:10] ah [18:10] thisfred, it does open the control panel though :) [18:10] i accepted a share [18:10] the evil one from alecu [18:10] :) [18:10] didn't get anything [18:11] kenvandine: ah, you don't get one on accepting, only when someone offers [18:11] alecu: can you offer something else to ken? [18:11] oh... [18:11] alecu, something smaller this time :) [18:11] you don;t have to accept it to test the messaging ;) [18:12] thisfred: you do!@ [18:12] you don; t have to subscribe to it ;-) [18:12] eh no [18:12] if you don't accept it, syncdaemon never knows about it [18:12] you just have to see that you get a message and that clicking on it opens the control panel on the volumes tab (and removes the message) [18:13] huh? [18:13] that's just not true [18:13] thisfred: exactly. You don't get the message if you don't accept it [18:13] thisfred: how come is not true? [18:13] I just got a message when alecu sent me a share [18:13] I'm pretty sure [18:13] thisfred: you accept it in the web [18:14] otherwise is impossible (tm) [18:14] hmm, maybe I confused the email with a message then? [18:14] it is confusing to me that you would get a message in the indicator if you already accepted it in an email [18:14] seems redundant [18:14] if that is the case, using the messaging menu for this is pointless [18:14] kenvandine: exactly [18:14] kenvandine: one thing is to accept a share, and another is subscribing to it in your device [18:15] so i guess the messaging menu message is just that there is something you accepted that you might want to subscribe too? [18:15] thisfred, kenvandine: accepting is not the same as subscribing to it [18:15] sure [18:15] i really think you need to find a way to make the messaging menu replace the need for that email [18:15] but I though accepting a share was also part of the messaging menu [18:15] kenvandine: in this cycle, there is no way to do it [18:15] thought [18:15] kenvandine, yes, we would like to tackle that for the next cycle. [18:15] ok [18:16] kenvandine: think about people you want to share something to but they don't have an U! account [18:16] U1* [18:16] as someone that hasn't reviewed the plans for this... it is rather confusing [18:16] so i suspect users will get confused too [18:16] nessita: well the web UI way doesn't need to go away, but for people who do it would be more convenient [18:16] nessita, agreed the email shouldn't go away [18:16] i hate opening a browser for anything :) [18:17] the thing is that subscribing (both to shares and to udfs) is per device... maybe it's not clear, but we have to work on improving it. [18:17] kenvandine, thisfred: right. We have an architectural problem with shares, the share as an object is not created in the server until is accepted. SO syncdaemon will not get any notification until that happens [18:17] and i just noticed the "more" button isn't working in firefox, i guess that is the bug nessita was talking about where it doesn't work in FF4 [18:18] the messaging menu currently only shows new shares and udfs pending subscription. [18:18] nessita: ok, I had everything backwards, as per usual ;) [18:18] kenvandine: yes, it is. FF4 and our web UI are going thru a fase [18:18] ok, so thisfred... after understanding that...i can still say i never got that "subscription" message in the indicator [18:19] kenvandine, so, it's not showing the subscription, and not showing an arrow. [18:19] kenvandine: right, but the thing is, you only get it once, and that would happen right after you accept it [18:19] those two things are surely related. [18:19] it is showing an arrow now [18:19] thisfred, perhaps you can add debugging logs for both [18:19] and clicking on it while running raises the control panel [18:19] which makes it even clunkier [18:20] thisfred, when i accept the share in the web UI, i got a notify-osd notification of the share [18:20] but nothing in the messaging menu [18:21] kenvandine: with this new version? [18:21] hmmm [18:21] yes [18:22] thisfred: good news is that the control panel is opened [18:22] kenvandine: can you please try, with thisfred's branch running, to do u1sdtool -q [18:22] yeah, at least I fixed half the problems [18:22] kenvandine: in my case, the messaging entry is duplicated when SD is not running [18:23] ok, quit and only have one entry [18:23] guh [18:23] I have two :-/ [18:23] :/ [18:24] kenvandine: do you have the new control panel package? the one you sponsored [18:24] nessita, perhaps you've "make installed" the control panel o u1-client? [18:24] *or [18:24] alecu: nopes, and I even uninstalled eveyrthing :-/ [18:24] and you have a dangling .desktop or something similar? [18:24] nessita, i do [18:25] i always only install from packages [18:25] i don't trust uninstalls of stuff i installed from source [18:25] * kenvandine adds some print statements [18:26] I can reproduce every time. When shutting down syncdaemon from thisfred's branch, bang, 2 entries [18:26] kenvandine, anyway, the entries in the messaging menu are put there at the moment that syncdaemon finds out that there are new folders. [18:27] kenvandine, if you shut down syncdaemon, then those entries will not appear again. [18:27] (unless someone shares a different folder, or you create an udf in a different device) [18:29] thisfred, your still creating the server in Messaging [18:29] so does that happen on ever new message? [18:30] no: Messaging is created at syncdaemon start [18:30] ok [18:30] so what adds the indicators? [18:31] Messaging.show_message [18:31] oh, just saw that [18:32] kenvandine: I removed the server.show() from there, does it also need one after all? [18:34] no [18:34] ok [18:37] dobey: https://code.launchpad.net/~nataliabidart/ubuntuone-client/better-debug/+merge/50967 [18:38] nessita, can you share something again? [18:38] kenvandine: sure [18:38] thx [18:38] i added some debugging [18:39] hey folks, I'm trying to get files to synch to the Ubuntu One cloud, having zero success with it...... [18:40] kenvandine: sent [18:41] i'm not even getting an option to synch when I right click on any of my folders [18:41] thx [18:41] Geoffrey2: what system are you running? [18:42] Ubuntu 10.10, Gnome desktop [18:42] Geoffrey2: do you know how to use a terminal? [18:43] Geoffrey2: you can open one going to Accessories -> Terminal. There, type: [18:43] u1sdtool -s [18:43] and paste the output in http://pastebin.ubuntu.com/, and give me the resulting URL please [18:43] weird, i set logging to debug and i am not getting any of my messages [18:43] it is like show_message isn't getting called [18:45] kenvandine: did you accept the share via web? [18:45] yes [18:46] nessita, can you send me another please? [18:46] kenvandine: sure [18:46] thx [18:46] kenvandine: done [18:48] nessita, http://pastebin.ubuntu.com/571292/ [18:49] Geoffrey2: your syncdaemon seems like is just starting [18:49] Geoffrey2: seems like it was not running before you run that command [18:50] Geoffrey2: can you please paste the contents of the log file located at: ~/.cache/ubuntuone/log/syncdaemon.log? [18:52] nessita, no email yet [18:52] now, one or two days ago, someone in #ubuntu tried to help me with this by linking me to a set of instructions by, Joshua Hoover, I believe, on how to totally remove and reinstall Ubuntu One...I didn't notice at first the instructions were from late 2009, and the process never created a new Ubuntu One folder.... [18:52] kenvandine: iI'll re share [18:54] dobey: what is the recommended way to handle noun numbers in i18n? I know single/plural is too simplistic for many languages, but is there a way to have gettext do lookups based on the number so the l10n-ers can figure it out? [18:54] Geoffrey2: not creating the Ubuntu One folder is correct [18:54] or do we just ignore languages like Polish? [18:55] Geoffrey2: can you please show me the contents of the log file I mentioned before? [18:55] this all in relation to bug #723327 [18:55] Launchpad bug 723327 in ubuntuone-client (Ubuntu) (and 1 other project) "Singular form is incorrect in notification (affects: 1) (heat: 6)" [Medium,Confirmed] https://launchpad.net/bugs/723327 [18:55] http://pastebin.ubuntu.com/571294/ [18:55] nessita, http://pastebin.ubuntu.com/571294/ [18:56] Geoffrey2: nice! let me ask for help. verterok, ping [18:56] thisfred: you mean like "monkey" vs. "monkeys" case? or what? [18:57] dobey: exactly [18:57] thisfred: ngettext [18:57] thx, I'll look that up [18:57] gettext.ngettext () [18:57] that would be in python i guess [18:57] awesome [18:58] nessita: whatsup? [18:58] verterok: http://pastebin.ubuntu.com/571294/ [18:58] verterok: isn't syncdaemon supposed to create the Ubuntu One dir itself? [18:58] thisfred: ngettext ("%d monkey", "%d monkeys", num_monkeys) [18:58] right [18:59] or dngettext since we use dgettext [18:59] nessita: yes [18:59] thisfred: and gettext has built-in support for weird plural cases in some languages [18:59] right, that's what I was hoping [18:59] verterok: so how do you explain the trace in that pastebin? [18:59] well yeah, dngettext depending on how/where it's used [18:59] nessita: which version of the client? [19:00] verterok: maverick [19:00] verterok: the user with problems is Geoffrey2, he said he's running Ubuntu 10.10, Gnome desktop [19:00] nessita: the user might have partially removed syncdaemon internal directories? [19:00] verterok: such as? [19:01] nessita: looks like syncdaemon thinks the metadata is in "None" version [19:01] Geoffrey2: what set of instructions did you follow? [19:01] verterok: and what does that mean? [19:02] nessita: probably a rm -Rf or something which left the directories in bad state [19:02] Geoffrey2: let me know when you're here [19:03] nessita: I don't really remember all the different combinations/layouts and the migration paths, but looks like syncdaemon thinks the metadata is in version None/0 [19:04] nessita: need to take a deep look at the code [19:04] verterok: ok, I'll check what set of instructions Geoffrey2 used to reset his account [19:04] nessita: oh, account reset(?) [19:04] sortof [19:04] nessita: probably something was left around that's confusing syncdaemon [19:04] maybe [19:05] im here [19:07] Geoffrey2: what instructions did you follow to reset your account? [19:08] looking for it... [19:11] got it.... http://ubuntuforums.org/showpost.php?p=8146023&postcount=2 === ralsina is now known as ralsina_doctor [19:21] thisfred, ping [19:21] kenvandine: pong [19:21] i think the problem is you aren't keeping a reference to the indicator created [19:21] in __init__, there was a self.indicators = {} [19:21] but never used [19:21] i changed that to a list and tried appending the indicator to it in show_messages [19:21] but now it never seems to be calling my code [19:21] joshuahoover: ping [19:22] verterok: the instructions used were http://ubuntuforums.org/showpost.php?p=8146023&postcount=2 [19:22] verterok: do they sound correct? [19:22] kenvandine: looking [19:22] thisfred, before i added it i had it printing out my debug statements, but now it isn't... not sure why [19:22] thisfred, this is hard code to debug [19:24] kenvandine: ah, but it is, it keeps the reference outside of messaging though: show_message returns an indicator [19:24] nessita: looks correct, I'ld suggest running the cleanup/reset steps again (and pastebin the output of the commands, in case something is failing and leaving cruft behind) [19:24] oh [19:24] kenvandine: and I agree it's hard to debug, but I blame the magic in libindicator. It's fine to blame my code back :) If I can improve something there I'd be glad to [19:24] hehe, not pointing blame [19:25] just having to get shares sent, etc [19:25] Geoffrey2: the instructions look correct, can you please run the cleanup/reset steps again (and pastebin the output of the commands, in case something is failing) [19:25] and my logging output gets randomly eaten [19:25] and i can't see why [19:25] nessita, any flag I need to attach to the commands to get output? [19:25] dobey: changes pushed to better-debug [19:25] Geoffrey2: nopes [19:25] kenvandine: what I do is take one of the tests that mocks the indicator, and then turn the mocking off by commenting out a mocker.patch line [19:26] thisfred, good idea [19:26] ok, from step 1? [19:26] that will make all the tests fail, so I can't forget to undo the commenting either ;) [19:26] :) [19:26] Geoffrey2: just run one by one and then, before running the app itself, show me the outcome [19:27] thisfred, but that is probably hard to do without all the u1 dev tools stuff installed [19:27] kenvandine: you'll also have to put in a pdb or something else to make the code stop when the message is shown [19:27] i am doing this from my running package [19:27] Geoffrey2: yes. quit the Ubuntu One client would be: u1sdtool -q [19:27] Geoffrey2: in a terminal, I mean [19:27] kenvandine: yeah, not sure how much of that will work with vanilla test tools [19:27] yeah [19:27] anyway [19:27] hmm, I may never have actually quit the client the first time.... [19:28] kenvandine: worth trying though, if you have twistedtrial installed [19:28] self.messaging.show_message(share.other_visible_name) [19:28] it isn't doing anything with the indicator that is returned [19:28] kenvandine: I removed the redundant self.indicators [19:29] hmm, now I've gone and confused myself [19:29] so you need to add a reference to it somewhere [19:29] hehe [19:29] you still had self.indicators in messaging.py [19:29] but didn't add anything too it [19:29] thisfred, i think that is the current problem [19:30] yep, looks like it. Possibly someone removed it because it looked redundant [19:30] yeah [19:30] I'll add it back in messaging [19:30] where others fear to tread [19:30] :) [19:30] i tried doing it... but then it never seemed to get called [19:31] can't tell why [19:31] nessita, now, Passwords and Encryption Keys isn't under the applications menu in 10.10, it's under System/Preferences....are the rest of the steps for that the same? [19:31] but i guess that is easier for you to test [19:31] Geoffrey2: I guess so [19:32] kenvandine: there is something seriously weird with trunk, I think some conflict merges went wrong or something like that [19:33] thisfred, ugh... hate it when that happens [19:33] but could explain stuff [19:38] I think it's probably my own stupid mistake though, I don't see anyone else touching this code in the history. Fixing. [19:39] the idea of needing to keep a reference to it is a little foreign when python hacking though [19:39] i think everyone that has integrated anything python in the messaging menu has missed that [19:45] nessita, ok, I'll have to come back tomorrow when I have more free time, but I'm not sure this accomplshed anything...u1sdtool is generating the same error message [19:46] Geoffrey2: ok, come back tomorrow and ping me [19:46] i didn't get any error messages doing those steps except the obvious error that Ubuntu One folder didn't exist...which it didn't, and still doesn't [19:49] Geoffrey2: the Ubuntu One folder should be created automatically [19:51] dobey: did I mention I fix the better-debug branch? [19:51] thisfred, kenvandine: did you find out what the problem is? [19:51] nessita: pong [19:52] nessita: yeah, I think so [19:52] nessita: I think something went wrong with my last branch last week [19:52] I seem to be missing some changes, and I'm pretty sure it got merged, so I must have done something wrong [19:52] thisfred: ah [19:53] nessita: yes, but i am trying to get a banshee feature done before freeze [19:53] thisfred: the exact same thing happened to me with a u1client branch [19:53] dobey: ok, no problem [19:54] nessita: weird and pretty worrying [19:55] thisfred: indeed [19:57] dobey: any idea why this branch (still says merged) is not in trunk? https://code.launchpad.net/~thisfred/ubuntuone-client/lp-702055/+merge/48987 [19:57] dobey: the last entry says u1autopilot failed the tests [19:58] looks like it was set to merged manually, maybe> [19:58] ? [19:59] kenvandine: looks like a branch got lost. I thought I was going crazy [20:01] or am I looking at the wrong trunk perhaps [20:01] thisfred: no, something screwed up. that branch is merged though [20:01] thisfred: what is missing? [20:02] also, that was 2 weeks ago [20:02] maybe I am going nuts [20:02] dobey: nm [20:03] http://bazaar.launchpad.net/~ubuntuone-control-tower/ubuntuone-client/trunk/revision/854 [20:04] dobey: yeah no I think I found the problem. [20:04] means I do need a closure [20:04] oh well [20:04] thisfred: your branch does seems landed [20:04] thisfred: when merging it into trunk I get no changes [20:05] nessita: yeah that's not the problem [20:05] It was an unlikely explanation for a weird phenomenon: I found a likelier one: [20:05] shoot [20:06] Shares are broken, but UDFs work, I bet you and alecu always used UDFs to test, whereas Ken used shares ... [20:06] We do keep a reference for the UDFs message around since we only ever have one udfs message for which we update the count when necessary [20:07] This is what happens when I stop working on something for 2 weeks, I forget everything I knew [20:07] thisfred: indeed, I used UDFs [20:09] nessita, i just logged out and back in... i got two U1 entries in the messaging menu for a few seconds [20:09] until syncdaemon started [20:09] nessita, so you aren't crazy :) [20:09] kenvandine: of course I'm not crazy! :-D [20:09] :-D [20:10] the dupe went away quickly here though [20:11] kenvandine: as soon as SD starts, I thikn [20:11] think* [20:11] yes [20:12] kenvandine: I notice it mor since I start syncdaemon by hand [20:12] I don't have it as a startup app === ivanka is now known as ivanka-train [20:38] nessita, alecu, kenvandine : r891 of https://code.launchpad.net/~thisfred/ubuntuone-client/fix-server-show/+merge/50962 has changed quite a bit and hopefully fixes the issue for shares as well as UDFs. (Ken: I now keep reference to all indicators in a list, and remove them from the callback, initial testing indicates this works) [20:39] thisfred, thx [20:39] i'll test it out in a few [20:39] awesome, thanks for all your help on this [20:41] kenvandine: ask for shares to me, I'm happy to do th,m [20:51] nessita, dobey: are we going to get a u1-client release (natty) for tomorrow? [20:51] verterok: I think dobey was working on that today, let's confirm [20:52] nessita: just wondering if revno >= 888 will be included :) [20:53] verterok: you should confirm with dobey. DO you have something you don't want to be released? [20:53] nessita: nope, I want ot be sure revno 888 is included :) [20:53] dobey: ^ [20:54] dobey: are we getting revno 888 of u1-client in natty for tomorrow release? === ralsina_doctor is now known as ralsina [20:59] is it in trunk? [21:00] i haven't done the release yet, so if it's already in trunk, the answer is 'yes' [21:00] i'm trying to finish up this banshee feature first [21:03] dobey: awesome, thanks! [21:07] sigh, i really do not like git :( [21:12] nessita: AAAARGH [21:12] thisfred: WHAT HAPPENED? [21:12] I just finished the singular/plural thing, only to find alecu just did it :( [21:13] rhi [21:13] thisfred: NO WAY [21:13] thisfred: alecu commented in the bug report [21:13] and I approved the branch he did it on [21:13] haha [21:13] you didn't see that? he even mentioned that in this channel, this morning [21:16] must have been before I showed up [21:16] nessita: anyway, he did it wrong, so my changes are still better [21:16] if a little unnecessary :) [21:16] but a ping would be nice if someone takes a bug away from me :) [21:16] I'm glad to see it wasn't completely wasted though [21:16] :-) [21:16] I'm still proposing it [21:18] nessita: https://code.launchpad.net/~thisfred/ubuntuone-client/fix-lp-723327/+merge/51009 [21:18] and dobey maybe, since he pointed me to the right way to do it. (Yes, that will make you think twice about helping me again ;) [21:28] thisfred: I trade a review, if you are up to https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/fix-buttons/+merge/51012 [21:28] ralsina: ping [21:28] nessita: ponele [21:29] ralsina: today, at the stand up, people was wondering if we should be booking tickets for the BA sprint (in particularly people traveling from the states) [21:29] nessita: I am waiting for marianna's response [21:30] I think tomorrow there will be a green light for booking. [21:30] ralsina: ah, ok. Do we know which hotel we're staying in? (pure curiosity) [21:30] nessita: no, that's one of the things where I need her help [21:30] ah [21:30] Lst sprint it was the ramada inn in olivos, IIRC [21:31] ralsina: yeah, becasue it was really close to facundobatista's and verterok's house [21:31] and mine [21:31] ralsina: maybe in this case we should be near you and/or alecu [21:31] ah [21:31] thisfred: ping [21:31] nessita: it's 15 minutes from mine, but very far for alecu [21:31] nessita: pong [21:32] ah [21:32] ralsina: are you working a bit longer? [21:32] nessita: yes [21:33] nessita: pong pong? [21:33] nessita: and I will do some work late tonight, so if you need reviews or whatever just ask [21:33] thisfred: I'm looking at your branch. Did you removed the translation capabilities from some strings? [21:33] ralsina: yes please! https://code.launchpad.net/~nataliabidart/ubuntuone-control-panel/fix-buttons/+merge/51012 [21:33] I don't think I did [21:33] nessita: If I did that's a mistake [21:33] nessita: I was already reviewing that one :-) [21:33] thisfred: the strings no longer have the Q_() [21:34] thisfred: for exmaple, FINAL_UPLOADED [21:34] ralsina: awesome! [21:34] nessita: they now have N_() right? [21:34] nopes, they now are nothing [21:34] wth [21:34] FINAL_UPLOADED_M = \ [21:34] "%(total_uploaded_files)d files were uploaded to your personal cloud." [21:34] eh, that string should be gone completely [21:34] eh no [21:34] wait [21:34] I am going nuts [21:35] nessita: the thing is [21:35] we need to do the l10n and thus the i18n when we know the number [21:35] thisfred: and I don't think that this will work: [21:35] if upload_done: [21:35] format_args = {'total_uploaded_files': upload_done} [21:35] parts.append(N_(FINAL_UPLOADED_S, FINAL_UPLOADED_M, [21:35] upload_done) % format_args) [21:35] thisfred: since the strings to translate are generated statically [21:35] nessita: since different languages handle plurality very differently [21:36] nessita: that's what ngettext/dngettext is for [21:36] nessita: the only thing I am not sure about, is whether N_ breaks the automatic launchpad extraction [21:37] for that matter I thought it had to be _ always, so I wonder if Q_ works, or if we configured that somewhere [21:38] nessita: but that's the reason the i18n moved into the methods itself [21:39] for instance in polish 1 < n < 10 has a different plural than n > 10 (I think, the exact numbers may be different) [21:45] nessita: hello? [21:47] thisfred: but but but [21:47] hehe [21:49] nessita: I don't see the strings from alecu's branch in launchpad, so I'm guessing the Q_ does not work either [21:49] * thisfred looks into launchpad extraction [21:51] the strings to translate will not be generated in this case === beuno is now known as 84XAAAATU === popey_ is now known as popey === karni_ is now known as karni