[02:27] 2/w 26 [03:09] hmm damn mhr3's gone [03:46] bschaefer: could I get you to review this please? https://code.launchpad.net/~thomir/unity/fix-failing-command-lens-test/+merge/102618 [03:46] thomi, yup! [03:47] approved [03:47] that was a very small change === jalcine is now known as JackyAlcine [08:11] didrocks: jenkins broken again? [08:11] tsdgeos: sorry, still adding a new option :) [08:11] tsdgeos: fixing that :) [08:11] ok [08:11] tsdgeos: we need a preprod at some point [08:11] yep :-) [08:12] didrocks: will you reapprove the unity2d merge that failed? [08:12] tsdgeos: sure, no worry, tracking that :) [08:12] cheers [08:13] argh [08:13] if [ ! -z "$pocket" ]; then [08:13] instead of [08:13] if [ -z "$pocket" ]; then [08:13] of course, intend something and do the contrary :p [08:13] happenz [08:24] tsdgeos: ok, tested the 3 different cases, runs fine, putting the production on back, sorry for the interruption (but now we can have SRU build, different ppas, different releases in // ;)) [08:25] didrocks: good stuff :-) [08:26] * thumper hands didrocks a beer [08:26] a real one in a week or so [08:26] * thumper wanders off again [08:26] thumper: \o/ === mpt_ is now known as mpt [09:00] woo \o/ [10:13] kamstrup: I was wondering, do you have lucene boards and flags at home? You're such a big fan! [10:13] totally [10:13] :-) [10:14] hey guys outta curiosity when you install an app why when a an app icon appears does it have an empty > on the left hand side of it? [10:14] kamstrup: thanks for relaying this post btw, it was really interesting :) [10:14] davmor2: it's on purpose to make it noticeable AFAIK [10:14] (it's also outlined) [10:14] didrocks: that's what I was assuming but was making sure it wasn't a bug [10:15] no, it's intended ;) [10:15] didrocks: I think you mean it's a feature right :D [10:16] right ;) === greyback is now known as greyback|lunch === MacSlow is now known as MacSlow|lunch [12:10] mhall119: I missed two packageges yesterday, that is the reason why it wouldn't build yesterday [12:12] mhall119: I checkd out this yesterday: https://myapps.developer.ubuntu.com/dev/ [12:12] must my apps be installed to /opt// [12:12] I cannot use /usr folder? [12:13] malin: for Unity Lenses and Scopes just link to your PPA, the ARB has been handling the packaging for them because it's different than normal apps [12:14] so they wil edit them to install to /opt instead of /usr ? [12:15] I can't see in the wizard how to link diectly to ppa [12:16] malin: just put it in the comments [12:16] I think that's all you need to do [12:17] mhall119: I will try to. But should I test it more before uloading it? I mean, there could be some bugs I haven't noticed yet, etc [12:18] malin: you can upload new versions after it is accepted [12:20] mhall119: I see :) it is not listed as version 1.0, so it means there might be glitches, bugs, etc [12:20] malin: that's fine === _salem is now known as salem_ [12:23] but mhall119 where exactly do I put in the link? All I can see is a way to upload the package from my computer [12:23] malin: join #ubuntu-arb, and ask wender or ajmitch where best to put that information [12:23] wendar, not wender [12:24] thanx [12:25] np === greyback|lunch is now known as greyback === rsalveti` is now known as rsalveti [12:39] :) === MacSlow|lunch is now known as MacSlow === bregma_ is now known as bregma [14:02] hi guys - if I open a tomboy window by clicking icon->search all notes, then maximise another window (or in my case with 2 monitors, maximise 1 app on each monitor). then click tomboy icon and click search all notes. - it doesn't raise the window - bug in tomboy, compiz, or other? [14:02] dropbox-preferences does similar results (except it also wiggles the icon in launcher - tomboy does not) [14:04] i'm told seb128 might be able to advise which app to file the bug under ;) [14:06] oimon, hey [14:07] oimon, I can't reproduce that but I would say compiz [14:07] ok thanks. popey managed to reproduce also [14:08] i'll update packages again to be sure [14:08] when oimon says "tomboy icon" he means "indicator" not the tomboy icon in the launcher [14:08] yeah, sorry [14:11] oimon: i can make the tomboy icon wobble [14:11] oimon: but it doesn't raise [14:13] can someone take a look at this usability bug report I've made? https://bugs.launchpad.net/unity/+bug/985675 thanks [14:13] Ubuntu bug 985675 in unity "Improved panel accordion effect behavior " [Wishlist,New] [14:13] oimon, popey: do you need a fullscreen dialog? [14:13] or it's just [14:13] open the search notes dialog [14:13] focus something else [14:14] use the indicator to select search notes again [14:14] and rather than being raised it just wobble? [14:14] in fact I can reproduce that [14:14] just unfocuss the tomboy dialog and try to open it from the indicator menu [14:14] i get it when I do: 1. click indicator, search for notes. Tomboy window opens. 2) click on a full screen app, tomboy goes to the back. 3) click indicator -> search for notes. Tomboy icon wobbles and blue triangle appears, but does not get raised [14:15] yup [14:15] alt-tabbing raises it, just the indicator doesn't [14:15] it's yet another of the 3 thousand focus stealing prevention bugs open in launchpad [14:16] there are zillion of cases and applications where it happens [14:21] ah ok [14:22] for some reason i wasn't typing the right words to find those in launchpad [14:25] bug #876723 [14:25] Launchpad bug 876723 in compiz (Ubuntu) "tomboy window doesn't move to front on activation" [Low,Incomplete] https://launchpad.net/bugs/876723 [14:26] oimon, bug #627195 [14:26] Launchpad bug 627195 in unity (Ubuntu) "Window management - Apps raised from indicators sometimes dont have the focus" [High,Triaged] https://launchpad.net/bugs/627195 [14:28] ah, very good [14:31] skal teste senere, men nå blir jeg vekke en stund. :) [14:31] hm.. [14:31] maybe use a proper language... [14:31] I will test later, but for now I am leaving for a while :) [14:31] thought I was in #ubuntu-no :p [14:31] thanks for letting me bother you chaps === salem_ is now known as _salem === _salem is now known as salem_ [15:20] seb128: also apps raised from launcher don't get focus for me. Disabling focus prevention helped but now all new windows get focus immediately, even when this is not appropriate === AlanChicken is now known as AlanBell [15:21] rye_, right, that's why we have focus prevention :p [15:23] seb128: compiz should get a brainwaves input module to determine whether to prevent focus stealing [15:23] but that's definitely not going into Precise [16:24] http://dreadknight666.deviantart.com/art/Improved-panel-accordion-effect-behavior-in-Ubuntu-297245940 [16:30] DreadKnight, https://lists.launchpad.net/unity-design/ might be a better media for design discussion than IRC [17:02] seb128, thanks; well not very eager to use forums or mailing lists usually xD [17:02] DreadKnight, and designers tend to not be very eager to use IRC ;-) [17:06] seb128, not even sure atm how to use that mailing list [17:06] guess I have to join the team first [17:10] seb128, how the heck do I reply to something like this? https://lists.launchpad.net/unity-design/msg08985.html [17:11] I'm a member and I'm subscribed to the mailing list; I see no button reply button.... [17:11] anyone? xD [17:12] DreadKnight, usually you reply by email to the message you got, I'm not sure you can reply to messages sent before you subscribed [17:13] seb128, epic fail then [17:13] was getting that idea as well [18:27] mhall119: ping [18:33] Teester: ping me again in about 30 minutes please [18:33] mhall119: Will do. [18:53] Teester: pong [18:55] mhall119: You were wondering about the askubuntu lens the other evening. [18:56] yes [18:56] ooh, i was too ☺ [18:56] jcastro sent me your direction [18:57] The version in my ppa is updated for precise. [18:57] There are no big changes in it apart from it working now. [18:58] awesome, does the ARB know to use your PPA version for precise? [18:59] No. My branch is based off stgraber's ARB version rather than the trunk. [18:59] Not sure if it can be merged with the trunk in that case. [18:59] Teester: do you have commit access to trunk? [19:00] No [19:00] who does? [19:00] Stefano Palazzo [19:01] ok [19:01] I think he also uploaded it to ARB. [19:02] for Oneiric, right? but not precise? [19:02] Yeah, for oneiric. [19:03] can you propose your Precise changes for merging into his branch, or does it depend on stuff from stgraber's branch? [19:04] The trunk doesn't use /opt/, the precise branch does, like stgraber's. === Cimi is now known as Cimi|tired [19:05] I suppose merging with the trunk would just update everything to the branch's version. [19:05] I'll propose a merge. [19:09] Ok. Merge proposed: https://code.launchpad.net/~markjtully/askubuntu-lens/precise/+merge/102745 [19:16] Teester: thanks, I'll try and get jcastro to ping Stefano about it [19:19] mhall119: Cool. [19:19] thanks for your work on this Teester === jalcine is now known as Guest31570 [20:20] Morning folks [20:25] good morning thomi [20:25] * thomi notices that bschaefer works really odd hours [20:25] you're online when I get on in the morning, and you're online when I finish work... do you ever sleep!?? [20:26] haha yeah [20:26] I have tutoring and classes in the middle of the day [20:26] so I split up work, in the morning and at night haha [20:34] bschaefer: here's a question for you: Inside Unity, do we get notified when the user initiates ibus? [20:35] our ibus tests are failing on jenkins, and I'm wondering if it's because we're not waiting for the ibus initiation to complete before start typing [20:35] thomi, it gets notified in nux, through text entry. There are bools you can check through the search bar though [20:35] oooh... [20:35] pango_entry_->im_active() [20:35] I think there is also a preedit bool check now too [20:36] also in IMTextEntry.cpp [20:36] Thanks [20:36] no problem, I should have mentioned that before... [20:36] with the timing problems === salem_ is now known as _salem [21:22] bschaefer: still around? [21:23] thomi, yup [21:23] bschaefer: I notice that when we deactivate ibus in the autopilot tests the pango_entry_->im_active() doesn't get set to False. [21:24] I think this might be causing some of the issues we're seeing in the ibus AP tests. Any ideas? [21:24] hmm that's odd. Let me look at the code really quick [21:24] bschaefer: I'm pushing my branch now so you can see what I'm doing... [21:25] alright cool, im wondering if some signals I changed would effect that [21:25] but I changed those a while ago... [21:26] bschaefer: https://code.launchpad.net/~thomir/unity/improve-ibus-testing/+merge/102767 [21:27] in Searchbar.cpp I add the im_entry property [21:27] alright cool, let me take a look at that [21:27] and then in activate_ibus and deactivate_ibus I assert that the im_active property is in the state I expect [21:28] What happens is that in the very first test, the last assert in deactivate_ibus fails [21:28] i.e.- pressing Ctrl+Space never turns ibus off [21:28] hmm that is odd, let me check if it turns it off when using it manually [21:29] through nux [21:31] * thomi gets more coffee, brb [21:31] thomi, hmm nux is getting the signal to set it to False [21:31] and alright, ill mess around with the ap test [21:31] and see if unity is doing something funny [21:33] bschaefer: Thanks! [21:37] no problem :) === jalcine_ is now known as JackyAlcine [21:45] thomi, hmm that is really odd. [21:45] oh rly? [21:46] because when ctrl+space is hit to deactivate it doesn't get to nux [21:47] only through the ap test though [21:47] so it works "in real life" but not in the autopilot test? [21:47] That is what it looks like, let me write a separate test to see if I can get an idea of whats really going on [21:52] thomi, really weird, I just made this test and nux gets the signal [21:53] http://paste.ubuntu.com/937532/ [21:53] * thomi looks [21:53] bschaefer: what happens if you add the assertions from my branch to the mix? [21:54] maybe the bug is in the text entry class not updating it's member variable? [21:54] you mean the addCleanup? [21:55] because im calling what your asserted [21:55] the activate and deactive ibus [22:01] thomi, ugg and now my ibus test don't feel like running. It just opens the dash then the keyboard indicator flashes [22:02] thomi, ooo what might be happening is how you add to CleanUp [22:02] you add ensure hidden first [22:02] * bschaefer not sure of the order [22:03] but if the dash gets hidden then you hit ctrl+space that could cause problems [22:06] bschaefer: hmmm [22:07] bschaefer: no, it's a stack [22:07] so ibus should be deactivated before the dash is hidden [22:07] * thomi verifies this [22:08] yeah, ugg my ibus test just stoped working, let me restart really quick as things started acting funny [22:11] hmmmm [22:11] now im_active is always True, and I can't turn it off :( [22:13] thomi yeah, with your test you pushed activate_ibus fails to work [22:13] for me at lease [22:13] thomi, could you try to deactive ibus without adding to clean up? After you do a keyboard.type? [22:14] hmmmm [22:15] because when I call the activate/deactive they seem to work on there own [22:15] it's when I call do_dash_engine... function it seems to fail [22:18] thomi, yeah it is stuck on for me too :( [22:18] when im printing im_active, when it is clearly off it is still 1 [22:18] so I wonder if the object isn't getting updated [22:18] ok, so that's indicative of a bug in Unity, right? [22:19] bschaefer: I'd be very surprised if the update issue is in autopilot. More likely there's some edge-case in Unity we're hitting that's casuing that variable to not update [22:20] thomi, yeah, im trying to think of the edge case though. As you just press Ctrl+space when the dash is open [22:20] and Ctrl+space to deactive it [22:20] A timing issue perhaps? [22:21] thomi, wired,if I comment out the asserts it works fine [22:21] and the signals start working again [22:22] the asserts in activate and de [22:22] ...... O.0 [22:23] wait but not when I print it... [22:23] ugg one sec [22:24] AlanBell, ping [22:24] thomi, ooooo I think whats happening is when you change the engine list it reset the im active [22:25] so if you comment out the deactivate ibus [22:25] it is still getting deactivated [22:29] thomi, I think you have found an edge case though [22:29] :( [22:29] I think when the engine gets reset to active, the ime_active isn't getting reset [22:29] so this could be an easy fix in nux [22:30] bschaefer: so I shouldn't bother deactivating ibus? [22:32] I think the problem is in nux, because when it the context gets destroyed it doesn't reset the ime_active [22:33] thomi, well if the problem is in nux, the way you have should work [22:34] bschaefer: so if I comment out the call to deactivate_ibus then everything seems to work [22:35] bschaefer: is the fix in nux trivial? [22:35] thomi, it seems to work, but the asserts should still fail [22:35] since it gets stuck as always true [22:35] bschaefer: well, I commented those out as well... the asserts in Activate_ibus work though [22:35] thomi, Im testing that right now [22:35] I guess because when the dash gets closed ibus is deactivated [22:36] thomi, no, I don't get a signal back from nux when that happens, I think it is this line of code that does it: self.activate_input_engine_or_skip(engine_name) [22:36] which sets the engine each time, which in a sense kills the ibus context [22:37] which ill test that in a sec, want to add some print statments [22:38] hmmmmm [22:38] I guess we should check what the current engine is and only change it if we need to [22:40] thomi, oo confirmed bug in nux :) [22:40] it does create a context each test [22:41] and i didn't reset ime_active, so when it get activated and then it would reset the engine leaving ime_active always true [22:41] ahhhh [22:42] thomi, now with the branch you gave me works perfectly :) [22:42] though destroying the context each time around seems costly for each pinyin text [22:42] it did catch a few crashes and bugs doing it though haha [22:43] bschaefer: I'll fix that in my branch [22:43] you got a MP for the nux branch for me to look at? [22:44] yeah let me remove some print statements and ill push it [22:44] thomi, just want to go through and make sure we don't need to reset anything else [22:44] fair enough :) [22:46] hi andyrock [22:47] AlanBell, hey I've a quick question about unity and accessibility [22:47] when I move the mouse on a launcher icon [22:47] blind folk tend not to use the mouse [22:48] ok so we just need to re-activate the screen reader during the launcher key navigation right? [22:48] touchscreens yes, mouse no [22:49] thomi, ok just pushed it, I also added ResetPreedit on Destroy just to be safe [22:49] alt+f1 puts focus on the launcher and in unity2d it reads it [22:49] unity3d is silent I think, are you working on that? [22:49] https://code.launchpad.net/~brandontschaefer/nux/ibus-active-reset [22:50] * AlanBell boots into unity3d with orca [22:50] thomi, branch above ^ [22:50] bschaefer: you want to propose it for merging? [22:50] AlanBell, I'll work on it [22:51] yeah doing right now! [22:51] * thomi is too impatient [22:51] * thomi gets some breakfast [22:51] andyrock: that would be great. Generally for orca testing you can throw the mouse away, any time you use it you are cheating :) [22:51] AlanBell, it's a high priority bug and we need to fix before the SRG [22:51] err... brunch now I guess [22:51] keyboard all the way [22:51] yeah, launcher and quicklists are silent [22:52] AlanBell, i think the dash is silent too... [22:52] HUD is silent in 3d (the hud options are silent in 2d as well) [22:52] but the panel accessibility is separate, so it should work [22:52] yeah, dash is silent [22:52] HUD has not accessibility support [22:53] thomi, haha sorry, I have to do the problem, the fix and say what test coverage was [22:53] .... [22:53] :) [22:53] im debating if I should put the reset in both Destroy and Create [22:53] or in Create or in just Destroy [22:53] I was pestering gord about HUD accessibility from the moment of release, I thought at one time it was reading the options out in unity2d [22:54] https://code.launchpad.net/~brandontschaefer/nux/ibus-active-reset/+merge/102774 [22:54] thomi, mp done [22:54] global menu is working OK andyrock, but most of the indicators are a bit busted still [22:54] bschaefer: trailing whitespace makes me sad [22:54] thomi, and break feast? isn't it around brunch lunchtime? [22:54] really? it's in ther? [22:55] diff line 18 [22:55] uug, I set vi to only to it to .py files [22:55] * bschaefer fixes [22:55] :) [22:55] bschaefer: yeah, it's too late for breakfast, and I'm having lunch with friends in town, so I can't eat too much either... [22:56] haha nice, and working? [22:56] (I'm sure you wanted that much detail about my eating habits) [22:56] always, I love details [22:56] heh [22:56] I posted some videos of using unity with orca http://www.youtube.com/user/alanbelltolc [22:57] thomi, hmm I think that fixed more whitespace problems [22:58] or it added more... [22:58] AlanBell, is alt+tab silent right= [22:58] ? [22:59] andyrock: yeah, that is broken too [23:00] bschaefer: approved from me, I've asked jay to look at it also [23:00] at least it got fixed for sighted users a couple of days ago [23:00] thomi, sweet, hmm I should talk with him about it a little [23:01] thomi, I almost feel like I should add the reset to both CreateContext and DestroyContext hmm [23:02] thanks though! and nice ap test, good catch haha [23:02] well, that branch should get merged soon [23:02] yeah, it'll be good, as if you destroy it and want to use ibus you have to create [23:03] so it looks good to me :) [23:07] andyrock: this stuff should work (quite a bit doesn't) https://docs.google.com/document/d/1g4K_nZ5zzQKrKExENU0OvfKja0L4UCBgLpFBsemoQF8/edit [23:07] AlanBell, first we need to un-mute unity 3d then we can add new "features" [23:08] AlanBell, btw alt+f1 and super+tab... [23:14] late here, if you have more questions andyrock then Pendulum might be able to help [23:14] really really pleased to see someone helping to make this work again :) [23:15] AlanBell, i'll work on it tomorrow (1:14 AM here and I'm still working :/ ) [23:15] you are an hour later than me then [23:38] bschaefer: could I get you to review this pelase? https://code.launchpad.net/~thomir/unity/improve-ibus-testing/+merge/102767 [23:38] thomi, yup, was just about to run the ibus test one more time to make sure [23:39] thomi, also sometimes the Ctrl+a you press at the beginning of the test is to fast [23:39] thomi, sometimes an 'a' just appears [23:42] ok, I gotta go to lunch... [23:44] bbl [23:48] thomi, alright...I got some failures, not sure if you got any