/srv/irclogs.ubuntu.com/2012/03/22/#ubuntu-unity.txt

thumperbschaefer: ping01:30
bschaeferthumper, hey01:32
=== jalcine is now known as jalcine_
=== jalcine_ is now known as jalcine
=== jalcine is now known as jalcine_
=== jalcine_ is now known as jalcine
=== jalcine is now known as jalcine_
=== jalcine_ is now known as jalcine
medocHi. I am writing a unity lens/scope for the recoll search engine. I know close to nothing about gtk/gnome/unity and I need a little help08:37
medocMy problem at the moment is that the recoll search (big disk-based index) is too slow for being started every time a character is typed08:38
medocI'd like to time-out a little and only launch an actual search after, maybe a few hundred ms after the user stopped typing.08:39
medocCould someone point me to a way to do this, some sample code, whatever ? Thanks08:39
medocAh and the lens is written in Python (which I don't know well either :) )08:40
=== alan_g is now known as alan_g|afk
=== jalcine_ is now known as Guest61506
=== Guest61506 is now known as JackyAlcine
=== JackyAlcine is now known as jalcine
=== jalcine is now known as Guest81870
=== fenris is now known as Guest95871
MCRDear Unity developers: Please read this and the comments as well ;) : http://www.webupd8.org/2012/03/unity-with-minimize-on-click-patch.html10:07
=== mmrazik is now known as mmrazik|afk
=== mmrazik|afk is now known as mmrazik
mhr3davidcalle, ping?10:51
davidcallemhr3, pong10:51
mhr3davidcalle, is the video lens even supposed to search in the home view?10:51
mhr3the files lens should pick up videos as well, so i'm not really sure...10:51
davidcallemhr3, It's supposed to act like the music lens.10:52
mhr3so no results for empty search, but searches work10:52
davidcalleYeah, and search only for the local videos scope.10:54
mhr3right10:55
davidcalleAnd the files lens picks up Music results as well, with the locate stuff.10:55
mhr3true10:55
davidcalleHopefully, when a Photo lens comes by default, we will be able to have something more clean on this side and propose lenses for almost each data type.10:57
davidcalleOr, right now, the files lens could avoid showing video and audio results in the home dash.10:58
davidcalleBut the old results from the video lens is the priority, I'll take care of it.10:59
=== _salem is now known as salem_
=== MacSlow is now known as MacSlow|lunch
apwdidrocks, i have another one for you: https://bugs.launchpad.net/ubuntu/+source/unity/+bug/96211912:29
ubot5Ubuntu bug 962119 in unity (Ubuntu) "monitor indicators stacked above the (x)(+)(-) buttons but below the title bar of the window" [Undecided,New]12:29
apwdidrocks, i will also note i hit the thing which said "display oriented" which _said_ it would file the bug against compiz instead of unity ... it did not however do that12:29
apw(opr at least thats how i interpreted the text i read)12:29
didrocksom26er: can you look please? ^ I'm really busy right now with the release stuff12:30
om26ersure12:30
om26erdidrocks, looking12:31
didrocksthanks12:31
apwdidrocks, in fact "Thanks for reporting this bug on unity.  Is the issue you are reporting purely graphical (will report more information about your graphic configuration and will report the bug against compiz)?"12:33
apwit does say it'll file the bug against compiz, and doesn't work12:33
didrocksapw: doesn't work? what did you get?12:33
didrocksah not genuine package?12:33
apwdidrocks, no i filed against unity, it asks that dialog text above, i hit YES, and it didn't move it to compiz like it says it will12:34
didrocksapw: it won't file it to compiz, however, it adds all the compiz hooks info12:34
didrocksI guess, I need to check12:34
didrockstoo old :)12:34
apwdidrocks, hmm then the text is wrong i guess .,,12:34
didrockswill have a look12:35
didrocksjust not… now :p12:35
apwno rush12:35
didrocksthanks for spotting it :)12:35
* apw is starting to see stacking issues again ... mostly in menus now12:35
om26eruh i broke my unity12:40
om26erwow dash looks hot now. just wow12:48
om26erapw, seems I can't reproduce the issue  here. do you know when it started happening?12:48
MCRIs smspillaz still part of your team ?12:50
om26erMCR, yes12:50
MCRCOOL, good to hear that !12:50
MCRom26er: Did you break MyUnity or your Unity ?12:51
apwom26er, its not an operation i use normally i only opened it to see if i had the new buttons in there12:51
apwom26er, it is easiest to reproduce by moving to focus follows mouse and moving the cursor in and out of the window.  happens about 1/4 of the time12:52
mhall119MCR: I can see how that can get confusing12:52
MCRHow can I check out Unity from source ? - Is there a wiki somewhere ?12:52
apwdidrocks, om26er, another stacking issue which might be compiz: https://bugs.launchpad.net/ubuntu/+source/unity/+bug/96213712:53
mhall119MCR: bzr branch lp:unity12:53
ubot5Ubuntu bug 962137 in unity (Ubuntu) "chromium bookmark menus are stacked incorrectly" [Undecided,New]12:53
MCRmhall119: thx a lot12:53
mhall119thomi: ping12:53
MCRmhall119: Compiling with cmake, standard ?12:54
mhall119MCR: I'm not the one to ask about that, I'm a Python guy12:55
om26erno test case again :p12:55
MCRmhall119: ok, thx - I will get it to work somehow ;)12:55
MCRmhall119: cmake seems to be it12:56
MCRups, seems I need some more packages :)12:58
mhall119MCR: someone else in here will be able to help you12:58
mhall119they helped me12:58
=== smb` is now known as smb
MCRCould someone tell me how to best install all the build dependencies for unity ?13:03
=== MacSlow|lunch is now known as MacSlow
=== yofel_ is now known as yofel
argesdbarth_, hello13:46
tsdgeosMCR: apt-get builddep unity13:49
tsdgeosbuild-dep actually13:50
dbarth_arges: hi13:55
MCRtsdgeos: thx a lot :)13:55
argesdbarth_, hey I saw that you marked pad.lv/960532 incomplete. this bug pretty much makes my laptop unusable, so I'm highly motivated to help debug this13:56
argesdbarth_, is there a debug guide / wiki / advice on how to get a better test case for this so it can be fixed13:56
MCRcmake .. worked, but sudo make install failed14:00
MCRfatal error: Nux/Nux.h: No such file or directory14:01
MCRany idea what I am missing ^^ ?14:01
dbarth_arges: ok14:13
dbarth_arges: for this kind of bug, there's not really a standard process14:14
dbarth_arges: my suggestion is to do DISPLAY=:0 xwininfo  -root -all > ~/xwininfo-all-stuck.log from a VT14:15
argesdbarth_, ok, any words of wisdom or perhaps tests  you can suggest that might give a better reproducer ... i can pretty much get it in that state within a 10-30m of usage14:15
argesok14:15
argesso do that when i can't click anymore14:16
dbarth_arges: right, when it's locked14:16
dbarth_arges: i've had the bug myself, and this is what i did to document the state14:17
dbarth_arges: also i think you can reproduce it more easily if you do ALT-F10 repeatadly (was trying to debug another issue and that happened to me again)14:17
dbarth_arges: https://bugs.launchpad.net/unity-distro-priority/+bug/94852214:17
ubot5Ubuntu bug 948522 in unity (Ubuntu) "Alt-F10 locks up Unity" [High,Confirmed]14:17
argesok let me fire it up, i'll check this out14:18
argesdbarth_, would unity 2d / unity make a difference? i've seen the lockup with both14:19
argesdbarth_, wow instant lockup hitting alt-f10 repeatably . but its different than the other lockup i've been experiencing14:20
zgregI noticed the alt-f10 freeze, too14:21
dbarth_arges: hmm, you mean it does the same with unity-2d as well?14:46
argesdbarth_, yes14:46
dbarth_arges: that would be very interesting, because a few components could be exonerated14:46
argesdbarth_, i am reproducing now and getting some logs14:46
dbarth_great14:46
elopioping greyback.15:06
elopiodo you have some time for me?15:06
greybackelopio: what can I do for you?15:06
elopiogreyback: it's just that testability can't find my application. I thought you might know how to debug this.15:09
elopiogreyback: http://paste.ubuntu.com/895187/15:09
elopioif you have a precise, it will return Run failed. Failed to launch application. [...] ApplicationNotAvailableError15:09
greybackelopio: try supplyingabsolute path to the execurable15:10
greybacktry supplying the absolute path to the executable (this time looking at what I type:) )15:10
greybackbut if it is in $PATH, that shouldn't matter15:11
elopiogreyback: just the same. The application is started, and I can see it from the visualizer. Still, it says that's not available.15:12
greybackelopio: hmm, have you the log file /logs/testability/qttasserver.log15:13
elopiogreyback: there's nothing on /logs/testability/15:16
elopiooh wait15:17
greybackelopio: you should create that directory so, and try agai15:17
greyback+n15:17
elopiogreyback: http://paste.ubuntu.com/895203/15:18
elopiothat's the right pid.15:19
greybackelopio: you have /usr/lib/x86_64-linux-gnu/qt4/plugins/testability/libtestability.so ?15:20
greyback(that's for 64bit machine)15:20
elopiogreyback: yes, it's there.15:21
greybackelopio: so strange. Could you supply ":sleep_time => 10" to the run function please15:23
greybackjust in case the app hasn't loaded the plugin in time15:23
elopiogreyback: same error, but 10 seconds later.15:24
greybackelopio: so the Visualizer sees the application, and you can inspect elements ok. But this fails15:25
greybackelopio: what code are you testing? I'd like to try this myself15:26
elopiogreyback: yes. I see the application opened by the script.15:26
elopiogreyback: it's just http://paste.ubuntu.com/895187/15:27
elopiofor ubuntuone-control-panel-qt you have to be on precise.15:27
greybackok let me try15:29
elopiothanks greyback.15:31
greybackelopio: yep reproduced anyway15:31
=== FloatingGoat_ is now known as Guest82575
elopiogreyback: greyep. If I'd know what the problem is, I could tell the u1 team to fix it. But I have no idea, and I still can't fully understand the ruby code for the run function.15:38
greybackelopio: I got it15:38
greybackelopio: http://paste.ubuntu.com/895228/15:39
elopiogreyback: awesome!15:39
* elopio trying...15:39
greybackelopio: testability checks the process name is running. But since you're app is in python, it's executable is named "python" - so testability never saw an app called "ubuntu-one.."15:40
greybackwhich confused it15:40
greybackelopio: https://projects.developer.nokia.com/dav/Testabilitydriver/doc/api/qt_linux/#GenericSut:run15:44
greybackelopio: the doc mentions a slight consequence, just fyi15:44
elopionow the problem seems to be my computer.15:45
* elopio smacking...15:45
elopioI'll read it. Thanks a lot greyback.15:46
greybackelopio: yw15:46
zgregI wonder if the workspace preview as it is currently implemented is useful (the small overview you get after ctrl-alt-up/down/left/right is pressed to move to another workspace)15:46
zgregthe live preview is prettyy useless, it's too small, the rendering is full of aliasing, and it needs considerable resources15:47
zgregthe current workspace is only highlighted, but it's too little contrast15:48
zgregs/only//15:48
zgregIMO the static preview of the compiz plugin is much more useful, and looks nicer15:48
zgregwhat do you think about this?15:49
bilal|webWhen is 5.8 coming out?15:49
zgregthe internal version number is already 5.8.0 :)15:53
=== zyga is now known as zyga-food
=== fenris is now known as Guest70667
=== Guest70667 is now known as ejat
=== jalcine_ is now known as Guest10125
=== Guest10125 is now known as JackyAlcine
=== JackyAlcine is now known as Guest91681
=== zyga-food is now known as zyga
thomimhall119: pong19:00
mhall119thomi: hey, was it you that helped me get unity compiled and run locally?19:01
mhall119if so, do yu have a written tutorial for that I can put on unity.u.c?19:01
thomimhall119: I don't think it was me, but I have a rough-and-ready blog post witten about it (although it'd need to be fixed if you're going to re-post it anywhere where people might actually read it)19:02
mhall119thomi: I can clean up/edit whatever it needs19:02
thomimhall119: http://tech-foo.blogspot.co.nz/2012/02/how-to-compile-unity-from-source.html19:03
mhall119I just don't remember all the steps19:03
mhall119ah, perfect, thanks19:03
thomino worries19:03
mhall119thomi: I'll get everybody to review these docs before they're published19:03
thomigood ;)19:04
thomiI'm quite happy writing on my blog where no one reads it. :)19:04
mhall119well, I can change that19:04
zgregapparently nux typically does 40 opengl calls and various small calculations just to draw a single texture quad onto the screen19:29
zgregno wonder performance isn't that great...19:29
thomizgreg: "apparently"?19:30
zgregif I got everything19:30
thomiok, you've measured this yourself?19:30
zgregI've read the source code, and tracked everything down, starting from QRP_ASM_1Tex19:31
zgreghmm the GLSL-based implementation is a little bit better, but not much19:33
zgregthat's not exactly helping either, typically slow machines need the ASM code path19:33
=== salem_ is now known as _salem
bschaeferthomi, ping21:32
thomibschaefer: sup?21:32
bschaeferthomi, hey, so I need to write some new anthy autopilot test for a new commit key21:33
bschaeferthomi, and was wonder if I should just copy paste the one with Ctrl+J and replace it with Enter21:33
thomibschaefer: does the old key still work?21:33
bschaeferyeah21:34
thomibschaefer: then just extend the scenarios on the anthy test case21:34
bschaeferthomi its just that Enter is the "perfered" way to commit text for Japanese/anthy I guess21:34
thomiso for each input string there'll be a scenario for Ctrl+J and a scenario for Enter21:35
bschaeferalright21:35
bschaeferWell I think that is what I did, just made a different function and replaced Ctrl+J with Enter21:36
thomibschaefer: you don't need to make a new function... just add a new scenario parameter21:36
bschaeferalright, was about to say that haha21:36
thomiso inside the one function we'll now do something like:21:36
thomiself.kb.press_and_release(self.commit_key)21:36
thomiand commit_key is a scenario parameter21:37
bschaeferthomi, ok!21:37
bschaeferthomi, also how was your fishing trip?21:37
thomibschaefer: was good, ended up not fishing at all, but went luging, bungy jumping, and clay pidgeon shooting21:37
thomigood stress releif :)21:38
bschaeferthomi, awesome! I keep forgetting your in summer time. I finally saw the sun this week!21:38
thomiheh, well, technically we're in autumn now21:38
bschaeferthomi, yeah, that makes sense as we just hit spring :)21:39
bschaeferthomi, alright I should have some autopilot test for you to look at soon!21:39
thomisweet!21:40
bschaeferthomi, https://code.launchpad.net/~brandontschaefer/unity/ibus-anthy-autopilot/+merge/9893321:52
bschaeferwhen you get a chance, diff still updating21:52
thomiok. running unity in valgrind at the moment21:53
bschaeferalso there is new code in nux that allows for 'Enter' to work, not sure if Jenkins will use the right nux right away21:53
thomiso everything is sloooooow21:53
bschaeferthomi, haha no worries, I keep forgetting how awesome python is. im actually fairly new to it haha21:54
bschaeferfor some reason I didn't think it was an associative array21:54
thomiheh21:55
* bschaefer gets slightly excited over languages 21:55
mhall119thumper: those bzr commit stats I was generating the other day?  I've incorporated them into http://reports.qa.ubuntu.com/reports/unity-stats/#commit_stats21:58
thumpernice21:59
mhall119it'll start showing something the next time there's a commit to one of the development focus branches21:59
mhall119it's grouping unity, unity-2d and nux21:59
mhall119I suppose I should add BAMF to the mix too22:00
thomibschaefer: hmmm,22:01
bschaeferthomi, o noes, did I do something wrong again haha?22:02
thomibschaefer: I found a much better way to do something.22:09
thomimind if I hack your branch around?22:09
bschaeferthomi, nope, go ahead! Should have pushed it to ~unity-teamm22:10
bschaeferbut the changes are small so go ahead :)22:10
thomiit turns out testscenarios provides a simple way to generate a dot-product of one or more parameters22:10
thomiso you can have every combination of param X and param Y22:10
bschaefero nice22:10
bschaeferso we can use Enter on all of them as well as Ctrl+J?22:11
thomiexactly22:18
thomi...but it's not available in the version of testscenarios in porecise :(22:18
bschaeferthomi, :(22:19
thomiI'll copy the code for now22:19
thomiit's pretty simple22:19
bschaeferthomi, yeah, I had to write something like that a while ago for school. permeations are fun haha22:20
thomibschaefer: ok, this is looking a bit harder than I anticipted. I'll work something out and propose a new merge22:26
thomiprobably this afternoon22:27
thomithis is something we need in a bunch of places, so it'll be good to get it in22:27
bschaeferthomi, awesome! Thanks :)22:28
thomibschaefer: ok, as soon I wrote that I figured it out22:31
thomithe test currently launches apps though... I guess we need that nux branch22:31
thumperthomi: scenarios can be a python property22:31
thumperthomi: so it can generate the scenarios when called22:31
thomithumper: yeah, but there's a better way :)22:32
thumperoh?22:32
thomithumper: if you have two axis, generate a dot product of everything from axis 1 with everything from axis 222:32
thomiWe've been doing this manually so far, which requires a lot of finger work22:32
bschaeferthomi, double awesome :)22:32
thomithis way it looks really nice22:33
bschaeferthomi, that nux branch should be merged with nux22:34
bschaeferthomi, did that earlier this morning22:34
thomiok, I guess I just don't have it locally then22:34
bschaeferthomi, also thats what Haskell does with list comprehension (the dot product)22:36
bschaeferthomi, not sure if you have dabbled there22:36
thomibschaefer: thumper check it out: https://code.launchpad.net/~thomir/unity/ibus-anthy-autopilot/+merge/9893922:37
bschaeferthomi, nice!22:40
bschaeferthomi, do you know if jenkins will recompile nux with an abi change?22:42
bschaeferwith out*22:42
thomibschaefer: I don't understand22:43
bschaeferthomi, well if we merge this, and nux isn't recompiled while jenkins is running that test it will fail22:43
thomibschaefer: ahh, yeah. That happens often22:43
bschaeferthomi, alright sweet :)22:43
thomiwe see a little spike on the jenkins graph. I'm not sure we can do much about it at this stage22:44
thomiI gotta go to lunch. be back online later though22:44
bschaeferthomi, cool! Have a good lunch22:44
=== chaoticuk_ is now known as chaoticuk

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!