/srv/irclogs.ubuntu.com/2011/04/21/#ayatana.txt

DBOping humphreybc05:22
humphreybchi DBO05:22
humphreybchow's it?05:22
DBOso I am trying out volley brawl05:22
DBO(finally)05:22
DBOis there any way to up the initial speed? its god awfully slow05:22
humphreybcI think you can hack the code, there are some really simple variables05:23
humphreybcBUT the good news is05:23
DBOit's python?05:23
humphreybcBart and Tommy are pushing out an update next week I think with hopefully an option to increase ball speed05:23
humphreybcIt's lua I think05:23
humphreybcI didn't write it05:23
humphreybcBut you can just find the .love file (opt/volleybrawl or something)05:23
humphreybcand then extract it05:23
DBOif I can modify it I can show you guys how to make it match in unity :)05:24
humphreybcI need to talk to Tommy/Bart but yeah we'd like an update out next week with "official" faster ball speed option, more characters, all that jazz05:24
humphreybcaha!05:24
humphreybcyes, do that05:24
humphreybcyou can email Bart and Tommy - sigmoid@ohso.co05:24
humphreybcI'll be on IRC each day for the next week anyway05:25
DBObasically the wm-class is set to "love-0.7"05:25
DBOyou need to set it to "volleybrawl"05:25
humphreybcah05:26
humphreybcI'll tell bart now so he gets it when he wakes up05:26
humphreybcDBO: Thanks. You should play around with the code, you can do some pretty funny stuff by messing with gravity05:27
DBOstill not sure how to extract a .love file05:27
RAOFWith flowers, surely?05:30
RAOFOr perhaps a nice dinner? :)05:30
DBOyou have to romance it first05:30
DBOoh I see05:32
DBOits just a zip file05:32
DBOoh hmmm05:39
DBOLove may not let them do that05:39
DBOnor does it seem to export any identifying information on the window05:39
DBOhow brutally evil05:40
humphreybclol05:42
humphreybcBart is the lead developer for love too05:42
humphreybcSo.. maybe he can actually fix the engine :P05:42
humphreybcDBO: are you working on that windows-spawning-in-the-dead-zone bug?05:42
DBOwhat bug?05:43
humphreybcclearly not then :D05:43
DBOyeah your easiest fix is to make love export the PID on teh window05:43
DBOand then have your wrapper script use exec05:43
DBO(thus preserving the launch pid)05:43
humphreybckk05:44
humphreybcI need to talk to Bart anyway so I'll mention it to him05:44
DBO_NET_WM_PID is what you are after05:44
humphreybcbrb05:45
DBOwhat a tease05:50
DBOwow06:10
DBOthe register *hates* us06:10
RAOFPeople read that rag?06:13
didrocksgood morning07:25
DBOmorning didrocks07:26
=== yofel_ is now known as yofel
DBOdidrocks, any bamf changes I make now are Oneiric, right?07:29
didrockshey DBO07:30
didrocksDBO: well, if you want to maintain an "Natty" branchc in //, it's fine:)07:31
didrocksDBO: if we still have important things, we can deal in a SRU07:31
DBOokay i want to make sweeping changes07:31
DBOits going to massively improve accuracy07:31
didrocksoneiric then, branch an natty one :)07:31
DBObut I dont know if you will be happy to ship it07:31
didrocksDBO: no sweeping change in natty, in oneiric, we have the time to fix them then ;)07:32
didrocksDBO: that's basically why I added a lot of fixme or TODO: in unity (for instance, removing unused variable07:32
DBOmaybe I can make a couple changes small enough to consider...07:33
didrocksDBO: if you can do that to improve the accuracy, in can be shipped as a 0day SRU07:34
DBOsweeping accuracy improvements on the way!07:35
DBOI dont know why I never thought of this before07:35
didrocksDBO: heh, a new matching way?07:49
DBOyeah07:49
DBOif I can remember how to turn a file full of null characters into a strv07:49
didrocksDBO: you have a bamf merge proposal btw07:57
didrocksin case you didn't see :)07:57
DBOlooking07:57
didrocksnot sure I like the workaround though :/07:57
didrocksDBO: don't you receive the branch merge proposal btw in your inbox?07:58
DBOdidrocks, the one from asac?07:58
didrocksDBO: from chrisccoulson07:58
* DBO checks07:58
DBOi dont see it07:58
didrocksDBO: sorry, it's in unity07:59
didrocksDBO: https://code.launchpad.net/~chrisccoulson/unity/lp767642/+merge/5856907:59
DBOO_o07:59
didrocksDBO: if you want, at UDS, I can show you how to automatically triage merge proposal to be in a special email folder :)08:02
DBOthat would be handy08:02
DBOactually08:02
DBOI get them08:02
DBOin a special folder08:03
DBOwhich I seem to mentally ignore08:03
DBOgo me08:03
didrocksDBO: I'll upgrade you so that you don't ignore them anymore :p08:05
didrocksDBO2.008:05
MacSlowgood morning folks08:06
MacSlowsalut didrocks08:06
didrockshey MacSlow ;)08:06
zniavregood morning08:14
zniavreim still stuck with nvidia 173 (legacy) driver , are they blacklisted ?08:14
didrockszniavre: you can run /usr/lib/nux/unity_support_test -p to see why unity isn't starting08:27
\shhmmm...strange...when I use ctrl+alt+numpad 1/3/etc. the window is moved to the screen edges, but when I move it from the position to another one, the window has not the same size as it was , it's grown automagically08:35
\shmoins btw08:36
zniavre Unity supported:          yes08:39
zniavre!paste08:40
ubot5For posting multi-line texts into the channel, please use http://paste.ubuntu.com | To post !screenshots use http://imagebin.org/?page=add | !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic.08:40
zniavrehttp://paste.ubuntu.com/596828/08:40
zniavrehttps://bugs.launchpad.net/bugs/76761308:41
ubot5Ubuntu bug 767613 in unity (Ubuntu) "unity does not start with nvidia173.14.30" [Undecided,New]08:41
zniavreafter gdm loggin unity can't start it blinks until i came back to gdm to choose gnome-classic08:42
zniavredidrocks,  i know you are french so to be technical, ça clignote ad-vitam sans demarer unity et sans crasher... (sorry for french i do not know to tell it in english)08:46
didrockszniavre: is this a recent regression?08:47
zniavrefrom yesterday and the new nvidia 173.14.30 driver08:47
zniavrewith nouveau it works quite well (at least unity, the rest is a pain due to poor performances of nouveau )08:48
didrocksRAOF: around ? ^^08:48
zniavreto be polite i m not against nouveau but nvidia173 is a bit better at least for my old hardware08:49
didrockszniavre: ok, will track that with RAOF and tseliot08:49
didrockszniavre: agreed, and we only support nvidia right now, nouveau has some mem leaks08:49
zniavrethats true08:49
didrockszniavre: ok, will ping the xorg people in charge of those, it's more a nvidia issue than unity08:50
didrockszniavre: oh for the record, does the ubuntu classic session work,08:50
zniavre<bjsnider> well i think you shuld report a unity bug08:50
didrockszniavre: the one using compiz (with nvidia)08:50
zniavreyes gnome-classic works (with compiz )08:51
didrocksok, but seems something regressed for unity (maybe FBO support)08:51
zniavreall opengl softs works too08:51
didrockszniavre: can you post those info and link your pastebin content as well, please?08:52
zniavrewhere ?08:52
didrockszniavre: in your bug report08:52
zniavreheu ? (sorry french again) il n'y a pas tout deja ? unity test etc...?08:54
didrockszniavre: argh, t'as raison, c'est moi qui l'ai ajouté en plus dans le hook apport. fin de cycle, fatigue… ;)08:55
zniavreje comprend ...08:55
didrockszniavre: j'ai souscris les bonnes personnes et je vais les pinguer aujourd'hui, merci pour l'info :)08:55
zniavreim sorry to report stuf like that...08:56
didrockszniavre: don't be sorry, sorry for the regression :-)08:56
zniavre:o)08:56
zniavrei need to go (baby sitting) but im often around if one of you will  needs more infos08:57
zniavrethank you08:57
didrockszniavre: yeah, I think we will need a stacktrace08:57
didrocksbut let's see with the xorg guys first :)08:58
janimoI started using unity yesterday and I sometimes encounter a strange behaviour on alt-tab. Instead of bringing the new window to the top, the sidebar scrolls in and the icon corresponding to the selected app wiggles09:12
janimopressing alt-tab again makes an other icon start moving as well, and  I can only select the apps by clicking them09:12
janimoI'd be inclined to think this is not a feature but this being Unity I may be so wrong :D09:13
=== daker_ is now known as daker
davidcalleKamstrup, you around?09:49
didrockszniavr: when you are back, can you make compiz crashing please and report a stacktrace?10:02
zniavre didrocks how can i do that please?10:07
didrockszniavre: so, sudo rm /var/crash/*crash (to clean the current crashes)10:08
didrockszniavre: then follow https://wiki.ubuntu.com/Apport#How to enable apport10:08
didrocksinstalling nvidia10:08
didrocksand reboot/try to log in the unity session10:08
didrockszniavre: once it's "blinking" (meaning it crashes but try to respawn)10:09
didrocksyou can logout10:09
didrocksthen, you should have a file /var/crash/*compiz*10:09
didrockszniavre: ubuntu-bug -c /var/crash/*compiz* to report this bug (choose "full version")10:09
zniavreok let me try (already with nvidia driver i need hav to crash compiz so)10:11
zniavredidrocks,  129mo is ok ?10:15
didrockszniavre: seems correct :)10:15
zniavreim sending it10:15
didrockszniavre: ping me back with the bug # then, please10:17
zniavredidrocks,  https://bugs.launchpad.net/ubuntu/+source/nux/+bug/76817810:18
ubot5Error: Launchpad(https://launchpad.net) bug 768178 not found10:18
zniavreho ?10:18
didrockszniavre: can you subscribe me to it? (it's private)10:18
zniavreas 'didrocks' ?10:19
didrockszniavre: right :)10:19
zniavreit looks ok10:19
didrockszniavre: can access to it, thanks!10:19
didrocksretracing to get all needed info is on the way10:20
zniavreim not sure to understand the last sentence but im trusting you    :o)10:21
didrockszniavre: what arch is it?10:21
didrockszniavre: don't mind, it's i386 :)10:22
zniavrdidrocks, yes 32bits10:28
=== API is now known as apinheiro
RAOFdidrocks: A new nvidia binary driver was uploaded recently; I think it's the first 173 driver that we've had which actually works against our X server.11:20
didrocksRAOF: loicm is on it, nux is failing11:20
RAOFHurray for getting a driver a week before release!11:22
zniavrefor maverick it was 4 days before ...   :o)11:22
ogra_RAOF, s/a week before release/on final freeze day/11:25
ogra_:)11:25
RAOFDBO: Oh, I think I've got a lead on the hang-on-vblank kernel problem.  We'll see once these kernels build.  Just a FYI.11:39
=== MacSlow is now known as MacSlow|lunch
=== MacSlow|lunch is now known as MacSlow
=== apinheiro is now known as apinheiro_lunch
sladentedg: can I introduce you to <evaluate> who's just choiced the channel and is the author of clipit, a Clipboard manager13:12
sladentedg: from bug #702316  it appears that there are some problems with the libindicate->notifcation area fallback13:12
ubot5Launchpad bug 702316 in libindicate (Ubuntu) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Undecided,Incomplete] https://launchpad.net/bugs/70231613:12
sladentedg: from running applications under a WM/Desktop Environment with libindicate support13:13
sladenevaluate: I think tedg is on US-time, we may have to wait a couple of hours13:14
sladenevalutate: in the mean-time would be able to put add/update your long explanation from #ubuntu-devel and add it to the bug report itself;  this would stop it getting lost13:14
evaluatesladen, will do so in a couple of minutes.13:40
davidcalleHey stefano-palazzo13:46
stefano-palazzohi davidcalle :)13:54
davidcallestefano-palazzo, https://code.launchpad.net/~davidc3/unity-books-lens/trunk # zeitgeist stuff in trunk : )13:56
stefano-palazzonice! I'll check it out right now13:56
davidcalleI'm pushing right now some comments to the code.13:57
=== m_conley_away is now known as m_conley
evaluatesladen, I looked at the bugreport again and from what I can tell you added all the relevant information already.14:17
evaluateWhat exactly would you want me to add to it?14:18
sladenmorning tedg14:20
sladen< sladen> tedg: can I introduce you to <evaluate> who's just choiced the channel and is the author of clipit, a Clipboard manager14:20
sladen< sladen> tedg: from bug #702316  it appears that there are some problems with the libindicate->notifcation area fallback14:21
ubot5Launchpad bug 702316 in libindicate (Ubuntu) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Undecided,Incomplete] https://launchpad.net/bugs/70231614:21
sladen< sladen> tedg: from running applications under a WM/Desktop Environment with libindicate support14:21
evaluatesladen, tedg, I'm sorry, but I have to go now. Could you please mail me or send me a message through launchpad (if that's possible) if there is any more information that I should provide?14:23
tedgsladen, evaluate, I put instructions on getting a backtrace14:23
evaluatetedg, ok, I will update the bug with that info when I'm back. Thanks for now!14:24
=== apinheiro_lunch is now known as apinheiro
lamalexcdbs, om26er are you guys working on the new queue at all? it really hasn't seemed to be going down. i've been otherwise occupied with some qa stuff and it's at 212 bugs. that's way higher than i should be15:21
cdbslamalex: I did a lot of squashing today15:21
cdbsand yesterday as well15:22
lamalexin the new queue?15:22
cdbslamalex: I'll speed up more tomorrow15:22
cdbslamalex: yes15:22
om26erlamalex, I did not, I just woke and was looking in there15:22
lamalexand you set the statuses?15:22
lamalexlike, the source package and the upstream package?15:22
cdbslamalex: I was focusing more on Incomplete bugs and dups15:22
lamalexok15:22
lamalexi'll try and get the new in order today15:23
lamalexkeep going on the incompletes and dups15:23
lamalexom26er, can you work on the new queue with me?15:23
lamalexbetween the two of us we should be able to knock it down quite a bit before the weekend15:23
cdbslamalex: As for the statuses, I didn't set the status to Incomplete or Confirmed on most of them because15:23
cdbslamalex: I commented on old NEW bugs whether he could reproduce in the latest versions15:24
om26erlamalex, yes defintely15:24
lamalexom26er, cdbs so when you touch a bug, basically it should always leave the new queue15:24
cdbslamalex: and many of them, such as bug #765447 apart from tons others were marked by the reporter themselves to be fixed15:24
lamalexi want that number to always be going down15:24
ubot5Launchpad bug 765447 in unity (Ubuntu) "compiz crashed with SIGSEGV in LauncherIcon::SetQuirk()" [Medium,Fix released] https://launchpad.net/bugs/76544715:24
lamalexso if you ask someone if it's stilll valid, mark it incomplete until they answer15:25
didrocksom26er: hey, btw, do you want today's mismatch report?15:25
cdbslamalex: okay, I'll take more care about that15:25
om26erdidrocks, that too ;)15:25
lamalexthanks15:25
didrocksom26er: here it is :) http://paste.ubuntu.com/596975/15:25
cdbsMy day ends in a few minutes, so more up tomorrow15:25
lamalexdidrocks, we should get your script to just shoot of an email to the unity-bugs team ;)15:25
didrockslamalex: hum, are you following this bugmail ML? I've the feeling that NEW bugs aren't handled15:26
lamalexdidrocks, that's what we're talking about right now15:26
didrocks(especially important ones, that's why I'm scanning this quickly every morning)15:26
didrocksand seb128 too15:26
lamalexom26er and i are going to devastate the new queue backlog that's formed today15:27
lamalexand tomorrow15:27
lamalexwe got it15:27
didrocksnice! :-)15:27
didrockslamalex: where is your hammer?15:27
* lamalex shows didrocks thorr's hammer15:28
didrocksheh :-)15:28
didrocks16:25:03         lamalex | so if you ask someone if it's stilll valid, mark it incomplete until they answer15:29
didrockslamalex: ensure then you are looking at those bugs again, because some people went upset on bugs you set as incomplete and they provided the info. Then no feedback :(15:30
seb128lamalex, oh, and also don't just ask if they still happen or bugs you could easily test on your install, that tend to annoy users as well15:31
cdbslamalex: not me?15:32
cdbs:)15:32
OmegaI think that we could use a 4th window button that when hovered over, makes the menu appear in the window decoration, that way people that want more screen-estate stay happy and those that want a menu in the window do too.15:41
didrockscdbs: I'm certain https://bugs.launchpad.net/bugs/761267 is a dup15:44
ubot5Ubuntu bug 761267 in unity (Ubuntu) "super key shortcuts don't take effect on places tiles until super key is released" [Low,Confirmed]15:44
didrockseven a seb128's bug IIRC :)15:44
cdbsdidrocks: I didn't find a master bug after searching for it15:45
* cdbs searches again, this time with lesser keywords15:45
didrockscdbs: ahah bug #747323 !15:48
ubot5Launchpad bug 747323 in unity (Ubuntu) "should not wait the key unpress to display lenses" [Low,Confirmed] https://launchpad.net/bugs/74732315:48
cdbs:o15:49
* cdbs marks15:49
didrockscdbs: edited the title :)15:49
didrocksbug #74732315:49
didrocksubot5: ?15:49
didrocks"should not wait the key unpress to display lenses / super wait to be released to show dash/places "15:50
* vish tickles ubot5 and bug #74732315:50
ubot5Launchpad bug 747323 in unity (Ubuntu) "should not wait the key unpress to display lenses / super wait to be released to show dash/places" [Low,Confirmed] https://launchpad.net/bugs/74732315:50
didrocksshould be easier to find now :)15:50
didrocksvish: not fair! :p15:50
cdbsdidrocks: I'm also sure bug #766127 is a dupe of something, can't find that bug now :( blame laptop changeover15:50
ubot5Launchpad bug 766127 in unity (Ubuntu) "When text is typed in the Dash Search - "loose focus" with mouse is only semi functional" [Undecided,New] https://launchpad.net/bugs/76612715:50
vish:D15:50
cdbsseb should fix his connection :)15:50
didrockscdbs: oh really? didn't see a dup for that one15:51
cdbsdidrocks: About clicking on the area where the dash formerly existed before resizing.. remember?15:51
cdbsI'll ask the reporter on whether he meant in that area or not15:51
didrockscdbs: oh, you're right!15:52
om26erlamalex, crashers related to nouveau experimental are invalid(if they specifically look live driver issues)?15:55
om26erbug 76732215:55
ubot5Launchpad bug 767322 in nux (Ubuntu) "unity_support_test crashed with SIGSEGV in nouveau_reloc_emit()" [Medium,New] https://launchpad.net/bugs/76732215:55
apwwhat does one file bugs against when the global menu isn't showing up menus for a specific application??15:55
om26erapw, indicator-appmenu might be a good place, I believe15:57
apwom26er, ok thanks15:57
apwbug #76839216:02
ubot5Launchpad bug 768392 in indicator-appmenu (Ubuntu) "global menu does not always have menus for an application" [Undecided,New] https://launchpad.net/bugs/76839216:02
om26erapw, you could be seeing bug 75458116:03
ubot5Launchpad bug 754581 in bamf (Ubuntu) "Some apps don't integrate to appmenu after having their windows closed (bamf side)" [Low,Triaged] https://launchpad.net/bugs/75458116:03
apwis anyone else seeing windows un-minimise the wrong height, such that the input areas are in the right place, but the app is rendered about the height of the menu bar too low16:03
apwsuch that to click on something yopu have to miss high16:03
didrocksapw: we were looking for reproducible testcase of this.16:04
didrockssmspillaz: you remember that I mentionned it? ^^16:05
om26erapw, something like this https://launchpadlibrarian.net/69486382/Capture.png maybe?16:05
apwom26er, yeah exactly like that.16:05
OmegaHmm, the maximize and minimize animations aren't each other's inverse, this kinda bugs me.16:05
OmegaWhat should I file the bug against?16:05
apwdidrocks, for me i seem to occur most often to chromium, when it is unminimised by something other than me16:05
didrocksapw: talk about it to smspillaz once he's around :)16:05
apwubuntu-bug for example, or opening a link in something like xchat16:06
didrocksapw: I got it a lot, but we had a patch in compiz which was supposed to fix it (and it did for me at least). I only saw that once since16:06
* apw suspects we need to be issued with aspestos shorts for uds16:06
didrocksapw: tried that 10 times with chromium, of course, didn't happen :/16:07
om26erdidrocks, did you write unity-preferences?16:07
didrocksom26er: I patched it, njpatel wrote it, why?16:07
apwdidrocks, funny happend opening the link to show the issue!16:07
apwom26er, which bug is that screen shot from16:07
om26erapw bug 76043616:08
ubot5Launchpad bug 760436 in unity (Ubuntu) "Window display shifted" [Medium,Confirmed] https://launchpad.net/bugs/76043616:08
didrocksom26er: don't tell me there is bug in the 50 lines :p16:08
om26erdidrocks, there is a crash ;)16:08
didrocksnoooooooooooooooooooooo16:08
didrocks:-)16:08
didrocksom26er: link? ;)16:08
om26erdidrocks, bug 76714016:09
ubot5Launchpad bug 767140 in unity (Ubuntu) "unity-preferences crashed with SIGABRT in g_object_newv()" [Medium,Confirmed] https://launchpad.net/bugs/76714016:09
smspillazdidrocks: yeah, I've got this one on the radar16:12
smspillazdidrocks: I know how to reproduce it, just not sure why it happens yet16:12
didrockssmspillaz: oh ok, is there a bug milestone for the SRU?16:12
didrockssmspillaz: unity 3.10.12?16:12
didrocks3.8.12*16:12
smspillazdidrocks: don't know16:13
didrockssmspillaz: can you ensure it's the case please?16:13
smspillazok, I'll do that asap16:13
cdbslamalex, om26er: My day ends today, will be back tomorrow, bye!16:15
cdbsmy day begins much before it begins for you both :)16:15
om26ercdbs, bye ;)16:15
Omegabye16:15
lamalexbye cdbs16:15
lamalexgoodnight16:15
lamalexhave a good weekend16:16
om26erdidrocks, what do you think about my question on if the crash caused by experimental nouveau(libgl1-mesa-dri-experiment) should be invalid? bug 76732216:16
ubot5Launchpad bug 767322 in nux (Ubuntu) "unity_support_test crashed with SIGSEGV in nouveau_reloc_emit()" [Medium,New] https://launchpad.net/bugs/76732216:16
cdbslamalex: I'm working on weekends16:16
cdbs:)16:16
cdbslamalex: non-stop since I came back from that trip16:17
didrocksom26er: one sec (just looked at the unity-preferences crash, seems a glib/gvariant crash, but will give a deeper look later)16:18
didrocksom26er: ok, on nouveau, I would say "we don't support nouveau as the testing and developping effort is really limited right now. However, we welcome all contribution to debug and fix this issue. Marking as triaged but prority low"16:19
didrocksom26er: does it sound good to you?16:19
didrocks(basically the same seem than the "no icon over launcher" with nouveau)16:19
om26erdidrocks, wfm, should I add 'mesa' to affects?16:20
lamalexcdbs, oh ok16:20
lamalexwell cool16:20
didrocksom26er: yes please ;)16:20
om26erdidrocks, thx16:21
didrocksyw16:21
stefano-palazzodavidcalle, are you around?16:23
davidcallestefano-palazzo, yes16:23
stefano-palazzodavidcalle, so, I've written a little function that suggest tags you might be interested in, even without a search query (based on running apps)16:23
stefano-palazzoNow, when do I run that function?16:23
stefano-palazzoIs there a callback that runs when you open up the lens?16:24
tedgkenvandine, Is there a way to recover gwibber posts that gwibber didn't actually post?16:24
kenvandineno16:24
tedgkenvandine, Like a log file or something?16:24
stefano-palazzoI saw in your video that results are displayed immediately; in what method do you add this stuff to the model?16:25
tedgkenvandine, :-(16:25
kenvandinetedg, i hope to be able to do that next cycle16:25
kenvandineif i get the time for the refactoring16:25
kenvandine:)16:25
davidcalleIt's the "active" signal (kenvandine uses it too) look at "on_active_change".16:25
* tedg will do no releases to help kenvandine16:25
stefano-palazzodavidcalle, thanks, I'll look into it :)16:25
kenvandinehehe... come to the session at UDS :)16:25
davidcallestefano-palazzo, but it's buggy. I've seen in sent during slow or long searches.16:26
davidcallestefano-palazzo, that's why I add a lot of conditions for it to send the call. Still some bugs though.16:27
didrockskenvandine: about the python lenses not autostarted, where did your debugging stopped?16:28
kenvandineadded some g_debug lines in unity16:32
kenvandineto make sure those assertions are really coming from the sections model16:32
kenvandineand to see when they get called16:32
didrockshum…16:33
didrockskenvandine: so, just to ensure16:33
didrockskenvandine: fresh login16:33
didrocksyou hover the place icon, click on it16:33
didrocksthere is no search result coming, right?16:33
kenvandineright16:33
didrocksif you then, closing it and click on it again16:33
kenvandineno sections populate16:33
didrocksdo they come?16:33
kenvandineand no sections in the quicklist16:33
kenvandineno16:33
kenvandinenever any results16:33
kenvandinebut16:33
didrocksok, so not a race while connecting then16:33
kenvandinelogging from my lens shows that it is calling all the right functions16:34
kenvandineand getting results16:34
didrockshum…16:34
kenvandineit's just unity isn't rendering them16:34
didrocksok, can you try something?16:34
kenvandinesure16:34
didrockslike, fresh login16:34
didrockshover the icon16:34
didrocks(not click on it)16:34
didrockswait for several seconds16:34
didrocksthen, click on it and look if the content is there16:34
kenvandinei did that16:35
kenvandinedidn't work16:35
didrocksok, so not a race at all16:35
kenvandinei think it is a race16:35
didrocksweird that reverting the commit for lazy load worked for you16:35
kenvandineit tries to get the sections16:35
kenvandineand it gets something that isn't a dee model16:35
kenvandineand it never tries again16:35
didrocksthat shouldn't happen if you hover, wait, cick16:35
didrocksclick*16:36
didrocksas basically, the code is equivalent to "connect at startup"16:36
didrocksI don't see the difference with the vala lense either :/16:36
didrocksreally puzzling, we need kamstrup for sure then :(16:36
=== bregma is now known as bregma|lunch
kenvandinethe hover produces that assertion too16:37
kenvandinenot just clicking it16:37
didrockswhy we didn't get that race when connecting at startup then? :/16:37
didrockshum hum, this reminds me my french fix which was refrenched fixed…16:38
didrockshum no in fact16:38
didrockskenvandine: the daemon are started, isn't it?16:38
didrocksdaemons*16:38
kenvandineyes16:38
kenvandinethey all start just fine16:38
didrocksso yeah, not the same…16:38
kenvandinethat's the thing, if the daemon is running16:39
kenvandineand you reload unity16:39
kenvandineit works fine16:39
kenvandinebecause it gets the sections model properly16:39
kenvandineoh... one more note16:39
kenvandineone sec16:39
didrocksyeah, I really don't see the difference with that and "connect later"16:39
kenvandine** (<unknown>:1706): WARNING **: Unable to call GetEntries() on: GDBus.Error:org.freedesktop.DBus.Error.UnknownMethod: No such interface `com.canonical.Unity.Place' on object at path /net/launchpad/unityplacepython16:39
kenvandinethat happens the first time you hove16:39
kenvandinehover16:39
didrocksahah16:39
didrocksso if it can't get the entries16:40
didrocksI think it discares the sections16:40
* popey queues up to speak to kenvandine 16:40
kenvandinei had forgotten i saw that last night16:40
kenvandineyeah, i think so16:40
kenvandinepopey, sure :)16:40
didrocksI think that's the start :)16:40
kenvandinedidrocks, and it never calls GetEntries again for that lens16:40
kenvandineso when that fails, it needs to retry when it is needed again16:40
didrockslet me see the merge request again16:41
kenvandinepopey, what's up?16:41
popeyyou know that scrolling error in gwibber?16:41
popeywhere it flickers up and down the timeline16:42
kenvandinepopey, yeah... :(16:42
popeythat16:42
popey:)16:42
popeyis it something you just don't see?16:42
kenvandinei rarely see it on my laptop16:42
kenvandinethe flicker that is16:42
kenvandinebut i can pretty reliably reproduce it on my netbook16:43
popeywonder why that is16:43
kenvandinewe just can't fix it without a huge overhaul of the UI16:43
popeyoh ok16:43
kenvandinethe rendering16:43
kenvandineor dropping the code that remembers your location16:43
popeyit makes gwibber pretty much unusable for me.16:43
kenvandinewhich that bug had like a thousand dupes16:43
popeyi generally never scroll it anymore16:44
kenvandineit never seems very bad for me, even on the netbook16:44
kenvandinei have to look for it to notice16:44
popeyin dont understand how that can be16:44
popeymine is a dual core 2.8GHz chip with 8GB RAM and a 7900 nvidia graphics card16:44
kenvandinethe suck of it is it has reposition itself with javascript16:44
kenvandineevery time16:44
kenvandineweird16:45
popeyso its not like its a slouch16:45
kenvandinei can usually not even see it on my laptop when i look for it16:45
didrockskenvandine: look at https://code.launchpad.net/~unity-team/unity/lazy-places/+merge/57525, line 328. We try to get the entries there (where we did that later previously)16:45
popeyok. I'll stop nagging you about it kenvandine :) until O16:45
kenvandinepopey, this should completely go away if we can get the new client done :)16:46
didrockskenvandine: i would remove that part16:46
kenvandineyeah, man i hate c++ syntax :)16:47
kenvandinedidrocks, probably better to just try to get the entries again if it fails16:47
kenvandinewhen they are needed that is16:47
kenvandinejust moving it later might not really fix it, just mask it16:48
kenvandinei am assuming the python lenses just take longer to get on the bus16:48
apwis the stacking of the window thingy which appears on pressing super behind everything else a known bug ?16:54
om26erinvisible window?16:55
apwno the one the dash produces on pressing and releasing super (win016:55
apw(win)16:55
apwwith the search box in the top,16:56
apwbah can't get a print screen of it, cause the stuupid window is up16:57
apwi'll assume its just a new bug16:59
Saammhow can i find my wine apps in unity? they are no where :(17:00
om26erSaamm, sam i thought wine apps showned up in search results17:01
seiflotfyhey ted17:01
Saammom26er, oh I have to use search?17:01
om26erSaamm, might also be visible in 'all applications' if I am not mistaken (but they are not categorised, for now :/)17:02
Saammom26er, oh too bad i dont remeber what things i installed :(17:02
davidcalleSaamm, what happens when searching "wine"?17:05
davidcallehmm17:06
seiflotfytedg, i am working on the https://bugs.launchpad.net/bugs/738555 looking good17:06
seiflotfy:)17:06
ubot5Ubuntu bug 738555 in Zeitgeist Framework "zeitgeist-daemon crashed with IOError in _write_to_disk(): [Errno 2] No such file or directory: '/home/royg/.local/share/zeitgeist/datasources.pickle'" [High,Triaged]17:06
tedgseiflotfy, Yeah, why can't you just connect to gnome-session?17:06
=== zyga is now known as zyga-reconfigure
Saamm_davidcalle, it shows wine apps but the problem is there 10-12 app shortcuts for wine and i don remember them...for searching to work you need to know what things you have installed17:06
tedgseiflotfy, It seems like catching the error is only removing the symptom17:06
Saamm_davidcalle, typing wine does not show all installed wine apps but a few17:08
seiflotfytedg, connecting to the gnome-sesstion is not the job of zeitgeist17:09
seiflotfybecause it will make it gnome dependant17:09
seiflotfytedg, also there is no dataloss happening17:09
davidcalleSaamm_, https://bugs.launchpad.net/unity/+bug/75327617:10
ubot5Ubuntu bug 753276 in unity (Ubuntu) "Currently no way to find wine apps in dash other than searching them from search bar" [Undecided,Confirmed]17:10
tedgseiflotfy, Okay, connect to XSMP.  Something that is in the user's session should connect to session manager.17:10
seiflotfytedg, this issue will happen with other applciations too17:10
doctormohey seiflotfy17:11
seiflotfytedg, will it guarantee that home directory is not re-encrypted before the session ends17:11
Saamm_davidcalle, uhm wil it be solved before release?17:11
tedgseiflotfy, Yes, you can inhibit until you're done saving.  I'd use gnome-session as an optional dep (it's dbus, so you can detect it) because the interface is WAY easier.17:12
seiflotfytedg, i can do it as an extension17:12
seiflotfymaking it an optional dependency17:12
seiflotfyor better yet17:12
seiflotfyimplement it in the datahub17:12
seiflotfysince the datahub is a gnome specific implementation17:12
davidcalleSaamm_, I don't think so.17:13
seiflotfytedg, the problem is i think the last unloading is anyhow unnecessary17:13
seiflotfybut let me check again17:13
tedgseiflotfy, It also may make sense to log session closing as an event in the DB.  For instance why I closed all those files.17:13
Saamm_davidcalle, oh :(17:14
seiflotfytedg, tyeah17:14
seiflotfytedg, ok i am on it17:14
seiflotfy:)17:14
seiflotfytedg, can u link me with XSMP docu?17:15
nerochiarosmspillaz: quick x11 question, for when you have a spare minute: is it always safe to call XCompositeRedirectSubwindows on the root window with CompositeRedirectAutomatic ? or are there cases when doing it can cause issues if the WM or other clients already requested redirection ?17:15
tedgseiflotfy, I can, but it may prove cruel :-)  I'd really suggest using gnome-session here....17:15
tedgseiflotfy, http://www.x.org/releases/X11R7.6/doc/libSM/xsmp.html17:16
seiflotfytedg,  the problem i dont think writing an extension will make it into natty17:16
seiflotfybut fixing this over a try catch could make the natty fix17:16
tedgseiflotfy, Eh, well at this point not much will make it into Natty.  I think we're talking SRU anyway.17:17
seiflotfysince really all we are doing is updating when the last time an application sent us a notification17:17
seiflotfytedg, true17:17
seiflotfyok so i am on it17:17
tedgseiflotfy, The catch should probably go in irregardless -- just incase the home directory is full for instance.17:17
seiflotfytedg, ok17:17
seiflotfyso i am writing an extension to log the start and the end of the session17:17
seiflotfy:)17:17
seiflotfyif we do it using XSMP i can make it as part of the engine17:18
seiflotfyusing gnome-session will make it land as an extension17:18
seiflotfyso i need to figure that out17:18
seiflotfytedg, can u open a bug abotu having to log the start and the end of a session17:18
seiflotfy?17:18
tedgseiflotfy, Yeah, just FYI, the GNOME guys *hate* XSMP and want to not support it anymore.  I think that they'll never be able to ditch it, but eh, they might anyway.17:18
tedgseiflotfy, Sure17:19
tedgseiflotfy, bug 76843717:21
ubot5Launchpad bug 768437 in Zeitgeist Framework "Zeitgeist should log the start and end of a session" [Undecided,New] https://launchpad.net/bugs/76843717:21
seiflotfytedg, awesome17:21
seiflotfyso this should make it with our next release17:21
seiflotfytedg, i am working on reducing the memory footprint17:21
tedgseiflotfy, Ah, always a good thing :-)17:22
tedgseiflotfy, Ditching Python? ;-)17:22
seiflotfytedg, no way17:23
seiflotfytedg, problem is ditching python wont make zeitgeist any faster17:23
seiflotfytedg, we spend most of our time in dbus and sqlite17:23
seiflotfy:)17:23
tedgseiflotfy, Would probably reduce memory though.17:24
seiflotfytedg, true17:24
seiflotfybut memory is cheap now adays17:24
seiflotfy:P17:24
seiflotfymaybe if i use couchdb zeitgeist will be faster17:24
tedgProbably not :-)17:24
seiflotfyi was kidding17:26
seiflotfy:P17:26
seiflotfybut i just discovered __slots__ in python17:26
seiflotfywhich is 6x more memory effciant than __dict__17:27
davidcallestefano-palazzo, I really like how you "ps aux" for your lens, this is clever!17:42
stefano-palazzodavidcalle, thanks :)17:42
stefano-palazzoI just pushed it in rev. 4117:42
stefano-palazzoI'm not at all sure it that it'll work perfectly, but I'm going to push it to the PPA still - while that is still acceptable17:43
stefano-palazzoIt won't be in a few days time :)17:43
stefano-palazzodavidcalle, I take it you saw the video?17:43
davidcallestefano-palazzo, nope, just looking at your commit.17:44
=== gabon_ is now known as gabon
stefano-palazzoyou need to get into the Ask Ubuntu chat ;-)17:44
stefano-palazzohttp://www.youtube.com/watch?v=qgWKcVzFuVI ← here it is17:44
davidcallestefano-palazzo, this is really cool17:45
davidcallestefano-palazzo, you should order the tags based on what's taking the most mem: when an app breaks or freezes it ususally does.17:48
stefano-palazzoI'm going to request a build and cross my fingers it doesn't break17:48
davidcalle:)17:48
stefano-palazzodavidcalle, that's a good call, I'll do that!17:49
davidcalleI'm trying to break evolution to see how it appears when broken in ps aux... and I can't! Usually, a click is enough :P17:51
=== bregma|lunch is now known as bregma
=== daker is now known as daker_
=== stefano-palazzo_ is now known as stefano-palazzo
evaluateHello.19:13
evaluatetedg, ping?19:14
evaluatetedg, I have attached the requested debug output to https://bugs.launchpad.net/ubuntu/+source/clipit/+bug/70231619:14
ubot5Ubuntu bug 702316 in libappindicator (Ubuntu) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Undecided,Incomplete]19:14
* tedg click19:15
evaluateIf you have the time to take a quick look at it and just tell me if I've done it correctly I'd be very thankful.19:15
om26ermy workspace switcher just changed to the old icon, bug or bling bling ?19:20
tedgevaluate, Hmm, is there any chance you could grab the libappindicator debug package?19:20
evaluatetedg, sure. How exactly do I do that? Just install the libappindicator-dbg ?19:21
tedgevaluate, I'd be libappindicator1-dbgsym19:21
tedgevaluate, https://wiki.ubuntu.com/DebuggingProgramCrash19:22
zniavrehttps://bugs.launchpad.net/ubuntu/+source/nux/+bug/768178  > thank you for the "workaround" at least i know i hav to wait19:23
ubot5Ubuntu bug 768178 in nux (Ubuntu) "compiz crashed with SIGSEGV in nux::IOpenGLShaderProgram::Begin()" [High,Triaged]19:23
zniavrewhat is SRU please?19:23
om26erzniavre, stable release update19:23
zniavreho i see thank you19:23
evaluatetedg, and after install the debug package, recompile my app and rerun that command?19:25
tedgevaluate, You shouldn't need to recompile19:26
evaluateok19:26
evaluatetedg, http://dl.dropbox.com/u/9596802/indicator-debug2.txt19:28
evaluateThere aren't any major differences between the two from what I can tell though...19:28
tedgevaluate, The big difference is the function name in #519:28
tedgevaluate, You said that this happens after opening the menu a few times?19:30
evaluateIt's not really opening. To reproduce the problem I just double/triple click the systray icon...19:30
tedgevaluate, What happens if you single click?19:31
evaluateI have done it slowly now. First click -> menu pops up, second click -> menu disappears, third click -> freeze19:32
tedgevaluate, Does this happen on Natty?19:33
evaluateFrom what the bug report says, yes. I couldn't test it on natty yet, because I don't have one installed with a DE that doesn't support indicators (if the icon is run in indicator 'mode', the problem doesn't exist, it only appears if the DE doesn't support indicators and thus the fallback gets applied).19:34
tedgevaluate, Okay, so I think this is the result of a patch that we've applied in Natty that I don't think has been backported yet.  It's basically that we're referencing the menu wrong, so the first time it's fine, but then when it gets unref'd on close it gets destroyed.19:35
tedgevaluate, So my thought is that it's fixed in Natty and we're waiting on a backport to Maverick.  But, I'd prefer if we *knew* that.19:36
evaluateok, let me see if I can run awn without indicator in maverick and reproduce that.19:36
evaluates/maverick/natty/19:37
tedgevaluate, Great, thanks.  I'm looking for the bug so we can bitch about the no-backport thing :-)19:37
tedgevaluate, bug 56927319:41
ubot5Launchpad bug 569273 in indicator-application (Ubuntu Lucid) "memory leak in gnome-power-manager" [Medium,Triaged] https://launchpad.net/bugs/56927319:41
tedgkenvandine, So it seems on bug 569273 the gpm patch was backported to Maverick instead of the libappindicator one...19:44
ubot5Launchpad bug 569273 in indicator-application (Ubuntu Lucid) "memory leak in gnome-power-manager" [Medium,Triaged] https://launchpad.net/bugs/56927319:44
tedgkenvandine, Which seems... backwards.19:44
* kenvandine doesn't understand19:44
tedgkenvandine, I'm a bit confused on where to go from here.19:44
evaluatetedg, do you want me to request a backport to maverick/lucid? I'd think it would have a higher change of getting backported fast if the request comes from *you*... :-p19:44
kenvandineevaluate, requests from tedg go to /dev/null19:45
kenvandine:)19:45
tedgkenvandine, mterry found a bug in libappindicator, but worked around it in gpm.  So then the gpm patch was put into Maverick instead of the libappindicator fix.19:45
evaluatekenvandine, lol19:45
tedgkenvandine, So, I think that evaluate's bug is because the libappindicator patch is needed.19:45
kenvandinetedg, you mean only one part of the patch...19:46
tedgkenvandine, Well, no.  The problem is that if libappindicator is fixed I think the gpm patch needs to be reverted...19:46
kenvandineoh19:47
kenvandinebut maverick doesn't have libappindicator does it?19:48
tedgI really don't even know *how* to write that SRU :-)19:48
kenvandineso was the fix uploaded to i-a for maverick?19:48
tedgWell it's a binary package that's built from indicator-application source.19:48
kenvandineright19:49
kenvandinebut was the fix SRU'd for maverick?19:49
evaluatetedg, although, if this is supposedly fixed in natty, I wonder what the bug is about...19:49
kenvandineif not the work around in gpm is the best we have still19:49
tedgkenvandine, The gpm, yes.19:49
evaluateI mean, the bug is submitted against natty+fluxbox...19:49
kenvandinetedg, it seems we just need to have one or the other in maverick19:49
evaluateand it seems to be the same error...19:50
tedgkenvandine, Correct, either one or the other.  But the libappindicator one is better IMO.19:50
kenvandinetedg, worth a SRU?19:50
tedgevaluate, Hmm, so we should make sure they're the same.19:50
kenvandinetedg, if that is the case, we create an SRU for that and also affects gpm package with a note that the patch needs to be reverted19:50
tedgkenvandine, I think that if evaluate's bug is the same, yes.  If not, perhaps not.19:51
* kenvandine thinks tedg is creating to much work19:51
evaluatetedg, sure, I'm downlading the latest natty nightly right now and will test against that19:51
tedgkenvandine, if all the fallbacks for any libappindicator using program are broken :-)19:51
kenvandinetedg, sure :)19:52
evaluateI've also had the same report for lucid (I think that's where the indicator was first introduced). If this is indeed the problem, shouldn't it also be backported to lucid, since that is a LTS?19:54
tedgevaluate, That'd be my thought -- but let's get Maverick first :-)19:56
seiflotfytedg, http://imgur.com/a/huEIc#FnpAa20:09
seiflotfywhat do u think20:09
akshatjseiflotfy, GTK3 please :P20:15
seiflotfyakshatj, yeah20:17
lamalexseiflotfy, can you have it only not log for chrome in private mode or is it all or nothing20:20
evaluatetedg, installing the nightly, will take a couple more minutes.20:21
tedgseiflotfy, Will it go into gnome control center?20:21
seiflotfylamalex, the chrome plugin respect the chrome private mode20:23
seiflotfylamalex, look at all the pictues20:23
seiflotfyu can add applciations not to be logged20:23
seiflotfyand u can go into incognito mode20:23
lamalexcool20:28
seiflotfy:)20:38
evaluatetedg, kenvandine, https://bugs.launchpad.net/ubuntu/+source/clipit/+bug/702316 -- updated with the log output from natty. The problem still exists in natty, so I guess this isn't the patch that tedg was talking about.21:59
ubot5Ubuntu bug 702316 in libappindicator (Ubuntu) "Generic Libindicate fallback support breaks applications (such as clipit) on non-Unity WMs/DEs" [Undecided,Incomplete]21:59
kenvandineevaluate, thx21:59
evaluatekenvandine, sure. I hope you guys can identify the problem...22:01
kenvandineme too :)22:01
kenvandinethx for verifying that22:01
evaluatekenvandine, I have to go to sleep now. If tedg shows up in the meantime, could you please let him know?22:07
=== m_conley is now known as m_conley_away
m4n1shtedg: ping22:31
m4n1shtedg: whenever you have time please, look at this https://bugs.launchpad.net/zeitgeist/+bug/76843722:32
tedgHey m4n1sh what's up?22:32
ubot5Ubuntu bug 768437 in Zeitgeist Framework "Zeitgeist should log the start and end of a session" [Undecided,New]22:32
m4n1shtedg: I saw you reported a bug about zeitgeist22:32
tedgm4n1sh, https://bugs.launchpad.net/zeitgeist/+bug/768437/comments/222:33
m4n1shtedg: thanks. That can be done22:34
m4n1shwrt to file open & close logging, it is already there22:35

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