UbuntuIRC / 2011 /09 /28 /#ayatana.txt
niansa
Initial commit
4aa5fce
[02:09] <Viper550> so I guess the Ubuntu font is too good for the boot screen :P
[02:18] <Viper550> jono, hello
[02:19] <jono> hey Viper550
[02:20] <Viper550> so yeah, you experienced with grub2 and mkfont? Its sorta related to design since I am sorta working on a refreshed boot screen, and well, I'm having issues
[04:29] <grvrulz> hello everybody!
[04:29] <grvrulz> Does unity-2d support integrated window controls in the panel??
[04:32] <grvrulz> anybody?
[04:54] <TheMuso> grvrulz: Yes it does.
[05:08] <grvrulz> TheMuso: but i dont have them
[05:08] <TheMuso> grvrulz: Ok, I can't help you I'm affraid.
[05:09] <grvrulz> TheMuso: np.. i'll try to find out.
[05:46] <didrocks> good morning
[06:43] <smspillaz> does anybody have a copy of matlab OR happen to know a crazy java application which displays this bug: https://bugs.launchpad.net/unity/+bug/857201
[06:43] <ubot5> Ubuntu bug 857201 in unity "Java application windows cut-off/truncated/not displayed properly" [High,Confirmed]
[06:47] <FloatingGoat> fantastic job with unity 11.10 beta 2 guys, addressed most of my UI fears cosmetic and functional.
[07:57] <kamstrup> jcastro: awake?
[08:08] <davidcalle> Hi kamstrup, do you think the PyGI <-> Dee situation will be fixed for the release?
[08:31] <kamstrup> davidcalle: i hope so, but right now it doesn't look too good :-/
[08:35] <davidcalle> kamstrup, ok. Just needed to know if I could get into the process of converting other python lenses to Oneiric and writing a python lens/scope tutorial on dev.ubuntu. Reordering my priorities ;-)
[08:40] <kamstrup> davidcalle: yeah, sorry, i will try and get it fixed, but no promises
[08:48] <davidcalle> kamstrup, no worry, thanks :)
[08:56] <njpatel> Trevinho, (looking at your merge)
[08:57] <njpatel> Trevinho, you shouldn't need to set/unset blending states in the ::Draw you added
[08:57] <njpatel> panel should work like dash and already has a SRC_OVER blending state
[08:57] <kamstrup> gord: can I assign this to you? https://bugs.launchpad.net/unity/+bug/861251
[08:57] <ubot5> Ubuntu bug 861251 in unity "Clicking a filter removes keyboard input focus from search entry" [Undecided,New]
[08:58] <gord> kamstrup, eeer you can, but i don't know how to fix it :) jay would know though
[08:58] <njpatel> gord, kamstrup I think we need to do what we did before, which is forward key events to the search entry if they aren't handled
[08:59] <gord> +1 but again, wouldn't know what we can hook in to nux for there
[09:00] <Trevinho> njpatel: I didn't add these blending values
[09:00] <kamstrup> gord: ok, i'll trust you to s/delegate/pass the monkey on/ then :-)
[09:00] <njpatel> kamstrup, can you assign it to jay please
[09:00] <Trevinho> They were alredy there
[09:00] <kamstrup> gord: how about this one? https://bugs.launchpad.net/unity/+bug/861255
[09:00] <ubot5> Ubuntu bug 861255 in unity "Ratings filter allows ratings > 1.0" [Undecided,New]
[09:00] <Trevinho> but I can move them out
[09:00] <njpatel> Trevinho, PanelIndicatorEntryView::Draw(n
[09:00] <kamstrup> njpatel, gord: ok jay on the first one then
[09:00] <njpatel> that is new
[09:00] <Trevinho> Ah.. ok
[09:00] <Trevinho> yes, sorry
[09:00] <Trevinho> I used for coherency with the rest of the code
[09:00] <gord> kamstrup, *cough* filters should be clamping ;) but yeah i can fix that
[09:00] <Trevinho> so I guess also WindowButtons doesn't need that...
[09:01] <njpatel> Trevinho, it shouldn't :/
[09:01] <njpatel> Trevinho, test panel opacity, if it still works then just leave it to avoid causing unknown issues
[09:01] <Trevinho> Ok, so I move it away from the windowbuttons and panelindicatorentry
[09:01] <njpatel> Trevinho, at the very least, that function is also missing a PaintBackground call
[09:02] <njpatel> otherwise on some systems it won't clear the background before painting
[09:02] <Trevinho> Well, I didn't since it cause to paint the dark background
[09:02] <Trevinho> Mh, I can use the "clear" tip... Used also in the PanelMenuView draw
[09:02] <Trevinho> using a white cololor
[09:03] <njpatel> okay, that should work too. it seems that the panel's drawing is not working right if DrawBackground is causing issues :/
[09:03] <Trevinho> otherwise paintbackground would lead to draw the dark bg, and so no cross-effect..
[09:03] <njpatel> anyway, we cna fix that next cycle
[09:03] <njpatel> okay
[09:03] <Trevinho> So, I'll all these things
[09:03] <Trevinho> I'll ping you when I'm done
[09:03] <njpatel> okay, I'll test it out some more
[09:05] <Trevinho> njpatel: are you ok with the timings?
[09:06] <Trevinho> I would have increased the fade-out value
[09:06] <Trevinho> but JohnLea preferred low values in both the directions
[09:06] <njpatel> Trevinho, yeah, timings are fine in this case
[09:07] <njpatel> we want it to be quick
[09:08] <Trevinho> Ok
[09:15] <Trevinho> njpatel: should the PanelMenuView::Draw set the blending instead?
[09:20] <njpatel> Trevinho, no, the blending is being set by panelview
[09:20] <Trevinho> so I move it out too
[09:20] <Trevinho> it was there...
[09:20] <njpatel> Trevinho, hmm, i'm a bit worried now
[09:20] <njpatel> someone has been changing that stuff, so maybe it's just better to leave it at this point
[09:21] <Trevinho> njpatel: maybe panelmenuview changes it just because it uses the gradient
[09:21] <Trevinho> that needs a different blending
[09:21] <Trevinho> then it resets to default
[09:21] <Trevinho> so I'll keep that
[09:22] <njpatel> Trevinho, well, it's using the same blending, i think keep it everywhere, including your code, for now
[09:22] <njpatel> we can fix it for P
[09:22] <Trevinho> OK
[09:23] <njpatel> sorry, I didn't think it was added back in :(
[09:26] <Trevinho> However njpatel without the SetBlend in my code, the effect doesn't work as expected
[09:26] <Trevinho> in fact, it keep showing the background until the menus aren't hidden. While I want to make semi-transparent only the entries
[09:27] <Trevinho> The only thing I can safely add to WindowButtons and IndicatorEntries, if you want is the code that is also in PanelMenuView under the comment "/* "Clear" out the background */"
[09:29] <njpatel> tremolux, okay, so let's leave it everywhere for now
[09:42] <Trevinho> njpatel, no need to clear that too,otherwise no cross-fade anymore!
[09:45] <kamstrup> didrocks: this bug is quite annoying for users, and I just committed fixes for both lenses. The files lens will get a release, but maybe just distro patch the apps lens? https://bugs.launchpad.net/unity-lens-applications/+bug/827414
[09:45] <ubot5> Ubuntu bug 827414 in unity (Ubuntu Oneiric) "gedit fails to start on first try, but does on the second" [High,Confirmed]
[09:45] <didrocks> kamstrup: yep, doing :)
[09:45] <kamstrup> didrocks: thanks!
[09:47] <didrocks> kamstrup: I don't get the fix?
[09:47] <didrocks> what's this, we don't have dbus activation for apps, isn't it?
[09:48] <kamstrup> didrocks: the lenses are started by dbus activation - and child processes inherit... http://bazaar.launchpad.net/~unity-team/unity-lens-applications/trunk/revision/242
[09:48] <kamstrup> so most apps wouldn't care
[09:48] <didrocks> kamstrup: yeah, I was on this page
[09:48] <kamstrup> but apparently you can start gedit "as a service" :-)
[09:48] <didrocks> oh, but then, ok, it tries to see if it's already started
[09:48] <didrocks> interesting :)
[09:49] <didrocks> kamstrup: backcporting
[09:50] <kamstrup> didrocks: when lp inserts a link automagically to lp:unity-lens-applications on the bug report, because I add --fixes=..., then why doesn't it link directly to the commit, when that info is readily available?
[09:50] <kamstrup> i can't count the times I've wanted that
[09:50] <didrocks> kamstrup: good question, I think there is a launchpad bug somewhere about it :)
[09:51] <kamstrup> probably :-)
[09:52] <kamstrup> didrocks: any high/critical lens related issues I've missed? My #1 prio right now otherwise is to fix the segv in pygobject3.0 with libdee, as that blocks *all*python lenses...
[09:52] <didrocks> kamstrup: enter doesn't activate the first entry, but I guess it's in the unity side
[09:53] <kamstrup> didrocks: i believe that's a unity thing
[09:53] <kamstrup> didrocks: but I can try and see if I stand a chance without spending days on it
[09:54] <didrocks> kamstrup: yeah, I see nothing else bad happening for the lenses themselves, I guess checking the filters can be some help
=== API is now known as apinheiro
[10:44] <andyrock> hi all
[10:44] <andyrock> JohnLea: around?
[10:45] <njpatel> Trevinho, why adding LauncherController to test-panel build? it's breaks make
[10:45] <andyrock> didrocks: i'm working on super + numpad bug
[10:45] <andyrock> it's a regression
[10:46] <didrocks> andyrock: nice! yeah, I fixed it for natty
[10:46] <didrocks> andyrock: I had to refix recently a lot of things for the numpad, compiz changed the way to do the grab
[10:46] <andyrock> do you remember if in Natty it's possible to use shift+super+numpad
[10:46] <didrocks> I guess it's linked
[10:46] <andyrock> ?
[10:46] <andyrock> you know
[10:46] <didrocks> I guess it was possible, I don't have a numpad but people on the bug told it worked after my natty fixes
[10:47] <andyrock> shift cancels numlocks
[10:47] <andyrock> *numlock
[10:47] <didrocks> ah nice ;)
[10:48] <andyrock> if it worked in Natty i've to study how i can avoid the "shift cancels numlock" behavior:)
[10:49] <didrocks> andyrock: I guess so ;) alt least, super + keypad is the first step :)
[10:50] <andyrock> didrocks: oh super + keypad is already fixed in my branch
[10:50] <didrocks> great :)
[10:50] <andyrock> it has been quite simple
[10:50] <andyrock> using boht 1 and KP_1
[10:50] <andyrock> works well
[10:51] <andyrock> but <shift><super>KP_1 doesn't work :/
[10:51] <didrocks> ok, it's just regestering KP_
[10:51] <andyrock> i mean XLookupString returns 0
[10:51] <didrocks> shift is interesting
[10:51] <andyrock> try to google "shift cancels numlock"
[10:52] <andyrock> didrocks: look here: https://code.launchpad.net/~andyrock/unity/fix-850792/+merge/76991
[10:54] <andyrock> The table looks weird try to edit it to make it readable
=== MacSlow is now known as MacSlow|lunch
[11:45] <JohnLea> andyrock; ping
[11:59] <didrocks> JohnLea: did you get my pm? :)
=== API is now known as Guest96864
=== Guest96864 is now known as apinheiro
[12:01] <JohnLea> didrocks; give me 10min, I'll get back to you
=== Amaranthus is now known as Amaranth
[12:25] <apinheiro> didrocks, are you here?
[12:25] <didrocks> apinheiro: yep
[12:26] <apinheiro> I'm trying to prepare a specific a11y unity ppa, with the code not included on the release due feature freeze
[12:26] <apinheiro> my first step was just
[12:26] <apinheiro> make a apt-get source unity
[12:26] <apinheiro> and try to create the package
[12:26] <apinheiro> with dpkg-buildpackage -b -rfakeroot
[12:26] <apinheiro> but I get this error:
[12:26] <apinheiro> dpkg-shlibdeps: error: no dependency information found for /usr/lib/libbamf3.so.0 (used by debian/unity/usr/lib/compiz/libunityshell.so)
[12:27] <apinheiro> at the end
[12:27] <apinheiro> didrocks, do you know if Im doing something wrong here?
[12:32] <didrocks> apinheiro:hum, that's weird, is libbamf3.so.0 a valid symlink?
[12:33] <apinheiro> didrocks, :
[12:33] <apinheiro> devel@atanagildo:/tmp/unity-4.18.0$ ls /usr/lib/libbamf3.so.0 -lash
[12:33] <apinheiro> 0 lrwxrwxrwx 1 root root 17 2011-07-12 12:57 /usr/lib/libbamf3.so.0 -> libbamf3.so.0.0.0
[12:33] <apinheiro> devel@atanagildo:/tmp/unity-4.18.0$ ls /usr/lib/libbamf3.so.0.0.0 -lash
[12:33] <apinheiro> 240K -rwxr-xr-x 1 root root 237K 2011-07-12 12:57 /usr/lib/libbamf3.so.0.0.0
[12:35] <didrocks> apinheiro: seems it's a locally installed libbamf, isn't it?
[12:35] <didrocks> apinheiro: we are using multi-arch on ubuntu and in oneric, libbamf should be installed to /usr/lib/i386-linux-gnu/libbamf3.so.0.0.0
[12:35] <didrocks> what package version of bamf do you have?
[12:36] <apinheiro> didrocks, hmm, odd, my /usr/lib/i386-linux-gnu is empty
[12:36] <apinheiro> ii libbamf3-dev 0.2.104-0ubuntu1 Window matching library - development files
[12:37] <didrocks> apinheiro: are you on oneiric? ;)
[12:38] <apinheiro> didrocks, well, my machine is 64 bits
[12:38] <apinheiro> devel@atanagildo:/tmp/unity-4.18.0$ ls /usr/lib/x86_64-linux-gnu/libbamf3.so
[12:38] <apinheiro> /usr/lib/x86_64-linux-gnu/libbamf3
[12:38] <apinheiro> didrocks, so yes it seems that I have that .so on two different subdirectories
[12:38] <didrocks> apinheiro: you shouldn't have on in /usr/lib/libbamf*
[12:38] <didrocks> apinheiro: maybe a local install? or remaining cruft?
[12:39] <apinheiro> didrocks, yeah probably, as my system started with natty, and then I keep doing upgrades to oneiric
[12:39] <apinheiro> didrocks, I will try to clean that stuff
[12:39] <apinheiro> didrocks, thanks
[12:42] <didrocks> apinheiro: yw! keep me posted
[12:45] <om26er__> smspillaz, minimizing a window leaves it in focus and have to click on the desktop or app behind, known already?
=== om26er__ is now known as om26er
[12:48] <smspillaz> om26er: you mean the one where it stays in focus but is not visible ?
[12:48] <smspillaz> I just pushed a branch to fix that, 3 people have pinged me about it today
[12:49] <Amaranth> smspillaz: oh, there you are
[12:49] <Amaranth> smspillaz: with your FBO work in unity did you ever run in to a problem where the screensaver didn't draw correctly?
[12:49] <Amaranth> I remember you mentioning disabling FBO drawing in certain cases
[12:49] <om26er> smspillaz, yes that
[12:52] <smspillaz> Amaranth: I had to disable it for transformed screens
[12:52] <smspillaz> om26er: yep, fixed it
[12:52] <Amaranth> hmm, screensaver active is not transformed
[12:53] <om26er> MacSlow|lunch, hey! dash scrollbar does not work on the far right edge (a la fitts law) targetted for final?
[12:53] <om26er> smspillaz, nice :)
=== MacSlow|lunch is now known as MacSlow
[12:55] <MacSlow> om26er, no
=== m_conley_away is now known as m_conley
=== bregma_ is now known as bregma
=== Guest54918 is now known as jbache
[14:01] <greyback> didrocks: should not this commit have a diff file: http://bazaar.launchpad.net/~kubuntu-packagers/kubuntu-packaging/qt/revision/235
[14:02] <didrocks> greyback: ohoh
[14:02] * didrocks checks for a missing bzr add :)
[14:02] <didrocks> greyback: fixed! thanks :)
[14:03] <greyback> didrocks: no problem :)
[14:03] <om26er> smspillaz, hey! just a heads up "show desktop" is still not working with latest unity :/
[14:03] <om26er> bug 836325
[14:03] <ubot5> Launchpad bug 836325 in unity (Ubuntu) "Clickable areas of previously active window remains on 'Show Desktop'" [High,Confirmed] https://launchpad.net/bugs/836325
[14:03] <om26er> https://bugs.launchpad.net/unity/+bug/836325/+attachment/2474569/+files/out-8.ogv
[14:04] <smspillaz> damn
[14:04] <smspillaz> ok
[14:04] <smspillaz> its on the list
[14:39] <smspillaz> om26er: ok, just got around to looking at this
[14:39] <smspillaz> om26er: it's working fine here, are there particular windows which cause problems ?
[14:39] <om26er> smspillaz, seems to happen with every
[14:39] <om26er> smspillaz, you use chromium?
[14:39] <smspillaz> chrome and chromium
[14:40] <om26er> smspillaz, and its running?
[14:40] <om26er> if i close chromium then the behavior is a bit different
[14:40] <smspillaz> I've got both open, and I switched from CSD to non CSD, don't see it
[14:40] <smspillaz> I have made tons of changes locally though
[14:41] <om26er> show desktop would show the desktop but the launcher would stay hidden as well as if something is there that is causing it to not show even if i click the desktop
[14:42] * om26er quits xchat to test if its the culprit
[14:43] <smspillaz> ah yeah
[14:43] <smspillaz> ah interesting, I can get it now
[14:44] <om26er> smspillaz, no particular app is responsible it seems
[14:44] <smspillaz> om26er: I think I know what it is
[14:44] <smspillaz> om26er: when you minimize a window then unminimize it, the unminimized window will be broken in show desktop
[14:44] <om26er> i should upgrade to latest compiz maybe one of stacking issue
[14:44] <smspillaz> it's not a stacking bug
[14:45] <om26er> smspillaz, seems your observation is the case
[14:45] <smspillaz> only seems to happen on gnome-terminal for me
[14:45] <smspillaz> anyways, I'll look into it
[14:45] <smspillaz> thanks
[14:46] <om26er> though i disabled 'minimized windows previews' a few days ago and the issue was happening so could be something aside from that
[14:46] <smspillaz> ah yeah
[14:46] <smspillaz> disabling minimized window previews won't turn off the fake show desktop handler
[14:46] <smspillaz> I need to fix that
[14:46] <om26er> smspillaz, alright then, so now you know the cause :)
[14:47] <smspillaz> :)
[14:47] <smspillaz> have you got the bug # for https://launchpadlibrarian.net/81308460/out-8.ogv ?
[14:47] <smspillaz> so I can add it to the 4.20 list ?
[14:47] <om26er> smspillaz, bug 836325
[14:47] <ubot5> Launchpad bug 836325 in unity (Ubuntu) "Clickable areas of previously active window remains on 'Show Desktop'" [High,Confirmed] https://launchpad.net/bugs/836325
[14:48] <om26er> yes on the list
[14:48] <smspillaz> cheers
[14:48] <om26er> or not, just updated the milestone :p
[14:48] <smspillaz> :)
[14:49] <smspillaz> alright, lets fix this
[14:53] <rodrigo_> diwic, hey, thanks for the patch for https://bugzilla.gnome.org/show_bug.cgi?id=657244 , but I'm preparing a (hopefully) better one, so can you replicate easily the crash?
[14:53] <ubot5> Gnome bug 657244 in media-keys "gnome-settings-daemon crashed with SIGSEGV in g_simple_async_result_complete()" [Critical,Needinfo]
[14:54] <diwic> rodrigo_, otp
[14:56] <rodrigo_> diwic, OTP? :)
[14:57] <diwic> rodrigo_, im on the phone
[14:57] <rodrigo_> diwic, ah ok :)
[14:57] <rodrigo_> diwic, ping me when you're done, please
[15:10] <diwic> rodrigo_, ok, phone conversation ended
[15:10] <rodrigo_> diwic, ok, so can you replicate the crash easily?
[15:11] <diwic> rodrigo_, so I don't know if it happens every time, I just logged in, saw it happening and noticed that since 963 other people were affected as well, it looked like it could use some analysis
[15:11] <diwic> rodrigo_, but I can try rebooting once and see if it happens again
[15:12] <rodrigo_> diwic, ok, just wanted to ask you to test my fix, as I haven't had the crash at all myself
[15:14] <diwic> rodrigo_, just rebooted and it didn't happen now, so not reproducable I assume
[15:14] <rodrigo_> yes, seems people are getting it only once when they update :(
[15:14] <diwic> aha
[15:15] <diwic> maybe I can try downgrading
[15:18] <gord> Trevinho, hey hey, having some issues with the menus on the panel today, releasing a click on an indicator causes the menu to go away. any idea whats going on there?
[15:19] <smspillaz> om26er: ok, fixed it (hopefully)
[15:19] <diwic> rodrigo_, nope, downgrading and upgrading did not reproduce the issue
[15:19] <rodrigo_> hmm
[15:19] <om26er> smspillaz, cool, which branch? I might test it ;)
[15:20] <om26er> smspillaz, also if you have a branch for bug 861411 please link it
[15:20] <ubot5> Launchpad bug 861411 in unity (Ubuntu) "minimized window leaves the titile bar words on panel" [High,In progress] https://launchpad.net/bugs/861411
[15:21] <om26er> yaay for the title :p
=== API is now known as Guest22350
=== Guest22350 is now known as apinheiro
[15:34] <rodrigo_> diwic, see the last comment on https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/832603 , if you can try it just to make sure nothing breaks?
[15:34] <ubot5> Ubuntu bug 832603 in gnome-settings-daemon (Ubuntu Oneiric) "gnome-settings-daemon crashed with SIGSEGV in g_simple_async_result_complete()" [Critical,Triaged]
[15:38] <diwic> rodrigo_, I'm looking at the patch - you set priv->bus_cancellable = NULL in _stop, then you check (priv->bus_cancellable != NULL) in on_bus_gotten, as part of precondition to returning early.
[15:38] <diwic> rodrigo_, this looks incorrect.
[15:38] <kenvandine> agateau, have you had Cimi_ look at this gtk patch? i don't see anything that looks like it would break but it is a pretty big refactoring of his patch
[15:38] <rodrigo_> diwic, on_bus_gotten will get called on the g_cancellable_cancel call
[15:39] <diwic> rodrigo_, immediately?
[15:39] <rodrigo_> diwic, so, at that time, it won't be NULL
[15:39] <Cimi_> kenvandine: exactly
[15:39] <rodrigo_> diwic, yes, since it's on the same thread
[15:39] <kenvandine> Cimi_, so you reviewed it?
[15:39] <Cimi_> kenvandine: better to blacklist vlc in overlay scrollbars
[15:39] <agateau> kenvandine: not yet, but I talked with him about it earlier
[15:39] <diwic> rodrigo_, ok. Then I understand.
[15:39] <agateau> Cimi_: this won't help with other apps like hp-systray
[15:40] <Cimi_> agateau: same issues?
[15:40] <agateau> Cimi_: yes: it sometimes load gtk first
[15:40] <diwic> rodrigo_, yeah I could try it, but not until tomorrow because my day is up
[15:40] <rodrigo_> ok
[15:41] <Cimi_> ok
[15:41] <Cimi_> agateau: we need the same for gtk3 then
[15:41] <agateau> Cimi_: it is not yet needed for Qt, but in the long run that makes sense
[15:41] <smspillaz> om26er: sorry I took so long https://code.launchpad.net/~smspillaz/unity/unity.fix_836325/+merge/77353
[15:42] <om26er> smspillaz, wrong branch? seems its for bug 832458
[15:42] <ubot5> Launchpad bug 832458 in compizconfig-settings-manager (Ubuntu) "ccsm crashed with KeyError in compizconfig.Plugin.ApplyStringExtensions (src/compizconfig.c:6780)(): '\xd0\t\x9b\n\x01'" [Medium,Confirmed] https://launchpad.net/bugs/832458
[15:43] <om26er> or the bug number might be wrong
[15:43] <Cimi_> agateau: why are you storing those static variables in gtkv/h scrollbar and scrolled window?
[15:43] <Cimi_> agateau: is it requires?
[15:43] <Cimi_> *REd
[15:44] <Cimi_> agateau: if it loads from gtkwidget.c, they are useless?
[15:44] <smspillaz> https://code.launchpad.net/~smspillaz/unity/unity.fix_836325/+merge/77353
[15:44] <smspillaz> should be it
[15:44] <smspillaz> maybe there's a typo in the bug #
[15:45] <agateau> Cimi_: which static vars? I removed static vars from the file you mentioned and did not add any new one
[15:45] <Cimi_> agateau: in the patch you posted, I can see static vars at the top of each file
[15:46] <Cimi_> https://launchpadlibrarian.net/81313776/100_overlay_scrollbar_loading.patch
[15:46] <agateau> Cimi_: I don't see any static var there
[15:46] <agateau> Cimi_: I see declaration of functions
[15:47] <Cimi_> indeed, not static
[15:47] <Cimi_> yep
[15:47] * Cimi_ tired
[15:47] <agateau> Cimi_: the only static vars are in gtkmain
[15:47] <agateau> Cimi_: they are needed because I didn't take the time to create a .h file for them
[15:48] <Cimi_> agateau: can we split that into a separate file then create it?
[15:48] <Cimi_> kenvandine: can we push that tomorrow?
[15:48] <Cimi_> kenvandine: I'd prefer to put the code into a separate file, will be cleaner
[15:49] <kenvandine> Cimi_, yeah, but early tomorrow ok?
[15:49] <agateau> Cimi_: I think you are quite ballsy to talk about cleaner code when your previous version contained code which was duplicated 3 times
[15:50] * agateau creates the header file anyway
[15:50] <kenvandine> hehe
[15:50] <Cimi_> agateau: I am not saying your code is not cleaner than mine, I am saying that since you had this great idea of sharing the functions, itps the right time to put it into a separate file as I said in #dx
[15:50] <smspillaz> hyperair: ping
[15:50] <smspillaz> om26er: actually
[15:51] <Cimi_> agateau: got what I mean?
[15:51] <smspillaz> om26er: are you able to reproduce https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/684731
[15:51] <hyperair> smspillaz: pong.
[15:51] <ubot5> Ubuntu bug 684731 in unity "Windows that hide themselves when closed don't appear in any "this workspace" switcher" [High,Confirmed]
[15:51] <hyperair> oh heh
[15:51] <Cimi_> agateau: even *more* cleaner
[15:51] <smspillaz> filed in 2010
[15:51] <smspillaz> hmmm
[15:51] <smspillaz> I think this popped up in my email queue somewhere
[15:51] <agateau> Cimi_: I just factored the thing because it was necessary for my fix. I usually don't do cleanup this close to deadlines
[15:51] <smspillaz> and then I saw it
[15:51] <smspillaz> and it's probably been fixed for ages
[15:51] <smspillaz> hyperair: is that still a problem ?
[15:51] <smspillaz> couldn't reproduce it using anything
[15:52] <Cimi_> agateau: we just need to create a ubuntuoverlayscrollbar.c/.h
=== m_conley is now known as m_conley_away
[15:52] <Cimi_> put all the code you wrote in gtkmain.c there
[15:52] <hyperair> smspillaz: i haven't see that in ages.
[15:52] <smspillaz> ok
[15:52] <agateau> Cimi_: yes, and mess with buildsystem, define proper copyright headers and so on
[15:52] <Cimi_> agateau: and add the ubuntuoverlayscrollbar.h headers
[15:52] <smspillaz> next bug!
[15:52] <hyperair> :)
[15:53] <Cimi_> agateau: I didn't know that, so, we can avoid adding headers for the moment
[15:53] <smspillaz> has anyone been able to reproduuuuuuuuce https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/851472
[15:53] <ubot5> Ubuntu bug 851472 in compiz (Ubuntu Oneiric) "compiz crashed with SIGSEGV in _XFreeEventCookies()" [Critical,Triaged]
[15:53] <om26er> smspillaz, me neither
[15:53] <Cimi_> agateau: I understood you didn't create the new file for time reasons, I didn't think at the required makefile and copyright changes
[15:54] <smspillaz> ok comment $15
[15:54] <agateau> Cimi_: ok
[15:54] <kenvandine> agateau, Cimi_: so stick with this version?
[15:54] <Cimi_> kenvandine: I am fine with this for oneiric
[15:54] <agateau> Cimi_: note that I agree with you it should be done, just now is not the best time
[15:54] <agateau> Cimi_: thanks
[15:54] <Cimi_> agateau: +1
[15:54] <om26er> smspillaz, try quitting apps from their quicklist
[15:55] <Cimi_> kenvandine: users won't notice, and it is stable
[15:55] <om26er> causes the crash here like 100 times a day
[15:55] <kenvandine> Cimi_, ok, thx
[15:55] <Cimi_> kenvandine: it's a cosmetic difference for coders
[15:55] <Cimi_> kenvandine: but as agateau said implies changes to the makefile and proper copyright assignment, better to postpone for P
[15:56] <Cimi_> I understimated that
[15:56] <Cimi_> but agateau was right
[15:57] <om26er> smspillaz, maximize chromium, open two tab in it, minimize it. right click on its icon in the launcher and select quit
[15:57] * kenvandine builds
[15:58] * hyperair wonders if the issue with the minimize animation in unity and multihead where the launcher ends up in the middle and windows end up minimizing to the non-existent launcher on the left is fixed
[15:58] * om26er crashed unity 7 times in last 5mins with a similar procedure :p
[15:59] <hyperair> D=
[15:59] * smspillaz paintstakingly does this in valgrind
[15:59] <hyperair> smspillaz: it's not pain until you run unity for ~24 hours in valgrind.
[16:00] <smspillaz> ah frick, I need to reset my settings
[16:00] <smspillaz> hang on
[16:04] <smspillaz> aaaaand valgrind goes NUTS
[16:06] <smspillaz> wow impressive
[16:06] <smspillaz> we actually manage to trash the entire heap
[16:06] <njpatel> Trevinho, ?
[16:09] <smspillaz> my gosh, if the fix for this is a 2 line null check, I am going to ...
[16:23] <kenvandine> Cimi_, agateau: the new patch does remove 3 symbols though
[16:28] <smspillaz> om26er: hyperair: https://code.launchpad.net/~smspillaz/unity/unity.fix_851472/+merge/77362
[16:29] * hyperair clicks
[16:29] <smspillaz> turns out that if an X call fails (eg a window was destroyed before you did it and the call is a synchronous one) the return values of anything inside that call will be to memory internal to X
[16:29] <hyperair> O_o
[16:29] <hyperair> to memory internal to X?
[16:29] <smspillaz> so if you don't check the return code and then free the memory (since it appears to be valid) it only, you know, frees the entire heap, since xlib is loaded before compiz
[16:29] <smspillaz> oh, sorry, internal to xlib
[16:30] <hyperair> oh heh
[16:30] <smspillaz> wouldn't be possible to free memory on the server
[16:30] <smspillaz> that would be a SIGSEGV ;-)
[16:30] <smspillaz> hyperair: I don't think xlib does this deliberately, but that's sort of the way it works out on compilation anyways
[16:31] <hyperair> hm
[16:31] * hyperair pretends to understand that
[16:32] <smspillaz> hyperair: pointers are usually predictable
[16:32] <smspillaz> depending on the way you compile stuff
[16:32] <smspillaz> because of the way OS caching works
[16:32] <smspillaz> it's why we need ASLR on linux really
[16:33] <hyperair> oh yeah that
[16:33] <hyperair> right
[16:33] <hyperair> well at least stuff allocated on the stack always ends up having the same address =\
[16:33] <hyperair> for some reason or other
[16:49] <apinheiro> njpatel, I have been testing the bug related with the keynav navigation
[16:49] <apinheiro> as it seems to be solved
[16:49] <apinheiro> but all what I said here:
[16:49] <apinheiro> https://bugs.launchpad.net/unity/+bug/852506/comments/2
[16:49] <ubot5> Ubuntu bug 852506 in unity (Ubuntu) "dash does not scroll down during keyboard navigation" [Critical,Fix committed]
[16:50] <apinheiro> is still valid
[16:50] <apinheiro> njpatel, could you test it?
[16:50] <apinheiro> or it is something that only happens if you do a manual unity --replace ?
[16:51] <njpatel> apinheiro, https://bugs.launchpad.net/unity/+bug/860805 :)
[16:51] <ubot5> Ubuntu bug 860805 in unity "Dash doesn't get the focus on pressing the super key first time after login" [High,Confirmed]
[16:53] <apinheiro> njpatel, ah ok, I thought that it was managed on the same bug
[16:53] <apinheiro> thanks
[17:05] <AlanBell> apinheiro: do you get orca reading out "button" for everything in the lenses?
[17:05] <apinheiro> AlanBell, do you usually read ubuntu-accessibility mailing lists?
[17:06] <AlanBell> yes
[17:06] <apinheiro> well, some weeks ago I sent a mail
[17:06] <apinheiro> in order to have a11y support for most of the dash etc
[17:06] <apinheiro> it is required these branches:
[17:06] <apinheiro> https://code.launchpad.net/~apinheiro
[17:06] <apinheiro> as I was not able to finish them before
[17:06] <apinheiro> feature freeze
=== yofel_ is now known as yofel
[17:07] <apinheiro> they will not be included on the official release
[17:07] <apinheiro> Im checking how to create a ppa including those changes
[17:07] <apinheiro> so the users could start to test them and give some feedback
[17:07] <AlanBell> ok, I recall that mail, I didn't realise the connection to the symptoms
[17:08] <AlanBell> there is some set of circumstances when it actually works right now, for the top row of icons in a category
[17:09] <apinheiro> AlanBell, I'm talking about Unity3D
[17:09] <ach1m> om26er, with which graphics chip do you test unity?
[17:09] <apinheiro> on Unity3D , withoug those branches
[17:09] <AlanBell> oh, OK that one is silent
[17:09] <apinheiro> it works the launcher
[17:10] <om26er> ach1m, nvidia and intel
[17:10] <AlanBell> yeah, unity2d everything is called "button"
[17:10] <ach1m> om26er, could it be that this bug #851337 only happens on intel HW?
[17:10] * om26er should get ati system too though
[17:10] <ubot5> Launchpad bug 851337 in unity "window decorations missing after expo" [Undecided,Confirmed] https://launchpad.net/bugs/851337
[17:10] <ach1m> I can reproduce this with g45 and gma945.
[17:11] <om26er> ach1m, i guess it to not be an issue when on live session
[17:11] <om26er> can you reproduce it on live session as well?
[17:11] <ach1m> sorry my Internet connection is not that fast, so I can get a liveCD very fast.
[17:12] <ach1m> can not*
[17:13] <ach1m> om26er, I tried unity --reset, it didn't help.
[17:13] <om26er> ach1m, also tested a new user?
[17:14] <om26er> ach1m, I have gma950 and it does not happenb
[17:14] <om26er> though i have yet to do thorough testing on that netbook
[17:14] <ach1m> Okay I can test a new user.
[17:15] * ach1m is back in a minute.
[17:20] <ach1m> om26er, also the same symptoms for a new user account.
[17:20] <om26er> ach1m, i'll grab the latest iso and put it on my gma950 netbook and will report in the bug
[17:20] <ach1m> thx.
[17:20] <om26er> internet is killing me though :D
[17:21] <om26er> ach1m, which system of your have gma950 ?
[17:21] <om26er> is that a netbook
[17:21] <ach1m> a notebook with gma945 and a Desktop system with an g45 aka x4500HD.
[17:22] <om26er> ach1m, both have the issue?
[17:23] <om26er> interesting
[17:23] <ach1m> yes.
[17:26] <om26er> ach1m, can you open a terminal and type unity-window-decorator --replace
[17:26] <om26er> and then let me know if the issue happens
[17:26] <om26er> seems i can reproduce with gtk-window-decorator
[17:27] <ach1m> om26er, http://paste.ubuntu.com/698610/
[17:27] <om26er> ach1m, test now if the issue is happening
[17:28] <ach1m> the problem is still there.
[17:29] <om26er> i'll test my netbook or i might end up pulling the nvidia card off and try with the integrated :D
[17:29] <ach1m> okay.
[17:30] <om26er> i just reproduced it but i had gtk-window-decorator as my window decorator
[17:31] <ach1m> om26er, it happens with gtk-window-decorator and unity-window-decorator.
[17:33] <om26er> reproduced on nvidia :)
[17:34] <ach1m> ah nice, or not so nice – but at least it seems not to be a driver issue.
[18:38] <njpatel> Trevinho, ping
=== m_conley_away is now known as m_conley
[19:59] <jono> DBO, about set for our call?
[19:59] <DBO> skype or mumble?
[19:59] <jono> DBO, lets do skype, give me one sec
[20:04] <Trevinho> njpatel: pong
[20:04] <njpatel> Trevinho, hey dude
[20:04] <Trevinho> Hi, I've seen the merge, thanks
[20:04] <njpatel> Trevinho, found an issue, first-click on a entry closes it again. I think it was with the fix for dealing with being able to keynav with the mouse on the panel
[20:05] <njpatel> Trevinho, no worries, awesome work dude
[20:05] <Trevinho> njpatel: I check
[20:05] <njpatel> thanks dude
[20:06] <Trevinho> But... First click... what do you mean exactly?
[20:06] <njpatel> just, when you click on a menu, it closes automatically
[20:06] <Trevinho> If I click over an indicator entry, it opens here correctly...
[20:06] <njpatel> then the next time you click, it works
[20:06] <njpatel> sorry, on an entry
[20:06] <Trevinho> Ahhh... ok
[20:06] <Trevinho> I try
[20:10] <Trevinho> So... njpatel to reproduce it I should click over an entry -> and I do a first click? Mhmhm... I can't understand how to reproduce it
[20:11] <njpatel> Trevinho, for me, with trunk unity, just clicking on an entry will cause the menu to show then hide
[20:12] <Trevinho> Mhmh... It doesn't here... I check if I forgot to merge some changes
[20:12] <njpatel> try a few times on different indicators, it seems reproducable
[20:19] <Trevinho> njpatel: but was this already present before the fading panel merge, isn't it?
[20:20] <njpatel> Trevinho, yep
[20:20] <njpatel> found it this morning
[20:20] <Trevinho> njpatel: but it closes and reopens to you or does it stay closed?
[20:22] <njpatel> tremolux, stay closed
[20:23] <njpatel> then I have to click again and it works fine
[20:23] <njpatel> it's really weird
[20:23] * tremolux waves to njpatel :)
[20:23] <njpatel> woop,s orry :)
[20:32] <Trevinho> njpatel: Mhmh... I can't reproduce it... The only issue I have is with the network indicator that quickly closes and re-opens its menu
[20:32] <Trevinho> but... The other menus seems to work fine
[20:33] <Trevinho> mhhmhm
[20:33] <njpatel> Trevinho, okay dude, no worries, I'll have a look when I get a chance
[20:37] <Trevinho> njpatel: if you can to do a quick screencast I maybe understand better what's going on there...
[20:39] <njpatel> Trevinho, I can't right now dude :/ don't worry we can fix later if necessary
[20:39] <Trevinho> Mh, ok...
[20:47] <Trevinho> however njpatel I guess that is mostly controlled by the panel-service.c's event_filter function
[20:48] <njpatel> tremolux, okay, I can check that
[20:48] * tremolux hugs njpatel
[20:48] <Trevinho> lol... Your tab is broken! :D
[20:48] * Trevinho is know as tremolux
[20:49] <Trevinho> I'm sorry but you need to press a "v" too :D
[20:51] <Trevinho> njpatel: however when you've time just merge the lp:~3v1n0/unity/fix-690143-v2 branch, it's a small fix, but needed for coherency :)
[20:58] <Trevinho> njpatel: I also experienced another bug that was present also before my changes, see bug #861279 however to test it just, open a terminal -> drag it on the right side of the screen -> make it semi-maximize -> oops, the panel will be empty and will stay empty until you won't switch the focus to another window and back to the terminal
[20:58] <ubot5> Launchpad bug 861279 in unity (Ubuntu) "Missing global menu with the first opened terminal" [Undecided,New] https://launchpad.net/bugs/861279
[20:59] <njpatel> Trevinho, wow
[20:59] <njpatel> okay, will take a look
[20:59] <Trevinho> njpatel: don't worry, I try to fix it now...
[20:59] <Trevinho> If I can, then the ball is all yours! :)
[20:59] <Trevinho> can't*
[21:00] <njpatel> thanks :)
[21:00] <Trevinho> however njpatel i guess that it's related to how the grid plugin works
[21:01] <Trevinho> maybe it simply tries to unmap the window... Who knows... I've to check :P
[21:01] <jcastro> hi guys
[21:04] <Trevinho> Hi jcastro!! :)
[21:05] <Trevinho> How are you?
[21:05] <jcastro> good, you?
[21:06] <Trevinho> Good... also if it's a rush... We to fix lots of things before the FF... :)
[21:10] <Trevinho> jcastro: have you been mostly working with the cloud related projects recently?
[21:10] <jcastro> yeah
[21:11] <jcastro> but I drop in when I can
[21:11] <jcastro> hey njpatel
[21:11] <jcastro> btw that broken twisty is still in the dash
[21:14] <njpatel> hey jcastro
[21:14] <njpatel> jcastro, twisty?
[21:14] <jcastro> yeah, the thing that used to collapse the dash
[21:15] <njpatel> yeah, it's broken for the time being dude, sorry
[21:15] <njpatel> oh, wait
[21:15] <njpatel> no, I fixed it this mornign, it won't show
[21:15] <jcastro> \o/
=== _thumper_ is now known as thumper
=== m_conley is now known as m_conley_away
[21:26] <jbicha> the Desktop icon is broken (as seen in Alt-Tab) or in Nautilus
[21:53] <Trevinho> njpatel: I've found the issue about the menus not being shown
[21:53] <njpatel> ah, awesome!
[21:54] <Trevinho> the fact is that if you move a window outside the monitor area
[21:54] <Trevinho> the menus have been hidden...
[21:54] <Trevinho> I've found how to make them show back
[21:54] <Trevinho> buuuutttt.... I guess we should tune this a little better
[21:55] <Trevinho> I mean... Now if you also move a terminal on the left... To make its center to stay outside the Viewport, then the menus doesn't show
[21:55] <Trevinho> I geuss we should only check if its completely outside....
[21:55] <njpatel> exactly
[21:55] <Trevinho> so, this is the fix... Ok
[21:56] <Trevinho> I change this thing
[21:58] <njpatel> okay, I can try it out tomorrow
[21:59] <Trevinho> cool
[21:59] <Trevinho> I'll send the merge req soon
[21:59] <Trevinho> then I see if I also I've the time for the latest fixes....
[22:00] <Trevinho> There's a very critical one here... If I open a terminal, I keep it above any other window, and I switch VP, then I can't re-focus it using the launcher icon.
[22:00] <Trevinho> jason wasn't able to reproduce it, but actually me and others including om26er yes
[22:08] <Trevinho> njpatel: also... If you take a window focussed with menus and you switch vp using ctrl+alt+arrows, then if you click over the empty panel... Wup! Here you are the menus
[22:08] <Trevinho> but they will hide after the first click
[22:08] <Trevinho> that mostly depends on libappmenus by the way
[22:08] <Trevinho> but maybe we could use a "disable" flag to the panelmenuview not to handle clicks if we don't want
[22:10] <njpatel> indeed
[22:12] <Trevinho> njpatel: another workaround would be to focus the panel itself when switching vp, but that could lead to issues...
[22:12] <Trevinho> So do I go for the disabled thing?
[22:12] <njpatel> yep
[22:13] <Trevinho> ok
[22:13] <Trevinho> njpatel: another question... :)
[22:13] <Trevinho> For the fix I was talking before (don't disable the menus/titles moving the window outside)
[22:13] <Trevinho> I actually check if the window intersects with the monitor / viewport...
[22:14] <Trevinho> do you want to set a minimum width/height?
[22:14] <Trevinho> I mean, It's currently impossible to move a window and not to intersect it with the current VP... Due to the grid plugin
[22:15] <Trevinho> so maybe, if a window shows only few pixels of itself on a vp, showing its title/menus is wrong
[22:16] <njpatel> Trevinho, well, at the moment we take the x + width/2 but only take y, not the midpoint of the height...that could be it
[22:16] <njpatel> x = x + width/2, I mean
[22:16] <njpatel> so the centre of the window, width wise, but not the centre of the height (to determine which panel it belongs too)
[22:16] <njpatel> to*
[22:18] <Trevinho> So, I'd say... That if less than a quarter of the window is not shown there, then hide in the panel
[22:19] <Trevinho> Also if... there could be windows that aren't in any vp
[22:19] <njpatel> right
[22:24] <sbte> smspillaz, about https://bugs.launchpad.net/unity/+bug/732997
[22:24] <ubot5> Ubuntu bug 732997 in Compiz Core "Cannot open a window that starts iconified" [High,Fix committed]
[22:24] <sbte> it still doesn't seem to work
[22:31] <sbte> smspillaz, you can test it with emesene
[22:34] <sbte> use the default settings and let someone talk to you
[22:35] <Trevinho> njpatel: I'd reduce that value... I'd put also 1/8 of the width/height...
[22:35] <Trevinho> Too little? It shouldn't for big windows
[22:36] <njpatel> not sure dude
[22:36] <njpatel> it might be a bit late to do that :/
[22:37] <njpatel> put up a m-p and I'll get the guys to test int he morning
[22:37] <Trevinho> Ok... I set it to 1/4 so...
[22:37] <Trevinho> Everyday I find a new issue :D
[22:39] <njpatel> heh :)
[22:47] <devfil> hi guys, in oneiric if I minimize the skype window it won't open anymore so I have to kill skype and reopen it
[22:48] <Trevinho> DBO: please merge this, since it's in your scope, https://code.launchpad.net/~3v1n0/unity/fix-690143-v2/+merge/77296 ;)
[22:48] <DBO> Trevinho, does it do that every time a window moves O_O
[22:48] <DBO> ?
[22:48] <Trevinho> DBO: there's no other way to do that
[22:49] <Trevinho> I also asked to Smspillaz....
[22:49] <Trevinho> I can add a timer to reduce
[22:49] <DBO> please
[22:49] <Trevinho> Ok, timer is coming
[22:50] <Trevinho> DBO: 200ms is it fine?
[22:50] <Trevinho> Or slower?
[22:51] <DBO> 250
[22:53] <Trevinho> ok
[22:58] <sbte> smspillaz, added an explanation of the problem and a test to https://bugs.launchpad.net/unity/+bug/732997
[22:58] <ubot5> Ubuntu bug 732997 in Compiz Core "Cannot open a window that starts iconified" [High,Fix committed]
[22:59] <sbte> I would be really happy if you could fix this so I don't have to arrange an SRU for emesene to hack around this
[23:02] <sbte> smspillaz, I'm going to bed now. Please contact devfil if you need any more information
[23:05] <Trevinho> DBO: maybe for keeping the consistency with the panel, we should also check if the window portion actually visible matches the one that the panel uses to show or hide the menus/titles
[23:05] <Trevinho> isn't it?
[23:06] <DBO> Trevinho, I dont know? :)
[23:06] <Trevinho> So, we have a coherent representation...
[23:07] <Trevinho> DBO: move a restored window (terminal) on the left... But don't make the grid plugin to semi-maximize it.
[23:07] <Trevinho> the title and menus are hidden
[23:07] <Trevinho> in that case the laucher should be coherent
[23:07] <Trevinho> however, I guess that there's no time...
[23:07] <Trevinho> IMHO in that case, the launcher should move to the VP containing the higher portion of that window
[23:07] <DBO> yeah not really
[23:08] <Trevinho> and draw itself with the right indicator
[23:08] <DBO> oh you mean for multi-monitor?
[23:08] <Trevinho> however... Since the "move" thing is harder, I can anyway change the launcher indicator
[23:08] <DBO> Trevinho, its that time of the cycle
[23:08] <DBO> where I trust you to make the right decision
[23:08] <DBO> I will review your code whenever it is ready
[23:08] <DBO> please ping me when it is :)
[23:09] <Trevinho> well, actually it works also for multi-viewport...
[23:09] <Trevinho> okkk
[23:09] <Trevinho> :)
[23:26] <Trevinho> DBO: I've done, https://code.launchpad.net/~3v1n0/unity/fix-690143-v2/+merge/77296 I've not added that thing about the 1/4 width because It's too late for that I guess...
[23:26] <Trevinho> Just fixing now :)
[23:27] <DBO> kk
[23:35] <DBO> Trevinho, is it time?
[23:45] <Trevinho> DBO: ehm?
[23:46] <DBO> when should I merge buddy ol pal ol friend ol spot of tea?
[23:46] <Trevinho> Now is fine...
[23:46] <Trevinho> I mean, I've added the timer...
[23:49] <DBO> okay
[23:49] <DBO> will do as soon as I get to it soon
[23:50] <Trevinho> :)
[23:52] <Trevinho> njpatel: the branch we were talking before is now up at https://code.launchpad.net/~3v1n0/unity/panel-active-window-fixes/+merge/77418