[01:36] chrisccoulson: ping === gabon_ is now known as gabon === m_conley is now known as m_conley_away [04:52] any libgtk2.0 packaging experts around? [04:53] * achiang is trying to figure out how to backport maverick's libgtk2.0 into lucid without getting caught in a spiraling maze of versioned build-dependencies, and could use some advice [05:04] achiang: You will probably have to wait till seb128 is around. [05:05] TheMuso: ah, thanks. i think i'm picking my way through the easier ones, perhaps by the time i get to the hard ones, seb128 will be awake again. :) [06:25] morning [06:54] kvalo: morning :) [07:41] good morning [07:44] didrocks: good morning. how's the life in france? [07:44] kvalo: cold, but I guess not as cold as for you :) [07:44] kvalo: and you? [07:45] it's warm here, 0 C :) not much sun though... [07:49] 0° C <-> warm. SIGSEV :) [07:49] <\sh> *grmpf* ../NuxGraphics/.libs/libnux-graphics-0.9.so: undefined reference to 'gluErrorString' , building nux from trunk whoosa [07:50] good morning [07:54] didrocks: :) [07:54] oSoMoN: good morning [07:57] hi kvalo, how is it going? [07:58] oSoMoN: I'm good, thanks. how are you? [07:59] kvalo: very good, thank you [08:05] hey folks [08:07] hey MacSlow [08:07] hi oSoMoN [08:37] are places menus known to have broken scrollbars? [08:37] also are they known to take a very long time to populate [08:38] apw: there is a bug report about it [08:38] for both [08:38] apw: try to look at unity bugs if you can add additional infos:) [08:39] didrocks, its hard to add info when you can't interact with any of your windows most of the time [08:39] didrocks, its a rather regualar failure mode for me [08:39] apw: oh you mean, you can't click on your window? [08:39] * apw dumps his natty laptop on the useless stack [08:39] but keyboard stil works? [08:39] didrocks, indeed [08:39] smspillaz: do you need live info ? ^^ [08:39] apw: this is a really annoying info, hard to reproduce though [08:40] apw: we have a bug about it if we are interested, one second, though [08:40] didrocks, i've had it 3-4 logins recently [08:40] apw: run xwininfo -all in a terminal and click on the "dead region" [08:40] apw: I have it sometimes a lot in the morning, and no more for the whole day… [08:40] also post your xwininfo -root -tree [08:40] smspillaz, the dead region is all of my screen [08:40] are you running unity ? [08:40] smspillaz, unity yes [08:40] apw: the bug is that there is some window that isn't painted that is there [08:41] and I'm currently trying to fix it :) [08:41] smspillaz, yeah and i think its a modal dialog [08:41] what application? [08:41] but i can't interact with it either, so i can't get rid of it [08:41] * apw can't tell as it is invisible, and uninteractible [08:42] apw: FYI, it's bug #709461 if you want to follow it [08:42] Launchpad bug 709461 in Unity "semi-random invisible window with x geometry on top layer possible, all viewport only (one ws though)" [Critical,Triaged] https://launchpad.net/bugs/709461 [08:42] smspillaz: some to be $random application in any case, as I had eog, ccsm… [08:43] didrocks, what does this window look like in xwininfo [08:43] apw: if you just run xwininfo (you need to ctrl + alt + t to open a terminal) you should be able to click on it [08:43] and it dumps the info of where you click [08:43] which should be the "invisible window" then [08:44] yeah [08:44] didrocks, oh its xchat apparently [08:44] i'll get the info on the bug [08:46] ok, was disconnected from freenode… [08:46] 09:44:47 didrocks | oh, it's really $random apps I guess :/ [08:46] 09:45:04 didrocks | thanks [08:46] 09:45:27 didrocks | apw: btw, restarting unity (running unity) should help [08:46] 09:45:37 didrocks | apw: if not, you can try metacity --replace and then run unity [08:48] didrocks, smspillaz, ok i have added the wininfo for the lost window and the complete stack, to the bug ... anything else you want off the machine? [08:49] also anyone got the bug number for the scrollbars not existing [08:49] apw: not handy, but looking for "scrollbar" in the unity bugs should give it to you pretty straightforward [08:49] finally :) are the windows incredibly slow to fill in for everyone else? mine take 15-20 before there is anything on them [08:49] I filed it on Monday [08:49] (windows == places windows) [08:50] apw: and you, same there is one about it, there is no optimization for this release [08:50] I also filed it on Monday [08:50] ok [08:50] <\sh> didrocks: when you find the time, please review bug #711717 (branch for merging attached) [08:50] apw: I think there are enough info, thanks! [08:50] Launchpad bug 711717 in Nux "Nux trunk FTBFS under natty" [Undecided,New] https://launchpad.net/bugs/711717 [08:50] * apw notes for this stacking bug, that although this is xchat running the launcher does not have the ><'s on it for xchat [08:50] \sh: excellent, thanks! [08:51] apw: oh? this is not known [08:51] apw: please file a bug about it [08:51] * apw tries to get a screen shot [08:51] apw: affecting unity and bamf [08:51] ack [08:53] ok to do that i am going to have to kill off this xchat to get to the bug dialogs ... so last chance for more info on it [08:53] apw: oh know, wait [08:53] * apw waits [08:54] apw: the no match on the launcher is because of the window not mapped [08:54] apw: so, it's no a bug [08:54] didrocks, ok i'll mention it in the original bug only [08:54] it's just a side effect :) [08:54] yeah, thanks! [08:55] didrocks, unity really needs a 'kill -USR1 ' to tell it to dump its stacking order information out or something, or turn on debug in general [08:56] apw: there is some work on that, we not totally finished [08:56] but* [08:57] apw: ccsm -> debug spewer -> change the keybinding [08:57] and we already have too much stuff on our plate for just features [08:57] smspillaz: it's crashing for me, do you remember? :) [08:57] right :/ [08:58] ok if there is nothing else i can add to the bug, i'll try killing off xchat and see if that resolves the issue or if minds are already lost [09:02] didrocks, ok i killed xchat and the 'gap' is still there terms of focus tracking [09:02] starting a new xchat means that whereever i put that on the screen it still gets focus for the old area, and its new area [09:02] apw: weird, you should rather rerun unity [09:03] can I have branches inside one repository, just like git does? I hate it that I need create a new directory for each branch [09:03] didrocks, so i assume that the input tracking is done by unity ? [09:03] and talking about bzr, of course [09:03] kvalo, i think there is an extension for that [09:03] apw: compiz has some [09:03] didrocks, so i suspect that that is where the issue is then [09:03] apw: great. do you happen to remember the name? :) [09:05] kvalo, hrm no sorry not a big bzr user [09:05] apw: thanks. you gave me motivation to find it :) [09:09] kvalo, possibly you can do it with bzr switch, "set up a shared repository with the --no-trees option together with lightweight checkout" [09:10] if those words mean anything to you [09:10] perhaps the 'colo' plugin [09:10] apw: heh, I wish :) [09:10] http://wiki.bazaar.canonical.com/GitStyleBranches [09:11] apw: excellent, thanks a lot! [09:11] didrocks: ARGH you were right [09:11] ls ~/.compiz-1/plugins -> libunityshell.so [09:11] *removes* [09:12] * didrocks stares at smspillaz :) [09:12] yay, now I have a unity back [09:12] apw: out of curiousity, which application is the one which had the invisible window which is blocking everything ? [09:12] smspillaz: I'll do a unity --distro :) [09:13] indeed :P [09:15] smspillaz, in my case it was xchat [09:15] xchats main window [09:17] ok [09:23] ah ok, reproduced it [09:27] ok, looks like the window is tracked initially and then unreparented [09:28] <\sh> hmm..this is strange [09:29] oh is the top left logo window meant to have anything in it? its empty for me [09:33] smspillaz, ^^ [09:36] apw: it is [09:36] apw: I only work on compiz :) so I wouldn't know much about the unity panels and launcher [09:36] smspillaz, ahh thanks :) [09:37] didrocks, whats it meant to have in it, and i assume its known that it does not so contain [09:37] apw: the same content than in maverick version, prefered apps and such [09:38] guys, building last nux from trunk fails: http://pastebin.com/UH7StuR4 [09:38] didrocks, ok so thats missing [09:38] yeah, still an early place draft [09:39] <\sh> cando_: bug #711717 [09:39] Launchpad bug 711717 in Nux "Nux trunk FTBFS under natty" [Undecided,New] https://launchpad.net/bugs/711717 [09:39] ok thanks [09:39] <\sh> cando_: fix is there [09:39] :) [09:41] smspillaz, just for you compiz just core dumped on login for me [09:46] are random hangs/compiz dumps as 3/4 logins known ? [09:49] apw: got a bug # for that? [09:49] apw: ah right, might be the places issue [09:49] no i've given filing bugs on it, as mostly the filing bug process dies too [09:50] and finally after 4 attempts i have a desktop with some interactible icons .. [09:51] smspillaz, if i ever manage to get one i'll let you know [09:51] ok [09:58] \sh, ok build flawlessy...thanks [09:58] kamstrup: does "G_PARAM_READABLE | G_PARAM_CONSTRUCT_ONLY" make sense? [09:58] kamstrup: for a property which is only set in the constructor and read otherwise [10:07] didrocks: I love it when this happens - I try to open geany to edit the source, and the bug happens to it -_- [10:07] see the compiz bugs have minds of their own ;-) [10:07] smspillaz: ahah, like for me with ccsm yesterday to set a debug shortcut? :) [10:07] yeah :p [10:08] kvalo: it needs to be G_PARAM_READWRITE | G_PARAM_CONSTRUCT_ONLY [10:08] kamstrup: oh. I'll fix that, thanks [10:09] kvalo: otherwise you'll get complaints from glib when you try and set the prop on construction time [10:09] kamstrup, njpatel anyone know a way to easily deep copy a GArray [10:09] can't see it in the api [10:09] didrocks: btw, do you know if there's any way to make bzr not rebuild the *entire* source package on bzr bd-do? [10:09] glist has a nice convenience function for it but not garray [10:10] smspillaz: unfortunately not and it removes the build-area [10:10] ronoc: GArray is a boxed type so you can do: [10:10] smspillaz: if you just want to test something [10:10] smspillaz: like, don't use the patch system [10:11] ronoc: GArray copy = g_boxed_copy (G_TYPE_ARRAY, my_array); [10:11] smspillaz: you can bzr bd-do ; quilt push -a ; make your changes ; debuild -nc [10:11] (and then make your changes ; debuild -nc …) [10:11] didrocks: cool [10:11] kamstrup, oh lovely thx [10:11] smspillaz: take care to not exit! [10:11] lovely [10:11] didrocks: yeah [10:11] smspillaz: as it will destroy the build-area :) [10:11] yeah [10:11] smspillaz: also, your changes will be in no patch, so take care! [10:12] kamstrup: that's black magic! ;) [10:13] kvalo: you need to learn to love the boxed types! ;-) [10:13] gobject: voodoo for the programmers [10:13] kamstrup: yeah, sure. after a lobotomy ;) [10:13] kvalo: that can be arranged === _LibertyZero is now known as LibertyZero === Cimi__ is now known as Cimi === smspillaz is now known as smspillaz|dinner [10:50] morning johnlea [10:51] klattimer; morning [10:51] johnlea: will want to pick your brains probably about timezone matching at some point [10:52] klattimer; ok, is it really urgent or can we do it towards the end of next week [10:53] next week will probably be fine [11:05] <\sh> didrocks: thx :) [11:05] \sh: thanks to you :) [11:10] <\sh> jcastro: https://wiki.ubuntu.com/Unity <- the instruction to build unity on maverick is wrong...first, there is no gsettings-desktop-schemas-dev in maverick, so this doesn't work, I wonder if we just drop this section and tell the people to not try to build it on maverick... === smspillaz|dinner is now known as smspillaz [11:44] kamstrup: g_value_dup_object()? huh, quite a misleading name [11:59] kvalo: yeah, but internally consistent with the rest of the gvalue api at least [11:59] kamstrup: true [12:17] njpatel, is there a log for the unity-panel-service I can tail ? [12:34] hi guys, i've a small problem: i can't run unity from trunk anymore...i mean: i've build nux and unity, setted up the env , typed compiz --replace, but always starts the "old" unity...what i'm doing wrong? [12:34] last week i was able to run unity from trunk... === MacSlow is now known as MacSlow|lunch [13:02] \sh, you there? [13:43] kamstrup: I have a problem with annotatations, any idea what's wrong: http://paste.ubuntu.com/561388/ [13:48] kvalo: looks odd... [13:49] kamstrup: g-ir-scanner is giving other warnings as well. I'll fix those first [13:52] maybe it magically disappers :) [13:55] didrocks, you got a moment? [13:56] cando_: not really know, fixing some crash for alpha2 [13:56] ok no problem..:) [13:56] cando_: in 3h, I think it should be fine [13:56] sorry :/ [13:56] ok thanks! [13:57] I have a problem with the installation of the "guest additions" in today's version of "natty" any suggestions? http://paste.ubuntu.com/561389/ [13:59] kamstrup: argh, a colon was missing after the function name... [13:59] hmm [13:59] rsajdok, that doesn't look good [14:01] spikeb: :) [14:02] man, i was going to try running natty (and the new unity) on virtualbox too. [14:02] * spikeb sighs [14:04] \sh: There are problems in natty also. [14:06] smspillaz: the fix is working [14:07] didrocks: nice [14:07] didrocks: so at least we know what is broken [14:08] smspillaz: yeah, I'll commit the fix in git if you want as the core components should rather use this as you told [14:08] indeed [14:08] just replace all of the s->getOptions with that [14:08] yeah, will do :) [14:08] err [14:08] s->getOption' [14:08] got it, no worry :) [14:08] :) [14:09] there is actually an s->getOptions () which serves a completely different purpose :p [14:15] argh ;) [14:15] naming :) === MacSlow|lunch is now known as MacSlow [14:34] gah, trapsing through the evolution source code I discover bugs bugs bugs... URI opening is broken because it wrongly detects "--import" on the command line, and the URI seems to not get passed correctly to running instances... [14:35] I hate evolution [14:36] * smspillaz -> bed [14:36] m_conley_away, there? [14:36] \sh: oh nice catch, thanks [14:41] <\sh> jcastro: updated as well the build-deps with some packages which were missing [14:42] hmm, DBO just updated those the other day [14:42] but adding more won't hurt I guess [14:42] njpatel, how do you set your borders to 0 px? [14:42] <\sh> jcastro: gnome-common and libpcre3-dev were definitly missing ;) [14:43] <\sh> jcastro: and I wonder if bug #692444 is really a bitesize ;) it needs IMHO a lot of refurbish of the startup code of the trash folder [14:43] Launchpad bug 692444 in Unity "clicking trash multiple times opens multiple instances of it." [Medium,In progress] https://launchpad.net/bugs/692444 [14:46] well, cando snagged it so who knows [14:46] are you looking for one no one is working on? [14:47] bug #692444 is definitively not a bitesize one [14:47] Launchpad bug 692444 in Unity "clicking trash multiple times opens multiple instances of it." [Medium,In progress] https://launchpad.net/bugs/692444 [14:47] but if cando_ can get it :) [14:47] cando_'s becoming large bug man [14:48] cando_: btw, I'm more available right now if you stil have your question [14:48] lamalex, /usr/share/themes/Ambiance/metacity-1/metacity-foo.xml, there are some settings for frame borders, you can set them to 0px. Or use my AmbianceThin theme http://bit.ly/h9XCYe [14:48] <\sh> didrocks: I was wondering if replacing g_spawn_command_line_async with g_spawn_async_with_pipes, catching the pid, adding a callback for cleaning up when pid is destroyed, and iterating over all gdk toplevel windows and focusing the right one is a good approach [14:49] \sh: no, that's evil and slow :) [14:49] <\sh> anyways...hp technician is waiting for me, bbl [14:50] njpatel, sweet thanks [14:51] lamalex, actually, grab it again, the latest has the shadow bits too [14:52] didrocks, i'm digging into that bug..and i'm very lucky! everytime i choose a bug it not so "bitesize"..:) [14:52] cando_: getting it done is just a real pleasure then :) [14:52] njpatel, I just grabbed it as I said "sweet thanks" so I'm probably ok [14:52] the shadow looks lighter [14:52] yeah my question is: i've a small problem: i can't run unity from trunk anymore...i mean: i've build nux and unity, setted up the env , typed compiz --replace, but always starts the "old" unity.. [14:52] lamalex, of the main window? yeah it should be lighter and have a bigger radius [14:53] now i'm copying lbunityshell.so from opt/unity/compiz/bin to /usr/lib/compiz and it works..but i guess it's not the right thing to do [14:53] cando_, where are you installing unity to? [14:53] cando_, I use this for my cmake generator line [14:53] cmake .. -DCMAKE_BUILD_TYPE=Debug -DCOMPIZ_PLUGIN_INSTALL_TYPE=local -DCMAKE_INSTALL_PREFIX=/usr/local [14:53] i'm following this: https://wiki.ubuntu.com/Unity/InstallationGuideFromSource [14:54] ah ok [14:54] that installs to .compiz-1/plugins or something [14:54] it's diffent..the page i've linked says to install in opt [14:54] yeah that's a way [14:54] Try using my cmake jawn [14:54] see if it works for you [14:54] ok trying...thanks lamalex [14:54] the difference is that you have to set up the env for the opt one [14:55] compiz will always prefer the one in .compiz-1/plugins [14:55] cando_, make sure -DCOMPIZ_PLUGIN_INSTALL_TYPE=local is on the cmake line for UNity, so it installs into ~/.compiz-1/plugins/ [14:55] cando_, that will ghost the system one and allow you to easily recovered if/when we break trunk :) [14:56] njpatel, building...:) i'll let you know in a second [14:57] oh, sorry, didn't see eariler comments [14:58] njpatel, lamalex , btw this page should be updated??https://wiki.ubuntu.com/Unity/InstallationGuideFromSource it doens't work in that way.. [14:58] MacSlow, I'm doing your keynav review now [14:58] lamalex, thanks! [14:59] lamalex, btw... adding the check for Shift for Shift-F10, which I forgot... so keep that in mind when reviewing/testing [15:07] MacSlow, where are you adding it? [15:08] lamalex, in the same branch... but it turned out that I first need to fix something in nux which is missing in order to allow checking for the key-modifier state [15:08] so I'm doing that (fixing nux) right now [15:08] k [15:09] *sigh* david barths internet has a shorter continuity period than a sentence [15:11] njpatel, ok works flawlessy..thanks [15:11] awesome! [15:12] MacSlow, do I need a different nux branch for this? [15:15] lamalex, nux trunk should do [15:16] MacSlow, ok, maybe im a few revs behind [15:19] lamalex, there are some patches that were merged recently to nux regarding keyboard-focus handling... loicm did them... check the bzr log of the nux-branch you have [15:24] haha MacSlow I get a massive compiler error trying to build your branch [15:25] lamalex, just checking the state of nux trunk [15:28] MacSlow, what nux trunk rev do you have? [15:29] lamalex, rev 191 two days old like I said [15:29] I didn't think I would care when running classic desktop but wow, that 50 pixels you lose to panels really makes a difference [15:30] Amaranth, 50 px ?! [15:30] Amaranth, how huge did you set your font? [15:30] MacSlow: 25 per panel [15:30] Or 24, forget [15:30] Amaranth, but we still have the panel at the top [15:30] so it would be only 25px gained [15:30] MacSlow: Not when the window is maximized :) [15:31] Amaranth, ah... that way you're going... ok [15:38] lamalex, still compiling the current nux trunk [15:41] lamalex, I'm getting similar compile-errors here... now looking at the commit history to figure out which commit is the culprit for htis === oubiwann is now known as oubiwann_ [15:43] lamalex, my guess is nux rev194 messed things up [15:45] loicm, did you test-compile unity against nux when you pushed nux r194? Did that work cleanly? [15:48] MacSlow: I had to break the nux API to make that change (r194) [15:48] MacSlow: Unity needs a fix I pushed at r813 [15:49] loicm, so we need to update unity then to reflect that?! [15:49] MacSlow: yes [15:49] ok [15:49] lamalex, ^^ [15:49] lamalex, since my branch isn't based on that it explains the break [15:50] lamalex, I'll try to rebase and see how that goes [15:50] MacSlow: If updating is problematic, that's a one line patch [15:51] ok [15:52] MacSlow, let me know when your branch is updated [16:01] cyphermox, what was the status of your multi-monitor branch? [16:03] lamalex, still not working [16:03] I haven't really had time to get back to it [16:03] cyphermox, do you have any idea what the problem was? [16:04] lamalex, not really. I was told by jaytaoko it might be related to how nux handles monitor size updates, but I couldn't dig into it [16:04] cyphermox, ok [16:04] thanks for the headstart, I'm going to take a look at it [16:04] it's driving me nuts :{ [16:04] lamalex, right now, it works fine when unity starts, but fails to redraw the regions after a size change [16:05] hm, well that's at least an improvement. [16:05] lamalex, the panel and launcher *do* react to mouse events for example at the right place, though, it's just not drawn correctly [16:05] oh weird [16:05] loicm, lamalex: rebased my branch on current unity trunk... just pushing it now [16:05] I think I wasn't far, just no idea what's missing ;) [16:05] loicm, lamalex: compiles and runs fine [16:07] lamalex, let me put two minutes to update the branch and clean up the unnecessary garbage :) [16:07] cyphermox, sure [16:07] cyphermox, if we can just have it working on start, it'd at least be SOMETHING until we figure out what's going on in nux with redraw [16:07] because for me right now, a good portion of my launcher is obscured [16:08] and it's almost unusable [16:08] need to alt tab to get to most things [16:09] same for me. it's annoying, but I got used to it ^.^ [16:09] lamalex, just cleaning up your stated issues [16:11] @anyone_from_design in gnome session there is a arrow on the ubuntu logo http://imagebin.org/index.php?mode=image&id=135775 and seems there is a patch rotting in bugzilla, if it works would it be fine to remove it in natty? [16:15] jcastro: around? [16:15] on a call, but keep typing. :) [16:15] jcastro: can you register "glew" as a launchpad project? [16:15] yep [16:15] excellent :) [16:15] belonging to ... ? [16:16] jcastro: it's an outside launchpad project, I just want to link a bug/patch [16:16] external* [16:17] ok I just need the license [16:17] oh, one sec [16:17] The original version of GLEW contained files under 4 licenses: [16:17] BSD, GPLv2+, SGI Free Software License B, GLX Public License [16:18] lamalex, it's updated, and still at lp:~mathieu-tl/unity/multimonitor/ (it deleted and recreated it) [16:19] https://launchpad.net/glew/ [16:19] done [16:21] jcastro: awesome, thanks! [16:36] MacSlow, updated the review for behavior [16:37] lamalex, pushed fixes regarding the initially stated the issues [16:39] are "long delays before indicators appear after login" a known issue ? [16:39] lamalex, see my reply [16:40] lamalex, only the key-nav of the quicklist will come in another branch... I just want to make sure this launcher-key-nav lands first [16:43] MacSlow, well im not talking about navigating the quicklists [16:43] when you right arrow to a ql, then hit down, you start back at the top of the launcher [16:43] that seems like a bug you'd hit ql nav or not [16:45] lamalex, oh... that should not happen... keyboard-focus should be dropped from the launcher *sigh* [16:45] lamalex, looking into this now [16:49] <\sh> jcastro: someone said, unity should run now under VirtualBox 4... hmm...trying to install the guest additions gives me: unsupported pre-release of Xorg server"... [16:49] you just ran into the Xorg transition [16:51] <\sh> jcastro: for sure :) well, I ran also into a non working laptop natty install ;) === JanC_ is now known as JanC [16:51] <\sh> but that has nothing to do with X or eventually it has to do with X it just stands at checking battery state ;) [16:51] yeah, yesterday wasn't a good day [16:52] I suspect it'll all be fine after today/tomorrow [16:52] well, other than the virtualbox thing, that's probably up to them to support X 1.10 [16:53] <\sh> WTH...fatal error: linux/autoconf.h: no such file... [16:55] <\sh> jcastro: but it is right, when you can't get a second instance of an app running under unity? I tested yesterday unity-2d and I'm not able to start two instances of gnome-terminal [16:55] <\sh> s/when/that/ [16:56] I am stuck in the same boat [16:56] I am just making new tabs in terminals [16:57] <\sh> jcastro: agreed, a nice workaround...but the behaviour is a bit strange [16:57] you're supposed to be able to right click and create new terminal like in unity 3d [16:57] it's just a bug [17:00] ronoc, hey, did you manage to take a look at the vala issue? [17:00] and471, no sorry I didn't , where is the code again ? [17:01] umm.. [17:01] lp:~and471/+junk/symphony [17:01] ronoc, ^ [17:02] and471, looking now [17:02] thanks [17:05] and471, compilation error -> src/backend/dbus.vala:88.73-88.73: error: syntax error, expected `;' [17:06] garr sorry, let me fix that [17:10] ronoc, do a bzr pull [17:10] ronoc, sorry about that === m_conley_away is now known as m_conley [17:16] and471, no worries, so I can't compile because I'm not natty and the dependencies have shifted [17:16] but [17:18] and471, i think you should place the libindicate in a separate class , [17:19] not quite sure why you are claiming ownership twice of the name [17:19] but all of your answers are in the source for the xnoise plugins [17:19] for maverick [17:19] http://code.google.com/p/xnoise/downloads/detail?name=xnoise-plugins-core-0.1.13.tar.bz2 [17:19] and471, ^ [17:19] ronoc, sorry, my problem was that it wouldn't compile [17:19] oh [17:20] hehe [17:20] and471, what was the problem again [17:20] /home/andrew/Software/Projects/symphony/src/backend/dbus.c:16: fatal error: libindicate/indicator-messages.h [17:20] No such file or directory [17:20] right so the vapi needed to be hacked [17:20] for libindicate [17:20] yeah but then it still didn't work [17:20] even with the inludedir stuff [17:21] *includedir [17:21] but the other errors were because you had more modifications to do no ? [17:21] nope [17:21] it just wouldn't compile [17:21] and we got stuck yesterday [17:21] what was the last error ? [17:21] /home/andrew/Software/Projects/symphony/src/backend/dbus.c:16: fatal error: libindicate/indicator-messages.h: No such file or directory [17:21] compilation terminated. [17:22] but I though you got around this problem by fixing the path in the vapi ? [17:22] *t [17:22] nope [17:22] ronoc, sorry I don't mean to be rude but I have to leave soon, so maybe I should take this up with the vala guys at a later date? [17:22] and471, yeah or developer for natty ? [17:22] ronoc, you probably have better things to do :) [17:23] its going be out very soon [17:23] just a few things alright ... [17:23] ronoc, thanks anyway :) [17:24] and471, no probs let me know how you get on. You should talk with the xnoise guys, they have a vala plugin for maverick which works fine [17:24] ok [17:40] unity used to work with libgl1-mesa-dri-experimental before the X update in natty. should the issue be reported for the xorg or maybe something need to be done in unity? bug 710588 [17:40] Launchpad bug 710588 in compiz (Ubuntu) "compiz assert failure: compiz: nv50_pc_emit.c:863: emit_flop: Assertion `STYPE(i, 0) == 0x09' failed." [Medium,New] https://launchpad.net/bugs/710588 [17:54] jaytaoko, How do you cancel a nux timer? Can I just delete it? [17:59] jaytaoko, nevermind, i found RemoveTimerHandler [17:59] chrisccoulson: ping [17:59] hi m_conley, how are you? [18:00] chrisccoulson: hey, things are good! I've kinda got a mystery on my hands here, and wondered if I could get some input [18:00] didrocks, ping [18:00] chrisccoulson: first off - have you tried compiling/running globalmenu-extension with the xulrunner b10 dev libraries? [18:00] nmarques: hey [18:01] chrisccoulson: and if so, have you noticed it segfaulting on startup (even without Lightning installed)? [18:01] didrocks, I've done as you said and cloned the profile from Natty [18:01] m_conley, yeah, i'm running that atm (but only in firefox) [18:01] nmarques: excellent [18:01] chrisccoulson: ah, haven't tried it in FF. Does it work for you in TB? [18:01] didrocks, unity doesn't launch (no segfaults), but I only get the wallpaper and pointer, nevertheless this is on ATI hardware with FireGL (11.1) and that might be problem based on some bug reports [18:02] m_conley, ah. you really need to run the extension with the same gecko version as it was built with [18:02] nmarques: FireGL is the proprietary driver isn't it? [18:02] i'm guessing what has happened is that one of the interfaces changed slightly between B10 and your tbird version [18:02] didrocks, yeap [18:02] chrisccoulson: hm - i think my TB is also on b10... [18:02] nmarques: yeah, there are bugs report about that, doesn't work with unity [18:03] m_conley, is it a nightly, or a milestone release? [18:03] didrocks, I'll try later on with Radeon DRM driver [18:03] chrisccoulson: yeah, i'm on a nightly [18:03] chrisccoulson: which version are you working with? [18:04] didrocks, I'm just installing the whole debugging stack to run unity_support_test on gdb and make sure everything is ok :) [18:04] m_conley, i would try compiling it against the exact thunderbird version that you're running. some of the functions it uses are inlined from headers, and if those change slightly, you get weird crashes [18:04] didrocks, but at least I do have a progress ;) [18:04] chrisccoulson: alright cool, thanks. :) [18:04] nmarques: yeah, all is related to the ATI driver… [18:05] that might not fix it, but that's probably the first thing i would try [18:05] didrocks, one question... does Unity3 rellies on clutk ? I don't have clutk installed [18:05] nmarques: no, it doesn't [18:07] didrocks, thanks, I'm going to debug through unity_support_test and check a couple of things in nux once I have the stuff required installed and will give a go with radeonDRM and later on with Intel chipset after I install Factory there :) [18:07] didrocks, once more, thanks for you help [18:07] nmarques: you're really welcome :) === dneary is now known as dneary_gone === ivanka is now known as ivanka-train [19:34] Hello can I post a REQUEST: unity to support tiling for developers... like awesome wm tiling === jaytaoko is now known as jaytaoko|brb === jaytaoko|brb is now known as jaytaoko === m_conley is now known as m_conley_away === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away === m_conley_away is now known as m_conley [21:49] can I speak to the dev of appmenu? [21:51] it's a bit late, he may already be asleep. [21:53] anyone else responsible for it? [21:54] you may have more luck with asking your questions on the ayatana-dev ML, most devs work in the european TZ, so it's already pretty late for them. [21:58] it's 11pm >< [21:58] and 10pm for the UK [21:58] but yeah, thx [21:59] ayatana-dev is empty... [22:06] I was talking of the mailing list [22:06] about even [22:21] oh [22:21] yo mpt [22:21] do you the appmenu? [22:22] do you use the appmenu? [22:36] m_conley, did you have any luck building the extension against your version of tbird? [22:37] chrisccoulson: hey - i'm running my Natty VM on a pretty weak machine this weak, plus I'm doing a debug build [22:37] so it's taking forever. :) [22:37] heh, that's ok :) [22:37] (i'm recompiling TB) [22:38] i need to really get our comm-central daily builds going [22:38] i keep saying i'm going to do it and then something else comes up ;) [22:38] sounds familiar. :D [22:39] chrisccoulson: hey, wouldn't ya know it - just finished the compile. Gonna see if it segfaults now...*fingers crossed*... [22:39] yeah, fingers crossed :) [22:39] i just had a crash in firefox actually, but i wasn't in gdb to get a trace :( [22:39] chrisccoulson: I've been a webby guy most of my working life - all this segfault business...memory management. And *compiling*. Cripes. :p [22:40] heh, i'm used to it now ;) [22:40] chrisccoulson: did the crash reporter send the data? [22:40] the crash reporter popped up, but the problem is that the symbols from the extension aren't on the server [22:40] so, it's not much use ;) [22:40] chrisccoulson: I've got a lot to learn about all of this. Hope it's cool if I keep pinging you periodically. [22:41] sure, no problem [22:41] chrisccoulson: bam - segfault [22:41] chrisccoulson: :/ [22:41] ah [22:41] did you manage to get a backtrace? [22:41] yeah, I think I know where things are going wrong - I just don't understand *why*. [22:41] give me a sec, i'll sum up my observations here... [22:44] oh, i just reproduced the firefox crash in gdb :) [22:44] chrisccoulson: It seems to be freaking out in the uGlobalMenuDocListener [22:45] chrisccoulson: during its Init [22:45] chrisccoulson: I was tooling around with this yesterday in gdb (I'm pretty new to gdb too) [22:45] chrisccoulson: and there's a point where mDocument is assigned rootNode->GetOwnerDoc() [22:46] chrisccoulson: the value assigned to mDocument gets filled with out-of-bounds pointers, so I figure that's where the problem starts there [22:47] chrisccoulson: but, when I, in gdb, manually call rootNode->GetOwnerDoc(), I get a nicely formed nsXULDocument returned, with no out of bounds addresses inside. [22:47] chrisccoulson: so, I'm wondering if maybe we've hit an inconsistency or problem in the XULRunner libraries [22:47] hmmm, interesting [22:47] chrisccoulson: I understand that nsXULDocument is an nsDocument - it's a subclass of some sort [22:48] chrisccoulson: but the guts get all messed up when assigning it to mDocument. :/ [22:48] q [22:48] yeah, that's strange. it's just returning a raw pointer too :/ [22:49] are you seeing something similar? [22:50] i need to try it in a build of comm-central really [22:50] i'm still running tbird 3.1 here ;) [22:50] ah [22:52] l [22:52] * m_conley keeps choosing the wrong window to type in. *sigh* [22:54] heh [22:54] oh, the firefox crash is an easy fix :) [22:55] uGlobalMenuIconLoader outlives the menu item [22:55] chrisccoulson: i suppose that'd do it, yes. :) [22:56] the extension you're running definately built with the thunderbird headers didn't it? [22:56] the only reason i ask is that GetOwnerDoc is an inline [22:57] it's just a bit strange that calling it manually in GDB seems to give the expected result :/ [22:57] chrisccoulson: I believe it did. If it didn't, how would I check? [22:57] i'm not sure :/ [22:57] actually, it might be possible to figure that out from autoconf.mk [22:57] chrisccoulson: perhaps I have duplicate headers stashed around. I imagine that sort of thing happens. [22:58] ac_configure_args in config/autoconf.mk tells you the build options [22:58] --with-libxul-sdk is the interesting one [22:59] xulrunner-devel-2.0b10 [22:59] that's the same one that I'm using for TB, I'm pretty sure. [23:00] are you building tbird on xulrunner? [23:00] the default will be with it's own copy [23:00] er, no, I suppose not [23:00] hm [23:00] that might cause something like that [23:00] i guess I assumed they'd be magically sync'd. :) [23:00] checking... [23:00] heh [23:00] they're out of sync, which is why we can't share the same xulrunner copy between firefox and tbird ;) [23:01] the releases are generally tagged at slightly different points in time [23:01] i wonder if i get the same crash as you if i run it in our latest ffox nightly. that will be the same gecko version as your tbird build [23:01] ok, hold up - how can I determine which version TB got built with? [23:02] are you building from hg? [23:02] yes. [23:02] did you just grab the latest? if so, it will probably now be 2.0b12pre [23:03] mozilla-central was tagged for b11 earlier today [23:03] I see, ok, checking... [23:03] it would be around 2 weeks newer than the xulrunner version you're building against [23:04] Do you recommend I try to get the latest xulrunner libraries to build with? [23:04] you could try that. but, if you've just built tbird, you should have a working SDK in mozilla/dist/sdk [23:05] you could try passing that path to --with-libxul-sdk [23:05] I'm experience a pretty bad bug. [23:05] I'm in a tty right now [23:05] chrisccoulson: cool, that sounds like a good idea. thanks. :) [23:05] you're welcome :) [23:05] i'm just grabbing our latest ffox nightly. i'll try running the b10 build of the extension in there [23:05] this is why i need to get comm-central nightlies working :) [23:06] cool [23:06] I closed my laptop lid, when I got back, my session was resumed, but I saw that I couldn't click or do anything. [23:06] But, I had a terminal window open, and it was tranparant [23:06] behind it I could still see (and type into) gdm [23:07] Omega, this channel isn't really for support [23:07] which ubuntu version are you using? [23:07] I'm not asking for support. [23:07] I'm on natty alpha 1. [23:07] Reporting a bug. [23:08] well, this isn't really the place to report a bug either ;) [23:08] you need launchpad for that [23:08] Maybe devs want me to do something while it's still in the state of problem. [23:08] I know this. [23:08] right, but a lot of people have already finished for the day already [23:08] Oh. [23:11] wow,my connection is going really slow tonight [23:11] it's probably telling me that it's the end of the day ;) [23:12] chrisccoulson: probably. :) [23:12] it's getting slower and slower ;) [23:12] chrisccoulson: /quit [23:12] might be time to restart my router [23:12] Oops. [23:12] (That wasn't directed at you, I was going to /quit) [23:12] heh ;) [23:13] sounds like Omega is also telling you that it's the end of the day. :) [23:14] Omega, quite a few of us are experiencing window stacking issues and dead spots on the screen with unity atm. perhaps your issue is related? [23:14] in any case, it's known to be quite broken currently ;) [23:17] chrisccoulson: No, GDM was visible behind my terminal (and I just unlocked through it) [23:18] Sounds like yet another stacking bug. [23:18] m_conley, oh, i don't get a crash when i run the b10 build in the latest ffox nightly [23:18] Well, this was after hibernation. [23:18] could be just coincidence though ;) [23:18] the new homepage is sweet! [23:19] yeah, I dig it too. :) [23:19] what's the new homepage o: [23:19] The FF team is really burning the midnight oil on this [23:19] yeah, i'm glad they stuck the restore session button on the homepage now [23:20] although, that doesn't help us much, as we use our own homepage ;) [23:20] you could do the same [23:20] chrisccoulson: hmm...ok, so I redirected to my comm-central sdk, and things were chugging along nicely with make, but then it error'd out, complaining that it can't find nsIRedirectChannelRegistrar.h in /include.... [23:20] and it's right - that file is missing. [23:20] m_conley, the issue with our home page is that it's all online, so it's not possible to have a button to restore the session [23:21] i think that's only available in chrome code [23:21] chrisccoulson: well, you could bundle it as part of the Ubuntu mod pack [23:21] yeah, i've been thinking about that :) [23:21] hm, yep, ok - so nsIRedirectChannelRegistrar.h is in my old xulrunner-dev include folder [23:22] but not in the new sdk in comm-central [23:22] maybe deprecated? [23:22] interesting. what does "find . -name nsIRedirectChannelRegistrar.*" show in the build tree? [23:23] by build tree, do you mean the gloablmenu-extension folder I have? [23:23] er, "globalmenu-extension" [23:23] or do you mean comm-central? [23:24] in comm-central [23:24] i don't think i'm pulling in that header [23:24] searching... ok, it found a .idl for it... [23:24] m_conley: Could you point me to the ppa with the builds you guys are talking about? [23:24] and that's it. [23:24] m_conley, oh, that might be a bug :/ [23:25] who's bug? [23:25] yours ;) [23:25] i'll take a quick look here actually [23:25] alright, cool - i'll start making noise on bugzilla [23:25] Omega: builds of what? [23:26] Omega: I'm using the nightly of Thunderbird, and the globalmenu-extension from Launchpad [23:26] Omega: and Monday's daily of Ubuntu. [23:26] m_conley, oh, did you build tbird with --enable-ipc? [23:27] no, I don't believe so [23:27] checking... [23:27] no, i didn't. [23:27] i'm not sure what the defaults for that are with comm-central [23:28] what does that flag do? [23:28] it enables ipc, but what is ipc? [23:28] ah, found it - reading... [23:28] IPC is to allow things to run out-of-process. does your tbird build have a separate libxul.so? [23:28] or is it just a big thunderbird-bin? [23:29] IPC only gets enabled for builds that create a libxul.so [23:29] (i think) [23:29] you can probably look in mozilla/dist/bin for that [23:30] chrisccoulson: it's symlinked to the libxul in ../../toolkit/library [23:30] so perhaps I pointed to the wrong sdk [23:30] ah, that's ok then. i wonder if comm-central still defaults to no IPC then? [23:30] what does MOZ_IPC say in config/autoconf.mk? [23:33] Hm - blank, so I'll assume false. [23:33] oh, i know :) [23:33] lets hear it [23:33] so, nsNetUtil.h #includes nsIRedirectChannelRegistrar if MOZ_IPC=1 [23:34] and nsIRedirectChannelRegistrar.h only exists if MOZ_IPC=1 too [23:34] so, it sounds like I need to set MOZ_IPC to 1. But why didn't I have to do that before? [23:34] however, when you include nsNetUtil.h, it's getting MOZ_IPC from config/autoconf.mk in globalmenu-extension [23:35] you need to pass --disable-ipc when building the extension [23:35] that should make it work [23:35] ok, testing... [23:35] fingers crossed :) [23:38] hrm, new problem - lots of undefined references to "PR_smprintf" and "PR_smprintf_free" [23:38] along with "PR_IntervalNow" and "PR_sscanf" [23:38] hmmm :/ [23:39] not sure what to make of that. [23:41] hm - I should also mention that those warnings being thrown from xpcom/glue/nsThreadUtils.cpp [23:41] ah, do you have a nspr folder in mozilla/dist/include ? [23:41] I have an nspr.h [23:42] but no folder. [23:42] what is NSPR_CFLAGS pointing to in config/autoconf.mk? [23:42] ....--disable-nspr ? [23:42] let me take a peek, hang on... [23:43] /mozilla/dist/inclued/nspr [23:43] which we've just determined doesn't exist. [23:43] "include" [23:44] ah, that would be why :) [23:44] hmmm, i'm not too sure how that happened though ;) [23:44] well, i'm not too sure *what* happened, so I guess we're even. :D [23:44] heh :) [23:45] so, that is the default value for NSPR_CFLAGS, which you get if you don't build with --without-system-nspr [23:45] but i'm not sure why the headers don't get copied :/ [23:46] kklimonda, bug 712186 = quality bug reporting ;) [23:46] Launchpad bug 712186 in transmission (Ubuntu) "i can' t download in the transmission" [Undecided,Incomplete] https://launchpad.net/bugs/712186 [23:47] that sounds like some of the bugs i get! [23:47] like, "i got an error" ;) [23:47] chrisccoulson: yeah, it's great [23:49] what's the sysrq command to call oom killer? [23:49] I can barely switch between windows right now :/ [23:51] chrisccoulson: oh, I'm blind - yes, there is an nspr folder in mozilla/dist/include/nspr [23:53] chrisccoulson: (globalmenu-extension is working very nicely for me in FF4.0b10, btw)