=== Amaranth is now known as c === c is now known as Amaranth [11:08] Hey guys im getting this error when I open totem in maverick http://pastebin.ubuntu.com/438744/ [11:08] any ideas what the problem is? [11:09] This probably reflects a bug in the program [11:09] jcristau: it says that but totem hasnt been updated in maverick yet [11:10] its still running the version in lucid [11:12] ah its a client side thing [11:31] I thought bratche had fixed that? Clearly not. [11:31] It's ARGB windows fallout. GDK_NATIVE_WINDOWS=1 totem should work. [11:31] RAOF: is there a bug report I can update [11:31] I remember there was a master bug report for lucid [11:32] Not that I remember offhand; it should be pretty easy to find, though. It's on the gtk package. [11:32] ill go look for it after college [14:04] RAOF: bryceh says you might help me with testing this bug that has bit eDP displays: https://bugs.freedesktop.org/show_bug.cgi?id=28070 [15:20] in Lucid, Compose sequences can only consist of two characters, and may only output one character. This makes them exactly equivalent to dead keys. but many useful example ~/.XCompose files online make use of 3 combining characters (in vitro w/ 2) or output multiple keys (digraphs) [15:25] that's not true. [15:29] jcristau: what do you mean? I have a .XCompose file .. its not working for me [15:29] jcristau: but like compose+t+m for ™ .. that does work [15:29] in the same file [15:32] i mean that compose sequences can output arbitrary strings, and have as many keys as you want. [15:33] jcristau: I tried setting to "ch" and it doesnt work. that same compose sequence will work for a single character like "*" asterisk .. I even tried it with dual fields like "ch" "ch" in case both fields were required [15:33] jcristau: what am I doing wrong? I'd love to have that working :) [15:34] there's already a sequence for compose c h [15:35] yeah, but I tried it for c+q too [15:35] in case it was that [15:35] jcristau: btw, where is compose c+q defined? it is not in /usr/share/X11/locale/iso8859-1/Compose [15:35] ** I mean c+h [15:45] oo lol, unbelievable, the es_EC.utf-8 doesn't map to one of the latin iso blocks, it maps to en_US.utf8 (*boggle*) [15:46] well now I know where compose c+h and compose c+c come from [15:48] why the hell would an utf8 locale map to the latin1 file.. [15:48] apparently if you define a .XCompose it does *not* load in place of the system compose file like normal, in ubuntu. (so there's no need for an "include" line at the front of .XCompose files in ubuntu) [15:48] jcristau: well, its the same as the keymap that way [15:48] symbols/es [15:49] jcristau: I admit, it would be *besT* to have an es_EC.UTF-8 and one fo rall the others, but that's a lot of files [15:52] jcristau: where would I suggest a bug fix for that? the compose.dir file even has a comment in it saying that the utf8 stuff is broken [15:54] i don't know what bug you're talking about [15:57] in /usr/share/X11/locale/compose.dir , the global compose directory, it says "Note: The UTF-8 locales don't work correctly yet. Work in progress." [15:58] the reason is because right now if you are using utf-8, generally you are asked to use the en_US.UTF-8 locale settings, which are inappropraite for the vast majority of /usr/share/X11/xkb/symbols/* files [15:59] jcristau: maybe its an incomplete feature instead of a bug, who would I talk to about how to fix it a good chunk of that? I've got a simple fix for the better part of all european locales [16:01] jcristau: not that this solves my issue with digraphs not working for .XCompose, but it solves the issue I found documented in the distribution files [16:56] in Lucid, at least in gnome apps Compose sequences can only consist of two characters, and may only output one character. This makes them exactly equivalent to dead keys. but many useful example ~/.XCompose files online make use of 3 combining characters (in vitro w/ 2) or output multiple keys (digraphs) [16:56] I just found out you can get multiple characters in qt though ! :) [17:38] : "by the way" -- works everywhere but in gnome/gtk. but other composes I set in .XCompose do work. what's going ? [17:57] hi [17:59] hi Bernardo [17:59] sup? [18:00] still fighting psb drivers... [18:01] that was before I came here [18:01] I'm still fighting gtk and xorg [18:01] :P [18:01] sadly, I want to get along with them .. why can't we all just get along?! :) [18:04] : "by the way" -- works everywhere but in gnome/gtk. but other composes I set in .XCompose do work. what's going on? [18:05] there is a bug about that somewhere [18:05] gtk bug [18:06] tjaalton: jesus is there? I saw the one aobut the xcompose file being ignored, but it's not that. the file is read but anything bigger than a dead key doesnt [18:06] or bigger than 1 character of output [18:07] at least it sounds familiar, i'll try to find it when i get home [18:09] thanks - yea this is hard to fid just searching for relavant terms like "gtk bug xcompose" or im-module .. [18:19] did you say it used to work? [18:20] I only wish the moblin/meego developers had done the sensible thing and put the driver as a separate module and not a huge kernel patch [18:21] tjaalton: no. its that it works everywhere -- EXCEPT gtk/gnome [18:21] at least now they are integrating with the Imagination Technologies ddk, so there is a hope that we'll have some decent openGL support [18:32] blue_anna: ok, then it should fit the bug aiui [18:35] tjaalton: alright -- wish I could find that bug myself -- I'm running searches like the wind here, andI can't find it [18:48] hi lucazade [18:48] hi! [18:50] Maybe we should start thinking on having a psb optimized kernel, adding the meego driver patch... :) [18:50] sounds good.. [18:51] is there a public changelog or a list of bugs fixed with this patch? [18:53] no [18:53] brb [18:53] ok [19:36] the gtk people tell me that xim is not the preferred im because it is buggy. I am getting one of those bugs with my particular set up. what is the suggested lucid alternative to xim for custom compose keys ? https://help.ubuntu.com/community/ComposeKey -- this says xim but like I said that's known not to work [19:40] blue_anna: ibus [19:40] and the bug i was referring to was different, and seems fixed (mod4 couldn't be used for shortcuts) [19:41] so i didn't understand the problem the first time ;) [19:41] tjaalton: there's a ton of them, I was just at gnome.bugzilla [19:42] if you don't limit the search to gtk+ and jsut search xim and xcompose, there are hundreds -- maybe thousands -- I dont know how they do it on volunteer manpower at all ;) [19:48] well there are >3000 bugs against the ubuntu X packages, 10000 against the kernel and almost 90000 in total (on ubuntu). I don't know if we are any better ;) [19:49] tjaalton: do you know how to get Xcompose working with ibus? there are some unanswered posts on ubuntuforums about that but no solutions [19:49] blue_anna: NO [19:49] sory [19:49] (caps) [19:49] +r [19:49] :) [19:49] nps [20:00] … [20:00] ooh, i found it :) [20:00] that character, that is [20:03] tjaalton: what character? [20:05] blue_anna: … [20:05] and 'multi minus minus minus' works here [20:05] — [20:06] so maybe it's just that your .XCompose is not read? [20:06] no well, I thought of that. [20:07] but in .Xcompose I created regular style compose keys and they do work: compose+t+m = ™ [20:07] ** ~/.XCompose [20:07] Sarvatt, did that powermizer tweak work? [20:09] blue_anna: ok [20:09] tjaalton: also in my system, those do work -- that's neat. compose minus minus minus works, didnt even know it had 3-key versions defined. [20:09] that doesn't work in qt apps [20:10] but it does work in xterm .. jeez, why does every possible library have different behavior :S [20:11] if I use ibus tables to emulate compose keys I will have to use only the ibus locale for that table ... [20:51] lucazade: I'm trying to build the xorg driver now, to see if it will work with our module [20:54] Bernardo: great.. hope you get it working! [20:55] is it difficult to build? [20:57] dependencies... They have a new lib (libwsbm), which needs some more headers we don't have in libdrm-poulsbo [20:57] The interesting part in the moblin kernel driver is that they implemented kms [20:58] so the psb_drv.h has a lot of kms related stuff [20:58] nice... [20:58] the pvr module? [20:58] no, the moblin one, psb [20:58] ok right [20:59] too much stuff :) [20:59] the meego version is the one they added the pvr module and made the psb use a lot of the pvr headers [20:59] yes, too much. They should have kept the module out of tree, like in mandriva or ubuntu [21:00] i should learn by doing.. so i've to try [21:05] I used to build a lpia kernel two years ago, when I had a aspire one, before the 1101HA [21:07] The tough part was getting the build structure right, but now there are a couple of walkthroughs out there in the web [21:08] i can imagine how hard was [21:10] if you have some hints on how to get a working build env, tell me [21:11] i'll try in a vm [21:14] brb [21:28] does anyone here understand /usr/share/X11/xkb well enough to describe how to set up a keyboard layout with two ground of 4 levels instead of the default one grounp of 4 levels? [21:29] * two groups [21:39] Bernardo i've to go [21:39] see you soon [21:49] https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/562802 => somebody here can check this ? it seems a weird issue, not happening with nouveau [21:49] Launchpad bug 562802 in firefox (Ubuntu) "xorg use cpu 100% sometime with firefox (affects: 4) (heat: 24)" [Undecided,Incomplete] [22:57] somebody alive here ? :(