[00:02] attente, bschaefer: Just read your conversation with interest. A few hours ago I dropped this comment to seb128: [00:02] "I have the impression that the im-config control in language-selector is now a no-op in Ubuntu. If that's the case, I suppose it should be hidden. At the same time, I assume that the im-config control still makes sense in Xubuntu and Lubuntu, and if so we should keep showing it in those derivatives." [00:02] That side of the problem seems to be related (to some extent) to your discussion. [00:05] GunnarHj, im-config still does stuff with upstart atm, but possibly it doesn't need to be there [00:06] and should be removed, and we'll need to take this into consideration on how its used else where...as we don't want to force ibus for all of ubuntu [00:07] one thing mentioned was, if you install fcitx, then upstart will see that and use fcitx as the default [00:07] the problem is, g-s-d is start ibus-daemon each time which fixes a big bug but causes some other problems [00:09] bschaefer: Well, I haven't tested yet, so my remark above i only a theory yet. Yes, I understand that the whole issue is about a conflict between the way g-c-c and g-s-d deals with input methods and "the old way" to do it. [00:10] GunnarHj, well right now im-config trumps g-s-d, as upstart looks at im-config, if im-config is set to say ibus, g-s-d doesn't do anything crazy [00:10] bschaefer: Right. But what if im-config is set so something else? [00:10] im more concerned about what happens when im-config is set to a different IMe, which with ibus it has the --replace which might trump over fcitx (which wont be good) [00:11] right, i was going to test that soon :) [00:11] bschaefer: Me to. ;-) [00:11] though, fcitx will get ran first through upstart, but ill have to see what happens... :), the more testing the better! [00:13] bschaefer: IIRC, im-config sets fcitx if it's installed *and* Chinese is the current display language. [00:15] hmm soo fcitx runs, as well as ibus... so they are both running [00:15] though with the ibus-daemon set to --panel disable it just doesn't work [00:15] at lease fcitx works [00:17] * bschaefer assumes some wont be happy that the ibus-daemon is running when fcitx is set to default and running === m_conley is now known as m_conley_away [00:56] bschaefer: Looks like I was wrong. After having Installed fcitx and fcitx-sunpinyin and selected fcitx in language-selector, I can successfully create those beautiful characters by help of fcitx-sunpinyin. Also the fcitx icon shows up. So the ability to set the input method framework in language-selector still applies in Unity. [00:58] GunnarHj, right, but ibus is still running [00:58] with fcitx [00:58] which it really shouldn't be... [00:59] bschaefer: Yes it is. But does it hurt? And if it didn't run, would the Text Entry window in g-c-c make any sense? [01:00] not so much, and that bug would still be around, the problem is...if you've selected xim [01:01] then want to us ibus [01:01] bschaefer: What then? [01:01] use* [01:01] bschaefer: Well, if you want to use ibus, why on earth would you select xim? [01:01] GunnarHj, the problem is that dbus command makes ibus not work, so the only way to get ibus to work if your im-config -n xim [01:02] GunnarHj, cause ibus still works with xim :) [01:02] if it is your XMOD* is set to none, which mean upstart doesn't start ibus, then the dbus call starts up a non working ibus daemon [01:05] bschaefer: Hmm... The "none" option in language-selector actually sets xim. Is that possibly a problem then? [01:06] bschaefer: Of is it maybe a good thing? [01:06] Or [01:06] GunnarHj, well XIM isn't an input engine [01:06] its an extra layer on top that most all input engines work through [01:11] bschaefer: In any case, if you want to use ibus, is there any reason not to select ibus? [01:13] GunnarHj, well we are attempting to assume the user does not know about im-config [01:13] if they don't...then it'll fun to try and get ibus working when you select chinese in the text entry [01:13] and only english is coming out [01:14] GunnarHj, would you mind testing that? Set the text entry to a chinese language and have im-config -n xim [01:14] which means g-s-d is starting the ibus-daemon (with options that makes ibus not work for me) [01:14] then I try to use chinese/pinyin and i only get english [01:15] bschaefer: Sure, I can try. Logging out first and get back in a few minutes. [01:16] GunnarHj, thank you [01:28] bschaefer: You are right, I couldn't use Chinese/Pinyin. Checking the env variables: [01:28] $ env | grep IM_ [01:28] CLUTTER_IM_MODULE=xim [01:28] IM_CONFIG_PHASE=1 [01:28] QT4_IM_MODULE=xim [01:28] GTK_IM_MODULE=xim [01:28] echo $XMODIFIERS [01:28] as well [01:29] GunnarHj, its not the problem with whats set, its the problem with the dbus activation [01:29] GunnarHj, do a "ps aux | grep ibus-daemon" [01:29] you'll see the --disable panel [01:35] bschaefer: But I'm still confused. The default option in language-selector is "default", which means that dbus/im-config starts ibus if only ibus is installed and one of the CJKV languages is the display language, or else dbus doesn't run im-config at all (but g-s-d does). In both cases it works for me. [01:36] GunnarHj, in both cases ibus/pinyin works for you? [01:36] bschaefer: Yes. Hmm... Maybe I should try again to confirm it... Getting back soon. [01:46] RAOF: hey ... i somehow have no X anymore on latest saucy [01:46] RAOF: not sure what happened, but i hoped you can help [01:46] * asac types from a console [01:46] asac: Hm. Let me see what changed while I was away... [01:46] RAOF: the lightdm upstart log says: [01:47] seat: cant create unsupported display server 'x' [01:47] i removed the xorg.conf.failsafe a few times [01:47] that didnt help [01:47] That's odd. [01:47] What's in /etc/lightdm/lightdm.conf.d? [01:48] 10-ubuntu.con [01:48] f [01:48] 50-greeter-wrapper.conf [01:48] 50-ubuntu.con [01:48] 50-unity-greeter.conf [01:48] 50-xserver-command.conf [01:48] 52-ubuntu-touch.conf... [01:48] hmm [01:48] whats that :)_ [01:49] wow ... that one surely has to go away :)? [01:49] RAOF: ^^ [01:49] I note that *I* don't have that config file. [01:49] wow [01:50] that helped [01:50] wtf :) [01:50] Where did that come from? [01:50] not sure... 3 month ago i had accidentially the phablet ppa enabled [01:50] however, ... the system was cleaned from that incident [01:50] and it worked since then for many reboots... so not sure [01:51] ubuntu-touch-session [01:51] wonder howthat can be instlaled [01:51] * asac purges that [01:52] Maybe lightdm got updated past the point that the config file started to take effect? [01:52] RAOF: thanks so much... i doubt thats anything you need to be worried about... except maybe how it ended up getting bad just now [01:52] yeah i assume so. thx [01:52] * asac goes to X rather than console [01:53] something is very odd [01:53] all looks different than before... but well, its unity running, so why would i complain :) [01:54] RAOF: thx! [01:54] Glad to be of service! [02:03] bschaefer: Yes, I can produce those beatiful characters using ibus/pinyin also when dbus doesn't run im-config at all, i.e. when "default" is selected in language selector and my display language is English. But when I select "ibus", and the proper env variables for ibus are set, a help window is shown which I don't see when "default" is selected. Also, to see the ibus/pinyin icon, "ibus" needs to be selected. [02:04] bschaefer: But this should really be tested by an im user, shouldn't it? [02:05] GunnarHj, whats your im-config set to? [02:06] GunnarHj, and when you say you select ibus, do you mean in that language app? [02:06] umm [02:06] bschaefer: What do you mean by "set to"? [02:06] language support app? [02:07] GunnarHj, i mean, if you type im-config and the gui pops up [02:07] whats the active IM? [02:07] or do: echo $XMODIFIERS [02:07] and let me know what it says... [02:09] bschaefer: When "default" is selected in l-s, the active IM is default, and when "IBus" is selected in l-s, the active IM is ibus. [02:10] GunnarHj, im not sure what language selector is then :) [02:11] bschaefer: Or when "default" is selected, there is no ~/.xinputrc file, while when "IBus" is selected, there is a ~/.xinputrc file saying "run_im ibus". [02:11] bschaefer: The IM selector in language-selector just sets the active IM. [02:12] GunnarHj, well the overall thing is to look how ibus-daemon is started [02:12] ps aux | grep ibus-daemon [02:12] will tell you the options on it [02:12] if it has --panel disable then its not going to work :), which if im-config is not set to ibus seems to always be on for it ... [02:16] bschaefer: I was the one who prepared im-config for Ubuntu. I chose to make it default to "cjkv mode" to avoid that ibus-daemon is started for everyone, also those users who don't know what an input method is. Now I wonder: What if we would change that, and make "ibus" the default for im-config? [02:17] GunnarHj, then all will work, but some don't want ibus-daemon to always be on [02:17] bschaefer: I mean, ibus-daemon is always started anyway nowadays. [02:17] GunnarHj, right, which I think thats a bug? [02:17] GunnarHj, but if im-config -n ibus was set then we would be all fine [02:18] bschaefer: That would be a simple thing that even might be possible to fix before the release... [02:19] very true, and i don't think it would cause any problems...hmm [02:19] bschaefer: Not saying that it's an ideal solution, but given the ibus integration in g-c-c, it may prevent user confusion. [02:20] GunnarHj, yeah, its better then what we have now [02:20] kylin want fcitx then it wont be a problem there either [02:21] as they should be able to set that them selfs (which you hope they do...) [02:21] bschaefer: I'm not sure how the achieve the fcitx default. [02:21] they [02:22] GunnarHj, if we actually set ibus to default as well we can drop the dbus activation thing with out causing that bug/crash? Along with not always having the ibus daemon [02:22] GunnarHj, hmm couldn't they have a patch? I mean how to they set fcitx to default atm anyway? [02:22] how do* [02:22] bschaefer: Don't know. [02:22] * bschaefer isn't super how they version of ubuntu works... [02:22] GunnarHj, but I think thats our best solution atm [02:23] bschaefer: It's over 4 a.m. here now... But I can file an im-config bug tomorrow, and involve e.g. happyaron in it. [02:23] GunnarHj, o geez! Sorry for keeping you up so late! [02:24] GunnarHj, thanks a ton! Have a good night! [02:24] its only 7:24 pm here :) [02:24] bschaefer: My choice; no problem. ;-) [02:24] GunnarHj, cya! [02:24] bschaefer: Good night to you as well. [02:25] thanks :) [05:30] Good morning [05:31] salut jibel [05:31] jibel: getting up earlier than few weeks ago? [05:32] Salut didrocks, ça va? [05:32] didrocks, yes, I'm starting earlier to finish my day earlier :) [05:32] heh :) [05:32] ça va bien, et toi? [05:33] and I realized there are living creatures outside ;) [05:33] didrocks, ça va ça va [05:35] ahah, really, living creatures outside? [08:05] hey! [08:11] hey Laney, hey desktopers [08:11] how is everyone? [08:12] great, good climbing session last night! [08:12] you? [08:13] I'm good thanks [08:15] hey seb128, Laney! [08:15] didrocks, lut, en forme ? [08:15] ça va, et toi? [08:16] nickel [08:16] hey didrocks [09:17] Laney, do you if the fact that you can't go to the settings list when you open "system-settings " is a feature or a bug? [09:18] pretty sure it was on purpose [09:18] so you can't go "back" confusingly [09:18] we had it that way originally [09:19] I think we should add it back [09:19] the current situation is a bit buggy [09:19] if you open a specific panel from an indicator [09:19] you get screwed until you force close the app [09:20] how would being able to go to the index help? [09:20] e.g battery indicator ->settings [09:20] go back to the dash [09:20] search for settings [09:20] click on it [09:20] it refocus the running instance open on battery [09:20] that should send the app back to the main page [09:20] ok, that's one other way to fix it [09:20] it'd also break if you then try to open another panel from a different indicator I suppose [09:20] though I'm not sure we are in control of what is happening there [09:21] I think it's upstart-app-launch doing the single instance [09:22] no, opening from a different panel works [09:22] is that because single instance is broken? [09:22] it doesn't seem so, I've only one preview in my recent apps list [09:23] it must kill the old one then [09:23] it seems it do [09:23] "upstart-app-launch ubuntu-system-settings" on my desktop closes the old one [09:24] you should talk to unity people [09:24] yeah, I'm going to talk to ted when he gets online [09:24] he's the one who has been working on the app launch stuff [09:24] Laney, thanks [09:29] phablet@ubuntu-phablet:~$ date [09:29] Thu Oct 3 10:29:08 BST 2013 [09:29] yay [09:29] up to date! [09:29] working and persistent? [09:29] Laney, I opened https://bugs.launchpad.net/ubuntu/+source/upstart-app-launch/+bug/1234588 [09:29] Launchpad bug 1234588 in ubuntu-system-settings (Ubuntu) "Can't go back to the summary after opening a specific subpanel" [High,Confirmed] [09:29] it updated /etc/timezone which I think was missing before [09:29] err I mean localtime [09:30] * Laney reboots [09:30] good [09:30] it's still not a symlink on first boot though [09:30] yeah, persisted [09:33] \o/ [09:34] shrug, why is firefox not remembering my google login, I've to log in again every time I reopen it [09:36] seb128: keyloggers need more chances on catching your password [09:45] Mirv: is gitorious the place to submit qtsystems patches? [09:50] Laney, I think it's on https://bugreports.qt-project.org [09:51] hmm [09:51] don't see qtsystems in the projects tehre [09:51] Laney, https://bugreports.qt-project.org/browse/QTBUG-31624 [09:51] " Systems: SystemInformation " it seems [09:51] (that's one I had in my awesome bar) [09:52] Laney, https://wiki.canonical.com/UbuntuEngineering/ContributingToQt btw [09:52] Laney: either bugzilla or a straight codereview request to gitorious (needs certain setupping) [09:54] yeah I don't know what applies to qtsystems [09:54] that seems to be a separate project [09:55] Laney: the same https://codereview.qt-project.org/#q,status:open+project:qt/qtsystems,n,z [09:55] aha [09:56] ok, doing that then [09:56] better test the patch first though ... [10:09] Mirv: could you add me to the group please? [10:10] Laney: done, should be visible now [10:11] it is, thanks [10:33] Laney, remember that problem I had a coupleof weeks back, where I'd login and get dumped back to the greeter? [10:34] And bcurtiswx had the same problem - well, I've got the same thing happening again now. I can get to a desktop in the guest session, but not using my proper account. [10:37] Nothing has been touched in ~/.cache/upstart - the datestamps are an hour old. [10:37] JamesTait: Try moving ~/.Xauthority [10:39] Laney, OK, giving that a go... [10:39] Laney, success! Thanks! [10:40] Laney, is this a known problem with a bug report I can contribute to? [10:41] It's fixed in lightdm already, just not in saucy yet [10:41] Excellent! :) [10:41] I'll be getting on with my work now then. :) Thanks again! === alf|xmir_devel is now known as alf|lunch === alf|lunch is now known as alf_ [13:42] chrisccoulson, hey, how are you? [13:43] chrisccoulson, do you know if that known issue with the current firefox and "session handling" [13:43] chrisccoulson, it keeps forgetting my google session (e.g I've to re-auth every day) and I often gets the "didn't close properly, do you want to restore the tabs" screen when simply closing and reopening a bit later [13:43] chrisccoulson, it looks like some stuff wouldn't get saved properly or would get corrupted [13:44] seb128, i haven't noticed anything like that before [13:46] chrisccoulson, hum, ok :/ [13:51] seb128, file permissions are ok? [13:54] chrisccoulson, from what I know yes, on what file should I check? [13:55] *** WARN addons.xpi: Exception running bootstrap method shutdown on webapps-team@lists.launchpad.net: [Exception... "Component returned failure code: 0x80004001 (NS_ERROR_NOT_IMPLEMENTED) [nsIComponentRegistrar.autoUnregister]" nsresult: "0x80004001 (NS_ERROR_NOT_IMPLEMENTED)" [13:55] I wonder if that's another webapp issue [13:55] the addon is turned off though [13:57] brb, session restart === psivaa is now known as psivaa-afk === mterry_ is now known as mterry [14:03] oops, got disconnected [14:03] attente, (assuming you didn't see my previous poke) What can we do to speed up landing of indicator-keyboard branches/package? [14:04] attente, is indicator-stack on manual saucy landing? [14:05] mterry, i'll try to find someone to review it === psivaa-afk is now known as psivaa === m_conley_away is now known as m_conley === mpt_ is now known as mpt [15:25] Laney, bah, there are bugs in your MR adding TRANSLATORS comments, that will teach me to just glance at those rather than do a proper review :p [15:25] NO! [15:25] Laney, http://bazaar.launchpad.net/~system-settings-touch/ubuntu-system-settings/trunk/revision/409/plugins/system-update/PageComponent.qml [15:26] Laney, see the 2 first chunks [15:26] Laney, (I'm just pushing a fix) [15:26] ahaha [15:26] the first one seems a middle click error [15:26] not sure what happened to the else [15:26] the first one is probably that stupid qt creator behaviour [15:26] the completion on alt-tab? [15:26] yes [15:26] yeah, I hate that one :/ [15:27] other one is general failure i guess [15:29] Laney, btw, did you get a reply for ev about whoopsie on touch? [15:29] oh, no [15:30] I forgot what the specific question was [15:30] I saw him talk on IRC yesterday and today [15:30] Laney, I think it was about enabling whoopsie from the security panel [15:30] in context of logind/polkit/ro? [15:30] oh yes [15:31] Laney, https://code.launchpad.net/~seb128/ubuntu-system-settings/update-working-check-update/+merge/189118 [15:32] pinged him in devel [15:34] Laney, thanks [15:46] is there some problem with screen blanking / unlocking in desktop atm? [15:47] when unlocking I see the contents of the display instead of the wallpaper on one of the monitors [15:51] I ran into that a few time as well [15:51] I'm sure mdeslaur doesn't like it [15:51] give access to whatever is under the lock for a second [15:51] I wonder if that's an xorg thing (e.g wrong buffer on screen or something) [15:53] don't know [15:53] mlankhorst: what do you think about this xorg theory? :P [15:54] it doesn't seem to happen when locking explicitly [15:54] only after the timeout [16:01] Laney, it's not easy to reproduce in any case [16:01] but I've a 2 screens setup, and I just had a case where I could see my right screen "unlocked" with the unlock dialog on the left [16:02] e.g I had xchat showing on one in the screens during the lock [16:02] yeah [16:02] was it updating? [16:02] or just what was there when it locked [16:02] I don't think so [16:02] it was a buffer of the lock time I think [16:12] merge ALL the things!! [16:12] * desrt loves that start-of-the-cycle feel [16:31] desrt, you live in the futur it seems, come back to real time, we still have some bug fix fun before getting there ;-) [16:32] * desrt should have kept his mouth shut [16:32] uh.... lunchtime! [16:32] * desrt runs [16:33] ;-) [16:35] seb128, hey [16:36] bschaefer, hello [16:36] seb128, soo something interesting about the ibus problem in g-s-d [16:37] seb128, if you're not aware...if g-s-d starts up ibus, CJK fails to work unless you restart the ibus-daemon with correct options [16:37] seb128, soo GunnarHj proposed this bug: https://bugs.launchpad.net/ubuntu/+source/im-config/+bug/1234768 [16:37] Launchpad bug 1234768 in im-config (Ubuntu) "Consider a change of the im-config default" [High,New] [16:38] happyaron, attente: ^ do you know what's going there [16:38] seb128, hey. any known issues with power saving on desktop? i recently switched back to xorg (after beta testing xmir for a long time), and now my screens never power down. they were on all last night [16:38] bschaefer, I'm pretty useless when it comes to understand CJK and what role im-config plays there [16:39] seb128, no worries :), thanks! [16:39] robru, not that I know, it could be a gnome-settings-daemon issue... what value do you have in g-c-c for "turn screen off if unactive for..." [16:39] seb128: it would be good to revert the triggerred startup of im-config, and revert im-config behavior to auto [16:39] this will let ibus running for everyone, but it will resolve all the headaches [16:39] happyaron, "triggered startup of im-config"? [16:40] * didrocks waves good evening and see you next thursday! [16:40] where did we add that? [16:40] (or before for some of you ;)) [16:40] didrocks, night, good luck for the next days, see you on saturday [16:40] seb128: I was once told it's started using upstart user session [16:40] see you on saturday seb128! [16:40] * happyaron on holidays right now, but anyway, ;-) [16:41] seb128, 10 minutes [16:41] upstart starts the selection of what im-config is set to, and +1 from me to setting it to auto [16:41] I short revert to upstream behavior is a working choice, but it's not cool... [16:41] happyaron, oh, I didn't know, enjoy! [16:41] :) [16:42] bschaefer, happyaron, attente: you guy just tell me what you think is best (or comment on https://bugs.launchpad.net/ubuntu/+source/im-config/+bug/1234768) and we can make that happen [16:42] Launchpad bug 1234768 in im-config (Ubuntu) "Consider a change of the im-config default" [High,New] [16:42] seb128, alright sounds good [16:43] bschaefer: I'm not sure if the upstart handled startup would affect users, but I think it's started when Xsession is run anyway. [16:44] happyaron, that could be it as well, I only started digging into upstart/im-config last week :) [16:47] bschaefer: In my experiment, Xsession runs im-config, and it works for cjkv, others run im-config when they click on something that triggers it, which will make indicator-keyboard not working. [16:47] if we switch to auto for default, triggered activation is not needed at all. [16:49] true, which is nice, the problem i was running into is the ibus-daemon was being started with --panel disable which was causing the ibus-daemon to not work at all [16:49] happyaron, but in either case, go enjoy your time off :) [16:50] :) [16:53] bschaefer, can't we just make g-s-d not drop --panel? or would that be wrong (e.g make an indicator appear for everyone not using the feature)? [16:54] seb128, at that point we might as well just set im-config to ibus or auto [16:54] same question [16:54] would that make everyone get an indicator? [16:54] yes [16:54] I think auto is better than ibus, though. [16:55] seb128, im starting to think it could be a possible bug in ibus... [16:55] happyaron, agreed [16:55] we don't want an ibus icon for users not have an im need/configured [16:56] * bschaefer wonders if the k-indicator is just stealing the super+space shortcut from ibus if --panel disable... [16:57] seb128, true, but the ibus daemon is already running anyway... [16:57] well, that's a problem by itself [16:57] (use resources for nothing) [16:58] right, hmm, its just when the ibus-daemon has --panel disable you can't actually use ibus at all (but its running!) [16:58] so if you were to go into ibus-setup, or even in the text-entry and add pinyin/chinese it wouldn't work [16:59] at lease from my testing... [17:00] bschaefer, why would ibus have a mode where it's not usable?/why would we call it in this mode? [17:01] seb128, that im not sure of...but thats what it seems to be in when im-config is set to none, and g-s-d starts it [17:01] soo this could very well be the overall issuse that needs to be addressed... [17:01] seb128, so right now, you have a default set up im wise? [17:02] and echo $XMODIFIERS [17:02] says none right? [17:02] $ echo $XMODIFIERS [17:02] $ [17:02] I've a "default set up" [17:02] I'm using french and I've added a GB layout [17:03] e.g just standard layout, no im [17:03] seb128, that seems good, and umm ps aux | grep ibus-daemon shows a --panel disable option? [17:03] seb128, if so, add chinese/pinyin to text entry [17:03] and you would think that since ibus-daemon is started, that it should work right? [17:05] seb128: that mode makes other stuff manages IM [17:05] happyaron, what is "other stuff"? g-s-d? [17:06] seb128: for people who'd like to put environmental variables in xsession start sequence [17:06] seb128: when im-config detects any of those variables are set it quits and do not set any of them. [17:07] don't you have to reboot though? I though im-config was only taken into consideration on a xsession restart? [17:07] thought* [17:08] * bschaefer still gets --panel disable when pinyin is set in the text entry...causing it to fail [17:08] though i should double check that [17:08] * bschaefer relogs [17:08] anytime a new Xsession is started im-config tries to do everything again. [17:08] right, i thought you were saying in the middle of session it would quit and try to do that again [17:09] * bschaefer mis read [17:09] (or be precise, everytime Xsession is executed) [17:11] soo i've chinese set in text entry which doesn't seem to be reflecting in ibus-setup.. ive im-config set to none [17:11] which is default (or so it seems?), and im not able to type in pinyin [17:12] bschaefer: mind run `env | grep IM` ? [17:12] /usr/bin/ibus-daemon --replace --xim --panel disable is started, as it doesn't detect the i've chinses set in my text entry [17:12] yeah [17:13] IM_CONFIG_PHASE=1 [17:13] which seems to be all... [17:14] what about `env | grep XMODIFIERS` ? [17:14] happyaron, XMO is not set, since im on none [17:15] im-config -n none [17:15] then it won't work... [17:15] happyaron, right, but thats the default... [17:16] im-config by default with an english layout to set it to none right? [17:16] that's a ubuntu default, to get rid of starting ibus for non-cjkv locales. [17:16] debian default is auto for all. [17:16] happyaron, right, but then I select a chinese language in my text entry...then reboot and its still none, and now chinese doesn't work :) [17:16] happyaron, yeah, this is why I would like auto, this means that the ibus panel will be there though [17:17] which is something seb128 doesn't want [17:17] bschaefer: but g-s-d cannot change those variables if I'm correct [17:18] when g-s-d is started, it's late to set those variables for the desktop session. [17:18] nope, but g-s-d is starting an ibus-daemon that makes it not work if the usr wants chinese... [17:18] i don't think it can change it [17:18] so you added a text entry, but nothing is working to set variables... [17:18] yup, and it still thinks you are a non cjkv user [17:18] so it starts ibus-daemon with a --panel disable [17:19] which means you cannot user CJK languages with out setting im-config your self or killing and restarting the ibus daemon [17:19] happyaron, so theres really 2 arguments, 1 we don't want the ibus panel if we don't use ibus, bug g-s-d need ibus to be running [17:19] don't think restarting ibus daemon will make Chinese input work, though other stuff may work (in g-c-c). [17:20] well it makes ibus work again at lease [17:20] happyaron, its a fun workaround that doesn't always work on some autopilot tests in unity atm...but i would like a more correct fix for this :) [17:21] so how do we ensure ibus works if we want to type chinese, but by default dont want to start the panel? [17:21]