Created attachment 524319 [details] Screenshot for Bengali Description of problem: after enabling keyboard layout from ibus for Bengali Language, it is showing both layout with same name. Benagli and Benagli Version-Release number of selected component (if applicable): xkeyboard-config-2.3-2.fc16.noarch ibus-1.3.99.20110817-5.fc16.x86_64 ibus-m17n-1.3.3-4.fc16.x86_64 How reproducible: everytime Steps to Reproduce: 1. add Bengali Keymap from Ibus Preference 2. there are two Bengali maps with same name 3. Actual results: there are two Bengali maps with same name Expected results: there is same difference between those two, so that user can make choice Additional info:
IBus loads /usr/share/X11/xkb/rules/evdev.xml and there are duplicated 'Bengali' in evdev.xml. <layout> <configItem> <name>bd</name> <shortDescription>bn</shortDescription> <description>Bengali</description> <languageList> <iso639Id>ben</iso639Id> </languageList> </configItem> ... ... <layout> <configItem> <name>in</name> <shortDescription>in</shortDescription> <description>Indian</description> </configItem> <variantList> <variant> <configItem> <name>ben</name> <shortDescription>bn</shortDescription> <description>Bengali</description> <languageList> <iso639Id>ben</iso639Id> </languageList> </configItem> </variant> <variant> I'd like to suggest to change the description in layout 'in' and variant 'ben' to distinguish layout 'bd'. Transferring to xkeyboard-config.
upstreamed here: https://bugs.freedesktop.org/show_bug.cgi?id=41204
From the upstream maintainer Sergey V. Udaltsov: "Actually, there was no promise that descriptions are going to be unique. But you could come up with the good different names - we could discuss that."
I don't mind any names. How about the following change? --- /usr/share/X11/xkb/rules/evdev.xml.orig +++ /usr/share/X11/xkb/rules/evdev.xml @@ -1714,7 +1714,7 @@ <configItem> <name>ben</name> <shortDescription>bn</shortDescription> - <description>Bengali</description> + <description>Indian (Bengali)</description> <languageList> <iso639Id>ben</iso639Id> </languageList> @@ -1724,7 +1724,7 @@ <configItem> <name>ben_probhat</name> <shortDescription>bn</shortDescription> - <description>Bengali (Probhat)</description> + <description>Indian (Bengali (Probhat))</description> <languageList> <iso639Id>ben</iso639Id> </languageList>
hmm, not a big fan of the double-nested parens. I just noticed that for German (Austria) and English (US) the style is to use a comma for variants so I think the right way here would be Indian (Bengali) and Indian (Bengali, probhat). I'll suggest that to svu upstream.
(In reply to comment #5) > hmm, not a big fan of the double-nested parens. I just noticed that for German > (Austria) and English (US) the style is to use a comma for variants so I think > the right way here would be Indian (Bengali) and Indian (Bengali, probhat). > I'll suggest that to svu upstream. Better it can "Bengali" and "Bengali Probhat" (skip Indian as script also used in Bangladesh)
Runa, Any suggestion regarding this?
(In reply to comment #6) > (In reply to comment #5) > > hmm, not a big fan of the double-nested parens. I just noticed that for German > > (Austria) and English (US) the style is to use a comma for variants so I think > > the right way here would be Indian (Bengali) and Indian (Bengali, probhat). > > I'll suggest that to svu upstream. > > Better it can "Bengali" and "Bengali Probhat" (skip Indian as script also used > in Bangladesh) The problem is the duplicated names. The "Bengali" is already used in layout 'bd' and the suggestion is for layout 'in', variant 'bn'. Probably it would be good to check the upstream bug. Personally I agree with the upstream suggestion. I can find 'Indian' for layout 'in' only in /usr/share/X11/xkb/rules/evdev.xml .
If we have to strictly distinguish based upon the 'in' and 'bd' groups, I'd have to second comment #5 here.
from a quick check, the in(ben) and bd layouts are different so a different name is certainly important here. You've probably seen the upstream comment, instead of what i suggested in comment #5 it should be "Bengali (India)" instead. Does that sound alright?
Given the regulations, the upstream suggestion sounds good to go. I really don't see anything else that we can use for the xkb maps.
Yes, I also agree with the upstream suggestion.
fwiw, this has been fixed since xkeyboard-config 2.4.1 which was somewhen in the F14 timeframe. Closing.