// uim-svn1643, LANG=ja_JP, XMODIFIERS=@im=uim a) immodule $ gedit $ uim-im-switcher It shows all CJK and m17n-* engines. b) XIM $ kedit $ uim-im-switcher It only shows Japanese engines.
This is not a bug. If you use XIM client with UTF-8 locale (e.g. ja_JP.UTF-8), you can get all available IM with uim-im-switcher. If you use XIM client with legacy encoding (e.g. ja_JP.eucJP), available IM are restricted to IM with correspoinding language and "*" since the client cannot show characters in other languages.
> since the client cannot show characters in other languages. Then uim-immodule should restrict other languages too. KWrite can manage UTF-8 encodings on ja_JP.eucJP, so I think KWrite can show UTF-8 characters on ja_JP.eucJP.
You should distinguish underlining mechanism between XIM and im-modules. With im-modules, string inputted by uim is directly passed to GUI library (GTK+ or Qt) with UTF-8 encode. So it doesn't matter wether application is working with UTF-8 locale or legacy locale. With XIM, string inputted by uim is transfered with CompoundText into Xlib, and Xlib convert this text to multibyte string and pass it to XIM client application. So if user uses clients with legacy encoding, only restricted characters can be available.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.