Does anyone know of a general purpose input method library
which is not dependant upon anything else? By that I mean
not dependant upon X-Windows, not dependant upon a console,
not relying upon locales whatsoever, and not tied to any
specific application, and doesnt even know about fonts.
What im imagining is a single input translation layer,
which will function whether I am logged into the machine
on a text tty, in an X session, or remotely through ssh.
It would work strictly on utf-8 input/output, as a way to
map the characters directly generatable by a keyboard into
the possible unicode characters. I am imagining that an
application would simply pre-filter its keystrokes through
some API, which would be totally self-contained except
for having backends for kanji-picking, etc. also switching
between languages and modes would be handled by the api
internally, the application would just set a starting
default (perhaps gleaned from locale).
for example it would be able to map:
"Tieesng Vieejt" to "Tiếng Việt" (assuming telex)
"kusuri" to "薬" (assuming backend available)
these are unsuitable:
-mule:
tied to emacs, knows about fonts to much
-XIM:
tied to locales and X windows
-console keymaps
-xkbmaps
It seems like so much work has already been done on input
methods, that someone must done something similar already...
�{�Q1|�_'����h��e��ޖ�+r�zm���j)g�X��+��b��n��?
