> Now I've installed the emacs 21.4,
Uh, oh, I've no idea whether my input method works with 21.4 at all!
Please use the current CVS of emacs. This is *far* better in all
respects than emacs 21.4.
> and then put the 4c-CNS.el.bz2 into my load path
Note that the file's encoding (EUC-TW) can't be handled with emacs
21.4. I've intentionally selected EUC-TW because the original one
(emacs-mule) isn't portable.
> (load-library "cjk-enc")
This is OK.
> (load-library "4c-CNS")
This doesn't work. To register the file as an input encoding, you
have to add the following code:
(register-input-method
"chinese-cns-4c" "Chinese-CNS" 'quail-use-package
"4c-CNS" "four corner CNS"
"4c-CNS")
> (autoload 'CJK-insert-space "cjkspace"
> "Insert tildes appropriately in CJK document." t)
> (defun cjk-LaTeX-mode-hook ()
> "CJK key definitions for LaTeX mode."
> (define-key LaTeX-mode-map " " 'CJK-insert-space))
> (add-hook 'LaTeX-mode-hook 'cjk-LaTeX-mode-hook)
This is fine. However, you are missing code to integrate cjk-enc into
the AUCTeX frame work (you need this anyway in case you are working
with preview-latex). Please following the documentation in cjk-enc.txt.
> When I use the emacs to open my tex files, I can.t see the big5
> chinese characters, and cjk-ext B charaters.
Well, even Emacs 22 won't have support for cjk-ext B characters. They
are preserved as correct Unicode values but are not displayable. You
probably should experiment with the unicode2 branch of Emacs (I don't
use this -- look up the emails in the emacs-devel list for more
details and how to check it out from the CVS). After the release of
Emacs 22 this unicode branch will be the basis for Emacs 23.
A good Unicode editor appears to be `mined'. I haven't tried it by
myself, but according to the announcement the level of Unicode support
seems to be quite astonishing.
> I want to make the emacs select encoding scheme according to my
> tex files automatically, that is, when my tex files are encoded with
> GBK, BIG5, GB and UTF8, the emacs should load the corresponding
> encoding character sets and display them correctly.
This is no problem, basically. Just insert a proper `coding:' file
variable, for example, as I've done in the 4c-CNS.el file I've sent to
you. Without a tag you can manually select the file's input encoding
during while visiting the file.
> Furthmore, your 4c-CNS.el.bz2 is for big5,
Not exactly. It is for CNS 11643, planes 1-7. It just uses big5
instead of planes 1 and 2 of CNS 11643 because it more common (and the
available bitmap fonts look better).
> how can I use it for GBK or UTF8, and as for the input method, can I
> use the more common, say, SCIM, to input the cjk-ext B characters
> with emacs?
`iconv' is your friend... You have to convert the euc-tw encoded file
to something you can use (this is, Unicode)! Then you have to write
your input encodings by yourself. I can't help you further. Be
careful to get an iconv version (or something similar) which is recent
enough to support cjk-ext A and B Unicode characters.
Werner
_______________________________________________
Cjk maillist - [email protected]
http://lists.ffii.org/mailman/listinfo/cjk
