Re: gnu emacs w32 & unicode

[Edward Casey]

[Note: view with UTF-8 encoding]

<jasonr (Jason Rumney) @ f2s.com> wrote in message
uwu1qyf6i.fsf@jasonrumney.net">news:uwu1qyf6i.fsf@jasonrumney.net...
> "Edward Casey" <ejmn@cpinternet.com> writes:
>
> > Has anyone out there using i386-mingw (21.3.1) successfully input
unicode
> > characters into buffers and saved them to files? I can see the Latin.4
> > characters in the read only "List Character Sets" buffer, and can
> > copypaste them into *scratch*, but I can't seem to produce them
whatever
> > combination of language environment, input methods, and coding systems
I
> > use. I suspect a font problem but I have about 200 different TT fonts
in
> > \windows\fonts and I know many of them have the glyphs for Latin-4
> > (Extended A), certainly Arial Unicode MS and Courier. Can anyone give
me a
> > fail-safe recipe to input these characters?
>
> C-u C-\ latin-4-postfix
>
> Then type the combination of characters that is needed to compose the
> character you want (eg o-)
>
> But that won't give you Unicode. For that, you will need Emacs from
> CVS AFAIK.
>

Thanks, Jason. That works. Now I can even get the characters "āēīōūĀĒĪŌŪ"
(should be macroned vowels under UTF-8 encoding) out to a UTF-8 file. This
I can get into MSWord but not directly into OpenOffice. I would like to
avoid non-copylefted software. The mule status to accomplish this is:
Multi-byte status "default t, current buffer t,"terminal coding
"emacs-mule," language environment "Latin-4," and, as you say, default
input method "latin-4-postfix." Now I'm wondering if running the Lisp
routine that is the goal of all this setting up of the environment will be
possible. Is a form like:

((eq c ?A) (delete-char) (insert ?Ā))

a legal lisp expression or is it necessary to use some escape sequence
with hex codes to represent a litteral unicode character outside the ascii
range? Anyway,

Thanks again,

Ed