[mew-int 01218] Re: iso-8859-15 (aka Latin-0 or Latin-9)

[Tatsuya Kinoshita]

On December 12, 2002, [mew-int 01217],
Kazu Yamamoto <kazu at example.com> wrote:

> I have implemented a translation mechanism between Latin 0 and Latin 1
> in Draft mode.
> 
> Please test this on Emacs 21 (not Emacs 20) with anon CVS version of
> Mew.

That's good.  But, I found some problems:

(1) When I set mew-charset-latin to "iso-8859-15", I reply mail
    which has iso-8859-1's CURRENCY SIGN, and I put iso-8859-15's
    characters, then Mew uses mew-charset-m17n (utf-8 or
    iso-2022-jp-2) even if I put iso-8859-1 compatible characters
    (e.g. A WITH GRAVE).  I hope that Mew uses iso-8859-1.

    Similarly, when I set mew-charset-latin to "iso-8859-1", I reply
    mail which has iso-8859-15's EURO SIGN, and I put iso-8859-1's
    characters, then Mew uses mew-charset-m17n (utf-8 or
    iso-2022-jp-2) even if I put iso-8859-15 compatible characters
    (e.g. A WITH GRAVE).  I hope that Mew uses iso-8859-15.

(2) The translation of other iso-8859 charsets (e.g. iso-8859-2) are
    not yet supported.

(3) When I set mew-charset-latin to "iso-8859-15", I reply mail
    which has iso-8859-15's EURO SIGN, and I put iso-8859-15's
    characters, then Mew uses mew-charset-m17n (utf-8 or
    iso-2022-jp-2) if ucs-tables.el is used.  I hope that Mew
    uses iso-8859-15.

I tell you about ucs-tables.el.  GNU Emacs 21.3 and Guns include
it.  You can get it by CVS.  

ucs-tables.el decodes iso-8859 characters into latin-iso8859-1
or mule-unicode-0100-24ff when `unify-8859-on-decoding-mode' is
enabled.

For example, ucs-tables.el expresses iso-8859-15's EURO SIGN as
mule-unicode-0100-24ff.  Latin-9 user input is expressed as
latin-iso8859-15.  If mule-unicode-0100-24ff and latin-iso8859-15
appeared in a single message, Mew uses mew-charset-m17n instead
of iso-8859-15.  Thus, the problem (3) occured.

-- 
Tatsuya Kinoshita