[Mew-dist 03682] (mew-release) accent letters

[Kazu Yamamoto ( 山本和彦 )]

To those who want to display accent letters in Summary mode on Emacs
20.

I received some reports on accent letters in Summary mode. If you use
Emacs 20 or later where multi-lingual feature of Mule was absorbed,
decoded accent letters appears weird in Summary mode. I should have
understand this correctly soon, sorry.

This is due to the protocol (i.e. mew-cs-scan) between Emacs and imls. 
Currently, imls is just decoded MIME-encoded strings typically found
in From: or Subject:. This means that European accent letters are
mapped to 8bit area.

If you use Emacs 19, this problem never occurs since Emacs 19 is just
bilingual from the character set point of view. It treats 8bit
characters as Latin-1 (aka ISO-8859-1).

But on Emacs 20, things are mixed up. First you should understand that
8bit area is used by many, many other character sets. EUC-JP is an
good example. So, we have to tell our strategy to Emacs 20.

By default, 'iso-2022-7bit-ss2' is specified for the character set of
imls's output. This is convenient at least for Japanese. However, it
is not so for European, I guess.

If you use accent letters in your daily life, please refer to
mew-mule3.el. Set some variables started with 'mew-cs' replacing
'iso-2022-7bit-ss2' to 'iso-8859-1 in your ".emacs".

	e.g. (setq mew-cs-scan 'iso-8859-1)

I will include some examples in the next version. (Please give me
reports whether or not this works.)

For a long time, we have discussed on this topic. A possible good
solution is create a new character set based on ISO 2022 but whose
8bit area is ISO-8859-1 for the protocol. This eliminates the local
configurations above.

--Kazu