Re: MIME text/plain attach with Shift_JIS (CP932)

2002-12-16 11:51:54
On December 16, 2002 at 23:07, Shinichiro HIDA wrote:

Now, I try to test MHonArc2002-12-16-snap.

Sometime, I would like to send the mail with some attachement in
the plain-text-file wrote with Shift_JIS for MS Windoze users.

MHonArc, including the snapshot builds, do not have support
for Shift_JIS (CP932).

I'm working on some more related encoding stuff with MHonArc right
now, so I'll try to add in the conversion table for CP932 for
MHonArc::CharEnt.  You may want to follow mhonarc-dev to see when
CP932 support gets checked into CVS.

Problem: unintelligible sequence of characters.
         plain/text file could not be attachement (only inline..).

MHonArc did not generate this error message.  No where in the
mhonarc code does this kind of error message exist.

By the way, text/plain attachments are not pre-processed, so if
the "attachcheck" MIMEARG option is set for m2h_text_plain::filter,
text/plain attachments will be saved as-is.

However, this may change with the development version where I am
trying to hook in text entity encoding support.  I.e. You will have
the ability to have all text entities (text/*) in mail converted to a
given character encoding, before any of the MIMEFILTERS are invoked.
This way, text entities, like text/html, can be converted to a single
encoding.  The processing of a text entity will be something like this:

  text/* entity -> text-encoder -> MIMEFILTERS

right now, it is:

  text/* entity -> MIMEFILTERS

which causes each MIMEFILTERS to be responsible for doing character
encoding conversion.  The MHonArc::CharEnt module is only used by
m2h_text_plain::filter, so data like text/html is left in the
their original character encoding.

This feature will have limitations depending on the version of Perl you
are using.  For Perl v5.8, I'm trying to leverage the Encode module.
For v5.6.x, the Unicode::MapUTF8 module (but I have noticed that it
does not support some charsets).  For other versions, I am trying to
code-up routines to allow for conversion to UTF-8.


P.S. Dicussions about snapshot builds should probably be directed
to mhonarc-dev.  Thanks.

To sign-off this list, send email to majordomo(_at_)mhonarc(_dot_)org with the

<Prev in Thread] Current Thread [Next in Thread>