<<< Date Index >>>     <<< Thread Index >>>

Re: File Identification



Hi Al,

 On Thursday, May 6, 2004 at 12:36:06 PM +0100, Al S. Budden wrote:

> when I attach the [two UTF-8 text] files to an email in mutt, I get
> the following
>| I     1 /tmp/mutt-morat-15379-0       [text/plain, 7bit, us-ascii, 0.4K]
>| A     2 ~/untyped_cds.txt             [text/plain, quoted, iso-8859-1, 0.9K]
>| A     3 ~/documents/list_of_music.txt [text/plain, quoted, utf-8, 103K]
> why mutt thinks that untyped_cds.txt isn't utf-8?

    The charset displayed here in compose menu is not the file's
original charset on disk, but the charset choosen by Mutt to send the
file. Mutt recodes text attachments from $charset to one of
$send_charset upon sending.

    If you get the JA patch, the $file_charset feature adds ability to
guess file's original charset (so it doesn't need to be $charset) before
this recoding. More infos in now closed Debian Bug #176332 at
<URL:http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=176332>. The
$file_charset feature is integrated in all this year Debian unstable
Mutt packages.

    Finally you can always manually force sending charset and recoding
or no recoding with <edit-type> (bound to ^T by default). Add a "%c" to
$attach_format to see the conv/noconv flag.


Bye!    Alain.
-- 
When you want to reply to a mailing list, please avoid doing so from a
digest. This often builds incorrect references and breaks threads.