On Tue, 06 Apr 1999, Vincent Lefevre wrote:
> > This changes the variable pipe_decode but when it was unset before
> > it will be unset after running urlview. I want pipe_decode to be
> > unset normally, so this is acceptable for me.
> I don't like this, because if one day, I set pipe_decode, things may
> fail to work, and I won't remember why.
I had no problems with this in the previous months. The only problem
is, that I have to remember, that pipe_decode will magically be unset
by running urlview. If I decide sometimes in the future to have
pipe_decode always set, I will change my macros in accordingly. I can
live with this.
> I think so too. The best would be something in Mutt, like local variables
> (something to save the environment) or options to pipe, e.g.
> |{pipe_decode} and |{nopipe_decode}
> so that the pipe always behaves as if pipe_decode were set or unset.
Features like this may blow up the mutt code...
> > maybe you want to have a look at mimedecode(1) which simply decodes
> > MIME to 8bit, so it could be used as a pre-filter for urlview.
> > mimedecode should be available at ftp://ftp.dde.dk/pub/mimedecode.c or
> > as a Debian package (source and binary, including a manpage).
> Before d/l'ing it, does it convert QP to 8bit in attachments too?
> I ask this, because I use emil (with procmail) to convert QP to 8bit,
> but when there is QP in attachments, it isn't converted.
The manpage tells the following:
NAME
mimedecode
SYNOPSIS
mimedecode
[-h|-d <debug level>] < encoded_msg > decoded_msg
DESCRIPTION
This manual page documents briefly the mimedecode command.
This manual page was written for the Debian GNU/Linux dis
tribution because the original program does not have a
manual page.
mimedecode performs the decoding of transfer encoded text
type mime messages. The message in its entirety is read
from stdin. The decoded message is written to stdout;
hence, this program behaves as a filter which may be
placed wherever convenient.
It is assumed that the message has reached its point of
final delivery and at that point 8-bit text types can be
handled natively. Hence, the need for transfer-encodings
is not present any more.
Only some cases are handled:
Encoded header fields are decoded from QP or B encoding.
The charset is assumed to be iso-8859-1
Part or subparts of content-type text only are decoded
All other content-types are passed transparently
OPTIONS
This program accepts the follwing options:
-h Switch on header logging
-d <debug level>
Enable debugging output at level: <debug level>.
Ciao
Roland
--
* [EMAIL PROTECTED] * http://www.spinnaker.de/ *
PGP: 1024/DD08DD6D 2D E7 CC DE D5 8D 78 BE 3C A0 A4 F1 4B 09 CE AF
