From xemacs-m  Thu Feb 27 14:44:19 1997
Received: from crystal.WonderWorks.COM (crystal.WonderWorks.com [192.203.206.1])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id OAA08751
	for <xemacs-beta@xemacs.org>; Thu, 27 Feb 1997 14:44:16 -0600 (CST)
Received: by crystal.WonderWorks.COM 
	id QQcete23720; Thu, 27 Feb 1997 15:44:15 -0500 (EST)
Date: Thu, 27 Feb 1997 15:44:15 -0500 (EST)
Message-Id: <QQcete23720.199702272044@crystal.WonderWorks.COM>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
From: Kyle Jones <kyle_jones@wonderworks.com>
To: xemacs-beta@xemacs.org
Subject: Incompatibilities between VM, TM, and PGP
In-Reply-To: <m3zpwqwngp.fsf@cortex.corpus.uni-muenster.de>
References: <m3zpwqwngp.fsf@cortex.corpus.uni-muenster.de>
X-Face: /cA45WHG7jWq>(O3&Z57Y<"WsX5ddc,4c#w0F*zrV#=M
        0@~@,s;b,aMtR5Sqs"+nU.z^CSFQ9t`z2>W,S,]:[+2^
        Nbf6v4g>!&,7R4Ot4Wg{&tm=WX7P["9%a)_da48-^tGy
        ,qz]Z,Zz\{E.,]'EO+F)@$KtF&V

Markus Gutschke writes:
 > I noticed an incompatibility between VM, TM, and PGP and I do
 > not really know, who is wrong. I currently suspect TM, but
 > that is just a wild guess.
 > 
 > In one of my messages, I included an entire old e-mail message
 > in message/rfc822 format. I also told TM, to sign my new
 > message. Usually, all of this works just fine, but in this
 > very case it resulted in a `bad signature'. After some
 > experimentation with different messages, I traced the problem
 > to the fact that VM escapes all lines, that start with 'From
 > '. Only, the escaping is done *after* the signature has been
 > computed. Of course, this breaks the checksum and the
 > signature becomes invalid. Manually removing the '>' escape
 > character yields a good signature. Unfortunately, I cannot
 > just remove the escape character from the INBOX, because VM
 > will then assume that the included message is actually the
 > header of a new message. So, a proper solution (or
 > work-around) would escape all lines before computing the PGP
 > signature.

This is most likely done by the transport system.  I.e. VM, TM
and PGP are either all blameless or all equally to blame for not
armoring the message against From line munging.  I will probably
make VM encode such messages using quoted-printable, or just
encode the 'F' in 'From ' as =46.

