From kmail-devel  Fri Aug 31 15:05:03 2001
From: Sven Carstens <s.carstens () gmx ! de>
Date: Fri, 31 Aug 2001 15:05:03 +0000
To: kmail-devel
Subject: Bug#31162: The e-mail in question
X-MARC-Message: https://marc.info/?l=kmail-devel&m=99927038611287

Am Sun, 31 Aug 2001 schrieben Sie:
> Well, I would definitely consider this an Eudora bug.
> They define here a MIME boundary and don't use it anywhere.
>=20
> Content-Type: multipart/alternative;
> =09boundary=3D"=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D_20793727=3D=3D_.ALT"
>=20
> >From the point of view of KMail, then everything is outside of these=20
> boundaries and therefore doesn't exist.
>=20
> If they really intend to send a multipart/alternative mail (with plain te=
xt=20
> and html), then that has to look very different.

The message is shurely not what was intended to be send.
But it's a valid message anyway. The parser in current KMail doesn't
seem to recognize boundaries from toplevel parts as recommended by
rfc 2046 5.1.2 .

The new parser in libkdenetwork will parse this message in a multipart/mixe=
d
message containing 5 sub parts:
text/plain (the strictly speaking empty multipart/alternative)
text/plain (the plain part of the intended multipart/alternative)
text/html (the html part of the intended multipart/alternative)
text/plain (the intended end of multipart/alternative)
application/msword (the intended second part of the multipart/mixed)

CU Sven

_______________________________________________
Kmail Developers mailing list
Kmail@mail.kde.org
http://mail.kde.org/mailman/listinfo/kmail