X-Spam-Check-By: sourceware.org
Message-ID: <44E5B4C8.9050206@ukf.net>
Date: Fri, 18 Aug 2006 13:38:32 +0100
From: Max Bowsher <maxb1 AT ukf DOT net>
Reply-To: cygwin AT cygwin DOT com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.12) Gecko/20050923 Thunderbird/1.0.7
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
Subject: Re: Is cygwin-ml softwere altering message content?
References: <00bc01c6c2bc$c3488db0$a501a8c0 AT CAM DOT ARTIMI DOT COM>
In-Reply-To: <00bc01c6c2bc$c3488db0$a501a8c0@CAM.ARTIMI.COM>
OpenPGP: id=C0F2C580
Content-Type: multipart/signed; micalg=pgp-sha1;  protocol="application/pgp-signature";  boundary="------------enig81DA614A85A0DEA24C201FDC"
X-IsSubscribed: yes
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

--------------enig81DA614A85A0DEA24C201FDC
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Dave Korn wrote:
> On 18 August 2006 12:42, Max Bowsher wrote:
>=20
>>
>> Let me try to force a re-wrap to occur here:
> 23456789A123456789B123456789C123456789D123456789E123456789F123456789G1234=
56789
> H
>=20
>   Didn't work.  Try some trailing spaces.


Yes it did.

The re-wrap which matters is the one in the quoted-printable encoded form.

The copy I sent is wrapped as follows:

>23456789A123456789B123456789C123456789D123456789E123456789F123456789G123=
=3D
456789H

However, the copy I received post list-munging was wrapped like this:

>23456789A123456789B123456789C123456789D123456789E123456789F123456789G12345=
=3D
6789H


>> The specific nature of the problem seems to be that quoted-printable
>> MIME parts seem to be getting unencoded and then re-encoded by the
>> sourceware mail system. In doing so, the wrapping policy applied by the
>> original mailer is destroyed, and sourceware's own is imposed.
>>
>> Clearly this is a bad thing, since unless the results of the policies
>> are identical, the validity of any signature is destroyed.
>=20
>   Are we really sure that there isn't incorrect decoding going on at the
> receive end, and that the two forms aren't actually supposed to decode to=
 the
> same result?

The RFCs indicate that it is the encoded form which is signed and verified.

The problem is that the sourceware mailer is decoding the MIME part and
re-encoding it into a subtly different form.

It's irrelevant that the two encoded forms happen to decode to the same
result, since it is the encoded form which is signed.


>> Is there a suitable sourceware administrator watching this thread, or
>> should I summarize the issue to overseers@ ?
>=20
>   I think we need to understand it better.  When I tried to decode your
> signature, that "looked good from here" to Al, I got an error.  So I think
> there's at least some client-dependency here.

I feel I do fully understand the situation.

Are you saying that you got a verification error for my message with
Message-ID <44E46C29 DOT 3050005 AT ukf DOT net> ?
That would be bizarre, since it verifies fine for me.

Max.


--------------enig81DA614A85A0DEA24C201FDC
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.1 (Cygwin)

iD8DBQFE5bTNfFNSmcDyxYARAoIhAJ9CbEZYqwlOOap/BnGFGaY+3xfYIgCgpcKu
/CBj7pF1dl08eyHXxG0OeZg=
=y8D7
-----END PGP SIGNATURE-----

--------------enig81DA614A85A0DEA24C201FDC--