Gnus doesn't understand its own message/rfc forwarding

Gnus doesn't understand its own message/rfc forwarding

[Steinar Bang]

Version: Gnus v5.8.4 (CVS checkout from march 17), XEmacs 20.4 (linux)

Gnus doesn't quite seem to understand its own message/rfc forwarding
these days.  When I do a C-d, this is what I see:

   [  94: Steinar Bang        ] <* mixed> fest med "baikerscømmet"
       [   2: Steinar Bang        ] <1 text>
       [  81: Steinar Bang        ] <2 rfc822>
           [  72: nobody              ] <2 text>

The rfc822 part looks like this at the start:

From: Steinar Bang <sb@dod.no>
Subject: fest med "baikerscømmet"
To: Bjørn Olav Listog <blistog@c2i.net>, Tore Reimers <treimers@c2i.net>
Date: 12 Apr 2000 08:21:52 +0200
Organization: Denizens of Doom, Norway Chapter

From: Steinar Bang <sb@dod.no>
Subject: innflyttingsfest
To: doders-announcements@math.uio.no
Cc: doders@dod.no
Date: 09 Apr 2000 21:32:36 +0200
Reply-To: sb@dod.no, tlf@dod.no
Organization: Denizens of Doom, Norway Chapter

ie. I don't get to the forwarded message in such a way that I can
reply to it, and as far as I can remember, this used to be possible.

The headers for the message/rfc822 part, looks like this:

Content-Type: message/rfc822; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

(BTW the charset parameter on this MIME type, seems to make problems
for the exchange server)

Does anyone know if this is fixed in the most recent CVS sources?

Thanx!


- Steinar

Re: Gnus doesn't understand its own message/rfc forwarding

[Kai Großjohann]

Steinar Bang <sb@metis.no> writes:

> The rfc822 part looks like this at the start:
> [...]
> ie. I don't get to the forwarded message in such a way that I can
> reply to it, and as far as I can remember, this used to be possible.

Can you reply to the (innermost) text part?

kai
-- 
The birch trees fly way too low these days.

Re: Gnus doesn't understand its own message/rfc forwarding

[Steinar Bang]

>>>>> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann):

> Steinar Bang <sb@metis.no> writes:
>> The rfc822 part looks like this at the start:
>> [...]
>> ie. I don't get to the forwarded message in such a way that I can
>> reply to it, and as far as I can remember, this used to be possible.

> Can you reply to the (innermost) text part?

Nope.  It doesn't show up as a message.  It's not even decoded from
q-p when I go down there.

This is what all headers of the part looks like after doing C-u g:

Content-Type: message/rfc822; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

X-From-Line: sb@dod.no  Sun Apr  9 21:34:32 2000
Return-Path: <sb@dod.no>
Received: from localhost (sb@localhost [127.0.0.1])
	by viffer.metis.no (8.9.3/8.9.3) with ESMTP id VAA12684
	for <sb@localhost>; Sun, 9 Apr 2000 21:34:31 +0200
Received: from tyr.metis.no
	by localhost with IMAP (fetchmail-5.1.0)
	for sb@localhost (single-drop); Sun, 09 Apr 2000 21:34:31 +0200 (MEST)=

Received: by metis.no (8.9.1/8.9.1) with ESMTP id VAA21223; Sun, 9 Apr =
2000 21:33:55 +0200 (MET DST)
Received: (from sb@localhost)
	by viffer.metis.no (8.9.3/8.9.3) id VAA12664;
	Sun, 9 Apr 2000 21:33:28 +0200
To: doders-announcements@math.uio.no
Cc: doders@dod.no
Subject: innflyttingsfest
Reply-To: sb@dod.no, tlf@dod.no
From: Steinar Bang <sb@dod.no>
Organization: Denizens of Doom, Norway Chapter
Date: 09 Apr 2000 21:32:36 +0200
X-Gnus-Mail-Source: file:/var/spool/mail/sb
Message-ID: <wh3dovxe7f.fsf@viffer.metis.no>
User-Agent: Gnus/5.0804 (Gnus v5.8.4) XEmacs/20.4 (Emerald)
X-MIME-Autoconverted: from quoted-printable to 8bit by metis.no id VAA2=
1223
Xref: viffer.metis.no mail.doders-announcements:8
MIME-Version: 1.0
Content-Type: text/plain; charset=3Diso-8859-1
Content-Transfer-Encoding: 8bit
Lines: 36

Re: Gnus doesn't understand its own message/rfc forwarding

[Steinar Bang]

>>>>> Steinar Bang <sb@metis.no>:

>>>>> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann):
>> Steinar Bang <sb@metis.no> writes:
>>> The rfc822 part looks like this at the start:
>>> [...]
>>> ie. I don't get to the forwarded message in such a way that I can
>>> reply to it, and as far as I can remember, this used to be possible.

>> Can you reply to the (innermost) text part?

> Nope.  It doesn't show up as a message.  It's not even decoded from
> q-p when I go down there.

> This is what all headers of the part looks like after doing C-u g:

[snip!]

Hm... it could be the fact that the message/rfc822 part is q-p, that
is the problem?  Ie. _all_ of the message is q-p, including the
headers.  Maybe it would be better to shove all encoding down to the
lowest part?

Isn't this the behavior that the below quoted paragraph from section
6.4 of http://www.ietf.org/rfc/rfc2045.txt would seem to indicate?

   Certain Content-Transfer-Encoding values may only be used on certain
   media types.  In particular, it is EXPRESSLY FORBIDDEN to use any
   encodings other than "7bit", "8bit", or "binary" with any composite
   media type, i.e. one that recursively includes other Content-Type
   fields.  Currently the only composite media types are "multipart" and
   "message".  All encodings that are desired for bodies of type
   multipart or message must be done at the innermost level, by encoding
   the actual body that needs to be encoded.

Re: Gnus doesn't understand its own message/rfc forwarding

[Carey Evans]

Steinar Bang <sb@metis.no> writes:

[snip: problems with message/rfc822]

This sounds like it's related to some problems I have with Gnus
5.8.3.  I haven't tried the latest CVS version to see if they still
exist though.

See:

    http://www.gnus.org/list-archives/ding/200001/msg00417.html
    http://www.gnus.org/list-archives/ding/200002/msg00069.html

[...]

> Isn't this the behavior that the below quoted paragraph from section
> 6.4 of http://www.ietf.org/rfc/rfc2045.txt would seem to indicate?
> 
>    Certain Content-Transfer-Encoding values may only be used on certain
>    media types.  In particular, it is EXPRESSLY FORBIDDEN to use any
>    encodings other than "7bit", "8bit", or "binary" with any composite
>    media type, i.e. one that recursively includes other Content-Type
>    fields.  Currently the only composite media types are "multipart" and
>    "message".  All encodings that are desired for bodies of type
>    multipart or message must be done at the innermost level, by encoding
>    the actual body that needs to be encoded.

It looks like that to me.

In fact, I don't think Gnus should be touching the content of the
forwarded message at all.  For S/MIME and PGP MIME, the entire MIME
part is signed, including the headers, so any changes from
quoted-printable to 7bit, or 8bit to quoted-printable, will make the
signature invalid.  Gnus should be able to forward the message intact
so that the signature can still be validated.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

"Perl Programmers are from Mars, Python Programmers are from Yorkshire?"
Catchy title.                                              - Alex in c.l.py

Re: Gnus doesn't understand its own message/rfc forwarding

[Shenghuo ZHU]

>>>>> "Steinar" == Steinar Bang <sb@metis.no> writes:

>>>>> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann):
>> Steinar Bang <sb@metis.no> writes:
>>> The rfc822 part looks like this at the start:
>>> [...]
>>> ie. I don't get to the forwarded message in such a way that I can
>>> reply to it, and as far as I can remember, this used to be possible.

>> Can you reply to the (innermost) text part?

Steinar> Nope.  It doesn't show up as a message.  It's not even
Steinar> decoded from q-p when I go down there.

I did a little change on the forwarding. Could you test your cases?

-- 
Shenghuo

Re: Gnus doesn't understand its own message/rfc forwarding

[Steinar Bang]

>>>>> Shenghuo ZHU <zsh@cs.rochester.edu>:

>>>>> "Steinar" == Steinar Bang <sb@metis.no> writes:
>>>>> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann):
>>> Steinar Bang <sb@metis.no> writes:
>>>> The rfc822 part looks like this at the start:
>>>> [...]
>>>> ie. I don't get to the forwarded message in such a way that I can
>>>> reply to it, and as far as I can remember, this used to be possible.

>>> Can you reply to the (innermost) text part?

Steinar> Nope.  It doesn't show up as a message.  It's not even
Steinar> decoded from q-p when I go down there.

> I did a little change on the forwarding. Could you test your cases?

I have now tested with the forwarded message being both CTE 8bit and
q-p (with US-ASCII) headers, and with the message with CTE 8bit and
RFC2047 headers, and all of them appeared fine in Gnus (no CTE or
charset on the message/rfc822 part and q-p encoding of the message
itself). 

I was able to go to the message by doing C-d, and reply to the
message. 

I've also forwarded a CTE 8bit, charset=iso-8859-1 message ot people
with Outlook98/Outlook2000, reading email through exchange servers,
and they saw the message as a forwarded message is supposed to be
displayed in that program.

Great work!  Thanx! :-)