Re: i28 proposed replacement text

On mån, 2008-05-12 at 02:45 +0200, Bjoern Hoehrmann wrote:
> * Henrik Nordstrom wrote:
> >3. If a Transfer-Encoding header field (Section 14.41) is present, and
> >   has a value other than "identity", then the transfer-length
> >   is defined by closing the connection.
> 
> I agree with your interpretation of the current text. I do note that,
> if we change the text, we should use the opportunity to clarify how
> to handle 
> 
>   Transfer-Encoding: identity;foo=bar
> 
> and perhaps
> 
>   Transfer-Encoding: gzip,identity
> 
> It is easy to misread "value other than".


Hmm... lets see now. We have removed the "identity" encoding in #16

http://www3.tools.ietf.org/wg/httpbis/trac/ticket/16

so 3 becomes simply

If a Transfer-Encoding header field (Section 14.41) is present, then the
transfer-length is defined by closing the connection.


Or perhaps by just clarifying the existing rule 2 so it can not be
misread when closing the connection is a valid delimiting metod.

From:

        If a Transfer-Encoding header field is present, then the
        transfer-length is defined by use of the "chunked"
        transfer-coding unless the message is terminated by closing the
        connection.

To:

        If a Transfer-Encoding header field is present, then the
        transfer-length is defined by use of the "chunked"
        transfer-coding, unless "chunked" is not the last encoding
        applied to the message-body in which case the message is
        terminated by closing the connection.

Regards
Henrik