[9131] in www-talk@info.cern.ch
Re: Termination of POST body
daemon@ATHENA.MIT.EDU (Darren New)
Tue Sep 26 08:32:08 1995
Date: Tue, 26 Sep 1995 08:20:20 +0100
From: Darren New <dnew@sgf.fv.com>
To: Paul Phillips <paulp@cerf.net>
Cc: www-talk@w3.org
In-Reply-To: <Pine.SUN.3.91.950925193337.22481A-100000@nic.cerf.net>
Resent-From: www-talk@w3.org
> If content-length is supplied, I can use that; no problem. If it is not,
> it appears that the "end-of-body delimiter" is the only mechanism for
> determining the end of the entity body. This appears to be \r\n based on
> these clients' behavior. Is this so?
I think the "end-of-body delimiter" they're talking about is the MIME
end-of-body, not CRLF. That is, if it's a multipart message, the
end-of-body is CRLF then two hyphens then the boundary parameter then
perhaps two more hyphens then another CRLF. If it's base64 encoded, it
would be the blank line at the end of the encoding. And so on.
