Re: NEW ISSUE: message-body in CONNECT response

Roy T. Fielding wrote:
> The standard requires an empty body on a non-closed connection to be
> indicated by one of the two message length indications (CL or TE chunked).
> In this case, the obvious solution is to require "Content-Length: 0" be
> included in the header fields of the 200 response.  It doesn't matter
> if some clients ignore that field.  What matters is that we don't add
> more method-specific parsing of response bodies.

I agree that it's stupid to have a special-case for CONNECT parsing, but
that's *already* how implementations are required to behave. A client
that tries to parse a CONNECT response according to the rules currently
given in RFC 2616 will be unable to create a tunnel. Requiring proxies
to add "Content-Length: 0" won't help, because clients will still want
to be able to deal with pre-2616bis proxies, which will require the
special case.

-- Dan