W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2005

Re: Pipelining and Expect: 100-continue interaction

From: Jamie Lokier <jamie@shareable.org>
Date: Fri, 22 Apr 2005 18:27:04 +0100
To: Marc Schneider <mschneider@opnet.com>
Cc: ietf-http-wg@w3.org
Message-ID: <20050422172704.GA22806@mail.shareable.org>

Marc Schneider wrote:

>    I could imagine that a server would attempt to parse the data as
>    if it was an HTTP request start line and discard the data if it
>    doesn't parse correctly.  If it does parse correctly, the server
>    would process it as a new request. In this case, pipelining of
>    request would be possible.

Nice idea, but the request body can start with text that looks exactly
the same as a request header - there's no reliable way to distinguish
them.

Also, if you wrote a client that did this, and even if there is a way
to distinguish them (as there is if the request is chunked), it would
not work with existing servers or proxies.

That's the main problem: to change this feature of the protocol to
allow clients to omit the request body and pipeline, and still be
compatible with existing servers, you need some way for the client
to discover that the server (or proxy) supports the feature.  It would
have to be a HTTP extension.

>    Is there something specific in the RFC that you could point to
>    that says that this is not legal?

It's not made very clear.  But I've asked exactly this question before
on this list, which is why I know the answer.

-- Jamie
Received on Friday, 22 April 2005 17:27:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:49:40 GMT