Re: A new suggestion on 100 CONTINUE

• From: Roy T. Fielding <fielding@kiwi.ICS.UCI.EDU>
• Date: Wed, 11 Jun 1997 13:28:11 -0700
• To: John Franks <john@math.nwu.edu>
• Cc: http-wg@cuckoo.hpl.hp.com
• Message-Id: <9706111327.aa08798@paris.ics.uci.edu>

>If the POST has no body the headers must contain a "Content-length: 0"
>line, so the server knows (i.e. this is not an implementation problem for
>my suggestion).
>
>This does illustrate my point though.  Isn't it a little bit brain dead
>to *require* a "100 CONTINUE" for a POST with no body?

That is brain dead.  The intended requirement was that it apply to
any request that includes a message body and not be tied to the method
definition at all.  The definition provided in 2068 is wrong.

>Incidentally, when a client waits for and gets a "100 CONTINUE" after
>sending the headers of a bodyless POST, what does it send?

Nothing [and that is explicitly covered in section 10.1.1]

....Roy

Received on Wednesday, 11 June 1997 13:34:23 UTC