W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2007

Re: Issue i17 (Revise description of the POST method)

From: Henrik Nordstrom <henrik@henriknordstrom.net>
Date: Mon, 04 Jun 2007 23:02:00 +0200
To: Julian Reschke <julian.reschke@gmx.de>
Cc: HTTP Working Group <ietf-http-wg@w3.org>
Message-Id: <1180990920.14442.48.camel@henriknordstrom.net>
mån 2007-06-04 klockan 22:14 +0200 skrev Julian Reschke:
> Henrik Nordstrom wrote:
> > I'm fine with both wordings. But it is the most fundamental difference
> > between the two methods.
> > ..
> OK, let me try to rephrase it.
> Is it OK to use POST for XML-RPC?

Yes, of course. XML-RPC is a perfect example where the Request-URI
identifies a data-handling process responsible for processing the
XML-RPC call and returning the result in a 200 response.

And is allowed by RFC2616 even if ther is that bit of confusing wording
about the submitted data being a subordinate.. The subordinate wording
is technically not that wrong, just confusing and makes completely wrong
associations which makes it a bad choice for describing the relation.

>  If yes, *could* I use PUT instead?

Not if you want the XML-RPC to be processed by the server and expect to
have the result returned.

I don't see how the definition of PUT could make anyone think that PUT
would be the appropriate method for XML-RPC calls to be processed online
by the server.

But PUT is the appropriate method you want to publish/store the XML-RPC
call as such (the XML-RPC request entity blob, not the result or the
XML-RPC call) as an entity on the server, at the location identified by
the request URI. In the XML-RPC context possibly for later batch or
offline processing by other agents, or simply as a permanent record of
the call contents.

In my eyes the RFC2616 PUT refeferences to POST behavior is perhaps
closer to the current POST definition than the original. Lets look at
that text again in isolation, should we:

   The URI in a
   POST request identifies the resource that will handle the enclosed
   entity. That resource might be a data-accepting process, a gateway to
   some other protocol, or a separate entity that accepts annotations.

No "subordinate" crap to confuse what is being talked about, and with
the "data-handling process" part which opens for pretty much any data
processing you might imagine.

It's actualy a quite good sentence, except that it has nothing to do
with PUT.. It can be moved straight up to POST with no changes imho. And
XML-RPC is a perfect match for the "data-accepting process" gategory.

Again, I don't quite get what it is you are worried about here. The
"data-accepting process" is just about anything which accepts data and
processes it in some manner..

What about the text is it you have concerns with?


Received on Monday, 4 June 2007 21:02:11 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 11:10:42 UTC