- From: Henrik Nordstrom <henrik@henriknordstrom.net>
- Date: Fri, 14 Mar 2008 00:50:16 +0100
- To: Brian Smith <brian@briansmith.org>
- Cc: "'HTTP Working Group'" <ietf-http-wg@w3.org>
On Sun, 2008-02-24 at 10:24 -0800, Brian Smith wrote: > > The server obviosly MAY inflate the > > entity or perform whatever processing on it it likes > > Right. > > > but that's not the primary expectations of HTTP for PUT. > > There aren't any agreed-upon expectations for how compressed request > bodies are processed; that is the problem. I disagree. Content-Encoding is an entity header. PUT is defined as storing the enclosed entity, which includes any entity-headers. The loophole is that it's not specified how PUT affects the state of a server, which by side-effect allows for servers to perform any processing they like of the entity enclosed in PUT before storage. But the fact that something is allowed by side-effect do not mean that thats the intended behavior. Regarding POST it's a more open question, but POST is an application level protocol exchange, and it can be expected the two endpoints have already negotiated the needed parameters by the same means the agent got the URI to POST to. This is already done in HTML, and expected by other applications making use of POST. Regards Henrik
Received on Thursday, 13 March 2008 23:51:36 UTC