Re: PATCH, Expect, Prefer, etc

I suppose the proxy *could* issue a subsequent GET, it would just be
impossible to ensure that the result is accurate.  E.g. the resource
might be modified between the initial response and the subsequent GET,
which somewhat defeats the purpose of the 209 response (at least when
used with PATCH).  I'm not sure we'd want to encourage this behavior at
all, even in highly controlled environments.

- James

Henrik Nordstrom wrote:
> On tor, 2007-08-02 at 23:18 -0700, James M Snell wrote:
> 
>> the client rather than the complete original response.  If the request
>> contains 209-content-returned and the server responds with 200 or 204
>> (or whatever), then it obviously will not be capable of fulfilling the
>> preference and must ignore it.
> 
> Can't it? What stops the proxy from issuing a GET in that case and
> returning the 200 response as a 209?
> 
> Note: Not saying it's generally a good thing or even desired for a proxy
> to do, and only makes sense in very special controlled deployments.
> 
> Regards
> Henrik