Re: Outlook and HTTP-mail server

Mike Barnes wrote:
>
> > draft-ieft-http-v11-spec-06, HTTP/1.1 draft standard
> > 13.10, last paragraph
> >
> > A cache that passes through requests for methods it does not
> > understand SHOULD invalidate any entities referred to by
> > the Request-URI.
>
> Doesn't the following section (13.11), imply that unknown methods MUST be
> passed through? This issue was raised on the WebDAV list a while back.

Nope. The Book of HTTP (bless it's protocol, and manifold methods)
explicitly states that you must write through IF you are dealing with
methods other than GET or HEAD [13:11] - which is a specific case of the
last paragraph of 13:10. So that's explicit and says nothing more.
[5:1:1] says "All other methods are OPTIONAL" (their emphasis), and that
we should return 501 for any unimplemented method. It's quite explicit.

> Or shoot me down in flames - I must admit to not having read the draft
> standard with full concentration going. Just a casual flip through. But I am
> biased in that I'm an avid follower of WebDAV and am working with it a lot
> these days. I think it'd be really nice if Squid would pass these things
> through. Especially when I'm trying to use DAV in a transparent cache
> environment and I have no choice in the matter. :(

I'm mostly a casual skimmer as well. I've printed it out and am taking
it more seriously now.

> (I feel like I'm in a biblical debate - "Oh, but does not the book of HTTP
> 13:11 state that unsupported methods shall be passed through?" "No good
> fellow, for if you refer to HTTP 13:10 you will see that those who pass
> through unsupported methods will be damned for all eternity!")

Nay, brother, for does not the Holy Book say in the prelude by the
prophets of IETF) that "It is inappropriate to use Internet-Drafts as
reference material or to cite them other than as "work in progress"." ;)

*g*

D
Received on Fri May 14 1999 - 21:16:13 MDT