W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1999

RE: Fragments in "Location" field

From: Henrik Frystyk Nielsen <henrikn@microsoft.com>
Date: Sun, 15 Aug 1999 08:09:20 -0700
Message-ID: <FD8F41F04D35D111899600805F19A16C1C3F4CB1@RED-MSG-48>
To: "'Roy T. Fielding'" <fielding@kiwi.ics.uci.edu>
Cc: http-wg@hplb.hpl.hp.com

> >Regarding allowing fragments, I believe the same is the case for
> >Content-Location:
> >
> >       Content-Location = "Content-Location" ":"
> >                         ( absoluteURI | relativeURI )
> >
> >should be
> >
> >       Content-Location = "Content-Location" ":"
> >                         ( absoluteURI | relativeURI ) [ "#" 
> fragment ]
> 
> No, that would be incorrect.  The reference in 
> Content-Location must be
> to a resource, not a fragment of a resource representation.

The fragment (or view) is a function of the media type of the representation
chosen and hence does not have to be defined the same way across the set of
representations available for content negotiation of any particular
resource. For example, a fragment in HTML pointing to page 3 does not have
to be defined the same way for PS.

It is therefore functionally correct that Content-Location can contain a
fragment. The problem is of course that HTTP servers currently do not see
the fragment under normal circumstances because clients do not send it but
this is the same as for the Location header field. In fact, I consider this
a bug in HTTP :)

Henrik Frystyk Nielsen,
mailto:frystyk@microsoft.com
Received on Sunday, 15 August 1999 18:16:47 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:33:32 EDT