Re: HEAD and 301/302/303 conflict?

Paul Leech reports:

>Section 5.2.2 on the HEAD method says that HEAD is just like GET, 
>except that the server must not return any Entity-Body in the response. 
>Section 6.2.3 on redirection status codes, 301 and 302 (Moved 
>Permanently and Moved Temporarily ) says that the Entity-Body in the 
>response should contain a short hypertext note. Similarly for 303 (See Other).

Ooops, contradiction.

>Presumably, 301 and 302 are reasonable responses for HEAD requests (303 
>is not so obvious -- it might be only OK as a response to a POST in 
>1.1?) and so require Entity-Bodies, which HEAD says can't be sent. In 
>the description for 300, it explictly says that 300 in response to a 
>HEAD request doesn't return an Entity-Body; there is no explicit 
>exemption for 301, 302, or 303.
>
>In section 5.2.2, was the intent to say that HEAD is like GET, except 
>that whenever GET would return 200 and an Entity-Body, HEAD will return 
>204 and no Entity-Body? That seems plausible, and reconciles all 
>behavioral difference except for when GET would return 300. Equally 
>plausible is that 301, 302, and 303 should contain the same exception 
>for HEAD that 300 contains.  I don't know the intent, so I can't make a 
>recommendation.
>
>Which should it be?

The latter -- I'll fix this in both specs.


 ....Roy T. Fielding  Department of ICS, University of California, Irvine USA
                      Visiting Scholar, MIT/LCS + World-Wide Web Consortium
                      (fielding@w3.org)                (fielding@ics.uci.edu)

Received on Friday, 25 August 1995 19:07:19 UTC