Randal L. Schwartz writes
> Uh, it seems a bit fishy to me. "nothing's changed, but by the way,
> set this cookie please". Why change a cookie if nothing else has
> changed?
If the HTTP headers were only about the document then I would agree.
However, if we look at this set of headers from a typical request
Date: Tue, 19 Jun 2001 01:03:02 GMT
Server: Apache/1.3.20 (Unix) mod_perl/1.25
Set-Cookie:
session=69643d343037266c6173746163636573733d393932393132353933; path=/
Last-Modified: Mon, 18 Jun 2001 17:09:26 GMT
ETag: "25c4f-6-3b2e35c6"
Accept-Ranges: bytes
Content-Length: 6
Connection: close
Content-Type: text/html
Then Date, Server, Accept-Ranges, and Connection are about the server
while Last-Modified, Content-Length, and Content-Type are about the
document. The Set-Cookie could be about either. (I don't know what
ETag is for.) In my case, the cookie is related to the user's
interaction with the server far more than with the document. So I do
think it is valid to return Set-Cookie headers on a 304.
-- Andrew
