Re: Adv. Col. Teleconference Minutes 2/16/99, relative

>One problem with the current idea of relative references: the notes say, "If
>reftarget is relative, the client has the burden of figuring out the value of
>the absolute URL.".  This sounds fine--clients already have code for relative
>URLs--but it won't work, because a 302's Location: header is required to
>provide an absolute URI (see RFC-2068, section 14.30).  (Plus, of course, it
>doesn't work at all for direct references.)

Right, the server must handle the relative to absolute conversion.
Most servers already do this for config or cgi-based redirects.

>I'd say that the Right Thing to do for expanding relative references is to
>base them off of the Request-URI of the request you're currently processing.
>This shouldn't be too hard: for the spec, point to RFC-1808 (with a note that
>the "base document" is the reference itself); for the implementation, lift
>some code from lynx or Mozilla or something.

That would be RFC 2396.


Received on Friday, 19 February 1999 00:48:23 UTC