David Wall wrote:
> > If a new version of HTTP specifies the use of ';' rather than '&', the
> > servlet engine could detect the HTTP version and decode appropriately.
>
> What I don't understand is why it's a "problem" for HTTP at all, since it
> seems to work just fine right now. I don't think it requires a new HTTP
> version at all, and is already "wrong" according to the standard, but works.
> I read about this being a bug already, and that if we wanted to use '&' in a
> param, then we should have used the escaped #nn format. Yet, all the
> browsers know how to handle this, so what's the fuss?!
My point exactly. The browsers all do it the common (non-compliant) way, which
means name-value pairs are separated by '&'. To implement ';' instead, you need to
specify a new version of HTTP so that clients and servers will agree upon which
encoding to use. This is especially important for query strings; there's already a
mime-type line being sent for POST body text ("text/x-www-urlencoded") that could
be changed to specify the new encoding.
To steal a line... standards are great, everyone should have one. :-)
Wes
===========================================================================
To unsubscribe: mailto [EMAIL PROTECTED] with body: "signoff JSP-INTEREST".
FAQs on JSP can be found at:
http://java.sun.com/products/jsp/faq.html
http://www.esperanto.org.nz/jsp/jspfaq.html
