RE: Encoding URLs

> Indeed. For that matter, how is an XSLT processor to know 
> that an arbitrary string (attribute value, #PCDATA, whatever) is even a
URI to 
> begin with? The "this is a URI" meaning is not inherent in any XML 
> vocabulary (except, of course, XHTML).

The escaping of URLs is only required with method="html", and is required on
those attributes defined in the HTML 4.0 spec to be URLs. This does raise
the question of how you get the equivalent behaviour with XHTML. It's also
caused some usability problems for Saxon users, e.g. where the URL is
already escaped, which is why my latest version introduces
disable-output-escaping for attribute values.

Mike Kay


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list