Re: Third release of Constraint SVG browser

Cameron McCormack wrote:
> Robin Berjon:
>>So do I! I would very much like to see it implemented on top of SVG 1.2 
>>instead of just Batik. Cameron should clamour for extension functions in 
>>the XPath API :-D
> 
> Yeah, that would be cool (the extension functions, if there were an
> interface to XPath from the DOM).  Maybe it would be best to wait for
> XSLT/XPath 2.0 for this, since you can define your own XPath functions
> in XSLT 2.0.  These specs are in final call now, I think.

Requiring XSLT 2.0 support seems very unlikely for SVG 1.2... it's a 
massive spec :) But given that SVG 1.2 has DOM 3 XPath, and that most 
XPath implementations have pluggable extensions (or at the very least a 
table of functions), the addition of something exposing those along the 
lines of:

    interface SVGXPathEvaluator : xpath::XPathEvaluator {
        void    registerFunction( in DOMString NamespaceURI,
                                  in DOMString FunctionName,
                                  in EventListener function );
    }

A listener taking an XPathResultList and returning an XPathResult would 
cut it. Wouldn't you then have all you need to implement CSVG on top of 
SVG 1.2?

-- 
Robin Berjon