Subject: RE: XSLT 1.1 From: Kay Michael <Michael.Kay@xxxxxxx> Date: Fri, 8 Sep 2000 18:20:35 +0100 |
> XSLT 1.1 specifies that it will only consider Java and > EcmaScript bindings. There is no XSLT 1.1 specification, only a requirements document. > It seems to me that the extension bindings don't have to be > language-specific. Why can't an XML-based binding be defined > to standardize on a cross-language specification? > Because there is no suitable cross-language definition of data types that it can refer to. SQL has exactly the same problem. If you make a call on my:proc(3.2) and proc is written in Java then the spec needs to say - where to look for a function called proc - what to do if it finds more than one - how to convert the argument 3.2 into the Java data type expected by the selected method (e.g. what if it expects a BigDecimal?) - how to convert the result of the selected method back to an XPath value. All these things are necessarily Java-specific. There are of course some things one can say that are not language-specific, and no doubt the spec, when it emerges, will try to achieve as much generality as possible. Mike Kay XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
XSLT 1.1, Kaganovich, Yevgeniy | Thread | MSXML and Namespaces ... not fully , Eckenberger Axel |
parameter count in xsl:number, Paul Terray | Date | RE: Axis problem, Kay Michael |
Month |