RE: [xsl] Re: XSLT 2.0: On xsl:sequence and xsl:copy-of

Subject: RE: [xsl] Re: XSLT 2.0: On xsl:sequence and xsl:copy-of
From: "Michael Kay" <mhk@xxxxxxxxx>
Date: Tue, 14 Oct 2003 11:15:37 +0100
> So rather than saying dear processor I want this function to 
> return an integer, its almost saying I hope it returns an 
> integer. I.e. its not a request for a cast, its .... just to 
> keep the typing system happy (for those that need one :-) 
> I.e. it does little for the end user.

It gives you an explicit error message rather than just "wrong output"
when you make mistakes, and it allows the system to make decisions
statically that otherwise would be made at run-time, which should lead
to faster execution.

There's nothing new about this; typed programming languages have been
with us for years, and are generally considered a "good thing" when you
need to write robust and scaleable software. The only thing that's new
in XPath 2.0/XSLT 2.0 is the ability to mix strong and weak typing in
the same language.
Michael Kay

 XSL-List info and archive:

Current Thread