Re: [xsl] Quasi-Literals and XML

Subject: Re: [xsl] Quasi-Literals and XML
From: David Carlisle <davidc@xxxxxxxxx>
Date: Wed, 31 Jan 2001 13:35:29 GMT
>  I thought the XSL gurus back
> then pretty much agreed that XSL was Turing-complete?

'tis so. In XSLT 1.0 you need to use string handling to implement the
tape, which is a bit of a pain, but given foo:node-set() or xslt 1.1
implicit rtf-node set conversion you can use node lists which makes
it a lot more reasonable.

But being turing complete doesn't really say much, that was turing's
point that essentially any type of machine/programming language would end up
being able to evaluate the same set of functions.

Of course the comment that you quote

> "XSL is a specialized language built specifically for transforming XML, into
> XML or other notations, but not for transforming other notations into XML.

is perfectly valid, and is explictly stated by the XSLT spec.
So I don't think that is necessarily a "criticism" of xsl, just a
statement of fact.
 
Being turing complete, one could write a regexep string matcher in XSLT,
if you had a spare month or two to write it, and your users had a
similar amount of time to run it.....

David



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


Current Thread