W3C home > Mailing lists > Public > public-qt-comments@w3.org > August 2003

RE: XSLT comments

From: Kay, Michael <Michael.Kay@softwareag.com>
Date: Thu, 14 Aug 2003 18:14:44 +0200
Message-ID: <DFF2AC9E3583D511A21F0008C7E62106073DD06D@daemsg02.software-ag.de>
To: Corneel Theben Tervile <corneel.theben-tervile@howest.be>, public-qt-comments@w3.org
> Dear Sir,
> 
> As I wanted to use XSLT for the transformation of a XML 
> document, containing in semantic sense a networkstructure, 
> into a XML document with the same information but in some 
> treestructure, I established that there is no instruction to 
> implement an iteration based on a condition (e.g. test the 
> value of a variable). Wouldn't it be useful to provide such 
> an instruction in future versions of XSLT?
> 

Thanks for the comment.

XSLT is a functional language designed to be side-effect-free. It therefore
relies heavily on recursion rather than iteration to implement such
algorithms. You will find that many complex graph-based algorithms have been
implemented in XSLT: look for example for Dimitre Novatchev's FXSL library.

If you have problems coding particular algorithms in XSLT, you can usually
get help from the xsl-list at www.mulberrytech.com.

XSLT 2.0 contains many features which make complex algorithms easier to
write than in XSLT 1.0 (for example, through the introduction of sequences
in the data model), but we don't intend to change the fact that it remains a
pure functional language.

Regards,

Michael Kay
Received on Thursday, 14 August 2003 12:16:29 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:45:13 UTC