RE: Iterative XML Handling

Can't comment on IE5 performance. SAXON optimises patterns consisting of a
single element name because they are so common. It will probably only make a
difference of 20% at worst, certainly there are more important things to
worry about in making a stylesheet go fast. Patterns only really get
expensive when you start using complex qualifying predicates. Unnecessary
naming of the ancestor nodes doesn't hurt too much if the node is going to
match anyway. But where the ancestors are actually necessary to discriminate
several rules, the impact will be bigger because you will be doing more
unsuccessful probing. If performance is a concern, avoid overloading element
names.

Mike Kay

> I am concerned about performance, although we are a bit away 
> from performance
> testing.  How much of a hit do you think that this could be?  
> (Using IE5)  The
> reason for the detailed addressing is that certain elements 
> (addresses, address)
> appear in several contexts, and are handled differently each time.
> 
> 
> (Do you really need to name all the ancestors? It will only 
> slow down the pattern matching...)
> 
> Mike Kay
> 
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 
> 
> 
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 


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