RE: Using a tree read with document() as a hash-table -- wor

> > On a final note - would a key() make this faster?  I still 
> > have a bit of a problem understanding how it works.
> > 
> Yes, it would be faster, assuming that the lookup table is of 
> some size and
> is used fairly often. (In fact, it's not a hash-lookup at all 
> unless you use
> a key - it's likely to be a serial search on most processors).

Thank you!

In case anyone is interested: 

The external XML-file has 78 ROW entries, and the code is used 75000 times.  The running time decreased to a third of the previous one with Saxon, which is quite satisfactory for me.  Apparently, using key() rules out XT, which is rather sad.

Now it is the time to attend to the memory footprint :-)

Kudos to the knowledge and responsiveness on this list.


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