Also I am guessing that
<xsl:key> would be a lot of overhead.
It trades space for time. You need to have space to make the index, but
if you have the space looking up nodes via key() can be orders of
magnitude faster than searching for them repeatedly using an axis such
as preceding-sibling.
David
________________________________________________________________________
This e-mail has been scanned for all viruses by Star Internet. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________