Re: [xsl] xslt2 grouping

[David Carlisle <davidc@xxxxxxxxx>]

>  The XPath 2.0 spec indicates that each node has its own node identity. So, 
>  since these nodes each has a separate identity they are not, in my view at 
>  least, duplicates.

That's nearly the answer I gave but I changed it just before posting to
say that they were not duplicates because they differ in the only
relevant property: the position in the original sequence.
Even if teh nodes had teh same node identity they would not be
duplicates in teh sense that Dave meant.

As it's XSLT2 and the top level select is selecting a sequence
rather than a set, the original sequence can have duplicates, in fact
it might be 
select="for i in 1 to 10 return ."
and be 10 copies of the current node.
Grouping this sequence may be pathologically strange but I think it
should be legal and you should end up with 10 copies of this node
distributed amongst the resulting group sequences.


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
________________________________________________________________________

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