My last question is the following:
In the example above the two sequences consist not of nodes,
but of *references to nodes*. This is a very useful new
datatype, not present in the XPath data model. Will it not be
nice to explain this in detail in the XSLT spec.?
I must admit I like to think in terms of "references to nodes" rather
than "nodes" but for some reason this is not the language that the data
model uses. When it talks of a sequence "containing nodes", this is
clearly not exclusive containment, since the same node can belong to
more than one sequence. But that's the terminology that's used.
Mike, this is not just a matter of terminology or a matter of our personal
tastes.
This is what items of a sequence really are: links or references to the
instances (the inclusion of atomic values here can be trivially defined --
by definition) that the sequence "contains".
So we know what an item of a sequence is but we are afraid to call it with
its real name?
I state that if the XSLT 2.0 specification fails to explain this in detail
and does not enforce the right terminology, this will affect the correct
understanding of sequences and will cause confusion leading to mistakes for
significant part of the XSLT programmer community in their daily work. For
most of us this ultimately means frustration, lost hours and money.
Can this be prevented?
=====
Cheers,
Dimitre Novatchev.
http://fxsl.sourceforge.net/ -- the home of FXSL
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list