RE: computed namespace nodes

The semantics of computed namespace nodes are broken in the Nov 12th
draft. Look at the correspondence in the public-qt-comments list to see
how it's been fixed. Basically it's been replaced by a facility that is
purely static (there is no longer any ability in XQuery to compute the
namespaces in the the result document dynamically - it was felt that not
many users needed this so it should be a v2 facility).
 
Michael Kay

-----Original Message-----
From: www-ql-request@w3.org [mailto:www-ql-request@w3.org] On Behalf Of
Damien Fisher
Sent: 07 February 2004 12:41
To: www-ql@w3.org
Subject: computed namespace nodes



As I read the current specification, the semantics of computed namespace
nodes seem a little peculiar.  The ability to bind namespace URIs at
runtime poses some problems:  suppose one has a computed namespace node
(which binds some URI, only known at runtime, to a fixed prefix), and
one then calls a function within the scope of this namespace
declaration, which makes use of the prefix.  How is one to know during
the static analysis phase which function is meant, given that the URI of
the function name is not known until runtime?

 

Am I reading this wrong?

 

Damien