attribute values as qnames?

[Mark D. Anderson]

> To reference the name of an element type in an
> attribute, one can use the namespace prefix: this is what
> XPaths do, for example.  So the XML schema processor 
> may indeed have to have the xmlns prefix->URI mappings 
> available. A namespace processor will not resolve values
> of attributes, merely names of elements and attributes (AFAIK,
> but I am easily confusable.)

Just to make this concrete...

Suppose (in the future) I have a document which indicates that a particular
attribute is of the qname type, via xmlschema.

Now suppose I process this via a SAX or XSLT processor. If that
processor is "xmlschema-aware", then it will expand those qname's
in attribute values in the same way it would element type names and
attribute names. I would write xslt patterns using whatever prefix
I like.

But if the SAX or XSLT processor is not "xmlschema-aware", then I
would have to match against them using a lexical pattern that happens
to have a colon in it (i.e. choose the same prefix).

Is that right?

It also seems like this would have implications for xsl:key, in
ways I can't quite suss out.

This is of interest to me not just for use of an attribute value
as a reference. I would like to have an "enum" list where the values
are qualified. These are basically URNs. Maybe that is an abuse of
the namespace mechanism though?

-mda


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)