- From: Florent Georges <fgeorges@fgeorges.org>
- Date: Tue, 1 Dec 2009 23:21:35 +0100
- To: Norman Walsh <ndw@nwalsh.com>
- Cc: public-xml-processing-model-comments@w3.org
2009/12/1 Norman Walsh wrote: > Florent Georges writes: >> Stricto sensu, that cannot be a QName for real, as this is in >> content. > XProc says explicitly that the content values that are > lexically QNames must be valid QNames with respect to the > in-scope namespaces on the element on which they occur. Well, in that case, of course... :-/ Thanks for the correction! Anyway, the initial worries were about the ability to generate the right QName, with the right in-scope bindings, for new-name, using XSLT. With XSLT 2.0 this is not a problem. Not sure being able to use XSLT 1.0 to generate XProc pipelines containing p:rename does really justify to change p:rename definition. But if I am right, @namespace was introduced in XSLT because of AVTs: to be able to use either a URI or a prefix provided at runtime, without having to get a static in-scope binding in the stylesheet. I guess there is the same problem here when using p:with-option for new-name, isn't there? Regards, -- Florent Georges http://www.fgeorges.org/
Received on Tuesday, 1 December 2009 22:22:08 UTC