> - get rid of unused namespaces by using exclude-result-prefixes on the > stylesheet > - if you create an element in a certain namespace, you can _not_ change the > namespace later on as the namespace have become an integral part of the > element. I'm already using exclude-result-prefixes. Also, this is not an "elements created in a certain namespace problem", I think. Here's an example:
XML document: <?xml version="1.0" encoding="UTF-8"?> <my:entry xmlns:my="http://mynamespace.org/"; xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; xsi:schemaLocation="test.xsd"> <my:content xmlns="http://www.w3.org/1999/xhtml";> <h1>example data</h1> <p>hello there</p> <p>and one more</p> </my:content> </my:entry> When I deep-copy the insides of this thing's <content> element, I get something like this: <h1 xmlns:my="http://mynamespace.org/"; xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance";>example data</h1> <p xmlns:my="http://mynamespace.org/"; xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance";>hello there</p> <p xmlns:my="http://mynamespace.org/"; xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance";>and one more</p> >From what I believe, the xhtml elements inside content are in the default (xhtml) namespace, and not in the my: and neither in the xsi: namespace.Also, the xmlns:* 'attributes' are just hints and are in no way related to the actual namespace the elements are in. The element name prefix does that. So anyways, the above is what I get if I don't use an identity transformation that strips namespaces. My original question was whether there is a cleaner solution than copying (manual, recursive, likely slower than the built-in xsl:copy-of operation) to remove these redundant namespace declarations from the xhtml output. _______________________________________________ xslt mailing list, project page http://xmlsoft.org/XSLT/ [email protected] http://mail.gnome.org/mailman/listinfo/xslt
