Removing empty text nodes before the document is transformed would actually solve the problem. I have worked around the problem in my code by avoiding to create these nodes.
Anders
Joseph Kesselman wrote:
Element c = document.createElement("C");
Text empty = document.createTextNode("");
c.appendChild(empty);
Hm.
Technically, XSLT *should* skip that text node, since -- as far as the XML Infoset and XPath Data Model are concerned -- it doesn't really exist; it's just an artifact of the DOM model. (Just as two adjacent Text nodes in the DOM are considered only a single text node by XSLT.)
We may or may not be enforcing that correctly. But I'd say that if there's an error here, it would be the fact that this Text node is ever visible to Xalan, not that it sometimes vanishes.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
