I'm seeing a behavior with the handling of namespaces that strikes me as a little odd. I don't remember seeing it discussed, and maybe it's by design, so I thought it would be worth running it by the list.
I have a document with its own schema and namespace that has elements that can contain XHTML children. I use the DOM to extract those children and import them into XHTML documents that I'm assembling. The root element of the target document declares the XHTML namespace as the default namespace. If the element in the source document that contains the XHTML fragment to be extracted makes the XHTML namespace the default, everything works as I'd expect. In particular, the imported fragment has the correct namespace and gets serialized without a namespace declaration. If, however, the XHTML namespace declaration in the source document is on an element that is extracted and then imported, the xmlns declaration is treated as an attribute and serialized. (Namespace processing also occurs correctly.) The end result is an output document that has a redundant default namespace declaration. It's not harmful, of course, but it's a bit surprising, and I'd like to avoid it if possible. I'm about to head off on vacation and I don't think it'll be trivial to work up a sample, so before I do, I thought I'd check whether this rings bells with anyone. Here are two contrived illustrations, source document followed by output document. The first pair illustrates the expected serialization that results when XHTML elements inherit the correct default namespace. <?xml version="1.0" encoding="UTF-8"?> <src:source xmlns ="http://www.w3.org/1999/xhtml"; xmlns:src="http://www.example.com/source";> <p>Some text to be extracted and imported into a target document.</p> </src:source> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd";> <html lang="en" xml:lang="en" xmlns="http://www.w3.org/1999/xhtml";> <body> <p>Some text to be extracted and imported into a target document.</p> </body> </html> The second pair illustrates the surprising serialization that results when XHTML elements declare their own default namespace. <?xml version="1.0" encoding="UTF-8"?> <source xmlns="http://www.example.com/source";> <p xmlns="http://www.w3.org/1999/xhtml";>Some text to be extracted and imported into a target document.</p> </source> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd";> <html lang="en" xml:lang="en" xmlns="http://www.w3.org/1999/xhtml";> <body> <p xmlns="http://www.w3.org/1999/xhtml";>Some text to be extracted and imported into a target document.</p> </body> </html>
