Re: Re: Xmlc: xHTML problem again

On St, 2003-04-02 at 10:53, David Li wrote:
> DocumentInfo is a XMLC interface, not a standard DOM. There is a way to  
> create DocumentInfo from a DOM Document object. It's used  
> GenericLinkedHTMLDocument to wrap the Document object and I think  
> GenericLinkedHTMLDocument implements the DocumentInfo interface.

I am not sure how this could help me fixing the problem. As I pointed
out earlier, there are various checks for "instanceof XMLObject" in the
Enhydra source code. And the XHTML is treated like an XMLObject. The
main question is whether XHTML should be more like an XMLObject or
rather like a standard HTML. I would vote for standard HTML like
handling.

Another question is what Document should be returned for a node that is
instance of both XMLObject and Document. Currently the code checks the
XMLObject first and then returns the getDocument of the node. Now I have
patched Enhydra so it returns the node itself (since XHTML is instance
of Document) and URLRewriter started working. Is this a good solution or
should we better fix the document of the XMLC compiled page to implement
the DocumentInfo? In such case it would be the XMLC that would have to
be patched? Or the xerces parser? Well, I don't understand this at all.

Petr