Namespaces and DTDs

[Ronald Bourret]

Elliotte Rusty Harold wrote:

> I have several DTDs with conflicting definitions of certain elements.
> (e.g one defines a HEAD as a TITLE followed by a META and another
> defines a HEAD as #PCDATA). I need to use all the DTDs and associated
> markup languages for a single document.
>
> To an extent I can disambiguate them with namespaces.  However, is there
> any way I can do this while still validating against the orignal DTDs?
> That is without rewriting the DTDs to use the qualified names instead of
> the orignal names that are in the DTDs? I've been trying to work with
> default values for xmlns attributes, and the like; but that doesn't seem
> to get me quite all the way to where I need to go. Am I going to have to
> break down and just rewrite the DTDs to use the qualified names?

If you want to use a namespace-unaware parser, I don't see how you can 
avoid rewriting the DTDs.  Unless the names in the DTDs are qualified, you 
will have two elements with the same name (e.g. "HEAD"), which is a 
validation error.  And even assuming that this isn't immediately flagged, I 
can see no way for a namespace-unaware parser to figure out which content 
model to validate against when it encounters one of the duplicated element 
names: If prefixes are used, the name won't match any of the DTD names; if 
prefixes are not used (due to use of defaults), the name will match 
multiple DTD names.

Note that this problem is not limited just to validation.  At the very 
least, it applies to retrieving default attribute values as well.

-- Ron Bourret


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)