Re: XSLT, XHTML, and default attribute values [somewha

> It's possible Mozilla implements its own custom entity resolver 
. in the eventit cannot reach a referenced DTD, 

That's a polite way of describing it:-)
It never even tries to fetch a dtd other than from its own local cache
directory if the last part of "the filename" in the system URI is the
same as the name of a file in mozillahome/res/dtd or it's public id is
one of half a dozen it knows (hardwired into the code, (variants of
xhtml and xhtml+mathml) then it uses the local copy otherwise it just
ignores the dtd and attributes don't get defaulted (this is conformant
with the xml spec) and entity refs generate an undefined entity error (this
arguably isn't conformant but the spec is a bit vague here)

> but I've never had any luck ignoring them using
> Xerces in Java without implementing a custom entity resolver.

whether or not you are supplied an easy hook, the xml spec certainly
allows such a hook, or to put it another way a non validating xml parser
that uses an entity resolver designed to return nothing for any uri is
conformant.

David
(I think we are getting more than somewhat OT]



________________________________________________________________________
This e-mail has been scanned for all viruses by Star. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
http://www.star.net.uk
________________________________________________________________________