Re: [xml-dev] External subset processing by browsers

["Andrew Welch" <andrew.j.welch@gmail.com>]

Hi George,

> One trick that you can use is to take advantage of the fact that a parser in
> non validating mode is not required to actually read the DTD. You need to
> have a DTD specified though in the XML file. You will get the entity
> notification on the ContentHandler skippedEntity callback.

Ahh ok thanks - almost there :) a couple of issues though:

- the root element can be rss, rdf or atom...

- the ContentHandler is already a TransformerHandler  (not sure that's
really an issue yet - but thought it's worth mentioning...)

It's shame that a DTD has to be specified, if it didn't (and the calls
were made to start/endEntity) it would be fine.  If I need to inspect
the content for the root element name then I may as well stick the
with the regex.

Also, if I'm adding a DTD, I may as well just point to the html 4 entites?

cheers
andrew