At one point, ViolaWWW used a real SGML parser (sgmls) entirely
(without CERN's parser) for its parsing of HTML (using Dan's DTD) and
also HMML (something I had cooked up; this was pre HTML+ DTD days).
Trouble was, it seemed few people were validating their HTML documents
with a DTD. Consequently, viola had a hard time with lots of the HTML
documents on the web...
So, in order to be more roubust, that SGML parsing process was put on ice,
and ViolaWWW is now back to using CERN's parser again. Thou, I ended up
making a bunch of modifications to the CERN parser s.t.it would know things
like what tag elements are valid within what tag elements. This is so that
the parser can generate implied close tags.
Awaiting for a DTD to solidify, *and people to actually validate with it*,
and an efficient SGML parser implementation...
John Franks wrote:
> Writing a browser is very difficult.
> It is not clear just how much of HTML+ will actually be
> implemented in any form in popular browsers.
> But unless you plan on writing your own browser and distributing it
> to your target audience, you may want to try to *persuade* browser
> writers that there is something in it for them if they put in all this
> extra work.
ViolaWWW supports lots of HTML+ now, including tables, but no math.
For the long run, if we can agree on some sort of architectural form,
and have a nice SGML parser to attach to the browser, then we're closer
to multiple DTD capable browsers.
-Pei