From: williamkempf_at_[hidden]
Date: 2001-09-26 12:31:01
--- In boost_at_y..., "Damien Fisher" <dfisher_at_u...> wrote:
> Perhaps it would be a good idea to ignore unicode until boost
finalizes its
> solution on the matter. Until then, an ASCII version would
definitely
> suffice, and I don't imagine it being difficult to move over. So
the next
> question would be: do we start with W3C interfaces first (DOM or
SAX or
> both) or come up with a new interface, and build DOM/SAX on top of
that?
Personally, I'd start with at least SAX. SAX is a very efficient
model that can easily be employed to implement any other model. It's
interface may not be overly palatable for general application
development in C++, but as a basis for a library that is it would be
examplary. I fear that going the other direction (i.e. develop a
different model first and then implement SAX on it) may result in
poor performance or require concessions in the new model to allow for
full implementation of SAX.
Bill Kempf