Hello --

On Fri, 16 Nov 2001, Tobias Reif wrote:

> David Alan Black wrote:
>
>
> >   Here's an ampersand and a greater-than sign: & >
> >
> > one would expect to end up with this, on output:
> >
> >   Here's an ampersand and a greater-than sign: & >

Funny, I never received my own mail.  Anyway...

> Hmmm... I have to go read. I thought it's OK for a parser+writer to
> leave them like they are. (like with NCRs). They're predefined named
> entities, that point to the corresponding NCRs, so perhaps substitute
> them with the NCRs?
> Otherwise
> <a>&lt;</a>
> in the input,
> becomes
> <a><</a>
> in the output, which is not well-formed.

But the output doesn't have to be well-formed, only the input.  As far
as I can tell, entity replacement is a pretty central thing to the
activity of an XML parser.  I guess in an XML transformation you'd
want to leave them intact, but for parsing as such I think it's
assumed that entities will get replaced.


David

-- 
David Alan Black
home: dblack / candle.superlink.net
work: blackdav / shu.edu
Web:  http://pirate.shu.edu/~blackdav