Re: startPrefixMapping with redundant namespace declarations

> Karl Waclawek wrote:
> >
> > I would say: As long as the spec doesn't treat redundant NS
> > declarations differently, then neither should the implementation.
> >
> 
> Fair point.  But we took the startPrefixMapping event to mean that a
> *new*prefix mapping was coming into effect.  I think this is best thought of
> as a logical rather than a lexical event, and therefore the redundant
> declaration is a non event.

With SAX there is always the problem of how close to the document's
lexical structure should it report. For some applications, like XML editors,
the lexical aspect is more important, and maybe there it would
make sense to have the duplicate declaration reported.

For example, the element with the redundant NS declaration might be taken out
as a fragement and passed to some other application, losing the context that
provides the namespace. It would then have to provide such a declaration on
its own. Editing such a file - you may want to preserve the redundant declaration.

> However, we don't wish to go out on a limb.  If the consensus is that
> redundant declarations should be reported, then we can simply make the
> useful ElCel behaviour controllable via an option.

Sounds reasonable.

Karl



-------------------------------------------------------
This SF.NET email is sponsored by: AMD - Your access to the experts
on Hammer Technology! Open Source & Linux Developers, register now
for the AMD Developer Symposium. Code: EX8664
http://www.developwithamd.com/developerlab
_______________________________________________
List: sax-devel, [email protected]
See:  http://www.saxproject.org/
https://lists.sourceforge.net/lists/listinfo/sax-devel