SAX TypeHandler was: Alternatives to XML Schemas

[Jonathan Borden <jborden@mediaone.net>]

Rick Jelliffe wrote:

>
> The reason why XML Schemas will initially be used for validation most is
> because, until we get DOM and SAX APIs for the PSVI or XSLT 2 and XPath2
> come along (for better or worse), all PSVI solutions are proprietary or
> hidden inside an application and so invisible.

Exactly. The Infoset is an abstraction which becomes real when you use a DOM
or SAX or whatever XML interface you choose. When type information is
exposed in DOM Level n, programs which have no use for it will still work
using DOM Level n-1. The immediate danger is that the PSVI, having no
physical incarnation, is just this, an abstraction.

For example if we define a SAX interface for type information such as
"TypeHandler", then we can see how such information is to be used in an XML
Schema independent fashion (for example a RELAX or TREX or Schematron
processor might emit the same type information as an XML Schema processor).

A piece of software might indicate its interest in type information by its
implementation of the TypeHandler interface (i.e. the default is a stub). To
an uninterested application, no harm done.

One could also define an interface by which a schema processor might
indicate per element/attribute validity -- though I'm still unsure what a
'partially valid' document means -- except to something like an editor.

-Jonathan