"Jeremy Hughes" <[EMAIL PROTECTED]> wrote on 16/08/2006 12:54:37:
> On 8/15/06, Graham Turrell <[EMAIL PROTECTED]> wrote:
>
> >
> > My reason for asking is germaine to current work I'm doing on the URI
> > resolver (WODEN-14 [1]): Sax provides a very similar resolver
framework,
> > and I need
> > to either plug my resolver directly into XMLSchemaCollection
>
> AFAIR XmlSchema has its own resolver mechanism see:
> org.apache.ws.commons.schema.resolver.* package. I guess the Woden
> resolver is going to have to work with the XmlSchema resolver in order
> to produce a fully resolved tree of WSDL and schema. Is this what
> you're getting at?
Yes you remember right. As a bit of background,
o.a.ws.commons.schema.XMLSchemaCollection is used in (D)OMWSDLReader to
"cache" schemas. It uses
org.apache.ws.commons.schema.resolver.DefaultURIResolver as its default
resolver (which basically just wraps the URI.resolve() method, virtually a
no-op). Since we don't replace it, it is this resolver currently used in
the Woden (D)OMWSDLReader classes. However the XMLSchemaCollection API
allows any custom resolver to be plugged in that conforms to the
URIResolver interface provided by that package. If we plan to use
XMLSchemaCollection as a central part of woden parsing then it is more
reasonable to base the WODEN-14 URIResolver framework on that interface.
If we were starting from scratch, i'd want a "greenfield" resolve method
with signature like:
(1)
public URI resolveURI(URI uri);
(In fact this is what I currently have in my sandbox).
In order to work directly with XMLSchemaCollection, a resolve method is
needed of signature:
(2)
org.xml.sax.InputSource resolveEntity(String targetNamespace,
String schemaLocation,
String baseUri
);
The differences are not vast, and the org.xml.sax package forms part of
JAXP bundled as part of J2SE 1.4.2 and above, so clearly it is here to
stay!
Jeremy, I think in answering your follow-up question, I have answered my
own. But I'd still like to hear any general preferences between (1) and (2)
above (especially with regard to returning a sax object)...
All comments welcome as ever!
Thanks,
Graham.
> Cheers,
> Jeremy
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
