Hi Edwin,
Here's what we did in Xerces1: since publicIds are (or at least were) very
DTD-specific, we don't use that field of an EntityResolver in this context.
We passed the namespace of the imported schema in the systemId field, and
only used the schemaLocation field ourselves if this other method didn't
provide us with a schema. The thought behind this, I conjecture, was that
the application-writer, who supplies both the schemas and the
EntityResolver, should be able to keep a mapping between his namespaces and
his files (or has some method of doing so; else why provide the entity
resolver?) A schema hint is really only useful to a default
implementation--like ours.
Hope that helps,
Neil
Neil Graham
XML Parser Development
IBM Toronto Lab
Phone: 905-413-3519, T/L 969-3519
E-mail: [EMAIL PROTECTED]
Edwin Goei <[EMAIL PROTECTED]> on 10/10/2001 04:52:32 AM
Please respond to [EMAIL PROTECTED]
To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
cc:
Subject: xsi:schemaLocation hint resolution
Do the {namespace-name schemaURI} pairs that are contained in the value
of an xsi:schemaLocation hint get passed to an EntityResolver? I assume
that the "publicID" parameter of the resolveEntity() method gets
assigned to the namespace-name in this case. What does "publicID" get
assigned to for the xsi:noNamespaceSchemaLocation case?
-Edwin
---------------------------------------------------------------------
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]
