- From: Bob Schloss <rschloss@us.ibm.com>
- Date: Mon, 29 Jul 2002 15:34:28 -0400
- To: Brenda Bell <bbell@juicesoftware.com>, xmlschema-dev@w3.org
- Message-ID: <OF20F100F3.79B25F22-ON85256C05.006B2CD7@us.ibm.com>
It is always legal to omit schemaLocation. But will that get what you want? The W3C XML Schema specification does not require a conforming processor to apply any particular approach to resolving the acquisition of schema components when <import> does not specify schemaLocation. (Even if schemaLocation is specified, there's lots of freedom). I suggest that you need to look at the schema processors you intend to use to see how they handle this. For example, Apache's Xerces processors provide an API so that you can supply an EntityResolver which allows your own code to recognize any well-known namespace URIs you care about and indicate to the processor where to get a schema document that corresponds to it. Regards, Bob Schloss IBM Thomas J. Watson Research Center Yorktown Heights, New York, USA Brenda Bell <bbell@juicesoftware.com>@w3.org on 07/29/2002 09:20:18 AM Sent by: xmlschema-dev-request@w3.org To: "Xmlschema-Dev@W3. Org (E-mail)" <xmlschema-dev@w3.org> cc: Subject: import w/o schemaLocation When you import a schema for a well-known namespace (http://www.w3. org/2001/XMLSchema), is it legal to omit the schemaLocation attribute? If the parser is "schema-aware", is it supposed to automatically handle the import as if it had been explicitely specified? Brenda Bell Sr. Software Architect Juice Software, Inc. Phone: 603.428.3994 Cell: 603.494.8206 Fax: 603.428.8713 Email: bbell@juicesoftware.com MSN: bbell@theotherbell.com
Received on Monday, 29 July 2002 15:36:07 UTC