- From: <noah_mendelsohn@us.ibm.com>
- Date: Tue, 24 Sep 2002 15:56:55 -0400
- To: Mark Feblowitz <mfeblowitz@frictionless.com>
- Cc: jeni@jenitennison.com, mrowell@openapplications.org, Ryan.Barr@ejgallo.com, xmlschema-dev@w3.org
Mark Feblowitz writes: >> As a practical matter, large namespaces are >> much more managable when defined across several, >> separate files Yes, which is why we provide a separate <include> mechanism that is intended to be used when putting together the several schema documents comprising the declarations of a single namespace. Significantly, you will note that schemaLocation on <include> is NOT a hint...it is mandatory. Multiple includes do what you want. The (often misunderstood) purpose of <import namespace="a"> is to make clear to the processor that namespace a is part of the vocabulary being defined, as opposed to the vocabularies used in the schema document itself (typically xsd:, but also potentially including namespaces used to extend xsd:, namespaces used within annotations, etc.) The schemaLocation on <import> is optional, and as you observe is a hint. It is NOT intended as a composition mechanism for multiple schema documents defining a single namespace. <include> and <redefine> are provided for that purpose. I hope this helps. ------------------------------------------------------------------ Noah Mendelsohn Voice: 1-617-693-4036 IBM Corporation Fax: 1-617-693-8676 One Rogers Street Cambridge, MA 02142 ------------------------------------------------------------------
Received on Tuesday, 24 September 2002 16:06:09 UTC