[
Lists Home |
Date Index |
Thread Index
]
No, there is no trouble creating the instance document. You add a dummy
empty global element (child of the <schema> element) to the overall schema
being created:
<element name="dummy-to-be-excised" type="string"/>
and make your instance document contain just this element. (If the overall
schema contains such an element already, just use it.)
When the validator reads this document, it must read the overall schema and
everything it imports and ensure they are self-consistent and valid (at
least xerces does), before looking at the instance. Thus the entire set of
constructs in the various schemas will be evaluated for completeness.
So the procedure I described does not depend upon a complex instance
document.
Jeff
----- Original Message -----
From: "Thomas B. Passin" <tpassin@comcast.net>
To: <xml-dev@lists.xml.org>
Sent: Wednesday, July 30, 2003 9:45 PM
Subject: Re: [xml-dev] Extract A Subset of a W3C XML Schema?
> Trouble is, you have to
> create the instance document (which might be a lot easier than messing
with
> the schema, though), and usually the generated schema's structure is not
> much like what you want to end up with.
>
|
