W3C home > Mailing lists > Public > public-xml-schema-testsuite@w3.org > January 2009

bug in MS-Wildcard/wildZ013a?

From: Tobias Koenig <tobias.koenig@trolltech.com>
Date: Mon, 19 Jan 2009 17:18:51 +0100
To: public-xml-schema-testsuite@w3.org
Message-Id: <200901191718.51896.tobias.koenig@trolltech.com>

Hej,

wildZ013a schema from the MS-Wildcard section contains the following entries:

<xs:complexType name="base">
	<xs:sequence>
		<xs:element name="sub" type="a:derived2" minOccurs="0"/>
		<xs:element name="sub2" type="a:derived3" minOccurs="0"/>
		<xs:element name="sub3" type="a:derived4" minOccurs="0"/>
		<xs:element name="sub4" type="a:derived5" minOccurs="0"/>
		<xs:element name="sub5" type="a:intersection1" minOccurs="0"/>
		<xs:element name="sub6" type="a:intersection2" minOccurs="0"/>
	</xs:sequence>
	<xs:attributeGroup ref="a:attG-a1"/>
 </xs:complexType>

 <xs:complexType name="derived">
	<xs:complexContent>
		<xs:extension base="a:base">
			<xs:sequence />
			<xs:attributeGroup ref="a:attG-a2"/>
		</xs:extension>
	</xs:complexContent>
 </xs:complexType>

 <xs:attributeGroup name="attG-a1">
	<xs:anyAttribute namespace="##other" processContents="lax"/>
 </xs:attributeGroup>

 <xs:attributeGroup name="attG-a2">
	<xs:anyAttribute namespace="##targetNamespace b c" processContents="lax"/>
 </xs:attributeGroup>

So one have to test if the namespace contsraint from anyAttribute of attG-a1 
is a valid subset if the anyAttribute of attG-a2 according to algorithm 
http://www.w3.org/TR/xmlschema11-1/#cos-ns-subset

In this case our _sub_ is a pair of 'not' and the targetNamespace and _super_ 
the targetNamespace, b and c.

However there is no rule for _sub_ having variety not and _super_ having 
variety enumeration, so the rule is false and therfor the whole schema 
afaiks.
But the meta data says it is valid, so where is the point I'm missing?

Ciao,
Tobias
Received on Monday, 19 January 2009 16:19:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 19 January 2009 16:19:30 GMT