Yes, the requirement is to identify a particle uniquely, not just an element
declaration.

However, some schema processors including XSV and Saxon use an algorithm
that only requires the element declaration to be identified uniquely, and
therefore let this one through.

Michael Kay
http://www.saxonica.com/
 

> -----Original Message-----
> From: xmlschema-dev-request@w... 
> [mailto:xmlschema-dev-request@w...] On Behalf Of Zafar Abbas
> Sent: 09 February 2005 18:49
> To: xmlschema-dev@w...
> Subject: Unique Particle Attribution
> 
> 
> >From my reading of the Unique Particle Attribution constraint in the
> spec, the following schema is a violation:
> 
> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema">
> 
> <xsd:complexType name="type">     
> <xsd:sequence>
> 	<xsd:element ref="a" minOccurs="0"/>
> 	<xsd:element ref="b" minOccurs="0"/>
> 	<xsd:element ref="a" maxOccurs="2"/>
> </xsd:sequence>
> </xsd:complexType>
> <xsd:element name="a" />
> <xsd:element name="b" />
> </xsd:schema>
> 
> 
> It can not be known which particle (a) to validate, even through they
> are references to the same element schema component. Is this
> understanding correct?
> 
> Thanks.
> 
> 


From noah_mendelsohn@u... Wed Feb 09 20:00:28 2005
Received: from lisa