extending existing complexType - xsd

Question

this topic is related to xsd: define an element that can be repeated an even number of times. i would like to create a new complex type using complexContent with extension. i tried this:
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" >
<xs:complexType name="evenOccurrence">
<xs:sequence minOccurs="0" maxOccurs="unbounded">
<xs:element name="B" maxOccurs="2" minOccurs="2"/>
</xs:sequence>
</xs:complexType>
<xs:complexType name="oddOcurrence">
<xs:complexContent>
<xs:extension base="evenOccurrence">
<xs:sequence>
<xs:element name="B"/>
</xs:sequence>
</xs:extension>
</xs:complexContent>
</xs:complexType>
this code generate the following error: cos-nonambig: B and B (or elements from their substitution group) violate "Unique Particle Attribution". During validation against this schema, ambiguity would be created for those two particles.. how i can fix the problem

Answer 1

This are two similar ways of doing it:
1- Without extension:
<xs:complexType name="oddOcurrence">
<xs:sequence>
<xs:element name="B" /> <!-- 1 time -->
<xs:sequence minOccurs="0" maxOccurs="unbounded">
<!-- 2n times -->
<xs:element name="B" maxOccurs="2" minOccurs="2" />
</xs:sequence>
</xs:sequence>
</xs:complexType>
2- Similar, with extension
<xs:complexType name="elementOneTime">
<xs:sequence>
<!-- 1 time -->
<xs:element name="B" />
</xs:sequence>
</xs:complexType>
<xs:complexType name="oddOcurrence">
<xs:complexContent>
<xs:extension base="elementOneTime">
<xs:sequence minOccurs="0" maxOccurs="unbounded">
<!-- 2n times -->
<xs:element name="B" maxOccurs="2" minOccurs="2" />
</xs:sequence>
</xs:extension>
</xs:complexContent>
</xs:complexType>
EDIT: ¿Why was it ambiguous?
As evenOccurrence can be empty, when the parser finds the first B element of and oddOccurance, it doesn't know if its parsing the first B of an evenOccurance or the B element that only appears once.
By putting first the element that can only appear once you remove the ambiguity, as the first B element would be analyzed and then the parser would enter in a state in which it would analyze the oddOccurence type (all the other B elements).