xs:language validation support

[Thropp, Shawn]

I am trying to figure out a message received from Xerces-J version 2.6.2 when the parser encountered a language code that was not valid according to the xs:language data type.  The error that I received was:

 

Error] ext_example1.xml:27:71: cvc-pattern-valid: Value 'eeeeeeeeeeeeeeeeeeeee' is not facet-valid with respect to pattern '([a-zA-Z]{1,8})(-[a-zA-Z0-9]{1,8})*' for type 'language'.

 

However when I look at

XML Schema Part 2: Datatypes, A Schema for Datatype Definitions (normative):

 

I see the following regular _expression_ defined (normative):

([a-zA-Z]{2}|[iI]-[a-zA-Z]+|[xX]-[a-zA-Z]{1,8})(-[a-zA-Z]{1,8})*

 

<xs:simpleType name="language" id="language">
    <xs:annotation>
      <xs:documentation
        source="http://www.w3.org/TR/xmlschema-2/#language"/>
    </xs:annotation>
    <xs:restriction base="xs:token">
      <xs:pattern
        value="([a-zA-Z]{2}|[iI]-[a-zA-Z]+|[xX]-[a-zA-Z]{1,8})(-[a-zA-Z]{1,8})*"
                id="language.pattern">
        <xs:annotation>
          <xs:documentation
                source="http://www.w3.org/TR/REC-xml#NT-LanguageID">
            pattern specifies the content of section 2.12 of XML 1.0e2
            and RFC 1766
          </xs:documentation>
        </xs:annotation>
      </xs:pattern>
    </xs:restriction>
  </xs:simpleType>

 

Obviously the two regular _expression_ used are different.  It seems as though, by the error poduced by Xerces-J, that Xerces-J parser code is overriding the normative data type (regular _expression_) with its own?  Why are the two different?  Or do I not understand what is going on?

 

Thanks for the help

Schawn