18654 2012-08-22 08:46:03 +0000 equality between anySimpleType and string 2012-09-21 16:10:48 +0000 1 1 1 Unclassified XML Schema Test Suite Sun Tests 2006-11-06 All All ASSIGNED needsAgreement P2 normal --- 1 xml.dev cmsmcq David_E3 public-xml-schema-testsuite oldest_to_newest 72531 0 xml.dev 2012-08-22 08:46:03 +0000 In test 'Positive' of group 'valueconstraint00901m1' a fixed value constraint "alpha beta" of type "xs:anySimpleType" must be compared with the value "alpha beta" of an instance specified type definition "xs:string". See http://www.w3.org/TR/xmlschema11-1/#cvc-elt: 5.2.2.2.2 If E's ·governing type definition· is a Simple Type Definition or a Complex Type Definition with {content type}.{variety} = simple, then the ·actual value· of E is equal or identical to D.{value constraint}.{value}. But the datatypes spec says: "values from different primitive data spaces are made artificially unequal even if they might otherwise be considered equal". xs:anySimpleType is a special data type, xs:string a primitive data type. So in my opinion they can't have equal or identical values. But this is a little bit different for XSD 1.0 - see http://www.w3.org/TR/xmlschema-1/#cvc-elt: 5.2.2.2.2 If the {content type} of the ·actual type definition· is a simple type definition, then the ·actual value· of the item must match the canonical lexical representation of the {value constraint} value. So in my opinion the expected result must be invalid for XSD 1.1, and valid for XSD 1.0. Best regards, Andreas Meissl 74234 1 David_E3 2012-09-21 16:10:09 +0000 (From Michael Sperberg-McQueen): We're not yet quite sure how to resolve this. Since anySimpleType is not a primitive but a special type, the rule quoted does not apply. The use of xsi:type may make a difference here, but so far the WG has not reached certainly on whether it does, or how. There is a rule in Structures 2.2.1.2 that says behavior of processors is unconstrained when validation rules implicate lexical mapping for types with undefined lexical mappings (like, for instance, the special types); that may mean that the behavior of 1.1 processors is unconstrained here, and the test case should be labeled accordingly.