RE: [xml-dev] Intuition and XML Schema restriction (was: XML Schema to relaxNG xslt)

[Jeff Lowery <jlowery@scenicsoft.com>]

> So true. Type derivation involving anonymous types is another example 
> where the derivation looks fine but is in fact incorrect.

And the funny thing is, I would have sworn that 6 mos. ago most validators
handled restrictions of anonymous types just fine.  Maybe I'm
misremembering. I know that XML Spy 4.3  doesn't complain. 

I still don't  know which validators can now correctly determine value space
subsets from pattern expressions, though (it may be a terrifically hard
problem to solve for all cases).  I can do a quick check (incorrectly
restricting pattern "[a-m]*" to "[a-z]*") with the three validators I have
here (all probably out of date by now):

XSV 1.203.2.37:  Nope, doesn't catch it.

SchemaQualityChecker version 2.0:  Nope.

XML Spy 4.3: Nada.

Anybody want to try the following with Xerces or MSXML?:

<?xml version="1.0" encoding="UTF-8"?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema";
xmlns:qux="http://www.example.com/foobar";
targetNamespace="http://www.example.com/foobar";
elementFormDefault="qualified"  attributeFormDefault="unqualified">
	
	<xs:simpleType name="name_t">
		<xs:restriction base="xs:string">
			<xs:pattern value="[a-m]*"/>
		</xs:restriction>
	</xs:simpleType>
	
	<xs:complexType name="foo">
		<xs:sequence>
			<xs:element name="firstName" type="qux:name_t"/>
		</xs:sequence>
	</xs:complexType>
	
	<xs:complexType name="bar">
		<xs:complexContent>
			<xs:restriction base="qux:foo">
				<xs:sequence>
					<xs:element name="firstName">
						<xs:simpleType>
							<xs:restriction
base="qux:name_t">
								<xs:pattern
value="[a-z]*"/>
							</xs:restriction>
						</xs:simpleType>
					</xs:element>
				</xs:sequence>
			</xs:restriction>
		</xs:complexContent>
	</xs:complexType>
</xs:schema>