This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
Testcases affected: msData/identityConstraint/idI014.xsd msData/identityConstraint/idI016.xsd msData/identityConstraint/idI022.xsd msData/identityConstraint/idJ015.xsd msData/identityConstraint/idJ017.xsd msData/identityConstraint/idJ030.xsd According to the "3.11.6 Constraints on Identity-constraint Definition Schema Components", the above schemas have valid xpath combinations and should be tested using valid scenarios.
14 and 16 at least look to me to have unbound ns prefixes in their XPaths, but I agree that 22 looks OK -- needs careful review
Hmmm, re-reading the XPath whitespace rules leaves me less sure about 22 -- care is clearly needed here
Michael Kay will run these tests through his processor and report the results.
Saxon reports all these tests as invalid. They all fail for the same reason: whitespace is used where it is not allowed, either within a NameTest or within the token "//". Specifically, the offending lines are: I014: <xsd:selector xpath="tid :*"/> I016: <xsd:selector xpath="tid : *"/> I022: <xsd:selector xpath="./ /."/> J015: <xsd:field xpath="xpns :*"/> J017: <xsd:field xpath="xpns : *"/> J030: <xsd:field xpath="./ /."/> XPath 1.0 does not allow a space in the middle of these constructs (NameTest and // are both tokens). XSD 1.0 says the same thing. The relevant rules are in 3.11.6: <quote> For readability, whitespace may be used in selector XPath expressions even though not explicitly allowed by the grammar: whitespace may be freely added within patterns before or after any token. [5] token ::= '.' | '/' | '//' | '|' | '@' | NameTest [6] whitespace ::= S </quote>
WG decided that the tests, marked as invalid, are appropriately so marked.