On Tue June 30 2009 12:01:34 pm Anthony Davidson wrote: > The deficiency in the class EndpointReferenceDomainExpression is still a > problem though. The ws addressing schema ( > http://www.w3.org/2006/03/addressing/ws-addr.xsd) indicates that xs:anyURI > is a valid value for the Address element.
Actually, the schema says the TYPE (actually baseType) of the Address element is anyURI. It doesn't say that anyURI is a valid value. Thus, the value in the type needs to be a valid URI according to XML Schema rules: http://www.w3.org/TR/xmlschema-2/#anyURI "The ·lexical space· of anyURI is finite-length character sequences which, when the algorithm defined in Section 5.4 of [XML Linking Language] is applied to them, result in strings which are legal URIs according to [RFC 2396], as amended by [RFC 2732]." I don't think "xsd:anyURI" (which is a QName) would meet that criteria. > Since CXF isn't honoring this value, it seems this is a bug. While not a bug, it's definitely something that I also consider a deficiency. I'm wondering if we could treat it as a regex pattern or something that could be applied. Thus, you could specify ".*" for everything, or "http://.*"; for all http and "https://"; for all https or similar. -- Daniel Kulp [email protected] http://www.dankulp.com/blog
