Hi Dominic, In the XMLBeans implementation you can declare the namespace prefix directly in the Xpath call. The Xpath expression we use are therefore constructed as follows: declare namespace foo='http://bar.org'; $this/foo:bar I did not find the place in the Xpath specification where the namespace declaration is defined, but http://www.w3.org/TR/xquery-xpath-parsing/ indicates that this should be standard conform. However, this expressions work for XMLBeans. I think other implementations should support this as well (Saxon?). Best regards, Oliver -----Ursprüngliche Nachricht----- Von: graap-wg-bounces@ogf.org [mailto:graap-wg-bounces@ogf.org] Im Auftrag von Dominic Battre Gesendet: Donnerstag, 23. August 2007 00:05 An: GRAAP-WG Betreff: [GRAAP-WG] XPath issues in Creation Constraints Hi. As we are looking for interoperability tests and implementations start showing up, I have a question about common practice in the evaluation of creation constraints. I have raised the issue before, but I think I have never received an answer. The issue is: A creation constraint might have an xpath "//wsag:foobar" but it is very difficult to get the namespace that is bound to the "wsag:" prefix. At least in case Axis 1 that is used. Does Axis 2 solve this problem or any other toolkit? Otherwise the question is whether it is technically feasible to evaluate XPath expressions like that at all. If it is not feasible, it might be considered a problem in the WS-Agreement spec. The reason for my writing this is that I found this bug report: http://bugzilla.globus.org/globus/show_bug.cgi?id=4227 The Globus Toolkit community faced the same issue and surrendered - or lets say, they came up with a very nice solution: Introducing a sane query language TargetedXPath. This is the Schema: http://src.gridindex.org/globus_4.1.1/xref/src/wsrf/schema/core/query/ This is the Implementation: http://src.gridindex.org/globus_4.1.1/xref/src/wsrf/java/core/source/src/org /globus/wsrf/query/targetedXPath/ I like this approach a lot. Would this be a viable approach for WS-Agreement? Best regards, Dominic -- graap-wg mailing list graap-wg@ogf.org http://www.ogf.org/mailman/listinfo/graap-wg