[ https://issues.apache.org/jira/browse/WSCOMMONS-120?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14632527#comment-14632527 ]
Hudson commented on WSCOMMONS-120: ---------------------------------- SUCCESS: Integrated in woden-trunk #1211 (See [https://builds.apache.org/job/woden-trunk/1211/]) Fix a regression for WODEN-13/WSCOMMONS-120. (veithen: rev 1691753) * /webservices/woden/trunk/java/woden-core/src/main/java/org/apache/woden/internal/wsdl20/ComponentModelBuilder.java * /webservices/woden/trunk/java/woden-core/src/test/java/org/apache/woden/MultiSchemaTest.java > XmlSchema.equals(Object) returns a false positive result > -------------------------------------------------------- > > Key: WSCOMMONS-120 > URL: https://issues.apache.org/jira/browse/WSCOMMONS-120 > Project: WS-Commons > Issue Type: Bug > Components: XmlSchema > Reporter: John Kaputin > Assignee: Davanum Srinivas > Attachments: XmlSchema_equals_patch.txt > > > The XmlSchema class inherits the equals(Object) method from XmlSchemaObject. > This method uses line number, line position and sourceURI for equality > checking, however line number and position are always zero (they are not > used yet?). This can cause a false positive when checking equality of two > different schemas that share the same source URI. > For example, the WSDL 2.0 testcase at [1] has two inlined schemas. For > XmlSchema1.equals(XmlSchema2), both XmlSchema objects will have line number > zero, line position zero and the WSDL URI as their sourceURI and > XmlSchemaObject.equals will return true, even though in this case the two > inlined schemas have different target namespaces and xs:include different > schema documents. > [1] > http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/test-suite/documents/good/Chameleon-4G/getBalance.wsdl > I think a solution is to create an equals method on XmlSchema which invokes > super.equals (i.e. from XmlSchemaObject) and adds further equality checking > specific to XmlSchema. I have tried the attached patch which fixes my > testcase, although it may require more equality checking logic in the equals > method > to be more useful generally. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@ws.apache.org For additional commands, e-mail: dev-h...@ws.apache.org