Hi Denis, I think this case is still valid. According to the XML Namespaces spec [1] a QName can be prefixed or unprefixed. In case of an unprefixed QName, the namespace is assumed to be either the default namespace or the null namespace. In our case its the latter. But its still a valid QName, isn't it?
Besides, since the fault is never declared nor caught it cannot cause any problems. It would even be allowed to catch it in a fault handler within BPEL. I think its only purpose in this case is, similar to JUnit's fail() to force the process instance to fail if it reaches a region that should not be reached because an expected exception has been thrown. Best, Tammo [1] http://www.w3.org/TR/REC-xml-names/#ns-qualnames On 28.12.2011 07:54, Denis Weerasiri wrote: > Hi, > Based on "10.6. Signaling Internal Faults – Throw" > http://docs.oasis-open.org/wsbpel/2.0/OS/wsbpel-v2.0-OS.html#_Toc164738508 in > WS-BPEL 2.0 specification, > the faultName should be ns qualified. But in some of the samples, faultName > is declared as ns unqualified. > > eg - > https://svn.apache.org/repos/asf/ode/trunk/bpel-test/src/test/resources/bpel/2.0/TestIMA/TestIMA.bpel > refer <bpel:throw faultName="error"/> > which should have been <bpel:throw faultName="tns:error"/> > > Is it ok to modify this sample with proper ns qualified faultNames? > -- Tammo van Lessen - http://www.taval.de
