Jim,
On Tuesday 14 December 2010 7:07:47 pm e...@apache.org wrote:
> MustUnderstandEndingInterceptor(); +
> public MustUnderstandInterceptor() {
> super(Phase.PRE_PROTOCOL);
> }
> @@ -80,12 +83,18 @@ public class MustUnderstandInterceptor e
> checkUnderstand(mustUnderstandHeaders, mustUnderstandQNames,
> notUnderstandHeaders);
>
> if (!notUnderstandHeaders.isEmpty()) {
> - throw new SoapFault(new Message("MUST_UNDERSTAND", BUNDLE,
> notUnderstandHeaders), -
> soapVersion.getMustUnderstand());
> + SoapFault soapFault = new SoapFault(new
> Message("MUST_UNDERSTAND", BUNDLE, notUnderstandHeaders), +
> soapVersion.getMustUnderstand()); +
> if (!isRequestor(soapMessage)) {
> + soapMessage.put(MustUnderstandInterceptor.FAULT,
> soapFault); + } else {
> + throw soapFault;
> + }
> }
> if (!ultimateReceiverHeaders.isEmpty() &&
> !isRequestor(soapMessage)) {
> checkUltimateReceiverHeaders(ultimateReceiverHeaders,
> mustUnderstandQNames, soapMessage); }
> + soapMessage.getInterceptorChain().add(ending);
> }
Can this be optimized a bit? Basically, could the call to:
soapMessage.getInterceptorChain().add(ending);
be placed into the
if (!isRequestor(soapMessage)) {
block? Since the interceptor only does anything if that block is called, we
don't need to add it for every invokation. Right?
--
Daniel Kulp
dk...@apache.org
http://dankulp.com/blog
