Sanjiva, Quoting from [1]. There is no guarantee that whoever sent the original message may not be the recipient of the reply.
/wsa:RelatesTo - This OPTIONAL (repeating) element information item contributes one abstract [relationship] property value, in the form of an (IRI, IRI) pair. The content of this element (of type xs:anyURI) conveys the [message id] of the related message. /wsa:RelatesTo/@RelationshipType - This OPTIONAL attribute (of type xs:anyURI) conveys the relationship type as an IRI. When absent, the implied value of this attribute is "http://www.w3.org/2005/08/addressing/reply". The test suite had only 2 nodes. Here we have 3. Which is not prohibited by the spec. The only reference in the SOAP binding [2] is how the fault messages are correlated to the original message. I racked my brain, but cannot see the spec prohibiting any of the examples / use cases that david mentions in [3]. Could you please do a quick review as well. Thanks, dims [1] http://www.w3.org/TR/2006/REC-ws-addr-core-20060509/ [2] http://www.w3.org/TR/2006/REC-ws-addr-soap-20060509/ [3] http://marc.theaimsgroup.com/?l=axis-dev&m=114915549611516&w=2 On 7/3/06, Sanjiva Weerawarana <[EMAIL PROTECTED]> wrote:
On Mon, 2006-07-03 at 01:45 -0400, Davanum Srinivas wrote: > Sanjiva, > > Can you please chime in on the problem itself [1]? Maybe if we > concentrate on the problem in question rather than dissect the > specific solution we will be better off? Interesting use-case. Is this the scenario that motivated David's proposal? David? Does the WS-Addr spec have anything to say about whether its legit to do this kinda stuff? The scenario is that the response message (containing a RelatesTo with @relationshipType=wsa:Response) is being used as an invocation message for someone else. Didn't the test suite have some stuff that checked proper handling of RelatesTo? Sanjiva. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
-- Davanum Srinivas : http://people.apache.org/~dims/ --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
