Next in thread → Next in month →

RE: [wsrm] Proposed definition for SubmitResponse Primitive

From
Jacques Durand <>
Date
2004-06-21T16:41:23+00:00
ID
Thread
RE: [wsrm] Proposed definition for SubmitResponse Primitive
Title: RE: [wsrm] Proposed definition for SubmitResponse Primitive

Tom: inline,

Jacques

-----Original Message-----

From: Tom Rutt [mailto:]

Sent: Thursday, June 17, 2004 2:28 PM

To: wsrm

Subject: [wsrm] Proposed definition for SubmitResponse Primitive

I propose we add a new primitive to the model:

Submit Response:

An abstract operation invoked by the consumer to convey response payload 

to the Reciving RMP (i.e.,  information intended for consumption by the 

producer of the original message request) to be carried in the

underlying protocol response to the reliable message request .

<JD> our mails crossed...

I suggested we keep the definition of these RMP operations "minimal",

i.e. only say what these do within the scope of a single RMP - the one to which they apply,

like we do for Submit, Deliver.

(so I would only state that the invocation of "Respond" must be related to a previous

invocation of "Deliver" on the same RMP)

We already made the relationship between RMP1.Submit and RMP2.Deliver a part of

an RM agreement (not of the operation definition), so it is only fitting that

 we also describe separately the relationship RMP2.Respond - RMP1.Notify .

which could be the object of an RM agreement too...

(and keeping it as an "intent" is too vague for a def)

Now, for the mention to the "underlying protocol response" - I proposed to use "SOAP MEP" instead 

of underlying protocol in a previous mail (though just a proposal). 

The association between RMP operations and the underlying SOAP MEP

would be a "binding" that we can describe separately as well.

</JD>

I tried to make this general, but I would like to clarify that submit 

response is not to be used for subsequent

one way callbacks conveying the response payload.

<JD>

We can convey this in the binding RMP -to- SOAP MEP.

That would be more appropriate, as we may as well be open later to other bindings

that would allow this... (e.g. if the Respond operation is allowed to bind

to a request of a separate request-response SOAP MEP, or to the message of

a separate one-way SOAP MEP).

</JD>

Tom Rutt

-- 

----------------------------------------------------

Tom Rutt        email: ; 

Tel: +1 732 801 5744          Fax: +1 732 774 5133

To unsubscribe from this mailing list (and be removed from the roster of the OASIS TC), go to http://www.oasis-open.org/apps/org/workgroup/wsrm/members/leave_workgroup.php.
Next in thread → Next in month →