ode-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lance Waterman" <lance.water...@gmail.com>
Subject Re: Re: Re: Asynchronous service invocation question
Date Thu, 01 Mar 2007 17:32:39 GMT

I think it might be helpful to have some bpel-tests around these MEPs (
perhaps there are some in another module that I am not aware of? ). What do
you think? I can open a JIRA for this if folks think its the right approach.


On 2/28/07, Alex Boisvert <boisvert@intalio.com> wrote:
> On 2/28/07, Jiang Liu <S3075163@student.rmit.edu.au> wrote:
> >
> > As you said: "As soon as you have a callback EPR, you should be able to
> > invoke it", if i am correct, i think "you" in your sentence means the
> > "external service", am i right? if so, :-), how does it be able to
> invoke
> > BPEL back since there is no implementation at external server side.
> The consumer (external service in this case) needs a service description
> (WSDL) for the callback.  It can be either static (known a priori) or
> obtained dynamically (from the EPR).  In the second case, the customer
> requires a webservice stack that supports the dynamic scenario.
> As WS-I suggests (?), WSDL only supports one-way and receive-reply
> patterns
> > (not sure about WSDL2), so technically, external service is not
> "required"
> > to send message back because BPEL's invocation is just a one-way
> pattern.
> If you mean a sequencing of one-way or request-response invocations then
> falls outside of the scope of WSDL.   BPEL itself can be used as an
> interface description language to describe such composition of simple
> message exchange patterns.  But you don't need BPEL per-se, all you need
> is
> an application that knows the composition and it able to participate in
> it.
> alex

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message