RE: Text related to fault handling discussion

Dear Gary,
 
But in your discussion with me about multiple responses to a request, your
solution, which I came to agree with was to allow different parts of the
same operation to be in different interactions (but within the same
workunit?  We need to state how far spread the operation can be) so we can
take advantage of the current implicit sequencing and the explicit choice
construct.  So does not this mean that CDL does have the same problem now?
 
Unconsciously, perhaps that is why I was preferring the solution of having a
slightly more complex syntax allowed within an interaction bracket so that
the grouping of the request, response(s) and fault message(s) was clear.
 
Best Regards     Tony
A M Fletcher
 
Cohesions  (TM)
 
Business transaction management software for application coordination
www.choreology.com <http://www.choreology.com/> 
 
Choreology Ltd., 68 Lombard Street, London EC3V 9LJ     UK
Tel: +44 (0) 1473 729537   Mobile: +44 (0) 7801 948219
tony.fletcher@choreology.com     (Home: amfletcher@iee.org)
-----Original Message-----
From: public-ws-chor-request@w3.org [mailto:public-ws-chor-request@w3.org]
On Behalf Of Gary Brown
Sent: 28 February 2005 17:41
To: public-ws-chor@w3.org
Subject: Text related to fault handling discussion


This text is from a recent WS-BPEL draft:
 
"Because WSDL does not require that fault names be unique within the
namespace where the service operation is defined, all faults sharing a
common name and defined in the same namespace are indistinguishable in
BPEL4WS. In WSDL 1.1 it is necessary to specify a portType name, an
operation name, and the fault name to uniquely identify a fault. This limits
the ability to use fault-handling mechanisms to deal with invocation faults.
This is an important shortcoming of the WSDL fault model that will be
removed in future versions of WSDL."
 
>From my understanding, BPEL is interested in being able to handle a fault
message independently from any specific operation, and that is why they have
raised this as an issue.
 
However, from a CDL perspective, fault messages are always handled within an
interaction that does have the relevant context (i.e. the port type and
operation name) - and therefore adding the faultName to a respond exchange
would "uniquely identify a fault". Unlike BPEL, CDL does not require faults
to be globally unique, as they are alway handled in the context of an
operation.
 
Regards
Gary