Re: Input-Only Operation

Pae, this operation type (input-only) is for applications where
subsequent information is not necessary (when the client doesn't care
whether there has been a failure on the receiving side) or where
subsequent information is being sent by different means, for example in
a different operation. In scenarios where a request/response model is
necessary, it's natural that input/output operations should be used.

Hope I've clarified it,

                   Jacek Kopecky

                   Senior Architect, Systinet Corporation
                   http://www.systinet.com/





On Sat, 2003-01-18 at 15:20, Pae Choi wrote:
> Under the section, "2.3 PortType Description Component", in the
> [1] WSDL v1.2 it stated as follows:
> 
> "Input-Only Operations: Operations which have only an input message. The
> semantics are that when a message is sent to the service, the service
> consumes it but does not produce any output message. There MUST NOT be any
> fault messages indicated in this case."
> 
> Say a client sends a message with the "Input-Only Operation"
> mode to a service. And there is a failure during the process at
> the service site and the service is not returning the "fault"
> message(s) according to the statement in the [1], how can the
> client make decisions without having the fault message(s) from
> service for the follow-up actions.
> 
> Any comments and/or pointers will be apprecated.
> 
> 
> Pae
> 
> 
> 
> [1] http://www.w3.org/TR/wsdl12
>

Received on Tuesday, 21 January 2003 12:33:15 UTC