In Muse, we throw SoapFault (or BaseFault, in WSRF) if the error is pre-defined and meant to be communicated back to the client. We use runtime exceptions to report configuration and logic errors that result in the programmer needing to go back and fix the endpoint before any clients can communicate with it.
Dan "Vinh Nguyen \(vinguye2\)" <[EMAIL PROTECTED]> wrote on 03/23/2007 08:04:51 PM: > What should be the convention for returning errors to the remote client? > Should we throw Exceptions or SoapFaults? > > I've been working on trying to standardize error handling from our code. > For some of our capability operations, we can explicitly throw a > SoapFault. This is good because we can set a specific error code and > message, and a client can catch the faults and respond according to the > error code. > > But in many other areas, both outside of our capability code and within, > general Exceptions are thrown instead. So we cannot control how the > errors are sent to the client. I'd like to know how to make the error > handling better so that a client can get errors in a consistent manner > and respond in an automated fashion. > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
