> Now coming to your question, not every subclass of the > "Exception" is treated as the "Application Exception" > but the one that the user has explicitly derived from > "Exception" (or from a subclass Application Exception > e.g. CreateException, FinderException). Because if one > inherits a "Custom Exception" class from "RuntimeException" > class (that's itself inherited from "Exception" class) or a > subclass of it e.g. EJBException, then that is treated as > a system exception.
Well, that's the behavior we thought we might see, but as I said, that is NOT the behavior that we do see. The app server is treating any exception which is an instance of the exception in the throws clause as an Application Exception. I don't see anything in chapter 12 which indicates that the app server is behaving incorrectly. The concept of System Exception is not clearly defined anywhere that I can see. The concept of Application Exception, however, is defined very clearly in the first sentence of chapter 12, though it doesn't discuss the implications of subclassing exceptions. It is possible, I think, to interpret the spec as saying that what I did (putting "throws Exception" on my method) was an illegal thing for a Bean Provider to do, for the spec says: An application exception must not be defined as a subclass of the java.lang.RuntimeException, or of the java.rmi.RemoteException. These are reserved for system exceptions. Does this mean that it is supposed to be illegal for a Bean Provider to use "throws Exception" on his bean method? thanks, bryan =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff EJB-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
