[jboss-dev-forums] [Design of Messaging on JBoss (Messaging/JBoss)] - Re: Justification for

[scott.stark@jboss.org]

"timfox" wrote : The philosophy I'm following is that JMSExceptions are for recoverable conditions, e.g. trying to send to a destination that doesn't exist, the database isn't up etc.
  | 
  | I'm treating RuntimeExceptions as programing errors - symptoms of some bug in the code - in which case we just want to throw it as is to the client.
  | 
  | 
This gets back to another question on whether implementation details should leak to clients. Given that there is no agreement on checked/unchecked exception usage, there is nothing semantically meaningful about a RuntimeException in terms of whether or not its a recoverable condition. It's just as likely to be a normal failure mode as an Exception depending on the frameworks in use on the server.

View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3961373#3961373

Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3961373