On Mon, 2007-02-12 at 13:32 +1300, Dennis Sosnoski wrote: > I'd personally support changing the message receiver interface to add a > ping() method, with the AbstractMessageReceiver class implementing it as > discussed. The reason is that I see this as something that should be a > mandatory part of the framework.
If we add the method then I'd prefer to have a module that supports the features we've discussed but not always engaged. The reason is that this is a non-standard feature and so its something we should encourage people to engage (globally) but not require. Are you ok with that? > But with the -1 from Nicholas Gallardo > on this approach, the alternative is to define a new interface for the > message receiver to implement if it's going to support the ping. The > engine would then need to check if the message receiver implements this > interface (which would be the case for AbstractMessageReceiver and > subclasses) before casting and making the call. If the message receiver > does not implement the interface, I'd say the engine should always > return a failure response to the ping. > > I suppose this message receiver interface could be the same as the > service class Pingable interface mentioned in the earlier email. That > way we avoid yet another added interface. To me this is an ugly hack to get around the right solution. Nick, you've been quiet since you put the -1. Please explain your position now given all the discussion we've had on this. Thanks, Sanjiva. -- Sanjiva Weerawarana, Ph.D. Founder & Director; Lanka Software Foundation; http://www.opensource.lk/ Founder, Chairman & CEO; WSO2, Inc.; http://www.wso2.com/ Director; Open Source Initiative; http://www.opensource.org/ Member; Apache Software Foundation; http://www.apache.org/ Visiting Lecturer; University of Moratuwa; http://www.cse.mrt.ac.lk/ --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
