Asynchronous retries 'whenFailure' broken in 0.4? about failsafe HOT 4 CLOSED

Your listener should only get called if doAnything throws an exception. I just pushed a test to verify this behavior and things seem to be working as expected. Two things to note:

The current API in master is a bit different than from the last release. The listener now accepts a result and a failure since failure could indicate an exception being thrown or a bad result being returned. That said, the listener is called as expected in the test.

Let me know what you think.

from failsafe.

I assumed that a exceeded retry policy would trigger the failure event. I use your library on a udp protocol, since there is not always a result. My solution at the moment is that I check the result to be present. If there no result, it must be a timeout. A "retryExceeded"-Handler would be nice.

Thank you!

from failsafe.

@Haardt I know this is an old issue, but I wanted to dig it back up again to make sure I understand your use case. Basically, you'd like your failure handler to be called if retries are exceeded and some unexpected result occurred? The unexpected result bit is important because that's how Recurrent knows to keep trying. Maybe an example policy:

retryPolicy = new RetryPolicy()
         .withDelay(100, TimeUnit.MILLISECONDS)
         .withMaxDuration(2, TimeUnit.SECONDS)
         .withMaxRetries(3)
         .whenResult(null);

This would perform retries so long as the result is null and the whenFailure handler would be called if a non-null result is never returned. Does this work for you?

from failsafe.

Also - See the new event listeners API that allows you to listen specifically for onCompleted, onFailed, etc. events:

https://github.com/jhalterman/recurrent#event-listeners

from failsafe.