Having a context object in the run(...) and get(...) methods

[pandoras-toolbox]

A suggestion about the method FailsafeExecutor#run(dev.failsafe.function.CheckedRunnable) and the similar get method.

I think it would be nice if a ExecutionEvent variable would be available like in the method RetryPolicyBuilder#onFailedAttempt(...)

There are some things which you cannot do with the available features of Failsafe if there is no such information in the execution block itself, I mean without workarounds or doing it in an awkward way.

What do you think?

[Tembrel]

Doesn't FailsafeExecutor::run(ContextualRunnable<Void>) give you that already?

Failsafe.with(...)
    .run(ctx -> {
        int count = ctx.getExecutionCount();
        ... do stuff with count, maybe throw checked exception ...
    });

As far as I know, the ExecutionEvent types just offer subsets of ExecutionContext information.

[pandoras-toolbox]

Oh, yes, you are right, I overlooked that method. Thank you!

[pandoras-toolbox]

But one question, I cannot obtain the duration for the next attempt? I wanted to log that it will wait now for that duration until retrying again.

[jhalterman]

You can configure an OnRetryScheduled event listener on the retry policy, and use event.getDelay(). See:

https://failsafe.dev/retry/#event-listeners
https://failsafe.dev/javadoc/core/dev/failsafe/RetryPolicyBuilder.html#onRetryScheduled-dev.failsafe.event.EventListener-

[pandoras-toolbox]

Thank you. I cannot do exactly what I wanted in an elegant way because the information is not available in the run(...) block, but nevermind.

[Tembrel]

There's no way, elegant or otherwise, to supply the delay before the next execution attempt within the current attempt, because you don't in general know whether the current attempt will be seen as a failure or, if a delay function is used, what delay will be computed. (The delay function can use the result of the failed attempt to determine the delay.)

[pandoras-toolbox]

Okay, I see.