Closed ggPeti closed 10 years ago
Well, the build failed, but that's because a spec for another issue is nondeterministic: it sets up a future with a 1 second sleep inside a thread, then waits 2 seconds for that thread to terminate. Looks like this time it wasn't enough. I added another 0.5 seconds of sleep to better our chances of the thread terminating in time.
Looks awesome, thanks!
Can you please release a new version with this fix? I'd love to get rid of the monkey patch I'm using for this right now :)
Of course!
Done :)
Because the
Future
class is a subclass ofDelegator
, it tries to delegate everything. Evenraise
! Thus, we need to explicitly callKernel.raise
, otherwise we get aSystemStackError
. The specs were incorrect as well, because they did not explicitly check for the error type and message, only expecting an::Exception
to be raised.