Documentation

Run a BasicGuest, and return when the Event returned by the BasicGuest fires.

Each call runs on a separate spider timeline, so you can launch multiple hosts via forkIO or forkOS and they will not mutex each other.

NOTE: If you want to capture values from a build before the network starts firing (e.g., to hand off event triggers to another thread), populate an MVar (if threading) or IORef as you build the network. If you receive errors about untouchable type variables while doing this, add type annotations to constrain the MVar or IORef contents before passing them to the function that returns your BasicGuest. See the Multithread.hs example for a demonstration of this pattern, and where to put the type annotations.

Run a BasicGuest without a quit Event.

basicHostForever guest = basicHostWithQuit $ never <$ guest

Augment a BasicGuest with an action that is repeatedly run until the provided Event fires. Each time the action completes, the returned Event will fire.

Example - providing a 'tick' Event to a network

myNetwork
  :: (Reflex t, MonadHold t m, MonadFix m)
  => Event t ()
  -> m (Dynamic t Int)
myNetwork = count

myGuest :: BasicGuestConstraints t m => BasicGuest t m (Event t ())
myGuest = mdo
  eTick <- repeatUntilQuit (void $ threadDelay 1000000) eQuit
  let
    eCountUpdated = updated dCount
    eQuit = () <$ ffilter (==5) eCountUpdated
  dCount <- myNetwork eTick

  performEvent_ $ liftIO . print <$> eCountUpdated
  pure eQuit

main :: IO ()
main = basicHostWithQuit myGuest

Like repeatUntilQuit, but it doesn't do anything with the result of the action. May be a little more efficient if you don't need it.