Data.Conduit.Async

• Introduction

Contains a combinator for concurrently joining a producer and a consumer, such that the producer may continue to produce (up to the queue size) as the consumer is concurrently consuming.

Documentation

buffer :: (MonadBaseControl IO m, MonadIO m) => Int -> Producer m a -> Consumer a m b -> m b

Concurrently join the producer and consumer, using a bounded queue of the given size. The producer will block when the queue is full, if it is producing faster than the consumers is taking from it. Likewise, if the consumer races ahead, it will block until more input is available.

Exceptions are properly managed and propagated between the two sides, so the net effect should be equivalent to not using buffer at all, save for the concurrent interleaving of effects.

($$&) :: (MonadIO m, MonadBaseControl IO m) => Producer m a -> Consumer a m b -> m b

An operator form of buffer. In general you should be able to replace any use of $$ with $$& and suddenly reap the benefit of concurrency, if your conduits were spending time waiting on each other.

gatherFrom

Arguments

:: (MonadIO m, MonadBaseControl IO m)

=> Int

Size of the queue to create

-> (TBQueue o -> m ())

Action that generates output values

-> Producer m o

Gather output values asynchronously from an action in the base monad and then yield them downstream. This provides a means of working around the restriction that ConduitM cannot be an instance of MonadBaseControl in order to, for example, yield values from within a Haskell callback function called from a C library.

drainTo

Arguments

:: (MonadIO m, MonadBaseControl IO m)

=> Int

Size of the queue to create

-> (TBQueue (Maybe i) -> m r)

Action to consume input values

-> Consumer i m r

Drain input values into an asynchronous action in the base monad via a bounded TBQueue. This is effectively the dual of gatherFrom.