Creates a simple, noncancelable F[A] instance that executes an asynchronous process on evaluation.

The given function is being injected with a side-effectful callback for signaling the final result of an asynchronous process.

Creates a cancelable F[A] instance that executes an asynchronous process on evaluation.

This builder accepts a registration function that is being injected with a side-effectful callback, to be called when the asynchronous process is complete with a final result.

The registration function is also supposed to return an IO[Unit] that captures the logic necessary for cancelling the asynchronous process, for as long as it is still active.

Example:

import java.util.concurrent.ScheduledExecutorService
import scala.concurrent.duration._

def sleep[F[_]](d: FiniteDuration)
  (implicit F: Concurrent[F], ec: ScheduledExecutorService): F[A] = {

  F.cancelable { cb =>
    // Note the callback is pure, so we need to trigger evaluation
    val run = new Runnable { def run() = cb(Right(())).unsafeRunSync }

    // Schedules task to run after delay
    val future = ec.schedule(run, d.length, d.unit)

    // Cancellation logic, suspended in IO
    IO(future.cancel())
  }
}

Returns a new F value that mirrors the source for normal termination, but that triggers the given error on cancelation.

This onCancelRaiseError operator transforms any task into one that on cancelation will terminate with the given error, thus transforming potentially non-terminating tasks into ones that yield a certain error.

import scala.concurrent.CancellationException

val F = Concurrent[IO]
val timer = Timer[IO]

val error = new CancellationException("Boo!")
val fa = F.onCancelRaiseError(timer.sleep(5.seconds, error))

fa.start.flatMap { fiber =>
  fiber.cancel *> fiber.join
}

Without "onCancelRaiseError" the sleep operation yields a non-terminating task on cancellation. But by applying "onCancelRaiseError", the yielded task above will terminate with the indicated "CancellationException" reference, which we can then also distinguish from other errors thrown in the F context.

Depending on the implementation, tasks that are canceled can become non-terminating. This operation ensures that when cancelation happens, the resulting task is terminated with an error, such that logic can be scheduled to happen after cancelation:

import scala.concurrent.CancellationException
val wasCanceled = new CancellationException()

F.onCancelRaiseError(fa, wasCanceled).attempt.flatMap {
  case Right(a) =>
    F.delay(println(s"Success: $a"))
  case Left(`wasCanceled`) =>
    F.delay(println("Was canceled!"))
  case Left(error) =>
    F.delay(println(s"Terminated in error: $error"))
}

This technique is how a "bracket" operation can be implemented.

Besides sending the cancelation signal, this operation also cuts the connection between the producer and the consumer. Example:

val F = Concurrent[IO]
val timer = Timer[IO]

// This task is uninterruptible ;-)
val tick = F.uncancelable(
  for {
    _ <- timer.sleep(5.seconds)
    _ <- IO(println("Tick!"))
  } yield ())

// Builds an value that triggers an exception on cancellation
val loud = F.onCancelRaiseError(tick, new CancellationException)

In this example the loud reference will be completed with a "CancellationException", as indicated via "onCancelRaiseError". The logic of the source won't get cancelled, because we've embedded it all in uncancelable. But its bind continuation is not allowed to continue after that, its final result not being allowed to be signaled.

Therefore this also transforms uncancelable values into ones that can be canceled. The logic of the source, its run-loop might not be interruptible, however cancel on a value on which onCancelRaiseError was applied will cut the connection from the producer, the consumer receiving the indicated error instead.

Run two tasks concurrently, creating a race between them and returns a pair containing both the winner's successful value and the loser represented as a still-unfinished fiber.

If the first task completes in error, then the result will complete in error, the other task being cancelled.

On usage the user has the option of cancelling the losing task, this being equivalent with plain race:

val ioA: IO[A] = ???
val ioB: IO[B] = ???

Concurrent[IO].racePair(ioA, ioB).flatMap {
  case Left((a, fiberB)) =>
    fiberB.cancel.map(_ => a)
  case Right((fiberA, b)) =>
    fiberA.cancel.map(_ => b)
}

See race for a simpler version that cancels the loser immediately.

Start concurrent execution of the source suspended in the F context.

Returns a Fiber that can be used to either join or cancel the running computation, being similar in spirit (but not in implementation) to starting a thread.

Suspends the evaluation of an F reference.

Equivalent to FlatMap.flatten for pure expressions, the purpose of this function is to suspend side effects in F.

Returns a new F that mirrors the source, but that is uninterruptible.

This means that the cancel signal has no effect on the result of join and that the cancelable token returned by ConcurrentEffect.runCancelable on evaluation will have no effect.

This operation is undoing the cancelation mechanism of cancelable, with this equivalence:

F.uncancelable(F.cancelable { cb => f(cb); io }) <-> F.async(f)

Sample:

val F = Concurrent[IO]
val timer = Timer[IO]

// Normally Timer#sleep yields cancelable tasks
val tick = F.uncancelable(timer.sleep(10.seconds))

// This prints "Tick!" after 10 seconds, even if we are
// cancelling the Fiber after start:
for {
  fiber <- F.start(tick)
  _ <- fiber.cancel
  _ <- fiber.join
} yield {
  println("Tick!")
}

Cancelable effects are great in race conditions, however sometimes this operation is necessary to ensure that the bind continuation of a task (the following flatMap operations) are also evaluated no matter what.

Lifts any by-name parameter into the F context.

Equivalent to Applicative.pure for pure expressions, the purpose of this function is to suspend side effects in F.

Inherited from LiftIO, defines a conversion from IO in terms of the Concurrent type class.

N.B. expressing this conversion in terms of Concurrent and its capabilities means that the resulting F is cancelable in case the source IO is.

To access this implementation as a standalone function, you can use Concurrent.liftIO (on the object companion).

Run two tasks concurrently and return the first to finish, either in success or error. The loser of the race is cancelled.

The two tasks are potentially executed in parallel, the winner being the first that signals a result.

As an example, this would be the implementation of a "timeout" operation:

import cats.effect._
import scala.concurrent.duration._

def timeoutTo[F[_], A](fa: F[A], after: FiniteDuration, fallback: F[A])
  (implicit F: Concurrent[F], timer: Timer[F]): F[A] = {

   F.race(fa, timer.sleep(timer)).flatMap {
     case Left((a, _)) => F.pure(a)
     case Right((_, _)) => fallback
   }
}

def timeout[F[_], A](fa: F[A], after: FiniteDuration)
  (implicit F: Concurrent[F], timer: Timer[F]): F[A] = {

   timeoutTo(fa, after,
     F.raiseError(new TimeoutException(after.toString)))
}

Also see racePair for a version that does not cancel the loser automatically on successful results.