Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Fetching contributors…

Cannot retrieve contributors at this time

410 lines (369 sloc) 12.15 kb
/* __ *\
** ________ ___ / / ___ Scala API **
** / __/ __// _ | / / / _ | (c) 2005-2013, LAMP/EPFL **
** __\ \/ /__/ __ |/ /__/ __ | http://scala-lang.org/ **
** /____/\___/_/ |_/____/_/ | | **
** |/ **
\* */
package scala.actors
import scala.util.control.ControlThrowable
import java.util.{Timer, TimerTask}
import scala.language.implicitConversions
/**
* Provides functions for the definition of actors, as well as actor
* operations, such as `receive`, `react`, `reply`, etc.
*
* @author Philipp Haller
*/
object Actor extends Combinators {
/** State of an actor.
*
* - '''New''' -
* Not yet started
* - '''Runnable''' -
* Executing
* - '''Suspended''' -
* Suspended, waiting in a `react`
* - '''TimedSuspended''' -
* Suspended, waiting in a `reactWithin`
* - '''Blocked''' -
* Blocked waiting in a `receive`
* - '''TimedBlocked''' -
* Blocked waiting in a `receiveWithin`
* - '''Terminated''' -
* Actor has terminated
*/
object State extends Enumeration {
val New,
Runnable,
Suspended,
TimedSuspended,
Blocked,
TimedBlocked,
Terminated = Value
}
private[actors] val tl = new ThreadLocal[InternalReplyReactor]
// timer thread runs as daemon
private[actors] val timer = new Timer(true)
private[actors] val suspendException = new SuspendActorControl
/**
* Returns the currently executing actor. Should be used instead
* of `'''this'''` in all blocks of code executed by actors.
*
* @return returns the currently executing actor.
*/
def self: Actor = self(Scheduler).asInstanceOf[Actor]
private[actors] def self(sched: IScheduler): InternalActor =
rawSelf(sched).asInstanceOf[InternalActor]
private[actors] def rawSelf: InternalReplyReactor =
rawSelf(Scheduler)
private[actors] def rawSelf(sched: IScheduler): InternalReplyReactor = {
val s = tl.get
if (s eq null) {
val r = new ActorProxy(Thread.currentThread, sched)
tl.set(r)
r
} else
s
}
private def parentScheduler: IScheduler = {
val s = tl.get
if (s eq null) Scheduler else s.scheduler
}
/**
* Resets an actor proxy associated with the current thread.
* It replaces the implicit `ActorProxy` instance
* of the current thread (if any) with a new instance.
*
* This permits to re-use the current thread as an actor
* even if its `ActorProxy` has died for some reason.
*/
def resetProxy() {
val a = tl.get
if ((null ne a) && a.isInstanceOf[ActorProxy])
tl.set(new ActorProxy(Thread.currentThread, parentScheduler))
}
/**
* Removes any reference to an `Actor` instance
* currently stored in thread-local storage.
*
* This allows to release references from threads that are potentially
* long-running or being re-used (e.g. inside a thread pool). Permanent
* references in thread-local storage are a potential memory leak.
*/
def clearSelf() {
tl set null
}
/**
* Factory method for creating and starting an actor.
*
* @example {{{
* import scala.actors.Actor._
* ...
* val a = actor {
* ...
* }
* }}}
*
* @param body the code block to be executed by the newly created actor
* @return the newly created actor. Note that it is automatically started.
*/
def actor(body: => Unit): Actor = {
val a = new Actor {
def act() = body
override final val scheduler: IScheduler = parentScheduler
}
a.start()
a
}
/**
* Factory method for creating actors whose
* body is defined using a `Responder`.
*
* @example {{{
* import scala.actors.Actor._
* import Responder.exec
* ...
* val a = reactor {
* for {
* res <- b !! MyRequest;
* if exec(println("result: "+res))
* } yield {}
* }
* }}}
*
* @param body the `Responder` to be executed by the newly created actor
* @return the newly created actor. Note that it is automatically started.
*/
def reactor(body: => Responder[Unit]): Actor = {
val a = new Actor {
def act() {
Responder.run(body)
}
override final val scheduler: IScheduler = parentScheduler
}
a.start()
a
}
/**
* Receives the next message from the mailbox of the current actor `self`.
*/
def ? : Any = self.?
/**
* Receives a message from the mailbox of `self`. Blocks if no message
* matching any of the cases of `f` can be received.
*
* @example {{{
* receive {
* case "exit" => println("exiting")
* case 42 => println("got the answer")
* case x:Int => println("got an answer")
* }
* }}}
*
* @param f a partial function specifying patterns and actions
* @return the result of processing the received message
*/
def receive[A](f: PartialFunction[Any, A]): A =
self.receive(f)
/**
* Receives a message from the mailbox of `self`. Blocks at most `msec`
* milliseconds if no message matching any of the cases of `f` can be
* received. If no message could be received the `TIMEOUT` action is
* executed if specified.
*
* @param msec the time span before timeout
* @param f a partial function specifying patterns and actions
* @return the result of processing the received message
*/
def receiveWithin[R](msec: Long)(f: PartialFunction[Any, R]): R =
self.receiveWithin(msec)(f)
/**
* Lightweight variant of `receive`.
*
* Actions in `f` have to contain the rest of the computation of `self`,
* as this method will never return.
*
* A common method of continuting the computation is to send a message
* to another actor:
* {{{
* react {
* case Get(from) =>
* react {
* case Put(x) => from ! x
* }
* }
* }}}
*
* Another common method is to use `loop` to continuously `react` to messages:
* {{{
* loop {
* react {
* case Msg(data) => // process data
* }
* }
* }}}
*
* @param f a partial function specifying patterns and actions
* @return this function never returns
*/
def react(f: PartialFunction[Any, Unit]): Nothing =
rawSelf.react(f)
/**
* Lightweight variant of `receiveWithin`.
*
* Actions in `f` have to contain the rest of the computation of `self`,
* as this method will never return.
*
* @param msec the time span before timeout
* @param f a partial function specifying patterns and actions
* @return this function never returns
*/
def reactWithin(msec: Long)(f: PartialFunction[Any, Unit]): Nothing =
self.reactWithin(msec)(f)
def eventloop(f: PartialFunction[Any, Unit]): Nothing =
rawSelf.react(new RecursiveProxyHandler(rawSelf, f))
private class RecursiveProxyHandler(a: InternalReplyReactor, f: PartialFunction[Any, Unit])
extends PartialFunction[Any, Unit] {
def isDefinedAt(m: Any): Boolean =
true // events are immediately removed from the mailbox
def apply(m: Any) {
if (f.isDefinedAt(m)) f(m)
a.react(this)
}
}
/**
* Returns the actor which sent the last received message.
*/
def sender: OutputChannel[Any] =
rawSelf.internalSender
/**
* Sends `msg` to the actor waiting in a call to `!?`.
*/
def reply(msg: Any): Unit =
rawSelf.reply(msg)
/**
* Sends `()` to the actor waiting in a call to `!?`.
*/
def reply(): Unit =
rawSelf.reply(())
/**
* Returns the number of messages in `self`'s mailbox
*
* @return the number of messages in `self`'s mailbox
*/
def mailboxSize: Int = rawSelf.mailboxSize
/**
* Converts a synchronous event-based operation into
* an asynchronous `Responder`.
*
* @example {{{
* val adder = reactor {
* for {
* _ <- respondOn(react) { case Add(a, b) => reply(a+b) }
* } yield {}
* }
* }}}
*/
def respondOn[A, B](fun: PartialFunction[A, Unit] => Nothing):
PartialFunction[A, B] => Responder[B] =
(caseBlock: PartialFunction[A, B]) => new Responder[B] {
def respond(k: B => Unit) = fun(caseBlock andThen k)
}
private[actors] trait Body[a] {
def andThen[b](other: => b): Unit
}
implicit def mkBody[a](body: => a) = new InternalActor.Body[a] {
def andThen[b](other: => b): Unit = rawSelf.seq(body, other)
}
/**
* Links `self` to actor `to`.
*
* @param to the actor to link to
* @return the parameter actor
*/
def link(to: AbstractActor): AbstractActor = self.link(to)
/**
* Links `self` to the actor defined by `body`.
*
* @param body the body of the actor to link to
* @return the parameter actor
*/
def link(body: => Unit): Actor = self.link(body)
/**
* Unlinks `self` from actor `from`.
*
* @param from the actor to unlink from
*/
def unlink(from: AbstractActor): Unit = self.unlink(from)
/**
* Terminates execution of `self` with the following effect on
* linked actors:
*
* For each linked actor `a` with `trapExit` set to `'''true'''`,
* send message `Exit(self, reason)` to `a`.
*
* For each linked actor `a` with `trapExit` set to `'''false'''`
* (default), call `a.exit(reason)` if `reason != 'normal`.
*/
def exit(reason: AnyRef): Nothing = self.exit(reason)
/**
* Terminates execution of `self` with the following effect on
* linked actors:
*
* For each linked actor `a` with `trapExit` set to `'''true'''`,
* send message `Exit(self, 'normal)` to `a`.
*/
def exit(): Nothing = rawSelf.exit()
}
/** Provides lightweight, concurrent actors. Actors are created by extending
* the `Actor` trait (alternatively, one of the factory methods in its
* companion object can be used). The behavior of an `Actor` subclass is
* defined by implementing its `act` method:
* {{{
* class MyActor extends Actor {
* def act() {
* // actor behavior goes here
* }
* }
* }}}
* A new `Actor` instance is started by invoking its `start` method.
*
* '''Note:''' care must be taken when invoking thread-blocking methods other
* than those provided by the `Actor` trait or its companion object (such as
* `receive`). Blocking the underlying thread inside an actor may lead to
* starvation of other actors. This also applies to actors hogging their
* thread for a long time between invoking `receive`/`react`.
*
* If actors use blocking operations (for example, methods for blocking I/O),
* there are several options:
*
* - The run-time system can be configured to use a larger thread pool size
* (for example, by setting the `actors.corePoolSize` JVM property).
* - The `scheduler` method of the `Actor` trait can be overridden to return a
* `ResizableThreadPoolScheduler`, which resizes its thread pool to
* avoid starvation caused by actors that invoke arbitrary blocking methods.
* - The `actors.enableForkJoin` JVM property can be set to `false`, in which
* case a `ResizableThreadPoolScheduler` is used by default to execute actors.
*
* The main ideas of the implementation are explained in the two papers
*
* - [[http://lampwww.epfl.ch/~odersky/papers/jmlc06.pdf Event-Based
* Programming without Inversion of Control]],
* Philipp Haller and Martin Odersky, ''Proc. JMLC 2006'', and
* - [[http://lamp.epfl.ch/~phaller/doc/haller07coord.pdf Actors that
* Unify Threads and Events]],
* Philipp Haller and Martin Odersky, ''Proc. COORDINATION 2007''.
*
* @author Philipp Haller
*
* @define actor actor
* @define channel actor's mailbox
*/
@SerialVersionUID(-781154067877019505L)
trait Actor extends InternalActor with ReplyReactor {
override def start(): Actor = synchronized {
super.start()
this
}
}
Jump to Line
Something went wrong with that request. Please try again.