Consider the following Kotlin code:
import kotlin.concurrent.thread
fun main() {
println("Press <Enter> to terminate.")
var interrupted = false
val worker = thread {
while (!interrupted) {
println("Working...")
Thread.sleep(1000L)
}
}
System.`in`.read()
println("Terminating...")
interrupted = true
worker.join()
println("Terminated.")
}
as well the same example rewritten using coroutines:
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.delay
import kotlinx.coroutines.launch
import kotlinx.coroutines.runBlocking
fun main() = runBlocking {
println("Press <Enter> to terminate.")
var interrupted = false
val worker = launch(Dispatchers.IO) {
while (!interrupted) {
println("Working...")
delay(1000L)
}
}
System.`in`.read()
println("Terminating...")
interrupted = true
worker.join()
println("Terminated.")
}
Both examples will work in most cases, and yet both are broken, because, at the bytecode level, a boolean variable accessed from more than a single thread is represented as a kotlin.jvm.internal.Ref.BooleanRef which is not thread-safe.
It's worth mentioning that a Java compiler will require interrupted to be final and the identical Java code will simply fail to compile.
Questions
What is the canonical way to rewrite the above code using just the standard library (i. e. w/o java.util.concurrent.atomic.AtomicBoolean or kotlinx.atomicfu.AtomicBoolean)?
How can the above code (the 2nd fragment which uses coroutines) be rewritten in the most portable way, so that it can target Kotlin/Multiplatform?
Based on Kotlin documentation
The first solution is a thread-safe data structure like AtmoicBoolean
import java.util.concurrent.atomic.AtomicBoolean
import kotlin.concurrent.thread
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.delay
import kotlinx.coroutines.launch
import kotlinx.coroutines.runBlocking
fun main() {
println("Press <Enter> to terminate.")
val interrupted = AtomicBoolean()
val worker = thread {
while (!interrupted.get()) {
println("Working...")
Thread.sleep(1000L)
}
}
System.`in`.read()
println("Terminating...")
interrupted.set(true)
worker.join()
println("Terminated.")
}
// coroutine way
fun main_2() = runBlocking {
println("Press <Enter> to terminate.")
val interrupted = AtomicBoolean()
val worker = launch(Dispatchers.IO) {
while (!interrupted.get()) {
println("Working...")
delay(1000L)
}
}
System.`in`.read()
println("Terminating...")
interrupted.set(true)
worker.join()
println("Terminated.")
}
Second solution is Mutual exclusion
import kotlinx.coroutines.Dispatchers
import kotlinx.coroutines.delay
import kotlinx.coroutines.launch
import kotlinx.coroutines.runBlocking
import kotlinx.coroutines.sync.Mutex
import kotlinx.coroutines.sync.withLock
val mutex = Mutex()
fun main() = runBlocking {
println("Press <Enter> to terminate.")
var interrupted = false
val worker = launch(Dispatchers.IO) {
while (mutex.withLock { !interrupted }) {
println("Working...")
delay(1000L)
}
}
System.`in`.read()
println("Terminating...")
mutex.withLock { interrupted = true }
worker.join()
println("Terminated.")
}
I am just using two solutions for this problem in here you can find another solution
How can the above code (the 2nd fragment, which uses coroutines) be rewritten in the most portable way so it can target Kotlin/Multiplatform?
I don't have much experience in kotlin-multiplatform, but you can't use `Dispacher.IO` in Kotlin multiplatform because it's bound to the JVM, so if you’re
using Kotlin/JavaScript or Kotlin/Native projects, you won’t be able to use it.
Related
First of all, I'm new in Kotlin, so please be nice :).
It's also my first time posting on StackOverflow
I want to literally STOP the current thread that I created but nothing works.
I tried quit(), quitSafely(), interrupt() but nothing works.
I created a class (Data.kt), in which I create and initialize a Handler and HandlerThread as follows :
class Dispatch(private val label: String = "main") {
var handler: Handler? = null
var handlerThread: HandlerThread? = null
init {
if (label == "main") {
handlerThread = null
handler = Handler(Looper.getMainLooper())
} else {
handlerThread = HandlerThread(label)
handlerThread!!.start()
handler = Handler(handlerThread!!.looper)
}
}
fun async(runnable: Runnable) = handler!!.post(runnable)
fun async(block: () -> (Unit)) = handler!!.post(block)
fun asyncAfter(milliseconds: Long, function: () -> (Unit)) {
handler!!.postDelayed(function, milliseconds)
}
fun asyncAfter(milliseconds: Long, runnable: Runnable) {
handler!!.postDelayed(runnable, milliseconds)
}
companion object {
val main = Dispatch()
private val global = Dispatch("global")
//fun global() = global
}
}
And now, in my DataManager, I use these to do asynchronous things :
fun getSomething(forceNetwork: Boolean ) {
val queue1 = Dispatch("thread1") // Create a thread called "thread1"
queue1.async {
for (i in 0..2_000_000) {
print("Hello World")
// Do everything i want in the current thread
}
// And on the main thread I call my callback
Dispatch.main.async {
//callback?.invoke(.........)
}
}
}
Now, in my MainActivity, I made 2 buttons :
One for running the function getSomething()
The other one is used for switching to another Controller View :
val button = findViewById<Button>(R.id.button)
button.setOnClickListener {
DataManager.getSomething(true)
}
val button2 = findViewById<Button>(R.id.button2)
button2.setOnClickListener {
val intent = Intent(this, Test::class.java) // Switch to my Test Controller
intent.setFlags(Intent.FLAG_ACTIVITY_NO_HISTORY)
startActivity(intent)
finish()
}
Is there a way to stop the thread, because when I switch to my second View, print("Hello World") is still triggered, unfortunately.
Thanks for helping me guys I hope that you understand !
A thread needs to periodically check a (global) flag and when it becomes true then the thread will break out from the loop. Java threads cannot be safely stopped without its consent.
Refer to page 252 here http://www.rjspm.com/PDF/JavaTheCompleteReference.pdf that describes the true story behind the legend.
I think that a truly interruptible thread is only possible through the support of the operating system kernel. The actual true lock is held deep down by the CPU hardware microprocessor.
I have one thread receiving packets from a local server:
// Shared object:
#Volatile lateinit var recentPacket: Packet
val recvMutex = Object()
// Thread code:
thread(isDaemon = true) {
while (true) {
val packet = readPacket()
synchronized(recvMutex) {
recentPacket = packet
recvMutex.notifyAll()
}
}
}
And I have multiple other threads waiting for packets, each of which should get the same packet that was just received:
suspend fun receive() {
return synchronized(recvMutex) {
recvMutex.wait() // non-blocking alternative to this?
recentPacket
}
}
It works, but Object.wait() blocks the thread. Is there a way to avoid this?
It works, but Object.wait() blocks the thread. Is there a way to avoid this?
Yes, but it means removing the complete wait-notify idiom you use now and replacing it with Kotlin's native BroadcastChannel.
Here is a basic example with two receivers and five packets being broadcast:
import kotlinx.coroutines.experimental.asCoroutineDispatcher
import kotlinx.coroutines.experimental.channels.BroadcastChannel
import kotlinx.coroutines.experimental.channels.SubscriptionReceiveChannel
import kotlinx.coroutines.experimental.delay
import kotlinx.coroutines.experimental.launch
import kotlinx.coroutines.experimental.runBlocking
import java.util.concurrent.ExecutorService
import java.util.concurrent.Executors
private val threadPool = Executors.newCachedThreadPool() as ExecutorService
val MyPool = threadPool.asCoroutineDispatcher()
fun main(args: Array<String>) {
val packetChannel = BroadcastChannel<Packet>(1)
(1..2).forEach {
launch(MyPool) {
receivePackets(it, packetChannel.openSubscription())
}
}
runBlocking {
(1..5).forEach {
packetChannel.send(Packet(it))
delay(100)
}
}
threadPool.shutdown()
}
suspend fun receivePackets(index: Int, packetChannel: SubscriptionReceiveChannel<Packet>) {
while (true) {
println("Receiver $index got packet ${packetChannel.receive().value}")
}
}
data class Packet(
val value: Int
)
Expect to see output such as this:
Receiver 1 got packet 1
Receiver 2 got packet 1
Receiver 2 got packet 2
Receiver 1 got packet 2
Receiver 1 got packet 3
Receiver 2 got packet 3
Receiver 1 got packet 4
Receiver 2 got packet 4
Receiver 1 got packet 5
Receiver 2 got packet 5
Coroutines seem to be in an experimental state; I suggest waiting until those mature before using them. See https://kotlinlang.org/docs/reference/coroutines.html#experimental-status-of-coroutines
I the mean time, you might try a ThreadPool:
import java.net.DatagramPacket
import java.net.DatagramSocket
import java.util.concurrent.Executors
import kotlin.concurrent.thread
fun start() {
val pool = Executors.newFixedThreadPool(10)
thread(isDaemon = true) {
val socket = DatagramSocket(12345)
while (!socket.isClosed) {
val packet = DatagramPacket(ByteArray(1000), 0)
socket.receive(packet)
pool.submit({
receive(packet)
})
}
}
pool.shutdown()
}
fun receive(packet: DatagramPacket) {
println(String(packet.data, 0, packet.length))
}
Asynchronous IO might be useful; You can look into Java Selectors
I've got a Sequence (from File.walkTopDown) and I need to run a long-running operation on each of them. I'd like to use Kotlin best practices / coroutines, but I either get no parallelism, or way too much parallelism and hit a "too many open files" IO error.
File("/Users/me/Pictures/").walkTopDown()
.onFail { file, ex -> println("ERROR: $file caused $ex") }
.filter { ... only big images... }
.map { file ->
async { // I *think* I want async and not "launch"...
ImageProcessor.fromFile(file)
}
}
This doesn't seem to run it in parallel, and my multi-core CPU never goes above 1 CPU's worth. Is there a way with coroutines to run "NumberOfCores parallel operations" worth of Deferred jobs?
I looked at Multithreading using Kotlin Coroutines which first creates ALL the jobs then joins them, but that means completing the Sequence/file tree walk completly bfore the heavy processing join step, and that seems... iffy! Splitting it into a collect and a process step means the collection could run way ahead of the processing.
val jobs = ... the Sequence above...
.toSet()
println("Found ${jobs.size}")
jobs.forEach { it.await() }
This isn't specific to your problem, but it does answer the question of, "how to cap kotlin coroutines maximum concurrency".
EDIT: As of kotlinx.coroutines 1.6.0 (https://github.com/Kotlin/kotlinx.coroutines/issues/2919), you can use limitedParallelism, e.g. Dispatchers.IO.limitedParallelism(123).
Old solution: I thought to use newFixedThreadPoolContext at first, but 1) it's deprecated and 2) it would use threads and I don't think that's necessary or desirable (same with Executors.newFixedThreadPool().asCoroutineDispatcher()). This solution might have flaws I'm not aware of by using Semaphore, but it's very simple:
import kotlinx.coroutines.async
import kotlinx.coroutines.awaitAll
import kotlinx.coroutines.coroutineScope
import kotlinx.coroutines.sync.Semaphore
import kotlinx.coroutines.sync.withPermit
/**
* Maps the inputs using [transform] at most [maxConcurrency] at a time until all Jobs are done.
*/
suspend fun <TInput, TOutput> Iterable<TInput>.mapConcurrently(
maxConcurrency: Int,
transform: suspend (TInput) -> TOutput,
) = coroutineScope {
val gate = Semaphore(maxConcurrency)
this#mapConcurrently.map {
async {
gate.withPermit {
transform(it)
}
}
}.awaitAll()
}
Tests (apologies, it uses Spek, hamcrest, and kotlin test):
import kotlinx.coroutines.ExperimentalCoroutinesApi
import kotlinx.coroutines.delay
import kotlinx.coroutines.launch
import kotlinx.coroutines.runBlocking
import kotlinx.coroutines.test.TestCoroutineDispatcher
import org.hamcrest.MatcherAssert.assertThat
import org.hamcrest.Matchers.greaterThanOrEqualTo
import org.hamcrest.Matchers.lessThanOrEqualTo
import org.spekframework.spek2.Spek
import org.spekframework.spek2.style.specification.describe
import java.util.concurrent.atomic.AtomicInteger
import kotlin.test.assertEquals
#OptIn(ExperimentalCoroutinesApi::class)
object AsyncHelpersKtTest : Spek({
val actionDelay: Long = 1_000 // arbitrary; obvious if non-test dispatcher is used on accident
val testDispatcher = TestCoroutineDispatcher()
afterEachTest {
// Clean up the TestCoroutineDispatcher to make sure no other work is running.
testDispatcher.cleanupTestCoroutines()
}
describe("mapConcurrently") {
it("should run all inputs concurrently if maxConcurrency >= size") {
val concurrentJobCounter = AtomicInteger(0)
val inputs = IntRange(1, 2).toList()
val maxConcurrency = inputs.size
// https://github.com/Kotlin/kotlinx.coroutines/issues/1266 has useful info & examples
runBlocking(testDispatcher) {
print("start runBlocking $coroutineContext\n")
// We have to run this async so that the code afterwards can advance the virtual clock
val job = launch {
testDispatcher.pauseDispatcher {
val result = inputs.mapConcurrently(maxConcurrency) {
print("action $it $coroutineContext\n")
// Sanity check that we never run more in parallel than max
assertThat(concurrentJobCounter.addAndGet(1), lessThanOrEqualTo(maxConcurrency))
// Allow for virtual clock adjustment
delay(actionDelay)
// Sanity check that we never run more in parallel than max
assertThat(concurrentJobCounter.getAndAdd(-1), lessThanOrEqualTo(maxConcurrency))
print("action $it after delay $coroutineContext\n")
it
}
// Order is not guaranteed, thus a Set
assertEquals(inputs.toSet(), result.toSet())
print("end mapConcurrently $coroutineContext\n")
}
}
print("before advanceTime $coroutineContext\n")
// Start the coroutines
testDispatcher.advanceTimeBy(0)
assertEquals(inputs.size, concurrentJobCounter.get(), "All jobs should have been started")
testDispatcher.advanceTimeBy(actionDelay)
print("after advanceTime $coroutineContext\n")
assertEquals(0, concurrentJobCounter.get(), "All jobs should have finished")
job.join()
}
}
it("should run one at a time if maxConcurrency = 1") {
val concurrentJobCounter = AtomicInteger(0)
val inputs = IntRange(1, 2).toList()
val maxConcurrency = 1
runBlocking(testDispatcher) {
val job = launch {
testDispatcher.pauseDispatcher {
inputs.mapConcurrently(maxConcurrency) {
assertThat(concurrentJobCounter.addAndGet(1), lessThanOrEqualTo(maxConcurrency))
delay(actionDelay)
assertThat(concurrentJobCounter.getAndAdd(-1), lessThanOrEqualTo(maxConcurrency))
it
}
}
}
testDispatcher.advanceTimeBy(0)
assertEquals(1, concurrentJobCounter.get(), "Only one job should have started")
val elapsedTime = testDispatcher.advanceUntilIdle()
print("elapsedTime=$elapsedTime")
assertThat(
"Virtual time should be at least as long as if all jobs ran sequentially",
elapsedTime,
greaterThanOrEqualTo(actionDelay * inputs.size)
)
job.join()
}
}
it("should handle cancellation") {
val jobCounter = AtomicInteger(0)
val inputs = IntRange(1, 2).toList()
val maxConcurrency = 1
runBlocking(testDispatcher) {
val job = launch {
testDispatcher.pauseDispatcher {
inputs.mapConcurrently(maxConcurrency) {
jobCounter.addAndGet(1)
delay(actionDelay)
it
}
}
}
testDispatcher.advanceTimeBy(0)
assertEquals(1, jobCounter.get(), "Only one job should have started")
job.cancel()
testDispatcher.advanceUntilIdle()
assertEquals(1, jobCounter.get(), "Only one job should have run")
job.join()
}
}
}
})
Per https://play.kotlinlang.org/hands-on/Introduction%20to%20Coroutines%20and%20Channels/09_Testing, you may also need to adjust compiler args for the tests to run:
compileTestKotlin {
kotlinOptions {
// Needed for runBlocking test coroutine dispatcher?
freeCompilerArgs += "-Xuse-experimental=kotlin.Experimental"
freeCompilerArgs += "-Xopt-in=kotlin.RequiresOptIn"
}
}
testImplementation 'org.jetbrains.kotlinx:kotlinx-coroutines-test:1.4.1'
The problem with your first snippet is that it doesn't run at all - remember, Sequence is lazy, and you have to use a terminal operation such as toSet() or forEach(). Additionally, you need to limit the number of threads that can be used for that task via constructing a newFixedThreadPoolContext context and using it in async:
val pictureContext = newFixedThreadPoolContext(nThreads = 10, name = "reading pictures in parallel")
File("/Users/me/Pictures/").walkTopDown()
.onFail { file, ex -> println("ERROR: $file caused $ex") }
.filter { ... only big images... }
.map { file ->
async(pictureContext) {
ImageProcessor.fromFile(file)
}
}
.toList()
.forEach { it.await() }
Edit:
You have to use a terminal operator (toList) befor awaiting the results
I got it working with a Channel. But maybe I'm being redundant with your way?
val pipe = ArrayChannel<Deferred<ImageFile>>(20)
launch {
while (!(pipe.isEmpty && pipe.isClosedForSend)) {
imageFiles.add(pipe.receive().await())
}
println("pipe closed")
}
File("/Users/me/").walkTopDown()
.onFail { file, ex -> println("ERROR: $file caused $ex") }
.forEach { pipe.send(async { ImageFile.fromFile(it) }) }
pipe.close()
This doesn't preserve the order of the projection but otherwise limits the throughput to at most maxDegreeOfParallelism. Expand and extend as you see fit.
suspend fun <TInput, TOutput> (Collection<TInput>).inParallel(
maxDegreeOfParallelism: Int,
action: suspend CoroutineScope.(input: TInput) -> TOutput
): Iterable<TOutput> = coroutineScope {
val list = this#inParallel
if (list.isEmpty())
return#coroutineScope listOf<TOutput>()
val brake = Channel<Unit>(maxDegreeOfParallelism)
val output = Channel<TOutput>()
val counter = AtomicInteger(0)
this.launch {
repeat(maxDegreeOfParallelism) {
brake.send(Unit)
}
for (input in list) {
val task = this.async {
action(input)
}
this.launch {
val result = task.await()
output.send(result)
val completed = counter.incrementAndGet()
if (completed == list.size) {
output.close()
} else brake.send(Unit)
}
brake.receive()
}
}
val results = mutableListOf<TOutput>()
for (item in output) {
results.add(item)
}
return#coroutineScope results
}
Example usage:
val output = listOf(1, 2, 3).inParallel(2) {
it + 1
} // Note that output may not be in same order as list.
Why not use the asFlow() operator and then use flatMapMerge?
someCoroutineScope.launch(Dispatchers.Default) {
File("/Users/me/Pictures/").walkTopDown()
.asFlow()
.filter { ... only big images... }
.flatMapMerge(concurrencyLimit) { file ->
flow {
emit(runInterruptable { ImageProcessor.fromFile(file) })
}
}.catch { ... }
.collect()
}
Then you can limit the simultaneous open files while still processing them concurrently.
To limit the parallelism to some value there is limitedParallelism function starting from the 1.6.0 version of the kotlinx.coroutines library. It can be called on CoroutineDispatcher object. So to limit threads for parallel execution we can write something like:
val parallelismLimit = Runtime.getRuntime().availableProcessors()
val limitedDispatcher = Dispatchers.Default.limitedParallelism(parallelismLimit)
val scope = CoroutineScope(limitedDispatcher) // we can set limitedDispatcher for the whole scope
scope.launch { // or we can set limitedDispatcher for a coroutine launch(limitedDispatcher)
File("/Users/me/Pictures/").walkTopDown()
.onFail { file, ex -> println("ERROR: $file caused $ex") }
.filter { ... only big images... }
.map { file ->
async {
ImageProcessor.fromFile(file)
}
}.toList().awaitAll()
}
ImageProcessor.fromFile(file) will be executed in parallel using parallelismLimit number of threads.
This will cap coroutines to workers. I'd recommend watching https://www.youtube.com/watch?v=3WGM-_MnPQA
package com.example.workers
import kotlinx.coroutines.*
import kotlinx.coroutines.channels.ReceiveChannel
import kotlinx.coroutines.channels.produce
import kotlin.system.measureTimeMillis
class ChannellibgradleApplication
fun main(args: Array<String>) {
var myList = mutableListOf<Int>(3000,1200,1400,3000,1200,1400,3000)
runBlocking {
var myChannel = produce(CoroutineName("MyInts")) {
myList.forEach { send(it) }
}
println("Starting coroutineScope ")
var time = measureTimeMillis {
coroutineScope {
var workers = 2
repeat(workers)
{
launch(CoroutineName("Sleep 1")) { theHardWork(myChannel) }
}
}
}
println("Ending coroutineScope $time ms")
}
}
suspend fun theHardWork(channel : ReceiveChannel<Int>)
{
for(m in channel) {
println("Starting Sleep $m")
delay(m.toLong())
println("Ending Sleep $m")
}
}
I call def activateReward by Akka actors and execution OracleClient.rewardActivate(user) sometimes is very slow (the database is outside of my responsibility and belongs to another company).
When database is slow the thread pool is exhausted and can not effectively allocate more threads to run callbacks future.onComplete because callbacks and futures works in the same execution context.
Please advise how to execute code in the callback asynchronously from threads which allocated for futures OracleClient.rewardActivate(user)
class RewardActivatorHelper {
private implicit val ec = new ExecutionContext {
val threadPool = Executors.newFixedThreadPool(1000)
def execute(runnable: Runnable) {threadPool.submit(runnable)}
def reportFailure(t: Throwable) {throw t}
}
case class FutureResult(spStart:Long, spFinish:Long)
def activateReward(msg:Msg, time:Long):Unit = {
msg.users.foreach {
user =>
val future:Future[FutureResult] = Future {
val (spStart, spFinish) = OracleClient.rewardActivate(user)
FutureResult(spStart, spFinish)
}
future.onComplete {
case Success(futureResult:FutureResult) =>
futureResult match {
case res:FutureResult => Logger.writeToLog(Logger.LogLevel.DEBUG,s"started:${res.spStart}finished:${res.spFinish}")
case _ => Logger.writeToLog(Logger.LogLevel.DEBUG, "some error")
}
case Failure(e:Throwable) => Logger.writeToLog(Logger.LogLevel.DEBUG, e.getMessage)
}
}
}
}
You can specify the execution context explicitly instead of implicitly for the onComplete callback by doing something along these lines:
import java.util.concurrent.Executors
import scala.concurrent.duration.Duration
object Example extends App {
import scala.concurrent._
private implicit val ec = new ExecutionContext {
val threadPool = Executors.newFixedThreadPool(1000)
def execute(runnable: Runnable) {threadPool.submit(runnable)}
def reportFailure(t: Throwable) {throw t}
}
val f = Future {
println("from future")
}
f.onComplete { _ =>
println("I'm done.")
}(scala.concurrent.ExecutionContext.Implicits.global)
Await.result(f, Duration.Inf)
}
This will of course not solve the underlying problem of a database not keeping up, but might be good to know anyway.
To clarify: I let the onComplete callback be handled by the standard global execution context. You might want to create a separate one.
For example
import scala.actors.Actor
import scala.actors.Actor._
object Main {
class Pong extends Actor {
def act() {
var pongCount = 0
while (true) {
receive {
case "Ping" =>
if (pongCount % 1000 == 0)
Console.println("Pong: ping "+pongCount)
sender ! "Pong"
pongCount = pongCount + 1
case "Stop" =>
Console.println("Pong: stop")
exit()
}
}
}
}
class Ping(count: Int, pong: Actor) extends Actor {
def act() {
var pingsLeft = count - 1
pong ! "Ping"
while (true) {
receive {
case "Pong" =>
if (pingsLeft % 1000 == 0)
Console.println("Ping: pong")
if (pingsLeft > 0) {
pong ! "Ping"
pingsLeft -= 1
} else {
Console.println("Ping: stop")
pong ! "Stop"
exit()
}
}
}
}
}
def main(args: Array[String]): Unit = {
val pong = new Pong
val ping = new Ping(100000, pong)
ping.start
pong.start
println("???")
}
}
I try to print "???" after the two actors call exit(), but now it is printed before "Ping: Stop" and "Pong stop"
I have try have a flag in the actor, flag is false while actor is running, and flag is true when actor stops, and in the main func, there is a while loop, such as while (actor.flag == false) {}, but it doesn't works, it is a endless loop:-/
So, please give me some advice.
If you need synchronous calls in akka, use ask pattern. Like
Await.result(ping ? "ping")
Also, you'd better use actor system to create actors.
import akka.actor.{ActorRef, Props, Actor, ActorSystem}
import akka.pattern.ask
import akka.util.Timeout
import scala.concurrent.Await
import scala.concurrent.duration._
import scala.concurrent.ExecutionContext.Implicits.global
object Test extends App {
implicit val timeout = Timeout(3 second)
val system = ActorSystem("ActorSystem")
class Pong extends Actor {
def receive: Receive = {
case "Ping" =>
println("ping")
context.stop(self)
}
}
lazy val pong = system.actorOf(Props(new Pong), "Pong")
val x = pong.ask("Ping")
val res = Await.result(x, timeout.duration)
println("????")
system.shutdown()
}