Użycie web workers

Dedykowane wątki robocze (Web Workers) zapewniają prosty sposób na uruchamiania skryptów w postaci wątków w tle treści internetowych. Po utworzeniu, wątek roboczy może przekazywać informacje do menedżera zadań, poprzez wysyłanie wiadomości do procesu obsługi zdarzeń, określonego przez twórcę. Jednak działają one w kontekście globalnym, który różni się od kontekstu bieżącego okna (wywoływanego za pomocą skrótu window, zamiast self, dlatego próba otrzymania obecnego zakresu globalnego komendą Worker zwróci błąd).

Wątki robocze  mogą wykonywać zadania bez interakcji z interfejsem użytkownika. Ponadto, mogą one obsługiwać wejście/wyjście używając XMLHttpRequest (chociaż responseXML i właściwości channel będą zawsze równe null).

Zobacz wątek w dokumentacji referencyjnej o wątkach roboczych; ten artykuł jest jej uzupełnieniem, które oferuje przykłady i dodaje szczegóły. Aby uzyskać listę funkcji dostępnych dla wątków, patrz Funkcje i interfejsy dostępne dla wątków.

O bezpieczeństwie wątków

Worker  jest interfejsem łączącym współbieżne wątki na na poziomie systemu operacyjnego i może spowodować ciekawe zmiany w kodzie, jeśli nie jesteś ostrożny. Jednak w przypadku wątków internetowych, punkty komunikacyjne z innych wątków są starannie kontrolowane, co oznacza, że ​w rzeczywistości bardzo trudno jest spowodować problemy współbieżności. Nie ma dostępu do bezwątkowych bezpiecznych komponentów lub obiektów DOM i trzeba by dopasować szczegółowe dane wejściowe i wyjściowe z wątku przez serializację obiektów. Więc trzeba się naprawdę mocno postarać, aby spowodować problemy w kodzie.

Mnożenie wątków

Tworzenie nowego wątku jest proste. Wszystko, co musisz zrobić, to wywołać konstruktor Worker(), podając URI skryptu do wykonania w wątku roboczym, a jeśli chcesz być w stanie otrzymywać powiadomienia od wątku, ustawić właściwości wątku Worker.onmessage na obsługę odpowiedniej funkcji zdarzenia.

var myWorker = new Worker("my_task.js");

myWorker.onmessage = function (oEvent) {
  console.log("Wywołane zwrotne, przez wątek!\n");
};

Alternatywnie, możesz użyć addEventListener() :

var myWorker = new Worker("my_task.js");

myWorker.addEventListener("message", function (oEvent) {
  console.log("Wywołane zwrotne, przez wątek!\n");
}, false);

myWorker.postMessage(""); // start wątku.

W powyższym przykładzie w linii 1 tworzymy nowy wątek roboczy. W linii 3 dodajemy do niego funkcję obsługi zdarzenia "message" pochodzącego z wątku. Funkcja ta będzie wywołana wtedy, kiedy wątek roboczy wywoła metodę Worker.postMessage(). W końcu w lini 7 uruchamiamy wątek.

Przekazywanie danych

Data passed between the main page and workers are copied, not shared. Objects are serialized as they're handed to the worker, and subsequently, de-serialized on the other end. The page and worker do not share the same instance, so the end result is that a duplicate is created on each end. Most browsers implement this feature as structured cloning.

Before continuing, let's create for didactical purpose a function named emulateMessage() which will simulate the behavior of a value which is cloned and not shared during the passage from a worker to the main page or vice versa:

function emulateMessage (vVal) {
    return eval("(" + JSON.stringify(vVal) + ")");
}

// Tests

// test #1
var example1 = new Number(3);
alert(typeof example1); // object
alert(typeof emulateMessage(example1)); // number

// test #2
var example2 = true;
alert(typeof example2); // boolean
alert(typeof emulateMessage(example2)); // boolean

// test #3
var example3 = new String("Hello World");
alert(typeof example3); // object
alert(typeof emulateMessage(example3)); // string

// test #4
var example4 = {
    "name": "John Smith",
    "age": 43
};
alert(typeof example4); // object
alert(typeof emulateMessage(example4)); // object

// test #5
function Animal (sType, nAge) {
    this.type = sType;
    this.age = nAge;
}
var example5 = new Animal("Cat", 3);
alert(example5.constructor); // Animal
alert(emulateMessage(example5).constructor); // Object

A value which is cloned and not shared is called message. Back to talk about workers, messages can be sent to and from the main thread by using postMessage(). The message event's data attribute contains data passed back from the worker.

example.html: (the main page):

var myWorker = new Worker("my_task.js");

myWorker.onmessage = function (oEvent) {
  console.log("Worker said : " + oEvent.data);
};

myWorker.postMessage("ali");

my_task.js (the worker):

postMessage("I\'m working before postMessage(\'ali\').");

onmessage = function (oEvent) {
  postMessage("Hi " + oEvent.data);
};

The structured cloning algorithm can accept JSON and a few things that JSON can't like circular references.

Przykłady przekazywania danych

Example #1: Create a generic "asynchronous eval()"

The following example shows how to use a worker in order to execute asynchronously any JavaScript code allowed in a Worker, through eval() within the worker:

// Syntax: asyncEval(code[, listener])

var asyncEval = (function () {

  var aListeners = [], oParser = new Worker("data:text/javascript;charset=US-ASCII,onmessage%20%3D%20function%20%28oEvent%29%20%7B%0A%09postMessage%28%7B%0A%09%09%22id%22%3A%20oEvent.data.id%2C%0A%09%09%22evaluated%22%3A%20eval%28oEvent.data.code%29%0A%09%7D%29%3B%0A%7D");

  oParser.onmessage = function (oEvent) {
    if (aListeners[oEvent.data.id]) { aListeners[oEvent.data.id](oEvent.data.evaluated); }
    delete aListeners[oEvent.data.id];
  };


  return function (sCode, fListener) {
    aListeners.push(fListener || null);
    oParser.postMessage({
      "id": aListeners.length - 1,
      "code": sCode
    });
  };

})();

(Note the data url is equivalent to a network request with the following response:)

onmessage = function (oEvent) {
	postMessage({
		"id": oEvent.data.id,
		"evaluated": eval(oEvent.data.code)
	});
}

Sample usage:

// asynchronous alert message...
asyncEval("3 + 2", function (sMessage) {
    alert("3 + 2 = " + sMessage);
});

// asynchronous print message...
asyncEval("\"Hello World!!!\"", function (sHTML) {
    document.body.appendChild(document.createTextNode(sHTML));
});

// asynchronous void...
asyncEval("(function () {\n\tvar oReq = new XMLHttpRequest();\n\toReq.open(\"get\", \"https://www.mozilla.org/\", false);\n\toReq.send(null);\n\treturn oReq.responseText;\n})()");

Example #2: Advanced passing JSON Data and creating a switching system

If you have to pass some complex data and have to call many different functions both in the main page and in the Worker, you can create a system like the following.

example.html (the main page):

<!doctype html>
<html>
<head>
<meta charset="UTF-8"  />
<title>MDN Example - Queryable worker</title>
<script type="text/javascript">
  /*
    QueryableWorker instances methods:
     * sendQuery(queryable function name, argument to pass 1, argument to pass 2, etc. etc): calls a Worker's queryable function
     * postMessage(string or JSON Data): see Worker.prototype.postMessage()
     * terminate(): terminates the Worker
     * addListener(name, function): adds a listener
     * removeListener(name): removes a listener
    QueryableWorker instances properties:
     * defaultListener: the default listener executed only when the Worker calls the postMessage() function directly
  */
  function QueryableWorker (sURL, fDefListener, fOnError) {
    var oInstance = this, oWorker = new Worker(sURL), oListeners = {};
    this.defaultListener = fDefListener || function () {};
    oWorker.onmessage = function (oEvent) {
      if (oEvent.data instanceof Object && oEvent.data.hasOwnProperty("vo42t30") && oEvent.data.hasOwnProperty("rnb93qh")) {
        oListeners[oEvent.data.vo42t30].apply(oInstance, oEvent.data.rnb93qh);
      } else {
        this.defaultListener.call(oInstance, oEvent.data);
      }
    };
    if (fOnError) { oWorker.onerror = fOnError; }
    this.sendQuery = function (/* queryable function name, argument to pass 1, argument to pass 2, etc. etc */) {
      if (arguments.length < 1) { throw new TypeError("QueryableWorker.sendQuery - not enough arguments"); return; }
      oWorker.postMessage({ "bk4e1h0": arguments[0], "ktp3fm1": Array.prototype.slice.call(arguments, 1) });
    };
    this.postMessage = function (vMsg) {
      //I just think there is no need to use call() method
      //how about just oWorker.postMessage(vMsg);
      //the same situation with terminate
      //well,just a little faster,no search up the prototye chain
      Worker.prototype.postMessage.call(oWorker, vMsg);
    };
    this.terminate = function () {
      Worker.prototype.terminate.call(oWorker);
    };
    this.addListener = function (sName, fListener) {
      oListeners[sName] = fListener;
    };
    this.removeListener = function (sName) {
      delete oListeners[sName];
    };
  };

  // your custom "queryable" worker
  var oMyTask = new QueryableWorker("my_task.js" /* , yourDefaultMessageListenerHere [optional], yourErrorListenerHere [optional] */);

  // your custom "listeners"

  oMyTask.addListener("printSomething", function (nResult) {
    document.getElementById("firstLink").parentNode.appendChild(document.createTextNode(" The difference is " + nResult + "!"));
  });

  oMyTask.addListener("alertSomething", function (nDeltaT, sUnit) {
    alert("Worker waited for " + nDeltaT + " " + sUnit + " :-)");
  });
</script>
</head>
<body>
  <ul>
    <li><a id="firstLink" href="javascript:oMyTask.sendQuery('getDifference', 5, 3);">What is the difference between 5 and 3?</a></li>
    <li><a href="javascript:oMyTask.sendQuery('waitSomething');">Wait 3 seconds</a></li>
    <li><a href="javascript:oMyTask.terminate();">terminate() the Worker</a></li>
  </ul>
</body>
</html>

my_task.js (the worker):

// your custom PRIVATE functions

function myPrivateFunc1 () {
  // do something
}

function myPrivateFunc2 () {
  // do something
}

// etc. etc.

// your custom PUBLIC functions (i.e. queryable from the main page)

var queryableFunctions = {
  // example #1: get the difference between two numbers:
  getDifference: function (nMinuend, nSubtrahend) {
      reply("printSomething", nMinuend - nSubtrahend);
  },
  // example #2: wait three seconds
  waitSomething: function () {
      setTimeout(function() { reply("alertSomething", 3, "seconds"); }, 3000);
  }
};

// system functions

function defaultQuery (vMsg) {
  // your default PUBLIC function executed only when main page calls the queryableWorker.postMessage() method directly
  // do something
}

function reply (/* listener name, argument to pass 1, argument to pass 2, etc. etc */) {
  if (arguments.length < 1) { throw new TypeError("reply - not enough arguments"); return; }
  postMessage({ "vo42t30": arguments[0], "rnb93qh": Array.prototype.slice.call(arguments, 1) });
}

onmessage = function (oEvent) {
  if (oEvent.data instanceof Object && oEvent.data.hasOwnProperty("bk4e1h0") && oEvent.data.hasOwnProperty("ktp3fm1")) {
    queryableFunctions[oEvent.data.bk4e1h0].apply(self, oEvent.data.ktp3fm1);
  } else {
    defaultQuery(oEvent.data);
  }
};

It is a possible method to switch the content of each mainpage-worker – and vice versa – message.

Przekazywanie danych przez przeniesienie własności (transferable objects)

Google Chrome 17 and Firefox 18 contain an additional way to pass certain kind of objects (Transferable objects, that is objects implementing the Transferable interface) to or from a worker with high performance. Transferable objects are transferred from one context to another with a zero-copy operation. This means a vast performance improvement when sending large data. Think of it as pass-by-reference if you're from the C/C++ world. However, unlike pass-by-reference, the 'version' from the calling context is no longer available once transferred. Its ownership is transferred to the new context. For example, when transferring an ArrayBuffer from your main app to Worker, the original ArrayBuffer is cleared and no longer usable. Its content is (quite literally) transferred to the Worker context.

// Create a 32MB "file" and fill it.
var uInt8Array = new Uint8Array(1024*1024*32); // 32MB
for (var i = 0; i < uInt8Array.length; ++i) {
  uInt8Array[i] = i;
}

worker.postMessage(uInt8Array.buffer, [uInt8Array.buffer]);

For more information on transferable objects, performance and feature-detection for this method, see the HTML5Rocks article.

Spawning subworkers

Workers may spawn more workers if they wish.  So-called subworkers must be hosted within the same origin as the parent page.  Also, the URIs for subworkers are resolved relative to the parent worker's location rather than that of the owning page.  This makes it easier for workers to keep track of where their dependencies are.

Subworkers are currently not supported in Chrome. See crbug.com/31666 .

Embedded workers

There is not an "official" way to embed the code of a worker within a web page as for the <script> elements. But a <script> element which does not have a src attribute and has a type attribute that does not identify an executable mime-type will be considered a data block element, that JavaScript could use.  "Data blocks" is a more general feature of HTML5 that can carry almost any textual data. So, a worker could be embedded in this way:

<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8" />
<title>MDN Example - Embedded worker</title>
<script type="text/js-worker">
  // This script WON'T be parsed by JS engines because its mime-type is text/js-worker.
  var myVar = "Hello World!";
  // Rest of your worker code goes here.
</script>
<script type="text/javascript">
  // This script WILL be parsed by JS engines because its mime-type is text/javascript.
  function pageLog (sMsg) {
    // Use a fragment: browser will only render/reflow once.
    var oFragm = document.createDocumentFragment();
    oFragm.appendChild(document.createTextNode(sMsg));
    oFragm.appendChild(document.createElement("br"));
    document.querySelector("#logDisplay").appendChild(oFragm);
  }
</script>
<script type="text/js-worker">
  // This script WON'T be parsed by JS engines because its mime-type is text/js-worker.
  onmessage = function (oEvent) {
    postMessage(myVar);
  };
  // Rest of your worker code goes here.
</script>
<script type="text/javascript">
  // This script WILL be parsed by JS engines because its mime-type is text/javascript.

  // In the past...:
  // blob builder existed
  // ...but now we use Blob...:
  var blob = new Blob(Array.prototype.map.call(document.querySelectorAll("script[type=\"text\/js-worker\"]"), function (oScript) { return oScript.textContent; }),{type: "text/javascript"});

  // Creating a new document.worker property containing all our "text/js-worker" scripts.
  document.worker = new Worker(window.URL.createObjectURL(blob));

  document.worker.onmessage = function (oEvent) {
    pageLog("Received: " + oEvent.data);
  };

  // Start the worker.
  window.onload = function() { document.worker.postMessage(""); };
</script>
</head>
<body><div id="logDisplay"></div></body>
</html>

The embedded worker is now nested into a new custom document.worker property.

Timeouts and intervals

Workers can use timeouts and intervals just like the main thread can.  This can be useful, for example, if you want to have your worker thread run code periodically instead of nonstop.

See setTimeout() , clearTimeout() , setInterval() , and clearInterval() for details. See also: JavaScript Timers.

Terminating a worker

If you need to immediately terminate a running worker, you can do so by calling the worker's terminate() method:

myWorker.terminate();

The worker thread is killed immediately without an opportunity to complete its operations or clean up after itself.

Workers may close themselves by calling their own nsIWorkerScope.close() method:

self.close();

Handling errors

When a runtime error occurs in worker, its onerror event handler is called.  It receives an event named error which implements the ErrorEvent interface.  The event doesn't bubble and is cancelable; to prevent the default action from taking place, the worker can call the error event's preventDefault() method.

The error event has the following three fields that are of interest:

message

A human-readable error message.

filename

The name of the script file in which the error occurred.

lineno

The line number of the script file on which the error occurred.

Accessing the navigator object

Workers may access the navigator object, which is available within their scope.  It contains the following strings which can be used to identify the browser, just as can be done from normal scripts:

• appName
• appVersion
• platform
• userAgent

Importing scripts and libraries

Worker threads have access to a global function, importScripts() , which lets them import scripts or libraries into their scope.  It accepts as parameters zero or more URIs to resources to import; all of the following examples are valid:

importScripts();                        /* imports nothing */
importScripts('foo.js');                /* imports just "foo.js" */
importScripts('foo.js', 'bar.js');      /* imports two scripts */

The browser loads each listed script and executes it. Any global objects from each script may then be used by the worker. If the script can't be loaded, NETWORK_ERROR is thrown, and subsequent code will not be executed. Previously executed code (including code deferred using window.setTimeout()) will still be functional though. Function declarations after the importScripts() method are also kept, since these are always evaluated before the rest of the code.

Examples

This section provides several examples of how to use DOM workers.

Performing computations in the background

One way workers are useful is to allow your code to perform processor-intensive calculations without blocking the user interface thread.  In this example, a worker is used to calculate Fibonacci numbers.

The JavaScript code

The following JavaScript code is stored in the "fibonacci.js" file referenced by the HTML in the next section.

var results = [];

function resultReceiver(event) {
  results.push(parseInt(event.data));
  if (results.length == 2) {
    postMessage(results[0] + results[1]);
  }
}

function errorReceiver(event) {
  throw event.data;
}

onmessage = function(event) {
  var n = parseInt(event.data);

  if (n == 0 || n == 1) {
    postMessage(n);
    return;
  }

  for (var i = 1; i <= 2; i++) {
    var worker = new Worker("fibonacci.js");
    worker.onmessage = resultReceiver;
    worker.onerror = errorReceiver;
    worker.postMessage(n - i);
  }
 };

The worker sets the property onmessage  to a function which will receive messages sent when the worker object's  postMessage() is called.  (Note that this differs from defining a global variable of that name, or defining a function with that name.   var onmessage and function onmessage will define global properties with those names, but they will not register the function to receive messages sent by the  web page that created the worker.)  This starts the recursion, spawning new copies of itself to handle each iteration of the calculation.

The HTML code

<!DOCTYPE html>
<html>
  <head>
    <meta charset="UTF-8"  />
    <title>Test threads fibonacci</title>
  </head>
  <body>

  <div id="result"></div>

  <script language="javascript">

    var worker = new Worker("fibonacci.js");

    worker.onmessage = function(event) {
      document.getElementById("result").textContent = event.data;
      dump("Got: " + event.data + "\n");
    };

    worker.onerror = function(error) {
      dump("Worker error: " + error.message + "\n");
      throw error;
    };

    worker.postMessage("5");

  </script>
  </body>
</html>

The web page creates a div element with the ID  result , which gets used to display the result, then spawns the worker.  After spawning the worker, the onmessage handler is configured to display the results by setting the contents of the div element, and the onerror handler is set to dump the error message.

Finally, a message is sent to the worker to start it.

Try this example .

Performing web I/O in the background

You can find an example of this in the article Using workers in extensions .

Dividing tasks among multiple workers

As multi-core computers become increasingly common, it's often useful to divide computationally complex tasks among multiple workers, which may then perform those tasks on multiple-processor cores.

example coming soon

Creating workers from within workers

The Fibonacci example shown previously demonstrates that workers can in fact spawn additional workers.  This makes it easy to create recursive routines.

Browser compatibility

See also

• File API Specification: Web WorkersED
• Worker interface
• SharedWorker interface
• Functions available to workers
• HTML5Rocks - The Basics of Web Workers
• Chrome has problems when using too many worker