Synchronous and asynchronous requests - Web APIs 编辑

XMLHttpRequest supports both synchronous and asynchronous communications. In general, however, asynchronous requests should be preferred to synchronous requests for performance reasons.

Synchronous requests block the execution of code which causes "freezing" on the screen and an unresponsive user experience.

Asynchronous request

If you use an asynchronous XMLHttpRequest, you receive a callback when the data has been received. This lets the browser continue to work as normal while your request is being handled.

Example: send a file to the console log

This is the simplest usage of asynchronous XMLHttpRequest.

var xhr = new XMLHttpRequest();
xhr.open("GET", "/bar/foo.txt", true);
xhr.onload = function (e) {
  if (xhr.readyState === 4) {
    if (xhr.status === 200) {
      console.log(xhr.responseText);
    } else {
      console.error(xhr.statusText);
    }
  }
};
xhr.onerror = function (e) {
  console.error(xhr.statusText);
};
xhr.send(null); 

Line 2 specifies true for its third parameter to indicate that the request should be handled asynchronously.

Line 3 creates an event handler function object and assigns it to the request's onload attribute. This handler looks at the request's readyState to see if the transaction is complete in line 4; if it is, and the HTTP status is 200, the handler dumps the received content. If an error occurred, an error message is displayed.

Line 15 actually initiates the request. The callback routine is called whenever the state of the request changes.

Example: writing a function to read an external file

In some cases, you must read many external files. This is a standard function which uses the XMLHttpRequest object asynchronously in order to switch the content of the read file to a specified listener.

function xhrSuccess() {
    this.callback.apply(this, this.arguments);
}

function xhrError() {
    console.error(this.statusText);
}

function loadFile(url, callback /*, opt_arg1, opt_arg2, ... */) {
    var xhr = new XMLHttpRequest();
    xhr.callback = callback;
    xhr.arguments = Array.prototype.slice.call(arguments, 2);
    xhr.onload = xhrSuccess;
    xhr.onerror = xhrError;
    xhr.open("GET", url, true);
    xhr.send(null);
}

Usage:

function showMessage(message) {
    console.log(message + this.responseText);
}

loadFile("message.txt", showMessage, "New message!\n\n");

The signature of the utility function loadFile declares (i) a target URL to read (via an HTTP GET request), (ii) a function to execute on successful completion of the XHR operation, and (iii) an arbitrary list of additional arguments that are passed through the XHR object (via the arguments property) to the success callback function.

Line 1 declares a function invoked when the XHR operation completes successfully. It, in turn, invokes the callback function specified in the invocation of the loadFile function (in this case, the function showMessage) which has been assigned to a property of the XHR object (Line 11). The additional arguments (if any) supplied to the invocation of function loadFile are "applied" to the running of the callback function.

Line 5 declares a function invoked when the XHR operation fails to complete successfully.

Line 11 stores the success callback given as the second argument to loadFile in the XHR object's callback property.

Line 12 slices the arguments array given to the invocation of loadFile. Starting with the third argument, all remaining arguments are collected, assigned to the arguments property of the variable xhr, passed to the success callback function xhrSuccess., and ultimately supplied to the callback function (in this case, showMessage) which is invoked by function xhrSuccess.

Line 15 specifies true for its third parameter to indicate that the request should be handled asynchronously.

Line 16 actually initiates the request.

Example: using a timeout

You can use a timeout to prevent your code from hanging while waiting for a read to finish. This is done by setting the value of the timeout property on the XMLHttpRequest object, as shown in the code below:

function loadFile(url, timeout, callback) {
    var args = Array.prototype.slice.call(arguments, 3);
    var xhr = new XMLHttpRequest();
    xhr.ontimeout = function () {
        console.error("The request for " + url + " timed out.");
    };
    xhr.onload = function() {
        if (xhr.readyState === 4) {
            if (xhr.status === 200) {
                callback.apply(xhr, args);
            } else {
                console.error(xhr.statusText);
            }
        }
    };
    xhr.open("GET", url, true);
    xhr.timeout = timeout;
    xhr.send(null);
}

Notice the addition of code to handle the "timeout" event by setting the ontimeout handler.

Usage:

function showMessage (message) {
    console.log(message + this.responseText);
}

loadFile("message.txt", 2000, showMessage, "New message!\n");

Here, we're specifying a timeout of 2000 ms.

Note: Support for timeout was added in Gecko 12.0.

Synchronous request

Note: Starting with Gecko 30.0 (Firefox 30.0 / Thunderbird 30.0 / SeaMonkey 2.27), Blink 39.0, and Edge 13, synchronous requests on the main thread have been deprecated due to their negative impact on the user experience.

Synchronous XHR requests often cause hangs on the web. But developers typically don't notice the problem because the hang only manifests with poor network conditions or when the remote server is slow to respond. Synchronous XHR is now in deprecation state. The recommendation is that developers move away from the synchronous API and instead use asynchronous requests.

All new XHR features such as timeout or abort are not allowed for synchronous XHR. Doing so will raise an InvalidAccessError.

Example: HTTP synchronous request

This example demonstrates how to make a simple synchronous request.

var request = new XMLHttpRequest();
request.open('GET', '/bar/foo.txt', false);  // `false` makes the request synchronous
request.send(null);

if (request.status === 200) {
  console.log(request.responseText);
}

Line 3 sends the request. The null parameter indicates that no body content is needed for the GET request.

Line 5 checks the status code after the transaction is completed. If the result is 200 -- HTTP's "OK" result -- the document's text content is output to the console.

Example: Synchronous HTTP request from a Worker

One of the few cases in which a synchronous request does not usually block execution is the use of XMLHttpRequest within a Worker.

example.html (the main page):

<!doctype html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<title>MDN Example</title>
<script type="text/javascript">
  var worker = new Worker("myTask.js");
  worker.onmessage = function(event) {
    alert("Worker said: " + event.data);
  };

  worker.postMessage("Hello");
</script>
</head>
<body></body>
</html>

myFile.txt (the target of the synchronous XMLHttpRequest invocation):

Hello World!!

myTask.js (the Worker):

self.onmessage = function (event) {
  if (event.data === "Hello") {
    var xhr = new XMLHttpRequest();
    xhr.open("GET", "myFile.txt", false);  // synchronous request
    xhr.send(null);
    self.postMessage(xhr.responseText);
  }
};

Note: The effect is asynchronous, because of the use of the Worker.

This pattern can be useful, for example in order to interact with the server in the background, or to preload content. See Using web workers for examples and details.

Adapting Sync XHR use cases to the Beacon API

There are some cases in which the synchronous usage of XMLHttpRequest is not replaceable, like during the unload, beforeunload, and pagehide events. You should consider using the fetch() API with the keepalive flag. When fetch with keepalive isn't available, you can consider using the navigator.sendBeacon() API, which can support these use cases while typically delivering a good UX.

The following example shows theoretical analytics code that attempts to submit data to a server by using a synchronous XMLHttpRequest in an unload handler. This results in the unloading of the page to be delayed.

window.addEventListener('unload', logData, false);

function logData() {
    var client = new XMLHttpRequest();
    client.open("POST", "/log", false); // third parameter indicates sync xhr. :(
    client.setRequestHeader("Content-Type", "text/plain;charset=UTF-8");
    client.send(analyticsData);
}

Using the sendBeacon() method, the data will be transmitted asynchronously to the web server when the User Agent has had an opportunity to do so, without delaying the unload or affecting the performance of the next navigation.

The following example shows a theoretical analytics code pattern that submits data to a server by using the sendBeacon() method.

window.addEventListener('unload', logData, false);

function logData() {
    navigator.sendBeacon("/log", analyticsData);
}

See also

• Using XMLHttpRequest
• AJAX
• navigator.sendBeacon