轮询、长轮询与 SSE
比较传统的轮询和长轮询方法与用于推送数据的服务器发送事件。
轮询、长轮询与 SSE 是 CoddyKit 上的免费 WebSockets & Realtime Systems Programming 课时。 这是第 2 节课,共 4 节。 你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。 这是 WebSockets & Realtime Systems Programming 学习路径的一部分,你的进度在网页和 CoddyKit 应用中同步。 WebSockets & Realtime Systems Programming 课程共包含 4 节课。
本课时的部分内容尚未翻译,以英文显示。
HTTP's Realtime Challenge
HTTP is stateless and unidirectional: the client asks, the server answers once. For live chat or tickers, constant asking or waiting is inefficient.
Introducing Polling
Polling is the simplest pseudo-realtime trick: the client asks the server for new data on a fixed interval, getting updates or an empty reply each time.
Polling Code Example
Here's a JavaScript client that polls a server every 2 seconds for updates — open your console to watch it fire.
<!DOCTYPE html>
<html>
<head>
<title>Polling Example</title>
</head>
<body>
<h1>Polling Status: <span id="status">Waiting...</span></h1>
<script>
function fetchData() {
// In a real app, this would be a server endpoint
fetch('https://jsonplaceholder.typicode.com/todos/1')
.then(response => response.json())
.then(data => {
const statusElement = document.getElementById('status');
statusElement.innerText = `Update at ${new Date().toLocaleTimeString()}`;
console.log('Polled:', data.title);
})
.catch(error => console.error('Polling error:', error));
}
// Poll every 2 seconds (2000 milliseconds)
setInterval(fetchData, 2000);
fetchData(); // Initial fetch to start
</script>
</body>
</html>Polling's Inefficiency
Polling is wasteful: high latency between polls, many empty responses burning resources, and poor scaling when many clients poll often.
Enter Long Polling
Long polling is smarter: instead of replying empty, the server holds the request open until data is ready or it times out, then the client reconnects.
How Long Polling Works
The long polling cycle: client requests, server waits for data or timeout, server responds and closes, client immediately re-requests to restart.
Long Polling Client Logic
Here's the long polling client pattern: it processes each response, then immediately reopens the connection (retrying after a delay on error).
function longPoll() {
console.log('Long polling for updates...');
fetch('/api/longpoll') // Imagine this endpoint holds the request
.then(response => response.json())
.then(data => {
if (data && data.message) {
console.log('Received update:', data.message);
// In a real app, update UI here
} else {
console.log('No new data, server likely timed out or sent empty.');
}
longPoll(); // Immediately send a new request
})
.catch(error => {
console.error('Long polling error:', error);
// Retry after a delay on error to prevent flooding
setTimeout(longPoll, 3000);
});
}
longPoll(); // Start the long polling processLong Polling's Pros & Cons
Long polling cuts latency and request count versus plain polling, but it still rides the request-response model — each update sets up and tears down a connection.
Server-Sent Events (SSE)
Server-Sent Events (SSE) give true server push over one long-lived HTTP connection. It's one-way (server to client) — perfect for feeds, tickers, notifications.
SSE Client Example
The native EventSource API makes subscribing to an SSE stream easy — the connection stays open until you close it or an error hits.
<!DOCTYPE html>
<html>
<head>
<title>SSE Example</title>
</head>
<body>
<h1>SSE Updates:</h1>
<ul id="events"></ul>
<script>
// Imagine a server at /sse-stream sending events
const eventSource = new EventSource('https://example.com/sse-stream'); // Replace with a real SSE endpoint
eventSource.onopen = () => {
console.log('SSE connection opened.');
const listItem = document.createElement('li');
listItem.textContent = `Connection opened at ${new Date().toLocaleTimeString()}`;
document.getElementById('events').appendChild(listItem);
};
eventSource.onmessage = (event) => {
const listItem = document.createElement('li');
listItem.textContent = `New message: ${event.data}`;
document.getElementById('events').appendChild(listItem);
console.log('Received SSE message:', event.data);
};
eventSource.onerror = (error) => {
console.error('SSE Error:', error);
const listItem = document.createElement('li');
listItem.textContent = `Error: ${error.message || 'Unknown'}`;
document.getElementById('events').appendChild(listItem);
eventSource.close(); // Close connection on error
};
// You can also listen for custom named events:
// eventSource.addEventListener('myCustomEvent', (event) => {
// console.log('Custom event data:', event.data);
// });
</script>
</body>
</html>Quick Check: Compare Methods
Which of the following statements about Polling, Long Polling, and Server-Sent Events (SSE) are true?
Recap: Unidirectional Push
Recap: polling repeatedly asks (simple but wasteful), long polling holds the request open, and SSE pushes one-way over a persistent connection.
常见问题解答
「轮询、长轮询与 SSE」课时是免费的吗?
是的 — 「轮询、长轮询与 SSE」的完整文本可在网页上免费阅读。要进行交互式练习(内置代码编辑器和全天候 AI 导师)并解锁 WebSockets & Realtime Systems Programming 课程的其余内容,请升级到 CoddyKit PRO。 WebSockets & Realtime Systems Programming 课程共包含 4 节课。
「轮询、长轮询与 SSE」这节课中我会学到什么?
比较传统的轮询和长轮询方法与用于推送数据的服务器发送事件。 你通过在浏览器中直接运行的动手代码来练习 WebSockets & Realtime Systems Programming,全天候 AI 导师会在你学习这节课的过程中回答你的问题。
学习 WebSockets & Realtime Systems Programming 需要有经验吗?
无需任何先前经验。CoddyKit 上的 WebSockets & Realtime Systems Programming 课程适合初学者到高级学习者,你可以从这里开始或从头开始,按照自己的节奏学习。 这是第 2 节课,共 4 节。
「轮询、长轮询与 SSE」课时需要多长时间?
大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动,所以你能稳步进步,并在网页和应用中从离开的地方继续。
我能在这节 WebSockets & Realtime Systems Programming 课中编写并运行代码吗?
能。每节 WebSockets & Realtime Systems Programming 课都包含内置代码编辑器,你可以在浏览器中直接编写并运行真实代码,并获得即时 AI 反馈 — 无需本地设置。
此课程中的所有课时
- 网络通信的演进
- 轮询、长轮询与 SSE
- WebSockets 简介:开启新时代
- 选择合适的实时技术