在浏览器中键入URL时会发生什么[关闭]

矫情吗；* 2022-11-27 00:36 159阅读 0赞

本文翻译自：[what happens when you type in a URL in browser \[closed\]][what happens when you type in a URL in browser _closed]

*Can somebody tell me what all happens behind the scenes from the time I type in a URL in the browser to the time when I get to see the page on the browser?* **从我在浏览器中输入URL到浏览器上的页面时，有人能告诉我幕后发生的一切吗？** *A detailed account of the process would be of great help.* **详细介绍该过程将有很大帮助。**

--------------------

### \#1楼 ###

参考：[https://stackoom.com/question/8Mmr/在浏览器中键入URL时会发生什么-关闭][https_stackoom.com_question_8Mmr_URL_-]

--------------------

### \#2楼 ###

*Look up the specification of HTTP.* **查找HTTP的规范。** *Or to get started, try [http://www.jmarshall.com/easy/http/][http_www.jmarshall.com_easy_http]* **或者要开始使用， [请访问http://www.jmarshall.com/easy/http/][http_www.jmarshall.com_easy_http]**

--------------------

### \#3楼 ###

*First the computer looks up the destination host.* **首先，计算机查找目标主机。** *If it exists in local DNS cache, it uses that information.* **如果它存在于本地DNS缓存中，则使用该信息。** *Otherwise, DNS querying is performed until the IP address is found.* **否则，执行DNS查询，直到找到IP地址。**

*Then, your browser opens a TCP connection to the destination host and sends the request according to HTTP 1.1 (or might use HTTP 1.0, but normal browsers don't do it any more).* **然后，您的浏览器打开与目标主机的TCP连接，并根据HTTP 1.1发送请求（或者可能使用HTTP 1.0，但普通浏览器不再执行此操作）。**

*The server looks up the required resource (if it exists) and responds using HTTP protocol, sends the data to the client (=your browser)* **服务器查找所需的资源（如果存在）并使用HTTP协议响应，将数据发送到客户端（=您的浏览器）**

*The browser then uses HTML parser to re-create document structure which is later presented to you on screen.* **然后，浏览器使用HTML解析器重新创建文档结构，稍后将在屏幕上显示该文档结构。** *If it finds references to external resources, such as pictures, css files, javascript files, these are is delivered the same way as the HTML document itself.* **如果它找到对外部资源的引用，例如图片，css文件，javascript文件，则它们的传递方式与HTML文档本身相同。**

--------------------

### \#4楼 ###

*Attention: this is an extremely rough and **oversimplified** sketch, assuming the simplest possible HTTP request (no HTTPS, no HTTP2, no extras), simplest possible DNS, no proxies, single-stack IPv4, one HTTP request only, a simple HTTP server on the other end, and no problems in any step.* **注意：这是一个非常粗略和过于简化的草图，假设最简单的HTTP请求（没有HTTPS，没有HTTP2，没有额外的），最简单的DNS，没有代理，单栈IPv4，只有一个HTTP请求，一个简单的HTTP服务器另一端，任何一步都没有问题。** *This is, for most contemporary intents and purposes, an unrealistic scenario;* **对于大多数当代意图和目的而言，这是一个不切实际的场景;** *all of these are far more complex in actual use, and the tech stack has become an order of magnitude more complicated since this was written.* **所有这些在实际使用中都要复杂得多，而且自编写以来，技术堆栈已变得更加复杂。** *With this in mind, the following timeline is still somewhat valid:* **考虑到这一点，以下时间表仍然有效：**

1.  *browser checks cache;* **浏览器检查缓存;** *if requested object is in cache and is fresh, skip to \#9* **如果请求的对象在缓存中并且是新鲜的，请跳到＃9**
2.  *browser asks OS for server's IP address* **浏览器询问操作系统的服务器IP地址**
3.  *OS makes a DNS lookup and replies the IP address to the browser* **操作系统进行DNS查找并将IP地址回复给浏览器**
4.  *browser opens a TCP connection to server (this step is much more complex with HTTPS)* **浏览器打开到服务器的TCP连接（这一步用HTTPS复杂得多）**
5.  *browser sends the HTTP request through TCP connection* **浏览器通过TCP连接发送HTTP请求**
6.  *browser receives HTTP response and may close the TCP connection, or reuse it for another request* **浏览器接收HTTP响应并可以关闭TCP连接，或者将其重新用于另一个请求**
7.  *browser checks if the response is a redirect or a conditional response (3xx result status codes), authorization request (401), error (4xx and 5xx), etc.;* **浏览器检查响应是重定向还是条件响应（3xx结果状态代码），授权请求（401），错误（4xx和5xx）等;** *these are handled differently from normal responses (2xx)* **这些处理方式与正常答案不同（2xx）**
8.  *if cacheable, response is stored in cache* **如果可缓存，则响应存储在缓存中**
9.  *browser decodes response (eg if it's gzipped)* **浏览器解码响应（例如，如果它被gzip压缩）**
10. *browser determines what to do with response (eg is it a HTML page, is it an image, is it a sound clip?)* **浏览器确定如何处理响应（例如，它是一个HTML页面，它是一个图像，它是一个声音剪辑吗？）**
11. *browser renders response, or offers a download dialog for unrecognized types* **浏览器呈现响应，或为无法识别的类型提供下载对话框**

*Again, discussion of each of these points have filled countless pages;* **再一次，对这些要点的讨论已经填满了无数页面;** *take this only as a summary, abridged for the sake of clarity.* **为了清楚起见，仅将其作为总结，删节。** *Also, there are many other things happening in parallel to this (processing typed-in address, speculative prefetching, adding page to browser history, displaying progress to user, notifying plugins and extensions, rendering the page while it's downloading, pipelining, connection tracking for keep-alive, cookie management, checking for malicious content etc.) - and the whole operation gets an order of magnitude more complex with HTTPS (certificates and ciphers and pinning, oh my!).* **此外，还有许多其他事情与此并行（处理输入的地址，推测预取，添加页面到浏览器历史记录，向用户显示进度，通知插件和扩展，在下载时呈现页面，流水线，连接跟踪保持活跃，cookie管理，检查恶意内容等） - 并且整个操作使用HTTPS（证书和密码和固定，哦，我的！）更复杂一个数量级。**

[what happens when you type in a URL in browser _closed]: https://oldbug.net/q/8Mmr/what-happens-when-you-type-in-a-URL-in-browser-closed
[https_stackoom.com_question_8Mmr_URL_-]: https://stackoom.com/question/8Mmr/%E5%9C%A8%E6%B5%8F%E8%A7%88%E5%99%A8%E4%B8%AD%E9%94%AE%E5%85%A5URL%E6%97%B6%E4%BC%9A%E5%8F%91%E7%94%9F%E4%BB%80%E4%B9%88-%E5%85%B3%E9%97%AD
[http_www.jmarshall.com_easy_http]: http://www.jmarshall.com/easy/http/