tomcat原理解析(一):一个简单的实现

it2022-05-05  192

一 概述

       前段时间去面试,被人问到了tomcat实现原理。由于平时没怎么关注容器的实现细节,这个问题基本没回答上来。所以最近花了很多时间一直在网上找资料和看tomcat的源码来研究里面处理一个HTTP请求的流程。网上讲tomcat的帖子比较多,大多都是直接切入主题看其源码,从我个人感受来说直接研究其源码实现比较难理解和非常枯燥,需要由简到难,慢慢深入。

二  一个简单tomcat服务器实现

        tomat是一个servlet容器,来处理http请求。在平时的使用中我们都会再浏览器中输入http地址来访问服务资源,比如格式http://host[":"port][abs_path]。从浏览器到服务端的一次请求都遵循http协议,在网络上其实走仍然是tcp协议,即我们常使用的socket来处理客户端和服务器的交互。根据输入的http地址可以知道服务器的IP地址和端口,根据这两个参数就可以定位到服务器的唯一地址。tomcat根据http地址端口后面的资源路径就可以知道反馈什么样的资源给浏览器。下面给出了一个非常简单的代码模拟了tomcat的简单实现

 

[html]  view plain  copy package com;    import java.io.*;  import java.net.ServerSocket;  import java.net.Socket;  import java.net.URLDecoder;  import java.util.StringTokenizer;    public class TomcatServer {        private final static int PORT = 8080;        public static void main(String[] args) {            try {              ServerSocket server = new ServerSocket(PORT);//根据端口号启动一个serverSocket              ServletHandler servletHandler=new ServletHandler(server);              servletHandler.start();          } catch (Exception e) {              e.printStackTrace();          }        }            private static class ServletHandler extends Thread{          ServerSocket server=null;          public ServletHandler(ServerSocket server){              this.server=server;          }              @Override          public void run() {              while (true) {                  try {                      Socket client = null;                      client = server.accept();//ServerSocket阻塞等待客户端请求数据                      if (client != null) {                          try {                              System.out.println("接收到一个客户端的请求");                                //根据客户端的Socket对象获取输入流对象。                              //封装字节流到字符流                              BufferedReader reader = new BufferedReader(new InputStreamReader(client.getInputStream()));                                // GET /test.jpg /HTTP1.1                              //http请求由三部分组成,分别是:请求行、消息报头、请求正文。                              //这里取的第一行数据就是请求行。http协议详解可以参考http://www.cnblogs.com/li0803/archive/2008/11/03/1324746.html说的很详细                              String line = reader.readLine();                                System.out.println("line: " + line);                                //拆分http请求路径,取http需要请求的资源完整路径                              String resource = line.substring(line.indexOf('/'),line.lastIndexOf('/') - 5);                                System.out.println("the resource you request is: "+ resource);                                resource = URLDecoder.decode(resource, "UTF-8");                                //获取到这次请求的方法类型,比如get或post请求                              String method = new StringTokenizer(line).nextElement().toString();                                System.out.println("the request method you send is: "+ method);                                //继续循环读取浏览器客户端发出的一行一行的数据                              while ((line = reader.readLine()) != null) {                                  if (line.equals("")) {//当line等于空行的时候标志Header消息结束                                      break;                                  }                                  System.out.println("the Http Header is : " + line);                              }                                //如果是POST的请求,直接打印POST提交上来的数据                              if ("post".equals(method.toLowerCase())) {                                  System.out.println("the post request body is: "                                          + reader.readLine());                              }else if("get".equals(method.toLowerCase())){                                  //判断是get类型的http请求处理                                  //根据http请求的资源后缀名来确定返回数据                                    //比如下载一个图片文件,我这里直接给定一个图片路径来模拟下载的情况                                  if (resource.endsWith(".jpg")) {                                      transferFileHandle("d://123.jpg", client);                                      closeSocket(client);                                      continue;                                    } else {                                 //直接返回一个网页数据                               //其实就是将html的代码以字节流的形式写到IO中反馈给客户端浏览器。                               //浏览器会根据http报文“Content-Type”来知道反馈给浏览器的数据是什么格式的,并进行什么样的处理                                 PrintStream writer = new PrintStream(client.getOutputStream(), true);                               writer.println("HTTP/1.0 200 OK");// 返回应答消息,并结束应答                               writer.println("Content-Type:text/html;charset=utf-8");                               writer.println();                               //writer.println("Content-Length:" + html.getBytes().length);// 返回内容字节数                               writer.println("<html><body>");                               writer.println("<href='www.baidu.com'>百度</a>");                               writer.println("<img src='https://ss0.bdstatic.com/5aV1bjqh_Q23odCf/static/superman/img/logo/bd_logo1_31bdc765.png'></img>");                               writer.println("</html></body>");                                   //writer.println("HTTP/1.0 404 Not found");// 返回应答消息,并结束应答                               writer.println();// 根据 HTTP 协议, 空行将结束头信息                               writer.close();                               closeSocket(client);//请求资源处理完毕,关闭socket链接                               continue;                                  }                              }                                } catch (Exception e) {                              System.out.println("HTTP服务器错误:"                                      + e.getLocalizedMessage());                          }                      }                  } catch (Exception e) {                      e.printStackTrace();                  }              }          }            private void closeSocket(Socket socket) {              try {                  socket.close();              } catch (IOException ex) {                  ex.printStackTrace();              }              System.out.println(socket + "离开了HTTP服务器");          }            private void transferFileHandle(String path, Socket client) {                File fileToSend = new File(path);                if (fileToSend.exists() && !fileToSend.isDirectory()) {                  try {                      //根据Socket获取输出流对象,将访问的资源数据写入到输出流中                      PrintStream writer = new PrintStream(client.getOutputStream());                      writer.println("HTTP/1.0 200 OK");// 返回应答消息,并结束应答                      writer.println("Content-Type:application/binary");                      writer.println("Content-Length:" + fileToSend.length());// 返回内容字节数                      writer.println();// 根据 HTTP 协议, 空行将结束头信息                        FileInputStream fis = new FileInputStream(fileToSend);                      byte[] buf = new byte[fis.available()];                      fis.read(buf);                      writer.write(buf);                      writer.close();                      fis.close();                  } catch (IOException e) {                      e.printStackTrace();                  }              }          }        }    }  

 

三  实践

    1.在浏览器中输入http://localhost:8080/123.jpg 链接,可以看到浏览器里面就将123.jpg下载到本地了。

    2.在浏览器中输入一个服务器不能识别的请求后缀比如http://localhost:8080/123.jpg1,可以看到浏览器打开了一个网页。如下图:点击里面的百度链接可以跳转

    3.后台tomcat服务器打印的http请求报文

 

      接收到一个客户端的请求line: GET /123.jpg1 HTTP/1.1the resource you request is: /123.jpg1the request method you send is: GETthe Http Header is : Host: localhost:8080the Http Header is : Connection: keep-alivethe Http Header is : Pragma: no-cachethe Http Header is : Cache-Control: no-cachethe Http Header is : Accept: text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8the Http Header is : Upgrade-Insecure-Requests: 1the Http Header is : User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/47.0.2526.106 Safari/537.36the Http Header is : Accept-Encoding: gzip, deflate, sdchthe Http Header is : Accept-Language: zh-CN,zh;q=0.8Socket[addr=/0:0:0:0:0:0:0:1,port=57864,localport=8080]离开了HTTP服务器

四  总结

从整个代码和测试情况来看,一次http请求其实就是一次socket套接字的处理。浏览器发起scoket的请求,tomcat服务器接受请求,并根据请求的路径定位客户端需要访问的资源。  只是socket客户端和服务器数据在交互时,都遵守着http协议规范。当然真正的tomcat容器比这个demo实现要复杂的很多,这个简易的tomcat服务器能够帮我们更好的理解tomcat源码。

转载于:https://www.cnblogs.com/csguo/p/7499395.html


最新回复(0)