在 python 中使用套接字编写代理 Web 服务器

Question

这是我在大学计算机网络课程研讨会上的代码。

要求：编写一个代理网络服务器。从浏览器端口 8888 接收 HTTP 请求，并将 HTTP 请求发送到 Web 服务器端口 80。似乎我向 Web 服务器发送请求时遇到了问题。

你能告诉我在这种情况下我的错误吗？

非常感谢

import socket
import sys
import _thread
import traceback
import ssl
def CreateServer(host, port): 
    Server = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    Server.bind((host,port))
    Server.listen(5)
    return Server

def ReadRequest(Client):
    re = ""
    Client.settimeout(10.0)
try:
    re = Client.recv(1024).decode()
    while (re):
        re = re + Client.recv(1024).decode()
except socket.timeout: # fail after 1 second of no activity
    if not re:
        print("Didn't receive data! [Timeout]")
finally:
    return re

#2. Client connect Server + 3. Read HTTP Request
def ReadHTTPRequest(Server): 
    re = ""
    while (re == ""):
        Client, address = Server.accept()
        print("Client: ", address," da ket noi toi Server")
        re = ReadRequest(Client)
    return Client,address, re

def proxy_server(webserver, port, conn, data, addr):
    print("{} {} {} {}".format(webserver, port, conn, addr))
    try:
        s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
        s.settimeout(10.0)
        s.connect((webserver, port))
        s.sendall(data)
        while 1:
            reply = s.recv(1024)
            if len(reply) > 0:
                conn.send(reply)
                print("[*] Request sent: {} > {}".format(addr[0],webserver))
            else:
                break        
    
        s.close()
        conn.close()
    
    except Exception as e:
        print(e)
        traceback.print_exc()
        s.close()
        conn.close()
        sys.exit(1)

def conn_string(Client,Request,addr):
    try:
        #print(addr)
        first_line=Request.split('\n')[0]
        url=first_line.split(" ")[1]
        http_pos=url.find("://")
        if http_pos==-1:
            temp=url
        else:
            temp=url[(http_pos+3):]
        port_pos=temp.find(":")
        webserver_pos=temp.find("/")
        if webserver_pos == -1:
            webserver_pos = len(temp)
        webserver = ""
        port = -1
        if port_pos == -1 or webserver_pos < port_pos:
            port = 80
            webserver = temp[:webserver_pos]
        else:
            port = 80
            #port = int(temp[(port_pos + 1):][:webserver_pos - port_pos -1])
            webserver = temp[:port_pos]
        proxy_server(webserver,port,Client,Request.encode(),addr)
    except Exception as e:
        print (e)
        traceback.print_exc()



if __name__=="__main__":
    try:
        Server=CreateServer("",8888)
        print("[*] Intializing socket. Done.")
        print("[*] Socket binded successfully...")
        print("[*] Server started successfully [{}]".format(8888))
    except Exception as e:
        print(e)
        sys.exit(2)
    while True:
        try:
             Client,addr,Request=ReadHTTPRequest(Server)
             print("---------HTTP request: ")
             print(Request)
             _thread.start_new_thread(conn_string,(Client,Request,addr))
        except KeyboardInterrupt:
            Server.close()
            print("\n[*] Shutting down..")
            sys.exit()
    Server.close()

  [1]: https://i.stack.imgur.com/216ZO.png

Answer 1

您只是将原始请求转发到服务器，即请求中包含完整的 URL：

GET http://ktdbcl.hcmus.edu.vn/ HTTP/1.1
Host: ktdbcl.hcmus.edu.vn
...

应该只转发路径：

GET / HTTP/1.1
Host: ktdbcl.hcmus.edu.vn
...

除此之外，请求和响应的读取被严重破坏。您没有正确解析 HTTP 协议，而是简单地等待一些时间并将几秒钟内没有数据视为消息结束。由于今天多个请求和响应将通过同一连接完成，因此您的代码将严重减慢任何浏览速度。

您应该正确解析 HTTP 协议并等待由 HTTP 标准定义的适当的请求结束标记，而不是等待数秒没有数据作为请求/响应结束的标记 - 请参阅 RFC 7230。