Python Socket实现TCP连接后仅获首字符，如何获取完整响应？

问题描述 #

我编写了一段使用Python socket库建立TCP连接、发送命令并接收响应的代码：

import socket

HOST = "MACHINE_IP_ADRESS"
PORT = "PORT_NUMBER"
COMMAND = "COMMAND_LINE"

# Connect to the server
with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
    s.connect((HOST, PORT))
    print("Connected to {}:{}".format(HOST, PORT))

    # Send the command
    s.sendall(COMMAND.encode('utf-8'))
    print("Sent command: {}".format(COMMAND))

    # Receive and print the answer
    answer = s.recv(1024).decode('utf-8')
    print('Received answer:', repr(answer))

在PuTTY终端中执行命令能得到完整响应>SERIAL NUMBER, XXXXXX，但运行该Python代码时仅收到第一个字符。我是Python新手，查阅资料后仍未解决，请问如何获取服务器的完整响应？

核心原因 #

TCP是流式协议，数据会被拆分成多个数据包分批发送，recv(1024)仅会读取当前socket缓冲区中最多1024字节的数据，无法保证一次性拿到完整响应。你遇到的情况就是服务器分批次返回数据，第一次recv只读取到了第一部分内容。

解决方法 #

1. 按响应结束标志循环读取 #

如果服务器返回的响应有固定结束标志（比如换行符\n、特定字符序列），可以循环读取直到触发该标志：

import socket

HOST = "MACHINE_IP_ADRESS"
PORT = 1234  # 注意：端口必须是整数，原代码中字符串类型会报错
COMMAND = "COMMAND_LINE"

with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
    s.connect((HOST, PORT))
    print("Connected to {}:{}".format(HOST, PORT))

    s.sendall(COMMAND.encode('utf-8'))
    print("Sent command: {}".format(COMMAND))

    # 初始化字节缓冲区存储完整响应
    full_response = b''
    while True:
        # 每次读取最多1024字节
        chunk = s.recv(1024)
        if not chunk:
            # 服务器关闭连接，停止读取
            break
        full_response += chunk
        # 假设响应以换行符结尾，可根据实际情况修改结束标志
        if b'\n' in chunk:
            break
    
    # 解码为字符串并去除多余空白符
    full_response = full_response.decode('utf-8').strip()
    print('Received answer:', repr(full_response))

2. 循环读取直到服务器关闭连接 #

如果服务器在发送完完整响应后会主动关闭TCP连接，可以循环读取直到recv返回空字节：

import socket

HOST = "MACHINE_IP_ADRESS"
PORT = 1234
COMMAND = "COMMAND_LINE"

with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
    s.connect((HOST, PORT))
    print("Connected to {}:{}".format(HOST, PORT))

    s.sendall(COMMAND.encode('utf-8'))
    print("Sent command: {}".format(COMMAND))

    full_response = b''
    while True:
        chunk = s.recv(1024)
        if not chunk:
            break
        full_response += chunk
    
    full_response = full_response.decode('utf-8')
    print('Received answer:', repr(full_response))

3. 按响应长度读取（需服务器协议支持） #

如果服务器会先返回响应的字节长度（比如先发送4字节表示长度，再发送对应长度的内容），可以先读取长度值，再精准读取对应字节数的内容。这种方式需要服务器遵循特定的协议约定，实现示例如下：

import socket
import struct

HOST = "MACHINE_IP_ADRESS"
PORT = 1234
COMMAND = "COMMAND_LINE"

with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
    s.connect((HOST, PORT))
    s.sendall(COMMAND.encode('utf-8'))

    # 先读取4字节的长度信息（假设用大端格式存储）
    length_data = s.recv(4)
    response_length = struct.unpack('!I', length_data)[0]

    # 读取对应长度的响应内容
    full_response = b''
    while len(full_response) < response_length:
        chunk = s.recv(min(1024, response_length - len(full_response)))
        if not chunk:
            break
        full_response += chunk
    
    full_response = full_response.decode('utf-8')
    print('Received answer:', repr(full_response))

内容的提问来源于stack exchange，提问作者Joan Molina