如何用Python处理Netcat服务器输出并返回算术运算结果？

Your current approach using subprocess to call nc has a couple of critical issues:

• Race condition: You might attempt to read ncoutput.txt before nc finishes writing the numbers to it, leading to empty or incomplete data.
• One-way communication: Once nc exits after sending the numbers, the connection closes—you can’t send the computed result back to the server.

A more reliable solution uses Python’s built-in socket module to handle the connection directly. This keeps the connection open for both receiving data and sending the result, eliminating the need for intermediate files or external processes. Here’s a complete, robust implementation:

#!/usr/bin/env python
import socket

# Server configuration
HOST = 'domain.com'
PORT = 1234

try:
    # Establish a socket connection to the server
    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
        s.connect((HOST, PORT))
        
        # Receive data from the server (adjust buffer size if needed)
        raw_data = s.recv(1024).decode('utf-8').strip()
        
        # Parse the received string into integers
        numbers = list(map(int, raw_data.split()))
        
        # Validate we received exactly three numbers
        if len(numbers) != 3:
            raise ValueError(f"Expected 3 numbers, got {len(numbers)}")
        
        num1, num2, num3 = numbers
        
        # Handle division by zero edge case
        if num3 == 0:
            raise ZeroDivisionError("Third number cannot be zero (division by zero)")
        
        # Calculate the result (use / for float division if required)
        result = (num1 * num2) // num3
        
        # Send the result back to the server (add newline if server expects it)
        s.sendall(str(result).encode('utf-8'))
        
        # Optional: Read a confirmation response from the server
        confirmation = s.recv(1024).decode('utf-8').strip()
        print(f"Server confirmation: {confirmation}")

except ConnectionRefusedError:
    print("Error: Could not connect to the server. Is it running?")
except ValueError as ve:
    print(f"Input error: {ve}")
except ZeroDivisionError as zde:
    print(f"Calculation error: {zde}")
except Exception as e:
    print(f"Unexpected error: {e}")

Key Improvements: #

• Persistent connection: Uses a single socket session to both receive numbers and send the result, avoiding subprocess/file I/O pitfalls.
• Error handling: Catches common issues like connection failures, invalid input formats, division by zero, and unexpected errors.
• Direct data flow: Reads server responses directly into memory, no intermediate files required.

Notes: #

• If the server sends more data than the 1024-byte buffer, increase the value in recv() (e.g., recv(4096)).
• If the server expects a newline after the result, modify the send line to s.sendall(f"{result}\n".encode('utf-8')).
• Swap // with / if you need floating-point division instead of integer division.

If you must use nc (not recommended for this use case), you can use bidirectional pipes to avoid race conditions:

#!/usr/bin/env python
import subprocess

with subprocess.Popen(
    ["nc", "domain.com", "1234"],
    stdin=subprocess.PIPE,
    stdout=subprocess.PIPE,
    text=True
) as proc:
    # Read numbers from nc's output
    output = proc.stdout.readline().strip()
    numbers = list(map(int, output.split()))
    
    # Compute result
    result = (numbers[0] * numbers[1]) // numbers[2]
    
    # Send result to nc's input (forwards to server)
    proc.stdin.write(f"{result}\n")
    proc.stdin.flush()
    
    # Optional: Read server response
    response = proc.stdout.readline().strip()
    print(f"Server response: {response}")

This version uses pipes to communicate with nc bidirectionally, but the socket module remains the more efficient and controllable option.

内容的提问来源于stack exchange，提问作者adam