如何实现Ruby后台TCPServer，不阻塞主脚本执行？

Got it, let's sort out this problem so your TCP server can run in the background while your main script keeps executing normally, and continuously accept and print responses on your target port.

The core issue with your original code is that server.accept is a blocking call—it pauses the entire script until a client connects. To fix this, we can use Ruby's threads to offload the server's listening logic to a background thread, freeing up the main thread to run your other code.

Modified Working Code #

require 'socket'

# Configure your server host and port
srvhost = '0.0.0.0' # Listen on all network interfaces
srvport = 12345      # Replace with your desired port

# Spin up a background thread to handle the TCP server
server_thread = Thread.new do
  server = TCPServer.open(srvhost, srvport)
  puts "TCP server started, listening on port #{srvport}..."

  # Loop indefinitely to accept multiple client connections
  loop do
    # This accept call blocks the background thread (not the main thread!)
    client = server.accept
    puts "New connection from: #{client.peeraddr[2]}:#{client.peeraddr[1]}"

    # Receive and print incoming data
    data = client.recv(1024)
    puts "Received data: #{data}"

    client.close
    puts "Connection closed\n"
  end
end

# Main thread continues executing your original logic
puts "Main thread running other tasks..."

# Your original CGI request code (unchanged)
res = send_request_cgi({ 
  'uri' => "/", 
  'method' => 'GET', 
  'vars_get' => {'cmd' => rand_text_alphanumeric(10)} 
})
print "#{res.headers}"

# Keep the main thread alive so the background server thread doesn't terminate
server_thread.join

Key Details to Note #

• Thread.new: This launches the server's listening loop in a separate background thread. The blocking accept call now only pauses this background thread, not your main script.
• loop do: Wrapping the accept logic in a loop ensures the server keeps accepting new connections indefinitely, instead of just one.
• server_thread.join: This tells the main thread to wait for the background server thread to finish (which it never will, thanks to the loop). If your main script already has a continuous running loop (like processing other tasks), you can omit this line—just make sure the main thread doesn't exit early.

Bonus: Handling Concurrent Clients #

If you need to handle multiple clients connecting at the same time, you can spin up a new thread for each client connection. This prevents one slow client from blocking others:

# Inside the server thread's loop
loop do
  client = server.accept
  # Create a new thread for each client
  Thread.new(client) do |conn|
    puts "New connection from: #{conn.peeraddr[2]}:#{conn.peeraddr[1]}"
    data = conn.recv(1024)
    puts "Received data: #{data}"
    conn.close
    puts "Connection closed\n"
  end
end

内容的提问来源于stack exchange，提问作者user8819420