Remote Automation (Socket Server)

Overview

WebRunner includes a multi-threaded TCP socket server for remote automation control. This enables cross-language support – any language that supports TCP sockets (Java, C#, Go, etc.) can send automation commands to WebRunner.

Starting the Server

from je_web_runner import start_web_runner_socket_server

server = start_web_runner_socket_server(host="localhost", port=9941)

The server starts in a background daemon thread and is ready to accept connections immediately.

You can also override the host and port via command-line arguments:

python your_script.py localhost 9941

Client Connection Example

import socket
import json

sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.connect(("localhost", 9941))

# Send actions as JSON (UTF-8 encoded)
actions = [
    ["WR_get_webdriver_manager", {"webdriver_name": "chrome"}],
    ["WR_to_url", {"url": "https://example.com"}],
    ["WR_quit"]
]
sock.send(json.dumps(actions).encode("utf-8"))

# Receive results (ends with "Return_Data_Over_JE\n")
response = sock.recv(4096).decode("utf-8")
print(response)

# Shutdown server
sock.send("quit_server".encode("utf-8"))

Protocol

Property	Value
Default host	localhost
Default port	9941
Encoding	UTF-8
Message format	JSON array of actions
Max receive buffer	8192 bytes
Response terminator	Return_Data_Over_JE
Shutdown command	quit_server
Threading model	Multi-threaded (socketserver.ThreadingMixIn)

Response Format

After executing actions, the server sends results back one by one. Each result is followed by a newline (\n). The final message is always Return_Data_Over_JE\n.

If an error occurs, the error message is sent to the client followed by the same terminator.

Server Classes

TCPServerHandler (extends socketserver.BaseRequestHandler): Handles incoming requests, parses JSON, executes actions, and returns results.

TCPServer (extends socketserver.ThreadingMixIn, socketserver.TCPServer): Multi-threaded TCP server with a close_flag attribute for shutdown control.