Python Automation Runner

A portable¹ python environment that exposes a basic but powerful set of API to write automation and assertion scripts to create testing scenarios for native/graphic applications. The environment will also function as a script runner, allowing the user to start and stop their script using windows hotkey. (this may be more of a script debug feature)

The main target are:

• provide a portable¹ environment to run python scripts via hotkey
• provide API in a domain language familiar to business/ testers/ QA to allow better business integration and easy transition to BDD.

Example testing script:

import logging
import time

import user_interaction as ui


logging.info("External script loaded")


def main_loop():

    logging.info("Current cursor position is: %s", str(ui.get_cursor_pos()))
    logging.info("Current foreground application executable is: %s", str(ui.get_foreground_window_executable()))

    logging.info("Clicking...")
    ui.click(100, 100)
    time.sleep(10)    

Description

The runner can be packaged as a windows executable (instructions below) that load, interpret, run and terminate an external python script via hotkeys.

If the external script will expose a function named main_loop, it will be run in loop indeed when CTRL-ALT-M will be pressed and terminated when CTRL-ALT-J is pressed.

The runner is also resilient to syntax/runtime errors in the external script, allowing the user to modify it and reload via hotkey without having to restart the runner.

NOTE: If you need to automate web application do NOT use this, please use Selenium.

Usage

Building the executable

python setup.py build

This will create a build/exe.win<PLATFORM>-<PYTHON_VERSION> folder that will contain the runner.exe

Running the runner...

runner.exe myscript.py

Starting and stopping the loop

• CTRL-ALT-M to start the main_loop
• CTRL-ALT-J to stop the main_loop

Roadmap: Future features

• Introduce more interaction API
• Allow user to bundle custom libraries into the executable or to load them from the custom script location
• Allow image matching to assert visual behaviour
  • Support dry run to to let the user capture the expected behaviour without having to capture the expected image manually
• Support Gherkin with behave
• Control the tested environment via REST to allow automations that involve multiple machines
• Reporting

¹ That can be copy/pasted in the target environment.