dtmcdona / dm_tasklist

Simple RPA System with Fast-API, OpenCV, Tesseract-OCR and a virtual display

DM_Tasklist

This project is a basic RPA system that uses Fast API endpoints as a backend for storing action sequences and streams screen data to the React.js frontend project (DM_React)

Docker setup

1. Docker compose up

      make dcu
   

2. To use the pytest tests:

      make tests
   

Local system setup:

1. This project uses a virtual xvfb display but if you would like to disable this feature then comment out or remove these from app/core/process_controller.py

   """Virtual display setup has to be setup before pyautogui is imported"""
   import Xlib.display
   from pyvirtualdisplay.display import Display
   disp = Display(visible=True, size=(1920, 1080), backend="xvfb", use_xauth=True)
   disp.start()
   

   """Virtual display"""
   pyautogui._pyautogui_x11._display = Xlib.display.Display(os.environ["DISPLAY"])
   

2. Install tesseract-ocr for pytesseract

   sudo apt install tesseract-ocr -y
   

3. Install this for pyenchant:

   sudo apt-get install libenchant1c2a -y
   

4. Make a new virtual environment and install requirements.txt

   python3 -m venv venv
   source venv/bin/activate
   pip install -r requirements.txt
   

5. This terminal commands will run the system on your local system:

   make local
   

Note:

I created two PostgreSQL adapter files with credentials in settings.py. The adapters are not part of the back-end. If you would like to use the PostgreSQL adapters then you need to setup a PostgreDB that match settings.py and connect the adapters to the endpoints. Currently, there is no need for a database but in the future it might be useful.