Self-healing Selenium Tests in Robot Framework using Healenium

Healenium is a tool based on Selenium that provides capability to heal (re-generate) a locator once the element could not be found. Read more about this tool at https://healenium.io

This is solution is based on the Healenium Docker containers for Python provided by the open-source project at https://github.com/healenium/healenium-example-python

Disclaimer

Please note that this is just a proof of concept, testing and debugging is still ongoing.

We would be happy to know your comments or issues on following e-mail: [email protected]

Prerequisities

1. install Python v3.9 or newer: https://www.python.org/downloads, including pip
2. using commandline terminal, navigate to the project root and create a new virtual environment and activate it
   • python -m venv venv or for MacOs python3 -m venv venv
   • .\venv\Scripts\activate or for MacOs source venv/bin/activate
3. install required libraries using pip:
   • pip install -r requirements.txt
4. install and run Docker Desktop: https://www.docker.com/products/docker-desktop
5. in the project directory, start commandline terminal and start the Healenium Docker containers using following commands:
   • docker-compose up -d - download Docker images and create containers
   • docker ps - list the created containers

Before running `docker-compose up -d`` on MacOs comment out lines 9 and 10:

volumes:
  - ./db/sql/init.sql:/docker-entrypoint-initdb.d/init.sql

New version of the lines should look like:

# volumes:
#   - ./db/sql/init.sql:/docker-entrypoint-initdb.d/init.sql

9. verify that following containers are up and running (either in the previous terminal output or directly in Docker Desktop client):
   • healenium/hlm-backend
   • postgres:11-alpine
   • healenium/hlm-proxy
   • healenium/hlm-selenium-4-standalone-xpra
   • healenium/hlm-selector-imitator

How to execute tests

Using commandline from project root:

• robot -d results tests/healenium_test.robot

To see the test execution, you can open http://localhost:8086 in your browser (replace localhost by the name of your server if running remotely)