Labra — Lightweight Automation & Baseline Reporting Architecture

Labra is a proof-of-concept framework that combines a simple Python test runner with optional reporting services. It is aimed at teams that want a lightweight starting point for automation while still being able to grow into a more complete continuous integration setup.

Features

• Plugin-style runner – src/labra/testengine.py loads configuration from config/config.yaml and executes tests using pytest.
• Multiple test scopes – tests are organised under test/unit, test/integration and test/e2e and can be filtered with pytest markers.
• Docker demo stack – demo_infra/docker-compose.yaml provisions GitLab CE, Allure reporting, Prometheus and Grafana for local experimentation.
• GitLab pipeline – the example .gitlab-ci.yml installs dependencies, runs the tests and uploads the resulting Allure report.

Repository layout

├── config/                 # YAML configuration used by the runner
├── demo_infra/             # docker‑compose stack for demo infrastructure
├── src/labra/              # test runner and reusable task helpers
├── test/                   # unit, integration and e2e tests
└── requirements.txt        # Python dependencies

src/labra/tasks/ contains small helper classes for the Playwright API and browser tasks used in the integration tests. These are intentionally minimal but show how additional tasks could be added.

Getting started

1. Install Python 3.11 and pip.

2. Install the required packages:

   pip install -r requirements.txt
   playwright install  # installs browser binaries for Playwright

3. Optionally start the demo services (Allure, GitLab, Prometheus and Grafana):

   docker compose -f demo_infra/docker-compose.yaml up -d

4. Run the tests using the provided configuration:

   python src/labra/testengine.py --config config/config.yaml

Test results will be written to the directory defined in the configuration file (by default output/). When the Allure service is running the runner will automatically post the results and trigger report generation.

Configuration

config/config.yaml controls which test scopes are executed and how results are reported. The key options are:

• enabled_scope – list of pytest markers to include.
• engine.test_root – directory that pytest should start in.
• reporting.allure_enabled – whether to upload results to Allure.
• reporting.allure_service – URL of the Allure Docker Service.

Feel free to adjust these values to match your environment.

Status

The project is intentionally small and many of the tests are placeholders. It acts as a template showing how a minimal automation framework can be assembled with Python, Playwright and standard DevOps tooling.