Gencovery Package Manager (GPM)

This repository allows managing Gencovery Web Services packages. It is used to create docker images to easily deploy GLab.

• Shell script init_lab.sh is run by the dockerfile entrypoint

• Python script init_lab.py is called by the init_lab.sh and allows pulling and intalling from Pip and Git any library described in an environment file config.json as given in folder tests/fixtures/config.json

Testing

Setup

1. Install dependencies:

   pip install -r requirements.txt

2. Create a .env file at project root with the following variables (optional for git pull tests):

   COMMUNITY_API_URL=<your_api_url>
   

Running Tests

The project uses pytest for testing. Tests are organized in a standard structure:

tests/
├── unit/              # Unit tests
├── integration/       # Integration tests
├── fixtures/          # Test data and configuration
└── conftest.py        # Shared fixtures

Run all tests:

pytest

Run specific test categories:

# Run only unit tests
pytest -m unit

# Run only integration tests
pytest -m integration

# Run tests in a specific file
pytest tests/unit/test_gpm.py

# Run tests with verbose output
pytest -v

# Run tests with coverage report (requires pytest-cov)
pytest --cov=init --cov-report=html

Run tests in VS Code:

• Use the Testing sidebar (flask icon)
• Tests are automatically discovered based on the pytest.ini configuration
• Click individual tests or test classes to run them

Test Markers

Tests are automatically marked based on their location and can also use custom markers:

• @pytest.mark.unit - Unit tests (auto-applied to tests/unit/*)
• @pytest.mark.integration - Integration tests (auto-applied to tests/integration/*)
• @pytest.mark.slow - Slow-running tests
• @pytest.mark.requires_network - Tests requiring network access
• @pytest.mark.requires_git - Tests requiring git operations

Filter tests by markers:

# Skip slow tests
pytest -m "not slow"

# Run only network tests
pytest -m requires_network

Testing the docker images

Testing the glab image locally : docker build -t glab:latest -f ./dockerfile/glab/Dockerfile . Testing the codelab image locally : docker build -t codelab:codelab -f ./dockerfile/codelab/Dockerfile .

Coding using the dev-env image : docker build -t lab-dev-env -f ./dockerfile/lab-dev-env/Dockerfile .

To run and tests the images, see lab-configurer repository.

Testing GPU the docker images

Testing the glab image locally : docker build --build-arg IMAGE="nvidia/cuda:12.4.1-runtime-ubuntu22.04" -t glab:gpu -f ./dockerfile/glab/Dockerfile .

Testing the codelab image locally : docker build --build-arg IMAGE="glab:gpu" -t local-codelab -f ./dockerfile/codelab/Dockerfile .

Building with buildx for multi-arch

From : https://itnext.io/building-multi-cpu-architecture-docker-images-for-arm-and-x86-2-building-in-gitlab-ci-295966b7185d

Create a new docker context for builder instance to use docker context create builder-context

Create a builder instance named "builderx" docker buildx create --name builderx --driver docker-container --use builder-context

docker buildx build -t glab:latest -f ./dockerfile/glab/Dockerfile --platform linux/amd64,linux/arm64 .