Skip to content

GAIA L3 Construct Python Tests

This directory contains Python unit tests for the GAIA L3 Construct shared Python modules.

Tested Modules

The test suite covers the following Python modules in the shared/ directory:

genai_core Package

  • Location: layers/python-sdk/python/genai_core/
  • Modules: auth, clients, workspaces, documents, embeddings, models, sessions, chunks, types, etc.
  • Subdirectories: aurora, bedrock_kb, kendra, langchain, opensearch, registry, utils, websites

File Import Batch Job

  • Location: file-import-batch-job/main.py
  • Purpose: Batch processing of file imports

Secrets Rotation Function

  • Location: secrets-rotation-function/index.py
  • Purpose: AWS Secrets Manager rotation functionality

Running Tests

# Navigate to python-tests directory
cd packages/constructs/L3/ai/gaia-l3-construct/lib/shared/python-tests

# Run all tests (uv handles setup automatically)
uv run pytest

# Run with coverage
uv run pytest --cov --cov-report=term-missing

Using npm Scripts

# Navigate to package root
cd packages/constructs/L3/ai/gaia-l3-construct

# Run all Python tests
npm run test:python

# Run with coverage
npm run test:python:coverage

# Run only unit tests
npm run test:python:unit

Test Categories

Tests are organized with markers for easy filtering:

# Run only genai_core module tests
uv run pytest -m genai_core

# Run only file import tests
uv run pytest -m file_import

# Run only secrets rotation tests
uv run pytest -m secrets

# Run only unit tests
uv run pytest -m unit

Test Structure

  • test_setup.py - Verifies Python path setup and module imports
  • test_genai_core_*.py - Tests for genai_core modules
  • test_file_import_main.py - Tests for file import batch job
  • test_secrets_rotation.py - Tests for secrets rotation function

Configuration

  • pyproject.toml - Modern Python project configuration with dependencies
  • conftest.py - Shared pytest fixtures and test setup

Dependencies

The test suite includes dependencies for: - AWS SDK testing (boto3, moto) - GenAI libraries (langchain, opensearch-py) - Database testing (psycopg2-binary) - Web scraping (beautifulsoup4, lxml) - HTTP mocking (responses)

Adding New Tests

  1. Create test files following the test_*.py naming convention
  2. Use appropriate markers (@pytest.mark.genai_core, etc.)
  3. Import modules using the paths set up in conftest.py
  4. Use the provided fixtures for AWS mocking and test data

Troubleshooting

If you encounter import errors: 1. Verify the Python paths in conftest.py are correct 2. Run test_setup.py to verify module imports 3. Check that all required dependencies are in pyproject.toml

For uv installation issues: 

# Install uv if not present
curl -LsSf https://astral.sh/uv/install.sh | sh
# or
pip install uv