feat: Set up comprehensive Python testing infrastructure

[llbbl]

Set up Python Testing Infrastructure

This PR establishes a comprehensive testing infrastructure for the OpenNRE project, providing all the necessary tools and configuration for developers to write and run tests effectively.

Summary of Changes

Package Management

• Added Poetry configuration via pyproject.toml with project metadata and dependencies
• Migrated from requirements.txt to Poetry for better dependency management
• Set up development dependencies separately from production dependencies

Testing Framework Setup

• pytest configuration with comprehensive settings in pyproject.toml:
  • Test discovery patterns for files, classes, and functions
  • Custom markers: unit, integration, slow
  • Coverage reporting with 80% threshold
  • Strict configuration and marker enforcement
• Coverage reporting configured for HTML, XML, and terminal output
• Testing dependencies: pytest, pytest-cov, pytest-mock

Directory Structure

• Created tests/ directory with proper __init__.py files
• Organized test structure:
  • tests/unit/ for unit tests
  • tests/integration/ for integration tests
  • tests/conftest.py for shared fixtures

Shared Testing Fixtures

• Comprehensive fixtures in conftest.py including:
  • Temporary directories and files
  • Mock models, encoders, and tokenizers
  • Sample data structures for relation extraction
  • Configuration and training argument fixtures
  • Random seed fixtures for reproducible tests
  • AWS credential mocking
  • Automatic test environment setup
• Optional dependency handling - fixtures gracefully handle missing numpy/torch

Configuration Files

• Updated .gitignore with testing artifacts, IDE files, and Claude settings
• pyproject.toml includes pytest configuration, coverage settings, and build system

Validation

• Created validation tests to verify the infrastructure setup
• Tested fixture accessibility and pytest functionality
• Verified coverage reporting and custom markers work correctly

Running Tests

After installing dependencies, you can run tests using:

# Install all dependencies
poetry install

# Run all tests with coverage
poetry run pytest

# Run tests with verbose output
poetry run pytest -v

# Run only unit tests
poetry run pytest -m unit

# Run tests excluding slow ones
poetry run pytest -m "not slow"

# Generate coverage reports
poetry run pytest --cov=opennre --cov-report=html

Coverage Reports

The infrastructure generates coverage reports in multiple formats:

• Terminal output with missing lines
• HTML report in htmlcov/ directory
• XML report as coverage.xml

Notes

• The testing infrastructure is designed to work with or without ML dependencies (torch, numpy)
• Fixtures will skip tests gracefully if required dependencies are not available
• The 80% coverage threshold can be adjusted in pyproject.toml if needed
• Additional ML dependencies can be installed later without affecting the testing setup

Dependencies

Development Dependencies Added:

• pytest ^7.4.0 - Main testing framework
• pytest-cov ^4.1.0 - Coverage reporting
• pytest-mock ^3.11.0 - Mocking utilities

The infrastructure is now ready for developers to start writing comprehensive tests for the OpenNRE codebase.