Phase 4 Combined Integration Test: Data Science Platform

Test ID

T-P4-003

Category

Combined Integration - Data Science with Platform-Specific Dependencies

Priority

P1

Description

This fixture simulates a data science/ML platform that combines platform-specific dependencies, heavy scientific packages, GPU support, and pre-release versions.

Features Combined

1. Platform Markers - Different dependencies for Linux/macOS/Windows
2. GPU Support - CUDA-specific packages on Linux
3. Heavy Dependencies - NumPy, Pandas, PyTorch, TensorFlow
4. Pre-release Versions - Testing with alpha/beta packages
5. Environment Markers - Python version and platform conditionals
6. Optional Dependencies - GPU, visualization, notebook extras

Real-World Inspiration

Based on patterns from:

• pytorch/pytorch installation requirements
• tensorflow/tensorflow platform-specific builds
• Data science projects with cross-platform support

Expected Dependency Tree Structure

data-science-platform
├── Core Scientific Stack (all platforms)
│   ├── numpy>=1.24.0
│   ├── pandas>=2.0.0
│   ├── scikit-learn>=1.3.0
│   ├── matplotlib>=3.7.0
│   └── scipy>=1.11.0
├── Platform-Specific Dependencies
│   ├── Linux + CUDA
│   │   ├── torch==2.1.0+cu118
│   │   └── nvidia-cudnn-cu11
│   ├── macOS (ARM)
│   │   └── torch==2.1.0
│   └── Windows
│       └── torch==2.1.0+cpu
├── Optional Features
│   ├── gpu group: torch with CUDA support
│   ├── viz group: plotly, seaborn, bokeh
│   └── notebook group: jupyter, ipykernel
└── Development Dependencies
    ├── pytest>=8.0.0
    ├── black>=24.0.0
    ├── isort>=5.12.0
    └── pre-commit>=3.5.0

Test Objectives

1. Verify platform marker evaluation (sys_platform, platform_machine)
2. Verify environment-specific dependencies are correctly filtered
3. Verify heavy dependencies with long transitive chains
4. Verify optional GPU support is properly grouped
5. Verify pre-release version handling
6. Verify cross-platform compatibility

Success Criteria

• Platform markers correctly evaluated
• Linux+CUDA dependencies only on appropriate platform
• macOS ARM-specific packages identified
• Windows-specific packages identified
• Heavy dependencies (torch, tensorflow) with correct versions
• Optional dependency groups properly separated
• Total dependency count varies by platform (30-50 packages)

UV Version Compatibility

• Minimum: UV 0.4.0+ (for enhanced platform support)
• Recommended: UV 0.7.0+

Files in this Fixture

• pyproject.toml - Project configuration with platform markers
• uv.lock - Lock file with platform-specific resolutions
• README.md - This file

Usage

# Generate lock file (platform-specific)
uv lock

# Sync with GPU support
uv sync --extra gpu

# Run dependency tree builder
dependency-tree-builder scan . > output.json

Platform-Specific Testing

This fixture should be tested on multiple platforms:

• Linux x86_64 with CUDA
• Linux x86_64 without CUDA
• macOS ARM64 (Apple Silicon)
• macOS x86_64 (Intel)
• Windows x86_64

Mend Integration Notes

This fixture tests Mend's ability to:

• Handle platform-specific dependencies
• Scan projects with conditional dependencies
• Report vulnerabilities in platform-specific packages
• Handle heavy scientific dependencies

Related Tests

• T-P2-009: Platform Markers
• T-P2-010: Pre-release Versions
• T-P2-008: Optional Dependencies