18_Testing_Philosophy

🧪 BioPro Testing Philosophy & Coverage

BioPro is built on a foundation of Test-Driven Development (TDD) and SOLID principles. Because the application handles sensitive scientific data and executes third-party plugin code, our testing suite is divided into three critical layers: Unit, Integration, and Security.

---

📊 Current Core Coverage (May 2026)

As of the latest audit, the BioPro core library maintains a 59% overall coverage with 108 tests passing. High-risk modules (Communication, Security, Data Integrity) are prioritized with coverage exceeding 85%.

Component	Coverage	Status
Event Bus (Nervous System)	96%	✅ Robust
Diagnostics Engine	94%	✅ Robust
Trust Architecture	88%	✅ High
History Manager (Undo/Redo)	86%	✅ High
Project & Asset Manager	75%	✅ Target Met
Network & Plugin Updates	65%	🎯 Targeting 80%
Resource Inspector	51%	🎯 Targeting 80%
SBOM & Supply Chain	0%	🎯 Targeting 90%

Note

Coverage of 0% in modules like preferences.py or sign_plugin.py is intentional; these are either purely aesthetic UI panels or CLI wrappers that are verified via end-to-end integration tests rather than unit tests.

---

🛠️ Testing Infrastructure

1. Headless Qt Testing

We use pytest-qt combined with an offscreen buffer to run UI tests without a physical display. This allows our CI/CD pipelines to verify signals, slots, and widget states.

# Run tests headlessly
QT_QPA_PLATFORM=offscreen .venv/bin/python3 -m pytest tests/core/

2. Cryptographic Verification

Tests in test_trust_architecture.py verify that the application correctly rejects tampered plugins, unauthorized authors, and corrupted manifests. We use mocked keys to simulate "Project" vs "Developer" trust levels.

3. Undo/Redo (The Time Machine)

The HistoryManager is subjected to heavy stress tests to ensure that complex scientific states can be serialized, saved, and restored without data loss or memory leaks.

---

🚀 Running the Suite

All developers are expected to run the suite locally before opening a Pull Request.

Standard Run

uv run pytest

With Coverage Report

uv run pytest --cov=biopro/core --cov-report=term-missing

---

📝 Best Practices

1. Mock External IO: Use monkeypatch or unittest.mock for network requests and home directory access.
2. Clean Teardown: Always use the temp_config_dir fixture to avoid polluting your local ~/.biopro folder.
3. Signal Tracking: Use qtbot.waitSignal to verify asynchronous UI behavior.

---

📈 Roadmap to 75% Coverage

We are currently executing an Acceleration Plan to move from 59% to 75% coverage by Q3 2026.

Priority Tracks:

1. Infrastructure Gaps: Closing missing logic paths in module_manager.py and network_updater.py.
2. Supply Chain Security: Achieving 90% coverage on the new SBOMGenerator and TrustManager.
3. Memory Stability: Increasing coverage of the ResourceInspector to 80% to ensure large-dataset reliability.