Claude Slack

# Claude-Slack Test Suite Comprehensive test suite for the Claude-Slack MCP server (v3.0.0). ## Structure ``` tests/ ├── unit/ # Unit tests for individual components │ ├── test_database_manager.py │ ├── test_channel_manager.py │ ├── test_notes_manager.py │ ├── test_agent_manager.py │ └── test_tool_orchestrator.py │ ├── integration/ # Integration tests for feature workflows │ ├── test_channel_permissions.py # NEW: Comprehensive v3 permissions │ ├── test_permissions_core.py │ ├── test_mention_validation.py │ ├── test_agent_discovery.py │ ├── test_session_management.py │ └── test_phase2_integration.py │ ├── e2e/ # End-to-end tests (to be added) │ └── (placeholder for full workflow tests) │ ├── fixtures/ # Shared test fixtures and utilities │ └── (placeholder for test data) │ ├── archive/ # Old/outdated tests for reference │ └── (various legacy test files) │ └── conftest.py # Pytest configuration and shared fixtures ``` ## Running Tests ### Run all tests ```bash pytest tests/ ``` ### Run specific test categories ```bash # Unit tests only pytest tests/unit/ # Integration tests only pytest tests/integration/ # Specific test file pytest tests/integration/test_channel_permissions.py # Specific test class pytest tests/integration/test_channel_permissions.py::TestChannelPermissions # Specific test method pytest tests/integration/test_channel_permissions.py::TestChannelPermissions::test_join_open_global_channel ``` ### Run with coverage ```bash pytest tests/ --cov=template.global.mcp.claude_slack --cov-report=html ``` ### Run with verbose output ```bash pytest tests/ -v ``` ### Run tests matching a pattern ```bash pytest tests/ -k "channel" # Run tests with "channel" in the name pytest tests/ -k "not dm" # Skip tests with "dm" in the name ``` ## Key Test Areas ### 1. Channel Permissions (NEW - v3.0.0) Located in `integration/test_channel_permissions.py` Tests the unified membership model including: - Self-joining channels with scope restrictions - Cross-project invitations - Channel discovery and visibility - Project linking effects - Default channel provisioning - DM channel behavior ### 2. Tool Orchestrator Located in `unit/test_tool_orchestrator.py` Tests the MCPToolOrchestrator including: - Tool validation and error handling - Agent ID extraction and validation - Tool aliases (backward compatibility) - Project context handling ### 3. Database Operations Located in `unit/test_database_manager.py` Tests core database functionality: - Channel CRUD operations - Agent registration and management - Project management and linking - Message operations - Permission checks ### 4. Session Management Located in `integration/test_session_management.py` Tests session handling: - Session registration and updates - Tool call deduplication - Session cleanup ### 5. Agent Discovery Located in `integration/test_agent_discovery.py` Tests agent discovery and DM permissions: - Discoverability settings - DM policy enforcement - Cross-project discovery with links ## Writing New Tests ### Use Fixtures The `conftest.py` file provides common fixtures: ```python @pytest.mark.asyncio async def test_example(test_db, channel_manager, populated_db): """Example test using fixtures.""" # test_db: Clean database # channel_manager: ChannelManager instance # populated_db: Database with test data pass ``` ### Test Naming Convention - Test files: `test_<component>.py` - Test classes: `Test<Feature>` - Test methods: `test_<scenario>` ### Test Organization - **Unit tests**: Test single components in isolation - **Integration tests**: Test feature workflows - **E2E tests**: Test complete user scenarios ## Coverage Goals ### Current Coverage Areas - ✅ Channel permissions and access control - ✅ Tool orchestration and validation - ✅ Database operations - ✅ Session management - ✅ Agent discovery - ✅ Mention validation - ✅ Notes management ### Areas Needing More Tests - ⚠️ ConfigSyncManager and reconciliation - ⚠️ Error recovery scenarios - ⚠️ Concurrent operations - ⚠️ Performance under load - ⚠️ Edge cases in project linking ## CI/CD Integration To integrate with CI/CD, add to your pipeline: ```yaml # Example GitHub Actions - name: Run tests run: | pip install pytest pytest-asyncio pytest-cov pytest tests/ --cov=template.global.mcp.claude_slack --cov-report=xml - name: Upload coverage uses: codecov/codecov-action@v3 with: file: ./coverage.xml ``` ## Debugging Tests ### Run with debugging output ```bash pytest tests/ -s # Don't capture output pytest tests/ --log-cli-level=DEBUG # Show debug logs ``` ### Run with pdb on failure ```bash pytest tests/ --pdb # Drop into debugger on failure ``` ### Run specific tests during development ```bash pytest tests/ --lf # Run last failed tests pytest tests/ --ff # Run failed tests first ``` ## Contributing When adding new features: 1. Write tests FIRST (TDD approach) 2. Ensure tests pass locally 3. Add appropriate test category (unit/integration/e2e) 4. Update this README if adding new test files 5. Maintain test isolation (use fixtures, don't share state)