Skip to content

testing.md

Latest commit

 

History

History
201 lines (149 loc) · 4.97 KB

testing.md

File metadata and controls

201 lines (149 loc) · 4.97 KB

Testing Sharpei

Sharpei has a comprehensive test suite covering the MCP server, REST API, and UI. All tests use temporary databases to avoid affecting your data.

Quick Start

# Activate virtual environment
source venv/bin/activate

# Run all tests
pytest

# Run with verbose output (default via pytest.ini)
pytest -v

Test Structure

tests/
├── conftest.py          # Shared fixtures
├── test_mcp_server.py   # MCP server tool tests
├── test_api.py          # FastAPI endpoint tests
└── test_ui.py           # Playwright UI tests

Test Categories

MCP Server Tests (test_mcp_server.py)

25 tests covering all MCP tools:

Category Tests Description
Categories 3 List, create, list with data
Tasks 9 CRUD operations, completion, archiving
Subtasks 3 Add subtask, inheritance, invalid parent
Filtering 6 By category, search, archived status
Archiving 2 Archive all, archive by category
Priority 1 Task ordering

API Tests (test_api.py)

28 tests covering all REST endpoints:

Category Tests Description
Root 1 HTML page load
Categories 5 CRUD, cascade delete behavior
Tasks 9 CRUD, 404 handling, cascade delete
Filtering 5 Category, search, case-insensitive, archived
Reordering 1 Drag-and-drop position updates
Archive 2 Archive completed, by category
Subtasks 3 Create, in response, hidden at top level
Position 2 Auto-assignment, per-priority

UI Tests (test_ui.py)

22 tests covering browser interactions:

Category Tests Description
Page Load 3 Title, banner, sidebar
Categories 3 Create, select, default selection
Tasks 5 Create, complete, uncomplete, delete
Expansion 3 Expand, edit description, change priority
Subtasks 1 Add subtask
Search 2 Search, clear search
Archiving 2 Archive completed, show archived
Show Details 1 Toggle details view
Hashtags 2 Add hashtag, filter by hashtag

Running Specific Tests

# Run only MCP tests
pytest tests/test_mcp_server.py

# Run only API tests
pytest tests/test_api.py

# Run only UI tests
pytest tests/test_ui.py

# Run a specific test class
pytest tests/test_api.py::TestTaskEndpoints

# Run a specific test
pytest tests/test_api.py::TestTaskEndpoints::test_create_task_minimal

# Run tests matching a pattern
pytest -k "search"
pytest -k "category"

Test Options

# Stop on first failure
pytest -x

# Show print statements
pytest -s

# Run failed tests from last run
pytest --lf

# Run tests in parallel (requires pytest-xdist)
pytest -n auto

# Generate coverage report (requires pytest-cov)
pytest --cov=app --cov-report=html

UI Test Options

# Run UI tests with browser visible (headed mode)
pytest tests/test_ui.py --headed

# Run in specific browser
pytest tests/test_ui.py --browser firefox
pytest tests/test_ui.py --browser webkit

# Slow down for debugging (milliseconds)
pytest tests/test_ui.py --slowmo 500

# Keep browser open on failure
pytest tests/test_ui.py --headed --slowmo 100 -x

Fixtures

Shared fixtures in conftest.py:

Fixture Scope Description
test_db function Temporary SQLite database
mcp_server function MCP server with test database
api_client function FastAPI TestClient
live_server function Running server for UI tests
ui_page function Playwright page navigated to app

Writing New Tests

API Test Example

def test_my_endpoint(self, api_client):
    """Test description."""
    # Create test data
    response = api_client.post("/api/tasks", json={"title": "Test"})
    assert response.status_code == 200

    # Verify behavior
    data = response.json()
    assert data["title"] == "Test"

UI Test Example

def test_my_feature(self, ui_page):
    """Test description."""
    # Interact with UI
    ui_page.locator("input[placeholder='Add a task...']").fill("Test")
    ui_page.locator("input[placeholder='Add a task...']").press("Enter")

    # Wait for changes
    ui_page.wait_for_selector(".task-title:has-text('Test')")

    # Verify
    expect(ui_page.locator(".task-list")).to_contain_text("Test")

Prerequisites

Install test dependencies:

pip install pytest pytest-playwright playwright
playwright install chromium

Continuous Integration

Example GitHub Actions workflow:

- name: Run tests
  run: |
    source venv/bin/activate
    playwright install chromium
    pytest --tb=short

Test Database

Tests use temporary SQLite databases created in the system temp directory. Each test gets a fresh database that is automatically deleted after the test completes. This ensures:

  • Tests are isolated from each other
  • No interference with your actual data
  • Tests can run in parallel safely