Writing and Running Tests

🧪 This guide explains how to write and run tests using pytest, following the structure used in the template-project.

Writing good tests helps you avoid bugs, refactor safely, and ensure your code does what you expect — even weeks or months later.

📁 Where Tests Go

All test code lives in the tests/ directory at the root of the project.

Each Python file in this folder starts with test_
Each test function inside starts with test_
You can organize tests by topic or module: test_utils.py, test_readers.py, etc.

✍️ Writing Your First Test

Tests are written as plain Python functions and typically use assertions to check behavior:

# tests/test_math.py


def test_addition():
    assert 1 + 1 == 2

To test your own project modules, just import them like normal:

# tests/test_processors.py
from template_project.processors import convert_units_var


def test_convert_units_basic():
    result = convert_units_var(10, "m", "km")
    assert result == 0.01

💡 All test functions must start with test_ so pytest can discover them.

▶️ Running Tests Locally

Make sure you’ve installed the dev dependencies:

pip install -e ".[dev]"

Then run:

pytest

This will automatically find and run any test_*.py files under the tests/ folder.

To run a specific test file:

pytest tests/test_processors.py

To run a specific test function:

pytest tests/test_processors.py::test_convert_units_basic

🛠 Recommended Conventions

Group related tests by file (e.g. test_utilities.py, test_standardise.py)
Name test functions to reflect what they check (e.g. test_raises_on_invalid_units)
Use fixtures or test classes to share setup code when needed
Keep each test focused on one thing

📊 Checking Test Coverage

Test coverage shows which lines of code are executed during testing. Higher coverage generally means better testing, but 100% coverage doesn’t guarantee perfect tests.

Quick Coverage Check

# Basic coverage report in terminal
pytest --cov=template_project --cov-report=term-missing

This shows:

Percentage covered for each file
Missing lines that aren’t tested

Detailed Coverage Analysis

For a comprehensive view, generate an HTML report:

pytest --cov=template_project --cov-report=html

Then open htmlcov/index.html in your browser.

Reading Coverage Reports

Terminal output example:

Name                    Stmts   Miss  Cover   Missing
-----------------------------------------------------
template_project/processors/units.py    20      3    85%   63, 96-97
template_project/utils.py    78     24    69%   79-80, 113-120
-----------------------------------------------------
TOTAL                       354     76    79%

What this means:

Stmts: Total executable statements
Miss: Statements not covered by tests
Cover: Percentage covered
Missing: Specific line numbers not tested

HTML report shows:

🟢 Green lines: Covered by tests
🔴 Red lines: Not covered by tests
🟡 Yellow lines: Partially covered (e.g., branches)

Coverage Guidelines

Target coverage levels:

80%+: Good coverage for most projects
90%+: Excellent coverage
100%: Often impractical (includes error handling, edge cases)

Focus on:

Core business logic functions
Complex algorithms and calculations
Error handling paths
Public API functions

Less critical to test:

Simple getters/setters
Third-party library integration
Configuration/setup code

Run pytest locally before pushing so you catch test failures early. The CI test job also runs the suite on every pull request. See the linting & formatting guide for the ruff checks CI enforces.

Summary Cheatsheet

Task	Command
Run all tests	`pytest`
Run specific test file	`pytest tests/test_xyz.py`
Run specific test	`pytest tests/test_xyz.py::test_name`
Coverage report (terminal)	`pytest --cov=template_project --cov-report=term-missing`
Coverage report (HTML)	`pytest --cov=template_project --cov-report=html`
Install test tools	`pip install -e ".[dev]"`

✅ Tests help you make confident changes. Use them early, and use them often!