HolobiomicsLab python-unit-test-execution
Summary
Execute a pytest unit-test suite against a Python module to verify that all tests pass with exit code 0 and no assertion failures, confirming correctness of the implementation under test.
When to use
After rewriting or modifying a Python module (such as calculate_feature_overlap.py in a metabolomics analysis tool) and you need to verify that the refactored code maintains backward compatibility and correctness against the original test suite.
When NOT to use
- The module has no existing test suite — use test_driven_development or manual code review instead.
- Tests are known to be broken or outdated — fix or update the test suite first before executing.
- The testing environment cannot be replicated (e.g., missing system dependencies or conflicting package versions that cannot be resolved).
Inputs
- Python module source code (e.g., calculate_feature_overlap.py)
- pytest test suite files (test_*.py)
- Project configuration (pyproject.toml or requirements.txt)
- Python environment with pytest installed
Outputs
- pytest exit code (0 for success)
- Test execution report (verbose output listing all tests and pass/fail status)
- No assertion failures or errors in the test output
How to apply
Clone or access the repository containing both the module under test and its pytest test suite. Install pytest and any dependencies listed in the project's requirements or setup configuration (e.g., pyproject.toml). Locate the test files targeting the module of interest. Execute pytest with verbose output to capture detailed test results using a command like pytest path/to/test_file.py -v. Verify that the exit code is 0 and that all assertions pass without errors. Review the verbose output to ensure 100% of tests passed and no skipped or xfailed tests unexpectedly occurred.
Related tools
- pytest (Test runner and framework for executing unit tests and capturing pass/fail results with verbose reporting) — https://github.com/pytest-dev/pytest
- FERMO (Metabolomics analysis tool containing the calculate_feature_overlap.py module and associated test suite) — https://github.com/fermo-metabolomics/FERMO
Examples
pytest fermo_gui/tests/test_calculate_feature_overlap.py -v
Evaluation signals
- pytest exit code is exactly 0 (indicates all tests passed)
- Verbose output shows '== X passed in Y.XXs ==' with no failures, errors, or skipped tests
- All assertion statements in test files execute without raising AssertionError
- No ImportError, ModuleNotFoundError, or dependency resolution errors occur during test execution
- The module's public API and return types match the expectations encoded in the test suite (e.g., function signatures, output shapes for feature overlap calculations)
Limitations
- Test execution only validates that the current implementation matches the test expectations; it does not validate that the tests themselves are correct or comprehensive.
- Exit code 0 confirms test passage but does not measure code coverage — all code paths may not be tested.
- pytest execution is environment-dependent; tests may pass locally but fail in CI/CD if dependencies or Python versions differ.
- Tests may be xfailed or skipped intentionally; a 'passing' run may include skipped tests that mask real issues.
Evidence
- [other] Locate and execute the pytest test suite targeting calculate_feature_overlap.py using pytest with verbose output to capture test results.: "Locate and execute the pytest test suite targeting calculate_feature_overlap.py using pytest with verbose output to capture test results."
- [other] Verify that all tests pass with exit code 0 and no assertion failures or errors.: "Verify that all tests pass with exit code 0 and no assertion failures or errors."
- [other] Install pytest and any dependencies listed in the project's requirements or setup configuration.: "Install pytest and any dependencies listed in the project's requirements or setup configuration."
- [readme] Dependencies including exact versions are specified in the pyproject.toml file.: "Dependencies including exact versions are specified in the pyproject.toml file."