python-test-and-fix

name	python-test-and-fix
description	Run a Python project's tests, diagnose failures, fix, and re-verify
version	1.0
metadata	{"pulsar":{"category":"coding","tags":["python","pytest","testing"]}}

Python: Test And Fix

Procedure

1. Detect the test runner: pyproject.toml with [tool.pytest.ini_options] or a tests/ directory means pytest; otherwise try python -m unittest discover.
2. Run the suite once with terminal (e.g. python -m pytest) before changing anything, to get a clean baseline of failures.
3. For each failure, read the failing test and the code under test with read_file before editing. Prefer patch for minimal, targeted fixes.
4. Fix one failure cluster at a time; re-run only the affected test file (python -m pytest tests/test_x.py -q) to iterate quickly.
5. When individual fixes pass, run the full suite again to catch regressions.

Pitfalls

• Do not "fix" a test by weakening its assertions unless the test itself is demonstrably wrong; state your reasoning when you do.
• Import errors often mean a missing dev dependency, not broken code; check pyproject.toml extras before editing sources.
• Beware of tests that depend on state (env vars, cwd, network); they can fail for environmental reasons unrelated to the change.

Verification

• The full test suite exits with code 0.
• No previously passing test now fails.
• Summarize which tests failed initially, what changed, and the final counts.