Pytest Coverage Skill

Checklist before writing tests

• Map critical modules: auth/session, todo CRUD/long-term goals, scheduler jobs, URL analyzers/blocker, notifications, CLI entrypoints. Note DB/network touchpoints to mock.
• Confirm coverage config or add it (see Config). Always exclude src/safe_family/templates/miscellaneous/.
• Identify fast, deterministic seams (pure functions like generate_time_slots, infer_tag, get_time_range, etc.) for unit coverage; avoid real DB/network/FS.

Config

• Prefer pyproject with coverage + pytest options:

  [tool.coverage.run]
  branch = true
  source = ["src"]
  omit = ["src/safe_family/templates/miscellaneous/*"]
  
  [tool.coverage.report]
  show_missing = true
  skip_covered = true
  fail_under = 80
  
  [tool.pytest.ini_options]
  addopts = "-q --disable-warnings --cov --cov-report=term-missing"
  

• If pyproject cannot be changed, pass flags: python -m pytest --cov=src --cov-report=term-missing --cov-fail-under=80 --cov-config=pyproject.toml.

Test-writing patterns

• Pure helpers: Assert exact outputs and boundary errors; freeze now via parameters or monkeypatch.
• DB-backed functions: Prefer patching get_db_connection/SQLAlchemy with fakes; avoid hitting real DB. Validate SQL parameters via mocks/spies.
• Scheduler/commands: Patch APScheduler, subprocess, requests, and timeouts. Assert functions called with expected args; do not perform network/router calls.
• Blueprint routes: Use Flask test client and app.test_client(); seed JWT/session via fixtures; patch external services (mail, requests.post).
• Notifications: Mock requests.post and mail.send; assert payload structure, no real HTTP.
• Long-term goals/time tracking: Use fixed datetimes; patch local_tz or inject today.

Workflow

1. Select targets that maximize uncovered lines while avoiding excluded templates. Start with pure utilities, then branchy routes/CLI, then scheduler logic.
2. Add/extend fixtures in tests/conftest.py for app factory, test client, and mock DB/requests/mail.
3. Write small, focused tests (arrange/act/assert) with descriptive names; cover success and error branches.
4. Run python -m pytest; iterate until coverage ≥ 80% and no flakiness. Use --maxfail=1 during iteration for speed.

CI

• Ensure workflow installs deps and runs python -m pytest with coverage flags or pyproject config. Keep runtime minimal (no services). Use PYTHONPATH=${{ github.workspace }}.