인터페이스는 현지화되어 있으며, Skill 소개와 설명 문서는 선택한 언어로 AI가 자동 번역합니다. 코드·명령어·경로·전문 용어는 원문을 유지하며 원작자 문서가 우선합니다.
Skills 둘러보기
개발 및 엔지니어링 · SKILL.md

django-targeted-mocking

django-targeted-mocking

Mock Django test boundaries safely with override_settings, pytest settings, autospecced unittest.mock objects, output/input capture, requests-mock, VCR.py, outbound HTTP blocking, and time-freezing tools. Use when tests need to replace settings, HTTP calls, time, command output, stdin, or external services without broad fragile MagicMock patches.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill 문서

GitHub

name: django-targeted-mocking description: Mock Django test boundaries safely with override_settings, pytest settings, autospecced unittest.mock objects, output/input capture, requests-mock, VCR.py, outbound HTTP blocking, and time-freezing tools. Use when tests need to replace settings, HTTP calls, time, command output, stdin, or external services without broad fragile MagicMock patches. license: MIT compatibility: Codex, Claude Code, and other Agent Skills-compatible clients. metadata: version: "0.1.0" displayName: Django Targeted Mocking category: Django tags: django,testing,mocking,http,pytest

Django Targeted Mocking

Mock only at boundaries that are slow, nondeterministic, external, or hard to trigger. Broad mocks can make tests faster while proving less, so prefer framework helpers and interface-checked replacements.

Boundary Workflow

  1. Decide whether mocking is necessary.

    • Prefer real code for local domain logic.
    • Mock external HTTP, time, settings, output/input, and expensive service edges.
  2. Patch where the code under test looks up the dependency.

    • Patch the imported symbol used by the module under test, not the original library path unless that is what the module reads.
  3. Prefer specific tools.

    • Settings: override_settings, modify_settings, or pytest-django's settings fixture.
    • Output: io.StringIO, pytest capsys, or command test helpers.
    • HTTP: requests-mock for hand-authored responses; VCR.py for recorded responses.
    • Time: pass time as a parameter when possible; otherwise use time-machine or a similar time-specific tool.
  4. Make mocks interface-aware.

    • Use autospec=True, spec, spec_set, create_autospec, or a small fake object.
    • Avoid bare Mock/MagicMock when the interface matters.
  5. Block unexpected outbound HTTP.

    • A normal test run should fail if a test reaches the network unintentionally.

Read mocking-patterns.md for concrete patterns.

Decision Rules

  • Use Django settings helpers instead of direct assignment to django.conf.settings.
  • Prefer a hand-built fake or types.SimpleNamespace when only a few attributes are needed.
  • Use requests-mock for deterministic API responses you control.
  • Use VCR.py when real recorded responses are valuable and safe to store.
  • On CI, configure VCR record mode so missing cassettes fail instead of recording.
  • Mock time consistently across the runtime; do not patch several datetime functions by hand.
  • Do not mock the component whose behavior the test is meant to prove.

Common Mistakes

  • Patching settings directly and leaking state across tests.
  • Using catch-all MagicMock objects that accept typoed attributes and wrong call signatures.
  • Allowing tests to hit real external services.
  • Recording VCR cassettes in CI.
  • Mocking a view's internal helpers so thoroughly that the test no longer covers request behavior.
  • Patching the wrong import path and thinking the mock is active.

Verification

Before finishing:

  • Each mock has a clear boundary reason.
  • Interface-sensitive mocks use specs or fakes.
  • Unexpected outbound HTTP is blocked in the normal suite.
  • Settings are restored automatically by Django or pytest helpers.
  • Tests still cover the behavior, not only mock calls.
저장한 Skills