tdd

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Test-Driven Development Skill

测试驱动开发技能

Overview

概述

Guides Test-Driven Development workflow: Red-Green-Refactor cycle for new features, bug fixes, legacy code, and test refactoring.

指导测试驱动开发工作流：针对新功能、Bug修复、遗留代码和测试重构的红-绿-重构循环。

Development Scenarios

开发场景

1. New Feature Development

1. 新功能开发

Follow the Red-Green-Refactor cycle:

遵循红-绿-重构循环：

🔴 RED - Write a Failing Test

🔴 红阶段 - 编写失败的测试

Write the smallest test defining desired behavior
Run test to confirm failure

bash

uv run pytest tests/feature/test_new_function.py -v

编写定义期望行为的最小测试用例
运行测试确认失败

bash

uv run pytest tests/feature/test_new_function.py -v

🟢 GREEN - Make the Test Pass

🟢 绿阶段 - 让测试通过

Write minimal production code to pass
Run test to confirm passes
Iterate until all tests pass

bash

uv run pytest tests/feature/test_new_function.py::test_new_function -v
uv run pytest

编写最少的生产代码使测试通过
运行测试确认通过
迭代直到所有测试通过

bash

uv run pytest tests/feature/test_new_function.py::test_new_function -v
uv run pytest

🔵 REFACTOR - Improve the Code

🔵 重构阶段 - 优化代码

Clean up test and production code
Ensure tests still pass
List refactoring opportunities (see Test Refactoring Opportunities)

bash

uv run pytest --cov=src --cov-report=term-missing

清理测试代码和生产代码
确保测试仍能通过
列出重构优化点（参见测试重构优化点）

bash

uv run pytest --cov=src --cov-report=term-missing

2. Bug Fixes

2. Bug修复

Never fix a bug without writing a test first.

Write test reproducing bug, verify fails
Fix bug in implementation
Run tests until green
Check refactoring opportunities

python

def test_calculate_total_with_negative_price_raises_error():
    items = [Item("Coke", -1.50)]
    with pytest.raises(ValueError, match="Price cannot be negative"):
        calculate_total(items)

修复Bug前务必先编写测试。

编写复现Bug的测试用例，验证测试失败
在实现代码中修复Bug
运行测试直到全部通过
检查重构优化点

python

def test_calculate_total_with_negative_price_raises_error():
    items = [Item("Coke", -1.50)]
    with pytest.raises(ValueError, match="Price cannot be negative"):
        calculate_total(items)

3. Legacy Code

3. 遗留代码

Write characterization tests capturing current behavior
Run tests to establish baseline
Make small changes with test coverage
Refactor incrementally

bash

uv run pytest tests/legacy/test_old_module.py -v
uv run pytest --cov=src/legacy_module --cov-report=term-missing

编写特征化测试捕获当前代码行为
运行测试建立基准
在测试覆盖下进行小幅度修改
逐步重构

bash

uv run pytest tests/legacy/test_old_module.py -v
uv run pytest --cov=src/legacy_module --cov-report=term-missing

Test Refactoring Opportunities

测试重构优化点

After each TDD cycle (or when reviewing existing tests), search for these improvement opportunities:

在每个TDD循环结束后（或评审现有测试时），寻找以下改进机会：

Identify Opportunities

识别优化点

Check for:

Redundant tests: Similar test logic that can be consolidated with parametrization
Duplicate fixtures: Common test data defined in multiple files
Missing parametrization: Multiple tests for similar scenarios
File organization: Tests that could be better consolidated
Best practice violations: Testing private methods, vague assertions, etc.
Slow tests: Tests that could be optimized or marked with
```
@pytest.mark.slow
```
Hardcoded values: Test data that should use fixtures or factories

检查以下内容：

冗余测试：可通过参数化合并的相似测试逻辑
重复夹具：在多个文件中定义的通用测试数据
缺少参数化：针对相似场景的多个重复测试
文件组织：可优化合并的测试文件
违反最佳实践：测试私有方法、模糊断言等
慢测试：可优化或标记为
```
@pytest.mark.slow
```
的测试
硬编码值：应使用夹具或工厂的测试数据

Refactoring Process

重构流程

When refactoring existing tests:

Capture baseline metrics:

bash

bin/ci-local 2>&1 | tee baseline.txt
grep "passed" baseline.txt  # Note test count
grep "Cover" baseline.txt   # Note coverage %

Prioritize changes by impact and isolation:
- High impact, isolated changes first (parametrization, shared fixtures)
- Medium impact changes (consolidating test files)
- Complex changes last (large file reorganization)
Apply refactoring patterns (see patterns.md)

Verify coverage maintained or improved:

bash

uv run pytest --cov=src --cov-report=term-missing

重构现有测试时：

捕获基准指标：

bash

bin/ci-local 2>&1 | tee baseline.txt
grep "passed" baseline.txt  # 记录测试数量
grep "Cover" baseline.txt   # 记录覆盖率百分比

按影响和独立性优先处理变更：
- 先处理高影响、独立的变更（参数化、共享夹具）
- 再处理中等影响的变更（合并测试文件）
- 最后处理复杂变更（大规模文件重组）
应用重构模式（参见patterns.md）

验证覆盖率维持或提升：

bash

uv run pytest --cov=src --cov-report=term-missing

Verification Checklist

验证清单

After completing TDD cycles or test refactoring, verify:

All tests pass:
```
uv run pytest
```

Coverage maintained or improved:

uv run pytest --cov=src --cov-report=term-missing

No mypy errors:
```
uv run mypy src/
```
No ruff errors:
```
uv run ruff check src/
```
Tests execute faster:
```
uv run pytest --durations=10
```

For complete verification, run the full CI pipeline:

bash

bin/ci-local

完成TDD循环或测试重构后，验证以下内容：

所有测试通过：
```
uv run pytest
```

覆盖率维持或提升：

uv run pytest --cov=src --cov-report=term-missing

无mypy错误：
```
uv run mypy src/
```
无ruff错误：
```
uv run ruff check src/
```
测试执行更快：
```
uv run pytest --durations=10
```

如需完整验证，运行完整CI流水线：

bash

bin/ci-local

Best Practices

最佳实践

See best-practices.md for test writing, organization, and code quality guidelines.

参见best-practices.md获取测试编写、组织和代码质量指南。

Verification Commands

验证命令

bash

undefined

bash

undefined

Run all tests

运行所有测试

uv run pytest

Run specific test file

运行指定测试文件

uv run pytest tests/models/test_registry.py -v

Run specific test

运行指定测试用例

uv run pytest tests/test_file.py::TestClass::test_function -vv

Run with coverage

带覆盖率运行测试

uv run pytest --cov=src --cov-report=term-missing --cov-report=html

Run full CI pipeline

运行完整CI流水线

bin/ci-local

Check test execution time

检查测试执行时间

uv run pytest --durations=10

Run fast tests only

仅运行快速测试

uv run pytest -m "not slow"

undefined

uv run pytest -m "not slow"

undefined

Important Notes

重要注意事项

Test First: Always write tests before implementation
Verify Red Phase: Confirm tests fail before writing implementation
One at a Time: Focus on one failing test before moving to the next
Maintain Coverage: Coverage must never decrease during refactoring
Small Changes: Make incremental changes and run tests frequently
List Refactoring Opportunities: After green, identify but wait for user request before implementing

先写测试：始终在实现代码前编写测试
验证红阶段：在编写实现代码前确认测试失败
一次处理一个：专注于一个失败的测试，再进行下一个
维持覆盖率：重构期间覆盖率绝不能下降
小步变更：进行增量变更并频繁运行测试
列出重构优化点：绿阶段结束后识别优化点，但需等待用户请求再实施