test-execution
暂无简介
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1.6k
- Forks
- 491
03 SKILL.md 正文
测试执行方法
强制要求:真实执行测试
你必须真正执行测试,禁止生成 Mock 数据!
测试执行流程
- 检测项目类型和测试框架
- 执行单元测试(含覆盖率)
- 执行集成测试
- 执行接口测试
- 执行 E2E 测试
- 解析测试输出(总数、通过数、失败数、覆盖率)
- 失败分级与处理
测试框架检测
JavaScript/TypeScript
| 框架 | 检测标志 | 配置文件 |
|---|---|---|
| Jest | "jest" in devDependencies |
jest.config.js / jest.config.ts |
| Vitest | "vitest" in devDependencies |
vitest.config.ts |
| Playwright | "@playwright/test" in devDependencies |
playwright.config.ts |
| Cypress | "cypress" in devDependencies |
cypress.config.ts |
Python
| 框架 | 检测标志 | 配置文件 |
|---|---|---|
| pytest | "pytest" in dependencies |
pytest.ini / pyproject.toml |
| unittest | 内置 | - |
Go
*_test.go文件存在 → 使用go test- 集成测试通常用 build tag 隔离:
//go:build integration
Java
| 框架 | 检测标志 |
|---|---|
| JUnit 5 | pom.xml 含 junit-jupiter |
| TestNG | pom.xml 含 testng |
测试命令矩阵
| 语言 | 单元测试 | 集成测试 | 接口测试 | E2E 测试 |
|---|---|---|---|---|
| Node.js | npx vitest run / npx jest |
npx vitest run tests/integration |
npx vitest run tests/api |
npx playwright test |
| Python | pytest tests/unit |
pytest tests/integration |
pytest tests/api |
pytest tests/e2e |
| Go | go test ./internal/... |
go test -tags=integration ./... |
go test ./tests/api/... |
- |
| Java | mvn test |
mvn verify |
mvn verify -Papi-test |
- |
接口测试方法论见
Skill(skill: "qa/api-testing"),集成测试方法论见Skill(skill: "qa/integration-testing")。
覆盖率度量与门禁
覆盖率采集
# Vitest
npx vitest run --coverage --reporter=json
# 输出 coverage/coverage-summary.json
# Jest
npx jest --coverage --coverageReporters=json-summary
# 输出 coverage/coverage-summary.json
# Python (pytest-cov)
pytest --cov=src --cov-report=json
# 输出 coverage.json
# Go
go test -coverprofile=coverage.out ./...
go tool cover -func=coverage.out
覆盖率报告解析
coverage-summary.json 结构:
{
"total": {
"lines": { "total": 100, "covered": 85, "pct": 85 },
"statements": { "total": 120, "covered": 100, "pct": 83.3 },
"functions": { "total": 40, "covered": 36, "pct": 90 },
"branches": { "total": 30, "covered": 22, "pct": 73.3 }
}
}
覆盖率门禁
| 指标 | 门槛 | 未达标处理 |
|---|---|---|
| 行覆盖率 | ≥ 80% | 列出未覆盖行,报 REVISION_NEEDED |
| 语句覆盖率 | ≥ 80% | 同上 |
| 函数覆盖率 | ≥ 80% | 同上 |
| 分支覆盖率 | ≥ 75% | 列出未覆盖分支 |
| 变更文件覆盖率 | ≥ 80% | 重点检查本次变更文件 |
覆盖率检查脚本(Node.js):
# 检查覆盖率是否达标
node -e "
const cov = require('./coverage/coverage-summary.json').total;
const thresholds = { lines: 80, statements: 80, functions: 80, branches: 75 };
let failed = false;
for (const [key, min] of Object.entries(thresholds)) {
const pct = cov[key].pct;
if (pct < min) {
console.log(\`❌ \${key}: \${pct}% < \${min}%\`);
failed = true;
} else {
console.log(\`✅ \${key}: \${pct}%\`);
}
}
process.exit(failed ? 1 : 0);
"
测试结果解析
Jest / Vitest JSON 输出
npx jest --json --outputFile=test-results.json
# 或
npx vitest run --reporter=json --outputFile=test-results.json
解析关键字段:
{
"numTotalTests": 50,
"numPassedTests": 48,
"numFailedTests": 2,
"numPendingTests": 0,
"success": false,
"testResults": [
{
"name": "tests/api/users.test.ts",
"status": "failed",
"assertionResults": [
{
"title": "creates user with valid data",
"status": "passed",
"duration": 15
},
{
"title": "rejects empty name with 400",
"status": "failed",
"failureMessages": ["Expected 400, received 201"]
}
]
}
]
}
pytest JSON 输出
pytest tests/ -v --json-report --json-report-file=test-results.json
解析:
{
"summary": {
"total": 30,
"passed": 28,
"failed": 2,
"skipped": 0,
"duration": 2.5
}
}
Go test 输出
go test ./... -json > test-results.json
解析(每行一个 JSON):
# 统计通过/失败
go test ./... -v 2>&1 | grep -E "^(=== RUN|--- (PASS|FAIL|SKIP))" | \
awk '/PASS/{p++} /FAIL/{f++} /SKIP/{s++} END{print "passed:"p" failed:"f" skipped:"s}'
Playwright JSON 输出
| 字段 | 说明 |
|---|---|
stats.expected |
通过的测试数 |
stats.unexpected |
失败的测试数 |
stats.flaky |
重试后通过的测试数 |
stats.skipped |
跳过的测试数 |
失败分级与处理
失败分级
| 级别 | 判定标准 | 处理方式 |
|---|---|---|
| Blocker | 核心功能不可用、数据丢失、安全漏洞 | 立即修复,阻止发布 |
| Critical | 主要功能异常、数据不一致 | 必须修复,阻塞发布 |
| Major | 次要功能异常、有 workaround | 修复后发布 |
| Minor | UI 展示问题、文案错误 | 可带 issue 发布 |
失败处理流程
测试失败
│
├─ 是环境问题吗?(DB 连接、端口占用、依赖未启动)
│ └─ 是 → 修复环境,重跑
│
├─ 是测试本身问题吗?(断言错误、测试数据过期、时序问题)
│ └─ 是 → 修复测试,重跑
│
├─ 是产品代码 bug 吗?
│ └─ 是 → 报 REVISION_NEEDED,交给开发修复
│
└─ 是 flaky 吗?(间歇性失败)
└─ 是 → 重试 1 次,仍失败则标记 flaky 并排查根因
禁止行为
- ❌ 禁止为通过测试而修改断言或删除用例
- ❌ 禁止用
.skip/xit跳过失败测试 - ❌ 禁止将失败测试标记为"预期失败"而不修复
- ❌ 禁止在报告中隐瞒失败
增量测试选择
为了加速反馈,只运行受变更影响的测试:
基于变更文件的测试选择
# 获取变更的源文件
git diff --name-only main...HEAD -- 'src/**/*.ts'
# Vitest: 只运行受影响的测试(需配置 coverage analysis)
npx vitest run --changed
# Jest: 只运行变更相关测试
npx jest --changedSince=main
测试目录约定
tests/
├── unit/ # 单元测试,快速,CI 必跑
├── integration/ # 集成测试,较慢,CI 必跑
├── api/ # 接口测试,CI 必跑
└── e2e/ # E2E 测试,最慢,PR 必跑
按标签选择性执行
// Vitest
import { describe, it } from 'vitest';
describe('@critical 用户认证', () => {
it('@smoke 登录成功', async () => {
/* ... */
});
it('登录失败', async () => {
/* ... */
});
});
# 只跑 critical 标签
npx vitest run -t @critical
# Playwright
npx playwright test --grep @critical
并行执行策略
Jest / Vitest 并行
// vitest.config.ts
export default defineConfig({
test: {
// 文件间并行(默认)
// 文件内串行(避免共享状态冲突)
pool: 'forks',
poolOptions: {
forks: {
singleFork: false, // 每个文件一个 fork
},
},
},
});
集成测试串行执行
集成测试通常操作数据库,需要串行避免冲突:
# Jest: 串行执行集成测试
npx jest tests/integration --runInBand
# Vitest: 串行
npx vitest run tests/integration --pool=forks --poolOptions.forks.singleFork=true
Go 并行控制
# 默认并行(CPU 核数)
go test ./...
# 限制并行数
go test -p 2 ./...
# 串行
go test -p 1 ./...
pytest 并行
# 使用 pytest-xdist 并行
pytest -n auto # 自动检测 CPU 核数
pytest -n 4 # 4 个 worker
pytest tests/integration -n 0 # 串行
接口测试执行
接口测试通常使用 Supertest(Node.js)或 requests(Python)直接调用应用,无需启动独立服务器。
# Node.js (Jest + Supertest)
npx jest tests/api --runInBand
# Python (pytest + requests)
pytest tests/api -v
# 输出覆盖率
npx jest tests/api --coverage
集成测试执行
集成测试需要数据库等依赖,使用 Testcontainers 或测试数据库:
# Node.js (需要 Docker 运行 Testcontainers)
npx jest tests/integration --runInBand
# Python
pytest tests/integration -v
# Go
go test ./tests/integration/... -v
集成测试通常
--runInBand串行执行,避免数据库并发冲突。
Playwright E2E 执行细节
完整方法论:详见
Skill(skill: "qa/e2e-playwright")
执行命令
# 安装浏览器(首次或 CI 环境)
npx playwright install --with-deps
# 运行全部 E2E 测试
npx playwright test
# 仅 critical 标签(门禁加速)
npx playwright test --grep @critical
# 指定浏览器
npx playwright test --project=chromium
# JSON 报告(门禁解析用)
npx playwright test --reporter=json
失败排查
# 查看 trace(失败时自动生成)
npx playwright show-trace <trace.zip路径>
# 打开 HTML 报告
npx playwright show-report
测试执行检查清单
- [ ] 已检测项目使用的测试框架
- [ ] 单元测试已执行并通过
- [ ] 集成测试已执行并通过
- [ ] 接口测试已执行并通过
- [ ] E2E 测试已执行并通过
- [ ] 覆盖率已采集并达标(≥ 80%)
- [ ] 所有失败用例已分类和处理
- [ ] 测试结果有完整记录(命令、退出码、通过/失败数、失败详情)
- [ ] 无 flaky 测试未处理
04 同仓库更多 Skills
worktable-review-with-annotations
Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.
worktable-find-and-synthesize-context
Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.
worktable-create-or-update-docs
Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.
worktable-create-or-manage-records
Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.
worktable-create-or-manage-interactive-html
Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.
worktable-collaborate-threads
Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.