pytest-attempt-summary 0.1.2

Enhanced Allure Attempt Summary for Playwright + Pytest

pytest-attempt-summary

中文	English
让 pytest 的重试过程在 Allure 报告中 看得见、讲得清、用得上	Make pytest retries visible, explainable, and actionable in Allure reports.
当测试启用 reruns 后，失败往往变得难以理解。Allure 只展示最终结果，却隐藏了每一次 Attempt 之间真正发生了什么。	When tests are retried, failures become harder to understand. Allure shows the final result — but hides what really happened between attempts.
pytest-attempt-summary 为 Allure 报告添加 Attempt Summary，让重试行为变得清晰、结构化、可分析。	pytest-attempt-summary adds an Attempt Summary section to Allure reports, making retry behavior clear, structured, and debuggable.

---

安装与快速开始 / Installation & Quick Start

中文	English
安装插件	Install the plugin
启用重试（示例）	Enable retries (example)
运行测试并生成 Allure 报告	Run tests and generate Allure report
在 Allure 中打开 失败用例 → Attempt Summary 会自动显示在 测试正文 中	Open a failed test case in Allure → Attempt Summary appears automatically in the Test Body

示例命令 / Example Commands:

pip install pytest-attempt-summary

# pytest.ini
[pytest]
reruns = 2

pytest --alluredir=allure-results
allure serve allure-results

---

为什么 Attempt Summary 很重要 / Why Attempt Summary Matters

中文	English
测试用例在重试后失败	A test fails after retries
你只能看到失败而已	That’s all you know
它每次都失败吗？错误在重试间是否变化？是不稳定用例还是确定性失败？截图、视频和 trace 在哪？	Did it fail every time? Did the error change between retries? Was it flaky or deterministic? Where are the screenshots, videos, and traces?

Without Attempt Summary / 没有 Attempt Summary:

FAILED

With pytest-attempt-summary / 使用 pytest-attempt-summary:

🔁 Attempts: 3 / 3 failed

🧠 Retry Insight
• Flaky behavior detected (error changed between attempts)

🔍 Attempt Diff Analysis
≠ Error Differences
≠ Duration Differences

Attempt 3 ❌ FAILED
🕑 Duration: 0.01s
▶ View Failure Details

---

中文	English
你可以立即看到	You instantly understand
❗ 测试失败了多次	❗ The test failed multiple times
🔁 错误在重试间发生变化	🔁 Errors changed between retries
🧩 这很可能是不稳定用例	🧩 This is likely a flaky test
🎯 明确下一步调查方向（截图、视频、trace）	🎯 Exactly where to investigate next (screenshots, video, trace)
重试不会掩盖问题，它们让问题变得可理解	Retries didn’t hide the problem. They explained it.

---

核心一句话 / One-line Summary

中文	English
Attempt Summary 帮助你理解 为什么 重试失败，而不仅仅是 失败了。	Attempt Summary helps you understand why retries fail — not just that they failed.

---

功能展示 / What You Get

失败用例在 Allure 中会附带 Attempt Summary：

🔁 Attempts: 3 / 3 failed

🧠 Retry Insight
• Flaky behavior detected (error changed between attempts)

🔍 Attempt Diff Analysis
≠ Error Differences
≠ Duration Differences

Attempt 1 | Attempt 2 | Attempt 3 - Latest
Attempt 3 ❌ FAILED
🕑 Duration: 0.01s
▼ View Failure Details (Attempt 3)
├─ Page URL
├─ Browser Console Errors
├─ Test Failure Errors
├─ Screenshot
├─ Video (see Tear down)
└─ Trace (see Tear down)

---

核心特性 / Key Features

中文	English
收集每个测试用例的所有重跑尝试	Collects all rerun attempts per test case
支持 pytest-rerunfailures	Supports pytest-rerunfailures
可交互展开 / 收起失败详情	Interactive expand / collapse Failure Details
高亮显示尝试之间的错误 / 耗时差异	Highlights error / duration differences between attempts
集成 Playwright 视频和 trace	Integrates with Playwright video and trace
轻量 HTML 附件，不让 Allure 报告臃肿	Lightweight HTML attachment (does not bloat Allure pages)

---

注意事项 / Important Notes

中文	English
视频和 Trace 文件由 UI 项目生成，可在 Allure → Tear down → context 查看	Video and Trace files are attached by your UI framework and are visible under Allure → Tear down → context
Attempt Summary 不直接嵌入视频或 trace，仅指引用户到哪里查看	Attempt Summary does not embed videos or traces, it only guides you where to find them
插件不会自动记录 attempt，需要 UI 项目自己收集每次尝试的数据	This plugin does not record attempts by itself, your UI project must collect attempt data (status, error, duration, artifacts)

---

预期 Attempt 数据格式 / Expected Attempt Data Format

[
  {
    "attempt": 1,
    "status": "FAILED",
    "duration": 0.5,
    "error": "AssertionError: ...",
    "url": "https://example.com",
    "has_video": true,
    "has_trace": true
  },
  {
    "attempt": 2,
    "status": "PASSED",
    "duration": 0.3
  }
]

---

适用场景 / Designed For

中文	English
Playwright + Pytest UI 自动化框架	Playwright + Pytest UI automation frameworks
使用 reruns 分析不稳定用例的团队	Teams using reruns to analyze flaky tests
关注重试洞察的工程师，而不仅仅是成功/失败	Engineers who want retry insight, not just pass/fail

---

许可 / License

MIT License