请问HN：现代测试自动化软件（Python/Go/TS）？
嗨，HN，
我正在寻找一些现代的测试自动化软件/框架，这些工具能够很好地与Python/Go/TS配合使用。在我的搜索中没有找到合适的选项，我也不想自己搭建测试自动化基础设施。
我的使用场景是在实验室环境中进行硬件和固件测试，我希望避免被迫进入特定供应商的硬件生态系统。
我所寻找的特性包括：
- Python/Go/TS兼容性（SDK、API或一流支持）
- 能够查看和查询历史测试运行（仪表板/日志和指标的趋势视图）
- 能够定义自定义测试序列/工作流，并能够并发运行步骤（不仅仅是一个平面的测试列表）。例如：条件步骤、重试、设置/拆卸阶段、多设备编排（电源、数字万用表、数据采集器和被测设备）
- 硬件无关性/无供应商锁定：我应该能够更换仪器/设备，而无需重写所有内容或被绑定到某个专有供应商（特别是NI）
- 理想情况下：还应该有Slack集成，用于启动测试和通知测试完成
问题：
1. 自己搭建软件架构是唯一的选择吗？如果你们构建过类似的东西，哪个技术栈效果最好（例如，Robot Framework、pytest + 插件、自定义调度器、Airflow/Prefect/Temporal等）？
2. 是否有专门构建的平台可以推荐，并且没有供应商锁定？
3. 你们使用什么工具来管理运行历史和报告？
4. 在可靠性、扩展到多个设备或维护驱动层方面，有没有什么“陷阱”？
如果需要，我很乐意自己组装序列逻辑，但我希望避免从头开始重新发明编排和运行历史/报告的过程。
查看原文
Hi HN,<p>I’m looking for recommendations for modern test automation software/frameworks that work well with Python/Go/TS. I wasn’t able to find any from my search and I don’t want to spin up my own test automation infrastructure.<p>My use case is hardware + firmware testing in a lab environment, where I want to avoid being forced into a specific vendor’s hardware ecosystem.<p>What I’m looking for:<p>- Python/Go/TS compatibility (SDK, API, or first-class support)<p>- Ability to see and query historical test runs (dashboards / trend views of logs and metrics)<p>- Ability to define custom test sequences/workflows with the ability to run steps concurrently (not just a flat list of tests). Examples: conditional steps, retries, setup/teardown phases, multi-device orchestration (PSUs, DMMs, DAQs, and DUTs)<p>- Hardware-agnostic / no vendor lock-in: I should be able to swap instruments/devices without rewriting everything or being tied to a proprietary vendor (looking at you NI)<p>- Ideally: also have a slack integration for initiating runs & notifications of run completions<p>Questions:<p>1. Is spinning up my own software architecture the only option? And If you’ve built something like this, what stack worked best (e.g., Robot Framework, pytest + plugins, custom orchestrator, Airflow/Prefect/Temporal, etc.)?<p>2. Are there purpose-built platforms you’d actually recommend that don’t vendor lock?<p>3. What do you use for run history + reporting?<p>4. Any “gotchas” with reliability, scaling to many devices, or maintaining driver layers?<p>I’m happy to assemble the sequence logic myself if needed, but I’d love to avoid reinventing orchestration and run history/reporting from scratch.