Алексей Консманов — Как особенности Python и Pytest могут поломать backend-автотесты

Learn more about the Heisenbug conference: https://jrg.su/xC2gbh — — Download the presentation from the website — https://jrg.su/PRLrcg This talk will focus on identifying, minimizing, and eliminating the impact of testing frameworks and language features of Python & Pytest on backend flaky tests. The goal is to systematize and demonstrate potential risks and how to avoid them before running tests. Examples will be provided with Python 3 and pytest. Automated test code contains nuances and hidden issues that can break tests. We'll discuss best practices for avoiding such issues. We'll examine which issues can break a single test, and which can break the entire test session and environment. We'll also explore how to find such tests and what best practices exist for writing flaky-free code. First, we'll discuss what flaky is and immediately limit the topic to flaky issues caused by language and framework features. Next, we'll look at specific code cases: why this code seems "okay," what the problem is, and how to avoid it. Finally, you'll receive a checklist to help you avoid the aforementioned problems. Target audience: Python AQA students of all levels, Python developers, Python SDETs, and DevOps. Jinor and mid-level engineers will learn new things, while senior specialists will be able to systematize their knowledge. The material will be in-depth, covering practical examples of bugs and fixes using Python and pytest. We'll also explore the specifics of the Python and pytest ecosystem.