E4b9a6's blog
有善始者实繁，能克终者盖寡。
作者：E4b9a6， 创建：2023-09-12， 字数：4294， 已阅：68， 最后更新：2023-09-12
这篇文章更新于 436 天前，文中部分信息可能失效，请自行甄别无效内容。
playwright是微软开源的自动化web应用程序测试框架，支持主流的Python、NodeJS、Java等语言，官方文档已经非常完善了，本文将快速分享一些生产开发中常见的用法
接下来使用的Python版本是3.8.6
Windows7最高支持到3.8.10版本，考虑到兼容性则尽量使用3.8.10以下版本
使用pip安装playwright如下
pip3 install playwright
playwright install
例子一：使用传统的同步写法
from playwright.sync_api import sync_playwright
with sync_playwright() as p:
browser = p.chromium.launch()
page = browser.new_page()
page.goto("http://playwright.dev")
print(page.title())
browser.close()
例子二：使用async/await异步语法
import asyncio
from playwright.async_api import async_playwright
async def main():
async with async_playwright() as p:
browser = await p.chromium.launch()
page = await browser.new_page()
await page.goto("http://playwright.dev")
print(await page.title())
await browser.close()
asyncio.run(main())
使用async能简单有效的提高并发，playwright在大部分场景下都属于网络IO密集型，但使用async写法对于开发人员的要求更高，同时debug也不如传统同步方法来的方便
是否使用异步写法取决于你的程序框架和开发需求
文档参考：https://playwright.dev/python/docs/api/class-locator
自动化测试非常需要的就是元素获取、元素点击、元素输入，下面以在bing.com搜索chancel.me并输出第一条搜索结果为例
from playwright.sync_api import sync_playwright
with sync_playwright() as p:
b = p.chromium.launch()
page = b.new_page()
page.goto('https://bing.com')
page.locator('#sb_form_q').fill('chancel.me') # fill方法填充字符串
page.locator('#search_icon').click() # id选择并点击
print(page.locator('[class=b_algo]').nth(0).text_content()) # class选择后获取文本
输出结果
syncmemo - Chancel's bloghttps://memo.chancel.me/help网页使用帮助 请记住本站网址 https://memo.chancel.me 使用步骤 访问本站时，会自动分配一个随机数（类似于97ND），请稍微花几秒钟记住这个ID，点击确认开始编辑便签内容
文档参考：https://playwright.dev/python/docs/api/class-browsercontext
自动化通常会涉及到cookies保存，playwright获取Cookies需要创建新的context
以下是cookies的get/set操作例子
from playwright.sync_api import sync_playwright
cookies = None
with sync_playwright() as p:
b = p.chromium.launch()
c = b.new_context()
page = c.new_page()
page.goto('https://www.chancel.me/')
cookies = c.cookies()
print(cookies)
with sync_playwright() as p:
b = p.chromium.launch()
c = b.new_context()
page = c.new_page()
c.add_cookies(cookies)
page.goto('https://www.chancel.me/')
cookies = c.cookies()
print(cookies)
输出结果
[{'name': 'TOKEN', 'value': 'c263e3a6-c88d-11ed-8979-6ee82343e005', 'domain': 'www.chancel.me', 'path': '/', 'expires': 1684658669.818441, 'httpOnly': True, 'secure': True, 'sameSite': 'Lax'}]
参考文档：https://playwright.dev/python/docs/api/class-download
以下载Github二进制包为例子
from playwright.sync_api import sync_playwright
cookies = None
with sync_playwright() as p:
b = p.chromium.launch()
page = b.new_page()
page.goto('https://github.com/chancelyg/godaddy-ddns/releases')
with page.expect_download() as download_info:
page.locator('[href="/chancelyg/godaddy-ddns/releases/download/v23.01.06/godaddy-ddns_Darwin_arm64.tar.gz"]').click()
download_info.value.save_as('godaddy-ddns_Darwin_arm64.tar.gz')
参考文档：https://playwright.dev/python/docs/api/class-video
在playwright开发时需要经常截图确认一下自动化流程是否正常，截图如下
from playwright.sync_api import sync_playwright
# 截图
with sync_playwright() as p:
b = p.chromium.launch()
page = b.new_page()
page.goto('https://www.chancel.me')
page.screenshot(path='screenshot.png')
# 录像
with sync_playwright() as p:
b = p.chromium.launch()
page = b.new_page(record_video_dir="videos", record_video_size={"width": 640, "height": 480})
page.goto('https://www.google.com')
page.goto('https://www.chancel.me')
参考文档：https://playwright.dev/docs/docker#image-tags
playwright有官方镜像，以mcr.microsoft.com/playwright/python:v1.31.0-focal为例，跑一个uvicorn网关应用的Dockerfile文件如下
FROM mcr.microsoft.com/playwright/python:v1.31.0-focal
ENV LANG en_US.UTF-8
WORKDIR /app
COPY ./web /app/web
COPY ./requirements.txt /app/requirements.txt
RUN pip3 install -r /app/requirements.txt && \
playwright install && \
ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
CMD ["uvicorn", "web:app", "--host=0.0.0.0", "--port=5000"]