微软开源最强Python自动化神器Playwright！不用写一行代码！自动生成代码还竟然如此流畅！妈妈再也不用担心我不会写代码了！

安装

# 安装playwright库
pip install playwright

# 安装浏览器驱动文件
python -m playwright install

#再安装
playwright install

要求：python版本3.7+

使用Playwright无需写一行代码，我们只需手动操作浏览器，它会录制我们的操作，然后自动生成代码脚本。 下面就是录制的命令codegen，仅仅一行。

python -m playwright codegen

codegen的用法可以使用–help查看，如果简单使用就是直接在命令后面加上url链接，如果有其他需要可以添加options。我就把结果粘贴出来：

Usage: npx playwright codegen [options] [url]

open page and generate code for user actions

Options:
  -o, --output      saves the generated script to a file
  --target           language to generate, one of javascript, test, python, python-async, csharp (default: "python")
  -b, --browser   browser to use, one of cr, chromium, ff, firefox, wk, webkit (default: "chromium")
  --channel           Chromium distribution channel, "chrome", "chrome-beta", "msedge-dev", etc
  --color-scheme       emulate preferred color scheme, "light" or "dark"
  --device         emulate device, for example  "iPhone 11"
  --geolocation   specify geolocation coordinates, for example "37.819722,-122.478611"
  --ignore-https-errors        ignore https errors
  --load-storage     load context storage state from the file, previously saved with --save-storage
  --lang             specify language / locale, for example "en-GB"
  --proxy-server        specify proxy server, for example "http://myproxy:3128" or "socks5://myproxy:8080"
  --save-storage     save context storage state at the end, for later use with --load-storage
  --timezone        time zone to emulate, for example "Europe/Rome"
  --timeout           timeout for Playwright actions in milliseconds (default: "10000")
  --user-agent      specify user agent string
  --viewport-size        specify browser viewport size in pixels, for example "1280, 720"
  -h, --help                   display help for command

Examples:

  $ codegen
  $ codegen --target=python
  $ codegen -b webkit https://example.com

解释： -o：将录制的脚本保存到一个文件 –target：规定生成脚本的语言，有JS和Python，java,c#等，默认为Python -b：指定浏览器驱动 举个例子： 比如，我要在baidu.com搜索，用chromium驱动，将结果保存为my.py的python文件。

python -m playwright codegen --target python -o 'my.py' -b chromium https://www.baidu.com

这就自动生成文件： 现在我还得改个名。大家可以试试。 当你在浏览器继续点击，他会继续更新生成新的代码： 这次我们执行下命令：

python -m playwright codegen --target python -o 'my.py' -b chromium https://www.baidu.comom

点击一下，代码就自动更新： 再点击一下，代码还会继续更新： 结束后自动关闭浏览器，保存生成的自动化脚本到py文件如下：

from playwright.sync_api import Playwright, sync_playwright


def run(playwright: Playwright) -> None:
    browser = playwright.chromium.launch(headless=False)
    context = browser.new_context()

    # Open new page
    page = context.new_page()

    # Go to https://www.baidu.com/
    page.goto("https://www.baidu.com/")

    # Click text=阿里女员工称被侵害 事发饭店回应
    with page.expect_popup() as popup_info:
        page.click("text=阿里女员工称被侵害 事发饭店回应")
    page1 = popup_info.value

    # Click em:has-text("阿里女员工称被侵害 事发饭店回应")
    with page1.expect_popup() as popup_info:
        page1.click("em:has-text(\"阿里女员工称被侵害 事发饭店回应\")")
    page2 = popup_info.value

    # Close page
    page2.close()

    # Close page
    page1.close()

    # Close page
    page.close()

    # ---------------------
    context.close()
    browser.close()


with sync_playwright() as playwright:
    run(playwright)

playwright还提供了同步和异步的API接口，这里也有官方文档，如果你英文还可以，可以参考文档：

https://playwright.dev/python/docs/intro/

这里我继续演示一个例子。 我就以访问我的CSDN为例子，我则需要在terminal执行以下命令：

python -m playwright codegen --target python -o 'my.py' -b chromium https://blog.csdn.net/weixin_46211269?spm=1000.2115.3001.5343

回车即可开始神操作！ 看效果，我多贴几个演示这个动态过程： 我点击了一下一篇文章，代码也跟着自动更新： 那么我还想再给自己点个赞，似乎被发现了猫腻？那我就用手机扫码登录，因为这是新的浏览器，没有登陆历史，所以这是正常的。 登陆了，我就给自己点了一个赞，代码也同时更新了我点赞的部分。 手残，点了一下图片，他还是把代码加了一部分来点击照片，下面是‘my.py’新的代码：

from playwright.sync_api import Playwright, sync_playwright


def run(playwright: Playwright) -> None:
    browser = playwright.chromium.launch(headless=False)
    context = browser.new_context()

    # Open new page
    page = context.new_page()

    # Go to https://blog.csdn.net/weixin_46211269?spm=1000.2115.3001.5343
    page.goto("https://blog.csdn.net/weixin_46211269?spm=1000.2115.3001.5343")

    # Click text=Django3.0入门教程:文章发布系统
    with page.expect_popup() as popup_info:
        page.click("text=Django3.0入门教程:文章发布系统")
    page1 = popup_info.value

    # Click #is-like-img
    page1.click("#is-like-img")

    # Click text=CSDN App扫码
    page1.frame(name="passport_iframe").click("text=CSDN App扫码")

    # Go to https://blog.csdn.net/weixin_46211269/article/details/119553344?spm=1001.2014.3001.5501
    page1.goto("https://blog.csdn.net/weixin_46211269/article/details/119553344?spm=1001.2014.3001.5501")

    # Click text=0 点赞 >> a
    page1.click("text=0 点赞 >> a")

    # Click img[alt="在这里插入图片描述"]
    page1.click("img[alt=\"在这里插入图片描述\"]", button="right")

    # Click text=在model.py复制粘贴以下代码：
    page1.click("text=在model.py复制粘贴以下代码：")

    # Click .imgViewDom img
    page1.click(".imgViewDom img")

    # Close page
    page1.close()

    # Close page
    page.close()

    # ---------------------
    context.close()
    browser.close()


with sync_playwright() as playwright:
    run(playwright)

你可以把文件名引号去掉，用这个代码运行，他则会执行相同的操作。 那么大家是不是还好奇如何分别使用同步和异步？ 那么问题来了，我先贴个我的群：970353786hhhh继续发车 由于这个生成代码如此之快，万一被发现怎么办？于是我想让他慢一点，比如这个火狐浏览器，使用slow_mo让他慢下来，而不能再用timeout,也不能用time . sleep (5) 来休息，而是可以用page.wait_for_timeout (5000)来代替，headless=False则表示无头模式

firefox.launch(headless=False, slow_mo=50)

直接说哦异步，因为同步我们已经写的太多了，如下就是异步的最简demo

import asyncio
from playwright.async_api import async_playwright

async def main():
    async with async_playwright() as p:
        browser = await p.chromium.launch(headless=False)
        await browser.close()

asyncio.run(main())

浏览器环境中也可以用来模拟多页场景涉及到移动设备,权限,语言环境和配色方案.比如：

import asyncio
from playwright.async_api import async_playwright

async def main():
    async with async_playwright() as p:
        iphone_11 = p.devices['iPhone 11 Pro']
        browser = await p.chromium.launch()
        context = await browser.new_context(
            **iphone_11,
            locale='de-DE',
            geolocation={ 'longitude': 12.492507, 'latitude': 41.889938 },
            permissions=['geolocation'],
            color_scheme='dark',
        )
        page = await browser.new_page()
        await browser.close()

asyncio.run(main())

浏览器中可以有多个页面。 一个 页面 是指一个标签或一个弹出窗口在浏览器中上下文。 它应该被用来导航到url页面内容并与之交互。比如以下代码，这是demo,你得根据你需要的网址进行修改，example.com则为demo

page = await context.new_page()

# Navigate explicitly, similar to entering a URL in the browser.
await page.goto('http://example.com')
# Fill an input.
await page.fill('#search', 'query')

# Navigate implicitly by clicking a link.
await page.click('#submit')
# Expect a new url.
print(page.url)

# Page can navigate from the script - this will be picked up by Playwright.
# window.location.href = 'https://example.com'

操作实在太多，不再继续演示了，如果你英文可以，可以看看上面的参考文档，这工具也是实在牛逼，爽爆了！