Python - Playwright安装与基础使用

前言：

Playwright 是专门为满足端到端测试的需要而创建的。Playwright 支持所有现代渲染引擎，包括 Chromium、WebKit(Safari 的浏览器引擎)和 Firefox。在 Windows、Linux 和 macOS 上进行本地测试或在 CI 上进行测试.

与Selenium+driver不同的是，Pw需要使用定制版的浏览器。如果本地已经安装了浏览器，那么可能会被覆盖掉，所以安装时请注意。官方会通过微软CDN自动下载，也支持配置外网代理下载，如果微软CDN在工作环境禁用，申请外网权限后，可以通过搭建一个跳板机进行代理下载。Playwright 安装的定制浏览器可以作为普通浏览器正常使用

 

一、安装：

 

安装Playwright

python3 -m pip install playwright -i https://mirrors.aliyun.com/pypi/simple/  # 安装playwright的python版本

 

安装Playwright自带的浏览器

python3 -m playwright install  # 安装playwright自带的浏览器和ffmepg

 

 

二、基本使用：

cmd输入命令

仅打开

playwright open  #打开Playwright

 

打开并开启录制

playwright codegen  #开启录制

#指定网址
playwright codegen www.baidu.com    

#移动端浏览
playwright codegen --device="iPhone 11"

#输出保存的代码文件名称
playwright codegen -o test.py

#浏览器的名称,默认Google chromium浏览器
playwright codegen -b chromium

#保留经过身份验证的状态
playwright codegen --save-storage=auth.json  URL

#取的状态文件路径，文件包含cookie信息
playwright codegen --load-storage=auth.json URL

 

 

三、录制的代码：

from playwright.sync_api import Playwright, sync_playwright, expect


def run(playwright: Playwright) -> None:
    browser = playwright.chromium.launch(headless=False)
    context = browser.new_context()
    page = context.new_page()
    page.goto("https://www.baidu.com/")
    page.locator("#kw").click()
    page.locator("#kw").fill("Playwright")
    age.wait_for_timeout(2000)   #等待
    page.locator("#kw").press("Enter")

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


with sync_playwright() as playwright:
    run(playwright)

 

 

from playwright.sync_api import Playwright, sync_playwright, expect


def run(playwright: Playwright) -> None:
    browser = playwright.chromium.launch(headless=False)
    context = browser.new_context()
    page = context.new_page()
    page.goto("https://www.baidu.com/")
    page.locator("#s-usersetting-top").hover()    #鼠标悬停设置
    page.wait_for_timeout(3000)
    page.get_by_role("link", name="搜索设置").click()
    page.wait_for_timeout(5000)

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


with sync_playwright() as playwright:
    run(playwright)

 

 

 

四、相关命令：

 

# 访问具体网站
page.goto(url)

#悬停
page.locator("#xx").hover()

# 监听事件, 比如close、console、load、request、response等
page.on(event, callback)

#页面等待
page.wait_for_timeout(2000)

# 等待页面500 ms没有网络请求   可切换到domcontentloaded - 等到加载DOMContentLoaded事件
page.wait_for_load_state(state='networkidle')

# 请求拦截
page.route(url, handler) # url可以是普通字符串（必须含url末位字符），也可以是正则pattern字符串，即re.compile(r'xxx')

# 设置页面大小
page.set_viewport_size({'width':1366, 'height':768})

# 执行js代码
data1 = page.evaluate('() => {return window.encrypt("xx", "yyy")}')
data2 = page.evaluate('([a, b]) => a + b', [3, 4]) # 7
data3 = page.evaluate('10+5') # 15

js_code1 = '''
var test = function(a,b){window.hello = 20;return a + b}; // 只能使用函数表达式，使用函数声明则会报错
test(30,60) // 不能使用return
'''
js_code2 = '''window.hello'''
js_code3 = '''
var a = 20;
var b = 30;
a + b
'''
print(page.evaluate(js_code1)) # 90
print(page.evaluate(js_code2)) # 20
print(page.evaluate(js_code3)) # 50

# 截图
page.screenshot(type=None, path=None)

# 点击页面元素
page.click(selector)

# 获取页面源码
page.content()

# 获取单个节点
element = page.query_selector(selector)
element.get_attribute('属性名') # 获取节点属性
element.text_content() # 获取节点文本

# 获取多个节点
elements = page.query_selector_all(selector)
for ele in elements:
    ele.get_attribute('属性名') # 获取属性
    ele.text_content() # 获取文本

# 文本输入
page.fill(selector, value, timeout=None) # 根据选择器，输入文本内容，timeout可设置对应节点的最长等待时间

 

 

五、事件监听：

Page对象提供了on方法，用来监听页面中的各个事件，比如close、console、load、request、response等

from playwright.sync_api import sync_playwright


def on_response(response):
    # 直接截获ajax请求数据
    if '/api/movie' in response.url and response.status == 200:
        print(response.json())


with sync_playwright() as playwright:
    browser = playwright.chromium.launch(headless=False)
    page = browser.new_page()
    # 事件监听，例如close、console、load、request、response等
    page.on('response', on_response)
    page.goto("https://spa6.scrape.center/") # 访问网址
    page.wait_for_load_state(state='networkidle') # 等待当前页面初始化和加载完成
    browser.close()