从环境搭建到实战详解Python中Playwright的入门教程

# 1. 安装核心库（推荐使用清华源加速）
pip install playwright -i https://pypi.tuna.tsinghua.edu.cn/simple

# 2. 下载浏览器二进制文件（自动匹配系统环境）
python -m playwright install

# 3. 安装pytest插件（可选，用于测试框架集成）
pip install pytest-playwright

from playwright.sync_api import sync_playwright

with sync_playwright() as p:
    browser = p.chromium.launch(headless=False)  # 有头模式便于调试
    page = browser.new_page()
    page.goto("https://playwright.dev")
    print(page.title())  # 应输出"Playwright"
    browser.close()

from playwright.sync_api import sync_playwright

def test_baidu_search():
    with sync_playwright() as p:
        # 启动浏览器（无头模式示例）
        browser = p.chromium.launch(headless=True)
        context = browser.new_context()  # 创建隔离环境
        page = context.new_page()
        
        # 执行操作
        page.goto("https://www.baidu.com")
        page.fill("input[name=wd]", "Playwright教程")
        page.click("text=百度一下")
        
        # 验证结果
        assert "Playwright" in page.title()
        
        # 资源清理
        context.close()
        browser.close()

# 通过角色定位按钮
page.get_by_role("button", name="登录").click()

# 通过标签文本定位
page.get_by_text("立即购买").click()

# 通过placeholder定位输入框
page.get_by_placeholder("请输入手机号").fill("13800138000")

# ID定位
page.locator("#submit-btn").click()

# 属性组合定位
page.locator("input[type='text'][class='form-control']").fill("test")

# 伪类选择（定位奇数行）
page.locator("tr:nth-of-type(odd)").click()

# 组合条件定位
page.locator("//input[@type='text' and contains(@placeholder,'密码')]").fill("123456")

# 轴定位（定位父元素）
page.locator("//button/parent::div").click()

def test_api_mock():
    with sync_playwright() as p:
        browser = p.chromium.launch()
        page = browser.new_page()
        
        # 拦截特定API请求
        def handle_route(route):
            if "api/data" in route.request.url:
                route.fulfill(
                    status=200,
                    json={"message": "Mock数据"}
                )
            else:
                route.continue_()
        
        page.route("**/*", handle_route)
        page.goto("https://example.com")
        assert "Mock数据" in page.content()
        browser.close()

def test_mobile_view():
    with sync_playwright() as p:
        # 使用预置设备参数
        iphone = p.devices["iPhone 12"]
        browser = p.chromium.launch()
        
        # 创建移动环境
        context = browser.new_context(
            **iphone,
            locale="zh-CN",
            timezone_id="Asia/Shanghai"
        )
        
        page = context.new_page()
        page.goto("https://m.taobao.com")
        page.screenshot(path="mobile_view.png")
        context.close()
        browser.close()

# pytest_playwright示例（conftest.py）
import pytest
from playwright.sync_api import sync_playwright

@pytest.fixture(scope="session")
def browser():
    pw = sync_playwright().start()
    browser = pw.chromium.launch(headless=True)
    yield browser
    browser.close()
    pw.stop()

def test_example(browser):
    page = browser.new_page()
    page.goto("https://bing.com")
    assert "Bing" in page.title()

<button data-testid="login-submit">登录</button>

page.get_by_test_id("login-submit").click()

frame = page.frame_locator("iframe.login-frame")
frame.locator("#username").fill("admin")

playwright codegen https://example.com

import logging
logging.basicConfig(level=logging.INFO)  # 显示详细日志

from playwright.sync_api import sync_playwright
import json
import os

class QuarkDownloader:
    def __init__(self):
        self.storage_path = "state.json"
        self.quark_url = "https://quark.cn/share"
        self.username = "your_username"  # 需替换为实际用户名

    def login_and_save_state(self):
        with sync_playwright() as p:
            browser = p.chromium.launch(headless=False)  # 显示浏览器界面
            page = browser.new_page()
            
            # 访问登录页
            page.goto(self.quark_url)
            
            # 等待扫码登录（实际项目中需替换为具体登录逻辑）
            page.wait_for_selector(f"span:text('{self.username}')")
            
            # 保存浏览器状态
            storage_state = browser.new_context().storage_state()
            with open(self.storage_path, "w") as f:
                json.dump(storage_state, f)
            
            browser.close()

    def download_file(self, share_url, extract_code):
        if not os.path.exists(self.storage_path):
            self.login_and_save_state()

        with sync_playwright() as p:
            browser = p.chromium.launch(headless=True)
            context = browser.new_context(storage_state=self.storage_path)
            page = context.new_page()
            
            # 监听下载事件
            def handle_download(route):
                if "download" in route.request.url:
                    route.continue_()
            
            page.route("**/*", handle_download)
            
            # 访问分享链接
            page.goto(share_url)
            page.fill("input[placeholder='提取码']", extract_code)
            page.click("text=提取文件")
            
            # 等待下载完成（实际项目中需根据具体下载逻辑调整）
            page.wait_for_timeout(10000)
            
            context.close()
            browser.close()

# 使用示例
downloader = QuarkDownloader()
downloader.download_file(
    share_url="https://quark.cn/share/xxxxxx",  # 替换为实际分享链接
    extract_code="1234"  # 替换为实际提取码
)