from playwright.sync_api import sync_playwright

with sync_playwright() as p:
    # 啟動Chrome
    chrome_browser = p.chromium.launch(headless=False)
    # 啟動Firefox
    firefox_browser = p.firefox.launch(headless=False)
    # 啟動WebKit（Safari內(nèi)核）
    webkit_browser = p.webkit.launch(headless=False)

page.get_by_role("button", name="登錄").click()  # 自動處理加載狀態(tài)

# 攔截特定API請求
def handle_route(route):
    if "api/products" in route.request.url:
        # 返回本地JSON文件
        with open("mock_data.json", "r") as f:
            mock_data = f.read()
        route.fulfill(body=mock_data, content_type="application/json")
    else:
        route.continue_()

page.route("**/*", handle_route)

from playwright.sync_api import sync_playwright
import json

def scrape_product_data(url):
    results = []
    
    with sync_playwright() as p:
        # 啟動瀏覽器（建議使用無頭模式時設(shè)置慢速動畫）
        browser = p.chromium.launch(headless=False, slow_mo=500)
        context = browser.new_context(
            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36...",
            ignore_https_errors=True
        )
        page = context.new_page()
        
        # 攔截圖片請求加速爬取
        page.route("**/*.{png,jpg,jpeg,gif}", lambda route: route.abort())
        
        page.goto(url, wait_until="networkidle")
        
        # 抓取第一頁數(shù)據(jù)
        products = page.query_selector_all(".product-item")
        for product in products:
            # 模擬鼠標(biāo)懸停顯示價格
            page.mouse.move(x=float(product.get_attribute("data-x")), 
                           y=float(product.get_attribute("data-y")))
            page.wait_for_selector(".price-popup", state="visible")
            
            data = {
                "name": product.get_by_text(".product-name").inner_text(),
                "price": product.get_by_text(".price-value").inner_text(),
                "sales": product.get_by_text(".sales-count").inner_text(),
                "shop": product.get_by_text(".shop-name").inner_text()
            }
            results.append(data)
        
        # 點(diǎn)擊下一頁直到抓取3頁
        for _ in range(2):
            next_button = page.get_by_role("button", name="下一頁")
            if next_button.is_disabled():
                break
            next_button.click()
            page.wait_for_network_idle()
            
            # 重復(fù)數(shù)據(jù)抓取邏輯...
            
        browser.close()
        return results

# 使用示例
if __name__ == "__main__":
    data = scrape_product_data("https://example.com/products")
    with open("products.json", "w", encoding="utf-8") as f:
        json.dump(data, f, ensure_ascii=False, indent=2)

# 推薦方式（按優(yōu)先級）
page.get_by_role("button", name="提交")  # ARIA角色定位
page.get_by_text("立即購買")            # 文本內(nèi)容定位
page.get_by_label("用戶名")             # 關(guān)聯(lián)標(biāo)簽定位
page.get_by_placeholder("請輸入密碼")    # 占位符定位
page.get_by_test_id("user-email")       # 測試ID定位
page.get_by_alt_text("品牌logo")        # 圖片替代文本定位

def scroll_to_bottom(page, max_scroll=10):
    last_height = page.evaluate("document.body.scrollHeight")
    for _ in range(max_scroll):
        page.evaluate("window.scrollTo(0, document.body.scrollHeight)")
        page.wait_for_timeout(1000)  # 等待內(nèi)容加載
        new_height = page.evaluate("document.body.scrollHeight")
        if new_height == last_height:
            break
        last_height = new_height

# 點(diǎn)擊導(dǎo)航后等待URL變化
page.get_by_text("分類").click()
page.wait_for_url("**/category/**", timeout=5000)

# 模擬移動設(shè)備
context = browser.new_context(
    viewport={"width": 375, "height": 667},
    user_agent="Mozilla/5.0 (iPhone; CPU iPhone OS 14_0 like Mac OS X) AppleWebKit...",
    device_scale_factor=2,
    is_mobile=True,
    has_touch=True
)

from playwright.sync_api import sync_playwright

def main():
    with sync_playwright() as p:
        # 啟動持久化瀏覽器上下文
        browser = p.chromium.launch_persistent_context(
            "./user_data_dir",
            headless=False,
            args=["--start-maximized"]
        )
        
        # 多次爬取任務(wù)復(fù)用同一個瀏覽器
        for url in ["https://example.com/page1", "https://example.com/page2"]:
            page = browser.new_page()
            page.goto(url)
            # 爬取邏輯...
            page.close()
        
        browser.close()

if __name__ == "__main__":
    main()

from concurrent.futures import ThreadPoolExecutor
from playwright.sync_api import sync_playwright

def scrape_task(url):
    with sync_playwright() as p:
        browser = p.chromium.launch(headless=True)
        page = browser.new_page()
        page.goto(url)
        # 爬取邏輯...
        browser.close()
        return data

urls = ["https://example.com/1", "https://example.com/2"]
with ThreadPoolExecutor(max_workers=3) as executor:
    results = list(executor.map(scrape_task, urls))