playwright将网页截图的方式

 from playwright.async_api import async_playwright

import os

import pytesseract

from PIL import Image

import asyncio

proxy = 'http://abc.com.cn:80'

async def download():

    async with async_playwright() as p:

        cookie = os.path.join(cookies_folder, "cookie.json")

        browser = await p.chromium.launch(

            headless=False,

            channel="chrome",

            proxy={"server": proxy}  # 设置代理?

        )

        context = await browser.new_context()

        page = await context.new_page()

        await page.goto("https://ddd.com/", timeout=180000)

        await page.wait_for_load_state("networkidle")

        # screenshot_path = 'fullpage.png'

        # await page.screenshot(path=screenshot_path, full_page=True)

        shadow_host = await page.query_selector('div#transcend-consent-manager')

        await shadow_host.screenshot(path='shadow_host.png')

        image = Image.open('shadow_host.png')

        data = pytesseract.image_to_data(image, lang='chi_sim+eng', output_type=pytesseract.Output.DICT)