如何使用Scrapeless抓取Google Lens结果

python --version

python -m venv google_lens

google_lens_env\Scripts\activate

source google_lens_env/bin/activate

pip install requests
pip install beautifulsoup4
pip install playwright

# 将 lens 信息存储在字典中
img_info = {
    'title': item.find("span").text,
    'thumbnail': item.find("img").attrs['src'],
}

import json

from bs4 import BeautifulSoup
from playwright.sync_api import sync_playwright


def scrape(url: str) -> str:
    with sync_playwright() as p:
        # 启动浏览器并禁用一些可能导致检测的功能
        browser = p.chromium.launch(
            headless=True,
            args=[
                "--disable-blink-features=AutomationControlled",
                "--disable-dev-shm-usage",
                "--disable-gpu",
                "--disable-extensions",
            ],
        )
        context = browser.new_context(
            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36",
            bypass_csp=True,
        )
        page = context.new_page()
        page.goto(url)
        page.wait_for_selector("body", state="attached")
        # 等待 2 秒以确保页面完全加载或渲染
        page.wait_for_timeout(2000)
        html_content = page.content()
        browser.close()
    return html_content


def main():
    url = "https://lens.google.com/uploadbyurl?url=https%3A%2F%2Fi.imgur.com%2FHBrB8p0.png"
    html_content = scrape(url)
    soup = BeautifulSoup(html_content, 'html.parser')
    # 获取页面的主要数据
    items = soup.find('div', {'jscontroller': 'M3v8m'}).find("div")
    # 循环组装
    assembly = lens_info(items)
    # 将结果保存到 JSON 文件
    with open('google_lens_data.json', 'w') as json_file:
        json.dump(assembly, json_file, indent=4)


def lens_info(items):
    lens_data = []
    for item in items:
        # 将 lens 信息存储在字典中
        img_info = {
            'title': item.find("span").text,
            'thumbnail': item.find("img").attrs['src'],
        }
        lens_data.append(img_info)
    return lens_data


if __name__ == "__main__":
    main()

[
  {
    "title": "Danny DeVito - Wikipedia",
    "thumbnail": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/2wCEAAkGBxAQEhAQEBAQEB
  },
  {
    "title": "Devito Danny Royalty-Free Images, Stock Photos & Pictures | Shutterstock",
    "thumbnail": "https://encrypted-tbn1.gstatic.com/images?q=tbn:ANd9GcSO6Pkv_UmXiianiCh52nD5s89d7KrlgQQox-f-K9FtXVILvHh_"
  },
  {
    "title": "DATA | Celebrity Stats | Page 62",
    "thumbnail": "https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcQ9juRVpW6sjE3OANTKIJzGEkiwUpjCI20Z1ydvJBCEDf3-NcQE"
  },
  {
    "title": "Danny DeVito, Grand opening of Buca di Beppo italian restaurant on Universal City Walk Universal City, California - 28.01.09 Stock Photo - Alamy",
    "thumbnail": "https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQq_07f-Unr7Y5BXjSJ224RlAidV9pzccqjucD4VF7VkJEJJqBk"
  }
]

import json
import requests

class Payload:
    def __init__(self, actor, input_data):
        self.actor = actor
        self.input = input_data

def send_request():
    host = "api.scrapeless.com"
    url = f"https://{host}/api/v1/scraper/request"
    token = "your_token"

    headers = {
        "x-api-token": token
    }

    input_data = {
        engine: "google_lens",
        hl: "en",
        country: "jp",
        url: "https://s3.zoommer.ge/zoommer-images/thumbs/0170510_apple-macbook-pro-13-inch-2022-mneh3lla-m2-chip-8gb256gb-ssd-space-grey-apple-m25nm-apple-8-core-gpu_550.jpeg",
    }

    payload = Payload("scraper.google.lens", input_data)

    json_payload = json.dumps(payload.__dict__)

    response = requests.post(url, headers=headers, data=json_payload)

    if response.status_code != 200:
        print("Error:", response.status_code, response.text)
        return

    print("body", response.text)


if __name__ == "__main__":
    send_request()