如何爬取 Kalodata：TikTok Shop 数据抓取指南

import asyncio
from playwright.async_api import async_playwright

async def scrape_kalodata():
    async with async_playwright() as p:
        # 使用类似 stealth 的参数来规避 Cloudflare 检测
        browser = await p.chromium.launch(headless=True)
        context = await browser.new_context(user_agent='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36')
        page = await context.new_page()
        
        # 导航至商品排名页面
        await page.goto('https://www.kalodata.com/product')
        
        # 等待表格行从内部 API 动态加载
        await page.wait_for_selector('.table-row-container', timeout=15000)
        
        # 提取商品名称和相关指标
        products = await page.query_selector_all('.product-name-class')
        for product in products:
            name = await product.inner_text()
            print(f'找到商品: {name}')
            
        await browser.close()

asyncio.run(scrape_kalodata())

import scrapy

class KalodataSpider(scrapy.Spider):
    name = 'kalodata_spider'
    start_urls = ['https://www.kalodata.com/shop']

    def parse(self, response):
        # 注意：对于这个重度依赖 JS 的网站，Scrapy 需要像 scrapy-playwright 这样的中间件
        for shop in response.css('.shop-list-item'):
            yield {
                'name': shop.css('.shop-name::text').get(),
                'revenue': shop.css('.revenue-value::text').get(),
                'sold': shop.css('.items-sold::text').get(),
            }

        # 针对页码的标准分页处理
        next_page = response.css('a.next-page-selector::attr(href)').get()
        if next_page:
            yield response.follow(next_page, self.parse)

const puppeteer = require('puppeteer-extra');
const StealthPlugin = require('puppeteer-extra-plugin-stealth');
puppeteer.use(StealthPlugin());

(async () => {
  const browser = await puppeteer.launch({ headless: true });
  const page = await browser.newPage();
  
  // 导航至达人洞察页面
  await page.goto('https://www.kalodata.com/creator', { waitUntil: 'networkidle2' });

  // 等待动态列表加载完成
  await page.waitForSelector('.creator-list-container');

  const creators = await page.evaluate(() => {
    const items = Array.from(document.querySelectorAll('.creator-item'));
    return items.map(item => ({
      name: item.querySelector('.name')?.innerText,
      followers: item.querySelector('.followers')?.innerText,
      category: item.querySelector('.category-tag')?.innerText
    }));
  });

  console.log(creators);
  await browser.close();
})();