Trustpilotをスクレイピングする方法：レビューと評価の抽出（2025年版）

import asyncio
from playwright.async_api import async_playwright

async def run():
    async with async_playwright() as p:
        browser = await p.chromium.launch(headless=True)
        context = await browser.new_context(user_agent='Mozilla/5.0')
        page = await context.new_page()
        
        await page.goto('https://www.trustpilot.com/review/www.apple.com')
        
        # 動的コンテンツの読み込みを待機
        await page.wait_for_selector('section.styles_reviewCard__hc_vR')
        
        reviews = await page.evaluate('''() => {
            return Array.from(document.querySelectorAll('section.styles_reviewCard__hc_vR')).map(card => ({
                author: card.querySelector('span.typography_appearance-default__S8ne3')?.innerText,
                rating: card.querySelector('.star-rating_starRating__Bdb_f img')?.alt,
                date: card.querySelector('time')?.getAttribute('datetime')
            }));
        }''')
        
        print(reviews)
        await browser.close()

asyncio.run(run())

import scrapy

class TrustpilotSpider(scrapy.Spider):
    name = 'trustpilot'
    start_urls = ['https://www.trustpilot.com/review/www.apple.com']

    def parse(self, response):
        # レビューカード要素のループ
        for review in response.css('section.styles_reviewCard__hc_vR'):
            yield {
                'author': review.css('span.typography_appearance-default__S8ne3::text').get(),
                'rating': review.css('div.star-rating_starRating__Bdb_f img::attr(alt)').get(),
                'title': review.css('h2.styles_reviewTitle__m9_V_::text').get(),
                'date': review.css('time::attr(datetime)').get()
            }

        # 次のページへの遷移
        next_page = response.css('a[name="pagination-button-next"]::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();
  await page.goto('https://www.trustpilot.com/review/www.apple.com', { waitUntil: 'networkidle2' });

  const data = await page.evaluate(() => {
    const items = document.querySelectorAll('section.styles_reviewCard__hc_vR');
    return Array.from(items).map(item => ({
      title: item.querySelector('h2')?.innerText,
      body: item.querySelector('p')?.innerText
    }));
  });

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