OnTheMarketのスクレイピング方法 | OnTheMarket Web Scraper

from playwright.sync_api import sync_playwright

def scrape_otm():
    with sync_playwright() as p:
        browser = p.chromium.launch(headless=True)
        # ステルスに近いコンテキストを使用
        context = browser.new_context(user_agent='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36')
        page = context.new_page()
        page.goto('https://www.onthemarket.com/for-sale/property/london/', wait_until='networkidle')
        
        # 結果がレンダリングされるのを待機
        page.wait_for_selector('li[id^="result-"]')
        
        listings = page.query_selector_all('li[id^="result-"]')
        for prop in listings:
            title = prop.query_selector('.text-sm.text-denim').inner_text()
            price = prop.query_selector('.text-xl.font-bold').inner_text()
            print({'title': title, 'price': price})
        
        browser.close()

scrape_otm()

import scrapy

class OnTheMarketSpider(scrapy.Spider):
    name = 'otm'
    start_urls = ['https://www.onthemarket.com/for-sale/property/london/']

    def parse(self, response):
        # メインのリストコンテナ内の各アイテムをターゲットにする
        for item in response.css('li[id^="result-"]'):
            yield {
                'price': item.css('.text-xl.font-bold::text').get(),
                'address': item.css('address span::text').get(),
                'agency': item.css('img::attr(alt)').get(),
                'link': response.urljoin(item.css('a::attr(href)').get())
            }

        next_page = response.css('link[rel="next"]::attr(href)').get()
        if next_page:
            yield response.follow(next_page, self.parse)

const puppeteer = require('puppeteer');

(async () => {
  const browser = await puppeteer.launch({ headless: true });
  const page = await browser.newPage();
  await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36');
  await page.goto('https://www.onthemarket.com/for-sale/property/london/', { waitUntil: 'networkidle2' });

  const data = await page.evaluate(() => {
    return Array.from(document.querySelectorAll('li[id^="result-"]')).map(li => ({
      price: li.querySelector('.text-xl')?.innerText.trim(),
      address: li.querySelector('address')?.innerText.trim()
    }));
  });

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