BureauxLocauxをスクレイピングする方法：商業用不動産データガイド

from playwright.sync_api import sync_playwright

def scrape_bureaux():
    with sync_playwright() as p:
        # ステルス設定または特定のUAでの起動を推奨
        browser = p.chromium.launch(headless=True)
        context = browser.new_context(user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36")
        page = context.new_page()
        
        # 検索結果ページへ移動
        page.goto("https://www.bureauxlocaux.com/immobilier-d-entreprise/annonces/location-bureaux", wait_until="networkidle")
        
        # リスティングが表示されるのを待機
        page.wait_for_selector(".AnnonceCard")
        
        listings = page.query_selector_all(".AnnonceCard")
        for item in listings:
            title = item.query_selector("h2").inner_text()
            price = item.query_selector(".price").inner_text() if item.query_selector(".price") else "エージェントに問い合わせ"
            print(f"{title}: {price}")
            
        browser.close()

scrape_bureaux()

import scrapy

class BureauxSpider(scrapy.Spider):
    name = 'bureaux_spider'
    start_urls = ['https://www.bureauxlocaux.com/immobilier-d-entreprise/annonces/location-bureaux']

    def parse(self, response):
        # ページ上の各物件カードをループ処理
        for ad in response.css('.AnnonceCard'):
            yield {
                'title': ad.css('h2::text').get(default='').strip(),
                'price': ad.css('.price::text').get(default='').strip(),
                'location': ad.css('.location::text').get(default='').strip(),
                'url': response.urljoin(ad.css('a::attr(href)').get())
            }

        # ページネーション: 「次へ」リンクを取得
        next_page = response.css('a.pagination-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();
  
  // 現実的なUser-Agentを設定
  await page.setUserAgent('Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36');

  await page.goto('https://www.bureauxlocaux.com/immobilier-d-entreprise/annonces/location-bureaux', { waitUntil: 'networkidle2' });
  
  // リスティング要素からデータを抽出
  const data = await page.evaluate(() => {
    const items = Array.from(document.querySelectorAll('.AnnonceCard'));
    return items.map(el => ({
      title: el.querySelector('h2')?.innerText.trim(),
      price: el.querySelector('.price')?.innerText.trim(),
      location: el.querySelector('.location-text')?.innerText.trim()
    }));
  });

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