cannaiq/backend/src/scripts/debug-dutchie-page.ts

/**
 * Debug Dutchie city page to see what data is available
 */

import puppeteer from 'puppeteer-extra';
import StealthPlugin from 'puppeteer-extra-plugin-stealth';

puppeteer.use(StealthPlugin());

async function main() {
  const cityUrl = process.argv[2] || 'https://dutchie.com/us/dispensaries/wa-bellevue';

  console.log(`Debugging page: ${cityUrl}`);

  const browser = await puppeteer.launch({
    headless: 'new',
    args: ['--no-sandbox', '--disable-setuid-sandbox', '--disable-dev-shm-usage'],
  });

  try {
    const page = await browser.newPage();
    await page.setUserAgent(
      'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'
    );

    console.log('Navigating...');
    await page.goto(cityUrl, {
      waitUntil: 'networkidle2',
      timeout: 60000,
    });

    await new Promise((r) => setTimeout(r, 5000));

    // Get page title
    const title = await page.title();
    console.log(`\nPage title: ${title}`);

    // Check for Cloudflare challenge
    const isCFChallenge = await page.evaluate(() => {
      return document.title.includes('Just a moment') ||
             document.body.textContent?.includes('Enable JavaScript');
    });

    if (isCFChallenge) {
      console.log('\n⚠️  CLOUDFLARE CHALLENGE DETECTED - waiting longer...');
      await new Promise((r) => setTimeout(r, 10000));
    }

    // Check for __NEXT_DATA__
    const nextData = await page.evaluate(() => {
      const script = document.querySelector('script#__NEXT_DATA__');
      if (script) {
        try {
          return JSON.parse(script.textContent || '{}');
        } catch {
          return { error: 'Failed to parse __NEXT_DATA__' };
        }
      }
      return null;
    });

    if (nextData) {
      console.log('\n✅ __NEXT_DATA__ found!');
      console.log('Keys:', Object.keys(nextData));
      if (nextData.props?.pageProps) {
        console.log('pageProps keys:', Object.keys(nextData.props.pageProps));
        if (nextData.props.pageProps.dispensaries) {
          console.log('Dispensaries count:', nextData.props.pageProps.dispensaries.length);
          // Show first dispensary structure
          const first = nextData.props.pageProps.dispensaries[0];
          if (first) {
            console.log('\nFirst dispensary keys:', Object.keys(first));
            console.log('First dispensary sample:', JSON.stringify(first, null, 2).slice(0, 1000));
          }
        }
      }
    } else {
      console.log('\n❌ No __NEXT_DATA__ found');

      // Check what scripts are on the page
      const scripts = await page.evaluate(() => {
        return Array.from(document.querySelectorAll('script[id]')).map(s => ({
          id: s.id,
          src: (s as HTMLScriptElement).src?.slice(0, 100),
        }));
      });
      console.log('Scripts with IDs:', scripts);

      // Try to find dispensary data in window object
      const windowData = await page.evaluate(() => {
        const w = window as any;
        const keys = ['__NEXT_DATA__', '__PRELOADED_STATE__', '__INITIAL_STATE__',
                      'dispensaries', '__data', 'pageData', '__remixContext'];
        const found: Record<string, any> = {};
        for (const key of keys) {
          if (w[key]) {
            found[key] = typeof w[key] === 'object' ? Object.keys(w[key]) : typeof w[key];
          }
        }
        return found;
      });
      console.log('Window data:', windowData);

      // Get some page content
      const bodyText = await page.evaluate(() => document.body.innerText.slice(0, 500));
      console.log('\nPage text preview:', bodyText);
    }

  } finally {
    await browser.close();
  }
}

main().catch(console.error);