cannaiq/backend/archive/scrape-azdhs-map.ts

import { chromium } from 'playwright-extra';
import stealth from 'puppeteer-extra-plugin-stealth';
import { pool } from './src/db/migrate';

chromium.use(stealth());

async function scrapeAZDHSMap() {
  console.log('🏛️  Scraping official AZDHS Arizona dispensary map...\n');

  const browser = await chromium.launch({
    headless: false,
  });

  const context = await browser.newContext({
    viewport: { width: 1920, height: 1080 },
    userAgent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36',
  });

  const page = await context.newPage();

  try {
    console.log('📄 Loading AZDHS map page...');

    await page.goto('https://azcarecheck.azdhs.gov/s/?facilityId=001t000000L0TApAAN', {
      waitUntil: 'networkidle',
      timeout: 60000
    });

    // Wait for page to fully load
    await page.waitForTimeout(10000);

    console.log('🔍 Extracting dispensary data...\n');

    // Extract all dispensary data
    const dispensaries = await page.evaluate(() => {
      const results: any[] = [];

      // This is a Salesforce Community page - look for the data
      // Check window object
      const windowKeys = Object.keys(window);
      console.log('Window keys sample:', windowKeys.filter(k => k.includes('SF') || k.includes('data') || k.includes('facility')).slice(0, 20));

      // Try to find all facility/location markers or data
      // Look for common patterns in Salesforce maps
      const facilityElements = document.querySelectorAll('[class*="facility"], [class*="location"], [class*="marker"]');
      console.log('Found facility elements:', facilityElements.length);

      return {
        debug: 'AZDHS page',
        windowKeysCount: windowKeys.length,
        facilityElements: facilityElements.length,
        sampleHTML: document.body.innerHTML.substring(0, 2000)
      };
    });

    console.log('Debug info:');
    console.log(JSON.stringify(dispensaries, null, 2).substring(0, 1500));

  } catch (error) {
    console.error(`❌ Error: ${error}`);
    throw error;
  } finally {
    await browser.close();
    await pool.end();
  }
}

scrapeAZDHSMap();