#!/usr/bin/env node

/**
 * Reonomy Scraper v4 - FINAL VERSION
 *
 * Key discoveries from browser inspection:
 * 1. Search for location → Get search-id from URL
 * 2. Extract all property IDs from search results
 * 3. Navigate to ownership view for each property:
 *    /search/{search-id}/property/{property-id}/ownership
 * 4. Extract emails/phones from mailto:/tel: links
 */

const puppeteer = require('puppeteer');
const fs = require('fs');
const path = require('path');

// Configuration
const REONOMY_EMAIL = process.env.REONOMY_EMAIL || 'henry@realestateenhanced.com';
const REONOMY_PASSWORD = process.env.REONOMY_PASSWORD || '9082166532';
const SEARCH_LOCATION = process.env.REONOMY_LOCATION || 'Eatontown, NJ';
const HEADLESS = process.env.HEADLESS === 'true';
const MAX_PROPERTIES = 20; // Number of properties to scrape
const PAGE_DELAY_MS = 3000; // Rate limiting delay between ownership pages

// Output files
const OUTPUT_FILE = path.join(__dirname, 'reonomy-leads-v4.json');
const LOG_FILE = path.join(__dirname, 'reonomy-scraper-v4.log');

function log(message) {
  const timestamp = new Date().toISOString();
  const logMessage = `[${timestamp}] ${message}\n`;
  console.log(message);
  fs.appendFileSync(LOG_FILE, logMessage);
}

function sleep(ms) {
  return new Promise(resolve => setTimeout(resolve, ms));
}

/**
 * Extract contact info from ownership page
 */
async function extractContactInfo(page, propertyUrl) {
  return await page.evaluate(() => {
    const info = {
      emails: [],
      phones: [],
      address: '',
    };

    // Extract emails from mailto: links
    document.querySelectorAll('a[href^="mailto:"]').forEach(a => {
      const email = a.href.replace('mailto:', '');
      if (email && email.length > 5) {
        info.emails.push(email);
      }
    });

    // Extract phones from tel: links
    document.querySelectorAll('a[href^="tel:"]').forEach(a => {
      const phone = a.href.replace('tel:', '');
      if (phone && phone.length > 7) {
        info.phones.push(phone);
      }
    });

    // Extract property address
    const addressMatch = document.body.innerText.match(/^(\d+[^,]+),\s*([A-Za-z\s]+),\s*([A-Z]{2})\s*(\d{5})/);
    if (addressMatch) {
      info.address = addressMatch[0];
    }

    return info;
  });
}

/**
 * Extract property IDs from search results page
 */
async function extractPropertyIds(page) {
  return await page.evaluate(() => {
    const propertyIds = [];
    const links = document.querySelectorAll('a[href*="/property/"]');

    links.forEach(link => {
      const href = link.href;
      const match = href.match(/property\/([a-f0-9-]+)/);

      if (match) {
        propertyIds.push({
          id: match[1],
          url: href
        });
      }
    });

    return propertyIds;
  });
}

/**
 * Main scraper
 */
async function scrapeLeads() {
  log('🚀 Starting Reonomy Scraper v4 (FINAL VERSION)...\n');

  const browser = await puppeteer.launch({
    headless: HEADLESS ? 'new' : false,
    args: ['--no-sandbox', '--disable-setuid-sandbox', '--window-size=1920,1080']
  });

  const page = await browser.newPage();
  await page.setViewport({ width: 1920, height: 1080 });

  const leads = [];

  try {
    // Login
    log('📍 Step 1: Logging into Reonomy...');
    await page.goto('https://app.reonomy.com/#!/account', {
      waitUntil: 'domcontentloaded',
      timeout: 60000
    });

    await sleep(2000);

    await page.type('input[type="email"]', REONOMY_EMAIL, { delay: 100 });
    await page.type('input[type="password"]', REONOMY_PASSWORD, { delay: 100 });
    await page.click('button[type="submit"]');

    log('⏳ Waiting for login...');
    await sleep(10000);

    // Check if logged in
    const url = page.url();
    if (url.includes('login') || url.includes('auth')) {
      throw new Error('Login failed. Please check credentials.');
    }

    log('✅ Successfully logged in!');

    // Navigate to search
    log(`\n📍 Step 2: Navigating to search...`);
    await page.goto('https://app.reonomy.com/#!/search', {
      waitUntil: 'networkidle2',
      timeout: 60000
    });

    await sleep(3000);

    // Perform search
    log(`📍 Step 3: Searching for: ${SEARCH_LOCATION}...`);

    const searchInput = await page.waitForSelector('input[placeholder*="address"], input[placeholder*="Search"]', {
      timeout: 10000
    }).catch(() => {
      return page.waitForSelector('input[type="text"]', { timeout: 5000 });
    });

    if (searchInput) {
      await searchInput.click({ clickCount: 3 });
      await searchInput.type(SEARCH_LOCATION, { delay: 100 });
      await sleep(1000);
      await page.keyboard.press('Enter');
      log('⏳ Searching...');
      await sleep(5000);
    }

    // Extract search ID from URL
    const urlMatch = page.url().match(/search\/([a-f0-9-]+)/);
    if (!urlMatch) {
      throw new Error('Could not extract search ID from URL');
    }
    const searchId = urlMatch[1];
    log(`✅ Search ID: ${searchId}`);

    // Extract property IDs from search results
    log('\n📍 Step 4: Extracting property IDs...');
    const propertyIds = await extractPropertyIds(page);
    log(`✅ Found ${propertyIds.length} property IDs`);

    if (propertyIds.length === 0) {
      log('⚠️  No property IDs found. The page structure may have changed.');
      throw new Error('No properties found on search page');
    }

    // Limit to MAX_PROPERTIES
    const propertiesToScrape = propertyIds.slice(0, MAX_PROPERTIES);

    // For each property, visit ownership page and extract contact info
    log(`\n📍 Step 5: Scraping ${propertiesToScrape.length} properties...`);

    for (let i = 0; i < propertiesToScrape.length; i++) {
      const prop = propertiesToScrape[i];
      log(`\n[${i + 1}/${propertiesToScrape.length}] Property ID: ${prop.id}`);

      // Build ownership URL
      const ownershipUrl = `https://app.reonomy.com/#!/search/${searchId}/property/${prop.id}/ownership`;
      log(`  🔗 Navigating to ownership page...`);

      await page.goto(ownershipUrl, {
        waitUntil: 'networkidle2',
        timeout: 30000
      });

      await sleep(2000);

      // Extract contact info
      const contactInfo = await extractContactInfo(page, prop.url);
      log(`  📧 Emails: ${contactInfo.emails.length} - ${contactInfo.emails.join(', ') || 'none'}`);
      log(`  📞 Phones: ${contactInfo.phones.length} - ${contactInfo.phones.join(', ') || 'none'}`);

      const lead = {
        scrapeDate: new Date().toISOString().split('T')[0],
        propertyId: prop.id,
        propertyUrl: prop.url,
        ownershipUrl: ownershipUrl,
        address: contactInfo.address || '',
        emails: contactInfo.emails,
        phones: contactInfo.phones,
        searchLocation: SEARCH_LOCATION,
        searchId: searchId
      };

      leads.push(lead);

      // Rate limiting
      if (i < propertiesToScrape.length - 1) {
        await sleep(PAGE_DELAY_MS);
      }
    }

    // Save results
    if (leads.length > 0) {
      log(`\n✅ Total leads scraped: ${leads.length}`);

      const outputData = {
        scrapeDate: new Date().toISOString(),
        location: SEARCH_LOCATION,
        searchId: searchId,
        leadCount: leads.length,
        leads: leads
      };

      fs.writeFileSync(OUTPUT_FILE, JSON.stringify(outputData, null, 2));
      log(`💾 Saved to: ${OUTPUT_FILE}`);
    } else {
      log('\n⚠️  No leads scraped.');
    }

    log('\n✅ Scraping complete!');

    return { leadCount: leads.length, outputFile: OUTPUT_FILE };

  } catch (error) {
    log(`\n❌ Error: ${error.message}`);
    log(error.stack);

    try {
      await page.screenshot({ path: '/tmp/reonomy-v4-error.png', fullPage: true });
      log('📸 Error screenshot saved: /tmp/reonomy-v4-error.png');
    } catch (e) {}

    throw error;

  } finally {
    await browser.close();
    log('\n🔚 Browser closed');
  }
}

// Run
scrapeLeads()
  .then(result => {
    log(`\n🎉 Success! ${result.leadCount} leads scraped.`);
    console.log(`\n💾 View your leads at: ${result.outputFile}`);
    process.exit(0);
  })
  .catch(error => {
    log(`\n💥 Scraper failed: ${error.message}`);
    process.exit(1);
  });