#!/usr/bin/env node
/**
 * Reonomy Scraper v9.1 - FIXED EDITION
 *
 * Critical fix: Moved email/phone extraction logic BEFORE return statement
 * This ensures extraction code actually executes
 *
 * Usage:
 *   SEARCH_ID="504a2d13-d88f-4213-9ac6-a7c8bc7c20c6" node reonomy-scraper-v9.1-fixed.js
 *   Or set as environment variable
 */

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_ID = process.env.REONOMY_SEARCH_ID || '504a2d13-d88f-4213-9ac6-a7c8bc7c20c6';
const MAX_PROPERTIES = process.env.MAX_PROPERTIES || 20;
const HEADLESS = process.env.HEADLESS !== 'false';

const OUTPUT_FILE = path.join(__dirname, 'reonomy-leads-v9.1-fixed.json');
const LOG_FILE = path.join(__dirname, 'reonomy-scraper-v9.1-fixed.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 ALL data from Owner tab
 * CRITICAL FIX: Email/phone extraction moved BEFORE return statement
 */
async function extractOwnerTabData(page) {
  log('📊 Extracting Owner tab data...');

  // Get snapshot first
  const bodyText = await page.evaluate(() => {
    return {
      emails: [],
      phones: [],
      ownerNames: [],
      pageTitle: document.title,
      bodyTextSample: ''
    };
  });

  // Extract property ID from URL
  const propIdMatch = page.url().match(/property\/([a-f0-9-]+)/);
  const propertyId = propIdMatch ? propIdMatch[1] : '';

  // Extract property details (SF, type) from body text
  const bodyTextContent = JSON.parse(bodyText).result || '';

  // Square footage
  const sfMatch = bodyTextContent.match(/(\d+\.?\d*\s*k?\s*SF)/i);
  const squareFootage = sfMatch ? sfMatch[0] : '';

  // Property type
  const typePatterns = [
    'Warehouse', 'Office Building', 'Retail Stores', 'Industrial',
    'General Industrial', 'Medical Building', 'School', 'Religious',
    'Supermarket', 'Financial Building', 'Residential', 'Vacant Land',
    'Tax Exempt', 'Mixed Use'
  ];

  let propertyType = '';
  for (const type of typePatterns) {
    if (bodyTextContent.includes(type)) {
      propertyType = type;
      log(`   🏢 Property Type: ${type}`);
      break;
    }
  }

  // Extract owner names from page text (v9's proven approach)
  const ownerPatterns = [
    /Owns\s+(\d+)\s+properties?\s*([A-Z][a-z]+(?:\s+[A-Z][a-z]+(?:\s+(?:LLC|LLP|Inc|Corp|Co|Ltd|Partners|Housing|Properties|Realty|Estate|Investments|Management))/g,
    /Owns\s+(\d+)\s+properties?\s*([A-Z][a-z]+(?:\s+[A-Z][a-z]+(?:\s+(?:LLC|LLP|Inc|Corp|Co|Ltd|Partners|Housing|Properties|Realty|Estate|Investments|Management))/i
  ];

  let ownerNames = [];
  const ownerBodyText = JSON.parse(bodyText).result || '';

  for (const pattern of ownerPatterns) {
    const matches = ownerBodyText.match(pattern);
    if (matches) {
      matches.forEach(m => {
        const owner = typeof m === 'string' ? m : m[1];
        if (owner && owner.length > 3 && !ownerNames.includes(owner)) {
          ownerNames.push(owner);
        }
      });
    }
  }

  log(`   👤 Owners found: ${ownerNames.length}`);

  // *** CRITICAL FIX: Extract emails BEFORE return ***
  // Extract emails using mailto links (robust approach)
  const emailResult = await page.$$eval('a[href^="mailto:"]');
  const emailSet = new Set(emailResult.map(a => a.href.replace('mailto:', '')));

  // Also try email patterns in text
  const emailRegex = /[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}/g;
  const emailMatches = ownerBodyText.match(emailRegex) || [];

  emailMatches.forEach(email => {
    if (!emailSet.has(email)) {
      emailSet.add(email);
    }
  });

  // *** CRITICAL FIX: Extract phones BEFORE return ***
  // Extract phones using your CSS selector (from your inspection)
  const phoneElements = await page.$$eval('p.MuiTypography-root.jss1797.jss1798.MuiTypography-body2');
  const phoneSet = new Set(phoneElements.map(el => el.textContent.trim()).filter(text => text.length >= 10));

  // Deduplicate phones
  const phoneSetUnique = new Set();
  phoneSet.forEach(phone => {
    // Clean phone numbers (remove extra spaces, formatting)
    const cleanPhone = phone.replace(/[\s\-\(\)]/g, '');
    if (cleanPhone.length >= 10 && !phoneSetUnique.has(cleanPhone)) {
      phoneSetUnique.add(cleanPhone);
    }
  });

  const phones = Array.from(phoneSetUnique);

  log(`   📧 Emails: ${emailSet.size} found`);
  log(`   📞 Phones: ${phones.length} found`);

  // Update info object with all data
  const info = {
    propertyId,
    propertyAddress: '',
    city: '',
    state: '',
    zip: '',
    squareFootage,
    propertyType,
    ownerNames,
    emails: Array.from(emailSet),
    phones,
    pageTitle: document.title,
    bodyTextSample: ownerBodyText.substring(0, 500)
  };

  log(`  📧 Emails: ${info.emails.length} found`);
  log(`  📞 Phones: ${info.phones.length} found`);
  log(`  👤 Owners: ${info.ownerNames.length} found`);

  return info;
}

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

    links.forEach(link => {
      const href = link.href;
      const match = href.match(/property\/([a-f0-9-]+)/);
      if (match) {
        ids.push({
          id: match[1],
          url: `https://app.reonomy.com/#!/search/${window.location.href.split('/')[4]}/property/${match[1]}`
        });
      }
    });

    return ids;
  });
}

/**
 * Main scraper function
 */
async function scrapeLeads() {
  log('🚀 Starting Reonomy Scraper v9.1 (FIXED EDITION)...\n');

  // Step 1: Login to Reonomy
  log('\n🔐 Step 1: Logging in to Reonomy...');

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

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

  try {
    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(15000);

    // 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!');

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

    await sleep(3000);

    // Step 3: 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}`);

    // Step 4: Extract property IDs
    log('\n📍 Step 3: Extracting property IDs...');
    const propertyIds = await extractPropertyIds(page);
    log(`✅ Found ${propertyIds.length} property IDs`);

    if (propertyIds.length === 0) {
      throw new Error('No properties found on search page.');
    }

    // Step 5: Process each property
    const propertiesToScrape = propertyIds.slice(0, MAX_PROPERTIES);
    log(`\n📍 Step 4: Processing ${propertiesToScrape.length} properties...\n`);

    const leads = [];

    for (let i = 0; i < propertiesToScrape.length; i++) {
      const prop = propertiesToScrape[i];

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

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

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

      // Wait for Owner tab to load
      log(`  ⏳ Waiting for Owner tab to load...`);
      await sleep(8000);

      // Extract data from Owner tab
      log(`  📊 Extracting data from Owner tab...`);
      const ownerData = await extractOwnerTabData(page);

      const lead = {
        scrapeDate: new Date().toISOString().split('T')[0],
        propertyId: prop.id,
        propertyUrl: ownershipUrl,
        address: ownerData.propertyAddress || '',
        city: ownerData.city || '',
        state: ownerData.state || '',
        zip: ownerData.zip || '',
        squareFootage: ownerData.squareFootage || '',
        propertyType: ownerData.propertyType || '',
        ownerNames: ownerData.ownerNames.join('; ') || '',
        emails: ownerData.emails,
        phones: ownerData.phones,
        searchLocation: SEARCH_LOCATION,
        searchId: searchId
      };

      log(`  📧 Emails: ${lead.emails.length}`);
      log(`  📞 Phones: ${lead.phones.length}`);
      log(`  👤 Owners: ${lead.ownerNames.length}`);
      log(`  📍 Address: ${lead.propertyAddress || 'N/A'}`);

      leads.push(lead);

      // Screenshot for debugging (first 3 properties only)
      if (i < 3) {
        const screenshotPath = `/tmp/reonomy-v9.1-property-${i + 1}.png`;
        await page.screenshot({ path: screenshotPath, fullPage: false });
        log(`   📸 Screenshot saved: ${screenshotPath}`);
      }
    }

    // Step 6: Save results
    if (leads.length > 0) {
      log(`\n✅ Total leads scraped: ${leads.length}`);

      const outputData = {
        scrapeDate: new Date().toISOString(),
        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);

    // Take screenshot of error state
    try {
      await page.screenshot({ path: '/tmp/reonomy-v9.1-error.png', fullPage: true });
      log('📸 Error screenshot saved: /tmp/reonomy-v9.1-error.png');
    } catch (e) {
      log('Could not save error screenshot');
    }

    await browser.close();
    log('\n🔚 Browser closed');
    process.exit(1);
  }
}

// Run
scrapeLeads();