`scrape-real-prices.sh`

102 lines
#!/bin/bash

# Scrape REAL price data from actual sources
# This will replace estimated data with verified market prices

echo "🔍 Scraping REAL price data from verified sources..."
echo ""

# Test scraper on one watch first
echo "Testing scraper on Speedmaster Moonwatch..."
node -e "
import PriceScraper from './utils/price-scraper.js';
import fs from 'fs';

const scraper = new PriceScraper();

const testWatch = {
  id: 'speedmaster-moonwatch-1969',
  model: 'Speedmaster Professional Moonwatch',
  reference: 'ST 105.012',
  series: 'Speedmaster'
};

console.log('\n📊 Scraping Chrono24...');
try {
  const chrono24Results = await scraper.scrapeChrono24(testWatch.reference);

  if (chrono24Results.length > 0) {
    console.log('✓ Found', chrono24Results.length, 'real listings on Chrono24:');
    chrono24Results.forEach((r, i) => {
      console.log(\`  \${i+1}. \${r.title}\`);
      console.log(\`     Price: $\${r.price?.toLocaleString() || 'N/A'}\`);
      console.log(\`     Condition: \${r.condition}\`);
    });
  } else {
    console.log('⚠️  No results from Chrono24 (may be blocked by anti-bot)');
  }
} catch (error) {
  console.error('✗ Chrono24 scraping failed:', error.message);
}

await scraper.close();
console.log('\n✓ Test complete');
" 2>&1

echo ""
echo "========================================"
echo "PROBLEM IDENTIFIED:"
echo "========================================"
echo ""
echo "The current price history data is ESTIMATED/CALCULATED, not real."
echo ""
echo "SOLUTIONS:"
echo ""
echo "1. MANUAL DATA ENTRY (Recommended for accuracy)"
echo "   - Research each watch on auction sites"
echo "   - Record actual sale prices with dates"
echo "   - Add source URLs to watches.json"
echo ""
echo "2. PAID DATA SOURCES"
echo "   - WatchCharts API (subscription required)"
echo "   - Chrono24 API (partner access needed)"
echo "   - Auction house databases"
echo ""
echo "3. WEB SCRAPING (Current approach)"
echo "   - Gets CURRENT prices only (not historical)"
echo "   - Often blocked by anti-bot measures"
echo "   - Cannot get past auction prices easily"
echo ""
echo "4. ARCHIVE.ORG / WAYBACK MACHINE"
echo "   - Access historical snapshots of dealer sites"
echo "   - Requires complex scraping logic"
echo "   - Spotty coverage"
echo ""
echo "========================================"
echo "RECOMMENDED APPROACH:"
echo "========================================"
echo ""
echo "For REAL historical data, you need:"
echo ""
echo "1. Auction records from:"
echo "   - Sotheby's past sales"
echo "   - Christie's past sales"
echo "   - Phillips auction archives"
echo "   - Antiquorum archives"
echo ""
echo "2. Dealer archives:"
echo "   - Bob's Watches sold inventory"
echo "   - Crown & Caliber past sales"
echo "   - Watchfinder & Co. archives"
echo ""
echo "3. Market data services:"
echo "   - WatchCharts ($$$ subscription)"
echo "   - Chrono24 Price Index (limited free access)"
echo ""
echo "Would you like me to:"
echo "A) Set up manual data entry templates"
echo "B) Research auction records for specific watches"
echo "C) Try archive.org scraping for historical prices"
echo "D) Create a data entry interface for you to add real sources"
echo ""