cannaiq/backend/src/dutchie-az/discovery/discovery-dt-locations-from-cities.ts

#!/usr/bin/env npx tsx
/**
 * Discovery Entrypoint: Dutchie Locations (From Cities)
 *
 * Reads from dutchie_discovery_cities (crawl_enabled = true)
 * and discovers store locations for each city.
 *
 * Geo coordinates are captured when available from Dutchie's payloads.
 *
 * Usage:
 *   npm run discovery:dt:locations
 *   npm run discovery:dt:locations -- --limit=10
 *   npm run discovery:dt:locations -- --delay=3000
 *   DATABASE_URL="..." npx tsx src/dutchie-az/discovery/discovery-dt-locations-from-cities.ts
 *
 * Options:
 *   --limit=N    Only process N cities (default: all)
 *   --delay=N    Delay between cities in ms (default: 2000)
 */

import { Pool } from 'pg';
import { DtLocationDiscoveryService } from './DtLocationDiscoveryService';

const DB_URL = process.env.DATABASE_URL || process.env.CANNAIQ_DB_URL ||
  'postgresql://dutchie:dutchie_local_pass@localhost:54320/dutchie_menus';

function parseArgs(): { limit?: number; delay?: number } {
  const args: { limit?: number; delay?: number } = {};

  for (const arg of process.argv.slice(2)) {
    const limitMatch = arg.match(/--limit=(\d+)/);
    if (limitMatch) args.limit = parseInt(limitMatch[1], 10);

    const delayMatch = arg.match(/--delay=(\d+)/);
    if (delayMatch) args.delay = parseInt(delayMatch[1], 10);
  }

  return args;
}

async function main() {
  const args = parseArgs();

  console.log('╔══════════════════════════════════════════════════╗');
  console.log('║   Dutchie Location Discovery (From Cities)       ║');
  console.log('║   Reads crawl_enabled cities, discovers stores   ║');
  console.log('╚══════════════════════════════════════════════════╝');
  console.log(`\nDatabase: ${DB_URL.replace(/:[^:@]+@/, ':****@')}`);
  if (args.limit) console.log(`City limit: ${args.limit}`);
  if (args.delay) console.log(`Delay: ${args.delay}ms`);

  const pool = new Pool({ connectionString: DB_URL });

  try {
    const { rows } = await pool.query('SELECT NOW() as time');
    console.log(`Connected at: ${rows[0].time}\n`);

    const service = new DtLocationDiscoveryService(pool);
    const result = await service.discoverAllEnabled({
      limit: args.limit,
      delayMs: args.delay ?? 2000,
    });

    console.log('\n' + '═'.repeat(50));
    console.log('SUMMARY');
    console.log('═'.repeat(50));
    console.log(`Cities processed:   ${result.totalCities}`);
    console.log(`Locations found:    ${result.totalLocationsFound}`);
    console.log(`Locations inserted: ${result.totalInserted}`);
    console.log(`Locations updated:  ${result.totalUpdated}`);
    console.log(`Locations skipped:  ${result.totalSkipped} (protected status)`);
    console.log(`Errors:             ${result.errors.length}`);
    console.log(`Duration:           ${(result.durationMs / 1000).toFixed(1)}s`);

    if (result.errors.length > 0) {
      console.log('\nErrors (first 10):');
      result.errors.slice(0, 10).forEach((e, i) => console.log(`  ${i + 1}. ${e}`));
      if (result.errors.length > 10) {
        console.log(`  ... and ${result.errors.length - 10} more`);
      }
    }

    // Get location stats including coordinates
    const stats = await service.getStats();
    console.log('\nCurrent Database Stats:');
    console.log(`  Total locations:     ${stats.total}`);
    console.log(`  With coordinates:    ${stats.withCoordinates}`);
    console.log(`  By status:`);
    stats.byStatus.forEach(s => console.log(`    ${s.status}: ${s.count}`));

    if (result.totalCities === 0) {
      console.log('\n⚠️  No crawl-enabled cities found.');
      console.log('   Seed cities first:');
      console.log('     npm run discovery:dt:cities:manual -- --city-slug=ny-hudson --city-name=Hudson --state-code=NY');
      process.exit(1);
    }

    if (result.errors.length > 0) {
      console.log('\n⚠️  Completed with errors');
      process.exit(1);
    }

    console.log('\n✅ Location discovery completed successfully');
    process.exit(0);
  } catch (error: any) {
    console.error('\n❌ Location discovery failed:', error.message);
    process.exit(1);
  } finally {
    await pool.end();
  }
}

main();