feat: implement geocoding evaluation scripts and enhance routing service with traffic-aware path processing

2026-07-16 14:31:07 +03:00
parent 3cf87999f4
commit 3773ab56a4
20 changed files with 885 additions and 137 deletions
@@ -0,0 +1,141 @@
+import * as fs from 'fs';
+import * as path from 'path';
+import axios from 'axios';
+
+// Constants
+const API_URL = process.env.TEST_API_URL || 'https://map-saas.intaleqapp.com/api/geocoding/search';
+const API_KEY = process.env.TEST_API_KEY || 'zP9vL5mK2nQ8xR7jT4wS1yB6hG3fV0cX';
+let DATA_FILE = path.join(__dirname, '../../../data/golden_tests/queries_syria.json');
+if (!fs.existsSync(DATA_FILE) && fs.existsSync('/data/data/golden_tests/queries_syria.json')) {
+  DATA_FILE = '/data/data/golden_tests/queries_syria.json'; // Docker environment fallback
+}
+
+// Interface for Test Query
+interface TestQuery {
+  query: string;
+  expected_location: {
+    lat: number;
+    lon: number;
+  };
+  expected_category?: string;
+  tolerance_meters: number;
+}
+
+// Interface for API Result
+interface SearchResult {
+  place_id: number;
+  name: string;
+  latitude: number | string;
+  longitude: number | string;
+  category: string;
+  type: string;
+  distance?: number;
+}
+
+// Haversine Formula to calculate distance in meters
+function calculateDistance(lat1: number, lon1: number, lat2: number, lon2: number): number {
+  const R = 6371e3; // Earth radius in meters
+  const φ1 = (lat1 * Math.PI) / 180;
+  const φ2 = (lat2 * Math.PI) / 180;
+  const Δφ = ((lat2 - lat1) * Math.PI) / 180;
+  const Δλ = ((lon2 - lon1) * Math.PI) / 180;
+
+  const a =
+    Math.sin(Δφ / 2) * Math.sin(Δφ / 2) +
+    Math.cos(φ1) * Math.cos(φ2) * Math.sin(Δλ / 2) * Math.sin(Δλ / 2);
+  const c = 2 * Math.atan2(Math.sqrt(a), Math.sqrt(1 - a));
+
+  return R * c;
+}
+
+async function runEvaluation() {
+  console.log('🚀 Starting Geocoding Evaluation (Golden Test Set)...\n');
+
+  if (!fs.existsSync(DATA_FILE)) {
+    console.error(`❌ Data file not found at ${DATA_FILE}`);
+    process.exit(1);
+  }
+
+  const rawData = fs.readFileSync(DATA_FILE, 'utf8');
+  const queries: TestQuery[] = JSON.parse(rawData);
+
+  let top1Hits = 0;
+  let top3Hits = 0;
+  let zeroResults = 0;
+  let latencies: number[] = [];
+
+  for (const test of queries) {
+    console.log(`Testing query: "${test.query}"`);
+    
+    const startTime = Date.now();
+    try {
+      const response = await axios.get(API_URL, {
+        headers: { 'x-api-key': API_KEY },
+        params: { q: test.query, country: 'syria' }
+      });
+      
+      const latency = Date.now() - startTime;
+      latencies.push(latency);
+      
+      const results: SearchResult[] = response.data.results || response.data;
+      
+      if (!results || results.length === 0) {
+        console.log(`   ❌ Zero results returned.`);
+        zeroResults++;
+        continue;
+      }
+
+      let foundInTop1 = false;
+      let foundInTop3 = false;
+
+      for (let i = 0; i < Math.min(results.length, 3); i++) {
+        const res = results[i];
+        const resLat = Number(res.latitude || (res as any).lat);
+        const resLon = Number(res.longitude || (res as any).lon);
+        
+        const dist = calculateDistance(test.expected_location.lat, test.expected_location.lon, resLat, resLon);
+        const isMatch = dist <= test.tolerance_meters;
+
+        console.log(`     [Rank ${i + 1}] Name: "${res.name_ar || res.name}" | Dist: ${Math.round(dist)}m | Source: ${res.source} | Category: ${res.category}`);
+
+        if (isMatch) {
+          if (i === 0) foundInTop1 = true;
+          foundInTop3 = true;
+          console.log(`   ✅ Match found at rank ${i + 1} (Distance: ${Math.round(dist)}m)`);
+          // We found the match, but we still print the rest for debugging
+        }
+      }
+
+      if (!foundInTop3) {
+        const topDist = calculateDistance(test.expected_location.lat, test.expected_location.lon, Number(results[0].latitude || (results[0] as any).lat), Number(results[0].longitude || (results[0] as any).lon));
+        console.log(`   ❌ No match in Top-3. Top result distance: ${Math.round(topDist)}m`);
+      }
+
+      if (foundInTop1) top1Hits++;
+      if (foundInTop3) top3Hits++;
+
+    } catch (error: any) {
+      console.error(`   ❌ API Error: ${error.message}`);
+    }
+  }
+
+  // Calculate metrics
+  latencies.sort((a, b) => a - b);
+  const p50 = latencies[Math.floor(latencies.length * 0.5)] || 0;
+  const p95 = latencies[Math.floor(latencies.length * 0.95)] || 0;
+  
+  const total = queries.length;
+
+  console.log('\n=======================================');
+  console.log('📊 EVALUATION RESULTS');
+  console.log('=======================================');
+  console.log(`Total Queries   : ${total}`);
+  console.log(`Top-1 Accuracy  : ${((top1Hits / total) * 100).toFixed(1)}% (${top1Hits}/${total})`);
+  console.log(`Top-3 Accuracy  : ${((top3Hits / total) * 100).toFixed(1)}% (${top3Hits}/${total})`);
+  console.log(`Zero Results    : ${((zeroResults / total) * 100).toFixed(1)}% (${zeroResults}/${total})`);
+  console.log(`p50 Latency     : ${p50} ms`);
+  console.log(`p95 Latency     : ${p95} ms`);
+  console.log('=======================================\n');
+}
+
+runEvaluation().catch(console.error);