Fix total review count detection - use robust selector on Overview tab

- Detect total BEFORE clicking reviews tab (element is on Overview)
- Use span[role="img"][aria-label*="review"] (robust, no class names)
- Extract count from aria-label (e.g., "260 reviews" → 260)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

modules/scraper_clean.py

@@ -234,6 +234,28 @@ def scrape_reviews(driver, url: str, max_reviews: int = 5000, timeout_no_new: in
             break
         time.sleep(0.01)  # 10ms - responsive but low CPU
 
+    # Extract total review count BEFORE clicking reviews tab (it's on Overview)
+    # ROBUST: Use aria-label="X reviews" on span[role="img"]
+    total_reviews = None
+    try:
+        total_reviews = driver.execute_script("""
+            // ROBUST: Find span[role="img"][aria-label*="review"] - contains "(X)" text
+            // aria-label format: "260 reviews" or "1,234 reviews"
+            var reviewSpans = document.querySelectorAll('span[role="img"][aria-label*="review"]');
+            for (var i = 0; i < reviewSpans.length; i++) {
+                var label = reviewSpans[i].getAttribute('aria-label') || '';
+                var match = label.match(/^([\\d,\\.]+)\\s*review/i);
+                if (match) {
+                    return parseInt(match[1].replace(/[,\\.]/g, ''));
+                }
+            }
+            return null;
+        """)
+        if total_reviews:
+            print(f"📊 Total reviews on page: {total_reviews}")
+    except:
+        pass
+
     # Click reviews tab - poll until found
     review_keywords = ["review", "reseña", "avis", "bewertung", "recensione", "opiniones"]
     start = time.time()
@@ -299,47 +321,6 @@ def scrape_reviews(driver, url: str, max_reviews: int = 5000, timeout_no_new: in
 
     print("✅ Found scroll container")
 
-    # Extract total review count from page (look in specific places)
-    total_reviews = None
-    try:
-        total_reviews = driver.execute_script("""
-            // Method 1: Sum up star rating counts (most accurate)
-            // Look for aria-labels like "5 stars, 171 reviews"
-            var total = 0;
-            var starLabels = document.querySelectorAll('[aria-label*="stars,"][aria-label*="review"]');
-            if (starLabels.length >= 5) {
-                for (var i = 0; i < starLabels.length; i++) {
-                    var match = starLabels[i].getAttribute('aria-label').match(/(\\d+)\\s*review/i);
-                    if (match) total += parseInt(match[1]);
-                }
-                if (total > 0) return total;
-            }
-
-            // Method 2: Look in reviews tab text (e.g., "Reviews (247)")
-            var tabs = document.querySelectorAll('button[role="tab"]');
-            for (var i = 0; i < tabs.length; i++) {
-                var text = tabs[i].textContent || '';
-                if (/review|reseña/i.test(text)) {
-                    var match = text.match(/\\(([\\d,\\.]+)\\)/);
-                    if (match) return parseInt(match[1].replace(/[,\\.]/g, ''));
-                }
-            }
-
-            // Method 3: Look for "X reviews" near rating
-            var spans = document.querySelectorAll('span');
-            for (var i = 0; i < spans.length; i++) {
-                var text = spans[i].textContent || '';
-                var match = text.match(/^([\\d,\\.]+)\\s*reviews?$/i);
-                if (match) return parseInt(match[1].replace(/[,\\.]/g, ''));
-            }
-
-            return null;
-        """)
-        if total_reviews:
-            print(f"📊 Total reviews on page: {total_reviews}")
-    except:
-        pass
-
     # PHASE 2: Inject API interceptor for scroll-loaded reviews
     print("🔌 Injecting API interceptor...")
     driver.execute_script("""