From 6038a936ec8fa1a4eaf96eaea0eec8cd1c02fd3e Mon Sep 17 00:00:00 2001
From: Silas Brito <silas.brito@br22tech.com.br>
Date: Wed, 26 Aug 2026 17:55:21 -0300
Subject: [PATCH] fix: extracao individual de nota de avaliacao (rating) e
 total de avaliacoes por estabelecimento

---
 backend/app/services/scraper_service.py | 79 ++++++++++++++++++-------
 1 file changed, 58 insertions(+), 21 deletions(-)

diff --git a/backend/app/services/scraper_service.py b/backend/app/services/scraper_service.py
index d847d32..0b65b51 100644
--- a/backend/app/services/scraper_service.py
+++ b/backend/app/services/scraper_service.py
@@ -12,6 +12,61 @@ logger = logging.getLogger(__name__)
 
 class ScraperService:
 
+    @staticmethod
+    def _extract_rating_and_reviews(detail_panel, text_fallback: str = "") -> tuple:
+        """
+        Extrai a nota de avaliação (ex: 4.8) e o total de avaliações (ex: 150)
+        específicos de cada estabelecimento no Google Maps.
+        """
+        rating = 0.0
+        total_avaliacoes = 0
+
+        if detail_panel:
+            # 1. Procurar elemento específico de nota (ex: <span aria-label="4,8 estrelas"> ou <span class="ceYyJf">)
+            star_elem = detail_panel.query_selector("[aria-label*='estrela'], [aria-label*='star'], span.ceYyJf, div.F7L8d")
+            if star_elem:
+                aria_txt = star_elem.get_attribute("aria-label") or star_elem.inner_text()
+                m = re.search(r'([1-5][.,]\d)', aria_txt)
+                if m:
+                    try:
+                        rating = float(m.group(1).replace(',', '.'))
+                    except ValueError:
+                        pass
+
+            # 2. Procurar elemento específico de total de avaliações (ex: <button aria-label="1.250 avaliações">)
+            rev_elem = detail_panel.query_selector("button[aria-label*='avaliaç'], span[aria-label*='avaliaç'], [aria-label*='avaliações']")
+            if rev_elem:
+                rev_txt = rev_elem.get_attribute("aria-label") or rev_elem.inner_text()
+                m = re.search(r'([\d.,]+)\s*avaliaç', rev_txt, re.IGNORECASE)
+                if m:
+                    try:
+                        total_avaliacoes = int(re.sub(r'\D', '', m.group(1)))
+                    except ValueError:
+                        pass
+
+        # Fallback usando o texto bruto do painel do lugar
+        if rating == 0.0 and text_fallback:
+            m = re.search(r'([1-5][.,]\d)\s*(?:estrelas?|★|\s*\(|\s*·)', text_fallback, re.IGNORECASE)
+            if not m:
+                m = re.search(r'([1-5][.,]\d)', text_fallback)
+            if m:
+                try:
+                    rating = float(m.group(1).replace(',', '.'))
+                except ValueError:
+                    pass
+
+        if total_avaliacoes == 0 and text_fallback:
+            m = re.search(r'([\d.,]+)\s*avaliaç', text_fallback, re.IGNORECASE)
+            if not m:
+                m = re.search(r'\(([\d.,]+)\)', text_fallback)
+            if m:
+                try:
+                    total_avaliacoes = int(re.sub(r'\D', '', m.group(1)))
+                except ValueError:
+                    pass
+
+        return rating, total_avaliacoes
+
     @staticmethod
     def scrape_google_maps(search_query: str, max_results: int = 15, headless: bool = True) -> List[Dict[str, Any]]:
         """
@@ -163,17 +218,8 @@ class ScraperService:
                             if site_btn:
                                 website = site_btn.get_attribute("href") or ""
 
-                            # 5. Rating e Avaliações
-                            rating = 0.0
-                            total_avaliacoes = 0
-                            rating_match = re.search(r'([1-5][.,]\d)\s*(?:\(([\d.,]+)\))?', detail_text)
-                            if rating_match:
-                                try:
-                                    rating = float(rating_match.group(1).replace(',', '.'))
-                                    if rating_match.group(2):
-                                        total_avaliacoes = int(re.sub(r'\D', '', rating_match.group(2)))
-                                except ValueError:
-                                    pass
+                            # 5. Rating e Avaliações reais por estabelecimento
+                            rating, total_avaliacoes = ScraperService._extract_rating_and_reviews(detail_panel, detail_text)
 
                             results.append({
                                 'nome_empresa': nome_empresa,
@@ -217,16 +263,7 @@ class ScraperService:
                             if site_btn:
                                 website = site_btn.get_attribute("href") or ""
 
-                            rating = 0.0
-                            total_avaliacoes = 0
-                            rating_match = re.search(r'([1-5][.,]\d)\s*(?:\(([\d.,]+)\))?', detail_text)
-                            if rating_match:
-                                try:
-                                    rating = float(rating_match.group(1).replace(',', '.'))
-                                    if rating_match.group(2):
-                                        total_avaliacoes = int(re.sub(r'\D', '', rating_match.group(2)))
-                                except ValueError:
-                                    pass
+                            rating, total_avaliacoes = ScraperService._extract_rating_and_reviews(detail_panel, detail_text)
 
                             results.append({
                                 'nome_empresa': nome_empresa,