feat: Implement new product parsers for Stokomani, Auchan, Carrefour, and Gifi, and add a centralized search configuration module.

app/services/parsers/auchan_parser.py

@@ -12,55 +12,70 @@ class AuchanParser(BaseParser):
         soup = BeautifulSoup(html, "html.parser")
         results = []
         
-        # Auchan products
-        # Config: article, div[class*='product-card']
+        # Auchan products - Ultra Robust Strategy
+        # The DOM is flat and dynamic. We rely on finding product links first.
+        # Links usually contain '/p-' in the href.
         
-        cards = soup.select("article, div[class*='product-card'], div[class*='list__item']")
+        links = soup.select("a[href*='/p-']")
+        seen_urls = set()
         
-        # Fallback: try finding links directly
-        if not cards:
-            links = soup.select("a[href*='/p-']")
-            cards = []
-            for link in links:
-                parent = link.find_parent("article") or link.find_parent("div", class_=lambda x: x and "product" in x)
-                if parent:
-                    cards.append(parent)
-                else:
-                    cards.append(link.parent)
-
-        for card in cards:
+        for link in links:
             try:
-                link_el = card.select_one("a[href*='/p-'], a")
-                if not link_el:
+                href = link.get('href')
+                if not href or href in seen_urls:
                     continue
-                    
-                href = link_el.get('href')
-                if not href or '/p-' not in href:
+                
+                # Filter out non-product links if any (e.g. facets)
+                if '/p-' not in href:
                     continue
                     
+                seen_urls.add(href)
                 url = self.make_absolute_url(href)
                 
-                title_el = card.select_one("h3, div[class*='title'], span[class*='title']")
-                title = title_el.get_text(strip=True) if title_el else link_el.get_text(strip=True)
+                # Find the container: usually the link itself or a close parent
+                # We look for a parent that contains price or image info
+                container = link
+                for _ in range(5): # Go up 5 levels max
+                    parent = container.parent
+                    if not parent:
+                        break
+                    container = parent
+                    # Stop if we hit a large container or list item
+                    if container.name == 'article' or (container.get('class') and any('list__item' in c for c in container.get('class'))):
+                        break
+                    # Stop if we find a price element inside this container that isn't the link itself
+                    if container.select_one("span[class*='price'], div[class*='price']"):
+                        break
+                
+                # Title
+                title = link.get('title')
+                if not title:
+                    title_el = container.select_one("h3, div[class*='title'], span[class*='title']")
+                    if title_el:
+                        title = title_el.get_text(strip=True)
+                if not title:
+                    title = link.get_text(strip=True)
                 
                 if not title:
                     continue
 
+                # Image
                 img_url = None
-                img_el = card.select_one("img")
+                img_el = container.select_one("img")
                 if img_el:
                     img_url = self._get_image_src(img_el)
                     if img_url:
                         img_url = self.make_absolute_url(img_url)
                 
+                # Price
                 price = None
-                price_el = card.select_one("div[class*='price'], span[class*='price'], .product-price")
+                price_el = container.select_one("div[class*='price'], span[class*='price'], .product-price")
                 if price_el:
                     price = self.parse_price_text(price_el.get_text())
                 
-                # Fallback price search in text
+                # Fallback price search in text of container
                 if not price:
-                    text = card.get_text(" ", strip=True)
+                    text = container.get_text(" ", strip=True)
                     price = self.parse_price_text(text)
                 
                 results.append(ProductResult(