codex2

pricewatch/app/stores/cdiscount/store.py

@@ -4,6 +4,7 @@ Store Cdiscount - Parsing de produits Cdiscount.com.
 Supporte l'extraction de: titre, prix, SKU, images, specs, etc.
 """
 
+import json
 import re
 from datetime import datetime
 from pathlib import Path
@@ -21,6 +22,7 @@ from pricewatch.app.core.schema import (
     StockStatus,
 )
 from pricewatch.app.stores.base import BaseStore
+from pricewatch.app.stores.price_parser import parse_price_text
 
 logger = get_logger("stores.cdiscount")
 
@@ -112,6 +114,8 @@ class CdiscountStore(BaseStore):
         images = self._extract_images(soup, debug_info)
         category = self._extract_category(soup, debug_info)
         specs = self._extract_specs(soup, debug_info)
+        description = self._extract_description(soup, debug_info)
+        msrp = self._extract_msrp(soup, debug_info)
         reference = self.extract_reference(url) or self._extract_sku_from_html(soup)
 
         # Déterminer le statut final
@@ -130,8 +134,10 @@ class CdiscountStore(BaseStore):
             stock_status=stock_status,
             reference=reference,
             category=category,
+            description=description,
             images=images,
             specs=specs,
+            msrp=msrp,
             debug=debug_info,
         )
 
@@ -158,6 +164,21 @@ class CdiscountStore(BaseStore):
         debug.errors.append("Titre non trouvé")
         return None
 
+    def _extract_description(self, soup: BeautifulSoup, debug: DebugInfo) -> Optional[str]:
+        """Extrait la description (meta tags)."""
+        meta = soup.find("meta", property="og:description") or soup.find(
+            "meta", attrs={"name": "description"}
+        )
+        if meta:
+            description = meta.get("content", "").strip()
+            if description:
+                return description
+        product_ld = self._find_product_ld(soup)
+        desc_ld = product_ld.get("description") if product_ld else None
+        if isinstance(desc_ld, str) and desc_ld.strip():
+            return desc_ld.strip()
+        return None
+
     def _extract_price(self, soup: BeautifulSoup, debug: DebugInfo) -> Optional[float]:
         """Extrait le prix."""
         selectors = self.get_selector("price", [])
@@ -170,20 +191,29 @@ class CdiscountStore(BaseStore):
                 # Attribut content (schema.org) ou texte
                 price_text = element.get("content") or element.get_text(strip=True)
 
-                # Extraire nombre (format: "299,99" ou "299.99")
-                match = re.search(r"(\d+)[.,]?(\d*)", price_text)
-                if match:
-                    integer_part = match.group(1)
-                    decimal_part = match.group(2) or "00"
-                    price_str = f"{integer_part}.{decimal_part}"
-                    try:
-                        return float(price_str)
-                    except ValueError:
-                        continue
+                price = parse_price_text(price_text)
+                if price is not None:
+                    return price
 
         debug.errors.append("Prix non trouvé")
         return None
 
+    def _extract_msrp(self, soup: BeautifulSoup, debug: DebugInfo) -> Optional[float]:
+        """Extrait le prix conseille."""
+        selectors = [
+            ".jsStrikePrice",
+            ".price__old",
+            ".c-price__strike",
+            ".price-strike",
+        ]
+        for selector in selectors:
+            element = soup.select_one(selector)
+            if element:
+                price = parse_price_text(element.get_text(strip=True))
+                if price is not None:
+                    return price
+        return None
+
     def _extract_currency(self, soup: BeautifulSoup, debug: DebugInfo) -> Optional[str]:
         """Extrait la devise."""
         selectors = self.get_selector("currency", [])
@@ -249,7 +279,14 @@ class CdiscountStore(BaseStore):
                         url = f"https:{url}"
                     images.append(url)
 
-        return list(set(images))  # Dédupliquer
+        ld_images = self._extract_ld_images(self._find_product_ld(soup))
+        for url in ld_images:
+            if url and url not in images:
+                if url.startswith("//"):
+                    url = f"https:{url}"
+                images.append(url)
+
+        return list(dict.fromkeys(images))  # Préserver l’ordre
 
     def _extract_category(self, soup: BeautifulSoup, debug: DebugInfo) -> Optional[str]:
         """Extrait la catégorie depuis les breadcrumbs."""
@@ -275,6 +312,53 @@ class CdiscountStore(BaseStore):
 
         return None
 
+    def _extract_json_ld_entries(self, soup: BeautifulSoup) -> list[dict]:
+        """Parse les scripts JSON-LD et retourne les objets."""
+        entries = []
+        scripts = soup.find_all("script", type="application/ld+json")
+        for script in scripts:
+            raw = script.string or script.text
+            if not raw:
+                continue
+            try:
+                payload = json.loads(raw.strip())
+            except (json.JSONDecodeError, TypeError):
+                continue
+            if isinstance(payload, list):
+                entries.extend(payload)
+            else:
+                entries.append(payload)
+        return entries
+
+    def _find_product_ld(self, soup: BeautifulSoup) -> dict:
+        """Retourne l’objet Product JSON-LD si présent."""
+        for entry in self._extract_json_ld_entries(soup):
+            if not isinstance(entry, dict):
+                continue
+            type_field = entry.get("@type") or entry.get("type")
+            if isinstance(type_field, str) and "product" in type_field.lower():
+                return entry
+        return {}
+
+    def _extract_ld_images(self, product_ld: dict) -> list[str]:
+        """Récupère les images listées dans le JSON-LD."""
+        if not product_ld:
+            return []
+        images = product_ld.get("image") or product_ld.get("images")
+        if not images:
+            return []
+        if isinstance(images, str):
+            images = [images]
+        extracted = []
+        for item in images:
+            if isinstance(item, str):
+                extracted.append(item)
+            elif isinstance(item, dict):
+                url = item.get("url")
+                if isinstance(url, str):
+                    extracted.append(url)
+        return extracted
+
     def _extract_specs(self, soup: BeautifulSoup, debug: DebugInfo) -> dict[str, str]:
         """Extrait les caractéristiques techniques."""
         specs = {}
@@ -298,6 +382,19 @@ class CdiscountStore(BaseStore):
                             if key and value:
                                 specs[key] = value
 
+        product_ld = self._find_product_ld(soup)
+        additional = product_ld.get("additionalProperty") if product_ld else None
+        if isinstance(additional, dict):
+            additional = [additional]
+        if isinstance(additional, list):
+            for item in additional:
+                if not isinstance(item, dict):
+                    continue
+                key = item.get("name") or item.get("propertyID")
+                value = item.get("value") or item.get("valueReference")
+                if key and value:
+                    specs[key] = value
+
         return specs
 
     def _extract_sku_from_html(self, soup: BeautifulSoup) -> Optional[str]: