fix dubizzle listing extraction

dubizzle_scraper/parsing/parser.py

@@ -8,6 +8,11 @@ from ..core.utils import LOT_RE, PRICE_RE, deep_find_all_keys, deep_find_key, fi
 
 logger = logging.getLogger("dubizzle_scraper.parsers")
 
+NEXT_DATA_RE = re.compile(
+	r'<script[^>]+id=["\']__NEXT_DATA__["\'][^>]*type=["\']application/json["\'][^>]*>(.*?)</script>',
+	re.IGNORECASE | re.DOTALL,
+)
+
 
 class VehicleParser:
 	# Парсер страницы авто.
@@ -157,6 +162,11 @@ class VehicleParser:
 		network_dump = network_dump or {}
 		responses = network_dump.get("json_responses", [])
 		payloads = [item.get("payload") for item in responses if isinstance(item.get("payload"), (dict, list))]
+		embedded = self._extract_embedded_json(page_html)
+		for item in embedded:
+			payload = item.get("payload")
+			if isinstance(payload, (dict, list)):
+				payloads.append(payload)
 		dom_kv = self._parse_dom_key_value_pairs(dom_text)
 		page_title = ""
 		title_match = re.search(r"<title[^>]*>(.*?)</title>", page_html or "", re.IGNORECASE | re.DOTALL)
@@ -194,11 +204,9 @@ class VehicleParser:
 		if not summary.get("current_bid") and len(prices) > 1:
 			summary["current_bid"] = prices[1]
 
-		embedded = self._extract_embedded_json(page_html)
 		for item in embedded:
 			p = item.get("payload")
 			if isinstance(p, (dict, list)):
-				payloads.append(p)
 				# Доп. проход по JSON.
 				extra = deep_find_all_keys([p], self.SUMMARY_KEY_MAP)
 				for field, vals in extra.items():
@@ -299,6 +307,11 @@ class VehicleParser:
 
 	@staticmethod
 	def _guess_currency(summary: dict[str, Any]) -> str:
+		for key in ["currency", "price", "buy_now", "current_bid", "actual_cash_value", "estimated_repair_cost"]:
+			value = str(summary.get(key) or "")
+			upper = value.upper()
+			if "AED" in upper or "د.إ" in value:
+				return "AED"
 		for key in ["buy_now", "current_bid", "actual_cash_value", "estimated_repair_cost"]:
 			value = str(summary.get(key) or "")
 			if "$" in value:
@@ -322,9 +335,18 @@ class VehicleParser:
 	def _extract_embedded_json(html: str) -> list[dict[str, Any]]:
 		scripts = re.findall(r"<script[^>]*>(.*?)</script>", html or "", flags=re.DOTALL | re.IGNORECASE)
 		extracted: list[dict[str, Any]] = []
+		next_match = NEXT_DATA_RE.search(html or "")
+		if next_match:
+			try:
+				next_data = json.loads(html_module.unescape(next_match.group(1).strip()))
+				extracted.append({"type": "next_data", "payload": next_data})
+			except Exception:
+				pass
 		for script_text in scripts:
 			if "{" not in script_text and "[" not in script_text:
 				continue
+			if "__NEXT_DATA__" in script_text:
+				continue
 			# Пропускаем большие блоки.
 			if len(script_text) > 51_200:
 				continue