update infra and schema

mobilede_scraper/storage/db.py

@@ -4,7 +4,7 @@ from contextlib import contextmanager
 from datetime import datetime, timezone
 from typing import Any, Iterator
 
-from sqlalchemy import create_engine, delete, or_, select, text, update
+from sqlalchemy import create_engine, delete, func, or_, select, text, update
 from sqlalchemy.dialects.postgresql import insert as pg_insert
 from sqlalchemy.orm import Session, sessionmaker
 
@@ -808,6 +808,34 @@ class PersistenceService:
 				if row and row[0] and row[1] and row[2]
 			]
 
+	def get_active_cars_batch_for_image_enrich(
+		self,
+		prefix: str | tuple[str, ...] = MOBILEDE_ORIGIN_PREFIXES,
+		*,
+		limit: int = 50,
+		max_existing_images: int = 1,
+	) -> list[tuple[int, str, str, int]]:
+		prefixes = (prefix,) if isinstance(prefix, str) else tuple(prefix)
+		with self.session_scope() as session:
+			image_count = func.count(Image.id)
+			result = session.execute(
+				select(Car.id, Car.origin_id, Car.origin_url, image_count.label("image_count"))
+				.outerjoin(Image, Image.car_id == Car.id)
+				.where(
+					_origin_prefix_filter(Car.origin_id, prefixes),
+					Car.is_sold == False,  # noqa: E712
+				)
+				.group_by(Car.id, Car.origin_id, Car.origin_url)
+				.having(image_count <= max(0, int(max_existing_images)))
+				.order_by(Car.last_seen_at.desc())
+				.limit(max(1, int(limit)))
+			)
+			return [
+				(int(row[0]), str(row[1]), str(row[2]), int(row[3] or 0))
+				for row in result
+				if row and row[0] and row[1] and row[2]
+			]
+
 	def mark_cars_sold_by_ids(self, car_ids: list[int], *, sold_at: datetime | None = None) -> int:
 		if not car_ids:
 			return 0