cleanup and fix runtime bugs

iaai_scraper/storage/db.py

@@ -8,44 +8,16 @@ from sqlalchemy import create_engine, or_, select
 from sqlalchemy.orm import Session, sessionmaker
 
 from ..core.config import Settings
-from .models import Base, Car, Image, SyncRun
+from .models import Base, Car, Image, ScrapeTask, SyncRun
 from .schemas import CarRecord
 
 logger = logging.getLogger("iaai_scraper.db")
 
 
+# Поля Car, которые приходят из CarRecord (без id, images, relationship).
 CAR_DB_FIELDS = {
-	"parser_id",
-	"brand",
-	"model",
-	"year",
-	"price",
-	"currency",
-	"mileage",
-	"country",
-	"is_sold",
-	"color",
-	"drive",
-	"gearbox",
-	"steering_wheel",
-	"body_type",
-	"engine_volume",
-	"selling_type",
-	"one_owner",
-	"new_car",
-	"is_hidden",
-	"origin",
-	"origin_url",
-	"origin_id",
-	"is_damaged",
-	"evaluation",
-	"non_smoking",
-	"rental",
-	"repair_history",
-	"slug",
-	"last_seen_at",
-	"content_hash",
-	"raw_attributes",
+	col.key for col in Car.__table__.columns
+	if col.key not in ("id",)
 }
 
 
@@ -54,11 +26,23 @@ class PersistenceService:
 	def __init__(self, settings: Settings) -> None:
 		# Инициализация engine и фабрики сессий.
 		self.settings = settings
-		self.engine = create_engine(settings.database.url, echo=settings.database.echo, future=True)
+		engine_kwargs = {
+			"echo": settings.database.echo,
+			"future": True,
+		}
+		# Пул только для PostgreSQL.
+		if "postgresql" in settings.database.url:
+			engine_kwargs["pool_size"] = settings.database.pool_size
+			engine_kwargs["max_overflow"] = settings.database.max_overflow
+		self.engine = create_engine(settings.database.url, **engine_kwargs)
 		self.session_factory = sessionmaker(bind=self.engine, expire_on_commit=False, future=True)
 
 	def create_tables(self) -> None:
-		Base.metadata.create_all(self.engine)
+		try:
+			Base.metadata.create_all(self.engine)
+		except Exception:
+			# Alembic уже создал таблицы/ENUM — пропускаем.
+			logger.debug("create_tables skipped (schema already exists)")
 
 	@contextmanager
 	def session_scope(self) -> Iterator[Session]:
@@ -181,3 +165,52 @@ class PersistenceService:
 			self._add_images(session, int(car.id), images)
 			session.flush()
 			return {"car_id": int(car.id), "images_upserted": len(images), "action": action}
+
+	# --- ScrapeTask.
+
+	def create_scrape_task(self, celery_task_id: str, task_type: str, vehicle_url: str | None = None) -> int:
+		"""Создаёт запись задачи Celery."""
+		with self.session_scope() as session:
+			task = ScrapeTask(
+				celery_task_id=celery_task_id,
+				task_type=task_type,
+				vehicle_url=vehicle_url,
+				status="pending",
+			)
+			session.add(task)
+			session.flush()
+			return int(task.id)
+
+	def update_scrape_task(self, celery_task_id: str, **kwargs) -> None:
+		"""Обновляет поля задачи по celery_task_id."""
+		with self.session_scope() as session:
+			task = session.execute(
+				select(ScrapeTask).where(ScrapeTask.celery_task_id == celery_task_id)
+			).scalars().first()
+			if task is None:
+				return
+			for key, value in kwargs.items():
+				if hasattr(task, key):
+					setattr(task, key, value)
+			session.flush()
+
+	def get_scrape_task(self, celery_task_id: str) -> dict | None:
+		"""Возвращает информацию о задаче."""
+		with self.session_scope() as session:
+			task = session.execute(
+				select(ScrapeTask).where(ScrapeTask.celery_task_id == celery_task_id)
+			).scalars().first()
+			if task is None:
+				return None
+			return {
+				"id": task.id,
+				"celery_task_id": task.celery_task_id,
+				"task_type": task.task_type,
+				"status": task.status,
+				"vehicle_url": task.vehicle_url,
+				"created_at": task.created_at.isoformat() if task.created_at else None,
+				"started_at": task.started_at.isoformat() if task.started_at else None,
+				"finished_at": task.finished_at.isoformat() if task.finished_at else None,
+				"result_summary": task.result_summary,
+				"error_message": task.error_message,
+			}