REFACTOR external service and improve websocket training

services/external/app/jobs/initialize_data.py

@@ -0,0 +1,54 @@
+# services/external/app/jobs/initialize_data.py
+"""
+Kubernetes Init Job - Initialize 24-month historical data
+"""
+
+import asyncio
+import argparse
+import sys
+import logging
+import structlog
+
+from app.ingestion.ingestion_manager import DataIngestionManager
+from app.core.database import database_manager
+
+logger = structlog.get_logger()
+
+
+async def main(months: int = 24):
+    """Initialize historical data for all enabled cities"""
+    logger.info("Starting data initialization job", months=months)
+
+    try:
+        manager = DataIngestionManager()
+        success = await manager.initialize_all_cities(months=months)
+
+        if success:
+            logger.info("✅ Data initialization completed successfully")
+            sys.exit(0)
+        else:
+            logger.error("❌ Data initialization failed")
+            sys.exit(1)
+
+    except Exception as e:
+        logger.error("❌ Fatal error during initialization", error=str(e))
+        sys.exit(1)
+    finally:
+        await database_manager.close_connections()
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Initialize historical data")
+    parser.add_argument("--months", type=int, default=24, help="Number of months to load")
+    parser.add_argument("--log-level", default="INFO", help="Log level")
+
+    args = parser.parse_args()
+
+    # Convert string log level to logging constant
+    log_level = getattr(logging, args.log_level.upper(), logging.INFO)
+
+    structlog.configure(
+        wrapper_class=structlog.make_filtering_bound_logger(log_level)
+    )
+
+    asyncio.run(main(months=args.months))