import schedule
import time
import subprocess
import logging
from datetime import datetime

# Setup basic logging for the scheduler
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - SCHEDULER - %(message)s',
    handlers=[
        logging.FileHandler("scheduler.log"),
        logging.StreamHandler()
    ]
)

def job():
    logging.info("Starting hourly crawler execution...")
    try:
        # Run crawler.py as a separate process to ensure clean memory state
        subprocess.run(["./venv/bin/python", "crawler.py"], check=True)
        logging.info("Crawler execution finished successfully.")
    except subprocess.CalledProcessError as e:
        logging.error(f"Crawler failed with exit code: {e.returncode}")
    except Exception as e:
        logging.error(f"Failed to run crawler: {e}")

if __name__ == "__main__":
    logging.info("--- Crawler Scheduler Started ---")
    
    # Run once immediately on startup
    job()
    
    # Schedule to run every 1 hour
    schedule.every(1).hours.do(job)
    
    logging.info("Scheduled to run every 1 hour. Waiting in background...")
    
    # Keep the script running
    try:
        while True:
            schedule.run_pending()
            time.sleep(60) # Check every minute
    except KeyboardInterrupt:
        logging.info("Scheduler stopped by user.")