Abeshith's picture
Scripts Stages Added
cbb53b6
"""CLI script to initialize Airflow database."""
import sys
from pathlib import Path
import subprocess
import os
# Add project root to path
project_root = Path(__file__).parent.parent
sys.path.insert(0, str(project_root))
from mlpipeline.logging.logger import get_logger
logger = get_logger(__name__)
def main():
"""Initialize Airflow database and create admin user."""
try:
airflow_home = project_root / "airflow"
# Set AIRFLOW_HOME environment variable
os.environ["AIRFLOW_HOME"] = str(airflow_home)
logger.info(f"AIRFLOW_HOME set to: {airflow_home}")
# Initialize database
logger.info("Initializing Airflow database...")
result = subprocess.run(
["airflow", "db", "init"],
capture_output=True,
text=True
)
if result.returncode != 0:
logger.error(f"Failed to initialize database: {result.stderr}")
return 1
logger.info("βœ… Database initialized successfully!")
# Create admin user
logger.info("Creating admin user...")
result = subprocess.run(
[
"airflow", "users", "create",
"--username", "admin",
"--firstname", "Admin",
"--lastname", "User",
"--role", "Admin",
"--email", "admin@example.com",
"--password", "admin"
],
capture_output=True,
text=True
)
if result.returncode != 0 and "already exists" not in result.stderr:
logger.warning(f"User creation note: {result.stderr}")
else:
logger.info("βœ… Admin user created (username: admin, password: admin)")
logger.info("\nπŸš€ Next steps:")
logger.info(" 1. Start scheduler: airflow scheduler")
logger.info(" 2. Start webserver: airflow webserver")
logger.info(" 3. Access UI: http://localhost:8080")
return 0
except FileNotFoundError:
logger.error("❌ Airflow not found. Install with: pip install apache-airflow")
return 1
except Exception as e:
logger.error(f"❌ Initialization failed: {e}")
return 1
if __name__ == "__main__":
exit(main())