backend/clean_db.py

import asyncio
from src.database import engine
from sqlalchemy import text
from src.config import get_settings
from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams, PayloadSchemaType

async def main():
    async with engine.begin() as conn:
        print('Wiping Postgres data natively...')
        await conn.execute(text('DROP SCHEMA public CASCADE'))
        await conn.execute(text('CREATE SCHEMA public'))
        print('Postgres schema wiped.')

    import qdrant_client
    settings = get_settings()
    try:
        q = QdrantClient(url=settings.qdrant_url, api_key=settings.qdrant_api_key)
        q.delete_collection(settings.collection_name)
        q.create_collection(
            collection_name=settings.collection_name, 
            vectors_config=VectorParams(size=settings.vector_size, distance=Distance.COSINE)
        )
        
        # Reinject indices required natively by the pipeline
        q.create_payload_index(
            collection_name=settings.collection_name,
            field_name="session_id",
            field_schema=PayloadSchemaType.KEYWORD
        )
        q.create_payload_index(
            collection_name=settings.collection_name,
            field_name="years_of_experience",
            field_schema=PayloadSchemaType.FLOAT
        )
        print('Qdrant collection wiped and re-indexed.')
    except Exception as e:
        print('Qdrant error:', e)

    print("\n------------------------------")
    print("Database is completely purged but empty.")
    print("WARNING: You MUST now run the following command to rebuild the tables:")
    print("         alembic upgrade head")
    print("------------------------------")

if __name__ == '__main__':
    asyncio.run(main())