scripts/delete_shadow_users.py

"""
delete_shadow_users.py
======================
One-time utility to wipe all shadow users (and their listings) from MongoDB.

Run from the AIDA/ directory:
    python scripts/delete_shadow_users.py

Pass --dry-run to preview what would be deleted without actually deleting:
    python scripts/delete_shadow_users.py --dry-run
"""

import asyncio
import logging
import os
import sys

sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from app.database import connect_db, get_db, disconnect_db

logging.basicConfig(level=logging.INFO, format="%(levelname)s  %(message)s")
logger = logging.getLogger(__name__)

DRY_RUN = "--dry-run" in sys.argv


async def run():
    await connect_db()
    try:
        db = await get_db()

        # ── 1. Find all shadow users ──────────────────────────────────────
        shadow_cursor = db.users.find(
            {"account_type": "shadow"},
            {"_id": 1, "display_name": 1, "whatsapp_number": 1},
        )
        shadow_users = await shadow_cursor.to_list(length=None)

        if not shadow_users:
            logger.info("No shadow users found — nothing to delete.")
            return

        logger.info(f"Found {len(shadow_users)} shadow user(s):")
        for u in shadow_users:
            logger.info(
                f"  • {u.get('display_name', '—')}  |  {u.get('whatsapp_number', '—')}  |  id={u['_id']}"
            )

        if DRY_RUN:
            logger.info("\n[DRY RUN] No changes made. Remove --dry-run to execute.")
            return

        # ── 2. Collect their IDs ──────────────────────────────────────────
        ids = [str(u["_id"]) for u in shadow_users]

        # ── 3. Delete their listings ─────────────────────────────────────
        listings_result = await db.listings.delete_many({"owner_id": {"$in": ids}})
        logger.info(f"Deleted {listings_result.deleted_count} listing(s) owned by shadow users.")

        # ── 4. Delete the shadow users themselves ─────────────────────────
        from bson import ObjectId
        oids = [u["_id"] for u in shadow_users]
        users_result = await db.users.delete_many({"_id": {"$in": oids}})
        logger.info(f"Deleted {users_result.deleted_count} shadow user(s).")

        logger.info("\n✅ Done. All shadow users and their listings have been removed.")

    finally:
        await disconnect_db()


if __name__ == "__main__":
    asyncio.run(run())