Delete documents with duplicate keys keeping 1

gistfile1.txt

# Finds documents from the "users" collection which have duplicate values for
# key "kazoo_id", and deletes as many as needed to keep only 1 document.

from pymongo.mongo_client import MongoClient
from pymongo.server_api import ServerApi

uri = "..."

client = MongoClient(uri, server_api=ServerApi('1'))
deleted = 0

try:
    db = client["cs_demo_prod"]
    collection = db["users"]

    result = collection.aggregate([
        {"$group": {
            "_id": "$kazoo_id",
            "count": {"$sum": 1},
        }},
        {"$match": {
            "count": {"$gt": 1},
            "_id": {"$ne": None}
        }}
    ])

    for doc in result:
        kazoo_id = doc["_id"]
        to_delete = doc["count"]
        for i in range(0, to_delete): # keeps 1 document of all the duplicates
            collection.delete_one({"kazoo_id": kazoo_id})
            deleted += 1



except Exception as e:
    print(e)

print("Deleted " + str(deleted) + " records")