How to accumulate SQL data using websocket with sqlalchemy on python

Question

I am using binance websocket to create sqlite database after closing of each candle on 1min interval. For this purpose I've create the following piece of code.

import pandas as pd
from binance.client import Client
from binance import BinanceSocketManager
from binance.enums import *
import sqlalchemy
import json
import random
import datetime
import time

engine= sqlalchemy.create_engine('sqlite:///app.db')
client = Client(api_public_key, api_private_key)
bsm = BinanceSocketManager(client)
socket = bsm.kline_socket('BTCUSDT', interval=KLINE_INTERVAL_1MINUTE)

def createframe(msg):
    df = pd.DataFrame([msg['k']])
    df = df.loc[:, ['t', 'o', 'h', 'l', 'c', 'V']]
    df.columns = ['time', 'open', 'high', 'low', 'close', 'volume']
    df.astype('float')
    df['time'] = datetime.datetime.fromtimestamp(df['time'] / 1000).strftime('%Y-%m-%d %H:%M:%S')
    return df

while True:
    await socket.__aenter__()
    msg = await socket.recv()
    if msg['k']['x']:
        frame = createframe(msg)
        frame.to_sql(msg['k']['s'], engine, if_exists='append', index=False)

Json object returned for kline websocket (msg):

{'e': 'kline', 'E': 1634470338549, 's': 'BTCUSDT', 'k': {'t': 1634470320000, 'T': 1634470379999, 's': 'BTCUSDT', 'i': '1m', 'f': 1101708308, 'L': 1101708646, 'o': '61104.36000000', 'c': '61110.07000000', 'h': '61118.44000000', 'l': '61104.35000000', 'v': '5.48572000', 'n': 339, 'x': False, 'q': '335219.68678330', 'V': '3.23284000', 'Q': '197550.98708780', 'B': '0'}}

After running the code for few minutes I get the following output:

pd.read_sql('BTCUSDT',engine)



time                        open            high            low             close          volume
0   2021-10-17 06:27:00     61019.99000000  61020.00000000  61013.51000000  61015.01000000  3.07002000
1   2021-10-17 06:28:00     61015.00000000  61062.52000000  61000.01000000  61056.70000000  13.05177000
2   2021-10-17 06:28:00     61015.00000000  61062.52000000  61000.01000000  61056.70000000  13.05177000
3   2021-10-17 06:29:00     61056.71000000  61084.60000000  61053.57000000  61060.35000000  6.09200000
4   2021-10-17 06:29:00     61056.71000000  61084.60000000  61053.57000000  61060.35000000  6.09200000
5   2021-10-17 06:30:00     61060.35000000  61069.00000000  61050.00000000  61050.01000000  4.71704000
6   2021-10-17 06:30:00     61060.35000000  61069.00000000  61050.00000000  61050.01000000  4.71704000
7   2021-10-17 06:31:00     61050.01000000  61059.78000000  61000.00000000  61033.33000000  4.17527000
8   2021-10-17 06:31:00     61050.01000000  61059.78000000  61000.00000000  61033.33000000  4.17527000
9   2021-10-17 06:32:00     61033.33000000  61047.42000000  61013.00000000  61013.01000000  4.55346000
10  2021-10-17 06:33:00     61013.01000000  61040.36000000  61013.00000000  61024.17000000  12.86187000
11  2021-10-17 06:33:00     61013.01000000  61040.36000000  61013.00000000  61024.17000000  12.86187000
12  2021-10-17 06:34:00     61024.17000000  61024.18000000  60963.04000000  60967.06000000  7.99450000

Any idea on how to avoid timeframe data duplicates?