from alpaca.data.historical import CryptoHistoricalDataClient, StockHistoricalDataClient
from alpaca.data.requests import CryptoLatestTradeRequest, StockLatestTradeRequest, StockLatestBarRequest, StockTradesRequest, StockBarsRequest
from alpaca.data.enums import DataFeed
from config import API_KEY, SECRET_KEY, MAX_BATCH_SIZE
import datetime
import time
from alpaca.data import Quote, Trade, Snapshot, Bar
from alpaca.data.models import BarSet, QuoteSet, TradeSet
from alpaca.data.timeframe import TimeFrame
import mplfinance as mpf
import pandas as pd

parametry = {}

# no keys required
#client = CryptoHistoricalDataClient()
client = StockHistoricalDataClient(API_KEY, SECRET_KEY, raw_data=False)
datetime_object_from = datetime.datetime(2023, 2, 27, 18, 51, 38, tzinfo=datetime.timezone.utc)
datetime_object_to = datetime.datetime(2023, 2, 27, 21, 51, 39, tzinfo=datetime.timezone.utc)
bar_request = StockBarsRequest(symbol_or_symbols="BAC",timeframe=TimeFrame.Hour, start=datetime_object_from, end=datetime_object_to, feed=DataFeed.SIP)

bars = client.get_stock_bars(bar_request).df
#bars = bars.drop(['symbol'])

#print(bars.df.close)
bars = bars.tz_convert('America/New_York')
print(bars)
print(bars.df.columns)
#Index(['open', 'high', 'low', 'close', 'volume', 'trade_count', 'vwap'], dtype='object')
bars.df.set_index('timestamp', inplace=True)

mpf.plot(bars.df, # the dataframe containing the OHLC (Open, High, Low and Close) data
         type='candle', # use candlesticks 
         volume=True, # also show the volume
         mav=(3,6,9), # use three different moving averages
         figratio=(3,1), # set the ratio of the figure
         style='yahoo',  # choose the yahoo style
         title='Prvni chart');

# #vrací se list od dict
# print(bars["BAC"])

# # k nemu muzeme pristupovat s 
# dict = bars["BAC"]
# print(type(dict))
# print(dict[2].timestamp)

# print(dict[2].close)

# print(dict[].close)