app.py

#! python
from ast import arg
from datetime import timedelta
from sys import argv
from time import strftime
import asyncio
from pprint import pprint as print
import pandas as pd
from src.polygon_util import polygon_url
from dotenv import load_dotenv
import os

load_dotenv()

TICKER = argv[2] if len(argv) > 2 else "TSLA"
START = int(argv[3]) if len(argv) > 3 else int(1)

class Bot:
  def clean(self):
    from src.polygon_util import import_options_tickers
    from src.models.contracts import Contracts
    import re, datetime
    contracts = Contracts()
    for c in contracts.search('underlying', TICKER):
      matches = re.match('O:(?P<ticker>[A-Z]{3,4})(?P<date>[0-9]+)(?P<type>[C|P])(?P<dollars>[0-9]{5})(?P<cents>[0-9]{3})', c['ticker'])
      if datetime.datetime.strptime(matches.group("date"), '%y%m%d') < datetime.datetime.today():
        print(contracts.delete(c['id']))

  def contracts(self):
    from src.polygon_util import import_options_tickers
    from src.models.contracts import Contracts
    import re, datetime

    contracts = Contracts()
    for ticker in import_options_tickers(TICKER):
      if not contracts.search("ticker", ticker, one_entry=True):
        print(f"Adding ticker: {ticker}")
      t = contracts.update({"ticker": ticker, "underlying": TICKER})
    for c in contracts.search('underlying', TICKER):
      matches = re.match('O:(?P<ticker>[A-Z]{3,4})(?P<date>[0-9]+)(?P<type>[C|P])(?P<dollars>[0-9]{5})(?P<cents>[0-9]{3})', c['ticker'])
      if datetime.datetime.strptime(matches.group("date"), '%y%m%d') < datetime.datetime.today():
        print(contracts.delete(c['id']))

  def agg(self):
    from src.models.contracts import Contracts
    from src.models.aggregrates import Aggregates
    from src.polygon_util import get_aggregate_bars, COLUMNS
    import pandas as pd
    contracts = Contracts()
    agg = Aggregates()
    import datetime
    from src.models.contracts import Contracts
    to_date = (datetime.datetime.today() - datetime.timedelta(days=START))
    #to_date = datetime.datetime.today()i
    from_date = (to_date - datetime.timedelta(days=7))
    print(f"{from_date.strftime('%D')} => {to_date.strftime('%D')}")
    # to_date = datetime.datetime.today()
    conts =  contracts.getSet(where=f'ticker LIKE "O:{TICKER}2%"', order='asc')
    put_call_ratio = {}
    prices = {}
    for c in conts:
      import re
      matches = re.match('O:(?P<ticker>[A-Z]{3,4})(?P<date>[0-9]+)(?P<type>[C|P])(?P<dollars>[0-9]{5})(?P<cents>[0-9]{3})', c['ticker'])
      data = get_aggregate_bars(c['ticker'], from_date=from_date, to_date=to_date)
      if datetime.datetime.strptime(matches.group("date"), '%y%m%d') < datetime.datetime.today(): continue
      print(c)
      if 'results' in data.keys():
        for result in data['results']:
          row = {
            'ticker': c['ticker'],
            'underlying': c['underlying']
          }
          # print(datetime.datetime.fromtimestamp(result['t'] / 1000).strftime('%Y-%m-%d'))
          result['t'] = datetime.datetime.fromtimestamp(result['t'] / 1000).strftime('%Y-%m-%d')
          for k in result.keys():
            row[COLUMNS[k]] = result[k]
          agg.update(row)

  def analyze(self):
    from src.models.aggregrates import Aggregates
    import pandas as pd
    lookback = 14
    agg = Aggregates()
    # calls = agg.getSet(where=f'ticker LIKE "O:{TICKER}22%C%"', offset=0, limit=11000, order='asc', order_by='ticker')
    # puts = agg.getSet(where=f'ticker LIKE "O:{TICKER}22%P%"', offset=0, limit=11000, order='asc', order_by='ticker')
    calls = agg.getResult(f"SELECT date, sum(volume), avg(vw) as vw, avg(high - low) as spread, avg(close), sum(number) FROM aggregates where ticker LIKE 'O:{TICKER}22%C%' GROUP BY date ORDER by date asc")
    puts = agg.getResult(f"SELECT date, sum(volume), avg(vw) as vw, avg(high - low) as spread, avg(close), sum(number) FROM aggregates where ticker LIKE 'O:{TICKER}22%P%' GROUP BY date ORDER by date asc")
    df_calls = pd.DataFrame(calls)
    df_calls['type'] = 'call'
    df_puts = pd.DataFrame(puts)
    df_puts['type'] = 'put'
    df = pd.concat([df_calls, df_puts])
    # df = df.drop(columns=['open', 'close', 'high', 'low'])
    print(df_puts)
    lookbackperiod = df_puts.iloc[-lookback:]
    cumret = (lookbackperiod.vw.pct_change() + 1).cumprod() - 1
    cum_change = cumret[cumret.last_valid_index()]

    print(cumret)

if __name__ == '__main__':
  bot = Bot()
  getattr(bot, argv[1])()