from datetime import date, datetime, timedelta, time import ujson import sqlite3 import pandas as pd import asyncio import aiohttp import pytz #Update Market Movers Price, ChangesPercentage, Volume and MarketCap regularly berlin_tz = pytz.timezone('Europe/Berlin') from dotenv import load_dotenv import os load_dotenv() api_key = os.getenv('FMP_API_KEY') def check_if_holiday(): hol1_date = datetime(2023, 5, 29) hol2_date = datetime(2023, 6, 19) hol2_next_day_date = datetime(2023,6,20) hol3_date = datetime(2023,9,4) hol3_next_day_date = datetime(2023,9,5) hol4_date = datetime(2023,11,23) hol5_date = datetime(2023,12,25) hol6_date = datetime(2024,1,1) hol7_date = datetime(2024,1,15) hol8_date = datetime(2024,2,19) current_datetime = datetime.now(berlin_tz) if current_datetime.year == hol1_date.year and current_datetime.month == hol1_date.month and current_datetime.day == hol1_date.day: holiday = 'memorial_day' elif current_datetime.year == hol2_date.year and current_datetime.month == hol2_date.month and current_datetime.day == hol2_date.day: holiday = 'independence_day' elif current_datetime.year == hol2_next_day_date.year and current_datetime.month == hol2_next_day_date.month and current_datetime.day == hol2_next_day_date.day: holiday = 'independence_day+1' elif current_datetime.year == hol3_date.year and current_datetime.month == hol3_date.month and current_datetime.day == hol3_date.day: holiday = 'labor_day' elif current_datetime.year == hol3_next_day_date.year and current_datetime.month == hol3_next_day_date.month and current_datetime.day == hol3_next_day_date.day: holiday = 'labor_day+1' elif current_datetime.year == hol4_date.year and current_datetime.month == hol4_date.month and current_datetime.day == hol4_date.day: holiday = 'thanks_giving' elif current_datetime.year == hol5_date.year and current_datetime.month == hol5_date.month and current_datetime.day == hol5_date.day: holiday = 'christmas' elif current_datetime.year == hol6_date.year and current_datetime.month == hol6_date.month and current_datetime.day == hol6_date.day: holiday = 'new_year' elif current_datetime.year == hol7_date.year and current_datetime.month == hol7_date.month and current_datetime.day == hol7_date.day: holiday = 'martin_luther_king' elif current_datetime.year == hol8_date.year and current_datetime.month == hol8_date.month and current_datetime.day == hol8_date.day: holiday = 'washington_birthday' else: holiday = None return holiday holiday = check_if_holiday() holiday_dates = { 'memorial_day': datetime(2023, 5, 26), 'independence_day': datetime(2023, 6, 16), 'independence_day+1': datetime(2023, 6, 16), 'labor_day': datetime(2023, 9, 1), 'labor_day+1': datetime(2023, 9, 1), 'thanks_giving': datetime(2023, 11, 22), 'christmas': datetime(2023, 12, 22), 'new_year': datetime(2023, 12, 29), 'martin_luther_king': datetime(2024, 1, 12), 'washington_birthday': datetime(2024,2,16) } def correct_1d_interval(): if holiday == 'memorial_day': start_date_1d = datetime(2023,5,26) elif holiday == 'independence_day' or holiday == 'independence_day+1': start_date_1d = datetime(2023, 6, 16) elif holiday == 'labor_day' or holiday == 'labor_day+1': start_date_1d = datetime(2023, 9, 1) elif holiday == 'thanks_giving': start_date_1d = datetime(2023, 11, 22) elif holiday == 'new_year': start_date_1d = datetime(2023, 12, 29) elif holiday == 'martin_luther_king': start_date_1d = datetime(2023, 1, 12) elif holiday == 'washington_birthday': start_date_1d = datetime(2024, 2, 16) else: current_time_berlin = datetime.now(berlin_tz) # Get the current weekday (Monday is 0 and Sunday is 6) current_weekday = current_time_berlin.weekday() is_afternoon = current_time_berlin.hour > 15 or (current_time_berlin.hour == 15 and current_time_berlin.minute >= 30) if current_weekday == 0: # It's Monday and before 15:30 PM start_date_1d = current_time_berlin if is_afternoon else current_time_berlin - timedelta(days=3) elif current_weekday in (5, 6): # Saturday or Sunday start_date_1d = current_time_berlin - timedelta(days=current_weekday % 5 + 1) else: start_date_1d = current_time_berlin if is_afternoon else current_time_berlin - timedelta(days=1) return start_date_1d async def get_todays_data(ticker): end_date = datetime.now(berlin_tz) current_weekday = end_date.weekday() current_time_berlin = datetime.now(berlin_tz) is_afternoon = current_time_berlin.hour > 15 or (current_time_berlin.hour == 15 and current_time_berlin.minute >= 30) start_date_1d = correct_1d_interval() if holiday in holiday_dates: if holiday in ['independence_day+1', 'labor_day+1', 'christmas_day+1'] and not is_afternoon: end_date_1d = holiday_dates[holiday] else: end_date_1d = holiday_dates[holiday] elif current_weekday == 0: # It's Monday and before 15:30 PM end_date_1d = current_time_berlin if is_afternoon else current_time_berlin - timedelta(days=3) else: end_date_1d = end_date start_date_1d = start_date_1d.strftime("%Y-%m-%d") end_date_1d = end_date_1d.strftime("%Y-%m-%d") url = f"https://financialmodelingprep.com/api/v3/historical-chart/1min/{ticker}?from={start_date_1d}&to={end_date_1d}&apikey={api_key}" df_1d = pd.DataFrame() current_date = correct_1d_interval() target_time = time(15,30) extract_date = current_date.strftime('%Y-%m-%d') async with aiohttp.ClientSession() as session: responses = await asyncio.gather(session.get(url)) for response in responses: try: json_data = await response.json() df_1d = pd.DataFrame(json_data).iloc[::-1].reset_index(drop=True) opening_price = df_1d['open'].iloc[0] df_1d = df_1d.drop(['open', 'high', 'low', 'volume'], axis=1) df_1d = df_1d.round(2).rename(columns={"date": "time", "close": "value"}) if current_weekday == 5 or current_weekday == 6: pass else: if current_date.time() < target_time: pass else: end_time = pd.to_datetime(f'{extract_date} 16:00:00') new_index = pd.date_range(start=df_1d['time'].iloc[-1], end=end_time, freq='1min') remaining_df = pd.DataFrame(index=new_index, columns=['value']) remaining_df = remaining_df.reset_index().rename(columns={"index": "time"}) remaining_df['time'] = remaining_df['time'].dt.strftime('%Y-%m-%d %H:%M:%S') remainind_df = remaining_df.set_index('time') df_1d = pd.concat([df_1d, remaining_df[1:: ]]) #To-do FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation. df_1d = ujson.loads(df_1d.to_json(orient="records")) except: df_1d = [] return df_1d async def get_jsonparsed_data(session, url): async with session.get(url) as response: data = await response.json() return data async def get_quote_of_stocks(ticker_list): ticker_str = ','.join(ticker_list) async with aiohttp.ClientSession() as session: url = f"https://financialmodelingprep.com/api/v3/quote/{ticker_str}?apikey={api_key}" async with session.get(url) as response: df = await response.json() return df async def get_gainer_loser_active_stocks(): #Database read 1y and 3y data query_fundamental_template = """ SELECT marketCap FROM stocks WHERE symbol = ? """ query_template = """ SELECT volume FROM "{ticker}" ORDER BY rowid DESC LIMIT 1 """ async with aiohttp.ClientSession(connector=aiohttp.TCPConnector(ssl=False)) as session: gainer_url = f"https://financialmodelingprep.com/api/v3/stock_market/gainers?apikey={api_key}" loser_url = f"https://financialmodelingprep.com/api/v3/stock_market/losers?apikey={api_key}" active_url = f"https://financialmodelingprep.com/api/v3/stock_market/actives?apikey={api_key}" # Gather all the HTTP requests concurrently tasks = [ get_jsonparsed_data(session, gainer_url), get_jsonparsed_data(session, loser_url), get_jsonparsed_data(session, active_url) ] gainer_json, loser_json, active_json = await asyncio.gather(*tasks) gainer_json = [{k: v for k, v in stock.items() if stock['symbol'] in symbols} for stock in gainer_json] gainer_json = [entry for entry in gainer_json if entry] loser_json = [{k: v for k, v in stock.items() if stock['symbol'] in symbols} for stock in loser_json] loser_json = [entry for entry in loser_json if entry] active_json = [{k: v for k, v in stock.items() if stock['symbol'] in symbols} for stock in active_json] active_json = [entry for entry in active_json if entry] for entry in active_json: try: symbol = entry['symbol'] query = query_template.format(ticker=symbol) fundamental_data = pd.read_sql_query(query_fundamental_template, con, params=(symbol,)) volume = pd.read_sql_query(query, con) entry['marketCap'] = int(fundamental_data['marketCap'].iloc[0]) entry['volume'] = int(volume['volume'].iloc[0]) except: entry['marketCap'] = None entry['volume'] = None active_json = sorted(active_json, key=lambda x: (x['marketCap'] >= 10**9, x['volume']), reverse=True) stocks = gainer_json[:20] + loser_json[:20] + active_json[:20] #remove change key element stocks = [{k: v for k, v in stock.items() if k != "change"} for stock in stocks] for entry in stocks: try: symbol = entry['symbol'] query = query_template.format(ticker=symbol) fundamental_data = pd.read_sql_query(query_fundamental_template, con, params=(symbol,)) volume = pd.read_sql_query(query, con) entry['marketCap'] = int(fundamental_data['marketCap'].iloc[0]) entry['volume'] = int(volume['volume'].iloc[0]) except: entry['marketCap'] = None entry['volume'] = None day_gainer_json = stocks[:20] day_loser_json = stocks[20:40] day_active_json = stocks[40:60] query_market_movers = """ SELECT gainer,loser,most_active FROM market_movers """ past_gainer = pd.read_sql_query(query_market_movers, con) gainer_json = eval(past_gainer['gainer'].iloc[0]) loser_json = eval(past_gainer['loser'].iloc[0]) active_json = eval(past_gainer['most_active'].iloc[0]) gainer_json['1D'] = day_gainer_json loser_json['1D'] = day_loser_json active_json['1D'] = day_active_json #sorted(day_active_json, key=lambda x: x.get('volume', 0) if x.get('volume') is not None else 0, reverse=True) data = {'gainers': gainer_json, 'losers': loser_json, 'active': active_json} #Extract all unique symbols from gainer,loser, active unique_symbols = set() # Iterate through time periods, categories, and symbols for time_period in data.keys(): for category in data[time_period].keys(): for stock_data in data[time_period][category]: symbol = stock_data["symbol"] unique_symbols.add(symbol) # Convert the set to a list if needed unique_symbols_list = list(unique_symbols) #Get the latest quote of all unique symbol and map it back to the original data list to update all values latest_quote = await get_quote_of_stocks(unique_symbols_list) # Updating values in the data list based on matching symbols from the quote list for time_period in data.keys(): for category in data[time_period].keys(): for stock_data in data[time_period][category]: symbol = stock_data["symbol"] quote_stock = next((item for item in latest_quote if item["symbol"] == symbol), None) if quote_stock: stock_data['price'] = quote_stock['price'] stock_data['changesPercentage'] = quote_stock['changesPercentage'] stock_data['marketCap'] = quote_stock['marketCap'] stock_data['volume'] = quote_stock['volume'] return data async def get_historical_data(): res_list = [] ticker_list = ['SPY', 'QQQ', 'DIA', 'IWM', 'IVV'] latest_quote = await get_quote_of_stocks(ticker_list) for quote in latest_quote: ticker = quote['symbol'] df = await get_todays_data(ticker) res_list.append({'symbol': ticker, 'priceData': df, 'changesPercentage': round(quote['changesPercentage'],2), 'previousClose': round(quote['previousClose'],2)}) return res_list try: con = sqlite3.connect('stocks.db') cursor = con.cursor() cursor.execute("PRAGMA journal_mode = wal") cursor.execute("SELECT DISTINCT symbol FROM stocks WHERE symbol != ?", ('%5EGSPC',)) symbols = [row[0] for row in cursor.fetchall()] data = asyncio.run(get_historical_data()) with open(f"json/mini-plots-index/data.json", 'w') as file: ujson.dump(data, file) data = asyncio.run(get_gainer_loser_active_stocks()) with open(f"json/market-movers/data.json", 'w') as file: ujson.dump(data, file) con.close() except Exception as e: print(e)