1
0
mirror of https://gitlab.com/simple-stock-bots/simple-discord-stock-bot.git synced 2025-06-15 14:56:40 +00:00
Simple-Discord-Stock-Bot/symbol_router.py

394 lines
12 KiB
Python

"""Function that routes symbols to the correct API provider.
"""
import datetime
import logging
import random
import re
import pandas as pd
import schedule
from cachetools import TTLCache, cached
from cg_Crypto import cg_Crypto
from MarketData import MarketData
from Symbol import Coin, Stock, Symbol
from typing import Dict
log = logging.getLogger(__name__)
class Router:
STOCK_REGEX = "(?:^|[^\\$])\\$([a-zA-Z.]{1,6})"
CRYPTO_REGEX = "[$]{2}([a-zA-Z]{1,20})"
trending_count: Dict[str, float] = {}
def __init__(self):
self.stock = MarketData()
self.crypto = cg_Crypto()
schedule.every().hour.do(self.trending_decay)
def trending_decay(self, decay=0.5):
"""Decays the value of each trending stock by a multiplier"""
t_copy = {}
dead_keys = []
if self.trending_count:
t_copy = self.trending_count.copy()
for key in t_copy.keys():
if t_copy[key] < 0.01:
# This just makes sure were not keeping around keys that havent been called in a very long time.
dead_keys.append(key)
else:
t_copy[key] = t_copy[key] * decay
for dead in dead_keys:
t_copy.pop(dead)
self.trending_count = t_copy.copy()
log.info("Decayed trending symbols.")
def find_symbols(self, text: str, *, trending_weight: int = 1) -> list[Stock | Symbol]:
"""Finds stock tickers starting with a dollar sign, and cryptocurrencies with two dollar signs
in a blob of text and returns them in a list.
Parameters
----------
text : str
Blob of text.
Returns
-------
list[Symbol]
List of stock symbols as Symbol objects
"""
schedule.run_pending()
symbols: list[Symbol] = []
stocks = set(re.findall(self.STOCK_REGEX, text))
for stock in stocks:
# Market data lacks tools to check if a symbol is valid.
symbols.append(Stock(stock))
coins = set(re.findall(self.CRYPTO_REGEX, text))
for coin in coins:
sym = self.crypto.symbol_list[self.crypto.symbol_list["symbol"].str.fullmatch(coin.lower(), case=False)]
if sym.empty:
log.info(f"{coin} is not in list of coins")
else:
symbols.append(Coin(sym))
if symbols:
for symbol in symbols:
self.trending_count[symbol.tag] = self.trending_count.get(symbol.tag, 0) + trending_weight
log.debug(self.trending_count)
return symbols
def status(self, bot_resp) -> str:
"""Checks for any issues with APIs.
Returns
-------
str
Human readable text on status of the bot and relevant APIs
"""
stats = f"""
Bot Status:
{bot_resp}
Stock Market Data:
{self.stock.status()}
Cryptocurrency Data:
{self.crypto.status()}
"""
log.warning(stats)
return stats
def inline_search(self, search: str, matches: int = 5) -> pd.DataFrame:
"""Searches based on the shortest symbol that contains the same string as the search.
Should be very fast compared to a fuzzy search.
Parameters
----------
search : str
String used to match against symbols.
Returns
-------
list[tuple[str, str]]
Each tuple contains: (Symbol, Issue Name).
"""
df = pd.concat([self.stock.symbol_list, self.crypto.symbol_list])
df = df[df["description"].str.contains(search, regex=False, case=False)].sort_values(
by="type_id", key=lambda x: x.str.len()
)
symbols = df.head(matches)
symbols["price_reply"] = symbols["type_id"].apply(
lambda sym: self.price_reply(self.find_symbols(sym, trending_weight=0))[0]
)
return symbols
def price_reply(self, symbols: list[Symbol]) -> list[str]:
"""Returns current market price or after hours if its available for a given stock symbol.
Parameters
----------
symbols : list
List of stock symbols.
Returns
-------
Dict[str, str]
Each symbol passed in is a key with its value being a human readable
markdown formatted string of the symbols price and movement.
"""
replies = []
for symbol in symbols:
log.info(symbol)
if isinstance(symbol, Stock):
replies.append(self.stock.price_reply(symbol))
elif isinstance(symbol, Coin):
replies.append(self.crypto.price_reply(symbol))
else:
log.info(f"{symbol} is not a Stock or Coin")
return replies
def info_reply(self, symbols: list) -> list[str]:
"""Gets information on stock symbols.
Parameters
----------
symbols : list[str]
List of stock symbols.
Returns
-------
Dict[str, str]
Each symbol passed in is a key with its value being a human readable formatted
string of the symbols information.
"""
replies = []
for symbol in symbols:
if isinstance(symbol, Stock):
replies.append(self.stock.info_reply(symbol))
elif isinstance(symbol, Coin):
replies.append(self.crypto.info_reply(symbol))
else:
log.debug(f"{symbol} is not a Stock or Coin")
return replies
def intra_reply(self, symbol: Symbol) -> pd.DataFrame:
"""Returns price data for a symbol since the last market open.
Parameters
----------
symbol : str
Stock symbol.
Returns
-------
pd.DataFrame
Returns a timeseries dataframe with high, low, and volume data if its available.
Otherwise returns empty pd.DataFrame.
"""
if isinstance(symbol, Stock):
return self.stock.intra_reply(symbol)
elif isinstance(symbol, Coin):
return self.crypto.intra_reply(symbol)
else:
log.debug(f"{symbol} is not a Stock or Coin")
return pd.DataFrame()
def chart_reply(self, symbol: Symbol) -> pd.DataFrame:
"""Returns price data for a symbol of the past month up until the previous trading days close.
Also caches multiple requests made in the same day.
Parameters
----------
symbol : str
Stock symbol.
Returns
-------
pd.DataFrame
Returns a timeseries dataframe with high, low, and volume data if its available.
Otherwise returns empty pd.DataFrame.
"""
if isinstance(symbol, Stock):
return self.stock.chart_reply(symbol)
elif isinstance(symbol, Coin):
return self.crypto.chart_reply(symbol)
else:
log.debug(f"{symbol} is not a Stock or Coin")
return pd.DataFrame()
def stat_reply(self, symbols: list[Symbol]) -> list[str]:
"""Gets key statistics for each symbol in the list
Parameters
----------
symbols : list[str]
List of stock symbols
Returns
-------
Dict[str, str]
Each symbol passed in is a key with its value being a human readable
formatted string of the symbols statistics.
"""
replies = []
for symbol in symbols:
if isinstance(symbol, Stock):
replies.append(self.stock.stat_reply(symbol))
elif isinstance(symbol, Coin):
replies.append(self.crypto.stat_reply(symbol))
else:
log.debug(f"{symbol} is not a Stock or Coin")
return replies
def cap_reply(self, symbols: list[Symbol]) -> list[str]:
"""Gets market cap for each symbol in the list
Parameters
----------
symbols : list[str]
List of stock symbols
Returns
-------
Dict[str, str]
Each symbol passed in is a key with its value being a human readable
formatted string of the symbols market cap.
"""
replies = []
for symbol in symbols:
if isinstance(symbol, Stock):
replies.append(self.stock.cap_reply(symbol))
elif isinstance(symbol, Coin):
replies.append(self.crypto.cap_reply(symbol))
else:
log.debug(f"{symbol} is not a Stock or Coin")
return replies
def spark_reply(self, symbols: list[Symbol]) -> list[str]:
"""Gets change for each symbol and returns it in a compact format
Parameters
----------
symbols : list[str]
List of stock symbols
Returns
-------
list[str]
List of human readable strings.
"""
replies = []
for symbol in symbols:
if isinstance(symbol, Stock):
replies.append(self.stock.spark_reply(symbol))
elif isinstance(symbol, Coin):
replies.append(self.crypto.spark_reply(symbol))
else:
log.debug(f"{symbol} is not a Stock or Coin")
return replies
@cached(cache=TTLCache(maxsize=1024, ttl=600))
def trending(self) -> str:
"""Checks APIs for trending symbols.
Returns
-------
list[str]
List of preformatted strings to be sent to user.
"""
# stocks = self.stock.trending()
coins = self.crypto.trending()
reply = ""
log.warning(self.trending_count)
if self.trending_count:
reply += "🔥Trending on the Stock Bot:\n`"
reply += "" * len("Trending on the Stock Bot:") + "`\n"
sorted_trending = [s[0] for s in sorted(self.trending_count.items(), key=lambda item: item[1])][::-1][0:5]
log.warning(sorted_trending)
for t in sorted_trending:
reply += self.spark_reply(self.find_symbols(t))[0] + "\n"
if coins:
reply += "\n\n🦎Trending Crypto:\n`"
reply += "" * len("Trending Crypto:") + "`\n"
for coin in coins:
reply += coin + "\n"
if "`$GME" in reply:
reply = reply.replace("🔥", "🦍")
if reply:
return reply
else:
log.warning("Failed to collect trending data.")
return "Trending data is not currently available."
def random_pick(self) -> str:
choice = random.choice(list(self.stock.symbol_list["description"]) + list(self.crypto.symbol_list["description"]))
hold = (datetime.date.today() + datetime.timedelta(random.randint(1, 365))).strftime("%b %d, %Y")
return f"{choice}\nBuy and hold until: {hold}"
def batch_price_reply(self, symbols: list[Symbol]) -> list[str]:
"""Returns current market price or after hours if its available for a given stock symbol.
Parameters
----------
symbols : list
List of stock symbols.
Returns
-------
Dict[str, str]
Each symbol passed in is a key with its value being a human readable
markdown formatted string of the symbols price and movement.
"""
replies = []
stocks = []
coins = []
for symbol in symbols:
if isinstance(symbol, Stock):
stocks.append(symbol)
elif isinstance(symbol, Coin):
coins.append(symbol)
else:
log.debug(f"{symbol} is not a Stock or Coin")
if stocks:
for stock in stocks:
replies.append(self.stock.price_reply(stock))
if coins:
replies = replies + self.crypto.batch_price(coins)
return replies