func_public.py

from constants import RESOLUTION
from func_utils import get_ISO_times

import pandas as pd
import numpy as np
import time
from pprint import pprint

# Get relevant time periods from ISO from and to
ISO_TIMES = get_ISO_times()

#pprint(ISO_TIMES)

# Get Candles recent
def get_candles_recent(client, market):

    # Define output
    close_prices = []

    # Protect API
    time.sleep(0.2)

    # Get data
    candles = client.public.get_candles(
        market=market,
        resolution=RESOLUTION,
        limit=100
        )

    # Structure data
    for candle in candles.data["candles"]:
        close_prices.append(candle["close"])

    # Construct and return close price series
    close_prices.reverse()
    prices_result = np.array(close_prices).astype(np.float)
    return prices_result

# Get Candles Historical
def get_candles_histoical(client, market):

    # Define output
    close_prices = []

    # Extract historical price data for each timeframe
    for timeframe in ISO_TIMES.keys():

        # Confirm times needed
        tf_ojb = ISO_TIMES[timeframe]
        from_iso = tf_ojb["from_iso"]
        to_iso = tf_ojb["to_iso"]

        # Protect rate limits
        time.sleep(0.2)

        # Get data
        candles = client.public.get_candles(
            market=market,
            resolution=RESOLUTION,
            from_iso=from_iso,
            to_iso=to_iso,
            limit=100
            )

        # Structure data
        for candle in candles.data["candles"]:
            close_prices.append({"datetime": candle["startedAt"], market: candle["close"]})

        # Construct and return DataFrame
        close_prices.reverse()
        return close_prices

# Construct market prices
def construct_market_prices(client):
    
    # Declare variables
    tradeable_markets = []
    markets = client.public.get_markets()

    # Find tradeable pairs
    for market in markets.data["markets"].keys():
        market_info = markets.data["markets"][market]
        if market_info["status"] == "ONLINE" and market_info["type"] == "PERPETUAL":
            tradeable_markets.append(market)

    # Send initial DataFrame
    close_prices = get_candles_histoical(client, tradeable_markets[0])
    
    #pprint(close_prices)

    df = pd.DataFrame(close_prices)
    df.set_index("datetime", inplace=True)

    # Append other prices to DataFrame
    # You an limit the amount to loop through here to save time in development
    # Change second number after :. Leave blank to include everything
    for market in tradeable_markets[1:]:
        close_prices_add = get_candles_histoical(client, market)
        df_add = pd.DataFrame(close_prices_add)
        df_add.set_index("datetime", inplace=True)
        df = pd.merge(df, df_add, how="outer", on="datetime", copy=False)
        del df_add

    # Check any columns with NaNs
    nans = df.columns[df.isna().any()].tolist()
    if len(nans) > 0:
        print("Dropping columns: ")
        print(nans)
        df.drop(columns=nans, inplace=True)

    # Return result
    return df