Refactoring Cryptocurrency Data Fetching with Python: A More Efficient Approach to CryptoCompare API

The provided solution is in Python and seems to be fetching historical cryptocurrency data from the CryptoCompare API. Here’s a refactored version with some improvements:

import requests
import pandas as pd

# Define the tickers and the API endpoint
tickers = ['BTC', 'ETH', 'XRP']
url = 'https://min-api.cryptocompare.com/data/histoday'

# Create an empty dictionary to store the data
data_dict = {}

# Loop through each ticker and fetch the data
for ticker in tickers:
    # Construct the API request URL
    url += '?fsym=' + ticker + '&tsym=USD&limit=600000000000&aggregate=1&e=CCCAGG'
    
    # Send a GET request to the API endpoint
    response = requests.get(url)
    
    # If the response was successful, parse the JSON data
    if response.status_code == 200:
        data = response.json()
        
        # Extract and clean the 'time' column
        time_column = pd.to_datetime(data['Data']['time'], unit='s')
        data['Time'] = time_column
        
        # Add the cleaned data to the dictionary
        data_dict[ticker] = data
    
    else:
        print(f"Failed to retrieve data for {ticker}")

# Create a DataFrame from the dictionary values and set 'Time' as the index
dfs = [df.set_index('Time') for df in data_dict.values()]
final_df = pd.concat(dfs, axis=1)

print(final_df.head())

In this version:

We’ve created a data_dict dictionary to store the fetched data, which is then used to create a final DataFrame.
The API request URL is constructed using string formatting, and the response status code is checked for success.
Pandas’ pd.to_datetime() function is used to clean and convert the ’time’ column into a pandas datetime object.
We’ve removed the use of global variables (globals()['df_' + str(t)]) in favor of encapsulating the data within the script.

Last modified on 2023-08-01