[FIXED] How do you make faster A.P.I calls using multithreading without using requests in Python?

Issue

I'm trying to receive historical stock data for every company in the S&P 500. The problem is that it is taking a really longtime to get the data.

from ApiStuff import ApiStuff
import fundamentalanalysis as fa
import pickle

tickers = pickle.load(open('S&P500_TICKERS.dat','rb'))

api_key = ApiStuff.api_key
data_from_tickers = []

for ticker in tickers:
    balance_sheet_annually  = fa.balance_sheet_statement(ticker, api_key, period="annual")
    data_from_tickers.append(balance_sheet_annually)

I tried searching on the internet on how to speed it up but they use other modules (i.e requests, aiohttp) for making the retrieval of the data faster and I am dependent on this module (fundamentalanalysis) to retrieve fundamental data.

Is there a way for me to still use this module and make api requests faster via the methods described?

Solution

You certainly can do this with multiple processes; concurrent.futures is made for this type of need. On the other hand, this is also a great learning opportunity for the use of open source. The source for fundamentalanalysis is available on Github. The function you're using, balance_sheet_statement, is very straightforward and basically consists of a GET request, a couple of data mappings, and the construction of a Pandas dataframe.

Replicating this logic using aiohttp or requests is going to be easier than wrangling the multiprocessing modules!

Answered By - asthasr