This repo contains the code for getting stocks close day prices based on the input currency. The code is written in Python and the APIs are built using FastAPI. There are also a few test cases written and can be found in test.py, these tests could be automated by using the same APIs for overall logic but for simplicity of the project and API rate limiting they are predefined.

Create a new virtual environment either using conda or python venv, and install all the requirements from requirements.txt file. After that, setup your credentials, create an .env file and add MARKETSTACK_API_KEY and EXCHANGE_RATE_API with their values.

You can test the API by Postman, run uvicorn main:app --reload command in the terminal and go to Postman, make a GET request to endpoint http:https://127.0.0.1:8000/stock-close-prices and pass the request data in JSON format.

Example Input:

    "stock_symbol": "AAPL",
    "currency": "EUR",
    "date_range": "2023/02/20-2023/03/21"
    }

After the setup, you can run pytest tests.py in your terminal to run 4 test cases. After successful setup you should be able to see passed test cases.

A quick overview of the files present in the repo:

• .gitignore: To ignore cache and environment files
• currency_data.py: Contains the logic to get the currency data from the API
• exceptions.py: Contains custom exception
• format.sh: Shell script with steps to format and run lint on all the files present
• main.py: For adding the API route
• requirements.txt: Requirements with versions used for the task
• schemas.py: Schema for request and response
• stock_close_price_api.py: API file with the endpoint calling utils and other methods
• stock_data.py: Logic for getting the stock data as well converting stock prices based on currency conversions
• tests.py: Contains test cases
• utils.py: Contains helper functions consumed by other functions

Some assumptions made are:

• The market is open on start and end dates, hence passing the date range. Method edge_case_market_closed_on_start_date in stock_data.py is something we can expand on but the initial idea is to use the http:https://api.marketstack.com/v1/eod/{previous_processed_date} endpoint to get the data of a particular day outside the date range.

• APIs return the data without limitation. Adding limits and offsets to the request calls is something that can be additionally done.

Currently we are handling multiple cases for dates possible for example:

• Month-Day-Year with leading zeros (02/17/2009)
• Day-Month-Year with leading zeros and dots as separators (17.02.2009)
• Month name Day, Year like this (February 17, 2009)

But the input is expected to be in string format separated by "-" for example 10.01.2022-10.02.2022.

A better approach could be to use datetime instead and separate inputs for start date and end date, this will also help to eliminate helper functions that are currently being used for processing dates from datetime to string and vice versa. We can modify StockPriceRequest in schemas.py to have two new fields start_date and end_date with type datetime and eliminate date_range. Also a better approach could be to make it non optional and having the user add dates in request before calling the endpoint.