code-challenge

ReST API to serve companies daily prices and recommendations stock data, using yfinance to fetch the data.

Table of Contents (most relevant files)

• src/stock/models.py - models and their relationships;
• src/stock/views.py - how endpoints are made and processed;
• src/stock/utils.py - how scalar value mapping is done;
• src/stock/fetch_data.py - how fetching stock data is done, via yfinance;
• src/stock/management/commands/fetch_daily_data.py - command that uses fetch_data.py logic

Build environment

Dependencies

• Docker >= 19.03.8
• docker-compose >= 1.25.5

# (optional) create an alias in your .bashrc/.zshrc or in your shell instance,
# to ease commands writing:
$ alias dc="docker-compose"

# change directory inside project
$ cd code-challenge

# start docker compose services in background
$ dc up -d

# "An error occurred when reading setup.py or setup.cfg: 'Attribute' object has no attribute 'id'" -> If this error occur, while docker image is building for first time, please ignore as it will not stop the image building process.

# check services are up and running
$ dc ps

# expected output:

             Name                            Command               State           Ports
-------------------------------------------------------------------------------------------------
code-challenge_api_1   /bin/sh -c while sleep 100 ...   Up      0.0.0.0:8000->8000/tcp
code-challenge_db_1    docker-entrypoint.sh postgres    Up      5432/tcp

# (first bootstrap only) run database migrations
$ dc exec api bash
$ poetry shell
$ python manage.py migrate

# launch django web server
$ dc exec api bash
$ poetry shell
$ python manage.py runserver 0.0.0.0:8000
# once it's running try open your browser at: http://localhost:8000

# (optional) create superuser, if you want to access backoffice
$ dc exec api bash
$ poetry shell
$ python manage.py createsuperuser --username <username>
# once it's finished try open your browser at http://localhost:8000/admin to login.

How to fetch daily data

$ dc exec api bash
$ poetry shell

# learn how to use the command
$ ./manage.py fetch_daily_data --help

# fetch all daily prices, for current date, and recommendations, for all specified tickers.
# NOTE: since you are specifying the tickers, even if they're not persisted on database yet, they will be automatically persisted.
$ ./manage.py fetch_daily_data --tickers FB,AAPL,NFLX,GOOG

# fetch all daily prices and recommendations, for current date,
# for all tickers that are already persisted on database.
# NOTE: make sure you've already persisted a company in database.
$ ./manage.py fetch_daily_data

# override daily data
$ ./manage.py fetch_daily_data 2020-04-30

# override daily data with specific tickers
$ ./manage.py fetch_daily_data 2020-04-30 --tickers FB

Access ReST endpoints

# list, create, update and get companies
# (depending on the HTTP verb - You can try it via your browser, there's a graphical user interface)
$ http://localhost:8000/companies

# query dailyprices using start_date and end_date
$ http://localhost:8000/dailyprices/?start_date=2020-05-01&end_date=2020-07-30

# query dailyprices using start_date and end_date AND companies
$ http://localhost:8000/dailyprices/?start_date=2020-05-01&end_date=2020-07-30&companies=FB

# query dailyprices using start_date and end_date AND MULTIPLE companies
$ http://localhost:8000/dailyprices/?start_date=2020-05-01&end_date=2020-07-30&companies=FB&companies=AAPL

# query recommendations using start_date and end_date
$ http://localhost:8000/recommendations/?start_date=2020-05-01&end_date=2020-07-30

# query recommendations using start_date and end_date AND companies
$ http://localhost:8000/recommendations/?start_date=2020-05-01&end_date=2020-07-30&companies=FB

# query recommendations using start_date and end_date AND MULTIPLE companies
$ http://localhost:8000/recommendations/?start_date=2020-05-01&end_date=2020-07-30&companies=FB&companies=AAPL

Relevant things missing but not in code challenge scope

• Test Driven Development;
• ReST endpoints authentication;

It was not done because it wasn't requested and it would take more time to finish the code challenge. For production environment though it would be necessary, at least the authentication.

Extra: Django Pandas (Tools for working with pandas in your Django projects)

# if you want to process the DailyPrice objects in a DataFrame/DataTime object
$ dc exec api bash
$ poetry shell
$ ./manage.py shell

from stock.models import DailyPrice as d

dp_qs = d.objects.all()
dp_qs.to_timeseries(index='created_at')

# (similar) expected output
#            id  open_value  high_value  low_value  close_value    volume company
# created_at
# 2020-07-01  17      365.12      367.36     363.91       364.11  26800491    AAPL
# 2020-04-30  18      206.92      209.69     201.57       204.71  46173300      FB

# Can find more in here: https://django-pandas.readthedocs.io/en/latest/

For detailed explanation on how things work, check out Django Docs, Docker Docs.