loganjhennessy/fatcat

Stock screener web app that will make me a fatcat... hah!

Fatcat

Stock screener tool to turn me into a fatcat.

Setup

Install dependencies

This app uses poetry for Python dependency version management.

# Install poetry globally
pip install poetry

# Within a virtualenv
cd fatcat
poetry install

The app includes a command-line tool to do setup and basic tasks.

Create database

To create the sqlite3 database and populate initial tables.

cd fatcat
python manage.py migrate

Populate database

This requires an API key for Financial Modeling Prep. To obtain one visit fmpcloud.io. Save that key to a text file called fmp_api_key.txt within the fatcat directory (same dir as manage.py).

The following call will use that key.

Note, be careful if you only signed up for the trial as this will burn through your limit pretty quick.

cd fatcat
python manage.py populate-companies

Local serving

# Backend
cd fatcat
uvicorn app:app

# Frontend
cd app
npm start

Todo

There is still a lot to do. Note that right now the API is not wired up to the database and instead returns one static object thousands of times.

The frontend is also not completely ready to hook up to the API anyway.

I don't think I want to start with a page listing all companies. I think the next step is to build a page that will show any one company. When you visit localhost:3000/DAL, for instance, it will show the page for DAL. The API call will go and get the data it needs from FMP if it does not already have it in the sqlite database.

This will progressively fill up the database with companies as I request them, rather than all at once. But if I change my mind and want to implement a batch job, I can.

Data for this page should include:

• Company profile information (name, description, exchange, etc.)
• Key metrics (BVPS, ROIC, etc)
• "The Big Five" calculated for 10, 5, and 1-year periods
• Sticker price
• Current price

Hypothetically, if I can generate and display this for one company, I can do it for any of them. Although, I'm sure I'll spend quite a bit of time working around corner-cases and missing data and the like.