Full Stack Web App Template - FARM Stack

This repository contains a full stack web application template that runs on the FARM solution stack, which stands for:

• Flask, a backend Python web framework.
• Amazon Web Services, a cloud service that can host software.
• React, a frontend JavaScript web framework.
• MySQL, a relational database management system.

What is FARM?

The FARM stack is more or less an extension of the LAMP stack with these notable changes:

• The middleware framework is explicit (Flask)
• Like the MEAN stack, it includes an application presentation layer (React)
• The environment which the software runs on is instead replaced by PaaS/IaaS cloud computing services (AWS)
• There is no opinion for which operating system to use

Flask

Flask is used to write the web APIs / middleware that can communicate between the web UI and the web backend / database. This is the core of the entire application that glues the pieces of the full stack application together.

Note that this particular template does not use an ORM, and instead uses the pymysql module. This means developers need to write SQL queries, but can maximize the complexity of the queries they need to run on the database.

Amazon Web Services

AWS is the cloud service which the application can be hosted on. We can host the entire software stack on the following services:

• AWS Lambda for hosting our applications on a serverless infrastructure. We can also choose to host our Flask APIs via Zappa, making deployment to production exceedingly easy.
• AWS RDS for hosting our database.
• AWS S3 for storing all sorts of other objects like images and files.

This means that developers can keep everything on one platform and minimize DevOps / IT work.

React

React is the framework which the web UI runs on. The React application may also be run on AWS Lambda with NodeJS.

This template uses JavaScript / JSX, but TypeScript may be used as well.

MySQL

MySQL is the relational database which the web backend uses. This can also be hosted on AWS through its RDS service.

How to Run Locally

To run this web application, the following needs to be done:

1. Setup a local MySQL instance using the schema in db/
2. Install modules for the API and UI
3. Run the API and UI locally

Database Setup

To set up the MySQL instance, follow these steps:

1. Create and connect to a MySQL database instance. The most common ways to do this are with MySQL Workbench or mysql on the Linux command line interface.
2. Run sample.sql within the instance. It will create a schema called SampleInventory and then initialize a table called Inventory.

API Setup

The codebase for the Flask API can be found in the api/ directory.

First-time Setup (macOS)

1. Create a Python virtual environment with virtualenv farm-stack after navigating to that directory.
2. Activate it with source farm-stack/bin/activate
3. Navigate to api/ and run pip3 install -r requirements.txt to install all Python modules necessary for the API.
4. Open dbcreds.py in the api/ directory and edit database credentials as needed.

Running the API

1. Activate the virtual environment from above.
2. Run python3 app.py to run the API on http://localhost:8080.

UI Setup

The codebase for the React UI can be found in the ui/ directory. This template uses yarn as its package manager; it is crucial that developers do not mix other package managers like npm when adding new modules.

Note that packages like react-bootstrap are missing from this template.

First-time Setup (macOS)

1. Run yarn install to install all dependencies required.

Running the UI

1. Run yarn start to run in development mode.
2. Navigate to http://localhost:3000 on the browser.