Principles
Environment variables management
Defining the variables
Defining the environment variables happens within an .env.* file created in the root of the directory preferably, following the UPPERCASE naming convention
Source control
The .env file should never be committed to any source control repository
Consuming the variables
- Using and consuming these variables from a file is not supported out of the box, for that reason, we utilize libraries to aid us in loading them and making them available for usage in our codebase, one of those libraries and the most used is dotenv which can be installed using
npm i dotenvand configured as such
import { config } from 'dotenv';
// this can take an optional config object
// like
const opts = { path: '/custom/path/to/.env' };
config(opts);
// if you do not need it in production, you might write
if (process.env.NODE_ENV !== 'production') {
config(); // you might pass the opts object
}
Variables in production
In production this library might not be needed, due to how hosting platforms pass the values to the node command to run your software, but this is unlikely depending on the path you take in deploying the software
Examples
Supplying env variables without the need for dotenv
# Here we set teh VAR value and node will make it accessible in `process.env.VAR`
VAR=value node index.js
Supplying env variables using dotenv
# Defining the variables in the `.env` file
VAR=value1
// using it in the code, example is in `index.ts` file
// this will run in development only,
// assuming there is a different way of supplying the variables to the code in production
if (process.env.NODE_ENV !== 'production') {
// this suppose the `.env` file is defined within the root of the repository
require('dotenv').config();
}
// accessing a variable
console.log(process.env.VAR); // will output `value1`