We want our app to run in different environments. When developing we may have a different set of data, when testing we have another different set of data, which we don't want to interfere with the data we have when running on testing or even production. You definitely DON'T wanna screw up your production or staging data. Keeping all these configuration stuffs in one place is going to save you a LOT of trouble setting up again and again later on.
So first, let's define our configurations here. This is a simple one, your configurations for our apps might be much more complicated!
confs =
"development":
"db": "mongodb://localhost/myapp-dev"
"testing":
"db": "mongodb://localhost/myapp-test"
"staging":
"db": "mongodb://localhost/myapp-stag"
"production":
"db": "mongodb://localhost/myapp-prod"Then we look for the environment variable as defined in your shell. If it doesn't exist, we default it to "development".
env = process.env.NODE_ENV || "development"Then we try to get the correct environment configuration based on the env variable. If some jackass puts in a dumb env variable and it doesn't exist, meh, let's just return the development configurations.
conf = confs[env] || confs["development"]Finally we export the correct configuration settings
module.exports = conf