config-envy

config-envy is both a command-line tool and a module to manage sensitive configuration files in Amazon AWS S3. The command-line tool is used to pull and store the files in S3, and the module is used to consume the configuration in your application.

Setup

npm install --save-dev config-envy # to install locally and use in your package.json scripts
npm install -g config-envy # to install globally if you prefer

You need to create a .config-envyrc file to configure your available configuration lanes and settings. It should look something like this:

{
  "projectName": "yourProjectName", // This is required
  "storage": "s3", // This is required. It can be a built in storage method (like s3) or a path to your own storage method adapter
  "defaults": { // This is the default config for your all of your lanes.
    "env": "{laneKey}", // env is the value to look for in NODE_ENV to use the specified .env file
    "local": "config/{laneKey}.env", // The local path to save a lane's .env
    // Each adapter has their own set of options. s3 requires these three for each lane
    "bucket": "com.company.xyz", // The s3 bucket to upload/download from
    "region": "us-west-2", //
    "key": "{projectName}/{laneKey}/.env" // The path to save into s3
  },
  "lanes": {
    // Each config lane needs to be declared here. Override the default options in each lane as needed
    "prod": {
      "env": "production"
    },
    "stage": {
      "env": "staging"
    },
    "development": {} // This this just uses the default config
  }
}

Once that file is in place in your project, you can use the command-line to save and retrieve environment files.

You'll want to add your config files to your .gitignore. So *.env, should be good to add to your .gitignore unless you use a different extension for this config files.

Command-Line

  Usage: config-envy [options] <lane|all> [get|put]

  Options:

    -h, --help     output usage information
    -V, --version  output the version number
    -l, --list     List the available lanes
    -c, --config   Show the paths to the configuration files that are being used
    -i, --init     Initialize the local .env files

If you're starting a new project, run config-envy --init to create a bunch of blank .env files for each of your configuration lanes. If you run it more than once, it won't override the files that already exist.

Go into each of your .env files and add your environment config as such:

SUPER_SECRET_URL=http://www.google.com
MY_PASSWORD=spike

If you want to push up config files, use this command:

config-envy [lane name] put

If you want to get a config file from s3, use this command:

config-envy [lane name] get

If you want to display the config for a given lane, use this command:

config-envy [lane name]

In place of [lane name], you can use all if you want to perform actions for all of the given config lanes.

Module

The module is used to consume those .env files and put them into your process.env at runtime. Default options are shown.

require('config-envy')({
  env: process.env.NODE_ENV, // The environment config to pull from. Will be based on the `env` property of the lane
  cwd: process.cwd(), // The current working directory to base storage options from
  localEnv: '.env', // The path to an overriding .env for the local environment
  overrideProcess: false, // True makes the .env files have a higher precedence. False makes the process.env take precedence
  extendProcess: true, // True sets sets the calculated properties back to process.env.
  silent: false, // Set to true if you want to be notified of any errors that happen when trying to read your .env files
});

Storage Adapters

There are a couple storage adapters built into config-envy. "Additional required options" refer to the additional lane options that are required for that given storage method. env and local are required for all of them. You can of course set defaults and not set them in your specific lane configs.

fs

This is a storage option that saves your original .env files somewhere else on you system.

Additional required options

• remotePath - The path to save the lane's config file "remotely" (still on your file system, just in a different place than your project).

s3

This is a storage option that saves your original .env files in s3. Note that with this adapter, you need to have your local credentials file set up on your machine as outlined here.

Additional required options

• region - The s3 region where the bucket lives.
• bucket - The bucket you wish to save the file to.
• key - The key or path you wish to save the file to in the given bucket.

Custom Adapters

If you wish to use a custom adapter, put in the path to your custom adapter in the storage option in your config-envyrc.

{
  "projectName": "myProject",
  "storage": "./node_modules/custom-adapter",
  ...
}

Your custom adapter module should have several properties on it. This is an example adapter. Pay attention to the comments as they outline what do expect:

/**
 * @function download
 * This function gets called when the `get` command is sent.
 * @param {Object} options
 * @param {string} options.env - The lane's environment variable for NODE_ENV
 * @param {string} options.local - The local (relative) path to save the file
 * @param {string} options.localPath - The local (absolute) path to save the file
 * @param {string} options.* - Any additional property that you required
 * @param {string} name - The name of the lane
 * @returns {Mixed} whatever you want! It can be asynchronous or whatever.
 */
exports.download = function(options, name) {

};

/**
 * @function upload
 * This function gets called when the `put` command is sent. Note that this
 * isn't required as you may have an adapter you don't want people overwriting
 * in your remote location.
 *
 * All of the @params are the same as download
 */
exports.upload = function(options, name) {

};

/**
 * @name requiredOptions
 * An array of strings, which are the required options you want for your
 * adapter. This isn't required.
 */
exports.requiredOptions = [];