Skip to content

Latest commit

 

History

History
292 lines (219 loc) · 8.77 KB

README.md

File metadata and controls

292 lines (219 loc) · 8.77 KB
Raw

Maatwebsite\Yamlenv

Reads env.yml and makes a validated list of variables available as environment variables.

This is package is based on vlucas/phpdotenv.

Build Status

Why choose Yaml over .env?

The benefits of using a env.yml file or similar has been proven a long time ago. The popularity of vlucas/phpdotenv and similar packages is a testament of this. There is no reason why you shouldn't use a package like that in your projects, especially if the amount of variables you need to manage is relatively small.

In case of larger, enterprise scale, applications, the amount of environment settings might very quickly become unmanageable. If it's a multi tenant system even more so. And if you're supporting different versions of the application with its own list of required environment settings, it soon becomes a necessity to automate this process.

These are largely the reasons we decided to move towards Yaml. It provides a few simple advantages over env.yml files:

  • It allows nesting, to allow grouping of certain settings ( like database connection credentials ).
  • It is easily readable by human as well as multiple automated deployment tools ( Ansible e.g. ).
  • There are at least as many ( probably more ) trustworthy packages that work with Yaml ( Symfony being the most important one )

In our initial use case it was the combination of nesting and Ansible compatibility that made us decide to build this package.

Installation with Composer

composer require maatwebsite/yamlenv ~1.0

Usage

The documentation below is, for a large part, the same as Vlucas/Dotenv. The filenames are changed of course and classnames. We tried to keep the package as close in usage to Dotenv as possible, for ease of use. So give credit to Vlucas for providing such a great base to work from!

Of course there are a few things unique to Yamlenv, and that has been added as well.

The env.yml file is generally kept out of version control since it can contain sensitive API keys and passwords. A separate env.yml.dist file can be created with all the required environment variables defined except for the sensitive ones, which are either user-supplied for their own development environments or are communicated elsewhere to project collaborators. The project collaborators then independently copy the env.yml.dist file to a local env.yml and ensure all the settings are correct for their local environment, filling in the secret keys or providing their own values when necessary. In this usage, the env.yml file should be added to the project's .gitignore file so that it will never be committed by collaborators. This usage ensures that no sensitive passwords or API keys will ever be in the version control history so there is less risk of a security breach, and production values will never have to be shared with all project collaborators.

Add your application configuration to a env.yml file in the root of your project. Make sure the env.yml file is added to your .gitignore so it is not checked-in the code

S3_BUCKET: "yamlenv"
SECRET_KEY: "secret_key"

Now create a file named env.yml.dist and check this into the project. This should have the ENV variables you need to have set, but the values should either be blank or filled with dummy data. The idea is to let people know what variables are required, but not give them the sensitive production values.

S3_BUCKET: "devbucket"
SECRET_KEY: "abc123"

You can then load env.yml in your application with:

$yamlenv = new Yamlenv\Yamlenv(__DIR__);
$yamlenv->load();

Optionally you can pass in a filename as the second parameter, if you would like to use something other than env.yml

$yamlenv = new Yamlenv\Yamlenv(__DIR__, 'myconfig');
$yamlenv->load();

All of the defined variables are now accessible with the getenv method, and are available in the $_ENV and $_SERVER super-globals.

$s3_bucket = getenv('S3_BUCKET');
$s3_bucket = $_ENV['S3_BUCKET'];
$s3_bucket = $_SERVER['S3_BUCKET'];

You should also be able to access them using your framework's Request class (if you are using a framework).

$s3_bucket = $request->env('S3_BUCKET');
$s3_bucket = $request->getEnv('S3_BUCKET');
$s3_bucket = $request->server->get('S3_BUCKET');
$s3_bucket = env('S3_BUCKET');

Uppercase keys

With Yamlenv you can pass boolean true as the third contructor parameter. This will make sure all your keys will be cast to uppercase

s3_bucket: "will_be_uppercase"
$yamlenv = new Yamlenv\Yamlenv(__DIR__, 'emv.yml', true);
$yamlenv->load();
$s3_bucket = getenv('S3_BUCKET'); // return will_be_uppercase
$s3_bucket = getenv('s3_bucket'); // returns null

Nesting Variables

Like with Dotenv it's possible to nest an environment variable within another. However, because we are using the Yaml format, nesting us supported natively.

DB:
  USER: username
  PASS: password
  HOST: localhost

These variables will be flattened into a single level before being added to the environment variables. The different keys will be concatenated into a single key, separated with underscores. So the above example wil give the same results as below:

DB_USER: username
DB_PASS: password
DB_HOST: localhost

This also works with multiple levels:

FOO:
  BAR: 
    LOREM:
        IPSUM: multilevelfun
FOO_BAR_LOREM_IPSUM: multilevelfun

Immutability

By default, Yamlenv will NOT overwrite existing environment variables that are already set in the environment.

If you want Yamlenv to overwrite existing environment variables, use overload instead of load:

$yamlenv = new Yamlenv\Yamlenv(__DIR__);
$yamlenv->overload();

Requiring Variables to be Set

Using Yamlenv, you can require specific ENV vars to be defined, and throw an Exception if they are not. This is particularly useful to let people know any explicit required variables that your app will not work without.

You can use a single string:

$yamlenv->required('DATABASE_DSN');

Or an array of strings:

$yamlenv->required(['DB_HOST', 'DB_NAME', 'DB_USER', 'DB_PASS']);

If any ENV vars are missing, Yamlenv will throw a RuntimeException like this:

One or more environment variables failed assertions: DATABASE_DSN is missing

Empty Variables

Beyond simply requiring a variable to be set, you might also need to ensure the variable is not empty:

$yamlenv->required('DATABASE_DSN')->notEmpty();

If the environment variable is empty, you'd get an Exception:

One or more environment variables failed assertions: DATABASE_DSN is empty

Integer Variables

You might also need to ensure the the variable is of an integer value. You may do the following:

$yamlenv->required('FOO')->isInteger();

If the environment variable is not an integer, you'd get an Exception:

One or more environment variables failed assertions: FOO is not an integer

Allowed Values

It is also possible to define a set of values that your environment variable should be. This is especially useful in situations where only a handful of options or drivers are actually supported by your code:

$yamlenv->required('SESSION_STORE')->allowedValues(['Filesystem', 'Memcached']);

If the environment variable wasn't in this list of allowed values, you'd get a similar Exception:

One or more environment variables failed assertions: SESSION_STORE is not an
allowed value

Comments

You can comment your env.yml file using the # character. E.g. This follows the normal Yaml syntax rules

# this is a comment
VAR: "value" # comment
VAR: value # comment

When nesting variables, it is important to always use key->value based children. While the below is valid Yaml, it does not result in usable variables.

FOO:
  - one
  - two

Correct would be:

FOO:
  BAR: one
  BAZ: two

Usage Notes

When a new developer clones your codebase, they will have an additional one-time step to manually copy the env.yml.dist file to env.yml and fill-in their own values (or get any sensitive values from a project co-worker).

Yamlenv is made for development environments, and generally should not be used in production. In production, the actual environment variables should be set so that there is no overhead of loading the env.yml file on each request. This can be achieved via an automated deployment process with tools like Vagrant, chef, or Puppet, or can be set manually with cloud hosts like Pagodabox and Heroku.