No release in over 3 years
Low commit activity in last 3 years
An implementation of role-based policies and operations to help controllers lose weight.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Project Readme

skinny_controllers

skinny_controllers is a thin layer on top of rails with the goal of allowing for much easier unit-testability, inspired by ember

A demo app can be found in the spec here.

Join the chat at https://gitter.im/NullVoxPopuli/skinny_controllers Gem Version Build Status Code Climate Test Coverage Dependency Status

An implementation of role-based policies and operations to help controllers lose weight.

The goal of this project is to help API apps be more slim, and separate logic as much as possible.

If you have an idea or suggestion for improved defaults, please submit an issue or pull request. :-)

Overview

  • Controllers only contain render logic. Typically, render json: model
  • Business logic is encapsulated in operations.
    • Without creating any new classes, render json: model will give you default CRUD functionality in your controller actions.
    • There is one operation per controller action.
  • Policies help determine whether or not the current_user is allowed to perform an action on an object.
  • Click here to see how this is different from trailblazer

Installation

gem 'skinny_controllers'

or

gem install skinny_controllers

Usage

In a controller:

include SkinnyControllers::Diet
# ...
# in your action
render json: model

and that's it!

What if you want to call your own operations?

Sometimes, magic is scary. You can call anything you want manually (operations and policies).

Here is an example that manually makes the call to the Host Operations and passes the subdomain parameter in to filter the Host object on the subdomain.

def show
  render json: host_from_subdomain, serializer: each_serializer
end

private

def host_from_subdomain
  @host ||= HostOperations::Read.new(current_user, params, host_params).run
end

def host_params
  params.permit(:subdomain)
end

The parameters for directly calling an operation are as follows:

# Parameter Default when directly calling an operation Implicit default Purpose
0 current_user n/a current_user the user performing the action
1 controller_params n/a params the full params hash from the controller
2 params_for_action controller_params create_params, index_params, etc e.g.: requiring a foreign key when looking up index
3 action controller_params[:action] action_name the name of the current action
4 options {} skinny_controllers_config options

For JSON-API

Strong parameters must be used on create/update actions.

Here is an example params method

private

def event_params
  params
    .require(:data)
    .require(:attributes)
    .permit(:name)
end

Note that we don't need the id under the data hash, because in a RESTful api, the id will be available to us through the top level params hash.


How is this different from trailblazer?

This may not be horribly apparent, but here is a table overviewing some highlevel differences

Feature - skinny_controllers trailblazer
Purpose - API - works very well with ActiveModel::Serializers General - additional features for server-side rendered views
Added Layers - Operations, Policies Operations, Policies, Forms
Validation - stay in models moved to operations via contract block
Additional objects - none contacts, representers, callbacks, cells
Rendering - done in the controller, and up to the dev to decide how that is done. ActiveModel::Serializers with JSON-API is highly recommended -
App Structure - same as rails. app/operations and app/policies are added encourages a new structure 'concepts', where cells, view templates, assets, operations, etc are all under concepts/{model-name}

Contributing

Please refer to each project's style guidelines and guidelines for submitting patches and additions. In general, we follow the "fork-and-pull" Git workflow.

  1. Fork the repo on GitHub
  2. Clone the project to your own machine
  3. Commit changes to your own branch
  4. Push your work back up to your fork
  5. Submit a Pull request so that we can review your changes