CRUD Helper

crudify method will automatically create CRUD actions that will map HTTP requests to corresponding methods on given Resource.

[ contents ↑ ]

Resource

First argument is required and should provide the CRUDified resource. Resource should respond to get and create methods. Objects created/returned by resource should respond to update/update_attributes and destroy methods.

If your resource/objects behaves differently, you can map its methods by passing them as options.

Example: - Force destroying objects

crudify ModelName, :delete => :destroy!

Example: - Using a resource that creating records by new instead of create

crudify ModelName, :post => :new

[ contents ↑ ]

Excluded Params

By default all params will be sent to resource.

Good enough, however sometimes you need to exclude some of them.

This is easily accomplished by using :exclude option.

To exclude a single param, pass it as a string.

crudify Resource, :exclude => '__stream_uuid__'

To exclude multiple params, pass them as an array.

crudify Resource, :exclude => ['__stream_uuid__', 'user']

[ contents ↑ ]

Root

By default, crudify will create actions that respond to controllers root path.

class App < E
  map '/'

  crudify SomeModel
end

This will create following actions:

• get_index
• head_index
• post_index
• put_index
• patch_index
• delete_index
• options_index

To route CRUD actions to a different path, simply pass it as second argument:

class App < E
  map '/'

  crudify UsersModel, :users
end

This will create following actions:

• get_users
• head_users
• post_users
• put_users
• patch_users
• delete_users
• options_users

IMPORTANT! When you need to override some CRUD action, define it using the corresponding verb. Verbless actions will have no effect cause verbified actions has priority over verbless ones, regardless definition order:

class App < E
  map '/'

  crudify SomeModel

  def index
    # will have no effect at all
  end

  def get_index
    # will override crudified method that responds to GET requests
  end
end

[ contents ↑ ]

Response

Crudifier will try to return the value of primary key of extracted/created/updated object.

By default :id is used as primary key.

To use a custom primary key, pass it via :pkey option:

crudify Resource, :pkey => 'prodID'

If objects created by your resource does respond to :[] and does contain the pkey column, the value of object[pkey column] will be returned.

Otherwise if objects created by your resource does respond to a method of same name as pkey, the value of object.send(pkey) will be returned.

Otherwise the object itself will be returned.

However, if you have a custom logic rather than simply return primary key, use a block.

The block will receive the object as first argument:

crudify UsersModel do |obj|
  case
    when post?, put?, patch?
      obj.id
    when head?
      last_modified obj.last_modified
    else
      content_type '.json'
      obj.to_json
  end
end

This will return object ID on POST, PUT, and PATCH requests.

On HEAD requests, the framework is always sending an empty body, so we only update the headers. This way the client may decide when to fetch the object.

On GET requests it will convert the object to JSON before it is sent to client. Also content_type is used to set proper content type.

DELETE action does not need a handler cause it ever returns an empty string.

[ contents ↑ ]

Error Handler

If your objects responds to :errors method, crudifier will try to extract and format errors accordingly.

In case of errors, crudifier will behave depending on given options and proc.

If a proc given, crudifier will NOT halt the request. It will pass error to given proc via second argument instead.

To halt the request when a proc given, set :halt_on_errors to true.

If no proc given, request will be halted unconditionally.

By default the 500 status code will be used when halting.

To use a custom status code pass it via :halt_with option.

Access Restriction

Using auth will instruct client to require authorization. Access can be restricted to some or all actions.

In example below we will restrict access to Create, Update and Delete actions:

class App < E
  # ...

  auth :post_index, :put_index, :patch_index, :delete_index do |user, pass|
    [user, pass] = ['admin', 'someReally?secretPass']
  end

  crudify ModelName
end

Now, when the client will want to POST, PUT, PATCH, DELETE, it will be asked for authorization.

And an OPTIONS request will return all actions for authorized clients and only GET, HEAD, OPTIONS for non-authorized clients.

If a root path given, crudify will create actions that responds to that root, thus actions name will contain given root.

In example below, crudify will create actions like get_users, post_users, put_users etc. That's why we should specify proper action name in auth for authorization to work:

class App < E
  # ...

  auth :post_users, :put_users, :patch_users, :delete_users do |user, pass|
    [user, pass] = ['admin', 'someReally?secretPass']
  end

  crudify UsersModel, :users
end

[ contents ↑ ]