Lotus::Model

A persistence framework for Lotus.

It delivers a convenient public API to execute queries and commands against a database. The architecture eases keeping the business logic (entities) separated from details such as persistence or validations.

It implements the following concepts:

Entity - An object defined by its identity.
Repository - An object that mediates between the entities and the persistence layer.
Data Mapper - A persistence mapper that keep entities independent from database details.
Adapter – A database adapter.
Query - An object that represents a database query.

Like all the other Lotus components, it can be used as a standalone framework or within a full Lotus application.

Status

Contact

Home page: http://lotusrb.org
Mailing List: http://lotusrb.org/mailing-list
API Doc: http://rdoc.info/gems/lotus-model
Bugs/Issues: https://github.com/lotus/model/issues
Support: http://stackoverflow.com/questions/tagged/lotus-ruby
Chat: https://gitter.im/lotus/chat

Rubies

Lotus::Model supports Ruby (MRI) 2+

Installation

Add this line to your application's Gemfile:

gem 'lotus-model'

And then execute:

$ bundle

Or install it yourself as:

$ gem install lotus-model

Usage

This class provides a DSL to configure adapter, mapping and collection.

require 'lotus/model'

class User
  include Lotus::Entity
  attributes :name, :age
end

class UserRepository
  include Lotus::Repository
end

Lotus::Model.configure do
  adapter type: :sql, uri: 'postgres://localhost/database'

  mapping do
    collection :users do
      entity      User
      repository UserRepository

      attribute :id,   Integer
      attribute :name, String
      attribute :age,  Integer
    end
  end
end

Lotus::Model.load!

user = User.new(name: 'Luca', age: 32)
user = UserRepository.create(user)

puts user.id # => 1

u = UserRepository.find(user.id)
u == user # => true

Concepts

Entities

An object that is defined by its identity. See "Domain Driven Design" by Eric Evans.

An entity is the core of an application, where the part of the domain logic is implemented. It's a small, cohesive object that expresses coherent and meaningful behaviors.

It deals with one and only one responsibility that is pertinent to the domain of the application, without caring about details such as persistence or validations.

This simplicity of design allows developers to focus on behaviors, or message passing if you will, which is the quintessence of Object Oriented Programming.

require 'lotus/model'

class Person
  include Lotus::Entity
  attributes :name, :age
end

When a class includes Lotus::Entity it receives the following interface:

#id
#id=
#initialize(attributes = {})

Lotus::Entity also provides the .attributes for defining attribute accessors for the given names.

If we expand the code above in pure Ruby, it would be:

class Person
  attr_accessor :id, :name, :age

  def initialize(attributes = {})
    @id, @name, @age = attributes.values_at(:id, :name, :age)
  end
end

Lotus::Model ships Lotus::Entity for developers's convenience.

Lotus::Model depends on a narrow and well-defined interface for an Entity - #id, #id=, #initialize(attributes={}). If your object implements that interface then that object can be used as an Entity in the Lotus::Model framework.

However, we suggest to implement this interface by including Lotus::Entity, in case that future versions of the framework will expand it.

See Dependency Inversion Principle for more on interfaces.

When a class extends a Lotus::Entity class, it will also inherit its mother's attributes.

require 'lotus/model'

class Article
  include Lotus::Entity
  attributes :name
end

class RareArticle < Article
  attributes :price
end

That is, RareArticle's attributes carry over :name attribute from Article, thus is :id, :name, :price.

Repositories

An object that mediates between entities and the persistence layer. It offers a standardized API to query and execute commands on a database.

A repository is storage independent, all the queries and commands are delegated to the current adapter.

This architecture has several advantages:

Applications depend on a standard API, instead of low level details (Dependency Inversion principle)
Applications depend on a stable API, that doesn't change if the storage changes
Developers can postpone storage decisions
Confines persistence logic at a low level
Multiple data sources can easily coexist in an application

When a class includes Lotus::Repository, it will receive the following interface:

.persist(entity) – Create or update an entity
.create(entity) – Create a record for the given entity
.update(entity) – Update the record corresponding to the given entity
.delete(entity) – Delete the record corresponding to the given entity
.all - Fetch all the entities from the collection
.find - Fetch an entity from the collection by its ID
.first - Fetch the first entity from the collection
.last - Fetch the last entity from the collection
.clear - Delete all the records from the collection
.query - Fabricates a query object

A collection is a homogenous set of records. It corresponds to a table for a SQL database or to a MongoDB collection.

All the queries are private. This decision forces developers to define intention revealing API, instead of leaking storage API details outside of a repository.

Look at the following code:

ArticleRepository.where(author_id: 23).order(:published_at).limit(8)

This is bad for a variety of reasons:

The caller has an intimate knowledge of the internal mechanisms of the Repository.
The caller works on several levels of abstraction.
It doesn't express a clear intent, it's just a chain of methods.
The caller can't be easily tested in isolation.
If we change the storage, we are forced to change the code of the caller(s).

There is a better way:

require 'lotus/model'

class ArticleRepository
  include Lotus::Repository

  def self.most_recent_by_author(author, limit = 8)
    query do
      where(author_id: author.id).
        order(:published_at)
    end.limit(limit)
  end
end

This is a huge improvement, because:

The caller doesn't know how the repository fetches the entities.
The caller works on a single level of abstraction. It doesn't even know about records, only works with entities.
It expresses a clear intent.
The caller can be easily tested in isolation. It's just a matter of stubbing this method.
If we change the storage, the callers aren't affected.

Here is an extended example of a repository that uses the SQL adapter.

class ArticleRepository
  include Lotus::Repository

  def self.most_recent_by_author(author, limit = 8)
    query do
      where(author_id: author.id).
        desc(:id).
        limit(limit)
    end
  end

  def self.most_recent_published_by_author(author, limit = 8)
    most_recent_by_author(author, limit).published
  end

  def self.published
    query do
      where(published: true)
    end
  end

  def self.drafts
    exclude published
  end

  def self.rank
    published.desc(:comments_count)
  end

  def self.best_article_ever
    rank.limit(1)
  end

  def self.comments_average
    query.average(:comments_count)
  end
end

You can also extract the common logic from your repository into a module to reuse it in other repositories. Here is a pagination example:

module RepositoryHelpers
  module Pagination
    def paginate(limit: 10, offset: 0)
      query do
        limit(limit).offset(offset)
      end
    end
  end
end

class ArticleRepository
  include Lotus::Repository
  extend RepositoryHelpers::Pagination

  def self.published
    query do
      where(published: true)
    end
  end

  # other repository-specific methods here
end

That will allow .paginate usage on ArticleRepository, for example: ArticleRepository.published.paginate(15, 0)

Your models and repositories have to be in the same namespace. Otherwise Lotus::Model::Mapper#load! will not initialize your repositories correctly.

class MyLotusApp::Model::User
  include Lotus::Entity
  # your code here
end

# This repository will work...
class MyLotusApp::Model::UserRepository
  include Lotus::Repository
  # your code here
end

# ...this will not!
class MyLotusApp::Repository::UserRepository
  include Lotus::Repository
  # your code here
end

Data Mapper

A persistence mapper that keeps entities independent from database details. It is database independent, it can work with SQL, document, and even with key/value stores.

The role of a data mapper is to translate database columns into the corresponding attribute of an entity.

require 'lotus/model'

mapper = Lotus::Model::Mapper.new do
  collection :users do
    entity User

    attribute :id,   Integer
    attribute :name, String
    attribute :age,  Integer
  end
end

For simplicity's sake, imagine that the mapper above is used with a SQL database. We use #collection to indicate the name of the table that we want to map, #entity to indicate the class that we want to associate. In the end, each call to #attribute associates the specified column with a corresponding Ruby type.

For advanced mapping and legacy databases, please have a look at the API doc.

Known limitations

Note there are limitations with inherited entities:

require 'lotus/model'

class Article
  include Lotus::Entity
  attributes :name
end

class RareArticle < Article
  attributes :price
end

mapper = Lotus::Model::Mapper.new do
  collection :articles do
    entity Article

    attribute :id,    Integer
    attribute :name,  String
    attribute :price, Integer
  end
end

In the example above, there are a few problems:

Article could not be fetched because mapping could not map price.
Finding a persisted RareArticle record, for eg. ArticleRepository.find(123), the result is an Article not RareArticle.

Adapter

An adapter is a concrete implementation of persistence logic for a specific database. Lotus::Model is shipped with three adapters:

SqlAdapter
MemoryAdapter
FileSystemAdapter

An adapter can be associated with one or multiple repositories.

require 'pg'
require 'lotus/model'
require 'lotus/model/adapters/sql_adapter'

mapper = Lotus::Model::Mapper.new do
  # ...
end

adapter = Lotus::Model::Adapters::SqlAdapter.new(mapper, 'postgres://host:port/database')

PersonRepository.adapter  = adapter
ArticleRepository.adapter = adapter

In the example above, we reuse the adapter because the target tables (people and articles) are defined in the same database. As rule of thumb, one adapter instance per database.

Query

An object that implements an interface for querying the database. This interface may vary, according to the adapter's specifications.

Here is common interface for existing class:

.all - Resolves the query by fetching records from the database and translating them into entities
.where, .and - Adds a condition that behaves like SQL WHERE
.or - Adds a condition that behaves like SQL OR
.exclude, .not - Logical negation of a #where condition
.select - Selects only the specified columns
.order, .asc - Specify the ascending order of the records, sorted by the given columns
.reverse_order, .desc - Specify the descending order of the records, sorted by the given columns
.limit - Limit the number of records to return
.offset - Specify an OFFSET clause. Due to SQL syntax restriction, offset MUST be used with #limit
.sum - Returns the sum of the values for the given column
.average, .avg - Returns the average of the values for the given column
.max - Returns the maximum value for the given column
.min - Returns the minimum value for the given column
.interval - Returns the difference between the MAX and MIN for the given column
.range - Returns a range of values between the MAX and the MIN for the given column
.exist? - Checks if at least one record exists for the current conditions
.count - Returns a count of the records for the current conditions

If you need more information regarding those methods, you can use comments from memory or sql adapters interface.

Think of an adapter for Redis, it will probably employ different strategies to filter records than an SQL query object.

Conventions

A repository must be named after an entity, by appending "Repository" to the entity class name (eg. Article => ArticleRepository).

Configurations

Non-standard repository can be configured for an entity, by setting repository on the collection.

  require 'lotus/model'

  mapper = Lotus::Model::Mapper.new do
    collection :users do
      entity User
      repository EmployeeRepository
    end
  end

Thread safety

Lotus::Model's is thread safe during the runtime, but it isn't during the loading process. The mapper compiles some code internally, be sure to safely load it before your application starts.

Mutex.new.synchronize do
  Lotus::Model.load!
end

This is not necessary, when Lotus::Model is used within a Lotus application.

Features

Timestamps

If an entity has the following accessors: :created_at and :updated_at, they will be automatically updated when the entity is persisted.

require 'lotus/model'

class User
  include Lotus::Entity
  attributes :name, :created_at, :updated_at
end

class UserRepository
  include Lotus::Repository
end

Lotus::Model.configure do
  adapter type: :memory, uri: 'memory://localhost/timestamps'

  mapping do
    collection :users do
      entity     User
      repository UserRepository

      attribute :id,         Integer
      attribute :name,       String
      attribute :created_at, DateTime
      attribute :updated_at, DateTime
    end
  end
end.load!

user = User.new(name: 'L')
puts user.created_at # => nil
puts user.updated_at # => nil

user = UserRepository.create(user)
puts user.created_at.to_s # => "2015-05-15T10:12:20+00:00"
puts user.updated_at.to_s # => "2015-05-15T10:12:20+00:00"

sleep 3
user.name = "Luca"
user      = UserRepository.update(user)
puts user.created_at.to_s # => "2015-05-15T10:12:20+00:00"
puts user.updated_at.to_s # => "2015-05-15T10:12:23+00:00"

Dirty Tracking

Entities are able to track changes of their data, if Lotus::Entity::DirtyTracking is included.

require 'lotus/model'

class User
  include Lotus::Entity
  include Lotus::Entity::DirtyTracking
  attributes :name, :age
end

class UserRepository
  include Lotus::Repository
end

Lotus::Model.configure do
  adapter type: :memory, uri: 'memory://localhost/dirty_tracking'

  mapping do
    collection :users do
      entity     User
      repository UserRepository

      attribute :id,   Integer
      attribute :name, String
      attribute :age,  String
    end
  end
end.load!

user = User.new(name: 'L')
user.changed? # => false

user.age = 33
user.changed?           # => true
user.changed_attributes # => {:age=>33}

user = UserRepository.create(user)
user.changed? # => false

user.update(name: 'Luca')
user.changed?           # => true
user.changed_attributes # => {:name=>"Luca"}

user = UserRepository.update(user)
user.changed? # => false

result = UserRepository.find(user.id)
result.changed? # => false

Example

For a full working example, have a look at EXAMPLE.md. Please remember that the setup code is only required for the standalone usage of Lotus::Model. A Lotus application will handle that configurations for you.

Versioning

Lotus::Model uses Semantic Versioning 2.0.0

Contributing

Fork it ( https://github.com/lotus/model/fork )
Create your feature branch (git checkout -b my-new-feature)
Commit your changes (git commit -am 'Add some feature')
Push to the branch (git push origin my-new-feature)
Create new Pull Request