NOTE: Hanami::Model was the persistence layer for Hanami 1.x. This library will not receive any updates.
For the persistence layer for Hanami 2.x, please see hanami/db
- Home page: http://hanamirb.org
- Mailing List: http://hanamirb.org/mailing-list
- API Doc: http://rubydoc.info/gems/hanami-model
- Chat: https://chat.hanamirb.org
Hanami::Model supports Hanami 1.x only, and Ruby (MRI) 2.6 and 2.7.
Add this line to your application's Gemfile:
gem 'hanami-model'And then execute:
$ bundle
Or install it yourself as:
$ gem install hanami-model
This class provides a DSL to configure the connection.
require 'hanami/model'
require 'hanami/model/sql'
class User < Hanami::Entity
end
class UserRepository < Hanami::Repository
end
Hanami::Model.configure do
adapter :sql, 'postgres://username:password@localhost/bookshelf'
end.load!
repository = UserRepository.new
user = repository.create(name: 'Luca')
puts user.id # => 1
found = repository.find(user.id)
found == user # => true
updated = repository.update(user.id, age: 34)
updated.age # => 34
repository.delete(user.id)A model domain 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 'hanami/model'
class Person < Hanami::Entity
endAn 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 inherits from Hanami::Repository, it will receive the following interface:
#create(data)– Create a record for the given data (or entity)#update(id, data)– Update the record corresponding to the given id by setting the given data (or entity)#delete(id)– Delete the record corresponding to the given id#all- Fetch all the entities from the relation#find- Fetch an entity from the relation by primary key#first- Fetch the first entity from the relation#last- Fetch the last entity from the relation#clear- Delete all the records from the relation
A relation 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.new.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 'hanami/model'
class ArticleRepository < Hanami::Repository
def most_recent_by_author(author, limit: 8)
articles.where(author_id: author.id).
order(:published_at).
limit(limit)
end
endThis 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.
Hanami::Model can automap columns from relations and entities attributes.
When using a sql adapter, you must require hanami/model/sql before Hanami::Model.load! is called so the relations are loaded correctly.
However, there are cases where columns and attribute names do not match (mainly legacy databases).
require 'hanami/model'
class UserRepository < Hanami::Repository
self.relation = :t_user_archive
mapping do
attribute :id, from: :i_user_id
attribute :name, from: :s_name
attribute :age, from: :i_age
end
endNOTE: This feature should be used only when automapping fails because of the naming mismatch.
- A repository must be named after an entity, by appending
"Repository"to the entity class name (eg.Article=>ArticleRepository).
Hanami::Model's is thread safe during the runtime, but it isn't during the loading process. The mapper compiles some code internally, so be sure to safely load it before your application starts.
Mutex.new.synchronize do
Hanami::Model.load!
endThis is not necessary when Hanami::Model is used within a Hanami application.
If an entity has the following accessors: :created_at and :updated_at, they will be automatically updated when the entity is persisted.
require 'hanami/model'
require 'hanami/model/sql'
class User < Hanami::Entity
end
class UserRepository < Hanami::Repository
end
Hanami::Model.configure do
adapter :sql, 'postgresql://localhost/bookshelf'
end.load!
repository = UserRepository.new
user = repository.create(name: 'Luca')
puts user.created_at.to_s # => "2016-09-19 13:40:13 UTC"
puts user.updated_at.to_s # => "2016-09-19 13:40:13 UTC"
sleep 3
user = repository.update(user.id, age: 34)
puts user.created_at.to_s # => "2016-09-19 13:40:13 UTC"
puts user.updated_at.to_s # => "2016-09-19 13:40:16 UTC"In order to log database operations, you can configure a logger:
Hanami::Model.configure do
# ...
logger "log/development.log", level: :debug
endIt accepts the following arguments:
stream: a Ruby StringIO object - it can be$stdoutor a path to file (eg."log/development.log") - Defaults to$stdout:level: logging level - it can be::debug,:info,:warn,:error,:fatal,:unknown- Defaults to:debug:formatter: logging formatter - it can be::defaultor:json- Defaults to:default
Hanami::Model uses Semantic Versioning 2.0.0
- Fork it ( https://github.com/hanami/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
Copyright © 2014-2021 Luca Guidi – Released under MIT License
This project was formerly known as Lotus (lotus-model).