Active Record – Object-relational mapping in Rails

Active Record connects classes to relational database tables to establish an almost zero-configuration persistence layer for applications. The library provides a base class that, when subclassed, sets up a mapping between the new class and an existing table in the database. In the context of an application, these classes are commonly referred to as models. Models can also be connected to other models; this is done by defining associations.

Active Record relies heavily on naming in that it uses class and association names to establish mappings between respective database tables and foreign key columns. Although these mappings can be defined explicitly, it’s recommended to follow naming conventions, especially when getting started with the library.

You can read more about Active Record in the Active Record Basics guide.

A short rundown of some of the major features:

  • Automated mapping between classes and tables, attributes and columns.

    class Product < ActiveRecord::Base
end

    The Product class is automatically mapped to the table named “products”, which might look like this:

    CREATE TABLE products (
  id bigint NOT NULL auto_increment,
  name varchar(255),
  PRIMARY KEY  (id)
);

    This would also define the following accessors: Product#name and Product#name=(new_name).

    Learn more

  • Associations between objects defined by simple class methods.

    class Firm < ActiveRecord::Base
  has_many   :clients
  has_one    :account
  belongs_to :conglomerate
end

    Learn more

  • Aggregations of value objects.

    class Account < ActiveRecord::Base
  composed_of :balance, class_name: 'Money',
              mapping: %w(balance amount)
  composed_of :address,
              mapping: [%w(address_street street), %w(address_city city)]
end

    Learn more

  • Validation rules that can differ for new or existing objects.

    class Account < ActiveRecord::Base
  validates :subdomain, :name, :email_address, :password, presence: true
  validates :subdomain, uniqueness: true
  validates :terms_of_service, acceptance: true, on: :create
  validates :password, :email_address, confirmation: true, on: :create
end

    Learn more

  • Callbacks available for the entire life cycle (instantiation, saving, destroying, validating, etc.).

    class Person < ActiveRecord::Base
  before_destroy :invalidate_payment_plan
  # the `invalidate_payment_plan` method gets called just before Person#destroy
end

    Learn more

  • Inheritance hierarchies.

    class Company < ActiveRecord::Base; end
class Firm < Company; end
class Client < Company; end
class PriorityClient < Client; end

    Learn more

  • Transactions.

    # Database transaction
Account.transaction do
  david.withdrawal(100)
  mary.deposit(100)
end

    Learn more

  • Reflections on columns, associations, and aggregations.

    reflection = Firm.reflect_on_association(:clients)
reflection.klass # => Client (class)
Firm.columns # Returns an array of column descriptors for the firms table

    Learn more

  • Database abstraction through simple adapters.

    # connect to SQLite3
ActiveRecord::Base.establish_connection(adapter: 'sqlite3', database: 'dbfile.sqlite3')

# connect to MySQL with authentication
ActiveRecord::Base.establish_connection(
  adapter:  'mysql2',
  host:     'localhost',
  username: 'me',
  password: 'secret',
  database: 'activerecord'
)

    Learn more and read about the built-in support for MySQL, PostgreSQL, and SQLite3.

  • Logging support for Log4r and Logger.

    ActiveRecord::Base.logger = ActiveSupport::Logger.new(STDOUT)
ActiveRecord::Base.logger = Log4r::Logger.new('Application Log')

  • Database agnostic schema management with Migrations.

    class AddSystemSettings < ActiveRecord::Migration[8.0]
  def up
    create_table :system_settings do |t|
      t.string  :name
      t.string  :label
      t.text    :value
      t.string  :type
      t.integer :position
    end

    SystemSetting.create name: 'notice', label: 'Use notice?', value: 1
  end

  def down
    drop_table :system_settings
  end
end

    Learn more

Philosophy

Active Record is an implementation of the object-relational mapping (ORM) pattern by the same name described by Martin Fowler:

“An object that wraps a row in a database table or view, encapsulates the database access, and adds domain logic on that data.”

Active Record attempts to provide a coherent wrapper as a solution for the inconvenience that is object-relational mapping. The prime directive for this mapping has been to minimize the amount of code needed to build a real-world domain model. This is made possible by relying on a number of conventions that make it easy for Active Record to infer complex relations and structures from a minimal amount of explicit direction.

Convention over Configuration:

  • No XML files!

  • Lots of reflection and run-time extension

  • Magic is not inherently a bad word

Admit the Database:

  • Lets you drop down to SQL for odd cases and performance

  • Doesn’t attempt to duplicate or replace data definitions

Download and installation

The latest version of Active Record can be installed with RubyGems:

$ gem install activerecord

Source code can be downloaded as part of the Rails project on GitHub:

License

Active Record is released under the MIT license:

Support

API documentation is at:

Bug reports for the Ruby on Rails project can be filed here:

Feature requests should be discussed on the rails-core mailing list here:

Validation error class to wrap association records’ errors, with index_errors support.

Namespace

Module

Class

Methods

Included Modules

Constants

MigrationProxy = Struct.new(:name, :version, :filename, :scope) do def initialize(name, version, filename, scope) super @migration = nil end def basename File.basename(filename) end delegate :migrate, :announce, :write, :disable_ddl_transaction, to: :migration private def migration @migration ||= load_migration end def load_migration Object.send(:remove_const, name) rescue nil load(File.expand_path(filename)) name.constantize.new(name, version) end end
 

MigrationProxy is used to defer loading of the actual migration classes until they are needed
Point = Struct.new(:x, :y)
UnknownAttributeError = ActiveModel::UnknownAttributeError
 

Active Model UnknownAttributeError

Raised when unknown attributes are supplied via mass assignment.

class Person
  include ActiveModel::AttributeAssignment
  include ActiveModel::Validations
end

person = Person.new
person.assign_attributes(name: 'Gorby')
# => ActiveModel::UnknownAttributeError: unknown attribute 'name' for Person.

Attributes

[RW] application_record_class
[RW] before_committed_on_all_records
[RW] belongs_to_required_validates_foreign_key
[RW] database_cli
[R] default_timezone
[RW] disable_prepared_statements
[RW] index_nested_attribute_errors
[RW] maintain_test_schema
[R] permanent_connection_checkout
[RW] query_transformers
[RW] raise_on_assign_to_attr_readonly
[RW] reading_role
[RW] run_after_transaction_callbacks_in_order_defined
[RW] writing_role

Class Public methods

action_on_strict_loading_violation

Set the application to log or raise when an association violates strict loading. Defaults to :raise.

📝 Source code 
# File activerecord/lib/active_record.rb, line 361
  singleton_class.attr_accessor :action_on_strict_loading_violation
🔎 See on GitHub

after_all_transactions_commit(&block)

Registers a block to be called after all the current transactions have been committed.

If there is no currently open transaction, the block is called immediately.

If there are multiple nested transactions, the block is called after the outermost one has been committed,

If any of the currently open transactions is rolled back, the block is never called.

If multiple transactions are open across multiple databases, the block will be invoked if and once all of them have been committed. But note that nesting transactions across two distinct databases is a sharding anti-pattern that comes with a world of hurts.

📝 Source code 
# File activerecord/lib/active_record.rb, line 527
  def self.after_all_transactions_commit(&block)
    open_transactions = all_open_transactions

    if open_transactions.empty?
      yield
    elsif open_transactions.size == 1
      open_transactions.first.after_commit(&block)
    else
      count = open_transactions.size
      callback = -> do
        count -= 1
        block.call if count.zero?
      end
      open_transactions.each do |t|
        t.after_commit(&callback)
      end
      open_transactions = nil # rubocop:disable Lint/UselessAssignment avoid holding it in the closure
    end
  end
🔎 See on GitHub

async_query_executor

Sets the async_query_executor for an application. By default the thread pool executor set to nil which will not run queries in the background. Applications must configure a thread pool executor to use this feature. Options are:

* nil - Does not initialize a thread pool executor. Any async calls will be
run in the foreground.
* :global_thread_pool - Initializes a single +Concurrent::ThreadPoolExecutor+
that uses the +async_query_concurrency+ for the +max_threads+ value.
* :multi_thread_pool - Initializes a +Concurrent::ThreadPoolExecutor+ for each
database connection. The initializer values are defined in the configuration hash.
📝 Source code 
# File activerecord/lib/active_record.rb, line 283
  singleton_class.attr_accessor :async_query_executor
🔎 See on GitHub

db_warnings_action

The action to take when database query produces warning. Must be one of :ignore, :log, :raise, :report, or a custom proc. The default is :ignore.

📝 Source code 
# File activerecord/lib/active_record.rb, line 233
  singleton_class.attr_reader :db_warnings_action
🔎 See on GitHub

db_warnings_action=(action)

📝 Source code 
# File activerecord/lib/active_record.rb, line 235
  def self.db_warnings_action=(action)
    @db_warnings_action =
      case action
      when :ignore
        nil
      when :log
        ->(warning) do
          warning_message = "[#{warning.class}] #{warning.message}"
          warning_message += " (#{warning.code})" if warning.code
          ActiveRecord::Base.logger.warn(warning_message)
        end
      when :raise
        ->(warning) { raise warning }
      when :report
        ->(warning) { Rails.error.report(warning, handled: true) }
      when Proc
        action
      else
        raise ArgumentError, "db_warnings_action must be one of :ignore, :log, :raise, :report, or a custom proc."
      end
  end
🔎 See on GitHub

db_warnings_ignore

Specify allowlist of database warnings.

📝 Source code 
# File activerecord/lib/active_record.rb, line 262
  singleton_class.attr_accessor :db_warnings_ignore
🔎 See on GitHub

default_timezone=(default_timezone)

Determines whether to use Time.utc (using :utc) or Time.local (using :local) when pulling dates and times from the database. This is set to :utc by default.

📝 Source code 
# File activerecord/lib/active_record.rb, line 218
  def self.default_timezone=(default_timezone)
    unless %i(local utc).include?(default_timezone)
      raise ArgumentError, "default_timezone must be either :utc (default) or :local."
    end

    @default_timezone = default_timezone
  end
🔎 See on GitHub

disconnect_all!()

Explicitly closes all database connections in all pools.

📝 Source code 
# File activerecord/lib/active_record.rb, line 510
  def self.disconnect_all!
    ConnectionAdapters::PoolConfig.disconnect_all!
  end
🔎 See on GitHub

dump_schema_after_migration

Specify whether schema dump should happen at the end of the bin/rails db:migrate command. This is true by default, which is useful for the development environment. This should ideally be false in the production environment where dumping schema is rarely needed.

📝 Source code 
# File activerecord/lib/active_record.rb, line 409
  singleton_class.attr_accessor :dump_schema_after_migration
🔎 See on GitHub

dump_schemas

Specifies which database schemas to dump when calling db:schema:dump. If the value is :schema_search_path (the default), any schemas listed in schema_search_path are dumped. Use :all to dump all schemas regardless of schema_search_path, or a string of comma separated schemas for a custom list.

📝 Source code 
# File activerecord/lib/active_record.rb, line 419
  singleton_class.attr_accessor :dump_schemas
🔎 See on GitHub

eager_load!()

📝 Source code 
# File activerecord/lib/active_record.rb, line 499
  def self.eager_load!
    super
    ActiveRecord::Locking.eager_load!
    ActiveRecord::Scoping.eager_load!
    ActiveRecord::Associations.eager_load!
    ActiveRecord::AttributeMethods.eager_load!
    ActiveRecord::ConnectionAdapters.eager_load!
    ActiveRecord::Encryption.eager_load!
  end
🔎 See on GitHub

error_on_ignored_order

Specifies if an error should be raised if the query has an order being ignored when doing batch queries. Useful in applications where the scope being ignored is error-worthy, rather than a warning.

📝 Source code 
# File activerecord/lib/active_record.rb, line 380
  singleton_class.attr_accessor :error_on_ignored_order
🔎 See on GitHub

gem_version()

Returns the currently loaded version of Active Record as a Gem::Version.

📝 Source code 
# File activerecord/lib/active_record/gem_version.rb, line 5
  def self.gem_version
    Gem::Version.new VERSION::STRING
  end
🔎 See on GitHub

generate_secure_token_on

Controls when to generate a value for has_secure_token declarations. Defaults to :create.

📝 Source code 
# File activerecord/lib/active_record.rb, line 460
  singleton_class.attr_accessor :generate_secure_token_on
🔎 See on GitHub

global_executor_concurrency=(global_executor_concurrency)

Set the global_executor_concurrency. This configuration value can only be used with the global thread pool async query executor.

📝 Source code 
# File activerecord/lib/active_record.rb, line 298
  def self.global_executor_concurrency=(global_executor_concurrency)
    if self.async_query_executor.nil? || self.async_query_executor == :multi_thread_pool
      raise ArgumentError, "`global_executor_concurrency` cannot be set when the executor is nil or set to `:multi_thread_pool`. For multiple thread pools, please set the concurrency in your database configuration."
    end

    @global_executor_concurrency = global_executor_concurrency
  end
🔎 See on GitHub

lazily_load_schema_cache

Lazily load the schema cache. This option will load the schema cache when a connection is established rather than on boot.

📝 Source code 
# File activerecord/lib/active_record.rb, line 189
  singleton_class.attr_accessor :lazily_load_schema_cache
🔎 See on GitHub

marshalling_format_version()

📝 Source code 
# File activerecord/lib/active_record.rb, line 463
  def self.marshalling_format_version
    Marshalling.format_version
  end
🔎 See on GitHub

marshalling_format_version=(value)

📝 Source code 
# File activerecord/lib/active_record.rb, line 467
  def self.marshalling_format_version=(value)
    Marshalling.format_version = value
  end
🔎 See on GitHub

migration_strategy

Specify strategy to use for executing migrations.

📝 Source code 
# File activerecord/lib/active_record.rb, line 400
  singleton_class.attr_accessor :migration_strategy
🔎 See on GitHub

permanent_connection_checkout=(value)

Defines whether ActiveRecord::Base.connection is allowed, deprecated, or entirely disallowed.

📝 Source code 
# File activerecord/lib/active_record.rb, line 314
  def self.permanent_connection_checkout=(value)
    unless [true, :deprecated, :disallowed].include?(value)
      raise ArgumentError, "permanent_connection_checkout must be one of: `true`, `:deprecated` or `:disallowed`"
    end
    @permanent_connection_checkout = value
  end
🔎 See on GitHub

protocol_adapters

Provides a mapping between database protocols/DBMSs and the underlying database adapter to be used. This is used only by the DATABASE_URL environment variable.

Example

DATABASE_URL="mysql://myuser:mypass@localhost/somedatabase"

The above URL specifies that MySQL is the desired protocol/DBMS, and the application configuration can then decide which adapter to use. For this example the default mapping is from mysql to mysql2, but :trilogy is also supported.

ActiveRecord.protocol_adapters.mysql = "mysql2"

The protocols names are arbitrary, and external database adapters can be registered and set here.

📝 Source code 
# File activerecord/lib/active_record.rb, line 490
  singleton_class.attr_accessor :protocol_adapters
🔎 See on GitHub

queues

Specifies the names of the queues used by background jobs.

📝 Source code 
# File activerecord/lib/active_record.rb, line 336
  singleton_class.attr_accessor :queues
🔎 See on GitHub

raise_int_wider_than_64bit

Application configurable boolean that denotes whether or not to raise an exception when the PostgreSQLAdapter is provided with an integer that is wider than signed 64bit representation

📝 Source code 
# File activerecord/lib/active_record.rb, line 446
  singleton_class.attr_accessor :raise_int_wider_than_64bit
🔎 See on GitHub

schema_cache_ignored_table?(table_name)

Checks to see if the table_name is ignored by checking against the schema_cache_ignored_tables option.

ActiveRecord.schema_cache_ignored_table?(:developers)
📝 Source code 
# File activerecord/lib/active_record.rb, line 205
  def self.schema_cache_ignored_table?(table_name)
    ActiveRecord.schema_cache_ignored_tables.any? do |ignored|
      ignored === table_name
    end
  end
🔎 See on GitHub

schema_cache_ignored_tables

A list of tables or regex’s to match tables to ignore when dumping the schema cache. For example if this is set to +[/^_/]+ the schema cache will not dump tables named with an underscore.

📝 Source code 
# File activerecord/lib/active_record.rb, line 197
  singleton_class.attr_accessor :schema_cache_ignored_tables
🔎 See on GitHub

schema_format

Specifies the format to use when dumping the database schema with Rails’ Rakefile. If :sql, the schema is dumped as (potentially database- specific) SQL statements. If :ruby, the schema is dumped as an ActiveRecord::Schema file which can be loaded into any database that supports migrations. Use :ruby if you want to have different database adapters for, e.g., your development and test environments.

📝 Source code 
# File activerecord/lib/active_record.rb, line 372
  singleton_class.attr_accessor :schema_format
🔎 See on GitHub

timestamped_migrations

Specify whether or not to use timestamps for migration versions

📝 Source code 
# File activerecord/lib/active_record.rb, line 386
  singleton_class.attr_accessor :timestamped_migrations
🔎 See on GitHub

use_yaml_unsafe_load

Application configurable boolean that instructs the YAML Coder to use an unsafe load if set to true.

📝 Source code 
# File activerecord/lib/active_record.rb, line 438
  singleton_class.attr_accessor :use_yaml_unsafe_load
🔎 See on GitHub

validate_migration_timestamps

Specify whether or not to validate migration timestamps. When set, an error will be raised if a timestamp is more than a day ahead of the timestamp associated with the current time. timestamped_migrations must be set to true.

📝 Source code 
# File activerecord/lib/active_record.rb, line 394
  singleton_class.attr_accessor :validate_migration_timestamps
🔎 See on GitHub

verbose_query_logs

Specifies if the methods calling database queries should be logged below their relevant queries. Defaults to false.

📝 Source code 
# File activerecord/lib/active_record.rb, line 329
  singleton_class.attr_accessor :verbose_query_logs
🔎 See on GitHub

verify_foreign_keys_for_fixtures

If true, Rails will verify all foreign keys in the database after loading fixtures. An error will be raised if there are any foreign key violations, indicating incorrectly written fixtures. Supported by PostgreSQL and SQLite.

📝 Source code 
# File activerecord/lib/active_record.rb, line 428
  singleton_class.attr_accessor :verify_foreign_keys_for_fixtures
🔎 See on GitHub

version()

Returns the currently loaded version of Active Record as a Gem::Version.

📝 Source code 
# File activerecord/lib/active_record/version.rb, line 7
  def self.version
    gem_version
  end
🔎 See on GitHub

yaml_column_permitted_classes

Application configurable array that provides additional permitted classes to Psych safe_load in the YAML Coder

📝 Source code 
# File activerecord/lib/active_record.rb, line 453
  singleton_class.attr_accessor :yaml_column_permitted_classes
🔎 See on GitHub