Active Record

Active Record objects don't specify their attributes directly, but rather infer them from the table definition with which they're linked. Adding, removing, and changing attributes and their type is done directly in the database. Any change is instantly reflected in the Active Record objects. The mapping that binds a given Active Record class to a certain database table will happen automatically in most common cases, but can be overwritten for the uncommon ones.

See the mapping rules in table_name and the full example in files/activerecord/README_rdoc.html for more insight.


Active Records accept constructor parameters either in a hash or as a block. The hash method is especially useful when you're receiving the data from somewhere else, like an HTTP request. It works like this:

user = => "David", :occupation => "Code Artist") # => "David"

You can also use block initialization:

user = do |u| = "David"
  u.occupation = "Code Artist"

And of course you can just create a bare object and specify the attributes after the fact:

user = = "David"
user.occupation = "Code Artist"


Conditions can either be specified as a string, array, or hash representing the WHERE-part of an SQL statement. The array form is to be used when the condition input is tainted and requires sanitization. The string form can be used for statements that don't involve tainted data. The hash form works much like the array form, except only equality and range is possible. Examples:

class User < ActiveRecord::Base
  def self.authenticate_unsafely(user_name, password)
    where("user_name = '#{user_name}' AND password = '#{password}'").first

  def self.authenticate_safely(user_name, password)
    where("user_name = ? AND password = ?", user_name, password).first

  def self.authenticate_safely_simply(user_name, password)
    where(:user_name => user_name, :password => password).first

The authenticate_unsafely method inserts the parameters directly into the query and is thus susceptible to SQL-injection attacks if the user_name and password parameters come directly from an HTTP request. The authenticate_safely and authenticate_safely_simply both will sanitize the user_name and password before inserting them in the query, which will ensure that an attacker can't escape the query and fake the login (or worse).

When using multiple parameters in the conditions, it can easily become hard to read exactly what the fourth or fifth question mark is supposed to represent. In those cases, you can resort to named bind variables instead. That's done by replacing the question marks with symbols and supplying a hash with values for the matching symbol keys:

  "id = :id AND name = :name AND division = :division AND created_at > :accounting_date",
  { :id => 3, :name => "37signals", :division => "First", :accounting_date => '2005-01-01' }

Similarly, a simple hash without a statement will generate conditions based on equality with the SQL AND operator. For instance:

Student.where(:first_name => "Harvey", :status => 1)

A range may be used in the hash to use the SQL BETWEEN operator:

Student.where(:grade => 9..12)

An array may be used in the hash to use the SQL IN operator:

Student.where(:grade => [9,11,12])

When joining tables, nested hashes or keys written in the form 'table_name.column_name' can be used to qualify the table name of a particular condition. For instance:

Student.joins(:schools).where(:schools => { :category => 'public' })
Student.joins(:schools).where('schools.category' => 'public' )

Overwriting default accessors

All column values are automatically available through basic accessors on the Active Record object, but sometimes you want to specialize this behavior. This can be done by overwriting the default accessors (using the same name as the attribute) and calling read_attribute(attr_name) and write_attribute(attr_name, value) to actually change things.

class Song < ActiveRecord::Base
  # Uses an integer of seconds to hold the length of the song

  def length=(minutes)
    write_attribute(:length, minutes.to_i * 60)

  def length
    read_attribute(:length) / 60

You can alternatively use self[:attribute]=(value) and self[:attribute] instead of write_attribute(:attribute, value) and read_attribute(:attribute).

Attribute query methods

In addition to the basic accessors, query methods are also automatically available on the Active Record object. Query methods allow you to test whether an attribute value is present.

For example, an Active Record User with the name attribute has a name? method that you can call to determine whether the user has a name:

user = => "David") # => true

anonymous = => "") # => false

Accessing attributes before they have been typecasted

Sometimes you want to be able to read the raw attribute data without having the column-determined typecast run its course first. That can be done by using the <attribute>_before_type_cast accessors that all attributes have. For example, if your Account model has a balance attribute, you can call account.balance_before_type_cast or account.id_before_type_cast.

This is especially useful in validation situations where the user might supply a string for an integer field and you want to display the original string back in an error message. Accessing the attribute normally would typecast the string to 0, which isn't what you want.

Dynamic attribute-based finders

Dynamic attribute-based finders are a cleaner way of getting (and/or creating) objects by simple queries without turning to SQL. They work by appending the name of an attribute to find_by_, find_last_by_, or find_all_by_ and thus produces finders like Person.find_by_user_name, Person.find_all_by_last_name, and Payment.find_by_transaction_id. Instead of writing Person.where(:user_name => user_name).first, you just do Person.find_by_user_name(user_name). And instead of writing Person.where(:last_name => last_name).all, you just do Person.find_all_by_last_name(last_name).

It's possible to add an exclamation point (!) on the end of the dynamic finders to get them to raise an ActiveRecord::RecordNotFound error if they do not return any records, like Person.find_by_last_name!.

It's also possible to use multiple attributes in the same find by separating them with “and”.

Person.where(:user_name => user_name, :password => password).first
Person.find_by_user_name_and_password(user_name, password) # with dynamic finder

It's even possible to call these dynamic finder methods on relations and named scopes.


The same dynamic finder style can be used to create the object if it doesn't already exist. This dynamic finder is called with find_or_create_by_ and will return the object if it already exists and otherwise creates it, then returns it. Protected attributes won't be set unless they are given in a block.

# No 'Summer' tag exists
Tag.find_or_create_by_name("Summer") # equal to Tag.create(:name => "Summer")

# Now the 'Summer' tag does exist
Tag.find_or_create_by_name("Summer") # equal to Tag.find_by_name("Summer")

# Now 'Bob' exist and is an 'admin'
User.find_or_create_by_name('Bob', :age => 40) { |u| u.admin = true }

Adding an exclamation point (!) on to the end of find_or_create_by_ will raise an ActiveRecord::RecordInvalid error if the new record is invalid.

Use the find_or_initialize_by_ finder if you want to return a new record without saving it first. Protected attributes won't be set unless they are given in a block.

# No 'Winter' tag exists
winter = Tag.find_or_initialize_by_name("Winter")
winter.persisted? # false

To find by a subset of the attributes to be used for instantiating a new object, pass a hash instead of a list of parameters.

Tag.find_or_create_by_name(:name => "rails", :creator => current_user)

That will either find an existing tag named “rails”, or create a new one while setting the user that created it.

Just like find_by_*, you can also use scoped_by_* to retrieve data. The good thing about using this feature is that the very first time result is returned using method_missing technique but after that the method is declared on the class. Henceforth method_missing will not be hit.


Saving arrays, hashes, and other non-mappable objects in text columns

Active Record can serialize any object in text columns using YAML. To do so, you must specify this with a call to the class method serialize. This makes it possible to store arrays, hashes, and other non-mappable objects without doing any additional work.

class User < ActiveRecord::Base
  serialize :preferences

user = User.create(:preferences => { "background" => "black", "display" => large })
User.find( # => { "background" => "black", "display" => large }

You can also specify a class option as the second parameter that'll raise an exception if a serialized object is retrieved as a descendant of a class not in the hierarchy.

class User < ActiveRecord::Base
  serialize :preferences, Hash

user = User.create(:preferences => %w( one two three ))
User.find(    # raises SerializationTypeMismatch

When you specify a class option, the default value for that attribute will be a new instance of that class.

class User < ActiveRecord::Base
  serialize :preferences, OpenStruct

user =
user.preferences.theme_color = "red"

Single table inheritance

Active Record allows inheritance by storing the name of the class in a column that by default is named “type” (can be changed by overwriting Base.inheritance_column). This means that an inheritance looking like this:

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

When you do Firm.create(:name => "37signals"), this record will be saved in the companies table with type = “Firm”. You can then fetch this row again using Company.where(:name => '37signals').first and it will return a Firm object.

If you don't have a type column defined in your table, single-table inheritance won't be triggered. In that case, it'll work just like normal subclasses with no special magic for differentiating between them or reloading the right type with find.

Note, all the attributes for all the cases are kept in the same table. Read more:

Connection to multiple databases in different models

Connections are usually created through ::establish_connection and retrieved by #connection. All classes inheriting from ActiveRecord::Base will use this connection. But you can also set a class-specific connection. For example, if Course is an ActiveRecord::Base, but resides in a different database, you can just say Course.establish_connection and Course and all of its subclasses will use this connection instead.

This feature is implemented by keeping a connection pool in ActiveRecord::Base that is a Hash indexed by the class. If a connection is requested, the ::retrieve_connection method will go up the class-hierarchy until a connection is found in the connection pool.


  • ActiveRecordError - Generic error class and superclass of all other errors raised by Active Record.

  • AdapterNotSpecified - The configuration hash used in establish_connection didn't include an :adapter key.

  • AdapterNotFound - The :adapter key used in establish_connection specified a non-existent adapter (or a bad spelling of an existing one).

  • AssociationTypeMismatch - The object assigned to the association wasn't of the type specified in the association definition.

  • SerializationTypeMismatch - The serialized object wasn't of the class specified as the second parameter.

  • ConnectionNotEstablished+ - No connection has been established. Use establish_connection before querying.

  • RecordNotFound - No record responded to the find method. Either the row with the given ID doesn't exist or the row didn't meet the additional restrictions. Some find calls do not raise this exception to signal nothing was found, please check its documentation for further details.

  • StatementInvalid - The database server rejected the SQL statement. The precise error is added in the message.

  • MultiparameterAssignmentErrors - Collection of errors that occurred during a mass assignment using the attributes= method. The errors property of this exception contains an array of AttributeAssignmentError objects that should be inspected to determine which attributes triggered the errors.

  • AttributeAssignmentError - An error occurred while doing a mass assignment through the attributes= method. You can inspect the attribute property of the exception object to determine which attribute triggered the error.

Note: The attributes listed are class-level attributes (accessible from both the class and instance level). So it's possible to assign a logger to the class through Base.logger= which will then be used by all instances in the current object space.

Included Modules
Class Public methods

Overwrite the default class equality method to provide support for association proxies.

# File activerecord/lib/active_record/base.rb, line 432
def ===(object)
# File activerecord/lib/active_record/base.rb, line 440
def arel_engine
  @arel_engine ||= begin
    if self == ActiveRecord::Base
      connection_handler.retrieve_connection_pool(self) ? self : superclass.arel_engine
# File activerecord/lib/active_record/base.rb, line 436
def arel_table
  @arel_table ||=, arel_engine)
# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 183
def clear_active_connections!

Contains the database configuration - as is typically stored in config/database.yml - as a Hash.

For example, the following database.yml…

  adapter: sqlite3
  database: db/development.sqlite3

  adapter: sqlite3
  database: db/production.sqlite3

…would result in ::configurations to look like this:

   'development' => {
      'adapter'  => 'sqlite3',
      'database' => 'db/development.sqlite3'
   'production' => {
      'adapter'  => 'sqlite3',
      'database' => 'db/production.sqlite3'
# File activerecord/lib/active_record/base.rb, line 368
cattr_accessor :configurations, :instance_writer => false

Returns true if Active Record is connected.

# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 175
def connected?

Returns the connection currently associated with the class. This can also be used to “borrow” the connection to do database work unrelated to any of the specific Active Records.

# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 144
def connection

Returns the configuration of the associated connection as a hash:

# => {:pool=>5, :timeout=>5000, :database=>"db/development.sqlite3", :adapter=>"sqlite3"}

Please use only for reading.

# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 162
def connection_config

The connection handler

# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 84
class_attribute :connection_handler, :instance_writer => false
# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 148
def connection_id
# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 152
def connection_id=(connection_id)
  Thread.current['ActiveRecord::Base.connection_id'] = connection_id
# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 166
def connection_pool
  connection_handler.retrieve_connection_pool(self) or raise ConnectionNotEstablished

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

# File activerecord/lib/active_record/base.rb, line 375
cattr_accessor :default_timezone, :instance_writer => false
establish_connection(spec = ENV["DATABASE_URL"])

Establishes the connection to the database. Accepts a hash as input where the :adapter key must be specified with the name of a database adapter (in lower-case) example for regular databases (MySQL, Postgresql, etc):

  :adapter  => "mysql",
  :host     => "localhost",
  :username => "myuser",
  :password => "mypass",
  :database => "somedatabase"

Example for SQLite database:

  :adapter => "sqlite",
  :database  => "path/to/dbfile"

Also accepts keys as strings (for parsing from YAML for example):

  "adapter" => "sqlite",
  "database"  => "path/to/dbfile"

Or a URL:


The exceptions AdapterNotSpecified, AdapterNotFound and ArgumentError may be returned on an error.

# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 128
def self.establish_connection(spec = ENV["DATABASE_URL"])
  resolver = spec, configurations
  spec = resolver.spec

  unless respond_to?(spec.adapter_method)
    raise AdapterNotFound, "database configuration specifies nonexistent #{spec.config[:adapter]} adapter"

  connection_handler.establish_connection name, spec
# File activerecord/lib/active_record/base.rb, line 409
def generated_feature_methods
  @generated_feature_methods ||= begin
    mod = const_set(:GeneratedFeatureMethods,
    include mod

Returns a string like 'Post(id:integer, title:string, body:text)'

# File activerecord/lib/active_record/base.rb, line 418
def inspect
  if self == Base
  elsif abstract_class?
  elsif table_exists?
    attr_list = { |c| "#{}: #{c.type}" } * ', '
    "#{super}(Table doesn't exist)"

Accepts a logger conforming to the interface of Log4r or the default Ruby 1.8+ Logger class, which is then passed on to any new database connections made and which can be retrieved on both a class and instance level by calling logger.

# File activerecord/lib/active_record/base.rb, line 339
cattr_accessor :logger, :instance_writer => false

Establishes a connection to the database that's used by all Active Record objects.

# File activerecord/lib/active_record/connection_adapters/mysql2_adapter.rb, line 9
def self.mysql2_connection(config)
  config[:username] = 'root' if config[:username].nil?

  if Mysql2::Client.const_defined? :FOUND_ROWS
    config[:flags] = Mysql2::Client::FOUND_ROWS

  client =
  options = [config[:host], config[:username], config[:password], config[:database], config[:port], config[:socket], 0], logger, options, config)
new(attributes = nil, options = {})

New objects can be instantiated as either empty (pass no construction parameter) or pre-set with attributes but not yet saved (pass a hash with key names matching the associated table column names). In both instances, valid attribute keys are determined by the column names of the associated table – hence you can't have attributes that aren't part of the table columns.

initialize respects mass-assignment security and accepts either :as or :without_protection options in the options parameter.


# Instantiates a single new object => 'Jamie')

# Instantiates a single new object using the :admin mass-assignment security role{ :first_name => 'Jamie', :is_admin => true }, :as => :admin)

# Instantiates a single new object bypassing mass-assignment security{ :first_name => 'Jamie', :is_admin => true }, :without_protection => true)
# File activerecord/lib/active_record/base.rb, line 481
def initialize(attributes = nil, options = {})
  defaults = Hash[ { |k, v| [k, v.duplicable? ? v.dup : v] }]
  @attributes = self.class.initialize_attributes(defaults)
  @association_cache = {}
  @aggregation_cache = {}
  @attributes_cache = {}
  @new_record = true
  @readonly = false
  @destroyed = false
  @marked_for_destruction = false
  @previously_changed = {}
  @changed_attributes = {}



  assign_attributes(attributes, options) if attributes

  yield self if block_given?
  run_callbacks :initialize
remove_connection(klass = self)
# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 179
def remove_connection(klass = self)
# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 170
def retrieve_connection

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.

# File activerecord/lib/active_record/base.rb, line 386
cattr_accessor :schema_format , :instance_writer => false

Specify whether or not to use timestamps for migration versions

# File activerecord/lib/active_record/base.rb, line 392
cattr_accessor :timestamped_migrations , :instance_writer => false
Instance Public methods

Allows sort on objects

# File activerecord/lib/active_record/base.rb, line 618
def <=>(other_object)
  if other_object.is_a?(self.class)
    self.to_key <=> other_object.to_key

Returns true if comparison_object is the same exact object, or comparison_object is of the same type and self has an ID and it is equal to

Note that new records are different from any other record by definition, unless the other record is the receiver itself. Besides, if you fetch existing records with select and leave the ID out, you're on your own, this predicate will return false.

Note also that destroying a record preserves its ID in the model instance, so deleted models are still comparable.

Also aliased as: eql?
# File activerecord/lib/active_record/base.rb, line 593
def ==(comparison_object)
  super ||
    comparison_object.instance_of?(self.class) &&
    id.present? && == id

Returns the connection currently associated with the class. This can also be used to “borrow” the connection to do database work that isn't easily done without going straight to SQL.

# File activerecord/lib/active_record/connection_adapters/abstract/connection_specification.rb, line 90
def connection

Populate coder with attributes about this record that should be serialized. The structure of coder defined in this method is guaranteed to match the structure of coder passed to the init_with method.


class Post < ActiveRecord::Base
coder = {}
coder # => { 'id' => nil, ... }
# File activerecord/lib/active_record/base.rb, line 580
def encode_with(coder)
  coder['attributes'] = attributes

Freeze the attributes hash such that associations are still accessible, even on destroyed records.

# File activerecord/lib/active_record/base.rb, line 608
def freeze
  @attributes.freeze; self

Returns true if the attributes hash has been frozen.

# File activerecord/lib/active_record/base.rb, line 613
def frozen?

Delegates to id in order to allow two records of the same type and id to work with something like:

[ Person.find(1), Person.find(2), Person.find(3) ] & [ Person.find(1), Person.find(4) ] # => [ Person.find(1) ]
# File activerecord/lib/active_record/base.rb, line 603
def hash

Initialize an empty model object from coder. coder must contain the attributes necessary for initializing an empty model object. For example:

class Post < ActiveRecord::Base

post = Post.allocate
post.init_with('attributes' => { 'title' => 'hello world' })
post.title # => 'hello world'
# File activerecord/lib/active_record/base.rb, line 514
def init_with(coder)
  @attributes = self.class.initialize_attributes(coder['attributes'])
  @relation = nil

  @attributes_cache, @previously_changed, @changed_attributes = {}, {}, {}
  @association_cache = {}
  @aggregation_cache = {}
  @readonly = @destroyed = @marked_for_destruction = false
  @new_record = false
  run_callbacks :find
  run_callbacks :initialize


Duped objects have no id assigned and are treated as new records. Note that this is a “shallow” copy as it copies the object's attributes only, not its associations. The extent of a “deep” copy is application specific and is therefore left to the application to implement according to its need. The dup method does not preserve the timestamps (created|updated)_(at|on).

# File activerecord/lib/active_record/base.rb, line 535
def initialize_dup(other)
  cloned_attributes = other.clone_attributes(:read_attribute_before_type_cast)
  self.class.initialize_attributes(cloned_attributes, :serialized => false)


  @attributes = cloned_attributes

  _run_after_initialize_callbacks if respond_to?(:_run_after_initialize_callbacks)

  @changed_attributes = {}
  self.class.column_defaults.each do |attr, orig_value|
    @changed_attributes[attr] = orig_value if _field_changed?(attr, orig_value, @attributes[attr])

  @aggregation_cache = {}
  @association_cache = {}
  @attributes_cache = {}
  @new_record  = true


Returns the contents of the record as a nicely formatted string.

# File activerecord/lib/active_record/base.rb, line 638
def inspect
  inspection = if @attributes
                 self.class.column_names.collect { |name|
                   if has_attribute?(name)
                     "#{name}: #{attribute_for_inspect(name)}"
                 }.compact.join(", ")
                 "not initialized"
  "#<#{self.class} #{inspection}>"

Marks this record as read only.

# File activerecord/lib/active_record/base.rb, line 633
def readonly!
  @readonly = true

Returns true if the record is read only. Records loaded through joins with piggy-back attributes will be marked as read only since they cannot be saved.

# File activerecord/lib/active_record/base.rb, line 628
def readonly?