ActiveRecord::ConnectionAdapters::TableDefinition

Represents the schema of an SQL table in an abstract way. This class provides methods for manipulating the schema representation.

Inside migration files, the t object in create_table is actually of this type:

class SomeMigration < ActiveRecord::Migration[5.0]
  def up
    create_table :foo do |t|
      puts t.class  # => "ActiveRecord::ConnectionAdapters::TableDefinition"
    end
  end

  def down
    ...
  end
end
Methods
#
B
C
I
N
R
T
Included Modules
Attributes
[R] as
[R] comment
[R] foreign_keys
[RW] indexes
[R] name
[R] options
[R] temporary
Class Public methods
new(name, temporary = false, options = nil, as = nil, comment: nil)
# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 219
def initialize(name, temporary = false, options = nil, as = nil, comment: nil)
  @columns_hash = {}
  @indexes = []
  @foreign_keys = []
  @primary_keys = nil
  @temporary = temporary
  @options = options
  @as = as
  @name = name
  @comment = comment
end
Instance Public methods
[](name)

Returns a ColumnDefinition for the column with name name.

# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 240
def [](name)
  @columns_hash[name.to_s]
end
belongs_to(*args, **options)
Alias for: references
column(name, type, options = {})

Instantiates a new column for the table. See connection.add_column for available options.

Additional options are:

  • :index - Create an index for the column. Can be either true or an options hash.

This method returns self.

Examples

# Assuming +td+ is an instance of TableDefinition
td.column(:granted, :boolean, index: true)

Short-hand examples

Instead of calling column directly, you can also work with the short-hand definitions for the default types. They use the type as the method name instead of as a parameter and allow for multiple columns to be defined in a single statement.

What can be written like this with the regular calls to column:

create_table :products do |t|
  t.column :shop_id,     :integer
  t.column :creator_id,  :integer
  t.column :item_number, :string
  t.column :name,        :string, default: "Untitled"
  t.column :value,       :string, default: "Untitled"
  t.column :created_at,  :datetime
  t.column :updated_at,  :datetime
end
add_index :products, :item_number

can also be written as follows using the short-hand:

create_table :products do |t|
  t.integer :shop_id, :creator_id
  t.string  :item_number, index: true
  t.string  :name, :value, default: "Untitled"
  t.timestamps null: false
end

There's a short-hand method for each of the type values declared at the top. And then there's #timestamps that'll add created_at and updated_at as datetimes.

#references will add an appropriately-named _id column, plus a corresponding _type column if the :polymorphic option is supplied. If :polymorphic is a hash of options, these will be used when creating the _type column. The :index option will also create an index, similar to calling add_index. So what can be written like this:

create_table :taggings do |t|
  t.integer :tag_id, :tagger_id, :taggable_id
  t.string  :tagger_type
  t.string  :taggable_type, default: 'Photo'
end
add_index :taggings, :tag_id, name: 'index_taggings_on_tag_id'
add_index :taggings, [:tagger_id, :tagger_type]

Can also be written as follows using references:

create_table :taggings do |t|
  t.references :tag, index: { name: 'index_taggings_on_tag_id' }
  t.references :tagger, polymorphic: true, index: true
  t.references :taggable, polymorphic: { default: 'Photo' }
end
# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 311
def column(name, type, options = {})
  name = name.to_s
  type = type.to_sym if type
  options = options.dup

  if @columns_hash[name] && @columns_hash[name].primary_key?
    raise ArgumentError, "you can't redefine the primary key column '#{name}'. To define a custom primary key, pass { id: false } to create_table."
  end

  index_options = options.delete(:index)
  index(name, index_options.is_a?(Hash) ? index_options : {}) if index_options
  @columns_hash[name] = new_column_definition(name, type, options)
  self
end
columns()

Returns an array of ColumnDefinition objects for the columns of the table.

# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 237
def columns; @columns_hash.values; end
index(column_name, options = {})

Adds index options to the indexes hash, keyed by column name This is primarily used to track indexes that need to be created after the table

index(:account_id, name: 'index_projects_on_account_id')
# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 336
def index(column_name, options = {})
  indexes << [column_name, options]
end
references(*args, **options)

Adds a reference.

t.references(:user)
t.belongs_to(:supplier, foreign_key: true)

See connection.add_reference for details of the options you can use.

Also aliased as: belongs_to
# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 364
def references(*args, **options)
  args.each do |ref_name|
    ReferenceDefinition.new(ref_name, options).add_to(self)
  end
end
remove_column(name)

remove the column name from the table.

remove_column(:account_id)
# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 328
def remove_column(name)
  @columns_hash.delete name.to_s
end
timestamps(**options)

Appends :datetime columns :created_at and :updated_at to the table. See connection.add_timestamps

t.timestamps null: false
# File activerecord/lib/active_record/connection_adapters/abstract/schema_definitions.rb, line 351
def timestamps(**options)
  options[:null] = false if options[:null].nil?

  column(:created_at, :datetime, options)
  column(:updated_at, :datetime, options)
end