module ActiveRecord::Core
Active Record Core
Attributes
Public Class Methods
Source
# File lib/active_record/core.rb, line 119 class_attribute :attributes_for_inspect, instance_accessor: false, default: :all
Specifies the attributes that will be included in the output of the inspect method:
Post.attributes_for_inspect = [:id, :title] Post.first.inspect #=> "#<Post id: 1, title: "Hello, World!">"
When set to â:all` inspect will list all the recordâs attributes:
Post.attributes_for_inspect = :all Post.first.inspect #=> "#<Post id: 1, title: "Hello, World!", published_at: "2023-10-23 14:28:11 +0000">"
Source
# File lib/active_record/core.rb, line 77 def self.configurations @@configurations end
Returns a fully resolved ActiveRecord::DatabaseConfigurations object.
Source
# File lib/active_record/core.rb, line 71 def self.configurations=(config) @@configurations = ActiveRecord::DatabaseConfigurations.new(config) end
Contains the database configuration - as is typically stored in config/database.yml - as an ActiveRecord::DatabaseConfigurations object.
For example, the following database.ymlâŚ
development: adapter: sqlite3 database: storage/development.sqlite3 production: adapter: sqlite3 database: storage/production.sqlite3
âŚwould result in ActiveRecord::Base.configurations to look like this:
#<ActiveRecord::DatabaseConfigurations:0x00007fd1acbdf800 @configurations=[
#<ActiveRecord::DatabaseConfigurations::HashConfig:0x00007fd1acbded10 @env_name="development",
@name="primary", @config={adapter: "sqlite3", database: "storage/development.sqlite3"}>,
#<ActiveRecord::DatabaseConfigurations::HashConfig:0x00007fd1acbdea90 @env_name="production",
@name="primary", @config={adapter: "sqlite3", database: "storage/production.sqlite3"}>
]>
Source
# File lib/active_record/core.rb, line 133 def self.connection_handler ActiveSupport::IsolatedExecutionState[:active_record_connection_handler] || default_connection_handler end
Source
# File lib/active_record/core.rb, line 137 def self.connection_handler=(handler) ActiveSupport::IsolatedExecutionState[:active_record_connection_handler] = handler end
Source
# File lib/active_record/core.rb, line 196 def self.current_preventing_writes connected_to_stack.reverse_each do |hash| return hash[:prevent_writes] if !hash[:prevent_writes].nil? && hash[:klasses].include?(Base) return hash[:prevent_writes] if !hash[:prevent_writes].nil? && hash[:klasses].include?(connection_class_for_self) end false end
Returns the symbol representing the current setting for preventing writes.
ActiveRecord::Base.connected_to(role: :reading) do ActiveRecord::Base.current_preventing_writes #=> true end ActiveRecord::Base.connected_to(role: :writing) do ActiveRecord::Base.current_preventing_writes #=> false end
Source
# File lib/active_record/core.rb, line 159 def self.current_role connected_to_stack.reverse_each do |hash| return hash[:role] if hash[:role] && hash[:klasses].include?(Base) return hash[:role] if hash[:role] && hash[:klasses].include?(connection_class_for_self) end default_role end
Returns the symbol representing the current connected role.
ActiveRecord::Base.connected_to(role: :writing) do ActiveRecord::Base.current_role #=> :writing end ActiveRecord::Base.connected_to(role: :reading) do ActiveRecord::Base.current_role #=> :reading end
Source
# File lib/active_record/core.rb, line 177 def self.current_shard connected_to_stack.reverse_each do |hash| return hash[:shard] if hash[:shard] && hash[:klasses].include?(Base) return hash[:shard] if hash[:shard] && hash[:klasses].include?(connection_class_for_self) end default_shard end
Returns the symbol representing the current connected shard.
ActiveRecord::Base.connected_to(role: :reading) do ActiveRecord::Base.current_shard #=> :default end ActiveRecord::Base.connected_to(role: :writing, shard: :one) do ActiveRecord::Base.current_shard #=> :one end
Source
# File lib/active_record/core.rb, line 47 class_attribute :destroy_association_async_batch_size, instance_writer: false, instance_predicate: false, default: nil
Specifies the maximum number of records that will be destroyed in a single background job by the dependent: :destroy_async association option. When nil (default), all dependent records will be destroyed in a single background job. If specified, the records to be destroyed will be split into multiple background jobs.
Source
# File lib/active_record/core.rb, line 27 def self.destroy_association_async_job if _destroy_association_async_job.is_a?(String) self._destroy_association_async_job = _destroy_association_async_job.constantize end _destroy_association_async_job rescue NameError => error raise NameError, "Unable to load destroy_association_async_job: #{error.message}" end
The job class used to destroy associations in the background.
Source
# File lib/active_record/core.rb, line 87 class_attribute :enumerate_columns_in_select_statements, instance_accessor: false, default: false
Force enumeration of all columns in SELECT statements. e.g. SELECT first_name, last_name FROM ... instead of SELECT * FROM ... This avoids PreparedStatementCacheExpired errors when a column is added to the database while the app is running.
Source
# File lib/active_record/core.rb, line 22 class_attribute :logger, instance_writer: false
Accepts a logger conforming to the interface of Log4r or the default Ruby Logger class, which is then passed on to any new database connections made. You can retrieve this logger by calling logger on either an Active Record model class or an Active Record model instance.
Source
# File lib/active_record/core.rb, line 471 def initialize(attributes = nil) @new_record = true @attributes = self.class._default_attributes.deep_dup init_internals initialize_internals_callback super yield self if block_given? _run_initialize_callbacks end
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 â hence you canât have attributes that arenât part of the table columns.
Example
# Instantiates a single new object User.new(first_name: 'Jamie')
Public Instance Methods
Source
# File lib/active_record/core.rb, line 665 def <=>(other_object) if other_object.is_a?(self.class) to_key <=> other_object.to_key else super end end
Allows sort on objects
Source
# File lib/active_record/core.rb, line 631 def ==(comparison_object) super || comparison_object.instance_of?(self.class) && primary_key_values_present? && comparison_object.id == id end
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 comparison_object.id.
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, youâre 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.
Source
# File lib/active_record/core.rb, line 523
Identical to Rubyâs clone method. This is a âshallowâ copy. Be warned that your attributes are not copied. That means that modifying attributes of the clone will modify the original, since they will both point to the same attributes hash. If you need a copy of your attributes hash, please use the dup method.
user = User.first new_user = user.clone user.name # => "Bob" new_user.name = "Joe" user.name # => "Joe" user.object_id == new_user.object_id # => false user.name.object_id == new_user.name.object_id # => true user.name.object_id == user.dup.name.object_id # => false
Source
# File lib/active_record/core.rb, line 770 def connection_handler self.class.connection_handler end
Source
# File lib/active_record/core.rb, line 540
Duped objects have no id assigned and are treated as new records. Note that this is a âshallowâ copy as it copies the objectâs attributes only, not its associations. The extent of a âdeepâ 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) and locking column.
Source
# File lib/active_record/core.rb, line 587 def encode_with(coder) self.class.yaml_encoder.encode(@attributes, coder) coder["new_record"] = new_record? coder["active_record_yaml_version"] = 2 end
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.
Example:
class Post < ActiveRecord::Base end coder = {} Post.new.encode_with(coder) coder # => {"attributes" => {"id" => nil, ... }}
Source
# File lib/active_record/core.rb, line 654 def freeze @attributes = @attributes.clone.freeze self end
Clone and freeze the attributes hash such that associations are still accessible, even on destroyed records, but cloned models will not be frozen.
Source
# File lib/active_record/core.rb, line 660 def frozen? @attributes.frozen? end
Returns true if the attributes hash has been frozen.
Source
# File lib/active_record/core.rb, line 794 def full_inspect inspect_with_attributes(all_attributes_for_inspect) end
Returns all attributes of the record as a nicely formatted string, ignoring .attributes_for_inspect.
Post.first.full_inspect #=> "#<Post id: 1, title: "Hello, World!", published_at: "2023-10-23 14:28:11 +0000">"
Source
# File lib/active_record/core.rb, line 641 def hash id = self.id if primary_key_values_present? self.class.hash ^ id.hash else super end end
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) ]
Source
# File lib/active_record/core.rb, line 498 def init_with(coder, &block) coder = LegacyYamlAdapter.convert(coder) attributes = self.class.yaml_encoder.decode(coder) init_with_attributes(attributes, coder["new_record"], &block) end
Initialize an empty model object from coder. coder should be the result of previously encoding an Active Record model, using encode_with.
class Post < ActiveRecord::Base end old_post = Post.new(title: "hello world") coder = {} old_post.encode_with(coder) post = Post.allocate post.init_with(coder) post.title # => 'hello world'
Source
# File lib/active_record/core.rb, line 784 def inspect inspect_with_attributes(attributes_for_inspect) end
Returns the attributes of the record as a nicely formatted string.
Post.first.inspect #=> "#<Post id: 1, title: "Hello, World!", published_at: "2023-10-23 14:28:11 +0000">"
The attributes can be limited by setting .attributes_for_inspect.
Post.attributes_for_inspect = [:id, :title] Post.first.inspect #=> "#<Post id: 1, title: "Hello, World!">"
Source
# File lib/active_record/core.rb, line 800 def pretty_print(pp) return super if custom_inspect_method_defined? pp.object_address_group(self) do if @attributes attr_names = attributes_for_inspect.select { |name| _has_attribute?(name.to_s) } pp.seplist(attr_names, proc { pp.text "," }) do |attr_name| attr_name = attr_name.to_s pp.breakable " " pp.group(1) do pp.text attr_name pp.text ":" pp.breakable value = attribute_for_inspect(attr_name) pp.text value end end else pp.breakable " " pp.text "not initialized" end end end
Takes a PP and prettily prints this record to it, allowing you to get a nice result from pp record when pp is required.
Source
# File lib/active_record/core.rb, line 766 def readonly! @readonly = true end
Prevents records from being written to the database:
customer = Customer.new customer.readonly! customer.save # raises ActiveRecord::ReadOnlyRecord customer = Customer.first customer.readonly! customer.update(name: 'New Name') # raises ActiveRecord::ReadOnlyRecord
Read-only records cannot be deleted from the database either:
customer = Customer.first customer.readonly! customer.destroy # raises ActiveRecord::ReadOnlyRecord
Please, note that the objects themselves are still mutable in memory:
customer = Customer.new customer.readonly! customer.name = 'New Name' # OK
but you wonât be able to persist the changes.
Source
# File lib/active_record/core.rb, line 682 def readonly? @readonly end
Returns true if the record is read only.
Source
# File lib/active_record/core.rb, line 594
Returns a hash of the given methods with their names as keys and returned values as values.
topic = Topic.new(title: "Budget", author_name: "Jason") topic.slice(:title, :author_name) # => { "title" => "Budget", "author_name" => "Jason" }
Source
# File lib/active_record/core.rb, line 722 def strict_loading!(value = true, mode: :all) unless [:all, :n_plus_one_only].include?(mode) raise ArgumentError, "The :mode option must be one of [:all, :n_plus_one_only] but #{mode.inspect} was provided." end @strict_loading_mode = mode @strict_loading = value end
Sets the record to strict_loading mode. This will raise an error if the record tries to lazily load an association.
NOTE: Strict loading is disabled during validation in order to let the record validate its association.
user = User.first user.strict_loading! # => true user.address.city # => ActiveRecord::StrictLoadingViolationError user.comments.to_a # => ActiveRecord::StrictLoadingViolationError
Parameters
-
value- Boolean specifying whether to enable or disable strict loading. -
:mode- Symbol specifying strict loading mode. Defaults to :all. Using :n_plus_one_only mode will only raise an error if an association that will lead to an n plus one query is lazily loaded.
Examples
user = User.first user.strict_loading!(false) # => false user.address.city # => "Tatooine" user.comments.to_a # => [#<Comment:0x00...] user.strict_loading!(mode: :n_plus_one_only) user.address.city # => "Tatooine" user.comments.to_a # => [#<Comment:0x00...] user.comments.first.ratings.to_a # => ActiveRecord::StrictLoadingViolationError
Source
# File lib/active_record/core.rb, line 687 def strict_loading? @strict_loading end
Returns true if the record is in strict_loading mode.
Source
# File lib/active_record/core.rb, line 739 def strict_loading_all? @strict_loading_mode == :all end
Returns true if the record uses strict_loading with :all mode enabled.
Source
# File lib/active_record/core.rb, line 734 def strict_loading_n_plus_one_only? @strict_loading_mode == :n_plus_one_only end
Returns true if the record uses strict_loading with :n_plus_one_only mode enabled.
Source
# File lib/active_record/core.rb, line 609
Returns an array of the values returned by the given methods.
topic = Topic.new(title: "Budget", author_name: "Jason") topic.values_at(:title, :author_name) # => ["Budget", "Jason"]
Private Instance Methods
Source
# File lib/active_record/core.rb, line 890 def all_attributes_for_inspect return [] unless @attributes attribute_names end
Source
# File lib/active_record/core.rb, line 886 def attributes_for_inspect self.class.attributes_for_inspect == :all ? all_attributes_for_inspect : self.class.attributes_for_inspect end
Source
# File lib/active_record/core.rb, line 856 def custom_inspect_method_defined? self.class.instance_method(:inspect).owner != ActiveRecord::Base.instance_method(:inspect).owner end
Source
# File lib/active_record/core.rb, line 836 def init_internals @readonly = false @previously_new_record = false @destroyed = false @marked_for_destruction = false @destroyed_by_association = nil @_start_transaction_state = nil klass = self.class @primary_key = klass.primary_key @strict_loading = klass.strict_loading_by_default @strict_loading_mode = klass.strict_loading_mode klass.define_attribute_methods end
Source
# File lib/active_record/core.rb, line 853 def initialize_internals_callback end
Source
# File lib/active_record/core.rb, line 871 def inspect_with_attributes(attributes_to_list) inspection = if @attributes attributes_to_list.filter_map do |name| name = name.to_s if _has_attribute?(name) "#{name}: #{attribute_for_inspect(name)}" end end.join(", ") else "not initialized" end "#<#{self.class} #{inspection}>" end
Source
# File lib/active_record/core.rb, line 867 def inspection_filter self.class.inspection_filter end
Source
# File lib/active_record/core.rb, line 832 def to_ary nil end
+Array#flatten+ will call to_ary (recursively) on each of the elements of the array, and then rescues from the possible NoMethodError. If those elements are ActiveRecord::Baseâs, then this triggers the various method_missingâs that we have, which significantly impacts upon performance.
So we can avoid the method_missing hit by explicitly defining to_ary as nil here.
See also tenderlovemaking.com/2011/06/28/til-its-ok-to-return-nil-from-to_ary.html