Callbacks are hooks into the lifecycle of an Active Record object that allow you to trigger logic before or after an alteration of the object state. This can be used to make sure that associated and dependent objects are deleted when destroy is called (by overwriting before_destroy) or to massage attributes before they‘re validated (by overwriting before_validation). As an example of the callbacks initiated, consider the Base#save call for a new record:

That‘s a total of eight callbacks, which gives you immense power to react and prepare for each state in the Active Record lifecycle. The sequence for calling Base#save an existing record is similar, except that each _on_create callback is replaced by the corresponding _on_update callback.

Examples:

  class CreditCard < ActiveRecord::Base
    # Strip everything but digits, so the user can specify "555 234 34" or
    # "5552-3434" or both will mean "55523434"
    def before_validation_on_create
      self.number = number.gsub(/[^0-9]/, "") if attribute_present?("number")
    end
  end

  class Subscription < ActiveRecord::Base
    before_create :record_signup

    private
      def record_signup
        self.signed_up_on = Date.today
      end
  end

  class Firm < ActiveRecord::Base
    # Destroys the associated clients and people when the firm is destroyed
    before_destroy { |record| Person.destroy_all "firm_id = #{record.id}"   }
    before_destroy { |record| Client.destroy_all "client_of = #{record.id}" }
  end

Inheritable callback queues

Besides the overwritable callback methods, it‘s also possible to register callbacks through the use of the callback macros. Their main advantage is that the macros add behavior into a callback queue that is kept intact down through an inheritance hierarchy. Example:

  class Topic < ActiveRecord::Base
    before_destroy :destroy_author
  end

  class Reply < Topic
    before_destroy :destroy_readers
  end

Now, when Topic#destroy is run only destroy_author is called. When Reply#destroy is run, both destroy_author and destroy_readers are called. Contrast this to the situation where we‘ve implemented the save behavior through overwriteable methods:

  class Topic < ActiveRecord::Base
    def before_destroy() destroy_author end
  end

  class Reply < Topic
    def before_destroy() destroy_readers end
  end

In that case, Reply#destroy would only run destroy_readers and not destroy_author. So, use the callback macros when you want to ensure that a certain callback is called for the entire hierarchy, and use the regular overwriteable methods when you want to leave it up to each descendant to decide whether they want to call super and trigger the inherited callbacks.

*IMPORTANT:* In order for inheritance to work for the callback queues, you must specify the callbacks before specifying the associations. Otherwise, you might trigger the loading of a child before the parent has registered the callbacks and they won‘t be inherited.

Types of callbacks

There are four types of callbacks accepted by the callback macros: Method references (symbol), callback objects, inline methods (using a proc), and inline eval methods (using a string). Method references and callback objects are the recommended approaches, inline methods using a proc are sometimes appropriate (such as for creating mix-ins), and inline eval methods are deprecated.

The method reference callbacks work by specifying a protected or private method available in the object, like this:

  class Topic < ActiveRecord::Base
    before_destroy :delete_parents

    private
      def delete_parents
        self.class.delete_all "parent_id = #{id}"
      end
  end

The callback objects have methods named after the callback called with the record as the only parameter, such as:

  class BankAccount < ActiveRecord::Base
    before_save      EncryptionWrapper.new
    after_save       EncryptionWrapper.new
    after_initialize EncryptionWrapper.new
  end

  class EncryptionWrapper
    def before_save(record)
      record.credit_card_number = encrypt(record.credit_card_number)
    end

    def after_save(record)
      record.credit_card_number = decrypt(record.credit_card_number)
    end

    alias_method :after_find, :after_save

    private
      def encrypt(value)
        # Secrecy is committed
      end

      def decrypt(value)
        # Secrecy is unveiled
      end
  end

So you specify the object you want messaged on a given callback. When that callback is triggered, the object has a method by the name of the callback messaged. You can make these callbacks more flexible by passing in other initialization data such as the name of the attribute to work with:

  class BankAccount < ActiveRecord::Base
    before_save      EncryptionWrapper.new("credit_card_number")
    after_save       EncryptionWrapper.new("credit_card_number")
    after_initialize EncryptionWrapper.new("credit_card_number")
  end

  class EncryptionWrapper
    def initialize(attribute)
      @attribute = attribute
    end

    def before_save(record)
      record.send("#{@attribute}=", encrypt(record.send("#{@attribute}")))
    end

    def after_save(record)
      record.send("#{@attribute}=", decrypt(record.send("#{@attribute}")))
    end

    alias_method :after_find, :after_save

    private
      def encrypt(value)
        # Secrecy is committed
      end

      def decrypt(value)
        # Secrecy is unveiled
      end
  end

The callback macros usually accept a symbol for the method they‘re supposed to run, but you can also pass a "method string", which will then be evaluated within the binding of the callback. Example:

  class Topic < ActiveRecord::Base
    before_destroy 'self.class.delete_all "parent_id = #{id}"'
  end

Notice that single quotes (’) are used so the #{id} part isn‘t evaluated until the callback is triggered. Also note that these inline callbacks can be stacked just like the regular ones:

  class Topic < ActiveRecord::Base
    before_destroy 'self.class.delete_all "parent_id = #{id}"',
                   'puts "Evaluated after parents are destroyed"'
  end

The after_find and after_initialize exceptions

Because after_find and after_initialize are called for each object found and instantiated by a finder, such as Base.find(:all), we‘ve had to implement a simple performance constraint (50% more speed on a simple test case). Unlike all the other callbacks, after_find and after_initialize will only be run if an explicit implementation is defined (def after_find). In that case, all of the callback types will be called.

before_validation* returning statements

If the returning value of a before_validation callback can be evaluated to false, the process will be aborted and Base#save will return false. If Base#save! is called it will raise a ActiveRecord::RecordInvalid exception. Nothing will be appended to the errors object.

Canceling callbacks

If a before_* callback returns false, all the later callbacks and the associated action are cancelled. If an after_* callback returns false, all the later callbacks are cancelled. Callbacks are generally run in the order they are defined, with the exception of callbacks defined as methods on the model, which are called last.

Transactions

The entire callback chain of a save, save!, or destroy call runs within a transaction. That includes after_* hooks. If everything goes fine a COMMIT is executed once the chain has been completed.

If a before_* callback cancels the action a ROLLBACK is issued. You can also trigger a ROLLBACK raising an exception in any of the callbacks, including after_* hooks. Note, however, that in that case the client needs to be aware of it because an ordinary save will raise such exception instead of quietly returning false.

Methods
Constants
CALLBACKS = %w( after_find after_initialize before_save after_save before_create after_create before_update after_update before_validation after_validation before_validation_on_create after_validation_on_create before_validation_on_update after_validation_on_update before_destroy after_destroy )
Public Instance methods
after_create()

Is called after Base.save on new objects that haven‘t been saved yet (no record exists). Note that this callback is still wrapped in the transaction around save. For example, if you invoke an external indexer at this point it won‘t see the changes in the database.

     # File activerecord/lib/active_record/callbacks.rb, line 263
263:     def after_create() end
after_destroy()

Is called after Base.destroy (and all the attributes have been frozen).

 class Contact < ActiveRecord::Base
   after_destroy { |record| logger.info( "Contact #{record.id} was destroyed." ) }
 end
     # File activerecord/lib/active_record/callbacks.rb, line 334
334:     def after_destroy()  end
after_save()

Is called after Base.save (regardless of whether it‘s a create or update save). Note that this callback is still wrapped in the transaction around save. For example, if you invoke an external indexer at this point it won‘t see the changes in the database.

 class Contact < ActiveRecord::Base
   after_save { logger.info( 'New contact saved!' ) }
 end
     # File activerecord/lib/active_record/callbacks.rb, line 247
247:     def after_save()  end
after_update()

Is called after Base.save on existing objects that have a record. Note that this callback is still wrapped in the transaction around save. For example, if you invoke an external indexer at this point it won‘t see the changes in the database.

     # File activerecord/lib/active_record/callbacks.rb, line 278
278:     def after_update() end
after_validation()

Is called after Validations.validate (which is part of the Base.save call).

     # File activerecord/lib/active_record/callbacks.rb, line 292
292:     def after_validation() end
after_validation_on_create()

Is called after Validations.validate (which is part of the Base.save call) on new objects that haven‘t been saved yet (no record exists).

     # File activerecord/lib/active_record/callbacks.rb, line 300
300:     def after_validation_on_create()  end
after_validation_on_update()

Is called after Validations.validate (which is part of the Base.save call) on existing objects that have a record.

     # File activerecord/lib/active_record/callbacks.rb, line 308
308:     def after_validation_on_update()  end
before_create()

Is called before Base.save on new objects that haven‘t been saved yet (no record exists).

     # File activerecord/lib/active_record/callbacks.rb, line 258
258:     def before_create() end
before_destroy()

Is called before Base.destroy.

Note: If you need to destroy or nullify associated records first, use the :dependent option on your associations.

     # File activerecord/lib/active_record/callbacks.rb, line 327
327:     def before_destroy() end
before_save()

Is called before Base.save (regardless of whether it‘s a create or update save).

     # File activerecord/lib/active_record/callbacks.rb, line 238
238:     def before_save() end
before_update()

Is called before Base.save on existing objects that have a record.

     # File activerecord/lib/active_record/callbacks.rb, line 273
273:     def before_update() end
before_validation()

Is called before Validations.validate (which is part of the Base.save call).

     # File activerecord/lib/active_record/callbacks.rb, line 289
289:     def before_validation() end
before_validation_on_create()

Is called before Validations.validate (which is part of the Base.save call) on new objects that haven‘t been saved yet (no record exists).

     # File activerecord/lib/active_record/callbacks.rb, line 296
296:     def before_validation_on_create() end
before_validation_on_update()

Is called before Validations.validate (which is part of the Base.save call) on existing objects that have a record.

     # File activerecord/lib/active_record/callbacks.rb, line 304
304:     def before_validation_on_update() end