Module | StateMachine::Integrations::ActiveModel |
In: |
lib/state_machine/integrations/active_model.rb
lib/state_machine/integrations/active_model/versions.rb lib/state_machine/integrations/active_model/observer.rb |
Adds support for integrating state machines with ActiveModel classes.
If using ActiveModel directly within your class, then any one of the following features need to be included in order for the integration to be detected:
Below is an example of a simple state machine defined within an ActiveModel class:
class Vehicle include ActiveModel::Observing include ActiveModel::Validations attr_accessor :state define_attribute_methods [:state] state_machine :initial => :parked do event :ignite do transition :parked => :idling end end end
The examples in the sections below will use the above class as a reference.
By default, no action will be invoked when a state is transitioned. This means that if you want to save changes when transitioning, you must define the action yourself like so:
class Vehicle include ActiveModel::Validations attr_accessor :state state_machine :action => :save do ... end def save # Save changes end end
As mentioned in StateMachine::Machine#state, you can define behaviors, like validations, that only execute for certain states. One important caveat here is that, due to a constraint in ActiveModel‘s validation framework, custom validators will not work as expected when defined to run in multiple states. For example:
class Vehicle include ActiveModel::Validations state_machine do ... state :first_gear, :second_gear do validate :speed_is_legal end end end
In this case, the :speed_is_legal validation will only get run for the :second_gear state. To avoid this, you can define your custom validation like so:
class Vehicle include ActiveModel::Validations state_machine do ... state :first_gear, :second_gear do validate {|vehicle| vehicle.speed_is_legal} end end end
In order to hook in validation support for your model, the ActiveModel::Validations feature must be included. If this is included and an event fails to successfully fire because there are no matching transitions for the object, a validation error is added to the object‘s state attribute to help in determining why it failed.
For example,
vehicle = Vehicle.new vehicle.ignite # => false vehicle.errors.full_messages # => ["State cannot transition via \"ignite\""]
In addition, if you‘re using the ignite! version of the event, then the failure reason (such as the current validation errors) will be included in the exception that gets raised when the event fails. For example, assuming there‘s a validation on a field called name on the class:
vehicle = Vehicle.new vehicle.ignite! # => StateMachine::InvalidTransition: Cannot transition state via :ignite from :parked (Reason(s): Name cannot be blank)
Beware that public event attributes mean that events can be fired whenever mass-assignment is being used. If you want to prevent malicious users from tampering with events through URLs / forms, the attribute should be protected like so:
class Vehicle include ActiveModel::MassAssignmentSecurity attr_accessor :state attr_protected :state_event # attr_accessible ... # Alternative technique state_machine do ... end end
If you want to only have some events be able to fire via mass-assignment, you can build two state machines (one public and one protected) like so:
class Vehicle include ActiveModel::MassAssignmentSecurity attr_accessor :state attr_protected :state_event # Prevent access to events in the first machine state_machine do # Define private events here end # Public machine targets the same state as the private machine state_machine :public_state, :attribute => :state do # Define public events here end end
All before/after transition callbacks defined for ActiveModel models behave in the same way that other ActiveSupport callbacks behave. The object involved in the transition is passed in as an argument.
For example,
class Vehicle include ActiveModel::Validations attr_accessor :state state_machine :initial => :parked do before_transition any => :idling do |vehicle| vehicle.put_on_seatbelt end before_transition do |vehicle, transition| # log message end event :ignite do transition :parked => :idling end end def put_on_seatbelt ... end end
Note, also, that the transition can be accessed by simply defining additional arguments in the callback block.
In order to hook in observer support for your application, the ActiveModel::Observing feature must be included. Because of the way ActiveModel observers are designed, there is less flexibility around the specific transitions that can be hooked in. However, a large number of hooks are supported. For example, if a transition for a object‘s state attribute changes the state from parked to idling via the ignite event, the following observer methods are supported:
The following class shows an example of some of these hooks:
class VehicleObserver < ActiveModel::Observer # Callback for :ignite event *before* the transition is performed def before_ignite(vehicle, transition) # log message end # Callback for :ignite event *after* the transition has been performed def after_ignite(vehicle, transition) # put on seatbelt end # Generic transition callback *before* the transition is performed def after_transition(vehicle, transition) Audit.log(vehicle, transition) end def after_failure_to_transition(vehicle, transition) Audit.error(vehicle, transition) end end
More flexible transition callbacks can be defined directly within the model as described in StateMachine::Machine#before_transition and StateMachine::Machine#after_transition.
To define a single observer for multiple state machines:
class StateMachineObserver < ActiveModel::Observer observe Vehicle, Switch, Project def after_transition(object, transition) Audit.log(object, transition) end end
Any error message that is generated from performing invalid transitions can be localized. The following default translations are used:
en: activemodel: errors: messages: invalid: "is invalid" # %{value} = attribute value, %{state} = Human state name invalid_event: "cannot transition when %{state}" # %{value} = attribute value, %{event} = Human event name, %{state} = Human current state name invalid_transition: "cannot transition via %{event}"
You can override these for a specific model like so:
en: activemodel: errors: models: user: invalid: "is not valid"
In addition to the above, you can also provide translations for the various states / events in each state machine. Using the Vehicle example, state translations will be looked for using the following keys, where model_name = "vehicle", machine_name = "state" and state_name = "parked":
Event translations will be looked for using the following keys, where model_name = "vehicle", machine_name = "state" and event_name = "ignite":
An example translation configuration might look like so:
es: activemodel: state_machines: states: parked: 'estacionado' events: park: 'estacionarse'
When using the ActiveModel::Dirty extension, your model will keep track of any changes that are made to attributes. Depending on your ORM, an object will only be saved when there are attributes that have changed on the object. When integrating with state_machine, typically the state field will be marked as dirty after a transition occurs. In some situations, however, this isn‘t the case.
If you define loopback transitions in your state machine, the value for the machine‘s attribute (e.g. state) will not change. Unless you explicitly indicate so, this means that your object won‘t persist anything on a loopback. For example:
class Vehicle include ActiveModel::Validations include ActiveModel::Dirty attr_accessor :state state_machine :initial => :parked do event :park do transition :parked => :parked, ... end end end
If, instead, you‘d like your object to always persist regardless of whether the value actually changed, you can do so by using the #{attribute}_will_change! helpers or defining a before_transition callback that actually changes an attribute on the model. For example:
class Vehicle ... state_machine :initial => :parked do before_transition all => same do |vehicle| vehicle.state_will_change! # Alternative solution, updating timestamp # vehicle.updated_at = Time.curent end end end
If you want to integrate state_machine with an ORM that implements parts or all of the ActiveModel API, only the machine defaults need to be specified. Otherwise, the implementation is similar to any other integration.
For example,
module StateMachine::Integrations::MyORM include StateMachine::Integrations::ActiveModel @defaults = {:action = > :persist} def self.matches?(klass) defined?(::MyORM::Base) && klass <= ::MyORM::Base end protected def runs_validations_on_action? action == :persist end end
If you wish to implement other features, such as attribute initialization with protected attributes, named scopes, or database transactions, you must add these independent of the ActiveModel integration. See the ActiveRecord implementation for examples of these customizations.
# File lib/state_machine/integrations/active_model/versions.rb, line 5 5: def self.active? 6: !defined?(::ActiveModel::VERSION) || ::ActiveModel::VERSION::MAJOR == 2 7: end
# File lib/state_machine/integrations/active_model/versions.rb, line 20 20: def self.active? 21: defined?(::ActiveModel::VERSION) && ::ActiveModel::VERSION::MAJOR == 3 && ::ActiveModel::VERSION::MINOR == 0 22: end
Whether this integration is available. Only true if ActiveModel is defined.
# File lib/state_machine/integrations/active_model.rb, line 371 371: def self.available? 372: defined?(::ActiveModel) 373: end
Should this integration be used for state machines in the given class? Classes that include ActiveModel::Observing or ActiveModel::Validations will automatically use the ActiveModel integration.
# File lib/state_machine/integrations/active_model.rb, line 378 378: def self.matches?(klass) 379: %w(Observing Validations).any? {|feature| ::ActiveModel.const_defined?(feature) && klass <= ::ActiveModel.const_get(feature)} 380: end
# File lib/state_machine/integrations/active_model/versions.rb, line 9 9: def define_validation_hook 10: define_helper :instance, "def valid?(*)\nself.class.state_machines.transitions(self, \#{action.inspect}, :after => false).perform { super }\nend\n", __FILE__, __LINE__ + 1 11: end
# File lib/state_machine/integrations/active_model/versions.rb, line 24 24: def define_validation_hook 25: # +around+ callbacks don't have direct access to results until AS 3.1 26: owner_class.set_callback(:validation, :after, 'value', :prepend => true) 27: super 28: end
Describes the current validation errors on the given object. If none are specific, then the default error is interpeted as a "halt".
# File lib/state_machine/integrations/active_model.rb, line 398 398: def errors_for(object) 399: object.errors.empty? ? 'Transition halted' : object.errors.full_messages * ', ' 400: end
Adds a validation error to the given object
# File lib/state_machine/integrations/active_model.rb, line 383 383: def invalidate(object, attribute, message, values = []) 384: if supports_validations? 385: attribute = self.attribute(attribute) 386: options = values.inject({}) do |options, (key, value)| 387: options[key] = value 388: options 389: end 390: 391: default_options = default_error_message_options(object, attribute, message) 392: object.errors.add(attribute, message, options.merge(default_options)) 393: end 394: end
Resets any errors previously added when invalidating the given object
# File lib/state_machine/integrations/active_model.rb, line 403 403: def reset(object) 404: object.errors.clear if supports_validations? 405: end
Creates a new callback in the callback chain, always inserting it before the default Observer callbacks that were created after initialization.
# File lib/state_machine/integrations/active_model.rb, line 526 526: def add_callback(type, options, &block) 527: options[:terminator] = callback_terminator 528: 529: if supports_observers? 530: @callbacks[type == :around ? :before : type].insert(-2, callback = Callback.new(type, options, &block)) 531: add_states(callback.known_states) 532: callback 533: else 534: super 535: end 536: end
Adds a set of default callbacks that utilize the Observer extensions
# File lib/state_machine/integrations/active_model.rb, line 488 488: def add_default_callbacks 489: if supports_observers? 490: callbacks[:before] << Callback.new(:before) {|object, transition| notify(:before, object, transition)} 491: callbacks[:after] << Callback.new(:after) {|object, transition| notify(:after, object, transition)} 492: callbacks[:failure] << Callback.new(:failure) {|object, transition| notify(:after_failure_to, object, transition)} 493: end 494: end
Configures new event with the built-in humanize scheme
# File lib/state_machine/integrations/active_model.rb, line 546 546: def add_events(new_events) 547: super.each do |event| 548: event.human_name = lambda {|event, klass| translate(klass, :event, event.name)} 549: end 550: end
Configures new states with the built-in humanize scheme
# File lib/state_machine/integrations/active_model.rb, line 539 539: def add_states(new_states) 540: super.each do |state| 541: state.human_name = lambda {|state, klass| translate(klass, :state, state.name)} 542: end 543: end
Initializes class-level extensions and defaults for this machine
# File lib/state_machine/integrations/active_model.rb, line 470 470: def after_initialize 471: super 472: load_locale 473: load_observer_extensions 474: add_default_callbacks 475: end
Build a list of ancestors for the given class to use when determining which localization key to use for a particular string.
# File lib/state_machine/integrations/active_model.rb, line 465 465: def ancestors_for(klass) 466: klass.lookup_ancestors 467: end
Runs state events around the object‘s validation process
# File lib/state_machine/integrations/active_model.rb, line 519 519: def around_validation(object) 520: object.class.state_machines.transitions(object, action, :after => false).perform { yield } 521: end
Gets the terminator to use for callbacks
# File lib/state_machine/integrations/active_model.rb, line 428 428: def callback_terminator 429: @terminator ||= lambda {|result| result == false} 430: end
The default options to use when generating messages for validation errors
# File lib/state_machine/integrations/active_model.rb, line 439 439: def default_error_message_options(object, attribute, message) 440: {:message => @messages[message]} 441: end
Adds hooks into validation for automatically firing events
# File lib/state_machine/integrations/active_model.rb, line 507 507: def define_action_helpers 508: super 509: define_validation_hook if runs_validations_on_action? 510: end
Skips defining reader/writer methods since this is done automatically
# File lib/state_machine/integrations/active_model.rb, line 497 497: def define_state_accessor 498: name = self.name 499: 500: owner_class.validates_each(attribute) do |object, attr, value| 501: machine = object.class.state_machine(name) 502: machine.invalidate(object, :state, :invalid) unless machine.states.match(object) 503: end if supports_validations? 504: end
Hooks into validations by defining around callbacks for the :validation event
# File lib/state_machine/integrations/active_model.rb, line 514 514: def define_validation_hook 515: owner_class.set_callback(:validation, :around, self, :prepend => true) 516: end
Determines the base scope to use when looking up translations
# File lib/state_machine/integrations/active_model.rb, line 433 433: def i18n_scope(klass) 434: klass.i18n_scope 435: end
Loads any locale files needed for translating validation errors
# File lib/state_machine/integrations/active_model.rb, line 478 478: def load_locale 479: I18n.load_path.unshift(@integration.locale_path) unless I18n.load_path.include?(@integration.locale_path) 480: end
Loads extensions to ActiveModel‘s Observers
# File lib/state_machine/integrations/active_model.rb, line 483 483: def load_observer_extensions 484: require 'state_machine/integrations/active_model/observer' 485: end
Notifies observers on the given object that a callback occurred involving the given transition. This will attempt to call the following methods on observers:
This will always return true regardless of the results of the callbacks.
# File lib/state_machine/integrations/active_model.rb, line 567 567: def notify(type, object, transition) 568: name = self.name 569: event = transition.qualified_event 570: from = transition.from_name || 'nil' 571: to = transition.to_name || 'nil' 572: 573: # Machine-specific updates 574: ["#{type}_#{event}", "#{type}_transition_#{name}"].each do |event_segment| 575: ["_from_#{from}", nil].each do |from_segment| 576: ["_to_#{to}", nil].each do |to_segment| 577: object.class.changed if object.class.respond_to?(:changed) 578: object.class.notify_observers('update_with_transition', [[event_segment, from_segment, to_segment].join, object, transition]) 579: end 580: end 581: end 582: 583: # Generic updates 584: object.class.changed if object.class.respond_to?(:changed) 585: object.class.notify_observers('update_with_transition', ["#{type}_transition", object, transition]) 586: 587: true 588: end
Do validations run when the action configured this machine is invoked? This is used to determine whether to fire off attribute-based event transitions when the action is run.
# File lib/state_machine/integrations/active_model.rb, line 423 423: def runs_validations_on_action? 424: false 425: end
Whether observers are supported in the integration. Only true if ActiveModel::Observer is available.
# File lib/state_machine/integrations/active_model.rb, line 410 410: def supports_observers? 411: defined?(::ActiveModel::Observing) && owner_class <= ::ActiveModel::Observing 412: end
Whether validations are supported in the integration. Only true if the ActiveModel feature is enabled on the owner class.
# File lib/state_machine/integrations/active_model.rb, line 416 416: def supports_validations? 417: defined?(::ActiveModel::Validations) && owner_class <= ::ActiveModel::Validations 418: end
Translates the given key / value combo. Translation keys are looked up in the following order:
If no keys are found, then the humanized value will be the fallback.
# File lib/state_machine/integrations/active_model.rb, line 451 451: def translate(klass, key, value) 452: ancestors = ancestors_for(klass) 453: group = key.to_s.pluralize 454: value = value ? value.to_s : 'nil' 455: 456: # Generate all possible translation keys 457: translations = ancestors.map {|ancestor| "#{ancestor.model_name.underscore}.#{name}.#{group}.#{value}""#{ancestor.model_name.underscore}.#{name}.#{group}.#{value}"} 458: translations.concat(ancestors.map {|ancestor| "#{ancestor.model_name.underscore}.#{group}.#{value}""#{ancestor.model_name.underscore}.#{group}.#{value}"}) 459: translations.concat(["#{name}.#{group}.#{value}""#{name}.#{group}.#{value}", "#{group}.#{value}""#{group}.#{value}", value.humanize.downcase]) 460: I18n.translate(translations.shift, :default => translations, :scope => [i18n_scope(klass), :state_machines]) 461: end