Decorator pattern

In object-oriented programming, the decorator pattern is a design pattern that allows behavior to be added to an individual object, dynamically, without affecting the behavior of other instances of the same class. The decorator pattern is often useful for adhering to the Single Responsibility Principle, as it allows functionality to be divided between classes with unique areas of concern as well as to the Open-Closed Principle, by allowing the functionality of a class to be extended without being modified. Decorator use can be more efficient than subclassing, because an object's behavior can be augmented without defining an entirely new object.

Overview
The decorator design pattern is one of the twenty-three well-known design patterns; these describe how to solve recurring design problems and design flexible and reusable object-oriented software—that is, objects which are easier to implement, change, test, and reuse.

What problems can it solve?

 * Responsibilities should be added to (and removed from) an object dynamically at run-time.
 * A flexible alternative to subclassing for extending functionality should be provided.

When using subclassing, different subclasses extend a class in different ways. But an extension is bound to the class at compile-time and can't be changed at run-time.

What solution does it describe?
Define  objects that
 * implement the interface of the extended (decorated) object transparently by forwarding all requests to it
 * perform additional functionality before/after forwarding a request.

This allows working with different  objects to extend the functionality of an object dynamically at run-time.

See also the UML class and sequence diagram below.

Intent
The decorator pattern can be used to extend (decorate) the functionality of a certain object statically, or in some cases at run-time, independently of other instances of the same class, provided some groundwork is done at design time. This is achieved by designing a new Decorator class that wraps the original class. This wrapping could be achieved by the following sequence of steps:
 * 1) Subclass the original Component class into a Decorator class (see UML diagram);
 * 2) In the Decorator class, add a Component pointer as a field;
 * 3) In the Decorator class, pass a Component to the Decorator constructor to initialize the Component pointer;
 * 4) In the Decorator class, forward all Component methods to the Component pointer; and
 * 5) In the ConcreteDecorator class, override any Component method(s) whose behavior needs to be modified.

This pattern is designed so that multiple decorators can be stacked on top of each other, each time adding a new functionality to the overridden method(s).

Note that decorators and the original class object share a common set of features. In the previous diagram, the operation method was available in both the decorated and undecorated versions.

The decoration features (e.g., methods, properties, or other members) are usually defined by an interface, mixin (a.k.a. trait) or class inheritance which is shared by the decorators and the decorated object. In the previous example, the class Component is inherited by both the ConcreteComponent and the subclasses that descend from Decorator.

The decorator pattern is an alternative to subclassing. Subclassing adds behavior at compile time, and the change affects all instances of the original class; decorating can provide new behavior at run-time for selected objects.

This difference becomes most important when there are several independent ways of extending functionality. In some object-oriented programming languages, classes cannot be created at runtime, and it is typically not possible to predict, at design time, what combinations of extensions will be needed. This would mean that a new class would have to be made for every possible combination. By contrast, decorators are objects, created at runtime, and can be combined on a per-use basis. The I/O Streams implementations of both Java and the .NET Framework incorporate the decorator pattern.

Motivation
As an example, consider a window in a windowing system. To allow scrolling of the window's contents, one may wish to add horizontal or vertical scrollbars to it, as appropriate. Assume windows are represented by instances of the Window interface, and assume this class has no functionality for adding scrollbars. One could create a subclass ScrollingWindow that provides them, or create a ScrollingWindowDecorator that adds this functionality to existing Window objects. At this point, either solution would be fine.

Now, assume one also desires the ability to add borders to windows. Again, the original Window class has no support. The ScrollingWindow subclass now poses a problem, because it has effectively created a new kind of window. If one wishes to add border support to many but not all windows, one must create subclasses WindowWithBorder and ScrollingWindowWithBorder, etc. This problem gets worse with every new feature or window subtype to be added. For the decorator solution, a new BorderedWindowDecorator is created. Any combination of ScrollingWindowDecorator or BorderedWindowDecorator can decorate existing windows. If the functionality needs to be added to all Windows, the base class can be modified. On the other hand, sometimes (e.g., using external frameworks) it is not possible, legal, or convenient to modify the base class.

In the previous example, the SimpleWindow and WindowDecorator classes implement the Window interface, which defines the draw method and the getDescription method that are required in this scenario, in order to decorate a window control.

Applying decorators
Adding or removing decorators on command (like a button press) is a common UI pattern, often implemented along with the Command design pattern. For example, a text editing application might have a button to highlight text. On button press, the individual text glyphs currently selected will all be wrapped in decorators that modify their draw function, causing them to be drawn in a highlighted manner (a real implementation would probably also use a demarcation system to maximize efficiency).

Applying or removing decorators based on changes in state is another common use case. Depending on the scope of the state, decorators can be applied or removed in bulk. Similarly, the State design pattern can be implemented using decorators instead of subclassed objects encapsulating the changing functionality. The use of decorators in this manner makes the State object's internal state and functionality more compositional and capable of handling arbitrary complexity.

Usage in Flyweight objects
Decoration is also often used in the Flyweight design pattern. Flyweight objects are divided into two components: an invariant component that is shared between all flyweight objects; and a variant, decorated component that may be partially shared or completely unshared. This partitioning of the flyweight object is intended to reduce memory consumption. The decorators are typically cached and reused as well. The decorators will all contain a common reference to the shared, invariant object. If the decorated state is only partially variant, then the decorators can also be shared to some degree - though care must be taken not to alter their state while they're being used. iOS's UITableView implements the flyweight pattern in this manner - a tableview's reusable cells are decorators that contains a references to a common tableview row object, and the cells are cached / reused.

Obstacles of interfacing with decorators
Applying combinations of decorators in diverse ways to a collection of objects introduces some problems interfacing with the collection in a way that takes full advantage of the functionality added by the decorators. The use of an Adapter or Visitor patterns can be useful in such cases. Interfacing with multiple layers of decorators poses additional challenges and logic of Adapters and Visitors must be designed to account for that.

Architectural relevance
Decorators support a compositional rather than a top-down, hierarchical approach to extending functionality. A decorator makes it possible to add or alter behavior of an interface at run-time. They can be used to wrap objects in a multilayered, arbitrary combination of ways. Doing the same with subclasses means implementing complex networks of multiple inheritance, which is memory-inefficient and at a certain point just cannot scale. Likewise, attempting to implement the same functionality with properties bloats each instance of the object with unnecessary properties. For the above reasons decorators are often considered a memory-efficient alternative to subclassing.

Decorators can also be used to specialize objects which are not subclassable, whose characteristics need to be altered at runtime (as mentioned elsewhere), or generally objects that are lacking in some needed functionality.

Usage in enhancing APIs
The decorator pattern also can augment the Facade pattern. A facade is designed to simply interface with the complex system it encapsulates, but it does not add functionality to the system. However, the wrapping of a complex system provides a space that may be used to introduce new functionality based on the coordination of subcomponents in the system. For example, a facade pattern may unify many different languages dictionaries under one multi-language dictionary interface. The new interface may also provide new functions for translating words between languages. This is a hybrid pattern - the unified interface provides a space for augmentation. Think of decorators as not being limited to wrapping individual objects, but capable of wrapping clusters of objects in this hybrid approach as well.

Alternatives to Decorators
As an alternative to the decorator pattern, the adapter can be used when the wrapper must respect a particular interface and must support polymorphic behavior, and the Facade when an easier or simpler interface to an underlying object is desired.

UML class and sequence diagram


In the above UML class diagram, the abstract  class maintains a reference to the decorated object and forwards all requests to it. This makes  transparent (invisible) to clients of.

Subclasses implement additional behavior that should be added to the  (before/after forwarding a request to it).

The sequence diagram shows the run-time interactions: The  object works through  and   objects to extend the functionality of a   object.

The  calls on, which forwards the request to. performs  after forwarding the request to  and returns to, which performs and returns to the.

C++
This implementation is based on the pre C++98 implementation in the book.

The program output is like

Full example can be tested on a godbolt page.

C++
Two options are presented here: first, a dynamic, runtime-composable decorator (has issues with calling decorated functions unless proxied explicitly) and a decorator that uses mixin inheritance.

Static Decorator (Mixin Inheritance)
This example demonstrates a static Decorator implementation, which is possible due to C++ ability to inherit from the template argument.

First example (window/scrolling scenario)
The following Java example illustrates the use of decorators using the window/scrolling scenario.

The following classes contain the decorators for all  classes, including the decorator classes themselves.

Here's a test program that creates a  instance which is fully decorated (i.e., with vertical and horizontal scrollbars), and prints its description:

The output of this program is "simple window, including vertical scrollbars, including horizontal scrollbars". Notice how the  method of the two decorators first retrieve the decorated  's description and decorates it with a suffix.

Below is the JUnit test class for the Test Driven Development

Second example (coffee making scenario)
The next Java example illustrates the use of decorators using coffee making scenario. In this example, the scenario only includes cost and ingredients.

The following classes contain the decorators for all Coffee classes, including the decorator classes themselves.

Here's a test program that creates a Coffee instance which is fully decorated (with milk and sprinkles), and calculate cost of coffee and prints its ingredients:

The output of this program is given below: Cost: 1.0; Ingredients: Coffee Cost: 1.5; Ingredients: Coffee, Milk Cost: 1.7; Ingredients: Coffee, Milk, Sprinkles

Python
The following Python example, taken from Python Wiki - DecoratorPattern, shows us how to pipeline decorators to dynamically add many behaviors in an object:

Note:

The Decorator Pattern (or an implementation of this design pattern in Python - as the above example) should not be confused with Python Decorators, a language feature of Python. They are different things.

Second to the Python Wiki:

The Decorator Pattern is a pattern described in the Design Patterns Book. It is a way of apparently modifying an object's behavior, by enclosing it inside a decorating object with a similar interface. This is not to be confused with Python Decorators, which is a language feature for dynamically modifying a function or class.

Crystal
Output: Cost: 1.0; Ingredients: Coffee Cost: 1.5; Ingredients: Coffee, Milk Cost: 1.7; Ingredients: Coffee, Milk, Sprinkles

C#
Output: Bike: 'Aluminium Bike + Sport Package + Security Package' Cost: 111

Ruby
Output: Cost: 1.0; Ingredients: Coffee Cost: 1.5; Ingredients: Coffee, Milk Cost: 1.7; Ingredients: Coffee, Milk, Sprinkles