Develop Reference

Noob question: in C# why would a user rely upon an empty method from an interface? [duplicate]

The reason for interfaces truly eludes me. From what I understand, it is kind of a work around for the non-existent multi-inheritance which doesn't exist in C# (or so I was told).
All I see is, you predefine some members and functions, which then have to be re-defined in the class again. Thus making the interface redundant. It just feels like syntactic… well, junk to me (Please no offense meant. Junk as in useless stuff).
In the example given below taken from a different C# interfaces thread on stack overflow, I would just create a base class called Pizza instead of an interface.
easy example (taken from a different stack overflow contribution)
public interface IPizza
{
public void Order();
}
public class PepperoniPizza : IPizza
{
public void Order()
{
//Order Pepperoni pizza
}
}
public class HawaiiPizza : IPizza
{
public void Order()
{
//Order HawaiiPizza
}
}

No one has really explained in plain terms how interfaces are useful, so I'm going to give it a shot (and steal an idea from Shamim's answer a bit).
Lets take the idea of a pizza ordering service. You can have multiple types of pizzas and a common action for each pizza is preparing the order in the system. Each pizza has to be prepared but each pizza is prepared differently. For example, when a stuffed crust pizza is ordered the system probably has to verify certain ingredients are available at the restaurant and set those aside that aren't needed for deep dish pizzas.
When writing this in code, technically you could just do
public class Pizza
{
public void Prepare(PizzaType tp)
{
switch (tp)
{
case PizzaType.StuffedCrust:
// prepare stuffed crust ingredients in system
break;
case PizzaType.DeepDish:
// prepare deep dish ingredients in system
break;
//.... etc.
}
}
}
However, deep dish pizzas (in C# terms) may require different properties to be set in the Prepare() method than stuffed crust, and thus you end up with a lot of optional properties, and the class doesn't scale well (what if you add new pizza types).
The proper way to solve this is to use interface. The interface declares that all Pizzas can be prepared, but each pizza can be prepared differently. So if you have the following interfaces:
public interface IPizza
{
void Prepare();
}
public class StuffedCrustPizza : IPizza
{
public void Prepare()
{
// Set settings in system for stuffed crust preparations
}
}
public class DeepDishPizza : IPizza
{
public void Prepare()
{
// Set settings in system for deep dish preparations
}
}
Now your order handling code does not need to know exactly what types of pizzas were ordered in order to handle the ingredients. It just has:
public PreparePizzas(IList<IPizza> pizzas)
{
foreach (IPizza pizza in pizzas)
pizza.Prepare();
}
Even though each type of pizza is prepared differently, this part of the code doesn't have to care what type of pizza we are dealing with, it just knows that it's being called for pizzas and therefore each call to Prepare will automatically prepare each pizza correctly based on its type, even if the collection has multiple types of pizzas.

The point is that the interface represents a contract. A set of public methods any implementing class has to have. Technically, the interface only governs syntax, i.e. what methods are there, what arguments they get and what they return. Usually they encapsulate semantics as well, although that only by documentation.
You can then have different implementations of an interface and swap them out at will. In your example, since every pizza instance is an IPizza you can use IPizza wherever you handle an instance of an unknown pizza type. Any instance whose type inherits from IPizza is guaranteed to be orderable, as it has an Order() method.
Python is not statically-typed, therefore types are kept and looked up at runtime. So you can try calling an Order() method on any object. The runtime is happy as long as the object has such a method and probably just shrugs and says »Meh.« if it doesn't. Not so in C#. The compiler is responsible for making the correct calls and if it just has some random object the compiler doesn't know yet whether the instance during runtime will have that method. From the compiler's point of view it's invalid since it cannot verify it. (You can do such things with reflection or the dynamic keyword, but that's going a bit far right now, I guess.)
Also note that an interface in the usual sense does not necessarily have to be a C# interface, it could be an abstract class as well or even a normal class (which can come in handy if all subclasses need to share some common code – in most cases, however, interface suffices).

For me, when starting out, the point to these only became clear when you stop looking at them as things to make your code easier/faster to write - this is not their purpose. They have a number of uses:
(This is going to lose the pizza analogy, as it's not very easy to visualise a use of this)
Say you are making a simple game on screen and It will have creatures with which you interact.
A: They can make your code easier to maintain in the future by introducing a loose coupling between your front end and your back end implementation.
You could write this to start with, as there are only going to be trolls:
// This is our back-end implementation of a troll
class Troll
{
void Walk(int distance)
{
//Implementation here
}
}
Front end:
function SpawnCreature()
{
Troll aTroll = new Troll();
aTroll.Walk(1);
}
Two weeks down the line, marketing decide you also need Orcs, as they read about them on twitter, so you would have to do something like:
class Orc
{
void Walk(int distance)
{
//Implementation (orcs are faster than trolls)
}
}
Front end:
void SpawnCreature(creatureType)
{
switch(creatureType)
{
case Orc:
Orc anOrc = new Orc();
anORc.Walk();
case Troll:
Troll aTroll = new Troll();
aTroll.Walk();
}
}
And you can see how this starts to get messy. You could use an interface here so that your front end would be written once and (here's the important bit) tested, and you can then plug in further back end items as required:
interface ICreature
{
void Walk(int distance)
}
public class Troll : ICreature
public class Orc : ICreature
//etc
Front end is then:
void SpawnCreature(creatureType)
{
ICreature creature;
switch(creatureType)
{
case Orc:
creature = new Orc();
case Troll:
creature = new Troll();
}
creature.Walk();
}
The front end now only cares about the interface ICreature - it's not bothered about the internal implementation of a troll or an orc, but only on the fact that they implement ICreature.
An important point to note when looking at this from this point of view is that you could also easily have used an abstract creature class, and from this perspective, this has the same effect.
And you could extract the creation out to a factory:
public class CreatureFactory {
public ICreature GetCreature(creatureType)
{
ICreature creature;
switch(creatureType)
{
case Orc:
creature = new Orc();
case Troll:
creature = new Troll();
}
return creature;
}
}
And our front end would then become:
CreatureFactory _factory;
void SpawnCreature(creatureType)
{
ICreature creature = _factory.GetCreature(creatureType);
creature.Walk();
}
The front end now does not even have to have a reference to the library where Troll and Orc are implemented (providing the factory is in a separate library) - it need know nothing about them whatsoever.
B: Say you have functionality that only some creatures will have in your otherwise homogenous data structure, e.g.
interface ICanTurnToStone
{
void TurnToStone();
}
public class Troll: ICreature, ICanTurnToStone
Front end could then be:
void SpawnCreatureInSunlight(creatureType)
{
ICreature creature = _factory.GetCreature(creatureType);
creature.Walk();
if (creature is ICanTurnToStone)
{
(ICanTurnToStone)creature.TurnToStone();
}
}
C: Usage for dependency injection
Most dependency injection frameworks work when there is a very loose coupling between the front end code and the back end implementation. If we take our factory example above and have our factory implement an interface:
public interface ICreatureFactory {
ICreature GetCreature(string creatureType);
}
Our front end could then have this injected (e.g an MVC API controller) through the constructor (typically):
public class CreatureController : Controller {
private readonly ICreatureFactory _factory;
public CreatureController(ICreatureFactory factory) {
_factory = factory;
}
public HttpResponseMessage TurnToStone(string creatureType) {
ICreature creature = _factory.GetCreature(creatureType);
creature.TurnToStone();
return Request.CreateResponse(HttpStatusCode.OK);
}
}
With our DI framework (e.g. Ninject or Autofac), we can set them up so that at runtime a instance of CreatureFactory will be created whenever an ICreatureFactory is needed in an constructor - this makes our code nice and simple.
It also means that when we write a unit test for our controller, we can provide a mocked ICreatureFactory (e.g. if the concrete implementation required DB access, we don't want our unit tests dependent on that) and easily test the code in our controller.
D: There are other uses e.g. you have two projects A and B that for 'legacy' reasons are not well structured, and A has a reference to B.
You then find functionality in B that needs to call a method already in A. You can't do it using concrete implementations as you get a circular reference.
You can have an interface declared in B that the class in A then implements. Your method in B can be passed an instance of a class that implements the interface with no problem, even though the concrete object is of a type in A.

Examples above don't make much sense. You could accomplish all above examples using classes (abstract class if you want it to behave only as a contract):
public abstract class Food {
public abstract void Prepare();
}
public class Pizza : Food {
public override void Prepare() { /* Prepare pizza */ }
}
public class Burger : Food {
public override void Prepare() { /* Prepare Burger */ }
}
You get the same behavior as with interface. You can create a List<Food> and iterate that w/o knowing what class sits on top.
More adequate example would be multiple inheritance:
public abstract class MenuItem {
public string Name { get; set; }
public abstract void BringToTable();
}
// Notice Soda only inherits from MenuItem
public class Soda : MenuItem {
public override void BringToTable() { /* Bring soda to table */ }
}
// All food needs to be cooked (real food) so we add this
// feature to all food menu items
public interface IFood {
void Cook();
}
public class Pizza : MenuItem, IFood {
public override void BringToTable() { /* Bring pizza to table */ }
public void Cook() { /* Cook Pizza */ }
}
public class Burger : MenuItem, IFood {
public override void BringToTable() { /* Bring burger to table */ }
public void Cook() { /* Cook Burger */ }
}
Then you can use all of them as MenuItem and don't care about how they handle each method call.
public class Waiter {
public void TakeOrder(IEnumerable<MenuItem> order)
{
// Cook first
// (all except soda because soda is not IFood)
foreach (var food in order.OfType<IFood>())
food.Cook();
// Bring them all to the table
// (everything, including soda, pizza and burger because they're all menu items)
foreach (var menuItem in order)
menuItem.BringToTable();
}
}

Simple Explanation with analogy
No interface (Example 1):
No interface (Example 2):
With an interface:
The Problem to Solve: What is the purpose of polymorphism?
Analogy: So I'm a foreperson on a construction site. I don't know which tradesperson is going to walk in. But I tell them what to do.
If it's a carpenter I say: build wooden scaffolding.
If it's a plumber, I say: Set up the pipes
If it's a BJP government bureaucrat, I say, three bags full of cash, sir.
The problem with the above approach is that I have to: (i) know who's walking in that door, and depending on who it is, I have to tell them what to do. This typically makes code harder to maintain or more error prone.
The implications of knowing what to do:
This means if the carpenter's code changes from: BuildScaffolding() to BuildScaffold() (i.e. a slight name change) then I will have to also change the calling class (i.e. the Foreperson class) as well - you'll have to make two changes to the code instead of (basically) just one. With polymorphism you (basically) only need to make one change to achieve the same result.
Secondly you won't have to constantly ask: who are you? ok do this...who are you? ok do that.....polymorphism - it DRYs that code, and is very effective in certain situations:
with polymorphism you can easily add additional classes of tradespeople without changing any existing code. (i.e. the second of the SOLID design principles: Open-close principle).
The solution
Imagine a scenario where, no matter who walks in the door, I can say: "Work()" and they do their respect jobs that they specialise in: the plumber would deal with pipes, and the electrician would deal with wires, and a bureaucrat could specialise in extracting bribes and making double work for everyone else.
The benefit of this approach is that: (i) I don't need to know exactly who is walking in through that door - all i need to know is that they will be a type of tradie and that they can do work, and secondly, (ii) i don't need to know anything about that particular trade. The tradie will take care of that.
So instead of this:
if(electrician) then electrician.FixCablesAndElectricity()
if(plumber) then plumber.IncreaseWaterPressureAndFixLeaks()
if(keralaCustoms) then keralaCustoms.askForBribes()
I can do something like this:
ITradesman tradie = Tradesman.Factory(); // in reality i know it's a plumber, but in the real world you won't know who's on the other side of the tradie assignment.
tradie.Work(); // and then tradie will do the work of a plumber, or electrician etc. depending on what type of tradesman he is. The foreman doesn't need to know anything, apart from telling the anonymous tradie to get to Work()!!
What's the benefit?
The benefit is that if the specific job requirements of the carpenter etc change, then the foreperson won't need to change his code - he doesn't need to know or care. All that matters is that the carpenter knows what is meant by Work(). Secondly, if a new type of construction worker comes onto the job site, then the foreman doesn't need to know anything about the trade - all the foreman cares is if the construction worker (.e.g Welder, Glazier, Tiler etc.) can get some Work() done.
Summary
An interface allows you to get the person to do the work they are assigned to, without you having the knowledge of exactly who they are or the specifics of what they can do. This allows you to easily add new types (of trade) without changing your existing code (well technically you do change it a tiny tiny bit), and that's the real benefit of an OOP approach vs. a more functional programming methodology.
If you don't understand any of the above or if it isn't clear ask in a comment and i'll try to make the answer better.

Here are your examples reexplained:
public interface IFood // not Pizza
{
public void Prepare();
}
public class Pizza : IFood
{
public void Prepare() // Not order for explanations sake
{
//Prepare Pizza
}
}
public class Burger : IFood
{
public void Prepare()
{
//Prepare Burger
}
}

In the absence of duck typing as you can use it in Python, C# relies on interfaces to provide abstractions. If the dependencies of a class were all concrete types, you could not pass in any other type - using interfaces you can pass in any type that implements the interface.

The Pizza example is bad because you should be using an abstract class that handles the ordering, and the pizzas should just override the pizza type, for example.
You use interfaces when you have a shared property, but your classes inherit from different places, or when you don't have any common code you could use. For instance, this is used things that can be disposed IDisposable, you know it will be disposed, you just don't know what will happen when it's disposed.
An interface is just a contract that tells you some things an object can do, what parameters and what return types to expect.

Consider the case where you don't control or own the base classes.
Take visual controls for instance, in .NET for Winforms they all inherit from the base class Control, that is completely defined in the .NET framework.
Let's assume you're in the business of creating custom controls. You want to build new buttons, textboxes, listviews, grids, whatnot and you'd like them all to have certain features unique to your set of controls.
For instance you might want a common way to handle theming, or a common way to handle localization.
In this case you can't "just create a base class" because if you do that, you have to reimplement everything that relates to controls.
Instead you will descend from Button, TextBox, ListView, GridView, etc. and add your code.
But this poses a problem, how can you now identify which controls are "yours", how can you build some code that says "for all the controls on the form that are mine, set the theme to X".
Enter interfaces.
Interfaces are a way to look at an object, to determine that the object adheres to a certain contract.
You would create "YourButton", descend from Button, and add support for all the interfaces you need.
This would allow you to write code like the following:
foreach (Control ctrl in Controls)
{
if (ctrl is IMyThemableControl)
((IMyThemableControl)ctrl).SetTheme(newTheme);
}
This would not be possible without interfaces, instead you would have to write code like this:
foreach (Control ctrl in Controls)
{
if (ctrl is MyThemableButton)
((MyThemableButton)ctrl).SetTheme(newTheme);
else if (ctrl is MyThemableTextBox)
((MyThemableTextBox)ctrl).SetTheme(newTheme);
else if (ctrl is MyThemableGridView)
((MyThemableGridView)ctrl).SetTheme(newTheme);
else ....
}

In this case, you could ( and probably would ) just define a Pizza base class and inherit from them. However, there are two reasons where Interfaces allow you to do things that cannot be achieved in other ways:
A class can implement multiple interfaces. It just defines features that the class must have. Implementing a range of interfaces means that a class can fulfil multiple functions in different places.
An interface can be defined in a hogher scope than the class or the caller. This means that you can separate the functionality, separate the project dependency, and keep the functionality in one project or class, and the implementation of this elsewhere.
One implication of 2 is that you can change the class that is being used, just requiring that it implements the appropriate interface.

Consider you can't use multiple inheritance in C#, and then look at your question again.

I did a search for the word "composition" on this page and didn't see it once. This answer is very much in addition to the answers aforementioned.
One of the absolutely crucial reasons for using interfaces in an Object Oriented Project is that they allow you to favour composition over inheritance. By implementing interfaces you can decouple your implementations from the various algorithms you are applying to them.
This superb "Decorator Pattern" tutorial by Derek Banas (which - funnily enough - also uses pizza as an example) is a worthwhile illustration:
https://www.youtube.com/watch?v=j40kRwSm4VE

Interface = contract, used for loose coupling (see GRASP).

If I am working on an API to draw shapes, I may want to use DirectX or graphic calls, or OpenGL. So, I will create an interface, which will abstract my implementation from what you call.
So you call a factory method: MyInterface i = MyGraphics.getInstance(). Then, you have a contract, so you know what functions you can expect in MyInterface. So, you can call i.drawRectangle or i.drawCube and know that if you swap one library out for another, that the functions are supported.
This becomes more important if you are using Dependency Injection, as then you can, in an XML file, swap implementations out.
So, you may have one crypto library that can be exported that is for general use, and another that is for sale only to American companies, and the difference is in that you change a config file, and the rest of the program isn't changed.
This is used a great deal with collections in .NET, as you should just use, for example, List variables, and don't worry whether it was an ArrayList or LinkedList.
As long as you code to the interface then the developer can change the actual implementation and the rest of the program is left unchanged.
This is also useful when unit testing, as you can mock out entire interfaces, so, I don't have to go to a database, but to a mocked out implementation that just returns static data, so I can test my method without worrying if the database is down for maintenance or not.

Interfaces are for applying connection between different classes. for example, you have a class for car and a tree;
public class Car { ... }
public class Tree { ... }
you want to add a burnable functionality for both classes. But each class have their own ways to burn. so you simply make;
public class Car : IBurnable
{
public void Burn() { ... }
}
public class Tree : IBurnable
{
public void Burn() { ... }
}

You will get interfaces, when you will need them :) You can study examples, but you need the Aha! effect to really get them.
Now that you know what interfaces are, just code without them. Sooner or later you will run into a problem, where the use of interfaces will be the most natural thing to do.

An interface is really a contract that the implementing classes must follow, it is in fact the base for pretty much every design pattern I know.
In your example, the interface is created because then anything that IS A Pizza, which means implements the Pizza interface, is guaranteed to have implemented
public void Order();
After your mentioned code you could have something like this:
public void orderMyPizza(IPizza myPizza) {
//This will always work, because everyone MUST implement order
myPizza.order();
}
This way you are using polymorphism and all you care is that your objects respond to order().

I'm surprised that not many posts contain the one most important reason for an interface: Design Patterns. It's the bigger picture into using contracts, and although it's a syntax decoration to machine code (to be honest, the compiler probably just ignores them), abstraction and interfaces are pivotal for OOP, human understanding, and complex system architectures.
Let's expand the pizza analogy to say a full fledge 3 course meal. We'll still have the core Prepare() interface for all our food categories, but we'd also have abstract declarations for course selections (starter, main, dessert), and differing properties for food types (savoury/sweet, vegetarian/non-vegetarian, gluten free etc).
Based on these specifications, we could implement the Abstract Factory pattern to conceptualise the whole process, but use interfaces to ensure that only the foundations were concrete. Everything else could become flexible or encourage polymorphism, yet maintain encapsulation between the different classes of Course that implement the ICourse interface.
If I had more time, I'd like to draw up a full example of this, or someone can extend this for me, but in summary, a C# interface would be the best tool in designing this type of system.

Here's an interface for objects that have a rectangular shape:
interface IRectangular
{
Int32 Width();
Int32 Height();
}
All it demands is that you implement ways to access the width and height of the object.
Now let's define a method that will work on any object that is IRectangular:
static class Utils
{
public static Int32 Area(IRectangular rect)
{
return rect.Width() * rect.Height();
}
}
That will return the area of any rectangular object.
Let's implement a class SwimmingPool that is rectangular:
class SwimmingPool : IRectangular
{
int width;
int height;
public SwimmingPool(int w, int h)
{ width = w; height = h; }
public int Width() { return width; }
public int Height() { return height; }
}
And another class House that is also rectangular:
class House : IRectangular
{
int width;
int height;
public House(int w, int h)
{ width = w; height = h; }
public int Width() { return width; }
public int Height() { return height; }
}
Given that, you can call the Area method on houses or swimming-pools:
var house = new House(2, 3);
var pool = new SwimmingPool(3, 4);
Console.WriteLine(Utils.Area(house));
Console.WriteLine(Utils.Area(pool));
In this way, your classes can "inherit" behavior (static-methods) from any number of interfaces.

What ?
Interfaces are basically a contract that all the classes implementing the Interface should follow. They looks like a class but has no implementation.
In C# Interface names by convention is defined by Prefixing an 'I' so if you want to have an interface called shapes, you would declare it as IShapes
Now Why ?
Improves code re-usability
Lets say you want to draw Circle, Triangle.
You can group them together and call them Shapesand have methods to draw Circle and Triangle
But having concrete implementation would be a bad idea because tomorrow you might decide to have 2 more Shapes Rectangle & Square. Now when you add them there is a great chance that you might break other parts of your code.
With Interface you isolate the different implementation from the Contract
Live Scenario Day 1
You were asked to create an App to Draw Circle and Triangle
interface IShapes
{
void DrawShape();
}
class Circle : IShapes
{
public void DrawShape()
{
Console.WriteLine("Implementation to Draw a Circle");
}
}
Class Triangle: IShapes
{
public void DrawShape()
{
Console.WriteLine("Implementation to draw a Triangle");
}
}
static void Main()
{
List <IShapes> shapes = new List<IShapes>();
shapes.Add(new Circle());
shapes.Add(new Triangle());
foreach(var shape in shapes)
{
shape.DrawShape();
}
}
Live Scenario Day 2
If you were asked add Square and Rectangle to it, all you have to do is create the implentation for it in class Square: IShapes and in Main add to list shapes.Add(new Square());

An interface defines a contract between the provider of a certain functionality and the correspondig consumers. It decouples the implementation from the contract (interface). You should have a look at object oriented architecture and design. You may want to start with wikipedia: http://en.wikipedia.org/wiki/Interface_(computing)

There are a lot of good answers here but I would like to try from a slightlt different perspective.
You may be familiar with the SOLID principles of object oriented design. In summary:
S - Single Responsibility Principle
O - Open/Closed Principle
L - Liskov Substitution Principle
I - Interface Segregation Principle
D - Dependency Inversion Principle
Following the SOLID principles helps to produce code that is clean, well factored, cohesive and loosely coupled. Given that:
"Dependency management is the key challenge in software at every scale" (Donald Knuth)
then anything that helps with dependency management is a big win. Interfaces and the Dependency Inversion Principle really help to decouple code from dependencies on concrete classes, so code can be written and reasoned about in terms of behaviours rather than implementations. This helps to break the code into components which can be composed at runtime rather than compile time and also means those components can be quite easily plugged in and out without having to alter the rest of the code.
Interfaces help in particular with the Dependency Inversion Principle, where code can be componentized into a collection of services, with each service being described by an interface. Services can then be "injected" into classes at runtime by passing them in as a constructor parameter. This technique really becomes critical if you start to write unit tests and use test driven development. Try it! You will quickly understand how interfaces help to break apart the code into manageable chunks that can be individually tested in isolation.

Soo many answers!
Giving my best shot. hehe.
So to begin, yes you could have used a concrete base and derived class here. In that case, you would have to do an empty or useless implementation for the Prepare method in the base class also making this method virtual and then the derived classes would override this Prepare method for themselves. This case, the implementation of Prepare in Base class is useless.
The reason why you chose to use an Interface is because you had to define a contract, not an implementation.
There is a IPizza type and it provides a functionality to Prepare. This is contract. How it is prepared is the implementation and it is not your lookout. It is responsibility of the various Pizza implementations.
An interface or an abstract class is preferred here over a concrete base class because you had to create an abstraction, i.e., the Prepare method. You cannot create an abstract method in a concrete base class.
Now you could say, why not use an abstract class?
So, when you need to achieve 100% abstraction, you need to go with Interface. But when you need some abstraction along with a concrete implementation, go with abstract class. It means.
Example: Lets say all your pizzas will have a base and base preparation will be the same process. However, all pizza types and toppings will vary. In this case you could create an Abstract class with an abstract method Prepare and a concrete method PreparePizzaBase.
public abstract class Pizza{
// concrete method which is common to all pizzas.
public PizzaBase PreparePizzaBase(){
// code for pizza base preparation.
}
public abstract void Prepare();
}
public class DeluxePizza: Pizza{
public void Prepare(){
var base=PreparePizzaBase();
// prepare deluxe pizza on pizza base.
}
}

The main purpose of the interfaces is that it makes a contract between you and any other class that implement that interface which makes your code decoupled and allows expandability.

Therese are ask really great examples.
Another, in the case of a switch statement, you no longer have the need to maintain and switch every time you want rio perform a task in a specific way.
In your pizza example, if want to make a pizza, the interface is all you need, from there each pizza takes care of it's own logic.
This helps to reduce coupling and cyclomatic complexity. You have to still implement the logic but there will be less you have to keep track of in the broader picture.
For each pizza you can then keep track of information specific to that pizza. What other pizzas have doesn't matter because only the other pizzas need to know.

The simplest way to think about interfaces is to recognize what inheritance means. If class CC inherits class C, it means both that:
Class CC can use any public or protected members of class C as though they were its own, and thus only needs to implement things which do not exist in the parent class.
A reference to a CC can be passed or assigned to a routine or variable that expects a reference to a C.
Those two function of inheritance are in some sense independent; although inheritance applies both simultaneously, it is also possible to apply the second without the first. This is useful because allowing an object to inherit members from two or more unrelated classes is much more complicated than allowing one type of thing to be substitutable for multiple types.
An interface is somewhat like an abstract base class, but with a key difference: an object which inherits a base class cannot inherit any other class. By contrast, an object may implement an interface without affecting its ability to inherit any desired class or implement any other interfaces.
One nice feature of this (underutilized in the .net framework, IMHO) is that they make it possible to indicate declaratively the things an object can do. Some objects, for example, will want data-source object from which they can retrieve things by index (as is possible with a List), but they won't need to store anything there. Other routines will need a data-depository object where they can store things not by index (as with Collection.Add), but they won't need to read anything back. Some data types will allow access by index, but won't allow writing; others will allow writing, but won't allow access by index. Some, of course, will allow both.
If ReadableByIndex and Appendable were unrelated base classes, it would be impossible to define a type which could be passed both to things expecting a ReadableByIndex and things expecting an Appendable. One could try to mitigate this by having ReadableByIndex or Appendable derive from the other; the derived class would have to make available public members for both purposes, but warn that some public members might not actually work. Some of Microsoft's classes and interfaces do that, but that's rather icky. A cleaner approach is to have interfaces for the different purposes, and then have objects implement interfaces for the things they can actually do. If one had an interface IReadableByIndex and another interface IAppendable, classes which could do one or the other could implement the appropriate interfaces for the things they can do.

Interfaces can also be daisy chained to create yet another interface. This ability to implement multiple Interfaces give the developer the advantage of adding functionality to their classes without having to change current class functionality (SOLID Principles)
O = "Classes should be open for extension but closed for modification"

To me an advantage/benefit of an interface is that it is more flexible than an abstract class. Since you can only inherit 1 abstract class but you can implement multiple interfaces, changes to a system that inherits an abstract class in many places becomes problematic. If it is inherited in 100 places, a change requires changes to all 100. But, with the interface, you can place the new change in a new interface and just use that interface where its needed (Interface Seq. from SOLID). Additionally, the memory usage seems like it would be less with the interface as an object in the interface example is used just once in memory despite how many places implement the interface.

Interfaces are used to drive consistency,in a manner that is loosely coupled which makes it different to abstract class which is tightly coupled.That's why its also commonly defined as a contract.Whichever classes that implements the interface has abide to "rules/syntax" defined by the interface and there is no concrete elements within it.
I'll just give an example supported by the graphic below.
Imagine in a factory there are 3 types of machines.A rectangle machine,a triangle machine and a polygon machine.Times are competitive and you want to streamline operator training.You just want to train them in one methodology of starting and stopping machines so you have a green start button and red stop button.So now across 3 different machines you have a consistent way of starting and stopping 3 different types of machines.Now imagine these machines are classes and the classes need to have start and stop methods,how you going to drive consistency across these classes which can be very different? Interface is the answer.
A simple example to help you visualize,one might ask why not use abstract class? With an interface the objects don't have to be directly related or inherited and you can still drive consistency across different classes.
public interface IMachine
{
bool Start();
bool Stop();
}
public class Car : IMachine
{
public bool Start()
{
Console.WriteLine("Car started");
return true;
}
public bool Stop()
{
Console.WriteLine("Car stopped");
return false;
}
}
public class Tank : IMachine
{
public bool Start()
{
Console.WriteLine("Tank started");
return true;
}
public bool Stop()
{
Console.WriteLine("Tank stopped");
return false;
}
}
class Program
{
static void Main(string[] args)
{
var car = new Car();
car.Start();
car.Stop();
var tank = new Tank();
tank.Start();
tank.Stop();
}
}

class Program {
static void Main(string[] args) {
IMachine machine = new Machine();
machine.Run();
Console.ReadKey();
}
}
class Machine : IMachine {
private void Run() {
Console.WriteLine("Running...");
}
void IMachine.Run() => Run();
}
interface IMachine
{
void Run();
}
Let me describe this by a different perspective. Let’s create a story according to the example which i have shown above;
Program, Machine and IMachine are the actors of our story. Program wants to run but it has not that ability and Machine knows how to run. Machine and IMachine are best friends but Program is not on speaking terms with Machine. So Program and IMachine make a deal and decided that IMachine will tell to Program how to run by looking Machine(like a reflector).
And Program learns how to run by help of IMachine.
Interface provides communication and developing loosely coupled projects.
PS: I’ve the method of concrete class as private. My aim in here is to achieve loosely coupled by preventing accessing concrete class properties and methods, and left only allowing way to reach them via interfaces. (So i defined interfaces’ methods explicitily).

c# interface segregation principle example confusion

I'm fairly new to programming and i'm having trouble to understand how to apply effectively the principle showed in the following link (the ATM one):
http://www.objectmentor.com/resources/articles/isp.pdf
Basically it starts with a design that does not complain the ISP (Interface Segregation Principle), and moves forward to refactor the behavior into different interfaces.
My question is: Don't we use interfaces to express common behaviour among not so (or not) related abstractions?
What's the point of encapsulating methods in an interface, if not even one is going to be shared with the classes that are going to implement them? In which scenario this could be consider useful?
If we continue the line of the example, the following code is given:
public interface ITransaction
{
void Execute();
}
public interface IDepositUi
{
void RequestDepositAmount();
}
public class DepositTransaction : ITransaction
{
private IDepositUi depositUI;
public DepositTransaction(IDepositUi ui)
{
depositUI = ui;
}
public virtual void Execute()
{
/*code*/
depositUI.RequestDepositAmount();
/*code*/
}
}
public interface WithdrawalUI
{
void RequestWithdrawalAmount();
}
public class WithdrawalTransaction : ITransaction
{
private WithdrawalUI withdrawalUI;
public WithdrawalTransaction(WithdrawalUI ui)
{
withdrawalUI = ui;
}
public virtual void Execute()
{
/*code*/
withdrawalUI.RequestWithdrawalAmount(); /*code*/
}
}
public interface TransferUI
{
void RequestTransferAmount();
}
public class TransferTransaction : ITransaction
{
private TransferUI transferUI;
public TransferTransaction(TransferUI ui)
{
transferUI = ui;
}
public virtual void Execute()
{
/*code*/
transferUI.RequestTransferAmount();
/*code*/
}
}
public interface UI : IDepositUi, WithdrawalUI, TransferUI
{
}
As far as i understand this, in order to use the previous design we should have something like:
UI impui = new IMPLEMENTATIONUI(); // Some UI implementation
DepositTransaction dt = new DepositTransaction(Gui);
dt.Execute();
Now, wouldn't we need that the IMPLEMENTATIONUI implements every single method? And if so, wouldn't it break the SRP?.

Don't we use interfaces to express common behaviour among not so (or not) related abstractions?
Yes, in SOLID, interfaces are required to express common behavior. Your Transaction interface is a superb example of this. Both the DepositTransaction and WithdrawlTransaction classes depend on it. ISP (Interface Segregation Principle) wants you to split this out because you may have a need to pass a Transaction object into a function to execute it. All of the SOLID principles are designFor example:
void ExecuteTransaction(Transaction transaction)
{
transaction.Execute();
}
Please note that this method does not depend on anything but the Transaction interface. This is dependency inversion.
If you do not create this interface you would need to create two different methods for executing a WithdrawlTransaction or a DepositTransaction; instead, you can use the ExecuteTransaction method and pass in anything that implements Transaction.
ExecuteTransation(withdrawl_TransactionObject);
or
ExecuteTransaction(deposit_TransactionObject);
or later in the future:
ExecuteTransaction(unanticipatedNewTypeOf_TransactionObject);
Now, wouldn't we need that the IMPLEMENTATIONUI implements every single method? And if so, wouldn't it break the SRP?
The Implementation UI is probably what the user uses to interact with the software. The user won't have a Single Responsibility, theoretically he/she will have to use all the interfaces that are required for the IMPLEMENTATIONUI class.
I doubt the implementation UI would implement all of the interfaces but it would likely use all of these interfaces in order to execute transactions. To paraphrase "Uncle Bob" Solid Principles your user interface should be full of Volatile code while your interfaces should be the most non-volatile.

Is having one interface inherit from three others an ISP violation?
public interface UI : IDepositUi, WithdrawalUI, TransferUI
{
}
The answer is that we can't tell. It depends on whether or not the client depends on the entire interface. If it does then this is not an ISP violation. If the client doesn't depend on all three inherited interfaces then this is a violation and the client should just depend on whichever interface(s) it does need.
It might, as you observed, violate some other principle, but that would be outside of the scope of discussing the ISP. But I don't think the point is that you should create that combined interface. The point is that you can while still preserving the smaller, segregated interfaces.
We might be tempted to create on giant interface and one giant class just because one client depends on all three interfaces. But if we do that, another client that only needs withdrawals or transfers would be forced to depend on the larger interface that it doesn't need.
In real life this sort of thing grows out of control because someone starts off with broad, vaguely named interface like ITransactionService and before you know it more developers throw the kitchen sink into it. In the example the interfaces are more specifically named. That won't enforce keeping them segregated, but it helps. Giving them such specific names up front communicates what should or shouldn't be in them. It suggests a developer who planned up front to keep interfaces segregated.

Is using var + basic dependency factory more loosely coupled than C# interfaces?

This is a general design question. We often use interfaces to decouple components, write to an interface not an implementation etc. Sometimes interfaces are used w/ a basic injection technique, such as,
interface IMyInterface
{
void DoSomething();
}
static class IMyInterfaceFactory
{
public static IMyInterface GetInstance()
{
return new MyInterfaceInstance();
}
}
class IMyInterfaceConsumer
{
IMyInterface mInterface;
public IMyInterfaceConsumer()
{
this.mInterface = IMyInterfaceFactory.GetInstance();
}
public void UseTheInterface()
{
this.mInterface.DoSomething();
}
}
My question is about using the var keyword instead. Not even using a true C# interface, but still creating an 'interface', in the design sense,
static class IMyInterfaceFactory
{
// of course, this doesnt need to be a single instance
static MyInterfaceInstance mSingleInstance;
// no longer programming to the interface, just returning the instance
public static MyInterfaceInstance GetInstance()
{
// null coalesce
return mSingleInstance ?? (mSingleInstance = new MyInterfaceInstance());
}
}
class IMyInterfaceConsumer
{
public void UseTheInterface()
{
// shorthand way, could also omit var, too
var myInterface = IMyInterfaceFactory.GetInstance();
myInterface.DoSomething();
}
}
This way I still only need to change the factory once, and as long as whatever instance it returns supports the methods that need to be consumed, it will work. The advantage however is that the producing and consuming objects dont need to even know about any explicit interface, none exists. It could also cleanly support an interface with more than just a couple methods (prevent bloated interface declarations).
One obvious downside is that everytime you want to consume a method from the 'interface', the factory will potentially have to re-instantiate the class, unless there is a single instance cached (as above) or some memoization technique used.
Pros/cons of this approach? Is this a common practice?

There is nothing dynamic or loose about the var keyword. It triggers static type inference at compile time.
Your second piece of code behaves identically to
public void UseTheInterface()
{
// shorthand way, could also omit var, too
MyInterfaceInstance myInterface = IMyInterfaceFactory.GetInstance();
myInterface.DoSomething();
}
The factory function is still strongly typed. In fact, by removing the interface, you've made consumer code much more tightly coupled.

Var keyword is still technically strongly typed, so your code does know what class/interface it is. If you planned on dumping it into an object then we are saying that the rest of your code has no clue what is coming out of that factory. I wouldn't suggest that though since that causes you to cast that object to utilize anything in it.
I'm not sure where you are trying to go with the "prevent bloated interface declarations" but you could do polymorphism through extending a base class or an abstract class as well. That would make it so any code that is common between the child classes could be left alone and any specific code (methods or properties) for each child class could be overridden.
If you are looking to change out the interface all together you will need to implement an interface in the interface, see this post. So you will basically have interface A only have the method DoStuff() and other interfaces that inherit from this interface could be used polymorphically like you are describing.
interface A
{
DoStuff();
}
interface B : A
{
DoSomethingElse();
}
class C : B
{
DoStuff(){}
DoSomethingElse(){}
}
By the way, your "single instance cached" code above is close to something called a singleton pattern.

Is it good practice to blank out inherited functionality that will not be used?

I'm wondering if I should change the software architecture of one of my projects.
I'm developing software for a project where two sides (in fact a host and a device) use shared code. That helps because shared data, e.g. enums can be stored in one central place.
I'm working with what we call a "channel" to transfer data between device and host. Each channel has to be implemented on device and host side. We have different kinds of channels, ordinary ones and special channels which transfer measurement data.
My current solution has the shared code in an abstract base class. From there on code is split between the two sides. As it has turned out there are a few cases when we would have shared code but we can't share it, we have to implement it on each side.
The principle of DRY (don't repeat yourself) says that you shouldn't have code twice.
My thought was now to concatenate the functionality of e.g. the abstract measurement channel on the device side and the host side in an abstract class with shared code. That means though that once we create an actual class for either the device or the host side for that channel we have to hide the functionality that is used by the other side.
Is this an acceptable thing to do:
public abstract class ChannelAbstract
{
protected void ChannelAbstractMethodUsedByDeviceSide() { }
protected void ChannelAbstractMethodUsedByHostSide() { }
}
public abstract class MeasurementChannelAbstract : ChannelAbstract
{
protected void MeasurementChannelAbstractMethodUsedByDeviceSide() { }
protected void MeasurementChannelAbstractMethodUsedByHostSide() { }
}
public class DeviceMeasurementChannel : MeasurementChannelAbstract
{
public new void MeasurementChannelAbstractMethodUsedByDeviceSide()
{
base.MeasurementChannelAbstractMethodUsedByDeviceSide();
}
public new void ChannelAbstractMethodUsedByDeviceSide()
{
base.ChannelAbstractMethodUsedByDeviceSide();
}
}
public class HostMeasurementChannel : MeasurementChannelAbstract
{
public new void MeasurementChannelAbstractMethodUsedByHostSide()
{
base.MeasurementChannelAbstractMethodUsedByHostSide();
}
public new void ChannelAbstractMethodUsedByHostSide()
{
base.ChannelAbstractMethodUsedByHostSide();
}
}
Now, DeviceMeasurementChannel is only using the functionality for the device side from MeasurementChannelAbstract. By declaring all methods/members of MeasurementChannelAbstract protected you have to use the new keyword to enable that functionality to be accessed from the outside.
Is that acceptable or are there any pitfalls, caveats, etc. that could arise later when using the code?

You can solve the problem with inheritance, like this:
public abstract class MeasurementChannelAbstract
{
protected abstract void Method();
}
public class DeviceMeasurementChannel : MeasurementChannelAbstract
{
public void Method()
{
// Device side implementation here.
}
}
public class HostMeasurementChannel : MeasurementChannelAbstract
{
public void Method()
{
// Host side implementation here.
}
}
... or by composition, using the Strategy pattern, like this:
public class MeasurementChannel
{
private MeasurementStrategyAbstract m_strategy;
public MeasurementChannel(MeasurementStrategyAbstract strategy)
{
m_strategy = strategy;
}
protected void Method()
{
m_strategy.Measure();
}
}
public abstract class MeasurementStrategyAbstract
{
protected abstract void Measure();
}
public class DeviceMeasurementStrategy : MeasurementStrategyAbstract
{
public void Measure()
{
// Device side implementation here.
}
}
public class HostMeasurementStrategy : MeasurementStrategyAbstract
{
public void Measure()
{
// Host side implementation here.
}
}
It seems to me that you want to divide your inheritance hierarchy between both Standard/Measurement channels and Device/Host channels. One way to do this is with multiple inheritance - but C# doesn't support multiple inheritance (except for interfaces), and in most cases a design based on composition will be simpler.

To me it looks a bit like you're confusing inheritance and composition. When you have to "blank out"/throw exception when inherited functionality does not overlap sanely, your inheritance graph is missing some intermediate class. And often this is because some functionality just should come from a member instance of an other class instead of being inherited.
Consider also practicality, mapping everything into perfect OOP is not the goal, the goal is a working program that is maintainable without huge pain.

Like Rich implies: Needing only ONE of the members declared ins MeasurementChannelAbstract in one of the concrete implementations is a very clear indication that your interface is badly defined as it has more than one responsibility. This means it is difficult for clients of your code (and readers) to understand the abstractions and see the difference between the various concrete implementations.
This is called the "Single Responsibility Principle" and it is is an important one for good OO design.
(For more on good OO design, I recommend working through all the SOLID principles).

It sounds to me as though you've not finished identifying which bits of code are shared. Wouldn't it make more sense to keep all the generic / shared stuff in MeasurementChannelAbstract rather than having different methods which the two sides call? I would have thought that those ought to be in inherited classes?

Therefore I ask myself if it is OK to
have a single chain of inheritance
(accepting that some classes will have
functionality for both host and device
side) and do the separation into
host/device channel at the last stage
(blanking out the functionality which
is not needed)
I think it's ok. But you may blanking out the functionality which is not needed by using different interfaces for SimpleChannel and MeasurementChannel.

NUnit - How to test all classes that implement a particular interface

If I have interface IFoo, and have several classes that implement it, what is the best/most elegant/cleverest way to test all those classes against the interface?
I'd like to reduce test code duplication, but still 'stay true' to the principles of Unit testing.
What would you consider best practice? I'm using NUnit, but I suppose examples from any Unit testing framework would be valid

If you have classes implement any one interface then they all need to implement the methods in that interface. In order to test these classes you need to create a unit test class for each of the classes.
Lets go with a smarter route instead; if your goal is to avoid code and test code duplication you might want to create an abstract class instead that handles the recurring code.
E.g. you have the following interface:
public interface IFoo {
public void CommonCode();
public void SpecificCode();
}
You might want to create an abstract class:
public abstract class AbstractFoo : IFoo {
public void CommonCode() {
SpecificCode();
}
public abstract void SpecificCode();
}
Testing that is easy; implement the abstract class in the test class either as an inner class:
[TestFixture]
public void TestClass {
private class TestFoo : AbstractFoo {
boolean hasCalledSpecificCode = false;
public void SpecificCode() {
hasCalledSpecificCode = true;
}
}
[Test]
public void testCommonCallsSpecificCode() {
TestFoo fooFighter = new TestFoo();
fooFighter.CommonCode();
Assert.That(fooFighter.hasCalledSpecificCode, Is.True());
}
}
...or let the test class extend the abstract class itself if that fits your fancy.
[TestFixture]
public void TestClass : AbstractFoo {
boolean hasCalledSpecificCode;
public void specificCode() {
hasCalledSpecificCode = true;
}
[Test]
public void testCommonCallsSpecificCode() {
AbstractFoo fooFighter = this;
hasCalledSpecificCode = false;
fooFighter.CommonCode();
Assert.That(fooFighter.hasCalledSpecificCode, Is.True());
}
}
Having an abstract class take care of common code that an interface implies gives a much cleaner code design.
I hope this makes sense to you.
As a side note, this is a common design pattern called the Template Method pattern. In the above example, the template method is the CommonCode method and SpecificCode is called a stub or a hook. The idea is that anyone can extend behavior without the need to know the behind the scenes stuff.
A lot of frameworks rely on this behavioral pattern, e.g. ASP.NET where you have to implement the hooks in a page or a user controls such as the generated Page_Load method which is called by the Load event, the template method calls the hooks behind the scenes. There are a lot more examples of this. Basically anything that you have to implement that is using the words "load", "init", or "render" is called by a template method.

I disagree with Jon Limjap when he says,
It is not a contract on either a.) how the method should be implemented and b.) what that method should be doing exactly (it only guarantees the return type), the two reasons that I glean would be your motive in wanting this kind of test.
There could be many parts of the contract not specified in the return type. A language-agnostic example:
public interface List {
// adds o and returns the list
public List add(Object o);
// removed the first occurrence of o and returns the list
public List remove(Object o);
}
Your unit tests on LinkedList, ArrayList, CircularlyLinkedList, and all the others should test not only that the lists themselves are returned, but also that they have been properly modified.
There was an earlier question on design-by-contract, which can help point you in the right direction on one way of DRYing up these tests.
If you don't want the overhead of contracts, I recommend test rigs, along the lines of what Spoike recommended:
abstract class BaseListTest {
abstract public List newListInstance();
public void testAddToList() {
// do some adding tests
}
public void testRemoveFromList() {
// do some removing tests
}
}
class ArrayListTest < BaseListTest {
List newListInstance() { new ArrayList(); }
public void arrayListSpecificTest1() {
// test something about ArrayLists beyond the List requirements
}
}

I don't think this is best practice.
The simple truth is that an interface is nothing more than a contract that a method is implemented. It is not a contract on either a.) how the method should be implemented and b.) what that method should be doing exactly (it only guarantees the return type), the two reasons that I glean would be your motive in wanting this kind of test.
If you really want to be in control of your method implementation, you have the option of:
Implementing it as a method in an abstract class, and inherit from that. You will still need to inherit it into a concrete class, but you are sure that unless it is explicitly overriden that method will do that correct thing.
In .NET 3.5/C# 3.0, implementing the method as an extension method referencing to the Interface
Example:
public static ReturnType MethodName (this IMyinterface myImplementation, SomeObject someParameter)
{
//method body goes here
}
Any implementation properly referencing to that extension method will emit precisely that extension method so you only need to test it once.

How about a hierarchy of [TestFixture]s classes? Put the common test code in the base test class and inherit it into child test classes..

When testing an interface or base class contract, I prefer to let the test framework automatically take care of finding all of the implementers. This lets you concentrate on the interface under test and be reasonably sure that all implementations will be tested, without having to do a lot of manual implementation.
For xUnit.net, I created a Type Resolver library to search for all implementations of a particular type (the xUnit.net extensions are just a thin wrapper over the Type Resolver functionality, so it can be adapted for use in other frameworks).
In MbUnit, you can use a CombinatorialTest with UsingImplementations attributes on the parameters.
For other frameworks, the base class pattern Spoike mentioned can be useful.
Beyond testing the basics of the interface, you should also test that each individual implementation follows its particular requirements.

I don't use NUnit but I have tested C++ interfaces. I would first test a TestFoo class which is a basic implementation of it to make sure the generic stuff works. Then you just need to test the stuff that is unique to each interface.