关键是接口代表一个契约。任何实现类都必须拥有的一组公共方法。从技术上讲，接口只控制语法，即有什么方法，它们得到什么参数以及它们返回什么。通常它们也封装语义，尽管只是通过文档。

然后，您可以拥有一个接口的不同实现，并随意交换它们。在您的示例中，由于每个披萨实例都是IPizza，因此在处理未知披萨类型实例的任何地方都可以使用IPizza。任何类型继承自IPizza的实例都保证是可排序的，因为它有一个Order()方法。

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.)

还要注意，通常意义上的接口不一定是c#接口，它也可以是一个抽象类，甚至是一个普通类(如果所有子类都需要共享一些公共代码，这可以派上用场——然而，在大多数情况下，接口就足够了)。

I share your sense that Interfaces are not necessary. Here is a quote from Cwalina pg 80 Framework Design Guidelines "I often here people saying that interfaces specify contracts. I believe this a dangerous myth. Interfaces by themselves do not specify much. ..." He and co-author Abrams managed 3 releases of .Net for Microsoft. He goes on to say that the 'contract' is "expressed" in an implementation of the class. IMHO watching this for decades, there were many people warning Microsoft that taking the engineering paradigm to the max in OLE/COM might seem good but its usefulness is more directly to hardware. Especially in a big way in the 80s and 90s getting interoperating standards codified. In our TCP/IP Internet world there is little appreciation of the hardware and software gymnastics we would jump through to get solutions 'wired up' between and among mainframes, minicomputers, and microprocessors of which PCs were just a small minority. So coding to interfaces and their protocols made computing work. And interfaces ruled. But what does solving making X.25 work with your application have in common with posting recipes for the holidays? I have been coding C++ and C# for many years and I never created one once.

下面是一个矩形对象的接口:

interface IRectangular
{
    Int32 Width();
    Int32 Height();
}

它所要求的是实现访问对象宽度和高度的方法。

现在让我们定义一个方法，它可以作用于任何irectangle对象:

static class Utils
{
    public static Int32 Area(IRectangular rect)
    {
        return rect.Width() * rect.Height();
    }
}

这将返回任意矩形物体的面积。

让我们实现一个矩形的类SwimmingPool:

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; }
}

另一个类House也是矩形的:

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; }
}

鉴于此，你可以在房屋或游泳池上调用Area方法:

var house = new House(2, 3);

var pool = new SwimmingPool(3, 4);

Console.WriteLine(Utils.Area(house));
Console.WriteLine(Utils.Area(pool));

通过这种方式，您的类可以从任意数量的接口“继承”行为(静态方法)。

接口定义了某种功能的提供者和相应的使用者之间的契约。它将实现与契约(接口)解耦。您应该了解一下面向对象的体系结构和设计。你可能想从维基百科开始:http://en.wikipedia.org/wiki/Interface_(computing)

关键是接口代表一个契约。任何实现类都必须拥有的一组公共方法。从技术上讲，接口只控制语法，即有什么方法，它们得到什么参数以及它们返回什么。通常它们也封装语义，尽管只是通过文档。

然后，您可以拥有一个接口的不同实现，并随意交换它们。在您的示例中，由于每个披萨实例都是IPizza，因此在处理未知披萨类型实例的任何地方都可以使用IPizza。任何类型继承自IPizza的实例都保证是可排序的，因为它有一个Order()方法。

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.)

还要注意，通常意义上的接口不一定是c#接口，它也可以是一个抽象类，甚至是一个普通类(如果所有子类都需要共享一些公共代码，这可以派上用场——然而，在大多数情况下，接口就足够了)。

对我来说，刚开始的时候，只有当你不再把它们看作是让你的代码更容易/更快编写的东西时，它们的意义才变得清晰——这不是它们的目的。它们有很多用途:

(这里就没有披萨的比喻了，因为这个比喻的用法不太容易想象)

假设你正在屏幕上制作一款简单的游戏，游戏中会有与你互动的生物。

答:通过在前端和后端实现之间引入松散耦合，它们可以使您的代码在将来更容易维护。

你可以这样写，因为这里只会有喷子:

// This is our back-end implementation of a troll
class Troll
{
    void Walk(int distance)
    {
        //Implementation here
    }
}

前端:

function SpawnCreature()
{
    Troll aTroll = new Troll();
    
    aTroll.Walk(1);
}

两周后，市场营销决定你也需要半兽人，因为他们在twitter上看到了他们，所以你必须做如下事情:

class Orc
{
    void Walk(int distance)
    {
        //Implementation (orcs are faster than trolls)
    }
}

前端:

void SpawnCreature(creatureType)
{
    switch(creatureType)
    {
         case Orc:

           Orc anOrc = new Orc();
           anORc.Walk();

          case Troll:

            Troll aTroll = new Troll();
             aTroll.Walk();
    }
}

你可以看到这是如何变得混乱的。你可以在这里使用一个接口，这样你的前端就会被编写一次(这里是重要的部分)测试，然后你可以根据需要插入更多的后端项目:

interface ICreature
{
    void Walk(int distance)
}

public class Troll : ICreature
public class Orc : ICreature 

//etc

前端则为:

void SpawnCreature(creatureType)
{
    ICreature creature;

    switch(creatureType)
    {
         case Orc:

           creature = new Orc();

          case Troll:

            creature = new Troll();
    }

    creature.Walk();
}

前端现在只关心接口ICreature -它不关心喷子或兽人的内部实现，而只关心他们实现ICreature的事实。

从这个角度来看，需要注意的一点是，您也可以很容易地使用抽象生物类，从这个角度来看，这具有相同的效果。

你可以将创建的内容提取到工厂:

public class CreatureFactory {

 public ICreature GetCreature(creatureType)
 {
    ICreature creature;

    switch(creatureType)
    {
         case Orc:

           creature = new Orc();

          case Troll:

            creature = new Troll();
    }

    return creature;
  }
}

我们的前端会变成:

CreatureFactory _factory;

void SpawnCreature(creatureType)
{
    ICreature creature = _factory.GetCreature(creatureType);

    creature.Walk();
}

现在，前端甚至不需要有实现Troll和Orc的库的引用(假设工厂在一个单独的库中)——它不需要知道任何关于它们的信息。

B:假设你拥有在你的同质数据结构中只有某些生物才有的功能，例如:

interface ICanTurnToStone
{
   void TurnToStone();
}

public class Troll: ICreature, ICanTurnToStone

前端可以是:

void SpawnCreatureInSunlight(creatureType)
{
    ICreature creature = _factory.GetCreature(creatureType);

    creature.Walk();

    if (creature is ICanTurnToStone)
    {
       (ICanTurnToStone)creature.TurnToStone();
    }
}

C:依赖注入用法

大多数依赖注入框架在前端代码和后端实现之间存在非常松散的耦合时才能工作。如果我们以上面的工厂为例，让我们的工厂实现一个接口:

public interface ICreatureFactory {
     ICreature GetCreature(string creatureType);
}

我们的前端可以通过构造函数注入(例如MVC API控制器)(通常):

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);
   }
}

使用我们的DI框架(例如Ninject或Autofac)，我们可以设置它们，以便在运行时在构造函数中需要ICreatureFactory时创建一个CreatureFactory实例——这使我们的代码美观而简单。

这也意味着当我们为控制器编写单元测试时，我们可以提供一个模拟的ICreatureFactory(例如，如果具体实现需要访问DB，我们不希望我们的单元测试依赖于它)，并轻松地测试控制器中的代码。

D:还有其他用途，例如，你有两个项目A和B，由于“遗留”原因，它们没有很好地组织起来，而A有B的参考。

然后在B中发现需要调用a中已经存在的方法的功能。由于您获得的是循环引用，因此不能使用具体实现来实现。

你可以在B中声明一个接口，然后由A中的类实现。你在B中的方法可以被传递一个实现接口的类的实例，即使具体对象是a中的类型。