Policy mode (strategy)

I. Introduction of the Strategy model

Policy mode: It defines the algorithm family, individually encapsulated, so that they can be replaced with each other, this mode allows the algorithm changes will not affect the use of the algorithm customers.

For example:

The cashier software in the mall, how to make different discounts for different products? Or, how to make different discount according to different membership level?

In fact, the internal definition of an algorithm family, to each member or commodity to specify a specific algorithm, according to different algorithms, to play a different discount.

Policy Mode UML diagram:

strategy (Abstract Algorithm Interface): Defines a common interface for all supported algorithms

Concretestrategya,concretestrategyb, and concretestrategyc: Inherited from strategy, encapsulating specific algorithms

context: Different algorithm implementations for maintaining different objects

Second, the Strategy mode code implementation

Here are the examples of different users buying merchandise discounts

First define an abstract algorithm interface:

12345

//抽象算法接口：定义了所有支持算法的公共接口publicinterface Strategy {    //算法方法:打印商品的价格（不同的实现打不同的折扣）    public double getPrice(doubleprice);}

And then define the specific individual algorithms

1234567891011121314151617181920212223242526272829303132

//具体的算法:普通用户，不打折public class GeneralUser implements Strategy {    @Override    public double getPrice(double price) {        System.out.println("普通用户，不打折");        return price;    }}//具体的算法:注册用户：打9折public class RegisterUser implements Strategy{    @Override    public double getPrice(double price) {        System.out.println("注册用户：打9折");        return price*0.9;    }}//具体的算法：普通会员：打8折public class RegisterVip implements Strategy{    @Override    public double getPrice(double price) {        System.out.println("普通会员：打8折");        return price*0.8;    }}//具体的算法：老会员：打5折public class OldVip implements Strategy{    @Override    public double getPrice(double price) {        System.out.println("老会员：打5折");        return price*0.5;    }}

The next step is to define a context that maintains different discounts for different users.

123456789101112

//上下文：用来维护不同对象的不同折扣public class Context {    private Strategy strategy;//持有算法族的引用    public Context(Strategy strategy) {        super();        this.strategy = strategy;    }    //打印价钱    public void printPrice(double price){        System.out.println("应付金额："+Math.round(strategy.getPrice(price)));    }}

Client Test Code:

1234567891011121314151617

public static void main(String[] args) {    double price = 998;//商品价格    Strategy generalUser = new GeneralUser();//普通用户    Strategy registerUser = new RegisterUser();//注册用户    Strategy registerVip = new RegisterVip();//普通会员    Strategy oldVip = new OldVip();//老会员        //根据不同的用户打不同的折扣    Context c1 = new Context(generalUser);    c1.printPrice(price);    Context c2 = new Context(registerUser);    c2.printPrice(price);    Context c3 = new Context(registerVip);    c3.printPrice(price);    Context c4 = new Context(oldVip);    c4.printPrice(price);}

The printing results are as follows:

Normal user, no discount

Amount Due: 998

Registered users: Call 90 percent

Amount Due: 898

Regular members: 80 percent off

Amount Due: 798

Old member: Call 50 percent

Amount Due: 499

Third, the application scenario

The essence of policy mode: separating algorithm, choosing different implementation.

Application Scenarios:

Layout management of Javase

Spring Framework, resource interface, resource access policy

Javax.servlet.http.httpservlet#service ();

From for notes (Wiz)

Policy mode (strategy)