7、scala面向对象-继承

一、继承

1、extends

Scala中，让子类继承父类，与Java一样，也是使用extends关键字 
继承就代表，子类可以从父类继承父类的field和method；然后子类可以在自己内部放入父类所没有，
子类特有的field和method；使用继承可以有效复用代码 
子类可以覆盖父类的field和method；但是如果父类用final修饰，field和method用final修饰，则该类是无法被继承的，
field和method是无法被覆盖的；


###
scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person {
  private var name = "leo"
  def getName = name
}

class Student extends Person {
 private var score  = "A"
 def getScore = score
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student
warning: previously defined object Person is not a companion to class Person.
Companions must be defined together; you may wish to use :paste mode for this.

scala> val s = new Student
s: Student = Student@985696

scala> s.getName
res11: String = leo

scala> s.getScore
res12: String = A

scala> s.getName
res13: String = leo

scala> s.name
<console>:14: error: value name is not a member of Student
       s.name
         ^

2、override和super

//Scala中，如果子类要覆盖一个父类中的非抽象方法，则必须使用override关键字;

//override关键字可以帮助我们尽早地发现代码里的错误，比如：override修饰的父类方法的方法名我们拼写错了；比如要覆盖的父类方法的参数我们写错了；等等

//此外，在子类覆盖父类方法之后，如果我们在子类中就是要调用父类的被覆盖的方法呢？
   那就可以使用super关键字，显式地指定要调用父类的方法


###
scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person {
  private var name = "leo"
  def getName = name
}

class Student extends Person {
  private var score = "A"
  def getScore = score
  override def getName = "Hi, I'm a student, my name is " + super.getName
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student
warning: previously defined object Person is not a companion to class Person.
Companions must be defined together; you may wish to use :paste mode for this.

scala> val s = new Student
s: Student = Student@618ad2aa

scala> s.getName
res15: String = Hi, I'm a student, my name is leo

3、override field

// Scala中，子类可以覆盖父类的val field，而且子类的val field还可以覆盖父类的val field的getter方法；只要在子类中使用override关键字即可

###
scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person {
  val name: String = "Person"
  def age: Int = 0
}

class Student extends Person { 
  override val name: String  = "leo"
  override val age: Int = 30
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student
warning: previously defined object Person is not a companion to class Person.
Companions must be defined together; you may wish to use :paste mode for this.

scala> val p = new Person
p: Person = Person@71d9cb05

scala> p.name
res16: String = Person

scala> p.age
res17: Int = 0

scala> val s = new Student
s: Student = Student@1ac4ccad

scala> s.name
res18: String = leo

scala> s.age
res19: Int = 30

4、isInstanceOf和asInstanceOf

// 如果我们创建了子类的对象，但是又将其赋予了父类类型的变量。则在后续的程序中，我们又需要将父类类型的变量转换为子类类型的变量，应该如何做？
// 首先，需要使用isInstanceOf判断对象是否是指定类的对象，如果是的话，则可以使用asInstanceOf将对象转换为指定类型
// 注意，如果对象是null，则isInstanceOf一定返回false，asInstanceOf一定返回null
// 注意，如果没有用isInstanceOf先判断对象是否为指定类的实例，就直接用asInstanceOf转换，则可能会抛出异常



###
scala> class Person
defined class Person
warning: previously defined object Person is not a companion to class Person.
Companions must be defined together; you may wish to use :paste mode for this.

scala> class Student extends Person
defined class Student

scala> val p: Person = new Student
p: Person = Student@f324455

scala> var s: Student = null
s: Student = null

scala> if (p.isInstanceOf[Student]) s = p.asInstanceOf[Student]

5、getClass和classOf

// isInstanceOf只能判断出对象是否是指定类以及其子类的对象，而不能精确判断出，对象就是指定类的对象
// 如果要求精确地判断对象就是指定类的对象，那么就只能使用getClass和classOf了
// 对象.getClass可以精确获取对象的类，classOf[类]可以精确获取类，然后使用==操作符即可判断


###
scala> class Person
defined class Person

scala> class Student extends Person
defined class Student

scala> val p: Person = new Student
p: Person = Student@77468bd9

scala> p.isInstanceOf[Person]
res0: Boolean = true

scala> p.getClass == classOf[Person]
res1: Boolean = false

scala> p.getClass == classOf[Student]
res2: Boolean = true

6、使用模式匹配进行类型判断

// 但是在实际开发中，比如spark的源码中，大量的地方都是使用了模式匹配的方式来进行类型的判断，这种方式更加地简洁明了，
而且代码得可维护性和可扩展性也非常的高
// 使用模式匹配，功能性上来说，与isInstanceOf一样，也是判断主要是该类以及该类的子类的对象即可，不是精准判断的



###
scala> class Person
defined class Person

scala> class Student extends Person
defined class Student

scala> val p: Person = new Student
p: Person = Student@192d3247

scala> p match {
     |   case per: Person => println("it's Person's object")
     |   case _ => println("unknown type")
     | }
it's Person's object

7、protected

// 跟java一样，scala中同样可以使用protected关键字来修饰field和method，这样在子类中就不需要super关键字，直接就可以访问field和method
// 还可以使用protected[this]，则只能在当前子类对象中访问父类的field和method，无法通过其他子类对象访问父类的field和method


###
scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person {
  protected var name: String = "leo"
}

class Student extends Person {
  def makeFriends(s: Student) {
    println("Hi, my name is " + name + ", your name is " + s.name)
  }
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student

scala> val s1 = new Student
s1: Student = Student@63355449

scala> val s2 = new Student
s2: Student = Student@1ab3a8c8

scala> s1.makeFriends(s2)
Hi, my name is leo, your name is leo




scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person {
  protected[this] val name: String  = "leo"
}

class Student extends Person {
  def makeFriends(s: Student) {
    println("my name is " + name + ", your name is " + s.name)
  }
}

// Exiting paste mode, now interpreting.

<console>:22: error: value name is not a member of Student
           println("my name is " + name + ", your name is " + s.name)
                                                                ^

8、调用父类的constructor

// Scala中，每个类可以有一个主constructor和任意多个辅助constructor，而每个辅助constructor的第一行都必须是调用其他辅助constructor或者是主constructor；
因此子类的辅助constructor是一定不可能直接调用父类的constructor的
// 只能在子类的主constructor中调用父类的constructor，以下这种语法，就是通过子类的主构造函数来调用父类的构造函数
// 注意！如果是父类中接收的参数，比如name和age，子类中接收时，就不要用任何val或var来修饰了，否则会认为是子类要覆盖父类的field


###
scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person(val name: String, val age: Int)

class Student(name: String, age: Int, var score: Double) extends Person(name, age) {
  def this(name: String) {
    this(name, 0, 0.0)
  }
  def this(age: Int) {
    this("leo", age, 0)
  }
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student

scala> val s = new Student("leo", 30, 100)
s: Student = Student@783a467b

scala> s.name
res6: String = leo

scala> s.age
res7: Int = 30

scala> s.score
res8: Double = 100.0

scala> val s2 = new Student("leo")
s2: Student = Student@6f204a1a

scala> s2.name
res9: String = leo

scala> s2.age
res10: Int = 0

scala> val s3 = new Student(30)
s3: Student = Student@3224a577

scala> s3.name
res11: String = leo

scala> s3.age
res12: Int = 30

scala> s3.score
res13: Double = 0.0

9、匿名子类

// 在Scala中，匿名子类是非常常见，而且非常强大的。Spark的源码中也大量使用了这种匿名子类。
// 匿名子类，也就是说，可以定义一个类的没有名称的子类，并直接创建其对象，然后将对象的引用赋予一个变量。之后甚至可以将该匿名子类的对象传递给其他函数。


###
scala> :paste
// Entering paste mode (ctrl-D to finish)

class Person(protected val name: String) {
  def sayHello = "Hello, I'm " + name
}

val p = new Person("leo") {
  override def sayHello = "Hi, I'm " + name
}

def greeting(p: Person {def sayHello: String}) {
  println(p.sayHello)
}

// Exiting paste mode, now interpreting.

defined class Person
p: Person = $anon$1@3f270e0a
greeting: (p: Person{def sayHello: String})Unit

scala> greeting(p)
Hi, I'm leo

10、抽象类

// 如果在父类中，有某些方法无法立即实现，而需要依赖不同的子来来覆盖，重写实现自己不同的方法实现。此时可以将父类中的这些方法不给出具体的实现，
只有方法签名，这种方法就是抽象方法。
// 而一个类中如果有一个抽象方法，那么类就必须用abstract来声明为抽象类，此时抽象类是不可以实例化的
// 在子类中覆盖抽象类的抽象方法时，不需要使用override关键字



###
scala> :paste
// Entering paste mode (ctrl-D to finish)

abstract class Person(val name: String) {
  def sayHello: Unit
}

class Student(name: String) extends Person(name) {
  def sayHello: Unit = println("Hello, " + name)
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student

scala> val p = new Person
<console>:11: error: class Person is abstract; cannot be instantiated
       val p = new Person
               ^

scala> val s = new Student("leo")
s: Student = Student@368239c8

scala> s.sayHello
Hello, leo

11、抽象field

// 如果在父类中，定义了field，但是没有给出初始值，则此field为抽象field
// 抽象field意味着，scala会根据自己的规则，为var或val类型的field生成对应的getter和setter方法，但是父类中是没有该field的
// 子类必须覆盖field，以定义自己的具体field，并且覆盖抽象field，不需要使用override关键字


###
scala> :paste
// Entering paste mode (ctrl-D to finish)

abstract class Person {
  val name: String
}

class Student extends Person {
  val name: String = "leo"
}

// Exiting paste mode, now interpreting.

defined class Person
defined class Student

scala> val s = new Student
s: Student = Student@1794d431

scala> s.name
res1: String = leo