Get value directly or via method in Scala?

问题:

One is able to define methods without paranthesis if they have no arguments. A special use-case is to use this to get values. Should I do this, or rather directly get the value?

So

class Complex(real: Double, imaginary: Double) {
  def re = real
  def im = imaginary
  override def toString() =
    "" + re + (if (im < 0) "" else "+") + im + "i"
}

or

class Complex(real: Double, imaginary: Double) {
  val re = real
  val im = imaginary
  override def toString() =
    "" + re + (if (im < 0) "" else "+") + im + "i"
}

回答1:

You can put val in front of the constructor arguments to make it a bit shorter, which is effectively the same as your second piece of code:

class Complex(val re: Double, val im: Double) {
  override def toString() = "" + re + (if (im < 0) "" else "+") + im + "i"
}

Note that def defines a method, while val defines a final member variable. Since the return value of these two methods is fixed, there's not really a reason to make them methods. So, in this case, use val instead of def.

回答2:

Even better: make it a case class:

case class Complex(re: Double, im: Double) {
   override def toString() = "%f%+fi".format(re,im)
}

This gives you re and im as members, plus some additional perks, such as copy:

val a = Complex(1,2)
val realA = a.copy(im = 0)

and unapply:

def isImaginary(a: Complex) = a match {
   case Complex(0, _) => true
   case _ => false
}

def abs(a: Complex) = a match {
   case Complex(re, im) => re*re + im*im
}

and also equals, hashCode, etc.