scala> val a = Array [Double] (10)
a: Array[Double] = Array(10.0)
scala> val a = new Array [Double] (10)
a: Array[Double] = Array(0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0)
Why these two expressions have different semantics?
Use the nextInt() method to fill the array with random values. It generates the random integer values, and to get filled, we used the fill keyword.
Constructors in Scala describe special methods used to initialize objects. When an object of that class needs to be created, it calls the constructor of the class. It can be used to set initial or default values for object attributes.
Creating an Array and Accessing Its ElementsScala translates the first line in the example above into a call to Array::apply(), defined in the Array companion object. Such a method takes a variable number of values as input and creates an instance of Array[T], where T is the type of the elements.
Scala constructor is used for creating an instance of a class. There are two types of constructor in Scala – Primary and Auxiliary. Not a special method, a constructor is different in Scala than in Java constructors. The class' body is the primary constructor and the parameter list follows the class name.
It's a bit confusing, but Scala has the notion of classes which you can create instances of, and objects, which are basically singleton instances of a class. It also has the notion of companion classes, which is a pair of a class and an object with the same name. This mechanism allows a "class" to essentially have static methods, which are otherwise not possible in Scala.
Array
has both a class and a companion object. Furthermore, the Array
object has an apply
method. apply
means you can create an object with Array(arg)
. But because Array
is a companion class, it also has a constructor that can be called via the more usual mechanism of new Array(arg)
.
The issue is that apply
in the the Array
object has different semantics than the Array
constructors. The apply
method creates an array out of the specified objects, so, for example, Array(1,2,3)
returns an array consisting of the objects 1
, 2
, and 3
. The constructors, on the other hand, take arguments that specify the size of the dimensions of the array (so you can create multidimensional arrays), and then initialize all slots to a default value.
So, basically:
val a = Array [Double] (10)
calls the apply
method on the Array
object, which creates a new array containing the given objects.val a = new Array [Double] (10)
calls the Array
constructor, which creates a new array with 10 slots, all initialized to a default value of 0.0
.new Array[Double](10)
is supposed to be equivalent to new double[10]
in Java.
But Scala also provides convenience methods on the singletons corresponding to its collection classes, and Array
is no exception.
Thus, if you can say List(1,2,3,4,5)
it seems natural that you could also say Array(1,2,3,4,5)
. And you can.
But it does leave one in the slightly awkward position of having rather different results depending on whether one adds the word new
or not. Given the competing interests, I think it's the best solution overall, but it does take a little getting used to.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With