John Meyer's SCJP 5 mock exam - casting

[Rizwan Patel]

Hi all,
When we try to assign the higher value literals to the lower value literals at that time we need to do the explicitly cast to avoid the compiler problem.

    class test
{
   int x;
   public static void main (  String [] args )
   {
      final int i;
      i = 127;
      byte b = i; //Should give unable to cast int to byte
      System.out.println(b);
 
   }
}

But above problems correct answer is stating that "Compiler error:loss of precesion".
I know this is the stupid question but I want to clear the doubt for this.

[Ankit Garg]

Rizwan I didn't get what you are trying to say. The error is correct.. It will say possible loss of precision. What are you expecting...

[Fidel Barro]

It will difinitely give you an compile error since your assigning a value which in int type to a variable which has byte type.. But of course, you can explicitly cast it (e.g byte b = (byte)i; ) but the precise value will be lost.. The value will be squeeze to byte and disregard the rest which will not be fitted to the size of byte(8 bits)

I hope I answered your question correctly.

[Rizwan Patel]

When run this program on my machine i get the below comppiler problem

Type mismatch: cannot convert from int to byte

I read about the loss of precision while down-casting (only reference in literals).

Whether the above compiler error message are equal.

[Rizwan Patel]

When run this program on my machine i get the below comppiler problem

Type mismatch: cannot convert from int to byte

I read about the loss of precision while down-casting (only reference in literals).

Whether the above compiler error message are equal.

[subhasish nag]

Because here i is not compile time constant.If you replace the line final int i;
i = 127; with int i=127 then it will be fine.

[Rizwan Patel]

When run this program on my machine i get the below comppiler problem

Type mismatch: cannot convert from int to byte

I read about the loss of precision while down-casting (only reference in literals).

Whether the above compiler error message are equal.

[sumi rankan]

Rizwan,
This sure will give you possible loss of precision error.But if you initialize final int i where you declared it, will compile fine.Because the compiler sees that 127 will surely fit in a byte and its value is not going to change since it is final.
But, if you just say int i=127 or anything <127 ,the compiler cannot be sure if that value is going to be same.So it flags an error eventhough the value fits comfortably in a byte.
Another thing,this holds true only if the value is <=127.If the value is above the range of byte,even if it is a comile time constant ,compiler flags an error.Hope this clears!