Question

I can’t seem to find the reason why the below code is breaking:

From my perspective what happens is that the integer lbnd is decreased

public class Test {
    public static void main(String[] args) {
        methode1();
    }
    static void methode1() {
        int ubnd = Integer.MAX_VALUE;
        int lbnd = ubnd;

        while ((float)lbnd == (float)ubnd) {
            --lbnd;
        }   
        System.out.println((++lbnd) + ".." + ubnd);
    }
}

the thing is that this loop should be an infinite loop from what I see, however it breaks after 64 loops because the ints changed in value. The outcome is:

2147483584..2147483647

but it should be an infinite loop with:

2147483647..2147483647

Answer 1

Remember that a float does not have sufficient precision to store all 32 bits of an integer precisely. So for an integer of high magnitude, (float) i and (float) (i-1) are identical for the purposes of a float comparison.

(Incidentally, when comparing an int and a float, it will be treated as a float comparison. So to illustrate an additional gotcha, you don’t even need the cast to float on one side of your == comparison.)