Let's follow the same flow from the previous problem. So, this time, the Function class will look as follows:

public class Function {

  private final int n;

  public Function(int n) {

    this.n = n;
  }

  public int yMinusX(int x, int y) {

    return y - x;
  }
}

Notice that the preceding snippet of code doesn't assume any range restrictions over x, y, and n. Now, let's impose the following ranges:

• n must be between 0 (inclusive) and 101 (exclusive), so n belongs to [0, 101].
• In the yMinusX() method, the range bounded by x and y, [x, y] must be a subrange of [0, n].

These ranges can be imposed in code via the if statements as follows:

public class Function {

  private static final int N_UPPER_BOUND = 101;
  private final int n;

  public Function(int n) {

    if (n < 0 || n >= N_UPPER_BOUND) {
      throw new IndexOutOfBoundsException("...");
    }

    this.n = n;
  }

  public int yMinusX(int x, int y) {

    if (x < 0 || x > y || y >= n) {
      throw new IndexOutOfBoundsException("...");
    }

    return y - x;
  }
}

Based on the previous problem, the condition for n can be replaced with Objects.checkIndex(). Moreover, the JDK 9 Objects class comes with a method named checkFromToIndex(int start, int end, int length) that checks whether the given subrange [given start, given end] is within the bounds of the range from [0, given length]. So, this method can be applied to the yMinusX() method to check that the range bounded by x and y, [x, y) is a subrange of [0, n]:

public class Function {

  private static final int N_UPPER_BOUND = 101;
  private final int n;

  public Function(int n) {

    this.n = Objects.checkIndex(n, N_UPPER_BOUND);
  }

  public int yMinusX(int x, int y) {

    Objects.checkFromToIndex(x, y, n);
    return y - x;
  }
}

For example, the following test will lead to IndexOutOfBoundsException since x is greater than y:

Function f = new Function(50);
int r = f.yMinusX(30, 20);