用 Java 合併排序實現

下面是使用泛型方法在 Java 中實現的。它是相同的演算法，如上所述。

public interface InPlaceSort<T extends Comparable<T>> {
void sort(final T[] elements); }

public class MergeSort < T extends Comparable < T >> implements InPlaceSort < T > {

@Override
public void sort(T[] elements) {
    T[] arr = (T[]) new Comparable[elements.length];
    sort(elements, arr, 0, elements.length - 1);
}

// We check both our sides and then merge them
private void sort(T[] elements, T[] arr, int low, int high) {
    if (low >= high) return;
    int mid = low + (high - low) / 2;
    sort(elements, arr, low, mid);
    sort(elements, arr, mid + 1, high);
    merge(elements, arr, low, high, mid);
}

private void merge(T[] a, T[] b, int low, int high, int mid) {
    int i = low;
    int j = mid + 1;

    // We select the smallest element of the two. And then we put it into b
    for (int k = low; k <= high; k++) {

        if (i <= mid && j <= high) {
            if (a[i].compareTo(a[j]) >= 0) {
                b[k] = a[j++];
            } else {
                b[k] = a[i++];
            }
        } else if (j > high && i <= mid) {
            b[k] = a[i++];
        } else if (i > mid && j <= high) {
            b[k] = a[j++];
        }
    }

    for (int n = low; n <= high; n++) {
        a[n] = b[n];
    }}}