Quicksort with Median of Medians Method.

Question

I am trying to test the MM, (Median of Medians method), to see that it has O(n) complexity. I can not get it to work, it is in part because I don't understand how to incorperate the MM code in to the Quicksort Method. Also note that there is an error in my Quick sort, the list sorts to some extent. I am trying to figure that out. So, all in all I have two problems... Here is what I have:
public class MMQuickSort
     private static long comparisons = 0;
     private static long exchanges = 0;
     static final int numberofgroups = 5;
     private static int groupsize = 0;
          public static void Quicksort(int[] a)
               // shuffle(a);           // to guard against worst-case and makes average = best = O(nlogn) without MM.
                groupsize = a.length / numberofgroups;
          QuickSort(a, 0, a.length);
     public static void QuickSort(int[] inarray, int low, int high)
          int thepivot, position = 0;
               if (low < high)
                    thepivot = Partition(inarray, low, high, high - 1);
                    QuickSort(inarray, low, thepivot - 1);
                    QuickSort(inarray, thepivot + 1, high);
     public static int Partition(int[] inarray, int low, int high, int pivot)
          int v = inarray[low];
          int j = low;
          int temp;
          //int moveleft = high + 1;
          //int v = inarray[low];
          for (int i = low + 1; i < high; i++)
               if (inarray[i] < v)
                    j++;
                    temp = inarray;
                    inarray[i] = inarray[j];
                    inarray[j] = temp;
               pivot = j;
               temp = inarray[low];
               inarray[low] = inarray[pivot];
               inarray[pivot] = temp;
          return pivot;
     public static int Select(int a[], int n, int k)
          int pivot= 0;
          if (n <= numberofgroups)
               Quicksort(a);
          else
               pivot = Select(a, a.length / numberofgroups, ((int)Math.ceil(a.length / numberofgroups / 2)));
          if (k == pivot)
               return k;
          else if (k < pivot)
               return Select(a, pivot - 1, k);
          else
               return Select(a, n - pivot, k - pivot);
To get it to work I call:
Quicksort(data) 
where "data" is a dis-ordered array of n elements.
Any thoughts would be much appreciated.
Thank you,
Brian                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           

Answer 1

it would be good if you posted the exception next time
the exception occurs in the partition() method, on the line
while (less(a[++i], v)) ;it appears that l = 13 and r = 12, and a[++i] = a[13], a[14], a[15]... are all less than v = a[12]
"l" probably shouldn't be greater than "r"
on closer examination, it appears that your quicksort() algorithm has no base case. it keeps partitioning and making recursive calls until the subarray you are partitioning over has negative size (?)