Skip to content
Related Articles
Get the best out of our app
GeeksforGeeks App
Open App

Related Articles

C Program For Pointing To Next Higher Value Node In A Linked List With An Arbitrary Pointer

Improve Article
Save Article
Like Article
Improve Article
Save Article
Like Article

Given singly linked list with every node having an additional “arbitrary” pointer that currently points to NULL. Need to make the “arbitrary” pointer point to the next higher value node.


We strongly recommend to minimize your browser and try this yourself first

A Simple Solution is to traverse all nodes one by one, for every node, find the node which has the next greater value of the current node and change the next pointer. Time Complexity of this solution is O(n2).

An Efficient Solution works in O(nLogn) time. The idea is to use Merge Sort for linked list
1) Traverse input list and copy next pointer to arbit pointer for every node. 
2) Do Merge Sort for the linked list formed by arbit pointers.

Below is the implementation of the above idea. All of the merger sort functions are taken from here. The taken functions are modified here so that they work on arbit pointers instead of next pointers.  


// C program to populate arbit pointers to
// next higher value using merge sort
// Link list node
struct Node
    int data;
    struct Node* next, *arbit;
// Function prototypes
struct Node* SortedMerge(struct Node* a,
                         struct Node* b);
void FrontBackSplit(struct Node* source,
                    struct Node** frontRef,
                    struct Node** backRef);
/* Sorts the linked list formed by
   arbit pointers (does not change
   next pointer or data) */
void MergeSort(struct Node** headRef)
    struct Node* head = *headRef;
    struct Node* a, *b;
    // Base case -- length 0 or 1
    if ((head == NULL) ||
        (head->arbit == NULL))
    // Split head into 'a' and 'b'
    // sublists
    FrontBackSplit(head, &a, &b);
    // Recursively sort the sublists
    // answer = merge the two sorted
    // lists together
    *headRef = SortedMerge(a, b);
   for details of this function */
struct Node* SortedMerge(struct Node* a,
                         struct Node* b)
    struct Node* result = NULL;
    // Base cases
    if (a == NULL)
        return (b);
    else if (b==NULL)
        return (a);
    // Pick either a or b, and recur
    if (a->data <= b->data)
        result = a;
        result->arbit =
        SortedMerge(a->arbit, b);
        result = b;
        result->arbit =
        SortedMerge(a, b->arbit);
    return (result);
/* Split the nodes of the given list into
   front and back halves, and return the
   two lists using the reference parameters.
   If the length is odd, the extra node
   should go in the front list. Uses the
   fast/slow pointer strategy.  */
void FrontBackSplit(struct Node* source,
                    struct Node** frontRef,
                    struct Node** backRef)
    struct Node* fast, *slow;
    if (source==NULL ||
        // length < 2 cases
        *frontRef = source;
        *backRef = NULL;
    slow = source, 
    fast = source->arbit;
    /* Advance 'fast' two nodes, and
       advance 'slow' one node */
    while (fast != NULL)
        fast = fast->arbit;
        if (fast != NULL)
            slow = slow->arbit;
            fast = fast->arbit;
    /* 'slow' is before the midpoint in
        the list, so split it in two at
        that point. */
    *frontRef = source;
    *backRef = slow->arbit;
    slow->arbit = NULL;
/* Function to insert a node at the
   beginning of the linked list */
void push(struct Node** head_ref,
          int new_data)
    // Allocate node
    struct Node* new_node =
           (struct Node*) malloc(sizeof(struct Node));
    // Put in the data 
    new_node->data = new_data;
    // Link the old list of the new node
    new_node->next = (*head_ref);
    new_node->arbit = NULL;
    // Move the head to point to
    // the new node
    (*head_ref) = new_node;
// Utility function to print
// result linked list
void printListafter(struct Node *node,
                    struct Node *anode)
    printf("Traversal using Next Pointer");
    while (node!=NULL)
        printf("%d, ", node->data);
        node = node->next;
    printf("Traversal using Arbit Pointer");
    while (anode!=NULL)
        printf("%d, ", anode->data);
        anode = anode->arbit;
// This function populates arbit pointer
// in every node to the next higher value.
// And returns pointer to the node with
// minimum value
struct Node* populateArbit(struct Node *head)
    // Copy next pointers to arbit pointers
    struct Node *temp = head;
    while (temp != NULL)
        temp->arbit = temp->next;
        temp = temp->next;
    // Do merge sort for arbitrary pointers
    // Return head of arbitrary pointer
    // linked list
    return head;
// Driver code
int main()
    // Start with the empty list
    struct Node* head = NULL;
    // Let us create the list shown above
    push(&head, 3);
    push(&head, 2);
    push(&head, 10);
    push(&head, 5);
    // Sort the above created Linked List
    struct Node *ahead = populateArbit(head);
    printf("Result Linked List is: ");
    printListafter(head, ahead);
    return 0;


Result Linked List is:
Traversal using Next Pointer
5, 10, 2, 3,
Traversal using Arbit Pointer
2, 3, 5, 10,

The time complexity of the populateArbit function is O(n log n), where n is the number of nodes in the linked list, due to the use of merge sort to sort the arbitrary pointers.

The space complexity is O(log n), which is the space used by the recursive stack during the merge sort algorithm.

Please refer complete article on Point to next higher value node in a linked list with an arbitrary pointer for more details!

My Personal Notes arrow_drop_up
Last Updated : 20 Mar, 2023
Like Article
Save Article
Similar Reads
Related Tutorials