How do you copy a linked list into another list?

Question

I'm studying data structures and linked lists, but I'm not getting the concept of how to make a copy of a linked list. Can someone explain this, possibly using pseudocode or C code?

Solution 1

The logic for duplicating a linked list is recursive and based on the following observations:

1. The clone of the empty list is the empty list.
2. The clone of a list with first node x and remaining nodes xs is a copy of x prepended to a clone of xs.

If you encode the linked list in C++, this can be very clean:

struct Node {
    int value;
    Node* next;
};

Node* Clone(Node* list) {
    if (list == NULL) return NULL;

    Node* result = new Node;
    result->value = list->value;
    result->next = Clone(list->next);
    return result;
}

Solution 2

Do you understand how to add a new node to an existing list? And do you understand how to traverse (i.e. iterate over) a list? Copying a list is just performing both of these operations simultaneously (traverse ListA; for each element, copy the element and add it as a new node to ListB).

Solution 3

The answer to this post has been given and accepted - all good! However, if someone is looking for an iterative approach in C#, here it is:

The Node class:

public class Node
{
    public Node(int val)
    {
        Val = val;
    }

    public Node Next { get; set; }
    public int Val { get; }
}

Here is the iterative implementation:

public Node CopyLinkedListIteratively(Node head)
{
    // validation:
    if (head == null) return null;

    // current node is always a 'new' node with value.
    Node currentNode = new Node(head.Val);

    // set copyList and previous to current node to start with - which is true at this point in time!
    Node copyList = currentNode;
    Node previous = currentNode;
    
    // move head one step forward as we already have copied its value.
    head = head.Next;

    // keep moving until we hit a null reference which is the end.
    while (head != null)
    {
        currentNode = new Node(head.Val); // create a new node every time as we move forward.
        previous.Next = currentNode; // set previous node's next to current node as previous node itself is one step behind the current.
        previous = previous.Next; // move prev pointer forward
        head = head.Next; // move head pointer forward as well
    }

    // return the reference to copyList.
    // copyList and previous both started off pointing to the currentNode, then in the while loop
    // previous kept moving forward till all nodes are copied.
    // copyList reference never moved from its position so its still pointing at the start.
    return copyList;
}

Time complexity: O(n)

Space complexity: O(n)

where n = number of nodes in the linked list.

Its personal preference to use recursive or iterative approach, however I would suggest to think about the function call stack when using recursive.

Unit test:

[Test]
public void CopyLinkedListIterativelyTest()
{
    Node head = new Node(1);
    head.Next = new Node(2);
    head.Next.Next = new Node(3);
    head.Next.Next.Next = new Node(4);
    head.Next.Next.Next.Next = new Node(5);

    var actual = runner.CopyLinkedListIteratively(head);

    while (actual != null)
    {
        Assert.AreEqual(head.Val, actual.Val);
        actual = actual.Next;
        head = head.Next;
    }
}