Given a linked list, rearrange it such that the converted list should be of the form a < b > c < d > e < f … where a, b, c… are consecutive data nodes of the linked list.
Examples:
Input: 1->2->3->4 Output: 1->3->2->4 Explanation: 1 and 3 should come first before 2 and 4 in zig-zag fashion, So resultant linked-list will be 1->3->2->4. Input: 11->15->20->5->10 Output: 11->20->5->15->10
We strongly recommend that you click here and practice it, before moving on to the solution.
A simple approach to do this is to sort the linked list using merge sort and then swap alternate, but that requires O(n Log n) time complexity. Here n is a number of elements in the linked list.
An efficient approach that requires O(n) time is, using a single scan similar to bubble sort and then maintain a flag for representing which order () currently we are. If the current two elements are not in that order then swap those elements otherwise not. Please refer to this for a detailed explanation of the swapping order.
// Java program to arrange linked // list in zigzag fashion class GfG
{ // Link list Node
static class Node
{
int data;
Node next;
}
static Node head = null ;
static int temp = 0 ;
// This function distributes
// the Node in zigzag fashion
static void zigZagList(Node head)
{
// If flag is true, then
// next node should be greater
// in the desired output.
boolean flag = true ;
// Traverse linked list starting
// from head.
Node current = head;
while (current != null &&
current.next != null )
{
// "<" relation expected
if (flag == true )
{
/* If we have a situation like
A > B > C where A, B and C
are consecutive Nodes in list
we get A > B < C by swapping B
and C */
if (current.data > current.next.data)
{
temp = current.data;
current.data = current.next.data;
current.next.data = temp;
}
}
// ">" relation expected
else
{
/* If we have a situation like
A < B < C where A, B and C
are consecutive Nodes in list
we get A < C > B by swapping
B and C */
if (current.data < current.next.data)
{
temp = current.data;
current.data = current.next.data;
current.next.data = temp;
}
}
current = current.next;
// flip flag for reverse checking
flag = !(flag);
}
}
// UTILITY FUNCTIONS
// Function to push a Node
static void push( int new_data)
{
// Allocate Node
Node new_Node = new Node();
// Put in the data
new_Node.data = new_data;
// Link the old list off the
// new Node
new_Node.next = (head);
// Move the head to point to
// the new Node
(head) = new_Node;
}
// Function to print linked list
static void printList(Node Node)
{
while (Node != null )
{
System.out.print(Node.data +
"->" );
Node = Node.next;
}
System.out.println( "NULL" );
}
// Driver code
public static void main(String[] args)
{
// Start with the empty list
// Node head = null;
// create a list 4 -> 3 -> 7 ->
// 8 -> 6 -> 2 -> 1
// answer should be -> 3 7 4 8
// 2 6 1
push( 1 );
push( 2 );
push( 6 );
push( 8 );
push( 7 );
push( 3 );
push( 4 );
System.out.println(
"Given linked list " );
printList(head);
zigZagList(head);
System.out.println(
"Zig Zag Linked list " );
printList(head);
}
} // This code is contributed by Prerna Saini. |
Output:
Given linked list 4->3->7->8->6->2->1->NULL Zig Zag Linked list 3->7->4->8->2->6->1->NULL
Time Complexity: O(n),The time complexity of this algorithm is O(n) as it only requires one iteration through the linked list.
Space Complexity: O(1),The space complexity is also O(1) as we only need a few variables to store data.
Another Approach:
In the above code, the push function pushes the node at the front of the linked list, the code can be easily modified for pushing the node at the end of the list. Another thing to note is, swapping of data between two nodes is done by swap by value not swap by links for simplicity, for the swap by links technique please see this.
This can be also be done recursively. The idea remains the same, let us suppose the value of the flag determines the condition we need to check for comparing the current element. So, if the flag is 0 (or false) the current element should be smaller than the next and if the flag is 1 ( or true ) then the current element should be greater than the next. If not, swap the values of nodes.
// Java program for the above approach import java.io.*;
// Node class class Node
{ int data;
Node next;
Node( int data)
{
this .data = data;
}
} public class GFG
{ private Node head;
// Print Linked List
public void printLL()
{
Node t = head;
while (t != null )
{
System.out.print(t.data +
" ->" );
t = t.next;
}
System.out.println();
}
// Swap both nodes
public void swap(Node a,
Node b)
{
if (a == null || b == null )
return ;
int temp = a.data;
a.data = b.data;
b.data = temp;
}
// Rearrange the linked list
// in zig zag way
public Node zigZag(Node node,
int flag)
{
if (node == null ||
node.next == null )
{
return node;
}
if (flag == 0 )
{
if (node.data >
node.next.data)
{
swap(node, node.next);
}
return zigZag(node.next, 1 );
}
else
{
if (node.data <
node.next.data)
{
swap(node, node.next);
}
return zigZag(node.next, 0 );
}
}
// Driver Code
public static void main(String[] args)
{
GFG lobj = new GFG();
lobj.head = new Node( 11 );
lobj.head.next = new Node( 15 );
lobj.head.next.next =
new Node( 20 );
lobj.head.next.next.next =
new Node( 5 );
lobj.head.next.next.next.next =
new Node( 10 );
lobj.printLL();
// 0 means the current element
// should be smaller than the next
int flag = 0 ;
lobj.zigZag(lobj.head, flag);
System.out.println(
"LL in zig zag fashion : " );
lobj.printLL();
}
} |
Output:
11 ->15 ->20 ->5 ->10 -> LL in zig zag fashion : 11 ->20 ->5 ->15 ->10 ->
Complexity Analysis:
-
Time Complexity: O(n).
Traversal of the list is done only once, and it has ‘n’ elements. -
Auxiliary Space: O(n).
O(n) extra space due to recursive stack.
Please refer complete article on Rearrange a Linked List in Zig-Zag fashion for more details!