Reference: LeetCode
Difficulty: Medium

Problem

Given a binary tree, flatten it to a linked list in-place.

Example:

1
2
3
4
5
    1
/ \
2 5
/ \ \
3 4 6

The flattened tree should look like:

1
2
3
4
5
6
7
8
9
10
11
1
\
2
\
3
\
4
\
5
\
6

Analysis

Preorder

1
2
3
4
5
    1
/ \
2 5
/ \ \
3 4 6

The idea is to use preorder traversal. For example, when root = 1, we want to set 1.right as 2 and set 4.right as 5.

Notice that to know where 4 is during the traversal, we can use a prev node to keep track of the previous node. The previous node of 1 is 4.

Note: I know the code could be shorter, but I think shorter code is barely readable!

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
private TreeNode prevNode = null;

public void flatten(TreeNode root) {
preorder(root);
}

private void preorder(TreeNode root) {
if (root == null) {
return;
}
prevNode = root; // visit root

preorder(root.left); // go left
TreeNode prevTemp = prevNode;
preorder(root.right); // go right

TreeNode rightTemp = root.right; // since prevTemp could be root
root.right = root.left;
root.left = null;

prevTemp.right = rightTemp;
prevTemp.left = null;
}

Time: $O(N)$
Space: $O(h)$

Recursive Function

Note: Just a bit modification on the previous solution.

This idea is a more recursive idea. Let’s say we have a flatten() function. For node 1, we can flatten its two children 2 and 5. At that time, we don’t need to know the detail but we are sure that we will have:

1
2
3
4
5
6
7
    1              1  // root
/ \ / \
2 5 2 5
/ \ \ \ \
3 4 6 3 6
\
4

Two subproblems have been flattened.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
private TreeNode prevNode = null;

public void flatten(TreeNode root) {
if (root == null) {
return;
}
prevNode = root; // visit root

flatten(root.left); // go left (preorder-like traversal)
TreeNode prevTemp = prevNode;
flatten(root.right); // go right

TreeNode rightTemp = root.right; // since prevTemp could be root
root.right = root.left;
root.left = null;

prevTemp.right = rightTemp;
prevTemp.left = null;
}

Time: $O(N)$
Space: $O(N)$

LeetCode Solution

More like reversed postorder!

I don’t like this following method neither.

1
2
3
4
5
6
7
8
9
10
11
12
private TreeNode prev = null;

public void flatten(TreeNode root) {
if (root == null) {
return;
}
flatten(root.right);
flatten(root.left);
root.right = prev;
root.left = null;
prev = root;
}

Other Solution: (not using field variable)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
public void flatten(TreeNode root) {
flatten(root, null);
}

private TreeNode flatten(TreeNode root, TreeNode prev) {
if (root == null) {
return prev;
}
prev = flatten(root.right, prev);
prev = flatten(root.left, prev);
root.right = prev;
root.left = null;
prev = root;
return prev;
}

Time: $O(N)$
Space: $O(h)$

Iteration (Preorder)

Use a stack to simulate the preorder traversal, but we also use peek operation to set current node’s right child to its left child.

1
2
3
4
5
    1
/ \
2 5
/ \ \
3 4 6

Note:

  • Before peeking, remember to check if the stack is empty.
  • Actually, we don’t have to check the stack’s size before calling peek(). If the stack is empty, it will just return null. But it does not hold true for Stack‘s peek().
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
public void flatten(TreeNode root) {
if (root == null) {
return;
}
Stack<TreeNode> stack = new Stack<>();
stack.push(root);
while (stack.size() > 0) {
TreeNode p = stack.pop();

if (p.right != null) { // push right first
stack.push(p.right);
}

if (p.left != null) { // push left then
stack.push(p.left);
}

if (stack.size() > 0) { // peek the left
p.right = stack.peek();
}

p.left = null; // Remember to set left as null
}
}

Time: $O(N)$
Space: $O(h)$