Given a non-empty binary search tree and a target value, find k values in the BST that are closest to the target.
Note:
- Given target value is a floating point.
- You may assume k is always valid, that is: k ≤ total nodes.
- You are guaranteed to have only one unique set of k values in the BST that are closest to the target.
Follow up:
Assume that the BST is balanced, could you solve it in less than O(n) runtime (where n = total nodes)?
Hint:
- Consider implement these two helper functions:
getPredecessor(N)
, which returns the next smaller node to N.getSuccessor(N)
, which returns the next larger node to N.
- Try to assume that each node has a parent pointer, it makes the problem much easier.
- Without parent pointer we just need to keep track of the path from the root to the current node using a stack.
- You would need two stacks to track the path in finding predecessor and successor node separately.
Solution:
Get Next:
if the node has right subtree, the next would be the left-most node of the right sub-tree.
if the node does not have right subtree, go to the parent(stack.peek()),
if the parent's value bigger than the nodes value, means the node is the left-child, the parent is the next node,
if the parent's value smaller than the nodes value, means the node is the right- child, go to parent's parent util when find the node is in the left subtree of an ancestor.
Get Prev is similar to get Next.
/**
* Definition for a binary tree node.
* public class TreeNode {
* int val;
* TreeNode left;
* TreeNode right;
* TreeNode(int x) { val = x; }
* }
*/
public class Solution {
public List<Integer> closestKValues(TreeNode root, double target, int k) {
Stack<TreeNode> pre = new Stack<>();
Stack<TreeNode> post = new Stack<>();
TreeNode cur = root;
double preDiff = 1.0 * Integer.MAX_VALUE;
TreeNode preNode = null;
double postDiff = 1.0 * Integer.MAX_VALUE;
TreeNode postNode = null;
// prepare stacks
while(cur!= null){
pre.push(cur);
post.push(cur);
if(cur.val > target){
if(postNode == null || cur.val - target < postDiff){
postDiff = cur.val - target;
postNode = cur;
}
cur = cur.left;
} else {
if(preNode == null || target - cur.val < preDiff){
preDiff = target - cur.val;
preNode = cur;
}
cur = cur.right;
}
}
// get the Node that is first one smaller than target
while(!pre.empty() && pre.peek() != preNode){
pre.pop();
}
// get the Node that is the first one bigger than target
while(!post.empty() && post.peek() != postNode){
post.pop();
}
//get result
List<Integer> ret = new LinkedList<>();
Integer bigger = getNext(post);
Integer smaller = getPrev(pre);
for(int i = 0; i < k; i++){
if(smaller == null){
ret.add(bigger);
bigger = getNext(post);
} else if(bigger == null){
ret.add(smaller);
smaller = getPrev(pre);
} else {
if( bigger - target < target - smaller){
ret.add(bigger);
bigger = getNext(post);
} else {
ret.add(smaller);
smaller = getPrev(pre);
}
}
}
return ret;
}
//get next, and prepare for the next next
Integer getNext(Stack<TreeNode> post){
if(post.empty()) return null;
TreeNode cur = post.pop();
int ret = cur.val;
if(cur.right != null){
pushAllLeft(post,cur.right);
} else {
while(!post.empty() && post.peek().val < ret){
post.pop();
}
}
return ret;
}
void pushAllLeft(Stack<TreeNode> post, TreeNode root){
while(root != null){
post.push(root);
root = root.left;
}
}
//get prev, and prepare for the prev prev
Integer getPrev(Stack<TreeNode> pre){
if(pre.empty()) return null;
TreeNode cur = pre.pop();
int ret = cur.val;
if(cur.left != null){
pushAllRight(pre,cur.left);
} else {
while(!pre.empty() && pre.peek().val > ret){
pre.pop();
}
}
return ret;
}
void pushAllRight(Stack<TreeNode> pre, TreeNode root){
while(root != null){
pre.push(root);
root = root.right;
}
}
}