Given inorder and postorder traversal of a tree, construct the binary tree.
Note:
You may assume that duplicates do not exist in the tree.
由于树中没有重复的值,那么我们可以通过从后序遍历中找到根节点,然后根据中序遍历中根的位置来确定左子树和右子树。
You may assume that duplicates do not exist in the tree.
public TreeNode buildTree(int[] inorder, int[] postorder) {
int n = inorder.length;
if(n == 0) return null;
if(n == 1) return new TreeNode(inorder[0]);
int key = postorder[n-1];
int ind = 0;
while(inorder[ind] != key) ind++;
int[] il = new int[ind];
int[] pl = new int[ind];
int[] ir = new int[n-ind-1];
int[] pr = new int[n-ind-1];
for(int i = 0; i < ind; i++) {
il[i] = inorder[i];
pl[i] = postorder[i];
}
for(int i = 0; i < n-ind-1; i++) {
ir[i] = inorder[ind+i+1];
pr[i] = postorder[ind+i];
}
TreeNode root = new TreeNode(key);
root.left = buildTree(il, pl);
root.right = buildTree(ir, pr);
return root;
}
没有评论:
发表评论