Given a binary tree, return the postorder traversal of its nodes' values.
For example: Given binary tree {1,#,2,3}
,
1 2 / 3
return [3,2,1]
.
Note: Recursive solution is trivial, could you do it iteratively?
思考:原思路是为每个节点增加一个标记。这个思路先序遍历不要标记,因为当前访问结点直接出栈,不用看是否访问过。不过此思路redefinition of 'struct TreeNode'。
struct TreeNode { int val; TreeNode *left; TreeNode *right; bool flag; TreeNode(int x) : val(x), left(NULL), right(NULL), flag(false) {} }; class Solution { private: vector<int> ret; public: vector<int> postorderTraversal(TreeNode *root) { // IMPORTANT: Please reset any member data you declared, as // the same Solution instance will be reused for each test case. ret.clear(); if(!root) return ret; stack<TreeNode *> s; s.push(root); while(!s.empty()) { TreeNode *cur=s.top(); if(cur->left&&!cur->left->flag) s.push(cur->left); else if(cur->right&&!cur->right->flag) s.push(cur->right); else { ret.push_back(cur->val); cur->flag=true; s.pop(); } } return ret; } };
既然不能增加标记结点,那么遍历过的结点直接“删掉”。不过这么做破坏了原树结构。
class Solution { private: vector<int> ret; public: vector<int> postorderTraversal(TreeNode *root) { // IMPORTANT: Please reset any member data you declared, as // the same Solution instance will be reused for each test case. ret.clear(); if(!root) return ret; stack<TreeNode *> s; s.push(root); TreeNode *cur=s.top(); while(!s.empty()) { TreeNode *cur=s.top(); if(!cur->left&&!cur->right) { ret.push_back(cur->val); s.pop(); } if(cur->right) { s.push(cur->right); cur->right=NULL; } if(cur->left) { s.push(cur->left); cur->left=NULL; } } return ret; } };
网上搜索有没有更好的方法,发现自己思维定势了,只知道不能递归就用stack代替。参考http://www.cnblogs.com/changchengxiao/p/3416402.html