一、二叉树
前序遍历、中序遍历、后序遍历:
1.前序遍历:
def preorderTraversal(self, root: TreeNode) -> List[int]:
if root == None:
return []
# 分别记录遍历结果,存储栈
result, istack = [], []
while root != None or len(istack) != 0:
# 嵌套循环
while root != None:
# 前序遍历,压入栈前先输出
result.append(root.val)
istack.append(root)
root=root.left
# 栈弹出
root = istack[-1]
istack = istack[:-1]
root = root.right
return result
2.中序遍历:
def inorderTraversal(self, root: TreeNode) -> List[int]:
if root == None:
return []
# 分别记录遍历结果,存储栈
result, istack = [], []
while root != None or len(istack) != 0:
while root != None:
istack.append(root)
root = root.left
# 中序遍历,等左子树压入栈完毕,再从栈输出
cur = istack[-1]
result.append(cur.val)
istack = istack[:-1]
root = cur.right
return result
3.后序遍历:
def postorderTraversal(self, root: TreeNode) -> List[int]:
if root == None:
return []
# 分别记录遍历结果,存储栈,后序遍历必要的前一个输出节点记录
result, istack, lastnode = [], [], None
while root != None or len(istack) != 0:
while root != None:
istack.append(root)
root = root.left
# 左子树压入栈后,判断当前节点是否可输出
cur = istack[-1]
# 如果当前节点无右子树,或者右子树已经输出,则该点可输出
if cur.right == None or cur.right == lastnode:
istack = istack[:-1]
result.append(cur.val)
lastnode = cur
# 否则继续将该点右子树入栈,再继续循环
else:
root = cur.right
return result
深度搜索DFS,层次搜索BFS:
1.层次搜索/遍历
def levelOrder(self, root: TreeNode) -> List[List[int]]:
if root == None:
return []
# 层序遍历,需要借助队列记录
result, iqueue = [], []
# 队列需要先入队
iqueue.append(root)
while len(iqueue) != 0:
# 遍历当前层之前,先记录该层的个数,以便队列边记录边遍历
l, temp = len(iqueue), []
for i in range(0, l):
# 遍历时,依次取出和输出队中的点,并将子树入队
cur = iqueue[0]
iqueue = iqueue[1:]
temp.append(cur.val)
if cur.left != None:
iqueue.append(cur.left)
if cur.right != None:
iqueue.append(cur.right)
result.append(temp)
return result
分治法:递归返回条件+分段处理+合并结果(快速排序、归并排序、二叉树)
1.简单分治法:用于计算二叉树的最大深度
def maxDepth(self, root: TreeNode) -> int:
if root == None:
return 0
left = self.maxDepth(root.left)
right = self.maxDepth(root.right)
return left + 1 if left > right else right + 1
2.判断是否为平衡二叉树:
def isBalanced(self, root: TreeNode) -> bool:
if self.judge(root) == -1:
return False
else:
return True
def judge(self, root: TreeNode) -> int:
if root == None:
return 0
left = self.judge(root.left)
right = self.judge(root.right)
# 借助层数来记录是否平衡,如果为-1,则直接返回不平衡
if left == -1 or right == -1 or left - right > 1 or right - left > 1:
return -1
else:
return left + 1 if left > right else right + 1
3.找出二叉树中最大路径和【hard】
def maxPathSum(self, root: TreeNode) -> int:
# 最大和:左/右子树最大和,左右子树加根节点的最大和
summax = self.find(root)
return summax[1]
def find(self, root: TreeNode) -> List[int]:
if root == None:
return [0, float('-inf')]
# 分治
left = self.find(root.left)
right = self.find(root.right)
result = []
# 合并规整,先求单边值,再求最大和
if left[0] > right[0]:
result.append(max(left[0] + root.val, 0))
else:
result.append(max(right[0] + root.val, 0))
result.append(max(max(left[1], right[1]), (left[0] + right[0] + root.val)))
return result
4.找出二叉树的公共祖先
def lowestCommonAncestor(self, root: 'TreeNode', p: 'TreeNode', q: 'TreeNode') -> 'TreeNode':
if root == None:
return root
if root == p or root == q:
return root
# 分治
left = self.lowestCommonAncestor(root.left, p, q)
right = self.lowestCommonAncestor(root.right, p, q)
# 归并整合,返回为空则未找到,返回不为空则找到
if left != None and right != None:
return root
if left != None:
return left
if right != None:
return right
return None
BFS层次遍历:
1.锯齿形层次遍历:
def zigzagLevelOrder(self, root: TreeNode) -> List[List[int]]:
if root == None:
return []
# 层次遍历需要借助队列顺序,锯齿形需要逆转标记
result, iqueue, iturn = [], [], True
iqueue.append(root)
while len(iqueue) > 0:
temp, l = [], len(iqueue)
for i in rang