mirror of
https://github.com/SunnyQjm/algorithm-review.git
synced 2026-06-03 08:16:43 +08:00
56 lines
1.8 KiB
Python
56 lines
1.8 KiB
Python
#!/usr/bin/env python
|
||
# coding=utf-8
|
||
|
||
#######################################################################################
|
||
# Leetcode 5 最长回文子串
|
||
#
|
||
# 给定一个字符串 s，找到 s 中最长的回文子串。你可以假设 s 的最大长度为 1000。
|
||
#
|
||
# 示例 1：
|
||
# 输入: "babad"
|
||
# 输出: "bab"
|
||
# 注意: "aba" 也是一个有效答案。
|
||
#
|
||
# 示例 2：
|
||
# 输入: "cbbd"
|
||
# 输出: "bb"
|
||
#######################################################################################
|
||
|
||
|
||
class Solution:
|
||
def longestPalindrome(self, s):
|
||
"""
|
||
:type s: str
|
||
:rtype str
|
||
|
||
(knowledge)
|
||
|
||
思路：
|
||
1. 使用动态规划
|
||
2. dp[i][j] => s[i:j] 子串是否是一个回文字符串
|
||
3. 状态转移方程
|
||
f(i, j) = True i == j
|
||
s[i] == s[j] i + 1 == j
|
||
f(i + 1, j - 1) && s[i] == s[j] i + 1 < j
|
||
"""
|
||
dp, resultStart, resultEnd = [[True if i == j else False for j in range(len(s))] for i in range(len(s))], 0, 1
|
||
|
||
for j in range(0, len(s) - 1):
|
||
dp[j][j + 1] = s[j] == s[j + 1]
|
||
if dp[j][j + 1]:
|
||
resultStart, resultEnd = j, j + 2
|
||
|
||
for i in range(2, len(s)):
|
||
for j in range(0, len(s) - i):
|
||
if dp[j + 1][j + i - 1] and s[j] == s[j + i]:
|
||
resultStart, resultEnd, dp[j][j + i] = j, j + i + 1, True
|
||
return s[resultStart: resultEnd]
|
||
|
||
|
||
if __name__ == '__main__':
|
||
solution = Solution()
|
||
print(solution.longestPalindrome("abcba"), "= abcba")
|
||
print(solution.longestPalindrome("babad"), "= aba")
|
||
print(solution.longestPalindrome("cbbd"), "= bb")
|
||
print(solution.longestPalindrome("ccc"), "= ccc")
|