leetcode126. Word Ladder II-CSDN博客

本文链接：https://blog.csdn.net/qq_30339595/article/details/103179402

本文探讨了如何使用宽度优先搜索(BFS)算法解决单词阶梯问题，通过逐个替换字母从初始单词到达目标单词，同时介绍了算法的优化技巧，如使用集合作为队列，记录每个层级的独特单词，以及利用ascii_lowercase快速查找可能的下一步。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

you are offered a list of words and an initial word and a destination word.your task is to change a letter at a time and reach the end word.
when I see this question,BFS suddenly bump upon me.but there are two tricks on BFS.first, you need not to record all paths as states,but to use a set as queue,which contains unique words at this level.second,when find the next word, use the pre for the next word to record its previous words.
the advantages of this way as following:
1.if a word is visited before,it should not be visited this level
2.words at the same level should only be calculated once.
the trick for finding the next word:
1.use string.ascii_lowercase to replace a letter in the word.
2.it helps cut down the time complexity for this process.
the contain time complexity in set is O(1), but if compare every word with another,it calls O(N*m)

from collections import defaultdict
import string
class Solution:
    def findLadders(self, beginWord: str, endWord: str, wordList: List[str]) -> List[List[str]]:
        self.pre=defaultdict(list)
        q={beginWord}
        self.pre[beginWord]=[[beginWord]]
        def caldiff(a,wl):
            ret=[]
            for i in range(len(a)):
                for c in string.ascii_lowercase:
                    w=a[:i]+c+a[i+1:]
                    if w in wl:
                        self.pre[w]+=[i+[w] for i in self.pre[a]]
                        ret.append(w)
            return ret
        
        
        while(len(q)):
            tmp=[]
            wordList=set(wordList)-q
            for i in q:
                tmp+=caldiff(i,wordList)
            q=set(tmp)
            if endWord in q:
                break
        return self.pre[endWord]

from collections import defaultdict
import string
class Solution:
    def findLadders(self, beginWord: str, endWord: str, wordList: List[str]) -> List[List[str]]:
        wordList=set(wordList)
        pw=defaultdict(list)
        for w in wordList:
            for i in range(len(w)):
                pw[w[:i]+'*'+w[i+1:]].append(w)
        q=[beginWord]
        pre=defaultdict(list)
        pre[beginWord]=[[beginWord]]
        visited={beginWord}
        while(len(q)):
            tmp=set()
            for w in q:
                for i in range(len(w)):
                    for u in pw[w[:i]+'*'+w[i+1:]]:
                        if u not in visited:
                            for j in pre[w]:
                                pre[u].append(j+[u])
                            tmp.add(u)
            q=tmp
            visited.update(tmp)
            if endWord in q:
                break
        return pre[endWord]