[LeetCode][Java] Word Ladder II

题目：

Given two words (start and end), and a dictionary, find all shortest transformation sequence(s) from start to end, such that:

1. Only one letter can be changed at a time
2. Each intermediate word must exist in the dictionary

For example,

Given:
start = "hit"
end = "cog"
dict = ["hot","dot","dog","lot","log"]

Return

  [
    ["hit","hot","dot","dog","cog"],
    ["hit","hot","lot","log","cog"]
  ]

Note:

• All words have the same length.
• All words contain only lowercase alphabetic characters.

题意：

与题目《Word Ladder》不同的地方是，这里不是求最短变换序列的长度，而是返回所有的满足条件的最短序列。

算法分析：

参考了博客http://blog.csdn.net/worldwindjp/article/details/19301355

先BFS生成找到end时的生成树，标记出每个单词所在的层数。然后从目标用DFS往回找，过了大数据

AC代码：

<span style="font-family:Microsoft YaHei;font-size:10px;">public class Solution
{
       //记录每个单词所在的层数
    HashMap<String,Integer> path = new HashMap<String,Integer>();
    //bfs生成path
    void bfs(String start, String end, HashSet<String> dict) 
    {
        Queue queue = new LinkedList<String>();
        queue.add(start);
        path.put(start,0);
        String current;
        while(!queue.isEmpty())
        {
            current = (String)queue.poll();
            if(current==end) 
            {
                continue;
            }
            for(int i=0;i<current.length();i++) 
            {
                char[] strCharArr = current.toCharArray();
                for(char ch='a';ch<='z';ch++) 
                {
                    if(strCharArr[i]==ch) 
                    {
                        continue;
                    }
                    strCharArr[i] = ch;
                    String newWord = new String(strCharArr);
                    if(newWord.equals(end)==true||dict.contains(newWord)) 
                    {
                        //每个单词在path中只能出现一次，也就是每个单词只能出现在一层中，这样就很巧妙的解决了环的问题。
            			if(path.get(newWord)==null) 
            			{
                            int depth = (int)path.get(current);
                            path.put(newWord,depth + 1);
                            queue.add(newWord);
                        }
                    }
                }
            }
        }
    }
    //从目标单词往回找开始单词，记录所有路径
    void dfs(String start, String end, HashSet<String> dict, ArrayList<String> pathArray,ArrayList<ArrayList<String>> result) 
    {
        //找到了，需要reverse加入的所有单词
    	if(start.equals(end)==true) 
    	{
            pathArray.add(start);
            Collections.reverse(pathArray);
            result.add(pathArray);
            return;
        }
        if(path.get(start)==null)
        {
            return;
        }
        pathArray.add(start);
        int nextDepth = (int)path.get(start) - 1;
        for(int i=0;i<start.length();i++) 
        {
            char[] strCharArr = start.toCharArray();
            for(char ch='a';ch<='z';ch++) 
            {
                if(strCharArr[i]==ch)
                {
                    continue;
                }
                strCharArr[i] = ch;
                String newWord = new String(strCharArr);
		        //只相差一个字母同时这个单词所在的层数也是当前单词的上一层
                if(path.get(newWord)!=null&&(path.get(newWord)==nextDepth))
                {
                    ArrayList<String> newPathArray = new ArrayList<String>(pathArray);
                    dfs(newWord,end,dict,newPathArray,result);
                }
            }
        }
    }
    
    public ArrayList<ArrayList<String>> findLadders(String start, String end, HashSet<String> dict)
    {
        ArrayList<ArrayList<String>> result = new ArrayList<ArrayList<String>>();
        ArrayList<String> path = new ArrayList<String>();
        if(start==null||end==null||start.length()!=end.length()) 
        {
            return result;
        }
        bfs(start, end, dict);
        dfs(end,start, dict, path, result);
        return result;
    }
}</span>