https://leetcode.com/problems/word-search-ii/description/
这一题,如果不懂原理想必会很麻烦,但是如果知道用trie tree做,就不值得一个hard。
这一题就是很纯粹的trie tree + dfs,为了避免dfs过度的延伸,我们在这里利用到了trie tree的截枝功能。如果这里利用HashSet<String>作为节点,那么你只能判断什么时候成功,但是你势必每个dfs要走完全图,不会知道什么时候是一个false终止的信号。这样的超级大指数级别的performance肯定不是我们想要的。这个就是trie tree能够给你的,一个个字符和子节点的匹配,可以有很有效率的进行dfs遍历的终结。还需要说什么的话就是大部分类似题目的inplace 记录走过的节点啊,之类的。。直接给出代码吧
class TrieNode {
boolean isWord;
TrieNode[] children;
public TrieNode() {
this.isWord = false;
this.children = new TrieNode[26];
}
}
class TrieTree {
TrieNode root;
public TrieTree() {
root = new TrieNode();
}
public void addWord(String word) {
TrieNode current = root;
for (int i = 0; i < word.length(); i++) {
int index = word.charAt(i) - 'a';
if (current.children[index] == null) {
current.children[index] = new TrieNode();
}
current = current.children[index];
}
current.isWord = true;
}
}
public List<String> findWords(char[][] board, String[] words) {
TrieTree dict = new TrieTree();
for (String word : words) dict.addWord(word);
Set<String> resSet = new HashSet<String>();
for (int i = 0; i < board.length; i++) {
for (int j = 0; j < board[0].length; j++) {
Set<String> curResSet = _DFSFindWords(board, i, j, new StringBuilder(), dict.root);
if (curResSet != null) resSet.addAll(curResSet);
}
}
return new LinkedList<String>(resSet);
}
static int[][] DIR = {{-1, 0}, {1, 0}, {0, -1}, {0, 1}};
private Set<String> _DFSFindWords(char[][] board, int x, int y, StringBuilder sb, TrieNode curNode) {
if (x < 0 || x >= board.length || y < 0 || y >= board[0].length || board[x][y] == '-') {
return null;
} else {
char c = board[x][y];
int index = c - 'a';
if (curNode.children[index] == null) {
return null;
} else {
Set<String> curRes = new HashSet<String>();
sb.append(c);
if (curNode.children[index].isWord) {
curRes.add(sb.toString());
}
board[x][y] = '-';
for (int[] dir : DIR) {
Set<String> nextRes = _DFSFindWords(board, x + dir[0], y + dir[1], sb, curNode.children[index]);
if (nextRes != null) {
curRes.addAll(nextRes);
}
}
board[x][y] = c;
sb.deleteCharAt(sb.length() - 1);
return curRes;
}
}
}
871

被折叠的 条评论
为什么被折叠?



