212. 单词搜索 II

212. 单词搜索 II

题目描述

给定一个 m x n 二维字符网格 board 和一个单词(字符串)列表 words,找出所有同时在二维网格和字典中出现的单词。

单词必须按照字母顺序,通过 相邻的单元格 内的字母构成,其中“相邻”单元格是那些水平相邻或垂直相邻的单元格。同一个单元格内的字母在一个单词中不允许被重复使用。

示例 1:

输入:board = [["o","a","a","n"],["e","t","a","e"],["i","h","k","r"],["i","f","l","v"]], words = ["oath","pea","eat","rain"]
输出:["eat","oath"]

示例 2:

输入:board = [["a","b"],["c","d"]], words = ["abcb"]
输出:[]

提示:

  • m == board.length

  • n == board[i].length

  • 1 <= m, n <= 12

  • board[i][j] 是一个小写英文字母

  • 1 <= words.length <= 3 * 10^4

  • 1 <= words[i].length <= 10

  • words[i] 由小写英文字母组成

  • words 中的所有字符串互不相同


题解:
法一:

字符串哈希 + 深搜。

记录 words 中每个单词的哈希值,并且记录最大长度。

然后枚举 board 中的每个位置,从该位置开始搜索,搜索的字符串长度不超过上面记录的最大值。

时间复杂度: O ( n ∗ m ∗ 3 L ∗ L ) O(n * m * 3^L * L) O(nm3LL)

额外空间复杂度: O ( n ∗ m ) O(n * m) O(nm)

class Solution {
    using ull = unsigned long long;
    const static int P = 131;
    unordered_set<ull> hash;
    unordered_set<ull> record;
    int max_len{0};
    vector<string> ret;
    vector<vector<bool>> vis;
    string ans;
    vector<int> dx, dy;
    int n, m;
public:
    void dfs( int x, int y, ull now, int len, const vector<vector<char>>& board ) {
        if ( hash.find(now) != hash.end() ) {
            if ( record.find(now) == record.end() ) {
                ret.emplace_back( ans );
                record.insert( now );
            }
        }
        if ( len >= max_len ) return;
        vis[x][y] = true;
        for ( int k = 0; k < 4; ++k ) {
            int tx = x + dx[k];
            int ty = y + dy[k];
            if ( tx < 0 || tx >= n || ty < 0 || ty >= m || vis[tx][ty] )
                continue;
            ans.push_back( board[tx][ty] );
            vis[tx][ty] = true;
            dfs( tx, ty, now * P + board[tx][ty], len + 1, board );
            vis[tx][ty] = false;
            ans.pop_back();
        }
        vis[x][y] = false;
    }
    vector<string> findWords(vector<vector<char>>& board, vector<string>& words) {
        if ( !board.size() || !board[0].size() ) return {};
        for ( const string& word : words ) {
            ull ans = 0;
            for ( const char& it : word )
                ans = ans * P + it;
            hash.insert( ans );
            if( word.size() > max_len )
                max_len = word.size();
        }
        n = board.size(), m = board[0].size();
        dx = vector<int>{-1, 0, 1, 0};
        dy = vector<int>{0, 1, 0, -1};
        vis = vector<vector<bool>>(n, vector<bool>(m));
        for ( int i = 0; i < n; ++i ) {
            for ( int j = 0; j < m; ++j ) {
                ans = board[i][j];
                dfs( i, j, board[i][j], 1, board );
            }
        }
        return ret;
    }
};
/*
时间:448ms,击败:29.78%
内存:8.3MB,击败:84.94%
*/

当然也可以遍历 words 中的每个单词,判断 board 中是否存在 word ,参考 单词搜索


class Solution {
private:
    vector<string> ret;
    vector<int> dx;
    vector<int> dy;
    vector<vector<bool>> vis;
    int n, m;
public:

    bool dfs( int x, int y, int p, const string& word, vector<vector<char>>& board ) {
        if ( p >= word.size() ) return true;
        vis[x][y] = true;
        for ( int k = 0; k < 4; ++k ) {
            int tx = x + dx[k];
            int ty = y + dy[k];
            if ( tx < 0 || tx >= n || ty < 0 || ty >= m || vis[tx][ty] )
                continue;
            if ( board[tx][ty] != word[p] ) continue;
            if ( dfs( tx, ty, p + 1, word, board ) ) return true;
        }
        vis[x][y] = false;
        return false;
    }

    vector<string> findWords(vector<vector<char>>& board, vector<string>& words) {
        if ( !board.size() || !board[0].size() ) return {};
        n = board.size(), m = board[0].size();
        dx = vector<int>{-1, 0, 1, 0};
        dy = vector<int>{0, 1, 0, -1};
        bool flag;
        for ( const string& word : words ) {
            vis = vector<vector<bool>>(n, vector<bool>(m, false));
            flag = false;
            for ( int i = 0; i < n; ++i ) {
                for ( int j = 0; j < m; ++j ) {
                    if ( board[i][j] != word[0] )
                        continue;
                    flag = dfs( i, j, 1, word, board );
                    if( flag ) {
                        ret.emplace_back( move(word) );
                        break;
                    }
                }
                if ( flag ) break;
            }
        }
        return ret;
    }
};
时间:8ms,击败:70.29%
内存:8.5MB,击败:55.77%
*/

速度出奇的快。。。

法二:

法一 中,我们是挨个判断相邻字符,在长度达到最大值时返回(剪枝),而如果我们只考虑与相邻字符构成的字符串在 words 中单词的前缀中,这样在搜索中就可以避免很多不必要的过程。

而判断前缀,就要用 trie树 了。

trie树 中,每个单词结束位置放置一个它在 words 中的下标(初值为 -1),在搜索过程中,如果遇到下标不为 -1 的节点,说明找到一个单词,直接保存结果,并将节点的下标置为 -1 ,防止结果重复。

时间复杂度: O ( n ∗ m ∗ 3 L ) O(n * m * 3^L) O(nm3L)

额外空间复杂度: O ( s ) O(s) O(s)strie树 中字母总数。

class TrieNode {
public:
    TrieNode *son[26]{nullptr};
    int id{-1};
public:
    TrieNode() { }
    void insert( const string& word, int id ) {
        auto root = this;
        int u;
        for ( const char& ch : word ) {
            u = ch - 'a';
            if( !root->son[u] )
                root->son[u] = new TrieNode();
            root = root->son[u];
        }
        root->id = id;
    }
};

class Solution {
private:
    TrieNode *root{nullptr};
    vector<vector<bool>> vis;
    vector<int> dx;
    vector<int> dy;
    vector<string> ret;
    int n, m;
public:
    vector<string> findWords(vector<vector<char>>& board, vector<string>& words) {
        if ( !board.size() || !board[0].size() ) return {};
        root = new TrieNode();
        for( int i = 0; i < words.size(); ++i )
            root->insert( words[i], i );
        n = board.size(), m = board[0].size();
        vis = vector<vector<bool>>( n, vector<bool>(m) );
        dx = vector<int>{-1, 0, 1, 0};
        dy = vector<int>{0, 1, 0, -1};
        int u;
        for ( int i = 0; i < n; ++i ) {
            for ( int j = 0; j < m; ++j ) {
                u = board[i][j] - 'a';
                if ( !root->son[u] )
                    continue;
                dfs( root->son[u], i, j, board, words );
            }
        }
        return ret;
    }

    void dfs( TrieNode* root, int x, int y, vector<vector<char>>& board, const vector<string>& words ) {
        if ( root->id != -1 ) {
            ret.emplace_back( move(words[root->id]) );
            root->id = -1;
        }
        vis[x][y] = true;
        int u;
        for ( int k = 0; k < 4; ++k ) {
            int tx = x + dx[k];
            int ty = y + dy[k];
            if ( tx < 0 || tx >= n || ty < 0 || ty >= m || vis[tx][ty] )
                continue;
            u = board[tx][ty] - 'a';
            if ( !root->son[u] ) continue;
            dfs( root->son[u], tx, ty, board, words );
        }
        vis[x][y] = false;
    }
};
/*
时间:348ms,击败:39.13%
内存:8.5MB,击败:63.62%
*/

优化:对于每个叶节点,其肯定是一个单词的结束节点,如果第一次遍历到一个叶节点,可以删除,避免重复遍历。

class TrieNode {
public:
    TrieNode *son[26]{nullptr};
    int id{-1};
    int sons{0};
public:
    TrieNode() { }
    void insert( const string& word, int id ) {
        auto root = this;
        for ( const char& ch : word ) {
            int u = ch - 'a';
            if( !root->son[u] ) {
                root->son[u] = new TrieNode();
                ++root->sons;
            }
            root = root->son[u];
        }
        root->id = id;
    }
};

class Solution {
private:
    TrieNode *root{nullptr};
    vector<int> dx;
    vector<int> dy;
    vector<string> ret;
    int n, m;
public:
    vector<string> findWords(vector<vector<char>>& board, vector<string>& words) {
        if ( !board.size() || !board[0].size() ) return {};
        root = new TrieNode();
        for( int i = 0; i < words.size(); ++i )
            root->insert( words[i], i );
        n = board.size(), m = board[0].size();
        dx = vector<int>{-1, 0, 1, 0};
        dy = vector<int>{0, 1, 0, -1};
        for ( int i = 0; i < n; ++i ) {
            for ( int j = 0; j < m; ++j ) {
                int u = board[i][j] - 'a';
                if ( !root->son[u] ) continue;
                dfs( root, i, j, board, words );
            }
        }
        return ret;
    }

    void dfs( TrieNode* root, int x, int y, vector<vector<char>>& board, const vector<string>& words ) {
        int u = board[x][y] - 'a';
        auto new_node = root->son[u];
        if ( new_node->id != -1 ) {
            ret.emplace_back( move(words[new_node->id]) );
            new_node->id = -1;
        }
        char tmp = board[x][y];
        board[x][y] = '#';
        for ( int k = 0; k < 4; ++k ) {
            int tx = x + dx[k];
            int ty = y + dy[k];
            if ( tx < 0 || tx >= n || ty < 0 || ty >= m || board[tx][ty] == '#' )
                continue;
            u = board[tx][ty] - 'a';
            if ( !new_node->son[u] ) continue;
            dfs( new_node, tx, ty, board, words );
        }
        board[x][y] = tmp;
        if ( !new_node->sons ) {
            --root->sons;
            delete new_node;
            u = board[x][y] - 'a';
            root->son[u] = nullptr;
        }
    }
};
/*
时间:0ms,击败:100.00%
内存:8.5MB,击败:63.62%
*/
  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值