Aho-Corasick 自动机

来源：互联网发布：制衣打飞软件编辑：程序博客网时间：2024/05/18 00:03
参考：《算法竞赛入门经典－训练指南》
在模式匹配问题中，如果模板有多个，KMP 算法就不太适合了
1、UVALive 4670 Dominating Patterns

#include <iostream>#include <cstdio>#include <cstdlib>#include <cstring>#include <cmath>#include <algorithm>#include <map>#include <queue>#include <stack>#include <set>#include <bitset>#include <ctime>#include <cctype>using namespace std;typedef long long ll;typedef unsigned long long ull;typedef pair<int, int> Pair;const int INF = 0x7fffffff;const int maxn = 1e6 + 10;map<string, int> mp;struct AC_Automata {    int ch[maxn][26], fail[maxn], val[maxn], last[maxn];    int cnt[160];    int size;    void init() {        size = 1;        memset(ch[0], 0, sizeof(ch[0]));        memset(cnt, 0, sizeof(cnt));        mp.clear();    }    int idx(char c) { return c - 'a'; };    void insert(char* s, int _val);    void find(char* T);    void cnt_time(int j);    void get_fail(void);};char text[maxn], P[160][80];int N;AC_Automata ac;int main() {#ifdef __AiR_H    freopen("in.txt", "r", stdin);#endif // __AiR_H_    while (scanf("%d", &N) != EOF && N != 0) {        ac.init();        for (int i = 0; i < N; ++i) {            scanf("%s", P[i]);            ac.insert(P[i], i + 1);        }        ac.get_fail();        scanf("%s", text);        ac.find(text);        int Max = 0;        for (int i = 1; i <= N; ++i) {            Max = max(Max, ac.cnt[i]);        }        printf("%d\n", Max);        for (int i = 0; i < N; ++i) {            if (ac.cnt[mp[P[i]]] == Max) {                printf("%s\n", P[i]);            }        }    }    return 0;}void AC_Automata::get_fail(void) {    queue<int> q;    fail[0] = 0;    for (int c = 0; c < 26; ++c) {        int u = ch[0][c];        if (u) {            fail[u] = 0; q.push(u); last[u] = 0;        }    }    while (!q.empty()) {        int f = q.front(); q.pop();        for (int c = 0; c < 26; ++c) {            int u = ch[f][c];            if (!u) {                ch[f][c] = ch[fail[f]][c];                continue;            }            q.push(u);            int v = fail[f];            fail[u] = ch[v][c];            last[u] = val[fail[u]] ? fail[u] : last[fail[u]];        }    }}void AC_Automata::find(char* T) {    int len = strlen(T);    int j = 0;    for (int i = 0; i < len; ++i) {        int c = idx(T[i]);        while (j && !ch[j][c]) {            j = fail[j];        }        j = ch[j][c];        if (val[j]) {            cnt_time(j);        } else if (last[j]) {            cnt_time(last[j]);        }    }}void AC_Automata::cnt_time(int j) {    if (j) {        ++cnt[val[j]];        cnt_time(last[j]);    }}void AC_Automata::insert(char* s, int _val) {    int u = 0, len = strlen(s);    for (int i = 0; i < len; ++i) {        int c = idx(s[i]);        if (!ch[u][c]) {            memset(ch[size], 0, sizeof(ch[size]));            val[size] = 0, ch[u][c] = size++;        }        u = ch[u][c];    }    val[u] = _val; mp[s] = _val;}
0 0