hdoj 3065 病毒侵袭持续中(AC自动机)

题目链接: http://acm.hdu.edu.cn/showproblem.php?pid=3065

思路分析:问题需要模式匹配多个模式串,需要注意的是模式串会包含和重叠,需要对AC自动机的匹配过程进行修改,对于每个节点,需要从该节点的失败指针回溯,

如果失败指针回溯后的节点为某个模式串的最后一个节点,则匹配了另一个模式串;

代码如下:

#include <queue>
#include <cstdio>
#include <cstring>
#include <iostream>
#include <algorithm>
using namespace std;

const int KIND = 128;
const int MAX_NODE = 1020 * 50;
const int MAX_N = 1000 + 10;
const int MAX_M = 2000000 + 100;
char str[MAX_M];
int visited[MAX_N];
int vir_match[MAX_N];
char vir[MAX_N][60];

struct Trie {
    int root, count;
    int next[MAX_NODE][KIND], fail[MAX_NODE], end[MAX_NODE];
    void Init()
    {
        count = 0;
        root = NewNode();
    }
    int NewNode()
    {
        for (int i = 0; i < KIND; ++i)
            next[count][i] = -1;
        end[count] = -1;
        return count++;
    }

    void Insert(char *str, int id)
    {
        int i = 0, k = 0;
        int now = root;

        while (str[i])
        {
            k = str[i];
            if (next[now][k] == -1)
                next[now][k] = NewNode();
            now = next[now][k];
            ++i;
        }
        end[now] = id;
    }

    void BuildAutomaton()
    {
        queue<int> Q;

        fail[root] = -1;
        Q.push(root);
        while (!Q.empty())
        {
            int now = Q.front();
            int p = -1;
            Q.pop();

            for (int i = 0; i < KIND; ++i)
            {
                if (next[now][i] != -1)
                {
                    if (now == root)
                        fail[next[now][i]] = root;
                    else
                    {
                        p = fail[now];
                        while (p != -1)
                        {
                            if (next[p][i] != -1)
                            {
                                fail[next[now][i]] = next[p][i];
                                break;
                            }
                            p = fail[p];
                        }
                        if (p == -1)
                            fail[next[now][i]] = root;
                    }
                    Q.push(next[now][i]);
                }
            }
        }
    }

    int Match(char *str)
    {
        int i = 0, k = 0, vir_count = 0;
        int p = root;

        while (str[i])
        {
            k = str[i];
            while (next[p][k] == -1 && p != root)
                p = fail[p];
            p = next[p][k];
            p = (p == -1) ? root : p;

            int temp = p;
            while (temp != root)
            {
                if (end[temp] != -1)
                {
                    if (visited[end[p]] == 0)
                        vir_match[vir_count++] = end[p];
                    visited[end[p]]++;
                }
                temp = fail[temp];
            }
            ++i;
        }
        return vir_count;
    }
};
Trie root;

int main()
{
    int vir_num = 0;
    int match_count = 0;

    while (scanf("%d
", &vir_num) != EOF)
    {
        root.Init();
        memset(vir_match, 0, sizeof(vir_match));
        memset(visited, 0, sizeof(visited));
        for (int i = 0; i < vir_num; ++i)
        {
            gets(str);
            strcpy(vir[i], str);
            root.Insert(str, i + 1);
        }

        match_count = 0;
        root.BuildAutomaton();
        gets(str);
        int ans = root.Match(str);
        sort(vir_match, vir_match + ans);
        if (ans)
        {
            for (int j = 0; j < ans - 1; ++j)
                printf("%s: %d
", vir[vir_match[j] - 1], visited[vir_match[j]]);
            printf("%s: %d
", vir[vir_match[ans - 1] - 1],
                visited[vir_match[ans - 1]]);
        }
    }
    return 0;
}
原文地址:https://www.cnblogs.com/tallisHe/p/4666084.html