POJ - 3261 Milk Patterns 后缀数组 二分

题目链接: POJ - 3261 Milk Patterns

题目大意

一个数组, 求至少出现k次的最长子串

思路

求出高度数组, 二分子串长度求解
check时, 对高度数组中每个所有元素的大于等于mid的区间, 统计区间长度如果大于等于k, 就返回true, 如果都小于k, false

代码

#include <iostream>
#include <cstdio>
#include <cstring>
#include <algorithm>

using namespace std;

const int maxn = 2e4 + 100;
int a[maxn], n;
int rk[maxn], tmp[maxn], sa[maxn], k, lcp[maxn];

bool cmp_sa(int i, int j)
{
    if (rk[i] != rk[j]) return rk[i] < rk[j];
    return (i + k <= n ? rk[i + k] : -1) < (j + k <= n ? rk[j + k] : -1);
}
void construct_sa(int *s)
{
    for (int i = 0; i <= n; ++i)
    {
        sa[i] = i;
        rk[i] = s[i];
    }
    rk[n] = -1;
    for (k = 1; k <= n; k *= 2)
    {
        sort(sa, sa + n + 1, cmp_sa);

        tmp[sa[0]] = 0;
        for (int i = 1; i <= n; ++i)
            tmp[sa[i]] = tmp[sa[i - 1]] + (cmp_sa(sa[i - 1], sa[i]) ? 1 : 0);
        copy(tmp, tmp + n + 1, rk);
    }
}
void construct_lcp(int *s)
{
    for (int i = 0; i <= n; ++i) rk[sa[i]] = i;

    int h = 0;
    lcp[0] = 0;
    for (int i = 0; i < n; ++i)
    {
        int j = sa[rk[i] - 1];

        if (h > 0) --h;
        for (; j + h < n && i + h < n; ++h)
            if (s[j + h] != s[i + h]) break;
        lcp[rk[i] - 1] = h;
    }
}


bool cal(int mid, int k)
{
    int cnt = 1;
    for(int i=0; i<n; ++i)
    {
        if(lcp[i] >= mid) ++cnt;
        else cnt = 1;
        if(cnt >= k) return 1;
    } 
    return 0;
}

int main()
{
    int k;
    scanf("%d%d", &n, &k);
    for(int i=0; i<n; ++i) scanf("%d", a+i);
    construct_sa(a);
    construct_lcp(a);
    int l = 0, r = n, mid;
    while(l < r)
    {
        mid = (l+r+1)/2;
        if(cal(mid, k)) l = mid;
        else r = mid-1;
    }
    printf("%d\n", l);

    return 0;
}