后缀数组(板子)

Posted 2021-02-08 maomao9173

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了后缀数组(板子)相关的知识，希望对你有一定的参考价值。

贴一下防止忘掉。

#include <bits/stdc++.h>
#define N 1000010
using namespace std;

char s[N];
int n, m, sa[N], tp[N], rk[N], _rk[N], bin[N];

void base_sort () {
    for (int i = 1; i <= m; ++i) bin[i] = 0;
    for (int i = 1; i <= n; ++i) bin[rk[tp[i]]]++; //每个出现的第一关键字++
    for (int i = 1; i <= m; ++i) bin[i] += bin[i - 1]; //前缀和 -> rk为i的后缀排名是多少
    for (int i = n; i >= 1; --i) sa[bin[rk[tp[i]]]--] = tp[i]; // 先记大的，然后名次顺位递减
}

void suffix_sort () {
    m = 255;
    for (int i = 1; i <= n; ++i) {
        rk[i] = s[i - 1], tp[i] = i;
    }
    base_sort ();
    for (int w = 1; w <= n; w <<= 1) {
        int cnt = 0;
        for (int i = n - w + 1; i <= n; ++i) {
            tp[++cnt] = i; //莫得第二关键字
        }
        for (int i = 1; i <= n; ++i) {
            if (sa[i] > w) { //如果有第二关键字
                tp[++cnt] = sa[i] - w; //把i当做第二关键字の后缀
            }
        }
        //printf ("sa : "); for (int i = 1; i <= n; ++i) printf ("%d ", sa[i]); printf ("
");
        //printf ("rk : "); for (int i = 1; i <= n; ++i) printf ("%d ", rk[i]); printf ("
");
        base_sort (); cnt = 1;
        //printf ("sa : "); for (int i = 1; i <= n; ++i) printf ("%d ", sa[i]); printf ("
");
        //printf ("rk : "); for (int i = 1; i <= n; ++i) printf ("%d ", rk[i]); printf ("
");
        memcpy (_rk, rk, sizeof (rk)); //把rk临时放一边
        rk [sa[1]] = 1;
        for (int i = 2; i <= n; ++i) {
            rk[sa[i]] = _rk[sa[i]] == _rk[sa[i - 1]] && _rk[sa[i] + w] == _rk[sa[i - 1] + w] ? cnt : ++cnt;
            //验证后缀sa[i]是不是和sa[i - 1]相等
        }
        if (cnt == n) break; //finished
        m = cnt; //m的可用范围
    }
}

int main () {
    scanf ("%s", s);
    n = strlen (s);
    printf ("n = %d
", n);
    suffix_sort ();
    for (int i = 1; i <= n; ++i) {
        printf ("%d ", sa[i]);
    }
    return 0;
}