这是一道板子题的改编,意在加深对求第k小子串的理解。首先先看一下最简单的SAM板子。相信应该都写过了才会写这题
//
// Created by acer on 2021/2/16.
//
//判断子串,不同子串个数,所有子串字典序第i大,最长公共子串
#include <cstring>
#include <iostream>
#include "string"
#define mem(x, i) memset(x,i,sizeof(x))
using namespace std;
const int MAXN = 3e5 + 10;
int len[MAXN << 1];
int ch[MAXN << 1][27];
int fa[MAXN << 1];
int last = 1;
int tot = 1;
int p;
void add(int c) {
p = last;
last = ++tot;
int np = last;
len[np] = len[p] + 1;
for (; p && !(ch[p][c]); p = fa[p]) ch[p][c] = np;
if (!p) fa[np] = 1;
else {
int q = ch[p][c];
if (len[q] == len[p] + 1) fa[np] = q;
else {
int nq = ++tot;
memcpy(ch[nq], ch[q], sizeof(ch[nq]));
fa[nq] = fa[q];
len[nq] = len[p] + 1;
fa[np] = fa[q] = nq;
for (; p && ch[p][c] == q; p = fa[p]) {
ch[p][c] = nq;
}
}
}
}
char s[MAXN];
int count[MAXN<<1];
void dfs(int k)
{
if (count[k]) return;
count[k] = 1;
for (int i = 0; i <= 26; ++i) { //此处是实现算法的关键,通过从a遍历到z来计算字典序
if (!ch[k][i]) continue;
dfs(ch[k][i]);
count[k]+=count[ch[k][i]]; //很明显,count的意义就是字典序为k的子串的首字母
}
}
int main() {
scanf("%s", s + 1);
int l = strlen(s + 1);
for (int i = 1; i <= l; ++i) {
add(s[i] - 'a');
}
int T;
dfs(1);
scanf("%d",&T);
while (T--){
int k;
int now = 1;
scanf("%d",&k);
while (k)
{
if (!k) break;
for (int i = 0; i <= 26; ++i) {
if (ch[now][i]) {
if (count[ch[now][i]] >= k) { //因为不能往前找,所以找到的第一个大于k的就肯定是属于答案的子串中
putchar(i+'a');
--k;
now = ch[now][i];
break;
} else
k-=count[ch[now][i]]; //这个字母包含的不够多
}
}
}puts("");
}
}
这套代码对应的题目的一个子问题,即找本质不同子串的情况,那么对于加上相同字串的贡献时,我们可以维护一个数组记录每个节点的子串重复出现了几次,同时用这个数组取替换count数组的值就行了。
//
// Created by acer on 2021/2/16.
//
//判断子串,不同子串个数,所有子串字典序第i大,最长公共子串
#include <cstring>
#include <iostream>
#include "string"
#define mem(x, i) memset(x,i,sizeof(x))
using namespace std;
const int MAXN = 3e5 + 10;
int len[MAXN << 1];
int ch[MAXN << 1][27];
int fa[MAXN << 1];
int siz[MAXN << 1];
int last = 1;
int tot = 1;
int p;
void add(int c) {
p = last;
last = ++tot;
int np = last;
len[np] = len[p] + 1;
for (; p && !(ch[p][c]); p = fa[p]) ch[p][c] = np;
if (!p) fa[np] = 1;
else {
int q = ch[p][c];
if (len[q] == len[p] + 1) fa[np] = q;
else {
int nq = ++tot;
memcpy(ch[nq], ch[q], sizeof(ch[nq]));
fa[nq] = fa[q];
len[nq] = len[p] + 1;
fa[np] = fa[q] = nq;
for (; p && ch[p][c] == q; p = fa[p]) {
ch[p][c] = nq;
}
}
}
siz[last] =1;
}
int id[MAXN<<1];
int weig[MAXN << 1];
void getTuopu()
{
for (int k = 1; k <= tot; ++k) //给每个点赋权值
weig[len[k]] ++;
for (int m = 1; m <= tot; ++m) //获得长度大于等于m的点的总个数,所以必然是不会相同的,故可以用来标记。
weig[m] += weig[m-1];
for (int n = 1; n <= tot; ++n) //根据点出现顺序获得拓扑序
id[weig[len[n]]--] = n;
}
char s[MAXN];
int count[MAXN << 1];
int t;
void dfs(int k) {
if (count[k]) return;
if (!t)
count[k] = 1;
else
count[k] = siz[k];
for (int i = 0; i <= 26; ++i) { //此处是实现算法的关键,通过从a遍历到z来计算字典序
if (!ch[k][i]) continue;
dfs(ch[k][i]);
count[k] += count[ch[k][i]]; //很明显,count的意义就是字典序为k的子串的首字母
}
}
int main() {
scanf("%s", s + 1);
int l = strlen(s + 1);
for (int i = 1; i <= l; ++i) {
add(s[i] - 'a');
}
getTuopu();
for (int i = tot; i > 1; --i) {
siz[fa[id[i]]] += siz[id[i]];
}
int k;
int now = 1;
cin >> t >> k;
dfs(1);
while (k) {
if (!k) break;
for (int i = 0; i <= 26; ++i) {
if (ch[now][i]) {
// cout << ch[now][i] << ' ' << char(i + 'a') << endl;
if (count[ch[now][i]] >= k) { //因为不能往前找,所以找到的第一个大于k的就肯定是属于答案的子串中
putchar(i + 'a');
now = ch[now][i];
k -= (t==1?siz[now]:1);
break;
} else
k -= count[ch[now][i]]; //这个字母包含的不够多
}
}
}
puts("");
}