POJ 3415

Common Substrings

Time Limit: 5000MS		Memory Limit: 65536K
Total Submissions: 8358		Accepted: 2761

Description

A substring of a string T is defined as:

T( i,  k)= T_iT_{i +1}... T_{i+k -1}, 1≤ i≤ i+k-1≤| T|.

Given two strings A, B and one integer K, we define S, a set of triples (i, j, k):

S = {( i,  j,  k) |  k≥ K,  A( i,  k)= B( j,  k)}.

You are to give the value of |S| for specific A, B and K.

Input

The input file contains several blocks of data. For each block, the first line contains one integer K, followed by two lines containing strings A and B, respectively. The input file is ended by K=0.

1 ≤ |A|, |B| ≤ 10⁵
1 ≤ K ≤ min{|A|, |B|}
Characters of A and B are all Latin letters.

Output

For each case, output an integer |S|.

Sample Input

2
aababaa
abaabaa
1
xx
xx
0

Sample Output

22
5

Source

POJ Monthly--2007.10.06, wintokk

题意：给出两个字符串SA, SB和一个数K，求两个字符串的长度不小于k的公共子串个数。

思路：两个字符串中间用一个没出现过的字符连接，之后求出height数组，根据K将后缀分组，对每一组判断每个后缀属于SA串还是SB串，对每个SA串，求前面的SB串与它的LCP长度，同样，对每个SB串，求前面的SA串与它的LCP长度。对于两个后缀，设它们的LCP值为L，则这两个后缀能产生L-K+1个公共字串。而求每个SA串与前面的SB串能产生多少公共子串可以用单调栈O(len)的实现：栈中维护每个后缀的height值，以及从起点扫描到该位置时，有多少个后缀的height值大于等于当前后缀的height值，于是每次扫描到一个后缀，将其加入单调栈中(需要弹出小于等于它的元素)，计算该后缀的cnt值。

#include <iostream>
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <string>
#include <algorithm>
#include <map>
#include <stack>

using namespace std;

#define INF 110000
#define N 201000
#define M 201000

int k, n, m, len1, len2;
char s[N];
int str[N], rank[N], sa[N], bucket[N], trank[N], sa2[N], height[N];

class Node
{
public:
    int val, cnt;
    Node(int _val, int _cnt)
    {
        val = _val;
        cnt = _cnt;
    }
};
stack<Node> sta;


int cmp(int p1, int p2, int len)
{
    return p1+len < n && p2+len < n && trank[p1] == trank[p2] && trank[p1+len] == trank[p2+len];
}

void getSa()
{
    for(int i = 0; i < m; i++) bucket[i] = 0;
    for(int i = 0; i < n; i++) bucket[rank[i] = str[i]]++;
    for(int i = 1; i < m; i++) bucket[i] += bucket[i-1];
    for(int i = n-1; i >= 0; i--) sa[--bucket[rank[i]]] = i;
    for(int j = 1, p = 0; p < n; j <<= 1, m = p)
    {
        p = 0;
        for(int i = n-j; i < n; i++) sa2[p++] = i;
        for(int i = 0; i < n; i++)
            if(sa[i] >= j) sa2[p++] = sa[i]-j;
        for(int i = 0; i < m; i++) bucket[i] = 0;
        for(int i = 0; i < n; i++) bucket[rank[i]]++;
        for(int i = 1; i < m; i++) bucket[i] += bucket[i-1];
        for(int i = n-1; i >= 0; i--) sa[--bucket[rank[sa2[i]]]] = sa2[i];
        for(int i = 0; i < n; i++) trank[i] = rank[i];
        p = 0; rank[sa[0]] = p++;
        for(int i = 1; i < n; i++)
            rank[sa[i]] = cmp(sa[i], sa[i-1], j) ? p-1 : p++;
    }
}
void getHeight()
{
    int h = 0;
    height[0] = 0;
    for(int i = 0; i < n; i++)
    {
        if(rank[i] == 0) continue;
        if(h) h--;
        int pre = sa[rank[i]-1];
        for(; pre+h < n && i+h < n && str[i+h] == str[pre+h]; h++);
        height[rank[i]] = h;
    }
}

inline int get(int pos)
{
    if(sa[pos] == len1) return 0;
    return sa[pos] < len1 ? 1 : 2;
}

long long cal(int l, int r, int flag)
{
    while(!sta.empty()) sta.pop();
    long long res = 0, sum = 0;
    if(get(l) != flag)
    {
        sta.push(Node(INF, 1));
        sum += INF-k+1;
    }
    for(int i = l+1; i <= r; i++)
    {
        int cnt = 0;
        while(!sta.empty() && sta.top().val >= height[i])
        {
            sum -= (sta.top().val-k+1)*sta.top().cnt;
            cnt += sta.top().cnt;
            sum += (height[i]-k+1)*sta.top().cnt;
            sta.pop();
        }
        if(cnt)
            sta.push(Node(height[i], cnt));
        if(get(i) != flag)
        {
            sum += INF-k+1;
            sta.push(Node(INF, 1));
        }

        if(get(i) == flag)
        {
            res += sum;
        }
    }
    return res;
}
int main()
{
    //freopen("C:\\Users\\zfh\\Desktop\\in.txt", "r", stdin);
    while(scanf("%d", &k) != -1 && k)
    {

        scanf(" %s", s);
        n = 0; m = 300;
        len1 = strlen(s);
        for(int i = 0; i < len1; i++) str[n++] = s[i];
        str[n++] = 0;
        scanf(" %s", s);
        len2 = strlen(s);
        for(int i = 0; i < len2; i++) str[n++] = s[i];
        str[n] = 1;

        getSa();
        getHeight();
        int sp = 0, tp = 0;
        long long ans = 0;
        for(int i = 0; i < n; i++)
        {
            if(height[i] >= k)
            {
                tp = i;
            }
            else
            {
                ans += cal(sp, tp, 1);
                ans += cal(sp, tp, 2);
                sp = tp = i;
            }
        }
        ans += cal(sp, tp, 1);
        ans += cal(sp, tp, 2);
        printf("%I64d\n", ans);
    }
    return 0;
}