B. Segment Occurrences

You are given two strings s and t

, both consisting only of lowercase Latin letters.

The substring s[l..r]

is the string which is obtained by taking characters sl,sl+1,…,sr

without changing the order.

Each of the occurrences of string a

in a string b is a position i (1≤i≤|b|−|a|+1) such that b[i..i+|a|−1]=a (|a| is the length of string a

).

You are asked q

queries: for the i-th query you are required to calculate the number of occurrences of string t in a substring s[li..ri]

.

Input

The first line contains three integer numbers n

, m and q (1≤n,m≤103, 1≤q≤105) — the length of string s, the length of string t

and the number of queries, respectively.

The second line is a string s

(|s|=n

), consisting only of lowercase Latin letters.

The third line is a string t

(|t|=m

), consisting only of lowercase Latin letters.

Each of the next q

lines contains two integer numbers li and ri (1≤li≤ri≤n) — the arguments for the i

-th query.

Output

Print q

lines — the i-th line should contain the answer to the i-th query, that is the number of occurrences of string t in a substring s[li..ri]

.

Examples

Input

Copy

10 3 4
codeforces
for
1 3
3 10
5 6
5 7

Output

Copy

0
1
0
1

Input

Copy

15 2 3
abacabadabacaba
ba
1 15
3 4
2 14

Output

Copy

4
0
3

Input

Copy

3 5 2
aaa
baaab
1 3
1 1

Output

Copy

0
0

Note

In the first example the queries are substrings: "cod", "deforces", "fo" and "for", respectively.

#include<bits/stdc++.h>
#define lowbit(x) (x&(-x))

#define rep(i,a,b) for(int i=a;i<b;i++)
using namespace std;
const int maxn=1e3+10;

char str[maxn],s[maxn];

struct node{
    char ch;
    int val;
    node(){}
    node(char _ch,int _val):ch(_ch),val(_val){}
};
node stk[maxn];
int top;
int f[maxn];

int tr[maxn];

int len1,len2;
void update(int x,int v){
//	printf("x:%d\n",x);
	while(x<=len1){
		tr[x]+=v;
		x+=lowbit(x);
	}
	//rep(i,1,len2+1)printf("i:%d %d\n",i,tr[i]);
}

int get_sum(int x){
	int ans=0;
	while(x>0){
		ans+=tr[x];
		x-=lowbit(x);
	}
	return ans;
}

/*
自己匹配自己的时候，谁是模版串？
[0,i-1]
*/
int fail[maxn];
void getFail(char* P,int len){
    fail[0]=fail[1]=0;
    for(int i=1;i<len;i++){
        //目的就是找以i号字符结尾的串 的后缀和模版串的前缀的最大长度
        int j=fail[i];//因为“模版串"没有i号字符，所以i肯定失配。fail[i]的含义就是以i-1号字符结尾的串中 1.模版的最大的前缀和文本后缀的匹配成功的长度，2.也就是状态机现在的下标 ，是正在匹配的那个模版的J号字符(从0开始）
        while(j&&P[i]!=P[j])j=fail[j];//没有对应的J；寻找和P[i]匹配的位置J
        fail[i+1]=(P[i]==P[j]?j+1:0);//如果找到了，长度就是J+1，没有找到长度就是0，
    }
}

int main()
{
	int m;
	scanf("%d %d %d",&len1,&len2,&m);
    scanf(" %s",str);
    scanf(" %s",s);

    getFail(s,len2);
    int j=0;
    for(int i=0;i<len1;i++){
    	while(j&&s[j]!=str[i])j=fail[j];
    	if(s[j]==str[i])++j;
		if(j==len2){
          int x=i-len2+1;
		  update(x+1,1);
		 // printf("i:%d %c\n",x+1,str[x]);
		}
	}
    rep(i,0,m){
    	int l,r;
    	scanf("%d %d",&l,&r);
    	//printf("l:%d r:%d\n",l,r);
    	if(r-l+1<len2){
    		printf("0\n");
		}
		else{
			printf("%d\n",get_sum(r-len2+1)-get_sum(l-1));
		}
	}
    return 0;
}