HackerRank - gcd-matrix

Alex has two arrays defined as  and . He created an  matrix, , where  for each  in . Recall that is the greatest common divisor of  and .

For example, if  and , he builds  like so:

Alex's friend Kiara loves matrices, so he gives her  questions about matrix where each question is in the form of some submatrix of  with its upper-left corner at  and its bottom-right corner at . For each question, find and print the number of distinct integers in the given submatrix on a new line.

Input Format

The first line contains three space-separated integers describing the respective values of  (the size of array ),  (the size of array ), and  (Alex's number of questions). 
The second line contains  space-separated integers describing . 
The third line contains  space-separated integers describing . 
Each line  of the  subsequent lines contains four space-separated integers describing the respective values of , , , and  for the  question (i.e., defining a submatrix with upper-left corner  and bottom-right corner ).

Constraints

Scoring

•  for  of score.
•  for  of score.

Output Format

For each of Alex's questions, print the number of distinct integers in the given submatrix on a new line.

Sample Input 0

3 3 3
1 2 3
2 4 6
0 0 1 1
0 0 2 2
1 1 2 2

Sample Output 0

2
3
3

Explanation 0

Given  and , we build the following :

The diagram below depicts the submatrices for each of the  questions in green:

1. For the submatrix between  and , the set of integers is . The number of distinct integers is .
2. For the submatrix between  and , the set of integers is . The number of distinct integers is .
3. For the submatrix between  and , the set of integers is . The number of distinct integers is .

这个题直接去求gcd显然并不现实。

在这里学习的一个比较巧妙的方法就是倒着来，我们选定i之后，对于两个区间的数分别找出来以i为约数的所有的数的个数num[i]。

对于序列A和序列B分别记为numa[i]和numb[i[。

之后numa[i]*numb[i]表示的是什么呢，自然是以i为最大公约数或者以i的倍数为最大公约数的所有数对的和。

然后我们再减去以i的倍数为最大公约数的数对个数，最后剩下的就是自然是以i为最大公约数的数对的总数了。

这样复杂度就能降低很多了，使n变成一个logn。

方法确实巧妙，倒着来想，我自己想的话估计要想好久。

#include <iostream>
#include <cstdio>
#include <algorithm>
#include <cstring>
using namespace std;
int n,m,q;
const int MAXN=1e5+7;
const int mod=1e9+7;
int a[MAXN],b[MAXN];
long long  numa[MAXN],numb[MAXN];
long long ans[MAXN];
int main()
{
    int i,j;
    scanf("%d%d%d",&n,&m,&q);
    for(i=0;i<n;++i)scanf("%d",&a[i]);
    for(i=0;i<m;++i)scanf("%d",&b[i]);
    int l1,r1,l2,r2;
    while(q--)
    {
        memset(numa,0,sizeof(numa));
        memset(numb,0,sizeof(numb));
        scanf("%d%d%d%d",&l1,&r1,&l2,&r2);
        for(i=l1;i<=l2;++i)numa[a[i]]++;
        for(i=r1;i<=r2;++i)numb[b[i]]++;
        for(i=1;i<=100000;++i)
        {
            for(j=i*2;j<=100000;j+=i)
            {
                numa[i]+=numa[j];
                numb[i]+=numb[j];
            }
        }
        int cnt=0;
        for(i=100000;i>=1;--i)
        {
            ans[i]=numa[i]*numb[i];
            for(j=i*2;j<=100000;j+=i)
            {
                ans[i]-=ans[j];
            }
            if(ans[i])cnt++;
        }
        printf("%d\n",cnt);
    }
}