Problem: Given an array that only contains "1" and "0", find the longest subarray which contains equal number of "1" and "0".
Solution: With hash table, we can have a O(N) solution. The detail is as follow:
- First convert all "0" to "-1", then calculate c[i] = sum(a[0], ... , a[i]). It takes O(N) to calculate all the c[i].
- Then our task is to find a c[i] and a c[j] such that c[i] = c[j] and |j-i| is maximum. With a hash table, we can finish this job by doing a linear scan with a time complexity of O(N). // 用hashtable存储具有相同key的position值
- There is a special case you need to handle. When c[N-1] = 0 (assume N is the size of a), the longest subarray is just a itself.
Idea: For every sum value record its leftmost occrrence in the left array and rightmost occrnce in the right array.such that in sum value at ith index == sum value at jth index it means that the subarray from index i to j has = no. of 1's and 0's.
Code:
#include<stdio.h>
#include<conio.h>
int main()
{
int arr[100], rite[200],lft[200],len[200],sum[100], n,i,tmp, lnth=0;
scanf("%d",&n);
printf("%d\n",n);
scanf("%d",&tmp);
if(tmp==0)
sum[0] = -1;
else
sum[0] = tmp;
for(i=1;i<n;i++)
{
scanf("%d",&tmp);
arr[i] = tmp;
if(tmp==0)
sum[i] = sum[i-1] - 1;
else
sum[i] = sum[i-1] + 1;
if(sum[i]==0)
lnth = i;
// printf("%d\n",sum[i]);
}
//printf("sum = %d\n",sum[n-1]);
for(i=0;i<(2*n-1);i++)
{
lft[i] = 0;
rite[i] = 0;
}
for(i=0;i<n;i++)
{
if(sum[i]>0)
rite[sum[i]+n-1] = i;
else if(sum[i]<0)
rite[sum[i]+n] = i;
}
for(i=n-1;i>=0;i--)
{
if(sum[i]>0)
lft[sum[i]+n-1] = i;
else if(sum[i]<0)
lft[sum[i]+n] = i;
}
for(i=0;i<(2*n-1);i++)
{
if(lnth<(rite[i]-lft[i]))
lnth = (rite[i]-lft[i]);
}
printf("length = %d",lnth);
getch();
return 0;
}
Ref:
http://tristan-interview.blogspot.com/2012/01/longest-subarray-with-equal-1-and-0.html
http://www.thealgorithmist.com/showthread.php/146-Array-of-0-s-and-1-s-Find-max-subarray-of-0-s-1-s