Let be the input array, and let us use -based indexing. Then the input values are stored in in[1], in[2], ..., in[n]. If we compute the product of all elements of the array in and then set out_i = p / in_i (where is the resulting array), this method fails when in_i = 0.
Let us consider a different approach. We construct two arrays:
• The prefix product array , where
and we define pref[0] = 1;
• The suffix product array , where
and we define suf[n + 1] = 1;
Then out_i = pref[i - 1] \cdot suf[i + 1].
Example
Algorithm implementation
Declare the arrays.
#define MAX 1000002
int in[MAX], pref[MAX], suf[MAX], res[MAX];Read input data.
scanf("%d",&n);
for(i = 1; i <= n; i++)
scanf("%d",&in[i]); // [1,2,3,4]Build the array of prefix products.
pref[0] = 1;
for(i = 1; i <= n; i++) // [1,2,6,24]
pref[i] = pref[i-1] * in[i];Build the array of suffix products.
suf[n+1] = 1;
for(i = n; i >= 1; i--) // [24,24,12,4]
suf[i] = suf[i+1] * in[i];Build the resulting array.
for(i = 1; i <= n; i++) // [24,12,8,6]
res[i] = pref[i-1] * suf[i+1];Print the answer.
for(i = 1; i <= n; i++)
printf("%d ",res[i]);
printf("\n");