Second smallest number in array

[deividas717]

Hello, I have this txt file

12.5
17.69
458.2
4892.3
7.02

And I need to find second smallest number in array, how I need to do it?

double min=0.0, min2;
[code]for(int i=0; i<n; i++){
if(S[i]<min){
min=S[i];
}
}

for(int i=0; i<n; i++)[
if(min<S[i] && S[i]>min2){
min2=S[i];
}
}

[Bourgond Aries]

Find the smallest first, then find the smallest again, but this time with the condition that this new smallest must be larger than the previous smallest.

float arr[5] = {1.f, 4.f, 0.5f, 3.f, 0.3f};
float *f = std::min_element(arr, arr + 5, [](const float &a, const float &b) -> bool {return a < b;});
float *s = std::min_element(arr, arr + 5, [f](const float &a, const float &b) -> bool {return a < b && a > *f;});
std::cout << *s << std::endl;

[deividas717]

Ok, thanks, but this is too much compicated for me. I don't know what is float and so on, because I'm just beginner. It is possible to write this code without float and so on ->?

[abhishekm71]

Just change your initial values for min and min2 to very large values (larger than the largest possible value in the list).

Also, you need the following change.

for(int i=0; i<n; i++)[
if(min<S[i] && S[i]<min2){ // not S[i] > min2
min2=S[i];
}
}

[Bourgond Aries]

float is a half-precision double. Here is an implementation using more elementary functionality.

double arr[5] {1., 4., 0.5, 3., 0.3};
double smallest = arr[0], secondsmallest = arr[0];

for (double &x : arr)
{
    if (x < smallest)
        smallest = x;
}

for (double &x : arr)
{
    if (x < secondsmallest && x > smallest)
        secondsmallest = x;
}

std::cout << secondsmallest << std::endl;

Perhaps you should read http://www.cplusplus.com/doc/tutorial/variables/

[abhishekm71]

@Bourgond Aries

Nice code! One minor point.
I think the initial value for secondsmallest should be a very large number. If arr[0] is the smallest number in the array, then the value of secondsmallest will not change ever.

[Bourgond Aries]

@abhishekm71

Indeed, you are right. I stand corrected. I have implemented your advice.

double arr[6] {0.2, 1., 4., 0.5, 3., 0.3};
double smallest = arr[0], secondsmallest = std::numeric_limits<double>::max();

for (double &x : arr)
{
    if (x < smallest)
        smallest = x;
}

for (double &x : arr)
{
    if (x < secondsmallest && x > smallest)
        secondsmallest = x;
}

std::cout << secondsmallest << std::endl;