finding the min and max element in vector using std::min_element, std::max_element

Question

I am new to C++, and I am trying to find the minimum and maximum elements of a std::vector , but both std::min_element() and std::max_element() are not working together. The given output is only the minimum value. In the output, only the minimum value is printed twice, instead of minimum first then maximum.

#include <iostream> 
#include <algorithm> 
#include <vector>
using namespace std;

bool comp1_(int a, int b) 
{ 
    return a > b; 
}

bool comp2(int a, int b) 
{ 
    return a < b; 
}

int main() 
{ 
    vector<int> myvector;
    vector<int>::iterator i1;
    vector<int>::iterator i2;
    int n, num;
    cin >> n;
    for(int i = 0; i < n; i++){
        cin >> num;
        myvector.push_back(num);
    }
    i2 = std::min_element(myvector.begin(), myvector.end(), comp2);
    cout << *i2 << " ";
    i1 = std::max_element(myvector.begin(), myvector.end(), comp1_); 
    cout << *i1; 
    return 0; 
} 

Answer 1

Your problem is simply that you use max_element incorrectly. It expects a comparison which returns ​true if the first argument is less than the second. You will need to use comp2_ in both cases. So in your program it should read

i1 = std::max_element(myvector.begin(), myvector.end(), comp2); 

The best practice to achieve what you want is using minmax_element like Jesper mentioned

#include <iostream> 
#include <algorithm> 
#include <vector>


bool compLess(int a, int b)
{
    return (a < b);
}

int main()
{
    using namespace std;
    vector<int> myvector;

    int n, num;
    cin >> n;
    for (int i = 0; i < n; i++) {
        cin >> num;
        myvector.push_back(num);
    }

    auto minmax = std::minmax_element(myvector.begin(), myvector.end(), compLess);
    cout << "min: " << *minmax.first << "\tmax:" << *minmax.second << "\n";
    return 0;
}

With this you do not have to iterate twice through the array.

Hint Do not use using namespace std; in global namespace. I would consider that to be bad practice.

Answer 2

i1 = std::max_element(myvector.begin(), myvector.end(), comp1_); 

is wrong

i1 = std::max_element(myvector.begin(), myvector.end(), comp2); 

would work.

By using comp1_ with max_element you are effectively asking for the minimum element because max_element finds the element x for which comp1_(x, y) is not true for any y.

You've effectively done a double negative, you've switched to asking for the maximum element but by also reversing the comparison function you've switched to asking for the minimum again.

What you should do when using min_element and max_element is pass a comparison function that means 'less than', then these functions will do what they say.

You might now realise that

i2 = std::min_element(myvector.begin(), myvector.end(), comp1_);

would get you the maximum element.

Answer 3

You need to use comp2 as well for the call of std::max_element. The comparison operator is for both min_element and max_element the same.

Alternatively you could also use i1 = std::min_element(myvector.begin(), myvector.end(), comp1_);

Check out also https://en.cppreference.com/w/cpp/algorithm/max_element .