# Compile error with vector sum

• Hi

I am trying to implement the below logic to find the largest sum of vector {5, 9, 7, 11} is sum of elements 9 and 11, which is 20.
Can anyone tell me what is wrong in my code below? Compilation gets terminated here.

 cpp
#include <stdexcept>
#include <iostream>
#include <vector>

class MaxSum
{
public:
static int findMaxSum(const std::vector<int>& numbers)
{ int curr = numbers[0];
int maxsum = curr;
for(int i = 1; i<numbers.size(); i++)
{
if(curr > 0)
{
curr = curr + numbers[i];
}
else
{
curr = numbers[i];
}
}
maxsum = std::max(maxsum, curr);
return maxsum;
}

 throw std::logic_error("Waiting to be implemented");
};


#ifndef RunTests
int main()
{
std:: vector<int> v {5, 9, 7, 11};
std::cout << MaxSum::findMaxSum(v);
}
#endif



• @tampere2021 In findMaxSum function, I am trying to implement logic to return largest sum of any 2 elements in given vector of numbers.

• Whatever you do, don't do any of the following:

• post the error message the compiler is giving you

Never, ever do one of those things.
</sarcasm>

But since you tried, a free tip: you have to use three , two are not enough.

• @hustbaer Compilation gets terminated and I can see the below error.
34 | throw std::logic_error("Waiting to be implemented");
| ^~~~~

• @tampere2021
Et näytä välittävän lähdekoodisi muotoilusta.

• std:: vector<int> v {5, 9, 7, 11};

Easiest way is using a partial sort:

const std::vector<int> v {5, 9, 7, 11};
std::array<int, 2> largest_values{};
std::partial_sort_copy(
std::begin(v), std::end(v),
std::begin(largest_values), std::end(largest_values),
std::greater<int>()
);
cout << std::accumulate(std::begin(largest_values), std::end(largest_values), 0) << "\n";


Edit: ginge auch mit nth_element, aber nur inplace.

• @wob I have changed the logic now.


#include <iostream>
using namespace std;

int main() {

int arr [4] = {5, 9, 7, 11};
int max = 0;
int len = sizeof(arr)/sizeof(arr[0]);
for (int i=0; i<len-2; i++){
for (int j=i+2; j<len; j++){
if ((arr[i]+arr[j]) > max)
max = arr[i]+arr[j];
}
}

cout << max;

return 0;
}



• @tampere2021 I think the below logic restricts me, so can I use just curr = curr + numbers[i]; But again I am not sure if this logic would work for me. Is there any other solution? As mentioned here, do I need to use your logic by replacing the line std:: vector<int> v {5, 9, 7, 11}; and use partial sort approach?

if(curr > 0)
{
curr = curr + numbers[i];
}
else
{
curr = numbers[i];
}


•   for (int i=0; i<len-2; i++){
for (int j=i+2; j<len; j++){
`

Why -2/+2 ?

With 4 elements there should be 6 $[= \sum_{i=1}^{(4-1)} i]$ comparisons inside the loops.