Project Euler Problem 11: Largest product in a grid

Here is my implementation to project Euler Problem 11. I did this problem a bit later, when I learned how to input from a txt file. The problem context can be seen here. I did add "\n" to the end of each line, after copying the numbers into a txt file, I'm not sure of any other way to do that. Any other improvements anyone can think of?

#include 
#include 
#include 
#include 
#include 

//splitting stream into ints.
std::vector split(std::string line){
    std::stringstream ss (line);
    std::vector result;
    std::string num;
    while(std::getline(ss, num, ' '))
        result.push_back(std::stoi(num));
    return result;
}
//comparing all int's that are horizontally next to each other
unsigned long long Horizontal(int a, int b, std::vector> grid){
    if(b > grid){
    if(a >grid){
    if (a >grid){
    b+=3; // b needs to be 3 larger for this function
    if(a > grid ){
    int gwidth = grid[0].size();
    int glength = grid.size();
    unsigned long long largest = 0;

    for(int a = 0; a > grid;

    //opening file.
    std::ifstream nums;
    nums.open("grid.txt");
    std::string row;

    //Calling function, and pushing back into the vector
    while(std::getline(nums, row, '\n')){
        grid.push_back(split(row));
    }

    std::cout << Largest(grid);

}

Your functions returning a unsigned long long are missing a return which leads to undefined behavior (more details on this question). You could throw an exception to handle this or you could just return 0.

Not a real issue and mostly a matter of personal preference but the way you check that you are not going out of the bounds of the array could be written in a more natural way.

To check that index b + 3 is in the array, I'd rather read b + 3 < a.size() than b < a.size() - 3.

Even more unusual is the pre-increment in the case of BackDiag() : you add 3 to b and then you consider b - 3.

Once your code re-written to take into account these comments, it looks like :

unsigned long long Horizontal(int a, int b, std::vector> grid){
    return (b + 3 > grid){
    return (a + 3 >grid){
    return (a + 3 >grid){
    return (a + 3 < grid.size() && b + 3 < grid[a].size()) ?
        (grid[a][b+3] * grid[a+1][b+2] * grid[a+2][b+1] * grid[a+3][b]) :
        0;
}

Now, one thing to notice is that this good is basically always the same. You should try to write a generic function that you can reuse.

The signature would be something like :

unsigned long long computeProduct(int a, int b, int incrA, int incrB, std::vector>grid)

StackExchange Code Review Q#47361, answer score: 5