I am trying to write a program find maximum element of each column in a matrix but getting errors as cannot convert 'int (*)[m]' to 'int (*)[100]'

Question

#include <bits/stdc++.h>
using namespace std;
const int MAX = 100;
//Function to calculate largest column
void largestInColumn(int mat[][MAX], int rows, int cols)
{
    for (int i = 0; i < cols; i++) {
// initialize the maximum element with 0
        int maxm = mat[0][i];
// Run the inner loop for rows
        for (int j = 1; j < rows; j++) {
// check if any element is greater than the maximum element of the column and replace it   

            if (mat[j][i] > maxm)
                maxm = mat[j][i];
        }
       
        cout << maxm << endl;
    }
}
// Driver code
int main()
{
    int n , m ;
   cin>>n>>m;
   int mat[n][m];
   for(int i=0;i<n;i++)
   {
   for(int j=0;j<m;j++)
   {
    cin>>mat[i][j];
   }    
   }
    largestInColumn(mat, n, m);
    return 0;
}

Answer 1

I will answer the question using valid C++ statements. There are no VLAs in C++. I will use a std::vector instead.

There is also no need to store any value of the matrix for this task. You can make the decision already during reading of the values.

#include <iostream>
#include <vector>
#include <limits>

int main() {

    // Read number of rows and columns
    if (size_t numberOfRows{}, numberOfColumns{}; std::cin >> numberOfRows >> numberOfColumns) {

        // Here we will store the max values per column, so, the result
        std::vector<int> maxColumnValue(numberOfColumns, std::numeric_limits<int>::lowest());

        // Read all rows and columns
        for (size_t row{}; row < numberOfRows; ++row)
            for (size_t col{}; col < numberOfColumns; ++col)

                // If the current value of the column is greater than the current max value, then use new value instead
                if (int value{ std::numeric_limits<int>::lowest() }; std::cin >> value)
                    if (value > maxColumnValue[col]) maxColumnValue[col] = value;

        // Show result to the user
        for (const int m : maxColumnValue) std::cout << m << '\n';
    }
    return 0;
}

Answer 2

The cause of the error is due to you trying to pass a variable-length-array to a function that requires a standard 2D array.

First, variable-length-arrays (VLA's) are not part of standard C++. Arrays in C++ require that the sizes of the array are known at compile-time, not runtime. So pretend they don't exist, because technically, they do not exist in standard C++.

Thus you have two choices:

1. Declare a non-variable-size 2D array and use that, or
2. Use a container that is built to have dynamic size, such as std::vector.

Since you did not specify how large n could be, then solution 2 is safer.

Given that, here is your code using std::vector:

#include <vector>
#include <iostream>

using Int1D = std::vector<int>;
using Int2D = std::vector<Int1D>;

//Function to calculate largest column
void largestInColumn(Int2D& mat, int rows, int cols)
{
    for (int i = 0; i < cols; i++) 
    {
        int maxm = mat[0][i];
        for (int j = 1; j < rows; j++) 
        {
            if (mat[j][i] > maxm)
                maxm = mat[j][i];
        }
        std::cout << maxm << std::endl;
    }
}

int main()
{
   int n , m ;
   std::cin >> n >> m;
   Int2D mat(n, Int1D(m));
   for(int i=0;i<n;i++)
   {
       for(int j=0;j<m;j++)
       {
         std::cin >> mat[i][j];
       }    
   }
   largestInColumn(mat, n, m);
}

Using the input:

3 3 
1 2 3
1 4 9
76 34 21

The output is:

76
34
21