fixed

For the record, OPs wrong solution is:

public class Solution{
  public static String diagonal(int[][] matrix){
    System.out.println(matrix[0][0]+" "+matrix[0][1]+" "+matrix[0][2]);
            System.out.println(matrix[1][0]+" "+matrix[1][1]+" "+matrix[1][2]);
            System.out.println(matrix[2][0]+" "+matrix[2][1]+" "+matrix[2][2]);
            System.out.println(matrix[0][0] + matrix[2][2] > matrix[0][2] + matrix[2][0] ? "Principal Diagonal win!" :
          matrix[0][0] + matrix[2][2] < matrix[0][2] + matrix[2][0] ? "Secondary Diagonal win!" : "Draw!");
    return matrix[0][0] + matrix[2][2] > matrix[0][2] + matrix[2][0] ? "Principal Diagonal win!" :
          matrix[0][0] + matrix[2][2] < matrix[0][2] + matrix[2][0] ? "Secondary Diagonal win!" : "Draw!";
  }
}

Example input it fails is:

matrix = [
  [1, 6, 1, 4, 9]
  [7, 2, 6, 2, 9]
  [1, 4, 8, 6, 5]
  [1, 6, 0, 1, 4]
  [1, 2, 5, 4, 1]
]

Your solution is shit broken, it can handle only matrices of size 3x3. It returns wrong result for matrices larger than 3x3.

Not a kata issue.

Shit ain't working

5 0 4
9 2 0
3 6 4
Principal Diagonal win!

right? according to the rules, it is. according to the random tests, secondary wins here though...

cool

PHP fork to fix this issue and enable PHP 8.

python new test framework is required. updated in this fork

Current state of C++ initial solution setup compiles correctly, closing!

OP solved it, closing

And you're starting with $arr[0], so in your loop you can skip first element by starting to count from $i = 1.

You don't have to check for equality here. Just checking if current number is less than min is enough.

Actual and expected are flipped in PHP

your code works for only one set of routes(

and apparently it is your solution that has bad time complexity.

Really? my solution is O(n^2) and passed in less than 1 second.