I suggest to do a rewrite on the random tests' logic to include tests below with optimal distribution of probability:

• 0 x

• More than 1 x

Also, do add assertion messages for random tests too, make sure they are readable and well-formatted

Are you need approve this code? Or me need rewrite?

i think i'm to blame for the confusion, a week ago or so i added if:c and if:javascript markdown blocks to the description because C returns a struct and returns {-1,-1} in case of missing/multiple 'x's. i just edited the description again to replace the if:javascript by if-not:c now that there are more than 2 languages

Wrote an answer.

1. I didn't really understand what it was about. I relied on a problem solved in javascript(https://www.codewars.com/kata/5777fe3f355edbf0a5000d11/train/javascript).

1. javascript

it('Return an empty array if more than one x found', () => {
    assert.deepEqual(xMarksTheSpot([['x', 'o'], ['o', 'x']]), []);
});

2. Julia

 @fact xMarksTheSpot(['x' 'o'; 'o' 'x'; 'o' 'x']) --> []

Tests, task descriptions, return values, everything was taken from the task using js as an example.

I added some remarks on the translation page.

Your return type is a 1-D array containing both x and y coordinates, which is different from other translations that return a 2-D array, each containg (x,y) pairs.

Did you invent this spec?

Return an empty array if more than one x found

Really Nice!!

Do you want review my golang translation?

Added Golang translation

Need to review and approve this code.