Python fork here:
https://www.codewars.com/kumite/6a279479b9242e8ca2d47e13?sel=6a279479b9242e8ca2d47e13

Which addresses the issue above. Notes:

1. Replaced author solution with one which parses non-regular (i.e. arbitrarily nested) chemical formulae
2. Added a non-regular example to fixed tests
3. Replaced random test generator with a tree-based generator which generates properly random formulae and their atom counts (obviates the need for a reference solution)

• kata ranks cannot be changed
• it is not a regular language since parentheses can be nested

kata ranks cannot be changed

Python random tests contain nested brackets but do not contain multipliers on them, whereas fixed tests contain multipliers on brackets but no truly nested brackets (each layer of brackets is a unique pair of symbols). This means that all the flat regex-based parsers in the Python solutions - of which there are many, including my first attempt - are failing to actually parse the random tests.

Give most of the top solutions for python this test case and they will not succeed:

("(Al(H2O)6)2(SO4)3(H2O)5", {'H': 34, 'O': 29, 'S': 3, 'Al': 2}, "Should parse alunogen: (Al(H2O)6)2(SO4)3(H2O)5")

More proof: a parser which just ignores the brackets entirely for the random tests
https://www.codewars.com/kata/reviews/55770da1dcdc57f27f00004b/groups/6a2638d8e4d51c69b2f53041

If the intention of the kata is to test formulae as a regular language, as I expect from the description/fixed tests, then random tests should not be generating brackets nested within brackets of the same type (at 5-kyu at least).

If the intention of the kata is to test formulae as a non-regular language, then that should be reflected in the fixed tests with nested brackets. This will invalidate many solutions.

Random tests should have multipliers on brackets regardless.

Haskell fork adding random tests.

    $this->assertFalse(validate_battlefield([
        [1, 1, 0, 0, 0, 1, 1, 0, 0, 0],
        [1, 0, 0, 0, 0, 0, 0, 0, 1, 0],
        [1, 0, 0, 0, 1, 1, 1, 0, 1, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 1, 0],
        [0, 0, 0, 0, 1, 1, 1, 0, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 1, 0],
        [0, 1, 0, 1, 0, 0, 0, 0, 0, 0],
        [0, 1, 0, 0, 0, 0, 0, 1, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
    ]));
    
    public function testAdditionalShipWithLengthFiveIsInvalid(): void
{
    $field = [
        [1, 1, 1, 1, 1, 0, 0, 0, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
        [1, 1, 1, 1, 0, 0, 0, 0, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
        [1, 1, 1, 0, 0, 0, 1, 1, 1, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
        [1, 1, 0, 0, 0, 1, 1, 0, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
        [1, 0, 1, 0, 1, 0, 1, 0, 0, 0],
        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
    ];

    $this->assertFalse(validate_battlefield($field));
}

Please add this test.

Why are you moving __ in your nickname?

there is green between them

Too hard to be yellow, too easy to be blue.

I replaced the author's solution with my own and uncommented those tests. the timeouts / OOM errors should no longer happen.

My first 3 kyu kata! Thanks alot :D

Updated the fork

• Improved random test generation to avoid having to re-parse the molecule
• Improved tests to provide more useful failure messages

this is one of katas that you should have hints about it, it they don't help you, get the solution and absorb it.

good kata :0