• Random tests aren't random
• Rounding the answer doesn't make sense
• Giving the formula in the description takes away part of the challenge

Please do not round, use comparisons with an error margin instead. This is documented.

Hi,

No random tests.

cheers

What is a? Apothem? Side? Perimeter?
Regular hexagon or any haxagon?

The hexagon is irrelevant, the kata is "translate this formula to python".

• Round up/down/nearest/...? (Spoiler: down.) It should be stated in description.
• The description should be language agnostic (no def something(..):, it's python specific).

P.S. don't round float results, use approximate equality.

1. The random tests are not random.
2. Formula in the description can be misunderstood by how it is written, it may be read as one of these:

   3 * sqrt(3 * a^2 / 2)
   3 * sqrt(3 * a^2) / 2
   3 * sqrt(3) * a^2 / 2
   3 * sqrt((3 * a)^2) / 2
   3 * sqrt((3 * a)^2 / 2)
   

I unpublished the kata because it has many technical issues:

• Tests handle floating point equality incorrectly, and some correct solutions fail due to float accuracy errors: 10643216.653600924 should equal 10643216.653600922.
• There is no random tests.
• Sample tests are missing proper organization with @describe and @test decorators.
• The kata contains the dummy JavaScript setup, which should be removed.
• The formula could be written using math markup.

Materials to read:

• Authoring guidelines
• Python authoring guide
• Floating point numbers

You can also visit Codewars Discord and one of its channels related to authoring.

• Sample tests are not using test framework

• No random tests

• Exact duplicate

Python: sample tests should use new test framework (import/describe/it).

Description:

If sum of element is 0 return empty list.

Test:

test.assert_equals(sum_of_elements([0]), 0)

P.S. returning different types (int/list) is bad practice.