The tests aren't currently isolating these validation requirements:

* The white king (K) and black king (k) must not be in adjacent spaces.
* Pawns (P or p) can’t be on the first rank (bottom row) or the eighth rank (top row).

It's also missing random tests.

Double-spaced for legibility and makes sure to clean up after itself. I don't know about best practices, but this definitely has good manners.

The first fixed case is passing an empty dict rather than a set.

Yeah, I believe once something's retired you can't republish. You'll have to make a new kata and publish that one.

Cool kata! Couple clarifications that should be made to the description:

• Specify that ' won't be used in text (unless you want to open the escape character can of worms)
• Specify what cd .. does (probably known to most solvers, but should still clarify)
• Explain how / will or won't be appended to calls to cd, it seems like both are used, but not specified
• Expand on "any characters are allowed in file and directory names", what about naming a directory .. or one foo and another foo/

You can of course create whatever difficulty you'd like, but just know that it's very difficult to get an 8kyu through the beta testing process. It tends to be easiest to learn the ropes of authoring on a 5 or 6 which are more likely to get approved.

As for uniqueness, reviewers are in general looking for a problem that is not equivalent (or too close) to currently existing kata. For example, this one might be considered too similar to this kata or this one. Naturally, these aren't identical, but especially at the 8kyu level the differences are often just cosmetic and the same skills are being tested.

You can also get some guidance from the discord channel before publishing; it's handy to get some eyes on your draft before having beta testers vote on it.

Hey, you've clearly put a lot of work into writing these, and they're likely to get auto-retired due to votes, so I wanted to fill you in on what the problems are.

• Sample tests: having only a single sample test isn't usually sufficient, as you want to show any special cases that are worth considering. In this case, maybe a test with an empty input, a length 1 array, and a length 5 array
• Submission tests: You need to include both fixed and random tests for submission test. This means multiple different inputs, as well as generating random inputs. Random tests are important to avoid hardcoded solutions like this.
• Description: Nothing wrong with having backstory, but be ready to have a lot of solvers not be interested in it (and potentially vote accordingly). It's a matter of opinion, but as long as you keep clear descriptions separate from the story you'll usually be alright.
• Kata novelty: Coming up with a unique idea is probably the hardest part of authoring. Ironically, authoring an easier kata is more difficult, as most simple ideas have already been done, and quality expectations are very high as new programmers will be solving them.

If you want to continue to try authoring, please read through the Creating your first kata tutorial and the rest of the relevant topics in the docs. They'll help you avoid learning all this via trial and error.

I'd recommend unpublishing these are fixing them if you want to try to get them approved, but due to the novelty issue it'll be an uphill battle. I'd strongly recommend a first kata be standalone (not part of a series) and not aiming to be an 8 kyu.

Approved, thanks for the translation!

Yeah, that should prevent solving via combining other patterns.

You should have additional larger tests (and possibly specifically test larger primes). I was able to pass with a solution that fails on ~20% of the domain.

Fixed, thanks.

I wasn't previously familiar with minterms, so I could be mistaken, but I believe the example case in the description is incorrect:

print(get_minterms("F(A,B,C)=A'B+BC'+A"))  # Output: [1, 3, 5, 7]

[[]] is a 1x0 matrix, not an NxN matrix.

Looks good!