waiting for the continuation...

finally i was able to use my sudoku days...
very good, thanks for writing this kata!!!

Thank you!)

Thanks!

this kata is very good !!!
thanks for writing it ... =)

go ahead ;) (note: you cannot "delete" it, you only can unpublish it, meaning it'll stay in your draft section... :/ )

I don't even know what unix is xD
But yeah, I worked it out quickly enough once I looked at the second test. In general you need two examples before you can deduce a pattern ;)

Too much work thrown away, regrettable!!! but it was a great learning experience, thanks to what helped me ... do I cancel it myself or is it done by you?

Yeah, it's pretty hard to find if a kata already exists or not, in the database... :/

The way you convey the data soesn't change the logic used to manipulate them. Same about the slight difference in the output format. So unfortunately, yours is effectively a duplicate of the other one, sorry.

The usual course of action is to unpublish the new beta kata (hence yours).

Cheers

before writing I looked and I didn't find it, after putting octal in the title I saw this yesterday, but I don't consider it duplicate, his entries are an object, and a different situation happens, don't you agree with me? what do you suggest?

This is very similar:
https://www.codewars.com/kata/57f4ccf0ab9a91c3d5000054/

Totally random tests have a chance of finding new edge cases. Pseudo-random tests have too much attempts at being too smart for that.

The innermost layer should always exist and always be an it. You can wrap that in describe to any level you want, including 0. That's the current situation. Used to be different, but wrapping describe in it already bit me when we went from 8.x to 12.x; won't do that again. Having describe as the outer layer might be recommended ( so minimum 1 level instead of 0 ), but I just don't know if it'll ever be required.

because if they aren't, people will exploit them. You know, just like people hardcoding the tests when there aren't random ones... ;p If you group random tests by edge cases, it's possible to hardocde some parts of them.

(note: the requirement is in the documentation ;p )

Why you want random tests to be totally random? Isn't it sometimes better to skew tests to have a sufficient amount of edge cases provided?

A meet-in-the-middle solution could be providing one "it" for each group of specific (edge) cases, nested under a "describe" block for the random tests.

no that's a bad idea for the random tests: that means they aren't totally random so they could be abused in some situations => Nope.