What language?

There's no performance requirement though so it's not ranked too low.

Duplicate:

https://www.codewars.com/kata/find-the-square-root

There are a number of numerical methods for finding the square root of a number. Due to the irregularities in rounding, consider implementing an allowance for a margin of error something to the magnitude of 1e-6, as in their square root must be within 1 one millionth of the actual value. Good luck!

Agreed! At least a 5kyu. Uff....

That's intentional as to not clutter the test cases.

Sample tests still pass in the same grid copy in C#.

Oh yeah. That should be fixed now. I've allowed the very big squares to be reused, as it would take to long to copy them.

The random tests and sample tests still pass in the same grid copy.

You're right. I've made the individual tests copy the original grid, to fix this. Thanks for pointing that out.

Because the grid is reference type, each run is not independent of the last as you are passing in the same grid for many tests. As such the tests will fail on subsequent runs if the data therein is manipulated unless the programmer starts by making a local copy of the grid on each run. This is out of spec. I suggest changing your testing methods such that every grid tested has a unique reference. It is a cool problem otherwise.

This Kata is ranked too low. In order to solve this efficiently you must be able to reason through many nested branches, loops, lists, and dictionaries.

I'll see about getting a different, https diagram.

I see it, http://strd6.com/wp-content/uploads/2010/06/circle_collisions.png. But these days links from https to http may cause problems...