Incredibly beautiful solution

2 kyu kata if solved without wiki, 6 kyu if with, 4 kyu in average.
Thanks!

This case was from random section, I can't reproduce it.

But I got the same error for another test grid and it looks like I once again confused the sequence of the log and error messages, the user interface is rather unfriendly.
This test grid really has multiple solutions, but my scripts return only one. My mistake.

Thanks for your help!

Something wrong with Python test cases.

"Invalid grid should raise an error" appears for

board = [
[0, 0, 2, 0, 0, 1, 0, 0, 0],
[3, 0, 0, 6, 8, 0, 7, 0, 0],
[0, 0, 0, 5, 0, 0, 6, 0, 0],
[7, 0, 5, 0, 2, 0, 0, 3, 0],
[0, 1, 0, 0, 0, 0, 0, 5, 0],
[0, 8, 0, 0, 4, 0, 9, 0, 1],
[0, 0, 7, 0, 0, 4, 0, 0, 0],
[0, 0, 4, 0, 1, 7, 0, 0, 9],
[0, 0, 0, 9, 0, 0, 8, 0, 0]]

This board has only one solution, I tested it with my script and a lot of other submitted scripts.
Can anyone suggest what the problem is?

6x6 solution scaled to 7x7 failed to bypass medved test case, so forced to implement trial/error "guess" mode.

Thanks for kata and for skyscrapers series!

Nice kata! Thought I could apply my naive 4x4 algorithm and failed...
Spent a whole day manually solving several 6x6 puzzles and generalizing the algorithm, then half day of coding and now I am happy :)
As a result, I do not quite understand what type of algorithm I got. It is not a backtracking, probably it is kind of optimized brute force.

Wow! Nice puzzle! But why just 4 kata? I think it would be at least 3.

I think this kata should be renamed to "Fight with zeroes", since the rest of solution is very simple.

What do you need to speed up? Brute force method is not acceptable - too many combinations.

For all who have difficulties with [[1, 4, 4, 3, 'a'], [3, 2, 4, 1], [4, 1, 3, 3], [2, 0, 1, 4], ['', False, None, '4']] test case and do not understand what the above explanations 'I read the wrong "output"-block' / 'I read the wrong block too' mean:

Invalid value types (boolean): True should equal False

message relates to [[True]] test case, not to [[1, 4, 4, 3, 'a'], [3, 2, 4, 1], [4, 1, 3, 3], [2, 0, 1, 4], ['', False, None, '4']]

Well-designed data schema is self-explanatory, poorly designed schema is very difficult to describe.
This kata is the second case.

0 is a square number? Crazy!

Python's default recursion limit is just 1000
Let test nb_year(10, 0.01, 1, 2000) with recursion soultion
I think recursive solutions are unacceptable for such tasks