no, this solution about a lot of copiyng

Would fail on "\n"

This can be done with trial and error. two character set is taken to avoid wrong mappings.

As '@' is the the ascii character before capital letter A, '@' is subtracted to get the correspending number from ascii set.
example :
'A'-'@' = 1
'B' -@' = 2 ..

This is no longer the case.

It would be 6 Kyu if the user is enforced to implement an efficient algorithm.

exactly

Now it's invalid.

The first step is to make it work, then you try to make it better... efficient, faster, etc....
I guess this guy forgot about that part.

Not a kata issue.

Good comment, not sure what's happening; c stuff is pretty stable since > ~2.7/x