I have an assumption that a compiler is likely to optimize the calls to sqrt by making the result a variable in the compiled program anyway. I heard it similarly does this with repeated strlen calls. While it would be good practice to tell the compiler to do that explicitly, it's unlikely to matter in the end anyway, right? And that's besides the question of whether this solution is meant as a code golf solution. Since best practice and code golf are often the antithesis of eachother.

Fixed

fixed by mauro-chabby combo

Good that you wrote it. I had the same problem :D

1 -> 2 -> 3 -> None should equal 1 -> 2 -> 3 -> None

0 -> 1 -> 4 -> 9 -> 16 -> None should equal 0 -> 1 -> 4 -> 9 -> 16 -> None

Oh data type was str, not int -> error message is a bit obscure

1 -> 2 -> 3 -> None should equal 1 -> 2 -> 3 -> None

Approved, thanks!

One of the most evil? How so? You obviously haven't seen much C code.
Might I suggest studying the results of the IOCCC?

fixed

Approved

approved

approved

approved

approved

well memmem is glibc-specific (not in standard C) and it looks for bytes patterns, not array elements, but it will happen to work there because the indices are < 128. you're the judge :P