I'd say it's better to just call lower() once you've matched, and use re.finditer() instead of re.findall(), so that the text is only swept once.

I think it's better than the first solution(because of r"[a-z'][a-z]+[a-z']").

this is beauty

Yours is more readable than mine, but my logic is a bit cleaner.

i think last line is redundant

yes

Thank you so much.
New for me - isalpha and enumerate
Very powerfull mechanics

Nicely Done!!

Thank you.

Why is creating a new object successful for this Kata whereas if chars[index] = chars[index].upper() is used all outputs are Uppercase?

Similar logic, but much cleaner than mine. Nice!