What command? Also perhaps you missed this part of the article:
> The same loop is in several ISO-8859 and Windows-125x encodings and has been there since Ruby 2.4.0. It is reached by upcase, swapcase, downcase(:fold) and casecmp?.
> UTF-8 strings take a different code path and are not affected.