stripAccents
stripAccents: (string?) -> string?input with its accents removed: the canonical decomposition (NFD, Unicode 10 data), Ł/ł as L/l, then every character of the Combining Diacritical Marks block (U+0300-U+036F) removed; other combining marks (a kana's voicing mark, Hebrew points) stay, decomposed. nil for nil. stripAccents("Łódź Café") is "Lodz Cafe". The length in UTF-16 units is kept for Latin-1 text (é to e), not the byte length: # drops by one for each accented Latin-1 letter, so an index into one is not an index into the other (Matcher indices on the stripped copy are UTF-16 units).