char_unicode_bug.hs

module Main (main) where

import GHC.IO.Encoding
import qualified Data.Char as Char

main :: IO ()
main = do
  setLocaleEncoding utf8

  print $ isDifferent '\4304' '\4308'

  putStrLn $ "isLower \4304: " ++ show (Char.isLower '\4304')
  putStrLn $ "isUpper \4308: " ++ show (Char.isUpper '\4308')

  putStrLn $ "isDigit \4304: " ++ show (Char.isDigit '\4304')
  putStrLn $ "isDigit \4308: " ++ show (Char.isDigit '\4308')

charClass :: Char -> Char
charClass c
  | Char.isLower c || Char.isUpper c = 'c'
  | Char.isDigit c = 'd'
  | otherwise = c

isDifferent :: Char -> Char -> Bool
isDifferent a b = charClass a /= charClass b

run.sh

for ghc in "ghc865" "ghc884" "ghc8102"
do
  nix-shell --pure -p haskell.compiler."$ghc" --run "echo $TERM && echo $LANG && ghc --version && ghc char_unicode_bug.hs -o Main && ./Main"
done

run_output.txt

alacritty
en_US.UTF-8
The Glorious Glasgow Haskell Compilation System, version 8.6.5
[1 of 1] Compiling Main             ( char_unicode_bug.hs, char_unicode_bug.o )
Linking Main ...
True
isLower ა: False
isUpper ე: False
isDigit ა: False
isDigit ე: False
alacritty
en_US.UTF-8
The Glorious Glasgow Haskell Compilation System, version 8.8.4
[1 of 1] Compiling Main             ( char_unicode_bug.hs, char_unicode_bug.o )
Linking Main ...
False
isLower ა: True
isUpper ე: False
isDigit ა: False
isDigit ე: False
alacritty
en_US.UTF-8
The Glorious Glasgow Haskell Compilation System, version 8.10.2
[1 of 1] Compiling Main             ( char_unicode_bug.hs, char_unicode_bug.o )
Linking Main ...
False
isLower ა: True
isUpper ე: False
isDigit ა: False
isDigit ე: False