Unicode strings are not parsed correctly

[dangmai]

Considering this code:

public class UnicodeCharacter {
    public void unicodeCharacterTest(){
        String a = '––––';

        /*
        system.debug('Hi');
        */
    }
}

It contains en-dash, which is represented by Unicode code point 8211. This code will fail to be formatted because apex-ast-serializer reads stdin as ASCII instead of UTF-8