Ruby

__ENCODING__ should respect the encoding of the given source code.

$ ruby -Ke -e 'p __ENCODING__' #<Encoding:EUC-JP> 

p eval("__ENCODING__".encode("euc-jp")) #=> #<Encoding:EUC-JP> 

But it is ignored when SCRIPT_LINES__ is set.

$ cat script_lines.rb SCRIPT_LINES__ = {} $ ruby -r./script_lines.rb -Ke -e 'p __ENCODING__' #<Encoding:UTF-8> 

Also similar when keep_script_lines is true.

pp RubyVM::AbstractSyntaxTree.parse("__ENCODING__".encode("euc-jp"), keep_script_lines: true) #=> (SCOPE@1:0-1:12 tbl: [] args: nil body: (LIT@1:0-1:12 #<Encoding:UTF-8>))