I'm posting this here for some advice on whether this is a bug, or
intentional behaviour.

ruby 1.9.2-preview1 is quite capable of distinguishing Unicode
upper-case characters. However this doesn't seem to be used when
deciding whether an identifier is a constant or a local variable.

>> SCH??N = 1   # constant
=> 1
>> ??BER = 2    # local variable!
=> 2
>> self.class.constants.grep(/SCH|BER/)
=> [:SCH??N]
>> local_variables.grep(/SCH|BER/)
=> [:??BER]

I am not sure from the source code whether this is intentional or not.
In parse.c it uses rb_enc_isupper which understands encodings:

        else if (rb_enc_isupper(m[0], enc)) {
            id = ID_CONST;
        }
        else {
            id = ID_LOCAL;

This is in rb_intern3. And yet it is called from rb_intern2 which says:

ID
rb_intern2(const char *name, long len)
{
    return rb_intern3(name, len, rb_usascii_encoding());
}

thereby ignoring the source encoding and saying always use ASCII rules.

Can someone confirm if this is the behaviour which is intended?
-- 
Posted via http://www.ruby-forum.com/.