Hi,

In message "Re: Subrange of String subclass => invalid object"
    on 02/01/30, Matt Armstrong <matt / lickey.com> writes:

|You are probably thinking of UTF8, which does allow the same Unicode
|number to be encoded multiple ways.  But I assume a String#char_at(i)
|method would return the Unicode number itself, not a UTF8 byte
|sequence.

Unicode itself is redundant, which means it has the "feature" to
represent same "character" in the different code point sequence.  It
requires normalization.

|So long as String#char_at can return a 4 byte numerical value, I think
|it pretty much covers everything.

I agree with this point anyway.

							matz.