What to do about character escapes outside the acceptable Unicode range?

The UTF-8 layer discussion has me thinking about how chr() and
character escapes handle some of the same illegal Unicode values we
are discussing in the layers thread.  Currently, it seems that illegal
Unicode non-characters warn, but that characters outside the Unicode
range beyond 0x10FFFF do not.  E.g.

  $ perl5137 -C -wE 'say chr($_) for 0x10FFFC .. 0x110001'
  􏿼
  􏿽
  Unicode non-character 0x10fffe is illegal for interchange at -e line 1.
  ��
  Unicode non-character 0x10ffff is illegal for interchange at -e line 1.
  ��
  ����
  ����

I wonder if it would be better to have them warn as being outside the
Unicode codepoint range.

I also wonder about someone manually constructing an overlong
character sequence and then trying to output it.  Does that warn?
Should it?

-- David

• What to do about character escapes outside the acceptable Unicode range? by David Golden
• Re: What to do about character escapes outside the acceptable Unicoderange? by karl williamson
• Re: What to do about character escapes outside the acceptable Unicode range? by Eric Brine