in reply to Re: converting character types
in thread converting character types

2**48? Just curious as to the rationale for this. I thought Unicode supported only about 1.1 million code points (0-0x10FFFF) and ISO 10646, 2**32 (or maybe it was 2**31) code points. Is there another charset/encoding standard that requires more, is this a consequence of something internal to Perl, or am I misinformed?

Replies are listed 'Best First'.
Re: converting character types
by Abigail-II (Bishop) on Jul 23, 2002 at 09:16 UTC
    It's just a natural extension - a UTF-8 character can take up to 6 bytes. It isn't quite 2**48 of course, as some bitpatterns are illegal, but it's in this ball park.

    Abigail