bug#24603: [RFC 02/18] Generate upcase and downcase tables from Unicode data

[Eli Zaretskii]

> From: Michal Nazarewicz <mina86@mina86.com>
> Cc: 24603@debbugs.gnu.org
> Date: Tue, 04 Oct 2016 16:54:31 +0200
> 
> +    ;; Ⅰ through Ⅻ had word syntax in the past so set it here as well.
> +    ;; General category of those characers is Number, Letter.
> +    (modify-syntax-entry '(#x2160 . #x216b) "w   " syn-tab)
> +
> +    ;; ⓐ thourgh ⓩ are symbols, other according to Unicode but Emacs set
> +    ;; their syntax to word in the past so keep backwards compatibility.
> +    (modify-syntax-entry '(#x24D0 . #x24E9) "w   " syn-tab))

I think we should document all the changes.  If the list of changes is
too long, and cannot be made short enough by summarizing (instead of
showing each individual character), then it probably should go into
some separate file in admin/unidata/.  If it can be short enough, then
a comment in characters.el is a better place, I think.

> I get the following (annotated) differences:

Can you add the name of each character (just one, the leftmost one) to
its line and post the result?  It's hard to read the report when it
only shows codepoints.

Thanks.