[\p{Emoji}] also matches numerical characters

[hachi8833]

Looks like this behavior is a bug:
https://rubular.com/r/XH1SVLPKgG7eM7

As far as I see, [\p{Emoji}] matches only 0123456789, and does not match any other N scripts such as ¼〇ⅲ៥

Could you please check this?

[747]

Just dropping by, but this seems a spec-conforming behavior, as described in this document:
https://www.unicode.org/Public/UCD/latest/ucd/emoji/emoji-data.txt

Not all emoji were added as new characters; some are just refashioning of (sequences of) existing characters.

"0".codepoints.map{ |c| sprintf("%04X", c) }
# => ["0030"]

"0️⃣".codepoints.map{ |c| sprintf("%04X", c) }
# => ["0030", "FE0F", "20E3"] (<- "0" + VS16 (emoji) + keycap)

\p{Emoji} represents a character that could be part of emoji. For matching a character that is unambiguously a part of emoji, \p{Emoji_Presentation} is a better choice.