Crystal Emoji Regex 💎

A set of Ruby regular expressions for matching Unicode Emoji symbols.

Background

This is based upon the fantastic work from Jessica Stokes' [ruby-emoji-regex](https://github.com/ticky/ruby-emoji-regex) which was based on Mathias Bynens' [emoji-regex](https://github.com/mathiasbynens/emoji-regex) Javascript package. emoji-regex is cleverly assembled based upon data from the Unicode Consortium.

The regular expressions provided herein are derived from that pacakge.

Installation

1. Add the dependency to your shard.yml:

   dependencies:
     emoji_regex:
       github: watzon/emoji_regex
   

2. Run shards install

Usage

emoji_regex provides these regular expressions:

• EmojiRegex::RGIEmoji is the regex you most likely want. It matches all emoji recommended for general interchange, as defined by the Unicode standard's RGI_Emoji property. In a future version, this regular expression will be renamed to EmojiRegex::Regex and all other regexes removed.

• EmojiRegex::Regex is deprecated, and will be replaced with RGIEmoji in a future major version. It matches emoji which present as emoji by default, and those which present as emoji when combined with U+FE0F VARIATION SELECTOR-16.

• EmojiRegex::Text is deprecated, and will be removed in a future major version. It matches emoji which present as text by default (regardless of variation selector), as well as those which present as emoji by default.

RGI vs Emoji vs Text Presentation

RGI_Emoji is a property of emoji symbols, defined in Unicode Technical Report #51 which marks emoji as being supported by major vendors and therefore expected to be usable generally. In most cases, this is the property you will want when seeking emoji characters.

Emoji_Presentation is another such property, defined in UTR#51 which controls whether symbols are intended to be rendered as emoji by default.

Generally, for emoji which re-use Unicode code points which existed before Emoji itself was introduced to Unicode, Emoji_Presentation is false. Emoji_Presentation may be true but RGI_Emoji false for characters with non-standard emoji-like representations in certain conditions. Notable cases are the Emoji Keycap Sequences (#️⃣, 1️⃣, 9️⃣, *️⃣, etc.) which are sequences composed of three characters; the base character, an U+FE0F VARIATION SELECTOR-16, and finally the U+20E3 COMBINING ENCLOSING KEYCAP.

These characters, therefore, are matched to varying degrees of precision by each of the regular expressions included in this package;

• # is matched only by EmojiRegex::Text as it is considered to be a text part of a possible emoji.
• #️ is matched by EmojiRegex::Regex as well as EmojiRegex::Text as it has Emoji_Presentation despite not being a generally accepted Emoji or recommended for general interchange.
• #️⃣ is matched by all three regular expressions, as it is recommended for general interchange.

It's most likely that the regular expression you want is EmojiRegex::RGIEmoji! ☺️

Example

require "emoji_regex"

text = <<-TEXT
\u{231A}: ⌚ default Emoji presentation character (Emoji_Presentation)
\u{2194}: ↔ default text presentation character
\u{2194}\u{FE0F}: ↔️ default text presentation character with Emoji variation selector
#: # default text presentation character
#\u{FE0F}: #️ default text presentation character with Emoji variation selector
#\u{FE0F}\u{20E3}: #️⃣ default text presentation character with Emoji variation selector and combining enclosing keycap
\u{1F469}: 👩 Emoji modifier base (Emoji_Modifier_Base)
\u{1F469}\u{1F3FF}: 👩🏿 Emoji modifier base followed by a modifier
TEXT

puts "EmojiRegex::RGIEmoji"
text.scan EmojiRegex::RGIEmoji do |match|
  puts "Matched sequence #{match[0]} — code points: #{match[0].size}"
end

puts

puts "EmojiRegex::Regex"
text.scan EmojiRegex::Regex do |match|
  puts "Matched sequence #{match[0]} — code points: #{match[0].size}"
end

puts

puts "EmojiRegex::Text"
text.scan EmojiRegex::Text do |match|
  puts "Matched sequence #{match[0]} — code points: #{match[0].size}"
end

Console output:

EmojiRegex::RGIEmoji
Matched sequence ⌚ — code points: 1
Matched sequence ⌚ — code points: 1
Matched sequence ↔️ — code points: 2
Matched sequence ↔️ — code points: 2
Matched sequence #️⃣ — code points: 3
Matched sequence #️⃣ — code points: 3
Matched sequence 👩 — code points: 1
Matched sequence 👩 — code points: 1
Matched sequence 👩🏿 — code points: 2
Matched sequence 👩🏿 — code points: 2

EmojiRegex::Regex
Matched sequence ⌚ — code points: 1
Matched sequence ⌚ — code points: 1
Matched sequence ↔️ — code points: 2
Matched sequence ↔️ — code points: 2
Matched sequence #️ — code points: 2
Matched sequence #️ — code points: 2
Matched sequence #️⃣ — code points: 3
Matched sequence #️⃣ — code points: 3
Matched sequence 👩 — code points: 1
Matched sequence 👩 — code points: 1
Matched sequence 👩🏿 — code points: 2
Matched sequence 👩🏿 — code points: 2

EmojiRegex::Text
Matched sequence ⌚ — code points: 1
Matched sequence ⌚ — code points: 1
Matched sequence ↔ — code points: 1
Matched sequence ↔ — code points: 1
Matched sequence ↔️ — code points: 2
Matched sequence ↔️ — code points: 2
Matched sequence # — code points: 1
Matched sequence # — code points: 1
Matched sequence #️ — code points: 2
Matched sequence #️ — code points: 2
Matched sequence #️⃣ — code points: 3
Matched sequence #️⃣ — code points: 3
Matched sequence 👩 — code points: 1
Matched sequence 👩 — code points: 1
Matched sequence 👩🏿 — code points: 2
Matched sequence 👩🏿 — code points: 2

Development

Requirements

• Ruby
• Node (v6 or newer)
• Yarn

Initial setup

To install all the Ruby and Javascript dependencies, you can run:

bin/setup

To update the Ruby source files based on the emoji-regex library:

bundle exec rake regenerate

Specs

A spec suite is provided, which can be run as:

crystal spec

Versioning Policy

The version of crystal-emoji-regex will always track the upstream version from ruby-emoji-regex.