Merge pull request #273 from github/dependabot-c67993ba-56b8-4f4e-a911-22640b185a14

Enable Dependabot

.github/dependabot.yaml

@@ -0,0 +1,11 @@
+---
+version: 2
+updates:
+  - package-ecosystem: 'bundler'
+    directory: '/'
+    schedule:
+        interval: 'weekly'
+  - package-ecosystem: 'github-actions'
+    directory: '/'
+    schedule:
+        interval: 'weekly'

.github/workflows/test.yml

@@ -0,0 +1,24 @@
+name: Test
+
+on: [push, pull_request]
+
+jobs:
+  test:
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby: ["2.7", "3.0", "3.1"]
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v3
+    
+    - name: Set up Ruby
+      uses: ruby/setup-ruby@v1
+      with:
+        ruby-version: ${{matrix.ruby}}
+        bundler-cache: true
+    
+    - name: Run tests
+      run: bundle exec rake

.travis.yml

@@ -1,12 +0,0 @@
-sudo: false
-script: script/test
-rvm:
-  - 1.9.3
-  - '2.1'
-  - '2.2'
-  - '2.3'
-  - '2.4'
-  - '2.5'
-  - '2.6'
-notifications:
-  email: false

Gemfile

@@ -2,5 +2,6 @@ source "https://rubygems.org"
 
 gem "rake", "~> 10.3.2"
 gem "minitest", "~> 5.3.5"
+gem "i18n", "~> 1.8.5"
 
 gemspec

Gemfile.lock

@@ -0,0 +1,25 @@
+PATH
+  remote: .
+  specs:
+    gemoji (4.1.0)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    concurrent-ruby (1.1.7)
+    i18n (1.8.5)
+      concurrent-ruby (~> 1.0)
+    minitest (5.3.5)
+    rake (10.3.2)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  gemoji!
+  i18n (~> 1.8.5)
+  minitest (~> 5.3.5)
+  rake (~> 10.3.2)
+
+BUNDLED WITH
+   2.4.10

Rakefile

@@ -20,5 +20,5 @@ namespace :db do
 end
 
 file 'vendor/unicode-emoji-test.txt' do |t|
-  system 'curl', '-fsSL', 'http://unicode.org/Public/emoji/12.0/emoji-test.txt', '-o', t.name
+  system 'curl', '-fsSL', 'http://unicode.org/Public/emoji/15.0/emoji-test.txt', '-o', t.name
 end

db/dump.rb

@@ -1,9 +1,11 @@
 # frozen_string_literal: true
 
+require "i18n"
 require 'emoji'
 require 'json'
 require_relative './emoji-test-parser'
 
+I18n.config.available_locales = :en
 items = []
 
 _, categories = EmojiTestParser.parse(File.expand_path("../../vendor/unicode-emoji-test.txt", __FILE__))
@@ -12,34 +14,32 @@ seen_existing = {}
 for category in categories
   for sub_category in category[:emoji]
     for emoji_item in sub_category[:emoji]
-      unicodes = emoji_item[:sequences].sort_by(&:bytesize)
-      existing_emoji = nil
-      unicodes.detect do |raw|
-        existing_emoji = Emoji.find_by_unicode(raw)
+      raw = emoji_item[:sequences][0]
+      existing_emoji = Emoji.find_by_unicode(raw) || Emoji.find_by_unicode("#{raw}\u{fe0f}")
+      if seen_existing.key?(existing_emoji)
+        existing_emoji = nil
+      else
+        seen_existing[existing_emoji] = true
       end
-      existing_emoji = nil if seen_existing.key?(existing_emoji)
+      description = emoji_item[:description].sub(/^E\d+(\.\d+)? /, '')
       output_item = {
-        emoji: unicodes[0],
-        description: emoji_item[:description],
+        emoji: raw,
+        description: description,
         category: category[:name],
       }
       if existing_emoji
-        eu = existing_emoji.unicode_aliases
-        preferred_raw = eu.size == 2 && eu[0] == "#{eu[1]}\u{fe0f}" ? eu[1] : eu[0]
         output_item.update(
-          emoji: preferred_raw,
           aliases: existing_emoji.aliases,
           tags: existing_emoji.tags,
           unicode_version: existing_emoji.unicode_version,
           ios_version: existing_emoji.ios_version,
         )
-        seen_existing[existing_emoji] = true
       else
         output_item.update(
-          aliases: [emoji_item[:description].gsub(/\W+/, '_').downcase],
+          aliases: [I18n.transliterate(description).gsub(/\W+/, '_').downcase],
           tags: [],
-          unicode_version: "12.0",
-          ios_version: "13.0",
+          unicode_version: "15.0",
+          ios_version: "16.4",
         )
       end
       output_item[:skin_tones] = true if emoji_item[:skin_tones]

db/emoji-test-parser.rb

@@ -9,12 +9,8 @@ module EmojiTestParser
     "\u{1F3FE}", # medium-dark skin tone
     "\u{1F3FF}", # dark skin tone
   ]
-  HAIR_MODIFIERS = [
-    "\u{1F9B0}", # red-haired
-    "\u{1F9B1}", # curly-haired
-    "\u{1F9B2}", # bald
-    "\u{1F9B3}", # white-haired
-  ]
+  SKIN_TONES_RE = /(#{SKIN_TONES.join("|")})/o
+  SKIP_TYPES = ["unqualified", "component"]
 
   module_function
 
@@ -52,12 +48,12 @@ module EmojiTestParser
         else
           row, desc = line.split("#", 2)
           desc = desc.strip.split(" ", 2)[1]
-          codepoints, _ = row.split(";", 2)
+          codepoints, qualification = row.split(";", 2)
+          next if SKIP_TYPES.include?(qualification.strip)
           emoji_raw = codepoints.strip.split.map { |c| c.hex }.pack("U*")
-          next if HAIR_MODIFIERS.include?(emoji_raw)
           emoji_normalized = emoji_raw
             .gsub(VARIATION_SELECTOR_16, "")
-            .gsub(/(#{SKIN_TONES.join("|")})/o, "")
+            .gsub(SKIN_TONES_RE, "")
           emoji_item = emoji_map[emoji_normalized]
           if SKIN_TONES.any? { |s| emoji_raw.include?(s) }
             emoji_item[:skin_tones] = true if emoji_item
@@ -91,7 +87,7 @@ if $0 == __FILE__
     html_output = true
   end
 
-  _, categories = EmojiTestParser.parse
+  _, categories = EmojiTestParser.parse(File.expand_path("../../vendor/unicode-emoji-test.txt", __FILE__))
 
   trap(:PIPE) { abort }
 

gemoji.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name    = "gemoji"
-  s.version = "4.0.0.pre0"
+  s.version = "4.1.0"
   s.summary = "Unicode emoji library"
   s.description = "Character information and metadata for Unicode emoji."
 

lib/emoji.rb

@@ -51,11 +51,12 @@ module Emoji
 
   # Public: Find an emoji by its unicode character. Return nil if missing.
   def find_by_unicode(unicode)
-    unicodes_index[unicode]
+    unicodes_index[unicode] || unicodes_index[unicode.sub(SKIN_TONE_RE, "")]
   end
 
   private
     VARIATION_SELECTOR_16 = "\u{fe0f}".freeze
+    SKIN_TONE_RE = /[\u{1F3FB}-\u{1F3FF}]/
 
     # Characters which must have VARIATION_SELECTOR_16 to render as color emoji:
     TEXT_GLYPHS = [
@@ -68,85 +69,9 @@ module Emoji
       "\u{00ae}",  # registered
       "\u{2122}",  # trade mark
       "\u{3030}",  # wavy dash
-      "\u{263a}",  # smiling face
-      "\u{261D}",  # index pointing up
-      "\u{270C}",  # victory hand
-      "\u{270D}",  # writing hand
-      "\u{2764}",  # red heart
-      "\u{2763}",  # heavy heart exclamation
-      "\u{2668}",  # hot springs
-      "\u{2708}",  # airplane
-      "\u{2600}",  # sun
-      "\u{2601}",  # cloud
-      "\u{2602}",  # umbrella
-      "\u{2744}",  # snowflake
-      "\u{2603}",  # snowman
-      "\u{2660}",  # spade suit
-      "\u{2665}",  # heart suit
-      "\u{2666}",  # diamond suit
-      "\u{2663}",  # club suit
-      "\u{260e}",  # telephone
-      "\u{2709}",  # envelope
-      "\u{270F}",  # pencil
-      "\u{2712}",  # black nib
-      "\u{2702}",  # scissors
-      "\u{26a0}",  # warning
-      "\u{2B06}",  # up arrow
-      "\u{2197}",  # up-right arrow
-      "\u{27A1}",  # right arrow
-      "\u{2198}",  # down-right arrow
-      "\u{2B07}",  # down arrow
-      "\u{2199}",  # down-left arrow
-      "\u{2B05}",  # left arrow
-      "\u{2196}",  # up-left arrow
-      "\u{2195}",  # up-down arrow
-      "\u{2194}",  # left-right arrow
-      "\u{21A9}",  # right arrow curving left
-      "\u{21AA}",  # left arrow curving right
-      "\u{2934}",  # right arrow curving up
-      "\u{2935}",  # right arrow curving down
-      "\u{2721}",  # star of David
-      "\u{262F}",  # yin yang
-      "\u{271D}",  # latin cross
-      "\u{25B6}",  # play button
-      "\u{25C0}",  # reverse button
-      "\u{23CF}",  # eject button
-      "\u{2640}",  # female sign
-      "\u{2642}",  # male sign
-      "\u{267B}",  # recycling symbol
-      "\u{2611}",  # ballot box with check
-      "\u{2714}",  # heavy check mark
-      "\u{2716}",  # heavy multiplication x
-      "\u{303D}",  # part alternation mark
-      "\u{2733}",  # eight-spoked asterisk
-      "\u{2734}",  # eight-pointed star
-      "\u{2747}",  # sparkle
-      "\u{203C}",  # double exclamation mark
-      "\u{2049}",  # exclamation question mark
-      "\u{23}\u{20E3}",  # keycap: #
-      "\u{2A}\u{20E3}",  # keycap: *
-      "\u{30}\u{20E3}",  # keycap: 0
-      "\u{31}\u{20E3}",  # keycap: 1
-      "\u{32}\u{20E3}",  # keycap: 2
-      "\u{33}\u{20E3}",  # keycap: 3
-      "\u{34}\u{20E3}",  # keycap: 4
-      "\u{35}\u{20E3}",  # keycap: 5
-      "\u{36}\u{20E3}",  # keycap: 6
-      "\u{37}\u{20E3}",  # keycap: 7
-      "\u{38}\u{20E3}",  # keycap: 8
-      "\u{39}\u{20E3}",  # keycap: 9
-      "\u{2139}",  # information
-      "\u{24C2}",  # circled M
-      "\u{1F17F}", # P button
-      "\u{3297}",  # Japanese “congratulations” button
-      "\u{3299}",  # Japanese “secret” button
-      "\u{25AA}",  # black small square
-      "\u{25AB}",  # white small square
-      "\u{25FB}",  # white medium square
-      "\u{25FC}",  # black medium square
     ].freeze
 
-    private_constant :VARIATION_SELECTOR_16, :TEXT_GLYPHS
+    private_constant :VARIATION_SELECTOR_16, :TEXT_GLYPHS, :SKIN_TONE_RE
 
     def parse_data_file
       data = File.open(data_file, 'r:UTF-8') do |file|
@@ -194,6 +119,7 @@ module Emoji
           emoji.description = dedup.call(raw_emoji[:description])
           emoji.unicode_version = dedup.call(raw_emoji[:unicode_version])
           emoji.ios_version = dedup.call(raw_emoji[:ios_version])
+          emoji.skin_tones = raw_emoji.fetch(:skin_tones, false)
         end
       end
     end

lib/emoji/character.rb

@@ -11,6 +11,11 @@ module Emoji
     # True if the emoji is not a standard Emoji character.
     def custom?() !raw end
 
+    # True if the emoji supports Fitzpatrick scale skin tone modifiers
+    def skin_tones?() @skin_tones end
+
+    attr_writer :skin_tones
+
     # A list of names uniquely referring to this emoji.
     attr_reader :aliases
 
@@ -38,6 +43,28 @@ module Emoji
     # Raw Unicode string for an emoji. Nil if emoji is non-standard.
     def raw() unicode_aliases.first end
 
+    # Raw Unicode strings for each skin tone variant of this emoji. The result is an empty array
+    # unless the emoji supports skin tones.
+    #
+    # Note: for emojis that depict multiple people (e.g. couples or families), this will not produce
+    # every possible permutation of skin tone per person.
+    def raw_skin_tone_variants
+      return [] if custom? || !skin_tones?
+      raw_normalized = raw.sub(VARIATION_SELECTOR_16, "")
+      idx = raw_normalized.index(ZERO_WIDTH_JOINER)
+      SKIN_TONES.map do |modifier|
+        if raw_normalized == PEOPLE_HOLDING_HANDS
+          # special case to apply the modifier to both persons
+          raw_normalized[0...idx] + modifier + raw_normalized[idx..nil] + modifier
+        elsif idx
+          # insert modifier before zero-width joiner
+          raw_normalized[0...idx] + modifier + raw_normalized[idx..nil]
+        else
+          raw_normalized + modifier
+        end
+      end
+    end
+
     def add_unicode_alias(str)
       unicode_aliases << str
     end
@@ -54,6 +81,7 @@ module Emoji
       @aliases = Array(name)
       @unicode_aliases = []
       @tags = []
+      @skin_tones = false
     end
 
     def inspect
@@ -77,6 +105,20 @@ module Emoji
 
     private
 
+    VARIATION_SELECTOR_16 = "\u{fe0f}".freeze
+    ZERO_WIDTH_JOINER = "\u{200d}".freeze
+    PEOPLE_HOLDING_HANDS = "\u{1f9d1}\u{200d}\u{1f91d}\u{200d}\u{1f9d1}".freeze
+
+    SKIN_TONES = [
+      "\u{1F3FB}", # light skin tone
+      "\u{1F3FC}", # medium-light skin tone
+      "\u{1F3FD}", # medium skin tone
+      "\u{1F3FE}", # medium-dark skin tone
+      "\u{1F3FF}", # dark skin tone
+    ]
+
+    private_constant :VARIATION_SELECTOR_16, :ZERO_WIDTH_JOINER, :PEOPLE_HOLDING_HANDS, :SKIN_TONES
+
     def default_image_filename
       if custom?
         '%s.png' % name

test/emoji_test.rb

@@ -50,14 +50,19 @@ class EmojiTest < TestCase
   GENDER_EXCEPTIONS = [
     "man_with_gua_pi_mao",
     "woman_with_headscarf",
-    "man_in_tuxedo",
-    "pregnant_woman",
     "isle_of_man",
-    "blonde_woman",
+    "blonde_woman", # blond_haired_man
     /^couple(kiss)?_/,
     /^family_/,
   ]
 
+  DASH_EXCEPTIONS = [
+    "-1",
+    "t-rex",
+    "e-mail",
+    "non-potable_water",
+  ]
+
   test "emojis have valid names" do
     aliases = Emoji.all.flat_map(&:aliases)
 
@@ -76,7 +81,7 @@ class EmojiTest < TestCase
     alias_count = Hash.new(0)
     aliases.each do |name|
       alias_count[name] += 1
-      invalid << name if name !~ /\A[\w+-]+\Z/
+      invalid << name if name !~ /\A[\w+]+\Z/ && !DASH_EXCEPTIONS.include?(name)
       another_gender = to_another_gender.call(name)
       gender_mismatch << another_gender unless aliases.include?(another_gender)
     end
@@ -97,12 +102,7 @@ class EmojiTest < TestCase
     message = "Missing or incorrect unicodes:\n"
     source_unicode_emoji.each do |emoji|
       emoji[:sequences].each do |raw|
-        found = Emoji.find_by_unicode(raw)
-        if text_glyphs.include?(raw)
-          assert_nil found, Emoji::Character.hex_inspect(raw)
-          next
-        end
-        next if found
+        next if text_glyphs.include?(raw) || Emoji.find_by_unicode(raw)
         message << "%s (%s)" % [Emoji::Character.hex_inspect(raw), emoji[:description]]
         if found = Emoji.find_by_unicode(raw.gsub("\u{fe0f}", ""))
           message << " - could be %s (:%s:)" % [found.hex_inspect, found.name]
@@ -115,13 +115,6 @@ class EmojiTest < TestCase
     assert_equal 0, missing, message
   end
 
-  test "raw representation does not include VARIATION SELECTOR 16 unless necessary" do
-    emoji = Emoji.all.select do |emoji|
-      !emoji.custom? && emoji.raw.end_with?("\u{fe0f}") && emoji.unicode_aliases.size == 2
-    end
-    assert_equal [], emoji
-  end
-
   test "emoji have category" do
     missing = Emoji.all.select { |e| e.category.to_s.empty? }
     assert_equal [], missing.map(&:name), "some emoji don't have a category"
@@ -164,6 +157,36 @@ class EmojiTest < TestCase
     assert_equal '8.3', emoji.ios_version
   end
 
+  test "skin tones" do
+    smiley = Emoji.find_by_alias("smiley")
+    assert_equal false, smiley.skin_tones?
+    assert_equal [], smiley.raw_skin_tone_variants
+
+    wave = Emoji.find_by_alias("wave")
+    assert_equal true, wave.skin_tones?
+
+    wave = Emoji.find_by_unicode("\u{1f44b}\u{1f3ff}") # wave + dark skin tone
+    assert_equal "wave", wave.name
+
+    woman_with_beard = Emoji.find_by_unicode("\u{1f9d4}\u{200d}\u{2640}\u{fe0f}")
+    assert_equal [
+      "1f9d4-1f3fb-200d-2640",
+      "1f9d4-1f3fc-200d-2640",
+      "1f9d4-1f3fd-200d-2640",
+      "1f9d4-1f3fe-200d-2640",
+      "1f9d4-1f3ff-200d-2640",
+    ], woman_with_beard.raw_skin_tone_variants.map { |u| Emoji::Character.hex_inspect(u) }
+
+    people_holding_hands = Emoji.find_by_alias("people_holding_hands")
+    assert_equal [
+      "1f9d1-1f3fb-200d-1f91d-200d-1f9d1-1f3fb",
+      "1f9d1-1f3fc-200d-1f91d-200d-1f9d1-1f3fc",
+      "1f9d1-1f3fd-200d-1f91d-200d-1f9d1-1f3fd",
+      "1f9d1-1f3fe-200d-1f91d-200d-1f9d1-1f3fe",
+      "1f9d1-1f3ff-200d-1f91d-200d-1f9d1-1f3ff",
+    ], people_holding_hands.raw_skin_tone_variants.map { |u| Emoji::Character.hex_inspect(u) }
+  end
+
   test "no custom emojis" do
     custom = Emoji.all.select(&:custom?)
     assert 0, custom.size