Merge branch 'fix/proper-emoji-qualification' into 'develop'

Emoji: implement full-qualifier using combinations

See merge request pleroma/pleroma!3709

lib/pleroma/emoji.ex

@@ -9,6 +9,7 @@ defmodule Pleroma.Emoji do
   """
   use GenServer
 
+  alias Pleroma.Emoji.Combinations
   alias Pleroma.Emoji.Loader
 
   require Logger
@@ -137,4 +138,17 @@ defmodule Pleroma.Emoji do
   end
 
   def is_unicode_emoji?(_), do: false
+
+  emoji_qualification_map =
+    emojis
+    |> Enum.filter(&String.contains?(&1, "\uFE0F"))
+    |> Combinations.variate_emoji_qualification()
+
+  for {qualified, unqualified_list} <- emoji_qualification_map do
+    for unqualified <- unqualified_list do
+      def fully_qualify_emoji(unquote(unqualified)), do: unquote(qualified)
+    end
+  end
+
+  def fully_qualify_emoji(emoji), do: emoji
 end

lib/pleroma/emoji/combinations.ex

@@ -0,0 +1,45 @@
+# Pleroma: A lightweight social networking server
+# Copyright © 2017-2022 Pleroma Authors <https://pleroma.social/>
+# SPDX-License-Identifier: AGPL-3.0-only
+
+defmodule Pleroma.Emoji.Combinations do
+  # FE0F is the emoji variation sequence. It is used for fully-qualifying
+  # emoji, and that includes emoji combinations.
+  # This code generates combinations per emoji: for each FE0F, all possible
+  # combinations of the character being removed or staying will be generated.
+  # This is made as an attempt to find all partially-qualified and unqualified
+  # versions of a fully-qualified emoji.
+  # I have found *no cases* for which this would be a problem, after browsing
+  # the entire emoji list in emoji-test.txt. This is safe, and, sadly, most
+  # likely sane too.
+
+  defp qualification_combinations(codepoints) do
+    qualification_combinations([[]], codepoints)
+  end
+
+  defp qualification_combinations(acc, []), do: acc
+
+  defp qualification_combinations(acc, ["\uFE0F" | tail]) do
+    acc
+    |> Enum.flat_map(fn x -> [x, x ++ ["\uFE0F"]] end)
+    |> qualification_combinations(tail)
+  end
+
+  defp qualification_combinations(acc, [codepoint | tail]) do
+    acc
+    |> Enum.map(&Kernel.++(&1, [codepoint]))
+    |> qualification_combinations(tail)
+  end
+
+  def variate_emoji_qualification(emoji) when is_binary(emoji) do
+    emoji
+    |> String.codepoints()
+    |> qualification_combinations()
+    |> Enum.map(&List.to_string/1)
+  end
+
+  def variate_emoji_qualification(emoji) when is_list(emoji) do
+    emoji
+    |> Enum.map(fn emoji -> {emoji, variate_emoji_qualification(emoji)} end)
+  end
+end

lib/pleroma/web/activity_pub/object_validators/emoji_react_validator.ex

@@ -63,8 +63,7 @@ defmodule Pleroma.Web.ActivityPub.ObjectValidators.EmojiReactValidator do
   end
 
   defp fix_emoji_qualification(%{"content" => emoji} = data) do
-    # Emoji variation sequence
-    new_emoji = emoji <> "\uFE0F"
+    new_emoji = Pleroma.Emoji.fully_qualify_emoji(emoji)
 
     cond do
       Pleroma.Emoji.is_unicode_emoji?(emoji) ->