Emoji: split qualification variation into a module
This commit is contained in:
parent
fb3f6e1975
commit
b99f5d6183
@ -9,6 +9,7 @@ defmodule Pleroma.Emoji do
|
||||
"""
|
||||
use GenServer
|
||||
|
||||
alias Pleroma.Emoji.Combinations
|
||||
alias Pleroma.Emoji.Loader
|
||||
|
||||
require Logger
|
||||
@ -138,42 +139,10 @@ defmodule Pleroma.Emoji do
|
||||
|
||||
def is_unicode_emoji?(_), do: false
|
||||
|
||||
# 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.
|
||||
emoji_qualification_map =
|
||||
emojis
|
||||
|> Enum.filter(&String.contains?(&1, "\uFE0F"))
|
||||
|> Enum.map(fn emoji ->
|
||||
combinate = fn x, combinate ->
|
||||
case x do
|
||||
[] ->
|
||||
[[]]
|
||||
|
||||
["\uFE0F" | tail] ->
|
||||
combinate.(tail, combinate)
|
||||
|> Enum.flat_map(fn x -> [x, ["\uFE0F" | x]] end)
|
||||
|
||||
[codepoint | tail] ->
|
||||
combinate.(tail, combinate)
|
||||
|> Enum.map(fn x -> [codepoint | x] end)
|
||||
end
|
||||
end
|
||||
|
||||
unqualified_list =
|
||||
emoji
|
||||
|> String.codepoints()
|
||||
|> combinate.(combinate)
|
||||
|> Enum.map(&List.to_string/1)
|
||||
|
||||
{emoji, unqualified_list}
|
||||
end)
|
||||
|> Combinations.variate_emoji_qualification()
|
||||
|
||||
for {qualified, unqualified_list} <- emoji_qualification_map do
|
||||
for unqualified <- unqualified_list do
|
||||
|
41
lib/pleroma/emoji/combinations.ex
Normal file
41
lib/pleroma/emoji/combinations.ex
Normal file
@ -0,0 +1,41 @@
|
||||
# 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([]), do: [[]]
|
||||
|
||||
defp qualification_combinations(["\uFE0F" | tail]) do
|
||||
tail
|
||||
|> qualification_combinations()
|
||||
|> Enum.flat_map(fn x -> [x, ["\uFE0F" | x]] end)
|
||||
end
|
||||
|
||||
defp qualification_combinations([codepoint | tail]) do
|
||||
tail
|
||||
|> qualification_combinations()
|
||||
|> Enum.map(fn x -> [codepoint | x] end)
|
||||
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
|
Loading…
Reference in New Issue
Block a user