From 54d76d6b56d90c17bc0ee8ab322fa05367018315 Mon Sep 17 00:00:00 2001
From: Alex Gleason <alex@alexgleason.me>
Date: Sun, 10 Apr 2022 15:25:07 -0500
Subject: [PATCH] Move emoji utils into its own module

---
 app/soapbox/components/ui/emoji/emoji.tsx | 28 +---------------
 app/soapbox/utils/__tests__/emoji.test.ts | 39 +++++++++++++++++++++++
 app/soapbox/utils/emoji.ts                | 35 ++++++++++++++++++++
 3 files changed, 75 insertions(+), 27 deletions(-)
 create mode 100644 app/soapbox/utils/__tests__/emoji.test.ts
 create mode 100644 app/soapbox/utils/emoji.ts

diff --git a/app/soapbox/components/ui/emoji/emoji.tsx b/app/soapbox/components/ui/emoji/emoji.tsx
index e9bcda2ed..59d00df2d 100644
--- a/app/soapbox/components/ui/emoji/emoji.tsx
+++ b/app/soapbox/components/ui/emoji/emoji.tsx
@@ -1,34 +1,8 @@
 import React from 'react';
 
+import { removeVS16s, toCodePoints } from 'soapbox/utils/emoji';
 import { joinPublicPath } from 'soapbox/utils/static';
 
-// Taken from twemoji-parser
-// https://github.com/twitter/twemoji-parser/blob/a97ef3994e4b88316812926844d51c296e889f76/src/index.js
-const removeVS16s = (rawEmoji: string): string => {
-  const vs16RegExp = /\uFE0F/g;
-  const zeroWidthJoiner = String.fromCharCode(0x200d);
-  return rawEmoji.indexOf(zeroWidthJoiner) < 0 ? rawEmoji.replace(vs16RegExp, '') : rawEmoji;
-};
-
-const toCodePoints = (unicodeSurrogates: string): string[] => {
-  const points = [];
-  let char = 0;
-  let previous = 0;
-  let i = 0;
-  while (i < unicodeSurrogates.length) {
-    char = unicodeSurrogates.charCodeAt(i++);
-    if (previous) {
-      points.push((0x10000 + ((previous - 0xd800) << 10) + (char - 0xdc00)).toString(16));
-      previous = 0;
-    } else if (char > 0xd800 && char <= 0xdbff) {
-      previous = char;
-    } else {
-      points.push(char.toString(16));
-    }
-  }
-  return points;
-};
-
 interface IEmoji extends React.ImgHTMLAttributes<HTMLImageElement> {
   emoji: string,
 }
diff --git a/app/soapbox/utils/__tests__/emoji.test.ts b/app/soapbox/utils/__tests__/emoji.test.ts
new file mode 100644
index 000000000..ab1a4ddaa
--- /dev/null
+++ b/app/soapbox/utils/__tests__/emoji.test.ts
@@ -0,0 +1,39 @@
+import {
+  removeVS16s,
+  toCodePoints,
+} from '../emoji';
+
+const ASCII_HEART = '❤'; // '\u2764\uFE0F'
+const RED_HEART_RGI = '❤️'; // '\u2764'
+const JOY = '😂';
+
+describe('removeVS16s()', () => {
+  it('removes Variation Selector-16 characters from emoji', () => {
+    // Sanity check
+    expect(ASCII_HEART).not.toBe(RED_HEART_RGI);
+
+    // It normalizes an emoji with VS16s
+    expect(removeVS16s(RED_HEART_RGI)).toBe(ASCII_HEART);
+
+    // Leaves a regular emoji alone
+    expect(removeVS16s(JOY)).toBe(JOY);
+  });
+});
+
+describe('toCodePoints()', () => {
+  it('converts a plain emoji', () => {
+    expect(toCodePoints('😂')).toEqual(['1f602']);
+  });
+
+  it('converts a VS16 emoji', () => {
+    expect(toCodePoints(RED_HEART_RGI)).toEqual(['2764', 'fe0f']);
+  });
+
+  it('converts an ASCII character', () => {
+    expect(toCodePoints(ASCII_HEART)).toEqual(['2764']);
+  });
+
+  it('converts a sequence emoji', () => {
+    expect(toCodePoints('🇺🇸')).toEqual(['1f1fa', '1f1f8']);
+  });
+});
diff --git a/app/soapbox/utils/emoji.ts b/app/soapbox/utils/emoji.ts
new file mode 100644
index 000000000..1d6da69d1
--- /dev/null
+++ b/app/soapbox/utils/emoji.ts
@@ -0,0 +1,35 @@
+// Taken from twemoji-parser
+// https://github.com/twitter/twemoji-parser/blob/a97ef3994e4b88316812926844d51c296e889f76/src/index.js
+
+/** Remove Variation Selector-16 characters from emoji */
+// https://emojipedia.org/variation-selector-16/
+const removeVS16s = (rawEmoji: string): string => {
+  const vs16RegExp = /\uFE0F/g;
+  const zeroWidthJoiner = String.fromCharCode(0x200d);
+  return rawEmoji.indexOf(zeroWidthJoiner) < 0 ? rawEmoji.replace(vs16RegExp, '') : rawEmoji;
+};
+
+/** Convert emoji into an array of Unicode codepoints */
+const toCodePoints = (unicodeSurrogates: string): string[] => {
+  const points = [];
+  let char = 0;
+  let previous = 0;
+  let i = 0;
+  while (i < unicodeSurrogates.length) {
+    char = unicodeSurrogates.charCodeAt(i++);
+    if (previous) {
+      points.push((0x10000 + ((previous - 0xd800) << 10) + (char - 0xdc00)).toString(16));
+      previous = 0;
+    } else if (char > 0xd800 && char <= 0xdbff) {
+      previous = char;
+    } else {
+      points.push(char.toString(16));
+    }
+  }
+  return points;
+};
+
+export {
+  removeVS16s,
+  toCodePoints,
+};