mirror of
https://github.com/renovatebot/renovate.git
synced 2025-01-12 23:16:26 +00:00
119 lines
3.3 KiB
TypeScript
119 lines
3.3 KiB
TypeScript
import is from '@sindresorhus/is';
|
||
import mathiasBynensEmojiRegex from 'emoji-regex';
|
||
import {
|
||
fromCodepointToUnicode,
|
||
fromHexcodeToCodepoint,
|
||
fromUnicodeToHexcode,
|
||
} from 'emojibase';
|
||
import emojibaseEmojiRegex from 'emojibase-regex/emoji.js';
|
||
import SHORTCODE_REGEX from 'emojibase-regex/shortcode.js';
|
||
import { z } from 'zod';
|
||
import type { RenovateConfig } from '../config/types';
|
||
import dataFiles from '../data-files.generated';
|
||
import { logger } from '../logger';
|
||
import { regEx } from './regex';
|
||
import { Json } from './schema-utils';
|
||
|
||
let unicodeEmoji = true;
|
||
|
||
let mappingsInitialized = false;
|
||
const shortCodesByHex = new Map<string, string>();
|
||
const hexCodesByShort = new Map<string, string>();
|
||
|
||
const EmojiShortcodesSchema = Json.pipe(
|
||
z.record(z.string(), z.union([z.string(), z.array(z.string())])),
|
||
);
|
||
|
||
function lazyInitMappings(): void {
|
||
if (!mappingsInitialized) {
|
||
const result = EmojiShortcodesSchema.safeParse(
|
||
dataFiles.get('node_modules/emojibase-data/en/shortcodes/github.json')!,
|
||
);
|
||
// istanbul ignore if: not easily testable
|
||
if (!result.success) {
|
||
logger.warn({ error: result.error }, 'Unable to parse emoji shortcodes');
|
||
return;
|
||
}
|
||
for (const [hex, val] of Object.entries(result.data)) {
|
||
const shortCodes = is.array(val) ? val : [val];
|
||
shortCodesByHex.set(hex, `:${shortCodes[0]}:`);
|
||
shortCodes.forEach((shortCode) => {
|
||
hexCodesByShort.set(`:${shortCode}:`, hex);
|
||
});
|
||
}
|
||
mappingsInitialized = true;
|
||
}
|
||
}
|
||
|
||
export function setEmojiConfig(config: RenovateConfig): void {
|
||
unicodeEmoji = !!config.unicodeEmoji;
|
||
}
|
||
|
||
const shortCodeRegex = regEx(SHORTCODE_REGEX.source, 'g');
|
||
|
||
export function emojify(text: string): string {
|
||
if (!unicodeEmoji) {
|
||
return text;
|
||
}
|
||
lazyInitMappings();
|
||
return text.replace(shortCodeRegex, (shortCode) => {
|
||
const hexCode = hexCodesByShort.get(shortCode);
|
||
return hexCode
|
||
? fromCodepointToUnicode(fromHexcodeToCodepoint(hexCode))
|
||
: shortCode;
|
||
});
|
||
}
|
||
|
||
const emojiRegexSrc = [emojibaseEmojiRegex, mathiasBynensEmojiRegex()].map(
|
||
({ source }) => source,
|
||
);
|
||
const emojiRegex = new RegExp(`(?:${emojiRegexSrc.join('|')})`, 'g'); // TODO #12875 cannot figure it out
|
||
const excludedModifiers = new Set([
|
||
'20E3',
|
||
'200D',
|
||
'FE0E',
|
||
'FE0F',
|
||
'1F3FB',
|
||
'1F3FC',
|
||
'1F3FD',
|
||
'1F3FE',
|
||
'1F3FF',
|
||
'1F9B0',
|
||
'1F9B1',
|
||
'1F9B2',
|
||
'1F9B3',
|
||
]);
|
||
|
||
export function stripHexCode(input: string): string {
|
||
return input
|
||
.split('-')
|
||
.filter((modifier) => !excludedModifiers.has(modifier))
|
||
.join('-');
|
||
}
|
||
|
||
export function unemojify(text: string): string {
|
||
if (unicodeEmoji) {
|
||
return text;
|
||
}
|
||
lazyInitMappings();
|
||
return text.replace(emojiRegex, (emoji) => {
|
||
const hexCode = stripHexCode(fromUnicodeToHexcode(emoji));
|
||
const shortCode = shortCodesByHex.get(hexCode);
|
||
return shortCode ?? '<27>';
|
||
});
|
||
}
|
||
|
||
function stripEmoji(emoji: string): string {
|
||
const hexCode = stripHexCode(fromUnicodeToHexcode(emoji));
|
||
// `hexCode` could be empty if `emoji` is a modifier character that isn't
|
||
// modifying anything
|
||
if (hexCode) {
|
||
const codePoint = fromHexcodeToCodepoint(hexCode);
|
||
return fromCodepointToUnicode(codePoint);
|
||
}
|
||
return '';
|
||
}
|
||
|
||
export function stripEmojis(input: string): string {
|
||
return input.replace(emojiRegex, stripEmoji);
|
||
}
|