taraskevizer 0.0.2 → 1.0.0

package/src/index.ts

@@ -1,3 +0,0 @@
-export * from './types';
-export { tarask, taraskSync } from './tarask';
-export { gobj } from './dict';

package/src/tarask.ts

@@ -1,258 +0,0 @@
-import {
-	wordlist,
-	softers,
-	arabLetters,
-	latinLetters,
-	latinLettersUpperCase,
-	gobj,
-} from './dict';
-import {
-	Tarask,
-	TaraskAsync,
-	Dict,
-	AlphabetDependentDict,
-	TaraskOptions,
-} from './types';
-import * as debug from './tools.debug';
-
-const isUpperCase = (str: string): boolean => str === str.toUpperCase();
-
-const NOFIX_CHAR = ' \uffff ';
-const NOFIX_REGEX = new RegExp(NOFIX_CHAR, 'g');
-const OPTIONAL_WORDS_REGEX = /\(.*?\)/g;
-const G_REGEX = /[Ґґ]/g;
-
-const CYRILLIC = 0,
-	LATIN = 1,
-	ARABIC = 2;
-const J_NEVER = 0,
-	J_RANDOM = 1,
-	J_ALWAYS = 2;
-
-const letters = {
-	[LATIN]: latinLetters,
-	[ARABIC]: arabLetters,
-} satisfies AlphabetDependentDict;
-const lettersUpperCase = {
-	[LATIN]: latinLettersUpperCase,
-} satisfies AlphabetDependentDict;
-const additionalReplacements = {
-	[CYRILLIC]: {
-		$1У: /([АЕЁІОУЫЭЮЯ])<tarF>Ў<\/tarF>/g,
-		' У': / <tarF>Ў<\/tarF>(?=\p{Lu})/gu,
-	},
-	[LATIN]: {
-		$1U: /([AEIOUY])<tarF>Ŭ<\/tarF>/g,
-		' U': / <tarF>Ŭ<\/tarF>(?=\p{Lu})/gu,
-	},
-	[ARABIC]: {},
-} satisfies AlphabetDependentDict;
-
-export const taraskSync: Tarask = (text, { abc = 0, j = 0, html }) => {
-	const noFix: string[] = [];
-
-	const LEFT_ANGLE_BRACKET = html ? '&lt;' : '<';
-
-	text = ` ${text.trim()}  `
-		.replace(/<([,.]?)((?:.|\s)*?)>/g, ($0, $1, $2) => {
-			if ($1 === ',') return LEFT_ANGLE_BRACKET + $2 + '>';
-			noFix[noFix.length] = $1 === '.' ? $2 : $0;
-			return NOFIX_CHAR;
-		})
-		.replace(/г'(?![еёіюя])/g, 'ґ')
-		.replace(/(\n|\t)/g, ' $1 ')
-		.replace(/ - /g, ' — ')
-		.replace(/(\p{P}|\p{S}|\d)/gu, ' $1 ')
-		.replace(/ ['`’] (?=\S)/g, 'ʼ')
-		.replace(/\(/g, '&#40');
-
-	let splittedOrig: string[], splitted: string[];
-	splittedOrig = replaceWithDict(
-		replaceWithDict(text, letters[abc]),
-		lettersUpperCase[abc]
-	).split(' ');
-
-	text = toTarask(text.toLowerCase());
-	if (j) text = replaceIbyJ(text, j === J_ALWAYS);
-	text = replaceWithDict(text, letters[abc]);
-
-	splitted = text.split(' ');
-	if (abc !== ARABIC) splitted = restoreCase(splitted, splittedOrig);
-	if (html) splitted = toHtmlTags(splitted, splittedOrig, abc);
-
-	text = splitted
-		.join(' ')
-		.replace(/&nbsp;/g, ' ')
-		.replace(/ (\p{P}|\p{S}|\d|&#40) /gu, '$1');
-
-	if (html) {
-		text = replaceWithDict(text, additionalReplacements[abc]);
-		if (abc === CYRILLIC)
-			text = text.replace(
-				G_REGEX,
-				// @ts-ignore
-				html.g ? '<tarH>$&</tarH>' : ($0) => `<tarH>${gobj[$0]}</tarH>`
-			);
-	}
-
-	if (noFix.length) text = text.replace(NOFIX_REGEX, () => noFix.shift());
-
-	return (html ? finalizer.html : finalizer.text)(text).trim();
-};
-
-export const tarask: TaraskAsync = (...args) =>
-	new Promise((res) => res(taraskSync(...args)));
-
-function restoreCase(text: string[], orig: string[]): string[] {
-	for (let i = 0; i < text.length; i++) {
-		const word = text[i];
-		const oWord = orig[i];
-		if (word === oWord) continue;
-		if (word === oWord.toLowerCase()) {
-			text[i] = oWord;
-			continue;
-		}
-		if (!oWord[0] || !isUpperCase(oWord[0])) continue;
-		if (word === 'зь') {
-			text[i] = isUpperCase(orig[i + 1]) ? 'ЗЬ' : 'Зь';
-		} else if (isUpperCase(oWord[oWord.length - 1])) {
-			text[i] = word.toUpperCase();
-		} else {
-			text[i] =
-				word[0] === '('
-					? word.replace(/.*?(?=\))/, ($0) =>
-							$0.replace(/[(|]./g, ($0) => $0.toUpperCase())
-					  )
-					: word[0].toUpperCase() + word.slice(1);
-		}
-	}
-
-	return text;
-}
-
-function toHtmlTags(
-	text: string[],
-	orig: string[],
-	abc: TaraskOptions['abc']
-): string[] {
-	for (let i = 0; i < text.length; i++) {
-		const word = text[i];
-		const oWord = orig[i];
-		if (oWord === word) continue;
-		const wordH = word.replace(G_REGEX, ($0) => gobj[$0]);
-		if (oWord === wordH) continue;
-		if (!/\(/.test(word)) {
-			if (word.length === oWord.length) {
-				const LettersText = word.split('');
-				for (let x = 0; x < LettersText.length; x++) {
-					if (LettersText[x] !== oWord[x])
-						LettersText[x] = `<tarF>${LettersText[x]}</tarF>`;
-				}
-				text[i] = LettersText.join('');
-				continue;
-			}
-			if (abc === CYRILLIC) {
-				const word1 = word.replace(/ь/g, '');
-				switch (oWord) {
-					case word1:
-						text[i] = word.replace(/ь/g, '<tarF>ь</tarF>');
-						continue;
-					case word1 + 'ь':
-						text[i] = word.slice(0, -1).replace(/ь/g, '<tarF>ь</tarF>') + 'ь';
-						continue;
-				}
-			}
-		}
-
-		const oWordEnd = oWord.length - 1;
-		let fromStart = 0;
-		let fromWordEnd = word.length - 1;
-		let fromOWordEnd = oWordEnd;
-
-		while (wordH[fromStart] === oWord[fromStart]) ++fromStart;
-		while (wordH[fromWordEnd] === oWord[fromOWordEnd])
-			--fromWordEnd, --fromOWordEnd;
-
-		if (oWord.length < word.length) {
-			if (fromOWordEnd === oWordEnd) {
-				text[i] = `<tarF>${word}</tarF>`;
-				continue;
-			}
-			if (fromWordEnd < 0) fromWordEnd = 0;
-		}
-
-		text[i] =
-			word.slice(0, fromStart) +
-			'<tarF>' +
-			word.slice(fromStart, fromWordEnd + 1) +
-			'</tarF>' +
-			word.slice(fromWordEnd + 1);
-	}
-
-	return text;
-}
-
-function toTarask(text: string): string {
-	text = replaceWithDict(text, wordlist);
-	loop: do {
-		text = replaceWithDict(text, softers);
-		for (const key in softers)
-			if (key !== '$1дзьдз' && softers[key].test(text)) continue loop;
-		break;
-	} while (true);
-
-	const iaReplacer = <TStart extends ' б' | ' н', T extends string>(
-		$0: `${TStart}е${T}`,
-		$1: TStart,
-		$2: T
-	) => ($2.match(/[аеёіоуыэюя]/g)?.length === 1 ? $1 + 'я' + $2 : $0);
-
-	return text
-		.replace(/ [уў]асьнігл /g, ' уаснігл ')
-		.replace(/ сь(?=нід |мі )/g, ' с')
-		.replace(/( б)е(зь? \S+)/g, iaReplacer)
-		.replace(/( н)е( \S+)/g, iaReplacer)
-		.replace(/( (?:б[ея]|пра|цера)?з) і(\S*)/g, ($0, $1, $2) =>
-			/([ая]ў|ну)$/.test($2) ? $1 + 'ь і' + $2 : $0
-		);
-}
-
-function replaceWithDict(text: string, dict: Dict = null): string {
-	for (const key in dict) text = text.replace(dict[key], key);
-
-	return text;
-}
-
-type Vow = 'а' | 'е' | 'ё' | 'і' | 'о' | 'у' | 'ы' | 'э' | 'ю' | 'я';
-
-type ToJ = <TVow extends `${Vow} `, TU extends '' | 'ў'>(
-	vow: TVow,
-	shortU: TU
-) => `${TVow}й ${TU extends 'ў' ? 'у' : ''}`;
-
-const toJ: ToJ = (vow, shortU) =>
-	(vow + 'й ' + (shortU ? 'у' : '')) as ReturnType<ToJ>;
-
-function replaceIbyJ(text: string, always = false): string {
-	return text.replace(
-		/([аеёіоуыэюя] )і (ў?)/g,
-		always
-			? ($0, $1, $2) => toJ($1, $2)
-			: ($0, $1, $2) => (Math.random() >= 0.5 ? toJ($1, $2) : $0)
-	);
-}
-
-const finalizer = {
-	html: (text) =>
-		text
-			.replace(OPTIONAL_WORDS_REGEX, ($0) => {
-				const options = $0.slice(1, -1).split('|');
-				const main = options.shift();
-				return `<tarL data-l='${options}'>${main}</tarL>`;
-			})
-			.replace(/ \n /g, '<br>'),
-	text: (text) =>
-		text
-			.replace(OPTIONAL_WORDS_REGEX, ($0) => $0.slice(1, -1).split('|')[0])
-			.replace(/&#40/g, '('),
-} satisfies Record<string, (text: string) => string>;

package/src/tools.debug.ts

@@ -1,20 +0,0 @@
-import { Dict } from './types';
-type LogFn = (...args: any[]) => void;
-declare const console: { log: LogFn };
-
-export function replaceWithDict(
-	text: string,
-	dict: Dict = null,
-	regex: RegExp
-): string {
-	for (const key in dict) {
-		text = text.replace(dict[key], key);
-		if (regex.test(text)) throw `'${key}': /${dict[key]}/`;
-	}
-
-	return text;
-}
-
-export const log: LogFn = (...msgs) => {
-	console.log(...msgs);
-};

package/src/tsconfig.json

@@ -1,5 +0,0 @@
-{
-	"compilerOptions": {
-		"lib": ["es6"]
-	}
-}

package/src/types.ts

@@ -1,15 +0,0 @@
-type Alphabet = 0 | 1 | 2;
-type J = 0 | 1 | 2;
-export type HtmlOptions = { g: boolean };
-export type TaraskOptions = {
-	abc: Alphabet;
-	j: J;
-	html: false | HtmlOptions;
-};
-type Promisify<T> = T extends (...args: infer TArgs) => infer TReturn
-	? (...args: TArgs) => Promise<TReturn>
-	: never;
-export type Tarask = (text: string, options: TaraskOptions) => string;
-export type TaraskAsync = Promisify<Tarask>;
-export type Dict = Record<string, RegExp>;
-export type AlphabetDependentDict = { [key in Alphabet]?: Dict };

package/tsup.config.js

@@ -1,24 +0,0 @@
-import { defineConfig } from 'tsup';
-import path from 'path';
-import { readFile, writeFile } from 'fs/promises';
-import postprocess from './postprocess';
-import noDebugFiles from './esbuild-plugins/no-debug-files';
-
-export default defineConfig({
-	entry: ['src/index.ts'],
-	clean: true,
-	sourcemap: false,
-	splitting: false,
-	minify: false,
-	format: ['cjs', 'esm'],
-	dts: true,
-	esbuildPlugins: [noDebugFiles],
-	async onSuccess() {
-		for (const ext of ['js', 'mjs']) {
-			const filePath = path.resolve(this.outDir, 'index.' + ext);
-			readFile(filePath, 'utf8').then((text) =>
-				writeFile(filePath, postprocess(text, ext))
-			);
-		}
-	},
-});