path: root/util/japanese.ts

import { UnicodeRange } from "./string.ts";
import "./number.ts";

declare global {
	interface String {
		/**
		 * @summary check if string is hiragana only
		 *
		 * @argument strict  don't allow ascii whitespace and punctuation (default: false)
		 *
		 * return `true` if at least one hiragana character is in string, and no other
		 * unicode ranges are found. ascii whitespace and punctuation is still allowed,
		 * but not counted as hiragana. this behavior can be turned off by setting
		 * `strict` to true
		 */
		hiraganaOnly(strict?: boolean): boolean
		/**
		 * @summary check if string is katakana only
		 *
		 * @argument strict  don't allow ascii whitespace and punctuation (default: false)
		 *
		 * return `true` if at least one katakana character is in string, and no other
		 * unicode ranges are found. ascii whitespace and punctuation is still allowed,
		 * but not counted as katakana. this behavior can be turned off by setting
		 * `strict` to true
		 */
		katakanaOnly(strict?: boolean): boolean
		/**
		 * @summary check if string is kanji only
		 *
		 * @argument strict  don't allow ascii whitespace and punctuation (default: false)
		 *
		 * return `true` if at least one kanji character is in string, and no other
		 * unicode ranges are found. ascii whitespace and punctuation is still allowed,
		 * but not counted as kanji. this behavior can be turned off by setting
		 * `strict` to true
		 */
		kanjiOnly(strict?: boolean): boolean
		/**
		 * @summary check if string is kana only
		 *
		 * @argument strict  don't allow ascii whitespace and punctuation (default: false)
		 *
		 * return `true` if at least one kana character is in string, and no other
		 * unicode ranges are found. ascii whitespace and punctuation is still allowed,
		 * but not counted as kana. this behavior can be turned off by setting `strict`
		 * to true
		 */
		kanaOnly(strict?: boolean): boolean
		/**
		 * @summary check if string is japanese only
		 *
		 * @argument strict  don't allow ascii whitespace and punctuation (default: false)
		 *
		 * return `true` if at least one japanese character is in string, and no other
		 * unicode ranges are found. ascii whitespace and punctuation is still allowed,
		 * but not counted as japanese. this behavior can be turned off by setting
		 * `strict` to true
		 */
		japaneseOnly(strict?: boolean): boolean

		/** @summary check if the string contains kanji characters */
		hasKanji(): boolean;

		/** @summary convert any half-width katakana to full-width */
		widenKatakana(): string;

		/** @summary convert any full-width katakana to hiragana */
		katakanaToHiragana(): string;

		/** @summary convert any kana (full and half-width) to full-width hiragana */
		normalizeKana(): string;
	}
}

enum StringOnlyReturnValue {
	TallyAdd,
	TallyIgnore,
	TallyStop,
}

/** @summary check tally for allowed scripts (internal use only) */
function stringOnly(input: string, check: (key: string, val: number) => StringOnlyReturnValue): boolean {
	var tally = input.rangeTally();
	var ok = false;
	for (var [key, val] of Object.entries(tally)) {
		switch(check(key, val)) {
			case StringOnlyReturnValue.TallyAdd: {
				ok = true;
				break;
			}
			case StringOnlyReturnValue.TallyIgnore: { break; }
			case StringOnlyReturnValue.TallyStop: { return false; }
		}
	}
	return ok;
}

String.prototype.hiraganaOnly = function(strict = false) {
	return stringOnly(this as string, (key, val) => {
		if (key == UnicodeRange.JapaneseFWHiragana)
			return StringOnlyReturnValue.TallyAdd; // count hiragana characters
		else if (!strict && key.startsWith("any-"))
			return StringOnlyReturnValue.TallyIgnore; // allow any- (ascii whitespace and punctuation)
		else if (val > 0)
			return StringOnlyReturnValue.TallyStop; // don't allow any other ranges
		return StringOnlyReturnValue.TallyIgnore;
	});
}

String.prototype.katakanaOnly = function(strict = false) {
	return stringOnly(this as string, (key, val) => {
		if ([UnicodeRange.JapaneseHWKatakana, UnicodeRange.JapaneseFWKatakana].includes(key as UnicodeRange))
			return StringOnlyReturnValue.TallyAdd; // count katakana characters
		else if (!strict && key.startsWith("any-"))
			return StringOnlyReturnValue.TallyIgnore; // allow any- (ascii whitespace and punctuation)
		else if (val > 0)
			return StringOnlyReturnValue.TallyStop; // don't allow any other ranges
		return StringOnlyReturnValue.TallyIgnore;
	});
}

String.prototype.kanjiOnly = function(strict = false) {
	return stringOnly(this as string, (key, val) => {
		if (key == UnicodeRange.JapaneseKanji)
			return StringOnlyReturnValue.TallyAdd; // count kanji characters
		else if (!strict && key.startsWith("any-"))
			return StringOnlyReturnValue.TallyIgnore; // allow any- (ascii whitespace and punctuation)
		else if (val > 0)
			return StringOnlyReturnValue.TallyStop; // don't allow any other ranges
		return StringOnlyReturnValue.TallyIgnore;
	});
}

String.prototype.kanaOnly = function(strict = false) {
	return stringOnly(this as string, (key, val) => {
		if ([UnicodeRange.JapaneseHWKatakana, UnicodeRange.JapaneseFWKatakana, UnicodeRange.JapaneseFWHiragana].includes(key as UnicodeRange))
			return StringOnlyReturnValue.TallyAdd; // count kana characters
		else if (!strict && key.startsWith("any-"))
			return StringOnlyReturnValue.TallyIgnore; // allow any- (ascii whitespace and punctuation)
		else if (val > 0)
			return StringOnlyReturnValue.TallyStop; // don't allow any other ranges
		return StringOnlyReturnValue.TallyIgnore;
	});
}

String.prototype.japaneseOnly = function(strict = false) {
	return stringOnly(this as string, (key, val) => {
		if (key.startsWith("jp-"))
			return StringOnlyReturnValue.TallyAdd; // count japanese characters
		else if (!strict && key.startsWith("any-"))
			return StringOnlyReturnValue.TallyIgnore; // allow any- (ascii whitespace and punctuation)
		else if (val > 0)
			return StringOnlyReturnValue.TallyStop; // don't allow any other ranges
		return StringOnlyReturnValue.TallyIgnore;
	});
}

String.prototype.widenKatakana = function() {
	const map: { [key: string]: string } = {
		"ｧ": "ァ",
		"ｱ": "ア",
		"ｨ": "ィ",
		"ｲ": "イ",
		"ｩ": "ゥ",
		"ｳ": "ウ",
		"ｪ": "ェ",
		"ｴ": "エ",
		"ｫ": "ォ",
		"ｵ": "オ",
		"ｶﾞ": "ガ",
		"ｶ": "カ",
		"ｷﾞ": "ギ",
		"ｷ": "キ",
		"ｸﾞ": "グ",
		"ｸ": "ク",
		"ｹﾞ": "ゲ",
		"ｹ": "ケ",
		"ｺﾞ": "ゴ",
		"ｺ": "コ",
		"ｻﾞ": "ザ",
		"ｻ": "サ",
		"ｼﾞ": "ジ",
		"ｼ": "シ",
		"ｽﾞ": "ズ",
		"ｽ": "ス",
		"ｾﾞ": "ゼ",
		"ｾ": "セ",
		"ｿﾞ": "ゾ",
		"ｿ": "ソ",
		"ﾀﾞ": "ダ",
		"ﾀ": "タ",
		"ﾁﾞ": "ヂ",
		"ﾁ": "チ",
		"ﾂﾞ": "ヅ",
		"ｯ": "ッ",
		"ﾂ": "ツ",
		"ﾃﾞ": "デ",
		"ﾃ": "テ",
		"ﾄﾞ": "ド",
		"ﾄ": "ト",
		"ﾅ": "ナ",
		"ﾆ": "ニ",
		"ﾇ": "ヌ",
		"ﾈ": "ネ",
		"ﾉ": "ノ",
		"ﾊﾞ": "バ",
		"ﾊﾟ": "パ",
		"ﾊ": "ハ",
		"ﾋﾞ": "ビ",
		"ﾋﾟ": "ピ",
		"ﾋ": "ヒ",
		"ﾌﾞ": "ブ",
		"ﾌﾟ": "プ",
		"ﾌ": "フ",
		"ﾍﾞ": "ベ",
		"ﾍﾟ": "ペ",
		"ﾍ": "ヘ",
		"ﾎﾞ": "ボ",
		"ﾎﾟ": "ポ",
		"ﾎ": "ホ",
		"ﾏ": "マ",
		"ﾐ": "ミ",
		"ﾑ": "ム",
		"ﾒ": "メ",
		"ﾓ": "モ",
		"ｬ": "ャ",
		"ﾔ": "ヤ",
		"ｭ": "ュ",
		"ﾕ": "ユ",
		"ｮ": "ョ",
		"ﾖ": "ヨ",
		"ﾗ": "ラ",
		"ﾘ": "リ",
		"ﾙ": "ル",
		"ﾚ": "レ",
		"ﾛ": "ロ",
		"ﾜ": "ワ",
		"ｦ": "ヲ",
		"ﾝ": "ン",
		"ｳﾞ": "ヴ",
		"ﾜﾞ": "ヷ",
		"ｲﾞ": "イ゙",
		"ｴﾞ": "エ゙",
		"ｦﾞ": "ヺ",
		"ｰ": "ー",
		" ": "　",
	};

	var out = "";
	outer:
	for (let i = 0; i < this.length; i++) {
		for (var key in map) {
			if (!this.substring(i).startsWith(key)) continue;
			out += map[key];
			i += key.length - 1;
			continue outer;
		}
		
		out += this[i];
	}
	return out;
}

String.prototype.katakanaToHiragana = function() {
	return this.map(char => {
		var code = char.codePointAt(0)!;
		if (0x30a1 <= code && code <= 0x30f6) return (code + (0x3041 - 0x30a1)).toChar();
		return char;
	})
}

String.prototype.normalizeKana = function() {
	return this.widenKatakana().katakanaToHiragana();
}

String.prototype.hasKanji = function() {
	for (var c of this)
		if ([UnicodeRange.JapaneseKanji].includes(c.range()))
			return true;
	return false;
}