2017-07-19 06:07:46 +00:00
|
|
|
|
/*
|
2020-01-01 17:00:00 +00:00
|
|
|
|
* Copyright (C) 2016-2020 Alex Yatskov <alex@foosoft.net>
|
2017-07-19 06:07:46 +00:00
|
|
|
|
* Author: Alex Yatskov <alex@foosoft.net>
|
|
|
|
|
*
|
|
|
|
|
* This program is free software: you can redistribute it and/or modify
|
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
|
* the Free Software Foundation, either version 3 of the License, or
|
|
|
|
|
* (at your option) any later version.
|
|
|
|
|
*
|
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
2017-08-26 19:21:28 +00:00
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
2017-07-19 06:07:46 +00:00
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
|
*
|
|
|
|
|
* You should have received a copy of the GNU General Public License
|
2020-01-01 17:00:31 +00:00
|
|
|
|
* along with this program. If not, see <https://www.gnu.org/licenses/>.
|
2017-07-19 06:07:46 +00:00
|
|
|
|
*/
|
|
|
|
|
|
2020-03-11 02:30:36 +00:00
|
|
|
|
/* global
|
2020-03-21 17:18:34 +00:00
|
|
|
|
* jp
|
2020-03-11 02:30:36 +00:00
|
|
|
|
* wanakana
|
|
|
|
|
*/
|
2017-07-19 06:07:46 +00:00
|
|
|
|
|
2020-03-21 17:18:34 +00:00
|
|
|
|
(() => {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
const HALFWIDTH_KATAKANA_MAPPING = new Map([
|
|
|
|
|
['ヲ', 'ヲヺ-'],
|
|
|
|
|
['ァ', 'ァ--'],
|
|
|
|
|
['ィ', 'ィ--'],
|
|
|
|
|
['ゥ', 'ゥ--'],
|
|
|
|
|
['ェ', 'ェ--'],
|
|
|
|
|
['ォ', 'ォ--'],
|
|
|
|
|
['ャ', 'ャ--'],
|
|
|
|
|
['ュ', 'ュ--'],
|
|
|
|
|
['ョ', 'ョ--'],
|
|
|
|
|
['ッ', 'ッ--'],
|
|
|
|
|
['ー', 'ー--'],
|
|
|
|
|
['ア', 'ア--'],
|
|
|
|
|
['イ', 'イ--'],
|
|
|
|
|
['ウ', 'ウヴ-'],
|
|
|
|
|
['エ', 'エ--'],
|
|
|
|
|
['オ', 'オ--'],
|
|
|
|
|
['カ', 'カガ-'],
|
|
|
|
|
['キ', 'キギ-'],
|
|
|
|
|
['ク', 'クグ-'],
|
|
|
|
|
['ケ', 'ケゲ-'],
|
|
|
|
|
['コ', 'コゴ-'],
|
|
|
|
|
['サ', 'サザ-'],
|
|
|
|
|
['シ', 'シジ-'],
|
|
|
|
|
['ス', 'スズ-'],
|
|
|
|
|
['セ', 'セゼ-'],
|
|
|
|
|
['ソ', 'ソゾ-'],
|
|
|
|
|
['タ', 'タダ-'],
|
|
|
|
|
['チ', 'チヂ-'],
|
|
|
|
|
['ツ', 'ツヅ-'],
|
|
|
|
|
['テ', 'テデ-'],
|
|
|
|
|
['ト', 'トド-'],
|
|
|
|
|
['ナ', 'ナ--'],
|
|
|
|
|
['ニ', 'ニ--'],
|
|
|
|
|
['ヌ', 'ヌ--'],
|
|
|
|
|
['ネ', 'ネ--'],
|
|
|
|
|
['ノ', 'ノ--'],
|
|
|
|
|
['ハ', 'ハバパ'],
|
|
|
|
|
['ヒ', 'ヒビピ'],
|
|
|
|
|
['フ', 'フブプ'],
|
|
|
|
|
['ヘ', 'ヘベペ'],
|
|
|
|
|
['ホ', 'ホボポ'],
|
|
|
|
|
['マ', 'マ--'],
|
|
|
|
|
['ミ', 'ミ--'],
|
|
|
|
|
['ム', 'ム--'],
|
|
|
|
|
['メ', 'メ--'],
|
|
|
|
|
['モ', 'モ--'],
|
|
|
|
|
['ヤ', 'ヤ--'],
|
|
|
|
|
['ユ', 'ユ--'],
|
|
|
|
|
['ヨ', 'ヨ--'],
|
|
|
|
|
['ラ', 'ラ--'],
|
|
|
|
|
['リ', 'リ--'],
|
|
|
|
|
['ル', 'ル--'],
|
|
|
|
|
['レ', 'レ--'],
|
|
|
|
|
['ロ', 'ロ--'],
|
|
|
|
|
['ワ', 'ワ--'],
|
|
|
|
|
['ン', 'ン--']
|
|
|
|
|
]);
|
|
|
|
|
|
|
|
|
|
const ITERATION_MARK_CODE_POINT = 0x3005;
|
|
|
|
|
|
2020-01-23 02:43:38 +00:00
|
|
|
|
|
2020-03-21 17:18:34 +00:00
|
|
|
|
// Existing functions
|
2020-01-23 02:43:38 +00:00
|
|
|
|
|
2020-03-21 17:18:34 +00:00
|
|
|
|
const isCodePointKanji = jp.isCodePointKanji;
|
|
|
|
|
const isStringEntirelyKana = jp.isStringEntirelyKana;
|
2020-01-23 02:43:38 +00:00
|
|
|
|
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
// Conversion functions
|
2020-01-23 02:43:38 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
function convertKatakanaToHiragana(text) {
|
|
|
|
|
let result = '';
|
|
|
|
|
for (const c of text) {
|
|
|
|
|
if (wanakana.isKatakana(c)) {
|
|
|
|
|
result += wanakana.toHiragana(c);
|
|
|
|
|
} else {
|
|
|
|
|
result += c;
|
|
|
|
|
}
|
2017-08-15 06:10:59 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
return result;
|
2019-11-13 18:24:11 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
function convertHiraganaToKatakana(text) {
|
|
|
|
|
let result = '';
|
|
|
|
|
for (const c of text) {
|
|
|
|
|
if (wanakana.isHiragana(c)) {
|
|
|
|
|
result += wanakana.toKatakana(c);
|
2019-11-13 18:24:11 +00:00
|
|
|
|
} else {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
result += c;
|
2019-11-13 18:24:11 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return result;
|
2019-11-13 18:24:11 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
function convertToRomaji(text) {
|
|
|
|
|
return wanakana.toRomaji(text);
|
2017-08-24 03:56:52 +00:00
|
|
|
|
}
|
2017-08-26 18:57:34 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
function convertReading(expressionFragment, readingFragment, readingMode) {
|
|
|
|
|
switch (readingMode) {
|
|
|
|
|
case 'hiragana':
|
|
|
|
|
return convertKatakanaToHiragana(readingFragment || '');
|
|
|
|
|
case 'katakana':
|
|
|
|
|
return convertHiraganaToKatakana(readingFragment || '');
|
|
|
|
|
case 'romaji':
|
|
|
|
|
if (readingFragment) {
|
|
|
|
|
return convertToRomaji(readingFragment);
|
|
|
|
|
} else {
|
|
|
|
|
if (isStringEntirelyKana(expressionFragment)) {
|
|
|
|
|
return convertToRomaji(expressionFragment);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
return readingFragment;
|
|
|
|
|
case 'none':
|
|
|
|
|
return null;
|
|
|
|
|
default:
|
|
|
|
|
return readingFragment;
|
2017-08-24 03:56:52 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
2017-08-24 03:56:52 +00:00
|
|
|
|
|
2020-03-14 20:38:12 +00:00
|
|
|
|
function convertNumericToFullWidth(text) {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
let result = '';
|
|
|
|
|
for (const char of text) {
|
|
|
|
|
let c = char.codePointAt(0);
|
|
|
|
|
if (c >= 0x30 && c <= 0x39) { // ['0', '9']
|
|
|
|
|
c += 0xff10 - 0x30; // 0xff10 = '0' full width
|
|
|
|
|
result += String.fromCodePoint(c);
|
|
|
|
|
} else {
|
|
|
|
|
result += char;
|
2017-08-26 18:57:34 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
|
|
|
|
return result;
|
|
|
|
|
}
|
|
|
|
|
|
2020-03-28 21:51:58 +00:00
|
|
|
|
function convertHalfWidthKanaToFullWidth(text, sourceMap=null) {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
let result = '';
|
|
|
|
|
|
|
|
|
|
// This function is safe to use charCodeAt instead of codePointAt, since all
|
|
|
|
|
// the relevant characters are represented with a single UTF-16 character code.
|
|
|
|
|
for (let i = 0, ii = text.length; i < ii; ++i) {
|
|
|
|
|
const c = text[i];
|
|
|
|
|
const mapping = HALFWIDTH_KATAKANA_MAPPING.get(c);
|
|
|
|
|
if (typeof mapping !== 'string') {
|
|
|
|
|
result += c;
|
|
|
|
|
continue;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
let index = 0;
|
|
|
|
|
switch (text.charCodeAt(i + 1)) {
|
|
|
|
|
case 0xff9e: // dakuten
|
|
|
|
|
index = 1;
|
|
|
|
|
break;
|
|
|
|
|
case 0xff9f: // handakuten
|
|
|
|
|
index = 2;
|
2019-11-26 23:20:04 +00:00
|
|
|
|
break;
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
let c2 = mapping[index];
|
|
|
|
|
if (index > 0) {
|
|
|
|
|
if (c2 === '-') { // invalid
|
|
|
|
|
index = 0;
|
|
|
|
|
c2 = mapping[0];
|
|
|
|
|
} else {
|
|
|
|
|
++i;
|
2017-08-24 03:56:52 +00:00
|
|
|
|
}
|
2017-08-26 18:57:34 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-28 21:51:58 +00:00
|
|
|
|
if (sourceMap !== null && index > 0) {
|
|
|
|
|
sourceMap.combine(result.length, 1);
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
|
|
|
|
result += c2;
|
2017-08-24 03:56:52 +00:00
|
|
|
|
}
|
2019-11-03 03:08:57 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
return result;
|
2019-11-03 03:08:57 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-28 21:51:58 +00:00
|
|
|
|
function convertAlphabeticToKana(text, sourceMap=null) {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
let part = '';
|
|
|
|
|
let result = '';
|
2019-12-22 19:07:30 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
for (const char of text) {
|
|
|
|
|
// Note: 0x61 is the character code for 'a'
|
|
|
|
|
let c = char.codePointAt(0);
|
|
|
|
|
if (c >= 0x41 && c <= 0x5a) { // ['A', 'Z']
|
|
|
|
|
c += (0x61 - 0x41);
|
|
|
|
|
} else if (c >= 0x61 && c <= 0x7a) { // ['a', 'z']
|
|
|
|
|
// NOP; c += (0x61 - 0x61);
|
|
|
|
|
} else if (c >= 0xff21 && c <= 0xff3a) { // ['A', 'Z'] fullwidth
|
|
|
|
|
c += (0x61 - 0xff21);
|
|
|
|
|
} else if (c >= 0xff41 && c <= 0xff5a) { // ['a', 'z'] fullwidth
|
|
|
|
|
c += (0x61 - 0xff41);
|
|
|
|
|
} else if (c === 0x2d || c === 0xff0d) { // '-' or fullwidth dash
|
|
|
|
|
c = 0x2d; // '-'
|
2019-12-22 19:07:30 +00:00
|
|
|
|
} else {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
if (part.length > 0) {
|
2020-03-28 21:51:58 +00:00
|
|
|
|
result += convertAlphabeticPartToKana(part, sourceMap, result.length);
|
2020-03-12 00:33:01 +00:00
|
|
|
|
part = '';
|
|
|
|
|
}
|
|
|
|
|
result += char;
|
|
|
|
|
continue;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
part += String.fromCodePoint(c);
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
if (part.length > 0) {
|
2020-03-28 21:51:58 +00:00
|
|
|
|
result += convertAlphabeticPartToKana(part, sourceMap, result.length);
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
return result;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-28 21:51:58 +00:00
|
|
|
|
function convertAlphabeticPartToKana(text, sourceMap, sourceMapStart) {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
const result = wanakana.toHiragana(text);
|
|
|
|
|
|
|
|
|
|
// Generate source mapping
|
2020-03-28 21:51:58 +00:00
|
|
|
|
if (sourceMap !== null) {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
let i = 0;
|
|
|
|
|
let resultPos = 0;
|
|
|
|
|
const ii = text.length;
|
|
|
|
|
while (i < ii) {
|
|
|
|
|
// Find smallest matching substring
|
|
|
|
|
let iNext = i + 1;
|
|
|
|
|
let resultPosNext = result.length;
|
|
|
|
|
while (iNext < ii) {
|
|
|
|
|
const t = wanakana.toHiragana(text.substring(0, iNext));
|
|
|
|
|
if (t === result.substring(0, t.length)) {
|
|
|
|
|
resultPosNext = t.length;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
++iNext;
|
|
|
|
|
}
|
2019-12-22 19:07:30 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
// Merge characters
|
|
|
|
|
const removals = iNext - i - 1;
|
|
|
|
|
if (removals > 0) {
|
2020-03-28 21:51:58 +00:00
|
|
|
|
sourceMap.combine(sourceMapStart, removals);
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
2020-03-28 21:51:58 +00:00
|
|
|
|
++sourceMapStart;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
// Empty elements
|
|
|
|
|
const additions = resultPosNext - resultPos - 1;
|
|
|
|
|
for (let j = 0; j < additions; ++j) {
|
2020-03-28 21:51:58 +00:00
|
|
|
|
sourceMap.insert(sourceMapStart, 0);
|
|
|
|
|
++sourceMapStart;
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
2019-12-22 19:07:30 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
i = iNext;
|
|
|
|
|
resultPos = resultPosNext;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
return result;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
// Furigana distribution
|
2019-12-22 19:07:30 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
function distributeFurigana(expression, reading) {
|
|
|
|
|
const fallback = [{furigana: reading, text: expression}];
|
|
|
|
|
if (!reading) {
|
|
|
|
|
return fallback;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
let isAmbiguous = false;
|
|
|
|
|
const segmentize = (reading2, groups) => {
|
|
|
|
|
if (groups.length === 0 || isAmbiguous) {
|
|
|
|
|
return [];
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
const group = groups[0];
|
|
|
|
|
if (group.mode === 'kana') {
|
|
|
|
|
if (convertKatakanaToHiragana(reading2).startsWith(convertKatakanaToHiragana(group.text))) {
|
|
|
|
|
const readingLeft = reading2.substring(group.text.length);
|
|
|
|
|
const segs = segmentize(readingLeft, groups.splice(1));
|
|
|
|
|
if (segs) {
|
|
|
|
|
return [{text: group.text}].concat(segs);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
let foundSegments = null;
|
|
|
|
|
for (let i = reading2.length; i >= group.text.length; --i) {
|
|
|
|
|
const readingUsed = reading2.substring(0, i);
|
|
|
|
|
const readingLeft = reading2.substring(i);
|
|
|
|
|
const segs = segmentize(readingLeft, groups.slice(1));
|
|
|
|
|
if (segs) {
|
|
|
|
|
if (foundSegments !== null) {
|
|
|
|
|
// more than one way to segmentize the tail, mark as ambiguous
|
|
|
|
|
isAmbiguous = true;
|
|
|
|
|
return null;
|
|
|
|
|
}
|
|
|
|
|
foundSegments = [{text: group.text, furigana: readingUsed}].concat(segs);
|
|
|
|
|
}
|
|
|
|
|
// there is only one way to segmentize the last non-kana group
|
|
|
|
|
if (groups.length === 1) {
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
return foundSegments;
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
const groups = [];
|
|
|
|
|
let modePrev = null;
|
|
|
|
|
for (const c of expression) {
|
|
|
|
|
const codePoint = c.codePointAt(0);
|
|
|
|
|
const modeCurr = isCodePointKanji(codePoint) || codePoint === ITERATION_MARK_CODE_POINT ? 'kanji' : 'kana';
|
|
|
|
|
if (modeCurr === modePrev) {
|
|
|
|
|
groups[groups.length - 1].text += c;
|
|
|
|
|
} else {
|
|
|
|
|
groups.push({mode: modeCurr, text: c});
|
|
|
|
|
modePrev = modeCurr;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
const segments = segmentize(reading, groups);
|
|
|
|
|
if (segments && !isAmbiguous) {
|
|
|
|
|
return segments;
|
|
|
|
|
}
|
|
|
|
|
return fallback;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
function distributeFuriganaInflected(expression, reading, source) {
|
|
|
|
|
const output = [];
|
|
|
|
|
|
|
|
|
|
let stemLength = 0;
|
|
|
|
|
const shortest = Math.min(source.length, expression.length);
|
|
|
|
|
const sourceHiragana = convertKatakanaToHiragana(source);
|
|
|
|
|
const expressionHiragana = convertKatakanaToHiragana(expression);
|
|
|
|
|
while (stemLength < shortest && sourceHiragana[stemLength] === expressionHiragana[stemLength]) {
|
|
|
|
|
++stemLength;
|
|
|
|
|
}
|
|
|
|
|
const offset = source.length - stemLength;
|
|
|
|
|
|
|
|
|
|
const stemExpression = source.substring(0, source.length - offset);
|
|
|
|
|
const stemReading = reading.substring(
|
|
|
|
|
0,
|
|
|
|
|
offset === 0 ? reading.length : reading.length - expression.length + stemLength
|
|
|
|
|
);
|
|
|
|
|
for (const segment of distributeFurigana(stemExpression, stemReading)) {
|
|
|
|
|
output.push(segment);
|
|
|
|
|
}
|
2019-12-22 19:07:30 +00:00
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
if (stemLength !== source.length) {
|
|
|
|
|
output.push({text: source.substring(stemLength)});
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
2020-03-12 00:33:01 +00:00
|
|
|
|
|
|
|
|
|
return output;
|
2019-12-22 19:07:30 +00:00
|
|
|
|
}
|
|
|
|
|
|
2020-03-12 00:33:01 +00:00
|
|
|
|
|
|
|
|
|
// Exports
|
|
|
|
|
|
2020-03-21 17:18:34 +00:00
|
|
|
|
Object.assign(jp, {
|
2020-03-12 00:33:01 +00:00
|
|
|
|
convertKatakanaToHiragana,
|
|
|
|
|
convertHiraganaToKatakana,
|
|
|
|
|
convertToRomaji,
|
|
|
|
|
convertReading,
|
2020-03-14 20:38:12 +00:00
|
|
|
|
convertNumericToFullWidth,
|
2020-03-12 00:33:01 +00:00
|
|
|
|
convertHalfWidthKanaToFullWidth,
|
|
|
|
|
convertAlphabeticToKana,
|
|
|
|
|
distributeFurigana,
|
|
|
|
|
distributeFuriganaInflected
|
2020-03-21 17:18:34 +00:00
|
|
|
|
});
|
2020-03-12 00:33:01 +00:00
|
|
|
|
})();
|