aboutsummaryrefslogtreecommitdiff
path: root/ext/js/language
diff options
context:
space:
mode:
authortoasted-nutbread <toasted-nutbread@users.noreply.github.com>2024-01-28 07:22:47 -0500
committerGitHub <noreply@github.com>2024-01-28 12:22:47 +0000
commitacc013a1a8051d34322f0f5f91d7bdecc0a18843 (patch)
tree601344b2f047f395548ddfb16a83319af10464f9 /ext/js/language
parenta51f1ab2dc675a49bfeeb08cc24b97eb8d888e4a (diff)
JapaneseUtil refactor (#555)
* Copy functions from JapaneseUtil * Remove JapaneseUtil * Update usages of JapaneseUtil functions
Diffstat (limited to 'ext/js/language')
-rw-r--r--ext/js/language/japanese-wanakana.js122
-rw-r--r--ext/js/language/japanese.js740
-rw-r--r--ext/js/language/sandbox/japanese-util.js885
-rw-r--r--ext/js/language/translator.js22
4 files changed, 872 insertions, 897 deletions
diff --git a/ext/js/language/japanese-wanakana.js b/ext/js/language/japanese-wanakana.js
new file mode 100644
index 00000000..b48ab6d6
--- /dev/null
+++ b/ext/js/language/japanese-wanakana.js
@@ -0,0 +1,122 @@
+/*
+ * Copyright (C) 2024 Yomitan Authors
+ *
+ * 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
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <https://www.gnu.org/licenses/>.
+ */
+
+import * as wanakana from '../../lib/wanakana.js';
+
+/**
+ * @param {string} text
+ * @param {?import('../general/text-source-map.js').TextSourceMap} sourceMap
+ * @param {number} sourceMapStart
+ * @returns {string}
+ */
+function convertAlphabeticPartToKana(text, sourceMap, sourceMapStart) {
+ const result = wanakana.toHiragana(text);
+
+ // Generate source mapping
+ if (sourceMap !== null) {
+ 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;
+ }
+
+ // Merge characters
+ const removals = iNext - i - 1;
+ if (removals > 0) {
+ sourceMap.combine(sourceMapStart, removals);
+ }
+ ++sourceMapStart;
+
+ // Empty elements
+ const additions = resultPosNext - resultPos - 1;
+ for (let j = 0; j < additions; ++j) {
+ sourceMap.insert(sourceMapStart, 0);
+ ++sourceMapStart;
+ }
+
+ i = iNext;
+ resultPos = resultPosNext;
+ }
+ }
+
+ return result;
+}
+
+/**
+ * @param {string} text
+ * @returns {string}
+ */
+export function convertToKana(text) {
+ return wanakana.toKana(text);
+}
+
+/**
+ * @param {string} text
+ * @returns {string}
+ */
+export function convertToRomaji(text) {
+ return wanakana.toRomaji(text);
+}
+
+/**
+ * @param {string} text
+ * @param {?import('../general/text-source-map.js').TextSourceMap} sourceMap
+ * @returns {string}
+ */
+export function convertAlphabeticToKana(text, sourceMap = null) {
+ let part = '';
+ let result = '';
+
+ for (const char of text) {
+ // Note: 0x61 is the character code for 'a'
+ let c = /** @type {number} */ (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; // '-'
+ } else {
+ if (part.length > 0) {
+ result += convertAlphabeticPartToKana(part, sourceMap, result.length);
+ part = '';
+ }
+ result += char;
+ continue;
+ }
+ part += String.fromCodePoint(c);
+ }
+
+ if (part.length > 0) {
+ result += convertAlphabeticPartToKana(part, sourceMap, result.length);
+ }
+ return result;
+}
diff --git a/ext/js/language/japanese.js b/ext/js/language/japanese.js
new file mode 100644
index 00000000..88eb5af5
--- /dev/null
+++ b/ext/js/language/japanese.js
@@ -0,0 +1,740 @@
+/*
+ * Copyright (C) 2024 Yomitan Authors
+ *
+ * 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
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <https://www.gnu.org/licenses/>.
+ */
+
+const HIRAGANA_SMALL_TSU_CODE_POINT = 0x3063;
+const KATAKANA_SMALL_TSU_CODE_POINT = 0x30c3;
+const KATAKANA_SMALL_KA_CODE_POINT = 0x30f5;
+const KATAKANA_SMALL_KE_CODE_POINT = 0x30f6;
+const KANA_PROLONGED_SOUND_MARK_CODE_POINT = 0x30fc;
+
+/** @type {import('japanese-util').CodepointRange} */
+const HIRAGANA_RANGE = [0x3040, 0x309f];
+/** @type {import('japanese-util').CodepointRange} */
+const KATAKANA_RANGE = [0x30a0, 0x30ff];
+
+/** @type {import('japanese-util').CodepointRange} */
+const HIRAGANA_CONVERSION_RANGE = [0x3041, 0x3096];
+/** @type {import('japanese-util').CodepointRange} */
+const KATAKANA_CONVERSION_RANGE = [0x30a1, 0x30f6];
+
+/** @type {import('japanese-util').CodepointRange[]} */
+const KANA_RANGES = [HIRAGANA_RANGE, KATAKANA_RANGE];
+
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_RANGE = [0x4e00, 0x9fff];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A_RANGE = [0x3400, 0x4dbf];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_B_RANGE = [0x20000, 0x2a6df];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_C_RANGE = [0x2a700, 0x2b73f];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_D_RANGE = [0x2b740, 0x2b81f];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_E_RANGE = [0x2b820, 0x2ceaf];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_F_RANGE = [0x2ceb0, 0x2ebef];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_COMPATIBILITY_IDEOGRAPHS_RANGE = [0xf900, 0xfaff];
+/** @type {import('japanese-util').CodepointRange} */
+const CJK_COMPATIBILITY_IDEOGRAPHS_SUPPLEMENT_RANGE = [0x2f800, 0x2fa1f];
+/** @type {import('japanese-util').CodepointRange[]} */
+const CJK_IDEOGRAPH_RANGES = [
+ CJK_UNIFIED_IDEOGRAPHS_RANGE,
+ CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A_RANGE,
+ CJK_UNIFIED_IDEOGRAPHS_EXTENSION_B_RANGE,
+ CJK_UNIFIED_IDEOGRAPHS_EXTENSION_C_RANGE,
+ CJK_UNIFIED_IDEOGRAPHS_EXTENSION_D_RANGE,
+ CJK_UNIFIED_IDEOGRAPHS_EXTENSION_E_RANGE,
+ CJK_UNIFIED_IDEOGRAPHS_EXTENSION_F_RANGE,
+ CJK_COMPATIBILITY_IDEOGRAPHS_RANGE,
+ CJK_COMPATIBILITY_IDEOGRAPHS_SUPPLEMENT_RANGE
+];
+
+/**
+ * Japanese character ranges, roughly ordered in order of expected frequency.
+ * @type {import('japanese-util').CodepointRange[]}
+ */
+const JAPANESE_RANGES = [
+ HIRAGANA_RANGE,
+ KATAKANA_RANGE,
+
+ ...CJK_IDEOGRAPH_RANGES,
+
+ [0xff66, 0xff9f], // Halfwidth katakana
+
+ [0x30fb, 0x30fc], // Katakana punctuation
+ [0xff61, 0xff65], // Kana punctuation
+ [0x3000, 0x303f], // CJK punctuation
+
+ [0xff10, 0xff19], // Fullwidth numbers
+ [0xff21, 0xff3a], // Fullwidth upper case Latin letters
+ [0xff41, 0xff5a], // Fullwidth lower case Latin letters
+
+ [0xff01, 0xff0f], // Fullwidth punctuation 1
+ [0xff1a, 0xff1f], // Fullwidth punctuation 2
+ [0xff3b, 0xff3f], // Fullwidth punctuation 3
+ [0xff5b, 0xff60], // Fullwidth punctuation 4
+ [0xffe0, 0xffee] // Currency markers
+];
+
+const SMALL_KANA_SET = new Set(Array.from('ぁぃぅぇぉゃゅょゎァィゥェォャュョヮ'));
+
+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 VOWEL_TO_KANA_MAPPING = new Map([
+ ['a', 'ぁあかがさざただなはばぱまゃやらゎわヵァアカガサザタダナハバパマャヤラヮワヵヷ'],
+ ['i', 'ぃいきぎしじちぢにひびぴみりゐィイキギシジチヂニヒビピミリヰヸ'],
+ ['u', 'ぅうくぐすずっつづぬふぶぷむゅゆるゥウクグスズッツヅヌフブプムュユルヴ'],
+ ['e', 'ぇえけげせぜてでねへべぺめれゑヶェエケゲセゼテデネヘベペメレヱヶヹ'],
+ ['o', 'ぉおこごそぞとどのほぼぽもょよろをォオコゴソゾトドノホボポモョヨロヲヺ'],
+ ['', 'のノ']
+]);
+
+/** @type {Map<string, string>} */
+const KANA_TO_VOWEL_MAPPING = new Map();
+for (const [vowel, characters] of VOWEL_TO_KANA_MAPPING) {
+ for (const character of characters) {
+ KANA_TO_VOWEL_MAPPING.set(character, vowel);
+ }
+}
+
+const kana = 'うゔ-かが-きぎ-くぐ-けげ-こご-さざ-しじ-すず-せぜ-そぞ-ただ-ちぢ-つづ-てで-とど-はばぱひびぴふぶぷへべぺほぼぽワヷ-ヰヸ-ウヴ-ヱヹ-ヲヺ-カガ-キギ-クグ-ケゲ-コゴ-サザ-シジ-スズ-セゼ-ソゾ-タダ-チヂ-ツヅ-テデ-トド-ハバパヒビピフブプヘベペホボポ';
+/** @type {Map<string, {character: string, type: import('japanese-util').DiacriticType}>} */
+const DIACRITIC_MAPPING = new Map();
+for (let i = 0, ii = kana.length; i < ii; i += 3) {
+ const character = kana[i];
+ const dakuten = kana[i + 1];
+ const handakuten = kana[i + 2];
+ DIACRITIC_MAPPING.set(dakuten, {character, type: 'dakuten'});
+ if (handakuten !== '-') {
+ DIACRITIC_MAPPING.set(handakuten, {character, type: 'handakuten'});
+ }
+}
+
+
+/**
+ * @param {number} codePoint
+ * @param {import('japanese-util').CodepointRange} range
+ * @returns {boolean}
+ */
+function isCodePointInRange(codePoint, [min, max]) {
+ return (codePoint >= min && codePoint <= max);
+}
+
+/**
+ * @param {number} codePoint
+ * @param {import('japanese-util').CodepointRange[]} ranges
+ * @returns {boolean}
+ */
+function isCodePointInRanges(codePoint, ranges) {
+ for (const [min, max] of ranges) {
+ if (codePoint >= min && codePoint <= max) {
+ return true;
+ }
+ }
+ return false;
+}
+
+/**
+ * @param {string} previousCharacter
+ * @returns {?string}
+ */
+function getProlongedHiragana(previousCharacter) {
+ switch (KANA_TO_VOWEL_MAPPING.get(previousCharacter)) {
+ case 'a': return 'あ';
+ case 'i': return 'い';
+ case 'u': return 'う';
+ case 'e': return 'え';
+ case 'o': return 'う';
+ default: return null;
+ }
+}
+
+/**
+ * @param {string} text
+ * @param {string} reading
+ * @returns {import('japanese-util').FuriganaSegment}
+ */
+function createFuriganaSegment(text, reading) {
+ return {text, reading};
+}
+
+/**
+ * @param {string} reading
+ * @param {string} readingNormalized
+ * @param {import('japanese-util').FuriganaGroup[]} groups
+ * @param {number} groupsStart
+ * @returns {?(import('japanese-util').FuriganaSegment[])}
+ */
+function segmentizeFurigana(reading, readingNormalized, groups, groupsStart) {
+ const groupCount = groups.length - groupsStart;
+ if (groupCount <= 0) {
+ return reading.length === 0 ? [] : null;
+ }
+
+ const group = groups[groupsStart];
+ const {isKana, text} = group;
+ const textLength = text.length;
+ if (isKana) {
+ const {textNormalized} = group;
+ if (textNormalized !== null && readingNormalized.startsWith(textNormalized)) {
+ const segments = segmentizeFurigana(
+ reading.substring(textLength),
+ readingNormalized.substring(textLength),
+ groups,
+ groupsStart + 1
+ );
+ if (segments !== null) {
+ if (reading.startsWith(text)) {
+ segments.unshift(createFuriganaSegment(text, ''));
+ } else {
+ segments.unshift(...getFuriganaKanaSegments(text, reading));
+ }
+ return segments;
+ }
+ }
+ return null;
+ } else {
+ let result = null;
+ for (let i = reading.length; i >= textLength; --i) {
+ const segments = segmentizeFurigana(
+ reading.substring(i),
+ readingNormalized.substring(i),
+ groups,
+ groupsStart + 1
+ );
+ if (segments !== null) {
+ if (result !== null) {
+ // More than one way to segmentize the tail; mark as ambiguous
+ return null;
+ }
+ const segmentReading = reading.substring(0, i);
+ segments.unshift(createFuriganaSegment(text, segmentReading));
+ result = segments;
+ }
+ // There is only one way to segmentize the last non-kana group
+ if (groupCount === 1) {
+ break;
+ }
+ }
+ return result;
+ }
+}
+
+/**
+ * @param {string} text
+ * @param {string} reading
+ * @returns {import('japanese-util').FuriganaSegment[]}
+ */
+function getFuriganaKanaSegments(text, reading) {
+ const textLength = text.length;
+ const newSegments = [];
+ let start = 0;
+ let state = (reading[0] === text[0]);
+ for (let i = 1; i < textLength; ++i) {
+ const newState = (reading[i] === text[i]);
+ if (state === newState) { continue; }
+ newSegments.push(createFuriganaSegment(text.substring(start, i), state ? '' : reading.substring(start, i)));
+ state = newState;
+ start = i;
+ }
+ newSegments.push(createFuriganaSegment(text.substring(start, textLength), state ? '' : reading.substring(start, textLength)));
+ return newSegments;
+}
+
+/**
+ * @param {string} text1
+ * @param {string} text2
+ * @returns {number}
+ */
+function getStemLength(text1, text2) {
+ const minLength = Math.min(text1.length, text2.length);
+ if (minLength === 0) { return 0; }
+
+ let i = 0;
+ while (true) {
+ const char1 = /** @type {number} */ (text1.codePointAt(i));
+ const char2 = /** @type {number} */ (text2.codePointAt(i));
+ if (char1 !== char2) { break; }
+ const charLength = String.fromCodePoint(char1).length;
+ i += charLength;
+ if (i >= minLength) {
+ if (i > minLength) {
+ i -= charLength; // Don't consume partial UTF16 surrogate characters
+ }
+ break;
+ }
+ }
+ return i;
+}
+
+
+// Character code testing functions
+
+/**
+ * @param {number} codePoint
+ * @returns {boolean}
+ */
+export function isCodePointKanji(codePoint) {
+ return isCodePointInRanges(codePoint, CJK_IDEOGRAPH_RANGES);
+}
+
+/**
+ * @param {number} codePoint
+ * @returns {boolean}
+ */
+export function isCodePointKana(codePoint) {
+ return isCodePointInRanges(codePoint, KANA_RANGES);
+}
+
+/**
+ * @param {number} codePoint
+ * @returns {boolean}
+ */
+export function isCodePointJapanese(codePoint) {
+ return isCodePointInRanges(codePoint, JAPANESE_RANGES);
+}
+
+
+// String testing functions
+
+/**
+ * @param {string} str
+ * @returns {boolean}
+ */
+export function isStringEntirelyKana(str) {
+ if (str.length === 0) { return false; }
+ for (const c of str) {
+ if (!isCodePointInRanges(/** @type {number} */ (c.codePointAt(0)), KANA_RANGES)) {
+ return false;
+ }
+ }
+ return true;
+}
+
+/**
+ * @param {string} str
+ * @returns {boolean}
+ */
+export function isStringPartiallyJapanese(str) {
+ if (str.length === 0) { return false; }
+ for (const c of str) {
+ if (isCodePointInRanges(/** @type {number} */ (c.codePointAt(0)), JAPANESE_RANGES)) {
+ return true;
+ }
+ }
+ return false;
+}
+
+
+// Mora functions
+
+/**
+ * @param {number} moraIndex
+ * @param {number} pitchAccentDownstepPosition
+ * @returns {boolean}
+ */
+export function isMoraPitchHigh(moraIndex, pitchAccentDownstepPosition) {
+ switch (pitchAccentDownstepPosition) {
+ case 0: return (moraIndex > 0);
+ case 1: return (moraIndex < 1);
+ default: return (moraIndex > 0 && moraIndex < pitchAccentDownstepPosition);
+ }
+}
+
+/**
+ * @param {string} text
+ * @param {number} pitchAccentDownstepPosition
+ * @param {boolean} isVerbOrAdjective
+ * @returns {?import('japanese-util').PitchCategory}
+ */
+export function getPitchCategory(text, pitchAccentDownstepPosition, isVerbOrAdjective) {
+ if (pitchAccentDownstepPosition === 0) {
+ return 'heiban';
+ }
+ if (isVerbOrAdjective) {
+ return pitchAccentDownstepPosition > 0 ? 'kifuku' : null;
+ }
+ if (pitchAccentDownstepPosition === 1) {
+ return 'atamadaka';
+ }
+ if (pitchAccentDownstepPosition > 1) {
+ return pitchAccentDownstepPosition >= getKanaMoraCount(text) ? 'odaka' : 'nakadaka';
+ }
+ return null;
+}
+
+/**
+ * @param {string} text
+ * @returns {string[]}
+ */
+export function getKanaMorae(text) {
+ const morae = [];
+ let i;
+ for (const c of text) {
+ if (SMALL_KANA_SET.has(c) && (i = morae.length) > 0) {
+ morae[i - 1] += c;
+ } else {
+ morae.push(c);
+ }
+ }
+ return morae;
+}
+
+/**
+ * @param {string} text
+ * @returns {number}
+ */
+export function getKanaMoraCount(text) {
+ let moraCount = 0;
+ for (const c of text) {
+ if (!(SMALL_KANA_SET.has(c) && moraCount > 0)) {
+ ++moraCount;
+ }
+ }
+ return moraCount;
+}
+
+
+// Conversion functions
+
+/**
+ * @param {string} text
+ * @param {boolean} [keepProlongedSoundMarks]
+ * @returns {string}
+ */
+export function convertKatakanaToHiragana(text, keepProlongedSoundMarks = false) {
+ let result = '';
+ const offset = (HIRAGANA_CONVERSION_RANGE[0] - KATAKANA_CONVERSION_RANGE[0]);
+ for (let char of text) {
+ const codePoint = /** @type {number} */ (char.codePointAt(0));
+ switch (codePoint) {
+ case KATAKANA_SMALL_KA_CODE_POINT:
+ case KATAKANA_SMALL_KE_CODE_POINT:
+ // No change
+ break;
+ case KANA_PROLONGED_SOUND_MARK_CODE_POINT:
+ if (!keepProlongedSoundMarks && result.length > 0) {
+ const char2 = getProlongedHiragana(result[result.length - 1]);
+ if (char2 !== null) { char = char2; }
+ }
+ break;
+ default:
+ if (isCodePointInRange(codePoint, KATAKANA_CONVERSION_RANGE)) {
+ char = String.fromCodePoint(codePoint + offset);
+ }
+ break;
+ }
+ result += char;
+ }
+ return result;
+}
+
+/**
+ * @param {string} text
+ * @returns {string}
+ */
+export function convertHiraganaToKatakana(text) {
+ let result = '';
+ const offset = (KATAKANA_CONVERSION_RANGE[0] - HIRAGANA_CONVERSION_RANGE[0]);
+ for (let char of text) {
+ const codePoint = /** @type {number} */ (char.codePointAt(0));
+ if (isCodePointInRange(codePoint, HIRAGANA_CONVERSION_RANGE)) {
+ char = String.fromCodePoint(codePoint + offset);
+ }
+ result += char;
+ }
+ return result;
+}
+
+/**
+ * @param {string} text
+ * @returns {string}
+ */
+export function convertNumericToFullWidth(text) {
+ let result = '';
+ for (const char of text) {
+ let c = /** @type {number} */ (char.codePointAt(0));
+ if (c >= 0x30 && c <= 0x39) { // ['0', '9']
+ c += 0xff10 - 0x30; // 0xff10 = '0' full width
+ result += String.fromCodePoint(c);
+ } else {
+ result += char;
+ }
+ }
+ return result;
+}
+
+/**
+ * @param {string} text
+ * @param {?import('../general/text-source-map.js').TextSourceMap} [sourceMap]
+ * @returns {string}
+ */
+export function convertHalfWidthKanaToFullWidth(text, sourceMap = null) {
+ 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;
+ break;
+ }
+
+ let c2 = mapping[index];
+ if (index > 0) {
+ if (c2 === '-') { // invalid
+ index = 0;
+ c2 = mapping[0];
+ } else {
+ ++i;
+ }
+ }
+
+ if (sourceMap !== null && index > 0) {
+ sourceMap.combine(result.length, 1);
+ }
+ result += c2;
+ }
+
+ return result;
+}
+
+/**
+ * @param {string} character
+ * @returns {?{character: string, type: import('japanese-util').DiacriticType}}
+ */
+export function getKanaDiacriticInfo(character) {
+ const info = DIACRITIC_MAPPING.get(character);
+ return typeof info !== 'undefined' ? {character: info.character, type: info.type} : null;
+}
+
+
+// Furigana distribution
+
+/**
+ * @param {string} term
+ * @param {string} reading
+ * @returns {import('japanese-util').FuriganaSegment[]}
+ */
+export function distributeFurigana(term, reading) {
+ if (reading === term) {
+ // Same
+ return [createFuriganaSegment(term, '')];
+ }
+
+ /** @type {import('japanese-util').FuriganaGroup[]} */
+ const groups = [];
+ /** @type {?import('japanese-util').FuriganaGroup} */
+ let groupPre = null;
+ let isKanaPre = null;
+ for (const c of term) {
+ const codePoint = /** @type {number} */ (c.codePointAt(0));
+ const isKana = isCodePointKana(codePoint);
+ if (isKana === isKanaPre) {
+ /** @type {import('japanese-util').FuriganaGroup} */ (groupPre).text += c;
+ } else {
+ groupPre = {isKana, text: c, textNormalized: null};
+ groups.push(groupPre);
+ isKanaPre = isKana;
+ }
+ }
+ for (const group of groups) {
+ if (group.isKana) {
+ group.textNormalized = convertKatakanaToHiragana(group.text);
+ }
+ }
+
+ const readingNormalized = convertKatakanaToHiragana(reading);
+ const segments = segmentizeFurigana(reading, readingNormalized, groups, 0);
+ if (segments !== null) {
+ return segments;
+ }
+
+ // Fallback
+ return [createFuriganaSegment(term, reading)];
+}
+
+/**
+ * @param {string} term
+ * @param {string} reading
+ * @param {string} source
+ * @returns {import('japanese-util').FuriganaSegment[]}
+ */
+export function distributeFuriganaInflected(term, reading, source) {
+ const termNormalized = convertKatakanaToHiragana(term);
+ const readingNormalized = convertKatakanaToHiragana(reading);
+ const sourceNormalized = convertKatakanaToHiragana(source);
+
+ let mainText = term;
+ let stemLength = getStemLength(termNormalized, sourceNormalized);
+
+ // Check if source is derived from the reading instead of the term
+ const readingStemLength = getStemLength(readingNormalized, sourceNormalized);
+ if (readingStemLength > 0 && readingStemLength >= stemLength) {
+ mainText = reading;
+ stemLength = readingStemLength;
+ reading = `${source.substring(0, stemLength)}${reading.substring(stemLength)}`;
+ }
+
+ const segments = [];
+ if (stemLength > 0) {
+ mainText = `${source.substring(0, stemLength)}${mainText.substring(stemLength)}`;
+ const segments2 = distributeFurigana(mainText, reading);
+ let consumed = 0;
+ for (const segment of segments2) {
+ const {text} = segment;
+ const start = consumed;
+ consumed += text.length;
+ if (consumed < stemLength) {
+ segments.push(segment);
+ } else if (consumed === stemLength) {
+ segments.push(segment);
+ break;
+ } else {
+ if (start < stemLength) {
+ segments.push(createFuriganaSegment(mainText.substring(start, stemLength), ''));
+ }
+ break;
+ }
+ }
+ }
+
+ if (stemLength < source.length) {
+ const remainder = source.substring(stemLength);
+ const segmentCount = segments.length;
+ if (segmentCount > 0 && segments[segmentCount - 1].reading.length === 0) {
+ // Append to the last segment if it has an empty reading
+ segments[segmentCount - 1].text += remainder;
+ } else {
+ // Otherwise, create a new segment
+ segments.push(createFuriganaSegment(remainder, ''));
+ }
+ }
+
+ return segments;
+}
+
+
+// Miscellaneous
+
+/**
+ * @param {string} text
+ * @param {boolean} fullCollapse
+ * @param {?import('../general/text-source-map.js').TextSourceMap} [sourceMap]
+ * @returns {string}
+ */
+export function collapseEmphaticSequences(text, fullCollapse, sourceMap = null) {
+ let result = '';
+ let collapseCodePoint = -1;
+ const hasSourceMap = (sourceMap !== null);
+ for (const char of text) {
+ const c = char.codePointAt(0);
+ if (
+ c === HIRAGANA_SMALL_TSU_CODE_POINT ||
+ c === KATAKANA_SMALL_TSU_CODE_POINT ||
+ c === KANA_PROLONGED_SOUND_MARK_CODE_POINT
+ ) {
+ if (collapseCodePoint !== c) {
+ collapseCodePoint = c;
+ if (!fullCollapse) {
+ result += char;
+ continue;
+ }
+ }
+ } else {
+ collapseCodePoint = -1;
+ result += char;
+ continue;
+ }
+
+ if (hasSourceMap) {
+ sourceMap.combine(Math.max(0, result.length - 1), 1);
+ }
+ }
+ return result;
+}
diff --git a/ext/js/language/sandbox/japanese-util.js b/ext/js/language/sandbox/japanese-util.js
deleted file mode 100644
index f9874cd4..00000000
--- a/ext/js/language/sandbox/japanese-util.js
+++ /dev/null
@@ -1,885 +0,0 @@
-/*
- * Copyright (C) 2023-2024 Yomitan Authors
- * Copyright (C) 2020-2022 Yomichan Authors
- *
- * 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
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program. If not, see <https://www.gnu.org/licenses/>.
- */
-
-const HIRAGANA_SMALL_TSU_CODE_POINT = 0x3063;
-const KATAKANA_SMALL_TSU_CODE_POINT = 0x30c3;
-const KATAKANA_SMALL_KA_CODE_POINT = 0x30f5;
-const KATAKANA_SMALL_KE_CODE_POINT = 0x30f6;
-const KANA_PROLONGED_SOUND_MARK_CODE_POINT = 0x30fc;
-
-/** @type {import('japanese-util').CodepointRange} */
-const HIRAGANA_RANGE = [0x3040, 0x309f];
-/** @type {import('japanese-util').CodepointRange} */
-const KATAKANA_RANGE = [0x30a0, 0x30ff];
-
-/** @type {import('japanese-util').CodepointRange} */
-const HIRAGANA_CONVERSION_RANGE = [0x3041, 0x3096];
-/** @type {import('japanese-util').CodepointRange} */
-const KATAKANA_CONVERSION_RANGE = [0x30a1, 0x30f6];
-
-/** @type {import('japanese-util').CodepointRange[]} */
-const KANA_RANGES = [HIRAGANA_RANGE, KATAKANA_RANGE];
-
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_RANGE = [0x4e00, 0x9fff];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A_RANGE = [0x3400, 0x4dbf];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_B_RANGE = [0x20000, 0x2a6df];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_C_RANGE = [0x2a700, 0x2b73f];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_D_RANGE = [0x2b740, 0x2b81f];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_E_RANGE = [0x2b820, 0x2ceaf];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_UNIFIED_IDEOGRAPHS_EXTENSION_F_RANGE = [0x2ceb0, 0x2ebef];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_COMPATIBILITY_IDEOGRAPHS_RANGE = [0xf900, 0xfaff];
-/** @type {import('japanese-util').CodepointRange} */
-const CJK_COMPATIBILITY_IDEOGRAPHS_SUPPLEMENT_RANGE = [0x2f800, 0x2fa1f];
-/** @type {import('japanese-util').CodepointRange[]} */
-const CJK_IDEOGRAPH_RANGES = [
- CJK_UNIFIED_IDEOGRAPHS_RANGE,
- CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A_RANGE,
- CJK_UNIFIED_IDEOGRAPHS_EXTENSION_B_RANGE,
- CJK_UNIFIED_IDEOGRAPHS_EXTENSION_C_RANGE,
- CJK_UNIFIED_IDEOGRAPHS_EXTENSION_D_RANGE,
- CJK_UNIFIED_IDEOGRAPHS_EXTENSION_E_RANGE,
- CJK_UNIFIED_IDEOGRAPHS_EXTENSION_F_RANGE,
- CJK_COMPATIBILITY_IDEOGRAPHS_RANGE,
- CJK_COMPATIBILITY_IDEOGRAPHS_SUPPLEMENT_RANGE
-];
-
-/**
- * Japanese character ranges, roughly ordered in order of expected frequency.
- * @type {import('japanese-util').CodepointRange[]}
- */
-const JAPANESE_RANGES = [
- HIRAGANA_RANGE,
- KATAKANA_RANGE,
-
- ...CJK_IDEOGRAPH_RANGES,
-
- [0xff66, 0xff9f], // Halfwidth katakana
-
- [0x30fb, 0x30fc], // Katakana punctuation
- [0xff61, 0xff65], // Kana punctuation
- [0x3000, 0x303f], // CJK punctuation
-
- [0xff10, 0xff19], // Fullwidth numbers
- [0xff21, 0xff3a], // Fullwidth upper case Latin letters
- [0xff41, 0xff5a], // Fullwidth lower case Latin letters
-
- [0xff01, 0xff0f], // Fullwidth punctuation 1
- [0xff1a, 0xff1f], // Fullwidth punctuation 2
- [0xff3b, 0xff3f], // Fullwidth punctuation 3
- [0xff5b, 0xff60], // Fullwidth punctuation 4
- [0xffe0, 0xffee] // Currency markers
-];
-
-const SMALL_KANA_SET = new Set(Array.from('ぁぃぅぇぉゃゅょゎァィゥェォャュョヮ'));
-
-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 VOWEL_TO_KANA_MAPPING = new Map([
- ['a', 'ぁあかがさざただなはばぱまゃやらゎわヵァアカガサザタダナハバパマャヤラヮワヵヷ'],
- ['i', 'ぃいきぎしじちぢにひびぴみりゐィイキギシジチヂニヒビピミリヰヸ'],
- ['u', 'ぅうくぐすずっつづぬふぶぷむゅゆるゥウクグスズッツヅヌフブプムュユルヴ'],
- ['e', 'ぇえけげせぜてでねへべぺめれゑヶェエケゲセゼテデネヘベペメレヱヶヹ'],
- ['o', 'ぉおこごそぞとどのほぼぽもょよろをォオコゴソゾトドノホボポモョヨロヲヺ'],
- ['', 'のノ']
-]);
-
-/** @type {Map<string, string>} */
-const KANA_TO_VOWEL_MAPPING = new Map();
-for (const [vowel, characters] of VOWEL_TO_KANA_MAPPING) {
- for (const character of characters) {
- KANA_TO_VOWEL_MAPPING.set(character, vowel);
- }
-}
-
-const kana = 'うゔ-かが-きぎ-くぐ-けげ-こご-さざ-しじ-すず-せぜ-そぞ-ただ-ちぢ-つづ-てで-とど-はばぱひびぴふぶぷへべぺほぼぽワヷ-ヰヸ-ウヴ-ヱヹ-ヲヺ-カガ-キギ-クグ-ケゲ-コゴ-サザ-シジ-スズ-セゼ-ソゾ-タダ-チヂ-ツヅ-テデ-トド-ハバパヒビピフブプヘベペホボポ';
-/** @type {Map<string, {character: string, type: import('japanese-util').DiacriticType}>} */
-const DIACRITIC_MAPPING = new Map();
-for (let i = 0, ii = kana.length; i < ii; i += 3) {
- const character = kana[i];
- const dakuten = kana[i + 1];
- const handakuten = kana[i + 2];
- DIACRITIC_MAPPING.set(dakuten, {character, type: 'dakuten'});
- if (handakuten !== '-') {
- DIACRITIC_MAPPING.set(handakuten, {character, type: 'handakuten'});
- }
-}
-
-
-/**
- * @param {number} codePoint
- * @param {import('japanese-util').CodepointRange} range
- * @returns {boolean}
- */
-function isCodePointInRange(codePoint, [min, max]) {
- return (codePoint >= min && codePoint <= max);
-}
-
-/**
- * @param {number} codePoint
- * @param {import('japanese-util').CodepointRange[]} ranges
- * @returns {boolean}
- */
-function isCodePointInRanges(codePoint, ranges) {
- for (const [min, max] of ranges) {
- if (codePoint >= min && codePoint <= max) {
- return true;
- }
- }
- return false;
-}
-
-/**
- * @param {string} previousCharacter
- * @returns {?string}
- */
-function getProlongedHiragana(previousCharacter) {
- switch (KANA_TO_VOWEL_MAPPING.get(previousCharacter)) {
- case 'a': return 'あ';
- case 'i': return 'い';
- case 'u': return 'う';
- case 'e': return 'え';
- case 'o': return 'う';
- default: return null;
- }
-}
-
-
-export class JapaneseUtil {
- /**
- * @param {?import('wanakana')|import('../../../lib/wanakana.js')} wanakana
- */
- constructor(wanakana = null) {
- /** @type {?import('wanakana')} */
- this._wanakana = /** @type {import('wanakana')} */ (wanakana);
- }
-
- // Character code testing functions
-
- /**
- * @param {number} codePoint
- * @returns {boolean}
- */
- isCodePointKanji(codePoint) {
- return isCodePointInRanges(codePoint, CJK_IDEOGRAPH_RANGES);
- }
-
- /**
- * @param {number} codePoint
- * @returns {boolean}
- */
- isCodePointKana(codePoint) {
- return isCodePointInRanges(codePoint, KANA_RANGES);
- }
-
- /**
- * @param {number} codePoint
- * @returns {boolean}
- */
- isCodePointJapanese(codePoint) {
- return isCodePointInRanges(codePoint, JAPANESE_RANGES);
- }
-
- // String testing functions
-
- /**
- * @param {string} str
- * @returns {boolean}
- */
- isStringEntirelyKana(str) {
- if (str.length === 0) { return false; }
- for (const c of str) {
- if (!isCodePointInRanges(/** @type {number} */ (c.codePointAt(0)), KANA_RANGES)) {
- return false;
- }
- }
- return true;
- }
-
- /**
- * @param {string} str
- * @returns {boolean}
- */
- isStringPartiallyJapanese(str) {
- if (str.length === 0) { return false; }
- for (const c of str) {
- if (isCodePointInRanges(/** @type {number} */ (c.codePointAt(0)), JAPANESE_RANGES)) {
- return true;
- }
- }
- return false;
- }
-
- // Mora functions
-
- /**
- * @param {number} moraIndex
- * @param {number} pitchAccentDownstepPosition
- * @returns {boolean}
- */
- isMoraPitchHigh(moraIndex, pitchAccentDownstepPosition) {
- switch (pitchAccentDownstepPosition) {
- case 0: return (moraIndex > 0);
- case 1: return (moraIndex < 1);
- default: return (moraIndex > 0 && moraIndex < pitchAccentDownstepPosition);
- }
- }
-
- /**
- * @param {string} text
- * @param {number} pitchAccentDownstepPosition
- * @param {boolean} isVerbOrAdjective
- * @returns {?import('japanese-util').PitchCategory}
- */
- getPitchCategory(text, pitchAccentDownstepPosition, isVerbOrAdjective) {
- if (pitchAccentDownstepPosition === 0) {
- return 'heiban';
- }
- if (isVerbOrAdjective) {
- return pitchAccentDownstepPosition > 0 ? 'kifuku' : null;
- }
- if (pitchAccentDownstepPosition === 1) {
- return 'atamadaka';
- }
- if (pitchAccentDownstepPosition > 1) {
- return pitchAccentDownstepPosition >= this.getKanaMoraCount(text) ? 'odaka' : 'nakadaka';
- }
- return null;
- }
-
- /**
- * @param {string} text
- * @returns {string[]}
- */
- getKanaMorae(text) {
- const morae = [];
- let i;
- for (const c of text) {
- if (SMALL_KANA_SET.has(c) && (i = morae.length) > 0) {
- morae[i - 1] += c;
- } else {
- morae.push(c);
- }
- }
- return morae;
- }
-
- /**
- * @param {string} text
- * @returns {number}
- */
- getKanaMoraCount(text) {
- let moraCount = 0;
- for (const c of text) {
- if (!(SMALL_KANA_SET.has(c) && moraCount > 0)) {
- ++moraCount;
- }
- }
- return moraCount;
- }
-
- // Conversion functions
-
- /**
- * @param {string} text
- * @returns {string}
- */
- convertToKana(text) {
- return this._getWanakana().toKana(text);
- }
-
- /**
- * @returns {boolean}
- */
- convertToKanaSupported() {
- return this._wanakana !== null;
- }
-
- /**
- * @param {string} text
- * @param {boolean} [keepProlongedSoundMarks]
- * @returns {string}
- */
- convertKatakanaToHiragana(text, keepProlongedSoundMarks = false) {
- let result = '';
- const offset = (HIRAGANA_CONVERSION_RANGE[0] - KATAKANA_CONVERSION_RANGE[0]);
- for (let char of text) {
- const codePoint = /** @type {number} */ (char.codePointAt(0));
- switch (codePoint) {
- case KATAKANA_SMALL_KA_CODE_POINT:
- case KATAKANA_SMALL_KE_CODE_POINT:
- // No change
- break;
- case KANA_PROLONGED_SOUND_MARK_CODE_POINT:
- if (!keepProlongedSoundMarks && result.length > 0) {
- const char2 = getProlongedHiragana(result[result.length - 1]);
- if (char2 !== null) { char = char2; }
- }
- break;
- default:
- if (isCodePointInRange(codePoint, KATAKANA_CONVERSION_RANGE)) {
- char = String.fromCodePoint(codePoint + offset);
- }
- break;
- }
- result += char;
- }
- return result;
- }
-
- /**
- * @param {string} text
- * @returns {string}
- */
- convertHiraganaToKatakana(text) {
- let result = '';
- const offset = (KATAKANA_CONVERSION_RANGE[0] - HIRAGANA_CONVERSION_RANGE[0]);
- for (let char of text) {
- const codePoint = /** @type {number} */ (char.codePointAt(0));
- if (isCodePointInRange(codePoint, HIRAGANA_CONVERSION_RANGE)) {
- char = String.fromCodePoint(codePoint + offset);
- }
- result += char;
- }
- return result;
- }
-
- /**
- * @param {string} text
- * @returns {string}
- */
- convertToRomaji(text) {
- const wanakana = this._getWanakana();
- return wanakana.toRomaji(text);
- }
-
- /**
- * @returns {boolean}
- */
- convertToRomajiSupported() {
- return this._wanakana !== null;
- }
-
- /**
- * @param {string} text
- * @returns {string}
- */
- convertNumericToFullWidth(text) {
- let result = '';
- for (const char of text) {
- let c = /** @type {number} */ (char.codePointAt(0));
- if (c >= 0x30 && c <= 0x39) { // ['0', '9']
- c += 0xff10 - 0x30; // 0xff10 = '0' full width
- result += String.fromCodePoint(c);
- } else {
- result += char;
- }
- }
- return result;
- }
-
- /**
- * @param {string} text
- * @param {?import('../../general/text-source-map.js').TextSourceMap} [sourceMap]
- * @returns {string}
- */
- convertHalfWidthKanaToFullWidth(text, sourceMap = null) {
- 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;
- break;
- }
-
- let c2 = mapping[index];
- if (index > 0) {
- if (c2 === '-') { // invalid
- index = 0;
- c2 = mapping[0];
- } else {
- ++i;
- }
- }
-
- if (sourceMap !== null && index > 0) {
- sourceMap.combine(result.length, 1);
- }
- result += c2;
- }
-
- return result;
- }
-
- /**
- * @param {string} text
- * @param {?import('../../general/text-source-map.js').TextSourceMap} sourceMap
- * @returns {string}
- */
- convertAlphabeticToKana(text, sourceMap = null) {
- let part = '';
- let result = '';
-
- for (const char of text) {
- // Note: 0x61 is the character code for 'a'
- let c = /** @type {number} */ (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; // '-'
- } else {
- if (part.length > 0) {
- result += this._convertAlphabeticPartToKana(part, sourceMap, result.length);
- part = '';
- }
- result += char;
- continue;
- }
- part += String.fromCodePoint(c);
- }
-
- if (part.length > 0) {
- result += this._convertAlphabeticPartToKana(part, sourceMap, result.length);
- }
- return result;
- }
-
- /**
- * @returns {boolean}
- */
- convertAlphabeticToKanaSupported() {
- return this._wanakana !== null;
- }
-
- /**
- * @param {string} character
- * @returns {?{character: string, type: import('japanese-util').DiacriticType}}
- */
- getKanaDiacriticInfo(character) {
- const info = DIACRITIC_MAPPING.get(character);
- return typeof info !== 'undefined' ? {character: info.character, type: info.type} : null;
- }
-
- // Furigana distribution
-
- /**
- * @param {string} term
- * @param {string} reading
- * @returns {import('japanese-util').FuriganaSegment[]}
- */
- distributeFurigana(term, reading) {
- if (reading === term) {
- // Same
- return [this._createFuriganaSegment(term, '')];
- }
-
- /** @type {import('japanese-util').FuriganaGroup[]} */
- const groups = [];
- /** @type {?import('japanese-util').FuriganaGroup} */
- let groupPre = null;
- let isKanaPre = null;
- for (const c of term) {
- const codePoint = /** @type {number} */ (c.codePointAt(0));
- const isKana = this.isCodePointKana(codePoint);
- if (isKana === isKanaPre) {
- /** @type {import('japanese-util').FuriganaGroup} */ (groupPre).text += c;
- } else {
- groupPre = {isKana, text: c, textNormalized: null};
- groups.push(groupPre);
- isKanaPre = isKana;
- }
- }
- for (const group of groups) {
- if (group.isKana) {
- group.textNormalized = this.convertKatakanaToHiragana(group.text);
- }
- }
-
- const readingNormalized = this.convertKatakanaToHiragana(reading);
- const segments = this._segmentizeFurigana(reading, readingNormalized, groups, 0);
- if (segments !== null) {
- return segments;
- }
-
- // Fallback
- return [this._createFuriganaSegment(term, reading)];
- }
-
- /**
- * @param {string} term
- * @param {string} reading
- * @param {string} source
- * @returns {import('japanese-util').FuriganaSegment[]}
- */
- distributeFuriganaInflected(term, reading, source) {
- const termNormalized = this.convertKatakanaToHiragana(term);
- const readingNormalized = this.convertKatakanaToHiragana(reading);
- const sourceNormalized = this.convertKatakanaToHiragana(source);
-
- let mainText = term;
- let stemLength = this._getStemLength(termNormalized, sourceNormalized);
-
- // Check if source is derived from the reading instead of the term
- const readingStemLength = this._getStemLength(readingNormalized, sourceNormalized);
- if (readingStemLength > 0 && readingStemLength >= stemLength) {
- mainText = reading;
- stemLength = readingStemLength;
- reading = `${source.substring(0, stemLength)}${reading.substring(stemLength)}`;
- }
-
- const segments = [];
- if (stemLength > 0) {
- mainText = `${source.substring(0, stemLength)}${mainText.substring(stemLength)}`;
- const segments2 = this.distributeFurigana(mainText, reading);
- let consumed = 0;
- for (const segment of segments2) {
- const {text} = segment;
- const start = consumed;
- consumed += text.length;
- if (consumed < stemLength) {
- segments.push(segment);
- } else if (consumed === stemLength) {
- segments.push(segment);
- break;
- } else {
- if (start < stemLength) {
- segments.push(this._createFuriganaSegment(mainText.substring(start, stemLength), ''));
- }
- break;
- }
- }
- }
-
- if (stemLength < source.length) {
- const remainder = source.substring(stemLength);
- const segmentCount = segments.length;
- if (segmentCount > 0 && segments[segmentCount - 1].reading.length === 0) {
- // Append to the last segment if it has an empty reading
- segments[segmentCount - 1].text += remainder;
- } else {
- // Otherwise, create a new segment
- segments.push(this._createFuriganaSegment(remainder, ''));
- }
- }
-
- return segments;
- }
-
- // Miscellaneous
-
- /**
- * @param {string} text
- * @param {boolean} fullCollapse
- * @param {?import('../../general/text-source-map.js').TextSourceMap} [sourceMap]
- * @returns {string}
- */
- collapseEmphaticSequences(text, fullCollapse, sourceMap = null) {
- let result = '';
- let collapseCodePoint = -1;
- const hasSourceMap = (sourceMap !== null);
- for (const char of text) {
- const c = char.codePointAt(0);
- if (
- c === HIRAGANA_SMALL_TSU_CODE_POINT ||
- c === KATAKANA_SMALL_TSU_CODE_POINT ||
- c === KANA_PROLONGED_SOUND_MARK_CODE_POINT
- ) {
- if (collapseCodePoint !== c) {
- collapseCodePoint = c;
- if (!fullCollapse) {
- result += char;
- continue;
- }
- }
- } else {
- collapseCodePoint = -1;
- result += char;
- continue;
- }
-
- if (hasSourceMap) {
- sourceMap.combine(Math.max(0, result.length - 1), 1);
- }
- }
- return result;
- }
-
- // Private
-
- /**
- * @param {string} text
- * @param {string} reading
- * @returns {import('japanese-util').FuriganaSegment}
- */
- _createFuriganaSegment(text, reading) {
- return {text, reading};
- }
-
- /**
- * @param {string} reading
- * @param {string} readingNormalized
- * @param {import('japanese-util').FuriganaGroup[]} groups
- * @param {number} groupsStart
- * @returns {?(import('japanese-util').FuriganaSegment[])}
- */
- _segmentizeFurigana(reading, readingNormalized, groups, groupsStart) {
- const groupCount = groups.length - groupsStart;
- if (groupCount <= 0) {
- return reading.length === 0 ? [] : null;
- }
-
- const group = groups[groupsStart];
- const {isKana, text} = group;
- const textLength = text.length;
- if (isKana) {
- const {textNormalized} = group;
- if (textNormalized !== null && readingNormalized.startsWith(textNormalized)) {
- const segments = this._segmentizeFurigana(
- reading.substring(textLength),
- readingNormalized.substring(textLength),
- groups,
- groupsStart + 1
- );
- if (segments !== null) {
- if (reading.startsWith(text)) {
- segments.unshift(this._createFuriganaSegment(text, ''));
- } else {
- segments.unshift(...this._getFuriganaKanaSegments(text, reading));
- }
- return segments;
- }
- }
- return null;
- } else {
- let result = null;
- for (let i = reading.length; i >= textLength; --i) {
- const segments = this._segmentizeFurigana(
- reading.substring(i),
- readingNormalized.substring(i),
- groups,
- groupsStart + 1
- );
- if (segments !== null) {
- if (result !== null) {
- // More than one way to segmentize the tail; mark as ambiguous
- return null;
- }
- const segmentReading = reading.substring(0, i);
- segments.unshift(this._createFuriganaSegment(text, segmentReading));
- result = segments;
- }
- // There is only one way to segmentize the last non-kana group
- if (groupCount === 1) {
- break;
- }
- }
- return result;
- }
- }
-
- /**
- * @param {string} text
- * @param {string} reading
- * @returns {import('japanese-util').FuriganaSegment[]}
- */
- _getFuriganaKanaSegments(text, reading) {
- const textLength = text.length;
- const newSegments = [];
- let start = 0;
- let state = (reading[0] === text[0]);
- for (let i = 1; i < textLength; ++i) {
- const newState = (reading[i] === text[i]);
- if (state === newState) { continue; }
- newSegments.push(this._createFuriganaSegment(text.substring(start, i), state ? '' : reading.substring(start, i)));
- state = newState;
- start = i;
- }
- newSegments.push(this._createFuriganaSegment(text.substring(start, textLength), state ? '' : reading.substring(start, textLength)));
- return newSegments;
- }
-
- /**
- * @returns {import('wanakana')}
- * @throws {Error}
- */
- _getWanakana() {
- const wanakana = this._wanakana;
- if (wanakana === null) { throw new Error('Functions which use WanaKana are not supported in this context'); }
- return wanakana;
- }
-
- /**
- * @param {string} text
- * @param {?import('../../general/text-source-map.js').TextSourceMap} sourceMap
- * @param {number} sourceMapStart
- * @returns {string}
- */
- _convertAlphabeticPartToKana(text, sourceMap, sourceMapStart) {
- const wanakana = this._getWanakana();
- const result = wanakana.toHiragana(text);
-
- // Generate source mapping
- if (sourceMap !== null) {
- 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;
- }
-
- // Merge characters
- const removals = iNext - i - 1;
- if (removals > 0) {
- sourceMap.combine(sourceMapStart, removals);
- }
- ++sourceMapStart;
-
- // Empty elements
- const additions = resultPosNext - resultPos - 1;
- for (let j = 0; j < additions; ++j) {
- sourceMap.insert(sourceMapStart, 0);
- ++sourceMapStart;
- }
-
- i = iNext;
- resultPos = resultPosNext;
- }
- }
-
- return result;
- }
-
- /**
- * @param {string} text1
- * @param {string} text2
- * @returns {number}
- */
- _getStemLength(text1, text2) {
- const minLength = Math.min(text1.length, text2.length);
- if (minLength === 0) { return 0; }
-
- let i = 0;
- while (true) {
- const char1 = /** @type {number} */ (text1.codePointAt(i));
- const char2 = /** @type {number} */ (text2.codePointAt(i));
- if (char1 !== char2) { break; }
- const charLength = String.fromCodePoint(char1).length;
- i += charLength;
- if (i >= minLength) {
- if (i > minLength) {
- i -= charLength; // Don't consume partial UTF16 surrogate characters
- }
- break;
- }
- }
- return i;
- }
-}
diff --git a/ext/js/language/translator.js b/ext/js/language/translator.js
index cedc7d3d..66eeb69f 100644
--- a/ext/js/language/translator.js
+++ b/ext/js/language/translator.js
@@ -19,6 +19,8 @@
import {RegexUtil} from '../general/regex-util.js';
import {TextSourceMap} from '../general/text-source-map.js';
import {Deinflector} from './deinflector.js';
+import {convertAlphabeticToKana} from './japanese-wanakana.js';
+import {collapseEmphaticSequences, convertHalfWidthKanaToFullWidth, convertHiraganaToKatakana, convertKatakanaToHiragana, convertNumericToFullWidth, isCodePointJapanese} from './japanese.js';
/**
* Class which finds term and kanji dictionary entries for text.
@@ -28,9 +30,7 @@ export class Translator {
* Creates a new Translator instance.
* @param {import('translator').ConstructorDetails} details The details for the class.
*/
- constructor({japaneseUtil, database}) {
- /** @type {import('./sandbox/japanese-util.js').JapaneseUtil} */
- this._japaneseUtil = japaneseUtil;
+ constructor({database}) {
/** @type {import('../dictionary/dictionary-database.js').DictionaryDatabase} */
this._database = database;
/** @type {?Deinflector} */
@@ -436,7 +436,6 @@ export class Translator {
this._getCollapseEmphaticOptions(options)
];
- const jp = this._japaneseUtil;
/** @type {import('translation-internal').DatabaseDeinflection[]} */
const deinflections = [];
const used = new Set();
@@ -447,22 +446,22 @@ export class Translator {
text2 = this._applyTextReplacements(text2, sourceMap, textReplacements);
}
if (halfWidth) {
- text2 = jp.convertHalfWidthKanaToFullWidth(text2, sourceMap);
+ text2 = convertHalfWidthKanaToFullWidth(text2, sourceMap);
}
if (numeric) {
- text2 = jp.convertNumericToFullWidth(text2);
+ text2 = convertNumericToFullWidth(text2);
}
if (alphabetic) {
- text2 = jp.convertAlphabeticToKana(text2, sourceMap);
+ text2 = convertAlphabeticToKana(text2, sourceMap);
}
if (katakana) {
- text2 = jp.convertHiraganaToKatakana(text2);
+ text2 = convertHiraganaToKatakana(text2);
}
if (hiragana) {
- text2 = jp.convertKatakanaToHiragana(text2);
+ text2 = convertKatakanaToHiragana(text2);
}
if (collapseEmphatic) {
- text2 = jp.collapseEmphaticSequences(text2, collapseEmphaticFull, sourceMap);
+ text2 = collapseEmphaticSequences(text2, collapseEmphaticFull, sourceMap);
}
for (
@@ -519,10 +518,9 @@ export class Translator {
* @returns {string}
*/
_getJapaneseOnlyText(text) {
- const jp = this._japaneseUtil;
let length = 0;
for (const c of text) {
- if (!jp.isCodePointJapanese(/** @type {number} */ (c.codePointAt(0)))) {
+ if (!isCodePointJapanese(/** @type {number} */ (c.codePointAt(0)))) {
return text.substring(0, length);
}
length += c.length;