Convert Japanese utilities to a module-like style

ext/bg/js/audio-uri-builder.js

@@ -17,7 +17,7 @@
  */
 
 /* global
- * jpIsStringEntirelyKana
+ * jp
  */
 
 class AudioUriBuilder {
@@ -66,7 +66,7 @@ class AudioUriBuilder {
         let kana = definition.reading;
         let kanji = definition.expression;
 
-        if (!kana && jpIsStringEntirelyKana(kanji)) {
+        if (!kana && jp.isStringEntirelyKana(kanji)) {
             kana = kanji;
             kanji = null;
         }

ext/bg/js/backend.js

@@ -32,9 +32,7 @@
  * dictEnabledSet
  * dictTermsSort
  * handlebarsRenderDynamic
- * jpConvertReading
+ * jp
- * jpDistributeFuriganaInflected
- * jpKatakanaToHiragana
  * optionsLoad
  * optionsSave
  * profileConditionsDescriptor
@@ -402,13 +400,13 @@ class Backend {
                 dictTermsSort(definitions);
                 const {expression, reading} = definitions[0];
                 const source = text.substring(0, sourceLength);
-                for (const {text: text2, furigana} of jpDistributeFuriganaInflected(expression, reading, source)) {
+                for (const {text: text2, furigana} of jp.distributeFuriganaInflected(expression, reading, source)) {
-                    const reading2 = jpConvertReading(text2, furigana, options.parsing.readingMode);
+                    const reading2 = jp.convertReading(text2, furigana, options.parsing.readingMode);
                     term.push({text: text2, reading: reading2});
                 }
                 text = text.substring(source.length);
             } else {
-                const reading = jpConvertReading(text[0], null, options.parsing.readingMode);
+                const reading = jp.convertReading(text[0], null, options.parsing.readingMode);
                 term.push({text: text[0], reading});
                 text = text.substring(1);
             }
@@ -427,16 +425,16 @@ class Backend {
                 for (const {expression, reading, source} of parsedLine) {
                     const term = [];
                     if (expression !== null && reading !== null) {
-                        for (const {text: text2, furigana} of jpDistributeFuriganaInflected(
+                        for (const {text: text2, furigana} of jp.distributeFuriganaInflected(
                             expression,
-                            jpKatakanaToHiragana(reading),
+                            jp.convertKatakanaToHiragana(reading),
                             source
                         )) {
-                            const reading2 = jpConvertReading(text2, furigana, options.parsing.readingMode);
+                            const reading2 = jp.convertReading(text2, furigana, options.parsing.readingMode);
                             term.push({text: text2, reading: reading2});
                         }
                     } else {
-                        const reading2 = jpConvertReading(source, null, options.parsing.readingMode);
+                        const reading2 = jp.convertReading(source, null, options.parsing.readingMode);
                         term.push({text: source, reading: reading2});
                     }
                     result.push(term);

ext/bg/js/clipboard-monitor.js

@@ -17,7 +17,7 @@
  */
 
 /* global
- * jpIsStringPartiallyJapanese
+ * jp
  */
 
 class ClipboardMonitor extends EventDispatcher {
@@ -54,7 +54,7 @@ class ClipboardMonitor extends EventDispatcher {
                 text !== this._previousText
             ) {
                 this._previousText = text;
-                if (jpIsStringPartiallyJapanese(text)) {
+                if (jp.isStringPartiallyJapanese(text)) {
                     this.trigger('change', {text});
                 }
             }

ext/bg/js/handlebars.js

@@ -18,8 +18,7 @@
 
 /* global
  * Handlebars
- * jpDistributeFurigana
+ * jp
- * jpIsCodePointKanji
  */
 
 function handlebarsEscape(text) {
@@ -33,7 +32,7 @@ function handlebarsDumpObject(options) {
 
 function handlebarsFurigana(options) {
     const definition = options.fn(this);
-    const segs = jpDistributeFurigana(definition.expression, definition.reading);
+    const segs = jp.distributeFurigana(definition.expression, definition.reading);
 
     let result = '';
     for (const seg of segs) {
@@ -49,7 +48,7 @@ function handlebarsFurigana(options) {
 
 function handlebarsFuriganaPlain(options) {
     const definition = options.fn(this);
-    const segs = jpDistributeFurigana(definition.expression, definition.reading);
+    const segs = jp.distributeFurigana(definition.expression, definition.reading);
 
     let result = '';
     for (const seg of segs) {
@@ -66,7 +65,7 @@ function handlebarsFuriganaPlain(options) {
 function handlebarsKanjiLinks(options) {
     let result = '';
     for (const c of options.fn(this)) {
-        if (jpIsCodePointKanji(c.codePointAt(0))) {
+        if (jp.isCodePointKanji(c.codePointAt(0))) {
             result += `<a href="#" class="kanji-link">${c}</a>`;
         } else {
             result += c;

ext/bg/js/japanese.js

@@ -20,7 +20,8 @@
  * wanakana
  */
 
-const JP_HALFWIDTH_KATAKANA_MAPPING = new Map([
+const jp = (() => {
+    const HALFWIDTH_KATAKANA_MAPPING = new Map([
         ['ｦ', 'ヲヺ-'],
         ['ｧ', 'ァ--'],
         ['ｨ', 'ィ--'],
@@ -77,25 +78,25 @@ const JP_HALFWIDTH_KATAKANA_MAPPING = new Map([
         ['ﾛ', 'ロ--'],
         ['ﾜ', 'ワ--'],
         ['ﾝ', 'ン--']
-]);
+    ]);
 
-const JP_HIRAGANA_RANGE = [0x3040, 0x309f];
+    const HIRAGANA_RANGE = [0x3040, 0x309f];
-const JP_KATAKANA_RANGE = [0x30a0, 0x30ff];
+    const KATAKANA_RANGE = [0x30a0, 0x30ff];
-const JP_KANA_RANGES = [JP_HIRAGANA_RANGE, JP_KATAKANA_RANGE];
+    const KANA_RANGES = [HIRAGANA_RANGE, KATAKANA_RANGE];
 
-const JP_CJK_COMMON_RANGE = [0x4e00, 0x9fff];
+    const CJK_COMMON_RANGE = [0x4e00, 0x9fff];
-const JP_CJK_RARE_RANGE = [0x3400, 0x4dbf];
+    const CJK_RARE_RANGE = [0x3400, 0x4dbf];
-const JP_CJK_RANGES = [JP_CJK_COMMON_RANGE, JP_CJK_RARE_RANGE];
+    const CJK_RANGES = [CJK_COMMON_RANGE, CJK_RARE_RANGE];
 
-const JP_ITERATION_MARK_CHAR_CODE = 0x3005;
+    const ITERATION_MARK_CODE_POINT = 0x3005;
 
-// Japanese character ranges, roughly ordered in order of expected frequency
+    // Japanese character ranges, roughly ordered in order of expected frequency
-const JP_JAPANESE_RANGES = [
+    const JAPANESE_RANGES = [
-    JP_HIRAGANA_RANGE,
+        HIRAGANA_RANGE,
-    JP_KATAKANA_RANGE,
+        KATAKANA_RANGE,
 
-    JP_CJK_COMMON_RANGE,
+        CJK_COMMON_RANGE,
-    JP_CJK_RARE_RANGE,
+        CJK_RARE_RANGE,
 
         [0xff66, 0xff9f], // Halfwidth katakana
 
@@ -112,62 +113,59 @@ const JP_JAPANESE_RANGES = [
         [0xff3b, 0xff3f], // Fullwidth punctuation 3
         [0xff5b, 0xff60], // Fullwidth punctuation 4
         [0xffe0, 0xffee]  // Currency markers
-];
+    ];
 
 
-// Helper functions
+    // Character code testing functions
 
-function _jpIsCodePointInRanges(codePoint, ranges) {
+    function isCodePointKanji(codePoint) {
+        return isCodePointInRanges(codePoint, CJK_RANGES);
+    }
+
+    function isCodePointKana(codePoint) {
+        return isCodePointInRanges(codePoint, KANA_RANGES);
+    }
+
+    function isCodePointJapanese(codePoint) {
+        return isCodePointInRanges(codePoint, JAPANESE_RANGES);
+    }
+
+    function isCodePointInRanges(codePoint, ranges) {
         for (const [min, max] of ranges) {
             if (codePoint >= min && codePoint <= max) {
                 return true;
             }
         }
         return false;
-}
+    }
 
 
-// Character code testing functions
+    // String testing functions
 
-function jpIsCodePointKanji(codePoint) {
+    function isStringEntirelyKana(str) {
-    return _jpIsCodePointInRanges(codePoint, JP_CJK_RANGES);
-}
-
-function jpIsCodePointKana(codePoint) {
-    return _jpIsCodePointInRanges(codePoint, JP_KANA_RANGES);
-}
-
-function jpIsCodePointJapanese(codePoint) {
-    return _jpIsCodePointInRanges(codePoint, JP_JAPANESE_RANGES);
-}
-
-
-// String testing functions
-
-function jpIsStringEntirelyKana(str) {
         if (str.length === 0) { return false; }
         for (const c of str) {
-        if (!jpIsCodePointKana(c.codePointAt(0))) {
+            if (!isCodePointKana(c.codePointAt(0))) {
                 return false;
             }
         }
         return true;
-}
+    }
 
-function jpIsStringPartiallyJapanese(str) {
+    function isStringPartiallyJapanese(str) {
         if (str.length === 0) { return false; }
         for (const c of str) {
-        if (jpIsCodePointJapanese(c.codePointAt(0))) {
+            if (isCodePointJapanese(c.codePointAt(0))) {
                 return true;
             }
         }
         return false;
-}
+    }
 
 
-// Conversion functions
+    // Conversion functions
 
-function jpKatakanaToHiragana(text) {
+    function convertKatakanaToHiragana(text) {
         let result = '';
         for (const c of text) {
             if (wanakana.isKatakana(c)) {
@@ -178,9 +176,9 @@ function jpKatakanaToHiragana(text) {
         }
 
         return result;
-}
+    }
 
-function jpHiraganaToKatakana(text) {
+    function convertHiraganaToKatakana(text) {
         let result = '';
         for (const c of text) {
             if (wanakana.isHiragana(c)) {
@@ -191,24 +189,24 @@ function jpHiraganaToKatakana(text) {
         }
 
         return result;
-}
+    }
 
-function jpToRomaji(text) {
+    function convertToRomaji(text) {
         return wanakana.toRomaji(text);
-}
+    }
 
-function jpConvertReading(expressionFragment, readingFragment, readingMode) {
+    function convertReading(expressionFragment, readingFragment, readingMode) {
         switch (readingMode) {
             case 'hiragana':
-            return jpKatakanaToHiragana(readingFragment || '');
+                return convertKatakanaToHiragana(readingFragment || '');
             case 'katakana':
-            return jpHiraganaToKatakana(readingFragment || '');
+                return convertHiraganaToKatakana(readingFragment || '');
             case 'romaji':
                 if (readingFragment) {
-                return jpToRomaji(readingFragment);
+                    return convertToRomaji(readingFragment);
                 } else {
-                if (jpIsStringEntirelyKana(expressionFragment)) {
+                    if (isStringEntirelyKana(expressionFragment)) {
-                    return jpToRomaji(expressionFragment);
+                        return convertToRomaji(expressionFragment);
                     }
                 }
                 return readingFragment;
@@ -217,101 +215,23 @@ function jpConvertReading(expressionFragment, readingFragment, readingMode) {
             default:
                 return readingFragment;
         }
-}
-
-function jpDistributeFurigana(expression, reading) {
-    const fallback = [{furigana: reading, text: expression}];
-    if (!reading) {
-        return fallback;
     }
 
-    let isAmbiguous = false;
+    function convertNumericTofullWidth(text) {
-    const segmentize = (reading2, groups) => {
+        let result = '';
-        if (groups.length === 0 || isAmbiguous) {
+        for (const char of text) {
-            return [];
+            let c = char.codePointAt(0);
-        }
+            if (c >= 0x30 && c <= 0x39) { // ['0', '9']
-
+                c += 0xff10 - 0x30; // 0xff10 = '0' full width
-        const group = groups[0];
+                result += String.fromCodePoint(c);
-        if (group.mode === 'kana') {
-            if (jpKatakanaToHiragana(reading2).startsWith(jpKatakanaToHiragana(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;
+                result += char;
-            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;
+        return result;
-        }
-    };
-
-    const groups = [];
-    let modePrev = null;
-    for (const c of expression) {
-        const codePoint = c.codePointAt(0);
-        const modeCurr = jpIsCodePointKanji(codePoint) || codePoint === JP_ITERATION_MARK_CHAR_CODE ? 'kanji' : 'kana';
-        if (modeCurr === modePrev) {
-            groups[groups.length - 1].text += c;
-        } else {
-            groups.push({mode: modeCurr, text: c});
-            modePrev = modeCurr;
-        }
     }
 
-    const segments = segmentize(reading, groups);
+    function convertHalfWidthKanaToFullWidth(text, sourceMapping) {
-    if (segments && !isAmbiguous) {
-        return segments;
-    }
-    return fallback;
-}
-
-function jpDistributeFuriganaInflected(expression, reading, source) {
-    const output = [];
-
-    let stemLength = 0;
-    const shortest = Math.min(source.length, expression.length);
-    const sourceHiragana = jpKatakanaToHiragana(source);
-    const expressionHiragana = jpKatakanaToHiragana(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 jpDistributeFurigana(stemExpression, stemReading)) {
-        output.push(segment);
-    }
-
-    if (stemLength !== source.length) {
-        output.push({text: source.substring(stemLength)});
-    }
-
-    return output;
-}
-
-function jpConvertHalfWidthKanaToFullWidth(text, sourceMapping) {
         let result = '';
         const hasSourceMapping = Array.isArray(sourceMapping);
 
@@ -319,7 +239,7 @@ function jpConvertHalfWidthKanaToFullWidth(text, sourceMapping) {
         // 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 = JP_HALFWIDTH_KATAKANA_MAPPING.get(c);
+            const mapping = HALFWIDTH_KATAKANA_MAPPING.get(c);
             if (typeof mapping !== 'string') {
                 result += c;
                 continue;
@@ -354,23 +274,9 @@ function jpConvertHalfWidthKanaToFullWidth(text, sourceMapping) {
         }
 
         return result;
-}
-
-function jpConvertNumericTofullWidth(text) {
-    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;
     }
-    }
-    return result;
-}
 
-function jpConvertAlphabeticToKana(text, sourceMapping) {
+    function convertAlphabeticToKana(text, sourceMapping) {
         let part = '';
         let result = '';
         const ii = text.length;
@@ -395,7 +301,7 @@ function jpConvertAlphabeticToKana(text, sourceMapping) {
                 c = 0x2d; // '-'
             } else {
                 if (part.length > 0) {
-                result += jpToHiragana(part, sourceMapping, result.length);
+                    result += convertAlphabeticPartToKana(part, sourceMapping, result.length);
                     part = '';
                 }
                 result += char;
@@ -405,12 +311,12 @@ function jpConvertAlphabeticToKana(text, sourceMapping) {
         }
 
         if (part.length > 0) {
-        result += jpToHiragana(part, sourceMapping, result.length);
+            result += convertAlphabeticPartToKana(part, sourceMapping, result.length);
         }
         return result;
-}
+    }
 
-function jpToHiragana(text, sourceMapping, sourceMappingStart) {
+    function convertAlphabeticPartToKana(text, sourceMapping, sourceMappingStart) {
         const result = wanakana.toHiragana(text);
 
         // Generate source mapping
@@ -455,4 +361,120 @@ function jpToHiragana(text, sourceMapping, sourceMappingStart) {
         }
 
         return result;
-}
+    }
+
+
+    // Furigana distribution
+
+    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 [];
+            }
+
+            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;
+            }
+        }
+
+        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);
+        }
+
+        if (stemLength !== source.length) {
+            output.push({text: source.substring(stemLength)});
+        }
+
+        return output;
+    }
+
+
+    // Exports
+
+    return {
+        isCodePointKanji,
+        isCodePointKana,
+        isCodePointJapanese,
+        isStringEntirelyKana,
+        isStringPartiallyJapanese,
+        convertKatakanaToHiragana,
+        convertHiraganaToKatakana,
+        convertToRomaji,
+        convertReading,
+        convertNumericTofullWidth,
+        convertHalfWidthKanaToFullWidth,
+        convertAlphabeticToKana,
+        distributeFurigana,
+        distributeFuriganaInflected
+    };
+})();

ext/bg/js/translator.js

@@ -29,13 +29,7 @@
  * dictTermsMergeBySequence
  * dictTermsSort
  * dictTermsUndupe
- * jpConvertAlphabeticToKana
+ * jp
- * jpConvertHalfWidthKanaToFullWidth
- * jpConvertNumericTofullWidth
- * jpDistributeFurigana
- * jpHiraganaToKatakana
- * jpIsCodePointJapanese
- * jpKatakanaToHiragana
  * requestJson
  */
 
@@ -275,7 +269,7 @@ class Translator {
                 const termTags = await this.expandTags(definition.termTags, definition.dictionary);
 
                 const {expression, reading} = definition;
-                const furiganaSegments = jpDistributeFurigana(expression, reading);
+                const furiganaSegments = jp.distributeFurigana(expression, reading);
 
                 definitions.push({
                     source: deinflection.source,
@@ -376,20 +370,20 @@ class Translator {
             let sourceMapping = null;
             if (halfWidth) {
                 if (sourceMapping === null) { sourceMapping = Translator.createTextSourceMapping(text2); }
-                text2 = jpConvertHalfWidthKanaToFullWidth(text2, sourceMapping);
+                text2 = jp.convertHalfWidthKanaToFullWidth(text2, sourceMapping);
             }
             if (numeric) {
-                text2 = jpConvertNumericTofullWidth(text2);
+                text2 = jp.convertNumericTofullWidth(text2);
             }
             if (alphabetic) {
                 if (sourceMapping === null) { sourceMapping = Translator.createTextSourceMapping(text2); }
-                text2 = jpConvertAlphabeticToKana(text2, sourceMapping);
+                text2 = jp.convertAlphabeticToKana(text2, sourceMapping);
             }
             if (katakana) {
-                text2 = jpHiraganaToKatakana(text2);
+                text2 = jp.convertHiraganaToKatakana(text2);
             }
             if (hiragana) {
-                text2 = jpKatakanaToHiragana(text2);
+                text2 = jp.convertKatakanaToHiragana(text2);
             }
 
             for (let i = text2.length; i > 0; --i) {
@@ -590,7 +584,7 @@ class Translator {
     }
 
     static createExpression(expression, reading, termTags=null, termFrequency=null) {
-        const furiganaSegments = jpDistributeFurigana(expression, reading);
+        const furiganaSegments = jp.distributeFurigana(expression, reading);
         return {
             expression,
             reading,
@@ -639,7 +633,7 @@ class Translator {
         if (!options.scanning.alphanumeric) {
             let newText = '';
             for (const c of text) {
-                if (!jpIsCodePointJapanese(c.codePointAt(0))) {
+                if (!jp.isCodePointJapanese(c.codePointAt(0))) {
                     break;
                 }
                 newText += c;