Fixed 咖啡 transliteration for Singapore dialect, fixed error with brow…

…ser import of the library, minor edits to words and vars datasets

package.json

@@ -1,6 +1,6 @@
 {
   "name": "taibun",
-  "version": "1.1.1",
+  "version": "1.1.2",
   "description": "Taiwanese Hokkien Transliterator and Tokeniser",
   "main": "taibun/index.js",
   "types": "taibun/index.d.ts",

taibun/data/vars.json

@@ -42,5 +42,6 @@
     "譁": "嘩",
     "廕": "蔭",
     "鍾": "鐘",
-    "歎": "嘆"
+    "歎": "嘆",
+    "儂": "人"
 }

taibun/data/words.json

@@ -20810,7 +20810,7 @@
     "群英": "kûn-ing",
     "裙裾": "kûn-ki",
     "拳擊": "kûn-kik",
-    "群墘儂": "kûn-kînn-lâng",
+    "群墘人": "kûn-kînn-lâng",
     "拳路": "kûn-lōo",
     "群山": "kûn-suann",
     "拳賽": "kûn-sài",

taibun/index.js

@@ -1,4 +1,4 @@
-let wordDict, tradDict, simpDict, varsDict;
+let wordDict, tradDict, simpDict, varsDict, pronsDict;
 
 if (typeof window === 'undefined') {
 	// Node.js
@@ -125,7 +125,7 @@ class Converter {
 
 	constructor({ system = 'Tailo', dialect = 'south', format = 'mark', delimiter = Converter.defaultDelimiter, sandhi = Converter.defaultSandhi, punctuation = 'format', convertNonCjk = false } = {}) {
 		this.system = system.toLowerCase();
-		this.dialect = dialect;
+		this.dialect = dialect.toLowerCase();
 		this.format = format;
 		this.delimiter = delimiter !== Converter.defaultDelimiter ? delimiter : this.setDefaultDelimiter();
 		this.sandhi = sandhi !== Converter.defaultSandhi ? sandhi : this.setDefaultSandhi();
@@ -175,12 +175,18 @@ class Converter {
 				if (!value || dialect === 'south') return value;
 				const parts = value.toLowerCase().split(/(--|-)/).filter(s => s);
 				const variations = Object.fromEntries(Array.from(property).map(char => [char, Object.fromEntries((pronsDictProxy[char] || []).map(v => v.split('/').length > 1 ? v.split('/') : [v, v]))]));
-				if (property in Converter.singaporeWords && dialect === 'singapore') {
-					Object.keys(Converter.singaporeWords[property]).forEach(char => {
-						if (char in variations) {
-							Object.assign(variations[char], Converter.singaporeWords[property][char]);
+				if (dialect === 'singapore') {
+					const substrings = new Set(
+						[...property].flatMap((_, i) => [...property.slice(i)].map((_, j) => property.slice(i, i + j + 1)))
+					);
+					substrings.forEach(substring => {
+						if (substring in Converter.singaporeWords) {
+							Object.entries(Converter.singaporeWords[substring]).forEach(([char, mappings]) => {
+								if (char in variations) Object.assign(variations[char], mappings);
+							});
 						}
 					});
+					value = value.split('').map(char => variations[char]?.[char] || char).join('');
 				}
 				let newParts = [];
 				let charIndex = 0;
@@ -404,10 +410,7 @@ class Converter {
 
 	// Helper to convert Taiwanese pronunciation to Singaporean
 	convertVariant(input) {
-		if (this.dialect === 'singapore') {
-			return input.replace('ing', 'eng');
-		}
-		return input;
+		return this.dialect === 'singapore' ? input.replace('ing', 'eng') : input;
 	}
 
 

tests/singapore.test.js

@@ -0,0 +1,57 @@
+const { Converter } = require('taibun');
+
+test('tones', () => {
+	const c = new Converter({ system: 'IPA', dialect: "Singapore", punctuation: 'none' });
+	const tones = { '衫': 'sã⁴⁴', '短': 'te⁴²', '褲': 'kʰɔ²¹', '闊': 'kʰuaʔ³²', '人': 'laŋ²⁴', '鼻': 'pʰĩ²²', '直': 'tit̚⁴' };
+	for (let [tone, expected] of Object.entries(tones)) {
+		expect(c.get(tone)).toBe(expected);
+	}
+});
+
+test('o conversion', () => {
+	const c = new Converter({ system: 'IPA', dialect: "Singapore", punctuation: 'none' });
+	const oWords = { '高': 'ko⁴⁴', '唔': 'ɔ̃⁴⁴', '烏': 'ɔ⁴⁴', '王': 'ɔŋ²⁴' };
+	for (let [word, expected] of Object.entries(oWords)) {
+		expect(c.get(word)).toBe(expected);
+	}
+});
+
+describe('eng conversion', () => {
+	const hanjiData = ['用', '冰', '兵', '幸啊', '無閒', '無政府'];
+	const testData = [
+		[['ēng', 'peng', 'peng', 'hēng--ah', 'bô-êng', 'bô tsèng-hú'], "Tailo"],
+		[['ēng', 'peng', 'peng', 'hēng--ah', 'bô-êng', 'bô chèng-hú'], "POJ"],
+		[['ㆤㄥ˫', 'ㄅㆤㄥ', 'ㄅㆤㄥ', 'ㄏㆤㄥ˫ ㄚ', 'ㆠㄜˊ ㆤㄥˊ', 'ㆠㄜˊ ㄗㆤㄥ˪ ㄏㄨˋ'], "Zhuyin"],
+		[['eng7', 'peng1', 'peng1', 'heng7 ah0', 'bo5 eng5', 'bo5 ceng3 hu2'], "TLPA"],
+		[['êng', 'bēng', 'bēng', 'hêng ah', 'bbóéng', 'bbó zènghǔ'], "Pingyim"],
+		[['ēng', 'beng', 'beng', 'hēng--åh', 'bhôr-ĕng', 'bhôr zèng-hù'], "Tongiong"],
+		[['eŋ²²', 'peŋ⁴⁴', 'peŋ⁴⁴', 'heŋ²² a', 'bo²⁴ eŋ²⁴', 'bo²⁴ tseŋ²¹ hu⁴²'], "IPA"]
+	];
+	testData.forEach(([transl, system]) => {
+		const data = hanjiData.map((h, i) => [h, transl[i]]);
+		test(`testing: ${system}`, () => {
+			const c = new Converter({ system, dialect: "Singapore", punctuation: 'none' });
+			data.forEach(([hanji, expected]) => { expect(c.get(hanji)).toBe(expected); });
+		});
+	});
+});
+
+test('sandhi', () => {
+	const sandhis = ['auto', 'none', 'excLast', 'inclLast'];
+	const expectedResults = ['Tài-uân', 'Tâi-uân', 'Tài-uân', 'Tài-uàn'];
+
+	sandhis.forEach((sandhi, index) => {
+		const c = new Converter({ dialect: "Singapore", punctuation: "none", sandhi });
+		expect(c.get('台灣')).toBe(expectedResults[index]);
+	});
+});
+
+test('kopi', () => {
+	const c = new Converter({ dialect: "Singapore", punctuation: 'none' });
+	const kos = ['咖啡', '烏咖啡', '咖啡杯', '咖哩', '咖咖仔'];
+	const expectedResults = ['ko-pi', 'oo-ko-pi', 'ko-pi-pue', 'ka-lí', 'ka-ka-á'];
+
+	kos.forEach((ko, index) => {
+		expect(c.get(ko)).toBe(expectedResults[index]);
+	});
+});