@echogarden/text-segmentation 0.2.1 → 0.3.1

package/dist/WordSequence.js

@@ -1,16 +1,17 @@
 export class WordSequence {
     entries = [];
-    addWordEntry(containingText, startOffset, endOffset, isPunctuation) {
-        const wordText = containingText.substring(startOffset, endOffset);
+    addWord(text, textStartOffset, isPunctuation) {
+        const startOffset = textStartOffset;
+        const endOffset = startOffset + text.length;
         this.entries.push({
-            text: wordText,
+            text,
             startOffset,
             endOffset,
             isPunctuation
         });
     }
     getWordRange(startIndex, endIndex) {
-        return [...this.iterateWordRange(startIndex, endIndex)];
+        return this.getEntryRange(startIndex, endIndex).map(entry => entry.text);
     }
     *iterateWordRange(startIndex, endIndex) {
         for (let i = startIndex; i < endIndex; i++) {
@@ -39,7 +40,7 @@ export class WordSequence {
         slicedSequence.entries = this.entries.slice(startIndex, endIndex);
         return slicedSequence;
     }
-    get words() {
+    get wordArray() {
         return this.entries.map(entry => entry.text);
     }
     get firstWord() {
@@ -54,6 +55,12 @@ export class WordSequence {
     get lastEntry() {
         return this.entries[this.length - 1];
     }
+    get punctuationEntries() {
+        return this.entries.filter(entry => entry.isPunctuation === true);
+    }
+    get punctuationWords() {
+        return this.entries.filter(entry => entry.isPunctuation === true).map(entry => entry.text);
+    }
     get nonPunctuationEntries() {
         return this.entries.filter(entry => entry.isPunctuation === false);
     }
@@ -61,7 +68,7 @@ export class WordSequence {
         return this.entries.filter(entry => entry.isPunctuation === false).map(entry => entry.text);
     }
     get text() {
-        return this.words.join('');
+        return this.wordArray.join('');
     }
     get length() {
         return this.entries.length;

package/dist/WordSequence.js.map

@@ -1 +1 @@
-{"version":3,"file":"WordSequence.js","sourceRoot":"","sources":["../src/WordSequence.ts"],"names":[],"mappings":"AAAA,MAAM,OAAO,YAAY;IACxB,OAAO,GAAgB,EAAE,CAAA;IAEzB,YAAY,CAAC,cAAsB,EAAE,WAAmB,EAAE,SAAiB,EAAE,aAAsB;QAClG,MAAM,QAAQ,GAAG,cAAc,CAAC,SAAS,CAAC,WAAW,EAAE,SAAS,CAAC,CAAA;QAEjE,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC;YACjB,IAAI,EAAE,QAAQ;YACd,WAAW;YACX,SAAS;YACT,aAAa;SACb,CAAC,CAAA;IACH,CAAC;IAED,YAAY,CAAC,UAAkB,EAAE,QAAgB;QAChD,OAAO,CAAC,GAAG,IAAI,CAAC,gBAAgB,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAC,CAAA;IACxD,CAAC;IAED,CAAC,gBAAgB,CAAC,UAAkB,EAAE,QAAgB;QACrD,KAAK,IAAI,CAAC,GAAG,UAAU,EAAE,CAAC,GAAG,QAAQ,EAAE,CAAC,EAAE,EAAE,CAAC;YAC5C,MAAM,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,CAAA;QACxB,CAAC;IACF,CAAC;IAED,SAAS,CAAC,KAAa;QACtB,OAAO,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,IAAI,CAAA;IAChC,CAAC;IAED,aAAa,CAAC,UAAkB,EAAE,QAAgB;QACjD,OAAO,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAA;IAChD,CAAC;IAED,CAAC,iBAAiB,CAAC,UAAkB,EAAE,QAAgB;QACtD,KAAK,IAAI,CAAC,GAAG,UAAU,EAAE,CAAC,GAAG,QAAQ,EAAE,CAAC,EAAE,EAAE,CAAC;YAC5C,MAAM,IAAI,CAAC,UAAU,CAAC,CAAC,CAAC,CAAA;QACzB,CAAC;IACF,CAAC;IAED,UAAU,CAAC,KAAa;QACvB,OAAO,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,CAAA;IAC3B,CAAC;IAED,KAAK;QACJ,OAAO,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAA;IAClC,CAAC;IAED,KAAK,CAAC,UAAkB,EAAE,QAAgB;QACzC,MAAM,cAAc,GAAG,IAAI,YAAY,EAAE,CAAA;QAEzC,cAAc,CAAC,OAAO,GAAG,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAA;QAEjE,OAAO,cAAc,CAAA;IACtB,CAAC;IAED,IAAI,KAAK;QACR,OAAO,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC7C,CAAC;IAED,IAAI,SAAS;QACZ,OAAO,IAAI,CAAC,UAAU,EAAE,IAAI,CAAA;IAC7B,CAAC;IAED,IAAI,QAAQ;QACX,OAAO,IAAI,CAAC,SAAS,EAAE,IAAI,CAAA;IAC5B,CAAC;IAED,IAAI,UAAU;QACb,OAAO,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAA;IACvB,CAAC;IAED,IAAI,SAAS;QACZ,OAAO,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAA;IACrC,CAAC;IAED,IAAI,qBAAqB;QACxB,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,aAAa,KAAK,KAAK,CAAC,CAAA;IACnE,CAAC;IAED,IAAI,mBAAmB;QACtB,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,aAAa,KAAK,KAAK,CAAC,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC5F,CAAC;IAED,IAAI,IAAI;QACP,OAAO,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;IAC3B,CAAC;IAED,IAAI,MAAM;QACT,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAA;IAC3B,CAAC;CACD"}
+{"version":3,"file":"WordSequence.js","sourceRoot":"","sources":["../src/WordSequence.ts"],"names":[],"mappings":"AAAA,MAAM,OAAO,YAAY;IACxB,OAAO,GAAgB,EAAE,CAAA;IAEzB,OAAO,CAAC,IAAY,EAAE,eAAuB,EAAE,aAAsB;QACpE,MAAM,WAAW,GAAG,eAAe,CAAA;QACnC,MAAM,SAAS,GAAG,WAAW,GAAG,IAAI,CAAC,MAAM,CAAA;QAE3C,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC;YACjB,IAAI;YACJ,WAAW;YACX,SAAS;YACT,aAAa;SACb,CAAC,CAAA;IACH,CAAC;IAED,YAAY,CAAC,UAAkB,EAAE,QAAgB;QAChD,OAAO,IAAI,CAAC,aAAa,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IACzE,CAAC;IAED,CAAC,gBAAgB,CAAC,UAAkB,EAAE,QAAgB;QACrD,KAAK,IAAI,CAAC,GAAG,UAAU,EAAE,CAAC,GAAG,QAAQ,EAAE,CAAC,EAAE,EAAE,CAAC;YAC5C,MAAM,IAAI,CAAC,SAAS,CAAC,CAAC,CAAC,CAAA;QACxB,CAAC;IACF,CAAC;IAED,SAAS,CAAC,KAAa;QACtB,OAAO,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,IAAI,CAAA;IAChC,CAAC;IAED,aAAa,CAAC,UAAkB,EAAE,QAAgB;QACjD,OAAO,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAA;IAChD,CAAC;IAED,CAAC,iBAAiB,CAAC,UAAkB,EAAE,QAAgB;QACtD,KAAK,IAAI,CAAC,GAAG,UAAU,EAAE,CAAC,GAAG,QAAQ,EAAE,CAAC,EAAE,EAAE,CAAC;YAC5C,MAAM,IAAI,CAAC,UAAU,CAAC,CAAC,CAAC,CAAA;QACzB,CAAC;IACF,CAAC;IAED,UAAU,CAAC,KAAa;QACvB,OAAO,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,CAAA;IAC3B,CAAC;IAED,KAAK;QACJ,OAAO,IAAI,CAAC,KAAK,CAAC,CAAC,EAAE,IAAI,CAAC,MAAM,CAAC,CAAA;IAClC,CAAC;IAED,KAAK,CAAC,UAAkB,EAAE,QAAgB;QACzC,MAAM,cAAc,GAAG,IAAI,YAAY,EAAE,CAAA;QAEzC,cAAc,CAAC,OAAO,GAAG,IAAI,CAAC,OAAO,CAAC,KAAK,CAAC,UAAU,EAAE,QAAQ,CAAC,CAAA;QAEjE,OAAO,cAAc,CAAA;IACtB,CAAC;IAED,IAAI,SAAS;QACZ,OAAO,IAAI,CAAC,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC7C,CAAC;IAED,IAAI,SAAS;QACZ,OAAO,IAAI,CAAC,UAAU,EAAE,IAAI,CAAA;IAC7B,CAAC;IAED,IAAI,QAAQ;QACX,OAAO,IAAI,CAAC,SAAS,EAAE,IAAI,CAAA;IAC5B,CAAC;IAED,IAAI,UAAU;QACb,OAAO,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAA;IACvB,CAAC;IAED,IAAI,SAAS;QACZ,OAAO,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAA;IACrC,CAAC;IAED,IAAI,kBAAkB;QACrB,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,aAAa,KAAK,IAAI,CAAC,CAAA;IAClE,CAAC;IAED,IAAI,gBAAgB;QACnB,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,aAAa,KAAK,IAAI,CAAC,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC3F,CAAC;IAED,IAAI,qBAAqB;QACxB,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,aAAa,KAAK,KAAK,CAAC,CAAA;IACnE,CAAC;IAED,IAAI,mBAAmB;QACtB,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,aAAa,KAAK,KAAK,CAAC,CAAC,GAAG,CAAC,KAAK,CAAC,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAA;IAC5F,CAAC;IAED,IAAI,IAAI;QACP,OAAO,IAAI,CAAC,SAAS,CAAC,IAAI,CAAC,EAAE,CAAC,CAAA;IAC/B,CAAC;IAED,IAAI,MAAM;QACT,OAAO,IAAI,CAAC,OAAO,CAAC,MAAM,CAAA;IAC3B,CAAC;CACD"}

package/dist/utilities/Utilities.d.ts

@@ -4,3 +4,4 @@ export declare function listAllCharsMatching(charPattern: Pattern): string[];
 export declare function extractSuppressions(entries: {
     suppression: string;
 }[]): string[];
+export declare function getShortLanguageCode(langCode: string): string;

package/dist/utilities/Utilities.js

@@ -21,4 +21,11 @@ export function extractSuppressions(entries) {
     }
     return suppressions;
 }
+export function getShortLanguageCode(langCode) {
+    const dashIndex = langCode.indexOf('-');
+    if (dashIndex == -1) {
+        return langCode;
+    }
+    return langCode.substring(0, dashIndex).toLowerCase();
+}
 //# sourceMappingURL=Utilities.js.map

package/dist/utilities/Utilities.js.map

@@ -1 +1 @@
-{"version":3,"file":"Utilities.js","sourceRoot":"","sources":["../../src/utilities/Utilities.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAW,MAAM,iBAAiB,CAAA;AAEtD,MAAM,UAAU,aAAa,CAAC,GAAW,EAAE,MAAM,GAAG,CAAC;IACpD,MAAM,UAAU,GAAG,EAAE,IAAI,MAAM,CAAA;IAE/B,OAAO,IAAI,CAAC,KAAK,CAAC,GAAG,GAAG,UAAU,CAAC,GAAG,UAAU,CAAA;AACjD,CAAC;AAED,MAAM,UAAU,oBAAoB,CAAC,WAAoB;IACxD,MAAM,MAAM,GAAG,WAAW,CAAC,WAAW,CAAC,CAAA;IAEvC,MAAM,aAAa,GAAa,EAAE,CAAA;IAElC,KAAK,IAAI,SAAS,GAAG,CAAC,EAAE,SAAS,GAAG,OAAO,EAAE,SAAS,EAAE,EAAE,CAAC;QAC1D,MAAM,IAAI,GAAG,MAAM,CAAC,aAAa,CAAC,SAAS,CAAC,CAAA;QAE5C,IAAI,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;YACvB,aAAa,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACzB,CAAC;IACF,CAAC;IAED,OAAO,aAAa,CAAA;AACrB,CAAC;AAED,MAAM,UAAU,mBAAmB,CAAC,OAAiC;IACpE,MAAM,YAAY,GAAa,EAAE,CAAA;IAEjC,KAAK,MAAM,KAAK,IAAI,OAAO,EAAE,CAAC;QAC7B,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,WAAW,CAAC,CAAA;IACrC,CAAC;IAED,OAAO,YAAY,CAAA;AACpB,CAAC"}
+{"version":3,"file":"Utilities.js","sourceRoot":"","sources":["../../src/utilities/Utilities.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,WAAW,EAAW,MAAM,iBAAiB,CAAA;AAEtD,MAAM,UAAU,aAAa,CAAC,GAAW,EAAE,MAAM,GAAG,CAAC;IACpD,MAAM,UAAU,GAAG,EAAE,IAAI,MAAM,CAAA;IAE/B,OAAO,IAAI,CAAC,KAAK,CAAC,GAAG,GAAG,UAAU,CAAC,GAAG,UAAU,CAAA;AACjD,CAAC;AAED,MAAM,UAAU,oBAAoB,CAAC,WAAoB;IACxD,MAAM,MAAM,GAAG,WAAW,CAAC,WAAW,CAAC,CAAA;IAEvC,MAAM,aAAa,GAAa,EAAE,CAAA;IAElC,KAAK,IAAI,SAAS,GAAG,CAAC,EAAE,SAAS,GAAG,OAAO,EAAE,SAAS,EAAE,EAAE,CAAC;QAC1D,MAAM,IAAI,GAAG,MAAM,CAAC,aAAa,CAAC,SAAS,CAAC,CAAA;QAE5C,IAAI,MAAM,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;YACvB,aAAa,CAAC,IAAI,CAAC,IAAI,CAAC,CAAA;QACzB,CAAC;IACF,CAAC;IAED,OAAO,aAAa,CAAA;AACrB,CAAC;AAED,MAAM,UAAU,mBAAmB,CAAC,OAAiC;IACpE,MAAM,YAAY,GAAa,EAAE,CAAA;IAEjC,KAAK,MAAM,KAAK,IAAI,OAAO,EAAE,CAAC;QAC7B,YAAY,CAAC,IAAI,CAAC,KAAK,CAAC,WAAW,CAAC,CAAA;IACrC,CAAC;IAED,OAAO,YAAY,CAAA;AACpB,CAAC;AAED,MAAM,UAAU,oBAAoB,CAAC,QAAgB;IACpD,MAAM,SAAS,GAAG,QAAQ,CAAC,OAAO,CAAC,GAAG,CAAC,CAAA;IAEvC,IAAI,SAAS,IAAI,CAAC,CAAC,EAAE,CAAC;QACrB,OAAO,QAAQ,CAAA;IAChB,CAAC;IAED,OAAO,QAAQ,CAAC,SAAS,CAAC,CAAC,EAAE,SAAS,CAAC,CAAC,WAAW,EAAE,CAAA;AACtD,CAAC"}

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@echogarden/text-segmentation",
-	"version": "0.2.1",
+	"version": "0.3.1",
 	"description": "A library for multilingual word, phrase and sentence segmentation.",
 	"author": "Rotem Dan",
 	"license": "MIT",
@@ -39,10 +39,10 @@
 		"regexp-composer": "../../regexp-composer"
 	},
 	"dependencies": {
-		"regexp-composer": "^0.2.2"
+		"regexp-composer": "^0.3.0"
 	},
 	"peerDependencies": {
-		"@echogarden/icu-segmentation-wasm": "^0.1.0"
+		"@echogarden/icu-segmentation-wasm": "^0.2.1"
 	},
 	"peerDependenciesMeta": {
 		"@echogarden/icu-segmentation-wasm": {
@@ -50,6 +50,6 @@
 		}
 	},
 	"devDependencies": {
-		"@types/node": "^22.15.16"
+		"@types/node": "^22.15.18"
 	}
 }

package/src/EastAsianCharacterPatterns.ts

@@ -1,12 +1,12 @@
 import { anyOf, buildRegExp, codepointRange } from 'regexp-composer'
 
-export const chineseCharacterRanges = anyOf(
+const chineseCharacterRanges = anyOf(
 	codepointRange('4E00', '9FFF'), // Main
 	codepointRange('3400', '4DBF'), // CJK Unified Ideographs Extension A
 	codepointRange('20000', '2A6DF') // CJK Unified Ideographs Extension B
 )
 
-export const japaneseHiraganaCharacterRanges = anyOf(
+const japaneseHiraganaCharacterRanges = anyOf(
 	codepointRange('3040', '309F'), // Hiragana
 	codepointRange('1AFF0', '1AFFF'), // Kana Extended-B
 	codepointRange('1B000', '1B0FF'), // Kana Supplement
@@ -14,7 +14,7 @@ export const japaneseHiraganaCharacterRanges = anyOf(
 	codepointRange('1B130', '1B16F'), // Small Kana Extension
 )
 
-export const japaneseKatakanaCharacterRanges = anyOf(
+const japaneseKatakanaCharacterRanges = anyOf(
 	codepointRange('30A0', '30FF'), // Katakana
 	codepointRange('31F0', '31FF'), // Katakana Phonetic Extensions
 	codepointRange('3200', '32FF'), // Enclosed CJK Letters and Months
@@ -25,16 +25,16 @@ export const japaneseKatakanaCharacterRanges = anyOf(
 	codepointRange('1B130', '1B16F'), // Small Kana Extension
 )
 
-export const thaiLetterRanges = anyOf(
+const thaiLetterRanges = anyOf(
 	codepointRange('0E00', '0E7F')
 )
 
-export const khmerLetterRanges = anyOf(
+const khmerLetterRanges = anyOf(
 	codepointRange('1780', '17FF'), // Letters
 	codepointRange('19E0', '19FF'), // Symbols
 )
 
-export const eastAsianCharRanges = anyOf(
+const eastAsianCharRanges = anyOf(
 	chineseCharacterRanges,
 	japaneseHiraganaCharacterRanges,
 	japaneseKatakanaCharacterRanges,

package/src/Patterns.ts

@@ -1,5 +1,8 @@
-import { anyOf, buildRegExp, charRange, inputEnd, inputStart, matches, oneOrMore, possibly, repeated, tab, unicodeProperty, whitespace } from 'regexp-composer'
+import { anyOf, buildRegExp, charRange, inputEnd, inputStart, matches, oneOrMore, possibly, repeated, tab, unicodeProperty, whitespace, zeroOrMore } from 'regexp-composer'
 
+////////////////////////////////////////////////////////////////////////////////////////////////
+// Pattern builder methods
+////////////////////////////////////////////////////////////////////////////////////////////////
 export function buildWordOrNumberPattern(suppressions: string[]) {
 	return anyOf(
 		buildSuppressionPattern(suppressions),
@@ -18,95 +21,142 @@ export function buildSuppressionPattern(suppressions: string[]) {
 }
 
 ////////////////////////////////////////////////////////////////////////////////////////////////
-// Numeric patterns
+// Single character patterns
 ////////////////////////////////////////////////////////////////////////////////////////////////
-export const punctuationPattern = unicodeProperty('Punctuation')
-export const digitPattern = unicodeProperty('Decimal_Number')
-export const arabicNumeralPattern = charRange('0', '9')
+const letterPattern = unicodeProperty('Letter')
+const markPattern = unicodeProperty('Mark')
+
+const letterOrMarkPattern = anyOf(
+	letterPattern,
+	markPattern,
+)
+
+const apostrophPattern = anyOf(`'`, `’`, `‘`)
+
+const punctuationPattern = unicodeProperty('Punctuation')
+const digitPattern = unicodeProperty('Decimal_Number')
+const arabicNumeralPattern = charRange('0', '9')
+
+const emojiPattern = unicodeProperty('Emoji')
 
-export const numericSeparatorPattern =
+const percentageCharacters = ['%']
+const currencyCharacters = ['$', '¥', '€', '£', '₩', '₭', '₽', '₫', '฿', '¢', '₮', '؋', '₦', '₱', '₴', '₪']
+
+const percentageOrCurrencyCharacterPattern = anyOf(...percentageCharacters, ...currencyCharacters)
+
+////////////////////////////////////////////////////////////////////////////////////////////////
+// Numeric patterns
+////////////////////////////////////////////////////////////////////////////////////////////////
+const numericSeparatorPattern =
 	matches(
 		anyOf('.', ',', '٬', '_'), {
 		ifPrecededBy: arabicNumeralPattern,
 		ifFollowedBy: arabicNumeralPattern
 	})
 
-export const dateTimeSeparatorPattern = matches(
-	anyOf('/', '-', ':'), {
-	ifPrecededBy: arabicNumeralPattern,
-	ifFollowedBy: arabicNumeralPattern
-})
+const dimensionsPattern = matches([
+	oneOrMore(arabicNumeralPattern),
 
-export const dateTimePattern =
-	oneOrMore(anyOf(
-		arabicNumeralPattern,
-		dateTimeSeparatorPattern,
-	))
+	oneOrMore([
+		'x',
+		oneOrMore(arabicNumeralPattern),
+	]),
+], {
+	ifPrecededBy: anyOf(whitespace, punctuationPattern, inputStart),
+	ifFollowedBy: anyOf(whitespace, punctuationPattern, inputEnd)
+})
 
-export const spacedThousandsSeparatorPattern =
+const spacedThousandsSeparatorPattern =
 	matches(
 		' ', {
 		ifPrecededBy: arabicNumeralPattern,
-		ifFollowedBy: repeated(2, arabicNumeralPattern)
+		ifFollowedBy: [
+			repeated(3, arabicNumeralPattern),
+			anyOf(whitespace, punctuationPattern, inputEnd)
+		]
 	})
 
-export const numericSignPattern =
+const numericSignPattern =
 	matches(
 		anyOf('-', '+'), {
-		ifPrecededBy: anyOf(whitespace, punctuationPattern),
+		ifPrecededBy: anyOf(whitespace, punctuationPattern, inputStart),
 		ifFollowedBy: arabicNumeralPattern,
 	})
 
-export const numberPattern = [
-	oneOrMore(anyOf(
+const numberPattern = [
+	possibly(numericSignPattern),
+	digitPattern,
+
+	zeroOrMore(anyOf(
 		digitPattern,
 		numericSeparatorPattern,
 		spacedThousandsSeparatorPattern,
-		numericSignPattern,
-	)),
+	))
+]
+
+const exponentPattern = [
+	anyOf('e', 'E'),
+	possibly(anyOf('+', '-')),
+	oneOrMore(arabicNumeralPattern),
 ]
 
-const percentageChars = ['%']
-const currencySpecialChars = ['$', '¥', '€', '£', '¥', '₩', '₭', '₽', '₫', '฿', '¢', '₮', '؋', '₦', '₱', '₴', '₪']
+const numberPossiblyFollowedByExponentOrLettersPattern = [
+	numberPattern,
+
+	possibly(anyOf(
+		exponentPattern,
 
-const percentageOrCurrencyPattern = anyOf(...percentageChars, ...currencySpecialChars)
+		matches(
+			zeroOrMore(unicodeProperty('Letter')), {
+			ifNotFollowedBy: digitPattern,
+		}),
+	))
+]
 
-export const prefixPercentageOrCurrencyPattern =
+const precedingPercentageOrCurrencyPattern =
 	matches([
-		percentageOrCurrencyPattern,
+		percentageOrCurrencyCharacterPattern,
 		numberPattern,
 	], {
 		ifNotPrecededBy: digitPattern,
-		ifFollowedBy: anyOf(whitespace, punctuationPattern),
+		ifFollowedBy: anyOf(whitespace, punctuationPattern, inputEnd),
 	})
 
-export const suffixPercentagePattern =
+const followingPercentageOrCurrencyPattern =
 	matches([
 		numberPattern,
-		percentageOrCurrencyPattern,
+		percentageOrCurrencyCharacterPattern,
 	], {
-		ifPrecededBy: anyOf(whitespace, punctuationPattern),
-		ifNotFollowedBy: digitPattern,
+		ifPrecededBy: anyOf(whitespace, punctuationPattern, inputStart),
+		ifNotFollowedBy: anyOf(digitPattern),
 	})
 
-export const percentagePattern = anyOf(
-	prefixPercentageOrCurrencyPattern,
-	suffixPercentagePattern
+const percentageOrCurrencyPattern = anyOf(
+	precedingPercentageOrCurrencyPattern,
+	followingPercentageOrCurrencyPattern
 )
 
 ////////////////////////////////////////////////////////////////////////////////////////////////
-// Letter patterns
+// Time patterns
 ////////////////////////////////////////////////////////////////////////////////////////////////
-export const letterPattern = unicodeProperty('Letter')
-export const markPattern = unicodeProperty('Mark')
-export const apostrophPattern = anyOf(`'`, `’`, `‘`)
-
-export const letterOrMarkPattern = anyOf(
-	letterPattern,
-	markPattern,
-)
+const timePattern = matches([
+	anyOf(arabicNumeralPattern, [charRange('0', '1'), arabicNumeralPattern], ['2', charRange('0', '3')]),
+
+	repeated([1, 2], [
+		':',
+		anyOf(arabicNumeralPattern, [charRange('0', '5'), arabicNumeralPattern]),
+	])
+], {
+	ifPrecededBy: anyOf(whitespace, punctuationPattern, inputStart),
+	ifNotPrecededBy: ':',
+	ifFollowedBy: anyOf(whitespace, punctuationPattern, inputEnd),
+	ifNotFollowedBy: ':',
+})
 
-export const dottedAbbreviationSequencePattern =
+////////////////////////////////////////////////////////////////////////////////////////////////
+// Letter patterns
+////////////////////////////////////////////////////////////////////////////////////////////////
+const dottedAbbreviationSequencePattern =
 	matches(
 		anyOf(
 			[
@@ -135,23 +185,14 @@ export const dottedAbbreviationSequencePattern =
 		ifNotFollowedBy: anyOf(letterOrMarkPattern, digitPattern)
 	})
 
-
-export const wordCharacterPattern =
+const wordCharacterPattern =
 	anyOf(
 		letterPattern,
 		markPattern,
 		digitPattern,
 	)
 
-export const wordSeparatorPattern =
-	matches(
-		anyOf('-', '_', '·', '‧', '&'), {
-
-		ifPrecededBy: letterOrMarkPattern,
-		ifFollowedBy: letterOrMarkPattern
-	})
-
-export const wordInnerApostrophPattern =
+const wordInnerApostrophPattern =
 	matches(
 		apostrophPattern, {
 
@@ -159,67 +200,119 @@ export const wordInnerApostrophPattern =
 		ifFollowedBy: letterOrMarkPattern,
 	})
 
-export const wordStartApostrophPattern =
+const wordStartApostrophPattern =
 	matches(
 		apostrophPattern, {
 
 		ifPrecededBy: whitespace,
-		ifFollowedBy: [letterOrMarkPattern, letterOrMarkPattern, whitespace]
+		ifFollowedBy: [letterOrMarkPattern, letterOrMarkPattern, anyOf(whitespace, inputEnd)]
 	})
 
-export const basicWordPattern =
+const basicWordPattern =
 	oneOrMore(
 		anyOf(
 			wordCharacterPattern,
-			wordSeparatorPattern,
 			wordInnerApostrophPattern,
 		),
 	)
 
-export const wordSegmentPattern = anyOf(
+const hyphenatedWordPattern = [
+	basicWordPattern,
+
+	oneOrMore([
+		'-',
+		basicWordPattern,
+	]),
+]
+
+const dotSeparatedWordPattern = matches([
+	basicWordPattern,
+
+	oneOrMore([
+		'.',
+		basicWordPattern,
+	]),
+], {
+	//ifPrecededBy: anyOf(whitespace, inputStart),
+	//ifFollowedBy: anyOf(whitespace, inputEnd)
+})
+
+const interpunctSeparatedWordPattern = [
+	basicWordPattern,
+
+	oneOrMore([
+		anyOf('·', '‧'),
+		basicWordPattern,
+	]),
+]
+
+const underscoreSeparatedWordPattern = [
+	basicWordPattern,
+
+	oneOrMore([
+		'_',
+		basicWordPattern,
+	]),
+]
+
+const wordSegmentPattern = anyOf(
+	timePattern,
+	dimensionsPattern,
+	percentageOrCurrencyPattern,
+	numberPossiblyFollowedByExponentOrLettersPattern,
+
+	emojiPattern,
+	interpunctSeparatedWordPattern,
+	dotSeparatedWordPattern,
 	dottedAbbreviationSequencePattern,
-	dateTimeSeparatorPattern,
-	percentagePattern,
-	numberPattern,
+	underscoreSeparatedWordPattern,
 	basicWordPattern,
 )
 
 ////////////////////////////////////////////////////////////////////////////////////////////////
-// Prebuilt regular expressions
+// Phrase separation patterns
 ////////////////////////////////////////////////////////////////////////////////////////////////
-export const phraseSeparators = [',', '、', '，', '،', ';', '；', ':', '：', '—']
+const phraseSeparatorCharacters = [',', '、', '，', '،', ';', '；', ':', '：', '—']
 
-export const phraseSeparatorRegExp = buildRegExp([
+const phraseSeparatorPattern = [
 	inputStart,
-	anyOf(...phraseSeparators),
+	anyOf(...phraseSeparatorCharacters),
 	inputEnd
-])
-
-export const sentenceSeparators = ['.', '。', '?', '？', '!', '！', '\n']
+]
 
-export const sentenceSeparatorRegExp = buildRegExp([
+const phraseSeparatorTrailingPunctuationPattern = [
 	inputStart,
-	anyOf(...sentenceSeparators),
+	anyOf(...phraseSeparatorCharacters, ' ', tab),
 	inputEnd
-])
+]
 
-export const sentenceSeparatorTrailingPunctuationRegExp = buildRegExp([
-	inputStart,
-	anyOf('"', '”', '’', ')', ']', '}', '»', ...sentenceSeparators, ...phraseSeparators, oneOrMore(whitespace)),
-	inputEnd
-])
+////////////////////////////////////////////////////////////////////////////////////////////////
+// Sentence separation patterns
+////////////////////////////////////////////////////////////////////////////////////////////////
+const sentenceSeparatorCharacters = ['.', '。', '?', '？', '!', '！', '\n']
 
-export const phraseSeparatorTrailingPunctuationRegExp = buildRegExp([
+const sentenceSeparatorPattern = [
 	inputStart,
-	anyOf(...phraseSeparators, ' ', tab),
+	anyOf(...sentenceSeparatorCharacters),
 	inputEnd
-])
+]
 
-export const oneOrMoreSpacesRegExp = buildRegExp([
+const sentenceSeparatorTrailingPunctuationPattern = [
 	inputStart,
-	oneOrMore(' '),
+	anyOf('"', '”', '’', ')', ']', '}', '»', ...sentenceSeparatorCharacters, ...phraseSeparatorCharacters, oneOrMore(whitespace)),
 	inputEnd
-])
+]
 
+////////////////////////////////////////////////////////////////////////////////////////////////
+// Prebuilt regular expressions
+////////////////////////////////////////////////////////////////////////////////////////////////
 export const wordCharacterRegExp = buildRegExp(wordCharacterPattern)
 export const whitespacePatternRegExp = buildRegExp(whitespace)
+
+export const letterPatternGlobalRegExp = buildRegExp(letterPattern, { global: true })
+
+export const phraseSeparatorRegExp = buildRegExp(phraseSeparatorPattern)
+export const phraseSeparatorTrailingPunctuationRegExp = buildRegExp(phraseSeparatorTrailingPunctuationPattern)
+
+export const sentenceSeparatorRegExp = buildRegExp(sentenceSeparatorPattern)
+export const sentenceSeparatorTrailingPunctuationRegExp = buildRegExp(sentenceSeparatorTrailingPunctuationPattern)