bitaboom 2.1.0 → 2.2.0

package/README.md

@@ -42,10 +42,22 @@ const plain = removeMarkdownFormatting('**Bold** _italic_ [link](https://example
 console.log(plain); // "Bold italic link"
 ```
 
+### High-performance Arabic preformatting
+
+If you need to normalize messy Arabic/OCR text at scale (spacing, punctuation, brackets, ellipses, references), use the single-pass preformatter:
+
+```typescript
+import { preformatArabicText } from 'bitaboom';
+
+preformatArabicText('بِسْمِ  اللَّهِ ( الرَّحْمَنِ ) 127 / 11 قَالَ ...');
+preformatArabicText(['صفحة 1 ...', 'صفحة 2 ...']); // batch mode
+```
+
 ## Feature highlights
 
 - **Arabic-first matching** – build diacritic-insensitive regular expressions, collapse tatweel, score Arabic content density, and replace Urdu glyphs.
 - **Rich typography normalisers** – more than 30 helpers to fix punctuation spacing, quotes, brackets, ellipses, smart quotes, uppercase detection, and whitespace quirks.
+- **Single-pass Arabic preformatter** – `preformatArabicText` consolidates the common formatting pipeline and is optimized for large datasets.
 - **Sanitisation pipelines** – strip references, URLs, part markers, markdown decorations, escaped spaces, or numbers in bilingual text.
 - **Parsing helpers** – validate JSON-ish blobs, split search queries by quotes, balance parentheses/quotes, and expand page range strings.
 - **Transliteration polish** – normalise common Arabic prefixes (`al-`, `wa-`, `bi-`), dedupe apostrophes, replace salutations with ﷺ, and extract initials from transliterated names.
@@ -151,6 +163,12 @@ All modules are exported from `src/index.ts`. Functions are grouped below by fea
 | `normalizeTransliteratedEnglish` | Combine prefix removal + diacritic stripping. |
 | `extractInitials` | Extract the first letters from up to two words (after normalisation). |
 
+### Preformatting pipeline (`src/preformat.ts`)
+
+| Function | Description |
+| --- | --- |
+| `preformatArabicText` | High-performance Arabic preformatting pipeline (single-pass, optimized for large datasets). Accepts a single string or an array of strings. |
+
 ## Build & development
 
 | Task | Command |

package/dist/index.d.ts

@@ -171,6 +171,15 @@ type MakeRegexOptions = {
  * rx.test('اَنا إلى الآفاق'); // true
  */
 declare const makeDiacriticInsensitiveRegex: (needle: string, opts?: MakeRegexOptions) => RegExp;
+/**
+ * Remove simple HTML/XML-like tags from a string.
+ *
+ * This is intentionally lightweight and does not attempt to parse HTML; it simply drops
+ * substrings that look like `<...>`.
+ *
+ * @param content Input string
+ * @returns String with tags removed
+ */
 declare const removeAllTags: (content: string) => string;
 //#endregion
 //#region src/constants.d.ts
@@ -181,6 +190,9 @@ declare const PATTERN_ENDS_WITH_PUNCTUATION: RegExp;
 /**
  * Adds line breaks after punctuation marks such as periods, exclamation points, and question marks.
  * Example: 'Text.' becomes 'Text.\n'.
+ *
+ * Note: For the full preformatting pipeline in one pass (significantly faster and more memory-friendly
+ * on very large inputs), use `preformatArabicText` from `src/preformat.ts`.
  * @param {string} text - The input text containing punctuation.
  * @returns {string} - The modified text with line breaks added after punctuation.
  */
@@ -494,6 +506,30 @@ declare const isBalanced: (str: string) => boolean;
  */
 declare const parsePageRanges: (pageInput: string) => number[];
 //#endregion
+//#region src/preformat.d.ts
+/**
+ * Hyperoptimized Arabic text preformatting entry point.
+ *
+ * The implementation lives in `src/preformat-core.ts` to keep the public surface
+ * area small while allowing internal benchmarking (buffer vs concat builders).
+ *
+ * @module preformat
+ */
+type PreformatArabicText = {
+  (text: string): string;
+  (texts: string[]): string[];
+};
+/**
+ * High-performance Arabic preformatting pipeline.
+ *
+ * Consolidates common formatting steps (spacing, punctuation normalization, reference formatting,
+ * bracket/quote cleanup, ellipsis condensation, newline normalization) into a single-pass formatter.
+ *
+ * @param text Input string or an array of strings
+ * @returns Preformatted string or array of strings (matching input shape)
+ */
+declare const preformatArabicText: PreformatArabicText;
+//#endregion
 //#region src/sanitization.d.ts
 /**
  * Removes various symbols, part references, and numerical markers from the text.
@@ -685,5 +721,5 @@ declare const normalizeTransliteratedEnglish: (text: string) => string;
  */
 declare const extractInitials: (fullName: string) => string;
 //#endregion
-export { MakeRegexOptions, PATTERN_ENDS_WITH_PUNCTUATION, addSpaceBeforeAndAfterPunctuation, addSpaceBetweenArabicTextAndNumbers, applySmartQuotes, arabicNumeralToNumber, cleanExtremeArabicUnderscores, cleanLiteralNewLines, cleanMultilines, cleanSpacesBeforePeriod, cleanSymbolsAndPartReferences, cleanTrailingPageNumbers, condenseAsterisks, condenseColons, condenseDashes, condenseEllipsis, condensePeriods, condenseUnderscores, convertUrduSymbolsToArabic, doubleToSingleBrackets, ensureSpaceBeforeBrackets, ensureSpaceBeforeQuotes, escapeRegex, extractInitials, findLastPunctuation, fixBracketTypos, fixCurlyBraces, fixMismatchedQuotationMarks, fixTrailingWow, formatStringBySentence, getArabicScore, hasWordInSingleLine, insertLineBreaksAfterPunctuation, isAllUppercase, isBalanced, isJsonStructureValid, isOnlyPunctuation, makeDiacriticInsensitive, makeDiacriticInsensitiveRegex, normalize, normalizeArabicPrefixesToAl, normalizeDoubleApostrophes, normalizeJsonSyntax, normalizeSlashInReferences, normalizeSpaces, normalizeTransliteratedEnglish, parsePageRanges, reduceMultilineBreaksToDouble, reduceMultilineBreaksToSingle, removeAllTags, removeArabicPrefixes, removeDeathYear, removeMarkdownFormatting, removeNonIndexSignatures, removeNumbersAndDashes, removeRedundantPunctuation, removeSingleDigitReferences, removeSingularCodes, removeSolitaryArabicLetters, removeSpaceInsideBrackets, removeUrls, replaceDoubleBracketsWithArrows, replaceEnglishPunctuationWithArabic, replaceLineBreaksWithSpaces, replaceSalutationsWithSymbol, splitByQuotes, stripAllDigits, stripBoldStyling, stripItalicsStyling, stripStyling, toTitleCase, trimSpaceInsideQuotes, truncate, truncateMiddle, unescapeSpaces };
+export { MakeRegexOptions, PATTERN_ENDS_WITH_PUNCTUATION, addSpaceBeforeAndAfterPunctuation, addSpaceBetweenArabicTextAndNumbers, applySmartQuotes, arabicNumeralToNumber, cleanExtremeArabicUnderscores, cleanLiteralNewLines, cleanMultilines, cleanSpacesBeforePeriod, cleanSymbolsAndPartReferences, cleanTrailingPageNumbers, condenseAsterisks, condenseColons, condenseDashes, condenseEllipsis, condensePeriods, condenseUnderscores, convertUrduSymbolsToArabic, doubleToSingleBrackets, ensureSpaceBeforeBrackets, ensureSpaceBeforeQuotes, escapeRegex, extractInitials, findLastPunctuation, fixBracketTypos, fixCurlyBraces, fixMismatchedQuotationMarks, fixTrailingWow, formatStringBySentence, getArabicScore, hasWordInSingleLine, insertLineBreaksAfterPunctuation, isAllUppercase, isBalanced, isJsonStructureValid, isOnlyPunctuation, makeDiacriticInsensitive, makeDiacriticInsensitiveRegex, normalize, normalizeArabicPrefixesToAl, normalizeDoubleApostrophes, normalizeJsonSyntax, normalizeSlashInReferences, normalizeSpaces, normalizeTransliteratedEnglish, parsePageRanges, preformatArabicText, reduceMultilineBreaksToDouble, reduceMultilineBreaksToSingle, removeAllTags, removeArabicPrefixes, removeDeathYear, removeMarkdownFormatting, removeNonIndexSignatures, removeNumbersAndDashes, removeRedundantPunctuation, removeSingleDigitReferences, removeSingularCodes, removeSolitaryArabicLetters, removeSpaceInsideBrackets, removeUrls, replaceDoubleBracketsWithArrows, replaceEnglishPunctuationWithArabic, replaceLineBreaksWithSpaces, replaceSalutationsWithSymbol, splitByQuotes, stripAllDigits, stripBoldStyling, stripItalicsStyling, stripStyling, toTitleCase, trimSpaceInsideQuotes, truncate, truncateMiddle, unescapeSpaces };
 //# sourceMappingURL=index.d.ts.map

package/dist/index.js

@@ -1,10 +1,12 @@
-const e=/[.!?؟؛]$/,t=e=>parseInt(e.replace(/[\u0660-\u0669]/g,e=>(e.charCodeAt(0)-1632).toString()),10),n=e=>e.replace(/(?<!\d ?ه|اه)ـ(?=\r?$)|^ـ(?!اهـ)/gm,``),r=e=>e.replace(/ھ/g,`ه`).replace(/ی/g,`ي`),i=e=>{if(!e)return 0;let t=/[\u0600-\u06FF\u0750-\u077F\u08A0-\u08FF\uFB50-\uFDFF\uFE70-\uFEFF]/g,n=/[0-9\u0660-\u0669\u06F0-\u06F9]/g,r=/[^\s0-9\u0660-\u0669\u06F0-\u06F9]/g,i=e.replace(n,``),a=i.match(t)||[],o=i.match(r)||[];return o.length===0?0:a.length/o.length},a=t=>{for(let n=t.length-1;n>=0;n--)if(e.test(t[n]))return n;return-1},o=e=>e.replace(/ و /g,` و`),s=e=>e.replace(/([\u0600-\u06FF]+)(\d+)/g,`$1 $2`),c=e=>e.replace(/(?<![0-9] ?)-|(?<=[\u0600-\u06FF])\s?\d\s?(?=[\u0600-\u06FF])/g,` `).replace(/(?<=[\u0600-\u06FF]\s)(\d+\s)+\d+(?=(\s[\u0600-\u06FF]|$))/g,` `),l=e=>e.replace(/[[({][\u0621-\u064A\u0660-\u0669][\])}]/g,``),u=e=>e.replace(/(^| )[\u0621-\u064A]( |$)/g,` `),d=e=>e.replace(/\?|؟\./g,`؟`).replace(/(;|؛)\s*(\1\s*)*/g,`؛`).replace(/,|-،/g,`،`),f=e=>e.replace(/[.*+?^${}()|[\]\\]/g,`\\$&`),p=(e,t={})=>{let{equivalences:n={alif:!0,taMarbutahHa:!0,alifMaqsurahYa:!0},allowTatweel:r=!0,ignoreDiacritics:i=!0,flexWhitespace:a=!0,flags:o=`u`}=t;if(e.length>5e3)throw Error(`makeDiacriticInsensitiveRegex: needle too long`);let s=e=>{switch(e){case`ا`:case`أ`:case`إ`:case`آ`:return n.alif?`[اأإآ]`:`ا`;case`ة`:case`ه`:return n.taMarbutahHa?`[هة]`:f(e);case`ى`:case`ي`:return n.alifMaqsurahYa?`[ىي]`:f(e);default:return f(e)}},c=`${i?`[\\u0610-\\u061A\\u064B-\\u065F\\u0670\\u06D6-\\u06ED]*`:``}${r?`\\u0640*`:``}`,l=``;for(let t of Array.from(e))/\s/.test(t)?l+=a?`\\s+`:`\\s*`:l+=`${s(t)}${c}`;return new RegExp(l,o)},m=e=>e.replace(/<[^>]*>/g,``),h=e=>e.replace(/([.?!؟])/g,`$1
+const e=/[.!?؟؛]$/,t=e=>parseInt(e.replace(/[\u0660-\u0669]/g,e=>(e.charCodeAt(0)-1632).toString()),10),n=e=>e.replace(/(?<!\d ?ه|اه)ـ(?=\r?$)|^ـ(?!اهـ)/gm,``),r=e=>e.replace(/ھ/g,`ه`).replace(/ی/g,`ي`),i=e=>{if(!e)return 0;let t=/[\u0600-\u06FF\u0750-\u077F\u08A0-\u08FF\uFB50-\uFDFF\uFE70-\uFEFF]/g,n=/[0-9\u0660-\u0669\u06F0-\u06F9]/g,r=/[^\s0-9\u0660-\u0669\u06F0-\u06F9]/g,i=e.replace(n,``),a=i.match(t)||[],o=i.match(r)||[];return o.length===0?0:a.length/o.length},a=t=>{for(let n=t.length-1;n>=0;n--)if(e.test(t[n]))return n;return-1},o=e=>e.replace(/ و /g,` و`),s=e=>e.replace(/([\u0600-\u06FF]+)(\d+)/g,`$1 $2`),c=e=>e.replace(/(?<![0-9] ?)-|(?<=[\u0600-\u06FF])\s?\d\s?(?=[\u0600-\u06FF])/g,` `).replace(/(?<=[\u0600-\u06FF]\s)(\d+\s)+\d+(?=(\s[\u0600-\u06FF]|$))/g,` `),l=e=>e.replace(/[[({][\u0621-\u064A\u0660-\u0669][\])}]/g,``),u=e=>e.replace(/(^| )[\u0621-\u064A]( |$)/g,` `),d=e=>e.replace(/\?|؟\./g,`؟`).replace(/(;|؛)\s*(\1\s*)*/g,`؛`).replace(/,|-،/g,`،`),f=e=>e.replace(/[.*+?^${}()|[\]\\]/g,`\\$&`),p=(e,t={})=>{let{equivalences:n={alif:!0,taMarbutahHa:!0,alifMaqsurahYa:!0},allowTatweel:r=!0,ignoreDiacritics:i=!0,flexWhitespace:a=!0,flags:o=`u`}=t;if(e.length>5e3)throw Error(`makeDiacriticInsensitiveRegex: needle too long`);let s=e=>{switch(e){case`ا`:case`أ`:case`إ`:case`آ`:return n.alif?`[اأإآ]`:`ا`;case`ة`:case`ه`:return n.taMarbutahHa?`[هة]`:f(e);case`ى`:case`ي`:return n.alifMaqsurahYa?`[ىي]`:f(e);default:return f(e)}},c=`${i?`[\\u0610-\\u061A\\u064B-\\u065F\\u0670\\u06D6-\\u06ED]*`:``}${r?`\\u0640*`:``}`,l=``;for(let t of Array.from(e))/\s/.test(t)?l+=a?`\\s+`:`\\s*`:l+=`${s(t)}${c}`;return new RegExp(l,o)},ee=e=>e.replace(/<[^>]*>/g,``),te=e=>e.replace(/([.?!؟])/g,`$1
 `).replace(/\n\s+/g,`
-`).trim(),ee=e=>e.replace(/( ?)([.!?,،؟;؛])((?![ '”“)"\]\n])|(?=\s{2,}))/g,`$1$2 `).replace(/\s([.!?,،؟;؛])\s*([ '”“)"\]\n])/g,`$1$2`).replace(/([^\s\w\d'”“)"\]]+)\s+([.!?,،؟;؛])|([.!?,،؟;؛])\s+$/g,`$1$2$3`).replace(/(?<=\D)( ?: ?)(?!(\d+:)|(:\d+))|(?<=\d) ?: ?(?=\D)|(?<=\D) ?: ?(?=\d)/g,`: `),te=e=>e.replace(/[“”]/g,`"`).replace(/"([^"]*)"/g,`“$1”`).replace(/^”/g,`“`),ne=e=>e.replace(/\\n|\r/g,`
-`),g=e=>e.replace(/^ +| +$/gm,``),_=e=>/^\s*\S+\s*$/gm.test(e),v=e=>/^[\u0020-\u002f\u003a-\u0040\u005b-\u0060\u007b-\u007e0-9٠-٩]+$/.test(e),y=e=>e.replace(/\s+([.؟!,،؛:?])/g,`$1`),b=e=>e.replace(/(\*\s*)+/g,`*`),x=e=>e.replace(/[.-]?:[.-]?/g,`:`),S=e=>e.replace(/-{2,}/g,`-`),C=e=>e.replace(/\.{2,}/g,`…`),w=e=>e.replace(/(\n\s*){3,}/g,`
+`).trim(),ne=e=>e.replace(/( ?)([.!?,،؟;؛])((?![ '”“)"\]\n])|(?=\s{2,}))/g,`$1$2 `).replace(/\s([.!?,،؟;؛])\s*([ '”“)"\]\n])/g,`$1$2`).replace(/([^\s\w\d'”“)"\]]+)\s+([.!?,،؟;؛])|([.!?,،؟;؛])\s+$/g,`$1$2$3`).replace(/(?<=\D)( ?: ?)(?!(\d+:)|(:\d+))|(?<=\d) ?: ?(?=\D)|(?<=\D) ?: ?(?=\d)/g,`: `),re=e=>e.replace(/[“”]/g,`"`).replace(/"([^"]*)"/g,`“$1”`).replace(/^”/g,`“`),ie=e=>e.replace(/\\n|\r/g,`
+`),ae=e=>e.replace(/^ +| +$/gm,``),oe=e=>/^\s*\S+\s*$/gm.test(e),se=e=>/^[\u0020-\u002f\u003a-\u0040\u005b-\u0060\u007b-\u007e0-9٠-٩]+$/.test(e),ce=e=>e.replace(/\s+([.؟!,،؛:?])/g,`$1`),le=e=>e.replace(/(\*\s*)+/g,`*`),ue=e=>e.replace(/[.-]?:[.-]?/g,`:`),de=e=>e.replace(/-{2,}/g,`-`),fe=e=>e.replace(/\.{2,}/g,`…`),pe=e=>e.replace(/(\n\s*){3,}/g,`
 
-`),T=e=>e.replace(/(\n\s*){2,}/g,`
-`),E=e=>e.replace(/\. +\./g,`.`),D=e=>e.replace(/ـ{2,}/g,`ـ`).replace(/_+/g,`_`),O=e=>e.replace(/(\(|\)){2,}|(\[|\]){2,}/g,`$1$2`),k=e=>e.replace(/(\S) *(\([^)]*\))/g,`$1 $2`),A=e=>e.replace(/(\S) *(«[^»]*»)/g,`$1 $2`),j=e=>e.replace(/\(«|\( \(/g,`«`).replace(/»\)|\) \)/g,`»`).replace(/\)([0-9\u0660-\u0669]+)\)/g,`($1)`).replace(/\)([0-9\u0660-\u0669]+)\(/g,`($1)`),M=e=>{let t=e;return t=t.replace(/\(([^(){}]+)\}/g,`{$1}`),t.replace(/\{([^(){}]+)\)/g,`{$1}`)},N=e=>e.replace(/«([^»)]+)\)/g,`«$1»`).replace(/\(([^()]+)»/g,`«$1»`).replace(/«([^»]+)(?=\s*$|$)/g,`«$1»`),P=e=>{let t=/^\((?:\d+|۱|۲|۳|۴|۵|۶|۷|۸|۹)\)\s/,n=[],r=e.split(`
+`),me=e=>e.replace(/(\n\s*){2,}/g,`
+`),he=e=>e.replace(/\. +\./g,`.`),ge=e=>e.replace(/ـ{2,}/g,`ـ`).replace(/_+/g,`_`),m=e=>e.replace(/(\(|\)){2,}|(\[|\]){2,}/g,`$1$2`),h=e=>e.replace(/(\S) *(\([^)]*\))/g,`$1 $2`),g=e=>e.replace(/(\S) *(«[^»]*»)/g,`$1 $2`),_=e=>e.replace(/\(«|\( \(/g,`«`).replace(/»\)|\) \)/g,`»`).replace(/\)([0-9\u0660-\u0669]+)\)/g,`($1)`).replace(/\)([0-9\u0660-\u0669]+)\(/g,`($1)`),v=e=>{let t=e;return t=t.replace(/\(([^(){}]+)\}/g,`{$1}`),t.replace(/\{([^(){}]+)\)/g,`{$1}`)},y=e=>e.replace(/«([^»)]+)\)/g,`«$1»`).replace(/\(([^()]+)»/g,`«$1»`).replace(/«([^»]+)(?=\s*$|$)/g,`«$1»`),b=e=>{let t=/^\((?:\d+|۱|۲|۳|۴|۵|۶|۷|۸|۹)\)\s/,n=[],r=e.split(`
 `),i=``;return r.forEach(e=>{let r=e.trim(),a=t.test(r),o=/^\(\d+\/\d+\)/.test(r);if(a&&!o)i&&=(n.push(i.trim()),``),n.push(r);else{i+=`${r} `;let e=i.trim().slice(-1);/[.!؟]/.test(e)&&(n.push(i.trim()),i=``)}}),i&&n.push(i.trim()),n.join(`
-`)},F=e=>{let t=e.replace(/[^\p{L}]/gu,``);return t.length===0?!1:t===t.toUpperCase()},I=e=>e.replace(/(\d+)\s?\/\s?(\d+)/g,`$1/$2`),L=e=>e.replace(/[ \t]+/g,` `),R=e=>e.replace(/([؟!])[.،]/g,`$1`),z=e=>e.replace(/([[(])\s*(.*?)\s*([\])])/g,`$1$2$3`),B=e=>e.replace(/\(\(\s?/g,`«`).replace(/\s?\)\)/g,`»`),V=e=>e.normalize(`NFKD`).replace(/[\u0300-\u036f]/g,``).trim(),H=e=>{let t={𝑎:`I`,𝑨:`g`,𝘼:`!`,𝑏:`J`,𝑩:`h`,𝘽:`?`,𝑐:`K`,𝑪:`i`,𝑑:`L`,𝑫:`j`,𝘿:`,`,𝑒:`M`,𝑬:`k`,𝙀:`.`,𝑓:`N`,𝑭:`l`,𝑔:`O`,𝑮:`m`,𝑯:`n`,𝑖:`Q`,𝑰:`o`,𝑗:`R`,𝑱:`p`,𝑘:`S`,𝑲:`q`,𝑙:`T`,𝑳:`r`,𝙇:`-`,𝑚:`U`,𝑴:`s`,𝑛:`V`,𝑵:`t`,𝑜:`W`,𝑶:`u`,𝑝:`X`,𝑷:`v`,𝑞:`Y`,𝑸:`w`,𝑟:`Z`,𝑹:`x`,𝑆:`A`,𝑺:`y`,𝑇:`B`,𝑻:`z`,𝑢:`a`,𝑈:`C`,𝑣:`b`,𝑉:`D`,𝑤:`c`,𝑊:`E`,𝑥:`d`,𝑋:`F`,𝑦:`e`,𝑌:`G`,𝑧:`f`,𝑍:`H`,"𝑕":`P`};return e.replace(/[\uD835\uDC62-\uD835\uDC7B\uD835\uDC46-\uD835\uDC5F\u{1D63C}-\u{1D647}]/gu,e=>t[e]||e)},U=e=>H(V(e)),W=e=>e.toLowerCase().split(` `).map(e=>{if(e.length===0)return e;let t=e.match(/\p{L}/u);if(!t||t.index===void 0)return e;let n=t.index;return e.slice(0,n)+e.charAt(n).toUpperCase()+e.slice(n+1)}).join(` `),G=e=>e.replace(/([“”"]|«) *(.*?) *([“”"]|»)/g,`$1$2$3`),K=e=>{let t=e.replace(/(\b\d+\b)(?=:)/g,`"$1"`);return t=t.replace(/:\s*'([^']+)'/g,`: "$1"`),t=t.replace(/:\s*"([^"]+)"/g,`: "$1"`),JSON.stringify(JSON.parse(t))},q=e=>/^{(\s*(\d+|'[^']*'|"[^"]*")\s*:\s*('|")[^'"]*\3\s*,)*(?:\s*(\d+|'[^']*'|"[^"]*")\s*:\s*('|")[^'"]*\5\s*)}$/.test(e.trim()),J=e=>(e.match(/(?:[^\s"]+|"(.*?)")+/g)||[]).map(e=>e.startsWith(`"`)?e.slice(1,-1):e),Y=e=>{let t=0;for(let n of e)n===`"`&&t++;return t%2==0},X={"(":`)`,"[":`]`,"{":`}`},re=new Set([`(`,`[`,`{`]),ie=new Set([`)`,`]`,`}`]),ae=e=>{let t=[];for(let n of e)if(re.has(n))t.push(n);else if(ie.has(n)){let e=t.pop();if(!e||X[e]!==n)return!1}return t.length===0},oe=e=>Y(e)&&ae(e),se=e=>{if(e.includes(`-`)){let[t,n]=e.split(`-`).map(Number);if(t>n)throw Error(`Start page cannot be greater than end page`);return Array.from({length:n-t+1},(e,n)=>t+n)}else return e.split(`,`).map(Number)},ce=e=>e.replace(/ *\(?:\d+(?:\/\d+){0,2}\)? *| *\[\d+(?:\/\d+)?\] *| *«\d+» *|\d+\/\d+(?:\/\d+)?|[،§{}۝؍‎﴿﴾<>;_؟»«:!،؛[\]…ـ¬.\\/*()"]/g,` `),le=e=>e.replace(/-\[\d+\]-/g,``),ue=e=>e.replace(/\s+/g,` `),de=e=>e.replace(/[0-9]/g,``),fe=e=>e.replace(/\[(d)\.\s*\d{1,4}[hH]\]\s*|\((d)\.\s*\d{1,4}[hH]\)\s*/g,``),pe=e=>e.replace(/[\d-]/g,``),me=e=>e.replace(/\(\d{1}\)|\[\d{1}\]|«\d»/g,``),he=e=>e.replace(/https?:\/\/(www\.)?[-a-zA-Z0-9@:%._+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()@:%_+.~#?&//=]*)/g,``),ge=e=>e.replace(/\*\*([^*]+)\*\*/g,`$1`).replace(/__([^_]+)__/g,`$1`).replace(/\*([^*]+)\*/g,`$1`).replace(/_([^_]+)_/g,`$1`).replace(/~~([^~]+)~~/g,`$1`).replace(/^\s*>\s?/gm,``).replace(/!\[[^\]]*]\([^)]*\)/g,``).replace(/\[([^\]]+)]\([^)]*\)/g,`$1`).replace(/^#+\s*/gm,``).replace(/^\s*[-*+]\s+/gm,``).replace(/^\s*\d+\.\s+/gm,``).replace(/`/gm,``),_e=(e,t=150)=>e.length>t?`${e.substring(0,t-1)}…`:e,ve=(e,t=50,n)=>{if(e.length<=t)return e;let r=Math.max(3,Math.floor(t/3)),i=n??r,a=t-1-i;return a<1?`${e.substring(0,t-1)}…`:`${e.substring(0,a)}…${e.substring(e.length-i)}`},ye=e=>e.replace(/\\ /g,` `).trim(),be=[[`ا`,`آ`,`أ`,`إ`],[`ة`,`ه`],[`ى`,`ي`]],xe=e=>{for(let t of be)if(t.includes(e))return`[${t.map(e=>f(e)).join(``)}]`;return f(e)},Se=e=>e.normalize(`NFC`).replace(/[\u200C\u200D]/g,``).replace(/\s+/g,` `).trim(),Ce=e=>{let t=Se(e);return Array.from(t).map(e=>xe(e)+`[ًٌٍَُِّْ]*`).join(``)},we=e=>e.replace(/(\b|\W)(Al |Al-|Ar-|As-|Adh-|Ad-|Ats-|Ath |Ath-|Az |Az-|az-|adh-|as-|ar-)/g,`$1al-`).replace(/(\b|\W)(Ash-S|ash-S)/g,`$1al-S`).replace(/al- (.+?)\b/g,`al-$1`),Te=e=>e.replace(/ʿʿ/g,`ʿ`).replace(/ʾʾ/g,`ʾ`),Ee=e=>e.replace(/\(peace be upon him\)|(Messenger of (Allah|Allāh)|Messenger|Prophet|Mu[hḥ]ammad) *\((s[^)]*m|peace[^)]*him|May[^)]*him|may[^)]*him)\)*/gi,`$1 ﷺ`).replace(/,\s*ﷺ\s*,/g,` ﷺ`),Z=e=>e.normalize(`NFKD`).replace(/[\u0300-\u036f]/g,``).replace(/`|ʾ|ʿ|-/g,``),Q=e=>L(e.replace(/(\bal-|\bli-|\bbi-|\bfī|\bwa[-\s]+|\bl-|\bliʿl|\Bʿalá|\Bʿan|\bb\.)/gi,``)),$=e=>Z(Q(e)),De=e=>$(e).trim().split(/[ -]/).slice(0,2).map(e=>e.charAt(0).toUpperCase()).join(``);export{e as PATTERN_ENDS_WITH_PUNCTUATION,ee as addSpaceBeforeAndAfterPunctuation,s as addSpaceBetweenArabicTextAndNumbers,te as applySmartQuotes,t as arabicNumeralToNumber,n as cleanExtremeArabicUnderscores,ne as cleanLiteralNewLines,g as cleanMultilines,y as cleanSpacesBeforePeriod,ce as cleanSymbolsAndPartReferences,le as cleanTrailingPageNumbers,b as condenseAsterisks,x as condenseColons,S as condenseDashes,C as condenseEllipsis,E as condensePeriods,D as condenseUnderscores,r as convertUrduSymbolsToArabic,O as doubleToSingleBrackets,k as ensureSpaceBeforeBrackets,A as ensureSpaceBeforeQuotes,f as escapeRegex,De as extractInitials,a as findLastPunctuation,j as fixBracketTypos,M as fixCurlyBraces,N as fixMismatchedQuotationMarks,o as fixTrailingWow,P as formatStringBySentence,i as getArabicScore,_ as hasWordInSingleLine,h as insertLineBreaksAfterPunctuation,F as isAllUppercase,oe as isBalanced,q as isJsonStructureValid,v as isOnlyPunctuation,Ce as makeDiacriticInsensitive,p as makeDiacriticInsensitiveRegex,Z as normalize,we as normalizeArabicPrefixesToAl,Te as normalizeDoubleApostrophes,K as normalizeJsonSyntax,I as normalizeSlashInReferences,L as normalizeSpaces,$ as normalizeTransliteratedEnglish,se as parsePageRanges,w as reduceMultilineBreaksToDouble,T as reduceMultilineBreaksToSingle,m as removeAllTags,Q as removeArabicPrefixes,fe as removeDeathYear,ge as removeMarkdownFormatting,c as removeNonIndexSignatures,pe as removeNumbersAndDashes,R as removeRedundantPunctuation,me as removeSingleDigitReferences,l as removeSingularCodes,u as removeSolitaryArabicLetters,z as removeSpaceInsideBrackets,he as removeUrls,B as replaceDoubleBracketsWithArrows,d as replaceEnglishPunctuationWithArabic,ue as replaceLineBreaksWithSpaces,Ee as replaceSalutationsWithSymbol,J as splitByQuotes,de as stripAllDigits,V as stripBoldStyling,H as stripItalicsStyling,U as stripStyling,W as toTitleCase,G as trimSpaceInsideQuotes,_e as truncate,ve as truncateMiddle,ye as unescapeSpaces};
+`)},x=e=>{let t=e.replace(/[^\p{L}]/gu,``);return t.length===0?!1:t===t.toUpperCase()},S=e=>e.replace(/(\d+)\s?\/\s?(\d+)/g,`$1/$2`),C=e=>e.replace(/[ \t]+/g,` `),w=e=>e.replace(/([؟!])[.،]/g,`$1`),T=e=>e.replace(/([[(])\s*(.*?)\s*([\])])/g,`$1$2$3`),E=e=>e.replace(/\(\(\s?/g,`«`).replace(/\s?\)\)/g,`»`),D=e=>e.normalize(`NFKD`).replace(/[\u0300-\u036f]/g,``).trim(),O=e=>{let t={𝑎:`I`,𝑨:`g`,𝘼:`!`,𝑏:`J`,𝑩:`h`,𝘽:`?`,𝑐:`K`,𝑪:`i`,𝑑:`L`,𝑫:`j`,𝘿:`,`,𝑒:`M`,𝑬:`k`,𝙀:`.`,𝑓:`N`,𝑭:`l`,𝑔:`O`,𝑮:`m`,𝑯:`n`,𝑖:`Q`,𝑰:`o`,𝑗:`R`,𝑱:`p`,𝑘:`S`,𝑲:`q`,𝑙:`T`,𝑳:`r`,𝙇:`-`,𝑚:`U`,𝑴:`s`,𝑛:`V`,𝑵:`t`,𝑜:`W`,𝑶:`u`,𝑝:`X`,𝑷:`v`,𝑞:`Y`,𝑸:`w`,𝑟:`Z`,𝑹:`x`,𝑆:`A`,𝑺:`y`,𝑇:`B`,𝑻:`z`,𝑢:`a`,𝑈:`C`,𝑣:`b`,𝑉:`D`,𝑤:`c`,𝑊:`E`,𝑥:`d`,𝑋:`F`,𝑦:`e`,𝑌:`G`,𝑧:`f`,𝑍:`H`,"𝑕":`P`};return e.replace(/[\uD835\uDC62-\uD835\uDC7B\uD835\uDC46-\uD835\uDC5F\u{1D63C}-\u{1D647}]/gu,e=>t[e]||e)},k=e=>O(D(e)),A=e=>e.toLowerCase().split(` `).map(e=>{if(e.length===0)return e;let t=e.match(/\p{L}/u);if(!t||t.index===void 0)return e;let n=t.index;return e.slice(0,n)+e.charAt(n).toUpperCase()+e.slice(n+1)}).join(` `),j=e=>e.replace(/([“”"]|«) *(.*?) *([“”"]|»)/g,`$1$2$3`),M=e=>{let t=e.replace(/(\b\d+\b)(?=:)/g,`"$1"`);return t=t.replace(/:\s*'([^']+)'/g,`: "$1"`),t=t.replace(/:\s*"([^"]+)"/g,`: "$1"`),JSON.stringify(JSON.parse(t))},N=e=>/^{(\s*(\d+|'[^']*'|"[^"]*")\s*:\s*('|")[^'"]*\3\s*,)*(?:\s*(\d+|'[^']*'|"[^"]*")\s*:\s*('|")[^'"]*\5\s*)}$/.test(e.trim()),P=e=>(e.match(/(?:[^\s"]+|"(.*?)")+/g)||[]).map(e=>e.startsWith(`"`)?e.slice(1,-1):e),F=e=>{let t=0;for(let n of e)n===`"`&&t++;return t%2==0},I={"(":`)`,"[":`]`,"{":`}`},L=new Set([`(`,`[`,`{`]),_e=new Set([`)`,`]`,`}`]),ve=e=>{let t=[];for(let n of e)if(L.has(n))t.push(n);else if(_e.has(n)){let e=t.pop();if(!e||I[e]!==n)return!1}return t.length===0},ye=e=>F(e)&&ve(e),be=e=>{if(e.includes(`-`)){let[t,n]=e.split(`-`).map(Number);if(t>n)throw Error(`Start page cannot be greater than end page`);return Array.from({length:n-t+1},(e,n)=>t+n)}else return e.split(`,`).map(Number)},R=new Uint8Array(65536),z=(e,t)=>{for(let n=0;n<e.length;n++)R[e.charCodeAt(n)]|=t};z(` 	`,1),z(`
+\r`,64);for(let e=48;e<=57;e++)R[e]|=16;for(let e=1536;e<=1791;e++)R[e]|=32;for(let e=1872;e<=1919;e++)R[e]|=32;z(`.!?,:;`,194),z(`،؛؟`,194),z(`([{"'«“`,68),z(`)]}"'»”`,74),z(`-_*و/`,64);const B=8221,V=1608,H=1548,U=1563,W=1567,G=1600,K=8230,q=e=>e===32||e===9||e===10||e===13;var xe=class{buffer;length;constructor(e){this.buffer=new Uint16Array(Math.max(16,e)),this.length=0}last(){return this.length>0?this.buffer[this.length-1]:0}secondLast(){return this.length>1?this.buffer[this.length-2]:0}push(e){this.ensureCapacity(1),this.buffer[this.length]=e,this.length++}pop(){this.length>0&&this.length--}ensureCapacity(e){let t=this.length+e;if(t<=this.buffer.length)return;let n=this.buffer.length*2;n<t&&(n=t);let r=new Uint16Array(n);r.set(this.buffer.subarray(0,this.length)),this.buffer=r}toStringTrimmed(){let e=0,t=this.length;for(;e<t&&q(this.buffer[e]);)e++;for(;t>e&&q(this.buffer[t-1]);)t--;if(t<=e)return``;let n=32768,r=``;for(let i=e;i<t;i+=n){let e=this.buffer.subarray(i,Math.min(t,i+n));r+=String.fromCharCode(...e)}return r}};const Se=e=>{if(!e)return``;let t=``,n=e.length,r=0,i=0,a=0;for(;r<n;){let o=e.charCodeAt(r),s=o,c=R[o];if(c&1){i++,r++;continue}if(c&64){if(o===10||o===13){for(i=0,a!==10&&(t+=`
+`,a=10),r++;r<n;){let t=e.charCodeAt(r);if(t===32||t===9||t===10||t===13)r++;else break}continue}if(o===63)o=W;else if(o===59)o=U;else if(o===44)o=H;else if(o===58){(a===46||a===45)&&(t=t.slice(0,-1),a=t.charCodeAt(t.length-1)||0);let n=e.charCodeAt(r+1);(n===46||n===45)&&r++}else if(o===40&&e.charCodeAt(r+1)===40)o=171,r++;else if(o===41&&e.charCodeAt(r+1)===41)o=187,r++;else if(o===46){if(a===K){r++;continue}if(a===46){t=t.slice(0,-1)+`…`,a=K,r++;continue}}else if(o===V&&a===32){for(t+=`و`,a=V,r++;r<n;){let t=e.charCodeAt(r);if(t===32||t===9)r++;else break}continue}else if(o===G&&a===G||o===95&&a===95||o===45&&a===45||o===42&&a===42){r++;continue}if((a===W||a===33)&&(o===46||o===H)){r++;continue}}if(i>0){let s=!0,c=R[o];if(c&2&&(s=!1),R[a]&4&&(s=!1),o===47){let t=0;for(let i=r+1;i<n;i++){let n=e.charCodeAt(i);if(n!==32&&n!==9&&n!==10&&n!==13){t=n;break}}R[a]&16&&R[t]&16&&(s=!1)}a===47&&c&16&&t.length>=2&&R[t.charCodeAt(t.length-2)]&16&&(s=!1),s&&(t+=` `,a=32),i=0}let l=R[o];l&16&&R[a]&32&&(t+=` `,a=32),l&4&&a!==32&&a!==10&&!(R[a]&4)&&a!==0&&(t+=` `,a=32),R[a]&128&&(l&13||o===32||o===10||o===13||l&8||o===34||o===39||o===187||o===B||o===a||(a===W||a===33)&&(o===46||o===H)||(t+=` `,a=32)),o===s?t+=e[r]:t+=String.fromCharCode(o),a=o,r++}return t.trim()},Ce=e=>{if(!e)return``;let t=e.length,n=0,r=new xe(t+(t>>3)+64),i=0,a=0;for(;n<t;){let o=e.charCodeAt(n),s=o,c=R[o];if(c&1){i++,n++;continue}if(c&64){if(o===10||o===13){for(i=0,a!==10&&(r.push(10),a=10),n++;n<t;){let t=e.charCodeAt(n);if(t===32||t===9||t===10||t===13)n++;else break}continue}if(o===63)o=W;else if(o===59)o=U;else if(o===44)o=H;else if(o===58){(a===46||a===45)&&(r.pop(),a=r.last());let t=e.charCodeAt(n+1);(t===46||t===45)&&n++}else if(o===40&&e.charCodeAt(n+1)===40)o=171,n++;else if(o===41&&e.charCodeAt(n+1)===41)o=187,n++;else if(o===46){if(a===K){n++;continue}if(a===46){r.pop(),r.push(K),a=K,n++;continue}}else if(o===V&&a===32){for(r.push(V),a=V,n++;n<t;){let t=e.charCodeAt(n);if(t===32||t===9)n++;else break}continue}else if(o===G&&a===G||o===95&&a===95||o===45&&a===45||o===42&&a===42){n++;continue}if((a===W||a===33)&&(o===46||o===H)){n++;continue}}if(i>0){let s=!0,c=R[o];if(c&2&&(s=!1),R[a]&4&&(s=!1),o===47){let r=0;for(let i=n+1;i<t;i++){let t=e.charCodeAt(i);if(t!==32&&t!==9&&t!==10&&t!==13){r=t;break}}R[a]&16&&R[r]&16&&(s=!1)}a===47&&c&16&&r.length>=2&&R[r.secondLast()]&16&&(s=!1),s&&(r.push(32),a=32),i=0}let l=R[o];l&16&&R[a]&32&&(r.push(32),a=32),l&4&&a!==32&&a!==10&&!(R[a]&4)&&a!==0&&(r.push(32),a=32),R[a]&128&&(l&13||o===32||o===10||o===13||l&8||o===34||o===39||o===187||o===B||o===a||(a===W||a===33)&&(o===46||o===H)||(r.push(32),a=32)),o===s?r.push(s):r.push(o),a=o,n++}return r.toStringTrimmed()},J=e=>Se(e),we=e=>Ce(e),Y=e=>{let t=process.env.BITABOOM_PREFORMAT_BUILDER;return t===`concat`?J(e):t===`buffer`?we(e):J(e)},Te=e=>Array.isArray(e)?e.map(Y):Y(e),Ee=e=>e.replace(/ *\(?:\d+(?:\/\d+){0,2}\)? *| *\[\d+(?:\/\d+)?\] *| *«\d+» *|\d+\/\d+(?:\/\d+)?|[،§{}۝؍‎﴿﴾<>;_؟»«:!،؛[\]…ـ¬.\\/*()"]/g,` `),De=e=>e.replace(/-\[\d+\]-/g,``),Oe=e=>e.replace(/\s+/g,` `),ke=e=>e.replace(/[0-9]/g,``),Ae=e=>e.replace(/\[(d)\.\s*\d{1,4}[hH]\]\s*|\((d)\.\s*\d{1,4}[hH]\)\s*/g,``),je=e=>e.replace(/[\d-]/g,``),Me=e=>e.replace(/\(\d{1}\)|\[\d{1}\]|«\d»/g,``),Ne=e=>e.replace(/https?:\/\/(www\.)?[-a-zA-Z0-9@:%._+~#=]{1,256}\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()@:%_+.~#?&//=]*)/g,``),Pe=e=>e.replace(/\*\*([^*]+)\*\*/g,`$1`).replace(/__([^_]+)__/g,`$1`).replace(/\*([^*]+)\*/g,`$1`).replace(/_([^_]+)_/g,`$1`).replace(/~~([^~]+)~~/g,`$1`).replace(/^\s*>\s?/gm,``).replace(/!\[[^\]]*]\([^)]*\)/g,``).replace(/\[([^\]]+)]\([^)]*\)/g,`$1`).replace(/^#+\s*/gm,``).replace(/^\s*[-*+]\s+/gm,``).replace(/^\s*\d+\.\s+/gm,``).replace(/`/gm,``),Fe=(e,t=150)=>e.length>t?`${e.substring(0,t-1)}…`:e,Ie=(e,t=50,n)=>{if(e.length<=t)return e;let r=Math.max(3,Math.floor(t/3)),i=n??r,a=t-1-i;return a<1?`${e.substring(0,t-1)}…`:`${e.substring(0,a)}…${e.substring(e.length-i)}`},Le=e=>e.replace(/\\ /g,` `).trim(),X=[[`ا`,`آ`,`أ`,`إ`],[`ة`,`ه`],[`ى`,`ي`]],Re=e=>{for(let t of X)if(t.includes(e))return`[${t.map(e=>f(e)).join(``)}]`;return f(e)},ze=e=>e.normalize(`NFC`).replace(/[\u200C\u200D]/g,``).replace(/\s+/g,` `).trim(),Be=e=>{let t=ze(e);return Array.from(t).map(e=>Re(e)+`[ًٌٍَُِّْ]*`).join(``)},Ve=e=>e.replace(/(\b|\W)(Al |Al-|Ar-|As-|Adh-|Ad-|Ats-|Ath |Ath-|Az |Az-|az-|adh-|as-|ar-)/g,`$1al-`).replace(/(\b|\W)(Ash-S|ash-S)/g,`$1al-S`).replace(/al- (.+?)\b/g,`al-$1`),He=e=>e.replace(/ʿʿ/g,`ʿ`).replace(/ʾʾ/g,`ʾ`),Ue=e=>e.replace(/\(peace be upon him\)|(Messenger of (Allah|Allāh)|Messenger|Prophet|Mu[hḥ]ammad) *\((s[^)]*m|peace[^)]*him|May[^)]*him|may[^)]*him)\)*/gi,`$1 ﷺ`).replace(/,\s*ﷺ\s*,/g,` ﷺ`),Z=e=>e.normalize(`NFKD`).replace(/[\u0300-\u036f]/g,``).replace(/`|ʾ|ʿ|-/g,``),Q=e=>C(e.replace(/(\bal-|\bli-|\bbi-|\bfī|\bwa[-\s]+|\bl-|\bliʿl|\Bʿalá|\Bʿan|\bb\.)/gi,``)),$=e=>Z(Q(e)),We=e=>$(e).trim().split(/[ -]/).slice(0,2).map(e=>e.charAt(0).toUpperCase()).join(``);export{e as PATTERN_ENDS_WITH_PUNCTUATION,ne as addSpaceBeforeAndAfterPunctuation,s as addSpaceBetweenArabicTextAndNumbers,re as applySmartQuotes,t as arabicNumeralToNumber,n as cleanExtremeArabicUnderscores,ie as cleanLiteralNewLines,ae as cleanMultilines,ce as cleanSpacesBeforePeriod,Ee as cleanSymbolsAndPartReferences,De as cleanTrailingPageNumbers,le as condenseAsterisks,ue as condenseColons,de as condenseDashes,fe as condenseEllipsis,he as condensePeriods,ge as condenseUnderscores,r as convertUrduSymbolsToArabic,m as doubleToSingleBrackets,h as ensureSpaceBeforeBrackets,g as ensureSpaceBeforeQuotes,f as escapeRegex,We as extractInitials,a as findLastPunctuation,_ as fixBracketTypos,v as fixCurlyBraces,y as fixMismatchedQuotationMarks,o as fixTrailingWow,b as formatStringBySentence,i as getArabicScore,oe as hasWordInSingleLine,te as insertLineBreaksAfterPunctuation,x as isAllUppercase,ye as isBalanced,N as isJsonStructureValid,se as isOnlyPunctuation,Be as makeDiacriticInsensitive,p as makeDiacriticInsensitiveRegex,Z as normalize,Ve as normalizeArabicPrefixesToAl,He as normalizeDoubleApostrophes,M as normalizeJsonSyntax,S as normalizeSlashInReferences,C as normalizeSpaces,$ as normalizeTransliteratedEnglish,be as parsePageRanges,Te as preformatArabicText,pe as reduceMultilineBreaksToDouble,me as reduceMultilineBreaksToSingle,ee as removeAllTags,Q as removeArabicPrefixes,Ae as removeDeathYear,Pe as removeMarkdownFormatting,c as removeNonIndexSignatures,je as removeNumbersAndDashes,w as removeRedundantPunctuation,Me as removeSingleDigitReferences,l as removeSingularCodes,u as removeSolitaryArabicLetters,T as removeSpaceInsideBrackets,Ne as removeUrls,E as replaceDoubleBracketsWithArrows,d as replaceEnglishPunctuationWithArabic,Oe as replaceLineBreaksWithSpaces,Ue as replaceSalutationsWithSymbol,P as splitByQuotes,ke as stripAllDigits,D as stripBoldStyling,O as stripItalicsStyling,k as stripStyling,A as toTitleCase,j as trimSpaceInsideQuotes,Fe as truncate,Ie as truncateMiddle,Le as unescapeSpaces};
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","names":["DIACRITICS_CLASS","sentences: string[]","italicMap: Record<string, string>","stack: string[]","EQUIV_GROUPS: string[][]"],"sources":["../src/constants.ts","../src/arabic.ts","../src/cleaning.ts","../src/formatting.ts","../src/parsing.ts","../src/sanitization.ts","../src/transliteration.ts"],"sourcesContent":["/** Matches text ending with common punctuation marks */\nexport const PATTERN_ENDS_WITH_PUNCTUATION = /[.!?؟؛]$/;\n","import { PATTERN_ENDS_WITH_PUNCTUATION } from './constants';\n\n/**\n * Converts Arabic-Indic numerals (٠-٩) to a JavaScript number.\n *\n * This function finds all Arabic-Indic digits in the input string and converts them\n * to their corresponding Arabic (Western) digits, then parses the result as an integer.\n *\n * Arabic-Indic digits mapping:\n * - ٠ → 0, ١ → 1, ٢ → 2, ٣ → 3, ٤ → 4\n * - ٥ → 5, ٦ → 6, ٧ → 7, ٨ → 8, ٩ → 9\n *\n * @param arabic - The string containing Arabic-Indic numerals to convert\n * @returns The parsed integer value of the converted numerals\n *\n * @example\n * ```typescript\n * arabicNumeralToNumber(\"١٢٣\"); // returns 123\n * arabicNumeralToNumber(\"٥٠\"); // returns 50\n * arabicNumeralToNumber(\"abc١٢٣xyz\"); // returns 123 (non-digits ignored)\n * arabicNumeralToNumber(\"\"); // returns NaN\n * ```\n *\n * Returns NaN if no valid Arabic-Indic digits are found\n */\nexport const arabicNumeralToNumber = (arabic: string) => {\n    return parseInt(\n        arabic.replace(/[\\u0660-\\u0669]/g, (c) => (c.charCodeAt(0) - 0x0660).toString()),\n        10,\n    );\n};\n\n/**\n * Removes extreme Arabic underscores (ـ) that appear at the beginning or end of a line or in text.\n * Does not affect Hijri dates (e.g., 1424هـ) or specific Arabic terms.\n * Example: \"ـThis is a textـ\" will be changed to \"This is a text\".\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with extreme underscores removed.\n */\nexport const cleanExtremeArabicUnderscores = (text: string) => {\n    return text.replace(/(?<!\\d ?ه|اه)ـ(?=\\r?$)|^ـ(?!اهـ)/gm, '');\n};\n\n/**\n * Converts Urdu symbols to their Arabic equivalents.\n * Example: 'ھذا' will be changed to 'هذا', 'ی' to 'ي'.\n * @param {string} text - The input text containing Urdu symbols.\n * @returns {string} - The modified text with Urdu symbols converted to Arabic symbols.\n */\nexport const convertUrduSymbolsToArabic = (text: string) => {\n    return text.replace(/ھ/g, 'ه').replace(/ی/g, 'ي');\n};\n\n/**\n * Calculates the proportion of Arabic characters in text relative to total non-whitespace, non-digit characters.\n * Digits (ASCII and Arabic-Indic variants) are excluded from both numerator and denominator.\n * @param text - The input text to analyze\n * @returns A decimal between 0-1 representing the Arabic character ratio (0 = no Arabic, 1 = all Arabic)\n */\nexport const getArabicScore = (text: string) => {\n    if (!text) {\n        return 0;\n    }\n    // Arabic letters (letters/ranges only)\n    const arabicLettersPattern = /[\\u0600-\\u06FF\\u0750-\\u077F\\u08A0-\\u08FF\\uFB50-\\uFDFF\\uFE70-\\uFEFF]/g;\n    // ASCII digits + Arabic-Indic digits + Extended Arabic-Indic digits\n    const allDigitPattern = /[0-9\\u0660-\\u0669\\u06F0-\\u06F9]/g;\n    // Counted characters exclude whitespace and all listed digits\n    const countedCharsPattern = /[^\\s0-9\\u0660-\\u0669\\u06F0-\\u06F9]/g;\n    const cleaned = text.replace(allDigitPattern, '');\n    const arabicMatches = cleaned.match(arabicLettersPattern) || [];\n    const totalMatches = cleaned.match(countedCharsPattern) || [];\n    return totalMatches.length === 0 ? 0 : arabicMatches.length / totalMatches.length;\n};\n\n/**\n * Finds the position of the last punctuation character in a string\n *\n * @param text - The text to search through\n * @returns The index of the last punctuation character, or -1 if none found\n *\n * @example\n * ```typescript\n * const text = \"Hello world! How are you?\";\n * const lastPuncIndex = findLastPunctuation(text);\n * // Result: 24 (position of the last '?')\n *\n * const noPuncText = \"Hello world\";\n * const notFound = findLastPunctuation(noPuncText);\n * // Result: -1 (no punctuation found)\n * ```\n */\nexport const findLastPunctuation = (text: string) => {\n    for (let i = text.length - 1; i >= 0; i--) {\n        if (PATTERN_ENDS_WITH_PUNCTUATION.test(text[i])) {\n            return i;\n        }\n    }\n\n    return -1;\n};\n\n/**\n * Fixes the trailing \"و\" (wow) in phrases such as \"عليكم و رحمة\" to \"عليكم ورحمة\".\n * This function attempts to correct phrases where \"و\" appears unnecessarily, particularly in greetings.\n * Example: 'السلام عليكم و رحمة' will be changed to 'السلام عليكم ورحمة'.\n * @param {string} text - The input text containing the \"و\" character.\n * @returns {string} - The modified text with unnecessary trailing \"و\" characters corrected.\n */\nexport const fixTrailingWow = (text: string) => {\n    return text.replace(/ و /g, ' و');\n};\n\n/**\n * Inserts a space between Arabic text and numbers.\n * Example: 'الآية37' will be changed to 'الآية 37'.\n * @param {string} text - The input text containing Arabic text followed by numbers.\n * @returns {string} - The modified text with spaces inserted between Arabic text and numbers.\n */\nexport const addSpaceBetweenArabicTextAndNumbers = (text: string) => {\n    return text.replace(/([\\u0600-\\u06FF]+)(\\d+)/g, '$1 $2');\n};\n\n/**\n * Removes single-digit numbers surrounded by Arabic text. Also removes dashes (-) not followed by a number.\n * For example, removes '3' from 'وهب 3 وقال' but does not remove '121' from 'لوحه 121 الجرح'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with non-index numbers and dashes removed.\n */\nexport const removeNonIndexSignatures = (text: string) => {\n    return text\n        .replace(/(?<![0-9] ?)-|(?<=[\\u0600-\\u06FF])\\s?\\d\\s?(?=[\\u0600-\\u06FF])/g, ' ')\n        .replace(/(?<=[\\u0600-\\u06FF]\\s)(\\d+\\s)+\\d+(?=(\\s[\\u0600-\\u06FF]|$))/g, ' ');\n};\n\n/**\n * Removes characters enclosed in square brackets [] or parentheses () if they are Arabic letters or Arabic-Indic numerals.\n * Example: '[س]' or '(س)' will be removed.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with singular codes removed.\n */\nexport const removeSingularCodes = (text: string) => {\n    return text.replace(/[[({][\\u0621-\\u064A\\u0660-\\u0669][\\])}]/g, '');\n};\n\n/**\n * Removes solitary Arabic letters unless they are the 'ha' letter, which is used in Hijri years.\n * Example: \"ب ا الكلمات ت\" will be changed to \"ا الكلمات\".\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with solitary Arabic letters removed.\n */\nexport const removeSolitaryArabicLetters = (text: string) => {\n    return text.replace(/(^| )[\\u0621-\\u064A]( |$)/g, ' ');\n};\n\n/**\n * Replaces English punctuation (question mark and semicolon) with their Arabic equivalents.\n * Example: '?' will be replaced with '؟', and ';' with '؛'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with English punctuation replaced by Arabic punctuation.\n */\nexport const replaceEnglishPunctuationWithArabic = (text: string) => {\n    return text\n        .replace(/\\?|؟\\./g, '؟')\n        .replace(/(;|؛)\\s*(\\1\\s*)*/g, '؛')\n        .replace(/,|-،/g, '،');\n};\n","/** Character class for Arabic diacritics (tashkīl/harakāt). */\nconst DIACRITICS_CLASS = '[\\\\u0610-\\\\u061A\\\\u064B-\\\\u065F\\\\u0670\\\\u06D6-\\\\u06ED]';\n/** Tatweel (kashīda) class. */\nconst TATWEEL_CLASS = '\\\\u0640';\n\n/**\n * Escape a string so it can be safely embedded into a RegExp source.\n *\n * @param s Any string\n * @returns Escaped string\n */\nexport const escapeRegex = (s: string): string => s.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&');\n\n/** Optional equivalence toggles for {@link makeDiacriticInsensitiveRegex}. */\ntype EquivOptions = {\n    /** Treat ا/أ/إ/آ as equivalent. @default true */\n    alif?: boolean;\n    /** Treat ة/ه as equivalent. @default true */\n    taMarbutahHa?: boolean;\n    /** Treat ى/ي as equivalent. @default true */\n    alifMaqsurahYa?: boolean;\n};\n\n/** Options for {@link makeDiacriticInsensitiveRegex}. */\nexport type MakeRegexOptions = {\n    /**\n     * Character equivalences to allow.\n     * @default { alif: true, taMarbutahHa: true, alifMaqsurahYa: true }\n     */\n    equivalences?: EquivOptions;\n\n    /**\n     * Allow tatweel between letters (tolerate decorative elongation).\n     * @default true\n     */\n    allowTatweel?: boolean;\n\n    /**\n     * Ignore diacritics by inserting a `DIACRITICS_CLASS*` after each letter.\n     * @default true\n     */\n    ignoreDiacritics?: boolean;\n\n    /**\n     * Treat any whitespace in the needle as `\\s+` for flexible matching.\n     * @default true\n     */\n    flexWhitespace?: boolean;\n\n    /**\n     * RegExp flags to use.\n     * @default 'u'\n     */\n    flags?: string;\n};\n\n/**\n * Build a **diacritic-insensitive**, **tatweel-tolerant** RegExp for Arabic text matching.\n *\n * Features:\n * - Optional character equivalences: ا~أ~إ~آ, ة~ه, ى~ي.\n * - Optional tolerance for tatweel between characters.\n * - Optional diacritic-insensitivity (by inserting a diacritics class after each char).\n * - Optional flexible whitespace (needle whitespace becomes `\\s+`).\n *\n * @param needle The Arabic text to match\n * @param opts See {@link MakeRegexOptions}\n * @returns A `RegExp` matching the needle with the desired tolerances\n *\n * @example\n * const rx = makeDiacriticInsensitiveRegex('أنا إلى الآفاق');\n * rx.test('انا الي الافاق'); // true\n * rx.test('اَنا إلى الآفاق'); // true\n */\nexport const makeDiacriticInsensitiveRegex = (needle: string, opts: MakeRegexOptions = {}): RegExp => {\n    const {\n        equivalences = { alif: true, taMarbutahHa: true, alifMaqsurahYa: true },\n        allowTatweel = true,\n        ignoreDiacritics = true,\n        flexWhitespace = true,\n        flags = 'u',\n    } = opts;\n\n    // Safety guard against extremely large inputs causing excessive pattern sizes\n    if (needle.length > 5000) {\n        throw new Error('makeDiacriticInsensitiveRegex: needle too long');\n    }\n\n    const charClass = (ch: string): string => {\n        switch (ch) {\n            case 'ا':\n            case 'أ':\n            case 'إ':\n            case 'آ':\n                return equivalences.alif ? '[اأإآ]' : 'ا';\n            case 'ة':\n            case 'ه':\n                return equivalences.taMarbutahHa ? '[هة]' : escapeRegex(ch);\n            case 'ى':\n            case 'ي':\n                return equivalences.alifMaqsurahYa ? '[ىي]' : escapeRegex(ch);\n            default:\n                return escapeRegex(ch);\n        }\n    };\n\n    const after = `${ignoreDiacritics ? `${DIACRITICS_CLASS}*` : ''}${allowTatweel ? `${TATWEEL_CLASS}*` : ''}`;\n\n    let pattern = '';\n    for (const ch of Array.from(needle)) {\n        if (/\\s/.test(ch)) {\n            pattern += flexWhitespace ? '\\\\s+' : '\\\\s*';\n        } else {\n            pattern += `${charClass(ch)}${after}`;\n        }\n    }\n\n    return new RegExp(pattern, flags);\n};\n\nexport const removeAllTags = (content: string) => content.replace(/<[^>]*>/g, '');\n","/**\n * Adds line breaks after punctuation marks such as periods, exclamation points, and question marks.\n * Example: 'Text.' becomes 'Text.\\n'.\n * @param {string} text - The input text containing punctuation.\n * @returns {string} - The modified text with line breaks added after punctuation.\n */\nexport const insertLineBreaksAfterPunctuation = (text: string) => {\n    // Define the punctuation marks that should trigger a new line\n    const punctuation = /([.?!؟])/g;\n\n    // Replace occurrences of punctuation marks followed by a space with the punctuation mark, a newline, and the space\n    const formattedText = text.replace(punctuation, '$1\\n').replace(/\\n\\s+/g, '\\n').trim();\n\n    return formattedText;\n};\n\n/**\n * Adds spaces before and after punctuation, except for certain cases like quoted text or ayah references.\n * Example: 'Text,word' becomes 'Text, word'.\n * @param {string} text - The input text containing punctuation.\n * @returns {string} - The modified text with spaces added before and after punctuation.\n */\nexport const addSpaceBeforeAndAfterPunctuation = (text: string) => {\n    return text\n        .replace(/( ?)([.!?,،؟;؛])((?![ '”“)\"\\]\\n])|(?=\\s{2,}))/g, '$1$2 ')\n        .replace(/\\s([.!?,،؟;؛])\\s*([ '”“)\"\\]\\n])/g, '$1$2')\n        .replace(/([^\\s\\w\\d'”“)\"\\]]+)\\s+([.!?,،؟;؛])|([.!?,،؟;؛])\\s+$/g, '$1$2$3')\n        .replace(/(?<=\\D)( ?: ?)(?!(\\d+:)|(:\\d+))|(?<=\\d) ?: ?(?=\\D)|(?<=\\D) ?: ?(?=\\d)/g, ': ');\n};\n\n/**\n * Turns regular double quotes surrounding a body of text into smart quotes.\n * Also fixes incorrect starting quotes by ensuring the string starts with an opening quote if needed.\n * Example: 'The \"quick brown\" fox' becomes 'The “quick brown” fox'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with smart quotes applied.\n */\nexport const applySmartQuotes = (text: string) => {\n    return text\n        .replace(/[“”]/g, '\"')\n        .replace(/\"([^\"]*)\"/g, '“$1”')\n        .replace(/^”/g, '“');\n};\n\n/**\n * Replaces literal new line characters (\\n) and carriage returns (\\r) with actual line breaks.\n * Example: 'A\\\\nB' becomes 'A\\nB'.\n * @param {string} text - The input text containing literal new lines.\n * @returns {string} - The modified text with actual line breaks.\n */\nexport const cleanLiteralNewLines = (text: string) => {\n    return text.replace(/\\\\n|\\r/g, '\\n');\n};\n\n/**\n * Removes trailing spaces from each line in a multiline string.\n * Example: \" This is a line   \\nAnother line   \" becomes \"This is a line\\nAnother line\".\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with trailing spaces removed.\n */\nexport const cleanMultilines = (text: string) => {\n    return text.replace(/^ +| +$/gm, '');\n};\n\n/**\n * Detects if a word is by itself in a line.\n * @param text The text to check.\n * @returns true if there exists a word in any of the lines in the text that is by itself.\n */\nexport const hasWordInSingleLine = (text: string): boolean => {\n    return /^\\s*\\S+\\s*$/gm.test(text);\n};\n\n/**\n * Checks if the input string consists of only punctuation characters.\n * @param {string} text - The input text to check.\n * @returns {boolean} - Returns true if the string contains only punctuation, false otherwise.\n */\nexport const isOnlyPunctuation = (text: string): boolean => {\n    const regex = /^[\\u0020-\\u002f\\u003a-\\u0040\\u005b-\\u0060\\u007b-\\u007e0-9٠-٩]+$/;\n    return regex.test(text);\n};\n\n/**\n * Cleans unnecessary spaces before punctuation marks such as periods, commas, and question marks.\n * Example: 'This is a sentence , with extra space .' becomes 'This is a sentence, with extra space.'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with cleaned spaces before punctuation.\n */\nexport const cleanSpacesBeforePeriod = (text: string) => {\n    return text.replace(/\\s+([.؟!,،؛:?])/g, '$1');\n};\n\n/**\n * Condenses multiple asterisks (*) into a single one.\n * Example: '***' becomes '*'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed asterisks.\n */\nexport const condenseAsterisks = (text: string) => {\n    return text.replace(/(\\*\\s*)+/g, '*');\n};\n\n/**\n * Replaces occurrences of colons surrounded by periods (e.g., '.:.' or ':') with a single colon.\n * Example: 'This.:. is a test' becomes 'This: is a test'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed colons.\n */\nexport const condenseColons = (text: string) => {\n    return text.replace(/[.-]?:[.-]?/g, ':');\n};\n\n/**\n * Condenses two or more dashes (--) into a single dash (-).\n * Example: 'This is some ---- text' becomes 'This is some - text'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed dashes.\n */\nexport const condenseDashes = (text: string) => {\n    return text.replace(/-{2,}/g, '-');\n};\n\n/**\n * Replaces sequences of two or more periods (e.g., '...') with an ellipsis character (…).\n * Example: 'This is a test...' becomes 'This is a test…'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with ellipses condensed.\n */\nexport const condenseEllipsis = (text: string) => {\n    return text.replace(/\\.{2,}/g, '…');\n};\n\n/**\n * Reduces multiple consecutive line breaks (3 or more) to exactly 2 line breaks.\n * Example: 'This is line 1\\n\\n\\n\\nThis is line 2' becomes 'This is line 1\\n\\nThis is line 2'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed line breaks.\n */\nexport const reduceMultilineBreaksToDouble = (text: string) => {\n    return text.replace(/(\\n\\s*){3,}/g, '\\n\\n');\n};\n\n/**\n * Reduces multiple consecutive line breaks (2 or more) to exactly 1 line break.\n * Example: 'This is line 1\\n\\nThis is line 2' becomes 'This is line 1\\nThis is line 2'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed line breaks.\n */\nexport const reduceMultilineBreaksToSingle = (text: string) => {\n    return text.replace(/(\\n\\s*){2,}/g, '\\n');\n};\n\n/**\n * Condenses multiple periods separated by spaces (e.g., '. . .') into a single period.\n * Example: 'This . . . is a test' becomes 'This. is a test'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed periods.\n */\nexport const condensePeriods = (text: string) => {\n    return text.replace(/\\. +\\./g, '.');\n};\n\n/**\n * Condenses multiple underscores (__) or Arabic Tatweel characters (ـــــ) into a single underscore or Tatweel.\n * Example: 'This is ـــ some text __' becomes 'This is ـ some text _'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed underscores.\n */\nexport const condenseUnderscores = (text: string) => {\n    return text.replace(/ـ{2,}/g, 'ـ').replace(/_+/g, '_');\n};\n\n/**\n * Replaces double parentheses or brackets with single ones.\n * Example: '((text))' becomes '(text)'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed brackets.\n */\nexport const doubleToSingleBrackets = (text: string) => {\n    return text.replace(/(\\(|\\)){2,}|(\\[|\\]){2,}/g, '$1$2');\n};\n\n/**\n * Ensures at most 1 space exists before any word before brackets.\n * Adds a space if there isn't one, or reduces multiple spaces to one.\n * @param {string} text - The input text to modify\n * @returns {string} - The modified text with proper spacing before brackets\n */\nexport const ensureSpaceBeforeBrackets = (text: string) => {\n    return text.replace(/(\\S) *(\\([^)]*\\))/g, '$1 $2');\n};\n\n/**\n * Ensures at most 1 space exists before any word before Arabic quotation marks.\n * Adds a space if there isn't one, or reduces multiple spaces to one.\n * @param {string} text - The input text to modify\n * @returns {string} - The modified text with proper spacing before Arabic quotes\n */\nexport const ensureSpaceBeforeQuotes = (text: string) => {\n    return text.replace(/(\\S) *(«[^»]*»)/g, '$1 $2');\n};\n\n/**\n * Fixes common bracket and quotation mark typos in text\n * Corrects malformed patterns like \"(«\", \"»)\", and misplaced digits in brackets\n * @param text - Input text that may contain bracket typos\n * @returns Text with corrected bracket and quotation mark combinations\n */\nexport const fixBracketTypos = (text: string) => {\n    return (\n        text\n            .replace(/\\(«|\\( \\(/g, '«')\n            .replace(/»\\)|\\) \\)/g, '»')\n            // Fix \")digit)\" pattern to \"(digit)\"\n            .replace(/\\)([0-9\\u0660-\\u0669]+)\\)/g, '($1)')\n            // Fix \")digit(\" pattern to \"(digit)\"\n            .replace(/\\)([0-9\\u0660-\\u0669]+)\\(/g, '($1)')\n    );\n};\n\n/**\n * Fixes mismatched curly braces by converting incorrect bracket/brace combinations\n * to proper curly braces { }\n * @param text - Input text that may contain mismatched curly braces\n * @returns Text with corrected curly brace pairs\n */\nexport const fixCurlyBraces = (text: string) => {\n    // Process each mismatch type separately to avoid interference\n    let result = text;\n\n    // Fix ( content } to { content }\n    result = result.replace(/\\(([^(){}]+)\\}/g, '{$1}');\n\n    // Fix { content ) to { content }\n    return result.replace(/\\{([^(){}]+)\\)/g, '{$1}');\n};\n\n/**\n * Fixes mismatched quotation marks in Arabic text by converting various\n * incorrect bracket/quote combinations to proper Arabic quotation marks (« »)\n * @param text - Input text that may contain mismatched quotation marks\n * @returns Text with corrected Arabic quotation marks\n */\nexport const fixMismatchedQuotationMarks = (text: string) => {\n    return (\n        text\n            // Matches mismatched quotation marks: « followed by content and closed with )\n            .replace(/«([^»)]+)\\)/g, '«$1»')\n            // Fix reverse mismatched ( content » to « content »\n            .replace(/\\(([^()]+)»/g, '«$1»')\n            // Matches any unclosed « quotation marks at end of content\n            .replace(/«([^»]+)(?=\\s*$|$)/g, '«$1»')\n    );\n};\n\n/**\n * Formats a multiline string by joining sentences and maintaining footnotes on their own lines.\n * Footnotes are identified by Arabic and English numerals.\n * Example: 'Sentence one.\\n(1) A footnote.\\nSentence two.' remains the same, while regular sentences are joined.\n * @param {string} input - The input text containing sentences and footnotes.\n * @returns {string} - The formatted text.\n */\nexport const formatStringBySentence = (input: string) => {\n    const footnoteRegex = /^\\((?:\\d+|۱|۲|۳|۴|۵|۶|۷|۸|۹)\\)\\s/;\n    const sentences: string[] = [];\n    const lines = input.split('\\n');\n    let currentSentence = '';\n\n    lines.forEach((line) => {\n        const trimmedLine = line.trim();\n        const isFootnote = footnoteRegex.test(trimmedLine);\n        const isNumber = /^\\(\\d+\\/\\d+\\)/.test(trimmedLine);\n\n        if (isFootnote && !isNumber) {\n            if (currentSentence) {\n                sentences.push(currentSentence.trim());\n                currentSentence = '';\n            }\n            sentences.push(trimmedLine);\n        } else {\n            currentSentence += `${trimmedLine} `;\n            const lastChar = currentSentence.trim().slice(-1);\n            if (/[.!؟]/.test(lastChar)) {\n                sentences.push(currentSentence.trim());\n                currentSentence = '';\n            }\n        }\n    });\n\n    // Add any remaining text to the output\n    if (currentSentence) {\n        sentences.push(currentSentence.trim());\n    }\n\n    return sentences.join('\\n');\n};\n\n/**\n * Detects if text is entirely in uppercase letters\n * @param text - The text to check\n * @returns true if all alphabetic characters are uppercase, false otherwise\n */\nexport const isAllUppercase = (text: string) => {\n    // Remove non-letter characters (including numbers, punctuation, spaces)\n    // \\p{L} matches any Unicode letter character\n    const lettersOnly = text.replace(/[^\\p{L}]/gu, '');\n\n    // If there are no letter characters, return false\n    if (lettersOnly.length === 0) {\n        return false;\n    }\n\n    return lettersOnly === lettersOnly.toUpperCase();\n};\n\n/**\n * Removes unnecessary spaces around slashes in references.\n * Example: '127 / 11' becomes '127/11'.\n * @param {string} text - The input text containing references.\n * @returns {string} - The modified text with spaces removed around slashes.\n */\nexport const normalizeSlashInReferences = (text: string) => {\n    return text.replace(/(\\d+)\\s?\\/\\s?(\\d+)/g, '$1/$2');\n};\n\n/**\n * Reduces multiple spaces or tabs to a single space.\n * Example: 'This   is a   text' becomes 'This is a text'.\n * @param {string} text - The input text containing extra spaces.\n * @returns {string} - The modified text with reduced spaces.\n */\nexport const normalizeSpaces = (text: string) => {\n    return text.replace(/[ \\t]+/g, ' ');\n};\n\n/**\n * Removes redundant punctuation marks that follow Arabic question marks or exclamation marks.\n * This function cleans up text by removing periods (.) or Arabic commas (،) that immediately\n * follow Arabic question marks (؟) or exclamation marks (!), as they are considered redundant\n * in proper Arabic punctuation.\n *\n * @param text - The Arabic text to clean up\n * @returns The text with redundant punctuation removed\n *\n * @example\n * ```typescript\n * removeRedundantPunctuation('كيف حالك؟.') // Returns: 'كيف حالك؟'\n * removeRedundantPunctuation('ممتاز!،') // Returns: 'ممتاز!'\n * removeRedundantPunctuation('هذا جيد.') // Returns: 'هذا جيد.' (unchanged)\n * ```\n */\nexport const removeRedundantPunctuation = (text: string) => {\n    return text.replace(/([؟!])[.،]/g, '$1');\n};\n\n/**\n * Removes spaces inside brackets, parentheses, or square brackets.\n * Example: '( a b )' becomes '(a b)'.\n * @param {string} text - The input text with spaces inside brackets.\n * @returns {string} - The modified text with spaces removed inside brackets.\n */\nexport const removeSpaceInsideBrackets = (text: string) => {\n    return text.replace(/([[(])\\s*(.*?)\\s*([\\])])/g, '$1$2$3');\n};\n\n/**\n * Replaces double parentheses single a single arrow variation.\n * Example: '((text))' becomes '«text»'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed brackets.\n */\nexport const replaceDoubleBracketsWithArrows = (text: string) => {\n    return text.replace(/\\(\\(\\s?/g, '«').replace(/\\s?\\)\\)/g, '»');\n};\n\n/**\n * Removes bold styling from text by normalizing the string and removing stylistic characters.\n * @param {string} text - The input text containing bold characters.\n * @returns {string} - The modified text with bold styling removed.\n */\nexport const stripBoldStyling = (text: string) => {\n    // Normalize the string to NFKD form\n    const normalizedString = text.normalize('NFKD');\n\n    // Remove combining marks (diacritics) and stylistic characters from the string\n    return normalizedString.replace(/[\\u0300-\\u036f]/g, '').trim();\n};\n\n/**\n * Removes italicized characters by replacing italic Unicode characters with their normal counterparts.\n * Example: '𝘼𝘽𝘾' becomes 'ABC'.\n * @param {string} text - The input text containing italicized characters.\n * @returns {string} - The modified text with italics removed.\n */\nexport const stripItalicsStyling = (text: string) => {\n    const italicMap: Record<string, string> = {\n        '\\uD835\\uDC4E': 'I',\n        '\\uD835\\uDC68': 'g',\n        '\\u{1D63C}': '!',\n        '\\uD835\\uDC4F': 'J',\n        '\\uD835\\uDC69': 'h',\n        '\\u{1D63D}': '?',\n        '\\uD835\\uDC50': 'K',\n        '\\uD835\\uDC6A': 'i',\n        '\\uD835\\uDC51': 'L',\n        '\\uD835\\uDC6B': 'j',\n        '\\u{1D63F}': ',',\n        '\\uD835\\uDC52': 'M',\n        '\\uD835\\uDC6C': 'k',\n        '\\u{1D640}': '.',\n        '\\uD835\\uDC53': 'N',\n        '\\uD835\\uDC6D': 'l',\n        '\\uD835\\uDC54': 'O',\n        '\\uD835\\uDC6E': 'm',\n        '\\uD835\\uDC6F': 'n',\n        '\\uD835\\uDC56': 'Q',\n        '\\uD835\\uDC70': 'o',\n        '\\uD835\\uDC57': 'R',\n        '\\uD835\\uDC71': 'p',\n        '\\uD835\\uDC58': 'S',\n        '\\uD835\\uDC72': 'q',\n        '\\uD835\\uDC59': 'T',\n        '\\uD835\\uDC73': 'r',\n        '\\u{1D647}': '-',\n        '\\uD835\\uDC5A': 'U',\n        '\\uD835\\uDC74': 's',\n        '\\uD835\\uDC5B': 'V',\n        '\\uD835\\uDC75': 't',\n        '\\uD835\\uDC5C': 'W',\n        '\\uD835\\uDC76': 'u',\n        '\\uD835\\uDC5D': 'X',\n        '\\uD835\\uDC77': 'v',\n        '\\uD835\\uDC5E': 'Y',\n        '\\uD835\\uDC78': 'w',\n        '\\uD835\\uDC5F': 'Z',\n        '\\uD835\\uDC79': 'x',\n        '\\uD835\\uDC46': 'A',\n        '\\uD835\\uDC7A': 'y',\n        '\\uD835\\uDC47': 'B',\n        '\\uD835\\uDC7B': 'z',\n        '\\uD835\\uDC62': 'a',\n        '\\uD835\\uDC48': 'C',\n        '\\uD835\\uDC63': 'b',\n        '\\uD835\\uDC49': 'D',\n        '\\uD835\\uDC64': 'c',\n        '\\uD835\\uDC4A': 'E',\n        '\\uD835\\uDC65': 'd',\n        '\\uD835\\uDC4B': 'F',\n        '\\uD835\\uDC66': 'e',\n        '\\uD835\\uDC4C': 'G',\n        '\\uD835\\uDC67': 'f',\n        '\\uD835\\uDC4D': 'H',\n        '\\uD835\\uDC55': 'P',\n    };\n\n    return text.replace(/[\\uD835\\uDC62-\\uD835\\uDC7B\\uD835\\uDC46-\\uD835\\uDC5F\\u{1D63C}-\\u{1D647}]/gu, (match) => {\n        return italicMap[match] || match;\n    });\n};\n\n/**\n * Removes all bold and italic styling from the input text.\n * @param {string} text - The input text to remove styling from.\n * @returns {string} - The modified text with all styling removed.\n */\nexport const stripStyling = (text: string) => {\n    return stripItalicsStyling(stripBoldStyling(text));\n};\n\n/**\n * Converts a string to title case (first letter of each word capitalized)\n * @param str - The input string to convert\n * @returns String with each word's first letter capitalized\n */\nexport const toTitleCase = (str: string) => {\n    return str\n        .toLowerCase()\n        .split(' ')\n        .map((word) => {\n            if (word.length === 0) return word;\n            // Find the first Unicode letter in the chunk\n            const match = word.match(/\\p{L}/u);\n            if (!match || match.index === undefined) return word;\n            const i = match.index;\n            return word.slice(0, i) + word.charAt(i).toUpperCase() + word.slice(i + 1);\n        })\n        .join(' ');\n};\n\n/**\n * Removes unnecessary spaces inside quotes.\n * Example: '“ Text ”' becomes '“Text”'.\n * @param {string} text - The input text with spaces inside quotes.\n * @returns {string} - The modified text with spaces removed inside quotes.\n */\nexport const trimSpaceInsideQuotes = (text: string) => {\n    return text.replace(/([“”\"]|«) *(.*?) *([“”\"]|»)/g, '$1$2$3');\n};\n","/**\n * Converts a string that resembles JSON but with numeric keys and single-quoted values\n * into valid JSON format. This function replaces numeric keys with quoted numeric keys\n * and ensures all values are double-quoted as required by JSON.\n *\n * @param {string} str - The input string that needs to be fixed into valid JSON.\n * @returns {string} - A valid JSON string.\n *\n * @example\n * const result = normalizeJsonSyntax(\"{10: 'abc', 20: 'def'}\");\n * console.log(result); // '{\"10\": \"abc\", \"20\": \"def\"}'\n */\nexport const normalizeJsonSyntax = (str: string) => {\n    let input = str.replace(/(\\b\\d+\\b)(?=:)/g, '\"$1\"');\n    input = input.replace(/:\\s*'([^']+)'/g, ': \"$1\"');\n    input = input.replace(/:\\s*\"([^\"]+)\"/g, ': \"$1\"');\n\n    return JSON.stringify(JSON.parse(input));\n};\n\n/**\n * Checks if a given string resembles a JSON object with numeric or quoted keys and values\n * that are single or double quoted. This is useful for detecting malformed JSON-like\n * structures that can be fixed by the `normalizeJsonSyntax` function.\n *\n * @param {string} str - The input string to check.\n * @returns {boolean} - Returns true if the string is JSON-like, false otherwise.\n *\n * @example\n * const result = isJsonStructureValid(\"{10: 'abc', 'key': 'value'}\");\n * console.log(result); // true\n */\nexport const isJsonStructureValid = (str: string) => {\n    // Checks for a pattern with numeric keys or quoted keys and values in quotes\n    const jsonLikePattern =\n        /^{(\\s*(\\d+|'[^']*'|\"[^\"]*\")\\s*:\\s*('|\")[^'\"]*\\3\\s*,)*(?:\\s*(\\d+|'[^']*'|\"[^\"]*\")\\s*:\\s*('|\")[^'\"]*\\5\\s*)}$/;\n    return jsonLikePattern.test(str.trim());\n};\n\n/**\n * Splits a string by spaces and quoted substrings.\n *\n * This function takes an input string and splits it into parts where substrings\n * enclosed in double quotes are treated as a single part. Other substrings\n * separated by spaces are split normally.\n *\n * @param {string} query - The input string to be split.\n * @returns {string[]} An array of strings, with quoted substrings kept intact.\n *\n * @example\n * const result = splitByQuotes('\"This is\" \"a part of the\" \"string and\"');\n * console.log(result); // [\"This is\", \"a part of the\", \"string and\"]\n */\nexport const splitByQuotes = (query: string): string[] => {\n    const regex = /(?:[^\\s\"]+|\"(.*?)\")+/g;\n    return (query.match(regex) || []).map((s: string) => (s.startsWith('\"') ? s.slice(1, -1) : s));\n};\n\n/**\n * Checks if all double quotes in a string are balanced (even count).\n * A string has balanced quotes if every opening quote has a corresponding closing quote.\n *\n * @param str - The string to check for balanced quotes\n * @returns True if quotes are balanced (even count), false otherwise\n *\n * @example\n * ```typescript\n * areQuotesBalanced('Hello \"world\"') // Returns: true\n * areQuotesBalanced('Hello \"world') // Returns: false\n * areQuotesBalanced('No quotes') // Returns: true\n * ```\n */\nconst areQuotesBalanced = (str: string) => {\n    let quoteCount = 0;\n    for (const char of str) {\n        if (char === '\"') {\n            quoteCount++;\n        }\n    }\n    return quoteCount % 2 === 0;\n};\n\nconst brackets = { '(': ')', '[': ']', '{': '}' };\nconst openBrackets = new Set(['(', '[', '{']);\nconst closeBrackets = new Set([')', ']', '}']);\n\n/**\n * Checks if all brackets in a string are properly balanced and matched.\n * This function validates that every opening bracket has a corresponding closing bracket\n * in the correct order and of the matching type.\n *\n * Supported bracket types: parentheses (), square brackets [], curly braces {}\n *\n * @param str - The string to check for balanced brackets\n * @returns True if all brackets are properly balanced and matched, false otherwise\n *\n * @example\n * ```typescript\n * areBracketsBalanced('(hello [world])') // Returns: true\n * areBracketsBalanced('(hello [world)') // Returns: false (mismatched)\n * areBracketsBalanced('((hello))') // Returns: true\n * areBracketsBalanced('(hello') // Returns: false (unclosed)\n * ```\n */\n\nconst areBracketsBalanced = (str: string) => {\n    const stack: string[] = [];\n\n    for (const char of str) {\n        if (openBrackets.has(char)) {\n            stack.push(char);\n        } else if (closeBrackets.has(char)) {\n            const lastOpen = stack.pop();\n            if (!lastOpen || brackets[lastOpen as keyof typeof brackets] !== char) {\n                return false;\n            }\n        }\n    }\n\n    return stack.length === 0;\n};\n\n/**\n * Checks if both quotes and brackets are balanced in a string.\n * This function combines quote balance checking and bracket balance checking\n * to ensure the entire string has properly balanced punctuation.\n *\n * A string is considered balanced when:\n * - All double quotes have matching pairs (even count)\n * - All brackets (parentheses, square brackets, curly braces) are properly matched and nested\n *\n * @param str - The string to check for balanced quotes and brackets\n * @returns True if both quotes and brackets are balanced, false otherwise\n *\n * @example\n * ```typescript\n * isBalanced('He said \"Hello (world)!\"') // Returns: true\n * isBalanced('He said \"Hello (world!\"') // Returns: false (unbalanced quote)\n * isBalanced('He said \"Hello (world)\"') // Returns: false (unbalanced quote)\n * isBalanced('Hello (world) [test]') // Returns: true\n * ```\n */\nexport const isBalanced = (str: string) => {\n    return areQuotesBalanced(str) && areBracketsBalanced(str);\n};\n\n/**\n * Parses page input string into array of page numbers, supporting ranges and lists\n * @param pageInput - Page specification string (e.g., \"1-5\" or \"1,3,5\")\n * @returns Array of page numbers\n * @throws Error when start page exceeds end page in range\n */\nexport const parsePageRanges = (pageInput: string): number[] => {\n    if (pageInput.includes('-')) {\n        const [start, end] = pageInput.split('-').map(Number);\n\n        if (start > end) {\n            throw new Error('Start page cannot be greater than end page');\n        }\n\n        return Array.from({ length: end - start + 1 }, (_, i) => start + i);\n    } else {\n        return pageInput.split(',').map(Number);\n    }\n};\n","import { escapeRegex } from './cleaning';\n\n/**\n * Removes various symbols, part references, and numerical markers from the text.\n * Example: '(1) (2/3)' becomes ''.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with symbols and part references removed.\n */\nexport const cleanSymbolsAndPartReferences = (text: string) => {\n    return text.replace(\n        / *\\(?:\\d+(?:\\/\\d+){0,2}\\)? *| *\\[\\d+(?:\\/\\d+)?\\] *| *«\\d+» *|\\d+\\/\\d+(?:\\/\\d+)?|[،§{}۝؍‎﴿﴾<>;_؟»«:!،؛[\\]…ـ¬.\\\\/*()\"]/g,\n        ' ',\n    );\n};\n\n/**\n * Removes trailing page numbers formatted as '-[46]-' from the text.\n * Example: 'This is some -[46]- text' becomes 'This is some text'.\n * @param {string} text - The input text with trailing page numbers.\n * @returns {string} - The modified text with page numbers removed.\n */\nexport const cleanTrailingPageNumbers = (text: string) => {\n    return text.replace(/-\\[\\d+\\]-/g, '');\n};\n\n/**\n * Replaces consecutive line breaks and whitespace characters with a single space.\n * Example: 'a\\nb' becomes 'a b'.\n * @param {string} text - The input text containing line breaks or multiple spaces.\n * @returns {string} - The modified text with spaces.\n */\nexport const replaceLineBreaksWithSpaces = (text: string) => {\n    return text.replace(/\\s+/g, ' ');\n};\n\n/**\n * Removes all numeric digits from the text.\n * Example: 'abc123' becomes 'abc'.\n * @param {string} text - The input text containing digits.\n * @returns {string} - The modified text with digits removed.\n */\nexport const stripAllDigits = (text: string) => {\n    return text.replace(/[0-9]/g, '');\n};\n\n/**\n * Removes death year references like \"(d. 390H)\" and \"[d. 100h]\" from the text.\n * Example: 'Sufyān ibn ‘Uyaynah (d. 198h)' becomes 'Sufyān ibn ‘Uyaynah'.\n * @param {string} text - The input text containing death year references.\n * @returns {string} - The modified text with death years removed.\n */\nexport const removeDeathYear = (text: string) => {\n    return text.replace(/\\[(d)\\.\\s*\\d{1,4}[hH]\\]\\s*|\\((d)\\.\\s*\\d{1,4}[hH]\\)\\s*/g, '');\n};\n\n/**\n * Removes numeric digits and dashes from the text.\n * Example: 'ABC 123-Xyz' becomes 'ABC Xyz'.\n * @param {string} text - The input text containing digits and dashes.\n * @returns {string} - The modified text with numbers and dashes removed.\n */\nexport const removeNumbersAndDashes = (text: string) => {\n    return text.replace(/[\\d-]/g, '');\n};\n\n/**\n * Removes single digit references like (1), «2», [3] from the text.\n * Example: 'Ref (1), Ref «2», Ref [3]' becomes 'Ref , Ref , Ref '.\n * @param {string} text - The input text containing single digit references.\n * @returns {string} - The modified text with single digit references removed.\n */\nexport const removeSingleDigitReferences = (text: string) => {\n    return text.replace(/\\(\\d{1}\\)|\\[\\d{1}\\]|«\\d»/g, '');\n};\n\n/**\n * Removes URLs from the text.\n * Example: 'Visit https://example.com' becomes 'Visit '.\n * @param {string} text - The input text containing URLs.\n * @returns {string} - The modified text with URLs removed.\n */\nexport const removeUrls = (text: string) => {\n    return text.replace(\n        /https?:\\/\\/(www\\.)?[-a-zA-Z0-9@:%._+~#=]{1,256}\\.[a-zA-Z0-9()]{1,6}\\b([-a-zA-Z0-9()@:%_+.~#?&//=]*)/g,\n        '',\n    );\n};\n\n/**\n * Removes common Markdown formatting syntax from text\n * @param text - The input text containing Markdown formatting\n * @returns Text with Markdown formatting removed (bold, italics, headers, lists, backticks)\n */\nexport const removeMarkdownFormatting = (text: string) => {\n    return (\n        text\n            // Remove bold first (**text**) - must come before italics\n            .replace(/\\*\\*([^*]+)\\*\\*/g, '$1')\n            // Remove bold with underscores (__text__)\n            .replace(/__([^_]+)__/g, '$1')\n            // Remove italics (*text*)\n            .replace(/\\*([^*]+)\\*/g, '$1')\n            // Remove italics with underscores (_text_)\n            .replace(/_([^_]+)_/g, '$1')\n            // Remove strikethrough (~~text~~)\n            .replace(/~~([^~]+)~~/g, '$1')\n            // Remove blockquotes\n            .replace(/^\\s*>\\s?/gm, '')\n            // Remove images ![alt](url)\n            .replace(/!\\[[^\\]]*]\\([^)]*\\)/g, '')\n            // Convert links [text](url) -> text\n            .replace(/\\[([^\\]]+)]\\([^)]*\\)/g, '$1')\n            // Remove headers (# ## ### etc.)\n            .replace(/^#+\\s*/gm, '')\n            // Remove unordered list markers (- * +)\n            .replace(/^\\s*[-*+]\\s+/gm, '')\n            // Remove ordered list markers (1. 2. etc.)\n            .replace(/^\\s*\\d+\\.\\s+/gm, '')\n            // Remove backticks\n            .replace(/`/gm, '')\n    );\n};\n\n/**\n * Truncates a string to a specified length, adding an ellipsis if truncated.\n *\n * @param val - The string to truncate\n * @param n - Maximum length of the string (default: 150)\n * @returns The truncated string with ellipsis if needed, otherwise the original string\n *\n * @example\n * ```javascript\n * truncate('The quick brown fox jumps over the lazy dog', 20);\n * // Output: 'The quick brown fox…'\n *\n * truncate('Short text', 50);\n * // Output: 'Short text'\n * ```\n */\nexport const truncate = (val: string, n = 150): string => (val.length > n ? `${val.substring(0, n - 1)}…` : val);\n\n/**\n * Truncates a string from the middle, preserving both the beginning and end portions.\n *\n * @param text - The string to truncate\n * @param maxLength - Maximum length of the resulting string (default: 50)\n * @param endLength - Number of characters to preserve at the end (default: 1/3 of maxLength, minimum 3)\n * @returns The truncated string with ellipsis in the middle if needed, otherwise the original string\n *\n * @example\n * ```javascript\n * truncateMiddle('The quick brown fox jumps right over the lazy dog', 20);\n * // Output: 'The quick bro…zy dog'\n *\n * truncateMiddle('The quick brown fox jumps right over the lazy dog', 25, 8);\n * // Output: 'The quick brown …lazy dog'\n *\n * truncateMiddle('Short text', 50);\n * // Output: 'Short text'\n * ```\n */\nexport const truncateMiddle = (text: string, maxLength: number = 50, endLength?: number) => {\n    if (text.length <= maxLength) {\n        return text;\n    }\n\n    // Default end length is roughly 1/3 of max length, minimum 3 characters\n    const defaultEndLength = Math.max(3, Math.floor(maxLength / 3));\n    const actualEndLength = endLength ?? defaultEndLength;\n\n    // Reserve space for the ellipsis character (1 char)\n    const availableLength = maxLength - 1;\n\n    // Calculate start length (remaining space after end portion)\n    const startLength = availableLength - actualEndLength;\n\n    // Ensure we have at least some characters at the start\n    if (startLength < 1) {\n        // If we can't fit both start and end, just truncate normally\n        return `${text.substring(0, maxLength - 1)}…`;\n    }\n\n    const startPortion = text.substring(0, startLength);\n    const endPortion = text.substring(text.length - actualEndLength);\n\n    return `${startPortion}…${endPortion}`;\n};\n\n/**\n * Unescapes backslash-escaped spaces and trims whitespace from both ends.\n * Commonly used to clean file paths that have been escaped when pasted into terminals.\n *\n * @param input - The string to unescape and clean\n * @returns The cleaned string with escaped spaces converted to regular spaces and trimmed\n *\n * @example\n * ```javascript\n * unescapeSpaces('My\\\\ Folder\\\\ Name');\n * // Output: 'My Folder Name'\n *\n * unescapeSpaces('  /path/to/My\\\\ Document.txt  ');\n * // Output: '/path/to/My Document.txt'\n *\n * unescapeSpaces('regular text');\n * // Output: 'regular text'\n * ```\n */\nexport const unescapeSpaces = (input: string) => input.replace(/\\\\ /g, ' ').trim();\n\n/**\n * Arabic diacritics (Tashkeel/Harakat).\n */\nconst DIACRITICS_CLASS = '[\\u064B\\u064C\\u064D\\u064E\\u064F\\u0650\\u0651\\u0652]';\n\n/**\n * Groups of equivalent Arabic characters — any character in a group should match\n * any other character in the same group.\n */\nconst EQUIV_GROUPS: string[][] = [\n    ['\\u0627', '\\u0622', '\\u0623', '\\u0625'], // ا, آ, أ, إ\n    ['\\u0629', '\\u0647'], // ة <-> ه\n    ['\\u0649', '\\u064A'], // ى <-> ي\n];\n\n/**\n * Return an escaped character class representing all equivalents for the given character.\n *\n * If the character belongs to one of the predefined equivalence groups (e.g. ا/آ/أ/إ),\n * the returned class will match any member of that group. Otherwise, the original\n * character is simply escaped for safe inclusion in a regular expression.\n *\n * @param ch - A single character to expand into its equivalence class\n * @returns A RegExp-safe string representing the character (and its equivalents when applicable)\n */\nconst getEquivClass = (ch: string): string => {\n    for (const group of EQUIV_GROUPS) {\n        if (group.includes(ch)) {\n            // join the group's members into a character class\n            return `[${group.map((c) => escapeRegex(c)).join('')}]`;\n        }\n    }\n    // not in equivalence groups -> return escaped character\n    return escapeRegex(ch);\n};\n\n/** Small safe normalization: NFC, remove ZWJ/ZWNJ, collapse spaces. */\nconst normalizeArabicLight = (str: string) => {\n    return str\n        .normalize('NFC')\n        .replace(/[\\u200C\\u200D]/g, '') // remove ZWJ/ZWNJ\n        .replace(/\\s+/g, ' ')\n        .trim();\n};\n\n/**\n * Creates a diacritic-insensitive regex pattern for Arabic text matching.\n * Normalizes text, handles character equivalences (ا/آ/أ/إ, ة/ه, ى/ي),\n * and makes each character tolerant of Arabic diacritics (Tashkeel/Harakat)\n * @param text - Input Arabic text to make diacritic-insensitive\n * @returns Regex pattern string that matches the text with or without diacritics and character variants\n */\nexport const makeDiacriticInsensitive = (text: string) => {\n    const diacriticsMatcher = `${DIACRITICS_CLASS}*`;\n    const norm = normalizeArabicLight(text);\n    // Use Array.from to iterate grapheme-safe over the string (works fine for Arabic letters)\n    return Array.from(norm)\n        .map((ch) => getEquivClass(ch) + diacriticsMatcher)\n        .join('');\n};\n","import { normalizeSpaces } from './formatting';\n\n/**\n * Replaces common Arabic prefixes (like 'Al-', 'Ar-', 'Ash-', etc.) with 'al-' in the text.\n * Handles different variations of prefixes such as Ash- and Al- but not when the second word\n * does not start with 'S'.\n * Example: 'Ash-Shafiee' becomes 'al-Shafiee'.\n *\n * @param {string} text - The input text containing Arabic prefixes.\n * @returns {string} - The modified text with standardized 'al-' prefixes.\n */\nexport const normalizeArabicPrefixesToAl = (text: string) => {\n    return text\n        .replace(/(\\b|\\W)(Al |Al-|Ar-|As-|Adh-|Ad-|Ats-|Ath |Ath-|Az |Az-|az-|adh-|as-|ar-)/g, '$1al-')\n        .replace(/(\\b|\\W)(Ash-S|ash-S)/g, '$1al-S')\n        .replace(/al- (.+?)\\b/g, 'al-$1');\n};\n\n/**\n * Removes double occurrences of Arabic apostrophes such as ʿʿ or ʾʾ in the text.\n * Example: 'ʿulamāʾʾ' becomes 'ʿulamāʾ'.\n *\n * @param {string} text - The input text containing double apostrophes.\n * @returns {string} - The modified text with condensed apostrophes.\n */\nexport const normalizeDoubleApostrophes = (text: string) => {\n    return text.replace(/ʿʿ/g, 'ʿ').replace(/ʾʾ/g, 'ʾ');\n};\n\n/**\n * Replaces common salutations such as \"sallahu alayhi wasallam\" with \"ﷺ\" in the text.\n * It also handles variations of the salutation phrase, including 'peace and blessings be upon him'.\n * Example: 'Then Muḥammad (sallahu alayhi wasallam)' becomes 'Then Muḥammad ﷺ'.\n *\n * @param {string} text - The input text containing salutations.\n * @returns {string} - The modified text with salutations replaced.\n */\nexport const replaceSalutationsWithSymbol = (text: string) => {\n    return text\n        .replace(\n            /\\(peace be upon him\\)|(Messenger of (Allah|Allāh)|Messenger|Prophet|Mu[hḥ]ammad) *\\((s[^)]*m|peace[^)]*him|May[^)]*him|may[^)]*him)\\)*/gi,\n            '$1 ﷺ',\n        )\n        .replace(/,\\s*ﷺ\\s*,/g, ' ﷺ');\n};\n\n/**\n * Normalizes the text by removing diacritics, apostrophes, and dashes.\n * Example: 'Al-Jadwal' becomes 'AlJadwal'.\n *\n * @param {string} input - The input text to normalize.\n * @returns {string} - The normalized text.\n */\nexport const normalize = (input: string) => {\n    return input\n        .normalize('NFKD')\n        .replace(/[\\u0300-\\u036f]/g, '')\n        .replace(/`|ʾ|ʿ|-/g, '');\n};\n\n/**\n * Strips common Arabic prefixes like 'al-', 'bi-', 'fī', 'wa-', etc. from the beginning of words.\n * Example: 'al-Bukhari' becomes 'Bukhari'.\n *\n * @param {string} text - The input text containing Arabic prefixes.\n * @returns {string} - The modified text with prefixes stripped.\n */\nexport const removeArabicPrefixes = (text: string) => {\n    return normalizeSpaces(text.replace(/(\\bal-|\\bli-|\\bbi-|\\bfī|\\bwa[-\\s]+|\\bl-|\\bliʿl|\\Bʿalá|\\Bʿan|\\bb\\.)/gi, ''));\n};\n\n/**\n * Simplifies English transliterations by removing diacritics, apostrophes, and common prefixes.\n * Example: 'Al-Jadwal' becomes 'Jadwal', and 'āḍġḥīṣṭū' becomes 'adghistu'.\n *\n * @param {string} text - The input text to simplify.\n * @returns {string} - The simplified text.\n */\nexport const normalizeTransliteratedEnglish = (text: string) => normalize(removeArabicPrefixes(text));\n\n/**\n * Extracts the initials from the input string, typically used for names or titles.\n * Example: 'Nayl al-Awtar' becomes 'NA'.\n *\n * @param {string} text - The input text to extract initials from.\n * @returns {string} - The extracted initials.\n */\nexport const extractInitials = (fullName: string) => {\n    const initials = normalizeTransliteratedEnglish(fullName)\n        .trim()\n        .split(/[ -]/)\n        .slice(0, 2)\n        .map((word) => {\n            return word.charAt(0).toUpperCase();\n        })\n        .join('');\n    return initials;\n};\n"],"mappings":"AACA,MAAa,EAAgC,WCwBhC,EAAyB,GAC3B,SACH,EAAO,QAAQ,mBAAqB,IAAO,EAAE,WAAW,EAAE,CAAG,MAAQ,UAAU,CAAC,CAChF,GACH,CAUQ,EAAiC,GACnC,EAAK,QAAQ,qCAAsC,GAAG,CASpD,EAA8B,GAChC,EAAK,QAAQ,KAAM,IAAI,CAAC,QAAQ,KAAM,IAAI,CASxC,EAAkB,GAAiB,CAC5C,GAAI,CAAC,EACD,MAAO,GAGX,IAAM,EAAuB,uEAEvB,EAAkB,mCAElB,EAAsB,sCACtB,EAAU,EAAK,QAAQ,EAAiB,GAAG,CAC3C,EAAgB,EAAQ,MAAM,EAAqB,EAAI,EAAE,CACzD,EAAe,EAAQ,MAAM,EAAoB,EAAI,EAAE,CAC7D,OAAO,EAAa,SAAW,EAAI,EAAI,EAAc,OAAS,EAAa,QAoBlE,EAAuB,GAAiB,CACjD,IAAK,IAAI,EAAI,EAAK,OAAS,EAAG,GAAK,EAAG,IAClC,GAAI,EAA8B,KAAK,EAAK,GAAG,CAC3C,OAAO,EAIf,MAAO,IAUE,EAAkB,GACpB,EAAK,QAAQ,OAAQ,KAAK,CASxB,EAAuC,GACzC,EAAK,QAAQ,2BAA4B,QAAQ,CAS/C,EAA4B,GAC9B,EACF,QAAQ,iEAAkE,IAAI,CAC9E,QAAQ,8DAA+D,IAAI,CASvE,EAAuB,GACzB,EAAK,QAAQ,2CAA4C,GAAG,CAS1D,EAA+B,GACjC,EAAK,QAAQ,6BAA8B,IAAI,CAS7C,EAAuC,GACzC,EACF,QAAQ,UAAW,IAAI,CACvB,QAAQ,oBAAqB,IAAI,CACjC,QAAQ,QAAS,IAAI,CC1JjB,EAAe,GAAsB,EAAE,QAAQ,sBAAuB,OAAO,CA+D7E,GAAiC,EAAgB,EAAyB,EAAE,GAAa,CAClG,GAAM,CACF,eAAe,CAAE,KAAM,GAAM,aAAc,GAAM,eAAgB,GAAM,CACvE,eAAe,GACf,mBAAmB,GACnB,iBAAiB,GACjB,QAAQ,KACR,EAGJ,GAAI,EAAO,OAAS,IAChB,MAAU,MAAM,iDAAiD,CAGrE,IAAM,EAAa,GAAuB,CACtC,OAAQ,EAAR,CACI,IAAK,IACL,IAAK,IACL,IAAK,IACL,IAAK,IACD,OAAO,EAAa,KAAO,SAAW,IAC1C,IAAK,IACL,IAAK,IACD,OAAO,EAAa,aAAe,OAAS,EAAY,EAAG,CAC/D,IAAK,IACL,IAAK,IACD,OAAO,EAAa,eAAiB,OAAS,EAAY,EAAG,CACjE,QACI,OAAO,EAAY,EAAG,GAI5B,EAAQ,GAAG,EAAmB,0DAAyB,KAAK,EAAe,WAAsB,KAEnG,EAAU,GACd,IAAK,IAAM,KAAM,MAAM,KAAK,EAAO,CAC3B,KAAK,KAAK,EAAG,CACb,GAAW,EAAiB,OAAS,OAErC,GAAW,GAAG,EAAU,EAAG,GAAG,IAItC,OAAO,IAAI,OAAO,EAAS,EAAM,EAGxB,EAAiB,GAAoB,EAAQ,QAAQ,WAAY,GAAG,CClHpE,EAAoC,GAKvB,EAAK,QAHP,YAG4B;EAAO,CAAC,QAAQ,SAAU;EAAK,CAAC,MAAM,CAW7E,GAAqC,GACvC,EACF,QAAQ,iDAAkD,QAAQ,CAClE,QAAQ,mCAAoC,OAAO,CACnD,QAAQ,uDAAwD,SAAS,CACzE,QAAQ,yEAA0E,KAAK,CAUnF,GAAoB,GACtB,EACF,QAAQ,QAAS,IAAI,CACrB,QAAQ,aAAc,OAAO,CAC7B,QAAQ,MAAO,IAAI,CASf,GAAwB,GAC1B,EAAK,QAAQ,UAAW;EAAK,CAS3B,EAAmB,GACrB,EAAK,QAAQ,YAAa,GAAG,CAQ3B,EAAuB,GACzB,gBAAgB,KAAK,EAAK,CAQxB,EAAqB,GAChB,kEACD,KAAK,EAAK,CASd,EAA2B,GAC7B,EAAK,QAAQ,mBAAoB,KAAK,CASpC,EAAqB,GACvB,EAAK,QAAQ,YAAa,IAAI,CAS5B,EAAkB,GACpB,EAAK,QAAQ,eAAgB,IAAI,CAS/B,EAAkB,GACpB,EAAK,QAAQ,SAAU,IAAI,CASzB,EAAoB,GACtB,EAAK,QAAQ,UAAW,IAAI,CAS1B,EAAiC,GACnC,EAAK,QAAQ,eAAgB;;EAAO,CASlC,EAAiC,GACnC,EAAK,QAAQ,eAAgB;EAAK,CAShC,EAAmB,GACrB,EAAK,QAAQ,UAAW,IAAI,CAS1B,EAAuB,GACzB,EAAK,QAAQ,SAAU,IAAI,CAAC,QAAQ,MAAO,IAAI,CAS7C,EAA0B,GAC5B,EAAK,QAAQ,2BAA4B,OAAO,CAS9C,EAA6B,GAC/B,EAAK,QAAQ,qBAAsB,QAAQ,CASzC,EAA2B,GAC7B,EAAK,QAAQ,mBAAoB,QAAQ,CASvC,EAAmB,GAExB,EACK,QAAQ,aAAc,IAAI,CAC1B,QAAQ,aAAc,IAAI,CAE1B,QAAQ,6BAA8B,OAAO,CAE7C,QAAQ,6BAA8B,OAAO,CAU7C,EAAkB,GAAiB,CAE5C,IAAI,EAAS,EAMb,MAHA,GAAS,EAAO,QAAQ,kBAAmB,OAAO,CAG3C,EAAO,QAAQ,kBAAmB,OAAO,EASvC,EAA+B,GAEpC,EAEK,QAAQ,eAAgB,OAAO,CAE/B,QAAQ,eAAgB,OAAO,CAE/B,QAAQ,sBAAuB,OAAO,CAWtC,EAA0B,GAAkB,CACrD,IAAM,EAAgB,mCAChBC,EAAsB,EAAE,CACxB,EAAQ,EAAM,MAAM;EAAK,CAC3B,EAAkB,GA4BtB,OA1BA,EAAM,QAAS,GAAS,CACpB,IAAM,EAAc,EAAK,MAAM,CACzB,EAAa,EAAc,KAAK,EAAY,CAC5C,EAAW,gBAAgB,KAAK,EAAY,CAElD,GAAI,GAAc,CAAC,EACf,AAEI,KADA,EAAU,KAAK,EAAgB,MAAM,CAAC,CACpB,IAEtB,EAAU,KAAK,EAAY,KACxB,CACH,GAAmB,GAAG,EAAY,GAClC,IAAM,EAAW,EAAgB,MAAM,CAAC,MAAM,GAAG,CAC7C,QAAQ,KAAK,EAAS,GACtB,EAAU,KAAK,EAAgB,MAAM,CAAC,CACtC,EAAkB,MAG5B,CAGE,GACA,EAAU,KAAK,EAAgB,MAAM,CAAC,CAGnC,EAAU,KAAK;EAAK,EAQlB,EAAkB,GAAiB,CAG5C,IAAM,EAAc,EAAK,QAAQ,aAAc,GAAG,CAOlD,OAJI,EAAY,SAAW,EAChB,GAGJ,IAAgB,EAAY,aAAa,EASvC,EAA8B,GAChC,EAAK,QAAQ,sBAAuB,QAAQ,CAS1C,EAAmB,GACrB,EAAK,QAAQ,UAAW,IAAI,CAmB1B,EAA8B,GAChC,EAAK,QAAQ,cAAe,KAAK,CAS/B,EAA6B,GAC/B,EAAK,QAAQ,4BAA6B,SAAS,CASjD,EAAmC,GACrC,EAAK,QAAQ,WAAY,IAAI,CAAC,QAAQ,WAAY,IAAI,CAQpD,EAAoB,GAEJ,EAAK,UAAU,OAAO,CAGvB,QAAQ,mBAAoB,GAAG,CAAC,MAAM,CASrD,EAAuB,GAAiB,CACjD,IAAMC,EAAoC,CACtC,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,KAAgB,IACnB,CAED,OAAO,EAAK,QAAQ,4EAA8E,GACvF,EAAU,IAAU,EAC7B,EAQO,EAAgB,GAClB,EAAoB,EAAiB,EAAK,CAAC,CAQzC,EAAe,GACjB,EACF,aAAa,CACb,MAAM,IAAI,CACV,IAAK,GAAS,CACX,GAAI,EAAK,SAAW,EAAG,OAAO,EAE9B,IAAM,EAAQ,EAAK,MAAM,SAAS,CAClC,GAAI,CAAC,GAAS,EAAM,QAAU,IAAA,GAAW,OAAO,EAChD,IAAM,EAAI,EAAM,MAChB,OAAO,EAAK,MAAM,EAAG,EAAE,CAAG,EAAK,OAAO,EAAE,CAAC,aAAa,CAAG,EAAK,MAAM,EAAI,EAAE,EAC5E,CACD,KAAK,IAAI,CASL,EAAyB,GAC3B,EAAK,QAAQ,+BAAgC,SAAS,CCrepD,EAAuB,GAAgB,CAChD,IAAI,EAAQ,EAAI,QAAQ,kBAAmB,OAAO,CAIlD,MAHA,GAAQ,EAAM,QAAQ,iBAAkB,SAAS,CACjD,EAAQ,EAAM,QAAQ,iBAAkB,SAAS,CAE1C,KAAK,UAAU,KAAK,MAAM,EAAM,CAAC,EAe/B,EAAwB,GAG7B,6GACmB,KAAK,EAAI,MAAM,CAAC,CAiB9B,EAAiB,IAElB,EAAM,MADA,wBACY,EAAI,EAAE,EAAE,IAAK,GAAe,EAAE,WAAW,IAAI,CAAG,EAAE,MAAM,EAAG,GAAG,CAAG,EAAG,CAiB5F,EAAqB,GAAgB,CACvC,IAAI,EAAa,EACjB,IAAK,IAAM,KAAQ,EACX,IAAS,KACT,IAGR,OAAO,EAAa,GAAM,GAGxB,EAAW,CAAE,IAAK,IAAK,IAAK,IAAK,IAAK,IAAK,CAC3C,GAAe,IAAI,IAAI,CAAC,IAAK,IAAK,IAAI,CAAC,CACvC,GAAgB,IAAI,IAAI,CAAC,IAAK,IAAK,IAAI,CAAC,CAqBxC,GAAuB,GAAgB,CACzC,IAAMC,EAAkB,EAAE,CAE1B,IAAK,IAAM,KAAQ,EACf,GAAI,GAAa,IAAI,EAAK,CACtB,EAAM,KAAK,EAAK,SACT,GAAc,IAAI,EAAK,CAAE,CAChC,IAAM,EAAW,EAAM,KAAK,CAC5B,GAAI,CAAC,GAAY,EAAS,KAAuC,EAC7D,MAAO,GAKnB,OAAO,EAAM,SAAW,GAuBf,GAAc,GAChB,EAAkB,EAAI,EAAI,GAAoB,EAAI,CAShD,GAAmB,GAAgC,CAC5D,GAAI,EAAU,SAAS,IAAI,CAAE,CACzB,GAAM,CAAC,EAAO,GAAO,EAAU,MAAM,IAAI,CAAC,IAAI,OAAO,CAErD,GAAI,EAAQ,EACR,MAAU,MAAM,6CAA6C,CAGjE,OAAO,MAAM,KAAK,CAAE,OAAQ,EAAM,EAAQ,EAAG,EAAG,EAAG,IAAM,EAAQ,EAAE,MAEnE,OAAO,EAAU,MAAM,IAAI,CAAC,IAAI,OAAO,EC1JlC,GAAiC,GACnC,EAAK,QACR,wHACA,IACH,CASQ,GAA4B,GAC9B,EAAK,QAAQ,aAAc,GAAG,CAS5B,GAA+B,GACjC,EAAK,QAAQ,OAAQ,IAAI,CASvB,GAAkB,GACpB,EAAK,QAAQ,SAAU,GAAG,CASxB,GAAmB,GACrB,EAAK,QAAQ,yDAA0D,GAAG,CASxE,GAA0B,GAC5B,EAAK,QAAQ,SAAU,GAAG,CASxB,GAA+B,GACjC,EAAK,QAAQ,4BAA6B,GAAG,CAS3C,GAAc,GAChB,EAAK,QACR,uGACA,GACH,CAQQ,GAA4B,GAEjC,EAEK,QAAQ,mBAAoB,KAAK,CAEjC,QAAQ,eAAgB,KAAK,CAE7B,QAAQ,eAAgB,KAAK,CAE7B,QAAQ,aAAc,KAAK,CAE3B,QAAQ,eAAgB,KAAK,CAE7B,QAAQ,aAAc,GAAG,CAEzB,QAAQ,uBAAwB,GAAG,CAEnC,QAAQ,wBAAyB,KAAK,CAEtC,QAAQ,WAAY,GAAG,CAEvB,QAAQ,iBAAkB,GAAG,CAE7B,QAAQ,iBAAkB,GAAG,CAE7B,QAAQ,MAAO,GAAG,CAoBlB,IAAY,EAAa,EAAI,MAAiB,EAAI,OAAS,EAAI,GAAG,EAAI,UAAU,EAAG,EAAI,EAAE,CAAC,GAAK,EAsB/F,IAAkB,EAAc,EAAoB,GAAI,IAAuB,CACxF,GAAI,EAAK,QAAU,EACf,OAAO,EAIX,IAAM,EAAmB,KAAK,IAAI,EAAG,KAAK,MAAM,EAAY,EAAE,CAAC,CACzD,EAAkB,GAAa,EAM/B,EAHkB,EAAY,EAGE,EAWtC,OARI,EAAc,EAEP,GAAG,EAAK,UAAU,EAAG,EAAY,EAAE,CAAC,GAMxC,GAHc,EAAK,UAAU,EAAG,EAAY,CAG5B,GAFJ,EAAK,UAAU,EAAK,OAAS,EAAgB,IAwBvD,GAAkB,GAAkB,EAAM,QAAQ,OAAQ,IAAI,CAAC,MAAM,CAW5EC,GAA2B,CAC7B,CAAC,IAAU,IAAU,IAAU,IAAS,CACxC,CAAC,IAAU,IAAS,CACpB,CAAC,IAAU,IAAS,CACvB,CAYK,GAAiB,GAAuB,CAC1C,IAAK,IAAM,KAAS,GAChB,GAAI,EAAM,SAAS,EAAG,CAElB,MAAO,IAAI,EAAM,IAAK,GAAM,EAAY,EAAE,CAAC,CAAC,KAAK,GAAG,CAAC,GAI7D,OAAO,EAAY,EAAG,EAIpB,GAAwB,GACnB,EACF,UAAU,MAAM,CAChB,QAAQ,kBAAmB,GAAG,CAC9B,QAAQ,OAAQ,IAAI,CACpB,MAAM,CAUF,GAA4B,GAAiB,CACtD,IACM,EAAO,GAAqB,EAAK,CAEvC,OAAO,MAAM,KAAK,EAAK,CAClB,IAAK,GAAO,GAAc,EAAG,CAAG,cAAkB,CAClD,KAAK,GAAG,EChQJ,GAA+B,GACjC,EACF,QAAQ,6EAA8E,QAAQ,CAC9F,QAAQ,wBAAyB,SAAS,CAC1C,QAAQ,eAAgB,QAAQ,CAU5B,GAA8B,GAChC,EAAK,QAAQ,MAAO,IAAI,CAAC,QAAQ,MAAO,IAAI,CAW1C,GAAgC,GAClC,EACF,QACG,2IACA,OACH,CACA,QAAQ,aAAc,KAAK,CAUvB,EAAa,GACf,EACF,UAAU,OAAO,CACjB,QAAQ,mBAAoB,GAAG,CAC/B,QAAQ,WAAY,GAAG,CAUnB,EAAwB,GAC1B,EAAgB,EAAK,QAAQ,uEAAwE,GAAG,CAAC,CAUvG,EAAkC,GAAiB,EAAU,EAAqB,EAAK,CAAC,CASxF,GAAmB,GACX,EAA+B,EAAS,CACpD,MAAM,CACN,MAAM,OAAO,CACb,MAAM,EAAG,EAAE,CACX,IAAK,GACK,EAAK,OAAO,EAAE,CAAC,aAAa,CACrC,CACD,KAAK,GAAG"}
+{"version":3,"file":"index.js","names":["DIACRITICS_CLASS","sentences: string[]","italicMap: Record<string, string>","stack: string[]","currentFlags","preformatArabicText: PreformatArabicText","EQUIV_GROUPS: string[][]"],"sources":["../src/constants.ts","../src/arabic.ts","../src/cleaning.ts","../src/formatting.ts","../src/parsing.ts","../src/preformat-core.ts","../src/preformat.ts","../src/sanitization.ts","../src/transliteration.ts"],"sourcesContent":["/** Matches text ending with common punctuation marks */\nexport const PATTERN_ENDS_WITH_PUNCTUATION = /[.!?؟؛]$/;\n","import { PATTERN_ENDS_WITH_PUNCTUATION } from './constants';\n\n/**\n * Converts Arabic-Indic numerals (٠-٩) to a JavaScript number.\n *\n * This function finds all Arabic-Indic digits in the input string and converts them\n * to their corresponding Arabic (Western) digits, then parses the result as an integer.\n *\n * Arabic-Indic digits mapping:\n * - ٠ → 0, ١ → 1, ٢ → 2, ٣ → 3, ٤ → 4\n * - ٥ → 5, ٦ → 6, ٧ → 7, ٨ → 8, ٩ → 9\n *\n * @param arabic - The string containing Arabic-Indic numerals to convert\n * @returns The parsed integer value of the converted numerals\n *\n * @example\n * ```typescript\n * arabicNumeralToNumber(\"١٢٣\"); // returns 123\n * arabicNumeralToNumber(\"٥٠\"); // returns 50\n * arabicNumeralToNumber(\"abc١٢٣xyz\"); // returns 123 (non-digits ignored)\n * arabicNumeralToNumber(\"\"); // returns NaN\n * ```\n *\n * Returns NaN if no valid Arabic-Indic digits are found\n */\nexport const arabicNumeralToNumber = (arabic: string) => {\n    return parseInt(\n        arabic.replace(/[\\u0660-\\u0669]/g, (c) => (c.charCodeAt(0) - 0x0660).toString()),\n        10,\n    );\n};\n\n/**\n * Removes extreme Arabic underscores (ـ) that appear at the beginning or end of a line or in text.\n * Does not affect Hijri dates (e.g., 1424هـ) or specific Arabic terms.\n * Example: \"ـThis is a textـ\" will be changed to \"This is a text\".\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with extreme underscores removed.\n */\nexport const cleanExtremeArabicUnderscores = (text: string) => {\n    return text.replace(/(?<!\\d ?ه|اه)ـ(?=\\r?$)|^ـ(?!اهـ)/gm, '');\n};\n\n/**\n * Converts Urdu symbols to their Arabic equivalents.\n * Example: 'ھذا' will be changed to 'هذا', 'ی' to 'ي'.\n * @param {string} text - The input text containing Urdu symbols.\n * @returns {string} - The modified text with Urdu symbols converted to Arabic symbols.\n */\nexport const convertUrduSymbolsToArabic = (text: string) => {\n    return text.replace(/ھ/g, 'ه').replace(/ی/g, 'ي');\n};\n\n/**\n * Calculates the proportion of Arabic characters in text relative to total non-whitespace, non-digit characters.\n * Digits (ASCII and Arabic-Indic variants) are excluded from both numerator and denominator.\n * @param text - The input text to analyze\n * @returns A decimal between 0-1 representing the Arabic character ratio (0 = no Arabic, 1 = all Arabic)\n */\nexport const getArabicScore = (text: string) => {\n    if (!text) {\n        return 0;\n    }\n    // Arabic letters (letters/ranges only)\n    const arabicLettersPattern = /[\\u0600-\\u06FF\\u0750-\\u077F\\u08A0-\\u08FF\\uFB50-\\uFDFF\\uFE70-\\uFEFF]/g;\n    // ASCII digits + Arabic-Indic digits + Extended Arabic-Indic digits\n    const allDigitPattern = /[0-9\\u0660-\\u0669\\u06F0-\\u06F9]/g;\n    // Counted characters exclude whitespace and all listed digits\n    const countedCharsPattern = /[^\\s0-9\\u0660-\\u0669\\u06F0-\\u06F9]/g;\n    const cleaned = text.replace(allDigitPattern, '');\n    const arabicMatches = cleaned.match(arabicLettersPattern) || [];\n    const totalMatches = cleaned.match(countedCharsPattern) || [];\n    return totalMatches.length === 0 ? 0 : arabicMatches.length / totalMatches.length;\n};\n\n/**\n * Finds the position of the last punctuation character in a string\n *\n * @param text - The text to search through\n * @returns The index of the last punctuation character, or -1 if none found\n *\n * @example\n * ```typescript\n * const text = \"Hello world! How are you?\";\n * const lastPuncIndex = findLastPunctuation(text);\n * // Result: 24 (position of the last '?')\n *\n * const noPuncText = \"Hello world\";\n * const notFound = findLastPunctuation(noPuncText);\n * // Result: -1 (no punctuation found)\n * ```\n */\nexport const findLastPunctuation = (text: string) => {\n    for (let i = text.length - 1; i >= 0; i--) {\n        if (PATTERN_ENDS_WITH_PUNCTUATION.test(text[i])) {\n            return i;\n        }\n    }\n\n    return -1;\n};\n\n/**\n * Fixes the trailing \"و\" (wow) in phrases such as \"عليكم و رحمة\" to \"عليكم ورحمة\".\n * This function attempts to correct phrases where \"و\" appears unnecessarily, particularly in greetings.\n * Example: 'السلام عليكم و رحمة' will be changed to 'السلام عليكم ورحمة'.\n * @param {string} text - The input text containing the \"و\" character.\n * @returns {string} - The modified text with unnecessary trailing \"و\" characters corrected.\n */\nexport const fixTrailingWow = (text: string) => {\n    return text.replace(/ و /g, ' و');\n};\n\n/**\n * Inserts a space between Arabic text and numbers.\n * Example: 'الآية37' will be changed to 'الآية 37'.\n * @param {string} text - The input text containing Arabic text followed by numbers.\n * @returns {string} - The modified text with spaces inserted between Arabic text and numbers.\n */\nexport const addSpaceBetweenArabicTextAndNumbers = (text: string) => {\n    return text.replace(/([\\u0600-\\u06FF]+)(\\d+)/g, '$1 $2');\n};\n\n/**\n * Removes single-digit numbers surrounded by Arabic text. Also removes dashes (-) not followed by a number.\n * For example, removes '3' from 'وهب 3 وقال' but does not remove '121' from 'لوحه 121 الجرح'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with non-index numbers and dashes removed.\n */\nexport const removeNonIndexSignatures = (text: string) => {\n    return text\n        .replace(/(?<![0-9] ?)-|(?<=[\\u0600-\\u06FF])\\s?\\d\\s?(?=[\\u0600-\\u06FF])/g, ' ')\n        .replace(/(?<=[\\u0600-\\u06FF]\\s)(\\d+\\s)+\\d+(?=(\\s[\\u0600-\\u06FF]|$))/g, ' ');\n};\n\n/**\n * Removes characters enclosed in square brackets [] or parentheses () if they are Arabic letters or Arabic-Indic numerals.\n * Example: '[س]' or '(س)' will be removed.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with singular codes removed.\n */\nexport const removeSingularCodes = (text: string) => {\n    return text.replace(/[[({][\\u0621-\\u064A\\u0660-\\u0669][\\])}]/g, '');\n};\n\n/**\n * Removes solitary Arabic letters unless they are the 'ha' letter, which is used in Hijri years.\n * Example: \"ب ا الكلمات ت\" will be changed to \"ا الكلمات\".\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with solitary Arabic letters removed.\n */\nexport const removeSolitaryArabicLetters = (text: string) => {\n    return text.replace(/(^| )[\\u0621-\\u064A]( |$)/g, ' ');\n};\n\n/**\n * Replaces English punctuation (question mark and semicolon) with their Arabic equivalents.\n * Example: '?' will be replaced with '؟', and ';' with '؛'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with English punctuation replaced by Arabic punctuation.\n */\nexport const replaceEnglishPunctuationWithArabic = (text: string) => {\n    return text\n        .replace(/\\?|؟\\./g, '؟')\n        .replace(/(;|؛)\\s*(\\1\\s*)*/g, '؛')\n        .replace(/,|-،/g, '،');\n};\n","/** Character class for Arabic diacritics (tashkīl/harakāt). */\nconst DIACRITICS_CLASS = '[\\\\u0610-\\\\u061A\\\\u064B-\\\\u065F\\\\u0670\\\\u06D6-\\\\u06ED]';\n/** Tatweel (kashīda) class. */\nconst TATWEEL_CLASS = '\\\\u0640';\n\n/**\n * Escape a string so it can be safely embedded into a RegExp source.\n *\n * @param s Any string\n * @returns Escaped string\n */\nexport const escapeRegex = (s: string): string => s.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&');\n\n/** Optional equivalence toggles for {@link makeDiacriticInsensitiveRegex}. */\ntype EquivOptions = {\n    /** Treat ا/أ/إ/آ as equivalent. @default true */\n    alif?: boolean;\n    /** Treat ة/ه as equivalent. @default true */\n    taMarbutahHa?: boolean;\n    /** Treat ى/ي as equivalent. @default true */\n    alifMaqsurahYa?: boolean;\n};\n\n/** Options for {@link makeDiacriticInsensitiveRegex}. */\nexport type MakeRegexOptions = {\n    /**\n     * Character equivalences to allow.\n     * @default { alif: true, taMarbutahHa: true, alifMaqsurahYa: true }\n     */\n    equivalences?: EquivOptions;\n\n    /**\n     * Allow tatweel between letters (tolerate decorative elongation).\n     * @default true\n     */\n    allowTatweel?: boolean;\n\n    /**\n     * Ignore diacritics by inserting a `DIACRITICS_CLASS*` after each letter.\n     * @default true\n     */\n    ignoreDiacritics?: boolean;\n\n    /**\n     * Treat any whitespace in the needle as `\\s+` for flexible matching.\n     * @default true\n     */\n    flexWhitespace?: boolean;\n\n    /**\n     * RegExp flags to use.\n     * @default 'u'\n     */\n    flags?: string;\n};\n\n/**\n * Build a **diacritic-insensitive**, **tatweel-tolerant** RegExp for Arabic text matching.\n *\n * Features:\n * - Optional character equivalences: ا~أ~إ~آ, ة~ه, ى~ي.\n * - Optional tolerance for tatweel between characters.\n * - Optional diacritic-insensitivity (by inserting a diacritics class after each char).\n * - Optional flexible whitespace (needle whitespace becomes `\\s+`).\n *\n * @param needle The Arabic text to match\n * @param opts See {@link MakeRegexOptions}\n * @returns A `RegExp` matching the needle with the desired tolerances\n *\n * @example\n * const rx = makeDiacriticInsensitiveRegex('أنا إلى الآفاق');\n * rx.test('انا الي الافاق'); // true\n * rx.test('اَنا إلى الآفاق'); // true\n */\nexport const makeDiacriticInsensitiveRegex = (needle: string, opts: MakeRegexOptions = {}): RegExp => {\n    const {\n        equivalences = { alif: true, taMarbutahHa: true, alifMaqsurahYa: true },\n        allowTatweel = true,\n        ignoreDiacritics = true,\n        flexWhitespace = true,\n        flags = 'u',\n    } = opts;\n\n    // Safety guard against extremely large inputs causing excessive pattern sizes\n    if (needle.length > 5000) {\n        throw new Error('makeDiacriticInsensitiveRegex: needle too long');\n    }\n\n    const charClass = (ch: string): string => {\n        switch (ch) {\n            case 'ا':\n            case 'أ':\n            case 'إ':\n            case 'آ':\n                return equivalences.alif ? '[اأإآ]' : 'ا';\n            case 'ة':\n            case 'ه':\n                return equivalences.taMarbutahHa ? '[هة]' : escapeRegex(ch);\n            case 'ى':\n            case 'ي':\n                return equivalences.alifMaqsurahYa ? '[ىي]' : escapeRegex(ch);\n            default:\n                return escapeRegex(ch);\n        }\n    };\n\n    const after = `${ignoreDiacritics ? `${DIACRITICS_CLASS}*` : ''}${allowTatweel ? `${TATWEEL_CLASS}*` : ''}`;\n\n    let pattern = '';\n    for (const ch of Array.from(needle)) {\n        if (/\\s/.test(ch)) {\n            pattern += flexWhitespace ? '\\\\s+' : '\\\\s*';\n        } else {\n            pattern += `${charClass(ch)}${after}`;\n        }\n    }\n\n    return new RegExp(pattern, flags);\n};\n\n/**\n * Remove simple HTML/XML-like tags from a string.\n *\n * This is intentionally lightweight and does not attempt to parse HTML; it simply drops\n * substrings that look like `<...>`.\n *\n * @param content Input string\n * @returns String with tags removed\n */\nexport const removeAllTags = (content: string) => content.replace(/<[^>]*>/g, '');\n","/**\n * Adds line breaks after punctuation marks such as periods, exclamation points, and question marks.\n * Example: 'Text.' becomes 'Text.\\n'.\n *\n * Note: For the full preformatting pipeline in one pass (significantly faster and more memory-friendly\n * on very large inputs), use `preformatArabicText` from `src/preformat.ts`.\n * @param {string} text - The input text containing punctuation.\n * @returns {string} - The modified text with line breaks added after punctuation.\n */\nexport const insertLineBreaksAfterPunctuation = (text: string) => {\n    // Define the punctuation marks that should trigger a new line\n    const punctuation = /([.?!؟])/g;\n\n    // Replace occurrences of punctuation marks followed by a space with the punctuation mark, a newline, and the space\n    const formattedText = text.replace(punctuation, '$1\\n').replace(/\\n\\s+/g, '\\n').trim();\n\n    return formattedText;\n};\n\n/**\n * Adds spaces before and after punctuation, except for certain cases like quoted text or ayah references.\n * Example: 'Text,word' becomes 'Text, word'.\n * @param {string} text - The input text containing punctuation.\n * @returns {string} - The modified text with spaces added before and after punctuation.\n */\nexport const addSpaceBeforeAndAfterPunctuation = (text: string) => {\n    return text\n        .replace(/( ?)([.!?,،؟;؛])((?![ '”“)\"\\]\\n])|(?=\\s{2,}))/g, '$1$2 ')\n        .replace(/\\s([.!?,،؟;؛])\\s*([ '”“)\"\\]\\n])/g, '$1$2')\n        .replace(/([^\\s\\w\\d'”“)\"\\]]+)\\s+([.!?,،؟;؛])|([.!?,،؟;؛])\\s+$/g, '$1$2$3')\n        .replace(/(?<=\\D)( ?: ?)(?!(\\d+:)|(:\\d+))|(?<=\\d) ?: ?(?=\\D)|(?<=\\D) ?: ?(?=\\d)/g, ': ');\n};\n\n/**\n * Turns regular double quotes surrounding a body of text into smart quotes.\n * Also fixes incorrect starting quotes by ensuring the string starts with an opening quote if needed.\n * Example: 'The \"quick brown\" fox' becomes 'The “quick brown” fox'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with smart quotes applied.\n */\nexport const applySmartQuotes = (text: string) => {\n    return text\n        .replace(/[“”]/g, '\"')\n        .replace(/\"([^\"]*)\"/g, '“$1”')\n        .replace(/^”/g, '“');\n};\n\n/**\n * Replaces literal new line characters (\\n) and carriage returns (\\r) with actual line breaks.\n * Example: 'A\\\\nB' becomes 'A\\nB'.\n * @param {string} text - The input text containing literal new lines.\n * @returns {string} - The modified text with actual line breaks.\n */\nexport const cleanLiteralNewLines = (text: string) => {\n    return text.replace(/\\\\n|\\r/g, '\\n');\n};\n\n/**\n * Removes trailing spaces from each line in a multiline string.\n * Example: \" This is a line   \\nAnother line   \" becomes \"This is a line\\nAnother line\".\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with trailing spaces removed.\n */\nexport const cleanMultilines = (text: string) => {\n    return text.replace(/^ +| +$/gm, '');\n};\n\n/**\n * Detects if a word is by itself in a line.\n * @param text The text to check.\n * @returns true if there exists a word in any of the lines in the text that is by itself.\n */\nexport const hasWordInSingleLine = (text: string): boolean => {\n    return /^\\s*\\S+\\s*$/gm.test(text);\n};\n\n/**\n * Checks if the input string consists of only punctuation characters.\n * @param {string} text - The input text to check.\n * @returns {boolean} - Returns true if the string contains only punctuation, false otherwise.\n */\nexport const isOnlyPunctuation = (text: string): boolean => {\n    const regex = /^[\\u0020-\\u002f\\u003a-\\u0040\\u005b-\\u0060\\u007b-\\u007e0-9٠-٩]+$/;\n    return regex.test(text);\n};\n\n/**\n * Cleans unnecessary spaces before punctuation marks such as periods, commas, and question marks.\n * Example: 'This is a sentence , with extra space .' becomes 'This is a sentence, with extra space.'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with cleaned spaces before punctuation.\n */\nexport const cleanSpacesBeforePeriod = (text: string) => {\n    return text.replace(/\\s+([.؟!,،؛:?])/g, '$1');\n};\n\n/**\n * Condenses multiple asterisks (*) into a single one.\n * Example: '***' becomes '*'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed asterisks.\n */\nexport const condenseAsterisks = (text: string) => {\n    return text.replace(/(\\*\\s*)+/g, '*');\n};\n\n/**\n * Replaces occurrences of colons surrounded by periods (e.g., '.:.' or ':') with a single colon.\n * Example: 'This.:. is a test' becomes 'This: is a test'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed colons.\n */\nexport const condenseColons = (text: string) => {\n    return text.replace(/[.-]?:[.-]?/g, ':');\n};\n\n/**\n * Condenses two or more dashes (--) into a single dash (-).\n * Example: 'This is some ---- text' becomes 'This is some - text'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed dashes.\n */\nexport const condenseDashes = (text: string) => {\n    return text.replace(/-{2,}/g, '-');\n};\n\n/**\n * Replaces sequences of two or more periods (e.g., '...') with an ellipsis character (…).\n * Example: 'This is a test...' becomes 'This is a test…'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with ellipses condensed.\n */\nexport const condenseEllipsis = (text: string) => {\n    return text.replace(/\\.{2,}/g, '…');\n};\n\n/**\n * Reduces multiple consecutive line breaks (3 or more) to exactly 2 line breaks.\n * Example: 'This is line 1\\n\\n\\n\\nThis is line 2' becomes 'This is line 1\\n\\nThis is line 2'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed line breaks.\n */\nexport const reduceMultilineBreaksToDouble = (text: string) => {\n    return text.replace(/(\\n\\s*){3,}/g, '\\n\\n');\n};\n\n/**\n * Reduces multiple consecutive line breaks (2 or more) to exactly 1 line break.\n * Example: 'This is line 1\\n\\nThis is line 2' becomes 'This is line 1\\nThis is line 2'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed line breaks.\n */\nexport const reduceMultilineBreaksToSingle = (text: string) => {\n    return text.replace(/(\\n\\s*){2,}/g, '\\n');\n};\n\n/**\n * Condenses multiple periods separated by spaces (e.g., '. . .') into a single period.\n * Example: 'This . . . is a test' becomes 'This. is a test'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed periods.\n */\nexport const condensePeriods = (text: string) => {\n    return text.replace(/\\. +\\./g, '.');\n};\n\n/**\n * Condenses multiple underscores (__) or Arabic Tatweel characters (ـــــ) into a single underscore or Tatweel.\n * Example: 'This is ـــ some text __' becomes 'This is ـ some text _'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed underscores.\n */\nexport const condenseUnderscores = (text: string) => {\n    return text.replace(/ـ{2,}/g, 'ـ').replace(/_+/g, '_');\n};\n\n/**\n * Replaces double parentheses or brackets with single ones.\n * Example: '((text))' becomes '(text)'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed brackets.\n */\nexport const doubleToSingleBrackets = (text: string) => {\n    return text.replace(/(\\(|\\)){2,}|(\\[|\\]){2,}/g, '$1$2');\n};\n\n/**\n * Ensures at most 1 space exists before any word before brackets.\n * Adds a space if there isn't one, or reduces multiple spaces to one.\n * @param {string} text - The input text to modify\n * @returns {string} - The modified text with proper spacing before brackets\n */\nexport const ensureSpaceBeforeBrackets = (text: string) => {\n    return text.replace(/(\\S) *(\\([^)]*\\))/g, '$1 $2');\n};\n\n/**\n * Ensures at most 1 space exists before any word before Arabic quotation marks.\n * Adds a space if there isn't one, or reduces multiple spaces to one.\n * @param {string} text - The input text to modify\n * @returns {string} - The modified text with proper spacing before Arabic quotes\n */\nexport const ensureSpaceBeforeQuotes = (text: string) => {\n    return text.replace(/(\\S) *(«[^»]*»)/g, '$1 $2');\n};\n\n/**\n * Fixes common bracket and quotation mark typos in text\n * Corrects malformed patterns like \"(«\", \"»)\", and misplaced digits in brackets\n * @param text - Input text that may contain bracket typos\n * @returns Text with corrected bracket and quotation mark combinations\n */\nexport const fixBracketTypos = (text: string) => {\n    return (\n        text\n            .replace(/\\(«|\\( \\(/g, '«')\n            .replace(/»\\)|\\) \\)/g, '»')\n            // Fix \")digit)\" pattern to \"(digit)\"\n            .replace(/\\)([0-9\\u0660-\\u0669]+)\\)/g, '($1)')\n            // Fix \")digit(\" pattern to \"(digit)\"\n            .replace(/\\)([0-9\\u0660-\\u0669]+)\\(/g, '($1)')\n    );\n};\n\n/**\n * Fixes mismatched curly braces by converting incorrect bracket/brace combinations\n * to proper curly braces { }\n * @param text - Input text that may contain mismatched curly braces\n * @returns Text with corrected curly brace pairs\n */\nexport const fixCurlyBraces = (text: string) => {\n    // Process each mismatch type separately to avoid interference\n    let result = text;\n\n    // Fix ( content } to { content }\n    result = result.replace(/\\(([^(){}]+)\\}/g, '{$1}');\n\n    // Fix { content ) to { content }\n    return result.replace(/\\{([^(){}]+)\\)/g, '{$1}');\n};\n\n/**\n * Fixes mismatched quotation marks in Arabic text by converting various\n * incorrect bracket/quote combinations to proper Arabic quotation marks (« »)\n * @param text - Input text that may contain mismatched quotation marks\n * @returns Text with corrected Arabic quotation marks\n */\nexport const fixMismatchedQuotationMarks = (text: string) => {\n    return (\n        text\n            // Matches mismatched quotation marks: « followed by content and closed with )\n            .replace(/«([^»)]+)\\)/g, '«$1»')\n            // Fix reverse mismatched ( content » to « content »\n            .replace(/\\(([^()]+)»/g, '«$1»')\n            // Matches any unclosed « quotation marks at end of content\n            .replace(/«([^»]+)(?=\\s*$|$)/g, '«$1»')\n    );\n};\n\n/**\n * Formats a multiline string by joining sentences and maintaining footnotes on their own lines.\n * Footnotes are identified by Arabic and English numerals.\n * Example: 'Sentence one.\\n(1) A footnote.\\nSentence two.' remains the same, while regular sentences are joined.\n * @param {string} input - The input text containing sentences and footnotes.\n * @returns {string} - The formatted text.\n */\nexport const formatStringBySentence = (input: string) => {\n    const footnoteRegex = /^\\((?:\\d+|۱|۲|۳|۴|۵|۶|۷|۸|۹)\\)\\s/;\n    const sentences: string[] = [];\n    const lines = input.split('\\n');\n    let currentSentence = '';\n\n    lines.forEach((line) => {\n        const trimmedLine = line.trim();\n        const isFootnote = footnoteRegex.test(trimmedLine);\n        const isNumber = /^\\(\\d+\\/\\d+\\)/.test(trimmedLine);\n\n        if (isFootnote && !isNumber) {\n            if (currentSentence) {\n                sentences.push(currentSentence.trim());\n                currentSentence = '';\n            }\n            sentences.push(trimmedLine);\n        } else {\n            currentSentence += `${trimmedLine} `;\n            const lastChar = currentSentence.trim().slice(-1);\n            if (/[.!؟]/.test(lastChar)) {\n                sentences.push(currentSentence.trim());\n                currentSentence = '';\n            }\n        }\n    });\n\n    // Add any remaining text to the output\n    if (currentSentence) {\n        sentences.push(currentSentence.trim());\n    }\n\n    return sentences.join('\\n');\n};\n\n/**\n * Detects if text is entirely in uppercase letters\n * @param text - The text to check\n * @returns true if all alphabetic characters are uppercase, false otherwise\n */\nexport const isAllUppercase = (text: string) => {\n    // Remove non-letter characters (including numbers, punctuation, spaces)\n    // \\p{L} matches any Unicode letter character\n    const lettersOnly = text.replace(/[^\\p{L}]/gu, '');\n\n    // If there are no letter characters, return false\n    if (lettersOnly.length === 0) {\n        return false;\n    }\n\n    return lettersOnly === lettersOnly.toUpperCase();\n};\n\n/**\n * Removes unnecessary spaces around slashes in references.\n * Example: '127 / 11' becomes '127/11'.\n * @param {string} text - The input text containing references.\n * @returns {string} - The modified text with spaces removed around slashes.\n */\nexport const normalizeSlashInReferences = (text: string) => {\n    return text.replace(/(\\d+)\\s?\\/\\s?(\\d+)/g, '$1/$2');\n};\n\n/**\n * Reduces multiple spaces or tabs to a single space.\n * Example: 'This   is a   text' becomes 'This is a text'.\n * @param {string} text - The input text containing extra spaces.\n * @returns {string} - The modified text with reduced spaces.\n */\nexport const normalizeSpaces = (text: string) => {\n    return text.replace(/[ \\t]+/g, ' ');\n};\n\n/**\n * Removes redundant punctuation marks that follow Arabic question marks or exclamation marks.\n * This function cleans up text by removing periods (.) or Arabic commas (،) that immediately\n * follow Arabic question marks (؟) or exclamation marks (!), as they are considered redundant\n * in proper Arabic punctuation.\n *\n * @param text - The Arabic text to clean up\n * @returns The text with redundant punctuation removed\n *\n * @example\n * ```typescript\n * removeRedundantPunctuation('كيف حالك؟.') // Returns: 'كيف حالك؟'\n * removeRedundantPunctuation('ممتاز!،') // Returns: 'ممتاز!'\n * removeRedundantPunctuation('هذا جيد.') // Returns: 'هذا جيد.' (unchanged)\n * ```\n */\nexport const removeRedundantPunctuation = (text: string) => {\n    return text.replace(/([؟!])[.،]/g, '$1');\n};\n\n/**\n * Removes spaces inside brackets, parentheses, or square brackets.\n * Example: '( a b )' becomes '(a b)'.\n * @param {string} text - The input text with spaces inside brackets.\n * @returns {string} - The modified text with spaces removed inside brackets.\n */\nexport const removeSpaceInsideBrackets = (text: string) => {\n    return text.replace(/([[(])\\s*(.*?)\\s*([\\])])/g, '$1$2$3');\n};\n\n/**\n * Replaces double parentheses single a single arrow variation.\n * Example: '((text))' becomes '«text»'.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with condensed brackets.\n */\nexport const replaceDoubleBracketsWithArrows = (text: string) => {\n    return text.replace(/\\(\\(\\s?/g, '«').replace(/\\s?\\)\\)/g, '»');\n};\n\n/**\n * Removes bold styling from text by normalizing the string and removing stylistic characters.\n * @param {string} text - The input text containing bold characters.\n * @returns {string} - The modified text with bold styling removed.\n */\nexport const stripBoldStyling = (text: string) => {\n    // Normalize the string to NFKD form\n    const normalizedString = text.normalize('NFKD');\n\n    // Remove combining marks (diacritics) and stylistic characters from the string\n    return normalizedString.replace(/[\\u0300-\\u036f]/g, '').trim();\n};\n\n/**\n * Removes italicized characters by replacing italic Unicode characters with their normal counterparts.\n * Example: '𝘼𝘽𝘾' becomes 'ABC'.\n * @param {string} text - The input text containing italicized characters.\n * @returns {string} - The modified text with italics removed.\n */\nexport const stripItalicsStyling = (text: string) => {\n    const italicMap: Record<string, string> = {\n        '\\uD835\\uDC4E': 'I',\n        '\\uD835\\uDC68': 'g',\n        '\\u{1D63C}': '!',\n        '\\uD835\\uDC4F': 'J',\n        '\\uD835\\uDC69': 'h',\n        '\\u{1D63D}': '?',\n        '\\uD835\\uDC50': 'K',\n        '\\uD835\\uDC6A': 'i',\n        '\\uD835\\uDC51': 'L',\n        '\\uD835\\uDC6B': 'j',\n        '\\u{1D63F}': ',',\n        '\\uD835\\uDC52': 'M',\n        '\\uD835\\uDC6C': 'k',\n        '\\u{1D640}': '.',\n        '\\uD835\\uDC53': 'N',\n        '\\uD835\\uDC6D': 'l',\n        '\\uD835\\uDC54': 'O',\n        '\\uD835\\uDC6E': 'm',\n        '\\uD835\\uDC6F': 'n',\n        '\\uD835\\uDC56': 'Q',\n        '\\uD835\\uDC70': 'o',\n        '\\uD835\\uDC57': 'R',\n        '\\uD835\\uDC71': 'p',\n        '\\uD835\\uDC58': 'S',\n        '\\uD835\\uDC72': 'q',\n        '\\uD835\\uDC59': 'T',\n        '\\uD835\\uDC73': 'r',\n        '\\u{1D647}': '-',\n        '\\uD835\\uDC5A': 'U',\n        '\\uD835\\uDC74': 's',\n        '\\uD835\\uDC5B': 'V',\n        '\\uD835\\uDC75': 't',\n        '\\uD835\\uDC5C': 'W',\n        '\\uD835\\uDC76': 'u',\n        '\\uD835\\uDC5D': 'X',\n        '\\uD835\\uDC77': 'v',\n        '\\uD835\\uDC5E': 'Y',\n        '\\uD835\\uDC78': 'w',\n        '\\uD835\\uDC5F': 'Z',\n        '\\uD835\\uDC79': 'x',\n        '\\uD835\\uDC46': 'A',\n        '\\uD835\\uDC7A': 'y',\n        '\\uD835\\uDC47': 'B',\n        '\\uD835\\uDC7B': 'z',\n        '\\uD835\\uDC62': 'a',\n        '\\uD835\\uDC48': 'C',\n        '\\uD835\\uDC63': 'b',\n        '\\uD835\\uDC49': 'D',\n        '\\uD835\\uDC64': 'c',\n        '\\uD835\\uDC4A': 'E',\n        '\\uD835\\uDC65': 'd',\n        '\\uD835\\uDC4B': 'F',\n        '\\uD835\\uDC66': 'e',\n        '\\uD835\\uDC4C': 'G',\n        '\\uD835\\uDC67': 'f',\n        '\\uD835\\uDC4D': 'H',\n        '\\uD835\\uDC55': 'P',\n    };\n\n    return text.replace(/[\\uD835\\uDC62-\\uD835\\uDC7B\\uD835\\uDC46-\\uD835\\uDC5F\\u{1D63C}-\\u{1D647}]/gu, (match) => {\n        return italicMap[match] || match;\n    });\n};\n\n/**\n * Removes all bold and italic styling from the input text.\n * @param {string} text - The input text to remove styling from.\n * @returns {string} - The modified text with all styling removed.\n */\nexport const stripStyling = (text: string) => {\n    return stripItalicsStyling(stripBoldStyling(text));\n};\n\n/**\n * Converts a string to title case (first letter of each word capitalized)\n * @param str - The input string to convert\n * @returns String with each word's first letter capitalized\n */\nexport const toTitleCase = (str: string) => {\n    return str\n        .toLowerCase()\n        .split(' ')\n        .map((word) => {\n            if (word.length === 0) {\n                return word;\n            }\n            // Find the first Unicode letter in the chunk\n            const match = word.match(/\\p{L}/u);\n            if (!match || match.index === undefined) {\n                return word;\n            }\n            const i = match.index;\n            return word.slice(0, i) + word.charAt(i).toUpperCase() + word.slice(i + 1);\n        })\n        .join(' ');\n};\n\n/**\n * Removes unnecessary spaces inside quotes.\n * Example: '“ Text ”' becomes '“Text”'.\n * @param {string} text - The input text with spaces inside quotes.\n * @returns {string} - The modified text with spaces removed inside quotes.\n */\nexport const trimSpaceInsideQuotes = (text: string) => {\n    return text.replace(/([“”\"]|«) *(.*?) *([“”\"]|»)/g, '$1$2$3');\n};\n","/**\n * Converts a string that resembles JSON but with numeric keys and single-quoted values\n * into valid JSON format. This function replaces numeric keys with quoted numeric keys\n * and ensures all values are double-quoted as required by JSON.\n *\n * @param {string} str - The input string that needs to be fixed into valid JSON.\n * @returns {string} - A valid JSON string.\n *\n * @example\n * const result = normalizeJsonSyntax(\"{10: 'abc', 20: 'def'}\");\n * console.log(result); // '{\"10\": \"abc\", \"20\": \"def\"}'\n */\nexport const normalizeJsonSyntax = (str: string) => {\n    let input = str.replace(/(\\b\\d+\\b)(?=:)/g, '\"$1\"');\n    input = input.replace(/:\\s*'([^']+)'/g, ': \"$1\"');\n    input = input.replace(/:\\s*\"([^\"]+)\"/g, ': \"$1\"');\n\n    return JSON.stringify(JSON.parse(input));\n};\n\n/**\n * Checks if a given string resembles a JSON object with numeric or quoted keys and values\n * that are single or double quoted. This is useful for detecting malformed JSON-like\n * structures that can be fixed by the `normalizeJsonSyntax` function.\n *\n * @param {string} str - The input string to check.\n * @returns {boolean} - Returns true if the string is JSON-like, false otherwise.\n *\n * @example\n * const result = isJsonStructureValid(\"{10: 'abc', 'key': 'value'}\");\n * console.log(result); // true\n */\nexport const isJsonStructureValid = (str: string) => {\n    // Checks for a pattern with numeric keys or quoted keys and values in quotes\n    const jsonLikePattern =\n        /^{(\\s*(\\d+|'[^']*'|\"[^\"]*\")\\s*:\\s*('|\")[^'\"]*\\3\\s*,)*(?:\\s*(\\d+|'[^']*'|\"[^\"]*\")\\s*:\\s*('|\")[^'\"]*\\5\\s*)}$/;\n    return jsonLikePattern.test(str.trim());\n};\n\n/**\n * Splits a string by spaces and quoted substrings.\n *\n * This function takes an input string and splits it into parts where substrings\n * enclosed in double quotes are treated as a single part. Other substrings\n * separated by spaces are split normally.\n *\n * @param {string} query - The input string to be split.\n * @returns {string[]} An array of strings, with quoted substrings kept intact.\n *\n * @example\n * const result = splitByQuotes('\"This is\" \"a part of the\" \"string and\"');\n * console.log(result); // [\"This is\", \"a part of the\", \"string and\"]\n */\nexport const splitByQuotes = (query: string): string[] => {\n    const regex = /(?:[^\\s\"]+|\"(.*?)\")+/g;\n    return (query.match(regex) || []).map((s: string) => (s.startsWith('\"') ? s.slice(1, -1) : s));\n};\n\n/**\n * Checks if all double quotes in a string are balanced (even count).\n * A string has balanced quotes if every opening quote has a corresponding closing quote.\n *\n * @param str - The string to check for balanced quotes\n * @returns True if quotes are balanced (even count), false otherwise\n *\n * @example\n * ```typescript\n * areQuotesBalanced('Hello \"world\"') // Returns: true\n * areQuotesBalanced('Hello \"world') // Returns: false\n * areQuotesBalanced('No quotes') // Returns: true\n * ```\n */\nconst areQuotesBalanced = (str: string) => {\n    let quoteCount = 0;\n    for (const char of str) {\n        if (char === '\"') {\n            quoteCount++;\n        }\n    }\n    return quoteCount % 2 === 0;\n};\n\nconst brackets = { '(': ')', '[': ']', '{': '}' };\nconst openBrackets = new Set(['(', '[', '{']);\nconst closeBrackets = new Set([')', ']', '}']);\n\n/**\n * Checks if all brackets in a string are properly balanced and matched.\n * This function validates that every opening bracket has a corresponding closing bracket\n * in the correct order and of the matching type.\n *\n * Supported bracket types: parentheses (), square brackets [], curly braces {}\n *\n * @param str - The string to check for balanced brackets\n * @returns True if all brackets are properly balanced and matched, false otherwise\n *\n * @example\n * ```typescript\n * areBracketsBalanced('(hello [world])') // Returns: true\n * areBracketsBalanced('(hello [world)') // Returns: false (mismatched)\n * areBracketsBalanced('((hello))') // Returns: true\n * areBracketsBalanced('(hello') // Returns: false (unclosed)\n * ```\n */\n\nconst areBracketsBalanced = (str: string) => {\n    const stack: string[] = [];\n\n    for (const char of str) {\n        if (openBrackets.has(char)) {\n            stack.push(char);\n        } else if (closeBrackets.has(char)) {\n            const lastOpen = stack.pop();\n            if (!lastOpen || brackets[lastOpen as keyof typeof brackets] !== char) {\n                return false;\n            }\n        }\n    }\n\n    return stack.length === 0;\n};\n\n/**\n * Checks if both quotes and brackets are balanced in a string.\n * This function combines quote balance checking and bracket balance checking\n * to ensure the entire string has properly balanced punctuation.\n *\n * A string is considered balanced when:\n * - All double quotes have matching pairs (even count)\n * - All brackets (parentheses, square brackets, curly braces) are properly matched and nested\n *\n * @param str - The string to check for balanced quotes and brackets\n * @returns True if both quotes and brackets are balanced, false otherwise\n *\n * @example\n * ```typescript\n * isBalanced('He said \"Hello (world)!\"') // Returns: true\n * isBalanced('He said \"Hello (world!\"') // Returns: false (unbalanced quote)\n * isBalanced('He said \"Hello (world)\"') // Returns: false (unbalanced quote)\n * isBalanced('Hello (world) [test]') // Returns: true\n * ```\n */\nexport const isBalanced = (str: string) => {\n    return areQuotesBalanced(str) && areBracketsBalanced(str);\n};\n\n/**\n * Parses page input string into array of page numbers, supporting ranges and lists\n * @param pageInput - Page specification string (e.g., \"1-5\" or \"1,3,5\")\n * @returns Array of page numbers\n * @throws Error when start page exceeds end page in range\n */\nexport const parsePageRanges = (pageInput: string): number[] => {\n    if (pageInput.includes('-')) {\n        const [start, end] = pageInput.split('-').map(Number);\n\n        if (start > end) {\n            throw new Error('Start page cannot be greater than end page');\n        }\n\n        return Array.from({ length: end - start + 1 }, (_, i) => start + i);\n    } else {\n        return pageInput.split(',').map(Number);\n    }\n};\n","/**\n * Internal implementation for `preformatArabicText`.\n *\n * This file intentionally exports additional helpers for benchmarking/testing,\n * but it is NOT exported from `src/index.ts` (public API).\n */\n\n// ==============================================================================\n// CONSTANTS & LOOKUP TABLE\n// ==============================================================================\n\nconst F_SPACE = 1;\nconst F_NO_SPACE_BEFORE = 2; // Punctuation/Closing that consumes preceding spaces\nconst F_OPENING = 4; // Opening brackets/quotes\nconst F_CLOSING = 8; // Closing brackets/quotes\nconst F_DIGIT = 16;\nconst F_ARABIC = 32;\nconst F_SPECIAL = 64; // Needs specific handling (transform, condense, etc)\nconst F_PUNCT = 128; // Punctuation that might need spacing after\n\nconst CHAR_MAP = new Uint8Array(65536);\n\n/**\n * Mark a set of characters with the given bit flags in {@link CHAR_MAP}.\n *\n * @param chars A string of individual characters to mark\n * @param flags Bitmask to OR into the lookup table for each character\n */\nconst setFlags = (chars: string, flags: number) => {\n    for (let i = 0; i < chars.length; i++) {\n        CHAR_MAP[chars.charCodeAt(i)] |= flags;\n    }\n};\n\n// Initialize Map\nsetFlags(' \\t', F_SPACE);\nsetFlags('\\n\\r', F_SPECIAL); // Newlines are special\n\n// Digits\nfor (let i = 48; i <= 57; i++) {\n    CHAR_MAP[i] |= F_DIGIT; // 0-9\n}\n\n// Arabic Letters (Common ranges)\nfor (let i = 0x0600; i <= 0x06ff; i++) {\n    CHAR_MAP[i] |= F_ARABIC;\n}\nfor (let i = 0x0750; i <= 0x077f; i++) {\n    CHAR_MAP[i] |= F_ARABIC;\n}\n\n// Punctuation & Brackets\n// No Space Before: . ! ? , : ; and Arabic equivalents\nsetFlags('.!?,:;', F_NO_SPACE_BEFORE | F_PUNCT | F_SPECIAL);\nsetFlags('،؛؟', F_NO_SPACE_BEFORE | F_PUNCT | F_SPECIAL);\n\n// Opening: ( [ { \" ' « “\nsetFlags('([{\"\\'«“', F_OPENING | F_SPECIAL);\n\n// Closing: ) ] } \" ' » ”\nsetFlags(')]}\"\\'»”', F_CLOSING | F_NO_SPACE_BEFORE | F_SPECIAL);\n\n// Specific Specials\nsetFlags('-_*و/', F_SPECIAL);\n\n// Codes for fast comparison\nconst C_TAB = 9;\nconst C_SPACE = 32;\nconst C_NEWLINE = 10;\nconst C_CR = 13;\nconst C_DOT = 46;\nconst C_COMMA = 44;\nconst C_COLON = 58;\nconst C_SEMICOLON = 59;\nconst C_Q_MARK = 63;\nconst C_EXCLAM = 33;\nconst C_SLASH = 47;\nconst C_DASH = 45;\nconst C_UNDERSCORE = 95;\nconst C_ASTERISK = 42;\nconst C_L_PAREN = 40;\nconst C_R_PAREN = 41;\nconst C_L_GUILLEMET = 171; // «\nconst C_R_GUILLEMET = 187; // »\nconst C_DQUOTE = 34; // \"\nconst C_SQUOTE = 39; // '\nconst C_RDQUOTE = 8221; // ”\nconst C_WOW = 1608; // و\nconst C_AR_COMMA = 1548; // ،\nconst C_AR_SEMICOLON = 1563; // ؛\nconst C_AR_Q_MARK = 1567; // ؟\nconst C_TATWEEL = 1600; // ـ\nconst C_ELLIPSIS = 8230; // …\n\n/**\n * Check whether a code unit should be treated as trim whitespace for final output.\n *\n * @param code UTF-16 code unit\n * @returns True if the code unit is a trim-whitespace character\n */\nconst isTrimWhitespace = (code: number): boolean => {\n    return code === C_SPACE || code === C_TAB || code === C_NEWLINE || code === C_CR;\n};\n\n/**\n * Growable UTF-16 output buffer used by the buffer-backed preformat implementation.\n *\n * Note: This builder stores UTF-16 code units; it assumes all emitted characters are within\n * the BMP (which is true for the transformations this pipeline performs).\n */\nclass Utf16Builder {\n    private buffer: Uint16Array;\n    public length: number;\n\n    public constructor(initialCapacity: number) {\n        this.buffer = new Uint16Array(Math.max(16, initialCapacity));\n        this.length = 0;\n    }\n\n    /**\n     * Return the last written UTF-16 code unit, or 0 if empty.\n     */\n    public last(): number {\n        return this.length > 0 ? this.buffer[this.length - 1] : 0;\n    }\n\n    /**\n     * Return the second-to-last written UTF-16 code unit, or 0 if not available.\n     */\n    public secondLast(): number {\n        return this.length > 1 ? this.buffer[this.length - 2] : 0;\n    }\n\n    /**\n     * Append a UTF-16 code unit to the buffer.\n     */\n    public push(code: number) {\n        this.ensureCapacity(1);\n        this.buffer[this.length] = code;\n        this.length++;\n    }\n\n    /**\n     * Remove the last UTF-16 code unit (no-op if empty).\n     */\n    public pop() {\n        if (this.length > 0) {\n            this.length--;\n        }\n    }\n\n    private ensureCapacity(extra: number) {\n        const needed = this.length + extra;\n        if (needed <= this.buffer.length) {\n            return;\n        }\n        let nextCap = this.buffer.length * 2;\n        if (nextCap < needed) {\n            nextCap = needed;\n        }\n        const next = new Uint16Array(nextCap);\n        next.set(this.buffer.subarray(0, this.length));\n        this.buffer = next;\n    }\n\n    /**\n     * Convert the buffer to a string, trimming leading/trailing whitespace (space/tab/newline/CR).\n     *\n     * Uses chunked `String.fromCharCode` to avoid stack limits on large outputs.\n     */\n    public toStringTrimmed(): string {\n        let start = 0;\n        let end = this.length;\n\n        while (start < end && isTrimWhitespace(this.buffer[start])) {\n            start++;\n        }\n        while (end > start && isTrimWhitespace(this.buffer[end - 1])) {\n            end--;\n        }\n\n        if (end <= start) {\n            return '';\n        }\n\n        // Chunked conversion to avoid call stack limits.\n        const CHUNK = 0x8000;\n        let out = '';\n        for (let i = start; i < end; i += CHUNK) {\n            const slice = this.buffer.subarray(i, Math.min(end, i + CHUNK));\n            out += String.fromCharCode(...slice);\n        }\n        return out;\n    }\n}\n\n// ==============================================================================\n// IMPLEMENTATIONS\n// ==============================================================================\n\n/**\n * NOTE ON MAINTAINABILITY:\n *\n * `processStringConcat` and `processStringBuffer` must remain byte-for-byte identical in behavior.\n * Any bug fix or behavior change MUST be applied to both implementations, and should be validated\n * by running the correctness tests (including `src/preformat.memory.test.ts` for large inputs).\n */\n\n/**\n * Preformat using string concatenation (`res += ...`).\n *\n * This is typically fastest for common page-sized inputs under Bun/V8.\n *\n * @param text Input string\n * @returns Preformatted string\n */\nconst processStringConcat = (text: string): string => {\n    if (!text) {\n        return '';\n    }\n\n    let res = '';\n    const len = text.length;\n    let i = 0;\n\n    let pendingSpaces = 0;\n    let lastCode = 0;\n\n    while (i < len) {\n        let code = text.charCodeAt(i);\n        const originalCode = code;\n        const flags = CHAR_MAP[code];\n\n        // ---------------------------------------------------------\n        // 1. Handle Whitespace\n        // ---------------------------------------------------------\n        if (flags & F_SPACE) {\n            pendingSpaces++;\n            i++;\n            continue;\n        }\n\n        // ---------------------------------------------------------\n        // 2. Handle Special Characters (Newlines, Transforms)\n        // ---------------------------------------------------------\n        if (flags & F_SPECIAL) {\n            // Newlines\n            if (code === C_NEWLINE || code === C_CR) {\n                pendingSpaces = 0;\n                if (lastCode !== C_NEWLINE) {\n                    res += '\\n';\n                    lastCode = C_NEWLINE;\n                }\n                // Skip subsequent whitespace\n                i++;\n                while (i < len) {\n                    const next = text.charCodeAt(i);\n                    if (next === C_SPACE || next === C_TAB || next === C_NEWLINE || next === C_CR) {\n                        i++;\n                    } else {\n                        break;\n                    }\n                }\n                continue;\n            }\n\n            // Transforms\n            if (code === C_Q_MARK) {\n                code = C_AR_Q_MARK;\n            } else if (code === C_SEMICOLON) {\n                code = C_AR_SEMICOLON;\n            } else if (code === C_COMMA) {\n                code = C_AR_COMMA;\n            }\n\n            // Condense Colons (.:. -> :)\n            else if (code === C_COLON) {\n                if (lastCode === C_DOT || lastCode === C_DASH) {\n                    res = res.slice(0, -1);\n                    lastCode = res.charCodeAt(res.length - 1) || 0;\n                }\n                const next = text.charCodeAt(i + 1);\n                if (next === C_DOT || next === C_DASH) {\n                    i++;\n                }\n            }\n\n            // Double Brackets\n            else if (code === C_L_PAREN && text.charCodeAt(i + 1) === C_L_PAREN) {\n                code = C_L_GUILLEMET; // « (\\u00AB)\n                i++;\n            } else if (code === C_R_PAREN && text.charCodeAt(i + 1) === C_R_PAREN) {\n                code = C_R_GUILLEMET; // » (\\u00BB)\n                i++;\n            }\n\n            // Ellipsis (...)\n            else if (code === C_DOT) {\n                if (lastCode === C_ELLIPSIS) {\n                    i++;\n                    continue;\n                }\n                if (lastCode === C_DOT) {\n                    res = res.slice(0, -1) + '…';\n                    lastCode = C_ELLIPSIS;\n                    i++;\n                    continue;\n                }\n            }\n\n            // Trailing Wow\n            else if (code === C_WOW && lastCode === C_SPACE) {\n                res += 'و';\n                lastCode = C_WOW;\n                i++;\n                while (i < len) {\n                    const next = text.charCodeAt(i);\n                    if (next === C_SPACE || next === C_TAB) {\n                        i++;\n                    } else {\n                        break;\n                    }\n                }\n                continue;\n            }\n\n            // Repeats (Tatweel, Underscore, Dash, Asterisk)\n            else if (\n                (code === C_TATWEEL && lastCode === C_TATWEEL) ||\n                (code === C_UNDERSCORE && lastCode === C_UNDERSCORE) ||\n                (code === C_DASH && lastCode === C_DASH) ||\n                (code === C_ASTERISK && lastCode === C_ASTERISK)\n            ) {\n                i++;\n                continue;\n            }\n\n            // Redundant Punctuation\n            if ((lastCode === C_AR_Q_MARK || lastCode === C_EXCLAM) && (code === C_DOT || code === C_AR_COMMA)) {\n                i++;\n                continue;\n            }\n        }\n\n        // ---------------------------------------------------------\n        // 3. Resolve Pending Spaces\n        // ---------------------------------------------------------\n        if (pendingSpaces > 0) {\n            let shouldEmitSpace = true;\n            const currentFlags = CHAR_MAP[code];\n\n            if (currentFlags & F_NO_SPACE_BEFORE) {\n                shouldEmitSpace = false;\n            }\n\n            if (CHAR_MAP[lastCode] & F_OPENING) {\n                shouldEmitSpace = false;\n            }\n\n            // Slash logic\n            if (code === C_SLASH) {\n                // Peek next non-space\n                let nextNonSpace = 0;\n                for (let k = i + 1; k < len; k++) {\n                    const nc = text.charCodeAt(k);\n                    if (nc !== C_SPACE && nc !== C_TAB && nc !== C_NEWLINE && nc !== C_CR) {\n                        nextNonSpace = nc;\n                        break;\n                    }\n                }\n\n                if (CHAR_MAP[lastCode] & F_DIGIT && CHAR_MAP[nextNonSpace] & F_DIGIT) {\n                    shouldEmitSpace = false;\n                }\n            }\n\n            // No space after slash in references\n            if (lastCode === C_SLASH && currentFlags & F_DIGIT) {\n                if (res.length >= 2 && CHAR_MAP[res.charCodeAt(res.length - 2)] & F_DIGIT) {\n                    shouldEmitSpace = false;\n                }\n            }\n\n            if (shouldEmitSpace) {\n                res += ' ';\n                lastCode = C_SPACE;\n            }\n            pendingSpaces = 0;\n        }\n\n        // ---------------------------------------------------------\n        // 4. Insert Missing Spaces\n        // ---------------------------------------------------------\n\n        const currentFlags = CHAR_MAP[code];\n\n        // Arabic <-> Number\n        if (currentFlags & F_DIGIT && CHAR_MAP[lastCode] & F_ARABIC) {\n            res += ' ';\n            lastCode = C_SPACE;\n        }\n\n        // Space before Opening\n        if (\n            currentFlags & F_OPENING &&\n            lastCode !== C_SPACE &&\n            lastCode !== C_NEWLINE &&\n            !(CHAR_MAP[lastCode] & F_OPENING) &&\n            lastCode !== 0\n        ) {\n            res += ' ';\n            lastCode = C_SPACE;\n        }\n\n        // Space after Punctuation\n        if (CHAR_MAP[lastCode] & F_PUNCT) {\n            const isSpecial =\n                currentFlags & (F_SPACE | F_CLOSING | F_OPENING) ||\n                code === C_SPACE ||\n                code === C_NEWLINE ||\n                code === C_CR ||\n                currentFlags & F_CLOSING ||\n                code === C_DQUOTE ||\n                code === C_SQUOTE ||\n                code === C_R_GUILLEMET ||\n                code === C_RDQUOTE ||\n                code === lastCode ||\n                ((lastCode === C_AR_Q_MARK || lastCode === C_EXCLAM) && (code === C_DOT || code === C_AR_COMMA));\n\n            if (!isSpecial) {\n                res += ' ';\n                lastCode = C_SPACE;\n            }\n        }\n\n        // ---------------------------------------------------------\n        // 5. Emit\n        // ---------------------------------------------------------\n        if (code !== originalCode) {\n            res += String.fromCharCode(code);\n        } else {\n            res += text[i];\n        }\n        lastCode = code;\n        i++;\n    }\n\n    return res.trim();\n};\n\n/**\n * Preformat using a growable UTF-16 buffer to reduce intermediate allocations.\n *\n * This is primarily intended for experimentation on extremely large inputs where GC pressure\n * may dominate (e.g. 100MB+ strings). It must remain byte-for-byte identical to the concat path.\n *\n * @param text Input string\n * @returns Preformatted string\n */\nconst processStringBuffer = (text: string): string => {\n    if (!text) {\n        return '';\n    }\n\n    const len = text.length;\n    let i = 0;\n\n    // Heuristic: output is usually close to input size; allocate a bit more to reduce growth.\n    const builder = new Utf16Builder(len + (len >> 3) + 64);\n\n    let pendingSpaces = 0;\n    let lastCode = 0;\n\n    while (i < len) {\n        let code = text.charCodeAt(i);\n        const originalCode = code;\n        const flags = CHAR_MAP[code];\n\n        // ---------------------------------------------------------\n        // 1. Handle Whitespace\n        // ---------------------------------------------------------\n        if (flags & F_SPACE) {\n            pendingSpaces++;\n            i++;\n            continue;\n        }\n\n        // ---------------------------------------------------------\n        // 2. Handle Special Characters (Newlines, Transforms)\n        // ---------------------------------------------------------\n        if (flags & F_SPECIAL) {\n            // Newlines\n            if (code === C_NEWLINE || code === C_CR) {\n                pendingSpaces = 0;\n                if (lastCode !== C_NEWLINE) {\n                    builder.push(C_NEWLINE);\n                    lastCode = C_NEWLINE;\n                }\n                // Skip subsequent whitespace\n                i++;\n                while (i < len) {\n                    const next = text.charCodeAt(i);\n                    if (next === C_SPACE || next === C_TAB || next === C_NEWLINE || next === C_CR) {\n                        i++;\n                    } else {\n                        break;\n                    }\n                }\n                continue;\n            }\n\n            // Transforms\n            if (code === C_Q_MARK) {\n                code = C_AR_Q_MARK;\n            } else if (code === C_SEMICOLON) {\n                code = C_AR_SEMICOLON;\n            } else if (code === C_COMMA) {\n                code = C_AR_COMMA;\n            }\n\n            // Condense Colons (.:. -> :)\n            else if (code === C_COLON) {\n                if (lastCode === C_DOT || lastCode === C_DASH) {\n                    builder.pop();\n                    lastCode = builder.last();\n                }\n                const next = text.charCodeAt(i + 1);\n                if (next === C_DOT || next === C_DASH) {\n                    i++;\n                }\n            }\n\n            // Double Brackets\n            else if (code === C_L_PAREN && text.charCodeAt(i + 1) === C_L_PAREN) {\n                code = C_L_GUILLEMET; // « (\\u00AB)\n                i++;\n            } else if (code === C_R_PAREN && text.charCodeAt(i + 1) === C_R_PAREN) {\n                code = C_R_GUILLEMET; // » (\\u00BB)\n                i++;\n            }\n\n            // Ellipsis (...)\n            else if (code === C_DOT) {\n                if (lastCode === C_ELLIPSIS) {\n                    i++;\n                    continue;\n                }\n                if (lastCode === C_DOT) {\n                    builder.pop();\n                    builder.push(C_ELLIPSIS);\n                    lastCode = C_ELLIPSIS;\n                    i++;\n                    continue;\n                }\n            }\n\n            // Trailing Wow\n            else if (code === C_WOW && lastCode === C_SPACE) {\n                builder.push(C_WOW);\n                lastCode = C_WOW;\n                i++;\n                while (i < len) {\n                    const next = text.charCodeAt(i);\n                    if (next === C_SPACE || next === C_TAB) {\n                        i++;\n                    } else {\n                        break;\n                    }\n                }\n                continue;\n            }\n\n            // Repeats (Tatweel, Underscore, Dash, Asterisk)\n            else if (\n                (code === C_TATWEEL && lastCode === C_TATWEEL) ||\n                (code === C_UNDERSCORE && lastCode === C_UNDERSCORE) ||\n                (code === C_DASH && lastCode === C_DASH) ||\n                (code === C_ASTERISK && lastCode === C_ASTERISK)\n            ) {\n                i++;\n                continue;\n            }\n\n            // Redundant Punctuation\n            if ((lastCode === C_AR_Q_MARK || lastCode === C_EXCLAM) && (code === C_DOT || code === C_AR_COMMA)) {\n                i++;\n                continue;\n            }\n        }\n\n        // ---------------------------------------------------------\n        // 3. Resolve Pending Spaces\n        // ---------------------------------------------------------\n        if (pendingSpaces > 0) {\n            let shouldEmitSpace = true;\n            const currentFlags = CHAR_MAP[code];\n\n            if (currentFlags & F_NO_SPACE_BEFORE) {\n                shouldEmitSpace = false;\n            }\n\n            if (CHAR_MAP[lastCode] & F_OPENING) {\n                shouldEmitSpace = false;\n            }\n\n            // Slash logic\n            if (code === C_SLASH) {\n                // Peek next non-space\n                let nextNonSpace = 0;\n                for (let k = i + 1; k < len; k++) {\n                    const nc = text.charCodeAt(k);\n                    if (nc !== C_SPACE && nc !== C_TAB && nc !== C_NEWLINE && nc !== C_CR) {\n                        nextNonSpace = nc;\n                        break;\n                    }\n                }\n\n                if (CHAR_MAP[lastCode] & F_DIGIT && CHAR_MAP[nextNonSpace] & F_DIGIT) {\n                    shouldEmitSpace = false;\n                }\n            }\n\n            // No space after slash in references\n            if (lastCode === C_SLASH && currentFlags & F_DIGIT) {\n                if (builder.length >= 2 && CHAR_MAP[builder.secondLast()] & F_DIGIT) {\n                    shouldEmitSpace = false;\n                }\n            }\n\n            if (shouldEmitSpace) {\n                builder.push(C_SPACE);\n                lastCode = C_SPACE;\n            }\n            pendingSpaces = 0;\n        }\n\n        // ---------------------------------------------------------\n        // 4. Insert Missing Spaces\n        // ---------------------------------------------------------\n\n        const currentFlags = CHAR_MAP[code];\n\n        // Arabic <-> Number\n        if (currentFlags & F_DIGIT && CHAR_MAP[lastCode] & F_ARABIC) {\n            builder.push(C_SPACE);\n            lastCode = C_SPACE;\n        }\n\n        // Space before Opening\n        if (\n            currentFlags & F_OPENING &&\n            lastCode !== C_SPACE &&\n            lastCode !== C_NEWLINE &&\n            !(CHAR_MAP[lastCode] & F_OPENING) &&\n            lastCode !== 0\n        ) {\n            builder.push(C_SPACE);\n            lastCode = C_SPACE;\n        }\n\n        // Space after Punctuation\n        if (CHAR_MAP[lastCode] & F_PUNCT) {\n            const isSpecial =\n                currentFlags & (F_SPACE | F_CLOSING | F_OPENING) ||\n                code === C_SPACE ||\n                code === C_NEWLINE ||\n                code === C_CR ||\n                currentFlags & F_CLOSING ||\n                code === C_DQUOTE ||\n                code === C_SQUOTE ||\n                code === C_R_GUILLEMET ||\n                code === C_RDQUOTE ||\n                code === lastCode ||\n                ((lastCode === C_AR_Q_MARK || lastCode === C_EXCLAM) && (code === C_DOT || code === C_AR_COMMA));\n\n            if (!isSpecial) {\n                builder.push(C_SPACE);\n                lastCode = C_SPACE;\n            }\n        }\n\n        // ---------------------------------------------------------\n        // 5. Emit\n        // ---------------------------------------------------------\n        if (code !== originalCode) {\n            builder.push(code);\n        } else {\n            builder.push(originalCode);\n        }\n        lastCode = code;\n        i++;\n    }\n\n    return builder.toStringTrimmed();\n};\n\n/**\n * Internal (non-public) variant: baseline `+=` builder.\n */\nexport const preformatArabicTextConcat = (text: string): string => processStringConcat(text);\n\n/**\n * Internal (non-public) variant: UTF-16 buffer builder to reduce intermediate allocations.\n */\nexport const preformatArabicTextBuffer = (text: string): string => processStringBuffer(text);\n\n\n","/**\n * Hyperoptimized Arabic text preformatting entry point.\n *\n * The implementation lives in `src/preformat-core.ts` to keep the public surface\n * area small while allowing internal benchmarking (buffer vs concat builders).\n *\n * @module preformat\n */\n\nimport { preformatArabicTextBuffer, preformatArabicTextConcat } from './preformat-core';\n\n/**\n * Preformat a single string.\n *\n * This is the internal implementation used by the public {@link preformatArabicText} API.\n * The builder can be forced for experiments via `BITABOOM_PREFORMAT_BUILDER`.\n *\n * @param text Input string\n * @returns Preformatted string\n */\nconst preformatOne = (text: string): string => {\n    // Allow forcing a builder for benchmarks/debugging without changing public API.\n    const forced = process.env.BITABOOM_PREFORMAT_BUILDER;\n    if (forced === 'concat') {\n        return preformatArabicTextConcat(text);\n    }\n    if (forced === 'buffer') {\n        return preformatArabicTextBuffer(text);\n    }\n\n    // Default: the concat builder is typically faster in Bun/V8 for common page-sized inputs.\n    // For experiments on extremely large inputs, force `BITABOOM_PREFORMAT_BUILDER=buffer`.\n    return preformatArabicTextConcat(text);\n};\n\n// ==============================================================================\n// PUBLIC API\n// ==============================================================================\n\ntype PreformatArabicText = {\n    (text: string): string;\n    (texts: string[]): string[];\n};\n\n/**\n * High-performance Arabic preformatting pipeline.\n *\n * Consolidates common formatting steps (spacing, punctuation normalization, reference formatting,\n * bracket/quote cleanup, ellipsis condensation, newline normalization) into a single-pass formatter.\n *\n * @param text Input string or an array of strings\n * @returns Preformatted string or array of strings (matching input shape)\n */\nexport const preformatArabicText: PreformatArabicText = (text: string | string[]): string | string[] => {\n    if (Array.isArray(text)) {\n        return text.map(preformatOne);\n    }\n    return preformatOne(text);\n};\n","import { escapeRegex } from './cleaning';\n\n/**\n * Removes various symbols, part references, and numerical markers from the text.\n * Example: '(1) (2/3)' becomes ''.\n * @param {string} text - The input text to apply the rule to.\n * @returns {string} - The modified text with symbols and part references removed.\n */\nexport const cleanSymbolsAndPartReferences = (text: string) => {\n    return text.replace(\n        / *\\(?:\\d+(?:\\/\\d+){0,2}\\)? *| *\\[\\d+(?:\\/\\d+)?\\] *| *«\\d+» *|\\d+\\/\\d+(?:\\/\\d+)?|[،§{}۝؍‎﴿﴾<>;_؟»«:!،؛[\\]…ـ¬.\\\\/*()\"]/g,\n        ' ',\n    );\n};\n\n/**\n * Removes trailing page numbers formatted as '-[46]-' from the text.\n * Example: 'This is some -[46]- text' becomes 'This is some text'.\n * @param {string} text - The input text with trailing page numbers.\n * @returns {string} - The modified text with page numbers removed.\n */\nexport const cleanTrailingPageNumbers = (text: string) => {\n    return text.replace(/-\\[\\d+\\]-/g, '');\n};\n\n/**\n * Replaces consecutive line breaks and whitespace characters with a single space.\n * Example: 'a\\nb' becomes 'a b'.\n * @param {string} text - The input text containing line breaks or multiple spaces.\n * @returns {string} - The modified text with spaces.\n */\nexport const replaceLineBreaksWithSpaces = (text: string) => {\n    return text.replace(/\\s+/g, ' ');\n};\n\n/**\n * Removes all numeric digits from the text.\n * Example: 'abc123' becomes 'abc'.\n * @param {string} text - The input text containing digits.\n * @returns {string} - The modified text with digits removed.\n */\nexport const stripAllDigits = (text: string) => {\n    return text.replace(/[0-9]/g, '');\n};\n\n/**\n * Removes death year references like \"(d. 390H)\" and \"[d. 100h]\" from the text.\n * Example: 'Sufyān ibn ‘Uyaynah (d. 198h)' becomes 'Sufyān ibn ‘Uyaynah'.\n * @param {string} text - The input text containing death year references.\n * @returns {string} - The modified text with death years removed.\n */\nexport const removeDeathYear = (text: string) => {\n    return text.replace(/\\[(d)\\.\\s*\\d{1,4}[hH]\\]\\s*|\\((d)\\.\\s*\\d{1,4}[hH]\\)\\s*/g, '');\n};\n\n/**\n * Removes numeric digits and dashes from the text.\n * Example: 'ABC 123-Xyz' becomes 'ABC Xyz'.\n * @param {string} text - The input text containing digits and dashes.\n * @returns {string} - The modified text with numbers and dashes removed.\n */\nexport const removeNumbersAndDashes = (text: string) => {\n    return text.replace(/[\\d-]/g, '');\n};\n\n/**\n * Removes single digit references like (1), «2», [3] from the text.\n * Example: 'Ref (1), Ref «2», Ref [3]' becomes 'Ref , Ref , Ref '.\n * @param {string} text - The input text containing single digit references.\n * @returns {string} - The modified text with single digit references removed.\n */\nexport const removeSingleDigitReferences = (text: string) => {\n    return text.replace(/\\(\\d{1}\\)|\\[\\d{1}\\]|«\\d»/g, '');\n};\n\n/**\n * Removes URLs from the text.\n * Example: 'Visit https://example.com' becomes 'Visit '.\n * @param {string} text - The input text containing URLs.\n * @returns {string} - The modified text with URLs removed.\n */\nexport const removeUrls = (text: string) => {\n    return text.replace(\n        /https?:\\/\\/(www\\.)?[-a-zA-Z0-9@:%._+~#=]{1,256}\\.[a-zA-Z0-9()]{1,6}\\b([-a-zA-Z0-9()@:%_+.~#?&//=]*)/g,\n        '',\n    );\n};\n\n/**\n * Removes common Markdown formatting syntax from text\n * @param text - The input text containing Markdown formatting\n * @returns Text with Markdown formatting removed (bold, italics, headers, lists, backticks)\n */\nexport const removeMarkdownFormatting = (text: string) => {\n    return (\n        text\n            // Remove bold first (**text**) - must come before italics\n            .replace(/\\*\\*([^*]+)\\*\\*/g, '$1')\n            // Remove bold with underscores (__text__)\n            .replace(/__([^_]+)__/g, '$1')\n            // Remove italics (*text*)\n            .replace(/\\*([^*]+)\\*/g, '$1')\n            // Remove italics with underscores (_text_)\n            .replace(/_([^_]+)_/g, '$1')\n            // Remove strikethrough (~~text~~)\n            .replace(/~~([^~]+)~~/g, '$1')\n            // Remove blockquotes\n            .replace(/^\\s*>\\s?/gm, '')\n            // Remove images ![alt](url)\n            .replace(/!\\[[^\\]]*]\\([^)]*\\)/g, '')\n            // Convert links [text](url) -> text\n            .replace(/\\[([^\\]]+)]\\([^)]*\\)/g, '$1')\n            // Remove headers (# ## ### etc.)\n            .replace(/^#+\\s*/gm, '')\n            // Remove unordered list markers (- * +)\n            .replace(/^\\s*[-*+]\\s+/gm, '')\n            // Remove ordered list markers (1. 2. etc.)\n            .replace(/^\\s*\\d+\\.\\s+/gm, '')\n            // Remove backticks\n            .replace(/`/gm, '')\n    );\n};\n\n/**\n * Truncates a string to a specified length, adding an ellipsis if truncated.\n *\n * @param val - The string to truncate\n * @param n - Maximum length of the string (default: 150)\n * @returns The truncated string with ellipsis if needed, otherwise the original string\n *\n * @example\n * ```javascript\n * truncate('The quick brown fox jumps over the lazy dog', 20);\n * // Output: 'The quick brown fox…'\n *\n * truncate('Short text', 50);\n * // Output: 'Short text'\n * ```\n */\nexport const truncate = (val: string, n = 150): string => (val.length > n ? `${val.substring(0, n - 1)}…` : val);\n\n/**\n * Truncates a string from the middle, preserving both the beginning and end portions.\n *\n * @param text - The string to truncate\n * @param maxLength - Maximum length of the resulting string (default: 50)\n * @param endLength - Number of characters to preserve at the end (default: 1/3 of maxLength, minimum 3)\n * @returns The truncated string with ellipsis in the middle if needed, otherwise the original string\n *\n * @example\n * ```javascript\n * truncateMiddle('The quick brown fox jumps right over the lazy dog', 20);\n * // Output: 'The quick bro…zy dog'\n *\n * truncateMiddle('The quick brown fox jumps right over the lazy dog', 25, 8);\n * // Output: 'The quick brown …lazy dog'\n *\n * truncateMiddle('Short text', 50);\n * // Output: 'Short text'\n * ```\n */\nexport const truncateMiddle = (text: string, maxLength: number = 50, endLength?: number) => {\n    if (text.length <= maxLength) {\n        return text;\n    }\n\n    // Default end length is roughly 1/3 of max length, minimum 3 characters\n    const defaultEndLength = Math.max(3, Math.floor(maxLength / 3));\n    const actualEndLength = endLength ?? defaultEndLength;\n\n    // Reserve space for the ellipsis character (1 char)\n    const availableLength = maxLength - 1;\n\n    // Calculate start length (remaining space after end portion)\n    const startLength = availableLength - actualEndLength;\n\n    // Ensure we have at least some characters at the start\n    if (startLength < 1) {\n        // If we can't fit both start and end, just truncate normally\n        return `${text.substring(0, maxLength - 1)}…`;\n    }\n\n    const startPortion = text.substring(0, startLength);\n    const endPortion = text.substring(text.length - actualEndLength);\n\n    return `${startPortion}…${endPortion}`;\n};\n\n/**\n * Unescapes backslash-escaped spaces and trims whitespace from both ends.\n * Commonly used to clean file paths that have been escaped when pasted into terminals.\n *\n * @param input - The string to unescape and clean\n * @returns The cleaned string with escaped spaces converted to regular spaces and trimmed\n *\n * @example\n * ```javascript\n * unescapeSpaces('My\\\\ Folder\\\\ Name');\n * // Output: 'My Folder Name'\n *\n * unescapeSpaces('  /path/to/My\\\\ Document.txt  ');\n * // Output: '/path/to/My Document.txt'\n *\n * unescapeSpaces('regular text');\n * // Output: 'regular text'\n * ```\n */\nexport const unescapeSpaces = (input: string) => input.replace(/\\\\ /g, ' ').trim();\n\n/**\n * Arabic diacritics (Tashkeel/Harakat).\n */\nconst DIACRITICS_CLASS = '[\\u064B\\u064C\\u064D\\u064E\\u064F\\u0650\\u0651\\u0652]';\n\n/**\n * Groups of equivalent Arabic characters — any character in a group should match\n * any other character in the same group.\n */\nconst EQUIV_GROUPS: string[][] = [\n    ['\\u0627', '\\u0622', '\\u0623', '\\u0625'], // ا, آ, أ, إ\n    ['\\u0629', '\\u0647'], // ة <-> ه\n    ['\\u0649', '\\u064A'], // ى <-> ي\n];\n\n/**\n * Return an escaped character class representing all equivalents for the given character.\n *\n * If the character belongs to one of the predefined equivalence groups (e.g. ا/آ/أ/إ),\n * the returned class will match any member of that group. Otherwise, the original\n * character is simply escaped for safe inclusion in a regular expression.\n *\n * @param ch - A single character to expand into its equivalence class\n * @returns A RegExp-safe string representing the character (and its equivalents when applicable)\n */\nconst getEquivClass = (ch: string): string => {\n    for (const group of EQUIV_GROUPS) {\n        if (group.includes(ch)) {\n            // join the group's members into a character class\n            return `[${group.map((c) => escapeRegex(c)).join('')}]`;\n        }\n    }\n    // not in equivalence groups -> return escaped character\n    return escapeRegex(ch);\n};\n\n/** Small safe normalization: NFC, remove ZWJ/ZWNJ, collapse spaces. */\nconst normalizeArabicLight = (str: string) => {\n    return str\n        .normalize('NFC')\n        .replace(/[\\u200C\\u200D]/g, '') // remove ZWJ/ZWNJ\n        .replace(/\\s+/g, ' ')\n        .trim();\n};\n\n/**\n * Creates a diacritic-insensitive regex pattern for Arabic text matching.\n * Normalizes text, handles character equivalences (ا/آ/أ/إ, ة/ه, ى/ي),\n * and makes each character tolerant of Arabic diacritics (Tashkeel/Harakat)\n * @param text - Input Arabic text to make diacritic-insensitive\n * @returns Regex pattern string that matches the text with or without diacritics and character variants\n */\nexport const makeDiacriticInsensitive = (text: string) => {\n    const diacriticsMatcher = `${DIACRITICS_CLASS}*`;\n    const norm = normalizeArabicLight(text);\n    // Use Array.from to iterate grapheme-safe over the string (works fine for Arabic letters)\n    return Array.from(norm)\n        .map((ch) => getEquivClass(ch) + diacriticsMatcher)\n        .join('');\n};\n","import { normalizeSpaces } from './formatting';\n\n/**\n * Replaces common Arabic prefixes (like 'Al-', 'Ar-', 'Ash-', etc.) with 'al-' in the text.\n * Handles different variations of prefixes such as Ash- and Al- but not when the second word\n * does not start with 'S'.\n * Example: 'Ash-Shafiee' becomes 'al-Shafiee'.\n *\n * @param {string} text - The input text containing Arabic prefixes.\n * @returns {string} - The modified text with standardized 'al-' prefixes.\n */\nexport const normalizeArabicPrefixesToAl = (text: string) => {\n    return text\n        .replace(/(\\b|\\W)(Al |Al-|Ar-|As-|Adh-|Ad-|Ats-|Ath |Ath-|Az |Az-|az-|adh-|as-|ar-)/g, '$1al-')\n        .replace(/(\\b|\\W)(Ash-S|ash-S)/g, '$1al-S')\n        .replace(/al- (.+?)\\b/g, 'al-$1');\n};\n\n/**\n * Removes double occurrences of Arabic apostrophes such as ʿʿ or ʾʾ in the text.\n * Example: 'ʿulamāʾʾ' becomes 'ʿulamāʾ'.\n *\n * @param {string} text - The input text containing double apostrophes.\n * @returns {string} - The modified text with condensed apostrophes.\n */\nexport const normalizeDoubleApostrophes = (text: string) => {\n    return text.replace(/ʿʿ/g, 'ʿ').replace(/ʾʾ/g, 'ʾ');\n};\n\n/**\n * Replaces common salutations such as \"sallahu alayhi wasallam\" with \"ﷺ\" in the text.\n * It also handles variations of the salutation phrase, including 'peace and blessings be upon him'.\n * Example: 'Then Muḥammad (sallahu alayhi wasallam)' becomes 'Then Muḥammad ﷺ'.\n *\n * @param {string} text - The input text containing salutations.\n * @returns {string} - The modified text with salutations replaced.\n */\nexport const replaceSalutationsWithSymbol = (text: string) => {\n    return text\n        .replace(\n            /\\(peace be upon him\\)|(Messenger of (Allah|Allāh)|Messenger|Prophet|Mu[hḥ]ammad) *\\((s[^)]*m|peace[^)]*him|May[^)]*him|may[^)]*him)\\)*/gi,\n            '$1 ﷺ',\n        )\n        .replace(/,\\s*ﷺ\\s*,/g, ' ﷺ');\n};\n\n/**\n * Normalizes the text by removing diacritics, apostrophes, and dashes.\n * Example: 'Al-Jadwal' becomes 'AlJadwal'.\n *\n * @param {string} input - The input text to normalize.\n * @returns {string} - The normalized text.\n */\nexport const normalize = (input: string) => {\n    return input\n        .normalize('NFKD')\n        .replace(/[\\u0300-\\u036f]/g, '')\n        .replace(/`|ʾ|ʿ|-/g, '');\n};\n\n/**\n * Strips common Arabic prefixes like 'al-', 'bi-', 'fī', 'wa-', etc. from the beginning of words.\n * Example: 'al-Bukhari' becomes 'Bukhari'.\n *\n * @param {string} text - The input text containing Arabic prefixes.\n * @returns {string} - The modified text with prefixes stripped.\n */\nexport const removeArabicPrefixes = (text: string) => {\n    return normalizeSpaces(text.replace(/(\\bal-|\\bli-|\\bbi-|\\bfī|\\bwa[-\\s]+|\\bl-|\\bliʿl|\\Bʿalá|\\Bʿan|\\bb\\.)/gi, ''));\n};\n\n/**\n * Simplifies English transliterations by removing diacritics, apostrophes, and common prefixes.\n * Example: 'Al-Jadwal' becomes 'Jadwal', and 'āḍġḥīṣṭū' becomes 'adghistu'.\n *\n * @param {string} text - The input text to simplify.\n * @returns {string} - The simplified text.\n */\nexport const normalizeTransliteratedEnglish = (text: string) => normalize(removeArabicPrefixes(text));\n\n/**\n * Extracts the initials from the input string, typically used for names or titles.\n * Example: 'Nayl al-Awtar' becomes 'NA'.\n *\n * @param {string} text - The input text to extract initials from.\n * @returns {string} - The extracted initials.\n */\nexport const extractInitials = (fullName: string) => {\n    const initials = normalizeTransliteratedEnglish(fullName)\n        .trim()\n        .split(/[ -]/)\n        .slice(0, 2)\n        .map((word) => {\n            return word.charAt(0).toUpperCase();\n        })\n        .join('');\n    return initials;\n};\n"],"mappings":"AACA,MAAa,EAAgC,WCwBhC,EAAyB,GAC3B,SACH,EAAO,QAAQ,mBAAqB,IAAO,EAAE,WAAW,EAAE,CAAG,MAAQ,UAAU,CAAC,CAChF,GACH,CAUQ,EAAiC,GACnC,EAAK,QAAQ,qCAAsC,GAAG,CASpD,EAA8B,GAChC,EAAK,QAAQ,KAAM,IAAI,CAAC,QAAQ,KAAM,IAAI,CASxC,EAAkB,GAAiB,CAC5C,GAAI,CAAC,EACD,MAAO,GAGX,IAAM,EAAuB,uEAEvB,EAAkB,mCAElB,EAAsB,sCACtB,EAAU,EAAK,QAAQ,EAAiB,GAAG,CAC3C,EAAgB,EAAQ,MAAM,EAAqB,EAAI,EAAE,CACzD,EAAe,EAAQ,MAAM,EAAoB,EAAI,EAAE,CAC7D,OAAO,EAAa,SAAW,EAAI,EAAI,EAAc,OAAS,EAAa,QAoBlE,EAAuB,GAAiB,CACjD,IAAK,IAAI,EAAI,EAAK,OAAS,EAAG,GAAK,EAAG,IAClC,GAAI,EAA8B,KAAK,EAAK,GAAG,CAC3C,OAAO,EAIf,MAAO,IAUE,EAAkB,GACpB,EAAK,QAAQ,OAAQ,KAAK,CASxB,EAAuC,GACzC,EAAK,QAAQ,2BAA4B,QAAQ,CAS/C,EAA4B,GAC9B,EACF,QAAQ,iEAAkE,IAAI,CAC9E,QAAQ,8DAA+D,IAAI,CASvE,EAAuB,GACzB,EAAK,QAAQ,2CAA4C,GAAG,CAS1D,EAA+B,GACjC,EAAK,QAAQ,6BAA8B,IAAI,CAS7C,EAAuC,GACzC,EACF,QAAQ,UAAW,IAAI,CACvB,QAAQ,oBAAqB,IAAI,CACjC,QAAQ,QAAS,IAAI,CC1JjB,EAAe,GAAsB,EAAE,QAAQ,sBAAuB,OAAO,CA+D7E,GAAiC,EAAgB,EAAyB,EAAE,GAAa,CAClG,GAAM,CACF,eAAe,CAAE,KAAM,GAAM,aAAc,GAAM,eAAgB,GAAM,CACvE,eAAe,GACf,mBAAmB,GACnB,iBAAiB,GACjB,QAAQ,KACR,EAGJ,GAAI,EAAO,OAAS,IAChB,MAAU,MAAM,iDAAiD,CAGrE,IAAM,EAAa,GAAuB,CACtC,OAAQ,EAAR,CACI,IAAK,IACL,IAAK,IACL,IAAK,IACL,IAAK,IACD,OAAO,EAAa,KAAO,SAAW,IAC1C,IAAK,IACL,IAAK,IACD,OAAO,EAAa,aAAe,OAAS,EAAY,EAAG,CAC/D,IAAK,IACL,IAAK,IACD,OAAO,EAAa,eAAiB,OAAS,EAAY,EAAG,CACjE,QACI,OAAO,EAAY,EAAG,GAI5B,EAAQ,GAAG,EAAmB,0DAAyB,KAAK,EAAe,WAAsB,KAEnG,EAAU,GACd,IAAK,IAAM,KAAM,MAAM,KAAK,EAAO,CAC3B,KAAK,KAAK,EAAG,CACb,GAAW,EAAiB,OAAS,OAErC,GAAW,GAAG,EAAU,EAAG,GAAG,IAItC,OAAO,IAAI,OAAO,EAAS,EAAM,EAYxB,GAAiB,GAAoB,EAAQ,QAAQ,WAAY,GAAG,CCxHpE,GAAoC,GAKvB,EAAK,QAHP,YAG4B;EAAO,CAAC,QAAQ,SAAU;EAAK,CAAC,MAAM,CAW7E,GAAqC,GACvC,EACF,QAAQ,iDAAkD,QAAQ,CAClE,QAAQ,mCAAoC,OAAO,CACnD,QAAQ,uDAAwD,SAAS,CACzE,QAAQ,yEAA0E,KAAK,CAUnF,GAAoB,GACtB,EACF,QAAQ,QAAS,IAAI,CACrB,QAAQ,aAAc,OAAO,CAC7B,QAAQ,MAAO,IAAI,CASf,GAAwB,GAC1B,EAAK,QAAQ,UAAW;EAAK,CAS3B,GAAmB,GACrB,EAAK,QAAQ,YAAa,GAAG,CAQ3B,GAAuB,GACzB,gBAAgB,KAAK,EAAK,CAQxB,GAAqB,GAChB,kEACD,KAAK,EAAK,CASd,GAA2B,GAC7B,EAAK,QAAQ,mBAAoB,KAAK,CASpC,GAAqB,GACvB,EAAK,QAAQ,YAAa,IAAI,CAS5B,GAAkB,GACpB,EAAK,QAAQ,eAAgB,IAAI,CAS/B,GAAkB,GACpB,EAAK,QAAQ,SAAU,IAAI,CASzB,GAAoB,GACtB,EAAK,QAAQ,UAAW,IAAI,CAS1B,GAAiC,GACnC,EAAK,QAAQ,eAAgB;;EAAO,CASlC,GAAiC,GACnC,EAAK,QAAQ,eAAgB;EAAK,CAShC,GAAmB,GACrB,EAAK,QAAQ,UAAW,IAAI,CAS1B,GAAuB,GACzB,EAAK,QAAQ,SAAU,IAAI,CAAC,QAAQ,MAAO,IAAI,CAS7C,EAA0B,GAC5B,EAAK,QAAQ,2BAA4B,OAAO,CAS9C,EAA6B,GAC/B,EAAK,QAAQ,qBAAsB,QAAQ,CASzC,EAA2B,GAC7B,EAAK,QAAQ,mBAAoB,QAAQ,CASvC,EAAmB,GAExB,EACK,QAAQ,aAAc,IAAI,CAC1B,QAAQ,aAAc,IAAI,CAE1B,QAAQ,6BAA8B,OAAO,CAE7C,QAAQ,6BAA8B,OAAO,CAU7C,EAAkB,GAAiB,CAE5C,IAAI,EAAS,EAMb,MAHA,GAAS,EAAO,QAAQ,kBAAmB,OAAO,CAG3C,EAAO,QAAQ,kBAAmB,OAAO,EASvC,EAA+B,GAEpC,EAEK,QAAQ,eAAgB,OAAO,CAE/B,QAAQ,eAAgB,OAAO,CAE/B,QAAQ,sBAAuB,OAAO,CAWtC,EAA0B,GAAkB,CACrD,IAAM,EAAgB,mCAChBC,EAAsB,EAAE,CACxB,EAAQ,EAAM,MAAM;EAAK,CAC3B,EAAkB,GA4BtB,OA1BA,EAAM,QAAS,GAAS,CACpB,IAAM,EAAc,EAAK,MAAM,CACzB,EAAa,EAAc,KAAK,EAAY,CAC5C,EAAW,gBAAgB,KAAK,EAAY,CAElD,GAAI,GAAc,CAAC,EACf,AAEI,KADA,EAAU,KAAK,EAAgB,MAAM,CAAC,CACpB,IAEtB,EAAU,KAAK,EAAY,KACxB,CACH,GAAmB,GAAG,EAAY,GAClC,IAAM,EAAW,EAAgB,MAAM,CAAC,MAAM,GAAG,CAC7C,QAAQ,KAAK,EAAS,GACtB,EAAU,KAAK,EAAgB,MAAM,CAAC,CACtC,EAAkB,MAG5B,CAGE,GACA,EAAU,KAAK,EAAgB,MAAM,CAAC,CAGnC,EAAU,KAAK;EAAK,EAQlB,EAAkB,GAAiB,CAG5C,IAAM,EAAc,EAAK,QAAQ,aAAc,GAAG,CAOlD,OAJI,EAAY,SAAW,EAChB,GAGJ,IAAgB,EAAY,aAAa,EASvC,EAA8B,GAChC,EAAK,QAAQ,sBAAuB,QAAQ,CAS1C,EAAmB,GACrB,EAAK,QAAQ,UAAW,IAAI,CAmB1B,EAA8B,GAChC,EAAK,QAAQ,cAAe,KAAK,CAS/B,EAA6B,GAC/B,EAAK,QAAQ,4BAA6B,SAAS,CASjD,EAAmC,GACrC,EAAK,QAAQ,WAAY,IAAI,CAAC,QAAQ,WAAY,IAAI,CAQpD,EAAoB,GAEJ,EAAK,UAAU,OAAO,CAGvB,QAAQ,mBAAoB,GAAG,CAAC,MAAM,CASrD,EAAuB,GAAiB,CACjD,IAAMC,EAAoC,CACtC,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAa,IACb,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,GAAgB,IAChB,KAAgB,IACnB,CAED,OAAO,EAAK,QAAQ,4EAA8E,GACvF,EAAU,IAAU,EAC7B,EAQO,EAAgB,GAClB,EAAoB,EAAiB,EAAK,CAAC,CAQzC,EAAe,GACjB,EACF,aAAa,CACb,MAAM,IAAI,CACV,IAAK,GAAS,CACX,GAAI,EAAK,SAAW,EAChB,OAAO,EAGX,IAAM,EAAQ,EAAK,MAAM,SAAS,CAClC,GAAI,CAAC,GAAS,EAAM,QAAU,IAAA,GAC1B,OAAO,EAEX,IAAM,EAAI,EAAM,MAChB,OAAO,EAAK,MAAM,EAAG,EAAE,CAAG,EAAK,OAAO,EAAE,CAAC,aAAa,CAAG,EAAK,MAAM,EAAI,EAAE,EAC5E,CACD,KAAK,IAAI,CASL,EAAyB,GAC3B,EAAK,QAAQ,+BAAgC,SAAS,CC5epD,EAAuB,GAAgB,CAChD,IAAI,EAAQ,EAAI,QAAQ,kBAAmB,OAAO,CAIlD,MAHA,GAAQ,EAAM,QAAQ,iBAAkB,SAAS,CACjD,EAAQ,EAAM,QAAQ,iBAAkB,SAAS,CAE1C,KAAK,UAAU,KAAK,MAAM,EAAM,CAAC,EAe/B,EAAwB,GAG7B,6GACmB,KAAK,EAAI,MAAM,CAAC,CAiB9B,EAAiB,IAElB,EAAM,MADA,wBACY,EAAI,EAAE,EAAE,IAAK,GAAe,EAAE,WAAW,IAAI,CAAG,EAAE,MAAM,EAAG,GAAG,CAAG,EAAG,CAiB5F,EAAqB,GAAgB,CACvC,IAAI,EAAa,EACjB,IAAK,IAAM,KAAQ,EACX,IAAS,KACT,IAGR,OAAO,EAAa,GAAM,GAGxB,EAAW,CAAE,IAAK,IAAK,IAAK,IAAK,IAAK,IAAK,CAC3C,EAAe,IAAI,IAAI,CAAC,IAAK,IAAK,IAAI,CAAC,CACvC,GAAgB,IAAI,IAAI,CAAC,IAAK,IAAK,IAAI,CAAC,CAqBxC,GAAuB,GAAgB,CACzC,IAAMC,EAAkB,EAAE,CAE1B,IAAK,IAAM,KAAQ,EACf,GAAI,EAAa,IAAI,EAAK,CACtB,EAAM,KAAK,EAAK,SACT,GAAc,IAAI,EAAK,CAAE,CAChC,IAAM,EAAW,EAAM,KAAK,CAC5B,GAAI,CAAC,GAAY,EAAS,KAAuC,EAC7D,MAAO,GAKnB,OAAO,EAAM,SAAW,GAuBf,GAAc,GAChB,EAAkB,EAAI,EAAI,GAAoB,EAAI,CAShD,GAAmB,GAAgC,CAC5D,GAAI,EAAU,SAAS,IAAI,CAAE,CACzB,GAAM,CAAC,EAAO,GAAO,EAAU,MAAM,IAAI,CAAC,IAAI,OAAO,CAErD,GAAI,EAAQ,EACR,MAAU,MAAM,6CAA6C,CAGjE,OAAO,MAAM,KAAK,CAAE,OAAQ,EAAM,EAAQ,EAAG,EAAG,EAAG,IAAM,EAAQ,EAAE,MAEnE,OAAO,EAAU,MAAM,IAAI,CAAC,IAAI,OAAO,EC9IzC,EAAW,IAAI,WAAW,MAAM,CAQhC,GAAY,EAAe,IAAkB,CAC/C,IAAK,IAAI,EAAI,EAAG,EAAI,EAAM,OAAQ,IAC9B,EAAS,EAAM,WAAW,EAAE,GAAK,GAKzC,EAAS,KAAO,EAAQ,CACxB,EAAS;IAAQ,GAAU,CAG3B,IAAK,IAAI,EAAI,GAAI,GAAK,GAAI,IACtB,EAAS,IAAM,GAInB,IAAK,IAAI,EAAI,KAAQ,GAAK,KAAQ,IAC9B,EAAS,IAAM,GAEnB,IAAK,IAAI,EAAI,KAAQ,GAAK,KAAQ,IAC9B,EAAS,IAAM,GAKnB,EAAS,SAA8B,IAAoB,CAC3D,EAAS,MAA2B,IAAoB,CAGxD,EAAS,UAAY,GAAsB,CAG3C,EAAS,UAAwB,GAA8B,CAG/D,EAAS,QAAS,GAAU,CAG5B,MAoBM,EAAY,KACZ,EAAQ,KACR,EAAa,KACb,EAAiB,KACjB,EAAc,KACd,EAAY,KACZ,EAAa,KAQb,EAAoB,GACf,IAAS,IAAW,IAAS,GAAS,IAAS,IAAa,IAAS,GAShF,IAAM,GAAN,KAAmB,CACf,OACA,OAEA,YAAmB,EAAyB,CACxC,KAAK,OAAS,IAAI,YAAY,KAAK,IAAI,GAAI,EAAgB,CAAC,CAC5D,KAAK,OAAS,EAMlB,MAAsB,CAClB,OAAO,KAAK,OAAS,EAAI,KAAK,OAAO,KAAK,OAAS,GAAK,EAM5D,YAA4B,CACxB,OAAO,KAAK,OAAS,EAAI,KAAK,OAAO,KAAK,OAAS,GAAK,EAM5D,KAAY,EAAc,CACtB,KAAK,eAAe,EAAE,CACtB,KAAK,OAAO,KAAK,QAAU,EAC3B,KAAK,SAMT,KAAa,CACL,KAAK,OAAS,GACd,KAAK,SAIb,eAAuB,EAAe,CAClC,IAAM,EAAS,KAAK,OAAS,EAC7B,GAAI,GAAU,KAAK,OAAO,OACtB,OAEJ,IAAI,EAAU,KAAK,OAAO,OAAS,EAC/B,EAAU,IACV,EAAU,GAEd,IAAM,EAAO,IAAI,YAAY,EAAQ,CACrC,EAAK,IAAI,KAAK,OAAO,SAAS,EAAG,KAAK,OAAO,CAAC,CAC9C,KAAK,OAAS,EAQlB,iBAAiC,CAC7B,IAAI,EAAQ,EACR,EAAM,KAAK,OAEf,KAAO,EAAQ,GAAO,EAAiB,KAAK,OAAO,GAAO,EACtD,IAEJ,KAAO,EAAM,GAAS,EAAiB,KAAK,OAAO,EAAM,GAAG,EACxD,IAGJ,GAAI,GAAO,EACP,MAAO,GAIX,IAAM,EAAQ,MACV,EAAM,GACV,IAAK,IAAI,EAAI,EAAO,EAAI,EAAK,GAAK,EAAO,CACrC,IAAM,EAAQ,KAAK,OAAO,SAAS,EAAG,KAAK,IAAI,EAAK,EAAI,EAAM,CAAC,CAC/D,GAAO,OAAO,aAAa,GAAG,EAAM,CAExC,OAAO,IAwBf,MAAM,GAAuB,GAAyB,CAClD,GAAI,CAAC,EACD,MAAO,GAGX,IAAI,EAAM,GACJ,EAAM,EAAK,OACb,EAAI,EAEJ,EAAgB,EAChB,EAAW,EAEf,KAAO,EAAI,GAAK,CACZ,IAAI,EAAO,EAAK,WAAW,EAAE,CACvB,EAAe,EACf,EAAQ,EAAS,GAKvB,GAAI,EAAQ,EAAS,CACjB,IACA,IACA,SAMJ,GAAI,EAAQ,GAAW,CAEnB,GAAI,IAAS,IAAa,IAAS,GAAM,CAQrC,IAPA,EAAgB,EACZ,IAAa,KACb,GAAO;EACP,EAAW,IAGf,IACO,EAAI,GAAK,CACZ,IAAM,EAAO,EAAK,WAAW,EAAE,CAC/B,GAAI,IAAS,IAAW,IAAS,GAAS,IAAS,IAAa,IAAS,GACrE,SAEA,MAGR,SAIJ,GAAI,IAAS,GACT,EAAO,UACA,IAAS,GAChB,EAAO,UACA,IAAS,GAChB,EAAO,UAIF,IAAS,GAAS,EACnB,IAAa,IAAS,IAAa,MACnC,EAAM,EAAI,MAAM,EAAG,GAAG,CACtB,EAAW,EAAI,WAAW,EAAI,OAAS,EAAE,EAAI,GAEjD,IAAM,EAAO,EAAK,WAAW,EAAI,EAAE,EAC/B,IAAS,IAAS,IAAS,KAC3B,YAKC,IAAS,IAAa,EAAK,WAAW,EAAI,EAAE,GAAK,GACtD,EAAO,IACP,YACO,IAAS,IAAa,EAAK,WAAW,EAAI,EAAE,GAAK,GACxD,EAAO,IACP,YAIK,IAAS,GAAO,CACrB,GAAI,IAAa,EAAY,CACzB,IACA,SAEJ,GAAI,IAAa,GAAO,CACpB,EAAM,EAAI,MAAM,EAAG,GAAG,CAAG,IACzB,EAAW,EACX,IACA,kBAKC,IAAS,GAAS,IAAa,GAAS,CAI7C,IAHA,GAAO,IACP,EAAW,EACX,IACO,EAAI,GAAK,CACZ,IAAM,EAAO,EAAK,WAAW,EAAE,CAC/B,GAAI,IAAS,IAAW,IAAS,EAC7B,SAEA,MAGR,iBAKC,IAAS,GAAa,IAAa,GACnC,IAAS,IAAgB,IAAa,IACtC,IAAS,IAAU,IAAa,IAChC,IAAS,IAAc,IAAa,GACvC,CACE,IACA,SAIJ,IAAK,IAAa,GAAe,IAAa,MAAc,IAAS,IAAS,IAAS,GAAa,CAChG,IACA,UAOR,GAAI,EAAgB,EAAG,CACnB,IAAI,EAAkB,GAChBC,EAAe,EAAS,GAW9B,GATIA,EAAe,IACf,EAAkB,IAGlB,EAAS,GAAY,IACrB,EAAkB,IAIlB,IAAS,GAAS,CAElB,IAAI,EAAe,EACnB,IAAK,IAAI,EAAI,EAAI,EAAG,EAAI,EAAK,IAAK,CAC9B,IAAM,EAAK,EAAK,WAAW,EAAE,CAC7B,GAAI,IAAO,IAAW,IAAO,GAAS,IAAO,IAAa,IAAO,GAAM,CACnE,EAAe,EACf,OAIJ,EAAS,GAAY,IAAW,EAAS,GAAgB,KACzD,EAAkB,IAKtB,IAAa,IAAWA,EAAe,IACnC,EAAI,QAAU,GAAK,EAAS,EAAI,WAAW,EAAI,OAAS,EAAE,EAAI,KAC9D,EAAkB,IAItB,IACA,GAAO,IACP,EAAW,IAEf,EAAgB,EAOpB,IAAM,EAAe,EAAS,GAG1B,EAAe,IAAW,EAAS,GAAY,KAC/C,GAAO,IACP,EAAW,IAKX,EAAe,GACf,IAAa,IACb,IAAa,IACb,EAAE,EAAS,GAAY,IACvB,IAAa,IAEb,GAAO,IACP,EAAW,IAIX,EAAS,GAAY,MAEjB,EAA0B,IAC1B,IAAS,IACT,IAAS,IACT,IAAS,IACT,EAAe,GACf,IAAS,IACT,IAAS,IACT,IAAS,KACT,IAAS,GACT,IAAS,IACP,IAAa,GAAe,IAAa,MAAc,IAAS,IAAS,IAAS,KAGpF,GAAO,IACP,EAAW,KAOf,IAAS,EAGT,GAAO,EAAK,GAFZ,GAAO,OAAO,aAAa,EAAK,CAIpC,EAAW,EACX,IAGJ,OAAO,EAAI,MAAM,EAYf,GAAuB,GAAyB,CAClD,GAAI,CAAC,EACD,MAAO,GAGX,IAAM,EAAM,EAAK,OACb,EAAI,EAGF,EAAU,IAAI,GAAa,GAAO,GAAO,GAAK,GAAG,CAEnD,EAAgB,EAChB,EAAW,EAEf,KAAO,EAAI,GAAK,CACZ,IAAI,EAAO,EAAK,WAAW,EAAE,CACvB,EAAe,EACf,EAAQ,EAAS,GAKvB,GAAI,EAAQ,EAAS,CACjB,IACA,IACA,SAMJ,GAAI,EAAQ,GAAW,CAEnB,GAAI,IAAS,IAAa,IAAS,GAAM,CAQrC,IAPA,EAAgB,EACZ,IAAa,KACb,EAAQ,KAAK,GAAU,CACvB,EAAW,IAGf,IACO,EAAI,GAAK,CACZ,IAAM,EAAO,EAAK,WAAW,EAAE,CAC/B,GAAI,IAAS,IAAW,IAAS,GAAS,IAAS,IAAa,IAAS,GACrE,SAEA,MAGR,SAIJ,GAAI,IAAS,GACT,EAAO,UACA,IAAS,GAChB,EAAO,UACA,IAAS,GAChB,EAAO,UAIF,IAAS,GAAS,EACnB,IAAa,IAAS,IAAa,MACnC,EAAQ,KAAK,CACb,EAAW,EAAQ,MAAM,EAE7B,IAAM,EAAO,EAAK,WAAW,EAAI,EAAE,EAC/B,IAAS,IAAS,IAAS,KAC3B,YAKC,IAAS,IAAa,EAAK,WAAW,EAAI,EAAE,GAAK,GACtD,EAAO,IACP,YACO,IAAS,IAAa,EAAK,WAAW,EAAI,EAAE,GAAK,GACxD,EAAO,IACP,YAIK,IAAS,GAAO,CACrB,GAAI,IAAa,EAAY,CACzB,IACA,SAEJ,GAAI,IAAa,GAAO,CACpB,EAAQ,KAAK,CACb,EAAQ,KAAK,EAAW,CACxB,EAAW,EACX,IACA,kBAKC,IAAS,GAAS,IAAa,GAAS,CAI7C,IAHA,EAAQ,KAAK,EAAM,CACnB,EAAW,EACX,IACO,EAAI,GAAK,CACZ,IAAM,EAAO,EAAK,WAAW,EAAE,CAC/B,GAAI,IAAS,IAAW,IAAS,EAC7B,SAEA,MAGR,iBAKC,IAAS,GAAa,IAAa,GACnC,IAAS,IAAgB,IAAa,IACtC,IAAS,IAAU,IAAa,IAChC,IAAS,IAAc,IAAa,GACvC,CACE,IACA,SAIJ,IAAK,IAAa,GAAe,IAAa,MAAc,IAAS,IAAS,IAAS,GAAa,CAChG,IACA,UAOR,GAAI,EAAgB,EAAG,CACnB,IAAI,EAAkB,GAChBA,EAAe,EAAS,GAW9B,GATIA,EAAe,IACf,EAAkB,IAGlB,EAAS,GAAY,IACrB,EAAkB,IAIlB,IAAS,GAAS,CAElB,IAAI,EAAe,EACnB,IAAK,IAAI,EAAI,EAAI,EAAG,EAAI,EAAK,IAAK,CAC9B,IAAM,EAAK,EAAK,WAAW,EAAE,CAC7B,GAAI,IAAO,IAAW,IAAO,GAAS,IAAO,IAAa,IAAO,GAAM,CACnE,EAAe,EACf,OAIJ,EAAS,GAAY,IAAW,EAAS,GAAgB,KACzD,EAAkB,IAKtB,IAAa,IAAWA,EAAe,IACnC,EAAQ,QAAU,GAAK,EAAS,EAAQ,YAAY,EAAI,KACxD,EAAkB,IAItB,IACA,EAAQ,KAAK,GAAQ,CACrB,EAAW,IAEf,EAAgB,EAOpB,IAAM,EAAe,EAAS,GAG1B,EAAe,IAAW,EAAS,GAAY,KAC/C,EAAQ,KAAK,GAAQ,CACrB,EAAW,IAKX,EAAe,GACf,IAAa,IACb,IAAa,IACb,EAAE,EAAS,GAAY,IACvB,IAAa,IAEb,EAAQ,KAAK,GAAQ,CACrB,EAAW,IAIX,EAAS,GAAY,MAEjB,EAA0B,IAC1B,IAAS,IACT,IAAS,IACT,IAAS,IACT,EAAe,GACf,IAAS,IACT,IAAS,IACT,IAAS,KACT,IAAS,GACT,IAAS,IACP,IAAa,GAAe,IAAa,MAAc,IAAS,IAAS,IAAS,KAGpF,EAAQ,KAAK,GAAQ,CACrB,EAAW,KAOf,IAAS,EAGT,EAAQ,KAAK,EAAa,CAF1B,EAAQ,KAAK,EAAK,CAItB,EAAW,EACX,IAGJ,OAAO,EAAQ,iBAAiB,EAMvB,EAA6B,GAAyB,GAAoB,EAAK,CAK/E,GAA6B,GAAyB,GAAoB,EAAK,CC5qBtF,EAAgB,GAAyB,CAE3C,IAAM,EAAS,QAAQ,IAAI,2BAU3B,OATI,IAAW,SACJ,EAA0B,EAAK,CAEtC,IAAW,SACJ,GAA0B,EAAK,CAKnC,EAA0B,EAAK,EAqB7BC,GAA4C,GACjD,MAAM,QAAQ,EAAK,CACZ,EAAK,IAAI,EAAa,CAE1B,EAAa,EAAK,CCjDhB,GAAiC,GACnC,EAAK,QACR,wHACA,IACH,CASQ,GAA4B,GAC9B,EAAK,QAAQ,aAAc,GAAG,CAS5B,GAA+B,GACjC,EAAK,QAAQ,OAAQ,IAAI,CASvB,GAAkB,GACpB,EAAK,QAAQ,SAAU,GAAG,CASxB,GAAmB,GACrB,EAAK,QAAQ,yDAA0D,GAAG,CASxE,GAA0B,GAC5B,EAAK,QAAQ,SAAU,GAAG,CASxB,GAA+B,GACjC,EAAK,QAAQ,4BAA6B,GAAG,CAS3C,GAAc,GAChB,EAAK,QACR,uGACA,GACH,CAQQ,GAA4B,GAEjC,EAEK,QAAQ,mBAAoB,KAAK,CAEjC,QAAQ,eAAgB,KAAK,CAE7B,QAAQ,eAAgB,KAAK,CAE7B,QAAQ,aAAc,KAAK,CAE3B,QAAQ,eAAgB,KAAK,CAE7B,QAAQ,aAAc,GAAG,CAEzB,QAAQ,uBAAwB,GAAG,CAEnC,QAAQ,wBAAyB,KAAK,CAEtC,QAAQ,WAAY,GAAG,CAEvB,QAAQ,iBAAkB,GAAG,CAE7B,QAAQ,iBAAkB,GAAG,CAE7B,QAAQ,MAAO,GAAG,CAoBlB,IAAY,EAAa,EAAI,MAAiB,EAAI,OAAS,EAAI,GAAG,EAAI,UAAU,EAAG,EAAI,EAAE,CAAC,GAAK,EAsB/F,IAAkB,EAAc,EAAoB,GAAI,IAAuB,CACxF,GAAI,EAAK,QAAU,EACf,OAAO,EAIX,IAAM,EAAmB,KAAK,IAAI,EAAG,KAAK,MAAM,EAAY,EAAE,CAAC,CACzD,EAAkB,GAAa,EAM/B,EAHkB,EAAY,EAGE,EAWtC,OARI,EAAc,EAEP,GAAG,EAAK,UAAU,EAAG,EAAY,EAAE,CAAC,GAMxC,GAHc,EAAK,UAAU,EAAG,EAAY,CAG5B,GAFJ,EAAK,UAAU,EAAK,OAAS,EAAgB,IAwBvD,GAAkB,GAAkB,EAAM,QAAQ,OAAQ,IAAI,CAAC,MAAM,CAW5EC,EAA2B,CAC7B,CAAC,IAAU,IAAU,IAAU,IAAS,CACxC,CAAC,IAAU,IAAS,CACpB,CAAC,IAAU,IAAS,CACvB,CAYK,GAAiB,GAAuB,CAC1C,IAAK,IAAM,KAAS,EAChB,GAAI,EAAM,SAAS,EAAG,CAElB,MAAO,IAAI,EAAM,IAAK,GAAM,EAAY,EAAE,CAAC,CAAC,KAAK,GAAG,CAAC,GAI7D,OAAO,EAAY,EAAG,EAIpB,GAAwB,GACnB,EACF,UAAU,MAAM,CAChB,QAAQ,kBAAmB,GAAG,CAC9B,QAAQ,OAAQ,IAAI,CACpB,MAAM,CAUF,GAA4B,GAAiB,CACtD,IACM,EAAO,GAAqB,EAAK,CAEvC,OAAO,MAAM,KAAK,EAAK,CAClB,IAAK,GAAO,GAAc,EAAG,CAAG,cAAkB,CAClD,KAAK,GAAG,EChQJ,GAA+B,GACjC,EACF,QAAQ,6EAA8E,QAAQ,CAC9F,QAAQ,wBAAyB,SAAS,CAC1C,QAAQ,eAAgB,QAAQ,CAU5B,GAA8B,GAChC,EAAK,QAAQ,MAAO,IAAI,CAAC,QAAQ,MAAO,IAAI,CAW1C,GAAgC,GAClC,EACF,QACG,2IACA,OACH,CACA,QAAQ,aAAc,KAAK,CAUvB,EAAa,GACf,EACF,UAAU,OAAO,CACjB,QAAQ,mBAAoB,GAAG,CAC/B,QAAQ,WAAY,GAAG,CAUnB,EAAwB,GAC1B,EAAgB,EAAK,QAAQ,uEAAwE,GAAG,CAAC,CAUvG,EAAkC,GAAiB,EAAU,EAAqB,EAAK,CAAC,CASxF,GAAmB,GACX,EAA+B,EAAS,CACpD,MAAM,CACN,MAAM,OAAO,CACb,MAAM,EAAG,EAAE,CACX,IAAK,GACK,EAAK,OAAO,EAAE,CAAC,aAAa,CACrC,CACD,KAAK,GAAG"}

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "bitaboom",
     "description": "Use string utils library to format Arabic and English translations.",
-    "version": "2.1.0",
+    "version": "2.2.0",
     "author": "Ragaeeb Haq",
     "license": "MIT",
     "private": false,
@@ -18,8 +18,8 @@
         "url": "https://github.com/ragaeeb/bitaboom/issues"
     },
     "engines": {
-        "bun": ">=1.3.2",
-        "node": ">=22.0.0"
+        "bun": ">=1.3.4",
+        "node": ">=24.0.0"
     },
     "files": [
         "dist/**"
@@ -35,11 +35,11 @@
         "arabic"
     ],
     "devDependencies": {
-        "@biomejs/biome": "^2.3.5",
-        "@types/bun": "^1.3.2",
-        "@types/node": "^24.10.1",
+        "@biomejs/biome": "^2.3.9",
+        "@types/bun": "^1.3.4",
+        "@types/node": "^25.0.3",
         "semantic-release": "^25.0.2",
-        "tsdown": "^0.16.3",
+        "tsdown": "^0.18.0",
         "typescript": "^5.9.3"
     }
 }