@nasser-sw/fabric 7.0.1-beta11 → 7.0.1-beta12

package/dist/src/text/scriptUtils.mjs

@@ -0,0 +1,212 @@
+/**
+ * Script Detection and Font Metrics Utilities
+ * 
+ * Provides script-aware text measurement for improved bounding box calculations
+ * across different writing systems like Arabic, Latin, CJK, etc.
+ */
+
+/**
+ * Unicode ranges for different scripts
+ */
+const SCRIPT_RANGES = {
+  latin: [[0x0000, 0x007F],
+  // Basic Latin
+  [0x0080, 0x00FF],
+  // Latin-1 Supplement
+  [0x0100, 0x017F],
+  // Latin Extended-A
+  [0x0180, 0x024F],
+  // Latin Extended-B
+  [0x1E00, 0x1EFF] // Latin Extended Additional
+  ],
+  arabic: [[0x0600, 0x06FF],
+  // Arabic
+  [0x0750, 0x077F],
+  // Arabic Supplement
+  [0x08A0, 0x08FF],
+  // Arabic Extended-A
+  [0xFB50, 0xFDFF],
+  // Arabic Presentation Forms-A
+  [0xFE70, 0xFEFF] // Arabic Presentation Forms-B
+  ],
+  hebrew: [[0x0590, 0x05FF],
+  // Hebrew
+  [0xFB1D, 0xFB4F] // Hebrew Presentation Forms
+  ],
+  cjk: [[0x2E80, 0x2EFF],
+  // CJK Radicals Supplement
+  [0x3000, 0x303F],
+  // CJK Symbols and Punctuation
+  [0x3400, 0x4DBF],
+  // CJK Extension A
+  [0x4E00, 0x9FFF],
+  // CJK Unified Ideographs
+  [0xF900, 0xFAFF],
+  // CJK Compatibility Ideographs
+  [0x20000, 0x2A6DF] // CJK Extension B
+  ],
+  devanagari: [[0x0900, 0x097F],
+  // Devanagari
+  [0xA8E0, 0xA8FF] // Devanagari Extended
+  ],
+  thai: [[0x0E00, 0x0E7F] // Thai
+  ]
+};
+
+/**
+ * Reference characters for font metrics by script
+ */
+const SCRIPT_REFERENCE_CHARS = {
+  latin: 'M',
+  arabic: 'ا',
+  // Arabic Letter Alef
+  hebrew: 'ם',
+  // Hebrew Letter Final Mem
+  cjk: '中',
+  // Chinese character for "middle"
+  devanagari: 'म',
+  // Devanagari Letter Ma
+  thai: 'ก',
+  // Thai Letter Ko Kai
+  mixed: 'M',
+  // Fallback to Latin
+  unknown: 'M' // Fallback to Latin
+};
+
+/**
+ * Script-specific fallback proportions for font metrics
+ */
+const SCRIPT_METRICS = {
+  latin: {
+    ascentRatio: 0.91,
+    descentRatio: 0.21,
+    actualAscentRatio: 0.716,
+    actualDescentRatio: 0.0
+  },
+  arabic: {
+    ascentRatio: 0.85,
+    // Arabic fonts typically have lower ascenders
+    descentRatio: 0.30,
+    // But deeper descenders for diacritics
+    actualAscentRatio: 0.65,
+    actualDescentRatio: 0.25
+  },
+  hebrew: {
+    ascentRatio: 0.88,
+    descentRatio: 0.25,
+    actualAscentRatio: 0.70,
+    actualDescentRatio: 0.15
+  },
+  cjk: {
+    ascentRatio: 0.88,
+    // CJK fonts are more square
+    descentRatio: 0.12,
+    actualAscentRatio: 0.80,
+    actualDescentRatio: 0.10
+  },
+  devanagari: {
+    ascentRatio: 1.0,
+    // Devanagari has tall ascenders
+    descentRatio: 0.25,
+    actualAscentRatio: 0.85,
+    actualDescentRatio: 0.15
+  },
+  thai: {
+    ascentRatio: 0.95,
+    // Thai has tall ascenders
+    descentRatio: 0.30,
+    // And some descenders
+    actualAscentRatio: 0.75,
+    actualDescentRatio: 0.20
+  },
+  mixed: {
+    ascentRatio: 0.91,
+    descentRatio: 0.21,
+    actualAscentRatio: 0.716,
+    actualDescentRatio: 0.0
+  },
+  unknown: {
+    ascentRatio: 0.91,
+    descentRatio: 0.21,
+    actualAscentRatio: 0.716,
+    actualDescentRatio: 0.0
+  }
+};
+
+/**
+ * Detect the primary script of a text string
+ */
+function detectScript(text) {
+  if (!text || text.length === 0) return 'unknown';
+  const scriptCounts = {
+    latin: 0,
+    arabic: 0,
+    hebrew: 0,
+    cjk: 0,
+    devanagari: 0,
+    thai: 0,
+    mixed: 0,
+    unknown: 0
+  };
+  for (const char of text) {
+    const codePoint = char.codePointAt(0);
+    if (codePoint === undefined) continue;
+    let detected = false;
+    for (const [script, ranges] of Object.entries(SCRIPT_RANGES)) {
+      for (const [start, end] of ranges) {
+        if (codePoint >= start && codePoint <= end) {
+          scriptCounts[script]++;
+          detected = true;
+          break;
+        }
+      }
+      if (detected) break;
+    }
+    if (!detected) {
+      scriptCounts.unknown++;
+    }
+  }
+
+  // Find the script with the highest count
+  const entries = Object.entries(scriptCounts);
+  entries.sort((a, b) => b[1] - a[1]);
+  const [topScript, topCount] = entries[0];
+  const [secondScript, secondCount] = entries[1];
+
+  // If we have significant counts in multiple scripts, consider it mixed
+  if (topCount > 0 && secondCount > topCount * 0.3) {
+    return 'mixed';
+  }
+  return topScript === 'unknown' && topCount === 0 ? 'latin' : topScript;
+}
+
+/**
+ * Get reference character for a given script
+ */
+function getScriptReferenceChar(script) {
+  return SCRIPT_REFERENCE_CHARS[script];
+}
+
+/**
+ * Get script-specific metric ratios
+ */
+function getScriptMetrics(script) {
+  return SCRIPT_METRICS[script];
+}
+
+/**
+ * Detect script for a single grapheme/character
+ */
+function detectGraphemeScript(grapheme) {
+  return detectScript(grapheme);
+}
+
+/**
+ * Check if text contains complex script characteristics
+ */
+function isComplexScript(script) {
+  return ['arabic', 'hebrew', 'devanagari', 'thai'].includes(script);
+}
+
+export { SCRIPT_METRICS, SCRIPT_REFERENCE_CHARS, detectGraphemeScript, detectScript, getScriptMetrics, getScriptReferenceChar, isComplexScript };
+//# sourceMappingURL=scriptUtils.mjs.map

package/dist/src/text/scriptUtils.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"scriptUtils.mjs","sources":["../../../src/text/scriptUtils.ts"],"sourcesContent":["/**\n * Script Detection and Font Metrics Utilities\n * \n * Provides script-aware text measurement for improved bounding box calculations\n * across different writing systems like Arabic, Latin, CJK, etc.\n */\n\nexport type ScriptType = 'latin' | 'arabic' | 'hebrew' | 'cjk' | 'devanagari' | 'thai' | 'mixed' | 'unknown';\n\n/**\n * Unicode ranges for different scripts\n */\nconst SCRIPT_RANGES = {\n  latin: [\n    [0x0000, 0x007F], // Basic Latin\n    [0x0080, 0x00FF], // Latin-1 Supplement\n    [0x0100, 0x017F], // Latin Extended-A\n    [0x0180, 0x024F], // Latin Extended-B\n    [0x1E00, 0x1EFF], // Latin Extended Additional\n  ],\n  arabic: [\n    [0x0600, 0x06FF], // Arabic\n    [0x0750, 0x077F], // Arabic Supplement\n    [0x08A0, 0x08FF], // Arabic Extended-A\n    [0xFB50, 0xFDFF], // Arabic Presentation Forms-A\n    [0xFE70, 0xFEFF], // Arabic Presentation Forms-B\n  ],\n  hebrew: [\n    [0x0590, 0x05FF], // Hebrew\n    [0xFB1D, 0xFB4F], // Hebrew Presentation Forms\n  ],\n  cjk: [\n    [0x2E80, 0x2EFF], // CJK Radicals Supplement\n    [0x3000, 0x303F], // CJK Symbols and Punctuation\n    [0x3400, 0x4DBF], // CJK Extension A\n    [0x4E00, 0x9FFF], // CJK Unified Ideographs\n    [0xF900, 0xFAFF], // CJK Compatibility Ideographs\n    [0x20000, 0x2A6DF], // CJK Extension B\n  ],\n  devanagari: [\n    [0x0900, 0x097F], // Devanagari\n    [0xA8E0, 0xA8FF], // Devanagari Extended\n  ],\n  thai: [\n    [0x0E00, 0x0E7F], // Thai\n  ],\n};\n\n/**\n * Reference characters for font metrics by script\n */\nexport const SCRIPT_REFERENCE_CHARS = {\n  latin: 'M',\n  arabic: 'ا', // Arabic Letter Alef\n  hebrew: 'ם', // Hebrew Letter Final Mem\n  cjk: '中', // Chinese character for \"middle\"\n  devanagari: 'म', // Devanagari Letter Ma\n  thai: 'ก', // Thai Letter Ko Kai\n  mixed: 'M', // Fallback to Latin\n  unknown: 'M', // Fallback to Latin\n} as const;\n\n/**\n * Script-specific fallback proportions for font metrics\n */\nexport const SCRIPT_METRICS = {\n  latin: {\n    ascentRatio: 0.91,\n    descentRatio: 0.21,\n    actualAscentRatio: 0.716,\n    actualDescentRatio: 0.0,\n  },\n  arabic: {\n    ascentRatio: 0.85, // Arabic fonts typically have lower ascenders\n    descentRatio: 0.30, // But deeper descenders for diacritics\n    actualAscentRatio: 0.65,\n    actualDescentRatio: 0.25,\n  },\n  hebrew: {\n    ascentRatio: 0.88,\n    descentRatio: 0.25,\n    actualAscentRatio: 0.70,\n    actualDescentRatio: 0.15,\n  },\n  cjk: {\n    ascentRatio: 0.88, // CJK fonts are more square\n    descentRatio: 0.12,\n    actualAscentRatio: 0.80,\n    actualDescentRatio: 0.10,\n  },\n  devanagari: {\n    ascentRatio: 1.0, // Devanagari has tall ascenders\n    descentRatio: 0.25,\n    actualAscentRatio: 0.85,\n    actualDescentRatio: 0.15,\n  },\n  thai: {\n    ascentRatio: 0.95, // Thai has tall ascenders\n    descentRatio: 0.30, // And some descenders\n    actualAscentRatio: 0.75,\n    actualDescentRatio: 0.20,\n  },\n  mixed: {\n    ascentRatio: 0.91,\n    descentRatio: 0.21,\n    actualAscentRatio: 0.716,\n    actualDescentRatio: 0.0,\n  },\n  unknown: {\n    ascentRatio: 0.91,\n    descentRatio: 0.21,\n    actualAscentRatio: 0.716,\n    actualDescentRatio: 0.0,\n  },\n} as const;\n\n/**\n * Detect the primary script of a text string\n */\nexport function detectScript(text: string): ScriptType {\n  if (!text || text.length === 0) return 'unknown';\n\n  const scriptCounts: Record<ScriptType, number> = {\n    latin: 0,\n    arabic: 0,\n    hebrew: 0,\n    cjk: 0,\n    devanagari: 0,\n    thai: 0,\n    mixed: 0,\n    unknown: 0,\n  };\n\n  for (const char of text) {\n    const codePoint = char.codePointAt(0);\n    if (codePoint === undefined) continue;\n\n    let detected = false;\n    \n    for (const [script, ranges] of Object.entries(SCRIPT_RANGES)) {\n      for (const [start, end] of ranges) {\n        if (codePoint >= start && codePoint <= end) {\n          scriptCounts[script as ScriptType]++;\n          detected = true;\n          break;\n        }\n      }\n      if (detected) break;\n    }\n\n    if (!detected) {\n      scriptCounts.unknown++;\n    }\n  }\n\n  // Find the script with the highest count\n  const entries = Object.entries(scriptCounts) as [ScriptType, number][];\n  entries.sort((a, b) => b[1] - a[1]);\n\n  const [topScript, topCount] = entries[0];\n  const [secondScript, secondCount] = entries[1];\n\n  // If we have significant counts in multiple scripts, consider it mixed\n  if (topCount > 0 && secondCount > topCount * 0.3) {\n    return 'mixed';\n  }\n\n  return topScript === 'unknown' && topCount === 0 ? 'latin' : topScript;\n}\n\n/**\n * Get reference character for a given script\n */\nexport function getScriptReferenceChar(script: ScriptType): string {\n  return SCRIPT_REFERENCE_CHARS[script];\n}\n\n/**\n * Get script-specific metric ratios\n */\nexport function getScriptMetrics(script: ScriptType) {\n  return SCRIPT_METRICS[script];\n}\n\n/**\n * Detect script for a single grapheme/character\n */\nexport function detectGraphemeScript(grapheme: string): ScriptType {\n  return detectScript(grapheme);\n}\n\n/**\n * Check if text contains complex script characteristics\n */\nexport function isComplexScript(script: ScriptType): boolean {\n  return ['arabic', 'hebrew', 'devanagari', 'thai'].includes(script);\n}\n\n/**\n * Check if script requires special glyph shaping\n */\nexport function requiresGlyphShaping(script: ScriptType): boolean {\n  return ['arabic', 'devanagari', 'thai'].includes(script);\n}\n\n/**\n * Get optimal sample text for measuring font metrics\n */\nexport function getSampleTextForScript(script: ScriptType): string {\n  const samples = {\n    latin: 'AaBbCcMmWwXxYyZz',\n    arabic: 'أابتجحدرسعفقلمنهويء',\n    hebrew: 'אבגדהוחטיכלמנעפקרשת',\n    cjk: '中文字体测试样本',\n    devanagari: 'अआइईउऊएऐओऔकगचजट',\n    thai: 'กขคงจชดตนบปพมยรลวสหอฮ',\n    mixed: 'AaBbCcMmWwXxYyZz',\n    unknown: 'AaBbCcMmWwXxYyZz',\n  };\n  \n  return samples[script];\n}"],"names":["SCRIPT_RANGES","latin","arabic","hebrew","cjk","devanagari","thai","SCRIPT_REFERENCE_CHARS","mixed","unknown","SCRIPT_METRICS","ascentRatio","descentRatio","actualAscentRatio","actualDescentRatio","detectScript","text","length","scriptCounts","char","codePoint","codePointAt","undefined","detected","script","ranges","Object","entries","start","end","sort","a","b","topScript","topCount","secondScript","secondCount","getScriptReferenceChar","getScriptMetrics","detectGraphemeScript","grapheme","isComplexScript","includes"],"mappings":"AAAA;AACA;AACA;AACA;AACA;AACA;;AAIA;AACA;AACA;AACA,MAAMA,aAAa,GAAG;AACpBC,EAAAA,KAAK,EAAE,CACL,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;AAClB,EAAA,CAAC,MAAM,EAAE,MAAM,CAAC;GACjB;AACDC,EAAAA,MAAM,EAAE,CACN,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;AAClB,EAAA,CAAC,MAAM,EAAE,MAAM,CAAC;GACjB;AACDC,EAAAA,MAAM,EAAE,CACN,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;AAClB,EAAA,CAAC,MAAM,EAAE,MAAM,CAAC;GACjB;AACDC,EAAAA,GAAG,EAAE,CACH,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;EAClB,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;AAClB,EAAA,CAAC,OAAO,EAAE,OAAO,CAAC;GACnB;AACDC,EAAAA,UAAU,EAAE,CACV,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;AAClB,EAAA,CAAC,MAAM,EAAE,MAAM,CAAC;GACjB;AACDC,EAAAA,IAAI,EAAE,CACJ,CAAC,MAAM,EAAE,MAAM,CAAC;AAAE;AAEtB,CAAC;;AAED;AACA;AACA;AACO,MAAMC,sBAAsB,GAAG;AACpCN,EAAAA,KAAK,EAAE,GAAG;AACVC,EAAAA,MAAM,EAAE,GAAG;AAAE;AACbC,EAAAA,MAAM,EAAE,GAAG;AAAE;AACbC,EAAAA,GAAG,EAAE,GAAG;AAAE;AACVC,EAAAA,UAAU,EAAE,GAAG;AAAE;AACjBC,EAAAA,IAAI,EAAE,GAAG;AAAE;AACXE,EAAAA,KAAK,EAAE,GAAG;AAAE;EACZC,OAAO,EAAE,GAAG;AACd;;AAEA;AACA;AACA;AACO,MAAMC,cAAc,GAAG;AAC5BT,EAAAA,KAAK,EAAE;AACLU,IAAAA,WAAW,EAAE,IAAI;AACjBC,IAAAA,YAAY,EAAE,IAAI;AAClBC,IAAAA,iBAAiB,EAAE,KAAK;AACxBC,IAAAA,kBAAkB,EAAE;GACrB;AACDZ,EAAAA,MAAM,EAAE;AACNS,IAAAA,WAAW,EAAE,IAAI;AAAE;AACnBC,IAAAA,YAAY,EAAE,IAAI;AAAE;AACpBC,IAAAA,iBAAiB,EAAE,IAAI;AACvBC,IAAAA,kBAAkB,EAAE;GACrB;AACDX,EAAAA,MAAM,EAAE;AACNQ,IAAAA,WAAW,EAAE,IAAI;AACjBC,IAAAA,YAAY,EAAE,IAAI;AAClBC,IAAAA,iBAAiB,EAAE,IAAI;AACvBC,IAAAA,kBAAkB,EAAE;GACrB;AACDV,EAAAA,GAAG,EAAE;AACHO,IAAAA,WAAW,EAAE,IAAI;AAAE;AACnBC,IAAAA,YAAY,EAAE,IAAI;AAClBC,IAAAA,iBAAiB,EAAE,IAAI;AACvBC,IAAAA,kBAAkB,EAAE;GACrB;AACDT,EAAAA,UAAU,EAAE;AACVM,IAAAA,WAAW,EAAE,GAAG;AAAE;AAClBC,IAAAA,YAAY,EAAE,IAAI;AAClBC,IAAAA,iBAAiB,EAAE,IAAI;AACvBC,IAAAA,kBAAkB,EAAE;GACrB;AACDR,EAAAA,IAAI,EAAE;AACJK,IAAAA,WAAW,EAAE,IAAI;AAAE;AACnBC,IAAAA,YAAY,EAAE,IAAI;AAAE;AACpBC,IAAAA,iBAAiB,EAAE,IAAI;AACvBC,IAAAA,kBAAkB,EAAE;GACrB;AACDN,EAAAA,KAAK,EAAE;AACLG,IAAAA,WAAW,EAAE,IAAI;AACjBC,IAAAA,YAAY,EAAE,IAAI;AAClBC,IAAAA,iBAAiB,EAAE,KAAK;AACxBC,IAAAA,kBAAkB,EAAE;GACrB;AACDL,EAAAA,OAAO,EAAE;AACPE,IAAAA,WAAW,EAAE,IAAI;AACjBC,IAAAA,YAAY,EAAE,IAAI;AAClBC,IAAAA,iBAAiB,EAAE,KAAK;AACxBC,IAAAA,kBAAkB,EAAE;AACtB;AACF;;AAEA;AACA;AACA;AACO,SAASC,YAAYA,CAACC,IAAY,EAAc;EACrD,IAAI,CAACA,IAAI,IAAIA,IAAI,CAACC,MAAM,KAAK,CAAC,EAAE,OAAO,SAAS;AAEhD,EAAA,MAAMC,YAAwC,GAAG;AAC/CjB,IAAAA,KAAK,EAAE,CAAC;AACRC,IAAAA,MAAM,EAAE,CAAC;AACTC,IAAAA,MAAM,EAAE,CAAC;AACTC,IAAAA,GAAG,EAAE,CAAC;AACNC,IAAAA,UAAU,EAAE,CAAC;AACbC,IAAAA,IAAI,EAAE,CAAC;AACPE,IAAAA,KAAK,EAAE,CAAC;AACRC,IAAAA,OAAO,EAAE;GACV;AAED,EAAA,KAAK,MAAMU,IAAI,IAAIH,IAAI,EAAE;AACvB,IAAA,MAAMI,SAAS,GAAGD,IAAI,CAACE,WAAW,CAAC,CAAC,CAAC;IACrC,IAAID,SAAS,KAAKE,SAAS,EAAE;IAE7B,IAAIC,QAAQ,GAAG,KAAK;AAEpB,IAAA,KAAK,MAAM,CAACC,MAAM,EAAEC,MAAM,CAAC,IAAIC,MAAM,CAACC,OAAO,CAAC3B,aAAa,CAAC,EAAE;MAC5D,KAAK,MAAM,CAAC4B,KAAK,EAAEC,GAAG,CAAC,IAAIJ,MAAM,EAAE;AACjC,QAAA,IAAIL,SAAS,IAAIQ,KAAK,IAAIR,SAAS,IAAIS,GAAG,EAAE;UAC1CX,YAAY,CAACM,MAAM,CAAe,EAAE;AACpCD,UAAAA,QAAQ,GAAG,IAAI;AACf,UAAA;AACF,QAAA;AACF,MAAA;AACA,MAAA,IAAIA,QAAQ,EAAE;AAChB,IAAA;IAEA,IAAI,CAACA,QAAQ,EAAE;MACbL,YAAY,CAACT,OAAO,EAAE;AACxB,IAAA;AACF,EAAA;;AAEA;AACA,EAAA,MAAMkB,OAAO,GAAGD,MAAM,CAACC,OAAO,CAACT,YAAY,CAA2B;AACtES,EAAAA,OAAO,CAACG,IAAI,CAAC,CAACC,CAAC,EAAEC,CAAC,KAAKA,CAAC,CAAC,CAAC,CAAC,GAAGD,CAAC,CAAC,CAAC,CAAC,CAAC;EAEnC,MAAM,CAACE,SAAS,EAAEC,QAAQ,CAAC,GAAGP,OAAO,CAAC,CAAC,CAAC;EACxC,MAAM,CAACQ,YAAY,EAAEC,WAAW,CAAC,GAAGT,OAAO,CAAC,CAAC,CAAC;;AAE9C;EACA,IAAIO,QAAQ,GAAG,CAAC,IAAIE,WAAW,GAAGF,QAAQ,GAAG,GAAG,EAAE;AAChD,IAAA,OAAO,OAAO;AAChB,EAAA;EAEA,OAAOD,SAAS,KAAK,SAAS,IAAIC,QAAQ,KAAK,CAAC,GAAG,OAAO,GAAGD,SAAS;AACxE;;AAEA;AACA;AACA;AACO,SAASI,sBAAsBA,CAACb,MAAkB,EAAU;EACjE,OAAOjB,sBAAsB,CAACiB,MAAM,CAAC;AACvC;;AAEA;AACA;AACA;AACO,SAASc,gBAAgBA,CAACd,MAAkB,EAAE;EACnD,OAAOd,cAAc,CAACc,MAAM,CAAC;AAC/B;;AAEA;AACA;AACA;AACO,SAASe,oBAAoBA,CAACC,QAAgB,EAAc;EACjE,OAAOzB,YAAY,CAACyB,QAAQ,CAAC;AAC/B;;AAEA;AACA;AACA;AACO,SAASC,eAAeA,CAACjB,MAAkB,EAAW;AAC3D,EAAA,OAAO,CAAC,QAAQ,EAAE,QAAQ,EAAE,YAAY,EAAE,MAAM,CAAC,CAACkB,QAAQ,CAAClB,MAAM,CAAC;AACpE;;;;"}

package/dist/src/text/unicode.min.mjs.map

@@ -1 +1 @@
-{"version":3,"file":"unicode.min.mjs","sources":["../../../src/text/unicode.ts"],"sourcesContent":["/**\n * Unicode and Internationalization Support\n * \n * Enhanced Unicode handling for complex scripts, RTL/LTR text,\n * and grapheme cluster boundary detection.\n */\n\nimport { graphemeSplit } from '../util/lang_string';\n\nexport interface BiDiRun {\n  text: string;\n  direction: 'ltr' | 'rtl';\n  level: number;\n  start: number;\n  end: number;\n}\n\nexport interface ScriptInfo {\n  script: string;\n  direction: 'ltr' | 'rtl';\n  needsShaping: boolean;\n  complexLayout: boolean;\n}\n\n// Unicode character categories for text processing\nconst UNICODE_CATEGORIES = {\n  // Bidirectional types\n  L: /[\\u0041-\\u005A\\u0061-\\u007A\\u00AA\\u00B5\\u00BA\\u00C0-\\u00D6\\u00D8-\\u00F6]/,\n  R: /[\\u05BE\\u05C0\\u05C3\\u05C6\\u05D0-\\u05EA\\u05F0-\\u05F4\\u0608\\u060B\\u060D]/,\n  AL: /[\\u0627\\u0629-\\u063A\\u0641-\\u064A\\u066D-\\u066F\\u0671-\\u06D3\\u06D5]/,\n  EN: /[\\u0030-\\u0039\\u00B2\\u00B3\\u00B9\\u06F0-\\u06F9]/,\n  ES: /[\\u002B\\u002D]/,\n  ET: /[\\u0023\\u0024\\u0025\\u00A2\\u00A3\\u00A4\\u00A5]/,\n  AN: /[\\u0660-\\u0669\\u066B\\u066C]/,\n  CS: /[\\u002C\\u002E\\u002F\\u003A\\u00A0]/,\n  NSM: /[\\u0300-\\u036F\\u0483-\\u0489\\u0591-\\u05BD\\u05BF\\u05C1\\u05C2]/,\n  BN: /[\\u0000-\\u0008\\u000E-\\u001B\\u007F-\\u0084\\u0086-\\u009F]/,\n  B: /[\\u000A\\u000D\\u001C-\\u001E\\u0085\\u2029]/,\n  S: /[\\u0009\\u000B\\u001F]/,\n  WS: /[\\u000C\\u0020\\u1680\\u2000-\\u200A\\u2028\\u205F\\u3000]/,\n  ON: /[\\u0021\\u0022\\u0026-\\u002A\\u003B-\\u0040\\u005B-\\u0060\\u007B-\\u007E]/,\n};\n\n// Script detection patterns\nconst SCRIPT_PATTERNS = {\n  latin: /[\\u0041-\\u005A\\u0061-\\u007A\\u00C0-\\u024F]/,\n  arabic: /[\\u0600-\\u06FF\\u0750-\\u077F\\u08A0-\\u08FF]/,\n  hebrew: /[\\u0590-\\u05FF]/,\n  cyrillic: /[\\u0400-\\u04FF\\u0500-\\u052F]/,\n  greek: /[\\u0370-\\u03FF\\u1F00-\\u1FFF]/,\n  cjk: /[\\u3400-\\u4DBF\\u4E00-\\u9FFF\\uF900-\\uFAFF]/,\n  hiragana: /[\\u3040-\\u309F]/,\n  katakana: /[\\u30A0-\\u30FF]/,\n  hangul: /[\\uAC00-\\uD7AF\\u1100-\\u11FF\\u3130-\\u318F]/,\n  thai: /[\\u0E00-\\u0E7F]/,\n  devanagari: /[\\u0900-\\u097F]/,\n  bengali: /[\\u0980-\\u09FF]/,\n  emoji: /[\\u1F600-\\u1F64F\\u1F300-\\u1F5FF\\u1F680-\\u1F6FF\\u1F700-\\u1F77F]/,\n};\n\n/**\n * Enhanced grapheme segmentation using Intl.Segmenter when available\n * with fallback to existing graphemeSplit implementation\n */\nexport function segmentGraphemes(text: string): string[] {\n  // Use native Intl.Segmenter if available\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\n    try {\n      const segmenter = new Intl.Segmenter(undefined, {\n        granularity: 'grapheme',\n      });\n      \n      const segments = segmenter.segment(text);\n      return Array.from(segments, segment => segment.segment);\n    } catch (e) {\n      // Fallback if Intl.Segmenter fails\n    }\n  }\n  \n  // Use existing Fabric.js implementation as fallback\n  return graphemeSplit(text);\n}\n\n/**\n * Analyze text for bidirectional runs using Unicode BiDi algorithm (simplified)\n */\nexport function analyzeBiDi(text: string, baseDirection: 'ltr' | 'rtl' = 'ltr'): BiDiRun[] {\n  if (!text) return [];\n\n  const runs: BiDiRun[] = [];\n  const chars = Array.from(text);\n  let currentRun: BiDiRun | null = null;\n\n  for (let i = 0; i < chars.length; i++) {\n    const char = chars[i];\n    const charDirection = getBidiDirection(char, baseDirection);\n    \n    // Start new run if direction changes\n    if (!currentRun || currentRun.direction !== charDirection) {\n      if (currentRun) {\n        runs.push(currentRun);\n      }\n      \n      currentRun = {\n        text: char,\n        direction: charDirection,\n        level: charDirection === 'rtl' ? 1 : 0,\n        start: i,\n        end: i + 1,\n      };\n    } else {\n      // Continue current run\n      currentRun.text += char;\n      currentRun.end = i + 1;\n    }\n  }\n\n  // Add final run\n  if (currentRun) {\n    runs.push(currentRun);\n  }\n\n  return runs.length > 0 ? runs : [{\n    text,\n    direction: baseDirection,\n    level: baseDirection === 'rtl' ? 1 : 0,\n    start: 0,\n    end: text.length,\n  }];\n}\n\n/**\n * Detect the primary script of text for font fallback and shaping\n */\nexport function detectScript(text: string): string {\n  const charCounts: Record<string, number> = {};\n  const chars = Array.from(text);\n\n  // Count characters by script\n  for (const char of chars) {\n    for (const [script, pattern] of Object.entries(SCRIPT_PATTERNS)) {\n      if (pattern.test(char)) {\n        charCounts[script] = (charCounts[script] || 0) + 1;\n        break;\n      }\n    }\n  }\n\n  // Find dominant script\n  let maxCount = 0;\n  let dominantScript = 'latin';\n\n  for (const [script, count] of Object.entries(charCounts)) {\n    if (count > maxCount) {\n      maxCount = count;\n      dominantScript = script;\n    }\n  }\n\n  return dominantScript;\n}\n\n/**\n * Get script information for layout decisions\n */\nexport function getScriptInfo(script: string): ScriptInfo {\n  const scriptMap: Record<string, ScriptInfo> = {\n    latin: {\n      script: 'latin',\n      direction: 'ltr',\n      needsShaping: false,\n      complexLayout: false,\n    },\n    arabic: {\n      script: 'arabic',\n      direction: 'rtl',\n      needsShaping: true,\n      complexLayout: true,\n    },\n    hebrew: {\n      script: 'hebrew',\n      direction: 'rtl',\n      needsShaping: false,\n      complexLayout: false,\n    },\n    thai: {\n      script: 'thai',\n      direction: 'ltr',\n      needsShaping: true,\n      complexLayout: true,\n    },\n    devanagari: {\n      script: 'devanagari',\n      direction: 'ltr',\n      needsShaping: true,\n      complexLayout: true,\n    },\n    cjk: {\n      script: 'cjk',\n      direction: 'ltr',\n      needsShaping: false,\n      complexLayout: false,\n    },\n  };\n\n  return scriptMap[script] || scriptMap.latin;\n}\n\n/**\n * Break words according to language-specific rules\n */\nexport function breakWords(text: string, locale?: string): string[] {\n  // Use native word segmentation if available\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\n    try {\n      const segmenter = new Intl.Segmenter(locale, {\n        granularity: 'word',\n      });\n      \n      const segments = segmenter.segment(text);\n      return Array.from(segments)\n        .filter(segment => segment.isWordLike)\n        .map(segment => segment.segment);\n    } catch (e) {\n      // Fallback if Intl.Segmenter fails\n    }\n  }\n\n  // Simple fallback word breaking\n  return text.split(/\\s+/).filter(word => word.length > 0);\n}\n\n/**\n * Character classification functions\n */\nexport function isWhitespace(grapheme: string): boolean {\n  return /\\s/.test(grapheme);\n}\n\nexport function isLineBreak(grapheme: string): boolean {\n  return grapheme === '\\n' || grapheme === '\\r' || grapheme === '\\r\\n';\n}\n\nexport function isWordSeparator(grapheme: string): boolean {\n  return /[\\s\\-\\u2010-\\u2015]/.test(grapheme);\n}\n\nexport function isPunctuation(grapheme: string): boolean {\n  return /[\\p{P}]/u.test(grapheme);\n}\n\nexport function isEmoji(grapheme: string): boolean {\n  return /[\\p{Emoji}]/u.test(grapheme);\n}\n\nexport function isRTLCharacter(char: string): boolean {\n  return UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char);\n}\n\n/**\n * Advanced grapheme cluster handling for complex scripts\n */\nexport function findGraphemeClusterBoundaries(text: string): number[] {\n  const boundaries: number[] = [0];\n  const graphemes = segmentGraphemes(text);\n  let position = 0;\n\n  for (const grapheme of graphemes) {\n    position += grapheme.length;\n    boundaries.push(position);\n  }\n\n  return boundaries;\n}\n\n/**\n * Check if a position is at a grapheme cluster boundary\n */\nexport function isGraphemeClusterBoundary(text: string, position: number): boolean {\n  const boundaries = findGraphemeClusterBoundaries(text);\n  return boundaries.includes(position);\n}\n\n/**\n * Find the next grapheme cluster boundary from a given position\n */\nexport function findNextGraphemeClusterBoundary(text: string, position: number): number {\n  const boundaries = findGraphemeClusterBoundaries(text);\n  return boundaries.find(boundary => boundary > position) || text.length;\n}\n\n/**\n * Find the previous grapheme cluster boundary from a given position  \n */\nexport function findPreviousGraphemeClusterBoundary(text: string, position: number): number {\n  const boundaries = findGraphemeClusterBoundaries(text);\n  const reversedBoundaries = [...boundaries].reverse();\n  return reversedBoundaries.find(boundary => boundary < position) || 0;\n}\n\n/**\n * Normalize text for consistent processing\n */\nexport function normalizeText(text: string, form: 'NFC' | 'NFD' | 'NFKC' | 'NFKD' = 'NFC'): string {\n  if (typeof text.normalize === 'function') {\n    return text.normalize(form);\n  }\n  return text;\n}\n\n/**\n * Detect if text contains complex script characters that need special handling\n */\nexport function needsComplexLayout(text: string): boolean {\n  const complexScripts = ['arabic', 'hebrew', 'thai', 'devanagari', 'bengali'];\n  const detectedScript = detectScript(text);\n  return complexScripts.includes(detectedScript);\n}\n\n/**\n * Get text direction based on content analysis\n */\nexport function getTextDirection(text: string, fallback: 'ltr' | 'rtl' = 'ltr'): 'ltr' | 'rtl' {\n  let rtlChars = 0;\n  let ltrChars = 0;\n\n  for (const char of text) {\n    if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\n      rtlChars++;\n    } else if (UNICODE_CATEGORIES.L.test(char)) {\n      ltrChars++;\n    }\n  }\n\n  if (rtlChars > ltrChars) return 'rtl';\n  if (ltrChars > rtlChars) return 'ltr';\n  return fallback;\n}\n\n/**\n * Split text into lines respecting complex script rules\n */\nexport function splitTextLines(text: string): string[] {\n  // Handle different line break types\n  return text.split(/\\r\\n|\\r|\\n/);\n}\n\n/**\n * Check if combining marks or modifiers should be kept with base character\n */\nexport function isCombiningMark(char: string): boolean {\n  return UNICODE_CATEGORIES.NSM.test(char);\n}\n\n/**\n * Get bidirectional character type\n */\nfunction getBidiDirection(char: string, baseDirection: 'ltr' | 'rtl'): 'ltr' | 'rtl' {\n  // Strong RTL characters\n  if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\n    return 'rtl';\n  }\n  \n  // Strong LTR characters  \n  if (UNICODE_CATEGORIES.L.test(char)) {\n    return 'ltr';\n  }\n  \n  // Numbers follow base direction in simplified algorithm\n  if (UNICODE_CATEGORIES.EN.test(char) || UNICODE_CATEGORIES.AN.test(char)) {\n    return baseDirection;\n  }\n  \n  // Neutral characters follow context\n  return baseDirection;\n}\n\n/**\n * Advanced word breaking for different scripts\n */\nexport function breakWordsAdvanced(text: string): string[] {\n  const script = detectScript(text);\n  \n  switch (script) {\n    case 'cjk':\n      // CJK can break at most characters\n      return Array.from(text);\n    \n    case 'thai':\n      // Thai doesn't use spaces - would need dictionary-based breaking\n      // For now, fall back to character-based breaking\n      return Array.from(text);\n    \n    case 'arabic':\n    case 'hebrew':\n      // Handle RTL scripts\n      return text.split(/\\s+/).filter(word => word.length > 0);\n    \n    default:\n      // Standard space-based word breaking\n      return text.split(/\\s+/).filter(word => word.length > 0);\n  }\n}\n\n/**\n * Check if text needs right-to-left processing\n */\nexport function needsRTLProcessing(text: string): boolean {\n  for (const char of text) {\n    if (isRTLCharacter(char)) {\n      return true;\n    }\n  }\n  return false;\n}\n\n/**\n * Reverse text for RTL display (simplified - real RTL is more complex)\n */\nexport function reverseForRTL(text: string): string {\n  // This is a simplified implementation\n  // Real RTL processing requires full BiDi algorithm\n  const graphemes = segmentGraphemes(text);\n  return graphemes.reverse().join('');\n}\n\n/**\n * Get appropriate line breaking class for character\n */\nexport function getLineBreakClass(char: string): string {\n  // Simplified line breaking classes\n  if (/\\s/.test(char)) return 'SP';\n  if (/[!-\\/:-@\\[-`{-~]/.test(char)) return 'BA'; // Break after\n  if (/[(\\[{]/.test(char)) return 'OP'; // Open punctuation\n  if (/[)\\]}]/.test(char)) return 'CL'; // Close punctuation\n  if (/\\d/.test(char)) return 'NU'; // Numeric\n  if (/[A-Za-z]/.test(char)) return 'AL'; // Alphabetic\n  \n  return 'AL'; // Default to alphabetic\n}\n\n/**\n * Check if line break is allowed between two characters\n */\nexport function isLineBreakAllowed(before: string, after: string): boolean {\n  const beforeClass = getLineBreakClass(before);\n  const afterClass = getLineBreakClass(after);\n  \n  // Simplified line breaking rules\n  if (beforeClass === 'OP') return false; // Don't break after opening\n  if (afterClass === 'CL') return false; // Don't break before closing\n  if (beforeClass === 'SP') return true; // Always allow break after space\n  \n  return true; // Default allow\n}"],"names":["segmentGraphemes","text","Intl","segments","Segmenter","undefined","granularity","segment","Array","from","e","graphemeSplit"],"mappings":"4DAgEO,SAASA,EAAiBC,GAE/B,GAAoB,oBAATC,MAAwB,cAAeA,KAChD,IACE,MAIMC,EAJY,IAAID,KAAKE,eAAUC,EAAW,CAC9CC,YAAa,aAGYC,QAAQN,GACnC,OAAOO,MAAMC,KAAKN,EAAUI,GAAWA,EAAQA,QACjD,CAAE,MAAOG,GACP,CAKJ,OAAOC,EAAcV,EACvB"}
+{"version":3,"file":"unicode.min.mjs","sources":["../../../src/text/unicode.ts"],"sourcesContent":["/**\r\n * Unicode and Internationalization Support\r\n * \r\n * Enhanced Unicode handling for complex scripts, RTL/LTR text,\r\n * and grapheme cluster boundary detection.\r\n */\r\n\r\nimport { graphemeSplit } from '../util/lang_string';\r\n\r\nexport interface BiDiRun {\r\n  text: string;\r\n  direction: 'ltr' | 'rtl';\r\n  level: number;\r\n  start: number;\r\n  end: number;\r\n}\r\n\r\nexport interface ScriptInfo {\r\n  script: string;\r\n  direction: 'ltr' | 'rtl';\r\n  needsShaping: boolean;\r\n  complexLayout: boolean;\r\n}\r\n\r\n// Unicode character categories for text processing\r\nconst UNICODE_CATEGORIES = {\r\n  // Bidirectional types\r\n  L: /[\\u0041-\\u005A\\u0061-\\u007A\\u00AA\\u00B5\\u00BA\\u00C0-\\u00D6\\u00D8-\\u00F6]/,\r\n  R: /[\\u05BE\\u05C0\\u05C3\\u05C6\\u05D0-\\u05EA\\u05F0-\\u05F4\\u0608\\u060B\\u060D]/,\r\n  AL: /[\\u0627\\u0629-\\u063A\\u0641-\\u064A\\u066D-\\u066F\\u0671-\\u06D3\\u06D5]/,\r\n  EN: /[\\u0030-\\u0039\\u00B2\\u00B3\\u00B9\\u06F0-\\u06F9]/,\r\n  ES: /[\\u002B\\u002D]/,\r\n  ET: /[\\u0023\\u0024\\u0025\\u00A2\\u00A3\\u00A4\\u00A5]/,\r\n  AN: /[\\u0660-\\u0669\\u066B\\u066C]/,\r\n  CS: /[\\u002C\\u002E\\u002F\\u003A\\u00A0]/,\r\n  NSM: /[\\u0300-\\u036F\\u0483-\\u0489\\u0591-\\u05BD\\u05BF\\u05C1\\u05C2]/,\r\n  BN: /[\\u0000-\\u0008\\u000E-\\u001B\\u007F-\\u0084\\u0086-\\u009F]/,\r\n  B: /[\\u000A\\u000D\\u001C-\\u001E\\u0085\\u2029]/,\r\n  S: /[\\u0009\\u000B\\u001F]/,\r\n  WS: /[\\u000C\\u0020\\u1680\\u2000-\\u200A\\u2028\\u205F\\u3000]/,\r\n  ON: /[\\u0021\\u0022\\u0026-\\u002A\\u003B-\\u0040\\u005B-\\u0060\\u007B-\\u007E]/,\r\n};\r\n\r\n// Script detection patterns\r\nconst SCRIPT_PATTERNS = {\r\n  latin: /[\\u0041-\\u005A\\u0061-\\u007A\\u00C0-\\u024F]/,\r\n  arabic: /[\\u0600-\\u06FF\\u0750-\\u077F\\u08A0-\\u08FF]/,\r\n  hebrew: /[\\u0590-\\u05FF]/,\r\n  cyrillic: /[\\u0400-\\u04FF\\u0500-\\u052F]/,\r\n  greek: /[\\u0370-\\u03FF\\u1F00-\\u1FFF]/,\r\n  cjk: /[\\u3400-\\u4DBF\\u4E00-\\u9FFF\\uF900-\\uFAFF]/,\r\n  hiragana: /[\\u3040-\\u309F]/,\r\n  katakana: /[\\u30A0-\\u30FF]/,\r\n  hangul: /[\\uAC00-\\uD7AF\\u1100-\\u11FF\\u3130-\\u318F]/,\r\n  thai: /[\\u0E00-\\u0E7F]/,\r\n  devanagari: /[\\u0900-\\u097F]/,\r\n  bengali: /[\\u0980-\\u09FF]/,\r\n  emoji: /[\\u1F600-\\u1F64F\\u1F300-\\u1F5FF\\u1F680-\\u1F6FF\\u1F700-\\u1F77F]/,\r\n};\r\n\r\n/**\r\n * Enhanced grapheme segmentation using Intl.Segmenter when available\r\n * with fallback to existing graphemeSplit implementation\r\n */\r\nexport function segmentGraphemes(text: string): string[] {\r\n  // Use native Intl.Segmenter if available\r\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\r\n    try {\r\n      const segmenter = new Intl.Segmenter(undefined, {\r\n        granularity: 'grapheme',\r\n      });\r\n      \r\n      const segments = segmenter.segment(text);\r\n      return Array.from(segments, segment => segment.segment);\r\n    } catch (e) {\r\n      // Fallback if Intl.Segmenter fails\r\n    }\r\n  }\r\n  \r\n  // Use existing Fabric.js implementation as fallback\r\n  return graphemeSplit(text);\r\n}\r\n\r\n/**\r\n * Analyze text for bidirectional runs using Unicode BiDi algorithm (simplified)\r\n */\r\nexport function analyzeBiDi(text: string, baseDirection: 'ltr' | 'rtl' = 'ltr'): BiDiRun[] {\r\n  if (!text) return [];\r\n\r\n  const runs: BiDiRun[] = [];\r\n  const chars = Array.from(text);\r\n  let currentRun: BiDiRun | null = null;\r\n\r\n  for (let i = 0; i < chars.length; i++) {\r\n    const char = chars[i];\r\n    const charDirection = getBidiDirection(char, baseDirection);\r\n    \r\n    // Start new run if direction changes\r\n    if (!currentRun || currentRun.direction !== charDirection) {\r\n      if (currentRun) {\r\n        runs.push(currentRun);\r\n      }\r\n      \r\n      currentRun = {\r\n        text: char,\r\n        direction: charDirection,\r\n        level: charDirection === 'rtl' ? 1 : 0,\r\n        start: i,\r\n        end: i + 1,\r\n      };\r\n    } else {\r\n      // Continue current run\r\n      currentRun.text += char;\r\n      currentRun.end = i + 1;\r\n    }\r\n  }\r\n\r\n  // Add final run\r\n  if (currentRun) {\r\n    runs.push(currentRun);\r\n  }\r\n\r\n  return runs.length > 0 ? runs : [{\r\n    text,\r\n    direction: baseDirection,\r\n    level: baseDirection === 'rtl' ? 1 : 0,\r\n    start: 0,\r\n    end: text.length,\r\n  }];\r\n}\r\n\r\n/**\r\n * Detect the primary script of text for font fallback and shaping\r\n */\r\nexport function detectScript(text: string): string {\r\n  const charCounts: Record<string, number> = {};\r\n  const chars = Array.from(text);\r\n\r\n  // Count characters by script\r\n  for (const char of chars) {\r\n    for (const [script, pattern] of Object.entries(SCRIPT_PATTERNS)) {\r\n      if (pattern.test(char)) {\r\n        charCounts[script] = (charCounts[script] || 0) + 1;\r\n        break;\r\n      }\r\n    }\r\n  }\r\n\r\n  // Find dominant script\r\n  let maxCount = 0;\r\n  let dominantScript = 'latin';\r\n\r\n  for (const [script, count] of Object.entries(charCounts)) {\r\n    if (count > maxCount) {\r\n      maxCount = count;\r\n      dominantScript = script;\r\n    }\r\n  }\r\n\r\n  return dominantScript;\r\n}\r\n\r\n/**\r\n * Get script information for layout decisions\r\n */\r\nexport function getScriptInfo(script: string): ScriptInfo {\r\n  const scriptMap: Record<string, ScriptInfo> = {\r\n    latin: {\r\n      script: 'latin',\r\n      direction: 'ltr',\r\n      needsShaping: false,\r\n      complexLayout: false,\r\n    },\r\n    arabic: {\r\n      script: 'arabic',\r\n      direction: 'rtl',\r\n      needsShaping: true,\r\n      complexLayout: true,\r\n    },\r\n    hebrew: {\r\n      script: 'hebrew',\r\n      direction: 'rtl',\r\n      needsShaping: false,\r\n      complexLayout: false,\r\n    },\r\n    thai: {\r\n      script: 'thai',\r\n      direction: 'ltr',\r\n      needsShaping: true,\r\n      complexLayout: true,\r\n    },\r\n    devanagari: {\r\n      script: 'devanagari',\r\n      direction: 'ltr',\r\n      needsShaping: true,\r\n      complexLayout: true,\r\n    },\r\n    cjk: {\r\n      script: 'cjk',\r\n      direction: 'ltr',\r\n      needsShaping: false,\r\n      complexLayout: false,\r\n    },\r\n  };\r\n\r\n  return scriptMap[script] || scriptMap.latin;\r\n}\r\n\r\n/**\r\n * Break words according to language-specific rules\r\n */\r\nexport function breakWords(text: string, locale?: string): string[] {\r\n  // Use native word segmentation if available\r\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\r\n    try {\r\n      const segmenter = new Intl.Segmenter(locale, {\r\n        granularity: 'word',\r\n      });\r\n      \r\n      const segments = segmenter.segment(text);\r\n      return Array.from(segments)\r\n        .filter(segment => segment.isWordLike)\r\n        .map(segment => segment.segment);\r\n    } catch (e) {\r\n      // Fallback if Intl.Segmenter fails\r\n    }\r\n  }\r\n\r\n  // Simple fallback word breaking\r\n  return text.split(/\\s+/).filter(word => word.length > 0);\r\n}\r\n\r\n/**\r\n * Character classification functions\r\n */\r\nexport function isWhitespace(grapheme: string): boolean {\r\n  return /\\s/.test(grapheme);\r\n}\r\n\r\nexport function isLineBreak(grapheme: string): boolean {\r\n  return grapheme === '\\n' || grapheme === '\\r' || grapheme === '\\r\\n';\r\n}\r\n\r\nexport function isWordSeparator(grapheme: string): boolean {\r\n  return /[\\s\\-\\u2010-\\u2015]/.test(grapheme);\r\n}\r\n\r\nexport function isPunctuation(grapheme: string): boolean {\r\n  return /[\\p{P}]/u.test(grapheme);\r\n}\r\n\r\nexport function isEmoji(grapheme: string): boolean {\r\n  return /[\\p{Emoji}]/u.test(grapheme);\r\n}\r\n\r\nexport function isRTLCharacter(char: string): boolean {\r\n  return UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char);\r\n}\r\n\r\n/**\r\n * Advanced grapheme cluster handling for complex scripts\r\n */\r\nexport function findGraphemeClusterBoundaries(text: string): number[] {\r\n  const boundaries: number[] = [0];\r\n  const graphemes = segmentGraphemes(text);\r\n  let position = 0;\r\n\r\n  for (const grapheme of graphemes) {\r\n    position += grapheme.length;\r\n    boundaries.push(position);\r\n  }\r\n\r\n  return boundaries;\r\n}\r\n\r\n/**\r\n * Check if a position is at a grapheme cluster boundary\r\n */\r\nexport function isGraphemeClusterBoundary(text: string, position: number): boolean {\r\n  const boundaries = findGraphemeClusterBoundaries(text);\r\n  return boundaries.includes(position);\r\n}\r\n\r\n/**\r\n * Find the next grapheme cluster boundary from a given position\r\n */\r\nexport function findNextGraphemeClusterBoundary(text: string, position: number): number {\r\n  const boundaries = findGraphemeClusterBoundaries(text);\r\n  return boundaries.find(boundary => boundary > position) || text.length;\r\n}\r\n\r\n/**\r\n * Find the previous grapheme cluster boundary from a given position  \r\n */\r\nexport function findPreviousGraphemeClusterBoundary(text: string, position: number): number {\r\n  const boundaries = findGraphemeClusterBoundaries(text);\r\n  const reversedBoundaries = [...boundaries].reverse();\r\n  return reversedBoundaries.find(boundary => boundary < position) || 0;\r\n}\r\n\r\n/**\r\n * Normalize text for consistent processing\r\n */\r\nexport function normalizeText(text: string, form: 'NFC' | 'NFD' | 'NFKC' | 'NFKD' = 'NFC'): string {\r\n  if (typeof text.normalize === 'function') {\r\n    return text.normalize(form);\r\n  }\r\n  return text;\r\n}\r\n\r\n/**\r\n * Detect if text contains complex script characters that need special handling\r\n */\r\nexport function needsComplexLayout(text: string): boolean {\r\n  const complexScripts = ['arabic', 'hebrew', 'thai', 'devanagari', 'bengali'];\r\n  const detectedScript = detectScript(text);\r\n  return complexScripts.includes(detectedScript);\r\n}\r\n\r\n/**\r\n * Get text direction based on content analysis\r\n */\r\nexport function getTextDirection(text: string, fallback: 'ltr' | 'rtl' = 'ltr'): 'ltr' | 'rtl' {\r\n  let rtlChars = 0;\r\n  let ltrChars = 0;\r\n\r\n  for (const char of text) {\r\n    if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\r\n      rtlChars++;\r\n    } else if (UNICODE_CATEGORIES.L.test(char)) {\r\n      ltrChars++;\r\n    }\r\n  }\r\n\r\n  if (rtlChars > ltrChars) return 'rtl';\r\n  if (ltrChars > rtlChars) return 'ltr';\r\n  return fallback;\r\n}\r\n\r\n/**\r\n * Split text into lines respecting complex script rules\r\n */\r\nexport function splitTextLines(text: string): string[] {\r\n  // Handle different line break types\r\n  return text.split(/\\r\\n|\\r|\\n/);\r\n}\r\n\r\n/**\r\n * Check if combining marks or modifiers should be kept with base character\r\n */\r\nexport function isCombiningMark(char: string): boolean {\r\n  return UNICODE_CATEGORIES.NSM.test(char);\r\n}\r\n\r\n/**\r\n * Get bidirectional character type\r\n */\r\nfunction getBidiDirection(char: string, baseDirection: 'ltr' | 'rtl'): 'ltr' | 'rtl' {\r\n  // Strong RTL characters\r\n  if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\r\n    return 'rtl';\r\n  }\r\n  \r\n  // Strong LTR characters  \r\n  if (UNICODE_CATEGORIES.L.test(char)) {\r\n    return 'ltr';\r\n  }\r\n  \r\n  // Numbers follow base direction in simplified algorithm\r\n  if (UNICODE_CATEGORIES.EN.test(char) || UNICODE_CATEGORIES.AN.test(char)) {\r\n    return baseDirection;\r\n  }\r\n  \r\n  // Neutral characters follow context\r\n  return baseDirection;\r\n}\r\n\r\n/**\r\n * Advanced word breaking for different scripts\r\n */\r\nexport function breakWordsAdvanced(text: string): string[] {\r\n  const script = detectScript(text);\r\n  \r\n  switch (script) {\r\n    case 'cjk':\r\n      // CJK can break at most characters\r\n      return Array.from(text);\r\n    \r\n    case 'thai':\r\n      // Thai doesn't use spaces - would need dictionary-based breaking\r\n      // For now, fall back to character-based breaking\r\n      return Array.from(text);\r\n    \r\n    case 'arabic':\r\n    case 'hebrew':\r\n      // Handle RTL scripts\r\n      return text.split(/\\s+/).filter(word => word.length > 0);\r\n    \r\n    default:\r\n      // Standard space-based word breaking\r\n      return text.split(/\\s+/).filter(word => word.length > 0);\r\n  }\r\n}\r\n\r\n/**\r\n * Check if text needs right-to-left processing\r\n */\r\nexport function needsRTLProcessing(text: string): boolean {\r\n  for (const char of text) {\r\n    if (isRTLCharacter(char)) {\r\n      return true;\r\n    }\r\n  }\r\n  return false;\r\n}\r\n\r\n/**\r\n * Reverse text for RTL display (simplified - real RTL is more complex)\r\n */\r\nexport function reverseForRTL(text: string): string {\r\n  // This is a simplified implementation\r\n  // Real RTL processing requires full BiDi algorithm\r\n  const graphemes = segmentGraphemes(text);\r\n  return graphemes.reverse().join('');\r\n}\r\n\r\n/**\r\n * Get appropriate line breaking class for character\r\n */\r\nexport function getLineBreakClass(char: string): string {\r\n  // Simplified line breaking classes\r\n  if (/\\s/.test(char)) return 'SP';\r\n  if (/[!-\\/:-@\\[-`{-~]/.test(char)) return 'BA'; // Break after\r\n  if (/[(\\[{]/.test(char)) return 'OP'; // Open punctuation\r\n  if (/[)\\]}]/.test(char)) return 'CL'; // Close punctuation\r\n  if (/\\d/.test(char)) return 'NU'; // Numeric\r\n  if (/[A-Za-z]/.test(char)) return 'AL'; // Alphabetic\r\n  \r\n  return 'AL'; // Default to alphabetic\r\n}\r\n\r\n/**\r\n * Check if line break is allowed between two characters\r\n */\r\nexport function isLineBreakAllowed(before: string, after: string): boolean {\r\n  const beforeClass = getLineBreakClass(before);\r\n  const afterClass = getLineBreakClass(after);\r\n  \r\n  // Simplified line breaking rules\r\n  if (beforeClass === 'OP') return false; // Don't break after opening\r\n  if (afterClass === 'CL') return false; // Don't break before closing\r\n  if (beforeClass === 'SP') return true; // Always allow break after space\r\n  \r\n  return true; // Default allow\r\n}"],"names":["segmentGraphemes","text","Intl","segments","Segmenter","undefined","granularity","segment","Array","from","e","graphemeSplit"],"mappings":"4DAgEO,SAASA,EAAiBC,GAE/B,GAAoB,oBAATC,MAAwB,cAAeA,KAChD,IACE,MAIMC,EAJY,IAAID,KAAKE,eAAUC,EAAW,CAC9CC,YAAa,aAGYC,QAAQN,GACnC,OAAOO,MAAMC,KAAKN,EAAUI,GAAWA,EAAQA,QACjD,CAAE,MAAOG,GACP,CAKJ,OAAOC,EAAcV,EACvB"}

package/dist/src/text/unicode.mjs

@@ -1,16 +1,16 @@
 import { graphemeSplit } from '../util/lang_string.mjs';
 
-/**
- * Unicode and Internationalization Support
- * 
- * Enhanced Unicode handling for complex scripts, RTL/LTR text,
- * and grapheme cluster boundary detection.
+/**
+ * Unicode and Internationalization Support
+ * 
+ * Enhanced Unicode handling for complex scripts, RTL/LTR text,
+ * and grapheme cluster boundary detection.
  */
 
 
-/**
- * Enhanced grapheme segmentation using Intl.Segmenter when available
- * with fallback to existing graphemeSplit implementation
+/**
+ * Enhanced grapheme segmentation using Intl.Segmenter when available
+ * with fallback to existing graphemeSplit implementation
  */
 function segmentGraphemes(text) {
   // Use native Intl.Segmenter if available

package/dist/src/text/unicode.mjs.map

@@ -1 +1 @@
-{"version":3,"file":"unicode.mjs","sources":["../../../src/text/unicode.ts"],"sourcesContent":["/**\n * Unicode and Internationalization Support\n * \n * Enhanced Unicode handling for complex scripts, RTL/LTR text,\n * and grapheme cluster boundary detection.\n */\n\nimport { graphemeSplit } from '../util/lang_string';\n\nexport interface BiDiRun {\n  text: string;\n  direction: 'ltr' | 'rtl';\n  level: number;\n  start: number;\n  end: number;\n}\n\nexport interface ScriptInfo {\n  script: string;\n  direction: 'ltr' | 'rtl';\n  needsShaping: boolean;\n  complexLayout: boolean;\n}\n\n// Unicode character categories for text processing\nconst UNICODE_CATEGORIES = {\n  // Bidirectional types\n  L: /[\\u0041-\\u005A\\u0061-\\u007A\\u00AA\\u00B5\\u00BA\\u00C0-\\u00D6\\u00D8-\\u00F6]/,\n  R: /[\\u05BE\\u05C0\\u05C3\\u05C6\\u05D0-\\u05EA\\u05F0-\\u05F4\\u0608\\u060B\\u060D]/,\n  AL: /[\\u0627\\u0629-\\u063A\\u0641-\\u064A\\u066D-\\u066F\\u0671-\\u06D3\\u06D5]/,\n  EN: /[\\u0030-\\u0039\\u00B2\\u00B3\\u00B9\\u06F0-\\u06F9]/,\n  ES: /[\\u002B\\u002D]/,\n  ET: /[\\u0023\\u0024\\u0025\\u00A2\\u00A3\\u00A4\\u00A5]/,\n  AN: /[\\u0660-\\u0669\\u066B\\u066C]/,\n  CS: /[\\u002C\\u002E\\u002F\\u003A\\u00A0]/,\n  NSM: /[\\u0300-\\u036F\\u0483-\\u0489\\u0591-\\u05BD\\u05BF\\u05C1\\u05C2]/,\n  BN: /[\\u0000-\\u0008\\u000E-\\u001B\\u007F-\\u0084\\u0086-\\u009F]/,\n  B: /[\\u000A\\u000D\\u001C-\\u001E\\u0085\\u2029]/,\n  S: /[\\u0009\\u000B\\u001F]/,\n  WS: /[\\u000C\\u0020\\u1680\\u2000-\\u200A\\u2028\\u205F\\u3000]/,\n  ON: /[\\u0021\\u0022\\u0026-\\u002A\\u003B-\\u0040\\u005B-\\u0060\\u007B-\\u007E]/,\n};\n\n// Script detection patterns\nconst SCRIPT_PATTERNS = {\n  latin: /[\\u0041-\\u005A\\u0061-\\u007A\\u00C0-\\u024F]/,\n  arabic: /[\\u0600-\\u06FF\\u0750-\\u077F\\u08A0-\\u08FF]/,\n  hebrew: /[\\u0590-\\u05FF]/,\n  cyrillic: /[\\u0400-\\u04FF\\u0500-\\u052F]/,\n  greek: /[\\u0370-\\u03FF\\u1F00-\\u1FFF]/,\n  cjk: /[\\u3400-\\u4DBF\\u4E00-\\u9FFF\\uF900-\\uFAFF]/,\n  hiragana: /[\\u3040-\\u309F]/,\n  katakana: /[\\u30A0-\\u30FF]/,\n  hangul: /[\\uAC00-\\uD7AF\\u1100-\\u11FF\\u3130-\\u318F]/,\n  thai: /[\\u0E00-\\u0E7F]/,\n  devanagari: /[\\u0900-\\u097F]/,\n  bengali: /[\\u0980-\\u09FF]/,\n  emoji: /[\\u1F600-\\u1F64F\\u1F300-\\u1F5FF\\u1F680-\\u1F6FF\\u1F700-\\u1F77F]/,\n};\n\n/**\n * Enhanced grapheme segmentation using Intl.Segmenter when available\n * with fallback to existing graphemeSplit implementation\n */\nexport function segmentGraphemes(text: string): string[] {\n  // Use native Intl.Segmenter if available\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\n    try {\n      const segmenter = new Intl.Segmenter(undefined, {\n        granularity: 'grapheme',\n      });\n      \n      const segments = segmenter.segment(text);\n      return Array.from(segments, segment => segment.segment);\n    } catch (e) {\n      // Fallback if Intl.Segmenter fails\n    }\n  }\n  \n  // Use existing Fabric.js implementation as fallback\n  return graphemeSplit(text);\n}\n\n/**\n * Analyze text for bidirectional runs using Unicode BiDi algorithm (simplified)\n */\nexport function analyzeBiDi(text: string, baseDirection: 'ltr' | 'rtl' = 'ltr'): BiDiRun[] {\n  if (!text) return [];\n\n  const runs: BiDiRun[] = [];\n  const chars = Array.from(text);\n  let currentRun: BiDiRun | null = null;\n\n  for (let i = 0; i < chars.length; i++) {\n    const char = chars[i];\n    const charDirection = getBidiDirection(char, baseDirection);\n    \n    // Start new run if direction changes\n    if (!currentRun || currentRun.direction !== charDirection) {\n      if (currentRun) {\n        runs.push(currentRun);\n      }\n      \n      currentRun = {\n        text: char,\n        direction: charDirection,\n        level: charDirection === 'rtl' ? 1 : 0,\n        start: i,\n        end: i + 1,\n      };\n    } else {\n      // Continue current run\n      currentRun.text += char;\n      currentRun.end = i + 1;\n    }\n  }\n\n  // Add final run\n  if (currentRun) {\n    runs.push(currentRun);\n  }\n\n  return runs.length > 0 ? runs : [{\n    text,\n    direction: baseDirection,\n    level: baseDirection === 'rtl' ? 1 : 0,\n    start: 0,\n    end: text.length,\n  }];\n}\n\n/**\n * Detect the primary script of text for font fallback and shaping\n */\nexport function detectScript(text: string): string {\n  const charCounts: Record<string, number> = {};\n  const chars = Array.from(text);\n\n  // Count characters by script\n  for (const char of chars) {\n    for (const [script, pattern] of Object.entries(SCRIPT_PATTERNS)) {\n      if (pattern.test(char)) {\n        charCounts[script] = (charCounts[script] || 0) + 1;\n        break;\n      }\n    }\n  }\n\n  // Find dominant script\n  let maxCount = 0;\n  let dominantScript = 'latin';\n\n  for (const [script, count] of Object.entries(charCounts)) {\n    if (count > maxCount) {\n      maxCount = count;\n      dominantScript = script;\n    }\n  }\n\n  return dominantScript;\n}\n\n/**\n * Get script information for layout decisions\n */\nexport function getScriptInfo(script: string): ScriptInfo {\n  const scriptMap: Record<string, ScriptInfo> = {\n    latin: {\n      script: 'latin',\n      direction: 'ltr',\n      needsShaping: false,\n      complexLayout: false,\n    },\n    arabic: {\n      script: 'arabic',\n      direction: 'rtl',\n      needsShaping: true,\n      complexLayout: true,\n    },\n    hebrew: {\n      script: 'hebrew',\n      direction: 'rtl',\n      needsShaping: false,\n      complexLayout: false,\n    },\n    thai: {\n      script: 'thai',\n      direction: 'ltr',\n      needsShaping: true,\n      complexLayout: true,\n    },\n    devanagari: {\n      script: 'devanagari',\n      direction: 'ltr',\n      needsShaping: true,\n      complexLayout: true,\n    },\n    cjk: {\n      script: 'cjk',\n      direction: 'ltr',\n      needsShaping: false,\n      complexLayout: false,\n    },\n  };\n\n  return scriptMap[script] || scriptMap.latin;\n}\n\n/**\n * Break words according to language-specific rules\n */\nexport function breakWords(text: string, locale?: string): string[] {\n  // Use native word segmentation if available\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\n    try {\n      const segmenter = new Intl.Segmenter(locale, {\n        granularity: 'word',\n      });\n      \n      const segments = segmenter.segment(text);\n      return Array.from(segments)\n        .filter(segment => segment.isWordLike)\n        .map(segment => segment.segment);\n    } catch (e) {\n      // Fallback if Intl.Segmenter fails\n    }\n  }\n\n  // Simple fallback word breaking\n  return text.split(/\\s+/).filter(word => word.length > 0);\n}\n\n/**\n * Character classification functions\n */\nexport function isWhitespace(grapheme: string): boolean {\n  return /\\s/.test(grapheme);\n}\n\nexport function isLineBreak(grapheme: string): boolean {\n  return grapheme === '\\n' || grapheme === '\\r' || grapheme === '\\r\\n';\n}\n\nexport function isWordSeparator(grapheme: string): boolean {\n  return /[\\s\\-\\u2010-\\u2015]/.test(grapheme);\n}\n\nexport function isPunctuation(grapheme: string): boolean {\n  return /[\\p{P}]/u.test(grapheme);\n}\n\nexport function isEmoji(grapheme: string): boolean {\n  return /[\\p{Emoji}]/u.test(grapheme);\n}\n\nexport function isRTLCharacter(char: string): boolean {\n  return UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char);\n}\n\n/**\n * Advanced grapheme cluster handling for complex scripts\n */\nexport function findGraphemeClusterBoundaries(text: string): number[] {\n  const boundaries: number[] = [0];\n  const graphemes = segmentGraphemes(text);\n  let position = 0;\n\n  for (const grapheme of graphemes) {\n    position += grapheme.length;\n    boundaries.push(position);\n  }\n\n  return boundaries;\n}\n\n/**\n * Check if a position is at a grapheme cluster boundary\n */\nexport function isGraphemeClusterBoundary(text: string, position: number): boolean {\n  const boundaries = findGraphemeClusterBoundaries(text);\n  return boundaries.includes(position);\n}\n\n/**\n * Find the next grapheme cluster boundary from a given position\n */\nexport function findNextGraphemeClusterBoundary(text: string, position: number): number {\n  const boundaries = findGraphemeClusterBoundaries(text);\n  return boundaries.find(boundary => boundary > position) || text.length;\n}\n\n/**\n * Find the previous grapheme cluster boundary from a given position  \n */\nexport function findPreviousGraphemeClusterBoundary(text: string, position: number): number {\n  const boundaries = findGraphemeClusterBoundaries(text);\n  const reversedBoundaries = [...boundaries].reverse();\n  return reversedBoundaries.find(boundary => boundary < position) || 0;\n}\n\n/**\n * Normalize text for consistent processing\n */\nexport function normalizeText(text: string, form: 'NFC' | 'NFD' | 'NFKC' | 'NFKD' = 'NFC'): string {\n  if (typeof text.normalize === 'function') {\n    return text.normalize(form);\n  }\n  return text;\n}\n\n/**\n * Detect if text contains complex script characters that need special handling\n */\nexport function needsComplexLayout(text: string): boolean {\n  const complexScripts = ['arabic', 'hebrew', 'thai', 'devanagari', 'bengali'];\n  const detectedScript = detectScript(text);\n  return complexScripts.includes(detectedScript);\n}\n\n/**\n * Get text direction based on content analysis\n */\nexport function getTextDirection(text: string, fallback: 'ltr' | 'rtl' = 'ltr'): 'ltr' | 'rtl' {\n  let rtlChars = 0;\n  let ltrChars = 0;\n\n  for (const char of text) {\n    if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\n      rtlChars++;\n    } else if (UNICODE_CATEGORIES.L.test(char)) {\n      ltrChars++;\n    }\n  }\n\n  if (rtlChars > ltrChars) return 'rtl';\n  if (ltrChars > rtlChars) return 'ltr';\n  return fallback;\n}\n\n/**\n * Split text into lines respecting complex script rules\n */\nexport function splitTextLines(text: string): string[] {\n  // Handle different line break types\n  return text.split(/\\r\\n|\\r|\\n/);\n}\n\n/**\n * Check if combining marks or modifiers should be kept with base character\n */\nexport function isCombiningMark(char: string): boolean {\n  return UNICODE_CATEGORIES.NSM.test(char);\n}\n\n/**\n * Get bidirectional character type\n */\nfunction getBidiDirection(char: string, baseDirection: 'ltr' | 'rtl'): 'ltr' | 'rtl' {\n  // Strong RTL characters\n  if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\n    return 'rtl';\n  }\n  \n  // Strong LTR characters  \n  if (UNICODE_CATEGORIES.L.test(char)) {\n    return 'ltr';\n  }\n  \n  // Numbers follow base direction in simplified algorithm\n  if (UNICODE_CATEGORIES.EN.test(char) || UNICODE_CATEGORIES.AN.test(char)) {\n    return baseDirection;\n  }\n  \n  // Neutral characters follow context\n  return baseDirection;\n}\n\n/**\n * Advanced word breaking for different scripts\n */\nexport function breakWordsAdvanced(text: string): string[] {\n  const script = detectScript(text);\n  \n  switch (script) {\n    case 'cjk':\n      // CJK can break at most characters\n      return Array.from(text);\n    \n    case 'thai':\n      // Thai doesn't use spaces - would need dictionary-based breaking\n      // For now, fall back to character-based breaking\n      return Array.from(text);\n    \n    case 'arabic':\n    case 'hebrew':\n      // Handle RTL scripts\n      return text.split(/\\s+/).filter(word => word.length > 0);\n    \n    default:\n      // Standard space-based word breaking\n      return text.split(/\\s+/).filter(word => word.length > 0);\n  }\n}\n\n/**\n * Check if text needs right-to-left processing\n */\nexport function needsRTLProcessing(text: string): boolean {\n  for (const char of text) {\n    if (isRTLCharacter(char)) {\n      return true;\n    }\n  }\n  return false;\n}\n\n/**\n * Reverse text for RTL display (simplified - real RTL is more complex)\n */\nexport function reverseForRTL(text: string): string {\n  // This is a simplified implementation\n  // Real RTL processing requires full BiDi algorithm\n  const graphemes = segmentGraphemes(text);\n  return graphemes.reverse().join('');\n}\n\n/**\n * Get appropriate line breaking class for character\n */\nexport function getLineBreakClass(char: string): string {\n  // Simplified line breaking classes\n  if (/\\s/.test(char)) return 'SP';\n  if (/[!-\\/:-@\\[-`{-~]/.test(char)) return 'BA'; // Break after\n  if (/[(\\[{]/.test(char)) return 'OP'; // Open punctuation\n  if (/[)\\]}]/.test(char)) return 'CL'; // Close punctuation\n  if (/\\d/.test(char)) return 'NU'; // Numeric\n  if (/[A-Za-z]/.test(char)) return 'AL'; // Alphabetic\n  \n  return 'AL'; // Default to alphabetic\n}\n\n/**\n * Check if line break is allowed between two characters\n */\nexport function isLineBreakAllowed(before: string, after: string): boolean {\n  const beforeClass = getLineBreakClass(before);\n  const afterClass = getLineBreakClass(after);\n  \n  // Simplified line breaking rules\n  if (beforeClass === 'OP') return false; // Don't break after opening\n  if (afterClass === 'CL') return false; // Don't break before closing\n  if (beforeClass === 'SP') return true; // Always allow break after space\n  \n  return true; // Default allow\n}"],"names":["segmentGraphemes","text","Intl","segmenter","Segmenter","undefined","granularity","segments","segment","Array","from","e","graphemeSplit"],"mappings":";;AAAA;AACA;AACA;AACA;AACA;AACA;;;AAuDA;AACA;AACA;AACA;AACO,SAASA,gBAAgBA,CAACC,IAAY,EAAY;AACvD;EACA,IAAI,OAAOC,IAAI,KAAK,WAAW,IAAI,WAAW,IAAIA,IAAI,EAAE;IACtD,IAAI;MACF,MAAMC,SAAS,GAAG,IAAID,IAAI,CAACE,SAAS,CAACC,SAAS,EAAE;AAC9CC,QAAAA,WAAW,EAAE;AACf,OAAC,CAAC;AAEF,MAAA,MAAMC,QAAQ,GAAGJ,SAAS,CAACK,OAAO,CAACP,IAAI,CAAC;MACxC,OAAOQ,KAAK,CAACC,IAAI,CAACH,QAAQ,EAAEC,OAAO,IAAIA,OAAO,CAACA,OAAO,CAAC;IACzD,CAAC,CAAC,OAAOG,CAAC,EAAE;AACV;AAAA,IAAA;AAEJ,EAAA;;AAEA;EACA,OAAOC,aAAa,CAACX,IAAI,CAAC;AAC5B;;;;"}
+{"version":3,"file":"unicode.mjs","sources":["../../../src/text/unicode.ts"],"sourcesContent":["/**\r\n * Unicode and Internationalization Support\r\n * \r\n * Enhanced Unicode handling for complex scripts, RTL/LTR text,\r\n * and grapheme cluster boundary detection.\r\n */\r\n\r\nimport { graphemeSplit } from '../util/lang_string';\r\n\r\nexport interface BiDiRun {\r\n  text: string;\r\n  direction: 'ltr' | 'rtl';\r\n  level: number;\r\n  start: number;\r\n  end: number;\r\n}\r\n\r\nexport interface ScriptInfo {\r\n  script: string;\r\n  direction: 'ltr' | 'rtl';\r\n  needsShaping: boolean;\r\n  complexLayout: boolean;\r\n}\r\n\r\n// Unicode character categories for text processing\r\nconst UNICODE_CATEGORIES = {\r\n  // Bidirectional types\r\n  L: /[\\u0041-\\u005A\\u0061-\\u007A\\u00AA\\u00B5\\u00BA\\u00C0-\\u00D6\\u00D8-\\u00F6]/,\r\n  R: /[\\u05BE\\u05C0\\u05C3\\u05C6\\u05D0-\\u05EA\\u05F0-\\u05F4\\u0608\\u060B\\u060D]/,\r\n  AL: /[\\u0627\\u0629-\\u063A\\u0641-\\u064A\\u066D-\\u066F\\u0671-\\u06D3\\u06D5]/,\r\n  EN: /[\\u0030-\\u0039\\u00B2\\u00B3\\u00B9\\u06F0-\\u06F9]/,\r\n  ES: /[\\u002B\\u002D]/,\r\n  ET: /[\\u0023\\u0024\\u0025\\u00A2\\u00A3\\u00A4\\u00A5]/,\r\n  AN: /[\\u0660-\\u0669\\u066B\\u066C]/,\r\n  CS: /[\\u002C\\u002E\\u002F\\u003A\\u00A0]/,\r\n  NSM: /[\\u0300-\\u036F\\u0483-\\u0489\\u0591-\\u05BD\\u05BF\\u05C1\\u05C2]/,\r\n  BN: /[\\u0000-\\u0008\\u000E-\\u001B\\u007F-\\u0084\\u0086-\\u009F]/,\r\n  B: /[\\u000A\\u000D\\u001C-\\u001E\\u0085\\u2029]/,\r\n  S: /[\\u0009\\u000B\\u001F]/,\r\n  WS: /[\\u000C\\u0020\\u1680\\u2000-\\u200A\\u2028\\u205F\\u3000]/,\r\n  ON: /[\\u0021\\u0022\\u0026-\\u002A\\u003B-\\u0040\\u005B-\\u0060\\u007B-\\u007E]/,\r\n};\r\n\r\n// Script detection patterns\r\nconst SCRIPT_PATTERNS = {\r\n  latin: /[\\u0041-\\u005A\\u0061-\\u007A\\u00C0-\\u024F]/,\r\n  arabic: /[\\u0600-\\u06FF\\u0750-\\u077F\\u08A0-\\u08FF]/,\r\n  hebrew: /[\\u0590-\\u05FF]/,\r\n  cyrillic: /[\\u0400-\\u04FF\\u0500-\\u052F]/,\r\n  greek: /[\\u0370-\\u03FF\\u1F00-\\u1FFF]/,\r\n  cjk: /[\\u3400-\\u4DBF\\u4E00-\\u9FFF\\uF900-\\uFAFF]/,\r\n  hiragana: /[\\u3040-\\u309F]/,\r\n  katakana: /[\\u30A0-\\u30FF]/,\r\n  hangul: /[\\uAC00-\\uD7AF\\u1100-\\u11FF\\u3130-\\u318F]/,\r\n  thai: /[\\u0E00-\\u0E7F]/,\r\n  devanagari: /[\\u0900-\\u097F]/,\r\n  bengali: /[\\u0980-\\u09FF]/,\r\n  emoji: /[\\u1F600-\\u1F64F\\u1F300-\\u1F5FF\\u1F680-\\u1F6FF\\u1F700-\\u1F77F]/,\r\n};\r\n\r\n/**\r\n * Enhanced grapheme segmentation using Intl.Segmenter when available\r\n * with fallback to existing graphemeSplit implementation\r\n */\r\nexport function segmentGraphemes(text: string): string[] {\r\n  // Use native Intl.Segmenter if available\r\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\r\n    try {\r\n      const segmenter = new Intl.Segmenter(undefined, {\r\n        granularity: 'grapheme',\r\n      });\r\n      \r\n      const segments = segmenter.segment(text);\r\n      return Array.from(segments, segment => segment.segment);\r\n    } catch (e) {\r\n      // Fallback if Intl.Segmenter fails\r\n    }\r\n  }\r\n  \r\n  // Use existing Fabric.js implementation as fallback\r\n  return graphemeSplit(text);\r\n}\r\n\r\n/**\r\n * Analyze text for bidirectional runs using Unicode BiDi algorithm (simplified)\r\n */\r\nexport function analyzeBiDi(text: string, baseDirection: 'ltr' | 'rtl' = 'ltr'): BiDiRun[] {\r\n  if (!text) return [];\r\n\r\n  const runs: BiDiRun[] = [];\r\n  const chars = Array.from(text);\r\n  let currentRun: BiDiRun | null = null;\r\n\r\n  for (let i = 0; i < chars.length; i++) {\r\n    const char = chars[i];\r\n    const charDirection = getBidiDirection(char, baseDirection);\r\n    \r\n    // Start new run if direction changes\r\n    if (!currentRun || currentRun.direction !== charDirection) {\r\n      if (currentRun) {\r\n        runs.push(currentRun);\r\n      }\r\n      \r\n      currentRun = {\r\n        text: char,\r\n        direction: charDirection,\r\n        level: charDirection === 'rtl' ? 1 : 0,\r\n        start: i,\r\n        end: i + 1,\r\n      };\r\n    } else {\r\n      // Continue current run\r\n      currentRun.text += char;\r\n      currentRun.end = i + 1;\r\n    }\r\n  }\r\n\r\n  // Add final run\r\n  if (currentRun) {\r\n    runs.push(currentRun);\r\n  }\r\n\r\n  return runs.length > 0 ? runs : [{\r\n    text,\r\n    direction: baseDirection,\r\n    level: baseDirection === 'rtl' ? 1 : 0,\r\n    start: 0,\r\n    end: text.length,\r\n  }];\r\n}\r\n\r\n/**\r\n * Detect the primary script of text for font fallback and shaping\r\n */\r\nexport function detectScript(text: string): string {\r\n  const charCounts: Record<string, number> = {};\r\n  const chars = Array.from(text);\r\n\r\n  // Count characters by script\r\n  for (const char of chars) {\r\n    for (const [script, pattern] of Object.entries(SCRIPT_PATTERNS)) {\r\n      if (pattern.test(char)) {\r\n        charCounts[script] = (charCounts[script] || 0) + 1;\r\n        break;\r\n      }\r\n    }\r\n  }\r\n\r\n  // Find dominant script\r\n  let maxCount = 0;\r\n  let dominantScript = 'latin';\r\n\r\n  for (const [script, count] of Object.entries(charCounts)) {\r\n    if (count > maxCount) {\r\n      maxCount = count;\r\n      dominantScript = script;\r\n    }\r\n  }\r\n\r\n  return dominantScript;\r\n}\r\n\r\n/**\r\n * Get script information for layout decisions\r\n */\r\nexport function getScriptInfo(script: string): ScriptInfo {\r\n  const scriptMap: Record<string, ScriptInfo> = {\r\n    latin: {\r\n      script: 'latin',\r\n      direction: 'ltr',\r\n      needsShaping: false,\r\n      complexLayout: false,\r\n    },\r\n    arabic: {\r\n      script: 'arabic',\r\n      direction: 'rtl',\r\n      needsShaping: true,\r\n      complexLayout: true,\r\n    },\r\n    hebrew: {\r\n      script: 'hebrew',\r\n      direction: 'rtl',\r\n      needsShaping: false,\r\n      complexLayout: false,\r\n    },\r\n    thai: {\r\n      script: 'thai',\r\n      direction: 'ltr',\r\n      needsShaping: true,\r\n      complexLayout: true,\r\n    },\r\n    devanagari: {\r\n      script: 'devanagari',\r\n      direction: 'ltr',\r\n      needsShaping: true,\r\n      complexLayout: true,\r\n    },\r\n    cjk: {\r\n      script: 'cjk',\r\n      direction: 'ltr',\r\n      needsShaping: false,\r\n      complexLayout: false,\r\n    },\r\n  };\r\n\r\n  return scriptMap[script] || scriptMap.latin;\r\n}\r\n\r\n/**\r\n * Break words according to language-specific rules\r\n */\r\nexport function breakWords(text: string, locale?: string): string[] {\r\n  // Use native word segmentation if available\r\n  if (typeof Intl !== 'undefined' && 'Segmenter' in Intl) {\r\n    try {\r\n      const segmenter = new Intl.Segmenter(locale, {\r\n        granularity: 'word',\r\n      });\r\n      \r\n      const segments = segmenter.segment(text);\r\n      return Array.from(segments)\r\n        .filter(segment => segment.isWordLike)\r\n        .map(segment => segment.segment);\r\n    } catch (e) {\r\n      // Fallback if Intl.Segmenter fails\r\n    }\r\n  }\r\n\r\n  // Simple fallback word breaking\r\n  return text.split(/\\s+/).filter(word => word.length > 0);\r\n}\r\n\r\n/**\r\n * Character classification functions\r\n */\r\nexport function isWhitespace(grapheme: string): boolean {\r\n  return /\\s/.test(grapheme);\r\n}\r\n\r\nexport function isLineBreak(grapheme: string): boolean {\r\n  return grapheme === '\\n' || grapheme === '\\r' || grapheme === '\\r\\n';\r\n}\r\n\r\nexport function isWordSeparator(grapheme: string): boolean {\r\n  return /[\\s\\-\\u2010-\\u2015]/.test(grapheme);\r\n}\r\n\r\nexport function isPunctuation(grapheme: string): boolean {\r\n  return /[\\p{P}]/u.test(grapheme);\r\n}\r\n\r\nexport function isEmoji(grapheme: string): boolean {\r\n  return /[\\p{Emoji}]/u.test(grapheme);\r\n}\r\n\r\nexport function isRTLCharacter(char: string): boolean {\r\n  return UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char);\r\n}\r\n\r\n/**\r\n * Advanced grapheme cluster handling for complex scripts\r\n */\r\nexport function findGraphemeClusterBoundaries(text: string): number[] {\r\n  const boundaries: number[] = [0];\r\n  const graphemes = segmentGraphemes(text);\r\n  let position = 0;\r\n\r\n  for (const grapheme of graphemes) {\r\n    position += grapheme.length;\r\n    boundaries.push(position);\r\n  }\r\n\r\n  return boundaries;\r\n}\r\n\r\n/**\r\n * Check if a position is at a grapheme cluster boundary\r\n */\r\nexport function isGraphemeClusterBoundary(text: string, position: number): boolean {\r\n  const boundaries = findGraphemeClusterBoundaries(text);\r\n  return boundaries.includes(position);\r\n}\r\n\r\n/**\r\n * Find the next grapheme cluster boundary from a given position\r\n */\r\nexport function findNextGraphemeClusterBoundary(text: string, position: number): number {\r\n  const boundaries = findGraphemeClusterBoundaries(text);\r\n  return boundaries.find(boundary => boundary > position) || text.length;\r\n}\r\n\r\n/**\r\n * Find the previous grapheme cluster boundary from a given position  \r\n */\r\nexport function findPreviousGraphemeClusterBoundary(text: string, position: number): number {\r\n  const boundaries = findGraphemeClusterBoundaries(text);\r\n  const reversedBoundaries = [...boundaries].reverse();\r\n  return reversedBoundaries.find(boundary => boundary < position) || 0;\r\n}\r\n\r\n/**\r\n * Normalize text for consistent processing\r\n */\r\nexport function normalizeText(text: string, form: 'NFC' | 'NFD' | 'NFKC' | 'NFKD' = 'NFC'): string {\r\n  if (typeof text.normalize === 'function') {\r\n    return text.normalize(form);\r\n  }\r\n  return text;\r\n}\r\n\r\n/**\r\n * Detect if text contains complex script characters that need special handling\r\n */\r\nexport function needsComplexLayout(text: string): boolean {\r\n  const complexScripts = ['arabic', 'hebrew', 'thai', 'devanagari', 'bengali'];\r\n  const detectedScript = detectScript(text);\r\n  return complexScripts.includes(detectedScript);\r\n}\r\n\r\n/**\r\n * Get text direction based on content analysis\r\n */\r\nexport function getTextDirection(text: string, fallback: 'ltr' | 'rtl' = 'ltr'): 'ltr' | 'rtl' {\r\n  let rtlChars = 0;\r\n  let ltrChars = 0;\r\n\r\n  for (const char of text) {\r\n    if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\r\n      rtlChars++;\r\n    } else if (UNICODE_CATEGORIES.L.test(char)) {\r\n      ltrChars++;\r\n    }\r\n  }\r\n\r\n  if (rtlChars > ltrChars) return 'rtl';\r\n  if (ltrChars > rtlChars) return 'ltr';\r\n  return fallback;\r\n}\r\n\r\n/**\r\n * Split text into lines respecting complex script rules\r\n */\r\nexport function splitTextLines(text: string): string[] {\r\n  // Handle different line break types\r\n  return text.split(/\\r\\n|\\r|\\n/);\r\n}\r\n\r\n/**\r\n * Check if combining marks or modifiers should be kept with base character\r\n */\r\nexport function isCombiningMark(char: string): boolean {\r\n  return UNICODE_CATEGORIES.NSM.test(char);\r\n}\r\n\r\n/**\r\n * Get bidirectional character type\r\n */\r\nfunction getBidiDirection(char: string, baseDirection: 'ltr' | 'rtl'): 'ltr' | 'rtl' {\r\n  // Strong RTL characters\r\n  if (UNICODE_CATEGORIES.R.test(char) || UNICODE_CATEGORIES.AL.test(char)) {\r\n    return 'rtl';\r\n  }\r\n  \r\n  // Strong LTR characters  \r\n  if (UNICODE_CATEGORIES.L.test(char)) {\r\n    return 'ltr';\r\n  }\r\n  \r\n  // Numbers follow base direction in simplified algorithm\r\n  if (UNICODE_CATEGORIES.EN.test(char) || UNICODE_CATEGORIES.AN.test(char)) {\r\n    return baseDirection;\r\n  }\r\n  \r\n  // Neutral characters follow context\r\n  return baseDirection;\r\n}\r\n\r\n/**\r\n * Advanced word breaking for different scripts\r\n */\r\nexport function breakWordsAdvanced(text: string): string[] {\r\n  const script = detectScript(text);\r\n  \r\n  switch (script) {\r\n    case 'cjk':\r\n      // CJK can break at most characters\r\n      return Array.from(text);\r\n    \r\n    case 'thai':\r\n      // Thai doesn't use spaces - would need dictionary-based breaking\r\n      // For now, fall back to character-based breaking\r\n      return Array.from(text);\r\n    \r\n    case 'arabic':\r\n    case 'hebrew':\r\n      // Handle RTL scripts\r\n      return text.split(/\\s+/).filter(word => word.length > 0);\r\n    \r\n    default:\r\n      // Standard space-based word breaking\r\n      return text.split(/\\s+/).filter(word => word.length > 0);\r\n  }\r\n}\r\n\r\n/**\r\n * Check if text needs right-to-left processing\r\n */\r\nexport function needsRTLProcessing(text: string): boolean {\r\n  for (const char of text) {\r\n    if (isRTLCharacter(char)) {\r\n      return true;\r\n    }\r\n  }\r\n  return false;\r\n}\r\n\r\n/**\r\n * Reverse text for RTL display (simplified - real RTL is more complex)\r\n */\r\nexport function reverseForRTL(text: string): string {\r\n  // This is a simplified implementation\r\n  // Real RTL processing requires full BiDi algorithm\r\n  const graphemes = segmentGraphemes(text);\r\n  return graphemes.reverse().join('');\r\n}\r\n\r\n/**\r\n * Get appropriate line breaking class for character\r\n */\r\nexport function getLineBreakClass(char: string): string {\r\n  // Simplified line breaking classes\r\n  if (/\\s/.test(char)) return 'SP';\r\n  if (/[!-\\/:-@\\[-`{-~]/.test(char)) return 'BA'; // Break after\r\n  if (/[(\\[{]/.test(char)) return 'OP'; // Open punctuation\r\n  if (/[)\\]}]/.test(char)) return 'CL'; // Close punctuation\r\n  if (/\\d/.test(char)) return 'NU'; // Numeric\r\n  if (/[A-Za-z]/.test(char)) return 'AL'; // Alphabetic\r\n  \r\n  return 'AL'; // Default to alphabetic\r\n}\r\n\r\n/**\r\n * Check if line break is allowed between two characters\r\n */\r\nexport function isLineBreakAllowed(before: string, after: string): boolean {\r\n  const beforeClass = getLineBreakClass(before);\r\n  const afterClass = getLineBreakClass(after);\r\n  \r\n  // Simplified line breaking rules\r\n  if (beforeClass === 'OP') return false; // Don't break after opening\r\n  if (afterClass === 'CL') return false; // Don't break before closing\r\n  if (beforeClass === 'SP') return true; // Always allow break after space\r\n  \r\n  return true; // Default allow\r\n}"],"names":["segmentGraphemes","text","Intl","segmenter","Segmenter","undefined","granularity","segments","segment","Array","from","e","graphemeSplit"],"mappings":";;AAAA;AACA;AACA;AACA;AACA;AACA;;;AAuDA;AACA;AACA;AACA;AACO,SAASA,gBAAgBA,CAACC,IAAY,EAAY;AACvD;EACA,IAAI,OAAOC,IAAI,KAAK,WAAW,IAAI,WAAW,IAAIA,IAAI,EAAE;IACtD,IAAI;MACF,MAAMC,SAAS,GAAG,IAAID,IAAI,CAACE,SAAS,CAACC,SAAS,EAAE;AAC9CC,QAAAA,WAAW,EAAE;AACf,OAAC,CAAC;AAEF,MAAA,MAAMC,QAAQ,GAAGJ,SAAS,CAACK,OAAO,CAACP,IAAI,CAAC;MACxC,OAAOQ,KAAK,CAACC,IAAI,CAACH,QAAQ,EAAEC,OAAO,IAAIA,OAAO,CAACA,OAAO,CAAC;IACzD,CAAC,CAAC,OAAOG,CAAC,EAAE;AACV;AAAA,IAAA;AAEJ,EAAA;;AAEA;EACA,OAAOC,aAAa,CAACX,IAAI,CAAC;AAC5B;;;;"}