@oxyhq/core 16.0.0 → 16.1.0

package/src/utils/__tests__/validationUtils.test.ts

@@ -13,6 +13,7 @@ import {
   DISPLAY_NAME_ALLOWED_SCRIPTS,
   DISPLAY_NAME_DISALLOWED_SOURCE,
   DISPLAY_NAME_ORPHANED_MARK_SOURCE,
+  DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE,
   isValidUUID,
   isValidDate,
   isValidFileSize,
@@ -21,6 +22,16 @@ import {
   sanitizeHTML,
   validateAndSanitizeUserInput
 } from '../validationUtils';
+// Imported from the generated module rather than re-exported through
+// `validationUtils`: the broad letters class is an internal operand of the
+// orphaned-mark lookbehind, and the denylist is a generation-time operand
+// already subtracted from the allowlist — neither is part of the policy's
+// public runtime surface.
+import {
+  DISPLAY_NAME_DENIED_SYMBOL_LETTERS_RANGES,
+  DISPLAY_NAME_LETTERS_RANGES,
+  DISPLAY_NAME_NAME_SEPARATORS_RANGES,
+} from '../displayNamePolicyRanges.generated';
 
 describe('Validation Utils', () => {
   describe('isRequiredString', () => {
@@ -214,6 +225,7 @@ describe('Validation Utils', () => {
       expect(PROPERTY_ESCAPE.test(DISPLAY_NAME_ALLOWED_SCRIPTS)).toBe(false);
       expect(PROPERTY_ESCAPE.test(DISPLAY_NAME_DISALLOWED_SOURCE)).toBe(false);
       expect(PROPERTY_ESCAPE.test(DISPLAY_NAME_ORPHANED_MARK_SOURCE)).toBe(false);
+      expect(PROPERTY_ESCAPE.test(DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE)).toBe(false);
     });
 
     it('only uses code-point escapes (\\x / \\u) in the class bodies', () => {
@@ -221,7 +233,11 @@ describe('Validation Utils', () => {
       // code-point escapes. Every backslash-escape in the runtime sources must
       // be one of those forms — never a property escape.
       const escapeLeads = new Set<string>();
-      for (const src of [DISPLAY_NAME_DISALLOWED_SOURCE, DISPLAY_NAME_ORPHANED_MARK_SOURCE]) {
+      for (const src of [
+        DISPLAY_NAME_DISALLOWED_SOURCE,
+        DISPLAY_NAME_ORPHANED_MARK_SOURCE,
+        DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE,
+      ]) {
         for (const [, lead] of src.matchAll(/\\(.)/g)) {
           escapeLeads.add(lead);
         }
@@ -299,6 +315,281 @@ describe('Validation Utils', () => {
     });
   });
 
+  // A Unicode script is not only its letters: `scx=X` also carries script X's
+  // own digits, punctuation and symbols. The allowlist is therefore generated as
+  // `scripts ∩ General_Category L`. Before that intersection existed the class
+  // admitted 1831 non-letter code points, so the documented policy ("digits,
+  // hyphens, dots, symbols are removed") held for ASCII input only — a federated
+  // actor could keep Arabic-Indic digits, a Bengali currency sign, or the U+061C
+  // bidi control in their display name. This block is the regression guard.
+  describe('display-name allowlist admits ONLY letters', () => {
+    const allowed = new RegExp(`[${DISPLAY_NAME_ALLOWED_SCRIPTS}]`, 'u');
+    // Built from the same generated ranges the policy ships, so this assertion
+    // stays property-escape-free and cannot drift from the runtime regex.
+    const letter = new RegExp(`[${DISPLAY_NAME_LETTERS_RANGES}]`, 'u');
+
+    // Scanning the FULL code-point space, not a sample: one leaked invisible
+    // control is a spoofing vector, and a sample cannot prove its absence.
+    const leaks: number[] = [];
+    let admitted = 0;
+    for (let cp = 0; cp <= 0x10ffff; cp++) {
+      if (cp >= 0xd800 && cp <= 0xdfff) continue; // lone surrogates
+      const ch = String.fromCodePoint(cp);
+      if (!allowed.test(ch)) continue;
+      admitted++;
+      if (!letter.test(ch)) leaks.push(cp);
+    }
+
+    it('admits no non-letter code point anywhere in Unicode', () => {
+      const named = leaks
+        .slice(0, 16)
+        .map((cp) => `U+${cp.toString(16).toUpperCase().padStart(4, '0')}`);
+      expect({ count: leaks.length, sample: named }).toEqual({ count: 0, sample: [] });
+    });
+
+    // Vacuity floor: an allowlist that collapsed to (near) nothing would satisfy
+    // the assertion above while silently rejecting every real name.
+    it('still admits the expected bulk of letters', () => {
+      expect(admitted).toBeGreaterThan(100_000);
+    });
+
+    it.each([
+      ['٥', 'Arabic-Indic digit (scx=Arabic, GC=Nd)'],
+      ['०', 'Devanagari digit (scx=Devanagari, GC=Nd)'],
+      ['۞', 'Arabic ornament (scx=Arabic, GC=So)'],
+      ['৳', 'Bengali rupee sign (scx=Bengali, GC=Sc)'],
+      ['।', 'Devanagari danda (scx=Devanagari, GC=Po)'],
+      ['،', 'Arabic comma (scx=Arabic, GC=Po)'],
+      ['؜', 'ARABIC LETTER MARK - invisible bidi control (scx=Arabic, GC=Cf)'],
+      ['᠎', 'MONGOLIAN VOWEL SEPARATOR - invisible (scx=Mongolian, GC=Cf)'],
+    ])('rejects %p, a non-letter from an allowlisted script (%s)', (ch) => {
+      expect(isValidDisplayName(ch)).toBe(false);
+    });
+
+    // The intersection narrows the allowlist, so every previously-accepted real
+    // name must still pass — these are the ones whose scripts own the code
+    // points removed above.
+    it.each([
+      ['مُحَمَد', 'Arabic letters + harakat survive'],
+      ['नमस्ते', 'Devanagari'],
+      ['ᠰᠣᠩᠭᠣᠯ', 'Mongolian'],
+      ['বাংলা', 'Bengali'],
+    ])('still accepts %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(true);
+    });
+  });
+
+  // A character policy classifies FORM, never MEANING. `卐` U+5350 and `卍`
+  // U+534D are CJK Unified Ideographs (GC=Lo, scx=Han) — to Unicode they are the
+  // same kind of thing as `山` in `山田太郎`, so no script- or category-level rule
+  // can reject them without also rejecting every real Chinese, Japanese and
+  // Korean name. They are therefore subtracted from the allowlist by an explicit
+  // code-point denylist at generation time. This block is the regression guard,
+  // and asserts the division of labour: the denylist covers ONLY what the
+  // intersection cannot.
+  describe('display-name denylist — letters that function as symbols', () => {
+    // The glyphs below are visually confusable with each other and with ordinary
+    // ideographs, so pin them to their code points before relying on them.
+    it('the fixtures below really are U+5350 and U+534D', () => {
+      expect('卐'.codePointAt(0)).toBe(0x5350);
+      expect('卍'.codePointAt(0)).toBe(0x534d);
+    });
+
+    it.each([
+      ['卐', 'U+5350 right-facing swastika (GC=Lo, scx=Han)'],
+      ['卍', 'U+534D left-facing swastika (GC=Lo, scx=Han)'],
+      ['卐 Glowniggers 卐', 'the production display name that motivated the denylist'],
+      ['卍 卐', 'both, alone'],
+      ['山田卍太郎', 'embedded mid-name, between real Han letters'],
+    ])('rejects %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(false);
+    });
+
+    // The whole point of a per-code-point denylist instead of a script rule.
+    it.each([
+      ['山田太郎', 'Han (Japanese)'],
+      ['김철수', 'Hangul (Korean)'],
+      ['王小明', 'Han (Chinese)'],
+      ['田中\u{20000}', 'Han incl. astral CJK Extension B'],
+      // The four immediate neighbours of the denied pair: the subtraction must
+      // punch out exactly two code points, not a range around them.
+      ['卌', 'U+534C, immediately below U+534D'],
+      ['华', 'U+534E, immediately above U+534D (as in 中华)'],
+      ['协', 'U+534F, immediately below U+5350'],
+      ['卑', 'U+5351, immediately above U+5350'],
+    ])('still accepts %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(true);
+    });
+
+    // Enumerate what is ACTUALLY denied from the generated class rather than
+    // trusting the literals above, so a future entry that is added to the list
+    // but not enforced by the emitted allowlist fails here.
+    const denied = new RegExp(`[${DISPLAY_NAME_DENIED_SYMBOL_LETTERS_RANGES}]`, 'u');
+    const deniedCodePoints: number[] = [];
+    for (let cp = 0; cp <= 0x10ffff; cp++) {
+      if (cp >= 0xd800 && cp <= 0xdfff) continue; // lone surrogates
+      if (denied.test(String.fromCodePoint(cp))) deniedCodePoints.push(cp);
+    }
+
+    it('denies exactly the two swastika ideographs', () => {
+      expect(deniedCodePoints).toEqual([0x534d, 0x5350]);
+    });
+
+    it('rejects every code point on the denylist', () => {
+      const accepted = deniedCodePoints.filter((cp) =>
+        isValidDisplayName(String.fromCodePoint(cp))
+      );
+      expect(
+        accepted.map((cp) => `U+${cp.toString(16).toUpperCase().padStart(4, '0')}`)
+      ).toEqual([]);
+      // Vacuity floor: an empty denylist would satisfy the assertion above while
+      // enforcing nothing.
+      expect(deniedCodePoints.length).toBeGreaterThanOrEqual(2);
+    });
+
+    // Hermes safety for the new class (see the property-escape block above).
+    it('the denylist ranges contain NO Unicode property escape', () => {
+      expect(/\\[pP]\{/.test(DISPLAY_NAME_DENIED_SYMBOL_LETTERS_RANGES)).toBe(false);
+      expect(() => new RegExp(`[${DISPLAY_NAME_DENIED_SYMBOL_LETTERS_RANGES}]`, 'u')).not.toThrow();
+    });
+
+    // Division of labour: the Tibetan svasti signs look like the entries above
+    // but are GC=So, so the `scripts ∩ General_Category L` intersection already
+    // excludes them. They are rejected — and deliberately NOT on the denylist,
+    // where they would be dead weight that reads as load-bearing. The generator
+    // fails the build if such a redundant entry is added.
+    it.each([
+      ['࿕', 'U+0FD5 RIGHT-FACING SVASTI SIGN'],
+      ['࿖', 'U+0FD6 LEFT-FACING SVASTI SIGN'],
+      ['࿗', 'U+0FD7 RIGHT-FACING SVASTI SIGN WITH DOTS'],
+      ['࿘', 'U+0FD8 LEFT-FACING SVASTI SIGN WITH DOTS'],
+    ])('rejects %p (%s) via the intersection, not the denylist', (ch) => {
+      expect(isValidDisplayName(ch)).toBe(false);
+      expect(denied.test(ch)).toBe(false);
+    });
+  });
+
+  // Four punctuation code points JOIN two letters inside one real name. All are
+  // General_Category P, so `scripts ∩ L` strips them by default and stripping
+  // SPLITS the name (`Codeur·euses` → `Codeur euses`). They are re-admitted, but
+  // ONLY between two letters — the same characters are also used as ornament
+  // (a trailing `Roberto ·`), which must keep being trimmed. The conditional is
+  // the whole rule, so it is tested from both sides.
+  describe('display-name name separators — allowed only between letters', () => {
+    // Visually confusable with each other and with ASCII punctuation.
+    it('the fixtures below really are U+00B7 U+05BE U+0F0B U+30FB', () => {
+      expect('·'.codePointAt(0)).toBe(0x00b7);
+      expect('־'.codePointAt(0)).toBe(0x05be);
+      expect('་'.codePointAt(0)).toBe(0x0f0b);
+      expect('・'.codePointAt(0)).toBe(0x30fb);
+    });
+
+    it.each([
+      ['Codeur·euses en Liberté', 'U+00B7 French inclusive writing (production value)'],
+      ['Codeur·euses', 'U+00B7 bare'],
+      ['Pouet·te', 'U+00B7 short form'],
+      ['お坐・エガード', 'U+30FB Japanese name separator (production value)'],
+      ['אייר אברמסקי־קרוננברג', 'U+05BE Hebrew maqaf compound surname (production value)'],
+      ['འོད་ཟེར', 'U+0F0B Tibetan intersyllabic tsheg (production value)'],
+    ])('accepts letter-flanked %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(true);
+    });
+
+    // A base letter can carry a combining mark, and that mark sits between the
+    // letter and the separator. If the flanking test only looked for a letter,
+    // these would be misread as unflanked and the separator stripped. Both
+    // fixtures use marks that do NOT recompose under NFC, so the mark really
+    // reaches the flanking test instead of being folded into a precomposed
+    // letter — the Tibetan one is the ordinary shape of Tibetan text.
+    it.each([
+      ['ཀི་ཁ', 'Tibetan letter + vowel sign U+0F72 + tsheg + letter'],
+      ['مُ·م', 'Arabic letter + damma U+064F + middle dot + letter'],
+    ])('accepts %p where a combining mark precedes the separator (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(true);
+    });
+
+    it.each([
+      ['Roberto ·', 'trailing ornament (production value)'],
+      ['Michał rysiek Woźniak ·', 'trailing ornament (production value)'],
+      ['·Roberto', 'leading'],
+      ['a··b', 'doubled — neither is letter-flanked on both sides'],
+      ['·', 'alone'],
+      ['お坐・', 'trailing katakana middle dot'],
+      ['・エガード', 'leading katakana middle dot'],
+      ['a ·b', 'space on the left'],
+      ['a· b', 'space on the right'],
+    ])('rejects unflanked %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(false);
+    });
+
+    // Enumerate what is ACTUALLY admitted from the generated class rather than
+    // trusting the literals above, and exercise both sides of the rule for each.
+    const separator = new RegExp(`[${DISPLAY_NAME_NAME_SEPARATORS_RANGES}]`, 'u');
+    const separatorCodePoints: number[] = [];
+    for (let cp = 0; cp <= 0x10ffff; cp++) {
+      if (cp >= 0xd800 && cp <= 0xdfff) continue; // lone surrogates
+      if (separator.test(String.fromCodePoint(cp))) separatorCodePoints.push(cp);
+    }
+
+    it('admits exactly the four name separators', () => {
+      expect(separatorCodePoints).toEqual([0x00b7, 0x05be, 0x0f0b, 0x30fb]);
+    });
+
+    it('every separator is valid between letters and invalid unflanked', () => {
+      const wrongWhenFlanked: string[] = [];
+      const wrongWhenUnflanked: string[] = [];
+      for (const cp of separatorCodePoints) {
+        const ch = String.fromCodePoint(cp);
+        const label = `U+${cp.toString(16).toUpperCase().padStart(4, '0')}`;
+        if (!isValidDisplayName(`a${ch}b`)) wrongWhenFlanked.push(label);
+        if (isValidDisplayName(`a ${ch}`)) wrongWhenUnflanked.push(label);
+      }
+      expect({ wrongWhenFlanked, wrongWhenUnflanked }).toEqual({
+        wrongWhenFlanked: [],
+        wrongWhenUnflanked: [],
+      });
+      // Vacuity floor: an empty separator class would satisfy both loops above.
+      expect(separatorCodePoints.length).toBe(4);
+    });
+
+    // The ASCII hyphen decision is untouched: U+05BE is General_Category Pd like
+    // U+002D, but admitting one says nothing about the other.
+    it('does not admit the ASCII hyphen', () => {
+      expect(separator.test('-')).toBe(false);
+      expect(isValidDisplayName('Jean-Luc')).toBe(false);
+      expect(isValidDisplayName('a-b')).toBe(false);
+    });
+
+    // Hermes safety for the new pattern (see the property-escape block above).
+    it('the unflanked-separator source contains NO Unicode property escape', () => {
+      expect(/\\[pP]\{/.test(DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE)).toBe(false);
+      expect(/\\[pP]\{/.test(DISPLAY_NAME_NAME_SEPARATORS_RANGES)).toBe(false);
+      expect(() => new RegExp(DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE, 'u')).not.toThrow();
+      // Global variant is what @oxyhq/api compiles for the strip path.
+      expect(() => new RegExp(DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE, 'gu')).not.toThrow();
+    });
+
+    // Nothing the policy already rejected may become valid.
+    it.each([
+      ['卐', 'denied symbol letter'],
+      ['卍', 'denied symbol letter'],
+      ['Agent007', 'digit'],
+      ['J.R.', 'dot'],
+      ['ᯅ', 'non-allowlisted script'],
+      ['nixCraft \u{1f427}', 'emoji'],
+    ])('still rejects %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(false);
+    });
+
+    it.each([
+      ['山田太郎', 'Han'],
+      ['김철수', 'Hangul'],
+      ["Renée O'Brien", 'Latin with accent + apostrophe'],
+    ])('still accepts %p (%s)', (name) => {
+      expect(isValidDisplayName(name)).toBe(true);
+    });
+  });
+
   describe('isValidUUID', () => {
     it('should return true for valid UUIDs', () => {
       expect(isValidUUID('123e4567-e89b-12d3-a456-426614174000')).toBe(true);

package/src/utils/displayNamePolicyRanges.generated.ts

@@ -17,7 +17,30 @@
  *
  * Classes captured (regexpu-core, u-mode):
  *   - DISPLAY_NAME_ALLOWED_SCRIPTS_RANGES: the 30-script Script_Extensions
- *     allowlist (scx=Latin, scx=Greek, … scx=Han).
+ *     allowlist (scx=Latin, scx=Greek, … scx=Han) INTERSECTED with
+ *     General_Category L, MINUS an explicit code-point denylist. The
+ *     intersection is load-bearing: `scx=X` also carries script X's digits,
+ *     punctuation and symbols, so without it the class admitted 1831 non-letter
+ *     code points — script digits, 1082 symbols, and 9 invisible format/control
+ *     characters including the U+061C bidi control. The generator fails if any
+ *     non-letter survives. The denylist covers the opposite case: code points
+ *     that ARE letters of an allowlisted script yet function as hate symbols
+ *     (卐 卍) — no script-level or category-level rule can exclude
+ *     them without also rejecting every real Chinese, Japanese and Korean name.
+ *   - DISPLAY_NAME_DENIED_SYMBOL_LETTERS_RANGES: that denylist on its own. NOT
+ *     used to build any runtime regex — the code points are already subtracted
+ *     from the allowlist above, so the policy enforces them with no extra probe.
+ *     It is emitted so tests can enumerate what is denied and assert each entry
+ *     is actually rejected.
+ *   - DISPLAY_NAME_NAME_SEPARATORS_RANGES: the four punctuation code points that
+ *     JOIN two letters inside one real name
+ *     (· ־ ་ ・). All are General_Category P,
+ *     so the letters intersection strips them by default; they are re-admitted
+ *     ONLY between two letters. Unlike the denylist, this class DOES build a
+ *     runtime regex — the conditional half lives in
+ *     `DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE`, which strips any of them that
+ *     is not letter-flanked on both sides, so decorative use
+ *     (a trailing `Roberto ·`) keeps being trimmed.
  *   - DISPLAY_NAME_COMBINING_MARKS_RANGES: General_Category M (combining marks).
  *   - DISPLAY_NAME_SPACE_SEPARATORS_RANGES: General_Category Zs (space
  *     separators).
@@ -28,7 +51,7 @@
  */
 
 export const DISPLAY_NAME_ALLOWED_SCRIPTS_RANGES =
-  'A-Za-z\\xAA\\xB7\\xBA\\xC0-\\xD6\\xD8-\\xF6\\xF8-\\u02B8\\u02BC\\u02C7\\u02C9-\\u02CB\\u02CD\\u02D7\\u02D9\\u02E0-\\u02E4\\u02EA\\u02EB\\u0300-\\u030E\\u0310\\u0311\\u0313\\u0323-\\u0325\\u032D\\u032E\\u0330\\u0331\\u0342\\u0345\\u0358\\u035E\\u0363-\\u0377\\u037A-\\u037D\\u037F\\u0384\\u0386\\u0388-\\u038A\\u038C\\u038E-\\u03A1\\u03A3-\\u03E1\\u03F0-\\u052F\\u0531-\\u0556\\u0559-\\u058A\\u058D-\\u058F\\u0591-\\u05C7\\u05D0-\\u05EA\\u05EF-\\u05F4\\u0600-\\u0604\\u0606-\\u06DC\\u06DE-\\u06FF\\u0750-\\u07B1\\u0870-\\u0891\\u0897-\\u08E1\\u08E3-\\u0952\\u0955-\\u0983\\u0985-\\u098C\\u098F\\u0990\\u0993-\\u09A8\\u09AA-\\u09B0\\u09B2\\u09B6-\\u09B9\\u09BC-\\u09C4\\u09C7\\u09C8\\u09CB-\\u09CE\\u09D7\\u09DC\\u09DD\\u09DF-\\u09E3\\u09E6-\\u09FE\\u0A01-\\u0A03\\u0A05-\\u0A0A\\u0A0F\\u0A10\\u0A13-\\u0A28\\u0A2A-\\u0A30\\u0A32\\u0A33\\u0A35\\u0A36\\u0A38\\u0A39\\u0A3C\\u0A3E-\\u0A42\\u0A47\\u0A48\\u0A4B-\\u0A4D\\u0A51\\u0A59-\\u0A5C\\u0A5E\\u0A66-\\u0A76\\u0A81-\\u0A83\\u0A85-\\u0A8D\\u0A8F-\\u0A91\\u0A93-\\u0AA8\\u0AAA-\\u0AB0\\u0AB2\\u0AB3\\u0AB5-\\u0AB9\\u0ABC-\\u0AC5\\u0AC7-\\u0AC9\\u0ACB-\\u0ACD\\u0AD0\\u0AE0-\\u0AE3\\u0AE6-\\u0AF1\\u0AF9-\\u0AFF\\u0B01-\\u0B03\\u0B05-\\u0B0C\\u0B0F\\u0B10\\u0B13-\\u0B28\\u0B2A-\\u0B30\\u0B32\\u0B33\\u0B35-\\u0B39\\u0B3C-\\u0B44\\u0B47\\u0B48\\u0B4B-\\u0B4D\\u0B55-\\u0B57\\u0B5C\\u0B5D\\u0B5F-\\u0B63\\u0B66-\\u0B77\\u0B82\\u0B83\\u0B85-\\u0B8A\\u0B8E-\\u0B90\\u0B92-\\u0B95\\u0B99\\u0B9A\\u0B9C\\u0B9E\\u0B9F\\u0BA3\\u0BA4\\u0BA8-\\u0BAA\\u0BAE-\\u0BB9\\u0BBE-\\u0BC2\\u0BC6-\\u0BC8\\u0BCA-\\u0BCD\\u0BD0\\u0BD7\\u0BE6-\\u0BFA\\u0C00-\\u0C0C\\u0C0E-\\u0C10\\u0C12-\\u0C28\\u0C2A-\\u0C39\\u0C3C-\\u0C44\\u0C46-\\u0C48\\u0C4A-\\u0C4D\\u0C55\\u0C56\\u0C58-\\u0C5A\\u0C5C\\u0C5D\\u0C60-\\u0C63\\u0C66-\\u0C6F\\u0C77-\\u0C8C\\u0C8E-\\u0C90\\u0C92-\\u0CA8\\u0CAA-\\u0CB3\\u0CB5-\\u0CB9\\u0CBC-\\u0CC4\\u0CC6-\\u0CC8\\u0CCA-\\u0CCD\\u0CD5\\u0CD6\\u0CDC-\\u0CDE\\u0CE0-\\u0CE3\\u0CE6-\\u0CEF\\u0CF1-\\u0CF3\\u0D00-\\u0D0C\\u0D0E-\\u0D10\\u0D12-\\u0D44\\u0D46-\\u0D48\\u0D4A-\\u0D4F\\u0D54-\\u0D63\\u0D66-\\u0D7F\\u0D81-\\u0D83\\u0D85-\\u0D96\\u0D9A-\\u0DB1\\u0DB3-\\u0DBB\\u0DBD\\u0DC0-\\u0DC6\\u0DCA\\u0DCF-\\u0DD4\\u0DD6\\u0DD8-\\u0DDF\\u0DE6-\\u0DEF\\u0DF2-\\u0DF4\\u0E01-\\u0E3A\\u0E40-\\u0E5B\\u0E81\\u0E82\\u0E84\\u0E86-\\u0E8A\\u0E8C-\\u0EA3\\u0EA5\\u0EA7-\\u0EBD\\u0EC0-\\u0EC4\\u0EC6\\u0EC8-\\u0ECE\\u0ED0-\\u0ED9\\u0EDC-\\u0EDF\\u0F00-\\u0F47\\u0F49-\\u0F6C\\u0F71-\\u0F97\\u0F99-\\u0FBC\\u0FBE-\\u0FCC\\u0FCE-\\u0FD4\\u0FD9\\u0FDA\\u1000-\\u10C5\\u10C7\\u10CD\\u10D0-\\u1248\\u124A-\\u124D\\u1250-\\u1256\\u1258\\u125A-\\u125D\\u1260-\\u1288\\u128A-\\u128D\\u1290-\\u12B0\\u12B2-\\u12B5\\u12B8-\\u12BE\\u12C0\\u12C2-\\u12C5\\u12C8-\\u12D6\\u12D8-\\u1310\\u1312-\\u1315\\u1318-\\u135A\\u135D-\\u137C\\u1380-\\u1399\\u13A0-\\u13F5\\u13F8-\\u13FD\\u1780-\\u17DD\\u17E0-\\u17E9\\u17F0-\\u17F9\\u1800-\\u1819\\u1820-\\u1878\\u1880-\\u18AA\\u19E0-\\u19FF\\u1C80-\\u1C8A\\u1C90-\\u1CBA\\u1CBD-\\u1CBF\\u1CD0-\\u1CF9\\u1D00-\\u1DC1\\u1DF8\\u1E00-\\u1F15\\u1F18-\\u1F1D\\u1F20-\\u1F45\\u1F48-\\u1F4D\\u1F50-\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F-\\u1F7D\\u1F80-\\u1FB4\\u1FB6-\\u1FC4\\u1FC6-\\u1FD3\\u1FD6-\\u1FDB\\u1FDD-\\u1FEF\\u1FF2-\\u1FF4\\u1FF6-\\u1FFE\\u202F\\u204F\\u205A\\u205D\\u2071\\u207F\\u2090-\\u209C\\u20F0\\u2126\\u212A\\u212B\\u2132\\u214E\\u2160-\\u2188\\u2C60-\\u2C7F\\u2D00-\\u2D25\\u2D27\\u2D2D\\u2D80-\\u2D96\\u2DA0-\\u2DA6\\u2DA8-\\u2DAE\\u2DB0-\\u2DB6\\u2DB8-\\u2DBE\\u2DC0-\\u2DC6\\u2DC8-\\u2DCE\\u2DD0-\\u2DD6\\u2DD8-\\u2DDE\\u2DE0-\\u2DFF\\u2E17\\u2E31\\u2E41\\u2E43\\u2E80-\\u2E99\\u2E9B-\\u2EF3\\u2F00-\\u2FD5\\u2FF0-\\u2FFF\\u3001-\\u3003\\u3005-\\u3011\\u3013-\\u301F\\u302A-\\u3035\\u3021-\\u302D\\u3037-\\u303F\\u3041-\\u3096\\u3099-\\u30FF\\u3105-\\u312F\\u3131-\\u318E\\u3190-\\u31E5\\u31EF-\\u321E\\u3220-\\u3247\\u3260-\\u327E\\u3280-\\u32B0\\u32C0-\\u32CB\\u32D0-\\u3370\\u337B-\\u337F\\u33E0-\\u33FE\\u3400-\\u4DBF\\u4E00-\\u9FFF\\uA640-\\uA69F\\uA700-\\uA707\\uA722-\\uA787\\uA78B-\\uA7DC\\uA7F1-\\uA7FF\\uA830-\\uA839\\uA8E0-\\uA8FF\\uA92E\\uA960-\\uA97C\\uA9E0-\\uA9FE\\uAA60-\\uAA7F\\uAB01-\\uAB06\\uAB09-\\uAB0E\\uAB11-\\uAB16\\uAB20-\\uAB26\\uAB28-\\uAB2E\\uAB30-\\uAB5A\\uAB5C-\\uAB69\\uAB70-\\uABBF\\uAC00-\\uD7A3\\uD7B0-\\uD7C6\\uD7CB-\\uD7FB\\uF900-\\uFA6D\\uFA70-\\uFAD9\\uFB00-\\uFB06\\uFB13-\\uFB17\\uFB1D-\\uFB36\\uFB38-\\uFB3C\\uFB3E\\uFB40\\uFB41\\uFB43\\uFB44\\uFB46-\\uFDCF\\uFDF0-\\uFDFF\\uFE2E\\uFE2F\\uFE45\\uFE46\\uFE70-\\uFE74\\uFE76-\\uFEFC\\uFF21-\\uFF3A\\uFF41-\\uFF5A\\uFF61-\\uFFBE\\uFFC2-\\uFFC7\\uFFCA-\\uFFCF\\uFFD2-\\uFFD7\\uFFDA-\\uFFDC\\u{10140}-\\u{1018E}\\u{101A0}\\u{102E0}-\\u{102FB}\\u{10780}-\\u{10785}\\u{10787}-\\u{107B0}\\u{107B2}-\\u{107BA}\\u{10E60}-\\u{10E7E}\\u{10EC2}-\\u{10EC7}\\u{10ED0}-\\u{10ED8}\\u{10EFA}-\\u{10EFF}\\u{111E1}-\\u{111F4}\\u{11301}\\u{11303}\\u{1133B}\\u{1133C}\\u{11660}-\\u{1166C}\\u{116D0}-\\u{116E3}\\u{11B00}-\\u{11B09}\\u{11FC0}-\\u{11FF1}\\u{11FFF}\\u{16FE2}\\u{16FE3}\\u{16FF0}-\\u{16FF6}\\u{1AFF0}-\\u{1AFF3}\\u{1AFF5}-\\u{1AFFB}\\u{1AFFD}\\u{1AFFE}\\u{1B000}-\\u{1B122}\\u{1B132}\\u{1B150}-\\u{1B152}\\u{1B155}\\u{1B164}-\\u{1B167}\\u{1D200}-\\u{1D245}\\u{1D360}-\\u{1D371}\\u{1DF00}-\\u{1DF1E}\\u{1DF25}-\\u{1DF2A}\\u{1E030}-\\u{1E06D}\\u{1E08F}\\u{1E7E0}-\\u{1E7E6}\\u{1E7E8}-\\u{1E7EB}\\u{1E7ED}\\u{1E7EE}\\u{1E7F0}-\\u{1E7FE}\\u{1EE00}-\\u{1EE03}\\u{1EE05}-\\u{1EE1F}\\u{1EE21}\\u{1EE22}\\u{1EE24}\\u{1EE27}\\u{1EE29}-\\u{1EE32}\\u{1EE34}-\\u{1EE37}\\u{1EE39}\\u{1EE3B}\\u{1EE42}\\u{1EE47}\\u{1EE49}\\u{1EE4B}\\u{1EE4D}-\\u{1EE4F}\\u{1EE51}\\u{1EE52}\\u{1EE54}\\u{1EE57}\\u{1EE59}\\u{1EE5B}\\u{1EE5D}\\u{1EE5F}\\u{1EE61}\\u{1EE62}\\u{1EE64}\\u{1EE67}-\\u{1EE6A}\\u{1EE6C}-\\u{1EE72}\\u{1EE74}-\\u{1EE77}\\u{1EE79}-\\u{1EE7C}\\u{1EE7E}\\u{1EE80}-\\u{1EE89}\\u{1EE8B}-\\u{1EE9B}\\u{1EEA1}-\\u{1EEA3}\\u{1EEA5}-\\u{1EEA9}\\u{1EEAB}-\\u{1EEBB}\\u{1EEF0}\\u{1EEF1}\\u{1F200}\\u{1F250}\\u{1F251}\\u{20000}-\\u{2A6DF}\\u{2A700}-\\u{2B81D}\\u{2B820}-\\u{2CEAD}\\u{2CEB0}-\\u{2EBE0}\\u{2EBF0}-\\u{2EE5D}\\u{2F800}-\\u{2FA1D}\\u{30000}-\\u{3134A}\\u{31350}-\\u{33479}';
+  'A-Za-z\\xAA\\xBA\\xC0-\\xD6\\xD8-\\xF6\\xF8-\\u02B8\\u02BC\\u02C7\\u02C9-\\u02CB\\u02CD\\u02E0-\\u02E4\\u0370-\\u0374\\u0376\\u0377\\u037A-\\u037D\\u037F\\u0386\\u0388-\\u038A\\u038C\\u038E-\\u03A1\\u03A3-\\u03E1\\u03F0-\\u03F5\\u03F7-\\u0481\\u048A-\\u052F\\u0531-\\u0556\\u0559\\u0560-\\u0588\\u05D0-\\u05EA\\u05EF-\\u05F2\\u0620-\\u064A\\u066E\\u066F\\u0671-\\u06D3\\u06D5\\u06E5\\u06E6\\u06EE\\u06EF\\u06FA-\\u06FC\\u06FF\\u0750-\\u07A5\\u07B1\\u0870-\\u0887\\u0889-\\u088F\\u08A0-\\u08C9\\u0904-\\u0939\\u093D\\u0950\\u0958-\\u0961\\u0971-\\u0980\\u0985-\\u098C\\u098F\\u0990\\u0993-\\u09A8\\u09AA-\\u09B0\\u09B2\\u09B6-\\u09B9\\u09BD\\u09CE\\u09DC\\u09DD\\u09DF-\\u09E1\\u09F0\\u09F1\\u09FC\\u0A05-\\u0A0A\\u0A0F\\u0A10\\u0A13-\\u0A28\\u0A2A-\\u0A30\\u0A32\\u0A33\\u0A35\\u0A36\\u0A38\\u0A39\\u0A59-\\u0A5C\\u0A5E\\u0A72-\\u0A74\\u0A85-\\u0A8D\\u0A8F-\\u0A91\\u0A93-\\u0AA8\\u0AAA-\\u0AB0\\u0AB2\\u0AB3\\u0AB5-\\u0AB9\\u0ABD\\u0AD0\\u0AE0\\u0AE1\\u0AF9\\u0B05-\\u0B0C\\u0B0F\\u0B10\\u0B13-\\u0B28\\u0B2A-\\u0B30\\u0B32\\u0B33\\u0B35-\\u0B39\\u0B3D\\u0B5C\\u0B5D\\u0B5F-\\u0B61\\u0B71\\u0B83\\u0B85-\\u0B8A\\u0B8E-\\u0B90\\u0B92-\\u0B95\\u0B99\\u0B9A\\u0B9C\\u0B9E\\u0B9F\\u0BA3\\u0BA4\\u0BA8-\\u0BAA\\u0BAE-\\u0BB9\\u0BD0\\u0C05-\\u0C0C\\u0C0E-\\u0C10\\u0C12-\\u0C28\\u0C2A-\\u0C39\\u0C3D\\u0C58-\\u0C5A\\u0C5C\\u0C5D\\u0C60\\u0C61\\u0C80\\u0C85-\\u0C8C\\u0C8E-\\u0C90\\u0C92-\\u0CA8\\u0CAA-\\u0CB3\\u0CB5-\\u0CB9\\u0CBD\\u0CDC-\\u0CDE\\u0CE0\\u0CE1\\u0CF1\\u0CF2\\u0D04-\\u0D0C\\u0D0E-\\u0D10\\u0D12-\\u0D3A\\u0D3D\\u0D4E\\u0D54-\\u0D56\\u0D5F-\\u0D61\\u0D7A-\\u0D7F\\u0D85-\\u0D96\\u0D9A-\\u0DB1\\u0DB3-\\u0DBB\\u0DBD\\u0DC0-\\u0DC6\\u0E01-\\u0E30\\u0E32\\u0E33\\u0E40-\\u0E46\\u0E81\\u0E82\\u0E84\\u0E86-\\u0E8A\\u0E8C-\\u0EA3\\u0EA5\\u0EA7-\\u0EB0\\u0EB2\\u0EB3\\u0EBD\\u0EC0-\\u0EC4\\u0EC6\\u0EDC-\\u0EDF\\u0F00\\u0F40-\\u0F47\\u0F49-\\u0F6C\\u0F88-\\u0F8C\\u1000-\\u102A\\u103F\\u1050-\\u1055\\u105A-\\u105D\\u1061\\u1065\\u1066\\u106E-\\u1070\\u1075-\\u1081\\u108E\\u10A0-\\u10C5\\u10C7\\u10CD\\u10D0-\\u10FA\\u10FC-\\u1248\\u124A-\\u124D\\u1250-\\u1256\\u1258\\u125A-\\u125D\\u1260-\\u1288\\u128A-\\u128D\\u1290-\\u12B0\\u12B2-\\u12B5\\u12B8-\\u12BE\\u12C0\\u12C2-\\u12C5\\u12C8-\\u12D6\\u12D8-\\u1310\\u1312-\\u1315\\u1318-\\u135A\\u1380-\\u138F\\u13A0-\\u13F5\\u13F8-\\u13FD\\u1780-\\u17B3\\u17D7\\u17DC\\u1820-\\u1878\\u1880-\\u1884\\u1887-\\u18A8\\u18AA\\u1C80-\\u1C8A\\u1C90-\\u1CBA\\u1CBD-\\u1CBF\\u1CE9-\\u1CEC\\u1CEE-\\u1CF3\\u1CF5\\u1CF6\\u1D00-\\u1DBF\\u1E00-\\u1F15\\u1F18-\\u1F1D\\u1F20-\\u1F45\\u1F48-\\u1F4D\\u1F50-\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F-\\u1F7D\\u1F80-\\u1FB4\\u1FB6-\\u1FBC\\u1FBE\\u1FC2-\\u1FC4\\u1FC6-\\u1FCC\\u1FD0-\\u1FD3\\u1FD6-\\u1FDB\\u1FE0-\\u1FEC\\u1FF2-\\u1FF4\\u1FF6-\\u1FFC\\u2071\\u207F\\u2090-\\u209C\\u2126\\u212A\\u212B\\u2132\\u214E\\u2183\\u2184\\u2C60-\\u2C7F\\u2D00-\\u2D25\\u2D27\\u2D2D\\u2D80-\\u2D96\\u2DA0-\\u2DA6\\u2DA8-\\u2DAE\\u2DB0-\\u2DB6\\u2DB8-\\u2DBE\\u2DC0-\\u2DC6\\u2DC8-\\u2DCE\\u2DD0-\\u2DD6\\u2DD8-\\u2DDE\\u3005\\u3006\\u3031-\\u3035\\u303B\\u303C\\u3041-\\u3096\\u309D-\\u309F\\u30A1-\\u30FA\\u30FC-\\u30FF\\u3105-\\u312F\\u3131-\\u318E\\u31A0-\\u31BF\\u31F0-\\u31FF\\u3400-\\u4DBF\\u4E00-\\u534C\\u534E\\u534F\\u5351-\\u9FFF\\uA640-\\uA66E\\uA67F-\\uA69D\\uA722-\\uA787\\uA78B-\\uA7DC\\uA7F1-\\uA7FF\\uA8F2-\\uA8F7\\uA8FB\\uA8FD\\uA8FE\\uA960-\\uA97C\\uA9E0-\\uA9E4\\uA9E6-\\uA9EF\\uA9FA-\\uA9FE\\uAA60-\\uAA76\\uAA7A\\uAA7E\\uAA7F\\uAB01-\\uAB06\\uAB09-\\uAB0E\\uAB11-\\uAB16\\uAB20-\\uAB26\\uAB28-\\uAB2E\\uAB30-\\uAB5A\\uAB5C-\\uAB69\\uAB70-\\uABBF\\uAC00-\\uD7A3\\uD7B0-\\uD7C6\\uD7CB-\\uD7FB\\uF900-\\uFA6D\\uFA70-\\uFAD9\\uFB00-\\uFB06\\uFB13-\\uFB17\\uFB1D\\uFB1F-\\uFB28\\uFB2A-\\uFB36\\uFB38-\\uFB3C\\uFB3E\\uFB40\\uFB41\\uFB43\\uFB44\\uFB46-\\uFBB1\\uFBD3-\\uFD3D\\uFD50-\\uFD8F\\uFD92-\\uFDC7\\uFDF0-\\uFDFB\\uFE70-\\uFE74\\uFE76-\\uFEFC\\uFF21-\\uFF3A\\uFF41-\\uFF5A\\uFF66-\\uFFBE\\uFFC2-\\uFFC7\\uFFCA-\\uFFCF\\uFFD2-\\uFFD7\\uFFDA-\\uFFDC\\u{10780}-\\u{10785}\\u{10787}-\\u{107B0}\\u{107B2}-\\u{107BA}\\u{10EC2}-\\u{10EC7}\\u{16FE3}\\u{16FF2}\\u{16FF3}\\u{1AFF0}-\\u{1AFF3}\\u{1AFF5}-\\u{1AFFB}\\u{1AFFD}\\u{1AFFE}\\u{1B000}-\\u{1B122}\\u{1B132}\\u{1B150}-\\u{1B152}\\u{1B155}\\u{1B164}-\\u{1B167}\\u{1DF00}-\\u{1DF1E}\\u{1DF25}-\\u{1DF2A}\\u{1E030}-\\u{1E06D}\\u{1E7E0}-\\u{1E7E6}\\u{1E7E8}-\\u{1E7EB}\\u{1E7ED}\\u{1E7EE}\\u{1E7F0}-\\u{1E7FE}\\u{1EE00}-\\u{1EE03}\\u{1EE05}-\\u{1EE1F}\\u{1EE21}\\u{1EE22}\\u{1EE24}\\u{1EE27}\\u{1EE29}-\\u{1EE32}\\u{1EE34}-\\u{1EE37}\\u{1EE39}\\u{1EE3B}\\u{1EE42}\\u{1EE47}\\u{1EE49}\\u{1EE4B}\\u{1EE4D}-\\u{1EE4F}\\u{1EE51}\\u{1EE52}\\u{1EE54}\\u{1EE57}\\u{1EE59}\\u{1EE5B}\\u{1EE5D}\\u{1EE5F}\\u{1EE61}\\u{1EE62}\\u{1EE64}\\u{1EE67}-\\u{1EE6A}\\u{1EE6C}-\\u{1EE72}\\u{1EE74}-\\u{1EE77}\\u{1EE79}-\\u{1EE7C}\\u{1EE7E}\\u{1EE80}-\\u{1EE89}\\u{1EE8B}-\\u{1EE9B}\\u{1EEA1}-\\u{1EEA3}\\u{1EEA5}-\\u{1EEA9}\\u{1EEAB}-\\u{1EEBB}\\u{20000}-\\u{2A6DF}\\u{2A700}-\\u{2B81D}\\u{2B820}-\\u{2CEAD}\\u{2CEB0}-\\u{2EBE0}\\u{2EBF0}-\\u{2EE5D}\\u{2F800}-\\u{2FA1D}\\u{30000}-\\u{3134A}\\u{31350}-\\u{33479}';
 
 export const DISPLAY_NAME_COMBINING_MARKS_RANGES =
   '\\u0300-\\u036F\\u0483-\\u0489\\u0591-\\u05BD\\u05BF\\u05C1\\u05C2\\u05C4\\u05C5\\u05C7\\u0610-\\u061A\\u064B-\\u065F\\u0670\\u06D6-\\u06DC\\u06DF-\\u06E4\\u06E7\\u06E8\\u06EA-\\u06ED\\u0711\\u0730-\\u074A\\u07A6-\\u07B0\\u07EB-\\u07F3\\u07FD\\u0816-\\u0819\\u081B-\\u0823\\u0825-\\u0827\\u0829-\\u082D\\u0859-\\u085B\\u0897-\\u089F\\u08CA-\\u08E1\\u08E3-\\u0903\\u093A-\\u093C\\u093E-\\u094F\\u0951-\\u0957\\u0962\\u0963\\u0981-\\u0983\\u09BC\\u09BE-\\u09C4\\u09C7\\u09C8\\u09CB-\\u09CD\\u09D7\\u09E2\\u09E3\\u09FE\\u0A01-\\u0A03\\u0A3C\\u0A3E-\\u0A42\\u0A47\\u0A48\\u0A4B-\\u0A4D\\u0A51\\u0A70\\u0A71\\u0A75\\u0A81-\\u0A83\\u0ABC\\u0ABE-\\u0AC5\\u0AC7-\\u0AC9\\u0ACB-\\u0ACD\\u0AE2\\u0AE3\\u0AFA-\\u0AFF\\u0B01-\\u0B03\\u0B3C\\u0B3E-\\u0B44\\u0B47\\u0B48\\u0B4B-\\u0B4D\\u0B55-\\u0B57\\u0B62\\u0B63\\u0B82\\u0BBE-\\u0BC2\\u0BC6-\\u0BC8\\u0BCA-\\u0BCD\\u0BD7\\u0C00-\\u0C04\\u0C3C\\u0C3E-\\u0C44\\u0C46-\\u0C48\\u0C4A-\\u0C4D\\u0C55\\u0C56\\u0C62\\u0C63\\u0C81-\\u0C83\\u0CBC\\u0CBE-\\u0CC4\\u0CC6-\\u0CC8\\u0CCA-\\u0CCD\\u0CD5\\u0CD6\\u0CE2\\u0CE3\\u0CF3\\u0D00-\\u0D03\\u0D3B\\u0D3C\\u0D3E-\\u0D44\\u0D46-\\u0D48\\u0D4A-\\u0D4D\\u0D57\\u0D62\\u0D63\\u0D81-\\u0D83\\u0DCA\\u0DCF-\\u0DD4\\u0DD6\\u0DD8-\\u0DDF\\u0DF2\\u0DF3\\u0E31\\u0E34-\\u0E3A\\u0E47-\\u0E4E\\u0EB1\\u0EB4-\\u0EBC\\u0EC8-\\u0ECE\\u0F18\\u0F19\\u0F35\\u0F37\\u0F39\\u0F3E\\u0F3F\\u0F71-\\u0F84\\u0F86\\u0F87\\u0F8D-\\u0F97\\u0F99-\\u0FBC\\u0FC6\\u102B-\\u103E\\u1056-\\u1059\\u105E-\\u1060\\u1062-\\u1064\\u1067-\\u106D\\u1071-\\u1074\\u1082-\\u108D\\u108F\\u109A-\\u109D\\u135D-\\u135F\\u1712-\\u1715\\u1732-\\u1734\\u1752\\u1753\\u1772\\u1773\\u17B4-\\u17D3\\u17DD\\u180B-\\u180D\\u180F\\u1885\\u1886\\u18A9\\u1920-\\u192B\\u1930-\\u193B\\u1A17-\\u1A1B\\u1A55-\\u1A5E\\u1A60-\\u1A7C\\u1A7F\\u1AB0-\\u1ADD\\u1AE0-\\u1AEB\\u1B00-\\u1B04\\u1B34-\\u1B44\\u1B6B-\\u1B73\\u1B80-\\u1B82\\u1BA1-\\u1BAD\\u1BE6-\\u1BF3\\u1C24-\\u1C37\\u1CD0-\\u1CD2\\u1CD4-\\u1CE8\\u1CED\\u1CF4\\u1CF7-\\u1CF9\\u1DC0-\\u1DFF\\u20D0-\\u20F0\\u2CEF-\\u2CF1\\u2D7F\\u2DE0-\\u2DFF\\u302A-\\u302F\\u3099\\u309A\\uA66F-\\uA672\\uA674-\\uA67D\\uA69E\\uA69F\\uA6F0\\uA6F1\\uA802\\uA806\\uA80B\\uA823-\\uA827\\uA82C\\uA880\\uA881\\uA8B4-\\uA8C5\\uA8E0-\\uA8F1\\uA8FF\\uA926-\\uA92D\\uA947-\\uA953\\uA980-\\uA983\\uA9B3-\\uA9C0\\uA9E5\\uAA29-\\uAA36\\uAA43\\uAA4C\\uAA4D\\uAA7B-\\uAA7D\\uAAB0\\uAAB2-\\uAAB4\\uAAB7\\uAAB8\\uAABE\\uAABF\\uAAC1\\uAAEB-\\uAAEF\\uAAF5\\uAAF6\\uABE3-\\uABEA\\uABEC\\uABED\\uFB1E\\uFE00-\\uFE0F\\uFE20-\\uFE2F\\u{101FD}\\u{102E0}\\u{10376}-\\u{1037A}\\u{10A01}-\\u{10A03}\\u{10A05}\\u{10A06}\\u{10A0C}-\\u{10A0F}\\u{10A38}-\\u{10A3A}\\u{10A3F}\\u{10AE5}\\u{10AE6}\\u{10D24}-\\u{10D27}\\u{10D69}-\\u{10D6D}\\u{10EAB}\\u{10EAC}\\u{10EFA}-\\u{10EFF}\\u{10F46}-\\u{10F50}\\u{10F82}-\\u{10F85}\\u{11000}-\\u{11002}\\u{11038}-\\u{11046}\\u{11070}\\u{11073}\\u{11074}\\u{1107F}-\\u{11082}\\u{110B0}-\\u{110BA}\\u{110C2}\\u{11100}-\\u{11102}\\u{11127}-\\u{11134}\\u{11145}\\u{11146}\\u{11173}\\u{11180}-\\u{11182}\\u{111B3}-\\u{111C0}\\u{111C9}-\\u{111CC}\\u{111CE}\\u{111CF}\\u{1122C}-\\u{11237}\\u{1123E}\\u{11241}\\u{112DF}-\\u{112EA}\\u{11300}-\\u{11303}\\u{1133B}\\u{1133C}\\u{1133E}-\\u{11344}\\u{11347}\\u{11348}\\u{1134B}-\\u{1134D}\\u{11357}\\u{11362}\\u{11363}\\u{11366}-\\u{1136C}\\u{11370}-\\u{11374}\\u{113B8}-\\u{113C0}\\u{113C2}\\u{113C5}\\u{113C7}-\\u{113CA}\\u{113CC}-\\u{113D0}\\u{113D2}\\u{113E1}\\u{113E2}\\u{11435}-\\u{11446}\\u{1145E}\\u{114B0}-\\u{114C3}\\u{115AF}-\\u{115B5}\\u{115B8}-\\u{115C0}\\u{115DC}\\u{115DD}\\u{11630}-\\u{11640}\\u{116AB}-\\u{116B7}\\u{1171D}-\\u{1172B}\\u{1182C}-\\u{1183A}\\u{11930}-\\u{11935}\\u{11937}\\u{11938}\\u{1193B}-\\u{1193E}\\u{11940}\\u{11942}\\u{11943}\\u{119D1}-\\u{119D7}\\u{119DA}-\\u{119E0}\\u{119E4}\\u{11A01}-\\u{11A0A}\\u{11A33}-\\u{11A39}\\u{11A3B}-\\u{11A3E}\\u{11A47}\\u{11A51}-\\u{11A5B}\\u{11A8A}-\\u{11A99}\\u{11B60}-\\u{11B67}\\u{11C2F}-\\u{11C36}\\u{11C38}-\\u{11C3F}\\u{11C92}-\\u{11CA7}\\u{11CA9}-\\u{11CB6}\\u{11D31}-\\u{11D36}\\u{11D3A}\\u{11D3C}\\u{11D3D}\\u{11D3F}-\\u{11D45}\\u{11D47}\\u{11D8A}-\\u{11D8E}\\u{11D90}\\u{11D91}\\u{11D93}-\\u{11D97}\\u{11EF3}-\\u{11EF6}\\u{11F00}\\u{11F01}\\u{11F03}\\u{11F34}-\\u{11F3A}\\u{11F3E}-\\u{11F42}\\u{11F5A}\\u{13440}\\u{13447}-\\u{13455}\\u{1611E}-\\u{1612F}\\u{16AF0}-\\u{16AF4}\\u{16B30}-\\u{16B36}\\u{16F4F}\\u{16F51}-\\u{16F87}\\u{16F8F}-\\u{16F92}\\u{16FE4}\\u{16FF0}\\u{16FF1}\\u{1BC9D}\\u{1BC9E}\\u{1CF00}-\\u{1CF2D}\\u{1CF30}-\\u{1CF46}\\u{1D165}-\\u{1D169}\\u{1D16D}-\\u{1D172}\\u{1D17B}-\\u{1D182}\\u{1D185}-\\u{1D18B}\\u{1D1AA}-\\u{1D1AD}\\u{1D242}-\\u{1D244}\\u{1DA00}-\\u{1DA36}\\u{1DA3B}-\\u{1DA6C}\\u{1DA75}\\u{1DA84}\\u{1DA9B}-\\u{1DA9F}\\u{1DAA1}-\\u{1DAAF}\\u{1E000}-\\u{1E006}\\u{1E008}-\\u{1E018}\\u{1E01B}-\\u{1E021}\\u{1E023}\\u{1E024}\\u{1E026}-\\u{1E02A}\\u{1E08F}\\u{1E130}-\\u{1E136}\\u{1E2AE}\\u{1E2EC}-\\u{1E2EF}\\u{1E4EC}-\\u{1E4EF}\\u{1E5EE}\\u{1E5EF}\\u{1E6E3}\\u{1E6E6}\\u{1E6EE}\\u{1E6EF}\\u{1E6F5}\\u{1E8D0}-\\u{1E8D6}\\u{1E944}-\\u{1E94A}\\u{E0100}-\\u{E01EF}';
@@ -38,3 +61,9 @@ export const DISPLAY_NAME_SPACE_SEPARATORS_RANGES =
 
 export const DISPLAY_NAME_LETTERS_RANGES =
   'A-Za-z\\xAA\\xB5\\xBA\\xC0-\\xD6\\xD8-\\xF6\\xF8-\\u02C1\\u02C6-\\u02D1\\u02E0-\\u02E4\\u02EC\\u02EE\\u0370-\\u0374\\u0376\\u0377\\u037A-\\u037D\\u037F\\u0386\\u0388-\\u038A\\u038C\\u038E-\\u03A1\\u03A3-\\u03F5\\u03F7-\\u0481\\u048A-\\u052F\\u0531-\\u0556\\u0559\\u0560-\\u0588\\u05D0-\\u05EA\\u05EF-\\u05F2\\u0620-\\u064A\\u066E\\u066F\\u0671-\\u06D3\\u06D5\\u06E5\\u06E6\\u06EE\\u06EF\\u06FA-\\u06FC\\u06FF\\u0710\\u0712-\\u072F\\u074D-\\u07A5\\u07B1\\u07CA-\\u07EA\\u07F4\\u07F5\\u07FA\\u0800-\\u0815\\u081A\\u0824\\u0828\\u0840-\\u0858\\u0860-\\u086A\\u0870-\\u0887\\u0889-\\u088F\\u08A0-\\u08C9\\u0904-\\u0939\\u093D\\u0950\\u0958-\\u0961\\u0971-\\u0980\\u0985-\\u098C\\u098F\\u0990\\u0993-\\u09A8\\u09AA-\\u09B0\\u09B2\\u09B6-\\u09B9\\u09BD\\u09CE\\u09DC\\u09DD\\u09DF-\\u09E1\\u09F0\\u09F1\\u09FC\\u0A05-\\u0A0A\\u0A0F\\u0A10\\u0A13-\\u0A28\\u0A2A-\\u0A30\\u0A32\\u0A33\\u0A35\\u0A36\\u0A38\\u0A39\\u0A59-\\u0A5C\\u0A5E\\u0A72-\\u0A74\\u0A85-\\u0A8D\\u0A8F-\\u0A91\\u0A93-\\u0AA8\\u0AAA-\\u0AB0\\u0AB2\\u0AB3\\u0AB5-\\u0AB9\\u0ABD\\u0AD0\\u0AE0\\u0AE1\\u0AF9\\u0B05-\\u0B0C\\u0B0F\\u0B10\\u0B13-\\u0B28\\u0B2A-\\u0B30\\u0B32\\u0B33\\u0B35-\\u0B39\\u0B3D\\u0B5C\\u0B5D\\u0B5F-\\u0B61\\u0B71\\u0B83\\u0B85-\\u0B8A\\u0B8E-\\u0B90\\u0B92-\\u0B95\\u0B99\\u0B9A\\u0B9C\\u0B9E\\u0B9F\\u0BA3\\u0BA4\\u0BA8-\\u0BAA\\u0BAE-\\u0BB9\\u0BD0\\u0C05-\\u0C0C\\u0C0E-\\u0C10\\u0C12-\\u0C28\\u0C2A-\\u0C39\\u0C3D\\u0C58-\\u0C5A\\u0C5C\\u0C5D\\u0C60\\u0C61\\u0C80\\u0C85-\\u0C8C\\u0C8E-\\u0C90\\u0C92-\\u0CA8\\u0CAA-\\u0CB3\\u0CB5-\\u0CB9\\u0CBD\\u0CDC-\\u0CDE\\u0CE0\\u0CE1\\u0CF1\\u0CF2\\u0D04-\\u0D0C\\u0D0E-\\u0D10\\u0D12-\\u0D3A\\u0D3D\\u0D4E\\u0D54-\\u0D56\\u0D5F-\\u0D61\\u0D7A-\\u0D7F\\u0D85-\\u0D96\\u0D9A-\\u0DB1\\u0DB3-\\u0DBB\\u0DBD\\u0DC0-\\u0DC6\\u0E01-\\u0E30\\u0E32\\u0E33\\u0E40-\\u0E46\\u0E81\\u0E82\\u0E84\\u0E86-\\u0E8A\\u0E8C-\\u0EA3\\u0EA5\\u0EA7-\\u0EB0\\u0EB2\\u0EB3\\u0EBD\\u0EC0-\\u0EC4\\u0EC6\\u0EDC-\\u0EDF\\u0F00\\u0F40-\\u0F47\\u0F49-\\u0F6C\\u0F88-\\u0F8C\\u1000-\\u102A\\u103F\\u1050-\\u1055\\u105A-\\u105D\\u1061\\u1065\\u1066\\u106E-\\u1070\\u1075-\\u1081\\u108E\\u10A0-\\u10C5\\u10C7\\u10CD\\u10D0-\\u10FA\\u10FC-\\u1248\\u124A-\\u124D\\u1250-\\u1256\\u1258\\u125A-\\u125D\\u1260-\\u1288\\u128A-\\u128D\\u1290-\\u12B0\\u12B2-\\u12B5\\u12B8-\\u12BE\\u12C0\\u12C2-\\u12C5\\u12C8-\\u12D6\\u12D8-\\u1310\\u1312-\\u1315\\u1318-\\u135A\\u1380-\\u138F\\u13A0-\\u13F5\\u13F8-\\u13FD\\u1401-\\u166C\\u166F-\\u167F\\u1681-\\u169A\\u16A0-\\u16EA\\u16F1-\\u16F8\\u1700-\\u1711\\u171F-\\u1731\\u1740-\\u1751\\u1760-\\u176C\\u176E-\\u1770\\u1780-\\u17B3\\u17D7\\u17DC\\u1820-\\u1878\\u1880-\\u1884\\u1887-\\u18A8\\u18AA\\u18B0-\\u18F5\\u1900-\\u191E\\u1950-\\u196D\\u1970-\\u1974\\u1980-\\u19AB\\u19B0-\\u19C9\\u1A00-\\u1A16\\u1A20-\\u1A54\\u1AA7\\u1B05-\\u1B33\\u1B45-\\u1B4C\\u1B83-\\u1BA0\\u1BAE\\u1BAF\\u1BBA-\\u1BE5\\u1C00-\\u1C23\\u1C4D-\\u1C4F\\u1C5A-\\u1C7D\\u1C80-\\u1C8A\\u1C90-\\u1CBA\\u1CBD-\\u1CBF\\u1CE9-\\u1CEC\\u1CEE-\\u1CF3\\u1CF5\\u1CF6\\u1CFA\\u1D00-\\u1DBF\\u1E00-\\u1F15\\u1F18-\\u1F1D\\u1F20-\\u1F45\\u1F48-\\u1F4D\\u1F50-\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F-\\u1F7D\\u1F80-\\u1FB4\\u1FB6-\\u1FBC\\u1FBE\\u1FC2-\\u1FC4\\u1FC6-\\u1FCC\\u1FD0-\\u1FD3\\u1FD6-\\u1FDB\\u1FE0-\\u1FEC\\u1FF2-\\u1FF4\\u1FF6-\\u1FFC\\u2071\\u207F\\u2090-\\u209C\\u2102\\u2107\\u210A-\\u2113\\u2115\\u2119-\\u211D\\u2124\\u2126\\u2128\\u212A-\\u212D\\u212F-\\u2139\\u213C-\\u213F\\u2145-\\u2149\\u214E\\u2183\\u2184\\u2C00-\\u2CE4\\u2CEB-\\u2CEE\\u2CF2\\u2CF3\\u2D00-\\u2D25\\u2D27\\u2D2D\\u2D30-\\u2D67\\u2D6F\\u2D80-\\u2D96\\u2DA0-\\u2DA6\\u2DA8-\\u2DAE\\u2DB0-\\u2DB6\\u2DB8-\\u2DBE\\u2DC0-\\u2DC6\\u2DC8-\\u2DCE\\u2DD0-\\u2DD6\\u2DD8-\\u2DDE\\u2E2F\\u3005\\u3006\\u3031-\\u3035\\u303B\\u303C\\u3041-\\u3096\\u309D-\\u309F\\u30A1-\\u30FA\\u30FC-\\u30FF\\u3105-\\u312F\\u3131-\\u318E\\u31A0-\\u31BF\\u31F0-\\u31FF\\u3400-\\u4DBF\\u4E00-\\uA48C\\uA4D0-\\uA4FD\\uA500-\\uA60C\\uA610-\\uA61F\\uA62A\\uA62B\\uA640-\\uA66E\\uA67F-\\uA69D\\uA6A0-\\uA6E5\\uA717-\\uA71F\\uA722-\\uA788\\uA78B-\\uA7DC\\uA7F1-\\uA801\\uA803-\\uA805\\uA807-\\uA80A\\uA80C-\\uA822\\uA840-\\uA873\\uA882-\\uA8B3\\uA8F2-\\uA8F7\\uA8FB\\uA8FD\\uA8FE\\uA90A-\\uA925\\uA930-\\uA946\\uA960-\\uA97C\\uA984-\\uA9B2\\uA9CF\\uA9E0-\\uA9E4\\uA9E6-\\uA9EF\\uA9FA-\\uA9FE\\uAA00-\\uAA28\\uAA40-\\uAA42\\uAA44-\\uAA4B\\uAA60-\\uAA76\\uAA7A\\uAA7E-\\uAAAF\\uAAB1\\uAAB5\\uAAB6\\uAAB9-\\uAABD\\uAAC0\\uAAC2\\uAADB-\\uAADD\\uAAE0-\\uAAEA\\uAAF2-\\uAAF4\\uAB01-\\uAB06\\uAB09-\\uAB0E\\uAB11-\\uAB16\\uAB20-\\uAB26\\uAB28-\\uAB2E\\uAB30-\\uAB5A\\uAB5C-\\uAB69\\uAB70-\\uABE2\\uAC00-\\uD7A3\\uD7B0-\\uD7C6\\uD7CB-\\uD7FB\\uF900-\\uFA6D\\uFA70-\\uFAD9\\uFB00-\\uFB06\\uFB13-\\uFB17\\uFB1D\\uFB1F-\\uFB28\\uFB2A-\\uFB36\\uFB38-\\uFB3C\\uFB3E\\uFB40\\uFB41\\uFB43\\uFB44\\uFB46-\\uFBB1\\uFBD3-\\uFD3D\\uFD50-\\uFD8F\\uFD92-\\uFDC7\\uFDF0-\\uFDFB\\uFE70-\\uFE74\\uFE76-\\uFEFC\\uFF21-\\uFF3A\\uFF41-\\uFF5A\\uFF66-\\uFFBE\\uFFC2-\\uFFC7\\uFFCA-\\uFFCF\\uFFD2-\\uFFD7\\uFFDA-\\uFFDC\\u{10000}-\\u{1000B}\\u{1000D}-\\u{10026}\\u{10028}-\\u{1003A}\\u{1003C}\\u{1003D}\\u{1003F}-\\u{1004D}\\u{10050}-\\u{1005D}\\u{10080}-\\u{100FA}\\u{10280}-\\u{1029C}\\u{102A0}-\\u{102D0}\\u{10300}-\\u{1031F}\\u{1032D}-\\u{10340}\\u{10342}-\\u{10349}\\u{10350}-\\u{10375}\\u{10380}-\\u{1039D}\\u{103A0}-\\u{103C3}\\u{103C8}-\\u{103CF}\\u{10400}-\\u{1049D}\\u{104B0}-\\u{104D3}\\u{104D8}-\\u{104FB}\\u{10500}-\\u{10527}\\u{10530}-\\u{10563}\\u{10570}-\\u{1057A}\\u{1057C}-\\u{1058A}\\u{1058C}-\\u{10592}\\u{10594}\\u{10595}\\u{10597}-\\u{105A1}\\u{105A3}-\\u{105B1}\\u{105B3}-\\u{105B9}\\u{105BB}\\u{105BC}\\u{105C0}-\\u{105F3}\\u{10600}-\\u{10736}\\u{10740}-\\u{10755}\\u{10760}-\\u{10767}\\u{10780}-\\u{10785}\\u{10787}-\\u{107B0}\\u{107B2}-\\u{107BA}\\u{10800}-\\u{10805}\\u{10808}\\u{1080A}-\\u{10835}\\u{10837}\\u{10838}\\u{1083C}\\u{1083F}-\\u{10855}\\u{10860}-\\u{10876}\\u{10880}-\\u{1089E}\\u{108E0}-\\u{108F2}\\u{108F4}\\u{108F5}\\u{10900}-\\u{10915}\\u{10920}-\\u{10939}\\u{10940}-\\u{10959}\\u{10980}-\\u{109B7}\\u{109BE}\\u{109BF}\\u{10A00}\\u{10A10}-\\u{10A13}\\u{10A15}-\\u{10A17}\\u{10A19}-\\u{10A35}\\u{10A60}-\\u{10A7C}\\u{10A80}-\\u{10A9C}\\u{10AC0}-\\u{10AC7}\\u{10AC9}-\\u{10AE4}\\u{10B00}-\\u{10B35}\\u{10B40}-\\u{10B55}\\u{10B60}-\\u{10B72}\\u{10B80}-\\u{10B91}\\u{10C00}-\\u{10C48}\\u{10C80}-\\u{10CB2}\\u{10CC0}-\\u{10CF2}\\u{10D00}-\\u{10D23}\\u{10D4A}-\\u{10D65}\\u{10D6F}-\\u{10D85}\\u{10E80}-\\u{10EA9}\\u{10EB0}\\u{10EB1}\\u{10EC2}-\\u{10EC7}\\u{10F00}-\\u{10F1C}\\u{10F27}\\u{10F30}-\\u{10F45}\\u{10F70}-\\u{10F81}\\u{10FB0}-\\u{10FC4}\\u{10FE0}-\\u{10FF6}\\u{11003}-\\u{11037}\\u{11071}\\u{11072}\\u{11075}\\u{11083}-\\u{110AF}\\u{110D0}-\\u{110E8}\\u{11103}-\\u{11126}\\u{11144}\\u{11147}\\u{11150}-\\u{11172}\\u{11176}\\u{11183}-\\u{111B2}\\u{111C1}-\\u{111C4}\\u{111DA}\\u{111DC}\\u{11200}-\\u{11211}\\u{11213}-\\u{1122B}\\u{1123F}\\u{11240}\\u{11280}-\\u{11286}\\u{11288}\\u{1128A}-\\u{1128D}\\u{1128F}-\\u{1129D}\\u{1129F}-\\u{112A8}\\u{112B0}-\\u{112DE}\\u{11305}-\\u{1130C}\\u{1130F}\\u{11310}\\u{11313}-\\u{11328}\\u{1132A}-\\u{11330}\\u{11332}\\u{11333}\\u{11335}-\\u{11339}\\u{1133D}\\u{11350}\\u{1135D}-\\u{11361}\\u{11380}-\\u{11389}\\u{1138B}\\u{1138E}\\u{11390}-\\u{113B5}\\u{113B7}\\u{113D1}\\u{113D3}\\u{11400}-\\u{11434}\\u{11447}-\\u{1144A}\\u{1145F}-\\u{11461}\\u{11480}-\\u{114AF}\\u{114C4}\\u{114C5}\\u{114C7}\\u{11580}-\\u{115AE}\\u{115D8}-\\u{115DB}\\u{11600}-\\u{1162F}\\u{11644}\\u{11680}-\\u{116AA}\\u{116B8}\\u{11700}-\\u{1171A}\\u{11740}-\\u{11746}\\u{11800}-\\u{1182B}\\u{118A0}-\\u{118DF}\\u{118FF}-\\u{11906}\\u{11909}\\u{1190C}-\\u{11913}\\u{11915}\\u{11916}\\u{11918}-\\u{1192F}\\u{1193F}\\u{11941}\\u{119A0}-\\u{119A7}\\u{119AA}-\\u{119D0}\\u{119E1}\\u{119E3}\\u{11A00}\\u{11A0B}-\\u{11A32}\\u{11A3A}\\u{11A50}\\u{11A5C}-\\u{11A89}\\u{11A9D}\\u{11AB0}-\\u{11AF8}\\u{11BC0}-\\u{11BE0}\\u{11C00}-\\u{11C08}\\u{11C0A}-\\u{11C2E}\\u{11C40}\\u{11C72}-\\u{11C8F}\\u{11D00}-\\u{11D06}\\u{11D08}\\u{11D09}\\u{11D0B}-\\u{11D30}\\u{11D46}\\u{11D60}-\\u{11D65}\\u{11D67}\\u{11D68}\\u{11D6A}-\\u{11D89}\\u{11D98}\\u{11DB0}-\\u{11DDB}\\u{11EE0}-\\u{11EF2}\\u{11F02}\\u{11F04}-\\u{11F10}\\u{11F12}-\\u{11F33}\\u{11FB0}\\u{12000}-\\u{12399}\\u{12480}-\\u{12543}\\u{12F90}-\\u{12FF0}\\u{13000}-\\u{1342F}\\u{13441}-\\u{13446}\\u{13460}-\\u{143FA}\\u{14400}-\\u{14646}\\u{16100}-\\u{1611D}\\u{16800}-\\u{16A38}\\u{16A40}-\\u{16A5E}\\u{16A70}-\\u{16ABE}\\u{16AD0}-\\u{16AED}\\u{16B00}-\\u{16B2F}\\u{16B40}-\\u{16B43}\\u{16B63}-\\u{16B77}\\u{16B7D}-\\u{16B8F}\\u{16D40}-\\u{16D6C}\\u{16E40}-\\u{16E7F}\\u{16EA0}-\\u{16EB8}\\u{16EBB}-\\u{16ED3}\\u{16F00}-\\u{16F4A}\\u{16F50}\\u{16F93}-\\u{16F9F}\\u{16FE0}\\u{16FE1}\\u{16FE3}\\u{16FF2}\\u{16FF3}\\u{17000}-\\u{18CD5}\\u{18CFF}-\\u{18D1E}\\u{18D80}-\\u{18DF2}\\u{1AFF0}-\\u{1AFF3}\\u{1AFF5}-\\u{1AFFB}\\u{1AFFD}\\u{1AFFE}\\u{1B000}-\\u{1B122}\\u{1B132}\\u{1B150}-\\u{1B152}\\u{1B155}\\u{1B164}-\\u{1B167}\\u{1B170}-\\u{1B2FB}\\u{1BC00}-\\u{1BC6A}\\u{1BC70}-\\u{1BC7C}\\u{1BC80}-\\u{1BC88}\\u{1BC90}-\\u{1BC99}\\u{1D400}-\\u{1D454}\\u{1D456}-\\u{1D49C}\\u{1D49E}\\u{1D49F}\\u{1D4A2}\\u{1D4A5}\\u{1D4A6}\\u{1D4A9}-\\u{1D4AC}\\u{1D4AE}-\\u{1D4B9}\\u{1D4BB}\\u{1D4BD}-\\u{1D4C3}\\u{1D4C5}-\\u{1D505}\\u{1D507}-\\u{1D50A}\\u{1D50D}-\\u{1D514}\\u{1D516}-\\u{1D51C}\\u{1D51E}-\\u{1D539}\\u{1D53B}-\\u{1D53E}\\u{1D540}-\\u{1D544}\\u{1D546}\\u{1D54A}-\\u{1D550}\\u{1D552}-\\u{1D6A5}\\u{1D6A8}-\\u{1D6C0}\\u{1D6C2}-\\u{1D6DA}\\u{1D6DC}-\\u{1D6FA}\\u{1D6FC}-\\u{1D714}\\u{1D716}-\\u{1D734}\\u{1D736}-\\u{1D74E}\\u{1D750}-\\u{1D76E}\\u{1D770}-\\u{1D788}\\u{1D78A}-\\u{1D7A8}\\u{1D7AA}-\\u{1D7C2}\\u{1D7C4}-\\u{1D7CB}\\u{1DF00}-\\u{1DF1E}\\u{1DF25}-\\u{1DF2A}\\u{1E030}-\\u{1E06D}\\u{1E100}-\\u{1E12C}\\u{1E137}-\\u{1E13D}\\u{1E14E}\\u{1E290}-\\u{1E2AD}\\u{1E2C0}-\\u{1E2EB}\\u{1E4D0}-\\u{1E4EB}\\u{1E5D0}-\\u{1E5ED}\\u{1E5F0}\\u{1E6C0}-\\u{1E6DE}\\u{1E6E0}-\\u{1E6E2}\\u{1E6E4}\\u{1E6E5}\\u{1E6E7}-\\u{1E6ED}\\u{1E6F0}-\\u{1E6F4}\\u{1E6FE}\\u{1E6FF}\\u{1E7E0}-\\u{1E7E6}\\u{1E7E8}-\\u{1E7EB}\\u{1E7ED}\\u{1E7EE}\\u{1E7F0}-\\u{1E7FE}\\u{1E800}-\\u{1E8C4}\\u{1E900}-\\u{1E943}\\u{1E94B}\\u{1EE00}-\\u{1EE03}\\u{1EE05}-\\u{1EE1F}\\u{1EE21}\\u{1EE22}\\u{1EE24}\\u{1EE27}\\u{1EE29}-\\u{1EE32}\\u{1EE34}-\\u{1EE37}\\u{1EE39}\\u{1EE3B}\\u{1EE42}\\u{1EE47}\\u{1EE49}\\u{1EE4B}\\u{1EE4D}-\\u{1EE4F}\\u{1EE51}\\u{1EE52}\\u{1EE54}\\u{1EE57}\\u{1EE59}\\u{1EE5B}\\u{1EE5D}\\u{1EE5F}\\u{1EE61}\\u{1EE62}\\u{1EE64}\\u{1EE67}-\\u{1EE6A}\\u{1EE6C}-\\u{1EE72}\\u{1EE74}-\\u{1EE77}\\u{1EE79}-\\u{1EE7C}\\u{1EE7E}\\u{1EE80}-\\u{1EE89}\\u{1EE8B}-\\u{1EE9B}\\u{1EEA1}-\\u{1EEA3}\\u{1EEA5}-\\u{1EEA9}\\u{1EEAB}-\\u{1EEBB}\\u{20000}-\\u{2A6DF}\\u{2A700}-\\u{2B81D}\\u{2B820}-\\u{2CEAD}\\u{2CEB0}-\\u{2EBE0}\\u{2EBF0}-\\u{2EE5D}\\u{2F800}-\\u{2FA1D}\\u{30000}-\\u{3134A}\\u{31350}-\\u{33479}';
+
+export const DISPLAY_NAME_DENIED_SYMBOL_LETTERS_RANGES =
+  '\\u{534D}\\u{5350}';
+
+export const DISPLAY_NAME_NAME_SEPARATORS_RANGES =
+  '\\u{B7}\\u{5BE}\\u{F0B}\\u{30FB}';

package/src/utils/validationUtils.ts

@@ -7,8 +7,15 @@ import {
   DISPLAY_NAME_COMBINING_MARKS_RANGES,
   DISPLAY_NAME_SPACE_SEPARATORS_RANGES,
   DISPLAY_NAME_LETTERS_RANGES,
+  DISPLAY_NAME_NAME_SEPARATORS_RANGES,
 } from './displayNamePolicyRanges.generated';
 
+/**
+ * Maximum stored length of a display name, in code units after cleaning.
+ * Shared by the API write path and client input surfaces.
+ */
+export const MAX_DISPLAY_NAME_LENGTH = 80;
+
 /**
  * Email validation regex
  */
@@ -61,7 +68,16 @@ export function isValidPassword(password: string): boolean {
  *     ideographic space, …) — but NOT control whitespace such as tab, newline,
  *     or carriage return, which would break layout or enable multi-line
  *     spoofing,
- *   - the straight apostrophe (`'`, e.g. "O'Brien").
+ *   - the straight apostrophe (`'`, e.g. "O'Brien"),
+ *   - four name SEPARATORS — `·` U+00B7, `־` U+05BE, `་` U+0F0B, `・` U+30FB —
+ *     but ONLY directly between two letters (`Codeur·euses`, `お坐・エガード`,
+ *     `אברמסקי־קרוננברג`, `འོད་ཟེར`). Every one is General_Category P, so the
+ *     letters intersection strips them by default; they are re-admitted because
+ *     stripping them SPLITS a real name in two. The flanking condition is not a
+ *     refinement but the whole point: the same characters are also used as
+ *     ornament (a trailing `Roberto ·`), and unconditional admission would let
+ *     that through. See {@link DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE}. The
+ *     ASCII hyphen is NOT among them — `Jean-Luc` stays rejected.
  *
  * Everything else is rejected: emoji (🐧), symbols (⁂ ⏚), `:emoji:` shortcodes,
  * digits, hyphens, dots, control whitespace (tab/newline/CR), letters from
@@ -71,7 +87,15 @@ export function isValidPassword(password: string): boolean {
  *
  * The allowlist is expressed with Unicode Script_Extensions (scx=…) so a letter
  * shared by several scripts (e.g. a Han ideograph used in both Chinese and
- * Japanese) still matches. It is the set of scripts Unicode UTS #39 marks
+ * Japanese) still matches, INTERSECTED with General_Category L so only the
+ * scripts' LETTERS are admitted. That intersection is load-bearing, not a
+ * refinement: `scx=X` also carries script X's own digits, punctuation and
+ * symbols, so the un-intersected allowlist admitted 1831 non-letter code points
+ * — script digits (`٠١٢`, `०१२`), 1082 symbols (`֍ ۞ ৳`), 180 punctuation marks
+ * (`։ ، ؛ ।`), and 9 INVISIBLE format/control characters, among them U+061C
+ * ARABIC LETTER MARK (a bidi control that can visually reorder a name) and
+ * U+180E MONGOLIAN VOWEL SEPARATOR. Without it the rejections listed above held
+ * for ASCII input only. It is the set of scripts Unicode UTS #39 marks
  * "Recommended" for general interchange / identifiers, plus Cherokee and
  * Mongolian (both in real modern name use). "Common" script is deliberately
  * EXCLUDED — that is where ASCII digits and general punctuation live, and this
@@ -79,6 +103,29 @@ export function isValidPassword(password: string): boolean {
  * name needs are added back explicitly. Limited-use / excluded / historic
  * scripts (Batak, Runic, Deseret, Adlam, …) are simply absent.
  *
+ * CODE-POINT DENYLIST — a character policy classifies FORM, never MEANING, so a
+ * code point can be a perfectly ordinary letter to Unicode and a hate symbol to
+ * a reader. `卐` U+5350 and `卍` U+534D are the case in point: both are CJK
+ * Unified Ideographs (General_Category Lo, Script_Extensions Han), i.e. the same
+ * kind of thing as `山` in `山田太郎`, so NO script-level or category-level rule
+ * can separate them — every rule that would exclude these two also excludes Han
+ * itself, rejecting every real Chinese, Japanese and Korean name. They are
+ * therefore enumerated and SUBTRACTED from the allowlist at generation time (see
+ * `SYMBOL_LETTER_DENYLIST` in `scripts/generateDisplayNamePolicyRanges.mjs`).
+ * Because the subtraction happens inside the one emitted allowlist, every
+ * consumer enforces it with no extra probe and none can forget it; the generator
+ * additionally REFUSES an entry that an existing lever already excludes (e.g.
+ * the Tibetan svasti signs U+0FD5–U+0FD8, General_Category So, already dropped
+ * by the intersection above), so the list cannot silently accumulate dead
+ * weight.
+ *
+ * REMAINING LIMIT — that closes the "letter that reads as a symbol" gap, not the
+ * other one: slurs and abusive phrasing spelled in ordinary allowlisted letters
+ * pass by construction, because they are built from exactly the characters every
+ * real name needs. No character-level rule — allowlist, intersection, or
+ * denylist — can reject them; that needs a word-level moderation layer, which is
+ * deliberately not attempted here.
+ *
  * HERMES / RANGES: the class bodies below are built from explicit Unicode
  * code-point RANGES ({@link DISPLAY_NAME_ALLOWED_SCRIPTS_RANGES} et al. from
  * `./displayNamePolicyRanges.generated`), NOT from `scx`/General_Category
@@ -100,21 +147,28 @@ export function isValidPassword(password: string): boolean {
 
 /**
  * The curated allowlist of Unicode scripts permitted in a display name, as a
- * character-class body of explicit code-point ranges (the compressed union of
- * the 30 allowlisted Script_Extensions, generated on V8). Interpolated into the
- * negated class below.
+ * character-class body of explicit code-point ranges: the union of the 30
+ * allowlisted Script_Extensions, INTERSECTED with General_Category L (so letters
+ * only), MINUS the symbol-letter denylist — 120823 code points. Interpolated
+ * into the negated class below, which is why both the reject gate here and the
+ * `@oxyhq/api` strip path inherit the denylist without a second pattern.
  */
 export const DISPLAY_NAME_ALLOWED_SCRIPTS = DISPLAY_NAME_ALLOWED_SCRIPTS_RANGES;
 
 /**
  * Source of the disallowed-character pattern: the negation of the full allowed
  * set (allowlisted scripts + combining marks + space separators + the straight
- * apostrophe). Consumers compile this with the `u` flag (and `g` for a global
- * strip). The whitespace class is space separators only (General_Category Zs),
- * NOT `\s` — the latter would admit tab/newline/carriage return, which break
- * layout and enable multi-line spoofing.
+ * apostrophe + the name separators). Consumers compile this with the `u` flag
+ * (and `g` for a global strip). The whitespace class is space separators only
+ * (General_Category Zs), NOT `\s` — the latter would admit tab/newline/carriage
+ * return, which break layout and enable multi-line spoofing.
+ *
+ * The name separators are admitted here only as CHARACTERS. Their position rule
+ * is a separate pattern ({@link DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE}), for
+ * the same reason combining marks are: a negated character class cannot express
+ * "allowed only in this context". Both halves must be applied together.
  */
-export const DISPLAY_NAME_DISALLOWED_SOURCE = `[^${DISPLAY_NAME_ALLOWED_SCRIPTS}${DISPLAY_NAME_COMBINING_MARKS_RANGES}${DISPLAY_NAME_SPACE_SEPARATORS_RANGES}']`;
+export const DISPLAY_NAME_DISALLOWED_SOURCE = `[^${DISPLAY_NAME_ALLOWED_SCRIPTS}${DISPLAY_NAME_COMBINING_MARKS_RANGES}${DISPLAY_NAME_SPACE_SEPARATORS_RANGES}${DISPLAY_NAME_NAME_SEPARATORS_RANGES}']`;
 
 /**
  * Source of the orphaned combining-mark pattern: a run of combining marks NOT
@@ -129,29 +183,68 @@ export const DISPLAY_NAME_DISALLOWED_SOURCE = `[^${DISPLAY_NAME_ALLOWED_SCRIPTS}
  */
 export const DISPLAY_NAME_ORPHANED_MARK_SOURCE = `(?<![${DISPLAY_NAME_LETTERS_RANGES}${DISPLAY_NAME_COMBINING_MARKS_RANGES}])[${DISPLAY_NAME_COMBINING_MARKS_RANGES}]+`;
 
+/**
+ * Source of the unflanked name-separator pattern — the positional half of the
+ * name-separator rule, and the exact counterpart of
+ * {@link DISPLAY_NAME_ORPHANED_MARK_SOURCE}: a combining mark is allowed only
+ * when it rides a base letter, and a name separator is allowed only when it JOINS
+ * two letters. Matches a separator that is missing a letter on either side, so
+ * consumers can reject it (non-global probe) or strip it (`g` flag).
+ *
+ * Two alternatives, one per side, because JS has no "not surrounded by" atom:
+ *   - `(?<![letters][marks])[sep]` — nothing letter-like immediately BEFORE.
+ *   - `[sep](?![letters])`        — no letter immediately AFTER.
+ *
+ * The left-hand class deliberately includes combining marks as well as letters:
+ * a base letter can carry an accent (`Renée·euses`, or a decomposed `e`+◌́), and
+ * that mark sits between the letter and the separator. Treating a mark as
+ * letter-like is what stops an accent from defeating the flanking test — the
+ * same reason the orphaned-mark lookbehind uses the identical pair of classes.
+ * The right-hand class is letters only: a mark AFTER a separator has no base and
+ * is removed by the orphaned-mark rule, so the separator is genuinely unflanked.
+ *
+ * Both lookarounds are single-character (no variable-length lookbehind), and the
+ * classes are explicit code-point ranges, so the compiled regex is within what
+ * Hermes accepts.
+ *
+ * `Codeur·euses` and `お坐・エガード` match NOTHING here and survive. A leading
+ * `·Roberto`, a trailing `Roberto ·`, and a doubled `a··b` each match and are
+ * stripped — which is what keeps a decorative middle dot from riding in on the
+ * back of the orthographic one.
+ */
+export const DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE = `(?<![${DISPLAY_NAME_LETTERS_RANGES}${DISPLAY_NAME_COMBINING_MARKS_RANGES}])[${DISPLAY_NAME_NAME_SEPARATORS_RANGES}]|[${DISPLAY_NAME_NAME_SEPARATORS_RANGES}](?![${DISPLAY_NAME_LETTERS_RANGES}])`;
+
 /** Non-global probe for the presence of a disallowed character. */
 const DISALLOWED_PROBE = new RegExp(DISPLAY_NAME_DISALLOWED_SOURCE, 'u');
 
 /** Non-global probe for the presence of an orphaned combining mark. */
 const ORPHANED_MARK_PROBE = new RegExp(DISPLAY_NAME_ORPHANED_MARK_SOURCE, 'u');
 
+/** Non-global probe for the presence of an unflanked name separator. */
+const UNFLANKED_SEPARATOR_PROBE = new RegExp(DISPLAY_NAME_UNFLANKED_SEPARATOR_SOURCE, 'u');
+
 /**
  * Whether `raw` already satisfies the display-name policy, i.e. it contains no
- * disallowed characters AND no orphaned combining marks. Used to REJECT native
- * (signup / profile edit) names with a 400 rather than silently stripping them,
- * and to validate inline in the client editor.
+ * disallowed characters, no orphaned combining marks, and no unflanked name
+ * separator. Used to REJECT native (signup / profile edit) names with a 400
+ * rather than silently stripping them, and to validate inline in the client
+ * editor.
  *
- * The orphaned-mark probe runs on the NFC-normalized form so a legitimate
- * decomposed accent (`e`+◌́) — which normalization recomposes into `é` — is NOT
- * rejected, while a lone, base-less mark (e.g. `"༘"`) IS.
+ * The two positional probes run on the NFC-normalized form, the character probe
+ * on the raw input. Normalization matters for both: a legitimate decomposed
+ * accent (`e`+◌́) recomposes into `é`, so it is NOT rejected as an orphaned mark,
+ * and a separator after that accent sees a base letter rather than a mark.
  *
- * The function only checks the character set; an empty or whitespace-only string
- * is considered valid (`true`). Call sites that require a non-empty name enforce
- * that separately.
+ * The function only checks the character set and separator placement; an empty
+ * or whitespace-only string is considered valid (`true`). Call sites that require
+ * a non-empty name enforce that separately.
  */
 export function isValidDisplayName(raw: string): boolean {
+  const normalized = raw.normalize('NFC');
   return (
-    !DISALLOWED_PROBE.test(raw) && !ORPHANED_MARK_PROBE.test(raw.normalize('NFC'))
+    !DISALLOWED_PROBE.test(raw) &&
+    !ORPHANED_MARK_PROBE.test(normalized) &&
+    !UNFLANKED_SEPARATOR_PROBE.test(normalized)
   );
 }