From 8e99c65cb17b203c0bf7e0a3fb9370fe15a36c26 Mon Sep 17 00:00:00 2001
From: dbarabashh <dbarabashdev@gmail.com>
Date: Tue, 29 Apr 2025 17:16:31 +0200
Subject: [PATCH] decode unicode escapes in identifier tokens

---
 packages/typescript-estree/src/node-utils.ts  | 14 +++++++-
 .../tests/lib/node-utils.test.ts              | 32 ++++++++++++++++++-
 2 files changed, 44 insertions(+), 2 deletions(-)

diff --git a/packages/typescript-estree/src/node-utils.ts b/packages/typescript-estree/src/node-utils.ts
index f2f60c777bd1..7c85249cecd0 100644
--- a/packages/typescript-estree/src/node-utils.ts
+++ b/packages/typescript-estree/src/node-utils.ts
@@ -577,6 +577,15 @@ export function getTokenType(
   return AST_TOKEN_TYPES.Identifier;
 }
 
+export function unescapeUnicodeIdentifier(text: string): string {
+  return text.replaceAll(
+    /\\u\{([0-9a-fA-F]+)\}|\\u([0-9a-fA-F]{4})/g,
+    (_match: string, curlyHex: string, shortHex: string) => {
+      const codePoint = parseInt(curlyHex || shortHex, 16);
+      return String.fromCodePoint(codePoint);
+    },
+  );
+}
 /**
  * Extends and formats a given ts.Token, for a given AST
  */
@@ -589,8 +598,11 @@ export function convertToken(
       ? token.getFullStart()
       : token.getStart(ast);
   const end = token.getEnd();
-  const value = ast.text.slice(start, end);
   const tokenType = getTokenType(token);
+  const value =
+    token.kind === SyntaxKind.Identifier
+      ? unescapeUnicodeIdentifier(ast.text.slice(start, end))
+      : ast.text.slice(start, end);
   const range: TSESTree.Range = [start, end];
   const loc = getLocFor(range, ast);
 
diff --git a/packages/typescript-estree/tests/lib/node-utils.test.ts b/packages/typescript-estree/tests/lib/node-utils.test.ts
index f1a1145685b7..c75b85aaa2b1 100644
--- a/packages/typescript-estree/tests/lib/node-utils.test.ts
+++ b/packages/typescript-estree/tests/lib/node-utils.test.ts
@@ -1,4 +1,7 @@
-import { unescapeStringLiteralText } from '../../src/node-utils';
+import {
+  unescapeStringLiteralText,
+  unescapeUnicodeIdentifier,
+} from '../../src/node-utils';
 
 describe(unescapeStringLiteralText, () => {
   it('should not modify content', () => {
@@ -42,3 +45,30 @@ describe(unescapeStringLiteralText, () => {
     ).toBe(`a\n<>"'&©∆&rx;😂\u0000\u0001`);
   });
 });
+
+describe('unescapeUnicodeIdentifier', () => {
+  it('should decode simple \\uXXXX escape', () => {
+    expect(unescapeUnicodeIdentifier('\\u0061')).toBe('a');
+    expect(unescapeUnicodeIdentifier('\\u0042')).toBe('B');
+  });
+
+  it('should decode \\u{X} escape', () => {
+    expect(unescapeUnicodeIdentifier('\\u{61}')).toBe('a');
+    expect(unescapeUnicodeIdentifier('\\u{1F602}')).toBe('😂');
+  });
+
+  it('should decode multiple escapes in one string', () => {
+    expect(unescapeUnicodeIdentifier('\\u0061\\u0062')).toBe('ab');
+    expect(unescapeUnicodeIdentifier('\\u{61}\\u{62}')).toBe('ab');
+  });
+
+  it('should leave non-escape text unchanged', () => {
+    expect(unescapeUnicodeIdentifier('foo')).toBe('foo');
+    expect(unescapeUnicodeIdentifier('a\\u0062c')).toBe('abc');
+  });
+
+  it('should not decode invalid escapes', () => {
+    expect(unescapeUnicodeIdentifier('\\u00ZZ')).toBe('\\u00ZZ');
+    expect(unescapeUnicodeIdentifier('\\u{ZZ}')).toBe('\\u{ZZ}');
+  });
+});