Apr 29, 2025 · May 2, 2025 · May 2, 2025 · May 2, 2025 · May 2, 2025
diff --git a/packages/typescript-estree/src/node-utils.ts b/packages/typescript-estree/src/node-utils.ts
  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
 */
      ? 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
 import { unescapeStringLiteralText } from '../../src/node-utils';
 import {
  unescapeStringLiteralText,
  unescapeUnicodeIdentifier,
 } from '../../src/node-utils';

 describe(unescapeStringLiteralText, () => {
  it('should not modify content', () => {
    ).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}');
  });
 });
Original file line number	Diff line number	Diff line change
Expand Up		@@ -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);
		},
		);
		}
Copy link Member JoshuaKGoldbergMay 2, 2025 Choose a reason for hiding this comment The reason will be displayed to describe this comment to others.Learn more. [Question] I'm not very familiar with these`\u` shenanigans and don't particularly feel a drive to be. Is there a standard package that does this? If not, maybe it would make sense to make one?
		/**
		* Extends and formats a given ts.Token, for a given AST
		*/
Expand All		@@ -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);

Expand Down
Original file line number	Diff line number	Diff line change
		@@ -1,4 +1,7 @@
		import { unescapeStringLiteralText } from '../../src/node-utils';
		import {
		unescapeStringLiteralText,
		unescapeUnicodeIdentifier,
		} from '../../src/node-utils';

		describe(unescapeStringLiteralText, () => {
		it('should not modify content', () => {
Expand DownExpand Up		@@ -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}');
		});
		});
Copy link Member JoshuaKGoldbergMay 2, 2025 Choose a reason for hiding this comment The reason will be displayed to describe this comment to others.Learn more. [Testing] Nit: I'm a fan of having one thing under test per`it()` case. I think it would make sense to split each of these two-paired ones into two`it()`s. But I wouldn't block on it, this file already doesn't align to that (and even more so in existing code!).