From b7715083a3c599f5caa17b87f0563ed9173f28f0 Mon Sep 17 00:00:00 2001
From: yogeshwaran-c <ycyogesh183@gmail.com>
Date: Tue, 24 Feb 2026 02:31:08 +0530
Subject: [PATCH] fix(compiler): parse named HTML entities containing digits

The lexer's isNamedEntityEnd function stopped scanning entity names
when encountering a digit character, causing 24 valid HTML named
entities with digits in their names (e.g. &sup1;, &frac12;, &blk34;)
to be treated as plain text instead of decoded to their corresponding
Unicode characters.

Fixes #51323
---
 packages/compiler/src/ml_parser/lexer.ts      |  7 ++++++-
 .../test/ml_parser/html_parser_spec.ts        | 11 ++++++++++
 .../compiler/test/ml_parser/lexer_spec.ts     | 21 +++++++++++++++++++
 3 files changed, 38 insertions(+), 1 deletion(-)

diff --git a/packages/compiler/src/ml_parser/lexer.ts b/packages/compiler/src/ml_parser/lexer.ts
index 3c8b04ab61b4..013b2eb1505f 100644
--- a/packages/compiler/src/ml_parser/lexer.ts
+++ b/packages/compiler/src/ml_parser/lexer.ts
@@ -1431,7 +1431,12 @@ function isDigitEntityEnd(code: number): boolean {
 }
 
 function isNamedEntityEnd(code: number): boolean {
-  return code === chars.$SEMICOLON || code === chars.$EOF || !chars.isAsciiLetter(code);
+  // Named entities may contain digits (e.g. &sup1;, &frac12;, &blk34;).
+  return (
+    code === chars.$SEMICOLON ||
+    code === chars.$EOF ||
+    !(chars.isAsciiLetter(code) || chars.isDigit(code))
+  );
 }
 
 function isExpansionCaseStart(peek: number): boolean {
diff --git a/packages/compiler/test/ml_parser/html_parser_spec.ts b/packages/compiler/test/ml_parser/html_parser_spec.ts
index 47b956aea40e..2ea1422e793d 100644
--- a/packages/compiler/test/ml_parser/html_parser_spec.ts
+++ b/packages/compiler/test/ml_parser/html_parser_spec.ts
@@ -68,6 +68,17 @@ describe('HtmlParser', () => {
         ]);
       });
 
+      it('should parse named HTML entities containing digits', () => {
+        expect(humanizeDom(parser.parse('<div>&sup1;</div>', 'TestComp'))).toEqual([
+          [html.Element, 'div', 0],
+          [html.Text, '\u00B9', 1, [''], ['\u00B9', '&sup1;'], ['']],
+        ]);
+        expect(humanizeDom(parser.parse('<div>&frac12;</div>', 'TestComp'))).toEqual([
+          [html.Element, 'div', 0],
+          [html.Text, '\u00BD', 1, [''], ['\u00BD', '&frac12;'], ['']],
+        ]);
+      });
+
       it('should normalize line endings within CDATA', () => {
         const parsed = parser.parse('<![CDATA[ line 1 \r\n line 2 ]]>', 'TestComp');
         expect(humanizeDom(parsed)).toEqual([
diff --git a/packages/compiler/test/ml_parser/lexer_spec.ts b/packages/compiler/test/ml_parser/lexer_spec.ts
index 9d7bbd1c5c74..e3ebac0e8270 100644
--- a/packages/compiler/test/ml_parser/lexer_spec.ts
+++ b/packages/compiler/test/ml_parser/lexer_spec.ts
@@ -2094,6 +2094,27 @@ describe('HtmlLexer', () => {
       ]);
     });
 
+    it('should parse named entities containing digits', () => {
+      expect(tokenizeAndHumanizeParts('&sup1;')).toEqual([
+        [TokenType.TEXT, ''],
+        [TokenType.ENCODED_ENTITY, '\u00B9', '&sup1;'],
+        [TokenType.TEXT, ''],
+        [TokenType.EOF],
+      ]);
+      expect(tokenizeAndHumanizeParts('&frac12;')).toEqual([
+        [TokenType.TEXT, ''],
+        [TokenType.ENCODED_ENTITY, '\u00BD', '&frac12;'],
+        [TokenType.TEXT, ''],
+        [TokenType.EOF],
+      ]);
+      expect(tokenizeAndHumanizeParts('&blk34;')).toEqual([
+        [TokenType.TEXT, ''],
+        [TokenType.ENCODED_ENTITY, '\u2593', '&blk34;'],
+        [TokenType.TEXT, ''],
+        [TokenType.EOF],
+      ]);
+    });
+
     it('should parse hexadecimal entities', () => {
       expect(tokenizeAndHumanizeParts('&#x41;&#X41;')).toEqual([
         [TokenType.TEXT, ''],