Mimic browser behavior in case of invalid numeric references, fixes #48

mdevils · Jan 25, 2021 · d0ed012 · d0ed012
1 parent 8042062
commit d0ed012
Show file tree

Hide file tree

Showing 4 changed files with 9 additions and 320 deletions.
diff --git a/package.json b/package.json
@@ -26,7 +26,7 @@
     "@typescript-eslint/parser": "^4.6.1",
     "benchmark": "^2.1.4",
     "chai": "^4.2.0",
-    "entities": "^2.1.0",
+    "entities": "^2.2.0",
     "eslint": "^7.12.1",
     "eslint-config-prettier": "^6.15.0",
     "eslint-plugin-import": "^2.22.1",

diff --git a/src/index.ts b/src/index.ts
@@ -90,6 +90,7 @@ const decodeRegExps: Record<Level, Record<DecodeScope, RegExp>> = {
 };
 
 const fromCharCode = String.fromCharCode;
+const outOfBoundsChar = fromCharCode(65533);
 
 export function decode(
     text: string | undefined | null,
@@ -113,7 +114,7 @@ export function decode(
             secondChar == 'x' || secondChar == 'X' ? parseInt(entity.substr(3), 16) : parseInt(entity.substr(2));
 
         return code >= 0x10ffff
-            ? entity
+            ? outOfBoundsChar
             : code > 65535
             ? fromCodePoint(code)
             : fromCharCode(numericUnicodeMap[code] || code);

diff --git a/test/index.test.ts b/test/index.test.ts
@@ -66,7 +66,7 @@ describe('decode()', () => {
         expect(decode('')).to.equal('');
     });
     it('should handle invalid numeric entities', () => {
-        expect(decode('&#2013266066;')).to.equal('&#2013266066;');
+        expect(decode('&#2013266066;')).to.equal(String.fromCharCode(65533));
     });
     it('should decode numeric entities without semicolon', () => {
         expect(decode('&#34C&#34')).to.equal('"C"');