gh-125682: Reject non-ASCII digits in the Python implementation of JSON decoder (GH-125687)

Author: nineteendo

Lib/json/scanner.py

@@ -9,7 +9,7 @@
 __all__ = ['make_scanner']
 
 NUMBER_RE = re.compile(
-    r'(-?(?:0|[1-9]\d*))(\.\d+)?([eE][-+]?\d+)?',
+    r'(-?(?:0|[1-9][0-9]*))(\.[0-9]+)?([eE][-+]?[0-9]+)?',
     (re.VERBOSE | re.MULTILINE | re.DOTALL))
 
 def py_make_scanner(context):

Lib/test/test_json/test_decode.py

@@ -16,6 +16,12 @@ def test_float(self):
         self.assertIsInstance(rval, float)
         self.assertEqual(rval, 1.0)
 
+    def test_nonascii_digits_rejected(self):
+        # JSON specifies only ascii digits, see gh-125687
+        for num in ["1\uff10", "0.\uff10", "0e\uff10"]:
+            with self.assertRaises(self.JSONDecodeError):
+                self.loads(num)
+
     def test_bytes(self):
         self.assertEqual(self.loads(b"1"), 1)
 

Misc/NEWS.d/next/Library/2024-10-18-09-51-29.gh-issue-125682.vsj4cU.rst

@@ -0,0 +1,2 @@
+Reject non-ASCII digits in the Python implementation of :func:`json.loads`
+conforming to the JSON specification.