gh-83461: Don't allow datetime parsing to accept non-ASCII digits

Doc/library/datetime.rst

@@ -2611,6 +2611,9 @@ differences between platforms in handling of unsupported format specifiers.
 .. versionadded:: 3.12
    ``%:z`` was added.
 
+.. versionchanged:: next
+   Non-ASCII digits are now rejected by ``strptime`` for numerical directives.
+
 Technical Detail
 ^^^^^^^^^^^^^^^^
 
@@ -2650,7 +2653,8 @@ Notes:
    Because the format depends on the current locale, care should be taken when
    making assumptions about the output value. Field orderings will vary (for
    example, "month/day/year" versus "day/month/year"), and the output may
-   contain non-ASCII characters.
+   contain non-ASCII characters. :meth:`~.datetime.strptime` rejects non-ASCII
+   digits for non-locale-specific numeric format codes (e.g. ``%Y``, ``%H``, etc).
 
 (2)
    The :meth:`~.datetime.strptime` method can parse years in the full [1, 9999] range, but

Doc/library/time.rst

@@ -563,6 +563,9 @@ Functions
       When used with the :func:`strptime` function, ``%U`` and ``%W`` are only used in
       calculations when the day of the week and the year are specified.
 
+   (5)
+      The :func:`strptime` function does not accept non-ASCII digits for numeric values.
+
    Here is an example, a format for dates compatible with that specified  in the
    :rfc:`2822` Internet email standard.  [1]_ ::
 
@@ -586,9 +589,8 @@ Functions
 
 .. function:: strptime(string[, format])
 
-   Parse a string representing a time according to a format.  The return value
-   is a :class:`struct_time` as returned by :func:`gmtime` or
-   :func:`localtime`.
+   Parse a string representing a time according to a format string.  The return
+   value is a :class:`struct_time` as returned by :func:`gmtime` or :func:`localtime`.
 
    The *format* parameter uses the same directives as those used by
    :func:`strftime`; it defaults to ``"%a %b %d %H:%M:%S %Y"`` which matches the
@@ -616,6 +618,9 @@ Functions
    and thus does not necessarily support all directives available that are not
    documented as supported.
 
+.. versionchanged:: next
+   Non-ASCII digits are now rejected by ``strptime``.
+
 
 .. class:: struct_time
 

Doc/whatsnew/3.14.rst

@@ -504,6 +504,13 @@ datetime
 * Add :meth:`datetime.time.strptime` and :meth:`datetime.date.strptime`.
   (Contributed by Wannes Boeykens in :gh:`41431`.)
 
+* When using digits in :meth:`datetime.date.strptime`,
+  :meth:`datetime.datetime.strptime`, or :meth:`datetime.time.strptime`, a
+  :exc:`ValueError` is raised if non-ASCII digits are specified for
+  non-locale-specific numeric format codes.
+  (Contributed by Stan Ulbrych in :gh:`131008`.)
+
+
 decimal
 -------
 
@@ -878,6 +885,15 @@ sys.monitoring
 * Two new events are added: :monitoring-event:`BRANCH_LEFT` and
   :monitoring-event:`BRANCH_RIGHT`. The ``BRANCH`` event is deprecated.
 
+
+time
+----
+
+* When using digits in :func:`time.strptime`, a :exc:`ValueError` is raised if
+  non-ASCII digits are specified for non-locale-specific numeric format codes.
+  (Contributed by Stan Ulbrych in :gh:`131008`.)
+
+
 threading
 ---------
 

Lib/_strptime.py

@@ -286,23 +286,23 @@ def __init__(self, locale_time=None):
         base = super()
         mapping = {
             # The " [1-9]" part of the regex is to make %c from ANSI C work
-            'd': r"(?P<d>3[0-1]|[1-2]\d|0[1-9]|[1-9]| [1-9])",
+            'd': r"(?P<d>3[0-1]|[1-2][0-9]|0[1-9]|[1-9]| [1-9])",
             'f': r"(?P<f>[0-9]{1,6})",
-            'H': r"(?P<H>2[0-3]|[0-1]\d|\d)",
+            'H': r"(?P<H>2[0-3]|[0-1][0-9]|[0-9])",
             'I': r"(?P<I>1[0-2]|0[1-9]|[1-9]| [1-9])",
-            'G': r"(?P<G>\d\d\d\d)",
-            'j': r"(?P<j>36[0-6]|3[0-5]\d|[1-2]\d\d|0[1-9]\d|00[1-9]|[1-9]\d|0[1-9]|[1-9])",
+            'G': r"(?P<G>[0-9][0-9][0-9][0-9])",
+            'j': r"(?P<j>36[0-6]|3[0-5][0-9]|[1-2][0-9][0-9]|0[1-9][0-9]|00[1-9]|[1-9][0-9]|0[1-9]|[1-9])",
             'm': r"(?P<m>1[0-2]|0[1-9]|[1-9])",
-            'M': r"(?P<M>[0-5]\d|\d)",
-            'S': r"(?P<S>6[0-1]|[0-5]\d|\d)",
-            'U': r"(?P<U>5[0-3]|[0-4]\d|\d)",
+            'M': r"(?P<M>[0-5][0-9]|[0-9])",
+            'S': r"(?P<S>6[0-1]|[0-5][0-9]|[0-9])",
+            'U': r"(?P<U>5[0-3]|[0-4][0-9]|[0-9])",
             'w': r"(?P<w>[0-6])",
             'u': r"(?P<u>[1-7])",
-            'V': r"(?P<V>5[0-3]|0[1-9]|[1-4]\d|\d)",
+            'V': r"(?P<V>5[0-3]|0[1-9]|[1-4][0-9]|[0-9])",
             # W is set below by using 'U'
-            'y': r"(?P<y>\d\d)",
-            'Y': r"(?P<Y>\d\d\d\d)",
-            'z': r"(?P<z>[+-]\d\d:?[0-5]\d(:?[0-5]\d(\.\d{1,6})?)?|(?-i:Z))",
+            'y': r"(?P<y>[0-9][0-9])",
+            'Y': r"(?P<Y>[0-9]{4})",
+            'z': r"(?P<z>[+-][0-9][0-9]:?[0-5][0-9](:?[0-5][0-9](\.[0-9]{1,6})?)?|(?-i:Z))",
             'A': self.__seqToRE(self.locale_time.f_weekday, 'A'),
             'a': self.__seqToRE(self.locale_time.a_weekday, 'a'),
             'B': self.__seqToRE(self.locale_time.f_month[1:], 'B'),
@@ -313,8 +313,8 @@ def __init__(self, locale_time=None):
                                 'Z'),
             '%': '%'}
         for d in 'dmyHIMS':
-            mapping['O' + d] = r'(?P<%s>\d\d|\d| \d)' % d
-        mapping['Ow'] = r'(?P<w>\d)'
+            mapping['O' + d] = r'(?P<%s>[0-9][0-9]|[0-9]| [0-9])' % d
+        mapping['Ow'] = r'(?P<w>[0-9])'
         mapping['W'] = mapping['U'].replace('U', 'W')
         base.__init__(mapping)
         base.__setitem__('X', self.pattern(self.locale_time.LC_time))

Lib/test/datetimetester.py

@@ -2916,6 +2916,14 @@ def test_strptime(self):
         with self.assertRaises(ValueError): strptime("-000", "%z")
         with self.assertRaises(ValueError): strptime("z", "%z")
 
+        # test only ascii is allowed
+        with self.assertRaises(ValueError): strptime('٢025-0٢-٢9', '%Y-%m-%d')
+        with self.assertRaises(ValueError): strptime('1٢:02:٢7', '%H:%M:%S')
+        with self.assertRaises(ValueError): strptime('٢5', '%y')
+        with self.assertRaises(ValueError): strptime('٢555', '%G')
+        with self.assertRaises(ValueError): strptime('٢/0٢ 0٢a٢', '%j/%y %I%p:%M:%S')
+        with self.assertRaises(ValueError): strptime('0٢/٢/200٢', '%U/%V')
+
     def test_strptime_single_digit(self):
         # bpo-34903: Check that single digit dates and times are allowed.
 
@@ -4036,7 +4044,7 @@ def test_strptime_tz(self):
         self.assertEqual(strptime("UTC", "%Z").tzinfo, None)
 
     def test_strptime_errors(self):
-        for tzstr in ("-2400", "-000", "z"):
+        for tzstr in ("-2400", "-000", "z", "٢"):
             with self.assertRaises(ValueError):
                 self.theclass.strptime(tzstr, "%z")
 

Lib/test/test_time.py

@@ -352,6 +352,10 @@ def test_strptime_leap_year(self):
                                    r'.*day of month without a year.*'):
             time.strptime('02-07 18:28', '%m-%d %H:%M')
 
+    def test_strptime_non_ascii(self):
+        with self.assertRaises(ValueError):
+            time.strptime('٢025', '%Y')
+
     def test_asctime(self):
         time.asctime(time.gmtime(self.t))
 

Misc/NEWS.d/next/Library/2025-03-09-11-01-00.gh-issue-83461.auwd13.rst

@@ -0,0 +1,3 @@
+Digits passed to :meth:`datetime.date.strptime`, :meth:`datetime.datetime.strptime`,
+:meth:`datetime.time.strptime` and :meth:`time.strptime` must now be ASCII in
+non-locale-specific numeric format codes.