bpo-30681: Support invalid date format or value in email Date header (GH-22090)

I am re-submitting an older PR which was abandoned but is still relevant, #10783 by @timb07.

The issue being solved () is still relevant. The original PR #10783 was closed as
the final request changes were not applied and since abandoned.

In this new PR I have re-used the original patch plus applied both comments from the review, by @maxking and @pganssle.


For reference, here is the original PR description:
In email.utils.parsedate_to_datetime(), a failure to parse the date, or invalid date components (such as hour outside 0..23) raises an exception. Document this behaviour, and add tests to test_email/test_utils.py to confirm this behaviour.

In email.headerregistry.DateHeader.parse(), check when parsedate_to_datetime() raises an exception and add a new defect InvalidDateDefect; preserve the invalid value as the string value of the header, but set the datetime attribute to None.

Add tests to test_email/test_headerregistry.py to confirm this behaviour; also added test to test_email/test_inversion.py to confirm emails with such defective date headers round trip successfully.

This pull request incorporates feedback gratefully received from @bitdancer, @brettcannon, @Mariatta and @warsaw, and replaces the earlier PR #2254.

Automerge-Triggered-By: GH:warsaw

Doc/library/email.errors.rst

@@ -112,3 +112,6 @@ All defect classes are subclassed from :class:`email.errors.MessageDefect`.
 * :class:`InvalidBase64LengthDefect` -- When decoding a block of base64 encoded
   bytes, the number of non-padding base64 characters was invalid (1 more than
   a multiple of 4).  The encoded block was kept as-is.
+
+* :class:`InvalidDateDefect` -- When decoding an invalid or unparsable date field.
+  The original value is kept as-is.

Doc/library/email.utils.rst

@@ -124,8 +124,10 @@ of the new API.
 .. function:: parsedate_to_datetime(date)
 
    The inverse of :func:`format_datetime`.  Performs the same function as
-   :func:`parsedate`, but on success returns a :mod:`~datetime.datetime`.  If
-   the input date has a timezone of ``-0000``, the ``datetime`` will be a naive
+   :func:`parsedate`, but on success returns a :mod:`~datetime.datetime`;
+   otherwise ``ValueError`` is raised if *date* contains an invalid value such
+   as an hour greater than 23 or a timezone offset not between -24 and 24 hours.
+   If the input date has a timezone of ``-0000``, the ``datetime`` will be a naive
    ``datetime``, and if the date is conforming to the RFCs it will represent a
    time in UTC but with no indication of the actual source timezone of the
    message the date comes from.  If the input date has any other valid timezone

Lib/email/_parseaddr.py

@@ -65,7 +65,7 @@ def _parsedate_tz(data):
 
     """
     if not data:
-        return
+        return None
     data = data.split()
     # The FWS after the comma after the day-of-week is optional, so search and
     # adjust for this.

Lib/email/errors.py

@@ -108,3 +108,6 @@ class NonASCIILocalPartDefect(HeaderDefect):
     """local_part contains non-ASCII characters"""
     # This defect only occurs during unicode parsing, not when
     # parsing messages decoded from binary.
+
+class InvalidDateDefect(HeaderDefect):
+    """Header has unparseable or invalid date"""

Lib/email/headerregistry.py

@@ -302,7 +302,14 @@ class DateHeader:
             kwds['parse_tree'] = parser.TokenList()
             return
         if isinstance(value, str):
+            kwds['decoded'] = value
+            try:
                 value = utils.parsedate_to_datetime(value)
+            except ValueError:
+                kwds['defects'].append(errors.InvalidDateDefect('Invalid date value or format'))
+                kwds['datetime'] = None
+                kwds['parse_tree'] = parser.TokenList()
+                return
         kwds['datetime'] = value
         kwds['decoded'] = utils.format_datetime(kwds['datetime'])
         kwds['parse_tree'] = cls.value_parser(kwds['decoded'])

Lib/email/utils.py

@@ -195,7 +195,10 @@ def make_msgid(idstring=None, domain=None):
 
 
 def parsedate_to_datetime(data):
-    *dtuple, tz = _parsedate_tz(data)
+    parsed_date_tz = _parsedate_tz(data)
+    if parsed_date_tz is None:
+        raise ValueError('Invalid date value or format "%s"' % str(data))
+    *dtuple, tz = parsed_date_tz
     if tz is None:
         return datetime.datetime(*dtuple[:6])
     return datetime.datetime(*dtuple[:6],

Lib/test/test_email/test_headerregistry.py

@@ -204,6 +204,22 @@ class TestDateHeader(TestHeaderBase):
         self.assertEqual(len(h.defects), 1)
         self.assertIsInstance(h.defects[0], errors.HeaderMissingRequiredValue)
 
+    def test_invalid_date_format(self):
+        s = 'Not a date header'
+        h = self.make_header('date', s)
+        self.assertEqual(h, s)
+        self.assertIsNone(h.datetime)
+        self.assertEqual(len(h.defects), 1)
+        self.assertIsInstance(h.defects[0], errors.InvalidDateDefect)
+
+    def test_invalid_date_value(self):
+        s = 'Tue, 06 Jun 2017 27:39:33 +0600'
+        h = self.make_header('date', s)
+        self.assertEqual(h, s)
+        self.assertIsNone(h.datetime)
+        self.assertEqual(len(h.defects), 1)
+        self.assertIsInstance(h.defects[0], errors.InvalidDateDefect)
+
     def test_datetime_read_only(self):
         h = self.make_header('date', self.datestring)
         with self.assertRaises(AttributeError):

Lib/test/test_email/test_inversion.py

@@ -46,6 +46,14 @@ class TestInversion(TestEmailBase):
             foo
             """),),
 
+        'header_with_invalid_date': (dedent(b"""\
+            Date: Tue, 06 Jun 2017 27:39:33 +0600
+            From: abc@xyz.com
+            Subject: timezones
+
+            How do they work even?
+            """),),
+
             }
 
     payload_params = {

Lib/test/test_email/test_utils.py

@@ -48,6 +48,16 @@ class DateTimeTests(unittest.TestCase):
             utils.parsedate_to_datetime(self.datestring + ' -0000'),
             self.naive_dt)
 
+    def test_parsedate_to_datetime_with_invalid_raises_valueerror(self):
+        invalid_dates = ['',
+                         '0',
+                         'A Complete Waste of Time'
+                         'Tue, 06 Jun 2017 27:39:33 +0600',
+                         'Tue, 06 Jun 2017 07:39:33 +2600',
+                         'Tue, 06 Jun 2017 27:39:33']
+        for dtstr in invalid_dates:
+            with self.subTest(dtstr=dtstr):
+                self.assertRaises(ValueError, utils.parsedate_to_datetime, dtstr)
 
 class LocaltimeTests(unittest.TestCase):
 

Misc/NEWS.d/next/Library/2020-09-04-17-33-04.bpo-30681.LR4fnY.rst

@@ -0,0 +1,2 @@
+Handle exceptions caused by unparseable date headers when using email
+"default" policy.  Patch by Tim Bell, Georges Toth