[3.7] bpo-22005: Fixed unpickling instances of datetime classes pickled by Python 2. (GH-11017) (GH-11022)

encoding='latin1' should be used for successful decoding.
(cherry picked from commit 8452ca1)

Author: serhiy-storchaka

Doc/library/pickle.rst

@@ -235,6 +235,9 @@ process more convenient:
    *errors* tell pickle how to decode 8-bit string instances pickled by Python
    2; these default to 'ASCII' and 'strict', respectively.  The *encoding* can
    be 'bytes' to read these 8-bit string instances as bytes objects.
+   Using ``encoding='latin1'`` is required for unpickling NumPy arrays and
+   instances of :class:`~datetime.datetime`, :class:`~datetime.date` and
+   :class:`~datetime.time` pickled by Python 2.
 
 .. function:: loads(bytes_object, \*, fix_imports=True, encoding="ASCII", errors="strict")
 
@@ -252,6 +255,9 @@ process more convenient:
    *errors* tell pickle how to decode 8-bit string instances pickled by Python
    2; these default to 'ASCII' and 'strict', respectively.  The *encoding* can
    be 'bytes' to read these 8-bit string instances as bytes objects.
+   Using ``encoding='latin1'`` is required for unpickling NumPy arrays and
+   instances of :class:`~datetime.datetime`, :class:`~datetime.date` and
+   :class:`~datetime.time` pickled by Python 2.
 
 
 The :mod:`pickle` module defines three exceptions:

Lib/datetime.py

@@ -808,9 +808,19 @@ def __new__(cls, year, month=None, day=None):
 
         year, month, day (required, base 1)
         """
-        if month is None and isinstance(year, bytes) and len(year) == 4 and \
-                1 <= year[2] <= 12:
+        if (month is None and
+            isinstance(year, (bytes, str)) and len(year) == 4 and
+            1 <= ord(year[2:3]) <= 12):
             # Pickle support
+            if isinstance(year, str):
+                try:
+                    year = year.encode('latin1')
+                except UnicodeEncodeError:
+                    # More informative error message.
+                    raise ValueError(
+                        "Failed to encode latin1 string when unpickling "
+                        "a date object. "
+                        "pickle.load(data, encoding='latin1') is assumed.")
             self = object.__new__(cls)
             self.__setstate(year)
             self._hashcode = -1
@@ -1184,8 +1194,18 @@ def __new__(cls, hour=0, minute=0, second=0, microsecond=0, tzinfo=None, *, fold
         tzinfo (default to None)
         fold (keyword only, default to zero)
         """
-        if isinstance(hour, bytes) and len(hour) == 6 and hour[0]&0x7F < 24:
+        if (isinstance(hour, (bytes, str)) and len(hour) == 6 and
+            ord(hour[0:1])&0x7F < 24):
             # Pickle support
+            if isinstance(hour, str):
+                try:
+                    hour = hour.encode('latin1')
+                except UnicodeEncodeError:
+                    # More informative error message.
+                    raise ValueError(
+                        "Failed to encode latin1 string when unpickling "
+                        "a time object. "
+                        "pickle.load(data, encoding='latin1') is assumed.")
             self = object.__new__(cls)
             self.__setstate(hour, minute or None)
             self._hashcode = -1
@@ -1496,8 +1516,18 @@ class datetime(date):
 
     def __new__(cls, year, month=None, day=None, hour=0, minute=0, second=0,
                 microsecond=0, tzinfo=None, *, fold=0):
-        if isinstance(year, bytes) and len(year) == 10 and 1 <= year[2]&0x7F <= 12:
+        if (isinstance(year, (bytes, str)) and len(year) == 10 and
+            1 <= ord(year[2:3])&0x7F <= 12):
             # Pickle support
+            if isinstance(year, str):
+                try:
+                    year = bytes(year, 'latin1')
+                except UnicodeEncodeError:
+                    # More informative error message.
+                    raise ValueError(
+                        "Failed to encode latin1 string when unpickling "
+                        "a datetime object. "
+                        "pickle.load(data, encoding='latin1') is assumed.")
             self = object.__new__(cls)
             self.__setstate(year, month)
             self._hashcode = -1

Lib/test/datetimetester.py

@@ -38,6 +38,7 @@
 import _strptime
 #
 
+pickle_loads = {pickle.loads, pickle._loads}
 
 pickle_choices = [(pickle, pickle, proto)
                   for proto in range(pickle.HIGHEST_PROTOCOL + 1)]
@@ -1434,6 +1435,19 @@ def test_pickling(self):
             self.assertEqual(orig, derived)
         self.assertEqual(orig.__reduce__(), orig.__reduce_ex__(2))
 
+    def test_compat_unpickle(self):
+        tests = [
+            b"cdatetime\ndate\n(S'\\x07\\xdf\\x0b\\x1b'\ntR.",
+            b'cdatetime\ndate\n(U\x04\x07\xdf\x0b\x1btR.',
+            b'\x80\x02cdatetime\ndate\nU\x04\x07\xdf\x0b\x1b\x85R.',
+        ]
+        args = 2015, 11, 27
+        expected = self.theclass(*args)
+        for data in tests:
+            for loads in pickle_loads:
+                derived = loads(data, encoding='latin1')
+                self.assertEqual(derived, expected)
+
     def test_compare(self):
         t1 = self.theclass(2, 3, 4)
         t2 = self.theclass(2, 3, 4)
@@ -2098,6 +2112,24 @@ def test_pickling_subclass_datetime(self):
             derived = unpickler.loads(green)
             self.assertEqual(orig, derived)
 
+    def test_compat_unpickle(self):
+        tests = [
+            b'cdatetime\ndatetime\n('
+            b"S'\\x07\\xdf\\x0b\\x1b\\x14;\\x01\\x00\\x10\\x00'\ntR.",
+
+            b'cdatetime\ndatetime\n('
+            b'U\n\x07\xdf\x0b\x1b\x14;\x01\x00\x10\x00tR.',
+
+            b'\x80\x02cdatetime\ndatetime\n'
+            b'U\n\x07\xdf\x0b\x1b\x14;\x01\x00\x10\x00\x85R.',
+        ]
+        args = 2015, 11, 27, 20, 59, 1, 64**2
+        expected = self.theclass(*args)
+        for data in tests:
+            for loads in pickle_loads:
+                derived = loads(data, encoding='latin1')
+                self.assertEqual(derived, expected)
+
     def test_more_compare(self):
         # The test_compare() inherited from TestDate covers the error cases.
         # We just want to test lexicographic ordering on the members datetime
@@ -3069,6 +3101,19 @@ def test_pickling_subclass_time(self):
             derived = unpickler.loads(green)
             self.assertEqual(orig, derived)
 
+    def test_compat_unpickle(self):
+        tests = [
+            b"cdatetime\ntime\n(S'\\x14;\\x10\\x00\\x10\\x00'\ntR.",
+            b'cdatetime\ntime\n(U\x06\x14;\x10\x00\x10\x00tR.',
+            b'\x80\x02cdatetime\ntime\nU\x06\x14;\x10\x00\x10\x00\x85R.',
+        ]
+        args = 20, 59, 16, 64**2
+        expected = self.theclass(*args)
+        for data in tests:
+            for loads in pickle_loads:
+                derived = loads(data, encoding='latin1')
+                self.assertEqual(derived, expected)
+
     def test_bool(self):
         # time is always True.
         cls = self.theclass
@@ -3441,6 +3486,40 @@ def test_pickling(self):
             self.assertEqual(derived.tzname(), 'cookie')
         self.assertEqual(orig.__reduce__(), orig.__reduce_ex__(2))
 
+    def test_compat_unpickle(self):
+        tests = [
+            b"cdatetime\ntime\n(S'\\x05\\x06\\x07\\x01\\xe2@'\n"
+            b"ctest.datetimetester\nPicklableFixedOffset\n(tR"
+            b"(dS'_FixedOffset__offset'\ncdatetime\ntimedelta\n"
+            b"(I-1\nI68400\nI0\ntRs"
+            b"S'_FixedOffset__dstoffset'\nNs"
+            b"S'_FixedOffset__name'\nS'cookie'\nsbtR.",
+
+            b'cdatetime\ntime\n(U\x06\x05\x06\x07\x01\xe2@'
+            b'ctest.datetimetester\nPicklableFixedOffset\n)R'
+            b'}(U\x14_FixedOffset__offsetcdatetime\ntimedelta\n'
+            b'(J\xff\xff\xff\xffJ0\x0b\x01\x00K\x00tR'
+            b'U\x17_FixedOffset__dstoffsetN'
+            b'U\x12_FixedOffset__nameU\x06cookieubtR.',
+
+            b'\x80\x02cdatetime\ntime\nU\x06\x05\x06\x07\x01\xe2@'
+            b'ctest.datetimetester\nPicklableFixedOffset\n)R'
+            b'}(U\x14_FixedOffset__offsetcdatetime\ntimedelta\n'
+            b'J\xff\xff\xff\xffJ0\x0b\x01\x00K\x00\x87R'
+            b'U\x17_FixedOffset__dstoffsetN'
+            b'U\x12_FixedOffset__nameU\x06cookieub\x86R.',
+        ]
+
+        tinfo = PicklableFixedOffset(-300, 'cookie')
+        expected = self.theclass(5, 6, 7, 123456, tzinfo=tinfo)
+        for data in tests:
+            for loads in pickle_loads:
+                derived = loads(data, encoding='latin1')
+                self.assertEqual(derived, expected, repr(data))
+                self.assertIsInstance(derived.tzinfo, PicklableFixedOffset)
+                self.assertEqual(derived.utcoffset(), timedelta(minutes=-300))
+                self.assertEqual(derived.tzname(), 'cookie')
+
     def test_more_bool(self):
         # time is always True.
         cls = self.theclass
@@ -3789,6 +3868,43 @@ def test_pickling(self):
             self.assertEqual(derived.tzname(), 'cookie')
         self.assertEqual(orig.__reduce__(), orig.__reduce_ex__(2))
 
+    def test_compat_unpickle(self):
+        tests = [
+            b'cdatetime\ndatetime\n'
+            b"(S'\\x07\\xdf\\x0b\\x1b\\x14;\\x01\\x01\\xe2@'\n"
+            b'ctest.datetimetester\nPicklableFixedOffset\n(tR'
+            b"(dS'_FixedOffset__offset'\ncdatetime\ntimedelta\n"
+            b'(I-1\nI68400\nI0\ntRs'
+            b"S'_FixedOffset__dstoffset'\nNs"
+            b"S'_FixedOffset__name'\nS'cookie'\nsbtR.",
+
+            b'cdatetime\ndatetime\n'
+            b'(U\n\x07\xdf\x0b\x1b\x14;\x01\x01\xe2@'
+            b'ctest.datetimetester\nPicklableFixedOffset\n)R'
+            b'}(U\x14_FixedOffset__offsetcdatetime\ntimedelta\n'
+            b'(J\xff\xff\xff\xffJ0\x0b\x01\x00K\x00tR'
+            b'U\x17_FixedOffset__dstoffsetN'
+            b'U\x12_FixedOffset__nameU\x06cookieubtR.',
+
+            b'\x80\x02cdatetime\ndatetime\n'
+            b'U\n\x07\xdf\x0b\x1b\x14;\x01\x01\xe2@'
+            b'ctest.datetimetester\nPicklableFixedOffset\n)R'
+            b'}(U\x14_FixedOffset__offsetcdatetime\ntimedelta\n'
+            b'J\xff\xff\xff\xffJ0\x0b\x01\x00K\x00\x87R'
+            b'U\x17_FixedOffset__dstoffsetN'
+            b'U\x12_FixedOffset__nameU\x06cookieub\x86R.',
+        ]
+        args = 2015, 11, 27, 20, 59, 1, 123456
+        tinfo = PicklableFixedOffset(-300, 'cookie')
+        expected = self.theclass(*args, **{'tzinfo': tinfo})
+        for data in tests:
+            for loads in pickle_loads:
+                derived = loads(data, encoding='latin1')
+                self.assertEqual(derived, expected)
+                self.assertIsInstance(derived.tzinfo, PicklableFixedOffset)
+                self.assertEqual(derived.utcoffset(), timedelta(minutes=-300))
+                self.assertEqual(derived.tzname(), 'cookie')
+
     def test_extreme_hashes(self):
         # If an attempt is made to hash these via subtracting the offset
         # then hashing a datetime object, OverflowError results.  The

Misc/NEWS.d/next/Library/2017-10-12-22-39-55.bpo-22005.lGP-sc.rst

@@ -0,0 +1,3 @@
+Implemented unpickling instances of :class:`~datetime.datetime`,
+:class:`~datetime.date` and :class:`~datetime.time` pickled by Python 2.
+``encoding='latin1'`` should be used for successful decoding.