Merge branch 'main' into patch-1

Author: sharktide

Doc/library/uuid.rst

@@ -11,9 +11,10 @@
 --------------
 
 This module provides immutable :class:`UUID` objects (the :class:`UUID` class)
-and the functions :func:`uuid1`, :func:`uuid3`, :func:`uuid4`, :func:`uuid5`,
-:func:`uuid6`, and :func:`uuid8` for generating version 1, 3, 4, 5, 6,
-and 8 UUIDs as specified in :rfc:`9562` (which supersedes :rfc:`4122`).
+and :ref:`functions <uuid-factory-functions>` for generating UUIDs corresponding
+to a specific UUID version as specified in :rfc:`9562` (which supersedes :rfc:`4122`),
+for example, :func:`uuid1` for UUID version 1, :func:`uuid3` for UUID version 3, and so on.
+Note that UUID version 2 is deliberately omitted as it is outside the scope of the RFC.
 
 If all you want is a unique ID, you should probably call :func:`uuid1` or
 :func:`uuid4`.  Note that :func:`uuid1` may compromise privacy since it creates
@@ -154,7 +155,7 @@ which relays any information about the UUID's safety, using this enumeration:
    :const:`RFC_4122`).
 
    .. versionchanged:: next
-      Added UUID versions 6 and 8.
+      Added UUID versions 6, 7 and 8.
 
 
 .. attribute:: UUID.is_safe
@@ -185,6 +186,8 @@ The :mod:`uuid` module defines the following functions:
       globally unique, while the latter are not.
 
 
+.. _uuid-factory-functions:
+
 .. function:: uuid1(node=None, clock_seq=None)
 
    Generate a UUID from a host ID, sequence number, and the current time. If *node*
@@ -228,6 +231,18 @@ The :mod:`uuid` module defines the following functions:
    .. versionadded:: next
 
 
+.. function:: uuid7()
+
+   Generate a time-based UUID according to
+   :rfc:`RFC 9562, §5.7 <9562#section-5.7>`.
+
+   For portability across platforms lacking sub-millisecond precision, UUIDs
+   produced by this function embed a 48-bit timestamp and use a 42-bit counter
+   to guarantee monotonicity within a millisecond.
+
+   .. versionadded:: next
+
+
 .. function:: uuid8(a=None, b=None, c=None)
 
    Generate a pseudo-random UUID according to
@@ -330,7 +345,7 @@ The :mod:`uuid` module can be executed as a script from the command line.
 
 .. code-block:: sh
 
-   python -m uuid [-h] [-u {uuid1,uuid3,uuid4,uuid5,uuid6,uuid8}] [-n NAMESPACE] [-N NAME]
+   python -m uuid [-h] [-u {uuid1,uuid3,uuid4,uuid5,uuid6,uuid7,uuid8}] [-n NAMESPACE] [-N NAME]
 
 The following options are accepted:
 
@@ -347,7 +362,7 @@ The following options are accepted:
    is used.
 
    .. versionchanged:: next
-      Allow generating UUID versions 6 and 8.
+      Allow generating UUID versions 6, 7 and 8.
 
 .. option:: -n <namespace>
             --namespace <namespace>

Doc/whatsnew/3.14.rst

@@ -924,8 +924,9 @@ urllib
 uuid
 ----
 
-* Add support for UUID versions 6 and 8 via :func:`uuid.uuid6` and
-  :func:`uuid.uuid8` respectively, as specified in :rfc:`9562`.
+* Add support for UUID versions 6, 7, and 8 via :func:`uuid.uuid6`,
+  :func:`uuid.uuid7`, and :func:`uuid.uuid8` respectively, as specified
+  in :rfc:`9562`.
   (Contributed by Bénédikt Tran in :gh:`89083`.)
 
 * :const:`uuid.NIL` and :const:`uuid.MAX` are now available to represent the

Lib/test/test_asyncio/test_base_events.py

@@ -233,20 +233,25 @@ def test_set_default_executor_error(self):
         self.assertIsNone(self.loop._default_executor)
 
     def test_shutdown_default_executor_timeout(self):
+        event = threading.Event()
+
         class DummyExecutor(concurrent.futures.ThreadPoolExecutor):
             def shutdown(self, wait=True, *, cancel_futures=False):
                 if wait:
-                    time.sleep(0.1)
+                    event.wait()
 
         self.loop._process_events = mock.Mock()
         self.loop._write_to_self = mock.Mock()
         executor = DummyExecutor()
         self.loop.set_default_executor(executor)
 
-        with self.assertWarnsRegex(RuntimeWarning,
-                                   "The executor did not finishing joining"):
-            self.loop.run_until_complete(
-                self.loop.shutdown_default_executor(timeout=0.01))
+        try:
+            with self.assertWarnsRegex(RuntimeWarning,
+                                       "The executor did not finishing joining"):
+                self.loop.run_until_complete(
+                    self.loop.shutdown_default_executor(timeout=0.01))
+        finally:
+            event.set()
 
     def test_call_soon(self):
         def cb():

Lib/test/test_hmac.py

@@ -566,32 +566,52 @@ def test_with_fallback(self):
             cache.pop('foo')
 
 
-class PyRFCTestCase(PyTestVectorsMixin, ThroughObjectMixin,
-                    RFCTestCasesMixin, unittest.TestCase):
-    """Python implementation of HMAC using hmac.HMAC()."""
-
-
-class PyDotNewRFCTestCase(PyTestVectorsMixin, ThroughModuleAPIMixin,
-                          RFCTestCasesMixin, unittest.TestCase):
-    """Python implementation of HMAC using hmac.new()."""
-
-
-@hashlib_helper.requires_hashlib()
-class OpenSSLRFCTestCase(OpenSSLTestVectorsMixin, RFCTestCasesMixin,
-                         unittest.TestCase):
-    """OpenSSL implementation of HMAC."""
+class RFCWithOpenSSLHashFunctionTestCasesMixin(RFCTestCasesMixin):
 
     def __init_subclass__(cls, *args, **kwargs):
         super().__init_subclass__(*args, **kwargs)
 
         for name in cls.ALGORITHMS:
             @property
+            @hashlib_helper.requires_hashlib()
             @hashlib_helper.requires_hashdigest(name, openssl=True)
             def func(self, *, __name=name):  # __name needed to bind 'name'
                 return getattr(_hashlib, f'openssl_{__name}')
             setattr(cls, name, func)
 
 
+class PyRFCTestCase(PyTestVectorsMixin, ThroughObjectMixin,
+                    RFCWithOpenSSLHashFunctionTestCasesMixin,
+                    unittest.TestCase):
+    """Python implementation of HMAC using hmac.HMAC().
+
+    The underlying hash functions are OpenSSL-based.
+    """
+
+
+class PyDotNewRFCTestCase(PyTestVectorsMixin, ThroughModuleAPIMixin,
+                          RFCWithOpenSSLHashFunctionTestCasesMixin,
+                          unittest.TestCase):
+    """Python implementation of HMAC using hmac.new().
+
+    The underlying hash functions are OpenSSL-based.
+    """
+
+
+class OpenSSLRFCTestCase(OpenSSLTestVectorsMixin,
+                         RFCWithOpenSSLHashFunctionTestCasesMixin,
+                         unittest.TestCase):
+    """OpenSSL implementation of HMAC.
+
+    The underlying hash functions are also OpenSSL-based."""
+
+
+# TODO(picnixz): once we have a HACL* HMAC, we should also test the Python
+# implementation of HMAC with a HACL*-based hash function. For now, we only
+# test it partially via the '_sha2' module, but for completeness we could
+# also test the RFC test vectors against all possible implementations.
+
+
 class DigestModTestCaseMixin(CreatorMixin, DigestMixin):
     """Tests for the 'digestmod' parameter."""
 

Lib/test/test_uuid.py

@@ -871,6 +871,206 @@ def test_uuid6_test_vectors(self):
             equal((u.int >> 80) & 0xffff, 0x232a)
             equal((u.int >> 96) & 0xffff_ffff, 0x1ec9_414c)
 
+    def test_uuid7(self):
+        equal = self.assertEqual
+        u = self.uuid.uuid7()
+        equal(u.variant, self.uuid.RFC_4122)
+        equal(u.version, 7)
+
+        # 1 Jan 2023 12:34:56.123_456_789
+        timestamp_ns = 1672533296_123_456_789  # ns precision
+        timestamp_ms, _ = divmod(timestamp_ns, 1_000_000)
+
+        for _ in range(100):
+            counter_hi = random.getrandbits(11)
+            counter_lo = random.getrandbits(30)
+            counter = (counter_hi << 30) | counter_lo
+
+            tail = random.getrandbits(32)
+            # effective number of bits is 32 + 30 + 11 = 73
+            random_bits = counter << 32 | tail
+
+            # set all remaining MSB of fake random bits to 1 to ensure that
+            # the implementation correctly removes them
+            random_bits = (((1 << 7) - 1) << 73) | random_bits
+            random_data = random_bits.to_bytes(10)
+
+            with (
+                mock.patch.multiple(
+                    self.uuid,
+                    _last_timestamp_v7=None,
+                    _last_counter_v7=0,
+                ),
+                mock.patch('time.time_ns', return_value=timestamp_ns),
+                mock.patch('os.urandom', return_value=random_data) as urand
+            ):
+                u = self.uuid.uuid7()
+                urand.assert_called_once_with(10)
+                equal(u.variant, self.uuid.RFC_4122)
+                equal(u.version, 7)
+
+                equal(self.uuid._last_timestamp_v7, timestamp_ms)
+                equal(self.uuid._last_counter_v7, counter)
+
+                unix_ts_ms = timestamp_ms & 0xffff_ffff_ffff
+                equal((u.int >> 80) & 0xffff_ffff_ffff, unix_ts_ms)
+
+                equal((u.int >> 75) & 1, 0)  # check that the MSB is 0
+                equal((u.int >> 64) & 0xfff, counter_hi)
+                equal((u.int >> 32) & 0x3fff_ffff, counter_lo)
+                equal(u.int & 0xffff_ffff, tail)
+
+    def test_uuid7_uniqueness(self):
+        # Test that UUIDv7-generated values are unique.
+        #
+        # While UUIDv8 has an entropy of 122 bits, those 122 bits may not
+        # necessarily be sampled from a PRNG. On the other hand, UUIDv7
+        # uses os.urandom() as a PRNG which features better randomness.
+        N = 1000
+        uuids = {self.uuid.uuid7() for _ in range(N)}
+        self.assertEqual(len(uuids), N)
+
+        versions = {u.version for u in uuids}
+        self.assertSetEqual(versions, {7})
+
+    def test_uuid7_monotonicity(self):
+        equal = self.assertEqual
+
+        us = [self.uuid.uuid7() for _ in range(10_000)]
+        equal(us, sorted(us))
+
+        with mock.patch.multiple(
+            self.uuid,
+            _last_timestamp_v7=0,
+            _last_counter_v7=0,
+        ):
+            # 1 Jan 2023 12:34:56.123_456_789
+            timestamp_ns = 1672533296_123_456_789  # ns precision
+            timestamp_ms, _ = divmod(timestamp_ns, 1_000_000)
+
+            # counter_{hi,lo} are chosen so that "counter + 1" does not overflow
+            counter_hi = random.getrandbits(11)
+            counter_lo = random.getrandbits(29)
+            counter = (counter_hi << 30) | counter_lo
+            self.assertLess(counter + 1, 0x3ff_ffff_ffff)
+
+            tail = random.getrandbits(32)
+            random_bits = counter << 32 | tail
+            random_data = random_bits.to_bytes(10)
+
+            with (
+                mock.patch('time.time_ns', return_value=timestamp_ns),
+                mock.patch('os.urandom', return_value=random_data) as urand
+            ):
+                u1 = self.uuid.uuid7()
+                urand.assert_called_once_with(10)
+                equal(self.uuid._last_timestamp_v7, timestamp_ms)
+                equal(self.uuid._last_counter_v7, counter)
+                equal((u1.int >> 64) & 0xfff, counter_hi)
+                equal((u1.int >> 32) & 0x3fff_ffff, counter_lo)
+                equal(u1.int & 0xffff_ffff, tail)
+
+            # 1 Jan 2023 12:34:56.123_457_032 (same millisecond but not same ns)
+            next_timestamp_ns = 1672533296_123_457_032
+            next_timestamp_ms, _ = divmod(timestamp_ns, 1_000_000)
+            equal(timestamp_ms, next_timestamp_ms)
+
+            next_tail_bytes = os.urandom(4)
+            next_fail = int.from_bytes(next_tail_bytes)
+
+            with (
+                mock.patch('time.time_ns', return_value=next_timestamp_ns),
+                mock.patch('os.urandom', return_value=next_tail_bytes) as urand
+            ):
+                u2 = self.uuid.uuid7()
+                urand.assert_called_once_with(4)
+                # same milli-second
+                equal(self.uuid._last_timestamp_v7, timestamp_ms)
+                # 42-bit counter advanced by 1
+                equal(self.uuid._last_counter_v7, counter + 1)
+                equal((u2.int >> 64) & 0xfff, counter_hi)
+                equal((u2.int >> 32) & 0x3fff_ffff, counter_lo + 1)
+                equal(u2.int & 0xffff_ffff, next_fail)
+
+            self.assertLess(u1, u2)
+
+    def test_uuid7_timestamp_backwards(self):
+        equal = self.assertEqual
+        # 1 Jan 2023 12:34:56.123_456_789
+        timestamp_ns = 1672533296_123_456_789  # ns precision
+        timestamp_ms, _ = divmod(timestamp_ns, 1_000_000)
+        fake_last_timestamp_v7 = timestamp_ms + 1
+
+        # counter_{hi,lo} are chosen so that "counter + 1" does not overflow
+        counter_hi = random.getrandbits(11)
+        counter_lo = random.getrandbits(29)
+        counter = (counter_hi << 30) | counter_lo
+        self.assertLess(counter + 1, 0x3ff_ffff_ffff)
+
+        tail_bytes = os.urandom(4)
+        tail = int.from_bytes(tail_bytes)
+
+        with (
+            mock.patch.multiple(
+                self.uuid,
+                _last_timestamp_v7=fake_last_timestamp_v7,
+                _last_counter_v7=counter,
+            ),
+            mock.patch('time.time_ns', return_value=timestamp_ns),
+            mock.patch('os.urandom', return_value=tail_bytes) as urand
+        ):
+            u = self.uuid.uuid7()
+            urand.assert_called_once_with(4)
+            equal(u.variant, self.uuid.RFC_4122)
+            equal(u.version, 7)
+            equal(self.uuid._last_timestamp_v7, fake_last_timestamp_v7 + 1)
+            unix_ts_ms = (fake_last_timestamp_v7 + 1) & 0xffff_ffff_ffff
+            equal((u.int >> 80) & 0xffff_ffff_ffff, unix_ts_ms)
+            # 42-bit counter advanced by 1
+            equal(self.uuid._last_counter_v7, counter + 1)
+            equal((u.int >> 64) & 0xfff, counter_hi)
+            # 42-bit counter advanced by 1 (counter_hi is untouched)
+            equal((u.int >> 32) & 0x3fff_ffff, counter_lo + 1)
+            equal(u.int & 0xffff_ffff, tail)
+
+    def test_uuid7_overflow_counter(self):
+        equal = self.assertEqual
+        # 1 Jan 2023 12:34:56.123_456_789
+        timestamp_ns = 1672533296_123_456_789  # ns precision
+        timestamp_ms, _ = divmod(timestamp_ns, 1_000_000)
+
+        new_counter_hi = random.getrandbits(11)
+        new_counter_lo = random.getrandbits(30)
+        new_counter = (new_counter_hi << 30) | new_counter_lo
+
+        tail = random.getrandbits(32)
+        random_bits = (new_counter << 32) | tail
+        random_data = random_bits.to_bytes(10)
+
+        with (
+            mock.patch.multiple(
+                self.uuid,
+                _last_timestamp_v7=timestamp_ms,
+                # same timestamp, but force an overflow on the counter
+                _last_counter_v7=0x3ff_ffff_ffff,
+            ),
+            mock.patch('time.time_ns', return_value=timestamp_ns),
+            mock.patch('os.urandom', return_value=random_data) as urand
+        ):
+            u = self.uuid.uuid7()
+            urand.assert_called_with(10)
+            equal(u.variant, self.uuid.RFC_4122)
+            equal(u.version, 7)
+            # timestamp advanced due to overflow
+            equal(self.uuid._last_timestamp_v7, timestamp_ms + 1)
+            unix_ts_ms = (timestamp_ms + 1) & 0xffff_ffff_ffff
+            equal((u.int >> 80) & 0xffff_ffff_ffff, unix_ts_ms)
+            # counter overflowed, so we picked a new one
+            equal(self.uuid._last_counter_v7, new_counter)
+            equal((u.int >> 64) & 0xfff, new_counter_hi)
+            equal((u.int >> 32) & 0x3fff_ffff, new_counter_lo)
+            equal(u.int & 0xffff_ffff, tail)
+
     def test_uuid8(self):
         equal = self.assertEqual
         u = self.uuid.uuid8()