From fe6f9b1a933673a7f926b1c0c2b7847b3d400464 Mon Sep 17 00:00:00 2001
From: Georgy Moiseev <georgy.moiseev@corp.mail.ru>
Date: Mon, 24 Oct 2022 11:54:16 +0300
Subject: [PATCH 1/4] test: fix dbapi test connection resource warnings

DBAPI2 compliance tests are not implemented here but inherited from
external module [1]. Two tests from this module open a connection and
forget to close it. The issue had been filed together with patch PR to
module repository [2], but the last update was 7 years ago so it is
possibly that it would never be merged. This patch adds this PR change
with method overwrite.

1. https://pypi.org/project/dbapi-compliance/
2. https://github.com/baztian/dbapi-compliance/issues/5

Part of #250
---
 test/suites/test_dbapi.py | 40 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/test/suites/test_dbapi.py b/test/suites/test_dbapi.py
index 3873fa84..0df06337 100644
--- a/test/suites/test_dbapi.py
+++ b/test/suites/test_dbapi.py
@@ -127,3 +127,43 @@ def test_setoutputsize(self):  # Do nothing
     @unittest.skip('Not implemented')
     def test_description(self):
         pass
+
+    def test_ExceptionsAsConnectionAttributes(self):
+        # Workaround for https://github.com/baztian/dbapi-compliance/issues/5
+
+        # OPTIONAL EXTENSION 
+        # Test for the optional DB API 2.0 extension, where the exceptions 
+        # are exposed as attributes on the Connection object 
+        # I figure this optional extension will be implemented by any 
+        # driver author who is using this test suite, so it is enabled 
+        # by default. 
+        drv = self.driver
+        con = self._connect()
+        try:
+            dbapi20._failUnless(self,con.Warning is drv.Warning)
+            dbapi20._failUnless(self,con.Error is drv.Error)
+            dbapi20._failUnless(self,con.InterfaceError is drv.InterfaceError)
+            dbapi20._failUnless(self,con.DatabaseError is drv.DatabaseError)
+            dbapi20._failUnless(self,con.OperationalError is drv.OperationalError)
+            dbapi20._failUnless(self,con.IntegrityError is drv.IntegrityError)
+            dbapi20._failUnless(self,con.InternalError is drv.InternalError)
+            dbapi20._failUnless(self,con.ProgrammingError is drv.ProgrammingError)
+            dbapi20. _failUnless(self,con.NotSupportedError is drv.NotSupportedError)
+        finally:
+            con.close()
+
+
+    def test_rollback(self):
+        # Workaround for https://github.com/baztian/dbapi-compliance/issues/5
+
+        con = self._connect()
+        try:
+            # If rollback is defined, it should either work or throw
+            # the documented exception
+            if hasattr(con,'rollback'):
+                try:
+                    con.rollback()
+                except self.driver.NotSupportedError:
+                    pass
+        finally:
+            con.close()

From b338a1fbb14d4a0cb55d29d6ecaf87b2cadd6cde Mon Sep 17 00:00:00 2001
From: Georgy Moiseev <georgy.moiseev@corp.mail.ru>
Date: Mon, 24 Oct 2022 15:46:21 +0300
Subject: [PATCH 2/4] api: pandas way to build datetime from timestamp

This option is required so it would be possible to decode Datetime with
external function without constructing excessive pandas.Timestamp
object.

Follows #204
---
 CHANGELOG.md                            | 45 ++++++++++++
 tarantool/msgpack_ext/types/datetime.py | 98 +++++++++++++++++++------
 test/suites/test_datetime.py            |  6 ++
 3 files changed, 128 insertions(+), 21 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6b8b2a75..1852f4d8 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -136,6 +136,51 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Support iproto feature discovery (#206).
 
+- Support pandas way to build datetime from timestamp (PR #252).
+
+  `timestamp_since_utc_epoch` is a parameter to set timestamp
+  convertion behavior for timezone-aware datetimes.
+
+  If ``False`` (default), behaves similar to Tarantool `datetime.new()`:
+
+  ```python
+  >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=False)
+  >>> dt
+  datetime: Timestamp('2022-01-01 00:00:00'), tz: ""
+  >>> dt.timestamp
+  1640995200.0
+  >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow',
+  ...                         timestamp_since_utc_epoch=False)
+  >>> dt
+  datetime: Timestamp('2022-01-01 00:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow"
+  >>> dt.timestamp
+  1640984400.0
+  ```
+
+  Thus, if ``False``, datetime is computed from timestamp
+  since epoch and then timezone is applied without any
+  convertion. In that case, `dt.timestamp` won't be equal to
+  initialization `timestamp` for all timezones with non-zero offset.
+
+  If ``True``, behaves similar to `pandas.Timestamp`:
+
+  ```python
+  >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=True)
+  >>> dt
+  datetime: Timestamp('2022-01-01 00:00:00'), tz: ""
+  >>> dt.timestamp
+  1640995200.0
+  >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow',
+  ...                         timestamp_since_utc_epoch=True)
+  >>> dt
+  datetime: Timestamp('2022-01-01 03:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow"
+  >>> dt.timestamp
+  1640995200.0
+  ```
+
+  Thus, if ``True``, datetime is computed in a way that `dt.timestamp` will
+  always be equal to initialization `timestamp`.
+
 ### Changed
 - Bump msgpack requirement to 1.0.4 (PR #223).
   The only reason of this bump is various vulnerability fixes,
diff --git a/tarantool/msgpack_ext/types/datetime.py b/tarantool/msgpack_ext/types/datetime.py
index d84352dd..b2dac8a9 100644
--- a/tarantool/msgpack_ext/types/datetime.py
+++ b/tarantool/msgpack_ext/types/datetime.py
@@ -279,7 +279,7 @@ class Datetime():
 
     def __init__(self, data=None, *, timestamp=None, year=None, month=None,
                  day=None, hour=None, minute=None, sec=None, nsec=None,
-                 tzoffset=0, tz=''):
+                 tzoffset=0, tz='', timestamp_since_utc_epoch=False):
         """
         :param data: MessagePack binary data to decode. If provided,
             all other parameters are ignored.
@@ -294,7 +294,10 @@ def __init__(self, data=None, *, timestamp=None, year=None, month=None,
             :paramref:`~tarantool.Datetime.params.minute`,
             :paramref:`~tarantool.Datetime.params.sec`.
             If :paramref:`~tarantool.Datetime.params.nsec` is provided,
-            it must be :obj:`int`.
+            it must be :obj:`int`. Refer to
+            :paramref:`~tarantool.Datetime.params.timestamp_since_utc_epoch`
+            to clarify how timezone-aware datetime is computed from
+            the timestamp.
         :type timestamp: :obj:`float` or :obj:`int`, optional
 
         :param year: Datetime year value. Must be a valid
@@ -344,8 +347,60 @@ def __init__(self, data=None, *, timestamp=None, year=None, month=None,
         :param tz: Timezone name from Olson timezone database.
         :type tz: :obj:`str`, optional
 
+        :param timestamp_since_utc_epoch: Parameter to set timestamp
+            convertion behavior for timezone-aware datetimes.
+
+            If ``False`` (default), behaves similar to Tarantool
+            `datetime.new()`_:
+
+            .. code-block:: python
+
+                >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=False)
+                >>> dt
+                datetime: Timestamp('2022-01-01 00:00:00'), tz: ""
+                >>> dt.timestamp
+                1640995200.0
+                >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow',
+                ...                         timestamp_since_utc_epoch=False)
+                >>> dt
+                datetime: Timestamp('2022-01-01 00:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow"
+                >>> dt.timestamp
+                1640984400.0
+
+            Thus, if ``False``, datetime is computed from timestamp
+            since epoch and then timezone is applied without any
+            convertion. In that case,
+            :attr:`~tarantool.Datetime.timestamp` won't be equal to
+            initialization
+            :paramref:`~tarantool.Datetime.params.timestamp` for all
+            timezones with non-zero offset.
+
+            If ``True``, behaves similar to :class:`pandas.Timestamp`:
+
+            .. code-block:: python
+
+                >>> dt = tarantool.Datetime(timestamp=1640995200, timestamp_since_utc_epoch=True)
+                >>> dt
+                datetime: Timestamp('2022-01-01 00:00:00'), tz: ""
+                >>> dt.timestamp
+                1640995200.0
+                >>> dt = tarantool.Datetime(timestamp=1640995200, tz='Europe/Moscow',
+                ...                         timestamp_since_utc_epoch=True)
+                >>> dt
+                datetime: Timestamp('2022-01-01 03:00:00+0300', tz='Europe/Moscow'), tz: "Europe/Moscow"
+                >>> dt.timestamp
+                1640995200.0
+
+            Thus, if ``True``, datetime is computed in a way that
+            :attr:`~tarantool.Datetime.timestamp` will always be equal
+            to initialization
+            :paramref:`~tarantool.Datetime.params.timestamp`.
+        :type timestamp_since_utc_epoch: :obj:`bool`, optional
+
         :raise: :exc:`ValueError`, :exc:`~tarantool.error.MsgpackError`,
             :class:`pandas.Timestamp` exceptions
+
+        .. _datetime.new(): https://www.tarantool.io/en/doc/latest/reference/reference_lua/datetime/new/
         """
 
         if data is not None:
@@ -358,6 +413,16 @@ def __init__(self, data=None, *, timestamp=None, year=None, month=None,
             self._tz = tz
             return
 
+        tzinfo = None
+        if tz != '':
+            if tz not in tt_timezones.timezoneToIndex:
+                raise ValueError(f'Unknown Tarantool timezone "{tz}"')
+
+            tzinfo = get_python_tzinfo(tz, ValueError)
+        elif tzoffset != 0:
+            tzinfo = pytz.FixedOffset(tzoffset)
+        self._tz = tz
+
         # The logic is same as in Tarantool, refer to datetime API.
         # https://www.tarantool.io/en/doc/latest/reference/reference_lua/datetime/new/
         if timestamp is not None:
@@ -375,6 +440,11 @@ def __init__(self, data=None, *, timestamp=None, year=None, month=None,
                 datetime = pandas.to_datetime(total_nsec, unit='ns')
             else:
                 datetime = pandas.to_datetime(timestamp, unit='s')
+
+            if not timestamp_since_utc_epoch:
+                self._datetime = datetime.replace(tzinfo=tzinfo)
+            else:
+                self._datetime = datetime.replace(tzinfo=pytz.UTC).tz_convert(tzinfo)
         else:
             if nsec is not None:
                 microsecond = nsec // NSEC_IN_MKSEC
@@ -383,25 +453,11 @@ def __init__(self, data=None, *, timestamp=None, year=None, month=None,
                 microsecond = 0
                 nanosecond = 0
 
-            datetime = pandas.Timestamp(year=year, month=month, day=day,
-                                        hour=hour, minute=minute, second=sec,
-                                        microsecond=microsecond,
-                                        nanosecond=nanosecond)
-
-        if tz != '':
-            if tz not in tt_timezones.timezoneToIndex:
-                raise ValueError(f'Unknown Tarantool timezone "{tz}"')
-
-            tzinfo = get_python_tzinfo(tz, ValueError)
-            self._datetime = datetime.replace(tzinfo=tzinfo)
-            self._tz = tz
-        elif tzoffset != 0:
-            tzinfo = pytz.FixedOffset(tzoffset)
-            self._datetime = datetime.replace(tzinfo=tzinfo)
-            self._tz = ''
-        else:
-            self._datetime = datetime
-            self._tz = ''
+            self._datetime = pandas.Timestamp(
+                year=year, month=month, day=day,
+                hour=hour, minute=minute, second=sec,
+                microsecond=microsecond,
+                nanosecond=nanosecond, tzinfo=tzinfo)
 
     def _interval_operation(self, other, sign=1):
         """
diff --git a/test/suites/test_datetime.py b/test/suites/test_datetime.py
index 3e4ee763..583f9058 100644
--- a/test/suites/test_datetime.py
+++ b/test/suites/test_datetime.py
@@ -270,6 +270,12 @@ def test_Datetime_class_invalid_init(self):
             'tarantool': r"datetime.new({year=2022, month=8, day=31, hour=18, min=7, sec=54, " +
                          r"nsec=308543321, tz='AZODT'})",
         },
+        'timestamp_since_utc_epoch': {
+            'python': tarantool.Datetime(timestamp=1661958474, nsec=308543321,
+                                         tz='Europe/Moscow', timestamp_since_utc_epoch=True),
+            'msgpack': (b'\x4a\x79\x0f\x63\x00\x00\x00\x00\x59\xff\x63\x12\xb4\x00\xb3\x03'),
+            'tarantool': r"datetime.new({timestamp=1661969274, nsec=308543321, tz='Europe/Moscow'})",
+        },
     }
 
     def test_msgpack_decode(self):

From 1f842558cda8451d80211e1388a19e497d9c6107 Mon Sep 17 00:00:00 2001
From: Georgy Moiseev <georgy.moiseev@corp.mail.ru>
Date: Mon, 24 Oct 2022 16:06:34 +0300
Subject: [PATCH 3/4] api: extract datetime encode/decode from class

Extract tarantool.Datetime encode and decode to external functions. This
is a breaking change, but since there is no tagged release with Datetime
yet and API was more internal rather than public, it shouldn't be an
issue.

Follows #204
---
 CHANGELOG.md                            |   2 +
 tarantool/msgpack_ext/datetime.py       | 146 +++++++++++++++++-
 tarantool/msgpack_ext/types/datetime.py | 191 +-----------------------
 test/suites/test_datetime.py            |   8 +-
 4 files changed, 153 insertions(+), 194 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 1852f4d8..ff7b56d8 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -189,6 +189,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Update API documentation strings (#67).
 - Update documentation index, quick start and guide pages (#67).
 - Use git version to set package version (#238).
+- Extract tarantool.Datetime encode and decode to external
+  functions (PR #252).
 
 ### Fixed
 - Package build (#238).
diff --git a/tarantool/msgpack_ext/datetime.py b/tarantool/msgpack_ext/datetime.py
index e47f162e..fc1045d4 100644
--- a/tarantool/msgpack_ext/datetime.py
+++ b/tarantool/msgpack_ext/datetime.py
@@ -1,18 +1,83 @@
 """
 Tarantool `datetime`_ extension type support module.
 
-Refer to :mod:`~tarantool.msgpack_ext.types.datetime`.
+The datetime MessagePack representation looks like this:
 
-.. _datetime: https://www.tarantool.io/en/doc/latest/dev_guide/internals/msgpack_extensions/#the-datetime-type
+.. code-block:: text
+
+    +---------+----------------+==========+-----------------+
+    | MP_EXT  | MP_DATETIME    | seconds  | nsec; tzoffset; |
+    | = d7/d8 | = 4            |          | tzindex;        |
+    +---------+----------------+==========+-----------------+
+
+MessagePack data contains:
+
+* Seconds (8 bytes) as an unencoded 64-bit signed integer stored in the
+  little-endian order.
+* The optional fields (8 bytes), if any of them have a non-zero value.
+  The fields include nsec (4 bytes), tzoffset (2 bytes), and
+  tzindex (2 bytes) packed in the little-endian order.
+
+``seconds`` is seconds since Epoch, where the epoch is the point where
+the time starts, and is platform dependent. For Unix, the epoch is
+January 1, 1970, 00:00:00 (UTC). Tarantool uses a ``double`` type, see a
+structure definition in src/lib/core/datetime.h and reasons in
+`datetime RFC`_.
+
+``nsec`` is nanoseconds, fractional part of seconds. Tarantool uses
+``int32_t``, see a definition in src/lib/core/datetime.h.
+
+``tzoffset`` is timezone offset in minutes from UTC. Tarantool uses
+``int16_t`` type, see a structure definition in src/lib/core/datetime.h.
+
+``tzindex`` is Olson timezone id. Tarantool uses ``int16_t`` type, see
+a structure definition in src/lib/core/datetime.h. If both
+``tzoffset`` and ``tzindex`` are specified, ``tzindex`` has the
+preference and the ``tzoffset`` value is ignored.
+
+.. _datetime RFC: https://github.com/tarantool/tarantool/wiki/Datetime-internals#intervals-in-c
 """
 
-from tarantool.msgpack_ext.types.datetime import Datetime
+from tarantool.msgpack_ext.types.datetime import (
+    NSEC_IN_SEC,
+    SEC_IN_MIN,
+    Datetime,
+)
+import tarantool.msgpack_ext.types.timezones as tt_timezones
+
+from tarantool.error import MsgpackError
 
 EXT_ID = 4
 """
 `datetime`_ type id.
 """
 
+BYTEORDER = 'little'
+
+SECONDS_SIZE_BYTES  = 8
+NSEC_SIZE_BYTES     = 4
+TZOFFSET_SIZE_BYTES = 2
+TZINDEX_SIZE_BYTES  = 2
+
+
+def get_int_as_bytes(data, size):
+    """
+    Get binary representation of integer value.
+
+    :param data: Integer value.
+    :type data: :obj:`int`
+
+    :param size: Integer size, in bytes.
+    :type size: :obj:`int`
+
+    :return: Encoded integer.
+    :rtype: :obj:`bytes`
+
+    :meta private:
+    """
+
+    return data.to_bytes(size, byteorder=BYTEORDER, signed=True)
+
 def encode(obj):
     """
     Encode a datetime object.
@@ -26,7 +91,48 @@ def encode(obj):
     :raise: :exc:`tarantool.Datetime.msgpack_encode` exceptions
     """
 
-    return obj.msgpack_encode()
+    seconds = obj.value // NSEC_IN_SEC
+    nsec = obj.nsec
+    tzoffset = obj.tzoffset
+
+    tz = obj.tz
+    if tz != '':
+        tzindex = tt_timezones.timezoneToIndex[tz]
+    else:
+        tzindex = 0
+
+    buf = get_int_as_bytes(seconds, SECONDS_SIZE_BYTES)
+
+    if (nsec != 0) or (tzoffset != 0) or (tzindex != 0):
+        buf = buf + get_int_as_bytes(nsec, NSEC_SIZE_BYTES)
+        buf = buf + get_int_as_bytes(tzoffset, TZOFFSET_SIZE_BYTES)
+        buf = buf + get_int_as_bytes(tzindex, TZINDEX_SIZE_BYTES)
+
+    return buf
+
+
+def get_bytes_as_int(data, cursor, size):
+    """
+    Get integer value from binary data.
+
+    :param data: MessagePack binary data.
+    :type data: :obj:`bytes`
+
+    :param cursor: Index after last parsed byte.
+    :type cursor: :obj:`int`
+
+    :param size: Integer size, in bytes.
+    :type size: :obj:`int`
+
+    :return: First value: parsed integer, second value: new cursor
+        position.
+    :rtype: first value: :obj:`int`, second value: :obj:`int`
+
+    :meta private:
+    """
+
+    part = data[cursor:cursor + size]
+    return int.from_bytes(part, BYTEORDER, signed=True), cursor + size
 
 def decode(data):
     """
@@ -38,7 +144,35 @@ def decode(data):
     :return: Decoded datetime.
     :rtype: :class:`tarantool.Datetime`
 
-    :raise: :exc:`tarantool.Datetime` exceptions
+    :raise: :exc:`~tarantool.error.MsgpackError`,
+        :exc:`tarantool.Datetime` exceptions
     """
 
-    return Datetime(data)
+    cursor = 0
+    seconds, cursor = get_bytes_as_int(data, cursor, SECONDS_SIZE_BYTES)
+
+    data_len = len(data)
+    if data_len == (SECONDS_SIZE_BYTES + NSEC_SIZE_BYTES + \
+                    TZOFFSET_SIZE_BYTES + TZINDEX_SIZE_BYTES):
+        nsec, cursor     = get_bytes_as_int(data, cursor, NSEC_SIZE_BYTES)
+        tzoffset, cursor = get_bytes_as_int(data, cursor, TZOFFSET_SIZE_BYTES)
+        tzindex, cursor  = get_bytes_as_int(data, cursor, TZINDEX_SIZE_BYTES)
+    elif data_len == SECONDS_SIZE_BYTES:
+        nsec = 0
+        tzoffset = 0
+        tzindex = 0
+    else:
+        raise MsgpackError(f'Unexpected datetime payload length {data_len}')
+
+    if tzindex != 0:
+        if tzindex not in tt_timezones.indexToTimezone:
+            raise MsgpackError(f'Failed to decode datetime with unknown tzindex "{tzindex}"')
+        tz = tt_timezones.indexToTimezone[tzindex]
+        return Datetime(timestamp=seconds, nsec=nsec, tz=tz,
+                        timestamp_since_utc_epoch=True)
+    elif tzoffset != 0:
+        return Datetime(timestamp=seconds, nsec=nsec, tzoffset=tzoffset,
+                        timestamp_since_utc_epoch=True)
+    else:
+        return Datetime(timestamp=seconds, nsec=nsec,
+                        timestamp_since_utc_epoch=True)
diff --git a/tarantool/msgpack_ext/types/datetime.py b/tarantool/msgpack_ext/types/datetime.py
index b2dac8a9..f5912dda 100644
--- a/tarantool/msgpack_ext/types/datetime.py
+++ b/tarantool/msgpack_ext/types/datetime.py
@@ -1,41 +1,5 @@
 """
-Tarantool `datetime`_ extension type support module.
-
-The datetime MessagePack representation looks like this:
-
-.. code-block:: text
-
-    +---------+----------------+==========+-----------------+
-    | MP_EXT  | MP_DATETIME    | seconds  | nsec; tzoffset; |
-    | = d7/d8 | = 4            |          | tzindex;        |
-    +---------+----------------+==========+-----------------+
-
-MessagePack data contains:
-
-* Seconds (8 bytes) as an unencoded 64-bit signed integer stored in the
-  little-endian order.
-* The optional fields (8 bytes), if any of them have a non-zero value.
-  The fields include nsec (4 bytes), tzoffset (2 bytes), and
-  tzindex (2 bytes) packed in the little-endian order.
-
-``seconds`` is seconds since Epoch, where the epoch is the point where
-the time starts, and is platform dependent. For Unix, the epoch is
-January 1, 1970, 00:00:00 (UTC). Tarantool uses a ``double`` type, see a
-structure definition in src/lib/core/datetime.h and reasons in
-`datetime RFC`_.
-
-``nsec`` is nanoseconds, fractional part of seconds. Tarantool uses
-``int32_t``, see a definition in src/lib/core/datetime.h.
-
-``tzoffset`` is timezone offset in minutes from UTC. Tarantool uses
-``int16_t`` type, see a structure definition in src/lib/core/datetime.h.
-
-``tzindex`` is Olson timezone id. Tarantool uses ``int16_t`` type, see
-a structure definition in src/lib/core/datetime.h. If both
-``tzoffset`` and ``tzindex`` are specified, ``tzindex`` has the
-preference and the ``tzoffset`` value is ignored.
-
-.. _datetime RFC: https://github.com/tarantool/tarantool/wiki/Datetime-internals#intervals-in-c
+Tarantool `datetime`_ extension type implementation module.
 """
 
 from copy import deepcopy
@@ -44,63 +8,14 @@
 import pytz
 
 import tarantool.msgpack_ext.types.timezones as tt_timezones
-from tarantool.error import MsgpackError
 
 from tarantool.msgpack_ext.types.interval import Interval, Adjust
 
-SECONDS_SIZE_BYTES  = 8
-NSEC_SIZE_BYTES     = 4
-TZOFFSET_SIZE_BYTES = 2
-TZINDEX_SIZE_BYTES  = 2
-
-BYTEORDER = 'little'
-
 NSEC_IN_SEC = 1000000000
 NSEC_IN_MKSEC = 1000
 SEC_IN_MIN = 60
 MONTH_IN_YEAR = 12
 
-def get_bytes_as_int(data, cursor, size):
-    """
-    Get integer value from binary data.
-
-    :param data: MessagePack binary data.
-    :type data: :obj:`bytes`
-
-    :param cursor: Index after last parsed byte.
-    :type cursor: :obj:`int`
-
-    :param size: Integer size, in bytes.
-    :type size: :obj:`int`
-
-    :return: First value: parsed integer, second value: new cursor
-        position.
-    :rtype: first value: :obj:`int`, second value: :obj:`int`
-
-    :meta private:
-    """
-
-    part = data[cursor:cursor + size]
-    return int.from_bytes(part, BYTEORDER, signed=True), cursor + size
-
-def get_int_as_bytes(data, size):
-    """
-    Get binary representation of integer value.
-
-    :param data: Integer value.
-    :type data: :obj:`int`
-
-    :param size: Integer size, in bytes.
-    :type size: :obj:`int`
-
-    :return: Encoded integer.
-    :rtype: :obj:`bytes`
-
-    :meta private:
-    """
-
-    return data.to_bytes(size, byteorder=BYTEORDER, signed=True)
-
 def compute_offset(timestamp):
     """
     Compute timezone offset. Offset is computed each time and not stored
@@ -126,7 +41,7 @@ def compute_offset(timestamp):
     # There is no precision loss since offset is in minutes
     return int(utc_offset.total_seconds()) // SEC_IN_MIN
 
-def get_python_tzinfo(tz, error_class):
+def get_python_tzinfo(tz):
     """
     All non-abbreviated Tarantool timezones are represented as pytz
     timezones (from :func:`pytz.timezone`). All non-ambiguous
@@ -138,9 +53,6 @@ def get_python_tzinfo(tz, error_class):
     :param tz: Tarantool timezone name.
     :type tz: :obj:`str`
 
-    :param error_class: Error class to raise in case of fail.
-    :type error_class: :obj:`Exception`
-
     :return: Timezone object.
     :rtype: :func:`pytz.timezone` result or :class:`pytz.FixedOffset`
 
@@ -155,66 +67,17 @@ def get_python_tzinfo(tz, error_class):
     # Checked with timezones/validate_timezones.py
     tt_tzinfo = tt_timezones.timezoneAbbrevInfo[tz]
     if (tt_tzinfo['category'] & tt_timezones.TZ_AMBIGUOUS) != 0:
-        raise error_class(f'Failed to create datetime with ambiguous timezone "{tz}"')
+        raise ValueError(f'Failed to create datetime with ambiguous timezone "{tz}"')
 
     return pytz.FixedOffset(tt_tzinfo['offset'])
 
-def msgpack_decode(data):
-    """
-    Decode MsgPack binary data to useful timestamp and timezone data.
-    For internal use of :class:`~tarantool.Datetime`.
-
-    :param data: MessagePack binary data to decode.
-    :type data: :obj:`bytes`
-
-    :return: First value: timestamp data with timezone info, second
-        value: Tarantool timezone name.
-    :rtype: first value: :class:`pandas.Timestamp`, second value:
-        :obj:`str`
-
-    :raises: :exc:`~tarantool.error.MsgpackError`
-
-    :meta private:
-    """
-
-    cursor = 0
-    seconds, cursor = get_bytes_as_int(data, cursor, SECONDS_SIZE_BYTES)
-
-    data_len = len(data)
-    if data_len == (SECONDS_SIZE_BYTES + NSEC_SIZE_BYTES + \
-                    TZOFFSET_SIZE_BYTES + TZINDEX_SIZE_BYTES):
-        nsec, cursor     = get_bytes_as_int(data, cursor, NSEC_SIZE_BYTES)
-        tzoffset, cursor = get_bytes_as_int(data, cursor, TZOFFSET_SIZE_BYTES)
-        tzindex, cursor  = get_bytes_as_int(data, cursor, TZINDEX_SIZE_BYTES)
-    elif data_len == SECONDS_SIZE_BYTES:
-        nsec = 0
-        tzoffset = 0
-        tzindex = 0
-    else:
-        raise MsgpackError(f'Unexpected datetime payload length {data_len}')
-
-    total_nsec = seconds * NSEC_IN_SEC + nsec
-    datetime = pandas.to_datetime(total_nsec, unit='ns')
-
-    if tzindex != 0:
-        if tzindex not in tt_timezones.indexToTimezone:
-            raise MsgpackError(f'Failed to decode datetime with unknown tzindex "{tzindex}"')
-        tz = tt_timezones.indexToTimezone[tzindex]
-        tzinfo = get_python_tzinfo(tz, MsgpackError)
-        return datetime.replace(tzinfo=pytz.UTC).tz_convert(tzinfo), tz
-    elif tzoffset != 0:
-        tzinfo = pytz.FixedOffset(tzoffset)
-        return datetime.replace(tzinfo=pytz.UTC).tz_convert(tzinfo), ''
-    else:
-        return datetime, ''
-
 class Datetime():
     """
     Class representing Tarantool `datetime`_ info. Internals are based
     on :class:`pandas.Timestamp`.
 
-    You can create :class:`~tarantool.Datetime` objects either from
-    MessagePack data or by using the same API as in Tarantool:
+    You can create :class:`~tarantool.Datetime` objects by using the
+    same API as in Tarantool:
 
     .. code-block:: python
 
@@ -277,14 +140,10 @@ class Datetime():
     .. _datetime: https://www.tarantool.io/en/doc/latest/dev_guide/internals/msgpack_extensions/#the-datetime-type
     """
 
-    def __init__(self, data=None, *, timestamp=None, year=None, month=None,
+    def __init__(self, *, timestamp=None, year=None, month=None,
                  day=None, hour=None, minute=None, sec=None, nsec=None,
                  tzoffset=0, tz='', timestamp_since_utc_epoch=False):
         """
-        :param data: MessagePack binary data to decode. If provided,
-            all other parameters are ignored.
-        :type data: :obj:`bytes`, optional
-
         :param timestamp: Timestamp since epoch. Cannot be provided
             together with
             :paramref:`~tarantool.Datetime.params.year`,
@@ -403,22 +262,12 @@ def __init__(self, data=None, *, timestamp=None, year=None, month=None,
         .. _datetime.new(): https://www.tarantool.io/en/doc/latest/reference/reference_lua/datetime/new/
         """
 
-        if data is not None:
-            if not isinstance(data, bytes):
-                raise ValueError('data argument (first positional argument) ' +
-                                 'expected to be a "bytes" instance')
-
-            datetime, tz = msgpack_decode(data)
-            self._datetime = datetime
-            self._tz = tz
-            return
-
         tzinfo = None
         if tz != '':
             if tz not in tt_timezones.timezoneToIndex:
                 raise ValueError(f'Unknown Tarantool timezone "{tz}"')
 
-            tzinfo = get_python_tzinfo(tz, ValueError)
+            tzinfo = get_python_tzinfo(tz)
         elif tzoffset != 0:
             tzinfo = pytz.FixedOffset(tzoffset)
         self._tz = tz
@@ -783,29 +632,3 @@ def value(self):
         """
 
         return self._datetime.value
-
-    def msgpack_encode(self):
-        """
-        Encode a datetime object.
-
-        :rtype: :obj:`bytes`
-        """
-
-        seconds = self.value // NSEC_IN_SEC
-        nsec = self.nsec
-        tzoffset = self.tzoffset
-
-        tz = self.tz
-        if tz != '':
-            tzindex = tt_timezones.timezoneToIndex[tz]
-        else:
-            tzindex = 0
-
-        buf = get_int_as_bytes(seconds, SECONDS_SIZE_BYTES)
-
-        if (nsec != 0) or (tzoffset != 0) or (tzindex != 0):
-            buf = buf + get_int_as_bytes(nsec, NSEC_SIZE_BYTES)
-            buf = buf + get_int_as_bytes(tzoffset, TZOFFSET_SIZE_BYTES)
-            buf = buf + get_int_as_bytes(tzindex, TZINDEX_SIZE_BYTES)
-
-        return buf
diff --git a/test/suites/test_datetime.py b/test/suites/test_datetime.py
index 583f9058..fb2d2855 100644
--- a/test/suites/test_datetime.py
+++ b/test/suites/test_datetime.py
@@ -104,14 +104,14 @@ def test_Datetime_class_API_wth_tz(self):
         'positional_year': {
             'args': [2022],
             'kwargs': {},
-            'type': ValueError,
-            'msg': 'data argument (first positional argument) expected to be a "bytes" instance'
+            'type': TypeError,
+            'msg': '__init__() takes 1 positional argument but 2 were given'
         },
         'positional_date': {
             'args': [2022, 8, 31],
             'kwargs': {},
             'type': TypeError,
-            'msg': '__init__() takes from 1 to 2 positional arguments but 4 were given'
+            'msg': '__init__() takes 1 positional argument but 4 were given'
         },
         'mixing_date_and_timestamp': {
             'args': [],
@@ -338,7 +338,7 @@ def test_msgpack_decode_unknown_tzindex(self):
     def test_msgpack_decode_ambiguous_tzindex(self):
         case = b'\x4a\x79\x0f\x63\x00\x00\x00\x00\x59\xff\x63\x12\x00\x00\x82\x00'
         self.assertRaisesRegex(
-            MsgpackError, 'Failed to create datetime with ambiguous timezone "AET"',
+            ValueError, 'Failed to create datetime with ambiguous timezone "AET"',
             lambda: unpacker_ext_hook(4, case))
 
 

From db9e5b864e4604b477259563487b2e9e85a3f069 Mon Sep 17 00:00:00 2001
From: Georgy Moiseev <georgy.moiseev@corp.mail.ru>
Date: Mon, 24 Oct 2022 16:27:05 +0300
Subject: [PATCH 4/4] api: extract interval encode/decode from class

Extract tarantool.Interval encode and decode to external functions. This
is a breaking change, but since there is no tagged release with Interval
yet and API was more internal rather than public, it shouldn't be an
issue.

Follows #229
---
 CHANGELOG.md                            |   2 +
 tarantool/msgpack_ext/interval.py       | 103 ++++++++++++++++--
 tarantool/msgpack_ext/types/interval.py | 136 +++---------------------
 test/suites/test_interval.py            |  43 +-------
 4 files changed, 112 insertions(+), 172 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index ff7b56d8..eed24c57 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -191,6 +191,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Use git version to set package version (#238).
 - Extract tarantool.Datetime encode and decode to external
   functions (PR #252).
+- Extract tarantool.Interval encode and decode to external
+  functions (PR #252).
 
 ### Fixed
 - Package build (#238).
diff --git a/tarantool/msgpack_ext/interval.py b/tarantool/msgpack_ext/interval.py
index 20a791ef..725edc9d 100644
--- a/tarantool/msgpack_ext/interval.py
+++ b/tarantool/msgpack_ext/interval.py
@@ -1,12 +1,50 @@
 """
 Tarantool `datetime.interval`_ extension type support module.
 
-Refer to :mod:`~tarantool.msgpack_ext.types.interval`.
+The interval MessagePack representation looks like this:
+
+.. code-block:: text
+
+    +--------+-------------------------+-------------+----------------+
+    | MP_EXT | Size of packed interval | MP_INTERVAL | PackedInterval |
+    +--------+-------------------------+-------------+----------------+
+
+Packed interval consists of:
+
+* Packed number of non-zero fields.
+* Packed non-null fields.
+
+Each packed field has the following structure:
+
+.. code-block:: text
+
+    +----------+=====================+
+    | field ID |     field value     |
+    +----------+=====================+
+
+The number of defined (non-null) fields can be zero. In this case,
+the packed interval will be encoded as integer 0.
+
+List of the field IDs:
+
+* 0 – year
+* 1 – month
+* 2 – week
+* 3 – day
+* 4 – hour
+* 5 – minute
+* 6 – second
+* 7 – nanosecond
+* 8 – adjust
 
 .. _datetime.interval: https://www.tarantool.io/en/doc/latest/dev_guide/internals/msgpack_extensions/#the-interval-type
 """
 
-from tarantool.msgpack_ext.types.interval import Interval
+import msgpack
+
+from tarantool.error import MsgpackError
+
+from tarantool.msgpack_ext.types.interval import Interval, Adjust, id_map
 
 EXT_ID = 6
 """
@@ -22,11 +60,25 @@ def encode(obj):
 
     :return: Encoded interval.
     :rtype: :obj:`bytes`
-
-    :raise: :exc:`tarantool.Interval.msgpack_encode` exceptions
     """
 
-    return obj.msgpack_encode()
+    buf = bytes()
+
+    count = 0
+    for field_id in id_map.keys():
+        field_name = id_map[field_id]
+        value = getattr(obj, field_name)
+
+        if field_name == 'adjust':
+            value = value.value
+
+        if value != 0:
+            buf = buf + msgpack.packb(field_id) + msgpack.packb(value)
+            count = count + 1
+
+    buf = msgpack.packb(count) + buf
+
+    return buf
 
 def decode(data):
     """
@@ -38,7 +90,44 @@ def decode(data):
     :return: Decoded interval.
     :rtype: :class:`tarantool.Interval`
 
-    :raise: :exc:`tarantool.Interval` exceptions
+    :raise: :exc:`MsgpackError`
     """
 
-    return Interval(data)
+    # If MessagePack data does not contain a field value, it is zero.
+    # If built not from MessagePack data, set argument values later.
+    kwargs = {
+        'year': 0,
+        'month': 0,
+        'week': 0,
+        'day': 0,
+        'hour': 0,
+        'minute': 0,
+        'sec': 0,
+        'nsec': 0,
+        'adjust': Adjust(0),
+    }
+
+    if len(data) != 0:
+        # To create an unpacker is the only way to parse
+        # a sequence of values in Python msgpack module.
+        unpacker = msgpack.Unpacker()
+        unpacker.feed(data)
+        field_count = unpacker.unpack()
+        for _ in range(field_count):
+            field_id = unpacker.unpack()
+            value = unpacker.unpack()
+
+            if field_id not in id_map:
+                raise MsgpackError(f'Unknown interval field id {field_id}')
+
+            field_name = id_map[field_id]
+
+            if field_name == 'adjust':
+                try:
+                    value = Adjust(value)
+                except ValueError as e:
+                    raise MsgpackError(e)
+
+            kwargs[id_map[field_id]] = value
+
+    return Interval(**kwargs)
diff --git a/tarantool/msgpack_ext/types/interval.py b/tarantool/msgpack_ext/types/interval.py
index 62d98145..e90ff0a0 100644
--- a/tarantool/msgpack_ext/types/interval.py
+++ b/tarantool/msgpack_ext/types/interval.py
@@ -1,48 +1,9 @@
 """
-Tarantool `datetime.interval`_ extension type support module.
-
-The interval MessagePack representation looks like this:
-
-.. code-block:: text
-
-    +--------+-------------------------+-------------+----------------+
-    | MP_EXT | Size of packed interval | MP_INTERVAL | PackedInterval |
-    +--------+-------------------------+-------------+----------------+
-
-Packed interval consists of:
-
-* Packed number of non-zero fields.
-* Packed non-null fields.
-
-Each packed field has the following structure:
-
-.. code-block:: text
-
-    +----------+=====================+
-    | field ID |     field value     |
-    +----------+=====================+
-
-The number of defined (non-null) fields can be zero. In this case,
-the packed interval will be encoded as integer 0.
-
-List of the field IDs:
-
-* 0 – year
-* 1 – month
-* 2 – week
-* 3 – day
-* 4 – hour
-* 5 – minute
-* 6 – second
-* 7 – nanosecond
-* 8 – adjust
+Tarantool `datetime.interval`_ extension type implementation module.
 """
 
-import msgpack
 from enum import Enum
 
-from tarantool.error import MsgpackError
-
 id_map = {
     0: 'year',
     1: 'month',
@@ -97,14 +58,10 @@ class Interval():
     .. _datetime.interval: https://www.tarantool.io/en/doc/latest/dev_guide/internals/msgpack_extensions/#the-interval-type
     """
 
-    def __init__(self, data=None, *, year=0, month=0, week=0,
+    def __init__(self, *, year=0, month=0, week=0,
                  day=0, hour=0, minute=0, sec=0,
                  nsec=0, adjust=Adjust.NONE):
         """
-        :param data: MessagePack binary data to decode. If provided,
-            all other parameters are ignored.
-        :type data: :obj:`bytes`, optional
-
         :param year: Interval year value.
         :type year: :obj:`int`, optional
 
@@ -132,61 +89,17 @@ def __init__(self, data=None, *, year=0, month=0, week=0,
         :param adjust: Interval adjustment rule. Refer to
             :meth:`~tarantool.Datetime.__add__`.
         :type adjust: :class:`~tarantool.IntervalAdjust`, optional
-
-        :raise: :exc:`ValueError`
         """
-
-        # If MessagePack data does not contain a field value, it is zero.
-        # If built not from MessagePack data, set argument values later.
-        self.year = 0
-        self.month = 0
-        self.week = 0
-        self.day = 0
-        self.hour = 0
-        self.minute = 0
-        self.sec = 0
-        self.nsec = 0
-        self.adjust = Adjust(0)
-
-        if data is not None:
-            if not isinstance(data, bytes):
-                raise ValueError('data argument (first positional argument) ' +
-                                 'expected to be a "bytes" instance')
-
-            if len(data) == 0:
-                return
-
-            # To create an unpacker is the only way to parse
-            # a sequence of values in Python msgpack module.
-            unpacker = msgpack.Unpacker()
-            unpacker.feed(data)
-            field_count = unpacker.unpack()
-            for _ in range(field_count):
-                field_id = unpacker.unpack()
-                value = unpacker.unpack()
-
-                if field_id not in id_map:
-                    raise MsgpackError(f'Unknown interval field id {field_id}')
-
-                field_name = id_map[field_id]
-
-                if field_name == 'adjust':
-                    try:
-                        value = Adjust(value)
-                    except ValueError as e:
-                        raise MsgpackError(e)
-
-                setattr(self, id_map[field_id], value)
-        else:
-            self.year = year
-            self.month = month
-            self.week = week
-            self.day = day
-            self.hour = hour
-            self.minute = minute
-            self.sec = sec
-            self.nsec = nsec
-            self.adjust = adjust
+        
+        self.year = year
+        self.month = month
+        self.week = week
+        self.day = day
+        self.hour = hour
+        self.minute = minute
+        self.sec = sec
+        self.nsec = nsec
+        self.adjust = adjust
 
     def __add__(self, other):
         """
@@ -319,28 +232,3 @@ def __repr__(self):
                f'nsec={self.nsec}, adjust={self.adjust})'
 
     __str__ = __repr__
-
-    def msgpack_encode(self):
-        """
-        Encode an interval object.
-
-        :rtype: :obj:`bytes`
-        """
-
-        buf = bytes()
-
-        count = 0
-        for field_id in id_map.keys():
-            field_name = id_map[field_id]
-            value = getattr(self, field_name)
-
-            if field_name == 'adjust':
-                value = value.value
-
-            if value != 0:
-                buf = buf + msgpack.packb(field_id) + msgpack.packb(value)
-                count = count + 1
-
-        buf = msgpack.packb(count) + buf
-
-        return buf
diff --git a/test/suites/test_interval.py b/test/suites/test_interval.py
index 63a3c0ca..2de70a11 100644
--- a/test/suites/test_interval.py
+++ b/test/suites/test_interval.py
@@ -57,50 +57,11 @@ def setUp(self):
 
         self.adm("box.space['test']:truncate()")
 
-    def test_Interval_bytes_init(self):
-        dt = tarantool.Interval(b'\x02\x00\x01\x08\x01')
-
-        self.assertEqual(dt.year, 1)
-        self.assertEqual(dt.month, 0)
-        self.assertEqual(dt.day, 0)
-        self.assertEqual(dt.hour, 0)
-        self.assertEqual(dt.minute, 0)
-        self.assertEqual(dt.sec, 0)
-        self.assertEqual(dt.nsec, 0)
-        self.assertEqual(dt.adjust, tarantool.IntervalAdjust.NONE)
-
-    def test_Interval_non_bytes_positional_init(self):
+    def test_Interval_positional_init(self):
         self.assertRaisesRegex(
-            ValueError, re.escape('data argument (first positional argument) ' +
-                                  'expected to be a "bytes" instance'),
+            TypeError, re.escape('__init__() takes 1 positional argument but 2 were given'),
             lambda: tarantool.Interval(1))
 
-    def test_Interval_bytes_init_ignore_other_fields(self):
-        dt = tarantool.Interval(b'\x02\x00\x01\x08\x01',
-                                year=2, month=2, day=3, hour=1, minute=2,
-                                sec=3000, nsec=10000000,
-                                adjust=tarantool.IntervalAdjust.LAST)
-
-        self.assertEqual(dt.year, 1)
-        self.assertEqual(dt.month, 0)
-        self.assertEqual(dt.day, 0)
-        self.assertEqual(dt.hour, 0)
-        self.assertEqual(dt.minute, 0)
-        self.assertEqual(dt.sec, 0)
-        self.assertEqual(dt.nsec, 0)
-        self.assertEqual(dt.adjust, tarantool.IntervalAdjust.NONE)
-
-    def test_Interval_bytes_init_unknown_field(self):
-        self.assertRaisesRegex(
-            MsgpackError, 'Unknown interval field id 9',
-            lambda: tarantool.Interval(b'\x01\x09\xce\x00\x98\x96\x80'))
-
-    def test_Interval_bytes_init_unknown_adjust(self):
-        self.assertRaisesRegex(
-            MsgpackError, '3 is not a valid Adjust',
-            lambda: tarantool.Interval(b'\x02\x07\xce\x00\x98\x96\x80\x08\x03'))
-
-
     cases = {
         'year': {
             'python': tarantool.Interval(year=1),