msgpack: support tzoffset in datetime

Support non-zero tzoffset in datetime extended type. If tzoffset and
tzindex are not specified, return object with timezone-naive
pandas.Timestamp internals. If tzoffset is specified, return object with
timezone-aware pandas.Timestamp with pytz.FixedOffset [1] timezone info.
pytz module is already a dependency of pandas, but this patch adds it as
a requirement just in case something will change in the future.

1. https://pypi.org/project/pytz/

Part of #204

Author: DifferentialOrange

CHANGELOG.md

@@ -31,6 +31,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   convert to a `pandas.Timestamp` and then use `to_datetime64()`
   or `to_datetime()` converter.
 
+- Offset in datetime type support (#204).
+
 ### Changed
 - Bump msgpack requirement to 1.0.4 (PR #223).
   The only reason of this bump is various vulnerability fixes,

requirements.txt

@@ -1,2 +1,3 @@
 msgpack>=1.0.4
 pandas
+pytz

tarantool/msgpack_ext/types/datetime.py

@@ -1,6 +1,7 @@
 from copy import deepcopy
 
 import pandas
+import pytz
 
 # https://www.tarantool.io/en/doc/latest/dev_guide/internals/msgpack_extensions/#the-datetime-type
 #
@@ -41,6 +42,8 @@
 BYTEORDER = 'little'
 
 NSEC_IN_SEC = 1000000000
+SEC_IN_MIN = 60
+MIN_IN_DAY = 60 * 24
 
 
 def get_bytes_as_int(data, cursor, size):
@@ -50,6 +53,14 @@ def get_bytes_as_int(data, cursor, size):
 def get_int_as_bytes(data, size):
     return data.to_bytes(size, byteorder=BYTEORDER, signed=True)
 
+def compute_offset(timestamp):
+    if timestamp.tz is None:
+        return 0
+
+    utc_offset = timestamp.tz.utcoffset(timestamp)
+    # There is no precision loss since offset is in minutes
+    return utc_offset.days * MIN_IN_DAY + utc_offset.seconds // SEC_IN_MIN
+
 def msgpack_decode(data):
     cursor = 0
     seconds, cursor = get_bytes_as_int(data, cursor, SECONDS_SIZE_BYTES)
@@ -63,12 +74,17 @@ def msgpack_decode(data):
         tzoffset = 0
         tzindex = 0
 
-    if (tzoffset != 0) or (tzindex != 0):
-        raise NotImplementedError
-
     total_nsec = seconds * NSEC_IN_SEC + nsec
 
-    timestamp = pandas.to_datetime(total_nsec, unit='ns')
+    if (tzindex != 0):
+        raise NotImplementedError
+    elif (tzoffset != 0):
+        tzinfo = pytz.FixedOffset(tzoffset)
+        timestamp = pandas.to_datetime(total_nsec, unit='ns').replace(tzinfo=pytz.utc).tz_convert(tzinfo)
+    else:
+        # return timezone-naive pandas.Timestamp
+        timestamp = pandas.to_datetime(total_nsec, unit='ns')
+
     return timestamp, tzoffset, tzindex
 
 class Datetime():
@@ -79,12 +95,16 @@ def __init__(self, *args, **kwargs):
                 timestamp, tzoffset, tzindex = msgpack_decode(data)
             elif isinstance(data, pandas.Timestamp):
                 timestamp = deepcopy(data)
+                tzoffset = compute_offset(timestamp)
             elif isinstance(data, Datetime):
                 timestamp = deepcopy(data._timestamp)
+                tzoffset = deepcopy(data._tzoffset)
         else:
             timestamp = pandas.Timestamp(*args, **kwargs)
+            tzoffset = compute_offset(timestamp)
 
         self._timestamp = timestamp
+        self._tzoffset = tzoffset
 
     def __eq__(self, other):
         if isinstance(other, Datetime):
@@ -93,11 +113,12 @@ def __eq__(self, other):
             return self._timestamp == other
         else:
             return False
+
     def __str__(self):
-        return f'tarantool.Datetime(timestamp={self._timestamp})'
+        return f'tarantool.Datetime(timestamp={self._timestamp}, tzoffset={self._tzoffset})'
 
     def __repr__(self):
-        return f'tarantool.Datetime(timestamp={self._timestamp})'
+        return f'tarantool.Datetime(timestamp={self._timestamp}, tzoffset={self._tzoffset})'
 
     def to_pd_timestamp(self):
         return deepcopy(self._timestamp)
@@ -107,7 +128,7 @@ def msgpack_encode(self):
 
         seconds = ts_value // NSEC_IN_SEC
         nsec = ts_value % NSEC_IN_SEC
-        tzoffset = 0
+        tzoffset = self._tzoffset
         tzindex = 0
 
         buf = get_int_as_bytes(seconds, SECONDS_SIZE_BYTES)

test/suites/test_datetime.py

@@ -8,6 +8,7 @@
 import warnings
 import tarantool
 import pandas
+import pytz
 
 from tarantool.msgpack_ext.packer import default as packer_default
 from tarantool.msgpack_ext.unpacker import ext_hook as unpacker_ext_hook
@@ -97,6 +98,38 @@ def setUp(self):
             'tarantool': r"datetime.new({year=2022, month=8, day=31, hour=18, min=7, sec=54, " +
                          r"nsec=308543321})",
         },
+        'datetime_with_positive_offset': {
+            'python': tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, second=54,
+                                         microsecond=308543, nanosecond=321,
+                                         tzinfo=pytz.FixedOffset(180)),
+            'msgpack': (b'\x4a\x79\x0f\x63\x00\x00\x00\x00\x59\xff\x63\x12\xb4\x00\x00\x00'),
+            'tarantool': r"datetime.new({year=2022, month=8, day=31, hour=18, min=7, sec=54, " +
+                         r"nsec=308543321, tzoffset=180})",
+        },
+        'datetime_with_negative_offset': {
+            'python': tarantool.Datetime(year=2022, month=8, day=31, hour=18, minute=7, second=54,
+                                         microsecond=308543, nanosecond=321,
+                                         tzinfo=pytz.FixedOffset(-60)),
+            'msgpack': (b'\x8a\xb1\x0f\x63\x00\x00\x00\x00\x59\xff\x63\x12\xc4\xff\x00\x00'),
+            'tarantool': r"datetime.new({year=2022, month=8, day=31, hour=18, min=7, sec=54, " +
+                         r"nsec=308543321, tzoffset=-60})",
+        },
+        'pandas_timestamp_with_positive_offset': {
+            'python': pandas.Timestamp(year=2022, month=8, day=31, hour=18, minute=7, second=54,
+                                       microsecond=308543, nanosecond=321,
+                                       tzinfo=pytz.FixedOffset(180)),
+            'msgpack': (b'\x4a\x79\x0f\x63\x00\x00\x00\x00\x59\xff\x63\x12\xb4\x00\x00\x00'),
+            'tarantool': r"datetime.new({year=2022, month=8, day=31, hour=18, min=7, sec=54, " +
+                         r"nsec=308543321, tzoffset=180})",
+        },
+        'pandas_timestamp_with_negative_offset': {
+            'python': pandas.Timestamp(year=2022, month=8, day=31, hour=18, minute=7, second=54,
+                                       microsecond=308543, nanosecond=321,
+                                       tzinfo=pytz.FixedOffset(-60)),
+            'msgpack': (b'\x8a\xb1\x0f\x63\x00\x00\x00\x00\x59\xff\x63\x12\xc4\xff\x00\x00'),
+            'tarantool': r"datetime.new({year=2022, month=8, day=31, hour=18, min=7, sec=54, " +
+                         r"nsec=308543321, tzoffset=-60})",
+        },
     }
 
     def test_msgpack_decode(self):