refactor(python): Parse fixed timezone offsets without pytz (#5769)

py-polars/polars/utils.py

@@ -5,7 +5,7 @@
 import os
 import sys
 import warnings
-from datetime import date, datetime, time, timedelta, timezone
+from datetime import date, datetime, time, timedelta, timezone, tzinfo
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, Callable, Iterable, Sequence, TypeVar, overload
 
@@ -259,38 +259,29 @@ def _to_python_datetime(
 # cache here as we have a single tz per column
 # and this function will be called on every conversion
 @functools.lru_cache(16)
-def _localize_offset(dt: datetime, offset: str) -> datetime:
+def _parse_fixed_tz_offset(offset: str) -> tzinfo:
     try:
-        import pytz
-    except ImportError:
-        raise ImportError("pytz needs to be installed to handle datetimes with offsets")
-    import re
-
-    if offset.startswith("-"):
-        g = re.search(r"-(\d\d):(\d\d)", offset)
-        if g is None:
-            raise ValueError(f"Offset: {offset} not understood.")
-        hours = -int(g.group(1))
-        minutes = -int(g.group(2))
-    elif offset.startswith("+"):
-        g = re.search(r"\+(\d\d):(\d\d)", offset)
-        if g is None:
-            raise ValueError(f"Offset: {offset} not understood.")
-        hours = int(g.group(1))
-        minutes = int(g.group(2))
-    else:
-        raise ValueError(f"Offset: {offset} not understood.")
+        # use fromisoformat to parse the offset
+        dt_offset = datetime.fromisoformat("2000-01-01T00:00:00" + offset)
 
-    tz = pytz.FixedOffset(hours * 60 + minutes)
-    return dt.astimezone(tz)
+        # alternatively, we parse the offset ourselves extracting hours and
+        # minutes, then we can construct:
+        # tzinfo=timezone(timedelta(hours=..., minutes=...))
+    except ValueError:
+        raise ValueError(f"Offset: {offset} not understood.") from None
+
+    return dt_offset.tzinfo  # type: ignore[return-value]
 
 
 def _localize(dt: datetime, tz: str) -> datetime:
     # zone info installation should already be checked
     try:
-        return dt.astimezone(zoneinfo.ZoneInfo(tz))
+        tzinfo = zoneinfo.ZoneInfo(tz)
     except zoneinfo.ZoneInfoNotFoundError:
-        return _localize_offset(dt, tz)
+        # try fixed offset, which is not supported by ZoneInfo
+        tzinfo = _parse_fixed_tz_offset(tz)  # type: ignore[assignment]
+
+    return dt.astimezone(tzinfo)
 
 
 def _in_notebook() -> bool:

py-polars/pyproject.toml

@@ -52,7 +52,6 @@ module = [
   "connectorx",
   "IPython.*",
   "zoneinfo",
-  "pytz",
 ]
 ignore_missing_imports = true
 

py-polars/requirements-dev.txt

@@ -8,7 +8,6 @@
 numpy
 pandas
 pyarrow
-pytz
 backports.zoneinfo; python_version < '3.9'
 tzdata; platform_system == 'Windows'
 xlsx2csv

py-polars/tests/unit/test_timezone.py

@@ -1,26 +1,30 @@
-from datetime import datetime
+from datetime import datetime, timedelta, timezone
 
-import pytz
+import pytest
 
 import polars as pl
 
 
-def test_timezone_aware_strptime() -> None:
+@pytest.mark.parametrize(
+    "tz_string,timedelta",
+    [("+01:00", timedelta(minutes=60)), ("-01:30", timedelta(hours=-1, minutes=-30))],
+)
+def test_timezone_aware_strptime(tz_string: str, timedelta: timedelta) -> None:
     times = pl.DataFrame(
         {
             "delivery_datetime": [
-                "2021-12-05 06:00:00+01:00",
-                "2021-12-05 07:00:00+01:00",
-                "2021-12-05 08:00:00+01:00",
+                "2021-12-05 06:00:00" + tz_string,
+                "2021-12-05 07:00:00" + tz_string,
+                "2021-12-05 08:00:00" + tz_string,
             ]
         }
     )
     assert times.with_column(
         pl.col("delivery_datetime").str.strptime(pl.Datetime, fmt="%Y-%m-%d %H:%M:%S%z")
     ).to_dict(False) == {
         "delivery_datetime": [
-            datetime(2021, 12, 5, 6, 0, tzinfo=pytz.FixedOffset(60)),
-            datetime(2021, 12, 5, 7, 0, tzinfo=pytz.FixedOffset(60)),
-            datetime(2021, 12, 5, 8, 0, tzinfo=pytz.FixedOffset(60)),
+            datetime(2021, 12, 5, 6, 0, tzinfo=timezone(timedelta)),
+            datetime(2021, 12, 5, 7, 0, tzinfo=timezone(timedelta)),
+            datetime(2021, 12, 5, 8, 0, tzinfo=timezone(timedelta)),
         ]
     }