correct for timezone information in python datetimes

py-polars/docs/source/reference/dataframe.rst

@@ -60,6 +60,7 @@ Descriptive stats
     DataFrame.is_unique
     DataFrame.n_chunks
     DataFrame.null_count
+    DataFrame.is_empty
 
 Computations
 ------------

py-polars/polars/eager/frame.py

@@ -3016,6 +3016,12 @@ def interpolate(self) -> "DataFrame":
         """
         return self.select(pl.col("*").interpolate())  # type: ignore
 
+    def is_empty(self) -> bool:
+        """
+        Check if the dataframe is empty
+        """
+        return self.height == 0
+
 
 class GroupBy:
     """

py-polars/polars/lazy/functions.py

@@ -1,5 +1,5 @@
 import typing as tp
-from datetime import datetime
+from datetime import datetime, timezone
 from typing import Any, Callable, Optional, Type, Union
 
 import numpy as np
@@ -354,7 +354,7 @@ def lit_date(dt: datetime) -> "pl.Expr":
     dt
         datetime.datetime
     """
-    return lit(int(dt.timestamp() * 1e3))
+    return lit(int((dt.replace(tzinfo=timezone.utc)).timestamp() * 1e3))
 
 
 def lit(

py-polars/tests/test_df.py

@@ -1022,3 +1022,13 @@ def test_h_agg():
     assert df.sum(axis=1, null_strategy="ignore").to_list() == [2, 2, 6]
     assert df.sum(axis=1, null_strategy="propagate").to_list() == [2, None, 6]
     assert df.mean(axis=1, null_strategy="propagate")[1] is None
+
+
+def test_filter_date():
+    dataset = pl.DataFrame(
+        {"date": ["2020-01-02", "2020-01-03", "2020-01-04"], "index": [1, 2, 3]}
+    )
+    df = dataset.with_column(pl.col("date").str.strptime(pl.Date32, "%Y-%m-%d"))
+
+    # filter out the data to match only records from the previous year
+    assert df.filter(col("date") <= pl.lit_date(datetime(2019, 1, 3))).is_empty()