fix(rust, python): allow +00:00 when loading from arrow (pola-rs#9747)

potzenhotz · Jul 11, 2023 · 5b92319 · 5b92319
1 parent dbe7b51
commit 5b92319
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 2 deletions.
diff --git a/polars/polars-core/src/series/from.rs b/polars/polars-core/src/series/from.rs
@@ -200,8 +200,9 @@ impl Series {
                 let mut tz = tz.clone();
                 if tz.as_deref() == Some("") {
                     tz = None;
-                }
-                if let Some(_tz) = &tz {
+                } else if tz.as_deref() == Some("+00:00") {
+                    tz = Some("UTC".to_string());
+                } else if let Some(_tz) = &tz {
                     #[cfg(feature = "timezones")]
                     validate_time_zone(_tz)?;
                 }

diff --git a/py-polars/tests/unit/io/files/tz_aware.parquet b/py-polars/tests/unit/io/files/tz_aware.parquet
diff --git a/py-polars/tests/unit/io/test_parquet.py b/py-polars/tests/unit/io/test_parquet.py
@@ -1,6 +1,8 @@
 from __future__ import annotations
 
 import io
+import os
+from datetime import datetime, timezone
 from typing import TYPE_CHECKING
 
 import numpy as np
@@ -507,3 +509,13 @@ def test_parquet_string_cache() -> None:
     # so polars should automatically set string cache
     f.seek(0)
     assert_series_equal(pl.read_parquet(f)["a"].cast(str), df["a"].cast(str))
+
+
+def test_tz_aware_parquet_9586() -> None:
+    result = pl.read_parquet(
+        os.path.join("tests", "unit", "io", "files", "tz_aware.parquet")
+    )
+    expected = pl.DataFrame(
+        {"UTC_DATETIME_ID": [datetime(2023, 6, 26, 14, 15, 0, tzinfo=timezone.utc)]}
+    ).select(pl.col("*").cast(pl.Datetime("ns", "UTC")))
+    assert_frame_equal(result, expected)