fix!: to_gbq uploads ArrowDtype(pa.timestamp(...) without timezon…

…e as `DATETIME` type (#832)

* fix!: `to_gbq` uploads `ArrowDtype(pa.timestamp(...)` without timezone as `DATETIME` type

Release-As: 0.25.0

* 🦉 Updates from OwlBot post-processor

See https://github.com/googleapis/repo-automation-bots/blob/main/packages/owl-bot/README.md

---------

Co-authored-by: Owl Bot <gcf-owl-bot[bot]@users.noreply.github.com>

pandas_gbq/schema/pyarrow_to_bigquery.py

@@ -38,6 +38,15 @@
 
 
 def arrow_type_to_bigquery_field(name, type_) -> Optional[schema.SchemaField]:
+    # Since both TIMESTAMP/DATETIME use pyarrow.timestamp(...), we need to use
+    # a special case to disambiguate them. See:
+    # https://github.com/googleapis/python-bigquery-pandas/issues/450
+    if pyarrow.types.is_timestamp(type_):
+        if type_.tz is None:
+            return schema.SchemaField(name, "DATETIME")
+        else:
+            return schema.SchemaField(name, "TIMESTAMP")
+
     detected_type = _ARROW_SCALAR_IDS_TO_BQ.get(type_.id, None)
     if detected_type is not None:
         return schema.SchemaField(name, detected_type)

tests/unit/schema/test_pyarrow_to_bigquery.py

@@ -2,21 +2,53 @@
 # Use of this source code is governed by a BSD-style
 # license that can be found in the LICENSE file.
 
+from google.cloud import bigquery
 import pyarrow
+import pytest
 
 from pandas_gbq.schema import pyarrow_to_bigquery
 
 
+@pytest.mark.parametrize(
+    (
+        "pyarrow_type",
+        "bigquery_type",
+    ),
+    (
+        # All integer types should map to BigQuery INT64 (or INTEGER since
+        # SchemaField uses the legacy SQL names). See:
+        # https://github.com/googleapis/python-bigquery-pandas/issues/616
+        (pyarrow.int8(), "INTEGER"),
+        (pyarrow.int16(), "INTEGER"),
+        (pyarrow.int32(), "INTEGER"),
+        (pyarrow.int64(), "INTEGER"),
+        (pyarrow.uint8(), "INTEGER"),
+        (pyarrow.uint16(), "INTEGER"),
+        (pyarrow.uint32(), "INTEGER"),
+        (pyarrow.uint64(), "INTEGER"),
+        # If there is no associated timezone, assume a naive (timezone-less)
+        # DATETIME. See:
+        # https://github.com/googleapis/python-bigquery-pandas/issues/450
+        (pyarrow.timestamp("ns"), "DATETIME"),
+        (pyarrow.timestamp("ns", tz="UTC"), "TIMESTAMP"),
+    ),
+)
+def test_arrow_type_to_bigquery_field_scalar_types(pyarrow_type, bigquery_type):
+    field: bigquery.SchemaField = pyarrow_to_bigquery.arrow_type_to_bigquery_field(
+        "test_name", pyarrow_type
+    )
+    assert field.name == "test_name"
+    assert field.field_type == bigquery_type
+
+
 def test_arrow_type_to_bigquery_field_unknown():
-    # Default types should be picked at a higher layer.
     assert (
         pyarrow_to_bigquery.arrow_type_to_bigquery_field("test_name", pyarrow.null())
         is None
     )
 
 
 def test_arrow_type_to_bigquery_field_list_of_unknown():
-    # Default types should be picked at a higher layer.
     assert (
         pyarrow_to_bigquery.arrow_type_to_bigquery_field(
             "test_name", pyarrow.list_(pyarrow.null())