REGR: fix period dtype <-> arrow roundtrip for pyarrow < 4 (pandas-dev#45524)

Author: jorisvandenbossche

pandas/core/internals/api.py

@@ -24,6 +24,7 @@
 from pandas.core.internals.blocks import (
     Block,
     DatetimeTZBlock,
+    ExtensionBlock,
     check_ndim,
     ensure_block_shape,
     extract_pandas_array,
@@ -51,6 +52,12 @@ def make_block(
 
     values, dtype = extract_pandas_array(values, dtype, ndim)
 
+    if klass is ExtensionBlock and is_period_dtype(values.dtype):
+        # GH-44681 changed PeriodArray to be stored in the 2D
+        # NDArrayBackedExtensionBlock instead of ExtensionBlock
+        # -> still allow ExtensionBlock to be passed in this case for back compat
+        klass = None
+
     if klass is None:
         dtype = dtype or values.dtype
         klass = get_block_type(dtype)

pandas/tests/arrays/period/test_arrow_compat.py

@@ -1,7 +1,5 @@
 import pytest
 
-from pandas.compat import pa_version_under4p0
-
 from pandas.core.dtypes.dtypes import PeriodDtype
 
 import pandas as pd
@@ -71,9 +69,6 @@ def test_arrow_array_missing():
     assert result.storage.equals(expected)
 
 
-@pytest.mark.xfail(
-    pa_version_under4p0, reason="pyarrow incorrectly uses pandas internals API"
-)
 def test_arrow_table_roundtrip():
     from pandas.core.arrays._arrow_utils import ArrowPeriodType
 
@@ -93,9 +88,6 @@ def test_arrow_table_roundtrip():
     tm.assert_frame_equal(result, expected)
 
 
-@pytest.mark.xfail(
-    pa_version_under4p0, reason="pyarrow incorrectly uses pandas internals API"
-)
 def test_arrow_load_from_zero_chunks():
     # GH-41040
 
@@ -114,9 +106,6 @@ def test_arrow_load_from_zero_chunks():
     tm.assert_frame_equal(result, df)
 
 
-@pytest.mark.xfail(
-    pa_version_under4p0, reason="pyarrow incorrectly uses pandas internals API"
-)
 def test_arrow_table_roundtrip_without_metadata():
     arr = PeriodArray([1, 2, 3], freq="H")
     arr[1] = pd.NaT

pandas/tests/io/test_feather.py

@@ -2,8 +2,6 @@
 import numpy as np
 import pytest
 
-from pandas.compat.pyarrow import pa_version_under4p0
-
 import pandas as pd
 import pandas._testing as tm
 
@@ -87,11 +85,7 @@ def test_basic(self):
                 ),
             }
         )
-        if not pa_version_under4p0:
-            # older pyarrow incorrectly uses pandas internal API, so
-            #  constructs invalid Block
-            df["periods"] = pd.period_range("2013", freq="M", periods=3)
-
+        df["periods"] = pd.period_range("2013", freq="M", periods=3)
         df["timedeltas"] = pd.timedelta_range("1 day", periods=3)
         df["intervals"] = pd.interval_range(0, 3, 3)
 

pandas/tests/io/test_parquet.py

@@ -15,7 +15,6 @@
 
 from pandas.compat.pyarrow import (
     pa_version_under2p0,
-    pa_version_under4p0,
     pa_version_under5p0,
     pa_version_under6p0,
 )
@@ -652,13 +651,7 @@ def test_use_nullable_dtypes(self, engine, request):
             "object",
             "datetime64[ns, UTC]",
             "float",
-            pytest.param(
-                "period[D]",
-                marks=pytest.mark.xfail(
-                    pa_version_under4p0,
-                    reason="pyarrow uses pandas internal API incorrectly",
-                ),
-            ),
+            "period[D]",
             "Float64",
             "string",
         ],
@@ -897,9 +890,6 @@ def test_pyarrow_backed_string_array(self, pa, string_storage):
             check_round_trip(df, pa, expected=df.astype(f"string[{string_storage}]"))
 
     @td.skip_if_no("pyarrow")
-    @pytest.mark.xfail(
-        pa_version_under4p0, reason="pyarrow uses pandas internal API incorrectly"
-    )
     def test_additional_extension_types(self, pa):
         # test additional ExtensionArrays that are supported through the
         # __arrow_array__ protocol + by defining a custom ExtensionType