Backport PR pandas-dev#54768 on branch 2.1.x (Fix roundtripping with pyarrow schema) (pandas-dev#54773)

Backport PR pandas-dev#54768: Fix roundtripping with pyarrow schema

Co-authored-by: Patrick Hoefler <[email protected]>

Author: meeseeksmachine

pandas/core/arrays/string_.py

@@ -14,6 +14,7 @@
     missing as libmissing,
 )
 from pandas._libs.arrays import NDArrayBacked
+from pandas._libs.lib import ensure_string_array
 from pandas.compat import pa_version_under7p0
 from pandas.compat.numpy import function as nv
 from pandas.util._decorators import doc
@@ -221,7 +222,7 @@ def __from_arrow__(
             arr = np.array([], dtype=object)
         else:
             arr = pyarrow.concat_arrays(chunks).to_numpy(zero_copy_only=False)
-            arr = lib.convert_nans_to_NA(arr)
+            arr = ensure_string_array(arr, na_value=libmissing.NA)
         # Bypass validation inside StringArray constructor, see GH#47781
         new_string_array = StringArray.__new__(StringArray)
         NDArrayBacked.__init__(

pandas/tests/io/test_parquet.py

@@ -1,5 +1,6 @@
 """ test parquet compat """
 import datetime
+from decimal import Decimal
 from io import BytesIO
 import os
 import pathlib
@@ -16,6 +17,7 @@
 from pandas.compat.pyarrow import (
     pa_version_under7p0,
     pa_version_under8p0,
+    pa_version_under11p0,
     pa_version_under13p0,
 )
 
@@ -1111,6 +1113,18 @@ def test_string_inference(self, tmp_path, pa):
         )
         tm.assert_frame_equal(result, expected)
 
+    @pytest.mark.skipif(pa_version_under11p0, reason="not supported before 11.0")
+    def test_roundtrip_decimal(self, tmp_path, pa):
+        # GH#54768
+        import pyarrow as pa
+
+        path = tmp_path / "decimal.p"
+        df = pd.DataFrame({"a": [Decimal("123.00")]}, dtype="string[pyarrow]")
+        df.to_parquet(path, schema=pa.schema([("a", pa.decimal128(5))]))
+        result = read_parquet(path)
+        expected = pd.DataFrame({"a": ["123"]}, dtype="string[python]")
+        tm.assert_frame_equal(result, expected)
+
 
 class TestParquetFastParquet(Base):
     def test_basic(self, fp, df_full):