From 4c5060738e132684d1def6152a151d3c112bf062 Mon Sep 17 00:00:00 2001
From: Patrick Hoefler <61934744+phofl@users.noreply.github.com>
Date: Fri, 9 Feb 2024 23:04:56 +0100
Subject: [PATCH] Backport PR #57121: REGR: Fix to_numpy for masked array with
 non-numeric dtype

---
 doc/source/whatsnew/v2.2.1.rst              |  3 +++
 pandas/core/arrays/masked.py                |  2 ++
 pandas/tests/arrays/masked/test_function.py | 17 +++++++++++++++++
 3 files changed, 22 insertions(+)

diff --git a/doc/source/whatsnew/v2.2.1.rst b/doc/source/whatsnew/v2.2.1.rst
index 883627bd4b19b..009d4794dbd1d 100644
--- a/doc/source/whatsnew/v2.2.1.rst
+++ b/doc/source/whatsnew/v2.2.1.rst
@@ -17,12 +17,15 @@ Fixed regressions
 - Fixed performance regression in :meth:`Series.combine_first` (:issue:`55845`)
 - Fixed regression in :func:`merge_ordered` raising ``TypeError`` for ``fill_method="ffill"`` and ``how="left"`` (:issue:`57010`)
 - Fixed regression in :func:`wide_to_long` raising an ``AttributeError`` for string columns (:issue:`57066`)
+- Fixed regression in :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmin`, :meth:`.SeriesGroupBy.idxmax` ignoring the ``skipna`` argument (:issue:`57040`)
+- Fixed regression in :meth:`.DataFrameGroupBy.idxmin`, :meth:`.DataFrameGroupBy.idxmax`, :meth:`.SeriesGroupBy.idxmin`, :meth:`.SeriesGroupBy.idxmax` where values containing the minimum or maximum value for the dtype could produce incorrect results (:issue:`57040`)
 - Fixed regression in :meth:`DataFrame.loc` raising ``IndexError`` for non-unique, masked dtype indexes where result has more than 10,000 rows (:issue:`57027`)
 - Fixed regression in :meth:`DataFrame.sort_index` not producing a stable sort for a index with duplicates (:issue:`57151`)
 - Fixed regression in :meth:`DataFrame.to_dict` with ``orient='list'`` and datetime or timedelta types returning integers (:issue:`54824`)
 - Fixed regression in :meth:`DataFrame.to_json` converting nullable integers to floats (:issue:`57224`)
 - Fixed regression in :meth:`DataFrameGroupBy.idxmin`, :meth:`DataFrameGroupBy.idxmax`, :meth:`SeriesGroupBy.idxmin`, :meth:`SeriesGroupBy.idxmax` ignoring the ``skipna`` argument (:issue:`57040`)
 - Fixed regression in :meth:`DataFrameGroupBy.idxmin`, :meth:`DataFrameGroupBy.idxmax`, :meth:`SeriesGroupBy.idxmin`, :meth:`SeriesGroupBy.idxmax` where values containing the minimum or maximum value for the dtype could produce incorrect results (:issue:`57040`)
+- Fixed regression in :meth:`ExtensionArray.to_numpy` raising for non-numeric masked dtypes (:issue:`56991`)
 - Fixed regression in :meth:`Index.join` raising ``TypeError`` when joining an empty index to a non-empty index containing mixed dtype values (:issue:`57048`)
 - Fixed regression in :meth:`Series.pct_change` raising a ``ValueError`` for an empty :class:`Series` (:issue:`57056`)
 - Fixed regression in :meth:`Series.to_numpy` when dtype is given as float and the data contains NaNs (:issue:`57121`)
diff --git a/pandas/core/arrays/masked.py b/pandas/core/arrays/masked.py
index 8c41de9c9fffa..7c49ce5343158 100644
--- a/pandas/core/arrays/masked.py
+++ b/pandas/core/arrays/masked.py
@@ -501,6 +501,8 @@ def to_numpy(
         """
         hasna = self._hasna
         dtype, na_value = to_numpy_dtype_inference(self, dtype, na_value, hasna)
+        if dtype is None:
+            dtype = object
 
         if hasna:
             if (
diff --git a/pandas/tests/arrays/masked/test_function.py b/pandas/tests/arrays/masked/test_function.py
index 4c7bd6e293ef4..b259018cd6121 100644
--- a/pandas/tests/arrays/masked/test_function.py
+++ b/pandas/tests/arrays/masked/test_function.py
@@ -5,6 +5,7 @@
 
 import pandas as pd
 import pandas._testing as tm
+from pandas.core.arrays import BaseMaskedArray
 
 arrays = [pd.array([1, 2, 3, None], dtype=dtype) for dtype in tm.ALL_INT_EA_DTYPES]
 arrays += [
@@ -55,3 +56,19 @@ def test_tolist(data):
     result = data.tolist()
     expected = list(data)
     tm.assert_equal(result, expected)
+
+
+def test_to_numpy():
+    # GH#56991
+
+    class MyStringArray(BaseMaskedArray):
+        dtype = pd.StringDtype()
+        _dtype_cls = pd.StringDtype
+        _internal_fill_value = pd.NA
+
+    arr = MyStringArray(
+        values=np.array(["a", "b", "c"]), mask=np.array([False, True, False])
+    )
+    result = arr.to_numpy()
+    expected = np.array(["a", pd.NA, "c"])
+    tm.assert_numpy_array_equal(result, expected)