BUG: astype to pyarrow does not copy np array

doc/source/whatsnew/v2.0.0.rst

@@ -1015,6 +1015,7 @@ Conversion
 - Bug in :class:`.arrays.ArrowExtensionArray` that would raise ``NotImplementedError`` when passed a sequence of strings or binary (:issue:`49172`)
 - Bug in :meth:`Series.astype` raising ``pyarrow.ArrowInvalid`` when converting from a non-pyarrow string dtype to a pyarrow numeric type (:issue:`50430`)
 - Bug in :meth:`Series.to_numpy` converting to NumPy array before applying ``na_value`` (:issue:`48951`)
+- Bug in :meth:`DataFrame.astype` not copying data when converting to pyarrow dtype (:issue:`50984`)
 - Bug in :func:`to_datetime` was not respecting ``exact`` argument when ``format`` was an ISO8601 format (:issue:`12649`)
 - Bug in :meth:`TimedeltaArray.astype` raising ``TypeError`` when converting to a pyarrow duration type (:issue:`49795`)
 -

pandas/core/arrays/arrow/array.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from copy import deepcopy
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -218,6 +219,9 @@ def _from_sequence(cls, scalars, *, dtype: Dtype | None = None, copy: bool = Fal
         if isinstance(scalars, cls):
             scalars = scalars._data
         elif not isinstance(scalars, (pa.Array, pa.ChunkedArray)):
+            if copy and is_array_like(scalars):
+                # pa array should not get updated when numpy array is updated
+                scalars = deepcopy(scalars)
             try:
                 scalars = pa.array(scalars, type=pa_dtype, from_pandas=True)
             except pa.ArrowInvalid:

pandas/tests/frame/methods/test_astype.py

@@ -3,6 +3,7 @@
 import numpy as np
 import pytest
 
+from pandas.compat import pa_version_under6p0
 import pandas.util._test_decorators as td
 
 import pandas as pd
@@ -867,3 +868,14 @@ def test_frame_astype_no_copy():
 
     assert result.a.dtype == pd.Int16Dtype()
     assert np.shares_memory(df.b.values, result.b.values)
+
+
+@pytest.mark.skipif(pa_version_under6p0, reason="pyarrow is required for this test")
+@pytest.mark.parametrize("dtype", ["int64", "Int64"])
+def test_astype_copies(dtype):
+    # GH#50984
+    df = DataFrame({"a": [1, 2, 3]}, dtype=dtype)
+    result = df.astype("int64[pyarrow]", copy=True)
+    df.iloc[0, 0] = 100
+    expected = DataFrame({"a": [1, 2, 3]}, dtype="int64[pyarrow]")
+    tm.assert_frame_equal(result, expected)