BUG: iter with readonly values, closes #28055 (#28074)

* BUG: iter with readonly values, closes #28055

* whatsnew

Author: jbrockmendel

doc/source/whatsnew/v0.25.1.rst

@@ -29,6 +29,7 @@ Datetimelike
 
 - Bug in :func:`to_datetime` where passing a timezone-naive :class:`DatetimeArray` or :class:`DatetimeIndex` and ``utc=True`` would incorrectly return a timezone-naive result (:issue:`27733`)
 - Bug in :meth:`Period.to_timestamp` where a :class:`Period` outside the :class:`Timestamp` implementation bounds (roughly 1677-09-21 to 2262-04-11) would return an incorrect :class:`Timestamp` instead of raising ``OutOfBoundsDatetime`` (:issue:`19643`)
+- Bug in iterating over :class:`DatetimeIndex` when the underlying data is read-only (:issue:`28055`)
 
 Timezones
 ^^^^^^^^^

pandas/_libs/tslib.pyx

@@ -71,7 +71,7 @@ cdef inline object create_time_from_ts(
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def ints_to_pydatetime(int64_t[:] arr, object tz=None, object freq=None,
+def ints_to_pydatetime(const int64_t[:] arr, object tz=None, object freq=None,
                        str box="datetime"):
     """
     Convert an i8 repr to an ndarray of datetimes, date, time or Timestamp

pandas/tests/indexes/datetimes/test_misc.py

@@ -377,3 +377,11 @@ def test_nanosecond_field(self):
         dti = DatetimeIndex(np.arange(10))
 
         tm.assert_index_equal(dti.nanosecond, pd.Index(np.arange(10, dtype=np.int64)))
+
+
+def test_iter_readonly():
+    # GH#28055 ints_to_pydatetime with readonly array
+    arr = np.array([np.datetime64("2012-02-15T12:00:00.000000000")])
+    arr.setflags(write=False)
+    dti = pd.to_datetime(arr)
+    list(dti)