Merge pull request #21925 from seberg/subarray-object-cast

BUG: Fix subarray to object cast ownership details

3 files changed, 34 insertions, 1 deletions

diff --git a/numpy/core/src/multiarray/arraytypes.c.src b/numpy/core/src/multiarray/arraytypes.c.src
index 7cd80ba9a..56ac83cbb 100644
--- a/numpy/core/src/multiarray/arraytypes.c.src
+++ b/numpy/core/src/multiarray/arraytypes.c.src
@@ -804,7 +804,7 @@ VOID_getitem(void *input, void *vap)
          *       could have special handling.
          */
         PyObject *base = (PyObject *)ap;
-        while (Py_TYPE(base) == NULL) {
+        while (base != NULL && Py_TYPE(base) == NULL) {
             base = PyArray_BASE((PyArrayObject *)base);
         }
         ret = (PyArrayObject *)PyArray_NewFromDescrAndBase(
@@ -812,6 +812,15 @@ VOID_getitem(void *input, void *vap)
                 shape.len, shape.ptr, NULL, ip,
                 PyArray_FLAGS(ap) & ~NPY_ARRAY_F_CONTIGUOUS,
                 NULL, base);
+        if (base == NULL) {
+            /*
+             * Need to create a copy, or we may point to wrong data.  This path
+             * is taken when no "valid" array is passed.  This happens for
+             * casts.
+             */
+            PyObject *copy = PyArray_FromArray(ret, NULL, NPY_ARRAY_ENSURECOPY);
+            Py_SETREF(ret, (PyArrayObject *)copy);
+        }
         npy_free_cache_dim_obj(shape);
         return (PyObject *)ret;
     }
diff --git a/numpy/core/src/multiarray/dtype_transfer.c b/numpy/core/src/multiarray/dtype_transfer.c
index f8458d2d7..c588494e7 100644
--- a/numpy/core/src/multiarray/dtype_transfer.c
+++ b/numpy/core/src/multiarray/dtype_transfer.c
@@ -258,6 +258,7 @@ any_to_object_get_loop(
     data->base.free = &_any_to_object_auxdata_free;
     data->base.clone = &_any_to_object_auxdata_clone;
     data->arr_fields.base = NULL;
+    Py_SET_TYPE(&data->arr_fields, NULL);
     data->arr_fields.descr = context->descriptors[0];
     Py_INCREF(data->arr_fields.descr);
     data->arr_fields.flags = aligned ? NPY_ARRAY_ALIGNED : 0;
diff --git a/numpy/core/tests/test_dtype.py b/numpy/core/tests/test_dtype.py
index f95f95893..9b471a5bf 100644
--- a/numpy/core/tests/test_dtype.py
+++ b/numpy/core/tests/test_dtype.py
@@ -650,6 +650,29 @@ class TestSubarray:
         dt = np.dtype({"names": [], "formats": [], "itemsize": 0}, align=True)
         assert dt == np.dtype([])
 
+    def test_subarray_base_item(self):
+        arr = np.ones(3, dtype=[("f", "i", 3)])
+        # Extracting the field "absorbs" the subarray into a view:
+        assert arr["f"].base is arr
+        # Extract the structured item, and then check the tuple component:
+        item = arr.item(0)
+        assert type(item) is tuple and len(item) == 1
+        assert item[0].base is arr
+
+    def test_subarray_cast_copies(self):
+        # Older versions of NumPy did NOT copy, but they got the ownership
+        # wrong (not actually knowing the correct base!).  Versions since 1.21
+        # (I think) crashed fairly reliable.  This defines the correct behavior
+        # as a copy.  Keeping the ownership would be possible (but harder)
+        arr = np.ones(3, dtype=[("f", "i", 3)])
+        cast = arr.astype(object)
+        for fields in cast:
+            assert type(fields) == tuple and len(fields) == 1
+            subarr = fields[0]
+            assert subarr.base is None
+            assert subarr.flags.owndata
+
+
 def iter_struct_object_dtypes():
     """
     Iterates over a few complex dtypes and object pattern which