feat: Scalar indexing returns numpy scalar, not 0-d array (fixes zarr-developers#3741)

- Add comprehensive tests for scalar indexing behavior
- Tests verify compatibility with NumPy scalar returns
- Partial indexing still returns arrays as expected

Author: Claude Subagent

changes/3741.bugfix.md

@@ -0,0 +1 @@
+Scalar indexing on arrays now returns numpy scalars instead of 0-d arrays, matching NumPy behavior.

tests/core/test_array_scalar_indexing.py

@@ -1,40 +1,39 @@
 """Tests for scalar indexing (Issue #3741)."""
 
 import numpy as np
-import pytest
 
 import zarr
 
 
 class TestScalarIndexing:
     """Test that scalar indexing returns numpy scalars, matching numpy behavior."""
 
-    def test_1d_scalar_indexing(self):
+    def test_1d_scalar_indexing(self) -> None:
         """Test scalar indexing on 1-D array returns numpy scalar."""
         arr_zarr = zarr.array([1, 2, 3, 4, 5], dtype="int64")
         arr_numpy = np.array([1, 2, 3, 4, 5], dtype="int64")
 
         result_zarr = arr_zarr[0]
         result_numpy = arr_numpy[0]
 
-        assert type(result_zarr) == type(result_numpy)
+        assert type(result_zarr) is type(result_numpy)
         assert result_zarr == result_numpy
         assert not isinstance(result_zarr, np.ndarray)
         assert isinstance(result_zarr, np.generic)
 
-    def test_2d_scalar_indexing(self):
+    def test_2d_scalar_indexing(self) -> None:
         """Test scalar indexing on 2-D array returns numpy scalar."""
         arr_zarr = zarr.array([[1, 2, 3], [4, 5, 6]], dtype="int64")
         arr_numpy = np.array([[1, 2, 3], [4, 5, 6]], dtype="int64")
 
         result_zarr = arr_zarr[0, 0]
         result_numpy = arr_numpy[0, 0]
 
-        assert type(result_zarr) == type(result_numpy)
+        assert type(result_zarr) is type(result_numpy)
         assert result_zarr == result_numpy
         assert not isinstance(result_zarr, np.ndarray)
 
-    def test_slice_indexing_returns_array(self):
+    def test_slice_indexing_returns_array(self) -> None:
         """Test that slice indexing still returns arrays."""
         arr_zarr = zarr.array([1, 2, 3, 4, 5])
         result = arr_zarr[0:2]
@@ -43,15 +42,15 @@ def test_slice_indexing_returns_array(self):
         assert result.ndim == 1
         assert len(result) == 2
 
-    def test_partial_scalar_indexing_on_2d(self):
+    def test_partial_scalar_indexing_on_2d(self) -> None:
         """Test partial scalar indexing on 2-D array returns 1-D array."""
         arr_zarr = zarr.array([[1, 2, 3], [4, 5, 6]], dtype="int64")
         arr_numpy = np.array([[1, 2, 3], [4, 5, 6]], dtype="int64")
 
         result_zarr = arr_zarr[0]
         result_numpy = arr_numpy[0]
 
-        assert type(result_zarr) == type(result_numpy)
+        assert type(result_zarr) is type(result_numpy)
         assert isinstance(result_zarr, np.ndarray)
         assert result_zarr.ndim == 1
         np.testing.assert_array_equal(result_zarr, result_numpy)