Add proper handling of unput array of usm_ndarray type in dpnp.ix_ (#2047) (#2050)

* Proper handling of input array as dpctl.tensor.usm_ndarray

* Update a page for Indexing routines

* Move dpnp.ix_ tests to keep lexycographical order

* Add tests to cover faulty use case

* Added entry to changelog

Author: antonwolfy

CHANGELOG.md

@@ -108,6 +108,7 @@ In addition, this release completes implementation of `dpnp.fft` module and adds
 * Resolved an issue with `dpnp.matmul` when an f_contiguous `out` keyword is passed to the the function [#1872](https://github.com/IntelPython/dpnp/pull/1872)
 * Resolved a possible race condition in `dpnp.inv` [#1940](https://github.com/IntelPython/dpnp/pull/1940)
 * Resolved an issue with failing tests for `dpnp.append` when running on a device without fp64 support [#2034](https://github.com/IntelPython/dpnp/pull/2034)
+* Resolved an issue with input array of `usm_ndarray` passed into `dpnp.ix_` [#2047](https://github.com/IntelPython/dpnp/pull/2047)
 
 
 ## [0.15.0] - 05/25/2024

doc/reference/indexing.rst

@@ -1,14 +1,13 @@
 .. _routines.indexing:
+.. _arrays.indexing:
 
-Array Indexing Routines
-=======================
+Indexing routines
+=================
 
-.. https://docs.scipy.org/doc/numpy/reference/arrays.indexing.html
+.. https://numpy.org/doc/stable/reference/routines.indexing.html
 
 Generating index arrays
 -----------------------
-
-
 .. autosummary::
    :toctree: generated/
    :nosignatures:

doc/reference/ndarray.rst

@@ -34,7 +34,7 @@ Indexing arrays
 Arrays can be indexed using an extended Python slicing syntax,
 ``array[selection]``.
 
-.. seealso:: :ref:`Array Indexing Routines <routines.indexing>`.
+.. seealso:: :ref:`Indexing routines <routines.indexing>`.
 
 
 Array attributes

dpnp/dpnp_iface_indexing.py

@@ -934,7 +934,7 @@ def ix_(*args):
     and the dimension with the non-unit shape value cycles through all
     N dimensions.
 
-    Using  :obj:`dpnp.ix_` one can quickly construct index arrays that will
+    Using :obj:`dpnp.ix_` one can quickly construct index arrays that will
     index the cross product. ``a[dpnp.ix_([1,3],[2,5])]`` returns the array
     ``[[a[1,2] a[1,5]], [a[3,2] a[3,5]]]``.
 
@@ -994,14 +994,15 @@ def ix_(*args):
     """
 
     dpnp.check_supported_arrays_type(*args)
+
     out = []
     nd = len(args)
     for k, new in enumerate(args):
         if new.ndim != 1:
             raise ValueError("Cross index must be 1 dimensional")
         if dpnp.issubdtype(new.dtype, dpnp.bool):
             (new,) = dpnp.nonzero(new)
-        new = new.reshape((1,) * k + (new.size,) + (1,) * (nd - k - 1))
+        new = dpnp.reshape(new, (1,) * k + (new.size,) + (1,) * (nd - k - 1))
         out.append(new)
     return tuple(out)
 

tests/test_indexing.py

@@ -1,5 +1,6 @@
 import functools
 
+import dpctl.tensor as dpt
 import numpy
 import pytest
 from dpctl.tensor._numpy_helper import AxisError
@@ -12,6 +13,7 @@
 )
 
 import dpnp
+from dpnp.dpnp_array import dpnp_array
 
 from .helper import get_all_dtypes, get_integer_dtypes, has_support_aspect64
 
@@ -272,6 +274,57 @@ def test_indexing_array_negative_strides(self):
         assert_array_equal(arr, 10.0)
 
 
+class TestIx:
+    @pytest.mark.parametrize(
+        "x0", [[0, 1], [True, True]], ids=["[0, 1]", "[True, True]"]
+    )
+    @pytest.mark.parametrize(
+        "x1",
+        [[2, 4], [False, False, True, False, True]],
+        ids=["[2, 4]", "[False, False, True, False, True]"],
+    )
+    def test_ix(self, x0, x1):
+        expected = dpnp.ix_(dpnp.array(x0), dpnp.array(x1))
+        result = numpy.ix_(numpy.array(x0), numpy.array(x1))
+
+        assert_array_equal(result[0], expected[0])
+        assert_array_equal(result[1], expected[1])
+
+    @pytest.mark.parametrize("dt", [dpnp.intp, dpnp.float32])
+    def test_ix_empty_out(self, dt):
+        a = numpy.array([], dtype=dt)
+        ia = dpnp.array(a)
+
+        (result,) = dpnp.ix_(ia)
+        (expected,) = numpy.ix_(a)
+        assert_array_equal(result, expected)
+        assert a.dtype == dt
+
+    def test_repeated_input(self):
+        a = numpy.arange(5)
+        ia = dpnp.array(a)
+
+        result = dpnp.ix_(ia, ia)
+        expected = numpy.ix_(a, a)
+        assert_array_equal(result[0], expected[0])
+        assert_array_equal(result[1], expected[1])
+
+    @pytest.mark.parametrize("arr", [[2, 4, 0, 1], [True, False, True, True]])
+    def test_usm_ndarray_input(self, arr):
+        a = numpy.array(arr)
+        ia = dpt.asarray(a)
+
+        (result,) = dpnp.ix_(ia)
+        (expected,) = numpy.ix_(a)
+        assert_array_equal(result, expected)
+        assert isinstance(result, dpnp_array)
+
+    @pytest.mark.parametrize("xp", [dpnp, numpy])
+    @pytest.mark.parametrize("shape", [(), (2, 2)])
+    def test_ix_error(self, xp, shape):
+        assert_raises(ValueError, xp.ix_, xp.ones(shape))
+
+
 class TestNonzero:
     @pytest.mark.parametrize("list_val", [[], [0], [1]])
     def test_trivial(self, list_val):
@@ -1143,37 +1196,6 @@ def test_empty_indices(self):
         )
 
 
-class TestIx:
-    @pytest.mark.parametrize(
-        "x0", [[0, 1], [True, True]], ids=["[0, 1]", "[True, True]"]
-    )
-    @pytest.mark.parametrize(
-        "x1",
-        [[2, 4], [False, False, True, False, True]],
-        ids=["[2, 4]", "[False, False, True, False, True]"],
-    )
-    def test_ix(self, x0, x1):
-        expected = dpnp.ix_(dpnp.array(x0), dpnp.array(x1))
-        result = numpy.ix_(numpy.array(x0), numpy.array(x1))
-
-        assert_array_equal(expected[0], result[0])
-        assert_array_equal(expected[1], result[1])
-
-    def test_ix_empty_out(self):
-        (a,) = dpnp.ix_(dpnp.array([], dtype=dpnp.intp))
-        assert_equal(a.dtype, dpnp.intp)
-
-        (a,) = dpnp.ix_(dpnp.array([], dtype=dpnp.float32))
-        assert_equal(a.dtype, dpnp.float32)
-
-    def test_ix_error(self):
-        with pytest.raises(ValueError):
-            dpnp.ix_(dpnp.ones(()))
-
-        with pytest.raises(ValueError):
-            dpnp.ix_(dpnp.ones((2, 2)))
-
-
 class TestSelect:
     choices_np = [
         numpy.array([1, 2, 3]),