Add backend statement and Examples section to cost_distance docstring

xrspatial/cost_distance.py

@@ -1226,6 +1226,12 @@ def cost_distance(
     to reach the nearest target pixel, where traversal cost along each
     edge equals ``geometric_distance * mean_friction_of_endpoints``.
 
+    Cost-distance supports NumPy, CuPy, Dask with NumPy, and Dask with CuPy
+    backed xarray DataArray. The return values of `cost_distance`
+    are of the same type as the input type: a NumPy-backed input gives a
+    NumPy-backed result, a CuPy-backed input gives a CuPy-backed result,
+    and a Dask-backed input gives a Dask-backed result.
+
     Parameters
     ----------
     raster : xr.DataArray or xr.Dataset
@@ -1254,6 +1260,38 @@ def cost_distance(
     xr.DataArray or xr.Dataset
         2-D array of accumulated cost-distance values (float32).
         Source pixels have cost 0.  Unreachable pixels are NaN.
+
+    Examples
+    --------
+    .. sourcecode:: python
+
+        >>> import numpy as np
+        >>> import xarray as xr
+        >>> source = np.array([
+        ...     [0., 0., 0.],
+        ...     [0., 1., 0.],
+        ...     [0., 0., 0.],
+        ... ])
+        >>> friction = np.ones((3, 3))
+        >>> n, m = source.shape
+        >>> raster = xr.DataArray(source, dims=['y', 'x'], name='raster')
+        >>> raster['y'] = np.arange(n)[::-1]
+        >>> raster['x'] = np.arange(m)
+        >>> friction_da = xr.DataArray(
+        ...     friction, dims=['y', 'x'], name='friction')
+        >>> friction_da['y'] = np.arange(n)[::-1]
+        >>> friction_da['x'] = np.arange(m)
+
+        >>> from xrspatial import cost_distance
+        >>> result = cost_distance(raster, friction_da)
+        >>> result
+        <xarray.DataArray (y: 3, x: 3)>
+        array([[1.4142135, 1.       , 1.4142135],
+               [1.       , 0.       , 1.       ],
+               [1.4142135, 1.       , 1.4142135]], dtype=float32)
+        Coordinates:
+          * y        (y) int64 2 1 0
+          * x        (x) int64 0 1 2
     """
     # --- validation ---
     _validate_raster(raster, func_name='cost_distance', name='raster')

xrspatial/tests/test_cost_distance.py

@@ -1276,3 +1276,38 @@ def test_iterative_heap_no_overflow_variable_friction(connectivity):
                 raster, friction, connectivity=connectivity))
         expected = _reference_dijkstra(source, friction_data, connectivity)
         np.testing.assert_allclose(out, expected, equal_nan=True, atol=1e-3)
+
+
+def test_docstring_states_all_backends():
+    doc = cost_distance.__doc__
+    assert "CuPy" in doc
+    assert "Dask with CuPy" in doc
+    assert "support NumPy backed, and Dask with NumPy backed" not in doc
+
+
+def test_docstring_example_matches_output():
+    source = np.array([
+        [0., 0., 0.],
+        [0., 1., 0.],
+        [0., 0., 0.],
+    ])
+    friction = np.ones((3, 3))
+    n, m = source.shape
+    raster = xr.DataArray(source, dims=['y', 'x'], name='raster')
+    raster['y'] = np.arange(n)[::-1]
+    raster['x'] = np.arange(m)
+    friction_da = xr.DataArray(
+        friction, dims=['y', 'x'], name='friction')
+    friction_da['y'] = np.arange(n)[::-1]
+    friction_da['x'] = np.arange(m)
+
+    result = cost_distance(raster, friction_da)
+    out = _compute(result)
+
+    expected = np.array([
+        [np.sqrt(2), 1., np.sqrt(2)],
+        [1., 0., 1.],
+        [np.sqrt(2), 1., np.sqrt(2)],
+    ], dtype=np.float32)
+    np.testing.assert_allclose(out, expected, atol=1e-5)
+    assert result.dtype == np.float32