Address warning in oldestdeps tests & increase coverage

reproject/adaptive/core.py

@@ -114,9 +114,9 @@ def _reproject_adaptive_2d(
     if wcs_in.low_level_wcs.pixel_n_dim != wcs_out.low_level_wcs.pixel_n_dim:
         raise ValueError("Number of dimensions between input and output WCS should match")
     elif len(shape_out) < wcs_out.low_level_wcs.pixel_n_dim:
-        raise ValueError("Length of shape_out too small for number of dimensions in wcs_out")
+        raise ValueError("Too few dimensions in shape_out")
     elif len(array_in.shape) < wcs_in.low_level_wcs.pixel_n_dim:
-        raise ValueError("Too few input-data dimensions for number of dimensions in wcs_out")
+        raise ValueError("Too few dimensions in input data")
     elif len(array_in.shape) != len(shape_out):
         raise ValueError("Number of dimensions in input and output data should match")
 

reproject/interpolation/tests/test_core.py

@@ -382,40 +382,6 @@ def test_slice_reprojection(roundtrip_coords):
     np.testing.assert_allclose(out_cube, ((inp_cube[:-1] + inp_cube[1:]) / 2.0))
 
 
-def test_too_few_dimens():
-    header_in = fits.Header.fromtextfile(
-        get_pkg_data_filename("data/cube.hdr", package="reproject.tests")
-    )
-
-    header_in["NAXIS"] = 3
-
-    header_out = header_in.copy()
-    w_in = WCS(header_in)
-    w_out = WCS(header_out)
-
-    array_in = np.zeros((2, 3, 4, 5))
-
-    # Create mis-matches between the input and output shapes
-
-    with pytest.raises(
-        ValueError, match="Number of dimensions in input and output data should match"
-    ):
-        array_out, footprint = reproject_interp((array_in, w_in), w_out, shape_out=[2, 3, 4, 5, 6])
-
-    with pytest.raises(
-        ValueError, match="Number of dimensions in input and output data should match"
-    ):
-        array_out, footprint = reproject_interp((array_in[0], w_in), w_out, shape_out=[3, 4, 5, 6])
-
-    # Give fewer array dimensions than WCS dimensions
-
-    with pytest.raises(ValueError, match="Too few dimensions in input data"):
-        array_out, footprint = reproject_interp((array_in[0, 0], w_in), w_out, shape_out=[4, 5, 6])
-
-    with pytest.raises(ValueError, match="Too few dimensions in shape_out"):
-        array_out, footprint = reproject_interp((array_in[0], w_in), w_out, shape_out=[5, 6])
-
-
 @pytest.mark.parametrize("roundtrip_coords", (False, True))
 def test_inequal_wcs_dims(roundtrip_coords):
     inp_cube = np.arange(3, dtype="float").repeat(4 * 5).reshape(3, 4, 5)
@@ -723,9 +689,18 @@ def test_blocked_broadcast_reprojection(input_extra_dims, output_shape, parallel
         # Provide the broadcast dimensions as part of the output shape
         output_shape = image_stack.shape
 
-    array_broadcast, footprint_broadcast = reproject_interp(
-        (image_stack, header_in), header_out, output_shape, parallel=parallel, block_size=[5, 5]
-    )
+    # the warning import and ignore is needed to keep pytest happy when running with
+    # older versions of astropy which don't have this fix:
+    # https://github.com/astropy/astropy/pull/12844
+    # All the warning code should be removed when old version no longer being used
+    # Using context manager ensure only blocked function has them ignored
+    import warnings
+
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore", category=FITSFixedWarning)
+        array_broadcast, footprint_broadcast = reproject_interp(
+            (image_stack, header_in), header_out, output_shape, parallel=parallel, block_size=[5, 5]
+        )
 
     np.testing.assert_array_equal(footprint_broadcast, footprint_ref)
     np.testing.assert_allclose(array_broadcast, array_ref)

reproject/tests/test_high_level.py

@@ -207,3 +207,53 @@ def test_identity_projection(projection_type):
     else:
         # The Gaussian kernel in the adaptive algorithm blurs the image
         assert np.all(data_in != data_out)
+
+
+@pytest.mark.parametrize(
+    "reproject_function", [reproject_interp, reproject_adaptive, reproject_exact]
+)
+def test_dimensions_checks(reproject_function):
+    header_in = fits.Header.fromtextfile(
+        get_pkg_data_filename("data/gc_eq.hdr", package="reproject.tests")
+    )
+
+    header_out = header_in.copy()
+    w_in = WCS(header_in)
+    w_out = WCS(header_out)
+
+    array_in = np.zeros((2, 3, 4, 5))
+
+    # Create mis-matches between the input and output shapes
+
+    with pytest.raises(
+        ValueError, match="Number of dimensions in input and output data should match"
+    ):
+        array_out, footprint = reproject_function(
+            (array_in, w_in), w_out, shape_out=[2, 3, 4, 5, 6]
+        )
+
+    with pytest.raises(
+        ValueError, match="Number of dimensions in input and output data should match"
+    ):
+        array_out, footprint = reproject_function(
+            (array_in[0], w_in), w_out, shape_out=[3, 4, 5, 6]
+        )
+
+    # Give fewer array dimensions than WCS dimensions
+
+    with pytest.raises(ValueError, match="Too few dimensions in input data"):
+        array_out, footprint = reproject_function(
+            (array_in[0, 0, 0], w_in), w_out, shape_out=[4, 5, 6]
+        )
+
+    with pytest.raises(ValueError, match="Too few dimensions in shape_out"):
+        array_out, footprint = reproject_function(
+            (array_in[0], w_in),
+            w_out,
+            shape_out=[
+                5,
+            ],
+        )
+
+    with pytest.raises(ValueError, match="Dimensions to be looped over must match exactly"):
+        array_out, footprint = reproject_function((array_in, w_in), w_out, shape_out=[3, 3, 4, 5])