Explicitly specify HDF5VirtualBackend for test parameter.

Author: sharkinsspatial

virtualizarr/tests/test_backend.py

@@ -11,6 +11,7 @@
 from virtualizarr import open_virtual_dataset
 from virtualizarr.backend import FileType, automatically_determine_filetype
 from virtualizarr.manifests import ManifestArray
+from virtualizarr.readers import HDF5VirtualBackend
 from virtualizarr.readers.hdf import HDFVirtualBackend
 from virtualizarr.tests import (
     has_astropy,
@@ -83,7 +84,7 @@ def test_FileType():
 
 
 @requires_kerchunk
-@pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+@pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
 class TestOpenVirtualDatasetIndexes:
     def test_no_indexes(self, netcdf4_file, hdf_backend):
         vds = open_virtual_dataset(netcdf4_file, indexes={}, backend=hdf_backend)
@@ -115,7 +116,7 @@ def index_mappings_equal(indexes1: Mapping[str, Index], indexes2: Mapping[str, I
 
 
 @requires_kerchunk
-@pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+@pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
 def test_cftime_index(tmpdir, hdf_backend):
     """Ensure a virtual dataset contains the same indexes as an Xarray dataset"""
     # Note: Test was created to debug: https://github.com/zarr-developers/VirtualiZarr/issues/168
@@ -145,7 +146,7 @@ def test_cftime_index(tmpdir, hdf_backend):
 
 
 @requires_kerchunk
-@pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+@pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
 class TestOpenVirtualDatasetAttrs:
     def test_drop_array_dimensions(self, netcdf4_file, hdf_backend):
         # regression test for GH issue #150
@@ -166,20 +167,16 @@ def test_coordinate_variable_attrs_preserved(self, netcdf4_file, hdf_backend):
 @network
 @requires_s3fs
 class TestReadFromS3:
-    @pytest.mark.parametrize(
-        "filetype", ["netcdf4", None], ids=["netcdf4 filetype", "None filetype"]
-    )
     @pytest.mark.parametrize(
         "indexes", [None, {}], ids=["None index", "empty dict index"]
     )
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
-    def test_anon_read_s3(self, filetype, indexes, hdf_backend):
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
+    def test_anon_read_s3(self, indexes, hdf_backend):
         """Parameterized tests for empty vs supplied indexes and filetypes."""
         # TODO: Switch away from this s3 url after minIO is implemented.
         fpath = "s3://carbonplan-share/virtualizarr/local.nc"
         vds = open_virtual_dataset(
             fpath,
-            filetype=filetype,
             indexes=indexes,
             reader_options={"storage_options": {"anon": True}},
             backend=hdf_backend,
@@ -191,7 +188,7 @@ def test_anon_read_s3(self, filetype, indexes, hdf_backend):
 
 
 @network
-@pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+@pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
 class TestReadFromURL:
     @pytest.mark.parametrize(
         "filetype, url",
@@ -295,7 +292,7 @@ def test_virtualizarr_vs_local_nisar(self, hdf_backend):
 
 @requires_kerchunk
 class TestLoadVirtualDataset:
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     def test_loadable_variables(self, netcdf4_file, hdf_backend):
         vars_to_load = ["air", "time"]
         vds = open_virtual_dataset(
@@ -330,7 +327,7 @@ def test_explicit_filetype_and_backend(self, netcdf4_file):
                 netcdf4_file, filetype="hdf", backend=HDFVirtualBackend
             )
 
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     def test_group_kwarg(self, hdf5_groups_file, hdf_backend):
         if hdf_backend:
             with pytest.raises(NotImplementedError, match="Nested groups"):
@@ -376,13 +373,13 @@ def test_open_virtual_dataset_passes_expected_args(
         }
         mock_read_kerchunk.assert_called_once_with(**args)
 
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     def test_open_dataset_with_empty(self, hdf5_empty, tmpdir, hdf_backend):
         vds = open_virtual_dataset(hdf5_empty, backend=hdf_backend)
         assert vds.empty.dims == ()
         assert vds.empty.attrs == {"empty": "true"}
 
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     def test_open_dataset_with_scalar(self, hdf5_scalar, tmpdir, hdf_backend):
         vds = open_virtual_dataset(hdf5_scalar, backend=hdf_backend)
         assert vds.scalar.dims == ()

virtualizarr/tests/test_integration.py

@@ -5,6 +5,7 @@
 
 from virtualizarr import open_virtual_dataset
 from virtualizarr.manifests import ChunkManifest, ManifestArray
+from virtualizarr.readers import HDF5VirtualBackend
 from virtualizarr.readers.hdf import HDFVirtualBackend
 from virtualizarr.tests import requires_kerchunk
 from virtualizarr.translators.kerchunk import (
@@ -64,7 +65,7 @@ def test_no_duplicates_find_var_names():
         ),
     ],
 )
-@pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+@pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
 def test_numpy_arrays_to_inlined_kerchunk_refs(
     netcdf4_file, inline_threshold, vars_to_inline, hdf_backend
 ):
@@ -92,7 +93,7 @@ def test_numpy_arrays_to_inlined_kerchunk_refs(
 @requires_kerchunk
 @pytest.mark.parametrize("format", ["dict", "json", "parquet"])
 class TestKerchunkRoundtrip:
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     def test_kerchunk_roundtrip_no_concat(self, tmpdir, format, hdf_backend):
         # set up example xarray dataset
         ds = xr.tutorial.open_dataset("air_temperature", decode_times=False)
@@ -125,7 +126,7 @@ def test_kerchunk_roundtrip_no_concat(self, tmpdir, format, hdf_backend):
         for coord in ds.coords:
             assert ds.coords[coord].attrs == roundtrip.coords[coord].attrs
 
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     @pytest.mark.parametrize("decode_times,time_vars", [(False, []), (True, ["time"])])
     def test_kerchunk_roundtrip_concat(
         self, tmpdir, format, hdf_backend, decode_times, time_vars
@@ -195,7 +196,7 @@ def test_kerchunk_roundtrip_concat(
             assert roundtrip.time.encoding["units"] == ds.time.encoding["units"]
             assert roundtrip.time.encoding["calendar"] == ds.time.encoding["calendar"]
 
-    @pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+    @pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
     def test_non_dimension_coordinates(self, tmpdir, format, hdf_backend):
         # regression test for GH issue #105
 
@@ -282,7 +283,7 @@ def test_datetime64_dtype_fill_value(self, tmpdir, format):
 
 
 @requires_kerchunk
-@pytest.mark.parametrize("hdf_backend", [None, HDFVirtualBackend])
+@pytest.mark.parametrize("hdf_backend", [HDF5VirtualBackend, HDFVirtualBackend])
 def test_open_scalar_variable(tmpdir, hdf_backend):
     # regression test for GH issue #100