replace np.random.random with new Generator api

src/yt_napari/_special_loaders.py

@@ -5,7 +5,8 @@
 
 
 def _ytnapari_load_grid():
- arr = np.random.random(size=(64, 64, 64))
+ rng = np.random.default_rng()
+ arr = rng.random(size=(64, 64, 64))
  d = dict(density=(arr, "g/cm**3"), temperature=(arr, "K"))
  bbox = np.array([[-1.5, 1.5], [-1.5, 1.5], [-1.5, 1.5]])
  shp = arr.shape

src/yt_napari/_tests/conftest.py

@@ -7,7 +7,8 @@
 def yt_ugrid_ds_fn(tmpdir_factory):
  # this fixture generates a random yt dataset saved to disk that can be
  # re-loaded and sampled.
- arr = np.random.random(size=(64, 64, 64))
+ rng = np.random.default_rng()
+ arr = rng.random(size=(64, 64, 64))
  d = dict(density=(arr, "g/cm**3"), temperature=(arr, "K"))
  bbox = np.array([[-1.5, 1.5], [-1.5, 1.5], [-1.5, 1.5]])
  shp = arr.shape

src/yt_napari/_tests/test_model_ingestor.py

@@ -211,9 +211,10 @@ def test_reference_layer(domains_to_test):
  im_type = "image"
 
  spatial_layer_list = []
+ rng = np.random.default_rng()
  for d in domains_to_test.domain_sets:
  layer_domain = _mi.LayerDomain(d.left_edge, d.right_edge, d.resolution)
- im = np.random.random(d.resolution)
+ im = rng.random(d.resolution)
  spatial_layer_list.append((im, {}, im_type, layer_domain))
 
  layer_for_ref = spatial_layer_list[0][3]
@@ -270,17 +271,18 @@ def test_ref_layer_selection(domains_to_test):
  im_type = "image"
 
  spatial_layer_list = []
+ rng = np.random.default_rng()
  for d in domains_to_test.domain_sets:
  layer_domain = _mi.LayerDomain(d.left_edge, d.right_edge, d.resolution)
- im = np.random.random(d.resolution)
+ im = rng.random(d.resolution)
  spatial_layer_list.append((im, {}, im_type, layer_domain))
 
  # add another small volume layer
  le = unyt.unyt_array([0.1, 0.1, 0.1], "m")
  re = unyt.unyt_array([0.2, 0.2, 0.2], "m")
  res = (3, 4, 5)
  small_vol_domain = _mi.LayerDomain(le, re, resolution=res)
- im = np.random.random(res)
+ im = rng.random(res)
  spatial_layer_list.append((im, {}, im_type, small_vol_domain))
 
  first_ref = _mi._choose_ref_layer(spatial_layer_list, method="first_in_list")
@@ -309,8 +311,8 @@ def test_2d_3d_mix():
  layer_2d = _mi.LayerDomain(
  le, re, res, n_d=2, new_dim_value=unyt.unyt_quantity(1, "km")
  )
-
- sp_layer = (np.random.random(res), {}, "testname", layer_2d)
+ rng = np.random.default_rng()
+ sp_layer = (rng.random(res), {}, "testname", layer_2d)
  new_layer_2d = ref.align_sanitize_layer(sp_layer)
  assert "scale" not in new_layer_2d[1] # no scale when it is all 1
 
@@ -392,9 +394,9 @@ def test_timeseries_container(selection_objs):
  domain = _mi.LayerDomain(
  unyt.unyt_array([0, 0, 0], "m"), unyt.unyt_array([1.0, 1.0, 1.0], "m"), shp
  )
- im = np.random.random(shp)
+ rng = np.random.default_rng()
+ im = rng.random(shp)
 
- print("what what")
  for _ in range(3):
  tc.add(reg_1, ("enzo", "temperature"), (im, im_kwargs, "image", domain))
 
@@ -408,7 +410,7 @@ def test_timeseries_container(selection_objs):
  shp,
  n_d=2,
  )
- im = np.random.random(shp)
+ im = rng.random(shp)
 
  for _ in range(2):
  tc.add(slc_1, ("enzo", "temperature"), (im, im_kwargs, "image", domain))
@@ -494,7 +496,8 @@ def test_yt_data_dir_check(tmp_path):
 
 
 def test_linear_rescale():
- data = 10 * np.random.random((5, 5))
+ rng = np.random.default_rng()
+ data = 10 * rng.random((5, 5))
  rsc = _mi._linear_rescale(data)
  assert rsc.min() == 0.0
  assert rsc.max() == 1.0
@@ -509,7 +512,7 @@ def test_linear_rescale():
  assert np.nanmin(rsc) == 0.0
  assert np.nanmax(rsc) == 1.0
 
- data = 10 * np.random.random((5, 5))
+ data = 10 * rng.random((5, 5))
  data[1, 1] = np.inf
  with pytest.warns(RuntimeWarning, match="invalid value"):
  rsc = _mi._linear_rescale(data, fill_inf=False)

src/yt_napari/_tests/test_timeseries.py

@@ -14,7 +14,8 @@
 def yt_ds_0():
  # this fixture generates a random yt dataset saved to disk that can be
  # re-loaded and sampled.
- arr = np.random.random(size=(16, 16, 16))
+ rng = np.random.default_rng()
+ arr = rng.random(size=(16, 16, 16))
  d = dict(density=(arr, "g/cm**3"), temperature=(arr, "K"))
  bbox = np.array([[-1.5, 1.5], [-1.5, 1.5], [-1.5, 1.5]])
  shp = arr.shape
@@ -179,7 +180,8 @@ def test_validate_scale(selection):
  assert kwargdict["scale"][0] == 10.0
 
  # check that existing scale is not over-ridden
- sc_scale = np.random.random((selection.nd,))
+ rng = np.random.default_rng()
+ sc_scale = rng.random((selection.nd,))
  kwargdict = {"scale": sc_scale}
  ts._validate_scale(selection, kwargdict, False, 1.0)
  assert np.all(kwargdict["scale"] == sc_scale)

src/yt_napari/_tests/test_widget_reader.py

@@ -46,7 +46,8 @@ def _rebuild_data(final_shape, data):
  # dataset will not have the full functionality of a ds. so here, we
  # inject a correctly shaped random array here. If we start using full
  # test datasets from yt in testing, this should be changed.
- return np.random.random(final_shape) * data.mean()
+ rng = np.random.default_rng()
+ return rng.random(final_shape) * data.mean()
 
 
 def test_save_widget_reader(make_napari_viewer, yt_ugrid_ds_fn, tmp_path):