roms-tools 3.2.0__py3-none-any.whl → 3.4.0__py3-none-any.whl

roms_tools/tests/test_analysis/test_roms_output.py

@@ -1,6 +1,3 @@
-import logging
-import os
-from datetime import datetime
 from pathlib import Path
 
 import matplotlib.pyplot as plt
@@ -9,7 +6,7 @@ import pytest
 import xarray as xr
 
 from roms_tools import Grid, ROMSOutput
-from roms_tools.download import download_test_data
+from roms_tools.datasets.download import download_test_data
 
 try:
     import xesmf  # type: ignore
@@ -58,20 +55,6 @@ def roms_output_from_restart_file_with_straddling_grid(use_dask):
     )
 
 
-@pytest.mark.parametrize(
-    "roms_output_fixture",
-    [
-        "roms_output_from_restart_file",
-        "roms_output_from_restart_file_adjusted_for_zeta",
-        "roms_output_from_restart_file_with_straddling_grid",
-    ],
-)
-def test_load_model_output_file(roms_output_fixture, request):
-    roms_output = request.getfixturevalue(roms_output_fixture)
-
-    assert isinstance(roms_output.ds, xr.Dataset)
-
-
 @pytest.fixture
 def roms_output_from_two_restart_files(use_dask):
     fname_grid = Path(download_test_data("epac25km_grd.nc"))
@@ -83,208 +66,6 @@ def roms_output_from_two_restart_files(use_dask):
     return ROMSOutput(grid=grid, path=[file1, file2], use_dask=use_dask)
 
 
-def test_load_model_output_file_list(roms_output_from_two_restart_files):
-    assert isinstance(roms_output_from_two_restart_files.ds, xr.Dataset)
-
-
-def test_load_model_output_with_wildcard(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    # Download at least two files, so these will be found within the pooch directory
-    Path(download_test_data("eastpac25km_rst.19980106000000.nc"))
-    Path(download_test_data("eastpac25km_rst.19980126000000.nc"))
-    directory = Path(
-        os.path.dirname(download_test_data("eastpac25km_rst.19980106000000.nc"))
-    )
-
-    output = ROMSOutput(grid=grid, path=directory / "*rst*.nc", use_dask=use_dask)
-    assert isinstance(output.ds, xr.Dataset)
-
-
-def test_invalid_path(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    # Non-existent file
-    with pytest.raises(FileNotFoundError):
-        ROMSOutput(
-            grid=grid,
-            path=Path("/path/to/nonexistent/file.nc"),
-            use_dask=use_dask,
-        )
-
-
-def test_set_correct_model_reference_date(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    output = ROMSOutput(
-        grid=grid,
-        path=Path(download_test_data("eastpac25km_rst.19980106000000.nc")),
-        use_dask=use_dask,
-    )
-    assert output.model_reference_date == datetime(1995, 1, 1)
-
-
-def test_model_reference_date_mismatch(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    # Create a ROMSOutput with a specified model_reference_date
-    model_ref_date = datetime(2020, 1, 1)
-    with pytest.raises(
-        ValueError, match="Mismatch between `self.model_reference_date`"
-    ):
-        ROMSOutput(
-            grid=grid,
-            path=Path(download_test_data("eastpac25km_rst.19980106000000.nc")),
-            model_reference_date=model_ref_date,
-            use_dask=use_dask,
-        )
-
-
-def test_model_reference_date_no_metadata(use_dask, tmp_path, caplog):
-    # Helper function to handle the test logic for cases where metadata is missing or invalid
-    def test_no_metadata(faulty_ocean_time_attr, expected_exception, log_message=None):
-        ds = xr.open_dataset(fname)
-        ds["ocean_time"].attrs = faulty_ocean_time_attr
-
-        # Write modified dataset to a new file
-        fname_mod = tmp_path / "eastpac25km_rst.19980106000000_without_metadata.nc"
-        ds.to_netcdf(fname_mod)
-
-        # Test case 1: Expecting a ValueError when metadata is missing or invalid
-        with pytest.raises(
-            expected_exception,
-            match="Model reference date could not be inferred from the metadata",
-        ):
-            ROMSOutput(grid=grid, path=fname_mod, use_dask=use_dask)
-
-        # Test case 2: When a model reference date is explicitly set, verify the warning
-        with caplog.at_level(logging.WARNING):
-            ROMSOutput(
-                grid=grid,
-                path=fname_mod,
-                model_reference_date=datetime(1995, 1, 1),
-                use_dask=use_dask,
-            )
-
-            if log_message:
-                # Verify the warning message in the log
-                assert log_message in caplog.text
-
-        fname_mod.unlink()
-
-    # Load grid and test data
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-    fname = download_test_data("eastpac25km_rst.19980106000000.nc")
-
-    # Test 1: Ocean time attribute 'long_name' is missing
-    test_no_metadata({}, ValueError)
-
-    # Test 2: Ocean time attribute 'long_name' contains invalid information
-    test_no_metadata(
-        {"long_name": "some random text"},
-        ValueError,
-        "Could not infer the model reference date from the metadata.",
-    )
-
-
-def test_compute_depth_coordinates(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-    fname_restart1 = Path(download_test_data("eastpac25km_rst.19980106000000.nc"))
-
-    for adjust_depth_for_sea_surface_height in [True, False]:
-        output = ROMSOutput(
-            grid=grid,
-            path=fname_restart1,
-            use_dask=use_dask,
-            adjust_depth_for_sea_surface_height=adjust_depth_for_sea_surface_height,
-        )
-
-        # Before calling get_vertical_coordinates, check if the dataset doesn't already have depth coordinates
-        assert "layer_depth_rho" not in output.ds_depth_coords.data_vars
-
-        # Call the method to get vertical coordinates
-        output._get_depth_coordinates(depth_type="layer")
-
-        # Check if the depth coordinates were added
-        assert "layer_depth_rho" in output.ds_depth_coords.data_vars
-
-
-def test_missing_zeta_gets_raised(use_dask):
-    """Test that a ValueError is raised when `zeta` is missing from the dataset and
-    `adjust_depth_for_sea_surface_height` is enabled.
-    """
-    # Load the grid
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    # Load the ROMS output
-    fname_restart1 = Path(download_test_data("eastpac25km_rst.19980106000000.nc"))
-    roms_output = ROMSOutput(
-        grid=grid,
-        path=fname_restart1,
-        use_dask=use_dask,
-        adjust_depth_for_sea_surface_height=True,
-    )
-
-    # Remove `zeta` from the dataset
-    object.__setattr__(
-        roms_output, "ds", roms_output.ds.drop_vars("zeta", errors="ignore")
-    )
-
-    # Expect ValueError when calling `_get_depth_coordinates`
-    with pytest.raises(
-        ValueError,
-        match="`zeta` is required in provided ROMS output when `adjust_depth_for_sea_surface_height` is enabled.",
-    ):
-        roms_output._get_depth_coordinates()
-
-
-def test_check_vertical_coordinate_mismatch(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    fname_restart1 = Path(download_test_data("eastpac25km_rst.19980106000000.nc"))
-    output = ROMSOutput(grid=grid, path=fname_restart1, use_dask=use_dask)
-
-    # create a mock dataset with inconsistent vertical coordinate parameters
-    ds_mock = output.ds.copy()
-
-    # Modify one of the vertical coordinate attributes to cause a mismatch
-    ds_mock.attrs["theta_s"] = 999
-
-    # Check if ValueError is raised due to mismatch
-    with pytest.raises(ValueError, match="theta_s from grid"):
-        output._check_vertical_coordinate(ds_mock)
-
-    # create a mock dataset with inconsistent vertical coordinate parameters
-    ds_mock = output.ds.copy()
-
-    # Modify one of the vertical coordinate attributes to cause a mismatch
-    ds_mock.attrs["Cs_w"] = ds_mock.attrs["Cs_w"] + 0.01
-
-    # Check if ValueError is raised due to mismatch
-    with pytest.raises(ValueError, match="Cs_w from grid"):
-        output._check_vertical_coordinate(ds_mock)
-
-
-def test_that_coordinates_are_added(use_dask):
-    fname_grid = Path(download_test_data("epac25km_grd.nc"))
-    grid = Grid.from_file(fname_grid)
-
-    fname_restart1 = Path(download_test_data("eastpac25km_rst.19980106000000.nc"))
-    output = ROMSOutput(grid=grid, path=fname_restart1, use_dask=use_dask)
-
-    assert "abs_time" in output.ds.coords
-    assert "lat_rho" in output.ds.coords
-    assert "lon_rho" in output.ds.coords
-
-
 @pytest.mark.parametrize(
     "roms_output_fixture",
     [

roms_tools/tests/{test_setup/test_datasets.py → test_datasets/test_lat_lon_datasets.py}

@@ -8,16 +8,15 @@ import numpy as np
 import pytest
 import xarray as xr
 
-from roms_tools.download import download_test_data
-from roms_tools.setup.datasets import (
+from roms_tools.datasets.download import download_test_data
+from roms_tools.datasets.lat_lon_datasets import (
     GLORYS_GLOBAL_GRID_PATH,
     CESMBGCDataset,
-    Dataset,
     ERA5ARCODataset,
     ERA5Correction,
     GLORYSDataset,
     GLORYSDefaultDataset,
-    RiverDataset,
+    LatLonDataset,
     TPXODataset,
     _concatenate_longitudes,
     choose_subdomain,
@@ -158,7 +157,7 @@ def test_select_times(data_fixture, expected_time_values, request, tmp_path, use
 
     filepath = tmp_path / "test.nc"
     dataset.to_netcdf(filepath)
-    dataset = Dataset(
+    dataset = LatLonDataset(
         filename=filepath,
         var_names={"var": "var"},
         start_time=start_time,
@@ -192,7 +191,7 @@ def test_select_times_valid_start_no_end_time(
     dataset.to_netcdf(filepath)
 
     # Instantiate Dataset object using the temporary file
-    dataset = Dataset(
+    dataset = LatLonDataset(
         filename=filepath,
         var_names={"var": "var"},
         start_time=start_time,
@@ -229,7 +228,7 @@ def test_select_times_invalid_start_no_end_time(
         ValueError,
         match="No exact match found ",
     ):
-        dataset = Dataset(
+        dataset = LatLonDataset(
             filename=filepath,
             var_names={"var": "var"},
             start_time=datetime(2022, 5, 1),
@@ -245,7 +244,7 @@ def test_multiple_matching_times(
     """
     filepath = tmp_path / "test.nc"
     global_dataset_with_multiple_times_per_day.to_netcdf(filepath)
-    dataset = Dataset(
+    dataset = LatLonDataset(
         filename=filepath,
         var_names={"var": "var"},
         start_time=datetime(2021, 12, 31, 22, 0),
@@ -265,7 +264,7 @@ def test_warnings_times(global_dataset, tmp_path, caplog, use_dask):
         start_time = datetime(2021, 1, 1)
         end_time = datetime(2021, 2, 1)
 
-        Dataset(
+        LatLonDataset(
             filename=filepath,
             var_names={"var": "var"},
             start_time=start_time,
@@ -279,7 +278,7 @@ def test_warnings_times(global_dataset, tmp_path, caplog, use_dask):
         start_time = datetime(2024, 1, 1)
         end_time = datetime(2024, 2, 1)
 
-        Dataset(
+        LatLonDataset(
             filename=filepath,
             var_names={"var": "var"},
             start_time=start_time,
@@ -291,13 +290,13 @@ def test_warnings_times(global_dataset, tmp_path, caplog, use_dask):
 
 
 def test_from_ds(global_dataset, global_dataset_with_noon_times, use_dask, tmp_path):
-    """Test the from_ds method of the Dataset class."""
+    """Test the from_ds method of the LatLonDataset class."""
     start_time = datetime(2022, 1, 1)
 
     filepath = tmp_path / "test.nc"
     global_dataset.to_netcdf(filepath)
 
-    dataset = Dataset(
+    dataset = LatLonDataset(
         filename=filepath,
         var_names={"var": "var"},
         dim_names={
@@ -311,9 +310,9 @@ def test_from_ds(global_dataset, global_dataset_with_noon_times, use_dask, tmp_p
         allow_flex_time=True,
     )
 
-    new_dataset = Dataset.from_ds(dataset, global_dataset_with_noon_times)
+    new_dataset = LatLonDataset.from_ds(dataset, global_dataset_with_noon_times)
 
-    assert isinstance(new_dataset, Dataset)  # Check that the new instance is a Dataset
+    assert isinstance(new_dataset, LatLonDataset)
     assert new_dataset.ds.equals(
         global_dataset_with_noon_times
     )  # Verify the new ds attribute is set correctly
@@ -330,7 +329,7 @@ def test_reverse_latitude_reverse_depth_choose_subdomain(
     global_dataset, tmp_path, use_dask
 ):
     """Test reversing latitude when it is not ascending, the choose_subdomain method,
-    and the convert_to_negative_depth method of the Dataset class.
+    and the convert_to_negative_depth method of the LatLonDataset class.
     """
     start_time = datetime(2022, 1, 1)
 
@@ -339,7 +338,7 @@ def test_reverse_latitude_reverse_depth_choose_subdomain(
     global_dataset["depth"] = global_dataset["depth"][::-1]
     global_dataset.to_netcdf(filepath)
 
-    dataset = Dataset(
+    dataset = LatLonDataset(
         filename=filepath,
         var_names={"var": "var"},
         dim_names={
@@ -393,7 +392,9 @@ def test_reverse_latitude_reverse_depth_choose_subdomain(
 def test_check_if_global_with_global_dataset(global_dataset, tmp_path, use_dask):
     filepath = tmp_path / "test.nc"
     global_dataset.to_netcdf(filepath)
-    dataset = Dataset(filename=filepath, var_names={"var": "var"}, use_dask=use_dask)
+    dataset = LatLonDataset(
+        filename=filepath, var_names={"var": "var"}, use_dask=use_dask
+    )
     is_global = dataset.check_if_global(dataset.ds)
     assert is_global
 
@@ -403,7 +404,9 @@ def test_check_if_global_with_non_global_dataset(
 ):
     filepath = tmp_path / "test.nc"
     non_global_dataset.to_netcdf(filepath)
-    dataset = Dataset(filename=filepath, var_names={"var": "var"}, use_dask=use_dask)
+    dataset = LatLonDataset(
+        filename=filepath, var_names={"var": "var"}, use_dask=use_dask
+    )
     is_global = dataset.check_if_global(dataset.ds)
 
     assert not is_global
@@ -421,7 +424,7 @@ def test_check_dataset(global_dataset, tmp_path, use_dask):
     with pytest.raises(
         ValueError, match="Dataset does not contain all required variables."
     ):
-        Dataset(
+        LatLonDataset(
             filename=filepath,
             var_names={"var": "var"},
             start_time=start_time,
@@ -440,7 +443,7 @@ def test_check_dataset(global_dataset, tmp_path, use_dask):
     with pytest.raises(
         ValueError, match="Dataset does not contain all required dimensions."
     ):
-        Dataset(
+        LatLonDataset(
             filename=filepath,
             var_names={"var": "var"},
             start_time=start_time,
@@ -601,14 +604,14 @@ def test_data_concatenation(use_dask):
 def test_time_validation(use_dask):
     fname = download_test_data("GLORYS_NA_2012.nc")
 
-    with pytest.raises(TypeError, match="start_time must be a datetime object"):
+    with pytest.raises(TypeError, match="`start_time` must be a datetime object"):
         GLORYSDataset(
             filename=fname,
             start_time="dummy",
             end_time=datetime(2013, 1, 1),
             use_dask=use_dask,
         )
-    with pytest.raises(TypeError, match="end_time must be a datetime object"):
+    with pytest.raises(TypeError, match="`end_time` must be a datetime object"):
         GLORYSDataset(
             filename=fname,
             start_time=datetime(2012, 1, 1),
@@ -760,48 +763,6 @@ class TestTPXODataset:
             global_tpxo_dataset.select_constituents(11, omega)
 
 
-class TestRiverDataset:
-    def test_deduplicate_river_names(self, tmp_path):
-        sample_dim_and_var_names = {
-            "dim_names": {"station": "station", "time": "time"},
-            "var_names": {
-                "latitude": "lat",
-                "longitude": "lon",
-                "flux": "flux",
-                "ratio": "ratio",
-                "name": "name",
-            },
-        }
-
-        data = {
-            "lat": (["station"], [10.0, 20.0, 30.0]),
-            "lon": (["station"], [100.0, 110.0, 120.0]),
-            "flux": (["time", "station"], np.random.rand(1, 3)),
-            "ratio": (["time", "station"], np.random.rand(1, 3)),
-            "name": (["station"], ["Amazon", "Nile", "Amazon"]),  # duplicate
-        }
-        coords = {"station": [0, 1, 2], "time": [0]}
-        ds = xr.Dataset(data, coords=coords)
-
-        # Write to temporary NetCDF file
-        file_path = tmp_path / "rivers.nc"
-        ds.to_netcdf(file_path)
-
-        river_dataset = RiverDataset(
-            filename=file_path,
-            start_time=datetime(2000, 1, 1),
-            end_time=datetime(2000, 1, 2),
-            dim_names=sample_dim_and_var_names["dim_names"],
-            var_names=sample_dim_and_var_names["var_names"],
-        )
-
-        names = river_dataset.ds["name"].values
-        assert "Amazon_1" in names
-        assert "Amazon_2" in names
-        assert "Nile" in names
-        assert len(set(names)) == len(names)  # all names must be unique
-
-
 # test _concatenate_longitudes
 
 

roms_tools/tests/test_datasets/test_river_datasets.py

@@ -0,0 +1,48 @@
+from datetime import datetime
+
+import numpy as np
+import xarray as xr
+
+from roms_tools.datasets.river_datasets import RiverDataset
+
+
+class TestRiverDataset:
+    def test_deduplicate_river_names(self, tmp_path):
+        sample_dim_and_var_names = {
+            "dim_names": {"station": "station", "time": "time"},
+            "var_names": {
+                "latitude": "lat",
+                "longitude": "lon",
+                "flux": "flux",
+                "ratio": "ratio",
+                "name": "name",
+            },
+        }
+
+        data = {
+            "lat": (["station"], [10.0, 20.0, 30.0]),
+            "lon": (["station"], [100.0, 110.0, 120.0]),
+            "flux": (["time", "station"], np.random.rand(1, 3)),
+            "ratio": (["time", "station"], np.random.rand(1, 3)),
+            "name": (["station"], ["Amazon", "Nile", "Amazon"]),  # duplicate
+        }
+        coords = {"station": [0, 1, 2], "time": [0]}
+        ds = xr.Dataset(data, coords=coords)
+
+        # Write to temporary NetCDF file
+        file_path = tmp_path / "rivers.nc"
+        ds.to_netcdf(file_path)
+
+        river_dataset = RiverDataset(
+            filename=file_path,
+            start_time=datetime(2000, 1, 1),
+            end_time=datetime(2000, 1, 2),
+            dim_names=sample_dim_and_var_names["dim_names"],
+            var_names=sample_dim_and_var_names["var_names"],
+        )
+
+        names = river_dataset.ds["name"].values
+        assert "Amazon_1" in names
+        assert "Amazon_2" in names
+        assert "Nile" in names
+        assert len(set(names)) == len(names)  # all names must be unique