def test_get_timeseries_point(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
r = ds.get_timeseries_point(13.0, -149.0, 0, 2031436800,
2034072000, "votemper")
self.assertAlmostEqual(r[0], 299.17, places=2)
self.assertAlmostEqual(r[1], 299.72, places=2)
def test_get_profile(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
p, d = n.get_profile(13.0, -149.0, 0, 'votemper')
self.assertAlmostEqual(p[0], 299.17, places=2)
self.assertAlmostEqual(p[10], 299.15, places=2)
self.assertAlmostEqual(p[20], 296.466766, places=6)
self.assertTrue(np.ma.is_masked(p[49]))
def test_get_point(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
self.assertAlmostEqual(ds.get_point(13.0, -149.0, 0, "votemper",
2031436800),
299.17,
places=2)
def test_bottom_point(self):
nc_data = NetCDFData('tests/testdata/nemo_test.nc')
with Nemo(nc_data) as ds:
self.assertAlmostEqual(ds.get_point(13.0, -149.0, 'bottom',
'votemper', 2031436800),
274.13,
places=2)
def test_no_depth_variable(self, patch_netcdfdata):
# This is a hack to trigger the no depth variable edge case
patch_netcdfdata.depth_dimensions.return_value = []
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
np.testing.assert_array_equal(ds.depths, np.array([0]))
def test_depths(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
expected = np.array(
[
0.494025,
1.54138,
2.64567,
3.81949,
5.07822,
6.44061,
7.92956,
9.573,
11.405,
13.4671,
15.8101,
18.4956,
21.5988,
25.2114,
29.4447,
34.4342,
40.3441,
47.3737,
55.7643,
65.8073,
77.8539,
92.3261,
109.729,
130.666,
155.851,
186.126,
222.475,
266.04,
318.127,
380.213,
453.938,
541.089,
643.567,
763.333,
902.339,
1062.44,
1245.29,
1452.25,
1684.28,
1941.89,
2225.08,
2533.34,
2865.7,
3220.82,
3597.03,
3992.48,
4405.22,
4833.29,
5274.78,
5727.92,
],
dtype=np.float32,
)
np.testing.assert_array_equal(ds.depths, expected)
def test_get_raw_point(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
lat, lon, data = n.get_raw_point(13.0, -149.0, 0, 0, 'votemper')
self.assertEqual(len(lat.values.ravel()), 12)
self.assertEqual(len(lon.values.ravel()), 12)
self.assertEqual(len(data.values.ravel()), 12)
self.assertAlmostEqual(data.values[1, 1], 299.3, places=1)
def test_get_profile_depths(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
p = n.get_profile_depths(13.0, -149.0, 0, 'votemper',
[0, 10, 25, 50, 100, 200, 500, 1000])
self.assertTrue(np.ma.is_masked(p[0]))
self.assertAlmostEqual(p[1], 299.15, places=2)
self.assertAlmostEqual(p[4], 292.48, places=2)
self.assertAlmostEqual(p[7], 277.90, places=2)
def test_get_profile(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
p, d = ds.get_profile(13.0, -149.0, "votemper", 2031436800)
self.assertAlmostEqual(p[0], 299.17, places=2)
self.assertAlmostEqual(p[10], 299.15, places=2)
self.assertAlmostEqual(p[20], 296.466766, places=6)
self.assertTrue(np.ma.is_masked(p[49]))
def test_variables(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
variables = n.variables
self.assertEqual(len(variables), 3)
self.assertTrue('votemper' in variables)
self.assertEqual(variables['votemper'].name,
'Water temperature at CMC')
self.assertEqual(variables['votemper'].unit, 'Kelvins')
def test_get_profile_depths(self):
nc_data = NetCDFData('tests/testdata/nemo_test.nc')
with Nemo(nc_data) as ds:
p = ds.get_profile_depths(13.0, -149.0, 2031436800, 'votemper',
[0, 10, 25, 50, 100, 200, 500, 1000])
self.assertTrue(np.ma.is_masked(p[0]))
self.assertAlmostEqual(p[1], 299.15, places=2)
self.assertAlmostEqual(p[4], 292.48, places=2)
self.assertAlmostEqual(p[7], 277.90, places=2)
def test_get_raw_point(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
lat, lon, data = ds.get_raw_point(13.0, -149.0, 0, 2031436800,
"votemper")
self.assertEqual(len(lat.values.ravel()), 12)
self.assertEqual(len(lon.values.ravel()), 12)
self.assertEqual(len(data.values.ravel()), 12)
self.assertAlmostEqual(data.values[1, 1], 299.3, places=1)
def test_get_path_profile(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
p, d, r, dep = n.get_path_profile(
[[13, -149], [14, -140], [15, -130]], 0, 'votemper', 10)
self.assertEqual(r.shape[0], 50)
self.assertGreater(r.shape[1], 10)
self.assertEqual(r.shape[1], p.shape[1])
self.assertEqual(r.shape[1], len(d))
self.assertEqual(d[0], 0)
def test_get_timeseries_profile(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
r, d = n.get_timeseries_profile(13.0, -149.0, 0, 1, 'votemper')
self.assertAlmostEqual(r[0, 0], 299.17, places=2)
self.assertAlmostEqual(r[0, 10], 299.15, places=2)
self.assertAlmostEqual(r[0, 20], 296.466766, places=6)
self.assertTrue(np.ma.is_masked(r[0, 49]))
self.assertNotEqual(r[0, 0], r[1, 0])
self.assertTrue(np.ma.is_masked(r[1, 49]))
def test_get_path_profile(self):
nc_data = NetCDFData('tests/testdata/nemo_test.nc')
with Nemo(nc_data) as ds:
p, d, r, dep = ds.get_path_profile(
[[13, -149], [14, -140], [15, -130]], 2031436800, 'votemper', 10)
self.assertEqual(r.shape[0], 50)
self.assertGreater(r.shape[1], 10)
self.assertEqual(r.shape[1], p.shape[1])
self.assertEqual(r.shape[1], len(d))
self.assertEqual(d[0], 0)
def test_get_timeseries_profile(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
r, d = ds.get_timeseries_profile(13.0, -149.0, 2031436800,
2034072000, "votemper")
self.assertAlmostEqual(r[0, 0], 299.17, places=2)
self.assertAlmostEqual(r[0, 10], 299.15, places=2)
self.assertAlmostEqual(r[0, 20], 296.466766, places=6)
self.assertTrue(np.ma.is_masked(r[0, 49]))
self.assertNotEqual(r[0, 0], r[1, 0])
self.assertTrue(np.ma.is_masked(r[1, 49]))
def test_variables(self):
nc_data = NetCDFData('tests/testdata/nemo_test.nc')
with Nemo(nc_data) as ds:
variables = ds.variables
self.assertEqual(len(variables), 3)
self.assertTrue('votemper' in variables)
self.assertEqual(variables['votemper'].name,
'Water temperature at CMC')
self.assertEqual(variables['votemper'].unit, 'Kelvins')
self.assertEqual(sorted(variables['votemper'].dimensions),
sorted(["deptht", "time_counter", "y", "x"]))
def test_variables(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
variables = ds.variables
self.assertEqual(len(variables), 3)
self.assertTrue("votemper" in variables)
self.assertEqual(variables["votemper"].name,
"Water temperature at CMC")
self.assertEqual(variables["votemper"].unit, "Kelvins")
self.assertEqual(
sorted(variables["votemper"].dimensions),
sorted(["deptht", "time_counter", "y", "x"]),
)
def test_timestamps(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
self.assertEqual(len(n.timestamps), 2)
self.assertEqual(
n.timestamps[0],
datetime.datetime(2014, 5, 17, 0, 0, 0, 0, pytz.UTC))
# Property is read-only
with self.assertRaises(AttributeError):
n.timestamps = []
# List is immutable
with self.assertRaises(ValueError):
n.timestamps[0] = 0
def test_get_area(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
a = np.array(
np.meshgrid(np.linspace(5, 10, 10),
np.linspace(-150, -160, 10)))
r = n.get_area(a, 0, 0, 'votemper', "gaussian", 25000, 10)
self.assertAlmostEqual(r[5, 5], 301.285, places=3)
r = n.get_area(a, 0, 0, 'votemper', "bilinear", 25000, 10)
self.assertAlmostEqual(r[5, 5], 301.269, places=3)
r = n.get_area(a, 0, 0, 'votemper', "nearest", 25000, 10)
self.assertAlmostEqual(r[5, 5], 301.28986, places=5)
r = n.get_area(a, 0, 0, 'votemper', "inverse", 25000, 10)
self.assertAlmostEqual(r[5, 5], 301.2795, places=4)
def test_get_area(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
with Nemo(nc_data) as ds:
a = np.array(
np.meshgrid(np.linspace(5, 10, 10),
np.linspace(-150, -160, 10)))
r = ds.get_area(a, 0, 2031436800, "votemper", "gaussian", 25000,
10)
self.assertAlmostEqual(r[5, 5], 301.285, places=3)
r = ds.get_area(a, 0, 2031436800, "votemper", "bilinear", 25000,
10)
self.assertAlmostEqual(r[5, 5], 301.269, places=3)
r = ds.get_area(a, 0, 2031436800, "votemper", "nearest", 25000, 10)
self.assertAlmostEqual(r[5, 5], 301.28986, places=5)
r = ds.get_area(a, 0, 2031436800, "votemper", "inverse", 25000, 10)
self.assertAlmostEqual(r[5, 5], 301.2795, places=4)
def open_dataset(dataset, **kwargs):
"""Open a dataset.
Creates a CalculatedData (derived from NetCDFData) instance to handle dataset file
access and calculation layer operations.
Then, determines the type of model the dataset is from and returns the appropriate
Model-derived instance (Nemo, Mercator, Fvcom) with the calculation layer instance
as an attribute.
Note: the returned model object will be LRU-cached internally so frequent calls
to open the "same" dataset will have minimal overhead.
Params:
* dataset -- Either a DatasetConfig object, or a string URL for the dataset
Optional Keyword Arguments:
* variable {str or list} -- String or list of strings of variable keys to be loaded
(e.g. 'votemper' or ['votemper', 'vosaline']).
* timestamp {int} -- Integer value of date/time for requested data (e.g. 2128723200).
When loading a range of timestamps, this argument serves as the starting time.
* endtime {int} -- Integer value of date/time. This argument is only used when
loading a range of timestamps, and should hold the ending time.
* nearest_timestamp {bool} -- When true, open_dataset will assume the given
starttime (and endtime) do not exactly correspond to a timestamp integer
in the dataset, and will perform a binary search to find the nearest timestamp
that is less-than-or-equal-to the given starttime (and endtime).
* meta_only {bool} -- Skip some dataset access operations in order to speed up
response.
"""
if not dataset:
raise ValueError("Unknown dataset.")
try:
url = dataset.url
calculated_vars = dataset.calculated_variables
except AttributeError:
url = dataset
calculated_vars = {}
if url is None:
raise ValueError("Dataset url is None.")
args = {
"calculated": calculated_vars,
"meta_only": kwargs.get("meta_only", False),
"grid_angle_file_url": getattr(dataset, "grid_angle_file_url", ""),
"dataset_key": getattr(dataset, "key", ""),
}
nc_data = CalculatedData(url, **args)
if not args["meta_only"]:
# Get required NC files from database and add to args
nc_data.get_nc_file_list(dataset, **kwargs)
dimension_list = nc_data.dimensions
if not dimension_list:
raise RuntimeError(f"Dataset not supported: {url}.")
if 'longitude' in dimension_list or 'latitude' in dimension_list:
return Mercator(nc_data)
if 'siglay' in dimension_list:
return Fvcom(nc_data)
return Nemo(nc_data)
def test_get_timeseries_point(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
r = n.get_timeseries_point(13.0, -149.0, 0, 0, 1, 'votemper')
self.assertAlmostEqual(r[0], 299.17, places=2)
self.assertAlmostEqual(r[1], 299.72, places=2)
def test_init(self):
Nemo(None)
def test_bottom_point(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
self.assertAlmostEqual(n.get_point(13.0, -149.0, 'bottom', 0,
'votemper'),
274.13,
places=2)
def test_get_profile_raises_when_surface_variable_requested(self):
nc_data = NetCDFData("tests/testdata/salishseacast_ssh_test.nc")
with Nemo(nc_data) as ds:
with self.assertRaises(APIError):
ds.get_profile(None, None, "ssh", None, None)
def test_get_point(self):
with Nemo('tests/testdata/nemo_test.nc') as n:
self.assertAlmostEqual(n.get_point(13.0, -149.0, 0, 0, 'votemper'),
299.17,
places=2)
def test_init(self):
nc_data = NetCDFData("tests/testdata/nemo_test.nc")
ds = Nemo(nc_data)
self.assertIs(ds.nc_data, nc_data)
self.assertEqual(ds.variables, nc_data.variables)
def test_open(self):
with Nemo('tests/testdata/nemo_test.nc'):
pass