def test_time_start_only(self):
da = open_dataset(self.nc_poslons).tas
yr_st = "2050"
# start date only
with pytest.warns(None):
out = subset.subset_time(da, start_date=f"{yr_st}-01")
np.testing.assert_array_equal(out.time.dt.year.min(), int(yr_st))
np.testing.assert_array_equal(out.time.dt.year.max(),
da.time.dt.year.max())
with pytest.warns(None):
out = subset.subset_time(da, start_date=f"{yr_st}-07")
np.testing.assert_array_equal(out.time.dt.year.min(), int(yr_st))
np.testing.assert_array_equal(out.time.min().dt.month, 7)
np.testing.assert_array_equal(out.time.dt.year.max(),
da.time.dt.year.max())
np.testing.assert_array_equal(out.time.max(), da.time.max())
with pytest.warns(None):
out = subset.subset_time(da, start_date=f"{yr_st}-07-15")
np.testing.assert_array_equal(out.time.dt.year.min(), int(yr_st))
np.testing.assert_array_equal(out.time.min().dt.month, 7)
np.testing.assert_array_equal(out.time.min().dt.day, 15)
np.testing.assert_array_equal(out.time.dt.year.max(),
da.time.dt.year.max())
np.testing.assert_array_equal(out.time.max(), da.time.max())
def test_warnings(self):
da = xr.open_dataset(self.nc_poslons).tas
with pytest.raises(ValueError) as record:
subset.subset_time(da, start_date="2059", end_date="2050")
with pytest.raises(TypeError):
subset.subset_time(da, start_yr=2050, end_yr=2059)
with pytest.warns(None) as record:
subset.subset_time(
da,
start_date=2050,
end_date=2055,
)
assert (
'start_date and end_date require dates in (type: str) using formats of "%Y", "%Y-%m" or "%Y-%m-%d".'
in [str(q.message) for q in record])
with pytest.warns(None) as record:
subset.subset_time(da,
start_date="2064-01-01T00:00:00",
end_date="2065-02-01T03:12:01")
assert [str(q.message) for q in record] == [
'"start_date" has been nudged to nearest valid time step in xarray object.',
'"end_date" has been nudged to nearest valid time step in xarray object.',
]
def test_simple(self):
da = open_dataset(self.nc_poslons).tas
yr_st = "2050"
yr_ed = "2059"
out = subset.subset_time(da, start_date=yr_st, end_date=yr_ed)
out1 = subset.subset_time(da,
start_date=f"{yr_st}-01",
end_date=f"{yr_ed}-12")
out2 = subset.subset_time(da,
start_date=f"{yr_st}-01-01",
end_date=f"{yr_ed}-12-31")
np.testing.assert_array_equal(out, out1)
np.testing.assert_array_equal(out, out2)
np.testing.assert_array_equal(len(np.unique(out.time.dt.year)), 10)
np.testing.assert_array_equal(out.time.dt.year.max(), int(yr_ed))
np.testing.assert_array_equal(out.time.dt.year.min(), int(yr_st))
def test_time_incomplete_years(self):
da = open_dataset(self.nc_poslons).tas
yr_st = "2050"
yr_ed = "2059"
out = subset.subset_time(da,
start_date=f"{yr_st}-07-01",
end_date=f"{yr_ed}-06-30")
out1 = subset.subset_time(da,
start_date=f"{yr_st}-07",
end_date=f"{yr_ed}-06")
np.testing.assert_array_equal(out, out1)
np.testing.assert_array_equal(out.time.dt.year.min(), int(yr_st))
np.testing.assert_array_equal(out.time.min().dt.month, 7)
np.testing.assert_array_equal(out.time.min().dt.day, 1)
np.testing.assert_array_equal(out.time.dt.year.max(), int(yr_ed))
np.testing.assert_array_equal(out.time.max().dt.month, 6)
np.testing.assert_array_equal(out.time.max().dt.day, 30)
def test_time_end_only(self):
da = open_dataset(self.nc_poslons).tas
yr_ed = "2059"
# end date only
with pytest.warns(None):
out = subset.subset_time(da, end_date=f"{yr_ed}-01")
np.testing.assert_array_equal(out.time.dt.year.max(), int(yr_ed))
np.testing.assert_array_equal(out.time.max().dt.month, 1)
np.testing.assert_array_equal(out.time.max().dt.day, 31)
np.testing.assert_array_equal(out.time.min(), da.time.min())
with pytest.warns(None):
out = subset.subset_time(da, end_date=f"{yr_ed}-06-15")
np.testing.assert_array_equal(out.time.dt.year.max(), int(yr_ed))
np.testing.assert_array_equal(out.time.max().dt.month, 6)
np.testing.assert_array_equal(out.time.max().dt.day, 15)
np.testing.assert_array_equal(out.time.min(), da.time.min())
def test_time_dates_outofbounds(self):
da = open_dataset(self.nc_poslons).tas
yr_st = "1776"
yr_ed = "2077"
with pytest.warns(None) as record:
out = subset.subset_time(da,
start_date=f"{yr_st}-01",
end_date=f"{yr_ed}-01")
np.testing.assert_array_equal(out.time.dt.year.min(),
da.time.dt.year.min())
np.testing.assert_array_equal(out.time.dt.year.max(),
da.time.dt.year.max())
assert (
'"start_date" not found within input date time range. Defaulting to minimum time step in xarray object.'
in [str(q.message) for q in record])
assert (
'"end_date" not found within input date time range. Defaulting to maximum time step in xarray object.'
in [str(q.message) for q in record])
def test_warnings(self):
da = open_dataset(self.nc_poslons).tas
with pytest.raises(ValueError) as record:
subset.subset_time(da, start_date="2059", end_date="2050")
with pytest.raises(TypeError):
subset.subset_time(da, start_yr=2050, end_yr=2059)
with pytest.warns(None) as record:
subset.subset_time(
da,
start_date=2050,
end_date=2055,
)
assert (
'start_date and end_date require dates in (type: str) using formats of "%Y", "%Y-%m" or "%Y-%m-%d".'
in [str(q.message) for q in record])
def finch_average_shape(
process: Process,
netcdf_inputs: List[ComplexInput],
request_inputs: RequestInputs,
) -> List[Path]:
"""Parse wps `request_inputs` based on their name and average `netcdf_inputs`.
The expected names of the request_inputs are as followed (taken from `wpsio.py`):
- shape: Polygon contour to average the data over.
- start_date: Initial date for temporal subsetting.
- end_date: Final date for temporal subsetting.
"""
shp = Path(request_inputs[wpsio.shape.identifier][0].file)
if shp.suffix == ".zip":
shp = extract_shp(shp)
start_date = single_input_or_none(request_inputs,
wpsio.start_date.identifier)
end_date = single_input_or_none(request_inputs, wpsio.end_date.identifier)
tolerance = single_input_or_none(request_inputs,
wpsio.tolerance.identifier)
variables = [r.data for r in request_inputs.get("variable", [])]
shape = gpd.read_file(shp)
if tolerance > 0:
shape['geometry'] = shape.simplify(tolerance)
n_files = len(netcdf_inputs)
count = 0
output_files = []
for resource in netcdf_inputs:
# if not subsetting by time, it's not necessary to decode times
time_subset = start_date is not None or end_date is not None
dataset = try_opendap(resource,
decode_times=time_subset,
chunk_dims=['time', 'realization'])
count += 1
write_log(
process,
f"Averaging file {count} of {n_files} ({getattr(resource, resource.prop)})",
subtask_percentage=(count - 1) * 100 // n_files,
)
dataset = dataset[variables] if variables else dataset
if time_subset:
dataset = subset_time(dataset,
start_date=start_date,
end_date=end_date)
averaged = average_shape(dataset, shape)
if not all(averaged.dims.values()):
LOGGER.warning(f"Average is empty for dataset: {resource.url}")
return
p = make_subset_file_name(resource, kind="avg")
output_filename = Path(process.workdir) / p
dataset_to_netcdf(averaged, output_filename)
output_files.append(output_filename)
return output_files
