def test_get_clean_interp_index_potential_overflow():
da = xr.DataArray(
[0, 1, 2],
dims=("time",),
coords={"time": xr.cftime_range("0000-01-01", periods=3, calendar="360_day")},
)
get_clean_interp_index(da, "time")
def test_get_clean_interp_index_dt(cf_da, calendar, freq):
"""In the gregorian case, the index should be proportional to normal datetimes."""
g = cf_da(calendar, freq=freq)
g["stime"] = xr.Variable(data=g.time.to_index().to_datetimeindex(), dims=("time",))
gi = get_clean_interp_index(g, "time")
si = get_clean_interp_index(g, "time", use_coordinate="stime")
np.testing.assert_array_equal(gi, si)
def test_get_clean_interp_index_strict(index):
da = xr.DataArray([0, 1, 2], dims=("x", ), coords={"x": index})
with pytest.raises(ValueError):
get_clean_interp_index(da, "x")
clean = get_clean_interp_index(da, "x", strict=False)
np.testing.assert_array_equal(index, clean)
assert clean.dtype == np.float64
def test_interpolate_na_nan_block_lengths(y, lengths):
arr = [[np.nan, np.nan, np.nan, 1, np.nan, np.nan, 4, np.nan, np.nan]]
da = xr.DataArray(arr * 2, dims=["x", "y"], coords={"x": [0, 1], "y": y})
index = get_clean_interp_index(da, dim="y", use_coordinate=True)
actual = _get_nan_block_lengths(da, dim="y", index=index)
expected = da.copy(data=lengths * 2)
assert_equal(actual, expected)
def test_get_clean_interp_index_cf_calendar(cf_da, calendar):
"""The index for CFTimeIndex is in units of days. This means that if two series using a 360 and 365 days
calendar each have a trend of .01C/year, the linear regression coefficients will be different because they
have different number of days.
Another option would be to have an index in units of years, but this would likely create other difficulties.
"""
i = get_clean_interp_index(cf_da(calendar), dim="time")
np.testing.assert_array_equal(i, np.arange(10) * 1e9 * 86400)