fix linting issues raised from new pylint and codespell versions (#1772)

### What kind of change does this PR introduce?

* Fixes new errors raised from the newer versions of `pylint` and
`codespell`
* Sets the doctest examples to all use `h5netcdf` with separate caches
to load datasets

### Does this PR introduce a breaking change?

No.

### Other information:

https://pypi.org/project/pylint/3.2.3/
https://pypi.org/project/codespell/2.3.0/

CHANGES.rst

@@ -15,6 +15,8 @@ Internal changes
 * Synchronized tooling versions across ``pyproject.toml`` and ``tox.ini`` and pinned them to the latest stable releases in GitHub Workflows. (:pull:`1744`).
 * Fixed a few small spelling and grammar issues that were causing errors with `codespell`. Now ignoring `SVG` files. (:pull:`1769`).
 * Temporarily skipping the ``test_hawkins_sutton_smoke`` test due to strange behaviour with `xarray`. (:pull:`1769`).
+* Fixed some previously uncaught errors raised from recent versions of `pylint` and `codespell`. (:pull:`1772`).
+* Set the `doctest` examples to all use `h5netcdf` with worker-separated caches to load datasets. (:pull:`1772`).
 
 v0.49.0 (2024-05-02)
 --------------------

pyproject.toml

@@ -156,7 +156,7 @@ values = [
 
 [tool.codespell]
 skip = 'xclim/data/*.json,docs/_build,docs/notebooks/xclim_training/*.ipynb,docs/references.bib,__pycache__,*.gz,*.nc,*.png,*.svg,*.whl'
-ignore-words-list = "absolue,bloc,bui,callendar,degreee,environnement,hanel,inferrable,lond,nam,nd,ressources,sie,vas"
+ignore-words-list = "absolue,bloc,bui,callendar,degreee,environnement,hanel,inferrable,lond,nam,nd,ressources,socio-economic,sie,vas"
 
 [tool.coverage.run]
 relative_files = true

tests/test_calendar.py

@@ -90,6 +90,8 @@ def test_time_bnds_irregular(typ):
     elif typ == "pd":
         start = pd.date_range("1990-01-01", periods=24, freq="MS")
         end = pd.date_range("1990-01-01 23:59:59.999999999", periods=24, freq="ME")
+    else:
+        raise ValueError("`typ` must be 'pd' or 'xr'")
 
     time = start + (end - start) / 2
 

tests/test_indices.py

@@ -2971,6 +2971,8 @@ def test_rain_season(pr_series, result_type, method_dry_start):
     elif result_type == "end_cond_fails":
         pr[{"time": 99 + 20 - 1}] = 5
         out_exp = [3, np.NaN, 363]
+    else:
+        raise ValueError(f"Unknown result_type: {result_type}")
 
     out = {}
     out["start"], out["end"], out["length"] = xci.rain_season(

xclim/indices/_agro.py

@@ -506,7 +506,7 @@ def cool_night_index(
     Examples
     --------
     >>> from xclim.indices import cool_night_index
-    >>> tasmin = xr.open_dataset(path_to_tasmin_file).tasmin
+    >>> tasmin = open_dataset(path_to_tasmin_file).tasmin
     >>> cni = cool_night_index(tasmin)
 
     References
@@ -1170,7 +1170,7 @@ def standardized_precipitation_index(
     -------
     >>> from datetime import datetime
     >>> from xclim.indices import standardized_precipitation_index
-    >>> ds = xr.open_dataset(path_to_pr_file)
+    >>> ds = open_dataset(path_to_pr_file)
     >>> pr = ds.pr
     >>> cal_start, cal_end = "1990-05-01", "1990-08-31"
     >>> spi_3 = standardized_precipitation_index(

xclim/indices/_anuclim.py

@@ -128,7 +128,7 @@ def temperature_seasonality(
     The following would compute for each grid cell of file `tas.day.nc` the annual temperature seasonality:
 
     >>> import xclim.indices as xci
-    >>> t = xr.open_dataset(path_to_tas_file).tas
+    >>> t = open_dataset(path_to_tas_file).tas
     >>> tday_seasonality = xci.temperature_seasonality(t)
     >>> t_weekly = xci.tg_mean(t, freq="7D")
     >>> tweek_seasonality = xci.temperature_seasonality(t_weekly)
@@ -179,7 +179,7 @@ def precip_seasonality(pr: xarray.DataArray, freq: str = "YS") -> xarray.DataArr
     The following would compute for each grid cell of file `pr.day.nc` the annual precipitation seasonality:
 
     >>> import xclim.indices as xci
-    >>> p = xr.open_dataset(path_to_pr_file).pr
+    >>> p = open_dataset(path_to_pr_file).pr
     >>> pday_seasonality = xci.precip_seasonality(p)
     >>> p_weekly = xci.precip_accumulation(p, freq="7D")
 
@@ -242,7 +242,7 @@ def tg_mean_warmcold_quarter(
     warmest quarter mean temperature:
 
     >>> from xclim.indices import tg_mean_warmcold_quarter
-    >>> t = xr.open_dataset(path_to_tas_file)
+    >>> t = open_dataset(path_to_tas_file)
     >>> t_warm_qrt = tg_mean_warmcold_quarter(tas=t.tas, op="warmest")
 
     Notes
@@ -353,7 +353,7 @@ def prcptot_wetdry_quarter(
     The following would compute for each grid cell of file `pr.day.nc` the annual wettest quarter total precipitation:
 
     >>> from xclim.indices import prcptot_wetdry_quarter
-    >>> p = xr.open_dataset(path_to_pr_file)
+    >>> p = open_dataset(path_to_pr_file)
     >>> pr_warm_qrt = prcptot_wetdry_quarter(pr=p.pr, op="wettest")
 
     Notes

xclim/indices/_multivariate.py

@@ -127,7 +127,7 @@ def cold_spell_duration_index(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import cold_spell_duration_index
-    >>> tasmin = xr.open_dataset(path_to_tasmin_file).tasmin.isel(lat=0, lon=0)
+    >>> tasmin = open_dataset(path_to_tasmin_file).tasmin.isel(lat=0, lon=0)
     >>> tn10 = percentile_doy(tasmin, per=10).sel(percentiles=10)
     >>> cold_spell_duration_index(tasmin, tn10)
 
@@ -971,7 +971,7 @@ def precip_accumulation(
     precipitation at the seasonal frequency, ie DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import precip_accumulation
-    >>> pr_day = xr.open_dataset(path_to_pr_file).pr
+    >>> pr_day = open_dataset(path_to_pr_file).pr
     >>> prcp_tot_seasonal = precip_accumulation(pr_day, freq="QS-DEC")
     """
     if phase == "liquid":
@@ -1034,7 +1034,7 @@ def precip_average(
     precipitation at the seasonal frequency, ie DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import precip_average
-    >>> pr_day = xr.open_dataset(path_to_pr_file).pr
+    >>> pr_day = open_dataset(path_to_pr_file).pr
     >>> prcp_tot_seasonal = precip_average(pr_day, freq="QS-DEC")
     """
     if phase == "liquid":
@@ -1146,8 +1146,8 @@ def high_precip_low_temp(
     Example
     -------
     To compute the number of days with intense rainfall while minimum temperatures dip below -0.2C:
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
-    >>> tasmin = xr.open_dataset(path_to_tasmin_file).tasmin
+    >>> pr = open_dataset(path_to_pr_file).pr
+    >>> tasmin = open_dataset(path_to_tasmin_file).tasmin
     >>> high_precip_low_temp(
     ...     pr, tas=tasmin, pr_thresh="10 mm/d", tas_thresh="-0.2 degC"
     ... )
@@ -1204,7 +1204,7 @@ def days_over_precip_thresh(
     Examples
     --------
     >>> from xclim.indices import days_over_precip_thresh
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> p75 = pr.quantile(0.75, dim="time", keep_attrs=True)
     >>> r75p = days_over_precip_thresh(pr, p75)
     """
@@ -1333,7 +1333,7 @@ def tg90p(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import tg90p
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
     >>> tas_per = percentile_doy(tas, per=90).sel(percentiles=90)
     >>> hot_days = tg90p(tas, tas_per)
     """
@@ -1391,7 +1391,7 @@ def tg10p(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import tg10p
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
     >>> tas_per = percentile_doy(tas, per=10).sel(percentiles=10)
     >>> cold_days = tg10p(tas, tas_per)
     """
@@ -1449,7 +1449,7 @@ def tn90p(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import tn90p
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
     >>> tas_per = percentile_doy(tas, per=90).sel(percentiles=90)
     >>> hot_days = tn90p(tas, tas_per)
     """
@@ -1507,7 +1507,7 @@ def tn10p(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import tn10p
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
     >>> tas_per = percentile_doy(tas, per=10).sel(percentiles=10)
     >>> cold_days = tn10p(tas, tas_per)
     """
@@ -1565,7 +1565,7 @@ def tx90p(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import tx90p
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
     >>> tasmax_per = percentile_doy(tas, per=90).sel(percentiles=90)
     >>> hot_days = tx90p(tas, tasmax_per)
     """
@@ -1623,7 +1623,7 @@ def tx10p(
     --------
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import tx10p
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
     >>> tasmax_per = percentile_doy(tas, per=10).sel(percentiles=10)
     >>> cold_days = tx10p(tas, tasmax_per)
     """
@@ -1762,7 +1762,7 @@ def warm_spell_duration_index(
     >>> from xclim.core.calendar import percentile_doy
     >>> from xclim.indices import warm_spell_duration_index
 
-    >>> tasmax = xr.open_dataset(path_to_tasmax_file).tasmax.isel(lat=0, lon=0)
+    >>> tasmax = open_dataset(path_to_tasmax_file).tasmax.isel(lat=0, lon=0)
     >>> tasmax_per = percentile_doy(tasmax, per=90).sel(percentiles=90)
     >>> warm_spell_duration_index(tasmax, tasmax_per)
     """

xclim/indices/_simple.py

@@ -103,7 +103,7 @@ def tg_mean(tas: xarray.DataArray, freq: str = "YS") -> xarray.DataArray:
     at the seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import tg_mean
-    >>> t = xr.open_dataset(path_to_tas_file).tas
+    >>> t = open_dataset(path_to_tas_file).tas
     >>> tg = tg_mean(t, freq="QS-DEC")
     """
     return select_resample_op(tas, op="mean", freq=freq)
@@ -428,7 +428,7 @@ def max_1day_precipitation_amount(
     The following would compute for each grid cell the highest 1-day total at an annual frequency:
 
     >>> from xclim.indices import max_1day_precipitation_amount
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> rx1day = max_1day_precipitation_amount(pr, freq="YS")
     """
     return select_resample_op(pr, op="max", freq=freq)
@@ -462,7 +462,7 @@ def max_n_day_precipitation_amount(
     The following would compute for each grid cell the highest 5-day total precipitation at an annual frequency:
 
     >>> from xclim.indices import max_n_day_precipitation_amount
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> out = max_n_day_precipitation_amount(pr, window=5, freq="YS")
     """
     # Rolling sum of the values
@@ -499,7 +499,7 @@ def max_pr_intensity(
     The following would compute the maximum 6-hour precipitation intensity at an annual frequency:
 
     >>> from xclim.indices import max_pr_intensity
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> out = max_pr_intensity(pr, window=5, freq="YS")
     """
     # Rolling sum of the values
@@ -569,7 +569,7 @@ def sfcWind_max(  # noqa: N802
     at the seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import sfcWind_max
-    >>> fg = xr.open_dataset(path_to_sfcWind_file).sfcWind
+    >>> fg = open_dataset(path_to_sfcWind_file).sfcWind
     >>> fg_max = sfcWind_max(fg, freq="QS-DEC")
     """
     return sfcWind.resample(time=freq).max(dim="time").assign_attrs(units=sfcWind.units)
@@ -610,7 +610,7 @@ def sfcWind_mean(  # noqa: N802
     at the seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import sfcWind_mean
-    >>> fg = xr.open_dataset(path_to_sfcWind_file).sfcWind
+    >>> fg = open_dataset(path_to_sfcWind_file).sfcWind
     >>> fg_mean = sfcWind_mean(fg, freq="QS-DEC")
     """
     return (
@@ -653,7 +653,7 @@ def sfcWind_min(  # noqa: N802
     at the seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import sfcWind_min
-    >>> fg = xr.open_dataset(path_to_sfcWind_file).sfcWind
+    >>> fg = open_dataset(path_to_sfcWind_file).sfcWind
     >>> fg_min = sfcWind_min(fg, freq="QS-DEC")
     """
     return sfcWind.resample(time=freq).min(dim="time").assign_attrs(units=sfcWind.units)

xclim/indices/_threshold.py

@@ -773,7 +773,7 @@ def daily_pr_intensity(
     precipitation >= 5 mm at seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import daily_pr_intensity
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> daily_int = daily_pr_intensity(pr, thresh="5 mm/day", freq="QS-DEC")
     """
     t = convert_units_to(thresh, pr, "hydro")
@@ -1142,7 +1142,7 @@ def growing_season_length(
     Examples
     --------
     >>> from xclim.indices import growing_season_length
-    >>> tas = xr.open_dataset(path_to_tas_file).tas
+    >>> tas = open_dataset(path_to_tas_file).tas
 
     For the Northern Hemisphere:
 
@@ -1228,7 +1228,7 @@ def frost_season_length(
     Examples
     --------
     >>> from xclim.indices import frost_season_length
-    >>> tasmin = xr.open_dataset(path_to_tasmin_file).tasmin
+    >>> tasmin = open_dataset(path_to_tasmin_file).tasmin
 
     For the Northern Hemisphere:
 
@@ -1411,7 +1411,7 @@ def frost_free_season_length(
     Examples
     --------
     >>> from xclim.indices import frost_season_length
-    >>> tasmin = xr.open_dataset(path_to_tasmin_file).tasmin
+    >>> tasmin = open_dataset(path_to_tasmin_file).tasmin
 
     For the Northern Hemisphere:
 
@@ -2604,7 +2604,7 @@ def wetdays(
     at the seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import wetdays
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> wd = wetdays(pr, thresh="5 mm/day", freq="QS-DEC")
     """
     thresh = convert_units_to(thresh, pr, "hydro")
@@ -2646,7 +2646,7 @@ def wetdays_prop(
     5 mm at the seasonal frequency, i.e. DJF, MAM, JJA, SON, DJF, etc.:
 
     >>> from xclim.indices import wetdays_prop
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> wd = wetdays_prop(pr, thresh="5 mm/day", freq="QS-DEC")
     """
     thresh = convert_units_to(thresh, pr, "hydro")
@@ -3179,7 +3179,7 @@ def dry_spell_frequency(
     Examples
     --------
     >>> from xclim.indices import dry_spell_frequency
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> dsf = dry_spell_frequency(pr=pr, op="sum")
     >>> dsf = dry_spell_frequency(pr=pr, op="max")
     """
@@ -3372,7 +3372,7 @@ def wet_spell_frequency(
     Examples
     --------
     >>> from xclim.indices import wet_spell_frequency
-    >>> pr = xr.open_dataset(path_to_pr_file).pr
+    >>> pr = open_dataset(path_to_pr_file).pr
     >>> dsf = wet_spell_frequency(pr=pr, op="sum")
     >>> dsf = wet_spell_frequency(pr=pr, op="max")
     """

xclim/indices/fire/_cffwis.py

@@ -597,6 +597,8 @@ def _fire_season(
     elif method in ["LA08", "GFWED"]:
         # In LA08, the check INCLUDES the current day,
         start_index = max(temp_condition_days, snow_condition_days)
+    else:
+        raise ValueError("`method` must be one of 'WF93', 'LA08' or 'GFWED'.")
 
     for it in range(start_index, tas.shape[-1]):
         if method == "WF93":
@@ -627,6 +629,8 @@ def _fire_season(
 
             # Shut down when mean snow OR mean temp are over/under threshold
             shut_down = (msnow >= snow_thresh) | (mtemp < temp_end_thresh)
+        else:
+            raise ValueError("`method` must be one of 'WF93', 'LA08' or 'GFWED'.")
 
         # Mask is on if the previous days was on OR is there is a start-up,  AND if it's not a shut-down,
         # Aka is off if either the previous day was or it is a shut-down.
@@ -687,6 +691,10 @@ def _fire_weather_calc(  # noqa: C901
 
     overwintering = params["overwintering"]
     dry_start = params["dry_start"]
+
+    ow_DC = dc0.copy() if dc0 is not None else None
+    ow_DMC = dmc0.copy() if dmc0 is not None else None
+
     if overwintering and "DC" in ind_prevs:
         # In overwintering, dc0 is understood as the previous season's last DC code.
         ow_DC = dc0.copy()

xclim/indices/fire/_ffdi.py

@@ -144,7 +144,7 @@ def _griffiths_drought_factor(p, smd, lim, df):  # pragma: no cover
                 # N = 0 defines a rainfall event since 9am today,
                 # so doesn't apply here, where p is the rainfall
                 # over previous 24 hours.
-                x_ = N**1.3 / (N**1.3 + P - 2.0)
+                x_ = N**1.3 / (N**1.3 + P - 2.0)  # pylint: disable=E0601
                 x = min(x_, x)
 
                 conseq = 0

xclim/sdba/adjustment.py

@@ -959,14 +959,18 @@ def _compute_transform_matrix(reference, historical):
             # This step needs vectorize with dask, but vectorize doesn't work with dask, argh.
             # Invert to get transformation matrix from hist to PC coords.
             Hinv = np.linalg.inv(H)
-            # Fancy tricks to choose best orientation on each axes
+            # Fancy tricks to choose the best orientation on each axis.
             # (using eigenvectors, the output axes orientation is undefined)
             if best_orientation == "simple":
                 orient = best_pc_orientation_simple(R, Hinv)
             elif best_orientation == "full":
                 orient = best_pc_orientation_full(
                     R, Hinv, reference.mean(axis=1), historical.mean(axis=1), historical
                 )
+            else:
+                raise ValueError(
+                    f"Unknown `best_orientation` method: {best_orientation}."
+                )
             # Get transformation matrix
             return (R * orient) @ Hinv