Merge and upd restrict_by_wl

.cruft.json

@@ -11,7 +11,7 @@
       "project_slug": "xscen",
       "project_short_description": "A climate change scenario-building analysis framework, built with xclim/xarray.",
       "pypi_username": "RondeauG",
-      "version": "0.7.12-beta",
+      "version": "0.7.13-beta",
       "use_pytest": "y",
       "use_black": "y",
       "add_pyup_badge": "n",

HISTORY.rst

@@ -17,7 +17,8 @@ New features and enhancements
 * ``xs.spatial_mean`` with ``method='xESMF'`` will also automatically segmentize polygons (down to a 1° resolution) to ensure a correct average (:pull:`260`).
 * Added documentation for `require_all_on` in `search_data_catalogs`. (:pull:`263`).
 * ``xs.save_to_table`` and ``xs.io.to_table`` to transform datasets and arrays to DataFrames, but with support for multi-columns, multi-sheets and localized table of content generation.
-* Added annual global tas timeseries for CMIP6's models CMCC-ESM2 (ssp245, ssp370, ssp585), EC-Earth3-CC (ssp245, ssp585), KACE-1-0-G (ssp245, ssp370, ssp585) and TaiESM1 (ssp245, ssp370). (:issue:`268`, :pull:`270`).
+* Better ``xs.extract.resample`` : support for weighted resampling operations when starting with frequencies coarser than daily and missing timesteps/values handling. (:issue:`80`, :issue:`93`, :pull:`265`).
+* Added annual global tas timeseries for CMIP6's models CMCC-ESM2 (ssp245, ssp370, ssp585), EC-Earth3-CC (ssp245, ssp585), KACE-1-0-G (ssp245, ssp370, ssp585) and TaiESM1 (ssp245, ssp370). Moved global tas database to a netCDF file. (:issue:`268`, :pull:`270`).
 
 Breaking changes
 ^^^^^^^^^^^^^^^^

docs/goodtoknow.rst

@@ -37,12 +37,12 @@ There are many ways to open data in xscen workflows. The list below tries to mak
 Which function to use when resampling data
 ------------------------------------------
 
-:extract_dataset: :py:func:`~xscen.extract.extract_dataset` has resampling capabilities to provide daily data from finer sources.
+:extract_dataset: :py:func:`~xscen.extract.extract_dataset`'s resampling capabilities are meant to provide daily data from finer sources.
 
-:xclim indicators: Through :py:func:`~xscen.indicators.compute_indicators`, xscen workflows can easily use `xclim indicators <https://xclim.readthedocs.io/en/stable/indicators.html>`_
-    to go from daily data to coarser (monthly, seasonal, annual).
+:resample: :py:func`xscen.extract.resample` extends xarray's `resample` methods with support for weighted resampling when starting from data coarser than daily and for handling of missing timesteps or values.
 
-What is currently not covered by either `xscen` or `xclim` is a method to resample from data coarser than daily, where the base period is non-uniform (ex: resampling from monthly to annual data, taking into account the number of days per month).
+:xclim indicators: Through :py:func:`~xscen.indicators.compute_indicators`, xscen workflows can easily use `xclim indicators <https://xclim.readthedocs.io/en/stable/indicators.html>`_
+    to go from daily data to coarser (monthly, seasonal, annual), with missing values handling. This option will add more metadata than the two firsts.
 
 
 Metadata translation

setup.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.7.12-beta
+current_version = 0.7.13-beta
 commit = True
 tag = False
 parse = (?P<major>\d+)\.(?P<minor>\d+).(?P<patch>\d+)(\-(?P<release>[a-z]+))?

setup.py

@@ -102,6 +102,6 @@ def run(self):
     test_suite="tests",
     extras_require={"dev": dev_requirements},
     url="https://github.com/Ouranosinc/xscen",
-    version="0.7.12-beta",
+    version="0.7.13-beta",
     zip_safe=False,
 )

tests/test_extract.py

@@ -3,6 +3,7 @@
 import numpy as np
 import pandas as pd
 import pytest
+import xarray as xr
 from conftest import notebooks
 from xclim.testing.helpers import test_timeseries as timeseries
 
@@ -413,3 +414,83 @@ def test_outofrange(self):
 
     def test_none(self):
         assert xs.subset_warming_level(TestSubsetWarmingLevel.ds, wl=20) is None
+
+
+class TestResample:
+    @pytest.mark.parametrize(
+        "infrq,meth,outfrq,exp",
+        [
+            ["MS", "mean", "QS-DEC", [0.47457627, 3, 6.01086957, 9, 11.96666667]],
+            [
+                "QS-DEC",
+                "mean",
+                "YS",
+                [1.49041096, 5.49041096, 9.49453552, 13.49041096, 17.49041096],
+            ],
+            ["MS", "std", "2YS", [6.92009239, 6.91557206]],
+            [
+                "QS",
+                "median",
+                "YS",
+                [1.516437, 5.516437, 9.516437, 13.51092864, 17.516437],
+            ],
+        ],
+    )
+    def test_weighted(self, infrq, meth, outfrq, exp):
+        da = timeseries(
+            np.arange(48),
+            variable="tas",
+            start="2001-01-01",
+            freq=infrq,
+        )
+        out = xs.extract.resample(da, outfrq, method=meth)
+        np.testing.assert_allclose(out.isel(time=slice(0, 5)), exp)
+
+    def test_weighted_wind(self):
+        uas = timeseries(
+            np.arange(48),
+            variable="uas",
+            start="2001-01-01",
+            freq="MS",
+        )
+        vas = timeseries(
+            np.arange(48),
+            variable="vas",
+            start="2001-01-01",
+            freq="MS",
+        )
+        ds = xr.merge([uas, vas])
+        out = xs.extract.resample(ds.uas, "YS", method="wind_direction", ds=ds)
+        np.testing.assert_allclose(out, [5.5260274, 17.5260274, 29.5260274, 41.5136612])
+
+    def test_missing(self):
+        da = timeseries(
+            np.arange(48),
+            variable="tas",
+            start="2001-01-01",
+            freq="MS",
+        )
+        out = xs.extract.resample(da, "QS-DEC", method="mean", missing="drop")
+        assert out.size == 15
+
+        out = xs.extract.resample(da, "QS-DEC", method="mean", missing="mask")
+        assert out.isel(time=0).isnull().all()
+
+    def test_missing_xclim(self):
+        arr = np.arange(48).astype(float)
+        arr[0] = np.nan
+        arr[40:] = np.nan
+        da = timeseries(
+            arr,
+            variable="tas",
+            start="2001-01-01",
+            freq="MS",
+        )
+        out = xs.extract.resample(da, "YS", method="mean", missing={"method": "any"})
+        assert out.isel(time=0).isnull().all()
+
+        out = xs.extract.resample(
+            da, "YS", method="mean", missing={"method": "pct", "tolerance": 0.6}
+        )
+        assert out.isel(time=0).notnull().all()
+        assert out.isel(time=-1).isnull().all()

tests/test_xscen.py

@@ -28,4 +28,4 @@ def test_package_metadata(self):
         contents = f.read()
         assert """Gabriel Rondeau-Genesse""" in contents
         assert '__email__ = "[email protected]"' in contents
-        assert '__version__ = "0.7.12-beta"' in contents
+        assert '__version__ = "0.7.13-beta"' in contents

xscen/__init__.py

@@ -52,7 +52,7 @@
 
 __author__ = """Gabriel Rondeau-Genesse"""
 __email__ = "[email protected]"
-__version__ = "0.7.12-beta"
+__version__ = "0.7.13-beta"
 
 
 # monkeypatch so that warnings.warn() doesn't mention itself