Fix pandas deprecation warnings

Author: spencerkclark

xarray/coding/calendar_ops.py

@@ -249,7 +249,7 @@ def _interpolate_day_of_year(times, target_calendar):
     source_calendar = times.dt.calendar
     return np.round(
         _days_in_year(times.dt.year, target_calendar)
-        * times.dt.dayofyear
+        * times.dt.day_of_year
         / _days_in_year(times.dt.year, source_calendar)
     ).astype(int)
 
@@ -272,7 +272,7 @@ def _random_day_of_year(time, target_calendar, use_cftime):
         new_doy = np.insert(new_doy, rm_idx - np.arange(5), -1)
         if _days_in_year(year, source_calendar) == 366:
             new_doy = np.insert(new_doy, 60, -1)
-    return new_doy[time.dt.dayofyear - 1]
+    return new_doy[time.dt.day_of_year - 1]
 
 
 def _convert_to_new_calendar_with_new_day_of_year(

xarray/coding/cftime_offsets.py

@@ -1623,10 +1623,18 @@ def date_range_like(source, calendar, use_cftime=None):
     end = convert_time_or_go_back(source_end, date_type)
 
     # For the cases where the source ends on the end of the month, we expect the same in the new calendar.
-    if source_end.day == source_end.daysinmonth and isinstance(
+    if isinstance(source_end, pd.Timestamp):
+        source_end_days_in_month = source_end.days_in_month
+    else:
+        source_end_days_in_month = source_end.daysinmonth
+    if isinstance(end, pd.Timestamp):
+        end_days_in_month = end.days_in_month
+    else:
+        end_days_in_month = end.daysinmonth
+    if source_end.day == source_end_days_in_month and isinstance(
         freq_as_offset, YearEnd | QuarterEnd | MonthEnd | Day
     ):
-        end = end.replace(day=end.daysinmonth)
+        end = end.replace(day=end_days_in_month)
 
     return date_range(
         start=start.isoformat(),

xarray/coding/cftimeindex.py

@@ -127,7 +127,12 @@ def get_date_field(datetimes, field):
     return np.array([getattr(date, field) for date in datetimes], dtype=np.int64)
 
 
-def _field_accessor(name, docstring=None, min_cftime_version="0.0"):
+def _field_accessor(
+    name: str,
+    docstring: str | None = None,
+    min_cftime_version: str = "0.0",
+    deprecation_pair: tuple[str, str] | None = None,
+):
     """Adapted from pandas.tseries.index._field_accessor"""
 
     def f(self, min_cftime_version=min_cftime_version):
@@ -136,6 +141,14 @@ def f(self, min_cftime_version=min_cftime_version):
         else:
             cftime = attempt_import("cftime")
 
+        if deprecation_pair is not None:
+            original, replacement = deprecation_pair
+            emit_user_level_warning(
+                f"CFTimeIndex.{original} is deprecated and will be removed in "
+                f"a future version. Use CFTimeIndex.{replacement} instead",
+                FutureWarning,
+            )
+
         if Version(cftime.__version__) >= Version(min_cftime_version):
             return get_date_field(self._data, name)
         else:
@@ -249,9 +262,23 @@ class CFTimeIndex(pd.Index):
     second = _field_accessor("second", "The seconds of the datetime")
     microsecond = _field_accessor("microsecond", "The microseconds of the datetime")
     dayofyear = _field_accessor(
+        "dayofyr",
+        "The ordinal day of year of the datetime",
+        "1.0.2.1",
+        ("dayofyear", "day_of_year"),
+    )
+    dayofweek = _field_accessor(
+        "dayofwk",
+        "The day of week of the datetime",
+        "1.0.2.1",
+        ("dayofweek", "day_of_week"),
+    )
+    day_of_year = _field_accessor(
         "dayofyr", "The ordinal day of year of the datetime", "1.0.2.1"
     )
-    dayofweek = _field_accessor("dayofwk", "The day of week of the datetime", "1.0.2.1")
+    day_of_week = _field_accessor(
+        "dayofwk", "The day of week of the datetime", "1.0.2.1"
+    )
     days_in_month = _field_accessor(
         "daysinmonth", "The number of days in the month of the datetime", "1.1.0.0"
     )

xarray/core/accessor_dt.py

@@ -16,6 +16,7 @@
     is_np_timedelta_like,
 )
 from xarray.core.types import T_DataArray
+from xarray.core.utils import emit_user_level_warning
 from xarray.core.variable import IndexVariable, Variable
 from xarray.namedarray.utils import is_duck_dask_array
 
@@ -465,17 +466,45 @@ def weekofyear(self) -> DataArray:
 
     week = weekofyear
 
+    @property
+    def day_of_week(self) -> T_DataArray:
+        """The day of the week with Monday=0, Sunday=6"""
+        return self._date_field("day_of_week", np.int64)
+
     @property
     def dayofweek(self) -> T_DataArray:
         """The day of the week with Monday=0, Sunday=6"""
-        return self._date_field("dayofweek", np.int64)
+        emit_user_level_warning(
+            "dt.dayofweek is deprecated and will be removed in a future "
+            "version. Use dt.day_of_week instead.",
+            FutureWarning,
+        )
+        return self._date_field("day_of_week", np.int64)
 
-    weekday = dayofweek
+    @property
+    def weekday(self) -> T_DataArray:
+        """The day of the week with Monday=0, Sunday=6"""
+        emit_user_level_warning(
+            "dt.weekday is deprecated and will be removed in a "
+            "future version. Use dt.day_of_week instead.",
+            FutureWarning,
+        )
+        return self._date_field("day_of_week", np.int64)
+
+    @property
+    def day_of_year(self) -> T_DataArray:
+        """The ordinal day of the year"""
+        return self._date_field("day_of_year", np.int64)
 
     @property
     def dayofyear(self) -> T_DataArray:
         """The ordinal day of the year"""
-        return self._date_field("dayofyear", np.int64)
+        emit_user_level_warning(
+            "dt.dayofyear is deprecated and will be removed in a future "
+            "version. Use dt.day_of_year instead.",
+            FutureWarning,
+        )
+        return self._date_field("day_of_year", np.int64)
 
     @property
     def quarter(self) -> T_DataArray:
@@ -487,7 +516,15 @@ def days_in_month(self) -> T_DataArray:
         """The number of days in the month"""
         return self._date_field("days_in_month", np.int64)
 
-    daysinmonth = days_in_month
+    @property
+    def daysinmonth(self) -> T_DataArray:
+        """The number of days in the month"""
+        emit_user_level_warning(
+            "dt.daysinmonth is deprecated and will be removed in a future "
+            "version. Use dt.days_in_month instead.",
+            FutureWarning,
+        )
+        return self._date_field("days_in_month", np.int64)
 
     @property
     def season(self) -> T_DataArray:

xarray/tests/test_accessor_dt.py

@@ -56,13 +56,11 @@ def setup(self):
             "nanosecond",
             "week",
             "weekofyear",
-            "dayofweek",
-            "weekday",
-            "dayofyear",
+            "day_of_week",
+            "day_of_year",
             "quarter",
             "date",
             "time",
-            "daysinmonth",
             "days_in_month",
             "is_month_start",
             "is_month_end",
@@ -104,6 +102,21 @@ def test_field_access(self, field) -> None:
         assert expected.dtype == actual.dtype
         assert_identical(expected, actual)
 
+    @pytest.mark.parametrize(
+        ("field", "replacement"),
+        [
+            ("daysinmonth", "days_in_month"),
+            ("dayofweek", "day_of_week"),
+            ("weekday", "day_of_week"),
+            ("dayofyear", "day_of_year"),
+        ],
+    )
+    def test_deprecated_field_access(self, field, replacement) -> None:
+        expected = getattr(self.data.time.dt, replacement)
+        with pytest.warns(FutureWarning, match=f"{field}.*{replacement}"):
+            actual = getattr(self.data.time.dt, field)
+        assert_identical(expected, actual)
+
     def test_total_seconds(self) -> None:
         # Subtract a value in the middle of the range to ensure that some values
         # are negative
@@ -177,9 +190,8 @@ def test_not_datetime_type(self) -> None:
             "nanosecond",
             "week",
             "weekofyear",
-            "dayofweek",
-            "weekday",
-            "dayofyear",
+            "day_of_week",
+            "day_of_year",
             "quarter",
             "date",
             "time",
@@ -441,7 +453,7 @@ def times_3d(times):
 
 @requires_cftime
 @pytest.mark.parametrize(
-    "field", ["year", "month", "day", "hour", "dayofyear", "dayofweek"]
+    "field", ["year", "month", "day", "hour", "day_of_year", "day_of_week"]
 )
 def test_field_access(data, field) -> None:
     result = getattr(data.time.dt, field)
@@ -533,7 +545,7 @@ def test_cftime_strftime_access(data) -> None:
 @requires_cftime
 @requires_dask
 @pytest.mark.parametrize(
-    "field", ["year", "month", "day", "hour", "dayofyear", "dayofweek"]
+    "field", ["year", "month", "day", "hour", "day_of_year", "day_of_week"]
 )
 def test_dask_field_access_1d(data, field) -> None:
     import dask.array as da
@@ -553,7 +565,7 @@ def test_dask_field_access_1d(data, field) -> None:
 @requires_cftime
 @requires_dask
 @pytest.mark.parametrize(
-    "field", ["year", "month", "day", "hour", "dayofyear", "dayofweek"]
+    "field", ["year", "month", "day", "hour", "day_of_year", "day_of_week"]
 )
 def test_dask_field_access(times_3d, data, field) -> None:
     import dask.array as da

xarray/tests/test_calendar_ops.py

@@ -159,7 +159,7 @@ def test_convert_calendar_360_days_random():
     # Ensure that added days are evenly distributed in the 5 fifths of each year
     conv = convert_calendar(da_360, "noleap", align_on="random", missing=np.nan)
     conv = conv.where(conv.isnull(), drop=True)
-    nandoys = conv.time.dt.dayofyear[:366]
+    nandoys = conv.time.dt.day_of_year[:366]
     assert all(nandoys < np.array([74, 147, 220, 293, 366]))
     assert all(nandoys > np.array([0, 73, 146, 219, 292]))
 

xarray/tests/test_cftime_offsets.py

@@ -1380,19 +1380,19 @@ def test_calendar_year_length(
 
 @pytest.mark.parametrize("freq", ["YE", "ME", "D"])
 def test_dayofweek_after_cftime(freq: str) -> None:
-    result = date_range("2000-02-01", periods=3, freq=freq, use_cftime=True).dayofweek
+    result = date_range("2000-02-01", periods=3, freq=freq, use_cftime=True).day_of_week
     # TODO: remove once requiring pandas 2.2+
     freq = _new_to_legacy_freq(freq)
-    expected = pd.date_range("2000-02-01", periods=3, freq=freq).dayofweek
+    expected = pd.date_range("2000-02-01", periods=3, freq=freq).day_of_week
     np.testing.assert_array_equal(result, expected)
 
 
 @pytest.mark.parametrize("freq", ["YE", "ME", "D"])
 def test_dayofyear_after_cftime(freq: str) -> None:
-    result = date_range("2000-02-01", periods=3, freq=freq, use_cftime=True).dayofyear
+    result = date_range("2000-02-01", periods=3, freq=freq, use_cftime=True).day_of_year
     # TODO: remove once requiring pandas 2.2+
     freq = _new_to_legacy_freq(freq)
-    expected = pd.date_range("2000-02-01", periods=3, freq=freq).dayofyear
+    expected = pd.date_range("2000-02-01", periods=3, freq=freq).day_of_year
     np.testing.assert_array_equal(result, expected)
 
 

xarray/tests/test_cftimeindex.py

@@ -315,8 +315,8 @@ def test_cftimeindex_field_accessors(index, field, expected):
         "minute",
         "second",
         "microsecond",
-        "dayofyear",
-        "dayofweek",
+        "day_of_year",
+        "day_of_week",
         "days_in_month",
     ],
 )
@@ -329,16 +329,30 @@ def test_empty_cftimeindex_field_accessors(field):
 
 
 @requires_cftime
-def test_cftimeindex_dayofyear_accessor(index):
-    result = index.dayofyear
+@pytest.mark.parametrize(
+    ("field", "replacement"), [("day_of_year", None), ("dayofyear", "day_of_year")]
+)
+def test_cftimeindex_dayofyear_accessor(index, field, replacement):
+    if replacement is not None:
+        with pytest.warns(FutureWarning, match=f"{field}.*{replacement}"):
+            result = getattr(index, field)
+    else:
+        result = getattr(index, field)
     expected = np.array([date.dayofyr for date in index], dtype=np.int64)
     assert_array_equal(result, expected)
     assert result.dtype == expected.dtype
 
 
 @requires_cftime
-def test_cftimeindex_dayofweek_accessor(index):
-    result = index.dayofweek
+@pytest.mark.parametrize(
+    ("field", "replacement"), [("day_of_week", None), ("dayofweek", "day_of_week")]
+)
+def test_cftimeindex_dayofweek_accessor(index, field, replacement):
+    if replacement is not None:
+        with pytest.warns(FutureWarning, match=f"{field}.*{replacement}"):
+            result = getattr(index, field)
+    else:
+        result = getattr(index, field)
     expected = np.array([date.dayofwk for date in index], dtype=np.int64)
     assert_array_equal(result, expected)
     assert result.dtype == expected.dtype

xarray/tests/test_dataarray.py

@@ -1481,8 +1481,8 @@ def test_virtual_time_components(self) -> None:
         dates = pd.date_range("2000-01-01", periods=10)
         da = DataArray(np.arange(1, 11), [("time", dates)])
 
-        assert_array_equal(da["time.dayofyear"], da.values)
-        assert_array_equal(da.coords["time.dayofyear"], da.values)
+        assert_array_equal(da["time.day_of_year"], da.values)
+        assert_array_equal(da.coords["time.day_of_year"], da.values)
 
     def test_coords(self) -> None:
         # use int64 to ensure repr() consistency on windows

xarray/tests/test_dataset.py

@@ -1729,7 +1729,7 @@ def test_sel(self) -> None:
         times = pd.date_range("2000-01-01", periods=3)
         assert_equal(data.isel(time=slice(3)), data.sel(time=times))
         assert_equal(
-            data.isel(time=slice(3)), data.sel(time=(data["time.dayofyear"] <= 3))
+            data.isel(time=slice(3)), data.sel(time=(data["time.day_of_year"] <= 3))
         )
 
         td = pd.to_timedelta(np.arange(3), unit="days")
@@ -4572,11 +4572,11 @@ def test_virtual_variables_time(self) -> None:
         assert_array_equal(data["time.month"].values, index.month)
         assert_array_equal(data["time.season"].values, "DJF")
         # test virtual variable math
-        assert_array_equal(data["time.dayofyear"] + 1, 2 + np.arange(20))
-        assert_array_equal(np.sin(data["time.dayofyear"]), np.sin(1 + np.arange(20)))
+        assert_array_equal(data["time.day_of_year"] + 1, 2 + np.arange(20))
+        assert_array_equal(np.sin(data["time.day_of_year"]), np.sin(1 + np.arange(20)))
         # ensure they become coordinates
-        expected = Dataset({}, {"dayofyear": data["time.dayofyear"]})
-        actual = data[["time.dayofyear"]]
+        expected = Dataset({}, {"day_of_year": data["time.day_of_year"]})
+        actual = data[["time.day_of_year"]]
         assert_equal(expected, actual)
         # non-coordinate variables
         ds = Dataset({"t": ("x", pd.date_range("2000-01-01", periods=3))})
@@ -4599,9 +4599,10 @@ def test_time_season(self) -> None:
     def test_slice_virtual_variable(self) -> None:
         data = create_test_data()
         assert_equal(
-            data["time.dayofyear"][:10].variable, Variable(["time"], 1 + np.arange(10))
+            data["time.day_of_year"][:10].variable,
+            Variable(["time"], 1 + np.arange(10)),
         )
-        assert_equal(data["time.dayofyear"][0].variable, Variable([], 1))
+        assert_equal(data["time.day_of_year"][0].variable, Variable([], 1))
 
     def test_setitem(self) -> None:
         # assign a variable