DEPR: Remove method and tolerance in Index.get_loc, bump xarray (#49630)

* DEPR: Remove method and tolerance in Index.get_loc

* note xarray bump

* Fix tests

* Fix refactor in period

* Lighter parameterization

* xfail xarray test

* Just use get_indexer

Author: mroeschke

ci/deps/actions-38-minimum_versions.yaml

@@ -54,7 +54,7 @@ dependencies:
   - sqlalchemy=1.4.16
   - tabulate=0.8.9
   - tzdata=2022a
-  - xarray=0.19.0
+  - xarray=0.21.0
   - xlrd=2.0.1
   - xlsxwriter=1.4.3
   - zstandard=0.15.2

doc/source/getting_started/install.rst

@@ -333,7 +333,7 @@ Installable with ``pip install "pandas[computation]"``.
 Dependency                Minimum Version    pip extra       Notes
 ========================= ================== =============== =============================================================
 SciPy                     1.7.1              computation     Miscellaneous statistical functions
-xarray                    0.19.0             computation     pandas-like API for N-dimensional data
+xarray                    0.21.0             computation     pandas-like API for N-dimensional data
 ========================= ================== =============== =============================================================
 
 Excel files

doc/source/whatsnew/v2.0.0.rst

@@ -339,6 +339,8 @@ Optional libraries below the lowest tested version may still work, but are not c
 +-----------------+-----------------+---------+
 | fastparquet     | 0.6.3           |    X    |
 +-----------------+-----------------+---------+
+| xarray          | 0.21.0          |    X    |
++-----------------+-----------------+---------+
 
 See :ref:`install.dependencies` and :ref:`install.optional_dependencies` for more.
 
@@ -521,6 +523,7 @@ Removal of prior version deprecations/changes
 - Removed the ``closed`` argument in :meth:`date_range` and :meth:`bdate_range` in favor of ``inclusive`` argument (:issue:`40245`)
 - Removed the ``center`` keyword in :meth:`DataFrame.expanding` (:issue:`20647`)
 - Removed the ``truediv`` keyword from :func:`eval` (:issue:`29812`)
+- Removed the ``method`` and ``tolerance`` arguments in :meth:`Index.get_loc`. Use ``index.get_indexer([label], method=..., tolerance=...)`` instead (:issue:`42269`)
 - Removed the ``pandas.datetime`` submodule (:issue:`30489`)
 - Removed the ``pandas.np`` submodule (:issue:`30296`)
 - Removed ``pandas.util.testing`` in favor of ``pandas.testing`` (:issue:`30745`)

pandas/compat/_optional.py

@@ -41,7 +41,7 @@
     "sqlalchemy": "1.4.16",
     "tables": "3.6.1",
     "tabulate": "0.8.9",
-    "xarray": "0.19.0",
+    "xarray": "0.21.0",
     "xlrd": "2.0.1",
     "xlsxwriter": "1.4.3",
     "zstandard": "0.15.2",

pandas/core/indexes/base.py

@@ -3429,27 +3429,13 @@ def _convert_can_do_setop(self, other) -> tuple[Index, Hashable]:
     # --------------------------------------------------------------------
     # Indexing Methods
 
-    def get_loc(self, key, method=None, tolerance=None):
+    def get_loc(self, key):
         """
         Get integer location, slice or boolean mask for requested label.
 
         Parameters
         ----------
         key : label
-        method : {None, 'pad'/'ffill', 'backfill'/'bfill', 'nearest'}, optional
-            * default: exact matches only.
-            * pad / ffill: find the PREVIOUS index value if no exact match.
-            * backfill / bfill: use NEXT index value if no exact match
-            * nearest: use the NEAREST index value if no exact match. Tied
-              distances are broken by preferring the larger index value.
-
-            .. deprecated:: 1.4
-                Use index.get_indexer([item], method=...) instead.
-
-        tolerance : int or float, optional
-            Maximum distance from index value for inexact matches. The value of
-            the index at the matching location must satisfy the equation
-            ``abs(index[loc] - key) <= tolerance``.
 
         Returns
         -------
@@ -3469,46 +3455,17 @@ def get_loc(self, key, method=None, tolerance=None):
         >>> non_monotonic_index.get_loc('b')
         array([False,  True, False,  True])
         """
-        if method is None:
-            if tolerance is not None:
-                raise ValueError(
-                    "tolerance argument only valid if using pad, "
-                    "backfill or nearest lookups"
-                )
-            casted_key = self._maybe_cast_indexer(key)
-            try:
-                return self._engine.get_loc(casted_key)
-            except KeyError as err:
-                raise KeyError(key) from err
-            except TypeError:
-                # If we have a listlike key, _check_indexing_error will raise
-                #  InvalidIndexError. Otherwise we fall through and re-raise
-                #  the TypeError.
-                self._check_indexing_error(key)
-                raise
-
-        # GH#42269
-        warnings.warn(
-            f"Passing method to {type(self).__name__}.get_loc is deprecated "
-            "and will raise in a future version. Use "
-            "index.get_indexer([item], method=...) instead.",
-            FutureWarning,
-            stacklevel=find_stack_level(),
-        )
-
-        if is_scalar(key) and isna(key) and not self.hasnans:
-            raise KeyError(key)
-
-        if tolerance is not None:
-            tolerance = self._convert_tolerance(tolerance, np.asarray(key))
-
-        indexer = self.get_indexer([key], method=method, tolerance=tolerance)
-        if indexer.ndim > 1 or indexer.size > 1:
-            raise TypeError("get_loc requires scalar valued input")
-        loc = indexer.item()
-        if loc == -1:
-            raise KeyError(key)
-        return loc
+        casted_key = self._maybe_cast_indexer(key)
+        try:
+            return self._engine.get_loc(casted_key)
+        except KeyError as err:
+            raise KeyError(key) from err
+        except TypeError:
+            # If we have a listlike key, _check_indexing_error will raise
+            #  InvalidIndexError. Otherwise we fall through and re-raise
+            #  the TypeError.
+            self._check_indexing_error(key)
+            raise
 
     _index_shared_docs[
         "get_indexer"

pandas/core/indexes/datetimes.py

@@ -556,7 +556,7 @@ def _disallow_mismatched_indexing(self, key) -> None:
         except TypeError as err:
             raise KeyError(key) from err
 
-    def get_loc(self, key, method=None, tolerance=None):
+    def get_loc(self, key):
         """
         Get integer location for requested label
 
@@ -587,8 +587,7 @@ def get_loc(self, key, method=None, tolerance=None):
                 try:
                     return self._partial_date_slice(reso, parsed)
                 except KeyError as err:
-                    if method is None:
-                        raise KeyError(key) from err
+                    raise KeyError(key) from err
 
             key = parsed
 
@@ -599,18 +598,14 @@ def get_loc(self, key, method=None, tolerance=None):
             )
 
         elif isinstance(key, dt.time):
-            if method is not None:
-                raise NotImplementedError(
-                    "cannot yet lookup inexact labels when key is a time object"
-                )
             return self.indexer_at_time(key)
 
         else:
             # unrecognized type
             raise KeyError(key)
 
         try:
-            return Index.get_loc(self, key, method, tolerance)
+            return Index.get_loc(self, key)
         except KeyError as err:
             raise KeyError(orig_key) from err
 

pandas/core/indexes/multi.py

@@ -2730,7 +2730,7 @@ def _get_loc_single_level_index(self, level_index: Index, key: Hashable) -> int:
         else:
             return level_index.get_loc(key)
 
-    def get_loc(self, key, method=None):
+    def get_loc(self, key):
         """
         Get location for a label or a tuple of labels.
 
@@ -2740,7 +2740,6 @@ def get_loc(self, key, method=None):
         Parameters
         ----------
         key : label or tuple of labels (one for each level)
-        method : None
 
         Returns
         -------
@@ -2772,12 +2771,6 @@ def get_loc(self, key, method=None):
         >>> mi.get_loc(('b', 'e'))
         1
         """
-        if method is not None:
-            raise NotImplementedError(
-                "only the default get_loc method is "
-                "currently supported for MultiIndex"
-            )
-
         self._check_indexing_error(key)
 
         def _maybe_to_slice(loc):

pandas/core/indexes/period.py

@@ -375,7 +375,7 @@ def _convert_tolerance(self, tolerance, target):
 
         return tolerance
 
-    def get_loc(self, key, method=None, tolerance=None):
+    def get_loc(self, key):
         """
         Get integer location for requested label.
 
@@ -421,10 +421,8 @@ def get_loc(self, key, method=None, tolerance=None):
                 # the reso < self._resolution_obj case goes
                 #  through _get_string_slice
                 key = self._cast_partial_indexing_scalar(parsed)
-            elif method is None:
-                raise KeyError(key)
             else:
-                key = self._cast_partial_indexing_scalar(parsed)
+                raise KeyError(key)
 
         elif isinstance(key, Period):
             self._disallow_mismatched_indexing(key)
@@ -437,7 +435,7 @@ def get_loc(self, key, method=None, tolerance=None):
             raise KeyError(key)
 
         try:
-            return Index.get_loc(self, key, method, tolerance)
+            return Index.get_loc(self, key)
         except KeyError as err:
             raise KeyError(orig_key) from err
 

pandas/core/indexes/range.py

@@ -328,17 +328,15 @@ def inferred_type(self) -> str:
     # Indexing Methods
 
     @doc(Int64Index.get_loc)
-    def get_loc(self, key, method=None, tolerance=None):
-        if method is None and tolerance is None:
-            if is_integer(key) or (is_float(key) and key.is_integer()):
-                new_key = int(key)
-                try:
-                    return self._range.index(new_key)
-                except ValueError as err:
-                    raise KeyError(key) from err
-            self._check_indexing_error(key)
-            raise KeyError(key)
-        return super().get_loc(key, method=method, tolerance=tolerance)
+    def get_loc(self, key):
+        if is_integer(key) or (is_float(key) and key.is_integer()):
+            new_key = int(key)
+            try:
+                return self._range.index(new_key)
+            except ValueError as err:
+                raise KeyError(key) from err
+        self._check_indexing_error(key)
+        raise KeyError(key)
 
     def _get_indexer(
         self,

pandas/core/indexes/timedeltas.py

@@ -174,7 +174,7 @@ def _is_comparable_dtype(self, dtype: DtypeObj) -> bool:
     # -------------------------------------------------------------------
     # Indexing Methods
 
-    def get_loc(self, key, method=None, tolerance=None):
+    def get_loc(self, key):
         """
         Get integer location for requested label
 
@@ -189,7 +189,7 @@ def get_loc(self, key, method=None, tolerance=None):
         except TypeError as err:
             raise KeyError(key) from err
 
-        return Index.get_loc(self, key, method, tolerance)
+        return Index.get_loc(self, key)
 
     def _parse_with_reso(self, label: str):
         # the "with_reso" is a no-op for TimedeltaIndex

pandas/tests/indexes/datetimes/test_indexing.py

@@ -8,8 +8,6 @@
 import numpy as np
 import pytest
 
-from pandas.errors import InvalidIndexError
-
 import pandas as pd
 from pandas import (
     DatetimeIndex,
@@ -405,75 +403,6 @@ def test_get_loc_key_unit_mismatch_not_castable(self):
 
         assert key not in dti
 
-    @pytest.mark.parametrize("method", [None, "pad", "backfill", "nearest"])
-    @pytest.mark.filterwarnings("ignore:Passing method:FutureWarning")
-    def test_get_loc_method_exact_match(self, method):
-        idx = date_range("2000-01-01", periods=3)
-        assert idx.get_loc(idx[1], method) == 1
-        assert idx.get_loc(idx[1].to_pydatetime(), method) == 1
-        assert idx.get_loc(str(idx[1]), method) == 1
-
-        if method is not None:
-            assert idx.get_loc(idx[1], method, tolerance=pd.Timedelta("0 days")) == 1
-
-    @pytest.mark.filterwarnings("ignore:Passing method:FutureWarning")
-    def test_get_loc(self):
-        idx = date_range("2000-01-01", periods=3)
-
-        assert idx.get_loc("2000-01-01", method="nearest") == 0
-        assert idx.get_loc("2000-01-01T12", method="nearest") == 1
-
-        assert idx.get_loc("2000-01-01T12", method="nearest", tolerance="1 day") == 1
-        assert (
-            idx.get_loc("2000-01-01T12", method="nearest", tolerance=pd.Timedelta("1D"))
-            == 1
-        )
-        assert (
-            idx.get_loc(
-                "2000-01-01T12", method="nearest", tolerance=np.timedelta64(1, "D")
-            )
-            == 1
-        )
-        assert (
-            idx.get_loc("2000-01-01T12", method="nearest", tolerance=timedelta(1)) == 1
-        )
-        with pytest.raises(ValueError, match="unit abbreviation w/o a number"):
-            idx.get_loc("2000-01-01T12", method="nearest", tolerance="foo")
-        with pytest.raises(KeyError, match="'2000-01-01T03'"):
-            idx.get_loc("2000-01-01T03", method="nearest", tolerance="2 hours")
-        with pytest.raises(
-            ValueError, match="tolerance size must match target index size"
-        ):
-            idx.get_loc(
-                "2000-01-01",
-                method="nearest",
-                tolerance=[
-                    pd.Timedelta("1day").to_timedelta64(),
-                    pd.Timedelta("1day").to_timedelta64(),
-                ],
-            )
-
-        assert idx.get_loc("2000", method="nearest") == slice(0, 3)
-        assert idx.get_loc("2000-01", method="nearest") == slice(0, 3)
-
-        assert idx.get_loc("1999", method="nearest") == 0
-        assert idx.get_loc("2001", method="nearest") == 2
-
-        with pytest.raises(KeyError, match="'1999'"):
-            idx.get_loc("1999", method="pad")
-        with pytest.raises(KeyError, match="'2001'"):
-            idx.get_loc("2001", method="backfill")
-
-        with pytest.raises(KeyError, match="'foobar'"):
-            idx.get_loc("foobar")
-        with pytest.raises(InvalidIndexError, match=r"slice\(None, 2, None\)"):
-            idx.get_loc(slice(2))
-
-        idx = DatetimeIndex(["2000-01-01", "2000-01-04"])
-        assert idx.get_loc("2000-01-02", method="nearest") == 0
-        assert idx.get_loc("2000-01-03", method="nearest") == 1
-        assert idx.get_loc("2000-01", method="nearest") == slice(0, 2)
-
     def test_get_loc_time_obj(self):
         # time indexing
         idx = date_range("2000-01-01", periods=24, freq="H")
@@ -486,11 +415,6 @@ def test_get_loc_time_obj(self):
         expected = np.array([])
         tm.assert_numpy_array_equal(result, expected, check_dtype=False)
 
-        msg = "cannot yet lookup inexact labels when key is a time object"
-        with pytest.raises(NotImplementedError, match=msg):
-            with tm.assert_produces_warning(FutureWarning, match="deprecated"):
-                idx.get_loc(time(12, 30), method="pad")
-
     def test_get_loc_time_obj2(self):
         # GH#8667
 
@@ -525,18 +449,6 @@ def test_get_loc_time_nat(self):
         expected = np.array([], dtype=np.intp)
         tm.assert_numpy_array_equal(loc, expected)
 
-    def test_get_loc_tz_aware(self):
-        # https://github.com/pandas-dev/pandas/issues/32140
-        dti = date_range(
-            Timestamp("2019-12-12 00:00:00", tz="US/Eastern"),
-            Timestamp("2019-12-13 00:00:00", tz="US/Eastern"),
-            freq="5s",
-        )
-        key = Timestamp("2019-12-12 10:19:25", tz="US/Eastern")
-        with tm.assert_produces_warning(FutureWarning, match="deprecated"):
-            result = dti.get_loc(key, method="nearest")
-        assert result == 7433
-
     def test_get_loc_nat(self):
         # GH#20464
         index = DatetimeIndex(["1/3/2000", "NaT"])