CLN: dtypes.concat (#39572)

Author: jbrockmendel

pandas/core/dtypes/concat.py

@@ -1,7 +1,7 @@
 """
 Utility functions related to concat.
 """
-from typing import Set, cast
+from typing import cast
 
 import numpy as np
 
@@ -14,49 +14,13 @@
     is_extension_array_dtype,
     is_sparse,
 )
-from pandas.core.dtypes.generic import ABCCategoricalIndex, ABCRangeIndex, ABCSeries
+from pandas.core.dtypes.generic import ABCCategoricalIndex, ABCSeries
 
 from pandas.core.arrays import ExtensionArray
 from pandas.core.arrays.sparse import SparseArray
 from pandas.core.construction import array, ensure_wrapped_if_datetimelike
 
 
-def _get_dtype_kinds(arrays) -> Set[str]:
-    """
-    Parameters
-    ----------
-    arrays : list of arrays
-
-    Returns
-    -------
-    set[str]
-        A set of kinds that exist in this list of arrays.
-    """
-    typs: Set[str] = set()
-    for arr in arrays:
-        # Note: we use dtype.kind checks because they are much more performant
-        #  than is_foo_dtype
-
-        dtype = arr.dtype
-        if not isinstance(dtype, np.dtype):
-            # ExtensionDtype so we get
-            #  e.g. "categorical", "datetime64[ns, US/Central]", "Sparse[itn64, 0]"
-            typ = str(dtype)
-        elif isinstance(arr, ABCRangeIndex):
-            typ = "range"
-        elif dtype.kind == "M":
-            typ = "datetime"
-        elif dtype.kind == "m":
-            typ = "timedelta"
-        elif dtype.kind in ["O", "b"]:
-            typ = str(dtype)  # i.e. "object", "bool"
-        else:
-            typ = dtype.kind
-
-        typs.add(typ)
-    return typs
-
-
 def _cast_to_common_type(arr: ArrayLike, dtype: DtypeObj) -> ArrayLike:
     """
     Helper function for `arr.astype(common_dtype)` but handling all special
@@ -130,8 +94,7 @@ def is_nonempty(x) -> bool:
     if non_empties and axis == 0:
         to_concat = non_empties
 
-    typs = _get_dtype_kinds(to_concat)
-    _contains_datetime = any(typ.startswith("datetime") for typ in typs)
+    kinds = {obj.dtype.kind for obj in to_concat}
 
     all_empty = not len(non_empties)
     single_dtype = len({x.dtype for x in to_concat}) == 1
@@ -150,17 +113,16 @@ def is_nonempty(x) -> bool:
         else:
             return np.concatenate(to_concat)
 
-    elif _contains_datetime or "timedelta" in typs:
+    elif any(kind in ["m", "M"] for kind in kinds):
         return _concat_datetime(to_concat, axis=axis)
 
     elif all_empty:
         # we have all empties, but may need to coerce the result dtype to
         # object if we have non-numeric type operands (numpy would otherwise
         # cast this to float)
-        typs = _get_dtype_kinds(to_concat)
-        if len(typs) != 1:
+        if len(kinds) != 1:
 
-            if not len(typs - {"i", "u", "f"}) or not len(typs - {"bool", "i", "u"}):
+            if not len(kinds - {"i", "u", "f"}) or not len(kinds - {"b", "i", "u"}):
                 # let numpy coerce
                 pass
             else:

pandas/tests/dtypes/test_concat.py

@@ -3,83 +3,10 @@
 import pandas.core.dtypes.concat as _concat
 
 import pandas as pd
-from pandas import DatetimeIndex, Period, PeriodIndex, Series, TimedeltaIndex
+from pandas import Series
 import pandas._testing as tm
 
 
-@pytest.mark.parametrize(
-    "to_concat, expected",
-    [
-        # int/float/str
-        ([["a"], [1, 2]], ["i", "object"]),
-        ([[3, 4], [1, 2]], ["i"]),
-        ([[3, 4], [1, 2.1]], ["i", "f"]),
-        # datetimelike
-        ([DatetimeIndex(["2011-01-01"]), DatetimeIndex(["2011-01-02"])], ["datetime"]),
-        ([TimedeltaIndex(["1 days"]), TimedeltaIndex(["2 days"])], ["timedelta"]),
-        # datetimelike object
-        (
-            [
-                DatetimeIndex(["2011-01-01"]),
-                DatetimeIndex(["2011-01-02"], tz="US/Eastern"),
-            ],
-            ["datetime", "datetime64[ns, US/Eastern]"],
-        ),
-        (
-            [
-                DatetimeIndex(["2011-01-01"], tz="Asia/Tokyo"),
-                DatetimeIndex(["2011-01-02"], tz="US/Eastern"),
-            ],
-            ["datetime64[ns, Asia/Tokyo]", "datetime64[ns, US/Eastern]"],
-        ),
-        ([TimedeltaIndex(["1 days"]), TimedeltaIndex(["2 hours"])], ["timedelta"]),
-        (
-            [
-                DatetimeIndex(["2011-01-01"], tz="Asia/Tokyo"),
-                TimedeltaIndex(["1 days"]),
-            ],
-            ["datetime64[ns, Asia/Tokyo]", "timedelta"],
-        ),
-    ],
-)
-def test_get_dtype_kinds(index_or_series, to_concat, expected):
-    to_concat_klass = [index_or_series(c) for c in to_concat]
-    result = _concat._get_dtype_kinds(to_concat_klass)
-    assert result == set(expected)
-
-
-@pytest.mark.parametrize(
-    "to_concat, expected",
-    [
-        (
-            [PeriodIndex(["2011-01"], freq="M"), PeriodIndex(["2011-01"], freq="M")],
-            ["period[M]"],
-        ),
-        (
-            [
-                Series([Period("2011-01", freq="M")]),
-                Series([Period("2011-02", freq="M")]),
-            ],
-            ["period[M]"],
-        ),
-        (
-            [PeriodIndex(["2011-01"], freq="M"), PeriodIndex(["2011-01"], freq="D")],
-            ["period[M]", "period[D]"],
-        ),
-        (
-            [
-                Series([Period("2011-01", freq="M")]),
-                Series([Period("2011-02", freq="D")]),
-            ],
-            ["period[M]", "period[D]"],
-        ),
-    ],
-)
-def test_get_dtype_kinds_period(to_concat, expected):
-    result = _concat._get_dtype_kinds(to_concat)
-    assert result == set(expected)
-
-
 def test_concat_mismatched_categoricals_with_empty():
     # concat_compat behavior on series._values should match pd.concat on series
     ser1 = Series(["a", "b", "c"], dtype="category")