PERF: DataFrame(dict) returns RangeIndex columns when possible (#57943)

* PERF: DataFrame(dict) returns RangeIndex columns when possible

* add whatsnew note

* Fix test failures

* Only 1 ndim

* Use infer_dtype

* Skip EA, skipna=False

Author: mroeschke

doc/source/whatsnew/v3.0.0.rst

@@ -267,6 +267,7 @@ Removal of prior version deprecations/changes
 Performance improvements
 ~~~~~~~~~~~~~~~~~~~~~~~~
 - :attr:`Categorical.categories` returns a :class:`RangeIndex` columns instead of an :class:`Index` if the constructed ``values`` was a ``range``. (:issue:`57787`)
+- :class:`DataFrame` returns a :class:`RangeIndex` columns when possible when ``data`` is a ``dict`` (:issue:`57943`)
 - :func:`concat` returns a :class:`RangeIndex` level in the :class:`MultiIndex` result when ``keys`` is a ``range`` or :class:`RangeIndex` (:issue:`57542`)
 - :meth:`RangeIndex.append` returns a :class:`RangeIndex` instead of a :class:`Index` when appending values that could continue the :class:`RangeIndex` (:issue:`57467`)
 - :meth:`Series.str.extract` returns a :class:`RangeIndex` columns instead of an :class:`Index` column when possible (:issue:`57542`)

pandas/core/indexes/api.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 
-import textwrap
 from typing import (
     TYPE_CHECKING,
     cast,
@@ -23,6 +22,7 @@
     ensure_index,
     ensure_index_from_sequences,
     get_unanimous_names,
+    maybe_sequence_to_range,
 )
 from pandas.core.indexes.category import CategoricalIndex
 from pandas.core.indexes.datetimes import DatetimeIndex
@@ -34,16 +34,6 @@
 
 if TYPE_CHECKING:
     from pandas._typing import Axis
-_sort_msg = textwrap.dedent(
-    """\
-Sorting because non-concatenation axis is not aligned. A future version
-of pandas will change to not sort by default.
-
-To accept the future behavior, pass 'sort=False'.
-
-To retain the current behavior and silence the warning, pass 'sort=True'.
-"""
-)
 
 
 __all__ = [
@@ -66,6 +56,7 @@
     "all_indexes_same",
     "default_index",
     "safe_sort_index",
+    "maybe_sequence_to_range",
 ]
 
 

pandas/core/indexes/base.py

@@ -7157,18 +7157,17 @@ def maybe_sequence_to_range(sequence) -> Any | range:
     -------
     Any : input or range
     """
-    if isinstance(sequence, (ABCSeries, Index, range)):
+    if isinstance(sequence, (ABCSeries, Index, range, ExtensionArray)):
         return sequence
-    np_sequence = np.asarray(sequence)
-    if np_sequence.dtype.kind != "i" or len(np_sequence) == 1:
+    elif len(sequence) == 1 or lib.infer_dtype(sequence, skipna=False) != "integer":
         return sequence
-    elif len(np_sequence) == 0:
+    elif len(sequence) == 0:
         return range(0)
-    diff = np_sequence[1] - np_sequence[0]
+    diff = sequence[1] - sequence[0]
     if diff == 0:
         return sequence
-    elif len(np_sequence) == 2 or lib.is_sequence_range(np_sequence, diff):
-        return range(np_sequence[0], np_sequence[-1] + diff, diff)
+    elif len(sequence) == 2 or lib.is_sequence_range(np.asarray(sequence), diff):
+        return range(sequence[0], sequence[-1] + diff, diff)
     else:
         return sequence
 

pandas/core/internals/construction.py

@@ -60,6 +60,7 @@
     default_index,
     ensure_index,
     get_objs_combined_axis,
+    maybe_sequence_to_range,
     union_indexes,
 )
 from pandas.core.internals.blocks import (
@@ -403,7 +404,7 @@ def dict_to_mgr(
                 arrays[i] = arr
 
     else:
-        keys = list(data.keys())
+        keys = maybe_sequence_to_range(list(data.keys()))
         columns = Index(keys) if keys else default_index(0)
         arrays = [com.maybe_iterable_to_list(data[k]) for k in keys]
 

pandas/tests/frame/test_constructors.py

@@ -2709,6 +2709,11 @@ def test_inference_on_pandas_objects(self):
         result = DataFrame({"a": ser})
         assert result.dtypes.iloc[0] == np.object_
 
+    def test_dict_keys_returns_rangeindex(self):
+        result = DataFrame({0: [1], 1: [2]}).columns
+        expected = RangeIndex(2)
+        tm.assert_index_equal(result, expected, exact=True)
+
 
 class TestDataFrameConstructorIndexInference:
     def test_frame_from_dict_of_series_overlapping_monthly_period_indexes(self):

pandas/tests/reshape/test_pivot.py

@@ -1738,6 +1738,7 @@ def test_daily(self):
             mask = ts.index.year == y
             expected[y] = Series(ts.values[mask], index=doy[mask])
         expected = DataFrame(expected, dtype=float).T
+        expected.index = expected.index.astype(np.int32)
         tm.assert_frame_equal(result, expected)
 
     def test_monthly(self):
@@ -1753,6 +1754,7 @@ def test_monthly(self):
             mask = ts.index.year == y
             expected[y] = Series(ts.values[mask], index=month[mask])
         expected = DataFrame(expected, dtype=float).T
+        expected.index = expected.index.astype(np.int32)
         tm.assert_frame_equal(result, expected)
 
     def test_pivot_table_with_iterator_values(self, data):