REF: avoid _with_infer constructor (#50001)

Author: jbrockmendel

pandas/_testing/__init__.py

@@ -274,7 +274,7 @@ def box_expected(expected, box_cls, transpose: bool = True):
  else:
  expected = pd.array(expected, copy=False)
  elif box_cls is Index:
- expected = Index._with_infer(expected)
+ expected = Index(expected)
  elif box_cls is Series:
  expected = Series(expected)
  elif box_cls is DataFrame:

pandas/core/algorithms.py

@@ -894,7 +894,7 @@ def value_counts(
 
  # For backwards compatibility, we let Index do its normal type
  # inference, _except_ for if if infers from object to bool.
- idx = Index._with_infer(keys)
+ idx = Index(keys)
  if idx.dtype == bool and keys.dtype == object:
  idx = idx.astype(object)
 

pandas/core/indexes/base.py

@@ -2678,6 +2678,7 @@ def fillna(self, value=None, downcast=None):
  if downcast is None:
  # no need to care metadata other than name
  # because it can't have freq if it has NaTs
+ # _with_infer needed for test_fillna_categorical
  return Index._with_infer(result, name=self.name)
  raise NotImplementedError(
  f"{type(self).__name__}.fillna does not support 'downcast' "
@@ -4230,10 +4231,10 @@ def _reindex_non_unique(
  new_indexer = np.arange(len(self.take(indexer)), dtype=np.intp)
  new_indexer[~check] = -1
 
- if isinstance(self, ABCMultiIndex):
- new_index = type(self).from_tuples(new_labels, names=self.names)
+ if not isinstance(self, ABCMultiIndex):
+ new_index = Index(new_labels, name=self.name)
  else:
- new_index = Index._with_infer(new_labels, name=self.name)
+ new_index = type(self).from_tuples(new_labels, names=self.names)
  return new_index, indexer, new_indexer
 
  # --------------------------------------------------------------------
@@ -6477,7 +6478,7 @@ def insert(self, loc: int, item) -> Index:
  if self._typ == "numericindex":
  # Use self._constructor instead of Index to retain NumericIndex GH#43921
  # TODO(2.0) can use Index instead of self._constructor
- return self._constructor._with_infer(new_values, name=self.name)
+ return self._constructor(new_values, name=self.name)
  else:
  return Index._with_infer(new_values, name=self.name)
 
@@ -6850,7 +6851,7 @@ def ensure_index_from_sequences(sequences, names=None) -> Index:
  if len(sequences) == 1:
  if names is not None:
  names = names[0]
- return Index._with_infer(sequences[0], name=names)
+ return Index(sequences[0], name=names)
  else:
  return MultiIndex.from_arrays(sequences, names=names)
 
@@ -6893,7 +6894,7 @@ def ensure_index(index_like: Axes, copy: bool = False) -> Index:
 
  if isinstance(index_like, ABCSeries):
  name = index_like.name
- return Index._with_infer(index_like, name=name, copy=copy)
+ return Index(index_like, name=name, copy=copy)
 
  if is_iterator(index_like):
  index_like = list(index_like)
@@ -6909,9 +6910,9 @@ def ensure_index(index_like: Axes, copy: bool = False) -> Index:
 
  return MultiIndex.from_arrays(index_like)
  else:
- return Index._with_infer(index_like, copy=copy, tupleize_cols=False)
+ return Index(index_like, copy=copy, tupleize_cols=False)
  else:
- return Index._with_infer(index_like, copy=copy)
+ return Index(index_like, copy=copy)
 
 
 def ensure_has_len(seq):

pandas/core/indexes/multi.py

@@ -2112,7 +2112,7 @@ def append(self, other):
  # setting names to None automatically
  return MultiIndex.from_tuples(new_tuples)
  except (TypeError, IndexError):
- return Index._with_infer(new_tuples)
+ return Index(new_tuples)
 
  def argsort(self, *args, **kwargs) -> npt.NDArray[np.intp]:
  if len(args) == 0 and len(kwargs) == 0:

pandas/core/strings/accessor.py

@@ -319,7 +319,7 @@ def cons_row(x):
  out = out.get_level_values(0)
  return out
  else:
- return Index._with_infer(result, name=name)
+ return Index(result, name=name)
  else:
  index = self._orig.index
  # This is a mess.

pandas/core/util/hashing.py

@@ -344,9 +344,7 @@ def _hash_ndarray(
  )
 
  codes, categories = factorize(vals, sort=False)
- cat = Categorical(
- codes, Index._with_infer(categories), ordered=False, fastpath=True
- )
+ cat = Categorical(codes, Index(categories), ordered=False, fastpath=True)
  return _hash_categorical(cat, encoding, hash_key)
 
  try:

pandas/tests/arithmetic/test_numeric.py

@@ -1147,6 +1147,9 @@ def test_numarr_with_dtype_add_nan(self, dtype, box_with_array):
 
  ser = tm.box_expected(ser, box)
  expected = tm.box_expected(expected, box)
+ if box is Index and dtype is object:
+ # TODO: avoid this; match behavior with Series
+ expected = expected.astype(np.float64)
 
  result = np.nan + ser
  tm.assert_equal(result, expected)
@@ -1162,6 +1165,9 @@ def test_numarr_with_dtype_add_int(self, dtype, box_with_array):
 
  ser = tm.box_expected(ser, box)
  expected = tm.box_expected(expected, box)
+ if box is Index and dtype is object:
+ # TODO: avoid this; match behavior with Series
+ expected = expected.astype(np.int64)
 
  result = 1 + ser
  tm.assert_equal(result, expected)

pandas/tests/arrays/integer/test_dtypes.py

@@ -89,7 +89,7 @@ def test_astype_index(all_data, dropna):
  other = all_data
 
  dtype = all_data.dtype
- idx = pd.Index._with_infer(np.array(other))
+ idx = pd.Index(np.array(other))
  assert isinstance(idx, ABCIndex)
 
  result = idx.astype(dtype)

pandas/tests/extension/base/groupby.py

@@ -33,7 +33,7 @@ def test_groupby_extension_agg(self, as_index, data_for_grouping):
  _, uniques = pd.factorize(data_for_grouping, sort=True)
 
  if as_index:
- index = pd.Index._with_infer(uniques, name="B")
+ index = pd.Index(uniques, name="B")
  expected = pd.Series([3.0, 1.0, 4.0], index=index, name="A")
  self.assert_series_equal(result, expected)
  else:
@@ -61,7 +61,7 @@ def test_groupby_extension_no_sort(self, data_for_grouping):
  result = df.groupby("B", sort=False).A.mean()
  _, index = pd.factorize(data_for_grouping, sort=False)
 
- index = pd.Index._with_infer(index, name="B")
+ index = pd.Index(index, name="B")
  expected = pd.Series([1.0, 3.0, 4.0], index=index, name="A")
  self.assert_series_equal(result, expected)
 

pandas/tests/extension/test_string.py

@@ -391,7 +391,7 @@ def test_groupby_extension_agg(self, as_index, data_for_grouping):
  _, uniques = pd.factorize(data_for_grouping, sort=True)
 
  if as_index:
- index = pd.Index._with_infer(uniques, name="B")
+ index = pd.Index(uniques, name="B")
  expected = pd.Series([3.0, 1.0, 4.0], index=index, name="A")
  self.assert_series_equal(result, expected)
  else: