CLN: Remove collections.abc classes from pandas.compat (#25957)

Author: jschendel

ci/code_checks.sh

@@ -111,8 +111,10 @@ fi
 if [[ -z "$CHECK" || "$CHECK" == "patterns" ]]; then
 
     # Check for imports from pandas.core.common instead of `import pandas.core.common as com`
+    # Check for imports from collections.abc instead of `from collections import abc`
     MSG='Check for non-standard imports' ; echo $MSG
     invgrep -R --include="*.py*" -E "from pandas.core.common import " pandas
+    invgrep -R --include="*.py*" -E "from collections.abc import " pandas
     # invgrep -R --include="*.py*" -E "from numpy import nan " pandas  # GH#24822 not yet implemented since the offending imports have not all been removed
     RET=$(($RET + $?)) ; echo $MSG "DONE"
 

pandas/compat/__init__.py

@@ -29,7 +29,6 @@
 import struct
 import inspect
 from collections import namedtuple
-import collections
 
 PY2 = sys.version_info[0] == 2
 PY3 = sys.version_info[0] >= 3
@@ -106,16 +105,6 @@ def lmap(*args, **kwargs):
 
     def lfilter(*args, **kwargs):
         return list(filter(*args, **kwargs))
-
-    Hashable = collections.abc.Hashable
-    Iterable = collections.abc.Iterable
-    Iterator = collections.abc.Iterator
-    Mapping = collections.abc.Mapping
-    MutableMapping = collections.abc.MutableMapping
-    Sequence = collections.abc.Sequence
-    Sized = collections.abc.Sized
-    Set = collections.abc.Set
-
 else:
     # Python 2
     _name_re = re.compile(r"[a-zA-Z_][a-zA-Z0-9_]*$")
@@ -138,15 +127,6 @@ def signature(f):
     lmap = builtins.map
     lfilter = builtins.filter
 
-    Hashable = collections.Hashable
-    Iterable = collections.Iterable
-    Iterator = collections.Iterator
-    Mapping = collections.Mapping
-    MutableMapping = collections.MutableMapping
-    Sequence = collections.Sequence
-    Sized = collections.Sized
-    Set = collections.Set
-
 if PY2:
     def iteritems(obj, **kw):
         return obj.iteritems(**kw)

pandas/core/arrays/sparse.py

@@ -1,6 +1,7 @@
 """
 SparseArray data structure
 """
+from collections import abc
 import numbers
 import operator
 import re
@@ -1377,7 +1378,7 @@ def map(self, mapper):
         if isinstance(mapper, ABCSeries):
             mapper = mapper.to_dict()
 
-        if isinstance(mapper, compat.Mapping):
+        if isinstance(mapper, abc.Mapping):
             fill_value = mapper.get(self.fill_value, self.fill_value)
             sp_values = [mapper.get(x, None) for x in self.sp_values]
         else:

pandas/core/common.py

@@ -5,7 +5,7 @@
 """
 
 import collections
-from collections import OrderedDict
+from collections import OrderedDict, abc
 from datetime import datetime, timedelta
 from functools import partial
 import inspect
@@ -14,7 +14,6 @@
 import numpy as np
 
 from pandas._libs import lib, tslibs
-import pandas.compat as compat
 from pandas.compat import PY36, iteritems
 
 from pandas.core.dtypes.cast import construct_1d_object_array_from_listlike
@@ -374,13 +373,13 @@ def standardize_mapping(into):
 
     Parameters
     ----------
-    into : instance or subclass of collections.Mapping
+    into : instance or subclass of collections.abc.Mapping
         Must be a class, an initialized collections.defaultdict,
-        or an instance of a collections.Mapping subclass.
+        or an instance of a collections.abc.Mapping subclass.
 
     Returns
     -------
-    mapping : a collections.Mapping subclass or other constructor
+    mapping : a collections.abc.Mapping subclass or other constructor
         a callable object that can accept an iterator to create
         the desired Mapping.
 
@@ -394,7 +393,7 @@ def standardize_mapping(into):
             return partial(
                 collections.defaultdict, into.default_factory)
         into = type(into)
-    if not issubclass(into, compat.Mapping):
+    if not issubclass(into, abc.Mapping):
         raise TypeError('unsupported type: {into}'.format(into=into))
     elif into == collections.defaultdict:
         raise TypeError(
@@ -471,7 +470,7 @@ def _get_rename_function(mapper):
     Returns a function that will map names/labels, dependent if mapper
     is a dict, Series or just a function.
     """
-    if isinstance(mapper, (compat.Mapping, ABCSeries)):
+    if isinstance(mapper, (abc.Mapping, ABCSeries)):
 
         def f(x):
             if x in mapper:

pandas/core/dtypes/inference.py

@@ -1,14 +1,13 @@
 """ basic inference routines """
 
+from collections import abc
 from numbers import Number
 import re
 
 import numpy as np
 
 from pandas._libs import lib
-from pandas.compat import PY2, Set, re_type
-
-from pandas import compat
+from pandas.compat import PY2, re_type
 
 is_bool = lib.is_bool
 
@@ -113,7 +112,7 @@ def _iterable_not_string(obj):
     False
     """
 
-    return isinstance(obj, compat.Iterable) and not isinstance(obj, str)
+    return isinstance(obj, abc.Iterable) and not isinstance(obj, str)
 
 
 def is_iterator(obj):
@@ -288,15 +287,15 @@ def is_list_like(obj, allow_sets=True):
     False
     """
 
-    return (isinstance(obj, compat.Iterable) and
+    return (isinstance(obj, abc.Iterable) and
             # we do not count strings/unicode/bytes as list-like
             not isinstance(obj, (str, bytes)) and
 
             # exclude zero-dimensional numpy arrays, effectively scalars
             not (isinstance(obj, np.ndarray) and obj.ndim == 0) and
 
             # exclude sets if allow_sets is False
-            not (allow_sets is False and isinstance(obj, Set)))
+            not (allow_sets is False and isinstance(obj, abc.Set)))
 
 
 def is_array_like(obj):
@@ -436,23 +435,23 @@ def is_named_tuple(obj):
 def is_hashable(obj):
     """Return True if hash(obj) will succeed, False otherwise.
 
-    Some types will pass a test against collections.Hashable but fail when they
-    are actually hashed with hash().
+    Some types will pass a test against collections.abc.Hashable but fail when
+    they are actually hashed with hash().
 
     Distinguish between these and other types by trying the call to hash() and
     seeing if they raise TypeError.
 
     Examples
     --------
     >>> a = ([],)
-    >>> isinstance(a, collections.Hashable)
+    >>> isinstance(a, collections.abc.Hashable)
     True
     >>> is_hashable(a)
     False
     """
-    # Unfortunately, we can't use isinstance(obj, collections.Hashable), which
-    # can be faster than calling hash. That is because numpy scalars on Python
-    # 3 fail this test.
+    # Unfortunately, we can't use isinstance(obj, collections.abc.Hashable),
+    # which can be faster than calling hash. That is because numpy scalars
+    # fail this test.
 
     # Reconsider this decision once this numpy bug is fixed:
     # https://github.com/numpy/numpy/issues/5562

pandas/core/frame.py

@@ -11,7 +11,7 @@
 labeling information
 """
 import collections
-from collections import OrderedDict
+from collections import OrderedDict, abc
 import functools
 import itertools
 import sys
@@ -33,8 +33,7 @@
                                      validate_axis_style_args)
 
 from pandas import compat
-from pandas.compat import (
-    PY36, Iterator, StringIO, lmap, lzip, raise_with_traceback)
+from pandas.compat import PY36, StringIO, lmap, lzip, raise_with_traceback
 from pandas.compat.numpy import function as nv
 from pandas.core.dtypes.cast import (
     maybe_upcast,
@@ -426,9 +425,9 @@ def __init__(self, data=None, index=None, columns=None, dtype=None,
                                    copy=copy)
 
         # For data is list-like, or Iterable (will consume into list)
-        elif (isinstance(data, compat.Iterable) and
+        elif (isinstance(data, abc.Iterable) and
               not isinstance(data, (str, bytes))):
-            if not isinstance(data, compat.Sequence):
+            if not isinstance(data, abc.Sequence):
                 data = list(data)
             if len(data) > 0:
                 if is_list_like(data[0]) and getattr(data[0], 'ndim', 1) == 1:
@@ -1203,7 +1202,7 @@ def to_dict(self, orient='dict', into=dict):
             indicates `split`.
 
         into : class, default dict
-            The collections.Mapping subclass used for all Mappings
+            The collections.abc.Mapping subclass used for all Mappings
             in the return value.  Can be the actual class or an empty
             instance of the mapping type you want.  If you want a
             collections.defaultdict, you must pass it initialized.
@@ -1212,8 +1211,8 @@ def to_dict(self, orient='dict', into=dict):
 
         Returns
         -------
-        dict, list or collections.Mapping
-            Return a collections.Mapping object representing the DataFrame.
+        dict, list or collections.abc.Mapping
+            Return a collections.abc.Mapping object representing the DataFrame.
             The resulting transformation depends on the `orient` parameter.
 
         See Also
@@ -4080,7 +4079,7 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
             the same length as the calling DataFrame, or a list containing an
             arbitrary combination of column keys and arrays. Here, "array"
             encompasses :class:`Series`, :class:`Index`, ``np.ndarray``, and
-            instances of :class:`abc.Iterator`.
+            instances of :class:`~collections.abc.Iterator`.
         drop : bool, default True
             Delete columns to be used as the new index.
         append : bool, default False
@@ -4166,7 +4165,7 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
         missing = []
         for col in keys:
             if isinstance(col, (ABCIndexClass, ABCSeries, np.ndarray,
-                                list, Iterator)):
+                                list, abc.Iterator)):
                 # arrays are fine as long as they are one-dimensional
                 # iterators get converted to list below
                 if getattr(col, 'ndim', 1) != 1:
@@ -4213,7 +4212,7 @@ def set_index(self, keys, drop=True, append=False, inplace=False,
             elif isinstance(col, (list, np.ndarray)):
                 arrays.append(col)
                 names.append(None)
-            elif isinstance(col, Iterator):
+            elif isinstance(col, abc.Iterator):
                 arrays.append(list(col))
                 names.append(None)
             # from here, col can only be a column label

pandas/core/groupby/generic.py

@@ -6,7 +6,7 @@
 which here returns a DataFrameGroupBy object.
 """
 
-import collections
+from collections import OrderedDict, abc
 import copy
 from functools import partial
 from textwrap import dedent
@@ -226,7 +226,7 @@ def _aggregate_generic(self, func, *args, **kwargs):
         axis = self.axis
         obj = self._obj_with_exclusions
 
-        result = collections.OrderedDict()
+        result = OrderedDict()
         if axis != obj._info_axis_number:
             try:
                 for name, data in self:
@@ -253,7 +253,7 @@ def _aggregate_item_by_item(self, func, *args, **kwargs):
         # only for axis==0
 
         obj = self._obj_with_exclusions
-        result = collections.OrderedDict()
+        result = OrderedDict()
         cannot_agg = []
         errors = None
         for item in obj:
@@ -770,7 +770,7 @@ def aggregate(self, func_or_funcs, *args, **kwargs):
         if isinstance(func_or_funcs, str):
             return getattr(self, func_or_funcs)(*args, **kwargs)
 
-        if isinstance(func_or_funcs, compat.Iterable):
+        if isinstance(func_or_funcs, abc.Iterable):
             # Catch instances of lists / tuples
             # but not the class list / tuple itself.
             ret = self._aggregate_multiple_funcs(func_or_funcs,
@@ -834,7 +834,7 @@ def _aggregate_multiple_funcs(self, arg, _level):
                     columns.append(com.get_callable_name(f))
             arg = lzip(columns, arg)
 
-        results = collections.OrderedDict()
+        results = OrderedDict()
         for name, func in arg:
             obj = self
             if name in results:
@@ -911,7 +911,7 @@ def _get_index():
                           name=self._selection_name)
 
     def _aggregate_named(self, func, *args, **kwargs):
-        result = collections.OrderedDict()
+        result = OrderedDict()
 
         for name, group in self:
             group.name = name
@@ -1507,7 +1507,7 @@ def _apply_to_column_groupbys(self, func):
     def _fill(self, direction, limit=None):
         """Overridden method to join grouped columns in output"""
         res = super(DataFrameGroupBy, self)._fill(direction, limit=limit)
-        output = collections.OrderedDict(
+        output = OrderedDict(
             (grp.name, grp.grouper) for grp in self.grouper.groupings)
 
         from pandas import concat

pandas/core/internals/construction.py

@@ -2,14 +2,13 @@
 Functions for preparing various inputs passed to the DataFrame or Series
 constructors before passing them to a BlockManager.
 """
-from collections import OrderedDict
+from collections import OrderedDict, abc
 
 import numpy as np
 import numpy.ma as ma
 
 from pandas._libs import lib
 from pandas._libs.tslibs import IncompatibleFrequency
-import pandas.compat as compat
 from pandas.compat import lmap, lrange, raise_with_traceback
 
 from pandas.core.dtypes.cast import (
@@ -395,7 +394,7 @@ def to_arrays(data, columns, coerce_float=False, dtype=None):
     if isinstance(data[0], (list, tuple)):
         return _list_to_arrays(data, columns, coerce_float=coerce_float,
                                dtype=dtype)
-    elif isinstance(data[0], compat.Mapping):
+    elif isinstance(data[0], abc.Mapping):
         return _list_of_dict_to_arrays(data, columns,
                                        coerce_float=coerce_float, dtype=dtype)
     elif isinstance(data[0], ABCSeries):

pandas/core/series.py

@@ -1,7 +1,7 @@
 """
 Data structure for 1-dimensional cross-sectional and time series data
 """
-from collections import OrderedDict
+from collections import OrderedDict, abc
 from shutil import get_terminal_size
 from textwrap import dedent
 import warnings
@@ -224,8 +224,8 @@ def __init__(self, data=None, index=None, dtype=None, name=None,
                 raise TypeError("{0!r} type is unordered"
                                 "".format(data.__class__.__name__))
             # If data is Iterable but not list-like, consume into list.
-            elif (isinstance(data, compat.Iterable)
-                  and not isinstance(data, compat.Sized)):
+            elif (isinstance(data, abc.Iterable) and
+                  not isinstance(data, abc.Sized)):
                 data = list(data)
             else:
 
@@ -1496,7 +1496,7 @@ def to_dict(self, into=dict):
         Parameters
         ----------
         into : class, default dict
-            The collections.Mapping subclass to use as the return
+            The collections.abc.Mapping subclass to use as the return
             object. Can be the actual class or an empty
             instance of the mapping type you want.  If you want a
             collections.defaultdict, you must pass it initialized.
@@ -1505,7 +1505,7 @@ def to_dict(self, into=dict):
 
         Returns
         -------
-        collections.Mapping
+        collections.abc.Mapping
             Key-value representation of Series.
 
         Examples