took care of review comments

Author: admin

doc/source/whatsnew/v1.0.0.rst

@@ -357,10 +357,10 @@ When :class:`Categorical` contains ``np.nan``,
    pd.Categorical([1, 2, np.nan], ordered=True).min()
 
 
-Default dtype of empty :class:`pandas.core.series.Series`
+Default dtype of empty :class:`pandas.Series`
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-Initialising an empty :class:`pandas.core.series.Series` without specifying a dtype will raise a `FutureWarning` now
+Initialising an empty :class:`pandas.Series` without specifying a dtype will raise a `DeprecationWarning` now
 (:issue:`17261`). The default dtype will change from ``float64`` to ``object`` in future releases so that it is
 consistent with the behaviour of :class:`DataFrame` and :class:`Index`.
 
@@ -370,7 +370,7 @@ consistent with the behaviour of :class:`DataFrame` and :class:`Index`.
 
    In [1]: pd.Series()
    Out[2]:
-   FutureWarning: The default dtype for empty Series will be 'object' instead of 'float64' in the next version. Specify a dtype explicitly to silence this warning.
+   DeprecationWarning: The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning.
    Series([], dtype: float64)
 
 .. _whatsnew_1000.api_breaking.deps:

pandas/core/algorithms.py

@@ -601,7 +601,7 @@ def _factorize_array(
 )
 @Appender(_shared_docs["factorize"])
 def factorize(
-    values, sort: bool = False, na_sentinel: int = -1, size_hint: Optional[int] = None
+    values, sort: bool = False, na_sentinel: int = -1, size_hint: Optional[int] = None,
 ) -> Tuple[np.ndarray, Union[np.ndarray, ABCIndex]]:
     # Implementation notes: This method is responsible for 3 things
     # 1.) coercing data to array-like (ndarray, Index, extension array)

pandas/core/construction.py

@@ -570,10 +570,21 @@ def _try_cast(
 # see gh-17261
 def is_empty_data(data):
     """
-    Utility to check if a Series is instantiated with empty data
+    Utility to check if a Series is instantiated with empty data,
+    which does not contain dtype information.
+
+    Parameters
+    ----------
+    data : array-like, Iterable, dict, or scalar value
+        Contains data stored in Series.
+
+    Returns
+    -------
+    bool
     """
     is_none = data is None
-    is_simple_empty = isinstance(data, (list, tuple, dict)) and not data
+    is_list_like_without_dtype = is_list_like(data) and not hasattr(data, "dtype")
+    is_simple_empty = is_list_like_without_dtype and not data
     return is_none or is_simple_empty
 
 
@@ -589,14 +600,29 @@ def create_series_with_explicit_dtype(
     """
     Helper to pass an explicit dtype when instantiating an empty Series.
 
-    The signature of this function mirrors the signature of Series.__init__
-    but adds the additional keyword argument `dtype_if_empty`.
-
-    This silences a FutureWarning described in the GitHub issue
+    This silences a DeprecationWarning described in the GitHub issue
     mentioned above.
+
+    Parameters
+    ----------
+    data : Mirrored from Series.__init__
+    index : Mirrored from Series.__init__
+    dtype : Mirrored from Series.__init__
+    name : Mirrored from Series.__init__
+    copy : Mirrored from Series.__init__
+    fastpath : Mirrored from Series.__init__
+    dtype_if_empty : str, numpy.dtype, or ExtensionDtype
+        This dtype will be passed explicitly if an empty Series will
+        be instantiated.
+
+    Returns
+    -------
+    Series
     """
     from pandas.core.series import Series
 
     if is_empty_data(data) and dtype is None:
         dtype = dtype_if_empty
-    return Series(data, index, dtype, name, copy, fastpath)
+    return Series(
+        data=data, index=index, dtype=dtype, name=name, copy=copy, fastpath=fastpath
+    )

pandas/core/generic.py

@@ -6111,7 +6111,9 @@ def fillna(
 
             if self.ndim == 1:
                 if isinstance(value, (dict, ABCSeries)):
-                    value = create_series_with_explicit_dtype(value)
+                    value = create_series_with_explicit_dtype(
+                        value, dtype_if_empty=object
+                    )
                 elif not is_list_like(value):
                     pass
                 else:

pandas/core/groupby/generic.py

@@ -260,7 +260,9 @@ def aggregate(self, func=None, *args, **kwargs):
                 result = self._aggregate_named(func, *args, **kwargs)
 
             index = Index(sorted(result), name=self.grouper.names[0])
-            ret = create_series_with_explicit_dtype(result, index=index)
+            ret = create_series_with_explicit_dtype(
+                result, index=index, dtype_if_empty=object
+            )
 
         if not self.as_index:  # pragma: no cover
             print("Warning, ignoring as_index=True")
@@ -1207,17 +1209,17 @@ def first_not_none(values):
                 return DataFrame()
             elif isinstance(v, NDFrame):
 
-                # this is to silence a FutureWarning
+                # this is to silence a DeprecationWarning
                 # TODO: Remove when default dtype of empty Series is object
                 kwargs = v._construct_axes_dict()
                 if v._constructor is Series:
-                    is_empty = "data" not in kwargs or not kwargs["data"]
-                    if "dtype" not in kwargs and is_empty:
-                        kwargs["dtype"] = object
+                    backup = create_series_with_explicit_dtype(
+                        **kwargs, dtype_if_empty=object
+                    )
+                else:
+                    backup = v._constructor(**kwargs)
 
-                values = [
-                    x if (x is not None) else v._constructor(**kwargs) for x in values
-                ]
+                values = [x if (x is not None) else backup for x in values]
 
             v = values[0]
 

pandas/core/series.py

@@ -180,19 +180,6 @@ class Series(base.IndexOpsMixin, generic.NDFrame):
     def __init__(
         self, data=None, index=None, dtype=None, name=None, copy=False, fastpath=False
     ):
-        if is_empty_data(data) and dtype is None:
-            # Empty Series should have dtype object to be consistent
-            # with the behaviour of DataFrame and Index
-            warnings.warn(
-                "The default dtype for empty Series will be 'object' instead"
-                " of 'float64' in the next version. Specify a dtype explicitly"
-                " to silence this warning.",
-                FutureWarning,
-                stacklevel=2,
-            )
-            # uncomment the line below when removing the FutureWarning
-            # dtype = np.dtype(object)
-
         # we are called internally, so short-circuit
         if fastpath:
 
@@ -206,6 +193,18 @@ def __init__(
 
         else:
 
+            if is_empty_data(data) and dtype is None:
+                # gh-17261
+                warnings.warn(
+                    "The default dtype for empty Series will be 'object' instead"
+                    " of 'float64' in a future version. Specify a dtype explicitly"
+                    " to silence this warning.",
+                    DeprecationWarning,
+                    stacklevel=2,
+                )
+                # uncomment the line below when removing the DeprecationWarning
+                # dtype = np.dtype(object)
+
             if index is not None:
                 index = ensure_index(index)
 
@@ -345,6 +344,8 @@ def _init_dict(self, data, index=None, dtype=None):
             keys, values = [], []
 
         # Input is now list-like, so rely on "standard" construction:
+
+        # TODO: passing np.float64 to not break anything yet. See GH-17261
         s = create_series_with_explicit_dtype(
             values, index=keys, dtype=dtype, dtype_if_empty=np.float64
         )

pandas/io/html.py

@@ -763,7 +763,7 @@ def _parse_tfoot_tr(self, table):
 
 def _expand_elements(body):
     data = [len(elem) for elem in body]
-    lens = create_series_with_explicit_dtype(data)
+    lens = create_series_with_explicit_dtype(data, dtype_if_empty=object)
     lens_max = lens.max()
     not_max = lens[lens != lens_max]
 

pandas/io/json/_json.py

@@ -1015,7 +1015,7 @@ def _parse_no_numpy(self):
             self.check_keys_split(decoded)
             self.obj = create_series_with_explicit_dtype(**decoded)
         else:
-            self.obj = create_series_with_explicit_dtype(data)
+            self.obj = create_series_with_explicit_dtype(data, dtype_if_empty=object)
 
     def _parse_numpy(self):
         load_kwargs = {
@@ -1033,9 +1033,9 @@ def _parse_numpy(self):
             self.check_keys_split(decoded)
             self.obj = create_series_with_explicit_dtype(**decoded)
         elif self.orient in ["columns", "index"]:
-            self.obj = create_series_with_explicit_dtype(*data)
+            self.obj = create_series_with_explicit_dtype(*data, dtype_if_empty=object)
         else:
-            self.obj = create_series_with_explicit_dtype(data)
+            self.obj = create_series_with_explicit_dtype(data, dtype_if_empty=object)
 
     def _try_convert_types(self):
         if self.obj is None:

pandas/tests/frame/test_constructors.py

@@ -1217,7 +1217,9 @@ def test_constructor_list_of_series(self):
             OrderedDict([["a", 1.5], ["b", 3], ["c", 4]]),
             OrderedDict([["b", 3], ["c", 4], ["d", 6]]),
         ]
-        data = [create_series_with_explicit_dtype(d) for d in data]
+        data = [
+            create_series_with_explicit_dtype(d, dtype_if_empty=object) for d in data
+        ]
 
         result = DataFrame(data)
         sdict = OrderedDict(zip(range(len(data)), data))

pandas/tests/io/formats/test_format.py

@@ -1017,7 +1017,7 @@ def test_east_asian_unicode_true(self):
     def test_to_string_buffer_all_unicode(self):
         buf = StringIO()
 
-        empty = DataFrame({"c/\u03c3": Series()})
+        empty = DataFrame({"c/\u03c3": Series(dtype=object)})
         nonempty = DataFrame({"c/\u03c3": Series([1, 2, 3])})
 
         print(empty, file=buf)

pandas/tests/plotting/test_misc.py

@@ -34,7 +34,7 @@ def test_get_accessor_args():
 
     with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
         x, y, kind, kwargs = func(
-            backend_name="", data=Series(), args=["line", None], kwargs={}
+            backend_name="", data=Series(dtype=object), args=["line", None], kwargs={}
         )
     assert x is None
     assert y is None

pandas/tests/series/test_constructors.py

@@ -52,7 +52,7 @@ class TestSeriesConstructors:
         ],
     )
     def test_empty_constructor(self, constructor, check_index_type):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             expected = Series()
             result = constructor()
 
@@ -78,7 +78,7 @@ def test_scalar_conversion(self):
         assert int(Series([1.0])) == 1
 
     def test_constructor(self, datetime_series):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             empty_series = Series()
         assert datetime_series.index.is_all_dates
 
@@ -96,7 +96,7 @@ def test_constructor(self, datetime_series):
         assert mixed[1] is np.NaN
 
         assert not empty_series.index.is_all_dates
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             assert not Series().index.is_all_dates
 
         # exception raised is of type Exception
@@ -116,7 +116,7 @@ def test_constructor(self, datetime_series):
 
     @pytest.mark.parametrize("input_class", [list, dict, OrderedDict])
     def test_constructor_empty(self, input_class):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             empty = Series()
             empty2 = Series(input_class())
 
@@ -136,7 +136,7 @@ def test_constructor_empty(self, input_class):
 
         if input_class is not list:
             # With index:
-            with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
                 empty = Series(index=range(10))
                 empty2 = Series(input_class(), index=range(10))
             tm.assert_series_equal(empty, empty2)
@@ -170,7 +170,7 @@ def test_constructor_dtype_only(self, dtype, index):
         assert len(result) == 0
 
     def test_constructor_no_data_index_order(self):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             result = pd.Series(index=["b", "a", "c"])
         assert result.index.tolist() == ["b", "a", "c"]
 
@@ -637,7 +637,7 @@ def test_constructor_limit_copies(self, index):
         assert s._data.blocks[0].values is not index
 
     def test_constructor_pass_none(self):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             s = Series(None, index=range(5))
         assert s.dtype == np.float64
 
@@ -646,7 +646,7 @@ def test_constructor_pass_none(self):
 
         # GH 7431
         # inference on the index
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             s = Series(index=np.array([None]))
             expected = Series(index=Index([None]))
         tm.assert_series_equal(s, expected)

pandas/tests/series/test_dtypes.py

@@ -412,7 +412,7 @@ def test_astype_empty_constructor_equality(self, dtype):
             "m",  # Generic timestamps raise a ValueError. Already tested.
         ):
             init_empty = Series([], dtype=dtype)
-            with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+            with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
                 as_type_empty = Series([]).astype(dtype)
             tm.assert_series_equal(init_empty, as_type_empty)
 
@@ -477,7 +477,7 @@ def test_infer_objects_series(self):
         tm.assert_series_equal(actual, expected)
 
     def test_is_homogeneous_type(self):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             empty = Series()
         assert empty._is_homogeneous_type
         assert Series([1, 2])._is_homogeneous_type

pandas/tests/series/test_duplicates.py

@@ -71,7 +71,7 @@ def test_unique_data_ownership():
 )
 def test_is_unique(data, expected):
     # GH11946 / GH25180
-    s = create_series_with_explicit_dtype(data)
+    s = create_series_with_explicit_dtype(data, dtype_if_empty=object)
     assert s.is_unique is expected
 
 

pandas/tests/series/test_replace.py

@@ -246,7 +246,7 @@ def test_replace_with_empty_dictlike(self):
         s = pd.Series(list("abcd"))
         tm.assert_series_equal(s, s.replace(dict()))
 
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             empty_series = pd.Series([])
         tm.assert_series_equal(s, s.replace(empty_series))
 

pandas/tests/series/test_subclass.py

@@ -32,6 +32,6 @@ def test_subclass_unstack(self):
         tm.assert_frame_equal(res, exp)
 
     def test_subclass_empty_repr(self):
-        with tm.assert_produces_warning(FutureWarning, check_stacklevel=False):
+        with tm.assert_produces_warning(DeprecationWarning, check_stacklevel=False):
             sub_series = tm.SubclassedSeries()
         assert "SubclassedSeries" in repr(sub_series)