CLN: comments, docstrings, depr NaT.freq (#45071)

Author: jbrockmendel

doc/source/whatsnew/v1.4.0.rst

@@ -547,6 +547,7 @@ Other Deprecations
 - Deprecated :meth:`Categorical.replace`, use :meth:`Series.replace` instead (:issue:`44929`)
 - Deprecated :meth:`Index.__getitem__` with a bool key; use ``index.values[key]`` to get the old behavior (:issue:`44051`)
 - Deprecated downcasting column-by-column in :meth:`DataFrame.where` with integer-dtypes (:issue:`44597`)
+- Deprecated :meth:`NaT.freq` (:issue:`45071`)
 -
 
 .. ---------------------------------------------------------------------------

pandas/_libs/hashtable_class_helper.pxi.in

@@ -1228,9 +1228,7 @@ cdef class PyObjectHashTable(HashTable):
             hash(val)
 
             if ignore_na and (
-                (val is C_NA)
-                or (val != val)
-                or (val is None)
+                checknull(val)
                 or (use_na_value and val == na_value)
             ):
                 # if missing values do not count as unique values (i.e. if

pandas/_libs/tslibs/nattype.pxd

@@ -10,7 +10,6 @@ cdef set c_nat_strings
 cdef class _NaT(datetime):
     cdef readonly:
         int64_t value
-        object freq
 
 cdef _NaT c_NaT
 

pandas/_libs/tslibs/nattype.pyx

@@ -357,10 +357,18 @@ class NaTType(_NaT):
 
         base = _NaT.__new__(cls, 1, 1, 1)
         base.value = NPY_NAT
-        base.freq = None
 
         return base
 
+    @property
+    def freq(self):
+        warnings.warn(
+            "NaT.freq is deprecated and will be removed in a future version.",
+            FutureWarning,
+            stacklevel=1,
+        )
+        return None
+
     def __reduce_ex__(self, protocol):
         # python 3.6 compat
         # https://bugs.python.org/issue28730

pandas/_testing/__init__.py

@@ -1078,6 +1078,8 @@ def shares_memory(left, right) -> bool:
         return shares_memory(left._ndarray, right)
     if isinstance(left, pd.core.arrays.SparseArray):
         return shares_memory(left.sp_values, right)
+    if isinstance(left, pd.core.arrays.IntervalArray):
+        return shares_memory(left._left, right) or shares_memory(left._right, right)
 
     if isinstance(left, ExtensionArray) and left.dtype == "string[pyarrow]":
         # https://github.com/pandas-dev/pandas/pull/43930#discussion_r736862669

pandas/core/algorithms.py

@@ -119,7 +119,7 @@ def _ensure_data(values: ArrayLike) -> np.ndarray:
     This will coerce:
     - ints -> int64
     - uint -> uint64
-    - bool -> uint64 (TODO this should be uint8)
+    - bool -> uint8
     - datetimelike -> i8
     - datetime64tz -> i8 (in local tz)
     - categorical -> codes
@@ -899,7 +899,6 @@ def value_counts_arraylike(values, dropna: bool):
     original = values
     values = _ensure_data(values)
 
-    # TODO: handle uint8
     keys, counts = htable.value_count(values, dropna)
 
     if needs_i8_conversion(original.dtype):

pandas/core/arrays/base.py

@@ -1500,6 +1500,11 @@ def _fill_mask_inplace(
     def _empty(cls, shape: Shape, dtype: ExtensionDtype):
         """
         Create an ExtensionArray with the given shape and dtype.
+
+        See also
+        --------
+        ExtensionDtype.empty
+            ExtensionDtype.empty is the 'official' public version of this API.
         """
         obj = cls._from_sequence([], dtype=dtype)
 

pandas/core/arrays/string_arrow.py

@@ -355,6 +355,8 @@ def copy(self) -> ArrowStringArray:
         """
         Return a shallow copy of the array.
 
+        Underlying ChunkedArray is immutable, so a deep copy is unnecessary.
+
         Returns
         -------
         ArrowStringArray

pandas/core/construction.py

@@ -388,10 +388,9 @@ def extract_array(
     ----------
     obj : object
         For Series / Index, the underlying ExtensionArray is unboxed.
-        For Numpy-backed ExtensionArrays, the ndarray is extracted.
 
     extract_numpy : bool, default False
-        Whether to extract the ndarray from a PandasArray
+        Whether to extract the ndarray from a PandasArray.
 
     extract_range : bool, default False
         If we have a RangeIndex, return range._values if True

pandas/core/dtypes/base.py

@@ -215,6 +215,10 @@ def empty(self, shape: Shape) -> type_t[ExtensionArray]:
 
         Analogous to numpy.empty.
 
+        Parameters
+        ----------
+        shape : int or tuple[int]
+
         Returns
         -------
         ExtensionArray

pandas/core/frame.py

@@ -5847,6 +5847,9 @@ def isna(self) -> DataFrame:
 
     @doc(NDFrame.isna, klass=_shared_doc_kwargs["klass"])
     def isnull(self) -> DataFrame:
+        """
+        DataFrame.isnull is an alias for DataFrame.isna.
+        """
         return self.isna()
 
     @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])
@@ -5855,6 +5858,9 @@ def notna(self) -> DataFrame:
 
     @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])
     def notnull(self) -> DataFrame:
+        """
+        DataFrame.notnull is an alias for DataFrame.notna.
+        """
         return ~self.isna()
 
     @deprecate_nonkeyword_arguments(version=None, allowed_args=["self"])

pandas/core/reshape/pivot.py

@@ -221,9 +221,7 @@ def __internal_pivot_table(
         table = table.sort_index(axis=1)
 
     if fill_value is not None:
-        _table = table.fillna(fill_value, downcast="infer")
-        assert _table is not None  # needed for mypy
-        table = _table
+        table = table.fillna(fill_value, downcast="infer")
 
     if margins:
         if dropna:

pandas/core/series.py

@@ -5231,6 +5231,9 @@ def isna(self) -> Series:
     # error: Cannot determine type of 'isna'
     @doc(NDFrame.isna, klass=_shared_doc_kwargs["klass"])  # type: ignore[has-type]
     def isnull(self) -> Series:
+        """
+        Series.isnull is an alias for Series.isna.
+        """
         return super().isnull()
 
     # error: Cannot determine type of 'notna'
@@ -5241,6 +5244,9 @@ def notna(self) -> Series:
     # error: Cannot determine type of 'notna'
     @doc(NDFrame.notna, klass=_shared_doc_kwargs["klass"])  # type: ignore[has-type]
     def notnull(self) -> Series:
+        """
+        Series.notnull is an alias for Series.notna.
+        """
         return super().notnull()
 
     @deprecate_nonkeyword_arguments(version=None, allowed_args=["self"])

pandas/tests/base/test_misc.py

@@ -21,6 +21,19 @@
 import pandas._testing as tm
 
 
+def test_isnull_notnull_docstrings():
+    # GH#41855 make sure its clear these are aliases
+    doc = pd.DataFrame.notnull.__doc__
+    assert doc.startswith("\nDataFrame.notnull is an alias for DataFrame.notna.\n")
+    doc = pd.DataFrame.isnull.__doc__
+    assert doc.startswith("\nDataFrame.isnull is an alias for DataFrame.isna.\n")
+
+    doc = Series.notnull.__doc__
+    assert doc.startswith("\nSeries.notnull is an alias for Series.notna.\n")
+    doc = Series.isnull.__doc__
+    assert doc.startswith("\nSeries.isnull is an alias for Series.isna.\n")
+
+
 @pytest.mark.parametrize(
     "op_name, op",
     [

pandas/tests/scalar/test_nat.py

@@ -718,3 +718,8 @@ def test_pickle():
     # GH#4606
     p = tm.round_trip_pickle(NaT)
     assert p is NaT
+
+
+def test_freq_deprecated():
+    with tm.assert_produces_warning(FutureWarning, match="deprecated"):
+        NaT.freq

pandas/tests/tools/test_to_datetime.py

@@ -1659,7 +1659,8 @@ def test_to_datetime_respects_dayfirst(self, cache):
         with pytest.raises(ValueError, match=msg):
             # if dayfirst is respected, then this would parse as month=13, which
             #  would raise
-            to_datetime("01-13-2012", dayfirst=True, cache=cache)
+            with tm.assert_produces_warning(UserWarning, match="Provide format"):
+                to_datetime("01-13-2012", dayfirst=True, cache=cache)
 
     def test_to_datetime_on_datetime64_series(self, cache):
         # #2699

pandas/tests/util/test_shares_memory.py

@@ -0,0 +1,13 @@
+import pandas as pd
+import pandas._testing as tm
+
+
+def test_shares_memory_interval():
+    obj = pd.interval_range(1, 5)
+
+    assert tm.shares_memory(obj, obj)
+    assert tm.shares_memory(obj, obj._data)
+    assert tm.shares_memory(obj, obj[::-1])
+    assert tm.shares_memory(obj, obj[:2])
+
+    assert not tm.shares_memory(obj, obj._data.copy())

setup.py

@@ -392,8 +392,8 @@ def run(self):
 # ----------------------------------------------------------------------
 # Specification of Dependencies
 
-# TODO: Need to check to see if e.g. `linetrace` has changed and possibly
-# re-compile.
+# TODO(cython#4518): Need to check to see if e.g. `linetrace` has changed and
+#  possibly re-compile.
 def maybe_cythonize(extensions, *args, **kwargs):
     """
     Render tempita templates before calling cythonize. This is skipped for