Merge pull request #74 from pandas-dev/master

Sync Fork from Upstream Repo

Author: sthagen

pandas/core/frame.py

@@ -3599,7 +3599,7 @@ def align(
         see_also_sub=" or columns",
     )
     @Appender(NDFrame.set_axis.__doc__)
-    def set_axis(self, labels, axis=0, inplace=False):
+    def set_axis(self, labels, axis: Axis = 0, inplace: bool = False):
         return super().set_axis(labels, axis=axis, inplace=inplace)
 
     @Substitution(**_shared_doc_kwargs)

pandas/core/generic.py

@@ -520,7 +520,7 @@ def _obj_with_exclusions(self: FrameOrSeries) -> FrameOrSeries:
         """ internal compat with SelectionMixin """
         return self
 
-    def set_axis(self, labels, axis=0, inplace=False):
+    def set_axis(self, labels, axis: Axis = 0, inplace: bool = False):
         """
         Assign desired index to given axis.
 
@@ -561,7 +561,8 @@ def set_axis(self, labels, axis=0, inplace=False):
             obj.set_axis(labels, axis=axis, inplace=True)
             return obj
 
-    def _set_axis(self, axis, labels) -> None:
+    def _set_axis(self, axis: int, labels: Index) -> None:
+        labels = ensure_index(labels)
         self._data.set_axis(axis, labels)
         self._clear_item_cache()
 

pandas/core/internals/managers.py

@@ -164,15 +164,15 @@ def __nonzero__(self):
     __bool__ = __nonzero__
 
     @property
-    def shape(self):
+    def shape(self) -> Tuple[int, ...]:
         return tuple(len(ax) for ax in self.axes)
 
     @property
     def ndim(self) -> int:
         return len(self.axes)
 
-    def set_axis(self, axis, new_labels):
-        new_labels = ensure_index(new_labels)
+    def set_axis(self, axis: int, new_labels: Index):
+        # Caller is responsible for ensuring we have an Index object.
         old_len = len(self.axes[axis])
         new_len = len(new_labels)
 
@@ -184,7 +184,9 @@ def set_axis(self, axis, new_labels):
 
         self.axes[axis] = new_labels
 
-    def rename_axis(self, mapper, axis, copy: bool = True, level=None):
+    def rename_axis(
+        self, mapper, axis: int, copy: bool = True, level=None
+    ) -> "BlockManager":
         """
         Rename one of axes.
 
@@ -193,7 +195,7 @@ def rename_axis(self, mapper, axis, copy: bool = True, level=None):
         mapper : unary callable
         axis : int
         copy : bool, default True
-        level : int, default None
+        level : int or None, default None
         """
         obj = self.copy(deep=copy)
         obj.set_axis(axis, _transform_index(self.axes[axis], mapper, level))
@@ -233,7 +235,7 @@ def _rebuild_blknos_and_blklocs(self):
         self._blklocs = new_blklocs
 
     @property
-    def items(self):
+    def items(self) -> Index:
         return self.axes[0]
 
     def _get_counts(self, f):
@@ -623,7 +625,7 @@ def comp(s, regex=False):
         bm._consolidate_inplace()
         return bm
 
-    def is_consolidated(self):
+    def is_consolidated(self) -> bool:
         """
         Return True if more than one block with the same dtype
         """
@@ -688,7 +690,7 @@ def get_numeric_data(self, copy: bool = False):
         self._consolidate_inplace()
         return self.combine([b for b in self.blocks if b.is_numeric], copy)
 
-    def combine(self, blocks, copy=True):
+    def combine(self, blocks: List[Block], copy: bool = True) -> "BlockManager":
         """ return a new manager with the blocks """
         if len(blocks) == 0:
             return self.make_empty()
@@ -992,7 +994,6 @@ def delete(self, item):
         self.blocks = tuple(
             b for blkno, b in enumerate(self.blocks) if not is_blk_deleted[blkno]
         )
-        self._shape = None
         self._rebuild_blknos_and_blklocs()
 
     def set(self, item, value):
@@ -1160,7 +1161,6 @@ def insert(self, loc: int, item, value, allow_duplicates: bool = False):
 
         self.axes[0] = new_axis
         self.blocks += (block,)
-        self._shape = None
 
         self._known_consolidated = False
 

pandas/core/series.py

@@ -9,7 +9,6 @@
     TYPE_CHECKING,
     Any,
     Callable,
-    Hashable,
     Iterable,
     List,
     Optional,
@@ -23,7 +22,7 @@
 from pandas._config import get_option
 
 from pandas._libs import lib, properties, reshape, tslibs
-from pandas._typing import Label
+from pandas._typing import Axis, DtypeObj, Label
 from pandas.compat.numpy import function as nv
 from pandas.util._decorators import Appender, Substitution, doc
 from pandas.util._validators import validate_bool_kwarg, validate_percentile
@@ -177,7 +176,7 @@ class Series(base.IndexOpsMixin, generic.NDFrame):
 
     _typ = "series"
 
-    _name: Optional[Hashable]
+    _name: Label
     _metadata: List[str] = ["name"]
     _internal_names_set = {"index"} | generic.NDFrame._internal_names_set
     _accessors = {"dt", "cat", "str", "sparse"}
@@ -391,9 +390,12 @@ def _can_hold_na(self):
 
     _index = None
 
-    def _set_axis(self, axis, labels, fastpath: bool = False) -> None:
+    def _set_axis(self, axis: int, labels, fastpath: bool = False) -> None:
         """
         Override generic, we want to set the _typ here.
+
+        This is called from the cython code when we set the `index` attribute
+        directly, e.g. `series.index = [1, 2, 3]`.
         """
         if not fastpath:
             labels = ensure_index(labels)
@@ -413,6 +415,7 @@ def _set_axis(self, axis, labels, fastpath: bool = False) -> None:
 
         object.__setattr__(self, "_index", labels)
         if not fastpath:
+            # The ensure_index call aabove ensures we have an Index object
             self._data.set_axis(axis, labels)
 
     def _update_inplace(self, result, **kwargs):
@@ -421,25 +424,25 @@ def _update_inplace(self, result, **kwargs):
 
     # ndarray compatibility
     @property
-    def dtype(self):
+    def dtype(self) -> DtypeObj:
         """
         Return the dtype object of the underlying data.
         """
         return self._data.dtype
 
     @property
-    def dtypes(self):
+    def dtypes(self) -> DtypeObj:
         """
         Return the dtype object of the underlying data.
         """
         return self._data.dtype
 
     @property
-    def name(self) -> Optional[Hashable]:
+    def name(self) -> Label:
         return self._name
 
     @name.setter
-    def name(self, value: Optional[Hashable]) -> None:
+    def name(self, value: Label) -> None:
         if not is_hashable(value):
             raise TypeError("Series.name must be a hashable type")
         object.__setattr__(self, "_name", value)
@@ -689,7 +692,7 @@ def __array_ufunc__(
         inputs = tuple(extract_array(x, extract_numpy=True) for x in inputs)
         result = getattr(ufunc, method)(*inputs, **kwargs)
 
-        name: Optional[Hashable]
+        name: Label
         if len(set(names)) == 1:
             name = names[0]
         else:
@@ -3983,7 +3986,7 @@ def rename(
         see_also_sub="",
     )
     @Appender(generic.NDFrame.set_axis.__doc__)
-    def set_axis(self, labels, axis=0, inplace=False):
+    def set_axis(self, labels, axis: Axis = 0, inplace: bool = False):
         return super().set_axis(labels, axis=axis, inplace=inplace)
 
     @Substitution(**_shared_doc_kwargs)

pandas/tests/frame/test_combine_concat.py

@@ -8,7 +8,7 @@
 import pandas._testing as tm
 
 
-class TestDataFrameConcatCommon:
+class TestDataFrameConcat:
     def test_concat_multiple_frames_dtypes(self):
 
         # GH 2759
@@ -107,77 +107,6 @@ def test_concat_tuple_keys(self):
         )
         tm.assert_frame_equal(results, expected)
 
-    def test_join_str_datetime(self):
-        str_dates = ["20120209", "20120222"]
-        dt_dates = [datetime(2012, 2, 9), datetime(2012, 2, 22)]
-
-        A = DataFrame(str_dates, index=range(2), columns=["aa"])
-        C = DataFrame([[1, 2], [3, 4]], index=str_dates, columns=dt_dates)
-
-        tst = A.join(C, on="aa")
-
-        assert len(tst.columns) == 3
-
-    def test_join_multiindex_leftright(self):
-        # GH 10741
-        df1 = pd.DataFrame(
-            [
-                ["a", "x", 0.471780],
-                ["a", "y", 0.774908],
-                ["a", "z", 0.563634],
-                ["b", "x", -0.353756],
-                ["b", "y", 0.368062],
-                ["b", "z", -1.721840],
-                ["c", "x", 1],
-                ["c", "y", 2],
-                ["c", "z", 3],
-            ],
-            columns=["first", "second", "value1"],
-        ).set_index(["first", "second"])
-
-        df2 = pd.DataFrame(
-            [["a", 10], ["b", 20]], columns=["first", "value2"]
-        ).set_index(["first"])
-
-        exp = pd.DataFrame(
-            [
-                [0.471780, 10],
-                [0.774908, 10],
-                [0.563634, 10],
-                [-0.353756, 20],
-                [0.368062, 20],
-                [-1.721840, 20],
-                [1.000000, np.nan],
-                [2.000000, np.nan],
-                [3.000000, np.nan],
-            ],
-            index=df1.index,
-            columns=["value1", "value2"],
-        )
-
-        # these must be the same results (but columns are flipped)
-        tm.assert_frame_equal(df1.join(df2, how="left"), exp)
-        tm.assert_frame_equal(df2.join(df1, how="right"), exp[["value2", "value1"]])
-
-        exp_idx = pd.MultiIndex.from_product(
-            [["a", "b"], ["x", "y", "z"]], names=["first", "second"]
-        )
-        exp = pd.DataFrame(
-            [
-                [0.471780, 10],
-                [0.774908, 10],
-                [0.563634, 10],
-                [-0.353756, 20],
-                [0.368062, 20],
-                [-1.721840, 20],
-            ],
-            index=exp_idx,
-            columns=["value1", "value2"],
-        )
-
-        tm.assert_frame_equal(df1.join(df2, how="right"), exp)
-        tm.assert_frame_equal(df2.join(df1, how="left"), exp[["value2", "value1"]])
-
     def test_concat_named_keys(self):
         # GH 14252
         df = pd.DataFrame({"foo": [1, 2], "bar": [0.1, 0.2]})

pandas/tests/frame/test_join.py

@@ -1,6 +1,9 @@
+from datetime import datetime
+
 import numpy as np
 import pytest
 
+import pandas as pd
 from pandas import DataFrame, Index, period_range
 import pandas._testing as tm
 
@@ -216,3 +219,76 @@ def test_suppress_future_warning_with_sort_kw(sort_kw):
     with tm.assert_produces_warning(None, check_stacklevel=False):
         result = a.join([b, c], how="outer", sort=sort_kw)
     tm.assert_frame_equal(result, expected)
+
+
+class TestDataFrameJoin:
+    def test_join_str_datetime(self):
+        str_dates = ["20120209", "20120222"]
+        dt_dates = [datetime(2012, 2, 9), datetime(2012, 2, 22)]
+
+        A = DataFrame(str_dates, index=range(2), columns=["aa"])
+        C = DataFrame([[1, 2], [3, 4]], index=str_dates, columns=dt_dates)
+
+        tst = A.join(C, on="aa")
+
+        assert len(tst.columns) == 3
+
+    def test_join_multiindex_leftright(self):
+        # GH 10741
+        df1 = pd.DataFrame(
+            [
+                ["a", "x", 0.471780],
+                ["a", "y", 0.774908],
+                ["a", "z", 0.563634],
+                ["b", "x", -0.353756],
+                ["b", "y", 0.368062],
+                ["b", "z", -1.721840],
+                ["c", "x", 1],
+                ["c", "y", 2],
+                ["c", "z", 3],
+            ],
+            columns=["first", "second", "value1"],
+        ).set_index(["first", "second"])
+
+        df2 = pd.DataFrame(
+            [["a", 10], ["b", 20]], columns=["first", "value2"]
+        ).set_index(["first"])
+
+        exp = pd.DataFrame(
+            [
+                [0.471780, 10],
+                [0.774908, 10],
+                [0.563634, 10],
+                [-0.353756, 20],
+                [0.368062, 20],
+                [-1.721840, 20],
+                [1.000000, np.nan],
+                [2.000000, np.nan],
+                [3.000000, np.nan],
+            ],
+            index=df1.index,
+            columns=["value1", "value2"],
+        )
+
+        # these must be the same results (but columns are flipped)
+        tm.assert_frame_equal(df1.join(df2, how="left"), exp)
+        tm.assert_frame_equal(df2.join(df1, how="right"), exp[["value2", "value1"]])
+
+        exp_idx = pd.MultiIndex.from_product(
+            [["a", "b"], ["x", "y", "z"]], names=["first", "second"]
+        )
+        exp = pd.DataFrame(
+            [
+                [0.471780, 10],
+                [0.774908, 10],
+                [0.563634, 10],
+                [-0.353756, 20],
+                [0.368062, 20],
+                [-1.721840, 20],
+            ],
+            index=exp_idx,
+            columns=["value1", "value2"],
+        )
+
+        tm.assert_frame_equal(df1.join(df2, how="right"), exp)
+        tm.assert_frame_equal(df2.join(df1, how="left"), exp[["value2", "value1"]])

pandas/tests/series/test_combine_concat.py

@@ -5,7 +5,7 @@
 from pandas import Series
 
 
-class TestSeriesCombine:
+class TestSeriesConcat:
     @pytest.mark.parametrize(
         "dtype", ["float64", "int8", "uint8", "bool", "m8[ns]", "M8[ns]"]
     )