Merge pull request #67 from pandas-dev/master

Sync Fork from Upstream Repo

Author: sthagen

doc/make.py

@@ -83,8 +83,8 @@ def _process_single_doc(self, single_doc):
                 obj = pandas  # noqa: F821
                 for name in single_doc.split("."):
                     obj = getattr(obj, name)
-            except AttributeError:
-                raise ImportError(f"Could not import {single_doc}")
+            except AttributeError as err:
+                raise ImportError(f"Could not import {single_doc}") from err
             else:
                 return single_doc[len("pandas.") :]
         else:

doc/source/user_guide/merging.rst

@@ -724,6 +724,27 @@ either the left or right tables, the values in the joined table will be
           labels=['left', 'right'], vertical=False);
    plt.close('all');
 
+You can merge a mult-indexed Series and a DataFrame, if the names of
+the MultiIndex correspond to the columns from the DataFrame. Transform
+the Series to a DataFrame using :meth:`Series.reset_index` before merging,
+as shown in the following example.
+
+.. ipython:: python
+
+   df = pd.DataFrame({"Let": ["A", "B", "C"], "Num": [1, 2, 3]})
+   df
+
+   ser = pd.Series(
+       ["a", "b", "c", "d", "e", "f"],
+       index=pd.MultiIndex.from_arrays(
+           [["A", "B", "C"] * 2, [1, 2, 3, 4, 5, 6]], names=["Let", "Num"]
+       ),
+   )
+   ser
+
+   result = pd.merge(df, ser.reset_index(), on=['Let', 'Num'])
+
+
 Here is another example with duplicate join keys in DataFrames:
 
 .. ipython:: python

doc/source/whatsnew/index.rst

@@ -7,7 +7,7 @@ Release Notes
 *************
 
 This is the list of changes to pandas between each release. For full details,
-see the commit logs at https://github.com/pandas-dev/pandas. For install and
+see the `commit logs <https://github.com/pandas-dev/pandas/commits/>`_. For install and
 upgrade instructions, see :ref:`install`.
 
 Version 1.1
@@ -24,9 +24,9 @@ Version 1.0
 .. toctree::
    :maxdepth: 2
 
-   v1.0.0
-   v1.0.1
    v1.0.2
+   v1.0.1
+   v1.0.0
 
 Version 0.25
 ------------

doc/source/whatsnew/v1.1.0.rst

@@ -90,7 +90,76 @@ Backwards incompatible API changes
   now raise a  ``TypeError`` if a not-accepted keyword argument is passed into it.
   Previously a ``UnsupportedFunctionCall`` was raised (``AssertionError`` if ``min_count`` passed into :meth:`~DataFrameGroupby.median``) (:issue:`31485`)
 - :meth:`DataFrame.at` and :meth:`Series.at` will raise a ``TypeError`` instead of a ``ValueError`` if an incompatible key is passed, and ``KeyError`` if a missing key is passed, matching the behavior of ``.loc[]`` (:issue:`31722`)
--
+
+.. _whatsnew_110.api_breaking.indexing_raises_key_errors:
+
+Failed Label-Based Lookups Always Raise KeyError
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Label lookups ``series[key]``, ``series.loc[key]`` and ``frame.loc[key]``
+used to raises either ``KeyError`` or ``TypeError`` depending on the type of
+key and type of :class:`Index`.  These now consistently raise ``KeyError`` (:issue:`31867`)
+
+.. ipython:: python
+
+    ser1 = pd.Series(range(3), index=[0, 1, 2])
+    ser2 = pd.Series(range(3), index=pd.date_range("2020-02-01", periods=3))
+
+*Previous behavior*:
+
+.. code-block:: ipython
+
+    In [3]: ser1[1.5]
+    ...
+    TypeError: cannot do label indexing on Int64Index with these indexers [1.5] of type float
+
+    In [4] ser1["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [5]: ser1.loc[1.5]
+    ...
+    TypeError: cannot do label indexing on Int64Index with these indexers [1.5] of type float
+
+    In [6]: ser1.loc["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [7]: ser2.loc[1]
+    ...
+    TypeError: cannot do label indexing on DatetimeIndex with these indexers [1] of type int
+
+    In [8]: ser2.loc[pd.Timestamp(0)]
+    ...
+    KeyError: Timestamp('1970-01-01 00:00:00')
+
+*New behavior*:
+
+.. code-block:: ipython
+
+    In [3]: ser1[1.5]
+    ...
+    KeyError: 1.5
+
+    In [4] ser1["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [5]: ser1.loc[1.5]
+    ...
+    KeyError: 1.5
+
+    In [6]: ser1.loc["foo"]
+    ...
+    KeyError: 'foo'
+
+    In [7]: ser2.loc[1]
+    ...
+    KeyError: 1
+
+    In [8]: ser2.loc[pd.Timestamp(0)]
+    ...
+    KeyError: Timestamp('1970-01-01 00:00:00')
 
 .. ---------------------------------------------------------------------------
 
@@ -126,9 +195,10 @@ Bug fixes
 
 Categorical
 ^^^^^^^^^^^
+
+- Bug where :func:`merge` was unable to join on non-unique categorical indices (:issue:`28189`)
 - Bug when passing categorical data to :class:`Index` constructor along with ``dtype=object`` incorrectly returning a :class:`CategoricalIndex` instead of object-dtype :class:`Index` (:issue:`32167`)
 -
--
 
 Datetimelike
 ^^^^^^^^^^^^

pandas/__init__.py

@@ -37,7 +37,7 @@
         f"C extension: {module} not built. If you want to import "
         "pandas from the source directory, you may need to run "
         "'python setup.py build_ext --inplace --force' to build the C extensions first."
-    )
+    ) from e
 
 from pandas._config import (
     get_option,
@@ -290,8 +290,8 @@ def __getattr__(self, item):
 
             try:
                 return getattr(self.np, item)
-            except AttributeError:
-                raise AttributeError(f"module numpy has no attribute {item}")
+            except AttributeError as err:
+                raise AttributeError(f"module numpy has no attribute {item}") from err
 
     np = __numpy()
 
@@ -306,8 +306,10 @@ def __getattr__(cls, item):
 
             try:
                 return getattr(cls.datetime, item)
-            except AttributeError:
-                raise AttributeError(f"module datetime has no attribute {item}")
+            except AttributeError as err:
+                raise AttributeError(
+                    f"module datetime has no attribute {item}"
+                ) from err
 
         def __instancecheck__(cls, other):
             return isinstance(other, cls.datetime)

pandas/_config/config.py

@@ -213,8 +213,8 @@ def __getattr__(self, key: str):
         prefix += key
         try:
             v = object.__getattribute__(self, "d")[key]
-        except KeyError:
-            raise OptionError("No such option")
+        except KeyError as err:
+            raise OptionError("No such option") from err
         if isinstance(v, dict):
             return DictWrapper(v, prefix)
         else:

pandas/_libs/hashtable.pyx

@@ -86,7 +86,10 @@ cdef class Factorizer:
         self, ndarray[object] values, sort=False, na_sentinel=-1, na_value=None
     ):
         """
+        Examples
+        --------
         Factorize values with nans replaced by na_sentinel
+
         >>> factorize(np.array([1,2,np.nan], dtype='O'), na_sentinel=20)
         array([ 0,  1, 20])
         """
@@ -131,7 +134,10 @@ cdef class Int64Factorizer:
     def factorize(self, const int64_t[:] values, sort=False,
                   na_sentinel=-1, na_value=None):
         """
+        Examples
+        --------
         Factorize values with nans replaced by na_sentinel
+
         >>> factorize(np.array([1,2,np.nan], dtype='O'), na_sentinel=20)
         array([ 0,  1, 20])
         """

pandas/_libs/index.pyx

@@ -115,8 +115,6 @@ cdef class IndexEngine:
     cdef _maybe_get_bool_indexer(self, object val):
         cdef:
             ndarray[uint8_t, ndim=1, cast=True] indexer
-            ndarray[intp_t, ndim=1] found
-            int count
 
         indexer = self._get_index_values() == val
         return self._unpack_bool_indexer(indexer, val)

pandas/_libs/internals.pyx

@@ -105,8 +105,7 @@ cdef class BlockPlacement:
             Py_ssize_t start, stop, end, _
         if not self._has_array:
             start, stop, step, _ = slice_get_indices_ex(self._as_slice)
-            self._as_array = np.arange(start, stop, step,
-                                       dtype=np.int64)
+            self._as_array = np.arange(start, stop, step, dtype=np.int64)
             self._has_array = True
         return self._as_array
 
@@ -283,8 +282,7 @@ cdef slice_getitem(slice slc, ind):
     s_start, s_stop, s_step, s_len = slice_get_indices_ex(slc)
 
     if isinstance(ind, slice):
-        ind_start, ind_stop, ind_step, ind_len = slice_get_indices_ex(ind,
-                                                                      s_len)
+        ind_start, ind_stop, ind_step, ind_len = slice_get_indices_ex(ind, s_len)
 
         if ind_step > 0 and ind_len == s_len:
             # short-cut for no-op slice

pandas/_libs/interval.pyx

@@ -481,8 +481,7 @@ cdef class Interval(IntervalMixin):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def intervals_to_interval_bounds(ndarray intervals,
-                                 bint validate_closed=True):
+def intervals_to_interval_bounds(ndarray intervals, bint validate_closed=True):
     """
     Parameters
     ----------
@@ -502,14 +501,14 @@ def intervals_to_interval_bounds(ndarray intervals,
     """
     cdef:
         object closed = None, interval
-        int64_t n = len(intervals)
+        Py_ssize_t i, n = len(intervals)
         ndarray left, right
         bint seen_closed = False
 
     left = np.empty(n, dtype=intervals.dtype)
     right = np.empty(n, dtype=intervals.dtype)
 
-    for i in range(len(intervals)):
+    for i in range(n):
         interval = intervals[i]
         if interval is None or util.is_nan(interval):
             left[i] = np.nan

pandas/_libs/join.pyx

@@ -254,6 +254,8 @@ ctypedef fused join_t:
     float64_t
     float32_t
     object
+    int8_t
+    int16_t
     int32_t
     int64_t
     uint64_t
@@ -815,18 +817,22 @@ def asof_join_nearest_on_X_by_Y(asof_t[:] left_values,
     right_indexer = np.empty(left_size, dtype=np.int64)
 
     # search both forward and backward
-    bli, bri = asof_join_backward_on_X_by_Y(left_values,
-                                            right_values,
-                                            left_by_values,
-                                            right_by_values,
-                                            allow_exact_matches,
-                                            tolerance)
-    fli, fri = asof_join_forward_on_X_by_Y(left_values,
-                                           right_values,
-                                           left_by_values,
-                                           right_by_values,
-                                           allow_exact_matches,
-                                           tolerance)
+    bli, bri = asof_join_backward_on_X_by_Y(
+        left_values,
+        right_values,
+        left_by_values,
+        right_by_values,
+        allow_exact_matches,
+        tolerance,
+    )
+    fli, fri = asof_join_forward_on_X_by_Y(
+        left_values,
+        right_values,
+        left_by_values,
+        right_by_values,
+        allow_exact_matches,
+        tolerance,
+    )
 
     for i in range(len(bri)):
         # choose timestamp from right with smaller difference