Merge branch 'master' into 31988_ReadPickle

Author: pedroreys

Diff for: doc/source/development/code_style.rst

@@ -9,6 +9,12 @@ pandas code style guide
 .. contents:: Table of contents:
    :local:
 
+*pandas* follows the `PEP8 <https://www.python.org/dev/peps/pep-0008/>`_
+standard and uses `Black <https://black.readthedocs.io/en/stable/>`_
+and `Flake8 <https://flake8.pycqa.org/en/latest/>`_ to ensure a
+consistent code format throughout the project. For details see the
+:ref:`contributing guide to pandas<contributing.code-formatting>`.
+
 Patterns
 ========
 

Diff for: doc/source/whatsnew/v1.0.2.rst

@@ -20,6 +20,8 @@ Fixed regressions
 - Fixed regression in :meth:`pandas.core.groupby.RollingGroupby.apply` where the ``raw`` parameter was ignored (:issue:`31754`)
 - Fixed regression in :meth:`rolling(..).corr() <pandas.core.window.Rolling.corr>` when using a time offset (:issue:`31789`)
 - Fixed regression where :func:`read_pickle` raised a ``UnicodeDecodeError`` when reading a py27 pickle with :class:`MultiIndex` column (:issue:`31988`).
+- Fixed regression in :class:`DataFrame` arithmetic operations with mis-matched columns (:issue:`31623`)
+-
 
 .. ---------------------------------------------------------------------------
 
@@ -31,6 +33,7 @@ Bug fixes
 **Categorical**
 
 - Fixed bug where :meth:`Categorical.from_codes` improperly raised a ``ValueError`` when passed nullable integer codes. (:issue:`31779`)
+- Bug in :class:`Categorical` that would ignore or crash when calling :meth:`Series.replace` with a list-like ``to_replace`` (:issue:`31720`)
 
 **I/O**
 

Diff for: environment.yml

@@ -26,6 +26,7 @@ dependencies:
 
   # documentation
   - gitpython  # obtain contributors from git for whatsnew
+  - gitdb2=2.0.6  # GH-32060
   - sphinx
 
   # documentation (jupyter notebooks)

Diff for: pandas/_libs/algos.pyx

@@ -1173,12 +1173,12 @@ ctypedef fused out_t:
 
 @cython.boundscheck(False)
 @cython.wraparound(False)
-def diff_2d(ndarray[diff_t, ndim=2] arr,
-            ndarray[out_t, ndim=2] out,
+def diff_2d(diff_t[:, :] arr,
+            out_t[:, :] out,
             Py_ssize_t periods, int axis):
     cdef:
         Py_ssize_t i, j, sx, sy, start, stop
-        bint f_contig = arr.flags.f_contiguous
+        bint f_contig = arr.is_f_contig()
 
     # Disable for unsupported dtype combinations,
     #  see https://github.com/cython/cython/issues/2646

Diff for: pandas/_libs/hashing.pyx

@@ -5,7 +5,7 @@ import cython
 from libc.stdlib cimport malloc, free
 
 import numpy as np
-from numpy cimport uint8_t, uint32_t, uint64_t, import_array
+from numpy cimport ndarray, uint8_t, uint32_t, uint64_t, import_array
 import_array()
 
 from pandas._libs.util cimport is_nan
@@ -15,7 +15,7 @@ DEF dROUNDS = 4
 
 
 @cython.boundscheck(False)
-def hash_object_array(object[:] arr, object key, object encoding='utf8'):
+def hash_object_array(ndarray[object] arr, object key, object encoding='utf8'):
     """
     Parameters
     ----------

Diff for: pandas/_libs/tslibs/conversion.pyx

@@ -152,7 +152,7 @@ def ensure_timedelta64ns(arr: ndarray, copy: bool=True):
 
 @cython.boundscheck(False)
 @cython.wraparound(False)
-def datetime_to_datetime64(object[:] values):
+def datetime_to_datetime64(ndarray[object] values):
     """
     Convert ndarray of datetime-like objects to int64 array representing
     nanosecond timestamps.

Diff for: pandas/_libs/window/aggregations.pyx

@@ -56,7 +56,7 @@ cdef:
 cdef inline int int_max(int a, int b): return a if a >= b else b
 cdef inline int int_min(int a, int b): return a if a <= b else b
 
-cdef inline bint is_monotonic_start_end_bounds(
+cdef bint is_monotonic_start_end_bounds(
     ndarray[int64_t, ndim=1] start, ndarray[int64_t, ndim=1] end
 ):
     return is_monotonic(start, False)[0] and is_monotonic(end, False)[0]

Diff for: pandas/_testing.py

@@ -1074,6 +1074,7 @@ def assert_series_equal(
     check_exact=False,
     check_datetimelike_compat=False,
     check_categorical=True,
+    check_category_order=True,
     obj="Series",
 ):
     """
@@ -1108,6 +1109,10 @@ def assert_series_equal(
         Compare datetime-like which is comparable ignoring dtype.
     check_categorical : bool, default True
         Whether to compare internal Categorical exactly.
+    check_category_order : bool, default True
+        Whether to compare category order of internal Categoricals
+
+        .. versionadded:: 1.0.2
     obj : str, default 'Series'
         Specify object name being compared, internally used to show appropriate
         assertion message.
@@ -1210,7 +1215,12 @@ def assert_series_equal(
 
     if check_categorical:
         if is_categorical_dtype(left) or is_categorical_dtype(right):
-            assert_categorical_equal(left.values, right.values, obj=f"{obj} category")
+            assert_categorical_equal(
+                left.values,
+                right.values,
+                obj=f"{obj} category",
+                check_category_order=check_category_order,
+            )
 
 
 # This could be refactored to use the NDFrame.equals method

Diff for: pandas/core/arrays/categorical.py

@@ -2441,18 +2441,30 @@ def replace(self, to_replace, value, inplace: bool = False):
         """
         inplace = validate_bool_kwarg(inplace, "inplace")
         cat = self if inplace else self.copy()
-        if to_replace in cat.categories:
-            if isna(value):
-                cat.remove_categories(to_replace, inplace=True)
-            else:
+
+        # build a dict of (to replace -> value) pairs
+        if is_list_like(to_replace):
+            # if to_replace is list-like and value is scalar
+            replace_dict = {replace_value: value for replace_value in to_replace}
+        else:
+            # if both to_replace and value are scalar
+            replace_dict = {to_replace: value}
+
+        # other cases, like if both to_replace and value are list-like or if
+        # to_replace is a dict, are handled separately in NDFrame
+        for replace_value, new_value in replace_dict.items():
+            if replace_value in cat.categories:
+                if isna(new_value):
+                    cat.remove_categories(replace_value, inplace=True)
+                    continue
                 categories = cat.categories.tolist()
-                index = categories.index(to_replace)
-                if value in cat.categories:
-                    value_index = categories.index(value)
+                index = categories.index(replace_value)
+                if new_value in cat.categories:
+                    value_index = categories.index(new_value)
                     cat._codes[cat._codes == index] = value_index
-                    cat.remove_categories(to_replace, inplace=True)
+                    cat.remove_categories(replace_value, inplace=True)
                 else:
-                    categories[index] = value
+                    categories[index] = new_value
                     cat.rename_categories(categories, inplace=True)
         if not inplace:
             return cat