DEPR: join_axes-kwarg in pd.concat (#22318)

Author: h-vetinari

doc/source/user_guide/merging.rst

@@ -70,9 +70,8 @@ some configurable handling of "what to do with the other axes":
 
 ::
 
-    pd.concat(objs, axis=0, join='outer', join_axes=None, ignore_index=False,
-              keys=None, levels=None, names=None, verify_integrity=False,
-              copy=True)
+    pd.concat(objs, axis=0, join='outer', ignore_index=False, keys=None,
+              levels=None, names=None, verify_integrity=False, copy=True)
 
 * ``objs`` : a sequence or mapping of Series or DataFrame objects. If a
   dict is passed, the sorted keys will be used as the `keys` argument, unless
@@ -87,8 +86,6 @@ some configurable handling of "what to do with the other axes":
   n - 1. This is useful if you are concatenating objects where the
   concatenation axis does not have meaningful indexing information. Note
   the index values on the other axes are still respected in the join.
-* ``join_axes`` : list of Index objects. Specific indexes to use for the other
-  n - 1 axes instead of performing inner/outer set logic.
 * ``keys`` : sequence, default None. Construct hierarchical index using the
   passed keys as the outermost level. If multiple levels passed, should
   contain tuples.
@@ -147,12 +144,11 @@ Set logic on the other axes
 
 When gluing together multiple DataFrames, you have a choice of how to handle
 the other axes (other than the one being concatenated). This can be done in
-the following three ways:
+the following two ways:
 
 * Take the union of them all, ``join='outer'``. This is the default
   option as it results in zero information loss.
 * Take the intersection, ``join='inner'``.
-* Use a specific index, as passed to the ``join_axes`` argument.
 
 Here is an example of each of these methods. First, the default ``join='outer'``
 behavior:
@@ -202,7 +198,13 @@ DataFrame:
 
 .. ipython:: python
 
-   result = pd.concat([df1, df4], axis=1, join_axes=[df1.index])
+   result = pd.concat([df1, df4], axis=1).reindex(df1.index)
+
+Similarly, we could index before the concatenation:
+
+.. ipython:: python
+
+    pd.concat([df1, df4.reindex(df1.index)], axis=1)
 
 .. ipython:: python
    :suppress:

doc/source/whatsnew/v0.25.0.rst

@@ -809,6 +809,7 @@ Other deprecations
 
 - The deprecated ``.ix[]`` indexer now raises a more visible ``FutureWarning`` instead of ``DeprecationWarning`` (:issue:`26438`).
 - Deprecated the ``units=M`` (months) and ``units=Y`` (year) parameters for ``units`` of :func:`pandas.to_timedelta`, :func:`pandas.Timedelta` and :func:`pandas.TimedeltaIndex` (:issue:`16344`)
+- :meth:`pandas.concat` has deprecated the ``join_axes``-keyword. Instead, use :meth:`DataFrame.reindex` or :meth:`DataFrame.reindex_like` on the result or on the inputs (:issue:`21951`)
 - The :attr:`SparseArray.values` attribute is deprecated. You can use ``np.asarray(...)`` or
   the :meth:`SparseArray.to_dense` method instead (:issue:`26421`).
 - The functions :func:`pandas.to_datetime` and :func:`pandas.to_timedelta` have deprecated the ``box`` keyword. Instead, use :meth:`to_numpy` or :meth:`Timestamp.to_datetime64` or :meth:`Timedelta.to_timedelta64`. (:issue:`24416`)

pandas/core/frame.py

@@ -6806,12 +6806,12 @@ def _join_compat(self, other, on=None, how='left', lsuffix='', rsuffix='',
             # join indexes only using concat
             if can_concat:
                 if how == 'left':
-                    how = 'outer'
-                    join_axes = [self.index]
+                    res = concat(frames, axis=1, join='outer',
+                                 verify_integrity=True)
+                    return res.reindex(self.index, copy=False)
                 else:
-                    join_axes = None
-                return concat(frames, axis=1, join=how, join_axes=join_axes,
-                              verify_integrity=True)
+                    return concat(frames, axis=1, join=how,
+                                  verify_integrity=True)
 
             joined = frames[0]
 

pandas/core/generic.py

@@ -9716,7 +9716,8 @@ def describe_1d(data):
                 if name not in names:
                     names.append(name)
 
-        d = pd.concat(ldesc, join_axes=pd.Index([names]), axis=1)
+        d = pd.concat([x.reindex(names, copy=False) for x in ldesc],
+                      axis=1, sort=False)
         d.columns = data.columns.copy()
         return d
 

pandas/core/groupby/generic.py

@@ -562,8 +562,10 @@ def _transform_general(self, func, *args, **kwargs):
                 applied.append(res)
 
         concat_index = obj.columns if self.axis == 0 else obj.index
-        concatenated = concat(applied, join_axes=[concat_index],
-                              axis=self.axis, verify_integrity=False)
+        other_axis = 1 if self.axis == 0 else 0  # switches between 0 & 1
+        concatenated = concat(applied, axis=self.axis, verify_integrity=False)
+        concatenated = concatenated.reindex(concat_index, axis=other_axis,
+                                            copy=False)
         return self._set_result_index_ordered(concatenated)
 
     @Substitution(klass='DataFrame', selected='')

pandas/core/reshape/concat.py

@@ -2,6 +2,8 @@
 concat routines
 """
 
+import warnings
+
 import numpy as np
 
 import pandas.core.dtypes.concat as _concat
@@ -44,8 +46,11 @@ def concat(objs, axis=0, join='outer', join_axes=None, ignore_index=False,
     join : {'inner', 'outer'}, default 'outer'
         How to handle indexes on other axis (or axes).
     join_axes : list of Index objects
+        .. deprecated:: 0.25.0
+
         Specific indexes to use for the other n - 1 axes instead of performing
-        inner/outer set logic.
+        inner/outer set logic. Use .reindex() before or after concatenation
+        as a replacement.
     ignore_index : bool, default False
         If True, do not use the index values along the concatenation axis. The
         resulting axis will be labeled 0, ..., n - 1. This is useful if you are
@@ -221,11 +226,11 @@ def concat(objs, axis=0, join='outer', join_axes=None, ignore_index=False,
         ...
     ValueError: Indexes have overlapping values: ['a']
     """
-    op = _Concatenator(objs, axis=axis, join_axes=join_axes,
-                       ignore_index=ignore_index, join=join,
-                       keys=keys, levels=levels, names=names,
-                       verify_integrity=verify_integrity,
+    op = _Concatenator(objs, axis=axis, ignore_index=ignore_index, join=join,
+                       join_axes=join_axes, keys=keys, levels=levels,
+                       names=names, verify_integrity=verify_integrity,
                        copy=copy, sort=sort)
+
     return op.get_result()
 
 
@@ -234,10 +239,9 @@ class _Concatenator:
     Orchestrates a concatenation operation for BlockManagers
     """
 
-    def __init__(self, objs, axis=0, join='outer', join_axes=None,
-                 keys=None, levels=None, names=None,
-                 ignore_index=False, verify_integrity=False, copy=True,
-                 sort=False):
+    def __init__(self, objs, axis=0, join='outer', join_axes=None, keys=None,
+                 levels=None, names=None, ignore_index=False,
+                 verify_integrity=False, copy=True, sort=False):
         if isinstance(objs, (NDFrame, str)):
             raise TypeError('first argument must be an iterable of pandas '
                             'objects, you passed an object of type '
@@ -310,9 +314,7 @@ def __init__(self, objs, axis=0, join='outer', join_axes=None,
                            if sum(obj.shape) > 0 or isinstance(obj, Series)]
 
             if (len(non_empties) and (keys is None and names is None and
-                                      levels is None and
-                                      join_axes is None and
-                                      not self.intersect)):
+                                      levels is None and not self.intersect)):
                 objs = non_empties
                 sample = objs[0]
 
@@ -446,7 +448,14 @@ def _get_new_axes(self):
                 if i == self.axis:
                     continue
                 new_axes[i] = self._get_comb_axis(i)
+
         else:
+            # GH 21951
+            warnings.warn(
+                'The join_axes-keyword is deprecated. Use .reindex or '
+                '.reindex_like on the result to achieve the same '
+                'functionality.', FutureWarning, stacklevel=4)
+
             if len(self.join_axes) != ndim - 1:
                 raise AssertionError("length of join_axes must be equal "
                                      "to {length}".format(length=ndim - 1))

pandas/tests/reshape/test_concat.py

@@ -722,6 +722,26 @@ def test_concat_categorical_empty(self):
         tm.assert_series_equal(pd.concat([s2, s1], ignore_index=True), exp)
         tm.assert_series_equal(s2.append(s1, ignore_index=True), exp)
 
+    def test_concat_join_axes_deprecated(self, axis):
+        # GH21951
+        one = pd.DataFrame([[0., 1.], [2., 3.]], columns=list('ab'))
+        two = pd.DataFrame([[10., 11.], [12., 13.]], index=[1, 2],
+                           columns=list('bc'))
+
+        expected = pd.concat([one, two],
+                             axis=1, sort=False).reindex(index=two.index)
+        with tm.assert_produces_warning(expected_warning=FutureWarning):
+            result = pd.concat([one, two],
+                               axis=1, sort=False, join_axes=[two.index])
+        tm.assert_frame_equal(result, expected)
+
+        expected = pd.concat([one, two],
+                             axis=0, sort=False).reindex(columns=two.columns)
+        with tm.assert_produces_warning(expected_warning=FutureWarning):
+            result = pd.concat([one, two],
+                               axis=0, sort=False, join_axes=[two.columns])
+        tm.assert_frame_equal(result, expected)
+
 
 class TestAppend: