Merge pull request #9877 from cpcloud/fix-index-dict-assign

Fix right hand side dict assignment for DataFrames

Author: cpcloud

doc/source/indexing.rst

@@ -249,6 +249,14 @@ new column.
 If you are using the IPython environment, you may also use tab-completion to
 see these accessible attributes.
 
+You can also assign a ``dict`` to a row of a ``DataFrame``:
+
+.. ipython:: python
+
+   x = pd.DataFrame({'x': [1, 2, 3], 'y': [3, 4, 5]})
+   x.iloc[1] = dict(x=9, y=99)
+   x
+
 Slicing ranges
 --------------
 

doc/source/whatsnew/v0.16.1.txt

@@ -134,6 +134,8 @@ Bug Fixes
 
 - Bug in unequal comparisons between categorical data and a scalar, which was not in the categories (e.g. ``Series(Categorical(list("abc"), ordered=True)) > "d"``. This returned ``False`` for all elements, but now raises a ``TypeError``. Equality comparisons also now return ``False`` for ``==`` and ``True`` for ``!=``. (:issue:`9848`)
 
+- Bug in DataFrame ``__setitem__`` when right hand side is a dictionary (:issue:`9874`)
+
 - Bug in ``MultiIndex.sortlevel()`` results in unicode level name breaks (:issue:`9875`)
 - Bug in which ``groupby.transform`` incorrectly enforced output dtypes to match input dtypes. (:issue:`9807`)
 

pandas/core/frame.py

@@ -4414,12 +4414,12 @@ def mode(self, axis=0, numeric_only=False):
         """
         Gets the mode(s) of each element along the axis selected. Empty if nothing
         has 2+ occurrences. Adds a row for each mode per label, fills in gaps
-        with nan. 
-        
+        with nan.
+
         Note that there could be multiple values returned for the selected
-        axis (when more than one item share the maximum frequency), which is the 
-        reason why a dataframe is returned. If you want to impute missing values 
-        with the mode in a dataframe ``df``, you can just do this: 
+        axis (when more than one item share the maximum frequency), which is the
+        reason why a dataframe is returned. If you want to impute missing values
+        with the mode in a dataframe ``df``, you can just do this:
         ``df.fillna(df.mode().iloc[0])``
 
         Parameters

pandas/core/indexing.py

@@ -200,7 +200,6 @@ def _has_valid_positional_setitem_indexer(self, indexer):
         return True
 
     def _setitem_with_indexer(self, indexer, value):
-
         self._has_valid_setitem_indexer(indexer)
 
         # also has the side effect of consolidating in-place
@@ -486,8 +485,8 @@ def can_do_equal_len():
                     self.obj[item_labels[indexer[info_axis]]] = value
                     return
 
-            if isinstance(value, ABCSeries):
-                value = self._align_series(indexer, value)
+            if isinstance(value, (ABCSeries, dict)):
+                value = self._align_series(indexer, Series(value))
 
             elif isinstance(value, ABCDataFrame):
                 value = self._align_frame(indexer, value)

pandas/tests/test_indexing.py

@@ -4411,6 +4411,16 @@ def test_slice_with_zero_step_raises(self):
         self.assertRaisesRegexp(ValueError, 'slice step cannot be zero',
                                 lambda: s.ix[::0])
 
+    def test_indexing_assignment_dict_already_exists(self):
+        df = pd.DataFrame({'x': [1, 2, 6],
+                           'y': [2, 2, 8],
+                           'z': [-5, 0, 5]}).set_index('z')
+        expected = df.copy()
+        rhs = dict(x=9, y=99)
+        df.loc[5] = rhs
+        expected.loc[5] = [9, 99]
+        tm.assert_frame_equal(df, expected)
+
 
 class TestSeriesNoneCoercion(tm.TestCase):
     EXPECTED_RESULTS = [