BUG: Fix strange behaviour of Series.iloc on MultiIndex Series (#17148)

Author: ante012

doc/source/whatsnew/v0.21.0.txt

@@ -329,6 +329,7 @@ Indexing
 - Fixes ``DataFrame.loc`` for setting with alignment and tz-aware ``DatetimeIndex`` (:issue:`16889`)
 - Avoids ``IndexError`` when passing an Index or Series to ``.iloc`` with older numpy (:issue:`17193`)
 - Allow unicode empty strings as placeholders in multilevel columns in Python 2 (:issue:`17099`)
+- Bug in :func:`Series.iloc` when used with increment or assignment and an int indexer on a ``MultiIndex`` ``Series`` causing the wrong indexes to be read from and written to (:issue:`17148`)
 
 I/O
 ^^^

pandas/core/indexing.py

@@ -146,7 +146,8 @@ def _get_setitem_indexer(self, key):
             return self._convert_tuple(key, is_setter=True)
 
         axis = self.obj._get_axis(0)
-        if isinstance(axis, MultiIndex):
+
+        if isinstance(axis, MultiIndex) and not is_integer(key):
             try:
                 return axis.get_loc(key)
             except Exception:

pandas/tests/indexing/test_iloc.py

@@ -269,6 +269,40 @@ def test_iloc_setitem(self):
         expected = Series([0, 1, 0], index=[4, 5, 6])
         tm.assert_series_equal(s, expected)
 
+    def test_iloc_setitem_int_multiindex_series(self):
+        # GH17148
+        def check_scenario(data, indexes, values, expected_k):
+            df = pd.DataFrame(
+                data=data,
+                columns=['i', 'j', 'k'])
+            df.set_index(['i', 'j'], inplace=True)
+
+            series = df.k.copy()
+            for i, v in zip(indexes, values):
+                series.iloc[i] += v
+
+            df.k = expected_k
+            expected = df.k.copy()
+            tm.assert_series_equal(series, expected)
+
+        check_scenario(
+            data=[[1, 22, 5], [1, 33, 6]],
+            indexes=[0, -1, 1],
+            values=[2, 3, 1],
+            expected_k=[7, 10])
+
+        check_scenario(
+            data=[[1, 3, 7], [2, 4, 8]],
+            indexes=[0, -1, 1],
+            values=[1, 1, 10],
+            expected_k=[8, 19])
+
+        check_scenario(
+            data=[[1, 11, 4], [2, 22, 5], [3, 33, 6]],
+            indexes=[0, -1, 1],
+            values=[4, 7, 10],
+            expected_k=[8, 15, 13])
+
     def test_iloc_setitem_list(self):
 
         # setitem with an iloc list