Merge pull request #10433 from behzadnouri/stack-dupl-indx

BUG: closes bug in stack when index is not unique

Author: jreback

doc/source/whatsnew/v0.17.0.txt

@@ -561,6 +561,7 @@ Bug Fixes
 - Bug in ``Table.select_column`` where name is not preserved (:issue:`10392`)
 - Bug in ``offsets.generate_range`` where ``start`` and ``end`` have finer precision than ``offset`` (:issue:`9907`)
 - Bug in ``pd.rolling_*`` where ``Series.name`` would be lost in the output (:issue:`10565`)
+- Bug in ``stack`` when index or columns are not unique. (:issue:`10417`)
 
 
 

pandas/core/reshape.py

@@ -461,6 +461,12 @@ def stack(frame, level=-1, dropna=True):
     -------
     stacked : Series
     """
+    def factorize(index):
+        if index.is_unique:
+            return index, np.arange(len(index))
+        cat = Categorical(index, ordered=True)
+        return cat.categories, cat.codes
+
     N, K = frame.shape
     if isinstance(frame.columns, MultiIndex):
         if frame.columns._reference_duplicate_name(level):
@@ -475,20 +481,22 @@ def stack(frame, level=-1, dropna=True):
         return _stack_multi_columns(frame, level_num=level_num, dropna=dropna)
     elif isinstance(frame.index, MultiIndex):
         new_levels = list(frame.index.levels)
-        new_levels.append(frame.columns)
-
         new_labels = [lab.repeat(K) for lab in frame.index.labels]
-        new_labels.append(np.tile(np.arange(K), N).ravel())
+
+        clev, clab = factorize(frame.columns)
+        new_levels.append(clev)
+        new_labels.append(np.tile(clab, N).ravel())
 
         new_names = list(frame.index.names)
         new_names.append(frame.columns.name)
         new_index = MultiIndex(levels=new_levels, labels=new_labels,
                                names=new_names, verify_integrity=False)
     else:
-        ilabels = np.arange(N).repeat(K)
-        clabels = np.tile(np.arange(K), N).ravel()
-        new_index = MultiIndex(levels=[frame.index, frame.columns],
-                               labels=[ilabels, clabels],
+        levels, (ilab, clab) = \
+                zip(*map(factorize, (frame.index, frame.columns)))
+        labels = ilab.repeat(K), np.tile(clab, N).ravel()
+        new_index = MultiIndex(levels=levels,
+                               labels=labels,
                                names=[frame.index.name, frame.columns.name],
                                verify_integrity=False)
 

pandas/tests/test_multilevel.py

@@ -964,6 +964,44 @@ def test_stack(self):
         result = self.ymd.unstack(0).stack(-2)
         expected = self.ymd.unstack(0).stack(0)
 
+        # GH10417
+        def check(left, right):
+            assert_series_equal(left, right)
+            self.assertFalse(left.index.is_unique)
+            li, ri = left.index, right.index
+            for i in range(ri.nlevels):
+                tm.assert_numpy_array_equal(li.levels[i], ri.levels[i])
+                tm.assert_numpy_array_equal(li.labels[i], ri.labels[i])
+
+        df = DataFrame(np.arange(12).reshape(4, 3),
+                       index=list('abab'),
+                       columns=['1st', '2nd', '3rd'])
+
+        mi = MultiIndex(levels=[['a', 'b'], ['1st', '2nd', '3rd']],
+                        labels=[np.tile(np.arange(2).repeat(3), 2),
+                                np.tile(np.arange(3), 4)])
+
+        left, right = df.stack(), Series(np.arange(12), index=mi)
+        check(left, right)
+
+        df.columns = ['1st', '2nd', '1st']
+        mi = MultiIndex(levels=[['a', 'b'], ['1st', '2nd']],
+                        labels=[np.tile(np.arange(2).repeat(3), 2),
+                                np.tile([0, 1, 0], 4)])
+
+        left, right = df.stack(), Series(np.arange(12), index=mi)
+        check(left, right)
+
+        tpls = ('a', 2), ('b', 1), ('a', 1), ('b', 2)
+        df.index = MultiIndex.from_tuples(tpls)
+        mi = MultiIndex(levels=[['a', 'b'], [1, 2], ['1st', '2nd']],
+                        labels=[np.tile(np.arange(2).repeat(3), 2),
+                                np.repeat([1, 0, 1], [3, 6, 3]),
+                                np.tile([0, 1, 0], 4)])
+
+        left, right = df.stack(), Series(np.arange(12), index=mi)
+        check(left, right)
+
     def test_unstack_odd_failure(self):
         data = """day,time,smoker,sum,len
 Fri,Dinner,No,8.25,3.