BUG: check for non-contiguous memory in SeriesGrouper, causing segfault

Author: wesm

pandas/core/frame.py

@@ -2856,7 +2856,7 @@ def append(self, other, ignore_index=False, verify_integrity=True):
             other = other.reindex(self.columns, copy=False)
             other = DataFrame(other.values.reshape((1, len(other))),
                               index=index, columns=self.columns)
-        elif isinstance(other, list):
+        elif isinstance(other, list) and not isinstance(other[0], DataFrame):
             other = DataFrame(other)
             if (self.columns.get_indexer(other.columns) >= 0).all():
                 other = other.ix[:, self.columns]

pandas/core/groupby.py

@@ -1331,7 +1331,7 @@ def _aggregate_series_fast(obj, func, group_index, ngroups):
         raise TypeError('Incompatible index for Cython grouper')
 
     # avoids object / Series creation overhead
-    dummy = obj[:0]
+    dummy = obj[:0].copy()
     indexer = lib.groupsort_indexer(group_index, ngroups)[0]
     obj = obj.take(indexer)
     group_index = group_index.take(indexer)

pandas/src/reduce.pyx

@@ -106,6 +106,8 @@ cdef class SeriesGrouper:
 
         self.labels = labels
         self.f = f
+        if not series.flags.c_contiguous:
+            series = series.copy('C')
         self.arr = series
         self.index = series.index
 
@@ -119,6 +121,9 @@ cdef class SeriesGrouper:
         else:
             if dummy.dtype != self.arr.dtype:
                 raise ValueError('Dummy array must be same dtype')
+            if not dummy.flags.contiguous:
+                dummy = dummy.copy()
+
         return dummy
 
     def get_result(self):

pandas/tests/test_groupby.py

@@ -1230,6 +1230,15 @@ def test_groupby_nonstring_columns(self):
         del expected[0]
         assert_frame_equal(result, expected)
 
+    def test_cython_grouper_series_bug_noncontig(self):
+        arr = np.empty((100, 100))
+        arr.fill(np.nan)
+        obj = Series(arr[:, 0], index=range(100))
+        inds = np.tile(range(10), 10)
+
+        result = obj.groupby(inds).agg(Series.median)
+        self.assert_(result.isnull().all())
+
 class TestPanelGroupBy(unittest.TestCase):
 
     def setUp(self):