Merge pull request #6748 from immerrr/fix-as-blocks-for-sparse

BUG: fix NDFrame.as_blocks() for sparse containers

Author: jreback

doc/source/release.rst

@@ -296,6 +296,7 @@ Bug Fixes
 - Bug in consistency of groupby aggregation when passing a custom function (:issue:`6715`)
 - Bug in resample when ``how=None`` resample freq is the same as the axis frequency (:issue:`5955`)
 - Bug in downcasting inference with empty arrays (:issue:`6733`)
+- Bug in ``obj.blocks`` on sparse containers dropping all but the last items of same for dtype (:issue:`6748`)
 
 pandas 0.13.1
 -------------

pandas/core/generic.py

@@ -2003,7 +2003,7 @@ def ftypes(self):
         return Series(self._data.get_ftypes(), index=self._info_axis,
                       dtype=np.object_)
 
-    def as_blocks(self, columns=None):
+    def as_blocks(self):
         """
         Convert the frame to a dict of dtype -> Constructor Types that each has
         a homogeneous dtype.
@@ -2025,12 +2025,18 @@ def as_blocks(self, columns=None):
         """
         self._consolidate_inplace()
 
-        bd = dict()
+        bd = {}
         for b in self._data.blocks:
-            b = b.reindex_items_from(columns or b.items)
-            bd[str(b.dtype)] = self._constructor(
-                BlockManager([b], [b.items, self.index])).__finalize__(self)
-        return bd
+            bd.setdefault(str(b.dtype), []).append(b)
+
+        result = {}
+        for dtype, blocks in bd.items():
+            # Must combine even after consolidation, because there may be
+            # sparse items which are never consolidated into one block.
+            combined = self._data.combine(blocks, copy=True)
+            result[dtype] = self._constructor(combined).__finalize__(self)
+
+        return result
 
     @property
     def blocks(self):

pandas/sparse/tests/test_sparse.py

@@ -1515,6 +1515,14 @@ def test_sparse_pow_issue(self):
 
         self.assertEqual(len(r2.sp_values), len(r1.sp_values))
 
+    def test_as_blocks(self):
+        df = SparseDataFrame({'A': [1.1, 3.3], 'B': [nan, -3.9]},
+                             dtype='float64')
+
+        df_blocks = df.blocks
+        self.assertEqual(list(df_blocks.keys()), ['float64'])
+        assert_frame_equal(df_blocks['float64'], df)
+
 
 def _dense_series_compare(s, f):
     result = f(s)