Fix repr when cols<=20 and rows>60 (#69)

* Fix repr when cols<=20 and rows>60 * Fix whitespace
modin-project · Aug 9, 2018 · 14240e1 · 14240e1
1 parent 4311f97
commit 14240e1
Show file tree

Hide file tree

Showing 2 changed files with 20 additions and 6 deletions.
diff --git a/modin/pandas/dataframe.py b/modin/pandas/dataframe.py
@@ -385,18 +385,16 @@ def col_dots_builder(full_front, full_back):
         if len(self.index) >= 60:
             head_blocks = self._head_block_builder(30)
             tail_blocks = self._tail_block_builder(30)
-            length_of_index = 30
+            index_of_head = self.index[:30]
+            index_of_tail = self.index[-30:]
         else:
             head_blocks = self._block_partitions
             # We set this to None so we know
             tail_blocks = None
-            length_of_index = len(self.index)
+            index_of_head = self.index
 
         # Get first and last 10 columns if there are more than 20 columns
         if len(self._col_metadata) >= 20:
-
-            index_of_head = self.index[:length_of_index]
-
             # Building the front blocks from head_blocks
             front_blocks = \
                 front_block_builder(head_blocks, 10, index_of_head)
@@ -410,7 +408,6 @@ def col_dots_builder(full_front, full_back):
 
             # True only if we have >60 rows in the DataFrame
             if tail_blocks is not None:
-                index_of_tail = self.index[-30:]
                 # Building the font blocks from tail_blocks
                 front_blocks = \
                     front_block_builder(tail_blocks, 10, index_of_tail)
@@ -435,6 +432,7 @@ def col_dots_builder(full_front, full_back):
             ]
             full_head = pandas.concat(list_of_head_rows)
             full_head.columns = self.columns
+            full_head.index = index_of_head
 
             # True only if we have >60 rows in the DataFrame
             if tail_blocks is not None:
@@ -444,6 +442,7 @@ def col_dots_builder(full_front, full_back):
                      for df in tail_blocks]
                 full_tail = pandas.concat(list_of_tail_rows)
                 full_tail.columns = self.columns
+                full_tail.index = index_of_tail
 
                 return row_dots_builder(full_head, full_tail)
             else:

diff --git a/modin/pandas/test/test_dataframe.py b/modin/pandas/test/test_dataframe.py
@@ -3415,18 +3415,33 @@ def test___repr__():
     #
     # assert repr(pandas_df) == repr(ray_df)
 
+    # ___repr___ method has a different code path depending on
+    # whether the number of rows is >60; and a different code path
+    # depending on the number of columns is >20.
+    # Previous test cases already check the case when cols>20
+    # and rows>60. The cases that follow exercise the other three
+    # combinations.
+    # rows <= 60, cols > 20
     frame_data = np.random.randint(0, 100, size=(10, 100))
     pandas_df = pandas.DataFrame(frame_data)
     ray_df = pd.DataFrame(frame_data)
 
     assert repr(pandas_df) == repr(ray_df)
 
+    # rows <= 60, cols <= 20
     frame_data = np.random.randint(0, 100, size=(10, 10))
     pandas_df = pandas.DataFrame(frame_data)
     ray_df = pd.DataFrame(frame_data)
 
     assert repr(pandas_df) == repr(ray_df)
 
+    # rows > 60, cols <= 20
+    frame_data = np.random.randint(0, 100, size=(100, 10))
+    pandas_df = pandas.DataFrame(frame_data)
+    ray_df = pd.DataFrame(frame_data)
+
+    assert repr(pandas_df) == repr(ray_df)
+
 
 @pytest.fixture
 def test_loc(ray_df, pd_df):