From 22a4a35ce369914fbf17f1d106f09e0e002c4bd9 Mon Sep 17 00:00:00 2001
From: Hi-king <hikingko1@gmail.com>
Date: Sun, 24 May 2020 17:42:22 +0900
Subject: [PATCH 1/2] :bug: allow gokart.task.TaskOnKart.load_data_frame to
 handle index only dataframe

---
 gokart/task.py            | 19 +++++++------------
 test/test_task_on_kart.py | 12 ++++++++++++
 2 files changed, 19 insertions(+), 12 deletions(-)

diff --git a/gokart/task.py b/gokart/task.py
index 2f3bda07..892aea61 100644
--- a/gokart/task.py
+++ b/gokart/task.py
@@ -191,21 +191,16 @@ def load_data_frame(self,
                         target: Union[None, str, TargetOnKart] = None,
                         required_columns: Optional[Set[str]] = None,
                         drop_columns: bool = False) -> pd.DataFrame:
-        data = self.load(target=target)
-        if isinstance(data, list):
-
-            def _pd_concat(dfs):
-                if isinstance(dfs, list):
-                    return pd.concat([_pd_concat(df) for df in dfs])
-                else:
-                    return dfs
-
-            data = _pd_concat(data)
+        def _pd_concat(dfs):
+            if isinstance(dfs, list):
+                return pd.concat([_pd_concat(df) for df in dfs])
+            else:
+                return dfs
+        data = _pd_concat(self.load(target=target))
 
         required_columns = required_columns or set()
-        if data.empty:
+        if data.empty and len(data.index) == 0:
             return pd.DataFrame(columns=required_columns)
-
         assert required_columns.issubset(set(data.columns)), f'data must have columns {required_columns}, but actually have only {data.columns}.'
         if drop_columns:
             data = data[required_columns]
diff --git a/test/test_task_on_kart.py b/test/test_task_on_kart.py
index dc6a510f..a99d0495 100644
--- a/test/test_task_on_kart.py
+++ b/test/test_task_on_kart.py
@@ -277,6 +277,18 @@ def test_load_data_frame_drop_columns(self):
         self.assertEqual(1, df.shape[0])
         self.assertSetEqual({'a', 'c'}, set(df.columns))
 
+    def test_load_index_only_dataframe(self):
+        task = _DummyTask()
+        task.load = MagicMock(return_value=pd.DataFrame(index=range(3)))
+
+        # connnot load index only frame with required_columns
+        self.assertRaises(AssertionError, lambda : task.load_data_frame(required_columns={'a', 'c'}))
+
+        df: pd.DataFrame = task.load_data_frame()
+        self.assertIsInstance(df, pd.DataFrame)
+        self.assertTrue(df.empty)
+        self.assertListEqual(list(range(3)), list(df.index))
+
     def test_use_rerun_with_inherits(self):
         # All tasks are completed.
         task_c = _DummyTaskC()

From 73c7ed486018d1e1b2def7e8cb8ec68b26ae9c06 Mon Sep 17 00:00:00 2001
From: Hi-king <hikingko1@gmail.com>
Date: Sun, 24 May 2020 17:50:11 +0900
Subject: [PATCH 2/2] :art: rename _flatten_recursively

---
 gokart/task.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/gokart/task.py b/gokart/task.py
index 892aea61..b4e7724e 100644
--- a/gokart/task.py
+++ b/gokart/task.py
@@ -191,12 +191,12 @@ def load_data_frame(self,
                         target: Union[None, str, TargetOnKart] = None,
                         required_columns: Optional[Set[str]] = None,
                         drop_columns: bool = False) -> pd.DataFrame:
-        def _pd_concat(dfs):
+        def _flatten_recursively(dfs):
             if isinstance(dfs, list):
-                return pd.concat([_pd_concat(df) for df in dfs])
+                return pd.concat([_flatten_recursively(df) for df in dfs])
             else:
                 return dfs
-        data = _pd_concat(self.load(target=target))
+        data = _flatten_recursively(self.load(target=target))
 
         required_columns = required_columns or set()
         if data.empty and len(data.index) == 0: