Explicitly disallow empty list as index_spark_colum_names and index_n…

…ames.

databricks/koalas/frame.py

@@ -3495,21 +3495,10 @@ def rename(index):
             scol_for(sdf, column).alias(name_like_string(name)) for column, name in new_index_map
         ]
 
-        if len(index_map) > 0:  # type: ignore
-            index_scols = [scol_for(sdf, column) for column in index_map]
-            sdf = sdf.select(
-                index_scols
-                + new_data_scols
-                + self._internal.data_spark_columns
-                + list(HIDDEN_COLUMNS)
-            )
-        else:
-            sdf = sdf.select(
-                new_data_scols + self._internal.data_spark_columns + list(HIDDEN_COLUMNS)
-            )
-
-            sdf = InternalFrame.attach_default_index(sdf)
-            index_map = OrderedDict({SPARK_DEFAULT_INDEX_NAME: None})
+        index_scols = [scol_for(sdf, column) for column in index_map]
+        sdf = sdf.select(
+            index_scols + new_data_scols + self._internal.data_spark_columns + list(HIDDEN_COLUMNS)
+        )
 
         if self._internal.column_labels_level > 1:
             column_depth = len(self._internal.column_labels[0])

databricks/koalas/groupby.py

@@ -61,7 +61,6 @@
 from databricks.koalas.config import get_option
 from databricks.koalas.utils import (
     align_diff_frames,
-    column_labels_level,
     is_name_like_tuple,
     is_name_like_value,
     name_like_string,
@@ -1318,29 +1317,28 @@ def _make_pandas_df_builder_func(kdf, func, return_schema, retain_index):
         index_names = kdf._internal.index_names
         data_columns = kdf._internal.data_spark_column_names
         column_labels = kdf._internal.column_labels
+        column_labels_level = kdf._internal.column_labels_level
 
         def rename_output(pdf):
             # TODO: This logic below was borrowed from `DataFrame.to_pandas_frame` to set the index
             #   within each pdf properly. we might have to deduplicate it.
             import pandas as pd
 
-            if len(index_columns) > 0:
-                append = False
-                for index_field in index_columns:
-                    drop = index_field not in data_columns
-                    pdf = pdf.set_index(index_field, drop=drop, append=append)
-                    append = True
-                pdf = pdf[data_columns]
+            append = False
+            for index_field in index_columns:
+                drop = index_field not in data_columns
+                pdf = pdf.set_index(index_field, drop=drop, append=append)
+                append = True
+            pdf = pdf[data_columns]
 
-            if column_labels_level(column_labels) > 1:
+            if column_labels_level > 1:
                 pdf.columns = pd.MultiIndex.from_tuples(column_labels)
             else:
                 pdf.columns = [None if label is None else label[0] for label in column_labels]
 
-            if len(index_names) > 0:
-                pdf.index.names = [
-                    name if name is None or len(name) > 1 else name[0] for name in index_names
-                ]
+            pdf.index.names = [
+                name if name is None or len(name) > 1 else name[0] for name in index_names
+            ]
 
             pdf = func(pdf)
 

databricks/koalas/internal.py

@@ -445,7 +445,7 @@ def __init__(
         assert isinstance(spark_frame, spark.DataFrame)
         assert not spark_frame.isStreaming, "Koalas does not support Structured Streaming."
 
-        if index_spark_column_names is None:
+        if not index_spark_column_names:
             assert not any(SPARK_INDEX_NAME_PATTERN.match(name) for name in spark_frame.columns), (
                 "Index columns should not appear in columns of the Spark DataFrame. Avoid "
                 "index column names [%s]." % SPARK_INDEX_NAME_PATTERN
@@ -470,7 +470,7 @@ def __init__(
                 NATURAL_ORDER_COLUMN_NAME, F.monotonically_increasing_id()
             )
 
-        if index_names is None:
+        if not index_names:
             index_names = [None] * len(index_spark_column_names)
 
         assert len(index_spark_column_names) == len(index_names), (
@@ -857,11 +857,10 @@ def to_pandas_frame(self) -> pd.DataFrame:
                 name=names[0],
             )
 
-        index_names = self.index_names
-        if len(index_names) > 0:
-            pdf.index.names = [
-                name if name is None or len(name) > 1 else name[0] for name in index_names
-            ]
+        pdf.index.names = [
+            name if name is None or len(name) > 1 else name[0] for name in self.index_names
+        ]
+
         return pdf
 
     @lazy_property

databricks/koalas/tests/test_dataframe.py

@@ -207,6 +207,7 @@ def test_reset_index(self):
         kdf = ks.from_pandas(pdf)
 
         self.assert_eq(kdf.reset_index(), pdf.reset_index())
+        self.assert_eq(kdf.reset_index().index, pdf.reset_index().index)
         self.assert_eq(kdf.reset_index(drop=True), pdf.reset_index(drop=True))
 
         pdf.index.name = "a"