SNOW-1569896 Refactor index to take in DataFrame/Series (#2020)

src/snowflake/snowpark/modin/pandas/base.py

@@ -667,7 +667,11 @@ def _get_index(self):
             The union of all indexes across the partitions.
         """
         # TODO: SNOW-1119855: Modin upgrade - modin.pandas.base.BasePandasDataset
-        return self._query_compiler.index
+        from snowflake.snowpark.modin.plugin.extensions.index import Index
+
+        if self._query_compiler.is_multiindex():
+            return self._query_compiler.index
+        return Index(data=self)
 
     index = property(_get_index, _set_index)
 

src/snowflake/snowpark/modin/plugin/extensions/index.py

@@ -35,6 +35,7 @@
 from pandas.core.dtypes.common import pandas_dtype
 
 from snowflake.snowpark.modin.pandas import DataFrame, Series
+from snowflake.snowpark.modin.pandas.base import BasePandasDataset
 from snowflake.snowpark.modin.pandas.utils import try_convert_index_to_native
 from snowflake.snowpark.modin.plugin._internal.telemetry import TelemetryMeta
 from snowflake.snowpark.modin.plugin.compiler.snowflake_query_compiler import (
@@ -50,7 +51,11 @@
 class Index(metaclass=TelemetryMeta):
     def __init__(
         self,
-        data: ArrayLike | SnowflakeQueryCompiler | None = None,
+        data: ArrayLike
+        | modin.pandas.DataFrame
+        | Series
+        | SnowflakeQueryCompiler
+        | None = None,
         dtype: str | np.dtype | ExtensionDtype | None = None,
         copy: bool = False,
         name: object = None,
@@ -63,7 +68,7 @@ def __init__(
 
         Parameters
         ----------
-        data : array-like (1-dimensional)
+        data : array-like (1-dimensional), modin.pandas.Series, modin.pandas.DataFrame, SnowflakeQueryCompiler, optional
         dtype : str, numpy.dtype, or ExtensionDtype, optional
             Data type for the output Index. If not specified, this will be
             inferred from `data`.
@@ -92,6 +97,8 @@ def __init__(
         >>> pd.Index([1, 2, 3], dtype="uint8")
         Index([1, 2, 3], dtype='int64')
         """
+        self._parent = data if isinstance(data, BasePandasDataset) else None
+        data = data._query_compiler if isinstance(data, BasePandasDataset) else data
         if isinstance(data, SnowflakeQueryCompiler):
             qc = data
         else:

tests/integ/modin/index/test_index_methods.py

@@ -3,6 +3,7 @@
 #
 
 import modin.pandas as pd
+import pandas as native_pd
 import pytest
 from numpy.testing import assert_equal
 from pandas._libs import lib
@@ -15,6 +16,7 @@
 )
 from tests.integ.modin.sql_counter import SqlCounter, sql_count_checker
 from tests.integ.modin.utils import (
+    assert_frame_equal,
     assert_index_equal,
     assert_series_equal,
     assert_snowpark_pandas_equals_to_pandas_without_dtypecheck,
@@ -341,3 +343,24 @@ def test_has_duplicates(index):
     with SqlCounter(query_count=1):
         snow_index = pd.Index(index)
         assert index.has_duplicates == snow_index.has_duplicates
+
+
+@sql_count_checker(query_count=6)
+def test_index_parent():
+    """
+    Check whether the parent field in Index is updated properly.
+    """
+    native_idx1 = native_pd.Index(["A", "B"], name="xyz")
+    native_idx2 = native_pd.Index(["A", "B", "D", "E", "G", "H"], name="CFI")
+
+    # DataFrame case.
+    df = pd.DataFrame([[1, 2], [3, 4]], index=native_idx1)
+    snow_idx = df.index
+    assert_frame_equal(snow_idx._parent, df)
+    assert_index_equal(snow_idx, native_idx1)
+
+    # Series case.
+    s = pd.Series([1, 2, 4, 5, 6, 7], index=native_idx2, name="zyx")
+    snow_idx = s.index
+    assert_series_equal(snow_idx._parent, s)
+    assert_index_equal(snow_idx, native_idx2)