SNOW-2362373: Fix bug with groupby.agg named aggregation path

CHANGELOG.md

@@ -67,6 +67,7 @@
 - Fixed a bug where writing Snowpark pandas dataframes on the pandas backend with a column multiindex to Snowflake with `to_snowflake` would raise `KeyError`.
 - Fixed a bug that `DataFrameReader.dbapi` (PuPr) is not compatible with oracledb 3.4.0.
 - Fixed a bug where `modin` would unintentionally be imported during session initialization in some scenarios.
+- Fixed a bug with `DataFrameGroupBy.agg` where tuples are treated as multiindex level using the named aggregation code path.
 
 #### Improvements
 

src/snowflake/snowpark/modin/plugin/_internal/aggregation_utils.py

@@ -899,7 +899,9 @@ def _is_supported_snowflake_agg_func(
     if isinstance(agg_func, tuple) and len(agg_func) == 2:
         # For named aggregations, like `df.agg(new_col=("old_col", "sum"))`,
         # take the second part of the named aggregation.
-        agg_func = agg_func[0]
+        agg_func = (
+            agg_func.func if isinstance(agg_func, AggFuncWithLabel) else agg_func[1]
+        )
 
     if get_snowflake_agg_func(agg_func, agg_kwargs, axis, _is_df_agg) is None:
         return AggregationSupportResult(
@@ -1378,13 +1380,18 @@ def get_agg_func_to_col_map(
     return agg_func_to_col_map
 
 
-def get_pandas_aggr_func_name(aggfunc: AggFuncTypeBase) -> str:
+def get_pandas_aggr_func_name(aggfunc: AggFuncTypeBase) -> Union[str, Any, None]:
     """
     Returns the friendly name for the aggr function.  For example, if it is a callable, it will return __name__
-    otherwise the same string name value.
+    otherwise the same string name value. If aggfunc is a tuple, treat as named aggregation and return
+    the first part of the name.
     """
     return (
-        getattr(aggfunc, "__name__", str(aggfunc))
+        getattr(
+            aggfunc,
+            "__name__",
+            aggfunc[0] if isinstance(aggfunc, tuple) else str(aggfunc),
+        )
         if not isinstance(aggfunc, str)
         else aggfunc
     )
@@ -1536,7 +1543,12 @@ def generate_column_agg_info(
         for func_info, label, identifier in zip(
             agg_func_list, agg_col_labels, agg_col_identifiers
         ):
-            func = func_info.func
+            # If func_info.func is a tuple, treat as named aggregation and return the aggregate function
+            func = (
+                func_info.func[1]
+                if isinstance(func_info.func, tuple)
+                else func_info.func
+            )
             is_dummy_agg = func_info.is_dummy_agg
             agg_func_col = pandas_lit(None) if is_dummy_agg else quoted_identifier
             snowflake_agg_func = get_snowflake_agg_func(func, agg_kwargs, axis=0)

tests/integ/modin/groupby/test_groupby_basic_agg.py

@@ -1321,3 +1321,23 @@ def test_groupby_timedelta_var(self):
             ),
             lambda df: df.groupby("A").var(),
         )
+
+    @sql_count_checker(query_count=1)
+    def test_groupby_agg_named_aggregation_implicit(self):
+        eval_snowpark_pandas_result(
+            *create_test_dfs(
+                {
+                    "team": ["A", "B"],
+                    "score": [10, 15],
+                }
+            ),
+            lambda df: df.groupby("team").agg({"score": [("total_score", "sum")]}),
+        )
+
+    @sql_count_checker(query_count=1)
+    def test_groupby_agg_named_aggregation_explicit(self):
+        agg1 = pd.NamedAgg(column="score", aggfunc="sum")
+        eval_snowpark_pandas_result(
+            *create_test_dfs({"team": ["A", "B"], "score": [10, 15]}),
+            lambda df: df.groupby("team").agg(total_score=agg1),
+        )