This is an automated email from the ASF dual-hosted git repository.
lgbo pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/incubator-gluten.git
The following commit(s) were added to refs/heads/main by this push:
new faa1efb295 fix (#8999)
faa1efb295 is described below
commit faa1efb2951868f739bb5f6ea0c663aa926bcfc1
Author: kevinyhzou <[email protected]>
AuthorDate: Tue Mar 18 18:02:05 2025 +0800
fix (#8999)
---
.../execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala | 6 ++++++
cpp-ch/local-engine/AggregateFunctions/GroupLimitFunctions.cpp | 1 -
cpp-ch/local-engine/Parser/RelParsers/GroupLimitRelParser.cpp | 5 +++--
3 files changed, 9 insertions(+), 3 deletions(-)
diff --git
a/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
b/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
index c30577c339..3ce03565e8 100644
---
a/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
+++
b/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
@@ -3390,5 +3390,11 @@ class GlutenClickHouseTPCHSaltNullParquetSuite extends
GlutenClickHouseTPCHAbstr
spark.sql("drop table test_tbl_8343")
}
+ test("GLUTEN-8995: Fix column not found in row_number") {
+ val select_sql =
+ "select id from (select id ,row_number() over (partition by id order by
id desc) as rank from range(1)) c1 where rank =1"
+ compareResultsAgainstVanillaSpark(select_sql, true, { _ => })
+ }
+
}
// scalastyle:on line.size.limit
diff --git a/cpp-ch/local-engine/AggregateFunctions/GroupLimitFunctions.cpp
b/cpp-ch/local-engine/AggregateFunctions/GroupLimitFunctions.cpp
index 67c9e8380a..20ef632ee6 100644
--- a/cpp-ch/local-engine/AggregateFunctions/GroupLimitFunctions.cpp
+++ b/cpp-ch/local-engine/AggregateFunctions/GroupLimitFunctions.cpp
@@ -271,7 +271,6 @@ private:
DB::ErrorCodes::BAD_ARGUMENTS, "Not found column {} in
tuple {}", ident_name, argument_types[0]->getName());
}
field.pos = std::distance(tuple_element_names.begin(), name_pos);
-
fields.push_back(field);
}
return fields;
diff --git a/cpp-ch/local-engine/Parser/RelParsers/GroupLimitRelParser.cpp
b/cpp-ch/local-engine/Parser/RelParsers/GroupLimitRelParser.cpp
index 597d90adaa..468a9a26f5 100644
--- a/cpp-ch/local-engine/Parser/RelParsers/GroupLimitRelParser.cpp
+++ b/cpp-ch/local-engine/Parser/RelParsers/GroupLimitRelParser.cpp
@@ -267,9 +267,10 @@ void
AggregateGroupLimitRelParser::prePrejectionForAggregateArguments(DB::QueryP
{
auto projection_actions =
std::make_shared<DB::ActionsDAG>(input_header.getColumnsWithTypeAndName());
-
auto partition_fields =
parsePartitionFields(win_rel_def->partition_expressions());
+ auto sort_fields = parseSortFields(win_rel_def->sorts());
std::set<size_t> unique_partition_fields(partition_fields.begin(),
partition_fields.end());
+ std::set<size_t> unique_sort_fields(sort_fields.begin(),
sort_fields.end());
DB::NameSet required_column_names;
auto build_tuple = [&](const DB::DataTypes & data_types,
const Strings & names,
@@ -295,7 +296,7 @@ void
AggregateGroupLimitRelParser::prePrejectionForAggregateArguments(DB::QueryP
for (size_t i = 0; i < input_header.columns(); ++i)
{
const auto & col = input_header.getByPosition(i);
- if (unique_partition_fields.count(i))
+ if (unique_partition_fields.count(i) && !unique_sort_fields.count(i))
{
required_column_names.insert(col.name);
aggregate_grouping_keys.push_back(col.name);
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
