Backend
CH (ClickHouse)
Bug description
For Array(Nullable(Array(T))), the code scans all nested elements and returns input_rows_count default nullable rows as soon as it sees one null inner array. This makes unrelated rows null. Example: [[[1], null], [[2]]] should only make the first row null under Spark-style flatten semantics, but the current implementation returns both rows as null.
Gluten version
No response
Spark version
None
Spark configurations
No response
System information
No response
Relevant logs
Backend
CH (ClickHouse)
Bug description
For
Array(Nullable(Array(T))), the code scans all nested elements and returnsinput_rows_countdefault nullable rows as soon as it sees one null inner array. This makes unrelated rows null. Example:[[[1], null], [[2]]]should only make the first row null under Spark-style flatten semantics, but the current implementation returns both rows as null.Gluten version
No response
Spark version
None
Spark configurations
No response
System information
No response
Relevant logs