bug: TypeError: 'NoneType' object is not callable while iterating Scanquery results

Bug Report

YDB Python SDK version:

ydb-3.3.5

Environment

Mac OS X

Current behavior:

for row_proto in message.rows:
row = _Row(message.columns)
for column, value, column_info in zip(message.columns, row_proto.items, column_parsers):
v_type = value.WhichOneof("value")
if v_type == "null_flag_value":
row[column.name] = None
continue

            while v_type == "nested_value":
                value = value.nested_value
                v_type = value.WhichOneof("value")

            column_parser, unwrapped_type = column_info

          row[column.name] = column_parser(unwrapped_type, value, table_client_settings)

E TypeError: 'NoneType' object is not callable

Expected behavior:

Steps to reproduce:

--!syntax_v1
pragma EmitAggApply;

SELECT
Pg::count(),Pg::count(key),Pg::min(key),Pg::max(key),Pg::sum(key),Pg::avg(key)
FROM (values (1),(2),(3)) as a(key)

        it = ydb_driver.table_client.scan_query(query, settings=settings)
        if it is None:
            return None

        while True:
            try:
                result = next(it)

Other information:

Start at the scan_query result iteration and the _Row assignment where column_parser is called, then reproduce the aggregate query shown in the issue. Trace why a parser is None for the returned values and verify that iterating these Scanquery results completes without TypeError.