[jira] [Created] (FLINK-20061) Row constructor unsupported in aggregation function

["Nico Kruber (Jira)" <ji...@apache.org>]

Nico Kruber created FLINK-20061:
-----------------------------------

             Summary: Row constructor unsupported in aggregation function
                 Key: FLINK-20061
                 URL: https://issues.apache.org/jira/browse/FLINK-20061
             Project: Flink
          Issue Type: Bug
          Components: Table SQL / API
    Affects Versions: 1.11.2
            Reporter: Nico Kruber


I was trying to use {{ROW}} in a user-defined aggregate function in a query like this:
{code}
  SELECT `id`, HOP_END(`timestamp`, INTERVAL '10' MINUTE, INTERVAL '1' MINUTE) AS `window_end`,
    RowMaxv0(`amount`, ROW(`timestamp`, `amount`, `payload`)) AS `max_amount`
  FROM `input`
  GROUP BY HOP(`timestamp`, INTERVAL '10' MINUTE, INTERVAL '1' MINUTE), `id`;
{code}

Eventually this resulted in an "unsupported" exception from Calcite:
{code}
Exception in thread "main" org.apache.flink.table.api.ValidationException: SQL validation failed. null
	at org.apache.flink.table.planner.calcite.FlinkPlannerImpl.org$apache$flink$table$planner$calcite$FlinkPlannerImpl$$validate(FlinkPlannerImpl.scala:146)
	at org.apache.flink.table.planner.calcite.FlinkPlannerImpl.validate(FlinkPlannerImpl.scala:108)
	at org.apache.flink.table.planner.operations.SqlToOperationConverter.convert(SqlToOperationConverter.java:187)
	at org.apache.flink.table.planner.delegation.ParserImpl.parse(ParserImpl.java:78)
	at org.apache.flink.table.api.internal.TableEnvironmentImpl.executeSql(TableEnvironmentImpl.java:684)
	at com.ververica.platform.sql.functions.RowMaxv0.main(RowMaxv0.java:93)
Caused by: java.lang.UnsupportedOperationException
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validateColumnListParams(SqlValidatorImpl.java:5689)
	at org.apache.calcite.sql.SqlFunction.deriveType(SqlFunction.java:268)
	at org.apache.calcite.sql.SqlFunction.deriveType(SqlFunction.java:218)
	at org.apache.calcite.sql.validate.SqlValidatorImpl$DeriveTypeVisitor.visit(SqlValidatorImpl.java:5858)
	at org.apache.calcite.sql.validate.SqlValidatorImpl$DeriveTypeVisitor.visit(SqlValidatorImpl.java:5845)
	at org.apache.calcite.sql.SqlCall.accept(SqlCall.java:139)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.deriveTypeImpl(SqlValidatorImpl.java:1800)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.deriveType(SqlValidatorImpl.java:1785)
	at org.apache.calcite.sql.SqlAsOperator.deriveType(SqlAsOperator.java:133)
	at org.apache.calcite.sql.validate.SqlValidatorImpl$DeriveTypeVisitor.visit(SqlValidatorImpl.java:5858)
	at org.apache.calcite.sql.validate.SqlValidatorImpl$DeriveTypeVisitor.visit(SqlValidatorImpl.java:5845)
	at org.apache.calcite.sql.SqlCall.accept(SqlCall.java:139)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.deriveTypeImpl(SqlValidatorImpl.java:1800)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.deriveType(SqlValidatorImpl.java:1785)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.expandSelectItem(SqlValidatorImpl.java:481)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validateSelectList(SqlValidatorImpl.java:4255)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validateSelect(SqlValidatorImpl.java:3523)
	at org.apache.calcite.sql.validate.SelectNamespace.validateImpl(SelectNamespace.java:60)
	at org.apache.calcite.sql.validate.AbstractNamespace.validate(AbstractNamespace.java:84)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validateNamespace(SqlValidatorImpl.java:1110)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validateQuery(SqlValidatorImpl.java:1084)
	at org.apache.calcite.sql.SqlSelect.validate(SqlSelect.java:232)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validateScopedExpression(SqlValidatorImpl.java:1059)
	at org.apache.calcite.sql.validate.SqlValidatorImpl.validate(SqlValidatorImpl.java:766)
	at org.apache.flink.table.planner.calcite.FlinkPlannerImpl.org$apache$flink$table$planner$calcite$FlinkPlannerImpl$$validate(FlinkPlannerImpl.scala:141)
	... 5 more
{code}

A workaround for this is to go via a subquery like the following but ultimately, this should result in the same thing (a simple projection).
{code}
  SELECT `id`, HOP_END(`timestamp`, INTERVAL '10' MINUTE, INTERVAL '1' MINUTE) AS `window_end`,
    RowMaxv0(`amount`, `row`) AS `max_amount`
  FROM (SELECT `id`, `timestamp`, `amount`, ROW(`timestamp`, `amount`, `payload`) AS `row` FROM `input`)
  GROUP BY HOP(`timestamp`, INTERVAL '10' MINUTE, INTERVAL '1' MINUTE), `id`
{code}



--
This message was sent by Atlassian Jira
(v8.3.4#803005)