Github user fhueske commented on a diff in the pull request: https://github.com/apache/flink/pull/2792#discussion_r89088819 --- Diff: flink-libraries/flink-table/src/main/scala/org/apache/flink/api/table/runtime/aggregate/AggregateUtil.scala --- @@ -39,66 +45,69 @@ object AggregateUtil { type JavaList[T] = java.util.List[T] /** - * Create Flink operator functions for aggregates. It includes 2 implementations of Flink - * operator functions: - * [[org.apache.flink.api.common.functions.MapFunction]] and - * [[org.apache.flink.api.common.functions.GroupReduceFunction]](if it's partial aggregate, - * should also implement [[org.apache.flink.api.common.functions.CombineFunction]] as well). - * The output of [[org.apache.flink.api.common.functions.MapFunction]] contains the - * intermediate aggregate values of all aggregate function, it's stored in Row by the following - * format: - * - * {{{ - * avg(x) aggOffsetInRow = 2 count(z) aggOffsetInRow = 5 - * | | - * v v - * +---------+---------+--------+--------+--------+--------+ - * |groupKey1|groupKey2| sum1 | count1 | sum2 | count2 | - * +---------+---------+--------+--------+--------+--------+ - * ^ - * | - * sum(y) aggOffsetInRow = 4 - * }}} - * - */ - def createOperatorFunctionsForAggregates( - namedAggregates: Seq[CalcitePair[AggregateCall, String]], - inputType: RelDataType, - outputType: RelDataType, - groupings: Array[Int]) - : (MapFunction[Any, Row], RichGroupReduceFunction[Row, Row]) = { - - val aggregateFunctionsAndFieldIndexes = - transformToAggregateFunctions(namedAggregates.map(_.getKey), inputType, groupings.length) - // store the aggregate fields of each aggregate function, by the same order of aggregates. - val aggFieldIndexes = aggregateFunctionsAndFieldIndexes._1 - val aggregates = aggregateFunctionsAndFieldIndexes._2 + * Create prepare MapFunction for aggregates. + * The output of [[org.apache.flink.api.common.functions.MapFunction]] contains the + * intermediate aggregate values of all aggregate function, it's stored in Row by the following + * format: + * + * {{{ + * avg(x) aggOffsetInRow = 2 count(z) aggOffsetInRow = 5 + * | | + * v v + * +---------+---------+--------+--------+--------+--------+ + * |groupKey1|groupKey2| sum1 | count1 | sum2 | count2 | + * +---------+---------+--------+--------+--------+--------+ + * ^ + * | + * sum(y) aggOffsetInRow = 4 + * }}} + * + */ + private[flink] def createPrepareMapFunction( + aggregates: Array[Aggregate[_ <: Any]], --- End diff -- I think we should make the arguments of the `create*Function` methods more consistent. Either we use Calcite's `namedAggregates` or we use the transformed `aggregates` and `aggFieldIndexes`.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---