Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/13494#discussion_r68377009 --- Diff: sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveMetastoreCatalog.scala --- @@ -457,6 +458,125 @@ private[hive] class HiveMetastoreCatalog(sparkSession: SparkSession) extends Log allowExisting) } } + + /** + * When scanning only partition columns, get results based on metadata without scanning files. + * It is used for distinct or distinct/Max/Min aggregations, example: max(partition). + */ + object MetadataOnlyOptimizer extends Rule[LogicalPlan] { + + private def canSupportMetadataOnly(a: Aggregate): Boolean = { + val aggregateExpressions = a.aggregateExpressions.flatMap { expr => + expr.collect { + case agg: AggregateExpression => agg + } + }.distinct + aggregateExpressions.forall { agg => + if (agg.isDistinct) { + true + } else { + agg.aggregateFunction match { + case max: Max => true + case min: Min => true + case _ => false + } + } + } + } + + private def findRelation(plan: LogicalPlan): (Option[LogicalPlan], Seq[Expression]) = { + plan match { + case relation @ LogicalRelation(files: HadoopFsRelation, _, table) + if files.partitionSchema.nonEmpty => + (Some(relation), Seq.empty[Expression]) + + case relation: MetastoreRelation if relation.partitionKeys.nonEmpty => + (Some(relation), Seq.empty[Expression]) + + case p @ Project(_, child) => --- End diff -- is it safe to keep traversing the plan tree through `Project`?
--- 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. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org