[ https://issues.apache.org/jira/browse/SPARK-15810?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15370494#comment-15370494 ]
Amit Sela commented on SPARK-15810: ----------------------------------- [~rxin] what's your take on the Java API issue with null ? does this belong here ? do you think it should be a separate issue ? > Aggregator doesn't play nice with Option > ---------------------------------------- > > Key: SPARK-15810 > URL: https://issues.apache.org/jira/browse/SPARK-15810 > Project: Spark > Issue Type: Sub-task > Components: SQL > Environment: spark 2.0.0-SNAPSHOT > Reporter: koert kuipers > > {code} > val ds1 = List(("a", 1), ("a", 2), ("a", 3)).toDS > val ds2 = ds1.map{ case (k, v) => (k, if (v > 1) Some(v) else None) } > val ds3 = ds2.groupByKey(_._1).agg(new Aggregator[(String, Option[Int]), > Option[Int], Option[Int]]{ > def zero: Option[Int] = None > def reduce(b: Option[Int], a: (String, Option[Int])): Option[Int] = > b.map(bv => a._2.map(av => bv + av).getOrElse(bv)).orElse(a._2) > def merge(b1: Option[Int], b2: Option[Int]): Option[Int] = b1.map(b1v => > b2.map(b2v => b1v + b2v).getOrElse(b1v)).orElse(b2) > def finish(reduction: Option[Int]): Option[Int] = reduction > def bufferEncoder: Encoder[Option[Int]] = implicitly[Encoder[Option[Int]]] > def outputEncoder: Encoder[Option[Int]] = implicitly[Encoder[Option[Int]]] > }.toColumn) > ds3.printSchema > ds3.show > {code} > i get as output a somewhat odd looking schema, and after that the program > just hangs pinning one cpu at 100%. the data never shows. > output: > {noformat} > root > |-- value: string (nullable = true) > |-- $anon$1(scala.Tuple2): struct (nullable = true) > | |-- value: integer (nullable = true) > {noformat} -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org