measure/src/main/scala/org/apache/griffin/measure/step/write/RecordWriteStep.scala

/*
Licensed to the Apache Software Foundation (ASF) under one
or more contributor license agreements.  See the NOTICE file
distributed with this work for additional information
regarding copyright ownership.  The ASF licenses this file
to you under the Apache License, Version 2.0 (the
"License"); you may not use this file except in compliance
with the License.  You may obtain a copy of the License at

  http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing,
software distributed under the License is distributed on an
"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
KIND, either express or implied.  See the License for the
specific language governing permissions and limitations
under the License.
*/
package org.apache.griffin.measure.step.write

import org.apache.spark.rdd.RDD
import org.apache.spark.sql._

import org.apache.griffin.measure.configuration.enums._
import org.apache.griffin.measure.context.DQContext
import org.apache.griffin.measure.step.builder.ConstantColumns
import org.apache.griffin.measure.utils.JsonUtil

/**
  * write records needs to be sink
  */
case class RecordWriteStep(name: String,
                           inputName: String,
                           filterTableNameOpt: Option[String] = None,
                           writeTimestampOpt: Option[Long] = None
                          ) extends WriteStep {

  def execute(context: DQContext): Boolean = {
    val timestamp = writeTimestampOpt.getOrElse(context.contextId.timestamp)

    val writeMode = writeTimestampOpt.map(_ => SimpleMode).getOrElse(context.writeMode)
    writeMode match {
      case SimpleMode =>
        // batch records
        val recordsOpt = getBatchRecords(context)
        // write records
        recordsOpt match {
          case Some(records) =>
            context.getSink(timestamp).sinkRecords(records, name)
          case _ =>
        }
      case TimestampMode =>
        // streaming records
        val (recordsOpt, emptyTimestamps) = getStreamingRecords(context)
        // write records
        recordsOpt.foreach { records =>
          records.foreach { pair =>
            val (t, strs) = pair
            context.getSink(t).sinkRecords(strs, name)
          }
        }
        emptyTimestamps.foreach { t =>
          context.getSink(t).sinkRecords(Nil, name)
        }
    }
    true
  }

  private def getTmst(row: Row, defTmst: Long): Long = {
    try {
      row.getAs[Long](ConstantColumns.tmst)
    } catch {
      case _: Throwable => defTmst
    }
  }

  private def getDataFrame(context: DQContext, name: String): Option[DataFrame] = {
    try {
      val df = context.sqlContext.table(s"`${name}`")
      Some(df)
    } catch {
      case e: Throwable =>
        error(s"get data frame ${name} fails", e)
        None
    }
  }

  private def getRecordDataFrame(context: DQContext): Option[DataFrame]
    = getDataFrame(context, inputName)

  private def getFilterTableDataFrame(context: DQContext): Option[DataFrame]
    = filterTableNameOpt.flatMap(getDataFrame(context, _))

  private def getBatchRecords(context: DQContext): Option[RDD[String]] = {
    getRecordDataFrame(context).map(_.toJSON.rdd);
  }

  private def getStreamingRecords(context: DQContext)
    : (Option[RDD[(Long, Iterable[String])]], Set[Long])
    = {
    implicit val encoder = Encoders.tuple(Encoders.scalaLong, Encoders.STRING)
    val defTimestamp = context.contextId.timestamp
    getRecordDataFrame(context) match {
      case Some(df) =>
        val (filterFuncOpt, emptyTimestamps) = getFilterTableDataFrame(context) match {
          case Some(filterDf) =>
            // timestamps with empty flag
            val tmsts: Array[(Long, Boolean)] = (filterDf.collect.flatMap { row =>
              try {
                val tmst = getTmst(row, defTimestamp)
                val empty = row.getAs[Boolean](ConstantColumns.empty)
                Some((tmst, empty))
              } catch {
                case _: Throwable => None
              }
            })
            val emptyTmsts = tmsts.filter(_._2).map(_._1).toSet
            val recordTmsts = tmsts.filter(!_._2).map(_._1).toSet
            val filterFuncOpt: Option[(Long) => Boolean] = if (recordTmsts.size > 0) {
              Some((t: Long) => recordTmsts.contains(t))
            } else None

            (filterFuncOpt, emptyTmsts)
          case _ => (Some((t: Long) => true), Set[Long]())
        }

        // filter timestamps need to record
        filterFuncOpt match {
          case Some(filterFunc) =>
            val records = df.flatMap { row =>
              val tmst = getTmst(row, defTimestamp)
              if (filterFunc(tmst)) {
                try {
                  val map = SparkRowFormatter.formatRow(row)
                  val str = JsonUtil.toJson(map)
                  Some((tmst, str))
                } catch {
                  case e: Throwable => None
                }
              } else None
            }
            (Some(records.rdd.groupByKey), emptyTimestamps)
          case _ => (None, emptyTimestamps)
        }
      case _ => (None, Set[Long]())
    }
  }

}