measure/src/test/scala/org/apache/griffin/measure/transformations/AccuracyTransformationsIntegrationTest.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.transformations

import org.apache.spark.sql.DataFrame
import org.scalatest._

import org.apache.griffin.measure.configuration.dqdefinition._
import org.apache.griffin.measure.configuration.enums.BatchProcessType
import org.apache.griffin.measure.context.{ContextId, DQContext}
import org.apache.griffin.measure.datasource.DataSourceFactory
import org.apache.griffin.measure.job.builder.DQJobBuilder
import org.apache.griffin.measure.SparkSuiteBase

case class AccuracyResult(total: Long, miss: Long, matched: Long, matchedFraction: Double)

class AccuracyTransformationsIntegrationTest extends FlatSpec with Matchers with SparkSuiteBase {
  private val EMPTY_PERSON_TABLE = "empty_person"
  private val PERSON_TABLE = "person"

  override def beforeAll(): Unit = {
    super.beforeAll()

    dropTables()
    createPersonTable()
    createEmptyPersonTable()

    spark.conf.set("spark.sql.crossJoin.enabled", "true")
  }

  override def afterAll(): Unit = {
    dropTables()
    super.afterAll()
  }

  "accuracy" should "basically work" in {
    checkAccuracy(
      sourceName = PERSON_TABLE,
      targetName = PERSON_TABLE,
      expectedResult = AccuracyResult(total = 2, miss = 0, matched = 2, matchedFraction = 1.0))
  }

  "accuracy" should "work with empty target" in {
    checkAccuracy(
      sourceName = PERSON_TABLE,
      targetName = EMPTY_PERSON_TABLE,
      expectedResult = AccuracyResult(total = 2, miss = 2, matched = 0, matchedFraction = 0.0)
    )
  }

  "accuracy" should "work with empty source" in {
    checkAccuracy(
      sourceName = EMPTY_PERSON_TABLE,
      targetName = PERSON_TABLE,
      expectedResult = AccuracyResult(total = 0, miss = 0, matched = 0, matchedFraction = 1.0))
  }

  "accuracy" should "work with empty source and target" in {
    checkAccuracy(
      sourceName = EMPTY_PERSON_TABLE,
      targetName = EMPTY_PERSON_TABLE,
      expectedResult = AccuracyResult(total = 0, miss = 0, matched = 0, matchedFraction = 1.0))
  }

  private def checkAccuracy(sourceName: String, targetName: String, expectedResult: AccuracyResult) = {
    val dqContext: DQContext = getDqContext(
      dataSourcesParam = List(
        DataSourceParam(
          name = "source",
          connectors = List(dataConnectorParam(tableName = sourceName))
        ),
        DataSourceParam(
          name = "target",
          connectors = List(dataConnectorParam(tableName = targetName))
        )
      ))

    val accuracyRule = RuleParam(
      dslType = "griffin-dsl",
      dqType = "ACCURACY",
      outDfName = "person_accuracy",
      rule = "source.name = target.name"
    )

    val spark = this.spark
    import spark.implicits._
    val res = getRuleResults(dqContext, accuracyRule)
      .as[AccuracyResult]
      .collect()

    res.length shouldBe 1

    res(0) shouldEqual expectedResult
  }

  private def getRuleResults(dqContext: DQContext, rule: RuleParam): DataFrame = {
    val dqJob = DQJobBuilder.buildDQJob(
      dqContext,
      evaluateRuleParam = EvaluateRuleParam(List(rule))
    )

    dqJob.execute(dqContext)

    spark.sql(s"select * from ${rule.getOutDfName()}")
  }

  private def createPersonTable(): Unit = {
    val personCsvPath = getClass.getResource("/hive/person_table.csv").getFile

    spark.sql(
      s"CREATE TABLE ${PERSON_TABLE} " +
        "( " +
        "  name String," +
        "  age int " +
        ") " +
        "ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n' " +
        "STORED AS TEXTFILE"
    )

    spark.sql(s"LOAD DATA LOCAL INPATH '$personCsvPath' OVERWRITE INTO TABLE ${PERSON_TABLE}")
  }

  private def createEmptyPersonTable(): Unit = {
    spark.sql(
      s"CREATE TABLE ${EMPTY_PERSON_TABLE} " +
        "( " +
        "  name String," +
        "  age int " +
        ") " +
        "ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' LINES TERMINATED BY '\n' " +
        "STORED AS TEXTFILE"
    )

    spark.sql(s"select * from ${EMPTY_PERSON_TABLE}").show()
  }

  private def dropTables(): Unit = {
    spark.sql(s"DROP TABLE IF EXISTS ${PERSON_TABLE} ")
    spark.sql(s"DROP TABLE IF EXISTS ${EMPTY_PERSON_TABLE} ")
  }

  private def getDqContext(dataSourcesParam: Seq[DataSourceParam], name: String = "test-context"): DQContext = {
    val dataSources = DataSourceFactory.getDataSources(spark, null, dataSourcesParam)
    dataSources.foreach(_.init())

    DQContext(
      ContextId(System.currentTimeMillis),
      name,
      dataSources,
      Nil,
      BatchProcessType
    )(spark)
  }

  private def dataConnectorParam(tableName: String) = {
    DataConnectorParam(
      conType = "HIVE",
      version = null,
      dataFrameName = null,
      config = Map("table.name" -> tableName),
      preProc = null
    )
  }
}