過濾器計數火花 dataframe

Question

我有兩個如下數據幀，我從 MySQL 表中讀取邏輯 DF

邏輯 DF：

slNo | filterCondtion |
-----------------------
1    | age > 100      |
2    | age > 50       |
3    | age > 10       |
4    | age > 20       |

InputDF - 從文件中讀取：

age   | name           |
------------------------
11    | suraj          |
22    | surjeth        |
33    | sam            |
43    | ram            |

我想從邏輯數據框中應用過濾器語句並添加這些過濾器的計數

結果 output：

slNo | filterCondtion | count |
------------------------------
1    | age > 100      |   10  |
2    | age > 50       |   2   |
3    | age > 10       |   5   |
4    | age > 20       |   6   |
-------------------------------

我嘗試過的代碼：

val LogicDF = spark.read.format("jdbc").option("url", "jdbc:mysql://localhost:3306/testDB").option("driver", "com.mysql.jdbc.Driver").option("dbtable", "logic_table").option("user", "root").option("password", "password").load()

def filterCount(str: String): Long ={
     val counte = inputDF.where(str).count()
counte
}

val filterCountUDF = udf[Long, String](filterCount)

LogicDF.withColumn("count",filterCountUDF(col("filterCondtion")))

錯誤跟蹤：

Caused by: org.apache.spark.SparkException: Failed to execute user defined function($anonfun$1: (string) => bigint)
  at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIteratorForCodegenStage1.processNext(Unknown Source)
  at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
  at org.apache.spark.sql.execution.WholeStageCodegenExec$$anonfun$11$$anon$1.hasNext(WholeStageCodegenExec.scala:619)
  at org.apache.spark.sql.execution.SparkPlan$$anonfun$2.apply(SparkPlan.scala:255)
  at org.apache.spark.sql.execution.SparkPlan$$anonfun$2.apply(SparkPlan.scala:247)
  at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$1$$anonfun$apply$24.apply(RDD.scala:836)
  at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$1$$anonfun$apply$24.apply(RDD.scala:836)
  at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
  at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
  at org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
  at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
  at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
  at org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
  at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90)
  at org.apache.spark.scheduler.Task.run(Task.scala:121)
  at org.apache.spark.executor.Executor$TaskRunner$$anonfun$10.apply(Executor.scala:402)
  at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360)
  at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:408)
  at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
  at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
  at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.NullPointerException
  at org.apache.spark.sql.Dataset.where(Dataset.scala:1525)
  at filterCount(<console>:28)
  at $anonfun$1.apply(<console>:25)
  at $anonfun$1.apply(<console>:25)
  ... 21 more

任何替代方案也可以...提前致謝。

Answer 1

沒有 UDF 的解決方案

只要您的 logicDF 小到可以收集到驅動程序中，這將起作用。

步驟1

將您的邏輯收集到Array[(Int, String)]中，如下所示：

val rules = logicDF.collect().map{ r: Row =>
  val slNo = r.getAs[Int](0)
  val condition = r.getAs[String](1)
  (slNo, condition)
}

第2步

使用條件值構建一個新列，將這些規則鏈接到 when Column中。 為此，請使用一些 scala 循環，例如：

val unused = when(lit(false), lit(false))
val filters: Column = rules.foldLeft(unused){
  case (acc: Column, (slNo: Int, cond: String)) =>
    acc.when(col("slNo") === slNo, expr(cond))
}

//You will get something like:
//when(col("slNo") === 1, expr("age > 10"))
//.when(col("slNo") === 2, expr("age > 20"))
//...

第 3 步

通過連接獲取兩個 DataFrame 的笛卡爾積，因此您可以將每個規則應用於數據中的每一行：

val joinDF = logicDF.join(inputDF, lit(true), "inner") //inner or whatever

第4步

使用帶有條件過濾器的前一Column進行過濾。

val withRulesDF = joinDF.filter(filters)

第 5 步

分組和計數：

val resultDF = withRulesDF
  .groupBy("slNo", "filterCondtion")
  .agg(count("*") as "count")

Answer 2

package spark

import org.apache.spark.sql.{DataFrame, SparkSession}
import org.apache.spark.sql.functions._

object LogicFilterDataFrame extends App {
  val spark = SparkSession.builder()
    .master("local")
    .appName("DataFrame-example")
    .getOrCreate()

  import spark.implicits._

  case class LogicFilter(slNo: Int, filterCondition: String)
  case class Data(age: Int, name:String)

  val logicDF = Seq(
    LogicFilter(1, "age > 100"),
    LogicFilter(2, "age > 50"),
    LogicFilter(3, "age > 10"),
    LogicFilter(4, "age > 20")
  ).toDF()

  val dataDF = Seq(
    Data(11, "suraj"),
    Data(22, "surjeth"),
    Data(33, "sam"),
    Data(43, "ram")
  ).toDF()

  val logicCount = udf{s: String => {
    dataDF.filter(s).count()
    }}
  val resDF = logicDF.filter('filterCondition.like("%age%")).withColumn("count", logicCount('filterCondition))
  resDF.show(false)

}