2

これは、sparkデータフレームにcsvファイルをロードする方法です

val sqlContext = new org.apache.spark.sql.SQLContext(sc)
import sqlContext.implicits._

import org.apache.spark.{ SparkConf, SparkContext }
import java.sql.{Date, Timestamp}
import org.apache.spark.sql.Row
import org.apache.spark.sql.types._
import org.apache.spark.sql.functions.udf



val get_cus_val = spark.udf.register("get_cus_val", (filePath: String) => filePath.split("\\.")(4))

val df1With_ = df.toDF(df.columns.map(_.replace(".", "_")): _*)
val column_to_keep = df1With_.columns.filter(v => (!v.contains("^") && !v.contains("!") && !v.contains("_c"))).toSeq
val df1result = df1With_.select(column_to_keep.head, column_to_keep.tail: _*)
val df1Final=df1result.withColumn("DataPartition", lit(null: String))

これは私の入力ファイル名の例です。

Fundamental.FinancialLineItem.FinancialLineItem.SelfSourcedPrivate.CUS.1.2017-09-07-1056.Full

Fundamental.FinancialLineItem.FinancialLineItem.Japan.CUS.1.2017-09-07-1056.Full.txt

今、私はこのファイルを読み込んで、「。」で分割したいと考えています。演算子を選択してから、 CUS を DataPartition の代わりに新しい列として追加します。

UDF なしで実行できますか。

これが既存のデータフレームのスキーマです

root
 |-- LineItem_organizationId: long (nullable = true)
 |-- LineItem_lineItemId: integer (nullable = true)
 |-- StatementTypeCode: string (nullable = true)
 |-- LineItemName: string (nullable = true)
 |-- LocalLanguageLabel: string (nullable = true)
 |-- FinancialConceptLocal: string (nullable = true)
 |-- FinancialConceptGlobal: string (nullable = true)
 |-- IsDimensional: boolean (nullable = true)
 |-- InstrumentId: string (nullable = true)
 |-- LineItemSequence: string (nullable = true)
 |-- PhysicalMeasureId: string (nullable = true)
 |-- FinancialConceptCodeGlobalSecondary: string (nullable = true)
 |-- IsRangeAllowed: boolean (nullable = true)
 |-- IsSegmentedByOrigin: boolean (nullable = true)
 |-- SegmentGroupDescription: string (nullable = true)
 |-- SegmentChildDescription: string (nullable = true)
 |-- SegmentChildLocalLanguageLabel: string (nullable = true)
 |-- LocalLanguageLabel_languageId: integer (nullable = true)
 |-- LineItemName_languageId: integer (nullable = true)
 |-- SegmentChildDescription_languageId: integer (nullable = true)
 |-- SegmentChildLocalLanguageLabel_languageId: integer (nullable = true)
 |-- SegmentGroupDescription_languageId: integer (nullable = true)
 |-- SegmentMultipleFundbDescription: string (nullable = true)
 |-- SegmentMultipleFundbDescription_languageId: integer (nullable = true)
 |-- IsCredit: boolean (nullable = true)
 |-- FinancialConceptLocalId: integer (nullable = true)
 |-- FinancialConceptGlobalId: integer (nullable = true)
 |-- FinancialConceptCodeGlobalSecondaryId: string (nullable = true)
 |-- FFAction: string (nullable = true)

提案された回答の後にコードを更新する

    val sqlContext = new org.apache.spark.sql.SQLContext(sc)
    import sqlContext.implicits._

    import org.apache.spark.{ SparkConf, SparkContext }
    import java.sql.{Date, Timestamp}
    import org.apache.spark.sql.Row
    import org.apache.spark.sql.types._
    import org.apache.spark.sql.functions.udf
    import org.apache.spark.sql.functions.{input_file_name, regexp_extract}

spark.udf.register("get_cus_val", (filePath: String) => filePath.split("\\.")(4))

import org.apache.spark.sql.functions.input_file_name

val df = sqlContext.read.format("csv").option("header", "true").option("delimiter", "|").option("inferSchema","true").load("s3://trfsdisu/SPARK/FinancialLineItem/MAIN")

val df1With_ = df.toDF(df.columns.map(_.replace(".", "_")): _*)
val column_to_keep = df1With_.columns.filter(v => (!v.contains("^") && !v.contains("!") && !v.contains("_c"))).toSeq
val df1result = df1With_.select(column_to_keep.head, column_to_keep.tail: _*)

df1result.withColumn("cus_val", get_cus_val(input_file_name))

df1result.printSchema()
4

1 に答える 1