Skip to content

[SPARK-10310][SQL]Using \t as the field delimeter and \n as the line delimeter #8476

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
zhichao-li wants to merge 9 commits into from
Closed

[SPARK-10310][SQL]Using \t as the field delimeter and \n as the line delimeter #8476

Changes from 1 commit
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
cafe301
tab as the field delimeter and 10 as the line delimeter
zhichao-li Aug 27, 2015
4939ed8
add some idea from hive
zhichao-li Aug 28, 2015
517f31f
style
zhichao-li Aug 31, 2015
23aaa04
fix cast
zhichao-li Aug 31, 2015
c6e9134
add unittest for script input and output format
zhichao-li Sep 7, 2015
9b3a8de
fix path of testing script
zhichao-li Sep 8, 2015
8fb75a6
style
zhichao-li Sep 8, 2015
2d1e604
fix script
zhichao-li Sep 8, 2015
ae0b68e
python3
zhichao-li Sep 8, 2015
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 16 additions & 15 deletions sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/ScriptTransformation.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,20 +21,21 @@ import java.io._
import java.util.Properties
import javax.annotation.Nullable

import org.apache.hadoop.util.LineReader

import scala.collection.JavaConverters._
import scala.util.control.NonFatal

import org.apache.hadoop.hive.serde.serdeConstants
import org.apache.hadoop.hive.serde2.AbstractSerDe
import org.apache.hadoop.hive.serde2.objectinspector._
import org.apache.hadoop.io.Writable
import org.apache.hadoop.io.{Text, Writable}

import org.apache.spark.rdd.RDD
import org.apache.spark.sql.catalyst.{CatalystTypeConverters, InternalRow}
import org.apache.spark.sql.catalyst.expressions._
import org.apache.spark.sql.catalyst.plans.logical.ScriptInputOutputSchema
import org.apache.spark.sql.execution._
import org.apache.spark.sql.hive.HiveShim._
import org.apache.spark.sql.hive.{HiveContext, HiveInspectors}
import org.apache.spark.sql.types.DataType
import org.apache.spark.util.{CircularBuffer, RedirectThread, Utils}
Expand Down Expand Up @@ -108,7 +109,7 @@ case class ScriptTransformation(
val reader = new BufferedReader(new InputStreamReader(inputStream))
val outputIterator: Iterator[InternalRow] = new Iterator[InternalRow] with HiveInspectors {
var curLine: String = null
val scriptOutputStream = new DataInputStream(inputStream)
val lineReader = new LineReader(inputStream)
var scriptOutputWritable: Writable = null
val reusedWritableObject: Writable = if (null != outputSerde) {
outputSerde.getSerializedClass().newInstance
Expand All @@ -134,15 +135,13 @@ case class ScriptTransformation(
}
} else if (scriptOutputWritable == null) {
scriptOutputWritable = reusedWritableObject
try {
scriptOutputWritable.readFields(scriptOutputStream)
if (0 == lineReader.readLine(scriptOutputWritable.asInstanceOf[Text])) {
if (writerThread.exception.isDefined) {
throw writerThread.exception.get
}
false
} else {
true
} catch {
case _: EOFException =>
if (writerThread.exception.isDefined) {
throw writerThread.exception.get
}
false
}
} else {
true
Expand Down Expand Up @@ -222,7 +221,7 @@ private class ScriptTransformationWriterThread(

override def run(): Unit = Utils.logUncaughtExceptions {
TaskContext.setTaskContext(taskContext)

val newLineCode = 10

@chenghao-intel chenghao-intel Aug 27, 2015

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Seems never used.

val dataOutputStream = new DataOutputStream(outputStream)

// We can't use Utils.tryWithSafeFinally here because we also need a `catch` block, so
Expand All @@ -248,9 +247,10 @@ private class ScriptTransformationWriterThread(
}
outputStream.write(data.getBytes("utf-8"))
} else {
val writable = inputSerde.serialize(
row.asInstanceOf[GenericInternalRow].values, inputSoi)
prepareWritable(writable, ioschema.outputSerdeProps).write(dataOutputStream)
val text = inputSerde.serialize(
row.asInstanceOf[GenericInternalRow].values, inputSoi).asInstanceOf[Text]
outputStream.write(text.getBytes(), 0, text.getLength())
outputStream.write(newLineCode)
}
}
outputStream.close()
Expand Down Expand Up @@ -340,6 +340,7 @@ case class HiveScriptIOSchema (

var propsMap = serdeProps.toMap + (serdeConstants.LIST_COLUMNS -> columns.mkString(","))
propsMap = propsMap + (serdeConstants.LIST_COLUMN_TYPES -> columnTypesNames)
propsMap = propsMap + (serdeConstants.FIELD_DELIM -> "\t")

@liancheng liancheng Sep 4, 2015

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shouldn't we specify line delimiter here?

@zhichao-li zhichao-li Sep 7, 2015

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

line delimiter is control by RecorderWriter, see TextRecordWriter as an example:

  public void write(Writable row) throws IOException {
    Text text = (Text) row;
    Text escapeText = text;

    if (HiveConf.getBoolVar(conf, HiveConf.ConfVars.HIVESCRIPTESCAPE)) {
      escapeText = HiveUtils.escapeText(text);
    }

    out.write(escapeText.getBytes(), 0, escapeText.getLength());
    out.write(Utilities.newLineCode);
  }


val properties = new Properties()
properties.putAll(propsMap.asJava)
Expand Down