• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    公众号

Scala TaskAttemptContext类代码示例

原作者: [db:作者] 来自: [db:来源] 收藏 邀请

本文整理汇总了Scala中org.apache.hadoop.mapreduce.TaskAttemptContext的典型用法代码示例。如果您正苦于以下问题:Scala TaskAttemptContext类的具体用法?Scala TaskAttemptContext怎么用?Scala TaskAttemptContext使用的例子?那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。



在下文中一共展示了TaskAttemptContext类的3个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于我们的系统推荐出更棒的Scala代码示例。

示例1: ParquetOutputWriter

//设置package包名称以及导入依赖的类
package com.newegg.eims.DataPorter.Parquet

import com.newegg.eims.DataPorter.Base.{DataSetSchema, IDataRow}
import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.fs.Path
import org.apache.hadoop.mapred.{JobConf, TaskAttemptContextImpl}
import org.apache.hadoop.mapreduce.TaskAttemptContext
import parquet.hadoop.ParquetOutputFormat
import parquet.hadoop.api.WriteSupport


class ParquetOutputWriter(dataSetSchema: DataSetSchema, path: Path, conf: JobConf) {

  class IDataRowParquetOutputFormat(support: ParquetWriteSupport, filePath: Path) extends ParquetOutputFormat[IDataRow]() {
    override def getWriteSupport(configuration: Configuration): WriteSupport[IDataRow] = {
      support
    }

    override def getDefaultWorkFile(context: TaskAttemptContext, extension: String): Path = {
      filePath
    }
  }

  private val context = new TaskAttemptContextImpl(conf, new org.apache.hadoop.mapred.TaskAttemptID())
  private val formatter = {
    val support = new ParquetOutputFormat[IDataRow]().getWriteSupport(conf).asInstanceOf[ParquetWriteSupport]
    support.setSchema(dataSetSchema)
    new IDataRowParquetOutputFormat(support, path)
  }

  private val recordWriter = formatter.getRecordWriter(context)

  def write(row: IDataRow): Unit = recordWriter.write(null, row)

  def close(): Unit = recordWriter.close(context)
} 
开发者ID:CodeBabyBear,项目名称:DataPorter,代码行数:37,代码来源:ParquetOutputWriter.scala


示例2: SuccinctHadoopMapReduceUtil

//设置package包名称以及导入依赖的类
package edu.berkeley.cs.succinct.annot.serde

import java.lang.{Boolean => JBoolean, Integer => JInteger}

import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.mapreduce.{TaskAttemptContext, TaskAttemptID}

object SuccinctHadoopMapReduceUtil {

  private def firstAvailableClass(first: String, second: String): Class[_] = {
    try {
      Class.forName(first)
    } catch {
      case e: ClassNotFoundException =>
        Class.forName(second)
    }
  }

  def newTaskAttemptID(jtIdentifier: String, jobId: Int, isMap: Boolean, taskId: Int,
                       attemptId: Int) = {
    val klass = Class.forName("org.apache.hadoop.mapreduce.TaskAttemptID")
    try {
      // First, attempt to use the old-style constructor that takes a boolean isMap
      // (not available in YARN)
      val ctor = klass.getDeclaredConstructor(classOf[String], classOf[Int], classOf[Boolean],
        classOf[Int], classOf[Int])
      ctor.newInstance(jtIdentifier, new JInteger(jobId), new JBoolean(isMap), new JInteger(taskId),
        new JInteger(attemptId)).asInstanceOf[TaskAttemptID]
    } catch {
      case exc: NoSuchMethodException =>
        // If that failed, look for the new constructor that takes a TaskType (not available in 1.x)
        val taskTypeClass = Class.forName("org.apache.hadoop.mapreduce.TaskType")
          .asInstanceOf[Class[Enum[_]]]
        val taskType = taskTypeClass.getMethod("valueOf", classOf[String]).invoke(
          taskTypeClass, if (isMap) "MAP" else "REDUCE")
        val ctor = klass.getDeclaredConstructor(classOf[String], classOf[Int], taskTypeClass,
          classOf[Int], classOf[Int])
        ctor.newInstance(jtIdentifier, new JInteger(jobId), taskType, new JInteger(taskId),
          new JInteger(attemptId)).asInstanceOf[TaskAttemptID]
    }
  }

  def newTaskAttemptContext(conf: Configuration, attemptId: TaskAttemptID): TaskAttemptContext = {
    val klass = firstAvailableClass(
      "org.apache.hadoop.mapreduce.task.TaskAttemptContextImpl", // hadoop2, hadoop2-yarn
      "org.apache.hadoop.mapreduce.TaskAttemptContext") // hadoop1
    val ctor = klass.getDeclaredConstructor(classOf[Configuration], classOf[TaskAttemptID])
    ctor.newInstance(conf, attemptId).asInstanceOf[TaskAttemptContext]
  }
} 
开发者ID:anuragkh,项目名称:annotation-search,代码行数:51,代码来源:SuccinctHadoopMapReduceUtil.scala


示例3: SuccinctAnnotationOutputFormat

//设置package包名称以及导入依赖的类
package edu.berkeley.cs.succinct.annot.serde

import java.io.File

import org.apache.hadoop.io.NullWritable
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat
import org.apache.hadoop.mapreduce.{RecordWriter, TaskAttemptContext}

class SuccinctAnnotationOutputFormat
  extends FileOutputFormat[NullWritable, (Int, Iterator[(String, String, String)])] {
  override def getRecordWriter(job: TaskAttemptContext):
  RecordWriter[NullWritable, (Int, Iterator[(String, String, String)])] = {
    val conf = job.getConfiguration
    val ignoreParseErrors = conf.get("succinct.annotations.ignoreParseErrors", "true").toBoolean
    val serializeInMemory = conf.get("succinct.annotations.serializeInMemory", "true").toBoolean
    val dirs = conf.get("spark.local.dir", System.getProperty("java.io.tmpdir")).split(",")
    println("ignoreParseErrors = " + ignoreParseErrors + " serializeInMemory = " + serializeInMemory
      + "Spark local dir = " + dirs(0) + " persistInMemory = false")
    val path = FileOutputFormat.getOutputPath(job)
    new SuccinctAnnotationRecordWriter(path, ignoreParseErrors, conf, (serializeInMemory, new File(dirs(0))))
  }
} 
开发者ID:anuragkh,项目名称:annotation-search,代码行数:23,代码来源:SuccinctAnnotationOutputFormat.scala



注:本文中的org.apache.hadoop.mapreduce.TaskAttemptContext类示例整理自Github/MSDocs等源码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。


鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
上一篇:
Scala SecretKeySpec类代码示例发布时间:2022-05-23
下一篇:
Scala UTF_8类代码示例发布时间:2022-05-23
热门推荐
热门话题
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap