pyspark的安装配置

1、搭建基本spark+Hadoop的本地环境

　　https://blog.csdn.net/u011513853/article/details/52865076?tdsourcetag=s_pcqq_aiomsg

2、下载对应的spark与pyspark的版本进行安装

　　https://pypi.org/project/pyspark/2.3.0/#history

3、单词统计测试

　　a、python版本

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

import os import shutil   from pyspark import SparkContext   inputpath = './data/wc.txt' outputpath = './data/out.txt'   sc = SparkContext('local', 'wordcount')   # 读取文件 input = sc.textFile(inputpath) # 切分单词 words = input.flatMap(lambda line: line.split(' ')) # 转换成键值对并计数 counts = words.map(lambda word: (word, 1)).reduceByKey(lambda x, y: x + y)   # 输出结果 counts.foreach(print)   # 删除输出目录 if os.path.exists(outputpath):     shutil.rmtree(outputpath, True)   # 将统计结果写入结果文件 counts.saveAsTextFile(outputpath)

　　

　　b、scala版本

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

package com.wcount   import java.io.{File, PrintWriter}   import org.apache.spark.rdd.RDD import org.apache.spark.{SparkConf, SparkContext}   object ScalaWordCount {     def main(args: Array[String]): Unit = {     /**       * SparkConf:表示spark application的参数，       *   setMaster：表示运行的模式：       *       *       local：本地模式，一般用于测试       *       standalone：spark集群自带的资源调度模式       *       yarn：hadoop       *       mesos:资源调度框架       *   setAppName:设置application的名称       */     val conf = new SparkConf().setMaster("local").setAppName("workJob")     /**       * SparkContext:spark application的上下文环境，通往集群的唯一入口       */     val sc = new SparkContext(conf)   //    val session: SparkSession = SparkSession.builder.appName("wc").master("local").getOrCreate()         val lines: RDD[String] = sc.textFile("./data/wc.txt")     val words: RDD[String] = lines.flatMap(line => {       println("flatmap...........")       line.split(" ")     })     val tuple: RDD[(String, Int)] = words.map(word => {       println("map............")       new Tuple2(word, 1)     })     val result: RDD[(String, Int)] = tuple.reduceByKey((v1: Int, v2: Int) => v1 + v2)     //result.foreach(println)       //文件写入     val outWriter = new PrintWriter(new File("./data/out.txt"))     var wt:String = ""       for (item<-result){       wt =item._1.toString+":"+item._2.toString+" "       println(wt)     }     println(wt)     outWriter.println(wt)     outWriter.close()       while (true){       }     //    sc.textFile("./data/wc").flatMap(line => {line.split(" ")}).map(word => {new Tuple2(word, 1)}).reduceByKey((v1: Int, v2: Int) => v1 + v2).foreach(println)     sc.stop()   } }