01 第一个 Spark 程序：WordCount

第一个 Spark 程序：WordCount

1. 使用 Spark-shell

1. 准备数据：创建文件夹 input，以及 Words.txt 文件

   [zgl@hadoop101 spark-2.1.1]$ mkdir input
   [zgl@hadoop101 input]$ vim Words.txt
   

   在文件中输入数据：

   hello spark
   hello scala
   hello world
   

2. 进入 spark-shell

   [zgl@hadoop101 spark-2.1.1]$ bin/spark-shell 
   

3. 编写 WordCount 程序并运行

   scala> sc.textFile("input/").flatMap(_.split(" ")).map((_,1)).reduceByKey(_+_).collect
   res0: Array[(String, Int)] = Array((scala,1), (hello,3), (world,1), (spark,1))  
   

   

2. 使用开发工具 IDEA

1. 创建 Maven 项目，并导入如下依赖

   <dependencies>
     <dependency>
       <groupId>org.apache.spark</groupId>
       <artifactId>spark-core_2.11</artifactId>
       <version>2.1.1</version>
     </dependency>
   </dependencies>
   
   <build>
     <plugins>
       <!-- 打包插件, 否则 scala 类不会编译并打包进去 -->
       <plugin>
         <groupId>net.alchim31.maven</groupId>
         <artifactId>scala-maven-plugin</artifactId>
         <version>3.4.6</version>
         <executions>
           <execution>
             <goals>
               <goal>compile</goal>
               <goal>testCompile</goal>
             </goals>
           </execution>
         </executions>
       </plugin>
     </plugins>
   </build>
   

2. 创建 WordCount.scala 文件，实现以下代码

   package com.guli
   
   import org.apache.spark.{SparkConf, SparkContext}
   
   object WordCount {
     def main(args: Array[String]): Unit = {
       val conf: SparkConf = new SparkConf().setAppName("WorldCount").setMaster("local[*]")
       val sc = new SparkContext(conf)
       val wcArray: Array[(String, Int)] = sc.textFile("/Users/zgl/Desktop/input").flatMap(_.split(" ")).map((_,1)).reduceByKey(_+_).collect()
       wcArray.foreach(println)
       sc.stop()
     }
   }
   

3. 运行结果

   (scala,1)
   (hello,3)
   (world,1)
   (spark,1)