(八)map,filter,flatMap算子-Java&Python版Spark

mapfilterflatMap算子

视频教程:

1、优酷

2、YouTube

 

1、map

map是将源JavaRDD的一个一个元素的传入call方法,并经过算法后一个一个的返回从而生成一个新的JavaRDD

 

java:

 

 1 package com.bean.spark.trans;
 2 
 3 import java.util.Arrays;
 4 import java.util.List;
 5 import org.apache.spark.SparkConf;
 6 import org.apache.spark.api.java.JavaRDD;
 7 import org.apache.spark.api.java.JavaSparkContext;
 8 import org.apache.spark.api.java.function.Function;
 9 /**
10  * 
11  * @author RedBean
12  *map
13  */
14 public class TraMap {
15     public static void main(String[] args) {
16         SparkConf conf = new SparkConf();
17         conf.setMaster("local");
18         conf.setAppName("map");
19         System.setProperty("hadoop.home.dir", "D:/tools/spark-2.0.0-bin-hadoop2.6");
20         JavaSparkContext sc = new JavaSparkContext(conf);
21         List<Integer> number = Arrays.asList(0,1,2,3,4,5,6,7,8,9);
22         JavaRDD<Integer> numberRDD = sc.parallelize(number);
23         JavaRDD<Integer> results = numberRDD.map(new Function<Integer, Integer>() {
24             @Override
25             public Integer call(Integer s) throws Exception {
26                 // TODO Auto-generated method stub
27                 return s * 5;
28             }
29         });
30         System.out.println(results.collect());
31     }
32 }

 

python:

 1 # -*- coding:utf-8 -*-
 2 
 3 
 4 from __future__ import print_function
 5 from pyspark import SparkConf
 6 from pyspark import SparkContext
 7 import os
 8 
 9 if __name__ == '__main__':
10     os.environ['SPARK_HOME'] = 'D:/tools/spark-2.0.0-bin-hadoop2.6'
11     conf = SparkConf().setAppName('mapTest').setMaster('local')
12     sc = SparkContext(conf=conf)
13     data = sc.parallelize([1,2,3,4,5,6])
14     def myMap(l):
15         return l * 5
16     print(data.map(myMap).collect())

 

2、filter

返回一个新的数据集,由经过func函数后返回值为true的原元素组成

 

java:

 1 package com.bean.spark.trans;
 2 
 3 import java.util.Arrays;
 4 import java.util.List;
 5 
 6 import org.apache.spark.SparkConf;
 7 import org.apache.spark.api.java.JavaRDD;
 8 import org.apache.spark.api.java.JavaSparkContext;
 9 import org.apache.spark.api.java.function.Function;
10 
11 public class TraFilter {
12     public static void main(String[] args) {
13         SparkConf conf = new SparkConf();
14         conf.setMaster("local");
15         conf.setAppName("filter");
16         System.setProperty("hadoop.home.dir", "D:/tools/spark-2.0.0-bin-hadoop2.6");
17         JavaSparkContext sc = new JavaSparkContext(conf);
18         List<Integer> number = Arrays.asList(0,1,2,3,4,5,6,7,8,9);
19         JavaRDD<Integer> numberRDD = sc.parallelize(number);
20         JavaRDD<Integer> results = numberRDD.filter(new Function<Integer, Boolean>() {
21             
22             @Override
23             public Boolean call(Integer s) throws Exception {
24                 // TODO Auto-generated method stub
25                 return s % 2 == 0;
26             }
27         });
28         System.out.println(results.collect());
29     }
30 }

python:

 1 # -*- coding:utf-8 -*-
 2 
 3 
 4 from __future__ import print_function
 5 from pyspark import SparkConf
 6 from pyspark import SparkContext
 7 import os
 8 
 9 if __name__ == '__main__':
10     os.environ['SPARK_HOME'] = 'D:/tools/spark-2.0.0-bin-hadoop2.6'
11     conf = SparkConf().setAppName('filterTest').setMaster('local')
12     sc = SparkContext(conf=conf)
13     data = sc.parallelize([1,2,3,4,5,6])
14     def filterFun(l):
15         return l > 2
16     print(data.filter(filterFun).collect())

 

3、flatMap

将一条 rdd数据使用你定义的函数给分解成多条 rdd数据。

java:

 1 package com.bean.spark.trans;
 2 
 3 import java.util.Arrays;
 4 import java.util.Iterator;
 5 
 6 import org.apache.spark.SparkConf;
 7 import org.apache.spark.api.java.JavaRDD;
 8 import org.apache.spark.api.java.JavaSparkContext;
 9 import org.apache.spark.api.java.function.FlatMapFunction;
10 
11 public class TraFlatMap {
12     public static void main(String[] args) {
13         SparkConf conf = new SparkConf();
14         conf.setMaster("local");
15         conf.setAppName("FlatMap");
16         System.setProperty("hadoop.home.dir", "D:/tools/spark-2.0.0-bin-hadoop2.6");
17         JavaSparkContext sc = new JavaSparkContext(conf);
18         JavaRDD<String> context = sc.textFile("D:/tools/data/flatMap/flatMap.txt");
19         JavaRDD<String> results = context.flatMap(new FlatMapFunction<String, String>() {
20             @Override
21             public Iterator<String> call(String s) throws Exception {
22                 // TODO Auto-generated method stub
23                 return Arrays.asList(s).iterator();
24             }
25         });
26         System.out.println(results.collect());
27         
28     }
29 }

python:

 

 1 # -*- coding:utf-8 -*-
 2 
 3 
 4 from __future__ import print_function
 5 from pyspark import SparkConf
 6 from pyspark import SparkContext
 7 import os
 8 
 9 if __name__ == '__main__':
10     os.environ['SPARK_HOME'] = 'D:/tools/spark-2.0.0-bin-hadoop2.6'
11     conf = SparkConf().setAppName('filterTest').setMaster('local')
12     sc = SparkContext(conf=conf)
13     data = sc.parallelize(["Hello World","Spark Hadoop Storm","java python c"])
14     def flatFun(l):
15         return l.split(" ")
16     print(data.flatMap(flatFun).collect())

 

 

 

 

    原文作者:spark
    原文地址: https://www.cnblogs.com/LgyBean/p/6251413.html
    本文转自网络文章,转载此文章仅为分享知识,如有侵权,请联系博主进行删除。
点赞