Spark之分区

import org.apache.spark.Partitioner;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.sql.SparkSession;
import scala.Tuple2;

import java.util.Arrays;
import java.util.List;
import java.util.regex.Pattern;

/**
 * Created by hadoop on 17-9-20.
 */
public class JavaSparkTest {



    //数据拆分的方式
    private static final Pattern SPACE = Pattern.compile(" ");

    public static void main(String[] args) throws Exception {
        //判断是否拿到对象
//        if (args.length < 1) {
//            System.err.println("Usage: JavaWordCount ");
//            System.exit(1);
//        }
        //sparkSession对象，为用户提供了一个统一的切入点来使用spark的功能
        SparkSession spark = SparkSession
                .builder()
                .master("local")
                .appName("JavaWordCount")
                .getOrCreate();
        System.out.println("star=========");

        JavaRDD lines = spark.read().textFile("file:///usr/local/spark/wordcount.txt").javaRDD();
        System.out.println("lines分区"+lines.getNumPartitions());
        System.out.println("lines分区器"+lines.partitioner());

        JavaRDD words = lines.flatMap(s -> Arrays.asList(SPACE.split(s)).iterator());

        words.collect().forEach(System.out::println);

        JavaPairRDD, Integer> ones = words.mapToPair(s -> new Tuple2<>(s, 1));
        //自定义分区器
        JavaPairRDD, Integer> tows=ones.partitionBy(new mupartition());
        System.out.println("分区器:"+tows.getNumPartitions()+"--->"+tows.partitioner());

        ones.collect().forEach(System.out::println);

        JavaPairRDD, Integer> counts = ones.reduceByKey((i1, i2) -> i1 + i2);
        System.out.println("counts分区"+counts.getNumPartitions());
        System.out.println("counts分区器"+counts.partitioner());

        List, Integer>> output = counts.collect();
        for (Tuple2,?> tuple : output) {
            System.out.println(tuple._1() + ": " + tuple._2());
        }
        spark.stop();
    }


    static class mupartition extends Partitioner {

        @Override
        public int numPartitions() {
            return 3;
        }

        @Override
        public int getPartition(Object key) {
            int index=key.hashCode()%numPartitions();
            if(index<0){
                index=0;
            }
            return index;
        }
    }


}