Examples of JavaPairRDD


Examples of org.apache.spark.api.java.JavaPairRDD

        JobConf hdpConf = HdpBootstrap.hadoopConfig();
        hdpConf.set(ConfigurationOptions.ES_RESOURCE, target);


        //JavaPairRDD data = sc.newAPIHadoopRDD(hdpConf, EsInputFormat.class, NullWritable.class, MapWritable.class);
        JavaPairRDD data = sc.hadoopRDD(hdpConf, EsInputFormat.class, NullWritable.class, MapWritable.class);

        long messages = data.filter(new Function<Tuple2<Text, MapWritable>, Boolean>() {
            public Boolean call(Tuple2<Text, MapWritable> t) { return t._2.containsKey(new Text("message")); }
        }).count();

        JavaRDD map = data.map(new Function<Tuple2<Text, MapWritable>, Map<String, Object>>() {
            public Map<String, Object> call(Tuple2<Text, MapWritable> v1) throws Exception {
                return (Map<String, Object>) WritableUtils.fromWritable(v1._2);
            }
        });

        JavaRDD fooBar = data.map(new Function<Tuple2<Text, MapWritable>, String>() {
            public String call(Tuple2<Text, MapWritable> v1) throws Exception {
                return v1._1.toString();
            }
        });

        assertThat((int) data.count(), is(2));
        System.out.println(data.take(10));
        System.out.println(messages);
        System.out.println(fooBar.take(2));
        System.out.println(map.take(10));
    }
View Full Code Here

Examples of org.apache.spark.api.java.JavaPairRDD

  private JavaRDDLike<?, ?> getJavaRDDLikeInternal(SparkRuntime runtime) {
    List<PCollectionImpl<?>> parents = getParents();
    JavaRDD[] rdds = new JavaRDD[parents.size()];
    for (int i = 0; i < rdds.length; i++) {
      if (parents.get(i) instanceof PTableBase) {
        JavaPairRDD prdd = (JavaPairRDD) ((SparkCollection) parents.get(i)).getJavaRDDLike(runtime);
        rdds[i] = prdd.mapPartitions(new FlatMapPairDoFn(IdentityFn.getInstance(), runtime.getRuntimeContext()));
      } else {
        rdds[i] = (JavaRDD) ((SparkCollection) parents.get(i)).getJavaRDDLike(runtime);
      }
    }
    return runtime.getSparkContext().union(rdds);
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.