Examples with Function org.apache.spark.api.java.function.Function used on opensource projects

Example 11 with Function

use of org.apache.spark.api.java.function.Function in project azure-tools-for-java by Microsoft.

the class JavaSparkPi method main.

public static void main(String[] args) throws Exception {
    // use this line if you want to run your application in the cluster
    // SparkConf sparkConf = new SparkConf().setAppName("JavaSparkPi");
    SparkConf sparkConf = new SparkConf().setAppName("JavaSparkPi").setMaster("local[2]");
    JavaSparkContext jsc = new JavaSparkContext(sparkConf);
    int slices = (args.length == 1) ? Integer.parseInt(args[0]) : 2;
    int n = 100000 * slices;
    List<Integer> l = new ArrayList<Integer>(n);
    for (int i = 0; i < n; i++) {
        l.add(i);
    }
    JavaRDD<Integer> dataSet = jsc.parallelize(l, slices);
    int count = dataSet.map(new Function<Integer, Integer>() {

        @Override
        public Integer call(Integer integer) {
            double x = Math.random() * 2 - 1;
            double y = Math.random() * 2 - 1;
            return (x * x + y * y < 1) ? 1 : 0;
        }
    }).reduce(new Function2<Integer, Integer, Integer>() {

        @Override
        public Integer call(Integer integer, Integer integer2) {
            return integer + integer2;
        }
    });
    System.out.println("Pi is roughly " + 4.0 * count / n);
    jsc.stop();
}

Example 12 with Function

use of org.apache.spark.api.java.function.Function in project learning-spark by databricks.

the class KeyValueMapFilter method main.

public static void main(String[] args) throws Exception {
    if (args.length != 2) {
        throw new Exception("Usage KeyValueMapFilter sparkMaster inputFile");
    }
    String master = args[0];
    String inputFile = args[1];
    JavaSparkContext sc = new JavaSparkContext(master, "KeyValueMapFilter", System.getenv("SPARK_HOME"), System.getenv("JARS"));
    JavaRDD<String> input = sc.textFile(inputFile);
    PairFunction<String, String, String> keyData = new PairFunction<String, String, String>() {

        @Override
        public Tuple2<String, String> call(String x) {
            return new Tuple2(x.split(" ")[0], x);
        }
    };
    Function<Tuple2<String, String>, Boolean> longWordFilter = new Function<Tuple2<String, String>, Boolean>() {

        @Override
        public Boolean call(Tuple2<String, String> input) {
            return (input._2().length() < 20);
        }
    };
    JavaPairRDD<String, String> rdd = input.mapToPair(keyData);
    JavaPairRDD<String, String> result = rdd.filter(longWordFilter);
    Map<String, String> resultMap = result.collectAsMap();
    for (Entry<String, String> entry : resultMap.entrySet()) {
        System.out.println(entry.getKey() + ":" + entry.getValue());
    }
}

Example 13 with Function

use of org.apache.spark.api.java.function.Function in project learning-spark by databricks.

the class PerKeyAvg method main.

public static void main(String[] args) throws Exception {
    String master;
    if (args.length > 0) {
        master = args[0];
    } else {
        master = "local";
    }
    JavaSparkContext sc = new JavaSparkContext(master, "PerKeyAvg", System.getenv("SPARK_HOME"), System.getenv("JARS"));
    List<Tuple2<String, Integer>> input = new ArrayList();
    input.add(new Tuple2("coffee", 1));
    input.add(new Tuple2("coffee", 2));
    input.add(new Tuple2("pandas", 3));
    JavaPairRDD<String, Integer> rdd = sc.parallelizePairs(input);
    Function<Integer, AvgCount> createAcc = new Function<Integer, AvgCount>() {

        @Override
        public AvgCount call(Integer x) {
            return new AvgCount(x, 1);
        }
    };
    Function2<AvgCount, Integer, AvgCount> addAndCount = new Function2<AvgCount, Integer, AvgCount>() {

        @Override
        public AvgCount call(AvgCount a, Integer x) {
            a.total_ += x;
            a.num_ += 1;
            return a;
        }
    };
    Function2<AvgCount, AvgCount, AvgCount> combine = new Function2<AvgCount, AvgCount, AvgCount>() {

        @Override
        public AvgCount call(AvgCount a, AvgCount b) {
            a.total_ += b.total_;
            a.num_ += b.num_;
            return a;
        }
    };
    AvgCount initial = new AvgCount(0, 0);
    JavaPairRDD<String, AvgCount> avgCounts = rdd.combineByKey(createAcc, addAndCount, combine);
    Map<String, AvgCount> countMap = avgCounts.collectAsMap();
    for (Entry<String, AvgCount> entry : countMap.entrySet()) {
        System.out.println(entry.getKey() + ":" + entry.getValue().avg());
    }
}

Example 14 with Function

use of org.apache.spark.api.java.function.Function in project learning-spark by databricks.

the class LogAnalyzerWindowed method processAccessLogs.

public void processAccessLogs(String outDir, JavaDStream<ApacheAccessLog> accessLogsDStream) {
    JavaDStream<ApacheAccessLog> windowDStream = accessLogsDStream.window(Flags.getInstance().getWindowLength(), Flags.getInstance().getSlideInterval());
    JavaDStream<String> ip = accessLogsDStream.map(new Function<ApacheAccessLog, String>() {

        public String call(ApacheAccessLog entry) {
            return entry.getIpAddress();
        }
    });
    // reduceByWindow
    JavaDStream<Long> requestCountRBW = accessLogsDStream.map(new Function<ApacheAccessLog, Long>() {

        public Long call(ApacheAccessLog entry) {
            return 1L;
        }
    }).reduceByWindow(new Function2<Long, Long, Long>() {

        public Long call(Long v1, Long v2) {
            return v1 + v2;
        }
    }, new Function2<Long, Long, Long>() {

        public Long call(Long v1, Long v2) {
            return v1 - v2;
        }
    }, Flags.getInstance().getWindowLength(), Flags.getInstance().getSlideInterval());
    requestCountRBW.print();
    // reducebykeyandwindow
    JavaPairDStream<String, Long> ipAddressPairDStream = accessLogsDStream.mapToPair(new PairFunction<ApacheAccessLog, String, Long>() {

        public Tuple2<String, Long> call(ApacheAccessLog entry) {
            return new Tuple2(entry.getIpAddress(), 1L);
        }
    });
    JavaPairDStream<String, Long> ipCountDStream = ipAddressPairDStream.reduceByKeyAndWindow(// Adding elements in the new slice
    new Function2<Long, Long, Long>() {

        public Long call(Long v1, Long v2) {
            return v1 + v2;
        }
    }, // Removing elements from the oldest slice
    new Function2<Long, Long, Long>() {

        public Long call(Long v1, Long v2) {
            return v1 - v2;
        }
    }, Flags.getInstance().getWindowLength(), Flags.getInstance().getSlideInterval());
    ipCountDStream.print();
    // Use countByWindow
    JavaDStream<Long> requestCount = accessLogsDStream.countByWindow(Flags.getInstance().getWindowLength(), Flags.getInstance().getSlideInterval());
    JavaPairDStream<String, Long> ipAddressRequestCount = ip.countByValueAndWindow(Flags.getInstance().getWindowLength(), Flags.getInstance().getSlideInterval());
    requestCount.print();
    ipAddressRequestCount.print();
    // use a transform for the response code count
    JavaPairDStream<Integer, Long> responseCodeCountTransform = accessLogsDStream.transformToPair(new Function<JavaRDD<ApacheAccessLog>, JavaPairRDD<Integer, Long>>() {

        public JavaPairRDD<Integer, Long> call(JavaRDD<ApacheAccessLog> logs) {
            return Functions.responseCodeCount(logs);
        }
    });
    windowDStream.foreachRDD(new Function<JavaRDD<ApacheAccessLog>, Void>() {

        public Void call(JavaRDD<ApacheAccessLog> accessLogs) {
            Tuple4<Long, Long, Long, Long> contentSizeStats = Functions.contentSizeStats(accessLogs);
            List<Tuple2<Integer, Long>> responseCodeToCount = Functions.responseCodeCount(accessLogs).take(100);
            JavaPairRDD<String, Long> ipAddressCounts = Functions.ipAddressCount(accessLogs);
            List<String> ip = Functions.filterIPAddress(ipAddressCounts).take(100);
            Object ordering = Ordering.natural();
            Comparator<Long> cmp = (Comparator<Long>) ordering;
            List<Tuple2<String, Long>> topEndpoints = Functions.endpointCount(accessLogs).top(10, new Functions.ValueComparator<String, Long>(cmp));
            logStatistics = new LogStatistics(contentSizeStats, responseCodeToCount, ip, topEndpoints);
            return null;
        }
    });
}

Example 15 with Function

use of org.apache.spark.api.java.function.Function in project learning-spark by databricks.

the class LogAnalyzerTotal method processAccessLogs.

public void processAccessLogs(String outDir, JavaDStream<ApacheAccessLog> accessLogsDStream) {
    // Calculate statistics based on the content size, and update the static variables to track this.
    accessLogsDStream.foreachRDD(new Function<JavaRDD<ApacheAccessLog>, Void>() {

        public Void call(JavaRDD<ApacheAccessLog> accessLogs) {
            Tuple4<Long, Long, Long, Long> stats = Functions.contentSizeStats(accessLogs);
            if (stats != null) {
                runningCount.getAndAdd(stats._1());
                runningSum.getAndAdd(stats._2());
                runningMin.set(Math.min(runningMin.get(), stats._3()));
                runningMax.set(Math.max(runningMax.get(), stats._4()));
            }
            return null;
        }
    });
    // A DStream of Resonse Code Counts;
    JavaPairDStream<Integer, Long> responseCodeCountDStream = accessLogsDStream.transformToPair(new Function<JavaRDD<ApacheAccessLog>, JavaPairRDD<Integer, Long>>() {

        public JavaPairRDD<Integer, Long> call(JavaRDD<ApacheAccessLog> rdd) {
            return Functions.responseCodeCount(rdd);
        }
    }).updateStateByKey(new Functions.ComputeRunningSum());
    responseCodeCountDStream.foreachRDD(new Function<JavaPairRDD<Integer, Long>, Void>() {

        public Void call(JavaPairRDD<Integer, Long> rdd) {
            currentResponseCodeCounts = rdd.take(100);
            return null;
        }
    });
    // A DStream of ipAddressCounts.
    JavaPairDStream<String, Long> ipRawDStream = accessLogsDStream.transformToPair(new Function<JavaRDD<ApacheAccessLog>, JavaPairRDD<String, Long>>() {

        public JavaPairRDD<String, Long> call(JavaRDD<ApacheAccessLog> rdd) {
            return Functions.ipAddressCount(rdd);
        }
    });
    JavaPairDStream<String, Long> ipCumDStream = ipRawDStream.updateStateByKey(new Functions.ComputeRunningSum());
    // A DStream of ipAddressCounts without transform
    JavaPairDStream<String, Long> ipDStream = accessLogsDStream.mapToPair(new Functions.IpTuple());
    JavaPairDStream<String, Long> ipCountsDStream = ipDStream.reduceByKey(new Functions.LongSumReducer());
    // and joining it with the transfer amount
    JavaPairDStream<String, Long> ipBytesDStream = accessLogsDStream.mapToPair(new Functions.IpContentTuple());
    JavaPairDStream<String, Long> ipBytesSumDStream = ipBytesDStream.reduceByKey(new Functions.LongSumReducer());
    JavaPairDStream<String, Tuple2<Long, Long>> ipBytesRequestCountDStream = ipBytesSumDStream.join(ipCountsDStream);
    // Save our dstream of ip address request counts
    JavaPairDStream<Text, LongWritable> writableDStream = ipDStream.mapToPair(new PairFunction<Tuple2<String, Long>, Text, LongWritable>() {

        public Tuple2<Text, LongWritable> call(Tuple2<String, Long> e) {
            return new Tuple2(new Text(e._1()), new LongWritable(e._2()));
        }
    });
    class OutFormat extends SequenceFileOutputFormat<Text, LongWritable> {
    }
    ;
    writableDStream.saveAsHadoopFiles(outDir, "pandas", Text.class, LongWritable.class, OutFormat.class);
    // All ips more than 10
    JavaDStream<String> ipAddressDStream = ipCumDStream.transform(new Function<JavaPairRDD<String, Long>, JavaRDD<String>>() {

        public JavaRDD<String> call(JavaPairRDD<String, Long> rdd) {
            return Functions.filterIPAddress(rdd);
        }
    });
    ipAddressDStream.foreachRDD(new Function<JavaRDD<String>, Void>() {

        public Void call(JavaRDD<String> rdd) {
            List<String> currentIPAddresses = rdd.take(100);
            return null;
        }
    });
    // A DStream of endpoint to count.
    JavaPairDStream<String, Long> endpointCountsDStream = accessLogsDStream.transformToPair(new Function<JavaRDD<ApacheAccessLog>, JavaPairRDD<String, Long>>() {

        public JavaPairRDD<String, Long> call(JavaRDD<ApacheAccessLog> rdd) {
            return Functions.endpointCount(rdd);
        }
    }).updateStateByKey(new Functions.ComputeRunningSum());
    Object ordering = Ordering.natural();
    final Comparator<Long> cmp = (Comparator<Long>) ordering;
    endpointCountsDStream.foreachRDD(new Function<JavaPairRDD<String, Long>, Void>() {

        public Void call(JavaPairRDD<String, Long> rdd) {
            currentTopEndpoints = rdd.takeOrdered(10, new Functions.ValueComparator<String, Long>(cmp));
            return null;
        }
    });
}