当前位置: 首页>>代码示例>>Java>>正文

Java KeyedStream类代码示例

本文整理汇总了Java中org.apache.flink.streaming.api.datastream.KeyedStream类的典型用法代码示例。如果您正苦于以下问题：Java KeyedStream类的具体用法？Java KeyedStream怎么用？Java KeyedStream使用的例子？那么, 这里精选的类代码示例或许可以为您提供帮助。


KeyedStream类属于org.apache.flink.streaming.api.datastream包，在下文中一共展示了KeyedStream类的15个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Java代码示例。

示例1: main

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public static void main(String[] args) throws Exception {
  // set up the execution environment
  final StreamExecutionEnvironment env = new StreamExecutionEnvBuilder().build();

  // Get the json config for parsing the raw input stream
  String parsingConfig = AppUtils.getParsingJsonConfig();

  KeyedStream<Tuple3<String, Long, String>, Tuple> kaydRawMessagesStream =
      setupKayedRawMessagesStream(env, parsingConfig);

  String outputStreamTopicName = configs.getStringProp("inputStreamTopicName");
  double streamDelayScale = configs.getDoubleProp("streamDelayScale");
  Properties producerProps = AppUtils.getKafkaProducerProperties();

  // replay the stream
  kaydRawMessagesStream.map(new StreamPlayer(streamDelayScale, outputStreamTopicName,
      producerProps)).setParallelism(1);

  // execute program
  env.execute("datAcron In-Situ Processing AIS Message Stream Simulator"
      + AppUtils.getAppVersion());
}
开发者ID:ehabqadah，项目名称:in-situ-processing-datAcron，代码行数:23，代码来源:RawStreamSimulator.java

示例2: setupKayedRawMessagesStream

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
/***
 * Setup the kayed stream of a raw stream.
 * 
 * @param env
 * @param streamSource
 * @param parsingConfig
 * @return
 */
private static KeyedStream<Tuple3<String, Long, String>, Tuple> setupKayedRawMessagesStream(
    final StreamExecutionEnvironment env, String parsingConfig) {
  DataStream<Tuple3<String, Long, String>> rawStream =
      env.addSource(
          new FileLinesStreamSource(configs.getStringProp("aisDataSetFilePath"), parsingConfig,true))
          .flatMap(new RawStreamMapper(parsingConfig)).setParallelism(1);

  // assign the timestamp of the AIS messages based on their timestamps
  DataStream<Tuple3<String, Long, String>> rawStreamWithTimeStamp =
      rawStream.assignTimestampsAndWatermarks(new RawMessageTimestampAssigner());

  // Construct the keyed stream (i.e., trajectories stream) of the raw messages by grouping them
  // based on the message ID (MMSI for vessels)
  KeyedStream<Tuple3<String, Long, String>, Tuple> kaydAisMessagesStream =
      rawStreamWithTimeStamp.keyBy(0).process(new RawMessagesSorter()).keyBy(0);
  return kaydAisMessagesStream;
}
开发者ID:ehabqadah，项目名称:in-situ-processing-datAcron，代码行数:26，代码来源:RawStreamSimulator.java

示例3: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector , KeySelector<V,Double> valueSelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);

         TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtCountWindFold<>(keySelector,valueFold, window, resultType),new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:26，代码来源:ExtNormalFreqAnomaly.java

示例4: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector , KeySelector<V,Double> valueSelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);

         TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtCountSumFold<>(keySelector, valueSelector, valueFold, resultType), new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:26，代码来源:ExtNormalValueAnomaly.java

示例5: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple2<K, AnomalyResult>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector , Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

         TypeInformation<Tuple2<K,Tuple4<Double,Double,Long,Long>>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple2.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO)});

        Tuple2<K,Tuple4<Double,Double,Long,Long>> init= new Tuple2<>(null,new Tuple4<>(0d,0d,0l,0l));
        KeyedStream<Tuple2<K,Tuple4<Double,Double,Long,Long>>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new CountWindFold<>(keySelector, window, resultType),new WindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:18，代码来源:NormalFreqAnomaly.java

示例6: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple2<K, AnomalyResult>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector , KeySelector<V,Double> valueSelector, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

         TypeInformation<Tuple2<K,Tuple4<Double,Double,Long,Long>>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple2.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO)});

        Tuple2<K,Tuple4<Double,Double,Long,Long>> init= new Tuple2<>(null,new Tuple4<>(0d,0d,0l,0l));
        KeyedStream<Tuple2<K,Tuple4<Double,Double,Long,Long>>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new CountSumFold<>(keySelector, valueSelector, resultType), new WindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:18，代码来源:NormalValueAnomaly.java

示例7: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple2<K, AnomalyResult>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector, KeySelector<V,Double> valueSelector, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Tuple2<K,Tuple4<Double,Double,Long,Long>>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple2.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO)});

        Tuple2<K,Tuple4<Double,Double,Long,Long>> init= new Tuple2<>(null,new Tuple4<>(0d,0d,0l,0l));
        KeyedStream<Tuple2<K,Tuple4<Double,Double,Long,Long>>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new CountSumFold<>(keySelector,valueSelector, resultType),new WindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:18，代码来源:PoissonValueAnomaly.java

示例8: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector, KeySelector<V,Double> valueSelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);


        TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtCountSumFold<>(keySelector,valueSelector,valueFold, resultType),new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:27，代码来源:ExtPoissonValueAnomaly.java

示例9: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple2<K, AnomalyResult>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector , Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);


        TypeInformation<Tuple2<K,Tuple4<Double,Double,Long,Long>>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple2.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO)});

        Tuple2<K,Tuple4<Double,Double,Long,Long>> init= new Tuple2<>(null,new Tuple4<>(0d,0d,0l,0l));
        KeyedStream<Tuple2<K,Tuple4<Double,Double,Long,Long>>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new CountWindFold<>(keySelector, window, resultType),new WindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:19，代码来源:PoissonFreqAnomaly.java

示例10: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector , PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);
 TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtCountWindFold<>(keySelector,valueFold, window, resultType),new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:25，代码来源:ExtPoissonFreqAnomaly.java

示例11: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector, KeySelector<V,Double> valueSelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);

        TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtLogCountSumFold<>(keySelector,valueSelector,valueFold, resultType),new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:26，代码来源:ExtLogNormalValueAnomaly.java

示例12: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector, KeySelector<V,Double> valueSelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);

        TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple2.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtLogCountSumFold<>(keySelector,valueSelector,valueFold, resultType),new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:26，代码来源:LogNormalValueAnomaly.java

示例13: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);

        TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtCountWindFold<>(keySelector,valueFold, window, resultType),new ExtWindowTimeExtractor(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:26，代码来源:ExtExponentialFreqAnomaly.java

示例14: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple2<K, AnomalyResult>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector,  Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Tuple2<K,Tuple4<Double,Double,Long,Long>>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple2.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO)});

        Tuple2<K,Tuple4<Double,Double,Long,Long>> init= new Tuple2<>(null,new Tuple4<>(0d,0d,0l,0l));
        KeyedStream<Tuple2<K,Tuple4<Double,Double,Long,Long>>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new CountWindFold<>(keySelector, window, resultType),new WindowTimeExtractor(resultType))
                .keyBy(0);


        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:19，代码来源:ExponentialFreqAnomaly.java

示例15: getAnomalySteam

import org.apache.flink.streaming.api.datastream.KeyedStream; //导入依赖的package包/类
public DataStream<Tuple3<K, AnomalyResult, RV>> getAnomalySteam(DataStream<V> ds, KeySelector<V, K> keySelector, KeySelector<V,Double> valueSelector, PayloadFold<V, RV> valueFold, Time window) {

        KeyedStream<V, K> keyedInput = ds
                .keyBy(keySelector);

        TypeInformation<Object> foldResultType = TypeExtractor.getUnaryOperatorReturnType(valueFold,
                PayloadFold.class,
                false,
                false,
                ds.getType(),
                "PayloadFold",
                false);

        TypeInformation<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>> resultType = (TypeInformation) new TupleTypeInfo<>(Tuple3.class,
                new TypeInformation[] {keyedInput.getKeyType(), new TupleTypeInfo(Tuple4.class,
                        BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.DOUBLE_TYPE_INFO, BasicTypeInfo.LONG_TYPE_INFO,BasicTypeInfo.LONG_TYPE_INFO), foldResultType});

        Tuple3<K,Tuple4<Double,Double,Long,Long>, RV> init= new Tuple3<>(null,new Tuple4<>(0d,0d,0l,0l), valueFold.getInit());
        KeyedStream<Tuple3<K,Tuple4<Double,Double,Long,Long>,RV>, Tuple> kPreStream = keyedInput
                .timeWindow(window)
                .apply(init, new ExtCountSumFold<>(keySelector,valueSelector,valueFold, resultType),new ExtWindowTimeExtractor<>(resultType))
                .keyBy(0);

        return kPreStream.flatMap(afm);
    }
开发者ID:sics-dna，项目名称:isc4flink，代码行数:26，代码来源:ExtExponentialValueAnomaly.java


注：本文中的org.apache.flink.streaming.api.datastream.KeyedStream类示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台，相关代码片段筛选自各路编程大神贡献的开源项目，源码版权归原作者所有，传播和使用请参考对应项目的License；未经允许，请勿转载。