.keyBy(LogLine::getIp) .valueBy(line -> 1L) .combineBy(Sums.ofLongs())
.keyBy(c -> toRegionIdUninitialized(bufferedKeys, endKeys, ByteBuffer.wrap(c.getRowArray(), c.getRowOffset(), c.getRowLength())))
.keyBy(Object::hashCode) .reduceBy((Stream<Cell> s, Collector<Cell> ctx) -> s.forEach(ctx::collect)) .outputValues();
Dataset<Pair<String, Integer>> pairs = ReduceByKey.of(input) .keyBy(e -> "") .valueBy(e -> 1) .combineBy(Sums.ofInts())
ReduceByKey.named("second") .of(mapped) .keyBy(e -> e) .reduceBy(values -> 1L) .output(ComputationHint.EXPENSIVE);
Dataset<Pair<String, Integer>> pairs = ReduceByKey.of(input) .keyBy(e -> "") .valueBy(e -> 1) .combineBy(Sums.ofInts())
.keyBy(String::toLowerCase) .valueBy(e -> 1L) .combineBy(Sums.ofLongs())
.keyBy(Pair::getFirst) .valueBy(Pair::getSecond) .combineBy(Sums.ofLongs())
.keyBy(e -> e.getFirst().word) .valueBy(Pair::getSecond) .combineBy(Sums.ofLongs())
Dataset<Pair<String, String>> reduced = ReduceByKey.of(f.createInput(source, Triple::getThird)) .keyBy(Triple::getFirst) .valueBy(Triple::getSecond) .combineBy(xs -> {
.keyBy(Pair::getFirst) .valueBy(e -> 1L) .combineBy(Sums.ofLongs())
Dataset<Pair<String, Long>> reduced = ReduceByKey.of(f.createInput(source, Pair::getSecond)) .keyBy(Pair::getFirst) .valueBy(e -> 1L) .reduceBy(Sums.ofLongs())
.keyBy(e -> "") .valueBy(e -> e) .reduceBy(s -> s.collect(Collectors.toSet()))
.named("reduceByKeyTwoHints") .of(mapped) .keyBy(e -> e) .reduceBy(values -> 1L) .windowBy(Time.of(Duration.ofSeconds(1)))
Dataset<Pair<String, Long>> reduced = ReduceByKey.of(input) .keyBy(Pair::getFirst) .valueBy(e -> 1L) .combineBy(Sums.ofLongs())