@Test
public void testChannelSelectors() {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStreamSource<Long> src = env.generateSequence(0, 0);
DataStream<Long> broadcast = src.broadcast();
DataStreamSink<Long> broadcastSink = broadcast.print();
StreamPartitioner<?> broadcastPartitioner =
env.getStreamGraph()
.getStreamEdges(src.getId(), broadcastSink.getTransformation().getId())
.get(0)
.getPartitioner();
assertTrue(broadcastPartitioner instanceof BroadcastPartitioner);
DataStream<Long> shuffle = src.shuffle();
DataStreamSink<Long> shuffleSink = shuffle.print();
StreamPartitioner<?> shufflePartitioner =
env.getStreamGraph()
.getStreamEdges(src.getId(), shuffleSink.getTransformation().getId())
.get(0)
.getPartitioner();
assertTrue(shufflePartitioner instanceof ShufflePartitioner);
DataStream<Long> forward = src.forward();
DataStreamSink<Long> forwardSink = forward.print();
StreamPartitioner<?> forwardPartitioner =
env.getStreamGraph()
.getStreamEdges(src.getId(), forwardSink.getTransformation().getId())
.get(0)
.getPartitioner();
assertTrue(forwardPartitioner instanceof ForwardPartitioner);
DataStream<Long> rebalance = src.rebalance();
DataStreamSink<Long> rebalanceSink = rebalance.print();
StreamPartitioner<?> rebalancePartitioner =
env.getStreamGraph()
.getStreamEdges(src.getId(), rebalanceSink.getTransformation().getId())
.get(0)
.getPartitioner();
assertTrue(rebalancePartitioner instanceof RebalancePartitioner);
DataStream<Long> global = src.global();
DataStreamSink<Long> globalSink = global.print();
StreamPartitioner<?> globalPartitioner =
env.getStreamGraph()
.getStreamEdges(src.getId(), globalSink.getTransformation().getId())
.get(0)
.getPartitioner();
assertTrue(globalPartitioner instanceof GlobalPartitioner);
}
@Test
public void testTypeInfo() {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStream<Long> src1 = env.generateSequence(0, 0);
assertEquals(TypeExtractor.getForClass(Long.class), src1.getType());
DataStream<Tuple2<Integer, String>> map =
src1.map(
new MapFunction<Long, Tuple2<Integer, String>>() {
@Override
public Tuple2<Integer, String> map(Long value) throws Exception {
return null;
}
});
assertEquals(TypeExtractor.getForObject(new Tuple2<>(0, "")), map.getType());
DataStream<String> window =
map.windowAll(GlobalWindows.create())
.trigger(PurgingTrigger.of(CountTrigger.of(5)))
.apply(
new AllWindowFunction<Tuple2<Integer, String>, String, GlobalWindow>() {
@Override
public void apply(
GlobalWindow window,
Iterable<Tuple2<Integer, String>> values,
Collector<String> out)
throws Exception {}
});
assertEquals(TypeExtractor.getForClass(String.class), window.getType());
DataStream<CustomPOJO> flatten =
window
.windowAll(GlobalWindows.create())
.trigger(PurgingTrigger.of(CountTrigger.of(5)))
.fold(
new CustomPOJO(),
new FoldFunction<String, CustomPOJO>() {
private static final long serialVersionUID = 1L;
@Override
public CustomPOJO fold(CustomPOJO accumulator, String value) throws Exception {
return null;
}
});
assertEquals(TypeExtractor.getForClass(CustomPOJO.class), flatten.getType());
}
@Test
public void testTypeInfo() {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStream<Long> src1 = env.generateSequence(0, 0);
assertEquals(TypeExtractor.getForClass(Long.class), src1.getType());
DataStream<Tuple2<Integer, String>> map =
src1.map(
new MapFunction<Long, Tuple2<Integer, String>>() {
@Override
public Tuple2<Integer, String> map(Long value) throws Exception {
return null;
}
});
assertEquals(TypeExtractor.getForObject(new Tuple2<Integer, String>(0, "")), map.getType());
WindowedDataStream<String> window =
map.window(Count.of(5))
.mapWindow(
new WindowMapFunction<Tuple2<Integer, String>, String>() {
@Override
public void mapWindow(
Iterable<Tuple2<Integer, String>> values, Collector<String> out)
throws Exception {}
});
assertEquals(TypeExtractor.getForClass(String.class), window.getType());
DataStream<CustomPOJO> flatten =
window
.foldWindow(
new CustomPOJO(),
new FoldFunction<String, CustomPOJO>() {
@Override
public CustomPOJO fold(CustomPOJO accumulator, String value) throws Exception {
return null;
}
})
.flatten();
assertEquals(TypeExtractor.getForClass(CustomPOJO.class), flatten.getType());
}
/**
* Tests {@link SingleOutputStreamOperator#name(String)} functionality.
*
* @throws Exception
*/
@Test
public void testNaming() throws Exception {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStream<Long> dataStream1 =
env.generateSequence(0, 0)
.name("testSource1")
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.name("testMap");
DataStream<Long> dataStream2 =
env.generateSequence(0, 0)
.name("testSource2")
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.name("testMap");
DataStreamSink<Long> connected =
dataStream1
.connect(dataStream2)
.flatMap(
new CoFlatMapFunction<Long, Long, Long>() {
private static final long serialVersionUID = 1L;
@Override
public void flatMap1(Long value, Collector<Long> out) throws Exception {}
@Override
public void flatMap2(Long value, Collector<Long> out) throws Exception {}
})
.name("testCoFlatMap")
.windowAll(GlobalWindows.create())
.trigger(PurgingTrigger.of(CountTrigger.of(10)))
.fold(
0L,
new FoldFunction<Long, Long>() {
private static final long serialVersionUID = 1L;
@Override
public Long fold(Long accumulator, Long value) throws Exception {
return null;
}
})
.name("testWindowFold")
.print();
// test functionality through the operator names in the execution plan
String plan = env.getExecutionPlan();
assertTrue(plan.contains("testSource1"));
assertTrue(plan.contains("testSource2"));
assertTrue(plan.contains("testMap"));
assertTrue(plan.contains("testMap"));
assertTrue(plan.contains("testCoFlatMap"));
assertTrue(plan.contains("testWindowFold"));
}
@Test
public void sinkKeyTest() {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStreamSink<Long> sink = env.generateSequence(1, 100).print();
assertTrue(
env.getStreamGraph().getStreamNode(sink.getTransformation().getId()).getStatePartitioner()
== null);
assertTrue(
env.getStreamGraph()
.getStreamNode(sink.getTransformation().getId())
.getInEdges()
.get(0)
.getPartitioner()
instanceof ForwardPartitioner);
KeySelector<Long, Long> key1 =
new KeySelector<Long, Long>() {
private static final long serialVersionUID = 1L;
@Override
public Long getKey(Long value) throws Exception {
return (long) 0;
}
};
DataStreamSink<Long> sink2 = env.generateSequence(1, 100).keyBy(key1).print();
assertTrue(
env.getStreamGraph().getStreamNode(sink2.getTransformation().getId()).getStatePartitioner()
!= null);
assertEquals(
key1,
env.getStreamGraph()
.getStreamNode(sink2.getTransformation().getId())
.getStatePartitioner());
assertTrue(
env.getStreamGraph()
.getStreamNode(sink2.getTransformation().getId())
.getInEdges()
.get(0)
.getPartitioner()
instanceof HashPartitioner);
KeySelector<Long, Long> key2 =
new KeySelector<Long, Long>() {
private static final long serialVersionUID = 1L;
@Override
public Long getKey(Long value) throws Exception {
return (long) 0;
}
};
DataStreamSink<Long> sink3 = env.generateSequence(1, 100).keyBy(key2).print();
assertTrue(
env.getStreamGraph().getStreamNode(sink3.getTransformation().getId()).getStatePartitioner()
!= null);
assertEquals(
key2,
env.getStreamGraph()
.getStreamNode(sink3.getTransformation().getId())
.getStatePartitioner());
assertTrue(
env.getStreamGraph()
.getStreamNode(sink3.getTransformation().getId())
.getInEdges()
.get(0)
.getPartitioner()
instanceof HashPartitioner);
}
@Test
public void operatorTest() {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStreamSource<Long> src = env.generateSequence(0, 0);
MapFunction<Long, Integer> mapFunction =
new MapFunction<Long, Integer>() {
@Override
public Integer map(Long value) throws Exception {
return null;
}
};
DataStream<Integer> map = src.map(mapFunction);
map.addSink(new NoOpSink<Integer>());
assertEquals(mapFunction, getFunctionForDataStream(map));
FlatMapFunction<Long, Integer> flatMapFunction =
new FlatMapFunction<Long, Integer>() {
private static final long serialVersionUID = 1L;
@Override
public void flatMap(Long value, Collector<Integer> out) throws Exception {}
};
DataStream<Integer> flatMap = src.flatMap(flatMapFunction);
flatMap.addSink(new NoOpSink<Integer>());
assertEquals(flatMapFunction, getFunctionForDataStream(flatMap));
FilterFunction<Integer> filterFunction =
new FilterFunction<Integer>() {
@Override
public boolean filter(Integer value) throws Exception {
return false;
}
};
DataStream<Integer> unionFilter = map.union(flatMap).filter(filterFunction);
unionFilter.addSink(new NoOpSink<Integer>());
assertEquals(filterFunction, getFunctionForDataStream(unionFilter));
try {
env.getStreamGraph().getStreamEdge(map.getId(), unionFilter.getId());
} catch (RuntimeException e) {
fail(e.getMessage());
}
try {
env.getStreamGraph().getStreamEdge(flatMap.getId(), unionFilter.getId());
} catch (RuntimeException e) {
fail(e.getMessage());
}
OutputSelector<Integer> outputSelector =
new OutputSelector<Integer>() {
@Override
public Iterable<String> select(Integer value) {
return null;
}
};
SplitStream<Integer> split = unionFilter.split(outputSelector);
split.select("dummy").addSink(new NoOpSink<Integer>());
List<OutputSelector<?>> outputSelectors =
env.getStreamGraph().getStreamNode(unionFilter.getId()).getOutputSelectors();
assertEquals(1, outputSelectors.size());
assertEquals(outputSelector, outputSelectors.get(0));
DataStream<Integer> select = split.select("a");
DataStreamSink<Integer> sink = select.print();
StreamEdge splitEdge =
env.getStreamGraph().getStreamEdge(unionFilter.getId(), sink.getTransformation().getId());
assertEquals("a", splitEdge.getSelectedNames().get(0));
ConnectedStreams<Integer, Integer> connect = map.connect(flatMap);
CoMapFunction<Integer, Integer, String> coMapper =
new CoMapFunction<Integer, Integer, String>() {
private static final long serialVersionUID = 1L;
@Override
public String map1(Integer value) {
return null;
}
@Override
public String map2(Integer value) {
return null;
}
};
DataStream<String> coMap = connect.map(coMapper);
coMap.addSink(new NoOpSink<String>());
assertEquals(coMapper, getFunctionForDataStream(coMap));
try {
env.getStreamGraph().getStreamEdge(map.getId(), coMap.getId());
} catch (RuntimeException e) {
fail(e.getMessage());
}
try {
env.getStreamGraph().getStreamEdge(flatMap.getId(), coMap.getId());
} catch (RuntimeException e) {
fail(e.getMessage());
}
}
/** Tests whether parallelism gets set. */
@Test
public void testParallelism() {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
DataStreamSource<Tuple2<Long, Long>> src = env.fromElements(new Tuple2<>(0L, 0L));
env.setParallelism(10);
SingleOutputStreamOperator<Long, ?> map =
src.map(
new MapFunction<Tuple2<Long, Long>, Long>() {
@Override
public Long map(Tuple2<Long, Long> value) throws Exception {
return null;
}
})
.name("MyMap");
DataStream<Long> windowed =
map.windowAll(GlobalWindows.create())
.trigger(PurgingTrigger.of(CountTrigger.of(10)))
.fold(
0L,
new FoldFunction<Long, Long>() {
@Override
public Long fold(Long accumulator, Long value) throws Exception {
return null;
}
});
windowed.addSink(new NoOpSink<Long>());
DataStreamSink<Long> sink =
map.addSink(
new SinkFunction<Long>() {
private static final long serialVersionUID = 1L;
@Override
public void invoke(Long value) throws Exception {}
});
assertEquals(1, env.getStreamGraph().getStreamNode(src.getId()).getParallelism());
assertEquals(10, env.getStreamGraph().getStreamNode(map.getId()).getParallelism());
assertEquals(1, env.getStreamGraph().getStreamNode(windowed.getId()).getParallelism());
assertEquals(
10, env.getStreamGraph().getStreamNode(sink.getTransformation().getId()).getParallelism());
env.setParallelism(7);
// Some parts, such as windowing rely on the fact that previous operators have a parallelism
// set when instantiating the Discretizer. This would break if we dynamically changed
// the parallelism of operations when changing the setting on the Execution Environment.
assertEquals(1, env.getStreamGraph().getStreamNode(src.getId()).getParallelism());
assertEquals(10, env.getStreamGraph().getStreamNode(map.getId()).getParallelism());
assertEquals(1, env.getStreamGraph().getStreamNode(windowed.getId()).getParallelism());
assertEquals(
10, env.getStreamGraph().getStreamNode(sink.getTransformation().getId()).getParallelism());
try {
src.setParallelism(3);
fail();
} catch (IllegalArgumentException success) {
// do nothing
}
DataStreamSource<Long> parallelSource = env.generateSequence(0, 0);
parallelSource.addSink(new NoOpSink<Long>());
assertEquals(7, env.getStreamGraph().getStreamNode(parallelSource.getId()).getParallelism());
parallelSource.setParallelism(3);
assertEquals(3, env.getStreamGraph().getStreamNode(parallelSource.getId()).getParallelism());
map.setParallelism(2);
assertEquals(2, env.getStreamGraph().getStreamNode(map.getId()).getParallelism());
sink.setParallelism(4);
assertEquals(
4, env.getStreamGraph().getStreamNode(sink.getTransformation().getId()).getParallelism());
}
/**
* Tests union functionality. This ensures that self-unions and unions of streams with differing
* parallelism work.
*
* @throws Exception
*/
@Test
public void testUnion() throws Exception {
StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
env.setParallelism(4);
DataStream<Long> input1 =
env.generateSequence(0, 0)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
});
DataStream<Long> selfUnion =
input1
.union(input1)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
});
DataStream<Long> input6 =
env.generateSequence(0, 0)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
});
DataStream<Long> selfUnionDifferentPartition =
input6
.broadcast()
.union(input6)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
});
DataStream<Long> input2 =
env.generateSequence(0, 0)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.setParallelism(4);
DataStream<Long> input3 =
env.generateSequence(0, 0)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.setParallelism(2);
DataStream<Long> unionDifferingParallelism =
input2
.union(input3)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.setParallelism(4);
DataStream<Long> input4 =
env.generateSequence(0, 0)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.setParallelism(2);
DataStream<Long> input5 =
env.generateSequence(0, 0)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.setParallelism(4);
DataStream<Long> unionDifferingPartitioning =
input4
.broadcast()
.union(input5)
.map(
new MapFunction<Long, Long>() {
@Override
public Long map(Long value) throws Exception {
return null;
}
})
.setParallelism(4);
StreamGraph streamGraph = env.getStreamGraph();
// verify self union
assertTrue(streamGraph.getStreamNode(selfUnion.getId()).getInEdges().size() == 2);
for (StreamEdge edge : streamGraph.getStreamNode(selfUnion.getId()).getInEdges()) {
assertTrue(edge.getPartitioner() instanceof ForwardPartitioner);
}
// verify self union with differnt partitioners
assertTrue(
streamGraph.getStreamNode(selfUnionDifferentPartition.getId()).getInEdges().size() == 2);
boolean hasForward = false;
boolean hasBroadcast = false;
for (StreamEdge edge :
streamGraph.getStreamNode(selfUnionDifferentPartition.getId()).getInEdges()) {
if (edge.getPartitioner() instanceof ForwardPartitioner) {
hasForward = true;
}
if (edge.getPartitioner() instanceof BroadcastPartitioner) {
hasBroadcast = true;
}
}
assertTrue(hasForward && hasBroadcast);
// verify union of streams with differing parallelism
assertTrue(
streamGraph.getStreamNode(unionDifferingParallelism.getId()).getInEdges().size() == 2);
for (StreamEdge edge :
streamGraph.getStreamNode(unionDifferingParallelism.getId()).getInEdges()) {
if (edge.getSourceId() == input2.getId()) {
assertTrue(edge.getPartitioner() instanceof ForwardPartitioner);
} else if (edge.getSourceId() == input3.getId()) {
assertTrue(edge.getPartitioner() instanceof RebalancePartitioner);
} else {
fail("Wrong input edge.");
}
}
// verify union of streams with differing partitionings
assertTrue(
streamGraph.getStreamNode(unionDifferingPartitioning.getId()).getInEdges().size() == 2);
for (StreamEdge edge :
streamGraph.getStreamNode(unionDifferingPartitioning.getId()).getInEdges()) {
if (edge.getSourceId() == input4.getId()) {
assertTrue(edge.getPartitioner() instanceof BroadcastPartitioner);
} else if (edge.getSourceId() == input5.getId()) {
assertTrue(edge.getPartitioner() instanceof ForwardPartitioner);
} else {
fail("Wrong input edge.");
}
}
}