private DataflowPipeline buildPipeline(DataflowPipelineOptions options) {
DataflowPipeline p = DataflowPipeline.create(options);
p.apply(TextIO.Read.named("ReadMyFile").from("gs://bucket/object"))
.apply(TextIO.Write.named("WriteMyFile").to("gs://bucket/object"));
return p;
}
@Override
protected void testProgram() throws Exception {
Pipeline p = FlinkTestPipeline.create();
PCollection<TableRow> input1 = p.apply(Create.of(EVENT_ARRAY));
PCollection<TableRow> input2 = p.apply(Create.of(CC_ARRAY));
PCollection<String> output = JoinExamples.joinEvents(input1, input2);
output.apply(TextIO.Write.to(resultPath));
p.run();
}
public static void main(String[] args) {
PipelineOptionsFactory.register(KafkaStreamingWordCountOptions.class);
KafkaStreamingWordCountOptions options =
PipelineOptionsFactory.fromArgs(args).as(KafkaStreamingWordCountOptions.class);
options.setJobName("KafkaExample - WindowSize: " + options.getWindowSize() + " seconds");
options.setStreaming(true);
options.setCheckpointingInterval(1000L);
options.setNumberOfExecutionRetries(5);
options.setExecutionRetryDelay(3000L);
options.setRunner(FlinkPipelineRunner.class);
System.out.println(
options.getKafkaTopic()
+ " "
+ options.getZookeeper()
+ " "
+ options.getBroker()
+ " "
+ options.getGroup());
Pipeline pipeline = Pipeline.create(options);
Properties p = new Properties();
p.setProperty("zookeeper.connect", options.getZookeeper());
p.setProperty("bootstrap.servers", options.getBroker());
p.setProperty("group.id", options.getGroup());
// this is the Flink consumer that reads the input to
// the program from a kafka topic.
FlinkKafkaConsumer08<String> kafkaConsumer =
new FlinkKafkaConsumer08<>(options.getKafkaTopic(), new SimpleStringSchema(), p);
PCollection<String> words =
pipeline
.apply(Read.from(new UnboundedFlinkSource<>(kafkaConsumer)).named("StreamingWordCount"))
.apply(ParDo.of(new ExtractWordsFn()))
.apply(
Window.<String>into(
FixedWindows.of(Duration.standardSeconds(options.getWindowSize())))
.triggering(AfterWatermark.pastEndOfWindow())
.withAllowedLateness(Duration.ZERO)
.discardingFiredPanes());
PCollection<KV<String, Long>> wordCounts = words.apply(Count.<String>perElement());
wordCounts.apply(ParDo.of(new FormatAsStringFn())).apply(TextIO.Write.to("./outputKafka.txt"));
pipeline.run();
}
/** Returns a Step for a DoFn by creating and translating a pipeline. */
private static Step createPredefinedStep() throws Exception {
DataflowPipelineOptions options = buildPipelineOptions();
DataflowPipelineTranslator translator = DataflowPipelineTranslator.fromOptions(options);
DataflowPipeline pipeline = DataflowPipeline.create(options);
String stepName = "DoFn1";
pipeline
.apply(TextIO.Read.named("ReadMyFile").from("gs://bucket/in"))
.apply(ParDo.of(new NoOpFn()).named(stepName))
.apply(TextIO.Write.named("WriteMyFile").to("gs://bucket/out"));
Job job = translator.translate(pipeline, Collections.<DataflowPackage>emptyList()).getJob();
assertEquals(3, job.getSteps().size());
Step step = job.getSteps().get(1);
assertEquals(stepName, getString(step.getProperties(), PropertyNames.USER_NAME));
return step;
}
@Test
public void testPredefinedAddStep() throws Exception {
DataflowPipelineOptions options = buildPipelineOptions();
DataflowPipelineTranslator translator = DataflowPipelineTranslator.fromOptions(options);
DataflowPipelineTranslator.registerTransformTranslator(
EmbeddedTransform.class, new EmbeddedTranslator());
// Create a predefined step using another pipeline
Step predefinedStep = createPredefinedStep();
// Create a pipeline that the predefined step will be embedded into
DataflowPipeline pipeline = DataflowPipeline.create(options);
pipeline
.apply(TextIO.Read.named("ReadMyFile").from("gs://bucket/in"))
.apply(ParDo.of(new NoOpFn()))
.apply(new EmbeddedTransform(predefinedStep.clone()))
.apply(TextIO.Write.named("WriteMyFile").to("gs://bucket/out"));
Job job = translator.translate(pipeline, Collections.<DataflowPackage>emptyList()).getJob();
List<Step> steps = job.getSteps();
assertEquals(4, steps.size());
// The input to the embedded step should match the output of the step before
Map<String, Object> step1Out = getOutputPortReference(steps.get(1));
Map<String, Object> step2In =
getDictionary(steps.get(2).getProperties(), PropertyNames.PARALLEL_INPUT);
assertEquals(step1Out, step2In);
// The output from the embedded step should match the input of the step after
Map<String, Object> step2Out = getOutputPortReference(steps.get(2));
Map<String, Object> step3In =
getDictionary(steps.get(3).getProperties(), PropertyNames.PARALLEL_INPUT);
assertEquals(step2Out, step3In);
// The step should not have been modified other than remapping the input
Step predefinedStepClone = predefinedStep.clone();
Step embeddedStepClone = steps.get(2).clone();
predefinedStepClone.getProperties().remove(PropertyNames.PARALLEL_INPUT);
embeddedStepClone.getProperties().remove(PropertyNames.PARALLEL_INPUT);
assertEquals(predefinedStepClone, embeddedStepClone);
}
@Override
public PDone apply(PCollection<KV<String, KV<URI, Double>>> wordToUriAndTfIdf) {
return wordToUriAndTfIdf
.apply(
ParDo.named("Format")
.of(
new DoFn<KV<String, KV<URI, Double>>, String>() {
private static final long serialVersionUID = 0;
@Override
public void processElement(ProcessContext c) {
c.output(
String.format(
"%s,\t%s,\t%f",
c.element().getKey(),
c.element().getValue().getKey(),
c.element().getValue().getValue()));
}
}))
.apply(TextIO.Write.to(output).withSuffix(".csv"));
}