将CoGroupByKey与自定义类型一起使用会导致编码器错误

时间：2017-09-26 07:40:36

标签： google-cloud-dataflow apache-beam apache-beam-io

我想加入两个PCollection（分别来自不同的输入）并按照此处描述的步骤实现，＆＃34;加入CoGroupByKey＆＃34;部分： https://cloud.google.com/dataflow/model/group-by-key

就我而言，我想加入GeoIP＆＃34; block＆＃34;信息和＆＃34;位置＆＃34;信息。所以我将Block和Location定义为一个自定义类，然后写如下：

final TupleTag<Block> t1 = new TupleTag<Block>();
final TupleTag<Location> t2 = new TupleTag<Location>();
PCollection<KV<Long, CoGbkResult>> coGbkResultColl = KeyedPCollectionTuple.of(t1, kvGeoNameIDBlock)
        .and(t2, kvGeoNameIDLocation).apply(CoGroupByKey.<Long>create());

密钥具有Long类型值。我认为已经完成但是当我运行mvn compile时，它会输出以下错误：

[ERROR] Failed to execute goal org.codehaus.mojo:exec-maven-plugin:1.4.0:java (default-cli) on project xxxx: An exception occured while executing the Java class. null: InvocationTargetException: Unable to return a default Coder for Extract GeoNameID-Block KV/ParMultiDo(ExtractGeoNameIDBlock).out0 [PCollection]. Correct one of the following root causes:
[ERROR]   No Coder has been manually specified;  you may do so using .setCoder().
[ERROR]   Inferring a Coder from the CoderRegistry failed: Cannot provide coder for parameterized type org.apache.beam.sdk.values.KV<java.lang.Long, com.xxx.platform.geoip2.Block>: Unable to provide a Coder for com.xxx.platform.geoip2.Block.
[ERROR]   Building a Coder using a registered CoderProvider failed.
[ERROR]   See suppressed exceptions for detailed failures.
[ERROR]   Using the default output Coder from the producing PTransform failed: Cannot provide coder for parameterized type org.apache.beam.sdk.values.KV<java.lang.Long, com.xxx.platform.geoip2.Block>: Unable to provide a Coder for com.xxx.platform.geoip2.Block.

输出错误的确切DoFn是ExtractGeoNameIDBlock，它只是创建其键（要连接）和自身的键值对。

// ExtractGeoNameIDBlock creates KV collection while reading from block CSV
static class ExtractGeoNameIDBlock extends DoFn<String, KV<Long, Block>> {
private static final long serialVersionUID = 1L;

  @ProcessElement
  public void processElement(ProcessContext c) throws Exception {
    String line = c.element();

    if (!line.startsWith("network,")) { // exclude headerline
      Block b = new Block();
      b.loadFromCsvLine(line);

      if (b.getGeonameId() != null) {
        c.output(KV.of(b.getGeonameId(), b));
      }
    }
  }
}

loadFromCsvLine只需解析CSV行，将字段转换为每个对应的类型并分配到其私有字段。

所以看起来我需要为我的自定义类设置一些编码器才能使它工作。 我发现了一个引用编码器的文件，但仍不确定如何实施我的文件。 https://cloud.google.com/dataflow/model/data-encoding

我可以遵循为自定义类创建自定义编码器的真实示例吗？

[更新13:02 09/26/2017] 我添加了

CoderRegistry cr = p.getCoderRegistry();
cr.registerCoderForClass(Block.class, AvroCoder.of(Block.class));

然后出错了

 java.lang.NullPointerException: in com.xxx.platform.geoip2.Block in long null of long in field representedCountryGeonameId of com.xxx.platform.geoip2.Block

[更新14:05 09/26/2017] 我改变了这样的实现：

@DefaultCoder(AvroCoder.class)
public class Block {
    private static final Logger LOG = LoggerFactory.getLogger(Block.class);

    @Nullable
    public String network;
    @Nullable
    public Long registeredCountryGeonameId;
:
:

（将@Nullable设置为所有属性）

但仍然有这个错误：

(22eeaf3dfb26f8cc): java.lang.RuntimeException: org.apache.beam.sdk.coders.CoderException: cannot encode a null Long
    at com.google.cloud.dataflow.worker.SimpleParDoFn$1.output(SimpleParDoFn.java:191)
    at org.apache.beam.runners.core.SimpleDoFnRunner.outputWindowedValue(SimpleDoFnRunner.java:211)
    at org.apache.beam.runners.core.SimpleDoFnRunner.access$700(SimpleDoFnRunner.java:66)
    at org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:436)
    at org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:424)
    at org.apache.beam.sdk.transforms.join.CoGroupByKey$ConstructUnionTableFn.processElement(CoGroupByKey.java:185)
Caused by: org.apache.beam.sdk.coders.CoderException: cannot encode a null Long
    at org.apache.beam.sdk.coders.VarLongCoder.encode(VarLongCoder.java:51)
    at org.apache.beam.sdk.coders.VarLongCoder.encode(VarLongCoder.java:35)
    at org.apache.beam.sdk.coders.Coder.encode(Coder.java:135)
    at com.google.cloud.dataflow.worker.ShuffleSink$ShuffleSinkWriter.encodeToChunk(ShuffleSink.java:320)
    at com.google.cloud.dataflow.worker.ShuffleSink$ShuffleSinkWriter.add(ShuffleSink.java:216)
    at com.google.cloud.dataflow.worker.ShuffleSink$ShuffleSinkWriter.add(ShuffleSink.java:178)
    at com.google.cloud.dataflow.worker.util.common.worker.WriteOperation.process(WriteOperation.java:80)
    at com.google.cloud.dataflow.worker.util.common.worker.OutputReceiver.process(OutputReceiver.java:52)
    at com.google.cloud.dataflow.worker.ReifyTimestampAndWindowsParDoFnFactory$ReifyTimestampAndWindowsParDoFn.processElement(ReifyTimestampAndWindowsParDoFnFactory.java:68)
    at com.google.cloud.dataflow.worker.util.common.worker.ParDoOperation.process(ParDoOperation.java:48)
    at com.google.cloud.dataflow.worker.util.common.worker.OutputReceiver.process(OutputReceiver.java:52)
    at com.google.cloud.dataflow.worker.SimpleParDoFn$1.output(SimpleParDoFn.java:183)
    at org.apache.beam.runners.core.SimpleDoFnRunner.outputWindowedValue(SimpleDoFnRunner.java:211)
    at org.apache.beam.runners.core.SimpleDoFnRunner.access$700(SimpleDoFnRunner.java:66)
    at org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:436)
    at org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:424)
    at org.apache.beam.sdk.transforms.join.CoGroupByKey$ConstructUnionTableFn.processElement(CoGroupByKey.java:185)
    at org.apache.beam.sdk.transforms.join.CoGroupByKey$ConstructUnionTableFn$DoFnInvoker.invokeProcessElement(Unknown Source)
    at org.apache.beam.runners.core.SimpleDoFnRunner.invokeProcessElement(SimpleDoFnRunner.java:177)
    at org.apache.beam.runners.core.SimpleDoFnRunner.processElement(SimpleDoFnRunner.java:141)
    at com.google.cloud.dataflow.worker.SimpleParDoFn.processElement(SimpleParDoFn.java:233)
    at com.google.cloud.dataflow.worker.util.common.worker.ParDoOperation.process(ParDoOperation.java:48)
    at com.google.cloud.dataflow.worker.util.common.worker.OutputReceiver.process(OutputReceiver.java:52)
    at com.google.cloud.dataflow.worker.SimpleParDoFn$1.output(SimpleParDoFn.java:183)
    at org.apache.beam.runners.core.SimpleDoFnRunner.outputWindowedValue(SimpleDoFnRunner.java:211)
    at org.apache.beam.runners.core.SimpleDoFnRunner.access$700(SimpleDoFnRunner.java:66)
    at org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:436)
    at org.apache.beam.runners.core.SimpleDoFnRunner$DoFnProcessContext.output(SimpleDoFnRunner.java:424)
    at com.bandainamcoent.platform.GeoIpPopulateTable$ExtractGeoNameIDBlock.processElement(GeoIpPopulateTable.java:79)
    at com.bandainamcoent.platform.GeoIpPopulateTable$ExtractGeoNameIDBlock$DoFnInvoker.invokeProcessElement(Unknown Source)
    at org.apache.beam.runners.core.SimpleDoFnRunner.invokeProcessElement(SimpleDoFnRunner.java:177)
    at org.apache.beam.runners.core.SimpleDoFnRunner.processElement(SimpleDoFnRunner.java:141)
    at com.google.cloud.dataflow.worker.SimpleParDoFn.processElement(SimpleParDoFn.java:233)
    at com.google.cloud.dataflow.worker.util.common.worker.ParDoOperation.process(ParDoOperation.java:48)
    at com.google.cloud.dataflow.worker.util.common.worker.OutputReceiver.process(OutputReceiver.java:52)
    at com.google.cloud.dataflow.worker.util.common.worker.ReadOperation.runReadLoop(ReadOperation.java:187)
    at com.google.cloud.dataflow.worker.util.common.worker.ReadOperation.start(ReadOperation.java:148)
    at com.google.cloud.dataflow.worker.util.common.worker.MapTaskExecutor.execute(MapTaskExecutor.java:68)
    at com.google.cloud.dataflow.worker.DataflowWorker.executeWork(DataflowWorker.java:336)
    at com.google.cloud.dataflow.worker.DataflowWorker.doWork(DataflowWorker.java:294)
    at com.google.cloud.dataflow.worker.DataflowWorker.getAndPerformWork(DataflowWorker.java:244)
    at com.google.cloud.dataflow.worker.DataflowBatchWorkerHarness$WorkerThread.doWork(DataflowBatchWorkerHarness.java:135)
    at com.google.cloud.dataflow.worker.DataflowBatchWorkerHarness$WorkerThread.call(DataflowBatchWorkerHarness.java:115)
    at com.google.cloud.dataflow.worker.DataflowBatchWorkerHarness$WorkerThread.call(DataflowBatchWorkerHarness.java:102)
    at java.util.concurrent.FutureTask.run(FutureTask.java:266)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
    at java.lang.Thread.run(Thread.java:745)

感谢。

2 个答案:

答案 0 :(得分：1)

您的自定义类Block看起来没有指定编码器。您可以创建自己的Coder，也可以使用AvroCoder之类的常规CoderRegistry。您还应该使用Block注册它，以便管道知道如何编码{{1}}。

答案 1 :(得分：0)

我终于通过使用AvroCoder + Nullable注释来实现它，因为我在14:05 09/26时发布更新 在我的问题中。

我看到的最后一个错误只是因为我的数据实际上有一个我没想到的空值。在我的Java代码中处理null值之后，一切正常。

我认为这篇关于另一个问题的帖子对这个问题非常有用： https://stackoverflow.com/a/32342403/2543803

相关问题
最新问题