RonBarkan opened a new issue, #29577:
URL: https://github.com/apache/beam/issues/29577

   ### What happened?
   
   When using 
`WithKeys.of(<lambda>).withKeyType(TypeDescriptor.of(TypeWithSchema.class)` for 
types annotated with `@DefaultSchema(JavaBeanSchema.class)`, Beam is not able 
to resolve the coder for the type, despite the schema being available.
   
   See also 
https://stackoverflow.com/questions/77575574/unable-to-return-a-default-coder-for-withkeys-when-using-defaultschemajavabean
   
   In particular, the docs say:
   
   > A PCollection with a schema does not need to have a Coder specified, as 
Beam knows how to encode and decode Schema rows;
   > Beam uses a special coder to encode schema types.
   
   However, when using such a type with `WithKeys`, I consistently get this 
exception:
   ```
   java.lang.IllegalStateException: Unable to return a default Coder for 
WithKeys/AddKeys/Map/ParMultiDo(Anonymous).output [PCollection@621486957]. 
Correct one of the following root causes:
     No Coder has been manually specified;  you may do so using .setCoder().
     Inferring a Coder from the CoderRegistry failed: Cannot provide coder for 
parameterized type org.apache.beam.sdk.values.KV<K, V>: Unable to provide a 
Coder for K.
     Building a Coder using a registered CoderProvider failed.
     See suppressed exceptions for detailed failures.
     Using the default output Coder from the producing PTransform failed: 
PTransform.getOutputCoder called.
   ```
   
   
   Example POJO:
   ```
   @DefaultSchema(JavaBeanSchema.class)
   public class Pojo {
     private final long num;
     private final String str;
   
     @SchemaCreate
     public Pojo(long num, String str) {
       this.num = num;
       this.str = str;
     }
   
     public long getNum() {
       return this.num;
     }
   
     public String getStr() {
       return this.str;
     }
   
     @Override
     public boolean equals(final Object o) {
        ...
     }
   
     @Override
     public int hashCode() {
       ...
     }
   }
   ```
   
   Example pipeline which generates this error:
   ```
    public static void main(String[] args) {
       PipelineOptions options = PipelineOptionsFactory.fromArgs(args).create();
       Pipeline p = Pipeline.create(options);
   
       PCollection<String> a =
           p.apply(Create.of(List.of("1", "2", 
"3")).withType(TypeDescriptors.strings()));
       PCollection<String> b =
           p.apply(Create.of(List.of("2", "3", 
"4")).withType(TypeDescriptors.strings()));
   
       PCollection<KV<Pojo, String>> a1 =
           a.apply(
               WithKeys.<Pojo, String>of(v -> new Pojo(1, v))
                   .withKeyType(TypeDescriptor.of(Pojo.class)));
   
       PCollection<KV<Pojo, String>> b1 =
           b.apply(
               WithKeys.<Pojo, String>of(v -> new Pojo(1, v))
                   .withKeyType(TypeDescriptor.of(Pojo.class)));
   
       var aTag = new TupleTag<String>();
       var bTag = new TupleTag<String>();
       KeyedPCollectionTuple.of(aTag, a1)  // <== Exception here (complaining 
about a1)
           .and(bTag, b1)
           .apply(CoGroupByKey.create())
           .apply(
               ParDo.of(
                   new DoFn<KV<Pojo, CoGbkResult>, Integer>() {
                     @ProcessElement
                     public void processElement(ProcessContext context) {
                       var element = context.element();
                       String a = element.getValue().getOnly(aTag, null);
                       String b = element.getValue().getOnly(bTag, null);
                       context.output(a != null && b != null ? 1 : 0);
                     }
                   }));
     }
   ```
   
   
   ### Issue Priority
   
   Priority: 2 (default / most bugs should be filed as P2)
   
   ### Issue Components
   
   - [ ] Component: Python SDK
   - [X] Component: Java SDK
   - [ ] Component: Go SDK
   - [ ] Component: Typescript SDK
   - [ ] Component: IO connector
   - [ ] Component: Beam YAML
   - [ ] Component: Beam examples
   - [ ] Component: Beam playground
   - [ ] Component: Beam katas
   - [ ] Component: Website
   - [ ] Component: Spark Runner
   - [ ] Component: Flink Runner
   - [ ] Component: Samza Runner
   - [ ] Component: Twister2 Runner
   - [ ] Component: Hazelcast Jet Runner
   - [ ] Component: Google Cloud Dataflow Runner


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to