使用flink将kafka数据以拼花格式存储在hdfs中？

Question

Store kafka data in hdfs as parquet format using flink, I am trying with fink documentation which is not working.

I am not finding any proper documentations to store it as parquet file

final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
env.setParallelism(1);
env.enableCheckpointing(100);

final List<Datum> data = Arrays.asList(new Datum("a", 1), new Datum("b", 2), new Datum("c", 3));

DataStream<Datum> stream = env.addSource(new FiniteTestSource<>(data), TypeInformation.of(Datum.class));


stream.addSink(
    StreamingFileSink.forBulkFormat(
        Path.fromLocalFile(new File("path")),
        ParquetAvroWriters.forReflectRecord(String.class))
        .build());
env.execute();

I have created a serializable class

public static class Datum implements Serializable {

        public String a;
        public int b;

        public Datum() {
        }

        public Datum(String a, int b) {
            this.a = a;
            this.b = b;
        }

        @Override
        public boolean equals(Object o) {
            if (this == o) {
                return true;
            }
            if (o == null || getClass() != o.getClass()) {
                return false;
            }

            Datum datum = (Datum) o;
            return b == datum.b && (a != null ? a.equals(datum.a) : datum.a == null);
        }

        @Override
        public int hashCode() {
            int result = a != null ? a.hashCode() : 0;
            result = 31 * result + b;
            return result;
        }
    }

The above code is not writing any data to file, it just keeps on creating many files.

If anyone can help with proper documentation or code

Answer 1

As written on the documentation of StreamingFileSink :

IMPORTANT: Checkpointing needs to be enabled when using the StreamingFileSink. Part files can only be finalized on successful checkpoints. If checkpointing is disabled part files will forever stay in in-progress or pending state and cannot be safely read by downstream systems.

To enable, just use

env.enableCheckpointing(1000);

You have quite a few options to tweak it.

---

Here is a complete example

final List<Address> data = Arrays.asList(
    new Address(1, "a", "b", "c", "12345"),
    new Address(2, "p", "q", "r", "12345"),
    new Address(3, "x", "y", "z", "12345")
);

final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
env.setParallelism(1);
env.enableCheckpointing(100);

DataStream<Address> stream = env.addSource(
    new FiniteTestSource<>(data), TypeInformation.of(Address.class));

stream.addSink(
    StreamingFileSink.forBulkFormat(
        Path.fromLocalFile(folder),
        ParquetAvroWriters.forSpecificRecord(Address.class))
        .build());

env.execute();