Examples with RewriteAvroPayload org.apache.hudi.common.model.RewriteAvroPayload used on opensource projects

Example 1 with RewriteAvroPayload

use of org.apache.hudi.common.model.RewriteAvroPayload in project hudi by apache.

the class JavaExecutionStrategy method transform.

/**
 * Transform IndexedRecord into HoodieRecord.
 */
private HoodieRecord<T> transform(IndexedRecord indexedRecord) {
    GenericRecord record = (GenericRecord) indexedRecord;
    Option<BaseKeyGenerator> keyGeneratorOpt = Option.empty();
    String key = KeyGenUtils.getRecordKeyFromGenericRecord(record, keyGeneratorOpt);
    String partition = KeyGenUtils.getPartitionPathFromGenericRecord(record, keyGeneratorOpt);
    HoodieKey hoodieKey = new HoodieKey(key, partition);
    HoodieRecordPayload avroPayload = new RewriteAvroPayload(record);
    HoodieRecord hoodieRecord = new HoodieAvroRecord(hoodieKey, avroPayload);
    return hoodieRecord;
}

Example 2 with RewriteAvroPayload

use of org.apache.hudi.common.model.RewriteAvroPayload in project hudi by apache.

the class MultipleSparkJobExecutionStrategy method transform.

/**
 * Transform IndexedRecord into HoodieRecord.
 */
private static <T> HoodieRecord<T> transform(IndexedRecord indexedRecord, HoodieWriteConfig writeConfig) {
    GenericRecord record = (GenericRecord) indexedRecord;
    Option<BaseKeyGenerator> keyGeneratorOpt = Option.empty();
    if (!writeConfig.populateMetaFields()) {
        try {
            keyGeneratorOpt = Option.of((BaseKeyGenerator) HoodieSparkKeyGeneratorFactory.createKeyGenerator(writeConfig.getProps()));
        } catch (IOException e) {
            throw new HoodieIOException("Only BaseKeyGenerators are supported when meta columns are disabled ", e);
        }
    }
    String key = KeyGenUtils.getRecordKeyFromGenericRecord(record, keyGeneratorOpt);
    String partition = KeyGenUtils.getPartitionPathFromGenericRecord(record, keyGeneratorOpt);
    HoodieKey hoodieKey = new HoodieKey(key, partition);
    HoodieRecordPayload avroPayload = new RewriteAvroPayload(record);
    HoodieRecord hoodieRecord = new HoodieAvroRecord(hoodieKey, avroPayload);
    return hoodieRecord;
}

Example 3 with RewriteAvroPayload

use of org.apache.hudi.common.model.RewriteAvroPayload in project hudi by apache.

the class SingleSparkJobExecutionStrategy method transform.

/**
 * Transform IndexedRecord into HoodieRecord.
 */
private HoodieRecord<T> transform(IndexedRecord indexedRecord) {
    GenericRecord record = (GenericRecord) indexedRecord;
    Option<BaseKeyGenerator> keyGeneratorOpt = Option.empty();
    if (!getWriteConfig().populateMetaFields()) {
        try {
            keyGeneratorOpt = Option.of((BaseKeyGenerator) HoodieSparkKeyGeneratorFactory.createKeyGenerator(new TypedProperties(getWriteConfig().getProps())));
        } catch (IOException e) {
            throw new HoodieIOException("Only BaseKeyGenerators are supported when meta columns are disabled ", e);
        }
    }
    String key = KeyGenUtils.getRecordKeyFromGenericRecord(record, keyGeneratorOpt);
    String partition = KeyGenUtils.getPartitionPathFromGenericRecord(record, keyGeneratorOpt);
    HoodieKey hoodieKey = new HoodieKey(key, partition);
    HoodieRecordPayload avroPayload = new RewriteAvroPayload(record);
    HoodieRecord hoodieRecord = new HoodieAvroRecord(hoodieKey, avroPayload);
    return hoodieRecord;
}

Example 4 with RewriteAvroPayload

use of org.apache.hudi.common.model.RewriteAvroPayload in project hudi by apache.

the class RDDSpatialCurveSortPartitioner method repartitionRecords.

@Override
public JavaRDD<HoodieRecord<T>> repartitionRecords(JavaRDD<HoodieRecord<T>> records, int outputSparkPartitions) {
    SerializableSchema serializableSchema = new SerializableSchema(schema);
    JavaRDD<GenericRecord> genericRecordsRDD = records.map(f -> (GenericRecord) f.getData().getInsertValue(serializableSchema.get()).get());
    Dataset<Row> sourceDataset = AvroConversionUtils.createDataFrame(genericRecordsRDD.rdd(), schema.toString(), sparkEngineContext.getSqlContext().sparkSession());
    Dataset<Row> sortedDataset = reorder(sourceDataset, outputSparkPartitions);
    return HoodieSparkUtils.createRdd(sortedDataset, schema.getName(), schema.getNamespace(), false, Option.empty()).toJavaRDD().map(record -> {
        String key = record.get(HoodieRecord.RECORD_KEY_METADATA_FIELD).toString();
        String partition = record.get(HoodieRecord.PARTITION_PATH_METADATA_FIELD).toString();
        HoodieKey hoodieKey = new HoodieKey(key, partition);
        HoodieRecord hoodieRecord = new HoodieAvroRecord(hoodieKey, new RewriteAvroPayload(record));
        return hoodieRecord;
    });
}