Java Code Examples for org.apache.iceberg.DataFiles#Builder

The following examples show how to use org.apache.iceberg.DataFiles#Builder . You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may check out the related API usage on the sidebar.
Example 1
Source File: ParquetRecordWriter.java    From dremio-oss with Apache License 2.0 6 votes vote down vote up 
private byte[] getIcebergMetaData() throws IOException {
  if (!this.isIcebergWriter) {
    return null;
  }

  final long fileSize = parquetFileWriter.getPos();
  DataFiles.Builder dataFileBuilder =
    DataFiles.builder(IcebergCatalog.getIcebergPartitionSpec(this.batchSchema, this.partitionColumns))
      .withPath(path.toString())
      .withFileSizeInBytes(fileSize)
      .withRecordCount(recordCount)
      .withFormat(FileFormat.PARQUET);

  // add partition info
  if (partitionColumns != null) {
    dataFileBuilder = dataFileBuilder.withPartition(partition.getIcebergPartitionData());
  }

  // add column level metrics
  Metrics metrics = footerMetricsToIcebergMetrics(parquetFileWriter.getFooter(), batchSchema);
  dataFileBuilder = dataFileBuilder.withMetrics(metrics);
  return IcebergSerDe.serializeDataFile(dataFileBuilder.build());
}
Example 2
Source File: IcebergMetadata.java    From presto with Apache License 2.0 5 votes vote down vote up 
@Override
public Optional<ConnectorOutputMetadata> finishInsert(ConnectorSession session, ConnectorInsertTableHandle insertHandle, Collection<Slice> fragments, Collection<ComputedStatistics> computedStatistics)
{
    IcebergWritableTableHandle table = (IcebergWritableTableHandle) insertHandle;
    org.apache.iceberg.Table icebergTable = transaction.table();

    List<CommitTaskData> commitTasks = fragments.stream()
            .map(slice -> commitTaskCodec.fromJson(slice.getBytes()))
            .collect(toImmutableList());

    Type[] partitionColumnTypes = icebergTable.spec().fields().stream()
            .map(field -> field.transform().getResultType(
                    icebergTable.schema().findType(field.sourceId())))
            .toArray(Type[]::new);

    AppendFiles appendFiles = transaction.newFastAppend();
    for (CommitTaskData task : commitTasks) {
        HdfsContext context = new HdfsContext(session, table.getSchemaName(), table.getTableName());
        Configuration configuration = hdfsEnvironment.getConfiguration(context, new Path(task.getPath()));

        DataFiles.Builder builder = DataFiles.builder(icebergTable.spec())
                .withInputFile(HadoopInputFile.fromLocation(task.getPath(), configuration))
                .withFormat(table.getFileFormat())
                .withMetrics(task.getMetrics().metrics());

        if (!icebergTable.spec().fields().isEmpty()) {
            String partitionDataJson = task.getPartitionDataJson()
                    .orElseThrow(() -> new VerifyException("No partition data for partitioned table"));
            builder.withPartition(PartitionData.fromJson(partitionDataJson, partitionColumnTypes));
        }

        appendFiles.appendFile(builder.build());
    }

    appendFiles.commit();
    transaction.commitTransaction();

    return Optional.of(new HiveWrittenPartitions(commitTasks.stream()
            .map(CommitTaskData::getPath)
            .collect(toImmutableList())));
}
Example 3
Source File: TestIcebergInputFormat.java    From iceberg with Apache License 2.0 4 votes vote down vote up 
private DataFile writeFile(
    Table table, StructLike partitionData, FileFormat fileFormat, List<Record> records) throws IOException {
  File file = temp.newFile();
  Assert.assertTrue(file.delete());
  FileAppender<Record> appender;
  switch (fileFormat) {
    case AVRO:
      appender = Avro.write(Files.localOutput(file))
          .schema(table.schema())
          .createWriterFunc(DataWriter::create)
          .named(fileFormat.name())
          .build();
      break;
    case PARQUET:
      appender = Parquet.write(Files.localOutput(file))
          .schema(table.schema())
          .createWriterFunc(GenericParquetWriter::buildWriter)
          .named(fileFormat.name())
          .build();
      break;
    case ORC:
      appender = ORC.write(Files.localOutput(file))
          .schema(table.schema())
          .createWriterFunc(GenericOrcWriter::buildWriter)
          .build();
      break;
    default:
      throw new UnsupportedOperationException("Cannot write format: " + fileFormat);
  }

  try {
    appender.addAll(records);
  } finally {
    appender.close();
  }

  DataFiles.Builder builder = DataFiles.builder(table.spec())
      .withPath(file.toString())
      .withFormat(format)
      .withFileSizeInBytes(file.length())
      .withMetrics(appender.metrics());
  if (partitionData != null) {
    builder.withPartition(partitionData);
  }
  return builder.build();
}