com.facebook.presto.orc

Class OrcBatchRecordReader

java.lang.Object

com.facebook.presto.orc.OrcBatchRecordReader

All Implemented Interfaces:

Closeable, AutoCloseable

public class OrcBatchRecordReader
extends Object

Constructor Summary

Constructors

Constructor and Description

OrcBatchRecordReader(Map<Integer,Type> includedColumns, OrcPredicate predicate, long numberOfRows, List<StripeInformation> fileStripes, List<ColumnStatistics> fileStats, List<StripeStatistics> stripeStats, OrcDataSource orcDataSource, long splitOffset, long splitLength, List<OrcType> types, Optional<OrcDecompressor> decompressor, int rowsInRowGroup, org.joda.time.DateTimeZone hiveStorageTimeZone, PostScript.HiveWriterVersion hiveWriterVersion, MetadataReader metadataReader, io.airlift.units.DataSize maxMergeDistance, io.airlift.units.DataSize tinyStripeThreshold, io.airlift.units.DataSize maxBlockSize, Map<String,io.airlift.slice.Slice> userMetadata, AggregatedMemoryContext systemMemoryUsage, Optional<OrcWriteValidation> writeValidation, int initialBatchSize, StripeMetadataSource stripeMetadataSource, HiveFileContext hiveFileContext)

Method Summary

Modifier and Type	Method and Description
protected void	adjustMaxBatchSize(long averageRowBytes)
protected void	batchRead(int batchSize)
void	close()
protected static StreamDescriptor	createStreamDescriptor(String parentStreamName, String fieldName, int typeId, List<OrcType> types, OrcDataSource dataSource)
long	getFilePosition() Return the row position relative to the start of the file.
long	getFileRowCount() Returns the total number of rows in the file.
long	getMaxCombinedBytesPerRow() Returns the sum of the largest cells in size from each column
protected int	getNextRowInGroup()
long	getReaderPosition() Return the row position within the stripes being read by this reader.
long	getReaderRowCount() Returns the total number of rows that can possibly be read by this reader.
protected long	getRetainedSizeInBytes()
long	getSplitLength()
protected T[]	getStreamReaders()
Map<String,io.airlift.slice.Slice>	getUserMetadata()
boolean	isColumnPresent(int hiveColumnIndex)
int	nextBatch()
protected int	prepareNextBatch()
Block	readBlock(int columnIndex)
protected boolean	shouldValidateWritePageChecksum()
protected void	updateMaxCombinedBytesPerRow(int columnIndex, Block block)
protected void	validateWritePageChecksum(Page page)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

OrcBatchRecordReader

public OrcBatchRecordReader(Map<Integer,Type> includedColumns,
                            OrcPredicate predicate,
                            long numberOfRows,
                            List<StripeInformation> fileStripes,
                            List<ColumnStatistics> fileStats,
                            List<StripeStatistics> stripeStats,
                            OrcDataSource orcDataSource,
                            long splitOffset,
                            long splitLength,
                            List<OrcType> types,
                            Optional<OrcDecompressor> decompressor,
                            int rowsInRowGroup,
                            org.joda.time.DateTimeZone hiveStorageTimeZone,
                            PostScript.HiveWriterVersion hiveWriterVersion,
                            MetadataReader metadataReader,
                            io.airlift.units.DataSize maxMergeDistance,
                            io.airlift.units.DataSize tinyStripeThreshold,
                            io.airlift.units.DataSize maxBlockSize,
                            Map<String,io.airlift.slice.Slice> userMetadata,
                            AggregatedMemoryContext systemMemoryUsage,
                            Optional<OrcWriteValidation> writeValidation,
                            int initialBatchSize,
                            StripeMetadataSource stripeMetadataSource,
                            HiveFileContext hiveFileContext)
                     throws OrcCorruptionException

Throws:

OrcCorruptionException

Method Detail

nextBatch

public int nextBatch()
              throws IOException

Throws:

IOException

readBlock

public Block readBlock(int columnIndex)
                throws IOException

Throws:

IOException

getRetainedSizeInBytes

protected long getRetainedSizeInBytes()

Returns:

The total size of memory retained by this OrcRecordReader

getMaxCombinedBytesPerRow

public long getMaxCombinedBytesPerRow()

Returns the sum of the largest cells in size from each column

updateMaxCombinedBytesPerRow

protected void updateMaxCombinedBytesPerRow(int columnIndex,
                                            Block block)

getStreamReaders

protected T[] getStreamReaders()

getFilePosition

public long getFilePosition()

Return the row position relative to the start of the file.

getFileRowCount

public long getFileRowCount()

Returns the total number of rows in the file. This count includes rows for stripes that were completely excluded due to stripe statistics.

getReaderPosition

public long getReaderPosition()

Return the row position within the stripes being read by this reader. This position will include rows that were never read due to row groups that are excluded due to row group statistics. Thus, it will advance faster than the number of rows actually read.

getReaderRowCount

public long getReaderRowCount()

Returns the total number of rows that can possibly be read by this reader. This count may be fewer than the number of rows in the file if some stripes were excluded due to stripe statistics, but may be more than the number of rows read if some row groups are excluded due to statistics.

getSplitLength

public long getSplitLength()

close

public void close()
           throws IOException

Specified by:

close in interface Closeable

Specified by:

close in interface AutoCloseable

Throws:

IOException

isColumnPresent

public boolean isColumnPresent(int hiveColumnIndex)

getUserMetadata

public Map<String,io.airlift.slice.Slice> getUserMetadata()

getNextRowInGroup

protected int getNextRowInGroup()

batchRead

protected void batchRead(int batchSize)

adjustMaxBatchSize

protected void adjustMaxBatchSize(long averageRowBytes)

prepareNextBatch

protected int prepareNextBatch()
                        throws IOException

Throws:

IOException

createStreamDescriptor

protected static StreamDescriptor createStreamDescriptor(String parentStreamName,
                                                         String fieldName,
                                                         int typeId,
                                                         List<OrcType> types,
                                                         OrcDataSource dataSource)

shouldValidateWritePageChecksum

protected boolean shouldValidateWritePageChecksum()

validateWritePageChecksum

protected void validateWritePageChecksum(Page page)