org.apache.hadoop.hdfs.client

Class HdfsDataOutputStream

java.lang.Object

java.io.OutputStream

java.io.FilterOutputStream

java.io.DataOutputStream

org.apache.hadoop.fs.FSDataOutputStream

org.apache.hadoop.hdfs.client.HdfsDataOutputStream

All Implemented Interfaces:

Closeable, DataOutput, Flushable, AutoCloseable, Abortable, CanSetDropBehind, org.apache.hadoop.fs.statistics.IOStatisticsSource, StreamCapabilities, Syncable

@InterfaceAudience.Public
 @InterfaceStability.Evolving
public class HdfsDataOutputStream
extends FSDataOutputStream

The Hdfs implementation of FSDataOutputStream.

Field Summary

Fields inherited from class java.io.DataOutputStream

written

Fields inherited from class java.io.FilterOutputStream

out

Fields inherited from interface org.apache.hadoop.fs.StreamCapabilities

ABORTABLE_STREAM, DROPBEHIND, HFLUSH, HSYNC, IOSTATISTICS, IOSTATISTICS_CONTEXT, PREADBYTEBUFFER, READAHEAD, READBYTEBUFFER, UNBUFFER, VECTOREDIO

Constructor Summary

Constructors

Constructor and Description
HdfsDataOutputStream(org.apache.hadoop.crypto.CryptoOutputStream out, org.apache.hadoop.fs.FileSystem.Statistics stats)
HdfsDataOutputStream(org.apache.hadoop.crypto.CryptoOutputStream out, org.apache.hadoop.fs.FileSystem.Statistics stats, long startPosition)
HdfsDataOutputStream(org.apache.hadoop.hdfs.DFSOutputStream out, org.apache.hadoop.fs.FileSystem.Statistics stats)
HdfsDataOutputStream(org.apache.hadoop.hdfs.DFSOutputStream out, org.apache.hadoop.fs.FileSystem.Statistics stats, long startPosition)

Method Summary

Modifier and Type	Method and Description
int	getCurrentBlockReplication() Get the actual number of replicas of the current block.
void	hsync(EnumSet<org.apache.hadoop.hdfs.client.HdfsDataOutputStream.SyncFlag> syncFlags) Sync buffered data to DataNodes (flush to disk devices).

Methods inherited from class org.apache.hadoop.fs.FSDataOutputStream

abort, close, getIOStatistics, getPos, hasCapability, hflush, hsync, setDropBehind, toString

Methods inherited from class java.io.DataOutputStream

flush, size, write, write, writeBoolean, writeByte, writeBytes, writeChar, writeChars, writeDouble, writeFloat, writeInt, writeLong, writeShort, writeUTF

Methods inherited from class java.io.FilterOutputStream

write

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.io.DataOutput

write

Constructor Detail

HdfsDataOutputStream

public HdfsDataOutputStream(org.apache.hadoop.hdfs.DFSOutputStream out,
                            org.apache.hadoop.fs.FileSystem.Statistics stats,
                            long startPosition)
                     throws IOException

Throws:

IOException

HdfsDataOutputStream

public HdfsDataOutputStream(org.apache.hadoop.hdfs.DFSOutputStream out,
                            org.apache.hadoop.fs.FileSystem.Statistics stats)
                     throws IOException

Throws:

IOException

HdfsDataOutputStream

public HdfsDataOutputStream(org.apache.hadoop.crypto.CryptoOutputStream out,
                            org.apache.hadoop.fs.FileSystem.Statistics stats,
                            long startPosition)
                     throws IOException

Throws:

IOException

HdfsDataOutputStream

public HdfsDataOutputStream(org.apache.hadoop.crypto.CryptoOutputStream out,
                            org.apache.hadoop.fs.FileSystem.Statistics stats)
                     throws IOException

Throws:

IOException

Method Detail

getCurrentBlockReplication

public int getCurrentBlockReplication()
                               throws IOException

Get the actual number of replicas of the current block. This can be different from the designated replication factor of the file because the namenode does not maintain replication for the blocks which are currently being written to. Depending on the configuration, the client may continue to write to a block even if a few datanodes in the write pipeline have failed, or the client may add a new datanodes once a datanode has failed.

Returns:

the number of valid replicas of the current block

Throws:

IOException

hsync

public void hsync(EnumSet<org.apache.hadoop.hdfs.client.HdfsDataOutputStream.SyncFlag> syncFlags)
           throws IOException

Sync buffered data to DataNodes (flush to disk devices).

Parameters:

syncFlags - Indicate the detailed semantic and actions of the hsync.

Throws:

IOException

See Also:

FSDataOutputStream.hsync()