com.asakusafw.runtime.directio.hadoop

Class StripedDataFormat.InputContext

java.lang.Object

com.asakusafw.runtime.directio.hadoop.StripedDataFormat.InputContext

Enclosing interface:

StripedDataFormat<T>

public static class StripedDataFormat.InputContext
extends Object

An input context for computing DirectInputFragments.

Since:

0.7.0

Constructor Summary

Constructors

Constructor and Description
InputContext(Class<?> dataType, Collection<? extends org.apache.hadoop.fs.FileStatus> inputFiles, org.apache.hadoop.fs.FileSystem fileSystem, long minimumFragmentSize, long preferredFragmentSize, boolean splitBlocks, boolean combineBlocks) Creates a new instance.

Method Summary

Modifier and Type	Method and Description
Class<?>	getDataType() Returns the target data type.
org.apache.hadoop.fs.FileSystem	getFileSystem() Returns the file system for the this datastore.
Collection<? extends org.apache.hadoop.fs.FileStatus>	getInputFiles() Returns the input files information.
long	getMinimumFragmentSize() Returns the minimum fragment size.
long	getPreferredFragmentSize() Returns the preferred fragment size.
boolean	isCombineBlocks() Returns whether combines multiple blocks into a fragment for optimization.
boolean	isSplitBlocks() Returns whether split DFS block into multiple splits for optimization or not.

Methods inherited from class java.lang.Object

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

Constructor Detail

InputContext

public InputContext(Class<?> dataType,
                    Collection<? extends org.apache.hadoop.fs.FileStatus> inputFiles,
                    org.apache.hadoop.fs.FileSystem fileSystem,
                    long minimumFragmentSize,
                    long preferredFragmentSize,
                    boolean splitBlocks,
                    boolean combineBlocks)

Creates a new instance.

Parameters:

dataType - the target datatype

inputFiles - the input files information

fileSystem - the file system for input files

minimumFragmentSize - the minimum fragment size, or < 0 if fragmentation is not expected

preferredFragmentSize - the preferred fragment size, or < 0 if fragmentation is not expected

splitBlocks - true to split a file block into multiple fragments

combineBlocks - true to combine multiple splits into a single fragment

Method Detail

getDataType

public Class<?> getDataType()

Returns the target data type.

Returns:

the data type

getFileSystem

public org.apache.hadoop.fs.FileSystem getFileSystem()

Returns the file system for the this datastore.

Returns:

the file system object

getInputFiles

public Collection<? extends org.apache.hadoop.fs.FileStatus> getInputFiles()

Returns the input files information.

Returns:

the input files information

getMinimumFragmentSize

public long getMinimumFragmentSize()

Returns the minimum fragment size.

Returns:

the minimum fragment size, or < 0 if fragmentation is restricted

getPreferredFragmentSize

public long getPreferredFragmentSize()

Returns the preferred fragment size.

Returns:

the preferred fragment size, or < 0 if fragmentation is restricted

isSplitBlocks

public boolean isSplitBlocks()

Returns whether split DFS block into multiple splits for optimization or not.

Returns:

the true to split, otherwise false

isCombineBlocks

public boolean isCombineBlocks()

Returns whether combines multiple blocks into a fragment for optimization.

Returns:

the true to combine, otherwise false