org.apache.hadoop.hive.ql.io

Class IOContext

java.lang.Object

org.apache.hadoop.hive.ql.io.IOContext

public class IOContext
extends Object

IOContext basically contains the position information of the current key/value. For blockCompressed files, isBlockPointer should return true, and currentBlockStart refers to the RCFile Block or SequenceFile Block. For non compressed files, isBlockPointer should return false, and currentBlockStart refers to the beginning offset of the current row, nextBlockStart refers the end of current row and beginning of next row.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	IOContext.Comparison

Constructor Summary

Constructors

Constructor and Description
IOContext()

Method Summary

Methods

Modifier and Type	Method and Description
static void	clear()
static IOContext	get(org.apache.hadoop.conf.Configuration conf)
IOContext.Comparison	getComparison()
long	getCurrentBlockStart()
long	getCurrentRow()
String	getGenericUDFClassName()
org.apache.hadoop.fs.Path	getInputPath()
boolean	getIOExceptions()
long	getNextBlockStart()
RecordIdentifier	getRecordIdentifier()
boolean	isBinarySearching()
boolean	isBlockPointer()
void	resetSortingValues() The thread local IOContext is static, we may need to restart the search if, for instance, multiple files are being searched as part of a CombinedHiveRecordReader
void	setBinarySearching(boolean isBinarySearching)
void	setBlockPointer(boolean isBlockPointer)
void	setComparison(Integer comparison)
void	setCurrentBlockStart(long currentBlockStart)
void	setCurrentRow(long currentRow)
void	setEndBinarySearch(boolean endBinarySearch)
void	setGenericUDFClassName(String genericUDFClassName)
void	setInputPath(org.apache.hadoop.fs.Path inputPath)
void	setIOExceptions(boolean ioe)
void	setNextBlockStart(long nextBlockStart)
void	setRecordIdentifier(RecordIdentifier ri)
void	setUseSorted(boolean useSorted)
boolean	shouldEndBinarySearch()
boolean	useSorted()

Methods inherited from class java.lang.Object

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

Constructor Detail

IOContext

public IOContext()

Method Detail

get

public static IOContext get(org.apache.hadoop.conf.Configuration conf)

clear

public static void clear()

getCurrentBlockStart

public long getCurrentBlockStart()

setCurrentBlockStart

public void setCurrentBlockStart(long currentBlockStart)

getNextBlockStart

public long getNextBlockStart()

setNextBlockStart

public void setNextBlockStart(long nextBlockStart)

getCurrentRow

public long getCurrentRow()

setCurrentRow

public void setCurrentRow(long currentRow)

isBlockPointer

public boolean isBlockPointer()

setBlockPointer

public void setBlockPointer(boolean isBlockPointer)

getInputPath

public org.apache.hadoop.fs.Path getInputPath()

setInputPath

public void setInputPath(org.apache.hadoop.fs.Path inputPath)

setIOExceptions

public void setIOExceptions(boolean ioe)

getIOExceptions

public boolean getIOExceptions()

useSorted

public boolean useSorted()

setUseSorted

public void setUseSorted(boolean useSorted)

isBinarySearching

public boolean isBinarySearching()

setBinarySearching

public void setBinarySearching(boolean isBinarySearching)

shouldEndBinarySearch

public boolean shouldEndBinarySearch()

setEndBinarySearch

public void setEndBinarySearch(boolean endBinarySearch)

getComparison

public IOContext.Comparison getComparison()

setComparison

public void setComparison(Integer comparison)

getGenericUDFClassName

public String getGenericUDFClassName()

setGenericUDFClassName

public void setGenericUDFClassName(String genericUDFClassName)

getRecordIdentifier

public RecordIdentifier getRecordIdentifier()

setRecordIdentifier

public void setRecordIdentifier(RecordIdentifier ri)

resetSortingValues

public void resetSortingValues()

The thread local IOContext is static, we may need to restart the search if, for instance, multiple files are being searched as part of a CombinedHiveRecordReader