org.apache.hadoop.hbase.mapreduce

Class TableSplit

java.lang.Object

org.apache.hadoop.mapreduce.InputSplit

org.apache.hadoop.hbase.mapreduce.TableSplit

All Implemented Interfaces:

Comparable<TableSplit>, org.apache.hadoop.io.Writable

@InterfaceAudience.Public
public class TableSplit
extends org.apache.hadoop.mapreduce.InputSplit
implements org.apache.hadoop.io.Writable, Comparable<TableSplit>

A table split corresponds to a key range (low, high) and an optional scanner. All references to row below refer to the key of the row.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
(package private) static class	TableSplit.Version

Field Summary

Fields

Modifier and Type	Field and Description
private String	encodedRegionName
private byte[]	endRow
private long	length
static org.slf4j.Logger	LOG Deprecated. LOG variable would be made private. fix in hbase 3.0
private String	regionLocation
private String	scan The scan object may be null but the serialized form of scan is never null or empty since we serialize the scan object with default values then.
private byte[]	startRow
private TableName	tableName
private static TableSplit.Version	VERSION

Constructor Summary

Constructors

Constructor and Description
TableSplit() Default constructor.
TableSplit(TableName tableName, byte[] startRow, byte[] endRow, String location) Creates a new instance without a scanner.
TableSplit(TableName tableName, byte[] startRow, byte[] endRow, String location, long length) Creates a new instance without a scanner.
TableSplit(TableName tableName, Scan scan, byte[] startRow, byte[] endRow, String location) Creates a new instance while assigning all variables.
TableSplit(TableName tableName, Scan scan, byte[] startRow, byte[] endRow, String location, long length) Creates a new instance while assigning all variables.
TableSplit(TableName tableName, Scan scan, byte[] startRow, byte[] endRow, String location, String encodedRegionName, long length) Creates a new instance while assigning all variables.

Method Summary

Modifier and Type	Method and Description
int	compareTo(TableSplit split) Compares this split against the given one.
boolean	equals(Object o)
String	getEncodedRegionName() Returns the region's encoded name.
byte[]	getEndRow() Returns the end row.
long	getLength() Returns the length of the split.
String[]	getLocations() Returns the region's location as an array.
String	getRegionLocation() Returns the region location.
Scan	getScan() Returns a Scan object from the stored string representation.
String	getScanAsString() Returns a scan string
byte[]	getStartRow() Returns the start row.
TableName	getTable() Returns the table name.
byte[]	getTableName() Returns the table name converted to a byte array.
int	hashCode()
void	readFields(DataInput in) Reads the values of each field.
String	toString() Returns the details about this instance as a string.
void	write(DataOutput out) Writes the field values to the output.

Methods inherited from class org.apache.hadoop.mapreduce.InputSplit

getLocationInfo

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Field Detail

LOG

@Deprecated
public static final org.slf4j.Logger LOG

Deprecated. LOG variable would be made private. fix in hbase 3.0

VERSION

private static final TableSplit.Version VERSION

tableName

private TableName tableName

startRow

private byte[] startRow

endRow

private byte[] endRow

regionLocation

private String regionLocation

encodedRegionName

private String encodedRegionName

scan

private String scan

The scan object may be null but the serialized form of scan is never null or empty since we serialize the scan object with default values then. Having no scanner in TableSplit doesn't necessarily mean there is no scanner for mapreduce job, it just means that we do not need to set it for each split. For example, it is not required to have a scan object for TableInputFormatBase since we use the scan from the job conf and scanner is supposed to be same for all the splits of table.

length

private long length

Constructor Detail

TableSplit

public TableSplit()

Default constructor.

TableSplit

public TableSplit(TableName tableName,
                  Scan scan,
                  byte[] startRow,
                  byte[] endRow,
                  String location)

Creates a new instance while assigning all variables. Length of region is set to 0 Encoded name of the region is set to blank

Parameters:

tableName - The name of the current table.

scan - The scan associated with this split.

startRow - The start row of the split.

endRow - The end row of the split.

location - The location of the region.

TableSplit

public TableSplit(TableName tableName,
                  Scan scan,
                  byte[] startRow,
                  byte[] endRow,
                  String location,
                  long length)

Creates a new instance while assigning all variables. Encoded name of region is set to blank

Parameters:

tableName - The name of the current table.

scan - The scan associated with this split.

startRow - The start row of the split.

endRow - The end row of the split.

location - The location of the region.

TableSplit

public TableSplit(TableName tableName,
                  Scan scan,
                  byte[] startRow,
                  byte[] endRow,
                  String location,
                  String encodedRegionName,
                  long length)

Creates a new instance while assigning all variables.

Parameters:

tableName - The name of the current table.

scan - The scan associated with this split.

startRow - The start row of the split.

endRow - The end row of the split.

encodedRegionName - The region ID.

location - The location of the region.

TableSplit

public TableSplit(TableName tableName,
                  byte[] startRow,
                  byte[] endRow,
                  String location)

Creates a new instance without a scanner. Length of region is set to 0

Parameters:

tableName - The name of the current table.

startRow - The start row of the split.

endRow - The end row of the split.

location - The location of the region.

TableSplit

public TableSplit(TableName tableName,
                  byte[] startRow,
                  byte[] endRow,
                  String location,
                  long length)

Creates a new instance without a scanner.

Parameters:

tableName - The name of the current table.

startRow - The start row of the split.

endRow - The end row of the split.

location - The location of the region.

length - Size of region in bytes

Method Detail

getScan

public Scan getScan()
             throws IOException

Returns a Scan object from the stored string representation.

Returns:

Returns a Scan object based on the stored scanner.

Throws:

IOException - throws IOException if deserialization fails

getScanAsString

@InterfaceAudience.Private
public String getScanAsString()

Returns a scan string

Returns:

scan as string. Should be noted that this is not same as getScan().toString() because Scan object will have the default values when empty scan string is deserialized. Thus, getScan().toString() can never be empty

getTableName

public byte[] getTableName()

Returns the table name converted to a byte array.

Returns:

The table name.

See Also:

getTable()

getTable

public TableName getTable()

Returns the table name.

Returns:

The table name.

getStartRow

public byte[] getStartRow()

Returns the start row.

Returns:

The start row.

getEndRow

public byte[] getEndRow()

Returns the end row.

Returns:

The end row.

getRegionLocation

public String getRegionLocation()

Returns the region location.

Returns:

The region's location.

getLocations

public String[] getLocations()

Returns the region's location as an array.

Specified by:

getLocations in class org.apache.hadoop.mapreduce.InputSplit

Returns:

The array containing the region location.

See Also:

InputSplit.getLocations()

getEncodedRegionName

public String getEncodedRegionName()

Returns the region's encoded name.

Returns:

The region's encoded name.

getLength

public long getLength()

Returns the length of the split.

Specified by:

getLength in class org.apache.hadoop.mapreduce.InputSplit

Returns:

The length of the split.

See Also:

InputSplit.getLength()

readFields

public void readFields(DataInput in)
                throws IOException

Reads the values of each field.

Specified by:

readFields in interface org.apache.hadoop.io.Writable

Parameters:

in - The input to read from.

Throws:

IOException - When reading the input fails.

write

public void write(DataOutput out)
           throws IOException

Writes the field values to the output.

Specified by:

write in interface org.apache.hadoop.io.Writable

Parameters:

out - The output to write to.

Throws:

IOException - When writing the values to the output fails.

toString

public String toString()

Returns the details about this instance as a string.

Overrides:

toString in class Object

Returns:

The values of this instance as a string.

See Also:

Object.toString()

compareTo

public int compareTo(TableSplit split)

Compares this split against the given one.

Specified by:

compareTo in interface Comparable<TableSplit>

Parameters:

split - The split to compare to.

Returns:

The result of the comparison.

See Also:

Comparable.compareTo(java.lang.Object)

equals

public boolean equals(Object o)

Overrides:

equals in class Object

hashCode

public int hashCode()

Overrides:

hashCode in class Object