Package org.apache.hadoop.hbase.io.hfile

Class BlockCacheKey

java.lang.Object

org.apache.hadoop.hbase.io.hfile.BlockCacheKey

All Implemented Interfaces:

Serializable, HeapSize

@Private
public class BlockCacheKey
extends Object
implements HeapSize, Serializable

Cache Key for use with implementations of BlockCache

See Also:

• Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
private final boolean	archived
private BlockType	blockType
private final int	cfId
static final long	FIXED_OVERHEAD
private final int	hfileNameId
private final boolean	isPrimaryReplicaBlock
private final long	offset
private final int	regionId
private static final long	serialVersionUID
private final FilePathStringPool	stringPool

Constructor Summary

Constructors

Constructor	Description
BlockCacheKey(String hfileName, long offset)	Constructs a new BlockCacheKey with the file name and offset only.
BlockCacheKey(String hfileName, long offset, boolean isPrimaryReplica, BlockType blockType)	Constructs a new BlockCacheKey with the file name, offset, replica and type only.
BlockCacheKey(String hfileName, String cfName, String regionName, long offset, boolean isPrimaryReplica, BlockType blockType, boolean archived)	Construct a new BlockCacheKey, with file, column family and region information.
BlockCacheKey(org.apache.hadoop.fs.Path hfilePath, long offset, boolean isPrimaryReplica, BlockType blockType)	Construct a new BlockCacheKey using a file path.

Method Summary

Modifier and Type	Method	Description
boolean	equals(Object o)
BlockType	getBlockType()
String	getCfName()	Returns the column family name portion of this cache key.
String	getHfileName()	Returns the hfileName portion of this cache key.
long	getOffset()
String	getRegionName()	Returns the region name portion of this cache key.
int	hashCode()
long	heapSize()	With the compressed format using integer file IDs, the heap size is significantly reduced.
boolean	isArchived()
boolean	isPrimary()
void	setBlockType(BlockType blockType)
String	toString()

Methods inherited from class java.lang.Object

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

Field Details

serialVersionUID

private static final long serialVersionUID

See Also:

• Constant Field Values

hfileNameId

private final int hfileNameId

stringPool

private final transient FilePathStringPool stringPool

regionId

private final int regionId

cfId

private final int cfId

offset

private final long offset

blockType

private BlockType blockType

isPrimaryReplicaBlock

private final boolean isPrimaryReplicaBlock

archived

private final boolean archived

FIXED_OVERHEAD

public static final long FIXED_OVERHEAD

Constructor Details

BlockCacheKey

public BlockCacheKey(String hfileName,
 long offset)

Constructs a new BlockCacheKey with the file name and offset only. To be used for cache lookups only, DO NOT use this for creating keys when inserting into the cache. Use either the overriding constructors with the path parameter or the region and cf parameters, otherwise, region cache metrics won't be recorded properly.

Parameters:

hfileName - The name of the HFile this block belongs to.

offset - Offset of the block into the file

BlockCacheKey

public BlockCacheKey(String hfileName,
 long offset,
 boolean isPrimaryReplica,
 BlockType blockType)

Constructs a new BlockCacheKey with the file name, offset, replica and type only. To be used for cache lookups only, DO NOT use this for creating keys when inserting into the cache. Use either the overriding constructors with the path parameter or the region and cf parameters, otherwise, region cache metrics won't be recorded properly.

Parameters:

hfileName - The name of the HFile this block belongs to.

offset - Offset of the block into the file

isPrimaryReplica - Whether this is from primary replica

blockType - Type of block

BlockCacheKey

public BlockCacheKey(String hfileName,
 String cfName,
 String regionName,
 long offset,
 boolean isPrimaryReplica,
 BlockType blockType,
 boolean archived)

Construct a new BlockCacheKey, with file, column family and region information. This should be used when inserting keys into the cache, so that region cache metrics are recorded properly.

Parameters:

hfileName - The name of the HFile this block belongs to.

cfName - The column family name

regionName - The region name

offset - Offset of the block into the file

isPrimaryReplica - Whether this is from primary replica

blockType - Type of block

BlockCacheKey

public BlockCacheKey(org.apache.hadoop.fs.Path hfilePath,
 long offset,
 boolean isPrimaryReplica,
 BlockType blockType)

Construct a new BlockCacheKey using a file path. File, column family and region information will be extracted from the passed path. This should be used when inserting keys into the cache, so that region cache metrics are recorded properly.

Parameters:

hfilePath - The path to the HFile

offset - Offset of the block into the file

isPrimaryReplica - Whether this is from primary replica

blockType - Type of block

Method Details

hashCode

public int hashCode()

Overrides:

hashCode in class Object

equals

public boolean equals(Object o)

Overrides:

equals in class Object

toString

public String toString()

Overrides:

toString in class Object

heapSize

public long heapSize()

With the compressed format using integer file IDs, the heap size is significantly reduced. We now only store a 4-byte integer instead of the full file name string.

Specified by:

heapSize in interface HeapSize

getHfileName

public String getHfileName()

Returns the hfileName portion of this cache key.

Returns:

The file name

getRegionName

public String getRegionName()

Returns the region name portion of this cache key.

Returns:

The region name

getCfName

public String getCfName()

Returns the column family name portion of this cache key.

Returns:

The column family name

isPrimary

public boolean isPrimary()

getOffset

public long getOffset()

getBlockType

public BlockType getBlockType()

setBlockType

public void setBlockType(BlockType blockType)

isArchived

public boolean isArchived()