org.elasticsearch.common.util

Class BloomFilter

java.lang.Object

org.elasticsearch.common.util.BloomFilter

public class BloomFilter
extends Object

A bloom filter. Inspired by Guava bloom filter implementation though with some optimizations.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	BloomFilter.Factory A factory that can use different fpp based on size.

Method Summary

Methods

Modifier and Type	Method and Description
static BloomFilter	create(int expectedInsertions, double fpp) Creates a bloom filter based on the with the expected number of insertions and expected false positive probability.
static BloomFilter	create(int expectedInsertions, double fpp, int numHashFunctions) Creates a bloom filter based on the expected number of insertions, expected false positive probability, and number of hash functions.
static BloomFilter	deserialize(org.apache.lucene.store.DataInput in)
protected static long	fmix(long k)
protected static long	getblock(byte[] key, int offset, int index)
int	getNumHashFunctions()
long	getSizeInBytes()
static long	hash3_x64_128(byte[] key, int offset, int length, long seed)
int	hashCode()
boolean	mightContain(org.apache.lucene.util.BytesRef value)
boolean	put(org.apache.lucene.util.BytesRef value)
static BloomFilter	readFrom(StreamInput in)
protected static long	rotl64(long v, int n)
static void	serilaize(BloomFilter filter, org.apache.lucene.store.DataOutput out)
static void	skipBloom(org.apache.lucene.store.IndexInput in)
static void	writeTo(BloomFilter filter, StreamOutput out)

Methods inherited from class java.lang.Object

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

Method Detail

create

public static BloomFilter create(int expectedInsertions,
                 double fpp)

Creates a bloom filter based on the with the expected number of insertions and expected false positive probability.

Parameters:

expectedInsertions - the number of expected insertions to the constructed

fpp - the desired false positive probability (must be positive and less than 1.0)

create

public static BloomFilter create(int expectedInsertions,
                 double fpp,
                 int numHashFunctions)

Creates a bloom filter based on the expected number of insertions, expected false positive probability, and number of hash functions.

Parameters:

expectedInsertions - the number of expected insertions to the constructed

fpp - the desired false positive probability (must be positive and less than 1.0)

numHashFunctions - the number of hash functions to use (must be less than or equal to 255)

skipBloom

public static void skipBloom(org.apache.lucene.store.IndexInput in)
                      throws IOException

Throws:

IOException

deserialize

public static BloomFilter deserialize(org.apache.lucene.store.DataInput in)
                               throws IOException

Throws:

IOException

serilaize

public static void serilaize(BloomFilter filter,
             org.apache.lucene.store.DataOutput out)
                      throws IOException

Throws:

IOException

readFrom

public static BloomFilter readFrom(StreamInput in)
                            throws IOException

Throws:

IOException

writeTo

public static void writeTo(BloomFilter filter,
           StreamOutput out)
                    throws IOException

Throws:

IOException

put

public boolean put(org.apache.lucene.util.BytesRef value)

mightContain

public boolean mightContain(org.apache.lucene.util.BytesRef value)

getNumHashFunctions

public int getNumHashFunctions()

getSizeInBytes

public long getSizeInBytes()

hashCode

public int hashCode()

Overrides:

hashCode in class Object

getblock

protected static long getblock(byte[] key,
            int offset,
            int index)

rotl64

protected static long rotl64(long v,
          int n)

fmix

protected static long fmix(long k)

hash3_x64_128

public static long hash3_x64_128(byte[] key,
                 int offset,
                 int length,
                 long seed)