Scala 3/scala/scala.util/scala.util.hashing/MurmurHash3

MurmurHash3

scala.util.hashing.MurmurHash3

An implementation of Austin Appleby's MurmurHash 3 algorithm (MurmurHash3_x86_32). This object contains methods that hash values of various types as well as means to construct Hashing objects.

This algorithm is designed to generate well-distributed non-cryptographic hashes. It is designed to hash data in 32 bit chunks (ints).

The mix method needs to be called at each step to update the intermediate hash value. For the last chunk to incorporate into the hash mixLast may be used instead, which is slightly faster. Finally finalizeHash needs to be called to compute the final hash value.

This is based on the earlier MurmurHash3 code by Rex Kerr, but the MurmurHash3 algorithm was since changed by its creator Austin Appleby to remedy some weaknesses and improve performance. This represents the latest and supposedly final version of the algorithm (revision 136). Even so, test the generated hashes in between Scala versions, even for point releases, as fast, non-cryptographic hashing algorithms evolve rapidly.

Attributes

See also: https://github.com/aappleby/smhasher
Source: MurmurHash3.scala
Graph
Supertypes: class Object

trait Matchable

class Any
Self type: MurmurHash3.type

Members list

Type members

Classlikes

Attributes

Source: MurmurHash3.scala
Supertypes: trait Hashing[Array[T]]

trait Serializable

class Object

trait Matchable

class Any

Value members

Concrete methods

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Computes the hashCode of a case class instance. This method returns the same value as x.hashCode if x is an instance of a case class with the default, synthetic hashCode.

This method can be used to implement case classes with a cached hashCode:

case class C(data: Data) {
 override lazy val hashCode: Int = MurmurHash3.caseClassHash(this)
}

NOTE: For case classes (or subclasses) that override productPrefix, the caseClassName parameter needs to be specified in order to obtain the same result as the synthetic hashCode. Otherwise, the value is not in sync with the case class equals method (scala/bug#13033).

scala> case class C(x: Int) { override def productPrefix = "Y" }

scala> C(1).hashCode
val res0: Int = -668012062

scala> MurmurHash3.caseClassHash(C(1))
val res1: Int = 1015658380

scala> MurmurHash3.caseClassHash(C(1), "C")
val res2: Int = -668012062

Value parameters

caseClassName: the case class name to use for hashing, or null to use x.productPrefix
x: the case class instance to hash

Attributes

Returns: the hash code, equivalent to the synthetic hashCode when caseClassName matches the declared class name
Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

To offer some potential for optimization.

Value parameters

xs: the sequence to hash (dispatches to specialized implementations for IndexedSeq and List)

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Attributes

Source: MurmurHash3.scala

Deprecated methods

Attributes

Deprecated: [Since version 2.13.17] use `caseClassHash` instead
Source: MurmurHash3.scala

Attributes

Deprecated: [Since version 2.13.17] use `caseClassHashing` instead
Source: MurmurHash3.scala

Inherited methods

Computes the hash of an array. Potential range hashes are recognized to produce a hash that is compatible with rangeHash.

Value parameters

a: the array to hash
seed: the initial seed for the hash computation

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes the hash of a byte array. Faster than arrayHash, because it hashes 4 bytes at once. Note that the result is not compatible with arrayHash!

Value parameters

data: the byte array to hash
seed: the initial seed for the hash computation

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

See the MurmurHash3.caseClassHash(x:Product,caseClassName:String) overload.

Value parameters

caseClassName: the case class name used for hashing, or null to fall back to x.productPrefix
seed: the initial seed for the hash computation
x: the case class instance to hash

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Finalize a hash to incorporate the length and make sure all bits avalanche.

Value parameters

hash: the intermediate hash value after all mix steps
length: the number of elements hashed

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes the hash of an IndexedSeq. Potential range hashes are recognized to produce a hash that is compatible with rangeHash.

Value parameters

a: the indexed sequence to hash
seed: the initial seed for the hash computation

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes the hash of a List. Potential range hashes are recognized to produce a hash that is compatible with rangeHash.

Value parameters

seed: the initial seed for the hash computation
xs: the list to hash

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Mix in a block of data into an intermediate hash value.

Value parameters

data: the new block of data to mix in
hash: the intermediate hash value

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

May optionally be used as the last mixing step. Is a little bit faster than mix, as it does no further mixing of the resulting hash. For the last element this is not necessary as the hash is thoroughly mixed during finalization anyway.

Value parameters

data: the last block of data to mix in
hash: the intermediate hash value

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes a hash that depends on the order of its arguments. Potential range hashes are recognized to produce a hash that is compatible with rangeHash.

Value parameters

seed: the initial seed for the hash computation
xs: the elements to hash in traversal order

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes the hash of a Range with at least 2 elements. Ranges with fewer elements need to use seqHash instead. The last parameter must be the actual last element produced by a Range, not the nominal end.

Value parameters

last: the actual last element produced by the range
seed: the initial seed for the hash computation
start: the first element of the range
step: the increment between successive elements

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes the hash of a string.

Value parameters

seed: the initial seed for the hash computation
str: the string to hash

Attributes

Inherited from:: MurmurHash3 (hidden)
Source: MurmurHash3.scala

Computes a hash that is symmetric in its arguments - that is a hash where the order of appearance of elements does not matter. This is useful for hashing sets, for example.