com.google.common.hash.Hashing Maven / Gradle / Ivy
Show all versions of guava Show documentation
/*
* Copyright (C) 2011 The Guava Authors
*
* Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
* in compliance with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software distributed under the License
* is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
* or implied. See the License for the specific language governing permissions and limitations under
* the License.
*/
package com.google.common.hash;
import static com.google.common.base.Preconditions.checkArgument;
import static com.google.common.base.Preconditions.checkNotNull;
import static com.google.common.base.Throwables.throwIfUnchecked;
import static java.lang.invoke.MethodType.methodType;
import com.google.errorprone.annotations.Immutable;
import com.google.j2objc.annotations.J2ObjCIncompatible;
import java.lang.invoke.MethodHandle;
import java.lang.invoke.MethodHandles;
import java.security.Key;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.Iterator;
import java.util.List;
import java.util.zip.Adler32;
import java.util.zip.CRC32;
import java.util.zip.Checksum;
import javax.annotation.CheckForNull;
import javax.crypto.spec.SecretKeySpec;
/**
* Static methods to obtain {@link HashFunction} instances, and other static hashing-related
* utilities.
*
* A comparison of the various hash functions can be found here.
*
* @author Kevin Bourrillion
* @author Dimitris Andreou
* @author Kurt Alfred Kluever
* @since 11.0
*/
@ElementTypesAreNonnullByDefault
public final class Hashing {
/**
* Returns a general-purpose, temporary-use, non-cryptographic hash function. The algorithm
* the returned function implements is unspecified and subject to change without notice.
*
*
Warning: a new random seed for these functions is chosen each time the {@code
* Hashing} class is loaded. Do not use this method if hash codes may escape the current
* process in any way, for example being sent over RPC, or saved to disk. For a general-purpose,
* non-cryptographic hash function that will never change behavior, we suggest {@link
* #murmur3_128}.
*
*
Repeated calls to this method on the same loaded {@code Hashing} class, using the same value
* for {@code minimumBits}, will return identically-behaving {@link HashFunction} instances.
*
* @param minimumBits a positive integer. This can be arbitrarily large. The returned {@link
* HashFunction} instance may use memory proportional to this integer.
* @return a hash function, described above, that produces hash codes of length {@code
* minimumBits} or greater
*/
public static HashFunction goodFastHash(int minimumBits) {
int bits = checkPositiveAndMakeMultipleOf32(minimumBits);
if (bits == 32) {
return Murmur3_32HashFunction.GOOD_FAST_HASH_32;
}
if (bits <= 128) {
return Murmur3_128HashFunction.GOOD_FAST_HASH_128;
}
// Otherwise, join together some 128-bit murmur3s
int hashFunctionsNeeded = (bits + 127) / 128;
HashFunction[] hashFunctions = new HashFunction[hashFunctionsNeeded];
hashFunctions[0] = Murmur3_128HashFunction.GOOD_FAST_HASH_128;
int seed = GOOD_FAST_HASH_SEED;
for (int i = 1; i < hashFunctionsNeeded; i++) {
seed += 1500450271; // a prime; shouldn't matter
hashFunctions[i] = murmur3_128(seed);
}
return new ConcatenatedHashFunction(hashFunctions);
}
/**
* Used to randomize {@link #goodFastHash} instances, so that programs which persist anything
* dependent on the hash codes they produce will fail sooner.
*/
@SuppressWarnings("GoodTime") // reading system time without TimeSource
static final int GOOD_FAST_HASH_SEED = (int) System.currentTimeMillis();
/**
* Returns a hash function implementing the 32-bit murmur3
* algorithm, x86 variant (little-endian variant), using the given seed value, with a known
* bug as described in the deprecation text.
*
*
The C++ equivalent is the MurmurHash3_x86_32 function (Murmur3A), which however does not
* have the bug.
*
* @deprecated This implementation produces incorrect hash values from the {@link
* HashFunction#hashString} method if the string contains non-BMP characters. Use {@link
* #murmur3_32_fixed(int)} instead.
*/
@Deprecated
public static HashFunction murmur3_32(int seed) {
return new Murmur3_32HashFunction(seed, /* supplementaryPlaneFix= */ false);
}
/**
* Returns a hash function implementing the 32-bit murmur3
* algorithm, x86 variant (little-endian variant), using the given seed value, with a known
* bug as described in the deprecation text.
*
*
The C++ equivalent is the MurmurHash3_x86_32 function (Murmur3A), which however does not
* have the bug.
*
* @deprecated This implementation produces incorrect hash values from the {@link
* HashFunction#hashString} method if the string contains non-BMP characters. Use {@link
* #murmur3_32_fixed()} instead.
*/
@Deprecated
public static HashFunction murmur3_32() {
return Murmur3_32HashFunction.MURMUR3_32;
}
/**
* Returns a hash function implementing the 32-bit murmur3
* algorithm, x86 variant (little-endian variant), using the given seed value.
*
*
The exact C++ equivalent is the MurmurHash3_x86_32 function (Murmur3A).
*
*
This method is called {@code murmur3_32_fixed} because it fixes a bug in the {@code
* HashFunction} returned by the original {@code murmur3_32} method.
*
* @since 31.0
*/
public static HashFunction murmur3_32_fixed(int seed) {
return new Murmur3_32HashFunction(seed, /* supplementaryPlaneFix= */ true);
}
/**
* Returns a hash function implementing the 32-bit murmur3
* algorithm, x86 variant (little-endian variant), using a seed value of zero.
*
*
The exact C++ equivalent is the MurmurHash3_x86_32 function (Murmur3A).
*
*
This method is called {@code murmur3_32_fixed} because it fixes a bug in the {@code
* HashFunction} returned by the original {@code murmur3_32} method.
*
* @since 31.0
*/
public static HashFunction murmur3_32_fixed() {
return Murmur3_32HashFunction.MURMUR3_32_FIXED;
}
/**
* Returns a hash function implementing the 128-bit murmur3
* algorithm, x64 variant (little-endian variant), using the given seed value.
*
*
The exact C++ equivalent is the MurmurHash3_x64_128 function (Murmur3F).
*/
public static HashFunction murmur3_128(int seed) {
return new Murmur3_128HashFunction(seed);
}
/**
* Returns a hash function implementing the 128-bit murmur3
* algorithm, x64 variant (little-endian variant), using a seed value of zero.
*
*
The exact C++ equivalent is the MurmurHash3_x64_128 function (Murmur3F).
*/
public static HashFunction murmur3_128() {
return Murmur3_128HashFunction.MURMUR3_128;
}
/**
* Returns a hash function implementing the 64-bit
* SipHash-2-4 algorithm using a seed value of {@code k = 00 01 02 ...}.
*
* @since 15.0
*/
public static HashFunction sipHash24() {
return SipHashFunction.SIP_HASH_24;
}
/**
* Returns a hash function implementing the 64-bit
* SipHash-2-4 algorithm using the given seed.
*
* @since 15.0
*/
public static HashFunction sipHash24(long k0, long k1) {
return new SipHashFunction(2, 4, k0, k1);
}
/**
* Returns a hash function implementing the MD5 hash algorithm (128 hash bits).
*
* @deprecated If you must interoperate with a system that requires MD5, then use this method,
* despite its deprecation. But if you can choose your hash function, avoid MD5, which is
* neither fast nor secure. As of January 2017, we suggest:
*
* - For security:
* {@link Hashing#sha256} or a higher-level API.
*
- For speed: {@link Hashing#goodFastHash}, though see its docs for caveats.
*
*/
@Deprecated
public static HashFunction md5() {
return Md5Holder.MD5;
}
private static class Md5Holder {
static final HashFunction MD5 = new MessageDigestHashFunction("MD5", "Hashing.md5()");
}
/**
* Returns a hash function implementing the SHA-1 algorithm (160 hash bits).
*
* @deprecated If you must interoperate with a system that requires SHA-1, then use this method,
* despite its deprecation. But if you can choose your hash function, avoid SHA-1, which is
* neither fast nor secure. As of January 2017, we suggest:
*
* - For security:
* {@link Hashing#sha256} or a higher-level API.
*
- For speed: {@link Hashing#goodFastHash}, though see its docs for caveats.
*
*/
@Deprecated
public static HashFunction sha1() {
return Sha1Holder.SHA_1;
}
private static class Sha1Holder {
static final HashFunction SHA_1 = new MessageDigestHashFunction("SHA-1", "Hashing.sha1()");
}
/** Returns a hash function implementing the SHA-256 algorithm (256 hash bits). */
public static HashFunction sha256() {
return Sha256Holder.SHA_256;
}
private static class Sha256Holder {
static final HashFunction SHA_256 =
new MessageDigestHashFunction("SHA-256", "Hashing.sha256()");
}
/**
* Returns a hash function implementing the SHA-384 algorithm (384 hash bits).
*
* @since 19.0
*/
public static HashFunction sha384() {
return Sha384Holder.SHA_384;
}
private static class Sha384Holder {
static final HashFunction SHA_384 =
new MessageDigestHashFunction("SHA-384", "Hashing.sha384()");
}
/** Returns a hash function implementing the SHA-512 algorithm (512 hash bits). */
public static HashFunction sha512() {
return Sha512Holder.SHA_512;
}
private static class Sha512Holder {
static final HashFunction SHA_512 =
new MessageDigestHashFunction("SHA-512", "Hashing.sha512()");
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* MD5 (128 hash bits) hash function and the given secret key.
*
* If you are designing a new system that needs HMAC, prefer {@link #hmacSha256} or other
* future-proof algorithms over {@code hmacMd5}.
*
* @param key the secret key
* @throws IllegalArgumentException if the given key is inappropriate for initializing this MAC
* @since 20.0
*/
public static HashFunction hmacMd5(Key key) {
return new MacHashFunction("HmacMD5", key, hmacToString("hmacMd5", key));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* MD5 (128 hash bits) hash function and a {@link SecretKeySpec} created from the given byte array
* and the MD5 algorithm.
*
*
If you are designing a new system that needs HMAC, prefer {@link #hmacSha256} or other
* future-proof algorithms over {@code hmacMd5}.
*
* @param key the key material of the secret key
* @since 20.0
*/
public static HashFunction hmacMd5(byte[] key) {
return hmacMd5(new SecretKeySpec(checkNotNull(key), "HmacMD5"));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* SHA-1 (160 hash bits) hash function and the given secret key.
*
* @param key the secret key
* @throws IllegalArgumentException if the given key is inappropriate for initializing this MAC
* @since 20.0
*/
public static HashFunction hmacSha1(Key key) {
return new MacHashFunction("HmacSHA1", key, hmacToString("hmacSha1", key));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* SHA-1 (160 hash bits) hash function and a {@link SecretKeySpec} created from the given byte
* array and the SHA-1 algorithm.
*
* @param key the key material of the secret key
* @since 20.0
*/
public static HashFunction hmacSha1(byte[] key) {
return hmacSha1(new SecretKeySpec(checkNotNull(key), "HmacSHA1"));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* SHA-256 (256 hash bits) hash function and the given secret key.
*
* @param key the secret key
* @throws IllegalArgumentException if the given key is inappropriate for initializing this MAC
* @since 20.0
*/
public static HashFunction hmacSha256(Key key) {
return new MacHashFunction("HmacSHA256", key, hmacToString("hmacSha256", key));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* SHA-256 (256 hash bits) hash function and a {@link SecretKeySpec} created from the given byte
* array and the SHA-256 algorithm.
*
* @param key the key material of the secret key
* @since 20.0
*/
public static HashFunction hmacSha256(byte[] key) {
return hmacSha256(new SecretKeySpec(checkNotNull(key), "HmacSHA256"));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* SHA-512 (512 hash bits) hash function and the given secret key.
*
* @param key the secret key
* @throws IllegalArgumentException if the given key is inappropriate for initializing this MAC
* @since 20.0
*/
public static HashFunction hmacSha512(Key key) {
return new MacHashFunction("HmacSHA512", key, hmacToString("hmacSha512", key));
}
/**
* Returns a hash function implementing the Message Authentication Code (MAC) algorithm, using the
* SHA-512 (512 hash bits) hash function and a {@link SecretKeySpec} created from the given byte
* array and the SHA-512 algorithm.
*
* @param key the key material of the secret key
* @since 20.0
*/
public static HashFunction hmacSha512(byte[] key) {
return hmacSha512(new SecretKeySpec(checkNotNull(key), "HmacSHA512"));
}
private static String hmacToString(String methodName, Key key) {
return "Hashing."
+ methodName
+ "(Key[algorithm="
+ key.getAlgorithm()
+ ", format="
+ key.getFormat()
+ "])";
}
/**
* Returns a hash function implementing the CRC32C checksum algorithm (32 hash bits) as described
* by RFC 3720, Section 12.1.
*
*
This function is best understood as a checksum rather than a true hash function.
*
* @since 18.0
*/
public static HashFunction crc32c() {
return Crc32CSupplier.HASH_FUNCTION;
}
@Immutable
private enum Crc32CSupplier implements ImmutableSupplier {
@J2ObjCIncompatible
JAVA_UTIL_ZIP {
@Override
public HashFunction get() {
return ChecksumType.CRC_32C.hashFunction;
}
},
ABSTRACT_HASH_FUNCTION {
@Override
public HashFunction get() {
return Crc32cHashFunction.CRC_32_C;
}
};
static final HashFunction HASH_FUNCTION = pickFunction().get();
private static Crc32CSupplier pickFunction() {
Crc32CSupplier[] functions = values();
if (functions.length == 1) {
// We're running under J2ObjC.
return functions[0];
}
// We can't refer to JAVA_UTIL_ZIP directly at compile time because of J2ObjC.
Crc32CSupplier javaUtilZip = functions[0];
try {
Class.forName("java.util.zip.CRC32C");
return javaUtilZip;
} catch (ClassNotFoundException runningUnderJava8) {
return ABSTRACT_HASH_FUNCTION;
}
}
}
/**
* Returns a hash function implementing the CRC-32 checksum algorithm (32 hash bits).
*
* To get the {@code long} value equivalent to {@link Checksum#getValue()} for a {@code
* HashCode} produced by this function, use {@link HashCode#padToLong()}.
*
*
This function is best understood as a checksum rather than a true hash function.
*
* @since 14.0
*/
public static HashFunction crc32() {
return ChecksumType.CRC_32.hashFunction;
}
/**
* Returns a hash function implementing the Adler-32 checksum algorithm (32 hash bits).
*
*
To get the {@code long} value equivalent to {@link Checksum#getValue()} for a {@code
* HashCode} produced by this function, use {@link HashCode#padToLong()}.
*
*
This function is best understood as a checksum rather than a true hash function.
*
* @since 14.0
*/
public static HashFunction adler32() {
return ChecksumType.ADLER_32.hashFunction;
}
@Immutable
enum ChecksumType implements ImmutableSupplier {
CRC_32("Hashing.crc32()") {
@Override
public Checksum get() {
return new CRC32();
}
},
@J2ObjCIncompatible
CRC_32C("Hashing.crc32c()") {
@Override
public Checksum get() {
return Crc32cMethodHandles.newCrc32c();
}
},
ADLER_32("Hashing.adler32()") {
@Override
public Checksum get() {
return new Adler32();
}
};
public final HashFunction hashFunction;
ChecksumType(String toString) {
this.hashFunction = new ChecksumHashFunction(this, 32, toString);
}
}
@J2ObjCIncompatible
@SuppressWarnings("unused")
private static final class Crc32cMethodHandles {
private static final MethodHandle CONSTRUCTOR = crc32cConstructor();
@IgnoreJRERequirement // https://github.com/mojohaus/animal-sniffer/issues/67
static Checksum newCrc32c() {
try {
return (Checksum) CONSTRUCTOR.invokeExact();
} catch (Throwable e) {
throwIfUnchecked(e);
// That constructor has no `throws` clause.
throw newLinkageError(e);
}
}
private static MethodHandle crc32cConstructor() {
try {
Class clazz = Class.forName("java.util.zip.CRC32C");
/*
* We can't cast to CRC32C at the call site because we support building with Java 8
* (https://github.com/google/guava/issues/6549). So we have to use asType() to change from
* CRC32C to Checksum. This may carry some performance cost
* (https://stackoverflow.com/a/22321671/28465), but I'd have to benchmark more carefully to
* even detect it.
*/
return MethodHandles.lookup()
.findConstructor(clazz, methodType(void.class))
.asType(methodType(Checksum.class));
} catch (ClassNotFoundException e) {
// We check that the class is available before calling this method.
throw new AssertionError(e);
} catch (IllegalAccessException e) {
// That API is public.
throw newLinkageError(e);
} catch (NoSuchMethodException e) {
// That constructor exists.
throw newLinkageError(e);
}
}
private static LinkageError newLinkageError(Throwable cause) {
return new LinkageError(cause.toString(), cause);
}
}
/**
* Returns a hash function implementing FarmHash's Fingerprint64, an open-source algorithm.
*
* This is designed for generating persistent fingerprints of strings. It isn't
* cryptographically secure, but it produces a high-quality hash with fewer collisions than some
* alternatives we've used in the past.
*
*
FarmHash fingerprints are encoded by {@link HashCode#asBytes} in little-endian order. This
* means {@link HashCode#asLong} is guaranteed to return the same value that
* farmhash::Fingerprint64() would for the same input (when compared using {@link
* com.google.common.primitives.UnsignedLongs}'s encoding of 64-bit unsigned numbers).
*
*
This function is best understood as a fingerprint rather than a true
* hash function.
*
* @since 20.0
*/
public static HashFunction farmHashFingerprint64() {
return FarmHashFingerprint64.FARMHASH_FINGERPRINT_64;
}
/**
* Returns a hash function implementing the Fingerprint2011 hashing function (64 hash bits).
*
*
This is designed for generating persistent fingerprints of strings. It isn't
* cryptographically secure, but it produces a high-quality hash with few collisions. Fingerprints
* generated using this are byte-wise identical to those created using the C++ version, but note
* that this uses unsigned integers (see {@link com.google.common.primitives.UnsignedInts}).
* Comparisons between the two should take this into account.
*
*
Fingerprint2011() is a form of Murmur2 on strings up to 32 bytes and a form of CityHash for
* longer strings. It could have been one or the other throughout. The main advantage of the
* combination is that CityHash has a bunch of special cases for short strings that don't need to
* be replicated here. The result will never be 0 or 1.
*
*
This function is best understood as a fingerprint rather than a true
* hash function.
*
* @since 31.1
*/
public static HashFunction fingerprint2011() {
return Fingerprint2011.FINGERPRINT_2011;
}
/**
* Assigns to {@code hashCode} a "bucket" in the range {@code [0, buckets)}, in a uniform manner
* that minimizes the need for remapping as {@code buckets} grows. That is, {@code
* consistentHash(h, n)} equals:
*
*
* - {@code n - 1}, with approximate probability {@code 1/n}
*
- {@code consistentHash(h, n - 1)}, otherwise (probability {@code 1 - 1/n})
*
*
* This method is suitable for the common use case of dividing work among buckets that meet the
* following conditions:
*
*
* - You want to assign the same fraction of inputs to each bucket.
*
- When you reduce the number of buckets, you can accept that the most recently added
* buckets will be removed first. More concretely, if you are dividing traffic among tasks,
* you can decrease the number of tasks from 15 and 10, killing off the final 5 tasks, and
* {@code consistentHash} will handle it. If, however, you are dividing traffic among
* servers {@code alpha}, {@code bravo}, and {@code charlie} and you occasionally need to
* take each of the servers offline, {@code consistentHash} will be a poor fit: It provides
* no way for you to specify which of the three buckets is disappearing. Thus, if your
* buckets change from {@code [alpha, bravo, charlie]} to {@code [bravo, charlie]}, it will
* assign all the old {@code alpha} traffic to {@code bravo} and all the old {@code bravo}
* traffic to {@code charlie}, rather than letting {@code bravo} keep its traffic.
*
*
* See the Wikipedia article on
* consistent hashing for more information.
*/
public static int consistentHash(HashCode hashCode, int buckets) {
return consistentHash(hashCode.padToLong(), buckets);
}
/**
* Assigns to {@code input} a "bucket" in the range {@code [0, buckets)}, in a uniform manner that
* minimizes the need for remapping as {@code buckets} grows. That is, {@code consistentHash(h,
* n)} equals:
*
*
* - {@code n - 1}, with approximate probability {@code 1/n}
*
- {@code consistentHash(h, n - 1)}, otherwise (probability {@code 1 - 1/n})
*
*
* This method is suitable for the common use case of dividing work among buckets that meet the
* following conditions:
*
*
* - You want to assign the same fraction of inputs to each bucket.
*
- When you reduce the number of buckets, you can accept that the most recently added
* buckets will be removed first. More concretely, if you are dividing traffic among tasks,
* you can decrease the number of tasks from 15 and 10, killing off the final 5 tasks, and
* {@code consistentHash} will handle it. If, however, you are dividing traffic among
* servers {@code alpha}, {@code bravo}, and {@code charlie} and you occasionally need to
* take each of the servers offline, {@code consistentHash} will be a poor fit: It provides
* no way for you to specify which of the three buckets is disappearing. Thus, if your
* buckets change from {@code [alpha, bravo, charlie]} to {@code [bravo, charlie]}, it will
* assign all the old {@code alpha} traffic to {@code bravo} and all the old {@code bravo}
* traffic to {@code charlie}, rather than letting {@code bravo} keep its traffic.
*
*
* See the Wikipedia article on
* consistent hashing for more information.
*/
public static int consistentHash(long input, int buckets) {
checkArgument(buckets > 0, "buckets must be positive: %s", buckets);
LinearCongruentialGenerator generator = new LinearCongruentialGenerator(input);
int candidate = 0;
int next;
// Jump from bucket to bucket until we go out of range
while (true) {
next = (int) ((candidate + 1) / generator.nextDouble());
if (next >= 0 && next < buckets) {
candidate = next;
} else {
return candidate;
}
}
}
/**
* Returns a hash code, having the same bit length as each of the input hash codes, that combines
* the information of these hash codes in an ordered fashion. That is, whenever two equal hash
* codes are produced by two calls to this method, it is as likely as possible that each
* was computed from the same input hash codes in the same order.
*
* @throws IllegalArgumentException if {@code hashCodes} is empty, or the hash codes do not all
* have the same bit length
*/
public static HashCode combineOrdered(Iterable hashCodes) {
Iterator iterator = hashCodes.iterator();
checkArgument(iterator.hasNext(), "Must be at least 1 hash code to combine.");
int bits = iterator.next().bits();
byte[] resultBytes = new byte[bits / 8];
for (HashCode hashCode : hashCodes) {
byte[] nextBytes = hashCode.asBytes();
checkArgument(
nextBytes.length == resultBytes.length, "All hashcodes must have the same bit length.");
for (int i = 0; i < nextBytes.length; i++) {
resultBytes[i] = (byte) (resultBytes[i] * 37 ^ nextBytes[i]);
}
}
return HashCode.fromBytesNoCopy(resultBytes);
}
/**
* Returns a hash code, having the same bit length as each of the input hash codes, that combines
* the information of these hash codes in an unordered fashion. That is, whenever two equal hash
* codes are produced by two calls to this method, it is as likely as possible that each
* was computed from the same input hash codes in some order.
*
* @throws IllegalArgumentException if {@code hashCodes} is empty, or the hash codes do not all
* have the same bit length
*/
public static HashCode combineUnordered(Iterable hashCodes) {
Iterator iterator = hashCodes.iterator();
checkArgument(iterator.hasNext(), "Must be at least 1 hash code to combine.");
byte[] resultBytes = new byte[iterator.next().bits() / 8];
for (HashCode hashCode : hashCodes) {
byte[] nextBytes = hashCode.asBytes();
checkArgument(
nextBytes.length == resultBytes.length, "All hashcodes must have the same bit length.");
for (int i = 0; i < nextBytes.length; i++) {
resultBytes[i] += nextBytes[i];
}
}
return HashCode.fromBytesNoCopy(resultBytes);
}
/** Checks that the passed argument is positive, and ceils it to a multiple of 32. */
static int checkPositiveAndMakeMultipleOf32(int bits) {
checkArgument(bits > 0, "Number of bits must be positive");
return (bits + 31) & ~31;
}
/**
* Returns a hash function which computes its hash code by concatenating the hash codes of the
* underlying hash functions together. This can be useful if you need to generate hash codes of a
* specific length.
*
* For example, if you need 1024-bit hash codes, you could join two {@link Hashing#sha512} hash
* functions together: {@code Hashing.concatenating(Hashing.sha512(), Hashing.sha512())}.
*
* @since 19.0
*/
public static HashFunction concatenating(
HashFunction first, HashFunction second, HashFunction... rest) {
// We can't use Lists.asList() here because there's no hash->collect dependency
List list = new ArrayList<>();
list.add(first);
list.add(second);
Collections.addAll(list, rest);
return new ConcatenatedHashFunction(list.toArray(new HashFunction[0]));
}
/**
* Returns a hash function which computes its hash code by concatenating the hash codes of the
* underlying hash functions together. This can be useful if you need to generate hash codes of a
* specific length.
*
* For example, if you need 1024-bit hash codes, you could join two {@link Hashing#sha512} hash
* functions together: {@code Hashing.concatenating(Hashing.sha512(), Hashing.sha512())}.
*
* @since 19.0
*/
public static HashFunction concatenating(Iterable hashFunctions) {
checkNotNull(hashFunctions);
// We can't use Iterables.toArray() here because there's no hash->collect dependency
List list = new ArrayList<>();
for (HashFunction hashFunction : hashFunctions) {
list.add(hashFunction);
}
checkArgument(!list.isEmpty(), "number of hash functions (%s) must be > 0", list.size());
return new ConcatenatedHashFunction(list.toArray(new HashFunction[0]));
}
private static final class ConcatenatedHashFunction extends AbstractCompositeHashFunction {
private ConcatenatedHashFunction(HashFunction... functions) {
super(functions);
for (HashFunction function : functions) {
checkArgument(
function.bits() % 8 == 0,
"the number of bits (%s) in hashFunction (%s) must be divisible by 8",
function.bits(),
function);
}
}
@Override
HashCode makeHash(Hasher[] hashers) {
byte[] bytes = new byte[bits() / 8];
int i = 0;
for (Hasher hasher : hashers) {
HashCode newHash = hasher.hash();
i += newHash.writeBytesTo(bytes, i, newHash.bits() / 8);
}
return HashCode.fromBytesNoCopy(bytes);
}
@Override
public int bits() {
int bitSum = 0;
for (HashFunction function : functions) {
bitSum += function.bits();
}
return bitSum;
}
@Override
public boolean equals(@CheckForNull Object object) {
if (object instanceof ConcatenatedHashFunction) {
ConcatenatedHashFunction other = (ConcatenatedHashFunction) object;
return Arrays.equals(functions, other.functions);
}
return false;
}
@Override
public int hashCode() {
return Arrays.hashCode(functions);
}
}
/**
* Linear CongruentialGenerator to use for consistent hashing. See
* http://en.wikipedia.org/wiki/Linear_congruential_generator
*/
private static final class LinearCongruentialGenerator {
private long state;
public LinearCongruentialGenerator(long seed) {
this.state = seed;
}
public double nextDouble() {
state = 2862933555777941757L * state + 1;
return ((double) ((int) (state >>> 33) + 1)) / 0x1.0p31;
}
}
private Hashing() {}
}