com.google.common.hash.Murmur3_32HashFunction Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of guava Show documentation
Show all versions of guava Show documentation
Guava is a suite of core and expanded libraries that include
utility classes, Google's collections, I/O classes, and
much more.
/*
* 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.
*/
/*
* MurmurHash3 was written by Austin Appleby, and is placed in the public
* domain. The author hereby disclaims copyright to this source code.
*/
/*
* Source:
* http://code.google.com/p/smhasher/source/browse/trunk/MurmurHash3.cpp
* (Modified to adapt to Guava coding conventions and to use the HashFunction interface)
*/
package com.google.common.hash;
import static com.google.common.base.Preconditions.checkPositionIndexes;
import static com.google.common.base.Preconditions.checkState;
import static com.google.common.primitives.UnsignedBytes.toInt;
import com.google.common.base.Charsets;
import com.google.common.primitives.Chars;
import com.google.common.primitives.Ints;
import com.google.common.primitives.Longs;
import com.google.errorprone.annotations.CanIgnoreReturnValue;
import com.google.errorprone.annotations.Immutable;
import java.io.Serializable;
import java.nio.ByteBuffer;
import java.nio.ByteOrder;
import java.nio.charset.Charset;
import javax.annotation.CheckForNull;
/**
* See MurmurHash3_x86_32 in the C++
* implementation.
*
* @author Austin Appleby
* @author Dimitris Andreou
* @author Kurt Alfred Kluever
*/
@Immutable
@ElementTypesAreNonnullByDefault
final class Murmur3_32HashFunction extends AbstractHashFunction implements Serializable {
static final HashFunction MURMUR3_32 =
new Murmur3_32HashFunction(0, /* supplementaryPlaneFix= */ false);
static final HashFunction MURMUR3_32_FIXED =
new Murmur3_32HashFunction(0, /* supplementaryPlaneFix= */ true);
// We can include the non-BMP fix here because Hashing.goodFastHash stresses that the hash is a
// temporary-use one. Therefore it shouldn't be persisted.
static final HashFunction GOOD_FAST_HASH_32 =
new Murmur3_32HashFunction(Hashing.GOOD_FAST_HASH_SEED, /* supplementaryPlaneFix= */ true);
private static final int CHUNK_SIZE = 4;
private static final int C1 = 0xcc9e2d51;
private static final int C2 = 0x1b873593;
private final int seed;
private final boolean supplementaryPlaneFix;
Murmur3_32HashFunction(int seed, boolean supplementaryPlaneFix) {
this.seed = seed;
this.supplementaryPlaneFix = supplementaryPlaneFix;
}
@Override
public int bits() {
return 32;
}
@Override
public Hasher newHasher() {
return new Murmur3_32Hasher(seed);
}
@Override
public String toString() {
return "Hashing.murmur3_32(" + seed + ")";
}
@Override
public boolean equals(@CheckForNull Object object) {
if (object instanceof Murmur3_32HashFunction) {
Murmur3_32HashFunction other = (Murmur3_32HashFunction) object;
return seed == other.seed && supplementaryPlaneFix == other.supplementaryPlaneFix;
}
return false;
}
@Override
public int hashCode() {
return getClass().hashCode() ^ seed;
}
@Override
public HashCode hashInt(int input) {
int k1 = mixK1(input);
int h1 = mixH1(seed, k1);
return fmix(h1, Ints.BYTES);
}
@Override
public HashCode hashLong(long input) {
int low = (int) input;
int high = (int) (input >>> 32);
int k1 = mixK1(low);
int h1 = mixH1(seed, k1);
k1 = mixK1(high);
h1 = mixH1(h1, k1);
return fmix(h1, Longs.BYTES);
}
@Override
public HashCode hashUnencodedChars(CharSequence input) {
int h1 = seed;
// step through the CharSequence 2 chars at a time
for (int i = 1; i < input.length(); i += 2) {
int k1 = input.charAt(i - 1) | (input.charAt(i) << 16);
k1 = mixK1(k1);
h1 = mixH1(h1, k1);
}
// deal with any remaining characters
if ((input.length() & 1) == 1) {
int k1 = input.charAt(input.length() - 1);
k1 = mixK1(k1);
h1 ^= k1;
}
return fmix(h1, Chars.BYTES * input.length());
}
@SuppressWarnings("deprecation") // need to use Charsets for Android tests to pass
@Override
public HashCode hashString(CharSequence input, Charset charset) {
if (Charsets.UTF_8.equals(charset)) {
int utf16Length = input.length();
int h1 = seed;
int i = 0;
int len = 0;
// This loop optimizes for pure ASCII.
while (i + 4 <= utf16Length) {
char c0 = input.charAt(i);
char c1 = input.charAt(i + 1);
char c2 = input.charAt(i + 2);
char c3 = input.charAt(i + 3);
if (c0 < 0x80 && c1 < 0x80 && c2 < 0x80 && c3 < 0x80) {
int k1 = c0 | (c1 << 8) | (c2 << 16) | (c3 << 24);
k1 = mixK1(k1);
h1 = mixH1(h1, k1);
i += 4;
len += 4;
} else {
break;
}
}
long buffer = 0;
int shift = 0;
for (; i < utf16Length; i++) {
char c = input.charAt(i);
if (c < 0x80) {
buffer |= (long) c << shift;
shift += 8;
len++;
} else if (c < 0x800) {
buffer |= charToTwoUtf8Bytes(c) << shift;
shift += 16;
len += 2;
} else if (c < Character.MIN_SURROGATE || c > Character.MAX_SURROGATE) {
buffer |= charToThreeUtf8Bytes(c) << shift;
shift += 24;
len += 3;
} else {
int codePoint = Character.codePointAt(input, i);
if (codePoint == c) {
// not a valid code point; let the JDK handle invalid Unicode
return hashBytes(input.toString().getBytes(charset));
}
i++;
buffer |= codePointToFourUtf8Bytes(codePoint) << shift;
if (supplementaryPlaneFix) { // bug compatibility: earlier versions did not have this add
shift += 32;
}
len += 4;
}
if (shift >= 32) {
int k1 = mixK1((int) buffer);
h1 = mixH1(h1, k1);
buffer = buffer >>> 32;
shift -= 32;
}
}
int k1 = mixK1((int) buffer);
h1 ^= k1;
return fmix(h1, len);
} else {
return hashBytes(input.toString().getBytes(charset));
}
}
@Override
public HashCode hashBytes(byte[] input, int off, int len) {
checkPositionIndexes(off, off + len, input.length);
int h1 = seed;
int i;
for (i = 0; i + CHUNK_SIZE <= len; i += CHUNK_SIZE) {
int k1 = mixK1(getIntLittleEndian(input, off + i));
h1 = mixH1(h1, k1);
}
int k1 = 0;
for (int shift = 0; i < len; i++, shift += 8) {
k1 ^= toInt(input[off + i]) << shift;
}
h1 ^= mixK1(k1);
return fmix(h1, len);
}
private static int getIntLittleEndian(byte[] input, int offset) {
return Ints.fromBytes(input[offset + 3], input[offset + 2], input[offset + 1], input[offset]);
}
private static int mixK1(int k1) {
k1 *= C1;
k1 = Integer.rotateLeft(k1, 15);
k1 *= C2;
return k1;
}
private static int mixH1(int h1, int k1) {
h1 ^= k1;
h1 = Integer.rotateLeft(h1, 13);
h1 = h1 * 5 + 0xe6546b64;
return h1;
}
// Finalization mix - force all bits of a hash block to avalanche
private static HashCode fmix(int h1, int length) {
h1 ^= length;
h1 ^= h1 >>> 16;
h1 *= 0x85ebca6b;
h1 ^= h1 >>> 13;
h1 *= 0xc2b2ae35;
h1 ^= h1 >>> 16;
return HashCode.fromInt(h1);
}
@CanIgnoreReturnValue
private static final class Murmur3_32Hasher extends AbstractHasher {
private int h1;
private long buffer;
private int shift;
private int length;
private boolean isDone;
Murmur3_32Hasher(int seed) {
this.h1 = seed;
this.length = 0;
isDone = false;
}
private void update(int nBytes, long update) {
// 1 <= nBytes <= 4
buffer |= (update & 0xFFFFFFFFL) << shift;
shift += nBytes * 8;
length += nBytes;
if (shift >= 32) {
h1 = mixH1(h1, mixK1((int) buffer));
buffer >>>= 32;
shift -= 32;
}
}
@Override
public Hasher putByte(byte b) {
update(1, b & 0xFF);
return this;
}
@Override
public Hasher putBytes(byte[] bytes, int off, int len) {
checkPositionIndexes(off, off + len, bytes.length);
int i;
for (i = 0; i + 4 <= len; i += 4) {
update(4, getIntLittleEndian(bytes, off + i));
}
for (; i < len; i++) {
putByte(bytes[off + i]);
}
return this;
}
@Override
public Hasher putBytes(ByteBuffer buffer) {
ByteOrder bo = buffer.order();
buffer.order(ByteOrder.LITTLE_ENDIAN);
while (buffer.remaining() >= 4) {
putInt(buffer.getInt());
}
while (buffer.hasRemaining()) {
putByte(buffer.get());
}
buffer.order(bo);
return this;
}
@Override
public Hasher putInt(int i) {
update(4, i);
return this;
}
@Override
public Hasher putLong(long l) {
update(4, (int) l);
update(4, l >>> 32);
return this;
}
@Override
public Hasher putChar(char c) {
update(2, c);
return this;
}
@SuppressWarnings("deprecation") // need to use Charsets for Android tests to pass
@Override
public Hasher putString(CharSequence input, Charset charset) {
if (Charsets.UTF_8.equals(charset)) {
int utf16Length = input.length();
int i = 0;
// This loop optimizes for pure ASCII.
while (i + 4 <= utf16Length) {
char c0 = input.charAt(i);
char c1 = input.charAt(i + 1);
char c2 = input.charAt(i + 2);
char c3 = input.charAt(i + 3);
if (c0 < 0x80 && c1 < 0x80 && c2 < 0x80 && c3 < 0x80) {
update(4, c0 | (c1 << 8) | (c2 << 16) | (c3 << 24));
i += 4;
} else {
break;
}
}
for (; i < utf16Length; i++) {
char c = input.charAt(i);
if (c < 0x80) {
update(1, c);
} else if (c < 0x800) {
update(2, charToTwoUtf8Bytes(c));
} else if (c < Character.MIN_SURROGATE || c > Character.MAX_SURROGATE) {
update(3, charToThreeUtf8Bytes(c));
} else {
int codePoint = Character.codePointAt(input, i);
if (codePoint == c) {
// fall back to JDK getBytes instead of trying to handle invalid surrogates ourselves
putBytes(input.subSequence(i, utf16Length).toString().getBytes(charset));
return this;
}
i++;
update(4, codePointToFourUtf8Bytes(codePoint));
}
}
return this;
} else {
return super.putString(input, charset);
}
}
@Override
public HashCode hash() {
checkState(!isDone);
isDone = true;
h1 ^= mixK1((int) buffer);
return fmix(h1, length);
}
}
private static long codePointToFourUtf8Bytes(int codePoint) {
// codePoint has at most 21 bits
return ((0xFL << 4) | (codePoint >>> 18))
| ((0x80L | (0x3F & (codePoint >>> 12))) << 8)
| ((0x80L | (0x3F & (codePoint >>> 6))) << 16)
| ((0x80L | (0x3F & codePoint)) << 24);
}
private static long charToThreeUtf8Bytes(char c) {
return ((0x7L << 5) | (c >>> 12))
| ((0x80 | (0x3F & (c >>> 6))) << 8)
| ((0x80 | (0x3F & c)) << 16);
}
private static long charToTwoUtf8Bytes(char c) {
// c has at most 11 bits
return ((0x3L << 6) | (c >>> 6)) | ((0x80 | (0x3F & c)) << 8);
}
private static final long serialVersionUID = 0L;
}