001/*
002 * Copyright (C) 2011 The Guava Authors
003 *
004 * Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
005 * in compliance with the License. You may obtain a copy of the License at
006 *
007 * http://www.apache.org/licenses/LICENSE-2.0
008 *
009 * Unless required by applicable law or agreed to in writing, software distributed under the License
010 * is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
011 * or implied. See the License for the specific language governing permissions and limitations under
012 * the License.
013 */
014
015package com.google.common.hash;
016
017import com.google.common.annotations.Beta;
018import com.google.errorprone.annotations.CanIgnoreReturnValue;
019import java.nio.ByteBuffer;
020import java.nio.charset.Charset;
021
022/**
023 * A {@link PrimitiveSink} that can compute a hash code after reading the input. Each hasher should
024 * translate all multibyte values ({@link #putInt(int)}, {@link #putLong(long)}, etc) to bytes in
025 * little-endian order.
026 *
027 * <p><b>Warning:</b> The result of calling any methods after calling {@link #hash} is undefined.
028 *
029 * <p><b>Warning:</b> Using a specific character encoding when hashing a {@link CharSequence} with
030 * {@link #putString(CharSequence, Charset)} is generally only useful for cross-language
031 * compatibility (otherwise prefer {@link #putUnencodedChars}). However, the character encodings
032 * must be identical across languages. Also beware that {@link Charset} definitions may occasionally
033 * change between Java releases.
034 *
035 * <p><b>Warning:</b> Chunks of data that are put into the {@link Hasher} are not delimited. The
036 * resulting {@link HashCode} is dependent only on the bytes inserted, and the order in which they
037 * were inserted, not how those bytes were chunked into discrete put() operations. For example, the
038 * following three expressions all generate colliding hash codes: <pre>   {@code
039 *
040 *   newHasher().putByte(b1).putByte(b2).putByte(b3).hash()
041 *   newHasher().putByte(b1).putBytes(new byte[] { b2, b3 }).hash()
042 *   newHasher().putBytes(new byte[] { b1, b2, b3 }).hash()}</pre>
043 *
044 * <p>If you wish to avoid this, you should either prepend or append the size of each chunk. Keep in
045 * mind that when dealing with char sequences, the encoded form of two concatenated char sequences
046 * is not equivalent to the concatenation of their encoded form. Therefore,
047 * {@link #putString(CharSequence, Charset)} should only be used consistently with <i>complete</i>
048 * sequences and not broken into chunks.
049 *
050 * @author Kevin Bourrillion
051 * @since 11.0
052 */
053@Beta
054@CanIgnoreReturnValue
055public interface Hasher extends PrimitiveSink {
056  @Override
057  Hasher putByte(byte b);
058
059  @Override
060  Hasher putBytes(byte[] bytes);
061
062  @Override
063  Hasher putBytes(byte[] bytes, int off, int len);
064
065  @Override
066  Hasher putBytes(ByteBuffer bytes);
067
068  @Override
069  Hasher putShort(short s);
070
071  @Override
072  Hasher putInt(int i);
073
074  @Override
075  Hasher putLong(long l);
076
077  /**
078   * Equivalent to {@code putInt(Float.floatToRawIntBits(f))}.
079   */
080  @Override
081  Hasher putFloat(float f);
082
083  /**
084   * Equivalent to {@code putLong(Double.doubleToRawLongBits(d))}.
085   */
086  @Override
087  Hasher putDouble(double d);
088
089  /**
090   * Equivalent to {@code putByte(b ? (byte) 1 : (byte) 0)}.
091   */
092  @Override
093  Hasher putBoolean(boolean b);
094
095  @Override
096  Hasher putChar(char c);
097
098  /**
099   * Equivalent to processing each {@code char} value in the {@code CharSequence}, in order. In
100   * other words, no character encoding is performed; the low byte and high byte of each {@code
101   * char} are hashed directly (in that order). The input must not be updated while this method is
102   * in progress.
103   *
104   * <p><b>Warning:</b> This method will produce different output than most other languages do when
105   * running the same hash function on the equivalent input. For cross-language compatibility, use
106   * {@link #putString}, usually with a charset of UTF-8. For other use cases, use {@code
107   * putUnencodedChars}.
108   *
109   * @since 15.0 (since 11.0 as putString(CharSequence)).
110   */
111  @Override
112  Hasher putUnencodedChars(CharSequence charSequence);
113
114  /**
115   * Equivalent to {@code putBytes(charSequence.toString().getBytes(charset))}.
116   *
117   * <p><b>Warning:</b> This method, which reencodes the input before hashing it, is useful only for
118   * cross-language compatibility. For other use cases, prefer {@link #putUnencodedChars}, which is
119   * faster, produces the same output across Java releases, and hashes every {@code char} in the
120   * input, even if some are invalid.
121   */
122  @Override
123  Hasher putString(CharSequence charSequence, Charset charset);
124
125  /**
126   * A simple convenience for {@code funnel.funnel(object, this)}.
127   */
128  <T> Hasher putObject(T instance, Funnel<? super T> funnel);
129
130  /**
131   * Computes a hash code based on the data that have been provided to this hasher. The result is
132   * unspecified if this method is called more than once on the same instance.
133   */
134  HashCode hash();
135
136  /**
137   * {@inheritDoc}
138   *
139   * @deprecated This returns {@link Object#hashCode()}; you almost certainly mean to call
140   *     {@code hash().asInt()}.
141   */
142  @Override
143  @Deprecated
144  int hashCode();
145}