| 1 | /* |
| 2 | * Copyright (C) 2005, 2006, 2008, 2010 Apple Inc. All rights reserved. |
| 3 | * |
| 4 | * This library is free software; you can redistribute it and/or |
| 5 | * modify it under the terms of the GNU Library General Public |
| 6 | * License as published by the Free Software Foundation; either |
| 7 | * version 2 of the License, or (at your option) any later version. |
| 8 | * |
| 9 | * This library is distributed in the hope that it will be useful, |
| 10 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 11 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 12 | * Library General Public License for more details. |
| 13 | * |
| 14 | * You should have received a copy of the GNU Library General Public License |
| 15 | * along with this library; see the file COPYING.LIB. If not, write to |
| 16 | * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, |
| 17 | * Boston, MA 02110-1301, USA. |
| 18 | * |
| 19 | */ |
| 20 | #ifndef WTF_StringHashFunctions_h |
| 21 | #define WTF_StringHashFunctions_h |
| 22 | |
| 23 | #include <wtf/unicode/Unicode.h> |
| 24 | |
| 25 | namespace WTF { |
| 26 | |
| 27 | // Golden ratio - arbitrary start value to avoid mapping all 0's to all 0's |
| 28 | static const unsigned stringHashingStartValue = 0x9e3779b9U; |
| 29 | |
| 30 | // stringHash methods based on Paul Hsieh's SuperFastHash. |
| 31 | // http://www.azillionmonkeys.com/qed/hash.html |
| 32 | // char* data is interpreted as latin-encoded (zero extended to 16 bits). |
| 33 | |
| 34 | inline unsigned stringHash(const UChar* data, unsigned length) |
| 35 | { |
| 36 | unsigned hash = WTF::stringHashingStartValue; |
| 37 | unsigned rem = length & 1; |
| 38 | length >>= 1; |
| 39 | |
| 40 | // Main loop |
| 41 | for (; length > 0; length--) { |
| 42 | hash += data[0]; |
| 43 | unsigned tmp = (data[1] << 11) ^ hash; |
| 44 | hash = (hash << 16) ^ tmp; |
| 45 | data += 2; |
| 46 | hash += hash >> 11; |
| 47 | } |
| 48 | |
| 49 | // Handle end case |
| 50 | if (rem) { |
| 51 | hash += data[0]; |
| 52 | hash ^= hash << 11; |
| 53 | hash += hash >> 17; |
| 54 | } |
| 55 | |
| 56 | // Force "avalanching" of final 127 bits |
| 57 | hash ^= hash << 3; |
| 58 | hash += hash >> 5; |
| 59 | hash ^= hash << 2; |
| 60 | hash += hash >> 15; |
| 61 | hash ^= hash << 10; |
| 62 | |
| 63 | hash &= 0x7fffffff; |
| 64 | |
| 65 | // this avoids ever returning a hash code of 0, since that is used to |
| 66 | // signal "hash not computed yet", using a value that is likely to be |
| 67 | // effectively the same as 0 when the low bits are masked |
| 68 | if (hash == 0) |
| 69 | hash = 0x40000000; |
| 70 | |
| 71 | return hash; |
| 72 | } |
| 73 | |
| 74 | inline unsigned stringHash(const char* data, unsigned length) |
| 75 | { |
| 76 | unsigned hash = WTF::stringHashingStartValue; |
| 77 | unsigned rem = length & 1; |
| 78 | length >>= 1; |
| 79 | |
| 80 | // Main loop |
| 81 | for (; length > 0; length--) { |
| 82 | hash += static_cast<unsigned char>(data[0]); |
| 83 | unsigned tmp = (static_cast<unsigned char>(data[1]) << 11) ^ hash; |
| 84 | hash = (hash << 16) ^ tmp; |
| 85 | data += 2; |
| 86 | hash += hash >> 11; |
| 87 | } |
| 88 | |
| 89 | // Handle end case |
| 90 | if (rem) { |
| 91 | hash += static_cast<unsigned char>(data[0]); |
| 92 | hash ^= hash << 11; |
| 93 | hash += hash >> 17; |
| 94 | } |
| 95 | |
| 96 | // Force "avalanching" of final 127 bits |
| 97 | hash ^= hash << 3; |
| 98 | hash += hash >> 5; |
| 99 | hash ^= hash << 2; |
| 100 | hash += hash >> 15; |
| 101 | hash ^= hash << 10; |
| 102 | |
| 103 | hash &= 0x7fffffff; |
| 104 | |
| 105 | // this avoids ever returning a hash code of 0, since that is used to |
| 106 | // signal "hash not computed yet", using a value that is likely to be |
| 107 | // effectively the same as 0 when the low bits are masked |
| 108 | if (hash == 0) |
| 109 | hash = 0x40000000; |
| 110 | |
| 111 | return hash; |
| 112 | } |
| 113 | |
| 114 | inline unsigned stringHash(const char* data) |
| 115 | { |
| 116 | unsigned hash = WTF::stringHashingStartValue; |
| 117 | |
| 118 | // Main loop |
| 119 | for (;;) { |
| 120 | unsigned char b0 = data[0]; |
| 121 | if (!b0) |
| 122 | break; |
| 123 | unsigned char b1 = data[1]; |
| 124 | if (!b1) { |
| 125 | hash += b0; |
| 126 | hash ^= hash << 11; |
| 127 | hash += hash >> 17; |
| 128 | break; |
| 129 | } |
| 130 | hash += b0; |
| 131 | unsigned tmp = (b1 << 11) ^ hash; |
| 132 | hash = (hash << 16) ^ tmp; |
| 133 | data += 2; |
| 134 | hash += hash >> 11; |
| 135 | } |
| 136 | |
| 137 | // Force "avalanching" of final 127 bits. |
| 138 | hash ^= hash << 3; |
| 139 | hash += hash >> 5; |
| 140 | hash ^= hash << 2; |
| 141 | hash += hash >> 15; |
| 142 | hash ^= hash << 10; |
| 143 | |
| 144 | hash &= 0x7fffffff; |
| 145 | |
| 146 | // This avoids ever returning a hash code of 0, since that is used to |
| 147 | // signal "hash not computed yet", using a value that is likely to be |
| 148 | // effectively the same as 0 when the low bits are masked. |
| 149 | if (hash == 0) |
| 150 | hash = 0x40000000; |
| 151 | |
| 152 | return hash; |
| 153 | } |
| 154 | |
| 155 | } // namespace WTF |
| 156 | |
| 157 | #endif // WTF_StringHashFunctions_h |
| 158 | |