| 1 | /********** |
| 2 | This library is free software; you can redistribute it and/or modify it under |
| 3 | the terms of the GNU Lesser General Public License as published by the |
| 4 | Free Software Foundation; either version 3 of the License, or (at your |
| 5 | option) any later version. (See <http://www.gnu.org/copyleft/lesser.html>.) |
| 6 | |
| 7 | This library is distributed in the hope that it will be useful, but WITHOUT |
| 8 | ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS |
| 9 | FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for |
| 10 | more details. |
| 11 | |
| 12 | You should have received a copy of the GNU Lesser General Public License |
| 13 | along with this library; if not, write to the Free Software Foundation, Inc., |
| 14 | 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
| 15 | **********/ |
| 16 | // Copyright (c) 1996-2020 Live Networks, Inc. All rights reserved. |
| 17 | // Basic Hash Table implementation |
| 18 | // Implementation |
| 19 | |
| 20 | #include "BasicHashTable.hh" |
| 21 | #include "strDup.hh" |
| 22 | |
| 23 | #if defined(__WIN32__) || defined(_WIN32) |
| 24 | #else |
| 25 | #include <stddef.h> |
| 26 | #endif |
| 27 | #include <string.h> |
| 28 | #include <stdio.h> |
| 29 | |
| 30 | // When there are this many entries per bucket, on average, rebuild |
| 31 | // the table to increase the number of buckets |
| 32 | #define REBUILD_MULTIPLIER 3 |
| 33 | |
| 34 | BasicHashTable::BasicHashTable(int keyType) |
| 35 | : fBuckets(fStaticBuckets), fNumBuckets(SMALL_HASH_TABLE_SIZE), |
| 36 | fNumEntries(0), fRebuildSize(SMALL_HASH_TABLE_SIZE*REBUILD_MULTIPLIER), |
| 37 | fDownShift(28), fMask(0x3), fKeyType(keyType) { |
| 38 | for (unsigned i = 0; i < SMALL_HASH_TABLE_SIZE; ++i) { |
| 39 | fStaticBuckets[i] = NULL; |
| 40 | } |
| 41 | } |
| 42 | |
| 43 | BasicHashTable::~BasicHashTable() { |
| 44 | // Free all the entries in the table: |
| 45 | for (unsigned i = 0; i < fNumBuckets; ++i) { |
| 46 | TableEntry* entry; |
| 47 | while ((entry = fBuckets[i]) != NULL) { |
| 48 | deleteEntry(i, entry); |
| 49 | } |
| 50 | } |
| 51 | |
| 52 | // Also free the bucket array, if it was dynamically allocated: |
| 53 | if (fBuckets != fStaticBuckets) delete[] fBuckets; |
| 54 | } |
| 55 | |
| 56 | void* BasicHashTable::Add(char const* key, void* value) { |
| 57 | void* oldValue; |
| 58 | unsigned index; |
| 59 | TableEntry* entry = lookupKey(key, index); |
| 60 | if (entry != NULL) { |
| 61 | // There's already an item with this key |
| 62 | oldValue = entry->value; |
| 63 | } else { |
| 64 | // There's no existing entry; create a new one: |
| 65 | entry = insertNewEntry(index, key); |
| 66 | oldValue = NULL; |
| 67 | } |
| 68 | entry->value = value; |
| 69 | |
| 70 | // If the table has become too large, rebuild it with more buckets: |
| 71 | if (fNumEntries >= fRebuildSize) rebuild(); |
| 72 | |
| 73 | return oldValue; |
| 74 | } |
| 75 | |
| 76 | Boolean BasicHashTable::Remove(char const* key) { |
| 77 | unsigned index; |
| 78 | TableEntry* entry = lookupKey(key, index); |
| 79 | if (entry == NULL) return False; // no such entry |
| 80 | |
| 81 | deleteEntry(index, entry); |
| 82 | |
| 83 | return True; |
| 84 | } |
| 85 | |
| 86 | void* BasicHashTable::Lookup(char const* key) const { |
| 87 | unsigned index; |
| 88 | TableEntry* entry = lookupKey(key, index); |
| 89 | if (entry == NULL) return NULL; // no such entry |
| 90 | |
| 91 | return entry->value; |
| 92 | } |
| 93 | |
| 94 | unsigned BasicHashTable::numEntries() const { |
| 95 | return fNumEntries; |
| 96 | } |
| 97 | |
| 98 | BasicHashTable::Iterator::Iterator(BasicHashTable const& table) |
| 99 | : fTable(table), fNextIndex(0), fNextEntry(NULL) { |
| 100 | } |
| 101 | |
| 102 | void* BasicHashTable::Iterator::next(char const*& key) { |
| 103 | while (fNextEntry == NULL) { |
| 104 | if (fNextIndex >= fTable.fNumBuckets) return NULL; |
| 105 | |
| 106 | fNextEntry = fTable.fBuckets[fNextIndex++]; |
| 107 | } |
| 108 | |
| 109 | BasicHashTable::TableEntry* entry = fNextEntry; |
| 110 | fNextEntry = entry->fNext; |
| 111 | |
| 112 | key = entry->key; |
| 113 | return entry->value; |
| 114 | } |
| 115 | |
| 116 | ////////// Implementation of HashTable creation functions ////////// |
| 117 | |
| 118 | HashTable* HashTable::create(int keyType) { |
| 119 | return new BasicHashTable(keyType); |
| 120 | } |
| 121 | |
| 122 | HashTable::Iterator* HashTable::Iterator::create(HashTable const& hashTable) { |
| 123 | // "hashTable" is assumed to be a BasicHashTable |
| 124 | return new BasicHashTable::Iterator((BasicHashTable const&)hashTable); |
| 125 | } |
| 126 | |
| 127 | ////////// Implementation of internal member functions ////////// |
| 128 | |
| 129 | BasicHashTable::TableEntry* BasicHashTable |
| 130 | ::lookupKey(char const* key, unsigned& index) const { |
| 131 | TableEntry* entry; |
| 132 | index = hashIndexFromKey(key); |
| 133 | |
| 134 | for (entry = fBuckets[index]; entry != NULL; entry = entry->fNext) { |
| 135 | if (keyMatches(key, entry->key)) break; |
| 136 | } |
| 137 | |
| 138 | return entry; |
| 139 | } |
| 140 | |
| 141 | Boolean BasicHashTable |
| 142 | ::keyMatches(char const* key1, char const* key2) const { |
| 143 | // The way we check the keys for a match depends upon their type: |
| 144 | if (fKeyType == STRING_HASH_KEYS) { |
| 145 | return (strcmp(key1, key2) == 0); |
| 146 | } else if (fKeyType == ONE_WORD_HASH_KEYS) { |
| 147 | return (key1 == key2); |
| 148 | } else { |
| 149 | unsigned* k1 = (unsigned*)key1; |
| 150 | unsigned* k2 = (unsigned*)key2; |
| 151 | |
| 152 | for (int i = 0; i < fKeyType; ++i) { |
| 153 | if (k1[i] != k2[i]) return False; // keys differ |
| 154 | } |
| 155 | return True; |
| 156 | } |
| 157 | } |
| 158 | |
| 159 | BasicHashTable::TableEntry* BasicHashTable |
| 160 | ::insertNewEntry(unsigned index, char const* key) { |
| 161 | TableEntry* entry = new TableEntry(); |
| 162 | entry->fNext = fBuckets[index]; |
| 163 | fBuckets[index] = entry; |
| 164 | |
| 165 | ++fNumEntries; |
| 166 | assignKey(entry, key); |
| 167 | |
| 168 | return entry; |
| 169 | } |
| 170 | |
| 171 | void BasicHashTable::assignKey(TableEntry* entry, char const* key) { |
| 172 | // The way we assign the key depends upon its type: |
| 173 | if (fKeyType == STRING_HASH_KEYS) { |
| 174 | entry->key = strDup(key); |
| 175 | } else if (fKeyType == ONE_WORD_HASH_KEYS) { |
| 176 | entry->key = key; |
| 177 | } else if (fKeyType > 0) { |
| 178 | unsigned* keyFrom = (unsigned*)key; |
| 179 | unsigned* keyTo = new unsigned[fKeyType]; |
| 180 | for (int i = 0; i < fKeyType; ++i) keyTo[i] = keyFrom[i]; |
| 181 | |
| 182 | entry->key = (char const*)keyTo; |
| 183 | } |
| 184 | } |
| 185 | |
| 186 | void BasicHashTable::deleteEntry(unsigned index, TableEntry* entry) { |
| 187 | TableEntry** ep = &fBuckets[index]; |
| 188 | |
| 189 | Boolean foundIt = False; |
| 190 | while (*ep != NULL) { |
| 191 | if (*ep == entry) { |
| 192 | foundIt = True; |
| 193 | *ep = entry->fNext; |
| 194 | break; |
| 195 | } |
| 196 | ep = &((*ep)->fNext); |
| 197 | } |
| 198 | |
| 199 | if (!foundIt) { // shouldn't happen |
| 200 | #ifdef DEBUG |
| 201 | fprintf(stderr, "BasicHashTable[%p]::deleteEntry(%d,%p): internal error - not found (first entry %p" , this, index, entry, fBuckets[index]); |
| 202 | if (fBuckets[index] != NULL) fprintf(stderr, ", next entry %p" , fBuckets[index]->fNext); |
| 203 | fprintf(stderr, ")\n" ); |
| 204 | #endif |
| 205 | } |
| 206 | |
| 207 | --fNumEntries; |
| 208 | deleteKey(entry); |
| 209 | delete entry; |
| 210 | } |
| 211 | |
| 212 | void BasicHashTable::deleteKey(TableEntry* entry) { |
| 213 | // The way we delete the key depends upon its type: |
| 214 | if (fKeyType == ONE_WORD_HASH_KEYS) { |
| 215 | entry->key = NULL; |
| 216 | } else { |
| 217 | delete[] (char*)entry->key; |
| 218 | entry->key = NULL; |
| 219 | } |
| 220 | } |
| 221 | |
| 222 | void BasicHashTable::rebuild() { |
| 223 | // Remember the existing table size: |
| 224 | unsigned oldSize = fNumBuckets; |
| 225 | TableEntry** oldBuckets = fBuckets; |
| 226 | |
| 227 | // Create the new sized table: |
| 228 | fNumBuckets *= 4; |
| 229 | fBuckets = new TableEntry*[fNumBuckets]; |
| 230 | for (unsigned i = 0; i < fNumBuckets; ++i) { |
| 231 | fBuckets[i] = NULL; |
| 232 | } |
| 233 | fRebuildSize *= 4; |
| 234 | fDownShift -= 2; |
| 235 | fMask = (fMask<<2)|0x3; |
| 236 | |
| 237 | // Rehash the existing entries into the new table: |
| 238 | for (TableEntry** oldChainPtr = oldBuckets; oldSize > 0; |
| 239 | --oldSize, ++oldChainPtr) { |
| 240 | for (TableEntry* hPtr = *oldChainPtr; hPtr != NULL; |
| 241 | hPtr = *oldChainPtr) { |
| 242 | *oldChainPtr = hPtr->fNext; |
| 243 | |
| 244 | unsigned index = hashIndexFromKey(hPtr->key); |
| 245 | |
| 246 | hPtr->fNext = fBuckets[index]; |
| 247 | fBuckets[index] = hPtr; |
| 248 | } |
| 249 | } |
| 250 | |
| 251 | // Free the old bucket array, if it was dynamically allocated: |
| 252 | if (oldBuckets != fStaticBuckets) delete[] oldBuckets; |
| 253 | } |
| 254 | |
| 255 | unsigned BasicHashTable::hashIndexFromKey(char const* key) const { |
| 256 | unsigned result = 0; |
| 257 | |
| 258 | if (fKeyType == STRING_HASH_KEYS) { |
| 259 | while (1) { |
| 260 | char c = *key++; |
| 261 | if (c == 0) break; |
| 262 | result += (result<<3) + (unsigned)c; |
| 263 | } |
| 264 | result &= fMask; |
| 265 | } else if (fKeyType == ONE_WORD_HASH_KEYS) { |
| 266 | result = randomIndex((uintptr_t)key); |
| 267 | } else { |
| 268 | unsigned* k = (unsigned*)key; |
| 269 | uintptr_t sum = 0; |
| 270 | for (int i = 0; i < fKeyType; ++i) { |
| 271 | sum += k[i]; |
| 272 | } |
| 273 | result = randomIndex(sum); |
| 274 | } |
| 275 | |
| 276 | return result; |
| 277 | } |
| 278 | |