Tue, 20 Sep 2022 10:36:04 +0200
use EXPECT_STREQ instead of strcmp
universe@549 | 1 | /* |
universe@549 | 2 | * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER. |
universe@549 | 3 | * |
universe@549 | 4 | * Copyright 2021 Mike Becker, Olaf Wintermann All rights reserved. |
universe@549 | 5 | * |
universe@549 | 6 | * Redistribution and use in source and binary forms, with or without |
universe@549 | 7 | * modification, are permitted provided that the following conditions are met: |
universe@549 | 8 | * |
universe@549 | 9 | * 1. Redistributions of source code must retain the above copyright |
universe@549 | 10 | * notice, this list of conditions and the following disclaimer. |
universe@549 | 11 | * |
universe@549 | 12 | * 2. Redistributions in binary form must reproduce the above copyright |
universe@549 | 13 | * notice, this list of conditions and the following disclaimer in the |
universe@549 | 14 | * documentation and/or other materials provided with the distribution. |
universe@549 | 15 | * |
universe@549 | 16 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" |
universe@549 | 17 | * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
universe@549 | 18 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
universe@549 | 19 | * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE |
universe@549 | 20 | * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR |
universe@549 | 21 | * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF |
universe@549 | 22 | * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS |
universe@549 | 23 | * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN |
universe@549 | 24 | * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) |
universe@549 | 25 | * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
universe@549 | 26 | * POSSIBILITY OF SUCH DAMAGE. |
universe@549 | 27 | */ |
universe@549 | 28 | |
universe@549 | 29 | #include <string.h> |
universe@549 | 30 | #include "cx/hash_map.h" |
universe@550 | 31 | #include "cx/utils.h" |
universe@549 | 32 | |
universe@550 | 33 | static void cx_hash_map_clear(struct cx_map_s *map) { |
universe@550 | 34 | struct cx_hash_map_s *hash_map = (struct cx_hash_map_s *) map; |
universe@550 | 35 | cx_for_n(i, hash_map->bucket_count) { |
universe@550 | 36 | struct cx_hash_map_element_s *elem = hash_map->buckets[i]; |
universe@550 | 37 | if (elem != NULL) { |
universe@550 | 38 | do { |
universe@550 | 39 | struct cx_hash_map_element_s *next = elem->next; |
universe@550 | 40 | // free the key data |
universe@563 | 41 | cxFree(map->allocator, elem->key.data.obj); |
universe@550 | 42 | // free the node |
universe@550 | 43 | cxFree(map->allocator, elem); |
universe@550 | 44 | // proceed |
universe@550 | 45 | elem = next; |
universe@550 | 46 | } while (elem != NULL); |
universe@550 | 47 | |
universe@550 | 48 | // do not leave a dangling pointer |
universe@550 | 49 | hash_map->buckets[i] = NULL; |
universe@550 | 50 | } |
universe@550 | 51 | } |
universe@550 | 52 | map->size = 0; |
universe@550 | 53 | } |
universe@550 | 54 | |
universe@550 | 55 | static void cx_hash_map_destructor(struct cx_map_s *map) { |
universe@550 | 56 | struct cx_hash_map_s *hash_map = (struct cx_hash_map_s *) map; |
universe@550 | 57 | |
universe@550 | 58 | // free the buckets |
universe@550 | 59 | cx_hash_map_clear(map); |
universe@550 | 60 | cxFree(map->allocator, hash_map->buckets); |
universe@550 | 61 | |
universe@550 | 62 | // free the map structure |
universe@550 | 63 | cxFree(map->allocator, map); |
universe@550 | 64 | } |
universe@550 | 65 | |
universe@550 | 66 | static int cx_hash_map_put( |
universe@550 | 67 | CxMap *map, |
universe@563 | 68 | CxHashKey key, |
universe@550 | 69 | void *value |
universe@550 | 70 | ) { |
universe@550 | 71 | struct cx_hash_map_s *hash_map = (struct cx_hash_map_s *) map; |
universe@550 | 72 | CxAllocator *allocator = map->allocator; |
universe@550 | 73 | |
universe@563 | 74 | unsigned hash = key.hash; |
universe@563 | 75 | if (hash == 0) { |
universe@563 | 76 | cx_hash_murmur(&key); |
universe@563 | 77 | hash = key.hash; |
universe@563 | 78 | } |
universe@550 | 79 | |
universe@550 | 80 | size_t slot = hash % hash_map->bucket_count; |
universe@550 | 81 | struct cx_hash_map_element_s *elm = hash_map->buckets[slot]; |
universe@550 | 82 | struct cx_hash_map_element_s *prev = NULL; |
universe@550 | 83 | |
universe@550 | 84 | while (elm != NULL && elm->key.hash < hash) { |
universe@550 | 85 | prev = elm; |
universe@550 | 86 | elm = elm->next; |
universe@550 | 87 | } |
universe@550 | 88 | |
universe@575 | 89 | if (elm != NULL && elm->key.hash == hash && elm->key.len == key.len && |
universe@575 | 90 | memcmp(elm->key.data.obj, key.data.obj, key.len) == 0) { |
universe@574 | 91 | // overwrite existing element |
universe@574 | 92 | elm->data = value; |
universe@574 | 93 | } else { |
universe@575 | 94 | // allocate new element |
universe@550 | 95 | struct cx_hash_map_element_s *e = cxMalloc(allocator, sizeof(struct cx_hash_map_element_s)); |
universe@550 | 96 | if (e == NULL) { |
universe@550 | 97 | return -1; |
universe@550 | 98 | } |
universe@550 | 99 | |
universe@550 | 100 | // write the value |
universe@550 | 101 | // TODO: depending on future map features, we may want to copy here |
universe@550 | 102 | e->data = value; |
universe@550 | 103 | |
universe@550 | 104 | // copy the key |
universe@550 | 105 | void *kd = cxMalloc(allocator, key.len); |
universe@550 | 106 | if (kd == NULL) { |
universe@550 | 107 | return -1; |
universe@550 | 108 | } |
universe@563 | 109 | memcpy(kd, key.data.obj, key.len); |
universe@563 | 110 | e->key.data.obj = kd; |
universe@550 | 111 | e->key.len = key.len; |
universe@550 | 112 | e->key.hash = hash; |
universe@550 | 113 | |
universe@550 | 114 | // insert the element into the linked list |
universe@550 | 115 | if (prev == NULL) { |
universe@550 | 116 | hash_map->buckets[slot] = e; |
universe@550 | 117 | } else { |
universe@550 | 118 | prev->next = e; |
universe@550 | 119 | } |
universe@550 | 120 | e->next = elm; |
universe@550 | 121 | |
universe@550 | 122 | // increase the size |
universe@550 | 123 | map->size++; |
universe@550 | 124 | } |
universe@550 | 125 | |
universe@550 | 126 | return 0; |
universe@550 | 127 | } |
universe@549 | 128 | |
universe@551 | 129 | static void cx_hash_map_unlink( |
universe@551 | 130 | struct cx_hash_map_s *hash_map, |
universe@551 | 131 | size_t slot, |
universe@551 | 132 | struct cx_hash_map_element_s *prev, |
universe@551 | 133 | struct cx_hash_map_element_s *elm |
universe@551 | 134 | ) { |
universe@551 | 135 | // unlink |
universe@551 | 136 | if (prev == NULL) { |
universe@551 | 137 | hash_map->buckets[slot] = elm->next; |
universe@551 | 138 | } else { |
universe@551 | 139 | prev->next = elm->next; |
universe@551 | 140 | } |
universe@551 | 141 | // free element |
universe@563 | 142 | cxFree(hash_map->base.allocator, elm->key.data.obj); |
universe@551 | 143 | cxFree(hash_map->base.allocator, elm); |
universe@551 | 144 | // decrease size |
universe@551 | 145 | hash_map->base.size--; |
universe@551 | 146 | } |
universe@551 | 147 | |
universe@549 | 148 | /** |
universe@550 | 149 | * Helper function to avoid code duplication. |
universe@549 | 150 | * |
universe@550 | 151 | * @param map the map |
universe@550 | 152 | * @param key the key to look up |
universe@550 | 153 | * @param remove flag indicating whether the looked up entry shall be removed |
universe@550 | 154 | * @return the value corresponding to the key or \c NULL |
universe@549 | 155 | */ |
universe@550 | 156 | static void *cx_hash_map_get_remove( |
universe@550 | 157 | CxMap *map, |
universe@563 | 158 | CxHashKey key, |
universe@550 | 159 | bool remove |
universe@550 | 160 | ) { |
universe@550 | 161 | struct cx_hash_map_s *hash_map = (struct cx_hash_map_s *) map; |
universe@550 | 162 | |
universe@563 | 163 | unsigned hash = key.hash; |
universe@563 | 164 | if (hash == 0) { |
universe@563 | 165 | cx_hash_murmur(&key); |
universe@563 | 166 | hash = key.hash; |
universe@563 | 167 | } |
universe@550 | 168 | |
universe@550 | 169 | size_t slot = hash % hash_map->bucket_count; |
universe@550 | 170 | struct cx_hash_map_element_s *elm = hash_map->buckets[slot]; |
universe@550 | 171 | struct cx_hash_map_element_s *prev = NULL; |
universe@550 | 172 | while (elm && elm->key.hash <= hash) { |
universe@550 | 173 | if (elm->key.hash == hash && elm->key.len == key.len) { |
universe@563 | 174 | if (memcmp(elm->key.data.obj, key.data.obj, key.len) == 0) { |
universe@550 | 175 | void *data = elm->data; |
universe@550 | 176 | if (remove) { |
universe@551 | 177 | cx_hash_map_unlink(hash_map, slot, prev, elm); |
universe@550 | 178 | } |
universe@550 | 179 | return data; |
universe@550 | 180 | } |
universe@550 | 181 | } |
universe@550 | 182 | prev = elm; |
universe@550 | 183 | elm = prev->next; |
universe@550 | 184 | } |
universe@550 | 185 | |
universe@550 | 186 | return NULL; |
universe@549 | 187 | } |
universe@550 | 188 | |
universe@550 | 189 | static void *cx_hash_map_get( |
universe@550 | 190 | CxMap const *map, |
universe@563 | 191 | CxHashKey key |
universe@550 | 192 | ) { |
universe@550 | 193 | // we can safely cast, because we know when remove=false, the map stays untouched |
universe@550 | 194 | return cx_hash_map_get_remove((CxMap *) map, key, false); |
universe@550 | 195 | } |
universe@550 | 196 | |
universe@550 | 197 | static void *cx_hash_map_remove( |
universe@550 | 198 | CxMap *map, |
universe@563 | 199 | CxHashKey key |
universe@550 | 200 | ) { |
universe@550 | 201 | return cx_hash_map_get_remove(map, key, true); |
universe@550 | 202 | } |
universe@550 | 203 | |
universe@551 | 204 | static void *cx_hash_map_iter_current_entry(CxIterator const *iter) { |
universe@551 | 205 | // struct has to have a compatible signature |
universe@573 | 206 | return (struct cx_map_entry_s *) &(iter->kv_data); |
universe@551 | 207 | } |
universe@551 | 208 | |
universe@551 | 209 | static void *cx_hash_map_iter_current_key(CxIterator const *iter) { |
universe@551 | 210 | struct cx_hash_map_element_s *elm = iter->elem_handle; |
universe@551 | 211 | return &elm->key; |
universe@551 | 212 | } |
universe@551 | 213 | |
universe@551 | 214 | static void *cx_hash_map_iter_current_value(CxIterator const *iter) { |
universe@551 | 215 | struct cx_hash_map_element_s *elm = iter->elem_handle; |
universe@551 | 216 | // TODO: return a pointer to data if this map is storing copies |
universe@551 | 217 | return elm->data; |
universe@551 | 218 | } |
universe@551 | 219 | |
universe@551 | 220 | static bool cx_hash_map_iter_valid(CxIterator const *iter) { |
universe@551 | 221 | return iter->elem_handle != NULL; |
universe@551 | 222 | } |
universe@551 | 223 | |
universe@551 | 224 | static void cx_hash_map_iter_next(CxIterator *iter) { |
universe@551 | 225 | struct cx_hash_map_s *map = iter->src_handle; |
universe@551 | 226 | struct cx_hash_map_element_s *elm = iter->elem_handle; |
universe@551 | 227 | |
universe@551 | 228 | // remove current element, if asked |
universe@551 | 229 | if (iter->remove) { |
universe@551 | 230 | // clear the flag |
universe@551 | 231 | iter->remove = false; |
universe@551 | 232 | |
universe@551 | 233 | // determine the next element |
universe@551 | 234 | struct cx_hash_map_element_s *next = elm->next; |
universe@551 | 235 | |
universe@551 | 236 | // search the previous element |
universe@551 | 237 | struct cx_hash_map_element_s *prev = NULL; |
universe@551 | 238 | if (map->buckets[iter->slot] != elm) { |
universe@551 | 239 | prev = map->buckets[iter->slot]; |
universe@551 | 240 | while (prev->next != elm) { |
universe@551 | 241 | prev = prev->next; |
universe@551 | 242 | } |
universe@551 | 243 | } |
universe@551 | 244 | |
universe@551 | 245 | // unlink |
universe@551 | 246 | cx_hash_map_unlink(map, iter->slot, prev, elm); |
universe@551 | 247 | |
universe@551 | 248 | // advance |
universe@551 | 249 | elm = next; |
universe@551 | 250 | } else { |
universe@551 | 251 | // just advance |
universe@551 | 252 | elm = elm->next; |
universe@560 | 253 | iter->index++; |
universe@551 | 254 | } |
universe@551 | 255 | |
universe@560 | 256 | // search the next bucket, if required |
universe@560 | 257 | while (elm == NULL && ++iter->slot < map->bucket_count) { |
universe@560 | 258 | elm = map->buckets[iter->slot]; |
universe@551 | 259 | } |
universe@551 | 260 | |
universe@560 | 261 | // fill the struct with the next element |
universe@551 | 262 | iter->elem_handle = elm; |
universe@551 | 263 | if (elm == NULL) { |
universe@551 | 264 | iter->kv_data.key = NULL; |
universe@551 | 265 | iter->kv_data.value = NULL; |
universe@551 | 266 | } else { |
universe@551 | 267 | iter->kv_data.key = &elm->key; |
universe@551 | 268 | // TODO: pointer to data if this map is storing copies |
universe@551 | 269 | iter->kv_data.value = elm->data; |
universe@551 | 270 | } |
universe@551 | 271 | } |
universe@551 | 272 | |
universe@551 | 273 | static CxIterator cx_hash_map_iterator(CxMap *map) { |
universe@550 | 274 | CxIterator iter; |
universe@551 | 275 | |
universe@551 | 276 | iter.src_handle = map; |
universe@551 | 277 | iter.valid = cx_hash_map_iter_valid; |
universe@551 | 278 | iter.next = cx_hash_map_iter_next; |
universe@551 | 279 | iter.current = cx_hash_map_iter_current_entry; |
universe@551 | 280 | |
universe@551 | 281 | iter.slot = 0; |
universe@551 | 282 | iter.index = 0; |
universe@551 | 283 | iter.remove = false; |
universe@551 | 284 | |
universe@551 | 285 | if (map->size > 0) { |
universe@551 | 286 | struct cx_hash_map_s *hash_map = (struct cx_hash_map_s *) map; |
universe@560 | 287 | struct cx_hash_map_element_s *elm = hash_map->buckets[0]; |
universe@554 | 288 | for (; elm == NULL; iter.slot++) { |
universe@554 | 289 | elm = hash_map->buckets[iter.slot]; |
universe@551 | 290 | } |
universe@554 | 291 | iter.elem_handle = elm; |
universe@554 | 292 | iter.kv_data.key = &elm->key; |
universe@554 | 293 | // TODO: pointer to data if this map is storing copies |
universe@554 | 294 | iter.kv_data.value = elm->data; |
universe@551 | 295 | } else { |
universe@551 | 296 | iter.elem_handle = NULL; |
universe@551 | 297 | iter.kv_data.key = NULL; |
universe@551 | 298 | iter.kv_data.value = NULL; |
universe@551 | 299 | } |
universe@551 | 300 | |
universe@550 | 301 | return iter; |
universe@550 | 302 | } |
universe@550 | 303 | |
universe@551 | 304 | static CxIterator cx_hash_map_iterator_keys(CxMap *map) { |
universe@551 | 305 | CxIterator iter = cx_hash_map_iterator(map); |
universe@551 | 306 | iter.current = cx_hash_map_iter_current_key; |
universe@550 | 307 | return iter; |
universe@550 | 308 | } |
universe@550 | 309 | |
universe@551 | 310 | static CxIterator cx_hash_map_iterator_values(CxMap *map) { |
universe@551 | 311 | CxIterator iter = cx_hash_map_iterator(map); |
universe@551 | 312 | iter.current = cx_hash_map_iter_current_value; |
universe@550 | 313 | return iter; |
universe@550 | 314 | } |
universe@550 | 315 | |
universe@550 | 316 | static cx_map_class cx_hash_map_class = { |
universe@550 | 317 | cx_hash_map_destructor, |
universe@550 | 318 | cx_hash_map_clear, |
universe@550 | 319 | cx_hash_map_put, |
universe@550 | 320 | cx_hash_map_get, |
universe@550 | 321 | cx_hash_map_remove, |
universe@550 | 322 | cx_hash_map_iterator, |
universe@550 | 323 | cx_hash_map_iterator_keys, |
universe@550 | 324 | cx_hash_map_iterator_values, |
universe@550 | 325 | }; |
universe@550 | 326 | |
universe@550 | 327 | CxMap *cxHashMapCreate( |
universe@550 | 328 | CxAllocator *allocator, |
universe@550 | 329 | size_t buckets |
universe@550 | 330 | ) { |
universe@550 | 331 | if (buckets == 0) { |
universe@550 | 332 | // implementation defined default |
universe@550 | 333 | buckets = 16; |
universe@550 | 334 | } |
universe@550 | 335 | |
universe@550 | 336 | struct cx_hash_map_s *map = cxMalloc(allocator, sizeof(struct cx_hash_map_s)); |
universe@550 | 337 | if (map == NULL) return NULL; |
universe@550 | 338 | |
universe@550 | 339 | // initialize hash map members |
universe@550 | 340 | map->bucket_count = buckets; |
universe@550 | 341 | map->buckets = cxCalloc(allocator, buckets, sizeof(struct cx_hash_map_element_s *)); |
universe@550 | 342 | if (map->buckets == NULL) { |
universe@550 | 343 | cxFree(allocator, map); |
universe@550 | 344 | return NULL; |
universe@550 | 345 | } |
universe@550 | 346 | |
universe@550 | 347 | // initialize base members |
universe@550 | 348 | map->base.cl = &cx_hash_map_class; |
universe@550 | 349 | map->base.allocator = allocator; |
universe@550 | 350 | map->base.size = 0; |
universe@550 | 351 | |
universe@550 | 352 | return (CxMap *) map; |
universe@550 | 353 | } |
universe@562 | 354 | |
universe@562 | 355 | int cxMapRehash(CxMap *map) { |
universe@562 | 356 | struct cx_hash_map_s *hash_map = (struct cx_hash_map_s *) map; |
universe@562 | 357 | if (map->size > ((hash_map->bucket_count * 3) >> 2)) { |
universe@562 | 358 | |
universe@562 | 359 | size_t new_bucket_count = (map->size * 5) >> 1; |
universe@562 | 360 | struct cx_hash_map_element_s **new_buckets = cxCalloc(map->allocator, |
universe@562 | 361 | new_bucket_count, sizeof(struct cx_hash_map_element_s *)); |
universe@562 | 362 | |
universe@562 | 363 | if (new_buckets == NULL) { |
universe@562 | 364 | return 1; |
universe@562 | 365 | } |
universe@562 | 366 | |
universe@562 | 367 | // iterate through the elements and assign them to their new slots |
universe@562 | 368 | cx_for_n(slot, hash_map->bucket_count) { |
universe@562 | 369 | struct cx_hash_map_element_s *elm = hash_map->buckets[slot]; |
universe@562 | 370 | while (elm != NULL) { |
universe@562 | 371 | struct cx_hash_map_element_s *next = elm->next; |
universe@562 | 372 | size_t new_slot = elm->key.hash % new_bucket_count; |
universe@562 | 373 | |
universe@562 | 374 | // find position where to insert |
universe@562 | 375 | struct cx_hash_map_element_s *bucket_next = new_buckets[new_slot]; |
universe@562 | 376 | struct cx_hash_map_element_s *bucket_prev = NULL; |
universe@562 | 377 | while (bucket_next != NULL && bucket_next->key.hash < elm->key.hash) { |
universe@562 | 378 | bucket_prev = bucket_next; |
universe@562 | 379 | bucket_next = bucket_next->next; |
universe@562 | 380 | } |
universe@562 | 381 | |
universe@562 | 382 | // insert |
universe@562 | 383 | if (bucket_prev == NULL) { |
universe@562 | 384 | elm->next = new_buckets[new_slot]; |
universe@562 | 385 | new_buckets[new_slot] = elm; |
universe@562 | 386 | } else { |
universe@562 | 387 | bucket_prev->next = elm; |
universe@562 | 388 | elm->next = bucket_next; |
universe@562 | 389 | } |
universe@562 | 390 | |
universe@562 | 391 | // advance |
universe@562 | 392 | elm = next; |
universe@562 | 393 | } |
universe@562 | 394 | } |
universe@562 | 395 | |
universe@562 | 396 | // assign result to the map |
universe@562 | 397 | hash_map->bucket_count = new_bucket_count; |
universe@562 | 398 | cxFree(map->allocator, hash_map->buckets); |
universe@562 | 399 | hash_map->buckets = new_buckets; |
universe@562 | 400 | } |
universe@562 | 401 | return 0; |
universe@562 | 402 | } |