1 /*
2 +----------------------------------------------------------------------+
3 | Zend Engine |
4 +----------------------------------------------------------------------+
5 | Copyright (c) 1998-2015 Zend Technologies Ltd. (http://www.zend.com) |
6 +----------------------------------------------------------------------+
7 | This source file is subject to version 2.00 of the Zend license, |
8 | that is bundled with this package in the file LICENSE, and is |
9 | available through the world-wide-web at the following url: |
10 | http://www.zend.com/license/2_00.txt. |
11 | If you did not receive a copy of the Zend license and are unable to |
12 | obtain it through the world-wide-web, please send a note to |
13 | license@zend.com so we can mail you a copy immediately. |
14 +----------------------------------------------------------------------+
15 | Authors: Andi Gutmans <andi@zend.com> |
16 | Zeev Suraski <zeev@zend.com> |
17 +----------------------------------------------------------------------+
18 */
19
20 /* $Id$ */
21
22 #ifndef ZEND_HASH_H
23 #define ZEND_HASH_H
24
25 #include <sys/types.h>
26 #include "zend.h"
27
28 #define HASH_KEY_IS_STRING 1
29 #define HASH_KEY_IS_LONG 2
30 #define HASH_KEY_NON_EXISTENT 3
31 #define HASH_KEY_NON_EXISTANT HASH_KEY_NON_EXISTENT /* Keeping old define (with typo) for backward compatibility */
32
33 #define HASH_UPDATE (1<<0)
34 #define HASH_ADD (1<<1)
35 #define HASH_NEXT_INSERT (1<<2)
36
37 #define HASH_DEL_KEY 0
38 #define HASH_DEL_INDEX 1
39 #define HASH_DEL_KEY_QUICK 2
40
41 #define HASH_UPDATE_KEY_IF_NONE 0
42 #define HASH_UPDATE_KEY_IF_BEFORE 1
43 #define HASH_UPDATE_KEY_IF_AFTER 2
44 #define HASH_UPDATE_KEY_ANYWAY 3
45
46 typedef ulong (*hash_func_t)(const char *arKey, uint nKeyLength);
47 typedef int (*compare_func_t)(const void *, const void * TSRMLS_DC);
48 typedef void (*sort_func_t)(void *, size_t, register size_t, compare_func_t TSRMLS_DC);
49 typedef void (*dtor_func_t)(void *pDest);
50 typedef void (*copy_ctor_func_t)(void *pElement);
51 typedef void (*copy_ctor_param_func_t)(void *pElement, void *pParam);
52
53 struct _hashtable;
54
55 typedef struct bucket {
56 ulong h; /* Used for numeric indexing */
57 uint nKeyLength;
58 void *pData;
59 void *pDataPtr;
60 struct bucket *pListNext;
61 struct bucket *pListLast;
62 struct bucket *pNext;
63 struct bucket *pLast;
64 const char *arKey;
65 } Bucket;
66
67 typedef struct _hashtable {
68 uint nTableSize;
69 uint nTableMask;
70 uint nNumOfElements;
71 ulong nNextFreeElement;
72 Bucket *pInternalPointer; /* Used for element traversal */
73 Bucket *pListHead;
74 Bucket *pListTail;
75 Bucket **arBuckets;
76 dtor_func_t pDestructor;
77 zend_bool persistent;
78 unsigned char nApplyCount;
79 zend_bool bApplyProtection;
80 #if ZEND_DEBUG
81 int inconsistent;
82 #endif
83 } HashTable;
84
85
86 typedef struct _zend_hash_key {
87 const char *arKey;
88 uint nKeyLength;
89 ulong h;
90 } zend_hash_key;
91
92
93 typedef zend_bool (*merge_checker_func_t)(HashTable *target_ht, void *source_data, zend_hash_key *hash_key, void *pParam);
94
95 typedef Bucket* HashPosition;
96
97 BEGIN_EXTERN_C()
98
99 /* startup/shutdown */
100 ZEND_API int _zend_hash_init(HashTable *ht, uint nSize, hash_func_t pHashFunction, dtor_func_t pDestructor, zend_bool persistent ZEND_FILE_LINE_DC);
101 ZEND_API int _zend_hash_init_ex(HashTable *ht, uint nSize, hash_func_t pHashFunction, dtor_func_t pDestructor, zend_bool persistent, zend_bool bApplyProtection ZEND_FILE_LINE_DC);
102 ZEND_API void zend_hash_destroy(HashTable *ht);
103 ZEND_API void zend_hash_clean(HashTable *ht);
104 #define zend_hash_init(ht, nSize, pHashFunction, pDestructor, persistent) _zend_hash_init((ht), (nSize), (pHashFunction), (pDestructor), (persistent) ZEND_FILE_LINE_CC)
105 #define zend_hash_init_ex(ht, nSize, pHashFunction, pDestructor, persistent, bApplyProtection) _zend_hash_init_ex((ht), (nSize), (pHashFunction), (pDestructor), (persistent), (bApplyProtection) ZEND_FILE_LINE_CC)
106
107 /* additions/updates/changes */
108 ZEND_API int _zend_hash_add_or_update(HashTable *ht, const char *arKey, uint nKeyLength, void *pData, uint nDataSize, void **pDest, int flag ZEND_FILE_LINE_DC);
109 #define zend_hash_update(ht, arKey, nKeyLength, pData, nDataSize, pDest) \
110 _zend_hash_add_or_update(ht, arKey, nKeyLength, pData, nDataSize, pDest, HASH_UPDATE ZEND_FILE_LINE_CC)
111 #define zend_hash_add(ht, arKey, nKeyLength, pData, nDataSize, pDest) \
112 _zend_hash_add_or_update(ht, arKey, nKeyLength, pData, nDataSize, pDest, HASH_ADD ZEND_FILE_LINE_CC)
113
114 ZEND_API int _zend_hash_quick_add_or_update(HashTable *ht, const char *arKey, uint nKeyLength, ulong h, void *pData, uint nDataSize, void **pDest, int flag ZEND_FILE_LINE_DC);
115 #define zend_hash_quick_update(ht, arKey, nKeyLength, h, pData, nDataSize, pDest) \
116 _zend_hash_quick_add_or_update(ht, arKey, nKeyLength, h, pData, nDataSize, pDest, HASH_UPDATE ZEND_FILE_LINE_CC)
117 #define zend_hash_quick_add(ht, arKey, nKeyLength, h, pData, nDataSize, pDest) \
118 _zend_hash_quick_add_or_update(ht, arKey, nKeyLength, h, pData, nDataSize, pDest, HASH_ADD ZEND_FILE_LINE_CC)
119
120 ZEND_API int _zend_hash_index_update_or_next_insert(HashTable *ht, ulong h, void *pData, uint nDataSize, void **pDest, int flag ZEND_FILE_LINE_DC);
121 #define zend_hash_index_update(ht, h, pData, nDataSize, pDest) \
122 _zend_hash_index_update_or_next_insert(ht, h, pData, nDataSize, pDest, HASH_UPDATE ZEND_FILE_LINE_CC)
123 #define zend_hash_next_index_insert(ht, pData, nDataSize, pDest) \
124 _zend_hash_index_update_or_next_insert(ht, 0, pData, nDataSize, pDest, HASH_NEXT_INSERT ZEND_FILE_LINE_CC)
125
126 ZEND_API int zend_hash_add_empty_element(HashTable *ht, const char *arKey, uint nKeyLength);
127
128
129 #define ZEND_HASH_APPLY_KEEP 0
130 #define ZEND_HASH_APPLY_REMOVE 1<<0
131 #define ZEND_HASH_APPLY_STOP 1<<1
132
133 typedef int (*apply_func_t)(void *pDest TSRMLS_DC);
134 typedef int (*apply_func_arg_t)(void *pDest, void *argument TSRMLS_DC);
135 typedef int (*apply_func_args_t)(void *pDest TSRMLS_DC, int num_args, va_list args, zend_hash_key *hash_key);
136
137 ZEND_API void zend_hash_graceful_destroy(HashTable *ht);
138 ZEND_API void zend_hash_graceful_reverse_destroy(HashTable *ht);
139 ZEND_API void zend_hash_apply(HashTable *ht, apply_func_t apply_func TSRMLS_DC);
140 ZEND_API void zend_hash_apply_with_argument(HashTable *ht, apply_func_arg_t apply_func, void * TSRMLS_DC);
141 ZEND_API void zend_hash_apply_with_arguments(HashTable *ht TSRMLS_DC, apply_func_args_t apply_func, int, ...);
142
143 /* This function should be used with special care (in other words,
144 * it should usually not be used). When used with the ZEND_HASH_APPLY_STOP
145 * return value, it assumes things about the order of the elements in the hash.
146 * Also, it does not provide the same kind of reentrancy protection that
147 * the standard apply functions do.
148 */
149 ZEND_API void zend_hash_reverse_apply(HashTable *ht, apply_func_t apply_func TSRMLS_DC);
150
151
152 /* Deletes */
153 ZEND_API int zend_hash_del_key_or_index(HashTable *ht, const char *arKey, uint nKeyLength, ulong h, int flag);
154 #define zend_hash_del(ht, arKey, nKeyLength) \
155 zend_hash_del_key_or_index(ht, arKey, nKeyLength, 0, HASH_DEL_KEY)
156 #define zend_hash_quick_del(ht, arKey, nKeyLength, h) \
157 zend_hash_del_key_or_index(ht, arKey, nKeyLength, h, HASH_DEL_KEY_QUICK)
158 #define zend_hash_index_del(ht, h) \
159 zend_hash_del_key_or_index(ht, NULL, 0, h, HASH_DEL_INDEX)
160
161 ZEND_API ulong zend_get_hash_value(const char *arKey, uint nKeyLength);
162
163 /* Data retreival */
164 ZEND_API int zend_hash_find(const HashTable *ht, const char *arKey, uint nKeyLength, void **pData);
165 ZEND_API int zend_hash_quick_find(const HashTable *ht, const char *arKey, uint nKeyLength, ulong h, void **pData);
166 ZEND_API int zend_hash_index_find(const HashTable *ht, ulong h, void **pData);
167
168 /* Misc */
169 ZEND_API int zend_hash_exists(const HashTable *ht, const char *arKey, uint nKeyLength);
170 ZEND_API int zend_hash_quick_exists(const HashTable *ht, const char *arKey, uint nKeyLength, ulong h);
171 ZEND_API int zend_hash_index_exists(const HashTable *ht, ulong h);
172 ZEND_API ulong zend_hash_next_free_element(const HashTable *ht);
173
174 /* traversing */
175 #define zend_hash_has_more_elements_ex(ht, pos) \
176 (zend_hash_get_current_key_type_ex(ht, pos) == HASH_KEY_NON_EXISTENT ? FAILURE : SUCCESS)
177 ZEND_API int zend_hash_move_forward_ex(HashTable *ht, HashPosition *pos);
178 ZEND_API int zend_hash_move_backwards_ex(HashTable *ht, HashPosition *pos);
179 ZEND_API int zend_hash_get_current_key_ex(const HashTable *ht, char **str_index, uint *str_length, ulong *num_index, zend_bool duplicate, HashPosition *pos);
180 ZEND_API void zend_hash_get_current_key_zval_ex(const HashTable *ht, zval *key, HashPosition *pos);
181 ZEND_API int zend_hash_get_current_key_type_ex(HashTable *ht, HashPosition *pos);
182 ZEND_API int zend_hash_get_current_data_ex(HashTable *ht, void **pData, HashPosition *pos);
183 ZEND_API void zend_hash_internal_pointer_reset_ex(HashTable *ht, HashPosition *pos);
184 ZEND_API void zend_hash_internal_pointer_end_ex(HashTable *ht, HashPosition *pos);
185 ZEND_API int zend_hash_update_current_key_ex(HashTable *ht, int key_type, const char *str_index, uint str_length, ulong num_index, int mode, HashPosition *pos);
186
187 typedef struct _HashPointer {
188 HashPosition pos;
189 ulong h;
190 } HashPointer;
191
192 ZEND_API int zend_hash_get_pointer(const HashTable *ht, HashPointer *ptr);
193 ZEND_API int zend_hash_set_pointer(HashTable *ht, const HashPointer *ptr);
194
195 #define zend_hash_has_more_elements(ht) \
196 zend_hash_has_more_elements_ex(ht, NULL)
197 #define zend_hash_move_forward(ht) \
198 zend_hash_move_forward_ex(ht, NULL)
199 #define zend_hash_move_backwards(ht) \
200 zend_hash_move_backwards_ex(ht, NULL)
201 #define zend_hash_get_current_key(ht, str_index, num_index, duplicate) \
202 zend_hash_get_current_key_ex(ht, str_index, NULL, num_index, duplicate, NULL)
203 #define zend_hash_get_current_key_zval(ht, key) \
204 zend_hash_get_current_key_zval_ex(ht, key, NULL)
205 #define zend_hash_get_current_key_type(ht) \
206 zend_hash_get_current_key_type_ex(ht, NULL)
207 #define zend_hash_get_current_data(ht, pData) \
208 zend_hash_get_current_data_ex(ht, pData, NULL)
209 #define zend_hash_internal_pointer_reset(ht) \
210 zend_hash_internal_pointer_reset_ex(ht, NULL)
211 #define zend_hash_internal_pointer_end(ht) \
212 zend_hash_internal_pointer_end_ex(ht, NULL)
213 #define zend_hash_update_current_key(ht, key_type, str_index, str_length, num_index) \
214 zend_hash_update_current_key_ex(ht, key_type, str_index, str_length, num_index, HASH_UPDATE_KEY_ANYWAY, NULL)
215
216 /* Copying, merging and sorting */
217 ZEND_API void zend_hash_copy(HashTable *target, HashTable *source, copy_ctor_func_t pCopyConstructor, void *tmp, uint size);
218 ZEND_API void _zend_hash_merge(HashTable *target, HashTable *source, copy_ctor_func_t pCopyConstructor, void *tmp, uint size, int overwrite ZEND_FILE_LINE_DC);
219 ZEND_API void zend_hash_merge_ex(HashTable *target, HashTable *source, copy_ctor_func_t pCopyConstructor, uint size, merge_checker_func_t pMergeSource, void *pParam);
220 ZEND_API int zend_hash_sort(HashTable *ht, sort_func_t sort_func, compare_func_t compare_func, int renumber TSRMLS_DC);
221 ZEND_API int zend_hash_compare(HashTable *ht1, HashTable *ht2, compare_func_t compar, zend_bool ordered TSRMLS_DC);
222 ZEND_API int zend_hash_minmax(const HashTable *ht, compare_func_t compar, int flag, void **pData TSRMLS_DC);
223
224 #define zend_hash_merge(target, source, pCopyConstructor, tmp, size, overwrite) \
225 _zend_hash_merge(target, source, pCopyConstructor, tmp, size, overwrite ZEND_FILE_LINE_CC)
226
227 ZEND_API int zend_hash_num_elements(const HashTable *ht);
228
229 ZEND_API int zend_hash_rehash(HashTable *ht);
230
231 /*
232 * DJBX33A (Daniel J. Bernstein, Times 33 with Addition)
233 *
234 * This is Daniel J. Bernstein's popular `times 33' hash function as
235 * posted by him years ago on comp.lang.c. It basically uses a function
236 * like ``hash(i) = hash(i-1) * 33 + str[i]''. This is one of the best
237 * known hash functions for strings. Because it is both computed very
238 * fast and distributes very well.
239 *
240 * The magic of number 33, i.e. why it works better than many other
241 * constants, prime or not, has never been adequately explained by
242 * anyone. So I try an explanation: if one experimentally tests all
243 * multipliers between 1 and 256 (as RSE did now) one detects that even
244 * numbers are not useable at all. The remaining 128 odd numbers
245 * (except for the number 1) work more or less all equally well. They
246 * all distribute in an acceptable way and this way fill a hash table
247 * with an average percent of approx. 86%.
248 *
249 * If one compares the Chi^2 values of the variants, the number 33 not
250 * even has the best value. But the number 33 and a few other equally
251 * good numbers like 17, 31, 63, 127 and 129 have nevertheless a great
252 * advantage to the remaining numbers in the large set of possible
253 * multipliers: their multiply operation can be replaced by a faster
254 * operation based on just one shift plus either a single addition
255 * or subtraction operation. And because a hash function has to both
256 * distribute good _and_ has to be very fast to compute, those few
257 * numbers should be preferred and seems to be the reason why Daniel J.
258 * Bernstein also preferred it.
259 *
260 *
261 * -- Ralf S. Engelschall <rse@engelschall.com>
262 */
263
zend_inline_hash_func(const char * arKey,uint nKeyLength)264 static inline ulong zend_inline_hash_func(const char *arKey, uint nKeyLength)
265 {
266 register ulong hash = 5381;
267
268 /* variant with the hash unrolled eight times */
269 for (; nKeyLength >= 8; nKeyLength -= 8) {
270 hash = ((hash << 5) + hash) + *arKey++;
271 hash = ((hash << 5) + hash) + *arKey++;
272 hash = ((hash << 5) + hash) + *arKey++;
273 hash = ((hash << 5) + hash) + *arKey++;
274 hash = ((hash << 5) + hash) + *arKey++;
275 hash = ((hash << 5) + hash) + *arKey++;
276 hash = ((hash << 5) + hash) + *arKey++;
277 hash = ((hash << 5) + hash) + *arKey++;
278 }
279 switch (nKeyLength) {
280 case 7: hash = ((hash << 5) + hash) + *arKey++; /* fallthrough... */
281 case 6: hash = ((hash << 5) + hash) + *arKey++; /* fallthrough... */
282 case 5: hash = ((hash << 5) + hash) + *arKey++; /* fallthrough... */
283 case 4: hash = ((hash << 5) + hash) + *arKey++; /* fallthrough... */
284 case 3: hash = ((hash << 5) + hash) + *arKey++; /* fallthrough... */
285 case 2: hash = ((hash << 5) + hash) + *arKey++; /* fallthrough... */
286 case 1: hash = ((hash << 5) + hash) + *arKey++; break;
287 case 0: break;
288 EMPTY_SWITCH_DEFAULT_CASE()
289 }
290 return hash;
291 }
292
293
294 ZEND_API ulong zend_hash_func(const char *arKey, uint nKeyLength);
295
296 #if ZEND_DEBUG
297 /* debug */
298 void zend_hash_display_pListTail(const HashTable *ht);
299 void zend_hash_display(const HashTable *ht);
300 #endif
301
END_EXTERN_C()302 END_EXTERN_C()
303
304 #define ZEND_INIT_SYMTABLE(ht) \
305 ZEND_INIT_SYMTABLE_EX(ht, 2, 0)
306
307 #define ZEND_INIT_SYMTABLE_EX(ht, n, persistent) \
308 zend_hash_init(ht, n, NULL, ZVAL_PTR_DTOR, persistent)
309
310 #define ZEND_HANDLE_NUMERIC_EX(key, length, idx, func) do { \
311 register const char *tmp = key; \
312 \
313 if (*tmp == '-') { \
314 tmp++; \
315 } \
316 if (*tmp >= '0' && *tmp <= '9') { /* possibly a numeric index */ \
317 const char *end = key + length - 1; \
318 \
319 if ((*end != '\0') /* not a null terminated string */ \
320 || (*tmp == '0' && length > 2) /* numbers with leading zeros */ \
321 || (end - tmp > MAX_LENGTH_OF_LONG - 1) /* number too long */ \
322 || (SIZEOF_LONG == 4 && \
323 end - tmp == MAX_LENGTH_OF_LONG - 1 && \
324 *tmp > '2')) { /* overflow */ \
325 break; \
326 } \
327 idx = (*tmp - '0'); \
328 while (++tmp != end && *tmp >= '0' && *tmp <= '9') { \
329 idx = (idx * 10) + (*tmp - '0'); \
330 } \
331 if (tmp == end) { \
332 if (*key == '-') { \
333 if (idx-1 > LONG_MAX) { /* overflow */ \
334 break; \
335 } \
336 idx = 0 - idx; \
337 } else if (idx > LONG_MAX) { /* overflow */ \
338 break; \
339 } \
340 func; \
341 } \
342 } \
343 } while (0)
344
345 #define ZEND_HANDLE_NUMERIC(key, length, func) do { \
346 ulong idx; \
347 \
348 ZEND_HANDLE_NUMERIC_EX(key, length, idx, return func); \
349 } while (0)
350
351 static inline int zend_symtable_update(HashTable *ht, const char *arKey, uint nKeyLength, void *pData, uint nDataSize, void **pDest) \
352 {
353 ZEND_HANDLE_NUMERIC(arKey, nKeyLength, zend_hash_index_update(ht, idx, pData, nDataSize, pDest));
354 return zend_hash_update(ht, arKey, nKeyLength, pData, nDataSize, pDest);
355 }
356
357
zend_symtable_del(HashTable * ht,const char * arKey,uint nKeyLength)358 static inline int zend_symtable_del(HashTable *ht, const char *arKey, uint nKeyLength)
359 {
360 ZEND_HANDLE_NUMERIC(arKey, nKeyLength, zend_hash_index_del(ht, idx));
361 return zend_hash_del(ht, arKey, nKeyLength);
362 }
363
364
zend_symtable_find(HashTable * ht,const char * arKey,uint nKeyLength,void ** pData)365 static inline int zend_symtable_find(HashTable *ht, const char *arKey, uint nKeyLength, void **pData)
366 {
367 ZEND_HANDLE_NUMERIC(arKey, nKeyLength, zend_hash_index_find(ht, idx, pData));
368 return zend_hash_find(ht, arKey, nKeyLength, pData);
369 }
370
371
zend_symtable_exists(HashTable * ht,const char * arKey,uint nKeyLength)372 static inline int zend_symtable_exists(HashTable *ht, const char *arKey, uint nKeyLength)
373 {
374 ZEND_HANDLE_NUMERIC(arKey, nKeyLength, zend_hash_index_exists(ht, idx));
375 return zend_hash_exists(ht, arKey, nKeyLength);
376 }
377
zend_symtable_update_current_key_ex(HashTable * ht,const char * arKey,uint nKeyLength,int mode,HashPosition * pos)378 static inline int zend_symtable_update_current_key_ex(HashTable *ht, const char *arKey, uint nKeyLength, int mode, HashPosition *pos)
379 {
380 ZEND_HANDLE_NUMERIC(arKey, nKeyLength, zend_hash_update_current_key_ex(ht, HASH_KEY_IS_LONG, NULL, 0, idx, mode, pos));
381 return zend_hash_update_current_key_ex(ht, HASH_KEY_IS_STRING, arKey, nKeyLength, 0, mode, pos);
382 }
383 #define zend_symtable_update_current_key(ht,arKey,nKeyLength,mode) \
384 zend_symtable_update_current_key_ex(ht, arKey, nKeyLength, mode, NULL)
385
386
387 #endif /* ZEND_HASH_H */
388
389 /*
390 * Local variables:
391 * tab-width: 4
392 * c-basic-offset: 4
393 * indent-tabs-mode: t
394 * End:
395 */
396
