/src/tinysparql/subprojects/libxml2-2.13.1/dict.c
Line | Count | Source (jump to first uncovered line) |
1 | | /* |
2 | | * dict.c: dictionary of reusable strings, just used to avoid allocation |
3 | | * and freeing operations. |
4 | | * |
5 | | * Copyright (C) 2003-2012 Daniel Veillard. |
6 | | * |
7 | | * Permission to use, copy, modify, and distribute this software for any |
8 | | * purpose with or without fee is hereby granted, provided that the above |
9 | | * copyright notice and this permission notice appear in all copies. |
10 | | * |
11 | | * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED |
12 | | * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF |
13 | | * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE AUTHORS AND |
14 | | * CONTRIBUTORS ACCEPT NO RESPONSIBILITY IN ANY CONCEIVABLE MANNER. |
15 | | * |
16 | | * Author: daniel@veillard.com |
17 | | */ |
18 | | |
19 | | #define IN_LIBXML |
20 | | #include "libxml.h" |
21 | | |
22 | | #include <errno.h> |
23 | | #include <limits.h> |
24 | | #include <stdlib.h> |
25 | | #include <string.h> |
26 | | |
27 | | #include "private/dict.h" |
28 | | #include "private/globals.h" |
29 | | #include "private/threads.h" |
30 | | |
31 | | #include <libxml/parser.h> |
32 | | #include <libxml/dict.h> |
33 | | #include <libxml/xmlmemory.h> |
34 | | #include <libxml/xmlstring.h> |
35 | | |
36 | | #ifndef SIZE_MAX |
37 | 0 | #define SIZE_MAX ((size_t) -1) |
38 | | #endif |
39 | | |
40 | 0 | #define MAX_FILL_NUM 7 |
41 | 0 | #define MAX_FILL_DENOM 8 |
42 | 0 | #define MIN_HASH_SIZE 8 |
43 | 0 | #define MAX_HASH_SIZE (1u << 31) |
44 | | |
45 | | typedef struct _xmlDictStrings xmlDictStrings; |
46 | | typedef xmlDictStrings *xmlDictStringsPtr; |
47 | | struct _xmlDictStrings { |
48 | | xmlDictStringsPtr next; |
49 | | xmlChar *free; |
50 | | xmlChar *end; |
51 | | size_t size; |
52 | | size_t nbStrings; |
53 | | xmlChar array[1]; |
54 | | }; |
55 | | |
56 | | typedef xmlHashedString xmlDictEntry; |
57 | | |
58 | | /* |
59 | | * The entire dictionary |
60 | | */ |
61 | | struct _xmlDict { |
62 | | int ref_counter; |
63 | | |
64 | | xmlDictEntry *table; |
65 | | size_t size; |
66 | | unsigned int nbElems; |
67 | | xmlDictStringsPtr strings; |
68 | | |
69 | | struct _xmlDict *subdict; |
70 | | /* used for randomization */ |
71 | | unsigned seed; |
72 | | /* used to impose a limit on size */ |
73 | | size_t limit; |
74 | | }; |
75 | | |
76 | | /* |
77 | | * A mutex for modifying the reference counter for shared |
78 | | * dictionaries. |
79 | | */ |
80 | | static xmlMutex xmlDictMutex; |
81 | | |
82 | | /** |
83 | | * xmlInitializeDict: |
84 | | * |
85 | | * DEPRECATED: Alias for xmlInitParser. |
86 | | * |
87 | | * Returns 0. |
88 | | */ |
89 | | int |
90 | 0 | xmlInitializeDict(void) { |
91 | 0 | xmlInitParser(); |
92 | 0 | return(0); |
93 | 0 | } |
94 | | |
95 | | /** |
96 | | * xmlInitDictInternal: |
97 | | * |
98 | | * Initialize mutex. |
99 | | */ |
100 | | void |
101 | 0 | xmlInitDictInternal(void) { |
102 | 0 | xmlInitMutex(&xmlDictMutex); |
103 | 0 | } |
104 | | |
105 | | /** |
106 | | * xmlDictCleanup: |
107 | | * |
108 | | * DEPRECATED: This function is a no-op. Call xmlCleanupParser |
109 | | * to free global state but see the warnings there. xmlCleanupParser |
110 | | * should be only called once at program exit. In most cases, you don't |
111 | | * have call cleanup functions at all. |
112 | | */ |
113 | | void |
114 | 0 | xmlDictCleanup(void) { |
115 | 0 | } |
116 | | |
117 | | /** |
118 | | * xmlCleanupDictInternal: |
119 | | * |
120 | | * Free the dictionary mutex. |
121 | | */ |
122 | | void |
123 | 0 | xmlCleanupDictInternal(void) { |
124 | 0 | xmlCleanupMutex(&xmlDictMutex); |
125 | 0 | } |
126 | | |
127 | | /* |
128 | | * xmlDictAddString: |
129 | | * @dict: the dictionary |
130 | | * @name: the name of the userdata |
131 | | * @len: the length of the name |
132 | | * |
133 | | * Add the string to the array[s] |
134 | | * |
135 | | * Returns the pointer of the local string, or NULL in case of error. |
136 | | */ |
137 | | static const xmlChar * |
138 | 0 | xmlDictAddString(xmlDictPtr dict, const xmlChar *name, unsigned int namelen) { |
139 | 0 | xmlDictStringsPtr pool; |
140 | 0 | const xmlChar *ret; |
141 | 0 | size_t size = 0; /* + sizeof(_xmlDictStrings) == 1024 */ |
142 | 0 | size_t limit = 0; |
143 | |
|
144 | 0 | pool = dict->strings; |
145 | 0 | while (pool != NULL) { |
146 | 0 | if ((size_t)(pool->end - pool->free) > namelen) |
147 | 0 | goto found_pool; |
148 | 0 | if (pool->size > size) size = pool->size; |
149 | 0 | limit += pool->size; |
150 | 0 | pool = pool->next; |
151 | 0 | } |
152 | | /* |
153 | | * Not found, need to allocate |
154 | | */ |
155 | 0 | if (pool == NULL) { |
156 | 0 | if ((dict->limit > 0) && (limit > dict->limit)) { |
157 | 0 | return(NULL); |
158 | 0 | } |
159 | | |
160 | 0 | if (size == 0) { |
161 | 0 | size = 1000; |
162 | 0 | } else { |
163 | 0 | if (size < (SIZE_MAX - sizeof(xmlDictStrings)) / 4) |
164 | 0 | size *= 4; /* exponential growth */ |
165 | 0 | else |
166 | 0 | size = SIZE_MAX - sizeof(xmlDictStrings); |
167 | 0 | } |
168 | 0 | if (size / 4 < namelen) { |
169 | 0 | if ((size_t) namelen + 0 < (SIZE_MAX - sizeof(xmlDictStrings)) / 4) |
170 | 0 | size = 4 * (size_t) namelen; /* just in case ! */ |
171 | 0 | else |
172 | 0 | return(NULL); |
173 | 0 | } |
174 | 0 | pool = (xmlDictStringsPtr) xmlMalloc(sizeof(xmlDictStrings) + size); |
175 | 0 | if (pool == NULL) |
176 | 0 | return(NULL); |
177 | 0 | pool->size = size; |
178 | 0 | pool->nbStrings = 0; |
179 | 0 | pool->free = &pool->array[0]; |
180 | 0 | pool->end = &pool->array[size]; |
181 | 0 | pool->next = dict->strings; |
182 | 0 | dict->strings = pool; |
183 | 0 | } |
184 | 0 | found_pool: |
185 | 0 | ret = pool->free; |
186 | 0 | memcpy(pool->free, name, namelen); |
187 | 0 | pool->free += namelen; |
188 | 0 | *(pool->free++) = 0; |
189 | 0 | pool->nbStrings++; |
190 | 0 | return(ret); |
191 | 0 | } |
192 | | |
193 | | /* |
194 | | * xmlDictAddQString: |
195 | | * @dict: the dictionary |
196 | | * @prefix: the prefix of the userdata |
197 | | * @plen: the prefix length |
198 | | * @name: the name of the userdata |
199 | | * @len: the length of the name |
200 | | * |
201 | | * Add the QName to the array[s] |
202 | | * |
203 | | * Returns the pointer of the local string, or NULL in case of error. |
204 | | */ |
205 | | static const xmlChar * |
206 | | xmlDictAddQString(xmlDictPtr dict, const xmlChar *prefix, unsigned int plen, |
207 | | const xmlChar *name, unsigned int namelen) |
208 | 0 | { |
209 | 0 | xmlDictStringsPtr pool; |
210 | 0 | const xmlChar *ret; |
211 | 0 | size_t size = 0; /* + sizeof(_xmlDictStrings) == 1024 */ |
212 | 0 | size_t limit = 0; |
213 | |
|
214 | 0 | pool = dict->strings; |
215 | 0 | while (pool != NULL) { |
216 | 0 | if ((size_t)(pool->end - pool->free) > namelen + plen + 1) |
217 | 0 | goto found_pool; |
218 | 0 | if (pool->size > size) size = pool->size; |
219 | 0 | limit += pool->size; |
220 | 0 | pool = pool->next; |
221 | 0 | } |
222 | | /* |
223 | | * Not found, need to allocate |
224 | | */ |
225 | 0 | if (pool == NULL) { |
226 | 0 | if ((dict->limit > 0) && (limit > dict->limit)) { |
227 | 0 | return(NULL); |
228 | 0 | } |
229 | | |
230 | 0 | if (size == 0) size = 1000; |
231 | 0 | else size *= 4; /* exponential growth */ |
232 | 0 | if (size < 4 * (namelen + plen + 1)) |
233 | 0 | size = 4 * (namelen + plen + 1); /* just in case ! */ |
234 | 0 | pool = (xmlDictStringsPtr) xmlMalloc(sizeof(xmlDictStrings) + size); |
235 | 0 | if (pool == NULL) |
236 | 0 | return(NULL); |
237 | 0 | pool->size = size; |
238 | 0 | pool->nbStrings = 0; |
239 | 0 | pool->free = &pool->array[0]; |
240 | 0 | pool->end = &pool->array[size]; |
241 | 0 | pool->next = dict->strings; |
242 | 0 | dict->strings = pool; |
243 | 0 | } |
244 | 0 | found_pool: |
245 | 0 | ret = pool->free; |
246 | 0 | memcpy(pool->free, prefix, plen); |
247 | 0 | pool->free += plen; |
248 | 0 | *(pool->free++) = ':'; |
249 | 0 | memcpy(pool->free, name, namelen); |
250 | 0 | pool->free += namelen; |
251 | 0 | *(pool->free++) = 0; |
252 | 0 | pool->nbStrings++; |
253 | 0 | return(ret); |
254 | 0 | } |
255 | | |
256 | | /** |
257 | | * xmlDictCreate: |
258 | | * |
259 | | * Create a new dictionary |
260 | | * |
261 | | * Returns the newly created dictionary, or NULL if an error occurred. |
262 | | */ |
263 | | xmlDictPtr |
264 | 0 | xmlDictCreate(void) { |
265 | 0 | xmlDictPtr dict; |
266 | |
|
267 | 0 | xmlInitParser(); |
268 | |
|
269 | 0 | dict = xmlMalloc(sizeof(xmlDict)); |
270 | 0 | if (dict == NULL) |
271 | 0 | return(NULL); |
272 | 0 | dict->ref_counter = 1; |
273 | 0 | dict->limit = 0; |
274 | |
|
275 | 0 | dict->size = 0; |
276 | 0 | dict->nbElems = 0; |
277 | 0 | dict->table = NULL; |
278 | 0 | dict->strings = NULL; |
279 | 0 | dict->subdict = NULL; |
280 | 0 | dict->seed = xmlRandom(); |
281 | 0 | #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION |
282 | 0 | dict->seed = 0; |
283 | 0 | #endif |
284 | 0 | return(dict); |
285 | 0 | } |
286 | | |
287 | | /** |
288 | | * xmlDictCreateSub: |
289 | | * @sub: an existing dictionary |
290 | | * |
291 | | * Create a new dictionary, inheriting strings from the read-only |
292 | | * dictionary @sub. On lookup, strings are first searched in the |
293 | | * new dictionary, then in @sub, and if not found are created in the |
294 | | * new dictionary. |
295 | | * |
296 | | * Returns the newly created dictionary, or NULL if an error occurred. |
297 | | */ |
298 | | xmlDictPtr |
299 | 0 | xmlDictCreateSub(xmlDictPtr sub) { |
300 | 0 | xmlDictPtr dict = xmlDictCreate(); |
301 | |
|
302 | 0 | if ((dict != NULL) && (sub != NULL)) { |
303 | 0 | dict->seed = sub->seed; |
304 | 0 | dict->subdict = sub; |
305 | 0 | xmlDictReference(dict->subdict); |
306 | 0 | } |
307 | 0 | return(dict); |
308 | 0 | } |
309 | | |
310 | | /** |
311 | | * xmlDictReference: |
312 | | * @dict: the dictionary |
313 | | * |
314 | | * Increment the reference counter of a dictionary |
315 | | * |
316 | | * Returns 0 in case of success and -1 in case of error |
317 | | */ |
318 | | int |
319 | 0 | xmlDictReference(xmlDictPtr dict) { |
320 | 0 | if (dict == NULL) return -1; |
321 | 0 | xmlMutexLock(&xmlDictMutex); |
322 | 0 | dict->ref_counter++; |
323 | 0 | xmlMutexUnlock(&xmlDictMutex); |
324 | 0 | return(0); |
325 | 0 | } |
326 | | |
327 | | /** |
328 | | * xmlDictFree: |
329 | | * @dict: the dictionary |
330 | | * |
331 | | * Free the hash @dict and its contents. The userdata is |
332 | | * deallocated with @f if provided. |
333 | | */ |
334 | | void |
335 | 0 | xmlDictFree(xmlDictPtr dict) { |
336 | 0 | xmlDictStringsPtr pool, nextp; |
337 | |
|
338 | 0 | if (dict == NULL) |
339 | 0 | return; |
340 | | |
341 | | /* decrement the counter, it may be shared by a parser and docs */ |
342 | 0 | xmlMutexLock(&xmlDictMutex); |
343 | 0 | dict->ref_counter--; |
344 | 0 | if (dict->ref_counter > 0) { |
345 | 0 | xmlMutexUnlock(&xmlDictMutex); |
346 | 0 | return; |
347 | 0 | } |
348 | | |
349 | 0 | xmlMutexUnlock(&xmlDictMutex); |
350 | |
|
351 | 0 | if (dict->subdict != NULL) { |
352 | 0 | xmlDictFree(dict->subdict); |
353 | 0 | } |
354 | |
|
355 | 0 | if (dict->table) { |
356 | 0 | xmlFree(dict->table); |
357 | 0 | } |
358 | 0 | pool = dict->strings; |
359 | 0 | while (pool != NULL) { |
360 | 0 | nextp = pool->next; |
361 | 0 | xmlFree(pool); |
362 | 0 | pool = nextp; |
363 | 0 | } |
364 | 0 | xmlFree(dict); |
365 | 0 | } |
366 | | |
367 | | /** |
368 | | * xmlDictOwns: |
369 | | * @dict: the dictionary |
370 | | * @str: the string |
371 | | * |
372 | | * check if a string is owned by the dictionary |
373 | | * |
374 | | * Returns 1 if true, 0 if false and -1 in case of error |
375 | | * -1 in case of error |
376 | | */ |
377 | | int |
378 | 0 | xmlDictOwns(xmlDictPtr dict, const xmlChar *str) { |
379 | 0 | xmlDictStringsPtr pool; |
380 | |
|
381 | 0 | if ((dict == NULL) || (str == NULL)) |
382 | 0 | return(-1); |
383 | 0 | pool = dict->strings; |
384 | 0 | while (pool != NULL) { |
385 | 0 | if ((str >= &pool->array[0]) && (str <= pool->free)) |
386 | 0 | return(1); |
387 | 0 | pool = pool->next; |
388 | 0 | } |
389 | 0 | if (dict->subdict) |
390 | 0 | return(xmlDictOwns(dict->subdict, str)); |
391 | 0 | return(0); |
392 | 0 | } |
393 | | |
394 | | /** |
395 | | * xmlDictSize: |
396 | | * @dict: the dictionary |
397 | | * |
398 | | * Query the number of elements installed in the hash @dict. |
399 | | * |
400 | | * Returns the number of elements in the dictionary or |
401 | | * -1 in case of error |
402 | | */ |
403 | | int |
404 | 0 | xmlDictSize(xmlDictPtr dict) { |
405 | 0 | if (dict == NULL) |
406 | 0 | return(-1); |
407 | 0 | if (dict->subdict) |
408 | 0 | return(dict->nbElems + dict->subdict->nbElems); |
409 | 0 | return(dict->nbElems); |
410 | 0 | } |
411 | | |
412 | | /** |
413 | | * xmlDictSetLimit: |
414 | | * @dict: the dictionary |
415 | | * @limit: the limit in bytes |
416 | | * |
417 | | * Set a size limit for the dictionary |
418 | | * Added in 2.9.0 |
419 | | * |
420 | | * Returns the previous limit of the dictionary or 0 |
421 | | */ |
422 | | size_t |
423 | 0 | xmlDictSetLimit(xmlDictPtr dict, size_t limit) { |
424 | 0 | size_t ret; |
425 | |
|
426 | 0 | if (dict == NULL) |
427 | 0 | return(0); |
428 | 0 | ret = dict->limit; |
429 | 0 | dict->limit = limit; |
430 | 0 | return(ret); |
431 | 0 | } |
432 | | |
433 | | /** |
434 | | * xmlDictGetUsage: |
435 | | * @dict: the dictionary |
436 | | * |
437 | | * Get how much memory is used by a dictionary for strings |
438 | | * Added in 2.9.0 |
439 | | * |
440 | | * Returns the amount of strings allocated |
441 | | */ |
442 | | size_t |
443 | 0 | xmlDictGetUsage(xmlDictPtr dict) { |
444 | 0 | xmlDictStringsPtr pool; |
445 | 0 | size_t limit = 0; |
446 | |
|
447 | 0 | if (dict == NULL) |
448 | 0 | return(0); |
449 | 0 | pool = dict->strings; |
450 | 0 | while (pool != NULL) { |
451 | 0 | limit += pool->size; |
452 | 0 | pool = pool->next; |
453 | 0 | } |
454 | 0 | return(limit); |
455 | 0 | } |
456 | | |
457 | | /***************************************************************** |
458 | | * |
459 | | * The code below was rewritten and is additionally licensed under |
460 | | * the main license in file 'Copyright'. |
461 | | * |
462 | | *****************************************************************/ |
463 | | |
464 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
465 | | static unsigned |
466 | | xmlDictHashName(unsigned seed, const xmlChar* data, size_t maxLen, |
467 | 0 | size_t *plen) { |
468 | 0 | unsigned h1, h2; |
469 | 0 | size_t i; |
470 | |
|
471 | 0 | HASH_INIT(h1, h2, seed); |
472 | |
|
473 | 0 | for (i = 0; i < maxLen && data[i]; i++) { |
474 | 0 | HASH_UPDATE(h1, h2, data[i]); |
475 | 0 | } |
476 | |
|
477 | 0 | HASH_FINISH(h1, h2); |
478 | |
|
479 | 0 | *plen = i; |
480 | 0 | return(h2 | MAX_HASH_SIZE); |
481 | 0 | } |
482 | | |
483 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
484 | | static unsigned |
485 | | xmlDictHashQName(unsigned seed, const xmlChar *prefix, const xmlChar *name, |
486 | 0 | size_t *pplen, size_t *plen) { |
487 | 0 | unsigned h1, h2; |
488 | 0 | size_t i; |
489 | |
|
490 | 0 | HASH_INIT(h1, h2, seed); |
491 | |
|
492 | 0 | for (i = 0; prefix[i] != 0; i++) { |
493 | 0 | HASH_UPDATE(h1, h2, prefix[i]); |
494 | 0 | } |
495 | 0 | *pplen = i; |
496 | |
|
497 | 0 | HASH_UPDATE(h1, h2, ':'); |
498 | |
|
499 | 0 | for (i = 0; name[i] != 0; i++) { |
500 | 0 | HASH_UPDATE(h1, h2, name[i]); |
501 | 0 | } |
502 | 0 | *plen = i; |
503 | |
|
504 | 0 | HASH_FINISH(h1, h2); |
505 | | |
506 | | /* |
507 | | * Always set the upper bit of hash values since 0 means an unoccupied |
508 | | * bucket. |
509 | | */ |
510 | 0 | return(h2 | MAX_HASH_SIZE); |
511 | 0 | } |
512 | | |
513 | | /** |
514 | | * xmlDictComputeHash: |
515 | | * @dict: dictionary |
516 | | * @string: C string |
517 | | * |
518 | | * Compute the hash value of a C string. |
519 | | * |
520 | | * Returns the hash value. |
521 | | */ |
522 | | unsigned |
523 | 0 | xmlDictComputeHash(const xmlDict *dict, const xmlChar *string) { |
524 | 0 | size_t len; |
525 | 0 | return(xmlDictHashName(dict->seed, string, SIZE_MAX, &len)); |
526 | 0 | } |
527 | | |
528 | 0 | #define HASH_ROL31(x,n) ((x) << (n) | ((x) & 0x7FFFFFFF) >> (31 - (n))) |
529 | | |
530 | | /** |
531 | | * xmlDictCombineHash: |
532 | | * @v1: first hash value |
533 | | * @v2: second hash value |
534 | | * |
535 | | * Combine two hash values. |
536 | | * |
537 | | * Returns the combined hash value. |
538 | | */ |
539 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
540 | | unsigned |
541 | 0 | xmlDictCombineHash(unsigned v1, unsigned v2) { |
542 | | /* |
543 | | * The upper bit of hash values is always set, so we have to operate on |
544 | | * 31-bit hashes here. |
545 | | */ |
546 | 0 | v1 ^= v2; |
547 | 0 | v1 += HASH_ROL31(v2, 5); |
548 | |
|
549 | 0 | return((v1 & 0xFFFFFFFF) | 0x80000000); |
550 | 0 | } |
551 | | |
552 | | /** |
553 | | * xmlDictFindEntry: |
554 | | * @dict: dict |
555 | | * @prefix: optional QName prefix |
556 | | * @name: string |
557 | | * @len: length of string |
558 | | * @hashValue: valid hash value of string |
559 | | * @pfound: result of search |
560 | | * |
561 | | * Try to find a matching hash table entry. If an entry was found, set |
562 | | * @found to 1 and return the entry. Otherwise, set @found to 0 and return |
563 | | * the location where a new entry should be inserted. |
564 | | */ |
565 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
566 | | static xmlDictEntry * |
567 | | xmlDictFindEntry(const xmlDict *dict, const xmlChar *prefix, |
568 | | const xmlChar *name, int len, unsigned hashValue, |
569 | 0 | int *pfound) { |
570 | 0 | xmlDictEntry *entry; |
571 | 0 | unsigned mask, pos, displ; |
572 | 0 | int found = 0; |
573 | |
|
574 | 0 | mask = dict->size - 1; |
575 | 0 | pos = hashValue & mask; |
576 | 0 | entry = &dict->table[pos]; |
577 | |
|
578 | 0 | if (entry->hashValue != 0) { |
579 | | /* |
580 | | * Robin hood hashing: abort if the displacement of the entry |
581 | | * is smaller than the displacement of the key we look for. |
582 | | * This also stops at the correct position when inserting. |
583 | | */ |
584 | 0 | displ = 0; |
585 | |
|
586 | 0 | do { |
587 | 0 | if (entry->hashValue == hashValue) { |
588 | 0 | if (prefix == NULL) { |
589 | | /* |
590 | | * name is not necessarily null-terminated. |
591 | | */ |
592 | 0 | if ((strncmp((const char *) entry->name, |
593 | 0 | (const char *) name, len) == 0) && |
594 | 0 | (entry->name[len] == 0)) { |
595 | 0 | found = 1; |
596 | 0 | break; |
597 | 0 | } |
598 | 0 | } else { |
599 | 0 | if (xmlStrQEqual(prefix, name, entry->name)) { |
600 | 0 | found = 1; |
601 | 0 | break; |
602 | 0 | } |
603 | 0 | } |
604 | 0 | } |
605 | | |
606 | 0 | displ++; |
607 | 0 | pos++; |
608 | 0 | entry++; |
609 | 0 | if ((pos & mask) == 0) |
610 | 0 | entry = dict->table; |
611 | 0 | } while ((entry->hashValue != 0) && |
612 | 0 | (((pos - entry->hashValue) & mask) >= displ)); |
613 | 0 | } |
614 | | |
615 | 0 | *pfound = found; |
616 | 0 | return(entry); |
617 | 0 | } |
618 | | |
619 | | /** |
620 | | * xmlDictGrow: |
621 | | * @dict: dictionary |
622 | | * @size: new size of the dictionary |
623 | | * |
624 | | * Resize the dictionary hash table. |
625 | | * |
626 | | * Returns 0 in case of success, -1 if a memory allocation failed. |
627 | | */ |
628 | | static int |
629 | 0 | xmlDictGrow(xmlDictPtr dict, unsigned size) { |
630 | 0 | const xmlDictEntry *oldentry, *oldend, *end; |
631 | 0 | xmlDictEntry *table; |
632 | 0 | unsigned oldsize, i; |
633 | | |
634 | | /* Add 0 to avoid spurious -Wtype-limits warning on 64-bit GCC */ |
635 | 0 | if ((size_t) size + 0 > SIZE_MAX / sizeof(table[0])) |
636 | 0 | return(-1); |
637 | 0 | table = xmlMalloc(size * sizeof(table[0])); |
638 | 0 | if (table == NULL) |
639 | 0 | return(-1); |
640 | 0 | memset(table, 0, size * sizeof(table[0])); |
641 | |
|
642 | 0 | oldsize = dict->size; |
643 | 0 | if (oldsize == 0) |
644 | 0 | goto done; |
645 | | |
646 | 0 | oldend = &dict->table[oldsize]; |
647 | 0 | end = &table[size]; |
648 | | |
649 | | /* |
650 | | * Robin Hood sorting order is maintained if we |
651 | | * |
652 | | * - compute dict indices with modulo |
653 | | * - resize by an integer factor |
654 | | * - start to copy from the beginning of a probe sequence |
655 | | */ |
656 | 0 | oldentry = dict->table; |
657 | 0 | while (oldentry->hashValue != 0) { |
658 | 0 | if (++oldentry >= oldend) |
659 | 0 | oldentry = dict->table; |
660 | 0 | } |
661 | |
|
662 | 0 | for (i = 0; i < oldsize; i++) { |
663 | 0 | if (oldentry->hashValue != 0) { |
664 | 0 | xmlDictEntry *entry = &table[oldentry->hashValue & (size - 1)]; |
665 | |
|
666 | 0 | while (entry->hashValue != 0) { |
667 | 0 | if (++entry >= end) |
668 | 0 | entry = table; |
669 | 0 | } |
670 | 0 | *entry = *oldentry; |
671 | 0 | } |
672 | |
|
673 | 0 | if (++oldentry >= oldend) |
674 | 0 | oldentry = dict->table; |
675 | 0 | } |
676 | |
|
677 | 0 | xmlFree(dict->table); |
678 | |
|
679 | 0 | done: |
680 | 0 | dict->table = table; |
681 | 0 | dict->size = size; |
682 | |
|
683 | 0 | return(0); |
684 | 0 | } |
685 | | |
686 | | /** |
687 | | * xmlDictLookupInternal: |
688 | | * @dict: dict |
689 | | * @prefix: optional QName prefix |
690 | | * @name: string |
691 | | * @maybeLen: length of string or -1 if unknown |
692 | | * @update: whether the string should be added |
693 | | * |
694 | | * Internal lookup and update function. |
695 | | */ |
696 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
697 | | static const xmlDictEntry * |
698 | | xmlDictLookupInternal(xmlDictPtr dict, const xmlChar *prefix, |
699 | 0 | const xmlChar *name, int maybeLen, int update) { |
700 | 0 | xmlDictEntry *entry = NULL; |
701 | 0 | const xmlChar *ret; |
702 | 0 | unsigned hashValue; |
703 | 0 | size_t maxLen, len, plen, klen; |
704 | 0 | int found = 0; |
705 | |
|
706 | 0 | if ((dict == NULL) || (name == NULL)) |
707 | 0 | return(NULL); |
708 | | |
709 | 0 | maxLen = (maybeLen < 0) ? SIZE_MAX : (size_t) maybeLen; |
710 | |
|
711 | 0 | if (prefix == NULL) { |
712 | 0 | hashValue = xmlDictHashName(dict->seed, name, maxLen, &len); |
713 | 0 | if (len > INT_MAX / 2) |
714 | 0 | return(NULL); |
715 | 0 | klen = len; |
716 | 0 | } else { |
717 | 0 | hashValue = xmlDictHashQName(dict->seed, prefix, name, &plen, &len); |
718 | 0 | if ((len > INT_MAX / 2) || (plen >= INT_MAX / 2 - len)) |
719 | 0 | return(NULL); |
720 | 0 | klen = plen + 1 + len; |
721 | 0 | } |
722 | | |
723 | 0 | if ((dict->limit > 0) && (klen >= dict->limit)) |
724 | 0 | return(NULL); |
725 | | |
726 | | /* |
727 | | * Check for an existing entry |
728 | | */ |
729 | 0 | if (dict->size > 0) |
730 | 0 | entry = xmlDictFindEntry(dict, prefix, name, klen, hashValue, &found); |
731 | 0 | if (found) |
732 | 0 | return(entry); |
733 | | |
734 | 0 | if ((dict->subdict != NULL) && (dict->subdict->size > 0)) { |
735 | 0 | xmlDictEntry *subEntry; |
736 | 0 | unsigned subHashValue; |
737 | |
|
738 | 0 | if (prefix == NULL) |
739 | 0 | subHashValue = xmlDictHashName(dict->subdict->seed, name, len, |
740 | 0 | &len); |
741 | 0 | else |
742 | 0 | subHashValue = xmlDictHashQName(dict->subdict->seed, prefix, name, |
743 | 0 | &plen, &len); |
744 | 0 | subEntry = xmlDictFindEntry(dict->subdict, prefix, name, klen, |
745 | 0 | subHashValue, &found); |
746 | 0 | if (found) |
747 | 0 | return(subEntry); |
748 | 0 | } |
749 | | |
750 | 0 | if (!update) |
751 | 0 | return(NULL); |
752 | | |
753 | | /* |
754 | | * Grow the hash table if needed |
755 | | */ |
756 | 0 | if (dict->nbElems + 1 > dict->size / MAX_FILL_DENOM * MAX_FILL_NUM) { |
757 | 0 | unsigned newSize, mask, displ, pos; |
758 | |
|
759 | 0 | if (dict->size == 0) { |
760 | 0 | newSize = MIN_HASH_SIZE; |
761 | 0 | } else { |
762 | 0 | if (dict->size >= MAX_HASH_SIZE) |
763 | 0 | return(NULL); |
764 | 0 | newSize = dict->size * 2; |
765 | 0 | } |
766 | 0 | if (xmlDictGrow(dict, newSize) != 0) |
767 | 0 | return(NULL); |
768 | | |
769 | | /* |
770 | | * Find new entry |
771 | | */ |
772 | 0 | mask = dict->size - 1; |
773 | 0 | displ = 0; |
774 | 0 | pos = hashValue & mask; |
775 | 0 | entry = &dict->table[pos]; |
776 | |
|
777 | 0 | while ((entry->hashValue != 0) && |
778 | 0 | ((pos - entry->hashValue) & mask) >= displ) { |
779 | 0 | displ++; |
780 | 0 | pos++; |
781 | 0 | entry++; |
782 | 0 | if ((pos & mask) == 0) |
783 | 0 | entry = dict->table; |
784 | 0 | } |
785 | 0 | } |
786 | | |
787 | 0 | if (prefix == NULL) |
788 | 0 | ret = xmlDictAddString(dict, name, len); |
789 | 0 | else |
790 | 0 | ret = xmlDictAddQString(dict, prefix, plen, name, len); |
791 | 0 | if (ret == NULL) |
792 | 0 | return(NULL); |
793 | | |
794 | | /* |
795 | | * Shift the remainder of the probe sequence to the right |
796 | | */ |
797 | 0 | if (entry->hashValue != 0) { |
798 | 0 | const xmlDictEntry *end = &dict->table[dict->size]; |
799 | 0 | const xmlDictEntry *cur = entry; |
800 | |
|
801 | 0 | do { |
802 | 0 | cur++; |
803 | 0 | if (cur >= end) |
804 | 0 | cur = dict->table; |
805 | 0 | } while (cur->hashValue != 0); |
806 | |
|
807 | 0 | if (cur < entry) { |
808 | | /* |
809 | | * If we traversed the end of the buffer, handle the part |
810 | | * at the start of the buffer. |
811 | | */ |
812 | 0 | memmove(&dict->table[1], dict->table, |
813 | 0 | (char *) cur - (char *) dict->table); |
814 | 0 | cur = end - 1; |
815 | 0 | dict->table[0] = *cur; |
816 | 0 | } |
817 | |
|
818 | 0 | memmove(&entry[1], entry, (char *) cur - (char *) entry); |
819 | 0 | } |
820 | | |
821 | | /* |
822 | | * Populate entry |
823 | | */ |
824 | 0 | entry->hashValue = hashValue; |
825 | 0 | entry->name = ret; |
826 | |
|
827 | 0 | dict->nbElems++; |
828 | |
|
829 | 0 | return(entry); |
830 | 0 | } |
831 | | |
832 | | /** |
833 | | * xmlDictLookup: |
834 | | * @dict: dictionary |
835 | | * @name: string key |
836 | | * @len: length of the key, if -1 it is recomputed |
837 | | * |
838 | | * Lookup a string and add it to the dictionary if it wasn't found. |
839 | | * |
840 | | * Returns the interned copy of the string or NULL if a memory allocation |
841 | | * failed. |
842 | | */ |
843 | | const xmlChar * |
844 | 0 | xmlDictLookup(xmlDictPtr dict, const xmlChar *name, int len) { |
845 | 0 | const xmlDictEntry *entry; |
846 | |
|
847 | 0 | entry = xmlDictLookupInternal(dict, NULL, name, len, 1); |
848 | 0 | if (entry == NULL) |
849 | 0 | return(NULL); |
850 | 0 | return(entry->name); |
851 | 0 | } |
852 | | |
853 | | /** |
854 | | * xmlDictLookupHashed: |
855 | | * @dict: dictionary |
856 | | * @name: string key |
857 | | * @len: length of the key, if -1 it is recomputed |
858 | | * |
859 | | * Lookup a dictionary entry and add the string to the dictionary if |
860 | | * it wasn't found. |
861 | | * |
862 | | * Returns the dictionary entry. |
863 | | */ |
864 | | xmlHashedString |
865 | 0 | xmlDictLookupHashed(xmlDictPtr dict, const xmlChar *name, int len) { |
866 | 0 | const xmlDictEntry *entry; |
867 | 0 | xmlHashedString ret; |
868 | |
|
869 | 0 | entry = xmlDictLookupInternal(dict, NULL, name, len, 1); |
870 | |
|
871 | 0 | if (entry == NULL) { |
872 | 0 | ret.name = NULL; |
873 | 0 | ret.hashValue = 0; |
874 | 0 | } else { |
875 | 0 | ret = *entry; |
876 | 0 | } |
877 | |
|
878 | 0 | return(ret); |
879 | 0 | } |
880 | | |
881 | | /** |
882 | | * xmlDictExists: |
883 | | * @dict: the dictionary |
884 | | * @name: the name of the userdata |
885 | | * @len: the length of the name, if -1 it is recomputed |
886 | | * |
887 | | * Check if a string exists in the dictionary. |
888 | | * |
889 | | * Returns the internal copy of the name or NULL if not found. |
890 | | */ |
891 | | const xmlChar * |
892 | 0 | xmlDictExists(xmlDictPtr dict, const xmlChar *name, int len) { |
893 | 0 | const xmlDictEntry *entry; |
894 | |
|
895 | 0 | entry = xmlDictLookupInternal(dict, NULL, name, len, 0); |
896 | 0 | if (entry == NULL) |
897 | 0 | return(NULL); |
898 | 0 | return(entry->name); |
899 | 0 | } |
900 | | |
901 | | /** |
902 | | * xmlDictQLookup: |
903 | | * @dict: the dictionary |
904 | | * @prefix: the prefix |
905 | | * @name: the name |
906 | | * |
907 | | * Lookup the QName @prefix:@name and add it to the dictionary if |
908 | | * it wasn't found. |
909 | | * |
910 | | * Returns the interned copy of the string or NULL if a memory allocation |
911 | | * failed. |
912 | | */ |
913 | | const xmlChar * |
914 | 0 | xmlDictQLookup(xmlDictPtr dict, const xmlChar *prefix, const xmlChar *name) { |
915 | 0 | const xmlDictEntry *entry; |
916 | |
|
917 | 0 | entry = xmlDictLookupInternal(dict, prefix, name, -1, 1); |
918 | 0 | if (entry == NULL) |
919 | 0 | return(NULL); |
920 | 0 | return(entry->name); |
921 | 0 | } |
922 | | |
923 | | /* |
924 | | * Pseudo-random generator |
925 | | */ |
926 | | |
927 | | #ifdef _WIN32 |
928 | | #define WIN32_LEAN_AND_MEAN |
929 | | #include <windows.h> |
930 | | #include <bcrypt.h> |
931 | | #elif defined(HAVE_GETENTROPY) |
932 | | #ifdef HAVE_UNISTD_H |
933 | | #include <unistd.h> |
934 | | #endif |
935 | | #ifdef HAVE_SYS_RANDOM_H |
936 | | #include <sys/random.h> |
937 | | #endif |
938 | | #else |
939 | | #include <time.h> |
940 | | #endif |
941 | | |
942 | | static xmlMutex xmlRngMutex; |
943 | | |
944 | | static unsigned globalRngState[2]; |
945 | | |
946 | | /* |
947 | | * xmlInitRandom: |
948 | | * |
949 | | * Initialize the PRNG. |
950 | | */ |
951 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
952 | | void |
953 | 0 | xmlInitRandom(void) { |
954 | 0 | xmlInitMutex(&xmlRngMutex); |
955 | |
|
956 | 0 | { |
957 | | #ifdef _WIN32 |
958 | | NTSTATUS status; |
959 | | |
960 | | status = BCryptGenRandom(NULL, (unsigned char *) globalRngState, |
961 | | sizeof(globalRngState), |
962 | | BCRYPT_USE_SYSTEM_PREFERRED_RNG); |
963 | | if (!BCRYPT_SUCCESS(status)) { |
964 | | fprintf(stderr, "libxml2: BCryptGenRandom failed with " |
965 | | "error code %lu\n", GetLastError()); |
966 | | abort(); |
967 | | } |
968 | | #elif defined(HAVE_GETENTROPY) |
969 | | while (1) { |
970 | | if (getentropy(globalRngState, sizeof(globalRngState)) == 0) |
971 | | break; |
972 | | |
973 | | if (errno != EINTR) { |
974 | | fprintf(stderr, "libxml2: getentropy failed with " |
975 | | "error code %d\n", errno); |
976 | | abort(); |
977 | | } |
978 | | } |
979 | | #else |
980 | 0 | int var; |
981 | |
|
982 | 0 | globalRngState[0] = |
983 | 0 | (unsigned) time(NULL) ^ |
984 | 0 | HASH_ROL((unsigned) ((size_t) &xmlInitRandom & 0xFFFFFFFF), 8); |
985 | 0 | globalRngState[1] = |
986 | 0 | HASH_ROL((unsigned) ((size_t) &xmlRngMutex & 0xFFFFFFFF), 16) ^ |
987 | 0 | HASH_ROL((unsigned) ((size_t) &var & 0xFFFFFFFF), 24); |
988 | 0 | #endif |
989 | 0 | } |
990 | 0 | } |
991 | | |
992 | | /* |
993 | | * xmlCleanupRandom: |
994 | | * |
995 | | * Clean up PRNG globals. |
996 | | */ |
997 | | void |
998 | 0 | xmlCleanupRandom(void) { |
999 | 0 | xmlCleanupMutex(&xmlRngMutex); |
1000 | 0 | } |
1001 | | |
1002 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
1003 | | static unsigned |
1004 | 0 | xoroshiro64ss(unsigned *s) { |
1005 | 0 | unsigned s0 = s[0]; |
1006 | 0 | unsigned s1 = s[1]; |
1007 | 0 | unsigned result = HASH_ROL(s0 * 0x9E3779BB, 5) * 5; |
1008 | |
|
1009 | 0 | s1 ^= s0; |
1010 | 0 | s[0] = HASH_ROL(s0, 26) ^ s1 ^ (s1 << 9); |
1011 | 0 | s[1] = HASH_ROL(s1, 13); |
1012 | |
|
1013 | 0 | return(result & 0xFFFFFFFF); |
1014 | 0 | } |
1015 | | |
1016 | | /* |
1017 | | * xmlGlobalRandom: |
1018 | | * |
1019 | | * Generate a pseudo-random value using the global PRNG. |
1020 | | * |
1021 | | * Returns a random value. |
1022 | | */ |
1023 | | unsigned |
1024 | 0 | xmlGlobalRandom(void) { |
1025 | 0 | unsigned ret; |
1026 | |
|
1027 | 0 | xmlMutexLock(&xmlRngMutex); |
1028 | 0 | ret = xoroshiro64ss(globalRngState); |
1029 | 0 | xmlMutexUnlock(&xmlRngMutex); |
1030 | |
|
1031 | 0 | return(ret); |
1032 | 0 | } |
1033 | | |
1034 | | /* |
1035 | | * xmlRandom: |
1036 | | * |
1037 | | * Generate a pseudo-random value using the thread-local PRNG. |
1038 | | * |
1039 | | * Returns a random value. |
1040 | | */ |
1041 | | unsigned |
1042 | 0 | xmlRandom(void) { |
1043 | 0 | #ifdef LIBXML_THREAD_ENABLED |
1044 | 0 | return(xoroshiro64ss(xmlGetLocalRngState())); |
1045 | | #else |
1046 | | return(xmlGlobalRandom()); |
1047 | | #endif |
1048 | 0 | } |
1049 | | |