/work/workdir/UnpackedTarball/libxml2/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/error.h" |
29 | | #include "private/globals.h" |
30 | | #include "private/threads.h" |
31 | | |
32 | | #include <libxml/parser.h> |
33 | | #include <libxml/dict.h> |
34 | | #include <libxml/xmlmemory.h> |
35 | | #include <libxml/xmlstring.h> |
36 | | |
37 | | #ifndef SIZE_MAX |
38 | | #define SIZE_MAX ((size_t) -1) |
39 | | #endif |
40 | | |
41 | 11.4M | #define MAX_FILL_NUM 7 |
42 | 11.4M | #define MAX_FILL_DENOM 8 |
43 | 272k | #define MIN_HASH_SIZE 8 |
44 | 96.9M | #define MAX_HASH_SIZE (1u << 31) |
45 | | |
46 | | typedef struct _xmlDictStrings xmlDictStrings; |
47 | | typedef xmlDictStrings *xmlDictStringsPtr; |
48 | | struct _xmlDictStrings { |
49 | | xmlDictStringsPtr next; |
50 | | xmlChar *free; |
51 | | xmlChar *end; |
52 | | size_t size; |
53 | | size_t nbStrings; |
54 | | xmlChar array[1]; |
55 | | }; |
56 | | |
57 | | typedef xmlHashedString xmlDictEntry; |
58 | | |
59 | | /* |
60 | | * The entire dictionary |
61 | | */ |
62 | | struct _xmlDict { |
63 | | int ref_counter; |
64 | | |
65 | | xmlDictEntry *table; |
66 | | size_t size; |
67 | | unsigned int nbElems; |
68 | | xmlDictStringsPtr strings; |
69 | | |
70 | | struct _xmlDict *subdict; |
71 | | /* used for randomization */ |
72 | | unsigned seed; |
73 | | /* used to impose a limit on size */ |
74 | | size_t limit; |
75 | | }; |
76 | | |
77 | | /* |
78 | | * A mutex for modifying the reference counter for shared |
79 | | * dictionaries. |
80 | | */ |
81 | | static xmlMutex xmlDictMutex; |
82 | | |
83 | | /** |
84 | | * xmlInitializeDict: |
85 | | * |
86 | | * DEPRECATED: Alias for xmlInitParser. |
87 | | * |
88 | | * Returns 0. |
89 | | */ |
90 | | int |
91 | 0 | xmlInitializeDict(void) { |
92 | 0 | xmlInitParser(); |
93 | 0 | return(0); |
94 | 0 | } |
95 | | |
96 | | /** |
97 | | * xmlInitDictInternal: |
98 | | * |
99 | | * Initialize mutex. |
100 | | */ |
101 | | void |
102 | 18 | xmlInitDictInternal(void) { |
103 | 18 | xmlInitMutex(&xmlDictMutex); |
104 | 18 | } |
105 | | |
106 | | /** |
107 | | * xmlDictCleanup: |
108 | | * |
109 | | * DEPRECATED: This function is a no-op. Call xmlCleanupParser |
110 | | * to free global state but see the warnings there. xmlCleanupParser |
111 | | * should be only called once at program exit. In most cases, you don't |
112 | | * have call cleanup functions at all. |
113 | | */ |
114 | | void |
115 | 0 | xmlDictCleanup(void) { |
116 | 0 | } |
117 | | |
118 | | /** |
119 | | * xmlCleanupDictInternal: |
120 | | * |
121 | | * Free the dictionary mutex. |
122 | | */ |
123 | | void |
124 | 0 | xmlCleanupDictInternal(void) { |
125 | 0 | xmlCleanupMutex(&xmlDictMutex); |
126 | 0 | } |
127 | | |
128 | | /* |
129 | | * xmlDictAddString: |
130 | | * @dict: the dictionary |
131 | | * @name: the name of the userdata |
132 | | * @len: the length of the name |
133 | | * |
134 | | * Add the string to the array[s] |
135 | | * |
136 | | * Returns the pointer of the local string, or NULL in case of error. |
137 | | */ |
138 | | static const xmlChar * |
139 | 11.7M | xmlDictAddString(xmlDictPtr dict, const xmlChar *name, unsigned int namelen) { |
140 | 11.7M | xmlDictStringsPtr pool; |
141 | 11.7M | const xmlChar *ret; |
142 | 11.7M | size_t size = 0; /* + sizeof(_xmlDictStrings) == 1024 */ |
143 | 11.7M | size_t limit = 0; |
144 | | |
145 | 11.7M | pool = dict->strings; |
146 | 11.7M | while (pool != NULL) { |
147 | 11.4M | if ((size_t)(pool->end - pool->free) > namelen) |
148 | 11.3M | goto found_pool; |
149 | 51.8k | if (pool->size > size) size = pool->size; |
150 | 51.8k | limit += pool->size; |
151 | 51.8k | pool = pool->next; |
152 | 51.8k | } |
153 | | /* |
154 | | * Not found, need to allocate |
155 | | */ |
156 | 322k | if (pool == NULL) { |
157 | 322k | if ((dict->limit > 0) && (limit > dict->limit)) { |
158 | 0 | return(NULL); |
159 | 0 | } |
160 | | |
161 | 322k | if (size == 0) { |
162 | 272k | size = 1000; |
163 | 272k | } else { |
164 | 50.0k | if (size < (SIZE_MAX - sizeof(xmlDictStrings)) / 4) |
165 | 50.0k | size *= 4; /* exponential growth */ |
166 | 0 | else |
167 | 0 | size = SIZE_MAX - sizeof(xmlDictStrings); |
168 | 50.0k | } |
169 | 322k | if (size / 4 < namelen) { |
170 | 593 | if ((size_t) namelen + 0 < (SIZE_MAX - sizeof(xmlDictStrings)) / 4) |
171 | 593 | size = 4 * (size_t) namelen; /* just in case ! */ |
172 | 0 | else |
173 | 0 | return(NULL); |
174 | 593 | } |
175 | 322k | pool = (xmlDictStringsPtr) xmlMalloc(sizeof(xmlDictStrings) + size); |
176 | 322k | if (pool == NULL) |
177 | 0 | return(NULL); |
178 | 322k | pool->size = size; |
179 | 322k | pool->nbStrings = 0; |
180 | 322k | pool->free = &pool->array[0]; |
181 | 322k | pool->end = &pool->array[size]; |
182 | 322k | pool->next = dict->strings; |
183 | 322k | dict->strings = pool; |
184 | 322k | } |
185 | 11.7M | found_pool: |
186 | 11.7M | ret = pool->free; |
187 | 11.7M | memcpy(pool->free, name, namelen); |
188 | 11.7M | pool->free += namelen; |
189 | 11.7M | *(pool->free++) = 0; |
190 | 11.7M | pool->nbStrings++; |
191 | 11.7M | return(ret); |
192 | 322k | } |
193 | | |
194 | | /* |
195 | | * xmlDictAddQString: |
196 | | * @dict: the dictionary |
197 | | * @prefix: the prefix of the userdata |
198 | | * @plen: the prefix length |
199 | | * @name: the name of the userdata |
200 | | * @len: the length of the name |
201 | | * |
202 | | * Add the QName to the array[s] |
203 | | * |
204 | | * Returns the pointer of the local string, or NULL in case of error. |
205 | | */ |
206 | | static const xmlChar * |
207 | | xmlDictAddQString(xmlDictPtr dict, const xmlChar *prefix, unsigned int plen, |
208 | | const xmlChar *name, unsigned int namelen) |
209 | 230 | { |
210 | 230 | xmlDictStringsPtr pool; |
211 | 230 | const xmlChar *ret; |
212 | 230 | size_t size = 0; /* + sizeof(_xmlDictStrings) == 1024 */ |
213 | 230 | size_t limit = 0; |
214 | | |
215 | 230 | pool = dict->strings; |
216 | 230 | while (pool != NULL) { |
217 | 230 | if ((size_t)(pool->end - pool->free) > namelen + plen + 1) |
218 | 230 | goto found_pool; |
219 | 0 | if (pool->size > size) size = pool->size; |
220 | 0 | limit += pool->size; |
221 | 0 | pool = pool->next; |
222 | 0 | } |
223 | | /* |
224 | | * Not found, need to allocate |
225 | | */ |
226 | 0 | if (pool == NULL) { |
227 | 0 | if ((dict->limit > 0) && (limit > dict->limit)) { |
228 | 0 | return(NULL); |
229 | 0 | } |
230 | | |
231 | 0 | if (size == 0) size = 1000; |
232 | 0 | else size *= 4; /* exponential growth */ |
233 | 0 | if (size < 4 * (namelen + plen + 1)) |
234 | 0 | size = 4 * (namelen + plen + 1); /* just in case ! */ |
235 | 0 | pool = (xmlDictStringsPtr) xmlMalloc(sizeof(xmlDictStrings) + size); |
236 | 0 | if (pool == NULL) |
237 | 0 | return(NULL); |
238 | 0 | pool->size = size; |
239 | 0 | pool->nbStrings = 0; |
240 | 0 | pool->free = &pool->array[0]; |
241 | 0 | pool->end = &pool->array[size]; |
242 | 0 | pool->next = dict->strings; |
243 | 0 | dict->strings = pool; |
244 | 0 | } |
245 | 230 | found_pool: |
246 | 230 | ret = pool->free; |
247 | 230 | memcpy(pool->free, prefix, plen); |
248 | 230 | pool->free += plen; |
249 | 230 | *(pool->free++) = ':'; |
250 | 230 | memcpy(pool->free, name, namelen); |
251 | 230 | pool->free += namelen; |
252 | 230 | *(pool->free++) = 0; |
253 | 230 | pool->nbStrings++; |
254 | 230 | return(ret); |
255 | 0 | } |
256 | | |
257 | | /** |
258 | | * xmlDictCreate: |
259 | | * |
260 | | * Create a new dictionary |
261 | | * |
262 | | * Returns the newly created dictionary, or NULL if an error occurred. |
263 | | */ |
264 | | xmlDictPtr |
265 | 272k | xmlDictCreate(void) { |
266 | 272k | xmlDictPtr dict; |
267 | | |
268 | 272k | xmlInitParser(); |
269 | | |
270 | 272k | dict = xmlMalloc(sizeof(xmlDict)); |
271 | 272k | if (dict == NULL) |
272 | 0 | return(NULL); |
273 | 272k | dict->ref_counter = 1; |
274 | 272k | dict->limit = 0; |
275 | | |
276 | 272k | dict->size = 0; |
277 | 272k | dict->nbElems = 0; |
278 | 272k | dict->table = NULL; |
279 | 272k | dict->strings = NULL; |
280 | 272k | dict->subdict = NULL; |
281 | 272k | dict->seed = xmlRandom(); |
282 | 272k | #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION |
283 | 272k | dict->seed = 0; |
284 | 272k | #endif |
285 | 272k | return(dict); |
286 | 272k | } |
287 | | |
288 | | /** |
289 | | * xmlDictCreateSub: |
290 | | * @sub: an existing dictionary |
291 | | * |
292 | | * Create a new dictionary, inheriting strings from the read-only |
293 | | * dictionary @sub. On lookup, strings are first searched in the |
294 | | * new dictionary, then in @sub, and if not found are created in the |
295 | | * new dictionary. |
296 | | * |
297 | | * Returns the newly created dictionary, or NULL if an error occurred. |
298 | | */ |
299 | | xmlDictPtr |
300 | 0 | xmlDictCreateSub(xmlDictPtr sub) { |
301 | 0 | xmlDictPtr dict = xmlDictCreate(); |
302 | |
|
303 | 0 | if ((dict != NULL) && (sub != NULL)) { |
304 | 0 | dict->seed = sub->seed; |
305 | 0 | dict->subdict = sub; |
306 | 0 | xmlDictReference(dict->subdict); |
307 | 0 | } |
308 | 0 | return(dict); |
309 | 0 | } |
310 | | |
311 | | /** |
312 | | * xmlDictReference: |
313 | | * @dict: the dictionary |
314 | | * |
315 | | * Increment the reference counter of a dictionary |
316 | | * |
317 | | * Returns 0 in case of success and -1 in case of error |
318 | | */ |
319 | | int |
320 | 14.3k | xmlDictReference(xmlDictPtr dict) { |
321 | 14.3k | if (dict == NULL) return -1; |
322 | 12.5k | xmlMutexLock(&xmlDictMutex); |
323 | 12.5k | dict->ref_counter++; |
324 | 12.5k | xmlMutexUnlock(&xmlDictMutex); |
325 | 12.5k | return(0); |
326 | 14.3k | } |
327 | | |
328 | | /** |
329 | | * xmlDictFree: |
330 | | * @dict: the dictionary |
331 | | * |
332 | | * Free the hash @dict and its contents. The userdata is |
333 | | * deallocated with @f if provided. |
334 | | */ |
335 | | void |
336 | 285k | xmlDictFree(xmlDictPtr dict) { |
337 | 285k | xmlDictStringsPtr pool, nextp; |
338 | | |
339 | 285k | if (dict == NULL) |
340 | 0 | return; |
341 | | |
342 | | /* decrement the counter, it may be shared by a parser and docs */ |
343 | 285k | xmlMutexLock(&xmlDictMutex); |
344 | 285k | dict->ref_counter--; |
345 | 285k | if (dict->ref_counter > 0) { |
346 | 12.5k | xmlMutexUnlock(&xmlDictMutex); |
347 | 12.5k | return; |
348 | 12.5k | } |
349 | | |
350 | 272k | xmlMutexUnlock(&xmlDictMutex); |
351 | | |
352 | 272k | if (dict->subdict != NULL) { |
353 | 0 | xmlDictFree(dict->subdict); |
354 | 0 | } |
355 | | |
356 | 272k | if (dict->table) { |
357 | 272k | xmlFree(dict->table); |
358 | 272k | } |
359 | 272k | pool = dict->strings; |
360 | 595k | while (pool != NULL) { |
361 | 322k | nextp = pool->next; |
362 | 322k | xmlFree(pool); |
363 | 322k | pool = nextp; |
364 | 322k | } |
365 | 272k | xmlFree(dict); |
366 | 272k | } |
367 | | |
368 | | /** |
369 | | * xmlDictOwns: |
370 | | * @dict: the dictionary |
371 | | * @str: the string |
372 | | * |
373 | | * check if a string is owned by the dictionary |
374 | | * |
375 | | * Returns 1 if true, 0 if false and -1 in case of error |
376 | | * -1 in case of error |
377 | | */ |
378 | | int |
379 | 4.22M | xmlDictOwns(xmlDictPtr dict, const xmlChar *str) { |
380 | 4.22M | xmlDictStringsPtr pool; |
381 | | |
382 | 4.22M | if ((dict == NULL) || (str == NULL)) |
383 | 0 | return(-1); |
384 | 4.22M | pool = dict->strings; |
385 | 5.41M | while (pool != NULL) { |
386 | 4.41M | if ((str >= &pool->array[0]) && (str <= pool->free)) |
387 | 3.22M | return(1); |
388 | 1.18M | pool = pool->next; |
389 | 1.18M | } |
390 | 1.00M | if (dict->subdict) |
391 | 0 | return(xmlDictOwns(dict->subdict, str)); |
392 | 1.00M | return(0); |
393 | 1.00M | } |
394 | | |
395 | | /** |
396 | | * xmlDictSize: |
397 | | * @dict: the dictionary |
398 | | * |
399 | | * Query the number of elements installed in the hash @dict. |
400 | | * |
401 | | * Returns the number of elements in the dictionary or |
402 | | * -1 in case of error |
403 | | */ |
404 | | int |
405 | 0 | xmlDictSize(xmlDictPtr dict) { |
406 | 0 | if (dict == NULL) |
407 | 0 | return(-1); |
408 | 0 | if (dict->subdict) |
409 | 0 | return(dict->nbElems + dict->subdict->nbElems); |
410 | 0 | return(dict->nbElems); |
411 | 0 | } |
412 | | |
413 | | /** |
414 | | * xmlDictSetLimit: |
415 | | * @dict: the dictionary |
416 | | * @limit: the limit in bytes |
417 | | * |
418 | | * Set a size limit for the dictionary |
419 | | * Added in 2.9.0 |
420 | | * |
421 | | * Returns the previous limit of the dictionary or 0 |
422 | | */ |
423 | | size_t |
424 | 538k | xmlDictSetLimit(xmlDictPtr dict, size_t limit) { |
425 | 538k | size_t ret; |
426 | | |
427 | 538k | if (dict == NULL) |
428 | 0 | return(0); |
429 | 538k | ret = dict->limit; |
430 | 538k | dict->limit = limit; |
431 | 538k | return(ret); |
432 | 538k | } |
433 | | |
434 | | /** |
435 | | * xmlDictGetUsage: |
436 | | * @dict: the dictionary |
437 | | * |
438 | | * Get how much memory is used by a dictionary for strings |
439 | | * Added in 2.9.0 |
440 | | * |
441 | | * Returns the amount of strings allocated |
442 | | */ |
443 | | size_t |
444 | 0 | xmlDictGetUsage(xmlDictPtr dict) { |
445 | 0 | xmlDictStringsPtr pool; |
446 | 0 | size_t limit = 0; |
447 | |
|
448 | 0 | if (dict == NULL) |
449 | 0 | return(0); |
450 | 0 | pool = dict->strings; |
451 | 0 | while (pool != NULL) { |
452 | 0 | limit += pool->size; |
453 | 0 | pool = pool->next; |
454 | 0 | } |
455 | 0 | return(limit); |
456 | 0 | } |
457 | | |
458 | | /***************************************************************** |
459 | | * |
460 | | * The code below was rewritten and is additionally licensed under |
461 | | * the main license in file 'Copyright'. |
462 | | * |
463 | | *****************************************************************/ |
464 | | |
465 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
466 | | static unsigned |
467 | | xmlDictHashName(unsigned seed, const xmlChar* data, size_t maxLen, |
468 | 96.2M | size_t *plen) { |
469 | 96.2M | unsigned h1, h2; |
470 | 96.2M | size_t i; |
471 | | |
472 | 96.2M | HASH_INIT(h1, h2, seed); |
473 | | |
474 | 755M | for (i = 0; i < maxLen && data[i]; i++) { |
475 | 659M | HASH_UPDATE(h1, h2, data[i]); |
476 | 659M | } |
477 | | |
478 | 96.2M | HASH_FINISH(h1, h2); |
479 | | |
480 | 96.2M | *plen = i; |
481 | 96.2M | return(h2 | MAX_HASH_SIZE); |
482 | 96.2M | } |
483 | | |
484 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
485 | | static unsigned |
486 | | xmlDictHashQName(unsigned seed, const xmlChar *prefix, const xmlChar *name, |
487 | 1.02k | size_t *pplen, size_t *plen) { |
488 | 1.02k | unsigned h1, h2; |
489 | 1.02k | size_t i; |
490 | | |
491 | 1.02k | HASH_INIT(h1, h2, seed); |
492 | | |
493 | 3.01k | for (i = 0; prefix[i] != 0; i++) { |
494 | 1.99k | HASH_UPDATE(h1, h2, prefix[i]); |
495 | 1.99k | } |
496 | 1.02k | *pplen = i; |
497 | | |
498 | 1.02k | HASH_UPDATE(h1, h2, ':'); |
499 | | |
500 | 9.13k | for (i = 0; name[i] != 0; i++) { |
501 | 8.10k | HASH_UPDATE(h1, h2, name[i]); |
502 | 8.10k | } |
503 | 1.02k | *plen = i; |
504 | | |
505 | 1.02k | HASH_FINISH(h1, h2); |
506 | | |
507 | | /* |
508 | | * Always set the upper bit of hash values since 0 means an unoccupied |
509 | | * bucket. |
510 | | */ |
511 | 1.02k | return(h2 | MAX_HASH_SIZE); |
512 | 1.02k | } |
513 | | |
514 | | /** |
515 | | * xmlDictComputeHash: |
516 | | * @dict: dictionary |
517 | | * @string: C string |
518 | | * |
519 | | * Compute the hash value of a C string. |
520 | | * |
521 | | * Returns the hash value. |
522 | | */ |
523 | | unsigned |
524 | 1.09M | xmlDictComputeHash(const xmlDict *dict, const xmlChar *string) { |
525 | 1.09M | size_t len; |
526 | 1.09M | return(xmlDictHashName(dict->seed, string, SIZE_MAX, &len)); |
527 | 1.09M | } |
528 | | |
529 | 20.8M | #define HASH_ROL31(x,n) ((x) << (n) | ((x) & 0x7FFFFFFF) >> (31 - (n))) |
530 | | |
531 | | /** |
532 | | * xmlDictCombineHash: |
533 | | * @v1: first hash value |
534 | | * @v2: second hash value |
535 | | * |
536 | | * Combine two hash values. |
537 | | * |
538 | | * Returns the combined hash value. |
539 | | */ |
540 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
541 | | unsigned |
542 | 20.8M | xmlDictCombineHash(unsigned v1, unsigned v2) { |
543 | | /* |
544 | | * The upper bit of hash values is always set, so we have to operate on |
545 | | * 31-bit hashes here. |
546 | | */ |
547 | 20.8M | v1 ^= v2; |
548 | 20.8M | v1 += HASH_ROL31(v2, 5); |
549 | | |
550 | 20.8M | return((v1 & 0xFFFFFFFF) | 0x80000000); |
551 | 20.8M | } |
552 | | |
553 | | /** |
554 | | * xmlDictFindEntry: |
555 | | * @dict: dict |
556 | | * @prefix: optional QName prefix |
557 | | * @name: string |
558 | | * @len: length of string |
559 | | * @hashValue: valid hash value of string |
560 | | * @pfound: result of search |
561 | | * |
562 | | * Try to find a matching hash table entry. If an entry was found, set |
563 | | * @found to 1 and return the entry. Otherwise, set @found to 0 and return |
564 | | * the location where a new entry should be inserted. |
565 | | */ |
566 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
567 | | static xmlDictEntry * |
568 | | xmlDictFindEntry(const xmlDict *dict, const xmlChar *prefix, |
569 | | const xmlChar *name, int len, unsigned hashValue, |
570 | 94.9M | int *pfound) { |
571 | 94.9M | xmlDictEntry *entry; |
572 | 94.9M | unsigned mask, pos, displ; |
573 | 94.9M | int found = 0; |
574 | | |
575 | 94.9M | mask = dict->size - 1; |
576 | 94.9M | pos = hashValue & mask; |
577 | 94.9M | entry = &dict->table[pos]; |
578 | | |
579 | 94.9M | if (entry->hashValue != 0) { |
580 | | /* |
581 | | * Robin hood hashing: abort if the displacement of the entry |
582 | | * is smaller than the displacement of the key we look for. |
583 | | * This also stops at the correct position when inserting. |
584 | | */ |
585 | 90.3M | displ = 0; |
586 | | |
587 | 161M | do { |
588 | 161M | if (entry->hashValue == hashValue) { |
589 | 83.4M | if (prefix == NULL) { |
590 | | /* |
591 | | * name is not necessarily null-terminated. |
592 | | */ |
593 | 83.4M | if ((strncmp((const char *) entry->name, |
594 | 83.4M | (const char *) name, len) == 0) && |
595 | 83.4M | (entry->name[len] == 0)) { |
596 | 83.4M | found = 1; |
597 | 83.4M | break; |
598 | 83.4M | } |
599 | 83.4M | } else { |
600 | 793 | if (xmlStrQEqual(prefix, name, entry->name)) { |
601 | 793 | found = 1; |
602 | 793 | break; |
603 | 793 | } |
604 | 786 | } |
605 | 83.4M | } |
606 | | |
607 | 78.4M | displ++; |
608 | 78.4M | pos++; |
609 | 78.4M | entry++; |
610 | 78.4M | if ((pos & mask) == 0) |
611 | 883k | entry = dict->table; |
612 | 78.4M | } while ((entry->hashValue != 0) && |
613 | 78.4M | (((pos - entry->hashValue) & mask) >= displ)); |
614 | 90.3M | } |
615 | | |
616 | 0 | *pfound = found; |
617 | 94.9M | return(entry); |
618 | 94.9M | } |
619 | | |
620 | | /** |
621 | | * xmlDictGrow: |
622 | | * @dict: dictionary |
623 | | * @size: new size of the dictionary |
624 | | * |
625 | | * Resize the dictionary hash table. |
626 | | * |
627 | | * Returns 0 in case of success, -1 if a memory allocation failed. |
628 | | */ |
629 | | static int |
630 | 910k | xmlDictGrow(xmlDictPtr dict, unsigned size) { |
631 | 910k | const xmlDictEntry *oldentry, *oldend, *end; |
632 | 910k | xmlDictEntry *table; |
633 | 910k | unsigned oldsize, i; |
634 | | |
635 | | /* Add 0 to avoid spurious -Wtype-limits warning on 64-bit GCC */ |
636 | 910k | if ((size_t) size + 0 > SIZE_MAX / sizeof(table[0])) |
637 | 0 | return(-1); |
638 | 910k | table = xmlMalloc(size * sizeof(table[0])); |
639 | 910k | if (table == NULL) |
640 | 0 | return(-1); |
641 | 910k | memset(table, 0, size * sizeof(table[0])); |
642 | | |
643 | 910k | oldsize = dict->size; |
644 | 910k | if (oldsize == 0) |
645 | 272k | goto done; |
646 | | |
647 | 637k | oldend = &dict->table[oldsize]; |
648 | 637k | end = &table[size]; |
649 | | |
650 | | /* |
651 | | * Robin Hood sorting order is maintained if we |
652 | | * |
653 | | * - compute dict indices with modulo |
654 | | * - resize by an integer factor |
655 | | * - start to copy from the beginning of a probe sequence |
656 | | */ |
657 | 637k | oldentry = dict->table; |
658 | 4.43M | while (oldentry->hashValue != 0) { |
659 | 3.80M | if (++oldentry >= oldend) |
660 | 0 | oldentry = dict->table; |
661 | 3.80M | } |
662 | | |
663 | 17.6M | for (i = 0; i < oldsize; i++) { |
664 | 17.0M | if (oldentry->hashValue != 0) { |
665 | 14.8M | xmlDictEntry *entry = &table[oldentry->hashValue & (size - 1)]; |
666 | | |
667 | 19.4M | while (entry->hashValue != 0) { |
668 | 4.58M | if (++entry >= end) |
669 | 96.8k | entry = table; |
670 | 4.58M | } |
671 | 14.8M | *entry = *oldentry; |
672 | 14.8M | } |
673 | | |
674 | 17.0M | if (++oldentry >= oldend) |
675 | 637k | oldentry = dict->table; |
676 | 17.0M | } |
677 | | |
678 | 637k | xmlFree(dict->table); |
679 | | |
680 | 910k | done: |
681 | 910k | dict->table = table; |
682 | 910k | dict->size = size; |
683 | | |
684 | 910k | return(0); |
685 | 637k | } |
686 | | |
687 | | /** |
688 | | * xmlDictLookupInternal: |
689 | | * @dict: dict |
690 | | * @prefix: optional QName prefix |
691 | | * @name: string |
692 | | * @maybeLen: length of string or -1 if unknown |
693 | | * @update: whether the string should be added |
694 | | * |
695 | | * Internal lookup and update function. |
696 | | */ |
697 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
698 | | static const xmlDictEntry * |
699 | | xmlDictLookupInternal(xmlDictPtr dict, const xmlChar *prefix, |
700 | 95.1M | const xmlChar *name, int maybeLen, int update) { |
701 | 95.1M | xmlDictEntry *entry = NULL; |
702 | 95.1M | const xmlChar *ret; |
703 | 95.1M | unsigned hashValue, newSize; |
704 | 95.1M | size_t maxLen, len, plen, klen; |
705 | 95.1M | int found = 0; |
706 | | |
707 | 95.1M | if ((dict == NULL) || (name == NULL)) |
708 | 0 | return(NULL); |
709 | | |
710 | 95.1M | maxLen = (maybeLen < 0) ? SIZE_MAX : (size_t) maybeLen; |
711 | | |
712 | 95.1M | if (prefix == NULL) { |
713 | 95.1M | hashValue = xmlDictHashName(dict->seed, name, maxLen, &len); |
714 | 95.1M | if (len > INT_MAX / 2) |
715 | 0 | return(NULL); |
716 | 95.1M | klen = len; |
717 | 95.1M | } else { |
718 | 939 | hashValue = xmlDictHashQName(dict->seed, prefix, name, &plen, &len); |
719 | 1.02k | if ((len > INT_MAX / 2) || (plen >= INT_MAX / 2 - len)) |
720 | 0 | return(NULL); |
721 | 939 | klen = plen + 1 + len; |
722 | 939 | } |
723 | | |
724 | 95.1M | if ((dict->limit > 0) && (klen >= dict->limit)) |
725 | 0 | return(NULL); |
726 | | |
727 | | /* |
728 | | * Check for an existing entry |
729 | | */ |
730 | 95.1M | if (dict->size == 0) { |
731 | 272k | newSize = MIN_HASH_SIZE; |
732 | 94.9M | } else { |
733 | 94.9M | entry = xmlDictFindEntry(dict, prefix, name, klen, hashValue, &found); |
734 | 94.9M | if (found) |
735 | 83.4M | return(entry); |
736 | | |
737 | 11.4M | if (dict->nbElems + 1 > dict->size / MAX_FILL_DENOM * MAX_FILL_NUM) { |
738 | 637k | if (dict->size >= MAX_HASH_SIZE) |
739 | 0 | return(NULL); |
740 | 637k | newSize = dict->size * 2; |
741 | 10.7M | } else { |
742 | 10.7M | newSize = 0; |
743 | 10.7M | } |
744 | 11.4M | } |
745 | | |
746 | 11.6M | if ((dict->subdict != NULL) && (dict->subdict->size > 0)) { |
747 | 0 | xmlDictEntry *subEntry; |
748 | 0 | unsigned subHashValue; |
749 | |
|
750 | 0 | if (prefix == NULL) |
751 | 0 | subHashValue = xmlDictHashName(dict->subdict->seed, name, len, |
752 | 0 | &len); |
753 | 0 | else |
754 | 0 | subHashValue = xmlDictHashQName(dict->subdict->seed, prefix, name, |
755 | 0 | &plen, &len); |
756 | 0 | subEntry = xmlDictFindEntry(dict->subdict, prefix, name, klen, |
757 | 0 | subHashValue, &found); |
758 | 0 | if (found) |
759 | 0 | return(subEntry); |
760 | 0 | } |
761 | | |
762 | 11.6M | if (!update) |
763 | 0 | return(NULL); |
764 | | |
765 | | /* |
766 | | * Grow the hash table if needed |
767 | | */ |
768 | 11.6M | if (newSize > 0) { |
769 | 910k | unsigned mask, displ, pos; |
770 | | |
771 | 910k | if (xmlDictGrow(dict, newSize) != 0) |
772 | 0 | return(NULL); |
773 | | |
774 | | /* |
775 | | * Find new entry |
776 | | */ |
777 | 910k | mask = dict->size - 1; |
778 | 910k | displ = 0; |
779 | 910k | pos = hashValue & mask; |
780 | 910k | entry = &dict->table[pos]; |
781 | | |
782 | 1.28M | while ((entry->hashValue != 0) && |
783 | 1.28M | ((pos - entry->hashValue) & mask) >= displ) { |
784 | 372k | displ++; |
785 | 372k | pos++; |
786 | 372k | entry++; |
787 | 372k | if ((pos & mask) == 0) |
788 | 13.1k | entry = dict->table; |
789 | 372k | } |
790 | 910k | } |
791 | | |
792 | 11.6M | if (prefix == NULL) |
793 | 11.7M | ret = xmlDictAddString(dict, name, len); |
794 | 18.4E | else |
795 | 18.4E | ret = xmlDictAddQString(dict, prefix, plen, name, len); |
796 | 11.6M | if (ret == NULL) |
797 | 0 | return(NULL); |
798 | | |
799 | | /* |
800 | | * Shift the remainder of the probe sequence to the right |
801 | | */ |
802 | 11.6M | if (entry->hashValue != 0) { |
803 | 3.74M | const xmlDictEntry *end = &dict->table[dict->size]; |
804 | 3.74M | const xmlDictEntry *cur = entry; |
805 | | |
806 | 18.0M | do { |
807 | 18.0M | cur++; |
808 | 18.0M | if (cur >= end) |
809 | 283k | cur = dict->table; |
810 | 18.0M | } while (cur->hashValue != 0); |
811 | | |
812 | 3.74M | if (cur < entry) { |
813 | | /* |
814 | | * If we traversed the end of the buffer, handle the part |
815 | | * at the start of the buffer. |
816 | | */ |
817 | 283k | memmove(&dict->table[1], dict->table, |
818 | 283k | (char *) cur - (char *) dict->table); |
819 | 283k | cur = end - 1; |
820 | 283k | dict->table[0] = *cur; |
821 | 283k | } |
822 | | |
823 | 3.74M | memmove(&entry[1], entry, (char *) cur - (char *) entry); |
824 | 3.74M | } |
825 | | |
826 | | /* |
827 | | * Populate entry |
828 | | */ |
829 | 11.6M | entry->hashValue = hashValue; |
830 | 11.6M | entry->name = ret; |
831 | | |
832 | 11.6M | dict->nbElems++; |
833 | | |
834 | 11.6M | return(entry); |
835 | 11.6M | } |
836 | | |
837 | | /** |
838 | | * xmlDictLookup: |
839 | | * @dict: dictionary |
840 | | * @name: string key |
841 | | * @len: length of the key, if -1 it is recomputed |
842 | | * |
843 | | * Lookup a string and add it to the dictionary if it wasn't found. |
844 | | * |
845 | | * Returns the interned copy of the string or NULL if a memory allocation |
846 | | * failed. |
847 | | */ |
848 | | const xmlChar * |
849 | 2.53M | xmlDictLookup(xmlDictPtr dict, const xmlChar *name, int len) { |
850 | 2.53M | const xmlDictEntry *entry; |
851 | | |
852 | 2.53M | entry = xmlDictLookupInternal(dict, NULL, name, len, 1); |
853 | 2.53M | if (entry == NULL) |
854 | 0 | return(NULL); |
855 | 2.53M | return(entry->name); |
856 | 2.53M | } |
857 | | |
858 | | /** |
859 | | * xmlDictLookupHashed: |
860 | | * @dict: dictionary |
861 | | * @name: string key |
862 | | * @len: length of the key, if -1 it is recomputed |
863 | | * |
864 | | * Lookup a dictionary entry and add the string to the dictionary if |
865 | | * it wasn't found. |
866 | | * |
867 | | * Returns the dictionary entry. |
868 | | */ |
869 | | xmlHashedString |
870 | 92.6M | xmlDictLookupHashed(xmlDictPtr dict, const xmlChar *name, int len) { |
871 | 92.6M | const xmlDictEntry *entry; |
872 | 92.6M | xmlHashedString ret; |
873 | | |
874 | 92.6M | entry = xmlDictLookupInternal(dict, NULL, name, len, 1); |
875 | | |
876 | 92.6M | if (entry == NULL) { |
877 | 0 | ret.name = NULL; |
878 | 0 | ret.hashValue = 0; |
879 | 92.6M | } else { |
880 | 92.6M | ret = *entry; |
881 | 92.6M | } |
882 | | |
883 | 92.6M | return(ret); |
884 | 92.6M | } |
885 | | |
886 | | /** |
887 | | * xmlDictExists: |
888 | | * @dict: the dictionary |
889 | | * @name: the name of the userdata |
890 | | * @len: the length of the name, if -1 it is recomputed |
891 | | * |
892 | | * Check if a string exists in the dictionary. |
893 | | * |
894 | | * Returns the internal copy of the name or NULL if not found. |
895 | | */ |
896 | | const xmlChar * |
897 | 0 | xmlDictExists(xmlDictPtr dict, const xmlChar *name, int len) { |
898 | 0 | const xmlDictEntry *entry; |
899 | |
|
900 | 0 | entry = xmlDictLookupInternal(dict, NULL, name, len, 0); |
901 | 0 | if (entry == NULL) |
902 | 0 | return(NULL); |
903 | 0 | return(entry->name); |
904 | 0 | } |
905 | | |
906 | | /** |
907 | | * xmlDictQLookup: |
908 | | * @dict: the dictionary |
909 | | * @prefix: the prefix |
910 | | * @name: the name |
911 | | * |
912 | | * Lookup the QName @prefix:@name and add it to the dictionary if |
913 | | * it wasn't found. |
914 | | * |
915 | | * Returns the interned copy of the string or NULL if a memory allocation |
916 | | * failed. |
917 | | */ |
918 | | const xmlChar * |
919 | 1.02k | xmlDictQLookup(xmlDictPtr dict, const xmlChar *prefix, const xmlChar *name) { |
920 | 1.02k | const xmlDictEntry *entry; |
921 | | |
922 | 1.02k | entry = xmlDictLookupInternal(dict, prefix, name, -1, 1); |
923 | 1.02k | if (entry == NULL) |
924 | 0 | return(NULL); |
925 | 1.02k | return(entry->name); |
926 | 1.02k | } |
927 | | |
928 | | /* |
929 | | * Pseudo-random generator |
930 | | */ |
931 | | |
932 | | #ifdef _WIN32 |
933 | | #define WIN32_LEAN_AND_MEAN |
934 | | #include <windows.h> |
935 | | #include <bcrypt.h> |
936 | | #else |
937 | | #if HAVE_DECL_GETENTROPY |
938 | | /* POSIX 2024 */ |
939 | | #include <unistd.h> |
940 | | /* Older platforms */ |
941 | | #include <sys/random.h> |
942 | | #endif |
943 | | #include <time.h> |
944 | | #endif |
945 | | |
946 | | static xmlMutex xmlRngMutex; |
947 | | |
948 | | static unsigned globalRngState[2]; |
949 | | |
950 | | /* |
951 | | * xmlInitRandom: |
952 | | * |
953 | | * Initialize the PRNG. |
954 | | */ |
955 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
956 | | void |
957 | 18 | xmlInitRandom(void) { |
958 | 18 | xmlInitMutex(&xmlRngMutex); |
959 | | |
960 | 18 | { |
961 | | #ifdef _WIN32 |
962 | | NTSTATUS status; |
963 | | |
964 | | /* |
965 | | * You can find many (recent as of 2025) discussions how |
966 | | * to get a pseudo-random seed on Windows in projects like |
967 | | * Golang, Rust, Chromium and Firefox. |
968 | | * |
969 | | * TODO: Support ProcessPrng available since Windows 10. |
970 | | */ |
971 | | status = BCryptGenRandom(NULL, (unsigned char *) globalRngState, |
972 | | sizeof(globalRngState), |
973 | | BCRYPT_USE_SYSTEM_PREFERRED_RNG); |
974 | | if (!BCRYPT_SUCCESS(status)) |
975 | | xmlAbort("libxml2: BCryptGenRandom failed with error code %lu\n", |
976 | | GetLastError()); |
977 | | #else |
978 | 18 | int var; |
979 | | |
980 | 18 | #if HAVE_DECL_GETENTROPY |
981 | 18 | while (1) { |
982 | 18 | if (getentropy(globalRngState, sizeof(globalRngState)) == 0) |
983 | 18 | return; |
984 | | |
985 | | /* |
986 | | * This most likely means that libxml2 was compiled on |
987 | | * a system supporting certain system calls and is running |
988 | | * on a system that doesn't support these calls, as can |
989 | | * be the case on Linux. |
990 | | */ |
991 | 0 | if (errno == ENOSYS) |
992 | 0 | break; |
993 | | |
994 | | /* |
995 | | * We really don't want to fallback to the unsafe PRNG |
996 | | * for possibly accidental reasons, so we abort on any |
997 | | * unknown error. |
998 | | */ |
999 | 0 | if (errno != EINTR) |
1000 | 0 | xmlAbort("libxml2: getentropy failed with error code %d\n", |
1001 | 0 | errno); |
1002 | 0 | } |
1003 | 0 | #endif |
1004 | | |
1005 | | /* |
1006 | | * TODO: Fallback to /dev/urandom for older POSIX systems. |
1007 | | */ |
1008 | 0 | globalRngState[0] = |
1009 | 0 | (unsigned) time(NULL) ^ |
1010 | 0 | HASH_ROL((unsigned) ((size_t) &xmlInitRandom & 0xFFFFFFFF), 8); |
1011 | 0 | globalRngState[1] = |
1012 | 0 | HASH_ROL((unsigned) ((size_t) &xmlRngMutex & 0xFFFFFFFF), 16) ^ |
1013 | 0 | HASH_ROL((unsigned) ((size_t) &var & 0xFFFFFFFF), 24); |
1014 | 0 | #endif |
1015 | 0 | } |
1016 | 0 | } |
1017 | | |
1018 | | /* |
1019 | | * xmlCleanupRandom: |
1020 | | * |
1021 | | * Clean up PRNG globals. |
1022 | | */ |
1023 | | void |
1024 | 0 | xmlCleanupRandom(void) { |
1025 | 0 | xmlCleanupMutex(&xmlRngMutex); |
1026 | 0 | } |
1027 | | |
1028 | | ATTRIBUTE_NO_SANITIZE_INTEGER |
1029 | | static unsigned |
1030 | 298k | xoroshiro64ss(unsigned *s) { |
1031 | 298k | unsigned s0 = s[0]; |
1032 | 298k | unsigned s1 = s[1]; |
1033 | 298k | unsigned result = HASH_ROL(s0 * 0x9E3779BB, 5) * 5; |
1034 | | |
1035 | 298k | s1 ^= s0; |
1036 | 298k | s[0] = HASH_ROL(s0, 26) ^ s1 ^ (s1 << 9); |
1037 | 298k | s[1] = HASH_ROL(s1, 13); |
1038 | | |
1039 | 298k | return(result & 0xFFFFFFFF); |
1040 | 298k | } |
1041 | | |
1042 | | /* |
1043 | | * xmlGlobalRandom: |
1044 | | * |
1045 | | * Generate a pseudo-random value using the global PRNG. |
1046 | | * |
1047 | | * Returns a random value. |
1048 | | */ |
1049 | | unsigned |
1050 | 17.5k | xmlGlobalRandom(void) { |
1051 | 17.5k | unsigned ret; |
1052 | | |
1053 | 17.5k | xmlMutexLock(&xmlRngMutex); |
1054 | 17.5k | ret = xoroshiro64ss(globalRngState); |
1055 | 17.5k | xmlMutexUnlock(&xmlRngMutex); |
1056 | | |
1057 | 17.5k | return(ret); |
1058 | 17.5k | } |
1059 | | |
1060 | | /* |
1061 | | * xmlRandom: |
1062 | | * |
1063 | | * Generate a pseudo-random value using the thread-local PRNG. |
1064 | | * |
1065 | | * Returns a random value. |
1066 | | */ |
1067 | | unsigned |
1068 | 280k | xmlRandom(void) { |
1069 | 280k | return(xoroshiro64ss(xmlGetLocalRngState())); |
1070 | 280k | } |
1071 | | |