Line | Count | Source (jump to first uncovered line) |
1 | | /* |
2 | | * parser.c : an XML 1.0 parser, namespaces and validity support are mostly |
3 | | * implemented on top of the SAX interfaces |
4 | | * |
5 | | * References: |
6 | | * The XML specification: |
7 | | * http://www.w3.org/TR/REC-xml |
8 | | * Original 1.0 version: |
9 | | * http://www.w3.org/TR/1998/REC-xml-19980210 |
10 | | * XML second edition working draft |
11 | | * http://www.w3.org/TR/2000/WD-xml-2e-20000814 |
12 | | * |
13 | | * Okay this is a big file, the parser core is around 7000 lines, then it |
14 | | * is followed by the progressive parser top routines, then the various |
15 | | * high level APIs to call the parser and a few miscellaneous functions. |
16 | | * A number of helper functions and deprecated ones have been moved to |
17 | | * parserInternals.c to reduce this file size. |
18 | | * As much as possible the functions are associated with their relative |
19 | | * production in the XML specification. A few productions defining the |
20 | | * different ranges of character are actually implanted either in |
21 | | * parserInternals.h or parserInternals.c |
22 | | * The DOM tree build is realized from the default SAX callbacks in |
23 | | * the module SAX2.c. |
24 | | * The routines doing the validation checks are in valid.c and called either |
25 | | * from the SAX callbacks or as standalone functions using a preparsed |
26 | | * document. |
27 | | * |
28 | | * See Copyright for the status of this software. |
29 | | * |
30 | | * Author: Daniel Veillard |
31 | | */ |
32 | | |
33 | | /* To avoid EBCDIC trouble when parsing on zOS */ |
34 | | #if defined(__MVS__) |
35 | | #pragma convert("ISO8859-1") |
36 | | #endif |
37 | | |
38 | | #define IN_LIBXML |
39 | | #include "libxml.h" |
40 | | |
41 | | #if defined(_WIN32) |
42 | | #define XML_DIR_SEP '\\' |
43 | | #else |
44 | | #define XML_DIR_SEP '/' |
45 | | #endif |
46 | | |
47 | | #include <stdlib.h> |
48 | | #include <limits.h> |
49 | | #include <string.h> |
50 | | #include <stdarg.h> |
51 | | #include <stddef.h> |
52 | | #include <ctype.h> |
53 | | #include <stdlib.h> |
54 | | #include <libxml/parser.h> |
55 | | #include <libxml/xmlmemory.h> |
56 | | #include <libxml/tree.h> |
57 | | #include <libxml/parserInternals.h> |
58 | | #include <libxml/valid.h> |
59 | | #include <libxml/entities.h> |
60 | | #include <libxml/xmlerror.h> |
61 | | #include <libxml/encoding.h> |
62 | | #include <libxml/xmlIO.h> |
63 | | #include <libxml/uri.h> |
64 | | #include <libxml/SAX2.h> |
65 | | #include <libxml/HTMLparser.h> |
66 | | #ifdef LIBXML_CATALOG_ENABLED |
67 | | #include <libxml/catalog.h> |
68 | | #endif |
69 | | |
70 | | #include "private/buf.h" |
71 | | #include "private/dict.h" |
72 | | #include "private/entities.h" |
73 | | #include "private/error.h" |
74 | | #include "private/html.h" |
75 | | #include "private/io.h" |
76 | | #include "private/memory.h" |
77 | | #include "private/parser.h" |
78 | | #include "private/tree.h" |
79 | | |
80 | 1.39M | #define NS_INDEX_EMPTY INT_MAX |
81 | 87.4k | #define NS_INDEX_XML (INT_MAX - 1) |
82 | 566k | #define URI_HASH_EMPTY 0xD943A04E |
83 | 30.5k | #define URI_HASH_XML 0xF0451F02 |
84 | | |
85 | | #ifndef STDIN_FILENO |
86 | 0 | #define STDIN_FILENO 0 |
87 | | #endif |
88 | | |
89 | | #ifndef SIZE_MAX |
90 | | #define SIZE_MAX ((size_t) -1) |
91 | | #endif |
92 | | |
93 | 288k | #define XML_MAX_ATTRS 100000000 /* 100 million */ |
94 | | |
95 | 651k | #define XML_SPECIAL_EXTERNAL (1 << 20) |
96 | 651k | #define XML_SPECIAL_TYPE_MASK (XML_SPECIAL_EXTERNAL - 1) |
97 | | |
98 | 727k | #define XML_ATTVAL_ALLOC (1 << 0) |
99 | 53.2k | #define XML_ATTVAL_NORM_CHANGE (1 << 1) |
100 | | |
101 | | struct _xmlStartTag { |
102 | | const xmlChar *prefix; |
103 | | const xmlChar *URI; |
104 | | int line; |
105 | | int nsNr; |
106 | | }; |
107 | | |
108 | | typedef struct { |
109 | | void *saxData; |
110 | | unsigned prefixHashValue; |
111 | | unsigned uriHashValue; |
112 | | unsigned elementId; |
113 | | int oldIndex; |
114 | | } xmlParserNsExtra; |
115 | | |
116 | | typedef struct { |
117 | | unsigned hashValue; |
118 | | int index; |
119 | | } xmlParserNsBucket; |
120 | | |
121 | | struct _xmlParserNsData { |
122 | | xmlParserNsExtra *extra; |
123 | | |
124 | | unsigned hashSize; |
125 | | unsigned hashElems; |
126 | | xmlParserNsBucket *hash; |
127 | | |
128 | | unsigned elementId; |
129 | | int defaultNsIndex; |
130 | | int minNsIndex; |
131 | | }; |
132 | | |
133 | | static int |
134 | | xmlParseElementStart(xmlParserCtxtPtr ctxt); |
135 | | |
136 | | static void |
137 | | xmlParseElementEnd(xmlParserCtxtPtr ctxt); |
138 | | |
139 | | static xmlEntityPtr |
140 | | xmlLookupGeneralEntity(xmlParserCtxtPtr ctxt, const xmlChar *name, int inAttr); |
141 | | |
142 | | static const xmlChar * |
143 | | xmlParseEntityRefInternal(xmlParserCtxtPtr ctxt); |
144 | | |
145 | | /************************************************************************ |
146 | | * * |
147 | | * Arbitrary limits set in the parser. See XML_PARSE_HUGE * |
148 | | * * |
149 | | ************************************************************************/ |
150 | | |
151 | | #define XML_PARSER_BIG_ENTITY 1000 |
152 | | #define XML_PARSER_LOT_ENTITY 5000 |
153 | | |
154 | | /* |
155 | | * Constants for protection against abusive entity expansion |
156 | | * ("billion laughs"). |
157 | | */ |
158 | | |
159 | | /* |
160 | | * A certain amount of entity expansion which is always allowed. |
161 | | */ |
162 | 1.15M | #define XML_PARSER_ALLOWED_EXPANSION 1000000 |
163 | | |
164 | | /* |
165 | | * Fixed cost for each entity reference. This crudely models processing time |
166 | | * as well to protect, for example, against exponential expansion of empty |
167 | | * or very short entities. |
168 | | */ |
169 | 1.15M | #define XML_ENT_FIXED_COST 20 |
170 | | |
171 | 27.7M | #define XML_PARSER_BIG_BUFFER_SIZE 300 |
172 | 177k | #define XML_PARSER_BUFFER_SIZE 100 |
173 | 33.0k | #define SAX_COMPAT_MODE BAD_CAST "SAX compatibility mode document" |
174 | | |
175 | | /** |
176 | | * XML_PARSER_CHUNK_SIZE |
177 | | * |
178 | | * When calling GROW that's the minimal amount of data |
179 | | * the parser expected to have received. It is not a hard |
180 | | * limit but an optimization when reading strings like Names |
181 | | * It is not strictly needed as long as inputs available characters |
182 | | * are followed by 0, which should be provided by the I/O level |
183 | | */ |
184 | | #define XML_PARSER_CHUNK_SIZE 100 |
185 | | |
186 | | /** |
187 | | * Constant string describing the internal version of the library |
188 | | */ |
189 | | const char *const |
190 | | xmlParserVersion = LIBXML_VERSION_STRING LIBXML_VERSION_EXTRA; |
191 | | |
192 | | /* |
193 | | * List of XML prefixed PI allowed by W3C specs |
194 | | */ |
195 | | |
196 | | static const char* const xmlW3CPIs[] = { |
197 | | "xml-stylesheet", |
198 | | "xml-model", |
199 | | NULL |
200 | | }; |
201 | | |
202 | | |
203 | | /* DEPR void xmlParserHandleReference(xmlParserCtxtPtr ctxt); */ |
204 | | static xmlEntityPtr xmlParseStringPEReference(xmlParserCtxtPtr ctxt, |
205 | | const xmlChar **str); |
206 | | |
207 | | static void |
208 | | xmlCtxtParseEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr ent); |
209 | | |
210 | | static int |
211 | | xmlLoadEntityContent(xmlParserCtxtPtr ctxt, xmlEntityPtr entity); |
212 | | |
213 | | static void |
214 | | xmlParsePERefInternal(xmlParserCtxt *ctxt, int markupDecl); |
215 | | |
216 | | /************************************************************************ |
217 | | * * |
218 | | * Some factorized error routines * |
219 | | * * |
220 | | ************************************************************************/ |
221 | | |
222 | | static void |
223 | 0 | xmlErrMemory(xmlParserCtxtPtr ctxt) { |
224 | 0 | xmlCtxtErrMemory(ctxt); |
225 | 0 | } |
226 | | |
227 | | /** |
228 | | * Handle a redefinition of attribute error |
229 | | * |
230 | | * @param ctxt an XML parser context |
231 | | * @param prefix the attribute prefix |
232 | | * @param localname the attribute localname |
233 | | */ |
234 | | static void |
235 | | xmlErrAttributeDup(xmlParserCtxtPtr ctxt, const xmlChar * prefix, |
236 | | const xmlChar * localname) |
237 | 369k | { |
238 | 369k | if (prefix == NULL) |
239 | 359k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, XML_ERR_ATTRIBUTE_REDEFINED, |
240 | 359k | XML_ERR_FATAL, localname, NULL, NULL, 0, |
241 | 359k | "Attribute %s redefined\n", localname); |
242 | 9.73k | else |
243 | 9.73k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, XML_ERR_ATTRIBUTE_REDEFINED, |
244 | 9.73k | XML_ERR_FATAL, prefix, localname, NULL, 0, |
245 | 9.73k | "Attribute %s:%s redefined\n", prefix, localname); |
246 | 369k | } |
247 | | |
248 | | /** |
249 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
250 | | * |
251 | | * @param ctxt an XML parser context |
252 | | * @param error the error number |
253 | | * @param msg the error message |
254 | | */ |
255 | | static void LIBXML_ATTR_FORMAT(3,0) |
256 | | xmlFatalErrMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
257 | | const char *msg) |
258 | 9.60M | { |
259 | 9.60M | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
260 | 9.60M | NULL, NULL, NULL, 0, "%s", msg); |
261 | 9.60M | } |
262 | | |
263 | | /** |
264 | | * Handle a warning. |
265 | | * |
266 | | * @param ctxt an XML parser context |
267 | | * @param error the error number |
268 | | * @param msg the error message |
269 | | * @param str1 extra data |
270 | | * @param str2 extra data |
271 | | */ |
272 | | void LIBXML_ATTR_FORMAT(3,0) |
273 | | xmlWarningMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
274 | | const char *msg, const xmlChar *str1, const xmlChar *str2) |
275 | 4.05k | { |
276 | 4.05k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_WARNING, |
277 | 4.05k | str1, str2, NULL, 0, msg, str1, str2); |
278 | 4.05k | } |
279 | | |
280 | | #ifdef LIBXML_VALID_ENABLED |
281 | | /** |
282 | | * Handle a validity error. |
283 | | * |
284 | | * @param ctxt an XML parser context |
285 | | * @param error the error number |
286 | | * @param msg the error message |
287 | | * @param str1 extra data |
288 | | * @param str2 extra data |
289 | | */ |
290 | | static void LIBXML_ATTR_FORMAT(3,0) |
291 | | xmlValidityError(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
292 | | const char *msg, const xmlChar *str1, const xmlChar *str2) |
293 | 0 | { |
294 | 0 | ctxt->valid = 0; |
295 | |
|
296 | 0 | xmlCtxtErr(ctxt, NULL, XML_FROM_DTD, error, XML_ERR_ERROR, |
297 | 0 | str1, str2, NULL, 0, msg, str1, str2); |
298 | 0 | } |
299 | | #endif |
300 | | |
301 | | /** |
302 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
303 | | * |
304 | | * @param ctxt an XML parser context |
305 | | * @param error the error number |
306 | | * @param msg the error message |
307 | | * @param val an integer value |
308 | | */ |
309 | | static void LIBXML_ATTR_FORMAT(3,0) |
310 | | xmlFatalErrMsgInt(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
311 | | const char *msg, int val) |
312 | 2.99M | { |
313 | 2.99M | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
314 | 2.99M | NULL, NULL, NULL, val, msg, val); |
315 | 2.99M | } |
316 | | |
317 | | /** |
318 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
319 | | * |
320 | | * @param ctxt an XML parser context |
321 | | * @param error the error number |
322 | | * @param msg the error message |
323 | | * @param str1 an string info |
324 | | * @param val an integer value |
325 | | * @param str2 an string info |
326 | | */ |
327 | | static void LIBXML_ATTR_FORMAT(3,0) |
328 | | xmlFatalErrMsgStrIntStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
329 | | const char *msg, const xmlChar *str1, int val, |
330 | | const xmlChar *str2) |
331 | 283k | { |
332 | 283k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
333 | 283k | str1, str2, NULL, val, msg, str1, val, str2); |
334 | 283k | } |
335 | | |
336 | | /** |
337 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
338 | | * |
339 | | * @param ctxt an XML parser context |
340 | | * @param error the error number |
341 | | * @param msg the error message |
342 | | * @param val a string value |
343 | | */ |
344 | | static void LIBXML_ATTR_FORMAT(3,0) |
345 | | xmlFatalErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
346 | | const char *msg, const xmlChar * val) |
347 | 460k | { |
348 | 460k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
349 | 460k | val, NULL, NULL, 0, msg, val); |
350 | 460k | } |
351 | | |
352 | | /** |
353 | | * Handle a non fatal parser error |
354 | | * |
355 | | * @param ctxt an XML parser context |
356 | | * @param error the error number |
357 | | * @param msg the error message |
358 | | * @param val a string value |
359 | | */ |
360 | | static void LIBXML_ATTR_FORMAT(3,0) |
361 | | xmlErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
362 | | const char *msg, const xmlChar * val) |
363 | 39.3k | { |
364 | 39.3k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_ERROR, |
365 | 39.3k | val, NULL, NULL, 0, msg, val); |
366 | 39.3k | } |
367 | | |
368 | | /** |
369 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
370 | | * |
371 | | * @param ctxt an XML parser context |
372 | | * @param error the error number |
373 | | * @param msg the message |
374 | | * @param info1 extra information string |
375 | | * @param info2 extra information string |
376 | | * @param info3 extra information string |
377 | | */ |
378 | | static void LIBXML_ATTR_FORMAT(3,0) |
379 | | xmlNsErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
380 | | const char *msg, |
381 | | const xmlChar * info1, const xmlChar * info2, |
382 | | const xmlChar * info3) |
383 | 377k | { |
384 | 377k | ctxt->nsWellFormed = 0; |
385 | | |
386 | 377k | xmlCtxtErr(ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_ERROR, |
387 | 377k | info1, info2, info3, 0, msg, info1, info2, info3); |
388 | 377k | } |
389 | | |
390 | | /** |
391 | | * Handle a namespace warning error |
392 | | * |
393 | | * @param ctxt an XML parser context |
394 | | * @param error the error number |
395 | | * @param msg the message |
396 | | * @param info1 extra information string |
397 | | * @param info2 extra information string |
398 | | * @param info3 extra information string |
399 | | */ |
400 | | static void LIBXML_ATTR_FORMAT(3,0) |
401 | | xmlNsWarn(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
402 | | const char *msg, |
403 | | const xmlChar * info1, const xmlChar * info2, |
404 | | const xmlChar * info3) |
405 | 19.3k | { |
406 | 19.3k | xmlCtxtErr(ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_WARNING, |
407 | 19.3k | info1, info2, info3, 0, msg, info1, info2, info3); |
408 | 19.3k | } |
409 | | |
410 | | static void |
411 | 3.46M | xmlSaturatedAdd(unsigned long *dst, unsigned long val) { |
412 | 3.46M | if (val > ULONG_MAX - *dst) |
413 | 0 | *dst = ULONG_MAX; |
414 | 3.46M | else |
415 | 3.46M | *dst += val; |
416 | 3.46M | } |
417 | | |
418 | | static void |
419 | 1.16M | xmlSaturatedAddSizeT(unsigned long *dst, unsigned long val) { |
420 | 1.16M | if (val > ULONG_MAX - *dst) |
421 | 0 | *dst = ULONG_MAX; |
422 | 1.16M | else |
423 | 1.16M | *dst += val; |
424 | 1.16M | } |
425 | | |
426 | | /** |
427 | | * Check for non-linear entity expansion behaviour. |
428 | | * |
429 | | * In some cases like xmlExpandEntityInAttValue, this function is called |
430 | | * for each, possibly nested entity and its unexpanded content length. |
431 | | * |
432 | | * In other cases like #xmlParseReference, it's only called for each |
433 | | * top-level entity with its unexpanded content length plus the sum of |
434 | | * the unexpanded content lengths (plus fixed cost) of all nested |
435 | | * entities. |
436 | | * |
437 | | * Summing the unexpanded lengths also adds the length of the reference. |
438 | | * This is by design. Taking the length of the entity name into account |
439 | | * discourages attacks that try to waste CPU time with abusively long |
440 | | * entity names. See test/recurse/lol6.xml for example. Each call also |
441 | | * adds some fixed cost XML_ENT_FIXED_COST to discourage attacks with |
442 | | * short entities. |
443 | | * |
444 | | * @param ctxt parser context |
445 | | * @param extra sum of unexpanded entity sizes |
446 | | * @returns 1 on error, 0 on success. |
447 | | */ |
448 | | static int |
449 | | xmlParserEntityCheck(xmlParserCtxtPtr ctxt, unsigned long extra) |
450 | 1.15M | { |
451 | 1.15M | unsigned long consumed; |
452 | 1.15M | unsigned long *expandedSize; |
453 | 1.15M | xmlParserInputPtr input = ctxt->input; |
454 | 1.15M | xmlEntityPtr entity = input->entity; |
455 | | |
456 | 1.15M | if ((entity) && (entity->flags & XML_ENT_CHECKED)) |
457 | 452 | return(0); |
458 | | |
459 | | /* |
460 | | * Compute total consumed bytes so far, including input streams of |
461 | | * external entities. |
462 | | */ |
463 | 1.15M | consumed = input->consumed; |
464 | 1.15M | xmlSaturatedAddSizeT(&consumed, input->cur - input->base); |
465 | 1.15M | xmlSaturatedAdd(&consumed, ctxt->sizeentities); |
466 | | |
467 | 1.15M | if (entity) |
468 | 3.33k | expandedSize = &entity->expandedSize; |
469 | 1.15M | else |
470 | 1.15M | expandedSize = &ctxt->sizeentcopy; |
471 | | |
472 | | /* |
473 | | * Add extra cost and some fixed cost. |
474 | | */ |
475 | 1.15M | xmlSaturatedAdd(expandedSize, extra); |
476 | 1.15M | xmlSaturatedAdd(expandedSize, XML_ENT_FIXED_COST); |
477 | | |
478 | | /* |
479 | | * It's important to always use saturation arithmetic when tracking |
480 | | * entity sizes to make the size checks reliable. If "sizeentcopy" |
481 | | * overflows, we have to abort. |
482 | | */ |
483 | 1.15M | if ((*expandedSize > XML_PARSER_ALLOWED_EXPANSION) && |
484 | 1.15M | ((*expandedSize >= ULONG_MAX) || |
485 | 76.0k | (*expandedSize / ctxt->maxAmpl > consumed))) { |
486 | 364 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
487 | 364 | "Maximum entity amplification factor exceeded, see " |
488 | 364 | "xmlCtxtSetMaxAmplification.\n"); |
489 | 364 | xmlHaltParser(ctxt); |
490 | 364 | return(1); |
491 | 364 | } |
492 | | |
493 | 1.15M | return(0); |
494 | 1.15M | } |
495 | | |
496 | | /************************************************************************ |
497 | | * * |
498 | | * Library wide options * |
499 | | * * |
500 | | ************************************************************************/ |
501 | | |
502 | | /** |
503 | | * Examines if the library has been compiled with a given feature. |
504 | | * |
505 | | * @param feature the feature to be examined |
506 | | * @returns zero (0) if the feature does not exist or an unknown |
507 | | * feature is requested, non-zero otherwise. |
508 | | */ |
509 | | int |
510 | | xmlHasFeature(xmlFeature feature) |
511 | 0 | { |
512 | 0 | switch (feature) { |
513 | 0 | case XML_WITH_THREAD: |
514 | 0 | #ifdef LIBXML_THREAD_ENABLED |
515 | 0 | return(1); |
516 | | #else |
517 | | return(0); |
518 | | #endif |
519 | 0 | case XML_WITH_TREE: |
520 | 0 | return(1); |
521 | 0 | case XML_WITH_OUTPUT: |
522 | 0 | #ifdef LIBXML_OUTPUT_ENABLED |
523 | 0 | return(1); |
524 | | #else |
525 | | return(0); |
526 | | #endif |
527 | 0 | case XML_WITH_PUSH: |
528 | 0 | #ifdef LIBXML_PUSH_ENABLED |
529 | 0 | return(1); |
530 | | #else |
531 | | return(0); |
532 | | #endif |
533 | 0 | case XML_WITH_READER: |
534 | 0 | #ifdef LIBXML_READER_ENABLED |
535 | 0 | return(1); |
536 | | #else |
537 | | return(0); |
538 | | #endif |
539 | 0 | case XML_WITH_PATTERN: |
540 | 0 | #ifdef LIBXML_PATTERN_ENABLED |
541 | 0 | return(1); |
542 | | #else |
543 | | return(0); |
544 | | #endif |
545 | 0 | case XML_WITH_WRITER: |
546 | 0 | #ifdef LIBXML_WRITER_ENABLED |
547 | 0 | return(1); |
548 | | #else |
549 | | return(0); |
550 | | #endif |
551 | 0 | case XML_WITH_SAX1: |
552 | 0 | #ifdef LIBXML_SAX1_ENABLED |
553 | 0 | return(1); |
554 | | #else |
555 | | return(0); |
556 | | #endif |
557 | 0 | case XML_WITH_HTTP: |
558 | 0 | return(0); |
559 | 0 | case XML_WITH_VALID: |
560 | 0 | #ifdef LIBXML_VALID_ENABLED |
561 | 0 | return(1); |
562 | | #else |
563 | | return(0); |
564 | | #endif |
565 | 0 | case XML_WITH_HTML: |
566 | 0 | #ifdef LIBXML_HTML_ENABLED |
567 | 0 | return(1); |
568 | | #else |
569 | | return(0); |
570 | | #endif |
571 | 0 | case XML_WITH_LEGACY: |
572 | 0 | return(0); |
573 | 0 | case XML_WITH_C14N: |
574 | 0 | #ifdef LIBXML_C14N_ENABLED |
575 | 0 | return(1); |
576 | | #else |
577 | | return(0); |
578 | | #endif |
579 | 0 | case XML_WITH_CATALOG: |
580 | 0 | #ifdef LIBXML_CATALOG_ENABLED |
581 | 0 | return(1); |
582 | | #else |
583 | | return(0); |
584 | | #endif |
585 | 0 | case XML_WITH_XPATH: |
586 | 0 | #ifdef LIBXML_XPATH_ENABLED |
587 | 0 | return(1); |
588 | | #else |
589 | | return(0); |
590 | | #endif |
591 | 0 | case XML_WITH_XPTR: |
592 | 0 | #ifdef LIBXML_XPTR_ENABLED |
593 | 0 | return(1); |
594 | | #else |
595 | | return(0); |
596 | | #endif |
597 | 0 | case XML_WITH_XINCLUDE: |
598 | 0 | #ifdef LIBXML_XINCLUDE_ENABLED |
599 | 0 | return(1); |
600 | | #else |
601 | | return(0); |
602 | | #endif |
603 | 0 | case XML_WITH_ICONV: |
604 | 0 | #ifdef LIBXML_ICONV_ENABLED |
605 | 0 | return(1); |
606 | | #else |
607 | | return(0); |
608 | | #endif |
609 | 0 | case XML_WITH_ISO8859X: |
610 | 0 | #ifdef LIBXML_ISO8859X_ENABLED |
611 | 0 | return(1); |
612 | | #else |
613 | | return(0); |
614 | | #endif |
615 | 0 | case XML_WITH_UNICODE: |
616 | 0 | return(0); |
617 | 0 | case XML_WITH_REGEXP: |
618 | 0 | #ifdef LIBXML_REGEXP_ENABLED |
619 | 0 | return(1); |
620 | | #else |
621 | | return(0); |
622 | | #endif |
623 | 0 | case XML_WITH_AUTOMATA: |
624 | 0 | #ifdef LIBXML_REGEXP_ENABLED |
625 | 0 | return(1); |
626 | | #else |
627 | | return(0); |
628 | | #endif |
629 | 0 | case XML_WITH_EXPR: |
630 | 0 | return(0); |
631 | 0 | case XML_WITH_RELAXNG: |
632 | 0 | #ifdef LIBXML_RELAXNG_ENABLED |
633 | 0 | return(1); |
634 | | #else |
635 | | return(0); |
636 | | #endif |
637 | 0 | case XML_WITH_SCHEMAS: |
638 | 0 | #ifdef LIBXML_SCHEMAS_ENABLED |
639 | 0 | return(1); |
640 | | #else |
641 | | return(0); |
642 | | #endif |
643 | 0 | case XML_WITH_SCHEMATRON: |
644 | | #ifdef LIBXML_SCHEMATRON_ENABLED |
645 | | return(1); |
646 | | #else |
647 | 0 | return(0); |
648 | 0 | #endif |
649 | 0 | case XML_WITH_MODULES: |
650 | 0 | #ifdef LIBXML_MODULES_ENABLED |
651 | 0 | return(1); |
652 | | #else |
653 | | return(0); |
654 | | #endif |
655 | 0 | case XML_WITH_DEBUG: |
656 | 0 | #ifdef LIBXML_DEBUG_ENABLED |
657 | 0 | return(1); |
658 | | #else |
659 | | return(0); |
660 | | #endif |
661 | 0 | case XML_WITH_DEBUG_MEM: |
662 | 0 | return(0); |
663 | 0 | case XML_WITH_ZLIB: |
664 | | #ifdef LIBXML_ZLIB_ENABLED |
665 | | return(1); |
666 | | #else |
667 | 0 | return(0); |
668 | 0 | #endif |
669 | 0 | case XML_WITH_LZMA: |
670 | | #ifdef LIBXML_LZMA_ENABLED |
671 | | return(1); |
672 | | #else |
673 | 0 | return(0); |
674 | 0 | #endif |
675 | 0 | case XML_WITH_ICU: |
676 | | #ifdef LIBXML_ICU_ENABLED |
677 | | return(1); |
678 | | #else |
679 | 0 | return(0); |
680 | 0 | #endif |
681 | 0 | default: |
682 | 0 | break; |
683 | 0 | } |
684 | 0 | return(0); |
685 | 0 | } |
686 | | |
687 | | /************************************************************************ |
688 | | * * |
689 | | * Simple string buffer * |
690 | | * * |
691 | | ************************************************************************/ |
692 | | |
693 | | typedef struct { |
694 | | xmlChar *mem; |
695 | | unsigned size; |
696 | | unsigned cap; /* size < cap */ |
697 | | unsigned max; /* size <= max */ |
698 | | xmlParserErrors code; |
699 | | } xmlSBuf; |
700 | | |
701 | | static void |
702 | 672k | xmlSBufInit(xmlSBuf *buf, unsigned max) { |
703 | 672k | buf->mem = NULL; |
704 | 672k | buf->size = 0; |
705 | 672k | buf->cap = 0; |
706 | 672k | buf->max = max; |
707 | 672k | buf->code = XML_ERR_OK; |
708 | 672k | } |
709 | | |
710 | | static int |
711 | 169k | xmlSBufGrow(xmlSBuf *buf, unsigned len) { |
712 | 169k | xmlChar *mem; |
713 | 169k | unsigned cap; |
714 | | |
715 | 169k | if (len >= UINT_MAX / 2 - buf->size) { |
716 | 0 | if (buf->code == XML_ERR_OK) |
717 | 0 | buf->code = XML_ERR_RESOURCE_LIMIT; |
718 | 0 | return(-1); |
719 | 0 | } |
720 | | |
721 | 169k | cap = (buf->size + len) * 2; |
722 | 169k | if (cap < 240) |
723 | 125k | cap = 240; |
724 | | |
725 | 169k | mem = xmlRealloc(buf->mem, cap); |
726 | 169k | if (mem == NULL) { |
727 | 0 | buf->code = XML_ERR_NO_MEMORY; |
728 | 0 | return(-1); |
729 | 0 | } |
730 | | |
731 | 169k | buf->mem = mem; |
732 | 169k | buf->cap = cap; |
733 | | |
734 | 169k | return(0); |
735 | 169k | } |
736 | | |
737 | | static void |
738 | 86.9M | xmlSBufAddString(xmlSBuf *buf, const xmlChar *str, unsigned len) { |
739 | 86.9M | if (buf->max - buf->size < len) { |
740 | 0 | if (buf->code == XML_ERR_OK) |
741 | 0 | buf->code = XML_ERR_RESOURCE_LIMIT; |
742 | 0 | return; |
743 | 0 | } |
744 | | |
745 | 86.9M | if (buf->cap - buf->size <= len) { |
746 | 159k | if (xmlSBufGrow(buf, len) < 0) |
747 | 0 | return; |
748 | 159k | } |
749 | | |
750 | 86.9M | if (len > 0) |
751 | 86.9M | memcpy(buf->mem + buf->size, str, len); |
752 | 86.9M | buf->size += len; |
753 | 86.9M | } |
754 | | |
755 | | static void |
756 | 85.1M | xmlSBufAddCString(xmlSBuf *buf, const char *str, unsigned len) { |
757 | 85.1M | xmlSBufAddString(buf, (const xmlChar *) str, len); |
758 | 85.1M | } |
759 | | |
760 | | static void |
761 | 142k | xmlSBufAddChar(xmlSBuf *buf, int c) { |
762 | 142k | xmlChar *end; |
763 | | |
764 | 142k | if (buf->max - buf->size < 4) { |
765 | 0 | if (buf->code == XML_ERR_OK) |
766 | 0 | buf->code = XML_ERR_RESOURCE_LIMIT; |
767 | 0 | return; |
768 | 0 | } |
769 | | |
770 | 142k | if (buf->cap - buf->size <= 4) { |
771 | 10.7k | if (xmlSBufGrow(buf, 4) < 0) |
772 | 0 | return; |
773 | 10.7k | } |
774 | | |
775 | 142k | end = buf->mem + buf->size; |
776 | | |
777 | 142k | if (c < 0x80) { |
778 | 59.6k | *end = (xmlChar) c; |
779 | 59.6k | buf->size += 1; |
780 | 82.4k | } else { |
781 | 82.4k | buf->size += xmlCopyCharMultiByte(end, c); |
782 | 82.4k | } |
783 | 142k | } |
784 | | |
785 | | static void |
786 | 64.8M | xmlSBufAddReplChar(xmlSBuf *buf) { |
787 | 64.8M | xmlSBufAddCString(buf, "\xEF\xBF\xBD", 3); |
788 | 64.8M | } |
789 | | |
790 | | static void |
791 | 0 | xmlSBufReportError(xmlSBuf *buf, xmlParserCtxtPtr ctxt, const char *errMsg) { |
792 | 0 | if (buf->code == XML_ERR_NO_MEMORY) |
793 | 0 | xmlCtxtErrMemory(ctxt); |
794 | 0 | else |
795 | 0 | xmlFatalErr(ctxt, buf->code, errMsg); |
796 | 0 | } |
797 | | |
798 | | static xmlChar * |
799 | | xmlSBufFinish(xmlSBuf *buf, int *sizeOut, xmlParserCtxtPtr ctxt, |
800 | 138k | const char *errMsg) { |
801 | 138k | if (buf->mem == NULL) { |
802 | 15.0k | buf->mem = xmlMalloc(1); |
803 | 15.0k | if (buf->mem == NULL) { |
804 | 0 | buf->code = XML_ERR_NO_MEMORY; |
805 | 15.0k | } else { |
806 | 15.0k | buf->mem[0] = 0; |
807 | 15.0k | } |
808 | 123k | } else { |
809 | 123k | buf->mem[buf->size] = 0; |
810 | 123k | } |
811 | | |
812 | 138k | if (buf->code == XML_ERR_OK) { |
813 | 138k | if (sizeOut != NULL) |
814 | 101k | *sizeOut = buf->size; |
815 | 138k | return(buf->mem); |
816 | 138k | } |
817 | | |
818 | 0 | xmlSBufReportError(buf, ctxt, errMsg); |
819 | |
|
820 | 0 | xmlFree(buf->mem); |
821 | |
|
822 | 0 | if (sizeOut != NULL) |
823 | 0 | *sizeOut = 0; |
824 | 0 | return(NULL); |
825 | 138k | } |
826 | | |
827 | | static void |
828 | 517k | xmlSBufCleanup(xmlSBuf *buf, xmlParserCtxtPtr ctxt, const char *errMsg) { |
829 | 517k | if (buf->code != XML_ERR_OK) |
830 | 0 | xmlSBufReportError(buf, ctxt, errMsg); |
831 | | |
832 | 517k | xmlFree(buf->mem); |
833 | 517k | } |
834 | | |
835 | | static int |
836 | | xmlUTF8MultibyteLen(xmlParserCtxtPtr ctxt, const xmlChar *str, |
837 | 91.4M | const char *errMsg) { |
838 | 91.4M | int c = str[0]; |
839 | 91.4M | int c1 = str[1]; |
840 | | |
841 | 91.4M | if ((c1 & 0xC0) != 0x80) |
842 | 3.90M | goto encoding_error; |
843 | | |
844 | 87.5M | if (c < 0xE0) { |
845 | | /* 2-byte sequence */ |
846 | 85.8M | if (c < 0xC2) |
847 | 53.3M | goto encoding_error; |
848 | | |
849 | 32.4M | return(2); |
850 | 85.8M | } else { |
851 | 1.66M | int c2 = str[2]; |
852 | | |
853 | 1.66M | if ((c2 & 0xC0) != 0x80) |
854 | 6.23k | goto encoding_error; |
855 | | |
856 | 1.65M | if (c < 0xF0) { |
857 | | /* 3-byte sequence */ |
858 | 1.64M | if (c == 0xE0) { |
859 | | /* overlong */ |
860 | 1.10M | if (c1 < 0xA0) |
861 | 899 | goto encoding_error; |
862 | 1.10M | } else if (c == 0xED) { |
863 | | /* surrogate */ |
864 | 768 | if (c1 >= 0xA0) |
865 | 396 | goto encoding_error; |
866 | 541k | } else if (c == 0xEF) { |
867 | | /* U+FFFE and U+FFFF are invalid Chars */ |
868 | 38.7k | if ((c1 == 0xBF) && (c2 >= 0xBE)) |
869 | 1.58k | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, errMsg); |
870 | 38.7k | } |
871 | | |
872 | 1.64M | return(3); |
873 | 1.64M | } else { |
874 | | /* 4-byte sequence */ |
875 | 10.4k | if ((str[3] & 0xC0) != 0x80) |
876 | 1.42k | goto encoding_error; |
877 | 9.00k | if (c == 0xF0) { |
878 | | /* overlong */ |
879 | 1.43k | if (c1 < 0x90) |
880 | 288 | goto encoding_error; |
881 | 7.57k | } else if (c >= 0xF4) { |
882 | | /* greater than 0x10FFFF */ |
883 | 3.60k | if ((c > 0xF4) || (c1 >= 0x90)) |
884 | 2.91k | goto encoding_error; |
885 | 3.60k | } |
886 | | |
887 | 5.80k | return(4); |
888 | 9.00k | } |
889 | 1.65M | } |
890 | | |
891 | 57.3M | encoding_error: |
892 | | /* Only report the first error */ |
893 | 57.3M | if ((ctxt->input->flags & XML_INPUT_ENCODING_ERROR) == 0) { |
894 | 1.69k | xmlCtxtErrIO(ctxt, XML_ERR_INVALID_ENCODING, NULL); |
895 | 1.69k | ctxt->input->flags |= XML_INPUT_ENCODING_ERROR; |
896 | 1.69k | } |
897 | | |
898 | 57.3M | return(0); |
899 | 87.5M | } |
900 | | |
901 | | /************************************************************************ |
902 | | * * |
903 | | * SAX2 defaulted attributes handling * |
904 | | * * |
905 | | ************************************************************************/ |
906 | | |
907 | | /** |
908 | | * Final initialization of the parser context before starting to parse. |
909 | | * |
910 | | * This accounts for users modifying struct members of parser context |
911 | | * directly. |
912 | | * |
913 | | * @param ctxt an XML parser context |
914 | | */ |
915 | | static void |
916 | 14.5k | xmlCtxtInitializeLate(xmlParserCtxtPtr ctxt) { |
917 | 14.5k | xmlSAXHandlerPtr sax; |
918 | | |
919 | | /* Avoid unused variable warning if features are disabled. */ |
920 | 14.5k | (void) sax; |
921 | | |
922 | | /* |
923 | | * Changing the SAX struct directly is still widespread practice |
924 | | * in internal and external code. |
925 | | */ |
926 | 14.5k | if (ctxt == NULL) return; |
927 | 14.5k | sax = ctxt->sax; |
928 | 14.5k | #ifdef LIBXML_SAX1_ENABLED |
929 | | /* |
930 | | * Only enable SAX2 if there SAX2 element handlers, except when there |
931 | | * are no element handlers at all. |
932 | | */ |
933 | 14.5k | if (((ctxt->options & XML_PARSE_SAX1) == 0) && |
934 | 14.5k | (sax) && |
935 | 14.5k | (sax->initialized == XML_SAX2_MAGIC) && |
936 | 14.5k | ((sax->startElementNs != NULL) || |
937 | 14.5k | (sax->endElementNs != NULL) || |
938 | 14.5k | ((sax->startElement == NULL) && (sax->endElement == NULL)))) |
939 | 14.5k | ctxt->sax2 = 1; |
940 | | #else |
941 | | ctxt->sax2 = 1; |
942 | | #endif /* LIBXML_SAX1_ENABLED */ |
943 | | |
944 | | /* |
945 | | * Some users replace the dictionary directly in the context struct. |
946 | | * We really need an API function to do that cleanly. |
947 | | */ |
948 | 14.5k | ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); |
949 | 14.5k | ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); |
950 | 14.5k | ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); |
951 | 14.5k | if ((ctxt->str_xml==NULL) || (ctxt->str_xmlns==NULL) || |
952 | 14.5k | (ctxt->str_xml_ns == NULL)) { |
953 | 0 | xmlErrMemory(ctxt); |
954 | 0 | } |
955 | | |
956 | 14.5k | xmlDictSetLimit(ctxt->dict, |
957 | 14.5k | (ctxt->options & XML_PARSE_HUGE) ? |
958 | 14.5k | 0 : |
959 | 14.5k | XML_MAX_DICTIONARY_LIMIT); |
960 | | |
961 | 14.5k | #ifdef LIBXML_VALID_ENABLED |
962 | 14.5k | if (ctxt->validate) |
963 | 0 | ctxt->vctxt.flags |= XML_VCTXT_VALIDATE; |
964 | 14.5k | else |
965 | 14.5k | ctxt->vctxt.flags &= ~XML_VCTXT_VALIDATE; |
966 | 14.5k | #endif /* LIBXML_VALID_ENABLED */ |
967 | 14.5k | } |
968 | | |
969 | | typedef struct { |
970 | | xmlHashedString prefix; |
971 | | xmlHashedString name; |
972 | | xmlHashedString value; |
973 | | const xmlChar *valueEnd; |
974 | | int external; |
975 | | int expandedSize; |
976 | | } xmlDefAttr; |
977 | | |
978 | | typedef struct _xmlDefAttrs xmlDefAttrs; |
979 | | typedef xmlDefAttrs *xmlDefAttrsPtr; |
980 | | struct _xmlDefAttrs { |
981 | | int nbAttrs; /* number of defaulted attributes on that element */ |
982 | | int maxAttrs; /* the size of the array */ |
983 | | #if __STDC_VERSION__ >= 199901L |
984 | | /* Using a C99 flexible array member avoids UBSan errors. */ |
985 | | xmlDefAttr attrs[] ATTRIBUTE_COUNTED_BY(maxAttrs); |
986 | | #else |
987 | | xmlDefAttr attrs[1]; |
988 | | #endif |
989 | | }; |
990 | | |
991 | | /** |
992 | | * Normalize the space in non CDATA attribute values: |
993 | | * If the attribute type is not CDATA, then the XML processor MUST further |
994 | | * process the normalized attribute value by discarding any leading and |
995 | | * trailing space (\#x20) characters, and by replacing sequences of space |
996 | | * (\#x20) characters by a single space (\#x20) character. |
997 | | * Note that the size of dst need to be at least src, and if one doesn't need |
998 | | * to preserve dst (and it doesn't come from a dictionary or read-only) then |
999 | | * passing src as dst is just fine. |
1000 | | * |
1001 | | * @param src the source string |
1002 | | * @param dst the target string |
1003 | | * @returns a pointer to the normalized value (dst) or NULL if no conversion |
1004 | | * is needed. |
1005 | | */ |
1006 | | static xmlChar * |
1007 | | xmlAttrNormalizeSpace(const xmlChar *src, xmlChar *dst) |
1008 | 14.8k | { |
1009 | 14.8k | if ((src == NULL) || (dst == NULL)) |
1010 | 0 | return(NULL); |
1011 | | |
1012 | 16.8k | while (*src == 0x20) src++; |
1013 | 14.4M | while (*src != 0) { |
1014 | 14.3M | if (*src == 0x20) { |
1015 | 146k | while (*src == 0x20) src++; |
1016 | 10.9k | if (*src != 0) |
1017 | 10.3k | *dst++ = 0x20; |
1018 | 14.3M | } else { |
1019 | 14.3M | *dst++ = *src++; |
1020 | 14.3M | } |
1021 | 14.3M | } |
1022 | 14.8k | *dst = 0; |
1023 | 14.8k | if (dst == src) |
1024 | 13.4k | return(NULL); |
1025 | 1.34k | return(dst); |
1026 | 14.8k | } |
1027 | | |
1028 | | /** |
1029 | | * Add a defaulted attribute for an element |
1030 | | * |
1031 | | * @param ctxt an XML parser context |
1032 | | * @param fullname the element fullname |
1033 | | * @param fullattr the attribute fullname |
1034 | | * @param value the attribute value |
1035 | | */ |
1036 | | static void |
1037 | | xmlAddDefAttrs(xmlParserCtxtPtr ctxt, |
1038 | | const xmlChar *fullname, |
1039 | | const xmlChar *fullattr, |
1040 | 15.8k | const xmlChar *value) { |
1041 | 15.8k | xmlDefAttrsPtr defaults; |
1042 | 15.8k | xmlDefAttr *attr; |
1043 | 15.8k | int len, expandedSize; |
1044 | 15.8k | xmlHashedString name; |
1045 | 15.8k | xmlHashedString prefix; |
1046 | 15.8k | xmlHashedString hvalue; |
1047 | 15.8k | const xmlChar *localname; |
1048 | | |
1049 | | /* |
1050 | | * Allows to detect attribute redefinitions |
1051 | | */ |
1052 | 15.8k | if (ctxt->attsSpecial != NULL) { |
1053 | 14.3k | if (xmlHashLookup2(ctxt->attsSpecial, fullname, fullattr) != NULL) |
1054 | 5.06k | return; |
1055 | 14.3k | } |
1056 | | |
1057 | 10.8k | if (ctxt->attsDefault == NULL) { |
1058 | 1.53k | ctxt->attsDefault = xmlHashCreateDict(10, ctxt->dict); |
1059 | 1.53k | if (ctxt->attsDefault == NULL) |
1060 | 0 | goto mem_error; |
1061 | 1.53k | } |
1062 | | |
1063 | | /* |
1064 | | * split the element name into prefix:localname , the string found |
1065 | | * are within the DTD and then not associated to namespace names. |
1066 | | */ |
1067 | 10.8k | localname = xmlSplitQName3(fullname, &len); |
1068 | 10.8k | if (localname == NULL) { |
1069 | 7.34k | name = xmlDictLookupHashed(ctxt->dict, fullname, -1); |
1070 | 7.34k | prefix.name = NULL; |
1071 | 7.34k | } else { |
1072 | 3.45k | name = xmlDictLookupHashed(ctxt->dict, localname, -1); |
1073 | 3.45k | prefix = xmlDictLookupHashed(ctxt->dict, fullname, len); |
1074 | 3.45k | if (prefix.name == NULL) |
1075 | 0 | goto mem_error; |
1076 | 3.45k | } |
1077 | 10.8k | if (name.name == NULL) |
1078 | 0 | goto mem_error; |
1079 | | |
1080 | | /* |
1081 | | * make sure there is some storage |
1082 | | */ |
1083 | 10.8k | defaults = xmlHashLookup2(ctxt->attsDefault, name.name, prefix.name); |
1084 | 10.8k | if ((defaults == NULL) || |
1085 | 10.8k | (defaults->nbAttrs >= defaults->maxAttrs)) { |
1086 | 3.71k | xmlDefAttrsPtr temp; |
1087 | 3.71k | int newSize; |
1088 | | |
1089 | 3.71k | if (defaults == NULL) { |
1090 | 2.40k | newSize = 4; |
1091 | 2.40k | } else { |
1092 | 1.30k | if ((defaults->maxAttrs >= XML_MAX_ATTRS) || |
1093 | 1.30k | ((size_t) defaults->maxAttrs > |
1094 | 1.30k | SIZE_MAX / 2 / sizeof(temp[0]) - sizeof(*defaults))) |
1095 | 0 | goto mem_error; |
1096 | | |
1097 | 1.30k | if (defaults->maxAttrs > XML_MAX_ATTRS / 2) |
1098 | 0 | newSize = XML_MAX_ATTRS; |
1099 | 1.30k | else |
1100 | 1.30k | newSize = defaults->maxAttrs * 2; |
1101 | 1.30k | } |
1102 | 3.71k | temp = xmlRealloc(defaults, |
1103 | 3.71k | sizeof(*defaults) + newSize * sizeof(xmlDefAttr)); |
1104 | 3.71k | if (temp == NULL) |
1105 | 0 | goto mem_error; |
1106 | 3.71k | if (defaults == NULL) |
1107 | 2.40k | temp->nbAttrs = 0; |
1108 | 3.71k | temp->maxAttrs = newSize; |
1109 | 3.71k | defaults = temp; |
1110 | 3.71k | if (xmlHashUpdateEntry2(ctxt->attsDefault, name.name, prefix.name, |
1111 | 3.71k | defaults, NULL) < 0) { |
1112 | 0 | xmlFree(defaults); |
1113 | 0 | goto mem_error; |
1114 | 0 | } |
1115 | 3.71k | } |
1116 | | |
1117 | | /* |
1118 | | * Split the attribute name into prefix:localname , the string found |
1119 | | * are within the DTD and hen not associated to namespace names. |
1120 | | */ |
1121 | 10.8k | localname = xmlSplitQName3(fullattr, &len); |
1122 | 10.8k | if (localname == NULL) { |
1123 | 5.98k | name = xmlDictLookupHashed(ctxt->dict, fullattr, -1); |
1124 | 5.98k | prefix.name = NULL; |
1125 | 5.98k | } else { |
1126 | 4.81k | name = xmlDictLookupHashed(ctxt->dict, localname, -1); |
1127 | 4.81k | prefix = xmlDictLookupHashed(ctxt->dict, fullattr, len); |
1128 | 4.81k | if (prefix.name == NULL) |
1129 | 0 | goto mem_error; |
1130 | 4.81k | } |
1131 | 10.8k | if (name.name == NULL) |
1132 | 0 | goto mem_error; |
1133 | | |
1134 | | /* intern the string and precompute the end */ |
1135 | 10.8k | len = strlen((const char *) value); |
1136 | 10.8k | hvalue = xmlDictLookupHashed(ctxt->dict, value, len); |
1137 | 10.8k | if (hvalue.name == NULL) |
1138 | 0 | goto mem_error; |
1139 | | |
1140 | 10.8k | expandedSize = strlen((const char *) name.name); |
1141 | 10.8k | if (prefix.name != NULL) |
1142 | 4.81k | expandedSize += strlen((const char *) prefix.name); |
1143 | 10.8k | expandedSize += len; |
1144 | | |
1145 | 10.8k | attr = &defaults->attrs[defaults->nbAttrs++]; |
1146 | 10.8k | attr->name = name; |
1147 | 10.8k | attr->prefix = prefix; |
1148 | 10.8k | attr->value = hvalue; |
1149 | 10.8k | attr->valueEnd = hvalue.name + len; |
1150 | 10.8k | attr->external = PARSER_EXTERNAL(ctxt); |
1151 | 10.8k | attr->expandedSize = expandedSize; |
1152 | | |
1153 | 10.8k | return; |
1154 | | |
1155 | 0 | mem_error: |
1156 | 0 | xmlErrMemory(ctxt); |
1157 | 0 | } |
1158 | | |
1159 | | /** |
1160 | | * Register this attribute type |
1161 | | * |
1162 | | * @param ctxt an XML parser context |
1163 | | * @param fullname the element fullname |
1164 | | * @param fullattr the attribute fullname |
1165 | | * @param type the attribute type |
1166 | | */ |
1167 | | static void |
1168 | | xmlAddSpecialAttr(xmlParserCtxtPtr ctxt, |
1169 | | const xmlChar *fullname, |
1170 | | const xmlChar *fullattr, |
1171 | | int type) |
1172 | 16.7k | { |
1173 | 16.7k | if (ctxt->attsSpecial == NULL) { |
1174 | 1.74k | ctxt->attsSpecial = xmlHashCreateDict(10, ctxt->dict); |
1175 | 1.74k | if (ctxt->attsSpecial == NULL) |
1176 | 0 | goto mem_error; |
1177 | 1.74k | } |
1178 | | |
1179 | 16.7k | if (PARSER_EXTERNAL(ctxt)) |
1180 | 0 | type |= XML_SPECIAL_EXTERNAL; |
1181 | | |
1182 | 16.7k | if (xmlHashAdd2(ctxt->attsSpecial, fullname, fullattr, |
1183 | 16.7k | XML_INT_TO_PTR(type)) < 0) |
1184 | 0 | goto mem_error; |
1185 | 16.7k | return; |
1186 | | |
1187 | 16.7k | mem_error: |
1188 | 0 | xmlErrMemory(ctxt); |
1189 | 0 | } |
1190 | | |
1191 | | /** |
1192 | | * Removes CDATA attributes from the special attribute table |
1193 | | */ |
1194 | | static void |
1195 | | xmlCleanSpecialAttrCallback(void *payload, void *data, |
1196 | | const xmlChar *fullname, const xmlChar *fullattr, |
1197 | 11.1k | const xmlChar *unused ATTRIBUTE_UNUSED) { |
1198 | 11.1k | xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) data; |
1199 | | |
1200 | 11.1k | if (XML_PTR_TO_INT(payload) == XML_ATTRIBUTE_CDATA) { |
1201 | 827 | xmlHashRemoveEntry2(ctxt->attsSpecial, fullname, fullattr, NULL); |
1202 | 827 | } |
1203 | 11.1k | } |
1204 | | |
1205 | | /** |
1206 | | * Trim the list of attributes defined to remove all those of type |
1207 | | * CDATA as they are not special. This call should be done when finishing |
1208 | | * to parse the DTD and before starting to parse the document root. |
1209 | | * |
1210 | | * @param ctxt an XML parser context |
1211 | | */ |
1212 | | static void |
1213 | | xmlCleanSpecialAttr(xmlParserCtxtPtr ctxt) |
1214 | 5.92k | { |
1215 | 5.92k | if (ctxt->attsSpecial == NULL) |
1216 | 4.17k | return; |
1217 | | |
1218 | 1.74k | xmlHashScanFull(ctxt->attsSpecial, xmlCleanSpecialAttrCallback, ctxt); |
1219 | | |
1220 | 1.74k | if (xmlHashSize(ctxt->attsSpecial) == 0) { |
1221 | 59 | xmlHashFree(ctxt->attsSpecial, NULL); |
1222 | 59 | ctxt->attsSpecial = NULL; |
1223 | 59 | } |
1224 | 1.74k | } |
1225 | | |
1226 | | /** |
1227 | | * Checks that the value conforms to the LanguageID production: |
1228 | | * |
1229 | | * @deprecated Internal function, do not use. |
1230 | | * |
1231 | | * NOTE: this is somewhat deprecated, those productions were removed from |
1232 | | * the XML Second edition. |
1233 | | * |
1234 | | * [33] LanguageID ::= Langcode ('-' Subcode)* |
1235 | | * [34] Langcode ::= ISO639Code | IanaCode | UserCode |
1236 | | * [35] ISO639Code ::= ([a-z] | [A-Z]) ([a-z] | [A-Z]) |
1237 | | * [36] IanaCode ::= ('i' | 'I') '-' ([a-z] | [A-Z])+ |
1238 | | * [37] UserCode ::= ('x' | 'X') '-' ([a-z] | [A-Z])+ |
1239 | | * [38] Subcode ::= ([a-z] | [A-Z])+ |
1240 | | * |
1241 | | * The current REC reference the successors of RFC 1766, currently 5646 |
1242 | | * |
1243 | | * http://www.rfc-editor.org/rfc/rfc5646.txt |
1244 | | * |
1245 | | * langtag = language |
1246 | | * ["-" script] |
1247 | | * ["-" region] |
1248 | | * *("-" variant) |
1249 | | * *("-" extension) |
1250 | | * ["-" privateuse] |
1251 | | * language = 2*3ALPHA ; shortest ISO 639 code |
1252 | | * ["-" extlang] ; sometimes followed by |
1253 | | * ; extended language subtags |
1254 | | * / 4ALPHA ; or reserved for future use |
1255 | | * / 5*8ALPHA ; or registered language subtag |
1256 | | * |
1257 | | * extlang = 3ALPHA ; selected ISO 639 codes |
1258 | | * *2("-" 3ALPHA) ; permanently reserved |
1259 | | * |
1260 | | * script = 4ALPHA ; ISO 15924 code |
1261 | | * |
1262 | | * region = 2ALPHA ; ISO 3166-1 code |
1263 | | * / 3DIGIT ; UN M.49 code |
1264 | | * |
1265 | | * variant = 5*8alphanum ; registered variants |
1266 | | * / (DIGIT 3alphanum) |
1267 | | * |
1268 | | * extension = singleton 1*("-" (2*8alphanum)) |
1269 | | * |
1270 | | * ; Single alphanumerics |
1271 | | * ; "x" reserved for private use |
1272 | | * singleton = DIGIT ; 0 - 9 |
1273 | | * / %x41-57 ; A - W |
1274 | | * / %x59-5A ; Y - Z |
1275 | | * / %x61-77 ; a - w |
1276 | | * / %x79-7A ; y - z |
1277 | | * |
1278 | | * it sounds right to still allow Irregular i-xxx IANA and user codes too |
1279 | | * The parser below doesn't try to cope with extension or privateuse |
1280 | | * that could be added but that's not interoperable anyway |
1281 | | * |
1282 | | * @param lang pointer to the string value |
1283 | | * @returns 1 if correct 0 otherwise |
1284 | | **/ |
1285 | | int |
1286 | | xmlCheckLanguageID(const xmlChar * lang) |
1287 | 0 | { |
1288 | 0 | const xmlChar *cur = lang, *nxt; |
1289 | |
|
1290 | 0 | if (cur == NULL) |
1291 | 0 | return (0); |
1292 | 0 | if (((cur[0] == 'i') && (cur[1] == '-')) || |
1293 | 0 | ((cur[0] == 'I') && (cur[1] == '-')) || |
1294 | 0 | ((cur[0] == 'x') && (cur[1] == '-')) || |
1295 | 0 | ((cur[0] == 'X') && (cur[1] == '-'))) { |
1296 | | /* |
1297 | | * Still allow IANA code and user code which were coming |
1298 | | * from the previous version of the XML-1.0 specification |
1299 | | * it's deprecated but we should not fail |
1300 | | */ |
1301 | 0 | cur += 2; |
1302 | 0 | while (((cur[0] >= 'A') && (cur[0] <= 'Z')) || |
1303 | 0 | ((cur[0] >= 'a') && (cur[0] <= 'z'))) |
1304 | 0 | cur++; |
1305 | 0 | return(cur[0] == 0); |
1306 | 0 | } |
1307 | 0 | nxt = cur; |
1308 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1309 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1310 | 0 | nxt++; |
1311 | 0 | if (nxt - cur >= 4) { |
1312 | | /* |
1313 | | * Reserved |
1314 | | */ |
1315 | 0 | if ((nxt - cur > 8) || (nxt[0] != 0)) |
1316 | 0 | return(0); |
1317 | 0 | return(1); |
1318 | 0 | } |
1319 | 0 | if (nxt - cur < 2) |
1320 | 0 | return(0); |
1321 | | /* we got an ISO 639 code */ |
1322 | 0 | if (nxt[0] == 0) |
1323 | 0 | return(1); |
1324 | 0 | if (nxt[0] != '-') |
1325 | 0 | return(0); |
1326 | | |
1327 | 0 | nxt++; |
1328 | 0 | cur = nxt; |
1329 | | /* now we can have extlang or script or region or variant */ |
1330 | 0 | if ((nxt[0] >= '0') && (nxt[0] <= '9')) |
1331 | 0 | goto region_m49; |
1332 | | |
1333 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1334 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1335 | 0 | nxt++; |
1336 | 0 | if (nxt - cur == 4) |
1337 | 0 | goto script; |
1338 | 0 | if (nxt - cur == 2) |
1339 | 0 | goto region; |
1340 | 0 | if ((nxt - cur >= 5) && (nxt - cur <= 8)) |
1341 | 0 | goto variant; |
1342 | 0 | if (nxt - cur != 3) |
1343 | 0 | return(0); |
1344 | | /* we parsed an extlang */ |
1345 | 0 | if (nxt[0] == 0) |
1346 | 0 | return(1); |
1347 | 0 | if (nxt[0] != '-') |
1348 | 0 | return(0); |
1349 | | |
1350 | 0 | nxt++; |
1351 | 0 | cur = nxt; |
1352 | | /* now we can have script or region or variant */ |
1353 | 0 | if ((nxt[0] >= '0') && (nxt[0] <= '9')) |
1354 | 0 | goto region_m49; |
1355 | | |
1356 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1357 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1358 | 0 | nxt++; |
1359 | 0 | if (nxt - cur == 2) |
1360 | 0 | goto region; |
1361 | 0 | if ((nxt - cur >= 5) && (nxt - cur <= 8)) |
1362 | 0 | goto variant; |
1363 | 0 | if (nxt - cur != 4) |
1364 | 0 | return(0); |
1365 | | /* we parsed a script */ |
1366 | 0 | script: |
1367 | 0 | if (nxt[0] == 0) |
1368 | 0 | return(1); |
1369 | 0 | if (nxt[0] != '-') |
1370 | 0 | return(0); |
1371 | | |
1372 | 0 | nxt++; |
1373 | 0 | cur = nxt; |
1374 | | /* now we can have region or variant */ |
1375 | 0 | if ((nxt[0] >= '0') && (nxt[0] <= '9')) |
1376 | 0 | goto region_m49; |
1377 | | |
1378 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1379 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1380 | 0 | nxt++; |
1381 | |
|
1382 | 0 | if ((nxt - cur >= 5) && (nxt - cur <= 8)) |
1383 | 0 | goto variant; |
1384 | 0 | if (nxt - cur != 2) |
1385 | 0 | return(0); |
1386 | | /* we parsed a region */ |
1387 | 0 | region: |
1388 | 0 | if (nxt[0] == 0) |
1389 | 0 | return(1); |
1390 | 0 | if (nxt[0] != '-') |
1391 | 0 | return(0); |
1392 | | |
1393 | 0 | nxt++; |
1394 | 0 | cur = nxt; |
1395 | | /* now we can just have a variant */ |
1396 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1397 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1398 | 0 | nxt++; |
1399 | |
|
1400 | 0 | if ((nxt - cur < 5) || (nxt - cur > 8)) |
1401 | 0 | return(0); |
1402 | | |
1403 | | /* we parsed a variant */ |
1404 | 0 | variant: |
1405 | 0 | if (nxt[0] == 0) |
1406 | 0 | return(1); |
1407 | 0 | if (nxt[0] != '-') |
1408 | 0 | return(0); |
1409 | | /* extensions and private use subtags not checked */ |
1410 | 0 | return (1); |
1411 | | |
1412 | 0 | region_m49: |
1413 | 0 | if (((nxt[1] >= '0') && (nxt[1] <= '9')) && |
1414 | 0 | ((nxt[2] >= '0') && (nxt[2] <= '9'))) { |
1415 | 0 | nxt += 3; |
1416 | 0 | goto region; |
1417 | 0 | } |
1418 | 0 | return(0); |
1419 | 0 | } |
1420 | | |
1421 | | /************************************************************************ |
1422 | | * * |
1423 | | * Parser stacks related functions and macros * |
1424 | | * * |
1425 | | ************************************************************************/ |
1426 | | |
1427 | | static xmlChar * |
1428 | | xmlParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar **str); |
1429 | | |
1430 | | /** |
1431 | | * Create a new namespace database. |
1432 | | * |
1433 | | * @returns the new obejct. |
1434 | | */ |
1435 | | xmlParserNsData * |
1436 | 14.5k | xmlParserNsCreate(void) { |
1437 | 14.5k | xmlParserNsData *nsdb = xmlMalloc(sizeof(*nsdb)); |
1438 | | |
1439 | 14.5k | if (nsdb == NULL) |
1440 | 0 | return(NULL); |
1441 | 14.5k | memset(nsdb, 0, sizeof(*nsdb)); |
1442 | 14.5k | nsdb->defaultNsIndex = INT_MAX; |
1443 | | |
1444 | 14.5k | return(nsdb); |
1445 | 14.5k | } |
1446 | | |
1447 | | /** |
1448 | | * Free a namespace database. |
1449 | | * |
1450 | | * @param nsdb namespace database |
1451 | | */ |
1452 | | void |
1453 | 14.5k | xmlParserNsFree(xmlParserNsData *nsdb) { |
1454 | 14.5k | if (nsdb == NULL) |
1455 | 0 | return; |
1456 | | |
1457 | 14.5k | xmlFree(nsdb->extra); |
1458 | 14.5k | xmlFree(nsdb->hash); |
1459 | 14.5k | xmlFree(nsdb); |
1460 | 14.5k | } |
1461 | | |
1462 | | /** |
1463 | | * Reset a namespace database. |
1464 | | * |
1465 | | * @param nsdb namespace database |
1466 | | */ |
1467 | | static void |
1468 | 0 | xmlParserNsReset(xmlParserNsData *nsdb) { |
1469 | 0 | if (nsdb == NULL) |
1470 | 0 | return; |
1471 | | |
1472 | 0 | nsdb->hashElems = 0; |
1473 | 0 | nsdb->elementId = 0; |
1474 | 0 | nsdb->defaultNsIndex = INT_MAX; |
1475 | |
|
1476 | 0 | if (nsdb->hash) |
1477 | 0 | memset(nsdb->hash, 0, nsdb->hashSize * sizeof(nsdb->hash[0])); |
1478 | 0 | } |
1479 | | |
1480 | | /** |
1481 | | * Signal that a new element has started. |
1482 | | * |
1483 | | * @param nsdb namespace database |
1484 | | * @returns 0 on success, -1 if the element counter overflowed. |
1485 | | */ |
1486 | | static int |
1487 | 2.21M | xmlParserNsStartElement(xmlParserNsData *nsdb) { |
1488 | 2.21M | if (nsdb->elementId == UINT_MAX) |
1489 | 0 | return(-1); |
1490 | 2.21M | nsdb->elementId++; |
1491 | | |
1492 | 2.21M | return(0); |
1493 | 2.21M | } |
1494 | | |
1495 | | /** |
1496 | | * Lookup namespace with given prefix. If `bucketPtr` is non-NULL, it will |
1497 | | * be set to the matching bucket, or the first empty bucket if no match |
1498 | | * was found. |
1499 | | * |
1500 | | * @param ctxt parser context |
1501 | | * @param prefix namespace prefix |
1502 | | * @param bucketPtr optional bucket (return value) |
1503 | | * @returns the namespace index on success, INT_MAX if no namespace was |
1504 | | * found. |
1505 | | */ |
1506 | | static int |
1507 | | xmlParserNsLookup(xmlParserCtxtPtr ctxt, const xmlHashedString *prefix, |
1508 | 2.08M | xmlParserNsBucket **bucketPtr) { |
1509 | 2.08M | xmlParserNsBucket *bucket, *tombstone; |
1510 | 2.08M | unsigned index, hashValue; |
1511 | | |
1512 | 2.08M | if (prefix->name == NULL) |
1513 | 586k | return(ctxt->nsdb->defaultNsIndex); |
1514 | | |
1515 | 1.49M | if (ctxt->nsdb->hashSize == 0) |
1516 | 32.6k | return(INT_MAX); |
1517 | | |
1518 | 1.46M | hashValue = prefix->hashValue; |
1519 | 1.46M | index = hashValue & (ctxt->nsdb->hashSize - 1); |
1520 | 1.46M | bucket = &ctxt->nsdb->hash[index]; |
1521 | 1.46M | tombstone = NULL; |
1522 | | |
1523 | 2.20M | while (bucket->hashValue) { |
1524 | 1.97M | if (bucket->index == INT_MAX) { |
1525 | 66.5k | if (tombstone == NULL) |
1526 | 60.2k | tombstone = bucket; |
1527 | 1.90M | } else if (bucket->hashValue == hashValue) { |
1528 | 1.23M | if (ctxt->nsTab[bucket->index * 2] == prefix->name) { |
1529 | 1.23M | if (bucketPtr != NULL) |
1530 | 1.07M | *bucketPtr = bucket; |
1531 | 1.23M | return(bucket->index); |
1532 | 1.23M | } |
1533 | 1.23M | } |
1534 | | |
1535 | 737k | index++; |
1536 | 737k | bucket++; |
1537 | 737k | if (index == ctxt->nsdb->hashSize) { |
1538 | 203k | index = 0; |
1539 | 203k | bucket = ctxt->nsdb->hash; |
1540 | 203k | } |
1541 | 737k | } |
1542 | | |
1543 | 228k | if (bucketPtr != NULL) |
1544 | 54.7k | *bucketPtr = tombstone ? tombstone : bucket; |
1545 | 228k | return(INT_MAX); |
1546 | 1.46M | } |
1547 | | |
1548 | | /** |
1549 | | * Lookup namespace URI with given prefix. |
1550 | | * |
1551 | | * @param ctxt parser context |
1552 | | * @param prefix namespace prefix |
1553 | | * @returns the namespace URI on success, NULL if no namespace was found. |
1554 | | */ |
1555 | | static const xmlChar * |
1556 | 605k | xmlParserNsLookupUri(xmlParserCtxtPtr ctxt, const xmlHashedString *prefix) { |
1557 | 605k | const xmlChar *ret; |
1558 | 605k | int nsIndex; |
1559 | | |
1560 | 605k | if (prefix->name == ctxt->str_xml) |
1561 | 531 | return(ctxt->str_xml_ns); |
1562 | | |
1563 | | /* |
1564 | | * minNsIndex is used when building an entity tree. We must |
1565 | | * ignore namespaces declared outside the entity. |
1566 | | */ |
1567 | 605k | nsIndex = xmlParserNsLookup(ctxt, prefix, NULL); |
1568 | 605k | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) |
1569 | 204k | return(NULL); |
1570 | | |
1571 | 400k | ret = ctxt->nsTab[nsIndex * 2 + 1]; |
1572 | 400k | if (ret[0] == 0) |
1573 | 28.8k | ret = NULL; |
1574 | 400k | return(ret); |
1575 | 605k | } |
1576 | | |
1577 | | /** |
1578 | | * Lookup extra data for the given prefix. This returns data stored |
1579 | | * with xmlParserNsUdpateSax. |
1580 | | * |
1581 | | * @param ctxt parser context |
1582 | | * @param prefix namespace prefix |
1583 | | * @returns the data on success, NULL if no namespace was found. |
1584 | | */ |
1585 | | void * |
1586 | 28.2k | xmlParserNsLookupSax(xmlParserCtxt *ctxt, const xmlChar *prefix) { |
1587 | 28.2k | xmlHashedString hprefix; |
1588 | 28.2k | int nsIndex; |
1589 | | |
1590 | 28.2k | if (prefix == ctxt->str_xml) |
1591 | 6.73k | return(NULL); |
1592 | | |
1593 | 21.5k | hprefix.name = prefix; |
1594 | 21.5k | if (prefix != NULL) |
1595 | 13.6k | hprefix.hashValue = xmlDictComputeHash(ctxt->dict, prefix); |
1596 | 7.92k | else |
1597 | 7.92k | hprefix.hashValue = 0; |
1598 | 21.5k | nsIndex = xmlParserNsLookup(ctxt, &hprefix, NULL); |
1599 | 21.5k | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) |
1600 | 0 | return(NULL); |
1601 | | |
1602 | 21.5k | return(ctxt->nsdb->extra[nsIndex].saxData); |
1603 | 21.5k | } |
1604 | | |
1605 | | /** |
1606 | | * Sets or updates extra data for the given prefix. This value will be |
1607 | | * returned by xmlParserNsLookupSax as long as the namespace with the |
1608 | | * given prefix is in scope. |
1609 | | * |
1610 | | * @param ctxt parser context |
1611 | | * @param prefix namespace prefix |
1612 | | * @param saxData extra data for SAX handler |
1613 | | * @returns the data on success, NULL if no namespace was found. |
1614 | | */ |
1615 | | int |
1616 | | xmlParserNsUpdateSax(xmlParserCtxt *ctxt, const xmlChar *prefix, |
1617 | 78.5k | void *saxData) { |
1618 | 78.5k | xmlHashedString hprefix; |
1619 | 78.5k | int nsIndex; |
1620 | | |
1621 | 78.5k | if (prefix == ctxt->str_xml) |
1622 | 0 | return(-1); |
1623 | | |
1624 | 78.5k | hprefix.name = prefix; |
1625 | 78.5k | if (prefix != NULL) |
1626 | 77.0k | hprefix.hashValue = xmlDictComputeHash(ctxt->dict, prefix); |
1627 | 1.49k | else |
1628 | 1.49k | hprefix.hashValue = 0; |
1629 | 78.5k | nsIndex = xmlParserNsLookup(ctxt, &hprefix, NULL); |
1630 | 78.5k | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) |
1631 | 0 | return(-1); |
1632 | | |
1633 | 78.5k | ctxt->nsdb->extra[nsIndex].saxData = saxData; |
1634 | 78.5k | return(0); |
1635 | 78.5k | } |
1636 | | |
1637 | | /** |
1638 | | * Grows the namespace tables. |
1639 | | * |
1640 | | * @param ctxt parser context |
1641 | | * @returns 0 on success, -1 if a memory allocation failed. |
1642 | | */ |
1643 | | static int |
1644 | 13.2k | xmlParserNsGrow(xmlParserCtxtPtr ctxt) { |
1645 | 13.2k | const xmlChar **table; |
1646 | 13.2k | xmlParserNsExtra *extra; |
1647 | 13.2k | int newSize; |
1648 | | |
1649 | 13.2k | newSize = xmlGrowCapacity(ctxt->nsMax, |
1650 | 13.2k | sizeof(table[0]) + sizeof(extra[0]), |
1651 | 13.2k | 16, XML_MAX_ITEMS); |
1652 | 13.2k | if (newSize < 0) |
1653 | 0 | goto error; |
1654 | | |
1655 | 13.2k | table = xmlRealloc(ctxt->nsTab, 2 * newSize * sizeof(table[0])); |
1656 | 13.2k | if (table == NULL) |
1657 | 0 | goto error; |
1658 | 13.2k | ctxt->nsTab = table; |
1659 | | |
1660 | 13.2k | extra = xmlRealloc(ctxt->nsdb->extra, newSize * sizeof(extra[0])); |
1661 | 13.2k | if (extra == NULL) |
1662 | 0 | goto error; |
1663 | 13.2k | ctxt->nsdb->extra = extra; |
1664 | | |
1665 | 13.2k | ctxt->nsMax = newSize; |
1666 | 13.2k | return(0); |
1667 | | |
1668 | 0 | error: |
1669 | 0 | xmlErrMemory(ctxt); |
1670 | 0 | return(-1); |
1671 | 13.2k | } |
1672 | | |
1673 | | /** |
1674 | | * Push a new namespace on the table. |
1675 | | * |
1676 | | * @param ctxt parser context |
1677 | | * @param prefix prefix with hash value |
1678 | | * @param uri uri with hash value |
1679 | | * @param saxData extra data for SAX handler |
1680 | | * @param defAttr whether the namespace comes from a default attribute |
1681 | | * @returns 1 if the namespace was pushed, 0 if the namespace was ignored, |
1682 | | * -1 if a memory allocation failed. |
1683 | | */ |
1684 | | static int |
1685 | | xmlParserNsPush(xmlParserCtxtPtr ctxt, const xmlHashedString *prefix, |
1686 | 702k | const xmlHashedString *uri, void *saxData, int defAttr) { |
1687 | 702k | xmlParserNsBucket *bucket = NULL; |
1688 | 702k | xmlParserNsExtra *extra; |
1689 | 702k | const xmlChar **ns; |
1690 | 702k | unsigned hashValue, nsIndex, oldIndex; |
1691 | | |
1692 | 702k | if ((prefix != NULL) && (prefix->name == ctxt->str_xml)) |
1693 | 148 | return(0); |
1694 | | |
1695 | 701k | if ((ctxt->nsNr >= ctxt->nsMax) && (xmlParserNsGrow(ctxt) < 0)) { |
1696 | 0 | xmlErrMemory(ctxt); |
1697 | 0 | return(-1); |
1698 | 0 | } |
1699 | | |
1700 | | /* |
1701 | | * Default namespace and 'xml' namespace |
1702 | | */ |
1703 | 701k | if ((prefix == NULL) || (prefix->name == NULL)) { |
1704 | 136k | oldIndex = ctxt->nsdb->defaultNsIndex; |
1705 | | |
1706 | 136k | if (oldIndex != INT_MAX) { |
1707 | 121k | extra = &ctxt->nsdb->extra[oldIndex]; |
1708 | | |
1709 | 121k | if (extra->elementId == ctxt->nsdb->elementId) { |
1710 | 19.1k | if (defAttr == 0) |
1711 | 18.5k | xmlErrAttributeDup(ctxt, NULL, BAD_CAST "xmlns"); |
1712 | 19.1k | return(0); |
1713 | 19.1k | } |
1714 | | |
1715 | 102k | if ((ctxt->options & XML_PARSE_NSCLEAN) && |
1716 | 102k | (uri->name == ctxt->nsTab[oldIndex * 2 + 1])) |
1717 | 0 | return(0); |
1718 | 102k | } |
1719 | | |
1720 | 117k | ctxt->nsdb->defaultNsIndex = ctxt->nsNr; |
1721 | 117k | goto populate_entry; |
1722 | 136k | } |
1723 | | |
1724 | | /* |
1725 | | * Hash table lookup |
1726 | | */ |
1727 | 565k | oldIndex = xmlParserNsLookup(ctxt, prefix, &bucket); |
1728 | 565k | if (oldIndex != INT_MAX) { |
1729 | 509k | extra = &ctxt->nsdb->extra[oldIndex]; |
1730 | | |
1731 | | /* |
1732 | | * Check for duplicate definitions on the same element. |
1733 | | */ |
1734 | 509k | if (extra->elementId == ctxt->nsdb->elementId) { |
1735 | 852 | if (defAttr == 0) |
1736 | 672 | xmlErrAttributeDup(ctxt, BAD_CAST "xmlns", prefix->name); |
1737 | 852 | return(0); |
1738 | 852 | } |
1739 | | |
1740 | 508k | if ((ctxt->options & XML_PARSE_NSCLEAN) && |
1741 | 508k | (uri->name == ctxt->nsTab[bucket->index * 2 + 1])) |
1742 | 0 | return(0); |
1743 | | |
1744 | 508k | bucket->index = ctxt->nsNr; |
1745 | 508k | goto populate_entry; |
1746 | 508k | } |
1747 | | |
1748 | | /* |
1749 | | * Insert new bucket |
1750 | | */ |
1751 | | |
1752 | 55.9k | hashValue = prefix->hashValue; |
1753 | | |
1754 | | /* |
1755 | | * Grow hash table, 50% fill factor |
1756 | | */ |
1757 | 55.9k | if (ctxt->nsdb->hashElems + 1 > ctxt->nsdb->hashSize / 2) { |
1758 | 1.78k | xmlParserNsBucket *newHash; |
1759 | 1.78k | unsigned newSize, i, index; |
1760 | | |
1761 | 1.78k | if (ctxt->nsdb->hashSize > UINT_MAX / 2) { |
1762 | 0 | xmlErrMemory(ctxt); |
1763 | 0 | return(-1); |
1764 | 0 | } |
1765 | 1.78k | newSize = ctxt->nsdb->hashSize ? ctxt->nsdb->hashSize * 2 : 16; |
1766 | 1.78k | newHash = xmlMalloc(newSize * sizeof(newHash[0])); |
1767 | 1.78k | if (newHash == NULL) { |
1768 | 0 | xmlErrMemory(ctxt); |
1769 | 0 | return(-1); |
1770 | 0 | } |
1771 | 1.78k | memset(newHash, 0, newSize * sizeof(newHash[0])); |
1772 | | |
1773 | 170k | for (i = 0; i < ctxt->nsdb->hashSize; i++) { |
1774 | 168k | unsigned hv = ctxt->nsdb->hash[i].hashValue; |
1775 | 168k | unsigned newIndex; |
1776 | | |
1777 | 168k | if ((hv == 0) || (ctxt->nsdb->hash[i].index == INT_MAX)) |
1778 | 166k | continue; |
1779 | 2.24k | newIndex = hv & (newSize - 1); |
1780 | | |
1781 | 3.09k | while (newHash[newIndex].hashValue != 0) { |
1782 | 843 | newIndex++; |
1783 | 843 | if (newIndex == newSize) |
1784 | 199 | newIndex = 0; |
1785 | 843 | } |
1786 | | |
1787 | 2.24k | newHash[newIndex] = ctxt->nsdb->hash[i]; |
1788 | 2.24k | } |
1789 | | |
1790 | 1.78k | xmlFree(ctxt->nsdb->hash); |
1791 | 1.78k | ctxt->nsdb->hash = newHash; |
1792 | 1.78k | ctxt->nsdb->hashSize = newSize; |
1793 | | |
1794 | | /* |
1795 | | * Relookup |
1796 | | */ |
1797 | 1.78k | index = hashValue & (newSize - 1); |
1798 | | |
1799 | 2.00k | while (newHash[index].hashValue != 0) { |
1800 | 225 | index++; |
1801 | 225 | if (index == newSize) |
1802 | 26 | index = 0; |
1803 | 225 | } |
1804 | | |
1805 | 1.78k | bucket = &newHash[index]; |
1806 | 1.78k | } |
1807 | | |
1808 | 55.9k | bucket->hashValue = hashValue; |
1809 | 55.9k | bucket->index = ctxt->nsNr; |
1810 | 55.9k | ctxt->nsdb->hashElems++; |
1811 | 55.9k | oldIndex = INT_MAX; |
1812 | | |
1813 | 681k | populate_entry: |
1814 | 681k | nsIndex = ctxt->nsNr; |
1815 | | |
1816 | 681k | ns = &ctxt->nsTab[nsIndex * 2]; |
1817 | 681k | ns[0] = prefix ? prefix->name : NULL; |
1818 | 681k | ns[1] = uri->name; |
1819 | | |
1820 | 681k | extra = &ctxt->nsdb->extra[nsIndex]; |
1821 | 681k | extra->saxData = saxData; |
1822 | 681k | extra->prefixHashValue = prefix ? prefix->hashValue : 0; |
1823 | 681k | extra->uriHashValue = uri->hashValue; |
1824 | 681k | extra->elementId = ctxt->nsdb->elementId; |
1825 | 681k | extra->oldIndex = oldIndex; |
1826 | | |
1827 | 681k | ctxt->nsNr++; |
1828 | | |
1829 | 681k | return(1); |
1830 | 55.9k | } |
1831 | | |
1832 | | /** |
1833 | | * Pops the top `nr` namespaces and restores the hash table. |
1834 | | * |
1835 | | * @param ctxt an XML parser context |
1836 | | * @param nr the number to pop |
1837 | | * @returns the number of namespaces popped. |
1838 | | */ |
1839 | | static int |
1840 | | xmlParserNsPop(xmlParserCtxtPtr ctxt, int nr) |
1841 | 228k | { |
1842 | 228k | int i; |
1843 | | |
1844 | | /* assert(nr <= ctxt->nsNr); */ |
1845 | | |
1846 | 908k | for (i = ctxt->nsNr - 1; i >= ctxt->nsNr - nr; i--) { |
1847 | 679k | const xmlChar *prefix = ctxt->nsTab[i * 2]; |
1848 | 679k | xmlParserNsExtra *extra = &ctxt->nsdb->extra[i]; |
1849 | | |
1850 | 679k | if (prefix == NULL) { |
1851 | 116k | ctxt->nsdb->defaultNsIndex = extra->oldIndex; |
1852 | 563k | } else { |
1853 | 563k | xmlHashedString hprefix; |
1854 | 563k | xmlParserNsBucket *bucket = NULL; |
1855 | | |
1856 | 563k | hprefix.name = prefix; |
1857 | 563k | hprefix.hashValue = extra->prefixHashValue; |
1858 | 563k | xmlParserNsLookup(ctxt, &hprefix, &bucket); |
1859 | | /* assert(bucket && bucket->hashValue); */ |
1860 | 563k | bucket->index = extra->oldIndex; |
1861 | 563k | } |
1862 | 679k | } |
1863 | | |
1864 | 228k | ctxt->nsNr -= nr; |
1865 | 228k | return(nr); |
1866 | 228k | } |
1867 | | |
1868 | | static int |
1869 | 5.64k | xmlCtxtGrowAttrs(xmlParserCtxtPtr ctxt) { |
1870 | 5.64k | const xmlChar **atts; |
1871 | 5.64k | unsigned *attallocs; |
1872 | 5.64k | int newSize; |
1873 | | |
1874 | 5.64k | newSize = xmlGrowCapacity(ctxt->maxatts / 5, |
1875 | 5.64k | sizeof(atts[0]) * 5 + sizeof(attallocs[0]), |
1876 | 5.64k | 10, XML_MAX_ATTRS); |
1877 | 5.64k | if (newSize < 0) { |
1878 | 0 | xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, |
1879 | 0 | "Maximum number of attributes exceeded"); |
1880 | 0 | return(-1); |
1881 | 0 | } |
1882 | | |
1883 | 5.64k | atts = xmlRealloc(ctxt->atts, newSize * sizeof(atts[0]) * 5); |
1884 | 5.64k | if (atts == NULL) |
1885 | 0 | goto mem_error; |
1886 | 5.64k | ctxt->atts = atts; |
1887 | | |
1888 | 5.64k | attallocs = xmlRealloc(ctxt->attallocs, |
1889 | 5.64k | newSize * sizeof(attallocs[0])); |
1890 | 5.64k | if (attallocs == NULL) |
1891 | 0 | goto mem_error; |
1892 | 5.64k | ctxt->attallocs = attallocs; |
1893 | | |
1894 | 5.64k | ctxt->maxatts = newSize * 5; |
1895 | | |
1896 | 5.64k | return(0); |
1897 | | |
1898 | 0 | mem_error: |
1899 | 0 | xmlErrMemory(ctxt); |
1900 | 0 | return(-1); |
1901 | 5.64k | } |
1902 | | |
1903 | | /** |
1904 | | * Pushes a new parser input on top of the input stack |
1905 | | * |
1906 | | * @param ctxt an XML parser context |
1907 | | * @param value the parser input |
1908 | | * @returns -1 in case of error, the index in the stack otherwise |
1909 | | */ |
1910 | | int |
1911 | | xmlCtxtPushInput(xmlParserCtxt *ctxt, xmlParserInput *value) |
1912 | 52.9k | { |
1913 | 52.9k | char *directory = NULL; |
1914 | 52.9k | int maxDepth; |
1915 | | |
1916 | 52.9k | if ((ctxt == NULL) || (value == NULL)) |
1917 | 0 | return(-1); |
1918 | | |
1919 | 52.9k | maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
1920 | | |
1921 | 52.9k | if (ctxt->inputNr >= ctxt->inputMax) { |
1922 | 1.45k | xmlParserInputPtr *tmp; |
1923 | 1.45k | int newSize; |
1924 | | |
1925 | 1.45k | newSize = xmlGrowCapacity(ctxt->inputMax, sizeof(tmp[0]), |
1926 | 1.45k | 5, maxDepth); |
1927 | 1.45k | if (newSize < 0) { |
1928 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
1929 | 0 | "Maximum entity nesting depth exceeded"); |
1930 | 0 | xmlHaltParser(ctxt); |
1931 | 0 | return(-1); |
1932 | 0 | } |
1933 | 1.45k | tmp = xmlRealloc(ctxt->inputTab, newSize * sizeof(tmp[0])); |
1934 | 1.45k | if (tmp == NULL) { |
1935 | 0 | xmlErrMemory(ctxt); |
1936 | 0 | return(-1); |
1937 | 0 | } |
1938 | 1.45k | ctxt->inputTab = tmp; |
1939 | 1.45k | ctxt->inputMax = newSize; |
1940 | 1.45k | } |
1941 | | |
1942 | 52.9k | if ((ctxt->inputNr == 0) && (value->filename != NULL)) { |
1943 | 0 | directory = xmlParserGetDirectory(value->filename); |
1944 | 0 | if (directory == NULL) { |
1945 | 0 | xmlErrMemory(ctxt); |
1946 | 0 | return(-1); |
1947 | 0 | } |
1948 | 0 | } |
1949 | | |
1950 | 52.9k | if (ctxt->input_id >= INT_MAX) { |
1951 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, "Input ID overflow\n"); |
1952 | 0 | return(-1); |
1953 | 0 | } |
1954 | | |
1955 | 52.9k | ctxt->inputTab[ctxt->inputNr] = value; |
1956 | 52.9k | ctxt->input = value; |
1957 | | |
1958 | 52.9k | if (ctxt->inputNr == 0) { |
1959 | 14.5k | xmlFree(ctxt->directory); |
1960 | 14.5k | ctxt->directory = directory; |
1961 | 14.5k | } |
1962 | | |
1963 | | /* |
1964 | | * The input ID is unused internally, but there are entity |
1965 | | * loaders in downstream code that detect the main document |
1966 | | * by checking for "input_id == 1". |
1967 | | */ |
1968 | 52.9k | value->id = ctxt->input_id++; |
1969 | | |
1970 | 52.9k | return(ctxt->inputNr++); |
1971 | 52.9k | } |
1972 | | |
1973 | | /** |
1974 | | * Pops the top parser input from the input stack |
1975 | | * |
1976 | | * @param ctxt an XML parser context |
1977 | | * @returns the input just removed |
1978 | | */ |
1979 | | xmlParserInput * |
1980 | | xmlCtxtPopInput(xmlParserCtxt *ctxt) |
1981 | 82.0k | { |
1982 | 82.0k | xmlParserInputPtr ret; |
1983 | | |
1984 | 82.0k | if (ctxt == NULL) |
1985 | 0 | return(NULL); |
1986 | 82.0k | if (ctxt->inputNr <= 0) |
1987 | 29.0k | return (NULL); |
1988 | 52.9k | ctxt->inputNr--; |
1989 | 52.9k | if (ctxt->inputNr > 0) |
1990 | 38.4k | ctxt->input = ctxt->inputTab[ctxt->inputNr - 1]; |
1991 | 14.5k | else |
1992 | 14.5k | ctxt->input = NULL; |
1993 | 52.9k | ret = ctxt->inputTab[ctxt->inputNr]; |
1994 | 52.9k | ctxt->inputTab[ctxt->inputNr] = NULL; |
1995 | 52.9k | return (ret); |
1996 | 82.0k | } |
1997 | | |
1998 | | /** |
1999 | | * Pushes a new element node on top of the node stack |
2000 | | * |
2001 | | * @deprecated Internal function, do not use. |
2002 | | * |
2003 | | * @param ctxt an XML parser context |
2004 | | * @param value the element node |
2005 | | * @returns -1 in case of error, the index in the stack otherwise |
2006 | | */ |
2007 | | int |
2008 | | nodePush(xmlParserCtxt *ctxt, xmlNode *value) |
2009 | 63.1k | { |
2010 | 63.1k | if (ctxt == NULL) |
2011 | 0 | return(0); |
2012 | | |
2013 | 63.1k | if (ctxt->nodeNr >= ctxt->nodeMax) { |
2014 | 5.10k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 2048 : 256; |
2015 | 5.10k | xmlNodePtr *tmp; |
2016 | 5.10k | int newSize; |
2017 | | |
2018 | 5.10k | newSize = xmlGrowCapacity(ctxt->nodeMax, sizeof(tmp[0]), |
2019 | 5.10k | 10, maxDepth); |
2020 | 5.10k | if (newSize < 0) { |
2021 | 6 | xmlFatalErrMsgInt(ctxt, XML_ERR_RESOURCE_LIMIT, |
2022 | 6 | "Excessive depth in document: %d," |
2023 | 6 | " use XML_PARSE_HUGE option\n", |
2024 | 6 | ctxt->nodeNr); |
2025 | 6 | xmlHaltParser(ctxt); |
2026 | 6 | return(-1); |
2027 | 6 | } |
2028 | | |
2029 | 5.09k | tmp = xmlRealloc(ctxt->nodeTab, newSize * sizeof(tmp[0])); |
2030 | 5.09k | if (tmp == NULL) { |
2031 | 0 | xmlErrMemory(ctxt); |
2032 | 0 | return (-1); |
2033 | 0 | } |
2034 | 5.09k | ctxt->nodeTab = tmp; |
2035 | 5.09k | ctxt->nodeMax = newSize; |
2036 | 5.09k | } |
2037 | | |
2038 | 63.1k | ctxt->nodeTab[ctxt->nodeNr] = value; |
2039 | 63.1k | ctxt->node = value; |
2040 | 63.1k | return (ctxt->nodeNr++); |
2041 | 63.1k | } |
2042 | | |
2043 | | /** |
2044 | | * Pops the top element node from the node stack |
2045 | | * |
2046 | | * @deprecated Internal function, do not use. |
2047 | | * |
2048 | | * @param ctxt an XML parser context |
2049 | | * @returns the node just removed |
2050 | | */ |
2051 | | xmlNode * |
2052 | | nodePop(xmlParserCtxt *ctxt) |
2053 | 319k | { |
2054 | 319k | xmlNodePtr ret; |
2055 | | |
2056 | 319k | if (ctxt == NULL) return(NULL); |
2057 | 319k | if (ctxt->nodeNr <= 0) |
2058 | 259k | return (NULL); |
2059 | 60.4k | ctxt->nodeNr--; |
2060 | 60.4k | if (ctxt->nodeNr > 0) |
2061 | 58.8k | ctxt->node = ctxt->nodeTab[ctxt->nodeNr - 1]; |
2062 | 1.60k | else |
2063 | 1.60k | ctxt->node = NULL; |
2064 | 60.4k | ret = ctxt->nodeTab[ctxt->nodeNr]; |
2065 | 60.4k | ctxt->nodeTab[ctxt->nodeNr] = NULL; |
2066 | 60.4k | return (ret); |
2067 | 319k | } |
2068 | | |
2069 | | /** |
2070 | | * Pushes a new element name/prefix/URL on top of the name stack |
2071 | | * |
2072 | | * @param ctxt an XML parser context |
2073 | | * @param value the element name |
2074 | | * @param prefix the element prefix |
2075 | | * @param URI the element namespace name |
2076 | | * @param line the current line number for error messages |
2077 | | * @param nsNr the number of namespaces pushed on the namespace table |
2078 | | * @returns -1 in case of error, the index in the stack otherwise |
2079 | | */ |
2080 | | static int |
2081 | | nameNsPush(xmlParserCtxtPtr ctxt, const xmlChar * value, |
2082 | | const xmlChar *prefix, const xmlChar *URI, int line, int nsNr) |
2083 | 612k | { |
2084 | 612k | xmlStartTag *tag; |
2085 | | |
2086 | 612k | if (ctxt->nameNr >= ctxt->nameMax) { |
2087 | 18.3k | const xmlChar **tmp; |
2088 | 18.3k | xmlStartTag *tmp2; |
2089 | 18.3k | int newSize; |
2090 | | |
2091 | 18.3k | newSize = xmlGrowCapacity(ctxt->nameMax, |
2092 | 18.3k | sizeof(tmp[0]) + sizeof(tmp2[0]), |
2093 | 18.3k | 10, XML_MAX_ITEMS); |
2094 | 18.3k | if (newSize < 0) |
2095 | 0 | goto mem_error; |
2096 | | |
2097 | 18.3k | tmp = xmlRealloc(ctxt->nameTab, newSize * sizeof(tmp[0])); |
2098 | 18.3k | if (tmp == NULL) |
2099 | 0 | goto mem_error; |
2100 | 18.3k | ctxt->nameTab = tmp; |
2101 | | |
2102 | 18.3k | tmp2 = xmlRealloc(ctxt->pushTab, newSize * sizeof(tmp2[0])); |
2103 | 18.3k | if (tmp2 == NULL) |
2104 | 0 | goto mem_error; |
2105 | 18.3k | ctxt->pushTab = tmp2; |
2106 | | |
2107 | 18.3k | ctxt->nameMax = newSize; |
2108 | 594k | } else if (ctxt->pushTab == NULL) { |
2109 | 9.07k | ctxt->pushTab = xmlMalloc(ctxt->nameMax * sizeof(ctxt->pushTab[0])); |
2110 | 9.07k | if (ctxt->pushTab == NULL) |
2111 | 0 | goto mem_error; |
2112 | 9.07k | } |
2113 | 612k | ctxt->nameTab[ctxt->nameNr] = value; |
2114 | 612k | ctxt->name = value; |
2115 | 612k | tag = &ctxt->pushTab[ctxt->nameNr]; |
2116 | 612k | tag->prefix = prefix; |
2117 | 612k | tag->URI = URI; |
2118 | 612k | tag->line = line; |
2119 | 612k | tag->nsNr = nsNr; |
2120 | 612k | return (ctxt->nameNr++); |
2121 | 0 | mem_error: |
2122 | 0 | xmlErrMemory(ctxt); |
2123 | 0 | return (-1); |
2124 | 612k | } |
2125 | | #ifdef LIBXML_PUSH_ENABLED |
2126 | | /** |
2127 | | * Pops the top element/prefix/URI name from the name stack |
2128 | | * |
2129 | | * @param ctxt an XML parser context |
2130 | | * @returns the name just removed |
2131 | | */ |
2132 | | static const xmlChar * |
2133 | | nameNsPop(xmlParserCtxtPtr ctxt) |
2134 | 0 | { |
2135 | 0 | const xmlChar *ret; |
2136 | |
|
2137 | 0 | if (ctxt->nameNr <= 0) |
2138 | 0 | return (NULL); |
2139 | 0 | ctxt->nameNr--; |
2140 | 0 | if (ctxt->nameNr > 0) |
2141 | 0 | ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; |
2142 | 0 | else |
2143 | 0 | ctxt->name = NULL; |
2144 | 0 | ret = ctxt->nameTab[ctxt->nameNr]; |
2145 | 0 | ctxt->nameTab[ctxt->nameNr] = NULL; |
2146 | 0 | return (ret); |
2147 | 0 | } |
2148 | | #endif /* LIBXML_PUSH_ENABLED */ |
2149 | | |
2150 | | /** |
2151 | | * Pops the top element name from the name stack |
2152 | | * |
2153 | | * @deprecated Internal function, do not use. |
2154 | | * |
2155 | | * @param ctxt an XML parser context |
2156 | | * @returns the name just removed |
2157 | | */ |
2158 | | static const xmlChar * |
2159 | | namePop(xmlParserCtxtPtr ctxt) |
2160 | 607k | { |
2161 | 607k | const xmlChar *ret; |
2162 | | |
2163 | 607k | if ((ctxt == NULL) || (ctxt->nameNr <= 0)) |
2164 | 0 | return (NULL); |
2165 | 607k | ctxt->nameNr--; |
2166 | 607k | if (ctxt->nameNr > 0) |
2167 | 603k | ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; |
2168 | 3.61k | else |
2169 | 3.61k | ctxt->name = NULL; |
2170 | 607k | ret = ctxt->nameTab[ctxt->nameNr]; |
2171 | 607k | ctxt->nameTab[ctxt->nameNr] = NULL; |
2172 | 607k | return (ret); |
2173 | 607k | } |
2174 | | |
2175 | 2.22M | static int spacePush(xmlParserCtxtPtr ctxt, int val) { |
2176 | 2.22M | if (ctxt->spaceNr >= ctxt->spaceMax) { |
2177 | 26.2k | int *tmp; |
2178 | 26.2k | int newSize; |
2179 | | |
2180 | 26.2k | newSize = xmlGrowCapacity(ctxt->spaceMax, sizeof(tmp[0]), |
2181 | 26.2k | 10, XML_MAX_ITEMS); |
2182 | 26.2k | if (newSize < 0) { |
2183 | 0 | xmlErrMemory(ctxt); |
2184 | 0 | return(-1); |
2185 | 0 | } |
2186 | | |
2187 | 26.2k | tmp = xmlRealloc(ctxt->spaceTab, newSize * sizeof(tmp[0])); |
2188 | 26.2k | if (tmp == NULL) { |
2189 | 0 | xmlErrMemory(ctxt); |
2190 | 0 | return(-1); |
2191 | 0 | } |
2192 | 26.2k | ctxt->spaceTab = tmp; |
2193 | | |
2194 | 26.2k | ctxt->spaceMax = newSize; |
2195 | 26.2k | } |
2196 | 2.22M | ctxt->spaceTab[ctxt->spaceNr] = val; |
2197 | 2.22M | ctxt->space = &ctxt->spaceTab[ctxt->spaceNr]; |
2198 | 2.22M | return(ctxt->spaceNr++); |
2199 | 2.22M | } |
2200 | | |
2201 | 2.21M | static int spacePop(xmlParserCtxtPtr ctxt) { |
2202 | 2.21M | int ret; |
2203 | 2.21M | if (ctxt->spaceNr <= 0) return(0); |
2204 | 2.21M | ctxt->spaceNr--; |
2205 | 2.21M | if (ctxt->spaceNr > 0) |
2206 | 2.21M | ctxt->space = &ctxt->spaceTab[ctxt->spaceNr - 1]; |
2207 | 0 | else |
2208 | 0 | ctxt->space = &ctxt->spaceTab[0]; |
2209 | 2.21M | ret = ctxt->spaceTab[ctxt->spaceNr]; |
2210 | 2.21M | ctxt->spaceTab[ctxt->spaceNr] = -1; |
2211 | 2.21M | return(ret); |
2212 | 2.21M | } |
2213 | | |
2214 | | /* |
2215 | | * Macros for accessing the content. Those should be used only by the parser, |
2216 | | * and not exported. |
2217 | | * |
2218 | | * Dirty macros, i.e. one often need to make assumption on the context to |
2219 | | * use them |
2220 | | * |
2221 | | * CUR_PTR return the current pointer to the xmlChar to be parsed. |
2222 | | * To be used with extreme caution since operations consuming |
2223 | | * characters may move the input buffer to a different location ! |
2224 | | * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled |
2225 | | * This should be used internally by the parser |
2226 | | * only to compare to ASCII values otherwise it would break when |
2227 | | * running with UTF-8 encoding. |
2228 | | * RAW same as CUR but in the input buffer, bypass any token |
2229 | | * extraction that may have been done |
2230 | | * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only |
2231 | | * to compare on ASCII based substring. |
2232 | | * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined |
2233 | | * strings without newlines within the parser. |
2234 | | * NEXT1(l) Skip 1 xmlChar, and must also be used only to skip 1 non-newline ASCII |
2235 | | * defined char within the parser. |
2236 | | * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding |
2237 | | * |
2238 | | * NEXT Skip to the next character, this does the proper decoding |
2239 | | * in UTF-8 mode. It also pop-up unfinished entities on the fly. |
2240 | | * NEXTL(l) Skip the current unicode character of l xmlChars long. |
2241 | | * COPY_BUF copy the current unicode char to the target buffer, increment |
2242 | | * the index |
2243 | | * GROW, SHRINK handling of input buffers |
2244 | | */ |
2245 | | |
2246 | 11.9M | #define RAW (*ctxt->input->cur) |
2247 | 171M | #define CUR (*ctxt->input->cur) |
2248 | 6.28M | #define NXT(val) ctxt->input->cur[(val)] |
2249 | 363M | #define CUR_PTR ctxt->input->cur |
2250 | 5.70M | #define BASE_PTR ctxt->input->base |
2251 | | |
2252 | | #define CMP4( s, c1, c2, c3, c4 ) \ |
2253 | 12.6M | ( ((unsigned char *) s)[ 0 ] == c1 && ((unsigned char *) s)[ 1 ] == c2 && \ |
2254 | 6.31M | ((unsigned char *) s)[ 2 ] == c3 && ((unsigned char *) s)[ 3 ] == c4 ) |
2255 | | #define CMP5( s, c1, c2, c3, c4, c5 ) \ |
2256 | 12.5M | ( CMP4( s, c1, c2, c3, c4 ) && ((unsigned char *) s)[ 4 ] == c5 ) |
2257 | | #define CMP6( s, c1, c2, c3, c4, c5, c6 ) \ |
2258 | 12.2M | ( CMP5( s, c1, c2, c3, c4, c5 ) && ((unsigned char *) s)[ 5 ] == c6 ) |
2259 | | #define CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) \ |
2260 | 12.0M | ( CMP6( s, c1, c2, c3, c4, c5, c6 ) && ((unsigned char *) s)[ 6 ] == c7 ) |
2261 | | #define CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) \ |
2262 | 11.9M | ( CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) && ((unsigned char *) s)[ 7 ] == c8 ) |
2263 | | #define CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) \ |
2264 | 5.96M | ( CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) && \ |
2265 | 5.96M | ((unsigned char *) s)[ 8 ] == c9 ) |
2266 | | #define CMP10( s, c1, c2, c3, c4, c5, c6, c7, c8, c9, c10 ) \ |
2267 | 2.51k | ( CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) && \ |
2268 | 2.51k | ((unsigned char *) s)[ 9 ] == c10 ) |
2269 | | |
2270 | 605k | #define SKIP(val) do { \ |
2271 | 605k | ctxt->input->cur += (val),ctxt->input->col+=(val); \ |
2272 | 605k | if (*ctxt->input->cur == 0) \ |
2273 | 605k | xmlParserGrow(ctxt); \ |
2274 | 605k | } while (0) |
2275 | | |
2276 | | #define SKIPL(val) do { \ |
2277 | | int skipl; \ |
2278 | | for(skipl=0; skipl<val; skipl++) { \ |
2279 | | if (*(ctxt->input->cur) == '\n') { \ |
2280 | | ctxt->input->line++; ctxt->input->col = 1; \ |
2281 | | } else ctxt->input->col++; \ |
2282 | | ctxt->input->cur++; \ |
2283 | | } \ |
2284 | | if (*ctxt->input->cur == 0) \ |
2285 | | xmlParserGrow(ctxt); \ |
2286 | | } while (0) |
2287 | | |
2288 | | #define SHRINK \ |
2289 | 9.73M | if (!PARSER_PROGRESSIVE(ctxt)) \ |
2290 | 9.73M | xmlParserShrink(ctxt); |
2291 | | |
2292 | | #define GROW \ |
2293 | 21.5M | if ((!PARSER_PROGRESSIVE(ctxt)) && \ |
2294 | 21.5M | (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \ |
2295 | 1.81M | xmlParserGrow(ctxt); |
2296 | | |
2297 | 3.33M | #define SKIP_BLANKS xmlSkipBlankChars(ctxt) |
2298 | | |
2299 | 271k | #define SKIP_BLANKS_PE xmlSkipBlankCharsPE(ctxt) |
2300 | | |
2301 | 12.3M | #define NEXT xmlNextChar(ctxt) |
2302 | | |
2303 | 2.54M | #define NEXT1 { \ |
2304 | 2.54M | ctxt->input->col++; \ |
2305 | 2.54M | ctxt->input->cur++; \ |
2306 | 2.54M | if (*ctxt->input->cur == 0) \ |
2307 | 2.54M | xmlParserGrow(ctxt); \ |
2308 | 2.54M | } |
2309 | | |
2310 | 218M | #define NEXTL(l) do { \ |
2311 | 218M | if (*(ctxt->input->cur) == '\n') { \ |
2312 | 14.5M | ctxt->input->line++; ctxt->input->col = 1; \ |
2313 | 204M | } else ctxt->input->col++; \ |
2314 | 218M | ctxt->input->cur += l; \ |
2315 | 218M | } while (0) |
2316 | | |
2317 | | #define COPY_BUF(b, i, v) \ |
2318 | 48.1M | if (v < 0x80) b[i++] = v; \ |
2319 | 48.1M | else i += xmlCopyCharMultiByte(&b[i],v) |
2320 | | |
2321 | | static int |
2322 | 44.4M | xmlCurrentCharRecover(xmlParserCtxtPtr ctxt, int *len) { |
2323 | 44.4M | int c = xmlCurrentChar(ctxt, len); |
2324 | | |
2325 | 44.4M | if (c == XML_INVALID_CHAR) |
2326 | 9.49M | c = 0xFFFD; /* replacement character */ |
2327 | | |
2328 | 44.4M | return(c); |
2329 | 44.4M | } |
2330 | | |
2331 | | /** |
2332 | | * Skip whitespace in the input stream. |
2333 | | * |
2334 | | * @deprecated Internal function, do not use. |
2335 | | * |
2336 | | * @param ctxt the XML parser context |
2337 | | * @returns the number of space chars skipped |
2338 | | */ |
2339 | | int |
2340 | 3.45M | xmlSkipBlankChars(xmlParserCtxt *ctxt) { |
2341 | 3.45M | const xmlChar *cur; |
2342 | 3.45M | int res = 0; |
2343 | | |
2344 | 3.45M | cur = ctxt->input->cur; |
2345 | 3.45M | while (IS_BLANK_CH(*cur)) { |
2346 | 2.32M | if (*cur == '\n') { |
2347 | 1.41M | ctxt->input->line++; ctxt->input->col = 1; |
2348 | 1.41M | } else { |
2349 | 907k | ctxt->input->col++; |
2350 | 907k | } |
2351 | 2.32M | cur++; |
2352 | 2.32M | if (res < INT_MAX) |
2353 | 2.32M | res++; |
2354 | 2.32M | if (*cur == 0) { |
2355 | 8.57k | ctxt->input->cur = cur; |
2356 | 8.57k | xmlParserGrow(ctxt); |
2357 | 8.57k | cur = ctxt->input->cur; |
2358 | 8.57k | } |
2359 | 2.32M | } |
2360 | 3.45M | ctxt->input->cur = cur; |
2361 | | |
2362 | 3.45M | if (res > 4) |
2363 | 8.97k | GROW; |
2364 | | |
2365 | 3.45M | return(res); |
2366 | 3.45M | } |
2367 | | |
2368 | | static void |
2369 | 31.3k | xmlPopPE(xmlParserCtxtPtr ctxt) { |
2370 | 31.3k | unsigned long consumed; |
2371 | 31.3k | xmlEntityPtr ent; |
2372 | | |
2373 | 31.3k | ent = ctxt->input->entity; |
2374 | | |
2375 | 31.3k | ent->flags &= ~XML_ENT_EXPANDING; |
2376 | | |
2377 | 31.3k | if ((ent->flags & XML_ENT_CHECKED) == 0) { |
2378 | 948 | int result; |
2379 | | |
2380 | | /* |
2381 | | * Read the rest of the stream in case of errors. We want |
2382 | | * to account for the whole entity size. |
2383 | | */ |
2384 | 948 | do { |
2385 | 948 | ctxt->input->cur = ctxt->input->end; |
2386 | 948 | xmlParserShrink(ctxt); |
2387 | 948 | result = xmlParserGrow(ctxt); |
2388 | 948 | } while (result > 0); |
2389 | | |
2390 | 948 | consumed = ctxt->input->consumed; |
2391 | 948 | xmlSaturatedAddSizeT(&consumed, |
2392 | 948 | ctxt->input->end - ctxt->input->base); |
2393 | | |
2394 | 948 | xmlSaturatedAdd(&ent->expandedSize, consumed); |
2395 | | |
2396 | | /* |
2397 | | * Add to sizeentities when parsing an external entity |
2398 | | * for the first time. |
2399 | | */ |
2400 | 948 | if (ent->etype == XML_EXTERNAL_PARAMETER_ENTITY) { |
2401 | 0 | xmlSaturatedAdd(&ctxt->sizeentities, consumed); |
2402 | 0 | } |
2403 | | |
2404 | 948 | ent->flags |= XML_ENT_CHECKED; |
2405 | 948 | } |
2406 | | |
2407 | 31.3k | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
2408 | | |
2409 | 31.3k | xmlParserEntityCheck(ctxt, ent->expandedSize); |
2410 | | |
2411 | 31.3k | GROW; |
2412 | 31.3k | } |
2413 | | |
2414 | | /** |
2415 | | * Skip whitespace in the input stream, also handling parameter |
2416 | | * entities. |
2417 | | * |
2418 | | * @param ctxt the XML parser context |
2419 | | * @returns the number of space chars skipped |
2420 | | */ |
2421 | | static int |
2422 | 271k | xmlSkipBlankCharsPE(xmlParserCtxtPtr ctxt) { |
2423 | 271k | int res = 0; |
2424 | 271k | int inParam; |
2425 | 271k | int expandParam; |
2426 | | |
2427 | 271k | inParam = PARSER_IN_PE(ctxt); |
2428 | 271k | expandParam = PARSER_EXTERNAL(ctxt); |
2429 | | |
2430 | 271k | if (!inParam && !expandParam) |
2431 | 120k | return(xmlSkipBlankChars(ctxt)); |
2432 | | |
2433 | | /* |
2434 | | * It's Okay to use CUR/NEXT here since all the blanks are on |
2435 | | * the ASCII range. |
2436 | | */ |
2437 | 7.37M | while (PARSER_STOPPED(ctxt) == 0) { |
2438 | 7.37M | if (IS_BLANK_CH(CUR)) { /* CHECKED tstblanks.xml */ |
2439 | 7.22M | NEXT; |
2440 | 7.22M | } else if (CUR == '%') { |
2441 | 458 | if ((expandParam == 0) || |
2442 | 458 | (IS_BLANK_CH(NXT(1))) || (NXT(1) == 0)) |
2443 | 458 | break; |
2444 | | |
2445 | | /* |
2446 | | * Expand parameter entity. We continue to consume |
2447 | | * whitespace at the start of the entity and possible |
2448 | | * even consume the whole entity and pop it. We might |
2449 | | * even pop multiple PEs in this loop. |
2450 | | */ |
2451 | 0 | xmlParsePERefInternal(ctxt, 0); |
2452 | |
|
2453 | 0 | inParam = PARSER_IN_PE(ctxt); |
2454 | 0 | expandParam = PARSER_EXTERNAL(ctxt); |
2455 | 150k | } else if (CUR == 0) { |
2456 | 9.48k | if (inParam == 0) |
2457 | 0 | break; |
2458 | | |
2459 | | /* |
2460 | | * Don't pop parameter entities that start a markup |
2461 | | * declaration to detect Well-formedness constraint: |
2462 | | * PE Between Declarations. |
2463 | | */ |
2464 | 9.48k | if (ctxt->input->flags & XML_INPUT_MARKUP_DECL) |
2465 | 9.48k | break; |
2466 | | |
2467 | 0 | xmlPopPE(ctxt); |
2468 | |
|
2469 | 0 | inParam = PARSER_IN_PE(ctxt); |
2470 | 0 | expandParam = PARSER_EXTERNAL(ctxt); |
2471 | 141k | } else { |
2472 | 141k | break; |
2473 | 141k | } |
2474 | | |
2475 | | /* |
2476 | | * Also increase the counter when entering or exiting a PERef. |
2477 | | * The spec says: "When a parameter-entity reference is recognized |
2478 | | * in the DTD and included, its replacement text MUST be enlarged |
2479 | | * by the attachment of one leading and one following space (#x20) |
2480 | | * character." |
2481 | | */ |
2482 | 7.22M | if (res < INT_MAX) |
2483 | 7.22M | res++; |
2484 | 7.22M | } |
2485 | | |
2486 | 151k | return(res); |
2487 | 271k | } |
2488 | | |
2489 | | /************************************************************************ |
2490 | | * * |
2491 | | * Commodity functions to handle entities * |
2492 | | * * |
2493 | | ************************************************************************/ |
2494 | | |
2495 | | /** |
2496 | | * @deprecated Internal function, don't use. |
2497 | | * |
2498 | | * @param ctxt an XML parser context |
2499 | | * @returns the current xmlChar in the parser context |
2500 | | */ |
2501 | | xmlChar |
2502 | 0 | xmlPopInput(xmlParserCtxt *ctxt) { |
2503 | 0 | xmlParserInputPtr input; |
2504 | |
|
2505 | 0 | if ((ctxt == NULL) || (ctxt->inputNr <= 1)) return(0); |
2506 | 0 | input = xmlCtxtPopInput(ctxt); |
2507 | 0 | xmlFreeInputStream(input); |
2508 | 0 | if (*ctxt->input->cur == 0) |
2509 | 0 | xmlParserGrow(ctxt); |
2510 | 0 | return(CUR); |
2511 | 0 | } |
2512 | | |
2513 | | /** |
2514 | | * Push an input stream onto the stack. |
2515 | | * |
2516 | | * @deprecated Internal function, don't use. |
2517 | | * |
2518 | | * @param ctxt an XML parser context |
2519 | | * @param input an XML parser input fragment (entity, XML fragment ...). |
2520 | | * @returns -1 in case of error or the index in the input stack |
2521 | | */ |
2522 | | int |
2523 | 0 | xmlPushInput(xmlParserCtxt *ctxt, xmlParserInput *input) { |
2524 | 0 | int ret; |
2525 | |
|
2526 | 0 | if ((ctxt == NULL) || (input == NULL)) |
2527 | 0 | return(-1); |
2528 | | |
2529 | 0 | ret = xmlCtxtPushInput(ctxt, input); |
2530 | 0 | if (ret >= 0) |
2531 | 0 | GROW; |
2532 | 0 | return(ret); |
2533 | 0 | } |
2534 | | |
2535 | | /** |
2536 | | * Parse a numeric character reference. Always consumes '&'. |
2537 | | * |
2538 | | * @deprecated Internal function, don't use. |
2539 | | * |
2540 | | * [66] CharRef ::= '&#' [0-9]+ ';' | |
2541 | | * '&#x' [0-9a-fA-F]+ ';' |
2542 | | * |
2543 | | * [ WFC: Legal Character ] |
2544 | | * Characters referred to using character references must match the |
2545 | | * production for Char. |
2546 | | * |
2547 | | * @param ctxt an XML parser context |
2548 | | * @returns the value parsed (as an int), 0 in case of error |
2549 | | */ |
2550 | | int |
2551 | 100k | xmlParseCharRef(xmlParserCtxt *ctxt) { |
2552 | 100k | int val = 0; |
2553 | 100k | int count = 0; |
2554 | | |
2555 | | /* |
2556 | | * Using RAW/CUR/NEXT is okay since we are working on ASCII range here |
2557 | | */ |
2558 | 100k | if ((RAW == '&') && (NXT(1) == '#') && |
2559 | 100k | (NXT(2) == 'x')) { |
2560 | 44.1k | SKIP(3); |
2561 | 44.1k | GROW; |
2562 | 170k | while ((RAW != ';') && (PARSER_STOPPED(ctxt) == 0)) { |
2563 | 131k | if (count++ > 20) { |
2564 | 461 | count = 0; |
2565 | 461 | GROW; |
2566 | 461 | } |
2567 | 131k | if ((RAW >= '0') && (RAW <= '9')) |
2568 | 8.80k | val = val * 16 + (CUR - '0'); |
2569 | 122k | else if ((RAW >= 'a') && (RAW <= 'f') && (count < 20)) |
2570 | 18.7k | val = val * 16 + (CUR - 'a') + 10; |
2571 | 103k | else if ((RAW >= 'A') && (RAW <= 'F') && (count < 20)) |
2572 | 98.5k | val = val * 16 + (CUR - 'A') + 10; |
2573 | 4.86k | else { |
2574 | 4.86k | xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); |
2575 | 4.86k | val = 0; |
2576 | 4.86k | break; |
2577 | 4.86k | } |
2578 | 126k | if (val > 0x110000) |
2579 | 9.28k | val = 0x110000; |
2580 | | |
2581 | 126k | NEXT; |
2582 | 126k | count++; |
2583 | 126k | } |
2584 | 44.1k | if (RAW == ';') { |
2585 | | /* on purpose to avoid reentrancy problems with NEXT and SKIP */ |
2586 | 39.2k | ctxt->input->col++; |
2587 | 39.2k | ctxt->input->cur++; |
2588 | 39.2k | } |
2589 | 56.2k | } else if ((RAW == '&') && (NXT(1) == '#')) { |
2590 | 56.2k | SKIP(2); |
2591 | 56.2k | GROW; |
2592 | 168k | while (RAW != ';') { /* loop blocked by count */ |
2593 | 126k | if (count++ > 20) { |
2594 | 1.54k | count = 0; |
2595 | 1.54k | GROW; |
2596 | 1.54k | } |
2597 | 126k | if ((RAW >= '0') && (RAW <= '9')) |
2598 | 111k | val = val * 10 + (CUR - '0'); |
2599 | 14.4k | else { |
2600 | 14.4k | xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); |
2601 | 14.4k | val = 0; |
2602 | 14.4k | break; |
2603 | 14.4k | } |
2604 | 111k | if (val > 0x110000) |
2605 | 13.1k | val = 0x110000; |
2606 | | |
2607 | 111k | NEXT; |
2608 | 111k | count++; |
2609 | 111k | } |
2610 | 56.2k | if (RAW == ';') { |
2611 | | /* on purpose to avoid reentrancy problems with NEXT and SKIP */ |
2612 | 41.8k | ctxt->input->col++; |
2613 | 41.8k | ctxt->input->cur++; |
2614 | 41.8k | } |
2615 | 56.2k | } else { |
2616 | 0 | if (RAW == '&') |
2617 | 0 | SKIP(1); |
2618 | 0 | xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); |
2619 | 0 | } |
2620 | | |
2621 | | /* |
2622 | | * [ WFC: Legal Character ] |
2623 | | * Characters referred to using character references must match the |
2624 | | * production for Char. |
2625 | | */ |
2626 | 100k | if (val >= 0x110000) { |
2627 | 351 | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2628 | 351 | "xmlParseCharRef: character reference out of bounds\n", |
2629 | 351 | val); |
2630 | 100k | } else if (IS_CHAR(val)) { |
2631 | 77.0k | return(val); |
2632 | 77.0k | } else { |
2633 | 22.9k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2634 | 22.9k | "xmlParseCharRef: invalid xmlChar value %d\n", |
2635 | 22.9k | val); |
2636 | 22.9k | } |
2637 | 23.3k | return(0); |
2638 | 100k | } |
2639 | | |
2640 | | /** |
2641 | | * parse Reference declarations, variant parsing from a string rather |
2642 | | * than an an input flow. |
2643 | | * |
2644 | | * [66] CharRef ::= '&#' [0-9]+ ';' | |
2645 | | * '&#x' [0-9a-fA-F]+ ';' |
2646 | | * |
2647 | | * [ WFC: Legal Character ] |
2648 | | * Characters referred to using character references must match the |
2649 | | * production for Char. |
2650 | | * |
2651 | | * @param ctxt an XML parser context |
2652 | | * @param str a pointer to an index in the string |
2653 | | * @returns the value parsed (as an int), 0 in case of error, str will be |
2654 | | * updated to the current value of the index |
2655 | | */ |
2656 | | static int |
2657 | 178k | xmlParseStringCharRef(xmlParserCtxtPtr ctxt, const xmlChar **str) { |
2658 | 178k | const xmlChar *ptr; |
2659 | 178k | xmlChar cur; |
2660 | 178k | int val = 0; |
2661 | | |
2662 | 178k | if ((str == NULL) || (*str == NULL)) return(0); |
2663 | 178k | ptr = *str; |
2664 | 178k | cur = *ptr; |
2665 | 178k | if ((cur == '&') && (ptr[1] == '#') && (ptr[2] == 'x')) { |
2666 | 44.5k | ptr += 3; |
2667 | 44.5k | cur = *ptr; |
2668 | 856k | while (cur != ';') { /* Non input consuming loop */ |
2669 | 813k | if ((cur >= '0') && (cur <= '9')) |
2670 | 738k | val = val * 16 + (cur - '0'); |
2671 | 75.1k | else if ((cur >= 'a') && (cur <= 'f')) |
2672 | 1.27k | val = val * 16 + (cur - 'a') + 10; |
2673 | 73.9k | else if ((cur >= 'A') && (cur <= 'F')) |
2674 | 72.7k | val = val * 16 + (cur - 'A') + 10; |
2675 | 1.17k | else { |
2676 | 1.17k | xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); |
2677 | 1.17k | val = 0; |
2678 | 1.17k | break; |
2679 | 1.17k | } |
2680 | 812k | if (val > 0x110000) |
2681 | 5.82k | val = 0x110000; |
2682 | | |
2683 | 812k | ptr++; |
2684 | 812k | cur = *ptr; |
2685 | 812k | } |
2686 | 44.5k | if (cur == ';') |
2687 | 43.3k | ptr++; |
2688 | 133k | } else if ((cur == '&') && (ptr[1] == '#')){ |
2689 | 133k | ptr += 2; |
2690 | 133k | cur = *ptr; |
2691 | 445k | while (cur != ';') { /* Non input consuming loops */ |
2692 | 313k | if ((cur >= '0') && (cur <= '9')) |
2693 | 312k | val = val * 10 + (cur - '0'); |
2694 | 1.86k | else { |
2695 | 1.86k | xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); |
2696 | 1.86k | val = 0; |
2697 | 1.86k | break; |
2698 | 1.86k | } |
2699 | 312k | if (val > 0x110000) |
2700 | 1.69k | val = 0x110000; |
2701 | | |
2702 | 312k | ptr++; |
2703 | 312k | cur = *ptr; |
2704 | 312k | } |
2705 | 133k | if (cur == ';') |
2706 | 131k | ptr++; |
2707 | 133k | } else { |
2708 | 0 | xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); |
2709 | 0 | return(0); |
2710 | 0 | } |
2711 | 178k | *str = ptr; |
2712 | | |
2713 | | /* |
2714 | | * [ WFC: Legal Character ] |
2715 | | * Characters referred to using character references must match the |
2716 | | * production for Char. |
2717 | | */ |
2718 | 178k | if (val >= 0x110000) { |
2719 | 248 | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2720 | 248 | "xmlParseStringCharRef: character reference out of bounds\n", |
2721 | 248 | val); |
2722 | 178k | } else if (IS_CHAR(val)) { |
2723 | 174k | return(val); |
2724 | 174k | } else { |
2725 | 3.96k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2726 | 3.96k | "xmlParseStringCharRef: invalid xmlChar value %d\n", |
2727 | 3.96k | val); |
2728 | 3.96k | } |
2729 | 4.21k | return(0); |
2730 | 178k | } |
2731 | | |
2732 | | /** |
2733 | | * [69] PEReference ::= '%' Name ';' |
2734 | | * |
2735 | | * @deprecated Internal function, do not use. |
2736 | | * |
2737 | | * [ WFC: No Recursion ] |
2738 | | * A parsed entity must not contain a recursive |
2739 | | * reference to itself, either directly or indirectly. |
2740 | | * |
2741 | | * [ WFC: Entity Declared ] |
2742 | | * In a document without any DTD, a document with only an internal DTD |
2743 | | * subset which contains no parameter entity references, or a document |
2744 | | * with "standalone='yes'", ... ... The declaration of a parameter |
2745 | | * entity must precede any reference to it... |
2746 | | * |
2747 | | * [ VC: Entity Declared ] |
2748 | | * In a document with an external subset or external parameter entities |
2749 | | * with "standalone='no'", ... ... The declaration of a parameter entity |
2750 | | * must precede any reference to it... |
2751 | | * |
2752 | | * [ WFC: In DTD ] |
2753 | | * Parameter-entity references may only appear in the DTD. |
2754 | | * NOTE: misleading but this is handled. |
2755 | | * |
2756 | | * A PEReference may have been detected in the current input stream |
2757 | | * the handling is done accordingly to |
2758 | | * http://www.w3.org/TR/REC-xml#entproc |
2759 | | * i.e. |
2760 | | * - Included in literal in entity values |
2761 | | * - Included as Parameter Entity reference within DTDs |
2762 | | * @param ctxt the parser context |
2763 | | */ |
2764 | | void |
2765 | 0 | xmlParserHandlePEReference(xmlParserCtxt *ctxt) { |
2766 | 0 | xmlParsePERefInternal(ctxt, 0); |
2767 | 0 | } |
2768 | | |
2769 | | /** |
2770 | | * @deprecated Internal function, don't use. |
2771 | | * |
2772 | | * @param ctxt the parser context |
2773 | | * @param str the input string |
2774 | | * @param len the string length |
2775 | | * @param what combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF |
2776 | | * @param end an end marker xmlChar, 0 if none |
2777 | | * @param end2 an end marker xmlChar, 0 if none |
2778 | | * @param end3 an end marker xmlChar, 0 if none |
2779 | | * @returns A newly allocated string with the substitution done. The caller |
2780 | | * must deallocate it ! |
2781 | | */ |
2782 | | xmlChar * |
2783 | | xmlStringLenDecodeEntities(xmlParserCtxt *ctxt, const xmlChar *str, int len, |
2784 | | int what ATTRIBUTE_UNUSED, |
2785 | 0 | xmlChar end, xmlChar end2, xmlChar end3) { |
2786 | 0 | if ((ctxt == NULL) || (str == NULL) || (len < 0)) |
2787 | 0 | return(NULL); |
2788 | | |
2789 | 0 | if ((str[len] != 0) || |
2790 | 0 | (end != 0) || (end2 != 0) || (end3 != 0)) |
2791 | 0 | return(NULL); |
2792 | | |
2793 | 0 | return(xmlExpandEntitiesInAttValue(ctxt, str, 0)); |
2794 | 0 | } |
2795 | | |
2796 | | /** |
2797 | | * @deprecated Internal function, don't use. |
2798 | | * |
2799 | | * @param ctxt the parser context |
2800 | | * @param str the input string |
2801 | | * @param what combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF |
2802 | | * @param end an end marker xmlChar, 0 if none |
2803 | | * @param end2 an end marker xmlChar, 0 if none |
2804 | | * @param end3 an end marker xmlChar, 0 if none |
2805 | | * @returns A newly allocated string with the substitution done. The caller |
2806 | | * must deallocate it ! |
2807 | | */ |
2808 | | xmlChar * |
2809 | | xmlStringDecodeEntities(xmlParserCtxt *ctxt, const xmlChar *str, |
2810 | | int what ATTRIBUTE_UNUSED, |
2811 | 0 | xmlChar end, xmlChar end2, xmlChar end3) { |
2812 | 0 | if ((ctxt == NULL) || (str == NULL)) |
2813 | 0 | return(NULL); |
2814 | | |
2815 | 0 | if ((end != 0) || (end2 != 0) || (end3 != 0)) |
2816 | 0 | return(NULL); |
2817 | | |
2818 | 0 | return(xmlExpandEntitiesInAttValue(ctxt, str, 0)); |
2819 | 0 | } |
2820 | | |
2821 | | /************************************************************************ |
2822 | | * * |
2823 | | * Commodity functions, cleanup needed ? * |
2824 | | * * |
2825 | | ************************************************************************/ |
2826 | | |
2827 | | /** |
2828 | | * Is this a sequence of blank chars that one can ignore ? |
2829 | | * |
2830 | | * @param ctxt an XML parser context |
2831 | | * @param str a xmlChar * |
2832 | | * @param len the size of `str` |
2833 | | * @param blank_chars we know the chars are blanks |
2834 | | * @returns 1 if ignorable 0 otherwise. |
2835 | | */ |
2836 | | |
2837 | | static int areBlanks(xmlParserCtxtPtr ctxt, const xmlChar *str, int len, |
2838 | 0 | int blank_chars) { |
2839 | 0 | int i; |
2840 | 0 | xmlNodePtr lastChild; |
2841 | | |
2842 | | /* |
2843 | | * Check for xml:space value. |
2844 | | */ |
2845 | 0 | if ((ctxt->space == NULL) || (*(ctxt->space) == 1) || |
2846 | 0 | (*(ctxt->space) == -2)) |
2847 | 0 | return(0); |
2848 | | |
2849 | | /* |
2850 | | * Check that the string is made of blanks |
2851 | | */ |
2852 | 0 | if (blank_chars == 0) { |
2853 | 0 | for (i = 0;i < len;i++) |
2854 | 0 | if (!(IS_BLANK_CH(str[i]))) return(0); |
2855 | 0 | } |
2856 | | |
2857 | | /* |
2858 | | * Look if the element is mixed content in the DTD if available |
2859 | | */ |
2860 | 0 | if (ctxt->node == NULL) return(0); |
2861 | 0 | if (ctxt->myDoc != NULL) { |
2862 | 0 | xmlElementPtr elemDecl = NULL; |
2863 | 0 | xmlDocPtr doc = ctxt->myDoc; |
2864 | 0 | const xmlChar *prefix = NULL; |
2865 | |
|
2866 | 0 | if (ctxt->node->ns) |
2867 | 0 | prefix = ctxt->node->ns->prefix; |
2868 | 0 | if (doc->intSubset != NULL) |
2869 | 0 | elemDecl = xmlHashLookup2(doc->intSubset->elements, ctxt->node->name, |
2870 | 0 | prefix); |
2871 | 0 | if ((elemDecl == NULL) && (doc->extSubset != NULL)) |
2872 | 0 | elemDecl = xmlHashLookup2(doc->extSubset->elements, ctxt->node->name, |
2873 | 0 | prefix); |
2874 | 0 | if (elemDecl != NULL) { |
2875 | 0 | if (elemDecl->etype == XML_ELEMENT_TYPE_ELEMENT) |
2876 | 0 | return(1); |
2877 | 0 | if ((elemDecl->etype == XML_ELEMENT_TYPE_ANY) || |
2878 | 0 | (elemDecl->etype == XML_ELEMENT_TYPE_MIXED)) |
2879 | 0 | return(0); |
2880 | 0 | } |
2881 | 0 | } |
2882 | | |
2883 | | /* |
2884 | | * Otherwise, heuristic :-\ |
2885 | | * |
2886 | | * When push parsing, we could be at the end of a chunk. |
2887 | | * This makes the look-ahead and consequently the NOBLANKS |
2888 | | * option unreliable. |
2889 | | */ |
2890 | 0 | if ((RAW != '<') && (RAW != 0xD)) return(0); |
2891 | 0 | if ((ctxt->node->children == NULL) && |
2892 | 0 | (RAW == '<') && (NXT(1) == '/')) return(0); |
2893 | | |
2894 | 0 | lastChild = xmlGetLastChild(ctxt->node); |
2895 | 0 | if (lastChild == NULL) { |
2896 | 0 | if ((ctxt->node->type != XML_ELEMENT_NODE) && |
2897 | 0 | (ctxt->node->content != NULL)) return(0); |
2898 | 0 | } else if (xmlNodeIsText(lastChild)) |
2899 | 0 | return(0); |
2900 | 0 | else if ((ctxt->node->children != NULL) && |
2901 | 0 | (xmlNodeIsText(ctxt->node->children))) |
2902 | 0 | return(0); |
2903 | 0 | return(1); |
2904 | 0 | } |
2905 | | |
2906 | | /************************************************************************ |
2907 | | * * |
2908 | | * Extra stuff for namespace support * |
2909 | | * Relates to http://www.w3.org/TR/WD-xml-names * |
2910 | | * * |
2911 | | ************************************************************************/ |
2912 | | |
2913 | | /** |
2914 | | * parse an UTF8 encoded XML qualified name string |
2915 | | * |
2916 | | * @deprecated Don't use. |
2917 | | * |
2918 | | * @param ctxt an XML parser context |
2919 | | * @param name an XML parser context |
2920 | | * @param prefixOut a xmlChar ** |
2921 | | * @returns the local part, and prefix is updated |
2922 | | * to get the Prefix if any. |
2923 | | */ |
2924 | | |
2925 | | xmlChar * |
2926 | 0 | xmlSplitQName(xmlParserCtxt *ctxt, const xmlChar *name, xmlChar **prefixOut) { |
2927 | 0 | xmlChar *ret; |
2928 | 0 | const xmlChar *localname; |
2929 | |
|
2930 | 0 | localname = xmlSplitQName4(name, prefixOut); |
2931 | 0 | if (localname == NULL) { |
2932 | 0 | xmlCtxtErrMemory(ctxt); |
2933 | 0 | return(NULL); |
2934 | 0 | } |
2935 | | |
2936 | 0 | ret = xmlStrdup(localname); |
2937 | 0 | if (ret == NULL) { |
2938 | 0 | xmlCtxtErrMemory(ctxt); |
2939 | 0 | xmlFree(*prefixOut); |
2940 | 0 | } |
2941 | |
|
2942 | 0 | return(ret); |
2943 | 0 | } |
2944 | | |
2945 | | /************************************************************************ |
2946 | | * * |
2947 | | * The parser itself * |
2948 | | * Relates to http://www.w3.org/TR/REC-xml * |
2949 | | * * |
2950 | | ************************************************************************/ |
2951 | | |
2952 | | /************************************************************************ |
2953 | | * * |
2954 | | * Routines to parse Name, NCName and NmToken * |
2955 | | * * |
2956 | | ************************************************************************/ |
2957 | | |
2958 | | /* |
2959 | | * The two following functions are related to the change of accepted |
2960 | | * characters for Name and NmToken in the Revision 5 of XML-1.0 |
2961 | | * They correspond to the modified production [4] and the new production [4a] |
2962 | | * changes in that revision. Also note that the macros used for the |
2963 | | * productions Letter, Digit, CombiningChar and Extender are not needed |
2964 | | * anymore. |
2965 | | * We still keep compatibility to pre-revision5 parsing semantic if the |
2966 | | * new XML_PARSE_OLD10 option is given to the parser. |
2967 | | */ |
2968 | | |
2969 | | static int |
2970 | 2.28M | xmlIsNameStartCharNew(int c) { |
2971 | | /* |
2972 | | * Use the new checks of production [4] [4a] amd [5] of the |
2973 | | * Update 5 of XML-1.0 |
2974 | | */ |
2975 | 2.28M | if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ |
2976 | 2.28M | (((c >= 'a') && (c <= 'z')) || |
2977 | 2.28M | ((c >= 'A') && (c <= 'Z')) || |
2978 | 2.28M | (c == '_') || (c == ':') || |
2979 | 2.28M | ((c >= 0xC0) && (c <= 0xD6)) || |
2980 | 2.28M | ((c >= 0xD8) && (c <= 0xF6)) || |
2981 | 2.28M | ((c >= 0xF8) && (c <= 0x2FF)) || |
2982 | 2.28M | ((c >= 0x370) && (c <= 0x37D)) || |
2983 | 2.28M | ((c >= 0x37F) && (c <= 0x1FFF)) || |
2984 | 2.28M | ((c >= 0x200C) && (c <= 0x200D)) || |
2985 | 2.28M | ((c >= 0x2070) && (c <= 0x218F)) || |
2986 | 2.28M | ((c >= 0x2C00) && (c <= 0x2FEF)) || |
2987 | 2.28M | ((c >= 0x3001) && (c <= 0xD7FF)) || |
2988 | 2.28M | ((c >= 0xF900) && (c <= 0xFDCF)) || |
2989 | 2.28M | ((c >= 0xFDF0) && (c <= 0xFFFD)) || |
2990 | 2.28M | ((c >= 0x10000) && (c <= 0xEFFFF)))) |
2991 | 475k | return(1); |
2992 | 1.80M | return(0); |
2993 | 2.28M | } |
2994 | | |
2995 | | static int |
2996 | 17.0M | xmlIsNameCharNew(int c) { |
2997 | | /* |
2998 | | * Use the new checks of production [4] [4a] amd [5] of the |
2999 | | * Update 5 of XML-1.0 |
3000 | | */ |
3001 | 17.0M | if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ |
3002 | 17.0M | (((c >= 'a') && (c <= 'z')) || |
3003 | 17.0M | ((c >= 'A') && (c <= 'Z')) || |
3004 | 17.0M | ((c >= '0') && (c <= '9')) || /* !start */ |
3005 | 17.0M | (c == '_') || (c == ':') || |
3006 | 17.0M | (c == '-') || (c == '.') || (c == 0xB7) || /* !start */ |
3007 | 17.0M | ((c >= 0xC0) && (c <= 0xD6)) || |
3008 | 17.0M | ((c >= 0xD8) && (c <= 0xF6)) || |
3009 | 17.0M | ((c >= 0xF8) && (c <= 0x2FF)) || |
3010 | 17.0M | ((c >= 0x300) && (c <= 0x36F)) || /* !start */ |
3011 | 17.0M | ((c >= 0x370) && (c <= 0x37D)) || |
3012 | 17.0M | ((c >= 0x37F) && (c <= 0x1FFF)) || |
3013 | 17.0M | ((c >= 0x200C) && (c <= 0x200D)) || |
3014 | 17.0M | ((c >= 0x203F) && (c <= 0x2040)) || /* !start */ |
3015 | 17.0M | ((c >= 0x2070) && (c <= 0x218F)) || |
3016 | 17.0M | ((c >= 0x2C00) && (c <= 0x2FEF)) || |
3017 | 17.0M | ((c >= 0x3001) && (c <= 0xD7FF)) || |
3018 | 17.0M | ((c >= 0xF900) && (c <= 0xFDCF)) || |
3019 | 17.0M | ((c >= 0xFDF0) && (c <= 0xFFFD)) || |
3020 | 17.0M | ((c >= 0x10000) && (c <= 0xEFFFF)))) |
3021 | 16.5M | return(1); |
3022 | 501k | return(0); |
3023 | 17.0M | } |
3024 | | |
3025 | | static int |
3026 | 0 | xmlIsNameStartCharOld(int c) { |
3027 | 0 | if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ |
3028 | 0 | ((IS_LETTER(c) || (c == '_') || (c == ':')))) |
3029 | 0 | return(1); |
3030 | 0 | return(0); |
3031 | 0 | } |
3032 | | |
3033 | | static int |
3034 | 0 | xmlIsNameCharOld(int c) { |
3035 | 0 | if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ |
3036 | 0 | ((IS_LETTER(c)) || (IS_DIGIT(c)) || |
3037 | 0 | (c == '.') || (c == '-') || |
3038 | 0 | (c == '_') || (c == ':') || |
3039 | 0 | (IS_COMBINING(c)) || |
3040 | 0 | (IS_EXTENDER(c)))) |
3041 | 0 | return(1); |
3042 | 0 | return(0); |
3043 | 0 | } |
3044 | | |
3045 | | static int |
3046 | 2.28M | xmlIsNameStartChar(int c, int old10) { |
3047 | 2.28M | if (!old10) |
3048 | 2.28M | return(xmlIsNameStartCharNew(c)); |
3049 | 0 | else |
3050 | 0 | return(xmlIsNameStartCharOld(c)); |
3051 | 2.28M | } |
3052 | | |
3053 | | static int |
3054 | 17.0M | xmlIsNameChar(int c, int old10) { |
3055 | 17.0M | if (!old10) |
3056 | 17.0M | return(xmlIsNameCharNew(c)); |
3057 | 0 | else |
3058 | 0 | return(xmlIsNameCharOld(c)); |
3059 | 17.0M | } |
3060 | | |
3061 | | /* |
3062 | | * Scan an XML Name, NCName or Nmtoken. |
3063 | | * |
3064 | | * Returns a pointer to the end of the name on success. If the |
3065 | | * name is invalid, returns `ptr`. If the name is longer than |
3066 | | * `maxSize` bytes, returns NULL. |
3067 | | * |
3068 | | * @param ptr pointer to the start of the name |
3069 | | * @param maxSize maximum size in bytes |
3070 | | * @param flags XML_SCAN_* flags |
3071 | | * @returns a pointer to the end of the name or NULL |
3072 | | */ |
3073 | | const xmlChar * |
3074 | 378k | xmlScanName(const xmlChar *ptr, size_t maxSize, int flags) { |
3075 | 378k | int stop = flags & XML_SCAN_NC ? ':' : 0; |
3076 | 378k | int old10 = flags & XML_SCAN_OLD10 ? 1 : 0; |
3077 | | |
3078 | 3.41M | while (1) { |
3079 | 3.41M | int c, len; |
3080 | | |
3081 | 3.41M | c = *ptr; |
3082 | 3.41M | if (c < 0x80) { |
3083 | 2.06M | if (c == stop) |
3084 | 207 | break; |
3085 | 2.06M | len = 1; |
3086 | 2.06M | } else { |
3087 | 1.34M | len = 4; |
3088 | 1.34M | c = xmlGetUTF8Char(ptr, &len); |
3089 | 1.34M | if (c < 0) |
3090 | 513 | break; |
3091 | 1.34M | } |
3092 | | |
3093 | 3.41M | if (flags & XML_SCAN_NMTOKEN ? |
3094 | 3.03M | !xmlIsNameChar(c, old10) : |
3095 | 3.41M | !xmlIsNameStartChar(c, old10)) |
3096 | 378k | break; |
3097 | | |
3098 | 3.03M | if ((size_t) len > maxSize) |
3099 | 0 | return(NULL); |
3100 | 3.03M | ptr += len; |
3101 | 3.03M | maxSize -= len; |
3102 | 3.03M | flags |= XML_SCAN_NMTOKEN; |
3103 | 3.03M | } |
3104 | | |
3105 | 378k | return(ptr); |
3106 | 378k | } |
3107 | | |
3108 | | static const xmlChar * |
3109 | 132k | xmlParseNameComplex(xmlParserCtxtPtr ctxt) { |
3110 | 132k | const xmlChar *ret; |
3111 | 132k | int len = 0, l; |
3112 | 132k | int c; |
3113 | 132k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3114 | 132k | XML_MAX_TEXT_LENGTH : |
3115 | 132k | XML_MAX_NAME_LENGTH; |
3116 | 132k | int old10 = (ctxt->options & XML_PARSE_OLD10) ? 1 : 0; |
3117 | | |
3118 | | /* |
3119 | | * Handler for more complex cases |
3120 | | */ |
3121 | 132k | c = xmlCurrentChar(ctxt, &l); |
3122 | 132k | if (!xmlIsNameStartChar(c, old10)) |
3123 | 99.9k | return(NULL); |
3124 | 32.3k | len += l; |
3125 | 32.3k | NEXTL(l); |
3126 | 32.3k | c = xmlCurrentChar(ctxt, &l); |
3127 | 3.94M | while (xmlIsNameChar(c, old10)) { |
3128 | 3.91M | if (len <= INT_MAX - l) |
3129 | 3.91M | len += l; |
3130 | 3.91M | NEXTL(l); |
3131 | 3.91M | c = xmlCurrentChar(ctxt, &l); |
3132 | 3.91M | } |
3133 | 32.3k | if (len > maxLength) { |
3134 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Name"); |
3135 | 0 | return(NULL); |
3136 | 0 | } |
3137 | 32.3k | if (ctxt->input->cur - ctxt->input->base < len) { |
3138 | | /* |
3139 | | * There were a couple of bugs where PERefs lead to to a change |
3140 | | * of the buffer. Check the buffer size to avoid passing an invalid |
3141 | | * pointer to xmlDictLookup. |
3142 | | */ |
3143 | 0 | xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, |
3144 | 0 | "unexpected change of input buffer"); |
3145 | 0 | return (NULL); |
3146 | 0 | } |
3147 | 32.3k | if ((*ctxt->input->cur == '\n') && (ctxt->input->cur[-1] == '\r')) |
3148 | 587 | ret = xmlDictLookup(ctxt->dict, ctxt->input->cur - (len + 1), len); |
3149 | 31.8k | else |
3150 | 31.8k | ret = xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len); |
3151 | 32.3k | if (ret == NULL) |
3152 | 0 | xmlErrMemory(ctxt); |
3153 | 32.3k | return(ret); |
3154 | 32.3k | } |
3155 | | |
3156 | | /** |
3157 | | * parse an XML name. |
3158 | | * |
3159 | | * @deprecated Internal function, don't use. |
3160 | | * |
3161 | | * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | |
3162 | | * CombiningChar | Extender |
3163 | | * |
3164 | | * [5] Name ::= (Letter | '_' | ':') (NameChar)* |
3165 | | * |
3166 | | * [6] Names ::= Name (#x20 Name)* |
3167 | | * |
3168 | | * @param ctxt an XML parser context |
3169 | | * @returns the Name parsed or NULL |
3170 | | */ |
3171 | | |
3172 | | const xmlChar * |
3173 | 680k | xmlParseName(xmlParserCtxt *ctxt) { |
3174 | 680k | const xmlChar *in; |
3175 | 680k | const xmlChar *ret; |
3176 | 680k | size_t count = 0; |
3177 | 680k | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3178 | 680k | XML_MAX_TEXT_LENGTH : |
3179 | 680k | XML_MAX_NAME_LENGTH; |
3180 | | |
3181 | 680k | GROW; |
3182 | | |
3183 | | /* |
3184 | | * Accelerator for simple ASCII names |
3185 | | */ |
3186 | 680k | in = ctxt->input->cur; |
3187 | 680k | if (((*in >= 0x61) && (*in <= 0x7A)) || |
3188 | 680k | ((*in >= 0x41) && (*in <= 0x5A)) || |
3189 | 680k | (*in == '_') || (*in == ':')) { |
3190 | 568k | in++; |
3191 | 11.9M | while (((*in >= 0x61) && (*in <= 0x7A)) || |
3192 | 11.9M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3193 | 11.9M | ((*in >= 0x30) && (*in <= 0x39)) || |
3194 | 11.9M | (*in == '_') || (*in == '-') || |
3195 | 11.9M | (*in == ':') || (*in == '.')) |
3196 | 11.3M | in++; |
3197 | 568k | if ((*in > 0) && (*in < 0x80)) { |
3198 | 548k | count = in - ctxt->input->cur; |
3199 | 548k | if (count > maxLength) { |
3200 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Name"); |
3201 | 0 | return(NULL); |
3202 | 0 | } |
3203 | 548k | ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count); |
3204 | 548k | ctxt->input->cur = in; |
3205 | 548k | ctxt->input->col += count; |
3206 | 548k | if (ret == NULL) |
3207 | 0 | xmlErrMemory(ctxt); |
3208 | 548k | return(ret); |
3209 | 548k | } |
3210 | 568k | } |
3211 | | /* accelerator for special cases */ |
3212 | 132k | return(xmlParseNameComplex(ctxt)); |
3213 | 680k | } |
3214 | | |
3215 | | static xmlHashedString |
3216 | 1.79M | xmlParseNCNameComplex(xmlParserCtxtPtr ctxt) { |
3217 | 1.79M | xmlHashedString ret; |
3218 | 1.79M | int len = 0, l; |
3219 | 1.79M | int c; |
3220 | 1.79M | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3221 | 1.79M | XML_MAX_TEXT_LENGTH : |
3222 | 1.79M | XML_MAX_NAME_LENGTH; |
3223 | 1.79M | int old10 = (ctxt->options & XML_PARSE_OLD10) ? 1 : 0; |
3224 | 1.79M | size_t startPosition = 0; |
3225 | | |
3226 | 1.79M | ret.name = NULL; |
3227 | 1.79M | ret.hashValue = 0; |
3228 | | |
3229 | | /* |
3230 | | * Handler for more complex cases |
3231 | | */ |
3232 | 1.79M | startPosition = CUR_PTR - BASE_PTR; |
3233 | 1.79M | c = xmlCurrentChar(ctxt, &l); |
3234 | 1.79M | if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */ |
3235 | 1.79M | (!xmlIsNameStartChar(c, old10) || (c == ':'))) { |
3236 | 1.74M | return(ret); |
3237 | 1.74M | } |
3238 | | |
3239 | 2.96M | while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */ |
3240 | 2.96M | (xmlIsNameChar(c, old10) && (c != ':'))) { |
3241 | 2.90M | if (len <= INT_MAX - l) |
3242 | 2.90M | len += l; |
3243 | 2.90M | NEXTL(l); |
3244 | 2.90M | c = xmlCurrentChar(ctxt, &l); |
3245 | 2.90M | } |
3246 | 53.0k | if (len > maxLength) { |
3247 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); |
3248 | 0 | return(ret); |
3249 | 0 | } |
3250 | 53.0k | ret = xmlDictLookupHashed(ctxt->dict, (BASE_PTR + startPosition), len); |
3251 | 53.0k | if (ret.name == NULL) |
3252 | 0 | xmlErrMemory(ctxt); |
3253 | 53.0k | return(ret); |
3254 | 53.0k | } |
3255 | | |
3256 | | /** |
3257 | | * parse an XML name. |
3258 | | * |
3259 | | * [4NS] NCNameChar ::= Letter | Digit | '.' | '-' | '_' | |
3260 | | * CombiningChar | Extender |
3261 | | * |
3262 | | * [5NS] NCName ::= (Letter | '_') (NCNameChar)* |
3263 | | * |
3264 | | * @param ctxt an XML parser context |
3265 | | * @returns the Name parsed or NULL |
3266 | | */ |
3267 | | |
3268 | | static xmlHashedString |
3269 | 3.15M | xmlParseNCName(xmlParserCtxtPtr ctxt) { |
3270 | 3.15M | const xmlChar *in, *e; |
3271 | 3.15M | xmlHashedString ret; |
3272 | 3.15M | size_t count = 0; |
3273 | 3.15M | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3274 | 3.15M | XML_MAX_TEXT_LENGTH : |
3275 | 3.15M | XML_MAX_NAME_LENGTH; |
3276 | | |
3277 | 3.15M | ret.name = NULL; |
3278 | | |
3279 | | /* |
3280 | | * Accelerator for simple ASCII names |
3281 | | */ |
3282 | 3.15M | in = ctxt->input->cur; |
3283 | 3.15M | e = ctxt->input->end; |
3284 | 3.15M | if ((((*in >= 0x61) && (*in <= 0x7A)) || |
3285 | 3.15M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3286 | 3.15M | (*in == '_')) && (in < e)) { |
3287 | 1.39M | in++; |
3288 | 5.46M | while ((((*in >= 0x61) && (*in <= 0x7A)) || |
3289 | 5.46M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3290 | 5.46M | ((*in >= 0x30) && (*in <= 0x39)) || |
3291 | 5.46M | (*in == '_') || (*in == '-') || |
3292 | 5.46M | (*in == '.')) && (in < e)) |
3293 | 4.06M | in++; |
3294 | 1.39M | if (in >= e) |
3295 | 1.00k | goto complex; |
3296 | 1.39M | if ((*in > 0) && (*in < 0x80)) { |
3297 | 1.35M | count = in - ctxt->input->cur; |
3298 | 1.35M | if (count > maxLength) { |
3299 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); |
3300 | 0 | return(ret); |
3301 | 0 | } |
3302 | 1.35M | ret = xmlDictLookupHashed(ctxt->dict, ctxt->input->cur, count); |
3303 | 1.35M | ctxt->input->cur = in; |
3304 | 1.35M | ctxt->input->col += count; |
3305 | 1.35M | if (ret.name == NULL) { |
3306 | 0 | xmlErrMemory(ctxt); |
3307 | 0 | } |
3308 | 1.35M | return(ret); |
3309 | 1.35M | } |
3310 | 1.39M | } |
3311 | 1.79M | complex: |
3312 | 1.79M | return(xmlParseNCNameComplex(ctxt)); |
3313 | 3.15M | } |
3314 | | |
3315 | | /** |
3316 | | * parse an XML name and compares for match |
3317 | | * (specialized for endtag parsing) |
3318 | | * |
3319 | | * @param ctxt an XML parser context |
3320 | | * @param other the name to compare with |
3321 | | * @returns NULL for an illegal name, (xmlChar*) 1 for success |
3322 | | * and the name for mismatch |
3323 | | */ |
3324 | | |
3325 | | static const xmlChar * |
3326 | 22.1k | xmlParseNameAndCompare(xmlParserCtxtPtr ctxt, xmlChar const *other) { |
3327 | 22.1k | register const xmlChar *cmp = other; |
3328 | 22.1k | register const xmlChar *in; |
3329 | 22.1k | const xmlChar *ret; |
3330 | | |
3331 | 22.1k | GROW; |
3332 | | |
3333 | 22.1k | in = ctxt->input->cur; |
3334 | 32.4k | while (*in != 0 && *in == *cmp) { |
3335 | 10.3k | ++in; |
3336 | 10.3k | ++cmp; |
3337 | 10.3k | } |
3338 | 22.1k | if (*cmp == 0 && (*in == '>' || IS_BLANK_CH (*in))) { |
3339 | | /* success */ |
3340 | 8.55k | ctxt->input->col += in - ctxt->input->cur; |
3341 | 8.55k | ctxt->input->cur = in; |
3342 | 8.55k | return (const xmlChar*) 1; |
3343 | 8.55k | } |
3344 | | /* failure (or end of input buffer), check with full function */ |
3345 | 13.5k | ret = xmlParseName (ctxt); |
3346 | | /* strings coming from the dictionary direct compare possible */ |
3347 | 13.5k | if (ret == other) { |
3348 | 1.18k | return (const xmlChar*) 1; |
3349 | 1.18k | } |
3350 | 12.3k | return ret; |
3351 | 13.5k | } |
3352 | | |
3353 | | /** |
3354 | | * Parse an XML name. |
3355 | | * |
3356 | | * @param ctxt an XML parser context |
3357 | | * @param str a pointer to the string pointer (IN/OUT) |
3358 | | * @returns the Name parsed or NULL. The `str` pointer |
3359 | | * is updated to the current location in the string. |
3360 | | */ |
3361 | | |
3362 | | static xmlChar * |
3363 | 373k | xmlParseStringName(xmlParserCtxtPtr ctxt, const xmlChar** str) { |
3364 | 373k | xmlChar *ret; |
3365 | 373k | const xmlChar *cur = *str; |
3366 | 373k | int flags = 0; |
3367 | 373k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3368 | 373k | XML_MAX_TEXT_LENGTH : |
3369 | 373k | XML_MAX_NAME_LENGTH; |
3370 | | |
3371 | 373k | if (ctxt->options & XML_PARSE_OLD10) |
3372 | 0 | flags |= XML_SCAN_OLD10; |
3373 | | |
3374 | 373k | cur = xmlScanName(*str, maxLength, flags); |
3375 | 373k | if (cur == NULL) { |
3376 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); |
3377 | 0 | return(NULL); |
3378 | 0 | } |
3379 | 373k | if (cur == *str) |
3380 | 997 | return(NULL); |
3381 | | |
3382 | 372k | ret = xmlStrndup(*str, cur - *str); |
3383 | 372k | if (ret == NULL) |
3384 | 0 | xmlErrMemory(ctxt); |
3385 | 372k | *str = cur; |
3386 | 372k | return(ret); |
3387 | 373k | } |
3388 | | |
3389 | | /** |
3390 | | * parse an XML Nmtoken. |
3391 | | * |
3392 | | * @deprecated Internal function, don't use. |
3393 | | * |
3394 | | * [7] Nmtoken ::= (NameChar)+ |
3395 | | * |
3396 | | * [8] Nmtokens ::= Nmtoken (#x20 Nmtoken)* |
3397 | | * |
3398 | | * @param ctxt an XML parser context |
3399 | | * @returns the Nmtoken parsed or NULL |
3400 | | */ |
3401 | | |
3402 | | xmlChar * |
3403 | 48.0k | xmlParseNmtoken(xmlParserCtxt *ctxt) { |
3404 | 48.0k | xmlChar buf[XML_MAX_NAMELEN + 5]; |
3405 | 48.0k | xmlChar *ret; |
3406 | 48.0k | int len = 0, l; |
3407 | 48.0k | int c; |
3408 | 48.0k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3409 | 48.0k | XML_MAX_TEXT_LENGTH : |
3410 | 48.0k | XML_MAX_NAME_LENGTH; |
3411 | 48.0k | int old10 = (ctxt->options & XML_PARSE_OLD10) ? 1 : 0; |
3412 | | |
3413 | 48.0k | c = xmlCurrentChar(ctxt, &l); |
3414 | | |
3415 | 213k | while (xmlIsNameChar(c, old10)) { |
3416 | 166k | COPY_BUF(buf, len, c); |
3417 | 166k | NEXTL(l); |
3418 | 166k | c = xmlCurrentChar(ctxt, &l); |
3419 | 166k | if (len >= XML_MAX_NAMELEN) { |
3420 | | /* |
3421 | | * Okay someone managed to make a huge token, so he's ready to pay |
3422 | | * for the processing speed. |
3423 | | */ |
3424 | 1.07k | xmlChar *buffer; |
3425 | 1.07k | int max = len * 2; |
3426 | | |
3427 | 1.07k | buffer = xmlMalloc(max); |
3428 | 1.07k | if (buffer == NULL) { |
3429 | 0 | xmlErrMemory(ctxt); |
3430 | 0 | return(NULL); |
3431 | 0 | } |
3432 | 1.07k | memcpy(buffer, buf, len); |
3433 | 6.88M | while (xmlIsNameChar(c, old10)) { |
3434 | 6.88M | if (len + 10 > max) { |
3435 | 2.89k | xmlChar *tmp; |
3436 | 2.89k | int newSize; |
3437 | | |
3438 | 2.89k | newSize = xmlGrowCapacity(max, 1, 1, maxLength); |
3439 | 2.89k | if (newSize < 0) { |
3440 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NmToken"); |
3441 | 0 | xmlFree(buffer); |
3442 | 0 | return(NULL); |
3443 | 0 | } |
3444 | 2.89k | tmp = xmlRealloc(buffer, newSize); |
3445 | 2.89k | if (tmp == NULL) { |
3446 | 0 | xmlErrMemory(ctxt); |
3447 | 0 | xmlFree(buffer); |
3448 | 0 | return(NULL); |
3449 | 0 | } |
3450 | 2.89k | buffer = tmp; |
3451 | 2.89k | max = newSize; |
3452 | 2.89k | } |
3453 | 6.88M | COPY_BUF(buffer, len, c); |
3454 | 6.88M | NEXTL(l); |
3455 | 6.88M | c = xmlCurrentChar(ctxt, &l); |
3456 | 6.88M | } |
3457 | 1.07k | buffer[len] = 0; |
3458 | 1.07k | return(buffer); |
3459 | 1.07k | } |
3460 | 166k | } |
3461 | 46.9k | if (len == 0) |
3462 | 17.4k | return(NULL); |
3463 | 29.5k | if (len > maxLength) { |
3464 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NmToken"); |
3465 | 0 | return(NULL); |
3466 | 0 | } |
3467 | 29.5k | ret = xmlStrndup(buf, len); |
3468 | 29.5k | if (ret == NULL) |
3469 | 0 | xmlErrMemory(ctxt); |
3470 | 29.5k | return(ret); |
3471 | 29.5k | } |
3472 | | |
3473 | | /** |
3474 | | * Validate an entity value and expand parameter entities. |
3475 | | * |
3476 | | * @param ctxt parser context |
3477 | | * @param buf string buffer |
3478 | | * @param str entity value |
3479 | | * @param length size of entity value |
3480 | | * @param depth nesting depth |
3481 | | */ |
3482 | | static void |
3483 | | xmlExpandPEsInEntityValue(xmlParserCtxtPtr ctxt, xmlSBuf *buf, |
3484 | 20.8k | const xmlChar *str, int length, int depth) { |
3485 | 20.8k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
3486 | 20.8k | const xmlChar *end, *chunk; |
3487 | 20.8k | int c, l; |
3488 | | |
3489 | 20.8k | if (str == NULL) |
3490 | 0 | return; |
3491 | | |
3492 | 20.8k | depth += 1; |
3493 | 20.8k | if (depth > maxDepth) { |
3494 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
3495 | 0 | "Maximum entity nesting depth exceeded"); |
3496 | 0 | return; |
3497 | 0 | } |
3498 | | |
3499 | 20.8k | end = str + length; |
3500 | 20.8k | chunk = str; |
3501 | | |
3502 | 69.0M | while ((str < end) && (!PARSER_STOPPED(ctxt))) { |
3503 | 69.0M | c = *str; |
3504 | | |
3505 | 69.0M | if (c >= 0x80) { |
3506 | 54.9M | l = xmlUTF8MultibyteLen(ctxt, str, |
3507 | 54.9M | "invalid character in entity value\n"); |
3508 | 54.9M | if (l == 0) { |
3509 | 24.7M | if (chunk < str) |
3510 | 55.2k | xmlSBufAddString(buf, chunk, str - chunk); |
3511 | 24.7M | xmlSBufAddReplChar(buf); |
3512 | 24.7M | str += 1; |
3513 | 24.7M | chunk = str; |
3514 | 30.1M | } else { |
3515 | 30.1M | str += l; |
3516 | 30.1M | } |
3517 | 54.9M | } else if (c == '&') { |
3518 | 29.0k | if (str[1] == '#') { |
3519 | 16.9k | if (chunk < str) |
3520 | 7.13k | xmlSBufAddString(buf, chunk, str - chunk); |
3521 | | |
3522 | 16.9k | c = xmlParseStringCharRef(ctxt, &str); |
3523 | 16.9k | if (c == 0) |
3524 | 4.20k | return; |
3525 | | |
3526 | 12.7k | xmlSBufAddChar(buf, c); |
3527 | | |
3528 | 12.7k | chunk = str; |
3529 | 12.7k | } else { |
3530 | 12.0k | xmlChar *name; |
3531 | | |
3532 | | /* |
3533 | | * General entity references are checked for |
3534 | | * syntactic validity. |
3535 | | */ |
3536 | 12.0k | str++; |
3537 | 12.0k | name = xmlParseStringName(ctxt, &str); |
3538 | | |
3539 | 12.0k | if ((name == NULL) || (*str++ != ';')) { |
3540 | 1.09k | xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_CHAR_ERROR, |
3541 | 1.09k | "EntityValue: '&' forbidden except for entities " |
3542 | 1.09k | "references\n"); |
3543 | 1.09k | xmlFree(name); |
3544 | 1.09k | return; |
3545 | 1.09k | } |
3546 | | |
3547 | 10.9k | xmlFree(name); |
3548 | 10.9k | } |
3549 | 14.0M | } else if (c == '%') { |
3550 | 1.38k | xmlEntityPtr ent; |
3551 | | |
3552 | 1.38k | if (chunk < str) |
3553 | 1.01k | xmlSBufAddString(buf, chunk, str - chunk); |
3554 | | |
3555 | 1.38k | ent = xmlParseStringPEReference(ctxt, &str); |
3556 | 1.38k | if (ent == NULL) |
3557 | 1.01k | return; |
3558 | | |
3559 | 376 | if (!PARSER_EXTERNAL(ctxt)) { |
3560 | 376 | xmlFatalErr(ctxt, XML_ERR_ENTITY_PE_INTERNAL, NULL); |
3561 | 376 | return; |
3562 | 376 | } |
3563 | | |
3564 | 0 | if (ent->content == NULL) { |
3565 | | /* |
3566 | | * Note: external parsed entities will not be loaded, |
3567 | | * it is not required for a non-validating parser to |
3568 | | * complete external PEReferences coming from the |
3569 | | * internal subset |
3570 | | */ |
3571 | 0 | if (((ctxt->options & XML_PARSE_NO_XXE) == 0) && |
3572 | 0 | ((ctxt->replaceEntities) || |
3573 | 0 | (ctxt->validate))) { |
3574 | 0 | xmlLoadEntityContent(ctxt, ent); |
3575 | 0 | } else { |
3576 | 0 | xmlWarningMsg(ctxt, XML_ERR_ENTITY_PROCESSING, |
3577 | 0 | "not validating will not read content for " |
3578 | 0 | "PE entity %s\n", ent->name, NULL); |
3579 | 0 | } |
3580 | 0 | } |
3581 | | |
3582 | | /* |
3583 | | * TODO: Skip if ent->content is still NULL. |
3584 | | */ |
3585 | |
|
3586 | 0 | if (xmlParserEntityCheck(ctxt, ent->length)) |
3587 | 0 | return; |
3588 | | |
3589 | 0 | if (ent->flags & XML_ENT_EXPANDING) { |
3590 | 0 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
3591 | 0 | xmlHaltParser(ctxt); |
3592 | 0 | return; |
3593 | 0 | } |
3594 | | |
3595 | 0 | ent->flags |= XML_ENT_EXPANDING; |
3596 | 0 | xmlExpandPEsInEntityValue(ctxt, buf, ent->content, ent->length, |
3597 | 0 | depth); |
3598 | 0 | ent->flags &= ~XML_ENT_EXPANDING; |
3599 | |
|
3600 | 0 | chunk = str; |
3601 | 14.0M | } else { |
3602 | | /* Normal ASCII char */ |
3603 | 14.0M | if (!IS_BYTE_CHAR(c)) { |
3604 | 2.94M | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, |
3605 | 2.94M | "invalid character in entity value\n"); |
3606 | 2.94M | if (chunk < str) |
3607 | 15.3k | xmlSBufAddString(buf, chunk, str - chunk); |
3608 | 2.94M | xmlSBufAddReplChar(buf); |
3609 | 2.94M | str += 1; |
3610 | 2.94M | chunk = str; |
3611 | 11.1M | } else { |
3612 | 11.1M | str += 1; |
3613 | 11.1M | } |
3614 | 14.0M | } |
3615 | 69.0M | } |
3616 | | |
3617 | 14.1k | if (chunk < str) |
3618 | 12.4k | xmlSBufAddString(buf, chunk, str - chunk); |
3619 | 14.1k | } |
3620 | | |
3621 | | /** |
3622 | | * parse a value for ENTITY declarations |
3623 | | * |
3624 | | * @deprecated Internal function, don't use. |
3625 | | * |
3626 | | * [9] EntityValue ::= '"' ([^%&"] | PEReference | Reference)* '"' | |
3627 | | * "'" ([^%&'] | PEReference | Reference)* "'" |
3628 | | * |
3629 | | * @param ctxt an XML parser context |
3630 | | * @param orig if non-NULL store a copy of the original entity value |
3631 | | * @returns the EntityValue parsed with reference substituted or NULL |
3632 | | */ |
3633 | | xmlChar * |
3634 | 20.8k | xmlParseEntityValue(xmlParserCtxt *ctxt, xmlChar **orig) { |
3635 | 20.8k | unsigned maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3636 | 20.8k | XML_MAX_HUGE_LENGTH : |
3637 | 20.8k | XML_MAX_TEXT_LENGTH; |
3638 | 20.8k | xmlSBuf buf; |
3639 | 20.8k | const xmlChar *start; |
3640 | 20.8k | int quote, length; |
3641 | | |
3642 | 20.8k | xmlSBufInit(&buf, maxLength); |
3643 | | |
3644 | 20.8k | GROW; |
3645 | | |
3646 | 20.8k | quote = CUR; |
3647 | 20.8k | if ((quote != '"') && (quote != '\'')) { |
3648 | 0 | xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL); |
3649 | 0 | return(NULL); |
3650 | 0 | } |
3651 | 20.8k | CUR_PTR++; |
3652 | | |
3653 | 20.8k | length = 0; |
3654 | | |
3655 | | /* |
3656 | | * Copy raw content of the entity into a buffer |
3657 | | */ |
3658 | 101M | while (1) { |
3659 | 101M | int c; |
3660 | | |
3661 | 101M | if (PARSER_STOPPED(ctxt)) |
3662 | 0 | goto error; |
3663 | | |
3664 | 101M | if (CUR_PTR >= ctxt->input->end) { |
3665 | 34 | xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_NOT_FINISHED, NULL); |
3666 | 34 | goto error; |
3667 | 34 | } |
3668 | | |
3669 | 101M | c = CUR; |
3670 | | |
3671 | 101M | if (c == 0) { |
3672 | 11 | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, |
3673 | 11 | "invalid character in entity value\n"); |
3674 | 11 | goto error; |
3675 | 11 | } |
3676 | 101M | if (c == quote) |
3677 | 20.8k | break; |
3678 | 101M | NEXTL(1); |
3679 | 101M | length += 1; |
3680 | | |
3681 | | /* |
3682 | | * TODO: Check growth threshold |
3683 | | */ |
3684 | 101M | if (ctxt->input->end - CUR_PTR < 10) |
3685 | 35.0k | GROW; |
3686 | 101M | } |
3687 | | |
3688 | 20.8k | start = CUR_PTR - length; |
3689 | | |
3690 | 20.8k | if (orig != NULL) { |
3691 | 20.8k | *orig = xmlStrndup(start, length); |
3692 | 20.8k | if (*orig == NULL) |
3693 | 0 | xmlErrMemory(ctxt); |
3694 | 20.8k | } |
3695 | | |
3696 | 20.8k | xmlExpandPEsInEntityValue(ctxt, &buf, start, length, ctxt->inputNr); |
3697 | | |
3698 | 20.8k | NEXTL(1); |
3699 | | |
3700 | 20.8k | return(xmlSBufFinish(&buf, NULL, ctxt, "entity length too long")); |
3701 | | |
3702 | 45 | error: |
3703 | 45 | xmlSBufCleanup(&buf, ctxt, "entity length too long"); |
3704 | 45 | return(NULL); |
3705 | 20.8k | } |
3706 | | |
3707 | | /** |
3708 | | * Check an entity reference in an attribute value for validity |
3709 | | * without expanding it. |
3710 | | * |
3711 | | * @param ctxt parser context |
3712 | | * @param pent entity |
3713 | | * @param depth nesting depth |
3714 | | */ |
3715 | | static void |
3716 | 0 | xmlCheckEntityInAttValue(xmlParserCtxtPtr ctxt, xmlEntityPtr pent, int depth) { |
3717 | 0 | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
3718 | 0 | const xmlChar *str; |
3719 | 0 | unsigned long expandedSize = pent->length; |
3720 | 0 | int c, flags; |
3721 | |
|
3722 | 0 | depth += 1; |
3723 | 0 | if (depth > maxDepth) { |
3724 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
3725 | 0 | "Maximum entity nesting depth exceeded"); |
3726 | 0 | return; |
3727 | 0 | } |
3728 | | |
3729 | 0 | if (pent->flags & XML_ENT_EXPANDING) { |
3730 | 0 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
3731 | 0 | xmlHaltParser(ctxt); |
3732 | 0 | return; |
3733 | 0 | } |
3734 | | |
3735 | | /* |
3736 | | * If we're parsing a default attribute value in DTD content, |
3737 | | * the entity might reference other entities which weren't |
3738 | | * defined yet, so the check isn't reliable. |
3739 | | */ |
3740 | 0 | if (ctxt->inSubset == 0) |
3741 | 0 | flags = XML_ENT_CHECKED | XML_ENT_VALIDATED; |
3742 | 0 | else |
3743 | 0 | flags = XML_ENT_VALIDATED; |
3744 | |
|
3745 | 0 | str = pent->content; |
3746 | 0 | if (str == NULL) |
3747 | 0 | goto done; |
3748 | | |
3749 | | /* |
3750 | | * Note that entity values are already validated. We only check |
3751 | | * for illegal less-than signs and compute the expanded size |
3752 | | * of the entity. No special handling for multi-byte characters |
3753 | | * is needed. |
3754 | | */ |
3755 | 0 | while (!PARSER_STOPPED(ctxt)) { |
3756 | 0 | c = *str; |
3757 | |
|
3758 | 0 | if (c != '&') { |
3759 | 0 | if (c == 0) |
3760 | 0 | break; |
3761 | | |
3762 | 0 | if (c == '<') |
3763 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, |
3764 | 0 | "'<' in entity '%s' is not allowed in attributes " |
3765 | 0 | "values\n", pent->name); |
3766 | |
|
3767 | 0 | str += 1; |
3768 | 0 | } else if (str[1] == '#') { |
3769 | 0 | int val; |
3770 | |
|
3771 | 0 | val = xmlParseStringCharRef(ctxt, &str); |
3772 | 0 | if (val == 0) { |
3773 | 0 | pent->content[0] = 0; |
3774 | 0 | break; |
3775 | 0 | } |
3776 | 0 | } else { |
3777 | 0 | xmlChar *name; |
3778 | 0 | xmlEntityPtr ent; |
3779 | |
|
3780 | 0 | name = xmlParseStringEntityRef(ctxt, &str); |
3781 | 0 | if (name == NULL) { |
3782 | 0 | pent->content[0] = 0; |
3783 | 0 | break; |
3784 | 0 | } |
3785 | | |
3786 | 0 | ent = xmlLookupGeneralEntity(ctxt, name, /* inAttr */ 1); |
3787 | 0 | xmlFree(name); |
3788 | |
|
3789 | 0 | if ((ent != NULL) && |
3790 | 0 | (ent->etype != XML_INTERNAL_PREDEFINED_ENTITY)) { |
3791 | 0 | if ((ent->flags & flags) != flags) { |
3792 | 0 | pent->flags |= XML_ENT_EXPANDING; |
3793 | 0 | xmlCheckEntityInAttValue(ctxt, ent, depth); |
3794 | 0 | pent->flags &= ~XML_ENT_EXPANDING; |
3795 | 0 | } |
3796 | |
|
3797 | 0 | xmlSaturatedAdd(&expandedSize, ent->expandedSize); |
3798 | 0 | xmlSaturatedAdd(&expandedSize, XML_ENT_FIXED_COST); |
3799 | 0 | } |
3800 | 0 | } |
3801 | 0 | } |
3802 | |
|
3803 | 0 | done: |
3804 | 0 | if (ctxt->inSubset == 0) |
3805 | 0 | pent->expandedSize = expandedSize; |
3806 | |
|
3807 | 0 | pent->flags |= flags; |
3808 | 0 | } |
3809 | | |
3810 | | /** |
3811 | | * Expand general entity references in an entity or attribute value. |
3812 | | * Perform attribute value normalization. |
3813 | | * |
3814 | | * @param ctxt parser context |
3815 | | * @param buf string buffer |
3816 | | * @param str entity or attribute value |
3817 | | * @param pent entity for entity value, NULL for attribute values |
3818 | | * @param normalize whether to collapse whitespace |
3819 | | * @param inSpace whitespace state |
3820 | | * @param depth nesting depth |
3821 | | * @param check whether to check for amplification |
3822 | | * @returns whether there was a normalization change |
3823 | | */ |
3824 | | static int |
3825 | | xmlExpandEntityInAttValue(xmlParserCtxtPtr ctxt, xmlSBuf *buf, |
3826 | | const xmlChar *str, xmlEntityPtr pent, int normalize, |
3827 | 285k | int *inSpace, int depth, int check) { |
3828 | 285k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
3829 | 285k | int c, chunkSize; |
3830 | 285k | int normChange = 0; |
3831 | | |
3832 | 285k | if (str == NULL) |
3833 | 0 | return(0); |
3834 | | |
3835 | 285k | depth += 1; |
3836 | 285k | if (depth > maxDepth) { |
3837 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
3838 | 0 | "Maximum entity nesting depth exceeded"); |
3839 | 0 | return(0); |
3840 | 0 | } |
3841 | | |
3842 | 285k | if (pent != NULL) { |
3843 | 285k | if (pent->flags & XML_ENT_EXPANDING) { |
3844 | 11 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
3845 | 11 | xmlHaltParser(ctxt); |
3846 | 11 | return(0); |
3847 | 11 | } |
3848 | | |
3849 | 285k | if (check) { |
3850 | 285k | if (xmlParserEntityCheck(ctxt, pent->length)) |
3851 | 59 | return(0); |
3852 | 285k | } |
3853 | 285k | } |
3854 | | |
3855 | 285k | chunkSize = 0; |
3856 | | |
3857 | | /* |
3858 | | * Note that entity values are already validated. No special |
3859 | | * handling for multi-byte characters is needed. |
3860 | | */ |
3861 | 436M | while (!PARSER_STOPPED(ctxt)) { |
3862 | 436M | c = *str; |
3863 | | |
3864 | 436M | if (c != '&') { |
3865 | 436M | if (c == 0) |
3866 | 242k | break; |
3867 | | |
3868 | | /* |
3869 | | * If this function is called without an entity, it is used to |
3870 | | * expand entities in an attribute content where less-than was |
3871 | | * already unscaped and is allowed. |
3872 | | */ |
3873 | 435M | if ((pent != NULL) && (c == '<')) { |
3874 | 43.1k | xmlFatalErrMsgStr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, |
3875 | 43.1k | "'<' in entity '%s' is not allowed in attributes " |
3876 | 43.1k | "values\n", pent->name); |
3877 | 43.1k | break; |
3878 | 43.1k | } |
3879 | | |
3880 | 435M | if (c <= 0x20) { |
3881 | 16.2M | if ((normalize) && (*inSpace)) { |
3882 | | /* Skip char */ |
3883 | 124k | if (chunkSize > 0) { |
3884 | 17.6k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3885 | 17.6k | chunkSize = 0; |
3886 | 17.6k | } |
3887 | 124k | normChange = 1; |
3888 | 16.0M | } else if (c < 0x20) { |
3889 | 15.7M | if (chunkSize > 0) { |
3890 | 487k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3891 | 487k | chunkSize = 0; |
3892 | 487k | } |
3893 | | |
3894 | 15.7M | xmlSBufAddCString(buf, " ", 1); |
3895 | 15.7M | } else { |
3896 | 342k | chunkSize += 1; |
3897 | 342k | } |
3898 | | |
3899 | 16.2M | *inSpace = 1; |
3900 | 419M | } else { |
3901 | 419M | chunkSize += 1; |
3902 | 419M | *inSpace = 0; |
3903 | 419M | } |
3904 | | |
3905 | 435M | str += 1; |
3906 | 435M | } else if (str[1] == '#') { |
3907 | 161k | int val; |
3908 | | |
3909 | 161k | if (chunkSize > 0) { |
3910 | 160k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3911 | 160k | chunkSize = 0; |
3912 | 160k | } |
3913 | | |
3914 | 161k | val = xmlParseStringCharRef(ctxt, &str); |
3915 | 161k | if (val == 0) { |
3916 | 7 | if (pent != NULL) |
3917 | 7 | pent->content[0] = 0; |
3918 | 7 | break; |
3919 | 7 | } |
3920 | | |
3921 | 161k | if (val == ' ') { |
3922 | 70.6k | if ((normalize) && (*inSpace)) |
3923 | 298 | normChange = 1; |
3924 | 70.4k | else |
3925 | 70.4k | xmlSBufAddCString(buf, " ", 1); |
3926 | 70.6k | *inSpace = 1; |
3927 | 90.6k | } else { |
3928 | 90.6k | xmlSBufAddChar(buf, val); |
3929 | 90.6k | *inSpace = 0; |
3930 | 90.6k | } |
3931 | 359k | } else { |
3932 | 359k | xmlChar *name; |
3933 | 359k | xmlEntityPtr ent; |
3934 | | |
3935 | 359k | if (chunkSize > 0) { |
3936 | 276k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3937 | 276k | chunkSize = 0; |
3938 | 276k | } |
3939 | | |
3940 | 359k | name = xmlParseStringEntityRef(ctxt, &str); |
3941 | 359k | if (name == NULL) { |
3942 | 6 | if (pent != NULL) |
3943 | 6 | pent->content[0] = 0; |
3944 | 6 | break; |
3945 | 6 | } |
3946 | | |
3947 | 359k | ent = xmlLookupGeneralEntity(ctxt, name, /* inAttr */ 1); |
3948 | 359k | xmlFree(name); |
3949 | | |
3950 | 359k | if ((ent != NULL) && |
3951 | 359k | (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { |
3952 | 107k | if (ent->content == NULL) { |
3953 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_INTERNAL_ERROR, |
3954 | 0 | "predefined entity has no content\n"); |
3955 | 0 | break; |
3956 | 0 | } |
3957 | | |
3958 | 107k | xmlSBufAddString(buf, ent->content, ent->length); |
3959 | | |
3960 | 107k | *inSpace = 0; |
3961 | 252k | } else if ((ent != NULL) && (ent->content != NULL)) { |
3962 | 191k | if (pent != NULL) |
3963 | 191k | pent->flags |= XML_ENT_EXPANDING; |
3964 | 191k | normChange |= xmlExpandEntityInAttValue(ctxt, buf, |
3965 | 191k | ent->content, ent, normalize, inSpace, depth, check); |
3966 | 191k | if (pent != NULL) |
3967 | 191k | pent->flags &= ~XML_ENT_EXPANDING; |
3968 | 191k | } |
3969 | 359k | } |
3970 | 436M | } |
3971 | | |
3972 | 285k | if (chunkSize > 0) |
3973 | 236k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3974 | | |
3975 | 285k | return(normChange); |
3976 | 285k | } |
3977 | | |
3978 | | /** |
3979 | | * Expand general entity references in an entity or attribute value. |
3980 | | * Perform attribute value normalization. |
3981 | | * |
3982 | | * @param ctxt parser context |
3983 | | * @param str entity or attribute value |
3984 | | * @param normalize whether to collapse whitespace |
3985 | | * @returns the expanded attribtue value. |
3986 | | */ |
3987 | | xmlChar * |
3988 | | xmlExpandEntitiesInAttValue(xmlParserCtxt *ctxt, const xmlChar *str, |
3989 | 0 | int normalize) { |
3990 | 0 | unsigned maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3991 | 0 | XML_MAX_HUGE_LENGTH : |
3992 | 0 | XML_MAX_TEXT_LENGTH; |
3993 | 0 | xmlSBuf buf; |
3994 | 0 | int inSpace = 1; |
3995 | |
|
3996 | 0 | xmlSBufInit(&buf, maxLength); |
3997 | |
|
3998 | 0 | xmlExpandEntityInAttValue(ctxt, &buf, str, NULL, normalize, &inSpace, |
3999 | 0 | ctxt->inputNr, /* check */ 0); |
4000 | |
|
4001 | 0 | if ((normalize) && (inSpace) && (buf.size > 0)) |
4002 | 0 | buf.size--; |
4003 | |
|
4004 | 0 | return(xmlSBufFinish(&buf, NULL, ctxt, "AttValue length too long")); |
4005 | 0 | } |
4006 | | |
4007 | | /** |
4008 | | * parse a value for an attribute. |
4009 | | * |
4010 | | * NOTE: if no normalization is needed, the routine will return pointers |
4011 | | * directly from the data buffer. |
4012 | | * |
4013 | | * 3.3.3 Attribute-Value Normalization: |
4014 | | * |
4015 | | * Before the value of an attribute is passed to the application or |
4016 | | * checked for validity, the XML processor must normalize it as follows: |
4017 | | * |
4018 | | * - a character reference is processed by appending the referenced |
4019 | | * character to the attribute value |
4020 | | * - an entity reference is processed by recursively processing the |
4021 | | * replacement text of the entity |
4022 | | * - a whitespace character (\#x20, \#xD, \#xA, \#x9) is processed by |
4023 | | * appending \#x20 to the normalized value, except that only a single |
4024 | | * \#x20 is appended for a "#xD#xA" sequence that is part of an external |
4025 | | * parsed entity or the literal entity value of an internal parsed entity |
4026 | | * - other characters are processed by appending them to the normalized value |
4027 | | * |
4028 | | * If the declared value is not CDATA, then the XML processor must further |
4029 | | * process the normalized attribute value by discarding any leading and |
4030 | | * trailing space (\#x20) characters, and by replacing sequences of space |
4031 | | * (\#x20) characters by a single space (\#x20) character. |
4032 | | * All attributes for which no declaration has been read should be treated |
4033 | | * by a non-validating parser as if declared CDATA. |
4034 | | * |
4035 | | * @param ctxt an XML parser context |
4036 | | * @param attlen attribute len result |
4037 | | * @param outFlags resulting XML_ATTVAL_* flags |
4038 | | * @param special value from attsSpecial |
4039 | | * @param isNamespace whether this is a namespace declaration |
4040 | | * @returns the AttValue parsed or NULL. The value has to be freed by the |
4041 | | * caller if it was copied, this can be detected by val[*len] == 0. |
4042 | | */ |
4043 | | static xmlChar * |
4044 | | xmlParseAttValueInternal(xmlParserCtxtPtr ctxt, int *attlen, int *outFlags, |
4045 | 651k | int special, int isNamespace) { |
4046 | 651k | unsigned maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4047 | 651k | XML_MAX_HUGE_LENGTH : |
4048 | 651k | XML_MAX_TEXT_LENGTH; |
4049 | 651k | xmlSBuf buf; |
4050 | 651k | xmlChar *ret; |
4051 | 651k | int c, l, quote, entFlags, chunkSize; |
4052 | 651k | int inSpace = 1; |
4053 | 651k | int replaceEntities; |
4054 | 651k | int normalize = (special & XML_SPECIAL_TYPE_MASK) != 0; |
4055 | 651k | int attvalFlags = 0; |
4056 | | |
4057 | | /* Always expand namespace URIs */ |
4058 | 651k | replaceEntities = (ctxt->replaceEntities) || (isNamespace); |
4059 | | |
4060 | 651k | xmlSBufInit(&buf, maxLength); |
4061 | | |
4062 | 651k | GROW; |
4063 | | |
4064 | 651k | quote = CUR; |
4065 | 651k | if ((quote != '"') && (quote != '\'')) { |
4066 | 17.0k | xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL); |
4067 | 17.0k | return(NULL); |
4068 | 17.0k | } |
4069 | 634k | NEXTL(1); |
4070 | | |
4071 | 634k | if (ctxt->inSubset == 0) |
4072 | 618k | entFlags = XML_ENT_CHECKED | XML_ENT_VALIDATED; |
4073 | 16.5k | else |
4074 | 16.5k | entFlags = XML_ENT_VALIDATED; |
4075 | | |
4076 | 634k | inSpace = 1; |
4077 | 634k | chunkSize = 0; |
4078 | | |
4079 | 58.5M | while (1) { |
4080 | 58.5M | if (PARSER_STOPPED(ctxt)) |
4081 | 70 | goto error; |
4082 | | |
4083 | 58.5M | if (CUR_PTR >= ctxt->input->end) { |
4084 | 1.58k | xmlFatalErrMsg(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED, |
4085 | 1.58k | "AttValue: ' expected\n"); |
4086 | 1.58k | goto error; |
4087 | 1.58k | } |
4088 | | |
4089 | | /* |
4090 | | * TODO: Check growth threshold |
4091 | | */ |
4092 | 58.5M | if (ctxt->input->end - CUR_PTR < 10) |
4093 | 26.0k | GROW; |
4094 | | |
4095 | 58.5M | c = CUR; |
4096 | | |
4097 | 58.5M | if (c >= 0x80) { |
4098 | 36.4M | l = xmlUTF8MultibyteLen(ctxt, CUR_PTR, |
4099 | 36.4M | "invalid character in attribute value\n"); |
4100 | 36.4M | if (l == 0) { |
4101 | 32.5M | if (chunkSize > 0) { |
4102 | 112k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4103 | 112k | chunkSize = 0; |
4104 | 112k | } |
4105 | 32.5M | xmlSBufAddReplChar(&buf); |
4106 | 32.5M | NEXTL(1); |
4107 | 32.5M | } else { |
4108 | 3.94M | chunkSize += l; |
4109 | 3.94M | NEXTL(l); |
4110 | 3.94M | } |
4111 | | |
4112 | 36.4M | inSpace = 0; |
4113 | 36.4M | } else if (c != '&') { |
4114 | 21.8M | if (c > 0x20) { |
4115 | 12.2M | if (c == quote) |
4116 | 625k | break; |
4117 | | |
4118 | 11.6M | if (c == '<') |
4119 | 323k | xmlFatalErr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, NULL); |
4120 | | |
4121 | 11.6M | chunkSize += 1; |
4122 | 11.6M | inSpace = 0; |
4123 | 11.6M | } else if (!IS_BYTE_CHAR(c)) { |
4124 | 4.62M | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, |
4125 | 4.62M | "invalid character in attribute value\n"); |
4126 | 4.62M | if (chunkSize > 0) { |
4127 | 42.1k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4128 | 42.1k | chunkSize = 0; |
4129 | 42.1k | } |
4130 | 4.62M | xmlSBufAddReplChar(&buf); |
4131 | 4.62M | inSpace = 0; |
4132 | 4.96M | } else { |
4133 | | /* Whitespace */ |
4134 | 4.96M | if ((normalize) && (inSpace)) { |
4135 | | /* Skip char */ |
4136 | 41.7k | if (chunkSize > 0) { |
4137 | 12.4k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4138 | 12.4k | chunkSize = 0; |
4139 | 12.4k | } |
4140 | 41.7k | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4141 | 4.92M | } else if (c < 0x20) { |
4142 | | /* Convert to space */ |
4143 | 4.47M | if (chunkSize > 0) { |
4144 | 61.3k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4145 | 61.3k | chunkSize = 0; |
4146 | 61.3k | } |
4147 | | |
4148 | 4.47M | xmlSBufAddCString(&buf, " ", 1); |
4149 | 4.47M | } else { |
4150 | 447k | chunkSize += 1; |
4151 | 447k | } |
4152 | | |
4153 | 4.96M | inSpace = 1; |
4154 | | |
4155 | 4.96M | if ((c == 0xD) && (NXT(1) == 0xA)) |
4156 | 2.42k | CUR_PTR++; |
4157 | 4.96M | } |
4158 | | |
4159 | 21.2M | NEXTL(1); |
4160 | 21.2M | } else if (NXT(1) == '#') { |
4161 | 50.1k | int val; |
4162 | | |
4163 | 50.1k | if (chunkSize > 0) { |
4164 | 23.3k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4165 | 23.3k | chunkSize = 0; |
4166 | 23.3k | } |
4167 | | |
4168 | 50.1k | val = xmlParseCharRef(ctxt); |
4169 | 50.1k | if (val == 0) |
4170 | 7.16k | goto error; |
4171 | | |
4172 | 42.9k | if ((val == '&') && (!replaceEntities)) { |
4173 | | /* |
4174 | | * The reparsing will be done in xmlNodeParseContent() |
4175 | | * called from SAX2.c |
4176 | | */ |
4177 | 0 | xmlSBufAddCString(&buf, "&", 5); |
4178 | 0 | inSpace = 0; |
4179 | 42.9k | } else if (val == ' ') { |
4180 | 4.33k | if ((normalize) && (inSpace)) |
4181 | 315 | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4182 | 4.02k | else |
4183 | 4.02k | xmlSBufAddCString(&buf, " ", 1); |
4184 | 4.33k | inSpace = 1; |
4185 | 38.6k | } else { |
4186 | 38.6k | xmlSBufAddChar(&buf, val); |
4187 | 38.6k | inSpace = 0; |
4188 | 38.6k | } |
4189 | 170k | } else { |
4190 | 170k | const xmlChar *name; |
4191 | 170k | xmlEntityPtr ent; |
4192 | | |
4193 | 170k | if (chunkSize > 0) { |
4194 | 75.9k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4195 | 75.9k | chunkSize = 0; |
4196 | 75.9k | } |
4197 | | |
4198 | 170k | name = xmlParseEntityRefInternal(ctxt); |
4199 | 170k | if (name == NULL) { |
4200 | | /* |
4201 | | * Probably a literal '&' which wasn't escaped. |
4202 | | * TODO: Handle gracefully in recovery mode. |
4203 | | */ |
4204 | 38.4k | continue; |
4205 | 38.4k | } |
4206 | | |
4207 | 131k | ent = xmlLookupGeneralEntity(ctxt, name, /* isAttr */ 1); |
4208 | 131k | if (ent == NULL) |
4209 | 32.9k | continue; |
4210 | | |
4211 | 98.7k | if (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY) { |
4212 | 4.18k | if ((ent->content[0] == '&') && (!replaceEntities)) |
4213 | 0 | xmlSBufAddCString(&buf, "&", 5); |
4214 | 4.18k | else |
4215 | 4.18k | xmlSBufAddString(&buf, ent->content, ent->length); |
4216 | 4.18k | inSpace = 0; |
4217 | 94.5k | } else if (replaceEntities) { |
4218 | 94.5k | if (xmlExpandEntityInAttValue(ctxt, &buf, |
4219 | 94.5k | ent->content, ent, normalize, &inSpace, ctxt->inputNr, |
4220 | 94.5k | /* check */ 1) > 0) |
4221 | 7.00k | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4222 | 94.5k | } else { |
4223 | 0 | if ((ent->flags & entFlags) != entFlags) |
4224 | 0 | xmlCheckEntityInAttValue(ctxt, ent, ctxt->inputNr); |
4225 | |
|
4226 | 0 | if (xmlParserEntityCheck(ctxt, ent->expandedSize)) { |
4227 | 0 | ent->content[0] = 0; |
4228 | 0 | goto error; |
4229 | 0 | } |
4230 | | |
4231 | | /* |
4232 | | * Just output the reference |
4233 | | */ |
4234 | 0 | xmlSBufAddCString(&buf, "&", 1); |
4235 | 0 | xmlSBufAddString(&buf, ent->name, xmlStrlen(ent->name)); |
4236 | 0 | xmlSBufAddCString(&buf, ";", 1); |
4237 | |
|
4238 | 0 | inSpace = 0; |
4239 | 0 | } |
4240 | 98.7k | } |
4241 | 58.5M | } |
4242 | | |
4243 | 625k | if ((buf.mem == NULL) && (outFlags != NULL)) { |
4244 | 508k | ret = (xmlChar *) CUR_PTR - chunkSize; |
4245 | | |
4246 | 508k | if (attlen != NULL) |
4247 | 508k | *attlen = chunkSize; |
4248 | 508k | if ((normalize) && (inSpace) && (chunkSize > 0)) { |
4249 | 333 | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4250 | 333 | *attlen -= 1; |
4251 | 333 | } |
4252 | | |
4253 | | /* Report potential error */ |
4254 | 508k | xmlSBufCleanup(&buf, ctxt, "AttValue length too long"); |
4255 | 508k | } else { |
4256 | 117k | if (chunkSize > 0) |
4257 | 77.1k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4258 | | |
4259 | 117k | if ((normalize) && (inSpace) && (buf.size > 0)) { |
4260 | 3.82k | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4261 | 3.82k | buf.size--; |
4262 | 3.82k | } |
4263 | | |
4264 | 117k | ret = xmlSBufFinish(&buf, attlen, ctxt, "AttValue length too long"); |
4265 | 117k | attvalFlags |= XML_ATTVAL_ALLOC; |
4266 | | |
4267 | 117k | if (ret != NULL) { |
4268 | 117k | if (attlen != NULL) |
4269 | 101k | *attlen = buf.size; |
4270 | 117k | } |
4271 | 117k | } |
4272 | | |
4273 | 625k | if (outFlags != NULL) |
4274 | 609k | *outFlags = attvalFlags; |
4275 | | |
4276 | 625k | NEXTL(1); |
4277 | | |
4278 | 625k | return(ret); |
4279 | | |
4280 | 8.81k | error: |
4281 | 8.81k | xmlSBufCleanup(&buf, ctxt, "AttValue length too long"); |
4282 | 8.81k | return(NULL); |
4283 | 634k | } |
4284 | | |
4285 | | /** |
4286 | | * parse a value for an attribute |
4287 | | * Note: the parser won't do substitution of entities here, this |
4288 | | * will be handled later in #xmlStringGetNodeList |
4289 | | * |
4290 | | * @deprecated Internal function, don't use. |
4291 | | * |
4292 | | * [10] AttValue ::= '"' ([^<&"] | Reference)* '"' | |
4293 | | * "'" ([^<&'] | Reference)* "'" |
4294 | | * |
4295 | | * 3.3.3 Attribute-Value Normalization: |
4296 | | * |
4297 | | * Before the value of an attribute is passed to the application or |
4298 | | * checked for validity, the XML processor must normalize it as follows: |
4299 | | * |
4300 | | * - a character reference is processed by appending the referenced |
4301 | | * character to the attribute value |
4302 | | * - an entity reference is processed by recursively processing the |
4303 | | * replacement text of the entity |
4304 | | * - a whitespace character (\#x20, \#xD, \#xA, \#x9) is processed by |
4305 | | * appending \#x20 to the normalized value, except that only a single |
4306 | | * \#x20 is appended for a "#xD#xA" sequence that is part of an external |
4307 | | * parsed entity or the literal entity value of an internal parsed entity |
4308 | | * - other characters are processed by appending them to the normalized value |
4309 | | * |
4310 | | * If the declared value is not CDATA, then the XML processor must further |
4311 | | * process the normalized attribute value by discarding any leading and |
4312 | | * trailing space (\#x20) characters, and by replacing sequences of space |
4313 | | * (\#x20) characters by a single space (\#x20) character. |
4314 | | * All attributes for which no declaration has been read should be treated |
4315 | | * by a non-validating parser as if declared CDATA. |
4316 | | * |
4317 | | * @param ctxt an XML parser context |
4318 | | * @returns the AttValue parsed or NULL. The value has to be freed by the |
4319 | | * caller. |
4320 | | */ |
4321 | | xmlChar * |
4322 | 18.5k | xmlParseAttValue(xmlParserCtxt *ctxt) { |
4323 | 18.5k | if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL); |
4324 | 18.5k | return(xmlParseAttValueInternal(ctxt, NULL, NULL, 0, 0)); |
4325 | 18.5k | } |
4326 | | |
4327 | | /** |
4328 | | * parse an XML Literal |
4329 | | * |
4330 | | * @deprecated Internal function, don't use. |
4331 | | * |
4332 | | * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") |
4333 | | * |
4334 | | * @param ctxt an XML parser context |
4335 | | * @returns the SystemLiteral parsed or NULL |
4336 | | */ |
4337 | | |
4338 | | xmlChar * |
4339 | 3.88k | xmlParseSystemLiteral(xmlParserCtxt *ctxt) { |
4340 | 3.88k | xmlChar *buf = NULL; |
4341 | 3.88k | int len = 0; |
4342 | 3.88k | int size = XML_PARSER_BUFFER_SIZE; |
4343 | 3.88k | int cur, l; |
4344 | 3.88k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4345 | 3.88k | XML_MAX_TEXT_LENGTH : |
4346 | 3.88k | XML_MAX_NAME_LENGTH; |
4347 | 3.88k | xmlChar stop; |
4348 | | |
4349 | 3.88k | if (RAW == '"') { |
4350 | 1.07k | NEXT; |
4351 | 1.07k | stop = '"'; |
4352 | 2.81k | } else if (RAW == '\'') { |
4353 | 1.94k | NEXT; |
4354 | 1.94k | stop = '\''; |
4355 | 1.94k | } else { |
4356 | 877 | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); |
4357 | 877 | return(NULL); |
4358 | 877 | } |
4359 | | |
4360 | 3.01k | buf = xmlMalloc(size); |
4361 | 3.01k | if (buf == NULL) { |
4362 | 0 | xmlErrMemory(ctxt); |
4363 | 0 | return(NULL); |
4364 | 0 | } |
4365 | 3.01k | cur = xmlCurrentCharRecover(ctxt, &l); |
4366 | 2.53M | while ((IS_CHAR(cur)) && (cur != stop)) { /* checked */ |
4367 | 2.53M | if (len + 5 >= size) { |
4368 | 1.70k | xmlChar *tmp; |
4369 | 1.70k | int newSize; |
4370 | | |
4371 | 1.70k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
4372 | 1.70k | if (newSize < 0) { |
4373 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "SystemLiteral"); |
4374 | 0 | xmlFree(buf); |
4375 | 0 | return(NULL); |
4376 | 0 | } |
4377 | 1.70k | tmp = xmlRealloc(buf, newSize); |
4378 | 1.70k | if (tmp == NULL) { |
4379 | 0 | xmlFree(buf); |
4380 | 0 | xmlErrMemory(ctxt); |
4381 | 0 | return(NULL); |
4382 | 0 | } |
4383 | 1.70k | buf = tmp; |
4384 | 1.70k | size = newSize; |
4385 | 1.70k | } |
4386 | 2.53M | COPY_BUF(buf, len, cur); |
4387 | 2.53M | NEXTL(l); |
4388 | 2.53M | cur = xmlCurrentCharRecover(ctxt, &l); |
4389 | 2.53M | } |
4390 | 3.01k | buf[len] = 0; |
4391 | 3.01k | if (!IS_CHAR(cur)) { |
4392 | 361 | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); |
4393 | 2.65k | } else { |
4394 | 2.65k | NEXT; |
4395 | 2.65k | } |
4396 | 3.01k | return(buf); |
4397 | 3.01k | } |
4398 | | |
4399 | | /** |
4400 | | * parse an XML public literal |
4401 | | * |
4402 | | * @deprecated Internal function, don't use. |
4403 | | * |
4404 | | * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" |
4405 | | * |
4406 | | * @param ctxt an XML parser context |
4407 | | * @returns the PubidLiteral parsed or NULL. |
4408 | | */ |
4409 | | |
4410 | | xmlChar * |
4411 | 5.93k | xmlParsePubidLiteral(xmlParserCtxt *ctxt) { |
4412 | 5.93k | xmlChar *buf = NULL; |
4413 | 5.93k | int len = 0; |
4414 | 5.93k | int size = XML_PARSER_BUFFER_SIZE; |
4415 | 5.93k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4416 | 5.93k | XML_MAX_TEXT_LENGTH : |
4417 | 5.93k | XML_MAX_NAME_LENGTH; |
4418 | 5.93k | xmlChar cur; |
4419 | 5.93k | xmlChar stop; |
4420 | | |
4421 | 5.93k | if (RAW == '"') { |
4422 | 1.98k | NEXT; |
4423 | 1.98k | stop = '"'; |
4424 | 3.95k | } else if (RAW == '\'') { |
4425 | 815 | NEXT; |
4426 | 815 | stop = '\''; |
4427 | 3.13k | } else { |
4428 | 3.13k | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); |
4429 | 3.13k | return(NULL); |
4430 | 3.13k | } |
4431 | 2.79k | buf = xmlMalloc(size); |
4432 | 2.79k | if (buf == NULL) { |
4433 | 0 | xmlErrMemory(ctxt); |
4434 | 0 | return(NULL); |
4435 | 0 | } |
4436 | 2.79k | cur = CUR; |
4437 | 544k | while ((IS_PUBIDCHAR_CH(cur)) && (cur != stop) && |
4438 | 544k | (PARSER_STOPPED(ctxt) == 0)) { /* checked */ |
4439 | 541k | if (len + 1 >= size) { |
4440 | 356 | xmlChar *tmp; |
4441 | 356 | int newSize; |
4442 | | |
4443 | 356 | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
4444 | 356 | if (newSize < 0) { |
4445 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Public ID"); |
4446 | 0 | xmlFree(buf); |
4447 | 0 | return(NULL); |
4448 | 0 | } |
4449 | 356 | tmp = xmlRealloc(buf, newSize); |
4450 | 356 | if (tmp == NULL) { |
4451 | 0 | xmlErrMemory(ctxt); |
4452 | 0 | xmlFree(buf); |
4453 | 0 | return(NULL); |
4454 | 0 | } |
4455 | 356 | buf = tmp; |
4456 | 356 | size = newSize; |
4457 | 356 | } |
4458 | 541k | buf[len++] = cur; |
4459 | 541k | NEXT; |
4460 | 541k | cur = CUR; |
4461 | 541k | } |
4462 | 2.79k | buf[len] = 0; |
4463 | 2.79k | if (cur != stop) { |
4464 | 947 | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); |
4465 | 1.85k | } else { |
4466 | 1.85k | NEXTL(1); |
4467 | 1.85k | } |
4468 | 2.79k | return(buf); |
4469 | 2.79k | } |
4470 | | |
4471 | | static void xmlParseCharDataComplex(xmlParserCtxtPtr ctxt, int partial); |
4472 | | |
4473 | | /* |
4474 | | * used for the test in the inner loop of the char data testing |
4475 | | */ |
4476 | | static const unsigned char test_char_data[256] = { |
4477 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4478 | | 0x00, 0x09, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x9, CR/LF separated */ |
4479 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4480 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4481 | | 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x00, 0x27, /* & */ |
4482 | | 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F, |
4483 | | 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, |
4484 | | 0x38, 0x39, 0x3A, 0x3B, 0x00, 0x3D, 0x3E, 0x3F, /* < */ |
4485 | | 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, |
4486 | | 0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, |
4487 | | 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, |
4488 | | 0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x00, 0x5E, 0x5F, /* ] */ |
4489 | | 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, |
4490 | | 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, |
4491 | | 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, |
4492 | | 0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F, |
4493 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* non-ascii */ |
4494 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4495 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4496 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4497 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4498 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4499 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4500 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4501 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4502 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4503 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4504 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4505 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4506 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4507 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
4508 | | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 |
4509 | | }; |
4510 | | |
4511 | | static void |
4512 | | xmlCharacters(xmlParserCtxtPtr ctxt, const xmlChar *buf, int size, |
4513 | 1.03M | int isBlank) { |
4514 | 1.03M | int checkBlanks; |
4515 | | |
4516 | 1.03M | if ((ctxt->sax == NULL) || (ctxt->disableSAX)) |
4517 | 938k | return; |
4518 | | |
4519 | 95.5k | checkBlanks = (!ctxt->keepBlanks) || |
4520 | 95.5k | (ctxt->sax->ignorableWhitespace != ctxt->sax->characters); |
4521 | | |
4522 | | /* |
4523 | | * Calling areBlanks with only parts of a text node |
4524 | | * is fundamentally broken, making the NOBLANKS option |
4525 | | * essentially unusable. |
4526 | | */ |
4527 | 95.5k | if ((checkBlanks) && |
4528 | 95.5k | (areBlanks(ctxt, buf, size, isBlank))) { |
4529 | 0 | if ((ctxt->sax->ignorableWhitespace != NULL) && |
4530 | 0 | (ctxt->keepBlanks)) |
4531 | 0 | ctxt->sax->ignorableWhitespace(ctxt->userData, buf, size); |
4532 | 95.5k | } else { |
4533 | 95.5k | if (ctxt->sax->characters != NULL) |
4534 | 95.5k | ctxt->sax->characters(ctxt->userData, buf, size); |
4535 | | |
4536 | | /* |
4537 | | * The old code used to update this value for "complex" data |
4538 | | * even if checkBlanks was false. This was probably a bug. |
4539 | | */ |
4540 | 95.5k | if ((checkBlanks) && (*ctxt->space == -1)) |
4541 | 0 | *ctxt->space = -2; |
4542 | 95.5k | } |
4543 | 95.5k | } |
4544 | | |
4545 | | /** |
4546 | | * Parse character data. Always makes progress if the first char isn't |
4547 | | * '<' or '&'. |
4548 | | * |
4549 | | * The right angle bracket (>) may be represented using the string ">", |
4550 | | * and must, for compatibility, be escaped using ">" or a character |
4551 | | * reference when it appears in the string "]]>" in content, when that |
4552 | | * string is not marking the end of a CDATA section. |
4553 | | * |
4554 | | * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*) |
4555 | | * @param ctxt an XML parser context |
4556 | | * @param partial buffer may contain partial UTF-8 sequences |
4557 | | */ |
4558 | | static void |
4559 | 3.44M | xmlParseCharDataInternal(xmlParserCtxtPtr ctxt, int partial) { |
4560 | 3.44M | const xmlChar *in; |
4561 | 3.44M | int line = ctxt->input->line; |
4562 | 3.44M | int col = ctxt->input->col; |
4563 | 3.44M | int ccol; |
4564 | | |
4565 | 3.44M | GROW; |
4566 | | /* |
4567 | | * Accelerated common case where input don't need to be |
4568 | | * modified before passing it to the handler. |
4569 | | */ |
4570 | 3.44M | in = ctxt->input->cur; |
4571 | 3.44M | do { |
4572 | 3.46M | get_more_space: |
4573 | 3.47M | while (*in == 0x20) { in++; ctxt->input->col++; } |
4574 | 3.46M | if (*in == 0xA) { |
4575 | 927k | do { |
4576 | 927k | ctxt->input->line++; ctxt->input->col = 1; |
4577 | 927k | in++; |
4578 | 927k | } while (*in == 0xA); |
4579 | 13.3k | goto get_more_space; |
4580 | 13.3k | } |
4581 | 3.44M | if (*in == '<') { |
4582 | 7.06k | while (in > ctxt->input->cur) { |
4583 | 3.53k | const xmlChar *tmp = ctxt->input->cur; |
4584 | 3.53k | size_t nbchar = in - tmp; |
4585 | | |
4586 | 3.53k | if (nbchar > XML_MAX_ITEMS) |
4587 | 0 | nbchar = XML_MAX_ITEMS; |
4588 | 3.53k | ctxt->input->cur += nbchar; |
4589 | | |
4590 | 3.53k | xmlCharacters(ctxt, tmp, nbchar, 1); |
4591 | 3.53k | } |
4592 | 3.53k | return; |
4593 | 3.53k | } |
4594 | | |
4595 | 3.49M | get_more: |
4596 | 3.49M | ccol = ctxt->input->col; |
4597 | 11.5M | while (test_char_data[*in]) { |
4598 | 8.00M | in++; |
4599 | 8.00M | ccol++; |
4600 | 8.00M | } |
4601 | 3.49M | ctxt->input->col = ccol; |
4602 | 3.49M | if (*in == 0xA) { |
4603 | 1.32M | do { |
4604 | 1.32M | ctxt->input->line++; ctxt->input->col = 1; |
4605 | 1.32M | in++; |
4606 | 1.32M | } while (*in == 0xA); |
4607 | 18.3k | goto get_more; |
4608 | 18.3k | } |
4609 | 3.47M | if (*in == ']') { |
4610 | 33.1k | if ((in[1] == ']') && (in[2] == '>')) { |
4611 | 2.82k | xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); |
4612 | 2.82k | ctxt->input->cur = in + 1; |
4613 | 2.82k | return; |
4614 | 2.82k | } |
4615 | 30.3k | if ((!partial) || (ctxt->input->end - in >= 2)) { |
4616 | 30.3k | in++; |
4617 | 30.3k | ctxt->input->col++; |
4618 | 30.3k | goto get_more; |
4619 | 30.3k | } |
4620 | 30.3k | } |
4621 | 3.93M | while (in > ctxt->input->cur) { |
4622 | 497k | const xmlChar *tmp = ctxt->input->cur; |
4623 | 497k | size_t nbchar = in - tmp; |
4624 | | |
4625 | 497k | if (nbchar > XML_MAX_ITEMS) |
4626 | 0 | nbchar = XML_MAX_ITEMS; |
4627 | 497k | ctxt->input->cur += nbchar; |
4628 | | |
4629 | 497k | xmlCharacters(ctxt, tmp, nbchar, 0); |
4630 | | |
4631 | 497k | line = ctxt->input->line; |
4632 | 497k | col = ctxt->input->col; |
4633 | 497k | } |
4634 | 3.44M | ctxt->input->cur = in; |
4635 | 3.44M | if (*in == 0xD) { |
4636 | 29.3k | in++; |
4637 | 29.3k | if (*in == 0xA) { |
4638 | 3.71k | ctxt->input->cur = in; |
4639 | 3.71k | in++; |
4640 | 3.71k | ctxt->input->line++; ctxt->input->col = 1; |
4641 | 3.71k | continue; /* while */ |
4642 | 3.71k | } |
4643 | 25.5k | in--; |
4644 | 25.5k | } |
4645 | 3.43M | if (*in == '<') { |
4646 | 219k | return; |
4647 | 219k | } |
4648 | 3.21M | if (*in == '&') { |
4649 | 52.6k | return; |
4650 | 52.6k | } |
4651 | 3.16M | if ((partial) && (*in == ']') && (ctxt->input->end - in < 2)) { |
4652 | 0 | return; |
4653 | 0 | } |
4654 | 3.16M | SHRINK; |
4655 | 3.16M | GROW; |
4656 | 3.16M | in = ctxt->input->cur; |
4657 | 3.16M | } while (((*in >= 0x20) && (*in <= 0x7F)) || |
4658 | 3.16M | (*in == 0x09) || (*in == 0x0a)); |
4659 | 3.16M | ctxt->input->line = line; |
4660 | 3.16M | ctxt->input->col = col; |
4661 | 3.16M | xmlParseCharDataComplex(ctxt, partial); |
4662 | 3.16M | } |
4663 | | |
4664 | | /** |
4665 | | * Always makes progress if the first char isn't '<' or '&'. |
4666 | | * |
4667 | | * parse a CharData section.this is the fallback function |
4668 | | * of #xmlParseCharData when the parsing requires handling |
4669 | | * of non-ASCII characters. |
4670 | | * |
4671 | | * @param ctxt an XML parser context |
4672 | | * @param partial whether the input can end with truncated UTF-8 |
4673 | | */ |
4674 | | static void |
4675 | 3.16M | xmlParseCharDataComplex(xmlParserCtxtPtr ctxt, int partial) { |
4676 | 3.16M | xmlChar buf[XML_PARSER_BIG_BUFFER_SIZE + 5]; |
4677 | 3.16M | int nbchar = 0; |
4678 | 3.16M | int cur, l; |
4679 | | |
4680 | 3.16M | cur = xmlCurrentCharRecover(ctxt, &l); |
4681 | 30.9M | while ((cur != '<') && /* checked */ |
4682 | 30.9M | (cur != '&') && |
4683 | 30.9M | ((!partial) || (cur != ']') || |
4684 | 30.7M | (ctxt->input->end - ctxt->input->cur >= 2)) && |
4685 | 30.9M | (IS_CHAR(cur))) { |
4686 | 27.7M | if ((cur == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { |
4687 | 819 | xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); |
4688 | 819 | } |
4689 | 27.7M | COPY_BUF(buf, nbchar, cur); |
4690 | | /* move current position before possible calling of ctxt->sax->characters */ |
4691 | 27.7M | NEXTL(l); |
4692 | 27.7M | if (nbchar >= XML_PARSER_BIG_BUFFER_SIZE) { |
4693 | 164k | buf[nbchar] = 0; |
4694 | | |
4695 | 164k | xmlCharacters(ctxt, buf, nbchar, 0); |
4696 | 164k | nbchar = 0; |
4697 | 164k | SHRINK; |
4698 | 164k | } |
4699 | 27.7M | cur = xmlCurrentCharRecover(ctxt, &l); |
4700 | 27.7M | } |
4701 | 3.16M | if (nbchar != 0) { |
4702 | 368k | buf[nbchar] = 0; |
4703 | | |
4704 | 368k | xmlCharacters(ctxt, buf, nbchar, 0); |
4705 | 368k | } |
4706 | | /* |
4707 | | * cur == 0 can mean |
4708 | | * |
4709 | | * - End of buffer. |
4710 | | * - An actual 0 character. |
4711 | | * - An incomplete UTF-8 sequence. This is allowed if partial is set. |
4712 | | */ |
4713 | 3.16M | if (ctxt->input->cur < ctxt->input->end) { |
4714 | 3.16M | if ((cur == 0) && (CUR != 0)) { |
4715 | 407 | if (partial == 0) { |
4716 | 407 | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4717 | 407 | "Incomplete UTF-8 sequence starting with %02X\n", CUR); |
4718 | 407 | NEXTL(1); |
4719 | 407 | } |
4720 | 3.16M | } else if ((cur != '<') && (cur != '&') && (cur != ']')) { |
4721 | | /* Generate the error and skip the offending character */ |
4722 | 2.93M | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4723 | 2.93M | "PCDATA invalid Char value %d\n", cur); |
4724 | 2.93M | NEXTL(l); |
4725 | 2.93M | } |
4726 | 3.16M | } |
4727 | 3.16M | } |
4728 | | |
4729 | | /** |
4730 | | * @deprecated Internal function, don't use. |
4731 | | * @param ctxt an XML parser context |
4732 | | * @param cdata unused |
4733 | | */ |
4734 | | void |
4735 | 0 | xmlParseCharData(xmlParserCtxt *ctxt, ATTRIBUTE_UNUSED int cdata) { |
4736 | 0 | xmlParseCharDataInternal(ctxt, 0); |
4737 | 0 | } |
4738 | | |
4739 | | /** |
4740 | | * Parse an External ID or a Public ID |
4741 | | * |
4742 | | * @deprecated Internal function, don't use. |
4743 | | * |
4744 | | * NOTE: Productions [75] and [83] interact badly since [75] can generate |
4745 | | * `'PUBLIC' S PubidLiteral S SystemLiteral` |
4746 | | * |
4747 | | * [75] ExternalID ::= 'SYSTEM' S SystemLiteral |
4748 | | * | 'PUBLIC' S PubidLiteral S SystemLiteral |
4749 | | * |
4750 | | * [83] PublicID ::= 'PUBLIC' S PubidLiteral |
4751 | | * |
4752 | | * @param ctxt an XML parser context |
4753 | | * @param publicId a xmlChar** receiving PubidLiteral |
4754 | | * @param strict indicate whether we should restrict parsing to only |
4755 | | * production [75], see NOTE below |
4756 | | * @returns the function returns SystemLiteral and in the second |
4757 | | * case publicID receives PubidLiteral, is strict is off |
4758 | | * it is possible to return NULL and have publicID set. |
4759 | | */ |
4760 | | |
4761 | | xmlChar * |
4762 | 16.1k | xmlParseExternalID(xmlParserCtxt *ctxt, xmlChar **publicId, int strict) { |
4763 | 16.1k | xmlChar *URI = NULL; |
4764 | | |
4765 | 16.1k | *publicId = NULL; |
4766 | 16.1k | if (CMP6(CUR_PTR, 'S', 'Y', 'S', 'T', 'E', 'M')) { |
4767 | 2.91k | SKIP(6); |
4768 | 2.91k | if (SKIP_BLANKS == 0) { |
4769 | 1.09k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
4770 | 1.09k | "Space required after 'SYSTEM'\n"); |
4771 | 1.09k | } |
4772 | 2.91k | URI = xmlParseSystemLiteral(ctxt); |
4773 | 2.91k | if (URI == NULL) { |
4774 | 597 | xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); |
4775 | 597 | } |
4776 | 13.2k | } else if (CMP6(CUR_PTR, 'P', 'U', 'B', 'L', 'I', 'C')) { |
4777 | 5.93k | SKIP(6); |
4778 | 5.93k | if (SKIP_BLANKS == 0) { |
4779 | 4.19k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
4780 | 4.19k | "Space required after 'PUBLIC'\n"); |
4781 | 4.19k | } |
4782 | 5.93k | *publicId = xmlParsePubidLiteral(ctxt); |
4783 | 5.93k | if (*publicId == NULL) { |
4784 | 3.13k | xmlFatalErr(ctxt, XML_ERR_PUBID_REQUIRED, NULL); |
4785 | 3.13k | } |
4786 | 5.93k | if (strict) { |
4787 | | /* |
4788 | | * We don't handle [83] so "S SystemLiteral" is required. |
4789 | | */ |
4790 | 712 | if (SKIP_BLANKS == 0) { |
4791 | 273 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
4792 | 273 | "Space required after the Public Identifier\n"); |
4793 | 273 | } |
4794 | 5.22k | } else { |
4795 | | /* |
4796 | | * We handle [83] so we return immediately, if |
4797 | | * "S SystemLiteral" is not detected. We skip blanks if no |
4798 | | * system literal was found, but this is harmless since we must |
4799 | | * be at the end of a NotationDecl. |
4800 | | */ |
4801 | 5.22k | if (SKIP_BLANKS == 0) return(NULL); |
4802 | 1.37k | if ((CUR != '\'') && (CUR != '"')) return(NULL); |
4803 | 1.37k | } |
4804 | 971 | URI = xmlParseSystemLiteral(ctxt); |
4805 | 971 | if (URI == NULL) { |
4806 | 280 | xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); |
4807 | 280 | } |
4808 | 971 | } |
4809 | 11.1k | return(URI); |
4810 | 16.1k | } |
4811 | | |
4812 | | /** |
4813 | | * Skip an XML (SGML) comment <!-- .... --> |
4814 | | * The spec says that "For compatibility, the string "--" (double-hyphen) |
4815 | | * must not occur within comments. " |
4816 | | * This is the slow routine in case the accelerator for ascii didn't work |
4817 | | * |
4818 | | * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' |
4819 | | * @param ctxt an XML parser context |
4820 | | * @param buf the already parsed part of the buffer |
4821 | | * @param len number of bytes in the buffer |
4822 | | * @param size allocated size of the buffer |
4823 | | */ |
4824 | | static void |
4825 | | xmlParseCommentComplex(xmlParserCtxtPtr ctxt, xmlChar *buf, |
4826 | 36.8k | size_t len, size_t size) { |
4827 | 36.8k | int q, ql; |
4828 | 36.8k | int r, rl; |
4829 | 36.8k | int cur, l; |
4830 | 36.8k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4831 | 36.8k | XML_MAX_HUGE_LENGTH : |
4832 | 36.8k | XML_MAX_TEXT_LENGTH; |
4833 | | |
4834 | 36.8k | if (buf == NULL) { |
4835 | 12.9k | len = 0; |
4836 | 12.9k | size = XML_PARSER_BUFFER_SIZE; |
4837 | 12.9k | buf = xmlMalloc(size); |
4838 | 12.9k | if (buf == NULL) { |
4839 | 0 | xmlErrMemory(ctxt); |
4840 | 0 | return; |
4841 | 0 | } |
4842 | 12.9k | } |
4843 | 36.8k | q = xmlCurrentCharRecover(ctxt, &ql); |
4844 | 36.8k | if (q == 0) |
4845 | 1.51k | goto not_terminated; |
4846 | 35.3k | if (!IS_CHAR(q)) { |
4847 | 10.7k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4848 | 10.7k | "xmlParseComment: invalid xmlChar value %d\n", |
4849 | 10.7k | q); |
4850 | 10.7k | xmlFree (buf); |
4851 | 10.7k | return; |
4852 | 10.7k | } |
4853 | 24.6k | NEXTL(ql); |
4854 | 24.6k | r = xmlCurrentCharRecover(ctxt, &rl); |
4855 | 24.6k | if (r == 0) |
4856 | 1.54k | goto not_terminated; |
4857 | 23.0k | if (!IS_CHAR(r)) { |
4858 | 2.13k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4859 | 2.13k | "xmlParseComment: invalid xmlChar value %d\n", |
4860 | 2.13k | r); |
4861 | 2.13k | xmlFree (buf); |
4862 | 2.13k | return; |
4863 | 2.13k | } |
4864 | 20.9k | NEXTL(rl); |
4865 | 20.9k | cur = xmlCurrentCharRecover(ctxt, &l); |
4866 | 20.9k | if (cur == 0) |
4867 | 892 | goto not_terminated; |
4868 | 1.84M | while (IS_CHAR(cur) && /* checked */ |
4869 | 1.84M | ((cur != '>') || |
4870 | 1.82M | (r != '-') || (q != '-'))) { |
4871 | 1.82M | if ((r == '-') && (q == '-')) { |
4872 | 244k | xmlFatalErr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, NULL); |
4873 | 244k | } |
4874 | 1.82M | if (len + 5 >= size) { |
4875 | 4.19k | xmlChar *tmp; |
4876 | 4.19k | int newSize; |
4877 | | |
4878 | 4.19k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
4879 | 4.19k | if (newSize < 0) { |
4880 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4881 | 0 | "Comment too big found", NULL); |
4882 | 0 | xmlFree (buf); |
4883 | 0 | return; |
4884 | 0 | } |
4885 | 4.19k | tmp = xmlRealloc(buf, newSize); |
4886 | 4.19k | if (tmp == NULL) { |
4887 | 0 | xmlErrMemory(ctxt); |
4888 | 0 | xmlFree(buf); |
4889 | 0 | return; |
4890 | 0 | } |
4891 | 4.19k | buf = tmp; |
4892 | 4.19k | size = newSize; |
4893 | 4.19k | } |
4894 | 1.82M | COPY_BUF(buf, len, q); |
4895 | | |
4896 | 1.82M | q = r; |
4897 | 1.82M | ql = rl; |
4898 | 1.82M | r = cur; |
4899 | 1.82M | rl = l; |
4900 | | |
4901 | 1.82M | NEXTL(l); |
4902 | 1.82M | cur = xmlCurrentCharRecover(ctxt, &l); |
4903 | | |
4904 | 1.82M | } |
4905 | 20.0k | buf[len] = 0; |
4906 | 20.0k | if (cur == 0) { |
4907 | 4.90k | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4908 | 4.90k | "Comment not terminated \n<!--%.50s\n", buf); |
4909 | 15.1k | } else if (!IS_CHAR(cur)) { |
4910 | 12.0k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4911 | 12.0k | "xmlParseComment: invalid xmlChar value %d\n", |
4912 | 12.0k | cur); |
4913 | 12.0k | } else { |
4914 | 3.06k | NEXT; |
4915 | 3.06k | if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && |
4916 | 3.06k | (!ctxt->disableSAX)) |
4917 | 1.35k | ctxt->sax->comment(ctxt->userData, buf); |
4918 | 3.06k | } |
4919 | 20.0k | xmlFree(buf); |
4920 | 20.0k | return; |
4921 | 3.94k | not_terminated: |
4922 | 3.94k | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4923 | 3.94k | "Comment not terminated\n", NULL); |
4924 | 3.94k | xmlFree(buf); |
4925 | 3.94k | } |
4926 | | |
4927 | | /** |
4928 | | * Parse an XML (SGML) comment. Always consumes '<!'. |
4929 | | * |
4930 | | * @deprecated Internal function, don't use. |
4931 | | * |
4932 | | * The spec says that "For compatibility, the string "--" (double-hyphen) |
4933 | | * must not occur within comments. " |
4934 | | * |
4935 | | * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' |
4936 | | * @param ctxt an XML parser context |
4937 | | */ |
4938 | | void |
4939 | 46.1k | xmlParseComment(xmlParserCtxt *ctxt) { |
4940 | 46.1k | xmlChar *buf = NULL; |
4941 | 46.1k | size_t size = XML_PARSER_BUFFER_SIZE; |
4942 | 46.1k | size_t len = 0; |
4943 | 46.1k | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4944 | 46.1k | XML_MAX_HUGE_LENGTH : |
4945 | 46.1k | XML_MAX_TEXT_LENGTH; |
4946 | 46.1k | const xmlChar *in; |
4947 | 46.1k | size_t nbchar = 0; |
4948 | 46.1k | int ccol; |
4949 | | |
4950 | | /* |
4951 | | * Check that there is a comment right here. |
4952 | | */ |
4953 | 46.1k | if ((RAW != '<') || (NXT(1) != '!')) |
4954 | 0 | return; |
4955 | 46.1k | SKIP(2); |
4956 | 46.1k | if ((RAW != '-') || (NXT(1) != '-')) |
4957 | 3 | return; |
4958 | 46.1k | SKIP(2); |
4959 | 46.1k | GROW; |
4960 | | |
4961 | | /* |
4962 | | * Accelerated common case where input don't need to be |
4963 | | * modified before passing it to the handler. |
4964 | | */ |
4965 | 46.1k | in = ctxt->input->cur; |
4966 | 46.5k | do { |
4967 | 46.5k | if (*in == 0xA) { |
4968 | 10.4k | do { |
4969 | 10.4k | ctxt->input->line++; ctxt->input->col = 1; |
4970 | 10.4k | in++; |
4971 | 10.4k | } while (*in == 0xA); |
4972 | 1.19k | } |
4973 | 311k | get_more: |
4974 | 311k | ccol = ctxt->input->col; |
4975 | 2.88M | while (((*in > '-') && (*in <= 0x7F)) || |
4976 | 2.88M | ((*in >= 0x20) && (*in < '-')) || |
4977 | 2.88M | (*in == 0x09)) { |
4978 | 2.57M | in++; |
4979 | 2.57M | ccol++; |
4980 | 2.57M | } |
4981 | 311k | ctxt->input->col = ccol; |
4982 | 311k | if (*in == 0xA) { |
4983 | 78.3k | do { |
4984 | 78.3k | ctxt->input->line++; ctxt->input->col = 1; |
4985 | 78.3k | in++; |
4986 | 78.3k | } while (*in == 0xA); |
4987 | 11.8k | goto get_more; |
4988 | 11.8k | } |
4989 | 299k | nbchar = in - ctxt->input->cur; |
4990 | | /* |
4991 | | * save current set of data |
4992 | | */ |
4993 | 299k | if (nbchar > 0) { |
4994 | 268k | if (nbchar > maxLength - len) { |
4995 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4996 | 0 | "Comment too big found", NULL); |
4997 | 0 | xmlFree(buf); |
4998 | 0 | return; |
4999 | 0 | } |
5000 | 268k | if (buf == NULL) { |
5001 | 27.9k | if ((*in == '-') && (in[1] == '-')) |
5002 | 6.05k | size = nbchar + 1; |
5003 | 21.9k | else |
5004 | 21.9k | size = XML_PARSER_BUFFER_SIZE + nbchar; |
5005 | 27.9k | buf = xmlMalloc(size); |
5006 | 27.9k | if (buf == NULL) { |
5007 | 0 | xmlErrMemory(ctxt); |
5008 | 0 | return; |
5009 | 0 | } |
5010 | 27.9k | len = 0; |
5011 | 240k | } else if (len + nbchar + 1 >= size) { |
5012 | 3.78k | xmlChar *new_buf; |
5013 | 3.78k | size += len + nbchar + XML_PARSER_BUFFER_SIZE; |
5014 | 3.78k | new_buf = xmlRealloc(buf, size); |
5015 | 3.78k | if (new_buf == NULL) { |
5016 | 0 | xmlErrMemory(ctxt); |
5017 | 0 | xmlFree(buf); |
5018 | 0 | return; |
5019 | 0 | } |
5020 | 3.78k | buf = new_buf; |
5021 | 3.78k | } |
5022 | 268k | memcpy(&buf[len], ctxt->input->cur, nbchar); |
5023 | 268k | len += nbchar; |
5024 | 268k | buf[len] = 0; |
5025 | 268k | } |
5026 | 299k | ctxt->input->cur = in; |
5027 | 299k | if (*in == 0xA) { |
5028 | 0 | in++; |
5029 | 0 | ctxt->input->line++; ctxt->input->col = 1; |
5030 | 0 | } |
5031 | 299k | if (*in == 0xD) { |
5032 | 10.7k | in++; |
5033 | 10.7k | if (*in == 0xA) { |
5034 | 5.36k | ctxt->input->cur = in; |
5035 | 5.36k | in++; |
5036 | 5.36k | ctxt->input->line++; ctxt->input->col = 1; |
5037 | 5.36k | goto get_more; |
5038 | 5.36k | } |
5039 | 5.35k | in--; |
5040 | 5.35k | } |
5041 | 293k | SHRINK; |
5042 | 293k | GROW; |
5043 | 293k | in = ctxt->input->cur; |
5044 | 293k | if (*in == '-') { |
5045 | 256k | if (in[1] == '-') { |
5046 | 238k | if (in[2] == '>') { |
5047 | 9.32k | SKIP(3); |
5048 | 9.32k | if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && |
5049 | 9.32k | (!ctxt->disableSAX)) { |
5050 | 865 | if (buf != NULL) |
5051 | 161 | ctxt->sax->comment(ctxt->userData, buf); |
5052 | 704 | else |
5053 | 704 | ctxt->sax->comment(ctxt->userData, BAD_CAST ""); |
5054 | 865 | } |
5055 | 9.32k | if (buf != NULL) |
5056 | 4.10k | xmlFree(buf); |
5057 | 9.32k | return; |
5058 | 9.32k | } |
5059 | 229k | if (buf != NULL) { |
5060 | 227k | xmlFatalErrMsgStr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, |
5061 | 227k | "Double hyphen within comment: " |
5062 | 227k | "<!--%.50s\n", |
5063 | 227k | buf); |
5064 | 227k | } else |
5065 | 1.69k | xmlFatalErrMsgStr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, |
5066 | 1.69k | "Double hyphen within comment\n", NULL); |
5067 | 229k | in++; |
5068 | 229k | ctxt->input->col++; |
5069 | 229k | } |
5070 | 247k | in++; |
5071 | 247k | ctxt->input->col++; |
5072 | 247k | goto get_more; |
5073 | 256k | } |
5074 | 293k | } while (((*in >= 0x20) && (*in <= 0x7F)) || (*in == 0x09) || (*in == 0x0a)); |
5075 | 36.8k | xmlParseCommentComplex(ctxt, buf, len, size); |
5076 | 36.8k | } |
5077 | | |
5078 | | |
5079 | | /** |
5080 | | * parse the name of a PI |
5081 | | * |
5082 | | * @deprecated Internal function, don't use. |
5083 | | * |
5084 | | * [17] PITarget ::= Name - (('X' | 'x') ('M' | 'm') ('L' | 'l')) |
5085 | | * |
5086 | | * @param ctxt an XML parser context |
5087 | | * @returns the PITarget name or NULL |
5088 | | */ |
5089 | | |
5090 | | const xmlChar * |
5091 | 52.5k | xmlParsePITarget(xmlParserCtxt *ctxt) { |
5092 | 52.5k | const xmlChar *name; |
5093 | | |
5094 | 52.5k | name = xmlParseName(ctxt); |
5095 | 52.5k | if ((name != NULL) && |
5096 | 52.5k | ((name[0] == 'x') || (name[0] == 'X')) && |
5097 | 52.5k | ((name[1] == 'm') || (name[1] == 'M')) && |
5098 | 52.5k | ((name[2] == 'l') || (name[2] == 'L'))) { |
5099 | 8.91k | int i; |
5100 | 8.91k | if ((name[0] == 'x') && (name[1] == 'm') && |
5101 | 8.91k | (name[2] == 'l') && (name[3] == 0)) { |
5102 | 4.75k | xmlFatalErrMsg(ctxt, XML_ERR_RESERVED_XML_NAME, |
5103 | 4.75k | "XML declaration allowed only at the start of the document\n"); |
5104 | 4.75k | return(name); |
5105 | 4.75k | } else if (name[3] == 0) { |
5106 | 490 | xmlFatalErr(ctxt, XML_ERR_RESERVED_XML_NAME, NULL); |
5107 | 490 | return(name); |
5108 | 490 | } |
5109 | 10.2k | for (i = 0;;i++) { |
5110 | 10.2k | if (xmlW3CPIs[i] == NULL) break; |
5111 | 7.11k | if (xmlStrEqual(name, (const xmlChar *)xmlW3CPIs[i])) |
5112 | 512 | return(name); |
5113 | 7.11k | } |
5114 | 3.15k | xmlWarningMsg(ctxt, XML_ERR_RESERVED_XML_NAME, |
5115 | 3.15k | "xmlParsePITarget: invalid name prefix 'xml'\n", |
5116 | 3.15k | NULL, NULL); |
5117 | 3.15k | } |
5118 | 46.8k | if ((name != NULL) && (xmlStrchr(name, ':') != NULL)) { |
5119 | 1.97k | xmlNsErr(ctxt, XML_NS_ERR_COLON, |
5120 | 1.97k | "colons are forbidden from PI names '%s'\n", name, NULL, NULL); |
5121 | 1.97k | } |
5122 | 46.8k | return(name); |
5123 | 52.5k | } |
5124 | | |
5125 | | #ifdef LIBXML_CATALOG_ENABLED |
5126 | | /** |
5127 | | * parse an XML Catalog Processing Instruction. |
5128 | | * |
5129 | | * <?oasis-xml-catalog catalog="http://example.com/catalog.xml"?> |
5130 | | * |
5131 | | * Occurs only if allowed by the user and if happening in the Misc |
5132 | | * part of the document before any doctype information |
5133 | | * This will add the given catalog to the parsing context in order |
5134 | | * to be used if there is a resolution need further down in the document |
5135 | | * |
5136 | | * @param ctxt an XML parser context |
5137 | | * @param catalog the PI value string |
5138 | | */ |
5139 | | |
5140 | | static void |
5141 | 0 | xmlParseCatalogPI(xmlParserCtxtPtr ctxt, const xmlChar *catalog) { |
5142 | 0 | xmlChar *URL = NULL; |
5143 | 0 | const xmlChar *tmp, *base; |
5144 | 0 | xmlChar marker; |
5145 | |
|
5146 | 0 | tmp = catalog; |
5147 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5148 | 0 | if (xmlStrncmp(tmp, BAD_CAST"catalog", 7)) |
5149 | 0 | goto error; |
5150 | 0 | tmp += 7; |
5151 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5152 | 0 | if (*tmp != '=') { |
5153 | 0 | return; |
5154 | 0 | } |
5155 | 0 | tmp++; |
5156 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5157 | 0 | marker = *tmp; |
5158 | 0 | if ((marker != '\'') && (marker != '"')) |
5159 | 0 | goto error; |
5160 | 0 | tmp++; |
5161 | 0 | base = tmp; |
5162 | 0 | while ((*tmp != 0) && (*tmp != marker)) tmp++; |
5163 | 0 | if (*tmp == 0) |
5164 | 0 | goto error; |
5165 | 0 | URL = xmlStrndup(base, tmp - base); |
5166 | 0 | tmp++; |
5167 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5168 | 0 | if (*tmp != 0) |
5169 | 0 | goto error; |
5170 | | |
5171 | 0 | if (URL != NULL) { |
5172 | | /* |
5173 | | * Unfortunately, the catalog API doesn't report OOM errors. |
5174 | | * xmlGetLastError isn't very helpful since we don't know |
5175 | | * where the last error came from. We'd have to reset it |
5176 | | * before this call and restore it afterwards. |
5177 | | */ |
5178 | 0 | ctxt->catalogs = xmlCatalogAddLocal(ctxt->catalogs, URL); |
5179 | 0 | xmlFree(URL); |
5180 | 0 | } |
5181 | 0 | return; |
5182 | | |
5183 | 0 | error: |
5184 | 0 | xmlWarningMsg(ctxt, XML_WAR_CATALOG_PI, |
5185 | 0 | "Catalog PI syntax error: %s\n", |
5186 | 0 | catalog, NULL); |
5187 | 0 | if (URL != NULL) |
5188 | 0 | xmlFree(URL); |
5189 | 0 | } |
5190 | | #endif |
5191 | | |
5192 | | /** |
5193 | | * parse an XML Processing Instruction. |
5194 | | * |
5195 | | * @deprecated Internal function, don't use. |
5196 | | * |
5197 | | * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>' |
5198 | | * |
5199 | | * The processing is transferred to SAX once parsed. |
5200 | | * |
5201 | | * @param ctxt an XML parser context |
5202 | | */ |
5203 | | |
5204 | | void |
5205 | 52.5k | xmlParsePI(xmlParserCtxt *ctxt) { |
5206 | 52.5k | xmlChar *buf = NULL; |
5207 | 52.5k | size_t len = 0; |
5208 | 52.5k | size_t size = XML_PARSER_BUFFER_SIZE; |
5209 | 52.5k | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
5210 | 52.5k | XML_MAX_HUGE_LENGTH : |
5211 | 52.5k | XML_MAX_TEXT_LENGTH; |
5212 | 52.5k | int cur, l; |
5213 | 52.5k | const xmlChar *target; |
5214 | | |
5215 | 52.5k | if ((RAW == '<') && (NXT(1) == '?')) { |
5216 | | /* |
5217 | | * this is a Processing Instruction. |
5218 | | */ |
5219 | 52.5k | SKIP(2); |
5220 | | |
5221 | | /* |
5222 | | * Parse the target name and check for special support like |
5223 | | * namespace. |
5224 | | */ |
5225 | 52.5k | target = xmlParsePITarget(ctxt); |
5226 | 52.5k | if (target != NULL) { |
5227 | 43.5k | if ((RAW == '?') && (NXT(1) == '>')) { |
5228 | 9.04k | SKIP(2); |
5229 | | |
5230 | | /* |
5231 | | * SAX: PI detected. |
5232 | | */ |
5233 | 9.04k | if ((ctxt->sax) && (!ctxt->disableSAX) && |
5234 | 9.04k | (ctxt->sax->processingInstruction != NULL)) |
5235 | 2.61k | ctxt->sax->processingInstruction(ctxt->userData, |
5236 | 2.61k | target, NULL); |
5237 | 9.04k | return; |
5238 | 9.04k | } |
5239 | 34.4k | buf = xmlMalloc(size); |
5240 | 34.4k | if (buf == NULL) { |
5241 | 0 | xmlErrMemory(ctxt); |
5242 | 0 | return; |
5243 | 0 | } |
5244 | 34.4k | if (SKIP_BLANKS == 0) { |
5245 | 23.8k | xmlFatalErrMsgStr(ctxt, XML_ERR_SPACE_REQUIRED, |
5246 | 23.8k | "ParsePI: PI %s space expected\n", target); |
5247 | 23.8k | } |
5248 | 34.4k | cur = xmlCurrentCharRecover(ctxt, &l); |
5249 | 4.86M | while (IS_CHAR(cur) && /* checked */ |
5250 | 4.86M | ((cur != '?') || (NXT(1) != '>'))) { |
5251 | 4.83M | if (len + 5 >= size) { |
5252 | 5.88k | xmlChar *tmp; |
5253 | 5.88k | int newSize; |
5254 | | |
5255 | 5.88k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
5256 | 5.88k | if (newSize < 0) { |
5257 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_PI_NOT_FINISHED, |
5258 | 0 | "PI %s too big found", target); |
5259 | 0 | xmlFree(buf); |
5260 | 0 | return; |
5261 | 0 | } |
5262 | 5.88k | tmp = xmlRealloc(buf, newSize); |
5263 | 5.88k | if (tmp == NULL) { |
5264 | 0 | xmlErrMemory(ctxt); |
5265 | 0 | xmlFree(buf); |
5266 | 0 | return; |
5267 | 0 | } |
5268 | 5.88k | buf = tmp; |
5269 | 5.88k | size = newSize; |
5270 | 5.88k | } |
5271 | 4.83M | COPY_BUF(buf, len, cur); |
5272 | 4.83M | NEXTL(l); |
5273 | 4.83M | cur = xmlCurrentCharRecover(ctxt, &l); |
5274 | 4.83M | } |
5275 | 34.4k | buf[len] = 0; |
5276 | 34.4k | if (cur != '?') { |
5277 | 16.2k | xmlFatalErrMsgStr(ctxt, XML_ERR_PI_NOT_FINISHED, |
5278 | 16.2k | "ParsePI: PI %s never end ...\n", target); |
5279 | 18.2k | } else { |
5280 | 18.2k | SKIP(2); |
5281 | | |
5282 | 18.2k | #ifdef LIBXML_CATALOG_ENABLED |
5283 | 18.2k | if ((ctxt->inSubset == 0) && |
5284 | 18.2k | (xmlStrEqual(target, XML_CATALOG_PI))) { |
5285 | 2.78k | xmlCatalogAllow allow = xmlCatalogGetDefaults(); |
5286 | | |
5287 | 2.78k | if ((ctxt->options & XML_PARSE_CATALOG_PI) && |
5288 | 2.78k | ((allow == XML_CATA_ALLOW_DOCUMENT) || |
5289 | 0 | (allow == XML_CATA_ALLOW_ALL))) |
5290 | 0 | xmlParseCatalogPI(ctxt, buf); |
5291 | 2.78k | } |
5292 | 18.2k | #endif |
5293 | | |
5294 | | /* |
5295 | | * SAX: PI detected. |
5296 | | */ |
5297 | 18.2k | if ((ctxt->sax) && (!ctxt->disableSAX) && |
5298 | 18.2k | (ctxt->sax->processingInstruction != NULL)) |
5299 | 443 | ctxt->sax->processingInstruction(ctxt->userData, |
5300 | 443 | target, buf); |
5301 | 18.2k | } |
5302 | 34.4k | xmlFree(buf); |
5303 | 34.4k | } else { |
5304 | 9.05k | xmlFatalErr(ctxt, XML_ERR_PI_NOT_STARTED, NULL); |
5305 | 9.05k | } |
5306 | 52.5k | } |
5307 | 52.5k | } |
5308 | | |
5309 | | /** |
5310 | | * Parse a notation declaration. Always consumes '<!'. |
5311 | | * |
5312 | | * @deprecated Internal function, don't use. |
5313 | | * |
5314 | | * [82] NotationDecl ::= '<!NOTATION' S Name S (ExternalID | PublicID) |
5315 | | * S? '>' |
5316 | | * |
5317 | | * Hence there is actually 3 choices: |
5318 | | * |
5319 | | * 'PUBLIC' S PubidLiteral |
5320 | | * 'PUBLIC' S PubidLiteral S SystemLiteral |
5321 | | * 'SYSTEM' S SystemLiteral |
5322 | | * |
5323 | | * See the NOTE on #xmlParseExternalID. |
5324 | | * |
5325 | | * @param ctxt an XML parser context |
5326 | | */ |
5327 | | |
5328 | | void |
5329 | 8.84k | xmlParseNotationDecl(xmlParserCtxt *ctxt) { |
5330 | 8.84k | const xmlChar *name; |
5331 | 8.84k | xmlChar *Pubid; |
5332 | 8.84k | xmlChar *Systemid; |
5333 | | |
5334 | 8.84k | if ((CUR != '<') || (NXT(1) != '!')) |
5335 | 0 | return; |
5336 | 8.84k | SKIP(2); |
5337 | | |
5338 | 8.84k | if (CMP8(CUR_PTR, 'N', 'O', 'T', 'A', 'T', 'I', 'O', 'N')) { |
5339 | 8.79k | #ifdef LIBXML_VALID_ENABLED |
5340 | 8.79k | int oldInputNr = ctxt->inputNr; |
5341 | 8.79k | #endif |
5342 | | |
5343 | 8.79k | SKIP(8); |
5344 | 8.79k | if (SKIP_BLANKS_PE == 0) { |
5345 | 645 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5346 | 645 | "Space required after '<!NOTATION'\n"); |
5347 | 645 | return; |
5348 | 645 | } |
5349 | | |
5350 | 8.14k | name = xmlParseName(ctxt); |
5351 | 8.14k | if (name == NULL) { |
5352 | 248 | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_STARTED, NULL); |
5353 | 248 | return; |
5354 | 248 | } |
5355 | 7.90k | if (xmlStrchr(name, ':') != NULL) { |
5356 | 2.82k | xmlNsErr(ctxt, XML_NS_ERR_COLON, |
5357 | 2.82k | "colons are forbidden from notation names '%s'\n", |
5358 | 2.82k | name, NULL, NULL); |
5359 | 2.82k | } |
5360 | 7.90k | if (SKIP_BLANKS_PE == 0) { |
5361 | 1.13k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5362 | 1.13k | "Space required after the NOTATION name'\n"); |
5363 | 1.13k | return; |
5364 | 1.13k | } |
5365 | | |
5366 | | /* |
5367 | | * Parse the IDs. |
5368 | | */ |
5369 | 6.76k | Systemid = xmlParseExternalID(ctxt, &Pubid, 0); |
5370 | 6.76k | SKIP_BLANKS_PE; |
5371 | | |
5372 | 6.76k | if (RAW == '>') { |
5373 | 807 | #ifdef LIBXML_VALID_ENABLED |
5374 | 807 | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
5375 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
5376 | 0 | "Notation declaration doesn't start and stop" |
5377 | 0 | " in the same entity\n", |
5378 | 0 | NULL, NULL); |
5379 | 0 | } |
5380 | 807 | #endif |
5381 | 807 | NEXT; |
5382 | 807 | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
5383 | 807 | (ctxt->sax->notationDecl != NULL)) |
5384 | 303 | ctxt->sax->notationDecl(ctxt->userData, name, Pubid, Systemid); |
5385 | 5.95k | } else { |
5386 | 5.95k | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_FINISHED, NULL); |
5387 | 5.95k | } |
5388 | 6.76k | if (Systemid != NULL) xmlFree(Systemid); |
5389 | 6.76k | if (Pubid != NULL) xmlFree(Pubid); |
5390 | 6.76k | } |
5391 | 8.84k | } |
5392 | | |
5393 | | /** |
5394 | | * Parse an entity declaration. Always consumes '<!'. |
5395 | | * |
5396 | | * @deprecated Internal function, don't use. |
5397 | | * |
5398 | | * [70] EntityDecl ::= GEDecl | PEDecl |
5399 | | * |
5400 | | * [71] GEDecl ::= '<!ENTITY' S Name S EntityDef S? '>' |
5401 | | * |
5402 | | * [72] PEDecl ::= '<!ENTITY' S '%' S Name S PEDef S? '>' |
5403 | | * |
5404 | | * [73] EntityDef ::= EntityValue | (ExternalID NDataDecl?) |
5405 | | * |
5406 | | * [74] PEDef ::= EntityValue | ExternalID |
5407 | | * |
5408 | | * [76] NDataDecl ::= S 'NDATA' S Name |
5409 | | * |
5410 | | * [ VC: Notation Declared ] |
5411 | | * The Name must match the declared name of a notation. |
5412 | | * |
5413 | | * @param ctxt an XML parser context |
5414 | | */ |
5415 | | |
5416 | | void |
5417 | 24.8k | xmlParseEntityDecl(xmlParserCtxt *ctxt) { |
5418 | 24.8k | const xmlChar *name = NULL; |
5419 | 24.8k | xmlChar *value = NULL; |
5420 | 24.8k | xmlChar *URI = NULL, *literal = NULL; |
5421 | 24.8k | const xmlChar *ndata = NULL; |
5422 | 24.8k | int isParameter = 0; |
5423 | 24.8k | xmlChar *orig = NULL; |
5424 | | |
5425 | 24.8k | if ((CUR != '<') || (NXT(1) != '!')) |
5426 | 0 | return; |
5427 | 24.8k | SKIP(2); |
5428 | | |
5429 | | /* GROW; done in the caller */ |
5430 | 24.8k | if (CMP6(CUR_PTR, 'E', 'N', 'T', 'I', 'T', 'Y')) { |
5431 | 24.7k | #ifdef LIBXML_VALID_ENABLED |
5432 | 24.7k | int oldInputNr = ctxt->inputNr; |
5433 | 24.7k | #endif |
5434 | | |
5435 | 24.7k | SKIP(6); |
5436 | 24.7k | if (SKIP_BLANKS_PE == 0) { |
5437 | 13.4k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5438 | 13.4k | "Space required after '<!ENTITY'\n"); |
5439 | 13.4k | } |
5440 | | |
5441 | 24.7k | if (RAW == '%') { |
5442 | 4.27k | NEXT; |
5443 | 4.27k | if (SKIP_BLANKS_PE == 0) { |
5444 | 297 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5445 | 297 | "Space required after '%%'\n"); |
5446 | 297 | } |
5447 | 4.27k | isParameter = 1; |
5448 | 4.27k | } |
5449 | | |
5450 | 24.7k | name = xmlParseName(ctxt); |
5451 | 24.7k | if (name == NULL) { |
5452 | 427 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
5453 | 427 | "xmlParseEntityDecl: no name\n"); |
5454 | 427 | return; |
5455 | 427 | } |
5456 | 24.3k | if (xmlStrchr(name, ':') != NULL) { |
5457 | 850 | xmlNsErr(ctxt, XML_NS_ERR_COLON, |
5458 | 850 | "colons are forbidden from entities names '%s'\n", |
5459 | 850 | name, NULL, NULL); |
5460 | 850 | } |
5461 | 24.3k | if (SKIP_BLANKS_PE == 0) { |
5462 | 10.2k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5463 | 10.2k | "Space required after the entity name\n"); |
5464 | 10.2k | } |
5465 | | |
5466 | | /* |
5467 | | * handle the various case of definitions... |
5468 | | */ |
5469 | 24.3k | if (isParameter) { |
5470 | 4.22k | if ((RAW == '"') || (RAW == '\'')) { |
5471 | 3.04k | value = xmlParseEntityValue(ctxt, &orig); |
5472 | 3.04k | if (value) { |
5473 | 3.03k | if ((ctxt->sax != NULL) && |
5474 | 3.03k | (!ctxt->disableSAX) && (ctxt->sax->entityDecl != NULL)) |
5475 | 1.46k | ctxt->sax->entityDecl(ctxt->userData, name, |
5476 | 1.46k | XML_INTERNAL_PARAMETER_ENTITY, |
5477 | 1.46k | NULL, NULL, value); |
5478 | 3.03k | } |
5479 | 3.04k | } else { |
5480 | 1.18k | URI = xmlParseExternalID(ctxt, &literal, 1); |
5481 | 1.18k | if ((URI == NULL) && (literal == NULL)) { |
5482 | 209 | xmlFatalErr(ctxt, XML_ERR_VALUE_REQUIRED, NULL); |
5483 | 209 | } |
5484 | 1.18k | if (URI) { |
5485 | 899 | if (xmlStrchr(URI, '#')) { |
5486 | 87 | xmlFatalErr(ctxt, XML_ERR_URI_FRAGMENT, NULL); |
5487 | 812 | } else { |
5488 | 812 | if ((ctxt->sax != NULL) && |
5489 | 812 | (!ctxt->disableSAX) && |
5490 | 812 | (ctxt->sax->entityDecl != NULL)) |
5491 | 438 | ctxt->sax->entityDecl(ctxt->userData, name, |
5492 | 438 | XML_EXTERNAL_PARAMETER_ENTITY, |
5493 | 438 | literal, URI, NULL); |
5494 | 812 | } |
5495 | 899 | } |
5496 | 1.18k | } |
5497 | 20.1k | } else { |
5498 | 20.1k | if ((RAW == '"') || (RAW == '\'')) { |
5499 | 17.8k | value = xmlParseEntityValue(ctxt, &orig); |
5500 | 17.8k | if ((ctxt->sax != NULL) && |
5501 | 17.8k | (!ctxt->disableSAX) && (ctxt->sax->entityDecl != NULL)) |
5502 | 939 | ctxt->sax->entityDecl(ctxt->userData, name, |
5503 | 939 | XML_INTERNAL_GENERAL_ENTITY, |
5504 | 939 | NULL, NULL, value); |
5505 | | /* |
5506 | | * For expat compatibility in SAX mode. |
5507 | | */ |
5508 | 17.8k | if ((ctxt->myDoc == NULL) || |
5509 | 17.8k | (xmlStrEqual(ctxt->myDoc->version, SAX_COMPAT_MODE))) { |
5510 | 9.63k | if (ctxt->myDoc == NULL) { |
5511 | 1.09k | ctxt->myDoc = xmlNewDoc(SAX_COMPAT_MODE); |
5512 | 1.09k | if (ctxt->myDoc == NULL) { |
5513 | 0 | xmlErrMemory(ctxt); |
5514 | 0 | goto done; |
5515 | 0 | } |
5516 | 1.09k | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
5517 | 1.09k | } |
5518 | 9.63k | if (ctxt->myDoc->intSubset == NULL) { |
5519 | 1.09k | ctxt->myDoc->intSubset = xmlNewDtd(ctxt->myDoc, |
5520 | 1.09k | BAD_CAST "fake", NULL, NULL); |
5521 | 1.09k | if (ctxt->myDoc->intSubset == NULL) { |
5522 | 0 | xmlErrMemory(ctxt); |
5523 | 0 | goto done; |
5524 | 0 | } |
5525 | 1.09k | } |
5526 | | |
5527 | 9.63k | xmlSAX2EntityDecl(ctxt, name, XML_INTERNAL_GENERAL_ENTITY, |
5528 | 9.63k | NULL, NULL, value); |
5529 | 9.63k | } |
5530 | 17.8k | } else { |
5531 | 2.26k | URI = xmlParseExternalID(ctxt, &literal, 1); |
5532 | 2.26k | if ((URI == NULL) && (literal == NULL)) { |
5533 | 991 | xmlFatalErr(ctxt, XML_ERR_VALUE_REQUIRED, NULL); |
5534 | 991 | } |
5535 | 2.26k | if (URI) { |
5536 | 1.10k | if (xmlStrchr(URI, '#')) { |
5537 | 409 | xmlFatalErr(ctxt, XML_ERR_URI_FRAGMENT, NULL); |
5538 | 409 | } |
5539 | 1.10k | } |
5540 | 2.26k | if ((RAW != '>') && (SKIP_BLANKS_PE == 0)) { |
5541 | 433 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5542 | 433 | "Space required before 'NDATA'\n"); |
5543 | 433 | } |
5544 | 2.26k | if (CMP5(CUR_PTR, 'N', 'D', 'A', 'T', 'A')) { |
5545 | 355 | SKIP(5); |
5546 | 355 | if (SKIP_BLANKS_PE == 0) { |
5547 | 199 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5548 | 199 | "Space required after 'NDATA'\n"); |
5549 | 199 | } |
5550 | 355 | ndata = xmlParseName(ctxt); |
5551 | 355 | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
5552 | 355 | (ctxt->sax->unparsedEntityDecl != NULL)) |
5553 | 23 | ctxt->sax->unparsedEntityDecl(ctxt->userData, name, |
5554 | 23 | literal, URI, ndata); |
5555 | 1.90k | } else { |
5556 | 1.90k | if ((ctxt->sax != NULL) && |
5557 | 1.90k | (!ctxt->disableSAX) && (ctxt->sax->entityDecl != NULL)) |
5558 | 78 | ctxt->sax->entityDecl(ctxt->userData, name, |
5559 | 78 | XML_EXTERNAL_GENERAL_PARSED_ENTITY, |
5560 | 78 | literal, URI, NULL); |
5561 | | /* |
5562 | | * For expat compatibility in SAX mode. |
5563 | | * assuming the entity replacement was asked for |
5564 | | */ |
5565 | 1.90k | if ((ctxt->replaceEntities != 0) && |
5566 | 1.90k | ((ctxt->myDoc == NULL) || |
5567 | 1.90k | (xmlStrEqual(ctxt->myDoc->version, SAX_COMPAT_MODE)))) { |
5568 | 757 | if (ctxt->myDoc == NULL) { |
5569 | 73 | ctxt->myDoc = xmlNewDoc(SAX_COMPAT_MODE); |
5570 | 73 | if (ctxt->myDoc == NULL) { |
5571 | 0 | xmlErrMemory(ctxt); |
5572 | 0 | goto done; |
5573 | 0 | } |
5574 | 73 | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
5575 | 73 | } |
5576 | | |
5577 | 757 | if (ctxt->myDoc->intSubset == NULL) { |
5578 | 73 | ctxt->myDoc->intSubset = xmlNewDtd(ctxt->myDoc, |
5579 | 73 | BAD_CAST "fake", NULL, NULL); |
5580 | 73 | if (ctxt->myDoc->intSubset == NULL) { |
5581 | 0 | xmlErrMemory(ctxt); |
5582 | 0 | goto done; |
5583 | 0 | } |
5584 | 73 | } |
5585 | 757 | xmlSAX2EntityDecl(ctxt, name, |
5586 | 757 | XML_EXTERNAL_GENERAL_PARSED_ENTITY, |
5587 | 757 | literal, URI, NULL); |
5588 | 757 | } |
5589 | 1.90k | } |
5590 | 2.26k | } |
5591 | 20.1k | } |
5592 | 24.3k | SKIP_BLANKS_PE; |
5593 | 24.3k | if (RAW != '>') { |
5594 | 680 | xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_NOT_FINISHED, |
5595 | 680 | "xmlParseEntityDecl: entity %s not terminated\n", name); |
5596 | 680 | xmlHaltParser(ctxt); |
5597 | 23.6k | } else { |
5598 | 23.6k | #ifdef LIBXML_VALID_ENABLED |
5599 | 23.6k | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
5600 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
5601 | 0 | "Entity declaration doesn't start and stop in" |
5602 | 0 | " the same entity\n", |
5603 | 0 | NULL, NULL); |
5604 | 0 | } |
5605 | 23.6k | #endif |
5606 | 23.6k | NEXT; |
5607 | 23.6k | } |
5608 | 24.3k | if (orig != NULL) { |
5609 | | /* |
5610 | | * Ugly mechanism to save the raw entity value. |
5611 | | */ |
5612 | 20.8k | xmlEntityPtr cur = NULL; |
5613 | | |
5614 | 20.8k | if (isParameter) { |
5615 | 3.03k | if ((ctxt->sax != NULL) && |
5616 | 3.03k | (ctxt->sax->getParameterEntity != NULL)) |
5617 | 3.03k | cur = ctxt->sax->getParameterEntity(ctxt->userData, name); |
5618 | 17.8k | } else { |
5619 | 17.8k | if ((ctxt->sax != NULL) && |
5620 | 17.8k | (ctxt->sax->getEntity != NULL)) |
5621 | 17.8k | cur = ctxt->sax->getEntity(ctxt->userData, name); |
5622 | 17.8k | if ((cur == NULL) && (ctxt->userData==ctxt)) { |
5623 | 5.03k | cur = xmlSAX2GetEntity(ctxt, name); |
5624 | 5.03k | } |
5625 | 17.8k | } |
5626 | 20.8k | if ((cur != NULL) && (cur->orig == NULL)) { |
5627 | 3.44k | cur->orig = orig; |
5628 | 3.44k | orig = NULL; |
5629 | 3.44k | } |
5630 | 20.8k | } |
5631 | | |
5632 | 24.3k | done: |
5633 | 24.3k | if (value != NULL) xmlFree(value); |
5634 | 24.3k | if (URI != NULL) xmlFree(URI); |
5635 | 24.3k | if (literal != NULL) xmlFree(literal); |
5636 | 24.3k | if (orig != NULL) xmlFree(orig); |
5637 | 24.3k | } |
5638 | 24.8k | } |
5639 | | |
5640 | | /** |
5641 | | * Parse an attribute default declaration |
5642 | | * |
5643 | | * @deprecated Internal function, don't use. |
5644 | | * |
5645 | | * [60] DefaultDecl ::= '#REQUIRED' | '#IMPLIED' | (('#FIXED' S)? AttValue) |
5646 | | * |
5647 | | * [ VC: Required Attribute ] |
5648 | | * if the default declaration is the keyword \#REQUIRED, then the |
5649 | | * attribute must be specified for all elements of the type in the |
5650 | | * attribute-list declaration. |
5651 | | * |
5652 | | * [ VC: Attribute Default Legal ] |
5653 | | * The declared default value must meet the lexical constraints of |
5654 | | * the declared attribute type c.f. #xmlValidateAttributeDecl |
5655 | | * |
5656 | | * [ VC: Fixed Attribute Default ] |
5657 | | * if an attribute has a default value declared with the \#FIXED |
5658 | | * keyword, instances of that attribute must match the default value. |
5659 | | * |
5660 | | * [ WFC: No < in Attribute Values ] |
5661 | | * handled in #xmlParseAttValue |
5662 | | * |
5663 | | * @param ctxt an XML parser context |
5664 | | * @param value Receive a possible fixed default value for the attribute |
5665 | | * @returns XML_ATTRIBUTE_NONE, XML_ATTRIBUTE_REQUIRED, XML_ATTRIBUTE_IMPLIED |
5666 | | * or XML_ATTRIBUTE_FIXED. |
5667 | | */ |
5668 | | |
5669 | | int |
5670 | 19.1k | xmlParseDefaultDecl(xmlParserCtxt *ctxt, xmlChar **value) { |
5671 | 19.1k | int val; |
5672 | 19.1k | xmlChar *ret; |
5673 | | |
5674 | 19.1k | *value = NULL; |
5675 | 19.1k | if (CMP9(CUR_PTR, '#', 'R', 'E', 'Q', 'U', 'I', 'R', 'E', 'D')) { |
5676 | 109 | SKIP(9); |
5677 | 109 | return(XML_ATTRIBUTE_REQUIRED); |
5678 | 109 | } |
5679 | 19.0k | if (CMP8(CUR_PTR, '#', 'I', 'M', 'P', 'L', 'I', 'E', 'D')) { |
5680 | 504 | SKIP(8); |
5681 | 504 | return(XML_ATTRIBUTE_IMPLIED); |
5682 | 504 | } |
5683 | 18.5k | val = XML_ATTRIBUTE_NONE; |
5684 | 18.5k | if (CMP6(CUR_PTR, '#', 'F', 'I', 'X', 'E', 'D')) { |
5685 | 967 | SKIP(6); |
5686 | 967 | val = XML_ATTRIBUTE_FIXED; |
5687 | 967 | if (SKIP_BLANKS_PE == 0) { |
5688 | 770 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5689 | 770 | "Space required after '#FIXED'\n"); |
5690 | 770 | } |
5691 | 967 | } |
5692 | 18.5k | ret = xmlParseAttValue(ctxt); |
5693 | 18.5k | if (ret == NULL) { |
5694 | 2.54k | xmlFatalErrMsg(ctxt, (xmlParserErrors)ctxt->errNo, |
5695 | 2.54k | "Attribute default value declaration error\n"); |
5696 | 2.54k | } else |
5697 | 15.9k | *value = ret; |
5698 | 18.5k | return(val); |
5699 | 19.0k | } |
5700 | | |
5701 | | /** |
5702 | | * parse an Notation attribute type. |
5703 | | * |
5704 | | * @deprecated Internal function, don't use. |
5705 | | * |
5706 | | * Note: the leading 'NOTATION' S part has already being parsed... |
5707 | | * |
5708 | | * [58] NotationType ::= 'NOTATION' S '(' S? Name (S? '|' S? Name)* S? ')' |
5709 | | * |
5710 | | * [ VC: Notation Attributes ] |
5711 | | * Values of this type must match one of the notation names included |
5712 | | * in the declaration; all notation names in the declaration must be declared. |
5713 | | * |
5714 | | * @param ctxt an XML parser context |
5715 | | * @returns the notation attribute tree built while parsing |
5716 | | */ |
5717 | | |
5718 | | xmlEnumeration * |
5719 | 765 | xmlParseNotationType(xmlParserCtxt *ctxt) { |
5720 | 765 | const xmlChar *name; |
5721 | 765 | xmlEnumerationPtr ret = NULL, last = NULL, cur, tmp; |
5722 | | |
5723 | 765 | if (RAW != '(') { |
5724 | 334 | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_STARTED, NULL); |
5725 | 334 | return(NULL); |
5726 | 334 | } |
5727 | 571 | do { |
5728 | 571 | NEXT; |
5729 | 571 | SKIP_BLANKS_PE; |
5730 | 571 | name = xmlParseName(ctxt); |
5731 | 571 | if (name == NULL) { |
5732 | 129 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
5733 | 129 | "Name expected in NOTATION declaration\n"); |
5734 | 129 | xmlFreeEnumeration(ret); |
5735 | 129 | return(NULL); |
5736 | 129 | } |
5737 | 442 | tmp = NULL; |
5738 | 442 | #ifdef LIBXML_VALID_ENABLED |
5739 | 442 | if (ctxt->validate) { |
5740 | 0 | tmp = ret; |
5741 | 0 | while (tmp != NULL) { |
5742 | 0 | if (xmlStrEqual(name, tmp->name)) { |
5743 | 0 | xmlValidityError(ctxt, XML_DTD_DUP_TOKEN, |
5744 | 0 | "standalone: attribute notation value token %s duplicated\n", |
5745 | 0 | name, NULL); |
5746 | 0 | if (!xmlDictOwns(ctxt->dict, name)) |
5747 | 0 | xmlFree((xmlChar *) name); |
5748 | 0 | break; |
5749 | 0 | } |
5750 | 0 | tmp = tmp->next; |
5751 | 0 | } |
5752 | 0 | } |
5753 | 442 | #endif /* LIBXML_VALID_ENABLED */ |
5754 | 442 | if (tmp == NULL) { |
5755 | 442 | cur = xmlCreateEnumeration(name); |
5756 | 442 | if (cur == NULL) { |
5757 | 0 | xmlErrMemory(ctxt); |
5758 | 0 | xmlFreeEnumeration(ret); |
5759 | 0 | return(NULL); |
5760 | 0 | } |
5761 | 442 | if (last == NULL) ret = last = cur; |
5762 | 136 | else { |
5763 | 136 | last->next = cur; |
5764 | 136 | last = cur; |
5765 | 136 | } |
5766 | 442 | } |
5767 | 442 | SKIP_BLANKS_PE; |
5768 | 442 | } while (RAW == '|'); |
5769 | 302 | if (RAW != ')') { |
5770 | 262 | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_FINISHED, NULL); |
5771 | 262 | xmlFreeEnumeration(ret); |
5772 | 262 | return(NULL); |
5773 | 262 | } |
5774 | 40 | NEXT; |
5775 | 40 | return(ret); |
5776 | 302 | } |
5777 | | |
5778 | | /** |
5779 | | * parse an Enumeration attribute type. |
5780 | | * |
5781 | | * @deprecated Internal function, don't use. |
5782 | | * |
5783 | | * [59] Enumeration ::= '(' S? Nmtoken (S? '|' S? Nmtoken)* S? ')' |
5784 | | * |
5785 | | * [ VC: Enumeration ] |
5786 | | * Values of this type must match one of the Nmtoken tokens in |
5787 | | * the declaration |
5788 | | * |
5789 | | * @param ctxt an XML parser context |
5790 | | * @returns the enumeration attribute tree built while parsing |
5791 | | */ |
5792 | | |
5793 | | xmlEnumeration * |
5794 | 4.77k | xmlParseEnumerationType(xmlParserCtxt *ctxt) { |
5795 | 4.77k | xmlChar *name; |
5796 | 4.77k | xmlEnumerationPtr ret = NULL, last = NULL, cur, tmp; |
5797 | | |
5798 | 4.77k | if (RAW != '(') { |
5799 | 891 | xmlFatalErr(ctxt, XML_ERR_ATTLIST_NOT_STARTED, NULL); |
5800 | 891 | return(NULL); |
5801 | 891 | } |
5802 | 9.13k | do { |
5803 | 9.13k | NEXT; |
5804 | 9.13k | SKIP_BLANKS_PE; |
5805 | 9.13k | name = xmlParseNmtoken(ctxt); |
5806 | 9.13k | if (name == NULL) { |
5807 | 800 | xmlFatalErr(ctxt, XML_ERR_NMTOKEN_REQUIRED, NULL); |
5808 | 800 | return(ret); |
5809 | 800 | } |
5810 | 8.33k | tmp = NULL; |
5811 | 8.33k | #ifdef LIBXML_VALID_ENABLED |
5812 | 8.33k | if (ctxt->validate) { |
5813 | 0 | tmp = ret; |
5814 | 0 | while (tmp != NULL) { |
5815 | 0 | if (xmlStrEqual(name, tmp->name)) { |
5816 | 0 | xmlValidityError(ctxt, XML_DTD_DUP_TOKEN, |
5817 | 0 | "standalone: attribute enumeration value token %s duplicated\n", |
5818 | 0 | name, NULL); |
5819 | 0 | if (!xmlDictOwns(ctxt->dict, name)) |
5820 | 0 | xmlFree(name); |
5821 | 0 | break; |
5822 | 0 | } |
5823 | 0 | tmp = tmp->next; |
5824 | 0 | } |
5825 | 0 | } |
5826 | 8.33k | #endif /* LIBXML_VALID_ENABLED */ |
5827 | 8.33k | if (tmp == NULL) { |
5828 | 8.33k | cur = xmlCreateEnumeration(name); |
5829 | 8.33k | if (!xmlDictOwns(ctxt->dict, name)) |
5830 | 8.33k | xmlFree(name); |
5831 | 8.33k | if (cur == NULL) { |
5832 | 0 | xmlErrMemory(ctxt); |
5833 | 0 | xmlFreeEnumeration(ret); |
5834 | 0 | return(NULL); |
5835 | 0 | } |
5836 | 8.33k | if (last == NULL) ret = last = cur; |
5837 | 4.50k | else { |
5838 | 4.50k | last->next = cur; |
5839 | 4.50k | last = cur; |
5840 | 4.50k | } |
5841 | 8.33k | } |
5842 | 8.33k | SKIP_BLANKS_PE; |
5843 | 8.33k | } while (RAW == '|'); |
5844 | 3.08k | if (RAW != ')') { |
5845 | 581 | xmlFatalErr(ctxt, XML_ERR_ATTLIST_NOT_FINISHED, NULL); |
5846 | 581 | return(ret); |
5847 | 581 | } |
5848 | 2.50k | NEXT; |
5849 | 2.50k | return(ret); |
5850 | 3.08k | } |
5851 | | |
5852 | | /** |
5853 | | * parse an Enumerated attribute type. |
5854 | | * |
5855 | | * @deprecated Internal function, don't use. |
5856 | | * |
5857 | | * [57] EnumeratedType ::= NotationType | Enumeration |
5858 | | * |
5859 | | * [58] NotationType ::= 'NOTATION' S '(' S? Name (S? '|' S? Name)* S? ')' |
5860 | | * |
5861 | | * @param ctxt an XML parser context |
5862 | | * @param tree the enumeration tree built while parsing |
5863 | | * @returns XML_ATTRIBUTE_ENUMERATION or XML_ATTRIBUTE_NOTATION |
5864 | | */ |
5865 | | |
5866 | | int |
5867 | 5.74k | xmlParseEnumeratedType(xmlParserCtxt *ctxt, xmlEnumeration **tree) { |
5868 | 5.74k | if (CMP8(CUR_PTR, 'N', 'O', 'T', 'A', 'T', 'I', 'O', 'N')) { |
5869 | 969 | SKIP(8); |
5870 | 969 | if (SKIP_BLANKS_PE == 0) { |
5871 | 204 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5872 | 204 | "Space required after 'NOTATION'\n"); |
5873 | 204 | return(0); |
5874 | 204 | } |
5875 | 765 | *tree = xmlParseNotationType(ctxt); |
5876 | 765 | if (*tree == NULL) return(0); |
5877 | 40 | return(XML_ATTRIBUTE_NOTATION); |
5878 | 765 | } |
5879 | 4.77k | *tree = xmlParseEnumerationType(ctxt); |
5880 | 4.77k | if (*tree == NULL) return(0); |
5881 | 3.83k | return(XML_ATTRIBUTE_ENUMERATION); |
5882 | 4.77k | } |
5883 | | |
5884 | | /** |
5885 | | * parse the Attribute list def for an element |
5886 | | * |
5887 | | * @deprecated Internal function, don't use. |
5888 | | * |
5889 | | * [54] AttType ::= StringType | TokenizedType | EnumeratedType |
5890 | | * |
5891 | | * [55] StringType ::= 'CDATA' |
5892 | | * |
5893 | | * [56] TokenizedType ::= 'ID' | 'IDREF' | 'IDREFS' | 'ENTITY' | |
5894 | | * 'ENTITIES' | 'NMTOKEN' | 'NMTOKENS' |
5895 | | * |
5896 | | * Validity constraints for attribute values syntax are checked in |
5897 | | * #xmlValidateAttributeValue |
5898 | | * |
5899 | | * [ VC: ID ] |
5900 | | * Values of type ID must match the Name production. A name must not |
5901 | | * appear more than once in an XML document as a value of this type; |
5902 | | * i.e., ID values must uniquely identify the elements which bear them. |
5903 | | * |
5904 | | * [ VC: One ID per Element Type ] |
5905 | | * No element type may have more than one ID attribute specified. |
5906 | | * |
5907 | | * [ VC: ID Attribute Default ] |
5908 | | * An ID attribute must have a declared default of \#IMPLIED or \#REQUIRED. |
5909 | | * |
5910 | | * [ VC: IDREF ] |
5911 | | * Values of type IDREF must match the Name production, and values |
5912 | | * of type IDREFS must match Names; each IDREF Name must match the value |
5913 | | * of an ID attribute on some element in the XML document; i.e. IDREF |
5914 | | * values must match the value of some ID attribute. |
5915 | | * |
5916 | | * [ VC: Entity Name ] |
5917 | | * Values of type ENTITY must match the Name production, values |
5918 | | * of type ENTITIES must match Names; each Entity Name must match the |
5919 | | * name of an unparsed entity declared in the DTD. |
5920 | | * |
5921 | | * [ VC: Name Token ] |
5922 | | * Values of type NMTOKEN must match the Nmtoken production; values |
5923 | | * of type NMTOKENS must match Nmtokens. |
5924 | | * |
5925 | | * @param ctxt an XML parser context |
5926 | | * @param tree the enumeration tree built while parsing |
5927 | | * @returns the attribute type |
5928 | | */ |
5929 | | int |
5930 | 24.5k | xmlParseAttributeType(xmlParserCtxt *ctxt, xmlEnumeration **tree) { |
5931 | 24.5k | if (CMP5(CUR_PTR, 'C', 'D', 'A', 'T', 'A')) { |
5932 | 1.37k | SKIP(5); |
5933 | 1.37k | return(XML_ATTRIBUTE_CDATA); |
5934 | 23.1k | } else if (CMP6(CUR_PTR, 'I', 'D', 'R', 'E', 'F', 'S')) { |
5935 | 484 | SKIP(6); |
5936 | 484 | return(XML_ATTRIBUTE_IDREFS); |
5937 | 22.7k | } else if (CMP5(CUR_PTR, 'I', 'D', 'R', 'E', 'F')) { |
5938 | 539 | SKIP(5); |
5939 | 539 | return(XML_ATTRIBUTE_IDREF); |
5940 | 22.1k | } else if ((RAW == 'I') && (NXT(1) == 'D')) { |
5941 | 13.0k | SKIP(2); |
5942 | 13.0k | return(XML_ATTRIBUTE_ID); |
5943 | 13.0k | } else if (CMP6(CUR_PTR, 'E', 'N', 'T', 'I', 'T', 'Y')) { |
5944 | 1.27k | SKIP(6); |
5945 | 1.27k | return(XML_ATTRIBUTE_ENTITY); |
5946 | 7.86k | } else if (CMP8(CUR_PTR, 'E', 'N', 'T', 'I', 'T', 'I', 'E', 'S')) { |
5947 | 1.40k | SKIP(8); |
5948 | 1.40k | return(XML_ATTRIBUTE_ENTITIES); |
5949 | 6.45k | } else if (CMP8(CUR_PTR, 'N', 'M', 'T', 'O', 'K', 'E', 'N', 'S')) { |
5950 | 502 | SKIP(8); |
5951 | 502 | return(XML_ATTRIBUTE_NMTOKENS); |
5952 | 5.95k | } else if (CMP7(CUR_PTR, 'N', 'M', 'T', 'O', 'K', 'E', 'N')) { |
5953 | 213 | SKIP(7); |
5954 | 213 | return(XML_ATTRIBUTE_NMTOKEN); |
5955 | 213 | } |
5956 | 5.74k | return(xmlParseEnumeratedType(ctxt, tree)); |
5957 | 24.5k | } |
5958 | | |
5959 | | /** |
5960 | | * Parse an attribute list declaration for an element. Always consumes '<!'. |
5961 | | * |
5962 | | * @deprecated Internal function, don't use. |
5963 | | * |
5964 | | * [52] AttlistDecl ::= '<!ATTLIST' S Name AttDef* S? '>' |
5965 | | * |
5966 | | * [53] AttDef ::= S Name S AttType S DefaultDecl |
5967 | | * @param ctxt an XML parser context |
5968 | | */ |
5969 | | void |
5970 | 24.8k | xmlParseAttributeListDecl(xmlParserCtxt *ctxt) { |
5971 | 24.8k | const xmlChar *elemName; |
5972 | 24.8k | const xmlChar *attrName; |
5973 | 24.8k | xmlEnumerationPtr tree; |
5974 | | |
5975 | 24.8k | if ((CUR != '<') || (NXT(1) != '!')) |
5976 | 0 | return; |
5977 | 24.8k | SKIP(2); |
5978 | | |
5979 | 24.8k | if (CMP7(CUR_PTR, 'A', 'T', 'T', 'L', 'I', 'S', 'T')) { |
5980 | 24.8k | #ifdef LIBXML_VALID_ENABLED |
5981 | 24.8k | int oldInputNr = ctxt->inputNr; |
5982 | 24.8k | #endif |
5983 | | |
5984 | 24.8k | SKIP(7); |
5985 | 24.8k | if (SKIP_BLANKS_PE == 0) { |
5986 | 14.9k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5987 | 14.9k | "Space required after '<!ATTLIST'\n"); |
5988 | 14.9k | } |
5989 | 24.8k | elemName = xmlParseName(ctxt); |
5990 | 24.8k | if (elemName == NULL) { |
5991 | 2.28k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
5992 | 2.28k | "ATTLIST: no name for Element\n"); |
5993 | 2.28k | return; |
5994 | 2.28k | } |
5995 | 22.5k | SKIP_BLANKS_PE; |
5996 | 22.5k | GROW; |
5997 | 39.2k | while ((RAW != '>') && (PARSER_STOPPED(ctxt) == 0)) { |
5998 | 35.1k | int type; |
5999 | 35.1k | int def; |
6000 | 35.1k | xmlChar *defaultValue = NULL; |
6001 | | |
6002 | 35.1k | GROW; |
6003 | 35.1k | tree = NULL; |
6004 | 35.1k | attrName = xmlParseName(ctxt); |
6005 | 35.1k | if (attrName == NULL) { |
6006 | 9.99k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6007 | 9.99k | "ATTLIST: no name for Attribute\n"); |
6008 | 9.99k | break; |
6009 | 9.99k | } |
6010 | 25.1k | GROW; |
6011 | 25.1k | if (SKIP_BLANKS_PE == 0) { |
6012 | 613 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6013 | 613 | "Space required after the attribute name\n"); |
6014 | 613 | break; |
6015 | 613 | } |
6016 | | |
6017 | 24.5k | type = xmlParseAttributeType(ctxt, &tree); |
6018 | 24.5k | if (type <= 0) { |
6019 | 1.86k | break; |
6020 | 1.86k | } |
6021 | | |
6022 | 22.7k | GROW; |
6023 | 22.7k | if (SKIP_BLANKS_PE == 0) { |
6024 | 3.56k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6025 | 3.56k | "Space required after the attribute type\n"); |
6026 | 3.56k | if (tree != NULL) |
6027 | 1.37k | xmlFreeEnumeration(tree); |
6028 | 3.56k | break; |
6029 | 3.56k | } |
6030 | | |
6031 | 19.1k | def = xmlParseDefaultDecl(ctxt, &defaultValue); |
6032 | 19.1k | if (def <= 0) { |
6033 | 0 | if (defaultValue != NULL) |
6034 | 0 | xmlFree(defaultValue); |
6035 | 0 | if (tree != NULL) |
6036 | 0 | xmlFreeEnumeration(tree); |
6037 | 0 | break; |
6038 | 0 | } |
6039 | 19.1k | if ((type != XML_ATTRIBUTE_CDATA) && (defaultValue != NULL)) |
6040 | 14.8k | xmlAttrNormalizeSpace(defaultValue, defaultValue); |
6041 | | |
6042 | 19.1k | GROW; |
6043 | 19.1k | if (RAW != '>') { |
6044 | 16.5k | if (SKIP_BLANKS_PE == 0) { |
6045 | 2.41k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6046 | 2.41k | "Space required after the attribute default value\n"); |
6047 | 2.41k | if (defaultValue != NULL) |
6048 | 113 | xmlFree(defaultValue); |
6049 | 2.41k | if (tree != NULL) |
6050 | 1.01k | xmlFreeEnumeration(tree); |
6051 | 2.41k | break; |
6052 | 2.41k | } |
6053 | 16.5k | } |
6054 | 16.7k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
6055 | 16.7k | (ctxt->sax->attributeDecl != NULL)) |
6056 | 9.93k | ctxt->sax->attributeDecl(ctxt->userData, elemName, attrName, |
6057 | 9.93k | type, def, defaultValue, tree); |
6058 | 6.80k | else if (tree != NULL) |
6059 | 717 | xmlFreeEnumeration(tree); |
6060 | | |
6061 | 16.7k | if ((ctxt->sax2) && (defaultValue != NULL) && |
6062 | 16.7k | (def != XML_ATTRIBUTE_IMPLIED) && |
6063 | 16.7k | (def != XML_ATTRIBUTE_REQUIRED)) { |
6064 | 15.8k | xmlAddDefAttrs(ctxt, elemName, attrName, defaultValue); |
6065 | 15.8k | } |
6066 | 16.7k | if (ctxt->sax2) { |
6067 | 16.7k | xmlAddSpecialAttr(ctxt, elemName, attrName, type); |
6068 | 16.7k | } |
6069 | 16.7k | if (defaultValue != NULL) |
6070 | 15.8k | xmlFree(defaultValue); |
6071 | 16.7k | GROW; |
6072 | 16.7k | } |
6073 | 22.5k | if (RAW == '>') { |
6074 | 4.14k | #ifdef LIBXML_VALID_ENABLED |
6075 | 4.14k | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
6076 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6077 | 0 | "Attribute list declaration doesn't start and" |
6078 | 0 | " stop in the same entity\n", |
6079 | 0 | NULL, NULL); |
6080 | 0 | } |
6081 | 4.14k | #endif |
6082 | 4.14k | NEXT; |
6083 | 4.14k | } |
6084 | 22.5k | } |
6085 | 24.8k | } |
6086 | | |
6087 | | /** |
6088 | | * Handle PEs and check that we don't pop the entity that started |
6089 | | * a balanced group. |
6090 | | * |
6091 | | * @param ctxt parser context |
6092 | | * @param openInputNr input nr of the entity with opening '(' |
6093 | | */ |
6094 | | static void |
6095 | 488k | xmlSkipBlankCharsPEBalanced(xmlParserCtxt *ctxt, int openInputNr) { |
6096 | 488k | SKIP_BLANKS; |
6097 | 488k | GROW; |
6098 | | |
6099 | 488k | (void) openInputNr; |
6100 | | |
6101 | 488k | if (!PARSER_EXTERNAL(ctxt) && !PARSER_IN_PE(ctxt)) |
6102 | 44.8k | return; |
6103 | | |
6104 | 463k | while (!PARSER_STOPPED(ctxt)) { |
6105 | 463k | if (ctxt->input->cur >= ctxt->input->end) { |
6106 | 8.59k | #ifdef LIBXML_VALID_ENABLED |
6107 | 8.59k | if ((ctxt->validate) && (ctxt->inputNr <= openInputNr)) { |
6108 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6109 | 0 | "Element content declaration doesn't start " |
6110 | 0 | "and stop in the same entity\n", |
6111 | 0 | NULL, NULL); |
6112 | 0 | } |
6113 | 8.59k | #endif |
6114 | 8.59k | if (PARSER_IN_PE(ctxt)) |
6115 | 8.54k | xmlPopPE(ctxt); |
6116 | 51 | else |
6117 | 51 | break; |
6118 | 455k | } else if (RAW == '%') { |
6119 | 11.3k | xmlParsePERefInternal(ctxt, 0); |
6120 | 443k | } else { |
6121 | 443k | break; |
6122 | 443k | } |
6123 | | |
6124 | 19.8k | SKIP_BLANKS; |
6125 | 19.8k | GROW; |
6126 | 19.8k | } |
6127 | 444k | } |
6128 | | |
6129 | | /** |
6130 | | * parse the declaration for a Mixed Element content |
6131 | | * The leading '(' and spaces have been skipped in #xmlParseElementContentDecl |
6132 | | * |
6133 | | * @deprecated Internal function, don't use. |
6134 | | * |
6135 | | * [51] Mixed ::= '(' S? '#PCDATA' (S? '|' S? Name)* S? ')*' | |
6136 | | * '(' S? '#PCDATA' S? ')' |
6137 | | * |
6138 | | * [ VC: Proper Group/PE Nesting ] applies to [51] too (see [49]) |
6139 | | * |
6140 | | * [ VC: No Duplicate Types ] |
6141 | | * The same name must not appear more than once in a single |
6142 | | * mixed-content declaration. |
6143 | | * |
6144 | | * @param ctxt an XML parser context |
6145 | | * @param openInputNr the input used for the current entity, needed for |
6146 | | * boundary checks |
6147 | | * @returns the list of the xmlElementContent describing the element choices |
6148 | | */ |
6149 | | xmlElementContent * |
6150 | 1.17k | xmlParseElementMixedContentDecl(xmlParserCtxt *ctxt, int openInputNr) { |
6151 | 1.17k | xmlElementContentPtr ret = NULL, cur = NULL, n; |
6152 | 1.17k | const xmlChar *elem = NULL; |
6153 | | |
6154 | 1.17k | GROW; |
6155 | 1.17k | if (CMP7(CUR_PTR, '#', 'P', 'C', 'D', 'A', 'T', 'A')) { |
6156 | 1.17k | SKIP(7); |
6157 | 1.17k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6158 | 1.17k | if (RAW == ')') { |
6159 | 296 | #ifdef LIBXML_VALID_ENABLED |
6160 | 296 | if ((ctxt->validate) && (ctxt->inputNr > openInputNr)) { |
6161 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6162 | 0 | "Element content declaration doesn't start " |
6163 | 0 | "and stop in the same entity\n", |
6164 | 0 | NULL, NULL); |
6165 | 0 | } |
6166 | 296 | #endif |
6167 | 296 | NEXT; |
6168 | 296 | ret = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_PCDATA); |
6169 | 296 | if (ret == NULL) |
6170 | 0 | goto mem_error; |
6171 | 296 | if (RAW == '*') { |
6172 | 70 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6173 | 70 | NEXT; |
6174 | 70 | } |
6175 | 296 | return(ret); |
6176 | 296 | } |
6177 | 882 | if ((RAW == '(') || (RAW == '|')) { |
6178 | 525 | ret = cur = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_PCDATA); |
6179 | 525 | if (ret == NULL) |
6180 | 0 | goto mem_error; |
6181 | 525 | } |
6182 | 2.33k | while ((RAW == '|') && (PARSER_STOPPED(ctxt) == 0)) { |
6183 | 1.71k | NEXT; |
6184 | 1.71k | n = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_OR); |
6185 | 1.71k | if (n == NULL) |
6186 | 0 | goto mem_error; |
6187 | 1.71k | if (elem == NULL) { |
6188 | 524 | n->c1 = cur; |
6189 | 524 | if (cur != NULL) |
6190 | 524 | cur->parent = n; |
6191 | 524 | ret = cur = n; |
6192 | 1.19k | } else { |
6193 | 1.19k | cur->c2 = n; |
6194 | 1.19k | n->parent = cur; |
6195 | 1.19k | n->c1 = xmlNewDocElementContent(ctxt->myDoc, elem, XML_ELEMENT_CONTENT_ELEMENT); |
6196 | 1.19k | if (n->c1 == NULL) |
6197 | 0 | goto mem_error; |
6198 | 1.19k | n->c1->parent = n; |
6199 | 1.19k | cur = n; |
6200 | 1.19k | } |
6201 | 1.71k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6202 | 1.71k | elem = xmlParseName(ctxt); |
6203 | 1.71k | if (elem == NULL) { |
6204 | 258 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6205 | 258 | "xmlParseElementMixedContentDecl : Name expected\n"); |
6206 | 258 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6207 | 258 | return(NULL); |
6208 | 258 | } |
6209 | 1.45k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6210 | 1.45k | } |
6211 | 624 | if ((RAW == ')') && (NXT(1) == '*')) { |
6212 | 208 | if (elem != NULL) { |
6213 | 208 | cur->c2 = xmlNewDocElementContent(ctxt->myDoc, elem, |
6214 | 208 | XML_ELEMENT_CONTENT_ELEMENT); |
6215 | 208 | if (cur->c2 == NULL) |
6216 | 0 | goto mem_error; |
6217 | 208 | cur->c2->parent = cur; |
6218 | 208 | } |
6219 | 208 | if (ret != NULL) |
6220 | 208 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6221 | 208 | #ifdef LIBXML_VALID_ENABLED |
6222 | 208 | if ((ctxt->validate) && (ctxt->inputNr > openInputNr)) { |
6223 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6224 | 0 | "Element content declaration doesn't start " |
6225 | 0 | "and stop in the same entity\n", |
6226 | 0 | NULL, NULL); |
6227 | 0 | } |
6228 | 208 | #endif |
6229 | 208 | SKIP(2); |
6230 | 416 | } else { |
6231 | 416 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6232 | 416 | xmlFatalErr(ctxt, XML_ERR_MIXED_NOT_STARTED, NULL); |
6233 | 416 | return(NULL); |
6234 | 416 | } |
6235 | | |
6236 | 624 | } else { |
6237 | 0 | xmlFatalErr(ctxt, XML_ERR_PCDATA_REQUIRED, NULL); |
6238 | 0 | } |
6239 | 208 | return(ret); |
6240 | | |
6241 | 0 | mem_error: |
6242 | 0 | xmlErrMemory(ctxt); |
6243 | 0 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6244 | 0 | return(NULL); |
6245 | 1.17k | } |
6246 | | |
6247 | | /** |
6248 | | * parse the declaration for a Mixed Element content |
6249 | | * The leading '(' and spaces have been skipped in #xmlParseElementContentDecl |
6250 | | * |
6251 | | * [47] children ::= (choice | seq) ('?' | '*' | '+')? |
6252 | | * |
6253 | | * [48] cp ::= (Name | choice | seq) ('?' | '*' | '+')? |
6254 | | * |
6255 | | * [49] choice ::= '(' S? cp ( S? '|' S? cp )* S? ')' |
6256 | | * |
6257 | | * [50] seq ::= '(' S? cp ( S? ',' S? cp )* S? ')' |
6258 | | * |
6259 | | * [ VC: Proper Group/PE Nesting ] applies to [49] and [50] |
6260 | | * TODO Parameter-entity replacement text must be properly nested |
6261 | | * with parenthesized groups. That is to say, if either of the |
6262 | | * opening or closing parentheses in a choice, seq, or Mixed |
6263 | | * construct is contained in the replacement text for a parameter |
6264 | | * entity, both must be contained in the same replacement text. For |
6265 | | * interoperability, if a parameter-entity reference appears in a |
6266 | | * choice, seq, or Mixed construct, its replacement text should not |
6267 | | * be empty, and neither the first nor last non-blank character of |
6268 | | * the replacement text should be a connector (| or ,). |
6269 | | * |
6270 | | * @param ctxt an XML parser context |
6271 | | * @param openInputNr the input used for the current entity, needed for |
6272 | | * boundary checks |
6273 | | * @param depth the level of recursion |
6274 | | * @returns the tree of xmlElementContent describing the element |
6275 | | * hierarchy. |
6276 | | */ |
6277 | | static xmlElementContentPtr |
6278 | | xmlParseElementChildrenContentDeclPriv(xmlParserCtxtPtr ctxt, int openInputNr, |
6279 | 104k | int depth) { |
6280 | 104k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 2048 : 256; |
6281 | 104k | xmlElementContentPtr ret = NULL, cur = NULL, last = NULL, op = NULL; |
6282 | 104k | const xmlChar *elem; |
6283 | 104k | xmlChar type = 0; |
6284 | | |
6285 | 104k | if (depth > maxDepth) { |
6286 | 1 | xmlFatalErrMsgInt(ctxt, XML_ERR_RESOURCE_LIMIT, |
6287 | 1 | "xmlParseElementChildrenContentDecl : depth %d too deep, " |
6288 | 1 | "use XML_PARSE_HUGE\n", depth); |
6289 | 1 | return(NULL); |
6290 | 1 | } |
6291 | 104k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6292 | 104k | if (RAW == '(') { |
6293 | 30.5k | int newInputNr = ctxt->inputNr; |
6294 | | |
6295 | | /* Recurse on first child */ |
6296 | 30.5k | NEXT; |
6297 | 30.5k | cur = ret = xmlParseElementChildrenContentDeclPriv(ctxt, newInputNr, |
6298 | 30.5k | depth + 1); |
6299 | 30.5k | if (cur == NULL) |
6300 | 30.0k | return(NULL); |
6301 | 74.3k | } else { |
6302 | 74.3k | elem = xmlParseName(ctxt); |
6303 | 74.3k | if (elem == NULL) { |
6304 | 1.09k | xmlFatalErr(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, NULL); |
6305 | 1.09k | return(NULL); |
6306 | 1.09k | } |
6307 | 73.2k | cur = ret = xmlNewDocElementContent(ctxt->myDoc, elem, XML_ELEMENT_CONTENT_ELEMENT); |
6308 | 73.2k | if (cur == NULL) { |
6309 | 0 | xmlErrMemory(ctxt); |
6310 | 0 | return(NULL); |
6311 | 0 | } |
6312 | 73.2k | GROW; |
6313 | 73.2k | if (RAW == '?') { |
6314 | 59.9k | cur->ocur = XML_ELEMENT_CONTENT_OPT; |
6315 | 59.9k | NEXT; |
6316 | 59.9k | } else if (RAW == '*') { |
6317 | 913 | cur->ocur = XML_ELEMENT_CONTENT_MULT; |
6318 | 913 | NEXT; |
6319 | 12.4k | } else if (RAW == '+') { |
6320 | 707 | cur->ocur = XML_ELEMENT_CONTENT_PLUS; |
6321 | 707 | NEXT; |
6322 | 11.7k | } else { |
6323 | 11.7k | cur->ocur = XML_ELEMENT_CONTENT_ONCE; |
6324 | 11.7k | } |
6325 | 73.2k | GROW; |
6326 | 73.2k | } |
6327 | 209k | while (!PARSER_STOPPED(ctxt)) { |
6328 | 209k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6329 | 209k | if (RAW == ')') |
6330 | 43.7k | break; |
6331 | | /* |
6332 | | * Each loop we parse one separator and one element. |
6333 | | */ |
6334 | 166k | if (RAW == ',') { |
6335 | 584 | if (type == 0) type = CUR; |
6336 | | |
6337 | | /* |
6338 | | * Detect "Name | Name , Name" error |
6339 | | */ |
6340 | 200 | else if (type != CUR) { |
6341 | 1 | xmlFatalErrMsgInt(ctxt, XML_ERR_SEPARATOR_REQUIRED, |
6342 | 1 | "xmlParseElementChildrenContentDecl : '%c' expected\n", |
6343 | 1 | type); |
6344 | 1 | if ((last != NULL) && (last != ret)) |
6345 | 1 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6346 | 1 | if (ret != NULL) |
6347 | 1 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6348 | 1 | return(NULL); |
6349 | 1 | } |
6350 | 583 | NEXT; |
6351 | | |
6352 | 583 | op = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_SEQ); |
6353 | 583 | if (op == NULL) { |
6354 | 0 | xmlErrMemory(ctxt); |
6355 | 0 | if ((last != NULL) && (last != ret)) |
6356 | 0 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6357 | 0 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6358 | 0 | return(NULL); |
6359 | 0 | } |
6360 | 583 | if (last == NULL) { |
6361 | 384 | op->c1 = ret; |
6362 | 384 | if (ret != NULL) |
6363 | 384 | ret->parent = op; |
6364 | 384 | ret = cur = op; |
6365 | 384 | } else { |
6366 | 199 | cur->c2 = op; |
6367 | 199 | if (op != NULL) |
6368 | 199 | op->parent = cur; |
6369 | 199 | op->c1 = last; |
6370 | 199 | if (last != NULL) |
6371 | 199 | last->parent = op; |
6372 | 199 | cur =op; |
6373 | 199 | last = NULL; |
6374 | 199 | } |
6375 | 165k | } else if (RAW == '|') { |
6376 | 157k | if (type == 0) type = CUR; |
6377 | | |
6378 | | /* |
6379 | | * Detect "Name , Name | Name" error |
6380 | | */ |
6381 | 87.7k | else if (type != CUR) { |
6382 | 1 | xmlFatalErrMsgInt(ctxt, XML_ERR_SEPARATOR_REQUIRED, |
6383 | 1 | "xmlParseElementChildrenContentDecl : '%c' expected\n", |
6384 | 1 | type); |
6385 | 1 | if ((last != NULL) && (last != ret)) |
6386 | 1 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6387 | 1 | if (ret != NULL) |
6388 | 1 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6389 | 1 | return(NULL); |
6390 | 1 | } |
6391 | 157k | NEXT; |
6392 | | |
6393 | 157k | op = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_OR); |
6394 | 157k | if (op == NULL) { |
6395 | 0 | xmlErrMemory(ctxt); |
6396 | 0 | if ((last != NULL) && (last != ret)) |
6397 | 0 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6398 | 0 | if (ret != NULL) |
6399 | 0 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6400 | 0 | return(NULL); |
6401 | 0 | } |
6402 | 157k | if (last == NULL) { |
6403 | 69.3k | op->c1 = ret; |
6404 | 69.3k | if (ret != NULL) |
6405 | 69.3k | ret->parent = op; |
6406 | 69.3k | ret = cur = op; |
6407 | 87.7k | } else { |
6408 | 87.7k | cur->c2 = op; |
6409 | 87.7k | if (op != NULL) |
6410 | 87.7k | op->parent = cur; |
6411 | 87.7k | op->c1 = last; |
6412 | 87.7k | if (last != NULL) |
6413 | 87.7k | last->parent = op; |
6414 | 87.7k | cur =op; |
6415 | 87.7k | last = NULL; |
6416 | 87.7k | } |
6417 | 157k | } else { |
6418 | 8.50k | xmlFatalErr(ctxt, XML_ERR_ELEMCONTENT_NOT_FINISHED, NULL); |
6419 | 8.50k | if ((last != NULL) && (last != ret)) |
6420 | 7.55k | xmlFreeDocElementContent(ctxt->myDoc, last); |
6421 | 8.50k | if (ret != NULL) |
6422 | 8.50k | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6423 | 8.50k | return(NULL); |
6424 | 8.50k | } |
6425 | 157k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6426 | 157k | if (RAW == '(') { |
6427 | 63.6k | int newInputNr = ctxt->inputNr; |
6428 | | |
6429 | | /* Recurse on second child */ |
6430 | 63.6k | NEXT; |
6431 | 63.6k | last = xmlParseElementChildrenContentDeclPriv(ctxt, newInputNr, |
6432 | 63.6k | depth + 1); |
6433 | 63.6k | if (last == NULL) { |
6434 | 20.9k | if (ret != NULL) |
6435 | 20.9k | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6436 | 20.9k | return(NULL); |
6437 | 20.9k | } |
6438 | 94.0k | } else { |
6439 | 94.0k | elem = xmlParseName(ctxt); |
6440 | 94.0k | if (elem == NULL) { |
6441 | 569 | xmlFatalErr(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, NULL); |
6442 | 569 | if (ret != NULL) |
6443 | 569 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6444 | 569 | return(NULL); |
6445 | 569 | } |
6446 | 93.5k | last = xmlNewDocElementContent(ctxt->myDoc, elem, XML_ELEMENT_CONTENT_ELEMENT); |
6447 | 93.5k | if (last == NULL) { |
6448 | 0 | xmlErrMemory(ctxt); |
6449 | 0 | if (ret != NULL) |
6450 | 0 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6451 | 0 | return(NULL); |
6452 | 0 | } |
6453 | 93.5k | if (RAW == '?') { |
6454 | 26.8k | last->ocur = XML_ELEMENT_CONTENT_OPT; |
6455 | 26.8k | NEXT; |
6456 | 66.6k | } else if (RAW == '*') { |
6457 | 19.1k | last->ocur = XML_ELEMENT_CONTENT_MULT; |
6458 | 19.1k | NEXT; |
6459 | 47.4k | } else if (RAW == '+') { |
6460 | 79 | last->ocur = XML_ELEMENT_CONTENT_PLUS; |
6461 | 79 | NEXT; |
6462 | 47.3k | } else { |
6463 | 47.3k | last->ocur = XML_ELEMENT_CONTENT_ONCE; |
6464 | 47.3k | } |
6465 | 93.5k | } |
6466 | 157k | } |
6467 | 43.7k | if ((cur != NULL) && (last != NULL)) { |
6468 | 40.6k | cur->c2 = last; |
6469 | 40.6k | if (last != NULL) |
6470 | 40.6k | last->parent = cur; |
6471 | 40.6k | } |
6472 | 43.7k | #ifdef LIBXML_VALID_ENABLED |
6473 | 43.7k | if ((ctxt->validate) && (ctxt->inputNr > openInputNr)) { |
6474 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6475 | 0 | "Element content declaration doesn't start " |
6476 | 0 | "and stop in the same entity\n", |
6477 | 0 | NULL, NULL); |
6478 | 0 | } |
6479 | 43.7k | #endif |
6480 | 43.7k | NEXT; |
6481 | 43.7k | if (RAW == '?') { |
6482 | 1.34k | if (ret != NULL) { |
6483 | 1.34k | if ((ret->ocur == XML_ELEMENT_CONTENT_PLUS) || |
6484 | 1.34k | (ret->ocur == XML_ELEMENT_CONTENT_MULT)) |
6485 | 1.07k | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6486 | 271 | else |
6487 | 271 | ret->ocur = XML_ELEMENT_CONTENT_OPT; |
6488 | 1.34k | } |
6489 | 1.34k | NEXT; |
6490 | 42.3k | } else if (RAW == '*') { |
6491 | 38.4k | if (ret != NULL) { |
6492 | 38.4k | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6493 | 38.4k | cur = ret; |
6494 | | /* |
6495 | | * Some normalization: |
6496 | | * (a | b* | c?)* == (a | b | c)* |
6497 | | */ |
6498 | 115k | while ((cur != NULL) && (cur->type == XML_ELEMENT_CONTENT_OR)) { |
6499 | 76.7k | if ((cur->c1 != NULL) && |
6500 | 76.7k | ((cur->c1->ocur == XML_ELEMENT_CONTENT_OPT) || |
6501 | 76.7k | (cur->c1->ocur == XML_ELEMENT_CONTENT_MULT))) |
6502 | 51.2k | cur->c1->ocur = XML_ELEMENT_CONTENT_ONCE; |
6503 | 76.7k | if ((cur->c2 != NULL) && |
6504 | 76.7k | ((cur->c2->ocur == XML_ELEMENT_CONTENT_OPT) || |
6505 | 76.7k | (cur->c2->ocur == XML_ELEMENT_CONTENT_MULT))) |
6506 | 38.1k | cur->c2->ocur = XML_ELEMENT_CONTENT_ONCE; |
6507 | 76.7k | cur = cur->c2; |
6508 | 76.7k | } |
6509 | 38.4k | } |
6510 | 38.4k | NEXT; |
6511 | 38.4k | } else if (RAW == '+') { |
6512 | 1.74k | if (ret != NULL) { |
6513 | 1.74k | int found = 0; |
6514 | | |
6515 | 1.74k | if ((ret->ocur == XML_ELEMENT_CONTENT_OPT) || |
6516 | 1.74k | (ret->ocur == XML_ELEMENT_CONTENT_MULT)) |
6517 | 719 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6518 | 1.03k | else |
6519 | 1.03k | ret->ocur = XML_ELEMENT_CONTENT_PLUS; |
6520 | | /* |
6521 | | * Some normalization: |
6522 | | * (a | b*)+ == (a | b)* |
6523 | | * (a | b?)+ == (a | b)* |
6524 | | */ |
6525 | 4.57k | while ((cur != NULL) && (cur->type == XML_ELEMENT_CONTENT_OR)) { |
6526 | 2.83k | if ((cur->c1 != NULL) && |
6527 | 2.83k | ((cur->c1->ocur == XML_ELEMENT_CONTENT_OPT) || |
6528 | 2.83k | (cur->c1->ocur == XML_ELEMENT_CONTENT_MULT))) { |
6529 | 756 | cur->c1->ocur = XML_ELEMENT_CONTENT_ONCE; |
6530 | 756 | found = 1; |
6531 | 756 | } |
6532 | 2.83k | if ((cur->c2 != NULL) && |
6533 | 2.83k | ((cur->c2->ocur == XML_ELEMENT_CONTENT_OPT) || |
6534 | 2.83k | (cur->c2->ocur == XML_ELEMENT_CONTENT_MULT))) { |
6535 | 877 | cur->c2->ocur = XML_ELEMENT_CONTENT_ONCE; |
6536 | 877 | found = 1; |
6537 | 877 | } |
6538 | 2.83k | cur = cur->c2; |
6539 | 2.83k | } |
6540 | 1.74k | if (found) |
6541 | 905 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6542 | 1.74k | } |
6543 | 1.74k | NEXT; |
6544 | 1.74k | } |
6545 | 43.7k | return(ret); |
6546 | 73.7k | } |
6547 | | |
6548 | | /** |
6549 | | * parse the declaration for a Mixed Element content |
6550 | | * The leading '(' and spaces have been skipped in #xmlParseElementContentDecl |
6551 | | * |
6552 | | * @deprecated Internal function, don't use. |
6553 | | * |
6554 | | * [47] children ::= (choice | seq) ('?' | '*' | '+')? |
6555 | | * |
6556 | | * [48] cp ::= (Name | choice | seq) ('?' | '*' | '+')? |
6557 | | * |
6558 | | * [49] choice ::= '(' S? cp ( S? '|' S? cp )* S? ')' |
6559 | | * |
6560 | | * [50] seq ::= '(' S? cp ( S? ',' S? cp )* S? ')' |
6561 | | * |
6562 | | * [ VC: Proper Group/PE Nesting ] applies to [49] and [50] |
6563 | | * TODO Parameter-entity replacement text must be properly nested |
6564 | | * with parenthesized groups. That is to say, if either of the |
6565 | | * opening or closing parentheses in a choice, seq, or Mixed |
6566 | | * construct is contained in the replacement text for a parameter |
6567 | | * entity, both must be contained in the same replacement text. For |
6568 | | * interoperability, if a parameter-entity reference appears in a |
6569 | | * choice, seq, or Mixed construct, its replacement text should not |
6570 | | * be empty, and neither the first nor last non-blank character of |
6571 | | * the replacement text should be a connector (| or ,). |
6572 | | * |
6573 | | * @param ctxt an XML parser context |
6574 | | * @param inputchk the input used for the current entity, needed for boundary checks |
6575 | | * @returns the tree of xmlElementContent describing the element |
6576 | | * hierarchy. |
6577 | | */ |
6578 | | xmlElementContent * |
6579 | 0 | xmlParseElementChildrenContentDecl(xmlParserCtxt *ctxt, int inputchk) { |
6580 | | /* stub left for API/ABI compat */ |
6581 | 0 | return(xmlParseElementChildrenContentDeclPriv(ctxt, inputchk, 1)); |
6582 | 0 | } |
6583 | | |
6584 | | /** |
6585 | | * parse the declaration for an Element content either Mixed or Children, |
6586 | | * the cases EMPTY and ANY are handled directly in #xmlParseElementDecl |
6587 | | * |
6588 | | * @deprecated Internal function, don't use. |
6589 | | * |
6590 | | * [46] contentspec ::= 'EMPTY' | 'ANY' | Mixed | children |
6591 | | * |
6592 | | * @param ctxt an XML parser context |
6593 | | * @param name the name of the element being defined. |
6594 | | * @param result the Element Content pointer will be stored here if any |
6595 | | * @returns an xmlElementTypeVal value or -1 on error |
6596 | | */ |
6597 | | |
6598 | | int |
6599 | | xmlParseElementContentDecl(xmlParserCtxt *ctxt, const xmlChar *name, |
6600 | 11.9k | xmlElementContent **result) { |
6601 | | |
6602 | 11.9k | xmlElementContentPtr tree = NULL; |
6603 | 11.9k | int openInputNr = ctxt->inputNr; |
6604 | 11.9k | int res; |
6605 | | |
6606 | 11.9k | *result = NULL; |
6607 | | |
6608 | 11.9k | if (RAW != '(') { |
6609 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, |
6610 | 0 | "xmlParseElementContentDecl : %s '(' expected\n", name); |
6611 | 0 | return(-1); |
6612 | 0 | } |
6613 | 11.9k | NEXT; |
6614 | 11.9k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6615 | 11.9k | if (CMP7(CUR_PTR, '#', 'P', 'C', 'D', 'A', 'T', 'A')) { |
6616 | 1.17k | tree = xmlParseElementMixedContentDecl(ctxt, openInputNr); |
6617 | 1.17k | res = XML_ELEMENT_TYPE_MIXED; |
6618 | 10.7k | } else { |
6619 | 10.7k | tree = xmlParseElementChildrenContentDeclPriv(ctxt, openInputNr, 1); |
6620 | 10.7k | res = XML_ELEMENT_TYPE_ELEMENT; |
6621 | 10.7k | } |
6622 | 11.9k | if (tree == NULL) |
6623 | 10.8k | return(-1); |
6624 | 1.07k | SKIP_BLANKS_PE; |
6625 | 1.07k | *result = tree; |
6626 | 1.07k | return(res); |
6627 | 11.9k | } |
6628 | | |
6629 | | /** |
6630 | | * Parse an element declaration. Always consumes '<!'. |
6631 | | * |
6632 | | * @deprecated Internal function, don't use. |
6633 | | * |
6634 | | * [45] elementdecl ::= '<!ELEMENT' S Name S contentspec S? '>' |
6635 | | * |
6636 | | * [ VC: Unique Element Type Declaration ] |
6637 | | * No element type may be declared more than once |
6638 | | * |
6639 | | * @param ctxt an XML parser context |
6640 | | * @returns the type of the element, or -1 in case of error |
6641 | | */ |
6642 | | int |
6643 | 20.5k | xmlParseElementDecl(xmlParserCtxt *ctxt) { |
6644 | 20.5k | const xmlChar *name; |
6645 | 20.5k | int ret = -1; |
6646 | 20.5k | xmlElementContentPtr content = NULL; |
6647 | | |
6648 | 20.5k | if ((CUR != '<') || (NXT(1) != '!')) |
6649 | 0 | return(ret); |
6650 | 20.5k | SKIP(2); |
6651 | | |
6652 | | /* GROW; done in the caller */ |
6653 | 20.5k | if (CMP7(CUR_PTR, 'E', 'L', 'E', 'M', 'E', 'N', 'T')) { |
6654 | 20.5k | #ifdef LIBXML_VALID_ENABLED |
6655 | 20.5k | int oldInputNr = ctxt->inputNr; |
6656 | 20.5k | #endif |
6657 | | |
6658 | 20.5k | SKIP(7); |
6659 | 20.5k | if (SKIP_BLANKS_PE == 0) { |
6660 | 6.94k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6661 | 6.94k | "Space required after 'ELEMENT'\n"); |
6662 | 6.94k | return(-1); |
6663 | 6.94k | } |
6664 | 13.6k | name = xmlParseName(ctxt); |
6665 | 13.6k | if (name == NULL) { |
6666 | 264 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6667 | 264 | "xmlParseElementDecl: no name for Element\n"); |
6668 | 264 | return(-1); |
6669 | 264 | } |
6670 | 13.3k | if (SKIP_BLANKS_PE == 0) { |
6671 | 11.6k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6672 | 11.6k | "Space required after the element name\n"); |
6673 | 11.6k | } |
6674 | 13.3k | if (CMP5(CUR_PTR, 'E', 'M', 'P', 'T', 'Y')) { |
6675 | 331 | SKIP(5); |
6676 | | /* |
6677 | | * Element must always be empty. |
6678 | | */ |
6679 | 331 | ret = XML_ELEMENT_TYPE_EMPTY; |
6680 | 13.0k | } else if ((RAW == 'A') && (NXT(1) == 'N') && |
6681 | 13.0k | (NXT(2) == 'Y')) { |
6682 | 601 | SKIP(3); |
6683 | | /* |
6684 | | * Element is a generic container. |
6685 | | */ |
6686 | 601 | ret = XML_ELEMENT_TYPE_ANY; |
6687 | 12.4k | } else if (RAW == '(') { |
6688 | 11.9k | ret = xmlParseElementContentDecl(ctxt, name, &content); |
6689 | 11.9k | if (ret <= 0) |
6690 | 10.8k | return(-1); |
6691 | 11.9k | } else { |
6692 | | /* |
6693 | | * [ WFC: PEs in Internal Subset ] error handling. |
6694 | | */ |
6695 | 517 | xmlFatalErrMsg(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, |
6696 | 517 | "xmlParseElementDecl: 'EMPTY', 'ANY' or '(' expected\n"); |
6697 | 517 | return(-1); |
6698 | 517 | } |
6699 | | |
6700 | 2.00k | SKIP_BLANKS_PE; |
6701 | | |
6702 | 2.00k | if (RAW != '>') { |
6703 | 1.28k | xmlFatalErr(ctxt, XML_ERR_GT_REQUIRED, NULL); |
6704 | 1.28k | if (content != NULL) { |
6705 | 722 | xmlFreeDocElementContent(ctxt->myDoc, content); |
6706 | 722 | } |
6707 | 1.28k | } else { |
6708 | 718 | #ifdef LIBXML_VALID_ENABLED |
6709 | 718 | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
6710 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6711 | 0 | "Element declaration doesn't start and stop in" |
6712 | 0 | " the same entity\n", |
6713 | 0 | NULL, NULL); |
6714 | 0 | } |
6715 | 718 | #endif |
6716 | | |
6717 | 718 | NEXT; |
6718 | 718 | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
6719 | 718 | (ctxt->sax->elementDecl != NULL)) { |
6720 | 479 | if (content != NULL) |
6721 | 300 | content->parent = NULL; |
6722 | 479 | ctxt->sax->elementDecl(ctxt->userData, name, ret, |
6723 | 479 | content); |
6724 | 479 | if ((content != NULL) && (content->parent == NULL)) { |
6725 | | /* |
6726 | | * this is a trick: if xmlAddElementDecl is called, |
6727 | | * instead of copying the full tree it is plugged directly |
6728 | | * if called from the parser. Avoid duplicating the |
6729 | | * interfaces or change the API/ABI |
6730 | | */ |
6731 | 154 | xmlFreeDocElementContent(ctxt->myDoc, content); |
6732 | 154 | } |
6733 | 479 | } else if (content != NULL) { |
6734 | 48 | xmlFreeDocElementContent(ctxt->myDoc, content); |
6735 | 48 | } |
6736 | 718 | } |
6737 | 2.00k | } |
6738 | 2.03k | return(ret); |
6739 | 20.5k | } |
6740 | | |
6741 | | /** |
6742 | | * Parse a conditional section. Always consumes '<!['. |
6743 | | * |
6744 | | * [61] conditionalSect ::= includeSect | ignoreSect |
6745 | | * [62] includeSect ::= '<![' S? 'INCLUDE' S? '[' extSubsetDecl ']]>' |
6746 | | * [63] ignoreSect ::= '<![' S? 'IGNORE' S? '[' ignoreSectContents* ']]>' |
6747 | | * [64] ignoreSectContents ::= Ignore ('<![' ignoreSectContents ']]>' |
6748 | | * Ignore)* |
6749 | | * [65] Ignore ::= Char* - (Char* ('<![' | ']]>') Char*) |
6750 | | * @param ctxt an XML parser context |
6751 | | */ |
6752 | | |
6753 | | static void |
6754 | 0 | xmlParseConditionalSections(xmlParserCtxtPtr ctxt) { |
6755 | 0 | size_t depth = 0; |
6756 | 0 | int isFreshPE = 0; |
6757 | 0 | int oldInputNr = ctxt->inputNr; |
6758 | 0 | int declInputNr = ctxt->inputNr; |
6759 | |
|
6760 | 0 | while (!PARSER_STOPPED(ctxt)) { |
6761 | 0 | if (ctxt->input->cur >= ctxt->input->end) { |
6762 | 0 | if (ctxt->inputNr <= oldInputNr) { |
6763 | 0 | xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL); |
6764 | 0 | return; |
6765 | 0 | } |
6766 | | |
6767 | 0 | xmlPopPE(ctxt); |
6768 | 0 | declInputNr = ctxt->inputNr; |
6769 | 0 | } else if ((RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
6770 | 0 | SKIP(3); |
6771 | 0 | SKIP_BLANKS_PE; |
6772 | |
|
6773 | 0 | isFreshPE = 0; |
6774 | |
|
6775 | 0 | if (CMP7(CUR_PTR, 'I', 'N', 'C', 'L', 'U', 'D', 'E')) { |
6776 | 0 | SKIP(7); |
6777 | 0 | SKIP_BLANKS_PE; |
6778 | 0 | if (RAW != '[') { |
6779 | 0 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID, NULL); |
6780 | 0 | return; |
6781 | 0 | } |
6782 | 0 | #ifdef LIBXML_VALID_ENABLED |
6783 | 0 | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6784 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6785 | 0 | "All markup of the conditional section is" |
6786 | 0 | " not in the same entity\n", |
6787 | 0 | NULL, NULL); |
6788 | 0 | } |
6789 | 0 | #endif |
6790 | 0 | NEXT; |
6791 | |
|
6792 | 0 | depth++; |
6793 | 0 | } else if (CMP6(CUR_PTR, 'I', 'G', 'N', 'O', 'R', 'E')) { |
6794 | 0 | size_t ignoreDepth = 0; |
6795 | |
|
6796 | 0 | SKIP(6); |
6797 | 0 | SKIP_BLANKS_PE; |
6798 | 0 | if (RAW != '[') { |
6799 | 0 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID, NULL); |
6800 | 0 | return; |
6801 | 0 | } |
6802 | 0 | #ifdef LIBXML_VALID_ENABLED |
6803 | 0 | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6804 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6805 | 0 | "All markup of the conditional section is" |
6806 | 0 | " not in the same entity\n", |
6807 | 0 | NULL, NULL); |
6808 | 0 | } |
6809 | 0 | #endif |
6810 | 0 | NEXT; |
6811 | |
|
6812 | 0 | while (PARSER_STOPPED(ctxt) == 0) { |
6813 | 0 | if (RAW == 0) { |
6814 | 0 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_NOT_FINISHED, NULL); |
6815 | 0 | return; |
6816 | 0 | } |
6817 | 0 | if ((RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
6818 | 0 | SKIP(3); |
6819 | 0 | ignoreDepth++; |
6820 | | /* Check for integer overflow */ |
6821 | 0 | if (ignoreDepth == 0) { |
6822 | 0 | xmlErrMemory(ctxt); |
6823 | 0 | return; |
6824 | 0 | } |
6825 | 0 | } else if ((RAW == ']') && (NXT(1) == ']') && |
6826 | 0 | (NXT(2) == '>')) { |
6827 | 0 | SKIP(3); |
6828 | 0 | if (ignoreDepth == 0) |
6829 | 0 | break; |
6830 | 0 | ignoreDepth--; |
6831 | 0 | } else { |
6832 | 0 | NEXT; |
6833 | 0 | } |
6834 | 0 | } |
6835 | | |
6836 | 0 | #ifdef LIBXML_VALID_ENABLED |
6837 | 0 | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6838 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6839 | 0 | "All markup of the conditional section is" |
6840 | 0 | " not in the same entity\n", |
6841 | 0 | NULL, NULL); |
6842 | 0 | } |
6843 | 0 | #endif |
6844 | 0 | } else { |
6845 | 0 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID_KEYWORD, NULL); |
6846 | 0 | return; |
6847 | 0 | } |
6848 | 0 | } else if ((depth > 0) && |
6849 | 0 | (RAW == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { |
6850 | 0 | if (isFreshPE) { |
6851 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_CONDSEC_INVALID, |
6852 | 0 | "Parameter entity must match " |
6853 | 0 | "extSubsetDecl\n"); |
6854 | 0 | return; |
6855 | 0 | } |
6856 | | |
6857 | 0 | depth--; |
6858 | 0 | #ifdef LIBXML_VALID_ENABLED |
6859 | 0 | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6860 | 0 | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6861 | 0 | "All markup of the conditional section is not" |
6862 | 0 | " in the same entity\n", |
6863 | 0 | NULL, NULL); |
6864 | 0 | } |
6865 | 0 | #endif |
6866 | 0 | SKIP(3); |
6867 | 0 | } else if ((RAW == '<') && ((NXT(1) == '!') || (NXT(1) == '?'))) { |
6868 | 0 | isFreshPE = 0; |
6869 | 0 | xmlParseMarkupDecl(ctxt); |
6870 | 0 | } else if (RAW == '%') { |
6871 | 0 | xmlParsePERefInternal(ctxt, 1); |
6872 | 0 | if (ctxt->inputNr > declInputNr) { |
6873 | 0 | isFreshPE = 1; |
6874 | 0 | declInputNr = ctxt->inputNr; |
6875 | 0 | } |
6876 | 0 | } else { |
6877 | 0 | xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL); |
6878 | 0 | return; |
6879 | 0 | } |
6880 | | |
6881 | 0 | if (depth == 0) |
6882 | 0 | break; |
6883 | | |
6884 | 0 | SKIP_BLANKS; |
6885 | 0 | SHRINK; |
6886 | 0 | GROW; |
6887 | 0 | } |
6888 | 0 | } |
6889 | | |
6890 | | /** |
6891 | | * Parse markup declarations. Always consumes '<!' or '<?'. |
6892 | | * |
6893 | | * @deprecated Internal function, don't use. |
6894 | | * |
6895 | | * [29] markupdecl ::= elementdecl | AttlistDecl | EntityDecl | |
6896 | | * NotationDecl | PI | Comment |
6897 | | * |
6898 | | * [ VC: Proper Declaration/PE Nesting ] |
6899 | | * Parameter-entity replacement text must be properly nested with |
6900 | | * markup declarations. That is to say, if either the first character |
6901 | | * or the last character of a markup declaration (markupdecl above) is |
6902 | | * contained in the replacement text for a parameter-entity reference, |
6903 | | * both must be contained in the same replacement text. |
6904 | | * |
6905 | | * [ WFC: PEs in Internal Subset ] |
6906 | | * In the internal DTD subset, parameter-entity references can occur |
6907 | | * only where markup declarations can occur, not within markup declarations. |
6908 | | * (This does not apply to references that occur in external parameter |
6909 | | * entities or to the external subset.) |
6910 | | * |
6911 | | * @param ctxt an XML parser context |
6912 | | */ |
6913 | | void |
6914 | 106k | xmlParseMarkupDecl(xmlParserCtxt *ctxt) { |
6915 | 106k | GROW; |
6916 | 106k | if (CUR == '<') { |
6917 | 106k | if (NXT(1) == '!') { |
6918 | 84.2k | switch (NXT(2)) { |
6919 | 45.4k | case 'E': |
6920 | 45.4k | if (NXT(3) == 'L') |
6921 | 20.5k | xmlParseElementDecl(ctxt); |
6922 | 24.8k | else if (NXT(3) == 'N') |
6923 | 24.8k | xmlParseEntityDecl(ctxt); |
6924 | 16 | else |
6925 | 16 | SKIP(2); |
6926 | 45.4k | break; |
6927 | 24.8k | case 'A': |
6928 | 24.8k | xmlParseAttributeListDecl(ctxt); |
6929 | 24.8k | break; |
6930 | 8.84k | case 'N': |
6931 | 8.84k | xmlParseNotationDecl(ctxt); |
6932 | 8.84k | break; |
6933 | 2.88k | case '-': |
6934 | 2.88k | xmlParseComment(ctxt); |
6935 | 2.88k | break; |
6936 | 2.27k | default: |
6937 | 2.27k | xmlFatalErr(ctxt, |
6938 | 2.27k | ctxt->inSubset == 2 ? |
6939 | 0 | XML_ERR_EXT_SUBSET_NOT_FINISHED : |
6940 | 2.27k | XML_ERR_INT_SUBSET_NOT_FINISHED, |
6941 | 2.27k | NULL); |
6942 | 2.27k | SKIP(2); |
6943 | 2.27k | break; |
6944 | 84.2k | } |
6945 | 84.2k | } else if (NXT(1) == '?') { |
6946 | 22.5k | xmlParsePI(ctxt); |
6947 | 22.5k | } |
6948 | 106k | } |
6949 | 106k | } |
6950 | | |
6951 | | /** |
6952 | | * parse an XML declaration header for external entities |
6953 | | * |
6954 | | * @deprecated Internal function, don't use. |
6955 | | * |
6956 | | * [77] TextDecl ::= '<?xml' VersionInfo? EncodingDecl S? '?>' |
6957 | | * @param ctxt an XML parser context |
6958 | | */ |
6959 | | |
6960 | | void |
6961 | 0 | xmlParseTextDecl(xmlParserCtxt *ctxt) { |
6962 | 0 | xmlChar *version; |
6963 | | |
6964 | | /* |
6965 | | * We know that '<?xml' is here. |
6966 | | */ |
6967 | 0 | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
6968 | 0 | SKIP(5); |
6969 | 0 | } else { |
6970 | 0 | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_STARTED, NULL); |
6971 | 0 | return; |
6972 | 0 | } |
6973 | | |
6974 | 0 | if (SKIP_BLANKS == 0) { |
6975 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6976 | 0 | "Space needed after '<?xml'\n"); |
6977 | 0 | } |
6978 | | |
6979 | | /* |
6980 | | * We may have the VersionInfo here. |
6981 | | */ |
6982 | 0 | version = xmlParseVersionInfo(ctxt); |
6983 | 0 | if (version == NULL) { |
6984 | 0 | version = xmlCharStrdup(XML_DEFAULT_VERSION); |
6985 | 0 | if (version == NULL) { |
6986 | 0 | xmlErrMemory(ctxt); |
6987 | 0 | return; |
6988 | 0 | } |
6989 | 0 | } else { |
6990 | 0 | if (SKIP_BLANKS == 0) { |
6991 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6992 | 0 | "Space needed here\n"); |
6993 | 0 | } |
6994 | 0 | } |
6995 | 0 | ctxt->input->version = version; |
6996 | | |
6997 | | /* |
6998 | | * We must have the encoding declaration |
6999 | | */ |
7000 | 0 | xmlParseEncodingDecl(ctxt); |
7001 | |
|
7002 | 0 | SKIP_BLANKS; |
7003 | 0 | if ((RAW == '?') && (NXT(1) == '>')) { |
7004 | 0 | SKIP(2); |
7005 | 0 | } else if (RAW == '>') { |
7006 | | /* Deprecated old WD ... */ |
7007 | 0 | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
7008 | 0 | NEXT; |
7009 | 0 | } else { |
7010 | 0 | int c; |
7011 | |
|
7012 | 0 | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
7013 | 0 | while ((PARSER_STOPPED(ctxt) == 0) && ((c = CUR) != 0)) { |
7014 | 0 | NEXT; |
7015 | 0 | if (c == '>') |
7016 | 0 | break; |
7017 | 0 | } |
7018 | 0 | } |
7019 | 0 | } |
7020 | | |
7021 | | /** |
7022 | | * parse Markup declarations from an external subset |
7023 | | * |
7024 | | * @deprecated Internal function, don't use. |
7025 | | * |
7026 | | * [30] extSubset ::= textDecl? extSubsetDecl |
7027 | | * |
7028 | | * [31] extSubsetDecl ::= (markupdecl | conditionalSect | |
7029 | | * PEReference | S) * |
7030 | | * @param ctxt an XML parser context |
7031 | | * @param publicId the public identifier |
7032 | | * @param systemId the system identifier (URL) |
7033 | | */ |
7034 | | void |
7035 | | xmlParseExternalSubset(xmlParserCtxt *ctxt, const xmlChar *publicId, |
7036 | 0 | const xmlChar *systemId) { |
7037 | 0 | int oldInputNr; |
7038 | |
|
7039 | 0 | xmlCtxtInitializeLate(ctxt); |
7040 | |
|
7041 | 0 | xmlDetectEncoding(ctxt); |
7042 | |
|
7043 | 0 | if (CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) { |
7044 | 0 | xmlParseTextDecl(ctxt); |
7045 | 0 | } |
7046 | 0 | if (ctxt->myDoc == NULL) { |
7047 | 0 | ctxt->myDoc = xmlNewDoc(BAD_CAST "1.0"); |
7048 | 0 | if (ctxt->myDoc == NULL) { |
7049 | 0 | xmlErrMemory(ctxt); |
7050 | 0 | return; |
7051 | 0 | } |
7052 | 0 | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
7053 | 0 | } |
7054 | 0 | if ((ctxt->myDoc->intSubset == NULL) && |
7055 | 0 | (xmlCreateIntSubset(ctxt->myDoc, NULL, publicId, systemId) == NULL)) { |
7056 | 0 | xmlErrMemory(ctxt); |
7057 | 0 | } |
7058 | |
|
7059 | 0 | ctxt->inSubset = 2; |
7060 | 0 | oldInputNr = ctxt->inputNr; |
7061 | |
|
7062 | 0 | SKIP_BLANKS; |
7063 | 0 | while (!PARSER_STOPPED(ctxt)) { |
7064 | 0 | if (ctxt->input->cur >= ctxt->input->end) { |
7065 | 0 | if (ctxt->inputNr <= oldInputNr) { |
7066 | 0 | xmlParserCheckEOF(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED); |
7067 | 0 | break; |
7068 | 0 | } |
7069 | | |
7070 | 0 | xmlPopPE(ctxt); |
7071 | 0 | } else if ((RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
7072 | 0 | xmlParseConditionalSections(ctxt); |
7073 | 0 | } else if ((RAW == '<') && ((NXT(1) == '!') || (NXT(1) == '?'))) { |
7074 | 0 | xmlParseMarkupDecl(ctxt); |
7075 | 0 | } else if (RAW == '%') { |
7076 | 0 | xmlParsePERefInternal(ctxt, 1); |
7077 | 0 | } else { |
7078 | 0 | xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL); |
7079 | |
|
7080 | 0 | while (ctxt->inputNr > oldInputNr) |
7081 | 0 | xmlPopPE(ctxt); |
7082 | 0 | break; |
7083 | 0 | } |
7084 | 0 | SKIP_BLANKS; |
7085 | 0 | SHRINK; |
7086 | 0 | GROW; |
7087 | 0 | } |
7088 | 0 | } |
7089 | | |
7090 | | /** |
7091 | | * parse and handle entity references in content, depending on the SAX |
7092 | | * interface, this may end-up in a call to character() if this is a |
7093 | | * CharRef, a predefined entity, if there is no reference() callback. |
7094 | | * or if the parser was asked to switch to that mode. |
7095 | | * |
7096 | | * @deprecated Internal function, don't use. |
7097 | | * |
7098 | | * Always consumes '&'. |
7099 | | * |
7100 | | * [67] Reference ::= EntityRef | CharRef |
7101 | | * @param ctxt an XML parser context |
7102 | | */ |
7103 | | void |
7104 | 166k | xmlParseReference(xmlParserCtxt *ctxt) { |
7105 | 166k | xmlEntityPtr ent = NULL; |
7106 | 166k | const xmlChar *name; |
7107 | 166k | xmlChar *val; |
7108 | | |
7109 | 166k | if (RAW != '&') |
7110 | 0 | return; |
7111 | | |
7112 | | /* |
7113 | | * Simple case of a CharRef |
7114 | | */ |
7115 | 166k | if (NXT(1) == '#') { |
7116 | 50.2k | int i = 0; |
7117 | 50.2k | xmlChar out[16]; |
7118 | 50.2k | int value = xmlParseCharRef(ctxt); |
7119 | | |
7120 | 50.2k | if (value == 0) |
7121 | 16.1k | return; |
7122 | | |
7123 | | /* |
7124 | | * Just encode the value in UTF-8 |
7125 | | */ |
7126 | 34.1k | COPY_BUF(out, i, value); |
7127 | 34.1k | out[i] = 0; |
7128 | 34.1k | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL) && |
7129 | 34.1k | (!ctxt->disableSAX)) |
7130 | 386 | ctxt->sax->characters(ctxt->userData, out, i); |
7131 | 34.1k | return; |
7132 | 50.2k | } |
7133 | | |
7134 | | /* |
7135 | | * We are seeing an entity reference |
7136 | | */ |
7137 | 116k | name = xmlParseEntityRefInternal(ctxt); |
7138 | 116k | if (name == NULL) |
7139 | 66.8k | return; |
7140 | 49.3k | ent = xmlLookupGeneralEntity(ctxt, name, /* isAttr */ 0); |
7141 | 49.3k | if (ent == NULL) { |
7142 | | /* |
7143 | | * Create a reference for undeclared entities. |
7144 | | */ |
7145 | 23.7k | if ((ctxt->replaceEntities == 0) && |
7146 | 23.7k | (ctxt->sax != NULL) && |
7147 | 23.7k | (ctxt->disableSAX == 0) && |
7148 | 23.7k | (ctxt->sax->reference != NULL)) { |
7149 | 0 | ctxt->sax->reference(ctxt->userData, name); |
7150 | 0 | } |
7151 | 23.7k | return; |
7152 | 23.7k | } |
7153 | 25.6k | if (!ctxt->wellFormed) |
7154 | 13.0k | return; |
7155 | | |
7156 | | /* special case of predefined entities */ |
7157 | 12.5k | if ((ent->name == NULL) || |
7158 | 12.5k | (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { |
7159 | 1.72k | val = ent->content; |
7160 | 1.72k | if (val == NULL) return; |
7161 | | /* |
7162 | | * inline the entity. |
7163 | | */ |
7164 | 1.72k | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL) && |
7165 | 1.72k | (!ctxt->disableSAX)) |
7166 | 1.72k | ctxt->sax->characters(ctxt->userData, val, xmlStrlen(val)); |
7167 | 1.72k | return; |
7168 | 1.72k | } |
7169 | | |
7170 | | /* |
7171 | | * Some users try to parse entities on their own and used to set |
7172 | | * the renamed "checked" member. Fix the flags to cover this |
7173 | | * case. |
7174 | | */ |
7175 | 10.8k | if (((ent->flags & XML_ENT_PARSED) == 0) && (ent->children != NULL)) |
7176 | 0 | ent->flags |= XML_ENT_PARSED; |
7177 | | |
7178 | | /* |
7179 | | * The first reference to the entity trigger a parsing phase |
7180 | | * where the ent->children is filled with the result from |
7181 | | * the parsing. |
7182 | | * Note: external parsed entities will not be loaded, it is not |
7183 | | * required for a non-validating parser, unless the parsing option |
7184 | | * of validating, or substituting entities were given. Doing so is |
7185 | | * far more secure as the parser will only process data coming from |
7186 | | * the document entity by default. |
7187 | | * |
7188 | | * FIXME: This doesn't work correctly since entities can be |
7189 | | * expanded with different namespace declarations in scope. |
7190 | | * For example: |
7191 | | * |
7192 | | * <!DOCTYPE doc [ |
7193 | | * <!ENTITY ent "<ns:elem/>"> |
7194 | | * ]> |
7195 | | * <doc> |
7196 | | * <decl1 xmlns:ns="urn:ns1"> |
7197 | | * &ent; |
7198 | | * </decl1> |
7199 | | * <decl2 xmlns:ns="urn:ns2"> |
7200 | | * &ent; |
7201 | | * </decl2> |
7202 | | * </doc> |
7203 | | * |
7204 | | * Proposed fix: |
7205 | | * |
7206 | | * - Ignore current namespace declarations when parsing the |
7207 | | * entity. If a prefix can't be resolved, don't report an error |
7208 | | * but mark it as unresolved. |
7209 | | * - Try to resolve these prefixes when expanding the entity. |
7210 | | * This will require a specialized version of xmlStaticCopyNode |
7211 | | * which can also make use of the namespace hash table to avoid |
7212 | | * quadratic behavior. |
7213 | | * |
7214 | | * Alternatively, we could simply reparse the entity on each |
7215 | | * expansion like we already do with custom SAX callbacks. |
7216 | | * External entity content should be cached in this case. |
7217 | | */ |
7218 | 10.8k | if ((ent->etype == XML_INTERNAL_GENERAL_ENTITY) || |
7219 | 10.8k | (((ctxt->options & XML_PARSE_NO_XXE) == 0) && |
7220 | 203 | ((ctxt->replaceEntities) || |
7221 | 10.6k | (ctxt->validate)))) { |
7222 | 10.6k | if ((ent->flags & XML_ENT_PARSED) == 0) { |
7223 | 583 | xmlCtxtParseEntity(ctxt, ent); |
7224 | 10.0k | } else if (ent->children == NULL) { |
7225 | | /* |
7226 | | * Probably running in SAX mode and the callbacks don't |
7227 | | * build the entity content. Parse the entity again. |
7228 | | * |
7229 | | * This will also be triggered in normal tree builder mode |
7230 | | * if an entity happens to be empty, causing unnecessary |
7231 | | * reloads. It's hard to come up with a reliable check in |
7232 | | * which mode we're running. |
7233 | | */ |
7234 | 6.60k | xmlCtxtParseEntity(ctxt, ent); |
7235 | 6.60k | } |
7236 | 10.6k | } |
7237 | | |
7238 | | /* |
7239 | | * We also check for amplification if entities aren't substituted. |
7240 | | * They might be expanded later. |
7241 | | */ |
7242 | 10.8k | if (xmlParserEntityCheck(ctxt, ent->expandedSize)) |
7243 | 39 | return; |
7244 | | |
7245 | 10.8k | if ((ctxt->sax == NULL) || (ctxt->disableSAX)) |
7246 | 141 | return; |
7247 | | |
7248 | 10.6k | if (ctxt->replaceEntities == 0) { |
7249 | | /* |
7250 | | * Create a reference |
7251 | | */ |
7252 | 0 | if (ctxt->sax->reference != NULL) |
7253 | 0 | ctxt->sax->reference(ctxt->userData, ent->name); |
7254 | 10.6k | } else if ((ent->children != NULL) && (ctxt->node != NULL)) { |
7255 | 3.83k | xmlNodePtr copy, cur; |
7256 | | |
7257 | | /* |
7258 | | * Seems we are generating the DOM content, copy the tree |
7259 | | */ |
7260 | 3.83k | cur = ent->children; |
7261 | | |
7262 | | /* |
7263 | | * Handle first text node with SAX to coalesce text efficiently |
7264 | | */ |
7265 | 3.83k | if ((cur->type == XML_TEXT_NODE) || |
7266 | 3.83k | (cur->type == XML_CDATA_SECTION_NODE)) { |
7267 | 3.07k | int len = xmlStrlen(cur->content); |
7268 | | |
7269 | 3.07k | if ((cur->type == XML_TEXT_NODE) || |
7270 | 3.07k | (ctxt->options & XML_PARSE_NOCDATA)) { |
7271 | 2.80k | if (ctxt->sax->characters != NULL) |
7272 | 2.80k | ctxt->sax->characters(ctxt, cur->content, len); |
7273 | 2.80k | } else { |
7274 | 269 | if (ctxt->sax->cdataBlock != NULL) |
7275 | 269 | ctxt->sax->cdataBlock(ctxt, cur->content, len); |
7276 | 269 | } |
7277 | | |
7278 | 3.07k | cur = cur->next; |
7279 | 3.07k | } |
7280 | | |
7281 | 10.1k | while (cur != NULL) { |
7282 | 8.53k | xmlNodePtr last; |
7283 | | |
7284 | | /* |
7285 | | * Handle last text node with SAX to coalesce text efficiently |
7286 | | */ |
7287 | 8.53k | if ((cur->next == NULL) && |
7288 | 8.53k | ((cur->type == XML_TEXT_NODE) || |
7289 | 2.62k | (cur->type == XML_CDATA_SECTION_NODE))) { |
7290 | 2.19k | int len = xmlStrlen(cur->content); |
7291 | | |
7292 | 2.19k | if ((cur->type == XML_TEXT_NODE) || |
7293 | 2.19k | (ctxt->options & XML_PARSE_NOCDATA)) { |
7294 | 2.03k | if (ctxt->sax->characters != NULL) |
7295 | 2.03k | ctxt->sax->characters(ctxt, cur->content, len); |
7296 | 2.03k | } else { |
7297 | 162 | if (ctxt->sax->cdataBlock != NULL) |
7298 | 162 | ctxt->sax->cdataBlock(ctxt, cur->content, len); |
7299 | 162 | } |
7300 | | |
7301 | 2.19k | break; |
7302 | 2.19k | } |
7303 | | |
7304 | | /* |
7305 | | * Reset coalesce buffer stats only for non-text nodes. |
7306 | | */ |
7307 | 6.33k | ctxt->nodemem = 0; |
7308 | 6.33k | ctxt->nodelen = 0; |
7309 | | |
7310 | 6.33k | copy = xmlDocCopyNode(cur, ctxt->myDoc, 1); |
7311 | | |
7312 | 6.33k | if (copy == NULL) { |
7313 | 0 | xmlErrMemory(ctxt); |
7314 | 0 | break; |
7315 | 0 | } |
7316 | | |
7317 | 6.33k | if (ctxt->parseMode == XML_PARSE_READER) { |
7318 | | /* Needed for reader */ |
7319 | 0 | copy->extra = cur->extra; |
7320 | | /* Maybe needed for reader */ |
7321 | 0 | copy->_private = cur->_private; |
7322 | 0 | } |
7323 | | |
7324 | 6.33k | copy->parent = ctxt->node; |
7325 | 6.33k | last = ctxt->node->last; |
7326 | 6.33k | if (last == NULL) { |
7327 | 248 | ctxt->node->children = copy; |
7328 | 6.09k | } else { |
7329 | 6.09k | last->next = copy; |
7330 | 6.09k | copy->prev = last; |
7331 | 6.09k | } |
7332 | 6.33k | ctxt->node->last = copy; |
7333 | | |
7334 | 6.33k | cur = cur->next; |
7335 | 6.33k | } |
7336 | 3.83k | } |
7337 | 10.6k | } |
7338 | | |
7339 | | static void |
7340 | 123k | xmlHandleUndeclaredEntity(xmlParserCtxtPtr ctxt, const xmlChar *name) { |
7341 | | /* |
7342 | | * [ WFC: Entity Declared ] |
7343 | | * In a document without any DTD, a document with only an |
7344 | | * internal DTD subset which contains no parameter entity |
7345 | | * references, or a document with "standalone='yes'", the |
7346 | | * Name given in the entity reference must match that in an |
7347 | | * entity declaration, except that well-formed documents |
7348 | | * need not declare any of the following entities: amp, lt, |
7349 | | * gt, apos, quot. |
7350 | | * The declaration of a parameter entity must precede any |
7351 | | * reference to it. |
7352 | | * Similarly, the declaration of a general entity must |
7353 | | * precede any reference to it which appears in a default |
7354 | | * value in an attribute-list declaration. Note that if |
7355 | | * entities are declared in the external subset or in |
7356 | | * external parameter entities, a non-validating processor |
7357 | | * is not obligated to read and process their declarations; |
7358 | | * for such documents, the rule that an entity must be |
7359 | | * declared is a well-formedness constraint only if |
7360 | | * standalone='yes'. |
7361 | | */ |
7362 | 123k | if ((ctxt->standalone == 1) || |
7363 | 123k | ((ctxt->hasExternalSubset == 0) && |
7364 | 123k | (ctxt->hasPErefs == 0))) { |
7365 | 84.5k | xmlFatalErrMsgStr(ctxt, XML_ERR_UNDECLARED_ENTITY, |
7366 | 84.5k | "Entity '%s' not defined\n", name); |
7367 | 84.5k | #ifdef LIBXML_VALID_ENABLED |
7368 | 84.5k | } else if (ctxt->validate) { |
7369 | | /* |
7370 | | * [ VC: Entity Declared ] |
7371 | | * In a document with an external subset or external |
7372 | | * parameter entities with "standalone='no'", ... |
7373 | | * ... The declaration of a parameter entity must |
7374 | | * precede any reference to it... |
7375 | | */ |
7376 | 0 | xmlValidityError(ctxt, XML_ERR_UNDECLARED_ENTITY, |
7377 | 0 | "Entity '%s' not defined\n", name, NULL); |
7378 | 0 | #endif |
7379 | 39.3k | } else if ((ctxt->loadsubset & ~XML_SKIP_IDS) || |
7380 | 39.3k | ((ctxt->replaceEntities) && |
7381 | 39.3k | ((ctxt->options & XML_PARSE_NO_XXE) == 0))) { |
7382 | | /* |
7383 | | * Also raise a non-fatal error |
7384 | | * |
7385 | | * - if the external subset is loaded and all entity declarations |
7386 | | * should be available, or |
7387 | | * - entity substition was requested without restricting |
7388 | | * external entity access. |
7389 | | */ |
7390 | 39.3k | xmlErrMsgStr(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7391 | 39.3k | "Entity '%s' not defined\n", name); |
7392 | 39.3k | } else { |
7393 | 0 | xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7394 | 0 | "Entity '%s' not defined\n", name, NULL); |
7395 | 0 | } |
7396 | | |
7397 | 123k | ctxt->valid = 0; |
7398 | 123k | } |
7399 | | |
7400 | | static xmlEntityPtr |
7401 | 540k | xmlLookupGeneralEntity(xmlParserCtxtPtr ctxt, const xmlChar *name, int inAttr) { |
7402 | 540k | xmlEntityPtr ent = NULL; |
7403 | | |
7404 | | /* |
7405 | | * Predefined entities override any extra definition |
7406 | | */ |
7407 | 540k | if ((ctxt->options & XML_PARSE_OLDSAX) == 0) { |
7408 | 540k | ent = xmlGetPredefinedEntity(name); |
7409 | 540k | if (ent != NULL) |
7410 | 116k | return(ent); |
7411 | 540k | } |
7412 | | |
7413 | | /* |
7414 | | * Ask first SAX for entity resolution, otherwise try the |
7415 | | * entities which may have stored in the parser context. |
7416 | | */ |
7417 | 423k | if (ctxt->sax != NULL) { |
7418 | 423k | if (ctxt->sax->getEntity != NULL) |
7419 | 423k | ent = ctxt->sax->getEntity(ctxt->userData, name); |
7420 | 423k | if ((ctxt->wellFormed == 1 ) && (ent == NULL) && |
7421 | 423k | (ctxt->options & XML_PARSE_OLDSAX)) |
7422 | 0 | ent = xmlGetPredefinedEntity(name); |
7423 | 423k | if ((ctxt->wellFormed == 1 ) && (ent == NULL) && |
7424 | 423k | (ctxt->userData==ctxt)) { |
7425 | 23.8k | ent = xmlSAX2GetEntity(ctxt, name); |
7426 | 23.8k | } |
7427 | 423k | } |
7428 | | |
7429 | 423k | if (ent == NULL) { |
7430 | 117k | xmlHandleUndeclaredEntity(ctxt, name); |
7431 | 117k | } |
7432 | | |
7433 | | /* |
7434 | | * [ WFC: Parsed Entity ] |
7435 | | * An entity reference must not contain the name of an |
7436 | | * unparsed entity |
7437 | | */ |
7438 | 306k | else if (ent->etype == XML_EXTERNAL_GENERAL_UNPARSED_ENTITY) { |
7439 | 198 | xmlFatalErrMsgStr(ctxt, XML_ERR_UNPARSED_ENTITY, |
7440 | 198 | "Entity reference to unparsed entity %s\n", name); |
7441 | 198 | ent = NULL; |
7442 | 198 | } |
7443 | | |
7444 | | /* |
7445 | | * [ WFC: No External Entity References ] |
7446 | | * Attribute values cannot contain direct or indirect |
7447 | | * entity references to external entities. |
7448 | | */ |
7449 | 306k | else if (ent->etype == XML_EXTERNAL_GENERAL_PARSED_ENTITY) { |
7450 | 747 | if (inAttr) { |
7451 | 237 | xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_IS_EXTERNAL, |
7452 | 237 | "Attribute references external entity '%s'\n", name); |
7453 | 237 | ent = NULL; |
7454 | 237 | } |
7455 | 747 | } |
7456 | | |
7457 | 423k | return(ent); |
7458 | 540k | } |
7459 | | |
7460 | | /** |
7461 | | * Parse an entity reference. Always consumes '&'. |
7462 | | * |
7463 | | * [68] EntityRef ::= '&' Name ';' |
7464 | | * |
7465 | | * @param ctxt an XML parser context |
7466 | | * @returns the name, or NULL in case of error. |
7467 | | */ |
7468 | | static const xmlChar * |
7469 | 286k | xmlParseEntityRefInternal(xmlParserCtxtPtr ctxt) { |
7470 | 286k | const xmlChar *name; |
7471 | | |
7472 | 286k | GROW; |
7473 | | |
7474 | 286k | if (RAW != '&') |
7475 | 0 | return(NULL); |
7476 | 286k | NEXT; |
7477 | 286k | name = xmlParseName(ctxt); |
7478 | 286k | if (name == NULL) { |
7479 | 64.2k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7480 | 64.2k | "xmlParseEntityRef: no name\n"); |
7481 | 64.2k | return(NULL); |
7482 | 64.2k | } |
7483 | 222k | if (RAW != ';') { |
7484 | 41.0k | xmlFatalErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING, NULL); |
7485 | 41.0k | return(NULL); |
7486 | 41.0k | } |
7487 | 181k | NEXT; |
7488 | | |
7489 | 181k | return(name); |
7490 | 222k | } |
7491 | | |
7492 | | /** |
7493 | | * @deprecated Internal function, don't use. |
7494 | | * |
7495 | | * @param ctxt an XML parser context |
7496 | | * @returns the xmlEntity if found, or NULL otherwise. |
7497 | | */ |
7498 | | xmlEntity * |
7499 | 0 | xmlParseEntityRef(xmlParserCtxt *ctxt) { |
7500 | 0 | const xmlChar *name; |
7501 | |
|
7502 | 0 | if (ctxt == NULL) |
7503 | 0 | return(NULL); |
7504 | | |
7505 | 0 | name = xmlParseEntityRefInternal(ctxt); |
7506 | 0 | if (name == NULL) |
7507 | 0 | return(NULL); |
7508 | | |
7509 | 0 | return(xmlLookupGeneralEntity(ctxt, name, /* inAttr */ 0)); |
7510 | 0 | } |
7511 | | |
7512 | | /** |
7513 | | * parse ENTITY references declarations, but this version parses it from |
7514 | | * a string value. |
7515 | | * |
7516 | | * [68] EntityRef ::= '&' Name ';' |
7517 | | * |
7518 | | * [ WFC: Entity Declared ] |
7519 | | * In a document without any DTD, a document with only an internal DTD |
7520 | | * subset which contains no parameter entity references, or a document |
7521 | | * with "standalone='yes'", the Name given in the entity reference |
7522 | | * must match that in an entity declaration, except that well-formed |
7523 | | * documents need not declare any of the following entities: amp, lt, |
7524 | | * gt, apos, quot. The declaration of a parameter entity must precede |
7525 | | * any reference to it. Similarly, the declaration of a general entity |
7526 | | * must precede any reference to it which appears in a default value in an |
7527 | | * attribute-list declaration. Note that if entities are declared in the |
7528 | | * external subset or in external parameter entities, a non-validating |
7529 | | * processor is not obligated to read and process their declarations; |
7530 | | * for such documents, the rule that an entity must be declared is a |
7531 | | * well-formedness constraint only if standalone='yes'. |
7532 | | * |
7533 | | * [ WFC: Parsed Entity ] |
7534 | | * An entity reference must not contain the name of an unparsed entity |
7535 | | * |
7536 | | * @param ctxt an XML parser context |
7537 | | * @param str a pointer to an index in the string |
7538 | | * @returns the xmlEntity if found, or NULL otherwise. The str pointer |
7539 | | * is updated to the current location in the string. |
7540 | | */ |
7541 | | static xmlChar * |
7542 | 359k | xmlParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar ** str) { |
7543 | 359k | xmlChar *name; |
7544 | 359k | const xmlChar *ptr; |
7545 | 359k | xmlChar cur; |
7546 | | |
7547 | 359k | if ((str == NULL) || (*str == NULL)) |
7548 | 0 | return(NULL); |
7549 | 359k | ptr = *str; |
7550 | 359k | cur = *ptr; |
7551 | 359k | if (cur != '&') |
7552 | 0 | return(NULL); |
7553 | | |
7554 | 359k | ptr++; |
7555 | 359k | name = xmlParseStringName(ctxt, &ptr); |
7556 | 359k | if (name == NULL) { |
7557 | 4 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7558 | 4 | "xmlParseStringEntityRef: no name\n"); |
7559 | 4 | *str = ptr; |
7560 | 4 | return(NULL); |
7561 | 4 | } |
7562 | 359k | if (*ptr != ';') { |
7563 | 2 | xmlFatalErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING, NULL); |
7564 | 2 | xmlFree(name); |
7565 | 2 | *str = ptr; |
7566 | 2 | return(NULL); |
7567 | 2 | } |
7568 | 359k | ptr++; |
7569 | | |
7570 | 359k | *str = ptr; |
7571 | 359k | return(name); |
7572 | 359k | } |
7573 | | |
7574 | | /** |
7575 | | * Parse a parameter entity reference. Always consumes '%'. |
7576 | | * |
7577 | | * The entity content is handled directly by pushing it's content as |
7578 | | * a new input stream. |
7579 | | * |
7580 | | * [69] PEReference ::= '%' Name ';' |
7581 | | * |
7582 | | * [ WFC: No Recursion ] |
7583 | | * A parsed entity must not contain a recursive |
7584 | | * reference to itself, either directly or indirectly. |
7585 | | * |
7586 | | * [ WFC: Entity Declared ] |
7587 | | * In a document without any DTD, a document with only an internal DTD |
7588 | | * subset which contains no parameter entity references, or a document |
7589 | | * with "standalone='yes'", ... ... The declaration of a parameter |
7590 | | * entity must precede any reference to it... |
7591 | | * |
7592 | | * [ VC: Entity Declared ] |
7593 | | * In a document with an external subset or external parameter entities |
7594 | | * with "standalone='no'", ... ... The declaration of a parameter entity |
7595 | | * must precede any reference to it... |
7596 | | * |
7597 | | * [ WFC: In DTD ] |
7598 | | * Parameter-entity references may only appear in the DTD. |
7599 | | * NOTE: misleading but this is handled. |
7600 | | * |
7601 | | * @param ctxt an XML parser context |
7602 | | * @param markupDecl whether the PERef starts a markup declaration |
7603 | | */ |
7604 | | static void |
7605 | 44.7k | xmlParsePERefInternal(xmlParserCtxt *ctxt, int markupDecl) { |
7606 | 44.7k | const xmlChar *name; |
7607 | 44.7k | xmlEntityPtr entity = NULL; |
7608 | 44.7k | xmlParserInputPtr input; |
7609 | | |
7610 | 44.7k | if (RAW != '%') |
7611 | 0 | return; |
7612 | 44.7k | NEXT; |
7613 | 44.7k | name = xmlParseName(ctxt); |
7614 | 44.7k | if (name == NULL) { |
7615 | 2.41k | xmlFatalErrMsg(ctxt, XML_ERR_PEREF_NO_NAME, "PEReference: no name\n"); |
7616 | 2.41k | return; |
7617 | 2.41k | } |
7618 | 42.3k | if (RAW != ';') { |
7619 | 3.96k | xmlFatalErr(ctxt, XML_ERR_PEREF_SEMICOL_MISSING, NULL); |
7620 | 3.96k | return; |
7621 | 3.96k | } |
7622 | | |
7623 | 38.3k | NEXT; |
7624 | | |
7625 | | /* Must be set before xmlHandleUndeclaredEntity */ |
7626 | 38.3k | ctxt->hasPErefs = 1; |
7627 | | |
7628 | | /* |
7629 | | * Request the entity from SAX |
7630 | | */ |
7631 | 38.3k | if ((ctxt->sax != NULL) && |
7632 | 38.3k | (ctxt->sax->getParameterEntity != NULL)) |
7633 | 38.3k | entity = ctxt->sax->getParameterEntity(ctxt->userData, name); |
7634 | | |
7635 | 38.3k | if (entity == NULL) { |
7636 | 6.18k | xmlHandleUndeclaredEntity(ctxt, name); |
7637 | 32.1k | } else { |
7638 | | /* |
7639 | | * Internal checking in case the entity quest barfed |
7640 | | */ |
7641 | 32.1k | if ((entity->etype != XML_INTERNAL_PARAMETER_ENTITY) && |
7642 | 32.1k | (entity->etype != XML_EXTERNAL_PARAMETER_ENTITY)) { |
7643 | 0 | xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7644 | 0 | "Internal: %%%s; is not a parameter entity\n", |
7645 | 0 | name, NULL); |
7646 | 32.1k | } else { |
7647 | 32.1k | if ((entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) && |
7648 | 32.1k | ((ctxt->options & XML_PARSE_NO_XXE) || |
7649 | 850 | ((ctxt->loadsubset == 0) && |
7650 | 0 | (ctxt->replaceEntities == 0) && |
7651 | 0 | (ctxt->validate == 0)))) |
7652 | 850 | return; |
7653 | | |
7654 | 31.3k | if (entity->flags & XML_ENT_EXPANDING) { |
7655 | 1 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
7656 | 1 | xmlHaltParser(ctxt); |
7657 | 1 | return; |
7658 | 1 | } |
7659 | | |
7660 | 31.3k | input = xmlNewEntityInputStream(ctxt, entity); |
7661 | 31.3k | if (xmlCtxtPushInput(ctxt, input) < 0) { |
7662 | 0 | xmlFreeInputStream(input); |
7663 | 0 | return; |
7664 | 0 | } |
7665 | | |
7666 | 31.3k | entity->flags |= XML_ENT_EXPANDING; |
7667 | | |
7668 | 31.3k | if (markupDecl) |
7669 | 23.1k | input->flags |= XML_INPUT_MARKUP_DECL; |
7670 | | |
7671 | 31.3k | GROW; |
7672 | | |
7673 | 31.3k | if (entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) { |
7674 | 0 | xmlDetectEncoding(ctxt); |
7675 | |
|
7676 | 0 | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && |
7677 | 0 | (IS_BLANK_CH(NXT(5)))) { |
7678 | 0 | xmlParseTextDecl(ctxt); |
7679 | 0 | } |
7680 | 0 | } |
7681 | 31.3k | } |
7682 | 32.1k | } |
7683 | 38.3k | } |
7684 | | |
7685 | | /** |
7686 | | * Parse a parameter entity reference. |
7687 | | * |
7688 | | * @deprecated Internal function, don't use. |
7689 | | * |
7690 | | * @param ctxt an XML parser context |
7691 | | */ |
7692 | | void |
7693 | 0 | xmlParsePEReference(xmlParserCtxt *ctxt) { |
7694 | 0 | xmlParsePERefInternal(ctxt, 0); |
7695 | 0 | } |
7696 | | |
7697 | | /** |
7698 | | * Load the content of an entity. |
7699 | | * |
7700 | | * @param ctxt an XML parser context |
7701 | | * @param entity an unloaded system entity |
7702 | | * @returns 0 in case of success and -1 in case of failure |
7703 | | */ |
7704 | | static int |
7705 | 0 | xmlLoadEntityContent(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { |
7706 | 0 | xmlParserInputPtr oldinput, input = NULL; |
7707 | 0 | xmlParserInputPtr *oldinputTab; |
7708 | 0 | const xmlChar *oldencoding; |
7709 | 0 | xmlChar *content = NULL; |
7710 | 0 | xmlResourceType rtype; |
7711 | 0 | size_t length, i; |
7712 | 0 | int oldinputNr, oldinputMax; |
7713 | 0 | int ret = -1; |
7714 | 0 | int res; |
7715 | |
|
7716 | 0 | if ((ctxt == NULL) || (entity == NULL) || |
7717 | 0 | ((entity->etype != XML_EXTERNAL_PARAMETER_ENTITY) && |
7718 | 0 | (entity->etype != XML_EXTERNAL_GENERAL_PARSED_ENTITY)) || |
7719 | 0 | (entity->content != NULL)) { |
7720 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, |
7721 | 0 | "xmlLoadEntityContent parameter error"); |
7722 | 0 | return(-1); |
7723 | 0 | } |
7724 | | |
7725 | 0 | if (entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) |
7726 | 0 | rtype = XML_RESOURCE_PARAMETER_ENTITY; |
7727 | 0 | else |
7728 | 0 | rtype = XML_RESOURCE_GENERAL_ENTITY; |
7729 | |
|
7730 | 0 | input = xmlLoadResource(ctxt, (char *) entity->URI, |
7731 | 0 | (char *) entity->ExternalID, rtype); |
7732 | 0 | if (input == NULL) |
7733 | 0 | return(-1); |
7734 | | |
7735 | 0 | oldinput = ctxt->input; |
7736 | 0 | oldinputNr = ctxt->inputNr; |
7737 | 0 | oldinputMax = ctxt->inputMax; |
7738 | 0 | oldinputTab = ctxt->inputTab; |
7739 | 0 | oldencoding = ctxt->encoding; |
7740 | |
|
7741 | 0 | ctxt->input = NULL; |
7742 | 0 | ctxt->inputNr = 0; |
7743 | 0 | ctxt->inputMax = 1; |
7744 | 0 | ctxt->encoding = NULL; |
7745 | 0 | ctxt->inputTab = xmlMalloc(sizeof(xmlParserInputPtr)); |
7746 | 0 | if (ctxt->inputTab == NULL) { |
7747 | 0 | xmlErrMemory(ctxt); |
7748 | 0 | xmlFreeInputStream(input); |
7749 | 0 | goto error; |
7750 | 0 | } |
7751 | | |
7752 | 0 | xmlBufResetInput(input->buf->buffer, input); |
7753 | |
|
7754 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
7755 | 0 | xmlFreeInputStream(input); |
7756 | 0 | goto error; |
7757 | 0 | } |
7758 | | |
7759 | 0 | xmlDetectEncoding(ctxt); |
7760 | | |
7761 | | /* |
7762 | | * Parse a possible text declaration first |
7763 | | */ |
7764 | 0 | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
7765 | 0 | xmlParseTextDecl(ctxt); |
7766 | | /* |
7767 | | * An XML-1.0 document can't reference an entity not XML-1.0 |
7768 | | */ |
7769 | 0 | if ((xmlStrEqual(ctxt->version, BAD_CAST "1.0")) && |
7770 | 0 | (!xmlStrEqual(ctxt->input->version, BAD_CAST "1.0"))) { |
7771 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_VERSION_MISMATCH, |
7772 | 0 | "Version mismatch between document and entity\n"); |
7773 | 0 | } |
7774 | 0 | } |
7775 | |
|
7776 | 0 | length = input->cur - input->base; |
7777 | 0 | xmlBufShrink(input->buf->buffer, length); |
7778 | 0 | xmlSaturatedAdd(&ctxt->sizeentities, length); |
7779 | |
|
7780 | 0 | while ((res = xmlParserInputBufferGrow(input->buf, 4096)) > 0) |
7781 | 0 | ; |
7782 | |
|
7783 | 0 | xmlBufResetInput(input->buf->buffer, input); |
7784 | |
|
7785 | 0 | if (res < 0) { |
7786 | 0 | xmlCtxtErrIO(ctxt, input->buf->error, NULL); |
7787 | 0 | goto error; |
7788 | 0 | } |
7789 | | |
7790 | 0 | length = xmlBufUse(input->buf->buffer); |
7791 | 0 | if (length > INT_MAX) { |
7792 | 0 | xmlErrMemory(ctxt); |
7793 | 0 | goto error; |
7794 | 0 | } |
7795 | | |
7796 | 0 | content = xmlStrndup(xmlBufContent(input->buf->buffer), length); |
7797 | 0 | if (content == NULL) { |
7798 | 0 | xmlErrMemory(ctxt); |
7799 | 0 | goto error; |
7800 | 0 | } |
7801 | | |
7802 | 0 | for (i = 0; i < length; ) { |
7803 | 0 | int clen = length - i; |
7804 | 0 | int c = xmlGetUTF8Char(content + i, &clen); |
7805 | |
|
7806 | 0 | if ((c < 0) || (!IS_CHAR(c))) { |
7807 | 0 | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
7808 | 0 | "xmlLoadEntityContent: invalid char value %d\n", |
7809 | 0 | content[i]); |
7810 | 0 | goto error; |
7811 | 0 | } |
7812 | 0 | i += clen; |
7813 | 0 | } |
7814 | | |
7815 | 0 | xmlSaturatedAdd(&ctxt->sizeentities, length); |
7816 | 0 | entity->content = content; |
7817 | 0 | entity->length = length; |
7818 | 0 | content = NULL; |
7819 | 0 | ret = 0; |
7820 | |
|
7821 | 0 | error: |
7822 | 0 | while (ctxt->inputNr > 0) |
7823 | 0 | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
7824 | 0 | xmlFree(ctxt->inputTab); |
7825 | 0 | xmlFree((xmlChar *) ctxt->encoding); |
7826 | |
|
7827 | 0 | ctxt->input = oldinput; |
7828 | 0 | ctxt->inputNr = oldinputNr; |
7829 | 0 | ctxt->inputMax = oldinputMax; |
7830 | 0 | ctxt->inputTab = oldinputTab; |
7831 | 0 | ctxt->encoding = oldencoding; |
7832 | |
|
7833 | 0 | xmlFree(content); |
7834 | |
|
7835 | 0 | return(ret); |
7836 | 0 | } |
7837 | | |
7838 | | /** |
7839 | | * parse PEReference declarations |
7840 | | * |
7841 | | * [69] PEReference ::= '%' Name ';' |
7842 | | * |
7843 | | * [ WFC: No Recursion ] |
7844 | | * A parsed entity must not contain a recursive |
7845 | | * reference to itself, either directly or indirectly. |
7846 | | * |
7847 | | * [ WFC: Entity Declared ] |
7848 | | * In a document without any DTD, a document with only an internal DTD |
7849 | | * subset which contains no parameter entity references, or a document |
7850 | | * with "standalone='yes'", ... ... The declaration of a parameter |
7851 | | * entity must precede any reference to it... |
7852 | | * |
7853 | | * [ VC: Entity Declared ] |
7854 | | * In a document with an external subset or external parameter entities |
7855 | | * with "standalone='no'", ... ... The declaration of a parameter entity |
7856 | | * must precede any reference to it... |
7857 | | * |
7858 | | * [ WFC: In DTD ] |
7859 | | * Parameter-entity references may only appear in the DTD. |
7860 | | * NOTE: misleading but this is handled. |
7861 | | * |
7862 | | * @param ctxt an XML parser context |
7863 | | * @param str a pointer to an index in the string |
7864 | | * @returns the string of the entity content. |
7865 | | * str is updated to the current value of the index |
7866 | | */ |
7867 | | static xmlEntityPtr |
7868 | 1.38k | xmlParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str) { |
7869 | 1.38k | const xmlChar *ptr; |
7870 | 1.38k | xmlChar cur; |
7871 | 1.38k | xmlChar *name; |
7872 | 1.38k | xmlEntityPtr entity = NULL; |
7873 | | |
7874 | 1.38k | if ((str == NULL) || (*str == NULL)) return(NULL); |
7875 | 1.38k | ptr = *str; |
7876 | 1.38k | cur = *ptr; |
7877 | 1.38k | if (cur != '%') |
7878 | 0 | return(NULL); |
7879 | 1.38k | ptr++; |
7880 | 1.38k | name = xmlParseStringName(ctxt, &ptr); |
7881 | 1.38k | if (name == NULL) { |
7882 | 173 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7883 | 173 | "xmlParseStringPEReference: no name\n"); |
7884 | 173 | *str = ptr; |
7885 | 173 | return(NULL); |
7886 | 173 | } |
7887 | 1.21k | cur = *ptr; |
7888 | 1.21k | if (cur != ';') { |
7889 | 411 | xmlFatalErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING, NULL); |
7890 | 411 | xmlFree(name); |
7891 | 411 | *str = ptr; |
7892 | 411 | return(NULL); |
7893 | 411 | } |
7894 | 804 | ptr++; |
7895 | | |
7896 | | /* Must be set before xmlHandleUndeclaredEntity */ |
7897 | 804 | ctxt->hasPErefs = 1; |
7898 | | |
7899 | | /* |
7900 | | * Request the entity from SAX |
7901 | | */ |
7902 | 804 | if ((ctxt->sax != NULL) && |
7903 | 804 | (ctxt->sax->getParameterEntity != NULL)) |
7904 | 804 | entity = ctxt->sax->getParameterEntity(ctxt->userData, name); |
7905 | | |
7906 | 804 | if (entity == NULL) { |
7907 | 428 | xmlHandleUndeclaredEntity(ctxt, name); |
7908 | 428 | } else { |
7909 | | /* |
7910 | | * Internal checking in case the entity quest barfed |
7911 | | */ |
7912 | 376 | if ((entity->etype != XML_INTERNAL_PARAMETER_ENTITY) && |
7913 | 376 | (entity->etype != XML_EXTERNAL_PARAMETER_ENTITY)) { |
7914 | 0 | xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7915 | 0 | "%%%s; is not a parameter entity\n", |
7916 | 0 | name, NULL); |
7917 | 0 | } |
7918 | 376 | } |
7919 | | |
7920 | 804 | xmlFree(name); |
7921 | 804 | *str = ptr; |
7922 | 804 | return(entity); |
7923 | 1.21k | } |
7924 | | |
7925 | | /** |
7926 | | * parse a DOCTYPE declaration |
7927 | | * |
7928 | | * @deprecated Internal function, don't use. |
7929 | | * |
7930 | | * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S? |
7931 | | * ('[' (markupdecl | PEReference | S)* ']' S?)? '>' |
7932 | | * |
7933 | | * [ VC: Root Element Type ] |
7934 | | * The Name in the document type declaration must match the element |
7935 | | * type of the root element. |
7936 | | * |
7937 | | * @param ctxt an XML parser context |
7938 | | */ |
7939 | | |
7940 | | void |
7941 | 5.92k | xmlParseDocTypeDecl(xmlParserCtxt *ctxt) { |
7942 | 5.92k | const xmlChar *name = NULL; |
7943 | 5.92k | xmlChar *publicId = NULL; |
7944 | 5.92k | xmlChar *URI = NULL; |
7945 | | |
7946 | | /* |
7947 | | * We know that '<!DOCTYPE' has been detected. |
7948 | | */ |
7949 | 5.92k | SKIP(9); |
7950 | | |
7951 | 5.92k | if (SKIP_BLANKS == 0) { |
7952 | 3.09k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
7953 | 3.09k | "Space required after 'DOCTYPE'\n"); |
7954 | 3.09k | } |
7955 | | |
7956 | | /* |
7957 | | * Parse the DOCTYPE name. |
7958 | | */ |
7959 | 5.92k | name = xmlParseName(ctxt); |
7960 | 5.92k | if (name == NULL) { |
7961 | 1.41k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7962 | 1.41k | "xmlParseDocTypeDecl : no DOCTYPE name !\n"); |
7963 | 1.41k | } |
7964 | 5.92k | ctxt->intSubName = name; |
7965 | | |
7966 | 5.92k | SKIP_BLANKS; |
7967 | | |
7968 | | /* |
7969 | | * Check for public and system identifier (URI) |
7970 | | */ |
7971 | 5.92k | URI = xmlParseExternalID(ctxt, &publicId, 1); |
7972 | | |
7973 | 5.92k | if ((URI != NULL) || (publicId != NULL)) { |
7974 | 298 | ctxt->hasExternalSubset = 1; |
7975 | 298 | } |
7976 | 5.92k | ctxt->extSubURI = URI; |
7977 | 5.92k | ctxt->extSubSystem = publicId; |
7978 | | |
7979 | 5.92k | SKIP_BLANKS; |
7980 | | |
7981 | | /* |
7982 | | * Create and update the internal subset. |
7983 | | */ |
7984 | 5.92k | if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) && |
7985 | 5.92k | (!ctxt->disableSAX)) |
7986 | 2.68k | ctxt->sax->internalSubset(ctxt->userData, name, publicId, URI); |
7987 | | |
7988 | 5.92k | if ((RAW != '[') && (RAW != '>')) { |
7989 | 175 | xmlFatalErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED, NULL); |
7990 | 175 | } |
7991 | 5.92k | } |
7992 | | |
7993 | | /** |
7994 | | * parse the internal subset declaration |
7995 | | * |
7996 | | * [28 end] ('[' (markupdecl | PEReference | S)* ']' S?)? '>' |
7997 | | * @param ctxt an XML parser context |
7998 | | */ |
7999 | | |
8000 | | static void |
8001 | 5.62k | xmlParseInternalSubset(xmlParserCtxtPtr ctxt) { |
8002 | | /* |
8003 | | * Is there any DTD definition ? |
8004 | | */ |
8005 | 5.62k | if (RAW == '[') { |
8006 | 5.62k | int oldInputNr = ctxt->inputNr; |
8007 | | |
8008 | 5.62k | NEXT; |
8009 | | /* |
8010 | | * Parse the succession of Markup declarations and |
8011 | | * PEReferences. |
8012 | | * Subsequence (markupdecl | PEReference | S)* |
8013 | | */ |
8014 | 5.62k | SKIP_BLANKS; |
8015 | 168k | while (1) { |
8016 | 168k | if (PARSER_STOPPED(ctxt)) { |
8017 | 694 | return; |
8018 | 167k | } else if (ctxt->input->cur >= ctxt->input->end) { |
8019 | 23.7k | if (ctxt->inputNr <= oldInputNr) { |
8020 | 1.03k | xmlFatalErr(ctxt, XML_ERR_INT_SUBSET_NOT_FINISHED, NULL); |
8021 | 1.03k | return; |
8022 | 1.03k | } |
8023 | 22.6k | xmlPopPE(ctxt); |
8024 | 144k | } else if ((RAW == ']') && (ctxt->inputNr <= oldInputNr)) { |
8025 | 1.95k | NEXT; |
8026 | 1.95k | SKIP_BLANKS; |
8027 | 1.95k | break; |
8028 | 142k | } else if ((PARSER_EXTERNAL(ctxt)) && |
8029 | 142k | (RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
8030 | | /* |
8031 | | * Conditional sections are allowed in external entities |
8032 | | * included by PE References in the internal subset. |
8033 | | */ |
8034 | 0 | xmlParseConditionalSections(ctxt); |
8035 | 142k | } else if ((RAW == '<') && ((NXT(1) == '!') || (NXT(1) == '?'))) { |
8036 | 106k | xmlParseMarkupDecl(ctxt); |
8037 | 106k | } else if (RAW == '%') { |
8038 | 33.3k | xmlParsePERefInternal(ctxt, 1); |
8039 | 33.3k | } else { |
8040 | 1.93k | xmlFatalErr(ctxt, XML_ERR_INT_SUBSET_NOT_FINISHED, NULL); |
8041 | | |
8042 | 2.00k | while (ctxt->inputNr > oldInputNr) |
8043 | 63 | xmlPopPE(ctxt); |
8044 | 1.93k | return; |
8045 | 1.93k | } |
8046 | 162k | SKIP_BLANKS; |
8047 | 162k | SHRINK; |
8048 | 162k | GROW; |
8049 | 162k | } |
8050 | 5.62k | } |
8051 | | |
8052 | | /* |
8053 | | * We should be at the end of the DOCTYPE declaration. |
8054 | | */ |
8055 | 1.95k | if (RAW != '>') { |
8056 | 158 | xmlFatalErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED, NULL); |
8057 | 158 | return; |
8058 | 158 | } |
8059 | 1.79k | NEXT; |
8060 | 1.79k | } |
8061 | | |
8062 | | #ifdef LIBXML_SAX1_ENABLED |
8063 | | /** |
8064 | | * parse an attribute |
8065 | | * |
8066 | | * @deprecated Internal function, don't use. |
8067 | | * |
8068 | | * [41] Attribute ::= Name Eq AttValue |
8069 | | * |
8070 | | * [ WFC: No External Entity References ] |
8071 | | * Attribute values cannot contain direct or indirect entity references |
8072 | | * to external entities. |
8073 | | * |
8074 | | * [ WFC: No < in Attribute Values ] |
8075 | | * The replacement text of any entity referred to directly or indirectly in |
8076 | | * an attribute value (other than "<") must not contain a <. |
8077 | | * |
8078 | | * [ VC: Attribute Value Type ] |
8079 | | * The attribute must have been declared; the value must be of the type |
8080 | | * declared for it. |
8081 | | * |
8082 | | * [25] Eq ::= S? '=' S? |
8083 | | * |
8084 | | * With namespace: |
8085 | | * |
8086 | | * [NS 11] Attribute ::= QName Eq AttValue |
8087 | | * |
8088 | | * Also the case QName == xmlns:??? is handled independently as a namespace |
8089 | | * definition. |
8090 | | * |
8091 | | * @param ctxt an XML parser context |
8092 | | * @param value a xmlChar ** used to store the value of the attribute |
8093 | | * @returns the attribute name, and the value in *value. |
8094 | | */ |
8095 | | |
8096 | | const xmlChar * |
8097 | 0 | xmlParseAttribute(xmlParserCtxt *ctxt, xmlChar **value) { |
8098 | 0 | const xmlChar *name; |
8099 | 0 | xmlChar *val; |
8100 | |
|
8101 | 0 | *value = NULL; |
8102 | 0 | GROW; |
8103 | 0 | name = xmlParseName(ctxt); |
8104 | 0 | if (name == NULL) { |
8105 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8106 | 0 | "error parsing attribute name\n"); |
8107 | 0 | return(NULL); |
8108 | 0 | } |
8109 | | |
8110 | | /* |
8111 | | * read the value |
8112 | | */ |
8113 | 0 | SKIP_BLANKS; |
8114 | 0 | if (RAW == '=') { |
8115 | 0 | NEXT; |
8116 | 0 | SKIP_BLANKS; |
8117 | 0 | val = xmlParseAttValue(ctxt); |
8118 | 0 | } else { |
8119 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE, |
8120 | 0 | "Specification mandates value for attribute %s\n", name); |
8121 | 0 | return(name); |
8122 | 0 | } |
8123 | | |
8124 | | /* |
8125 | | * Check that xml:lang conforms to the specification |
8126 | | * No more registered as an error, just generate a warning now |
8127 | | * since this was deprecated in XML second edition |
8128 | | */ |
8129 | 0 | if ((ctxt->pedantic) && (xmlStrEqual(name, BAD_CAST "xml:lang"))) { |
8130 | 0 | if (!xmlCheckLanguageID(val)) { |
8131 | 0 | xmlWarningMsg(ctxt, XML_WAR_LANG_VALUE, |
8132 | 0 | "Malformed value for xml:lang : %s\n", |
8133 | 0 | val, NULL); |
8134 | 0 | } |
8135 | 0 | } |
8136 | | |
8137 | | /* |
8138 | | * Check that xml:space conforms to the specification |
8139 | | */ |
8140 | 0 | if (xmlStrEqual(name, BAD_CAST "xml:space")) { |
8141 | 0 | if (xmlStrEqual(val, BAD_CAST "default")) |
8142 | 0 | *(ctxt->space) = 0; |
8143 | 0 | else if (xmlStrEqual(val, BAD_CAST "preserve")) |
8144 | 0 | *(ctxt->space) = 1; |
8145 | 0 | else { |
8146 | 0 | xmlWarningMsg(ctxt, XML_WAR_SPACE_VALUE, |
8147 | 0 | "Invalid value \"%s\" for xml:space : \"default\" or \"preserve\" expected\n", |
8148 | 0 | val, NULL); |
8149 | 0 | } |
8150 | 0 | } |
8151 | |
|
8152 | 0 | *value = val; |
8153 | 0 | return(name); |
8154 | 0 | } |
8155 | | |
8156 | | /** |
8157 | | * Parse a start tag. Always consumes '<'. |
8158 | | * |
8159 | | * @deprecated Internal function, don't use. |
8160 | | * |
8161 | | * [40] STag ::= '<' Name (S Attribute)* S? '>' |
8162 | | * |
8163 | | * [ WFC: Unique Att Spec ] |
8164 | | * No attribute name may appear more than once in the same start-tag or |
8165 | | * empty-element tag. |
8166 | | * |
8167 | | * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>' |
8168 | | * |
8169 | | * [ WFC: Unique Att Spec ] |
8170 | | * No attribute name may appear more than once in the same start-tag or |
8171 | | * empty-element tag. |
8172 | | * |
8173 | | * With namespace: |
8174 | | * |
8175 | | * [NS 8] STag ::= '<' QName (S Attribute)* S? '>' |
8176 | | * |
8177 | | * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>' |
8178 | | * |
8179 | | * @param ctxt an XML parser context |
8180 | | * @returns the element name parsed |
8181 | | */ |
8182 | | |
8183 | | const xmlChar * |
8184 | 0 | xmlParseStartTag(xmlParserCtxt *ctxt) { |
8185 | 0 | const xmlChar *name; |
8186 | 0 | const xmlChar *attname; |
8187 | 0 | xmlChar *attvalue; |
8188 | 0 | const xmlChar **atts = ctxt->atts; |
8189 | 0 | int nbatts = 0; |
8190 | 0 | int maxatts = ctxt->maxatts; |
8191 | 0 | int i; |
8192 | |
|
8193 | 0 | if (RAW != '<') return(NULL); |
8194 | 0 | NEXT1; |
8195 | |
|
8196 | 0 | name = xmlParseName(ctxt); |
8197 | 0 | if (name == NULL) { |
8198 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8199 | 0 | "xmlParseStartTag: invalid element name\n"); |
8200 | 0 | return(NULL); |
8201 | 0 | } |
8202 | | |
8203 | | /* |
8204 | | * Now parse the attributes, it ends up with the ending |
8205 | | * |
8206 | | * (S Attribute)* S? |
8207 | | */ |
8208 | 0 | SKIP_BLANKS; |
8209 | 0 | GROW; |
8210 | |
|
8211 | 0 | while (((RAW != '>') && |
8212 | 0 | ((RAW != '/') || (NXT(1) != '>')) && |
8213 | 0 | (IS_BYTE_CHAR(RAW))) && (PARSER_STOPPED(ctxt) == 0)) { |
8214 | 0 | attname = xmlParseAttribute(ctxt, &attvalue); |
8215 | 0 | if (attname == NULL) |
8216 | 0 | break; |
8217 | 0 | if (attvalue != NULL) { |
8218 | | /* |
8219 | | * [ WFC: Unique Att Spec ] |
8220 | | * No attribute name may appear more than once in the same |
8221 | | * start-tag or empty-element tag. |
8222 | | */ |
8223 | 0 | for (i = 0; i < nbatts;i += 2) { |
8224 | 0 | if (xmlStrEqual(atts[i], attname)) { |
8225 | 0 | xmlErrAttributeDup(ctxt, NULL, attname); |
8226 | 0 | goto failed; |
8227 | 0 | } |
8228 | 0 | } |
8229 | | /* |
8230 | | * Add the pair to atts |
8231 | | */ |
8232 | 0 | if (nbatts + 4 > maxatts) { |
8233 | 0 | const xmlChar **n; |
8234 | 0 | int newSize; |
8235 | |
|
8236 | 0 | newSize = xmlGrowCapacity(maxatts, sizeof(n[0]) * 2, |
8237 | 0 | 11, XML_MAX_ATTRS); |
8238 | 0 | if (newSize < 0) { |
8239 | 0 | xmlErrMemory(ctxt); |
8240 | 0 | goto failed; |
8241 | 0 | } |
8242 | 0 | #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION |
8243 | 0 | if (newSize < 2) |
8244 | 0 | newSize = 2; |
8245 | 0 | #endif |
8246 | 0 | n = xmlRealloc(atts, newSize * sizeof(n[0]) * 2); |
8247 | 0 | if (n == NULL) { |
8248 | 0 | xmlErrMemory(ctxt); |
8249 | 0 | goto failed; |
8250 | 0 | } |
8251 | 0 | atts = n; |
8252 | 0 | maxatts = newSize * 2; |
8253 | 0 | ctxt->atts = atts; |
8254 | 0 | ctxt->maxatts = maxatts; |
8255 | 0 | } |
8256 | | |
8257 | 0 | atts[nbatts++] = attname; |
8258 | 0 | atts[nbatts++] = attvalue; |
8259 | 0 | atts[nbatts] = NULL; |
8260 | 0 | atts[nbatts + 1] = NULL; |
8261 | |
|
8262 | 0 | attvalue = NULL; |
8263 | 0 | } |
8264 | | |
8265 | 0 | failed: |
8266 | |
|
8267 | 0 | if (attvalue != NULL) |
8268 | 0 | xmlFree(attvalue); |
8269 | |
|
8270 | 0 | GROW |
8271 | 0 | if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>')))) |
8272 | 0 | break; |
8273 | 0 | if (SKIP_BLANKS == 0) { |
8274 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
8275 | 0 | "attributes construct error\n"); |
8276 | 0 | } |
8277 | 0 | SHRINK; |
8278 | 0 | GROW; |
8279 | 0 | } |
8280 | | |
8281 | | /* |
8282 | | * SAX: Start of Element ! |
8283 | | */ |
8284 | 0 | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL) && |
8285 | 0 | (!ctxt->disableSAX)) { |
8286 | 0 | if (nbatts > 0) |
8287 | 0 | ctxt->sax->startElement(ctxt->userData, name, atts); |
8288 | 0 | else |
8289 | 0 | ctxt->sax->startElement(ctxt->userData, name, NULL); |
8290 | 0 | } |
8291 | |
|
8292 | 0 | if (atts != NULL) { |
8293 | | /* Free only the content strings */ |
8294 | 0 | for (i = 1;i < nbatts;i+=2) |
8295 | 0 | if (atts[i] != NULL) |
8296 | 0 | xmlFree((xmlChar *) atts[i]); |
8297 | 0 | } |
8298 | 0 | return(name); |
8299 | 0 | } |
8300 | | |
8301 | | /** |
8302 | | * Parse an end tag. Always consumes '</'. |
8303 | | * |
8304 | | * [42] ETag ::= '</' Name S? '>' |
8305 | | * |
8306 | | * With namespace |
8307 | | * |
8308 | | * [NS 9] ETag ::= '</' QName S? '>' |
8309 | | * @param ctxt an XML parser context |
8310 | | * @param line line of the start tag |
8311 | | */ |
8312 | | |
8313 | | static void |
8314 | 0 | xmlParseEndTag1(xmlParserCtxtPtr ctxt, int line) { |
8315 | 0 | const xmlChar *name; |
8316 | |
|
8317 | 0 | GROW; |
8318 | 0 | if ((RAW != '<') || (NXT(1) != '/')) { |
8319 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_LTSLASH_REQUIRED, |
8320 | 0 | "xmlParseEndTag: '</' not found\n"); |
8321 | 0 | return; |
8322 | 0 | } |
8323 | 0 | SKIP(2); |
8324 | |
|
8325 | 0 | name = xmlParseNameAndCompare(ctxt,ctxt->name); |
8326 | | |
8327 | | /* |
8328 | | * We should definitely be at the ending "S? '>'" part |
8329 | | */ |
8330 | 0 | GROW; |
8331 | 0 | SKIP_BLANKS; |
8332 | 0 | if ((!IS_BYTE_CHAR(RAW)) || (RAW != '>')) { |
8333 | 0 | xmlFatalErr(ctxt, XML_ERR_GT_REQUIRED, NULL); |
8334 | 0 | } else |
8335 | 0 | NEXT1; |
8336 | | |
8337 | | /* |
8338 | | * [ WFC: Element Type Match ] |
8339 | | * The Name in an element's end-tag must match the element type in the |
8340 | | * start-tag. |
8341 | | * |
8342 | | */ |
8343 | 0 | if (name != (xmlChar*)1) { |
8344 | 0 | if (name == NULL) name = BAD_CAST "unparsable"; |
8345 | 0 | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NAME_MISMATCH, |
8346 | 0 | "Opening and ending tag mismatch: %s line %d and %s\n", |
8347 | 0 | ctxt->name, line, name); |
8348 | 0 | } |
8349 | | |
8350 | | /* |
8351 | | * SAX: End of Tag |
8352 | | */ |
8353 | 0 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL) && |
8354 | 0 | (!ctxt->disableSAX)) |
8355 | 0 | ctxt->sax->endElement(ctxt->userData, ctxt->name); |
8356 | |
|
8357 | 0 | namePop(ctxt); |
8358 | 0 | spacePop(ctxt); |
8359 | 0 | } |
8360 | | |
8361 | | /** |
8362 | | * parse an end of tag |
8363 | | * |
8364 | | * @deprecated Internal function, don't use. |
8365 | | * |
8366 | | * [42] ETag ::= '</' Name S? '>' |
8367 | | * |
8368 | | * With namespace |
8369 | | * |
8370 | | * [NS 9] ETag ::= '</' QName S? '>' |
8371 | | * @param ctxt an XML parser context |
8372 | | */ |
8373 | | |
8374 | | void |
8375 | 0 | xmlParseEndTag(xmlParserCtxt *ctxt) { |
8376 | 0 | xmlParseEndTag1(ctxt, 0); |
8377 | 0 | } |
8378 | | #endif /* LIBXML_SAX1_ENABLED */ |
8379 | | |
8380 | | /************************************************************************ |
8381 | | * * |
8382 | | * SAX 2 specific operations * |
8383 | | * * |
8384 | | ************************************************************************/ |
8385 | | |
8386 | | /** |
8387 | | * parse an XML Namespace QName |
8388 | | * |
8389 | | * [6] QName ::= (Prefix ':')? LocalPart |
8390 | | * [7] Prefix ::= NCName |
8391 | | * [8] LocalPart ::= NCName |
8392 | | * |
8393 | | * @param ctxt an XML parser context |
8394 | | * @param prefix pointer to store the prefix part |
8395 | | * @returns the Name parsed or NULL |
8396 | | */ |
8397 | | |
8398 | | static xmlHashedString |
8399 | 3.00M | xmlParseQNameHashed(xmlParserCtxtPtr ctxt, xmlHashedString *prefix) { |
8400 | 3.00M | xmlHashedString l, p; |
8401 | 3.00M | int start, isNCName = 0; |
8402 | | |
8403 | 3.00M | l.name = NULL; |
8404 | 3.00M | p.name = NULL; |
8405 | | |
8406 | 3.00M | GROW; |
8407 | 3.00M | start = CUR_PTR - BASE_PTR; |
8408 | | |
8409 | 3.00M | l = xmlParseNCName(ctxt); |
8410 | 3.00M | if (l.name != NULL) { |
8411 | 1.27M | isNCName = 1; |
8412 | 1.27M | if (CUR == ':') { |
8413 | 151k | NEXT; |
8414 | 151k | p = l; |
8415 | 151k | l = xmlParseNCName(ctxt); |
8416 | 151k | } |
8417 | 1.27M | } |
8418 | 3.00M | if ((l.name == NULL) || (CUR == ':')) { |
8419 | 1.74M | xmlChar *tmp; |
8420 | | |
8421 | 1.74M | l.name = NULL; |
8422 | 1.74M | p.name = NULL; |
8423 | 1.74M | if ((isNCName == 0) && (CUR != ':')) |
8424 | 1.71M | return(l); |
8425 | 38.8k | tmp = xmlParseNmtoken(ctxt); |
8426 | 38.8k | if (tmp != NULL) |
8427 | 22.2k | xmlFree(tmp); |
8428 | 38.8k | l = xmlDictLookupHashed(ctxt->dict, BASE_PTR + start, |
8429 | 38.8k | CUR_PTR - (BASE_PTR + start)); |
8430 | 38.8k | if (l.name == NULL) { |
8431 | 0 | xmlErrMemory(ctxt); |
8432 | 0 | return(l); |
8433 | 0 | } |
8434 | 38.8k | xmlNsErr(ctxt, XML_NS_ERR_QNAME, |
8435 | 38.8k | "Failed to parse QName '%s'\n", l.name, NULL, NULL); |
8436 | 38.8k | } |
8437 | | |
8438 | 1.28M | *prefix = p; |
8439 | 1.28M | return(l); |
8440 | 3.00M | } |
8441 | | |
8442 | | /** |
8443 | | * parse an XML Namespace QName |
8444 | | * |
8445 | | * [6] QName ::= (Prefix ':')? LocalPart |
8446 | | * [7] Prefix ::= NCName |
8447 | | * [8] LocalPart ::= NCName |
8448 | | * |
8449 | | * @param ctxt an XML parser context |
8450 | | * @param prefix pointer to store the prefix part |
8451 | | * @returns the Name parsed or NULL |
8452 | | */ |
8453 | | |
8454 | | static const xmlChar * |
8455 | 10.3k | xmlParseQName(xmlParserCtxtPtr ctxt, const xmlChar **prefix) { |
8456 | 10.3k | xmlHashedString n, p; |
8457 | | |
8458 | 10.3k | n = xmlParseQNameHashed(ctxt, &p); |
8459 | 10.3k | if (n.name == NULL) |
8460 | 4.23k | return(NULL); |
8461 | 6.15k | *prefix = p.name; |
8462 | 6.15k | return(n.name); |
8463 | 10.3k | } |
8464 | | |
8465 | | /** |
8466 | | * parse an XML name and compares for match |
8467 | | * (specialized for endtag parsing) |
8468 | | * |
8469 | | * @param ctxt an XML parser context |
8470 | | * @param name the localname |
8471 | | * @param prefix the prefix, if any. |
8472 | | * @returns NULL for an illegal name, (xmlChar*) 1 for success |
8473 | | * and the name for mismatch |
8474 | | */ |
8475 | | |
8476 | | static const xmlChar * |
8477 | | xmlParseQNameAndCompare(xmlParserCtxtPtr ctxt, xmlChar const *name, |
8478 | 13.6k | xmlChar const *prefix) { |
8479 | 13.6k | const xmlChar *cmp; |
8480 | 13.6k | const xmlChar *in; |
8481 | 13.6k | const xmlChar *ret; |
8482 | 13.6k | const xmlChar *prefix2; |
8483 | | |
8484 | 13.6k | if (prefix == NULL) return(xmlParseNameAndCompare(ctxt, name)); |
8485 | | |
8486 | 13.6k | GROW; |
8487 | 13.6k | in = ctxt->input->cur; |
8488 | | |
8489 | 13.6k | cmp = prefix; |
8490 | 22.9k | while (*in != 0 && *in == *cmp) { |
8491 | 9.28k | ++in; |
8492 | 9.28k | ++cmp; |
8493 | 9.28k | } |
8494 | 13.6k | if ((*cmp == 0) && (*in == ':')) { |
8495 | 7.09k | in++; |
8496 | 7.09k | cmp = name; |
8497 | 12.9k | while (*in != 0 && *in == *cmp) { |
8498 | 5.87k | ++in; |
8499 | 5.87k | ++cmp; |
8500 | 5.87k | } |
8501 | 7.09k | if (*cmp == 0 && (*in == '>' || IS_BLANK_CH (*in))) { |
8502 | | /* success */ |
8503 | 3.26k | ctxt->input->col += in - ctxt->input->cur; |
8504 | 3.26k | ctxt->input->cur = in; |
8505 | 3.26k | return((const xmlChar*) 1); |
8506 | 3.26k | } |
8507 | 7.09k | } |
8508 | | /* |
8509 | | * all strings coms from the dictionary, equality can be done directly |
8510 | | */ |
8511 | 10.3k | ret = xmlParseQName (ctxt, &prefix2); |
8512 | 10.3k | if (ret == NULL) |
8513 | 4.23k | return(NULL); |
8514 | 6.15k | if ((ret == name) && (prefix == prefix2)) |
8515 | 2.05k | return((const xmlChar*) 1); |
8516 | 4.09k | return ret; |
8517 | 6.15k | } |
8518 | | |
8519 | | /** |
8520 | | * parse an attribute in the new SAX2 framework. |
8521 | | * |
8522 | | * @param ctxt an XML parser context |
8523 | | * @param pref the element prefix |
8524 | | * @param elem the element name |
8525 | | * @param hprefix resulting attribute prefix |
8526 | | * @param value resulting value of the attribute |
8527 | | * @param len resulting length of the attribute |
8528 | | * @param alloc resulting indicator if the attribute was allocated |
8529 | | * @returns the attribute name, and the value in *value, . |
8530 | | */ |
8531 | | |
8532 | | static xmlHashedString |
8533 | | xmlParseAttribute2(xmlParserCtxtPtr ctxt, |
8534 | | const xmlChar * pref, const xmlChar * elem, |
8535 | | xmlHashedString * hprefix, xmlChar ** value, |
8536 | | int *len, int *alloc) |
8537 | 774k | { |
8538 | 774k | xmlHashedString hname; |
8539 | 774k | const xmlChar *prefix, *name; |
8540 | 774k | xmlChar *val = NULL, *internal_val = NULL; |
8541 | 774k | int special = 0; |
8542 | 774k | int isNamespace; |
8543 | 774k | int flags; |
8544 | | |
8545 | 774k | *value = NULL; |
8546 | 774k | GROW; |
8547 | 774k | hname = xmlParseQNameHashed(ctxt, hprefix); |
8548 | 774k | if (hname.name == NULL) { |
8549 | 97.0k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8550 | 97.0k | "error parsing attribute name\n"); |
8551 | 97.0k | return(hname); |
8552 | 97.0k | } |
8553 | 677k | name = hname.name; |
8554 | 677k | prefix = hprefix->name; |
8555 | | |
8556 | | /* |
8557 | | * get the type if needed |
8558 | | */ |
8559 | 677k | if (ctxt->attsSpecial != NULL) { |
8560 | 126k | special = XML_PTR_TO_INT(xmlHashQLookup2(ctxt->attsSpecial, pref, elem, |
8561 | 126k | prefix, name)); |
8562 | 126k | } |
8563 | | |
8564 | | /* |
8565 | | * read the value |
8566 | | */ |
8567 | 677k | SKIP_BLANKS; |
8568 | 677k | if (RAW != '=') { |
8569 | 44.1k | xmlFatalErrMsgStr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE, |
8570 | 44.1k | "Specification mandates value for attribute %s\n", |
8571 | 44.1k | name); |
8572 | 44.1k | goto error; |
8573 | 44.1k | } |
8574 | | |
8575 | | |
8576 | 633k | NEXT; |
8577 | 633k | SKIP_BLANKS; |
8578 | 633k | flags = 0; |
8579 | 633k | isNamespace = (((prefix == NULL) && (name == ctxt->str_xmlns)) || |
8580 | 633k | (prefix == ctxt->str_xmlns)); |
8581 | 633k | val = xmlParseAttValueInternal(ctxt, len, &flags, special, |
8582 | 633k | isNamespace); |
8583 | 633k | if (val == NULL) |
8584 | 23.2k | goto error; |
8585 | | |
8586 | 609k | *alloc = (flags & XML_ATTVAL_ALLOC) != 0; |
8587 | | |
8588 | 609k | #ifdef LIBXML_VALID_ENABLED |
8589 | 609k | if ((ctxt->validate) && |
8590 | 609k | (ctxt->standalone) && |
8591 | 609k | (special & XML_SPECIAL_EXTERNAL) && |
8592 | 609k | (flags & XML_ATTVAL_NORM_CHANGE)) { |
8593 | 0 | xmlValidityError(ctxt, XML_DTD_NOT_STANDALONE, |
8594 | 0 | "standalone: normalization of attribute %s on %s " |
8595 | 0 | "by external subset declaration\n", |
8596 | 0 | name, elem); |
8597 | 0 | } |
8598 | 609k | #endif |
8599 | | |
8600 | 609k | if (prefix == ctxt->str_xml) { |
8601 | | /* |
8602 | | * Check that xml:lang conforms to the specification |
8603 | | * No more registered as an error, just generate a warning now |
8604 | | * since this was deprecated in XML second edition |
8605 | | */ |
8606 | 16.8k | if ((ctxt->pedantic) && (xmlStrEqual(name, BAD_CAST "lang"))) { |
8607 | 0 | internal_val = xmlStrndup(val, *len); |
8608 | 0 | if (internal_val == NULL) |
8609 | 0 | goto mem_error; |
8610 | 0 | if (!xmlCheckLanguageID(internal_val)) { |
8611 | 0 | xmlWarningMsg(ctxt, XML_WAR_LANG_VALUE, |
8612 | 0 | "Malformed value for xml:lang : %s\n", |
8613 | 0 | internal_val, NULL); |
8614 | 0 | } |
8615 | 0 | } |
8616 | | |
8617 | | /* |
8618 | | * Check that xml:space conforms to the specification |
8619 | | */ |
8620 | 16.8k | if (xmlStrEqual(name, BAD_CAST "space")) { |
8621 | 1.62k | internal_val = xmlStrndup(val, *len); |
8622 | 1.62k | if (internal_val == NULL) |
8623 | 0 | goto mem_error; |
8624 | 1.62k | if (xmlStrEqual(internal_val, BAD_CAST "default")) |
8625 | 618 | *(ctxt->space) = 0; |
8626 | 1.00k | else if (xmlStrEqual(internal_val, BAD_CAST "preserve")) |
8627 | 377 | *(ctxt->space) = 1; |
8628 | 631 | else { |
8629 | 631 | xmlWarningMsg(ctxt, XML_WAR_SPACE_VALUE, |
8630 | 631 | "Invalid value \"%s\" for xml:space : \"default\" or \"preserve\" expected\n", |
8631 | 631 | internal_val, NULL); |
8632 | 631 | } |
8633 | 1.62k | } |
8634 | 16.8k | if (internal_val) { |
8635 | 1.62k | xmlFree(internal_val); |
8636 | 1.62k | } |
8637 | 16.8k | } |
8638 | | |
8639 | 609k | *value = val; |
8640 | 609k | return (hname); |
8641 | | |
8642 | 0 | mem_error: |
8643 | 0 | xmlErrMemory(ctxt); |
8644 | 67.4k | error: |
8645 | 67.4k | if ((val != NULL) && (*alloc != 0)) |
8646 | 0 | xmlFree(val); |
8647 | 67.4k | return(hname); |
8648 | 0 | } |
8649 | | |
8650 | | /** |
8651 | | * Inserts a new attribute into the hash table. |
8652 | | * |
8653 | | * @param ctxt parser context |
8654 | | * @param size size of the hash table |
8655 | | * @param name attribute name |
8656 | | * @param uri namespace uri |
8657 | | * @param hashValue combined hash value of name and uri |
8658 | | * @param aindex attribute index (this is a multiple of 5) |
8659 | | * @returns INT_MAX if no existing attribute was found, the attribute |
8660 | | * index if an attribute was found, -1 if a memory allocation failed. |
8661 | | */ |
8662 | | static int |
8663 | | xmlAttrHashInsert(xmlParserCtxtPtr ctxt, unsigned size, const xmlChar *name, |
8664 | 622k | const xmlChar *uri, unsigned hashValue, int aindex) { |
8665 | 622k | xmlAttrHashBucket *table = ctxt->attrHash; |
8666 | 622k | xmlAttrHashBucket *bucket; |
8667 | 622k | unsigned hindex; |
8668 | | |
8669 | 622k | hindex = hashValue & (size - 1); |
8670 | 622k | bucket = &table[hindex]; |
8671 | | |
8672 | 692k | while (bucket->index >= 0) { |
8673 | 435k | const xmlChar **atts = &ctxt->atts[bucket->index]; |
8674 | | |
8675 | 435k | if (name == atts[0]) { |
8676 | 386k | int nsIndex = XML_PTR_TO_INT(atts[2]); |
8677 | | |
8678 | 386k | if ((nsIndex == NS_INDEX_EMPTY) ? (uri == NULL) : |
8679 | 386k | (nsIndex == NS_INDEX_XML) ? (uri == ctxt->str_xml_ns) : |
8680 | 24.9k | (uri == ctxt->nsTab[nsIndex * 2 + 1])) |
8681 | 365k | return(bucket->index); |
8682 | 386k | } |
8683 | | |
8684 | 70.8k | hindex++; |
8685 | 70.8k | bucket++; |
8686 | 70.8k | if (hindex >= size) { |
8687 | 14.5k | hindex = 0; |
8688 | 14.5k | bucket = table; |
8689 | 14.5k | } |
8690 | 70.8k | } |
8691 | | |
8692 | 256k | bucket->index = aindex; |
8693 | | |
8694 | 256k | return(INT_MAX); |
8695 | 622k | } |
8696 | | |
8697 | | static int |
8698 | | xmlAttrHashInsertQName(xmlParserCtxtPtr ctxt, unsigned size, |
8699 | | const xmlChar *name, const xmlChar *prefix, |
8700 | 9.53k | unsigned hashValue, int aindex) { |
8701 | 9.53k | xmlAttrHashBucket *table = ctxt->attrHash; |
8702 | 9.53k | xmlAttrHashBucket *bucket; |
8703 | 9.53k | unsigned hindex; |
8704 | | |
8705 | 9.53k | hindex = hashValue & (size - 1); |
8706 | 9.53k | bucket = &table[hindex]; |
8707 | | |
8708 | 12.8k | while (bucket->index >= 0) { |
8709 | 8.63k | const xmlChar **atts = &ctxt->atts[bucket->index]; |
8710 | | |
8711 | 8.63k | if ((name == atts[0]) && (prefix == atts[1])) |
8712 | 5.29k | return(bucket->index); |
8713 | | |
8714 | 3.33k | hindex++; |
8715 | 3.33k | bucket++; |
8716 | 3.33k | if (hindex >= size) { |
8717 | 258 | hindex = 0; |
8718 | 258 | bucket = table; |
8719 | 258 | } |
8720 | 3.33k | } |
8721 | | |
8722 | 4.23k | bucket->index = aindex; |
8723 | | |
8724 | 4.23k | return(INT_MAX); |
8725 | 9.53k | } |
8726 | | /** |
8727 | | * Parse a start tag. Always consumes '<'. |
8728 | | * |
8729 | | * This routine is called when running SAX2 parsing |
8730 | | * |
8731 | | * [40] STag ::= '<' Name (S Attribute)* S? '>' |
8732 | | * |
8733 | | * [ WFC: Unique Att Spec ] |
8734 | | * No attribute name may appear more than once in the same start-tag or |
8735 | | * empty-element tag. |
8736 | | * |
8737 | | * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>' |
8738 | | * |
8739 | | * [ WFC: Unique Att Spec ] |
8740 | | * No attribute name may appear more than once in the same start-tag or |
8741 | | * empty-element tag. |
8742 | | * |
8743 | | * With namespace: |
8744 | | * |
8745 | | * [NS 8] STag ::= '<' QName (S Attribute)* S? '>' |
8746 | | * |
8747 | | * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>' |
8748 | | * |
8749 | | * @param ctxt an XML parser context |
8750 | | * @param pref resulting namespace prefix |
8751 | | * @param URI resulting namespace URI |
8752 | | * @param nbNsPtr resulting number of namespace declarations |
8753 | | * @returns the element name parsed |
8754 | | */ |
8755 | | |
8756 | | static const xmlChar * |
8757 | | xmlParseStartTag2(xmlParserCtxtPtr ctxt, const xmlChar **pref, |
8758 | 2.21M | const xmlChar **URI, int *nbNsPtr) { |
8759 | 2.21M | xmlHashedString hlocalname; |
8760 | 2.21M | xmlHashedString hprefix; |
8761 | 2.21M | xmlHashedString hattname; |
8762 | 2.21M | xmlHashedString haprefix; |
8763 | 2.21M | const xmlChar *localname; |
8764 | 2.21M | const xmlChar *prefix; |
8765 | 2.21M | const xmlChar *attname; |
8766 | 2.21M | const xmlChar *aprefix; |
8767 | 2.21M | const xmlChar *uri; |
8768 | 2.21M | xmlChar *attvalue = NULL; |
8769 | 2.21M | const xmlChar **atts = ctxt->atts; |
8770 | 2.21M | unsigned attrHashSize = 0; |
8771 | 2.21M | int maxatts = ctxt->maxatts; |
8772 | 2.21M | int nratts, nbatts, nbdef; |
8773 | 2.21M | int i, j, nbNs, nbTotalDef, attval, nsIndex, maxAtts; |
8774 | 2.21M | int alloc = 0; |
8775 | 2.21M | int numNsErr = 0; |
8776 | 2.21M | int numDupErr = 0; |
8777 | | |
8778 | 2.21M | if (RAW != '<') return(NULL); |
8779 | 2.21M | NEXT1; |
8780 | | |
8781 | 2.21M | nbatts = 0; |
8782 | 2.21M | nratts = 0; |
8783 | 2.21M | nbdef = 0; |
8784 | 2.21M | nbNs = 0; |
8785 | 2.21M | nbTotalDef = 0; |
8786 | 2.21M | attval = 0; |
8787 | | |
8788 | 2.21M | if (xmlParserNsStartElement(ctxt->nsdb) < 0) { |
8789 | 0 | xmlErrMemory(ctxt); |
8790 | 0 | return(NULL); |
8791 | 0 | } |
8792 | | |
8793 | 2.21M | hlocalname = xmlParseQNameHashed(ctxt, &hprefix); |
8794 | 2.21M | if (hlocalname.name == NULL) { |
8795 | 1.60M | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8796 | 1.60M | "StartTag: invalid element name\n"); |
8797 | 1.60M | return(NULL); |
8798 | 1.60M | } |
8799 | 605k | localname = hlocalname.name; |
8800 | 605k | prefix = hprefix.name; |
8801 | | |
8802 | | /* |
8803 | | * Now parse the attributes, it ends up with the ending |
8804 | | * |
8805 | | * (S Attribute)* S? |
8806 | | */ |
8807 | 605k | SKIP_BLANKS; |
8808 | 605k | GROW; |
8809 | | |
8810 | | /* |
8811 | | * The ctxt->atts array will be ultimately passed to the SAX callback |
8812 | | * containing five xmlChar pointers for each attribute: |
8813 | | * |
8814 | | * [0] attribute name |
8815 | | * [1] attribute prefix |
8816 | | * [2] namespace URI |
8817 | | * [3] attribute value |
8818 | | * [4] end of attribute value |
8819 | | * |
8820 | | * To save memory, we reuse this array temporarily and store integers |
8821 | | * in these pointer variables. |
8822 | | * |
8823 | | * [0] attribute name |
8824 | | * [1] attribute prefix |
8825 | | * [2] hash value of attribute prefix, and later namespace index |
8826 | | * [3] for non-allocated values: ptrdiff_t offset into input buffer |
8827 | | * [4] for non-allocated values: ptrdiff_t offset into input buffer |
8828 | | * |
8829 | | * The ctxt->attallocs array contains an additional unsigned int for |
8830 | | * each attribute, containing the hash value of the attribute name |
8831 | | * and the alloc flag in bit 31. |
8832 | | */ |
8833 | | |
8834 | 1.03M | while (((RAW != '>') && |
8835 | 1.03M | ((RAW != '/') || (NXT(1) != '>')) && |
8836 | 1.03M | (IS_BYTE_CHAR(RAW))) && (PARSER_STOPPED(ctxt) == 0)) { |
8837 | 774k | int len = -1; |
8838 | | |
8839 | 774k | hattname = xmlParseAttribute2(ctxt, prefix, localname, |
8840 | 774k | &haprefix, &attvalue, &len, |
8841 | 774k | &alloc); |
8842 | 774k | if (hattname.name == NULL) |
8843 | 97.0k | break; |
8844 | 677k | if (attvalue == NULL) |
8845 | 67.4k | goto next_attr; |
8846 | 609k | attname = hattname.name; |
8847 | 609k | aprefix = haprefix.name; |
8848 | 609k | if (len < 0) len = xmlStrlen(attvalue); |
8849 | | |
8850 | 609k | if ((attname == ctxt->str_xmlns) && (aprefix == NULL)) { |
8851 | 130k | xmlHashedString huri; |
8852 | 130k | xmlURIPtr parsedUri; |
8853 | | |
8854 | 130k | huri = xmlDictLookupHashed(ctxt->dict, attvalue, len); |
8855 | 130k | uri = huri.name; |
8856 | 130k | if (uri == NULL) { |
8857 | 0 | xmlErrMemory(ctxt); |
8858 | 0 | goto next_attr; |
8859 | 0 | } |
8860 | 130k | if (*uri != 0) { |
8861 | 125k | if (xmlParseURISafe((const char *) uri, &parsedUri) < 0) { |
8862 | 0 | xmlErrMemory(ctxt); |
8863 | 0 | goto next_attr; |
8864 | 0 | } |
8865 | 125k | if (parsedUri == NULL) { |
8866 | 84.5k | xmlNsErr(ctxt, XML_WAR_NS_URI, |
8867 | 84.5k | "xmlns: '%s' is not a valid URI\n", |
8868 | 84.5k | uri, NULL, NULL); |
8869 | 84.5k | } else { |
8870 | 40.8k | if (parsedUri->scheme == NULL) { |
8871 | 19.3k | xmlNsWarn(ctxt, XML_WAR_NS_URI_RELATIVE, |
8872 | 19.3k | "xmlns: URI %s is not absolute\n", |
8873 | 19.3k | uri, NULL, NULL); |
8874 | 19.3k | } |
8875 | 40.8k | xmlFreeURI(parsedUri); |
8876 | 40.8k | } |
8877 | 125k | if (uri == ctxt->str_xml_ns) { |
8878 | 726 | if (attname != ctxt->str_xml) { |
8879 | 726 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8880 | 726 | "xml namespace URI cannot be the default namespace\n", |
8881 | 726 | NULL, NULL, NULL); |
8882 | 726 | } |
8883 | 726 | goto next_attr; |
8884 | 726 | } |
8885 | 124k | if ((len == 29) && |
8886 | 124k | (xmlStrEqual(uri, |
8887 | 1.81k | BAD_CAST "http://www.w3.org/2000/xmlns/"))) { |
8888 | 276 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8889 | 276 | "reuse of the xmlns namespace name is forbidden\n", |
8890 | 276 | NULL, NULL, NULL); |
8891 | 276 | goto next_attr; |
8892 | 276 | } |
8893 | 124k | } |
8894 | | |
8895 | 129k | if (xmlParserNsPush(ctxt, NULL, &huri, NULL, 0) > 0) |
8896 | 110k | nbNs++; |
8897 | 479k | } else if (aprefix == ctxt->str_xmlns) { |
8898 | 35.5k | xmlHashedString huri; |
8899 | 35.5k | xmlURIPtr parsedUri; |
8900 | | |
8901 | 35.5k | huri = xmlDictLookupHashed(ctxt->dict, attvalue, len); |
8902 | 35.5k | uri = huri.name; |
8903 | 35.5k | if (uri == NULL) { |
8904 | 0 | xmlErrMemory(ctxt); |
8905 | 0 | goto next_attr; |
8906 | 0 | } |
8907 | | |
8908 | 35.5k | if (attname == ctxt->str_xml) { |
8909 | 1.64k | if (uri != ctxt->str_xml_ns) { |
8910 | 896 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8911 | 896 | "xml namespace prefix mapped to wrong URI\n", |
8912 | 896 | NULL, NULL, NULL); |
8913 | 896 | } |
8914 | | /* |
8915 | | * Do not keep a namespace definition node |
8916 | | */ |
8917 | 1.64k | goto next_attr; |
8918 | 1.64k | } |
8919 | 33.9k | if (uri == ctxt->str_xml_ns) { |
8920 | 4.47k | if (attname != ctxt->str_xml) { |
8921 | 4.47k | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8922 | 4.47k | "xml namespace URI mapped to wrong prefix\n", |
8923 | 4.47k | NULL, NULL, NULL); |
8924 | 4.47k | } |
8925 | 4.47k | goto next_attr; |
8926 | 4.47k | } |
8927 | 29.4k | if (attname == ctxt->str_xmlns) { |
8928 | 7.75k | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8929 | 7.75k | "redefinition of the xmlns prefix is forbidden\n", |
8930 | 7.75k | NULL, NULL, NULL); |
8931 | 7.75k | goto next_attr; |
8932 | 7.75k | } |
8933 | 21.6k | if ((len == 29) && |
8934 | 21.6k | (xmlStrEqual(uri, |
8935 | 919 | BAD_CAST "http://www.w3.org/2000/xmlns/"))) { |
8936 | 343 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8937 | 343 | "reuse of the xmlns namespace name is forbidden\n", |
8938 | 343 | NULL, NULL, NULL); |
8939 | 343 | goto next_attr; |
8940 | 343 | } |
8941 | 21.3k | if ((uri == NULL) || (uri[0] == 0)) { |
8942 | 721 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8943 | 721 | "xmlns:%s: Empty XML namespace is not allowed\n", |
8944 | 721 | attname, NULL, NULL); |
8945 | 721 | goto next_attr; |
8946 | 20.6k | } else { |
8947 | 20.6k | if (xmlParseURISafe((const char *) uri, &parsedUri) < 0) { |
8948 | 0 | xmlErrMemory(ctxt); |
8949 | 0 | goto next_attr; |
8950 | 0 | } |
8951 | 20.6k | if (parsedUri == NULL) { |
8952 | 9.56k | xmlNsErr(ctxt, XML_WAR_NS_URI, |
8953 | 9.56k | "xmlns:%s: '%s' is not a valid URI\n", |
8954 | 9.56k | attname, uri, NULL); |
8955 | 11.0k | } else { |
8956 | 11.0k | if ((ctxt->pedantic) && (parsedUri->scheme == NULL)) { |
8957 | 0 | xmlNsWarn(ctxt, XML_WAR_NS_URI_RELATIVE, |
8958 | 0 | "xmlns:%s: URI %s is not absolute\n", |
8959 | 0 | attname, uri, NULL); |
8960 | 0 | } |
8961 | 11.0k | xmlFreeURI(parsedUri); |
8962 | 11.0k | } |
8963 | 20.6k | } |
8964 | | |
8965 | 20.6k | if (xmlParserNsPush(ctxt, &hattname, &huri, NULL, 0) > 0) |
8966 | 19.9k | nbNs++; |
8967 | 443k | } else { |
8968 | | /* |
8969 | | * Populate attributes array, see above for repurposing |
8970 | | * of xmlChar pointers. |
8971 | | */ |
8972 | 443k | if ((atts == NULL) || (nbatts + 5 > maxatts)) { |
8973 | 4.10k | int res = xmlCtxtGrowAttrs(ctxt); |
8974 | | |
8975 | 4.10k | maxatts = ctxt->maxatts; |
8976 | 4.10k | atts = ctxt->atts; |
8977 | | |
8978 | 4.10k | if (res < 0) |
8979 | 0 | goto next_attr; |
8980 | 4.10k | } |
8981 | 443k | ctxt->attallocs[nratts++] = (hattname.hashValue & 0x7FFFFFFF) | |
8982 | 443k | ((unsigned) alloc << 31); |
8983 | 443k | atts[nbatts++] = attname; |
8984 | 443k | atts[nbatts++] = aprefix; |
8985 | 443k | atts[nbatts++] = XML_INT_TO_PTR(haprefix.hashValue); |
8986 | 443k | if (alloc) { |
8987 | 55.1k | atts[nbatts++] = attvalue; |
8988 | 55.1k | attvalue += len; |
8989 | 55.1k | atts[nbatts++] = attvalue; |
8990 | 388k | } else { |
8991 | | /* |
8992 | | * attvalue points into the input buffer which can be |
8993 | | * reallocated. Store differences to input->base instead. |
8994 | | * The pointers will be reconstructed later. |
8995 | | */ |
8996 | 388k | atts[nbatts++] = XML_INT_TO_PTR(attvalue - BASE_PTR); |
8997 | 388k | attvalue += len; |
8998 | 388k | atts[nbatts++] = XML_INT_TO_PTR(attvalue - BASE_PTR); |
8999 | 388k | } |
9000 | | /* |
9001 | | * tag if some deallocation is needed |
9002 | | */ |
9003 | 443k | if (alloc != 0) attval = 1; |
9004 | 443k | attvalue = NULL; /* moved into atts */ |
9005 | 443k | } |
9006 | | |
9007 | 677k | next_attr: |
9008 | 677k | if ((attvalue != NULL) && (alloc != 0)) { |
9009 | 46.5k | xmlFree(attvalue); |
9010 | 46.5k | attvalue = NULL; |
9011 | 46.5k | } |
9012 | | |
9013 | 677k | GROW |
9014 | 677k | if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>')))) |
9015 | 92.3k | break; |
9016 | 584k | if (SKIP_BLANKS == 0) { |
9017 | 153k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
9018 | 153k | "attributes construct error\n"); |
9019 | 153k | break; |
9020 | 153k | } |
9021 | 431k | GROW; |
9022 | 431k | } |
9023 | | |
9024 | | /* |
9025 | | * Namespaces from default attributes |
9026 | | */ |
9027 | 605k | if (ctxt->attsDefault != NULL) { |
9028 | 232k | xmlDefAttrsPtr defaults; |
9029 | | |
9030 | 232k | defaults = xmlHashLookup2(ctxt->attsDefault, localname, prefix); |
9031 | 232k | if (defaults != NULL) { |
9032 | 956k | for (i = 0; i < defaults->nbAttrs; i++) { |
9033 | 831k | xmlDefAttr *attr = &defaults->attrs[i]; |
9034 | | |
9035 | 831k | attname = attr->name.name; |
9036 | 831k | aprefix = attr->prefix.name; |
9037 | | |
9038 | 831k | if ((attname == ctxt->str_xmlns) && (aprefix == NULL)) { |
9039 | 6.94k | xmlParserEntityCheck(ctxt, attr->expandedSize); |
9040 | | |
9041 | 6.94k | if (xmlParserNsPush(ctxt, NULL, &attr->value, NULL, 1) > 0) |
9042 | 6.33k | nbNs++; |
9043 | 824k | } else if (aprefix == ctxt->str_xmlns) { |
9044 | 544k | xmlParserEntityCheck(ctxt, attr->expandedSize); |
9045 | | |
9046 | 544k | if (xmlParserNsPush(ctxt, &attr->name, &attr->value, |
9047 | 544k | NULL, 1) > 0) |
9048 | 544k | nbNs++; |
9049 | 544k | } else { |
9050 | 280k | if (nratts + nbTotalDef >= XML_MAX_ATTRS) { |
9051 | 0 | xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, |
9052 | 0 | "Maximum number of attributes exceeded"); |
9053 | 0 | break; |
9054 | 0 | } |
9055 | 280k | nbTotalDef += 1; |
9056 | 280k | } |
9057 | 831k | } |
9058 | 124k | } |
9059 | 232k | } |
9060 | | |
9061 | | /* |
9062 | | * Resolve attribute namespaces |
9063 | | */ |
9064 | 1.04M | for (i = 0; i < nbatts; i += 5) { |
9065 | 443k | attname = atts[i]; |
9066 | 443k | aprefix = atts[i+1]; |
9067 | | |
9068 | | /* |
9069 | | * The default namespace does not apply to attribute names. |
9070 | | */ |
9071 | 443k | if (aprefix == NULL) { |
9072 | 392k | nsIndex = NS_INDEX_EMPTY; |
9073 | 392k | } else if (aprefix == ctxt->str_xml) { |
9074 | 16.8k | nsIndex = NS_INDEX_XML; |
9075 | 34.1k | } else { |
9076 | 34.1k | haprefix.name = aprefix; |
9077 | 34.1k | haprefix.hashValue = (size_t) atts[i+2]; |
9078 | 34.1k | nsIndex = xmlParserNsLookup(ctxt, &haprefix, NULL); |
9079 | | |
9080 | 34.1k | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) { |
9081 | 18.6k | xmlNsErr(ctxt, XML_NS_ERR_UNDEFINED_NAMESPACE, |
9082 | 18.6k | "Namespace prefix %s for %s on %s is not defined\n", |
9083 | 18.6k | aprefix, attname, localname); |
9084 | 18.6k | nsIndex = NS_INDEX_EMPTY; |
9085 | 18.6k | } |
9086 | 34.1k | } |
9087 | | |
9088 | 443k | atts[i+2] = XML_INT_TO_PTR(nsIndex); |
9089 | 443k | } |
9090 | | |
9091 | | /* |
9092 | | * Maximum number of attributes including default attributes. |
9093 | | */ |
9094 | 605k | maxAtts = nratts + nbTotalDef; |
9095 | | |
9096 | | /* |
9097 | | * Verify that attribute names are unique. |
9098 | | */ |
9099 | 605k | if (maxAtts > 1) { |
9100 | 85.2k | attrHashSize = 4; |
9101 | 159k | while (attrHashSize / 2 < (unsigned) maxAtts) |
9102 | 73.9k | attrHashSize *= 2; |
9103 | | |
9104 | 85.2k | if (attrHashSize > ctxt->attrHashMax) { |
9105 | 1.28k | xmlAttrHashBucket *tmp; |
9106 | | |
9107 | 1.28k | tmp = xmlRealloc(ctxt->attrHash, attrHashSize * sizeof(tmp[0])); |
9108 | 1.28k | if (tmp == NULL) { |
9109 | 0 | xmlErrMemory(ctxt); |
9110 | 0 | goto done; |
9111 | 0 | } |
9112 | | |
9113 | 1.28k | ctxt->attrHash = tmp; |
9114 | 1.28k | ctxt->attrHashMax = attrHashSize; |
9115 | 1.28k | } |
9116 | | |
9117 | 85.2k | memset(ctxt->attrHash, -1, attrHashSize * sizeof(ctxt->attrHash[0])); |
9118 | | |
9119 | 487k | for (i = 0, j = 0; j < nratts; i += 5, j++) { |
9120 | 402k | const xmlChar *nsuri; |
9121 | 402k | unsigned hashValue, nameHashValue, uriHashValue; |
9122 | 402k | int res; |
9123 | | |
9124 | 402k | attname = atts[i]; |
9125 | 402k | aprefix = atts[i+1]; |
9126 | 402k | nsIndex = XML_PTR_TO_INT(atts[i+2]); |
9127 | | /* Hash values always have bit 31 set, see dict.c */ |
9128 | 402k | nameHashValue = ctxt->attallocs[j] | 0x80000000; |
9129 | | |
9130 | 402k | if (nsIndex == NS_INDEX_EMPTY) { |
9131 | | /* |
9132 | | * Prefix with empty namespace means an undeclared |
9133 | | * prefix which was already reported above. |
9134 | | */ |
9135 | 382k | if (aprefix != NULL) |
9136 | 15.0k | continue; |
9137 | 367k | nsuri = NULL; |
9138 | 367k | uriHashValue = URI_HASH_EMPTY; |
9139 | 367k | } else if (nsIndex == NS_INDEX_XML) { |
9140 | 4.16k | nsuri = ctxt->str_xml_ns; |
9141 | 4.16k | uriHashValue = URI_HASH_XML; |
9142 | 15.1k | } else { |
9143 | 15.1k | nsuri = ctxt->nsTab[nsIndex * 2 + 1]; |
9144 | 15.1k | uriHashValue = ctxt->nsdb->extra[nsIndex].uriHashValue; |
9145 | 15.1k | } |
9146 | | |
9147 | 387k | hashValue = xmlDictCombineHash(nameHashValue, uriHashValue); |
9148 | 387k | res = xmlAttrHashInsert(ctxt, attrHashSize, attname, nsuri, |
9149 | 387k | hashValue, i); |
9150 | 387k | if (res < 0) |
9151 | 0 | continue; |
9152 | | |
9153 | | /* |
9154 | | * [ WFC: Unique Att Spec ] |
9155 | | * No attribute name may appear more than once in the same |
9156 | | * start-tag or empty-element tag. |
9157 | | * As extended by the Namespace in XML REC. |
9158 | | */ |
9159 | 387k | if (res < INT_MAX) { |
9160 | 351k | if (aprefix == atts[res+1]) { |
9161 | 344k | xmlErrAttributeDup(ctxt, aprefix, attname); |
9162 | 344k | numDupErr += 1; |
9163 | 344k | } else { |
9164 | 6.69k | xmlNsErr(ctxt, XML_NS_ERR_ATTRIBUTE_REDEFINED, |
9165 | 6.69k | "Namespaced Attribute %s in '%s' redefined\n", |
9166 | 6.69k | attname, nsuri, NULL); |
9167 | 6.69k | numNsErr += 1; |
9168 | 6.69k | } |
9169 | 351k | } |
9170 | 387k | } |
9171 | 85.2k | } |
9172 | | |
9173 | | /* |
9174 | | * Default attributes |
9175 | | */ |
9176 | 605k | if (ctxt->attsDefault != NULL) { |
9177 | 232k | xmlDefAttrsPtr defaults; |
9178 | | |
9179 | 232k | defaults = xmlHashLookup2(ctxt->attsDefault, localname, prefix); |
9180 | 232k | if (defaults != NULL) { |
9181 | 956k | for (i = 0; i < defaults->nbAttrs; i++) { |
9182 | 831k | xmlDefAttr *attr = &defaults->attrs[i]; |
9183 | 831k | const xmlChar *nsuri = NULL; |
9184 | 831k | unsigned hashValue, uriHashValue = 0; |
9185 | 831k | int res; |
9186 | | |
9187 | 831k | attname = attr->name.name; |
9188 | 831k | aprefix = attr->prefix.name; |
9189 | | |
9190 | 831k | if ((attname == ctxt->str_xmlns) && (aprefix == NULL)) |
9191 | 6.94k | continue; |
9192 | 824k | if (aprefix == ctxt->str_xmlns) |
9193 | 544k | continue; |
9194 | | |
9195 | 280k | if (aprefix == NULL) { |
9196 | 35.7k | nsIndex = NS_INDEX_EMPTY; |
9197 | 35.7k | nsuri = NULL; |
9198 | 35.7k | uriHashValue = URI_HASH_EMPTY; |
9199 | 244k | } else if (aprefix == ctxt->str_xml) { |
9200 | 26.3k | nsIndex = NS_INDEX_XML; |
9201 | 26.3k | nsuri = ctxt->str_xml_ns; |
9202 | 26.3k | uriHashValue = URI_HASH_XML; |
9203 | 217k | } else { |
9204 | 217k | nsIndex = xmlParserNsLookup(ctxt, &attr->prefix, NULL); |
9205 | 217k | if ((nsIndex == INT_MAX) || |
9206 | 217k | (nsIndex < ctxt->nsdb->minNsIndex)) { |
9207 | 163k | xmlNsErr(ctxt, XML_NS_ERR_UNDEFINED_NAMESPACE, |
9208 | 163k | "Namespace prefix %s for %s on %s is not " |
9209 | 163k | "defined\n", |
9210 | 163k | aprefix, attname, localname); |
9211 | 163k | nsIndex = NS_INDEX_EMPTY; |
9212 | 163k | nsuri = NULL; |
9213 | 163k | uriHashValue = URI_HASH_EMPTY; |
9214 | 163k | } else { |
9215 | 54.7k | nsuri = ctxt->nsTab[nsIndex * 2 + 1]; |
9216 | 54.7k | uriHashValue = ctxt->nsdb->extra[nsIndex].uriHashValue; |
9217 | 54.7k | } |
9218 | 217k | } |
9219 | | |
9220 | | /* |
9221 | | * Check whether the attribute exists |
9222 | | */ |
9223 | 280k | if (maxAtts > 1) { |
9224 | 234k | hashValue = xmlDictCombineHash(attr->name.hashValue, |
9225 | 234k | uriHashValue); |
9226 | 234k | res = xmlAttrHashInsert(ctxt, attrHashSize, attname, nsuri, |
9227 | 234k | hashValue, nbatts); |
9228 | 234k | if (res < 0) |
9229 | 0 | continue; |
9230 | 234k | if (res < INT_MAX) { |
9231 | 13.6k | if (aprefix == atts[res+1]) |
9232 | 3.27k | continue; |
9233 | 10.4k | xmlNsErr(ctxt, XML_NS_ERR_ATTRIBUTE_REDEFINED, |
9234 | 10.4k | "Namespaced Attribute %s in '%s' redefined\n", |
9235 | 10.4k | attname, nsuri, NULL); |
9236 | 10.4k | } |
9237 | 234k | } |
9238 | | |
9239 | 276k | xmlParserEntityCheck(ctxt, attr->expandedSize); |
9240 | | |
9241 | 276k | if ((atts == NULL) || (nbatts + 5 > maxatts)) { |
9242 | 1.54k | res = xmlCtxtGrowAttrs(ctxt); |
9243 | | |
9244 | 1.54k | maxatts = ctxt->maxatts; |
9245 | 1.54k | atts = ctxt->atts; |
9246 | | |
9247 | 1.54k | if (res < 0) { |
9248 | 0 | localname = NULL; |
9249 | 0 | goto done; |
9250 | 0 | } |
9251 | 1.54k | } |
9252 | | |
9253 | 276k | atts[nbatts++] = attname; |
9254 | 276k | atts[nbatts++] = aprefix; |
9255 | 276k | atts[nbatts++] = XML_INT_TO_PTR(nsIndex); |
9256 | 276k | atts[nbatts++] = attr->value.name; |
9257 | 276k | atts[nbatts++] = attr->valueEnd; |
9258 | | |
9259 | 276k | #ifdef LIBXML_VALID_ENABLED |
9260 | | /* |
9261 | | * This should be moved to valid.c, but we don't keep track |
9262 | | * whether an attribute was defaulted. |
9263 | | */ |
9264 | 276k | if ((ctxt->validate) && |
9265 | 276k | (ctxt->standalone == 1) && |
9266 | 276k | (attr->external != 0)) { |
9267 | 0 | xmlValidityError(ctxt, XML_DTD_STANDALONE_DEFAULTED, |
9268 | 0 | "standalone: attribute %s on %s defaulted " |
9269 | 0 | "from external subset\n", |
9270 | 0 | attname, localname); |
9271 | 0 | } |
9272 | 276k | #endif |
9273 | 276k | nbdef++; |
9274 | 276k | } |
9275 | 124k | } |
9276 | 232k | } |
9277 | | |
9278 | | /* |
9279 | | * Using a single hash table for nsUri/localName pairs cannot |
9280 | | * detect duplicate QNames reliably. The following example will |
9281 | | * only result in two namespace errors. |
9282 | | * |
9283 | | * <doc xmlns:a="a" xmlns:b="a"> |
9284 | | * <elem a:a="" b:a="" b:a=""/> |
9285 | | * </doc> |
9286 | | * |
9287 | | * If we saw more than one namespace error but no duplicate QNames |
9288 | | * were found, we have to scan for duplicate QNames. |
9289 | | */ |
9290 | 605k | if ((numDupErr == 0) && (numNsErr > 1)) { |
9291 | 1.22k | memset(ctxt->attrHash, -1, |
9292 | 1.22k | attrHashSize * sizeof(ctxt->attrHash[0])); |
9293 | | |
9294 | 11.8k | for (i = 0, j = 0; j < nratts; i += 5, j++) { |
9295 | 10.5k | unsigned hashValue, nameHashValue, prefixHashValue; |
9296 | 10.5k | int res; |
9297 | | |
9298 | 10.5k | aprefix = atts[i+1]; |
9299 | 10.5k | if (aprefix == NULL) |
9300 | 1.06k | continue; |
9301 | | |
9302 | 9.53k | attname = atts[i]; |
9303 | | /* Hash values always have bit 31 set, see dict.c */ |
9304 | 9.53k | nameHashValue = ctxt->attallocs[j] | 0x80000000; |
9305 | 9.53k | prefixHashValue = xmlDictComputeHash(ctxt->dict, aprefix); |
9306 | | |
9307 | 9.53k | hashValue = xmlDictCombineHash(nameHashValue, prefixHashValue); |
9308 | 9.53k | res = xmlAttrHashInsertQName(ctxt, attrHashSize, attname, |
9309 | 9.53k | aprefix, hashValue, i); |
9310 | 9.53k | if (res < INT_MAX) |
9311 | 5.29k | xmlErrAttributeDup(ctxt, aprefix, attname); |
9312 | 9.53k | } |
9313 | 1.22k | } |
9314 | | |
9315 | | /* |
9316 | | * Reconstruct attribute pointers |
9317 | | */ |
9318 | 1.32M | for (i = 0, j = 0; i < nbatts; i += 5, j++) { |
9319 | | /* namespace URI */ |
9320 | 720k | nsIndex = XML_PTR_TO_INT(atts[i+2]); |
9321 | 720k | if (nsIndex == INT_MAX) |
9322 | 608k | atts[i+2] = NULL; |
9323 | 112k | else if (nsIndex == INT_MAX - 1) |
9324 | 42.6k | atts[i+2] = ctxt->str_xml_ns; |
9325 | 69.5k | else |
9326 | 69.5k | atts[i+2] = ctxt->nsTab[nsIndex * 2 + 1]; |
9327 | | |
9328 | 720k | if ((j < nratts) && (ctxt->attallocs[j] & 0x80000000) == 0) { |
9329 | 388k | atts[i+3] = BASE_PTR + XML_PTR_TO_INT(atts[i+3]); /* value */ |
9330 | 388k | atts[i+4] = BASE_PTR + XML_PTR_TO_INT(atts[i+4]); /* valuend */ |
9331 | 388k | } |
9332 | 720k | } |
9333 | | |
9334 | 605k | uri = xmlParserNsLookupUri(ctxt, &hprefix); |
9335 | 605k | if ((prefix != NULL) && (uri == NULL)) { |
9336 | 24.2k | xmlNsErr(ctxt, XML_NS_ERR_UNDEFINED_NAMESPACE, |
9337 | 24.2k | "Namespace prefix %s on %s is not defined\n", |
9338 | 24.2k | prefix, localname, NULL); |
9339 | 24.2k | } |
9340 | 605k | *pref = prefix; |
9341 | 605k | *URI = uri; |
9342 | | |
9343 | | /* |
9344 | | * SAX callback |
9345 | | */ |
9346 | 605k | if ((ctxt->sax != NULL) && (ctxt->sax->startElementNs != NULL) && |
9347 | 605k | (!ctxt->disableSAX)) { |
9348 | 56.0k | if (nbNs > 0) |
9349 | 24.5k | ctxt->sax->startElementNs(ctxt->userData, localname, prefix, uri, |
9350 | 24.5k | nbNs, ctxt->nsTab + 2 * (ctxt->nsNr - nbNs), |
9351 | 24.5k | nbatts / 5, nbdef, atts); |
9352 | 31.5k | else |
9353 | 31.5k | ctxt->sax->startElementNs(ctxt->userData, localname, prefix, uri, |
9354 | 31.5k | 0, NULL, nbatts / 5, nbdef, atts); |
9355 | 56.0k | } |
9356 | | |
9357 | 605k | done: |
9358 | | /* |
9359 | | * Free allocated attribute values |
9360 | | */ |
9361 | 605k | if (attval != 0) { |
9362 | 405k | for (i = 0, j = 0; j < nratts; i += 5, j++) |
9363 | 370k | if (ctxt->attallocs[j] & 0x80000000) |
9364 | 55.1k | xmlFree((xmlChar *) atts[i+3]); |
9365 | 34.6k | } |
9366 | | |
9367 | 605k | *nbNsPtr = nbNs; |
9368 | 605k | return(localname); |
9369 | 605k | } |
9370 | | |
9371 | | /** |
9372 | | * Parse an end tag. Always consumes '</'. |
9373 | | * |
9374 | | * [42] ETag ::= '</' Name S? '>' |
9375 | | * |
9376 | | * With namespace |
9377 | | * |
9378 | | * [NS 9] ETag ::= '</' QName S? '>' |
9379 | | * @param ctxt an XML parser context |
9380 | | * @param tag the corresponding start tag |
9381 | | */ |
9382 | | |
9383 | | static void |
9384 | 35.9k | xmlParseEndTag2(xmlParserCtxtPtr ctxt, const xmlStartTag *tag) { |
9385 | 35.9k | const xmlChar *name; |
9386 | | |
9387 | 35.9k | GROW; |
9388 | 35.9k | if ((RAW != '<') || (NXT(1) != '/')) { |
9389 | 148 | xmlFatalErr(ctxt, XML_ERR_LTSLASH_REQUIRED, NULL); |
9390 | 148 | return; |
9391 | 148 | } |
9392 | 35.7k | SKIP(2); |
9393 | | |
9394 | 35.7k | if (tag->prefix == NULL) |
9395 | 22.1k | name = xmlParseNameAndCompare(ctxt, ctxt->name); |
9396 | 13.6k | else |
9397 | 13.6k | name = xmlParseQNameAndCompare(ctxt, ctxt->name, tag->prefix); |
9398 | | |
9399 | | /* |
9400 | | * We should definitely be at the ending "S? '>'" part |
9401 | | */ |
9402 | 35.7k | GROW; |
9403 | 35.7k | SKIP_BLANKS; |
9404 | 35.7k | if ((!IS_BYTE_CHAR(RAW)) || (RAW != '>')) { |
9405 | 27.3k | xmlFatalErr(ctxt, XML_ERR_GT_REQUIRED, NULL); |
9406 | 27.3k | } else |
9407 | 8.43k | NEXT1; |
9408 | | |
9409 | | /* |
9410 | | * [ WFC: Element Type Match ] |
9411 | | * The Name in an element's end-tag must match the element type in the |
9412 | | * start-tag. |
9413 | | * |
9414 | | */ |
9415 | 35.7k | if (name != (xmlChar*)1) { |
9416 | 20.7k | if (name == NULL) name = BAD_CAST "unparsable"; |
9417 | 20.7k | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NAME_MISMATCH, |
9418 | 20.7k | "Opening and ending tag mismatch: %s line %d and %s\n", |
9419 | 20.7k | ctxt->name, tag->line, name); |
9420 | 20.7k | } |
9421 | | |
9422 | | /* |
9423 | | * SAX: End of Tag |
9424 | | */ |
9425 | 35.7k | if ((ctxt->sax != NULL) && (ctxt->sax->endElementNs != NULL) && |
9426 | 35.7k | (!ctxt->disableSAX)) |
9427 | 868 | ctxt->sax->endElementNs(ctxt->userData, ctxt->name, tag->prefix, |
9428 | 868 | tag->URI); |
9429 | | |
9430 | 35.7k | spacePop(ctxt); |
9431 | 35.7k | if (tag->nsNr != 0) |
9432 | 4.90k | xmlParserNsPop(ctxt, tag->nsNr); |
9433 | 35.7k | } |
9434 | | |
9435 | | /** |
9436 | | * Parse escaped pure raw content. Always consumes '<!['. |
9437 | | * |
9438 | | * @deprecated Internal function, don't use. |
9439 | | * |
9440 | | * [18] CDSect ::= CDStart CData CDEnd |
9441 | | * |
9442 | | * [19] CDStart ::= '<![CDATA[' |
9443 | | * |
9444 | | * [20] Data ::= (Char* - (Char* ']]>' Char*)) |
9445 | | * |
9446 | | * [21] CDEnd ::= ']]>' |
9447 | | * @param ctxt an XML parser context |
9448 | | */ |
9449 | | void |
9450 | 29.8k | xmlParseCDSect(xmlParserCtxt *ctxt) { |
9451 | 29.8k | xmlChar *buf = NULL; |
9452 | 29.8k | int len = 0; |
9453 | 29.8k | int size = XML_PARSER_BUFFER_SIZE; |
9454 | 29.8k | int r, rl; |
9455 | 29.8k | int s, sl; |
9456 | 29.8k | int cur, l; |
9457 | 29.8k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
9458 | 29.8k | XML_MAX_HUGE_LENGTH : |
9459 | 29.8k | XML_MAX_TEXT_LENGTH; |
9460 | | |
9461 | 29.8k | if ((CUR != '<') || (NXT(1) != '!') || (NXT(2) != '[')) |
9462 | 0 | return; |
9463 | 29.8k | SKIP(3); |
9464 | | |
9465 | 29.8k | if (!CMP6(CUR_PTR, 'C', 'D', 'A', 'T', 'A', '[')) |
9466 | 0 | return; |
9467 | 29.8k | SKIP(6); |
9468 | | |
9469 | 29.8k | r = xmlCurrentCharRecover(ctxt, &rl); |
9470 | 29.8k | if (!IS_CHAR(r)) { |
9471 | 5.52k | xmlFatalErr(ctxt, XML_ERR_CDATA_NOT_FINISHED, NULL); |
9472 | 5.52k | goto out; |
9473 | 5.52k | } |
9474 | 24.2k | NEXTL(rl); |
9475 | 24.2k | s = xmlCurrentCharRecover(ctxt, &sl); |
9476 | 24.2k | if (!IS_CHAR(s)) { |
9477 | 11.4k | xmlFatalErr(ctxt, XML_ERR_CDATA_NOT_FINISHED, NULL); |
9478 | 11.4k | goto out; |
9479 | 11.4k | } |
9480 | 12.8k | NEXTL(sl); |
9481 | 12.8k | cur = xmlCurrentCharRecover(ctxt, &l); |
9482 | 12.8k | buf = xmlMalloc(size); |
9483 | 12.8k | if (buf == NULL) { |
9484 | 0 | xmlErrMemory(ctxt); |
9485 | 0 | goto out; |
9486 | 0 | } |
9487 | 4.11M | while (IS_CHAR(cur) && |
9488 | 4.11M | ((r != ']') || (s != ']') || (cur != '>'))) { |
9489 | 4.10M | if (len + 5 >= size) { |
9490 | 7.31k | xmlChar *tmp; |
9491 | 7.31k | int newSize; |
9492 | | |
9493 | 7.31k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
9494 | 7.31k | if (newSize < 0) { |
9495 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_CDATA_NOT_FINISHED, |
9496 | 0 | "CData section too big found\n"); |
9497 | 0 | goto out; |
9498 | 0 | } |
9499 | 7.31k | tmp = xmlRealloc(buf, newSize); |
9500 | 7.31k | if (tmp == NULL) { |
9501 | 0 | xmlErrMemory(ctxt); |
9502 | 0 | goto out; |
9503 | 0 | } |
9504 | 7.31k | buf = tmp; |
9505 | 7.31k | size = newSize; |
9506 | 7.31k | } |
9507 | 4.10M | COPY_BUF(buf, len, r); |
9508 | 4.10M | r = s; |
9509 | 4.10M | rl = sl; |
9510 | 4.10M | s = cur; |
9511 | 4.10M | sl = l; |
9512 | 4.10M | NEXTL(l); |
9513 | 4.10M | cur = xmlCurrentCharRecover(ctxt, &l); |
9514 | 4.10M | } |
9515 | 12.8k | buf[len] = 0; |
9516 | 12.8k | if (cur != '>') { |
9517 | 8.51k | xmlFatalErrMsgStr(ctxt, XML_ERR_CDATA_NOT_FINISHED, |
9518 | 8.51k | "CData section not finished\n%.50s\n", buf); |
9519 | 8.51k | goto out; |
9520 | 8.51k | } |
9521 | 4.28k | NEXTL(l); |
9522 | | |
9523 | | /* |
9524 | | * OK the buffer is to be consumed as cdata. |
9525 | | */ |
9526 | 4.28k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { |
9527 | 3.40k | if ((ctxt->sax->cdataBlock != NULL) && |
9528 | 3.40k | ((ctxt->options & XML_PARSE_NOCDATA) == 0)) { |
9529 | 3.40k | ctxt->sax->cdataBlock(ctxt->userData, buf, len); |
9530 | 3.40k | } else if (ctxt->sax->characters != NULL) { |
9531 | 0 | ctxt->sax->characters(ctxt->userData, buf, len); |
9532 | 0 | } |
9533 | 3.40k | } |
9534 | | |
9535 | 29.8k | out: |
9536 | 29.8k | xmlFree(buf); |
9537 | 29.8k | } |
9538 | | |
9539 | | /** |
9540 | | * Parse a content sequence. Stops at EOF or '</'. Leaves checking of |
9541 | | * unexpected EOF to the caller. |
9542 | | * |
9543 | | * @param ctxt an XML parser context |
9544 | | */ |
9545 | | |
9546 | | static void |
9547 | 12.9k | xmlParseContentInternal(xmlParserCtxtPtr ctxt) { |
9548 | 12.9k | int oldNameNr = ctxt->nameNr; |
9549 | 12.9k | int oldSpaceNr = ctxt->spaceNr; |
9550 | 12.9k | int oldNodeNr = ctxt->nodeNr; |
9551 | | |
9552 | 12.9k | GROW; |
9553 | 5.96M | while ((ctxt->input->cur < ctxt->input->end) && |
9554 | 5.96M | (PARSER_STOPPED(ctxt) == 0)) { |
9555 | 5.95M | const xmlChar *cur = ctxt->input->cur; |
9556 | | |
9557 | | /* |
9558 | | * First case : a Processing Instruction. |
9559 | | */ |
9560 | 5.95M | if ((*cur == '<') && (cur[1] == '?')) { |
9561 | 27.8k | xmlParsePI(ctxt); |
9562 | 27.8k | } |
9563 | | |
9564 | | /* |
9565 | | * Second case : a CDSection |
9566 | | */ |
9567 | | /* 2.6.0 test was *cur not RAW */ |
9568 | 5.92M | else if (CMP9(CUR_PTR, '<', '!', '[', 'C', 'D', 'A', 'T', 'A', '[')) { |
9569 | 29.8k | xmlParseCDSect(ctxt); |
9570 | 29.8k | } |
9571 | | |
9572 | | /* |
9573 | | * Third case : a comment |
9574 | | */ |
9575 | 5.89M | else if ((*cur == '<') && (NXT(1) == '!') && |
9576 | 5.89M | (NXT(2) == '-') && (NXT(3) == '-')) { |
9577 | 42.4k | xmlParseComment(ctxt); |
9578 | 42.4k | } |
9579 | | |
9580 | | /* |
9581 | | * Fourth case : a sub-element. |
9582 | | */ |
9583 | 5.85M | else if (*cur == '<') { |
9584 | 2.24M | if (NXT(1) == '/') { |
9585 | 35.7k | if (ctxt->nameNr <= oldNameNr) |
9586 | 206 | break; |
9587 | 35.5k | xmlParseElementEnd(ctxt); |
9588 | 2.20M | } else { |
9589 | 2.20M | xmlParseElementStart(ctxt); |
9590 | 2.20M | } |
9591 | 2.24M | } |
9592 | | |
9593 | | /* |
9594 | | * Fifth case : a reference. If if has not been resolved, |
9595 | | * parsing returns it's Name, create the node |
9596 | | */ |
9597 | | |
9598 | 3.61M | else if (*cur == '&') { |
9599 | 166k | xmlParseReference(ctxt); |
9600 | 166k | } |
9601 | | |
9602 | | /* |
9603 | | * Last case, text. Note that References are handled directly. |
9604 | | */ |
9605 | 3.44M | else { |
9606 | 3.44M | xmlParseCharDataInternal(ctxt, 0); |
9607 | 3.44M | } |
9608 | | |
9609 | 5.95M | SHRINK; |
9610 | 5.95M | GROW; |
9611 | 5.95M | } |
9612 | | |
9613 | 12.9k | if ((ctxt->nameNr > oldNameNr) && |
9614 | 12.9k | (ctxt->input->cur >= ctxt->input->end) && |
9615 | 12.9k | (ctxt->wellFormed)) { |
9616 | 186 | const xmlChar *name = ctxt->nameTab[ctxt->nameNr - 1]; |
9617 | 186 | int line = ctxt->pushTab[ctxt->nameNr - 1].line; |
9618 | 186 | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NOT_FINISHED, |
9619 | 186 | "Premature end of data in tag %s line %d\n", |
9620 | 186 | name, line, NULL); |
9621 | 186 | } |
9622 | | |
9623 | | /* |
9624 | | * Clean up in error case |
9625 | | */ |
9626 | | |
9627 | 58.6k | while (ctxt->nodeNr > oldNodeNr) |
9628 | 45.6k | nodePop(ctxt); |
9629 | | |
9630 | 291k | while (ctxt->nameNr > oldNameNr) { |
9631 | 278k | xmlStartTag *tag = &ctxt->pushTab[ctxt->nameNr - 1]; |
9632 | | |
9633 | 278k | if (tag->nsNr != 0) |
9634 | 130k | xmlParserNsPop(ctxt, tag->nsNr); |
9635 | | |
9636 | 278k | namePop(ctxt); |
9637 | 278k | } |
9638 | | |
9639 | 291k | while (ctxt->spaceNr > oldSpaceNr) |
9640 | 278k | spacePop(ctxt); |
9641 | 12.9k | } |
9642 | | |
9643 | | /** |
9644 | | * Parse XML element content. This is useful if you're only interested |
9645 | | * in custom SAX callbacks. If you want a node list, use |
9646 | | * #xmlCtxtParseContent. |
9647 | | * |
9648 | | * @param ctxt an XML parser context |
9649 | | */ |
9650 | | void |
9651 | 0 | xmlParseContent(xmlParserCtxt *ctxt) { |
9652 | 0 | if ((ctxt == NULL) || (ctxt->input == NULL)) |
9653 | 0 | return; |
9654 | | |
9655 | 0 | xmlCtxtInitializeLate(ctxt); |
9656 | |
|
9657 | 0 | xmlParseContentInternal(ctxt); |
9658 | |
|
9659 | 0 | xmlParserCheckEOF(ctxt, XML_ERR_NOT_WELL_BALANCED); |
9660 | 0 | } |
9661 | | |
9662 | | /** |
9663 | | * parse an XML element |
9664 | | * |
9665 | | * @deprecated Internal function, don't use. |
9666 | | * |
9667 | | * [39] element ::= EmptyElemTag | STag content ETag |
9668 | | * |
9669 | | * [ WFC: Element Type Match ] |
9670 | | * The Name in an element's end-tag must match the element type in the |
9671 | | * start-tag. |
9672 | | * |
9673 | | * @param ctxt an XML parser context |
9674 | | */ |
9675 | | |
9676 | | void |
9677 | 9.43k | xmlParseElement(xmlParserCtxt *ctxt) { |
9678 | 9.43k | if (xmlParseElementStart(ctxt) != 0) |
9679 | 3.62k | return; |
9680 | | |
9681 | 5.81k | xmlParseContentInternal(ctxt); |
9682 | | |
9683 | 5.81k | if (ctxt->input->cur >= ctxt->input->end) { |
9684 | 5.46k | if (ctxt->wellFormed) { |
9685 | 450 | const xmlChar *name = ctxt->nameTab[ctxt->nameNr - 1]; |
9686 | 450 | int line = ctxt->pushTab[ctxt->nameNr - 1].line; |
9687 | 450 | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NOT_FINISHED, |
9688 | 450 | "Premature end of data in tag %s line %d\n", |
9689 | 450 | name, line, NULL); |
9690 | 450 | } |
9691 | 5.46k | return; |
9692 | 5.46k | } |
9693 | | |
9694 | 353 | xmlParseElementEnd(ctxt); |
9695 | 353 | } |
9696 | | |
9697 | | /** |
9698 | | * Parse the start of an XML element. Returns -1 in case of error, 0 if an |
9699 | | * opening tag was parsed, 1 if an empty element was parsed. |
9700 | | * |
9701 | | * Always consumes '<'. |
9702 | | * |
9703 | | * @param ctxt an XML parser context |
9704 | | */ |
9705 | | static int |
9706 | 2.21M | xmlParseElementStart(xmlParserCtxtPtr ctxt) { |
9707 | 2.21M | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 2048 : 256; |
9708 | 2.21M | const xmlChar *name; |
9709 | 2.21M | const xmlChar *prefix = NULL; |
9710 | 2.21M | const xmlChar *URI = NULL; |
9711 | 2.21M | xmlParserNodeInfo node_info; |
9712 | 2.21M | int line; |
9713 | 2.21M | xmlNodePtr cur; |
9714 | 2.21M | int nbNs = 0; |
9715 | | |
9716 | 2.21M | if (ctxt->nameNr > maxDepth) { |
9717 | 11 | xmlFatalErrMsgInt(ctxt, XML_ERR_RESOURCE_LIMIT, |
9718 | 11 | "Excessive depth in document: %d use XML_PARSE_HUGE option\n", |
9719 | 11 | ctxt->nameNr); |
9720 | 11 | xmlHaltParser(ctxt); |
9721 | 11 | return(-1); |
9722 | 11 | } |
9723 | | |
9724 | | /* Capture start position */ |
9725 | 2.21M | if (ctxt->record_info) { |
9726 | 0 | node_info.begin_pos = ctxt->input->consumed + |
9727 | 0 | (CUR_PTR - ctxt->input->base); |
9728 | 0 | node_info.begin_line = ctxt->input->line; |
9729 | 0 | } |
9730 | | |
9731 | 2.21M | if (ctxt->spaceNr == 0) |
9732 | 0 | spacePush(ctxt, -1); |
9733 | 2.21M | else if (*ctxt->space == -2) |
9734 | 0 | spacePush(ctxt, -1); |
9735 | 2.21M | else |
9736 | 2.21M | spacePush(ctxt, *ctxt->space); |
9737 | | |
9738 | 2.21M | line = ctxt->input->line; |
9739 | 2.21M | #ifdef LIBXML_SAX1_ENABLED |
9740 | 2.21M | if (ctxt->sax2) |
9741 | 2.21M | #endif /* LIBXML_SAX1_ENABLED */ |
9742 | 2.21M | name = xmlParseStartTag2(ctxt, &prefix, &URI, &nbNs); |
9743 | 0 | #ifdef LIBXML_SAX1_ENABLED |
9744 | 0 | else |
9745 | 0 | name = xmlParseStartTag(ctxt); |
9746 | 2.21M | #endif /* LIBXML_SAX1_ENABLED */ |
9747 | 2.21M | if (name == NULL) { |
9748 | 1.60M | spacePop(ctxt); |
9749 | 1.60M | return(-1); |
9750 | 1.60M | } |
9751 | 605k | nameNsPush(ctxt, name, prefix, URI, line, nbNs); |
9752 | 605k | cur = ctxt->node; |
9753 | | |
9754 | 605k | #ifdef LIBXML_VALID_ENABLED |
9755 | | /* |
9756 | | * [ VC: Root Element Type ] |
9757 | | * The Name in the document type declaration must match the element |
9758 | | * type of the root element. |
9759 | | */ |
9760 | 605k | if (ctxt->validate && ctxt->wellFormed && ctxt->myDoc && |
9761 | 605k | ctxt->node && (ctxt->node == ctxt->myDoc->children)) |
9762 | 0 | ctxt->valid &= xmlValidateRoot(&ctxt->vctxt, ctxt->myDoc); |
9763 | 605k | #endif /* LIBXML_VALID_ENABLED */ |
9764 | | |
9765 | | /* |
9766 | | * Check for an Empty Element. |
9767 | | */ |
9768 | 605k | if ((RAW == '/') && (NXT(1) == '>')) { |
9769 | 23.4k | SKIP(2); |
9770 | 23.4k | if (ctxt->sax2) { |
9771 | 23.4k | if ((ctxt->sax != NULL) && (ctxt->sax->endElementNs != NULL) && |
9772 | 23.4k | (!ctxt->disableSAX)) |
9773 | 3.31k | ctxt->sax->endElementNs(ctxt->userData, name, prefix, URI); |
9774 | 23.4k | #ifdef LIBXML_SAX1_ENABLED |
9775 | 23.4k | } else { |
9776 | 0 | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL) && |
9777 | 0 | (!ctxt->disableSAX)) |
9778 | 0 | ctxt->sax->endElement(ctxt->userData, name); |
9779 | 0 | #endif /* LIBXML_SAX1_ENABLED */ |
9780 | 0 | } |
9781 | 23.4k | namePop(ctxt); |
9782 | 23.4k | spacePop(ctxt); |
9783 | 23.4k | if (nbNs > 0) |
9784 | 3.20k | xmlParserNsPop(ctxt, nbNs); |
9785 | 23.4k | if (cur != NULL && ctxt->record_info) { |
9786 | 0 | node_info.node = cur; |
9787 | 0 | node_info.end_pos = ctxt->input->consumed + |
9788 | 0 | (CUR_PTR - ctxt->input->base); |
9789 | 0 | node_info.end_line = ctxt->input->line; |
9790 | 0 | xmlParserAddNodeInfo(ctxt, &node_info); |
9791 | 0 | } |
9792 | 23.4k | return(1); |
9793 | 23.4k | } |
9794 | 582k | if (RAW == '>') { |
9795 | 319k | NEXT1; |
9796 | 319k | if (cur != NULL && ctxt->record_info) { |
9797 | 0 | node_info.node = cur; |
9798 | 0 | node_info.end_pos = 0; |
9799 | 0 | node_info.end_line = 0; |
9800 | 0 | xmlParserAddNodeInfo(ctxt, &node_info); |
9801 | 0 | } |
9802 | 319k | } else { |
9803 | 262k | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_GT_REQUIRED, |
9804 | 262k | "Couldn't find end of Start Tag %s line %d\n", |
9805 | 262k | name, line, NULL); |
9806 | | |
9807 | | /* |
9808 | | * end of parsing of this node. |
9809 | | */ |
9810 | 262k | nodePop(ctxt); |
9811 | 262k | namePop(ctxt); |
9812 | 262k | spacePop(ctxt); |
9813 | 262k | if (nbNs > 0) |
9814 | 89.7k | xmlParserNsPop(ctxt, nbNs); |
9815 | 262k | return(-1); |
9816 | 262k | } |
9817 | | |
9818 | 319k | return(0); |
9819 | 582k | } |
9820 | | |
9821 | | /** |
9822 | | * Parse the end of an XML element. Always consumes '</'. |
9823 | | * |
9824 | | * @param ctxt an XML parser context |
9825 | | */ |
9826 | | static void |
9827 | 35.9k | xmlParseElementEnd(xmlParserCtxtPtr ctxt) { |
9828 | 35.9k | xmlNodePtr cur = ctxt->node; |
9829 | | |
9830 | 35.9k | if (ctxt->nameNr <= 0) { |
9831 | 0 | if ((RAW == '<') && (NXT(1) == '/')) |
9832 | 0 | SKIP(2); |
9833 | 0 | return; |
9834 | 0 | } |
9835 | | |
9836 | | /* |
9837 | | * parse the end of tag: '</' should be here. |
9838 | | */ |
9839 | 35.9k | if (ctxt->sax2) { |
9840 | 35.9k | xmlParseEndTag2(ctxt, &ctxt->pushTab[ctxt->nameNr - 1]); |
9841 | 35.9k | namePop(ctxt); |
9842 | 35.9k | } |
9843 | 0 | #ifdef LIBXML_SAX1_ENABLED |
9844 | 0 | else |
9845 | 0 | xmlParseEndTag1(ctxt, 0); |
9846 | 35.9k | #endif /* LIBXML_SAX1_ENABLED */ |
9847 | | |
9848 | | /* |
9849 | | * Capture end position |
9850 | | */ |
9851 | 35.9k | if (cur != NULL && ctxt->record_info) { |
9852 | 0 | xmlParserNodeInfoPtr node_info; |
9853 | |
|
9854 | 0 | node_info = (xmlParserNodeInfoPtr) xmlParserFindNodeInfo(ctxt, cur); |
9855 | 0 | if (node_info != NULL) { |
9856 | 0 | node_info->end_pos = ctxt->input->consumed + |
9857 | 0 | (CUR_PTR - ctxt->input->base); |
9858 | 0 | node_info->end_line = ctxt->input->line; |
9859 | 0 | } |
9860 | 0 | } |
9861 | 35.9k | } |
9862 | | |
9863 | | /** |
9864 | | * parse the XML version value. |
9865 | | * |
9866 | | * @deprecated Internal function, don't use. |
9867 | | * |
9868 | | * [26] VersionNum ::= '1.' [0-9]+ |
9869 | | * |
9870 | | * In practice allow [0-9].[0-9]+ at that level |
9871 | | * |
9872 | | * @param ctxt an XML parser context |
9873 | | * @returns the string giving the XML version number, or NULL |
9874 | | */ |
9875 | | xmlChar * |
9876 | 638 | xmlParseVersionNum(xmlParserCtxt *ctxt) { |
9877 | 638 | xmlChar *buf = NULL; |
9878 | 638 | int len = 0; |
9879 | 638 | int size = 10; |
9880 | 638 | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
9881 | 638 | XML_MAX_TEXT_LENGTH : |
9882 | 638 | XML_MAX_NAME_LENGTH; |
9883 | 638 | xmlChar cur; |
9884 | | |
9885 | 638 | buf = xmlMalloc(size); |
9886 | 638 | if (buf == NULL) { |
9887 | 0 | xmlErrMemory(ctxt); |
9888 | 0 | return(NULL); |
9889 | 0 | } |
9890 | 638 | cur = CUR; |
9891 | 638 | if (!((cur >= '0') && (cur <= '9'))) { |
9892 | 35 | xmlFree(buf); |
9893 | 35 | return(NULL); |
9894 | 35 | } |
9895 | 603 | buf[len++] = cur; |
9896 | 603 | NEXT; |
9897 | 603 | cur=CUR; |
9898 | 603 | if (cur != '.') { |
9899 | 14 | xmlFree(buf); |
9900 | 14 | return(NULL); |
9901 | 14 | } |
9902 | 589 | buf[len++] = cur; |
9903 | 589 | NEXT; |
9904 | 589 | cur=CUR; |
9905 | 46.8k | while ((cur >= '0') && (cur <= '9')) { |
9906 | 46.2k | if (len + 1 >= size) { |
9907 | 129 | xmlChar *tmp; |
9908 | 129 | int newSize; |
9909 | | |
9910 | 129 | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
9911 | 129 | if (newSize < 0) { |
9912 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "VersionNum"); |
9913 | 0 | xmlFree(buf); |
9914 | 0 | return(NULL); |
9915 | 0 | } |
9916 | 129 | tmp = xmlRealloc(buf, newSize); |
9917 | 129 | if (tmp == NULL) { |
9918 | 0 | xmlErrMemory(ctxt); |
9919 | 0 | xmlFree(buf); |
9920 | 0 | return(NULL); |
9921 | 0 | } |
9922 | 129 | buf = tmp; |
9923 | 129 | size = newSize; |
9924 | 129 | } |
9925 | 46.2k | buf[len++] = cur; |
9926 | 46.2k | NEXT; |
9927 | 46.2k | cur=CUR; |
9928 | 46.2k | } |
9929 | 589 | buf[len] = 0; |
9930 | 589 | return(buf); |
9931 | 589 | } |
9932 | | |
9933 | | /** |
9934 | | * parse the XML version. |
9935 | | * |
9936 | | * @deprecated Internal function, don't use. |
9937 | | * |
9938 | | * [24] VersionInfo ::= S 'version' Eq (' VersionNum ' | " VersionNum ") |
9939 | | * |
9940 | | * [25] Eq ::= S? '=' S? |
9941 | | * |
9942 | | * @param ctxt an XML parser context |
9943 | | * @returns the version string, e.g. "1.0" |
9944 | | */ |
9945 | | |
9946 | | xmlChar * |
9947 | 2.58k | xmlParseVersionInfo(xmlParserCtxt *ctxt) { |
9948 | 2.58k | xmlChar *version = NULL; |
9949 | | |
9950 | 2.58k | if (CMP7(CUR_PTR, 'v', 'e', 'r', 's', 'i', 'o', 'n')) { |
9951 | 719 | SKIP(7); |
9952 | 719 | SKIP_BLANKS; |
9953 | 719 | if (RAW != '=') { |
9954 | 16 | xmlFatalErr(ctxt, XML_ERR_EQUAL_REQUIRED, NULL); |
9955 | 16 | return(NULL); |
9956 | 16 | } |
9957 | 703 | NEXT; |
9958 | 703 | SKIP_BLANKS; |
9959 | 703 | if (RAW == '"') { |
9960 | 146 | NEXT; |
9961 | 146 | version = xmlParseVersionNum(ctxt); |
9962 | 146 | if (RAW != '"') { |
9963 | 18 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
9964 | 18 | } else |
9965 | 128 | NEXT; |
9966 | 557 | } else if (RAW == '\''){ |
9967 | 492 | NEXT; |
9968 | 492 | version = xmlParseVersionNum(ctxt); |
9969 | 492 | if (RAW != '\'') { |
9970 | 54 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
9971 | 54 | } else |
9972 | 438 | NEXT; |
9973 | 492 | } else { |
9974 | 65 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_STARTED, NULL); |
9975 | 65 | } |
9976 | 703 | } |
9977 | 2.56k | return(version); |
9978 | 2.58k | } |
9979 | | |
9980 | | /** |
9981 | | * parse the XML encoding name |
9982 | | * |
9983 | | * @deprecated Internal function, don't use. |
9984 | | * |
9985 | | * [81] EncName ::= [A-Za-z] ([A-Za-z0-9._] | '-')* |
9986 | | * |
9987 | | * @param ctxt an XML parser context |
9988 | | * @returns the encoding name value or NULL |
9989 | | */ |
9990 | | xmlChar * |
9991 | 1.12k | xmlParseEncName(xmlParserCtxt *ctxt) { |
9992 | 1.12k | xmlChar *buf = NULL; |
9993 | 1.12k | int len = 0; |
9994 | 1.12k | int size = 10; |
9995 | 1.12k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
9996 | 1.12k | XML_MAX_TEXT_LENGTH : |
9997 | 1.12k | XML_MAX_NAME_LENGTH; |
9998 | 1.12k | xmlChar cur; |
9999 | | |
10000 | 1.12k | cur = CUR; |
10001 | 1.12k | if (((cur >= 'a') && (cur <= 'z')) || |
10002 | 1.12k | ((cur >= 'A') && (cur <= 'Z'))) { |
10003 | 1.11k | buf = xmlMalloc(size); |
10004 | 1.11k | if (buf == NULL) { |
10005 | 0 | xmlErrMemory(ctxt); |
10006 | 0 | return(NULL); |
10007 | 0 | } |
10008 | | |
10009 | 1.11k | buf[len++] = cur; |
10010 | 1.11k | NEXT; |
10011 | 1.11k | cur = CUR; |
10012 | 1.05M | while (((cur >= 'a') && (cur <= 'z')) || |
10013 | 1.05M | ((cur >= 'A') && (cur <= 'Z')) || |
10014 | 1.05M | ((cur >= '0') && (cur <= '9')) || |
10015 | 1.05M | (cur == '.') || (cur == '_') || |
10016 | 1.05M | (cur == '-')) { |
10017 | 1.05M | if (len + 1 >= size) { |
10018 | 823 | xmlChar *tmp; |
10019 | 823 | int newSize; |
10020 | | |
10021 | 823 | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
10022 | 823 | if (newSize < 0) { |
10023 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "EncName"); |
10024 | 0 | xmlFree(buf); |
10025 | 0 | return(NULL); |
10026 | 0 | } |
10027 | 823 | tmp = xmlRealloc(buf, newSize); |
10028 | 823 | if (tmp == NULL) { |
10029 | 0 | xmlErrMemory(ctxt); |
10030 | 0 | xmlFree(buf); |
10031 | 0 | return(NULL); |
10032 | 0 | } |
10033 | 823 | buf = tmp; |
10034 | 823 | size = newSize; |
10035 | 823 | } |
10036 | 1.05M | buf[len++] = cur; |
10037 | 1.05M | NEXT; |
10038 | 1.05M | cur = CUR; |
10039 | 1.05M | } |
10040 | 1.11k | buf[len] = 0; |
10041 | 1.11k | } else { |
10042 | 14 | xmlFatalErr(ctxt, XML_ERR_ENCODING_NAME, NULL); |
10043 | 14 | } |
10044 | 1.12k | return(buf); |
10045 | 1.12k | } |
10046 | | |
10047 | | /** |
10048 | | * parse the XML encoding declaration |
10049 | | * |
10050 | | * @deprecated Internal function, don't use. |
10051 | | * |
10052 | | * [80] EncodingDecl ::= S 'encoding' Eq ('"' EncName '"' | |
10053 | | * "'" EncName "'") |
10054 | | * |
10055 | | * this setups the conversion filters. |
10056 | | * |
10057 | | * @param ctxt an XML parser context |
10058 | | * @returns the encoding value or NULL |
10059 | | */ |
10060 | | |
10061 | | const xmlChar * |
10062 | 2.56k | xmlParseEncodingDecl(xmlParserCtxt *ctxt) { |
10063 | 2.56k | xmlChar *encoding = NULL; |
10064 | | |
10065 | 2.56k | SKIP_BLANKS; |
10066 | 2.56k | if (CMP8(CUR_PTR, 'e', 'n', 'c', 'o', 'd', 'i', 'n', 'g') == 0) |
10067 | 1.42k | return(NULL); |
10068 | | |
10069 | 1.13k | SKIP(8); |
10070 | 1.13k | SKIP_BLANKS; |
10071 | 1.13k | if (RAW != '=') { |
10072 | 8 | xmlFatalErr(ctxt, XML_ERR_EQUAL_REQUIRED, NULL); |
10073 | 8 | return(NULL); |
10074 | 8 | } |
10075 | 1.12k | NEXT; |
10076 | 1.12k | SKIP_BLANKS; |
10077 | 1.12k | if (RAW == '"') { |
10078 | 671 | NEXT; |
10079 | 671 | encoding = xmlParseEncName(ctxt); |
10080 | 671 | if (RAW != '"') { |
10081 | 93 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10082 | 93 | xmlFree((xmlChar *) encoding); |
10083 | 93 | return(NULL); |
10084 | 93 | } else |
10085 | 578 | NEXT; |
10086 | 671 | } else if (RAW == '\''){ |
10087 | 453 | NEXT; |
10088 | 453 | encoding = xmlParseEncName(ctxt); |
10089 | 453 | if (RAW != '\'') { |
10090 | 31 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10091 | 31 | xmlFree((xmlChar *) encoding); |
10092 | 31 | return(NULL); |
10093 | 31 | } else |
10094 | 422 | NEXT; |
10095 | 453 | } else { |
10096 | 4 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_STARTED, NULL); |
10097 | 4 | } |
10098 | | |
10099 | 1.00k | if (encoding == NULL) |
10100 | 5 | return(NULL); |
10101 | | |
10102 | 999 | xmlSetDeclaredEncoding(ctxt, encoding); |
10103 | | |
10104 | 999 | return(ctxt->encoding); |
10105 | 1.00k | } |
10106 | | |
10107 | | /** |
10108 | | * parse the XML standalone declaration |
10109 | | * |
10110 | | * @deprecated Internal function, don't use. |
10111 | | * |
10112 | | * [32] SDDecl ::= S 'standalone' Eq |
10113 | | * (("'" ('yes' | 'no') "'") | ('"' ('yes' | 'no')'"')) |
10114 | | * |
10115 | | * [ VC: Standalone Document Declaration ] |
10116 | | * TODO The standalone document declaration must have the value "no" |
10117 | | * if any external markup declarations contain declarations of: |
10118 | | * - attributes with default values, if elements to which these |
10119 | | * attributes apply appear in the document without specifications |
10120 | | * of values for these attributes, or |
10121 | | * - entities (other than amp, lt, gt, apos, quot), if references |
10122 | | * to those entities appear in the document, or |
10123 | | * - attributes with values subject to normalization, where the |
10124 | | * attribute appears in the document with a value which will change |
10125 | | * as a result of normalization, or |
10126 | | * - element types with element content, if white space occurs directly |
10127 | | * within any instance of those types. |
10128 | | * |
10129 | | * @param ctxt an XML parser context |
10130 | | * @returns |
10131 | | * 1 if standalone="yes" |
10132 | | * 0 if standalone="no" |
10133 | | * -2 if standalone attribute is missing or invalid |
10134 | | * (A standalone value of -2 means that the XML declaration was found, |
10135 | | * but no value was specified for the standalone attribute). |
10136 | | */ |
10137 | | |
10138 | | int |
10139 | 2.51k | xmlParseSDDecl(xmlParserCtxt *ctxt) { |
10140 | 2.51k | int standalone = -2; |
10141 | | |
10142 | 2.51k | SKIP_BLANKS; |
10143 | 2.51k | if (CMP10(CUR_PTR, 's', 't', 'a', 'n', 'd', 'a', 'l', 'o', 'n', 'e')) { |
10144 | 349 | SKIP(10); |
10145 | 349 | SKIP_BLANKS; |
10146 | 349 | if (RAW != '=') { |
10147 | 8 | xmlFatalErr(ctxt, XML_ERR_EQUAL_REQUIRED, NULL); |
10148 | 8 | return(standalone); |
10149 | 8 | } |
10150 | 341 | NEXT; |
10151 | 341 | SKIP_BLANKS; |
10152 | 341 | if (RAW == '\''){ |
10153 | 296 | NEXT; |
10154 | 296 | if ((RAW == 'n') && (NXT(1) == 'o')) { |
10155 | 284 | standalone = 0; |
10156 | 284 | SKIP(2); |
10157 | 284 | } else if ((RAW == 'y') && (NXT(1) == 'e') && |
10158 | 12 | (NXT(2) == 's')) { |
10159 | 3 | standalone = 1; |
10160 | 3 | SKIP(3); |
10161 | 9 | } else { |
10162 | 9 | xmlFatalErr(ctxt, XML_ERR_STANDALONE_VALUE, NULL); |
10163 | 9 | } |
10164 | 296 | if (RAW != '\'') { |
10165 | 15 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10166 | 15 | } else |
10167 | 281 | NEXT; |
10168 | 296 | } else if (RAW == '"'){ |
10169 | 40 | NEXT; |
10170 | 40 | if ((RAW == 'n') && (NXT(1) == 'o')) { |
10171 | 2 | standalone = 0; |
10172 | 2 | SKIP(2); |
10173 | 38 | } else if ((RAW == 'y') && (NXT(1) == 'e') && |
10174 | 38 | (NXT(2) == 's')) { |
10175 | 19 | standalone = 1; |
10176 | 19 | SKIP(3); |
10177 | 19 | } else { |
10178 | 19 | xmlFatalErr(ctxt, XML_ERR_STANDALONE_VALUE, NULL); |
10179 | 19 | } |
10180 | 40 | if (RAW != '"') { |
10181 | 23 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10182 | 23 | } else |
10183 | 17 | NEXT; |
10184 | 40 | } else { |
10185 | 5 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_STARTED, NULL); |
10186 | 5 | } |
10187 | 341 | } |
10188 | 2.51k | return(standalone); |
10189 | 2.51k | } |
10190 | | |
10191 | | /** |
10192 | | * parse an XML declaration header |
10193 | | * |
10194 | | * @deprecated Internal function, don't use. |
10195 | | * |
10196 | | * [23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl? SDDecl? S? '?>' |
10197 | | * @param ctxt an XML parser context |
10198 | | */ |
10199 | | |
10200 | | void |
10201 | 2.58k | xmlParseXMLDecl(xmlParserCtxt *ctxt) { |
10202 | 2.58k | xmlChar *version; |
10203 | | |
10204 | | /* |
10205 | | * This value for standalone indicates that the document has an |
10206 | | * XML declaration but it does not have a standalone attribute. |
10207 | | * It will be overwritten later if a standalone attribute is found. |
10208 | | */ |
10209 | | |
10210 | 2.58k | ctxt->standalone = -2; |
10211 | | |
10212 | | /* |
10213 | | * We know that '<?xml' is here. |
10214 | | */ |
10215 | 2.58k | SKIP(5); |
10216 | | |
10217 | 2.58k | if (!IS_BLANK_CH(RAW)) { |
10218 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
10219 | 0 | "Blank needed after '<?xml'\n"); |
10220 | 0 | } |
10221 | 2.58k | SKIP_BLANKS; |
10222 | | |
10223 | | /* |
10224 | | * We must have the VersionInfo here. |
10225 | | */ |
10226 | 2.58k | version = xmlParseVersionInfo(ctxt); |
10227 | 2.58k | if (version == NULL) { |
10228 | 1.99k | xmlFatalErr(ctxt, XML_ERR_VERSION_MISSING, NULL); |
10229 | 1.99k | } else { |
10230 | 589 | if (!xmlStrEqual(version, (const xmlChar *) XML_DEFAULT_VERSION)) { |
10231 | | /* |
10232 | | * Changed here for XML-1.0 5th edition |
10233 | | */ |
10234 | 272 | if (ctxt->options & XML_PARSE_OLD10) { |
10235 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_UNKNOWN_VERSION, |
10236 | 0 | "Unsupported version '%s'\n", |
10237 | 0 | version); |
10238 | 272 | } else { |
10239 | 272 | if ((version[0] == '1') && ((version[1] == '.'))) { |
10240 | 260 | xmlWarningMsg(ctxt, XML_WAR_UNKNOWN_VERSION, |
10241 | 260 | "Unsupported version '%s'\n", |
10242 | 260 | version, NULL); |
10243 | 260 | } else { |
10244 | 12 | xmlFatalErrMsgStr(ctxt, XML_ERR_UNKNOWN_VERSION, |
10245 | 12 | "Unsupported version '%s'\n", |
10246 | 12 | version); |
10247 | 12 | } |
10248 | 272 | } |
10249 | 272 | } |
10250 | 589 | if (ctxt->version != NULL) |
10251 | 0 | xmlFree((void *) ctxt->version); |
10252 | 589 | ctxt->version = version; |
10253 | 589 | } |
10254 | | |
10255 | | /* |
10256 | | * We may have the encoding declaration |
10257 | | */ |
10258 | 2.58k | if (!IS_BLANK_CH(RAW)) { |
10259 | 2.01k | if ((RAW == '?') && (NXT(1) == '>')) { |
10260 | 22 | SKIP(2); |
10261 | 22 | return; |
10262 | 22 | } |
10263 | 1.99k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Blank needed here\n"); |
10264 | 1.99k | } |
10265 | 2.56k | xmlParseEncodingDecl(ctxt); |
10266 | | |
10267 | | /* |
10268 | | * We may have the standalone status. |
10269 | | */ |
10270 | 2.56k | if ((ctxt->encoding != NULL) && (!IS_BLANK_CH(RAW))) { |
10271 | 564 | if ((RAW == '?') && (NXT(1) == '>')) { |
10272 | 42 | SKIP(2); |
10273 | 42 | return; |
10274 | 42 | } |
10275 | 522 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Blank needed here\n"); |
10276 | 522 | } |
10277 | | |
10278 | | /* |
10279 | | * We can grow the input buffer freely at that point |
10280 | | */ |
10281 | 2.51k | GROW; |
10282 | | |
10283 | 2.51k | SKIP_BLANKS; |
10284 | 2.51k | ctxt->standalone = xmlParseSDDecl(ctxt); |
10285 | | |
10286 | 2.51k | SKIP_BLANKS; |
10287 | 2.51k | if ((RAW == '?') && (NXT(1) == '>')) { |
10288 | 347 | SKIP(2); |
10289 | 2.17k | } else if (RAW == '>') { |
10290 | | /* Deprecated old WD ... */ |
10291 | 433 | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
10292 | 433 | NEXT; |
10293 | 1.73k | } else { |
10294 | 1.73k | int c; |
10295 | | |
10296 | 1.73k | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
10297 | 1.38M | while ((PARSER_STOPPED(ctxt) == 0) && |
10298 | 1.38M | ((c = CUR) != 0)) { |
10299 | 1.38M | NEXT; |
10300 | 1.38M | if (c == '>') |
10301 | 1.25k | break; |
10302 | 1.38M | } |
10303 | 1.73k | } |
10304 | 2.51k | } |
10305 | | |
10306 | | /** |
10307 | | * @since 2.14.0 |
10308 | | * |
10309 | | * @param ctxt parser context |
10310 | | * @returns the version from the XML declaration. |
10311 | | */ |
10312 | | const xmlChar * |
10313 | 0 | xmlCtxtGetVersion(xmlParserCtxt *ctxt) { |
10314 | 0 | if (ctxt == NULL) |
10315 | 0 | return(NULL); |
10316 | | |
10317 | 0 | return(ctxt->version); |
10318 | 0 | } |
10319 | | |
10320 | | /** |
10321 | | * @since 2.14.0 |
10322 | | * |
10323 | | * @param ctxt parser context |
10324 | | * @returns the value from the standalone document declaration. |
10325 | | */ |
10326 | | int |
10327 | 0 | xmlCtxtGetStandalone(xmlParserCtxt *ctxt) { |
10328 | 0 | if (ctxt == NULL) |
10329 | 0 | return(0); |
10330 | | |
10331 | 0 | return(ctxt->standalone); |
10332 | 0 | } |
10333 | | |
10334 | | /** |
10335 | | * parse an XML Misc* optional field. |
10336 | | * |
10337 | | * @deprecated Internal function, don't use. |
10338 | | * |
10339 | | * [27] Misc ::= Comment | PI | S |
10340 | | * @param ctxt an XML parser context |
10341 | | */ |
10342 | | |
10343 | | void |
10344 | 29.8k | xmlParseMisc(xmlParserCtxt *ctxt) { |
10345 | 32.8k | while (PARSER_STOPPED(ctxt) == 0) { |
10346 | 31.9k | SKIP_BLANKS; |
10347 | 31.9k | GROW; |
10348 | 31.9k | if ((RAW == '<') && (NXT(1) == '?')) { |
10349 | 2.23k | xmlParsePI(ctxt); |
10350 | 29.7k | } else if (CMP4(CUR_PTR, '<', '!', '-', '-')) { |
10351 | 820 | xmlParseComment(ctxt); |
10352 | 28.9k | } else { |
10353 | 28.9k | break; |
10354 | 28.9k | } |
10355 | 31.9k | } |
10356 | 29.8k | } |
10357 | | |
10358 | | static void |
10359 | 14.4k | xmlFinishDocument(xmlParserCtxtPtr ctxt) { |
10360 | 14.4k | xmlDocPtr doc; |
10361 | | |
10362 | | /* |
10363 | | * SAX: end of the document processing. |
10364 | | */ |
10365 | 14.4k | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
10366 | 14.4k | ctxt->sax->endDocument(ctxt->userData); |
10367 | | |
10368 | | /* |
10369 | | * Remove locally kept entity definitions if the tree was not built |
10370 | | */ |
10371 | 14.4k | doc = ctxt->myDoc; |
10372 | 14.4k | if ((doc != NULL) && |
10373 | 14.4k | (xmlStrEqual(doc->version, SAX_COMPAT_MODE))) { |
10374 | 1.16k | xmlFreeDoc(doc); |
10375 | 1.16k | ctxt->myDoc = NULL; |
10376 | 1.16k | } |
10377 | 14.4k | } |
10378 | | |
10379 | | /** |
10380 | | * Parse an XML document and invoke the SAX handlers. This is useful |
10381 | | * if you're only interested in custom SAX callbacks. If you want a |
10382 | | * document tree, use #xmlCtxtParseDocument. |
10383 | | * |
10384 | | * @param ctxt an XML parser context |
10385 | | * @returns 0, -1 in case of error. |
10386 | | */ |
10387 | | |
10388 | | int |
10389 | 14.5k | xmlParseDocument(xmlParserCtxt *ctxt) { |
10390 | 14.5k | if ((ctxt == NULL) || (ctxt->input == NULL)) |
10391 | 0 | return(-1); |
10392 | | |
10393 | 14.5k | GROW; |
10394 | | |
10395 | | /* |
10396 | | * SAX: detecting the level. |
10397 | | */ |
10398 | 14.5k | xmlCtxtInitializeLate(ctxt); |
10399 | | |
10400 | 14.5k | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) { |
10401 | 14.5k | ctxt->sax->setDocumentLocator(ctxt->userData, |
10402 | 14.5k | (xmlSAXLocator *) &xmlDefaultSAXLocator); |
10403 | 14.5k | } |
10404 | | |
10405 | 14.5k | xmlDetectEncoding(ctxt); |
10406 | | |
10407 | 14.5k | if (CUR == 0) { |
10408 | 36 | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); |
10409 | 36 | return(-1); |
10410 | 36 | } |
10411 | | |
10412 | 14.4k | GROW; |
10413 | 14.4k | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
10414 | | |
10415 | | /* |
10416 | | * Note that we will switch encoding on the fly. |
10417 | | */ |
10418 | 2.58k | xmlParseXMLDecl(ctxt); |
10419 | 2.58k | SKIP_BLANKS; |
10420 | 11.8k | } else { |
10421 | 11.8k | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10422 | 11.8k | if (ctxt->version == NULL) { |
10423 | 0 | xmlErrMemory(ctxt); |
10424 | 0 | return(-1); |
10425 | 0 | } |
10426 | 11.8k | } |
10427 | 14.4k | if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) |
10428 | 12.1k | ctxt->sax->startDocument(ctxt->userData); |
10429 | 14.4k | if ((ctxt->myDoc != NULL) && (ctxt->input != NULL) && |
10430 | 14.4k | (ctxt->input->buf != NULL) && (ctxt->input->buf->compressed >= 0)) { |
10431 | 0 | ctxt->myDoc->compression = ctxt->input->buf->compressed; |
10432 | 0 | } |
10433 | | |
10434 | | /* |
10435 | | * The Misc part of the Prolog |
10436 | | */ |
10437 | 14.4k | xmlParseMisc(ctxt); |
10438 | | |
10439 | | /* |
10440 | | * Then possibly doc type declaration(s) and more Misc |
10441 | | * (doctypedecl Misc*)? |
10442 | | */ |
10443 | 14.4k | GROW; |
10444 | 14.4k | if (CMP9(CUR_PTR, '<', '!', 'D', 'O', 'C', 'T', 'Y', 'P', 'E')) { |
10445 | | |
10446 | 5.92k | ctxt->inSubset = 1; |
10447 | 5.92k | xmlParseDocTypeDecl(ctxt); |
10448 | 5.92k | if (RAW == '[') { |
10449 | 5.62k | xmlParseInternalSubset(ctxt); |
10450 | 5.62k | } else if (RAW == '>') { |
10451 | 124 | NEXT; |
10452 | 124 | } |
10453 | | |
10454 | | /* |
10455 | | * Create and update the external subset. |
10456 | | */ |
10457 | 5.92k | ctxt->inSubset = 2; |
10458 | 5.92k | if ((ctxt->sax != NULL) && (ctxt->sax->externalSubset != NULL) && |
10459 | 5.92k | (!ctxt->disableSAX)) |
10460 | 849 | ctxt->sax->externalSubset(ctxt->userData, ctxt->intSubName, |
10461 | 849 | ctxt->extSubSystem, ctxt->extSubURI); |
10462 | 5.92k | ctxt->inSubset = 0; |
10463 | | |
10464 | 5.92k | xmlCleanSpecialAttr(ctxt); |
10465 | | |
10466 | 5.92k | xmlParseMisc(ctxt); |
10467 | 5.92k | } |
10468 | | |
10469 | | /* |
10470 | | * Time to start parsing the tree itself |
10471 | | */ |
10472 | 14.4k | GROW; |
10473 | 14.4k | if (RAW != '<') { |
10474 | 5.03k | if (ctxt->wellFormed) |
10475 | 280 | xmlFatalErrMsg(ctxt, XML_ERR_DOCUMENT_EMPTY, |
10476 | 280 | "Start tag expected, '<' not found\n"); |
10477 | 9.43k | } else { |
10478 | 9.43k | xmlParseElement(ctxt); |
10479 | | |
10480 | | /* |
10481 | | * The Misc part at the end |
10482 | | */ |
10483 | 9.43k | xmlParseMisc(ctxt); |
10484 | | |
10485 | 9.43k | xmlParserCheckEOF(ctxt, XML_ERR_DOCUMENT_END); |
10486 | 9.43k | } |
10487 | | |
10488 | 14.4k | ctxt->instate = XML_PARSER_EOF; |
10489 | 14.4k | xmlFinishDocument(ctxt); |
10490 | | |
10491 | 14.4k | if (! ctxt->wellFormed) { |
10492 | 14.4k | ctxt->valid = 0; |
10493 | 14.4k | return(-1); |
10494 | 14.4k | } |
10495 | | |
10496 | 30 | return(0); |
10497 | 14.4k | } |
10498 | | |
10499 | | /** |
10500 | | * parse a general parsed entity |
10501 | | * An external general parsed entity is well-formed if it matches the |
10502 | | * production labeled extParsedEnt. |
10503 | | * |
10504 | | * @deprecated Internal function, don't use. |
10505 | | * |
10506 | | * [78] extParsedEnt ::= TextDecl? content |
10507 | | * |
10508 | | * @param ctxt an XML parser context |
10509 | | * @returns 0, -1 in case of error. the parser context is augmented |
10510 | | * as a result of the parsing. |
10511 | | */ |
10512 | | |
10513 | | int |
10514 | 0 | xmlParseExtParsedEnt(xmlParserCtxt *ctxt) { |
10515 | 0 | if ((ctxt == NULL) || (ctxt->input == NULL)) |
10516 | 0 | return(-1); |
10517 | | |
10518 | 0 | xmlCtxtInitializeLate(ctxt); |
10519 | |
|
10520 | 0 | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) { |
10521 | 0 | ctxt->sax->setDocumentLocator(ctxt->userData, |
10522 | 0 | (xmlSAXLocator *) &xmlDefaultSAXLocator); |
10523 | 0 | } |
10524 | |
|
10525 | 0 | xmlDetectEncoding(ctxt); |
10526 | |
|
10527 | 0 | if (CUR == 0) { |
10528 | 0 | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); |
10529 | 0 | } |
10530 | | |
10531 | | /* |
10532 | | * Check for the XMLDecl in the Prolog. |
10533 | | */ |
10534 | 0 | GROW; |
10535 | 0 | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
10536 | | |
10537 | | /* |
10538 | | * Note that we will switch encoding on the fly. |
10539 | | */ |
10540 | 0 | xmlParseXMLDecl(ctxt); |
10541 | 0 | SKIP_BLANKS; |
10542 | 0 | } else { |
10543 | 0 | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10544 | 0 | } |
10545 | 0 | if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) |
10546 | 0 | ctxt->sax->startDocument(ctxt->userData); |
10547 | | |
10548 | | /* |
10549 | | * Doing validity checking on chunk doesn't make sense |
10550 | | */ |
10551 | 0 | ctxt->options &= ~XML_PARSE_DTDVALID; |
10552 | 0 | ctxt->validate = 0; |
10553 | 0 | ctxt->depth = 0; |
10554 | |
|
10555 | 0 | xmlParseContentInternal(ctxt); |
10556 | |
|
10557 | 0 | if (ctxt->input->cur < ctxt->input->end) |
10558 | 0 | xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); |
10559 | | |
10560 | | /* |
10561 | | * SAX: end of the document processing. |
10562 | | */ |
10563 | 0 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
10564 | 0 | ctxt->sax->endDocument(ctxt->userData); |
10565 | |
|
10566 | 0 | if (! ctxt->wellFormed) return(-1); |
10567 | 0 | return(0); |
10568 | 0 | } |
10569 | | |
10570 | | #ifdef LIBXML_PUSH_ENABLED |
10571 | | /************************************************************************ |
10572 | | * * |
10573 | | * Progressive parsing interfaces * |
10574 | | * * |
10575 | | ************************************************************************/ |
10576 | | |
10577 | | /** |
10578 | | * Check whether the input buffer contains a character. |
10579 | | * |
10580 | | * @param ctxt an XML parser context |
10581 | | * @param c character |
10582 | | */ |
10583 | | static int |
10584 | 0 | xmlParseLookupChar(xmlParserCtxtPtr ctxt, int c) { |
10585 | 0 | const xmlChar *cur; |
10586 | |
|
10587 | 0 | if (ctxt->checkIndex == 0) { |
10588 | 0 | cur = ctxt->input->cur + 1; |
10589 | 0 | } else { |
10590 | 0 | cur = ctxt->input->cur + ctxt->checkIndex; |
10591 | 0 | } |
10592 | |
|
10593 | 0 | if (memchr(cur, c, ctxt->input->end - cur) == NULL) { |
10594 | 0 | size_t index = ctxt->input->end - ctxt->input->cur; |
10595 | |
|
10596 | 0 | if (index > LONG_MAX) { |
10597 | 0 | ctxt->checkIndex = 0; |
10598 | 0 | return(1); |
10599 | 0 | } |
10600 | 0 | ctxt->checkIndex = index; |
10601 | 0 | return(0); |
10602 | 0 | } else { |
10603 | 0 | ctxt->checkIndex = 0; |
10604 | 0 | return(1); |
10605 | 0 | } |
10606 | 0 | } |
10607 | | |
10608 | | /** |
10609 | | * Check whether the input buffer contains a string. |
10610 | | * |
10611 | | * @param ctxt an XML parser context |
10612 | | * @param startDelta delta to apply at the start |
10613 | | * @param str string |
10614 | | * @param strLen length of string |
10615 | | */ |
10616 | | static const xmlChar * |
10617 | | xmlParseLookupString(xmlParserCtxtPtr ctxt, size_t startDelta, |
10618 | 0 | const char *str, size_t strLen) { |
10619 | 0 | const xmlChar *cur, *term; |
10620 | |
|
10621 | 0 | if (ctxt->checkIndex == 0) { |
10622 | 0 | cur = ctxt->input->cur + startDelta; |
10623 | 0 | } else { |
10624 | 0 | cur = ctxt->input->cur + ctxt->checkIndex; |
10625 | 0 | } |
10626 | |
|
10627 | 0 | term = BAD_CAST strstr((const char *) cur, str); |
10628 | 0 | if (term == NULL) { |
10629 | 0 | const xmlChar *end = ctxt->input->end; |
10630 | 0 | size_t index; |
10631 | | |
10632 | | /* Rescan (strLen - 1) characters. */ |
10633 | 0 | if ((size_t) (end - cur) < strLen) |
10634 | 0 | end = cur; |
10635 | 0 | else |
10636 | 0 | end -= strLen - 1; |
10637 | 0 | index = end - ctxt->input->cur; |
10638 | 0 | if (index > LONG_MAX) { |
10639 | 0 | ctxt->checkIndex = 0; |
10640 | 0 | return(ctxt->input->end - strLen); |
10641 | 0 | } |
10642 | 0 | ctxt->checkIndex = index; |
10643 | 0 | } else { |
10644 | 0 | ctxt->checkIndex = 0; |
10645 | 0 | } |
10646 | | |
10647 | 0 | return(term); |
10648 | 0 | } |
10649 | | |
10650 | | /** |
10651 | | * Check whether the input buffer contains terminated char data. |
10652 | | * |
10653 | | * @param ctxt an XML parser context |
10654 | | */ |
10655 | | static int |
10656 | 0 | xmlParseLookupCharData(xmlParserCtxtPtr ctxt) { |
10657 | 0 | const xmlChar *cur = ctxt->input->cur + ctxt->checkIndex; |
10658 | 0 | const xmlChar *end = ctxt->input->end; |
10659 | 0 | size_t index; |
10660 | |
|
10661 | 0 | while (cur < end) { |
10662 | 0 | if ((*cur == '<') || (*cur == '&')) { |
10663 | 0 | ctxt->checkIndex = 0; |
10664 | 0 | return(1); |
10665 | 0 | } |
10666 | 0 | cur++; |
10667 | 0 | } |
10668 | | |
10669 | 0 | index = cur - ctxt->input->cur; |
10670 | 0 | if (index > LONG_MAX) { |
10671 | 0 | ctxt->checkIndex = 0; |
10672 | 0 | return(1); |
10673 | 0 | } |
10674 | 0 | ctxt->checkIndex = index; |
10675 | 0 | return(0); |
10676 | 0 | } |
10677 | | |
10678 | | /** |
10679 | | * Check whether there's enough data in the input buffer to finish parsing |
10680 | | * a start tag. This has to take quotes into account. |
10681 | | * |
10682 | | * @param ctxt an XML parser context |
10683 | | */ |
10684 | | static int |
10685 | 0 | xmlParseLookupGt(xmlParserCtxtPtr ctxt) { |
10686 | 0 | const xmlChar *cur; |
10687 | 0 | const xmlChar *end = ctxt->input->end; |
10688 | 0 | int state = ctxt->endCheckState; |
10689 | 0 | size_t index; |
10690 | |
|
10691 | 0 | if (ctxt->checkIndex == 0) |
10692 | 0 | cur = ctxt->input->cur + 1; |
10693 | 0 | else |
10694 | 0 | cur = ctxt->input->cur + ctxt->checkIndex; |
10695 | |
|
10696 | 0 | while (cur < end) { |
10697 | 0 | if (state) { |
10698 | 0 | if (*cur == state) |
10699 | 0 | state = 0; |
10700 | 0 | } else if (*cur == '\'' || *cur == '"') { |
10701 | 0 | state = *cur; |
10702 | 0 | } else if (*cur == '>') { |
10703 | 0 | ctxt->checkIndex = 0; |
10704 | 0 | ctxt->endCheckState = 0; |
10705 | 0 | return(1); |
10706 | 0 | } |
10707 | 0 | cur++; |
10708 | 0 | } |
10709 | | |
10710 | 0 | index = cur - ctxt->input->cur; |
10711 | 0 | if (index > LONG_MAX) { |
10712 | 0 | ctxt->checkIndex = 0; |
10713 | 0 | ctxt->endCheckState = 0; |
10714 | 0 | return(1); |
10715 | 0 | } |
10716 | 0 | ctxt->checkIndex = index; |
10717 | 0 | ctxt->endCheckState = state; |
10718 | 0 | return(0); |
10719 | 0 | } |
10720 | | |
10721 | | /** |
10722 | | * Check whether there's enough data in the input buffer to finish parsing |
10723 | | * the internal subset. |
10724 | | * |
10725 | | * @param ctxt an XML parser context |
10726 | | */ |
10727 | | static int |
10728 | 0 | xmlParseLookupInternalSubset(xmlParserCtxtPtr ctxt) { |
10729 | | /* |
10730 | | * Sorry, but progressive parsing of the internal subset is not |
10731 | | * supported. We first check that the full content of the internal |
10732 | | * subset is available and parsing is launched only at that point. |
10733 | | * Internal subset ends with "']' S? '>'" in an unescaped section and |
10734 | | * not in a ']]>' sequence which are conditional sections. |
10735 | | */ |
10736 | 0 | const xmlChar *cur, *start; |
10737 | 0 | const xmlChar *end = ctxt->input->end; |
10738 | 0 | int state = ctxt->endCheckState; |
10739 | 0 | size_t index; |
10740 | |
|
10741 | 0 | if (ctxt->checkIndex == 0) { |
10742 | 0 | cur = ctxt->input->cur + 1; |
10743 | 0 | } else { |
10744 | 0 | cur = ctxt->input->cur + ctxt->checkIndex; |
10745 | 0 | } |
10746 | 0 | start = cur; |
10747 | |
|
10748 | 0 | while (cur < end) { |
10749 | 0 | if (state == '-') { |
10750 | 0 | if ((*cur == '-') && |
10751 | 0 | (cur[1] == '-') && |
10752 | 0 | (cur[2] == '>')) { |
10753 | 0 | state = 0; |
10754 | 0 | cur += 3; |
10755 | 0 | start = cur; |
10756 | 0 | continue; |
10757 | 0 | } |
10758 | 0 | } |
10759 | 0 | else if (state == ']') { |
10760 | 0 | if (*cur == '>') { |
10761 | 0 | ctxt->checkIndex = 0; |
10762 | 0 | ctxt->endCheckState = 0; |
10763 | 0 | return(1); |
10764 | 0 | } |
10765 | 0 | if (IS_BLANK_CH(*cur)) { |
10766 | 0 | state = ' '; |
10767 | 0 | } else if (*cur != ']') { |
10768 | 0 | state = 0; |
10769 | 0 | start = cur; |
10770 | 0 | continue; |
10771 | 0 | } |
10772 | 0 | } |
10773 | 0 | else if (state == ' ') { |
10774 | 0 | if (*cur == '>') { |
10775 | 0 | ctxt->checkIndex = 0; |
10776 | 0 | ctxt->endCheckState = 0; |
10777 | 0 | return(1); |
10778 | 0 | } |
10779 | 0 | if (!IS_BLANK_CH(*cur)) { |
10780 | 0 | state = 0; |
10781 | 0 | start = cur; |
10782 | 0 | continue; |
10783 | 0 | } |
10784 | 0 | } |
10785 | 0 | else if (state != 0) { |
10786 | 0 | if (*cur == state) { |
10787 | 0 | state = 0; |
10788 | 0 | start = cur + 1; |
10789 | 0 | } |
10790 | 0 | } |
10791 | 0 | else if (*cur == '<') { |
10792 | 0 | if ((cur[1] == '!') && |
10793 | 0 | (cur[2] == '-') && |
10794 | 0 | (cur[3] == '-')) { |
10795 | 0 | state = '-'; |
10796 | 0 | cur += 4; |
10797 | | /* Don't treat <!--> as comment */ |
10798 | 0 | start = cur; |
10799 | 0 | continue; |
10800 | 0 | } |
10801 | 0 | } |
10802 | 0 | else if ((*cur == '"') || (*cur == '\'') || (*cur == ']')) { |
10803 | 0 | state = *cur; |
10804 | 0 | } |
10805 | | |
10806 | 0 | cur++; |
10807 | 0 | } |
10808 | | |
10809 | | /* |
10810 | | * Rescan the three last characters to detect "<!--" and "-->" |
10811 | | * split across chunks. |
10812 | | */ |
10813 | 0 | if ((state == 0) || (state == '-')) { |
10814 | 0 | if (cur - start < 3) |
10815 | 0 | cur = start; |
10816 | 0 | else |
10817 | 0 | cur -= 3; |
10818 | 0 | } |
10819 | 0 | index = cur - ctxt->input->cur; |
10820 | 0 | if (index > LONG_MAX) { |
10821 | 0 | ctxt->checkIndex = 0; |
10822 | 0 | ctxt->endCheckState = 0; |
10823 | 0 | return(1); |
10824 | 0 | } |
10825 | 0 | ctxt->checkIndex = index; |
10826 | 0 | ctxt->endCheckState = state; |
10827 | 0 | return(0); |
10828 | 0 | } |
10829 | | |
10830 | | /** |
10831 | | * Try to progress on parsing |
10832 | | * |
10833 | | * @param ctxt an XML parser context |
10834 | | * @param terminate last chunk indicator |
10835 | | * @returns zero if no parsing was possible |
10836 | | */ |
10837 | | static int |
10838 | 0 | xmlParseTryOrFinish(xmlParserCtxtPtr ctxt, int terminate) { |
10839 | 0 | int ret = 0; |
10840 | 0 | size_t avail; |
10841 | 0 | xmlChar cur, next; |
10842 | |
|
10843 | 0 | if (ctxt->input == NULL) |
10844 | 0 | return(0); |
10845 | | |
10846 | 0 | if ((ctxt->input != NULL) && |
10847 | 0 | (ctxt->input->cur - ctxt->input->base > 4096)) { |
10848 | 0 | xmlParserShrink(ctxt); |
10849 | 0 | } |
10850 | |
|
10851 | 0 | while (ctxt->disableSAX == 0) { |
10852 | 0 | avail = ctxt->input->end - ctxt->input->cur; |
10853 | 0 | if (avail < 1) |
10854 | 0 | goto done; |
10855 | 0 | switch (ctxt->instate) { |
10856 | 0 | case XML_PARSER_EOF: |
10857 | | /* |
10858 | | * Document parsing is done ! |
10859 | | */ |
10860 | 0 | goto done; |
10861 | 0 | case XML_PARSER_START: |
10862 | | /* |
10863 | | * Very first chars read from the document flow. |
10864 | | */ |
10865 | 0 | if ((!terminate) && (avail < 4)) |
10866 | 0 | goto done; |
10867 | | |
10868 | | /* |
10869 | | * We need more bytes to detect EBCDIC code pages. |
10870 | | * See xmlDetectEBCDIC. |
10871 | | */ |
10872 | 0 | if ((CMP4(CUR_PTR, 0x4C, 0x6F, 0xA7, 0x94)) && |
10873 | 0 | (!terminate) && (avail < 200)) |
10874 | 0 | goto done; |
10875 | | |
10876 | 0 | xmlDetectEncoding(ctxt); |
10877 | 0 | ctxt->instate = XML_PARSER_XML_DECL; |
10878 | 0 | break; |
10879 | | |
10880 | 0 | case XML_PARSER_XML_DECL: |
10881 | 0 | if ((!terminate) && (avail < 2)) |
10882 | 0 | goto done; |
10883 | 0 | cur = ctxt->input->cur[0]; |
10884 | 0 | next = ctxt->input->cur[1]; |
10885 | 0 | if ((cur == '<') && (next == '?')) { |
10886 | | /* PI or XML decl */ |
10887 | 0 | if ((!terminate) && |
10888 | 0 | (!xmlParseLookupString(ctxt, 2, "?>", 2))) |
10889 | 0 | goto done; |
10890 | 0 | if ((ctxt->input->cur[2] == 'x') && |
10891 | 0 | (ctxt->input->cur[3] == 'm') && |
10892 | 0 | (ctxt->input->cur[4] == 'l') && |
10893 | 0 | (IS_BLANK_CH(ctxt->input->cur[5]))) { |
10894 | 0 | ret += 5; |
10895 | 0 | xmlParseXMLDecl(ctxt); |
10896 | 0 | } else { |
10897 | 0 | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10898 | 0 | if (ctxt->version == NULL) { |
10899 | 0 | xmlErrMemory(ctxt); |
10900 | 0 | break; |
10901 | 0 | } |
10902 | 0 | } |
10903 | 0 | } else { |
10904 | 0 | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10905 | 0 | if (ctxt->version == NULL) { |
10906 | 0 | xmlErrMemory(ctxt); |
10907 | 0 | break; |
10908 | 0 | } |
10909 | 0 | } |
10910 | 0 | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) { |
10911 | 0 | ctxt->sax->setDocumentLocator(ctxt->userData, |
10912 | 0 | (xmlSAXLocator *) &xmlDefaultSAXLocator); |
10913 | 0 | } |
10914 | 0 | if ((ctxt->sax) && (ctxt->sax->startDocument) && |
10915 | 0 | (!ctxt->disableSAX)) |
10916 | 0 | ctxt->sax->startDocument(ctxt->userData); |
10917 | 0 | ctxt->instate = XML_PARSER_MISC; |
10918 | 0 | break; |
10919 | 0 | case XML_PARSER_START_TAG: { |
10920 | 0 | const xmlChar *name; |
10921 | 0 | const xmlChar *prefix = NULL; |
10922 | 0 | const xmlChar *URI = NULL; |
10923 | 0 | int line = ctxt->input->line; |
10924 | 0 | int nbNs = 0; |
10925 | |
|
10926 | 0 | if ((!terminate) && (avail < 2)) |
10927 | 0 | goto done; |
10928 | 0 | cur = ctxt->input->cur[0]; |
10929 | 0 | if (cur != '<') { |
10930 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_DOCUMENT_EMPTY, |
10931 | 0 | "Start tag expected, '<' not found"); |
10932 | 0 | ctxt->instate = XML_PARSER_EOF; |
10933 | 0 | xmlFinishDocument(ctxt); |
10934 | 0 | goto done; |
10935 | 0 | } |
10936 | 0 | if ((!terminate) && (!xmlParseLookupGt(ctxt))) |
10937 | 0 | goto done; |
10938 | 0 | if (ctxt->spaceNr == 0) |
10939 | 0 | spacePush(ctxt, -1); |
10940 | 0 | else if (*ctxt->space == -2) |
10941 | 0 | spacePush(ctxt, -1); |
10942 | 0 | else |
10943 | 0 | spacePush(ctxt, *ctxt->space); |
10944 | 0 | #ifdef LIBXML_SAX1_ENABLED |
10945 | 0 | if (ctxt->sax2) |
10946 | 0 | #endif /* LIBXML_SAX1_ENABLED */ |
10947 | 0 | name = xmlParseStartTag2(ctxt, &prefix, &URI, &nbNs); |
10948 | 0 | #ifdef LIBXML_SAX1_ENABLED |
10949 | 0 | else |
10950 | 0 | name = xmlParseStartTag(ctxt); |
10951 | 0 | #endif /* LIBXML_SAX1_ENABLED */ |
10952 | 0 | if (name == NULL) { |
10953 | 0 | spacePop(ctxt); |
10954 | 0 | ctxt->instate = XML_PARSER_EOF; |
10955 | 0 | xmlFinishDocument(ctxt); |
10956 | 0 | goto done; |
10957 | 0 | } |
10958 | 0 | #ifdef LIBXML_VALID_ENABLED |
10959 | | /* |
10960 | | * [ VC: Root Element Type ] |
10961 | | * The Name in the document type declaration must match |
10962 | | * the element type of the root element. |
10963 | | */ |
10964 | 0 | if (ctxt->validate && ctxt->wellFormed && ctxt->myDoc && |
10965 | 0 | ctxt->node && (ctxt->node == ctxt->myDoc->children)) |
10966 | 0 | ctxt->valid &= xmlValidateRoot(&ctxt->vctxt, ctxt->myDoc); |
10967 | 0 | #endif /* LIBXML_VALID_ENABLED */ |
10968 | | |
10969 | | /* |
10970 | | * Check for an Empty Element. |
10971 | | */ |
10972 | 0 | if ((RAW == '/') && (NXT(1) == '>')) { |
10973 | 0 | SKIP(2); |
10974 | |
|
10975 | 0 | if (ctxt->sax2) { |
10976 | 0 | if ((ctxt->sax != NULL) && |
10977 | 0 | (ctxt->sax->endElementNs != NULL) && |
10978 | 0 | (!ctxt->disableSAX)) |
10979 | 0 | ctxt->sax->endElementNs(ctxt->userData, name, |
10980 | 0 | prefix, URI); |
10981 | 0 | if (nbNs > 0) |
10982 | 0 | xmlParserNsPop(ctxt, nbNs); |
10983 | 0 | #ifdef LIBXML_SAX1_ENABLED |
10984 | 0 | } else { |
10985 | 0 | if ((ctxt->sax != NULL) && |
10986 | 0 | (ctxt->sax->endElement != NULL) && |
10987 | 0 | (!ctxt->disableSAX)) |
10988 | 0 | ctxt->sax->endElement(ctxt->userData, name); |
10989 | 0 | #endif /* LIBXML_SAX1_ENABLED */ |
10990 | 0 | } |
10991 | 0 | spacePop(ctxt); |
10992 | 0 | } else if (RAW == '>') { |
10993 | 0 | NEXT; |
10994 | 0 | nameNsPush(ctxt, name, prefix, URI, line, nbNs); |
10995 | 0 | } else { |
10996 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_GT_REQUIRED, |
10997 | 0 | "Couldn't find end of Start Tag %s\n", |
10998 | 0 | name); |
10999 | 0 | nodePop(ctxt); |
11000 | 0 | spacePop(ctxt); |
11001 | 0 | if (nbNs > 0) |
11002 | 0 | xmlParserNsPop(ctxt, nbNs); |
11003 | 0 | } |
11004 | |
|
11005 | 0 | if (ctxt->nameNr == 0) |
11006 | 0 | ctxt->instate = XML_PARSER_EPILOG; |
11007 | 0 | else |
11008 | 0 | ctxt->instate = XML_PARSER_CONTENT; |
11009 | 0 | break; |
11010 | 0 | } |
11011 | 0 | case XML_PARSER_CONTENT: { |
11012 | 0 | cur = ctxt->input->cur[0]; |
11013 | |
|
11014 | 0 | if (cur == '<') { |
11015 | 0 | if ((!terminate) && (avail < 2)) |
11016 | 0 | goto done; |
11017 | 0 | next = ctxt->input->cur[1]; |
11018 | |
|
11019 | 0 | if (next == '/') { |
11020 | 0 | ctxt->instate = XML_PARSER_END_TAG; |
11021 | 0 | break; |
11022 | 0 | } else if (next == '?') { |
11023 | 0 | if ((!terminate) && |
11024 | 0 | (!xmlParseLookupString(ctxt, 2, "?>", 2))) |
11025 | 0 | goto done; |
11026 | 0 | xmlParsePI(ctxt); |
11027 | 0 | ctxt->instate = XML_PARSER_CONTENT; |
11028 | 0 | break; |
11029 | 0 | } else if (next == '!') { |
11030 | 0 | if ((!terminate) && (avail < 3)) |
11031 | 0 | goto done; |
11032 | 0 | next = ctxt->input->cur[2]; |
11033 | |
|
11034 | 0 | if (next == '-') { |
11035 | 0 | if ((!terminate) && (avail < 4)) |
11036 | 0 | goto done; |
11037 | 0 | if (ctxt->input->cur[3] == '-') { |
11038 | 0 | if ((!terminate) && |
11039 | 0 | (!xmlParseLookupString(ctxt, 4, "-->", 3))) |
11040 | 0 | goto done; |
11041 | 0 | xmlParseComment(ctxt); |
11042 | 0 | ctxt->instate = XML_PARSER_CONTENT; |
11043 | 0 | break; |
11044 | 0 | } |
11045 | 0 | } else if (next == '[') { |
11046 | 0 | if ((!terminate) && (avail < 9)) |
11047 | 0 | goto done; |
11048 | 0 | if ((ctxt->input->cur[2] == '[') && |
11049 | 0 | (ctxt->input->cur[3] == 'C') && |
11050 | 0 | (ctxt->input->cur[4] == 'D') && |
11051 | 0 | (ctxt->input->cur[5] == 'A') && |
11052 | 0 | (ctxt->input->cur[6] == 'T') && |
11053 | 0 | (ctxt->input->cur[7] == 'A') && |
11054 | 0 | (ctxt->input->cur[8] == '[')) { |
11055 | 0 | if ((!terminate) && |
11056 | 0 | (!xmlParseLookupString(ctxt, 9, "]]>", 3))) |
11057 | 0 | goto done; |
11058 | 0 | ctxt->instate = XML_PARSER_CDATA_SECTION; |
11059 | 0 | xmlParseCDSect(ctxt); |
11060 | 0 | ctxt->instate = XML_PARSER_CONTENT; |
11061 | 0 | break; |
11062 | 0 | } |
11063 | 0 | } |
11064 | 0 | } |
11065 | 0 | } else if (cur == '&') { |
11066 | 0 | if ((!terminate) && (!xmlParseLookupChar(ctxt, ';'))) |
11067 | 0 | goto done; |
11068 | 0 | xmlParseReference(ctxt); |
11069 | 0 | break; |
11070 | 0 | } else { |
11071 | | /* TODO Avoid the extra copy, handle directly !!! */ |
11072 | | /* |
11073 | | * Goal of the following test is: |
11074 | | * - minimize calls to the SAX 'character' callback |
11075 | | * when they are mergeable |
11076 | | * - handle an problem for isBlank when we only parse |
11077 | | * a sequence of blank chars and the next one is |
11078 | | * not available to check against '<' presence. |
11079 | | * - tries to homogenize the differences in SAX |
11080 | | * callbacks between the push and pull versions |
11081 | | * of the parser. |
11082 | | */ |
11083 | 0 | if (avail < XML_PARSER_BIG_BUFFER_SIZE) { |
11084 | 0 | if ((!terminate) && (!xmlParseLookupCharData(ctxt))) |
11085 | 0 | goto done; |
11086 | 0 | } |
11087 | 0 | ctxt->checkIndex = 0; |
11088 | 0 | xmlParseCharDataInternal(ctxt, !terminate); |
11089 | 0 | break; |
11090 | 0 | } |
11091 | | |
11092 | 0 | ctxt->instate = XML_PARSER_START_TAG; |
11093 | 0 | break; |
11094 | 0 | } |
11095 | 0 | case XML_PARSER_END_TAG: |
11096 | 0 | if ((!terminate) && (!xmlParseLookupChar(ctxt, '>'))) |
11097 | 0 | goto done; |
11098 | 0 | if (ctxt->sax2) { |
11099 | 0 | xmlParseEndTag2(ctxt, &ctxt->pushTab[ctxt->nameNr - 1]); |
11100 | 0 | nameNsPop(ctxt); |
11101 | 0 | } |
11102 | 0 | #ifdef LIBXML_SAX1_ENABLED |
11103 | 0 | else |
11104 | 0 | xmlParseEndTag1(ctxt, 0); |
11105 | 0 | #endif /* LIBXML_SAX1_ENABLED */ |
11106 | 0 | if (ctxt->nameNr == 0) { |
11107 | 0 | ctxt->instate = XML_PARSER_EPILOG; |
11108 | 0 | } else { |
11109 | 0 | ctxt->instate = XML_PARSER_CONTENT; |
11110 | 0 | } |
11111 | 0 | break; |
11112 | 0 | case XML_PARSER_MISC: |
11113 | 0 | case XML_PARSER_PROLOG: |
11114 | 0 | case XML_PARSER_EPILOG: |
11115 | 0 | SKIP_BLANKS; |
11116 | 0 | avail = ctxt->input->end - ctxt->input->cur; |
11117 | 0 | if (avail < 1) |
11118 | 0 | goto done; |
11119 | 0 | if (ctxt->input->cur[0] == '<') { |
11120 | 0 | if ((!terminate) && (avail < 2)) |
11121 | 0 | goto done; |
11122 | 0 | next = ctxt->input->cur[1]; |
11123 | 0 | if (next == '?') { |
11124 | 0 | if ((!terminate) && |
11125 | 0 | (!xmlParseLookupString(ctxt, 2, "?>", 2))) |
11126 | 0 | goto done; |
11127 | 0 | xmlParsePI(ctxt); |
11128 | 0 | break; |
11129 | 0 | } else if (next == '!') { |
11130 | 0 | if ((!terminate) && (avail < 3)) |
11131 | 0 | goto done; |
11132 | | |
11133 | 0 | if (ctxt->input->cur[2] == '-') { |
11134 | 0 | if ((!terminate) && (avail < 4)) |
11135 | 0 | goto done; |
11136 | 0 | if (ctxt->input->cur[3] == '-') { |
11137 | 0 | if ((!terminate) && |
11138 | 0 | (!xmlParseLookupString(ctxt, 4, "-->", 3))) |
11139 | 0 | goto done; |
11140 | 0 | xmlParseComment(ctxt); |
11141 | 0 | break; |
11142 | 0 | } |
11143 | 0 | } else if (ctxt->instate == XML_PARSER_MISC) { |
11144 | 0 | if ((!terminate) && (avail < 9)) |
11145 | 0 | goto done; |
11146 | 0 | if ((ctxt->input->cur[2] == 'D') && |
11147 | 0 | (ctxt->input->cur[3] == 'O') && |
11148 | 0 | (ctxt->input->cur[4] == 'C') && |
11149 | 0 | (ctxt->input->cur[5] == 'T') && |
11150 | 0 | (ctxt->input->cur[6] == 'Y') && |
11151 | 0 | (ctxt->input->cur[7] == 'P') && |
11152 | 0 | (ctxt->input->cur[8] == 'E')) { |
11153 | 0 | if ((!terminate) && (!xmlParseLookupGt(ctxt))) |
11154 | 0 | goto done; |
11155 | 0 | ctxt->inSubset = 1; |
11156 | 0 | xmlParseDocTypeDecl(ctxt); |
11157 | 0 | if (RAW == '[') { |
11158 | 0 | ctxt->instate = XML_PARSER_DTD; |
11159 | 0 | } else { |
11160 | 0 | if (RAW == '>') |
11161 | 0 | NEXT; |
11162 | | /* |
11163 | | * Create and update the external subset. |
11164 | | */ |
11165 | 0 | ctxt->inSubset = 2; |
11166 | 0 | if ((ctxt->sax != NULL) && |
11167 | 0 | (!ctxt->disableSAX) && |
11168 | 0 | (ctxt->sax->externalSubset != NULL)) |
11169 | 0 | ctxt->sax->externalSubset( |
11170 | 0 | ctxt->userData, |
11171 | 0 | ctxt->intSubName, |
11172 | 0 | ctxt->extSubSystem, |
11173 | 0 | ctxt->extSubURI); |
11174 | 0 | ctxt->inSubset = 0; |
11175 | 0 | xmlCleanSpecialAttr(ctxt); |
11176 | 0 | ctxt->instate = XML_PARSER_PROLOG; |
11177 | 0 | } |
11178 | 0 | break; |
11179 | 0 | } |
11180 | 0 | } |
11181 | 0 | } |
11182 | 0 | } |
11183 | | |
11184 | 0 | if (ctxt->instate == XML_PARSER_EPILOG) { |
11185 | 0 | if (ctxt->errNo == XML_ERR_OK) |
11186 | 0 | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_END, NULL); |
11187 | 0 | ctxt->instate = XML_PARSER_EOF; |
11188 | 0 | xmlFinishDocument(ctxt); |
11189 | 0 | } else { |
11190 | 0 | ctxt->instate = XML_PARSER_START_TAG; |
11191 | 0 | } |
11192 | 0 | break; |
11193 | 0 | case XML_PARSER_DTD: { |
11194 | 0 | if ((!terminate) && (!xmlParseLookupInternalSubset(ctxt))) |
11195 | 0 | goto done; |
11196 | 0 | xmlParseInternalSubset(ctxt); |
11197 | 0 | ctxt->inSubset = 2; |
11198 | 0 | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
11199 | 0 | (ctxt->sax->externalSubset != NULL)) |
11200 | 0 | ctxt->sax->externalSubset(ctxt->userData, ctxt->intSubName, |
11201 | 0 | ctxt->extSubSystem, ctxt->extSubURI); |
11202 | 0 | ctxt->inSubset = 0; |
11203 | 0 | xmlCleanSpecialAttr(ctxt); |
11204 | 0 | ctxt->instate = XML_PARSER_PROLOG; |
11205 | 0 | break; |
11206 | 0 | } |
11207 | 0 | default: |
11208 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_INTERNAL_ERROR, |
11209 | 0 | "PP: internal error\n"); |
11210 | 0 | ctxt->instate = XML_PARSER_EOF; |
11211 | 0 | break; |
11212 | 0 | } |
11213 | 0 | } |
11214 | 0 | done: |
11215 | 0 | return(ret); |
11216 | 0 | } |
11217 | | |
11218 | | /** |
11219 | | * Parse a chunk of memory in push parser mode. |
11220 | | * |
11221 | | * Assumes that the parser context was initialized with |
11222 | | * #xmlCreatePushParserCtxt. |
11223 | | * |
11224 | | * The last chunk, which will often be empty, must be marked with |
11225 | | * the `terminate` flag. With the default SAX callbacks, the resulting |
11226 | | * document will be available in ctxt->myDoc. This pointer will not |
11227 | | * be freed when calling #xmlFreeParserCtxt and must be freed by the |
11228 | | * caller. If the document isn't well-formed, it will still be returned |
11229 | | * in ctxt->myDoc. |
11230 | | * |
11231 | | * As an exception, #xmlCtxtResetPush will free the document in |
11232 | | * ctxt->myDoc. So ctxt->myDoc should be set to NULL after extracting |
11233 | | * the document. |
11234 | | * |
11235 | | * Since 2.14.0, #xmlCtxtGetDocument can be used to retrieve the |
11236 | | * result document. |
11237 | | * |
11238 | | * @param ctxt an XML parser context |
11239 | | * @param chunk chunk of memory |
11240 | | * @param size size of chunk in bytes |
11241 | | * @param terminate last chunk indicator |
11242 | | * @returns an xmlParserErrors code (0 on success). |
11243 | | */ |
11244 | | int |
11245 | | xmlParseChunk(xmlParserCtxt *ctxt, const char *chunk, int size, |
11246 | 0 | int terminate) { |
11247 | 0 | size_t curBase; |
11248 | 0 | size_t maxLength; |
11249 | 0 | size_t pos; |
11250 | 0 | int end_in_lf = 0; |
11251 | 0 | int res; |
11252 | |
|
11253 | 0 | if ((ctxt == NULL) || (size < 0)) |
11254 | 0 | return(XML_ERR_ARGUMENT); |
11255 | 0 | if ((chunk == NULL) && (size > 0)) |
11256 | 0 | return(XML_ERR_ARGUMENT); |
11257 | 0 | if ((ctxt->input == NULL) || (ctxt->input->buf == NULL)) |
11258 | 0 | return(XML_ERR_ARGUMENT); |
11259 | 0 | if (ctxt->disableSAX != 0) |
11260 | 0 | return(ctxt->errNo); |
11261 | | |
11262 | 0 | ctxt->input->flags |= XML_INPUT_PROGRESSIVE; |
11263 | 0 | if (ctxt->instate == XML_PARSER_START) |
11264 | 0 | xmlCtxtInitializeLate(ctxt); |
11265 | 0 | if ((size > 0) && (chunk != NULL) && (!terminate) && |
11266 | 0 | (chunk[size - 1] == '\r')) { |
11267 | 0 | end_in_lf = 1; |
11268 | 0 | size--; |
11269 | 0 | } |
11270 | | |
11271 | | /* |
11272 | | * Also push an empty chunk to make sure that the raw buffer |
11273 | | * will be flushed if there is an encoder. |
11274 | | */ |
11275 | 0 | pos = ctxt->input->cur - ctxt->input->base; |
11276 | 0 | res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk); |
11277 | 0 | xmlBufUpdateInput(ctxt->input->buf->buffer, ctxt->input, pos); |
11278 | 0 | if (res < 0) { |
11279 | 0 | xmlCtxtErrIO(ctxt, ctxt->input->buf->error, NULL); |
11280 | 0 | xmlHaltParser(ctxt); |
11281 | 0 | return(ctxt->errNo); |
11282 | 0 | } |
11283 | | |
11284 | 0 | xmlParseTryOrFinish(ctxt, terminate); |
11285 | |
|
11286 | 0 | curBase = ctxt->input->cur - ctxt->input->base; |
11287 | 0 | maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
11288 | 0 | XML_MAX_HUGE_LENGTH : |
11289 | 0 | XML_MAX_LOOKUP_LIMIT; |
11290 | 0 | if (curBase > maxLength) { |
11291 | 0 | xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, |
11292 | 0 | "Buffer size limit exceeded, try XML_PARSE_HUGE\n"); |
11293 | 0 | xmlHaltParser(ctxt); |
11294 | 0 | } |
11295 | |
|
11296 | 0 | if ((ctxt->errNo != XML_ERR_OK) && (ctxt->disableSAX != 0)) |
11297 | 0 | return(ctxt->errNo); |
11298 | | |
11299 | 0 | if (end_in_lf == 1) { |
11300 | 0 | pos = ctxt->input->cur - ctxt->input->base; |
11301 | 0 | res = xmlParserInputBufferPush(ctxt->input->buf, 1, "\r"); |
11302 | 0 | xmlBufUpdateInput(ctxt->input->buf->buffer, ctxt->input, pos); |
11303 | 0 | if (res < 0) { |
11304 | 0 | xmlCtxtErrIO(ctxt, ctxt->input->buf->error, NULL); |
11305 | 0 | xmlHaltParser(ctxt); |
11306 | 0 | return(ctxt->errNo); |
11307 | 0 | } |
11308 | 0 | } |
11309 | 0 | if (terminate) { |
11310 | | /* |
11311 | | * Check for termination |
11312 | | */ |
11313 | 0 | if ((ctxt->instate != XML_PARSER_EOF) && |
11314 | 0 | (ctxt->instate != XML_PARSER_EPILOG)) { |
11315 | 0 | if (ctxt->nameNr > 0) { |
11316 | 0 | const xmlChar *name = ctxt->nameTab[ctxt->nameNr - 1]; |
11317 | 0 | int line = ctxt->pushTab[ctxt->nameNr - 1].line; |
11318 | 0 | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NOT_FINISHED, |
11319 | 0 | "Premature end of data in tag %s line %d\n", |
11320 | 0 | name, line, NULL); |
11321 | 0 | } else if (ctxt->instate == XML_PARSER_START) { |
11322 | 0 | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); |
11323 | 0 | } else { |
11324 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_DOCUMENT_EMPTY, |
11325 | 0 | "Start tag expected, '<' not found\n"); |
11326 | 0 | } |
11327 | 0 | } else { |
11328 | 0 | xmlParserCheckEOF(ctxt, XML_ERR_DOCUMENT_END); |
11329 | 0 | } |
11330 | 0 | if (ctxt->instate != XML_PARSER_EOF) { |
11331 | 0 | ctxt->instate = XML_PARSER_EOF; |
11332 | 0 | xmlFinishDocument(ctxt); |
11333 | 0 | } |
11334 | 0 | } |
11335 | 0 | if (ctxt->wellFormed == 0) |
11336 | 0 | return((xmlParserErrors) ctxt->errNo); |
11337 | 0 | else |
11338 | 0 | return(0); |
11339 | 0 | } |
11340 | | |
11341 | | /************************************************************************ |
11342 | | * * |
11343 | | * I/O front end functions to the parser * |
11344 | | * * |
11345 | | ************************************************************************/ |
11346 | | |
11347 | | /** |
11348 | | * Create a parser context for using the XML parser in push mode. |
11349 | | * See #xmlParseChunk. |
11350 | | * |
11351 | | * Passing an initial chunk is useless and deprecated. |
11352 | | * |
11353 | | * The push parser doesn't support recovery mode or the |
11354 | | * XML_PARSE_NOBLANKS option. |
11355 | | * |
11356 | | * `filename` is used as base URI to fetch external entities and for |
11357 | | * error reports. |
11358 | | * |
11359 | | * @param sax a SAX handler (optional) |
11360 | | * @param user_data user data for SAX callbacks (optional) |
11361 | | * @param chunk initial chunk (optional, deprecated) |
11362 | | * @param size size of initial chunk in bytes |
11363 | | * @param filename file name or URI (optional) |
11364 | | * @returns the new parser context or NULL if a memory allocation |
11365 | | * failed. |
11366 | | */ |
11367 | | |
11368 | | xmlParserCtxt * |
11369 | | xmlCreatePushParserCtxt(xmlSAXHandler *sax, void *user_data, |
11370 | 0 | const char *chunk, int size, const char *filename) { |
11371 | 0 | xmlParserCtxtPtr ctxt; |
11372 | 0 | xmlParserInputPtr input; |
11373 | |
|
11374 | 0 | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
11375 | 0 | if (ctxt == NULL) |
11376 | 0 | return(NULL); |
11377 | | |
11378 | 0 | ctxt->options &= ~XML_PARSE_NODICT; |
11379 | 0 | ctxt->dictNames = 1; |
11380 | |
|
11381 | 0 | input = xmlNewPushInput(filename, chunk, size); |
11382 | 0 | if (input == NULL) { |
11383 | 0 | xmlFreeParserCtxt(ctxt); |
11384 | 0 | return(NULL); |
11385 | 0 | } |
11386 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
11387 | 0 | xmlFreeInputStream(input); |
11388 | 0 | xmlFreeParserCtxt(ctxt); |
11389 | 0 | return(NULL); |
11390 | 0 | } |
11391 | | |
11392 | 0 | return(ctxt); |
11393 | 0 | } |
11394 | | #endif /* LIBXML_PUSH_ENABLED */ |
11395 | | |
11396 | | /** |
11397 | | * Blocks further parser processing |
11398 | | * |
11399 | | * @param ctxt an XML parser context |
11400 | | */ |
11401 | | void |
11402 | 0 | xmlStopParser(xmlParserCtxt *ctxt) { |
11403 | 0 | if (ctxt == NULL) |
11404 | 0 | return; |
11405 | 0 | xmlHaltParser(ctxt); |
11406 | 0 | if (ctxt->errNo != XML_ERR_NO_MEMORY) |
11407 | 0 | ctxt->errNo = XML_ERR_USER_STOP; |
11408 | 0 | } |
11409 | | |
11410 | | /** |
11411 | | * Create a parser context for using the XML parser with an existing |
11412 | | * I/O stream |
11413 | | * |
11414 | | * @param sax a SAX handler (optional) |
11415 | | * @param user_data user data for SAX callbacks (optional) |
11416 | | * @param ioread an I/O read function |
11417 | | * @param ioclose an I/O close function (optional) |
11418 | | * @param ioctx an I/O handler |
11419 | | * @param enc the charset encoding if known (deprecated) |
11420 | | * @returns the new parser context or NULL |
11421 | | */ |
11422 | | xmlParserCtxt * |
11423 | | xmlCreateIOParserCtxt(xmlSAXHandler *sax, void *user_data, |
11424 | | xmlInputReadCallback ioread, |
11425 | | xmlInputCloseCallback ioclose, |
11426 | 0 | void *ioctx, xmlCharEncoding enc) { |
11427 | 0 | xmlParserCtxtPtr ctxt; |
11428 | 0 | xmlParserInputPtr input; |
11429 | 0 | const char *encoding; |
11430 | |
|
11431 | 0 | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
11432 | 0 | if (ctxt == NULL) |
11433 | 0 | return(NULL); |
11434 | | |
11435 | 0 | encoding = xmlGetCharEncodingName(enc); |
11436 | 0 | input = xmlCtxtNewInputFromIO(ctxt, NULL, ioread, ioclose, ioctx, |
11437 | 0 | encoding, 0); |
11438 | 0 | if (input == NULL) { |
11439 | 0 | xmlFreeParserCtxt(ctxt); |
11440 | 0 | return (NULL); |
11441 | 0 | } |
11442 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
11443 | 0 | xmlFreeInputStream(input); |
11444 | 0 | xmlFreeParserCtxt(ctxt); |
11445 | 0 | return(NULL); |
11446 | 0 | } |
11447 | | |
11448 | 0 | return(ctxt); |
11449 | 0 | } |
11450 | | |
11451 | | #ifdef LIBXML_VALID_ENABLED |
11452 | | /************************************************************************ |
11453 | | * * |
11454 | | * Front ends when parsing a DTD * |
11455 | | * * |
11456 | | ************************************************************************/ |
11457 | | |
11458 | | /** |
11459 | | * Parse a DTD. |
11460 | | * |
11461 | | * Option XML_PARSE_DTDLOAD should be enabled in the parser context |
11462 | | * to make external entities work. |
11463 | | * |
11464 | | * @since 2.14.0 |
11465 | | * |
11466 | | * @param ctxt a parser context |
11467 | | * @param input a parser input |
11468 | | * @param publicId public ID of the DTD (optional) |
11469 | | * @param systemId system ID of the DTD (optional) |
11470 | | * @returns the resulting xmlDtd or NULL in case of error. |
11471 | | * `input` will be freed by the function in any case. |
11472 | | */ |
11473 | | xmlDtd * |
11474 | | xmlCtxtParseDtd(xmlParserCtxt *ctxt, xmlParserInput *input, |
11475 | 0 | const xmlChar *publicId, const xmlChar *systemId) { |
11476 | 0 | xmlDtdPtr ret = NULL; |
11477 | |
|
11478 | 0 | if ((ctxt == NULL) || (input == NULL)) { |
11479 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
11480 | 0 | xmlFreeInputStream(input); |
11481 | 0 | return(NULL); |
11482 | 0 | } |
11483 | | |
11484 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
11485 | 0 | xmlFreeInputStream(input); |
11486 | 0 | return(NULL); |
11487 | 0 | } |
11488 | | |
11489 | 0 | if (publicId == NULL) |
11490 | 0 | publicId = BAD_CAST "none"; |
11491 | 0 | if (systemId == NULL) |
11492 | 0 | systemId = BAD_CAST "none"; |
11493 | |
|
11494 | 0 | ctxt->myDoc = xmlNewDoc(BAD_CAST "1.0"); |
11495 | 0 | if (ctxt->myDoc == NULL) { |
11496 | 0 | xmlErrMemory(ctxt); |
11497 | 0 | goto error; |
11498 | 0 | } |
11499 | 0 | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
11500 | 0 | ctxt->myDoc->extSubset = xmlNewDtd(ctxt->myDoc, BAD_CAST "none", |
11501 | 0 | publicId, systemId); |
11502 | 0 | if (ctxt->myDoc->extSubset == NULL) { |
11503 | 0 | xmlErrMemory(ctxt); |
11504 | 0 | xmlFreeDoc(ctxt->myDoc); |
11505 | 0 | goto error; |
11506 | 0 | } |
11507 | | |
11508 | 0 | xmlParseExternalSubset(ctxt, publicId, systemId); |
11509 | |
|
11510 | 0 | if (ctxt->wellFormed) { |
11511 | 0 | ret = ctxt->myDoc->extSubset; |
11512 | 0 | ctxt->myDoc->extSubset = NULL; |
11513 | 0 | if (ret != NULL) { |
11514 | 0 | xmlNodePtr tmp; |
11515 | |
|
11516 | 0 | ret->doc = NULL; |
11517 | 0 | tmp = ret->children; |
11518 | 0 | while (tmp != NULL) { |
11519 | 0 | tmp->doc = NULL; |
11520 | 0 | tmp = tmp->next; |
11521 | 0 | } |
11522 | 0 | } |
11523 | 0 | } else { |
11524 | 0 | ret = NULL; |
11525 | 0 | } |
11526 | 0 | xmlFreeDoc(ctxt->myDoc); |
11527 | 0 | ctxt->myDoc = NULL; |
11528 | |
|
11529 | 0 | error: |
11530 | 0 | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
11531 | |
|
11532 | 0 | return(ret); |
11533 | 0 | } |
11534 | | |
11535 | | /** |
11536 | | * Load and parse a DTD |
11537 | | * |
11538 | | * @deprecated Use #xmlCtxtParseDtd. |
11539 | | * |
11540 | | * @param sax the SAX handler block or NULL |
11541 | | * @param input an Input Buffer |
11542 | | * @param enc the charset encoding if known |
11543 | | * @returns the resulting xmlDtd or NULL in case of error. |
11544 | | * `input` will be freed by the function in any case. |
11545 | | */ |
11546 | | |
11547 | | xmlDtd * |
11548 | | xmlIOParseDTD(xmlSAXHandler *sax, xmlParserInputBuffer *input, |
11549 | 0 | xmlCharEncoding enc) { |
11550 | 0 | xmlDtdPtr ret = NULL; |
11551 | 0 | xmlParserCtxtPtr ctxt; |
11552 | 0 | xmlParserInputPtr pinput = NULL; |
11553 | |
|
11554 | 0 | if (input == NULL) |
11555 | 0 | return(NULL); |
11556 | | |
11557 | 0 | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
11558 | 0 | if (ctxt == NULL) { |
11559 | 0 | xmlFreeParserInputBuffer(input); |
11560 | 0 | return(NULL); |
11561 | 0 | } |
11562 | 0 | xmlCtxtSetOptions(ctxt, XML_PARSE_DTDLOAD); |
11563 | | |
11564 | | /* |
11565 | | * generate a parser input from the I/O handler |
11566 | | */ |
11567 | |
|
11568 | 0 | pinput = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); |
11569 | 0 | if (pinput == NULL) { |
11570 | 0 | xmlFreeParserInputBuffer(input); |
11571 | 0 | xmlFreeParserCtxt(ctxt); |
11572 | 0 | return(NULL); |
11573 | 0 | } |
11574 | | |
11575 | 0 | if (enc != XML_CHAR_ENCODING_NONE) { |
11576 | 0 | xmlSwitchEncoding(ctxt, enc); |
11577 | 0 | } |
11578 | |
|
11579 | 0 | ret = xmlCtxtParseDtd(ctxt, pinput, NULL, NULL); |
11580 | |
|
11581 | 0 | xmlFreeParserCtxt(ctxt); |
11582 | 0 | return(ret); |
11583 | 0 | } |
11584 | | |
11585 | | /** |
11586 | | * Load and parse an external subset. |
11587 | | * |
11588 | | * @deprecated Use #xmlCtxtParseDtd. |
11589 | | * |
11590 | | * @param sax the SAX handler block |
11591 | | * @param publicId public identifier of the DTD (optional) |
11592 | | * @param systemId system identifier (URL) of the DTD |
11593 | | * @returns the resulting xmlDtd or NULL in case of error. |
11594 | | */ |
11595 | | |
11596 | | xmlDtd * |
11597 | | xmlSAXParseDTD(xmlSAXHandler *sax, const xmlChar *publicId, |
11598 | 0 | const xmlChar *systemId) { |
11599 | 0 | xmlDtdPtr ret = NULL; |
11600 | 0 | xmlParserCtxtPtr ctxt; |
11601 | 0 | xmlParserInputPtr input = NULL; |
11602 | 0 | xmlChar* systemIdCanonic; |
11603 | |
|
11604 | 0 | if ((publicId == NULL) && (systemId == NULL)) return(NULL); |
11605 | | |
11606 | 0 | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
11607 | 0 | if (ctxt == NULL) { |
11608 | 0 | return(NULL); |
11609 | 0 | } |
11610 | 0 | xmlCtxtSetOptions(ctxt, XML_PARSE_DTDLOAD); |
11611 | | |
11612 | | /* |
11613 | | * Canonicalise the system ID |
11614 | | */ |
11615 | 0 | systemIdCanonic = xmlCanonicPath(systemId); |
11616 | 0 | if ((systemId != NULL) && (systemIdCanonic == NULL)) { |
11617 | 0 | xmlFreeParserCtxt(ctxt); |
11618 | 0 | return(NULL); |
11619 | 0 | } |
11620 | | |
11621 | | /* |
11622 | | * Ask the Entity resolver to load the damn thing |
11623 | | */ |
11624 | | |
11625 | 0 | if ((ctxt->sax != NULL) && (ctxt->sax->resolveEntity != NULL)) |
11626 | 0 | input = ctxt->sax->resolveEntity(ctxt->userData, publicId, |
11627 | 0 | systemIdCanonic); |
11628 | 0 | if (input == NULL) { |
11629 | 0 | xmlFreeParserCtxt(ctxt); |
11630 | 0 | if (systemIdCanonic != NULL) |
11631 | 0 | xmlFree(systemIdCanonic); |
11632 | 0 | return(NULL); |
11633 | 0 | } |
11634 | | |
11635 | 0 | if (input->filename == NULL) |
11636 | 0 | input->filename = (char *) systemIdCanonic; |
11637 | 0 | else |
11638 | 0 | xmlFree(systemIdCanonic); |
11639 | |
|
11640 | 0 | ret = xmlCtxtParseDtd(ctxt, input, publicId, systemId); |
11641 | |
|
11642 | 0 | xmlFreeParserCtxt(ctxt); |
11643 | 0 | return(ret); |
11644 | 0 | } |
11645 | | |
11646 | | |
11647 | | /** |
11648 | | * Load and parse an external subset. |
11649 | | * |
11650 | | * @param publicId public identifier of the DTD (optional) |
11651 | | * @param systemId system identifier (URL) of the DTD |
11652 | | * @returns the resulting xmlDtd or NULL in case of error. |
11653 | | */ |
11654 | | |
11655 | | xmlDtd * |
11656 | 0 | xmlParseDTD(const xmlChar *publicId, const xmlChar *systemId) { |
11657 | 0 | return(xmlSAXParseDTD(NULL, publicId, systemId)); |
11658 | 0 | } |
11659 | | #endif /* LIBXML_VALID_ENABLED */ |
11660 | | |
11661 | | /************************************************************************ |
11662 | | * * |
11663 | | * Front ends when parsing an Entity * |
11664 | | * * |
11665 | | ************************************************************************/ |
11666 | | |
11667 | | static xmlNodePtr |
11668 | | xmlCtxtParseContentInternal(xmlParserCtxtPtr ctxt, xmlParserInputPtr input, |
11669 | 7.17k | int hasTextDecl, int buildTree) { |
11670 | 7.17k | xmlNodePtr root = NULL; |
11671 | 7.17k | xmlNodePtr list = NULL; |
11672 | 7.17k | xmlChar *rootName = BAD_CAST "#root"; |
11673 | 7.17k | int result; |
11674 | | |
11675 | 7.17k | if (buildTree) { |
11676 | 7.17k | root = xmlNewDocNode(ctxt->myDoc, NULL, rootName, NULL); |
11677 | 7.17k | if (root == NULL) { |
11678 | 0 | xmlErrMemory(ctxt); |
11679 | 0 | goto error; |
11680 | 0 | } |
11681 | 7.17k | } |
11682 | | |
11683 | 7.17k | if (xmlCtxtPushInput(ctxt, input) < 0) |
11684 | 0 | goto error; |
11685 | | |
11686 | 7.17k | nameNsPush(ctxt, rootName, NULL, NULL, 0, 0); |
11687 | 7.17k | spacePush(ctxt, -1); |
11688 | | |
11689 | 7.17k | if (buildTree) |
11690 | 7.17k | nodePush(ctxt, root); |
11691 | | |
11692 | 7.17k | if (hasTextDecl) { |
11693 | 0 | xmlDetectEncoding(ctxt); |
11694 | | |
11695 | | /* |
11696 | | * Parse a possible text declaration first |
11697 | | */ |
11698 | 0 | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && |
11699 | 0 | (IS_BLANK_CH(NXT(5)))) { |
11700 | 0 | xmlParseTextDecl(ctxt); |
11701 | | /* |
11702 | | * An XML-1.0 document can't reference an entity not XML-1.0 |
11703 | | */ |
11704 | 0 | if ((xmlStrEqual(ctxt->version, BAD_CAST "1.0")) && |
11705 | 0 | (!xmlStrEqual(ctxt->input->version, BAD_CAST "1.0"))) { |
11706 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_VERSION_MISMATCH, |
11707 | 0 | "Version mismatch between document and " |
11708 | 0 | "entity\n"); |
11709 | 0 | } |
11710 | 0 | } |
11711 | 0 | } |
11712 | | |
11713 | 7.17k | xmlParseContentInternal(ctxt); |
11714 | | |
11715 | 7.17k | if (ctxt->input->cur < ctxt->input->end) |
11716 | 35 | xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); |
11717 | | |
11718 | 7.17k | if ((ctxt->wellFormed) || |
11719 | 7.17k | ((ctxt->recovery) && (!xmlCtxtIsCatastrophicError(ctxt)))) { |
11720 | 7.04k | if (root != NULL) { |
11721 | 7.04k | xmlNodePtr cur; |
11722 | | |
11723 | | /* |
11724 | | * Unlink newly created node list. |
11725 | | */ |
11726 | 7.04k | list = root->children; |
11727 | 7.04k | root->children = NULL; |
11728 | 7.04k | root->last = NULL; |
11729 | 8.06k | for (cur = list; cur != NULL; cur = cur->next) |
11730 | 1.02k | cur->parent = NULL; |
11731 | 7.04k | } |
11732 | 7.04k | } |
11733 | | |
11734 | | /* |
11735 | | * Read the rest of the stream in case of errors. We want |
11736 | | * to account for the whole entity size. |
11737 | | */ |
11738 | 7.17k | do { |
11739 | 7.17k | ctxt->input->cur = ctxt->input->end; |
11740 | 7.17k | xmlParserShrink(ctxt); |
11741 | 7.17k | result = xmlParserGrow(ctxt); |
11742 | 7.17k | } while (result > 0); |
11743 | | |
11744 | 7.17k | if (buildTree) |
11745 | 7.17k | nodePop(ctxt); |
11746 | | |
11747 | 7.17k | namePop(ctxt); |
11748 | 7.17k | spacePop(ctxt); |
11749 | | |
11750 | 7.17k | xmlCtxtPopInput(ctxt); |
11751 | | |
11752 | 7.17k | error: |
11753 | 7.17k | xmlFreeNode(root); |
11754 | | |
11755 | 7.17k | return(list); |
11756 | 7.17k | } |
11757 | | |
11758 | | static void |
11759 | 7.18k | xmlCtxtParseEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr ent) { |
11760 | 7.18k | xmlParserInputPtr input; |
11761 | 7.18k | xmlNodePtr list; |
11762 | 7.18k | unsigned long consumed; |
11763 | 7.18k | int isExternal; |
11764 | 7.18k | int buildTree; |
11765 | 7.18k | int oldMinNsIndex; |
11766 | 7.18k | int oldNodelen, oldNodemem; |
11767 | | |
11768 | 7.18k | isExternal = (ent->etype == XML_EXTERNAL_GENERAL_PARSED_ENTITY); |
11769 | 7.18k | buildTree = (ctxt->node != NULL); |
11770 | | |
11771 | | /* |
11772 | | * Recursion check |
11773 | | */ |
11774 | 7.18k | if (ent->flags & XML_ENT_EXPANDING) { |
11775 | 10 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
11776 | 10 | xmlHaltParser(ctxt); |
11777 | 10 | goto error; |
11778 | 10 | } |
11779 | | |
11780 | | /* |
11781 | | * Load entity |
11782 | | */ |
11783 | 7.17k | input = xmlNewEntityInputStream(ctxt, ent); |
11784 | 7.17k | if (input == NULL) |
11785 | 0 | goto error; |
11786 | | |
11787 | | /* |
11788 | | * When building a tree, we need to limit the scope of namespace |
11789 | | * declarations, so that entities don't reference xmlNs structs |
11790 | | * from the parent of a reference. |
11791 | | */ |
11792 | 7.17k | oldMinNsIndex = ctxt->nsdb->minNsIndex; |
11793 | 7.17k | if (buildTree) |
11794 | 7.17k | ctxt->nsdb->minNsIndex = ctxt->nsNr; |
11795 | | |
11796 | 7.17k | oldNodelen = ctxt->nodelen; |
11797 | 7.17k | oldNodemem = ctxt->nodemem; |
11798 | 7.17k | ctxt->nodelen = 0; |
11799 | 7.17k | ctxt->nodemem = 0; |
11800 | | |
11801 | | /* |
11802 | | * Parse content |
11803 | | * |
11804 | | * This initiates a recursive call chain: |
11805 | | * |
11806 | | * - xmlCtxtParseContentInternal |
11807 | | * - xmlParseContentInternal |
11808 | | * - xmlParseReference |
11809 | | * - xmlCtxtParseEntity |
11810 | | * |
11811 | | * The nesting depth is limited by the maximum number of inputs, |
11812 | | * see xmlCtxtPushInput. |
11813 | | * |
11814 | | * It's possible to make this non-recursive (minNsIndex must be |
11815 | | * stored in the input struct) at the expense of code readability. |
11816 | | */ |
11817 | | |
11818 | 7.17k | ent->flags |= XML_ENT_EXPANDING; |
11819 | | |
11820 | 7.17k | list = xmlCtxtParseContentInternal(ctxt, input, isExternal, buildTree); |
11821 | | |
11822 | 7.17k | ent->flags &= ~XML_ENT_EXPANDING; |
11823 | | |
11824 | 7.17k | ctxt->nsdb->minNsIndex = oldMinNsIndex; |
11825 | 7.17k | ctxt->nodelen = oldNodelen; |
11826 | 7.17k | ctxt->nodemem = oldNodemem; |
11827 | | |
11828 | | /* |
11829 | | * Entity size accounting |
11830 | | */ |
11831 | 7.17k | consumed = input->consumed; |
11832 | 7.17k | xmlSaturatedAddSizeT(&consumed, input->end - input->base); |
11833 | | |
11834 | 7.17k | if ((ent->flags & XML_ENT_CHECKED) == 0) |
11835 | 563 | xmlSaturatedAdd(&ent->expandedSize, consumed); |
11836 | | |
11837 | 7.17k | if ((ent->flags & XML_ENT_PARSED) == 0) { |
11838 | 563 | if (isExternal) |
11839 | 0 | xmlSaturatedAdd(&ctxt->sizeentities, consumed); |
11840 | | |
11841 | 563 | ent->children = list; |
11842 | | |
11843 | 1.58k | while (list != NULL) { |
11844 | 1.02k | list->parent = (xmlNodePtr) ent; |
11845 | | |
11846 | | /* |
11847 | | * Downstream code like the nginx xslt module can set |
11848 | | * ctxt->myDoc->extSubset to a separate DTD, so the entity |
11849 | | * might have a different or a NULL document. |
11850 | | */ |
11851 | 1.02k | if (list->doc != ent->doc) |
11852 | 0 | xmlSetTreeDoc(list, ent->doc); |
11853 | | |
11854 | 1.02k | if (list->next == NULL) |
11855 | 397 | ent->last = list; |
11856 | 1.02k | list = list->next; |
11857 | 1.02k | } |
11858 | 6.61k | } else { |
11859 | 6.61k | xmlFreeNodeList(list); |
11860 | 6.61k | } |
11861 | | |
11862 | 7.17k | xmlFreeInputStream(input); |
11863 | | |
11864 | 7.18k | error: |
11865 | 7.18k | ent->flags |= XML_ENT_PARSED | XML_ENT_CHECKED; |
11866 | 7.18k | } |
11867 | | |
11868 | | /** |
11869 | | * Parse an external general entity within an existing parsing context |
11870 | | * An external general parsed entity is well-formed if it matches the |
11871 | | * production labeled extParsedEnt. |
11872 | | * |
11873 | | * [78] extParsedEnt ::= TextDecl? content |
11874 | | * |
11875 | | * @param ctxt the existing parsing context |
11876 | | * @param URL the URL for the entity to load |
11877 | | * @param ID the System ID for the entity to load |
11878 | | * @param listOut the return value for the set of parsed nodes |
11879 | | * @returns 0 if the entity is well formed, -1 in case of args problem and |
11880 | | * the parser error code otherwise |
11881 | | */ |
11882 | | |
11883 | | int |
11884 | | xmlParseCtxtExternalEntity(xmlParserCtxt *ctxt, const xmlChar *URL, |
11885 | 0 | const xmlChar *ID, xmlNode **listOut) { |
11886 | 0 | xmlParserInputPtr input; |
11887 | 0 | xmlNodePtr list; |
11888 | |
|
11889 | 0 | if (listOut != NULL) |
11890 | 0 | *listOut = NULL; |
11891 | |
|
11892 | 0 | if (ctxt == NULL) |
11893 | 0 | return(XML_ERR_ARGUMENT); |
11894 | | |
11895 | 0 | input = xmlLoadResource(ctxt, (char *) URL, (char *) ID, |
11896 | 0 | XML_RESOURCE_GENERAL_ENTITY); |
11897 | 0 | if (input == NULL) |
11898 | 0 | return(ctxt->errNo); |
11899 | | |
11900 | 0 | xmlCtxtInitializeLate(ctxt); |
11901 | |
|
11902 | 0 | list = xmlCtxtParseContentInternal(ctxt, input, /* hasTextDecl */ 1, 1); |
11903 | 0 | if (listOut != NULL) |
11904 | 0 | *listOut = list; |
11905 | 0 | else |
11906 | 0 | xmlFreeNodeList(list); |
11907 | |
|
11908 | 0 | xmlFreeInputStream(input); |
11909 | 0 | return(ctxt->errNo); |
11910 | 0 | } |
11911 | | |
11912 | | #ifdef LIBXML_SAX1_ENABLED |
11913 | | /** |
11914 | | * Parse an external general entity |
11915 | | * An external general parsed entity is well-formed if it matches the |
11916 | | * production labeled extParsedEnt. |
11917 | | * |
11918 | | * @deprecated Use #xmlParseCtxtExternalEntity. |
11919 | | * |
11920 | | * [78] extParsedEnt ::= TextDecl? content |
11921 | | * |
11922 | | * @param doc the document the chunk pertains to |
11923 | | * @param sax the SAX handler block (possibly NULL) |
11924 | | * @param user_data The user data returned on SAX callbacks (possibly NULL) |
11925 | | * @param depth Used for loop detection, use 0 |
11926 | | * @param URL the URL for the entity to load |
11927 | | * @param ID the System ID for the entity to load |
11928 | | * @param list the return value for the set of parsed nodes |
11929 | | * @returns 0 if the entity is well formed, -1 in case of args problem and |
11930 | | * the parser error code otherwise |
11931 | | */ |
11932 | | |
11933 | | int |
11934 | | xmlParseExternalEntity(xmlDoc *doc, xmlSAXHandler *sax, void *user_data, |
11935 | 0 | int depth, const xmlChar *URL, const xmlChar *ID, xmlNode **list) { |
11936 | 0 | xmlParserCtxtPtr ctxt; |
11937 | 0 | int ret; |
11938 | |
|
11939 | 0 | if (list != NULL) |
11940 | 0 | *list = NULL; |
11941 | |
|
11942 | 0 | if (doc == NULL) |
11943 | 0 | return(XML_ERR_ARGUMENT); |
11944 | | |
11945 | 0 | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
11946 | 0 | if (ctxt == NULL) |
11947 | 0 | return(XML_ERR_NO_MEMORY); |
11948 | | |
11949 | 0 | ctxt->depth = depth; |
11950 | 0 | ctxt->myDoc = doc; |
11951 | 0 | ret = xmlParseCtxtExternalEntity(ctxt, URL, ID, list); |
11952 | |
|
11953 | 0 | xmlFreeParserCtxt(ctxt); |
11954 | 0 | return(ret); |
11955 | 0 | } |
11956 | | |
11957 | | /** |
11958 | | * Parse a well-balanced chunk of an XML document |
11959 | | * called by the parser |
11960 | | * The allowed sequence for the Well Balanced Chunk is the one defined by |
11961 | | * the content production in the XML grammar: |
11962 | | * |
11963 | | * [43] content ::= (element | CharData | Reference | CDSect | PI | |
11964 | | * Comment)* |
11965 | | * |
11966 | | * @param doc the document the chunk pertains to (must not be NULL) |
11967 | | * @param sax the SAX handler block (possibly NULL) |
11968 | | * @param user_data The user data returned on SAX callbacks (possibly NULL) |
11969 | | * @param depth Used for loop detection, use 0 |
11970 | | * @param string the input string in UTF8 or ISO-Latin (zero terminated) |
11971 | | * @param lst the return value for the set of parsed nodes |
11972 | | * @returns 0 if the chunk is well balanced, -1 in case of args problem and |
11973 | | * the parser error code otherwise |
11974 | | */ |
11975 | | |
11976 | | int |
11977 | | xmlParseBalancedChunkMemory(xmlDoc *doc, xmlSAXHandler *sax, |
11978 | 0 | void *user_data, int depth, const xmlChar *string, xmlNode **lst) { |
11979 | 0 | return xmlParseBalancedChunkMemoryRecover( doc, sax, user_data, |
11980 | 0 | depth, string, lst, 0 ); |
11981 | 0 | } |
11982 | | #endif /* LIBXML_SAX1_ENABLED */ |
11983 | | |
11984 | | /** |
11985 | | * Parse a well-balanced chunk of XML matching the 'content' production. |
11986 | | * |
11987 | | * Namespaces in scope of `node` and entities of `node`'s document are |
11988 | | * recognized. When validating, the DTD of `node`'s document is used. |
11989 | | * |
11990 | | * Always consumes `input` even in error case. |
11991 | | * |
11992 | | * @since 2.14.0 |
11993 | | * |
11994 | | * @param ctxt parser context |
11995 | | * @param input parser input |
11996 | | * @param node target node or document |
11997 | | * @param hasTextDecl whether to parse text declaration |
11998 | | * @returns a node list or NULL in case of error. |
11999 | | */ |
12000 | | xmlNode * |
12001 | | xmlCtxtParseContent(xmlParserCtxt *ctxt, xmlParserInput *input, |
12002 | 0 | xmlNode *node, int hasTextDecl) { |
12003 | 0 | xmlDocPtr doc; |
12004 | 0 | xmlNodePtr cur, list = NULL; |
12005 | 0 | int nsnr = 0; |
12006 | 0 | xmlDictPtr oldDict; |
12007 | 0 | int oldOptions, oldDictNames, oldLoadSubset; |
12008 | |
|
12009 | 0 | if ((ctxt == NULL) || (input == NULL) || (node == NULL)) { |
12010 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
12011 | 0 | goto exit; |
12012 | 0 | } |
12013 | | |
12014 | 0 | doc = node->doc; |
12015 | 0 | if (doc == NULL) { |
12016 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
12017 | 0 | goto exit; |
12018 | 0 | } |
12019 | | |
12020 | 0 | switch (node->type) { |
12021 | 0 | case XML_ELEMENT_NODE: |
12022 | 0 | case XML_DOCUMENT_NODE: |
12023 | 0 | case XML_HTML_DOCUMENT_NODE: |
12024 | 0 | break; |
12025 | | |
12026 | 0 | case XML_ATTRIBUTE_NODE: |
12027 | 0 | case XML_TEXT_NODE: |
12028 | 0 | case XML_CDATA_SECTION_NODE: |
12029 | 0 | case XML_ENTITY_REF_NODE: |
12030 | 0 | case XML_PI_NODE: |
12031 | 0 | case XML_COMMENT_NODE: |
12032 | 0 | for (cur = node->parent; cur != NULL; cur = node->parent) { |
12033 | 0 | if ((cur->type == XML_ELEMENT_NODE) || |
12034 | 0 | (cur->type == XML_DOCUMENT_NODE) || |
12035 | 0 | (cur->type == XML_HTML_DOCUMENT_NODE)) { |
12036 | 0 | node = cur; |
12037 | 0 | break; |
12038 | 0 | } |
12039 | 0 | } |
12040 | 0 | break; |
12041 | | |
12042 | 0 | default: |
12043 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
12044 | 0 | goto exit; |
12045 | 0 | } |
12046 | | |
12047 | 0 | xmlCtxtReset(ctxt); |
12048 | |
|
12049 | 0 | oldDict = ctxt->dict; |
12050 | 0 | oldOptions = ctxt->options; |
12051 | 0 | oldDictNames = ctxt->dictNames; |
12052 | 0 | oldLoadSubset = ctxt->loadsubset; |
12053 | | |
12054 | | /* |
12055 | | * Use input doc's dict if present, else assure XML_PARSE_NODICT is set. |
12056 | | */ |
12057 | 0 | if (doc->dict != NULL) { |
12058 | 0 | ctxt->dict = doc->dict; |
12059 | 0 | } else { |
12060 | 0 | ctxt->options |= XML_PARSE_NODICT; |
12061 | 0 | ctxt->dictNames = 0; |
12062 | 0 | } |
12063 | | |
12064 | | /* |
12065 | | * Disable IDs |
12066 | | */ |
12067 | 0 | ctxt->loadsubset |= XML_SKIP_IDS; |
12068 | |
|
12069 | 0 | ctxt->myDoc = doc; |
12070 | |
|
12071 | 0 | #ifdef LIBXML_HTML_ENABLED |
12072 | 0 | if (ctxt->html) { |
12073 | | /* |
12074 | | * When parsing in context, it makes no sense to add implied |
12075 | | * elements like html/body/etc... |
12076 | | */ |
12077 | 0 | ctxt->options |= HTML_PARSE_NOIMPLIED; |
12078 | |
|
12079 | 0 | list = htmlCtxtParseContentInternal(ctxt, input); |
12080 | 0 | } else |
12081 | 0 | #endif |
12082 | 0 | { |
12083 | 0 | xmlCtxtInitializeLate(ctxt); |
12084 | | |
12085 | | /* |
12086 | | * initialize the SAX2 namespaces stack |
12087 | | */ |
12088 | 0 | cur = node; |
12089 | 0 | while ((cur != NULL) && (cur->type == XML_ELEMENT_NODE)) { |
12090 | 0 | xmlNsPtr ns = cur->nsDef; |
12091 | 0 | xmlHashedString hprefix, huri; |
12092 | |
|
12093 | 0 | while (ns != NULL) { |
12094 | 0 | hprefix = xmlDictLookupHashed(ctxt->dict, ns->prefix, -1); |
12095 | 0 | huri = xmlDictLookupHashed(ctxt->dict, ns->href, -1); |
12096 | 0 | if (xmlParserNsPush(ctxt, &hprefix, &huri, ns, 1) > 0) |
12097 | 0 | nsnr++; |
12098 | 0 | ns = ns->next; |
12099 | 0 | } |
12100 | 0 | cur = cur->parent; |
12101 | 0 | } |
12102 | |
|
12103 | 0 | list = xmlCtxtParseContentInternal(ctxt, input, hasTextDecl, 1); |
12104 | |
|
12105 | 0 | if (nsnr > 0) |
12106 | 0 | xmlParserNsPop(ctxt, nsnr); |
12107 | 0 | } |
12108 | |
|
12109 | 0 | ctxt->dict = oldDict; |
12110 | 0 | ctxt->options = oldOptions; |
12111 | 0 | ctxt->dictNames = oldDictNames; |
12112 | 0 | ctxt->loadsubset = oldLoadSubset; |
12113 | 0 | ctxt->myDoc = NULL; |
12114 | 0 | ctxt->node = NULL; |
12115 | |
|
12116 | 0 | exit: |
12117 | 0 | xmlFreeInputStream(input); |
12118 | 0 | return(list); |
12119 | 0 | } |
12120 | | |
12121 | | /** |
12122 | | * Parse a well-balanced chunk of an XML document |
12123 | | * within the context (DTD, namespaces, etc ...) of the given node. |
12124 | | * |
12125 | | * The allowed sequence for the data is a Well Balanced Chunk defined by |
12126 | | * the content production in the XML grammar: |
12127 | | * |
12128 | | * [43] content ::= (element | CharData | Reference | CDSect | PI | |
12129 | | * Comment)* |
12130 | | * |
12131 | | * This function assumes the encoding of `node`'s document which is |
12132 | | * typically not what you want. A better alternative is |
12133 | | * #xmlCtxtParseContent. |
12134 | | * |
12135 | | * @param node the context node |
12136 | | * @param data the input string |
12137 | | * @param datalen the input string length in bytes |
12138 | | * @param options a combination of xmlParserOption |
12139 | | * @param listOut the return value for the set of parsed nodes |
12140 | | * @returns XML_ERR_OK if the chunk is well balanced, and the parser |
12141 | | * error code otherwise |
12142 | | */ |
12143 | | xmlParserErrors |
12144 | | xmlParseInNodeContext(xmlNode *node, const char *data, int datalen, |
12145 | 0 | int options, xmlNode **listOut) { |
12146 | 0 | xmlParserCtxtPtr ctxt; |
12147 | 0 | xmlParserInputPtr input; |
12148 | 0 | xmlDocPtr doc; |
12149 | 0 | xmlNodePtr list; |
12150 | 0 | xmlParserErrors ret; |
12151 | |
|
12152 | 0 | if (listOut == NULL) |
12153 | 0 | return(XML_ERR_INTERNAL_ERROR); |
12154 | 0 | *listOut = NULL; |
12155 | |
|
12156 | 0 | if ((node == NULL) || (data == NULL) || (datalen < 0)) |
12157 | 0 | return(XML_ERR_INTERNAL_ERROR); |
12158 | | |
12159 | 0 | doc = node->doc; |
12160 | 0 | if (doc == NULL) |
12161 | 0 | return(XML_ERR_INTERNAL_ERROR); |
12162 | | |
12163 | 0 | #ifdef LIBXML_HTML_ENABLED |
12164 | 0 | if (doc->type == XML_HTML_DOCUMENT_NODE) { |
12165 | 0 | ctxt = htmlNewParserCtxt(); |
12166 | 0 | } |
12167 | 0 | else |
12168 | 0 | #endif |
12169 | 0 | ctxt = xmlNewParserCtxt(); |
12170 | |
|
12171 | 0 | if (ctxt == NULL) |
12172 | 0 | return(XML_ERR_NO_MEMORY); |
12173 | | |
12174 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, NULL, data, datalen, |
12175 | 0 | (const char *) doc->encoding, |
12176 | 0 | XML_INPUT_BUF_STATIC); |
12177 | 0 | if (input == NULL) { |
12178 | 0 | xmlFreeParserCtxt(ctxt); |
12179 | 0 | return(XML_ERR_NO_MEMORY); |
12180 | 0 | } |
12181 | | |
12182 | 0 | xmlCtxtUseOptions(ctxt, options); |
12183 | |
|
12184 | 0 | list = xmlCtxtParseContent(ctxt, input, node, /* hasTextDecl */ 0); |
12185 | |
|
12186 | 0 | if (list == NULL) { |
12187 | 0 | ret = ctxt->errNo; |
12188 | 0 | if (ret == XML_ERR_ARGUMENT) |
12189 | 0 | ret = XML_ERR_INTERNAL_ERROR; |
12190 | 0 | } else { |
12191 | 0 | ret = XML_ERR_OK; |
12192 | 0 | *listOut = list; |
12193 | 0 | } |
12194 | |
|
12195 | 0 | xmlFreeParserCtxt(ctxt); |
12196 | |
|
12197 | 0 | return(ret); |
12198 | 0 | } |
12199 | | |
12200 | | #ifdef LIBXML_SAX1_ENABLED |
12201 | | /** |
12202 | | * Parse a well-balanced chunk of an XML document |
12203 | | * |
12204 | | * The allowed sequence for the Well Balanced Chunk is the one defined by |
12205 | | * the content production in the XML grammar: |
12206 | | * |
12207 | | * [43] content ::= (element | CharData | Reference | CDSect | PI | |
12208 | | * Comment)* |
12209 | | * |
12210 | | * In case recover is set to 1, the nodelist will not be empty even if |
12211 | | * the parsed chunk is not well balanced, assuming the parsing succeeded to |
12212 | | * some extent. |
12213 | | * |
12214 | | * @param doc the document the chunk pertains to (must not be NULL) |
12215 | | * @param sax the SAX handler block (possibly NULL) |
12216 | | * @param user_data The user data returned on SAX callbacks (possibly NULL) |
12217 | | * @param depth Used for loop detection, use 0 |
12218 | | * @param string the input string in UTF8 or ISO-Latin (zero terminated) |
12219 | | * @param listOut the return value for the set of parsed nodes |
12220 | | * @param recover return nodes even if the data is broken (use 0) |
12221 | | * @returns 0 if the chunk is well balanced, or thehe parser error code |
12222 | | * otherwise. |
12223 | | */ |
12224 | | int |
12225 | | xmlParseBalancedChunkMemoryRecover(xmlDoc *doc, xmlSAXHandler *sax, |
12226 | | void *user_data, int depth, const xmlChar *string, xmlNode **listOut, |
12227 | 0 | int recover) { |
12228 | 0 | xmlParserCtxtPtr ctxt; |
12229 | 0 | xmlParserInputPtr input; |
12230 | 0 | xmlNodePtr list; |
12231 | 0 | int ret; |
12232 | |
|
12233 | 0 | if (listOut != NULL) |
12234 | 0 | *listOut = NULL; |
12235 | |
|
12236 | 0 | if (string == NULL) |
12237 | 0 | return(XML_ERR_ARGUMENT); |
12238 | | |
12239 | 0 | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
12240 | 0 | if (ctxt == NULL) |
12241 | 0 | return(XML_ERR_NO_MEMORY); |
12242 | | |
12243 | 0 | xmlCtxtInitializeLate(ctxt); |
12244 | |
|
12245 | 0 | ctxt->depth = depth; |
12246 | 0 | ctxt->myDoc = doc; |
12247 | 0 | if (recover) { |
12248 | 0 | ctxt->options |= XML_PARSE_RECOVER; |
12249 | 0 | ctxt->recovery = 1; |
12250 | 0 | } |
12251 | |
|
12252 | 0 | input = xmlNewStringInputStream(ctxt, string); |
12253 | 0 | if (input == NULL) { |
12254 | 0 | ret = ctxt->errNo; |
12255 | 0 | goto error; |
12256 | 0 | } |
12257 | | |
12258 | 0 | list = xmlCtxtParseContentInternal(ctxt, input, /* hasTextDecl */ 0, 1); |
12259 | 0 | if (listOut != NULL) |
12260 | 0 | *listOut = list; |
12261 | 0 | else |
12262 | 0 | xmlFreeNodeList(list); |
12263 | |
|
12264 | 0 | if (!ctxt->wellFormed) |
12265 | 0 | ret = ctxt->errNo; |
12266 | 0 | else |
12267 | 0 | ret = XML_ERR_OK; |
12268 | |
|
12269 | 0 | error: |
12270 | 0 | xmlFreeInputStream(input); |
12271 | 0 | xmlFreeParserCtxt(ctxt); |
12272 | 0 | return(ret); |
12273 | 0 | } |
12274 | | |
12275 | | /** |
12276 | | * parse an XML external entity out of context and build a tree. |
12277 | | * It use the given SAX function block to handle the parsing callback. |
12278 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12279 | | * |
12280 | | * @deprecated Don't use. |
12281 | | * |
12282 | | * [78] extParsedEnt ::= TextDecl? content |
12283 | | * |
12284 | | * This correspond to a "Well Balanced" chunk |
12285 | | * |
12286 | | * @param sax the SAX handler block |
12287 | | * @param filename the filename |
12288 | | * @returns the resulting document tree |
12289 | | */ |
12290 | | |
12291 | | xmlDoc * |
12292 | 0 | xmlSAXParseEntity(xmlSAXHandler *sax, const char *filename) { |
12293 | 0 | xmlDocPtr ret; |
12294 | 0 | xmlParserCtxtPtr ctxt; |
12295 | |
|
12296 | 0 | ctxt = xmlCreateFileParserCtxt(filename); |
12297 | 0 | if (ctxt == NULL) { |
12298 | 0 | return(NULL); |
12299 | 0 | } |
12300 | 0 | if (sax != NULL) { |
12301 | 0 | if (sax->initialized == XML_SAX2_MAGIC) { |
12302 | 0 | *ctxt->sax = *sax; |
12303 | 0 | } else { |
12304 | 0 | memset(ctxt->sax, 0, sizeof(*ctxt->sax)); |
12305 | 0 | memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1)); |
12306 | 0 | } |
12307 | 0 | ctxt->userData = NULL; |
12308 | 0 | } |
12309 | |
|
12310 | 0 | xmlParseExtParsedEnt(ctxt); |
12311 | |
|
12312 | 0 | if (ctxt->wellFormed) { |
12313 | 0 | ret = ctxt->myDoc; |
12314 | 0 | } else { |
12315 | 0 | ret = NULL; |
12316 | 0 | xmlFreeDoc(ctxt->myDoc); |
12317 | 0 | } |
12318 | |
|
12319 | 0 | xmlFreeParserCtxt(ctxt); |
12320 | |
|
12321 | 0 | return(ret); |
12322 | 0 | } |
12323 | | |
12324 | | /** |
12325 | | * parse an XML external entity out of context and build a tree. |
12326 | | * |
12327 | | * [78] extParsedEnt ::= TextDecl? content |
12328 | | * |
12329 | | * This correspond to a "Well Balanced" chunk |
12330 | | * |
12331 | | * @param filename the filename |
12332 | | * @returns the resulting document tree |
12333 | | */ |
12334 | | |
12335 | | xmlDoc * |
12336 | 0 | xmlParseEntity(const char *filename) { |
12337 | 0 | return(xmlSAXParseEntity(NULL, filename)); |
12338 | 0 | } |
12339 | | #endif /* LIBXML_SAX1_ENABLED */ |
12340 | | |
12341 | | /** |
12342 | | * Create a parser context for an external entity |
12343 | | * Automatic support for ZLIB/Compress compressed document is provided |
12344 | | * by default if found at compile-time. |
12345 | | * |
12346 | | * @deprecated Don't use. |
12347 | | * |
12348 | | * @param URL the entity URL |
12349 | | * @param ID the entity PUBLIC ID |
12350 | | * @param base a possible base for the target URI |
12351 | | * @returns the new parser context or NULL |
12352 | | */ |
12353 | | xmlParserCtxt * |
12354 | | xmlCreateEntityParserCtxt(const xmlChar *URL, const xmlChar *ID, |
12355 | 0 | const xmlChar *base) { |
12356 | 0 | xmlParserCtxtPtr ctxt; |
12357 | 0 | xmlParserInputPtr input; |
12358 | 0 | xmlChar *uri = NULL; |
12359 | |
|
12360 | 0 | ctxt = xmlNewParserCtxt(); |
12361 | 0 | if (ctxt == NULL) |
12362 | 0 | return(NULL); |
12363 | | |
12364 | 0 | if (base != NULL) { |
12365 | 0 | if (xmlBuildURISafe(URL, base, &uri) < 0) |
12366 | 0 | goto error; |
12367 | 0 | if (uri != NULL) |
12368 | 0 | URL = uri; |
12369 | 0 | } |
12370 | | |
12371 | 0 | input = xmlLoadResource(ctxt, (char *) URL, (char *) ID, |
12372 | 0 | XML_RESOURCE_UNKNOWN); |
12373 | 0 | if (input == NULL) |
12374 | 0 | goto error; |
12375 | | |
12376 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12377 | 0 | xmlFreeInputStream(input); |
12378 | 0 | goto error; |
12379 | 0 | } |
12380 | | |
12381 | 0 | xmlFree(uri); |
12382 | 0 | return(ctxt); |
12383 | | |
12384 | 0 | error: |
12385 | 0 | xmlFree(uri); |
12386 | 0 | xmlFreeParserCtxt(ctxt); |
12387 | 0 | return(NULL); |
12388 | 0 | } |
12389 | | |
12390 | | /************************************************************************ |
12391 | | * * |
12392 | | * Front ends when parsing from a file * |
12393 | | * * |
12394 | | ************************************************************************/ |
12395 | | |
12396 | | /** |
12397 | | * Create a parser context for a file or URL content. |
12398 | | * Automatic support for ZLIB/Compress compressed document is provided |
12399 | | * by default if found at compile-time and for file accesses |
12400 | | * |
12401 | | * @deprecated Use #xmlNewParserCtxt and #xmlCtxtReadFile. |
12402 | | * |
12403 | | * @param filename the filename or URL |
12404 | | * @param options a combination of xmlParserOption |
12405 | | * @returns the new parser context or NULL |
12406 | | */ |
12407 | | xmlParserCtxt * |
12408 | | xmlCreateURLParserCtxt(const char *filename, int options) |
12409 | 0 | { |
12410 | 0 | xmlParserCtxtPtr ctxt; |
12411 | 0 | xmlParserInputPtr input; |
12412 | |
|
12413 | 0 | ctxt = xmlNewParserCtxt(); |
12414 | 0 | if (ctxt == NULL) |
12415 | 0 | return(NULL); |
12416 | | |
12417 | 0 | options |= XML_PARSE_UNZIP; |
12418 | |
|
12419 | 0 | xmlCtxtUseOptions(ctxt, options); |
12420 | |
|
12421 | 0 | input = xmlLoadResource(ctxt, filename, NULL, XML_RESOURCE_MAIN_DOCUMENT); |
12422 | 0 | if (input == NULL) { |
12423 | 0 | xmlFreeParserCtxt(ctxt); |
12424 | 0 | return(NULL); |
12425 | 0 | } |
12426 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12427 | 0 | xmlFreeInputStream(input); |
12428 | 0 | xmlFreeParserCtxt(ctxt); |
12429 | 0 | return(NULL); |
12430 | 0 | } |
12431 | | |
12432 | 0 | return(ctxt); |
12433 | 0 | } |
12434 | | |
12435 | | /** |
12436 | | * Create a parser context for a file content. |
12437 | | * Automatic support for ZLIB/Compress compressed document is provided |
12438 | | * by default if found at compile-time. |
12439 | | * |
12440 | | * @deprecated Use #xmlNewParserCtxt and #xmlCtxtReadFile. |
12441 | | * |
12442 | | * @param filename the filename |
12443 | | * @returns the new parser context or NULL |
12444 | | */ |
12445 | | xmlParserCtxt * |
12446 | | xmlCreateFileParserCtxt(const char *filename) |
12447 | 0 | { |
12448 | 0 | return(xmlCreateURLParserCtxt(filename, 0)); |
12449 | 0 | } |
12450 | | |
12451 | | #ifdef LIBXML_SAX1_ENABLED |
12452 | | /** |
12453 | | * parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12454 | | * compressed document is provided by default if found at compile-time. |
12455 | | * It use the given SAX function block to handle the parsing callback. |
12456 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12457 | | * |
12458 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadFile. |
12459 | | * |
12460 | | * User data (void *) is stored within the parser context in the |
12461 | | * context's _private member, so it is available nearly everywhere in libxml |
12462 | | * |
12463 | | * @param sax the SAX handler block |
12464 | | * @param filename the filename |
12465 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12466 | | * documents |
12467 | | * @param data the userdata |
12468 | | * @returns the resulting document tree |
12469 | | */ |
12470 | | |
12471 | | xmlDoc * |
12472 | | xmlSAXParseFileWithData(xmlSAXHandler *sax, const char *filename, |
12473 | 0 | int recovery, void *data) { |
12474 | 0 | xmlDocPtr ret = NULL; |
12475 | 0 | xmlParserCtxtPtr ctxt; |
12476 | 0 | xmlParserInputPtr input; |
12477 | |
|
12478 | 0 | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
12479 | 0 | if (ctxt == NULL) |
12480 | 0 | return(NULL); |
12481 | | |
12482 | 0 | if (data != NULL) |
12483 | 0 | ctxt->_private = data; |
12484 | |
|
12485 | 0 | if (recovery) { |
12486 | 0 | ctxt->options |= XML_PARSE_RECOVER; |
12487 | 0 | ctxt->recovery = 1; |
12488 | 0 | } |
12489 | |
|
12490 | 0 | if ((filename != NULL) && (filename[0] == '-') && (filename[1] == 0)) |
12491 | 0 | input = xmlCtxtNewInputFromFd(ctxt, filename, STDIN_FILENO, NULL, 0); |
12492 | 0 | else |
12493 | 0 | input = xmlCtxtNewInputFromUrl(ctxt, filename, NULL, NULL, 0); |
12494 | |
|
12495 | 0 | if (input != NULL) |
12496 | 0 | ret = xmlCtxtParseDocument(ctxt, input); |
12497 | |
|
12498 | 0 | xmlFreeParserCtxt(ctxt); |
12499 | 0 | return(ret); |
12500 | 0 | } |
12501 | | |
12502 | | /** |
12503 | | * parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12504 | | * compressed document is provided by default if found at compile-time. |
12505 | | * It use the given SAX function block to handle the parsing callback. |
12506 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12507 | | * |
12508 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadFile. |
12509 | | * |
12510 | | * @param sax the SAX handler block |
12511 | | * @param filename the filename |
12512 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12513 | | * documents |
12514 | | * @returns the resulting document tree |
12515 | | */ |
12516 | | |
12517 | | xmlDoc * |
12518 | | xmlSAXParseFile(xmlSAXHandler *sax, const char *filename, |
12519 | 0 | int recovery) { |
12520 | 0 | return(xmlSAXParseFileWithData(sax,filename,recovery,NULL)); |
12521 | 0 | } |
12522 | | |
12523 | | /** |
12524 | | * parse an XML in-memory document and build a tree. |
12525 | | * In the case the document is not Well Formed, a attempt to build a |
12526 | | * tree is tried anyway |
12527 | | * |
12528 | | * @deprecated Use #xmlReadDoc with XML_PARSE_RECOVER. |
12529 | | * |
12530 | | * @param cur a pointer to an array of xmlChar |
12531 | | * @returns the resulting document tree or NULL in case of failure |
12532 | | */ |
12533 | | |
12534 | | xmlDoc * |
12535 | 0 | xmlRecoverDoc(const xmlChar *cur) { |
12536 | 0 | return(xmlSAXParseDoc(NULL, cur, 1)); |
12537 | 0 | } |
12538 | | |
12539 | | /** |
12540 | | * parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12541 | | * compressed document is provided by default if found at compile-time. |
12542 | | * |
12543 | | * @deprecated Use #xmlReadFile. |
12544 | | * |
12545 | | * @param filename the filename |
12546 | | * @returns the resulting document tree if the file was wellformed, |
12547 | | * NULL otherwise. |
12548 | | */ |
12549 | | |
12550 | | xmlDoc * |
12551 | 0 | xmlParseFile(const char *filename) { |
12552 | 0 | return(xmlSAXParseFile(NULL, filename, 0)); |
12553 | 0 | } |
12554 | | |
12555 | | /** |
12556 | | * parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12557 | | * compressed document is provided by default if found at compile-time. |
12558 | | * In the case the document is not Well Formed, it attempts to build |
12559 | | * a tree anyway |
12560 | | * |
12561 | | * @deprecated Use #xmlReadFile with XML_PARSE_RECOVER. |
12562 | | * |
12563 | | * @param filename the filename |
12564 | | * @returns the resulting document tree or NULL in case of failure |
12565 | | */ |
12566 | | |
12567 | | xmlDoc * |
12568 | 0 | xmlRecoverFile(const char *filename) { |
12569 | 0 | return(xmlSAXParseFile(NULL, filename, 1)); |
12570 | 0 | } |
12571 | | |
12572 | | |
12573 | | /** |
12574 | | * Setup the parser context to parse a new buffer; Clears any prior |
12575 | | * contents from the parser context. The buffer parameter must not be |
12576 | | * NULL, but the filename parameter can be |
12577 | | * |
12578 | | * @deprecated Don't use. |
12579 | | * |
12580 | | * @param ctxt an XML parser context |
12581 | | * @param buffer a xmlChar * buffer |
12582 | | * @param filename a file name |
12583 | | */ |
12584 | | void |
12585 | | xmlSetupParserForBuffer(xmlParserCtxt *ctxt, const xmlChar* buffer, |
12586 | | const char* filename) |
12587 | 0 | { |
12588 | 0 | xmlParserInputPtr input; |
12589 | |
|
12590 | 0 | if ((ctxt == NULL) || (buffer == NULL)) |
12591 | 0 | return; |
12592 | | |
12593 | 0 | xmlCtxtReset(ctxt); |
12594 | |
|
12595 | 0 | input = xmlCtxtNewInputFromString(ctxt, filename, (const char *) buffer, |
12596 | 0 | NULL, 0); |
12597 | 0 | if (input == NULL) |
12598 | 0 | return; |
12599 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) |
12600 | 0 | xmlFreeInputStream(input); |
12601 | 0 | } |
12602 | | |
12603 | | /** |
12604 | | * parse an XML file and call the given SAX handler routines. |
12605 | | * Automatic support for ZLIB/Compress compressed document is provided |
12606 | | * |
12607 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadFile. |
12608 | | * |
12609 | | * @param sax a SAX handler |
12610 | | * @param user_data The user data returned on SAX callbacks |
12611 | | * @param filename a file name |
12612 | | * @returns 0 in case of success or a error number otherwise |
12613 | | */ |
12614 | | int |
12615 | | xmlSAXUserParseFile(xmlSAXHandler *sax, void *user_data, |
12616 | 0 | const char *filename) { |
12617 | 0 | int ret = 0; |
12618 | 0 | xmlParserCtxtPtr ctxt; |
12619 | |
|
12620 | 0 | ctxt = xmlCreateFileParserCtxt(filename); |
12621 | 0 | if (ctxt == NULL) return -1; |
12622 | 0 | if (sax != NULL) { |
12623 | 0 | if (sax->initialized == XML_SAX2_MAGIC) { |
12624 | 0 | *ctxt->sax = *sax; |
12625 | 0 | } else { |
12626 | 0 | memset(ctxt->sax, 0, sizeof(*ctxt->sax)); |
12627 | 0 | memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1)); |
12628 | 0 | } |
12629 | 0 | ctxt->userData = user_data; |
12630 | 0 | } |
12631 | |
|
12632 | 0 | xmlParseDocument(ctxt); |
12633 | |
|
12634 | 0 | if (ctxt->wellFormed) |
12635 | 0 | ret = 0; |
12636 | 0 | else { |
12637 | 0 | if (ctxt->errNo != 0) |
12638 | 0 | ret = ctxt->errNo; |
12639 | 0 | else |
12640 | 0 | ret = -1; |
12641 | 0 | } |
12642 | 0 | if (ctxt->myDoc != NULL) { |
12643 | 0 | xmlFreeDoc(ctxt->myDoc); |
12644 | 0 | ctxt->myDoc = NULL; |
12645 | 0 | } |
12646 | 0 | xmlFreeParserCtxt(ctxt); |
12647 | |
|
12648 | 0 | return ret; |
12649 | 0 | } |
12650 | | #endif /* LIBXML_SAX1_ENABLED */ |
12651 | | |
12652 | | /************************************************************************ |
12653 | | * * |
12654 | | * Front ends when parsing from memory * |
12655 | | * * |
12656 | | ************************************************************************/ |
12657 | | |
12658 | | /** |
12659 | | * Create a parser context for an XML in-memory document. The input buffer |
12660 | | * must not contain a terminating null byte. |
12661 | | * |
12662 | | * @param buffer a pointer to a char array |
12663 | | * @param size the size of the array |
12664 | | * @returns the new parser context or NULL |
12665 | | */ |
12666 | | xmlParserCtxt * |
12667 | 14.5k | xmlCreateMemoryParserCtxt(const char *buffer, int size) { |
12668 | 14.5k | xmlParserCtxtPtr ctxt; |
12669 | 14.5k | xmlParserInputPtr input; |
12670 | | |
12671 | 14.5k | if (size < 0) |
12672 | 0 | return(NULL); |
12673 | | |
12674 | 14.5k | ctxt = xmlNewParserCtxt(); |
12675 | 14.5k | if (ctxt == NULL) |
12676 | 0 | return(NULL); |
12677 | | |
12678 | 14.5k | input = xmlCtxtNewInputFromMemory(ctxt, NULL, buffer, size, NULL, 0); |
12679 | 14.5k | if (input == NULL) { |
12680 | 0 | xmlFreeParserCtxt(ctxt); |
12681 | 0 | return(NULL); |
12682 | 0 | } |
12683 | 14.5k | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12684 | 0 | xmlFreeInputStream(input); |
12685 | 0 | xmlFreeParserCtxt(ctxt); |
12686 | 0 | return(NULL); |
12687 | 0 | } |
12688 | | |
12689 | 14.5k | return(ctxt); |
12690 | 14.5k | } |
12691 | | |
12692 | | #ifdef LIBXML_SAX1_ENABLED |
12693 | | /** |
12694 | | * parse an XML in-memory block and use the given SAX function block |
12695 | | * to handle the parsing callback. If sax is NULL, fallback to the default |
12696 | | * DOM tree building routines. |
12697 | | * |
12698 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadMemory. |
12699 | | * |
12700 | | * User data (void *) is stored within the parser context in the |
12701 | | * context's _private member, so it is available nearly everywhere in libxml |
12702 | | * |
12703 | | * @param sax the SAX handler block |
12704 | | * @param buffer an pointer to a char array |
12705 | | * @param size the size of the array |
12706 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12707 | | * documents |
12708 | | * @param data the userdata |
12709 | | * @returns the resulting document tree |
12710 | | */ |
12711 | | |
12712 | | xmlDoc * |
12713 | | xmlSAXParseMemoryWithData(xmlSAXHandler *sax, const char *buffer, |
12714 | 0 | int size, int recovery, void *data) { |
12715 | 0 | xmlDocPtr ret = NULL; |
12716 | 0 | xmlParserCtxtPtr ctxt; |
12717 | 0 | xmlParserInputPtr input; |
12718 | |
|
12719 | 0 | if (size < 0) |
12720 | 0 | return(NULL); |
12721 | | |
12722 | 0 | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
12723 | 0 | if (ctxt == NULL) |
12724 | 0 | return(NULL); |
12725 | | |
12726 | 0 | if (data != NULL) |
12727 | 0 | ctxt->_private=data; |
12728 | |
|
12729 | 0 | if (recovery) { |
12730 | 0 | ctxt->options |= XML_PARSE_RECOVER; |
12731 | 0 | ctxt->recovery = 1; |
12732 | 0 | } |
12733 | |
|
12734 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, NULL, buffer, size, NULL, |
12735 | 0 | XML_INPUT_BUF_STATIC); |
12736 | |
|
12737 | 0 | if (input != NULL) |
12738 | 0 | ret = xmlCtxtParseDocument(ctxt, input); |
12739 | |
|
12740 | 0 | xmlFreeParserCtxt(ctxt); |
12741 | 0 | return(ret); |
12742 | 0 | } |
12743 | | |
12744 | | /** |
12745 | | * parse an XML in-memory block and use the given SAX function block |
12746 | | * to handle the parsing callback. If sax is NULL, fallback to the default |
12747 | | * DOM tree building routines. |
12748 | | * |
12749 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadMemory. |
12750 | | * |
12751 | | * @param sax the SAX handler block |
12752 | | * @param buffer an pointer to a char array |
12753 | | * @param size the size of the array |
12754 | | * @param recovery work in recovery mode, i.e. tries to read not Well Formed |
12755 | | * documents |
12756 | | * @returns the resulting document tree |
12757 | | */ |
12758 | | xmlDoc * |
12759 | | xmlSAXParseMemory(xmlSAXHandler *sax, const char *buffer, |
12760 | 0 | int size, int recovery) { |
12761 | 0 | return xmlSAXParseMemoryWithData(sax, buffer, size, recovery, NULL); |
12762 | 0 | } |
12763 | | |
12764 | | /** |
12765 | | * parse an XML in-memory block and build a tree. |
12766 | | * |
12767 | | * @deprecated Use #xmlReadMemory. |
12768 | | * |
12769 | | * @param buffer an pointer to a char array |
12770 | | * @param size the size of the array |
12771 | | * @returns the resulting document tree |
12772 | | */ |
12773 | | |
12774 | 0 | xmlDoc *xmlParseMemory(const char *buffer, int size) { |
12775 | 0 | return(xmlSAXParseMemory(NULL, buffer, size, 0)); |
12776 | 0 | } |
12777 | | |
12778 | | /** |
12779 | | * parse an XML in-memory block and build a tree. |
12780 | | * In the case the document is not Well Formed, an attempt to |
12781 | | * build a tree is tried anyway |
12782 | | * |
12783 | | * @deprecated Use #xmlReadMemory with XML_PARSE_RECOVER. |
12784 | | * |
12785 | | * @param buffer an pointer to a char array |
12786 | | * @param size the size of the array |
12787 | | * @returns the resulting document tree or NULL in case of error |
12788 | | */ |
12789 | | |
12790 | 0 | xmlDoc *xmlRecoverMemory(const char *buffer, int size) { |
12791 | 0 | return(xmlSAXParseMemory(NULL, buffer, size, 1)); |
12792 | 0 | } |
12793 | | |
12794 | | /** |
12795 | | * parse an XML in-memory buffer and call the given SAX handler routines. |
12796 | | * |
12797 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadMemory. |
12798 | | * |
12799 | | * @param sax a SAX handler |
12800 | | * @param user_data The user data returned on SAX callbacks |
12801 | | * @param buffer an in-memory XML document input |
12802 | | * @param size the length of the XML document in bytes |
12803 | | * @returns 0 in case of success or a error number otherwise |
12804 | | */ |
12805 | | int xmlSAXUserParseMemory(xmlSAXHandler *sax, void *user_data, |
12806 | 0 | const char *buffer, int size) { |
12807 | 0 | int ret = 0; |
12808 | 0 | xmlParserCtxtPtr ctxt; |
12809 | |
|
12810 | 0 | ctxt = xmlCreateMemoryParserCtxt(buffer, size); |
12811 | 0 | if (ctxt == NULL) return -1; |
12812 | 0 | if (sax != NULL) { |
12813 | 0 | if (sax->initialized == XML_SAX2_MAGIC) { |
12814 | 0 | *ctxt->sax = *sax; |
12815 | 0 | } else { |
12816 | 0 | memset(ctxt->sax, 0, sizeof(*ctxt->sax)); |
12817 | 0 | memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1)); |
12818 | 0 | } |
12819 | 0 | ctxt->userData = user_data; |
12820 | 0 | } |
12821 | |
|
12822 | 0 | xmlParseDocument(ctxt); |
12823 | |
|
12824 | 0 | if (ctxt->wellFormed) |
12825 | 0 | ret = 0; |
12826 | 0 | else { |
12827 | 0 | if (ctxt->errNo != 0) |
12828 | 0 | ret = ctxt->errNo; |
12829 | 0 | else |
12830 | 0 | ret = -1; |
12831 | 0 | } |
12832 | 0 | if (ctxt->myDoc != NULL) { |
12833 | 0 | xmlFreeDoc(ctxt->myDoc); |
12834 | 0 | ctxt->myDoc = NULL; |
12835 | 0 | } |
12836 | 0 | xmlFreeParserCtxt(ctxt); |
12837 | |
|
12838 | 0 | return ret; |
12839 | 0 | } |
12840 | | #endif /* LIBXML_SAX1_ENABLED */ |
12841 | | |
12842 | | /** |
12843 | | * Creates a parser context for an XML in-memory document. |
12844 | | * |
12845 | | * @param str a pointer to an array of xmlChar |
12846 | | * @returns the new parser context or NULL |
12847 | | */ |
12848 | | xmlParserCtxt * |
12849 | 0 | xmlCreateDocParserCtxt(const xmlChar *str) { |
12850 | 0 | xmlParserCtxtPtr ctxt; |
12851 | 0 | xmlParserInputPtr input; |
12852 | |
|
12853 | 0 | ctxt = xmlNewParserCtxt(); |
12854 | 0 | if (ctxt == NULL) |
12855 | 0 | return(NULL); |
12856 | | |
12857 | 0 | input = xmlCtxtNewInputFromString(ctxt, NULL, (const char *) str, NULL, 0); |
12858 | 0 | if (input == NULL) { |
12859 | 0 | xmlFreeParserCtxt(ctxt); |
12860 | 0 | return(NULL); |
12861 | 0 | } |
12862 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12863 | 0 | xmlFreeInputStream(input); |
12864 | 0 | xmlFreeParserCtxt(ctxt); |
12865 | 0 | return(NULL); |
12866 | 0 | } |
12867 | | |
12868 | 0 | return(ctxt); |
12869 | 0 | } |
12870 | | |
12871 | | #ifdef LIBXML_SAX1_ENABLED |
12872 | | /** |
12873 | | * parse an XML in-memory document and build a tree. |
12874 | | * It use the given SAX function block to handle the parsing callback. |
12875 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12876 | | * |
12877 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadDoc. |
12878 | | * |
12879 | | * @param sax the SAX handler block |
12880 | | * @param cur a pointer to an array of xmlChar |
12881 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12882 | | * documents |
12883 | | * @returns the resulting document tree |
12884 | | */ |
12885 | | |
12886 | | xmlDoc * |
12887 | 0 | xmlSAXParseDoc(xmlSAXHandler *sax, const xmlChar *cur, int recovery) { |
12888 | 0 | xmlDocPtr ret; |
12889 | 0 | xmlParserCtxtPtr ctxt; |
12890 | 0 | xmlSAXHandlerPtr oldsax = NULL; |
12891 | |
|
12892 | 0 | if (cur == NULL) return(NULL); |
12893 | | |
12894 | | |
12895 | 0 | ctxt = xmlCreateDocParserCtxt(cur); |
12896 | 0 | if (ctxt == NULL) return(NULL); |
12897 | 0 | if (sax != NULL) { |
12898 | 0 | oldsax = ctxt->sax; |
12899 | 0 | ctxt->sax = sax; |
12900 | 0 | ctxt->userData = NULL; |
12901 | 0 | } |
12902 | |
|
12903 | 0 | xmlParseDocument(ctxt); |
12904 | 0 | if ((ctxt->wellFormed) || recovery) ret = ctxt->myDoc; |
12905 | 0 | else { |
12906 | 0 | ret = NULL; |
12907 | 0 | xmlFreeDoc(ctxt->myDoc); |
12908 | 0 | ctxt->myDoc = NULL; |
12909 | 0 | } |
12910 | 0 | if (sax != NULL) |
12911 | 0 | ctxt->sax = oldsax; |
12912 | 0 | xmlFreeParserCtxt(ctxt); |
12913 | |
|
12914 | 0 | return(ret); |
12915 | 0 | } |
12916 | | |
12917 | | /** |
12918 | | * parse an XML in-memory document and build a tree. |
12919 | | * |
12920 | | * @deprecated Use #xmlReadDoc. |
12921 | | * |
12922 | | * @param cur a pointer to an array of xmlChar |
12923 | | * @returns the resulting document tree |
12924 | | */ |
12925 | | |
12926 | | xmlDoc * |
12927 | 0 | xmlParseDoc(const xmlChar *cur) { |
12928 | 0 | return(xmlSAXParseDoc(NULL, cur, 0)); |
12929 | 0 | } |
12930 | | #endif /* LIBXML_SAX1_ENABLED */ |
12931 | | |
12932 | | /************************************************************************ |
12933 | | * * |
12934 | | * New set (2.6.0) of simpler and more flexible APIs * |
12935 | | * * |
12936 | | ************************************************************************/ |
12937 | | |
12938 | | /** |
12939 | | * Free a string if it is not owned by the "dict" dictionary in the |
12940 | | * current scope |
12941 | | * |
12942 | | * @param str a string |
12943 | | */ |
12944 | | #define DICT_FREE(str) \ |
12945 | 0 | if ((str) && ((!dict) || \ |
12946 | 0 | (xmlDictOwns(dict, (const xmlChar *)(str)) == 0))) \ |
12947 | 0 | xmlFree((char *)(str)); |
12948 | | |
12949 | | /** |
12950 | | * Reset a parser context |
12951 | | * |
12952 | | * @param ctxt an XML parser context |
12953 | | */ |
12954 | | void |
12955 | | xmlCtxtReset(xmlParserCtxt *ctxt) |
12956 | 0 | { |
12957 | 0 | xmlParserInputPtr input; |
12958 | 0 | xmlDictPtr dict; |
12959 | |
|
12960 | 0 | if (ctxt == NULL) |
12961 | 0 | return; |
12962 | | |
12963 | 0 | dict = ctxt->dict; |
12964 | |
|
12965 | 0 | while ((input = xmlCtxtPopInput(ctxt)) != NULL) { /* Non consuming */ |
12966 | 0 | xmlFreeInputStream(input); |
12967 | 0 | } |
12968 | 0 | ctxt->inputNr = 0; |
12969 | 0 | ctxt->input = NULL; |
12970 | |
|
12971 | 0 | ctxt->spaceNr = 0; |
12972 | 0 | if (ctxt->spaceTab != NULL) { |
12973 | 0 | ctxt->spaceTab[0] = -1; |
12974 | 0 | ctxt->space = &ctxt->spaceTab[0]; |
12975 | 0 | } else { |
12976 | 0 | ctxt->space = NULL; |
12977 | 0 | } |
12978 | | |
12979 | |
|
12980 | 0 | ctxt->nodeNr = 0; |
12981 | 0 | ctxt->node = NULL; |
12982 | |
|
12983 | 0 | ctxt->nameNr = 0; |
12984 | 0 | ctxt->name = NULL; |
12985 | |
|
12986 | 0 | ctxt->nsNr = 0; |
12987 | 0 | xmlParserNsReset(ctxt->nsdb); |
12988 | |
|
12989 | 0 | DICT_FREE(ctxt->version); |
12990 | 0 | ctxt->version = NULL; |
12991 | 0 | DICT_FREE(ctxt->encoding); |
12992 | 0 | ctxt->encoding = NULL; |
12993 | 0 | DICT_FREE(ctxt->extSubURI); |
12994 | 0 | ctxt->extSubURI = NULL; |
12995 | 0 | DICT_FREE(ctxt->extSubSystem); |
12996 | 0 | ctxt->extSubSystem = NULL; |
12997 | |
|
12998 | 0 | if (ctxt->directory != NULL) { |
12999 | 0 | xmlFree(ctxt->directory); |
13000 | 0 | ctxt->directory = NULL; |
13001 | 0 | } |
13002 | |
|
13003 | 0 | if (ctxt->myDoc != NULL) |
13004 | 0 | xmlFreeDoc(ctxt->myDoc); |
13005 | 0 | ctxt->myDoc = NULL; |
13006 | |
|
13007 | 0 | ctxt->standalone = -1; |
13008 | 0 | ctxt->hasExternalSubset = 0; |
13009 | 0 | ctxt->hasPErefs = 0; |
13010 | 0 | ctxt->html = ctxt->html ? 1 : 0; |
13011 | 0 | ctxt->instate = XML_PARSER_START; |
13012 | |
|
13013 | 0 | ctxt->wellFormed = 1; |
13014 | 0 | ctxt->nsWellFormed = 1; |
13015 | 0 | ctxt->disableSAX = 0; |
13016 | 0 | ctxt->valid = 1; |
13017 | 0 | ctxt->record_info = 0; |
13018 | 0 | ctxt->checkIndex = 0; |
13019 | 0 | ctxt->endCheckState = 0; |
13020 | 0 | ctxt->inSubset = 0; |
13021 | 0 | ctxt->errNo = XML_ERR_OK; |
13022 | 0 | ctxt->depth = 0; |
13023 | 0 | ctxt->catalogs = NULL; |
13024 | 0 | ctxt->sizeentities = 0; |
13025 | 0 | ctxt->sizeentcopy = 0; |
13026 | 0 | xmlInitNodeInfoSeq(&ctxt->node_seq); |
13027 | |
|
13028 | 0 | if (ctxt->attsDefault != NULL) { |
13029 | 0 | xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator); |
13030 | 0 | ctxt->attsDefault = NULL; |
13031 | 0 | } |
13032 | 0 | if (ctxt->attsSpecial != NULL) { |
13033 | 0 | xmlHashFree(ctxt->attsSpecial, NULL); |
13034 | 0 | ctxt->attsSpecial = NULL; |
13035 | 0 | } |
13036 | |
|
13037 | 0 | #ifdef LIBXML_CATALOG_ENABLED |
13038 | 0 | if (ctxt->catalogs != NULL) |
13039 | 0 | xmlCatalogFreeLocal(ctxt->catalogs); |
13040 | 0 | #endif |
13041 | 0 | ctxt->nbErrors = 0; |
13042 | 0 | ctxt->nbWarnings = 0; |
13043 | 0 | if (ctxt->lastError.code != XML_ERR_OK) |
13044 | 0 | xmlResetError(&ctxt->lastError); |
13045 | 0 | } |
13046 | | |
13047 | | /** |
13048 | | * Reset a push parser context |
13049 | | * |
13050 | | * @param ctxt an XML parser context |
13051 | | * @param chunk a pointer to an array of chars |
13052 | | * @param size number of chars in the array |
13053 | | * @param filename an optional file name or URI |
13054 | | * @param encoding the document encoding, or NULL |
13055 | | * @returns 0 in case of success and 1 in case of error |
13056 | | */ |
13057 | | int |
13058 | | xmlCtxtResetPush(xmlParserCtxt *ctxt, const char *chunk, |
13059 | | int size, const char *filename, const char *encoding) |
13060 | 0 | { |
13061 | 0 | xmlParserInputPtr input; |
13062 | |
|
13063 | 0 | if (ctxt == NULL) |
13064 | 0 | return(1); |
13065 | | |
13066 | 0 | xmlCtxtReset(ctxt); |
13067 | |
|
13068 | 0 | input = xmlNewPushInput(filename, chunk, size); |
13069 | 0 | if (input == NULL) |
13070 | 0 | return(1); |
13071 | | |
13072 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
13073 | 0 | xmlFreeInputStream(input); |
13074 | 0 | return(1); |
13075 | 0 | } |
13076 | | |
13077 | 0 | if (encoding != NULL) |
13078 | 0 | xmlSwitchEncodingName(ctxt, encoding); |
13079 | |
|
13080 | 0 | return(0); |
13081 | 0 | } |
13082 | | |
13083 | | static int |
13084 | | xmlCtxtSetOptionsInternal(xmlParserCtxtPtr ctxt, int options, int keepMask) |
13085 | 29.0k | { |
13086 | 29.0k | int allMask; |
13087 | | |
13088 | 29.0k | if (ctxt == NULL) |
13089 | 0 | return(-1); |
13090 | | |
13091 | | /* |
13092 | | * XInclude options aren't handled by the parser. |
13093 | | * |
13094 | | * XML_PARSE_XINCLUDE |
13095 | | * XML_PARSE_NOXINCNODE |
13096 | | * XML_PARSE_NOBASEFIX |
13097 | | */ |
13098 | 29.0k | allMask = XML_PARSE_RECOVER | |
13099 | 29.0k | XML_PARSE_NOENT | |
13100 | 29.0k | XML_PARSE_DTDLOAD | |
13101 | 29.0k | XML_PARSE_DTDATTR | |
13102 | 29.0k | XML_PARSE_DTDVALID | |
13103 | 29.0k | XML_PARSE_NOERROR | |
13104 | 29.0k | XML_PARSE_NOWARNING | |
13105 | 29.0k | XML_PARSE_PEDANTIC | |
13106 | 29.0k | XML_PARSE_NOBLANKS | |
13107 | 29.0k | #ifdef LIBXML_SAX1_ENABLED |
13108 | 29.0k | XML_PARSE_SAX1 | |
13109 | 29.0k | #endif |
13110 | 29.0k | XML_PARSE_NONET | |
13111 | 29.0k | XML_PARSE_NODICT | |
13112 | 29.0k | XML_PARSE_NSCLEAN | |
13113 | 29.0k | XML_PARSE_NOCDATA | |
13114 | 29.0k | XML_PARSE_COMPACT | |
13115 | 29.0k | XML_PARSE_OLD10 | |
13116 | 29.0k | XML_PARSE_HUGE | |
13117 | 29.0k | XML_PARSE_OLDSAX | |
13118 | 29.0k | XML_PARSE_IGNORE_ENC | |
13119 | 29.0k | XML_PARSE_BIG_LINES | |
13120 | 29.0k | XML_PARSE_NO_XXE | |
13121 | 29.0k | XML_PARSE_UNZIP | |
13122 | 29.0k | XML_PARSE_NO_SYS_CATALOG | |
13123 | 29.0k | XML_PARSE_CATALOG_PI; |
13124 | | |
13125 | 29.0k | ctxt->options = (ctxt->options & keepMask) | (options & allMask); |
13126 | | |
13127 | | /* |
13128 | | * For some options, struct members are historically the source |
13129 | | * of truth. The values are initalized from global variables and |
13130 | | * old code could also modify them directly. Several older API |
13131 | | * functions that don't take an options argument rely on these |
13132 | | * deprecated mechanisms. |
13133 | | * |
13134 | | * Once public access to struct members and the globals are |
13135 | | * disabled, we can use the options bitmask as source of |
13136 | | * truth, making all these struct members obsolete. |
13137 | | * |
13138 | | * The XML_DETECT_IDS flags is misnamed. It simply enables |
13139 | | * loading of the external subset. |
13140 | | */ |
13141 | 29.0k | ctxt->recovery = (options & XML_PARSE_RECOVER) ? 1 : 0; |
13142 | 29.0k | ctxt->replaceEntities = (options & XML_PARSE_NOENT) ? 1 : 0; |
13143 | 29.0k | ctxt->loadsubset = (options & XML_PARSE_DTDLOAD) ? XML_DETECT_IDS : 0; |
13144 | 29.0k | ctxt->loadsubset |= (options & XML_PARSE_DTDATTR) ? XML_COMPLETE_ATTRS : 0; |
13145 | 29.0k | ctxt->validate = (options & XML_PARSE_DTDVALID) ? 1 : 0; |
13146 | 29.0k | ctxt->pedantic = (options & XML_PARSE_PEDANTIC) ? 1 : 0; |
13147 | 29.0k | ctxt->keepBlanks = (options & XML_PARSE_NOBLANKS) ? 0 : 1; |
13148 | 29.0k | ctxt->dictNames = (options & XML_PARSE_NODICT) ? 0 : 1; |
13149 | | |
13150 | 29.0k | if (options & XML_PARSE_HUGE) { |
13151 | 29.0k | if (ctxt->dict != NULL) |
13152 | 29.0k | xmlDictSetLimit(ctxt->dict, 0); |
13153 | 29.0k | } |
13154 | | |
13155 | 29.0k | return(options & ~allMask); |
13156 | 29.0k | } |
13157 | | |
13158 | | /** |
13159 | | * Applies the options to the parser context. Unset options are |
13160 | | * cleared. |
13161 | | * |
13162 | | * @since 2.13.0 |
13163 | | * |
13164 | | * With older versions, you can use #xmlCtxtUseOptions. |
13165 | | * |
13166 | | * @param ctxt an XML parser context |
13167 | | * @param options a bitmask of xmlParserOption values |
13168 | | * @returns 0 in case of success, the set of unknown or unimplemented options |
13169 | | * in case of error. |
13170 | | */ |
13171 | | int |
13172 | | xmlCtxtSetOptions(xmlParserCtxt *ctxt, int options) |
13173 | 14.5k | { |
13174 | 14.5k | #ifdef LIBXML_HTML_ENABLED |
13175 | 14.5k | if ((ctxt != NULL) && (ctxt->html)) |
13176 | 0 | return(htmlCtxtSetOptions(ctxt, options)); |
13177 | 14.5k | #endif |
13178 | | |
13179 | 14.5k | return(xmlCtxtSetOptionsInternal(ctxt, options, 0)); |
13180 | 14.5k | } |
13181 | | |
13182 | | /** |
13183 | | * Get the current options of the parser context. |
13184 | | * |
13185 | | * @since 2.14.0 |
13186 | | * |
13187 | | * @param ctxt an XML parser context |
13188 | | * @returns the current options set in the parser context, or -1 if ctxt is NULL. |
13189 | | */ |
13190 | | int |
13191 | | xmlCtxtGetOptions(xmlParserCtxt *ctxt) |
13192 | 0 | { |
13193 | 0 | if (ctxt == NULL) |
13194 | 0 | return(-1); |
13195 | | |
13196 | 0 | return(ctxt->options); |
13197 | 0 | } |
13198 | | |
13199 | | /** |
13200 | | * Applies the options to the parser context. The following options |
13201 | | * are never cleared and can only be enabled: |
13202 | | * |
13203 | | * @deprecated Use #xmlCtxtSetOptions. |
13204 | | * |
13205 | | * - XML_PARSE_NOERROR |
13206 | | * - XML_PARSE_NOWARNING |
13207 | | * - XML_PARSE_NONET |
13208 | | * - XML_PARSE_NSCLEAN |
13209 | | * - XML_PARSE_NOCDATA |
13210 | | * - XML_PARSE_COMPACT |
13211 | | * - XML_PARSE_OLD10 |
13212 | | * - XML_PARSE_HUGE |
13213 | | * - XML_PARSE_OLDSA- X |
13214 | | * - XML_PARSE_IGNORE_ENC |
13215 | | * - XML_PARSE_BIG_LINES |
13216 | | * |
13217 | | * @param ctxt an XML parser context |
13218 | | * @param options a combination of xmlParserOption |
13219 | | * @returns 0 in case of success, the set of unknown or unimplemented options |
13220 | | * in case of error. |
13221 | | */ |
13222 | | int |
13223 | | xmlCtxtUseOptions(xmlParserCtxt *ctxt, int options) |
13224 | 14.5k | { |
13225 | 14.5k | int keepMask; |
13226 | | |
13227 | 14.5k | #ifdef LIBXML_HTML_ENABLED |
13228 | 14.5k | if ((ctxt != NULL) && (ctxt->html)) |
13229 | 0 | return(htmlCtxtUseOptions(ctxt, options)); |
13230 | 14.5k | #endif |
13231 | | |
13232 | | /* |
13233 | | * For historic reasons, some options can only be enabled. |
13234 | | */ |
13235 | 14.5k | keepMask = XML_PARSE_NOERROR | |
13236 | 14.5k | XML_PARSE_NOWARNING | |
13237 | 14.5k | XML_PARSE_NONET | |
13238 | 14.5k | XML_PARSE_NSCLEAN | |
13239 | 14.5k | XML_PARSE_NOCDATA | |
13240 | 14.5k | XML_PARSE_COMPACT | |
13241 | 14.5k | XML_PARSE_OLD10 | |
13242 | 14.5k | XML_PARSE_HUGE | |
13243 | 14.5k | XML_PARSE_OLDSAX | |
13244 | 14.5k | XML_PARSE_IGNORE_ENC | |
13245 | 14.5k | XML_PARSE_BIG_LINES; |
13246 | | |
13247 | 14.5k | return(xmlCtxtSetOptionsInternal(ctxt, options, keepMask)); |
13248 | 14.5k | } |
13249 | | |
13250 | | /** |
13251 | | * To protect against exponential entity expansion ("billion laughs"), the |
13252 | | * size of serialized output is (roughly) limited to the input size |
13253 | | * multiplied by this factor. The default value is 5. |
13254 | | * |
13255 | | * When working with documents making heavy use of entity expansion, it can |
13256 | | * be necessary to increase the value. For security reasons, this should only |
13257 | | * be considered when processing trusted input. |
13258 | | * |
13259 | | * @param ctxt an XML parser context |
13260 | | * @param maxAmpl maximum amplification factor |
13261 | | */ |
13262 | | void |
13263 | | xmlCtxtSetMaxAmplification(xmlParserCtxt *ctxt, unsigned maxAmpl) |
13264 | 0 | { |
13265 | 0 | if (ctxt == NULL) |
13266 | 0 | return; |
13267 | 0 | ctxt->maxAmpl = maxAmpl; |
13268 | 0 | } |
13269 | | |
13270 | | /** |
13271 | | * Parse an XML document and return the resulting document tree. |
13272 | | * Takes ownership of the input object. |
13273 | | * |
13274 | | * @since 2.13.0 |
13275 | | * |
13276 | | * @param ctxt an XML parser context |
13277 | | * @param input parser input |
13278 | | * @returns the resulting document tree or NULL |
13279 | | */ |
13280 | | xmlDoc * |
13281 | | xmlCtxtParseDocument(xmlParserCtxt *ctxt, xmlParserInput *input) |
13282 | 0 | { |
13283 | 0 | xmlDocPtr ret = NULL; |
13284 | |
|
13285 | 0 | if ((ctxt == NULL) || (input == NULL)) { |
13286 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
13287 | 0 | xmlFreeInputStream(input); |
13288 | 0 | return(NULL); |
13289 | 0 | } |
13290 | | |
13291 | | /* assert(ctxt->inputNr == 0); */ |
13292 | 0 | while (ctxt->inputNr > 0) |
13293 | 0 | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
13294 | |
|
13295 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
13296 | 0 | xmlFreeInputStream(input); |
13297 | 0 | return(NULL); |
13298 | 0 | } |
13299 | | |
13300 | 0 | xmlParseDocument(ctxt); |
13301 | |
|
13302 | 0 | ret = xmlCtxtGetDocument(ctxt); |
13303 | | |
13304 | | /* assert(ctxt->inputNr == 1); */ |
13305 | 0 | while (ctxt->inputNr > 0) |
13306 | 0 | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
13307 | |
|
13308 | 0 | return(ret); |
13309 | 0 | } |
13310 | | |
13311 | | /** |
13312 | | * Convenience function to parse an XML document from a |
13313 | | * zero-terminated string. |
13314 | | * |
13315 | | * See #xmlCtxtReadDoc for details. |
13316 | | * |
13317 | | * @param cur a pointer to a zero terminated string |
13318 | | * @param URL base URL (optional) |
13319 | | * @param encoding the document encoding (optional) |
13320 | | * @param options a combination of xmlParserOption |
13321 | | * @returns the resulting document tree |
13322 | | */ |
13323 | | xmlDoc * |
13324 | | xmlReadDoc(const xmlChar *cur, const char *URL, const char *encoding, |
13325 | | int options) |
13326 | 0 | { |
13327 | 0 | xmlParserCtxtPtr ctxt; |
13328 | 0 | xmlParserInputPtr input; |
13329 | 0 | xmlDocPtr doc = NULL; |
13330 | |
|
13331 | 0 | ctxt = xmlNewParserCtxt(); |
13332 | 0 | if (ctxt == NULL) |
13333 | 0 | return(NULL); |
13334 | | |
13335 | 0 | xmlCtxtUseOptions(ctxt, options); |
13336 | |
|
13337 | 0 | input = xmlCtxtNewInputFromString(ctxt, URL, (const char *) cur, encoding, |
13338 | 0 | XML_INPUT_BUF_STATIC); |
13339 | |
|
13340 | 0 | if (input != NULL) |
13341 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13342 | |
|
13343 | 0 | xmlFreeParserCtxt(ctxt); |
13344 | 0 | return(doc); |
13345 | 0 | } |
13346 | | |
13347 | | /** |
13348 | | * Convenience function to parse an XML file from the filesystem, |
13349 | | * the network or a global user-define resource loader. |
13350 | | * |
13351 | | * This function always enables the XML_PARSE_UNZIP option for |
13352 | | * backward compatibility. If a "-" filename is passed, it will |
13353 | | * read from stdin. Both of these features are potentially |
13354 | | * insecure and might be removed from later versions. |
13355 | | * |
13356 | | * See #xmlCtxtReadFile for details. |
13357 | | * |
13358 | | * @param filename a file or URL |
13359 | | * @param encoding the document encoding (optional) |
13360 | | * @param options a combination of xmlParserOption |
13361 | | * @returns the resulting document tree |
13362 | | */ |
13363 | | xmlDoc * |
13364 | | xmlReadFile(const char *filename, const char *encoding, int options) |
13365 | 0 | { |
13366 | 0 | xmlParserCtxtPtr ctxt; |
13367 | 0 | xmlParserInputPtr input; |
13368 | 0 | xmlDocPtr doc = NULL; |
13369 | |
|
13370 | 0 | ctxt = xmlNewParserCtxt(); |
13371 | 0 | if (ctxt == NULL) |
13372 | 0 | return(NULL); |
13373 | | |
13374 | 0 | options |= XML_PARSE_UNZIP; |
13375 | |
|
13376 | 0 | xmlCtxtUseOptions(ctxt, options); |
13377 | | |
13378 | | /* |
13379 | | * Backward compatibility for users of command line utilities like |
13380 | | * xmlstarlet expecting "-" to mean stdin. This is dangerous and |
13381 | | * should be removed at some point. |
13382 | | */ |
13383 | 0 | if ((filename != NULL) && (filename[0] == '-') && (filename[1] == 0)) |
13384 | 0 | input = xmlCtxtNewInputFromFd(ctxt, filename, STDIN_FILENO, |
13385 | 0 | encoding, 0); |
13386 | 0 | else |
13387 | 0 | input = xmlCtxtNewInputFromUrl(ctxt, filename, NULL, encoding, 0); |
13388 | |
|
13389 | 0 | if (input != NULL) |
13390 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13391 | |
|
13392 | 0 | xmlFreeParserCtxt(ctxt); |
13393 | 0 | return(doc); |
13394 | 0 | } |
13395 | | |
13396 | | /** |
13397 | | * Parse an XML in-memory document and build a tree. The input buffer must |
13398 | | * not contain a terminating null byte. |
13399 | | * |
13400 | | * See #xmlCtxtReadMemory for details. |
13401 | | * |
13402 | | * @param buffer a pointer to a char array |
13403 | | * @param size the size of the array |
13404 | | * @param url base URL (optional) |
13405 | | * @param encoding the document encoding (optional) |
13406 | | * @param options a combination of xmlParserOption |
13407 | | * @returns the resulting document tree |
13408 | | */ |
13409 | | xmlDoc * |
13410 | | xmlReadMemory(const char *buffer, int size, const char *url, |
13411 | | const char *encoding, int options) |
13412 | 0 | { |
13413 | 0 | xmlParserCtxtPtr ctxt; |
13414 | 0 | xmlParserInputPtr input; |
13415 | 0 | xmlDocPtr doc = NULL; |
13416 | |
|
13417 | 0 | if (size < 0) |
13418 | 0 | return(NULL); |
13419 | | |
13420 | 0 | ctxt = xmlNewParserCtxt(); |
13421 | 0 | if (ctxt == NULL) |
13422 | 0 | return(NULL); |
13423 | | |
13424 | 0 | xmlCtxtUseOptions(ctxt, options); |
13425 | |
|
13426 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, url, buffer, size, encoding, |
13427 | 0 | XML_INPUT_BUF_STATIC); |
13428 | |
|
13429 | 0 | if (input != NULL) |
13430 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13431 | |
|
13432 | 0 | xmlFreeParserCtxt(ctxt); |
13433 | 0 | return(doc); |
13434 | 0 | } |
13435 | | |
13436 | | /** |
13437 | | * Parse an XML from a file descriptor and build a tree. |
13438 | | * |
13439 | | * See #xmlCtxtReadFd for details. |
13440 | | * |
13441 | | * NOTE that the file descriptor will not be closed when the |
13442 | | * context is freed or reset. |
13443 | | * |
13444 | | * @param fd an open file descriptor |
13445 | | * @param URL base URL (optional) |
13446 | | * @param encoding the document encoding (optional) |
13447 | | * @param options a combination of xmlParserOption |
13448 | | * @returns the resulting document tree |
13449 | | */ |
13450 | | xmlDoc * |
13451 | | xmlReadFd(int fd, const char *URL, const char *encoding, int options) |
13452 | 0 | { |
13453 | 0 | xmlParserCtxtPtr ctxt; |
13454 | 0 | xmlParserInputPtr input; |
13455 | 0 | xmlDocPtr doc = NULL; |
13456 | |
|
13457 | 0 | ctxt = xmlNewParserCtxt(); |
13458 | 0 | if (ctxt == NULL) |
13459 | 0 | return(NULL); |
13460 | | |
13461 | 0 | xmlCtxtUseOptions(ctxt, options); |
13462 | |
|
13463 | 0 | input = xmlCtxtNewInputFromFd(ctxt, URL, fd, encoding, 0); |
13464 | |
|
13465 | 0 | if (input != NULL) |
13466 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13467 | |
|
13468 | 0 | xmlFreeParserCtxt(ctxt); |
13469 | 0 | return(doc); |
13470 | 0 | } |
13471 | | |
13472 | | /** |
13473 | | * Parse an XML document from I/O functions and context and build a tree. |
13474 | | * |
13475 | | * See #xmlCtxtReadIO for details. |
13476 | | * |
13477 | | * @param ioread an I/O read function |
13478 | | * @param ioclose an I/O close function (optional) |
13479 | | * @param ioctx an I/O handler |
13480 | | * @param URL base URL (optional) |
13481 | | * @param encoding the document encoding (optional) |
13482 | | * @param options a combination of xmlParserOption |
13483 | | * @returns the resulting document tree |
13484 | | */ |
13485 | | xmlDoc * |
13486 | | xmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose, |
13487 | | void *ioctx, const char *URL, const char *encoding, int options) |
13488 | 0 | { |
13489 | 0 | xmlParserCtxtPtr ctxt; |
13490 | 0 | xmlParserInputPtr input; |
13491 | 0 | xmlDocPtr doc = NULL; |
13492 | |
|
13493 | 0 | ctxt = xmlNewParserCtxt(); |
13494 | 0 | if (ctxt == NULL) |
13495 | 0 | return(NULL); |
13496 | | |
13497 | 0 | xmlCtxtUseOptions(ctxt, options); |
13498 | |
|
13499 | 0 | input = xmlCtxtNewInputFromIO(ctxt, URL, ioread, ioclose, ioctx, |
13500 | 0 | encoding, 0); |
13501 | |
|
13502 | 0 | if (input != NULL) |
13503 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13504 | |
|
13505 | 0 | xmlFreeParserCtxt(ctxt); |
13506 | 0 | return(doc); |
13507 | 0 | } |
13508 | | |
13509 | | /** |
13510 | | * Parse an XML in-memory document and build a tree. |
13511 | | * |
13512 | | * `URL` is used as base to resolve external entities and for error |
13513 | | * reporting. |
13514 | | * |
13515 | | * See #xmlCtxtUseOptions for details. |
13516 | | * |
13517 | | * @param ctxt an XML parser context |
13518 | | * @param str a pointer to a zero terminated string |
13519 | | * @param URL base URL (optional) |
13520 | | * @param encoding the document encoding (optional) |
13521 | | * @param options a combination of xmlParserOption |
13522 | | * @returns the resulting document tree |
13523 | | */ |
13524 | | xmlDoc * |
13525 | | xmlCtxtReadDoc(xmlParserCtxt *ctxt, const xmlChar *str, |
13526 | | const char *URL, const char *encoding, int options) |
13527 | 0 | { |
13528 | 0 | xmlParserInputPtr input; |
13529 | |
|
13530 | 0 | if (ctxt == NULL) |
13531 | 0 | return(NULL); |
13532 | | |
13533 | 0 | xmlCtxtReset(ctxt); |
13534 | 0 | xmlCtxtUseOptions(ctxt, options); |
13535 | |
|
13536 | 0 | input = xmlCtxtNewInputFromString(ctxt, URL, (const char *) str, encoding, |
13537 | 0 | XML_INPUT_BUF_STATIC); |
13538 | 0 | if (input == NULL) |
13539 | 0 | return(NULL); |
13540 | | |
13541 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13542 | 0 | } |
13543 | | |
13544 | | /** |
13545 | | * Parse an XML file from the filesystem, the network or a user-defined |
13546 | | * resource loader. |
13547 | | * |
13548 | | * This function always enables the XML_PARSE_UNZIP option for |
13549 | | * backward compatibility. This feature is potentially insecure |
13550 | | * and might be removed from later versions. |
13551 | | * |
13552 | | * @param ctxt an XML parser context |
13553 | | * @param filename a file or URL |
13554 | | * @param encoding the document encoding (optional) |
13555 | | * @param options a combination of xmlParserOption |
13556 | | * @returns the resulting document tree |
13557 | | */ |
13558 | | xmlDoc * |
13559 | | xmlCtxtReadFile(xmlParserCtxt *ctxt, const char *filename, |
13560 | | const char *encoding, int options) |
13561 | 0 | { |
13562 | 0 | xmlParserInputPtr input; |
13563 | |
|
13564 | 0 | if (ctxt == NULL) |
13565 | 0 | return(NULL); |
13566 | | |
13567 | 0 | options |= XML_PARSE_UNZIP; |
13568 | |
|
13569 | 0 | xmlCtxtReset(ctxt); |
13570 | 0 | xmlCtxtUseOptions(ctxt, options); |
13571 | |
|
13572 | 0 | input = xmlCtxtNewInputFromUrl(ctxt, filename, NULL, encoding, 0); |
13573 | 0 | if (input == NULL) |
13574 | 0 | return(NULL); |
13575 | | |
13576 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13577 | 0 | } |
13578 | | |
13579 | | /** |
13580 | | * Parse an XML in-memory document and build a tree. The input buffer must |
13581 | | * not contain a terminating null byte. |
13582 | | * |
13583 | | * `URL` is used as base to resolve external entities and for error |
13584 | | * reporting. |
13585 | | * |
13586 | | * See #xmlCtxtUseOptions for details. |
13587 | | * |
13588 | | * @param ctxt an XML parser context |
13589 | | * @param buffer a pointer to a char array |
13590 | | * @param size the size of the array |
13591 | | * @param URL base URL (optional) |
13592 | | * @param encoding the document encoding (optional) |
13593 | | * @param options a combination of xmlParserOption |
13594 | | * @returns the resulting document tree |
13595 | | */ |
13596 | | xmlDoc * |
13597 | | xmlCtxtReadMemory(xmlParserCtxt *ctxt, const char *buffer, int size, |
13598 | | const char *URL, const char *encoding, int options) |
13599 | 0 | { |
13600 | 0 | xmlParserInputPtr input; |
13601 | |
|
13602 | 0 | if ((ctxt == NULL) || (size < 0)) |
13603 | 0 | return(NULL); |
13604 | | |
13605 | 0 | xmlCtxtReset(ctxt); |
13606 | 0 | xmlCtxtUseOptions(ctxt, options); |
13607 | |
|
13608 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, URL, buffer, size, encoding, |
13609 | 0 | XML_INPUT_BUF_STATIC); |
13610 | 0 | if (input == NULL) |
13611 | 0 | return(NULL); |
13612 | | |
13613 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13614 | 0 | } |
13615 | | |
13616 | | /** |
13617 | | * Parse an XML document from a file descriptor and build a tree. |
13618 | | * |
13619 | | * NOTE that the file descriptor will not be closed when the |
13620 | | * context is freed or reset. |
13621 | | * |
13622 | | * `URL` is used as base to resolve external entities and for error |
13623 | | * reporting. |
13624 | | * |
13625 | | * See #xmlCtxtUseOptions for details. |
13626 | | * |
13627 | | * @param ctxt an XML parser context |
13628 | | * @param fd an open file descriptor |
13629 | | * @param URL base URL (optional) |
13630 | | * @param encoding the document encoding (optional) |
13631 | | * @param options a combination of xmlParserOption |
13632 | | * @returns the resulting document tree |
13633 | | */ |
13634 | | xmlDoc * |
13635 | | xmlCtxtReadFd(xmlParserCtxt *ctxt, int fd, |
13636 | | const char *URL, const char *encoding, int options) |
13637 | 0 | { |
13638 | 0 | xmlParserInputPtr input; |
13639 | |
|
13640 | 0 | if (ctxt == NULL) |
13641 | 0 | return(NULL); |
13642 | | |
13643 | 0 | xmlCtxtReset(ctxt); |
13644 | 0 | xmlCtxtUseOptions(ctxt, options); |
13645 | |
|
13646 | 0 | input = xmlCtxtNewInputFromFd(ctxt, URL, fd, encoding, 0); |
13647 | 0 | if (input == NULL) |
13648 | 0 | return(NULL); |
13649 | | |
13650 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13651 | 0 | } |
13652 | | |
13653 | | /** |
13654 | | * parse an XML document from I/O functions and source and build a tree. |
13655 | | * This reuses the existing `ctxt` parser context |
13656 | | * |
13657 | | * `URL` is used as base to resolve external entities and for error |
13658 | | * reporting. |
13659 | | * |
13660 | | * See #xmlCtxtUseOptions for details. |
13661 | | * |
13662 | | * @param ctxt an XML parser context |
13663 | | * @param ioread an I/O read function |
13664 | | * @param ioclose an I/O close function |
13665 | | * @param ioctx an I/O handler |
13666 | | * @param URL the base URL to use for the document |
13667 | | * @param encoding the document encoding, or NULL |
13668 | | * @param options a combination of xmlParserOption |
13669 | | * @returns the resulting document tree |
13670 | | */ |
13671 | | xmlDoc * |
13672 | | xmlCtxtReadIO(xmlParserCtxt *ctxt, xmlInputReadCallback ioread, |
13673 | | xmlInputCloseCallback ioclose, void *ioctx, |
13674 | | const char *URL, |
13675 | | const char *encoding, int options) |
13676 | 0 | { |
13677 | 0 | xmlParserInputPtr input; |
13678 | |
|
13679 | 0 | if (ctxt == NULL) |
13680 | 0 | return(NULL); |
13681 | | |
13682 | 0 | xmlCtxtReset(ctxt); |
13683 | 0 | xmlCtxtUseOptions(ctxt, options); |
13684 | |
|
13685 | 0 | input = xmlCtxtNewInputFromIO(ctxt, URL, ioread, ioclose, ioctx, |
13686 | 0 | encoding, 0); |
13687 | 0 | if (input == NULL) |
13688 | 0 | return(NULL); |
13689 | | |
13690 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13691 | 0 | } |
13692 | | |