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 | 5.37M | #define NS_INDEX_EMPTY INT_MAX |
81 | 255k | #define NS_INDEX_XML (INT_MAX - 1) |
82 | 1.03M | #define URI_HASH_EMPTY 0xD943A04E |
83 | 124k | #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 | 1.05M | #define XML_MAX_ATTRS 100000000 /* 100 million */ |
94 | | |
95 | 5.33M | #define XML_SPECIAL_EXTERNAL (1 << 20) |
96 | 5.30M | #define XML_SPECIAL_TYPE_MASK (XML_SPECIAL_EXTERNAL - 1) |
97 | | |
98 | 5.56M | #define XML_ATTVAL_ALLOC (1 << 0) |
99 | 112k | #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 | 814k | #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 | 817k | #define XML_ENT_FIXED_COST 20 |
170 | | |
171 | 274M | #define XML_PARSER_BIG_BUFFER_SIZE 300 |
172 | 730k | #define XML_PARSER_BUFFER_SIZE 100 |
173 | 865k | #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 version of the library used at |
188 | | * run-time. |
189 | | */ |
190 | | const char *const |
191 | | xmlParserVersion = LIBXML_VERSION_STRING LIBXML_VERSION_EXTRA; |
192 | | |
193 | | /* |
194 | | * List of XML prefixed PI allowed by W3C specs |
195 | | */ |
196 | | |
197 | | static const char* const xmlW3CPIs[] = { |
198 | | "xml-stylesheet", |
199 | | "xml-model", |
200 | | NULL |
201 | | }; |
202 | | |
203 | | |
204 | | /* DEPR void xmlParserHandleReference(xmlParserCtxtPtr ctxt); */ |
205 | | static xmlEntityPtr xmlParseStringPEReference(xmlParserCtxtPtr ctxt, |
206 | | const xmlChar **str); |
207 | | |
208 | | static void |
209 | | xmlCtxtParseEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr ent); |
210 | | |
211 | | static int |
212 | | xmlLoadEntityContent(xmlParserCtxtPtr ctxt, xmlEntityPtr entity); |
213 | | |
214 | | static void |
215 | | xmlParsePERefInternal(xmlParserCtxt *ctxt, int markupDecl); |
216 | | |
217 | | /************************************************************************ |
218 | | * * |
219 | | * Some factorized error routines * |
220 | | * * |
221 | | ************************************************************************/ |
222 | | |
223 | | static void |
224 | 2.28k | xmlErrMemory(xmlParserCtxtPtr ctxt) { |
225 | 2.28k | xmlCtxtErrMemory(ctxt); |
226 | 2.28k | } |
227 | | |
228 | | /** |
229 | | * Handle a redefinition of attribute error |
230 | | * |
231 | | * @param ctxt an XML parser context |
232 | | * @param prefix the attribute prefix |
233 | | * @param localname the attribute localname |
234 | | */ |
235 | | static void |
236 | | xmlErrAttributeDup(xmlParserCtxtPtr ctxt, const xmlChar * prefix, |
237 | | const xmlChar * localname) |
238 | 17.0k | { |
239 | 17.0k | if (prefix == NULL) |
240 | 3.24k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, XML_ERR_ATTRIBUTE_REDEFINED, |
241 | 3.24k | XML_ERR_FATAL, localname, NULL, NULL, 0, |
242 | 3.24k | "Attribute %s redefined\n", localname); |
243 | 13.7k | else |
244 | 13.7k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, XML_ERR_ATTRIBUTE_REDEFINED, |
245 | 13.7k | XML_ERR_FATAL, prefix, localname, NULL, 0, |
246 | 13.7k | "Attribute %s:%s redefined\n", prefix, localname); |
247 | 17.0k | } |
248 | | |
249 | | /** |
250 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
251 | | * |
252 | | * @param ctxt an XML parser context |
253 | | * @param error the error number |
254 | | * @param msg the error message |
255 | | */ |
256 | | static void LIBXML_ATTR_FORMAT(3,0) |
257 | | xmlFatalErrMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
258 | | const char *msg) |
259 | 72.0M | { |
260 | 72.0M | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
261 | 72.0M | NULL, NULL, NULL, 0, "%s", msg); |
262 | 72.0M | } |
263 | | |
264 | | /** |
265 | | * Handle a warning. |
266 | | * |
267 | | * @param ctxt an XML parser context |
268 | | * @param error the error number |
269 | | * @param msg the error message |
270 | | * @param str1 extra data |
271 | | * @param str2 extra data |
272 | | */ |
273 | | void LIBXML_ATTR_FORMAT(3,0) |
274 | | xmlWarningMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
275 | | const char *msg, const xmlChar *str1, const xmlChar *str2) |
276 | 57.7k | { |
277 | 57.7k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_WARNING, |
278 | 57.7k | str1, str2, NULL, 0, msg, str1, str2); |
279 | 57.7k | } |
280 | | |
281 | | #ifdef LIBXML_VALID_ENABLED |
282 | | /** |
283 | | * Handle a validity error. |
284 | | * |
285 | | * @param ctxt an XML parser context |
286 | | * @param error the error number |
287 | | * @param msg the error message |
288 | | * @param str1 extra data |
289 | | * @param str2 extra data |
290 | | */ |
291 | | static void LIBXML_ATTR_FORMAT(3,0) |
292 | | xmlValidityError(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
293 | | const char *msg, const xmlChar *str1, const xmlChar *str2) |
294 | | { |
295 | | ctxt->valid = 0; |
296 | | |
297 | | xmlCtxtErr(ctxt, NULL, XML_FROM_DTD, error, XML_ERR_ERROR, |
298 | | str1, str2, NULL, 0, msg, str1, str2); |
299 | | } |
300 | | #endif |
301 | | |
302 | | /** |
303 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
304 | | * |
305 | | * @param ctxt an XML parser context |
306 | | * @param error the error number |
307 | | * @param msg the error message |
308 | | * @param val an integer value |
309 | | */ |
310 | | static void LIBXML_ATTR_FORMAT(3,0) |
311 | | xmlFatalErrMsgInt(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
312 | | const char *msg, int val) |
313 | 8.12M | { |
314 | 8.12M | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
315 | 8.12M | NULL, NULL, NULL, val, msg, val); |
316 | 8.12M | } |
317 | | |
318 | | /** |
319 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
320 | | * |
321 | | * @param ctxt an XML parser context |
322 | | * @param error the error number |
323 | | * @param msg the error message |
324 | | * @param str1 an string info |
325 | | * @param val an integer value |
326 | | * @param str2 an string info |
327 | | */ |
328 | | static void LIBXML_ATTR_FORMAT(3,0) |
329 | | xmlFatalErrMsgStrIntStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
330 | | const char *msg, const xmlChar *str1, int val, |
331 | | const xmlChar *str2) |
332 | 3.03M | { |
333 | 3.03M | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
334 | 3.03M | str1, str2, NULL, val, msg, str1, val, str2); |
335 | 3.03M | } |
336 | | |
337 | | /** |
338 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
339 | | * |
340 | | * @param ctxt an XML parser context |
341 | | * @param error the error number |
342 | | * @param msg the error message |
343 | | * @param val a string value |
344 | | */ |
345 | | static void LIBXML_ATTR_FORMAT(3,0) |
346 | | xmlFatalErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
347 | | const char *msg, const xmlChar * val) |
348 | 540k | { |
349 | 540k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, |
350 | 540k | val, NULL, NULL, 0, msg, val); |
351 | 540k | } |
352 | | |
353 | | /** |
354 | | * Handle a non fatal parser error |
355 | | * |
356 | | * @param ctxt an XML parser context |
357 | | * @param error the error number |
358 | | * @param msg the error message |
359 | | * @param val a string value |
360 | | */ |
361 | | static void LIBXML_ATTR_FORMAT(3,0) |
362 | | xmlErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
363 | | const char *msg, const xmlChar * val) |
364 | 19.7k | { |
365 | 19.7k | xmlCtxtErr(ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_ERROR, |
366 | 19.7k | val, NULL, NULL, 0, msg, val); |
367 | 19.7k | } |
368 | | |
369 | | /** |
370 | | * Handle a fatal parser error, i.e. violating Well-Formedness constraints |
371 | | * |
372 | | * @param ctxt an XML parser context |
373 | | * @param error the error number |
374 | | * @param msg the message |
375 | | * @param info1 extra information string |
376 | | * @param info2 extra information string |
377 | | * @param info3 extra information string |
378 | | */ |
379 | | static void LIBXML_ATTR_FORMAT(3,0) |
380 | | xmlNsErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
381 | | const char *msg, |
382 | | const xmlChar * info1, const xmlChar * info2, |
383 | | const xmlChar * info3) |
384 | 806k | { |
385 | 806k | ctxt->nsWellFormed = 0; |
386 | | |
387 | 806k | xmlCtxtErr(ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_ERROR, |
388 | 806k | info1, info2, info3, 0, msg, info1, info2, info3); |
389 | 806k | } |
390 | | |
391 | | /** |
392 | | * Handle a namespace warning error |
393 | | * |
394 | | * @param ctxt an XML parser context |
395 | | * @param error the error number |
396 | | * @param msg the message |
397 | | * @param info1 extra information string |
398 | | * @param info2 extra information string |
399 | | * @param info3 extra information string |
400 | | */ |
401 | | static void LIBXML_ATTR_FORMAT(3,0) |
402 | | xmlNsWarn(xmlParserCtxtPtr ctxt, xmlParserErrors error, |
403 | | const char *msg, |
404 | | const xmlChar * info1, const xmlChar * info2, |
405 | | const xmlChar * info3) |
406 | 21.6k | { |
407 | 21.6k | xmlCtxtErr(ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_WARNING, |
408 | 21.6k | info1, info2, info3, 0, msg, info1, info2, info3); |
409 | 21.6k | } |
410 | | |
411 | | static void |
412 | 2.50M | xmlSaturatedAdd(unsigned long *dst, unsigned long val) { |
413 | 2.50M | if (val > ULONG_MAX - *dst) |
414 | 0 | *dst = ULONG_MAX; |
415 | 2.50M | else |
416 | 2.50M | *dst += val; |
417 | 2.50M | } |
418 | | |
419 | | static void |
420 | 843k | xmlSaturatedAddSizeT(unsigned long *dst, unsigned long val) { |
421 | 843k | if (val > ULONG_MAX - *dst) |
422 | 0 | *dst = ULONG_MAX; |
423 | 843k | else |
424 | 843k | *dst += val; |
425 | 843k | } |
426 | | |
427 | | /** |
428 | | * Check for non-linear entity expansion behaviour. |
429 | | * |
430 | | * In some cases like xmlExpandEntityInAttValue, this function is called |
431 | | * for each, possibly nested entity and its unexpanded content length. |
432 | | * |
433 | | * In other cases like #xmlParseReference, it's only called for each |
434 | | * top-level entity with its unexpanded content length plus the sum of |
435 | | * the unexpanded content lengths (plus fixed cost) of all nested |
436 | | * entities. |
437 | | * |
438 | | * Summing the unexpanded lengths also adds the length of the reference. |
439 | | * This is by design. Taking the length of the entity name into account |
440 | | * discourages attacks that try to waste CPU time with abusively long |
441 | | * entity names. See test/recurse/lol6.xml for example. Each call also |
442 | | * adds some fixed cost XML_ENT_FIXED_COST to discourage attacks with |
443 | | * short entities. |
444 | | * |
445 | | * @param ctxt parser context |
446 | | * @param extra sum of unexpanded entity sizes |
447 | | * @returns 1 on error, 0 on success. |
448 | | */ |
449 | | static int |
450 | | xmlParserEntityCheck(xmlParserCtxtPtr ctxt, unsigned long extra) |
451 | 817k | { |
452 | 817k | unsigned long consumed; |
453 | 817k | unsigned long *expandedSize; |
454 | 817k | xmlParserInputPtr input = ctxt->input; |
455 | 817k | xmlEntityPtr entity = input->entity; |
456 | | |
457 | 817k | if ((entity) && (entity->flags & XML_ENT_CHECKED)) |
458 | 2.46k | return(0); |
459 | | |
460 | | /* |
461 | | * Compute total consumed bytes so far, including input streams of |
462 | | * external entities. |
463 | | */ |
464 | 814k | consumed = input->consumed; |
465 | 814k | xmlSaturatedAddSizeT(&consumed, input->cur - input->base); |
466 | 814k | xmlSaturatedAdd(&consumed, ctxt->sizeentities); |
467 | | |
468 | 814k | if (entity) |
469 | 32.6k | expandedSize = &entity->expandedSize; |
470 | 782k | else |
471 | 782k | expandedSize = &ctxt->sizeentcopy; |
472 | | |
473 | | /* |
474 | | * Add extra cost and some fixed cost. |
475 | | */ |
476 | 814k | xmlSaturatedAdd(expandedSize, extra); |
477 | 814k | xmlSaturatedAdd(expandedSize, XML_ENT_FIXED_COST); |
478 | | |
479 | | /* |
480 | | * It's important to always use saturation arithmetic when tracking |
481 | | * entity sizes to make the size checks reliable. If "sizeentcopy" |
482 | | * overflows, we have to abort. |
483 | | */ |
484 | 814k | if ((*expandedSize > XML_PARSER_ALLOWED_EXPANSION) && |
485 | 814k | ((*expandedSize >= ULONG_MAX) || |
486 | 22.1k | (*expandedSize / ctxt->maxAmpl > consumed))) { |
487 | 1.20k | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
488 | 1.20k | "Maximum entity amplification factor exceeded, see " |
489 | 1.20k | "xmlCtxtSetMaxAmplification.\n"); |
490 | 1.20k | xmlHaltParser(ctxt); |
491 | 1.20k | return(1); |
492 | 1.20k | } |
493 | | |
494 | 813k | return(0); |
495 | 814k | } |
496 | | |
497 | | /************************************************************************ |
498 | | * * |
499 | | * Library wide options * |
500 | | * * |
501 | | ************************************************************************/ |
502 | | |
503 | | /** |
504 | | * Examines if the library has been compiled with a given feature. |
505 | | * |
506 | | * @param feature the feature to be examined |
507 | | * @returns zero (0) if the feature does not exist or an unknown |
508 | | * feature is requested, non-zero otherwise. |
509 | | */ |
510 | | int |
511 | | xmlHasFeature(xmlFeature feature) |
512 | 0 | { |
513 | 0 | switch (feature) { |
514 | 0 | case XML_WITH_THREAD: |
515 | 0 | #ifdef LIBXML_THREAD_ENABLED |
516 | 0 | return(1); |
517 | | #else |
518 | | return(0); |
519 | | #endif |
520 | 0 | case XML_WITH_TREE: |
521 | 0 | return(1); |
522 | 0 | case XML_WITH_OUTPUT: |
523 | 0 | #ifdef LIBXML_OUTPUT_ENABLED |
524 | 0 | return(1); |
525 | | #else |
526 | | return(0); |
527 | | #endif |
528 | 0 | case XML_WITH_PUSH: |
529 | | #ifdef LIBXML_PUSH_ENABLED |
530 | | return(1); |
531 | | #else |
532 | 0 | return(0); |
533 | 0 | #endif |
534 | 0 | case XML_WITH_READER: |
535 | | #ifdef LIBXML_READER_ENABLED |
536 | | return(1); |
537 | | #else |
538 | 0 | return(0); |
539 | 0 | #endif |
540 | 0 | case XML_WITH_PATTERN: |
541 | 0 | #ifdef LIBXML_PATTERN_ENABLED |
542 | 0 | return(1); |
543 | | #else |
544 | | return(0); |
545 | | #endif |
546 | 0 | case XML_WITH_WRITER: |
547 | | #ifdef LIBXML_WRITER_ENABLED |
548 | | return(1); |
549 | | #else |
550 | 0 | return(0); |
551 | 0 | #endif |
552 | 0 | case XML_WITH_SAX1: |
553 | | #ifdef LIBXML_SAX1_ENABLED |
554 | | return(1); |
555 | | #else |
556 | 0 | return(0); |
557 | 0 | #endif |
558 | 0 | case XML_WITH_HTTP: |
559 | 0 | return(0); |
560 | 0 | case XML_WITH_VALID: |
561 | | #ifdef LIBXML_VALID_ENABLED |
562 | | return(1); |
563 | | #else |
564 | 0 | return(0); |
565 | 0 | #endif |
566 | 0 | case XML_WITH_HTML: |
567 | 0 | #ifdef LIBXML_HTML_ENABLED |
568 | 0 | return(1); |
569 | | #else |
570 | | return(0); |
571 | | #endif |
572 | 0 | case XML_WITH_LEGACY: |
573 | 0 | return(0); |
574 | 0 | case XML_WITH_C14N: |
575 | | #ifdef LIBXML_C14N_ENABLED |
576 | | return(1); |
577 | | #else |
578 | 0 | return(0); |
579 | 0 | #endif |
580 | 0 | case XML_WITH_CATALOG: |
581 | 0 | #ifdef LIBXML_CATALOG_ENABLED |
582 | 0 | return(1); |
583 | | #else |
584 | | return(0); |
585 | | #endif |
586 | 0 | case XML_WITH_XPATH: |
587 | 0 | #ifdef LIBXML_XPATH_ENABLED |
588 | 0 | return(1); |
589 | | #else |
590 | | return(0); |
591 | | #endif |
592 | 0 | case XML_WITH_XPTR: |
593 | 0 | #ifdef LIBXML_XPTR_ENABLED |
594 | 0 | return(1); |
595 | | #else |
596 | | return(0); |
597 | | #endif |
598 | 0 | case XML_WITH_XINCLUDE: |
599 | 0 | #ifdef LIBXML_XINCLUDE_ENABLED |
600 | 0 | return(1); |
601 | | #else |
602 | | return(0); |
603 | | #endif |
604 | 0 | case XML_WITH_ICONV: |
605 | 0 | #ifdef LIBXML_ICONV_ENABLED |
606 | 0 | return(1); |
607 | | #else |
608 | | return(0); |
609 | | #endif |
610 | 0 | case XML_WITH_ISO8859X: |
611 | 0 | #ifdef LIBXML_ISO8859X_ENABLED |
612 | 0 | return(1); |
613 | | #else |
614 | | return(0); |
615 | | #endif |
616 | 0 | case XML_WITH_UNICODE: |
617 | 0 | return(0); |
618 | 0 | case XML_WITH_REGEXP: |
619 | | #ifdef LIBXML_REGEXP_ENABLED |
620 | | return(1); |
621 | | #else |
622 | 0 | return(0); |
623 | 0 | #endif |
624 | 0 | case XML_WITH_AUTOMATA: |
625 | | #ifdef LIBXML_REGEXP_ENABLED |
626 | | return(1); |
627 | | #else |
628 | 0 | return(0); |
629 | 0 | #endif |
630 | 0 | case XML_WITH_EXPR: |
631 | 0 | return(0); |
632 | 0 | case XML_WITH_RELAXNG: |
633 | | #ifdef LIBXML_RELAXNG_ENABLED |
634 | | return(1); |
635 | | #else |
636 | 0 | return(0); |
637 | 0 | #endif |
638 | 0 | case XML_WITH_SCHEMAS: |
639 | | #ifdef LIBXML_SCHEMAS_ENABLED |
640 | | return(1); |
641 | | #else |
642 | 0 | return(0); |
643 | 0 | #endif |
644 | 0 | case XML_WITH_SCHEMATRON: |
645 | | #ifdef LIBXML_SCHEMATRON_ENABLED |
646 | | return(1); |
647 | | #else |
648 | 0 | return(0); |
649 | 0 | #endif |
650 | 0 | case XML_WITH_MODULES: |
651 | 0 | #ifdef LIBXML_MODULES_ENABLED |
652 | 0 | return(1); |
653 | | #else |
654 | | return(0); |
655 | | #endif |
656 | 0 | case XML_WITH_DEBUG: |
657 | 0 | #ifdef LIBXML_DEBUG_ENABLED |
658 | 0 | return(1); |
659 | | #else |
660 | | return(0); |
661 | | #endif |
662 | 0 | case XML_WITH_DEBUG_MEM: |
663 | 0 | return(0); |
664 | 0 | case XML_WITH_ZLIB: |
665 | | #ifdef LIBXML_ZLIB_ENABLED |
666 | | return(1); |
667 | | #else |
668 | 0 | return(0); |
669 | 0 | #endif |
670 | 0 | case XML_WITH_LZMA: |
671 | | #ifdef LIBXML_LZMA_ENABLED |
672 | | return(1); |
673 | | #else |
674 | 0 | return(0); |
675 | 0 | #endif |
676 | 0 | case XML_WITH_ICU: |
677 | | #ifdef LIBXML_ICU_ENABLED |
678 | | return(1); |
679 | | #else |
680 | 0 | return(0); |
681 | 0 | #endif |
682 | 0 | default: |
683 | 0 | break; |
684 | 0 | } |
685 | 0 | return(0); |
686 | 0 | } |
687 | | |
688 | | /************************************************************************ |
689 | | * * |
690 | | * Simple string buffer * |
691 | | * * |
692 | | ************************************************************************/ |
693 | | |
694 | | typedef struct { |
695 | | xmlChar *mem; |
696 | | unsigned size; |
697 | | unsigned cap; /* size < cap */ |
698 | | unsigned max; /* size <= max */ |
699 | | xmlParserErrors code; |
700 | | } xmlSBuf; |
701 | | |
702 | | static void |
703 | 5.42M | xmlSBufInit(xmlSBuf *buf, unsigned max) { |
704 | 5.42M | buf->mem = NULL; |
705 | 5.42M | buf->size = 0; |
706 | 5.42M | buf->cap = 0; |
707 | 5.42M | buf->max = max; |
708 | 5.42M | buf->code = XML_ERR_OK; |
709 | 5.42M | } |
710 | | |
711 | | static int |
712 | 682k | xmlSBufGrow(xmlSBuf *buf, unsigned len) { |
713 | 682k | xmlChar *mem; |
714 | 682k | unsigned cap; |
715 | | |
716 | 682k | if (len >= UINT_MAX / 2 - buf->size) { |
717 | 0 | if (buf->code == XML_ERR_OK) |
718 | 0 | buf->code = XML_ERR_RESOURCE_LIMIT; |
719 | 0 | return(-1); |
720 | 0 | } |
721 | | |
722 | 682k | cap = (buf->size + len) * 2; |
723 | 682k | if (cap < 240) |
724 | 440k | cap = 240; |
725 | | |
726 | 682k | mem = xmlRealloc(buf->mem, cap); |
727 | 682k | if (mem == NULL) { |
728 | 137k | buf->code = XML_ERR_NO_MEMORY; |
729 | 137k | return(-1); |
730 | 137k | } |
731 | | |
732 | 545k | buf->mem = mem; |
733 | 545k | buf->cap = cap; |
734 | | |
735 | 545k | return(0); |
736 | 682k | } |
737 | | |
738 | | static void |
739 | 145M | xmlSBufAddString(xmlSBuf *buf, const xmlChar *str, unsigned len) { |
740 | 145M | if (buf->max - buf->size < len) { |
741 | 2.72k | if (buf->code == XML_ERR_OK) |
742 | 42 | buf->code = XML_ERR_RESOURCE_LIMIT; |
743 | 2.72k | return; |
744 | 2.72k | } |
745 | | |
746 | 145M | if (buf->cap - buf->size <= len) { |
747 | 674k | if (xmlSBufGrow(buf, len) < 0) |
748 | 136k | return; |
749 | 674k | } |
750 | | |
751 | 145M | if (len > 0) |
752 | 145M | memcpy(buf->mem + buf->size, str, len); |
753 | 145M | buf->size += len; |
754 | 145M | } |
755 | | |
756 | | static void |
757 | 141M | xmlSBufAddCString(xmlSBuf *buf, const char *str, unsigned len) { |
758 | 141M | xmlSBufAddString(buf, (const xmlChar *) str, len); |
759 | 141M | } |
760 | | |
761 | | static void |
762 | 300k | xmlSBufAddChar(xmlSBuf *buf, int c) { |
763 | 300k | xmlChar *end; |
764 | | |
765 | 300k | if (buf->max - buf->size < 4) { |
766 | 105 | if (buf->code == XML_ERR_OK) |
767 | 0 | buf->code = XML_ERR_RESOURCE_LIMIT; |
768 | 105 | return; |
769 | 105 | } |
770 | | |
771 | 300k | if (buf->cap - buf->size <= 4) { |
772 | 7.89k | if (xmlSBufGrow(buf, 4) < 0) |
773 | 516 | return; |
774 | 7.89k | } |
775 | | |
776 | 300k | end = buf->mem + buf->size; |
777 | | |
778 | 300k | if (c < 0x80) { |
779 | 148k | *end = (xmlChar) c; |
780 | 148k | buf->size += 1; |
781 | 151k | } else { |
782 | 151k | buf->size += xmlCopyCharMultiByte(end, c); |
783 | 151k | } |
784 | 300k | } |
785 | | |
786 | | static void |
787 | 131M | xmlSBufAddReplChar(xmlSBuf *buf) { |
788 | 131M | xmlSBufAddCString(buf, "\xEF\xBF\xBD", 3); |
789 | 131M | } |
790 | | |
791 | | static void |
792 | 230 | xmlSBufReportError(xmlSBuf *buf, xmlParserCtxtPtr ctxt, const char *errMsg) { |
793 | 230 | if (buf->code == XML_ERR_NO_MEMORY) |
794 | 188 | xmlCtxtErrMemory(ctxt); |
795 | 42 | else |
796 | 42 | xmlFatalErr(ctxt, buf->code, errMsg); |
797 | 230 | } |
798 | | |
799 | | static xmlChar * |
800 | | xmlSBufFinish(xmlSBuf *buf, int *sizeOut, xmlParserCtxtPtr ctxt, |
801 | 465k | const char *errMsg) { |
802 | 465k | if (buf->mem == NULL) { |
803 | 36.4k | buf->mem = xmlMalloc(1); |
804 | 36.4k | if (buf->mem == NULL) { |
805 | 28 | buf->code = XML_ERR_NO_MEMORY; |
806 | 36.4k | } else { |
807 | 36.4k | buf->mem[0] = 0; |
808 | 36.4k | } |
809 | 428k | } else { |
810 | 428k | buf->mem[buf->size] = 0; |
811 | 428k | } |
812 | | |
813 | 465k | if (buf->code == XML_ERR_OK) { |
814 | 465k | if (sizeOut != NULL) |
815 | 298k | *sizeOut = buf->size; |
816 | 465k | return(buf->mem); |
817 | 465k | } |
818 | | |
819 | 93 | xmlSBufReportError(buf, ctxt, errMsg); |
820 | | |
821 | 93 | xmlFree(buf->mem); |
822 | | |
823 | 93 | if (sizeOut != NULL) |
824 | 41 | *sizeOut = 0; |
825 | 93 | return(NULL); |
826 | 465k | } |
827 | | |
828 | | static void |
829 | 4.93M | xmlSBufCleanup(xmlSBuf *buf, xmlParserCtxtPtr ctxt, const char *errMsg) { |
830 | 4.93M | if (buf->code != XML_ERR_OK) |
831 | 137 | xmlSBufReportError(buf, ctxt, errMsg); |
832 | | |
833 | 4.93M | xmlFree(buf->mem); |
834 | 4.93M | } |
835 | | |
836 | | static int |
837 | | xmlUTF8MultibyteLen(xmlParserCtxtPtr ctxt, const xmlChar *str, |
838 | 337M | const char *errMsg) { |
839 | 337M | int c = str[0]; |
840 | 337M | int c1 = str[1]; |
841 | | |
842 | 337M | if ((c1 & 0xC0) != 0x80) |
843 | 24.8M | goto encoding_error; |
844 | | |
845 | 313M | if (c < 0xE0) { |
846 | | /* 2-byte sequence */ |
847 | 86.3M | if (c < 0xC2) |
848 | 38.2M | goto encoding_error; |
849 | | |
850 | 48.1M | return(2); |
851 | 226M | } else { |
852 | 226M | int c2 = str[2]; |
853 | | |
854 | 226M | if ((c2 & 0xC0) != 0x80) |
855 | 38.3k | goto encoding_error; |
856 | | |
857 | 226M | if (c < 0xF0) { |
858 | | /* 3-byte sequence */ |
859 | 226M | if (c == 0xE0) { |
860 | | /* overlong */ |
861 | 3.83M | if (c1 < 0xA0) |
862 | 484 | goto encoding_error; |
863 | 222M | } else if (c == 0xED) { |
864 | | /* surrogate */ |
865 | 2.74k | if (c1 >= 0xA0) |
866 | 1.02k | goto encoding_error; |
867 | 222M | } else if (c == 0xEF) { |
868 | | /* U+FFFE and U+FFFF are invalid Chars */ |
869 | 322k | if ((c1 == 0xBF) && (c2 >= 0xBE)) |
870 | 7.69k | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, errMsg); |
871 | 322k | } |
872 | | |
873 | 226M | return(3); |
874 | 226M | } else { |
875 | | /* 4-byte sequence */ |
876 | 76.8k | if ((str[3] & 0xC0) != 0x80) |
877 | 9.06k | goto encoding_error; |
878 | 67.8k | if (c == 0xF0) { |
879 | | /* overlong */ |
880 | 23.1k | if (c1 < 0x90) |
881 | 3.03k | goto encoding_error; |
882 | 44.6k | } else if (c >= 0xF4) { |
883 | | /* greater than 0x10FFFF */ |
884 | 33.7k | if ((c > 0xF4) || (c1 >= 0x90)) |
885 | 33.4k | goto encoding_error; |
886 | 33.7k | } |
887 | | |
888 | 31.3k | return(4); |
889 | 67.8k | } |
890 | 226M | } |
891 | | |
892 | 63.1M | encoding_error: |
893 | | /* Only report the first error */ |
894 | 63.1M | if ((ctxt->input->flags & XML_INPUT_ENCODING_ERROR) == 0) { |
895 | 20.5k | xmlCtxtErrIO(ctxt, XML_ERR_INVALID_ENCODING, NULL); |
896 | 20.5k | ctxt->input->flags |= XML_INPUT_ENCODING_ERROR; |
897 | 20.5k | } |
898 | | |
899 | 63.1M | return(0); |
900 | 313M | } |
901 | | |
902 | | /************************************************************************ |
903 | | * * |
904 | | * SAX2 defaulted attributes handling * |
905 | | * * |
906 | | ************************************************************************/ |
907 | | |
908 | | /** |
909 | | * Final initialization of the parser context before starting to parse. |
910 | | * |
911 | | * This accounts for users modifying struct members of parser context |
912 | | * directly. |
913 | | * |
914 | | * @param ctxt an XML parser context |
915 | | */ |
916 | | static void |
917 | 851k | xmlCtxtInitializeLate(xmlParserCtxtPtr ctxt) { |
918 | 851k | xmlSAXHandlerPtr sax; |
919 | | |
920 | | /* Avoid unused variable warning if features are disabled. */ |
921 | 851k | (void) sax; |
922 | | |
923 | | /* |
924 | | * Changing the SAX struct directly is still widespread practice |
925 | | * in internal and external code. |
926 | | */ |
927 | 851k | if (ctxt == NULL) return; |
928 | 851k | sax = ctxt->sax; |
929 | | #ifdef LIBXML_SAX1_ENABLED |
930 | | /* |
931 | | * Only enable SAX2 if there SAX2 element handlers, except when there |
932 | | * are no element handlers at all. |
933 | | */ |
934 | | if (((ctxt->options & XML_PARSE_SAX1) == 0) && |
935 | | (sax) && |
936 | | (sax->initialized == XML_SAX2_MAGIC) && |
937 | | ((sax->startElementNs != NULL) || |
938 | | (sax->endElementNs != NULL) || |
939 | | ((sax->startElement == NULL) && (sax->endElement == NULL)))) |
940 | | ctxt->sax2 = 1; |
941 | | #else |
942 | 851k | ctxt->sax2 = 1; |
943 | 851k | #endif /* LIBXML_SAX1_ENABLED */ |
944 | | |
945 | | /* |
946 | | * Some users replace the dictionary directly in the context struct. |
947 | | * We really need an API function to do that cleanly. |
948 | | */ |
949 | 851k | ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); |
950 | 851k | ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); |
951 | 851k | ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); |
952 | 851k | if ((ctxt->str_xml==NULL) || (ctxt->str_xmlns==NULL) || |
953 | 851k | (ctxt->str_xml_ns == NULL)) { |
954 | 0 | xmlErrMemory(ctxt); |
955 | 0 | } |
956 | | |
957 | 851k | xmlDictSetLimit(ctxt->dict, |
958 | 851k | (ctxt->options & XML_PARSE_HUGE) ? |
959 | 0 | 0 : |
960 | 851k | XML_MAX_DICTIONARY_LIMIT); |
961 | | |
962 | | #ifdef LIBXML_VALID_ENABLED |
963 | | if (ctxt->validate) |
964 | | ctxt->vctxt.flags |= XML_VCTXT_VALIDATE; |
965 | | else |
966 | | ctxt->vctxt.flags &= ~XML_VCTXT_VALIDATE; |
967 | | #endif /* LIBXML_VALID_ENABLED */ |
968 | 851k | } |
969 | | |
970 | | typedef struct { |
971 | | xmlHashedString prefix; |
972 | | xmlHashedString name; |
973 | | xmlHashedString value; |
974 | | const xmlChar *valueEnd; |
975 | | int external; |
976 | | int expandedSize; |
977 | | } xmlDefAttr; |
978 | | |
979 | | typedef struct _xmlDefAttrs xmlDefAttrs; |
980 | | typedef xmlDefAttrs *xmlDefAttrsPtr; |
981 | | struct _xmlDefAttrs { |
982 | | int nbAttrs; /* number of defaulted attributes on that element */ |
983 | | int maxAttrs; /* the size of the array */ |
984 | | #if __STDC_VERSION__ >= 199901L |
985 | | /* Using a C99 flexible array member avoids UBSan errors. */ |
986 | | xmlDefAttr attrs[] ATTRIBUTE_COUNTED_BY(maxAttrs); |
987 | | #else |
988 | | xmlDefAttr attrs[1]; |
989 | | #endif |
990 | | }; |
991 | | |
992 | | /** |
993 | | * Normalize the space in non CDATA attribute values: |
994 | | * If the attribute type is not CDATA, then the XML processor MUST further |
995 | | * process the normalized attribute value by discarding any leading and |
996 | | * trailing space (\#x20) characters, and by replacing sequences of space |
997 | | * (\#x20) characters by a single space (\#x20) character. |
998 | | * Note that the size of dst need to be at least src, and if one doesn't need |
999 | | * to preserve dst (and it doesn't come from a dictionary or read-only) then |
1000 | | * passing src as dst is just fine. |
1001 | | * |
1002 | | * @param src the source string |
1003 | | * @param dst the target string |
1004 | | * @returns a pointer to the normalized value (dst) or NULL if no conversion |
1005 | | * is needed. |
1006 | | */ |
1007 | | static xmlChar * |
1008 | | xmlAttrNormalizeSpace(const xmlChar *src, xmlChar *dst) |
1009 | 41.2k | { |
1010 | 41.2k | if ((src == NULL) || (dst == NULL)) |
1011 | 0 | return(NULL); |
1012 | | |
1013 | 42.1k | while (*src == 0x20) src++; |
1014 | 14.3M | while (*src != 0) { |
1015 | 14.3M | if (*src == 0x20) { |
1016 | 32.8k | while (*src == 0x20) src++; |
1017 | 14.1k | if (*src != 0) |
1018 | 13.0k | *dst++ = 0x20; |
1019 | 14.3M | } else { |
1020 | 14.3M | *dst++ = *src++; |
1021 | 14.3M | } |
1022 | 14.3M | } |
1023 | 41.2k | *dst = 0; |
1024 | 41.2k | if (dst == src) |
1025 | 39.5k | return(NULL); |
1026 | 1.65k | return(dst); |
1027 | 41.2k | } |
1028 | | |
1029 | | /** |
1030 | | * Add a defaulted attribute for an element |
1031 | | * |
1032 | | * @param ctxt an XML parser context |
1033 | | * @param fullname the element fullname |
1034 | | * @param fullattr the attribute fullname |
1035 | | * @param value the attribute value |
1036 | | */ |
1037 | | static void |
1038 | | xmlAddDefAttrs(xmlParserCtxtPtr ctxt, |
1039 | | const xmlChar *fullname, |
1040 | | const xmlChar *fullattr, |
1041 | 51.9k | const xmlChar *value) { |
1042 | 51.9k | xmlDefAttrsPtr defaults; |
1043 | 51.9k | xmlDefAttr *attr; |
1044 | 51.9k | int len, expandedSize; |
1045 | 51.9k | xmlHashedString name; |
1046 | 51.9k | xmlHashedString prefix; |
1047 | 51.9k | xmlHashedString hvalue; |
1048 | 51.9k | const xmlChar *localname; |
1049 | | |
1050 | | /* |
1051 | | * Allows to detect attribute redefinitions |
1052 | | */ |
1053 | 51.9k | if (ctxt->attsSpecial != NULL) { |
1054 | 38.8k | if (xmlHashLookup2(ctxt->attsSpecial, fullname, fullattr) != NULL) |
1055 | 5.32k | return; |
1056 | 38.8k | } |
1057 | | |
1058 | 46.6k | if (ctxt->attsDefault == NULL) { |
1059 | 15.8k | ctxt->attsDefault = xmlHashCreateDict(10, ctxt->dict); |
1060 | 15.8k | if (ctxt->attsDefault == NULL) |
1061 | 21 | goto mem_error; |
1062 | 15.8k | } |
1063 | | |
1064 | | /* |
1065 | | * split the element name into prefix:localname , the string found |
1066 | | * are within the DTD and then not associated to namespace names. |
1067 | | */ |
1068 | 46.5k | localname = xmlSplitQName3(fullname, &len); |
1069 | 46.5k | if (localname == NULL) { |
1070 | 45.7k | name = xmlDictLookupHashed(ctxt->dict, fullname, -1); |
1071 | 45.7k | prefix.name = NULL; |
1072 | 45.7k | } else { |
1073 | 847 | name = xmlDictLookupHashed(ctxt->dict, localname, -1); |
1074 | 847 | prefix = xmlDictLookupHashed(ctxt->dict, fullname, len); |
1075 | 847 | if (prefix.name == NULL) |
1076 | 0 | goto mem_error; |
1077 | 847 | } |
1078 | 46.5k | if (name.name == NULL) |
1079 | 0 | goto mem_error; |
1080 | | |
1081 | | /* |
1082 | | * make sure there is some storage |
1083 | | */ |
1084 | 46.5k | defaults = xmlHashLookup2(ctxt->attsDefault, name.name, prefix.name); |
1085 | 46.5k | if ((defaults == NULL) || |
1086 | 46.5k | (defaults->nbAttrs >= defaults->maxAttrs)) { |
1087 | 20.7k | xmlDefAttrsPtr temp; |
1088 | 20.7k | int newSize; |
1089 | | |
1090 | 20.7k | if (defaults == NULL) { |
1091 | 16.8k | newSize = 4; |
1092 | 16.8k | } else { |
1093 | 3.81k | if ((defaults->maxAttrs >= XML_MAX_ATTRS) || |
1094 | 3.81k | ((size_t) defaults->maxAttrs > |
1095 | 3.81k | SIZE_MAX / 2 / sizeof(temp[0]) - sizeof(*defaults))) |
1096 | 0 | goto mem_error; |
1097 | | |
1098 | 3.81k | if (defaults->maxAttrs > XML_MAX_ATTRS / 2) |
1099 | 0 | newSize = XML_MAX_ATTRS; |
1100 | 3.81k | else |
1101 | 3.81k | newSize = defaults->maxAttrs * 2; |
1102 | 3.81k | } |
1103 | 20.7k | temp = xmlRealloc(defaults, |
1104 | 20.7k | sizeof(*defaults) + newSize * sizeof(xmlDefAttr)); |
1105 | 20.7k | if (temp == NULL) |
1106 | 1 | goto mem_error; |
1107 | 20.7k | if (defaults == NULL) |
1108 | 16.8k | temp->nbAttrs = 0; |
1109 | 20.7k | temp->maxAttrs = newSize; |
1110 | 20.7k | defaults = temp; |
1111 | 20.7k | if (xmlHashUpdateEntry2(ctxt->attsDefault, name.name, prefix.name, |
1112 | 20.7k | defaults, NULL) < 0) { |
1113 | 0 | xmlFree(defaults); |
1114 | 0 | goto mem_error; |
1115 | 0 | } |
1116 | 20.7k | } |
1117 | | |
1118 | | /* |
1119 | | * Split the attribute name into prefix:localname , the string found |
1120 | | * are within the DTD and hen not associated to namespace names. |
1121 | | */ |
1122 | 46.5k | localname = xmlSplitQName3(fullattr, &len); |
1123 | 46.5k | if (localname == NULL) { |
1124 | 21.5k | name = xmlDictLookupHashed(ctxt->dict, fullattr, -1); |
1125 | 21.5k | prefix.name = NULL; |
1126 | 25.0k | } else { |
1127 | 25.0k | name = xmlDictLookupHashed(ctxt->dict, localname, -1); |
1128 | 25.0k | prefix = xmlDictLookupHashed(ctxt->dict, fullattr, len); |
1129 | 25.0k | if (prefix.name == NULL) |
1130 | 1 | goto mem_error; |
1131 | 25.0k | } |
1132 | 46.5k | if (name.name == NULL) |
1133 | 1 | goto mem_error; |
1134 | | |
1135 | | /* intern the string and precompute the end */ |
1136 | 46.5k | len = strlen((const char *) value); |
1137 | 46.5k | hvalue = xmlDictLookupHashed(ctxt->dict, value, len); |
1138 | 46.5k | if (hvalue.name == NULL) |
1139 | 1 | goto mem_error; |
1140 | | |
1141 | 46.5k | expandedSize = strlen((const char *) name.name); |
1142 | 46.5k | if (prefix.name != NULL) |
1143 | 25.0k | expandedSize += strlen((const char *) prefix.name); |
1144 | 46.5k | expandedSize += len; |
1145 | | |
1146 | 46.5k | attr = &defaults->attrs[defaults->nbAttrs++]; |
1147 | 46.5k | attr->name = name; |
1148 | 46.5k | attr->prefix = prefix; |
1149 | 46.5k | attr->value = hvalue; |
1150 | 46.5k | attr->valueEnd = hvalue.name + len; |
1151 | 46.5k | attr->external = PARSER_EXTERNAL(ctxt); |
1152 | 46.5k | attr->expandedSize = expandedSize; |
1153 | | |
1154 | 46.5k | return; |
1155 | | |
1156 | 25 | mem_error: |
1157 | 25 | xmlErrMemory(ctxt); |
1158 | 25 | } |
1159 | | |
1160 | | /** |
1161 | | * Register this attribute type |
1162 | | * |
1163 | | * @param ctxt an XML parser context |
1164 | | * @param fullname the element fullname |
1165 | | * @param fullattr the attribute fullname |
1166 | | * @param type the attribute type |
1167 | | */ |
1168 | | static void |
1169 | | xmlAddSpecialAttr(xmlParserCtxtPtr ctxt, |
1170 | | const xmlChar *fullname, |
1171 | | const xmlChar *fullattr, |
1172 | | int type) |
1173 | 83.1k | { |
1174 | 83.1k | if (ctxt->attsSpecial == NULL) { |
1175 | 19.6k | ctxt->attsSpecial = xmlHashCreateDict(10, ctxt->dict); |
1176 | 19.6k | if (ctxt->attsSpecial == NULL) |
1177 | 26 | goto mem_error; |
1178 | 19.6k | } |
1179 | | |
1180 | 83.0k | if (PARSER_EXTERNAL(ctxt)) |
1181 | 35.5k | type |= XML_SPECIAL_EXTERNAL; |
1182 | | |
1183 | 83.0k | if (xmlHashAdd2(ctxt->attsSpecial, fullname, fullattr, |
1184 | 83.0k | XML_INT_TO_PTR(type)) < 0) |
1185 | 2 | goto mem_error; |
1186 | 83.0k | return; |
1187 | | |
1188 | 83.0k | mem_error: |
1189 | 28 | xmlErrMemory(ctxt); |
1190 | 28 | } |
1191 | | |
1192 | | /** |
1193 | | * Removes CDATA attributes from the special attribute table |
1194 | | */ |
1195 | | static void |
1196 | | xmlCleanSpecialAttrCallback(void *payload, void *data, |
1197 | | const xmlChar *fullname, const xmlChar *fullattr, |
1198 | 76.1k | const xmlChar *unused ATTRIBUTE_UNUSED) { |
1199 | 76.1k | xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) data; |
1200 | | |
1201 | 76.1k | if (XML_PTR_TO_INT(payload) == XML_ATTRIBUTE_CDATA) { |
1202 | 2.73k | xmlHashRemoveEntry2(ctxt->attsSpecial, fullname, fullattr, NULL); |
1203 | 2.73k | } |
1204 | 76.1k | } |
1205 | | |
1206 | | /** |
1207 | | * Trim the list of attributes defined to remove all those of type |
1208 | | * CDATA as they are not special. This call should be done when finishing |
1209 | | * to parse the DTD and before starting to parse the document root. |
1210 | | * |
1211 | | * @param ctxt an XML parser context |
1212 | | */ |
1213 | | static void |
1214 | | xmlCleanSpecialAttr(xmlParserCtxtPtr ctxt) |
1215 | 148k | { |
1216 | 148k | if (ctxt->attsSpecial == NULL) |
1217 | 129k | return; |
1218 | | |
1219 | 19.6k | xmlHashScanFull(ctxt->attsSpecial, xmlCleanSpecialAttrCallback, ctxt); |
1220 | | |
1221 | 19.6k | if (xmlHashSize(ctxt->attsSpecial) == 0) { |
1222 | 270 | xmlHashFree(ctxt->attsSpecial, NULL); |
1223 | 270 | ctxt->attsSpecial = NULL; |
1224 | 270 | } |
1225 | 19.6k | } |
1226 | | |
1227 | | /** |
1228 | | * Checks that the value conforms to the LanguageID production: |
1229 | | * |
1230 | | * @deprecated Internal function, do not use. |
1231 | | * |
1232 | | * NOTE: this is somewhat deprecated, those productions were removed from |
1233 | | * the XML Second edition. |
1234 | | * |
1235 | | * [33] LanguageID ::= Langcode ('-' Subcode)* |
1236 | | * [34] Langcode ::= ISO639Code | IanaCode | UserCode |
1237 | | * [35] ISO639Code ::= ([a-z] | [A-Z]) ([a-z] | [A-Z]) |
1238 | | * [36] IanaCode ::= ('i' | 'I') '-' ([a-z] | [A-Z])+ |
1239 | | * [37] UserCode ::= ('x' | 'X') '-' ([a-z] | [A-Z])+ |
1240 | | * [38] Subcode ::= ([a-z] | [A-Z])+ |
1241 | | * |
1242 | | * The current REC reference the successors of RFC 1766, currently 5646 |
1243 | | * |
1244 | | * http://www.rfc-editor.org/rfc/rfc5646.txt |
1245 | | * |
1246 | | * langtag = language |
1247 | | * ["-" script] |
1248 | | * ["-" region] |
1249 | | * *("-" variant) |
1250 | | * *("-" extension) |
1251 | | * ["-" privateuse] |
1252 | | * language = 2*3ALPHA ; shortest ISO 639 code |
1253 | | * ["-" extlang] ; sometimes followed by |
1254 | | * ; extended language subtags |
1255 | | * / 4ALPHA ; or reserved for future use |
1256 | | * / 5*8ALPHA ; or registered language subtag |
1257 | | * |
1258 | | * extlang = 3ALPHA ; selected ISO 639 codes |
1259 | | * *2("-" 3ALPHA) ; permanently reserved |
1260 | | * |
1261 | | * script = 4ALPHA ; ISO 15924 code |
1262 | | * |
1263 | | * region = 2ALPHA ; ISO 3166-1 code |
1264 | | * / 3DIGIT ; UN M.49 code |
1265 | | * |
1266 | | * variant = 5*8alphanum ; registered variants |
1267 | | * / (DIGIT 3alphanum) |
1268 | | * |
1269 | | * extension = singleton 1*("-" (2*8alphanum)) |
1270 | | * |
1271 | | * ; Single alphanumerics |
1272 | | * ; "x" reserved for private use |
1273 | | * singleton = DIGIT ; 0 - 9 |
1274 | | * / %x41-57 ; A - W |
1275 | | * / %x59-5A ; Y - Z |
1276 | | * / %x61-77 ; a - w |
1277 | | * / %x79-7A ; y - z |
1278 | | * |
1279 | | * it sounds right to still allow Irregular i-xxx IANA and user codes too |
1280 | | * The parser below doesn't try to cope with extension or privateuse |
1281 | | * that could be added but that's not interoperable anyway |
1282 | | * |
1283 | | * @param lang pointer to the string value |
1284 | | * @returns 1 if correct 0 otherwise |
1285 | | **/ |
1286 | | int |
1287 | | xmlCheckLanguageID(const xmlChar * lang) |
1288 | 0 | { |
1289 | 0 | const xmlChar *cur = lang, *nxt; |
1290 | |
|
1291 | 0 | if (cur == NULL) |
1292 | 0 | return (0); |
1293 | 0 | if (((cur[0] == 'i') && (cur[1] == '-')) || |
1294 | 0 | ((cur[0] == 'I') && (cur[1] == '-')) || |
1295 | 0 | ((cur[0] == 'x') && (cur[1] == '-')) || |
1296 | 0 | ((cur[0] == 'X') && (cur[1] == '-'))) { |
1297 | | /* |
1298 | | * Still allow IANA code and user code which were coming |
1299 | | * from the previous version of the XML-1.0 specification |
1300 | | * it's deprecated but we should not fail |
1301 | | */ |
1302 | 0 | cur += 2; |
1303 | 0 | while (((cur[0] >= 'A') && (cur[0] <= 'Z')) || |
1304 | 0 | ((cur[0] >= 'a') && (cur[0] <= 'z'))) |
1305 | 0 | cur++; |
1306 | 0 | return(cur[0] == 0); |
1307 | 0 | } |
1308 | 0 | nxt = cur; |
1309 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1310 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1311 | 0 | nxt++; |
1312 | 0 | if (nxt - cur >= 4) { |
1313 | | /* |
1314 | | * Reserved |
1315 | | */ |
1316 | 0 | if ((nxt - cur > 8) || (nxt[0] != 0)) |
1317 | 0 | return(0); |
1318 | 0 | return(1); |
1319 | 0 | } |
1320 | 0 | if (nxt - cur < 2) |
1321 | 0 | return(0); |
1322 | | /* we got an ISO 639 code */ |
1323 | 0 | if (nxt[0] == 0) |
1324 | 0 | return(1); |
1325 | 0 | if (nxt[0] != '-') |
1326 | 0 | return(0); |
1327 | | |
1328 | 0 | nxt++; |
1329 | 0 | cur = nxt; |
1330 | | /* now we can have extlang or script or region or variant */ |
1331 | 0 | if ((nxt[0] >= '0') && (nxt[0] <= '9')) |
1332 | 0 | goto region_m49; |
1333 | | |
1334 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1335 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1336 | 0 | nxt++; |
1337 | 0 | if (nxt - cur == 4) |
1338 | 0 | goto script; |
1339 | 0 | if (nxt - cur == 2) |
1340 | 0 | goto region; |
1341 | 0 | if ((nxt - cur >= 5) && (nxt - cur <= 8)) |
1342 | 0 | goto variant; |
1343 | 0 | if (nxt - cur != 3) |
1344 | 0 | return(0); |
1345 | | /* we parsed an extlang */ |
1346 | 0 | if (nxt[0] == 0) |
1347 | 0 | return(1); |
1348 | 0 | if (nxt[0] != '-') |
1349 | 0 | return(0); |
1350 | | |
1351 | 0 | nxt++; |
1352 | 0 | cur = nxt; |
1353 | | /* now we can have script or region or variant */ |
1354 | 0 | if ((nxt[0] >= '0') && (nxt[0] <= '9')) |
1355 | 0 | goto region_m49; |
1356 | | |
1357 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1358 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1359 | 0 | nxt++; |
1360 | 0 | if (nxt - cur == 2) |
1361 | 0 | goto region; |
1362 | 0 | if ((nxt - cur >= 5) && (nxt - cur <= 8)) |
1363 | 0 | goto variant; |
1364 | 0 | if (nxt - cur != 4) |
1365 | 0 | return(0); |
1366 | | /* we parsed a script */ |
1367 | 0 | script: |
1368 | 0 | if (nxt[0] == 0) |
1369 | 0 | return(1); |
1370 | 0 | if (nxt[0] != '-') |
1371 | 0 | return(0); |
1372 | | |
1373 | 0 | nxt++; |
1374 | 0 | cur = nxt; |
1375 | | /* now we can have region or variant */ |
1376 | 0 | if ((nxt[0] >= '0') && (nxt[0] <= '9')) |
1377 | 0 | goto region_m49; |
1378 | | |
1379 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1380 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1381 | 0 | nxt++; |
1382 | |
|
1383 | 0 | if ((nxt - cur >= 5) && (nxt - cur <= 8)) |
1384 | 0 | goto variant; |
1385 | 0 | if (nxt - cur != 2) |
1386 | 0 | return(0); |
1387 | | /* we parsed a region */ |
1388 | 0 | region: |
1389 | 0 | if (nxt[0] == 0) |
1390 | 0 | return(1); |
1391 | 0 | if (nxt[0] != '-') |
1392 | 0 | return(0); |
1393 | | |
1394 | 0 | nxt++; |
1395 | 0 | cur = nxt; |
1396 | | /* now we can just have a variant */ |
1397 | 0 | while (((nxt[0] >= 'A') && (nxt[0] <= 'Z')) || |
1398 | 0 | ((nxt[0] >= 'a') && (nxt[0] <= 'z'))) |
1399 | 0 | nxt++; |
1400 | |
|
1401 | 0 | if ((nxt - cur < 5) || (nxt - cur > 8)) |
1402 | 0 | return(0); |
1403 | | |
1404 | | /* we parsed a variant */ |
1405 | 0 | variant: |
1406 | 0 | if (nxt[0] == 0) |
1407 | 0 | return(1); |
1408 | 0 | if (nxt[0] != '-') |
1409 | 0 | return(0); |
1410 | | /* extensions and private use subtags not checked */ |
1411 | 0 | return (1); |
1412 | | |
1413 | 0 | region_m49: |
1414 | 0 | if (((nxt[1] >= '0') && (nxt[1] <= '9')) && |
1415 | 0 | ((nxt[2] >= '0') && (nxt[2] <= '9'))) { |
1416 | 0 | nxt += 3; |
1417 | 0 | goto region; |
1418 | 0 | } |
1419 | 0 | return(0); |
1420 | 0 | } |
1421 | | |
1422 | | /************************************************************************ |
1423 | | * * |
1424 | | * Parser stacks related functions and macros * |
1425 | | * * |
1426 | | ************************************************************************/ |
1427 | | |
1428 | | static xmlChar * |
1429 | | xmlParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar **str); |
1430 | | |
1431 | | /** |
1432 | | * Create a new namespace database. |
1433 | | * |
1434 | | * @returns the new obejct. |
1435 | | */ |
1436 | | xmlParserNsData * |
1437 | 1.00M | xmlParserNsCreate(void) { |
1438 | 1.00M | xmlParserNsData *nsdb = xmlMalloc(sizeof(*nsdb)); |
1439 | | |
1440 | 1.00M | if (nsdb == NULL) |
1441 | 43 | return(NULL); |
1442 | 1.00M | memset(nsdb, 0, sizeof(*nsdb)); |
1443 | 1.00M | nsdb->defaultNsIndex = INT_MAX; |
1444 | | |
1445 | 1.00M | return(nsdb); |
1446 | 1.00M | } |
1447 | | |
1448 | | /** |
1449 | | * Free a namespace database. |
1450 | | * |
1451 | | * @param nsdb namespace database |
1452 | | */ |
1453 | | void |
1454 | 1.00M | xmlParserNsFree(xmlParserNsData *nsdb) { |
1455 | 1.00M | if (nsdb == NULL) |
1456 | 0 | return; |
1457 | | |
1458 | 1.00M | xmlFree(nsdb->extra); |
1459 | 1.00M | xmlFree(nsdb->hash); |
1460 | 1.00M | xmlFree(nsdb); |
1461 | 1.00M | } |
1462 | | |
1463 | | /** |
1464 | | * Reset a namespace database. |
1465 | | * |
1466 | | * @param nsdb namespace database |
1467 | | */ |
1468 | | static void |
1469 | 0 | xmlParserNsReset(xmlParserNsData *nsdb) { |
1470 | 0 | if (nsdb == NULL) |
1471 | 0 | return; |
1472 | | |
1473 | 0 | nsdb->hashElems = 0; |
1474 | 0 | nsdb->elementId = 0; |
1475 | 0 | nsdb->defaultNsIndex = INT_MAX; |
1476 | |
|
1477 | 0 | if (nsdb->hash) |
1478 | 0 | memset(nsdb->hash, 0, nsdb->hashSize * sizeof(nsdb->hash[0])); |
1479 | 0 | } |
1480 | | |
1481 | | /** |
1482 | | * Signal that a new element has started. |
1483 | | * |
1484 | | * @param nsdb namespace database |
1485 | | * @returns 0 on success, -1 if the element counter overflowed. |
1486 | | */ |
1487 | | static int |
1488 | 8.28M | xmlParserNsStartElement(xmlParserNsData *nsdb) { |
1489 | 8.28M | if (nsdb->elementId == UINT_MAX) |
1490 | 0 | return(-1); |
1491 | 8.28M | nsdb->elementId++; |
1492 | | |
1493 | 8.28M | return(0); |
1494 | 8.28M | } |
1495 | | |
1496 | | /** |
1497 | | * Lookup namespace with given prefix. If `bucketPtr` is non-NULL, it will |
1498 | | * be set to the matching bucket, or the first empty bucket if no match |
1499 | | * was found. |
1500 | | * |
1501 | | * @param ctxt parser context |
1502 | | * @param prefix namespace prefix |
1503 | | * @param bucketPtr optional bucket (return value) |
1504 | | * @returns the namespace index on success, INT_MAX if no namespace was |
1505 | | * found. |
1506 | | */ |
1507 | | static int |
1508 | | xmlParserNsLookup(xmlParserCtxtPtr ctxt, const xmlHashedString *prefix, |
1509 | 13.1M | xmlParserNsBucket **bucketPtr) { |
1510 | 13.1M | xmlParserNsBucket *bucket, *tombstone; |
1511 | 13.1M | unsigned index, hashValue; |
1512 | | |
1513 | 13.1M | if (prefix->name == NULL) |
1514 | 4.15M | return(ctxt->nsdb->defaultNsIndex); |
1515 | | |
1516 | 9.04M | if (ctxt->nsdb->hashSize == 0) |
1517 | 715k | return(INT_MAX); |
1518 | | |
1519 | 8.32M | hashValue = prefix->hashValue; |
1520 | 8.32M | index = hashValue & (ctxt->nsdb->hashSize - 1); |
1521 | 8.32M | bucket = &ctxt->nsdb->hash[index]; |
1522 | 8.32M | tombstone = NULL; |
1523 | | |
1524 | 8.94M | while (bucket->hashValue) { |
1525 | 8.37M | if (bucket->index == INT_MAX) { |
1526 | 161k | if (tombstone == NULL) |
1527 | 147k | tombstone = bucket; |
1528 | 8.21M | } else if (bucket->hashValue == hashValue) { |
1529 | 7.76M | if (ctxt->nsTab[bucket->index * 2] == prefix->name) { |
1530 | 7.76M | if (bucketPtr != NULL) |
1531 | 1.08M | *bucketPtr = bucket; |
1532 | 7.76M | return(bucket->index); |
1533 | 7.76M | } |
1534 | 7.76M | } |
1535 | | |
1536 | 614k | index++; |
1537 | 614k | bucket++; |
1538 | 614k | if (index == ctxt->nsdb->hashSize) { |
1539 | 20.4k | index = 0; |
1540 | 20.4k | bucket = ctxt->nsdb->hash; |
1541 | 20.4k | } |
1542 | 614k | } |
1543 | | |
1544 | 564k | if (bucketPtr != NULL) |
1545 | 172k | *bucketPtr = tombstone ? tombstone : bucket; |
1546 | 564k | return(INT_MAX); |
1547 | 8.32M | } |
1548 | | |
1549 | | /** |
1550 | | * Lookup namespace URI with given prefix. |
1551 | | * |
1552 | | * @param ctxt parser context |
1553 | | * @param prefix namespace prefix |
1554 | | * @returns the namespace URI on success, NULL if no namespace was found. |
1555 | | */ |
1556 | | static const xmlChar * |
1557 | 7.71M | xmlParserNsLookupUri(xmlParserCtxtPtr ctxt, const xmlHashedString *prefix) { |
1558 | 7.71M | const xmlChar *ret; |
1559 | 7.71M | int nsIndex; |
1560 | | |
1561 | 7.71M | if (prefix->name == ctxt->str_xml) |
1562 | 17.9k | return(ctxt->str_xml_ns); |
1563 | | |
1564 | | /* |
1565 | | * minNsIndex is used when building an entity tree. We must |
1566 | | * ignore namespaces declared outside the entity. |
1567 | | */ |
1568 | 7.69M | nsIndex = xmlParserNsLookup(ctxt, prefix, NULL); |
1569 | 7.69M | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) |
1570 | 3.74M | return(NULL); |
1571 | | |
1572 | 3.94M | ret = ctxt->nsTab[nsIndex * 2 + 1]; |
1573 | 3.94M | if (ret[0] == 0) |
1574 | 19.0k | ret = NULL; |
1575 | 3.94M | return(ret); |
1576 | 7.69M | } |
1577 | | |
1578 | | /** |
1579 | | * Lookup extra data for the given prefix. This returns data stored |
1580 | | * with xmlParserNsUdpateSax. |
1581 | | * |
1582 | | * @param ctxt parser context |
1583 | | * @param prefix namespace prefix |
1584 | | * @returns the data on success, NULL if no namespace was found. |
1585 | | */ |
1586 | | void * |
1587 | 2.88M | xmlParserNsLookupSax(xmlParserCtxt *ctxt, const xmlChar *prefix) { |
1588 | 2.88M | xmlHashedString hprefix; |
1589 | 2.88M | int nsIndex; |
1590 | | |
1591 | 2.88M | if (prefix == ctxt->str_xml) |
1592 | 123k | return(NULL); |
1593 | | |
1594 | 2.76M | hprefix.name = prefix; |
1595 | 2.76M | if (prefix != NULL) |
1596 | 2.55M | hprefix.hashValue = xmlDictComputeHash(ctxt->dict, prefix); |
1597 | 207k | else |
1598 | 207k | hprefix.hashValue = 0; |
1599 | 2.76M | nsIndex = xmlParserNsLookup(ctxt, &hprefix, NULL); |
1600 | 2.76M | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) |
1601 | 0 | return(NULL); |
1602 | | |
1603 | 2.76M | return(ctxt->nsdb->extra[nsIndex].saxData); |
1604 | 2.76M | } |
1605 | | |
1606 | | /** |
1607 | | * Sets or updates extra data for the given prefix. This value will be |
1608 | | * returned by xmlParserNsLookupSax as long as the namespace with the |
1609 | | * given prefix is in scope. |
1610 | | * |
1611 | | * @param ctxt parser context |
1612 | | * @param prefix namespace prefix |
1613 | | * @param saxData extra data for SAX handler |
1614 | | * @returns the data on success, NULL if no namespace was found. |
1615 | | */ |
1616 | | int |
1617 | | xmlParserNsUpdateSax(xmlParserCtxt *ctxt, const xmlChar *prefix, |
1618 | 743k | void *saxData) { |
1619 | 743k | xmlHashedString hprefix; |
1620 | 743k | int nsIndex; |
1621 | | |
1622 | 743k | if (prefix == ctxt->str_xml) |
1623 | 0 | return(-1); |
1624 | | |
1625 | 743k | hprefix.name = prefix; |
1626 | 743k | if (prefix != NULL) |
1627 | 720k | hprefix.hashValue = xmlDictComputeHash(ctxt->dict, prefix); |
1628 | 23.2k | else |
1629 | 23.2k | hprefix.hashValue = 0; |
1630 | 743k | nsIndex = xmlParserNsLookup(ctxt, &hprefix, NULL); |
1631 | 743k | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) |
1632 | 0 | return(-1); |
1633 | | |
1634 | 743k | ctxt->nsdb->extra[nsIndex].saxData = saxData; |
1635 | 743k | return(0); |
1636 | 743k | } |
1637 | | |
1638 | | /** |
1639 | | * Grows the namespace tables. |
1640 | | * |
1641 | | * @param ctxt parser context |
1642 | | * @returns 0 on success, -1 if a memory allocation failed. |
1643 | | */ |
1644 | | static int |
1645 | 769k | xmlParserNsGrow(xmlParserCtxtPtr ctxt) { |
1646 | 769k | const xmlChar **table; |
1647 | 769k | xmlParserNsExtra *extra; |
1648 | 769k | int newSize; |
1649 | | |
1650 | 769k | newSize = xmlGrowCapacity(ctxt->nsMax, |
1651 | 769k | sizeof(table[0]) + sizeof(extra[0]), |
1652 | 769k | 16, XML_MAX_ITEMS); |
1653 | 769k | if (newSize < 0) |
1654 | 0 | goto error; |
1655 | | |
1656 | 769k | table = xmlRealloc(ctxt->nsTab, 2 * newSize * sizeof(table[0])); |
1657 | 769k | if (table == NULL) |
1658 | 75 | goto error; |
1659 | 769k | ctxt->nsTab = table; |
1660 | | |
1661 | 769k | extra = xmlRealloc(ctxt->nsdb->extra, newSize * sizeof(extra[0])); |
1662 | 769k | if (extra == NULL) |
1663 | 29 | goto error; |
1664 | 769k | ctxt->nsdb->extra = extra; |
1665 | | |
1666 | 769k | ctxt->nsMax = newSize; |
1667 | 769k | return(0); |
1668 | | |
1669 | 104 | error: |
1670 | 104 | xmlErrMemory(ctxt); |
1671 | 104 | return(-1); |
1672 | 769k | } |
1673 | | |
1674 | | /** |
1675 | | * Push a new namespace on the table. |
1676 | | * |
1677 | | * @param ctxt parser context |
1678 | | * @param prefix prefix with hash value |
1679 | | * @param uri uri with hash value |
1680 | | * @param saxData extra data for SAX handler |
1681 | | * @param defAttr whether the namespace comes from a default attribute |
1682 | | * @returns 1 if the namespace was pushed, 0 if the namespace was ignored, |
1683 | | * -1 if a memory allocation failed. |
1684 | | */ |
1685 | | static int |
1686 | | xmlParserNsPush(xmlParserCtxtPtr ctxt, const xmlHashedString *prefix, |
1687 | 1.03M | const xmlHashedString *uri, void *saxData, int defAttr) { |
1688 | 1.03M | xmlParserNsBucket *bucket = NULL; |
1689 | 1.03M | xmlParserNsExtra *extra; |
1690 | 1.03M | const xmlChar **ns; |
1691 | 1.03M | unsigned hashValue, nsIndex, oldIndex; |
1692 | | |
1693 | 1.03M | if ((prefix != NULL) && (prefix->name == ctxt->str_xml)) |
1694 | 302 | return(0); |
1695 | | |
1696 | 1.02M | if ((ctxt->nsNr >= ctxt->nsMax) && (xmlParserNsGrow(ctxt) < 0)) { |
1697 | 104 | xmlErrMemory(ctxt); |
1698 | 104 | return(-1); |
1699 | 104 | } |
1700 | | |
1701 | | /* |
1702 | | * Default namespace and 'xml' namespace |
1703 | | */ |
1704 | 1.02M | if ((prefix == NULL) || (prefix->name == NULL)) { |
1705 | 80.9k | oldIndex = ctxt->nsdb->defaultNsIndex; |
1706 | | |
1707 | 80.9k | if (oldIndex != INT_MAX) { |
1708 | 49.8k | extra = &ctxt->nsdb->extra[oldIndex]; |
1709 | | |
1710 | 49.8k | if (extra->elementId == ctxt->nsdb->elementId) { |
1711 | 2.09k | if (defAttr == 0) |
1712 | 768 | xmlErrAttributeDup(ctxt, NULL, BAD_CAST "xmlns"); |
1713 | 2.09k | return(0); |
1714 | 2.09k | } |
1715 | | |
1716 | 47.7k | if ((ctxt->options & XML_PARSE_NSCLEAN) && |
1717 | 47.7k | (uri->name == ctxt->nsTab[oldIndex * 2 + 1])) |
1718 | 0 | return(0); |
1719 | 47.7k | } |
1720 | | |
1721 | 78.8k | ctxt->nsdb->defaultNsIndex = ctxt->nsNr; |
1722 | 78.8k | goto populate_entry; |
1723 | 80.9k | } |
1724 | | |
1725 | | /* |
1726 | | * Hash table lookup |
1727 | | */ |
1728 | 948k | oldIndex = xmlParserNsLookup(ctxt, prefix, &bucket); |
1729 | 948k | if (oldIndex != INT_MAX) { |
1730 | 164k | extra = &ctxt->nsdb->extra[oldIndex]; |
1731 | | |
1732 | | /* |
1733 | | * Check for duplicate definitions on the same element. |
1734 | | */ |
1735 | 164k | if (extra->elementId == ctxt->nsdb->elementId) { |
1736 | 5.68k | if (defAttr == 0) |
1737 | 5.48k | xmlErrAttributeDup(ctxt, BAD_CAST "xmlns", prefix->name); |
1738 | 5.68k | return(0); |
1739 | 5.68k | } |
1740 | | |
1741 | 158k | if ((ctxt->options & XML_PARSE_NSCLEAN) && |
1742 | 158k | (uri->name == ctxt->nsTab[bucket->index * 2 + 1])) |
1743 | 0 | return(0); |
1744 | | |
1745 | 158k | bucket->index = ctxt->nsNr; |
1746 | 158k | goto populate_entry; |
1747 | 158k | } |
1748 | | |
1749 | | /* |
1750 | | * Insert new bucket |
1751 | | */ |
1752 | | |
1753 | 784k | hashValue = prefix->hashValue; |
1754 | | |
1755 | | /* |
1756 | | * Grow hash table, 50% fill factor |
1757 | | */ |
1758 | 784k | if (ctxt->nsdb->hashElems + 1 > ctxt->nsdb->hashSize / 2) { |
1759 | 615k | xmlParserNsBucket *newHash; |
1760 | 615k | unsigned newSize, i, index; |
1761 | | |
1762 | 615k | if (ctxt->nsdb->hashSize > UINT_MAX / 2) { |
1763 | 0 | xmlErrMemory(ctxt); |
1764 | 0 | return(-1); |
1765 | 0 | } |
1766 | 615k | newSize = ctxt->nsdb->hashSize ? ctxt->nsdb->hashSize * 2 : 16; |
1767 | 615k | newHash = xmlMalloc(newSize * sizeof(newHash[0])); |
1768 | 615k | if (newHash == NULL) { |
1769 | 46 | xmlErrMemory(ctxt); |
1770 | 46 | return(-1); |
1771 | 46 | } |
1772 | 615k | memset(newHash, 0, newSize * sizeof(newHash[0])); |
1773 | | |
1774 | 798k | for (i = 0; i < ctxt->nsdb->hashSize; i++) { |
1775 | 183k | unsigned hv = ctxt->nsdb->hash[i].hashValue; |
1776 | 183k | unsigned newIndex; |
1777 | | |
1778 | 183k | if ((hv == 0) || (ctxt->nsdb->hash[i].index == INT_MAX)) |
1779 | 172k | continue; |
1780 | 10.7k | newIndex = hv & (newSize - 1); |
1781 | | |
1782 | 15.4k | while (newHash[newIndex].hashValue != 0) { |
1783 | 4.69k | newIndex++; |
1784 | 4.69k | if (newIndex == newSize) |
1785 | 1.32k | newIndex = 0; |
1786 | 4.69k | } |
1787 | | |
1788 | 10.7k | newHash[newIndex] = ctxt->nsdb->hash[i]; |
1789 | 10.7k | } |
1790 | | |
1791 | 615k | xmlFree(ctxt->nsdb->hash); |
1792 | 615k | ctxt->nsdb->hash = newHash; |
1793 | 615k | ctxt->nsdb->hashSize = newSize; |
1794 | | |
1795 | | /* |
1796 | | * Relookup |
1797 | | */ |
1798 | 615k | index = hashValue & (newSize - 1); |
1799 | | |
1800 | 618k | while (newHash[index].hashValue != 0) { |
1801 | 3.16k | index++; |
1802 | 3.16k | if (index == newSize) |
1803 | 490 | index = 0; |
1804 | 3.16k | } |
1805 | | |
1806 | 615k | bucket = &newHash[index]; |
1807 | 615k | } |
1808 | | |
1809 | 784k | bucket->hashValue = hashValue; |
1810 | 784k | bucket->index = ctxt->nsNr; |
1811 | 784k | ctxt->nsdb->hashElems++; |
1812 | 784k | oldIndex = INT_MAX; |
1813 | | |
1814 | 1.02M | populate_entry: |
1815 | 1.02M | nsIndex = ctxt->nsNr; |
1816 | | |
1817 | 1.02M | ns = &ctxt->nsTab[nsIndex * 2]; |
1818 | 1.02M | ns[0] = prefix ? prefix->name : NULL; |
1819 | 1.02M | ns[1] = uri->name; |
1820 | | |
1821 | 1.02M | extra = &ctxt->nsdb->extra[nsIndex]; |
1822 | 1.02M | extra->saxData = saxData; |
1823 | 1.02M | extra->prefixHashValue = prefix ? prefix->hashValue : 0; |
1824 | 1.02M | extra->uriHashValue = uri->hashValue; |
1825 | 1.02M | extra->elementId = ctxt->nsdb->elementId; |
1826 | 1.02M | extra->oldIndex = oldIndex; |
1827 | | |
1828 | 1.02M | ctxt->nsNr++; |
1829 | | |
1830 | 1.02M | return(1); |
1831 | 784k | } |
1832 | | |
1833 | | /** |
1834 | | * Pops the top `nr` namespaces and restores the hash table. |
1835 | | * |
1836 | | * @param ctxt an XML parser context |
1837 | | * @param nr the number to pop |
1838 | | * @returns the number of namespaces popped. |
1839 | | */ |
1840 | | static int |
1841 | | xmlParserNsPop(xmlParserCtxtPtr ctxt, int nr) |
1842 | 739k | { |
1843 | 739k | int i; |
1844 | | |
1845 | | /* assert(nr <= ctxt->nsNr); */ |
1846 | | |
1847 | 1.73M | for (i = ctxt->nsNr - 1; i >= ctxt->nsNr - nr; i--) { |
1848 | 997k | const xmlChar *prefix = ctxt->nsTab[i * 2]; |
1849 | 997k | xmlParserNsExtra *extra = &ctxt->nsdb->extra[i]; |
1850 | | |
1851 | 997k | if (prefix == NULL) { |
1852 | 73.3k | ctxt->nsdb->defaultNsIndex = extra->oldIndex; |
1853 | 924k | } else { |
1854 | 924k | xmlHashedString hprefix; |
1855 | 924k | xmlParserNsBucket *bucket = NULL; |
1856 | | |
1857 | 924k | hprefix.name = prefix; |
1858 | 924k | hprefix.hashValue = extra->prefixHashValue; |
1859 | 924k | xmlParserNsLookup(ctxt, &hprefix, &bucket); |
1860 | | /* assert(bucket && bucket->hashValue); */ |
1861 | 924k | bucket->index = extra->oldIndex; |
1862 | 924k | } |
1863 | 997k | } |
1864 | | |
1865 | 739k | ctxt->nsNr -= nr; |
1866 | 739k | return(nr); |
1867 | 739k | } |
1868 | | |
1869 | | static int |
1870 | 887k | xmlCtxtGrowAttrs(xmlParserCtxtPtr ctxt) { |
1871 | 887k | const xmlChar **atts; |
1872 | 887k | unsigned *attallocs; |
1873 | 887k | int newSize; |
1874 | | |
1875 | 887k | newSize = xmlGrowCapacity(ctxt->maxatts / 5, |
1876 | 887k | sizeof(atts[0]) * 5 + sizeof(attallocs[0]), |
1877 | 887k | 10, XML_MAX_ATTRS); |
1878 | 887k | if (newSize < 0) { |
1879 | 0 | xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, |
1880 | 0 | "Maximum number of attributes exceeded"); |
1881 | 0 | return(-1); |
1882 | 0 | } |
1883 | | |
1884 | 887k | atts = xmlRealloc(ctxt->atts, newSize * sizeof(atts[0]) * 5); |
1885 | 887k | if (atts == NULL) |
1886 | 60 | goto mem_error; |
1887 | 887k | ctxt->atts = atts; |
1888 | | |
1889 | 887k | attallocs = xmlRealloc(ctxt->attallocs, |
1890 | 887k | newSize * sizeof(attallocs[0])); |
1891 | 887k | if (attallocs == NULL) |
1892 | 45 | goto mem_error; |
1893 | 887k | ctxt->attallocs = attallocs; |
1894 | | |
1895 | 887k | ctxt->maxatts = newSize * 5; |
1896 | | |
1897 | 887k | return(0); |
1898 | | |
1899 | 105 | mem_error: |
1900 | 105 | xmlErrMemory(ctxt); |
1901 | 105 | return(-1); |
1902 | 887k | } |
1903 | | |
1904 | | /** |
1905 | | * Pushes a new parser input on top of the input stack |
1906 | | * |
1907 | | * @param ctxt an XML parser context |
1908 | | * @param value the parser input |
1909 | | * @returns -1 in case of error, the index in the stack otherwise |
1910 | | */ |
1911 | | int |
1912 | | xmlCtxtPushInput(xmlParserCtxt *ctxt, xmlParserInput *value) |
1913 | 907k | { |
1914 | 907k | char *directory = NULL; |
1915 | 907k | int maxDepth; |
1916 | | |
1917 | 907k | if ((ctxt == NULL) || (value == NULL)) |
1918 | 9.01k | return(-1); |
1919 | | |
1920 | 898k | maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
1921 | | |
1922 | 898k | if (ctxt->inputNr >= ctxt->inputMax) { |
1923 | 33.0k | xmlParserInputPtr *tmp; |
1924 | 33.0k | int newSize; |
1925 | | |
1926 | 33.0k | newSize = xmlGrowCapacity(ctxt->inputMax, sizeof(tmp[0]), |
1927 | 33.0k | 5, maxDepth); |
1928 | 33.0k | if (newSize < 0) { |
1929 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, |
1930 | 0 | "Maximum entity nesting depth exceeded"); |
1931 | 0 | xmlHaltParser(ctxt); |
1932 | 0 | return(-1); |
1933 | 0 | } |
1934 | 33.0k | tmp = xmlRealloc(ctxt->inputTab, newSize * sizeof(tmp[0])); |
1935 | 33.0k | if (tmp == NULL) { |
1936 | 5 | xmlErrMemory(ctxt); |
1937 | 5 | return(-1); |
1938 | 5 | } |
1939 | 33.0k | ctxt->inputTab = tmp; |
1940 | 33.0k | ctxt->inputMax = newSize; |
1941 | 33.0k | } |
1942 | | |
1943 | 898k | if ((ctxt->inputNr == 0) && (value->filename != NULL)) { |
1944 | 66.4k | directory = xmlParserGetDirectory(value->filename); |
1945 | 66.4k | if (directory == NULL) { |
1946 | 0 | xmlErrMemory(ctxt); |
1947 | 0 | return(-1); |
1948 | 0 | } |
1949 | 66.4k | } |
1950 | | |
1951 | 898k | if (ctxt->input_id >= INT_MAX) { |
1952 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_RESOURCE_LIMIT, "Input ID overflow\n"); |
1953 | 0 | return(-1); |
1954 | 0 | } |
1955 | | |
1956 | 898k | ctxt->inputTab[ctxt->inputNr] = value; |
1957 | 898k | ctxt->input = value; |
1958 | | |
1959 | 898k | if (ctxt->inputNr == 0) { |
1960 | 857k | xmlFree(ctxt->directory); |
1961 | 857k | ctxt->directory = directory; |
1962 | 857k | } |
1963 | | |
1964 | | /* |
1965 | | * The input ID is unused internally, but there are entity |
1966 | | * loaders in downstream code that detect the main document |
1967 | | * by checking for "input_id == 1". |
1968 | | */ |
1969 | 898k | value->id = ctxt->input_id++; |
1970 | | |
1971 | 898k | return(ctxt->inputNr++); |
1972 | 898k | } |
1973 | | |
1974 | | /** |
1975 | | * Pops the top parser input from the input stack |
1976 | | * |
1977 | | * @param ctxt an XML parser context |
1978 | | * @returns the input just removed |
1979 | | */ |
1980 | | xmlParserInput * |
1981 | | xmlCtxtPopInput(xmlParserCtxt *ctxt) |
1982 | 2.89M | { |
1983 | 2.89M | xmlParserInputPtr ret; |
1984 | | |
1985 | 2.89M | if (ctxt == NULL) |
1986 | 0 | return(NULL); |
1987 | 2.89M | if (ctxt->inputNr <= 0) |
1988 | 2.01M | return (NULL); |
1989 | 883k | ctxt->inputNr--; |
1990 | 883k | if (ctxt->inputNr > 0) |
1991 | 40.2k | ctxt->input = ctxt->inputTab[ctxt->inputNr - 1]; |
1992 | 843k | else |
1993 | 843k | ctxt->input = NULL; |
1994 | 883k | ret = ctxt->inputTab[ctxt->inputNr]; |
1995 | 883k | ctxt->inputTab[ctxt->inputNr] = NULL; |
1996 | 883k | return (ret); |
1997 | 2.89M | } |
1998 | | |
1999 | | /** |
2000 | | * Pushes a new element node on top of the node stack |
2001 | | * |
2002 | | * @deprecated Internal function, do not use. |
2003 | | * |
2004 | | * @param ctxt an XML parser context |
2005 | | * @param value the element node |
2006 | | * @returns -1 in case of error, the index in the stack otherwise |
2007 | | */ |
2008 | | int |
2009 | | nodePush(xmlParserCtxt *ctxt, xmlNode *value) |
2010 | 6.56M | { |
2011 | 6.56M | if (ctxt == NULL) |
2012 | 0 | return(0); |
2013 | | |
2014 | 6.56M | if (ctxt->nodeNr >= ctxt->nodeMax) { |
2015 | 837k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 2048 : 256; |
2016 | 837k | xmlNodePtr *tmp; |
2017 | 837k | int newSize; |
2018 | | |
2019 | 837k | newSize = xmlGrowCapacity(ctxt->nodeMax, sizeof(tmp[0]), |
2020 | 837k | 10, maxDepth); |
2021 | 837k | if (newSize < 0) { |
2022 | 324 | xmlFatalErrMsgInt(ctxt, XML_ERR_RESOURCE_LIMIT, |
2023 | 324 | "Excessive depth in document: %d," |
2024 | 324 | " use XML_PARSE_HUGE option\n", |
2025 | 324 | ctxt->nodeNr); |
2026 | 324 | xmlHaltParser(ctxt); |
2027 | 324 | return(-1); |
2028 | 324 | } |
2029 | | |
2030 | 837k | tmp = xmlRealloc(ctxt->nodeTab, newSize * sizeof(tmp[0])); |
2031 | 837k | if (tmp == NULL) { |
2032 | 48 | xmlErrMemory(ctxt); |
2033 | 48 | return (-1); |
2034 | 48 | } |
2035 | 837k | ctxt->nodeTab = tmp; |
2036 | 837k | ctxt->nodeMax = newSize; |
2037 | 837k | } |
2038 | | |
2039 | 6.56M | ctxt->nodeTab[ctxt->nodeNr] = value; |
2040 | 6.56M | ctxt->node = value; |
2041 | 6.56M | return (ctxt->nodeNr++); |
2042 | 6.56M | } |
2043 | | |
2044 | | /** |
2045 | | * Pops the top element node from the node stack |
2046 | | * |
2047 | | * @deprecated Internal function, do not use. |
2048 | | * |
2049 | | * @param ctxt an XML parser context |
2050 | | * @returns the node just removed |
2051 | | */ |
2052 | | xmlNode * |
2053 | | nodePop(xmlParserCtxt *ctxt) |
2054 | 6.89M | { |
2055 | 6.89M | xmlNodePtr ret; |
2056 | | |
2057 | 6.89M | if (ctxt == NULL) return(NULL); |
2058 | 6.89M | if (ctxt->nodeNr <= 0) |
2059 | 349k | return (NULL); |
2060 | 6.54M | ctxt->nodeNr--; |
2061 | 6.54M | if (ctxt->nodeNr > 0) |
2062 | 5.91M | ctxt->node = ctxt->nodeTab[ctxt->nodeNr - 1]; |
2063 | 626k | else |
2064 | 626k | ctxt->node = NULL; |
2065 | 6.54M | ret = ctxt->nodeTab[ctxt->nodeNr]; |
2066 | 6.54M | ctxt->nodeTab[ctxt->nodeNr] = NULL; |
2067 | 6.54M | return (ret); |
2068 | 6.89M | } |
2069 | | |
2070 | | /** |
2071 | | * Pushes a new element name/prefix/URL on top of the name stack |
2072 | | * |
2073 | | * @param ctxt an XML parser context |
2074 | | * @param value the element name |
2075 | | * @param prefix the element prefix |
2076 | | * @param URI the element namespace name |
2077 | | * @param line the current line number for error messages |
2078 | | * @param nsNr the number of namespaces pushed on the namespace table |
2079 | | * @returns -1 in case of error, the index in the stack otherwise |
2080 | | */ |
2081 | | static int |
2082 | | nameNsPush(xmlParserCtxtPtr ctxt, const xmlChar * value, |
2083 | | const xmlChar *prefix, const xmlChar *URI, int line, int nsNr) |
2084 | 7.72M | { |
2085 | 7.72M | xmlStartTag *tag; |
2086 | | |
2087 | 7.72M | if (ctxt->nameNr >= ctxt->nameMax) { |
2088 | 930k | const xmlChar **tmp; |
2089 | 930k | xmlStartTag *tmp2; |
2090 | 930k | int newSize; |
2091 | | |
2092 | 930k | newSize = xmlGrowCapacity(ctxt->nameMax, |
2093 | 930k | sizeof(tmp[0]) + sizeof(tmp2[0]), |
2094 | 930k | 10, XML_MAX_ITEMS); |
2095 | 930k | if (newSize < 0) |
2096 | 0 | goto mem_error; |
2097 | | |
2098 | 930k | tmp = xmlRealloc(ctxt->nameTab, newSize * sizeof(tmp[0])); |
2099 | 930k | if (tmp == NULL) |
2100 | 316 | goto mem_error; |
2101 | 930k | ctxt->nameTab = tmp; |
2102 | | |
2103 | 930k | tmp2 = xmlRealloc(ctxt->pushTab, newSize * sizeof(tmp2[0])); |
2104 | 930k | if (tmp2 == NULL) |
2105 | 32 | goto mem_error; |
2106 | 930k | ctxt->pushTab = tmp2; |
2107 | | |
2108 | 930k | ctxt->nameMax = newSize; |
2109 | 6.79M | } else if (ctxt->pushTab == NULL) { |
2110 | 685k | ctxt->pushTab = xmlMalloc(ctxt->nameMax * sizeof(ctxt->pushTab[0])); |
2111 | 685k | if (ctxt->pushTab == NULL) |
2112 | 754 | goto mem_error; |
2113 | 685k | } |
2114 | 7.72M | ctxt->nameTab[ctxt->nameNr] = value; |
2115 | 7.72M | ctxt->name = value; |
2116 | 7.72M | tag = &ctxt->pushTab[ctxt->nameNr]; |
2117 | 7.72M | tag->prefix = prefix; |
2118 | 7.72M | tag->URI = URI; |
2119 | 7.72M | tag->line = line; |
2120 | 7.72M | tag->nsNr = nsNr; |
2121 | 7.72M | return (ctxt->nameNr++); |
2122 | 1.10k | mem_error: |
2123 | 1.10k | xmlErrMemory(ctxt); |
2124 | 1.10k | return (-1); |
2125 | 7.72M | } |
2126 | | #ifdef LIBXML_PUSH_ENABLED |
2127 | | /** |
2128 | | * Pops the top element/prefix/URI name from the name stack |
2129 | | * |
2130 | | * @param ctxt an XML parser context |
2131 | | * @returns the name just removed |
2132 | | */ |
2133 | | static const xmlChar * |
2134 | | nameNsPop(xmlParserCtxtPtr ctxt) |
2135 | | { |
2136 | | const xmlChar *ret; |
2137 | | |
2138 | | if (ctxt->nameNr <= 0) |
2139 | | return (NULL); |
2140 | | ctxt->nameNr--; |
2141 | | if (ctxt->nameNr > 0) |
2142 | | ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; |
2143 | | else |
2144 | | ctxt->name = NULL; |
2145 | | ret = ctxt->nameTab[ctxt->nameNr]; |
2146 | | ctxt->nameTab[ctxt->nameNr] = NULL; |
2147 | | return (ret); |
2148 | | } |
2149 | | #endif /* LIBXML_PUSH_ENABLED */ |
2150 | | |
2151 | | /** |
2152 | | * Pops the top element name from the name stack |
2153 | | * |
2154 | | * @deprecated Internal function, do not use. |
2155 | | * |
2156 | | * @param ctxt an XML parser context |
2157 | | * @returns the name just removed |
2158 | | */ |
2159 | | static const xmlChar * |
2160 | | namePop(xmlParserCtxtPtr ctxt) |
2161 | 7.67M | { |
2162 | 7.67M | const xmlChar *ret; |
2163 | | |
2164 | 7.67M | if ((ctxt == NULL) || (ctxt->nameNr <= 0)) |
2165 | 268 | return (NULL); |
2166 | 7.67M | ctxt->nameNr--; |
2167 | 7.67M | if (ctxt->nameNr > 0) |
2168 | 7.03M | ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; |
2169 | 640k | else |
2170 | 640k | ctxt->name = NULL; |
2171 | 7.67M | ret = ctxt->nameTab[ctxt->nameNr]; |
2172 | 7.67M | ctxt->nameTab[ctxt->nameNr] = NULL; |
2173 | 7.67M | return (ret); |
2174 | 7.67M | } |
2175 | | |
2176 | 8.29M | static int spacePush(xmlParserCtxtPtr ctxt, int val) { |
2177 | 8.29M | if (ctxt->spaceNr >= ctxt->spaceMax) { |
2178 | 1.58M | int *tmp; |
2179 | 1.58M | int newSize; |
2180 | | |
2181 | 1.58M | newSize = xmlGrowCapacity(ctxt->spaceMax, sizeof(tmp[0]), |
2182 | 1.58M | 10, XML_MAX_ITEMS); |
2183 | 1.58M | if (newSize < 0) { |
2184 | 0 | xmlErrMemory(ctxt); |
2185 | 0 | return(-1); |
2186 | 0 | } |
2187 | | |
2188 | 1.58M | tmp = xmlRealloc(ctxt->spaceTab, newSize * sizeof(tmp[0])); |
2189 | 1.58M | if (tmp == NULL) { |
2190 | 242 | xmlErrMemory(ctxt); |
2191 | 242 | return(-1); |
2192 | 242 | } |
2193 | 1.57M | ctxt->spaceTab = tmp; |
2194 | | |
2195 | 1.57M | ctxt->spaceMax = newSize; |
2196 | 1.57M | } |
2197 | 8.29M | ctxt->spaceTab[ctxt->spaceNr] = val; |
2198 | 8.29M | ctxt->space = &ctxt->spaceTab[ctxt->spaceNr]; |
2199 | 8.29M | return(ctxt->spaceNr++); |
2200 | 8.29M | } |
2201 | | |
2202 | 8.25M | static int spacePop(xmlParserCtxtPtr ctxt) { |
2203 | 8.25M | int ret; |
2204 | 8.25M | if (ctxt->spaceNr <= 0) return(0); |
2205 | 8.25M | ctxt->spaceNr--; |
2206 | 8.25M | if (ctxt->spaceNr > 0) |
2207 | 8.25M | ctxt->space = &ctxt->spaceTab[ctxt->spaceNr - 1]; |
2208 | 203 | else |
2209 | 203 | ctxt->space = &ctxt->spaceTab[0]; |
2210 | 8.25M | ret = ctxt->spaceTab[ctxt->spaceNr]; |
2211 | 8.25M | ctxt->spaceTab[ctxt->spaceNr] = -1; |
2212 | 8.25M | return(ret); |
2213 | 8.25M | } |
2214 | | |
2215 | | /* |
2216 | | * Macros for accessing the content. Those should be used only by the parser, |
2217 | | * and not exported. |
2218 | | * |
2219 | | * Dirty macros, i.e. one often need to make assumption on the context to |
2220 | | * use them |
2221 | | * |
2222 | | * CUR_PTR return the current pointer to the xmlChar to be parsed. |
2223 | | * To be used with extreme caution since operations consuming |
2224 | | * characters may move the input buffer to a different location ! |
2225 | | * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled |
2226 | | * This should be used internally by the parser |
2227 | | * only to compare to ASCII values otherwise it would break when |
2228 | | * running with UTF-8 encoding. |
2229 | | * RAW same as CUR but in the input buffer, bypass any token |
2230 | | * extraction that may have been done |
2231 | | * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only |
2232 | | * to compare on ASCII based substring. |
2233 | | * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined |
2234 | | * strings without newlines within the parser. |
2235 | | * NEXT1(l) Skip 1 xmlChar, and must also be used only to skip 1 non-newline ASCII |
2236 | | * defined char within the parser. |
2237 | | * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding |
2238 | | * |
2239 | | * NEXT Skip to the next character, this does the proper decoding |
2240 | | * in UTF-8 mode. It also pop-up unfinished entities on the fly. |
2241 | | * NEXTL(l) Skip the current unicode character of l xmlChars long. |
2242 | | * COPY_BUF copy the current unicode char to the target buffer, increment |
2243 | | * the index |
2244 | | * GROW, SHRINK handling of input buffers |
2245 | | */ |
2246 | | |
2247 | 70.7M | #define RAW (*ctxt->input->cur) |
2248 | 817M | #define CUR (*ctxt->input->cur) |
2249 | 33.7M | #define NXT(val) ctxt->input->cur[(val)] |
2250 | 1.78G | #define CUR_PTR ctxt->input->cur |
2251 | 28.4M | #define BASE_PTR ctxt->input->base |
2252 | | |
2253 | | #define CMP4( s, c1, c2, c3, c4 ) \ |
2254 | 56.9M | ( ((unsigned char *) s)[ 0 ] == c1 && ((unsigned char *) s)[ 1 ] == c2 && \ |
2255 | 29.3M | ((unsigned char *) s)[ 2 ] == c3 && ((unsigned char *) s)[ 3 ] == c4 ) |
2256 | | #define CMP5( s, c1, c2, c3, c4, c5 ) \ |
2257 | 54.1M | ( CMP4( s, c1, c2, c3, c4 ) && ((unsigned char *) s)[ 4 ] == c5 ) |
2258 | | #define CMP6( s, c1, c2, c3, c4, c5, c6 ) \ |
2259 | 52.1M | ( CMP5( s, c1, c2, c3, c4, c5 ) && ((unsigned char *) s)[ 5 ] == c6 ) |
2260 | | #define CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) \ |
2261 | 50.9M | ( CMP6( s, c1, c2, c3, c4, c5, c6 ) && ((unsigned char *) s)[ 6 ] == c7 ) |
2262 | | #define CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) \ |
2263 | 50.1M | ( CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) && ((unsigned char *) s)[ 7 ] == c8 ) |
2264 | | #define CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) \ |
2265 | 24.9M | ( CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) && \ |
2266 | 24.9M | ((unsigned char *) s)[ 8 ] == c9 ) |
2267 | | #define CMP10( s, c1, c2, c3, c4, c5, c6, c7, c8, c9, c10 ) \ |
2268 | 82.0k | ( CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) && \ |
2269 | 82.0k | ((unsigned char *) s)[ 9 ] == c10 ) |
2270 | | |
2271 | 7.33M | #define SKIP(val) do { \ |
2272 | 7.33M | ctxt->input->cur += (val),ctxt->input->col+=(val); \ |
2273 | 7.33M | if (*ctxt->input->cur == 0) \ |
2274 | 7.33M | xmlParserGrow(ctxt); \ |
2275 | 7.33M | } while (0) |
2276 | | |
2277 | | #define SKIPL(val) do { \ |
2278 | | int skipl; \ |
2279 | | for(skipl=0; skipl<val; skipl++) { \ |
2280 | | if (*(ctxt->input->cur) == '\n') { \ |
2281 | | ctxt->input->line++; ctxt->input->col = 1; \ |
2282 | | } else ctxt->input->col++; \ |
2283 | | ctxt->input->cur++; \ |
2284 | | } \ |
2285 | | if (*ctxt->input->cur == 0) \ |
2286 | | xmlParserGrow(ctxt); \ |
2287 | | } while (0) |
2288 | | |
2289 | | #define SHRINK \ |
2290 | 34.6M | if (!PARSER_PROGRESSIVE(ctxt)) \ |
2291 | 34.6M | xmlParserShrink(ctxt); |
2292 | | |
2293 | | #define GROW \ |
2294 | 108M | if ((!PARSER_PROGRESSIVE(ctxt)) && \ |
2295 | 108M | (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \ |
2296 | 35.9M | xmlParserGrow(ctxt); |
2297 | | |
2298 | 27.0M | #define SKIP_BLANKS xmlSkipBlankChars(ctxt) |
2299 | | |
2300 | 1.35M | #define SKIP_BLANKS_PE xmlSkipBlankCharsPE(ctxt) |
2301 | | |
2302 | 40.4M | #define NEXT xmlNextChar(ctxt) |
2303 | | |
2304 | 12.0M | #define NEXT1 { \ |
2305 | 12.0M | ctxt->input->col++; \ |
2306 | 12.0M | ctxt->input->cur++; \ |
2307 | 12.0M | if (*ctxt->input->cur == 0) \ |
2308 | 12.0M | xmlParserGrow(ctxt); \ |
2309 | 12.0M | } |
2310 | | |
2311 | 1.16G | #define NEXTL(l) do { \ |
2312 | 1.16G | if (*(ctxt->input->cur) == '\n') { \ |
2313 | 94.0M | ctxt->input->line++; ctxt->input->col = 1; \ |
2314 | 1.07G | } else ctxt->input->col++; \ |
2315 | 1.16G | ctxt->input->cur += l; \ |
2316 | 1.16G | } while (0) |
2317 | | |
2318 | | #define COPY_BUF(b, i, v) \ |
2319 | 366M | if (v < 0x80) b[i++] = v; \ |
2320 | 366M | else i += xmlCopyCharMultiByte(&b[i],v) |
2321 | | |
2322 | | static int |
2323 | 360M | xmlCurrentCharRecover(xmlParserCtxtPtr ctxt, int *len) { |
2324 | 360M | int c = xmlCurrentChar(ctxt, len); |
2325 | | |
2326 | 360M | if (c == XML_INVALID_CHAR) |
2327 | 75.9M | c = 0xFFFD; /* replacement character */ |
2328 | | |
2329 | 360M | return(c); |
2330 | 360M | } |
2331 | | |
2332 | | /** |
2333 | | * Skip whitespace in the input stream. |
2334 | | * |
2335 | | * @deprecated Internal function, do not use. |
2336 | | * |
2337 | | * @param ctxt the XML parser context |
2338 | | * @returns the number of space chars skipped |
2339 | | */ |
2340 | | int |
2341 | 27.8M | xmlSkipBlankChars(xmlParserCtxt *ctxt) { |
2342 | 27.8M | const xmlChar *cur; |
2343 | 27.8M | int res = 0; |
2344 | | |
2345 | 27.8M | cur = ctxt->input->cur; |
2346 | 27.8M | while (IS_BLANK_CH(*cur)) { |
2347 | 15.0M | if (*cur == '\n') { |
2348 | 2.89M | ctxt->input->line++; ctxt->input->col = 1; |
2349 | 12.1M | } else { |
2350 | 12.1M | ctxt->input->col++; |
2351 | 12.1M | } |
2352 | 15.0M | cur++; |
2353 | 15.0M | if (res < INT_MAX) |
2354 | 15.0M | res++; |
2355 | 15.0M | if (*cur == 0) { |
2356 | 580k | ctxt->input->cur = cur; |
2357 | 580k | xmlParserGrow(ctxt); |
2358 | 580k | cur = ctxt->input->cur; |
2359 | 580k | } |
2360 | 15.0M | } |
2361 | 27.8M | ctxt->input->cur = cur; |
2362 | | |
2363 | 27.8M | if (res > 4) |
2364 | 454k | GROW; |
2365 | | |
2366 | 27.8M | return(res); |
2367 | 27.8M | } |
2368 | | |
2369 | | static void |
2370 | 25.6k | xmlPopPE(xmlParserCtxtPtr ctxt) { |
2371 | 25.6k | unsigned long consumed; |
2372 | 25.6k | xmlEntityPtr ent; |
2373 | | |
2374 | 25.6k | ent = ctxt->input->entity; |
2375 | | |
2376 | 25.6k | ent->flags &= ~XML_ENT_EXPANDING; |
2377 | | |
2378 | 25.6k | if ((ent->flags & XML_ENT_CHECKED) == 0) { |
2379 | 18.7k | int result; |
2380 | | |
2381 | | /* |
2382 | | * Read the rest of the stream in case of errors. We want |
2383 | | * to account for the whole entity size. |
2384 | | */ |
2385 | 23.5k | do { |
2386 | 23.5k | ctxt->input->cur = ctxt->input->end; |
2387 | 23.5k | xmlParserShrink(ctxt); |
2388 | 23.5k | result = xmlParserGrow(ctxt); |
2389 | 23.5k | } while (result > 0); |
2390 | | |
2391 | 18.7k | consumed = ctxt->input->consumed; |
2392 | 18.7k | xmlSaturatedAddSizeT(&consumed, |
2393 | 18.7k | ctxt->input->end - ctxt->input->base); |
2394 | | |
2395 | 18.7k | xmlSaturatedAdd(&ent->expandedSize, consumed); |
2396 | | |
2397 | | /* |
2398 | | * Add to sizeentities when parsing an external entity |
2399 | | * for the first time. |
2400 | | */ |
2401 | 18.7k | if (ent->etype == XML_EXTERNAL_PARAMETER_ENTITY) { |
2402 | 15.5k | xmlSaturatedAdd(&ctxt->sizeentities, consumed); |
2403 | 15.5k | } |
2404 | | |
2405 | 18.7k | ent->flags |= XML_ENT_CHECKED; |
2406 | 18.7k | } |
2407 | | |
2408 | 25.6k | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
2409 | | |
2410 | 25.6k | xmlParserEntityCheck(ctxt, ent->expandedSize); |
2411 | | |
2412 | 25.6k | GROW; |
2413 | 25.6k | } |
2414 | | |
2415 | | /** |
2416 | | * Skip whitespace in the input stream, also handling parameter |
2417 | | * entities. |
2418 | | * |
2419 | | * @param ctxt the XML parser context |
2420 | | * @returns the number of space chars skipped |
2421 | | */ |
2422 | | static int |
2423 | 1.35M | xmlSkipBlankCharsPE(xmlParserCtxtPtr ctxt) { |
2424 | 1.35M | int res = 0; |
2425 | 1.35M | int inParam; |
2426 | 1.35M | int expandParam; |
2427 | | |
2428 | 1.35M | inParam = PARSER_IN_PE(ctxt); |
2429 | 1.35M | expandParam = PARSER_EXTERNAL(ctxt); |
2430 | | |
2431 | 1.35M | if (!inParam && !expandParam) |
2432 | 781k | return(xmlSkipBlankChars(ctxt)); |
2433 | | |
2434 | | /* |
2435 | | * It's Okay to use CUR/NEXT here since all the blanks are on |
2436 | | * the ASCII range. |
2437 | | */ |
2438 | 1.88M | while (PARSER_STOPPED(ctxt) == 0) { |
2439 | 1.87M | if (IS_BLANK_CH(CUR)) { /* CHECKED tstblanks.xml */ |
2440 | 1.28M | NEXT; |
2441 | 1.28M | } else if (CUR == '%') { |
2442 | 86.1k | if ((expandParam == 0) || |
2443 | 86.1k | (IS_BLANK_CH(NXT(1))) || (NXT(1) == 0)) |
2444 | 60.9k | break; |
2445 | | |
2446 | | /* |
2447 | | * Expand parameter entity. We continue to consume |
2448 | | * whitespace at the start of the entity and possible |
2449 | | * even consume the whole entity and pop it. We might |
2450 | | * even pop multiple PEs in this loop. |
2451 | | */ |
2452 | 25.1k | xmlParsePERefInternal(ctxt, 0); |
2453 | | |
2454 | 25.1k | inParam = PARSER_IN_PE(ctxt); |
2455 | 25.1k | expandParam = PARSER_EXTERNAL(ctxt); |
2456 | 510k | } else if (CUR == 0) { |
2457 | 10.2k | if (inParam == 0) |
2458 | 1.04k | break; |
2459 | | |
2460 | | /* |
2461 | | * Don't pop parameter entities that start a markup |
2462 | | * declaration to detect Well-formedness constraint: |
2463 | | * PE Between Declarations. |
2464 | | */ |
2465 | 9.23k | if (ctxt->input->flags & XML_INPUT_MARKUP_DECL) |
2466 | 3.59k | break; |
2467 | | |
2468 | 5.64k | xmlPopPE(ctxt); |
2469 | | |
2470 | 5.64k | inParam = PARSER_IN_PE(ctxt); |
2471 | 5.64k | expandParam = PARSER_EXTERNAL(ctxt); |
2472 | 500k | } else { |
2473 | 500k | break; |
2474 | 500k | } |
2475 | | |
2476 | | /* |
2477 | | * Also increase the counter when entering or exiting a PERef. |
2478 | | * The spec says: "When a parameter-entity reference is recognized |
2479 | | * in the DTD and included, its replacement text MUST be enlarged |
2480 | | * by the attachment of one leading and one following space (#x20) |
2481 | | * character." |
2482 | | */ |
2483 | 1.31M | if (res < INT_MAX) |
2484 | 1.31M | res++; |
2485 | 1.31M | } |
2486 | | |
2487 | 568k | return(res); |
2488 | 1.35M | } |
2489 | | |
2490 | | /************************************************************************ |
2491 | | * * |
2492 | | * Commodity functions to handle entities * |
2493 | | * * |
2494 | | ************************************************************************/ |
2495 | | |
2496 | | /** |
2497 | | * @deprecated Internal function, don't use. |
2498 | | * |
2499 | | * @param ctxt an XML parser context |
2500 | | * @returns the current xmlChar in the parser context |
2501 | | */ |
2502 | | xmlChar |
2503 | 0 | xmlPopInput(xmlParserCtxt *ctxt) { |
2504 | 0 | xmlParserInputPtr input; |
2505 | |
|
2506 | 0 | if ((ctxt == NULL) || (ctxt->inputNr <= 1)) return(0); |
2507 | 0 | input = xmlCtxtPopInput(ctxt); |
2508 | 0 | xmlFreeInputStream(input); |
2509 | 0 | if (*ctxt->input->cur == 0) |
2510 | 0 | xmlParserGrow(ctxt); |
2511 | 0 | return(CUR); |
2512 | 0 | } |
2513 | | |
2514 | | /** |
2515 | | * Push an input stream onto the stack. |
2516 | | * |
2517 | | * @deprecated Internal function, don't use. |
2518 | | * |
2519 | | * @param ctxt an XML parser context |
2520 | | * @param input an XML parser input fragment (entity, XML fragment ...). |
2521 | | * @returns -1 in case of error or the index in the input stack |
2522 | | */ |
2523 | | int |
2524 | 0 | xmlPushInput(xmlParserCtxt *ctxt, xmlParserInput *input) { |
2525 | 0 | int ret; |
2526 | |
|
2527 | 0 | if ((ctxt == NULL) || (input == NULL)) |
2528 | 0 | return(-1); |
2529 | | |
2530 | 0 | ret = xmlCtxtPushInput(ctxt, input); |
2531 | 0 | if (ret >= 0) |
2532 | 0 | GROW; |
2533 | 0 | return(ret); |
2534 | 0 | } |
2535 | | |
2536 | | /** |
2537 | | * Parse a numeric character reference. Always consumes '&'. |
2538 | | * |
2539 | | * @deprecated Internal function, don't use. |
2540 | | * |
2541 | | * [66] CharRef ::= '&#' [0-9]+ ';' | |
2542 | | * '&#x' [0-9a-fA-F]+ ';' |
2543 | | * |
2544 | | * [ WFC: Legal Character ] |
2545 | | * Characters referred to using character references must match the |
2546 | | * production for Char. |
2547 | | * |
2548 | | * @param ctxt an XML parser context |
2549 | | * @returns the value parsed (as an int), 0 in case of error |
2550 | | */ |
2551 | | int |
2552 | 178k | xmlParseCharRef(xmlParserCtxt *ctxt) { |
2553 | 178k | int val = 0; |
2554 | 178k | int count = 0; |
2555 | | |
2556 | | /* |
2557 | | * Using RAW/CUR/NEXT is okay since we are working on ASCII range here |
2558 | | */ |
2559 | 178k | if ((RAW == '&') && (NXT(1) == '#') && |
2560 | 178k | (NXT(2) == 'x')) { |
2561 | 28.0k | SKIP(3); |
2562 | 28.0k | GROW; |
2563 | 101k | while ((RAW != ';') && (PARSER_STOPPED(ctxt) == 0)) { |
2564 | 83.1k | if (count++ > 20) { |
2565 | 848 | count = 0; |
2566 | 848 | GROW; |
2567 | 848 | } |
2568 | 83.1k | if ((RAW >= '0') && (RAW <= '9')) |
2569 | 54.5k | val = val * 16 + (CUR - '0'); |
2570 | 28.6k | else if ((RAW >= 'a') && (RAW <= 'f') && (count < 20)) |
2571 | 7.99k | val = val * 16 + (CUR - 'a') + 10; |
2572 | 20.6k | else if ((RAW >= 'A') && (RAW <= 'F') && (count < 20)) |
2573 | 11.3k | val = val * 16 + (CUR - 'A') + 10; |
2574 | 9.37k | else { |
2575 | 9.37k | xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); |
2576 | 9.37k | val = 0; |
2577 | 9.37k | break; |
2578 | 9.37k | } |
2579 | 73.8k | if (val > 0x110000) |
2580 | 10.0k | val = 0x110000; |
2581 | | |
2582 | 73.8k | NEXT; |
2583 | 73.8k | count++; |
2584 | 73.8k | } |
2585 | 28.0k | if (RAW == ';') { |
2586 | | /* on purpose to avoid reentrancy problems with NEXT and SKIP */ |
2587 | 18.6k | ctxt->input->col++; |
2588 | 18.6k | ctxt->input->cur++; |
2589 | 18.6k | } |
2590 | 150k | } else if ((RAW == '&') && (NXT(1) == '#')) { |
2591 | 150k | SKIP(2); |
2592 | 150k | GROW; |
2593 | 419k | while (RAW != ';') { /* loop blocked by count */ |
2594 | 278k | if (count++ > 20) { |
2595 | 1.19k | count = 0; |
2596 | 1.19k | GROW; |
2597 | 1.19k | } |
2598 | 278k | if ((RAW >= '0') && (RAW <= '9')) |
2599 | 268k | val = val * 10 + (CUR - '0'); |
2600 | 9.79k | else { |
2601 | 9.79k | xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); |
2602 | 9.79k | val = 0; |
2603 | 9.79k | break; |
2604 | 9.79k | } |
2605 | 268k | if (val > 0x110000) |
2606 | 14.7k | val = 0x110000; |
2607 | | |
2608 | 268k | NEXT; |
2609 | 268k | count++; |
2610 | 268k | } |
2611 | 150k | if (RAW == ';') { |
2612 | | /* on purpose to avoid reentrancy problems with NEXT and SKIP */ |
2613 | 140k | ctxt->input->col++; |
2614 | 140k | ctxt->input->cur++; |
2615 | 140k | } |
2616 | 150k | } else { |
2617 | 0 | if (RAW == '&') |
2618 | 0 | SKIP(1); |
2619 | 0 | xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); |
2620 | 0 | } |
2621 | | |
2622 | | /* |
2623 | | * [ WFC: Legal Character ] |
2624 | | * Characters referred to using character references must match the |
2625 | | * production for Char. |
2626 | | */ |
2627 | 178k | if (val >= 0x110000) { |
2628 | 304 | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2629 | 304 | "xmlParseCharRef: character reference out of bounds\n", |
2630 | 304 | val); |
2631 | 304 | val = 0xFFFD; |
2632 | 177k | } else if (!IS_CHAR(val)) { |
2633 | 20.1k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2634 | 20.1k | "xmlParseCharRef: invalid xmlChar value %d\n", |
2635 | 20.1k | val); |
2636 | 20.1k | } |
2637 | 178k | return(val); |
2638 | 178k | } |
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 | 189k | xmlParseStringCharRef(xmlParserCtxtPtr ctxt, const xmlChar **str) { |
2658 | 189k | const xmlChar *ptr; |
2659 | 189k | xmlChar cur; |
2660 | 189k | int val = 0; |
2661 | | |
2662 | 189k | if ((str == NULL) || (*str == NULL)) return(0); |
2663 | 189k | ptr = *str; |
2664 | 189k | cur = *ptr; |
2665 | 189k | if ((cur == '&') && (ptr[1] == '#') && (ptr[2] == 'x')) { |
2666 | 150k | ptr += 3; |
2667 | 150k | cur = *ptr; |
2668 | 731k | while (cur != ';') { /* Non input consuming loop */ |
2669 | 581k | if ((cur >= '0') && (cur <= '9')) |
2670 | 449k | val = val * 16 + (cur - '0'); |
2671 | 131k | else if ((cur >= 'a') && (cur <= 'f')) |
2672 | 3.73k | val = val * 16 + (cur - 'a') + 10; |
2673 | 128k | else if ((cur >= 'A') && (cur <= 'F')) |
2674 | 126k | val = val * 16 + (cur - 'A') + 10; |
2675 | 1.18k | else { |
2676 | 1.18k | xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); |
2677 | 1.18k | val = 0; |
2678 | 1.18k | break; |
2679 | 1.18k | } |
2680 | 580k | if (val > 0x110000) |
2681 | 12.3k | val = 0x110000; |
2682 | | |
2683 | 580k | ptr++; |
2684 | 580k | cur = *ptr; |
2685 | 580k | } |
2686 | 150k | if (cur == ';') |
2687 | 149k | ptr++; |
2688 | 150k | } else if ((cur == '&') && (ptr[1] == '#')){ |
2689 | 38.1k | ptr += 2; |
2690 | 38.1k | cur = *ptr; |
2691 | 134k | while (cur != ';') { /* Non input consuming loops */ |
2692 | 99.7k | if ((cur >= '0') && (cur <= '9')) |
2693 | 96.6k | val = val * 10 + (cur - '0'); |
2694 | 3.11k | else { |
2695 | 3.11k | xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); |
2696 | 3.11k | val = 0; |
2697 | 3.11k | break; |
2698 | 3.11k | } |
2699 | 96.6k | if (val > 0x110000) |
2700 | 2.83k | val = 0x110000; |
2701 | | |
2702 | 96.6k | ptr++; |
2703 | 96.6k | cur = *ptr; |
2704 | 96.6k | } |
2705 | 38.1k | if (cur == ';') |
2706 | 35.0k | ptr++; |
2707 | 38.1k | } else { |
2708 | 0 | xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); |
2709 | 0 | return(0); |
2710 | 0 | } |
2711 | 189k | *str = ptr; |
2712 | | |
2713 | | /* |
2714 | | * [ WFC: Legal Character ] |
2715 | | * Characters referred to using character references must match the |
2716 | | * production for Char. |
2717 | | */ |
2718 | 189k | if (val >= 0x110000) { |
2719 | 264 | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2720 | 264 | "xmlParseStringCharRef: character reference out of bounds\n", |
2721 | 264 | val); |
2722 | 188k | } else if (IS_CHAR(val)) { |
2723 | 183k | return(val); |
2724 | 183k | } else { |
2725 | 4.99k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
2726 | 4.99k | "xmlParseStringCharRef: invalid xmlChar value %d\n", |
2727 | 4.99k | val); |
2728 | 4.99k | } |
2729 | 5.26k | return(0); |
2730 | 189k | } |
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 | 9.65M | 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 | 9.65M | if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ |
2976 | 9.65M | (((c >= 'a') && (c <= 'z')) || |
2977 | 9.54M | ((c >= 'A') && (c <= 'Z')) || |
2978 | 9.54M | (c == '_') || (c == ':') || |
2979 | 9.54M | ((c >= 0xC0) && (c <= 0xD6)) || |
2980 | 9.54M | ((c >= 0xD8) && (c <= 0xF6)) || |
2981 | 9.54M | ((c >= 0xF8) && (c <= 0x2FF)) || |
2982 | 9.54M | ((c >= 0x370) && (c <= 0x37D)) || |
2983 | 9.54M | ((c >= 0x37F) && (c <= 0x1FFF)) || |
2984 | 9.54M | ((c >= 0x200C) && (c <= 0x200D)) || |
2985 | 9.54M | ((c >= 0x2070) && (c <= 0x218F)) || |
2986 | 9.54M | ((c >= 0x2C00) && (c <= 0x2FEF)) || |
2987 | 9.54M | ((c >= 0x3001) && (c <= 0xD7FF)) || |
2988 | 9.54M | ((c >= 0xF900) && (c <= 0xFDCF)) || |
2989 | 9.54M | ((c >= 0xFDF0) && (c <= 0xFFFD)) || |
2990 | 9.54M | ((c >= 0x10000) && (c <= 0xEFFFF)))) |
2991 | 5.64M | return(1); |
2992 | 4.01M | return(0); |
2993 | 9.65M | } |
2994 | | |
2995 | | static int |
2996 | 108M | 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 | 108M | if ((c != ' ') && (c != '>') && (c != '/') && /* accelerators */ |
3002 | 108M | (((c >= 'a') && (c <= 'z')) || |
3003 | 108M | ((c >= 'A') && (c <= 'Z')) || |
3004 | 108M | ((c >= '0') && (c <= '9')) || /* !start */ |
3005 | 108M | (c == '_') || (c == ':') || |
3006 | 108M | (c == '-') || (c == '.') || (c == 0xB7) || /* !start */ |
3007 | 108M | ((c >= 0xC0) && (c <= 0xD6)) || |
3008 | 108M | ((c >= 0xD8) && (c <= 0xF6)) || |
3009 | 108M | ((c >= 0xF8) && (c <= 0x2FF)) || |
3010 | 108M | ((c >= 0x300) && (c <= 0x36F)) || /* !start */ |
3011 | 108M | ((c >= 0x370) && (c <= 0x37D)) || |
3012 | 108M | ((c >= 0x37F) && (c <= 0x1FFF)) || |
3013 | 108M | ((c >= 0x200C) && (c <= 0x200D)) || |
3014 | 108M | ((c >= 0x203F) && (c <= 0x2040)) || /* !start */ |
3015 | 108M | ((c >= 0x2070) && (c <= 0x218F)) || |
3016 | 108M | ((c >= 0x2C00) && (c <= 0x2FEF)) || |
3017 | 108M | ((c >= 0x3001) && (c <= 0xD7FF)) || |
3018 | 108M | ((c >= 0xF900) && (c <= 0xFDCF)) || |
3019 | 108M | ((c >= 0xFDF0) && (c <= 0xFFFD)) || |
3020 | 108M | ((c >= 0x10000) && (c <= 0xEFFFF)))) |
3021 | 103M | return(1); |
3022 | 5.09M | return(0); |
3023 | 108M | } |
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 | 9.65M | xmlIsNameStartChar(int c, int old10) { |
3047 | 9.65M | if (!old10) |
3048 | 9.65M | return(xmlIsNameStartCharNew(c)); |
3049 | 0 | else |
3050 | 0 | return(xmlIsNameStartCharOld(c)); |
3051 | 9.65M | } |
3052 | | |
3053 | | static int |
3054 | 108M | xmlIsNameChar(int c, int old10) { |
3055 | 108M | if (!old10) |
3056 | 108M | return(xmlIsNameCharNew(c)); |
3057 | 0 | else |
3058 | 0 | return(xmlIsNameCharOld(c)); |
3059 | 108M | } |
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 | 5.72M | xmlScanName(const xmlChar *ptr, size_t maxSize, int flags) { |
3075 | 5.72M | int stop = flags & XML_SCAN_NC ? ':' : 0; |
3076 | 5.72M | int old10 = flags & XML_SCAN_OLD10 ? 1 : 0; |
3077 | | |
3078 | 63.4M | while (1) { |
3079 | 63.4M | int c, len; |
3080 | | |
3081 | 63.4M | c = *ptr; |
3082 | 63.4M | if (c < 0x80) { |
3083 | 59.0M | if (c == stop) |
3084 | 690k | break; |
3085 | 58.3M | len = 1; |
3086 | 58.3M | } else { |
3087 | 4.39M | len = 4; |
3088 | 4.39M | c = xmlGetUTF8Char(ptr, &len); |
3089 | 4.39M | if (c < 0) |
3090 | 2.39k | break; |
3091 | 4.39M | } |
3092 | | |
3093 | 62.7M | if (flags & XML_SCAN_NMTOKEN ? |
3094 | 57.1M | !xmlIsNameChar(c, old10) : |
3095 | 62.7M | !xmlIsNameStartChar(c, old10)) |
3096 | 5.02M | break; |
3097 | | |
3098 | 57.7M | if ((size_t) len > maxSize) |
3099 | 613 | return(NULL); |
3100 | 57.7M | ptr += len; |
3101 | 57.7M | maxSize -= len; |
3102 | 57.7M | flags |= XML_SCAN_NMTOKEN; |
3103 | 57.7M | } |
3104 | | |
3105 | 5.72M | return(ptr); |
3106 | 5.72M | } |
3107 | | |
3108 | | static const xmlChar * |
3109 | 579k | xmlParseNameComplex(xmlParserCtxtPtr ctxt) { |
3110 | 579k | const xmlChar *ret; |
3111 | 579k | int len = 0, l; |
3112 | 579k | int c; |
3113 | 579k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3114 | 0 | XML_MAX_TEXT_LENGTH : |
3115 | 579k | XML_MAX_NAME_LENGTH; |
3116 | 579k | int old10 = (ctxt->options & XML_PARSE_OLD10) ? 1 : 0; |
3117 | | |
3118 | | /* |
3119 | | * Handler for more complex cases |
3120 | | */ |
3121 | 579k | c = xmlCurrentChar(ctxt, &l); |
3122 | 579k | if (!xmlIsNameStartChar(c, old10)) |
3123 | 500k | return(NULL); |
3124 | 78.4k | len += l; |
3125 | 78.4k | NEXTL(l); |
3126 | 78.4k | c = xmlCurrentChar(ctxt, &l); |
3127 | 21.0M | while (xmlIsNameChar(c, old10)) { |
3128 | 20.9M | if (len <= INT_MAX - l) |
3129 | 20.9M | len += l; |
3130 | 20.9M | NEXTL(l); |
3131 | 20.9M | c = xmlCurrentChar(ctxt, &l); |
3132 | 20.9M | } |
3133 | 78.4k | if (len > maxLength) { |
3134 | 252 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Name"); |
3135 | 252 | return(NULL); |
3136 | 252 | } |
3137 | 78.2k | 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 | 78.2k | if ((*ctxt->input->cur == '\n') && (ctxt->input->cur[-1] == '\r')) |
3148 | 838 | ret = xmlDictLookup(ctxt->dict, ctxt->input->cur - (len + 1), len); |
3149 | 77.4k | else |
3150 | 77.4k | ret = xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len); |
3151 | 78.2k | if (ret == NULL) |
3152 | 1 | xmlErrMemory(ctxt); |
3153 | 78.2k | return(ret); |
3154 | 78.2k | } |
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 | 2.06M | xmlParseName(xmlParserCtxt *ctxt) { |
3174 | 2.06M | const xmlChar *in; |
3175 | 2.06M | const xmlChar *ret; |
3176 | 2.06M | size_t count = 0; |
3177 | 2.06M | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3178 | 0 | XML_MAX_TEXT_LENGTH : |
3179 | 2.06M | XML_MAX_NAME_LENGTH; |
3180 | | |
3181 | 2.06M | GROW; |
3182 | | |
3183 | | /* |
3184 | | * Accelerator for simple ASCII names |
3185 | | */ |
3186 | 2.06M | in = ctxt->input->cur; |
3187 | 2.06M | if (((*in >= 0x61) && (*in <= 0x7A)) || |
3188 | 2.06M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3189 | 2.06M | (*in == '_') || (*in == ':')) { |
3190 | 1.55M | in++; |
3191 | 233M | while (((*in >= 0x61) && (*in <= 0x7A)) || |
3192 | 233M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3193 | 233M | ((*in >= 0x30) && (*in <= 0x39)) || |
3194 | 233M | (*in == '_') || (*in == '-') || |
3195 | 233M | (*in == ':') || (*in == '.')) |
3196 | 231M | in++; |
3197 | 1.55M | if ((*in > 0) && (*in < 0x80)) { |
3198 | 1.48M | count = in - ctxt->input->cur; |
3199 | 1.48M | if (count > maxLength) { |
3200 | 2.02k | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "Name"); |
3201 | 2.02k | return(NULL); |
3202 | 2.02k | } |
3203 | 1.48M | ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count); |
3204 | 1.48M | ctxt->input->cur = in; |
3205 | 1.48M | ctxt->input->col += count; |
3206 | 1.48M | if (ret == NULL) |
3207 | 3 | xmlErrMemory(ctxt); |
3208 | 1.48M | return(ret); |
3209 | 1.48M | } |
3210 | 1.55M | } |
3211 | | /* accelerator for special cases */ |
3212 | 579k | return(xmlParseNameComplex(ctxt)); |
3213 | 2.06M | } |
3214 | | |
3215 | | static xmlHashedString |
3216 | 3.47M | xmlParseNCNameComplex(xmlParserCtxtPtr ctxt) { |
3217 | 3.47M | xmlHashedString ret; |
3218 | 3.47M | int len = 0, l; |
3219 | 3.47M | int c; |
3220 | 3.47M | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3221 | 0 | XML_MAX_TEXT_LENGTH : |
3222 | 3.47M | XML_MAX_NAME_LENGTH; |
3223 | 3.47M | int old10 = (ctxt->options & XML_PARSE_OLD10) ? 1 : 0; |
3224 | 3.47M | size_t startPosition = 0; |
3225 | | |
3226 | 3.47M | ret.name = NULL; |
3227 | 3.47M | ret.hashValue = 0; |
3228 | | |
3229 | | /* |
3230 | | * Handler for more complex cases |
3231 | | */ |
3232 | 3.47M | startPosition = CUR_PTR - BASE_PTR; |
3233 | 3.47M | c = xmlCurrentChar(ctxt, &l); |
3234 | 3.47M | if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */ |
3235 | 3.47M | (!xmlIsNameStartChar(c, old10) || (c == ':'))) { |
3236 | 3.29M | return(ret); |
3237 | 3.29M | } |
3238 | | |
3239 | 14.9M | while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */ |
3240 | 14.9M | (xmlIsNameChar(c, old10) && (c != ':'))) { |
3241 | 14.8M | if (len <= INT_MAX - l) |
3242 | 14.8M | len += l; |
3243 | 14.8M | NEXTL(l); |
3244 | 14.8M | c = xmlCurrentChar(ctxt, &l); |
3245 | 14.8M | } |
3246 | 175k | if (len > maxLength) { |
3247 | 424 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); |
3248 | 424 | return(ret); |
3249 | 424 | } |
3250 | 175k | ret = xmlDictLookupHashed(ctxt->dict, (BASE_PTR + startPosition), len); |
3251 | 175k | if (ret.name == NULL) |
3252 | 1 | xmlErrMemory(ctxt); |
3253 | 175k | return(ret); |
3254 | 175k | } |
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 | 21.2M | xmlParseNCName(xmlParserCtxtPtr ctxt) { |
3270 | 21.2M | const xmlChar *in, *e; |
3271 | 21.2M | xmlHashedString ret; |
3272 | 21.2M | size_t count = 0; |
3273 | 21.2M | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3274 | 0 | XML_MAX_TEXT_LENGTH : |
3275 | 21.2M | XML_MAX_NAME_LENGTH; |
3276 | | |
3277 | 21.2M | ret.name = NULL; |
3278 | | |
3279 | | /* |
3280 | | * Accelerator for simple ASCII names |
3281 | | */ |
3282 | 21.2M | in = ctxt->input->cur; |
3283 | 21.2M | e = ctxt->input->end; |
3284 | 21.2M | if ((((*in >= 0x61) && (*in <= 0x7A)) || |
3285 | 21.2M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3286 | 21.2M | (*in == '_')) && (in < e)) { |
3287 | 17.9M | in++; |
3288 | 95.2M | while ((((*in >= 0x61) && (*in <= 0x7A)) || |
3289 | 95.2M | ((*in >= 0x41) && (*in <= 0x5A)) || |
3290 | 95.2M | ((*in >= 0x30) && (*in <= 0x39)) || |
3291 | 95.2M | (*in == '_') || (*in == '-') || |
3292 | 95.2M | (*in == '.')) && (in < e)) |
3293 | 77.2M | in++; |
3294 | 17.9M | if (in >= e) |
3295 | 3.68k | goto complex; |
3296 | 17.9M | if ((*in > 0) && (*in < 0x80)) { |
3297 | 17.8M | count = in - ctxt->input->cur; |
3298 | 17.8M | if (count > maxLength) { |
3299 | 276 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); |
3300 | 276 | return(ret); |
3301 | 276 | } |
3302 | 17.8M | ret = xmlDictLookupHashed(ctxt->dict, ctxt->input->cur, count); |
3303 | 17.8M | ctxt->input->cur = in; |
3304 | 17.8M | ctxt->input->col += count; |
3305 | 17.8M | if (ret.name == NULL) { |
3306 | 4 | xmlErrMemory(ctxt); |
3307 | 4 | } |
3308 | 17.8M | return(ret); |
3309 | 17.8M | } |
3310 | 17.9M | } |
3311 | 3.47M | complex: |
3312 | 3.47M | return(xmlParseNCNameComplex(ctxt)); |
3313 | 21.2M | } |
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 | 510k | xmlParseNameAndCompare(xmlParserCtxtPtr ctxt, xmlChar const *other) { |
3327 | 510k | register const xmlChar *cmp = other; |
3328 | 510k | register const xmlChar *in; |
3329 | 510k | const xmlChar *ret; |
3330 | | |
3331 | 510k | GROW; |
3332 | | |
3333 | 510k | in = ctxt->input->cur; |
3334 | 1.73M | while (*in != 0 && *in == *cmp) { |
3335 | 1.22M | ++in; |
3336 | 1.22M | ++cmp; |
3337 | 1.22M | } |
3338 | 510k | if (*cmp == 0 && (*in == '>' || IS_BLANK_CH (*in))) { |
3339 | | /* success */ |
3340 | 368k | ctxt->input->col += in - ctxt->input->cur; |
3341 | 368k | ctxt->input->cur = in; |
3342 | 368k | return (const xmlChar*) 1; |
3343 | 368k | } |
3344 | | /* failure (or end of input buffer), check with full function */ |
3345 | 142k | ret = xmlParseName (ctxt); |
3346 | | /* strings coming from the dictionary direct compare possible */ |
3347 | 142k | if (ret == other) { |
3348 | 978 | return (const xmlChar*) 1; |
3349 | 978 | } |
3350 | 141k | return ret; |
3351 | 142k | } |
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 | 503k | xmlParseStringName(xmlParserCtxtPtr ctxt, const xmlChar** str) { |
3364 | 503k | xmlChar *ret; |
3365 | 503k | const xmlChar *cur = *str; |
3366 | 503k | int flags = 0; |
3367 | 503k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3368 | 0 | XML_MAX_TEXT_LENGTH : |
3369 | 503k | XML_MAX_NAME_LENGTH; |
3370 | | |
3371 | 503k | if (ctxt->options & XML_PARSE_OLD10) |
3372 | 0 | flags |= XML_SCAN_OLD10; |
3373 | | |
3374 | 503k | cur = xmlScanName(*str, maxLength, flags); |
3375 | 503k | if (cur == NULL) { |
3376 | 101 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NCName"); |
3377 | 101 | return(NULL); |
3378 | 101 | } |
3379 | 503k | if (cur == *str) |
3380 | 7.63k | return(NULL); |
3381 | | |
3382 | 495k | ret = xmlStrndup(*str, cur - *str); |
3383 | 495k | if (ret == NULL) |
3384 | 13 | xmlErrMemory(ctxt); |
3385 | 495k | *str = cur; |
3386 | 495k | return(ret); |
3387 | 503k | } |
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 | 148k | xmlParseNmtoken(xmlParserCtxt *ctxt) { |
3404 | 148k | xmlChar buf[XML_MAX_NAMELEN + 5]; |
3405 | 148k | xmlChar *ret; |
3406 | 148k | int len = 0, l; |
3407 | 148k | int c; |
3408 | 148k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3409 | 0 | XML_MAX_TEXT_LENGTH : |
3410 | 148k | XML_MAX_NAME_LENGTH; |
3411 | 148k | int old10 = (ctxt->options & XML_PARSE_OLD10) ? 1 : 0; |
3412 | | |
3413 | 148k | c = xmlCurrentChar(ctxt, &l); |
3414 | | |
3415 | 690k | while (xmlIsNameChar(c, old10)) { |
3416 | 543k | COPY_BUF(buf, len, c); |
3417 | 543k | NEXTL(l); |
3418 | 543k | c = xmlCurrentChar(ctxt, &l); |
3419 | 543k | 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 | 1.92k | xmlChar *buffer; |
3425 | 1.92k | int max = len * 2; |
3426 | | |
3427 | 1.92k | buffer = xmlMalloc(max); |
3428 | 1.92k | if (buffer == NULL) { |
3429 | 1 | xmlErrMemory(ctxt); |
3430 | 1 | return(NULL); |
3431 | 1 | } |
3432 | 1.92k | memcpy(buffer, buf, len); |
3433 | 15.0M | while (xmlIsNameChar(c, old10)) { |
3434 | 15.0M | if (len + 10 > max) { |
3435 | 6.61k | xmlChar *tmp; |
3436 | 6.61k | int newSize; |
3437 | | |
3438 | 6.61k | newSize = xmlGrowCapacity(max, 1, 1, maxLength); |
3439 | 6.61k | if (newSize < 0) { |
3440 | 312 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NmToken"); |
3441 | 312 | xmlFree(buffer); |
3442 | 312 | return(NULL); |
3443 | 312 | } |
3444 | 6.30k | tmp = xmlRealloc(buffer, newSize); |
3445 | 6.30k | if (tmp == NULL) { |
3446 | 2 | xmlErrMemory(ctxt); |
3447 | 2 | xmlFree(buffer); |
3448 | 2 | return(NULL); |
3449 | 2 | } |
3450 | 6.29k | buffer = tmp; |
3451 | 6.29k | max = newSize; |
3452 | 6.29k | } |
3453 | 15.0M | COPY_BUF(buffer, len, c); |
3454 | 15.0M | NEXTL(l); |
3455 | 15.0M | c = xmlCurrentChar(ctxt, &l); |
3456 | 15.0M | } |
3457 | 1.60k | buffer[len] = 0; |
3458 | 1.60k | return(buffer); |
3459 | 1.92k | } |
3460 | 543k | } |
3461 | 146k | if (len == 0) |
3462 | 60.4k | return(NULL); |
3463 | 86.5k | if (len > maxLength) { |
3464 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "NmToken"); |
3465 | 0 | return(NULL); |
3466 | 0 | } |
3467 | 86.5k | ret = xmlStrndup(buf, len); |
3468 | 86.5k | if (ret == NULL) |
3469 | 6 | xmlErrMemory(ctxt); |
3470 | 86.5k | return(ret); |
3471 | 86.5k | } |
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 | 129k | const xmlChar *str, int length, int depth) { |
3485 | 129k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
3486 | 129k | const xmlChar *end, *chunk; |
3487 | 129k | int c, l; |
3488 | | |
3489 | 129k | if (str == NULL) |
3490 | 5.09k | return; |
3491 | | |
3492 | 124k | depth += 1; |
3493 | 124k | 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 | 124k | end = str + length; |
3500 | 124k | chunk = str; |
3501 | | |
3502 | 94.6M | while ((str < end) && (!PARSER_STOPPED(ctxt))) { |
3503 | 94.5M | c = *str; |
3504 | | |
3505 | 94.5M | if (c >= 0x80) { |
3506 | 80.9M | l = xmlUTF8MultibyteLen(ctxt, str, |
3507 | 80.9M | "invalid character in entity value\n"); |
3508 | 80.9M | if (l == 0) { |
3509 | 17.5M | if (chunk < str) |
3510 | 223k | xmlSBufAddString(buf, chunk, str - chunk); |
3511 | 17.5M | xmlSBufAddReplChar(buf); |
3512 | 17.5M | str += 1; |
3513 | 17.5M | chunk = str; |
3514 | 63.4M | } else { |
3515 | 63.4M | str += l; |
3516 | 63.4M | } |
3517 | 80.9M | } else if (c == '&') { |
3518 | 103k | if (str[1] == '#') { |
3519 | 46.3k | if (chunk < str) |
3520 | 33.6k | xmlSBufAddString(buf, chunk, str - chunk); |
3521 | | |
3522 | 46.3k | c = xmlParseStringCharRef(ctxt, &str); |
3523 | 46.3k | if (c == 0) |
3524 | 4.49k | return; |
3525 | | |
3526 | 41.8k | xmlSBufAddChar(buf, c); |
3527 | | |
3528 | 41.8k | chunk = str; |
3529 | 57.6k | } else { |
3530 | 57.6k | xmlChar *name; |
3531 | | |
3532 | | /* |
3533 | | * General entity references are checked for |
3534 | | * syntactic validity. |
3535 | | */ |
3536 | 57.6k | str++; |
3537 | 57.6k | name = xmlParseStringName(ctxt, &str); |
3538 | | |
3539 | 57.6k | if ((name == NULL) || (*str++ != ';')) { |
3540 | 3.20k | xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_CHAR_ERROR, |
3541 | 3.20k | "EntityValue: '&' forbidden except for entities " |
3542 | 3.20k | "references\n"); |
3543 | 3.20k | xmlFree(name); |
3544 | 3.20k | return; |
3545 | 3.20k | } |
3546 | | |
3547 | 54.4k | xmlFree(name); |
3548 | 54.4k | } |
3549 | 13.4M | } else if (c == '%') { |
3550 | 25.5k | xmlEntityPtr ent; |
3551 | | |
3552 | 25.5k | if (chunk < str) |
3553 | 14.4k | xmlSBufAddString(buf, chunk, str - chunk); |
3554 | | |
3555 | 25.5k | ent = xmlParseStringPEReference(ctxt, &str); |
3556 | 25.5k | if (ent == NULL) |
3557 | 8.35k | return; |
3558 | | |
3559 | 17.1k | if (!PARSER_EXTERNAL(ctxt)) { |
3560 | 309 | xmlFatalErr(ctxt, XML_ERR_ENTITY_PE_INTERNAL, NULL); |
3561 | 309 | return; |
3562 | 309 | } |
3563 | | |
3564 | 16.8k | 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 | 6.93k | if (((ctxt->options & XML_PARSE_NO_XXE) == 0) && |
3572 | 6.93k | ((ctxt->replaceEntities) || |
3573 | 6.93k | (ctxt->validate))) { |
3574 | 6.93k | xmlLoadEntityContent(ctxt, ent); |
3575 | 6.93k | } 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 | 6.93k | } |
3581 | | |
3582 | | /* |
3583 | | * TODO: Skip if ent->content is still NULL. |
3584 | | */ |
3585 | | |
3586 | 16.8k | if (xmlParserEntityCheck(ctxt, ent->length)) |
3587 | 9 | return; |
3588 | | |
3589 | 16.8k | if (ent->flags & XML_ENT_EXPANDING) { |
3590 | 1.30k | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
3591 | 1.30k | xmlHaltParser(ctxt); |
3592 | 1.30k | return; |
3593 | 1.30k | } |
3594 | | |
3595 | 15.5k | ent->flags |= XML_ENT_EXPANDING; |
3596 | 15.5k | xmlExpandPEsInEntityValue(ctxt, buf, ent->content, ent->length, |
3597 | 15.5k | depth); |
3598 | 15.5k | ent->flags &= ~XML_ENT_EXPANDING; |
3599 | | |
3600 | 15.5k | chunk = str; |
3601 | 13.4M | } else { |
3602 | | /* Normal ASCII char */ |
3603 | 13.4M | if (!IS_BYTE_CHAR(c)) { |
3604 | 3.68M | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, |
3605 | 3.68M | "invalid character in entity value\n"); |
3606 | 3.68M | if (chunk < str) |
3607 | 50.4k | xmlSBufAddString(buf, chunk, str - chunk); |
3608 | 3.68M | xmlSBufAddReplChar(buf); |
3609 | 3.68M | str += 1; |
3610 | 3.68M | chunk = str; |
3611 | 9.78M | } else { |
3612 | 9.78M | str += 1; |
3613 | 9.78M | } |
3614 | 13.4M | } |
3615 | 94.5M | } |
3616 | | |
3617 | 106k | if (chunk < str) |
3618 | 99.3k | xmlSBufAddString(buf, chunk, str - chunk); |
3619 | 106k | } |
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 | 116k | xmlParseEntityValue(xmlParserCtxt *ctxt, xmlChar **orig) { |
3635 | 116k | unsigned maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
3636 | 0 | XML_MAX_HUGE_LENGTH : |
3637 | 116k | XML_MAX_TEXT_LENGTH; |
3638 | 116k | xmlSBuf buf; |
3639 | 116k | const xmlChar *start; |
3640 | 116k | int quote, length; |
3641 | | |
3642 | 116k | xmlSBufInit(&buf, maxLength); |
3643 | | |
3644 | 116k | GROW; |
3645 | | |
3646 | 116k | quote = CUR; |
3647 | 116k | if ((quote != '"') && (quote != '\'')) { |
3648 | 0 | xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL); |
3649 | 0 | return(NULL); |
3650 | 0 | } |
3651 | 116k | CUR_PTR++; |
3652 | | |
3653 | 116k | length = 0; |
3654 | | |
3655 | | /* |
3656 | | * Copy raw content of the entity into a buffer |
3657 | | */ |
3658 | 218M | while (1) { |
3659 | 218M | int c; |
3660 | | |
3661 | 218M | if (PARSER_STOPPED(ctxt)) |
3662 | 77 | goto error; |
3663 | | |
3664 | 218M | if (CUR_PTR >= ctxt->input->end) { |
3665 | 1.05k | xmlFatalErrMsg(ctxt, XML_ERR_ENTITY_NOT_FINISHED, NULL); |
3666 | 1.05k | goto error; |
3667 | 1.05k | } |
3668 | | |
3669 | 218M | c = CUR; |
3670 | | |
3671 | 218M | if (c == 0) { |
3672 | 1.00k | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, |
3673 | 1.00k | "invalid character in entity value\n"); |
3674 | 1.00k | goto error; |
3675 | 1.00k | } |
3676 | 218M | if (c == quote) |
3677 | 114k | break; |
3678 | 218M | NEXTL(1); |
3679 | 218M | length += 1; |
3680 | | |
3681 | | /* |
3682 | | * TODO: Check growth threshold |
3683 | | */ |
3684 | 218M | if (ctxt->input->end - CUR_PTR < 10) |
3685 | 38.0k | GROW; |
3686 | 218M | } |
3687 | | |
3688 | 114k | start = CUR_PTR - length; |
3689 | | |
3690 | 114k | if (orig != NULL) { |
3691 | 114k | *orig = xmlStrndup(start, length); |
3692 | 114k | if (*orig == NULL) |
3693 | 4 | xmlErrMemory(ctxt); |
3694 | 114k | } |
3695 | | |
3696 | 114k | xmlExpandPEsInEntityValue(ctxt, &buf, start, length, ctxt->inputNr); |
3697 | | |
3698 | 114k | NEXTL(1); |
3699 | | |
3700 | 114k | return(xmlSBufFinish(&buf, NULL, ctxt, "entity length too long")); |
3701 | | |
3702 | 2.13k | error: |
3703 | 2.13k | xmlSBufCleanup(&buf, ctxt, "entity length too long"); |
3704 | 2.13k | return(NULL); |
3705 | 116k | } |
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 | 52 | xmlCheckEntityInAttValue(xmlParserCtxtPtr ctxt, xmlEntityPtr pent, int depth) { |
3717 | 52 | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
3718 | 52 | const xmlChar *str; |
3719 | 52 | unsigned long expandedSize = pent->length; |
3720 | 52 | int c, flags; |
3721 | | |
3722 | 52 | depth += 1; |
3723 | 52 | 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 | 52 | 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 | 52 | if (ctxt->inSubset == 0) |
3741 | 52 | flags = XML_ENT_CHECKED | XML_ENT_VALIDATED; |
3742 | 0 | else |
3743 | 0 | flags = XML_ENT_VALIDATED; |
3744 | | |
3745 | 52 | str = pent->content; |
3746 | 52 | 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 | 3.28M | while (!PARSER_STOPPED(ctxt)) { |
3756 | 3.28M | c = *str; |
3757 | | |
3758 | 3.28M | if (c != '&') { |
3759 | 3.28M | if (c == 0) |
3760 | 51 | break; |
3761 | | |
3762 | 3.28M | if (c == '<') |
3763 | 7.67k | xmlFatalErrMsgStr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, |
3764 | 7.67k | "'<' in entity '%s' is not allowed in attributes " |
3765 | 7.67k | "values\n", pent->name); |
3766 | | |
3767 | 3.28M | str += 1; |
3768 | 3.28M | } else if (str[1] == '#') { |
3769 | 1 | int val; |
3770 | | |
3771 | 1 | val = xmlParseStringCharRef(ctxt, &str); |
3772 | 1 | if (val == 0) { |
3773 | 1 | pent->content[0] = 0; |
3774 | 1 | break; |
3775 | 1 | } |
3776 | 4.24k | } else { |
3777 | 4.24k | xmlChar *name; |
3778 | 4.24k | xmlEntityPtr ent; |
3779 | | |
3780 | 4.24k | name = xmlParseStringEntityRef(ctxt, &str); |
3781 | 4.24k | if (name == NULL) { |
3782 | 0 | pent->content[0] = 0; |
3783 | 0 | break; |
3784 | 0 | } |
3785 | | |
3786 | 4.24k | ent = xmlLookupGeneralEntity(ctxt, name, /* inAttr */ 1); |
3787 | 4.24k | xmlFree(name); |
3788 | | |
3789 | 4.24k | if ((ent != NULL) && |
3790 | 4.24k | (ent->etype != XML_INTERNAL_PREDEFINED_ENTITY)) { |
3791 | 2.76k | if ((ent->flags & flags) != flags) { |
3792 | 34 | pent->flags |= XML_ENT_EXPANDING; |
3793 | 34 | xmlCheckEntityInAttValue(ctxt, ent, depth); |
3794 | 34 | pent->flags &= ~XML_ENT_EXPANDING; |
3795 | 34 | } |
3796 | | |
3797 | 2.76k | xmlSaturatedAdd(&expandedSize, ent->expandedSize); |
3798 | 2.76k | xmlSaturatedAdd(&expandedSize, XML_ENT_FIXED_COST); |
3799 | 2.76k | } |
3800 | 4.24k | } |
3801 | 3.28M | } |
3802 | | |
3803 | 52 | done: |
3804 | 52 | if (ctxt->inSubset == 0) |
3805 | 52 | pent->expandedSize = expandedSize; |
3806 | | |
3807 | 52 | pent->flags |= flags; |
3808 | 52 | } |
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 | 491k | int *inSpace, int depth, int check) { |
3828 | 491k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 40 : 20; |
3829 | 491k | int c, chunkSize; |
3830 | 491k | int normChange = 0; |
3831 | | |
3832 | 491k | if (str == NULL) |
3833 | 0 | return(0); |
3834 | | |
3835 | 491k | depth += 1; |
3836 | 491k | 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 | 491k | if (pent != NULL) { |
3843 | 491k | if (pent->flags & XML_ENT_EXPANDING) { |
3844 | 264 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
3845 | 264 | xmlHaltParser(ctxt); |
3846 | 264 | return(0); |
3847 | 264 | } |
3848 | | |
3849 | 491k | if (check) { |
3850 | 491k | if (xmlParserEntityCheck(ctxt, pent->length)) |
3851 | 453 | return(0); |
3852 | 491k | } |
3853 | 491k | } |
3854 | | |
3855 | 490k | chunkSize = 0; |
3856 | | |
3857 | | /* |
3858 | | * Note that entity values are already validated. No special |
3859 | | * handling for multi-byte characters is needed. |
3860 | | */ |
3861 | 1.20G | while (!PARSER_STOPPED(ctxt)) { |
3862 | 1.20G | c = *str; |
3863 | | |
3864 | 1.20G | if (c != '&') { |
3865 | 1.20G | if (c == 0) |
3866 | 449k | 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 | 1.20G | if ((pent != NULL) && (c == '<')) { |
3874 | 39.0k | xmlFatalErrMsgStr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, |
3875 | 39.0k | "'<' in entity '%s' is not allowed in attributes " |
3876 | 39.0k | "values\n", pent->name); |
3877 | 39.0k | break; |
3878 | 39.0k | } |
3879 | | |
3880 | 1.20G | if (c <= 0x20) { |
3881 | 4.01M | if ((normalize) && (*inSpace)) { |
3882 | | /* Skip char */ |
3883 | 177k | if (chunkSize > 0) { |
3884 | 37.0k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3885 | 37.0k | chunkSize = 0; |
3886 | 37.0k | } |
3887 | 177k | normChange = 1; |
3888 | 3.83M | } else if (c < 0x20) { |
3889 | 3.24M | if (chunkSize > 0) { |
3890 | 393k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3891 | 393k | chunkSize = 0; |
3892 | 393k | } |
3893 | | |
3894 | 3.24M | xmlSBufAddCString(buf, " ", 1); |
3895 | 3.24M | } else { |
3896 | 590k | chunkSize += 1; |
3897 | 590k | } |
3898 | | |
3899 | 4.01M | *inSpace = 1; |
3900 | 1.19G | } else { |
3901 | 1.19G | chunkSize += 1; |
3902 | 1.19G | *inSpace = 0; |
3903 | 1.19G | } |
3904 | | |
3905 | 1.20G | str += 1; |
3906 | 1.20G | } else if (str[1] == '#') { |
3907 | 142k | int val; |
3908 | | |
3909 | 142k | if (chunkSize > 0) { |
3910 | 132k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3911 | 132k | chunkSize = 0; |
3912 | 132k | } |
3913 | | |
3914 | 142k | val = xmlParseStringCharRef(ctxt, &str); |
3915 | 142k | if (val == 0) { |
3916 | 769 | if (pent != NULL) |
3917 | 769 | pent->content[0] = 0; |
3918 | 769 | break; |
3919 | 769 | } |
3920 | | |
3921 | 142k | if (val == ' ') { |
3922 | 8.70k | if ((normalize) && (*inSpace)) |
3923 | 0 | normChange = 1; |
3924 | 8.70k | else |
3925 | 8.70k | xmlSBufAddCString(buf, " ", 1); |
3926 | 8.70k | *inSpace = 1; |
3927 | 133k | } else { |
3928 | 133k | xmlSBufAddChar(buf, val); |
3929 | 133k | *inSpace = 0; |
3930 | 133k | } |
3931 | 415k | } else { |
3932 | 415k | xmlChar *name; |
3933 | 415k | xmlEntityPtr ent; |
3934 | | |
3935 | 415k | if (chunkSize > 0) { |
3936 | 275k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3937 | 275k | chunkSize = 0; |
3938 | 275k | } |
3939 | | |
3940 | 415k | name = xmlParseStringEntityRef(ctxt, &str); |
3941 | 415k | if (name == NULL) { |
3942 | 1.13k | if (pent != NULL) |
3943 | 1.13k | pent->content[0] = 0; |
3944 | 1.13k | break; |
3945 | 1.13k | } |
3946 | | |
3947 | 414k | ent = xmlLookupGeneralEntity(ctxt, name, /* inAttr */ 1); |
3948 | 414k | xmlFree(name); |
3949 | | |
3950 | 414k | if ((ent != NULL) && |
3951 | 414k | (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { |
3952 | 36.2k | 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 | 36.2k | xmlSBufAddString(buf, ent->content, ent->length); |
3959 | | |
3960 | 36.2k | *inSpace = 0; |
3961 | 378k | } else if ((ent != NULL) && (ent->content != NULL)) { |
3962 | 301k | if (pent != NULL) |
3963 | 301k | pent->flags |= XML_ENT_EXPANDING; |
3964 | 301k | normChange |= xmlExpandEntityInAttValue(ctxt, buf, |
3965 | 301k | ent->content, ent, normalize, inSpace, depth, check); |
3966 | 301k | if (pent != NULL) |
3967 | 301k | pent->flags &= ~XML_ENT_EXPANDING; |
3968 | 301k | } |
3969 | 414k | } |
3970 | 1.20G | } |
3971 | | |
3972 | 490k | if (chunkSize > 0) |
3973 | 394k | xmlSBufAddString(buf, str - chunkSize, chunkSize); |
3974 | | |
3975 | 490k | return(normChange); |
3976 | 491k | } |
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 | 5.30M | int special, int isNamespace) { |
4046 | 5.30M | unsigned maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4047 | 0 | XML_MAX_HUGE_LENGTH : |
4048 | 5.30M | XML_MAX_TEXT_LENGTH; |
4049 | 5.30M | xmlSBuf buf; |
4050 | 5.30M | xmlChar *ret; |
4051 | 5.30M | int c, l, quote, entFlags, chunkSize; |
4052 | 5.30M | int inSpace = 1; |
4053 | 5.30M | int replaceEntities; |
4054 | 5.30M | int normalize = (special & XML_SPECIAL_TYPE_MASK) != 0; |
4055 | 5.30M | int attvalFlags = 0; |
4056 | | |
4057 | | /* Always expand namespace URIs */ |
4058 | 5.30M | replaceEntities = (ctxt->replaceEntities) || (isNamespace); |
4059 | | |
4060 | 5.30M | xmlSBufInit(&buf, maxLength); |
4061 | | |
4062 | 5.30M | GROW; |
4063 | | |
4064 | 5.30M | quote = CUR; |
4065 | 5.30M | if ((quote != '"') && (quote != '\'')) { |
4066 | 22.2k | xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL); |
4067 | 22.2k | return(NULL); |
4068 | 22.2k | } |
4069 | 5.28M | NEXTL(1); |
4070 | | |
4071 | 5.28M | if (ctxt->inSubset == 0) |
4072 | 5.22M | entFlags = XML_ENT_CHECKED | XML_ENT_VALIDATED; |
4073 | 53.3k | else |
4074 | 53.3k | entFlags = XML_ENT_VALIDATED; |
4075 | | |
4076 | 5.28M | inSpace = 1; |
4077 | 5.28M | chunkSize = 0; |
4078 | | |
4079 | 530M | while (1) { |
4080 | 530M | if (PARSER_STOPPED(ctxt)) |
4081 | 794 | goto error; |
4082 | | |
4083 | 530M | if (CUR_PTR >= ctxt->input->end) { |
4084 | 12.0k | xmlFatalErrMsg(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED, |
4085 | 12.0k | "AttValue: ' expected\n"); |
4086 | 12.0k | goto error; |
4087 | 12.0k | } |
4088 | | |
4089 | | /* |
4090 | | * TODO: Check growth threshold |
4091 | | */ |
4092 | 530M | if (ctxt->input->end - CUR_PTR < 10) |
4093 | 260k | GROW; |
4094 | | |
4095 | 530M | c = CUR; |
4096 | | |
4097 | 530M | if (c >= 0x80) { |
4098 | 256M | l = xmlUTF8MultibyteLen(ctxt, CUR_PTR, |
4099 | 256M | "invalid character in attribute value\n"); |
4100 | 256M | if (l == 0) { |
4101 | 45.5M | if (chunkSize > 0) { |
4102 | 423k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4103 | 423k | chunkSize = 0; |
4104 | 423k | } |
4105 | 45.5M | xmlSBufAddReplChar(&buf); |
4106 | 45.5M | NEXTL(1); |
4107 | 211M | } else { |
4108 | 211M | chunkSize += l; |
4109 | 211M | NEXTL(l); |
4110 | 211M | } |
4111 | | |
4112 | 256M | inSpace = 0; |
4113 | 273M | } else if (c != '&') { |
4114 | 272M | if (c > 0x20) { |
4115 | 198M | if (c == quote) |
4116 | 5.26M | break; |
4117 | | |
4118 | 193M | if (c == '<') |
4119 | 2.05M | xmlFatalErr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, NULL); |
4120 | | |
4121 | 193M | chunkSize += 1; |
4122 | 193M | inSpace = 0; |
4123 | 193M | } else if (!IS_BYTE_CHAR(c)) { |
4124 | 64.3M | xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, |
4125 | 64.3M | "invalid character in attribute value\n"); |
4126 | 64.3M | if (chunkSize > 0) { |
4127 | 199k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4128 | 199k | chunkSize = 0; |
4129 | 199k | } |
4130 | 64.3M | xmlSBufAddReplChar(&buf); |
4131 | 64.3M | inSpace = 0; |
4132 | 64.3M | } else { |
4133 | | /* Whitespace */ |
4134 | 9.91M | if ((normalize) && (inSpace)) { |
4135 | | /* Skip char */ |
4136 | 101k | if (chunkSize > 0) { |
4137 | 19.1k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4138 | 19.1k | chunkSize = 0; |
4139 | 19.1k | } |
4140 | 101k | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4141 | 9.81M | } else if (c < 0x20) { |
4142 | | /* Convert to space */ |
4143 | 7.36M | if (chunkSize > 0) { |
4144 | 582k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4145 | 582k | chunkSize = 0; |
4146 | 582k | } |
4147 | | |
4148 | 7.36M | xmlSBufAddCString(&buf, " ", 1); |
4149 | 7.36M | } else { |
4150 | 2.45M | chunkSize += 1; |
4151 | 2.45M | } |
4152 | | |
4153 | 9.91M | inSpace = 1; |
4154 | | |
4155 | 9.91M | if ((c == 0xD) && (NXT(1) == 0xA)) |
4156 | 29.8k | CUR_PTR++; |
4157 | 9.91M | } |
4158 | | |
4159 | 267M | NEXTL(1); |
4160 | 267M | } else if (NXT(1) == '#') { |
4161 | 134k | int val; |
4162 | | |
4163 | 134k | if (chunkSize > 0) { |
4164 | 75.3k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4165 | 75.3k | chunkSize = 0; |
4166 | 75.3k | } |
4167 | | |
4168 | 134k | val = xmlParseCharRef(ctxt); |
4169 | 134k | if (val == 0) |
4170 | 2.81k | goto error; |
4171 | | |
4172 | 131k | if ((val == '&') && (!replaceEntities)) { |
4173 | | /* |
4174 | | * The reparsing will be done in xmlNodeParseContent() |
4175 | | * called from SAX2.c |
4176 | | */ |
4177 | 119 | xmlSBufAddCString(&buf, "&", 5); |
4178 | 119 | inSpace = 0; |
4179 | 131k | } else if (val == ' ') { |
4180 | 6.13k | if ((normalize) && (inSpace)) |
4181 | 783 | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4182 | 5.35k | else |
4183 | 5.35k | xmlSBufAddCString(&buf, " ", 1); |
4184 | 6.13k | inSpace = 1; |
4185 | 125k | } else { |
4186 | 125k | xmlSBufAddChar(&buf, val); |
4187 | 125k | inSpace = 0; |
4188 | 125k | } |
4189 | 586k | } else { |
4190 | 586k | const xmlChar *name; |
4191 | 586k | xmlEntityPtr ent; |
4192 | | |
4193 | 586k | if (chunkSize > 0) { |
4194 | 251k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4195 | 251k | chunkSize = 0; |
4196 | 251k | } |
4197 | | |
4198 | 586k | name = xmlParseEntityRefInternal(ctxt); |
4199 | 586k | if (name == NULL) { |
4200 | | /* |
4201 | | * Probably a literal '&' which wasn't escaped. |
4202 | | * TODO: Handle gracefully in recovery mode. |
4203 | | */ |
4204 | 247k | continue; |
4205 | 247k | } |
4206 | | |
4207 | 339k | ent = xmlLookupGeneralEntity(ctxt, name, /* isAttr */ 1); |
4208 | 339k | if (ent == NULL) |
4209 | 92.5k | continue; |
4210 | | |
4211 | 246k | if (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY) { |
4212 | 35.4k | if ((ent->content[0] == '&') && (!replaceEntities)) |
4213 | 433 | xmlSBufAddCString(&buf, "&", 5); |
4214 | 34.9k | else |
4215 | 34.9k | xmlSBufAddString(&buf, ent->content, ent->length); |
4216 | 35.4k | inSpace = 0; |
4217 | 211k | } else if (replaceEntities) { |
4218 | 189k | if (xmlExpandEntityInAttValue(ctxt, &buf, |
4219 | 189k | ent->content, ent, normalize, &inSpace, ctxt->inputNr, |
4220 | 189k | /* check */ 1) > 0) |
4221 | 9.17k | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4222 | 189k | } else { |
4223 | 21.2k | if ((ent->flags & entFlags) != entFlags) |
4224 | 18 | xmlCheckEntityInAttValue(ctxt, ent, ctxt->inputNr); |
4225 | | |
4226 | 21.2k | if (xmlParserEntityCheck(ctxt, ent->expandedSize)) { |
4227 | 12 | ent->content[0] = 0; |
4228 | 12 | goto error; |
4229 | 12 | } |
4230 | | |
4231 | | /* |
4232 | | * Just output the reference |
4233 | | */ |
4234 | 21.2k | xmlSBufAddCString(&buf, "&", 1); |
4235 | 21.2k | xmlSBufAddString(&buf, ent->name, xmlStrlen(ent->name)); |
4236 | 21.2k | xmlSBufAddCString(&buf, ";", 1); |
4237 | | |
4238 | 21.2k | inSpace = 0; |
4239 | 21.2k | } |
4240 | 246k | } |
4241 | 530M | } |
4242 | | |
4243 | 5.26M | if ((buf.mem == NULL) && (outFlags != NULL)) { |
4244 | 4.91M | ret = (xmlChar *) CUR_PTR - chunkSize; |
4245 | | |
4246 | 4.91M | if (attlen != NULL) |
4247 | 4.91M | *attlen = chunkSize; |
4248 | 4.91M | if ((normalize) && (inSpace) && (chunkSize > 0)) { |
4249 | 304 | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4250 | 304 | *attlen -= 1; |
4251 | 304 | } |
4252 | | |
4253 | | /* Report potential error */ |
4254 | 4.91M | xmlSBufCleanup(&buf, ctxt, "AttValue length too long"); |
4255 | 4.91M | } else { |
4256 | 351k | if (chunkSize > 0) |
4257 | 208k | xmlSBufAddString(&buf, CUR_PTR - chunkSize, chunkSize); |
4258 | | |
4259 | 351k | if ((normalize) && (inSpace) && (buf.size > 0)) { |
4260 | 528 | attvalFlags |= XML_ATTVAL_NORM_CHANGE; |
4261 | 528 | buf.size--; |
4262 | 528 | } |
4263 | | |
4264 | 351k | ret = xmlSBufFinish(&buf, attlen, ctxt, "AttValue length too long"); |
4265 | 351k | attvalFlags |= XML_ATTVAL_ALLOC; |
4266 | | |
4267 | 351k | if (ret != NULL) { |
4268 | 351k | if (attlen != NULL) |
4269 | 298k | *attlen = buf.size; |
4270 | 351k | } |
4271 | 351k | } |
4272 | | |
4273 | 5.26M | if (outFlags != NULL) |
4274 | 5.21M | *outFlags = attvalFlags; |
4275 | | |
4276 | 5.26M | NEXTL(1); |
4277 | | |
4278 | 5.26M | return(ret); |
4279 | | |
4280 | 15.7k | error: |
4281 | 15.7k | xmlSBufCleanup(&buf, ctxt, "AttValue length too long"); |
4282 | 15.7k | return(NULL); |
4283 | 5.28M | } |
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 | 60.1k | xmlParseAttValue(xmlParserCtxt *ctxt) { |
4323 | 60.1k | if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL); |
4324 | 60.1k | return(xmlParseAttValueInternal(ctxt, NULL, NULL, 0, 0)); |
4325 | 60.1k | } |
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 | 94.2k | xmlParseSystemLiteral(xmlParserCtxt *ctxt) { |
4340 | 94.2k | xmlChar *buf = NULL; |
4341 | 94.2k | int len = 0; |
4342 | 94.2k | int size = XML_PARSER_BUFFER_SIZE; |
4343 | 94.2k | int cur, l; |
4344 | 94.2k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4345 | 0 | XML_MAX_TEXT_LENGTH : |
4346 | 94.2k | XML_MAX_NAME_LENGTH; |
4347 | 94.2k | xmlChar stop; |
4348 | | |
4349 | 94.2k | if (RAW == '"') { |
4350 | 76.8k | NEXT; |
4351 | 76.8k | stop = '"'; |
4352 | 76.8k | } else if (RAW == '\'') { |
4353 | 14.6k | NEXT; |
4354 | 14.6k | stop = '\''; |
4355 | 14.6k | } else { |
4356 | 2.73k | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); |
4357 | 2.73k | return(NULL); |
4358 | 2.73k | } |
4359 | | |
4360 | 91.4k | buf = xmlMalloc(size); |
4361 | 91.4k | if (buf == NULL) { |
4362 | 15 | xmlErrMemory(ctxt); |
4363 | 15 | return(NULL); |
4364 | 15 | } |
4365 | 91.4k | cur = xmlCurrentCharRecover(ctxt, &l); |
4366 | 11.2M | while ((IS_CHAR(cur)) && (cur != stop)) { /* checked */ |
4367 | 11.1M | if (len + 5 >= size) { |
4368 | 16.1k | xmlChar *tmp; |
4369 | 16.1k | int newSize; |
4370 | | |
4371 | 16.1k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
4372 | 16.1k | if (newSize < 0) { |
4373 | 342 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "SystemLiteral"); |
4374 | 342 | xmlFree(buf); |
4375 | 342 | return(NULL); |
4376 | 342 | } |
4377 | 15.7k | tmp = xmlRealloc(buf, newSize); |
4378 | 15.7k | if (tmp == NULL) { |
4379 | 3 | xmlFree(buf); |
4380 | 3 | xmlErrMemory(ctxt); |
4381 | 3 | return(NULL); |
4382 | 3 | } |
4383 | 15.7k | buf = tmp; |
4384 | 15.7k | size = newSize; |
4385 | 15.7k | } |
4386 | 11.1M | COPY_BUF(buf, len, cur); |
4387 | 11.1M | NEXTL(l); |
4388 | 11.1M | cur = xmlCurrentCharRecover(ctxt, &l); |
4389 | 11.1M | } |
4390 | 91.1k | buf[len] = 0; |
4391 | 91.1k | if (!IS_CHAR(cur)) { |
4392 | 1.18k | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); |
4393 | 89.9k | } else { |
4394 | 89.9k | NEXT; |
4395 | 89.9k | } |
4396 | 91.1k | return(buf); |
4397 | 91.4k | } |
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 | 25.9k | xmlParsePubidLiteral(xmlParserCtxt *ctxt) { |
4412 | 25.9k | xmlChar *buf = NULL; |
4413 | 25.9k | int len = 0; |
4414 | 25.9k | int size = XML_PARSER_BUFFER_SIZE; |
4415 | 25.9k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4416 | 0 | XML_MAX_TEXT_LENGTH : |
4417 | 25.9k | XML_MAX_NAME_LENGTH; |
4418 | 25.9k | xmlChar cur; |
4419 | 25.9k | xmlChar stop; |
4420 | | |
4421 | 25.9k | if (RAW == '"') { |
4422 | 9.78k | NEXT; |
4423 | 9.78k | stop = '"'; |
4424 | 16.2k | } else if (RAW == '\'') { |
4425 | 15.2k | NEXT; |
4426 | 15.2k | stop = '\''; |
4427 | 15.2k | } else { |
4428 | 956 | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); |
4429 | 956 | return(NULL); |
4430 | 956 | } |
4431 | 25.0k | buf = xmlMalloc(size); |
4432 | 25.0k | if (buf == NULL) { |
4433 | 1 | xmlErrMemory(ctxt); |
4434 | 1 | return(NULL); |
4435 | 1 | } |
4436 | 25.0k | cur = CUR; |
4437 | 124k | while ((IS_PUBIDCHAR_CH(cur)) && (cur != stop) && |
4438 | 124k | (PARSER_STOPPED(ctxt) == 0)) { /* checked */ |
4439 | 99.1k | if (len + 1 >= size) { |
4440 | 354 | xmlChar *tmp; |
4441 | 354 | int newSize; |
4442 | | |
4443 | 354 | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
4444 | 354 | 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 | 354 | tmp = xmlRealloc(buf, newSize); |
4450 | 354 | if (tmp == NULL) { |
4451 | 1 | xmlErrMemory(ctxt); |
4452 | 1 | xmlFree(buf); |
4453 | 1 | return(NULL); |
4454 | 1 | } |
4455 | 353 | buf = tmp; |
4456 | 353 | size = newSize; |
4457 | 353 | } |
4458 | 99.1k | buf[len++] = cur; |
4459 | 99.1k | NEXT; |
4460 | 99.1k | cur = CUR; |
4461 | 99.1k | } |
4462 | 25.0k | buf[len] = 0; |
4463 | 25.0k | if (cur != stop) { |
4464 | 1.42k | xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); |
4465 | 23.6k | } else { |
4466 | 23.6k | NEXTL(1); |
4467 | 23.6k | } |
4468 | 25.0k | return(buf); |
4469 | 25.0k | } |
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 | 8.75M | int isBlank) { |
4514 | 8.75M | int checkBlanks; |
4515 | | |
4516 | 8.75M | if ((ctxt->sax == NULL) || (ctxt->disableSAX)) |
4517 | 3.37M | return; |
4518 | | |
4519 | 5.38M | checkBlanks = (!ctxt->keepBlanks) || |
4520 | 5.38M | (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 | 5.38M | if ((checkBlanks) && |
4528 | 5.38M | (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 | 5.38M | } else { |
4533 | 5.38M | if (ctxt->sax->characters != NULL) |
4534 | 5.38M | 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 | 5.38M | if ((checkBlanks) && (*ctxt->space == -1)) |
4541 | 0 | *ctxt->space = -2; |
4542 | 5.38M | } |
4543 | 5.38M | } |
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 | 13.8M | xmlParseCharDataInternal(xmlParserCtxtPtr ctxt, int partial) { |
4560 | 13.8M | const xmlChar *in; |
4561 | 13.8M | int line = ctxt->input->line; |
4562 | 13.8M | int col = ctxt->input->col; |
4563 | 13.8M | int ccol; |
4564 | 13.8M | int terminate = 0; |
4565 | | |
4566 | 13.8M | GROW; |
4567 | | /* |
4568 | | * Accelerated common case where input don't need to be |
4569 | | * modified before passing it to the handler. |
4570 | | */ |
4571 | 13.8M | in = ctxt->input->cur; |
4572 | 14.0M | do { |
4573 | 16.5M | get_more_space: |
4574 | 22.5M | while (*in == 0x20) { in++; ctxt->input->col++; } |
4575 | 16.5M | if (*in == 0xA) { |
4576 | 40.5M | do { |
4577 | 40.5M | ctxt->input->line++; ctxt->input->col = 1; |
4578 | 40.5M | in++; |
4579 | 40.5M | } while (*in == 0xA); |
4580 | 2.50M | goto get_more_space; |
4581 | 2.50M | } |
4582 | 14.0M | if (*in == '<') { |
4583 | 6.24M | while (in > ctxt->input->cur) { |
4584 | 3.12M | const xmlChar *tmp = ctxt->input->cur; |
4585 | 3.12M | size_t nbchar = in - tmp; |
4586 | | |
4587 | 3.12M | if (nbchar > XML_MAX_ITEMS) |
4588 | 0 | nbchar = XML_MAX_ITEMS; |
4589 | 3.12M | ctxt->input->cur += nbchar; |
4590 | | |
4591 | 3.12M | xmlCharacters(ctxt, tmp, nbchar, 1); |
4592 | 3.12M | } |
4593 | 3.12M | return; |
4594 | 3.12M | } |
4595 | | |
4596 | 11.8M | get_more: |
4597 | 11.8M | ccol = ctxt->input->col; |
4598 | 314M | while (test_char_data[*in]) { |
4599 | 302M | in++; |
4600 | 302M | ccol++; |
4601 | 302M | } |
4602 | 11.8M | ctxt->input->col = ccol; |
4603 | 11.8M | if (*in == 0xA) { |
4604 | 17.7M | do { |
4605 | 17.7M | ctxt->input->line++; ctxt->input->col = 1; |
4606 | 17.7M | in++; |
4607 | 17.7M | } while (*in == 0xA); |
4608 | 794k | goto get_more; |
4609 | 794k | } |
4610 | 11.0M | if (*in == ']') { |
4611 | 128k | size_t avail = ctxt->input->end - in; |
4612 | | |
4613 | 128k | if (partial && avail < 2) { |
4614 | 0 | terminate = 1; |
4615 | 0 | goto invoke_callback; |
4616 | 0 | } |
4617 | 128k | if (in[1] == ']') { |
4618 | 14.0k | if (partial && avail < 3) { |
4619 | 0 | terminate = 1; |
4620 | 0 | goto invoke_callback; |
4621 | 0 | } |
4622 | 14.0k | if (in[2] == '>') |
4623 | 1.32k | xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); |
4624 | 14.0k | } |
4625 | | |
4626 | 128k | in++; |
4627 | 128k | ctxt->input->col++; |
4628 | 128k | goto get_more; |
4629 | 128k | } |
4630 | | |
4631 | 10.9M | invoke_callback: |
4632 | 13.7M | while (in > ctxt->input->cur) { |
4633 | 2.82M | const xmlChar *tmp = ctxt->input->cur; |
4634 | 2.82M | size_t nbchar = in - tmp; |
4635 | | |
4636 | 2.82M | if (nbchar > XML_MAX_ITEMS) |
4637 | 0 | nbchar = XML_MAX_ITEMS; |
4638 | 2.82M | ctxt->input->cur += nbchar; |
4639 | | |
4640 | 2.82M | xmlCharacters(ctxt, tmp, nbchar, 0); |
4641 | | |
4642 | 2.82M | line = ctxt->input->line; |
4643 | 2.82M | col = ctxt->input->col; |
4644 | 2.82M | } |
4645 | 10.9M | ctxt->input->cur = in; |
4646 | 10.9M | if (*in == 0xD) { |
4647 | 217k | in++; |
4648 | 217k | if (*in == 0xA) { |
4649 | 192k | ctxt->input->cur = in; |
4650 | 192k | in++; |
4651 | 192k | ctxt->input->line++; ctxt->input->col = 1; |
4652 | 192k | continue; /* while */ |
4653 | 192k | } |
4654 | 24.6k | in--; |
4655 | 24.6k | } |
4656 | 10.7M | if (*in == '<') { |
4657 | 1.73M | return; |
4658 | 1.73M | } |
4659 | 8.97M | if (*in == '&') { |
4660 | 135k | return; |
4661 | 135k | } |
4662 | 8.84M | if (terminate) { |
4663 | 0 | return; |
4664 | 0 | } |
4665 | 8.84M | SHRINK; |
4666 | 8.84M | GROW; |
4667 | 8.84M | in = ctxt->input->cur; |
4668 | 9.03M | } while (((*in >= 0x20) && (*in <= 0x7F)) || |
4669 | 9.03M | (*in == 0x09) || (*in == 0x0a)); |
4670 | 8.85M | ctxt->input->line = line; |
4671 | 8.85M | ctxt->input->col = col; |
4672 | 8.85M | xmlParseCharDataComplex(ctxt, partial); |
4673 | 8.85M | } |
4674 | | |
4675 | | /** |
4676 | | * Always makes progress if the first char isn't '<' or '&'. |
4677 | | * |
4678 | | * parse a CharData section.this is the fallback function |
4679 | | * of #xmlParseCharData when the parsing requires handling |
4680 | | * of non-ASCII characters. |
4681 | | * |
4682 | | * @param ctxt an XML parser context |
4683 | | * @param partial whether the input can end with truncated UTF-8 |
4684 | | */ |
4685 | | static void |
4686 | 8.85M | xmlParseCharDataComplex(xmlParserCtxtPtr ctxt, int partial) { |
4687 | 8.85M | xmlChar buf[XML_PARSER_BIG_BUFFER_SIZE + 5]; |
4688 | 8.85M | int nbchar = 0; |
4689 | 8.85M | int cur, l; |
4690 | | |
4691 | 8.85M | cur = xmlCurrentCharRecover(ctxt, &l); |
4692 | 283M | while ((cur != '<') && /* checked */ |
4693 | 283M | (cur != '&') && |
4694 | 283M | (IS_CHAR(cur))) { |
4695 | 274M | if (cur == ']') { |
4696 | 250k | size_t avail = ctxt->input->end - ctxt->input->cur; |
4697 | | |
4698 | 250k | if (partial && avail < 2) |
4699 | 0 | break; |
4700 | 250k | if (NXT(1) == ']') { |
4701 | 141k | if (partial && avail < 3) |
4702 | 0 | break; |
4703 | 141k | if (NXT(2) == '>') |
4704 | 7.75k | xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); |
4705 | 141k | } |
4706 | 250k | } |
4707 | | |
4708 | 274M | COPY_BUF(buf, nbchar, cur); |
4709 | | /* move current position before possible calling of ctxt->sax->characters */ |
4710 | 274M | NEXTL(l); |
4711 | 274M | if (nbchar >= XML_PARSER_BIG_BUFFER_SIZE) { |
4712 | 1.52M | buf[nbchar] = 0; |
4713 | | |
4714 | 1.52M | xmlCharacters(ctxt, buf, nbchar, 0); |
4715 | 1.52M | nbchar = 0; |
4716 | 1.52M | SHRINK; |
4717 | 1.52M | } |
4718 | 274M | cur = xmlCurrentCharRecover(ctxt, &l); |
4719 | 274M | } |
4720 | 8.85M | if (nbchar != 0) { |
4721 | 1.28M | buf[nbchar] = 0; |
4722 | | |
4723 | 1.28M | xmlCharacters(ctxt, buf, nbchar, 0); |
4724 | 1.28M | } |
4725 | | /* |
4726 | | * cur == 0 can mean |
4727 | | * |
4728 | | * - End of buffer. |
4729 | | * - An actual 0 character. |
4730 | | * - An incomplete UTF-8 sequence. This is allowed if partial is set. |
4731 | | */ |
4732 | 8.85M | if (ctxt->input->cur < ctxt->input->end) { |
4733 | 8.82M | if ((cur == 0) && (CUR != 0)) { |
4734 | 6.76k | if (partial == 0) { |
4735 | 6.76k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4736 | 6.76k | "Incomplete UTF-8 sequence starting with %02X\n", CUR); |
4737 | 6.76k | NEXTL(1); |
4738 | 6.76k | } |
4739 | 8.81M | } else if ((cur != '<') && (cur != '&') && (cur != ']')) { |
4740 | | /* Generate the error and skip the offending character */ |
4741 | 8.07M | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4742 | 8.07M | "PCDATA invalid Char value %d\n", cur); |
4743 | 8.07M | NEXTL(l); |
4744 | 8.07M | } |
4745 | 8.82M | } |
4746 | 8.85M | } |
4747 | | |
4748 | | /** |
4749 | | * @deprecated Internal function, don't use. |
4750 | | * @param ctxt an XML parser context |
4751 | | * @param cdata unused |
4752 | | */ |
4753 | | void |
4754 | 0 | xmlParseCharData(xmlParserCtxt *ctxt, ATTRIBUTE_UNUSED int cdata) { |
4755 | 0 | xmlParseCharDataInternal(ctxt, 0); |
4756 | 0 | } |
4757 | | |
4758 | | /** |
4759 | | * Parse an External ID or a Public ID |
4760 | | * |
4761 | | * @deprecated Internal function, don't use. |
4762 | | * |
4763 | | * NOTE: Productions [75] and [83] interact badly since [75] can generate |
4764 | | * `'PUBLIC' S PubidLiteral S SystemLiteral` |
4765 | | * |
4766 | | * [75] ExternalID ::= 'SYSTEM' S SystemLiteral |
4767 | | * | 'PUBLIC' S PubidLiteral S SystemLiteral |
4768 | | * |
4769 | | * [83] PublicID ::= 'PUBLIC' S PubidLiteral |
4770 | | * |
4771 | | * @param ctxt an XML parser context |
4772 | | * @param publicId a xmlChar** receiving PubidLiteral |
4773 | | * @param strict indicate whether we should restrict parsing to only |
4774 | | * production [75], see NOTE below |
4775 | | * @returns the function returns SystemLiteral and in the second |
4776 | | * case publicID receives PubidLiteral, is strict is off |
4777 | | * it is possible to return NULL and have publicID set. |
4778 | | */ |
4779 | | |
4780 | | xmlChar * |
4781 | 217k | xmlParseExternalID(xmlParserCtxt *ctxt, xmlChar **publicId, int strict) { |
4782 | 217k | xmlChar *URI = NULL; |
4783 | | |
4784 | 217k | *publicId = NULL; |
4785 | 217k | if (CMP6(CUR_PTR, 'S', 'Y', 'S', 'T', 'E', 'M')) { |
4786 | 73.1k | SKIP(6); |
4787 | 73.1k | if (SKIP_BLANKS == 0) { |
4788 | 682 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
4789 | 682 | "Space required after 'SYSTEM'\n"); |
4790 | 682 | } |
4791 | 73.1k | URI = xmlParseSystemLiteral(ctxt); |
4792 | 73.1k | if (URI == NULL) { |
4793 | 865 | xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); |
4794 | 865 | } |
4795 | 144k | } else if (CMP6(CUR_PTR, 'P', 'U', 'B', 'L', 'I', 'C')) { |
4796 | 25.9k | SKIP(6); |
4797 | 25.9k | if (SKIP_BLANKS == 0) { |
4798 | 1.11k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
4799 | 1.11k | "Space required after 'PUBLIC'\n"); |
4800 | 1.11k | } |
4801 | 25.9k | *publicId = xmlParsePubidLiteral(ctxt); |
4802 | 25.9k | if (*publicId == NULL) { |
4803 | 958 | xmlFatalErr(ctxt, XML_ERR_PUBID_REQUIRED, NULL); |
4804 | 958 | } |
4805 | 25.9k | if (strict) { |
4806 | | /* |
4807 | | * We don't handle [83] so "S SystemLiteral" is required. |
4808 | | */ |
4809 | 19.9k | if (SKIP_BLANKS == 0) { |
4810 | 2.46k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
4811 | 2.46k | "Space required after the Public Identifier\n"); |
4812 | 2.46k | } |
4813 | 19.9k | } else { |
4814 | | /* |
4815 | | * We handle [83] so we return immediately, if |
4816 | | * "S SystemLiteral" is not detected. We skip blanks if no |
4817 | | * system literal was found, but this is harmless since we must |
4818 | | * be at the end of a NotationDecl. |
4819 | | */ |
4820 | 6.03k | if (SKIP_BLANKS == 0) return(NULL); |
4821 | 1.37k | if ((CUR != '\'') && (CUR != '"')) return(NULL); |
4822 | 1.37k | } |
4823 | 21.0k | URI = xmlParseSystemLiteral(ctxt); |
4824 | 21.0k | if (URI == NULL) { |
4825 | 2.22k | xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); |
4826 | 2.22k | } |
4827 | 21.0k | } |
4828 | 212k | return(URI); |
4829 | 217k | } |
4830 | | |
4831 | | /** |
4832 | | * Skip an XML (SGML) comment <!-- .... --> |
4833 | | * The spec says that "For compatibility, the string "--" (double-hyphen) |
4834 | | * must not occur within comments. " |
4835 | | * This is the slow routine in case the accelerator for ascii didn't work |
4836 | | * |
4837 | | * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' |
4838 | | * @param ctxt an XML parser context |
4839 | | * @param buf the already parsed part of the buffer |
4840 | | * @param len number of bytes in the buffer |
4841 | | * @param size allocated size of the buffer |
4842 | | */ |
4843 | | static void |
4844 | | xmlParseCommentComplex(xmlParserCtxtPtr ctxt, xmlChar *buf, |
4845 | 56.7k | size_t len, size_t size) { |
4846 | 56.7k | int q, ql; |
4847 | 56.7k | int r, rl; |
4848 | 56.7k | int cur, l; |
4849 | 56.7k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4850 | 0 | XML_MAX_HUGE_LENGTH : |
4851 | 56.7k | XML_MAX_TEXT_LENGTH; |
4852 | | |
4853 | 56.7k | if (buf == NULL) { |
4854 | 3.32k | len = 0; |
4855 | 3.32k | size = XML_PARSER_BUFFER_SIZE; |
4856 | 3.32k | buf = xmlMalloc(size); |
4857 | 3.32k | if (buf == NULL) { |
4858 | 1 | xmlErrMemory(ctxt); |
4859 | 1 | return; |
4860 | 1 | } |
4861 | 3.32k | } |
4862 | 56.7k | q = xmlCurrentCharRecover(ctxt, &ql); |
4863 | 56.7k | if (q == 0) |
4864 | 5.53k | goto not_terminated; |
4865 | 51.2k | if (!IS_CHAR(q)) { |
4866 | 4.17k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4867 | 4.17k | "xmlParseComment: invalid xmlChar value %d\n", |
4868 | 4.17k | q); |
4869 | 4.17k | xmlFree (buf); |
4870 | 4.17k | return; |
4871 | 4.17k | } |
4872 | 47.0k | NEXTL(ql); |
4873 | 47.0k | r = xmlCurrentCharRecover(ctxt, &rl); |
4874 | 47.0k | if (r == 0) |
4875 | 2.95k | goto not_terminated; |
4876 | 44.0k | if (!IS_CHAR(r)) { |
4877 | 1.41k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4878 | 1.41k | "xmlParseComment: invalid xmlChar value %d\n", |
4879 | 1.41k | r); |
4880 | 1.41k | xmlFree (buf); |
4881 | 1.41k | return; |
4882 | 1.41k | } |
4883 | 42.6k | NEXTL(rl); |
4884 | 42.6k | cur = xmlCurrentCharRecover(ctxt, &l); |
4885 | 42.6k | if (cur == 0) |
4886 | 1.52k | goto not_terminated; |
4887 | 5.34M | while (IS_CHAR(cur) && /* checked */ |
4888 | 5.34M | ((cur != '>') || |
4889 | 5.33M | (r != '-') || (q != '-'))) { |
4890 | 5.30M | if ((r == '-') && (q == '-')) { |
4891 | 27.9k | xmlFatalErr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, NULL); |
4892 | 27.9k | } |
4893 | 5.30M | if (len + 5 >= size) { |
4894 | 17.7k | xmlChar *tmp; |
4895 | 17.7k | int newSize; |
4896 | | |
4897 | 17.7k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
4898 | 17.7k | if (newSize < 0) { |
4899 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4900 | 0 | "Comment too big found", NULL); |
4901 | 0 | xmlFree (buf); |
4902 | 0 | return; |
4903 | 0 | } |
4904 | 17.7k | tmp = xmlRealloc(buf, newSize); |
4905 | 17.7k | if (tmp == NULL) { |
4906 | 7 | xmlErrMemory(ctxt); |
4907 | 7 | xmlFree(buf); |
4908 | 7 | return; |
4909 | 7 | } |
4910 | 17.7k | buf = tmp; |
4911 | 17.7k | size = newSize; |
4912 | 17.7k | } |
4913 | 5.30M | COPY_BUF(buf, len, q); |
4914 | | |
4915 | 5.30M | q = r; |
4916 | 5.30M | ql = rl; |
4917 | 5.30M | r = cur; |
4918 | 5.30M | rl = l; |
4919 | | |
4920 | 5.30M | NEXTL(l); |
4921 | 5.30M | cur = xmlCurrentCharRecover(ctxt, &l); |
4922 | | |
4923 | 5.30M | } |
4924 | 41.1k | buf[len] = 0; |
4925 | 41.1k | if (cur == 0) { |
4926 | 4.32k | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4927 | 4.32k | "Comment not terminated \n<!--%.50s\n", buf); |
4928 | 36.8k | } else if (!IS_CHAR(cur)) { |
4929 | 3.91k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
4930 | 3.91k | "xmlParseComment: invalid xmlChar value %d\n", |
4931 | 3.91k | cur); |
4932 | 32.8k | } else { |
4933 | 32.8k | NEXT; |
4934 | 32.8k | if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && |
4935 | 32.8k | (!ctxt->disableSAX)) |
4936 | 23.9k | ctxt->sax->comment(ctxt->userData, buf); |
4937 | 32.8k | } |
4938 | 41.1k | xmlFree(buf); |
4939 | 41.1k | return; |
4940 | 10.0k | not_terminated: |
4941 | 10.0k | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
4942 | 10.0k | "Comment not terminated\n", NULL); |
4943 | 10.0k | xmlFree(buf); |
4944 | 10.0k | } |
4945 | | |
4946 | | /** |
4947 | | * Parse an XML (SGML) comment. Always consumes '<!'. |
4948 | | * |
4949 | | * @deprecated Internal function, don't use. |
4950 | | * |
4951 | | * The spec says that "For compatibility, the string "--" (double-hyphen) |
4952 | | * must not occur within comments. " |
4953 | | * |
4954 | | * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->' |
4955 | | * @param ctxt an XML parser context |
4956 | | */ |
4957 | | void |
4958 | 229k | xmlParseComment(xmlParserCtxt *ctxt) { |
4959 | 229k | xmlChar *buf = NULL; |
4960 | 229k | size_t size = XML_PARSER_BUFFER_SIZE; |
4961 | 229k | size_t len = 0; |
4962 | 229k | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
4963 | 0 | XML_MAX_HUGE_LENGTH : |
4964 | 229k | XML_MAX_TEXT_LENGTH; |
4965 | 229k | const xmlChar *in; |
4966 | 229k | size_t nbchar = 0; |
4967 | 229k | int ccol; |
4968 | | |
4969 | | /* |
4970 | | * Check that there is a comment right here. |
4971 | | */ |
4972 | 229k | if ((RAW != '<') || (NXT(1) != '!')) |
4973 | 0 | return; |
4974 | 229k | SKIP(2); |
4975 | 229k | if ((RAW != '-') || (NXT(1) != '-')) |
4976 | 907 | return; |
4977 | 228k | SKIP(2); |
4978 | 228k | GROW; |
4979 | | |
4980 | | /* |
4981 | | * Accelerated common case where input don't need to be |
4982 | | * modified before passing it to the handler. |
4983 | | */ |
4984 | 228k | in = ctxt->input->cur; |
4985 | 229k | do { |
4986 | 229k | if (*in == 0xA) { |
4987 | 23.1k | do { |
4988 | 23.1k | ctxt->input->line++; ctxt->input->col = 1; |
4989 | 23.1k | in++; |
4990 | 23.1k | } while (*in == 0xA); |
4991 | 22.9k | } |
4992 | 681k | get_more: |
4993 | 681k | ccol = ctxt->input->col; |
4994 | 12.5M | while (((*in > '-') && (*in <= 0x7F)) || |
4995 | 12.5M | ((*in >= 0x20) && (*in < '-')) || |
4996 | 12.5M | (*in == 0x09)) { |
4997 | 11.8M | in++; |
4998 | 11.8M | ccol++; |
4999 | 11.8M | } |
5000 | 681k | ctxt->input->col = ccol; |
5001 | 681k | if (*in == 0xA) { |
5002 | 115k | do { |
5003 | 115k | ctxt->input->line++; ctxt->input->col = 1; |
5004 | 115k | in++; |
5005 | 115k | } while (*in == 0xA); |
5006 | 96.2k | goto get_more; |
5007 | 96.2k | } |
5008 | 584k | nbchar = in - ctxt->input->cur; |
5009 | | /* |
5010 | | * save current set of data |
5011 | | */ |
5012 | 584k | if (nbchar > 0) { |
5013 | 562k | if (nbchar > maxLength - len) { |
5014 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, |
5015 | 0 | "Comment too big found", NULL); |
5016 | 0 | xmlFree(buf); |
5017 | 0 | return; |
5018 | 0 | } |
5019 | 562k | if (buf == NULL) { |
5020 | 222k | if ((*in == '-') && (in[1] == '-')) |
5021 | 101k | size = nbchar + 1; |
5022 | 121k | else |
5023 | 121k | size = XML_PARSER_BUFFER_SIZE + nbchar; |
5024 | 222k | buf = xmlMalloc(size); |
5025 | 222k | if (buf == NULL) { |
5026 | 7 | xmlErrMemory(ctxt); |
5027 | 7 | return; |
5028 | 7 | } |
5029 | 222k | len = 0; |
5030 | 340k | } else if (len + nbchar + 1 >= size) { |
5031 | 37.3k | xmlChar *new_buf; |
5032 | 37.3k | size += len + nbchar + XML_PARSER_BUFFER_SIZE; |
5033 | 37.3k | new_buf = xmlRealloc(buf, size); |
5034 | 37.3k | if (new_buf == NULL) { |
5035 | 2 | xmlErrMemory(ctxt); |
5036 | 2 | xmlFree(buf); |
5037 | 2 | return; |
5038 | 2 | } |
5039 | 37.3k | buf = new_buf; |
5040 | 37.3k | } |
5041 | 562k | memcpy(&buf[len], ctxt->input->cur, nbchar); |
5042 | 562k | len += nbchar; |
5043 | 562k | buf[len] = 0; |
5044 | 562k | } |
5045 | 584k | ctxt->input->cur = in; |
5046 | 584k | if (*in == 0xA) { |
5047 | 0 | in++; |
5048 | 0 | ctxt->input->line++; ctxt->input->col = 1; |
5049 | 0 | } |
5050 | 584k | if (*in == 0xD) { |
5051 | 145k | in++; |
5052 | 145k | if (*in == 0xA) { |
5053 | 135k | ctxt->input->cur = in; |
5054 | 135k | in++; |
5055 | 135k | ctxt->input->line++; ctxt->input->col = 1; |
5056 | 135k | goto get_more; |
5057 | 135k | } |
5058 | 9.75k | in--; |
5059 | 9.75k | } |
5060 | 449k | SHRINK; |
5061 | 449k | GROW; |
5062 | 449k | in = ctxt->input->cur; |
5063 | 449k | if (*in == '-') { |
5064 | 392k | if (in[1] == '-') { |
5065 | 261k | if (in[2] == '>') { |
5066 | 171k | SKIP(3); |
5067 | 171k | if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && |
5068 | 171k | (!ctxt->disableSAX)) { |
5069 | 75.7k | if (buf != NULL) |
5070 | 73.5k | ctxt->sax->comment(ctxt->userData, buf); |
5071 | 2.21k | else |
5072 | 2.21k | ctxt->sax->comment(ctxt->userData, BAD_CAST ""); |
5073 | 75.7k | } |
5074 | 171k | if (buf != NULL) |
5075 | 169k | xmlFree(buf); |
5076 | 171k | return; |
5077 | 171k | } |
5078 | 89.2k | if (buf != NULL) { |
5079 | 86.8k | xmlFatalErrMsgStr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, |
5080 | 86.8k | "Double hyphen within comment: " |
5081 | 86.8k | "<!--%.50s\n", |
5082 | 86.8k | buf); |
5083 | 86.8k | } else |
5084 | 2.32k | xmlFatalErrMsgStr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, |
5085 | 2.32k | "Double hyphen within comment\n", NULL); |
5086 | 89.2k | in++; |
5087 | 89.2k | ctxt->input->col++; |
5088 | 89.2k | } |
5089 | 220k | in++; |
5090 | 220k | ctxt->input->col++; |
5091 | 220k | goto get_more; |
5092 | 392k | } |
5093 | 449k | } while (((*in >= 0x20) && (*in <= 0x7F)) || (*in == 0x09) || (*in == 0x0a)); |
5094 | 56.7k | xmlParseCommentComplex(ctxt, buf, len, size); |
5095 | 56.7k | } |
5096 | | |
5097 | | |
5098 | | /** |
5099 | | * Parse the name of a PI |
5100 | | * |
5101 | | * @deprecated Internal function, don't use. |
5102 | | * |
5103 | | * [17] PITarget ::= Name - (('X' | 'x') ('M' | 'm') ('L' | 'l')) |
5104 | | * |
5105 | | * @param ctxt an XML parser context |
5106 | | * @returns the PITarget name or NULL |
5107 | | */ |
5108 | | |
5109 | | const xmlChar * |
5110 | 132k | xmlParsePITarget(xmlParserCtxt *ctxt) { |
5111 | 132k | const xmlChar *name; |
5112 | | |
5113 | 132k | name = xmlParseName(ctxt); |
5114 | 132k | if ((name != NULL) && |
5115 | 132k | ((name[0] == 'x') || (name[0] == 'X')) && |
5116 | 132k | ((name[1] == 'm') || (name[1] == 'M')) && |
5117 | 132k | ((name[2] == 'l') || (name[2] == 'L'))) { |
5118 | 13.0k | int i; |
5119 | 13.0k | if ((name[0] == 'x') && (name[1] == 'm') && |
5120 | 13.0k | (name[2] == 'l') && (name[3] == 0)) { |
5121 | 4.96k | xmlFatalErrMsg(ctxt, XML_ERR_RESERVED_XML_NAME, |
5122 | 4.96k | "XML declaration allowed only at the start of the document\n"); |
5123 | 4.96k | return(name); |
5124 | 8.10k | } else if (name[3] == 0) { |
5125 | 1.26k | xmlFatalErr(ctxt, XML_ERR_RESERVED_XML_NAME, NULL); |
5126 | 1.26k | return(name); |
5127 | 1.26k | } |
5128 | 20.0k | for (i = 0;;i++) { |
5129 | 20.0k | if (xmlW3CPIs[i] == NULL) break; |
5130 | 13.4k | if (xmlStrEqual(name, (const xmlChar *)xmlW3CPIs[i])) |
5131 | 304 | return(name); |
5132 | 13.4k | } |
5133 | 6.54k | xmlWarningMsg(ctxt, XML_ERR_RESERVED_XML_NAME, |
5134 | 6.54k | "xmlParsePITarget: invalid name prefix 'xml'\n", |
5135 | 6.54k | NULL, NULL); |
5136 | 6.54k | } |
5137 | 126k | if ((name != NULL) && (xmlStrchr(name, ':') != NULL)) { |
5138 | 584 | xmlNsErr(ctxt, XML_NS_ERR_COLON, |
5139 | 584 | "colons are forbidden from PI names '%s'\n", name, NULL, NULL); |
5140 | 584 | } |
5141 | 126k | return(name); |
5142 | 132k | } |
5143 | | |
5144 | | #ifdef LIBXML_CATALOG_ENABLED |
5145 | | /** |
5146 | | * Parse an XML Catalog Processing Instruction. |
5147 | | * |
5148 | | * <?oasis-xml-catalog catalog="http://example.com/catalog.xml"?> |
5149 | | * |
5150 | | * Occurs only if allowed by the user and if happening in the Misc |
5151 | | * part of the document before any doctype information |
5152 | | * This will add the given catalog to the parsing context in order |
5153 | | * to be used if there is a resolution need further down in the document |
5154 | | * |
5155 | | * @param ctxt an XML parser context |
5156 | | * @param catalog the PI value string |
5157 | | */ |
5158 | | |
5159 | | static void |
5160 | 0 | xmlParseCatalogPI(xmlParserCtxtPtr ctxt, const xmlChar *catalog) { |
5161 | 0 | xmlChar *URL = NULL; |
5162 | 0 | const xmlChar *tmp, *base; |
5163 | 0 | xmlChar marker; |
5164 | |
|
5165 | 0 | tmp = catalog; |
5166 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5167 | 0 | if (xmlStrncmp(tmp, BAD_CAST"catalog", 7)) |
5168 | 0 | goto error; |
5169 | 0 | tmp += 7; |
5170 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5171 | 0 | if (*tmp != '=') { |
5172 | 0 | return; |
5173 | 0 | } |
5174 | 0 | tmp++; |
5175 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5176 | 0 | marker = *tmp; |
5177 | 0 | if ((marker != '\'') && (marker != '"')) |
5178 | 0 | goto error; |
5179 | 0 | tmp++; |
5180 | 0 | base = tmp; |
5181 | 0 | while ((*tmp != 0) && (*tmp != marker)) tmp++; |
5182 | 0 | if (*tmp == 0) |
5183 | 0 | goto error; |
5184 | 0 | URL = xmlStrndup(base, tmp - base); |
5185 | 0 | tmp++; |
5186 | 0 | while (IS_BLANK_CH(*tmp)) tmp++; |
5187 | 0 | if (*tmp != 0) |
5188 | 0 | goto error; |
5189 | | |
5190 | 0 | if (URL != NULL) { |
5191 | | /* |
5192 | | * Unfortunately, the catalog API doesn't report OOM errors. |
5193 | | * xmlGetLastError isn't very helpful since we don't know |
5194 | | * where the last error came from. We'd have to reset it |
5195 | | * before this call and restore it afterwards. |
5196 | | */ |
5197 | 0 | ctxt->catalogs = xmlCatalogAddLocal(ctxt->catalogs, URL); |
5198 | 0 | xmlFree(URL); |
5199 | 0 | } |
5200 | 0 | return; |
5201 | | |
5202 | 0 | error: |
5203 | 0 | xmlWarningMsg(ctxt, XML_WAR_CATALOG_PI, |
5204 | 0 | "Catalog PI syntax error: %s\n", |
5205 | 0 | catalog, NULL); |
5206 | 0 | if (URL != NULL) |
5207 | 0 | xmlFree(URL); |
5208 | 0 | } |
5209 | | #endif |
5210 | | |
5211 | | /** |
5212 | | * Parse an XML Processing Instruction. |
5213 | | * |
5214 | | * @deprecated Internal function, don't use. |
5215 | | * |
5216 | | * [16] PI ::= '<?' PITarget (S (Char* - (Char* '?>' Char*)))? '?>' |
5217 | | * |
5218 | | * The processing is transferred to SAX once parsed. |
5219 | | * |
5220 | | * @param ctxt an XML parser context |
5221 | | */ |
5222 | | |
5223 | | void |
5224 | 132k | xmlParsePI(xmlParserCtxt *ctxt) { |
5225 | 132k | xmlChar *buf = NULL; |
5226 | 132k | size_t len = 0; |
5227 | 132k | size_t size = XML_PARSER_BUFFER_SIZE; |
5228 | 132k | size_t maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
5229 | 0 | XML_MAX_HUGE_LENGTH : |
5230 | 132k | XML_MAX_TEXT_LENGTH; |
5231 | 132k | int cur, l; |
5232 | 132k | const xmlChar *target; |
5233 | | |
5234 | 132k | if ((RAW == '<') && (NXT(1) == '?')) { |
5235 | | /* |
5236 | | * this is a Processing Instruction. |
5237 | | */ |
5238 | 132k | SKIP(2); |
5239 | | |
5240 | | /* |
5241 | | * Parse the target name and check for special support like |
5242 | | * namespace. |
5243 | | */ |
5244 | 132k | target = xmlParsePITarget(ctxt); |
5245 | 132k | if (target != NULL) { |
5246 | 128k | if ((RAW == '?') && (NXT(1) == '>')) { |
5247 | 48.8k | SKIP(2); |
5248 | | |
5249 | | /* |
5250 | | * SAX: PI detected. |
5251 | | */ |
5252 | 48.8k | if ((ctxt->sax) && (!ctxt->disableSAX) && |
5253 | 48.8k | (ctxt->sax->processingInstruction != NULL)) |
5254 | 43.3k | ctxt->sax->processingInstruction(ctxt->userData, |
5255 | 43.3k | target, NULL); |
5256 | 48.8k | return; |
5257 | 48.8k | } |
5258 | 79.5k | buf = xmlMalloc(size); |
5259 | 79.5k | if (buf == NULL) { |
5260 | 12 | xmlErrMemory(ctxt); |
5261 | 12 | return; |
5262 | 12 | } |
5263 | 79.5k | if (SKIP_BLANKS == 0) { |
5264 | 15.3k | xmlFatalErrMsgStr(ctxt, XML_ERR_SPACE_REQUIRED, |
5265 | 15.3k | "ParsePI: PI %s space expected\n", target); |
5266 | 15.3k | } |
5267 | 79.5k | cur = xmlCurrentCharRecover(ctxt, &l); |
5268 | 15.4M | while (IS_CHAR(cur) && /* checked */ |
5269 | 15.4M | ((cur != '?') || (NXT(1) != '>'))) { |
5270 | 15.3M | if (len + 5 >= size) { |
5271 | 11.5k | xmlChar *tmp; |
5272 | 11.5k | int newSize; |
5273 | | |
5274 | 11.5k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
5275 | 11.5k | if (newSize < 0) { |
5276 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_PI_NOT_FINISHED, |
5277 | 0 | "PI %s too big found", target); |
5278 | 0 | xmlFree(buf); |
5279 | 0 | return; |
5280 | 0 | } |
5281 | 11.5k | tmp = xmlRealloc(buf, newSize); |
5282 | 11.5k | if (tmp == NULL) { |
5283 | 2 | xmlErrMemory(ctxt); |
5284 | 2 | xmlFree(buf); |
5285 | 2 | return; |
5286 | 2 | } |
5287 | 11.5k | buf = tmp; |
5288 | 11.5k | size = newSize; |
5289 | 11.5k | } |
5290 | 15.3M | COPY_BUF(buf, len, cur); |
5291 | 15.3M | NEXTL(l); |
5292 | 15.3M | cur = xmlCurrentCharRecover(ctxt, &l); |
5293 | 15.3M | } |
5294 | 79.5k | buf[len] = 0; |
5295 | 79.5k | if (cur != '?') { |
5296 | 15.0k | xmlFatalErrMsgStr(ctxt, XML_ERR_PI_NOT_FINISHED, |
5297 | 15.0k | "ParsePI: PI %s never end ...\n", target); |
5298 | 64.4k | } else { |
5299 | 64.4k | SKIP(2); |
5300 | | |
5301 | 64.4k | #ifdef LIBXML_CATALOG_ENABLED |
5302 | 64.4k | if ((ctxt->inSubset == 0) && |
5303 | 64.4k | (xmlStrEqual(target, XML_CATALOG_PI))) { |
5304 | 314 | xmlCatalogAllow allow = xmlCatalogGetDefaults(); |
5305 | | |
5306 | 314 | if ((ctxt->options & XML_PARSE_CATALOG_PI) && |
5307 | 314 | ((allow == XML_CATA_ALLOW_DOCUMENT) || |
5308 | 0 | (allow == XML_CATA_ALLOW_ALL))) |
5309 | 0 | xmlParseCatalogPI(ctxt, buf); |
5310 | 314 | } |
5311 | 64.4k | #endif |
5312 | | |
5313 | | /* |
5314 | | * SAX: PI detected. |
5315 | | */ |
5316 | 64.4k | if ((ctxt->sax) && (!ctxt->disableSAX) && |
5317 | 64.4k | (ctxt->sax->processingInstruction != NULL)) |
5318 | 54.9k | ctxt->sax->processingInstruction(ctxt->userData, |
5319 | 54.9k | target, buf); |
5320 | 64.4k | } |
5321 | 79.5k | xmlFree(buf); |
5322 | 79.5k | } else { |
5323 | 4.41k | xmlFatalErr(ctxt, XML_ERR_PI_NOT_STARTED, NULL); |
5324 | 4.41k | } |
5325 | 132k | } |
5326 | 132k | } |
5327 | | |
5328 | | /** |
5329 | | * Parse a notation declaration. Always consumes '<!'. |
5330 | | * |
5331 | | * @deprecated Internal function, don't use. |
5332 | | * |
5333 | | * [82] NotationDecl ::= '<!NOTATION' S Name S (ExternalID | PublicID) |
5334 | | * S? '>' |
5335 | | * |
5336 | | * Hence there is actually 3 choices: |
5337 | | * |
5338 | | * 'PUBLIC' S PubidLiteral |
5339 | | * 'PUBLIC' S PubidLiteral S SystemLiteral |
5340 | | * 'SYSTEM' S SystemLiteral |
5341 | | * |
5342 | | * See the NOTE on #xmlParseExternalID. |
5343 | | * |
5344 | | * @param ctxt an XML parser context |
5345 | | */ |
5346 | | |
5347 | | void |
5348 | 13.3k | xmlParseNotationDecl(xmlParserCtxt *ctxt) { |
5349 | 13.3k | const xmlChar *name; |
5350 | 13.3k | xmlChar *Pubid; |
5351 | 13.3k | xmlChar *Systemid; |
5352 | | |
5353 | 13.3k | if ((CUR != '<') || (NXT(1) != '!')) |
5354 | 0 | return; |
5355 | 13.3k | SKIP(2); |
5356 | | |
5357 | 13.3k | if (CMP8(CUR_PTR, 'N', 'O', 'T', 'A', 'T', 'I', 'O', 'N')) { |
5358 | | #ifdef LIBXML_VALID_ENABLED |
5359 | | int oldInputNr = ctxt->inputNr; |
5360 | | #endif |
5361 | | |
5362 | 10.9k | SKIP(8); |
5363 | 10.9k | if (SKIP_BLANKS_PE == 0) { |
5364 | 1.06k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5365 | 1.06k | "Space required after '<!NOTATION'\n"); |
5366 | 1.06k | return; |
5367 | 1.06k | } |
5368 | | |
5369 | 9.92k | name = xmlParseName(ctxt); |
5370 | 9.92k | if (name == NULL) { |
5371 | 371 | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_STARTED, NULL); |
5372 | 371 | return; |
5373 | 371 | } |
5374 | 9.55k | if (xmlStrchr(name, ':') != NULL) { |
5375 | 584 | xmlNsErr(ctxt, XML_NS_ERR_COLON, |
5376 | 584 | "colons are forbidden from notation names '%s'\n", |
5377 | 584 | name, NULL, NULL); |
5378 | 584 | } |
5379 | 9.55k | if (SKIP_BLANKS_PE == 0) { |
5380 | 227 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5381 | 227 | "Space required after the NOTATION name'\n"); |
5382 | 227 | return; |
5383 | 227 | } |
5384 | | |
5385 | | /* |
5386 | | * Parse the IDs. |
5387 | | */ |
5388 | 9.33k | Systemid = xmlParseExternalID(ctxt, &Pubid, 0); |
5389 | 9.33k | SKIP_BLANKS_PE; |
5390 | | |
5391 | 9.33k | if (RAW == '>') { |
5392 | | #ifdef LIBXML_VALID_ENABLED |
5393 | | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
5394 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
5395 | | "Notation declaration doesn't start and stop" |
5396 | | " in the same entity\n", |
5397 | | NULL, NULL); |
5398 | | } |
5399 | | #endif |
5400 | 6.52k | NEXT; |
5401 | 6.52k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
5402 | 6.52k | (ctxt->sax->notationDecl != NULL)) |
5403 | 6.25k | ctxt->sax->notationDecl(ctxt->userData, name, Pubid, Systemid); |
5404 | 6.52k | } else { |
5405 | 2.80k | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_FINISHED, NULL); |
5406 | 2.80k | } |
5407 | 9.33k | if (Systemid != NULL) xmlFree(Systemid); |
5408 | 9.33k | if (Pubid != NULL) xmlFree(Pubid); |
5409 | 9.33k | } |
5410 | 13.3k | } |
5411 | | |
5412 | | /** |
5413 | | * Parse an entity declaration. Always consumes '<!'. |
5414 | | * |
5415 | | * @deprecated Internal function, don't use. |
5416 | | * |
5417 | | * [70] EntityDecl ::= GEDecl | PEDecl |
5418 | | * |
5419 | | * [71] GEDecl ::= '<!ENTITY' S Name S EntityDef S? '>' |
5420 | | * |
5421 | | * [72] PEDecl ::= '<!ENTITY' S '%' S Name S PEDef S? '>' |
5422 | | * |
5423 | | * [73] EntityDef ::= EntityValue | (ExternalID NDataDecl?) |
5424 | | * |
5425 | | * [74] PEDef ::= EntityValue | ExternalID |
5426 | | * |
5427 | | * [76] NDataDecl ::= S 'NDATA' S Name |
5428 | | * |
5429 | | * [ VC: Notation Declared ] |
5430 | | * The Name must match the declared name of a notation. |
5431 | | * |
5432 | | * @param ctxt an XML parser context |
5433 | | */ |
5434 | | |
5435 | | void |
5436 | 184k | xmlParseEntityDecl(xmlParserCtxt *ctxt) { |
5437 | 184k | const xmlChar *name = NULL; |
5438 | 184k | xmlChar *value = NULL; |
5439 | 184k | xmlChar *URI = NULL, *literal = NULL; |
5440 | 184k | const xmlChar *ndata = NULL; |
5441 | 184k | int isParameter = 0; |
5442 | 184k | xmlChar *orig = NULL; |
5443 | | |
5444 | 184k | if ((CUR != '<') || (NXT(1) != '!')) |
5445 | 0 | return; |
5446 | 184k | SKIP(2); |
5447 | | |
5448 | | /* GROW; done in the caller */ |
5449 | 184k | if (CMP6(CUR_PTR, 'E', 'N', 'T', 'I', 'T', 'Y')) { |
5450 | | #ifdef LIBXML_VALID_ENABLED |
5451 | | int oldInputNr = ctxt->inputNr; |
5452 | | #endif |
5453 | | |
5454 | 181k | SKIP(6); |
5455 | 181k | if (SKIP_BLANKS_PE == 0) { |
5456 | 5.79k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5457 | 5.79k | "Space required after '<!ENTITY'\n"); |
5458 | 5.79k | } |
5459 | | |
5460 | 181k | if (RAW == '%') { |
5461 | 106k | NEXT; |
5462 | 106k | if (SKIP_BLANKS_PE == 0) { |
5463 | 7.74k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5464 | 7.74k | "Space required after '%%'\n"); |
5465 | 7.74k | } |
5466 | 106k | isParameter = 1; |
5467 | 106k | } |
5468 | | |
5469 | 181k | name = xmlParseName(ctxt); |
5470 | 181k | if (name == NULL) { |
5471 | 5.74k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
5472 | 5.74k | "xmlParseEntityDecl: no name\n"); |
5473 | 5.74k | return; |
5474 | 5.74k | } |
5475 | 176k | if (xmlStrchr(name, ':') != NULL) { |
5476 | 510 | xmlNsErr(ctxt, XML_NS_ERR_COLON, |
5477 | 510 | "colons are forbidden from entities names '%s'\n", |
5478 | 510 | name, NULL, NULL); |
5479 | 510 | } |
5480 | 176k | if (SKIP_BLANKS_PE == 0) { |
5481 | 2.86k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5482 | 2.86k | "Space required after the entity name\n"); |
5483 | 2.86k | } |
5484 | | |
5485 | | /* |
5486 | | * handle the various case of definitions... |
5487 | | */ |
5488 | 176k | if (isParameter) { |
5489 | 102k | if ((RAW == '"') || (RAW == '\'')) { |
5490 | 64.3k | value = xmlParseEntityValue(ctxt, &orig); |
5491 | 64.3k | if (value) { |
5492 | 63.4k | if ((ctxt->sax != NULL) && |
5493 | 63.4k | (!ctxt->disableSAX) && (ctxt->sax->entityDecl != NULL)) |
5494 | 43.7k | ctxt->sax->entityDecl(ctxt->userData, name, |
5495 | 43.7k | XML_INTERNAL_PARAMETER_ENTITY, |
5496 | 43.7k | NULL, NULL, value); |
5497 | 63.4k | } |
5498 | 64.3k | } else { |
5499 | 38.6k | URI = xmlParseExternalID(ctxt, &literal, 1); |
5500 | 38.6k | if ((URI == NULL) && (literal == NULL)) { |
5501 | 1.75k | xmlFatalErr(ctxt, XML_ERR_VALUE_REQUIRED, NULL); |
5502 | 1.75k | } |
5503 | 38.6k | if (URI) { |
5504 | 36.5k | if (xmlStrchr(URI, '#')) { |
5505 | 381 | xmlFatalErr(ctxt, XML_ERR_URI_FRAGMENT, NULL); |
5506 | 36.1k | } else { |
5507 | 36.1k | if ((ctxt->sax != NULL) && |
5508 | 36.1k | (!ctxt->disableSAX) && |
5509 | 36.1k | (ctxt->sax->entityDecl != NULL)) |
5510 | 34.0k | ctxt->sax->entityDecl(ctxt->userData, name, |
5511 | 34.0k | XML_EXTERNAL_PARAMETER_ENTITY, |
5512 | 34.0k | literal, URI, NULL); |
5513 | 36.1k | } |
5514 | 36.5k | } |
5515 | 38.6k | } |
5516 | 102k | } else { |
5517 | 73.3k | if ((RAW == '"') || (RAW == '\'')) { |
5518 | 51.9k | value = xmlParseEntityValue(ctxt, &orig); |
5519 | 51.9k | if ((ctxt->sax != NULL) && |
5520 | 51.9k | (!ctxt->disableSAX) && (ctxt->sax->entityDecl != NULL)) |
5521 | 26.0k | ctxt->sax->entityDecl(ctxt->userData, name, |
5522 | 26.0k | XML_INTERNAL_GENERAL_ENTITY, |
5523 | 26.0k | NULL, NULL, value); |
5524 | | /* |
5525 | | * For expat compatibility in SAX mode. |
5526 | | */ |
5527 | 51.9k | if ((ctxt->myDoc == NULL) || |
5528 | 51.9k | (xmlStrEqual(ctxt->myDoc->version, SAX_COMPAT_MODE))) { |
5529 | 20.2k | if (ctxt->myDoc == NULL) { |
5530 | 15.1k | ctxt->myDoc = xmlNewDoc(SAX_COMPAT_MODE); |
5531 | 15.1k | if (ctxt->myDoc == NULL) { |
5532 | 22 | xmlErrMemory(ctxt); |
5533 | 22 | goto done; |
5534 | 22 | } |
5535 | 15.0k | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
5536 | 15.0k | } |
5537 | 20.2k | if (ctxt->myDoc->intSubset == NULL) { |
5538 | 15.0k | ctxt->myDoc->intSubset = xmlNewDtd(ctxt->myDoc, |
5539 | 15.0k | BAD_CAST "fake", NULL, NULL); |
5540 | 15.0k | if (ctxt->myDoc->intSubset == NULL) { |
5541 | 5 | xmlErrMemory(ctxt); |
5542 | 5 | goto done; |
5543 | 5 | } |
5544 | 15.0k | } |
5545 | | |
5546 | 20.2k | xmlSAX2EntityDecl(ctxt, name, XML_INTERNAL_GENERAL_ENTITY, |
5547 | 20.2k | NULL, NULL, value); |
5548 | 20.2k | } |
5549 | 51.9k | } else { |
5550 | 21.3k | URI = xmlParseExternalID(ctxt, &literal, 1); |
5551 | 21.3k | if ((URI == NULL) && (literal == NULL)) { |
5552 | 4.87k | xmlFatalErr(ctxt, XML_ERR_VALUE_REQUIRED, NULL); |
5553 | 4.87k | } |
5554 | 21.3k | if (URI) { |
5555 | 15.7k | if (xmlStrchr(URI, '#')) { |
5556 | 314 | xmlFatalErr(ctxt, XML_ERR_URI_FRAGMENT, NULL); |
5557 | 314 | } |
5558 | 15.7k | } |
5559 | 21.3k | if ((RAW != '>') && (SKIP_BLANKS_PE == 0)) { |
5560 | 5.97k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5561 | 5.97k | "Space required before 'NDATA'\n"); |
5562 | 5.97k | } |
5563 | 21.3k | if (CMP5(CUR_PTR, 'N', 'D', 'A', 'T', 'A')) { |
5564 | 1.28k | SKIP(5); |
5565 | 1.28k | if (SKIP_BLANKS_PE == 0) { |
5566 | 767 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5567 | 767 | "Space required after 'NDATA'\n"); |
5568 | 767 | } |
5569 | 1.28k | ndata = xmlParseName(ctxt); |
5570 | 1.28k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
5571 | 1.28k | (ctxt->sax->unparsedEntityDecl != NULL)) |
5572 | 223 | ctxt->sax->unparsedEntityDecl(ctxt->userData, name, |
5573 | 223 | literal, URI, ndata); |
5574 | 20.0k | } else { |
5575 | 20.0k | if ((ctxt->sax != NULL) && |
5576 | 20.0k | (!ctxt->disableSAX) && (ctxt->sax->entityDecl != NULL)) |
5577 | 12.8k | ctxt->sax->entityDecl(ctxt->userData, name, |
5578 | 12.8k | XML_EXTERNAL_GENERAL_PARSED_ENTITY, |
5579 | 12.8k | literal, URI, NULL); |
5580 | | /* |
5581 | | * For expat compatibility in SAX mode. |
5582 | | * assuming the entity replacement was asked for |
5583 | | */ |
5584 | 20.0k | if ((ctxt->replaceEntities != 0) && |
5585 | 20.0k | ((ctxt->myDoc == NULL) || |
5586 | 20.0k | (xmlStrEqual(ctxt->myDoc->version, SAX_COMPAT_MODE)))) { |
5587 | 2.38k | if (ctxt->myDoc == NULL) { |
5588 | 903 | ctxt->myDoc = xmlNewDoc(SAX_COMPAT_MODE); |
5589 | 903 | if (ctxt->myDoc == NULL) { |
5590 | 1 | xmlErrMemory(ctxt); |
5591 | 1 | goto done; |
5592 | 1 | } |
5593 | 902 | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
5594 | 902 | } |
5595 | | |
5596 | 2.38k | if (ctxt->myDoc->intSubset == NULL) { |
5597 | 902 | ctxt->myDoc->intSubset = xmlNewDtd(ctxt->myDoc, |
5598 | 902 | BAD_CAST "fake", NULL, NULL); |
5599 | 902 | if (ctxt->myDoc->intSubset == NULL) { |
5600 | 1 | xmlErrMemory(ctxt); |
5601 | 1 | goto done; |
5602 | 1 | } |
5603 | 902 | } |
5604 | 2.38k | xmlSAX2EntityDecl(ctxt, name, |
5605 | 2.38k | XML_EXTERNAL_GENERAL_PARSED_ENTITY, |
5606 | 2.38k | literal, URI, NULL); |
5607 | 2.38k | } |
5608 | 20.0k | } |
5609 | 21.3k | } |
5610 | 73.3k | } |
5611 | 176k | SKIP_BLANKS_PE; |
5612 | 176k | if (RAW != '>') { |
5613 | 21.7k | xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_NOT_FINISHED, |
5614 | 21.7k | "xmlParseEntityDecl: entity %s not terminated\n", name); |
5615 | 21.7k | xmlHaltParser(ctxt); |
5616 | 154k | } else { |
5617 | | #ifdef LIBXML_VALID_ENABLED |
5618 | | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
5619 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
5620 | | "Entity declaration doesn't start and stop in" |
5621 | | " the same entity\n", |
5622 | | NULL, NULL); |
5623 | | } |
5624 | | #endif |
5625 | 154k | NEXT; |
5626 | 154k | } |
5627 | 176k | if (orig != NULL) { |
5628 | | /* |
5629 | | * Ugly mechanism to save the raw entity value. |
5630 | | */ |
5631 | 114k | xmlEntityPtr cur = NULL; |
5632 | | |
5633 | 114k | if (isParameter) { |
5634 | 63.4k | if ((ctxt->sax != NULL) && |
5635 | 63.4k | (ctxt->sax->getParameterEntity != NULL)) |
5636 | 63.4k | cur = ctxt->sax->getParameterEntity(ctxt->userData, name); |
5637 | 63.4k | } else { |
5638 | 50.6k | if ((ctxt->sax != NULL) && |
5639 | 50.6k | (ctxt->sax->getEntity != NULL)) |
5640 | 50.6k | cur = ctxt->sax->getEntity(ctxt->userData, name); |
5641 | 50.6k | if ((cur == NULL) && (ctxt->userData==ctxt)) { |
5642 | 5.14k | cur = xmlSAX2GetEntity(ctxt, name); |
5643 | 5.14k | } |
5644 | 50.6k | } |
5645 | 114k | if ((cur != NULL) && (cur->orig == NULL)) { |
5646 | 73.9k | cur->orig = orig; |
5647 | 73.9k | orig = NULL; |
5648 | 73.9k | } |
5649 | 114k | } |
5650 | | |
5651 | 176k | done: |
5652 | 176k | if (value != NULL) xmlFree(value); |
5653 | 176k | if (URI != NULL) xmlFree(URI); |
5654 | 176k | if (literal != NULL) xmlFree(literal); |
5655 | 176k | if (orig != NULL) xmlFree(orig); |
5656 | 176k | } |
5657 | 184k | } |
5658 | | |
5659 | | /** |
5660 | | * Parse an attribute default declaration |
5661 | | * |
5662 | | * @deprecated Internal function, don't use. |
5663 | | * |
5664 | | * [60] DefaultDecl ::= '#REQUIRED' | '#IMPLIED' | (('#FIXED' S)? AttValue) |
5665 | | * |
5666 | | * [ VC: Required Attribute ] |
5667 | | * if the default declaration is the keyword \#REQUIRED, then the |
5668 | | * attribute must be specified for all elements of the type in the |
5669 | | * attribute-list declaration. |
5670 | | * |
5671 | | * [ VC: Attribute Default Legal ] |
5672 | | * The declared default value must meet the lexical constraints of |
5673 | | * the declared attribute type c.f. #xmlValidateAttributeDecl |
5674 | | * |
5675 | | * [ VC: Fixed Attribute Default ] |
5676 | | * if an attribute has a default value declared with the \#FIXED |
5677 | | * keyword, instances of that attribute must match the default value. |
5678 | | * |
5679 | | * [ WFC: No < in Attribute Values ] |
5680 | | * handled in #xmlParseAttValue |
5681 | | * |
5682 | | * @param ctxt an XML parser context |
5683 | | * @param value Receive a possible fixed default value for the attribute |
5684 | | * @returns XML_ATTRIBUTE_NONE, XML_ATTRIBUTE_REQUIRED, XML_ATTRIBUTE_IMPLIED |
5685 | | * or XML_ATTRIBUTE_FIXED. |
5686 | | */ |
5687 | | |
5688 | | int |
5689 | 91.5k | xmlParseDefaultDecl(xmlParserCtxt *ctxt, xmlChar **value) { |
5690 | 91.5k | int val; |
5691 | 91.5k | xmlChar *ret; |
5692 | | |
5693 | 91.5k | *value = NULL; |
5694 | 91.5k | if (CMP9(CUR_PTR, '#', 'R', 'E', 'Q', 'U', 'I', 'R', 'E', 'D')) { |
5695 | 1.19k | SKIP(9); |
5696 | 1.19k | return(XML_ATTRIBUTE_REQUIRED); |
5697 | 1.19k | } |
5698 | 90.3k | if (CMP8(CUR_PTR, '#', 'I', 'M', 'P', 'L', 'I', 'E', 'D')) { |
5699 | 30.2k | SKIP(8); |
5700 | 30.2k | return(XML_ATTRIBUTE_IMPLIED); |
5701 | 30.2k | } |
5702 | 60.1k | val = XML_ATTRIBUTE_NONE; |
5703 | 60.1k | if (CMP6(CUR_PTR, '#', 'F', 'I', 'X', 'E', 'D')) { |
5704 | 8.29k | SKIP(6); |
5705 | 8.29k | val = XML_ATTRIBUTE_FIXED; |
5706 | 8.29k | if (SKIP_BLANKS_PE == 0) { |
5707 | 251 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5708 | 251 | "Space required after '#FIXED'\n"); |
5709 | 251 | } |
5710 | 8.29k | } |
5711 | 60.1k | ret = xmlParseAttValue(ctxt); |
5712 | 60.1k | if (ret == NULL) { |
5713 | 7.26k | xmlFatalErrMsg(ctxt, (xmlParserErrors)ctxt->errNo, |
5714 | 7.26k | "Attribute default value declaration error\n"); |
5715 | 7.26k | } else |
5716 | 52.8k | *value = ret; |
5717 | 60.1k | return(val); |
5718 | 90.3k | } |
5719 | | |
5720 | | /** |
5721 | | * Parse an Notation attribute type. |
5722 | | * |
5723 | | * @deprecated Internal function, don't use. |
5724 | | * |
5725 | | * Note: the leading 'NOTATION' S part has already being parsed... |
5726 | | * |
5727 | | * [58] NotationType ::= 'NOTATION' S '(' S? Name (S? '|' S? Name)* S? ')' |
5728 | | * |
5729 | | * [ VC: Notation Attributes ] |
5730 | | * Values of this type must match one of the notation names included |
5731 | | * in the declaration; all notation names in the declaration must be declared. |
5732 | | * |
5733 | | * @param ctxt an XML parser context |
5734 | | * @returns the notation attribute tree built while parsing |
5735 | | */ |
5736 | | |
5737 | | xmlEnumeration * |
5738 | 928 | xmlParseNotationType(xmlParserCtxt *ctxt) { |
5739 | 928 | const xmlChar *name; |
5740 | 928 | xmlEnumerationPtr ret = NULL, last = NULL, cur, tmp; |
5741 | | |
5742 | 928 | if (RAW != '(') { |
5743 | 212 | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_STARTED, NULL); |
5744 | 212 | return(NULL); |
5745 | 212 | } |
5746 | 1.08k | do { |
5747 | 1.08k | NEXT; |
5748 | 1.08k | SKIP_BLANKS_PE; |
5749 | 1.08k | name = xmlParseName(ctxt); |
5750 | 1.08k | if (name == NULL) { |
5751 | 229 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
5752 | 229 | "Name expected in NOTATION declaration\n"); |
5753 | 229 | xmlFreeEnumeration(ret); |
5754 | 229 | return(NULL); |
5755 | 229 | } |
5756 | 851 | tmp = NULL; |
5757 | | #ifdef LIBXML_VALID_ENABLED |
5758 | | if (ctxt->validate) { |
5759 | | tmp = ret; |
5760 | | while (tmp != NULL) { |
5761 | | if (xmlStrEqual(name, tmp->name)) { |
5762 | | xmlValidityError(ctxt, XML_DTD_DUP_TOKEN, |
5763 | | "standalone: attribute notation value token %s duplicated\n", |
5764 | | name, NULL); |
5765 | | if (!xmlDictOwns(ctxt->dict, name)) |
5766 | | xmlFree((xmlChar *) name); |
5767 | | break; |
5768 | | } |
5769 | | tmp = tmp->next; |
5770 | | } |
5771 | | } |
5772 | | #endif /* LIBXML_VALID_ENABLED */ |
5773 | 851 | if (tmp == NULL) { |
5774 | 851 | cur = xmlCreateEnumeration(name); |
5775 | 851 | if (cur == NULL) { |
5776 | 1 | xmlErrMemory(ctxt); |
5777 | 1 | xmlFreeEnumeration(ret); |
5778 | 1 | return(NULL); |
5779 | 1 | } |
5780 | 850 | if (last == NULL) ret = last = cur; |
5781 | 358 | else { |
5782 | 358 | last->next = cur; |
5783 | 358 | last = cur; |
5784 | 358 | } |
5785 | 850 | } |
5786 | 850 | SKIP_BLANKS_PE; |
5787 | 850 | } while (RAW == '|'); |
5788 | 486 | if (RAW != ')') { |
5789 | 253 | xmlFatalErr(ctxt, XML_ERR_NOTATION_NOT_FINISHED, NULL); |
5790 | 253 | xmlFreeEnumeration(ret); |
5791 | 253 | return(NULL); |
5792 | 253 | } |
5793 | 233 | NEXT; |
5794 | 233 | return(ret); |
5795 | 486 | } |
5796 | | |
5797 | | /** |
5798 | | * Parse an Enumeration attribute type. |
5799 | | * |
5800 | | * @deprecated Internal function, don't use. |
5801 | | * |
5802 | | * [59] Enumeration ::= '(' S? Nmtoken (S? '|' S? Nmtoken)* S? ')' |
5803 | | * |
5804 | | * [ VC: Enumeration ] |
5805 | | * Values of this type must match one of the Nmtoken tokens in |
5806 | | * the declaration |
5807 | | * |
5808 | | * @param ctxt an XML parser context |
5809 | | * @returns the enumeration attribute tree built while parsing |
5810 | | */ |
5811 | | |
5812 | | xmlEnumeration * |
5813 | 13.8k | xmlParseEnumerationType(xmlParserCtxt *ctxt) { |
5814 | 13.8k | xmlChar *name; |
5815 | 13.8k | xmlEnumerationPtr ret = NULL, last = NULL, cur, tmp; |
5816 | | |
5817 | 13.8k | if (RAW != '(') { |
5818 | 8.19k | xmlFatalErr(ctxt, XML_ERR_ATTLIST_NOT_STARTED, NULL); |
5819 | 8.19k | return(NULL); |
5820 | 8.19k | } |
5821 | 20.9k | do { |
5822 | 20.9k | NEXT; |
5823 | 20.9k | SKIP_BLANKS_PE; |
5824 | 20.9k | name = xmlParseNmtoken(ctxt); |
5825 | 20.9k | if (name == NULL) { |
5826 | 307 | xmlFatalErr(ctxt, XML_ERR_NMTOKEN_REQUIRED, NULL); |
5827 | 307 | return(ret); |
5828 | 307 | } |
5829 | 20.6k | tmp = NULL; |
5830 | | #ifdef LIBXML_VALID_ENABLED |
5831 | | if (ctxt->validate) { |
5832 | | tmp = ret; |
5833 | | while (tmp != NULL) { |
5834 | | if (xmlStrEqual(name, tmp->name)) { |
5835 | | xmlValidityError(ctxt, XML_DTD_DUP_TOKEN, |
5836 | | "standalone: attribute enumeration value token %s duplicated\n", |
5837 | | name, NULL); |
5838 | | if (!xmlDictOwns(ctxt->dict, name)) |
5839 | | xmlFree(name); |
5840 | | break; |
5841 | | } |
5842 | | tmp = tmp->next; |
5843 | | } |
5844 | | } |
5845 | | #endif /* LIBXML_VALID_ENABLED */ |
5846 | 20.6k | if (tmp == NULL) { |
5847 | 20.6k | cur = xmlCreateEnumeration(name); |
5848 | 20.6k | if (!xmlDictOwns(ctxt->dict, name)) |
5849 | 20.6k | xmlFree(name); |
5850 | 20.6k | if (cur == NULL) { |
5851 | 5 | xmlErrMemory(ctxt); |
5852 | 5 | xmlFreeEnumeration(ret); |
5853 | 5 | return(NULL); |
5854 | 5 | } |
5855 | 20.6k | if (last == NULL) ret = last = cur; |
5856 | 15.2k | else { |
5857 | 15.2k | last->next = cur; |
5858 | 15.2k | last = cur; |
5859 | 15.2k | } |
5860 | 20.6k | } |
5861 | 20.6k | SKIP_BLANKS_PE; |
5862 | 20.6k | } while (RAW == '|'); |
5863 | 5.33k | if (RAW != ')') { |
5864 | 698 | xmlFatalErr(ctxt, XML_ERR_ATTLIST_NOT_FINISHED, NULL); |
5865 | 698 | return(ret); |
5866 | 698 | } |
5867 | 4.63k | NEXT; |
5868 | 4.63k | return(ret); |
5869 | 5.33k | } |
5870 | | |
5871 | | /** |
5872 | | * Parse an Enumerated attribute type. |
5873 | | * |
5874 | | * @deprecated Internal function, don't use. |
5875 | | * |
5876 | | * [57] EnumeratedType ::= NotationType | Enumeration |
5877 | | * |
5878 | | * [58] NotationType ::= 'NOTATION' S '(' S? Name (S? '|' S? Name)* S? ')' |
5879 | | * |
5880 | | * @param ctxt an XML parser context |
5881 | | * @param tree the enumeration tree built while parsing |
5882 | | * @returns XML_ATTRIBUTE_ENUMERATION or XML_ATTRIBUTE_NOTATION |
5883 | | */ |
5884 | | |
5885 | | int |
5886 | 15.0k | xmlParseEnumeratedType(xmlParserCtxt *ctxt, xmlEnumeration **tree) { |
5887 | 15.0k | if (CMP8(CUR_PTR, 'N', 'O', 'T', 'A', 'T', 'I', 'O', 'N')) { |
5888 | 1.16k | SKIP(8); |
5889 | 1.16k | if (SKIP_BLANKS_PE == 0) { |
5890 | 235 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
5891 | 235 | "Space required after 'NOTATION'\n"); |
5892 | 235 | return(0); |
5893 | 235 | } |
5894 | 928 | *tree = xmlParseNotationType(ctxt); |
5895 | 928 | if (*tree == NULL) return(0); |
5896 | 233 | return(XML_ATTRIBUTE_NOTATION); |
5897 | 928 | } |
5898 | 13.8k | *tree = xmlParseEnumerationType(ctxt); |
5899 | 13.8k | if (*tree == NULL) return(0); |
5900 | 5.41k | return(XML_ATTRIBUTE_ENUMERATION); |
5901 | 13.8k | } |
5902 | | |
5903 | | /** |
5904 | | * Parse the Attribute list def for an element |
5905 | | * |
5906 | | * @deprecated Internal function, don't use. |
5907 | | * |
5908 | | * [54] AttType ::= StringType | TokenizedType | EnumeratedType |
5909 | | * |
5910 | | * [55] StringType ::= 'CDATA' |
5911 | | * |
5912 | | * [56] TokenizedType ::= 'ID' | 'IDREF' | 'IDREFS' | 'ENTITY' | |
5913 | | * 'ENTITIES' | 'NMTOKEN' | 'NMTOKENS' |
5914 | | * |
5915 | | * Validity constraints for attribute values syntax are checked in |
5916 | | * #xmlValidateAttributeValue |
5917 | | * |
5918 | | * [ VC: ID ] |
5919 | | * Values of type ID must match the Name production. A name must not |
5920 | | * appear more than once in an XML document as a value of this type; |
5921 | | * i.e., ID values must uniquely identify the elements which bear them. |
5922 | | * |
5923 | | * [ VC: One ID per Element Type ] |
5924 | | * No element type may have more than one ID attribute specified. |
5925 | | * |
5926 | | * [ VC: ID Attribute Default ] |
5927 | | * An ID attribute must have a declared default of \#IMPLIED or \#REQUIRED. |
5928 | | * |
5929 | | * [ VC: IDREF ] |
5930 | | * Values of type IDREF must match the Name production, and values |
5931 | | * of type IDREFS must match Names; each IDREF Name must match the value |
5932 | | * of an ID attribute on some element in the XML document; i.e. IDREF |
5933 | | * values must match the value of some ID attribute. |
5934 | | * |
5935 | | * [ VC: Entity Name ] |
5936 | | * Values of type ENTITY must match the Name production, values |
5937 | | * of type ENTITIES must match Names; each Entity Name must match the |
5938 | | * name of an unparsed entity declared in the DTD. |
5939 | | * |
5940 | | * [ VC: Name Token ] |
5941 | | * Values of type NMTOKEN must match the Nmtoken production; values |
5942 | | * of type NMTOKENS must match Nmtokens. |
5943 | | * |
5944 | | * @param ctxt an XML parser context |
5945 | | * @param tree the enumeration tree built while parsing |
5946 | | * @returns the attribute type |
5947 | | */ |
5948 | | int |
5949 | 106k | xmlParseAttributeType(xmlParserCtxt *ctxt, xmlEnumeration **tree) { |
5950 | 106k | if (CMP5(CUR_PTR, 'C', 'D', 'A', 'T', 'A')) { |
5951 | 27.7k | SKIP(5); |
5952 | 27.7k | return(XML_ATTRIBUTE_CDATA); |
5953 | 78.8k | } else if (CMP6(CUR_PTR, 'I', 'D', 'R', 'E', 'F', 'S')) { |
5954 | 598 | SKIP(6); |
5955 | 598 | return(XML_ATTRIBUTE_IDREFS); |
5956 | 78.2k | } else if (CMP5(CUR_PTR, 'I', 'D', 'R', 'E', 'F')) { |
5957 | 1.75k | SKIP(5); |
5958 | 1.75k | return(XML_ATTRIBUTE_IDREF); |
5959 | 76.4k | } else if ((RAW == 'I') && (NXT(1) == 'D')) { |
5960 | 48.0k | SKIP(2); |
5961 | 48.0k | return(XML_ATTRIBUTE_ID); |
5962 | 48.0k | } else if (CMP6(CUR_PTR, 'E', 'N', 'T', 'I', 'T', 'Y')) { |
5963 | 450 | SKIP(6); |
5964 | 450 | return(XML_ATTRIBUTE_ENTITY); |
5965 | 27.9k | } else if (CMP8(CUR_PTR, 'E', 'N', 'T', 'I', 'T', 'I', 'E', 'S')) { |
5966 | 2.06k | SKIP(8); |
5967 | 2.06k | return(XML_ATTRIBUTE_ENTITIES); |
5968 | 25.8k | } else if (CMP8(CUR_PTR, 'N', 'M', 'T', 'O', 'K', 'E', 'N', 'S')) { |
5969 | 7.46k | SKIP(8); |
5970 | 7.46k | return(XML_ATTRIBUTE_NMTOKENS); |
5971 | 18.3k | } else if (CMP7(CUR_PTR, 'N', 'M', 'T', 'O', 'K', 'E', 'N')) { |
5972 | 3.39k | SKIP(7); |
5973 | 3.39k | return(XML_ATTRIBUTE_NMTOKEN); |
5974 | 3.39k | } |
5975 | 15.0k | return(xmlParseEnumeratedType(ctxt, tree)); |
5976 | 106k | } |
5977 | | |
5978 | | /** |
5979 | | * Parse an attribute list declaration for an element. Always consumes '<!'. |
5980 | | * |
5981 | | * @deprecated Internal function, don't use. |
5982 | | * |
5983 | | * [52] AttlistDecl ::= '<!ATTLIST' S Name AttDef* S? '>' |
5984 | | * |
5985 | | * [53] AttDef ::= S Name S AttType S DefaultDecl |
5986 | | * @param ctxt an XML parser context |
5987 | | */ |
5988 | | void |
5989 | 61.0k | xmlParseAttributeListDecl(xmlParserCtxt *ctxt) { |
5990 | 61.0k | const xmlChar *elemName; |
5991 | 61.0k | const xmlChar *attrName; |
5992 | 61.0k | xmlEnumerationPtr tree; |
5993 | | |
5994 | 61.0k | if ((CUR != '<') || (NXT(1) != '!')) |
5995 | 0 | return; |
5996 | 61.0k | SKIP(2); |
5997 | | |
5998 | 61.0k | if (CMP7(CUR_PTR, 'A', 'T', 'T', 'L', 'I', 'S', 'T')) { |
5999 | | #ifdef LIBXML_VALID_ENABLED |
6000 | | int oldInputNr = ctxt->inputNr; |
6001 | | #endif |
6002 | | |
6003 | 55.8k | SKIP(7); |
6004 | 55.8k | if (SKIP_BLANKS_PE == 0) { |
6005 | 942 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6006 | 942 | "Space required after '<!ATTLIST'\n"); |
6007 | 942 | } |
6008 | 55.8k | elemName = xmlParseName(ctxt); |
6009 | 55.8k | if (elemName == NULL) { |
6010 | 359 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6011 | 359 | "ATTLIST: no name for Element\n"); |
6012 | 359 | return; |
6013 | 359 | } |
6014 | 55.4k | SKIP_BLANKS_PE; |
6015 | 55.4k | GROW; |
6016 | 138k | while ((RAW != '>') && (PARSER_STOPPED(ctxt) == 0)) { |
6017 | 110k | int type; |
6018 | 110k | int def; |
6019 | 110k | xmlChar *defaultValue = NULL; |
6020 | | |
6021 | 110k | GROW; |
6022 | 110k | tree = NULL; |
6023 | 110k | attrName = xmlParseName(ctxt); |
6024 | 110k | if (attrName == NULL) { |
6025 | 1.03k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6026 | 1.03k | "ATTLIST: no name for Attribute\n"); |
6027 | 1.03k | break; |
6028 | 1.03k | } |
6029 | 109k | GROW; |
6030 | 109k | if (SKIP_BLANKS_PE == 0) { |
6031 | 2.67k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6032 | 2.67k | "Space required after the attribute name\n"); |
6033 | 2.67k | break; |
6034 | 2.67k | } |
6035 | | |
6036 | 106k | type = xmlParseAttributeType(ctxt, &tree); |
6037 | 106k | if (type <= 0) { |
6038 | 9.35k | break; |
6039 | 9.35k | } |
6040 | | |
6041 | 97.1k | GROW; |
6042 | 97.1k | if (SKIP_BLANKS_PE == 0) { |
6043 | 5.62k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6044 | 5.62k | "Space required after the attribute type\n"); |
6045 | 5.62k | if (tree != NULL) |
6046 | 794 | xmlFreeEnumeration(tree); |
6047 | 5.62k | break; |
6048 | 5.62k | } |
6049 | | |
6050 | 91.5k | def = xmlParseDefaultDecl(ctxt, &defaultValue); |
6051 | 91.5k | if (def <= 0) { |
6052 | 0 | if (defaultValue != NULL) |
6053 | 0 | xmlFree(defaultValue); |
6054 | 0 | if (tree != NULL) |
6055 | 0 | xmlFreeEnumeration(tree); |
6056 | 0 | break; |
6057 | 0 | } |
6058 | 91.5k | if ((type != XML_ATTRIBUTE_CDATA) && (defaultValue != NULL)) |
6059 | 41.2k | xmlAttrNormalizeSpace(defaultValue, defaultValue); |
6060 | | |
6061 | 91.5k | GROW; |
6062 | 91.5k | if (RAW != '>') { |
6063 | 80.9k | if (SKIP_BLANKS_PE == 0) { |
6064 | 8.43k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6065 | 8.43k | "Space required after the attribute default value\n"); |
6066 | 8.43k | if (defaultValue != NULL) |
6067 | 920 | xmlFree(defaultValue); |
6068 | 8.43k | if (tree != NULL) |
6069 | 594 | xmlFreeEnumeration(tree); |
6070 | 8.43k | break; |
6071 | 8.43k | } |
6072 | 80.9k | } |
6073 | 83.1k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
6074 | 83.1k | (ctxt->sax->attributeDecl != NULL)) |
6075 | 52.3k | ctxt->sax->attributeDecl(ctxt->userData, elemName, attrName, |
6076 | 52.3k | type, def, defaultValue, tree); |
6077 | 30.8k | else if (tree != NULL) |
6078 | 1.18k | xmlFreeEnumeration(tree); |
6079 | | |
6080 | 83.1k | if ((ctxt->sax2) && (defaultValue != NULL) && |
6081 | 83.1k | (def != XML_ATTRIBUTE_IMPLIED) && |
6082 | 83.1k | (def != XML_ATTRIBUTE_REQUIRED)) { |
6083 | 51.9k | xmlAddDefAttrs(ctxt, elemName, attrName, defaultValue); |
6084 | 51.9k | } |
6085 | 83.1k | if (ctxt->sax2) { |
6086 | 83.1k | xmlAddSpecialAttr(ctxt, elemName, attrName, type); |
6087 | 83.1k | } |
6088 | 83.1k | if (defaultValue != NULL) |
6089 | 51.9k | xmlFree(defaultValue); |
6090 | 83.1k | GROW; |
6091 | 83.1k | } |
6092 | 55.4k | if (RAW == '>') { |
6093 | | #ifdef LIBXML_VALID_ENABLED |
6094 | | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
6095 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6096 | | "Attribute list declaration doesn't start and" |
6097 | | " stop in the same entity\n", |
6098 | | NULL, NULL); |
6099 | | } |
6100 | | #endif |
6101 | 29.3k | NEXT; |
6102 | 29.3k | } |
6103 | 55.4k | } |
6104 | 61.0k | } |
6105 | | |
6106 | | /** |
6107 | | * Handle PEs and check that we don't pop the entity that started |
6108 | | * a balanced group. |
6109 | | * |
6110 | | * @param ctxt parser context |
6111 | | * @param openInputNr input nr of the entity with opening '(' |
6112 | | */ |
6113 | | static void |
6114 | 304k | xmlSkipBlankCharsPEBalanced(xmlParserCtxt *ctxt, int openInputNr) { |
6115 | 304k | SKIP_BLANKS; |
6116 | 304k | GROW; |
6117 | | |
6118 | 304k | (void) openInputNr; |
6119 | | |
6120 | 304k | if (!PARSER_EXTERNAL(ctxt) && !PARSER_IN_PE(ctxt)) |
6121 | 223k | return; |
6122 | | |
6123 | 81.9k | while (!PARSER_STOPPED(ctxt)) { |
6124 | 81.0k | if (ctxt->input->cur >= ctxt->input->end) { |
6125 | | #ifdef LIBXML_VALID_ENABLED |
6126 | | if ((ctxt->validate) && (ctxt->inputNr <= openInputNr)) { |
6127 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6128 | | "Element content declaration doesn't start " |
6129 | | "and stop in the same entity\n", |
6130 | | NULL, NULL); |
6131 | | } |
6132 | | #endif |
6133 | 592 | if (PARSER_IN_PE(ctxt)) |
6134 | 285 | xmlPopPE(ctxt); |
6135 | 307 | else |
6136 | 307 | break; |
6137 | 80.4k | } else if (RAW == '%') { |
6138 | 1.13k | xmlParsePERefInternal(ctxt, 0); |
6139 | 79.3k | } else { |
6140 | 79.3k | break; |
6141 | 79.3k | } |
6142 | | |
6143 | 1.41k | SKIP_BLANKS; |
6144 | 1.41k | GROW; |
6145 | 1.41k | } |
6146 | 80.4k | } |
6147 | | |
6148 | | /** |
6149 | | * Parse the declaration for a Mixed Element content |
6150 | | * The leading '(' and spaces have been skipped in #xmlParseElementContentDecl |
6151 | | * |
6152 | | * @deprecated Internal function, don't use. |
6153 | | * |
6154 | | * [51] Mixed ::= '(' S? '#PCDATA' (S? '|' S? Name)* S? ')*' | |
6155 | | * '(' S? '#PCDATA' S? ')' |
6156 | | * |
6157 | | * [ VC: Proper Group/PE Nesting ] applies to [51] too (see [49]) |
6158 | | * |
6159 | | * [ VC: No Duplicate Types ] |
6160 | | * The same name must not appear more than once in a single |
6161 | | * mixed-content declaration. |
6162 | | * |
6163 | | * @param ctxt an XML parser context |
6164 | | * @param openInputNr the input used for the current entity, needed for |
6165 | | * boundary checks |
6166 | | * @returns the list of the xmlElementContent describing the element choices |
6167 | | */ |
6168 | | xmlElementContent * |
6169 | 11.1k | xmlParseElementMixedContentDecl(xmlParserCtxt *ctxt, int openInputNr) { |
6170 | 11.1k | xmlElementContentPtr ret = NULL, cur = NULL, n; |
6171 | 11.1k | const xmlChar *elem = NULL; |
6172 | | |
6173 | 11.1k | GROW; |
6174 | 11.1k | if (CMP7(CUR_PTR, '#', 'P', 'C', 'D', 'A', 'T', 'A')) { |
6175 | 11.1k | SKIP(7); |
6176 | 11.1k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6177 | 11.1k | if (RAW == ')') { |
6178 | | #ifdef LIBXML_VALID_ENABLED |
6179 | | if ((ctxt->validate) && (ctxt->inputNr > openInputNr)) { |
6180 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6181 | | "Element content declaration doesn't start " |
6182 | | "and stop in the same entity\n", |
6183 | | NULL, NULL); |
6184 | | } |
6185 | | #endif |
6186 | 8.06k | NEXT; |
6187 | 8.06k | ret = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_PCDATA); |
6188 | 8.06k | if (ret == NULL) |
6189 | 3 | goto mem_error; |
6190 | 8.06k | if (RAW == '*') { |
6191 | 204 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6192 | 204 | NEXT; |
6193 | 204 | } |
6194 | 8.06k | return(ret); |
6195 | 8.06k | } |
6196 | 3.12k | if ((RAW == '(') || (RAW == '|')) { |
6197 | 2.51k | ret = cur = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_PCDATA); |
6198 | 2.51k | if (ret == NULL) |
6199 | 2 | goto mem_error; |
6200 | 2.51k | } |
6201 | 5.43k | while ((RAW == '|') && (PARSER_STOPPED(ctxt) == 0)) { |
6202 | 2.62k | NEXT; |
6203 | 2.62k | n = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_OR); |
6204 | 2.62k | if (n == NULL) |
6205 | 1 | goto mem_error; |
6206 | 2.62k | if (elem == NULL) { |
6207 | 2.28k | n->c1 = cur; |
6208 | 2.28k | if (cur != NULL) |
6209 | 2.28k | cur->parent = n; |
6210 | 2.28k | ret = cur = n; |
6211 | 2.28k | } else { |
6212 | 337 | cur->c2 = n; |
6213 | 337 | n->parent = cur; |
6214 | 337 | n->c1 = xmlNewDocElementContent(ctxt->myDoc, elem, XML_ELEMENT_CONTENT_ELEMENT); |
6215 | 337 | if (n->c1 == NULL) |
6216 | 1 | goto mem_error; |
6217 | 336 | n->c1->parent = n; |
6218 | 336 | cur = n; |
6219 | 336 | } |
6220 | 2.61k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6221 | 2.61k | elem = xmlParseName(ctxt); |
6222 | 2.61k | if (elem == NULL) { |
6223 | 304 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6224 | 304 | "xmlParseElementMixedContentDecl : Name expected\n"); |
6225 | 304 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6226 | 304 | return(NULL); |
6227 | 304 | } |
6228 | 2.31k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6229 | 2.31k | } |
6230 | 2.81k | if ((RAW == ')') && (NXT(1) == '*')) { |
6231 | 1.36k | if (elem != NULL) { |
6232 | 1.36k | cur->c2 = xmlNewDocElementContent(ctxt->myDoc, elem, |
6233 | 1.36k | XML_ELEMENT_CONTENT_ELEMENT); |
6234 | 1.36k | if (cur->c2 == NULL) |
6235 | 1 | goto mem_error; |
6236 | 1.36k | cur->c2->parent = cur; |
6237 | 1.36k | } |
6238 | 1.36k | if (ret != NULL) |
6239 | 1.36k | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6240 | | #ifdef LIBXML_VALID_ENABLED |
6241 | | if ((ctxt->validate) && (ctxt->inputNr > openInputNr)) { |
6242 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6243 | | "Element content declaration doesn't start " |
6244 | | "and stop in the same entity\n", |
6245 | | NULL, NULL); |
6246 | | } |
6247 | | #endif |
6248 | 1.36k | SKIP(2); |
6249 | 1.44k | } else { |
6250 | 1.44k | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6251 | 1.44k | xmlFatalErr(ctxt, XML_ERR_MIXED_NOT_STARTED, NULL); |
6252 | 1.44k | return(NULL); |
6253 | 1.44k | } |
6254 | | |
6255 | 2.81k | } else { |
6256 | 0 | xmlFatalErr(ctxt, XML_ERR_PCDATA_REQUIRED, NULL); |
6257 | 0 | } |
6258 | 1.36k | return(ret); |
6259 | | |
6260 | 8 | mem_error: |
6261 | 8 | xmlErrMemory(ctxt); |
6262 | 8 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6263 | 8 | return(NULL); |
6264 | 11.1k | } |
6265 | | |
6266 | | /** |
6267 | | * Parse the declaration for a Mixed Element content |
6268 | | * The leading '(' and spaces have been skipped in #xmlParseElementContentDecl |
6269 | | * |
6270 | | * [47] children ::= (choice | seq) ('?' | '*' | '+')? |
6271 | | * |
6272 | | * [48] cp ::= (Name | choice | seq) ('?' | '*' | '+')? |
6273 | | * |
6274 | | * [49] choice ::= '(' S? cp ( S? '|' S? cp )* S? ')' |
6275 | | * |
6276 | | * [50] seq ::= '(' S? cp ( S? ',' S? cp )* S? ')' |
6277 | | * |
6278 | | * [ VC: Proper Group/PE Nesting ] applies to [49] and [50] |
6279 | | * TODO Parameter-entity replacement text must be properly nested |
6280 | | * with parenthesized groups. That is to say, if either of the |
6281 | | * opening or closing parentheses in a choice, seq, or Mixed |
6282 | | * construct is contained in the replacement text for a parameter |
6283 | | * entity, both must be contained in the same replacement text. For |
6284 | | * interoperability, if a parameter-entity reference appears in a |
6285 | | * choice, seq, or Mixed construct, its replacement text should not |
6286 | | * be empty, and neither the first nor last non-blank character of |
6287 | | * the replacement text should be a connector (| or ,). |
6288 | | * |
6289 | | * @param ctxt an XML parser context |
6290 | | * @param openInputNr the input used for the current entity, needed for |
6291 | | * boundary checks |
6292 | | * @param depth the level of recursion |
6293 | | * @returns the tree of xmlElementContent describing the element |
6294 | | * hierarchy. |
6295 | | */ |
6296 | | static xmlElementContentPtr |
6297 | | xmlParseElementChildrenContentDeclPriv(xmlParserCtxtPtr ctxt, int openInputNr, |
6298 | 110k | int depth) { |
6299 | 110k | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 2048 : 256; |
6300 | 110k | xmlElementContentPtr ret = NULL, cur = NULL, last = NULL, op = NULL; |
6301 | 110k | const xmlChar *elem; |
6302 | 110k | xmlChar type = 0; |
6303 | | |
6304 | 110k | if (depth > maxDepth) { |
6305 | 229 | xmlFatalErrMsgInt(ctxt, XML_ERR_RESOURCE_LIMIT, |
6306 | 229 | "xmlParseElementChildrenContentDecl : depth %d too deep, " |
6307 | 229 | "use XML_PARSE_HUGE\n", depth); |
6308 | 229 | return(NULL); |
6309 | 229 | } |
6310 | 110k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6311 | 110k | if (RAW == '(') { |
6312 | 70.1k | int newInputNr = ctxt->inputNr; |
6313 | | |
6314 | | /* Recurse on first child */ |
6315 | 70.1k | NEXT; |
6316 | 70.1k | cur = ret = xmlParseElementChildrenContentDeclPriv(ctxt, newInputNr, |
6317 | 70.1k | depth + 1); |
6318 | 70.1k | if (cur == NULL) |
6319 | 63.4k | return(NULL); |
6320 | 70.1k | } else { |
6321 | 40.2k | elem = xmlParseName(ctxt); |
6322 | 40.2k | if (elem == NULL) { |
6323 | 2.43k | xmlFatalErr(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, NULL); |
6324 | 2.43k | return(NULL); |
6325 | 2.43k | } |
6326 | 37.8k | cur = ret = xmlNewDocElementContent(ctxt->myDoc, elem, XML_ELEMENT_CONTENT_ELEMENT); |
6327 | 37.8k | if (cur == NULL) { |
6328 | 6 | xmlErrMemory(ctxt); |
6329 | 6 | return(NULL); |
6330 | 6 | } |
6331 | 37.8k | GROW; |
6332 | 37.8k | if (RAW == '?') { |
6333 | 3.51k | cur->ocur = XML_ELEMENT_CONTENT_OPT; |
6334 | 3.51k | NEXT; |
6335 | 34.3k | } else if (RAW == '*') { |
6336 | 2.85k | cur->ocur = XML_ELEMENT_CONTENT_MULT; |
6337 | 2.85k | NEXT; |
6338 | 31.4k | } else if (RAW == '+') { |
6339 | 561 | cur->ocur = XML_ELEMENT_CONTENT_PLUS; |
6340 | 561 | NEXT; |
6341 | 30.8k | } else { |
6342 | 30.8k | cur->ocur = XML_ELEMENT_CONTENT_ONCE; |
6343 | 30.8k | } |
6344 | 37.8k | GROW; |
6345 | 37.8k | } |
6346 | 87.7k | while (!PARSER_STOPPED(ctxt)) { |
6347 | 86.8k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6348 | 86.8k | if (RAW == ')') |
6349 | 33.8k | break; |
6350 | | /* |
6351 | | * Each loop we parse one separator and one element. |
6352 | | */ |
6353 | 53.0k | if (RAW == ',') { |
6354 | 31.6k | if (type == 0) type = CUR; |
6355 | | |
6356 | | /* |
6357 | | * Detect "Name | Name , Name" error |
6358 | | */ |
6359 | 20.7k | else if (type != CUR) { |
6360 | 222 | xmlFatalErrMsgInt(ctxt, XML_ERR_SEPARATOR_REQUIRED, |
6361 | 222 | "xmlParseElementChildrenContentDecl : '%c' expected\n", |
6362 | 222 | type); |
6363 | 222 | if ((last != NULL) && (last != ret)) |
6364 | 222 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6365 | 222 | if (ret != NULL) |
6366 | 222 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6367 | 222 | return(NULL); |
6368 | 222 | } |
6369 | 31.4k | NEXT; |
6370 | | |
6371 | 31.4k | op = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_SEQ); |
6372 | 31.4k | if (op == NULL) { |
6373 | 4 | xmlErrMemory(ctxt); |
6374 | 4 | if ((last != NULL) && (last != ret)) |
6375 | 3 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6376 | 4 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6377 | 4 | return(NULL); |
6378 | 4 | } |
6379 | 31.4k | if (last == NULL) { |
6380 | 10.9k | op->c1 = ret; |
6381 | 10.9k | if (ret != NULL) |
6382 | 10.9k | ret->parent = op; |
6383 | 10.9k | ret = cur = op; |
6384 | 20.5k | } else { |
6385 | 20.5k | cur->c2 = op; |
6386 | 20.5k | if (op != NULL) |
6387 | 20.5k | op->parent = cur; |
6388 | 20.5k | op->c1 = last; |
6389 | 20.5k | if (last != NULL) |
6390 | 20.5k | last->parent = op; |
6391 | 20.5k | cur =op; |
6392 | 20.5k | last = NULL; |
6393 | 20.5k | } |
6394 | 31.4k | } else if (RAW == '|') { |
6395 | 17.4k | if (type == 0) type = CUR; |
6396 | | |
6397 | | /* |
6398 | | * Detect "Name , Name | Name" error |
6399 | | */ |
6400 | 4.23k | else if (type != CUR) { |
6401 | 234 | xmlFatalErrMsgInt(ctxt, XML_ERR_SEPARATOR_REQUIRED, |
6402 | 234 | "xmlParseElementChildrenContentDecl : '%c' expected\n", |
6403 | 234 | type); |
6404 | 234 | if ((last != NULL) && (last != ret)) |
6405 | 234 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6406 | 234 | if (ret != NULL) |
6407 | 234 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6408 | 234 | return(NULL); |
6409 | 234 | } |
6410 | 17.2k | NEXT; |
6411 | | |
6412 | 17.2k | op = xmlNewDocElementContent(ctxt->myDoc, NULL, XML_ELEMENT_CONTENT_OR); |
6413 | 17.2k | if (op == NULL) { |
6414 | 3 | xmlErrMemory(ctxt); |
6415 | 3 | if ((last != NULL) && (last != ret)) |
6416 | 1 | xmlFreeDocElementContent(ctxt->myDoc, last); |
6417 | 3 | if (ret != NULL) |
6418 | 3 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6419 | 3 | return(NULL); |
6420 | 3 | } |
6421 | 17.2k | if (last == NULL) { |
6422 | 13.2k | op->c1 = ret; |
6423 | 13.2k | if (ret != NULL) |
6424 | 13.2k | ret->parent = op; |
6425 | 13.2k | ret = cur = op; |
6426 | 13.2k | } else { |
6427 | 4.00k | cur->c2 = op; |
6428 | 4.00k | if (op != NULL) |
6429 | 4.00k | op->parent = cur; |
6430 | 4.00k | op->c1 = last; |
6431 | 4.00k | if (last != NULL) |
6432 | 4.00k | last->parent = op; |
6433 | 4.00k | cur =op; |
6434 | 4.00k | last = NULL; |
6435 | 4.00k | } |
6436 | 17.2k | } else { |
6437 | 3.90k | xmlFatalErr(ctxt, XML_ERR_ELEMCONTENT_NOT_FINISHED, NULL); |
6438 | 3.90k | if ((last != NULL) && (last != ret)) |
6439 | 1.09k | xmlFreeDocElementContent(ctxt->myDoc, last); |
6440 | 3.90k | if (ret != NULL) |
6441 | 3.90k | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6442 | 3.90k | return(NULL); |
6443 | 3.90k | } |
6444 | 48.7k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6445 | 48.7k | if (RAW == '(') { |
6446 | 9.73k | int newInputNr = ctxt->inputNr; |
6447 | | |
6448 | | /* Recurse on second child */ |
6449 | 9.73k | NEXT; |
6450 | 9.73k | last = xmlParseElementChildrenContentDeclPriv(ctxt, newInputNr, |
6451 | 9.73k | depth + 1); |
6452 | 9.73k | if (last == NULL) { |
6453 | 4.81k | if (ret != NULL) |
6454 | 4.81k | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6455 | 4.81k | return(NULL); |
6456 | 4.81k | } |
6457 | 38.9k | } else { |
6458 | 38.9k | elem = xmlParseName(ctxt); |
6459 | 38.9k | if (elem == NULL) { |
6460 | 729 | xmlFatalErr(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, NULL); |
6461 | 729 | if (ret != NULL) |
6462 | 729 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6463 | 729 | return(NULL); |
6464 | 729 | } |
6465 | 38.2k | last = xmlNewDocElementContent(ctxt->myDoc, elem, XML_ELEMENT_CONTENT_ELEMENT); |
6466 | 38.2k | if (last == NULL) { |
6467 | 4 | xmlErrMemory(ctxt); |
6468 | 4 | if (ret != NULL) |
6469 | 4 | xmlFreeDocElementContent(ctxt->myDoc, ret); |
6470 | 4 | return(NULL); |
6471 | 4 | } |
6472 | 38.2k | if (RAW == '?') { |
6473 | 14.0k | last->ocur = XML_ELEMENT_CONTENT_OPT; |
6474 | 14.0k | NEXT; |
6475 | 24.1k | } else if (RAW == '*') { |
6476 | 8.69k | last->ocur = XML_ELEMENT_CONTENT_MULT; |
6477 | 8.69k | NEXT; |
6478 | 15.4k | } else if (RAW == '+') { |
6479 | 447 | last->ocur = XML_ELEMENT_CONTENT_PLUS; |
6480 | 447 | NEXT; |
6481 | 15.0k | } else { |
6482 | 15.0k | last->ocur = XML_ELEMENT_CONTENT_ONCE; |
6483 | 15.0k | } |
6484 | 38.2k | } |
6485 | 48.7k | } |
6486 | 34.6k | if ((cur != NULL) && (last != NULL)) { |
6487 | 17.0k | cur->c2 = last; |
6488 | 17.0k | if (last != NULL) |
6489 | 17.0k | last->parent = cur; |
6490 | 17.0k | } |
6491 | | #ifdef LIBXML_VALID_ENABLED |
6492 | | if ((ctxt->validate) && (ctxt->inputNr > openInputNr)) { |
6493 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6494 | | "Element content declaration doesn't start " |
6495 | | "and stop in the same entity\n", |
6496 | | NULL, NULL); |
6497 | | } |
6498 | | #endif |
6499 | 34.6k | NEXT; |
6500 | 34.6k | if (RAW == '?') { |
6501 | 1.50k | if (ret != NULL) { |
6502 | 1.50k | if ((ret->ocur == XML_ELEMENT_CONTENT_PLUS) || |
6503 | 1.50k | (ret->ocur == XML_ELEMENT_CONTENT_MULT)) |
6504 | 524 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6505 | 979 | else |
6506 | 979 | ret->ocur = XML_ELEMENT_CONTENT_OPT; |
6507 | 1.50k | } |
6508 | 1.50k | NEXT; |
6509 | 33.1k | } else if (RAW == '*') { |
6510 | 6.84k | if (ret != NULL) { |
6511 | 6.84k | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6512 | 6.84k | cur = ret; |
6513 | | /* |
6514 | | * Some normalization: |
6515 | | * (a | b* | c?)* == (a | b | c)* |
6516 | | */ |
6517 | 14.4k | while ((cur != NULL) && (cur->type == XML_ELEMENT_CONTENT_OR)) { |
6518 | 7.56k | if ((cur->c1 != NULL) && |
6519 | 7.56k | ((cur->c1->ocur == XML_ELEMENT_CONTENT_OPT) || |
6520 | 7.56k | (cur->c1->ocur == XML_ELEMENT_CONTENT_MULT))) |
6521 | 435 | cur->c1->ocur = XML_ELEMENT_CONTENT_ONCE; |
6522 | 7.56k | if ((cur->c2 != NULL) && |
6523 | 7.56k | ((cur->c2->ocur == XML_ELEMENT_CONTENT_OPT) || |
6524 | 7.56k | (cur->c2->ocur == XML_ELEMENT_CONTENT_MULT))) |
6525 | 1.17k | cur->c2->ocur = XML_ELEMENT_CONTENT_ONCE; |
6526 | 7.56k | cur = cur->c2; |
6527 | 7.56k | } |
6528 | 6.84k | } |
6529 | 6.84k | NEXT; |
6530 | 26.3k | } else if (RAW == '+') { |
6531 | 1.95k | if (ret != NULL) { |
6532 | 1.95k | int found = 0; |
6533 | | |
6534 | 1.95k | if ((ret->ocur == XML_ELEMENT_CONTENT_OPT) || |
6535 | 1.95k | (ret->ocur == XML_ELEMENT_CONTENT_MULT)) |
6536 | 495 | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6537 | 1.46k | else |
6538 | 1.46k | ret->ocur = XML_ELEMENT_CONTENT_PLUS; |
6539 | | /* |
6540 | | * Some normalization: |
6541 | | * (a | b*)+ == (a | b)* |
6542 | | * (a | b?)+ == (a | b)* |
6543 | | */ |
6544 | 5.21k | while ((cur != NULL) && (cur->type == XML_ELEMENT_CONTENT_OR)) { |
6545 | 3.25k | if ((cur->c1 != NULL) && |
6546 | 3.25k | ((cur->c1->ocur == XML_ELEMENT_CONTENT_OPT) || |
6547 | 3.25k | (cur->c1->ocur == XML_ELEMENT_CONTENT_MULT))) { |
6548 | 957 | cur->c1->ocur = XML_ELEMENT_CONTENT_ONCE; |
6549 | 957 | found = 1; |
6550 | 957 | } |
6551 | 3.25k | if ((cur->c2 != NULL) && |
6552 | 3.25k | ((cur->c2->ocur == XML_ELEMENT_CONTENT_OPT) || |
6553 | 3.25k | (cur->c2->ocur == XML_ELEMENT_CONTENT_MULT))) { |
6554 | 777 | cur->c2->ocur = XML_ELEMENT_CONTENT_ONCE; |
6555 | 777 | found = 1; |
6556 | 777 | } |
6557 | 3.25k | cur = cur->c2; |
6558 | 3.25k | } |
6559 | 1.95k | if (found) |
6560 | 1.32k | ret->ocur = XML_ELEMENT_CONTENT_MULT; |
6561 | 1.95k | } |
6562 | 1.95k | NEXT; |
6563 | 1.95k | } |
6564 | 34.6k | return(ret); |
6565 | 44.5k | } |
6566 | | |
6567 | | /** |
6568 | | * Parse the declaration for a Mixed Element content |
6569 | | * The leading '(' and spaces have been skipped in #xmlParseElementContentDecl |
6570 | | * |
6571 | | * @deprecated Internal function, don't use. |
6572 | | * |
6573 | | * [47] children ::= (choice | seq) ('?' | '*' | '+')? |
6574 | | * |
6575 | | * [48] cp ::= (Name | choice | seq) ('?' | '*' | '+')? |
6576 | | * |
6577 | | * [49] choice ::= '(' S? cp ( S? '|' S? cp )* S? ')' |
6578 | | * |
6579 | | * [50] seq ::= '(' S? cp ( S? ',' S? cp )* S? ')' |
6580 | | * |
6581 | | * [ VC: Proper Group/PE Nesting ] applies to [49] and [50] |
6582 | | * TODO Parameter-entity replacement text must be properly nested |
6583 | | * with parenthesized groups. That is to say, if either of the |
6584 | | * opening or closing parentheses in a choice, seq, or Mixed |
6585 | | * construct is contained in the replacement text for a parameter |
6586 | | * entity, both must be contained in the same replacement text. For |
6587 | | * interoperability, if a parameter-entity reference appears in a |
6588 | | * choice, seq, or Mixed construct, its replacement text should not |
6589 | | * be empty, and neither the first nor last non-blank character of |
6590 | | * the replacement text should be a connector (| or ,). |
6591 | | * |
6592 | | * @param ctxt an XML parser context |
6593 | | * @param inputchk the input used for the current entity, needed for boundary checks |
6594 | | * @returns the tree of xmlElementContent describing the element |
6595 | | * hierarchy. |
6596 | | */ |
6597 | | xmlElementContent * |
6598 | 0 | xmlParseElementChildrenContentDecl(xmlParserCtxt *ctxt, int inputchk) { |
6599 | | /* stub left for API/ABI compat */ |
6600 | 0 | return(xmlParseElementChildrenContentDeclPriv(ctxt, inputchk, 1)); |
6601 | 0 | } |
6602 | | |
6603 | | /** |
6604 | | * Parse the declaration for an Element content either Mixed or Children, |
6605 | | * the cases EMPTY and ANY are handled directly in #xmlParseElementDecl |
6606 | | * |
6607 | | * @deprecated Internal function, don't use. |
6608 | | * |
6609 | | * [46] contentspec ::= 'EMPTY' | 'ANY' | Mixed | children |
6610 | | * |
6611 | | * @param ctxt an XML parser context |
6612 | | * @param name the name of the element being defined. |
6613 | | * @param result the Element Content pointer will be stored here if any |
6614 | | * @returns an xmlElementTypeVal value or -1 on error |
6615 | | */ |
6616 | | |
6617 | | int |
6618 | | xmlParseElementContentDecl(xmlParserCtxt *ctxt, const xmlChar *name, |
6619 | 41.9k | xmlElementContent **result) { |
6620 | | |
6621 | 41.9k | xmlElementContentPtr tree = NULL; |
6622 | 41.9k | int openInputNr = ctxt->inputNr; |
6623 | 41.9k | int res; |
6624 | | |
6625 | 41.9k | *result = NULL; |
6626 | | |
6627 | 41.9k | if (RAW != '(') { |
6628 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, |
6629 | 0 | "xmlParseElementContentDecl : %s '(' expected\n", name); |
6630 | 0 | return(-1); |
6631 | 0 | } |
6632 | 41.9k | NEXT; |
6633 | 41.9k | xmlSkipBlankCharsPEBalanced(ctxt, openInputNr); |
6634 | 41.9k | if (CMP7(CUR_PTR, '#', 'P', 'C', 'D', 'A', 'T', 'A')) { |
6635 | 11.1k | tree = xmlParseElementMixedContentDecl(ctxt, openInputNr); |
6636 | 11.1k | res = XML_ELEMENT_TYPE_MIXED; |
6637 | 30.7k | } else { |
6638 | 30.7k | tree = xmlParseElementChildrenContentDeclPriv(ctxt, openInputNr, 1); |
6639 | 30.7k | res = XML_ELEMENT_TYPE_ELEMENT; |
6640 | 30.7k | } |
6641 | 41.9k | if (tree == NULL) |
6642 | 9.52k | return(-1); |
6643 | 32.4k | SKIP_BLANKS_PE; |
6644 | 32.4k | *result = tree; |
6645 | 32.4k | return(res); |
6646 | 41.9k | } |
6647 | | |
6648 | | /** |
6649 | | * Parse an element declaration. Always consumes '<!'. |
6650 | | * |
6651 | | * @deprecated Internal function, don't use. |
6652 | | * |
6653 | | * [45] elementdecl ::= '<!ELEMENT' S Name S contentspec S? '>' |
6654 | | * |
6655 | | * [ VC: Unique Element Type Declaration ] |
6656 | | * No element type may be declared more than once |
6657 | | * |
6658 | | * @param ctxt an XML parser context |
6659 | | * @returns the type of the element, or -1 in case of error |
6660 | | */ |
6661 | | int |
6662 | 65.2k | xmlParseElementDecl(xmlParserCtxt *ctxt) { |
6663 | 65.2k | const xmlChar *name; |
6664 | 65.2k | int ret = -1; |
6665 | 65.2k | xmlElementContentPtr content = NULL; |
6666 | | |
6667 | 65.2k | if ((CUR != '<') || (NXT(1) != '!')) |
6668 | 0 | return(ret); |
6669 | 65.2k | SKIP(2); |
6670 | | |
6671 | | /* GROW; done in the caller */ |
6672 | 65.2k | if (CMP7(CUR_PTR, 'E', 'L', 'E', 'M', 'E', 'N', 'T')) { |
6673 | | #ifdef LIBXML_VALID_ENABLED |
6674 | | int oldInputNr = ctxt->inputNr; |
6675 | | #endif |
6676 | | |
6677 | 62.1k | SKIP(7); |
6678 | 62.1k | if (SKIP_BLANKS_PE == 0) { |
6679 | 2.20k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6680 | 2.20k | "Space required after 'ELEMENT'\n"); |
6681 | 2.20k | return(-1); |
6682 | 2.20k | } |
6683 | 59.9k | name = xmlParseName(ctxt); |
6684 | 59.9k | if (name == NULL) { |
6685 | 272 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
6686 | 272 | "xmlParseElementDecl: no name for Element\n"); |
6687 | 272 | return(-1); |
6688 | 272 | } |
6689 | 59.6k | if (SKIP_BLANKS_PE == 0) { |
6690 | 2.01k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6691 | 2.01k | "Space required after the element name\n"); |
6692 | 2.01k | } |
6693 | 59.6k | if (CMP5(CUR_PTR, 'E', 'M', 'P', 'T', 'Y')) { |
6694 | 11.9k | SKIP(5); |
6695 | | /* |
6696 | | * Element must always be empty. |
6697 | | */ |
6698 | 11.9k | ret = XML_ELEMENT_TYPE_EMPTY; |
6699 | 47.7k | } else if ((RAW == 'A') && (NXT(1) == 'N') && |
6700 | 47.7k | (NXT(2) == 'Y')) { |
6701 | 2.02k | SKIP(3); |
6702 | | /* |
6703 | | * Element is a generic container. |
6704 | | */ |
6705 | 2.02k | ret = XML_ELEMENT_TYPE_ANY; |
6706 | 45.7k | } else if (RAW == '(') { |
6707 | 41.9k | ret = xmlParseElementContentDecl(ctxt, name, &content); |
6708 | 41.9k | if (ret <= 0) |
6709 | 9.52k | return(-1); |
6710 | 41.9k | } else { |
6711 | | /* |
6712 | | * [ WFC: PEs in Internal Subset ] error handling. |
6713 | | */ |
6714 | 3.81k | xmlFatalErrMsg(ctxt, XML_ERR_ELEMCONTENT_NOT_STARTED, |
6715 | 3.81k | "xmlParseElementDecl: 'EMPTY', 'ANY' or '(' expected\n"); |
6716 | 3.81k | return(-1); |
6717 | 3.81k | } |
6718 | | |
6719 | 46.3k | SKIP_BLANKS_PE; |
6720 | | |
6721 | 46.3k | if (RAW != '>') { |
6722 | 2.89k | xmlFatalErr(ctxt, XML_ERR_GT_REQUIRED, NULL); |
6723 | 2.89k | if (content != NULL) { |
6724 | 1.95k | xmlFreeDocElementContent(ctxt->myDoc, content); |
6725 | 1.95k | } |
6726 | 43.4k | } else { |
6727 | | #ifdef LIBXML_VALID_ENABLED |
6728 | | if ((ctxt->validate) && (ctxt->inputNr > oldInputNr)) { |
6729 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6730 | | "Element declaration doesn't start and stop in" |
6731 | | " the same entity\n", |
6732 | | NULL, NULL); |
6733 | | } |
6734 | | #endif |
6735 | | |
6736 | 43.4k | NEXT; |
6737 | 43.4k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
6738 | 43.4k | (ctxt->sax->elementDecl != NULL)) { |
6739 | 36.6k | if (content != NULL) |
6740 | 25.5k | content->parent = NULL; |
6741 | 36.6k | ctxt->sax->elementDecl(ctxt->userData, name, ret, |
6742 | 36.6k | content); |
6743 | 36.6k | if ((content != NULL) && (content->parent == NULL)) { |
6744 | | /* |
6745 | | * this is a trick: if xmlAddElementDecl is called, |
6746 | | * instead of copying the full tree it is plugged directly |
6747 | | * if called from the parser. Avoid duplicating the |
6748 | | * interfaces or change the API/ABI |
6749 | | */ |
6750 | 760 | xmlFreeDocElementContent(ctxt->myDoc, content); |
6751 | 760 | } |
6752 | 36.6k | } else if (content != NULL) { |
6753 | 4.87k | xmlFreeDocElementContent(ctxt->myDoc, content); |
6754 | 4.87k | } |
6755 | 43.4k | } |
6756 | 46.3k | } |
6757 | 49.4k | return(ret); |
6758 | 65.2k | } |
6759 | | |
6760 | | /** |
6761 | | * Parse a conditional section. Always consumes '<!['. |
6762 | | * |
6763 | | * [61] conditionalSect ::= includeSect | ignoreSect |
6764 | | * [62] includeSect ::= '<![' S? 'INCLUDE' S? '[' extSubsetDecl ']]>' |
6765 | | * [63] ignoreSect ::= '<![' S? 'IGNORE' S? '[' ignoreSectContents* ']]>' |
6766 | | * [64] ignoreSectContents ::= Ignore ('<![' ignoreSectContents ']]>' |
6767 | | * Ignore)* |
6768 | | * [65] Ignore ::= Char* - (Char* ('<![' | ']]>') Char*) |
6769 | | * @param ctxt an XML parser context |
6770 | | */ |
6771 | | |
6772 | | static void |
6773 | 8.40k | xmlParseConditionalSections(xmlParserCtxtPtr ctxt) { |
6774 | 8.40k | size_t depth = 0; |
6775 | 8.40k | int isFreshPE = 0; |
6776 | 8.40k | int oldInputNr = ctxt->inputNr; |
6777 | 8.40k | int declInputNr = ctxt->inputNr; |
6778 | | |
6779 | 18.2k | while (!PARSER_STOPPED(ctxt)) { |
6780 | 18.1k | if (ctxt->input->cur >= ctxt->input->end) { |
6781 | 304 | if (ctxt->inputNr <= oldInputNr) { |
6782 | 304 | xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL); |
6783 | 304 | return; |
6784 | 304 | } |
6785 | | |
6786 | 0 | xmlPopPE(ctxt); |
6787 | 0 | declInputNr = ctxt->inputNr; |
6788 | 17.8k | } else if ((RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
6789 | 9.61k | SKIP(3); |
6790 | 9.61k | SKIP_BLANKS_PE; |
6791 | | |
6792 | 9.61k | isFreshPE = 0; |
6793 | | |
6794 | 9.61k | if (CMP7(CUR_PTR, 'I', 'N', 'C', 'L', 'U', 'D', 'E')) { |
6795 | 5.43k | SKIP(7); |
6796 | 5.43k | SKIP_BLANKS_PE; |
6797 | 5.43k | if (RAW != '[') { |
6798 | 513 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID, NULL); |
6799 | 513 | return; |
6800 | 513 | } |
6801 | | #ifdef LIBXML_VALID_ENABLED |
6802 | | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6803 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6804 | | "All markup of the conditional section is" |
6805 | | " not in the same entity\n", |
6806 | | NULL, NULL); |
6807 | | } |
6808 | | #endif |
6809 | 4.92k | NEXT; |
6810 | | |
6811 | 4.92k | depth++; |
6812 | 4.92k | } else if (CMP6(CUR_PTR, 'I', 'G', 'N', 'O', 'R', 'E')) { |
6813 | 1.15k | size_t ignoreDepth = 0; |
6814 | | |
6815 | 1.15k | SKIP(6); |
6816 | 1.15k | SKIP_BLANKS_PE; |
6817 | 1.15k | if (RAW != '[') { |
6818 | 304 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID, NULL); |
6819 | 304 | return; |
6820 | 304 | } |
6821 | | #ifdef LIBXML_VALID_ENABLED |
6822 | | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6823 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6824 | | "All markup of the conditional section is" |
6825 | | " not in the same entity\n", |
6826 | | NULL, NULL); |
6827 | | } |
6828 | | #endif |
6829 | 851 | NEXT; |
6830 | | |
6831 | 62.7k | while (PARSER_STOPPED(ctxt) == 0) { |
6832 | 62.7k | if (RAW == 0) { |
6833 | 521 | xmlFatalErr(ctxt, XML_ERR_CONDSEC_NOT_FINISHED, NULL); |
6834 | 521 | return; |
6835 | 521 | } |
6836 | 62.1k | if ((RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
6837 | 885 | SKIP(3); |
6838 | 885 | ignoreDepth++; |
6839 | | /* Check for integer overflow */ |
6840 | 885 | if (ignoreDepth == 0) { |
6841 | 0 | xmlErrMemory(ctxt); |
6842 | 0 | return; |
6843 | 0 | } |
6844 | 61.3k | } else if ((RAW == ']') && (NXT(1) == ']') && |
6845 | 61.3k | (NXT(2) == '>')) { |
6846 | 831 | SKIP(3); |
6847 | 831 | if (ignoreDepth == 0) |
6848 | 328 | break; |
6849 | 503 | ignoreDepth--; |
6850 | 60.4k | } else { |
6851 | 60.4k | NEXT; |
6852 | 60.4k | } |
6853 | 62.1k | } |
6854 | | |
6855 | | #ifdef LIBXML_VALID_ENABLED |
6856 | | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6857 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6858 | | "All markup of the conditional section is" |
6859 | | " not in the same entity\n", |
6860 | | NULL, NULL); |
6861 | | } |
6862 | | #endif |
6863 | 3.02k | } else { |
6864 | 3.02k | xmlFatalErr(ctxt, XML_ERR_CONDSEC_INVALID_KEYWORD, NULL); |
6865 | 3.02k | return; |
6866 | 3.02k | } |
6867 | 9.61k | } else if ((depth > 0) && |
6868 | 8.22k | (RAW == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { |
6869 | 477 | if (isFreshPE) { |
6870 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_CONDSEC_INVALID, |
6871 | 0 | "Parameter entity must match " |
6872 | 0 | "extSubsetDecl\n"); |
6873 | 0 | return; |
6874 | 0 | } |
6875 | | |
6876 | 477 | depth--; |
6877 | | #ifdef LIBXML_VALID_ENABLED |
6878 | | if ((ctxt->validate) && (ctxt->inputNr > declInputNr)) { |
6879 | | xmlValidityError(ctxt, XML_ERR_ENTITY_BOUNDARY, |
6880 | | "All markup of the conditional section is not" |
6881 | | " in the same entity\n", |
6882 | | NULL, NULL); |
6883 | | } |
6884 | | #endif |
6885 | 477 | SKIP(3); |
6886 | 7.74k | } else if ((RAW == '<') && ((NXT(1) == '!') || (NXT(1) == '?'))) { |
6887 | 3.89k | isFreshPE = 0; |
6888 | 3.89k | xmlParseMarkupDecl(ctxt); |
6889 | 3.89k | } else if (RAW == '%') { |
6890 | 510 | xmlParsePERefInternal(ctxt, 1); |
6891 | 510 | if (ctxt->inputNr > declInputNr) { |
6892 | 0 | isFreshPE = 1; |
6893 | 0 | declInputNr = ctxt->inputNr; |
6894 | 0 | } |
6895 | 3.33k | } else { |
6896 | 3.33k | xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL); |
6897 | 3.33k | return; |
6898 | 3.33k | } |
6899 | | |
6900 | 10.1k | if (depth == 0) |
6901 | 334 | break; |
6902 | | |
6903 | 9.80k | SKIP_BLANKS; |
6904 | 9.80k | SHRINK; |
6905 | 9.80k | GROW; |
6906 | 9.80k | } |
6907 | 8.40k | } |
6908 | | |
6909 | | /** |
6910 | | * Parse markup declarations. Always consumes '<!' or '<?'. |
6911 | | * |
6912 | | * @deprecated Internal function, don't use. |
6913 | | * |
6914 | | * [29] markupdecl ::= elementdecl | AttlistDecl | EntityDecl | |
6915 | | * NotationDecl | PI | Comment |
6916 | | * |
6917 | | * [ VC: Proper Declaration/PE Nesting ] |
6918 | | * Parameter-entity replacement text must be properly nested with |
6919 | | * markup declarations. That is to say, if either the first character |
6920 | | * or the last character of a markup declaration (markupdecl above) is |
6921 | | * contained in the replacement text for a parameter-entity reference, |
6922 | | * both must be contained in the same replacement text. |
6923 | | * |
6924 | | * [ WFC: PEs in Internal Subset ] |
6925 | | * In the internal DTD subset, parameter-entity references can occur |
6926 | | * only where markup declarations can occur, not within markup declarations. |
6927 | | * (This does not apply to references that occur in external parameter |
6928 | | * entities or to the external subset.) |
6929 | | * |
6930 | | * @param ctxt an XML parser context |
6931 | | */ |
6932 | | void |
6933 | 390k | xmlParseMarkupDecl(xmlParserCtxt *ctxt) { |
6934 | 390k | GROW; |
6935 | 390k | if (CUR == '<') { |
6936 | 390k | if (NXT(1) == '!') { |
6937 | 383k | switch (NXT(2)) { |
6938 | 250k | case 'E': |
6939 | 250k | if (NXT(3) == 'L') |
6940 | 65.2k | xmlParseElementDecl(ctxt); |
6941 | 185k | else if (NXT(3) == 'N') |
6942 | 184k | xmlParseEntityDecl(ctxt); |
6943 | 286 | else |
6944 | 286 | SKIP(2); |
6945 | 250k | break; |
6946 | 61.0k | case 'A': |
6947 | 61.0k | xmlParseAttributeListDecl(ctxt); |
6948 | 61.0k | break; |
6949 | 13.3k | case 'N': |
6950 | 13.3k | xmlParseNotationDecl(ctxt); |
6951 | 13.3k | break; |
6952 | 56.1k | case '-': |
6953 | 56.1k | xmlParseComment(ctxt); |
6954 | 56.1k | break; |
6955 | 2.60k | default: |
6956 | 2.60k | xmlFatalErr(ctxt, |
6957 | 2.60k | ctxt->inSubset == 2 ? |
6958 | 559 | XML_ERR_EXT_SUBSET_NOT_FINISHED : |
6959 | 2.60k | XML_ERR_INT_SUBSET_NOT_FINISHED, |
6960 | 2.60k | NULL); |
6961 | 2.60k | SKIP(2); |
6962 | 2.60k | break; |
6963 | 383k | } |
6964 | 383k | } else if (NXT(1) == '?') { |
6965 | 7.27k | xmlParsePI(ctxt); |
6966 | 7.27k | } |
6967 | 390k | } |
6968 | 390k | } |
6969 | | |
6970 | | /** |
6971 | | * Parse an XML declaration header for external entities |
6972 | | * |
6973 | | * @deprecated Internal function, don't use. |
6974 | | * |
6975 | | * [77] TextDecl ::= '<?xml' VersionInfo? EncodingDecl S? '?>' |
6976 | | * @param ctxt an XML parser context |
6977 | | */ |
6978 | | |
6979 | | void |
6980 | 15.3k | xmlParseTextDecl(xmlParserCtxt *ctxt) { |
6981 | 15.3k | xmlChar *version; |
6982 | | |
6983 | | /* |
6984 | | * We know that '<?xml' is here. |
6985 | | */ |
6986 | 15.3k | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
6987 | 15.0k | SKIP(5); |
6988 | 15.0k | } else { |
6989 | 258 | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_STARTED, NULL); |
6990 | 258 | return; |
6991 | 258 | } |
6992 | | |
6993 | 15.0k | if (SKIP_BLANKS == 0) { |
6994 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
6995 | 0 | "Space needed after '<?xml'\n"); |
6996 | 0 | } |
6997 | | |
6998 | | /* |
6999 | | * We may have the VersionInfo here. |
7000 | | */ |
7001 | 15.0k | version = xmlParseVersionInfo(ctxt); |
7002 | 15.0k | if (version == NULL) { |
7003 | 5.75k | version = xmlCharStrdup(XML_DEFAULT_VERSION); |
7004 | 5.75k | if (version == NULL) { |
7005 | 4 | xmlErrMemory(ctxt); |
7006 | 4 | return; |
7007 | 4 | } |
7008 | 9.31k | } else { |
7009 | 9.31k | if (SKIP_BLANKS == 0) { |
7010 | 967 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
7011 | 967 | "Space needed here\n"); |
7012 | 967 | } |
7013 | 9.31k | } |
7014 | 15.0k | ctxt->input->version = version; |
7015 | | |
7016 | | /* |
7017 | | * We must have the encoding declaration |
7018 | | */ |
7019 | 15.0k | xmlParseEncodingDecl(ctxt); |
7020 | | |
7021 | 15.0k | SKIP_BLANKS; |
7022 | 15.0k | if ((RAW == '?') && (NXT(1) == '>')) { |
7023 | 6.47k | SKIP(2); |
7024 | 8.59k | } else if (RAW == '>') { |
7025 | | /* Deprecated old WD ... */ |
7026 | 296 | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
7027 | 296 | NEXT; |
7028 | 8.29k | } else { |
7029 | 8.29k | int c; |
7030 | | |
7031 | 8.29k | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
7032 | 280k | while ((PARSER_STOPPED(ctxt) == 0) && ((c = CUR) != 0)) { |
7033 | 278k | NEXT; |
7034 | 278k | if (c == '>') |
7035 | 6.90k | break; |
7036 | 278k | } |
7037 | 8.29k | } |
7038 | 15.0k | } |
7039 | | |
7040 | | /** |
7041 | | * Parse Markup declarations from an external subset |
7042 | | * |
7043 | | * @deprecated Internal function, don't use. |
7044 | | * |
7045 | | * [30] extSubset ::= textDecl? extSubsetDecl |
7046 | | * |
7047 | | * [31] extSubsetDecl ::= (markupdecl | conditionalSect | |
7048 | | * PEReference | S) * |
7049 | | * @param ctxt an XML parser context |
7050 | | * @param publicId the public identifier |
7051 | | * @param systemId the system identifier (URL) |
7052 | | */ |
7053 | | void |
7054 | | xmlParseExternalSubset(xmlParserCtxt *ctxt, const xmlChar *publicId, |
7055 | 14.7k | const xmlChar *systemId) { |
7056 | 14.7k | int oldInputNr; |
7057 | | |
7058 | 14.7k | xmlCtxtInitializeLate(ctxt); |
7059 | | |
7060 | 14.7k | xmlDetectEncoding(ctxt); |
7061 | | |
7062 | 14.7k | if (CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) { |
7063 | 9.16k | xmlParseTextDecl(ctxt); |
7064 | 9.16k | } |
7065 | 14.7k | if (ctxt->myDoc == NULL) { |
7066 | 0 | ctxt->myDoc = xmlNewDoc(BAD_CAST "1.0"); |
7067 | 0 | if (ctxt->myDoc == NULL) { |
7068 | 0 | xmlErrMemory(ctxt); |
7069 | 0 | return; |
7070 | 0 | } |
7071 | 0 | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
7072 | 0 | } |
7073 | 14.7k | if ((ctxt->myDoc->intSubset == NULL) && |
7074 | 14.7k | (xmlCreateIntSubset(ctxt->myDoc, NULL, publicId, systemId) == NULL)) { |
7075 | 0 | xmlErrMemory(ctxt); |
7076 | 0 | } |
7077 | | |
7078 | 14.7k | ctxt->inSubset = 2; |
7079 | 14.7k | oldInputNr = ctxt->inputNr; |
7080 | | |
7081 | 14.7k | SKIP_BLANKS; |
7082 | 55.1k | while (!PARSER_STOPPED(ctxt)) { |
7083 | 52.4k | if (ctxt->input->cur >= ctxt->input->end) { |
7084 | 1.58k | if (ctxt->inputNr <= oldInputNr) { |
7085 | 989 | xmlParserCheckEOF(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED); |
7086 | 989 | break; |
7087 | 989 | } |
7088 | | |
7089 | 600 | xmlPopPE(ctxt); |
7090 | 50.8k | } else if ((RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
7091 | 4.89k | xmlParseConditionalSections(ctxt); |
7092 | 45.9k | } else if ((RAW == '<') && ((NXT(1) == '!') || (NXT(1) == '?'))) { |
7093 | 33.1k | xmlParseMarkupDecl(ctxt); |
7094 | 33.1k | } else if (RAW == '%') { |
7095 | 1.74k | xmlParsePERefInternal(ctxt, 1); |
7096 | 11.0k | } else { |
7097 | 11.0k | xmlFatalErr(ctxt, XML_ERR_EXT_SUBSET_NOT_FINISHED, NULL); |
7098 | | |
7099 | 11.4k | while (ctxt->inputNr > oldInputNr) |
7100 | 348 | xmlPopPE(ctxt); |
7101 | 11.0k | break; |
7102 | 11.0k | } |
7103 | 40.3k | SKIP_BLANKS; |
7104 | 40.3k | SHRINK; |
7105 | 40.3k | GROW; |
7106 | 40.3k | } |
7107 | 14.7k | } |
7108 | | |
7109 | | /** |
7110 | | * Parse and handle entity references in content, depending on the SAX |
7111 | | * interface, this may end-up in a call to character() if this is a |
7112 | | * CharRef, a predefined entity, if there is no reference() callback. |
7113 | | * or if the parser was asked to switch to that mode. |
7114 | | * |
7115 | | * @deprecated Internal function, don't use. |
7116 | | * |
7117 | | * Always consumes '&'. |
7118 | | * |
7119 | | * [67] Reference ::= EntityRef | CharRef |
7120 | | * @param ctxt an XML parser context |
7121 | | */ |
7122 | | void |
7123 | 530k | xmlParseReference(xmlParserCtxt *ctxt) { |
7124 | 530k | xmlEntityPtr ent = NULL; |
7125 | 530k | const xmlChar *name; |
7126 | 530k | xmlChar *val; |
7127 | | |
7128 | 530k | if (RAW != '&') |
7129 | 0 | return; |
7130 | | |
7131 | | /* |
7132 | | * Simple case of a CharRef |
7133 | | */ |
7134 | 530k | if (NXT(1) == '#') { |
7135 | 43.5k | int i = 0; |
7136 | 43.5k | xmlChar out[16]; |
7137 | 43.5k | int value = xmlParseCharRef(ctxt); |
7138 | | |
7139 | 43.5k | if (value == 0) |
7140 | 16.7k | return; |
7141 | | |
7142 | | /* |
7143 | | * Just encode the value in UTF-8 |
7144 | | */ |
7145 | 26.7k | COPY_BUF(out, i, value); |
7146 | 26.7k | out[i] = 0; |
7147 | 26.7k | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL) && |
7148 | 26.7k | (!ctxt->disableSAX)) |
7149 | 9.00k | ctxt->sax->characters(ctxt->userData, out, i); |
7150 | 26.7k | return; |
7151 | 43.5k | } |
7152 | | |
7153 | | /* |
7154 | | * We are seeing an entity reference |
7155 | | */ |
7156 | 487k | name = xmlParseEntityRefInternal(ctxt); |
7157 | 487k | if (name == NULL) |
7158 | 313k | return; |
7159 | 173k | ent = xmlLookupGeneralEntity(ctxt, name, /* isAttr */ 0); |
7160 | 173k | if (ent == NULL) { |
7161 | | /* |
7162 | | * Create a reference for undeclared entities. |
7163 | | */ |
7164 | 56.6k | if ((ctxt->replaceEntities == 0) && |
7165 | 56.6k | (ctxt->sax != NULL) && |
7166 | 56.6k | (ctxt->disableSAX == 0) && |
7167 | 56.6k | (ctxt->sax->reference != NULL)) { |
7168 | 3.75k | ctxt->sax->reference(ctxt->userData, name); |
7169 | 3.75k | } |
7170 | 56.6k | return; |
7171 | 56.6k | } |
7172 | 116k | if (!ctxt->wellFormed) |
7173 | 68.5k | return; |
7174 | | |
7175 | | /* special case of predefined entities */ |
7176 | 48.1k | if ((ent->name == NULL) || |
7177 | 48.1k | (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { |
7178 | 35.0k | val = ent->content; |
7179 | 35.0k | if (val == NULL) return; |
7180 | | /* |
7181 | | * inline the entity. |
7182 | | */ |
7183 | 35.0k | if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL) && |
7184 | 35.0k | (!ctxt->disableSAX)) |
7185 | 35.0k | ctxt->sax->characters(ctxt->userData, val, xmlStrlen(val)); |
7186 | 35.0k | return; |
7187 | 35.0k | } |
7188 | | |
7189 | | /* |
7190 | | * Some users try to parse entities on their own and used to set |
7191 | | * the renamed "checked" member. Fix the flags to cover this |
7192 | | * case. |
7193 | | */ |
7194 | 13.0k | if (((ent->flags & XML_ENT_PARSED) == 0) && (ent->children != NULL)) |
7195 | 0 | ent->flags |= XML_ENT_PARSED; |
7196 | | |
7197 | | /* |
7198 | | * The first reference to the entity trigger a parsing phase |
7199 | | * where the ent->children is filled with the result from |
7200 | | * the parsing. |
7201 | | * Note: external parsed entities will not be loaded, it is not |
7202 | | * required for a non-validating parser, unless the parsing option |
7203 | | * of validating, or substituting entities were given. Doing so is |
7204 | | * far more secure as the parser will only process data coming from |
7205 | | * the document entity by default. |
7206 | | * |
7207 | | * FIXME: This doesn't work correctly since entities can be |
7208 | | * expanded with different namespace declarations in scope. |
7209 | | * For example: |
7210 | | * |
7211 | | * <!DOCTYPE doc [ |
7212 | | * <!ENTITY ent "<ns:elem/>"> |
7213 | | * ]> |
7214 | | * <doc> |
7215 | | * <decl1 xmlns:ns="urn:ns1"> |
7216 | | * &ent; |
7217 | | * </decl1> |
7218 | | * <decl2 xmlns:ns="urn:ns2"> |
7219 | | * &ent; |
7220 | | * </decl2> |
7221 | | * </doc> |
7222 | | * |
7223 | | * Proposed fix: |
7224 | | * |
7225 | | * - Ignore current namespace declarations when parsing the |
7226 | | * entity. If a prefix can't be resolved, don't report an error |
7227 | | * but mark it as unresolved. |
7228 | | * - Try to resolve these prefixes when expanding the entity. |
7229 | | * This will require a specialized version of xmlStaticCopyNode |
7230 | | * which can also make use of the namespace hash table to avoid |
7231 | | * quadratic behavior. |
7232 | | * |
7233 | | * Alternatively, we could simply reparse the entity on each |
7234 | | * expansion like we already do with custom SAX callbacks. |
7235 | | * External entity content should be cached in this case. |
7236 | | */ |
7237 | 13.0k | if ((ent->etype == XML_INTERNAL_GENERAL_ENTITY) || |
7238 | 13.0k | (((ctxt->options & XML_PARSE_NO_XXE) == 0) && |
7239 | 9.91k | ((ctxt->replaceEntities) || |
7240 | 13.0k | (ctxt->validate)))) { |
7241 | 13.0k | if ((ent->flags & XML_ENT_PARSED) == 0) { |
7242 | 11.2k | xmlCtxtParseEntity(ctxt, ent); |
7243 | 11.2k | } else if (ent->children == NULL) { |
7244 | | /* |
7245 | | * Probably running in SAX mode and the callbacks don't |
7246 | | * build the entity content. Parse the entity again. |
7247 | | * |
7248 | | * This will also be triggered in normal tree builder mode |
7249 | | * if an entity happens to be empty, causing unnecessary |
7250 | | * reloads. It's hard to come up with a reliable check in |
7251 | | * which mode we're running. |
7252 | | */ |
7253 | 335 | xmlCtxtParseEntity(ctxt, ent); |
7254 | 335 | } |
7255 | 13.0k | } |
7256 | | |
7257 | | /* |
7258 | | * We also check for amplification if entities aren't substituted. |
7259 | | * They might be expanded later. |
7260 | | */ |
7261 | 13.0k | if (xmlParserEntityCheck(ctxt, ent->expandedSize)) |
7262 | 2 | return; |
7263 | | |
7264 | 13.0k | if ((ctxt->sax == NULL) || (ctxt->disableSAX)) |
7265 | 7.72k | return; |
7266 | | |
7267 | 5.34k | if (ctxt->replaceEntities == 0) { |
7268 | | /* |
7269 | | * Create a reference |
7270 | | */ |
7271 | 1 | if (ctxt->sax->reference != NULL) |
7272 | 1 | ctxt->sax->reference(ctxt->userData, ent->name); |
7273 | 5.34k | } else if ((ent->children != NULL) && (ctxt->node != NULL)) { |
7274 | 3.61k | xmlNodePtr copy, cur; |
7275 | | |
7276 | | /* |
7277 | | * Seems we are generating the DOM content, copy the tree |
7278 | | */ |
7279 | 3.61k | cur = ent->children; |
7280 | | |
7281 | | /* |
7282 | | * Handle first text node with SAX to coalesce text efficiently |
7283 | | */ |
7284 | 3.61k | if ((cur->type == XML_TEXT_NODE) || |
7285 | 3.61k | (cur->type == XML_CDATA_SECTION_NODE)) { |
7286 | 2.04k | int len = xmlStrlen(cur->content); |
7287 | | |
7288 | 2.04k | if ((cur->type == XML_TEXT_NODE) || |
7289 | 2.04k | (ctxt->options & XML_PARSE_NOCDATA)) { |
7290 | 2.04k | if (ctxt->sax->characters != NULL) |
7291 | 2.04k | ctxt->sax->characters(ctxt, cur->content, len); |
7292 | 2.04k | } else { |
7293 | 0 | if (ctxt->sax->cdataBlock != NULL) |
7294 | 0 | ctxt->sax->cdataBlock(ctxt, cur->content, len); |
7295 | 0 | } |
7296 | | |
7297 | 2.04k | cur = cur->next; |
7298 | 2.04k | } |
7299 | | |
7300 | 11.3k | while (cur != NULL) { |
7301 | 10.4k | xmlNodePtr last; |
7302 | | |
7303 | | /* |
7304 | | * Handle last text node with SAX to coalesce text efficiently |
7305 | | */ |
7306 | 10.4k | if ((cur->next == NULL) && |
7307 | 10.4k | ((cur->type == XML_TEXT_NODE) || |
7308 | 2.96k | (cur->type == XML_CDATA_SECTION_NODE))) { |
7309 | 2.71k | int len = xmlStrlen(cur->content); |
7310 | | |
7311 | 2.71k | if ((cur->type == XML_TEXT_NODE) || |
7312 | 2.71k | (ctxt->options & XML_PARSE_NOCDATA)) { |
7313 | 2.71k | if (ctxt->sax->characters != NULL) |
7314 | 2.71k | ctxt->sax->characters(ctxt, cur->content, len); |
7315 | 2.71k | } else { |
7316 | 0 | if (ctxt->sax->cdataBlock != NULL) |
7317 | 0 | ctxt->sax->cdataBlock(ctxt, cur->content, len); |
7318 | 0 | } |
7319 | | |
7320 | 2.71k | break; |
7321 | 2.71k | } |
7322 | | |
7323 | | /* |
7324 | | * Reset coalesce buffer stats only for non-text nodes. |
7325 | | */ |
7326 | 7.75k | ctxt->nodemem = 0; |
7327 | 7.75k | ctxt->nodelen = 0; |
7328 | | |
7329 | 7.75k | copy = xmlDocCopyNode(cur, ctxt->myDoc, 1); |
7330 | | |
7331 | 7.75k | if (copy == NULL) { |
7332 | 23 | xmlErrMemory(ctxt); |
7333 | 23 | break; |
7334 | 23 | } |
7335 | | |
7336 | 7.73k | if (ctxt->parseMode == XML_PARSE_READER) { |
7337 | | /* Needed for reader */ |
7338 | 0 | copy->extra = cur->extra; |
7339 | | /* Maybe needed for reader */ |
7340 | 0 | copy->_private = cur->_private; |
7341 | 0 | } |
7342 | | |
7343 | 7.73k | copy->parent = ctxt->node; |
7344 | 7.73k | last = ctxt->node->last; |
7345 | 7.73k | if (last == NULL) { |
7346 | 430 | ctxt->node->children = copy; |
7347 | 7.30k | } else { |
7348 | 7.30k | last->next = copy; |
7349 | 7.30k | copy->prev = last; |
7350 | 7.30k | } |
7351 | 7.73k | ctxt->node->last = copy; |
7352 | | |
7353 | 7.73k | cur = cur->next; |
7354 | 7.73k | } |
7355 | 3.61k | } |
7356 | 5.34k | } |
7357 | | |
7358 | | static void |
7359 | 244k | xmlHandleUndeclaredEntity(xmlParserCtxtPtr ctxt, const xmlChar *name) { |
7360 | | /* |
7361 | | * [ WFC: Entity Declared ] |
7362 | | * In a document without any DTD, a document with only an |
7363 | | * internal DTD subset which contains no parameter entity |
7364 | | * references, or a document with "standalone='yes'", the |
7365 | | * Name given in the entity reference must match that in an |
7366 | | * entity declaration, except that well-formed documents |
7367 | | * need not declare any of the following entities: amp, lt, |
7368 | | * gt, apos, quot. |
7369 | | * The declaration of a parameter entity must precede any |
7370 | | * reference to it. |
7371 | | * Similarly, the declaration of a general entity must |
7372 | | * precede any reference to it which appears in a default |
7373 | | * value in an attribute-list declaration. Note that if |
7374 | | * entities are declared in the external subset or in |
7375 | | * external parameter entities, a non-validating processor |
7376 | | * is not obligated to read and process their declarations; |
7377 | | * for such documents, the rule that an entity must be |
7378 | | * declared is a well-formedness constraint only if |
7379 | | * standalone='yes'. |
7380 | | */ |
7381 | 244k | if ((ctxt->standalone == 1) || |
7382 | 244k | ((ctxt->hasExternalSubset == 0) && |
7383 | 244k | (ctxt->hasPErefs == 0))) { |
7384 | 225k | xmlFatalErrMsgStr(ctxt, XML_ERR_UNDECLARED_ENTITY, |
7385 | 225k | "Entity '%s' not defined\n", name); |
7386 | | #ifdef LIBXML_VALID_ENABLED |
7387 | | } else if (ctxt->validate) { |
7388 | | /* |
7389 | | * [ VC: Entity Declared ] |
7390 | | * In a document with an external subset or external |
7391 | | * parameter entities with "standalone='no'", ... |
7392 | | * ... The declaration of a parameter entity must |
7393 | | * precede any reference to it... |
7394 | | */ |
7395 | | xmlValidityError(ctxt, XML_ERR_UNDECLARED_ENTITY, |
7396 | | "Entity '%s' not defined\n", name, NULL); |
7397 | | #endif |
7398 | 225k | } else if ((ctxt->loadsubset & ~XML_SKIP_IDS) || |
7399 | 19.8k | ((ctxt->replaceEntities) && |
7400 | 19.7k | ((ctxt->options & XML_PARSE_NO_XXE) == 0))) { |
7401 | | /* |
7402 | | * Also raise a non-fatal error |
7403 | | * |
7404 | | * - if the external subset is loaded and all entity declarations |
7405 | | * should be available, or |
7406 | | * - entity substition was requested without restricting |
7407 | | * external entity access. |
7408 | | */ |
7409 | 19.7k | xmlErrMsgStr(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7410 | 19.7k | "Entity '%s' not defined\n", name); |
7411 | 19.7k | } else { |
7412 | 67 | xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7413 | 67 | "Entity '%s' not defined\n", name, NULL); |
7414 | 67 | } |
7415 | | |
7416 | 244k | ctxt->valid = 0; |
7417 | 244k | } |
7418 | | |
7419 | | static xmlEntityPtr |
7420 | 931k | xmlLookupGeneralEntity(xmlParserCtxtPtr ctxt, const xmlChar *name, int inAttr) { |
7421 | 931k | xmlEntityPtr ent = NULL; |
7422 | | |
7423 | | /* |
7424 | | * Predefined entities override any extra definition |
7425 | | */ |
7426 | 931k | if ((ctxt->options & XML_PARSE_OLDSAX) == 0) { |
7427 | 931k | ent = xmlGetPredefinedEntity(name); |
7428 | 931k | if (ent != NULL) |
7429 | 123k | return(ent); |
7430 | 931k | } |
7431 | | |
7432 | | /* |
7433 | | * Ask first SAX for entity resolution, otherwise try the |
7434 | | * entities which may have stored in the parser context. |
7435 | | */ |
7436 | 807k | if (ctxt->sax != NULL) { |
7437 | 807k | if (ctxt->sax->getEntity != NULL) |
7438 | 807k | ent = ctxt->sax->getEntity(ctxt->userData, name); |
7439 | 807k | if ((ctxt->wellFormed == 1 ) && (ent == NULL) && |
7440 | 807k | (ctxt->options & XML_PARSE_OLDSAX)) |
7441 | 0 | ent = xmlGetPredefinedEntity(name); |
7442 | 807k | if ((ctxt->wellFormed == 1 ) && (ent == NULL) && |
7443 | 807k | (ctxt->userData==ctxt)) { |
7444 | 2.20k | ent = xmlSAX2GetEntity(ctxt, name); |
7445 | 2.20k | } |
7446 | 807k | } |
7447 | | |
7448 | 807k | if (ent == NULL) { |
7449 | 225k | xmlHandleUndeclaredEntity(ctxt, name); |
7450 | 225k | } |
7451 | | |
7452 | | /* |
7453 | | * [ WFC: Parsed Entity ] |
7454 | | * An entity reference must not contain the name of an |
7455 | | * unparsed entity |
7456 | | */ |
7457 | 582k | else if (ent->etype == XML_EXTERNAL_GENERAL_UNPARSED_ENTITY) { |
7458 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_UNPARSED_ENTITY, |
7459 | 0 | "Entity reference to unparsed entity %s\n", name); |
7460 | 0 | ent = NULL; |
7461 | 0 | } |
7462 | | |
7463 | | /* |
7464 | | * [ WFC: No External Entity References ] |
7465 | | * Attribute values cannot contain direct or indirect |
7466 | | * entity references to external entities. |
7467 | | */ |
7468 | 582k | else if (ent->etype == XML_EXTERNAL_GENERAL_PARSED_ENTITY) { |
7469 | 14.2k | if (inAttr) { |
7470 | 964 | xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_IS_EXTERNAL, |
7471 | 964 | "Attribute references external entity '%s'\n", name); |
7472 | 964 | ent = NULL; |
7473 | 964 | } |
7474 | 14.2k | } |
7475 | | |
7476 | 807k | return(ent); |
7477 | 931k | } |
7478 | | |
7479 | | /** |
7480 | | * Parse an entity reference. Always consumes '&'. |
7481 | | * |
7482 | | * [68] EntityRef ::= '&' Name ';' |
7483 | | * |
7484 | | * @param ctxt an XML parser context |
7485 | | * @returns the name, or NULL in case of error. |
7486 | | */ |
7487 | | static const xmlChar * |
7488 | 1.07M | xmlParseEntityRefInternal(xmlParserCtxtPtr ctxt) { |
7489 | 1.07M | const xmlChar *name; |
7490 | | |
7491 | 1.07M | GROW; |
7492 | | |
7493 | 1.07M | if (RAW != '&') |
7494 | 0 | return(NULL); |
7495 | 1.07M | NEXT; |
7496 | 1.07M | name = xmlParseName(ctxt); |
7497 | 1.07M | if (name == NULL) { |
7498 | 447k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7499 | 447k | "xmlParseEntityRef: no name\n"); |
7500 | 447k | return(NULL); |
7501 | 447k | } |
7502 | 626k | if (RAW != ';') { |
7503 | 113k | xmlFatalErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING, NULL); |
7504 | 113k | return(NULL); |
7505 | 113k | } |
7506 | 512k | NEXT; |
7507 | | |
7508 | 512k | return(name); |
7509 | 626k | } |
7510 | | |
7511 | | /** |
7512 | | * @deprecated Internal function, don't use. |
7513 | | * |
7514 | | * @param ctxt an XML parser context |
7515 | | * @returns the xmlEntity if found, or NULL otherwise. |
7516 | | */ |
7517 | | xmlEntity * |
7518 | 0 | xmlParseEntityRef(xmlParserCtxt *ctxt) { |
7519 | 0 | const xmlChar *name; |
7520 | |
|
7521 | 0 | if (ctxt == NULL) |
7522 | 0 | return(NULL); |
7523 | | |
7524 | 0 | name = xmlParseEntityRefInternal(ctxt); |
7525 | 0 | if (name == NULL) |
7526 | 0 | return(NULL); |
7527 | | |
7528 | 0 | return(xmlLookupGeneralEntity(ctxt, name, /* inAttr */ 0)); |
7529 | 0 | } |
7530 | | |
7531 | | /** |
7532 | | * Parse ENTITY references declarations, but this version parses it from |
7533 | | * a string value. |
7534 | | * |
7535 | | * [68] EntityRef ::= '&' Name ';' |
7536 | | * |
7537 | | * [ WFC: Entity Declared ] |
7538 | | * In a document without any DTD, a document with only an internal DTD |
7539 | | * subset which contains no parameter entity references, or a document |
7540 | | * with "standalone='yes'", the Name given in the entity reference |
7541 | | * must match that in an entity declaration, except that well-formed |
7542 | | * documents need not declare any of the following entities: amp, lt, |
7543 | | * gt, apos, quot. The declaration of a parameter entity must precede |
7544 | | * any reference to it. Similarly, the declaration of a general entity |
7545 | | * must precede any reference to it which appears in a default value in an |
7546 | | * attribute-list declaration. Note that if entities are declared in the |
7547 | | * external subset or in external parameter entities, a non-validating |
7548 | | * processor is not obligated to read and process their declarations; |
7549 | | * for such documents, the rule that an entity must be declared is a |
7550 | | * well-formedness constraint only if standalone='yes'. |
7551 | | * |
7552 | | * [ WFC: Parsed Entity ] |
7553 | | * An entity reference must not contain the name of an unparsed entity |
7554 | | * |
7555 | | * @param ctxt an XML parser context |
7556 | | * @param str a pointer to an index in the string |
7557 | | * @returns the xmlEntity if found, or NULL otherwise. The str pointer |
7558 | | * is updated to the current location in the string. |
7559 | | */ |
7560 | | static xmlChar * |
7561 | 420k | xmlParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar ** str) { |
7562 | 420k | xmlChar *name; |
7563 | 420k | const xmlChar *ptr; |
7564 | 420k | xmlChar cur; |
7565 | | |
7566 | 420k | if ((str == NULL) || (*str == NULL)) |
7567 | 0 | return(NULL); |
7568 | 420k | ptr = *str; |
7569 | 420k | cur = *ptr; |
7570 | 420k | if (cur != '&') |
7571 | 0 | return(NULL); |
7572 | | |
7573 | 420k | ptr++; |
7574 | 420k | name = xmlParseStringName(ctxt, &ptr); |
7575 | 420k | if (name == NULL) { |
7576 | 811 | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7577 | 811 | "xmlParseStringEntityRef: no name\n"); |
7578 | 811 | *str = ptr; |
7579 | 811 | return(NULL); |
7580 | 811 | } |
7581 | 419k | if (*ptr != ';') { |
7582 | 321 | xmlFatalErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING, NULL); |
7583 | 321 | xmlFree(name); |
7584 | 321 | *str = ptr; |
7585 | 321 | return(NULL); |
7586 | 321 | } |
7587 | 418k | ptr++; |
7588 | | |
7589 | 418k | *str = ptr; |
7590 | 418k | return(name); |
7591 | 419k | } |
7592 | | |
7593 | | /** |
7594 | | * Parse a parameter entity reference. Always consumes '%'. |
7595 | | * |
7596 | | * The entity content is handled directly by pushing it's content as |
7597 | | * a new input stream. |
7598 | | * |
7599 | | * [69] PEReference ::= '%' Name ';' |
7600 | | * |
7601 | | * [ WFC: No Recursion ] |
7602 | | * A parsed entity must not contain a recursive |
7603 | | * reference to itself, either directly or indirectly. |
7604 | | * |
7605 | | * [ WFC: Entity Declared ] |
7606 | | * In a document without any DTD, a document with only an internal DTD |
7607 | | * subset which contains no parameter entity references, or a document |
7608 | | * with "standalone='yes'", ... ... The declaration of a parameter |
7609 | | * entity must precede any reference to it... |
7610 | | * |
7611 | | * [ VC: Entity Declared ] |
7612 | | * In a document with an external subset or external parameter entities |
7613 | | * with "standalone='no'", ... ... The declaration of a parameter entity |
7614 | | * must precede any reference to it... |
7615 | | * |
7616 | | * [ WFC: In DTD ] |
7617 | | * Parameter-entity references may only appear in the DTD. |
7618 | | * NOTE: misleading but this is handled. |
7619 | | * |
7620 | | * @param ctxt an XML parser context |
7621 | | * @param markupDecl whether the PERef starts a markup declaration |
7622 | | */ |
7623 | | static void |
7624 | 64.9k | xmlParsePERefInternal(xmlParserCtxt *ctxt, int markupDecl) { |
7625 | 64.9k | const xmlChar *name; |
7626 | 64.9k | xmlEntityPtr entity = NULL; |
7627 | 64.9k | xmlParserInputPtr input; |
7628 | | |
7629 | 64.9k | if (RAW != '%') |
7630 | 0 | return; |
7631 | 64.9k | NEXT; |
7632 | 64.9k | name = xmlParseName(ctxt); |
7633 | 64.9k | if (name == NULL) { |
7634 | 6.06k | xmlFatalErrMsg(ctxt, XML_ERR_PEREF_NO_NAME, "PEReference: no name\n"); |
7635 | 6.06k | return; |
7636 | 6.06k | } |
7637 | 58.9k | if (RAW != ';') { |
7638 | 1.25k | xmlFatalErr(ctxt, XML_ERR_PEREF_SEMICOL_MISSING, NULL); |
7639 | 1.25k | return; |
7640 | 1.25k | } |
7641 | | |
7642 | 57.6k | NEXT; |
7643 | | |
7644 | | /* Must be set before xmlHandleUndeclaredEntity */ |
7645 | 57.6k | ctxt->hasPErefs = 1; |
7646 | | |
7647 | | /* |
7648 | | * Request the entity from SAX |
7649 | | */ |
7650 | 57.6k | if ((ctxt->sax != NULL) && |
7651 | 57.6k | (ctxt->sax->getParameterEntity != NULL)) |
7652 | 57.6k | entity = ctxt->sax->getParameterEntity(ctxt->userData, name); |
7653 | | |
7654 | 57.6k | if (entity == NULL) { |
7655 | 17.5k | xmlHandleUndeclaredEntity(ctxt, name); |
7656 | 40.1k | } else { |
7657 | | /* |
7658 | | * Internal checking in case the entity quest barfed |
7659 | | */ |
7660 | 40.1k | if ((entity->etype != XML_INTERNAL_PARAMETER_ENTITY) && |
7661 | 40.1k | (entity->etype != XML_EXTERNAL_PARAMETER_ENTITY)) { |
7662 | 0 | xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7663 | 0 | "Internal: %%%s; is not a parameter entity\n", |
7664 | 0 | name, NULL); |
7665 | 40.1k | } else { |
7666 | 40.1k | if ((entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) && |
7667 | 40.1k | ((ctxt->options & XML_PARSE_NO_XXE) || |
7668 | 31.6k | (((ctxt->loadsubset & ~XML_SKIP_IDS) == 0) && |
7669 | 31.6k | (ctxt->replaceEntities == 0) && |
7670 | 31.6k | (ctxt->validate == 0)))) |
7671 | 0 | return; |
7672 | | |
7673 | 40.1k | if (entity->flags & XML_ENT_EXPANDING) { |
7674 | 758 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
7675 | 758 | xmlHaltParser(ctxt); |
7676 | 758 | return; |
7677 | 758 | } |
7678 | | |
7679 | 39.3k | input = xmlNewEntityInputStream(ctxt, entity); |
7680 | 39.3k | if (xmlCtxtPushInput(ctxt, input) < 0) { |
7681 | 9.02k | xmlFreeInputStream(input); |
7682 | 9.02k | return; |
7683 | 9.02k | } |
7684 | | |
7685 | 30.3k | entity->flags |= XML_ENT_EXPANDING; |
7686 | | |
7687 | 30.3k | if (markupDecl) |
7688 | 21.6k | input->flags |= XML_INPUT_MARKUP_DECL; |
7689 | | |
7690 | 30.3k | GROW; |
7691 | | |
7692 | 30.3k | if (entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) { |
7693 | 21.8k | xmlDetectEncoding(ctxt); |
7694 | | |
7695 | 21.8k | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && |
7696 | 21.8k | (IS_BLANK_CH(NXT(5)))) { |
7697 | 2.64k | xmlParseTextDecl(ctxt); |
7698 | 2.64k | } |
7699 | 21.8k | } |
7700 | 30.3k | } |
7701 | 40.1k | } |
7702 | 57.6k | } |
7703 | | |
7704 | | /** |
7705 | | * Parse a parameter entity reference. |
7706 | | * |
7707 | | * @deprecated Internal function, don't use. |
7708 | | * |
7709 | | * @param ctxt an XML parser context |
7710 | | */ |
7711 | | void |
7712 | 0 | xmlParsePEReference(xmlParserCtxt *ctxt) { |
7713 | 0 | xmlParsePERefInternal(ctxt, 0); |
7714 | 0 | } |
7715 | | |
7716 | | /** |
7717 | | * Load the content of an entity. |
7718 | | * |
7719 | | * @param ctxt an XML parser context |
7720 | | * @param entity an unloaded system entity |
7721 | | * @returns 0 in case of success and -1 in case of failure |
7722 | | */ |
7723 | | static int |
7724 | 6.93k | xmlLoadEntityContent(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { |
7725 | 6.93k | xmlParserInputPtr oldinput, input = NULL; |
7726 | 6.93k | xmlParserInputPtr *oldinputTab; |
7727 | 6.93k | xmlChar *oldencoding; |
7728 | 6.93k | xmlChar *content = NULL; |
7729 | 6.93k | xmlResourceType rtype; |
7730 | 6.93k | size_t length, i; |
7731 | 6.93k | int oldinputNr, oldinputMax; |
7732 | 6.93k | int ret = -1; |
7733 | 6.93k | int res; |
7734 | | |
7735 | 6.93k | if ((ctxt == NULL) || (entity == NULL) || |
7736 | 6.93k | ((entity->etype != XML_EXTERNAL_PARAMETER_ENTITY) && |
7737 | 6.93k | (entity->etype != XML_EXTERNAL_GENERAL_PARSED_ENTITY)) || |
7738 | 6.93k | (entity->content != NULL)) { |
7739 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, |
7740 | 0 | "xmlLoadEntityContent parameter error"); |
7741 | 0 | return(-1); |
7742 | 0 | } |
7743 | | |
7744 | 6.93k | if (entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) |
7745 | 6.93k | rtype = XML_RESOURCE_PARAMETER_ENTITY; |
7746 | 0 | else |
7747 | 0 | rtype = XML_RESOURCE_GENERAL_ENTITY; |
7748 | | |
7749 | 6.93k | input = xmlLoadResource(ctxt, (char *) entity->URI, |
7750 | 6.93k | (char *) entity->ExternalID, rtype); |
7751 | 6.93k | if (input == NULL) |
7752 | 761 | return(-1); |
7753 | | |
7754 | 6.17k | oldinput = ctxt->input; |
7755 | 6.17k | oldinputNr = ctxt->inputNr; |
7756 | 6.17k | oldinputMax = ctxt->inputMax; |
7757 | 6.17k | oldinputTab = ctxt->inputTab; |
7758 | 6.17k | oldencoding = ctxt->encoding; |
7759 | | |
7760 | 6.17k | ctxt->input = NULL; |
7761 | 6.17k | ctxt->inputNr = 0; |
7762 | 6.17k | ctxt->inputMax = 1; |
7763 | 6.17k | ctxt->encoding = NULL; |
7764 | 6.17k | ctxt->inputTab = xmlMalloc(sizeof(xmlParserInputPtr)); |
7765 | 6.17k | if (ctxt->inputTab == NULL) { |
7766 | 1 | xmlErrMemory(ctxt); |
7767 | 1 | xmlFreeInputStream(input); |
7768 | 1 | goto error; |
7769 | 1 | } |
7770 | | |
7771 | 6.17k | xmlBufResetInput(input->buf->buffer, input); |
7772 | | |
7773 | 6.17k | if (xmlCtxtPushInput(ctxt, input) < 0) { |
7774 | 0 | xmlFreeInputStream(input); |
7775 | 0 | goto error; |
7776 | 0 | } |
7777 | | |
7778 | 6.17k | xmlDetectEncoding(ctxt); |
7779 | | |
7780 | | /* |
7781 | | * Parse a possible text declaration first |
7782 | | */ |
7783 | 6.17k | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
7784 | 976 | xmlParseTextDecl(ctxt); |
7785 | | /* |
7786 | | * An XML-1.0 document can't reference an entity not XML-1.0 |
7787 | | */ |
7788 | 976 | if ((xmlStrEqual(ctxt->version, BAD_CAST "1.0")) && |
7789 | 976 | (!xmlStrEqual(ctxt->input->version, BAD_CAST "1.0"))) { |
7790 | 309 | xmlFatalErrMsg(ctxt, XML_ERR_VERSION_MISMATCH, |
7791 | 309 | "Version mismatch between document and entity\n"); |
7792 | 309 | } |
7793 | 976 | } |
7794 | | |
7795 | 6.17k | length = input->cur - input->base; |
7796 | 6.17k | xmlBufShrink(input->buf->buffer, length); |
7797 | 6.17k | xmlSaturatedAdd(&ctxt->sizeentities, length); |
7798 | | |
7799 | 8.87k | while ((res = xmlParserInputBufferGrow(input->buf, 4096)) > 0) |
7800 | 2.70k | ; |
7801 | | |
7802 | 6.17k | xmlBufResetInput(input->buf->buffer, input); |
7803 | | |
7804 | 6.17k | if (res < 0) { |
7805 | 839 | xmlCtxtErrIO(ctxt, input->buf->error, NULL); |
7806 | 839 | goto error; |
7807 | 839 | } |
7808 | | |
7809 | 5.33k | length = xmlBufUse(input->buf->buffer); |
7810 | 5.33k | if (length > INT_MAX) { |
7811 | 0 | xmlErrMemory(ctxt); |
7812 | 0 | goto error; |
7813 | 0 | } |
7814 | | |
7815 | 5.33k | content = xmlStrndup(xmlBufContent(input->buf->buffer), length); |
7816 | 5.33k | if (content == NULL) { |
7817 | 3 | xmlErrMemory(ctxt); |
7818 | 3 | goto error; |
7819 | 3 | } |
7820 | | |
7821 | 3.94M | for (i = 0; i < length; ) { |
7822 | 3.94M | int clen = length - i; |
7823 | 3.94M | int c = xmlGetUTF8Char(content + i, &clen); |
7824 | | |
7825 | 3.94M | if ((c < 0) || (!IS_CHAR(c))) { |
7826 | 4.79k | xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, |
7827 | 4.79k | "xmlLoadEntityContent: invalid char value %d\n", |
7828 | 4.79k | content[i]); |
7829 | 4.79k | goto error; |
7830 | 4.79k | } |
7831 | 3.94M | i += clen; |
7832 | 3.94M | } |
7833 | | |
7834 | 534 | xmlSaturatedAdd(&ctxt->sizeentities, length); |
7835 | 534 | entity->content = content; |
7836 | 534 | entity->length = length; |
7837 | 534 | content = NULL; |
7838 | 534 | ret = 0; |
7839 | | |
7840 | 6.17k | error: |
7841 | 12.3k | while (ctxt->inputNr > 0) |
7842 | 6.17k | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
7843 | 6.17k | xmlFree(ctxt->inputTab); |
7844 | 6.17k | xmlFree(ctxt->encoding); |
7845 | | |
7846 | 6.17k | ctxt->input = oldinput; |
7847 | 6.17k | ctxt->inputNr = oldinputNr; |
7848 | 6.17k | ctxt->inputMax = oldinputMax; |
7849 | 6.17k | ctxt->inputTab = oldinputTab; |
7850 | 6.17k | ctxt->encoding = oldencoding; |
7851 | | |
7852 | 6.17k | xmlFree(content); |
7853 | | |
7854 | 6.17k | return(ret); |
7855 | 534 | } |
7856 | | |
7857 | | /** |
7858 | | * Parse PEReference declarations |
7859 | | * |
7860 | | * [69] PEReference ::= '%' Name ';' |
7861 | | * |
7862 | | * [ WFC: No Recursion ] |
7863 | | * A parsed entity must not contain a recursive |
7864 | | * reference to itself, either directly or indirectly. |
7865 | | * |
7866 | | * [ WFC: Entity Declared ] |
7867 | | * In a document without any DTD, a document with only an internal DTD |
7868 | | * subset which contains no parameter entity references, or a document |
7869 | | * with "standalone='yes'", ... ... The declaration of a parameter |
7870 | | * entity must precede any reference to it... |
7871 | | * |
7872 | | * [ VC: Entity Declared ] |
7873 | | * In a document with an external subset or external parameter entities |
7874 | | * with "standalone='no'", ... ... The declaration of a parameter entity |
7875 | | * must precede any reference to it... |
7876 | | * |
7877 | | * [ WFC: In DTD ] |
7878 | | * Parameter-entity references may only appear in the DTD. |
7879 | | * NOTE: misleading but this is handled. |
7880 | | * |
7881 | | * @param ctxt an XML parser context |
7882 | | * @param str a pointer to an index in the string |
7883 | | * @returns the string of the entity content. |
7884 | | * str is updated to the current value of the index |
7885 | | */ |
7886 | | static xmlEntityPtr |
7887 | 25.5k | xmlParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str) { |
7888 | 25.5k | const xmlChar *ptr; |
7889 | 25.5k | xmlChar cur; |
7890 | 25.5k | xmlChar *name; |
7891 | 25.5k | xmlEntityPtr entity = NULL; |
7892 | | |
7893 | 25.5k | if ((str == NULL) || (*str == NULL)) return(NULL); |
7894 | 25.5k | ptr = *str; |
7895 | 25.5k | cur = *ptr; |
7896 | 25.5k | if (cur != '%') |
7897 | 0 | return(NULL); |
7898 | 25.5k | ptr++; |
7899 | 25.5k | name = xmlParseStringName(ctxt, &ptr); |
7900 | 25.5k | if (name == NULL) { |
7901 | 4.98k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7902 | 4.98k | "xmlParseStringPEReference: no name\n"); |
7903 | 4.98k | *str = ptr; |
7904 | 4.98k | return(NULL); |
7905 | 4.98k | } |
7906 | 20.5k | cur = *ptr; |
7907 | 20.5k | if (cur != ';') { |
7908 | 1.16k | xmlFatalErr(ctxt, XML_ERR_ENTITYREF_SEMICOL_MISSING, NULL); |
7909 | 1.16k | xmlFree(name); |
7910 | 1.16k | *str = ptr; |
7911 | 1.16k | return(NULL); |
7912 | 1.16k | } |
7913 | 19.3k | ptr++; |
7914 | | |
7915 | | /* Must be set before xmlHandleUndeclaredEntity */ |
7916 | 19.3k | ctxt->hasPErefs = 1; |
7917 | | |
7918 | | /* |
7919 | | * Request the entity from SAX |
7920 | | */ |
7921 | 19.3k | if ((ctxt->sax != NULL) && |
7922 | 19.3k | (ctxt->sax->getParameterEntity != NULL)) |
7923 | 19.3k | entity = ctxt->sax->getParameterEntity(ctxt->userData, name); |
7924 | | |
7925 | 19.3k | if (entity == NULL) { |
7926 | 2.21k | xmlHandleUndeclaredEntity(ctxt, name); |
7927 | 17.1k | } else { |
7928 | | /* |
7929 | | * Internal checking in case the entity quest barfed |
7930 | | */ |
7931 | 17.1k | if ((entity->etype != XML_INTERNAL_PARAMETER_ENTITY) && |
7932 | 17.1k | (entity->etype != XML_EXTERNAL_PARAMETER_ENTITY)) { |
7933 | 0 | xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, |
7934 | 0 | "%%%s; is not a parameter entity\n", |
7935 | 0 | name, NULL); |
7936 | 0 | } |
7937 | 17.1k | } |
7938 | | |
7939 | 19.3k | xmlFree(name); |
7940 | 19.3k | *str = ptr; |
7941 | 19.3k | return(entity); |
7942 | 20.5k | } |
7943 | | |
7944 | | /** |
7945 | | * Parse a DOCTYPE declaration |
7946 | | * |
7947 | | * @deprecated Internal function, don't use. |
7948 | | * |
7949 | | * [28] doctypedecl ::= '<!DOCTYPE' S Name (S ExternalID)? S? |
7950 | | * ('[' (markupdecl | PEReference | S)* ']' S?)? '>' |
7951 | | * |
7952 | | * [ VC: Root Element Type ] |
7953 | | * The Name in the document type declaration must match the element |
7954 | | * type of the root element. |
7955 | | * |
7956 | | * @param ctxt an XML parser context |
7957 | | */ |
7958 | | |
7959 | | void |
7960 | 148k | xmlParseDocTypeDecl(xmlParserCtxt *ctxt) { |
7961 | 148k | const xmlChar *name = NULL; |
7962 | 148k | xmlChar *publicId = NULL; |
7963 | 148k | xmlChar *URI = NULL; |
7964 | | |
7965 | | /* |
7966 | | * We know that '<!DOCTYPE' has been detected. |
7967 | | */ |
7968 | 148k | SKIP(9); |
7969 | | |
7970 | 148k | if (SKIP_BLANKS == 0) { |
7971 | 20.4k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
7972 | 20.4k | "Space required after 'DOCTYPE'\n"); |
7973 | 20.4k | } |
7974 | | |
7975 | | /* |
7976 | | * Parse the DOCTYPE name. |
7977 | | */ |
7978 | 148k | name = xmlParseName(ctxt); |
7979 | 148k | if (name == NULL) { |
7980 | 6.00k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
7981 | 6.00k | "xmlParseDocTypeDecl : no DOCTYPE name !\n"); |
7982 | 6.00k | } |
7983 | 148k | ctxt->intSubName = name; |
7984 | | |
7985 | 148k | SKIP_BLANKS; |
7986 | | |
7987 | | /* |
7988 | | * Check for public and system identifier (URI) |
7989 | | */ |
7990 | 148k | URI = xmlParseExternalID(ctxt, &publicId, 1); |
7991 | | |
7992 | 148k | if ((URI != NULL) || (publicId != NULL)) { |
7993 | 36.9k | ctxt->hasExternalSubset = 1; |
7994 | 36.9k | } |
7995 | 148k | ctxt->extSubURI = URI; |
7996 | 148k | ctxt->extSubSystem = publicId; |
7997 | | |
7998 | 148k | SKIP_BLANKS; |
7999 | | |
8000 | | /* |
8001 | | * Create and update the internal subset. |
8002 | | */ |
8003 | 148k | if ((ctxt->sax != NULL) && (ctxt->sax->internalSubset != NULL) && |
8004 | 148k | (!ctxt->disableSAX)) |
8005 | 108k | ctxt->sax->internalSubset(ctxt->userData, name, publicId, URI); |
8006 | | |
8007 | 148k | if ((RAW != '[') && (RAW != '>')) { |
8008 | 5.13k | xmlFatalErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED, NULL); |
8009 | 5.13k | } |
8010 | 148k | } |
8011 | | |
8012 | | /** |
8013 | | * Parse the internal subset declaration |
8014 | | * |
8015 | | * [28 end] ('[' (markupdecl | PEReference | S)* ']' S?)? '>' |
8016 | | * @param ctxt an XML parser context |
8017 | | */ |
8018 | | |
8019 | | static void |
8020 | 136k | xmlParseInternalSubset(xmlParserCtxtPtr ctxt) { |
8021 | | /* |
8022 | | * Is there any DTD definition ? |
8023 | | */ |
8024 | 136k | if (RAW == '[') { |
8025 | 136k | int oldInputNr = ctxt->inputNr; |
8026 | | |
8027 | 136k | NEXT; |
8028 | | /* |
8029 | | * Parse the succession of Markup declarations and |
8030 | | * PEReferences. |
8031 | | * Subsequence (markupdecl | PEReference | S)* |
8032 | | */ |
8033 | 136k | SKIP_BLANKS; |
8034 | 534k | while (1) { |
8035 | 534k | if (PARSER_STOPPED(ctxt)) { |
8036 | 20.1k | return; |
8037 | 514k | } else if (ctxt->input->cur >= ctxt->input->end) { |
8038 | 6.13k | if (ctxt->inputNr <= oldInputNr) { |
8039 | 1.50k | xmlFatalErr(ctxt, XML_ERR_INT_SUBSET_NOT_FINISHED, NULL); |
8040 | 1.50k | return; |
8041 | 1.50k | } |
8042 | 4.63k | xmlPopPE(ctxt); |
8043 | 508k | } else if ((RAW == ']') && (ctxt->inputNr <= oldInputNr)) { |
8044 | 42.0k | NEXT; |
8045 | 42.0k | SKIP_BLANKS; |
8046 | 42.0k | break; |
8047 | 466k | } else if ((PARSER_EXTERNAL(ctxt)) && |
8048 | 466k | (RAW == '<') && (NXT(1) == '!') && (NXT(2) == '[')) { |
8049 | | /* |
8050 | | * Conditional sections are allowed in external entities |
8051 | | * included by PE References in the internal subset. |
8052 | | */ |
8053 | 3.51k | xmlParseConditionalSections(ctxt); |
8054 | 462k | } else if ((RAW == '<') && ((NXT(1) == '!') || (NXT(1) == '?'))) { |
8055 | 353k | xmlParseMarkupDecl(ctxt); |
8056 | 353k | } else if (RAW == '%') { |
8057 | 36.4k | xmlParsePERefInternal(ctxt, 1); |
8058 | 72.2k | } else { |
8059 | 72.2k | xmlFatalErr(ctxt, XML_ERR_INT_SUBSET_NOT_FINISHED, NULL); |
8060 | | |
8061 | 86.4k | while (ctxt->inputNr > oldInputNr) |
8062 | 14.1k | xmlPopPE(ctxt); |
8063 | 72.2k | return; |
8064 | 72.2k | } |
8065 | 398k | SKIP_BLANKS; |
8066 | 398k | SHRINK; |
8067 | 398k | GROW; |
8068 | 398k | } |
8069 | 136k | } |
8070 | | |
8071 | | /* |
8072 | | * We should be at the end of the DOCTYPE declaration. |
8073 | | */ |
8074 | 42.0k | if (RAW != '>') { |
8075 | 526 | xmlFatalErr(ctxt, XML_ERR_DOCTYPE_NOT_FINISHED, NULL); |
8076 | 526 | return; |
8077 | 526 | } |
8078 | 41.5k | NEXT; |
8079 | 41.5k | } |
8080 | | |
8081 | | #ifdef LIBXML_SAX1_ENABLED |
8082 | | /** |
8083 | | * Parse an attribute |
8084 | | * |
8085 | | * @deprecated Internal function, don't use. |
8086 | | * |
8087 | | * [41] Attribute ::= Name Eq AttValue |
8088 | | * |
8089 | | * [ WFC: No External Entity References ] |
8090 | | * Attribute values cannot contain direct or indirect entity references |
8091 | | * to external entities. |
8092 | | * |
8093 | | * [ WFC: No < in Attribute Values ] |
8094 | | * The replacement text of any entity referred to directly or indirectly in |
8095 | | * an attribute value (other than "<") must not contain a <. |
8096 | | * |
8097 | | * [ VC: Attribute Value Type ] |
8098 | | * The attribute must have been declared; the value must be of the type |
8099 | | * declared for it. |
8100 | | * |
8101 | | * [25] Eq ::= S? '=' S? |
8102 | | * |
8103 | | * With namespace: |
8104 | | * |
8105 | | * [NS 11] Attribute ::= QName Eq AttValue |
8106 | | * |
8107 | | * Also the case QName == xmlns:??? is handled independently as a namespace |
8108 | | * definition. |
8109 | | * |
8110 | | * @param ctxt an XML parser context |
8111 | | * @param value a xmlChar ** used to store the value of the attribute |
8112 | | * @returns the attribute name, and the value in *value. |
8113 | | */ |
8114 | | |
8115 | | const xmlChar * |
8116 | | xmlParseAttribute(xmlParserCtxt *ctxt, xmlChar **value) { |
8117 | | const xmlChar *name; |
8118 | | xmlChar *val; |
8119 | | |
8120 | | *value = NULL; |
8121 | | GROW; |
8122 | | name = xmlParseName(ctxt); |
8123 | | if (name == NULL) { |
8124 | | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8125 | | "error parsing attribute name\n"); |
8126 | | return(NULL); |
8127 | | } |
8128 | | |
8129 | | /* |
8130 | | * read the value |
8131 | | */ |
8132 | | SKIP_BLANKS; |
8133 | | if (RAW == '=') { |
8134 | | NEXT; |
8135 | | SKIP_BLANKS; |
8136 | | val = xmlParseAttValue(ctxt); |
8137 | | } else { |
8138 | | xmlFatalErrMsgStr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE, |
8139 | | "Specification mandates value for attribute %s\n", name); |
8140 | | return(name); |
8141 | | } |
8142 | | |
8143 | | /* |
8144 | | * Check that xml:lang conforms to the specification |
8145 | | * No more registered as an error, just generate a warning now |
8146 | | * since this was deprecated in XML second edition |
8147 | | */ |
8148 | | if ((ctxt->pedantic) && (xmlStrEqual(name, BAD_CAST "xml:lang"))) { |
8149 | | if (!xmlCheckLanguageID(val)) { |
8150 | | xmlWarningMsg(ctxt, XML_WAR_LANG_VALUE, |
8151 | | "Malformed value for xml:lang : %s\n", |
8152 | | val, NULL); |
8153 | | } |
8154 | | } |
8155 | | |
8156 | | /* |
8157 | | * Check that xml:space conforms to the specification |
8158 | | */ |
8159 | | if (xmlStrEqual(name, BAD_CAST "xml:space")) { |
8160 | | if (xmlStrEqual(val, BAD_CAST "default")) |
8161 | | *(ctxt->space) = 0; |
8162 | | else if (xmlStrEqual(val, BAD_CAST "preserve")) |
8163 | | *(ctxt->space) = 1; |
8164 | | else { |
8165 | | xmlWarningMsg(ctxt, XML_WAR_SPACE_VALUE, |
8166 | | "Invalid value \"%s\" for xml:space : \"default\" or \"preserve\" expected\n", |
8167 | | val, NULL); |
8168 | | } |
8169 | | } |
8170 | | |
8171 | | *value = val; |
8172 | | return(name); |
8173 | | } |
8174 | | |
8175 | | /** |
8176 | | * Parse a start tag. Always consumes '<'. |
8177 | | * |
8178 | | * @deprecated Internal function, don't use. |
8179 | | * |
8180 | | * [40] STag ::= '<' Name (S Attribute)* S? '>' |
8181 | | * |
8182 | | * [ WFC: Unique Att Spec ] |
8183 | | * No attribute name may appear more than once in the same start-tag or |
8184 | | * empty-element tag. |
8185 | | * |
8186 | | * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>' |
8187 | | * |
8188 | | * [ WFC: Unique Att Spec ] |
8189 | | * No attribute name may appear more than once in the same start-tag or |
8190 | | * empty-element tag. |
8191 | | * |
8192 | | * With namespace: |
8193 | | * |
8194 | | * [NS 8] STag ::= '<' QName (S Attribute)* S? '>' |
8195 | | * |
8196 | | * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>' |
8197 | | * |
8198 | | * @param ctxt an XML parser context |
8199 | | * @returns the element name parsed |
8200 | | */ |
8201 | | |
8202 | | const xmlChar * |
8203 | | xmlParseStartTag(xmlParserCtxt *ctxt) { |
8204 | | const xmlChar *name; |
8205 | | const xmlChar *attname; |
8206 | | xmlChar *attvalue; |
8207 | | const xmlChar **atts = ctxt->atts; |
8208 | | int nbatts = 0; |
8209 | | int maxatts = ctxt->maxatts; |
8210 | | int i; |
8211 | | |
8212 | | if (RAW != '<') return(NULL); |
8213 | | NEXT1; |
8214 | | |
8215 | | name = xmlParseName(ctxt); |
8216 | | if (name == NULL) { |
8217 | | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8218 | | "xmlParseStartTag: invalid element name\n"); |
8219 | | return(NULL); |
8220 | | } |
8221 | | |
8222 | | /* |
8223 | | * Now parse the attributes, it ends up with the ending |
8224 | | * |
8225 | | * (S Attribute)* S? |
8226 | | */ |
8227 | | SKIP_BLANKS; |
8228 | | GROW; |
8229 | | |
8230 | | while (((RAW != '>') && |
8231 | | ((RAW != '/') || (NXT(1) != '>')) && |
8232 | | (IS_BYTE_CHAR(RAW))) && (PARSER_STOPPED(ctxt) == 0)) { |
8233 | | attname = xmlParseAttribute(ctxt, &attvalue); |
8234 | | if (attname == NULL) |
8235 | | break; |
8236 | | if (attvalue != NULL) { |
8237 | | /* |
8238 | | * [ WFC: Unique Att Spec ] |
8239 | | * No attribute name may appear more than once in the same |
8240 | | * start-tag or empty-element tag. |
8241 | | */ |
8242 | | for (i = 0; i < nbatts;i += 2) { |
8243 | | if (xmlStrEqual(atts[i], attname)) { |
8244 | | xmlErrAttributeDup(ctxt, NULL, attname); |
8245 | | goto failed; |
8246 | | } |
8247 | | } |
8248 | | /* |
8249 | | * Add the pair to atts |
8250 | | */ |
8251 | | if (nbatts + 4 > maxatts) { |
8252 | | const xmlChar **n; |
8253 | | int newSize; |
8254 | | |
8255 | | newSize = xmlGrowCapacity(maxatts, sizeof(n[0]) * 2, |
8256 | | 11, XML_MAX_ATTRS); |
8257 | | if (newSize < 0) { |
8258 | | xmlErrMemory(ctxt); |
8259 | | goto failed; |
8260 | | } |
8261 | | #ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION |
8262 | | if (newSize < 2) |
8263 | | newSize = 2; |
8264 | | #endif |
8265 | | n = xmlRealloc(atts, newSize * sizeof(n[0]) * 2); |
8266 | | if (n == NULL) { |
8267 | | xmlErrMemory(ctxt); |
8268 | | goto failed; |
8269 | | } |
8270 | | atts = n; |
8271 | | maxatts = newSize * 2; |
8272 | | ctxt->atts = atts; |
8273 | | ctxt->maxatts = maxatts; |
8274 | | } |
8275 | | |
8276 | | atts[nbatts++] = attname; |
8277 | | atts[nbatts++] = attvalue; |
8278 | | atts[nbatts] = NULL; |
8279 | | atts[nbatts + 1] = NULL; |
8280 | | |
8281 | | attvalue = NULL; |
8282 | | } |
8283 | | |
8284 | | failed: |
8285 | | |
8286 | | if (attvalue != NULL) |
8287 | | xmlFree(attvalue); |
8288 | | |
8289 | | GROW |
8290 | | if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>')))) |
8291 | | break; |
8292 | | if (SKIP_BLANKS == 0) { |
8293 | | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
8294 | | "attributes construct error\n"); |
8295 | | } |
8296 | | SHRINK; |
8297 | | GROW; |
8298 | | } |
8299 | | |
8300 | | /* |
8301 | | * SAX: Start of Element ! |
8302 | | */ |
8303 | | if ((ctxt->sax != NULL) && (ctxt->sax->startElement != NULL) && |
8304 | | (!ctxt->disableSAX)) { |
8305 | | if (nbatts > 0) |
8306 | | ctxt->sax->startElement(ctxt->userData, name, atts); |
8307 | | else |
8308 | | ctxt->sax->startElement(ctxt->userData, name, NULL); |
8309 | | } |
8310 | | |
8311 | | if (atts != NULL) { |
8312 | | /* Free only the content strings */ |
8313 | | for (i = 1;i < nbatts;i+=2) |
8314 | | if (atts[i] != NULL) |
8315 | | xmlFree((xmlChar *) atts[i]); |
8316 | | } |
8317 | | return(name); |
8318 | | } |
8319 | | |
8320 | | /** |
8321 | | * Parse an end tag. Always consumes '</'. |
8322 | | * |
8323 | | * [42] ETag ::= '</' Name S? '>' |
8324 | | * |
8325 | | * With namespace |
8326 | | * |
8327 | | * [NS 9] ETag ::= '</' QName S? '>' |
8328 | | * @param ctxt an XML parser context |
8329 | | * @param line line of the start tag |
8330 | | */ |
8331 | | |
8332 | | static void |
8333 | | xmlParseEndTag1(xmlParserCtxtPtr ctxt, int line) { |
8334 | | const xmlChar *name; |
8335 | | |
8336 | | GROW; |
8337 | | if ((RAW != '<') || (NXT(1) != '/')) { |
8338 | | xmlFatalErrMsg(ctxt, XML_ERR_LTSLASH_REQUIRED, |
8339 | | "xmlParseEndTag: '</' not found\n"); |
8340 | | return; |
8341 | | } |
8342 | | SKIP(2); |
8343 | | |
8344 | | name = xmlParseNameAndCompare(ctxt,ctxt->name); |
8345 | | |
8346 | | /* |
8347 | | * We should definitely be at the ending "S? '>'" part |
8348 | | */ |
8349 | | GROW; |
8350 | | SKIP_BLANKS; |
8351 | | if ((!IS_BYTE_CHAR(RAW)) || (RAW != '>')) { |
8352 | | xmlFatalErr(ctxt, XML_ERR_GT_REQUIRED, NULL); |
8353 | | } else |
8354 | | NEXT1; |
8355 | | |
8356 | | /* |
8357 | | * [ WFC: Element Type Match ] |
8358 | | * The Name in an element's end-tag must match the element type in the |
8359 | | * start-tag. |
8360 | | * |
8361 | | */ |
8362 | | if (name != (xmlChar*)1) { |
8363 | | if (name == NULL) name = BAD_CAST "unparsable"; |
8364 | | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NAME_MISMATCH, |
8365 | | "Opening and ending tag mismatch: %s line %d and %s\n", |
8366 | | ctxt->name, line, name); |
8367 | | } |
8368 | | |
8369 | | /* |
8370 | | * SAX: End of Tag |
8371 | | */ |
8372 | | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL) && |
8373 | | (!ctxt->disableSAX)) |
8374 | | ctxt->sax->endElement(ctxt->userData, ctxt->name); |
8375 | | |
8376 | | namePop(ctxt); |
8377 | | spacePop(ctxt); |
8378 | | } |
8379 | | |
8380 | | /** |
8381 | | * Parse an end of tag |
8382 | | * |
8383 | | * @deprecated Internal function, don't use. |
8384 | | * |
8385 | | * [42] ETag ::= '</' Name S? '>' |
8386 | | * |
8387 | | * With namespace |
8388 | | * |
8389 | | * [NS 9] ETag ::= '</' QName S? '>' |
8390 | | * @param ctxt an XML parser context |
8391 | | */ |
8392 | | |
8393 | | void |
8394 | | xmlParseEndTag(xmlParserCtxt *ctxt) { |
8395 | | xmlParseEndTag1(ctxt, 0); |
8396 | | } |
8397 | | #endif /* LIBXML_SAX1_ENABLED */ |
8398 | | |
8399 | | /************************************************************************ |
8400 | | * * |
8401 | | * SAX 2 specific operations * |
8402 | | * * |
8403 | | ************************************************************************/ |
8404 | | |
8405 | | /** |
8406 | | * Parse an XML Namespace QName |
8407 | | * |
8408 | | * [6] QName ::= (Prefix ':')? LocalPart |
8409 | | * [7] Prefix ::= NCName |
8410 | | * [8] LocalPart ::= NCName |
8411 | | * |
8412 | | * @param ctxt an XML parser context |
8413 | | * @param prefix pointer to store the prefix part |
8414 | | * @returns the Name parsed or NULL |
8415 | | */ |
8416 | | |
8417 | | static xmlHashedString |
8418 | 16.2M | xmlParseQNameHashed(xmlParserCtxtPtr ctxt, xmlHashedString *prefix) { |
8419 | 16.2M | xmlHashedString l, p; |
8420 | 16.2M | int start, isNCName = 0; |
8421 | | |
8422 | 16.2M | l.name = NULL; |
8423 | 16.2M | p.name = NULL; |
8424 | | |
8425 | 16.2M | GROW; |
8426 | 16.2M | start = CUR_PTR - BASE_PTR; |
8427 | | |
8428 | 16.2M | l = xmlParseNCName(ctxt); |
8429 | 16.2M | if (l.name != NULL) { |
8430 | 13.0M | isNCName = 1; |
8431 | 13.0M | if (CUR == ':') { |
8432 | 4.97M | NEXT; |
8433 | 4.97M | p = l; |
8434 | 4.97M | l = xmlParseNCName(ctxt); |
8435 | 4.97M | } |
8436 | 13.0M | } |
8437 | 16.2M | if ((l.name == NULL) || (CUR == ':')) { |
8438 | 3.31M | xmlChar *tmp; |
8439 | | |
8440 | 3.31M | l.name = NULL; |
8441 | 3.31M | p.name = NULL; |
8442 | 3.31M | if ((isNCName == 0) && (CUR != ':')) |
8443 | 3.19M | return(l); |
8444 | 127k | tmp = xmlParseNmtoken(ctxt); |
8445 | 127k | if (tmp != NULL) |
8446 | 67.4k | xmlFree(tmp); |
8447 | 127k | l = xmlDictLookupHashed(ctxt->dict, BASE_PTR + start, |
8448 | 127k | CUR_PTR - (BASE_PTR + start)); |
8449 | 127k | if (l.name == NULL) { |
8450 | 3 | xmlErrMemory(ctxt); |
8451 | 3 | return(l); |
8452 | 3 | } |
8453 | 127k | xmlNsErr(ctxt, XML_NS_ERR_QNAME, |
8454 | 127k | "Failed to parse QName '%s'\n", l.name, NULL, NULL); |
8455 | 127k | } |
8456 | | |
8457 | 13.1M | *prefix = p; |
8458 | 13.1M | return(l); |
8459 | 16.2M | } |
8460 | | |
8461 | | /** |
8462 | | * Parse an XML Namespace QName |
8463 | | * |
8464 | | * [6] QName ::= (Prefix ':')? LocalPart |
8465 | | * [7] Prefix ::= NCName |
8466 | | * [8] LocalPart ::= NCName |
8467 | | * |
8468 | | * @param ctxt an XML parser context |
8469 | | * @param prefix pointer to store the prefix part |
8470 | | * @returns the Name parsed or NULL |
8471 | | */ |
8472 | | |
8473 | | static const xmlChar * |
8474 | 58.1k | xmlParseQName(xmlParserCtxtPtr ctxt, const xmlChar **prefix) { |
8475 | 58.1k | xmlHashedString n, p; |
8476 | | |
8477 | 58.1k | n = xmlParseQNameHashed(ctxt, &p); |
8478 | 58.1k | if (n.name == NULL) |
8479 | 3.37k | return(NULL); |
8480 | 54.7k | *prefix = p.name; |
8481 | 54.7k | return(n.name); |
8482 | 58.1k | } |
8483 | | |
8484 | | /** |
8485 | | * Parse an XML name and compares for match |
8486 | | * (specialized for endtag parsing) |
8487 | | * |
8488 | | * @param ctxt an XML parser context |
8489 | | * @param name the localname |
8490 | | * @param prefix the prefix, if any. |
8491 | | * @returns NULL for an illegal name, (xmlChar*) 1 for success |
8492 | | * and the name for mismatch |
8493 | | */ |
8494 | | |
8495 | | static const xmlChar * |
8496 | | xmlParseQNameAndCompare(xmlParserCtxtPtr ctxt, xmlChar const *name, |
8497 | 1.19M | xmlChar const *prefix) { |
8498 | 1.19M | const xmlChar *cmp; |
8499 | 1.19M | const xmlChar *in; |
8500 | 1.19M | const xmlChar *ret; |
8501 | 1.19M | const xmlChar *prefix2; |
8502 | | |
8503 | 1.19M | if (prefix == NULL) return(xmlParseNameAndCompare(ctxt, name)); |
8504 | | |
8505 | 1.19M | GROW; |
8506 | 1.19M | in = ctxt->input->cur; |
8507 | | |
8508 | 1.19M | cmp = prefix; |
8509 | 4.28M | while (*in != 0 && *in == *cmp) { |
8510 | 3.09M | ++in; |
8511 | 3.09M | ++cmp; |
8512 | 3.09M | } |
8513 | 1.19M | if ((*cmp == 0) && (*in == ':')) { |
8514 | 1.17M | in++; |
8515 | 1.17M | cmp = name; |
8516 | 11.2M | while (*in != 0 && *in == *cmp) { |
8517 | 10.1M | ++in; |
8518 | 10.1M | ++cmp; |
8519 | 10.1M | } |
8520 | 1.17M | if (*cmp == 0 && (*in == '>' || IS_BLANK_CH (*in))) { |
8521 | | /* success */ |
8522 | 1.13M | ctxt->input->col += in - ctxt->input->cur; |
8523 | 1.13M | ctxt->input->cur = in; |
8524 | 1.13M | return((const xmlChar*) 1); |
8525 | 1.13M | } |
8526 | 1.17M | } |
8527 | | /* |
8528 | | * all strings coms from the dictionary, equality can be done directly |
8529 | | */ |
8530 | 58.1k | ret = xmlParseQName (ctxt, &prefix2); |
8531 | 58.1k | if (ret == NULL) |
8532 | 3.37k | return(NULL); |
8533 | 54.7k | if ((ret == name) && (prefix == prefix2)) |
8534 | 20.9k | return((const xmlChar*) 1); |
8535 | 33.8k | return ret; |
8536 | 54.7k | } |
8537 | | |
8538 | | /** |
8539 | | * Parse an attribute in the new SAX2 framework. |
8540 | | * |
8541 | | * @param ctxt an XML parser context |
8542 | | * @param pref the element prefix |
8543 | | * @param elem the element name |
8544 | | * @param hprefix resulting attribute prefix |
8545 | | * @param value resulting value of the attribute |
8546 | | * @param len resulting length of the attribute |
8547 | | * @param alloc resulting indicator if the attribute was allocated |
8548 | | * @returns the attribute name, and the value in *value, . |
8549 | | */ |
8550 | | |
8551 | | static xmlHashedString |
8552 | | xmlParseAttribute2(xmlParserCtxtPtr ctxt, |
8553 | | const xmlChar * pref, const xmlChar * elem, |
8554 | | xmlHashedString * hprefix, xmlChar ** value, |
8555 | | int *len, int *alloc) |
8556 | 7.95M | { |
8557 | 7.95M | xmlHashedString hname; |
8558 | 7.95M | const xmlChar *prefix, *name; |
8559 | 7.95M | xmlChar *val = NULL, *internal_val = NULL; |
8560 | 7.95M | int special = 0; |
8561 | 7.95M | int isNamespace; |
8562 | 7.95M | int flags; |
8563 | | |
8564 | 7.95M | *value = NULL; |
8565 | 7.95M | GROW; |
8566 | 7.95M | hname = xmlParseQNameHashed(ctxt, hprefix); |
8567 | 7.95M | if (hname.name == NULL) { |
8568 | 2.61M | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8569 | 2.61M | "error parsing attribute name\n"); |
8570 | 2.61M | return(hname); |
8571 | 2.61M | } |
8572 | 5.34M | name = hname.name; |
8573 | 5.34M | prefix = hprefix->name; |
8574 | | |
8575 | | /* |
8576 | | * get the type if needed |
8577 | | */ |
8578 | 5.34M | if (ctxt->attsSpecial != NULL) { |
8579 | 53.6k | special = XML_PTR_TO_INT(xmlHashQLookup2(ctxt->attsSpecial, pref, elem, |
8580 | 53.6k | prefix, name)); |
8581 | 53.6k | } |
8582 | | |
8583 | | /* |
8584 | | * read the value |
8585 | | */ |
8586 | 5.34M | SKIP_BLANKS; |
8587 | 5.34M | if (RAW != '=') { |
8588 | 97.5k | xmlFatalErrMsgStr(ctxt, XML_ERR_ATTRIBUTE_WITHOUT_VALUE, |
8589 | 97.5k | "Specification mandates value for attribute %s\n", |
8590 | 97.5k | name); |
8591 | 97.5k | goto error; |
8592 | 97.5k | } |
8593 | | |
8594 | | |
8595 | 5.24M | NEXT; |
8596 | 5.24M | SKIP_BLANKS; |
8597 | 5.24M | flags = 0; |
8598 | 5.24M | isNamespace = (((prefix == NULL) && (name == ctxt->str_xmlns)) || |
8599 | 5.24M | (prefix == ctxt->str_xmlns)); |
8600 | 5.24M | val = xmlParseAttValueInternal(ctxt, len, &flags, special, |
8601 | 5.24M | isNamespace); |
8602 | 5.24M | if (val == NULL) |
8603 | 30.7k | goto error; |
8604 | | |
8605 | 5.21M | *alloc = (flags & XML_ATTVAL_ALLOC) != 0; |
8606 | | |
8607 | | #ifdef LIBXML_VALID_ENABLED |
8608 | | if ((ctxt->validate) && |
8609 | | (ctxt->standalone) && |
8610 | | (special & XML_SPECIAL_EXTERNAL) && |
8611 | | (flags & XML_ATTVAL_NORM_CHANGE)) { |
8612 | | xmlValidityError(ctxt, XML_DTD_NOT_STANDALONE, |
8613 | | "standalone: normalization of attribute %s on %s " |
8614 | | "by external subset declaration\n", |
8615 | | name, elem); |
8616 | | } |
8617 | | #endif |
8618 | | |
8619 | 5.21M | if (prefix == ctxt->str_xml) { |
8620 | | /* |
8621 | | * Check that xml:lang conforms to the specification |
8622 | | * No more registered as an error, just generate a warning now |
8623 | | * since this was deprecated in XML second edition |
8624 | | */ |
8625 | 103k | if ((ctxt->pedantic) && (xmlStrEqual(name, BAD_CAST "lang"))) { |
8626 | 0 | internal_val = xmlStrndup(val, *len); |
8627 | 0 | if (internal_val == NULL) |
8628 | 0 | goto mem_error; |
8629 | 0 | if (!xmlCheckLanguageID(internal_val)) { |
8630 | 0 | xmlWarningMsg(ctxt, XML_WAR_LANG_VALUE, |
8631 | 0 | "Malformed value for xml:lang : %s\n", |
8632 | 0 | internal_val, NULL); |
8633 | 0 | } |
8634 | 0 | } |
8635 | | |
8636 | | /* |
8637 | | * Check that xml:space conforms to the specification |
8638 | | */ |
8639 | 103k | if (xmlStrEqual(name, BAD_CAST "space")) { |
8640 | 984 | internal_val = xmlStrndup(val, *len); |
8641 | 984 | if (internal_val == NULL) |
8642 | 2 | goto mem_error; |
8643 | 982 | if (xmlStrEqual(internal_val, BAD_CAST "default")) |
8644 | 319 | *(ctxt->space) = 0; |
8645 | 663 | else if (xmlStrEqual(internal_val, BAD_CAST "preserve")) |
8646 | 339 | *(ctxt->space) = 1; |
8647 | 324 | else { |
8648 | 324 | xmlWarningMsg(ctxt, XML_WAR_SPACE_VALUE, |
8649 | 324 | "Invalid value \"%s\" for xml:space : \"default\" or \"preserve\" expected\n", |
8650 | 324 | internal_val, NULL); |
8651 | 324 | } |
8652 | 982 | } |
8653 | 103k | if (internal_val) { |
8654 | 982 | xmlFree(internal_val); |
8655 | 982 | } |
8656 | 103k | } |
8657 | | |
8658 | 5.21M | *value = val; |
8659 | 5.21M | return (hname); |
8660 | | |
8661 | 2 | mem_error: |
8662 | 2 | xmlErrMemory(ctxt); |
8663 | 128k | error: |
8664 | 128k | if ((val != NULL) && (*alloc != 0)) |
8665 | 1 | xmlFree(val); |
8666 | 128k | return(hname); |
8667 | 2 | } |
8668 | | |
8669 | | /** |
8670 | | * Inserts a new attribute into the hash table. |
8671 | | * |
8672 | | * @param ctxt parser context |
8673 | | * @param size size of the hash table |
8674 | | * @param name attribute name |
8675 | | * @param uri namespace uri |
8676 | | * @param hashValue combined hash value of name and uri |
8677 | | * @param aindex attribute index (this is a multiple of 5) |
8678 | | * @returns INT_MAX if no existing attribute was found, the attribute |
8679 | | * index if an attribute was found, -1 if a memory allocation failed. |
8680 | | */ |
8681 | | static int |
8682 | | xmlAttrHashInsert(xmlParserCtxtPtr ctxt, unsigned size, const xmlChar *name, |
8683 | 1.15M | const xmlChar *uri, unsigned hashValue, int aindex) { |
8684 | 1.15M | xmlAttrHashBucket *table = ctxt->attrHash; |
8685 | 1.15M | xmlAttrHashBucket *bucket; |
8686 | 1.15M | unsigned hindex; |
8687 | | |
8688 | 1.15M | hindex = hashValue & (size - 1); |
8689 | 1.15M | bucket = &table[hindex]; |
8690 | | |
8691 | 1.28M | while (bucket->index >= 0) { |
8692 | 156k | const xmlChar **atts = &ctxt->atts[bucket->index]; |
8693 | | |
8694 | 156k | if (name == atts[0]) { |
8695 | 28.3k | int nsIndex = XML_PTR_TO_INT(atts[2]); |
8696 | | |
8697 | 28.3k | if ((nsIndex == NS_INDEX_EMPTY) ? (uri == NULL) : |
8698 | 28.3k | (nsIndex == NS_INDEX_XML) ? (uri == ctxt->str_xml_ns) : |
8699 | 11.5k | (uri == ctxt->nsTab[nsIndex * 2 + 1])) |
8700 | 26.3k | return(bucket->index); |
8701 | 28.3k | } |
8702 | | |
8703 | 129k | hindex++; |
8704 | 129k | bucket++; |
8705 | 129k | if (hindex >= size) { |
8706 | 22.8k | hindex = 0; |
8707 | 22.8k | bucket = table; |
8708 | 22.8k | } |
8709 | 129k | } |
8710 | | |
8711 | 1.12M | bucket->index = aindex; |
8712 | | |
8713 | 1.12M | return(INT_MAX); |
8714 | 1.15M | } |
8715 | | |
8716 | | static int |
8717 | | xmlAttrHashInsertQName(xmlParserCtxtPtr ctxt, unsigned size, |
8718 | | const xmlChar *name, const xmlChar *prefix, |
8719 | 2.62k | unsigned hashValue, int aindex) { |
8720 | 2.62k | xmlAttrHashBucket *table = ctxt->attrHash; |
8721 | 2.62k | xmlAttrHashBucket *bucket; |
8722 | 2.62k | unsigned hindex; |
8723 | | |
8724 | 2.62k | hindex = hashValue & (size - 1); |
8725 | 2.62k | bucket = &table[hindex]; |
8726 | | |
8727 | 4.23k | while (bucket->index >= 0) { |
8728 | 2.19k | const xmlChar **atts = &ctxt->atts[bucket->index]; |
8729 | | |
8730 | 2.19k | if ((name == atts[0]) && (prefix == atts[1])) |
8731 | 574 | return(bucket->index); |
8732 | | |
8733 | 1.61k | hindex++; |
8734 | 1.61k | bucket++; |
8735 | 1.61k | if (hindex >= size) { |
8736 | 354 | hindex = 0; |
8737 | 354 | bucket = table; |
8738 | 354 | } |
8739 | 1.61k | } |
8740 | | |
8741 | 2.04k | bucket->index = aindex; |
8742 | | |
8743 | 2.04k | return(INT_MAX); |
8744 | 2.62k | } |
8745 | | /** |
8746 | | * Parse a start tag. Always consumes '<'. |
8747 | | * |
8748 | | * This routine is called when running SAX2 parsing |
8749 | | * |
8750 | | * [40] STag ::= '<' Name (S Attribute)* S? '>' |
8751 | | * |
8752 | | * [ WFC: Unique Att Spec ] |
8753 | | * No attribute name may appear more than once in the same start-tag or |
8754 | | * empty-element tag. |
8755 | | * |
8756 | | * [44] EmptyElemTag ::= '<' Name (S Attribute)* S? '/>' |
8757 | | * |
8758 | | * [ WFC: Unique Att Spec ] |
8759 | | * No attribute name may appear more than once in the same start-tag or |
8760 | | * empty-element tag. |
8761 | | * |
8762 | | * With namespace: |
8763 | | * |
8764 | | * [NS 8] STag ::= '<' QName (S Attribute)* S? '>' |
8765 | | * |
8766 | | * [NS 10] EmptyElement ::= '<' QName (S Attribute)* S? '/>' |
8767 | | * |
8768 | | * @param ctxt an XML parser context |
8769 | | * @param pref resulting namespace prefix |
8770 | | * @param URI resulting namespace URI |
8771 | | * @param nbNsPtr resulting number of namespace declarations |
8772 | | * @returns the element name parsed |
8773 | | */ |
8774 | | |
8775 | | static const xmlChar * |
8776 | | xmlParseStartTag2(xmlParserCtxtPtr ctxt, const xmlChar **pref, |
8777 | 8.28M | const xmlChar **URI, int *nbNsPtr) { |
8778 | 8.28M | xmlHashedString hlocalname; |
8779 | 8.28M | xmlHashedString hprefix; |
8780 | 8.28M | xmlHashedString hattname; |
8781 | 8.28M | xmlHashedString haprefix; |
8782 | 8.28M | const xmlChar *localname; |
8783 | 8.28M | const xmlChar *prefix; |
8784 | 8.28M | const xmlChar *attname; |
8785 | 8.28M | const xmlChar *aprefix; |
8786 | 8.28M | const xmlChar *uri; |
8787 | 8.28M | xmlChar *attvalue = NULL; |
8788 | 8.28M | const xmlChar **atts = ctxt->atts; |
8789 | 8.28M | unsigned attrHashSize = 0; |
8790 | 8.28M | int maxatts = ctxt->maxatts; |
8791 | 8.28M | int nratts, nbatts, nbdef; |
8792 | 8.28M | int i, j, nbNs, nbTotalDef, attval, nsIndex, maxAtts; |
8793 | 8.28M | int alloc = 0; |
8794 | 8.28M | int numNsErr = 0; |
8795 | 8.28M | int numDupErr = 0; |
8796 | | |
8797 | 8.28M | if (RAW != '<') return(NULL); |
8798 | 8.28M | NEXT1; |
8799 | | |
8800 | 8.28M | nbatts = 0; |
8801 | 8.28M | nratts = 0; |
8802 | 8.28M | nbdef = 0; |
8803 | 8.28M | nbNs = 0; |
8804 | 8.28M | nbTotalDef = 0; |
8805 | 8.28M | attval = 0; |
8806 | | |
8807 | 8.28M | if (xmlParserNsStartElement(ctxt->nsdb) < 0) { |
8808 | 0 | xmlErrMemory(ctxt); |
8809 | 0 | return(NULL); |
8810 | 0 | } |
8811 | | |
8812 | 8.28M | hlocalname = xmlParseQNameHashed(ctxt, &hprefix); |
8813 | 8.28M | if (hlocalname.name == NULL) { |
8814 | 576k | xmlFatalErrMsg(ctxt, XML_ERR_NAME_REQUIRED, |
8815 | 576k | "StartTag: invalid element name\n"); |
8816 | 576k | return(NULL); |
8817 | 576k | } |
8818 | 7.71M | localname = hlocalname.name; |
8819 | 7.71M | prefix = hprefix.name; |
8820 | | |
8821 | | /* |
8822 | | * Now parse the attributes, it ends up with the ending |
8823 | | * |
8824 | | * (S Attribute)* S? |
8825 | | */ |
8826 | 7.71M | SKIP_BLANKS; |
8827 | 7.71M | GROW; |
8828 | | |
8829 | | /* |
8830 | | * The ctxt->atts array will be ultimately passed to the SAX callback |
8831 | | * containing five xmlChar pointers for each attribute: |
8832 | | * |
8833 | | * [0] attribute name |
8834 | | * [1] attribute prefix |
8835 | | * [2] namespace URI |
8836 | | * [3] attribute value |
8837 | | * [4] end of attribute value |
8838 | | * |
8839 | | * To save memory, we reuse this array temporarily and store integers |
8840 | | * in these pointer variables. |
8841 | | * |
8842 | | * [0] attribute name |
8843 | | * [1] attribute prefix |
8844 | | * [2] hash value of attribute prefix, and later namespace index |
8845 | | * [3] for non-allocated values: ptrdiff_t offset into input buffer |
8846 | | * [4] for non-allocated values: ptrdiff_t offset into input buffer |
8847 | | * |
8848 | | * The ctxt->attallocs array contains an additional unsigned int for |
8849 | | * each attribute, containing the hash value of the attribute name |
8850 | | * and the alloc flag in bit 31. |
8851 | | */ |
8852 | | |
8853 | 9.75M | while (((RAW != '>') && |
8854 | 9.75M | ((RAW != '/') || (NXT(1) != '>')) && |
8855 | 9.75M | (IS_BYTE_CHAR(RAW))) && (PARSER_STOPPED(ctxt) == 0)) { |
8856 | 7.95M | int len = -1; |
8857 | | |
8858 | 7.95M | hattname = xmlParseAttribute2(ctxt, prefix, localname, |
8859 | 7.95M | &haprefix, &attvalue, &len, |
8860 | 7.95M | &alloc); |
8861 | 7.95M | if (hattname.name == NULL) |
8862 | 2.61M | break; |
8863 | 5.34M | if (attvalue == NULL) |
8864 | 128k | goto next_attr; |
8865 | 5.21M | attname = hattname.name; |
8866 | 5.21M | aprefix = haprefix.name; |
8867 | 5.21M | if (len < 0) len = xmlStrlen(attvalue); |
8868 | | |
8869 | 5.21M | if ((attname == ctxt->str_xmlns) && (aprefix == NULL)) { |
8870 | 63.5k | xmlHashedString huri; |
8871 | 63.5k | xmlURIPtr parsedUri; |
8872 | | |
8873 | 63.5k | huri = xmlDictLookupHashed(ctxt->dict, attvalue, len); |
8874 | 63.5k | uri = huri.name; |
8875 | 63.5k | if (uri == NULL) { |
8876 | 2 | xmlErrMemory(ctxt); |
8877 | 2 | goto next_attr; |
8878 | 2 | } |
8879 | 63.5k | if (*uri != 0) { |
8880 | 61.6k | if (xmlParseURISafe((const char *) uri, &parsedUri) < 0) { |
8881 | 6 | xmlErrMemory(ctxt); |
8882 | 6 | goto next_attr; |
8883 | 6 | } |
8884 | 61.6k | if (parsedUri == NULL) { |
8885 | 34.5k | xmlNsErr(ctxt, XML_WAR_NS_URI, |
8886 | 34.5k | "xmlns: '%s' is not a valid URI\n", |
8887 | 34.5k | uri, NULL, NULL); |
8888 | 34.5k | } else { |
8889 | 27.1k | if (parsedUri->scheme == NULL) { |
8890 | 21.6k | xmlNsWarn(ctxt, XML_WAR_NS_URI_RELATIVE, |
8891 | 21.6k | "xmlns: URI %s is not absolute\n", |
8892 | 21.6k | uri, NULL, NULL); |
8893 | 21.6k | } |
8894 | 27.1k | xmlFreeURI(parsedUri); |
8895 | 27.1k | } |
8896 | 61.6k | if (uri == ctxt->str_xml_ns) { |
8897 | 347 | if (attname != ctxt->str_xml) { |
8898 | 347 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8899 | 347 | "xml namespace URI cannot be the default namespace\n", |
8900 | 347 | NULL, NULL, NULL); |
8901 | 347 | } |
8902 | 347 | goto next_attr; |
8903 | 347 | } |
8904 | 61.3k | if ((len == 29) && |
8905 | 61.3k | (xmlStrEqual(uri, |
8906 | 578 | BAD_CAST "http://www.w3.org/2000/xmlns/"))) { |
8907 | 229 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8908 | 229 | "reuse of the xmlns namespace name is forbidden\n", |
8909 | 229 | NULL, NULL, NULL); |
8910 | 229 | goto next_attr; |
8911 | 229 | } |
8912 | 61.3k | } |
8913 | | |
8914 | 62.9k | if (xmlParserNsPush(ctxt, NULL, &huri, NULL, 0) > 0) |
8915 | 62.1k | nbNs++; |
8916 | 5.14M | } else if (aprefix == ctxt->str_xmlns) { |
8917 | 877k | xmlHashedString huri; |
8918 | 877k | xmlURIPtr parsedUri; |
8919 | | |
8920 | 877k | huri = xmlDictLookupHashed(ctxt->dict, attvalue, len); |
8921 | 877k | uri = huri.name; |
8922 | 877k | if (uri == NULL) { |
8923 | 2 | xmlErrMemory(ctxt); |
8924 | 2 | goto next_attr; |
8925 | 2 | } |
8926 | | |
8927 | 877k | if (attname == ctxt->str_xml) { |
8928 | 747 | if (uri != ctxt->str_xml_ns) { |
8929 | 383 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8930 | 383 | "xml namespace prefix mapped to wrong URI\n", |
8931 | 383 | NULL, NULL, NULL); |
8932 | 383 | } |
8933 | | /* |
8934 | | * Do not keep a namespace definition node |
8935 | | */ |
8936 | 747 | goto next_attr; |
8937 | 747 | } |
8938 | 876k | if (uri == ctxt->str_xml_ns) { |
8939 | 281 | if (attname != ctxt->str_xml) { |
8940 | 281 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8941 | 281 | "xml namespace URI mapped to wrong prefix\n", |
8942 | 281 | NULL, NULL, NULL); |
8943 | 281 | } |
8944 | 281 | goto next_attr; |
8945 | 281 | } |
8946 | 876k | if (attname == ctxt->str_xmlns) { |
8947 | 306 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8948 | 306 | "redefinition of the xmlns prefix is forbidden\n", |
8949 | 306 | NULL, NULL, NULL); |
8950 | 306 | goto next_attr; |
8951 | 306 | } |
8952 | 875k | if ((len == 29) && |
8953 | 875k | (xmlStrEqual(uri, |
8954 | 3.67k | BAD_CAST "http://www.w3.org/2000/xmlns/"))) { |
8955 | 364 | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8956 | 364 | "reuse of the xmlns namespace name is forbidden\n", |
8957 | 364 | NULL, NULL, NULL); |
8958 | 364 | goto next_attr; |
8959 | 364 | } |
8960 | 875k | if ((uri == NULL) || (uri[0] == 0)) { |
8961 | 6.39k | xmlNsErr(ctxt, XML_NS_ERR_XML_NAMESPACE, |
8962 | 6.39k | "xmlns:%s: Empty XML namespace is not allowed\n", |
8963 | 6.39k | attname, NULL, NULL); |
8964 | 6.39k | goto next_attr; |
8965 | 869k | } else { |
8966 | 869k | if (xmlParseURISafe((const char *) uri, &parsedUri) < 0) { |
8967 | 158 | xmlErrMemory(ctxt); |
8968 | 158 | goto next_attr; |
8969 | 158 | } |
8970 | 869k | if (parsedUri == NULL) { |
8971 | 126k | xmlNsErr(ctxt, XML_WAR_NS_URI, |
8972 | 126k | "xmlns:%s: '%s' is not a valid URI\n", |
8973 | 126k | attname, uri, NULL); |
8974 | 742k | } else { |
8975 | 742k | if ((ctxt->pedantic) && (parsedUri->scheme == NULL)) { |
8976 | 0 | xmlNsWarn(ctxt, XML_WAR_NS_URI_RELATIVE, |
8977 | 0 | "xmlns:%s: URI %s is not absolute\n", |
8978 | 0 | attname, uri, NULL); |
8979 | 0 | } |
8980 | 742k | xmlFreeURI(parsedUri); |
8981 | 742k | } |
8982 | 869k | } |
8983 | | |
8984 | 869k | if (xmlParserNsPush(ctxt, &hattname, &huri, NULL, 0) > 0) |
8985 | 863k | nbNs++; |
8986 | 4.27M | } else { |
8987 | | /* |
8988 | | * Populate attributes array, see above for repurposing |
8989 | | * of xmlChar pointers. |
8990 | | */ |
8991 | 4.27M | if ((atts == NULL) || (nbatts + 5 > maxatts)) { |
8992 | 878k | int res = xmlCtxtGrowAttrs(ctxt); |
8993 | | |
8994 | 878k | maxatts = ctxt->maxatts; |
8995 | 878k | atts = ctxt->atts; |
8996 | | |
8997 | 878k | if (res < 0) |
8998 | 96 | goto next_attr; |
8999 | 878k | } |
9000 | 4.27M | ctxt->attallocs[nratts++] = (hattname.hashValue & 0x7FFFFFFF) | |
9001 | 4.27M | ((unsigned) alloc << 31); |
9002 | 4.27M | atts[nbatts++] = attname; |
9003 | 4.27M | atts[nbatts++] = aprefix; |
9004 | 4.27M | atts[nbatts++] = XML_INT_TO_PTR(haprefix.hashValue); |
9005 | 4.27M | if (alloc) { |
9006 | 162k | atts[nbatts++] = attvalue; |
9007 | 162k | attvalue += len; |
9008 | 162k | atts[nbatts++] = attvalue; |
9009 | 4.10M | } else { |
9010 | | /* |
9011 | | * attvalue points into the input buffer which can be |
9012 | | * reallocated. Store differences to input->base instead. |
9013 | | * The pointers will be reconstructed later. |
9014 | | */ |
9015 | 4.10M | atts[nbatts++] = XML_INT_TO_PTR(attvalue - BASE_PTR); |
9016 | 4.10M | attvalue += len; |
9017 | 4.10M | atts[nbatts++] = XML_INT_TO_PTR(attvalue - BASE_PTR); |
9018 | 4.10M | } |
9019 | | /* |
9020 | | * tag if some deallocation is needed |
9021 | | */ |
9022 | 4.27M | if (alloc != 0) attval = 1; |
9023 | 4.27M | attvalue = NULL; /* moved into atts */ |
9024 | 4.27M | } |
9025 | | |
9026 | 5.34M | next_attr: |
9027 | 5.34M | if ((attvalue != NULL) && (alloc != 0)) { |
9028 | 135k | xmlFree(attvalue); |
9029 | 135k | attvalue = NULL; |
9030 | 135k | } |
9031 | | |
9032 | 5.34M | GROW |
9033 | 5.34M | if ((RAW == '>') || (((RAW == '/') && (NXT(1) == '>')))) |
9034 | 3.08M | break; |
9035 | 2.25M | if (SKIP_BLANKS == 0) { |
9036 | 214k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
9037 | 214k | "attributes construct error\n"); |
9038 | 214k | break; |
9039 | 214k | } |
9040 | 2.04M | GROW; |
9041 | 2.04M | } |
9042 | | |
9043 | | /* |
9044 | | * Namespaces from default attributes |
9045 | | */ |
9046 | 7.71M | if (ctxt->attsDefault != NULL) { |
9047 | 111k | xmlDefAttrsPtr defaults; |
9048 | | |
9049 | 111k | defaults = xmlHashLookup2(ctxt->attsDefault, localname, prefix); |
9050 | 111k | if (defaults != NULL) { |
9051 | 311k | for (i = 0; i < defaults->nbAttrs; i++) { |
9052 | 252k | xmlDefAttr *attr = &defaults->attrs[i]; |
9053 | | |
9054 | 252k | attname = attr->name.name; |
9055 | 252k | aprefix = attr->prefix.name; |
9056 | | |
9057 | 252k | if ((attname == ctxt->str_xmlns) && (aprefix == NULL)) { |
9058 | 17.9k | xmlParserEntityCheck(ctxt, attr->expandedSize); |
9059 | | |
9060 | 17.9k | if (xmlParserNsPush(ctxt, NULL, &attr->value, NULL, 1) > 0) |
9061 | 16.6k | nbNs++; |
9062 | 234k | } else if (aprefix == ctxt->str_xmlns) { |
9063 | 80.0k | xmlParserEntityCheck(ctxt, attr->expandedSize); |
9064 | | |
9065 | 80.0k | if (xmlParserNsPush(ctxt, &attr->name, &attr->value, |
9066 | 80.0k | NULL, 1) > 0) |
9067 | 79.4k | nbNs++; |
9068 | 154k | } else { |
9069 | 154k | if (nratts + nbTotalDef >= XML_MAX_ATTRS) { |
9070 | 0 | xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, |
9071 | 0 | "Maximum number of attributes exceeded"); |
9072 | 0 | break; |
9073 | 0 | } |
9074 | 154k | nbTotalDef += 1; |
9075 | 154k | } |
9076 | 252k | } |
9077 | 58.9k | } |
9078 | 111k | } |
9079 | | |
9080 | | /* |
9081 | | * Resolve attribute namespaces |
9082 | | */ |
9083 | 11.9M | for (i = 0; i < nbatts; i += 5) { |
9084 | 4.27M | attname = atts[i]; |
9085 | 4.27M | aprefix = atts[i+1]; |
9086 | | |
9087 | | /* |
9088 | | * The default namespace does not apply to attribute names. |
9089 | | */ |
9090 | 4.27M | if (aprefix == NULL) { |
9091 | 4.10M | nsIndex = NS_INDEX_EMPTY; |
9092 | 4.10M | } else if (aprefix == ctxt->str_xml) { |
9093 | 103k | nsIndex = NS_INDEX_XML; |
9094 | 103k | } else { |
9095 | 62.0k | haprefix.name = aprefix; |
9096 | 62.0k | haprefix.hashValue = (size_t) atts[i+2]; |
9097 | 62.0k | nsIndex = xmlParserNsLookup(ctxt, &haprefix, NULL); |
9098 | | |
9099 | 62.0k | if ((nsIndex == INT_MAX) || (nsIndex < ctxt->nsdb->minNsIndex)) { |
9100 | 45.1k | xmlNsErr(ctxt, XML_NS_ERR_UNDEFINED_NAMESPACE, |
9101 | 45.1k | "Namespace prefix %s for %s on %s is not defined\n", |
9102 | 45.1k | aprefix, attname, localname); |
9103 | 45.1k | nsIndex = NS_INDEX_EMPTY; |
9104 | 45.1k | } |
9105 | 62.0k | } |
9106 | | |
9107 | 4.27M | atts[i+2] = XML_INT_TO_PTR(nsIndex); |
9108 | 4.27M | } |
9109 | | |
9110 | | /* |
9111 | | * Maximum number of attributes including default attributes. |
9112 | | */ |
9113 | 7.71M | maxAtts = nratts + nbTotalDef; |
9114 | | |
9115 | | /* |
9116 | | * Verify that attribute names are unique. |
9117 | | */ |
9118 | 7.71M | if (maxAtts > 1) { |
9119 | 479k | attrHashSize = 4; |
9120 | 648k | while (attrHashSize / 2 < (unsigned) maxAtts) |
9121 | 168k | attrHashSize *= 2; |
9122 | | |
9123 | 479k | if (attrHashSize > ctxt->attrHashMax) { |
9124 | 194k | xmlAttrHashBucket *tmp; |
9125 | | |
9126 | 194k | tmp = xmlRealloc(ctxt->attrHash, attrHashSize * sizeof(tmp[0])); |
9127 | 194k | if (tmp == NULL) { |
9128 | 36 | xmlErrMemory(ctxt); |
9129 | 36 | goto done; |
9130 | 36 | } |
9131 | | |
9132 | 194k | ctxt->attrHash = tmp; |
9133 | 194k | ctxt->attrHashMax = attrHashSize; |
9134 | 194k | } |
9135 | | |
9136 | 479k | memset(ctxt->attrHash, -1, attrHashSize * sizeof(ctxt->attrHash[0])); |
9137 | | |
9138 | 1.54M | for (i = 0, j = 0; j < nratts; i += 5, j++) { |
9139 | 1.06M | const xmlChar *nsuri; |
9140 | 1.06M | unsigned hashValue, nameHashValue, uriHashValue; |
9141 | 1.06M | int res; |
9142 | | |
9143 | 1.06M | attname = atts[i]; |
9144 | 1.06M | aprefix = atts[i+1]; |
9145 | 1.06M | nsIndex = XML_PTR_TO_INT(atts[i+2]); |
9146 | | /* Hash values always have bit 31 set, see dict.c */ |
9147 | 1.06M | nameHashValue = ctxt->attallocs[j] | 0x80000000; |
9148 | | |
9149 | 1.06M | if (nsIndex == NS_INDEX_EMPTY) { |
9150 | | /* |
9151 | | * Prefix with empty namespace means an undeclared |
9152 | | * prefix which was already reported above. |
9153 | | */ |
9154 | 948k | if (aprefix != NULL) |
9155 | 40.4k | continue; |
9156 | 907k | nsuri = NULL; |
9157 | 907k | uriHashValue = URI_HASH_EMPTY; |
9158 | 907k | } else if (nsIndex == NS_INDEX_XML) { |
9159 | 101k | nsuri = ctxt->str_xml_ns; |
9160 | 101k | uriHashValue = URI_HASH_XML; |
9161 | 101k | } else { |
9162 | 15.9k | nsuri = ctxt->nsTab[nsIndex * 2 + 1]; |
9163 | 15.9k | uriHashValue = ctxt->nsdb->extra[nsIndex].uriHashValue; |
9164 | 15.9k | } |
9165 | | |
9166 | 1.02M | hashValue = xmlDictCombineHash(nameHashValue, uriHashValue); |
9167 | 1.02M | res = xmlAttrHashInsert(ctxt, attrHashSize, attname, nsuri, |
9168 | 1.02M | hashValue, i); |
9169 | 1.02M | if (res < 0) |
9170 | 0 | continue; |
9171 | | |
9172 | | /* |
9173 | | * [ WFC: Unique Att Spec ] |
9174 | | * No attribute name may appear more than once in the same |
9175 | | * start-tag or empty-element tag. |
9176 | | * As extended by the Namespace in XML REC. |
9177 | | */ |
9178 | 1.02M | if (res < INT_MAX) { |
9179 | 11.4k | if (aprefix == atts[res+1]) { |
9180 | 10.1k | xmlErrAttributeDup(ctxt, aprefix, attname); |
9181 | 10.1k | numDupErr += 1; |
9182 | 10.1k | } else { |
9183 | 1.26k | xmlNsErr(ctxt, XML_NS_ERR_ATTRIBUTE_REDEFINED, |
9184 | 1.26k | "Namespaced Attribute %s in '%s' redefined\n", |
9185 | 1.26k | attname, nsuri, NULL); |
9186 | 1.26k | numNsErr += 1; |
9187 | 1.26k | } |
9188 | 11.4k | } |
9189 | 1.02M | } |
9190 | 479k | } |
9191 | | |
9192 | | /* |
9193 | | * Default attributes |
9194 | | */ |
9195 | 7.71M | if (ctxt->attsDefault != NULL) { |
9196 | 111k | xmlDefAttrsPtr defaults; |
9197 | | |
9198 | 111k | defaults = xmlHashLookup2(ctxt->attsDefault, localname, prefix); |
9199 | 111k | if (defaults != NULL) { |
9200 | 311k | for (i = 0; i < defaults->nbAttrs; i++) { |
9201 | 252k | xmlDefAttr *attr = &defaults->attrs[i]; |
9202 | 252k | const xmlChar *nsuri = NULL; |
9203 | 252k | unsigned hashValue, uriHashValue = 0; |
9204 | 252k | int res; |
9205 | | |
9206 | 252k | attname = attr->name.name; |
9207 | 252k | aprefix = attr->prefix.name; |
9208 | | |
9209 | 252k | if ((attname == ctxt->str_xmlns) && (aprefix == NULL)) |
9210 | 17.9k | continue; |
9211 | 234k | if (aprefix == ctxt->str_xmlns) |
9212 | 79.9k | continue; |
9213 | | |
9214 | 154k | if (aprefix == NULL) { |
9215 | 67.8k | nsIndex = NS_INDEX_EMPTY; |
9216 | 67.8k | nsuri = NULL; |
9217 | 67.8k | uriHashValue = URI_HASH_EMPTY; |
9218 | 87.0k | } else if (aprefix == ctxt->str_xml) { |
9219 | 22.8k | nsIndex = NS_INDEX_XML; |
9220 | 22.8k | nsuri = ctxt->str_xml_ns; |
9221 | 22.8k | uriHashValue = URI_HASH_XML; |
9222 | 64.2k | } else { |
9223 | 64.2k | nsIndex = xmlParserNsLookup(ctxt, &attr->prefix, NULL); |
9224 | 64.2k | if ((nsIndex == INT_MAX) || |
9225 | 64.2k | (nsIndex < ctxt->nsdb->minNsIndex)) { |
9226 | 61.0k | xmlNsErr(ctxt, XML_NS_ERR_UNDEFINED_NAMESPACE, |
9227 | 61.0k | "Namespace prefix %s for %s on %s is not " |
9228 | 61.0k | "defined\n", |
9229 | 61.0k | aprefix, attname, localname); |
9230 | 61.0k | nsIndex = NS_INDEX_EMPTY; |
9231 | 61.0k | nsuri = NULL; |
9232 | 61.0k | uriHashValue = URI_HASH_EMPTY; |
9233 | 61.0k | } else { |
9234 | 3.18k | nsuri = ctxt->nsTab[nsIndex * 2 + 1]; |
9235 | 3.18k | uriHashValue = ctxt->nsdb->extra[nsIndex].uriHashValue; |
9236 | 3.18k | } |
9237 | 64.2k | } |
9238 | | |
9239 | | /* |
9240 | | * Check whether the attribute exists |
9241 | | */ |
9242 | 154k | if (maxAtts > 1) { |
9243 | 129k | hashValue = xmlDictCombineHash(attr->name.hashValue, |
9244 | 129k | uriHashValue); |
9245 | 129k | res = xmlAttrHashInsert(ctxt, attrHashSize, attname, nsuri, |
9246 | 129k | hashValue, nbatts); |
9247 | 129k | if (res < 0) |
9248 | 0 | continue; |
9249 | 129k | if (res < INT_MAX) { |
9250 | 14.8k | if (aprefix == atts[res+1]) |
9251 | 3.86k | continue; |
9252 | 10.9k | xmlNsErr(ctxt, XML_NS_ERR_ATTRIBUTE_REDEFINED, |
9253 | 10.9k | "Namespaced Attribute %s in '%s' redefined\n", |
9254 | 10.9k | attname, nsuri, NULL); |
9255 | 10.9k | } |
9256 | 129k | } |
9257 | | |
9258 | 151k | xmlParserEntityCheck(ctxt, attr->expandedSize); |
9259 | | |
9260 | 151k | if ((atts == NULL) || (nbatts + 5 > maxatts)) { |
9261 | 9.24k | res = xmlCtxtGrowAttrs(ctxt); |
9262 | | |
9263 | 9.24k | maxatts = ctxt->maxatts; |
9264 | 9.24k | atts = ctxt->atts; |
9265 | | |
9266 | 9.24k | if (res < 0) { |
9267 | 9 | localname = NULL; |
9268 | 9 | goto done; |
9269 | 9 | } |
9270 | 9.24k | } |
9271 | | |
9272 | 151k | atts[nbatts++] = attname; |
9273 | 151k | atts[nbatts++] = aprefix; |
9274 | 151k | atts[nbatts++] = XML_INT_TO_PTR(nsIndex); |
9275 | 151k | atts[nbatts++] = attr->value.name; |
9276 | 151k | atts[nbatts++] = attr->valueEnd; |
9277 | | |
9278 | | #ifdef LIBXML_VALID_ENABLED |
9279 | | /* |
9280 | | * This should be moved to valid.c, but we don't keep track |
9281 | | * whether an attribute was defaulted. |
9282 | | */ |
9283 | | if ((ctxt->validate) && |
9284 | | (ctxt->standalone == 1) && |
9285 | | (attr->external != 0)) { |
9286 | | xmlValidityError(ctxt, XML_DTD_STANDALONE_DEFAULTED, |
9287 | | "standalone: attribute %s on %s defaulted " |
9288 | | "from external subset\n", |
9289 | | attname, localname); |
9290 | | } |
9291 | | #endif |
9292 | 151k | nbdef++; |
9293 | 151k | } |
9294 | 58.9k | } |
9295 | 111k | } |
9296 | | |
9297 | | /* |
9298 | | * Using a single hash table for nsUri/localName pairs cannot |
9299 | | * detect duplicate QNames reliably. The following example will |
9300 | | * only result in two namespace errors. |
9301 | | * |
9302 | | * <doc xmlns:a="a" xmlns:b="a"> |
9303 | | * <elem a:a="" b:a="" b:a=""/> |
9304 | | * </doc> |
9305 | | * |
9306 | | * If we saw more than one namespace error but no duplicate QNames |
9307 | | * were found, we have to scan for duplicate QNames. |
9308 | | */ |
9309 | 7.71M | if ((numDupErr == 0) && (numNsErr > 1)) { |
9310 | 538 | memset(ctxt->attrHash, -1, |
9311 | 538 | attrHashSize * sizeof(ctxt->attrHash[0])); |
9312 | | |
9313 | 4.41k | for (i = 0, j = 0; j < nratts; i += 5, j++) { |
9314 | 3.87k | unsigned hashValue, nameHashValue, prefixHashValue; |
9315 | 3.87k | int res; |
9316 | | |
9317 | 3.87k | aprefix = atts[i+1]; |
9318 | 3.87k | if (aprefix == NULL) |
9319 | 1.25k | continue; |
9320 | | |
9321 | 2.62k | attname = atts[i]; |
9322 | | /* Hash values always have bit 31 set, see dict.c */ |
9323 | 2.62k | nameHashValue = ctxt->attallocs[j] | 0x80000000; |
9324 | 2.62k | prefixHashValue = xmlDictComputeHash(ctxt->dict, aprefix); |
9325 | | |
9326 | 2.62k | hashValue = xmlDictCombineHash(nameHashValue, prefixHashValue); |
9327 | 2.62k | res = xmlAttrHashInsertQName(ctxt, attrHashSize, attname, |
9328 | 2.62k | aprefix, hashValue, i); |
9329 | 2.62k | if (res < INT_MAX) |
9330 | 574 | xmlErrAttributeDup(ctxt, aprefix, attname); |
9331 | 2.62k | } |
9332 | 538 | } |
9333 | | |
9334 | | /* |
9335 | | * Reconstruct attribute pointers |
9336 | | */ |
9337 | 12.1M | for (i = 0, j = 0; i < nbatts; i += 5, j++) { |
9338 | | /* namespace URI */ |
9339 | 4.42M | nsIndex = XML_PTR_TO_INT(atts[i+2]); |
9340 | 4.42M | if (nsIndex == INT_MAX) |
9341 | 4.27M | atts[i+2] = NULL; |
9342 | 146k | else if (nsIndex == INT_MAX - 1) |
9343 | 126k | atts[i+2] = ctxt->str_xml_ns; |
9344 | 20.0k | else |
9345 | 20.0k | atts[i+2] = ctxt->nsTab[nsIndex * 2 + 1]; |
9346 | | |
9347 | 4.42M | if ((j < nratts) && (ctxt->attallocs[j] & 0x80000000) == 0) { |
9348 | 4.10M | atts[i+3] = BASE_PTR + XML_PTR_TO_INT(atts[i+3]); /* value */ |
9349 | 4.10M | atts[i+4] = BASE_PTR + XML_PTR_TO_INT(atts[i+4]); /* valuend */ |
9350 | 4.10M | } |
9351 | 4.42M | } |
9352 | | |
9353 | 7.71M | uri = xmlParserNsLookupUri(ctxt, &hprefix); |
9354 | 7.71M | if ((prefix != NULL) && (uri == NULL)) { |
9355 | 389k | xmlNsErr(ctxt, XML_NS_ERR_UNDEFINED_NAMESPACE, |
9356 | 389k | "Namespace prefix %s on %s is not defined\n", |
9357 | 389k | prefix, localname, NULL); |
9358 | 389k | } |
9359 | 7.71M | *pref = prefix; |
9360 | 7.71M | *URI = uri; |
9361 | | |
9362 | | /* |
9363 | | * SAX callback |
9364 | | */ |
9365 | 7.71M | if ((ctxt->sax != NULL) && (ctxt->sax->startElementNs != NULL) && |
9366 | 7.71M | (!ctxt->disableSAX)) { |
9367 | 6.55M | if (nbNs > 0) |
9368 | 625k | ctxt->sax->startElementNs(ctxt->userData, localname, prefix, uri, |
9369 | 625k | nbNs, ctxt->nsTab + 2 * (ctxt->nsNr - nbNs), |
9370 | 625k | nbatts / 5, nbdef, atts); |
9371 | 5.92M | else |
9372 | 5.92M | ctxt->sax->startElementNs(ctxt->userData, localname, prefix, uri, |
9373 | 5.92M | 0, NULL, nbatts / 5, nbdef, atts); |
9374 | 6.55M | } |
9375 | | |
9376 | 7.71M | done: |
9377 | | /* |
9378 | | * Free allocated attribute values |
9379 | | */ |
9380 | 7.71M | if (attval != 0) { |
9381 | 381k | for (i = 0, j = 0; j < nratts; i += 5, j++) |
9382 | 231k | if (ctxt->attallocs[j] & 0x80000000) |
9383 | 162k | xmlFree((xmlChar *) atts[i+3]); |
9384 | 150k | } |
9385 | | |
9386 | 7.71M | *nbNsPtr = nbNs; |
9387 | 7.71M | return(localname); |
9388 | 7.71M | } |
9389 | | |
9390 | | /** |
9391 | | * Parse an end tag. Always consumes '</'. |
9392 | | * |
9393 | | * [42] ETag ::= '</' Name S? '>' |
9394 | | * |
9395 | | * With namespace |
9396 | | * |
9397 | | * [NS 9] ETag ::= '</' QName S? '>' |
9398 | | * @param ctxt an XML parser context |
9399 | | * @param tag the corresponding start tag |
9400 | | */ |
9401 | | |
9402 | | static void |
9403 | 1.70M | xmlParseEndTag2(xmlParserCtxtPtr ctxt, const xmlStartTag *tag) { |
9404 | 1.70M | const xmlChar *name; |
9405 | | |
9406 | 1.70M | GROW; |
9407 | 1.70M | if ((RAW != '<') || (NXT(1) != '/')) { |
9408 | 2.91k | xmlFatalErr(ctxt, XML_ERR_LTSLASH_REQUIRED, NULL); |
9409 | 2.91k | return; |
9410 | 2.91k | } |
9411 | 1.70M | SKIP(2); |
9412 | | |
9413 | 1.70M | if (tag->prefix == NULL) |
9414 | 510k | name = xmlParseNameAndCompare(ctxt, ctxt->name); |
9415 | 1.19M | else |
9416 | 1.19M | name = xmlParseQNameAndCompare(ctxt, ctxt->name, tag->prefix); |
9417 | | |
9418 | | /* |
9419 | | * We should definitely be at the ending "S? '>'" part |
9420 | | */ |
9421 | 1.70M | GROW; |
9422 | 1.70M | SKIP_BLANKS; |
9423 | 1.70M | if ((!IS_BYTE_CHAR(RAW)) || (RAW != '>')) { |
9424 | 114k | xmlFatalErr(ctxt, XML_ERR_GT_REQUIRED, NULL); |
9425 | 114k | } else |
9426 | 1.58M | NEXT1; |
9427 | | |
9428 | | /* |
9429 | | * [ WFC: Element Type Match ] |
9430 | | * The Name in an element's end-tag must match the element type in the |
9431 | | * start-tag. |
9432 | | * |
9433 | | */ |
9434 | 1.70M | if (name != (xmlChar*)1) { |
9435 | 178k | if (name == NULL) name = BAD_CAST "unparsable"; |
9436 | 178k | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NAME_MISMATCH, |
9437 | 178k | "Opening and ending tag mismatch: %s line %d and %s\n", |
9438 | 178k | ctxt->name, tag->line, name); |
9439 | 178k | } |
9440 | | |
9441 | | /* |
9442 | | * SAX: End of Tag |
9443 | | */ |
9444 | 1.70M | if ((ctxt->sax != NULL) && (ctxt->sax->endElementNs != NULL) && |
9445 | 1.70M | (!ctxt->disableSAX)) |
9446 | 1.33M | ctxt->sax->endElementNs(ctxt->userData, ctxt->name, tag->prefix, |
9447 | 1.33M | tag->URI); |
9448 | | |
9449 | 1.70M | spacePop(ctxt); |
9450 | 1.70M | if (tag->nsNr != 0) |
9451 | 601k | xmlParserNsPop(ctxt, tag->nsNr); |
9452 | 1.70M | } |
9453 | | |
9454 | | /** |
9455 | | * Parse escaped pure raw content. Always consumes '<!['. |
9456 | | * |
9457 | | * @deprecated Internal function, don't use. |
9458 | | * |
9459 | | * [18] CDSect ::= CDStart CData CDEnd |
9460 | | * |
9461 | | * [19] CDStart ::= '<![CDATA[' |
9462 | | * |
9463 | | * [20] Data ::= (Char* - (Char* ']]>' Char*)) |
9464 | | * |
9465 | | * [21] CDEnd ::= ']]>' |
9466 | | * @param ctxt an XML parser context |
9467 | | */ |
9468 | | void |
9469 | 85.9k | xmlParseCDSect(xmlParserCtxt *ctxt) { |
9470 | 85.9k | xmlChar *buf = NULL; |
9471 | 85.9k | int len = 0; |
9472 | 85.9k | int size = XML_PARSER_BUFFER_SIZE; |
9473 | 85.9k | int r, rl; |
9474 | 85.9k | int s, sl; |
9475 | 85.9k | int cur, l; |
9476 | 85.9k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
9477 | 0 | XML_MAX_HUGE_LENGTH : |
9478 | 85.9k | XML_MAX_TEXT_LENGTH; |
9479 | | |
9480 | 85.9k | if ((CUR != '<') || (NXT(1) != '!') || (NXT(2) != '[')) |
9481 | 0 | return; |
9482 | 85.9k | SKIP(3); |
9483 | | |
9484 | 85.9k | if (!CMP6(CUR_PTR, 'C', 'D', 'A', 'T', 'A', '[')) |
9485 | 0 | return; |
9486 | 85.9k | SKIP(6); |
9487 | | |
9488 | 85.9k | r = xmlCurrentCharRecover(ctxt, &rl); |
9489 | 85.9k | if (!IS_CHAR(r)) { |
9490 | 2.84k | xmlFatalErr(ctxt, XML_ERR_CDATA_NOT_FINISHED, NULL); |
9491 | 2.84k | goto out; |
9492 | 2.84k | } |
9493 | 83.0k | NEXTL(rl); |
9494 | 83.0k | s = xmlCurrentCharRecover(ctxt, &sl); |
9495 | 83.0k | if (!IS_CHAR(s)) { |
9496 | 10.1k | xmlFatalErr(ctxt, XML_ERR_CDATA_NOT_FINISHED, NULL); |
9497 | 10.1k | goto out; |
9498 | 10.1k | } |
9499 | 72.9k | NEXTL(sl); |
9500 | 72.9k | cur = xmlCurrentCharRecover(ctxt, &l); |
9501 | 72.9k | buf = xmlMalloc(size); |
9502 | 72.9k | if (buf == NULL) { |
9503 | 3 | xmlErrMemory(ctxt); |
9504 | 3 | goto out; |
9505 | 3 | } |
9506 | 44.4M | while (IS_CHAR(cur) && |
9507 | 44.4M | ((r != ']') || (s != ']') || (cur != '>'))) { |
9508 | 44.3M | if (len + 5 >= size) { |
9509 | 7.53k | xmlChar *tmp; |
9510 | 7.53k | int newSize; |
9511 | | |
9512 | 7.53k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
9513 | 7.53k | if (newSize < 0) { |
9514 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_CDATA_NOT_FINISHED, |
9515 | 0 | "CData section too big found\n"); |
9516 | 0 | goto out; |
9517 | 0 | } |
9518 | 7.53k | tmp = xmlRealloc(buf, newSize); |
9519 | 7.53k | if (tmp == NULL) { |
9520 | 1 | xmlErrMemory(ctxt); |
9521 | 1 | goto out; |
9522 | 1 | } |
9523 | 7.53k | buf = tmp; |
9524 | 7.53k | size = newSize; |
9525 | 7.53k | } |
9526 | 44.3M | COPY_BUF(buf, len, r); |
9527 | 44.3M | r = s; |
9528 | 44.3M | rl = sl; |
9529 | 44.3M | s = cur; |
9530 | 44.3M | sl = l; |
9531 | 44.3M | NEXTL(l); |
9532 | 44.3M | cur = xmlCurrentCharRecover(ctxt, &l); |
9533 | 44.3M | } |
9534 | 72.8k | buf[len] = 0; |
9535 | 72.8k | if (cur != '>') { |
9536 | 5.59k | xmlFatalErrMsgStr(ctxt, XML_ERR_CDATA_NOT_FINISHED, |
9537 | 5.59k | "CData section not finished\n%.50s\n", buf); |
9538 | 5.59k | goto out; |
9539 | 5.59k | } |
9540 | 67.3k | NEXTL(l); |
9541 | | |
9542 | | /* |
9543 | | * OK the buffer is to be consumed as cdata. |
9544 | | */ |
9545 | 67.3k | if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { |
9546 | 65.7k | if ((ctxt->sax->cdataBlock != NULL) && |
9547 | 65.7k | ((ctxt->options & XML_PARSE_NOCDATA) == 0)) { |
9548 | 3.11k | ctxt->sax->cdataBlock(ctxt->userData, buf, len); |
9549 | 62.5k | } else if (ctxt->sax->characters != NULL) { |
9550 | 62.5k | ctxt->sax->characters(ctxt->userData, buf, len); |
9551 | 62.5k | } |
9552 | 65.7k | } |
9553 | | |
9554 | 85.9k | out: |
9555 | 85.9k | xmlFree(buf); |
9556 | 85.9k | } |
9557 | | |
9558 | | /** |
9559 | | * Parse a content sequence. Stops at EOF or '</'. Leaves checking of |
9560 | | * unexpected EOF to the caller. |
9561 | | * |
9562 | | * @param ctxt an XML parser context |
9563 | | */ |
9564 | | |
9565 | | static void |
9566 | 662k | xmlParseContentInternal(xmlParserCtxtPtr ctxt) { |
9567 | 662k | int oldNameNr = ctxt->nameNr; |
9568 | 662k | int oldSpaceNr = ctxt->spaceNr; |
9569 | 662k | int oldNodeNr = ctxt->nodeNr; |
9570 | | |
9571 | 662k | GROW; |
9572 | 24.0M | while ((ctxt->input->cur < ctxt->input->end) && |
9573 | 24.0M | (PARSER_STOPPED(ctxt) == 0)) { |
9574 | 23.9M | const xmlChar *cur = ctxt->input->cur; |
9575 | | |
9576 | | /* |
9577 | | * First case : a Processing Instruction. |
9578 | | */ |
9579 | 23.9M | if ((*cur == '<') && (cur[1] == '?')) { |
9580 | 60.9k | xmlParsePI(ctxt); |
9581 | 60.9k | } |
9582 | | |
9583 | | /* |
9584 | | * Second case : a CDSection |
9585 | | */ |
9586 | | /* 2.6.0 test was *cur not RAW */ |
9587 | 23.9M | else if (CMP9(CUR_PTR, '<', '!', '[', 'C', 'D', 'A', 'T', 'A', '[')) { |
9588 | 85.9k | xmlParseCDSect(ctxt); |
9589 | 85.9k | } |
9590 | | |
9591 | | /* |
9592 | | * Third case : a comment |
9593 | | */ |
9594 | 23.8M | else if ((*cur == '<') && (NXT(1) == '!') && |
9595 | 23.8M | (NXT(2) == '-') && (NXT(3) == '-')) { |
9596 | 148k | xmlParseComment(ctxt); |
9597 | 148k | } |
9598 | | |
9599 | | /* |
9600 | | * Fourth case : a sub-element. |
9601 | | */ |
9602 | 23.6M | else if (*cur == '<') { |
9603 | 9.29M | if (NXT(1) == '/') { |
9604 | 1.70M | if (ctxt->nameNr <= oldNameNr) |
9605 | 606k | break; |
9606 | 1.09M | xmlParseElementEnd(ctxt); |
9607 | 7.59M | } else { |
9608 | 7.59M | xmlParseElementStart(ctxt); |
9609 | 7.59M | } |
9610 | 9.29M | } |
9611 | | |
9612 | | /* |
9613 | | * Fifth case : a reference. If if has not been resolved, |
9614 | | * parsing returns it's Name, create the node |
9615 | | */ |
9616 | | |
9617 | 14.3M | else if (*cur == '&') { |
9618 | 530k | xmlParseReference(ctxt); |
9619 | 530k | } |
9620 | | |
9621 | | /* |
9622 | | * Last case, text. Note that References are handled directly. |
9623 | | */ |
9624 | 13.8M | else { |
9625 | 13.8M | xmlParseCharDataInternal(ctxt, 0); |
9626 | 13.8M | } |
9627 | | |
9628 | 23.3M | SHRINK; |
9629 | 23.3M | GROW; |
9630 | 23.3M | } |
9631 | | |
9632 | 662k | if ((ctxt->nameNr > oldNameNr) && |
9633 | 662k | (ctxt->input->cur >= ctxt->input->end) && |
9634 | 662k | (ctxt->wellFormed)) { |
9635 | 583 | const xmlChar *name = ctxt->nameTab[ctxt->nameNr - 1]; |
9636 | 583 | int line = ctxt->pushTab[ctxt->nameNr - 1].line; |
9637 | 583 | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NOT_FINISHED, |
9638 | 583 | "Premature end of data in tag %s line %d\n", |
9639 | 583 | name, line, NULL); |
9640 | 583 | } |
9641 | | |
9642 | | /* |
9643 | | * Clean up in error case |
9644 | | */ |
9645 | | |
9646 | 844k | while (ctxt->nodeNr > oldNodeNr) |
9647 | 181k | nodePop(ctxt); |
9648 | | |
9649 | 1.10M | while (ctxt->nameNr > oldNameNr) { |
9650 | 443k | xmlStartTag *tag = &ctxt->pushTab[ctxt->nameNr - 1]; |
9651 | | |
9652 | 443k | if (tag->nsNr != 0) |
9653 | 44.2k | xmlParserNsPop(ctxt, tag->nsNr); |
9654 | | |
9655 | 443k | namePop(ctxt); |
9656 | 443k | } |
9657 | | |
9658 | 1.10M | while (ctxt->spaceNr > oldSpaceNr) |
9659 | 443k | spacePop(ctxt); |
9660 | 662k | } |
9661 | | |
9662 | | /** |
9663 | | * Parse XML element content. This is useful if you're only interested |
9664 | | * in custom SAX callbacks. If you want a node list, use |
9665 | | * #xmlCtxtParseContent. |
9666 | | * |
9667 | | * @param ctxt an XML parser context |
9668 | | */ |
9669 | | void |
9670 | 0 | xmlParseContent(xmlParserCtxt *ctxt) { |
9671 | 0 | if ((ctxt == NULL) || (ctxt->input == NULL)) |
9672 | 0 | return; |
9673 | | |
9674 | 0 | xmlCtxtInitializeLate(ctxt); |
9675 | |
|
9676 | 0 | xmlParseContentInternal(ctxt); |
9677 | |
|
9678 | 0 | xmlParserCheckEOF(ctxt, XML_ERR_NOT_WELL_BALANCED); |
9679 | 0 | } |
9680 | | |
9681 | | /** |
9682 | | * Parse an XML element |
9683 | | * |
9684 | | * @deprecated Internal function, don't use. |
9685 | | * |
9686 | | * [39] element ::= EmptyElemTag | STag content ETag |
9687 | | * |
9688 | | * [ WFC: Element Type Match ] |
9689 | | * The Name in an element's end-tag must match the element type in the |
9690 | | * start-tag. |
9691 | | * |
9692 | | * @param ctxt an XML parser context |
9693 | | */ |
9694 | | |
9695 | | void |
9696 | 698k | xmlParseElement(xmlParserCtxt *ctxt) { |
9697 | 698k | if (xmlParseElementStart(ctxt) != 0) |
9698 | 45.0k | return; |
9699 | | |
9700 | 653k | xmlParseContentInternal(ctxt); |
9701 | | |
9702 | 653k | if (ctxt->input->cur >= ctxt->input->end) { |
9703 | 43.5k | if (ctxt->wellFormed) { |
9704 | 566 | const xmlChar *name = ctxt->nameTab[ctxt->nameNr - 1]; |
9705 | 566 | int line = ctxt->pushTab[ctxt->nameNr - 1].line; |
9706 | 566 | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NOT_FINISHED, |
9707 | 566 | "Premature end of data in tag %s line %d\n", |
9708 | 566 | name, line, NULL); |
9709 | 566 | } |
9710 | 43.5k | return; |
9711 | 43.5k | } |
9712 | | |
9713 | 609k | xmlParseElementEnd(ctxt); |
9714 | 609k | } |
9715 | | |
9716 | | /** |
9717 | | * Parse the start of an XML element. Returns -1 in case of error, 0 if an |
9718 | | * opening tag was parsed, 1 if an empty element was parsed. |
9719 | | * |
9720 | | * Always consumes '<'. |
9721 | | * |
9722 | | * @param ctxt an XML parser context |
9723 | | */ |
9724 | | static int |
9725 | 8.28M | xmlParseElementStart(xmlParserCtxtPtr ctxt) { |
9726 | 8.28M | int maxDepth = (ctxt->options & XML_PARSE_HUGE) ? 2048 : 256; |
9727 | 8.28M | const xmlChar *name; |
9728 | 8.28M | const xmlChar *prefix = NULL; |
9729 | 8.28M | const xmlChar *URI = NULL; |
9730 | 8.28M | xmlParserNodeInfo node_info; |
9731 | 8.28M | int line; |
9732 | 8.28M | xmlNodePtr cur; |
9733 | 8.28M | int nbNs = 0; |
9734 | | |
9735 | 8.28M | if (ctxt->nameNr > maxDepth) { |
9736 | 370 | xmlFatalErrMsgInt(ctxt, XML_ERR_RESOURCE_LIMIT, |
9737 | 370 | "Excessive depth in document: %d use XML_PARSE_HUGE option\n", |
9738 | 370 | ctxt->nameNr); |
9739 | 370 | xmlHaltParser(ctxt); |
9740 | 370 | return(-1); |
9741 | 370 | } |
9742 | | |
9743 | | /* Capture start position */ |
9744 | 8.28M | if (ctxt->record_info) { |
9745 | 0 | node_info.begin_pos = ctxt->input->consumed + |
9746 | 0 | (CUR_PTR - ctxt->input->base); |
9747 | 0 | node_info.begin_line = ctxt->input->line; |
9748 | 0 | } |
9749 | | |
9750 | 8.28M | if (ctxt->spaceNr == 0) |
9751 | 0 | spacePush(ctxt, -1); |
9752 | 8.28M | else if (*ctxt->space == -2) |
9753 | 0 | spacePush(ctxt, -1); |
9754 | 8.28M | else |
9755 | 8.28M | spacePush(ctxt, *ctxt->space); |
9756 | | |
9757 | 8.28M | line = ctxt->input->line; |
9758 | | #ifdef LIBXML_SAX1_ENABLED |
9759 | | if (ctxt->sax2) |
9760 | | #endif /* LIBXML_SAX1_ENABLED */ |
9761 | 8.28M | name = xmlParseStartTag2(ctxt, &prefix, &URI, &nbNs); |
9762 | | #ifdef LIBXML_SAX1_ENABLED |
9763 | | else |
9764 | | name = xmlParseStartTag(ctxt); |
9765 | | #endif /* LIBXML_SAX1_ENABLED */ |
9766 | 8.28M | if (name == NULL) { |
9767 | 576k | spacePop(ctxt); |
9768 | 576k | return(-1); |
9769 | 576k | } |
9770 | 7.71M | nameNsPush(ctxt, name, prefix, URI, line, nbNs); |
9771 | 7.71M | cur = ctxt->node; |
9772 | | |
9773 | | #ifdef LIBXML_VALID_ENABLED |
9774 | | /* |
9775 | | * [ VC: Root Element Type ] |
9776 | | * The Name in the document type declaration must match the element |
9777 | | * type of the root element. |
9778 | | */ |
9779 | | if (ctxt->validate && ctxt->wellFormed && ctxt->myDoc && |
9780 | | ctxt->node && (ctxt->node == ctxt->myDoc->children)) |
9781 | | ctxt->valid &= xmlValidateRoot(&ctxt->vctxt, ctxt->myDoc); |
9782 | | #endif /* LIBXML_VALID_ENABLED */ |
9783 | | |
9784 | | /* |
9785 | | * Check for an Empty Element. |
9786 | | */ |
9787 | 7.71M | if ((RAW == '/') && (NXT(1) == '>')) { |
9788 | 2.66M | SKIP(2); |
9789 | 2.66M | if (ctxt->sax2) { |
9790 | 2.66M | if ((ctxt->sax != NULL) && (ctxt->sax->endElementNs != NULL) && |
9791 | 2.66M | (!ctxt->disableSAX)) |
9792 | 2.50M | ctxt->sax->endElementNs(ctxt->userData, name, prefix, URI); |
9793 | | #ifdef LIBXML_SAX1_ENABLED |
9794 | | } else { |
9795 | | if ((ctxt->sax != NULL) && (ctxt->sax->endElement != NULL) && |
9796 | | (!ctxt->disableSAX)) |
9797 | | ctxt->sax->endElement(ctxt->userData, name); |
9798 | | #endif /* LIBXML_SAX1_ENABLED */ |
9799 | 2.66M | } |
9800 | 2.66M | namePop(ctxt); |
9801 | 2.66M | spacePop(ctxt); |
9802 | 2.66M | if (nbNs > 0) |
9803 | 14.1k | xmlParserNsPop(ctxt, nbNs); |
9804 | 2.66M | if (cur != NULL && ctxt->record_info) { |
9805 | 0 | node_info.node = cur; |
9806 | 0 | node_info.end_pos = ctxt->input->consumed + |
9807 | 0 | (CUR_PTR - ctxt->input->base); |
9808 | 0 | node_info.end_line = ctxt->input->line; |
9809 | 0 | xmlParserAddNodeInfo(ctxt, &node_info); |
9810 | 0 | } |
9811 | 2.66M | return(1); |
9812 | 2.66M | } |
9813 | 5.05M | if (RAW == '>') { |
9814 | 2.19M | NEXT1; |
9815 | 2.19M | if (cur != NULL && ctxt->record_info) { |
9816 | 0 | node_info.node = cur; |
9817 | 0 | node_info.end_pos = 0; |
9818 | 0 | node_info.end_line = 0; |
9819 | 0 | xmlParserAddNodeInfo(ctxt, &node_info); |
9820 | 0 | } |
9821 | 2.85M | } else { |
9822 | 2.85M | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_GT_REQUIRED, |
9823 | 2.85M | "Couldn't find end of Start Tag %s line %d\n", |
9824 | 2.85M | name, line, NULL); |
9825 | | |
9826 | | /* |
9827 | | * end of parsing of this node. |
9828 | | */ |
9829 | 2.85M | nodePop(ctxt); |
9830 | 2.85M | namePop(ctxt); |
9831 | 2.85M | spacePop(ctxt); |
9832 | 2.85M | if (nbNs > 0) |
9833 | 79.2k | xmlParserNsPop(ctxt, nbNs); |
9834 | 2.85M | return(-1); |
9835 | 2.85M | } |
9836 | | |
9837 | 2.19M | return(0); |
9838 | 5.05M | } |
9839 | | |
9840 | | /** |
9841 | | * Parse the end of an XML element. Always consumes '</'. |
9842 | | * |
9843 | | * @param ctxt an XML parser context |
9844 | | */ |
9845 | | static void |
9846 | 1.70M | xmlParseElementEnd(xmlParserCtxtPtr ctxt) { |
9847 | 1.70M | xmlNodePtr cur = ctxt->node; |
9848 | | |
9849 | 1.70M | if (ctxt->nameNr <= 0) { |
9850 | 731 | if ((RAW == '<') && (NXT(1) == '/')) |
9851 | 3 | SKIP(2); |
9852 | 731 | return; |
9853 | 731 | } |
9854 | | |
9855 | | /* |
9856 | | * parse the end of tag: '</' should be here. |
9857 | | */ |
9858 | 1.70M | if (ctxt->sax2) { |
9859 | 1.70M | xmlParseEndTag2(ctxt, &ctxt->pushTab[ctxt->nameNr - 1]); |
9860 | 1.70M | namePop(ctxt); |
9861 | 1.70M | } |
9862 | | #ifdef LIBXML_SAX1_ENABLED |
9863 | | else |
9864 | | xmlParseEndTag1(ctxt, 0); |
9865 | | #endif /* LIBXML_SAX1_ENABLED */ |
9866 | | |
9867 | | /* |
9868 | | * Capture end position |
9869 | | */ |
9870 | 1.70M | if (cur != NULL && ctxt->record_info) { |
9871 | 0 | xmlParserNodeInfoPtr node_info; |
9872 | |
|
9873 | 0 | node_info = (xmlParserNodeInfoPtr) xmlParserFindNodeInfo(ctxt, cur); |
9874 | 0 | if (node_info != NULL) { |
9875 | 0 | node_info->end_pos = ctxt->input->consumed + |
9876 | 0 | (CUR_PTR - ctxt->input->base); |
9877 | 0 | node_info->end_line = ctxt->input->line; |
9878 | 0 | } |
9879 | 0 | } |
9880 | 1.70M | } |
9881 | | |
9882 | | /** |
9883 | | * Parse the XML version value. |
9884 | | * |
9885 | | * @deprecated Internal function, don't use. |
9886 | | * |
9887 | | * [26] VersionNum ::= '1.' [0-9]+ |
9888 | | * |
9889 | | * In practice allow [0-9].[0-9]+ at that level |
9890 | | * |
9891 | | * @param ctxt an XML parser context |
9892 | | * @returns the string giving the XML version number, or NULL |
9893 | | */ |
9894 | | xmlChar * |
9895 | 174k | xmlParseVersionNum(xmlParserCtxt *ctxt) { |
9896 | 174k | xmlChar *buf = NULL; |
9897 | 174k | int len = 0; |
9898 | 174k | int size = 10; |
9899 | 174k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
9900 | 0 | XML_MAX_TEXT_LENGTH : |
9901 | 174k | XML_MAX_NAME_LENGTH; |
9902 | 174k | xmlChar cur; |
9903 | | |
9904 | 174k | buf = xmlMalloc(size); |
9905 | 174k | if (buf == NULL) { |
9906 | 11 | xmlErrMemory(ctxt); |
9907 | 11 | return(NULL); |
9908 | 11 | } |
9909 | 174k | cur = CUR; |
9910 | 174k | if (!((cur >= '0') && (cur <= '9'))) { |
9911 | 1.90k | xmlFree(buf); |
9912 | 1.90k | return(NULL); |
9913 | 1.90k | } |
9914 | 172k | buf[len++] = cur; |
9915 | 172k | NEXT; |
9916 | 172k | cur=CUR; |
9917 | 172k | if (cur != '.') { |
9918 | 3.44k | xmlFree(buf); |
9919 | 3.44k | return(NULL); |
9920 | 3.44k | } |
9921 | 168k | buf[len++] = cur; |
9922 | 168k | NEXT; |
9923 | 168k | cur=CUR; |
9924 | 327k | while ((cur >= '0') && (cur <= '9')) { |
9925 | 158k | if (len + 1 >= size) { |
9926 | 3.05k | xmlChar *tmp; |
9927 | 3.05k | int newSize; |
9928 | | |
9929 | 3.05k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
9930 | 3.05k | if (newSize < 0) { |
9931 | 0 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "VersionNum"); |
9932 | 0 | xmlFree(buf); |
9933 | 0 | return(NULL); |
9934 | 0 | } |
9935 | 3.05k | tmp = xmlRealloc(buf, newSize); |
9936 | 3.05k | if (tmp == NULL) { |
9937 | 1 | xmlErrMemory(ctxt); |
9938 | 1 | xmlFree(buf); |
9939 | 1 | return(NULL); |
9940 | 1 | } |
9941 | 3.05k | buf = tmp; |
9942 | 3.05k | size = newSize; |
9943 | 3.05k | } |
9944 | 158k | buf[len++] = cur; |
9945 | 158k | NEXT; |
9946 | 158k | cur=CUR; |
9947 | 158k | } |
9948 | 168k | buf[len] = 0; |
9949 | 168k | return(buf); |
9950 | 168k | } |
9951 | | |
9952 | | /** |
9953 | | * Parse the XML version. |
9954 | | * |
9955 | | * @deprecated Internal function, don't use. |
9956 | | * |
9957 | | * [24] VersionInfo ::= S 'version' Eq (' VersionNum ' | " VersionNum ") |
9958 | | * |
9959 | | * [25] Eq ::= S? '=' S? |
9960 | | * |
9961 | | * @param ctxt an XML parser context |
9962 | | * @returns the version string, e.g. "1.0" |
9963 | | */ |
9964 | | |
9965 | | xmlChar * |
9966 | 199k | xmlParseVersionInfo(xmlParserCtxt *ctxt) { |
9967 | 199k | xmlChar *version = NULL; |
9968 | | |
9969 | 199k | if (CMP7(CUR_PTR, 'v', 'e', 'r', 's', 'i', 'o', 'n')) { |
9970 | 177k | SKIP(7); |
9971 | 177k | SKIP_BLANKS; |
9972 | 177k | if (RAW != '=') { |
9973 | 1.82k | xmlFatalErr(ctxt, XML_ERR_EQUAL_REQUIRED, NULL); |
9974 | 1.82k | return(NULL); |
9975 | 1.82k | } |
9976 | 175k | NEXT; |
9977 | 175k | SKIP_BLANKS; |
9978 | 175k | if (RAW == '"') { |
9979 | 168k | NEXT; |
9980 | 168k | version = xmlParseVersionNum(ctxt); |
9981 | 168k | if (RAW != '"') { |
9982 | 5.19k | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
9983 | 5.19k | } else |
9984 | 163k | NEXT; |
9985 | 168k | } else if (RAW == '\''){ |
9986 | 5.34k | NEXT; |
9987 | 5.34k | version = xmlParseVersionNum(ctxt); |
9988 | 5.34k | if (RAW != '\'') { |
9989 | 1.20k | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
9990 | 1.20k | } else |
9991 | 4.14k | NEXT; |
9992 | 5.34k | } else { |
9993 | 1.36k | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_STARTED, NULL); |
9994 | 1.36k | } |
9995 | 175k | } |
9996 | 197k | return(version); |
9997 | 199k | } |
9998 | | |
9999 | | /** |
10000 | | * Parse the XML encoding name |
10001 | | * |
10002 | | * @deprecated Internal function, don't use. |
10003 | | * |
10004 | | * [81] EncName ::= [A-Za-z] ([A-Za-z0-9._] | '-')* |
10005 | | * |
10006 | | * @param ctxt an XML parser context |
10007 | | * @returns the encoding name value or NULL |
10008 | | */ |
10009 | | xmlChar * |
10010 | 136k | xmlParseEncName(xmlParserCtxt *ctxt) { |
10011 | 136k | xmlChar *buf = NULL; |
10012 | 136k | int len = 0; |
10013 | 136k | int size = 10; |
10014 | 136k | int maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
10015 | 0 | XML_MAX_TEXT_LENGTH : |
10016 | 136k | XML_MAX_NAME_LENGTH; |
10017 | 136k | xmlChar cur; |
10018 | | |
10019 | 136k | cur = CUR; |
10020 | 136k | if (((cur >= 'a') && (cur <= 'z')) || |
10021 | 136k | ((cur >= 'A') && (cur <= 'Z'))) { |
10022 | 134k | buf = xmlMalloc(size); |
10023 | 134k | if (buf == NULL) { |
10024 | 13 | xmlErrMemory(ctxt); |
10025 | 13 | return(NULL); |
10026 | 13 | } |
10027 | | |
10028 | 134k | buf[len++] = cur; |
10029 | 134k | NEXT; |
10030 | 134k | cur = CUR; |
10031 | 21.8M | while (((cur >= 'a') && (cur <= 'z')) || |
10032 | 21.8M | ((cur >= 'A') && (cur <= 'Z')) || |
10033 | 21.8M | ((cur >= '0') && (cur <= '9')) || |
10034 | 21.8M | (cur == '.') || (cur == '_') || |
10035 | 21.8M | (cur == '-')) { |
10036 | 21.7M | if (len + 1 >= size) { |
10037 | 90.3k | xmlChar *tmp; |
10038 | 90.3k | int newSize; |
10039 | | |
10040 | 90.3k | newSize = xmlGrowCapacity(size, 1, 1, maxLength); |
10041 | 90.3k | if (newSize < 0) { |
10042 | 220 | xmlFatalErr(ctxt, XML_ERR_NAME_TOO_LONG, "EncName"); |
10043 | 220 | xmlFree(buf); |
10044 | 220 | return(NULL); |
10045 | 220 | } |
10046 | 90.0k | tmp = xmlRealloc(buf, newSize); |
10047 | 90.0k | if (tmp == NULL) { |
10048 | 3 | xmlErrMemory(ctxt); |
10049 | 3 | xmlFree(buf); |
10050 | 3 | return(NULL); |
10051 | 3 | } |
10052 | 90.0k | buf = tmp; |
10053 | 90.0k | size = newSize; |
10054 | 90.0k | } |
10055 | 21.7M | buf[len++] = cur; |
10056 | 21.7M | NEXT; |
10057 | 21.7M | cur = CUR; |
10058 | 21.7M | } |
10059 | 134k | buf[len] = 0; |
10060 | 134k | } else { |
10061 | 1.64k | xmlFatalErr(ctxt, XML_ERR_ENCODING_NAME, NULL); |
10062 | 1.64k | } |
10063 | 135k | return(buf); |
10064 | 136k | } |
10065 | | |
10066 | | /** |
10067 | | * Parse the XML encoding declaration |
10068 | | * |
10069 | | * @deprecated Internal function, don't use. |
10070 | | * |
10071 | | * [80] EncodingDecl ::= S 'encoding' Eq ('"' EncName '"' | |
10072 | | * "'" EncName "'") |
10073 | | * |
10074 | | * this setups the conversion filters. |
10075 | | * |
10076 | | * @param ctxt an XML parser context |
10077 | | * @returns the encoding value or NULL |
10078 | | */ |
10079 | | |
10080 | | const xmlChar * |
10081 | 173k | xmlParseEncodingDecl(xmlParserCtxt *ctxt) { |
10082 | 173k | xmlChar *encoding = NULL; |
10083 | | |
10084 | 173k | SKIP_BLANKS; |
10085 | 173k | if (CMP8(CUR_PTR, 'e', 'n', 'c', 'o', 'd', 'i', 'n', 'g') == 0) |
10086 | 36.7k | return(NULL); |
10087 | | |
10088 | 137k | SKIP(8); |
10089 | 137k | SKIP_BLANKS; |
10090 | 137k | if (RAW != '=') { |
10091 | 668 | xmlFatalErr(ctxt, XML_ERR_EQUAL_REQUIRED, NULL); |
10092 | 668 | return(NULL); |
10093 | 668 | } |
10094 | 136k | NEXT; |
10095 | 136k | SKIP_BLANKS; |
10096 | 136k | if (RAW == '"') { |
10097 | 134k | NEXT; |
10098 | 134k | encoding = xmlParseEncName(ctxt); |
10099 | 134k | if (RAW != '"') { |
10100 | 3.16k | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10101 | 3.16k | xmlFree(encoding); |
10102 | 3.16k | return(NULL); |
10103 | 3.16k | } else |
10104 | 130k | NEXT; |
10105 | 134k | } else if (RAW == '\''){ |
10106 | 2.01k | NEXT; |
10107 | 2.01k | encoding = xmlParseEncName(ctxt); |
10108 | 2.01k | if (RAW != '\'') { |
10109 | 222 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10110 | 222 | xmlFree(encoding); |
10111 | 222 | return(NULL); |
10112 | 222 | } else |
10113 | 1.79k | NEXT; |
10114 | 2.01k | } else { |
10115 | 497 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_STARTED, NULL); |
10116 | 497 | } |
10117 | | |
10118 | 133k | if (encoding == NULL) |
10119 | 501 | return(NULL); |
10120 | | |
10121 | 132k | xmlSetDeclaredEncoding(ctxt, encoding); |
10122 | | |
10123 | 132k | return(ctxt->encoding); |
10124 | 133k | } |
10125 | | |
10126 | | /** |
10127 | | * Parse the XML standalone declaration |
10128 | | * |
10129 | | * @deprecated Internal function, don't use. |
10130 | | * |
10131 | | * [32] SDDecl ::= S 'standalone' Eq |
10132 | | * (("'" ('yes' | 'no') "'") | ('"' ('yes' | 'no')'"')) |
10133 | | * |
10134 | | * [ VC: Standalone Document Declaration ] |
10135 | | * TODO The standalone document declaration must have the value "no" |
10136 | | * if any external markup declarations contain declarations of: |
10137 | | * - attributes with default values, if elements to which these |
10138 | | * attributes apply appear in the document without specifications |
10139 | | * of values for these attributes, or |
10140 | | * - entities (other than amp, lt, gt, apos, quot), if references |
10141 | | * to those entities appear in the document, or |
10142 | | * - attributes with values subject to normalization, where the |
10143 | | * attribute appears in the document with a value which will change |
10144 | | * as a result of normalization, or |
10145 | | * - element types with element content, if white space occurs directly |
10146 | | * within any instance of those types. |
10147 | | * |
10148 | | * @param ctxt an XML parser context |
10149 | | * @returns |
10150 | | * 1 if standalone="yes" |
10151 | | * 0 if standalone="no" |
10152 | | * -2 if standalone attribute is missing or invalid |
10153 | | * (A standalone value of -2 means that the XML declaration was found, |
10154 | | * but no value was specified for the standalone attribute). |
10155 | | */ |
10156 | | |
10157 | | int |
10158 | 82.0k | xmlParseSDDecl(xmlParserCtxt *ctxt) { |
10159 | 82.0k | int standalone = -2; |
10160 | | |
10161 | 82.0k | SKIP_BLANKS; |
10162 | 82.0k | if (CMP10(CUR_PTR, 's', 't', 'a', 'n', 'd', 'a', 'l', 'o', 'n', 'e')) { |
10163 | 13.8k | SKIP(10); |
10164 | 13.8k | SKIP_BLANKS; |
10165 | 13.8k | if (RAW != '=') { |
10166 | 1.67k | xmlFatalErr(ctxt, XML_ERR_EQUAL_REQUIRED, NULL); |
10167 | 1.67k | return(standalone); |
10168 | 1.67k | } |
10169 | 12.2k | NEXT; |
10170 | 12.2k | SKIP_BLANKS; |
10171 | 12.2k | if (RAW == '\''){ |
10172 | 2.55k | NEXT; |
10173 | 2.55k | if ((RAW == 'n') && (NXT(1) == 'o')) { |
10174 | 1.07k | standalone = 0; |
10175 | 1.07k | SKIP(2); |
10176 | 1.47k | } else if ((RAW == 'y') && (NXT(1) == 'e') && |
10177 | 1.47k | (NXT(2) == 's')) { |
10178 | 532 | standalone = 1; |
10179 | 532 | SKIP(3); |
10180 | 944 | } else { |
10181 | 944 | xmlFatalErr(ctxt, XML_ERR_STANDALONE_VALUE, NULL); |
10182 | 944 | } |
10183 | 2.55k | if (RAW != '\'') { |
10184 | 1.97k | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10185 | 1.97k | } else |
10186 | 571 | NEXT; |
10187 | 9.65k | } else if (RAW == '"'){ |
10188 | 9.12k | NEXT; |
10189 | 9.12k | if ((RAW == 'n') && (NXT(1) == 'o')) { |
10190 | 4.63k | standalone = 0; |
10191 | 4.63k | SKIP(2); |
10192 | 4.63k | } else if ((RAW == 'y') && (NXT(1) == 'e') && |
10193 | 4.49k | (NXT(2) == 's')) { |
10194 | 470 | standalone = 1; |
10195 | 470 | SKIP(3); |
10196 | 4.02k | } else { |
10197 | 4.02k | xmlFatalErr(ctxt, XML_ERR_STANDALONE_VALUE, NULL); |
10198 | 4.02k | } |
10199 | 9.12k | if (RAW != '"') { |
10200 | 4.46k | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_CLOSED, NULL); |
10201 | 4.46k | } else |
10202 | 4.66k | NEXT; |
10203 | 9.12k | } else { |
10204 | 528 | xmlFatalErr(ctxt, XML_ERR_STRING_NOT_STARTED, NULL); |
10205 | 528 | } |
10206 | 12.2k | } |
10207 | 80.3k | return(standalone); |
10208 | 82.0k | } |
10209 | | |
10210 | | /** |
10211 | | * Parse an XML declaration header |
10212 | | * |
10213 | | * @deprecated Internal function, don't use. |
10214 | | * |
10215 | | * [23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl? SDDecl? S? '?>' |
10216 | | * @param ctxt an XML parser context |
10217 | | */ |
10218 | | |
10219 | | void |
10220 | 184k | xmlParseXMLDecl(xmlParserCtxt *ctxt) { |
10221 | 184k | xmlChar *version; |
10222 | | |
10223 | | /* |
10224 | | * This value for standalone indicates that the document has an |
10225 | | * XML declaration but it does not have a standalone attribute. |
10226 | | * It will be overwritten later if a standalone attribute is found. |
10227 | | */ |
10228 | | |
10229 | 184k | ctxt->standalone = -2; |
10230 | | |
10231 | | /* |
10232 | | * We know that '<?xml' is here. |
10233 | | */ |
10234 | 184k | SKIP(5); |
10235 | | |
10236 | 184k | if (!IS_BLANK_CH(RAW)) { |
10237 | 0 | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, |
10238 | 0 | "Blank needed after '<?xml'\n"); |
10239 | 0 | } |
10240 | 184k | SKIP_BLANKS; |
10241 | | |
10242 | | /* |
10243 | | * We must have the VersionInfo here. |
10244 | | */ |
10245 | 184k | version = xmlParseVersionInfo(ctxt); |
10246 | 184k | if (version == NULL) { |
10247 | 25.2k | xmlFatalErr(ctxt, XML_ERR_VERSION_MISSING, NULL); |
10248 | 159k | } else { |
10249 | 159k | if (!xmlStrEqual(version, (const xmlChar *) XML_DEFAULT_VERSION)) { |
10250 | | /* |
10251 | | * Changed here for XML-1.0 5th edition |
10252 | | */ |
10253 | 59.0k | if (ctxt->options & XML_PARSE_OLD10) { |
10254 | 0 | xmlFatalErrMsgStr(ctxt, XML_ERR_UNKNOWN_VERSION, |
10255 | 0 | "Unsupported version '%s'\n", |
10256 | 0 | version); |
10257 | 59.0k | } else { |
10258 | 59.0k | if ((version[0] == '1') && ((version[1] == '.'))) { |
10259 | 50.4k | xmlWarningMsg(ctxt, XML_WAR_UNKNOWN_VERSION, |
10260 | 50.4k | "Unsupported version '%s'\n", |
10261 | 50.4k | version, NULL); |
10262 | 50.4k | } else { |
10263 | 8.61k | xmlFatalErrMsgStr(ctxt, XML_ERR_UNKNOWN_VERSION, |
10264 | 8.61k | "Unsupported version '%s'\n", |
10265 | 8.61k | version); |
10266 | 8.61k | } |
10267 | 59.0k | } |
10268 | 59.0k | } |
10269 | 159k | if (ctxt->version != NULL) |
10270 | 0 | xmlFree(ctxt->version); |
10271 | 159k | ctxt->version = version; |
10272 | 159k | } |
10273 | | |
10274 | | /* |
10275 | | * We may have the encoding declaration |
10276 | | */ |
10277 | 184k | if (!IS_BLANK_CH(RAW)) { |
10278 | 51.1k | if ((RAW == '?') && (NXT(1) == '>')) { |
10279 | 25.8k | SKIP(2); |
10280 | 25.8k | return; |
10281 | 25.8k | } |
10282 | 25.3k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Blank needed here\n"); |
10283 | 25.3k | } |
10284 | 158k | xmlParseEncodingDecl(ctxt); |
10285 | | |
10286 | | /* |
10287 | | * We may have the standalone status. |
10288 | | */ |
10289 | 158k | if ((ctxt->encoding != NULL) && (!IS_BLANK_CH(RAW))) { |
10290 | 78.7k | if ((RAW == '?') && (NXT(1) == '>')) { |
10291 | 76.8k | SKIP(2); |
10292 | 76.8k | return; |
10293 | 76.8k | } |
10294 | 1.83k | xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Blank needed here\n"); |
10295 | 1.83k | } |
10296 | | |
10297 | | /* |
10298 | | * We can grow the input buffer freely at that point |
10299 | | */ |
10300 | 82.0k | GROW; |
10301 | | |
10302 | 82.0k | SKIP_BLANKS; |
10303 | 82.0k | ctxt->standalone = xmlParseSDDecl(ctxt); |
10304 | | |
10305 | 82.0k | SKIP_BLANKS; |
10306 | 82.0k | if ((RAW == '?') && (NXT(1) == '>')) { |
10307 | 32.3k | SKIP(2); |
10308 | 49.6k | } else if (RAW == '>') { |
10309 | | /* Deprecated old WD ... */ |
10310 | 1.18k | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
10311 | 1.18k | NEXT; |
10312 | 48.4k | } else { |
10313 | 48.4k | int c; |
10314 | | |
10315 | 48.4k | xmlFatalErr(ctxt, XML_ERR_XMLDECL_NOT_FINISHED, NULL); |
10316 | 2.01M | while ((PARSER_STOPPED(ctxt) == 0) && |
10317 | 2.01M | ((c = CUR) != 0)) { |
10318 | 2.00M | NEXT; |
10319 | 2.00M | if (c == '>') |
10320 | 36.8k | break; |
10321 | 2.00M | } |
10322 | 48.4k | } |
10323 | 82.0k | } |
10324 | | |
10325 | | /** |
10326 | | * @since 2.14.0 |
10327 | | * |
10328 | | * @param ctxt parser context |
10329 | | * @returns the version from the XML declaration. |
10330 | | */ |
10331 | | const xmlChar * |
10332 | 0 | xmlCtxtGetVersion(xmlParserCtxt *ctxt) { |
10333 | 0 | if (ctxt == NULL) |
10334 | 0 | return(NULL); |
10335 | | |
10336 | 0 | return(ctxt->version); |
10337 | 0 | } |
10338 | | |
10339 | | /** |
10340 | | * @since 2.14.0 |
10341 | | * |
10342 | | * @param ctxt parser context |
10343 | | * @returns the value from the standalone document declaration. |
10344 | | */ |
10345 | | int |
10346 | 0 | xmlCtxtGetStandalone(xmlParserCtxt *ctxt) { |
10347 | 0 | if (ctxt == NULL) |
10348 | 0 | return(0); |
10349 | | |
10350 | 0 | return(ctxt->standalone); |
10351 | 0 | } |
10352 | | |
10353 | | /** |
10354 | | * Parse an XML Misc* optional field. |
10355 | | * |
10356 | | * @deprecated Internal function, don't use. |
10357 | | * |
10358 | | * [27] Misc ::= Comment | PI | S |
10359 | | * @param ctxt an XML parser context |
10360 | | */ |
10361 | | |
10362 | | void |
10363 | 1.68M | xmlParseMisc(xmlParserCtxt *ctxt) { |
10364 | 1.77M | while (PARSER_STOPPED(ctxt) == 0) { |
10365 | 1.74M | SKIP_BLANKS; |
10366 | 1.74M | GROW; |
10367 | 1.74M | if ((RAW == '<') && (NXT(1) == '?')) { |
10368 | 64.6k | xmlParsePI(ctxt); |
10369 | 1.67M | } else if (CMP4(CUR_PTR, '<', '!', '-', '-')) { |
10370 | 25.1k | xmlParseComment(ctxt); |
10371 | 1.65M | } else { |
10372 | 1.65M | break; |
10373 | 1.65M | } |
10374 | 1.74M | } |
10375 | 1.68M | } |
10376 | | |
10377 | | static void |
10378 | 833k | xmlFinishDocument(xmlParserCtxtPtr ctxt) { |
10379 | 833k | xmlDocPtr doc; |
10380 | | |
10381 | | /* |
10382 | | * SAX: end of the document processing. |
10383 | | */ |
10384 | 833k | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
10385 | 833k | ctxt->sax->endDocument(ctxt->userData); |
10386 | | |
10387 | | /* |
10388 | | * Remove locally kept entity definitions if the tree was not built |
10389 | | */ |
10390 | 833k | doc = ctxt->myDoc; |
10391 | 833k | if ((doc != NULL) && |
10392 | 833k | (xmlStrEqual(doc->version, SAX_COMPAT_MODE))) { |
10393 | 15.9k | xmlFreeDoc(doc); |
10394 | 15.9k | ctxt->myDoc = NULL; |
10395 | 15.9k | } |
10396 | 833k | } |
10397 | | |
10398 | | /** |
10399 | | * Parse an XML document and invoke the SAX handlers. This is useful |
10400 | | * if you're only interested in custom SAX callbacks. If you want a |
10401 | | * document tree, use #xmlCtxtParseDocument. |
10402 | | * |
10403 | | * @param ctxt an XML parser context |
10404 | | * @returns 0, -1 in case of error. |
10405 | | */ |
10406 | | |
10407 | | int |
10408 | 836k | xmlParseDocument(xmlParserCtxt *ctxt) { |
10409 | 836k | if ((ctxt == NULL) || (ctxt->input == NULL)) |
10410 | 0 | return(-1); |
10411 | | |
10412 | 836k | GROW; |
10413 | | |
10414 | | /* |
10415 | | * SAX: detecting the level. |
10416 | | */ |
10417 | 836k | xmlCtxtInitializeLate(ctxt); |
10418 | | |
10419 | 836k | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) { |
10420 | 836k | ctxt->sax->setDocumentLocator(ctxt->userData, |
10421 | 836k | (xmlSAXLocator *) &xmlDefaultSAXLocator); |
10422 | 836k | } |
10423 | | |
10424 | 836k | xmlDetectEncoding(ctxt); |
10425 | | |
10426 | 836k | if (CUR == 0) { |
10427 | 3.27k | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); |
10428 | 3.27k | return(-1); |
10429 | 3.27k | } |
10430 | | |
10431 | 833k | GROW; |
10432 | 833k | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
10433 | | |
10434 | | /* |
10435 | | * Note that we will switch encoding on the fly. |
10436 | | */ |
10437 | 184k | xmlParseXMLDecl(ctxt); |
10438 | 184k | SKIP_BLANKS; |
10439 | 649k | } else { |
10440 | 649k | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10441 | 649k | if (ctxt->version == NULL) { |
10442 | 53 | xmlErrMemory(ctxt); |
10443 | 53 | return(-1); |
10444 | 53 | } |
10445 | 649k | } |
10446 | 833k | if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) |
10447 | 777k | ctxt->sax->startDocument(ctxt->userData); |
10448 | 833k | if ((ctxt->myDoc != NULL) && (ctxt->input != NULL) && |
10449 | 833k | (ctxt->input->buf != NULL) && (ctxt->input->buf->compressed >= 0)) { |
10450 | 0 | ctxt->myDoc->compression = ctxt->input->buf->compressed; |
10451 | 0 | } |
10452 | | |
10453 | | /* |
10454 | | * The Misc part of the Prolog |
10455 | | */ |
10456 | 833k | xmlParseMisc(ctxt); |
10457 | | |
10458 | | /* |
10459 | | * Then possibly doc type declaration(s) and more Misc |
10460 | | * (doctypedecl Misc*)? |
10461 | | */ |
10462 | 833k | GROW; |
10463 | 833k | if (CMP9(CUR_PTR, '<', '!', 'D', 'O', 'C', 'T', 'Y', 'P', 'E')) { |
10464 | | |
10465 | 148k | ctxt->inSubset = 1; |
10466 | 148k | xmlParseDocTypeDecl(ctxt); |
10467 | 148k | if (RAW == '[') { |
10468 | 136k | xmlParseInternalSubset(ctxt); |
10469 | 136k | } else if (RAW == '>') { |
10470 | 7.51k | NEXT; |
10471 | 7.51k | } |
10472 | | |
10473 | | /* |
10474 | | * Create and update the external subset. |
10475 | | */ |
10476 | 148k | ctxt->inSubset = 2; |
10477 | 148k | if ((ctxt->sax != NULL) && (ctxt->sax->externalSubset != NULL) && |
10478 | 148k | (!ctxt->disableSAX)) |
10479 | 37.6k | ctxt->sax->externalSubset(ctxt->userData, ctxt->intSubName, |
10480 | 37.6k | ctxt->extSubSystem, ctxt->extSubURI); |
10481 | 148k | ctxt->inSubset = 0; |
10482 | | |
10483 | 148k | xmlCleanSpecialAttr(ctxt); |
10484 | | |
10485 | 148k | xmlParseMisc(ctxt); |
10486 | 148k | } |
10487 | | |
10488 | | /* |
10489 | | * Time to start parsing the tree itself |
10490 | | */ |
10491 | 833k | GROW; |
10492 | 833k | if (RAW != '<') { |
10493 | 135k | if (ctxt->wellFormed) |
10494 | 6.85k | xmlFatalErrMsg(ctxt, XML_ERR_DOCUMENT_EMPTY, |
10495 | 6.85k | "Start tag expected, '<' not found\n"); |
10496 | 698k | } else { |
10497 | 698k | xmlParseElement(ctxt); |
10498 | | |
10499 | | /* |
10500 | | * The Misc part at the end |
10501 | | */ |
10502 | 698k | xmlParseMisc(ctxt); |
10503 | | |
10504 | 698k | xmlParserCheckEOF(ctxt, XML_ERR_DOCUMENT_END); |
10505 | 698k | } |
10506 | | |
10507 | 833k | ctxt->instate = XML_PARSER_EOF; |
10508 | 833k | xmlFinishDocument(ctxt); |
10509 | | |
10510 | 833k | if (! ctxt->wellFormed) { |
10511 | 232k | ctxt->valid = 0; |
10512 | 232k | return(-1); |
10513 | 232k | } |
10514 | | |
10515 | 601k | return(0); |
10516 | 833k | } |
10517 | | |
10518 | | /** |
10519 | | * Parse a general parsed entity |
10520 | | * An external general parsed entity is well-formed if it matches the |
10521 | | * production labeled extParsedEnt. |
10522 | | * |
10523 | | * @deprecated Internal function, don't use. |
10524 | | * |
10525 | | * [78] extParsedEnt ::= TextDecl? content |
10526 | | * |
10527 | | * @param ctxt an XML parser context |
10528 | | * @returns 0, -1 in case of error. the parser context is augmented |
10529 | | * as a result of the parsing. |
10530 | | */ |
10531 | | |
10532 | | int |
10533 | 0 | xmlParseExtParsedEnt(xmlParserCtxt *ctxt) { |
10534 | 0 | if ((ctxt == NULL) || (ctxt->input == NULL)) |
10535 | 0 | return(-1); |
10536 | | |
10537 | 0 | xmlCtxtInitializeLate(ctxt); |
10538 | |
|
10539 | 0 | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) { |
10540 | 0 | ctxt->sax->setDocumentLocator(ctxt->userData, |
10541 | 0 | (xmlSAXLocator *) &xmlDefaultSAXLocator); |
10542 | 0 | } |
10543 | |
|
10544 | 0 | xmlDetectEncoding(ctxt); |
10545 | |
|
10546 | 0 | if (CUR == 0) { |
10547 | 0 | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); |
10548 | 0 | } |
10549 | | |
10550 | | /* |
10551 | | * Check for the XMLDecl in the Prolog. |
10552 | | */ |
10553 | 0 | GROW; |
10554 | 0 | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && (IS_BLANK_CH(NXT(5)))) { |
10555 | | |
10556 | | /* |
10557 | | * Note that we will switch encoding on the fly. |
10558 | | */ |
10559 | 0 | xmlParseXMLDecl(ctxt); |
10560 | 0 | SKIP_BLANKS; |
10561 | 0 | } else { |
10562 | 0 | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10563 | 0 | } |
10564 | 0 | if ((ctxt->sax) && (ctxt->sax->startDocument) && (!ctxt->disableSAX)) |
10565 | 0 | ctxt->sax->startDocument(ctxt->userData); |
10566 | | |
10567 | | /* |
10568 | | * Doing validity checking on chunk doesn't make sense |
10569 | | */ |
10570 | 0 | ctxt->options &= ~XML_PARSE_DTDVALID; |
10571 | 0 | ctxt->validate = 0; |
10572 | 0 | ctxt->depth = 0; |
10573 | |
|
10574 | 0 | xmlParseContentInternal(ctxt); |
10575 | |
|
10576 | 0 | if (ctxt->input->cur < ctxt->input->end) |
10577 | 0 | xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); |
10578 | | |
10579 | | /* |
10580 | | * SAX: end of the document processing. |
10581 | | */ |
10582 | 0 | if ((ctxt->sax) && (ctxt->sax->endDocument != NULL)) |
10583 | 0 | ctxt->sax->endDocument(ctxt->userData); |
10584 | |
|
10585 | 0 | if (! ctxt->wellFormed) return(-1); |
10586 | 0 | return(0); |
10587 | 0 | } |
10588 | | |
10589 | | #ifdef LIBXML_PUSH_ENABLED |
10590 | | /************************************************************************ |
10591 | | * * |
10592 | | * Progressive parsing interfaces * |
10593 | | * * |
10594 | | ************************************************************************/ |
10595 | | |
10596 | | /** |
10597 | | * Check whether the input buffer contains a character. |
10598 | | * |
10599 | | * @param ctxt an XML parser context |
10600 | | * @param c character |
10601 | | */ |
10602 | | static int |
10603 | | xmlParseLookupChar(xmlParserCtxtPtr ctxt, int c) { |
10604 | | const xmlChar *cur; |
10605 | | |
10606 | | if (ctxt->checkIndex == 0) { |
10607 | | cur = ctxt->input->cur + 1; |
10608 | | } else { |
10609 | | cur = ctxt->input->cur + ctxt->checkIndex; |
10610 | | } |
10611 | | |
10612 | | if (memchr(cur, c, ctxt->input->end - cur) == NULL) { |
10613 | | size_t index = ctxt->input->end - ctxt->input->cur; |
10614 | | |
10615 | | if (index > LONG_MAX) { |
10616 | | ctxt->checkIndex = 0; |
10617 | | return(1); |
10618 | | } |
10619 | | ctxt->checkIndex = index; |
10620 | | return(0); |
10621 | | } else { |
10622 | | ctxt->checkIndex = 0; |
10623 | | return(1); |
10624 | | } |
10625 | | } |
10626 | | |
10627 | | /** |
10628 | | * Check whether the input buffer contains a string. |
10629 | | * |
10630 | | * @param ctxt an XML parser context |
10631 | | * @param startDelta delta to apply at the start |
10632 | | * @param str string |
10633 | | * @param strLen length of string |
10634 | | */ |
10635 | | static const xmlChar * |
10636 | | xmlParseLookupString(xmlParserCtxtPtr ctxt, size_t startDelta, |
10637 | | const char *str, size_t strLen) { |
10638 | | const xmlChar *cur, *term; |
10639 | | |
10640 | | if (ctxt->checkIndex == 0) { |
10641 | | cur = ctxt->input->cur + startDelta; |
10642 | | } else { |
10643 | | cur = ctxt->input->cur + ctxt->checkIndex; |
10644 | | } |
10645 | | |
10646 | | term = BAD_CAST strstr((const char *) cur, str); |
10647 | | if (term == NULL) { |
10648 | | const xmlChar *end = ctxt->input->end; |
10649 | | size_t index; |
10650 | | |
10651 | | /* Rescan (strLen - 1) characters. */ |
10652 | | if ((size_t) (end - cur) < strLen) |
10653 | | end = cur; |
10654 | | else |
10655 | | end -= strLen - 1; |
10656 | | index = end - ctxt->input->cur; |
10657 | | if (index > LONG_MAX) { |
10658 | | ctxt->checkIndex = 0; |
10659 | | return(ctxt->input->end - strLen); |
10660 | | } |
10661 | | ctxt->checkIndex = index; |
10662 | | } else { |
10663 | | ctxt->checkIndex = 0; |
10664 | | } |
10665 | | |
10666 | | return(term); |
10667 | | } |
10668 | | |
10669 | | /** |
10670 | | * Check whether the input buffer contains terminated char data. |
10671 | | * |
10672 | | * @param ctxt an XML parser context |
10673 | | */ |
10674 | | static int |
10675 | | xmlParseLookupCharData(xmlParserCtxtPtr ctxt) { |
10676 | | const xmlChar *cur = ctxt->input->cur + ctxt->checkIndex; |
10677 | | const xmlChar *end = ctxt->input->end; |
10678 | | size_t index; |
10679 | | |
10680 | | while (cur < end) { |
10681 | | if ((*cur == '<') || (*cur == '&')) { |
10682 | | ctxt->checkIndex = 0; |
10683 | | return(1); |
10684 | | } |
10685 | | cur++; |
10686 | | } |
10687 | | |
10688 | | index = cur - ctxt->input->cur; |
10689 | | if (index > LONG_MAX) { |
10690 | | ctxt->checkIndex = 0; |
10691 | | return(1); |
10692 | | } |
10693 | | ctxt->checkIndex = index; |
10694 | | return(0); |
10695 | | } |
10696 | | |
10697 | | /** |
10698 | | * Check whether there's enough data in the input buffer to finish parsing |
10699 | | * a start tag. This has to take quotes into account. |
10700 | | * |
10701 | | * @param ctxt an XML parser context |
10702 | | */ |
10703 | | static int |
10704 | | xmlParseLookupGt(xmlParserCtxtPtr ctxt) { |
10705 | | const xmlChar *cur; |
10706 | | const xmlChar *end = ctxt->input->end; |
10707 | | int state = ctxt->endCheckState; |
10708 | | size_t index; |
10709 | | |
10710 | | if (ctxt->checkIndex == 0) |
10711 | | cur = ctxt->input->cur + 1; |
10712 | | else |
10713 | | cur = ctxt->input->cur + ctxt->checkIndex; |
10714 | | |
10715 | | while (cur < end) { |
10716 | | if (state) { |
10717 | | if (*cur == state) |
10718 | | state = 0; |
10719 | | } else if (*cur == '\'' || *cur == '"') { |
10720 | | state = *cur; |
10721 | | } else if (*cur == '>') { |
10722 | | ctxt->checkIndex = 0; |
10723 | | ctxt->endCheckState = 0; |
10724 | | return(1); |
10725 | | } |
10726 | | cur++; |
10727 | | } |
10728 | | |
10729 | | index = cur - ctxt->input->cur; |
10730 | | if (index > LONG_MAX) { |
10731 | | ctxt->checkIndex = 0; |
10732 | | ctxt->endCheckState = 0; |
10733 | | return(1); |
10734 | | } |
10735 | | ctxt->checkIndex = index; |
10736 | | ctxt->endCheckState = state; |
10737 | | return(0); |
10738 | | } |
10739 | | |
10740 | | /** |
10741 | | * Check whether there's enough data in the input buffer to finish parsing |
10742 | | * the internal subset. |
10743 | | * |
10744 | | * @param ctxt an XML parser context |
10745 | | */ |
10746 | | static int |
10747 | | xmlParseLookupInternalSubset(xmlParserCtxtPtr ctxt) { |
10748 | | /* |
10749 | | * Sorry, but progressive parsing of the internal subset is not |
10750 | | * supported. We first check that the full content of the internal |
10751 | | * subset is available and parsing is launched only at that point. |
10752 | | * Internal subset ends with "']' S? '>'" in an unescaped section and |
10753 | | * not in a ']]>' sequence which are conditional sections. |
10754 | | */ |
10755 | | const xmlChar *cur, *start; |
10756 | | const xmlChar *end = ctxt->input->end; |
10757 | | int state = ctxt->endCheckState; |
10758 | | size_t index; |
10759 | | |
10760 | | if (ctxt->checkIndex == 0) { |
10761 | | cur = ctxt->input->cur + 1; |
10762 | | } else { |
10763 | | cur = ctxt->input->cur + ctxt->checkIndex; |
10764 | | } |
10765 | | start = cur; |
10766 | | |
10767 | | while (cur < end) { |
10768 | | if (state == '-') { |
10769 | | if ((*cur == '-') && |
10770 | | (cur[1] == '-') && |
10771 | | (cur[2] == '>')) { |
10772 | | state = 0; |
10773 | | cur += 3; |
10774 | | start = cur; |
10775 | | continue; |
10776 | | } |
10777 | | } |
10778 | | else if (state == ']') { |
10779 | | if (*cur == '>') { |
10780 | | ctxt->checkIndex = 0; |
10781 | | ctxt->endCheckState = 0; |
10782 | | return(1); |
10783 | | } |
10784 | | if (IS_BLANK_CH(*cur)) { |
10785 | | state = ' '; |
10786 | | } else if (*cur != ']') { |
10787 | | state = 0; |
10788 | | start = cur; |
10789 | | continue; |
10790 | | } |
10791 | | } |
10792 | | else if (state == ' ') { |
10793 | | if (*cur == '>') { |
10794 | | ctxt->checkIndex = 0; |
10795 | | ctxt->endCheckState = 0; |
10796 | | return(1); |
10797 | | } |
10798 | | if (!IS_BLANK_CH(*cur)) { |
10799 | | state = 0; |
10800 | | start = cur; |
10801 | | continue; |
10802 | | } |
10803 | | } |
10804 | | else if (state != 0) { |
10805 | | if (*cur == state) { |
10806 | | state = 0; |
10807 | | start = cur + 1; |
10808 | | } |
10809 | | } |
10810 | | else if (*cur == '<') { |
10811 | | if ((cur[1] == '!') && |
10812 | | (cur[2] == '-') && |
10813 | | (cur[3] == '-')) { |
10814 | | state = '-'; |
10815 | | cur += 4; |
10816 | | /* Don't treat <!--> as comment */ |
10817 | | start = cur; |
10818 | | continue; |
10819 | | } |
10820 | | } |
10821 | | else if ((*cur == '"') || (*cur == '\'') || (*cur == ']')) { |
10822 | | state = *cur; |
10823 | | } |
10824 | | |
10825 | | cur++; |
10826 | | } |
10827 | | |
10828 | | /* |
10829 | | * Rescan the three last characters to detect "<!--" and "-->" |
10830 | | * split across chunks. |
10831 | | */ |
10832 | | if ((state == 0) || (state == '-')) { |
10833 | | if (cur - start < 3) |
10834 | | cur = start; |
10835 | | else |
10836 | | cur -= 3; |
10837 | | } |
10838 | | index = cur - ctxt->input->cur; |
10839 | | if (index > LONG_MAX) { |
10840 | | ctxt->checkIndex = 0; |
10841 | | ctxt->endCheckState = 0; |
10842 | | return(1); |
10843 | | } |
10844 | | ctxt->checkIndex = index; |
10845 | | ctxt->endCheckState = state; |
10846 | | return(0); |
10847 | | } |
10848 | | |
10849 | | /** |
10850 | | * Try to progress on parsing |
10851 | | * |
10852 | | * @param ctxt an XML parser context |
10853 | | * @param terminate last chunk indicator |
10854 | | * @returns zero if no parsing was possible |
10855 | | */ |
10856 | | static int |
10857 | | xmlParseTryOrFinish(xmlParserCtxtPtr ctxt, int terminate) { |
10858 | | int ret = 0; |
10859 | | size_t avail; |
10860 | | xmlChar cur, next; |
10861 | | |
10862 | | if (ctxt->input == NULL) |
10863 | | return(0); |
10864 | | |
10865 | | if ((ctxt->input != NULL) && |
10866 | | (ctxt->input->cur - ctxt->input->base > 4096)) { |
10867 | | xmlParserShrink(ctxt); |
10868 | | } |
10869 | | |
10870 | | while (ctxt->disableSAX == 0) { |
10871 | | avail = ctxt->input->end - ctxt->input->cur; |
10872 | | if (avail < 1) |
10873 | | goto done; |
10874 | | switch (ctxt->instate) { |
10875 | | case XML_PARSER_EOF: |
10876 | | /* |
10877 | | * Document parsing is done ! |
10878 | | */ |
10879 | | goto done; |
10880 | | case XML_PARSER_START: |
10881 | | /* |
10882 | | * Very first chars read from the document flow. |
10883 | | */ |
10884 | | if ((!terminate) && (avail < 4)) |
10885 | | goto done; |
10886 | | |
10887 | | /* |
10888 | | * We need more bytes to detect EBCDIC code pages. |
10889 | | * See xmlDetectEBCDIC. |
10890 | | */ |
10891 | | if ((CMP4(CUR_PTR, 0x4C, 0x6F, 0xA7, 0x94)) && |
10892 | | (!terminate) && (avail < 200)) |
10893 | | goto done; |
10894 | | |
10895 | | xmlDetectEncoding(ctxt); |
10896 | | ctxt->instate = XML_PARSER_XML_DECL; |
10897 | | break; |
10898 | | |
10899 | | case XML_PARSER_XML_DECL: |
10900 | | if ((!terminate) && (avail < 2)) |
10901 | | goto done; |
10902 | | cur = ctxt->input->cur[0]; |
10903 | | next = ctxt->input->cur[1]; |
10904 | | if ((cur == '<') && (next == '?')) { |
10905 | | /* PI or XML decl */ |
10906 | | if ((!terminate) && |
10907 | | (!xmlParseLookupString(ctxt, 2, "?>", 2))) |
10908 | | goto done; |
10909 | | if ((ctxt->input->cur[2] == 'x') && |
10910 | | (ctxt->input->cur[3] == 'm') && |
10911 | | (ctxt->input->cur[4] == 'l') && |
10912 | | (IS_BLANK_CH(ctxt->input->cur[5]))) { |
10913 | | ret += 5; |
10914 | | xmlParseXMLDecl(ctxt); |
10915 | | } else { |
10916 | | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10917 | | if (ctxt->version == NULL) { |
10918 | | xmlErrMemory(ctxt); |
10919 | | break; |
10920 | | } |
10921 | | } |
10922 | | } else { |
10923 | | ctxt->version = xmlCharStrdup(XML_DEFAULT_VERSION); |
10924 | | if (ctxt->version == NULL) { |
10925 | | xmlErrMemory(ctxt); |
10926 | | break; |
10927 | | } |
10928 | | } |
10929 | | if ((ctxt->sax) && (ctxt->sax->setDocumentLocator)) { |
10930 | | ctxt->sax->setDocumentLocator(ctxt->userData, |
10931 | | (xmlSAXLocator *) &xmlDefaultSAXLocator); |
10932 | | } |
10933 | | if ((ctxt->sax) && (ctxt->sax->startDocument) && |
10934 | | (!ctxt->disableSAX)) |
10935 | | ctxt->sax->startDocument(ctxt->userData); |
10936 | | ctxt->instate = XML_PARSER_MISC; |
10937 | | break; |
10938 | | case XML_PARSER_START_TAG: { |
10939 | | const xmlChar *name; |
10940 | | const xmlChar *prefix = NULL; |
10941 | | const xmlChar *URI = NULL; |
10942 | | int line = ctxt->input->line; |
10943 | | int nbNs = 0; |
10944 | | |
10945 | | if ((!terminate) && (avail < 2)) |
10946 | | goto done; |
10947 | | cur = ctxt->input->cur[0]; |
10948 | | if (cur != '<') { |
10949 | | xmlFatalErrMsg(ctxt, XML_ERR_DOCUMENT_EMPTY, |
10950 | | "Start tag expected, '<' not found"); |
10951 | | ctxt->instate = XML_PARSER_EOF; |
10952 | | xmlFinishDocument(ctxt); |
10953 | | goto done; |
10954 | | } |
10955 | | if ((!terminate) && (!xmlParseLookupGt(ctxt))) |
10956 | | goto done; |
10957 | | if (ctxt->spaceNr == 0) |
10958 | | spacePush(ctxt, -1); |
10959 | | else if (*ctxt->space == -2) |
10960 | | spacePush(ctxt, -1); |
10961 | | else |
10962 | | spacePush(ctxt, *ctxt->space); |
10963 | | #ifdef LIBXML_SAX1_ENABLED |
10964 | | if (ctxt->sax2) |
10965 | | #endif /* LIBXML_SAX1_ENABLED */ |
10966 | | name = xmlParseStartTag2(ctxt, &prefix, &URI, &nbNs); |
10967 | | #ifdef LIBXML_SAX1_ENABLED |
10968 | | else |
10969 | | name = xmlParseStartTag(ctxt); |
10970 | | #endif /* LIBXML_SAX1_ENABLED */ |
10971 | | if (name == NULL) { |
10972 | | spacePop(ctxt); |
10973 | | ctxt->instate = XML_PARSER_EOF; |
10974 | | xmlFinishDocument(ctxt); |
10975 | | goto done; |
10976 | | } |
10977 | | #ifdef LIBXML_VALID_ENABLED |
10978 | | /* |
10979 | | * [ VC: Root Element Type ] |
10980 | | * The Name in the document type declaration must match |
10981 | | * the element type of the root element. |
10982 | | */ |
10983 | | if (ctxt->validate && ctxt->wellFormed && ctxt->myDoc && |
10984 | | ctxt->node && (ctxt->node == ctxt->myDoc->children)) |
10985 | | ctxt->valid &= xmlValidateRoot(&ctxt->vctxt, ctxt->myDoc); |
10986 | | #endif /* LIBXML_VALID_ENABLED */ |
10987 | | |
10988 | | /* |
10989 | | * Check for an Empty Element. |
10990 | | */ |
10991 | | if ((RAW == '/') && (NXT(1) == '>')) { |
10992 | | SKIP(2); |
10993 | | |
10994 | | if (ctxt->sax2) { |
10995 | | if ((ctxt->sax != NULL) && |
10996 | | (ctxt->sax->endElementNs != NULL) && |
10997 | | (!ctxt->disableSAX)) |
10998 | | ctxt->sax->endElementNs(ctxt->userData, name, |
10999 | | prefix, URI); |
11000 | | if (nbNs > 0) |
11001 | | xmlParserNsPop(ctxt, nbNs); |
11002 | | #ifdef LIBXML_SAX1_ENABLED |
11003 | | } else { |
11004 | | if ((ctxt->sax != NULL) && |
11005 | | (ctxt->sax->endElement != NULL) && |
11006 | | (!ctxt->disableSAX)) |
11007 | | ctxt->sax->endElement(ctxt->userData, name); |
11008 | | #endif /* LIBXML_SAX1_ENABLED */ |
11009 | | } |
11010 | | spacePop(ctxt); |
11011 | | } else if (RAW == '>') { |
11012 | | NEXT; |
11013 | | nameNsPush(ctxt, name, prefix, URI, line, nbNs); |
11014 | | } else { |
11015 | | xmlFatalErrMsgStr(ctxt, XML_ERR_GT_REQUIRED, |
11016 | | "Couldn't find end of Start Tag %s\n", |
11017 | | name); |
11018 | | nodePop(ctxt); |
11019 | | spacePop(ctxt); |
11020 | | if (nbNs > 0) |
11021 | | xmlParserNsPop(ctxt, nbNs); |
11022 | | } |
11023 | | |
11024 | | if (ctxt->nameNr == 0) |
11025 | | ctxt->instate = XML_PARSER_EPILOG; |
11026 | | else |
11027 | | ctxt->instate = XML_PARSER_CONTENT; |
11028 | | break; |
11029 | | } |
11030 | | case XML_PARSER_CONTENT: { |
11031 | | cur = ctxt->input->cur[0]; |
11032 | | |
11033 | | if (cur == '<') { |
11034 | | if ((!terminate) && (avail < 2)) |
11035 | | goto done; |
11036 | | next = ctxt->input->cur[1]; |
11037 | | |
11038 | | if (next == '/') { |
11039 | | ctxt->instate = XML_PARSER_END_TAG; |
11040 | | break; |
11041 | | } else if (next == '?') { |
11042 | | if ((!terminate) && |
11043 | | (!xmlParseLookupString(ctxt, 2, "?>", 2))) |
11044 | | goto done; |
11045 | | xmlParsePI(ctxt); |
11046 | | ctxt->instate = XML_PARSER_CONTENT; |
11047 | | break; |
11048 | | } else if (next == '!') { |
11049 | | if ((!terminate) && (avail < 3)) |
11050 | | goto done; |
11051 | | next = ctxt->input->cur[2]; |
11052 | | |
11053 | | if (next == '-') { |
11054 | | if ((!terminate) && (avail < 4)) |
11055 | | goto done; |
11056 | | if (ctxt->input->cur[3] == '-') { |
11057 | | if ((!terminate) && |
11058 | | (!xmlParseLookupString(ctxt, 4, "-->", 3))) |
11059 | | goto done; |
11060 | | xmlParseComment(ctxt); |
11061 | | ctxt->instate = XML_PARSER_CONTENT; |
11062 | | break; |
11063 | | } |
11064 | | } else if (next == '[') { |
11065 | | if ((!terminate) && (avail < 9)) |
11066 | | goto done; |
11067 | | if ((ctxt->input->cur[2] == '[') && |
11068 | | (ctxt->input->cur[3] == 'C') && |
11069 | | (ctxt->input->cur[4] == 'D') && |
11070 | | (ctxt->input->cur[5] == 'A') && |
11071 | | (ctxt->input->cur[6] == 'T') && |
11072 | | (ctxt->input->cur[7] == 'A') && |
11073 | | (ctxt->input->cur[8] == '[')) { |
11074 | | if ((!terminate) && |
11075 | | (!xmlParseLookupString(ctxt, 9, "]]>", 3))) |
11076 | | goto done; |
11077 | | ctxt->instate = XML_PARSER_CDATA_SECTION; |
11078 | | xmlParseCDSect(ctxt); |
11079 | | ctxt->instate = XML_PARSER_CONTENT; |
11080 | | break; |
11081 | | } |
11082 | | } |
11083 | | } |
11084 | | } else if (cur == '&') { |
11085 | | if ((!terminate) && (!xmlParseLookupChar(ctxt, ';'))) |
11086 | | goto done; |
11087 | | xmlParseReference(ctxt); |
11088 | | break; |
11089 | | } else { |
11090 | | /* TODO Avoid the extra copy, handle directly !!! */ |
11091 | | /* |
11092 | | * Goal of the following test is: |
11093 | | * - minimize calls to the SAX 'character' callback |
11094 | | * when they are mergeable |
11095 | | * - handle an problem for isBlank when we only parse |
11096 | | * a sequence of blank chars and the next one is |
11097 | | * not available to check against '<' presence. |
11098 | | * - tries to homogenize the differences in SAX |
11099 | | * callbacks between the push and pull versions |
11100 | | * of the parser. |
11101 | | */ |
11102 | | if (avail < XML_PARSER_BIG_BUFFER_SIZE) { |
11103 | | if ((!terminate) && (!xmlParseLookupCharData(ctxt))) |
11104 | | goto done; |
11105 | | } |
11106 | | ctxt->checkIndex = 0; |
11107 | | xmlParseCharDataInternal(ctxt, !terminate); |
11108 | | break; |
11109 | | } |
11110 | | |
11111 | | ctxt->instate = XML_PARSER_START_TAG; |
11112 | | break; |
11113 | | } |
11114 | | case XML_PARSER_END_TAG: |
11115 | | if ((!terminate) && (!xmlParseLookupChar(ctxt, '>'))) |
11116 | | goto done; |
11117 | | if (ctxt->sax2) { |
11118 | | xmlParseEndTag2(ctxt, &ctxt->pushTab[ctxt->nameNr - 1]); |
11119 | | nameNsPop(ctxt); |
11120 | | } |
11121 | | #ifdef LIBXML_SAX1_ENABLED |
11122 | | else |
11123 | | xmlParseEndTag1(ctxt, 0); |
11124 | | #endif /* LIBXML_SAX1_ENABLED */ |
11125 | | if (ctxt->nameNr == 0) { |
11126 | | ctxt->instate = XML_PARSER_EPILOG; |
11127 | | } else { |
11128 | | ctxt->instate = XML_PARSER_CONTENT; |
11129 | | } |
11130 | | break; |
11131 | | case XML_PARSER_MISC: |
11132 | | case XML_PARSER_PROLOG: |
11133 | | case XML_PARSER_EPILOG: |
11134 | | SKIP_BLANKS; |
11135 | | avail = ctxt->input->end - ctxt->input->cur; |
11136 | | if (avail < 1) |
11137 | | goto done; |
11138 | | if (ctxt->input->cur[0] == '<') { |
11139 | | if ((!terminate) && (avail < 2)) |
11140 | | goto done; |
11141 | | next = ctxt->input->cur[1]; |
11142 | | if (next == '?') { |
11143 | | if ((!terminate) && |
11144 | | (!xmlParseLookupString(ctxt, 2, "?>", 2))) |
11145 | | goto done; |
11146 | | xmlParsePI(ctxt); |
11147 | | break; |
11148 | | } else if (next == '!') { |
11149 | | if ((!terminate) && (avail < 3)) |
11150 | | goto done; |
11151 | | |
11152 | | if (ctxt->input->cur[2] == '-') { |
11153 | | if ((!terminate) && (avail < 4)) |
11154 | | goto done; |
11155 | | if (ctxt->input->cur[3] == '-') { |
11156 | | if ((!terminate) && |
11157 | | (!xmlParseLookupString(ctxt, 4, "-->", 3))) |
11158 | | goto done; |
11159 | | xmlParseComment(ctxt); |
11160 | | break; |
11161 | | } |
11162 | | } else if (ctxt->instate == XML_PARSER_MISC) { |
11163 | | if ((!terminate) && (avail < 9)) |
11164 | | goto done; |
11165 | | if ((ctxt->input->cur[2] == 'D') && |
11166 | | (ctxt->input->cur[3] == 'O') && |
11167 | | (ctxt->input->cur[4] == 'C') && |
11168 | | (ctxt->input->cur[5] == 'T') && |
11169 | | (ctxt->input->cur[6] == 'Y') && |
11170 | | (ctxt->input->cur[7] == 'P') && |
11171 | | (ctxt->input->cur[8] == 'E')) { |
11172 | | if ((!terminate) && (!xmlParseLookupGt(ctxt))) |
11173 | | goto done; |
11174 | | ctxt->inSubset = 1; |
11175 | | xmlParseDocTypeDecl(ctxt); |
11176 | | if (RAW == '[') { |
11177 | | ctxt->instate = XML_PARSER_DTD; |
11178 | | } else { |
11179 | | if (RAW == '>') |
11180 | | NEXT; |
11181 | | /* |
11182 | | * Create and update the external subset. |
11183 | | */ |
11184 | | ctxt->inSubset = 2; |
11185 | | if ((ctxt->sax != NULL) && |
11186 | | (!ctxt->disableSAX) && |
11187 | | (ctxt->sax->externalSubset != NULL)) |
11188 | | ctxt->sax->externalSubset( |
11189 | | ctxt->userData, |
11190 | | ctxt->intSubName, |
11191 | | ctxt->extSubSystem, |
11192 | | ctxt->extSubURI); |
11193 | | ctxt->inSubset = 0; |
11194 | | xmlCleanSpecialAttr(ctxt); |
11195 | | ctxt->instate = XML_PARSER_PROLOG; |
11196 | | } |
11197 | | break; |
11198 | | } |
11199 | | } |
11200 | | } |
11201 | | } |
11202 | | |
11203 | | if (ctxt->instate == XML_PARSER_EPILOG) { |
11204 | | if (ctxt->errNo == XML_ERR_OK) |
11205 | | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_END, NULL); |
11206 | | ctxt->instate = XML_PARSER_EOF; |
11207 | | xmlFinishDocument(ctxt); |
11208 | | } else { |
11209 | | ctxt->instate = XML_PARSER_START_TAG; |
11210 | | } |
11211 | | break; |
11212 | | case XML_PARSER_DTD: { |
11213 | | if ((!terminate) && (!xmlParseLookupInternalSubset(ctxt))) |
11214 | | goto done; |
11215 | | xmlParseInternalSubset(ctxt); |
11216 | | ctxt->inSubset = 2; |
11217 | | if ((ctxt->sax != NULL) && (!ctxt->disableSAX) && |
11218 | | (ctxt->sax->externalSubset != NULL)) |
11219 | | ctxt->sax->externalSubset(ctxt->userData, ctxt->intSubName, |
11220 | | ctxt->extSubSystem, ctxt->extSubURI); |
11221 | | ctxt->inSubset = 0; |
11222 | | xmlCleanSpecialAttr(ctxt); |
11223 | | ctxt->instate = XML_PARSER_PROLOG; |
11224 | | break; |
11225 | | } |
11226 | | default: |
11227 | | xmlFatalErrMsg(ctxt, XML_ERR_INTERNAL_ERROR, |
11228 | | "PP: internal error\n"); |
11229 | | ctxt->instate = XML_PARSER_EOF; |
11230 | | break; |
11231 | | } |
11232 | | } |
11233 | | done: |
11234 | | return(ret); |
11235 | | } |
11236 | | |
11237 | | /** |
11238 | | * Parse a chunk of memory in push parser mode. |
11239 | | * |
11240 | | * Assumes that the parser context was initialized with |
11241 | | * #xmlCreatePushParserCtxt. |
11242 | | * |
11243 | | * The last chunk, which will often be empty, must be marked with |
11244 | | * the `terminate` flag. With the default SAX callbacks, the resulting |
11245 | | * document will be available in ctxt->myDoc. This pointer will not |
11246 | | * be freed when calling #xmlFreeParserCtxt and must be freed by the |
11247 | | * caller. If the document isn't well-formed, it will still be returned |
11248 | | * in ctxt->myDoc. |
11249 | | * |
11250 | | * As an exception, #xmlCtxtResetPush will free the document in |
11251 | | * ctxt->myDoc. So ctxt->myDoc should be set to NULL after extracting |
11252 | | * the document. |
11253 | | * |
11254 | | * Since 2.14.0, #xmlCtxtGetDocument can be used to retrieve the |
11255 | | * result document. |
11256 | | * |
11257 | | * @param ctxt an XML parser context |
11258 | | * @param chunk chunk of memory |
11259 | | * @param size size of chunk in bytes |
11260 | | * @param terminate last chunk indicator |
11261 | | * @returns an xmlParserErrors code (0 on success). |
11262 | | */ |
11263 | | int |
11264 | | xmlParseChunk(xmlParserCtxt *ctxt, const char *chunk, int size, |
11265 | | int terminate) { |
11266 | | size_t curBase; |
11267 | | size_t maxLength; |
11268 | | size_t pos; |
11269 | | int end_in_lf = 0; |
11270 | | int res; |
11271 | | |
11272 | | if ((ctxt == NULL) || (size < 0)) |
11273 | | return(XML_ERR_ARGUMENT); |
11274 | | if ((chunk == NULL) && (size > 0)) |
11275 | | return(XML_ERR_ARGUMENT); |
11276 | | if ((ctxt->input == NULL) || (ctxt->input->buf == NULL)) |
11277 | | return(XML_ERR_ARGUMENT); |
11278 | | if (ctxt->disableSAX != 0) |
11279 | | return(ctxt->errNo); |
11280 | | |
11281 | | ctxt->input->flags |= XML_INPUT_PROGRESSIVE; |
11282 | | if (ctxt->instate == XML_PARSER_START) |
11283 | | xmlCtxtInitializeLate(ctxt); |
11284 | | if ((size > 0) && (chunk != NULL) && (!terminate) && |
11285 | | (chunk[size - 1] == '\r')) { |
11286 | | end_in_lf = 1; |
11287 | | size--; |
11288 | | } |
11289 | | |
11290 | | /* |
11291 | | * Also push an empty chunk to make sure that the raw buffer |
11292 | | * will be flushed if there is an encoder. |
11293 | | */ |
11294 | | pos = ctxt->input->cur - ctxt->input->base; |
11295 | | res = xmlParserInputBufferPush(ctxt->input->buf, size, chunk); |
11296 | | xmlBufUpdateInput(ctxt->input->buf->buffer, ctxt->input, pos); |
11297 | | if (res < 0) { |
11298 | | xmlCtxtErrIO(ctxt, ctxt->input->buf->error, NULL); |
11299 | | xmlHaltParser(ctxt); |
11300 | | return(ctxt->errNo); |
11301 | | } |
11302 | | |
11303 | | xmlParseTryOrFinish(ctxt, terminate); |
11304 | | |
11305 | | curBase = ctxt->input->cur - ctxt->input->base; |
11306 | | maxLength = (ctxt->options & XML_PARSE_HUGE) ? |
11307 | | XML_MAX_HUGE_LENGTH : |
11308 | | XML_MAX_LOOKUP_LIMIT; |
11309 | | if (curBase > maxLength) { |
11310 | | xmlFatalErr(ctxt, XML_ERR_RESOURCE_LIMIT, |
11311 | | "Buffer size limit exceeded, try XML_PARSE_HUGE\n"); |
11312 | | xmlHaltParser(ctxt); |
11313 | | } |
11314 | | |
11315 | | if ((ctxt->errNo != XML_ERR_OK) && (ctxt->disableSAX != 0)) |
11316 | | return(ctxt->errNo); |
11317 | | |
11318 | | if (end_in_lf == 1) { |
11319 | | pos = ctxt->input->cur - ctxt->input->base; |
11320 | | res = xmlParserInputBufferPush(ctxt->input->buf, 1, "\r"); |
11321 | | xmlBufUpdateInput(ctxt->input->buf->buffer, ctxt->input, pos); |
11322 | | if (res < 0) { |
11323 | | xmlCtxtErrIO(ctxt, ctxt->input->buf->error, NULL); |
11324 | | xmlHaltParser(ctxt); |
11325 | | return(ctxt->errNo); |
11326 | | } |
11327 | | } |
11328 | | if (terminate) { |
11329 | | /* |
11330 | | * Check for termination |
11331 | | */ |
11332 | | if ((ctxt->instate != XML_PARSER_EOF) && |
11333 | | (ctxt->instate != XML_PARSER_EPILOG)) { |
11334 | | if (ctxt->nameNr > 0) { |
11335 | | const xmlChar *name = ctxt->nameTab[ctxt->nameNr - 1]; |
11336 | | int line = ctxt->pushTab[ctxt->nameNr - 1].line; |
11337 | | xmlFatalErrMsgStrIntStr(ctxt, XML_ERR_TAG_NOT_FINISHED, |
11338 | | "Premature end of data in tag %s line %d\n", |
11339 | | name, line, NULL); |
11340 | | } else if (ctxt->instate == XML_PARSER_START) { |
11341 | | xmlFatalErr(ctxt, XML_ERR_DOCUMENT_EMPTY, NULL); |
11342 | | } else { |
11343 | | xmlFatalErrMsg(ctxt, XML_ERR_DOCUMENT_EMPTY, |
11344 | | "Start tag expected, '<' not found\n"); |
11345 | | } |
11346 | | } else { |
11347 | | xmlParserCheckEOF(ctxt, XML_ERR_DOCUMENT_END); |
11348 | | } |
11349 | | if (ctxt->instate != XML_PARSER_EOF) { |
11350 | | ctxt->instate = XML_PARSER_EOF; |
11351 | | xmlFinishDocument(ctxt); |
11352 | | } |
11353 | | } |
11354 | | if (ctxt->wellFormed == 0) |
11355 | | return((xmlParserErrors) ctxt->errNo); |
11356 | | else |
11357 | | return(0); |
11358 | | } |
11359 | | |
11360 | | /************************************************************************ |
11361 | | * * |
11362 | | * I/O front end functions to the parser * |
11363 | | * * |
11364 | | ************************************************************************/ |
11365 | | |
11366 | | /** |
11367 | | * Create a parser context for using the XML parser in push mode. |
11368 | | * See #xmlParseChunk. |
11369 | | * |
11370 | | * Passing an initial chunk is useless and deprecated. |
11371 | | * |
11372 | | * The push parser doesn't support recovery mode or the |
11373 | | * XML_PARSE_NOBLANKS option. |
11374 | | * |
11375 | | * `filename` is used as base URI to fetch external entities and for |
11376 | | * error reports. |
11377 | | * |
11378 | | * @param sax a SAX handler (optional) |
11379 | | * @param user_data user data for SAX callbacks (optional) |
11380 | | * @param chunk initial chunk (optional, deprecated) |
11381 | | * @param size size of initial chunk in bytes |
11382 | | * @param filename file name or URI (optional) |
11383 | | * @returns the new parser context or NULL if a memory allocation |
11384 | | * failed. |
11385 | | */ |
11386 | | |
11387 | | xmlParserCtxt * |
11388 | | xmlCreatePushParserCtxt(xmlSAXHandler *sax, void *user_data, |
11389 | | const char *chunk, int size, const char *filename) { |
11390 | | xmlParserCtxtPtr ctxt; |
11391 | | xmlParserInputPtr input; |
11392 | | |
11393 | | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
11394 | | if (ctxt == NULL) |
11395 | | return(NULL); |
11396 | | |
11397 | | ctxt->options &= ~XML_PARSE_NODICT; |
11398 | | ctxt->dictNames = 1; |
11399 | | |
11400 | | input = xmlNewPushInput(filename, chunk, size); |
11401 | | if (input == NULL) { |
11402 | | xmlFreeParserCtxt(ctxt); |
11403 | | return(NULL); |
11404 | | } |
11405 | | if (xmlCtxtPushInput(ctxt, input) < 0) { |
11406 | | xmlFreeInputStream(input); |
11407 | | xmlFreeParserCtxt(ctxt); |
11408 | | return(NULL); |
11409 | | } |
11410 | | |
11411 | | return(ctxt); |
11412 | | } |
11413 | | #endif /* LIBXML_PUSH_ENABLED */ |
11414 | | |
11415 | | /** |
11416 | | * Blocks further parser processing |
11417 | | * |
11418 | | * @param ctxt an XML parser context |
11419 | | */ |
11420 | | void |
11421 | 22.9M | xmlStopParser(xmlParserCtxt *ctxt) { |
11422 | 22.9M | if (ctxt == NULL) |
11423 | 22.9M | return; |
11424 | 0 | xmlHaltParser(ctxt); |
11425 | | /* |
11426 | | * TODO: Update ctxt->lastError and ctxt->wellFormed? |
11427 | | */ |
11428 | 0 | if (ctxt->errNo != XML_ERR_NO_MEMORY) |
11429 | 0 | ctxt->errNo = XML_ERR_USER_STOP; |
11430 | 0 | } |
11431 | | |
11432 | | /** |
11433 | | * Create a parser context for using the XML parser with an existing |
11434 | | * I/O stream |
11435 | | * |
11436 | | * @param sax a SAX handler (optional) |
11437 | | * @param user_data user data for SAX callbacks (optional) |
11438 | | * @param ioread an I/O read function |
11439 | | * @param ioclose an I/O close function (optional) |
11440 | | * @param ioctx an I/O handler |
11441 | | * @param enc the charset encoding if known (deprecated) |
11442 | | * @returns the new parser context or NULL |
11443 | | */ |
11444 | | xmlParserCtxt * |
11445 | | xmlCreateIOParserCtxt(xmlSAXHandler *sax, void *user_data, |
11446 | | xmlInputReadCallback ioread, |
11447 | | xmlInputCloseCallback ioclose, |
11448 | 0 | void *ioctx, xmlCharEncoding enc) { |
11449 | 0 | xmlParserCtxtPtr ctxt; |
11450 | 0 | xmlParserInputPtr input; |
11451 | 0 | const char *encoding; |
11452 | |
|
11453 | 0 | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
11454 | 0 | if (ctxt == NULL) |
11455 | 0 | return(NULL); |
11456 | | |
11457 | 0 | encoding = xmlGetCharEncodingName(enc); |
11458 | 0 | input = xmlCtxtNewInputFromIO(ctxt, NULL, ioread, ioclose, ioctx, |
11459 | 0 | encoding, 0); |
11460 | 0 | if (input == NULL) { |
11461 | 0 | xmlFreeParserCtxt(ctxt); |
11462 | 0 | return (NULL); |
11463 | 0 | } |
11464 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
11465 | 0 | xmlFreeInputStream(input); |
11466 | 0 | xmlFreeParserCtxt(ctxt); |
11467 | 0 | return(NULL); |
11468 | 0 | } |
11469 | | |
11470 | 0 | return(ctxt); |
11471 | 0 | } |
11472 | | |
11473 | | #ifdef LIBXML_VALID_ENABLED |
11474 | | /************************************************************************ |
11475 | | * * |
11476 | | * Front ends when parsing a DTD * |
11477 | | * * |
11478 | | ************************************************************************/ |
11479 | | |
11480 | | /** |
11481 | | * Parse a DTD. |
11482 | | * |
11483 | | * Option XML_PARSE_DTDLOAD should be enabled in the parser context |
11484 | | * to make external entities work. |
11485 | | * |
11486 | | * @since 2.14.0 |
11487 | | * |
11488 | | * @param ctxt a parser context |
11489 | | * @param input a parser input |
11490 | | * @param publicId public ID of the DTD (optional) |
11491 | | * @param systemId system ID of the DTD (optional) |
11492 | | * @returns the resulting xmlDtd or NULL in case of error. |
11493 | | * `input` will be freed by the function in any case. |
11494 | | */ |
11495 | | xmlDtd * |
11496 | | xmlCtxtParseDtd(xmlParserCtxt *ctxt, xmlParserInput *input, |
11497 | | const xmlChar *publicId, const xmlChar *systemId) { |
11498 | | xmlDtdPtr ret = NULL; |
11499 | | |
11500 | | if ((ctxt == NULL) || (input == NULL)) { |
11501 | | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
11502 | | xmlFreeInputStream(input); |
11503 | | return(NULL); |
11504 | | } |
11505 | | |
11506 | | if (xmlCtxtPushInput(ctxt, input) < 0) { |
11507 | | xmlFreeInputStream(input); |
11508 | | return(NULL); |
11509 | | } |
11510 | | |
11511 | | if (publicId == NULL) |
11512 | | publicId = BAD_CAST "none"; |
11513 | | if (systemId == NULL) |
11514 | | systemId = BAD_CAST "none"; |
11515 | | |
11516 | | ctxt->myDoc = xmlNewDoc(BAD_CAST "1.0"); |
11517 | | if (ctxt->myDoc == NULL) { |
11518 | | xmlErrMemory(ctxt); |
11519 | | goto error; |
11520 | | } |
11521 | | ctxt->myDoc->properties = XML_DOC_INTERNAL; |
11522 | | ctxt->myDoc->extSubset = xmlNewDtd(ctxt->myDoc, BAD_CAST "none", |
11523 | | publicId, systemId); |
11524 | | if (ctxt->myDoc->extSubset == NULL) { |
11525 | | xmlErrMemory(ctxt); |
11526 | | xmlFreeDoc(ctxt->myDoc); |
11527 | | goto error; |
11528 | | } |
11529 | | |
11530 | | xmlParseExternalSubset(ctxt, publicId, systemId); |
11531 | | |
11532 | | if (ctxt->wellFormed) { |
11533 | | ret = ctxt->myDoc->extSubset; |
11534 | | ctxt->myDoc->extSubset = NULL; |
11535 | | if (ret != NULL) { |
11536 | | xmlNodePtr tmp; |
11537 | | |
11538 | | ret->doc = NULL; |
11539 | | tmp = ret->children; |
11540 | | while (tmp != NULL) { |
11541 | | tmp->doc = NULL; |
11542 | | tmp = tmp->next; |
11543 | | } |
11544 | | } |
11545 | | } else { |
11546 | | ret = NULL; |
11547 | | } |
11548 | | xmlFreeDoc(ctxt->myDoc); |
11549 | | ctxt->myDoc = NULL; |
11550 | | |
11551 | | error: |
11552 | | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
11553 | | |
11554 | | return(ret); |
11555 | | } |
11556 | | |
11557 | | /** |
11558 | | * Load and parse a DTD |
11559 | | * |
11560 | | * @deprecated Use #xmlCtxtParseDtd. |
11561 | | * |
11562 | | * @param sax the SAX handler block or NULL |
11563 | | * @param input an Input Buffer |
11564 | | * @param enc the charset encoding if known |
11565 | | * @returns the resulting xmlDtd or NULL in case of error. |
11566 | | * `input` will be freed by the function in any case. |
11567 | | */ |
11568 | | |
11569 | | xmlDtd * |
11570 | | xmlIOParseDTD(xmlSAXHandler *sax, xmlParserInputBuffer *input, |
11571 | | xmlCharEncoding enc) { |
11572 | | xmlDtdPtr ret = NULL; |
11573 | | xmlParserCtxtPtr ctxt; |
11574 | | xmlParserInputPtr pinput = NULL; |
11575 | | |
11576 | | if (input == NULL) |
11577 | | return(NULL); |
11578 | | |
11579 | | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
11580 | | if (ctxt == NULL) { |
11581 | | xmlFreeParserInputBuffer(input); |
11582 | | return(NULL); |
11583 | | } |
11584 | | xmlCtxtSetOptions(ctxt, XML_PARSE_DTDLOAD); |
11585 | | |
11586 | | /* |
11587 | | * generate a parser input from the I/O handler |
11588 | | */ |
11589 | | |
11590 | | pinput = xmlNewIOInputStream(ctxt, input, XML_CHAR_ENCODING_NONE); |
11591 | | if (pinput == NULL) { |
11592 | | xmlFreeParserInputBuffer(input); |
11593 | | xmlFreeParserCtxt(ctxt); |
11594 | | return(NULL); |
11595 | | } |
11596 | | |
11597 | | if (enc != XML_CHAR_ENCODING_NONE) { |
11598 | | xmlSwitchEncoding(ctxt, enc); |
11599 | | } |
11600 | | |
11601 | | ret = xmlCtxtParseDtd(ctxt, pinput, NULL, NULL); |
11602 | | |
11603 | | xmlFreeParserCtxt(ctxt); |
11604 | | return(ret); |
11605 | | } |
11606 | | |
11607 | | /** |
11608 | | * Load and parse an external subset. |
11609 | | * |
11610 | | * @deprecated Use #xmlCtxtParseDtd. |
11611 | | * |
11612 | | * @param sax the SAX handler block |
11613 | | * @param publicId public identifier of the DTD (optional) |
11614 | | * @param systemId system identifier (URL) of the DTD |
11615 | | * @returns the resulting xmlDtd or NULL in case of error. |
11616 | | */ |
11617 | | |
11618 | | xmlDtd * |
11619 | | xmlSAXParseDTD(xmlSAXHandler *sax, const xmlChar *publicId, |
11620 | | const xmlChar *systemId) { |
11621 | | xmlDtdPtr ret = NULL; |
11622 | | xmlParserCtxtPtr ctxt; |
11623 | | xmlParserInputPtr input = NULL; |
11624 | | xmlChar* systemIdCanonic; |
11625 | | |
11626 | | if ((publicId == NULL) && (systemId == NULL)) return(NULL); |
11627 | | |
11628 | | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
11629 | | if (ctxt == NULL) { |
11630 | | return(NULL); |
11631 | | } |
11632 | | xmlCtxtSetOptions(ctxt, XML_PARSE_DTDLOAD); |
11633 | | |
11634 | | /* |
11635 | | * Canonicalise the system ID |
11636 | | */ |
11637 | | systemIdCanonic = xmlCanonicPath(systemId); |
11638 | | if ((systemId != NULL) && (systemIdCanonic == NULL)) { |
11639 | | xmlFreeParserCtxt(ctxt); |
11640 | | return(NULL); |
11641 | | } |
11642 | | |
11643 | | /* |
11644 | | * Ask the Entity resolver to load the damn thing |
11645 | | */ |
11646 | | |
11647 | | if ((ctxt->sax != NULL) && (ctxt->sax->resolveEntity != NULL)) |
11648 | | input = ctxt->sax->resolveEntity(ctxt->userData, publicId, |
11649 | | systemIdCanonic); |
11650 | | if (input == NULL) { |
11651 | | xmlFreeParserCtxt(ctxt); |
11652 | | if (systemIdCanonic != NULL) |
11653 | | xmlFree(systemIdCanonic); |
11654 | | return(NULL); |
11655 | | } |
11656 | | |
11657 | | if (input->filename == NULL) |
11658 | | input->filename = (char *) systemIdCanonic; |
11659 | | else |
11660 | | xmlFree(systemIdCanonic); |
11661 | | |
11662 | | ret = xmlCtxtParseDtd(ctxt, input, publicId, systemId); |
11663 | | |
11664 | | xmlFreeParserCtxt(ctxt); |
11665 | | return(ret); |
11666 | | } |
11667 | | |
11668 | | |
11669 | | /** |
11670 | | * Load and parse an external subset. |
11671 | | * |
11672 | | * @param publicId public identifier of the DTD (optional) |
11673 | | * @param systemId system identifier (URL) of the DTD |
11674 | | * @returns the resulting xmlDtd or NULL in case of error. |
11675 | | */ |
11676 | | |
11677 | | xmlDtd * |
11678 | | xmlParseDTD(const xmlChar *publicId, const xmlChar *systemId) { |
11679 | | return(xmlSAXParseDTD(NULL, publicId, systemId)); |
11680 | | } |
11681 | | #endif /* LIBXML_VALID_ENABLED */ |
11682 | | |
11683 | | /************************************************************************ |
11684 | | * * |
11685 | | * Front ends when parsing an Entity * |
11686 | | * * |
11687 | | ************************************************************************/ |
11688 | | |
11689 | | static xmlNodePtr |
11690 | | xmlCtxtParseContentInternal(xmlParserCtxtPtr ctxt, xmlParserInputPtr input, |
11691 | 9.90k | int hasTextDecl, int buildTree) { |
11692 | 9.90k | xmlNodePtr root = NULL; |
11693 | 9.90k | xmlNodePtr list = NULL; |
11694 | 9.90k | xmlChar *rootName = BAD_CAST "#root"; |
11695 | 9.90k | int result; |
11696 | | |
11697 | 9.90k | if (buildTree) { |
11698 | 9.90k | root = xmlNewDocNode(ctxt->myDoc, NULL, rootName, NULL); |
11699 | 9.90k | if (root == NULL) { |
11700 | 1 | xmlErrMemory(ctxt); |
11701 | 1 | goto error; |
11702 | 1 | } |
11703 | 9.90k | } |
11704 | | |
11705 | 9.90k | if (xmlCtxtPushInput(ctxt, input) < 0) |
11706 | 1 | goto error; |
11707 | | |
11708 | 9.90k | nameNsPush(ctxt, rootName, NULL, NULL, 0, 0); |
11709 | 9.90k | spacePush(ctxt, -1); |
11710 | | |
11711 | 9.90k | if (buildTree) |
11712 | 9.90k | nodePush(ctxt, root); |
11713 | | |
11714 | 9.90k | if (hasTextDecl) { |
11715 | 8.24k | xmlDetectEncoding(ctxt); |
11716 | | |
11717 | | /* |
11718 | | * Parse a possible text declaration first |
11719 | | */ |
11720 | 8.24k | if ((CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l')) && |
11721 | 8.24k | (IS_BLANK_CH(NXT(5)))) { |
11722 | 2.54k | xmlParseTextDecl(ctxt); |
11723 | | /* |
11724 | | * An XML-1.0 document can't reference an entity not XML-1.0 |
11725 | | */ |
11726 | 2.54k | if ((xmlStrEqual(ctxt->version, BAD_CAST "1.0")) && |
11727 | 2.54k | (!xmlStrEqual(ctxt->input->version, BAD_CAST "1.0"))) { |
11728 | 1.29k | xmlFatalErrMsg(ctxt, XML_ERR_VERSION_MISMATCH, |
11729 | 1.29k | "Version mismatch between document and " |
11730 | 1.29k | "entity\n"); |
11731 | 1.29k | } |
11732 | 2.54k | } |
11733 | 8.24k | } |
11734 | | |
11735 | 9.90k | xmlParseContentInternal(ctxt); |
11736 | | |
11737 | 9.90k | if (ctxt->input->cur < ctxt->input->end) |
11738 | 841 | xmlFatalErr(ctxt, XML_ERR_NOT_WELL_BALANCED, NULL); |
11739 | | |
11740 | 9.90k | if ((ctxt->wellFormed) || |
11741 | 9.90k | ((ctxt->recovery) && (!xmlCtxtIsCatastrophicError(ctxt)))) { |
11742 | 2.44k | if (root != NULL) { |
11743 | 2.44k | xmlNodePtr cur; |
11744 | | |
11745 | | /* |
11746 | | * Unlink newly created node list. |
11747 | | */ |
11748 | 2.44k | list = root->children; |
11749 | 2.44k | root->children = NULL; |
11750 | 2.44k | root->last = NULL; |
11751 | 8.83k | for (cur = list; cur != NULL; cur = cur->next) |
11752 | 6.39k | cur->parent = NULL; |
11753 | 2.44k | } |
11754 | 2.44k | } |
11755 | | |
11756 | | /* |
11757 | | * Read the rest of the stream in case of errors. We want |
11758 | | * to account for the whole entity size. |
11759 | | */ |
11760 | 10.1k | do { |
11761 | 10.1k | ctxt->input->cur = ctxt->input->end; |
11762 | 10.1k | xmlParserShrink(ctxt); |
11763 | 10.1k | result = xmlParserGrow(ctxt); |
11764 | 10.1k | } while (result > 0); |
11765 | | |
11766 | 9.90k | if (buildTree) |
11767 | 9.90k | nodePop(ctxt); |
11768 | | |
11769 | 9.90k | namePop(ctxt); |
11770 | 9.90k | spacePop(ctxt); |
11771 | | |
11772 | 9.90k | xmlCtxtPopInput(ctxt); |
11773 | | |
11774 | 9.90k | error: |
11775 | 9.90k | xmlFreeNode(root); |
11776 | | |
11777 | 9.90k | return(list); |
11778 | 9.90k | } |
11779 | | |
11780 | | static void |
11781 | 11.5k | xmlCtxtParseEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr ent) { |
11782 | 11.5k | xmlParserInputPtr input; |
11783 | 11.5k | xmlNodePtr list; |
11784 | 11.5k | unsigned long consumed; |
11785 | 11.5k | int isExternal; |
11786 | 11.5k | int buildTree; |
11787 | 11.5k | int oldMinNsIndex; |
11788 | 11.5k | int oldNodelen, oldNodemem; |
11789 | | |
11790 | 11.5k | isExternal = (ent->etype == XML_EXTERNAL_GENERAL_PARSED_ENTITY); |
11791 | 11.5k | buildTree = (ctxt->node != NULL); |
11792 | | |
11793 | | /* |
11794 | | * Recursion check |
11795 | | */ |
11796 | 11.5k | if (ent->flags & XML_ENT_EXPANDING) { |
11797 | 243 | xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); |
11798 | 243 | xmlHaltParser(ctxt); |
11799 | 243 | goto error; |
11800 | 243 | } |
11801 | | |
11802 | | /* |
11803 | | * Load entity |
11804 | | */ |
11805 | 11.3k | input = xmlNewEntityInputStream(ctxt, ent); |
11806 | 11.3k | if (input == NULL) |
11807 | 1.42k | goto error; |
11808 | | |
11809 | | /* |
11810 | | * When building a tree, we need to limit the scope of namespace |
11811 | | * declarations, so that entities don't reference xmlNs structs |
11812 | | * from the parent of a reference. |
11813 | | */ |
11814 | 9.90k | oldMinNsIndex = ctxt->nsdb->minNsIndex; |
11815 | 9.90k | if (buildTree) |
11816 | 9.90k | ctxt->nsdb->minNsIndex = ctxt->nsNr; |
11817 | | |
11818 | 9.90k | oldNodelen = ctxt->nodelen; |
11819 | 9.90k | oldNodemem = ctxt->nodemem; |
11820 | 9.90k | ctxt->nodelen = 0; |
11821 | 9.90k | ctxt->nodemem = 0; |
11822 | | |
11823 | | /* |
11824 | | * Parse content |
11825 | | * |
11826 | | * This initiates a recursive call chain: |
11827 | | * |
11828 | | * - xmlCtxtParseContentInternal |
11829 | | * - xmlParseContentInternal |
11830 | | * - xmlParseReference |
11831 | | * - xmlCtxtParseEntity |
11832 | | * |
11833 | | * The nesting depth is limited by the maximum number of inputs, |
11834 | | * see xmlCtxtPushInput. |
11835 | | * |
11836 | | * It's possible to make this non-recursive (minNsIndex must be |
11837 | | * stored in the input struct) at the expense of code readability. |
11838 | | */ |
11839 | | |
11840 | 9.90k | ent->flags |= XML_ENT_EXPANDING; |
11841 | | |
11842 | 9.90k | list = xmlCtxtParseContentInternal(ctxt, input, isExternal, buildTree); |
11843 | | |
11844 | 9.90k | ent->flags &= ~XML_ENT_EXPANDING; |
11845 | | |
11846 | 9.90k | ctxt->nsdb->minNsIndex = oldMinNsIndex; |
11847 | 9.90k | ctxt->nodelen = oldNodelen; |
11848 | 9.90k | ctxt->nodemem = oldNodemem; |
11849 | | |
11850 | | /* |
11851 | | * Entity size accounting |
11852 | | */ |
11853 | 9.90k | consumed = input->consumed; |
11854 | 9.90k | xmlSaturatedAddSizeT(&consumed, input->end - input->base); |
11855 | | |
11856 | 9.90k | if ((ent->flags & XML_ENT_CHECKED) == 0) |
11857 | 9.58k | xmlSaturatedAdd(&ent->expandedSize, consumed); |
11858 | | |
11859 | 9.90k | if ((ent->flags & XML_ENT_PARSED) == 0) { |
11860 | 9.58k | if (isExternal) |
11861 | 8.00k | xmlSaturatedAdd(&ctxt->sizeentities, consumed); |
11862 | | |
11863 | 9.58k | ent->children = list; |
11864 | | |
11865 | 15.9k | while (list != NULL) { |
11866 | 6.39k | list->parent = (xmlNodePtr) ent; |
11867 | | |
11868 | | /* |
11869 | | * Downstream code like the nginx xslt module can set |
11870 | | * ctxt->myDoc->extSubset to a separate DTD, so the entity |
11871 | | * might have a different or a NULL document. |
11872 | | */ |
11873 | 6.39k | if (list->doc != ent->doc) |
11874 | 0 | xmlSetTreeDoc(list, ent->doc); |
11875 | | |
11876 | 6.39k | if (list->next == NULL) |
11877 | 2.14k | ent->last = list; |
11878 | 6.39k | list = list->next; |
11879 | 6.39k | } |
11880 | 9.58k | } else { |
11881 | 320 | xmlFreeNodeList(list); |
11882 | 320 | } |
11883 | | |
11884 | 9.90k | xmlFreeInputStream(input); |
11885 | | |
11886 | 11.5k | error: |
11887 | 11.5k | ent->flags |= XML_ENT_PARSED | XML_ENT_CHECKED; |
11888 | 11.5k | } |
11889 | | |
11890 | | /** |
11891 | | * Parse an external general entity within an existing parsing context |
11892 | | * An external general parsed entity is well-formed if it matches the |
11893 | | * production labeled extParsedEnt. |
11894 | | * |
11895 | | * [78] extParsedEnt ::= TextDecl? content |
11896 | | * |
11897 | | * @param ctxt the existing parsing context |
11898 | | * @param URL the URL for the entity to load |
11899 | | * @param ID the System ID for the entity to load |
11900 | | * @param listOut the return value for the set of parsed nodes |
11901 | | * @returns 0 if the entity is well formed, -1 in case of args problem and |
11902 | | * the parser error code otherwise |
11903 | | */ |
11904 | | |
11905 | | int |
11906 | | xmlParseCtxtExternalEntity(xmlParserCtxt *ctxt, const xmlChar *URL, |
11907 | 0 | const xmlChar *ID, xmlNode **listOut) { |
11908 | 0 | xmlParserInputPtr input; |
11909 | 0 | xmlNodePtr list; |
11910 | |
|
11911 | 0 | if (listOut != NULL) |
11912 | 0 | *listOut = NULL; |
11913 | |
|
11914 | 0 | if (ctxt == NULL) |
11915 | 0 | return(XML_ERR_ARGUMENT); |
11916 | | |
11917 | 0 | input = xmlLoadResource(ctxt, (char *) URL, (char *) ID, |
11918 | 0 | XML_RESOURCE_GENERAL_ENTITY); |
11919 | 0 | if (input == NULL) |
11920 | 0 | return(ctxt->errNo); |
11921 | | |
11922 | 0 | xmlCtxtInitializeLate(ctxt); |
11923 | |
|
11924 | 0 | list = xmlCtxtParseContentInternal(ctxt, input, /* hasTextDecl */ 1, 1); |
11925 | 0 | if (listOut != NULL) |
11926 | 0 | *listOut = list; |
11927 | 0 | else |
11928 | 0 | xmlFreeNodeList(list); |
11929 | |
|
11930 | 0 | xmlFreeInputStream(input); |
11931 | 0 | return(ctxt->errNo); |
11932 | 0 | } |
11933 | | |
11934 | | #ifdef LIBXML_SAX1_ENABLED |
11935 | | /** |
11936 | | * Parse an external general entity |
11937 | | * An external general parsed entity is well-formed if it matches the |
11938 | | * production labeled extParsedEnt. |
11939 | | * |
11940 | | * @deprecated Use #xmlParseCtxtExternalEntity. |
11941 | | * |
11942 | | * [78] extParsedEnt ::= TextDecl? content |
11943 | | * |
11944 | | * @param doc the document the chunk pertains to |
11945 | | * @param sax the SAX handler block (possibly NULL) |
11946 | | * @param user_data The user data returned on SAX callbacks (possibly NULL) |
11947 | | * @param depth Used for loop detection, use 0 |
11948 | | * @param URL the URL for the entity to load |
11949 | | * @param ID the System ID for the entity to load |
11950 | | * @param list the return value for the set of parsed nodes |
11951 | | * @returns 0 if the entity is well formed, -1 in case of args problem and |
11952 | | * the parser error code otherwise |
11953 | | */ |
11954 | | |
11955 | | int |
11956 | | xmlParseExternalEntity(xmlDoc *doc, xmlSAXHandler *sax, void *user_data, |
11957 | | int depth, const xmlChar *URL, const xmlChar *ID, xmlNode **list) { |
11958 | | xmlParserCtxtPtr ctxt; |
11959 | | int ret; |
11960 | | |
11961 | | if (list != NULL) |
11962 | | *list = NULL; |
11963 | | |
11964 | | if (doc == NULL) |
11965 | | return(XML_ERR_ARGUMENT); |
11966 | | |
11967 | | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
11968 | | if (ctxt == NULL) |
11969 | | return(XML_ERR_NO_MEMORY); |
11970 | | |
11971 | | ctxt->depth = depth; |
11972 | | ctxt->myDoc = doc; |
11973 | | ret = xmlParseCtxtExternalEntity(ctxt, URL, ID, list); |
11974 | | |
11975 | | xmlFreeParserCtxt(ctxt); |
11976 | | return(ret); |
11977 | | } |
11978 | | |
11979 | | /** |
11980 | | * Parse a well-balanced chunk of an XML document |
11981 | | * called by the parser |
11982 | | * The allowed sequence for the Well Balanced Chunk is the one defined by |
11983 | | * the content production in the XML grammar: |
11984 | | * |
11985 | | * [43] content ::= (element | CharData | Reference | CDSect | PI | |
11986 | | * Comment)* |
11987 | | * |
11988 | | * @param doc the document the chunk pertains to (must not be NULL) |
11989 | | * @param sax the SAX handler block (possibly NULL) |
11990 | | * @param user_data The user data returned on SAX callbacks (possibly NULL) |
11991 | | * @param depth Used for loop detection, use 0 |
11992 | | * @param string the input string in UTF8 or ISO-Latin (zero terminated) |
11993 | | * @param lst the return value for the set of parsed nodes |
11994 | | * @returns 0 if the chunk is well balanced, -1 in case of args problem and |
11995 | | * the parser error code otherwise |
11996 | | */ |
11997 | | |
11998 | | int |
11999 | | xmlParseBalancedChunkMemory(xmlDoc *doc, xmlSAXHandler *sax, |
12000 | | void *user_data, int depth, const xmlChar *string, xmlNode **lst) { |
12001 | | return xmlParseBalancedChunkMemoryRecover( doc, sax, user_data, |
12002 | | depth, string, lst, 0 ); |
12003 | | } |
12004 | | #endif /* LIBXML_SAX1_ENABLED */ |
12005 | | |
12006 | | /** |
12007 | | * Parse a well-balanced chunk of XML matching the 'content' production. |
12008 | | * |
12009 | | * Namespaces in scope of `node` and entities of `node`'s document are |
12010 | | * recognized. When validating, the DTD of `node`'s document is used. |
12011 | | * |
12012 | | * Always consumes `input` even in error case. |
12013 | | * |
12014 | | * @since 2.14.0 |
12015 | | * |
12016 | | * @param ctxt parser context |
12017 | | * @param input parser input |
12018 | | * @param node target node or document |
12019 | | * @param hasTextDecl whether to parse text declaration |
12020 | | * @returns a node list or NULL in case of error. |
12021 | | */ |
12022 | | xmlNode * |
12023 | | xmlCtxtParseContent(xmlParserCtxt *ctxt, xmlParserInput *input, |
12024 | 0 | xmlNode *node, int hasTextDecl) { |
12025 | 0 | xmlDocPtr doc; |
12026 | 0 | xmlNodePtr cur, list = NULL; |
12027 | 0 | int nsnr = 0; |
12028 | 0 | xmlDictPtr oldDict; |
12029 | 0 | int oldOptions, oldDictNames, oldLoadSubset; |
12030 | |
|
12031 | 0 | if ((ctxt == NULL) || (input == NULL) || (node == NULL)) { |
12032 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
12033 | 0 | goto exit; |
12034 | 0 | } |
12035 | | |
12036 | 0 | doc = node->doc; |
12037 | 0 | if (doc == NULL) { |
12038 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
12039 | 0 | goto exit; |
12040 | 0 | } |
12041 | | |
12042 | 0 | switch (node->type) { |
12043 | 0 | case XML_ELEMENT_NODE: |
12044 | 0 | case XML_DOCUMENT_NODE: |
12045 | 0 | case XML_HTML_DOCUMENT_NODE: |
12046 | 0 | break; |
12047 | | |
12048 | 0 | case XML_ATTRIBUTE_NODE: |
12049 | 0 | case XML_TEXT_NODE: |
12050 | 0 | case XML_CDATA_SECTION_NODE: |
12051 | 0 | case XML_ENTITY_REF_NODE: |
12052 | 0 | case XML_PI_NODE: |
12053 | 0 | case XML_COMMENT_NODE: |
12054 | 0 | for (cur = node->parent; cur != NULL; cur = node->parent) { |
12055 | 0 | if ((cur->type == XML_ELEMENT_NODE) || |
12056 | 0 | (cur->type == XML_DOCUMENT_NODE) || |
12057 | 0 | (cur->type == XML_HTML_DOCUMENT_NODE)) { |
12058 | 0 | node = cur; |
12059 | 0 | break; |
12060 | 0 | } |
12061 | 0 | } |
12062 | 0 | break; |
12063 | | |
12064 | 0 | default: |
12065 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
12066 | 0 | goto exit; |
12067 | 0 | } |
12068 | | |
12069 | 0 | xmlCtxtReset(ctxt); |
12070 | |
|
12071 | 0 | oldDict = ctxt->dict; |
12072 | 0 | oldOptions = ctxt->options; |
12073 | 0 | oldDictNames = ctxt->dictNames; |
12074 | 0 | oldLoadSubset = ctxt->loadsubset; |
12075 | | |
12076 | | /* |
12077 | | * Use input doc's dict if present, else assure XML_PARSE_NODICT is set. |
12078 | | */ |
12079 | 0 | if (doc->dict != NULL) { |
12080 | 0 | ctxt->dict = doc->dict; |
12081 | 0 | } else { |
12082 | 0 | ctxt->options |= XML_PARSE_NODICT; |
12083 | 0 | ctxt->dictNames = 0; |
12084 | 0 | } |
12085 | | |
12086 | | /* |
12087 | | * Disable IDs |
12088 | | */ |
12089 | 0 | ctxt->loadsubset |= XML_SKIP_IDS; |
12090 | 0 | ctxt->options |= XML_PARSE_SKIP_IDS; |
12091 | |
|
12092 | 0 | ctxt->myDoc = doc; |
12093 | |
|
12094 | 0 | #ifdef LIBXML_HTML_ENABLED |
12095 | 0 | if (ctxt->html) { |
12096 | | /* |
12097 | | * When parsing in context, it makes no sense to add implied |
12098 | | * elements like html/body/etc... |
12099 | | */ |
12100 | 0 | ctxt->options |= HTML_PARSE_NOIMPLIED; |
12101 | |
|
12102 | 0 | list = htmlCtxtParseContentInternal(ctxt, input); |
12103 | 0 | } else |
12104 | 0 | #endif |
12105 | 0 | { |
12106 | 0 | xmlCtxtInitializeLate(ctxt); |
12107 | | |
12108 | | /* |
12109 | | * initialize the SAX2 namespaces stack |
12110 | | */ |
12111 | 0 | cur = node; |
12112 | 0 | while ((cur != NULL) && (cur->type == XML_ELEMENT_NODE)) { |
12113 | 0 | xmlNsPtr ns = cur->nsDef; |
12114 | 0 | xmlHashedString hprefix, huri; |
12115 | |
|
12116 | 0 | while (ns != NULL) { |
12117 | 0 | hprefix = xmlDictLookupHashed(ctxt->dict, ns->prefix, -1); |
12118 | 0 | huri = xmlDictLookupHashed(ctxt->dict, ns->href, -1); |
12119 | 0 | if (xmlParserNsPush(ctxt, &hprefix, &huri, ns, 1) > 0) |
12120 | 0 | nsnr++; |
12121 | 0 | ns = ns->next; |
12122 | 0 | } |
12123 | 0 | cur = cur->parent; |
12124 | 0 | } |
12125 | |
|
12126 | 0 | list = xmlCtxtParseContentInternal(ctxt, input, hasTextDecl, 1); |
12127 | |
|
12128 | 0 | if (nsnr > 0) |
12129 | 0 | xmlParserNsPop(ctxt, nsnr); |
12130 | 0 | } |
12131 | |
|
12132 | 0 | ctxt->dict = oldDict; |
12133 | 0 | ctxt->options = oldOptions; |
12134 | 0 | ctxt->dictNames = oldDictNames; |
12135 | 0 | ctxt->loadsubset = oldLoadSubset; |
12136 | 0 | ctxt->myDoc = NULL; |
12137 | 0 | ctxt->node = NULL; |
12138 | |
|
12139 | 0 | exit: |
12140 | 0 | xmlFreeInputStream(input); |
12141 | 0 | return(list); |
12142 | 0 | } |
12143 | | |
12144 | | /** |
12145 | | * Parse a well-balanced chunk of an XML document |
12146 | | * within the context (DTD, namespaces, etc ...) of the given node. |
12147 | | * |
12148 | | * The allowed sequence for the data is a Well Balanced Chunk defined by |
12149 | | * the content production in the XML grammar: |
12150 | | * |
12151 | | * [43] content ::= (element | CharData | Reference | CDSect | PI | |
12152 | | * Comment)* |
12153 | | * |
12154 | | * This function assumes the encoding of `node`'s document which is |
12155 | | * typically not what you want. A better alternative is |
12156 | | * #xmlCtxtParseContent. |
12157 | | * |
12158 | | * @param node the context node |
12159 | | * @param data the input string |
12160 | | * @param datalen the input string length in bytes |
12161 | | * @param options a combination of xmlParserOption |
12162 | | * @param listOut the return value for the set of parsed nodes |
12163 | | * @returns XML_ERR_OK if the chunk is well balanced, and the parser |
12164 | | * error code otherwise |
12165 | | */ |
12166 | | xmlParserErrors |
12167 | | xmlParseInNodeContext(xmlNode *node, const char *data, int datalen, |
12168 | 0 | int options, xmlNode **listOut) { |
12169 | 0 | xmlParserCtxtPtr ctxt; |
12170 | 0 | xmlParserInputPtr input; |
12171 | 0 | xmlDocPtr doc; |
12172 | 0 | xmlNodePtr list; |
12173 | 0 | xmlParserErrors ret; |
12174 | |
|
12175 | 0 | if (listOut == NULL) |
12176 | 0 | return(XML_ERR_INTERNAL_ERROR); |
12177 | 0 | *listOut = NULL; |
12178 | |
|
12179 | 0 | if ((node == NULL) || (data == NULL) || (datalen < 0)) |
12180 | 0 | return(XML_ERR_INTERNAL_ERROR); |
12181 | | |
12182 | 0 | doc = node->doc; |
12183 | 0 | if (doc == NULL) |
12184 | 0 | return(XML_ERR_INTERNAL_ERROR); |
12185 | | |
12186 | 0 | #ifdef LIBXML_HTML_ENABLED |
12187 | 0 | if (doc->type == XML_HTML_DOCUMENT_NODE) { |
12188 | 0 | ctxt = htmlNewParserCtxt(); |
12189 | 0 | } |
12190 | 0 | else |
12191 | 0 | #endif |
12192 | 0 | ctxt = xmlNewParserCtxt(); |
12193 | |
|
12194 | 0 | if (ctxt == NULL) |
12195 | 0 | return(XML_ERR_NO_MEMORY); |
12196 | | |
12197 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, NULL, data, datalen, |
12198 | 0 | (const char *) doc->encoding, |
12199 | 0 | XML_INPUT_BUF_STATIC); |
12200 | 0 | if (input == NULL) { |
12201 | 0 | xmlFreeParserCtxt(ctxt); |
12202 | 0 | return(XML_ERR_NO_MEMORY); |
12203 | 0 | } |
12204 | | |
12205 | 0 | xmlCtxtUseOptions(ctxt, options); |
12206 | |
|
12207 | 0 | list = xmlCtxtParseContent(ctxt, input, node, /* hasTextDecl */ 0); |
12208 | |
|
12209 | 0 | if (list == NULL) { |
12210 | 0 | ret = ctxt->errNo; |
12211 | 0 | if (ret == XML_ERR_ARGUMENT) |
12212 | 0 | ret = XML_ERR_INTERNAL_ERROR; |
12213 | 0 | } else { |
12214 | 0 | ret = XML_ERR_OK; |
12215 | 0 | *listOut = list; |
12216 | 0 | } |
12217 | |
|
12218 | 0 | xmlFreeParserCtxt(ctxt); |
12219 | |
|
12220 | 0 | return(ret); |
12221 | 0 | } |
12222 | | |
12223 | | #ifdef LIBXML_SAX1_ENABLED |
12224 | | /** |
12225 | | * Parse a well-balanced chunk of an XML document |
12226 | | * |
12227 | | * The allowed sequence for the Well Balanced Chunk is the one defined by |
12228 | | * the content production in the XML grammar: |
12229 | | * |
12230 | | * [43] content ::= (element | CharData | Reference | CDSect | PI | |
12231 | | * Comment)* |
12232 | | * |
12233 | | * In case recover is set to 1, the nodelist will not be empty even if |
12234 | | * the parsed chunk is not well balanced, assuming the parsing succeeded to |
12235 | | * some extent. |
12236 | | * |
12237 | | * @param doc the document the chunk pertains to (must not be NULL) |
12238 | | * @param sax the SAX handler block (possibly NULL) |
12239 | | * @param user_data The user data returned on SAX callbacks (possibly NULL) |
12240 | | * @param depth Used for loop detection, use 0 |
12241 | | * @param string the input string in UTF8 or ISO-Latin (zero terminated) |
12242 | | * @param listOut the return value for the set of parsed nodes |
12243 | | * @param recover return nodes even if the data is broken (use 0) |
12244 | | * @returns 0 if the chunk is well balanced, or thehe parser error code |
12245 | | * otherwise. |
12246 | | */ |
12247 | | int |
12248 | | xmlParseBalancedChunkMemoryRecover(xmlDoc *doc, xmlSAXHandler *sax, |
12249 | | void *user_data, int depth, const xmlChar *string, xmlNode **listOut, |
12250 | | int recover) { |
12251 | | xmlParserCtxtPtr ctxt; |
12252 | | xmlParserInputPtr input; |
12253 | | xmlNodePtr list; |
12254 | | int ret; |
12255 | | |
12256 | | if (listOut != NULL) |
12257 | | *listOut = NULL; |
12258 | | |
12259 | | if (string == NULL) |
12260 | | return(XML_ERR_ARGUMENT); |
12261 | | |
12262 | | ctxt = xmlNewSAXParserCtxt(sax, user_data); |
12263 | | if (ctxt == NULL) |
12264 | | return(XML_ERR_NO_MEMORY); |
12265 | | |
12266 | | xmlCtxtInitializeLate(ctxt); |
12267 | | |
12268 | | ctxt->depth = depth; |
12269 | | ctxt->myDoc = doc; |
12270 | | if (recover) { |
12271 | | ctxt->options |= XML_PARSE_RECOVER; |
12272 | | ctxt->recovery = 1; |
12273 | | } |
12274 | | |
12275 | | input = xmlNewStringInputStream(ctxt, string); |
12276 | | if (input == NULL) { |
12277 | | ret = ctxt->errNo; |
12278 | | goto error; |
12279 | | } |
12280 | | |
12281 | | list = xmlCtxtParseContentInternal(ctxt, input, /* hasTextDecl */ 0, 1); |
12282 | | if (listOut != NULL) |
12283 | | *listOut = list; |
12284 | | else |
12285 | | xmlFreeNodeList(list); |
12286 | | |
12287 | | if (!ctxt->wellFormed) |
12288 | | ret = ctxt->errNo; |
12289 | | else |
12290 | | ret = XML_ERR_OK; |
12291 | | |
12292 | | error: |
12293 | | xmlFreeInputStream(input); |
12294 | | xmlFreeParserCtxt(ctxt); |
12295 | | return(ret); |
12296 | | } |
12297 | | |
12298 | | /** |
12299 | | * Parse an XML external entity out of context and build a tree. |
12300 | | * It use the given SAX function block to handle the parsing callback. |
12301 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12302 | | * |
12303 | | * @deprecated Don't use. |
12304 | | * |
12305 | | * [78] extParsedEnt ::= TextDecl? content |
12306 | | * |
12307 | | * This correspond to a "Well Balanced" chunk |
12308 | | * |
12309 | | * @param sax the SAX handler block |
12310 | | * @param filename the filename |
12311 | | * @returns the resulting document tree |
12312 | | */ |
12313 | | |
12314 | | xmlDoc * |
12315 | | xmlSAXParseEntity(xmlSAXHandler *sax, const char *filename) { |
12316 | | xmlDocPtr ret; |
12317 | | xmlParserCtxtPtr ctxt; |
12318 | | |
12319 | | ctxt = xmlCreateFileParserCtxt(filename); |
12320 | | if (ctxt == NULL) { |
12321 | | return(NULL); |
12322 | | } |
12323 | | if (sax != NULL) { |
12324 | | if (sax->initialized == XML_SAX2_MAGIC) { |
12325 | | *ctxt->sax = *sax; |
12326 | | } else { |
12327 | | memset(ctxt->sax, 0, sizeof(*ctxt->sax)); |
12328 | | memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1)); |
12329 | | } |
12330 | | ctxt->userData = NULL; |
12331 | | } |
12332 | | |
12333 | | xmlParseExtParsedEnt(ctxt); |
12334 | | |
12335 | | if (ctxt->wellFormed) { |
12336 | | ret = ctxt->myDoc; |
12337 | | } else { |
12338 | | ret = NULL; |
12339 | | xmlFreeDoc(ctxt->myDoc); |
12340 | | } |
12341 | | |
12342 | | xmlFreeParserCtxt(ctxt); |
12343 | | |
12344 | | return(ret); |
12345 | | } |
12346 | | |
12347 | | /** |
12348 | | * Parse an XML external entity out of context and build a tree. |
12349 | | * |
12350 | | * [78] extParsedEnt ::= TextDecl? content |
12351 | | * |
12352 | | * This correspond to a "Well Balanced" chunk |
12353 | | * |
12354 | | * @param filename the filename |
12355 | | * @returns the resulting document tree |
12356 | | */ |
12357 | | |
12358 | | xmlDoc * |
12359 | | xmlParseEntity(const char *filename) { |
12360 | | return(xmlSAXParseEntity(NULL, filename)); |
12361 | | } |
12362 | | #endif /* LIBXML_SAX1_ENABLED */ |
12363 | | |
12364 | | /** |
12365 | | * Create a parser context for an external entity |
12366 | | * Automatic support for ZLIB/Compress compressed document is provided |
12367 | | * by default if found at compile-time. |
12368 | | * |
12369 | | * @deprecated Don't use. |
12370 | | * |
12371 | | * @param URL the entity URL |
12372 | | * @param ID the entity PUBLIC ID |
12373 | | * @param base a possible base for the target URI |
12374 | | * @returns the new parser context or NULL |
12375 | | */ |
12376 | | xmlParserCtxt * |
12377 | | xmlCreateEntityParserCtxt(const xmlChar *URL, const xmlChar *ID, |
12378 | 0 | const xmlChar *base) { |
12379 | 0 | xmlParserCtxtPtr ctxt; |
12380 | 0 | xmlParserInputPtr input; |
12381 | 0 | xmlChar *uri = NULL; |
12382 | |
|
12383 | 0 | ctxt = xmlNewParserCtxt(); |
12384 | 0 | if (ctxt == NULL) |
12385 | 0 | return(NULL); |
12386 | | |
12387 | 0 | if (base != NULL) { |
12388 | 0 | if (xmlBuildURISafe(URL, base, &uri) < 0) |
12389 | 0 | goto error; |
12390 | 0 | if (uri != NULL) |
12391 | 0 | URL = uri; |
12392 | 0 | } |
12393 | | |
12394 | 0 | input = xmlLoadResource(ctxt, (char *) URL, (char *) ID, |
12395 | 0 | XML_RESOURCE_UNKNOWN); |
12396 | 0 | if (input == NULL) |
12397 | 0 | goto error; |
12398 | | |
12399 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12400 | 0 | xmlFreeInputStream(input); |
12401 | 0 | goto error; |
12402 | 0 | } |
12403 | | |
12404 | 0 | xmlFree(uri); |
12405 | 0 | return(ctxt); |
12406 | | |
12407 | 0 | error: |
12408 | 0 | xmlFree(uri); |
12409 | 0 | xmlFreeParserCtxt(ctxt); |
12410 | 0 | return(NULL); |
12411 | 0 | } |
12412 | | |
12413 | | /************************************************************************ |
12414 | | * * |
12415 | | * Front ends when parsing from a file * |
12416 | | * * |
12417 | | ************************************************************************/ |
12418 | | |
12419 | | /** |
12420 | | * Create a parser context for a file or URL content. |
12421 | | * Automatic support for ZLIB/Compress compressed document is provided |
12422 | | * by default if found at compile-time and for file accesses |
12423 | | * |
12424 | | * @deprecated Use #xmlNewParserCtxt and #xmlCtxtReadFile. |
12425 | | * |
12426 | | * @param filename the filename or URL |
12427 | | * @param options a combination of xmlParserOption |
12428 | | * @returns the new parser context or NULL |
12429 | | */ |
12430 | | xmlParserCtxt * |
12431 | | xmlCreateURLParserCtxt(const char *filename, int options) |
12432 | 0 | { |
12433 | 0 | xmlParserCtxtPtr ctxt; |
12434 | 0 | xmlParserInputPtr input; |
12435 | |
|
12436 | 0 | ctxt = xmlNewParserCtxt(); |
12437 | 0 | if (ctxt == NULL) |
12438 | 0 | return(NULL); |
12439 | | |
12440 | 0 | options |= XML_PARSE_UNZIP; |
12441 | |
|
12442 | 0 | xmlCtxtUseOptions(ctxt, options); |
12443 | |
|
12444 | 0 | input = xmlLoadResource(ctxt, filename, NULL, XML_RESOURCE_MAIN_DOCUMENT); |
12445 | 0 | if (input == NULL) { |
12446 | 0 | xmlFreeParserCtxt(ctxt); |
12447 | 0 | return(NULL); |
12448 | 0 | } |
12449 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12450 | 0 | xmlFreeInputStream(input); |
12451 | 0 | xmlFreeParserCtxt(ctxt); |
12452 | 0 | return(NULL); |
12453 | 0 | } |
12454 | | |
12455 | 0 | return(ctxt); |
12456 | 0 | } |
12457 | | |
12458 | | /** |
12459 | | * Create a parser context for a file content. |
12460 | | * Automatic support for ZLIB/Compress compressed document is provided |
12461 | | * by default if found at compile-time. |
12462 | | * |
12463 | | * @deprecated Use #xmlNewParserCtxt and #xmlCtxtReadFile. |
12464 | | * |
12465 | | * @param filename the filename |
12466 | | * @returns the new parser context or NULL |
12467 | | */ |
12468 | | xmlParserCtxt * |
12469 | | xmlCreateFileParserCtxt(const char *filename) |
12470 | 0 | { |
12471 | 0 | return(xmlCreateURLParserCtxt(filename, 0)); |
12472 | 0 | } |
12473 | | |
12474 | | #ifdef LIBXML_SAX1_ENABLED |
12475 | | /** |
12476 | | * Parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12477 | | * compressed document is provided by default if found at compile-time. |
12478 | | * It use the given SAX function block to handle the parsing callback. |
12479 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12480 | | * |
12481 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadFile. |
12482 | | * |
12483 | | * User data (void *) is stored within the parser context in the |
12484 | | * context's _private member, so it is available nearly everywhere in libxml |
12485 | | * |
12486 | | * @param sax the SAX handler block |
12487 | | * @param filename the filename |
12488 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12489 | | * documents |
12490 | | * @param data the userdata |
12491 | | * @returns the resulting document tree |
12492 | | */ |
12493 | | |
12494 | | xmlDoc * |
12495 | | xmlSAXParseFileWithData(xmlSAXHandler *sax, const char *filename, |
12496 | | int recovery, void *data) { |
12497 | | xmlDocPtr ret = NULL; |
12498 | | xmlParserCtxtPtr ctxt; |
12499 | | xmlParserInputPtr input; |
12500 | | |
12501 | | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
12502 | | if (ctxt == NULL) |
12503 | | return(NULL); |
12504 | | |
12505 | | if (data != NULL) |
12506 | | ctxt->_private = data; |
12507 | | |
12508 | | if (recovery) { |
12509 | | ctxt->options |= XML_PARSE_RECOVER; |
12510 | | ctxt->recovery = 1; |
12511 | | } |
12512 | | |
12513 | | if ((filename != NULL) && (filename[0] == '-') && (filename[1] == 0)) |
12514 | | input = xmlCtxtNewInputFromFd(ctxt, filename, STDIN_FILENO, NULL, 0); |
12515 | | else |
12516 | | input = xmlCtxtNewInputFromUrl(ctxt, filename, NULL, NULL, 0); |
12517 | | |
12518 | | if (input != NULL) |
12519 | | ret = xmlCtxtParseDocument(ctxt, input); |
12520 | | |
12521 | | xmlFreeParserCtxt(ctxt); |
12522 | | return(ret); |
12523 | | } |
12524 | | |
12525 | | /** |
12526 | | * Parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12527 | | * compressed document is provided by default if found at compile-time. |
12528 | | * It use the given SAX function block to handle the parsing callback. |
12529 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12530 | | * |
12531 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadFile. |
12532 | | * |
12533 | | * @param sax the SAX handler block |
12534 | | * @param filename the filename |
12535 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12536 | | * documents |
12537 | | * @returns the resulting document tree |
12538 | | */ |
12539 | | |
12540 | | xmlDoc * |
12541 | | xmlSAXParseFile(xmlSAXHandler *sax, const char *filename, |
12542 | | int recovery) { |
12543 | | return(xmlSAXParseFileWithData(sax,filename,recovery,NULL)); |
12544 | | } |
12545 | | |
12546 | | /** |
12547 | | * Parse an XML in-memory document and build a tree. |
12548 | | * In the case the document is not Well Formed, a attempt to build a |
12549 | | * tree is tried anyway |
12550 | | * |
12551 | | * @deprecated Use #xmlReadDoc with XML_PARSE_RECOVER. |
12552 | | * |
12553 | | * @param cur a pointer to an array of xmlChar |
12554 | | * @returns the resulting document tree or NULL in case of failure |
12555 | | */ |
12556 | | |
12557 | | xmlDoc * |
12558 | | xmlRecoverDoc(const xmlChar *cur) { |
12559 | | return(xmlSAXParseDoc(NULL, cur, 1)); |
12560 | | } |
12561 | | |
12562 | | /** |
12563 | | * Parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12564 | | * compressed document is provided by default if found at compile-time. |
12565 | | * |
12566 | | * @deprecated Use #xmlReadFile. |
12567 | | * |
12568 | | * @param filename the filename |
12569 | | * @returns the resulting document tree if the file was wellformed, |
12570 | | * NULL otherwise. |
12571 | | */ |
12572 | | |
12573 | | xmlDoc * |
12574 | | xmlParseFile(const char *filename) { |
12575 | | return(xmlSAXParseFile(NULL, filename, 0)); |
12576 | | } |
12577 | | |
12578 | | /** |
12579 | | * Parse an XML file and build a tree. Automatic support for ZLIB/Compress |
12580 | | * compressed document is provided by default if found at compile-time. |
12581 | | * In the case the document is not Well Formed, it attempts to build |
12582 | | * a tree anyway |
12583 | | * |
12584 | | * @deprecated Use #xmlReadFile with XML_PARSE_RECOVER. |
12585 | | * |
12586 | | * @param filename the filename |
12587 | | * @returns the resulting document tree or NULL in case of failure |
12588 | | */ |
12589 | | |
12590 | | xmlDoc * |
12591 | | xmlRecoverFile(const char *filename) { |
12592 | | return(xmlSAXParseFile(NULL, filename, 1)); |
12593 | | } |
12594 | | |
12595 | | |
12596 | | /** |
12597 | | * Setup the parser context to parse a new buffer; Clears any prior |
12598 | | * contents from the parser context. The buffer parameter must not be |
12599 | | * NULL, but the filename parameter can be |
12600 | | * |
12601 | | * @deprecated Don't use. |
12602 | | * |
12603 | | * @param ctxt an XML parser context |
12604 | | * @param buffer a xmlChar * buffer |
12605 | | * @param filename a file name |
12606 | | */ |
12607 | | void |
12608 | | xmlSetupParserForBuffer(xmlParserCtxt *ctxt, const xmlChar* buffer, |
12609 | | const char* filename) |
12610 | | { |
12611 | | xmlParserInputPtr input; |
12612 | | |
12613 | | if ((ctxt == NULL) || (buffer == NULL)) |
12614 | | return; |
12615 | | |
12616 | | xmlCtxtReset(ctxt); |
12617 | | |
12618 | | input = xmlCtxtNewInputFromString(ctxt, filename, (const char *) buffer, |
12619 | | NULL, 0); |
12620 | | if (input == NULL) |
12621 | | return; |
12622 | | if (xmlCtxtPushInput(ctxt, input) < 0) |
12623 | | xmlFreeInputStream(input); |
12624 | | } |
12625 | | |
12626 | | /** |
12627 | | * Parse an XML file and call the given SAX handler routines. |
12628 | | * Automatic support for ZLIB/Compress compressed document is provided |
12629 | | * |
12630 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadFile. |
12631 | | * |
12632 | | * @param sax a SAX handler |
12633 | | * @param user_data The user data returned on SAX callbacks |
12634 | | * @param filename a file name |
12635 | | * @returns 0 in case of success or a error number otherwise |
12636 | | */ |
12637 | | int |
12638 | | xmlSAXUserParseFile(xmlSAXHandler *sax, void *user_data, |
12639 | | const char *filename) { |
12640 | | int ret = 0; |
12641 | | xmlParserCtxtPtr ctxt; |
12642 | | |
12643 | | ctxt = xmlCreateFileParserCtxt(filename); |
12644 | | if (ctxt == NULL) return -1; |
12645 | | if (sax != NULL) { |
12646 | | if (sax->initialized == XML_SAX2_MAGIC) { |
12647 | | *ctxt->sax = *sax; |
12648 | | } else { |
12649 | | memset(ctxt->sax, 0, sizeof(*ctxt->sax)); |
12650 | | memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1)); |
12651 | | } |
12652 | | ctxt->userData = user_data; |
12653 | | } |
12654 | | |
12655 | | xmlParseDocument(ctxt); |
12656 | | |
12657 | | if (ctxt->wellFormed) |
12658 | | ret = 0; |
12659 | | else { |
12660 | | if (ctxt->errNo != 0) |
12661 | | ret = ctxt->errNo; |
12662 | | else |
12663 | | ret = -1; |
12664 | | } |
12665 | | if (ctxt->myDoc != NULL) { |
12666 | | xmlFreeDoc(ctxt->myDoc); |
12667 | | ctxt->myDoc = NULL; |
12668 | | } |
12669 | | xmlFreeParserCtxt(ctxt); |
12670 | | |
12671 | | return ret; |
12672 | | } |
12673 | | #endif /* LIBXML_SAX1_ENABLED */ |
12674 | | |
12675 | | /************************************************************************ |
12676 | | * * |
12677 | | * Front ends when parsing from memory * |
12678 | | * * |
12679 | | ************************************************************************/ |
12680 | | |
12681 | | /** |
12682 | | * Create a parser context for an XML in-memory document. The input buffer |
12683 | | * must not contain a terminating null byte. |
12684 | | * |
12685 | | * @param buffer a pointer to a char array |
12686 | | * @param size the size of the array |
12687 | | * @returns the new parser context or NULL |
12688 | | */ |
12689 | | xmlParserCtxt * |
12690 | 0 | xmlCreateMemoryParserCtxt(const char *buffer, int size) { |
12691 | 0 | xmlParserCtxtPtr ctxt; |
12692 | 0 | xmlParserInputPtr input; |
12693 | |
|
12694 | 0 | if (size < 0) |
12695 | 0 | return(NULL); |
12696 | | |
12697 | 0 | ctxt = xmlNewParserCtxt(); |
12698 | 0 | if (ctxt == NULL) |
12699 | 0 | return(NULL); |
12700 | | |
12701 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, NULL, buffer, size, NULL, 0); |
12702 | 0 | if (input == NULL) { |
12703 | 0 | xmlFreeParserCtxt(ctxt); |
12704 | 0 | return(NULL); |
12705 | 0 | } |
12706 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12707 | 0 | xmlFreeInputStream(input); |
12708 | 0 | xmlFreeParserCtxt(ctxt); |
12709 | 0 | return(NULL); |
12710 | 0 | } |
12711 | | |
12712 | 0 | return(ctxt); |
12713 | 0 | } |
12714 | | |
12715 | | #ifdef LIBXML_SAX1_ENABLED |
12716 | | /** |
12717 | | * Parse an XML in-memory block and use the given SAX function block |
12718 | | * to handle the parsing callback. If sax is NULL, fallback to the default |
12719 | | * DOM tree building routines. |
12720 | | * |
12721 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadMemory. |
12722 | | * |
12723 | | * User data (void *) is stored within the parser context in the |
12724 | | * context's _private member, so it is available nearly everywhere in libxml |
12725 | | * |
12726 | | * @param sax the SAX handler block |
12727 | | * @param buffer an pointer to a char array |
12728 | | * @param size the size of the array |
12729 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12730 | | * documents |
12731 | | * @param data the userdata |
12732 | | * @returns the resulting document tree |
12733 | | */ |
12734 | | |
12735 | | xmlDoc * |
12736 | | xmlSAXParseMemoryWithData(xmlSAXHandler *sax, const char *buffer, |
12737 | | int size, int recovery, void *data) { |
12738 | | xmlDocPtr ret = NULL; |
12739 | | xmlParserCtxtPtr ctxt; |
12740 | | xmlParserInputPtr input; |
12741 | | |
12742 | | if (size < 0) |
12743 | | return(NULL); |
12744 | | |
12745 | | ctxt = xmlNewSAXParserCtxt(sax, NULL); |
12746 | | if (ctxt == NULL) |
12747 | | return(NULL); |
12748 | | |
12749 | | if (data != NULL) |
12750 | | ctxt->_private=data; |
12751 | | |
12752 | | if (recovery) { |
12753 | | ctxt->options |= XML_PARSE_RECOVER; |
12754 | | ctxt->recovery = 1; |
12755 | | } |
12756 | | |
12757 | | input = xmlCtxtNewInputFromMemory(ctxt, NULL, buffer, size, NULL, |
12758 | | XML_INPUT_BUF_STATIC); |
12759 | | |
12760 | | if (input != NULL) |
12761 | | ret = xmlCtxtParseDocument(ctxt, input); |
12762 | | |
12763 | | xmlFreeParserCtxt(ctxt); |
12764 | | return(ret); |
12765 | | } |
12766 | | |
12767 | | /** |
12768 | | * Parse an XML in-memory block and use the given SAX function block |
12769 | | * to handle the parsing callback. If sax is NULL, fallback to the default |
12770 | | * DOM tree building routines. |
12771 | | * |
12772 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadMemory. |
12773 | | * |
12774 | | * @param sax the SAX handler block |
12775 | | * @param buffer an pointer to a char array |
12776 | | * @param size the size of the array |
12777 | | * @param recovery work in recovery mode, i.e. tries to read not Well Formed |
12778 | | * documents |
12779 | | * @returns the resulting document tree |
12780 | | */ |
12781 | | xmlDoc * |
12782 | | xmlSAXParseMemory(xmlSAXHandler *sax, const char *buffer, |
12783 | | int size, int recovery) { |
12784 | | return xmlSAXParseMemoryWithData(sax, buffer, size, recovery, NULL); |
12785 | | } |
12786 | | |
12787 | | /** |
12788 | | * Parse an XML in-memory block and build a tree. |
12789 | | * |
12790 | | * @deprecated Use #xmlReadMemory. |
12791 | | * |
12792 | | * @param buffer an pointer to a char array |
12793 | | * @param size the size of the array |
12794 | | * @returns the resulting document tree |
12795 | | */ |
12796 | | |
12797 | | xmlDoc *xmlParseMemory(const char *buffer, int size) { |
12798 | | return(xmlSAXParseMemory(NULL, buffer, size, 0)); |
12799 | | } |
12800 | | |
12801 | | /** |
12802 | | * Parse an XML in-memory block and build a tree. |
12803 | | * In the case the document is not Well Formed, an attempt to |
12804 | | * build a tree is tried anyway |
12805 | | * |
12806 | | * @deprecated Use #xmlReadMemory with XML_PARSE_RECOVER. |
12807 | | * |
12808 | | * @param buffer an pointer to a char array |
12809 | | * @param size the size of the array |
12810 | | * @returns the resulting document tree or NULL in case of error |
12811 | | */ |
12812 | | |
12813 | | xmlDoc *xmlRecoverMemory(const char *buffer, int size) { |
12814 | | return(xmlSAXParseMemory(NULL, buffer, size, 1)); |
12815 | | } |
12816 | | |
12817 | | /** |
12818 | | * Parse an XML in-memory buffer and call the given SAX handler routines. |
12819 | | * |
12820 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadMemory. |
12821 | | * |
12822 | | * @param sax a SAX handler |
12823 | | * @param user_data The user data returned on SAX callbacks |
12824 | | * @param buffer an in-memory XML document input |
12825 | | * @param size the length of the XML document in bytes |
12826 | | * @returns 0 in case of success or a error number otherwise |
12827 | | */ |
12828 | | int xmlSAXUserParseMemory(xmlSAXHandler *sax, void *user_data, |
12829 | | const char *buffer, int size) { |
12830 | | int ret = 0; |
12831 | | xmlParserCtxtPtr ctxt; |
12832 | | |
12833 | | ctxt = xmlCreateMemoryParserCtxt(buffer, size); |
12834 | | if (ctxt == NULL) return -1; |
12835 | | if (sax != NULL) { |
12836 | | if (sax->initialized == XML_SAX2_MAGIC) { |
12837 | | *ctxt->sax = *sax; |
12838 | | } else { |
12839 | | memset(ctxt->sax, 0, sizeof(*ctxt->sax)); |
12840 | | memcpy(ctxt->sax, sax, sizeof(xmlSAXHandlerV1)); |
12841 | | } |
12842 | | ctxt->userData = user_data; |
12843 | | } |
12844 | | |
12845 | | xmlParseDocument(ctxt); |
12846 | | |
12847 | | if (ctxt->wellFormed) |
12848 | | ret = 0; |
12849 | | else { |
12850 | | if (ctxt->errNo != 0) |
12851 | | ret = ctxt->errNo; |
12852 | | else |
12853 | | ret = -1; |
12854 | | } |
12855 | | if (ctxt->myDoc != NULL) { |
12856 | | xmlFreeDoc(ctxt->myDoc); |
12857 | | ctxt->myDoc = NULL; |
12858 | | } |
12859 | | xmlFreeParserCtxt(ctxt); |
12860 | | |
12861 | | return ret; |
12862 | | } |
12863 | | #endif /* LIBXML_SAX1_ENABLED */ |
12864 | | |
12865 | | /** |
12866 | | * Creates a parser context for an XML in-memory document. |
12867 | | * |
12868 | | * @param str a pointer to an array of xmlChar |
12869 | | * @returns the new parser context or NULL |
12870 | | */ |
12871 | | xmlParserCtxt * |
12872 | 0 | xmlCreateDocParserCtxt(const xmlChar *str) { |
12873 | 0 | xmlParserCtxtPtr ctxt; |
12874 | 0 | xmlParserInputPtr input; |
12875 | |
|
12876 | 0 | ctxt = xmlNewParserCtxt(); |
12877 | 0 | if (ctxt == NULL) |
12878 | 0 | return(NULL); |
12879 | | |
12880 | 0 | input = xmlCtxtNewInputFromString(ctxt, NULL, (const char *) str, NULL, 0); |
12881 | 0 | if (input == NULL) { |
12882 | 0 | xmlFreeParserCtxt(ctxt); |
12883 | 0 | return(NULL); |
12884 | 0 | } |
12885 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
12886 | 0 | xmlFreeInputStream(input); |
12887 | 0 | xmlFreeParserCtxt(ctxt); |
12888 | 0 | return(NULL); |
12889 | 0 | } |
12890 | | |
12891 | 0 | return(ctxt); |
12892 | 0 | } |
12893 | | |
12894 | | #ifdef LIBXML_SAX1_ENABLED |
12895 | | /** |
12896 | | * Parse an XML in-memory document and build a tree. |
12897 | | * It use the given SAX function block to handle the parsing callback. |
12898 | | * If sax is NULL, fallback to the default DOM tree building routines. |
12899 | | * |
12900 | | * @deprecated Use #xmlNewSAXParserCtxt and #xmlCtxtReadDoc. |
12901 | | * |
12902 | | * @param sax the SAX handler block |
12903 | | * @param cur a pointer to an array of xmlChar |
12904 | | * @param recovery work in recovery mode, i.e. tries to read no Well Formed |
12905 | | * documents |
12906 | | * @returns the resulting document tree |
12907 | | */ |
12908 | | |
12909 | | xmlDoc * |
12910 | | xmlSAXParseDoc(xmlSAXHandler *sax, const xmlChar *cur, int recovery) { |
12911 | | xmlDocPtr ret; |
12912 | | xmlParserCtxtPtr ctxt; |
12913 | | xmlSAXHandlerPtr oldsax = NULL; |
12914 | | |
12915 | | if (cur == NULL) return(NULL); |
12916 | | |
12917 | | |
12918 | | ctxt = xmlCreateDocParserCtxt(cur); |
12919 | | if (ctxt == NULL) return(NULL); |
12920 | | if (sax != NULL) { |
12921 | | oldsax = ctxt->sax; |
12922 | | ctxt->sax = sax; |
12923 | | ctxt->userData = NULL; |
12924 | | } |
12925 | | |
12926 | | xmlParseDocument(ctxt); |
12927 | | if ((ctxt->wellFormed) || recovery) ret = ctxt->myDoc; |
12928 | | else { |
12929 | | ret = NULL; |
12930 | | xmlFreeDoc(ctxt->myDoc); |
12931 | | ctxt->myDoc = NULL; |
12932 | | } |
12933 | | if (sax != NULL) |
12934 | | ctxt->sax = oldsax; |
12935 | | xmlFreeParserCtxt(ctxt); |
12936 | | |
12937 | | return(ret); |
12938 | | } |
12939 | | |
12940 | | /** |
12941 | | * Parse an XML in-memory document and build a tree. |
12942 | | * |
12943 | | * @deprecated Use #xmlReadDoc. |
12944 | | * |
12945 | | * @param cur a pointer to an array of xmlChar |
12946 | | * @returns the resulting document tree |
12947 | | */ |
12948 | | |
12949 | | xmlDoc * |
12950 | | xmlParseDoc(const xmlChar *cur) { |
12951 | | return(xmlSAXParseDoc(NULL, cur, 0)); |
12952 | | } |
12953 | | #endif /* LIBXML_SAX1_ENABLED */ |
12954 | | |
12955 | | /************************************************************************ |
12956 | | * * |
12957 | | * New set (2.6.0) of simpler and more flexible APIs * |
12958 | | * * |
12959 | | ************************************************************************/ |
12960 | | |
12961 | | /** |
12962 | | * Reset a parser context |
12963 | | * |
12964 | | * @param ctxt an XML parser context |
12965 | | */ |
12966 | | void |
12967 | | xmlCtxtReset(xmlParserCtxt *ctxt) |
12968 | 0 | { |
12969 | 0 | xmlParserInputPtr input; |
12970 | |
|
12971 | 0 | if (ctxt == NULL) |
12972 | 0 | return; |
12973 | | |
12974 | 0 | while ((input = xmlCtxtPopInput(ctxt)) != NULL) { /* Non consuming */ |
12975 | 0 | xmlFreeInputStream(input); |
12976 | 0 | } |
12977 | 0 | ctxt->inputNr = 0; |
12978 | 0 | ctxt->input = NULL; |
12979 | |
|
12980 | 0 | ctxt->spaceNr = 0; |
12981 | 0 | if (ctxt->spaceTab != NULL) { |
12982 | 0 | ctxt->spaceTab[0] = -1; |
12983 | 0 | ctxt->space = &ctxt->spaceTab[0]; |
12984 | 0 | } else { |
12985 | 0 | ctxt->space = NULL; |
12986 | 0 | } |
12987 | | |
12988 | |
|
12989 | 0 | ctxt->nodeNr = 0; |
12990 | 0 | ctxt->node = NULL; |
12991 | |
|
12992 | 0 | ctxt->nameNr = 0; |
12993 | 0 | ctxt->name = NULL; |
12994 | |
|
12995 | 0 | ctxt->nsNr = 0; |
12996 | 0 | xmlParserNsReset(ctxt->nsdb); |
12997 | |
|
12998 | 0 | if (ctxt->version != NULL) { |
12999 | 0 | xmlFree(ctxt->version); |
13000 | 0 | ctxt->version = NULL; |
13001 | 0 | } |
13002 | 0 | if (ctxt->encoding != NULL) { |
13003 | 0 | xmlFree(ctxt->encoding); |
13004 | 0 | ctxt->encoding = NULL; |
13005 | 0 | } |
13006 | 0 | if (ctxt->extSubURI != NULL) { |
13007 | 0 | xmlFree(ctxt->extSubURI); |
13008 | 0 | ctxt->extSubURI = NULL; |
13009 | 0 | } |
13010 | 0 | if (ctxt->extSubSystem != NULL) { |
13011 | 0 | xmlFree(ctxt->extSubSystem); |
13012 | 0 | ctxt->extSubSystem = NULL; |
13013 | 0 | } |
13014 | 0 | if (ctxt->directory != NULL) { |
13015 | 0 | xmlFree(ctxt->directory); |
13016 | 0 | ctxt->directory = NULL; |
13017 | 0 | } |
13018 | |
|
13019 | 0 | if (ctxt->myDoc != NULL) |
13020 | 0 | xmlFreeDoc(ctxt->myDoc); |
13021 | 0 | ctxt->myDoc = NULL; |
13022 | |
|
13023 | 0 | ctxt->standalone = -1; |
13024 | 0 | ctxt->hasExternalSubset = 0; |
13025 | 0 | ctxt->hasPErefs = 0; |
13026 | 0 | ctxt->html = ctxt->html ? 1 : 0; |
13027 | 0 | ctxt->instate = XML_PARSER_START; |
13028 | |
|
13029 | 0 | ctxt->wellFormed = 1; |
13030 | 0 | ctxt->nsWellFormed = 1; |
13031 | 0 | ctxt->disableSAX = 0; |
13032 | 0 | ctxt->valid = 1; |
13033 | 0 | ctxt->record_info = 0; |
13034 | 0 | ctxt->checkIndex = 0; |
13035 | 0 | ctxt->endCheckState = 0; |
13036 | 0 | ctxt->inSubset = 0; |
13037 | 0 | ctxt->errNo = XML_ERR_OK; |
13038 | 0 | ctxt->depth = 0; |
13039 | 0 | ctxt->catalogs = NULL; |
13040 | 0 | ctxt->sizeentities = 0; |
13041 | 0 | ctxt->sizeentcopy = 0; |
13042 | 0 | xmlInitNodeInfoSeq(&ctxt->node_seq); |
13043 | |
|
13044 | 0 | if (ctxt->attsDefault != NULL) { |
13045 | 0 | xmlHashFree(ctxt->attsDefault, xmlHashDefaultDeallocator); |
13046 | 0 | ctxt->attsDefault = NULL; |
13047 | 0 | } |
13048 | 0 | if (ctxt->attsSpecial != NULL) { |
13049 | 0 | xmlHashFree(ctxt->attsSpecial, NULL); |
13050 | 0 | ctxt->attsSpecial = NULL; |
13051 | 0 | } |
13052 | |
|
13053 | 0 | #ifdef LIBXML_CATALOG_ENABLED |
13054 | 0 | if (ctxt->catalogs != NULL) |
13055 | 0 | xmlCatalogFreeLocal(ctxt->catalogs); |
13056 | 0 | #endif |
13057 | 0 | ctxt->nbErrors = 0; |
13058 | 0 | ctxt->nbWarnings = 0; |
13059 | 0 | if (ctxt->lastError.code != XML_ERR_OK) |
13060 | 0 | xmlResetError(&ctxt->lastError); |
13061 | 0 | } |
13062 | | |
13063 | | /** |
13064 | | * Reset a push parser context |
13065 | | * |
13066 | | * @param ctxt an XML parser context |
13067 | | * @param chunk a pointer to an array of chars |
13068 | | * @param size number of chars in the array |
13069 | | * @param filename an optional file name or URI |
13070 | | * @param encoding the document encoding, or NULL |
13071 | | * @returns 0 in case of success and 1 in case of error |
13072 | | */ |
13073 | | int |
13074 | | xmlCtxtResetPush(xmlParserCtxt *ctxt, const char *chunk, |
13075 | | int size, const char *filename, const char *encoding) |
13076 | 0 | { |
13077 | 0 | xmlParserInputPtr input; |
13078 | |
|
13079 | 0 | if (ctxt == NULL) |
13080 | 0 | return(1); |
13081 | | |
13082 | 0 | xmlCtxtReset(ctxt); |
13083 | |
|
13084 | 0 | input = xmlNewPushInput(filename, chunk, size); |
13085 | 0 | if (input == NULL) |
13086 | 0 | return(1); |
13087 | | |
13088 | 0 | if (xmlCtxtPushInput(ctxt, input) < 0) { |
13089 | 0 | xmlFreeInputStream(input); |
13090 | 0 | return(1); |
13091 | 0 | } |
13092 | | |
13093 | 0 | if (encoding != NULL) |
13094 | 0 | xmlSwitchEncodingName(ctxt, encoding); |
13095 | |
|
13096 | 0 | return(0); |
13097 | 0 | } |
13098 | | |
13099 | | static int |
13100 | | xmlCtxtSetOptionsInternal(xmlParserCtxtPtr ctxt, int options, int keepMask) |
13101 | 1.00M | { |
13102 | 1.00M | int allMask; |
13103 | | |
13104 | 1.00M | if (ctxt == NULL) |
13105 | 0 | return(-1); |
13106 | | |
13107 | | /* |
13108 | | * XInclude options aren't handled by the parser. |
13109 | | * |
13110 | | * XML_PARSE_XINCLUDE |
13111 | | * XML_PARSE_NOXINCNODE |
13112 | | * XML_PARSE_NOBASEFIX |
13113 | | */ |
13114 | 1.00M | allMask = XML_PARSE_RECOVER | |
13115 | 1.00M | XML_PARSE_NOENT | |
13116 | 1.00M | XML_PARSE_DTDLOAD | |
13117 | 1.00M | XML_PARSE_DTDATTR | |
13118 | 1.00M | XML_PARSE_DTDVALID | |
13119 | 1.00M | XML_PARSE_NOERROR | |
13120 | 1.00M | XML_PARSE_NOWARNING | |
13121 | 1.00M | XML_PARSE_PEDANTIC | |
13122 | 1.00M | XML_PARSE_NOBLANKS | |
13123 | | #ifdef LIBXML_SAX1_ENABLED |
13124 | | XML_PARSE_SAX1 | |
13125 | | #endif |
13126 | 1.00M | XML_PARSE_NONET | |
13127 | 1.00M | XML_PARSE_NODICT | |
13128 | 1.00M | XML_PARSE_NSCLEAN | |
13129 | 1.00M | XML_PARSE_NOCDATA | |
13130 | 1.00M | XML_PARSE_COMPACT | |
13131 | 1.00M | XML_PARSE_OLD10 | |
13132 | 1.00M | XML_PARSE_HUGE | |
13133 | 1.00M | XML_PARSE_OLDSAX | |
13134 | 1.00M | XML_PARSE_IGNORE_ENC | |
13135 | 1.00M | XML_PARSE_BIG_LINES | |
13136 | 1.00M | XML_PARSE_NO_XXE | |
13137 | 1.00M | XML_PARSE_UNZIP | |
13138 | 1.00M | XML_PARSE_NO_SYS_CATALOG | |
13139 | 1.00M | XML_PARSE_CATALOG_PI; |
13140 | | |
13141 | 1.00M | ctxt->options = (ctxt->options & keepMask) | (options & allMask); |
13142 | | |
13143 | | /* |
13144 | | * For some options, struct members are historically the source |
13145 | | * of truth. The values are initalized from global variables and |
13146 | | * old code could also modify them directly. Several older API |
13147 | | * functions that don't take an options argument rely on these |
13148 | | * deprecated mechanisms. |
13149 | | * |
13150 | | * Once public access to struct members and the globals are |
13151 | | * disabled, we can use the options bitmask as source of |
13152 | | * truth, making all these struct members obsolete. |
13153 | | * |
13154 | | * The XML_DETECT_IDS flags is misnamed. It simply enables |
13155 | | * loading of the external subset. |
13156 | | */ |
13157 | 1.00M | ctxt->recovery = (options & XML_PARSE_RECOVER) ? 1 : 0; |
13158 | 1.00M | ctxt->replaceEntities = (options & XML_PARSE_NOENT) ? 1 : 0; |
13159 | 1.00M | ctxt->loadsubset = (options & XML_PARSE_DTDLOAD) ? XML_DETECT_IDS : 0; |
13160 | 1.00M | ctxt->loadsubset |= (options & XML_PARSE_DTDATTR) ? XML_COMPLETE_ATTRS : 0; |
13161 | 1.00M | ctxt->loadsubset |= (options & XML_PARSE_SKIP_IDS) ? XML_SKIP_IDS : 0; |
13162 | 1.00M | ctxt->validate = (options & XML_PARSE_DTDVALID) ? 1 : 0; |
13163 | 1.00M | ctxt->pedantic = (options & XML_PARSE_PEDANTIC) ? 1 : 0; |
13164 | 1.00M | ctxt->keepBlanks = (options & XML_PARSE_NOBLANKS) ? 0 : 1; |
13165 | 1.00M | ctxt->dictNames = (options & XML_PARSE_NODICT) ? 0 : 1; |
13166 | | |
13167 | 1.00M | return(options & ~allMask); |
13168 | 1.00M | } |
13169 | | |
13170 | | /** |
13171 | | * Applies the options to the parser context. Unset options are |
13172 | | * cleared. |
13173 | | * |
13174 | | * @since 2.13.0 |
13175 | | * |
13176 | | * With older versions, you can use #xmlCtxtUseOptions. |
13177 | | * |
13178 | | * @param ctxt an XML parser context |
13179 | | * @param options a bitmask of xmlParserOption values |
13180 | | * @returns 0 in case of success, the set of unknown or unimplemented options |
13181 | | * in case of error. |
13182 | | */ |
13183 | | int |
13184 | | xmlCtxtSetOptions(xmlParserCtxt *ctxt, int options) |
13185 | 0 | { |
13186 | 0 | #ifdef LIBXML_HTML_ENABLED |
13187 | 0 | if ((ctxt != NULL) && (ctxt->html)) |
13188 | 0 | return(htmlCtxtSetOptions(ctxt, options)); |
13189 | 0 | #endif |
13190 | | |
13191 | 0 | return(xmlCtxtSetOptionsInternal(ctxt, options, 0)); |
13192 | 0 | } |
13193 | | |
13194 | | /** |
13195 | | * Get the current options of the parser context. |
13196 | | * |
13197 | | * @since 2.14.0 |
13198 | | * |
13199 | | * @param ctxt an XML parser context |
13200 | | * @returns the current options set in the parser context, or -1 if ctxt is NULL. |
13201 | | */ |
13202 | | int |
13203 | | xmlCtxtGetOptions(xmlParserCtxt *ctxt) |
13204 | 0 | { |
13205 | 0 | if (ctxt == NULL) |
13206 | 0 | return(-1); |
13207 | | |
13208 | 0 | return(ctxt->options); |
13209 | 0 | } |
13210 | | |
13211 | | /** |
13212 | | * Applies the options to the parser context. The following options |
13213 | | * are never cleared and can only be enabled: |
13214 | | * |
13215 | | * - XML_PARSE_NOERROR |
13216 | | * - XML_PARSE_NOWARNING |
13217 | | * - XML_PARSE_NONET |
13218 | | * - XML_PARSE_NSCLEAN |
13219 | | * - XML_PARSE_NOCDATA |
13220 | | * - XML_PARSE_COMPACT |
13221 | | * - XML_PARSE_OLD10 |
13222 | | * - XML_PARSE_HUGE |
13223 | | * - XML_PARSE_OLDSAX |
13224 | | * - XML_PARSE_IGNORE_ENC |
13225 | | * - XML_PARSE_BIG_LINES |
13226 | | * |
13227 | | * @deprecated Use #xmlCtxtSetOptions. |
13228 | | * |
13229 | | * @param ctxt an XML parser context |
13230 | | * @param options a combination of xmlParserOption |
13231 | | * @returns 0 in case of success, the set of unknown or unimplemented options |
13232 | | * in case of error. |
13233 | | */ |
13234 | | int |
13235 | | xmlCtxtUseOptions(xmlParserCtxt *ctxt, int options) |
13236 | 1.00M | { |
13237 | 1.00M | int keepMask; |
13238 | | |
13239 | 1.00M | #ifdef LIBXML_HTML_ENABLED |
13240 | 1.00M | if ((ctxt != NULL) && (ctxt->html)) |
13241 | 0 | return(htmlCtxtUseOptions(ctxt, options)); |
13242 | 1.00M | #endif |
13243 | | |
13244 | | /* |
13245 | | * For historic reasons, some options can only be enabled. |
13246 | | */ |
13247 | 1.00M | keepMask = XML_PARSE_NOERROR | |
13248 | 1.00M | XML_PARSE_NOWARNING | |
13249 | 1.00M | XML_PARSE_NONET | |
13250 | 1.00M | XML_PARSE_NSCLEAN | |
13251 | 1.00M | XML_PARSE_NOCDATA | |
13252 | 1.00M | XML_PARSE_COMPACT | |
13253 | 1.00M | XML_PARSE_OLD10 | |
13254 | 1.00M | XML_PARSE_HUGE | |
13255 | 1.00M | XML_PARSE_OLDSAX | |
13256 | 1.00M | XML_PARSE_IGNORE_ENC | |
13257 | 1.00M | XML_PARSE_BIG_LINES; |
13258 | | |
13259 | 1.00M | return(xmlCtxtSetOptionsInternal(ctxt, options, keepMask)); |
13260 | 1.00M | } |
13261 | | |
13262 | | /** |
13263 | | * To protect against exponential entity expansion ("billion laughs"), the |
13264 | | * size of serialized output is (roughly) limited to the input size |
13265 | | * multiplied by this factor. The default value is 5. |
13266 | | * |
13267 | | * When working with documents making heavy use of entity expansion, it can |
13268 | | * be necessary to increase the value. For security reasons, this should only |
13269 | | * be considered when processing trusted input. |
13270 | | * |
13271 | | * @param ctxt an XML parser context |
13272 | | * @param maxAmpl maximum amplification factor |
13273 | | */ |
13274 | | void |
13275 | | xmlCtxtSetMaxAmplification(xmlParserCtxt *ctxt, unsigned maxAmpl) |
13276 | 0 | { |
13277 | 0 | if (ctxt == NULL) |
13278 | 0 | return; |
13279 | 0 | ctxt->maxAmpl = maxAmpl; |
13280 | 0 | } |
13281 | | |
13282 | | /** |
13283 | | * Parse an XML document and return the resulting document tree. |
13284 | | * Takes ownership of the input object. |
13285 | | * |
13286 | | * @since 2.13.0 |
13287 | | * |
13288 | | * @param ctxt an XML parser context |
13289 | | * @param input parser input |
13290 | | * @returns the resulting document tree or NULL |
13291 | | */ |
13292 | | xmlDoc * |
13293 | | xmlCtxtParseDocument(xmlParserCtxt *ctxt, xmlParserInput *input) |
13294 | 836k | { |
13295 | 836k | xmlDocPtr ret = NULL; |
13296 | | |
13297 | 836k | if ((ctxt == NULL) || (input == NULL)) { |
13298 | 0 | xmlFatalErr(ctxt, XML_ERR_ARGUMENT, NULL); |
13299 | 0 | xmlFreeInputStream(input); |
13300 | 0 | return(NULL); |
13301 | 0 | } |
13302 | | |
13303 | | /* assert(ctxt->inputNr == 0); */ |
13304 | 836k | while (ctxt->inputNr > 0) |
13305 | 0 | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
13306 | | |
13307 | 836k | if (xmlCtxtPushInput(ctxt, input) < 0) { |
13308 | 0 | xmlFreeInputStream(input); |
13309 | 0 | return(NULL); |
13310 | 0 | } |
13311 | | |
13312 | 836k | xmlParseDocument(ctxt); |
13313 | | |
13314 | 836k | ret = xmlCtxtGetDocument(ctxt); |
13315 | | |
13316 | | /* assert(ctxt->inputNr == 1); */ |
13317 | 1.67M | while (ctxt->inputNr > 0) |
13318 | 839k | xmlFreeInputStream(xmlCtxtPopInput(ctxt)); |
13319 | | |
13320 | 836k | return(ret); |
13321 | 836k | } |
13322 | | |
13323 | | /** |
13324 | | * Convenience function to parse an XML document from a |
13325 | | * zero-terminated string. |
13326 | | * |
13327 | | * See #xmlCtxtReadDoc for details. |
13328 | | * |
13329 | | * @param cur a pointer to a zero terminated string |
13330 | | * @param URL base URL (optional) |
13331 | | * @param encoding the document encoding (optional) |
13332 | | * @param options a combination of xmlParserOption |
13333 | | * @returns the resulting document tree |
13334 | | */ |
13335 | | xmlDoc * |
13336 | | xmlReadDoc(const xmlChar *cur, const char *URL, const char *encoding, |
13337 | | int options) |
13338 | 0 | { |
13339 | 0 | xmlParserCtxtPtr ctxt; |
13340 | 0 | xmlParserInputPtr input; |
13341 | 0 | xmlDocPtr doc = NULL; |
13342 | |
|
13343 | 0 | ctxt = xmlNewParserCtxt(); |
13344 | 0 | if (ctxt == NULL) |
13345 | 0 | return(NULL); |
13346 | | |
13347 | 0 | xmlCtxtUseOptions(ctxt, options); |
13348 | |
|
13349 | 0 | input = xmlCtxtNewInputFromString(ctxt, URL, (const char *) cur, encoding, |
13350 | 0 | XML_INPUT_BUF_STATIC); |
13351 | |
|
13352 | 0 | if (input != NULL) |
13353 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13354 | |
|
13355 | 0 | xmlFreeParserCtxt(ctxt); |
13356 | 0 | return(doc); |
13357 | 0 | } |
13358 | | |
13359 | | /** |
13360 | | * Convenience function to parse an XML file from the filesystem |
13361 | | * or a global, user-defined resource loader. |
13362 | | * |
13363 | | * This function always enables the XML_PARSE_UNZIP option for |
13364 | | * backward compatibility. If a "-" filename is passed, it will |
13365 | | * read from stdin. Both of these features are potentially |
13366 | | * insecure and might be removed from later versions. |
13367 | | * |
13368 | | * See #xmlCtxtReadFile for details. |
13369 | | * |
13370 | | * @param filename a file or URL |
13371 | | * @param encoding the document encoding (optional) |
13372 | | * @param options a combination of xmlParserOption |
13373 | | * @returns the resulting document tree |
13374 | | */ |
13375 | | xmlDoc * |
13376 | | xmlReadFile(const char *filename, const char *encoding, int options) |
13377 | 0 | { |
13378 | 0 | xmlParserCtxtPtr ctxt; |
13379 | 0 | xmlParserInputPtr input; |
13380 | 0 | xmlDocPtr doc = NULL; |
13381 | |
|
13382 | 0 | ctxt = xmlNewParserCtxt(); |
13383 | 0 | if (ctxt == NULL) |
13384 | 0 | return(NULL); |
13385 | | |
13386 | 0 | options |= XML_PARSE_UNZIP; |
13387 | |
|
13388 | 0 | xmlCtxtUseOptions(ctxt, options); |
13389 | | |
13390 | | /* |
13391 | | * Backward compatibility for users of command line utilities like |
13392 | | * xmlstarlet expecting "-" to mean stdin. This is dangerous and |
13393 | | * should be removed at some point. |
13394 | | */ |
13395 | 0 | if ((filename != NULL) && (filename[0] == '-') && (filename[1] == 0)) |
13396 | 0 | input = xmlCtxtNewInputFromFd(ctxt, filename, STDIN_FILENO, |
13397 | 0 | encoding, 0); |
13398 | 0 | else |
13399 | 0 | input = xmlCtxtNewInputFromUrl(ctxt, filename, NULL, encoding, 0); |
13400 | |
|
13401 | 0 | if (input != NULL) |
13402 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13403 | |
|
13404 | 0 | xmlFreeParserCtxt(ctxt); |
13405 | 0 | return(doc); |
13406 | 0 | } |
13407 | | |
13408 | | /** |
13409 | | * Parse an XML in-memory document and build a tree. The input buffer must |
13410 | | * not contain a terminating null byte. |
13411 | | * |
13412 | | * See #xmlCtxtReadMemory for details. |
13413 | | * |
13414 | | * @param buffer a pointer to a char array |
13415 | | * @param size the size of the array |
13416 | | * @param url base URL (optional) |
13417 | | * @param encoding the document encoding (optional) |
13418 | | * @param options a combination of xmlParserOption |
13419 | | * @returns the resulting document tree |
13420 | | */ |
13421 | | xmlDoc * |
13422 | | xmlReadMemory(const char *buffer, int size, const char *url, |
13423 | | const char *encoding, int options) |
13424 | 66.7k | { |
13425 | 66.7k | xmlParserCtxtPtr ctxt; |
13426 | 66.7k | xmlParserInputPtr input; |
13427 | 66.7k | xmlDocPtr doc = NULL; |
13428 | | |
13429 | 66.7k | if (size < 0) |
13430 | 0 | return(NULL); |
13431 | | |
13432 | 66.7k | ctxt = xmlNewParserCtxt(); |
13433 | 66.7k | if (ctxt == NULL) |
13434 | 0 | return(NULL); |
13435 | | |
13436 | 66.7k | xmlCtxtUseOptions(ctxt, options); |
13437 | | |
13438 | 66.7k | input = xmlCtxtNewInputFromMemory(ctxt, url, buffer, size, encoding, |
13439 | 66.7k | XML_INPUT_BUF_STATIC); |
13440 | | |
13441 | 66.7k | if (input != NULL) |
13442 | 66.7k | doc = xmlCtxtParseDocument(ctxt, input); |
13443 | | |
13444 | 66.7k | xmlFreeParserCtxt(ctxt); |
13445 | 66.7k | return(doc); |
13446 | 66.7k | } |
13447 | | |
13448 | | /** |
13449 | | * Parse an XML from a file descriptor and build a tree. |
13450 | | * |
13451 | | * See #xmlCtxtReadFd for details. |
13452 | | * |
13453 | | * NOTE that the file descriptor will not be closed when the |
13454 | | * context is freed or reset. |
13455 | | * |
13456 | | * @param fd an open file descriptor |
13457 | | * @param URL base URL (optional) |
13458 | | * @param encoding the document encoding (optional) |
13459 | | * @param options a combination of xmlParserOption |
13460 | | * @returns the resulting document tree |
13461 | | */ |
13462 | | xmlDoc * |
13463 | | xmlReadFd(int fd, const char *URL, const char *encoding, int options) |
13464 | 0 | { |
13465 | 0 | xmlParserCtxtPtr ctxt; |
13466 | 0 | xmlParserInputPtr input; |
13467 | 0 | xmlDocPtr doc = NULL; |
13468 | |
|
13469 | 0 | ctxt = xmlNewParserCtxt(); |
13470 | 0 | if (ctxt == NULL) |
13471 | 0 | return(NULL); |
13472 | | |
13473 | 0 | xmlCtxtUseOptions(ctxt, options); |
13474 | |
|
13475 | 0 | input = xmlCtxtNewInputFromFd(ctxt, URL, fd, encoding, 0); |
13476 | |
|
13477 | 0 | if (input != NULL) |
13478 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13479 | |
|
13480 | 0 | xmlFreeParserCtxt(ctxt); |
13481 | 0 | return(doc); |
13482 | 0 | } |
13483 | | |
13484 | | /** |
13485 | | * Parse an XML document from I/O functions and context and build a tree. |
13486 | | * |
13487 | | * See #xmlCtxtReadIO for details. |
13488 | | * |
13489 | | * @param ioread an I/O read function |
13490 | | * @param ioclose an I/O close function (optional) |
13491 | | * @param ioctx an I/O handler |
13492 | | * @param URL base URL (optional) |
13493 | | * @param encoding the document encoding (optional) |
13494 | | * @param options a combination of xmlParserOption |
13495 | | * @returns the resulting document tree |
13496 | | */ |
13497 | | xmlDoc * |
13498 | | xmlReadIO(xmlInputReadCallback ioread, xmlInputCloseCallback ioclose, |
13499 | | void *ioctx, const char *URL, const char *encoding, int options) |
13500 | 0 | { |
13501 | 0 | xmlParserCtxtPtr ctxt; |
13502 | 0 | xmlParserInputPtr input; |
13503 | 0 | xmlDocPtr doc = NULL; |
13504 | |
|
13505 | 0 | ctxt = xmlNewParserCtxt(); |
13506 | 0 | if (ctxt == NULL) |
13507 | 0 | return(NULL); |
13508 | | |
13509 | 0 | xmlCtxtUseOptions(ctxt, options); |
13510 | |
|
13511 | 0 | input = xmlCtxtNewInputFromIO(ctxt, URL, ioread, ioclose, ioctx, |
13512 | 0 | encoding, 0); |
13513 | |
|
13514 | 0 | if (input != NULL) |
13515 | 0 | doc = xmlCtxtParseDocument(ctxt, input); |
13516 | |
|
13517 | 0 | xmlFreeParserCtxt(ctxt); |
13518 | 0 | return(doc); |
13519 | 0 | } |
13520 | | |
13521 | | /** |
13522 | | * Parse an XML in-memory document and build a tree. |
13523 | | * |
13524 | | * `URL` is used as base to resolve external entities and for error |
13525 | | * reporting. |
13526 | | * |
13527 | | * @param ctxt an XML parser context |
13528 | | * @param str a pointer to a zero terminated string |
13529 | | * @param URL base URL (optional) |
13530 | | * @param encoding the document encoding (optional) |
13531 | | * @param options a combination of xmlParserOption |
13532 | | * @returns the resulting document tree |
13533 | | */ |
13534 | | xmlDoc * |
13535 | | xmlCtxtReadDoc(xmlParserCtxt *ctxt, const xmlChar *str, |
13536 | | const char *URL, const char *encoding, int options) |
13537 | 0 | { |
13538 | 0 | xmlParserInputPtr input; |
13539 | |
|
13540 | 0 | if (ctxt == NULL) |
13541 | 0 | return(NULL); |
13542 | | |
13543 | 0 | xmlCtxtReset(ctxt); |
13544 | 0 | xmlCtxtUseOptions(ctxt, options); |
13545 | |
|
13546 | 0 | input = xmlCtxtNewInputFromString(ctxt, URL, (const char *) str, encoding, |
13547 | 0 | XML_INPUT_BUF_STATIC); |
13548 | 0 | if (input == NULL) |
13549 | 0 | return(NULL); |
13550 | | |
13551 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13552 | 0 | } |
13553 | | |
13554 | | /** |
13555 | | * Parse an XML file from the filesystem or a global, user-defined |
13556 | | * resource loader. |
13557 | | * |
13558 | | * This function always enables the XML_PARSE_UNZIP option for |
13559 | | * backward compatibility. This feature is potentially insecure |
13560 | | * and might be removed from later versions. |
13561 | | * |
13562 | | * @param ctxt an XML parser context |
13563 | | * @param filename a file or URL |
13564 | | * @param encoding the document encoding (optional) |
13565 | | * @param options a combination of xmlParserOption |
13566 | | * @returns the resulting document tree |
13567 | | */ |
13568 | | xmlDoc * |
13569 | | xmlCtxtReadFile(xmlParserCtxt *ctxt, const char *filename, |
13570 | | const char *encoding, int options) |
13571 | 0 | { |
13572 | 0 | xmlParserInputPtr input; |
13573 | |
|
13574 | 0 | if (ctxt == NULL) |
13575 | 0 | return(NULL); |
13576 | | |
13577 | 0 | options |= XML_PARSE_UNZIP; |
13578 | |
|
13579 | 0 | xmlCtxtReset(ctxt); |
13580 | 0 | xmlCtxtUseOptions(ctxt, options); |
13581 | |
|
13582 | 0 | input = xmlCtxtNewInputFromUrl(ctxt, filename, NULL, encoding, 0); |
13583 | 0 | if (input == NULL) |
13584 | 0 | return(NULL); |
13585 | | |
13586 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13587 | 0 | } |
13588 | | |
13589 | | /** |
13590 | | * Parse an XML in-memory document and build a tree. The input buffer must |
13591 | | * not contain a terminating null byte. |
13592 | | * |
13593 | | * `URL` is used as base to resolve external entities and for error |
13594 | | * reporting. |
13595 | | * |
13596 | | * @param ctxt an XML parser context |
13597 | | * @param buffer a pointer to a char array |
13598 | | * @param size the size of the array |
13599 | | * @param URL base URL (optional) |
13600 | | * @param encoding the document encoding (optional) |
13601 | | * @param options a combination of xmlParserOption |
13602 | | * @returns the resulting document tree |
13603 | | */ |
13604 | | xmlDoc * |
13605 | | xmlCtxtReadMemory(xmlParserCtxt *ctxt, const char *buffer, int size, |
13606 | | const char *URL, const char *encoding, int options) |
13607 | 0 | { |
13608 | 0 | xmlParserInputPtr input; |
13609 | |
|
13610 | 0 | if ((ctxt == NULL) || (size < 0)) |
13611 | 0 | return(NULL); |
13612 | | |
13613 | 0 | xmlCtxtReset(ctxt); |
13614 | 0 | xmlCtxtUseOptions(ctxt, options); |
13615 | |
|
13616 | 0 | input = xmlCtxtNewInputFromMemory(ctxt, URL, buffer, size, encoding, |
13617 | 0 | XML_INPUT_BUF_STATIC); |
13618 | 0 | if (input == NULL) |
13619 | 0 | return(NULL); |
13620 | | |
13621 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13622 | 0 | } |
13623 | | |
13624 | | /** |
13625 | | * Parse an XML document from a file descriptor and build a tree. |
13626 | | * |
13627 | | * NOTE that the file descriptor will not be closed when the |
13628 | | * context is freed or reset. |
13629 | | * |
13630 | | * `URL` is used as base to resolve external entities and for error |
13631 | | * reporting. |
13632 | | * |
13633 | | * @param ctxt an XML parser context |
13634 | | * @param fd an open file descriptor |
13635 | | * @param URL base URL (optional) |
13636 | | * @param encoding the document encoding (optional) |
13637 | | * @param options a combination of xmlParserOption |
13638 | | * @returns the resulting document tree |
13639 | | */ |
13640 | | xmlDoc * |
13641 | | xmlCtxtReadFd(xmlParserCtxt *ctxt, int fd, |
13642 | | const char *URL, const char *encoding, int options) |
13643 | 0 | { |
13644 | 0 | xmlParserInputPtr input; |
13645 | |
|
13646 | 0 | if (ctxt == NULL) |
13647 | 0 | return(NULL); |
13648 | | |
13649 | 0 | xmlCtxtReset(ctxt); |
13650 | 0 | xmlCtxtUseOptions(ctxt, options); |
13651 | |
|
13652 | 0 | input = xmlCtxtNewInputFromFd(ctxt, URL, fd, encoding, 0); |
13653 | 0 | if (input == NULL) |
13654 | 0 | return(NULL); |
13655 | | |
13656 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13657 | 0 | } |
13658 | | |
13659 | | /** |
13660 | | * Parse an XML document from I/O functions and source and build a tree. |
13661 | | * This reuses the existing `ctxt` parser context |
13662 | | * |
13663 | | * `URL` is used as base to resolve external entities and for error |
13664 | | * reporting. |
13665 | | * |
13666 | | * @param ctxt an XML parser context |
13667 | | * @param ioread an I/O read function |
13668 | | * @param ioclose an I/O close function |
13669 | | * @param ioctx an I/O handler |
13670 | | * @param URL the base URL to use for the document |
13671 | | * @param encoding the document encoding, or NULL |
13672 | | * @param options a combination of xmlParserOption |
13673 | | * @returns the resulting document tree |
13674 | | */ |
13675 | | xmlDoc * |
13676 | | xmlCtxtReadIO(xmlParserCtxt *ctxt, xmlInputReadCallback ioread, |
13677 | | xmlInputCloseCallback ioclose, void *ioctx, |
13678 | | const char *URL, |
13679 | | const char *encoding, int options) |
13680 | 0 | { |
13681 | 0 | xmlParserInputPtr input; |
13682 | |
|
13683 | 0 | if (ctxt == NULL) |
13684 | 0 | return(NULL); |
13685 | | |
13686 | 0 | xmlCtxtReset(ctxt); |
13687 | 0 | xmlCtxtUseOptions(ctxt, options); |
13688 | |
|
13689 | 0 | input = xmlCtxtNewInputFromIO(ctxt, URL, ioread, ioclose, ioctx, |
13690 | 0 | encoding, 0); |
13691 | 0 | if (input == NULL) |
13692 | 0 | return(NULL); |
13693 | | |
13694 | 0 | return(xmlCtxtParseDocument(ctxt, input)); |
13695 | 0 | } |
13696 | | |