]>
Commit | Line | Data |
---|---|---|
1 | /* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd | |
2 | See the file COPYING for copying permission. | |
3 | */ | |
4 | ||
5 | #include <stddef.h> | |
6 | #include <string.h> /* memset(), memcpy() */ | |
7 | #include <assert.h> | |
8 | #include <limits.h> /* UINT_MAX */ | |
9 | #include <time.h> /* time() */ | |
10 | ||
11 | #define XML_BUILDING_EXPAT 1 | |
12 | ||
13 | #ifdef COMPILED_FROM_DSP | |
14 | #include "winconfig.h" | |
15 | #elif defined(OS2_32) | |
16 | ||
17 | #include "os2config.h" | |
18 | #include "expat.h" | |
19 | ||
20 | #elif defined(__MSDOS__) | |
21 | ||
22 | #include "dosconfig.h" | |
23 | #include "expat.h" | |
24 | ||
25 | #elif defined(MACOS_CLASSIC) | |
26 | #include "macconfig.h" | |
27 | #elif defined(__amigaos__) | |
28 | #include "amigaconfig.h" | |
29 | #elif defined(__WATCOMC__) | |
30 | #include "watcomconfig.h" | |
31 | #elif defined(HAVE_EXPAT_CONFIG_H) | |
32 | #include "expat_config.h" | |
33 | #endif /* ndef COMPILED_FROM_DSP */ | |
34 | ||
35 | #include "ascii.h" | |
36 | #include "expat.h" | |
37 | ||
38 | #ifdef XML_UNICODE | |
39 | #define XML_ENCODE_MAX XML_UTF16_ENCODE_MAX | |
40 | #define XmlConvert XmlUtf16Convert | |
41 | #define XmlGetInternalEncoding XmlGetUtf16InternalEncoding | |
42 | #define XmlGetInternalEncodingNS XmlGetUtf16InternalEncodingNS | |
43 | #define XmlEncode XmlUtf16Encode | |
44 | /* Using pointer subtraction to convert to integer type. */ | |
45 | #define MUST_CONVERT(enc, s) (!(enc)->isUtf16 || (((char *)(s) - (char *)NULL) & 1)) | |
46 | typedef unsigned short ICHAR; | |
47 | #else | |
48 | #define XML_ENCODE_MAX XML_UTF8_ENCODE_MAX | |
49 | #define XmlConvert XmlUtf8Convert | |
50 | #define XmlGetInternalEncoding XmlGetUtf8InternalEncoding | |
51 | #define XmlGetInternalEncodingNS XmlGetUtf8InternalEncodingNS | |
52 | #define XmlEncode XmlUtf8Encode | |
53 | #define MUST_CONVERT(enc, s) (!(enc)->isUtf8) | |
54 | typedef char ICHAR; | |
55 | #endif | |
56 | ||
57 | ||
58 | #ifndef XML_NS | |
59 | ||
60 | #define XmlInitEncodingNS XmlInitEncoding | |
61 | #define XmlInitUnknownEncodingNS XmlInitUnknownEncoding | |
62 | #undef XmlGetInternalEncodingNS | |
63 | #define XmlGetInternalEncodingNS XmlGetInternalEncoding | |
64 | #define XmlParseXmlDeclNS XmlParseXmlDecl | |
65 | ||
66 | #endif | |
67 | ||
68 | #ifdef XML_UNICODE | |
69 | ||
70 | #ifdef XML_UNICODE_WCHAR_T | |
71 | #define XML_T(x) (const wchar_t)x | |
72 | #define XML_L(x) L ## x | |
73 | #else | |
74 | #define XML_T(x) (const unsigned short)x | |
75 | #define XML_L(x) x | |
76 | #endif | |
77 | ||
78 | #else | |
79 | ||
80 | #define XML_T(x) x | |
81 | #define XML_L(x) x | |
82 | ||
83 | #endif | |
84 | ||
85 | /* Round up n to be a multiple of sz, where sz is a power of 2. */ | |
86 | #define ROUND_UP(n, sz) (((n) + ((sz) - 1)) & ~((sz) - 1)) | |
87 | ||
88 | /* Handle the case where memmove() doesn't exist. */ | |
89 | #ifndef HAVE_MEMMOVE | |
90 | #ifdef HAVE_BCOPY | |
91 | #define memmove(d,s,l) bcopy((s),(d),(l)) | |
92 | #else | |
93 | #error memmove does not exist on this platform, nor is a substitute available | |
94 | #endif /* HAVE_BCOPY */ | |
95 | #endif /* HAVE_MEMMOVE */ | |
96 | ||
97 | #include "internal.h" | |
98 | #include "xmltok.h" | |
99 | #include "xmlrole.h" | |
100 | ||
101 | typedef const XML_Char *KEY; | |
102 | ||
103 | typedef struct { | |
104 | KEY name; | |
105 | } NAMED; | |
106 | ||
107 | typedef struct { | |
108 | NAMED **v; | |
109 | unsigned char power; | |
110 | size_t size; | |
111 | size_t used; | |
112 | const XML_Memory_Handling_Suite *mem; | |
113 | } HASH_TABLE; | |
114 | ||
115 | /* Basic character hash algorithm, taken from Python's string hash: | |
116 | h = h * 1000003 ^ character, the constant being a prime number. | |
117 | ||
118 | */ | |
119 | #ifdef XML_UNICODE | |
120 | #define CHAR_HASH(h, c) \ | |
121 | (((h) * 0xF4243) ^ (unsigned short)(c)) | |
122 | #else | |
123 | #define CHAR_HASH(h, c) \ | |
124 | (((h) * 0xF4243) ^ (unsigned char)(c)) | |
125 | #endif | |
126 | ||
127 | /* For probing (after a collision) we need a step size relative prime | |
128 | to the hash table size, which is a power of 2. We use double-hashing, | |
129 | since we can calculate a second hash value cheaply by taking those bits | |
130 | of the first hash value that were discarded (masked out) when the table | |
131 | index was calculated: index = hash & mask, where mask = table->size - 1. | |
132 | We limit the maximum step size to table->size / 4 (mask >> 2) and make | |
133 | it odd, since odd numbers are always relative prime to a power of 2. | |
134 | */ | |
135 | #define SECOND_HASH(hash, mask, power) \ | |
136 | ((((hash) & ~(mask)) >> ((power) - 1)) & ((mask) >> 2)) | |
137 | #define PROBE_STEP(hash, mask, power) \ | |
138 | ((unsigned char)((SECOND_HASH(hash, mask, power)) | 1)) | |
139 | ||
140 | typedef struct { | |
141 | NAMED **p; | |
142 | NAMED **end; | |
143 | } HASH_TABLE_ITER; | |
144 | ||
145 | #define INIT_TAG_BUF_SIZE 32 /* must be a multiple of sizeof(XML_Char) */ | |
146 | #define INIT_DATA_BUF_SIZE 1024 | |
147 | #define INIT_ATTS_SIZE 16 | |
148 | #define INIT_ATTS_VERSION 0xFFFFFFFF | |
149 | #define INIT_BLOCK_SIZE 1024 | |
150 | #define INIT_BUFFER_SIZE 1024 | |
151 | ||
152 | #define EXPAND_SPARE 24 | |
153 | ||
154 | typedef struct binding { | |
155 | struct prefix *prefix; | |
156 | struct binding *nextTagBinding; | |
157 | struct binding *prevPrefixBinding; | |
158 | const struct attribute_id *attId; | |
159 | XML_Char *uri; | |
160 | int uriLen; | |
161 | int uriAlloc; | |
162 | } BINDING; | |
163 | ||
164 | typedef struct prefix { | |
165 | const XML_Char *name; | |
166 | BINDING *binding; | |
167 | } PREFIX; | |
168 | ||
169 | typedef struct { | |
170 | const XML_Char *str; | |
171 | const XML_Char *localPart; | |
172 | const XML_Char *prefix; | |
173 | int strLen; | |
174 | int uriLen; | |
175 | int prefixLen; | |
176 | } TAG_NAME; | |
177 | ||
178 | /* TAG represents an open element. | |
179 | The name of the element is stored in both the document and API | |
180 | encodings. The memory buffer 'buf' is a separately-allocated | |
181 | memory area which stores the name. During the XML_Parse()/ | |
182 | XMLParseBuffer() when the element is open, the memory for the 'raw' | |
183 | version of the name (in the document encoding) is shared with the | |
184 | document buffer. If the element is open across calls to | |
185 | XML_Parse()/XML_ParseBuffer(), the buffer is re-allocated to | |
186 | contain the 'raw' name as well. | |
187 | ||
188 | A parser re-uses these structures, maintaining a list of allocated | |
189 | TAG objects in a free list. | |
190 | */ | |
191 | typedef struct tag { | |
192 | struct tag *parent; /* parent of this element */ | |
193 | const char *rawName; /* tagName in the original encoding */ | |
194 | int rawNameLength; | |
195 | TAG_NAME name; /* tagName in the API encoding */ | |
196 | char *buf; /* buffer for name components */ | |
197 | char *bufEnd; /* end of the buffer */ | |
198 | BINDING *bindings; | |
199 | } TAG; | |
200 | ||
201 | typedef struct { | |
202 | const XML_Char *name; | |
203 | const XML_Char *textPtr; | |
204 | int textLen; /* length in XML_Chars */ | |
205 | int processed; /* # of processed bytes - when suspended */ | |
206 | const XML_Char *systemId; | |
207 | const XML_Char *base; | |
208 | const XML_Char *publicId; | |
209 | const XML_Char *notation; | |
210 | XML_Bool open; | |
211 | XML_Bool is_param; | |
212 | XML_Bool is_internal; /* true if declared in internal subset outside PE */ | |
213 | } ENTITY; | |
214 | ||
215 | typedef struct { | |
216 | enum XML_Content_Type type; | |
217 | enum XML_Content_Quant quant; | |
218 | const XML_Char * name; | |
219 | int firstchild; | |
220 | int lastchild; | |
221 | int childcnt; | |
222 | int nextsib; | |
223 | } CONTENT_SCAFFOLD; | |
224 | ||
225 | #define INIT_SCAFFOLD_ELEMENTS 32 | |
226 | ||
227 | typedef struct block { | |
228 | struct block *next; | |
229 | int size; | |
230 | XML_Char s[1]; | |
231 | } BLOCK; | |
232 | ||
233 | typedef struct { | |
234 | BLOCK *blocks; | |
235 | BLOCK *freeBlocks; | |
236 | const XML_Char *end; | |
237 | XML_Char *ptr; | |
238 | XML_Char *start; | |
239 | const XML_Memory_Handling_Suite *mem; | |
240 | } STRING_POOL; | |
241 | ||
242 | /* The XML_Char before the name is used to determine whether | |
243 | an attribute has been specified. */ | |
244 | typedef struct attribute_id { | |
245 | XML_Char *name; | |
246 | PREFIX *prefix; | |
247 | XML_Bool maybeTokenized; | |
248 | XML_Bool xmlns; | |
249 | } ATTRIBUTE_ID; | |
250 | ||
251 | typedef struct { | |
252 | const ATTRIBUTE_ID *id; | |
253 | XML_Bool isCdata; | |
254 | const XML_Char *value; | |
255 | } DEFAULT_ATTRIBUTE; | |
256 | ||
257 | typedef struct { | |
258 | unsigned long version; | |
259 | unsigned long hash; | |
260 | const XML_Char *uriName; | |
261 | } NS_ATT; | |
262 | ||
263 | typedef struct { | |
264 | const XML_Char *name; | |
265 | PREFIX *prefix; | |
266 | const ATTRIBUTE_ID *idAtt; | |
267 | int nDefaultAtts; | |
268 | int allocDefaultAtts; | |
269 | DEFAULT_ATTRIBUTE *defaultAtts; | |
270 | } ELEMENT_TYPE; | |
271 | ||
272 | typedef struct { | |
273 | HASH_TABLE generalEntities; | |
274 | HASH_TABLE elementTypes; | |
275 | HASH_TABLE attributeIds; | |
276 | HASH_TABLE prefixes; | |
277 | STRING_POOL pool; | |
278 | STRING_POOL entityValuePool; | |
279 | /* false once a parameter entity reference has been skipped */ | |
280 | XML_Bool keepProcessing; | |
281 | /* true once an internal or external PE reference has been encountered; | |
282 | this includes the reference to an external subset */ | |
283 | XML_Bool hasParamEntityRefs; | |
284 | XML_Bool standalone; | |
285 | #ifdef XML_DTD | |
286 | /* indicates if external PE has been read */ | |
287 | XML_Bool paramEntityRead; | |
288 | HASH_TABLE paramEntities; | |
289 | #endif /* XML_DTD */ | |
290 | PREFIX defaultPrefix; | |
291 | /* === scaffolding for building content model === */ | |
292 | XML_Bool in_eldecl; | |
293 | CONTENT_SCAFFOLD *scaffold; | |
294 | unsigned contentStringLen; | |
295 | unsigned scaffSize; | |
296 | unsigned scaffCount; | |
297 | int scaffLevel; | |
298 | int *scaffIndex; | |
299 | } DTD; | |
300 | ||
301 | typedef struct open_internal_entity { | |
302 | const char *internalEventPtr; | |
303 | const char *internalEventEndPtr; | |
304 | struct open_internal_entity *next; | |
305 | ENTITY *entity; | |
306 | int startTagLevel; | |
307 | XML_Bool betweenDecl; /* WFC: PE Between Declarations */ | |
308 | } OPEN_INTERNAL_ENTITY; | |
309 | ||
310 | typedef enum XML_Error PTRCALL Processor(XML_Parser parser, | |
311 | const char *start, | |
312 | const char *end, | |
313 | const char **endPtr); | |
314 | ||
315 | static Processor prologProcessor; | |
316 | static Processor prologInitProcessor; | |
317 | static Processor contentProcessor; | |
318 | static Processor cdataSectionProcessor; | |
319 | #ifdef XML_DTD | |
320 | static Processor ignoreSectionProcessor; | |
321 | static Processor externalParEntProcessor; | |
322 | static Processor externalParEntInitProcessor; | |
323 | static Processor entityValueProcessor; | |
324 | static Processor entityValueInitProcessor; | |
325 | #endif /* XML_DTD */ | |
326 | static Processor epilogProcessor; | |
327 | static Processor errorProcessor; | |
328 | static Processor externalEntityInitProcessor; | |
329 | static Processor externalEntityInitProcessor2; | |
330 | static Processor externalEntityInitProcessor3; | |
331 | static Processor externalEntityContentProcessor; | |
332 | static Processor internalEntityProcessor; | |
333 | ||
334 | static enum XML_Error | |
335 | handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName); | |
336 | static enum XML_Error | |
337 | processXmlDecl(XML_Parser parser, int isGeneralTextEntity, | |
338 | const char *s, const char *next); | |
339 | static enum XML_Error | |
340 | initializeEncoding(XML_Parser parser); | |
341 | static enum XML_Error | |
342 | doProlog(XML_Parser parser, const ENCODING *enc, const char *s, | |
343 | const char *end, int tok, const char *next, const char **nextPtr, | |
344 | XML_Bool haveMore); | |
345 | static enum XML_Error | |
346 | processInternalEntity(XML_Parser parser, ENTITY *entity, | |
347 | XML_Bool betweenDecl); | |
348 | static enum XML_Error | |
349 | doContent(XML_Parser parser, int startTagLevel, const ENCODING *enc, | |
350 | const char *start, const char *end, const char **endPtr, | |
351 | XML_Bool haveMore); | |
352 | static enum XML_Error | |
353 | doCdataSection(XML_Parser parser, const ENCODING *, const char **startPtr, | |
354 | const char *end, const char **nextPtr, XML_Bool haveMore); | |
355 | #ifdef XML_DTD | |
356 | static enum XML_Error | |
357 | doIgnoreSection(XML_Parser parser, const ENCODING *, const char **startPtr, | |
358 | const char *end, const char **nextPtr, XML_Bool haveMore); | |
359 | #endif /* XML_DTD */ | |
360 | ||
361 | static enum XML_Error | |
362 | storeAtts(XML_Parser parser, const ENCODING *, const char *s, | |
363 | TAG_NAME *tagNamePtr, BINDING **bindingsPtr); | |
364 | static enum XML_Error | |
365 | addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, | |
366 | const XML_Char *uri, BINDING **bindingsPtr); | |
367 | static int | |
368 | defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *, XML_Bool isCdata, | |
369 | XML_Bool isId, const XML_Char *dfltValue, XML_Parser parser); | |
370 | static enum XML_Error | |
371 | storeAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata, | |
372 | const char *, const char *, STRING_POOL *); | |
373 | static enum XML_Error | |
374 | appendAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata, | |
375 | const char *, const char *, STRING_POOL *); | |
376 | static ATTRIBUTE_ID * | |
377 | getAttributeId(XML_Parser parser, const ENCODING *enc, const char *start, | |
378 | const char *end); | |
379 | static int | |
380 | setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *); | |
381 | static enum XML_Error | |
382 | storeEntityValue(XML_Parser parser, const ENCODING *enc, const char *start, | |
383 | const char *end); | |
384 | static int | |
385 | reportProcessingInstruction(XML_Parser parser, const ENCODING *enc, | |
386 | const char *start, const char *end); | |
387 | static int | |
388 | reportComment(XML_Parser parser, const ENCODING *enc, const char *start, | |
389 | const char *end); | |
390 | static void | |
391 | reportDefault(XML_Parser parser, const ENCODING *enc, const char *start, | |
392 | const char *end); | |
393 | ||
394 | static const XML_Char * getContext(XML_Parser parser); | |
395 | static XML_Bool | |
396 | setContext(XML_Parser parser, const XML_Char *context); | |
397 | ||
398 | static void FASTCALL normalizePublicId(XML_Char *s); | |
399 | ||
400 | static DTD * dtdCreate(const XML_Memory_Handling_Suite *ms); | |
401 | /* do not call if parentParser != NULL */ | |
402 | static void dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms); | |
403 | static void | |
404 | dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms); | |
405 | static int | |
406 | dtdCopy(XML_Parser oldParser, | |
407 | DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms); | |
408 | static int | |
409 | copyEntityTable(XML_Parser oldParser, | |
410 | HASH_TABLE *, STRING_POOL *, const HASH_TABLE *); | |
411 | static NAMED * | |
412 | lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize); | |
413 | static void FASTCALL | |
414 | hashTableInit(HASH_TABLE *, const XML_Memory_Handling_Suite *ms); | |
415 | static void FASTCALL hashTableClear(HASH_TABLE *); | |
416 | static void FASTCALL hashTableDestroy(HASH_TABLE *); | |
417 | static void FASTCALL | |
418 | hashTableIterInit(HASH_TABLE_ITER *, const HASH_TABLE *); | |
419 | static NAMED * FASTCALL hashTableIterNext(HASH_TABLE_ITER *); | |
420 | ||
421 | static void FASTCALL | |
422 | poolInit(STRING_POOL *, const XML_Memory_Handling_Suite *ms); | |
423 | static void FASTCALL poolClear(STRING_POOL *); | |
424 | static void FASTCALL poolDestroy(STRING_POOL *); | |
425 | static XML_Char * | |
426 | poolAppend(STRING_POOL *pool, const ENCODING *enc, | |
427 | const char *ptr, const char *end); | |
428 | static XML_Char * | |
429 | poolStoreString(STRING_POOL *pool, const ENCODING *enc, | |
430 | const char *ptr, const char *end); | |
431 | static XML_Bool FASTCALL poolGrow(STRING_POOL *pool); | |
432 | static const XML_Char * FASTCALL | |
433 | poolCopyString(STRING_POOL *pool, const XML_Char *s); | |
434 | static const XML_Char * | |
435 | poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n); | |
436 | static const XML_Char * FASTCALL | |
437 | poolAppendString(STRING_POOL *pool, const XML_Char *s); | |
438 | ||
439 | static int FASTCALL nextScaffoldPart(XML_Parser parser); | |
440 | static XML_Content * build_model(XML_Parser parser); | |
441 | static ELEMENT_TYPE * | |
442 | getElementType(XML_Parser parser, const ENCODING *enc, | |
443 | const char *ptr, const char *end); | |
444 | ||
445 | static unsigned long generate_hash_secret_salt(void); | |
446 | static XML_Bool startParsing(XML_Parser parser); | |
447 | ||
448 | static XML_Parser | |
449 | parserCreate(const XML_Char *encodingName, | |
450 | const XML_Memory_Handling_Suite *memsuite, | |
451 | const XML_Char *nameSep, | |
452 | DTD *dtd); | |
453 | ||
454 | static void | |
455 | parserInit(XML_Parser parser, const XML_Char *encodingName); | |
456 | ||
457 | #define poolStart(pool) ((pool)->start) | |
458 | #define poolEnd(pool) ((pool)->ptr) | |
459 | #define poolLength(pool) ((pool)->ptr - (pool)->start) | |
460 | #define poolChop(pool) ((void)--(pool->ptr)) | |
461 | #define poolLastChar(pool) (((pool)->ptr)[-1]) | |
462 | #define poolDiscard(pool) ((pool)->ptr = (pool)->start) | |
463 | #define poolFinish(pool) ((pool)->start = (pool)->ptr) | |
464 | #define poolAppendChar(pool, c) \ | |
465 | (((pool)->ptr == (pool)->end && !poolGrow(pool)) \ | |
466 | ? 0 \ | |
467 | : ((*((pool)->ptr)++ = c), 1)) | |
468 | ||
469 | struct XML_ParserStruct { | |
470 | /* The first member must be userData so that the XML_GetUserData | |
471 | macro works. */ | |
472 | void *m_userData; | |
473 | void *m_handlerArg; | |
474 | char *m_buffer; | |
475 | const XML_Memory_Handling_Suite m_mem; | |
476 | /* first character to be parsed */ | |
477 | const char *m_bufferPtr; | |
478 | /* past last character to be parsed */ | |
479 | char *m_bufferEnd; | |
480 | /* allocated end of buffer */ | |
481 | const char *m_bufferLim; | |
482 | XML_Index m_parseEndByteIndex; | |
483 | const char *m_parseEndPtr; | |
484 | XML_Char *m_dataBuf; | |
485 | XML_Char *m_dataBufEnd; | |
486 | XML_StartElementHandler m_startElementHandler; | |
487 | XML_EndElementHandler m_endElementHandler; | |
488 | XML_CharacterDataHandler m_characterDataHandler; | |
489 | XML_ProcessingInstructionHandler m_processingInstructionHandler; | |
490 | XML_CommentHandler m_commentHandler; | |
491 | XML_StartCdataSectionHandler m_startCdataSectionHandler; | |
492 | XML_EndCdataSectionHandler m_endCdataSectionHandler; | |
493 | XML_DefaultHandler m_defaultHandler; | |
494 | XML_StartDoctypeDeclHandler m_startDoctypeDeclHandler; | |
495 | XML_EndDoctypeDeclHandler m_endDoctypeDeclHandler; | |
496 | XML_UnparsedEntityDeclHandler m_unparsedEntityDeclHandler; | |
497 | XML_NotationDeclHandler m_notationDeclHandler; | |
498 | XML_StartNamespaceDeclHandler m_startNamespaceDeclHandler; | |
499 | XML_EndNamespaceDeclHandler m_endNamespaceDeclHandler; | |
500 | XML_NotStandaloneHandler m_notStandaloneHandler; | |
501 | XML_ExternalEntityRefHandler m_externalEntityRefHandler; | |
502 | XML_Parser m_externalEntityRefHandlerArg; | |
503 | XML_SkippedEntityHandler m_skippedEntityHandler; | |
504 | XML_UnknownEncodingHandler m_unknownEncodingHandler; | |
505 | XML_ElementDeclHandler m_elementDeclHandler; | |
506 | XML_AttlistDeclHandler m_attlistDeclHandler; | |
507 | XML_EntityDeclHandler m_entityDeclHandler; | |
508 | XML_XmlDeclHandler m_xmlDeclHandler; | |
509 | const ENCODING *m_encoding; | |
510 | INIT_ENCODING m_initEncoding; | |
511 | const ENCODING *m_internalEncoding; | |
512 | const XML_Char *m_protocolEncodingName; | |
513 | XML_Bool m_ns; | |
514 | XML_Bool m_ns_triplets; | |
515 | void *m_unknownEncodingMem; | |
516 | void *m_unknownEncodingData; | |
517 | void *m_unknownEncodingHandlerData; | |
518 | void (XMLCALL *m_unknownEncodingRelease)(void *); | |
519 | PROLOG_STATE m_prologState; | |
520 | Processor *m_processor; | |
521 | enum XML_Error m_errorCode; | |
522 | const char *m_eventPtr; | |
523 | const char *m_eventEndPtr; | |
524 | const char *m_positionPtr; | |
525 | OPEN_INTERNAL_ENTITY *m_openInternalEntities; | |
526 | OPEN_INTERNAL_ENTITY *m_freeInternalEntities; | |
527 | XML_Bool m_defaultExpandInternalEntities; | |
528 | int m_tagLevel; | |
529 | ENTITY *m_declEntity; | |
530 | const XML_Char *m_doctypeName; | |
531 | const XML_Char *m_doctypeSysid; | |
532 | const XML_Char *m_doctypePubid; | |
533 | const XML_Char *m_declAttributeType; | |
534 | const XML_Char *m_declNotationName; | |
535 | const XML_Char *m_declNotationPublicId; | |
536 | ELEMENT_TYPE *m_declElementType; | |
537 | ATTRIBUTE_ID *m_declAttributeId; | |
538 | XML_Bool m_declAttributeIsCdata; | |
539 | XML_Bool m_declAttributeIsId; | |
540 | DTD *m_dtd; | |
541 | const XML_Char *m_curBase; | |
542 | TAG *m_tagStack; | |
543 | TAG *m_freeTagList; | |
544 | BINDING *m_inheritedBindings; | |
545 | BINDING *m_freeBindingList; | |
546 | int m_attsSize; | |
547 | int m_nSpecifiedAtts; | |
548 | int m_idAttIndex; | |
549 | ATTRIBUTE *m_atts; | |
550 | NS_ATT *m_nsAtts; | |
551 | unsigned long m_nsAttsVersion; | |
552 | unsigned char m_nsAttsPower; | |
553 | #ifdef XML_ATTR_INFO | |
554 | XML_AttrInfo *m_attInfo; | |
555 | #endif | |
556 | POSITION m_position; | |
557 | STRING_POOL m_tempPool; | |
558 | STRING_POOL m_temp2Pool; | |
559 | char *m_groupConnector; | |
560 | unsigned int m_groupSize; | |
561 | XML_Char m_namespaceSeparator; | |
562 | XML_Parser m_parentParser; | |
563 | XML_ParsingStatus m_parsingStatus; | |
564 | #ifdef XML_DTD | |
565 | XML_Bool m_isParamEntity; | |
566 | XML_Bool m_useForeignDTD; | |
567 | enum XML_ParamEntityParsing m_paramEntityParsing; | |
568 | #endif | |
569 | unsigned long m_hash_secret_salt; | |
570 | }; | |
571 | ||
572 | #define MALLOC(s) (parser->m_mem.malloc_fcn((s))) | |
573 | #define REALLOC(p,s) (parser->m_mem.realloc_fcn((p),(s))) | |
574 | #define FREE(p) (parser->m_mem.free_fcn((p))) | |
575 | ||
576 | #define userData (parser->m_userData) | |
577 | #define handlerArg (parser->m_handlerArg) | |
578 | #define startElementHandler (parser->m_startElementHandler) | |
579 | #define endElementHandler (parser->m_endElementHandler) | |
580 | #define characterDataHandler (parser->m_characterDataHandler) | |
581 | #define processingInstructionHandler \ | |
582 | (parser->m_processingInstructionHandler) | |
583 | #define commentHandler (parser->m_commentHandler) | |
584 | #define startCdataSectionHandler \ | |
585 | (parser->m_startCdataSectionHandler) | |
586 | #define endCdataSectionHandler (parser->m_endCdataSectionHandler) | |
587 | #define defaultHandler (parser->m_defaultHandler) | |
588 | #define startDoctypeDeclHandler (parser->m_startDoctypeDeclHandler) | |
589 | #define endDoctypeDeclHandler (parser->m_endDoctypeDeclHandler) | |
590 | #define unparsedEntityDeclHandler \ | |
591 | (parser->m_unparsedEntityDeclHandler) | |
592 | #define notationDeclHandler (parser->m_notationDeclHandler) | |
593 | #define startNamespaceDeclHandler \ | |
594 | (parser->m_startNamespaceDeclHandler) | |
595 | #define endNamespaceDeclHandler (parser->m_endNamespaceDeclHandler) | |
596 | #define notStandaloneHandler (parser->m_notStandaloneHandler) | |
597 | #define externalEntityRefHandler \ | |
598 | (parser->m_externalEntityRefHandler) | |
599 | #define externalEntityRefHandlerArg \ | |
600 | (parser->m_externalEntityRefHandlerArg) | |
601 | #define internalEntityRefHandler \ | |
602 | (parser->m_internalEntityRefHandler) | |
603 | #define skippedEntityHandler (parser->m_skippedEntityHandler) | |
604 | #define unknownEncodingHandler (parser->m_unknownEncodingHandler) | |
605 | #define elementDeclHandler (parser->m_elementDeclHandler) | |
606 | #define attlistDeclHandler (parser->m_attlistDeclHandler) | |
607 | #define entityDeclHandler (parser->m_entityDeclHandler) | |
608 | #define xmlDeclHandler (parser->m_xmlDeclHandler) | |
609 | #define encoding (parser->m_encoding) | |
610 | #define initEncoding (parser->m_initEncoding) | |
611 | #define internalEncoding (parser->m_internalEncoding) | |
612 | #define unknownEncodingMem (parser->m_unknownEncodingMem) | |
613 | #define unknownEncodingData (parser->m_unknownEncodingData) | |
614 | #define unknownEncodingHandlerData \ | |
615 | (parser->m_unknownEncodingHandlerData) | |
616 | #define unknownEncodingRelease (parser->m_unknownEncodingRelease) | |
617 | #define protocolEncodingName (parser->m_protocolEncodingName) | |
618 | #define ns (parser->m_ns) | |
619 | #define ns_triplets (parser->m_ns_triplets) | |
620 | #define prologState (parser->m_prologState) | |
621 | #define processor (parser->m_processor) | |
622 | #define errorCode (parser->m_errorCode) | |
623 | #define eventPtr (parser->m_eventPtr) | |
624 | #define eventEndPtr (parser->m_eventEndPtr) | |
625 | #define positionPtr (parser->m_positionPtr) | |
626 | #define position (parser->m_position) | |
627 | #define openInternalEntities (parser->m_openInternalEntities) | |
628 | #define freeInternalEntities (parser->m_freeInternalEntities) | |
629 | #define defaultExpandInternalEntities \ | |
630 | (parser->m_defaultExpandInternalEntities) | |
631 | #define tagLevel (parser->m_tagLevel) | |
632 | #define buffer (parser->m_buffer) | |
633 | #define bufferPtr (parser->m_bufferPtr) | |
634 | #define bufferEnd (parser->m_bufferEnd) | |
635 | #define parseEndByteIndex (parser->m_parseEndByteIndex) | |
636 | #define parseEndPtr (parser->m_parseEndPtr) | |
637 | #define bufferLim (parser->m_bufferLim) | |
638 | #define dataBuf (parser->m_dataBuf) | |
639 | #define dataBufEnd (parser->m_dataBufEnd) | |
640 | #define _dtd (parser->m_dtd) | |
641 | #define curBase (parser->m_curBase) | |
642 | #define declEntity (parser->m_declEntity) | |
643 | #define doctypeName (parser->m_doctypeName) | |
644 | #define doctypeSysid (parser->m_doctypeSysid) | |
645 | #define doctypePubid (parser->m_doctypePubid) | |
646 | #define declAttributeType (parser->m_declAttributeType) | |
647 | #define declNotationName (parser->m_declNotationName) | |
648 | #define declNotationPublicId (parser->m_declNotationPublicId) | |
649 | #define declElementType (parser->m_declElementType) | |
650 | #define declAttributeId (parser->m_declAttributeId) | |
651 | #define declAttributeIsCdata (parser->m_declAttributeIsCdata) | |
652 | #define declAttributeIsId (parser->m_declAttributeIsId) | |
653 | #define freeTagList (parser->m_freeTagList) | |
654 | #define freeBindingList (parser->m_freeBindingList) | |
655 | #define inheritedBindings (parser->m_inheritedBindings) | |
656 | #define tagStack (parser->m_tagStack) | |
657 | #define atts (parser->m_atts) | |
658 | #define attsSize (parser->m_attsSize) | |
659 | #define nSpecifiedAtts (parser->m_nSpecifiedAtts) | |
660 | #define idAttIndex (parser->m_idAttIndex) | |
661 | #define nsAtts (parser->m_nsAtts) | |
662 | #define nsAttsVersion (parser->m_nsAttsVersion) | |
663 | #define nsAttsPower (parser->m_nsAttsPower) | |
664 | #define attInfo (parser->m_attInfo) | |
665 | #define tempPool (parser->m_tempPool) | |
666 | #define temp2Pool (parser->m_temp2Pool) | |
667 | #define groupConnector (parser->m_groupConnector) | |
668 | #define groupSize (parser->m_groupSize) | |
669 | #define namespaceSeparator (parser->m_namespaceSeparator) | |
670 | #define parentParser (parser->m_parentParser) | |
671 | #define ps_parsing (parser->m_parsingStatus.parsing) | |
672 | #define ps_finalBuffer (parser->m_parsingStatus.finalBuffer) | |
673 | #ifdef XML_DTD | |
674 | #define isParamEntity (parser->m_isParamEntity) | |
675 | #define useForeignDTD (parser->m_useForeignDTD) | |
676 | #define paramEntityParsing (parser->m_paramEntityParsing) | |
677 | #endif /* XML_DTD */ | |
678 | #define hash_secret_salt (parser->m_hash_secret_salt) | |
679 | ||
680 | XML_Parser XMLCALL | |
681 | XML_ParserCreate(const XML_Char *encodingName) | |
682 | { | |
683 | return XML_ParserCreate_MM(encodingName, NULL, NULL); | |
684 | } | |
685 | ||
686 | XML_Parser XMLCALL | |
687 | XML_ParserCreateNS(const XML_Char *encodingName, XML_Char nsSep) | |
688 | { | |
689 | XML_Char tmp[2]; | |
690 | *tmp = nsSep; | |
691 | return XML_ParserCreate_MM(encodingName, NULL, tmp); | |
692 | } | |
693 | ||
694 | static const XML_Char implicitContext[] = { | |
695 | ASCII_x, ASCII_m, ASCII_l, ASCII_EQUALS, ASCII_h, ASCII_t, ASCII_t, ASCII_p, | |
696 | ASCII_COLON, ASCII_SLASH, ASCII_SLASH, ASCII_w, ASCII_w, ASCII_w, | |
697 | ASCII_PERIOD, ASCII_w, ASCII_3, ASCII_PERIOD, ASCII_o, ASCII_r, ASCII_g, | |
698 | ASCII_SLASH, ASCII_X, ASCII_M, ASCII_L, ASCII_SLASH, ASCII_1, ASCII_9, | |
699 | ASCII_9, ASCII_8, ASCII_SLASH, ASCII_n, ASCII_a, ASCII_m, ASCII_e, | |
700 | ASCII_s, ASCII_p, ASCII_a, ASCII_c, ASCII_e, '\0' | |
701 | }; | |
702 | ||
703 | static unsigned long | |
704 | generate_hash_secret_salt(void) | |
705 | { | |
706 | unsigned int seed = time(NULL) % UINT_MAX; | |
707 | srand(seed); | |
708 | return rand(); | |
709 | } | |
710 | ||
711 | static XML_Bool /* only valid for root parser */ | |
712 | startParsing(XML_Parser parser) | |
713 | { | |
714 | /* hash functions must be initialized before setContext() is called */ | |
715 | if (hash_secret_salt == 0) | |
716 | hash_secret_salt = generate_hash_secret_salt(); | |
717 | if (ns) { | |
718 | /* implicit context only set for root parser, since child | |
719 | parsers (i.e. external entity parsers) will inherit it | |
720 | */ | |
721 | return setContext(parser, implicitContext); | |
722 | } | |
723 | return XML_TRUE; | |
724 | } | |
725 | ||
726 | XML_Parser XMLCALL | |
727 | XML_ParserCreate_MM(const XML_Char *encodingName, | |
728 | const XML_Memory_Handling_Suite *memsuite, | |
729 | const XML_Char *nameSep) | |
730 | { | |
731 | return parserCreate(encodingName, memsuite, nameSep, NULL); | |
732 | } | |
733 | ||
734 | static XML_Parser | |
735 | parserCreate(const XML_Char *encodingName, | |
736 | const XML_Memory_Handling_Suite *memsuite, | |
737 | const XML_Char *nameSep, | |
738 | DTD *dtd) | |
739 | { | |
740 | XML_Parser parser; | |
741 | ||
742 | if (memsuite) { | |
743 | XML_Memory_Handling_Suite *mtemp; | |
744 | parser = (XML_Parser) | |
745 | memsuite->malloc_fcn(sizeof(struct XML_ParserStruct)); | |
746 | if (parser != NULL) { | |
747 | mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem); | |
748 | mtemp->malloc_fcn = memsuite->malloc_fcn; | |
749 | mtemp->realloc_fcn = memsuite->realloc_fcn; | |
750 | mtemp->free_fcn = memsuite->free_fcn; | |
751 | } | |
752 | } | |
753 | else { | |
754 | XML_Memory_Handling_Suite *mtemp; | |
755 | parser = (XML_Parser)malloc(sizeof(struct XML_ParserStruct)); | |
756 | if (parser != NULL) { | |
757 | mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem); | |
758 | mtemp->malloc_fcn = malloc; | |
759 | mtemp->realloc_fcn = realloc; | |
760 | mtemp->free_fcn = free; | |
761 | } | |
762 | } | |
763 | ||
764 | if (!parser) | |
765 | return parser; | |
766 | ||
767 | buffer = NULL; | |
768 | bufferLim = NULL; | |
769 | ||
770 | attsSize = INIT_ATTS_SIZE; | |
771 | atts = (ATTRIBUTE *)MALLOC(attsSize * sizeof(ATTRIBUTE)); | |
772 | if (atts == NULL) { | |
773 | FREE(parser); | |
774 | return NULL; | |
775 | } | |
776 | #ifdef XML_ATTR_INFO | |
777 | attInfo = (XML_AttrInfo*)MALLOC(attsSize * sizeof(XML_AttrInfo)); | |
778 | if (attInfo == NULL) { | |
779 | FREE(atts); | |
780 | FREE(parser); | |
781 | return NULL; | |
782 | } | |
783 | #endif | |
784 | dataBuf = (XML_Char *)MALLOC(INIT_DATA_BUF_SIZE * sizeof(XML_Char)); | |
785 | if (dataBuf == NULL) { | |
786 | FREE(atts); | |
787 | #ifdef XML_ATTR_INFO | |
788 | FREE(attInfo); | |
789 | #endif | |
790 | FREE(parser); | |
791 | return NULL; | |
792 | } | |
793 | dataBufEnd = dataBuf + INIT_DATA_BUF_SIZE; | |
794 | ||
795 | if (dtd) | |
796 | _dtd = dtd; | |
797 | else { | |
798 | _dtd = dtdCreate(&parser->m_mem); | |
799 | if (_dtd == NULL) { | |
800 | FREE(dataBuf); | |
801 | FREE(atts); | |
802 | #ifdef XML_ATTR_INFO | |
803 | FREE(attInfo); | |
804 | #endif | |
805 | FREE(parser); | |
806 | return NULL; | |
807 | } | |
808 | } | |
809 | ||
810 | freeBindingList = NULL; | |
811 | freeTagList = NULL; | |
812 | freeInternalEntities = NULL; | |
813 | ||
814 | groupSize = 0; | |
815 | groupConnector = NULL; | |
816 | ||
817 | unknownEncodingHandler = NULL; | |
818 | unknownEncodingHandlerData = NULL; | |
819 | ||
820 | namespaceSeparator = ASCII_EXCL; | |
821 | ns = XML_FALSE; | |
822 | ns_triplets = XML_FALSE; | |
823 | ||
824 | nsAtts = NULL; | |
825 | nsAttsVersion = 0; | |
826 | nsAttsPower = 0; | |
827 | ||
828 | poolInit(&tempPool, &(parser->m_mem)); | |
829 | poolInit(&temp2Pool, &(parser->m_mem)); | |
830 | parserInit(parser, encodingName); | |
831 | ||
832 | if (encodingName && !protocolEncodingName) { | |
833 | XML_ParserFree(parser); | |
834 | return NULL; | |
835 | } | |
836 | ||
837 | if (nameSep) { | |
838 | ns = XML_TRUE; | |
839 | internalEncoding = XmlGetInternalEncodingNS(); | |
840 | namespaceSeparator = *nameSep; | |
841 | } | |
842 | else { | |
843 | internalEncoding = XmlGetInternalEncoding(); | |
844 | } | |
845 | ||
846 | return parser; | |
847 | } | |
848 | ||
849 | static void | |
850 | parserInit(XML_Parser parser, const XML_Char *encodingName) | |
851 | { | |
852 | processor = prologInitProcessor; | |
853 | XmlPrologStateInit(&prologState); | |
854 | protocolEncodingName = (encodingName != NULL | |
855 | ? poolCopyString(&tempPool, encodingName) | |
856 | : NULL); | |
857 | curBase = NULL; | |
858 | XmlInitEncoding(&initEncoding, &encoding, 0); | |
859 | userData = NULL; | |
860 | handlerArg = NULL; | |
861 | startElementHandler = NULL; | |
862 | endElementHandler = NULL; | |
863 | characterDataHandler = NULL; | |
864 | processingInstructionHandler = NULL; | |
865 | commentHandler = NULL; | |
866 | startCdataSectionHandler = NULL; | |
867 | endCdataSectionHandler = NULL; | |
868 | defaultHandler = NULL; | |
869 | startDoctypeDeclHandler = NULL; | |
870 | endDoctypeDeclHandler = NULL; | |
871 | unparsedEntityDeclHandler = NULL; | |
872 | notationDeclHandler = NULL; | |
873 | startNamespaceDeclHandler = NULL; | |
874 | endNamespaceDeclHandler = NULL; | |
875 | notStandaloneHandler = NULL; | |
876 | externalEntityRefHandler = NULL; | |
877 | externalEntityRefHandlerArg = parser; | |
878 | skippedEntityHandler = NULL; | |
879 | elementDeclHandler = NULL; | |
880 | attlistDeclHandler = NULL; | |
881 | entityDeclHandler = NULL; | |
882 | xmlDeclHandler = NULL; | |
883 | bufferPtr = buffer; | |
884 | bufferEnd = buffer; | |
885 | parseEndByteIndex = 0; | |
886 | parseEndPtr = NULL; | |
887 | declElementType = NULL; | |
888 | declAttributeId = NULL; | |
889 | declEntity = NULL; | |
890 | doctypeName = NULL; | |
891 | doctypeSysid = NULL; | |
892 | doctypePubid = NULL; | |
893 | declAttributeType = NULL; | |
894 | declNotationName = NULL; | |
895 | declNotationPublicId = NULL; | |
896 | declAttributeIsCdata = XML_FALSE; | |
897 | declAttributeIsId = XML_FALSE; | |
898 | memset(&position, 0, sizeof(POSITION)); | |
899 | errorCode = XML_ERROR_NONE; | |
900 | eventPtr = NULL; | |
901 | eventEndPtr = NULL; | |
902 | positionPtr = NULL; | |
903 | openInternalEntities = NULL; | |
904 | defaultExpandInternalEntities = XML_TRUE; | |
905 | tagLevel = 0; | |
906 | tagStack = NULL; | |
907 | inheritedBindings = NULL; | |
908 | nSpecifiedAtts = 0; | |
909 | unknownEncodingMem = NULL; | |
910 | unknownEncodingRelease = NULL; | |
911 | unknownEncodingData = NULL; | |
912 | parentParser = NULL; | |
913 | ps_parsing = XML_INITIALIZED; | |
914 | #ifdef XML_DTD | |
915 | isParamEntity = XML_FALSE; | |
916 | useForeignDTD = XML_FALSE; | |
917 | paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER; | |
918 | #endif | |
919 | hash_secret_salt = 0; | |
920 | } | |
921 | ||
922 | /* moves list of bindings to freeBindingList */ | |
923 | static void FASTCALL | |
924 | moveToFreeBindingList(XML_Parser parser, BINDING *bindings) | |
925 | { | |
926 | while (bindings) { | |
927 | BINDING *b = bindings; | |
928 | bindings = bindings->nextTagBinding; | |
929 | b->nextTagBinding = freeBindingList; | |
930 | freeBindingList = b; | |
931 | } | |
932 | } | |
933 | ||
934 | XML_Bool XMLCALL | |
935 | XML_ParserReset(XML_Parser parser, const XML_Char *encodingName) | |
936 | { | |
937 | TAG *tStk; | |
938 | OPEN_INTERNAL_ENTITY *openEntityList; | |
939 | if (parentParser) | |
940 | return XML_FALSE; | |
941 | /* move tagStack to freeTagList */ | |
942 | tStk = tagStack; | |
943 | while (tStk) { | |
944 | TAG *tag = tStk; | |
945 | tStk = tStk->parent; | |
946 | tag->parent = freeTagList; | |
947 | moveToFreeBindingList(parser, tag->bindings); | |
948 | tag->bindings = NULL; | |
949 | freeTagList = tag; | |
950 | } | |
951 | /* move openInternalEntities to freeInternalEntities */ | |
952 | openEntityList = openInternalEntities; | |
953 | while (openEntityList) { | |
954 | OPEN_INTERNAL_ENTITY *openEntity = openEntityList; | |
955 | openEntityList = openEntity->next; | |
956 | openEntity->next = freeInternalEntities; | |
957 | freeInternalEntities = openEntity; | |
958 | } | |
959 | moveToFreeBindingList(parser, inheritedBindings); | |
960 | FREE(unknownEncodingMem); | |
961 | if (unknownEncodingRelease) | |
962 | unknownEncodingRelease(unknownEncodingData); | |
963 | poolClear(&tempPool); | |
964 | poolClear(&temp2Pool); | |
965 | parserInit(parser, encodingName); | |
966 | dtdReset(_dtd, &parser->m_mem); | |
967 | return XML_TRUE; | |
968 | } | |
969 | ||
970 | enum XML_Status XMLCALL | |
971 | XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName) | |
972 | { | |
973 | /* Block after XML_Parse()/XML_ParseBuffer() has been called. | |
974 | XXX There's no way for the caller to determine which of the | |
975 | XXX possible error cases caused the XML_STATUS_ERROR return. | |
976 | */ | |
977 | if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) | |
978 | return XML_STATUS_ERROR; | |
979 | if (encodingName == NULL) | |
980 | protocolEncodingName = NULL; | |
981 | else { | |
982 | protocolEncodingName = poolCopyString(&tempPool, encodingName); | |
983 | if (!protocolEncodingName) | |
984 | return XML_STATUS_ERROR; | |
985 | } | |
986 | return XML_STATUS_OK; | |
987 | } | |
988 | ||
989 | XML_Parser XMLCALL | |
990 | XML_ExternalEntityParserCreate(XML_Parser oldParser, | |
991 | const XML_Char *context, | |
992 | const XML_Char *encodingName) | |
993 | { | |
994 | XML_Parser parser = oldParser; | |
995 | DTD *newDtd = NULL; | |
996 | DTD *oldDtd = _dtd; | |
997 | XML_StartElementHandler oldStartElementHandler = startElementHandler; | |
998 | XML_EndElementHandler oldEndElementHandler = endElementHandler; | |
999 | XML_CharacterDataHandler oldCharacterDataHandler = characterDataHandler; | |
1000 | XML_ProcessingInstructionHandler oldProcessingInstructionHandler | |
1001 | = processingInstructionHandler; | |
1002 | XML_CommentHandler oldCommentHandler = commentHandler; | |
1003 | XML_StartCdataSectionHandler oldStartCdataSectionHandler | |
1004 | = startCdataSectionHandler; | |
1005 | XML_EndCdataSectionHandler oldEndCdataSectionHandler | |
1006 | = endCdataSectionHandler; | |
1007 | XML_DefaultHandler oldDefaultHandler = defaultHandler; | |
1008 | XML_UnparsedEntityDeclHandler oldUnparsedEntityDeclHandler | |
1009 | = unparsedEntityDeclHandler; | |
1010 | XML_NotationDeclHandler oldNotationDeclHandler = notationDeclHandler; | |
1011 | XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler | |
1012 | = startNamespaceDeclHandler; | |
1013 | XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler | |
1014 | = endNamespaceDeclHandler; | |
1015 | XML_NotStandaloneHandler oldNotStandaloneHandler = notStandaloneHandler; | |
1016 | XML_ExternalEntityRefHandler oldExternalEntityRefHandler | |
1017 | = externalEntityRefHandler; | |
1018 | XML_SkippedEntityHandler oldSkippedEntityHandler = skippedEntityHandler; | |
1019 | XML_UnknownEncodingHandler oldUnknownEncodingHandler | |
1020 | = unknownEncodingHandler; | |
1021 | XML_ElementDeclHandler oldElementDeclHandler = elementDeclHandler; | |
1022 | XML_AttlistDeclHandler oldAttlistDeclHandler = attlistDeclHandler; | |
1023 | XML_EntityDeclHandler oldEntityDeclHandler = entityDeclHandler; | |
1024 | XML_XmlDeclHandler oldXmlDeclHandler = xmlDeclHandler; | |
1025 | ELEMENT_TYPE * oldDeclElementType = declElementType; | |
1026 | ||
1027 | void *oldUserData = userData; | |
1028 | void *oldHandlerArg = handlerArg; | |
1029 | XML_Bool oldDefaultExpandInternalEntities = defaultExpandInternalEntities; | |
1030 | XML_Parser oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg; | |
1031 | #ifdef XML_DTD | |
1032 | enum XML_ParamEntityParsing oldParamEntityParsing = paramEntityParsing; | |
1033 | int oldInEntityValue = prologState.inEntityValue; | |
1034 | #endif | |
1035 | XML_Bool oldns_triplets = ns_triplets; | |
1036 | /* Note that the new parser shares the same hash secret as the old | |
1037 | parser, so that dtdCopy and copyEntityTable can lookup values | |
1038 | from hash tables associated with either parser without us having | |
1039 | to worry which hash secrets each table has. | |
1040 | */ | |
1041 | unsigned long oldhash_secret_salt = hash_secret_salt; | |
1042 | ||
1043 | #ifdef XML_DTD | |
1044 | if (!context) | |
1045 | newDtd = oldDtd; | |
1046 | #endif /* XML_DTD */ | |
1047 | ||
1048 | /* Note that the magical uses of the pre-processor to make field | |
1049 | access look more like C++ require that `parser' be overwritten | |
1050 | here. This makes this function more painful to follow than it | |
1051 | would be otherwise. | |
1052 | */ | |
1053 | if (ns) { | |
1054 | XML_Char tmp[2]; | |
1055 | *tmp = namespaceSeparator; | |
1056 | parser = parserCreate(encodingName, &parser->m_mem, tmp, newDtd); | |
1057 | } | |
1058 | else { | |
1059 | parser = parserCreate(encodingName, &parser->m_mem, NULL, newDtd); | |
1060 | } | |
1061 | ||
1062 | if (!parser) | |
1063 | return NULL; | |
1064 | ||
1065 | startElementHandler = oldStartElementHandler; | |
1066 | endElementHandler = oldEndElementHandler; | |
1067 | characterDataHandler = oldCharacterDataHandler; | |
1068 | processingInstructionHandler = oldProcessingInstructionHandler; | |
1069 | commentHandler = oldCommentHandler; | |
1070 | startCdataSectionHandler = oldStartCdataSectionHandler; | |
1071 | endCdataSectionHandler = oldEndCdataSectionHandler; | |
1072 | defaultHandler = oldDefaultHandler; | |
1073 | unparsedEntityDeclHandler = oldUnparsedEntityDeclHandler; | |
1074 | notationDeclHandler = oldNotationDeclHandler; | |
1075 | startNamespaceDeclHandler = oldStartNamespaceDeclHandler; | |
1076 | endNamespaceDeclHandler = oldEndNamespaceDeclHandler; | |
1077 | notStandaloneHandler = oldNotStandaloneHandler; | |
1078 | externalEntityRefHandler = oldExternalEntityRefHandler; | |
1079 | skippedEntityHandler = oldSkippedEntityHandler; | |
1080 | unknownEncodingHandler = oldUnknownEncodingHandler; | |
1081 | elementDeclHandler = oldElementDeclHandler; | |
1082 | attlistDeclHandler = oldAttlistDeclHandler; | |
1083 | entityDeclHandler = oldEntityDeclHandler; | |
1084 | xmlDeclHandler = oldXmlDeclHandler; | |
1085 | declElementType = oldDeclElementType; | |
1086 | userData = oldUserData; | |
1087 | if (oldUserData == oldHandlerArg) | |
1088 | handlerArg = userData; | |
1089 | else | |
1090 | handlerArg = parser; | |
1091 | if (oldExternalEntityRefHandlerArg != oldParser) | |
1092 | externalEntityRefHandlerArg = oldExternalEntityRefHandlerArg; | |
1093 | defaultExpandInternalEntities = oldDefaultExpandInternalEntities; | |
1094 | ns_triplets = oldns_triplets; | |
1095 | hash_secret_salt = oldhash_secret_salt; | |
1096 | parentParser = oldParser; | |
1097 | #ifdef XML_DTD | |
1098 | paramEntityParsing = oldParamEntityParsing; | |
1099 | prologState.inEntityValue = oldInEntityValue; | |
1100 | if (context) { | |
1101 | #endif /* XML_DTD */ | |
1102 | if (!dtdCopy(oldParser, _dtd, oldDtd, &parser->m_mem) | |
1103 | || !setContext(parser, context)) { | |
1104 | XML_ParserFree(parser); | |
1105 | return NULL; | |
1106 | } | |
1107 | processor = externalEntityInitProcessor; | |
1108 | #ifdef XML_DTD | |
1109 | } | |
1110 | else { | |
1111 | /* The DTD instance referenced by _dtd is shared between the document's | |
1112 | root parser and external PE parsers, therefore one does not need to | |
1113 | call setContext. In addition, one also *must* not call setContext, | |
1114 | because this would overwrite existing prefix->binding pointers in | |
1115 | _dtd with ones that get destroyed with the external PE parser. | |
1116 | This would leave those prefixes with dangling pointers. | |
1117 | */ | |
1118 | isParamEntity = XML_TRUE; | |
1119 | XmlPrologStateInitExternalEntity(&prologState); | |
1120 | processor = externalParEntInitProcessor; | |
1121 | } | |
1122 | #endif /* XML_DTD */ | |
1123 | return parser; | |
1124 | } | |
1125 | ||
1126 | static void FASTCALL | |
1127 | destroyBindings(BINDING *bindings, XML_Parser parser) | |
1128 | { | |
1129 | for (;;) { | |
1130 | BINDING *b = bindings; | |
1131 | if (!b) | |
1132 | break; | |
1133 | bindings = b->nextTagBinding; | |
1134 | FREE(b->uri); | |
1135 | FREE(b); | |
1136 | } | |
1137 | } | |
1138 | ||
1139 | void XMLCALL | |
1140 | XML_ParserFree(XML_Parser parser) | |
1141 | { | |
1142 | TAG *tagList; | |
1143 | OPEN_INTERNAL_ENTITY *entityList; | |
1144 | if (parser == NULL) | |
1145 | return; | |
1146 | /* free tagStack and freeTagList */ | |
1147 | tagList = tagStack; | |
1148 | for (;;) { | |
1149 | TAG *p; | |
1150 | if (tagList == NULL) { | |
1151 | if (freeTagList == NULL) | |
1152 | break; | |
1153 | tagList = freeTagList; | |
1154 | freeTagList = NULL; | |
1155 | } | |
1156 | p = tagList; | |
1157 | tagList = tagList->parent; | |
1158 | FREE(p->buf); | |
1159 | destroyBindings(p->bindings, parser); | |
1160 | FREE(p); | |
1161 | } | |
1162 | /* free openInternalEntities and freeInternalEntities */ | |
1163 | entityList = openInternalEntities; | |
1164 | for (;;) { | |
1165 | OPEN_INTERNAL_ENTITY *openEntity; | |
1166 | if (entityList == NULL) { | |
1167 | if (freeInternalEntities == NULL) | |
1168 | break; | |
1169 | entityList = freeInternalEntities; | |
1170 | freeInternalEntities = NULL; | |
1171 | } | |
1172 | openEntity = entityList; | |
1173 | entityList = entityList->next; | |
1174 | FREE(openEntity); | |
1175 | } | |
1176 | ||
1177 | destroyBindings(freeBindingList, parser); | |
1178 | destroyBindings(inheritedBindings, parser); | |
1179 | poolDestroy(&tempPool); | |
1180 | poolDestroy(&temp2Pool); | |
1181 | #ifdef XML_DTD | |
1182 | /* external parameter entity parsers share the DTD structure | |
1183 | parser->m_dtd with the root parser, so we must not destroy it | |
1184 | */ | |
1185 | if (!isParamEntity && _dtd) | |
1186 | #else | |
1187 | if (_dtd) | |
1188 | #endif /* XML_DTD */ | |
1189 | dtdDestroy(_dtd, (XML_Bool)!parentParser, &parser->m_mem); | |
1190 | FREE((void *)atts); | |
1191 | #ifdef XML_ATTR_INFO | |
1192 | FREE((void *)attInfo); | |
1193 | #endif | |
1194 | FREE(groupConnector); | |
1195 | FREE(buffer); | |
1196 | FREE(dataBuf); | |
1197 | FREE(nsAtts); | |
1198 | FREE(unknownEncodingMem); | |
1199 | if (unknownEncodingRelease) | |
1200 | unknownEncodingRelease(unknownEncodingData); | |
1201 | FREE(parser); | |
1202 | } | |
1203 | ||
1204 | void XMLCALL | |
1205 | XML_UseParserAsHandlerArg(XML_Parser parser) | |
1206 | { | |
1207 | handlerArg = parser; | |
1208 | } | |
1209 | ||
1210 | enum XML_Error XMLCALL | |
1211 | XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD) | |
1212 | { | |
1213 | #ifdef XML_DTD | |
1214 | /* block after XML_Parse()/XML_ParseBuffer() has been called */ | |
1215 | if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) | |
1216 | return XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING; | |
1217 | useForeignDTD = useDTD; | |
1218 | return XML_ERROR_NONE; | |
1219 | #else | |
1220 | return XML_ERROR_FEATURE_REQUIRES_XML_DTD; | |
1221 | #endif | |
1222 | } | |
1223 | ||
1224 | void XMLCALL | |
1225 | XML_SetReturnNSTriplet(XML_Parser parser, int do_nst) | |
1226 | { | |
1227 | /* block after XML_Parse()/XML_ParseBuffer() has been called */ | |
1228 | if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) | |
1229 | return; | |
1230 | ns_triplets = do_nst ? XML_TRUE : XML_FALSE; | |
1231 | } | |
1232 | ||
1233 | void XMLCALL | |
1234 | XML_SetUserData(XML_Parser parser, void *p) | |
1235 | { | |
1236 | if (handlerArg == userData) | |
1237 | handlerArg = userData = p; | |
1238 | else | |
1239 | userData = p; | |
1240 | } | |
1241 | ||
1242 | enum XML_Status XMLCALL | |
1243 | XML_SetBase(XML_Parser parser, const XML_Char *p) | |
1244 | { | |
1245 | if (p) { | |
1246 | p = poolCopyString(&_dtd->pool, p); | |
1247 | if (!p) | |
1248 | return XML_STATUS_ERROR; | |
1249 | curBase = p; | |
1250 | } | |
1251 | else | |
1252 | curBase = NULL; | |
1253 | return XML_STATUS_OK; | |
1254 | } | |
1255 | ||
1256 | const XML_Char * XMLCALL | |
1257 | XML_GetBase(XML_Parser parser) | |
1258 | { | |
1259 | return curBase; | |
1260 | } | |
1261 | ||
1262 | int XMLCALL | |
1263 | XML_GetSpecifiedAttributeCount(XML_Parser parser) | |
1264 | { | |
1265 | return nSpecifiedAtts; | |
1266 | } | |
1267 | ||
1268 | int XMLCALL | |
1269 | XML_GetIdAttributeIndex(XML_Parser parser) | |
1270 | { | |
1271 | return idAttIndex; | |
1272 | } | |
1273 | ||
1274 | #ifdef XML_ATTR_INFO | |
1275 | const XML_AttrInfo * XMLCALL | |
1276 | XML_GetAttributeInfo(XML_Parser parser) | |
1277 | { | |
1278 | return attInfo; | |
1279 | } | |
1280 | #endif | |
1281 | ||
1282 | void XMLCALL | |
1283 | XML_SetElementHandler(XML_Parser parser, | |
1284 | XML_StartElementHandler start, | |
1285 | XML_EndElementHandler end) | |
1286 | { | |
1287 | startElementHandler = start; | |
1288 | endElementHandler = end; | |
1289 | } | |
1290 | ||
1291 | void XMLCALL | |
1292 | XML_SetStartElementHandler(XML_Parser parser, | |
1293 | XML_StartElementHandler start) { | |
1294 | startElementHandler = start; | |
1295 | } | |
1296 | ||
1297 | void XMLCALL | |
1298 | XML_SetEndElementHandler(XML_Parser parser, | |
1299 | XML_EndElementHandler end) { | |
1300 | endElementHandler = end; | |
1301 | } | |
1302 | ||
1303 | void XMLCALL | |
1304 | XML_SetCharacterDataHandler(XML_Parser parser, | |
1305 | XML_CharacterDataHandler handler) | |
1306 | { | |
1307 | characterDataHandler = handler; | |
1308 | } | |
1309 | ||
1310 | void XMLCALL | |
1311 | XML_SetProcessingInstructionHandler(XML_Parser parser, | |
1312 | XML_ProcessingInstructionHandler handler) | |
1313 | { | |
1314 | processingInstructionHandler = handler; | |
1315 | } | |
1316 | ||
1317 | void XMLCALL | |
1318 | XML_SetCommentHandler(XML_Parser parser, | |
1319 | XML_CommentHandler handler) | |
1320 | { | |
1321 | commentHandler = handler; | |
1322 | } | |
1323 | ||
1324 | void XMLCALL | |
1325 | XML_SetCdataSectionHandler(XML_Parser parser, | |
1326 | XML_StartCdataSectionHandler start, | |
1327 | XML_EndCdataSectionHandler end) | |
1328 | { | |
1329 | startCdataSectionHandler = start; | |
1330 | endCdataSectionHandler = end; | |
1331 | } | |
1332 | ||
1333 | void XMLCALL | |
1334 | XML_SetStartCdataSectionHandler(XML_Parser parser, | |
1335 | XML_StartCdataSectionHandler start) { | |
1336 | startCdataSectionHandler = start; | |
1337 | } | |
1338 | ||
1339 | void XMLCALL | |
1340 | XML_SetEndCdataSectionHandler(XML_Parser parser, | |
1341 | XML_EndCdataSectionHandler end) { | |
1342 | endCdataSectionHandler = end; | |
1343 | } | |
1344 | ||
1345 | void XMLCALL | |
1346 | XML_SetDefaultHandler(XML_Parser parser, | |
1347 | XML_DefaultHandler handler) | |
1348 | { | |
1349 | defaultHandler = handler; | |
1350 | defaultExpandInternalEntities = XML_FALSE; | |
1351 | } | |
1352 | ||
1353 | void XMLCALL | |
1354 | XML_SetDefaultHandlerExpand(XML_Parser parser, | |
1355 | XML_DefaultHandler handler) | |
1356 | { | |
1357 | defaultHandler = handler; | |
1358 | defaultExpandInternalEntities = XML_TRUE; | |
1359 | } | |
1360 | ||
1361 | void XMLCALL | |
1362 | XML_SetDoctypeDeclHandler(XML_Parser parser, | |
1363 | XML_StartDoctypeDeclHandler start, | |
1364 | XML_EndDoctypeDeclHandler end) | |
1365 | { | |
1366 | startDoctypeDeclHandler = start; | |
1367 | endDoctypeDeclHandler = end; | |
1368 | } | |
1369 | ||
1370 | void XMLCALL | |
1371 | XML_SetStartDoctypeDeclHandler(XML_Parser parser, | |
1372 | XML_StartDoctypeDeclHandler start) { | |
1373 | startDoctypeDeclHandler = start; | |
1374 | } | |
1375 | ||
1376 | void XMLCALL | |
1377 | XML_SetEndDoctypeDeclHandler(XML_Parser parser, | |
1378 | XML_EndDoctypeDeclHandler end) { | |
1379 | endDoctypeDeclHandler = end; | |
1380 | } | |
1381 | ||
1382 | void XMLCALL | |
1383 | XML_SetUnparsedEntityDeclHandler(XML_Parser parser, | |
1384 | XML_UnparsedEntityDeclHandler handler) | |
1385 | { | |
1386 | unparsedEntityDeclHandler = handler; | |
1387 | } | |
1388 | ||
1389 | void XMLCALL | |
1390 | XML_SetNotationDeclHandler(XML_Parser parser, | |
1391 | XML_NotationDeclHandler handler) | |
1392 | { | |
1393 | notationDeclHandler = handler; | |
1394 | } | |
1395 | ||
1396 | void XMLCALL | |
1397 | XML_SetNamespaceDeclHandler(XML_Parser parser, | |
1398 | XML_StartNamespaceDeclHandler start, | |
1399 | XML_EndNamespaceDeclHandler end) | |
1400 | { | |
1401 | startNamespaceDeclHandler = start; | |
1402 | endNamespaceDeclHandler = end; | |
1403 | } | |
1404 | ||
1405 | void XMLCALL | |
1406 | XML_SetStartNamespaceDeclHandler(XML_Parser parser, | |
1407 | XML_StartNamespaceDeclHandler start) { | |
1408 | startNamespaceDeclHandler = start; | |
1409 | } | |
1410 | ||
1411 | void XMLCALL | |
1412 | XML_SetEndNamespaceDeclHandler(XML_Parser parser, | |
1413 | XML_EndNamespaceDeclHandler end) { | |
1414 | endNamespaceDeclHandler = end; | |
1415 | } | |
1416 | ||
1417 | void XMLCALL | |
1418 | XML_SetNotStandaloneHandler(XML_Parser parser, | |
1419 | XML_NotStandaloneHandler handler) | |
1420 | { | |
1421 | notStandaloneHandler = handler; | |
1422 | } | |
1423 | ||
1424 | void XMLCALL | |
1425 | XML_SetExternalEntityRefHandler(XML_Parser parser, | |
1426 | XML_ExternalEntityRefHandler handler) | |
1427 | { | |
1428 | externalEntityRefHandler = handler; | |
1429 | } | |
1430 | ||
1431 | void XMLCALL | |
1432 | XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg) | |
1433 | { | |
1434 | if (arg) | |
1435 | externalEntityRefHandlerArg = (XML_Parser)arg; | |
1436 | else | |
1437 | externalEntityRefHandlerArg = parser; | |
1438 | } | |
1439 | ||
1440 | void XMLCALL | |
1441 | XML_SetSkippedEntityHandler(XML_Parser parser, | |
1442 | XML_SkippedEntityHandler handler) | |
1443 | { | |
1444 | skippedEntityHandler = handler; | |
1445 | } | |
1446 | ||
1447 | void XMLCALL | |
1448 | XML_SetUnknownEncodingHandler(XML_Parser parser, | |
1449 | XML_UnknownEncodingHandler handler, | |
1450 | void *data) | |
1451 | { | |
1452 | unknownEncodingHandler = handler; | |
1453 | unknownEncodingHandlerData = data; | |
1454 | } | |
1455 | ||
1456 | void XMLCALL | |
1457 | XML_SetElementDeclHandler(XML_Parser parser, | |
1458 | XML_ElementDeclHandler eldecl) | |
1459 | { | |
1460 | elementDeclHandler = eldecl; | |
1461 | } | |
1462 | ||
1463 | void XMLCALL | |
1464 | XML_SetAttlistDeclHandler(XML_Parser parser, | |
1465 | XML_AttlistDeclHandler attdecl) | |
1466 | { | |
1467 | attlistDeclHandler = attdecl; | |
1468 | } | |
1469 | ||
1470 | void XMLCALL | |
1471 | XML_SetEntityDeclHandler(XML_Parser parser, | |
1472 | XML_EntityDeclHandler handler) | |
1473 | { | |
1474 | entityDeclHandler = handler; | |
1475 | } | |
1476 | ||
1477 | void XMLCALL | |
1478 | XML_SetXmlDeclHandler(XML_Parser parser, | |
1479 | XML_XmlDeclHandler handler) { | |
1480 | xmlDeclHandler = handler; | |
1481 | } | |
1482 | ||
1483 | int XMLCALL | |
1484 | XML_SetParamEntityParsing(XML_Parser parser, | |
1485 | enum XML_ParamEntityParsing peParsing) | |
1486 | { | |
1487 | /* block after XML_Parse()/XML_ParseBuffer() has been called */ | |
1488 | if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) | |
1489 | return 0; | |
1490 | #ifdef XML_DTD | |
1491 | paramEntityParsing = peParsing; | |
1492 | return 1; | |
1493 | #else | |
1494 | return peParsing == XML_PARAM_ENTITY_PARSING_NEVER; | |
1495 | #endif | |
1496 | } | |
1497 | ||
1498 | int XMLCALL | |
1499 | XML_SetHashSalt(XML_Parser parser, | |
1500 | unsigned long hash_salt) | |
1501 | { | |
1502 | /* block after XML_Parse()/XML_ParseBuffer() has been called */ | |
1503 | if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED) | |
1504 | return 0; | |
1505 | hash_secret_salt = hash_salt; | |
1506 | return 1; | |
1507 | } | |
1508 | ||
1509 | enum XML_Status XMLCALL | |
1510 | XML_Parse(XML_Parser parser, const char *s, int len, int isFinal) | |
1511 | { | |
1512 | switch (ps_parsing) { | |
1513 | case XML_SUSPENDED: | |
1514 | errorCode = XML_ERROR_SUSPENDED; | |
1515 | return XML_STATUS_ERROR; | |
1516 | case XML_FINISHED: | |
1517 | errorCode = XML_ERROR_FINISHED; | |
1518 | return XML_STATUS_ERROR; | |
1519 | case XML_INITIALIZED: | |
1520 | if (parentParser == NULL && !startParsing(parser)) { | |
1521 | errorCode = XML_ERROR_NO_MEMORY; | |
1522 | return XML_STATUS_ERROR; | |
1523 | } | |
1524 | default: | |
1525 | ps_parsing = XML_PARSING; | |
1526 | } | |
1527 | ||
1528 | if (len == 0) { | |
1529 | ps_finalBuffer = (XML_Bool)isFinal; | |
1530 | if (!isFinal) | |
1531 | return XML_STATUS_OK; | |
1532 | positionPtr = bufferPtr; | |
1533 | parseEndPtr = bufferEnd; | |
1534 | ||
1535 | /* If data are left over from last buffer, and we now know that these | |
1536 | data are the final chunk of input, then we have to check them again | |
1537 | to detect errors based on that fact. | |
1538 | */ | |
1539 | errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr); | |
1540 | ||
1541 | if (errorCode == XML_ERROR_NONE) { | |
1542 | switch (ps_parsing) { | |
1543 | case XML_SUSPENDED: | |
1544 | XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position); | |
1545 | positionPtr = bufferPtr; | |
1546 | return XML_STATUS_SUSPENDED; | |
1547 | case XML_INITIALIZED: | |
1548 | case XML_PARSING: | |
1549 | ps_parsing = XML_FINISHED; | |
1550 | /* fall through */ | |
1551 | default: | |
1552 | return XML_STATUS_OK; | |
1553 | } | |
1554 | } | |
1555 | eventEndPtr = eventPtr; | |
1556 | processor = errorProcessor; | |
1557 | return XML_STATUS_ERROR; | |
1558 | } | |
1559 | #ifndef XML_CONTEXT_BYTES | |
1560 | else if (bufferPtr == bufferEnd) { | |
1561 | const char *end; | |
1562 | int nLeftOver; | |
1563 | enum XML_Error result; | |
1564 | parseEndByteIndex += len; | |
1565 | positionPtr = s; | |
1566 | ps_finalBuffer = (XML_Bool)isFinal; | |
1567 | ||
1568 | errorCode = processor(parser, s, parseEndPtr = s + len, &end); | |
1569 | ||
1570 | if (errorCode != XML_ERROR_NONE) { | |
1571 | eventEndPtr = eventPtr; | |
1572 | processor = errorProcessor; | |
1573 | return XML_STATUS_ERROR; | |
1574 | } | |
1575 | else { | |
1576 | switch (ps_parsing) { | |
1577 | case XML_SUSPENDED: | |
1578 | result = XML_STATUS_SUSPENDED; | |
1579 | break; | |
1580 | case XML_INITIALIZED: | |
1581 | case XML_PARSING: | |
1582 | if (isFinal) { | |
1583 | ps_parsing = XML_FINISHED; | |
1584 | return XML_STATUS_OK; | |
1585 | } | |
1586 | /* fall through */ | |
1587 | default: | |
1588 | result = XML_STATUS_OK; | |
1589 | } | |
1590 | } | |
1591 | ||
1592 | XmlUpdatePosition(encoding, positionPtr, end, &position); | |
1593 | nLeftOver = s + len - end; | |
1594 | if (nLeftOver) { | |
1595 | if (buffer == NULL || nLeftOver > bufferLim - buffer) { | |
1596 | /* FIXME avoid integer overflow */ | |
1597 | char *temp; | |
1598 | temp = (buffer == NULL | |
1599 | ? (char *)MALLOC(len * 2) | |
1600 | : (char *)REALLOC(buffer, len * 2)); | |
1601 | if (temp == NULL) { | |
1602 | errorCode = XML_ERROR_NO_MEMORY; | |
1603 | eventPtr = eventEndPtr = NULL; | |
1604 | processor = errorProcessor; | |
1605 | return XML_STATUS_ERROR; | |
1606 | } | |
1607 | buffer = temp; | |
1608 | bufferLim = buffer + len * 2; | |
1609 | } | |
1610 | memcpy(buffer, end, nLeftOver); | |
1611 | } | |
1612 | bufferPtr = buffer; | |
1613 | bufferEnd = buffer + nLeftOver; | |
1614 | positionPtr = bufferPtr; | |
1615 | parseEndPtr = bufferEnd; | |
1616 | eventPtr = bufferPtr; | |
1617 | eventEndPtr = bufferPtr; | |
1618 | return result; | |
1619 | } | |
1620 | #endif /* not defined XML_CONTEXT_BYTES */ | |
1621 | else { | |
1622 | void *buff = XML_GetBuffer(parser, len); | |
1623 | if (buff == NULL) | |
1624 | return XML_STATUS_ERROR; | |
1625 | else { | |
1626 | memcpy(buff, s, len); | |
1627 | return XML_ParseBuffer(parser, len, isFinal); | |
1628 | } | |
1629 | } | |
1630 | } | |
1631 | ||
1632 | enum XML_Status XMLCALL | |
1633 | XML_ParseBuffer(XML_Parser parser, int len, int isFinal) | |
1634 | { | |
1635 | const char *start; | |
1636 | enum XML_Status result = XML_STATUS_OK; | |
1637 | ||
1638 | switch (ps_parsing) { | |
1639 | case XML_SUSPENDED: | |
1640 | errorCode = XML_ERROR_SUSPENDED; | |
1641 | return XML_STATUS_ERROR; | |
1642 | case XML_FINISHED: | |
1643 | errorCode = XML_ERROR_FINISHED; | |
1644 | return XML_STATUS_ERROR; | |
1645 | case XML_INITIALIZED: | |
1646 | if (parentParser == NULL && !startParsing(parser)) { | |
1647 | errorCode = XML_ERROR_NO_MEMORY; | |
1648 | return XML_STATUS_ERROR; | |
1649 | } | |
1650 | default: | |
1651 | ps_parsing = XML_PARSING; | |
1652 | } | |
1653 | ||
1654 | start = bufferPtr; | |
1655 | positionPtr = start; | |
1656 | bufferEnd += len; | |
1657 | parseEndPtr = bufferEnd; | |
1658 | parseEndByteIndex += len; | |
1659 | ps_finalBuffer = (XML_Bool)isFinal; | |
1660 | ||
1661 | errorCode = processor(parser, start, parseEndPtr, &bufferPtr); | |
1662 | ||
1663 | if (errorCode != XML_ERROR_NONE) { | |
1664 | eventEndPtr = eventPtr; | |
1665 | processor = errorProcessor; | |
1666 | return XML_STATUS_ERROR; | |
1667 | } | |
1668 | else { | |
1669 | switch (ps_parsing) { | |
1670 | case XML_SUSPENDED: | |
1671 | result = XML_STATUS_SUSPENDED; | |
1672 | break; | |
1673 | case XML_INITIALIZED: | |
1674 | case XML_PARSING: | |
1675 | if (isFinal) { | |
1676 | ps_parsing = XML_FINISHED; | |
1677 | return result; | |
1678 | } | |
1679 | default: ; /* should not happen */ | |
1680 | } | |
1681 | } | |
1682 | ||
1683 | XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position); | |
1684 | positionPtr = bufferPtr; | |
1685 | return result; | |
1686 | } | |
1687 | ||
1688 | void * XMLCALL | |
1689 | XML_GetBuffer(XML_Parser parser, int len) | |
1690 | { | |
1691 | switch (ps_parsing) { | |
1692 | case XML_SUSPENDED: | |
1693 | errorCode = XML_ERROR_SUSPENDED; | |
1694 | return NULL; | |
1695 | case XML_FINISHED: | |
1696 | errorCode = XML_ERROR_FINISHED; | |
1697 | return NULL; | |
1698 | default: ; | |
1699 | } | |
1700 | ||
1701 | if (len > bufferLim - bufferEnd) { | |
1702 | /* FIXME avoid integer overflow */ | |
1703 | int neededSize = len + (int)(bufferEnd - bufferPtr); | |
1704 | #ifdef XML_CONTEXT_BYTES | |
1705 | int keep = (int)(bufferPtr - buffer); | |
1706 | ||
1707 | if (keep > XML_CONTEXT_BYTES) | |
1708 | keep = XML_CONTEXT_BYTES; | |
1709 | neededSize += keep; | |
1710 | #endif /* defined XML_CONTEXT_BYTES */ | |
1711 | if (neededSize <= bufferLim - buffer) { | |
1712 | #ifdef XML_CONTEXT_BYTES | |
1713 | if (keep < bufferPtr - buffer) { | |
1714 | int offset = (int)(bufferPtr - buffer) - keep; | |
1715 | memmove(buffer, &buffer[offset], bufferEnd - bufferPtr + keep); | |
1716 | bufferEnd -= offset; | |
1717 | bufferPtr -= offset; | |
1718 | } | |
1719 | #else | |
1720 | memmove(buffer, bufferPtr, bufferEnd - bufferPtr); | |
1721 | bufferEnd = buffer + (bufferEnd - bufferPtr); | |
1722 | bufferPtr = buffer; | |
1723 | #endif /* not defined XML_CONTEXT_BYTES */ | |
1724 | } | |
1725 | else { | |
1726 | char *newBuf; | |
1727 | int bufferSize = (int)(bufferLim - bufferPtr); | |
1728 | if (bufferSize == 0) | |
1729 | bufferSize = INIT_BUFFER_SIZE; | |
1730 | do { | |
1731 | bufferSize *= 2; | |
1732 | } while (bufferSize < neededSize); | |
1733 | newBuf = (char *)MALLOC(bufferSize); | |
1734 | if (newBuf == 0) { | |
1735 | errorCode = XML_ERROR_NO_MEMORY; | |
1736 | return NULL; | |
1737 | } | |
1738 | bufferLim = newBuf + bufferSize; | |
1739 | #ifdef XML_CONTEXT_BYTES | |
1740 | if (bufferPtr) { | |
1741 | int keep = (int)(bufferPtr - buffer); | |
1742 | if (keep > XML_CONTEXT_BYTES) | |
1743 | keep = XML_CONTEXT_BYTES; | |
1744 | memcpy(newBuf, &bufferPtr[-keep], bufferEnd - bufferPtr + keep); | |
1745 | FREE(buffer); | |
1746 | buffer = newBuf; | |
1747 | bufferEnd = buffer + (bufferEnd - bufferPtr) + keep; | |
1748 | bufferPtr = buffer + keep; | |
1749 | } | |
1750 | else { | |
1751 | bufferEnd = newBuf + (bufferEnd - bufferPtr); | |
1752 | bufferPtr = buffer = newBuf; | |
1753 | } | |
1754 | #else | |
1755 | if (bufferPtr) { | |
1756 | memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr); | |
1757 | FREE(buffer); | |
1758 | } | |
1759 | bufferEnd = newBuf + (bufferEnd - bufferPtr); | |
1760 | bufferPtr = buffer = newBuf; | |
1761 | #endif /* not defined XML_CONTEXT_BYTES */ | |
1762 | } | |
1763 | eventPtr = eventEndPtr = NULL; | |
1764 | positionPtr = NULL; | |
1765 | } | |
1766 | return bufferEnd; | |
1767 | } | |
1768 | ||
1769 | enum XML_Status XMLCALL | |
1770 | XML_StopParser(XML_Parser parser, XML_Bool resumable) | |
1771 | { | |
1772 | switch (ps_parsing) { | |
1773 | case XML_SUSPENDED: | |
1774 | if (resumable) { | |
1775 | errorCode = XML_ERROR_SUSPENDED; | |
1776 | return XML_STATUS_ERROR; | |
1777 | } | |
1778 | ps_parsing = XML_FINISHED; | |
1779 | break; | |
1780 | case XML_FINISHED: | |
1781 | errorCode = XML_ERROR_FINISHED; | |
1782 | return XML_STATUS_ERROR; | |
1783 | default: | |
1784 | if (resumable) { | |
1785 | #ifdef XML_DTD | |
1786 | if (isParamEntity) { | |
1787 | errorCode = XML_ERROR_SUSPEND_PE; | |
1788 | return XML_STATUS_ERROR; | |
1789 | } | |
1790 | #endif | |
1791 | ps_parsing = XML_SUSPENDED; | |
1792 | } | |
1793 | else | |
1794 | ps_parsing = XML_FINISHED; | |
1795 | } | |
1796 | return XML_STATUS_OK; | |
1797 | } | |
1798 | ||
1799 | enum XML_Status XMLCALL | |
1800 | XML_ResumeParser(XML_Parser parser) | |
1801 | { | |
1802 | enum XML_Status result = XML_STATUS_OK; | |
1803 | ||
1804 | if (ps_parsing != XML_SUSPENDED) { | |
1805 | errorCode = XML_ERROR_NOT_SUSPENDED; | |
1806 | return XML_STATUS_ERROR; | |
1807 | } | |
1808 | ps_parsing = XML_PARSING; | |
1809 | ||
1810 | errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr); | |
1811 | ||
1812 | if (errorCode != XML_ERROR_NONE) { | |
1813 | eventEndPtr = eventPtr; | |
1814 | processor = errorProcessor; | |
1815 | return XML_STATUS_ERROR; | |
1816 | } | |
1817 | else { | |
1818 | switch (ps_parsing) { | |
1819 | case XML_SUSPENDED: | |
1820 | result = XML_STATUS_SUSPENDED; | |
1821 | break; | |
1822 | case XML_INITIALIZED: | |
1823 | case XML_PARSING: | |
1824 | if (ps_finalBuffer) { | |
1825 | ps_parsing = XML_FINISHED; | |
1826 | return result; | |
1827 | } | |
1828 | default: ; | |
1829 | } | |
1830 | } | |
1831 | ||
1832 | XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position); | |
1833 | positionPtr = bufferPtr; | |
1834 | return result; | |
1835 | } | |
1836 | ||
1837 | void XMLCALL | |
1838 | XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status) | |
1839 | { | |
1840 | assert(status != NULL); | |
1841 | *status = parser->m_parsingStatus; | |
1842 | } | |
1843 | ||
1844 | enum XML_Error XMLCALL | |
1845 | XML_GetErrorCode(XML_Parser parser) | |
1846 | { | |
1847 | return errorCode; | |
1848 | } | |
1849 | ||
1850 | XML_Index XMLCALL | |
1851 | XML_GetCurrentByteIndex(XML_Parser parser) | |
1852 | { | |
1853 | if (eventPtr) | |
1854 | return parseEndByteIndex - (parseEndPtr - eventPtr); | |
1855 | return -1; | |
1856 | } | |
1857 | ||
1858 | int XMLCALL | |
1859 | XML_GetCurrentByteCount(XML_Parser parser) | |
1860 | { | |
1861 | if (eventEndPtr && eventPtr) | |
1862 | return (int)(eventEndPtr - eventPtr); | |
1863 | return 0; | |
1864 | } | |
1865 | ||
1866 | const char * XMLCALL | |
1867 | XML_GetInputContext(XML_Parser parser, int *offset, int *size) | |
1868 | { | |
1869 | #ifdef XML_CONTEXT_BYTES | |
1870 | if (eventPtr && buffer) { | |
1871 | *offset = (int)(eventPtr - buffer); | |
1872 | *size = (int)(bufferEnd - buffer); | |
1873 | return buffer; | |
1874 | } | |
1875 | #endif /* defined XML_CONTEXT_BYTES */ | |
1876 | return (char *) 0; | |
1877 | } | |
1878 | ||
1879 | XML_Size XMLCALL | |
1880 | XML_GetCurrentLineNumber(XML_Parser parser) | |
1881 | { | |
1882 | if (eventPtr && eventPtr >= positionPtr) { | |
1883 | XmlUpdatePosition(encoding, positionPtr, eventPtr, &position); | |
1884 | positionPtr = eventPtr; | |
1885 | } | |
1886 | return position.lineNumber + 1; | |
1887 | } | |
1888 | ||
1889 | XML_Size XMLCALL | |
1890 | XML_GetCurrentColumnNumber(XML_Parser parser) | |
1891 | { | |
1892 | if (eventPtr && eventPtr >= positionPtr) { | |
1893 | XmlUpdatePosition(encoding, positionPtr, eventPtr, &position); | |
1894 | positionPtr = eventPtr; | |
1895 | } | |
1896 | return position.columnNumber; | |
1897 | } | |
1898 | ||
1899 | void XMLCALL | |
1900 | XML_FreeContentModel(XML_Parser parser, XML_Content *model) | |
1901 | { | |
1902 | FREE(model); | |
1903 | } | |
1904 | ||
1905 | void * XMLCALL | |
1906 | XML_MemMalloc(XML_Parser parser, size_t size) | |
1907 | { | |
1908 | return MALLOC(size); | |
1909 | } | |
1910 | ||
1911 | void * XMLCALL | |
1912 | XML_MemRealloc(XML_Parser parser, void *ptr, size_t size) | |
1913 | { | |
1914 | return REALLOC(ptr, size); | |
1915 | } | |
1916 | ||
1917 | void XMLCALL | |
1918 | XML_MemFree(XML_Parser parser, void *ptr) | |
1919 | { | |
1920 | FREE(ptr); | |
1921 | } | |
1922 | ||
1923 | void XMLCALL | |
1924 | XML_DefaultCurrent(XML_Parser parser) | |
1925 | { | |
1926 | if (defaultHandler) { | |
1927 | if (openInternalEntities) | |
1928 | reportDefault(parser, | |
1929 | internalEncoding, | |
1930 | openInternalEntities->internalEventPtr, | |
1931 | openInternalEntities->internalEventEndPtr); | |
1932 | else | |
1933 | reportDefault(parser, encoding, eventPtr, eventEndPtr); | |
1934 | } | |
1935 | } | |
1936 | ||
1937 | const XML_LChar * XMLCALL | |
1938 | XML_ErrorString(enum XML_Error code) | |
1939 | { | |
1940 | static const XML_LChar* const message[] = { | |
1941 | 0, | |
1942 | XML_L("out of memory"), | |
1943 | XML_L("syntax error"), | |
1944 | XML_L("no element found"), | |
1945 | XML_L("not well-formed (invalid token)"), | |
1946 | XML_L("unclosed token"), | |
1947 | XML_L("partial character"), | |
1948 | XML_L("mismatched tag"), | |
1949 | XML_L("duplicate attribute"), | |
1950 | XML_L("junk after document element"), | |
1951 | XML_L("illegal parameter entity reference"), | |
1952 | XML_L("undefined entity"), | |
1953 | XML_L("recursive entity reference"), | |
1954 | XML_L("asynchronous entity"), | |
1955 | XML_L("reference to invalid character number"), | |
1956 | XML_L("reference to binary entity"), | |
1957 | XML_L("reference to external entity in attribute"), | |
1958 | XML_L("XML or text declaration not at start of entity"), | |
1959 | XML_L("unknown encoding"), | |
1960 | XML_L("encoding specified in XML declaration is incorrect"), | |
1961 | XML_L("unclosed CDATA section"), | |
1962 | XML_L("error in processing external entity reference"), | |
1963 | XML_L("document is not standalone"), | |
1964 | XML_L("unexpected parser state - please send a bug report"), | |
1965 | XML_L("entity declared in parameter entity"), | |
1966 | XML_L("requested feature requires XML_DTD support in Expat"), | |
1967 | XML_L("cannot change setting once parsing has begun"), | |
1968 | XML_L("unbound prefix"), | |
1969 | XML_L("must not undeclare prefix"), | |
1970 | XML_L("incomplete markup in parameter entity"), | |
1971 | XML_L("XML declaration not well-formed"), | |
1972 | XML_L("text declaration not well-formed"), | |
1973 | XML_L("illegal character(s) in public id"), | |
1974 | XML_L("parser suspended"), | |
1975 | XML_L("parser not suspended"), | |
1976 | XML_L("parsing aborted"), | |
1977 | XML_L("parsing finished"), | |
1978 | XML_L("cannot suspend in external parameter entity"), | |
1979 | XML_L("reserved prefix (xml) must not be undeclared or bound to another namespace name"), | |
1980 | XML_L("reserved prefix (xmlns) must not be declared or undeclared"), | |
1981 | XML_L("prefix must not be bound to one of the reserved namespace names") | |
1982 | }; | |
1983 | if (code > 0 && code < sizeof(message)/sizeof(message[0])) | |
1984 | return message[code]; | |
1985 | return NULL; | |
1986 | } | |
1987 | ||
1988 | const XML_LChar * XMLCALL | |
1989 | XML_ExpatVersion(void) { | |
1990 | ||
1991 | /* V1 is used to string-ize the version number. However, it would | |
1992 | string-ize the actual version macro *names* unless we get them | |
1993 | substituted before being passed to V1. CPP is defined to expand | |
1994 | a macro, then rescan for more expansions. Thus, we use V2 to expand | |
1995 | the version macros, then CPP will expand the resulting V1() macro | |
1996 | with the correct numerals. */ | |
1997 | /* ### I'm assuming cpp is portable in this respect... */ | |
1998 | ||
1999 | #define V1(a,b,c) XML_L(#a)XML_L(".")XML_L(#b)XML_L(".")XML_L(#c) | |
2000 | #define V2(a,b,c) XML_L("expat_")V1(a,b,c) | |
2001 | ||
2002 | return V2(XML_MAJOR_VERSION, XML_MINOR_VERSION, XML_MICRO_VERSION); | |
2003 | ||
2004 | #undef V1 | |
2005 | #undef V2 | |
2006 | } | |
2007 | ||
2008 | XML_Expat_Version XMLCALL | |
2009 | XML_ExpatVersionInfo(void) | |
2010 | { | |
2011 | XML_Expat_Version version; | |
2012 | ||
2013 | version.major = XML_MAJOR_VERSION; | |
2014 | version.minor = XML_MINOR_VERSION; | |
2015 | version.micro = XML_MICRO_VERSION; | |
2016 | ||
2017 | return version; | |
2018 | } | |
2019 | ||
2020 | const XML_Feature * XMLCALL | |
2021 | XML_GetFeatureList(void) | |
2022 | { | |
2023 | static const XML_Feature features[] = { | |
2024 | {XML_FEATURE_SIZEOF_XML_CHAR, XML_L("sizeof(XML_Char)"), | |
2025 | sizeof(XML_Char)}, | |
2026 | {XML_FEATURE_SIZEOF_XML_LCHAR, XML_L("sizeof(XML_LChar)"), | |
2027 | sizeof(XML_LChar)}, | |
2028 | #ifdef XML_UNICODE | |
2029 | {XML_FEATURE_UNICODE, XML_L("XML_UNICODE"), 0}, | |
2030 | #endif | |
2031 | #ifdef XML_UNICODE_WCHAR_T | |
2032 | {XML_FEATURE_UNICODE_WCHAR_T, XML_L("XML_UNICODE_WCHAR_T"), 0}, | |
2033 | #endif | |
2034 | #ifdef XML_DTD | |
2035 | {XML_FEATURE_DTD, XML_L("XML_DTD"), 0}, | |
2036 | #endif | |
2037 | #ifdef XML_CONTEXT_BYTES | |
2038 | {XML_FEATURE_CONTEXT_BYTES, XML_L("XML_CONTEXT_BYTES"), | |
2039 | XML_CONTEXT_BYTES}, | |
2040 | #endif | |
2041 | #ifdef XML_MIN_SIZE | |
2042 | {XML_FEATURE_MIN_SIZE, XML_L("XML_MIN_SIZE"), 0}, | |
2043 | #endif | |
2044 | #ifdef XML_NS | |
2045 | {XML_FEATURE_NS, XML_L("XML_NS"), 0}, | |
2046 | #endif | |
2047 | #ifdef XML_LARGE_SIZE | |
2048 | {XML_FEATURE_LARGE_SIZE, XML_L("XML_LARGE_SIZE"), 0}, | |
2049 | #endif | |
2050 | #ifdef XML_ATTR_INFO | |
2051 | {XML_FEATURE_ATTR_INFO, XML_L("XML_ATTR_INFO"), 0}, | |
2052 | #endif | |
2053 | {XML_FEATURE_END, NULL, 0} | |
2054 | }; | |
2055 | ||
2056 | return features; | |
2057 | } | |
2058 | ||
2059 | /* Initially tag->rawName always points into the parse buffer; | |
2060 | for those TAG instances opened while the current parse buffer was | |
2061 | processed, and not yet closed, we need to store tag->rawName in a more | |
2062 | permanent location, since the parse buffer is about to be discarded. | |
2063 | */ | |
2064 | static XML_Bool | |
2065 | storeRawNames(XML_Parser parser) | |
2066 | { | |
2067 | TAG *tag = tagStack; | |
2068 | while (tag) { | |
2069 | int bufSize; | |
2070 | int nameLen = sizeof(XML_Char) * (tag->name.strLen + 1); | |
2071 | char *rawNameBuf = tag->buf + nameLen; | |
2072 | /* Stop if already stored. Since tagStack is a stack, we can stop | |
2073 | at the first entry that has already been copied; everything | |
2074 | below it in the stack is already been accounted for in a | |
2075 | previous call to this function. | |
2076 | */ | |
2077 | if (tag->rawName == rawNameBuf) | |
2078 | break; | |
2079 | /* For re-use purposes we need to ensure that the | |
2080 | size of tag->buf is a multiple of sizeof(XML_Char). | |
2081 | */ | |
2082 | bufSize = nameLen + ROUND_UP(tag->rawNameLength, sizeof(XML_Char)); | |
2083 | if (bufSize > tag->bufEnd - tag->buf) { | |
2084 | char *temp = (char *)REALLOC(tag->buf, bufSize); | |
2085 | if (temp == NULL) | |
2086 | return XML_FALSE; | |
2087 | /* if tag->name.str points to tag->buf (only when namespace | |
2088 | processing is off) then we have to update it | |
2089 | */ | |
2090 | if (tag->name.str == (XML_Char *)tag->buf) | |
2091 | tag->name.str = (XML_Char *)temp; | |
2092 | /* if tag->name.localPart is set (when namespace processing is on) | |
2093 | then update it as well, since it will always point into tag->buf | |
2094 | */ | |
2095 | if (tag->name.localPart) | |
2096 | tag->name.localPart = (XML_Char *)temp + (tag->name.localPart - | |
2097 | (XML_Char *)tag->buf); | |
2098 | tag->buf = temp; | |
2099 | tag->bufEnd = temp + bufSize; | |
2100 | rawNameBuf = temp + nameLen; | |
2101 | } | |
2102 | memcpy(rawNameBuf, tag->rawName, tag->rawNameLength); | |
2103 | tag->rawName = rawNameBuf; | |
2104 | tag = tag->parent; | |
2105 | } | |
2106 | return XML_TRUE; | |
2107 | } | |
2108 | ||
2109 | static enum XML_Error PTRCALL | |
2110 | contentProcessor(XML_Parser parser, | |
2111 | const char *start, | |
2112 | const char *end, | |
2113 | const char **endPtr) | |
2114 | { | |
2115 | enum XML_Error result = doContent(parser, 0, encoding, start, end, | |
2116 | endPtr, (XML_Bool)!ps_finalBuffer); | |
2117 | if (result == XML_ERROR_NONE) { | |
2118 | if (!storeRawNames(parser)) | |
2119 | return XML_ERROR_NO_MEMORY; | |
2120 | } | |
2121 | return result; | |
2122 | } | |
2123 | ||
2124 | static enum XML_Error PTRCALL | |
2125 | externalEntityInitProcessor(XML_Parser parser, | |
2126 | const char *start, | |
2127 | const char *end, | |
2128 | const char **endPtr) | |
2129 | { | |
2130 | enum XML_Error result = initializeEncoding(parser); | |
2131 | if (result != XML_ERROR_NONE) | |
2132 | return result; | |
2133 | processor = externalEntityInitProcessor2; | |
2134 | return externalEntityInitProcessor2(parser, start, end, endPtr); | |
2135 | } | |
2136 | ||
2137 | static enum XML_Error PTRCALL | |
2138 | externalEntityInitProcessor2(XML_Parser parser, | |
2139 | const char *start, | |
2140 | const char *end, | |
2141 | const char **endPtr) | |
2142 | { | |
2143 | const char *next = start; /* XmlContentTok doesn't always set the last arg */ | |
2144 | int tok = XmlContentTok(encoding, start, end, &next); | |
2145 | switch (tok) { | |
2146 | case XML_TOK_BOM: | |
2147 | /* If we are at the end of the buffer, this would cause the next stage, | |
2148 | i.e. externalEntityInitProcessor3, to pass control directly to | |
2149 | doContent (by detecting XML_TOK_NONE) without processing any xml text | |
2150 | declaration - causing the error XML_ERROR_MISPLACED_XML_PI in doContent. | |
2151 | */ | |
2152 | if (next == end && !ps_finalBuffer) { | |
2153 | *endPtr = next; | |
2154 | return XML_ERROR_NONE; | |
2155 | } | |
2156 | start = next; | |
2157 | break; | |
2158 | case XML_TOK_PARTIAL: | |
2159 | if (!ps_finalBuffer) { | |
2160 | *endPtr = start; | |
2161 | return XML_ERROR_NONE; | |
2162 | } | |
2163 | eventPtr = start; | |
2164 | return XML_ERROR_UNCLOSED_TOKEN; | |
2165 | case XML_TOK_PARTIAL_CHAR: | |
2166 | if (!ps_finalBuffer) { | |
2167 | *endPtr = start; | |
2168 | return XML_ERROR_NONE; | |
2169 | } | |
2170 | eventPtr = start; | |
2171 | return XML_ERROR_PARTIAL_CHAR; | |
2172 | } | |
2173 | processor = externalEntityInitProcessor3; | |
2174 | return externalEntityInitProcessor3(parser, start, end, endPtr); | |
2175 | } | |
2176 | ||
2177 | static enum XML_Error PTRCALL | |
2178 | externalEntityInitProcessor3(XML_Parser parser, | |
2179 | const char *start, | |
2180 | const char *end, | |
2181 | const char **endPtr) | |
2182 | { | |
2183 | int tok; | |
2184 | const char *next = start; /* XmlContentTok doesn't always set the last arg */ | |
2185 | eventPtr = start; | |
2186 | tok = XmlContentTok(encoding, start, end, &next); | |
2187 | eventEndPtr = next; | |
2188 | ||
2189 | switch (tok) { | |
2190 | case XML_TOK_XML_DECL: | |
2191 | { | |
2192 | enum XML_Error result; | |
2193 | result = processXmlDecl(parser, 1, start, next); | |
2194 | if (result != XML_ERROR_NONE) | |
2195 | return result; | |
2196 | switch (ps_parsing) { | |
2197 | case XML_SUSPENDED: | |
2198 | *endPtr = next; | |
2199 | return XML_ERROR_NONE; | |
2200 | case XML_FINISHED: | |
2201 | return XML_ERROR_ABORTED; | |
2202 | default: | |
2203 | start = next; | |
2204 | } | |
2205 | } | |
2206 | break; | |
2207 | case XML_TOK_PARTIAL: | |
2208 | if (!ps_finalBuffer) { | |
2209 | *endPtr = start; | |
2210 | return XML_ERROR_NONE; | |
2211 | } | |
2212 | return XML_ERROR_UNCLOSED_TOKEN; | |
2213 | case XML_TOK_PARTIAL_CHAR: | |
2214 | if (!ps_finalBuffer) { | |
2215 | *endPtr = start; | |
2216 | return XML_ERROR_NONE; | |
2217 | } | |
2218 | return XML_ERROR_PARTIAL_CHAR; | |
2219 | } | |
2220 | processor = externalEntityContentProcessor; | |
2221 | tagLevel = 1; | |
2222 | return externalEntityContentProcessor(parser, start, end, endPtr); | |
2223 | } | |
2224 | ||
2225 | static enum XML_Error PTRCALL | |
2226 | externalEntityContentProcessor(XML_Parser parser, | |
2227 | const char *start, | |
2228 | const char *end, | |
2229 | const char **endPtr) | |
2230 | { | |
2231 | enum XML_Error result = doContent(parser, 1, encoding, start, end, | |
2232 | endPtr, (XML_Bool)!ps_finalBuffer); | |
2233 | if (result == XML_ERROR_NONE) { | |
2234 | if (!storeRawNames(parser)) | |
2235 | return XML_ERROR_NO_MEMORY; | |
2236 | } | |
2237 | return result; | |
2238 | } | |
2239 | ||
2240 | static enum XML_Error | |
2241 | doContent(XML_Parser parser, | |
2242 | int startTagLevel, | |
2243 | const ENCODING *enc, | |
2244 | const char *s, | |
2245 | const char *end, | |
2246 | const char **nextPtr, | |
2247 | XML_Bool haveMore) | |
2248 | { | |
2249 | /* save one level of indirection */ | |
2250 | DTD * const dtd = _dtd; | |
2251 | ||
2252 | const char **eventPP; | |
2253 | const char **eventEndPP; | |
2254 | if (enc == encoding) { | |
2255 | eventPP = &eventPtr; | |
2256 | eventEndPP = &eventEndPtr; | |
2257 | } | |
2258 | else { | |
2259 | eventPP = &(openInternalEntities->internalEventPtr); | |
2260 | eventEndPP = &(openInternalEntities->internalEventEndPtr); | |
2261 | } | |
2262 | *eventPP = s; | |
2263 | ||
2264 | for (;;) { | |
2265 | const char *next = s; /* XmlContentTok doesn't always set the last arg */ | |
2266 | int tok = XmlContentTok(enc, s, end, &next); | |
2267 | *eventEndPP = next; | |
2268 | switch (tok) { | |
2269 | case XML_TOK_TRAILING_CR: | |
2270 | if (haveMore) { | |
2271 | *nextPtr = s; | |
2272 | return XML_ERROR_NONE; | |
2273 | } | |
2274 | *eventEndPP = end; | |
2275 | if (characterDataHandler) { | |
2276 | XML_Char c = 0xA; | |
2277 | characterDataHandler(handlerArg, &c, 1); | |
2278 | } | |
2279 | else if (defaultHandler) | |
2280 | reportDefault(parser, enc, s, end); | |
2281 | /* We are at the end of the final buffer, should we check for | |
2282 | XML_SUSPENDED, XML_FINISHED? | |
2283 | */ | |
2284 | if (startTagLevel == 0) | |
2285 | return XML_ERROR_NO_ELEMENTS; | |
2286 | if (tagLevel != startTagLevel) | |
2287 | return XML_ERROR_ASYNC_ENTITY; | |
2288 | *nextPtr = end; | |
2289 | return XML_ERROR_NONE; | |
2290 | case XML_TOK_NONE: | |
2291 | if (haveMore) { | |
2292 | *nextPtr = s; | |
2293 | return XML_ERROR_NONE; | |
2294 | } | |
2295 | if (startTagLevel > 0) { | |
2296 | if (tagLevel != startTagLevel) | |
2297 | return XML_ERROR_ASYNC_ENTITY; | |
2298 | *nextPtr = s; | |
2299 | return XML_ERROR_NONE; | |
2300 | } | |
2301 | return XML_ERROR_NO_ELEMENTS; | |
2302 | case XML_TOK_INVALID: | |
2303 | *eventPP = next; | |
2304 | return XML_ERROR_INVALID_TOKEN; | |
2305 | case XML_TOK_PARTIAL: | |
2306 | if (haveMore) { | |
2307 | *nextPtr = s; | |
2308 | return XML_ERROR_NONE; | |
2309 | } | |
2310 | return XML_ERROR_UNCLOSED_TOKEN; | |
2311 | case XML_TOK_PARTIAL_CHAR: | |
2312 | if (haveMore) { | |
2313 | *nextPtr = s; | |
2314 | return XML_ERROR_NONE; | |
2315 | } | |
2316 | return XML_ERROR_PARTIAL_CHAR; | |
2317 | case XML_TOK_ENTITY_REF: | |
2318 | { | |
2319 | const XML_Char *name; | |
2320 | ENTITY *entity; | |
2321 | XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc, | |
2322 | s + enc->minBytesPerChar, | |
2323 | next - enc->minBytesPerChar); | |
2324 | if (ch) { | |
2325 | if (characterDataHandler) | |
2326 | characterDataHandler(handlerArg, &ch, 1); | |
2327 | else if (defaultHandler) | |
2328 | reportDefault(parser, enc, s, next); | |
2329 | break; | |
2330 | } | |
2331 | name = poolStoreString(&dtd->pool, enc, | |
2332 | s + enc->minBytesPerChar, | |
2333 | next - enc->minBytesPerChar); | |
2334 | if (!name) | |
2335 | return XML_ERROR_NO_MEMORY; | |
2336 | entity = (ENTITY *)lookup(parser, &dtd->generalEntities, name, 0); | |
2337 | poolDiscard(&dtd->pool); | |
2338 | /* First, determine if a check for an existing declaration is needed; | |
2339 | if yes, check that the entity exists, and that it is internal, | |
2340 | otherwise call the skipped entity or default handler. | |
2341 | */ | |
2342 | if (!dtd->hasParamEntityRefs || dtd->standalone) { | |
2343 | if (!entity) | |
2344 | return XML_ERROR_UNDEFINED_ENTITY; | |
2345 | else if (!entity->is_internal) | |
2346 | return XML_ERROR_ENTITY_DECLARED_IN_PE; | |
2347 | } | |
2348 | else if (!entity) { | |
2349 | if (skippedEntityHandler) | |
2350 | skippedEntityHandler(handlerArg, name, 0); | |
2351 | else if (defaultHandler) | |
2352 | reportDefault(parser, enc, s, next); | |
2353 | break; | |
2354 | } | |
2355 | if (entity->open) | |
2356 | return XML_ERROR_RECURSIVE_ENTITY_REF; | |
2357 | if (entity->notation) | |
2358 | return XML_ERROR_BINARY_ENTITY_REF; | |
2359 | if (entity->textPtr) { | |
2360 | enum XML_Error result; | |
2361 | if (!defaultExpandInternalEntities) { | |
2362 | if (skippedEntityHandler) | |
2363 | skippedEntityHandler(handlerArg, entity->name, 0); | |
2364 | else if (defaultHandler) | |
2365 | reportDefault(parser, enc, s, next); | |
2366 | break; | |
2367 | } | |
2368 | result = processInternalEntity(parser, entity, XML_FALSE); | |
2369 | if (result != XML_ERROR_NONE) | |
2370 | return result; | |
2371 | } | |
2372 | else if (externalEntityRefHandler) { | |
2373 | const XML_Char *context; | |
2374 | entity->open = XML_TRUE; | |
2375 | context = getContext(parser); | |
2376 | entity->open = XML_FALSE; | |
2377 | if (!context) | |
2378 | return XML_ERROR_NO_MEMORY; | |
2379 | if (!externalEntityRefHandler(externalEntityRefHandlerArg, | |
2380 | context, | |
2381 | entity->base, | |
2382 | entity->systemId, | |
2383 | entity->publicId)) | |
2384 | return XML_ERROR_EXTERNAL_ENTITY_HANDLING; | |
2385 | poolDiscard(&tempPool); | |
2386 | } | |
2387 | else if (defaultHandler) | |
2388 | reportDefault(parser, enc, s, next); | |
2389 | break; | |
2390 | } | |
2391 | case XML_TOK_START_TAG_NO_ATTS: | |
2392 | /* fall through */ | |
2393 | case XML_TOK_START_TAG_WITH_ATTS: | |
2394 | { | |
2395 | TAG *tag; | |
2396 | enum XML_Error result; | |
2397 | XML_Char *toPtr; | |
2398 | if (freeTagList) { | |
2399 | tag = freeTagList; | |
2400 | freeTagList = freeTagList->parent; | |
2401 | } | |
2402 | else { | |
2403 | tag = (TAG *)MALLOC(sizeof(TAG)); | |
2404 | if (!tag) | |
2405 | return XML_ERROR_NO_MEMORY; | |
2406 | tag->buf = (char *)MALLOC(INIT_TAG_BUF_SIZE); | |
2407 | if (!tag->buf) { | |
2408 | FREE(tag); | |
2409 | return XML_ERROR_NO_MEMORY; | |
2410 | } | |
2411 | tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE; | |
2412 | } | |
2413 | tag->bindings = NULL; | |
2414 | tag->parent = tagStack; | |
2415 | tagStack = tag; | |
2416 | tag->name.localPart = NULL; | |
2417 | tag->name.prefix = NULL; | |
2418 | tag->rawName = s + enc->minBytesPerChar; | |
2419 | tag->rawNameLength = XmlNameLength(enc, tag->rawName); | |
2420 | ++tagLevel; | |
2421 | { | |
2422 | const char *rawNameEnd = tag->rawName + tag->rawNameLength; | |
2423 | const char *fromPtr = tag->rawName; | |
2424 | toPtr = (XML_Char *)tag->buf; | |
2425 | for (;;) { | |
2426 | int bufSize; | |
2427 | int convLen; | |
2428 | XmlConvert(enc, | |
2429 | &fromPtr, rawNameEnd, | |
2430 | (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1); | |
2431 | convLen = (int)(toPtr - (XML_Char *)tag->buf); | |
2432 | if (fromPtr == rawNameEnd) { | |
2433 | tag->name.strLen = convLen; | |
2434 | break; | |
2435 | } | |
2436 | bufSize = (int)(tag->bufEnd - tag->buf) << 1; | |
2437 | { | |
2438 | char *temp = (char *)REALLOC(tag->buf, bufSize); | |
2439 | if (temp == NULL) | |
2440 | return XML_ERROR_NO_MEMORY; | |
2441 | tag->buf = temp; | |
2442 | tag->bufEnd = temp + bufSize; | |
2443 | toPtr = (XML_Char *)temp + convLen; | |
2444 | } | |
2445 | } | |
2446 | } | |
2447 | tag->name.str = (XML_Char *)tag->buf; | |
2448 | *toPtr = XML_T('\0'); | |
2449 | result = storeAtts(parser, enc, s, &(tag->name), &(tag->bindings)); | |
2450 | if (result) | |
2451 | return result; | |
2452 | if (startElementHandler) | |
2453 | startElementHandler(handlerArg, tag->name.str, | |
2454 | (const XML_Char **)atts); | |
2455 | else if (defaultHandler) | |
2456 | reportDefault(parser, enc, s, next); | |
2457 | poolClear(&tempPool); | |
2458 | break; | |
2459 | } | |
2460 | case XML_TOK_EMPTY_ELEMENT_NO_ATTS: | |
2461 | /* fall through */ | |
2462 | case XML_TOK_EMPTY_ELEMENT_WITH_ATTS: | |
2463 | { | |
2464 | const char *rawName = s + enc->minBytesPerChar; | |
2465 | enum XML_Error result; | |
2466 | BINDING *bindings = NULL; | |
2467 | XML_Bool noElmHandlers = XML_TRUE; | |
2468 | TAG_NAME name; | |
2469 | name.str = poolStoreString(&tempPool, enc, rawName, | |
2470 | rawName + XmlNameLength(enc, rawName)); | |
2471 | if (!name.str) | |
2472 | return XML_ERROR_NO_MEMORY; | |
2473 | poolFinish(&tempPool); | |
2474 | result = storeAtts(parser, enc, s, &name, &bindings); | |
2475 | if (result) | |
2476 | return result; | |
2477 | poolFinish(&tempPool); | |
2478 | if (startElementHandler) { | |
2479 | startElementHandler(handlerArg, name.str, (const XML_Char **)atts); | |
2480 | noElmHandlers = XML_FALSE; | |
2481 | } | |
2482 | if (endElementHandler) { | |
2483 | if (startElementHandler) | |
2484 | *eventPP = *eventEndPP; | |
2485 | endElementHandler(handlerArg, name.str); | |
2486 | noElmHandlers = XML_FALSE; | |
2487 | } | |
2488 | if (noElmHandlers && defaultHandler) | |
2489 | reportDefault(parser, enc, s, next); | |
2490 | poolClear(&tempPool); | |
2491 | while (bindings) { | |
2492 | BINDING *b = bindings; | |
2493 | if (endNamespaceDeclHandler) | |
2494 | endNamespaceDeclHandler(handlerArg, b->prefix->name); | |
2495 | bindings = bindings->nextTagBinding; | |
2496 | b->nextTagBinding = freeBindingList; | |
2497 | freeBindingList = b; | |
2498 | b->prefix->binding = b->prevPrefixBinding; | |
2499 | } | |
2500 | } | |
2501 | if (tagLevel == 0) | |
2502 | return epilogProcessor(parser, next, end, nextPtr); | |
2503 | break; | |
2504 | case XML_TOK_END_TAG: | |
2505 | if (tagLevel == startTagLevel) | |
2506 | return XML_ERROR_ASYNC_ENTITY; | |
2507 | else { | |
2508 | int len; | |
2509 | const char *rawName; | |
2510 | TAG *tag = tagStack; | |
2511 | tagStack = tag->parent; | |
2512 | tag->parent = freeTagList; | |
2513 | freeTagList = tag; | |
2514 | rawName = s + enc->minBytesPerChar*2; | |
2515 | len = XmlNameLength(enc, rawName); | |
2516 | if (len != tag->rawNameLength | |
2517 | || memcmp(tag->rawName, rawName, len) != 0) { | |
2518 | *eventPP = rawName; | |
2519 | return XML_ERROR_TAG_MISMATCH; | |
2520 | } | |
2521 | --tagLevel; | |
2522 | if (endElementHandler) { | |
2523 | const XML_Char *localPart; | |
2524 | const XML_Char *prefix; | |
2525 | XML_Char *uri; | |
2526 | localPart = tag->name.localPart; | |
2527 | if (ns && localPart) { | |
2528 | /* localPart and prefix may have been overwritten in | |
2529 | tag->name.str, since this points to the binding->uri | |
2530 | buffer which gets re-used; so we have to add them again | |
2531 | */ | |
2532 | uri = (XML_Char *)tag->name.str + tag->name.uriLen; | |
2533 | /* don't need to check for space - already done in storeAtts() */ | |
2534 | while (*localPart) *uri++ = *localPart++; | |
2535 | prefix = (XML_Char *)tag->name.prefix; | |
2536 | if (ns_triplets && prefix) { | |
2537 | *uri++ = namespaceSeparator; | |
2538 | while (*prefix) *uri++ = *prefix++; | |
2539 | } | |
2540 | *uri = XML_T('\0'); | |
2541 | } | |
2542 | endElementHandler(handlerArg, tag->name.str); | |
2543 | } | |
2544 | else if (defaultHandler) | |
2545 | reportDefault(parser, enc, s, next); | |
2546 | while (tag->bindings) { | |
2547 | BINDING *b = tag->bindings; | |
2548 | if (endNamespaceDeclHandler) | |
2549 | endNamespaceDeclHandler(handlerArg, b->prefix->name); | |
2550 | tag->bindings = tag->bindings->nextTagBinding; | |
2551 | b->nextTagBinding = freeBindingList; | |
2552 | freeBindingList = b; | |
2553 | b->prefix->binding = b->prevPrefixBinding; | |
2554 | } | |
2555 | if (tagLevel == 0) | |
2556 | return epilogProcessor(parser, next, end, nextPtr); | |
2557 | } | |
2558 | break; | |
2559 | case XML_TOK_CHAR_REF: | |
2560 | { | |
2561 | int n = XmlCharRefNumber(enc, s); | |
2562 | if (n < 0) | |
2563 | return XML_ERROR_BAD_CHAR_REF; | |
2564 | if (characterDataHandler) { | |
2565 | XML_Char buf[XML_ENCODE_MAX]; | |
2566 | characterDataHandler(handlerArg, buf, XmlEncode(n, (ICHAR *)buf)); | |
2567 | } | |
2568 | else if (defaultHandler) | |
2569 | reportDefault(parser, enc, s, next); | |
2570 | } | |
2571 | break; | |
2572 | case XML_TOK_XML_DECL: | |
2573 | return XML_ERROR_MISPLACED_XML_PI; | |
2574 | case XML_TOK_DATA_NEWLINE: | |
2575 | if (characterDataHandler) { | |
2576 | XML_Char c = 0xA; | |
2577 | characterDataHandler(handlerArg, &c, 1); | |
2578 | } | |
2579 | else if (defaultHandler) | |
2580 | reportDefault(parser, enc, s, next); | |
2581 | break; | |
2582 | case XML_TOK_CDATA_SECT_OPEN: | |
2583 | { | |
2584 | enum XML_Error result; | |
2585 | if (startCdataSectionHandler) | |
2586 | startCdataSectionHandler(handlerArg); | |
2587 | #if 0 | |
2588 | /* Suppose you doing a transformation on a document that involves | |
2589 | changing only the character data. You set up a defaultHandler | |
2590 | and a characterDataHandler. The defaultHandler simply copies | |
2591 | characters through. The characterDataHandler does the | |
2592 | transformation and writes the characters out escaping them as | |
2593 | necessary. This case will fail to work if we leave out the | |
2594 | following two lines (because & and < inside CDATA sections will | |
2595 | be incorrectly escaped). | |
2596 | ||
2597 | However, now we have a start/endCdataSectionHandler, so it seems | |
2598 | easier to let the user deal with this. | |
2599 | */ | |
2600 | else if (characterDataHandler) | |
2601 | characterDataHandler(handlerArg, dataBuf, 0); | |
2602 | #endif | |
2603 | else if (defaultHandler) | |
2604 | reportDefault(parser, enc, s, next); | |
2605 | result = doCdataSection(parser, enc, &next, end, nextPtr, haveMore); | |
2606 | if (result != XML_ERROR_NONE) | |
2607 | return result; | |
2608 | else if (!next) { | |
2609 | processor = cdataSectionProcessor; | |
2610 | return result; | |
2611 | } | |
2612 | } | |
2613 | break; | |
2614 | case XML_TOK_TRAILING_RSQB: | |
2615 | if (haveMore) { | |
2616 | *nextPtr = s; | |
2617 | return XML_ERROR_NONE; | |
2618 | } | |
2619 | if (characterDataHandler) { | |
2620 | if (MUST_CONVERT(enc, s)) { | |
2621 | ICHAR *dataPtr = (ICHAR *)dataBuf; | |
2622 | XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd); | |
2623 | characterDataHandler(handlerArg, dataBuf, | |
2624 | (int)(dataPtr - (ICHAR *)dataBuf)); | |
2625 | } | |
2626 | else | |
2627 | characterDataHandler(handlerArg, | |
2628 | (XML_Char *)s, | |
2629 | (int)((XML_Char *)end - (XML_Char *)s)); | |
2630 | } | |
2631 | else if (defaultHandler) | |
2632 | reportDefault(parser, enc, s, end); | |
2633 | /* We are at the end of the final buffer, should we check for | |
2634 | XML_SUSPENDED, XML_FINISHED? | |
2635 | */ | |
2636 | if (startTagLevel == 0) { | |
2637 | *eventPP = end; | |
2638 | return XML_ERROR_NO_ELEMENTS; | |
2639 | } | |
2640 | if (tagLevel != startTagLevel) { | |
2641 | *eventPP = end; | |
2642 | return XML_ERROR_ASYNC_ENTITY; | |
2643 | } | |
2644 | *nextPtr = end; | |
2645 | return XML_ERROR_NONE; | |
2646 | case XML_TOK_DATA_CHARS: | |
2647 | { | |
2648 | XML_CharacterDataHandler charDataHandler = characterDataHandler; | |
2649 | if (charDataHandler) { | |
2650 | if (MUST_CONVERT(enc, s)) { | |
2651 | for (;;) { | |
2652 | ICHAR *dataPtr = (ICHAR *)dataBuf; | |
2653 | XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); | |
2654 | *eventEndPP = s; | |
2655 | charDataHandler(handlerArg, dataBuf, | |
2656 | (int)(dataPtr - (ICHAR *)dataBuf)); | |
2657 | if (s == next) | |
2658 | break; | |
2659 | *eventPP = s; | |
2660 | } | |
2661 | } | |
2662 | else | |
2663 | charDataHandler(handlerArg, | |
2664 | (XML_Char *)s, | |
2665 | (int)((XML_Char *)next - (XML_Char *)s)); | |
2666 | } | |
2667 | else if (defaultHandler) | |
2668 | reportDefault(parser, enc, s, next); | |
2669 | } | |
2670 | break; | |
2671 | case XML_TOK_PI: | |
2672 | if (!reportProcessingInstruction(parser, enc, s, next)) | |
2673 | return XML_ERROR_NO_MEMORY; | |
2674 | break; | |
2675 | case XML_TOK_COMMENT: | |
2676 | if (!reportComment(parser, enc, s, next)) | |
2677 | return XML_ERROR_NO_MEMORY; | |
2678 | break; | |
2679 | default: | |
2680 | if (defaultHandler) | |
2681 | reportDefault(parser, enc, s, next); | |
2682 | break; | |
2683 | } | |
2684 | *eventPP = s = next; | |
2685 | switch (ps_parsing) { | |
2686 | case XML_SUSPENDED: | |
2687 | *nextPtr = next; | |
2688 | return XML_ERROR_NONE; | |
2689 | case XML_FINISHED: | |
2690 | return XML_ERROR_ABORTED; | |
2691 | default: ; | |
2692 | } | |
2693 | } | |
2694 | /* not reached */ | |
2695 | } | |
2696 | ||
2697 | /* Precondition: all arguments must be non-NULL; | |
2698 | Purpose: | |
2699 | - normalize attributes | |
2700 | - check attributes for well-formedness | |
2701 | - generate namespace aware attribute names (URI, prefix) | |
2702 | - build list of attributes for startElementHandler | |
2703 | - default attributes | |
2704 | - process namespace declarations (check and report them) | |
2705 | - generate namespace aware element name (URI, prefix) | |
2706 | */ | |
2707 | static enum XML_Error | |
2708 | storeAtts(XML_Parser parser, const ENCODING *enc, | |
2709 | const char *attStr, TAG_NAME *tagNamePtr, | |
2710 | BINDING **bindingsPtr) | |
2711 | { | |
2712 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
2713 | ELEMENT_TYPE *elementType; | |
2714 | int nDefaultAtts; | |
2715 | const XML_Char **appAtts; /* the attribute list for the application */ | |
2716 | int attIndex = 0; | |
2717 | int prefixLen; | |
2718 | int i; | |
2719 | int n; | |
2720 | XML_Char *uri; | |
2721 | int nPrefixes = 0; | |
2722 | BINDING *binding; | |
2723 | const XML_Char *localPart; | |
2724 | ||
2725 | /* lookup the element type name */ | |
2726 | elementType = (ELEMENT_TYPE *)lookup(parser, &dtd->elementTypes, tagNamePtr->str,0); | |
2727 | if (!elementType) { | |
2728 | const XML_Char *name = poolCopyString(&dtd->pool, tagNamePtr->str); | |
2729 | if (!name) | |
2730 | return XML_ERROR_NO_MEMORY; | |
2731 | elementType = (ELEMENT_TYPE *)lookup(parser, &dtd->elementTypes, name, | |
2732 | sizeof(ELEMENT_TYPE)); | |
2733 | if (!elementType) | |
2734 | return XML_ERROR_NO_MEMORY; | |
2735 | if (ns && !setElementTypePrefix(parser, elementType)) | |
2736 | return XML_ERROR_NO_MEMORY; | |
2737 | } | |
2738 | nDefaultAtts = elementType->nDefaultAtts; | |
2739 | ||
2740 | /* get the attributes from the tokenizer */ | |
2741 | n = XmlGetAttributes(enc, attStr, attsSize, atts); | |
2742 | if (n + nDefaultAtts > attsSize) { | |
2743 | int oldAttsSize = attsSize; | |
2744 | ATTRIBUTE *temp; | |
2745 | #ifdef XML_ATTR_INFO | |
2746 | XML_AttrInfo *temp2; | |
2747 | #endif | |
2748 | attsSize = n + nDefaultAtts + INIT_ATTS_SIZE; | |
2749 | temp = (ATTRIBUTE *)REALLOC((void *)atts, attsSize * sizeof(ATTRIBUTE)); | |
2750 | if (temp == NULL) | |
2751 | return XML_ERROR_NO_MEMORY; | |
2752 | atts = temp; | |
2753 | #ifdef XML_ATTR_INFO | |
2754 | temp2 = (XML_AttrInfo *)REALLOC((void *)attInfo, attsSize * sizeof(XML_AttrInfo)); | |
2755 | if (temp2 == NULL) | |
2756 | return XML_ERROR_NO_MEMORY; | |
2757 | attInfo = temp2; | |
2758 | #endif | |
2759 | if (n > oldAttsSize) | |
2760 | XmlGetAttributes(enc, attStr, n, atts); | |
2761 | } | |
2762 | ||
2763 | appAtts = (const XML_Char **)atts; | |
2764 | for (i = 0; i < n; i++) { | |
2765 | ATTRIBUTE *currAtt = &atts[i]; | |
2766 | #ifdef XML_ATTR_INFO | |
2767 | XML_AttrInfo *currAttInfo = &attInfo[i]; | |
2768 | #endif | |
2769 | /* add the name and value to the attribute list */ | |
2770 | ATTRIBUTE_ID *attId = getAttributeId(parser, enc, currAtt->name, | |
2771 | currAtt->name | |
2772 | + XmlNameLength(enc, currAtt->name)); | |
2773 | if (!attId) | |
2774 | return XML_ERROR_NO_MEMORY; | |
2775 | #ifdef XML_ATTR_INFO | |
2776 | currAttInfo->nameStart = parseEndByteIndex - (parseEndPtr - currAtt->name); | |
2777 | currAttInfo->nameEnd = currAttInfo->nameStart + | |
2778 | XmlNameLength(enc, currAtt->name); | |
2779 | currAttInfo->valueStart = parseEndByteIndex - | |
2780 | (parseEndPtr - currAtt->valuePtr); | |
2781 | currAttInfo->valueEnd = parseEndByteIndex - (parseEndPtr - currAtt->valueEnd); | |
2782 | #endif | |
2783 | /* Detect duplicate attributes by their QNames. This does not work when | |
2784 | namespace processing is turned on and different prefixes for the same | |
2785 | namespace are used. For this case we have a check further down. | |
2786 | */ | |
2787 | if ((attId->name)[-1]) { | |
2788 | if (enc == encoding) | |
2789 | eventPtr = atts[i].name; | |
2790 | return XML_ERROR_DUPLICATE_ATTRIBUTE; | |
2791 | } | |
2792 | (attId->name)[-1] = 1; | |
2793 | appAtts[attIndex++] = attId->name; | |
2794 | if (!atts[i].normalized) { | |
2795 | enum XML_Error result; | |
2796 | XML_Bool isCdata = XML_TRUE; | |
2797 | ||
2798 | /* figure out whether declared as other than CDATA */ | |
2799 | if (attId->maybeTokenized) { | |
2800 | int j; | |
2801 | for (j = 0; j < nDefaultAtts; j++) { | |
2802 | if (attId == elementType->defaultAtts[j].id) { | |
2803 | isCdata = elementType->defaultAtts[j].isCdata; | |
2804 | break; | |
2805 | } | |
2806 | } | |
2807 | } | |
2808 | ||
2809 | /* normalize the attribute value */ | |
2810 | result = storeAttributeValue(parser, enc, isCdata, | |
2811 | atts[i].valuePtr, atts[i].valueEnd, | |
2812 | &tempPool); | |
2813 | if (result) | |
2814 | return result; | |
2815 | appAtts[attIndex] = poolStart(&tempPool); | |
2816 | poolFinish(&tempPool); | |
2817 | } | |
2818 | else { | |
2819 | /* the value did not need normalizing */ | |
2820 | appAtts[attIndex] = poolStoreString(&tempPool, enc, atts[i].valuePtr, | |
2821 | atts[i].valueEnd); | |
2822 | if (appAtts[attIndex] == 0) | |
2823 | return XML_ERROR_NO_MEMORY; | |
2824 | poolFinish(&tempPool); | |
2825 | } | |
2826 | /* handle prefixed attribute names */ | |
2827 | if (attId->prefix) { | |
2828 | if (attId->xmlns) { | |
2829 | /* deal with namespace declarations here */ | |
2830 | enum XML_Error result = addBinding(parser, attId->prefix, attId, | |
2831 | appAtts[attIndex], bindingsPtr); | |
2832 | if (result) | |
2833 | return result; | |
2834 | --attIndex; | |
2835 | } | |
2836 | else { | |
2837 | /* deal with other prefixed names later */ | |
2838 | attIndex++; | |
2839 | nPrefixes++; | |
2840 | (attId->name)[-1] = 2; | |
2841 | } | |
2842 | } | |
2843 | else | |
2844 | attIndex++; | |
2845 | } | |
2846 | ||
2847 | /* set-up for XML_GetSpecifiedAttributeCount and XML_GetIdAttributeIndex */ | |
2848 | nSpecifiedAtts = attIndex; | |
2849 | if (elementType->idAtt && (elementType->idAtt->name)[-1]) { | |
2850 | for (i = 0; i < attIndex; i += 2) | |
2851 | if (appAtts[i] == elementType->idAtt->name) { | |
2852 | idAttIndex = i; | |
2853 | break; | |
2854 | } | |
2855 | } | |
2856 | else | |
2857 | idAttIndex = -1; | |
2858 | ||
2859 | /* do attribute defaulting */ | |
2860 | for (i = 0; i < nDefaultAtts; i++) { | |
2861 | const DEFAULT_ATTRIBUTE *da = elementType->defaultAtts + i; | |
2862 | if (!(da->id->name)[-1] && da->value) { | |
2863 | if (da->id->prefix) { | |
2864 | if (da->id->xmlns) { | |
2865 | enum XML_Error result = addBinding(parser, da->id->prefix, da->id, | |
2866 | da->value, bindingsPtr); | |
2867 | if (result) | |
2868 | return result; | |
2869 | } | |
2870 | else { | |
2871 | (da->id->name)[-1] = 2; | |
2872 | nPrefixes++; | |
2873 | appAtts[attIndex++] = da->id->name; | |
2874 | appAtts[attIndex++] = da->value; | |
2875 | } | |
2876 | } | |
2877 | else { | |
2878 | (da->id->name)[-1] = 1; | |
2879 | appAtts[attIndex++] = da->id->name; | |
2880 | appAtts[attIndex++] = da->value; | |
2881 | } | |
2882 | } | |
2883 | } | |
2884 | appAtts[attIndex] = 0; | |
2885 | ||
2886 | /* expand prefixed attribute names, check for duplicates, | |
2887 | and clear flags that say whether attributes were specified */ | |
2888 | i = 0; | |
2889 | if (nPrefixes) { | |
2890 | int j; /* hash table index */ | |
2891 | unsigned long version = nsAttsVersion; | |
2892 | int nsAttsSize = (int)1 << nsAttsPower; | |
2893 | /* size of hash table must be at least 2 * (# of prefixed attributes) */ | |
2894 | if ((nPrefixes << 1) >> nsAttsPower) { /* true for nsAttsPower = 0 */ | |
2895 | NS_ATT *temp; | |
2896 | /* hash table size must also be a power of 2 and >= 8 */ | |
2897 | while (nPrefixes >> nsAttsPower++); | |
2898 | if (nsAttsPower < 3) | |
2899 | nsAttsPower = 3; | |
2900 | nsAttsSize = (int)1 << nsAttsPower; | |
2901 | temp = (NS_ATT *)REALLOC(nsAtts, nsAttsSize * sizeof(NS_ATT)); | |
2902 | if (!temp) | |
2903 | return XML_ERROR_NO_MEMORY; | |
2904 | nsAtts = temp; | |
2905 | version = 0; /* force re-initialization of nsAtts hash table */ | |
2906 | } | |
2907 | /* using a version flag saves us from initializing nsAtts every time */ | |
2908 | if (!version) { /* initialize version flags when version wraps around */ | |
2909 | version = INIT_ATTS_VERSION; | |
2910 | for (j = nsAttsSize; j != 0; ) | |
2911 | nsAtts[--j].version = version; | |
2912 | } | |
2913 | nsAttsVersion = --version; | |
2914 | ||
2915 | /* expand prefixed names and check for duplicates */ | |
2916 | for (; i < attIndex; i += 2) { | |
2917 | const XML_Char *s = appAtts[i]; | |
2918 | if (s[-1] == 2) { /* prefixed */ | |
2919 | ATTRIBUTE_ID *id; | |
2920 | const BINDING *b; | |
2921 | unsigned long uriHash = hash_secret_salt; | |
2922 | ((XML_Char *)s)[-1] = 0; /* clear flag */ | |
2923 | id = (ATTRIBUTE_ID *)lookup(parser, &dtd->attributeIds, s, 0); | |
2924 | b = id->prefix->binding; | |
2925 | if (!b) | |
2926 | return XML_ERROR_UNBOUND_PREFIX; | |
2927 | ||
2928 | /* as we expand the name we also calculate its hash value */ | |
2929 | for (j = 0; j < b->uriLen; j++) { | |
2930 | const XML_Char c = b->uri[j]; | |
2931 | if (!poolAppendChar(&tempPool, c)) | |
2932 | return XML_ERROR_NO_MEMORY; | |
2933 | uriHash = CHAR_HASH(uriHash, c); | |
2934 | } | |
2935 | while (*s++ != XML_T(ASCII_COLON)) | |
2936 | ; | |
2937 | do { /* copies null terminator */ | |
2938 | const XML_Char c = *s; | |
2939 | if (!poolAppendChar(&tempPool, *s)) | |
2940 | return XML_ERROR_NO_MEMORY; | |
2941 | uriHash = CHAR_HASH(uriHash, c); | |
2942 | } while (*s++); | |
2943 | ||
2944 | { /* Check hash table for duplicate of expanded name (uriName). | |
2945 | Derived from code in lookup(parser, HASH_TABLE *table, ...). | |
2946 | */ | |
2947 | unsigned char step = 0; | |
2948 | unsigned long mask = nsAttsSize - 1; | |
2949 | j = uriHash & mask; /* index into hash table */ | |
2950 | while (nsAtts[j].version == version) { | |
2951 | /* for speed we compare stored hash values first */ | |
2952 | if (uriHash == nsAtts[j].hash) { | |
2953 | const XML_Char *s1 = poolStart(&tempPool); | |
2954 | const XML_Char *s2 = nsAtts[j].uriName; | |
2955 | /* s1 is null terminated, but not s2 */ | |
2956 | for (; *s1 == *s2 && *s1 != 0; s1++, s2++); | |
2957 | if (*s1 == 0) | |
2958 | return XML_ERROR_DUPLICATE_ATTRIBUTE; | |
2959 | } | |
2960 | if (!step) | |
2961 | step = PROBE_STEP(uriHash, mask, nsAttsPower); | |
2962 | j < step ? (j += nsAttsSize - step) : (j -= step); | |
2963 | } | |
2964 | } | |
2965 | ||
2966 | if (ns_triplets) { /* append namespace separator and prefix */ | |
2967 | tempPool.ptr[-1] = namespaceSeparator; | |
2968 | s = b->prefix->name; | |
2969 | do { | |
2970 | if (!poolAppendChar(&tempPool, *s)) | |
2971 | return XML_ERROR_NO_MEMORY; | |
2972 | } while (*s++); | |
2973 | } | |
2974 | ||
2975 | /* store expanded name in attribute list */ | |
2976 | s = poolStart(&tempPool); | |
2977 | poolFinish(&tempPool); | |
2978 | appAtts[i] = s; | |
2979 | ||
2980 | /* fill empty slot with new version, uriName and hash value */ | |
2981 | nsAtts[j].version = version; | |
2982 | nsAtts[j].hash = uriHash; | |
2983 | nsAtts[j].uriName = s; | |
2984 | ||
2985 | if (!--nPrefixes) { | |
2986 | i += 2; | |
2987 | break; | |
2988 | } | |
2989 | } | |
2990 | else /* not prefixed */ | |
2991 | ((XML_Char *)s)[-1] = 0; /* clear flag */ | |
2992 | } | |
2993 | } | |
2994 | /* clear flags for the remaining attributes */ | |
2995 | for (; i < attIndex; i += 2) | |
2996 | ((XML_Char *)(appAtts[i]))[-1] = 0; | |
2997 | for (binding = *bindingsPtr; binding; binding = binding->nextTagBinding) | |
2998 | binding->attId->name[-1] = 0; | |
2999 | ||
3000 | if (!ns) | |
3001 | return XML_ERROR_NONE; | |
3002 | ||
3003 | /* expand the element type name */ | |
3004 | if (elementType->prefix) { | |
3005 | binding = elementType->prefix->binding; | |
3006 | if (!binding) | |
3007 | return XML_ERROR_UNBOUND_PREFIX; | |
3008 | localPart = tagNamePtr->str; | |
3009 | while (*localPart++ != XML_T(ASCII_COLON)) | |
3010 | ; | |
3011 | } | |
3012 | else if (dtd->defaultPrefix.binding) { | |
3013 | binding = dtd->defaultPrefix.binding; | |
3014 | localPart = tagNamePtr->str; | |
3015 | } | |
3016 | else | |
3017 | return XML_ERROR_NONE; | |
3018 | prefixLen = 0; | |
3019 | if (ns_triplets && binding->prefix->name) { | |
3020 | for (; binding->prefix->name[prefixLen++];) | |
3021 | ; /* prefixLen includes null terminator */ | |
3022 | } | |
3023 | tagNamePtr->localPart = localPart; | |
3024 | tagNamePtr->uriLen = binding->uriLen; | |
3025 | tagNamePtr->prefix = binding->prefix->name; | |
3026 | tagNamePtr->prefixLen = prefixLen; | |
3027 | for (i = 0; localPart[i++];) | |
3028 | ; /* i includes null terminator */ | |
3029 | n = i + binding->uriLen + prefixLen; | |
3030 | if (n > binding->uriAlloc) { | |
3031 | TAG *p; | |
3032 | uri = (XML_Char *)MALLOC((n + EXPAND_SPARE) * sizeof(XML_Char)); | |
3033 | if (!uri) | |
3034 | return XML_ERROR_NO_MEMORY; | |
3035 | binding->uriAlloc = n + EXPAND_SPARE; | |
3036 | memcpy(uri, binding->uri, binding->uriLen * sizeof(XML_Char)); | |
3037 | for (p = tagStack; p; p = p->parent) | |
3038 | if (p->name.str == binding->uri) | |
3039 | p->name.str = uri; | |
3040 | FREE(binding->uri); | |
3041 | binding->uri = uri; | |
3042 | } | |
3043 | /* if namespaceSeparator != '\0' then uri includes it already */ | |
3044 | uri = binding->uri + binding->uriLen; | |
3045 | memcpy(uri, localPart, i * sizeof(XML_Char)); | |
3046 | /* we always have a namespace separator between localPart and prefix */ | |
3047 | if (prefixLen) { | |
3048 | uri += i - 1; | |
3049 | *uri = namespaceSeparator; /* replace null terminator */ | |
3050 | memcpy(uri + 1, binding->prefix->name, prefixLen * sizeof(XML_Char)); | |
3051 | } | |
3052 | tagNamePtr->str = binding->uri; | |
3053 | return XML_ERROR_NONE; | |
3054 | } | |
3055 | ||
3056 | /* addBinding() overwrites the value of prefix->binding without checking. | |
3057 | Therefore one must keep track of the old value outside of addBinding(). | |
3058 | */ | |
3059 | static enum XML_Error | |
3060 | addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, | |
3061 | const XML_Char *uri, BINDING **bindingsPtr) | |
3062 | { | |
3063 | static const XML_Char xmlNamespace[] = { | |
3064 | ASCII_h, ASCII_t, ASCII_t, ASCII_p, ASCII_COLON, ASCII_SLASH, ASCII_SLASH, | |
3065 | ASCII_w, ASCII_w, ASCII_w, ASCII_PERIOD, ASCII_w, ASCII_3, ASCII_PERIOD, | |
3066 | ASCII_o, ASCII_r, ASCII_g, ASCII_SLASH, ASCII_X, ASCII_M, ASCII_L, | |
3067 | ASCII_SLASH, ASCII_1, ASCII_9, ASCII_9, ASCII_8, ASCII_SLASH, | |
3068 | ASCII_n, ASCII_a, ASCII_m, ASCII_e, ASCII_s, ASCII_p, ASCII_a, ASCII_c, | |
3069 | ASCII_e, '\0' | |
3070 | }; | |
3071 | static const int xmlLen = | |
3072 | (int)sizeof(xmlNamespace)/sizeof(XML_Char) - 1; | |
3073 | static const XML_Char xmlnsNamespace[] = { | |
3074 | ASCII_h, ASCII_t, ASCII_t, ASCII_p, ASCII_COLON, ASCII_SLASH, ASCII_SLASH, | |
3075 | ASCII_w, ASCII_w, ASCII_w, ASCII_PERIOD, ASCII_w, ASCII_3, ASCII_PERIOD, | |
3076 | ASCII_o, ASCII_r, ASCII_g, ASCII_SLASH, ASCII_2, ASCII_0, ASCII_0, | |
3077 | ASCII_0, ASCII_SLASH, ASCII_x, ASCII_m, ASCII_l, ASCII_n, ASCII_s, | |
3078 | ASCII_SLASH, '\0' | |
3079 | }; | |
3080 | static const int xmlnsLen = | |
3081 | (int)sizeof(xmlnsNamespace)/sizeof(XML_Char) - 1; | |
3082 | ||
3083 | XML_Bool mustBeXML = XML_FALSE; | |
3084 | XML_Bool isXML = XML_TRUE; | |
3085 | XML_Bool isXMLNS = XML_TRUE; | |
3086 | ||
3087 | BINDING *b; | |
3088 | int len; | |
3089 | ||
3090 | /* empty URI is only valid for default namespace per XML NS 1.0 (not 1.1) */ | |
3091 | if (*uri == XML_T('\0') && prefix->name) | |
3092 | return XML_ERROR_UNDECLARING_PREFIX; | |
3093 | ||
3094 | if (prefix->name | |
3095 | && prefix->name[0] == XML_T(ASCII_x) | |
3096 | && prefix->name[1] == XML_T(ASCII_m) | |
3097 | && prefix->name[2] == XML_T(ASCII_l)) { | |
3098 | ||
3099 | /* Not allowed to bind xmlns */ | |
3100 | if (prefix->name[3] == XML_T(ASCII_n) | |
3101 | && prefix->name[4] == XML_T(ASCII_s) | |
3102 | && prefix->name[5] == XML_T('\0')) | |
3103 | return XML_ERROR_RESERVED_PREFIX_XMLNS; | |
3104 | ||
3105 | if (prefix->name[3] == XML_T('\0')) | |
3106 | mustBeXML = XML_TRUE; | |
3107 | } | |
3108 | ||
3109 | for (len = 0; uri[len]; len++) { | |
3110 | if (isXML && (len > xmlLen || uri[len] != xmlNamespace[len])) | |
3111 | isXML = XML_FALSE; | |
3112 | ||
3113 | if (!mustBeXML && isXMLNS | |
3114 | && (len > xmlnsLen || uri[len] != xmlnsNamespace[len])) | |
3115 | isXMLNS = XML_FALSE; | |
3116 | } | |
3117 | isXML = isXML && len == xmlLen; | |
3118 | isXMLNS = isXMLNS && len == xmlnsLen; | |
3119 | ||
3120 | if (mustBeXML != isXML) | |
3121 | return mustBeXML ? XML_ERROR_RESERVED_PREFIX_XML | |
3122 | : XML_ERROR_RESERVED_NAMESPACE_URI; | |
3123 | ||
3124 | if (isXMLNS) | |
3125 | return XML_ERROR_RESERVED_NAMESPACE_URI; | |
3126 | ||
3127 | if (namespaceSeparator) | |
3128 | len++; | |
3129 | if (freeBindingList) { | |
3130 | b = freeBindingList; | |
3131 | if (len > b->uriAlloc) { | |
3132 | XML_Char *temp = (XML_Char *)REALLOC(b->uri, | |
3133 | sizeof(XML_Char) * (len + EXPAND_SPARE)); | |
3134 | if (temp == NULL) | |
3135 | return XML_ERROR_NO_MEMORY; | |
3136 | b->uri = temp; | |
3137 | b->uriAlloc = len + EXPAND_SPARE; | |
3138 | } | |
3139 | freeBindingList = b->nextTagBinding; | |
3140 | } | |
3141 | else { | |
3142 | b = (BINDING *)MALLOC(sizeof(BINDING)); | |
3143 | if (!b) | |
3144 | return XML_ERROR_NO_MEMORY; | |
3145 | b->uri = (XML_Char *)MALLOC(sizeof(XML_Char) * (len + EXPAND_SPARE)); | |
3146 | if (!b->uri) { | |
3147 | FREE(b); | |
3148 | return XML_ERROR_NO_MEMORY; | |
3149 | } | |
3150 | b->uriAlloc = len + EXPAND_SPARE; | |
3151 | } | |
3152 | b->uriLen = len; | |
3153 | memcpy(b->uri, uri, len * sizeof(XML_Char)); | |
3154 | if (namespaceSeparator) | |
3155 | b->uri[len - 1] = namespaceSeparator; | |
3156 | b->prefix = prefix; | |
3157 | b->attId = attId; | |
3158 | b->prevPrefixBinding = prefix->binding; | |
3159 | /* NULL binding when default namespace undeclared */ | |
3160 | if (*uri == XML_T('\0') && prefix == &_dtd->defaultPrefix) | |
3161 | prefix->binding = NULL; | |
3162 | else | |
3163 | prefix->binding = b; | |
3164 | b->nextTagBinding = *bindingsPtr; | |
3165 | *bindingsPtr = b; | |
3166 | /* if attId == NULL then we are not starting a namespace scope */ | |
3167 | if (attId && startNamespaceDeclHandler) | |
3168 | startNamespaceDeclHandler(handlerArg, prefix->name, | |
3169 | prefix->binding ? uri : 0); | |
3170 | return XML_ERROR_NONE; | |
3171 | } | |
3172 | ||
3173 | /* The idea here is to avoid using stack for each CDATA section when | |
3174 | the whole file is parsed with one call. | |
3175 | */ | |
3176 | static enum XML_Error PTRCALL | |
3177 | cdataSectionProcessor(XML_Parser parser, | |
3178 | const char *start, | |
3179 | const char *end, | |
3180 | const char **endPtr) | |
3181 | { | |
3182 | enum XML_Error result = doCdataSection(parser, encoding, &start, end, | |
3183 | endPtr, (XML_Bool)!ps_finalBuffer); | |
3184 | if (result != XML_ERROR_NONE) | |
3185 | return result; | |
3186 | if (start) { | |
3187 | if (parentParser) { /* we are parsing an external entity */ | |
3188 | processor = externalEntityContentProcessor; | |
3189 | return externalEntityContentProcessor(parser, start, end, endPtr); | |
3190 | } | |
3191 | else { | |
3192 | processor = contentProcessor; | |
3193 | return contentProcessor(parser, start, end, endPtr); | |
3194 | } | |
3195 | } | |
3196 | return result; | |
3197 | } | |
3198 | ||
3199 | /* startPtr gets set to non-null if the section is closed, and to null if | |
3200 | the section is not yet closed. | |
3201 | */ | |
3202 | static enum XML_Error | |
3203 | doCdataSection(XML_Parser parser, | |
3204 | const ENCODING *enc, | |
3205 | const char **startPtr, | |
3206 | const char *end, | |
3207 | const char **nextPtr, | |
3208 | XML_Bool haveMore) | |
3209 | { | |
3210 | const char *s = *startPtr; | |
3211 | const char **eventPP; | |
3212 | const char **eventEndPP; | |
3213 | if (enc == encoding) { | |
3214 | eventPP = &eventPtr; | |
3215 | *eventPP = s; | |
3216 | eventEndPP = &eventEndPtr; | |
3217 | } | |
3218 | else { | |
3219 | eventPP = &(openInternalEntities->internalEventPtr); | |
3220 | eventEndPP = &(openInternalEntities->internalEventEndPtr); | |
3221 | } | |
3222 | *eventPP = s; | |
3223 | *startPtr = NULL; | |
3224 | ||
3225 | for (;;) { | |
3226 | const char *next; | |
3227 | int tok = XmlCdataSectionTok(enc, s, end, &next); | |
3228 | *eventEndPP = next; | |
3229 | switch (tok) { | |
3230 | case XML_TOK_CDATA_SECT_CLOSE: | |
3231 | if (endCdataSectionHandler) | |
3232 | endCdataSectionHandler(handlerArg); | |
3233 | #if 0 | |
3234 | /* see comment under XML_TOK_CDATA_SECT_OPEN */ | |
3235 | else if (characterDataHandler) | |
3236 | characterDataHandler(handlerArg, dataBuf, 0); | |
3237 | #endif | |
3238 | else if (defaultHandler) | |
3239 | reportDefault(parser, enc, s, next); | |
3240 | *startPtr = next; | |
3241 | *nextPtr = next; | |
3242 | if (ps_parsing == XML_FINISHED) | |
3243 | return XML_ERROR_ABORTED; | |
3244 | else | |
3245 | return XML_ERROR_NONE; | |
3246 | case XML_TOK_DATA_NEWLINE: | |
3247 | if (characterDataHandler) { | |
3248 | XML_Char c = 0xA; | |
3249 | characterDataHandler(handlerArg, &c, 1); | |
3250 | } | |
3251 | else if (defaultHandler) | |
3252 | reportDefault(parser, enc, s, next); | |
3253 | break; | |
3254 | case XML_TOK_DATA_CHARS: | |
3255 | { | |
3256 | XML_CharacterDataHandler charDataHandler = characterDataHandler; | |
3257 | if (charDataHandler) { | |
3258 | if (MUST_CONVERT(enc, s)) { | |
3259 | for (;;) { | |
3260 | ICHAR *dataPtr = (ICHAR *)dataBuf; | |
3261 | XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd); | |
3262 | *eventEndPP = next; | |
3263 | charDataHandler(handlerArg, dataBuf, | |
3264 | (int)(dataPtr - (ICHAR *)dataBuf)); | |
3265 | if (s == next) | |
3266 | break; | |
3267 | *eventPP = s; | |
3268 | } | |
3269 | } | |
3270 | else | |
3271 | charDataHandler(handlerArg, | |
3272 | (XML_Char *)s, | |
3273 | (int)((XML_Char *)next - (XML_Char *)s)); | |
3274 | } | |
3275 | else if (defaultHandler) | |
3276 | reportDefault(parser, enc, s, next); | |
3277 | } | |
3278 | break; | |
3279 | case XML_TOK_INVALID: | |
3280 | *eventPP = next; | |
3281 | return XML_ERROR_INVALID_TOKEN; | |
3282 | case XML_TOK_PARTIAL_CHAR: | |
3283 | if (haveMore) { | |
3284 | *nextPtr = s; | |
3285 | return XML_ERROR_NONE; | |
3286 | } | |
3287 | return XML_ERROR_PARTIAL_CHAR; | |
3288 | case XML_TOK_PARTIAL: | |
3289 | case XML_TOK_NONE: | |
3290 | if (haveMore) { | |
3291 | *nextPtr = s; | |
3292 | return XML_ERROR_NONE; | |
3293 | } | |
3294 | return XML_ERROR_UNCLOSED_CDATA_SECTION; | |
3295 | default: | |
3296 | *eventPP = next; | |
3297 | return XML_ERROR_UNEXPECTED_STATE; | |
3298 | } | |
3299 | ||
3300 | *eventPP = s = next; | |
3301 | switch (ps_parsing) { | |
3302 | case XML_SUSPENDED: | |
3303 | *nextPtr = next; | |
3304 | return XML_ERROR_NONE; | |
3305 | case XML_FINISHED: | |
3306 | return XML_ERROR_ABORTED; | |
3307 | default: ; | |
3308 | } | |
3309 | } | |
3310 | /* not reached */ | |
3311 | } | |
3312 | ||
3313 | #ifdef XML_DTD | |
3314 | ||
3315 | /* The idea here is to avoid using stack for each IGNORE section when | |
3316 | the whole file is parsed with one call. | |
3317 | */ | |
3318 | static enum XML_Error PTRCALL | |
3319 | ignoreSectionProcessor(XML_Parser parser, | |
3320 | const char *start, | |
3321 | const char *end, | |
3322 | const char **endPtr) | |
3323 | { | |
3324 | enum XML_Error result = doIgnoreSection(parser, encoding, &start, end, | |
3325 | endPtr, (XML_Bool)!ps_finalBuffer); | |
3326 | if (result != XML_ERROR_NONE) | |
3327 | return result; | |
3328 | if (start) { | |
3329 | processor = prologProcessor; | |
3330 | return prologProcessor(parser, start, end, endPtr); | |
3331 | } | |
3332 | return result; | |
3333 | } | |
3334 | ||
3335 | /* startPtr gets set to non-null is the section is closed, and to null | |
3336 | if the section is not yet closed. | |
3337 | */ | |
3338 | static enum XML_Error | |
3339 | doIgnoreSection(XML_Parser parser, | |
3340 | const ENCODING *enc, | |
3341 | const char **startPtr, | |
3342 | const char *end, | |
3343 | const char **nextPtr, | |
3344 | XML_Bool haveMore) | |
3345 | { | |
3346 | const char *next; | |
3347 | int tok; | |
3348 | const char *s = *startPtr; | |
3349 | const char **eventPP; | |
3350 | const char **eventEndPP; | |
3351 | if (enc == encoding) { | |
3352 | eventPP = &eventPtr; | |
3353 | *eventPP = s; | |
3354 | eventEndPP = &eventEndPtr; | |
3355 | } | |
3356 | else { | |
3357 | eventPP = &(openInternalEntities->internalEventPtr); | |
3358 | eventEndPP = &(openInternalEntities->internalEventEndPtr); | |
3359 | } | |
3360 | *eventPP = s; | |
3361 | *startPtr = NULL; | |
3362 | tok = XmlIgnoreSectionTok(enc, s, end, &next); | |
3363 | *eventEndPP = next; | |
3364 | switch (tok) { | |
3365 | case XML_TOK_IGNORE_SECT: | |
3366 | if (defaultHandler) | |
3367 | reportDefault(parser, enc, s, next); | |
3368 | *startPtr = next; | |
3369 | *nextPtr = next; | |
3370 | if (ps_parsing == XML_FINISHED) | |
3371 | return XML_ERROR_ABORTED; | |
3372 | else | |
3373 | return XML_ERROR_NONE; | |
3374 | case XML_TOK_INVALID: | |
3375 | *eventPP = next; | |
3376 | return XML_ERROR_INVALID_TOKEN; | |
3377 | case XML_TOK_PARTIAL_CHAR: | |
3378 | if (haveMore) { | |
3379 | *nextPtr = s; | |
3380 | return XML_ERROR_NONE; | |
3381 | } | |
3382 | return XML_ERROR_PARTIAL_CHAR; | |
3383 | case XML_TOK_PARTIAL: | |
3384 | case XML_TOK_NONE: | |
3385 | if (haveMore) { | |
3386 | *nextPtr = s; | |
3387 | return XML_ERROR_NONE; | |
3388 | } | |
3389 | return XML_ERROR_SYNTAX; /* XML_ERROR_UNCLOSED_IGNORE_SECTION */ | |
3390 | default: | |
3391 | *eventPP = next; | |
3392 | return XML_ERROR_UNEXPECTED_STATE; | |
3393 | } | |
3394 | /* not reached */ | |
3395 | } | |
3396 | ||
3397 | #endif /* XML_DTD */ | |
3398 | ||
3399 | static enum XML_Error | |
3400 | initializeEncoding(XML_Parser parser) | |
3401 | { | |
3402 | const char *s; | |
3403 | #ifdef XML_UNICODE | |
3404 | char encodingBuf[128]; | |
3405 | if (!protocolEncodingName) | |
3406 | s = NULL; | |
3407 | else { | |
3408 | int i; | |
3409 | for (i = 0; protocolEncodingName[i]; i++) { | |
3410 | if (i == sizeof(encodingBuf) - 1 | |
3411 | || (protocolEncodingName[i] & ~0x7f) != 0) { | |
3412 | encodingBuf[0] = '\0'; | |
3413 | break; | |
3414 | } | |
3415 | encodingBuf[i] = (char)protocolEncodingName[i]; | |
3416 | } | |
3417 | encodingBuf[i] = '\0'; | |
3418 | s = encodingBuf; | |
3419 | } | |
3420 | #else | |
3421 | s = protocolEncodingName; | |
3422 | #endif | |
3423 | if ((ns ? XmlInitEncodingNS : XmlInitEncoding)(&initEncoding, &encoding, s)) | |
3424 | return XML_ERROR_NONE; | |
3425 | return handleUnknownEncoding(parser, protocolEncodingName); | |
3426 | } | |
3427 | ||
3428 | static enum XML_Error | |
3429 | processXmlDecl(XML_Parser parser, int isGeneralTextEntity, | |
3430 | const char *s, const char *next) | |
3431 | { | |
3432 | const char *encodingName = NULL; | |
3433 | const XML_Char *storedEncName = NULL; | |
3434 | const ENCODING *newEncoding = NULL; | |
3435 | const char *version = NULL; | |
3436 | const char *versionend; | |
3437 | const XML_Char *storedversion = NULL; | |
3438 | int standalone = -1; | |
3439 | if (!(ns | |
3440 | ? XmlParseXmlDeclNS | |
3441 | : XmlParseXmlDecl)(isGeneralTextEntity, | |
3442 | encoding, | |
3443 | s, | |
3444 | next, | |
3445 | &eventPtr, | |
3446 | &version, | |
3447 | &versionend, | |
3448 | &encodingName, | |
3449 | &newEncoding, | |
3450 | &standalone)) { | |
3451 | if (isGeneralTextEntity) | |
3452 | return XML_ERROR_TEXT_DECL; | |
3453 | else | |
3454 | return XML_ERROR_XML_DECL; | |
3455 | } | |
3456 | if (!isGeneralTextEntity && standalone == 1) { | |
3457 | _dtd->standalone = XML_TRUE; | |
3458 | #ifdef XML_DTD | |
3459 | if (paramEntityParsing == XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE) | |
3460 | paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER; | |
3461 | #endif /* XML_DTD */ | |
3462 | } | |
3463 | if (xmlDeclHandler) { | |
3464 | if (encodingName != NULL) { | |
3465 | storedEncName = poolStoreString(&temp2Pool, | |
3466 | encoding, | |
3467 | encodingName, | |
3468 | encodingName | |
3469 | + XmlNameLength(encoding, encodingName)); | |
3470 | if (!storedEncName) | |
3471 | return XML_ERROR_NO_MEMORY; | |
3472 | poolFinish(&temp2Pool); | |
3473 | } | |
3474 | if (version) { | |
3475 | storedversion = poolStoreString(&temp2Pool, | |
3476 | encoding, | |
3477 | version, | |
3478 | versionend - encoding->minBytesPerChar); | |
3479 | if (!storedversion) | |
3480 | return XML_ERROR_NO_MEMORY; | |
3481 | } | |
3482 | xmlDeclHandler(handlerArg, storedversion, storedEncName, standalone); | |
3483 | } | |
3484 | else if (defaultHandler) | |
3485 | reportDefault(parser, encoding, s, next); | |
3486 | if (protocolEncodingName == NULL) { | |
3487 | if (newEncoding) { | |
3488 | if (newEncoding->minBytesPerChar != encoding->minBytesPerChar) { | |
3489 | eventPtr = encodingName; | |
3490 | return XML_ERROR_INCORRECT_ENCODING; | |
3491 | } | |
3492 | encoding = newEncoding; | |
3493 | } | |
3494 | else if (encodingName) { | |
3495 | enum XML_Error result; | |
3496 | if (!storedEncName) { | |
3497 | storedEncName = poolStoreString( | |
3498 | &temp2Pool, encoding, encodingName, | |
3499 | encodingName + XmlNameLength(encoding, encodingName)); | |
3500 | if (!storedEncName) | |
3501 | return XML_ERROR_NO_MEMORY; | |
3502 | } | |
3503 | result = handleUnknownEncoding(parser, storedEncName); | |
3504 | poolClear(&temp2Pool); | |
3505 | if (result == XML_ERROR_UNKNOWN_ENCODING) | |
3506 | eventPtr = encodingName; | |
3507 | return result; | |
3508 | } | |
3509 | } | |
3510 | ||
3511 | if (storedEncName || storedversion) | |
3512 | poolClear(&temp2Pool); | |
3513 | ||
3514 | return XML_ERROR_NONE; | |
3515 | } | |
3516 | ||
3517 | static enum XML_Error | |
3518 | handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName) | |
3519 | { | |
3520 | if (unknownEncodingHandler) { | |
3521 | XML_Encoding info; | |
3522 | int i; | |
3523 | for (i = 0; i < 256; i++) | |
3524 | info.map[i] = -1; | |
3525 | info.convert = NULL; | |
3526 | info.data = NULL; | |
3527 | info.release = NULL; | |
3528 | if (unknownEncodingHandler(unknownEncodingHandlerData, encodingName, | |
3529 | &info)) { | |
3530 | ENCODING *enc; | |
3531 | unknownEncodingMem = MALLOC(XmlSizeOfUnknownEncoding()); | |
3532 | if (!unknownEncodingMem) { | |
3533 | if (info.release) | |
3534 | info.release(info.data); | |
3535 | return XML_ERROR_NO_MEMORY; | |
3536 | } | |
3537 | enc = (ns | |
3538 | ? XmlInitUnknownEncodingNS | |
3539 | : XmlInitUnknownEncoding)(unknownEncodingMem, | |
3540 | info.map, | |
3541 | info.convert, | |
3542 | info.data); | |
3543 | if (enc) { | |
3544 | unknownEncodingData = info.data; | |
3545 | unknownEncodingRelease = info.release; | |
3546 | encoding = enc; | |
3547 | return XML_ERROR_NONE; | |
3548 | } | |
3549 | } | |
3550 | if (info.release != NULL) | |
3551 | info.release(info.data); | |
3552 | } | |
3553 | return XML_ERROR_UNKNOWN_ENCODING; | |
3554 | } | |
3555 | ||
3556 | static enum XML_Error PTRCALL | |
3557 | prologInitProcessor(XML_Parser parser, | |
3558 | const char *s, | |
3559 | const char *end, | |
3560 | const char **nextPtr) | |
3561 | { | |
3562 | enum XML_Error result = initializeEncoding(parser); | |
3563 | if (result != XML_ERROR_NONE) | |
3564 | return result; | |
3565 | processor = prologProcessor; | |
3566 | return prologProcessor(parser, s, end, nextPtr); | |
3567 | } | |
3568 | ||
3569 | #ifdef XML_DTD | |
3570 | ||
3571 | static enum XML_Error PTRCALL | |
3572 | externalParEntInitProcessor(XML_Parser parser, | |
3573 | const char *s, | |
3574 | const char *end, | |
3575 | const char **nextPtr) | |
3576 | { | |
3577 | enum XML_Error result = initializeEncoding(parser); | |
3578 | if (result != XML_ERROR_NONE) | |
3579 | return result; | |
3580 | ||
3581 | /* we know now that XML_Parse(Buffer) has been called, | |
3582 | so we consider the external parameter entity read */ | |
3583 | _dtd->paramEntityRead = XML_TRUE; | |
3584 | ||
3585 | if (prologState.inEntityValue) { | |
3586 | processor = entityValueInitProcessor; | |
3587 | return entityValueInitProcessor(parser, s, end, nextPtr); | |
3588 | } | |
3589 | else { | |
3590 | processor = externalParEntProcessor; | |
3591 | return externalParEntProcessor(parser, s, end, nextPtr); | |
3592 | } | |
3593 | } | |
3594 | ||
3595 | static enum XML_Error PTRCALL | |
3596 | entityValueInitProcessor(XML_Parser parser, | |
3597 | const char *s, | |
3598 | const char *end, | |
3599 | const char **nextPtr) | |
3600 | { | |
3601 | int tok; | |
3602 | const char *start = s; | |
3603 | const char *next = start; | |
3604 | eventPtr = start; | |
3605 | ||
3606 | for (;;) { | |
3607 | tok = XmlPrologTok(encoding, start, end, &next); | |
3608 | eventEndPtr = next; | |
3609 | if (tok <= 0) { | |
3610 | if (!ps_finalBuffer && tok != XML_TOK_INVALID) { | |
3611 | *nextPtr = s; | |
3612 | return XML_ERROR_NONE; | |
3613 | } | |
3614 | switch (tok) { | |
3615 | case XML_TOK_INVALID: | |
3616 | return XML_ERROR_INVALID_TOKEN; | |
3617 | case XML_TOK_PARTIAL: | |
3618 | return XML_ERROR_UNCLOSED_TOKEN; | |
3619 | case XML_TOK_PARTIAL_CHAR: | |
3620 | return XML_ERROR_PARTIAL_CHAR; | |
3621 | case XML_TOK_NONE: /* start == end */ | |
3622 | default: | |
3623 | break; | |
3624 | } | |
3625 | /* found end of entity value - can store it now */ | |
3626 | return storeEntityValue(parser, encoding, s, end); | |
3627 | } | |
3628 | else if (tok == XML_TOK_XML_DECL) { | |
3629 | enum XML_Error result; | |
3630 | result = processXmlDecl(parser, 0, start, next); | |
3631 | if (result != XML_ERROR_NONE) | |
3632 | return result; | |
3633 | switch (ps_parsing) { | |
3634 | case XML_SUSPENDED: | |
3635 | *nextPtr = next; | |
3636 | return XML_ERROR_NONE; | |
3637 | case XML_FINISHED: | |
3638 | return XML_ERROR_ABORTED; | |
3639 | default: | |
3640 | *nextPtr = next; | |
3641 | } | |
3642 | /* stop scanning for text declaration - we found one */ | |
3643 | processor = entityValueProcessor; | |
3644 | return entityValueProcessor(parser, next, end, nextPtr); | |
3645 | } | |
3646 | /* If we are at the end of the buffer, this would cause XmlPrologTok to | |
3647 | return XML_TOK_NONE on the next call, which would then cause the | |
3648 | function to exit with *nextPtr set to s - that is what we want for other | |
3649 | tokens, but not for the BOM - we would rather like to skip it; | |
3650 | then, when this routine is entered the next time, XmlPrologTok will | |
3651 | return XML_TOK_INVALID, since the BOM is still in the buffer | |
3652 | */ | |
3653 | else if (tok == XML_TOK_BOM && next == end && !ps_finalBuffer) { | |
3654 | *nextPtr = next; | |
3655 | return XML_ERROR_NONE; | |
3656 | } | |
3657 | start = next; | |
3658 | eventPtr = start; | |
3659 | } | |
3660 | } | |
3661 | ||
3662 | static enum XML_Error PTRCALL | |
3663 | externalParEntProcessor(XML_Parser parser, | |
3664 | const char *s, | |
3665 | const char *end, | |
3666 | const char **nextPtr) | |
3667 | { | |
3668 | const char *next = s; | |
3669 | int tok; | |
3670 | ||
3671 | tok = XmlPrologTok(encoding, s, end, &next); | |
3672 | if (tok <= 0) { | |
3673 | if (!ps_finalBuffer && tok != XML_TOK_INVALID) { | |
3674 | *nextPtr = s; | |
3675 | return XML_ERROR_NONE; | |
3676 | } | |
3677 | switch (tok) { | |
3678 | case XML_TOK_INVALID: | |
3679 | return XML_ERROR_INVALID_TOKEN; | |
3680 | case XML_TOK_PARTIAL: | |
3681 | return XML_ERROR_UNCLOSED_TOKEN; | |
3682 | case XML_TOK_PARTIAL_CHAR: | |
3683 | return XML_ERROR_PARTIAL_CHAR; | |
3684 | case XML_TOK_NONE: /* start == end */ | |
3685 | default: | |
3686 | break; | |
3687 | } | |
3688 | } | |
3689 | /* This would cause the next stage, i.e. doProlog to be passed XML_TOK_BOM. | |
3690 | However, when parsing an external subset, doProlog will not accept a BOM | |
3691 | as valid, and report a syntax error, so we have to skip the BOM | |
3692 | */ | |
3693 | else if (tok == XML_TOK_BOM) { | |
3694 | s = next; | |
3695 | tok = XmlPrologTok(encoding, s, end, &next); | |
3696 | } | |
3697 | ||
3698 | processor = prologProcessor; | |
3699 | return doProlog(parser, encoding, s, end, tok, next, | |
3700 | nextPtr, (XML_Bool)!ps_finalBuffer); | |
3701 | } | |
3702 | ||
3703 | static enum XML_Error PTRCALL | |
3704 | entityValueProcessor(XML_Parser parser, | |
3705 | const char *s, | |
3706 | const char *end, | |
3707 | const char **nextPtr) | |
3708 | { | |
3709 | const char *start = s; | |
3710 | const char *next = s; | |
3711 | const ENCODING *enc = encoding; | |
3712 | int tok; | |
3713 | ||
3714 | for (;;) { | |
3715 | tok = XmlPrologTok(enc, start, end, &next); | |
3716 | if (tok <= 0) { | |
3717 | if (!ps_finalBuffer && tok != XML_TOK_INVALID) { | |
3718 | *nextPtr = s; | |
3719 | return XML_ERROR_NONE; | |
3720 | } | |
3721 | switch (tok) { | |
3722 | case XML_TOK_INVALID: | |
3723 | return XML_ERROR_INVALID_TOKEN; | |
3724 | case XML_TOK_PARTIAL: | |
3725 | return XML_ERROR_UNCLOSED_TOKEN; | |
3726 | case XML_TOK_PARTIAL_CHAR: | |
3727 | return XML_ERROR_PARTIAL_CHAR; | |
3728 | case XML_TOK_NONE: /* start == end */ | |
3729 | default: | |
3730 | break; | |
3731 | } | |
3732 | /* found end of entity value - can store it now */ | |
3733 | return storeEntityValue(parser, enc, s, end); | |
3734 | } | |
3735 | start = next; | |
3736 | } | |
3737 | } | |
3738 | ||
3739 | #endif /* XML_DTD */ | |
3740 | ||
3741 | static enum XML_Error PTRCALL | |
3742 | prologProcessor(XML_Parser parser, | |
3743 | const char *s, | |
3744 | const char *end, | |
3745 | const char **nextPtr) | |
3746 | { | |
3747 | const char *next = s; | |
3748 | int tok = XmlPrologTok(encoding, s, end, &next); | |
3749 | return doProlog(parser, encoding, s, end, tok, next, | |
3750 | nextPtr, (XML_Bool)!ps_finalBuffer); | |
3751 | } | |
3752 | ||
3753 | static enum XML_Error | |
3754 | doProlog(XML_Parser parser, | |
3755 | const ENCODING *enc, | |
3756 | const char *s, | |
3757 | const char *end, | |
3758 | int tok, | |
3759 | const char *next, | |
3760 | const char **nextPtr, | |
3761 | XML_Bool haveMore) | |
3762 | { | |
3763 | #ifdef XML_DTD | |
3764 | static const XML_Char externalSubsetName[] = { ASCII_HASH , '\0' }; | |
3765 | #endif /* XML_DTD */ | |
3766 | static const XML_Char atypeCDATA[] = | |
3767 | { ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' }; | |
3768 | static const XML_Char atypeID[] = { ASCII_I, ASCII_D, '\0' }; | |
3769 | static const XML_Char atypeIDREF[] = | |
3770 | { ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, '\0' }; | |
3771 | static const XML_Char atypeIDREFS[] = | |
3772 | { ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, ASCII_S, '\0' }; | |
3773 | static const XML_Char atypeENTITY[] = | |
3774 | { ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_Y, '\0' }; | |
3775 | static const XML_Char atypeENTITIES[] = { ASCII_E, ASCII_N, | |
3776 | ASCII_T, ASCII_I, ASCII_T, ASCII_I, ASCII_E, ASCII_S, '\0' }; | |
3777 | static const XML_Char atypeNMTOKEN[] = { | |
3778 | ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, '\0' }; | |
3779 | static const XML_Char atypeNMTOKENS[] = { ASCII_N, ASCII_M, ASCII_T, | |
3780 | ASCII_O, ASCII_K, ASCII_E, ASCII_N, ASCII_S, '\0' }; | |
3781 | static const XML_Char notationPrefix[] = { ASCII_N, ASCII_O, ASCII_T, | |
3782 | ASCII_A, ASCII_T, ASCII_I, ASCII_O, ASCII_N, ASCII_LPAREN, '\0' }; | |
3783 | static const XML_Char enumValueSep[] = { ASCII_PIPE, '\0' }; | |
3784 | static const XML_Char enumValueStart[] = { ASCII_LPAREN, '\0' }; | |
3785 | ||
3786 | /* save one level of indirection */ | |
3787 | DTD * const dtd = _dtd; | |
3788 | ||
3789 | const char **eventPP; | |
3790 | const char **eventEndPP; | |
3791 | enum XML_Content_Quant quant; | |
3792 | ||
3793 | if (enc == encoding) { | |
3794 | eventPP = &eventPtr; | |
3795 | eventEndPP = &eventEndPtr; | |
3796 | } | |
3797 | else { | |
3798 | eventPP = &(openInternalEntities->internalEventPtr); | |
3799 | eventEndPP = &(openInternalEntities->internalEventEndPtr); | |
3800 | } | |
3801 | ||
3802 | for (;;) { | |
3803 | int role; | |
3804 | XML_Bool handleDefault = XML_TRUE; | |
3805 | *eventPP = s; | |
3806 | *eventEndPP = next; | |
3807 | if (tok <= 0) { | |
3808 | if (haveMore && tok != XML_TOK_INVALID) { | |
3809 | *nextPtr = s; | |
3810 | return XML_ERROR_NONE; | |
3811 | } | |
3812 | switch (tok) { | |
3813 | case XML_TOK_INVALID: | |
3814 | *eventPP = next; | |
3815 | return XML_ERROR_INVALID_TOKEN; | |
3816 | case XML_TOK_PARTIAL: | |
3817 | return XML_ERROR_UNCLOSED_TOKEN; | |
3818 | case XML_TOK_PARTIAL_CHAR: | |
3819 | return XML_ERROR_PARTIAL_CHAR; | |
3820 | case -XML_TOK_PROLOG_S: | |
3821 | tok = -tok; | |
3822 | break; | |
3823 | case XML_TOK_NONE: | |
3824 | #ifdef XML_DTD | |
3825 | /* for internal PE NOT referenced between declarations */ | |
3826 | if (enc != encoding && !openInternalEntities->betweenDecl) { | |
3827 | *nextPtr = s; | |
3828 | return XML_ERROR_NONE; | |
3829 | } | |
3830 | /* WFC: PE Between Declarations - must check that PE contains | |
3831 | complete markup, not only for external PEs, but also for | |
3832 | internal PEs if the reference occurs between declarations. | |
3833 | */ | |
3834 | if (isParamEntity || enc != encoding) { | |
3835 | if (XmlTokenRole(&prologState, XML_TOK_NONE, end, end, enc) | |
3836 | == XML_ROLE_ERROR) | |
3837 | return XML_ERROR_INCOMPLETE_PE; | |
3838 | *nextPtr = s; | |
3839 | return XML_ERROR_NONE; | |
3840 | } | |
3841 | #endif /* XML_DTD */ | |
3842 | return XML_ERROR_NO_ELEMENTS; | |
3843 | default: | |
3844 | tok = -tok; | |
3845 | next = end; | |
3846 | break; | |
3847 | } | |
3848 | } | |
3849 | role = XmlTokenRole(&prologState, tok, s, next, enc); | |
3850 | switch (role) { | |
3851 | case XML_ROLE_XML_DECL: | |
3852 | { | |
3853 | enum XML_Error result = processXmlDecl(parser, 0, s, next); | |
3854 | if (result != XML_ERROR_NONE) | |
3855 | return result; | |
3856 | enc = encoding; | |
3857 | handleDefault = XML_FALSE; | |
3858 | } | |
3859 | break; | |
3860 | case XML_ROLE_DOCTYPE_NAME: | |
3861 | if (startDoctypeDeclHandler) { | |
3862 | doctypeName = poolStoreString(&tempPool, enc, s, next); | |
3863 | if (!doctypeName) | |
3864 | return XML_ERROR_NO_MEMORY; | |
3865 | poolFinish(&tempPool); | |
3866 | doctypePubid = NULL; | |
3867 | handleDefault = XML_FALSE; | |
3868 | } | |
3869 | doctypeSysid = NULL; /* always initialize to NULL */ | |
3870 | break; | |
3871 | case XML_ROLE_DOCTYPE_INTERNAL_SUBSET: | |
3872 | if (startDoctypeDeclHandler) { | |
3873 | startDoctypeDeclHandler(handlerArg, doctypeName, doctypeSysid, | |
3874 | doctypePubid, 1); | |
3875 | doctypeName = NULL; | |
3876 | poolClear(&tempPool); | |
3877 | handleDefault = XML_FALSE; | |
3878 | } | |
3879 | break; | |
3880 | #ifdef XML_DTD | |
3881 | case XML_ROLE_TEXT_DECL: | |
3882 | { | |
3883 | enum XML_Error result = processXmlDecl(parser, 1, s, next); | |
3884 | if (result != XML_ERROR_NONE) | |
3885 | return result; | |
3886 | enc = encoding; | |
3887 | handleDefault = XML_FALSE; | |
3888 | } | |
3889 | break; | |
3890 | #endif /* XML_DTD */ | |
3891 | case XML_ROLE_DOCTYPE_PUBLIC_ID: | |
3892 | #ifdef XML_DTD | |
3893 | useForeignDTD = XML_FALSE; | |
3894 | declEntity = (ENTITY *)lookup(parser, | |
3895 | &dtd->paramEntities, | |
3896 | externalSubsetName, | |
3897 | sizeof(ENTITY)); | |
3898 | if (!declEntity) | |
3899 | return XML_ERROR_NO_MEMORY; | |
3900 | #endif /* XML_DTD */ | |
3901 | dtd->hasParamEntityRefs = XML_TRUE; | |
3902 | if (startDoctypeDeclHandler) { | |
3903 | XML_Char *pubId; | |
3904 | if (!XmlIsPublicId(enc, s, next, eventPP)) | |
3905 | return XML_ERROR_PUBLICID; | |
3906 | pubId = poolStoreString(&tempPool, enc, | |
3907 | s + enc->minBytesPerChar, | |
3908 | next - enc->minBytesPerChar); | |
3909 | if (!pubId) | |
3910 | return XML_ERROR_NO_MEMORY; | |
3911 | normalizePublicId(pubId); | |
3912 | poolFinish(&tempPool); | |
3913 | doctypePubid = pubId; | |
3914 | handleDefault = XML_FALSE; | |
3915 | goto alreadyChecked; | |
3916 | } | |
3917 | /* fall through */ | |
3918 | case XML_ROLE_ENTITY_PUBLIC_ID: | |
3919 | if (!XmlIsPublicId(enc, s, next, eventPP)) | |
3920 | return XML_ERROR_PUBLICID; | |
3921 | alreadyChecked: | |
3922 | if (dtd->keepProcessing && declEntity) { | |
3923 | XML_Char *tem = poolStoreString(&dtd->pool, | |
3924 | enc, | |
3925 | s + enc->minBytesPerChar, | |
3926 | next - enc->minBytesPerChar); | |
3927 | if (!tem) | |
3928 | return XML_ERROR_NO_MEMORY; | |
3929 | normalizePublicId(tem); | |
3930 | declEntity->publicId = tem; | |
3931 | poolFinish(&dtd->pool); | |
3932 | if (entityDeclHandler) | |
3933 | handleDefault = XML_FALSE; | |
3934 | } | |
3935 | break; | |
3936 | case XML_ROLE_DOCTYPE_CLOSE: | |
3937 | if (doctypeName) { | |
3938 | startDoctypeDeclHandler(handlerArg, doctypeName, | |
3939 | doctypeSysid, doctypePubid, 0); | |
3940 | poolClear(&tempPool); | |
3941 | handleDefault = XML_FALSE; | |
3942 | } | |
3943 | /* doctypeSysid will be non-NULL in the case of a previous | |
3944 | XML_ROLE_DOCTYPE_SYSTEM_ID, even if startDoctypeDeclHandler | |
3945 | was not set, indicating an external subset | |
3946 | */ | |
3947 | #ifdef XML_DTD | |
3948 | if (doctypeSysid || useForeignDTD) { | |
3949 | XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs; | |
3950 | dtd->hasParamEntityRefs = XML_TRUE; | |
3951 | if (paramEntityParsing && externalEntityRefHandler) { | |
3952 | ENTITY *entity = (ENTITY *)lookup(parser, | |
3953 | &dtd->paramEntities, | |
3954 | externalSubsetName, | |
3955 | sizeof(ENTITY)); | |
3956 | if (!entity) | |
3957 | return XML_ERROR_NO_MEMORY; | |
3958 | if (useForeignDTD) | |
3959 | entity->base = curBase; | |
3960 | dtd->paramEntityRead = XML_FALSE; | |
3961 | if (!externalEntityRefHandler(externalEntityRefHandlerArg, | |
3962 | 0, | |
3963 | entity->base, | |
3964 | entity->systemId, | |
3965 | entity->publicId)) | |
3966 | return XML_ERROR_EXTERNAL_ENTITY_HANDLING; | |
3967 | if (dtd->paramEntityRead) { | |
3968 | if (!dtd->standalone && | |
3969 | notStandaloneHandler && | |
3970 | !notStandaloneHandler(handlerArg)) | |
3971 | return XML_ERROR_NOT_STANDALONE; | |
3972 | } | |
3973 | /* if we didn't read the foreign DTD then this means that there | |
3974 | is no external subset and we must reset dtd->hasParamEntityRefs | |
3975 | */ | |
3976 | else if (!doctypeSysid) | |
3977 | dtd->hasParamEntityRefs = hadParamEntityRefs; | |
3978 | /* end of DTD - no need to update dtd->keepProcessing */ | |
3979 | } | |
3980 | useForeignDTD = XML_FALSE; | |
3981 | } | |
3982 | #endif /* XML_DTD */ | |
3983 | if (endDoctypeDeclHandler) { | |
3984 | endDoctypeDeclHandler(handlerArg); | |
3985 | handleDefault = XML_FALSE; | |
3986 | } | |
3987 | break; | |
3988 | case XML_ROLE_INSTANCE_START: | |
3989 | #ifdef XML_DTD | |
3990 | /* if there is no DOCTYPE declaration then now is the | |
3991 | last chance to read the foreign DTD | |
3992 | */ | |
3993 | if (useForeignDTD) { | |
3994 | XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs; | |
3995 | dtd->hasParamEntityRefs = XML_TRUE; | |
3996 | if (paramEntityParsing && externalEntityRefHandler) { | |
3997 | ENTITY *entity = (ENTITY *)lookup(parser, &dtd->paramEntities, | |
3998 | externalSubsetName, | |
3999 | sizeof(ENTITY)); | |
4000 | if (!entity) | |
4001 | return XML_ERROR_NO_MEMORY; | |
4002 | entity->base = curBase; | |
4003 | dtd->paramEntityRead = XML_FALSE; | |
4004 | if (!externalEntityRefHandler(externalEntityRefHandlerArg, | |
4005 | 0, | |
4006 | entity->base, | |
4007 | entity->systemId, | |
4008 | entity->publicId)) | |
4009 | return XML_ERROR_EXTERNAL_ENTITY_HANDLING; | |
4010 | if (dtd->paramEntityRead) { | |
4011 | if (!dtd->standalone && | |
4012 | notStandaloneHandler && | |
4013 | !notStandaloneHandler(handlerArg)) | |
4014 | return XML_ERROR_NOT_STANDALONE; | |
4015 | } | |
4016 | /* if we didn't read the foreign DTD then this means that there | |
4017 | is no external subset and we must reset dtd->hasParamEntityRefs | |
4018 | */ | |
4019 | else | |
4020 | dtd->hasParamEntityRefs = hadParamEntityRefs; | |
4021 | /* end of DTD - no need to update dtd->keepProcessing */ | |
4022 | } | |
4023 | } | |
4024 | #endif /* XML_DTD */ | |
4025 | processor = contentProcessor; | |
4026 | return contentProcessor(parser, s, end, nextPtr); | |
4027 | case XML_ROLE_ATTLIST_ELEMENT_NAME: | |
4028 | declElementType = getElementType(parser, enc, s, next); | |
4029 | if (!declElementType) | |
4030 | return XML_ERROR_NO_MEMORY; | |
4031 | goto checkAttListDeclHandler; | |
4032 | case XML_ROLE_ATTRIBUTE_NAME: | |
4033 | declAttributeId = getAttributeId(parser, enc, s, next); | |
4034 | if (!declAttributeId) | |
4035 | return XML_ERROR_NO_MEMORY; | |
4036 | declAttributeIsCdata = XML_FALSE; | |
4037 | declAttributeType = NULL; | |
4038 | declAttributeIsId = XML_FALSE; | |
4039 | goto checkAttListDeclHandler; | |
4040 | case XML_ROLE_ATTRIBUTE_TYPE_CDATA: | |
4041 | declAttributeIsCdata = XML_TRUE; | |
4042 | declAttributeType = atypeCDATA; | |
4043 | goto checkAttListDeclHandler; | |
4044 | case XML_ROLE_ATTRIBUTE_TYPE_ID: | |
4045 | declAttributeIsId = XML_TRUE; | |
4046 | declAttributeType = atypeID; | |
4047 | goto checkAttListDeclHandler; | |
4048 | case XML_ROLE_ATTRIBUTE_TYPE_IDREF: | |
4049 | declAttributeType = atypeIDREF; | |
4050 | goto checkAttListDeclHandler; | |
4051 | case XML_ROLE_ATTRIBUTE_TYPE_IDREFS: | |
4052 | declAttributeType = atypeIDREFS; | |
4053 | goto checkAttListDeclHandler; | |
4054 | case XML_ROLE_ATTRIBUTE_TYPE_ENTITY: | |
4055 | declAttributeType = atypeENTITY; | |
4056 | goto checkAttListDeclHandler; | |
4057 | case XML_ROLE_ATTRIBUTE_TYPE_ENTITIES: | |
4058 | declAttributeType = atypeENTITIES; | |
4059 | goto checkAttListDeclHandler; | |
4060 | case XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN: | |
4061 | declAttributeType = atypeNMTOKEN; | |
4062 | goto checkAttListDeclHandler; | |
4063 | case XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS: | |
4064 | declAttributeType = atypeNMTOKENS; | |
4065 | checkAttListDeclHandler: | |
4066 | if (dtd->keepProcessing && attlistDeclHandler) | |
4067 | handleDefault = XML_FALSE; | |
4068 | break; | |
4069 | case XML_ROLE_ATTRIBUTE_ENUM_VALUE: | |
4070 | case XML_ROLE_ATTRIBUTE_NOTATION_VALUE: | |
4071 | if (dtd->keepProcessing && attlistDeclHandler) { | |
4072 | const XML_Char *prefix; | |
4073 | if (declAttributeType) { | |
4074 | prefix = enumValueSep; | |
4075 | } | |
4076 | else { | |
4077 | prefix = (role == XML_ROLE_ATTRIBUTE_NOTATION_VALUE | |
4078 | ? notationPrefix | |
4079 | : enumValueStart); | |
4080 | } | |
4081 | if (!poolAppendString(&tempPool, prefix)) | |
4082 | return XML_ERROR_NO_MEMORY; | |
4083 | if (!poolAppend(&tempPool, enc, s, next)) | |
4084 | return XML_ERROR_NO_MEMORY; | |
4085 | declAttributeType = tempPool.start; | |
4086 | handleDefault = XML_FALSE; | |
4087 | } | |
4088 | break; | |
4089 | case XML_ROLE_IMPLIED_ATTRIBUTE_VALUE: | |
4090 | case XML_ROLE_REQUIRED_ATTRIBUTE_VALUE: | |
4091 | if (dtd->keepProcessing) { | |
4092 | if (!defineAttribute(declElementType, declAttributeId, | |
4093 | declAttributeIsCdata, declAttributeIsId, | |
4094 | 0, parser)) | |
4095 | return XML_ERROR_NO_MEMORY; | |
4096 | if (attlistDeclHandler && declAttributeType) { | |
4097 | if (*declAttributeType == XML_T(ASCII_LPAREN) | |
4098 | || (*declAttributeType == XML_T(ASCII_N) | |
4099 | && declAttributeType[1] == XML_T(ASCII_O))) { | |
4100 | /* Enumerated or Notation type */ | |
4101 | if (!poolAppendChar(&tempPool, XML_T(ASCII_RPAREN)) | |
4102 | || !poolAppendChar(&tempPool, XML_T('\0'))) | |
4103 | return XML_ERROR_NO_MEMORY; | |
4104 | declAttributeType = tempPool.start; | |
4105 | poolFinish(&tempPool); | |
4106 | } | |
4107 | *eventEndPP = s; | |
4108 | attlistDeclHandler(handlerArg, declElementType->name, | |
4109 | declAttributeId->name, declAttributeType, | |
4110 | 0, role == XML_ROLE_REQUIRED_ATTRIBUTE_VALUE); | |
4111 | poolClear(&tempPool); | |
4112 | handleDefault = XML_FALSE; | |
4113 | } | |
4114 | } | |
4115 | break; | |
4116 | case XML_ROLE_DEFAULT_ATTRIBUTE_VALUE: | |
4117 | case XML_ROLE_FIXED_ATTRIBUTE_VALUE: | |
4118 | if (dtd->keepProcessing) { | |
4119 | const XML_Char *attVal; | |
4120 | enum XML_Error result = | |
4121 | storeAttributeValue(parser, enc, declAttributeIsCdata, | |
4122 | s + enc->minBytesPerChar, | |
4123 | next - enc->minBytesPerChar, | |
4124 | &dtd->pool); | |
4125 | if (result) | |
4126 | return result; | |
4127 | attVal = poolStart(&dtd->pool); | |
4128 | poolFinish(&dtd->pool); | |
4129 | /* ID attributes aren't allowed to have a default */ | |
4130 | if (!defineAttribute(declElementType, declAttributeId, | |
4131 | declAttributeIsCdata, XML_FALSE, attVal, parser)) | |
4132 | return XML_ERROR_NO_MEMORY; | |
4133 | if (attlistDeclHandler && declAttributeType) { | |
4134 | if (*declAttributeType == XML_T(ASCII_LPAREN) | |
4135 | || (*declAttributeType == XML_T(ASCII_N) | |
4136 | && declAttributeType[1] == XML_T(ASCII_O))) { | |
4137 | /* Enumerated or Notation type */ | |
4138 | if (!poolAppendChar(&tempPool, XML_T(ASCII_RPAREN)) | |
4139 | || !poolAppendChar(&tempPool, XML_T('\0'))) | |
4140 | return XML_ERROR_NO_MEMORY; | |
4141 | declAttributeType = tempPool.start; | |
4142 | poolFinish(&tempPool); | |
4143 | } | |
4144 | *eventEndPP = s; | |
4145 | attlistDeclHandler(handlerArg, declElementType->name, | |
4146 | declAttributeId->name, declAttributeType, | |
4147 | attVal, | |
4148 | role == XML_ROLE_FIXED_ATTRIBUTE_VALUE); | |
4149 | poolClear(&tempPool); | |
4150 | handleDefault = XML_FALSE; | |
4151 | } | |
4152 | } | |
4153 | break; | |
4154 | case XML_ROLE_ENTITY_VALUE: | |
4155 | if (dtd->keepProcessing) { | |
4156 | enum XML_Error result = storeEntityValue(parser, enc, | |
4157 | s + enc->minBytesPerChar, | |
4158 | next - enc->minBytesPerChar); | |
4159 | if (declEntity) { | |
4160 | declEntity->textPtr = poolStart(&dtd->entityValuePool); | |
4161 | declEntity->textLen = (int)(poolLength(&dtd->entityValuePool)); | |
4162 | poolFinish(&dtd->entityValuePool); | |
4163 | if (entityDeclHandler) { | |
4164 | *eventEndPP = s; | |
4165 | entityDeclHandler(handlerArg, | |
4166 | declEntity->name, | |
4167 | declEntity->is_param, | |
4168 | declEntity->textPtr, | |
4169 | declEntity->textLen, | |
4170 | curBase, 0, 0, 0); | |
4171 | handleDefault = XML_FALSE; | |
4172 | } | |
4173 | } | |
4174 | else | |
4175 | poolDiscard(&dtd->entityValuePool); | |
4176 | if (result != XML_ERROR_NONE) | |
4177 | return result; | |
4178 | } | |
4179 | break; | |
4180 | case XML_ROLE_DOCTYPE_SYSTEM_ID: | |
4181 | #ifdef XML_DTD | |
4182 | useForeignDTD = XML_FALSE; | |
4183 | #endif /* XML_DTD */ | |
4184 | dtd->hasParamEntityRefs = XML_TRUE; | |
4185 | if (startDoctypeDeclHandler) { | |
4186 | doctypeSysid = poolStoreString(&tempPool, enc, | |
4187 | s + enc->minBytesPerChar, | |
4188 | next - enc->minBytesPerChar); | |
4189 | if (doctypeSysid == NULL) | |
4190 | return XML_ERROR_NO_MEMORY; | |
4191 | poolFinish(&tempPool); | |
4192 | handleDefault = XML_FALSE; | |
4193 | } | |
4194 | #ifdef XML_DTD | |
4195 | else | |
4196 | /* use externalSubsetName to make doctypeSysid non-NULL | |
4197 | for the case where no startDoctypeDeclHandler is set */ | |
4198 | doctypeSysid = externalSubsetName; | |
4199 | #endif /* XML_DTD */ | |
4200 | if (!dtd->standalone | |
4201 | #ifdef XML_DTD | |
4202 | && !paramEntityParsing | |
4203 | #endif /* XML_DTD */ | |
4204 | && notStandaloneHandler | |
4205 | && !notStandaloneHandler(handlerArg)) | |
4206 | return XML_ERROR_NOT_STANDALONE; | |
4207 | #ifndef XML_DTD | |
4208 | break; | |
4209 | #else /* XML_DTD */ | |
4210 | if (!declEntity) { | |
4211 | declEntity = (ENTITY *)lookup(parser, | |
4212 | &dtd->paramEntities, | |
4213 | externalSubsetName, | |
4214 | sizeof(ENTITY)); | |
4215 | if (!declEntity) | |
4216 | return XML_ERROR_NO_MEMORY; | |
4217 | declEntity->publicId = NULL; | |
4218 | } | |
4219 | /* fall through */ | |
4220 | #endif /* XML_DTD */ | |
4221 | case XML_ROLE_ENTITY_SYSTEM_ID: | |
4222 | if (dtd->keepProcessing && declEntity) { | |
4223 | declEntity->systemId = poolStoreString(&dtd->pool, enc, | |
4224 | s + enc->minBytesPerChar, | |
4225 | next - enc->minBytesPerChar); | |
4226 | if (!declEntity->systemId) | |
4227 | return XML_ERROR_NO_MEMORY; | |
4228 | declEntity->base = curBase; | |
4229 | poolFinish(&dtd->pool); | |
4230 | if (entityDeclHandler) | |
4231 | handleDefault = XML_FALSE; | |
4232 | } | |
4233 | break; | |
4234 | case XML_ROLE_ENTITY_COMPLETE: | |
4235 | if (dtd->keepProcessing && declEntity && entityDeclHandler) { | |
4236 | *eventEndPP = s; | |
4237 | entityDeclHandler(handlerArg, | |
4238 | declEntity->name, | |
4239 | declEntity->is_param, | |
4240 | 0,0, | |
4241 | declEntity->base, | |
4242 | declEntity->systemId, | |
4243 | declEntity->publicId, | |
4244 | 0); | |
4245 | handleDefault = XML_FALSE; | |
4246 | } | |
4247 | break; | |
4248 | case XML_ROLE_ENTITY_NOTATION_NAME: | |
4249 | if (dtd->keepProcessing && declEntity) { | |
4250 | declEntity->notation = poolStoreString(&dtd->pool, enc, s, next); | |
4251 | if (!declEntity->notation) | |
4252 | return XML_ERROR_NO_MEMORY; | |
4253 | poolFinish(&dtd->pool); | |
4254 | if (unparsedEntityDeclHandler) { | |
4255 | *eventEndPP = s; | |
4256 | unparsedEntityDeclHandler(handlerArg, | |
4257 | declEntity->name, | |
4258 | declEntity->base, | |
4259 | declEntity->systemId, | |
4260 | declEntity->publicId, | |
4261 | declEntity->notation); | |
4262 | handleDefault = XML_FALSE; | |
4263 | } | |
4264 | else if (entityDeclHandler) { | |
4265 | *eventEndPP = s; | |
4266 | entityDeclHandler(handlerArg, | |
4267 | declEntity->name, | |
4268 | 0,0,0, | |
4269 | declEntity->base, | |
4270 | declEntity->systemId, | |
4271 | declEntity->publicId, | |
4272 | declEntity->notation); | |
4273 | handleDefault = XML_FALSE; | |
4274 | } | |
4275 | } | |
4276 | break; | |
4277 | case XML_ROLE_GENERAL_ENTITY_NAME: | |
4278 | { | |
4279 | if (XmlPredefinedEntityName(enc, s, next)) { | |
4280 | declEntity = NULL; | |
4281 | break; | |
4282 | } | |
4283 | if (dtd->keepProcessing) { | |
4284 | const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next); | |
4285 | if (!name) | |
4286 | return XML_ERROR_NO_MEMORY; | |
4287 | declEntity = (ENTITY *)lookup(parser, &dtd->generalEntities, name, | |
4288 | sizeof(ENTITY)); | |
4289 | if (!declEntity) | |
4290 | return XML_ERROR_NO_MEMORY; | |
4291 | if (declEntity->name != name) { | |
4292 | poolDiscard(&dtd->pool); | |
4293 | declEntity = NULL; | |
4294 | } | |
4295 | else { | |
4296 | poolFinish(&dtd->pool); | |
4297 | declEntity->publicId = NULL; | |
4298 | declEntity->is_param = XML_FALSE; | |
4299 | /* if we have a parent parser or are reading an internal parameter | |
4300 | entity, then the entity declaration is not considered "internal" | |
4301 | */ | |
4302 | declEntity->is_internal = !(parentParser || openInternalEntities); | |
4303 | if (entityDeclHandler) | |
4304 | handleDefault = XML_FALSE; | |
4305 | } | |
4306 | } | |
4307 | else { | |
4308 | poolDiscard(&dtd->pool); | |
4309 | declEntity = NULL; | |
4310 | } | |
4311 | } | |
4312 | break; | |
4313 | case XML_ROLE_PARAM_ENTITY_NAME: | |
4314 | #ifdef XML_DTD | |
4315 | if (dtd->keepProcessing) { | |
4316 | const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next); | |
4317 | if (!name) | |
4318 | return XML_ERROR_NO_MEMORY; | |
4319 | declEntity = (ENTITY *)lookup(parser, &dtd->paramEntities, | |
4320 | name, sizeof(ENTITY)); | |
4321 | if (!declEntity) | |
4322 | return XML_ERROR_NO_MEMORY; | |
4323 | if (declEntity->name != name) { | |
4324 | poolDiscard(&dtd->pool); | |
4325 | declEntity = NULL; | |
4326 | } | |
4327 | else { | |
4328 | poolFinish(&dtd->pool); | |
4329 | declEntity->publicId = NULL; | |
4330 | declEntity->is_param = XML_TRUE; | |
4331 | /* if we have a parent parser or are reading an internal parameter | |
4332 | entity, then the entity declaration is not considered "internal" | |
4333 | */ | |
4334 | declEntity->is_internal = !(parentParser || openInternalEntities); | |
4335 | if (entityDeclHandler) | |
4336 | handleDefault = XML_FALSE; | |
4337 | } | |
4338 | } | |
4339 | else { | |
4340 | poolDiscard(&dtd->pool); | |
4341 | declEntity = NULL; | |
4342 | } | |
4343 | #else /* not XML_DTD */ | |
4344 | declEntity = NULL; | |
4345 | #endif /* XML_DTD */ | |
4346 | break; | |
4347 | case XML_ROLE_NOTATION_NAME: | |
4348 | declNotationPublicId = NULL; | |
4349 | declNotationName = NULL; | |
4350 | if (notationDeclHandler) { | |
4351 | declNotationName = poolStoreString(&tempPool, enc, s, next); | |
4352 | if (!declNotationName) | |
4353 | return XML_ERROR_NO_MEMORY; | |
4354 | poolFinish(&tempPool); | |
4355 | handleDefault = XML_FALSE; | |
4356 | } | |
4357 | break; | |
4358 | case XML_ROLE_NOTATION_PUBLIC_ID: | |
4359 | if (!XmlIsPublicId(enc, s, next, eventPP)) | |
4360 | return XML_ERROR_PUBLICID; | |
4361 | if (declNotationName) { /* means notationDeclHandler != NULL */ | |
4362 | XML_Char *tem = poolStoreString(&tempPool, | |
4363 | enc, | |
4364 | s + enc->minBytesPerChar, | |
4365 | next - enc->minBytesPerChar); | |
4366 | if (!tem) | |
4367 | return XML_ERROR_NO_MEMORY; | |
4368 | normalizePublicId(tem); | |
4369 | declNotationPublicId = tem; | |
4370 | poolFinish(&tempPool); | |
4371 | handleDefault = XML_FALSE; | |
4372 | } | |
4373 | break; | |
4374 | case XML_ROLE_NOTATION_SYSTEM_ID: | |
4375 | if (declNotationName && notationDeclHandler) { | |
4376 | const XML_Char *systemId | |
4377 | = poolStoreString(&tempPool, enc, | |
4378 | s + enc->minBytesPerChar, | |
4379 | next - enc->minBytesPerChar); | |
4380 | if (!systemId) | |
4381 | return XML_ERROR_NO_MEMORY; | |
4382 | *eventEndPP = s; | |
4383 | notationDeclHandler(handlerArg, | |
4384 | declNotationName, | |
4385 | curBase, | |
4386 | systemId, | |
4387 | declNotationPublicId); | |
4388 | handleDefault = XML_FALSE; | |
4389 | } | |
4390 | poolClear(&tempPool); | |
4391 | break; | |
4392 | case XML_ROLE_NOTATION_NO_SYSTEM_ID: | |
4393 | if (declNotationPublicId && notationDeclHandler) { | |
4394 | *eventEndPP = s; | |
4395 | notationDeclHandler(handlerArg, | |
4396 | declNotationName, | |
4397 | curBase, | |
4398 | 0, | |
4399 | declNotationPublicId); | |
4400 | handleDefault = XML_FALSE; | |
4401 | } | |
4402 | poolClear(&tempPool); | |
4403 | break; | |
4404 | case XML_ROLE_ERROR: | |
4405 | switch (tok) { | |
4406 | case XML_TOK_PARAM_ENTITY_REF: | |
4407 | /* PE references in internal subset are | |
4408 | not allowed within declarations. */ | |
4409 | return XML_ERROR_PARAM_ENTITY_REF; | |
4410 | case XML_TOK_XML_DECL: | |
4411 | return XML_ERROR_MISPLACED_XML_PI; | |
4412 | default: | |
4413 | return XML_ERROR_SYNTAX; | |
4414 | } | |
4415 | #ifdef XML_DTD | |
4416 | case XML_ROLE_IGNORE_SECT: | |
4417 | { | |
4418 | enum XML_Error result; | |
4419 | if (defaultHandler) | |
4420 | reportDefault(parser, enc, s, next); | |
4421 | handleDefault = XML_FALSE; | |
4422 | result = doIgnoreSection(parser, enc, &next, end, nextPtr, haveMore); | |
4423 | if (result != XML_ERROR_NONE) | |
4424 | return result; | |
4425 | else if (!next) { | |
4426 | processor = ignoreSectionProcessor; | |
4427 | return result; | |
4428 | } | |
4429 | } | |
4430 | break; | |
4431 | #endif /* XML_DTD */ | |
4432 | case XML_ROLE_GROUP_OPEN: | |
4433 | if (prologState.level >= groupSize) { | |
4434 | if (groupSize) { | |
4435 | char *temp = (char *)REALLOC(groupConnector, groupSize *= 2); | |
4436 | if (temp == NULL) | |
4437 | return XML_ERROR_NO_MEMORY; | |
4438 | groupConnector = temp; | |
4439 | if (dtd->scaffIndex) { | |
4440 | int *temp = (int *)REALLOC(dtd->scaffIndex, | |
4441 | groupSize * sizeof(int)); | |
4442 | if (temp == NULL) | |
4443 | return XML_ERROR_NO_MEMORY; | |
4444 | dtd->scaffIndex = temp; | |
4445 | } | |
4446 | } | |
4447 | else { | |
4448 | groupConnector = (char *)MALLOC(groupSize = 32); | |
4449 | if (!groupConnector) | |
4450 | return XML_ERROR_NO_MEMORY; | |
4451 | } | |
4452 | } | |
4453 | groupConnector[prologState.level] = 0; | |
4454 | if (dtd->in_eldecl) { | |
4455 | int myindex = nextScaffoldPart(parser); | |
4456 | if (myindex < 0) | |
4457 | return XML_ERROR_NO_MEMORY; | |
4458 | dtd->scaffIndex[dtd->scaffLevel] = myindex; | |
4459 | dtd->scaffLevel++; | |
4460 | dtd->scaffold[myindex].type = XML_CTYPE_SEQ; | |
4461 | if (elementDeclHandler) | |
4462 | handleDefault = XML_FALSE; | |
4463 | } | |
4464 | break; | |
4465 | case XML_ROLE_GROUP_SEQUENCE: | |
4466 | if (groupConnector[prologState.level] == ASCII_PIPE) | |
4467 | return XML_ERROR_SYNTAX; | |
4468 | groupConnector[prologState.level] = ASCII_COMMA; | |
4469 | if (dtd->in_eldecl && elementDeclHandler) | |
4470 | handleDefault = XML_FALSE; | |
4471 | break; | |
4472 | case XML_ROLE_GROUP_CHOICE: | |
4473 | if (groupConnector[prologState.level] == ASCII_COMMA) | |
4474 | return XML_ERROR_SYNTAX; | |
4475 | if (dtd->in_eldecl | |
4476 | && !groupConnector[prologState.level] | |
4477 | && (dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type | |
4478 | != XML_CTYPE_MIXED) | |
4479 | ) { | |
4480 | dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type | |
4481 | = XML_CTYPE_CHOICE; | |
4482 | if (elementDeclHandler) | |
4483 | handleDefault = XML_FALSE; | |
4484 | } | |
4485 | groupConnector[prologState.level] = ASCII_PIPE; | |
4486 | break; | |
4487 | case XML_ROLE_PARAM_ENTITY_REF: | |
4488 | #ifdef XML_DTD | |
4489 | case XML_ROLE_INNER_PARAM_ENTITY_REF: | |
4490 | dtd->hasParamEntityRefs = XML_TRUE; | |
4491 | if (!paramEntityParsing) | |
4492 | dtd->keepProcessing = dtd->standalone; | |
4493 | else { | |
4494 | const XML_Char *name; | |
4495 | ENTITY *entity; | |
4496 | name = poolStoreString(&dtd->pool, enc, | |
4497 | s + enc->minBytesPerChar, | |
4498 | next - enc->minBytesPerChar); | |
4499 | if (!name) | |
4500 | return XML_ERROR_NO_MEMORY; | |
4501 | entity = (ENTITY *)lookup(parser, &dtd->paramEntities, name, 0); | |
4502 | poolDiscard(&dtd->pool); | |
4503 | /* first, determine if a check for an existing declaration is needed; | |
4504 | if yes, check that the entity exists, and that it is internal, | |
4505 | otherwise call the skipped entity handler | |
4506 | */ | |
4507 | if (prologState.documentEntity && | |
4508 | (dtd->standalone | |
4509 | ? !openInternalEntities | |
4510 | : !dtd->hasParamEntityRefs)) { | |
4511 | if (!entity) | |
4512 | return XML_ERROR_UNDEFINED_ENTITY; | |
4513 | else if (!entity->is_internal) | |
4514 | return XML_ERROR_ENTITY_DECLARED_IN_PE; | |
4515 | } | |
4516 | else if (!entity) { | |
4517 | dtd->keepProcessing = dtd->standalone; | |
4518 | /* cannot report skipped entities in declarations */ | |
4519 | if ((role == XML_ROLE_PARAM_ENTITY_REF) && skippedEntityHandler) { | |
4520 | skippedEntityHandler(handlerArg, name, 1); | |
4521 | handleDefault = XML_FALSE; | |
4522 | } | |
4523 | break; | |
4524 | } | |
4525 | if (entity->open) | |
4526 | return XML_ERROR_RECURSIVE_ENTITY_REF; | |
4527 | if (entity->textPtr) { | |
4528 | enum XML_Error result; | |
4529 | XML_Bool betweenDecl = | |
4530 | (role == XML_ROLE_PARAM_ENTITY_REF ? XML_TRUE : XML_FALSE); | |
4531 | result = processInternalEntity(parser, entity, betweenDecl); | |
4532 | if (result != XML_ERROR_NONE) | |
4533 | return result; | |
4534 | handleDefault = XML_FALSE; | |
4535 | break; | |
4536 | } | |
4537 | if (externalEntityRefHandler) { | |
4538 | dtd->paramEntityRead = XML_FALSE; | |
4539 | entity->open = XML_TRUE; | |
4540 | if (!externalEntityRefHandler(externalEntityRefHandlerArg, | |
4541 | 0, | |
4542 | entity->base, | |
4543 | entity->systemId, | |
4544 | entity->publicId)) { | |
4545 | entity->open = XML_FALSE; | |
4546 | return XML_ERROR_EXTERNAL_ENTITY_HANDLING; | |
4547 | } | |
4548 | entity->open = XML_FALSE; | |
4549 | handleDefault = XML_FALSE; | |
4550 | if (!dtd->paramEntityRead) { | |
4551 | dtd->keepProcessing = dtd->standalone; | |
4552 | break; | |
4553 | } | |
4554 | } | |
4555 | else { | |
4556 | dtd->keepProcessing = dtd->standalone; | |
4557 | break; | |
4558 | } | |
4559 | } | |
4560 | #endif /* XML_DTD */ | |
4561 | if (!dtd->standalone && | |
4562 | notStandaloneHandler && | |
4563 | !notStandaloneHandler(handlerArg)) | |
4564 | return XML_ERROR_NOT_STANDALONE; | |
4565 | break; | |
4566 | ||
4567 | /* Element declaration stuff */ | |
4568 | ||
4569 | case XML_ROLE_ELEMENT_NAME: | |
4570 | if (elementDeclHandler) { | |
4571 | declElementType = getElementType(parser, enc, s, next); | |
4572 | if (!declElementType) | |
4573 | return XML_ERROR_NO_MEMORY; | |
4574 | dtd->scaffLevel = 0; | |
4575 | dtd->scaffCount = 0; | |
4576 | dtd->in_eldecl = XML_TRUE; | |
4577 | handleDefault = XML_FALSE; | |
4578 | } | |
4579 | break; | |
4580 | ||
4581 | case XML_ROLE_CONTENT_ANY: | |
4582 | case XML_ROLE_CONTENT_EMPTY: | |
4583 | if (dtd->in_eldecl) { | |
4584 | if (elementDeclHandler) { | |
4585 | XML_Content * content = (XML_Content *) MALLOC(sizeof(XML_Content)); | |
4586 | if (!content) | |
4587 | return XML_ERROR_NO_MEMORY; | |
4588 | content->quant = XML_CQUANT_NONE; | |
4589 | content->name = NULL; | |
4590 | content->numchildren = 0; | |
4591 | content->children = NULL; | |
4592 | content->type = ((role == XML_ROLE_CONTENT_ANY) ? | |
4593 | XML_CTYPE_ANY : | |
4594 | XML_CTYPE_EMPTY); | |
4595 | *eventEndPP = s; | |
4596 | elementDeclHandler(handlerArg, declElementType->name, content); | |
4597 | handleDefault = XML_FALSE; | |
4598 | } | |
4599 | dtd->in_eldecl = XML_FALSE; | |
4600 | } | |
4601 | break; | |
4602 | ||
4603 | case XML_ROLE_CONTENT_PCDATA: | |
4604 | if (dtd->in_eldecl) { | |
4605 | dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type | |
4606 | = XML_CTYPE_MIXED; | |
4607 | if (elementDeclHandler) | |
4608 | handleDefault = XML_FALSE; | |
4609 | } | |
4610 | break; | |
4611 | ||
4612 | case XML_ROLE_CONTENT_ELEMENT: | |
4613 | quant = XML_CQUANT_NONE; | |
4614 | goto elementContent; | |
4615 | case XML_ROLE_CONTENT_ELEMENT_OPT: | |
4616 | quant = XML_CQUANT_OPT; | |
4617 | goto elementContent; | |
4618 | case XML_ROLE_CONTENT_ELEMENT_REP: | |
4619 | quant = XML_CQUANT_REP; | |
4620 | goto elementContent; | |
4621 | case XML_ROLE_CONTENT_ELEMENT_PLUS: | |
4622 | quant = XML_CQUANT_PLUS; | |
4623 | elementContent: | |
4624 | if (dtd->in_eldecl) { | |
4625 | ELEMENT_TYPE *el; | |
4626 | const XML_Char *name; | |
4627 | int nameLen; | |
4628 | const char *nxt = (quant == XML_CQUANT_NONE | |
4629 | ? next | |
4630 | : next - enc->minBytesPerChar); | |
4631 | int myindex = nextScaffoldPart(parser); | |
4632 | if (myindex < 0) | |
4633 | return XML_ERROR_NO_MEMORY; | |
4634 | dtd->scaffold[myindex].type = XML_CTYPE_NAME; | |
4635 | dtd->scaffold[myindex].quant = quant; | |
4636 | el = getElementType(parser, enc, s, nxt); | |
4637 | if (!el) | |
4638 | return XML_ERROR_NO_MEMORY; | |
4639 | name = el->name; | |
4640 | dtd->scaffold[myindex].name = name; | |
4641 | nameLen = 0; | |
4642 | for (; name[nameLen++]; ) | |
4643 | ; | |
4644 | dtd->contentStringLen += nameLen; | |
4645 | if (elementDeclHandler) | |
4646 | handleDefault = XML_FALSE; | |
4647 | } | |
4648 | break; | |
4649 | ||
4650 | case XML_ROLE_GROUP_CLOSE: | |
4651 | quant = XML_CQUANT_NONE; | |
4652 | goto closeGroup; | |
4653 | case XML_ROLE_GROUP_CLOSE_OPT: | |
4654 | quant = XML_CQUANT_OPT; | |
4655 | goto closeGroup; | |
4656 | case XML_ROLE_GROUP_CLOSE_REP: | |
4657 | quant = XML_CQUANT_REP; | |
4658 | goto closeGroup; | |
4659 | case XML_ROLE_GROUP_CLOSE_PLUS: | |
4660 | quant = XML_CQUANT_PLUS; | |
4661 | closeGroup: | |
4662 | if (dtd->in_eldecl) { | |
4663 | if (elementDeclHandler) | |
4664 | handleDefault = XML_FALSE; | |
4665 | dtd->scaffLevel--; | |
4666 | dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel]].quant = quant; | |
4667 | if (dtd->scaffLevel == 0) { | |
4668 | if (!handleDefault) { | |
4669 | XML_Content *model = build_model(parser); | |
4670 | if (!model) | |
4671 | return XML_ERROR_NO_MEMORY; | |
4672 | *eventEndPP = s; | |
4673 | elementDeclHandler(handlerArg, declElementType->name, model); | |
4674 | } | |
4675 | dtd->in_eldecl = XML_FALSE; | |
4676 | dtd->contentStringLen = 0; | |
4677 | } | |
4678 | } | |
4679 | break; | |
4680 | /* End element declaration stuff */ | |
4681 | ||
4682 | case XML_ROLE_PI: | |
4683 | if (!reportProcessingInstruction(parser, enc, s, next)) | |
4684 | return XML_ERROR_NO_MEMORY; | |
4685 | handleDefault = XML_FALSE; | |
4686 | break; | |
4687 | case XML_ROLE_COMMENT: | |
4688 | if (!reportComment(parser, enc, s, next)) | |
4689 | return XML_ERROR_NO_MEMORY; | |
4690 | handleDefault = XML_FALSE; | |
4691 | break; | |
4692 | case XML_ROLE_NONE: | |
4693 | switch (tok) { | |
4694 | case XML_TOK_BOM: | |
4695 | handleDefault = XML_FALSE; | |
4696 | break; | |
4697 | } | |
4698 | break; | |
4699 | case XML_ROLE_DOCTYPE_NONE: | |
4700 | if (startDoctypeDeclHandler) | |
4701 | handleDefault = XML_FALSE; | |
4702 | break; | |
4703 | case XML_ROLE_ENTITY_NONE: | |
4704 | if (dtd->keepProcessing && entityDeclHandler) | |
4705 | handleDefault = XML_FALSE; | |
4706 | break; | |
4707 | case XML_ROLE_NOTATION_NONE: | |
4708 | if (notationDeclHandler) | |
4709 | handleDefault = XML_FALSE; | |
4710 | break; | |
4711 | case XML_ROLE_ATTLIST_NONE: | |
4712 | if (dtd->keepProcessing && attlistDeclHandler) | |
4713 | handleDefault = XML_FALSE; | |
4714 | break; | |
4715 | case XML_ROLE_ELEMENT_NONE: | |
4716 | if (elementDeclHandler) | |
4717 | handleDefault = XML_FALSE; | |
4718 | break; | |
4719 | } /* end of big switch */ | |
4720 | ||
4721 | if (handleDefault && defaultHandler) | |
4722 | reportDefault(parser, enc, s, next); | |
4723 | ||
4724 | switch (ps_parsing) { | |
4725 | case XML_SUSPENDED: | |
4726 | *nextPtr = next; | |
4727 | return XML_ERROR_NONE; | |
4728 | case XML_FINISHED: | |
4729 | return XML_ERROR_ABORTED; | |
4730 | default: | |
4731 | s = next; | |
4732 | tok = XmlPrologTok(enc, s, end, &next); | |
4733 | } | |
4734 | } | |
4735 | /* not reached */ | |
4736 | } | |
4737 | ||
4738 | static enum XML_Error PTRCALL | |
4739 | epilogProcessor(XML_Parser parser, | |
4740 | const char *s, | |
4741 | const char *end, | |
4742 | const char **nextPtr) | |
4743 | { | |
4744 | processor = epilogProcessor; | |
4745 | eventPtr = s; | |
4746 | for (;;) { | |
4747 | const char *next = NULL; | |
4748 | int tok = XmlPrologTok(encoding, s, end, &next); | |
4749 | eventEndPtr = next; | |
4750 | switch (tok) { | |
4751 | /* report partial linebreak - it might be the last token */ | |
4752 | case -XML_TOK_PROLOG_S: | |
4753 | if (defaultHandler) { | |
4754 | reportDefault(parser, encoding, s, next); | |
4755 | if (ps_parsing == XML_FINISHED) | |
4756 | return XML_ERROR_ABORTED; | |
4757 | } | |
4758 | *nextPtr = next; | |
4759 | return XML_ERROR_NONE; | |
4760 | case XML_TOK_NONE: | |
4761 | *nextPtr = s; | |
4762 | return XML_ERROR_NONE; | |
4763 | case XML_TOK_PROLOG_S: | |
4764 | if (defaultHandler) | |
4765 | reportDefault(parser, encoding, s, next); | |
4766 | break; | |
4767 | case XML_TOK_PI: | |
4768 | if (!reportProcessingInstruction(parser, encoding, s, next)) | |
4769 | return XML_ERROR_NO_MEMORY; | |
4770 | break; | |
4771 | case XML_TOK_COMMENT: | |
4772 | if (!reportComment(parser, encoding, s, next)) | |
4773 | return XML_ERROR_NO_MEMORY; | |
4774 | break; | |
4775 | case XML_TOK_INVALID: | |
4776 | eventPtr = next; | |
4777 | return XML_ERROR_INVALID_TOKEN; | |
4778 | case XML_TOK_PARTIAL: | |
4779 | if (!ps_finalBuffer) { | |
4780 | *nextPtr = s; | |
4781 | return XML_ERROR_NONE; | |
4782 | } | |
4783 | return XML_ERROR_UNCLOSED_TOKEN; | |
4784 | case XML_TOK_PARTIAL_CHAR: | |
4785 | if (!ps_finalBuffer) { | |
4786 | *nextPtr = s; | |
4787 | return XML_ERROR_NONE; | |
4788 | } | |
4789 | return XML_ERROR_PARTIAL_CHAR; | |
4790 | default: | |
4791 | return XML_ERROR_JUNK_AFTER_DOC_ELEMENT; | |
4792 | } | |
4793 | eventPtr = s = next; | |
4794 | switch (ps_parsing) { | |
4795 | case XML_SUSPENDED: | |
4796 | *nextPtr = next; | |
4797 | return XML_ERROR_NONE; | |
4798 | case XML_FINISHED: | |
4799 | return XML_ERROR_ABORTED; | |
4800 | default: ; | |
4801 | } | |
4802 | } | |
4803 | } | |
4804 | ||
4805 | static enum XML_Error | |
4806 | processInternalEntity(XML_Parser parser, ENTITY *entity, | |
4807 | XML_Bool betweenDecl) | |
4808 | { | |
4809 | const char *textStart, *textEnd; | |
4810 | const char *next; | |
4811 | enum XML_Error result; | |
4812 | OPEN_INTERNAL_ENTITY *openEntity; | |
4813 | ||
4814 | if (freeInternalEntities) { | |
4815 | openEntity = freeInternalEntities; | |
4816 | freeInternalEntities = openEntity->next; | |
4817 | } | |
4818 | else { | |
4819 | openEntity = (OPEN_INTERNAL_ENTITY *)MALLOC(sizeof(OPEN_INTERNAL_ENTITY)); | |
4820 | if (!openEntity) | |
4821 | return XML_ERROR_NO_MEMORY; | |
4822 | } | |
4823 | entity->open = XML_TRUE; | |
4824 | entity->processed = 0; | |
4825 | openEntity->next = openInternalEntities; | |
4826 | openInternalEntities = openEntity; | |
4827 | openEntity->entity = entity; | |
4828 | openEntity->startTagLevel = tagLevel; | |
4829 | openEntity->betweenDecl = betweenDecl; | |
4830 | openEntity->internalEventPtr = NULL; | |
4831 | openEntity->internalEventEndPtr = NULL; | |
4832 | textStart = (char *)entity->textPtr; | |
4833 | textEnd = (char *)(entity->textPtr + entity->textLen); | |
4834 | ||
4835 | #ifdef XML_DTD | |
4836 | if (entity->is_param) { | |
4837 | int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next); | |
4838 | result = doProlog(parser, internalEncoding, textStart, textEnd, tok, | |
4839 | next, &next, XML_FALSE); | |
4840 | } | |
4841 | else | |
4842 | #endif /* XML_DTD */ | |
4843 | result = doContent(parser, tagLevel, internalEncoding, textStart, | |
4844 | textEnd, &next, XML_FALSE); | |
4845 | ||
4846 | if (result == XML_ERROR_NONE) { | |
4847 | if (textEnd != next && ps_parsing == XML_SUSPENDED) { | |
4848 | entity->processed = (int)(next - textStart); | |
4849 | processor = internalEntityProcessor; | |
4850 | } | |
4851 | else { | |
4852 | entity->open = XML_FALSE; | |
4853 | openInternalEntities = openEntity->next; | |
4854 | /* put openEntity back in list of free instances */ | |
4855 | openEntity->next = freeInternalEntities; | |
4856 | freeInternalEntities = openEntity; | |
4857 | } | |
4858 | } | |
4859 | return result; | |
4860 | } | |
4861 | ||
4862 | static enum XML_Error PTRCALL | |
4863 | internalEntityProcessor(XML_Parser parser, | |
4864 | const char *s, | |
4865 | const char *end, | |
4866 | const char **nextPtr) | |
4867 | { | |
4868 | ENTITY *entity; | |
4869 | const char *textStart, *textEnd; | |
4870 | const char *next; | |
4871 | enum XML_Error result; | |
4872 | OPEN_INTERNAL_ENTITY *openEntity = openInternalEntities; | |
4873 | if (!openEntity) | |
4874 | return XML_ERROR_UNEXPECTED_STATE; | |
4875 | ||
4876 | entity = openEntity->entity; | |
4877 | textStart = ((char *)entity->textPtr) + entity->processed; | |
4878 | textEnd = (char *)(entity->textPtr + entity->textLen); | |
4879 | ||
4880 | #ifdef XML_DTD | |
4881 | if (entity->is_param) { | |
4882 | int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next); | |
4883 | result = doProlog(parser, internalEncoding, textStart, textEnd, tok, | |
4884 | next, &next, XML_FALSE); | |
4885 | } | |
4886 | else | |
4887 | #endif /* XML_DTD */ | |
4888 | result = doContent(parser, openEntity->startTagLevel, internalEncoding, | |
4889 | textStart, textEnd, &next, XML_FALSE); | |
4890 | ||
4891 | if (result != XML_ERROR_NONE) | |
4892 | return result; | |
4893 | else if (textEnd != next && ps_parsing == XML_SUSPENDED) { | |
4894 | entity->processed = (int)(next - (char *)entity->textPtr); | |
4895 | return result; | |
4896 | } | |
4897 | else { | |
4898 | entity->open = XML_FALSE; | |
4899 | openInternalEntities = openEntity->next; | |
4900 | /* put openEntity back in list of free instances */ | |
4901 | openEntity->next = freeInternalEntities; | |
4902 | freeInternalEntities = openEntity; | |
4903 | } | |
4904 | ||
4905 | #ifdef XML_DTD | |
4906 | if (entity->is_param) { | |
4907 | int tok; | |
4908 | processor = prologProcessor; | |
4909 | tok = XmlPrologTok(encoding, s, end, &next); | |
4910 | return doProlog(parser, encoding, s, end, tok, next, nextPtr, | |
4911 | (XML_Bool)!ps_finalBuffer); | |
4912 | } | |
4913 | else | |
4914 | #endif /* XML_DTD */ | |
4915 | { | |
4916 | processor = contentProcessor; | |
4917 | /* see externalEntityContentProcessor vs contentProcessor */ | |
4918 | return doContent(parser, parentParser ? 1 : 0, encoding, s, end, | |
4919 | nextPtr, (XML_Bool)!ps_finalBuffer); | |
4920 | } | |
4921 | } | |
4922 | ||
4923 | static enum XML_Error PTRCALL | |
4924 | errorProcessor(XML_Parser parser, | |
4925 | const char *s, | |
4926 | const char *end, | |
4927 | const char **nextPtr) | |
4928 | { | |
4929 | return errorCode; | |
4930 | } | |
4931 | ||
4932 | static enum XML_Error | |
4933 | storeAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata, | |
4934 | const char *ptr, const char *end, | |
4935 | STRING_POOL *pool) | |
4936 | { | |
4937 | enum XML_Error result = appendAttributeValue(parser, enc, isCdata, ptr, | |
4938 | end, pool); | |
4939 | if (result) | |
4940 | return result; | |
4941 | if (!isCdata && poolLength(pool) && poolLastChar(pool) == 0x20) | |
4942 | poolChop(pool); | |
4943 | if (!poolAppendChar(pool, XML_T('\0'))) | |
4944 | return XML_ERROR_NO_MEMORY; | |
4945 | return XML_ERROR_NONE; | |
4946 | } | |
4947 | ||
4948 | static enum XML_Error | |
4949 | appendAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata, | |
4950 | const char *ptr, const char *end, | |
4951 | STRING_POOL *pool) | |
4952 | { | |
4953 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
4954 | for (;;) { | |
4955 | const char *next; | |
4956 | int tok = XmlAttributeValueTok(enc, ptr, end, &next); | |
4957 | switch (tok) { | |
4958 | case XML_TOK_NONE: | |
4959 | return XML_ERROR_NONE; | |
4960 | case XML_TOK_INVALID: | |
4961 | if (enc == encoding) | |
4962 | eventPtr = next; | |
4963 | return XML_ERROR_INVALID_TOKEN; | |
4964 | case XML_TOK_PARTIAL: | |
4965 | if (enc == encoding) | |
4966 | eventPtr = ptr; | |
4967 | return XML_ERROR_INVALID_TOKEN; | |
4968 | case XML_TOK_CHAR_REF: | |
4969 | { | |
4970 | XML_Char buf[XML_ENCODE_MAX]; | |
4971 | int i; | |
4972 | int n = XmlCharRefNumber(enc, ptr); | |
4973 | if (n < 0) { | |
4974 | if (enc == encoding) | |
4975 | eventPtr = ptr; | |
4976 | return XML_ERROR_BAD_CHAR_REF; | |
4977 | } | |
4978 | if (!isCdata | |
4979 | && n == 0x20 /* space */ | |
4980 | && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20)) | |
4981 | break; | |
4982 | n = XmlEncode(n, (ICHAR *)buf); | |
4983 | if (!n) { | |
4984 | if (enc == encoding) | |
4985 | eventPtr = ptr; | |
4986 | return XML_ERROR_BAD_CHAR_REF; | |
4987 | } | |
4988 | for (i = 0; i < n; i++) { | |
4989 | if (!poolAppendChar(pool, buf[i])) | |
4990 | return XML_ERROR_NO_MEMORY; | |
4991 | } | |
4992 | } | |
4993 | break; | |
4994 | case XML_TOK_DATA_CHARS: | |
4995 | if (!poolAppend(pool, enc, ptr, next)) | |
4996 | return XML_ERROR_NO_MEMORY; | |
4997 | break; | |
4998 | case XML_TOK_TRAILING_CR: | |
4999 | next = ptr + enc->minBytesPerChar; | |
5000 | /* fall through */ | |
5001 | case XML_TOK_ATTRIBUTE_VALUE_S: | |
5002 | case XML_TOK_DATA_NEWLINE: | |
5003 | if (!isCdata && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20)) | |
5004 | break; | |
5005 | if (!poolAppendChar(pool, 0x20)) | |
5006 | return XML_ERROR_NO_MEMORY; | |
5007 | break; | |
5008 | case XML_TOK_ENTITY_REF: | |
5009 | { | |
5010 | const XML_Char *name; | |
5011 | ENTITY *entity; | |
5012 | char checkEntityDecl; | |
5013 | XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc, | |
5014 | ptr + enc->minBytesPerChar, | |
5015 | next - enc->minBytesPerChar); | |
5016 | if (ch) { | |
5017 | if (!poolAppendChar(pool, ch)) | |
5018 | return XML_ERROR_NO_MEMORY; | |
5019 | break; | |
5020 | } | |
5021 | name = poolStoreString(&temp2Pool, enc, | |
5022 | ptr + enc->minBytesPerChar, | |
5023 | next - enc->minBytesPerChar); | |
5024 | if (!name) | |
5025 | return XML_ERROR_NO_MEMORY; | |
5026 | entity = (ENTITY *)lookup(parser, &dtd->generalEntities, name, 0); | |
5027 | poolDiscard(&temp2Pool); | |
5028 | /* First, determine if a check for an existing declaration is needed; | |
5029 | if yes, check that the entity exists, and that it is internal. | |
5030 | */ | |
5031 | if (pool == &dtd->pool) /* are we called from prolog? */ | |
5032 | checkEntityDecl = | |
5033 | #ifdef XML_DTD | |
5034 | prologState.documentEntity && | |
5035 | #endif /* XML_DTD */ | |
5036 | (dtd->standalone | |
5037 | ? !openInternalEntities | |
5038 | : !dtd->hasParamEntityRefs); | |
5039 | else /* if (pool == &tempPool): we are called from content */ | |
5040 | checkEntityDecl = !dtd->hasParamEntityRefs || dtd->standalone; | |
5041 | if (checkEntityDecl) { | |
5042 | if (!entity) | |
5043 | return XML_ERROR_UNDEFINED_ENTITY; | |
5044 | else if (!entity->is_internal) | |
5045 | return XML_ERROR_ENTITY_DECLARED_IN_PE; | |
5046 | } | |
5047 | else if (!entity) { | |
5048 | /* Cannot report skipped entity here - see comments on | |
5049 | skippedEntityHandler. | |
5050 | if (skippedEntityHandler) | |
5051 | skippedEntityHandler(handlerArg, name, 0); | |
5052 | */ | |
5053 | /* Cannot call the default handler because this would be | |
5054 | out of sync with the call to the startElementHandler. | |
5055 | if ((pool == &tempPool) && defaultHandler) | |
5056 | reportDefault(parser, enc, ptr, next); | |
5057 | */ | |
5058 | break; | |
5059 | } | |
5060 | if (entity->open) { | |
5061 | if (enc == encoding) | |
5062 | eventPtr = ptr; | |
5063 | return XML_ERROR_RECURSIVE_ENTITY_REF; | |
5064 | } | |
5065 | if (entity->notation) { | |
5066 | if (enc == encoding) | |
5067 | eventPtr = ptr; | |
5068 | return XML_ERROR_BINARY_ENTITY_REF; | |
5069 | } | |
5070 | if (!entity->textPtr) { | |
5071 | if (enc == encoding) | |
5072 | eventPtr = ptr; | |
5073 | return XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF; | |
5074 | } | |
5075 | else { | |
5076 | enum XML_Error result; | |
5077 | const XML_Char *textEnd = entity->textPtr + entity->textLen; | |
5078 | entity->open = XML_TRUE; | |
5079 | result = appendAttributeValue(parser, internalEncoding, isCdata, | |
5080 | (char *)entity->textPtr, | |
5081 | (char *)textEnd, pool); | |
5082 | entity->open = XML_FALSE; | |
5083 | if (result) | |
5084 | return result; | |
5085 | } | |
5086 | } | |
5087 | break; | |
5088 | default: | |
5089 | if (enc == encoding) | |
5090 | eventPtr = ptr; | |
5091 | return XML_ERROR_UNEXPECTED_STATE; | |
5092 | } | |
5093 | ptr = next; | |
5094 | } | |
5095 | /* not reached */ | |
5096 | } | |
5097 | ||
5098 | static enum XML_Error | |
5099 | storeEntityValue(XML_Parser parser, | |
5100 | const ENCODING *enc, | |
5101 | const char *entityTextPtr, | |
5102 | const char *entityTextEnd) | |
5103 | { | |
5104 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
5105 | STRING_POOL *pool = &(dtd->entityValuePool); | |
5106 | enum XML_Error result = XML_ERROR_NONE; | |
5107 | #ifdef XML_DTD | |
5108 | int oldInEntityValue = prologState.inEntityValue; | |
5109 | prologState.inEntityValue = 1; | |
5110 | #endif /* XML_DTD */ | |
5111 | /* never return Null for the value argument in EntityDeclHandler, | |
5112 | since this would indicate an external entity; therefore we | |
5113 | have to make sure that entityValuePool.start is not null */ | |
5114 | if (!pool->blocks) { | |
5115 | if (!poolGrow(pool)) | |
5116 | return XML_ERROR_NO_MEMORY; | |
5117 | } | |
5118 | ||
5119 | for (;;) { | |
5120 | const char *next; | |
5121 | int tok = XmlEntityValueTok(enc, entityTextPtr, entityTextEnd, &next); | |
5122 | switch (tok) { | |
5123 | case XML_TOK_PARAM_ENTITY_REF: | |
5124 | #ifdef XML_DTD | |
5125 | if (isParamEntity || enc != encoding) { | |
5126 | const XML_Char *name; | |
5127 | ENTITY *entity; | |
5128 | name = poolStoreString(&tempPool, enc, | |
5129 | entityTextPtr + enc->minBytesPerChar, | |
5130 | next - enc->minBytesPerChar); | |
5131 | if (!name) { | |
5132 | result = XML_ERROR_NO_MEMORY; | |
5133 | goto endEntityValue; | |
5134 | } | |
5135 | entity = (ENTITY *)lookup(parser, &dtd->paramEntities, name, 0); | |
5136 | poolDiscard(&tempPool); | |
5137 | if (!entity) { | |
5138 | /* not a well-formedness error - see XML 1.0: WFC Entity Declared */ | |
5139 | /* cannot report skipped entity here - see comments on | |
5140 | skippedEntityHandler | |
5141 | if (skippedEntityHandler) | |
5142 | skippedEntityHandler(handlerArg, name, 0); | |
5143 | */ | |
5144 | dtd->keepProcessing = dtd->standalone; | |
5145 | goto endEntityValue; | |
5146 | } | |
5147 | if (entity->open) { | |
5148 | if (enc == encoding) | |
5149 | eventPtr = entityTextPtr; | |
5150 | result = XML_ERROR_RECURSIVE_ENTITY_REF; | |
5151 | goto endEntityValue; | |
5152 | } | |
5153 | if (entity->systemId) { | |
5154 | if (externalEntityRefHandler) { | |
5155 | dtd->paramEntityRead = XML_FALSE; | |
5156 | entity->open = XML_TRUE; | |
5157 | if (!externalEntityRefHandler(externalEntityRefHandlerArg, | |
5158 | 0, | |
5159 | entity->base, | |
5160 | entity->systemId, | |
5161 | entity->publicId)) { | |
5162 | entity->open = XML_FALSE; | |
5163 | result = XML_ERROR_EXTERNAL_ENTITY_HANDLING; | |
5164 | goto endEntityValue; | |
5165 | } | |
5166 | entity->open = XML_FALSE; | |
5167 | if (!dtd->paramEntityRead) | |
5168 | dtd->keepProcessing = dtd->standalone; | |
5169 | } | |
5170 | else | |
5171 | dtd->keepProcessing = dtd->standalone; | |
5172 | } | |
5173 | else { | |
5174 | entity->open = XML_TRUE; | |
5175 | result = storeEntityValue(parser, | |
5176 | internalEncoding, | |
5177 | (char *)entity->textPtr, | |
5178 | (char *)(entity->textPtr | |
5179 | + entity->textLen)); | |
5180 | entity->open = XML_FALSE; | |
5181 | if (result) | |
5182 | goto endEntityValue; | |
5183 | } | |
5184 | break; | |
5185 | } | |
5186 | #endif /* XML_DTD */ | |
5187 | /* In the internal subset, PE references are not legal | |
5188 | within markup declarations, e.g entity values in this case. */ | |
5189 | eventPtr = entityTextPtr; | |
5190 | result = XML_ERROR_PARAM_ENTITY_REF; | |
5191 | goto endEntityValue; | |
5192 | case XML_TOK_NONE: | |
5193 | result = XML_ERROR_NONE; | |
5194 | goto endEntityValue; | |
5195 | case XML_TOK_ENTITY_REF: | |
5196 | case XML_TOK_DATA_CHARS: | |
5197 | if (!poolAppend(pool, enc, entityTextPtr, next)) { | |
5198 | result = XML_ERROR_NO_MEMORY; | |
5199 | goto endEntityValue; | |
5200 | } | |
5201 | break; | |
5202 | case XML_TOK_TRAILING_CR: | |
5203 | next = entityTextPtr + enc->minBytesPerChar; | |
5204 | /* fall through */ | |
5205 | case XML_TOK_DATA_NEWLINE: | |
5206 | if (pool->end == pool->ptr && !poolGrow(pool)) { | |
5207 | result = XML_ERROR_NO_MEMORY; | |
5208 | goto endEntityValue; | |
5209 | } | |
5210 | *(pool->ptr)++ = 0xA; | |
5211 | break; | |
5212 | case XML_TOK_CHAR_REF: | |
5213 | { | |
5214 | XML_Char buf[XML_ENCODE_MAX]; | |
5215 | int i; | |
5216 | int n = XmlCharRefNumber(enc, entityTextPtr); | |
5217 | if (n < 0) { | |
5218 | if (enc == encoding) | |
5219 | eventPtr = entityTextPtr; | |
5220 | result = XML_ERROR_BAD_CHAR_REF; | |
5221 | goto endEntityValue; | |
5222 | } | |
5223 | n = XmlEncode(n, (ICHAR *)buf); | |
5224 | if (!n) { | |
5225 | if (enc == encoding) | |
5226 | eventPtr = entityTextPtr; | |
5227 | result = XML_ERROR_BAD_CHAR_REF; | |
5228 | goto endEntityValue; | |
5229 | } | |
5230 | for (i = 0; i < n; i++) { | |
5231 | if (pool->end == pool->ptr && !poolGrow(pool)) { | |
5232 | result = XML_ERROR_NO_MEMORY; | |
5233 | goto endEntityValue; | |
5234 | } | |
5235 | *(pool->ptr)++ = buf[i]; | |
5236 | } | |
5237 | } | |
5238 | break; | |
5239 | case XML_TOK_PARTIAL: | |
5240 | if (enc == encoding) | |
5241 | eventPtr = entityTextPtr; | |
5242 | result = XML_ERROR_INVALID_TOKEN; | |
5243 | goto endEntityValue; | |
5244 | case XML_TOK_INVALID: | |
5245 | if (enc == encoding) | |
5246 | eventPtr = next; | |
5247 | result = XML_ERROR_INVALID_TOKEN; | |
5248 | goto endEntityValue; | |
5249 | default: | |
5250 | if (enc == encoding) | |
5251 | eventPtr = entityTextPtr; | |
5252 | result = XML_ERROR_UNEXPECTED_STATE; | |
5253 | goto endEntityValue; | |
5254 | } | |
5255 | entityTextPtr = next; | |
5256 | } | |
5257 | endEntityValue: | |
5258 | #ifdef XML_DTD | |
5259 | prologState.inEntityValue = oldInEntityValue; | |
5260 | #endif /* XML_DTD */ | |
5261 | return result; | |
5262 | } | |
5263 | ||
5264 | static void FASTCALL | |
5265 | normalizeLines(XML_Char *s) | |
5266 | { | |
5267 | XML_Char *p; | |
5268 | for (;; s++) { | |
5269 | if (*s == XML_T('\0')) | |
5270 | return; | |
5271 | if (*s == 0xD) | |
5272 | break; | |
5273 | } | |
5274 | p = s; | |
5275 | do { | |
5276 | if (*s == 0xD) { | |
5277 | *p++ = 0xA; | |
5278 | if (*++s == 0xA) | |
5279 | s++; | |
5280 | } | |
5281 | else | |
5282 | *p++ = *s++; | |
5283 | } while (*s); | |
5284 | *p = XML_T('\0'); | |
5285 | } | |
5286 | ||
5287 | static int | |
5288 | reportProcessingInstruction(XML_Parser parser, const ENCODING *enc, | |
5289 | const char *start, const char *end) | |
5290 | { | |
5291 | const XML_Char *target; | |
5292 | XML_Char *data; | |
5293 | const char *tem; | |
5294 | if (!processingInstructionHandler) { | |
5295 | if (defaultHandler) | |
5296 | reportDefault(parser, enc, start, end); | |
5297 | return 1; | |
5298 | } | |
5299 | start += enc->minBytesPerChar * 2; | |
5300 | tem = start + XmlNameLength(enc, start); | |
5301 | target = poolStoreString(&tempPool, enc, start, tem); | |
5302 | if (!target) | |
5303 | return 0; | |
5304 | poolFinish(&tempPool); | |
5305 | data = poolStoreString(&tempPool, enc, | |
5306 | XmlSkipS(enc, tem), | |
5307 | end - enc->minBytesPerChar*2); | |
5308 | if (!data) | |
5309 | return 0; | |
5310 | normalizeLines(data); | |
5311 | processingInstructionHandler(handlerArg, target, data); | |
5312 | poolClear(&tempPool); | |
5313 | return 1; | |
5314 | } | |
5315 | ||
5316 | static int | |
5317 | reportComment(XML_Parser parser, const ENCODING *enc, | |
5318 | const char *start, const char *end) | |
5319 | { | |
5320 | XML_Char *data; | |
5321 | if (!commentHandler) { | |
5322 | if (defaultHandler) | |
5323 | reportDefault(parser, enc, start, end); | |
5324 | return 1; | |
5325 | } | |
5326 | data = poolStoreString(&tempPool, | |
5327 | enc, | |
5328 | start + enc->minBytesPerChar * 4, | |
5329 | end - enc->minBytesPerChar * 3); | |
5330 | if (!data) | |
5331 | return 0; | |
5332 | normalizeLines(data); | |
5333 | commentHandler(handlerArg, data); | |
5334 | poolClear(&tempPool); | |
5335 | return 1; | |
5336 | } | |
5337 | ||
5338 | static void | |
5339 | reportDefault(XML_Parser parser, const ENCODING *enc, | |
5340 | const char *s, const char *end) | |
5341 | { | |
5342 | if (MUST_CONVERT(enc, s)) { | |
5343 | const char **eventPP; | |
5344 | const char **eventEndPP; | |
5345 | if (enc == encoding) { | |
5346 | eventPP = &eventPtr; | |
5347 | eventEndPP = &eventEndPtr; | |
5348 | } | |
5349 | else { | |
5350 | eventPP = &(openInternalEntities->internalEventPtr); | |
5351 | eventEndPP = &(openInternalEntities->internalEventEndPtr); | |
5352 | } | |
5353 | do { | |
5354 | ICHAR *dataPtr = (ICHAR *)dataBuf; | |
5355 | XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd); | |
5356 | *eventEndPP = s; | |
5357 | defaultHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf)); | |
5358 | *eventPP = s; | |
5359 | } while (s != end); | |
5360 | } | |
5361 | else | |
5362 | defaultHandler(handlerArg, (XML_Char *)s, (int)((XML_Char *)end - (XML_Char *)s)); | |
5363 | } | |
5364 | ||
5365 | ||
5366 | static int | |
5367 | defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *attId, XML_Bool isCdata, | |
5368 | XML_Bool isId, const XML_Char *value, XML_Parser parser) | |
5369 | { | |
5370 | DEFAULT_ATTRIBUTE *att; | |
5371 | if (value || isId) { | |
5372 | /* The handling of default attributes gets messed up if we have | |
5373 | a default which duplicates a non-default. */ | |
5374 | int i; | |
5375 | for (i = 0; i < type->nDefaultAtts; i++) | |
5376 | if (attId == type->defaultAtts[i].id) | |
5377 | return 1; | |
5378 | if (isId && !type->idAtt && !attId->xmlns) | |
5379 | type->idAtt = attId; | |
5380 | } | |
5381 | if (type->nDefaultAtts == type->allocDefaultAtts) { | |
5382 | if (type->allocDefaultAtts == 0) { | |
5383 | type->allocDefaultAtts = 8; | |
5384 | type->defaultAtts = (DEFAULT_ATTRIBUTE *)MALLOC(type->allocDefaultAtts | |
5385 | * sizeof(DEFAULT_ATTRIBUTE)); | |
5386 | if (!type->defaultAtts) | |
5387 | return 0; | |
5388 | } | |
5389 | else { | |
5390 | DEFAULT_ATTRIBUTE *temp; | |
5391 | int count = type->allocDefaultAtts * 2; | |
5392 | temp = (DEFAULT_ATTRIBUTE *) | |
5393 | REALLOC(type->defaultAtts, (count * sizeof(DEFAULT_ATTRIBUTE))); | |
5394 | if (temp == NULL) | |
5395 | return 0; | |
5396 | type->allocDefaultAtts = count; | |
5397 | type->defaultAtts = temp; | |
5398 | } | |
5399 | } | |
5400 | att = type->defaultAtts + type->nDefaultAtts; | |
5401 | att->id = attId; | |
5402 | att->value = value; | |
5403 | att->isCdata = isCdata; | |
5404 | if (!isCdata) | |
5405 | attId->maybeTokenized = XML_TRUE; | |
5406 | type->nDefaultAtts += 1; | |
5407 | return 1; | |
5408 | } | |
5409 | ||
5410 | static int | |
5411 | setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *elementType) | |
5412 | { | |
5413 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
5414 | const XML_Char *name; | |
5415 | for (name = elementType->name; *name; name++) { | |
5416 | if (*name == XML_T(ASCII_COLON)) { | |
5417 | PREFIX *prefix; | |
5418 | const XML_Char *s; | |
5419 | for (s = elementType->name; s != name; s++) { | |
5420 | if (!poolAppendChar(&dtd->pool, *s)) | |
5421 | return 0; | |
5422 | } | |
5423 | if (!poolAppendChar(&dtd->pool, XML_T('\0'))) | |
5424 | return 0; | |
5425 | prefix = (PREFIX *)lookup(parser, &dtd->prefixes, poolStart(&dtd->pool), | |
5426 | sizeof(PREFIX)); | |
5427 | if (!prefix) | |
5428 | return 0; | |
5429 | if (prefix->name == poolStart(&dtd->pool)) | |
5430 | poolFinish(&dtd->pool); | |
5431 | else | |
5432 | poolDiscard(&dtd->pool); | |
5433 | elementType->prefix = prefix; | |
5434 | ||
5435 | } | |
5436 | } | |
5437 | return 1; | |
5438 | } | |
5439 | ||
5440 | static ATTRIBUTE_ID * | |
5441 | getAttributeId(XML_Parser parser, const ENCODING *enc, | |
5442 | const char *start, const char *end) | |
5443 | { | |
5444 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
5445 | ATTRIBUTE_ID *id; | |
5446 | const XML_Char *name; | |
5447 | if (!poolAppendChar(&dtd->pool, XML_T('\0'))) | |
5448 | return NULL; | |
5449 | name = poolStoreString(&dtd->pool, enc, start, end); | |
5450 | if (!name) | |
5451 | return NULL; | |
5452 | /* skip quotation mark - its storage will be re-used (like in name[-1]) */ | |
5453 | ++name; | |
5454 | id = (ATTRIBUTE_ID *)lookup(parser, &dtd->attributeIds, name, sizeof(ATTRIBUTE_ID)); | |
5455 | if (!id) | |
5456 | return NULL; | |
5457 | if (id->name != name) | |
5458 | poolDiscard(&dtd->pool); | |
5459 | else { | |
5460 | poolFinish(&dtd->pool); | |
5461 | if (!ns) | |
5462 | ; | |
5463 | else if (name[0] == XML_T(ASCII_x) | |
5464 | && name[1] == XML_T(ASCII_m) | |
5465 | && name[2] == XML_T(ASCII_l) | |
5466 | && name[3] == XML_T(ASCII_n) | |
5467 | && name[4] == XML_T(ASCII_s) | |
5468 | && (name[5] == XML_T('\0') || name[5] == XML_T(ASCII_COLON))) { | |
5469 | if (name[5] == XML_T('\0')) | |
5470 | id->prefix = &dtd->defaultPrefix; | |
5471 | else | |
5472 | id->prefix = (PREFIX *)lookup(parser, &dtd->prefixes, name + 6, sizeof(PREFIX)); | |
5473 | id->xmlns = XML_TRUE; | |
5474 | } | |
5475 | else { | |
5476 | int i; | |
5477 | for (i = 0; name[i]; i++) { | |
5478 | /* attributes without prefix are *not* in the default namespace */ | |
5479 | if (name[i] == XML_T(ASCII_COLON)) { | |
5480 | int j; | |
5481 | for (j = 0; j < i; j++) { | |
5482 | if (!poolAppendChar(&dtd->pool, name[j])) | |
5483 | return NULL; | |
5484 | } | |
5485 | if (!poolAppendChar(&dtd->pool, XML_T('\0'))) | |
5486 | return NULL; | |
5487 | id->prefix = (PREFIX *)lookup(parser, &dtd->prefixes, poolStart(&dtd->pool), | |
5488 | sizeof(PREFIX)); | |
5489 | if (id->prefix->name == poolStart(&dtd->pool)) | |
5490 | poolFinish(&dtd->pool); | |
5491 | else | |
5492 | poolDiscard(&dtd->pool); | |
5493 | break; | |
5494 | } | |
5495 | } | |
5496 | } | |
5497 | } | |
5498 | return id; | |
5499 | } | |
5500 | ||
5501 | #define CONTEXT_SEP XML_T(ASCII_FF) | |
5502 | ||
5503 | static const XML_Char * | |
5504 | getContext(XML_Parser parser) | |
5505 | { | |
5506 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
5507 | HASH_TABLE_ITER iter; | |
5508 | XML_Bool needSep = XML_FALSE; | |
5509 | ||
5510 | if (dtd->defaultPrefix.binding) { | |
5511 | int i; | |
5512 | int len; | |
5513 | if (!poolAppendChar(&tempPool, XML_T(ASCII_EQUALS))) | |
5514 | return NULL; | |
5515 | len = dtd->defaultPrefix.binding->uriLen; | |
5516 | if (namespaceSeparator) | |
5517 | len--; | |
5518 | for (i = 0; i < len; i++) | |
5519 | if (!poolAppendChar(&tempPool, dtd->defaultPrefix.binding->uri[i])) | |
5520 | return NULL; | |
5521 | needSep = XML_TRUE; | |
5522 | } | |
5523 | ||
5524 | hashTableIterInit(&iter, &(dtd->prefixes)); | |
5525 | for (;;) { | |
5526 | int i; | |
5527 | int len; | |
5528 | const XML_Char *s; | |
5529 | PREFIX *prefix = (PREFIX *)hashTableIterNext(&iter); | |
5530 | if (!prefix) | |
5531 | break; | |
5532 | if (!prefix->binding) | |
5533 | continue; | |
5534 | if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP)) | |
5535 | return NULL; | |
5536 | for (s = prefix->name; *s; s++) | |
5537 | if (!poolAppendChar(&tempPool, *s)) | |
5538 | return NULL; | |
5539 | if (!poolAppendChar(&tempPool, XML_T(ASCII_EQUALS))) | |
5540 | return NULL; | |
5541 | len = prefix->binding->uriLen; | |
5542 | if (namespaceSeparator) | |
5543 | len--; | |
5544 | for (i = 0; i < len; i++) | |
5545 | if (!poolAppendChar(&tempPool, prefix->binding->uri[i])) | |
5546 | return NULL; | |
5547 | needSep = XML_TRUE; | |
5548 | } | |
5549 | ||
5550 | ||
5551 | hashTableIterInit(&iter, &(dtd->generalEntities)); | |
5552 | for (;;) { | |
5553 | const XML_Char *s; | |
5554 | ENTITY *e = (ENTITY *)hashTableIterNext(&iter); | |
5555 | if (!e) | |
5556 | break; | |
5557 | if (!e->open) | |
5558 | continue; | |
5559 | if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP)) | |
5560 | return NULL; | |
5561 | for (s = e->name; *s; s++) | |
5562 | if (!poolAppendChar(&tempPool, *s)) | |
5563 | return 0; | |
5564 | needSep = XML_TRUE; | |
5565 | } | |
5566 | ||
5567 | if (!poolAppendChar(&tempPool, XML_T('\0'))) | |
5568 | return NULL; | |
5569 | return tempPool.start; | |
5570 | } | |
5571 | ||
5572 | static XML_Bool | |
5573 | setContext(XML_Parser parser, const XML_Char *context) | |
5574 | { | |
5575 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
5576 | const XML_Char *s = context; | |
5577 | ||
5578 | while (*context != XML_T('\0')) { | |
5579 | if (*s == CONTEXT_SEP || *s == XML_T('\0')) { | |
5580 | ENTITY *e; | |
5581 | if (!poolAppendChar(&tempPool, XML_T('\0'))) | |
5582 | return XML_FALSE; | |
5583 | e = (ENTITY *)lookup(parser, &dtd->generalEntities, poolStart(&tempPool), 0); | |
5584 | if (e) | |
5585 | e->open = XML_TRUE; | |
5586 | if (*s != XML_T('\0')) | |
5587 | s++; | |
5588 | context = s; | |
5589 | poolDiscard(&tempPool); | |
5590 | } | |
5591 | else if (*s == XML_T(ASCII_EQUALS)) { | |
5592 | PREFIX *prefix; | |
5593 | if (poolLength(&tempPool) == 0) | |
5594 | prefix = &dtd->defaultPrefix; | |
5595 | else { | |
5596 | if (!poolAppendChar(&tempPool, XML_T('\0'))) | |
5597 | return XML_FALSE; | |
5598 | prefix = (PREFIX *)lookup(parser, &dtd->prefixes, poolStart(&tempPool), | |
5599 | sizeof(PREFIX)); | |
5600 | if (!prefix) | |
5601 | return XML_FALSE; | |
5602 | if (prefix->name == poolStart(&tempPool)) { | |
5603 | prefix->name = poolCopyString(&dtd->pool, prefix->name); | |
5604 | if (!prefix->name) | |
5605 | return XML_FALSE; | |
5606 | } | |
5607 | poolDiscard(&tempPool); | |
5608 | } | |
5609 | for (context = s + 1; | |
5610 | *context != CONTEXT_SEP && *context != XML_T('\0'); | |
5611 | context++) | |
5612 | if (!poolAppendChar(&tempPool, *context)) | |
5613 | return XML_FALSE; | |
5614 | if (!poolAppendChar(&tempPool, XML_T('\0'))) | |
5615 | return XML_FALSE; | |
5616 | if (addBinding(parser, prefix, NULL, poolStart(&tempPool), | |
5617 | &inheritedBindings) != XML_ERROR_NONE) | |
5618 | return XML_FALSE; | |
5619 | poolDiscard(&tempPool); | |
5620 | if (*context != XML_T('\0')) | |
5621 | ++context; | |
5622 | s = context; | |
5623 | } | |
5624 | else { | |
5625 | if (!poolAppendChar(&tempPool, *s)) | |
5626 | return XML_FALSE; | |
5627 | s++; | |
5628 | } | |
5629 | } | |
5630 | return XML_TRUE; | |
5631 | } | |
5632 | ||
5633 | static void FASTCALL | |
5634 | normalizePublicId(XML_Char *publicId) | |
5635 | { | |
5636 | XML_Char *p = publicId; | |
5637 | XML_Char *s; | |
5638 | for (s = publicId; *s; s++) { | |
5639 | switch (*s) { | |
5640 | case 0x20: | |
5641 | case 0xD: | |
5642 | case 0xA: | |
5643 | if (p != publicId && p[-1] != 0x20) | |
5644 | *p++ = 0x20; | |
5645 | break; | |
5646 | default: | |
5647 | *p++ = *s; | |
5648 | } | |
5649 | } | |
5650 | if (p != publicId && p[-1] == 0x20) | |
5651 | --p; | |
5652 | *p = XML_T('\0'); | |
5653 | } | |
5654 | ||
5655 | static DTD * | |
5656 | dtdCreate(const XML_Memory_Handling_Suite *ms) | |
5657 | { | |
5658 | DTD *p = (DTD *)ms->malloc_fcn(sizeof(DTD)); | |
5659 | if (p == NULL) | |
5660 | return p; | |
5661 | poolInit(&(p->pool), ms); | |
5662 | poolInit(&(p->entityValuePool), ms); | |
5663 | hashTableInit(&(p->generalEntities), ms); | |
5664 | hashTableInit(&(p->elementTypes), ms); | |
5665 | hashTableInit(&(p->attributeIds), ms); | |
5666 | hashTableInit(&(p->prefixes), ms); | |
5667 | #ifdef XML_DTD | |
5668 | p->paramEntityRead = XML_FALSE; | |
5669 | hashTableInit(&(p->paramEntities), ms); | |
5670 | #endif /* XML_DTD */ | |
5671 | p->defaultPrefix.name = NULL; | |
5672 | p->defaultPrefix.binding = NULL; | |
5673 | ||
5674 | p->in_eldecl = XML_FALSE; | |
5675 | p->scaffIndex = NULL; | |
5676 | p->scaffold = NULL; | |
5677 | p->scaffLevel = 0; | |
5678 | p->scaffSize = 0; | |
5679 | p->scaffCount = 0; | |
5680 | p->contentStringLen = 0; | |
5681 | ||
5682 | p->keepProcessing = XML_TRUE; | |
5683 | p->hasParamEntityRefs = XML_FALSE; | |
5684 | p->standalone = XML_FALSE; | |
5685 | return p; | |
5686 | } | |
5687 | ||
5688 | static void | |
5689 | dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms) | |
5690 | { | |
5691 | HASH_TABLE_ITER iter; | |
5692 | hashTableIterInit(&iter, &(p->elementTypes)); | |
5693 | for (;;) { | |
5694 | ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter); | |
5695 | if (!e) | |
5696 | break; | |
5697 | if (e->allocDefaultAtts != 0) | |
5698 | ms->free_fcn(e->defaultAtts); | |
5699 | } | |
5700 | hashTableClear(&(p->generalEntities)); | |
5701 | #ifdef XML_DTD | |
5702 | p->paramEntityRead = XML_FALSE; | |
5703 | hashTableClear(&(p->paramEntities)); | |
5704 | #endif /* XML_DTD */ | |
5705 | hashTableClear(&(p->elementTypes)); | |
5706 | hashTableClear(&(p->attributeIds)); | |
5707 | hashTableClear(&(p->prefixes)); | |
5708 | poolClear(&(p->pool)); | |
5709 | poolClear(&(p->entityValuePool)); | |
5710 | p->defaultPrefix.name = NULL; | |
5711 | p->defaultPrefix.binding = NULL; | |
5712 | ||
5713 | p->in_eldecl = XML_FALSE; | |
5714 | ||
5715 | ms->free_fcn(p->scaffIndex); | |
5716 | p->scaffIndex = NULL; | |
5717 | ms->free_fcn(p->scaffold); | |
5718 | p->scaffold = NULL; | |
5719 | ||
5720 | p->scaffLevel = 0; | |
5721 | p->scaffSize = 0; | |
5722 | p->scaffCount = 0; | |
5723 | p->contentStringLen = 0; | |
5724 | ||
5725 | p->keepProcessing = XML_TRUE; | |
5726 | p->hasParamEntityRefs = XML_FALSE; | |
5727 | p->standalone = XML_FALSE; | |
5728 | } | |
5729 | ||
5730 | static void | |
5731 | dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms) | |
5732 | { | |
5733 | HASH_TABLE_ITER iter; | |
5734 | hashTableIterInit(&iter, &(p->elementTypes)); | |
5735 | for (;;) { | |
5736 | ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter); | |
5737 | if (!e) | |
5738 | break; | |
5739 | if (e->allocDefaultAtts != 0) | |
5740 | ms->free_fcn(e->defaultAtts); | |
5741 | } | |
5742 | hashTableDestroy(&(p->generalEntities)); | |
5743 | #ifdef XML_DTD | |
5744 | hashTableDestroy(&(p->paramEntities)); | |
5745 | #endif /* XML_DTD */ | |
5746 | hashTableDestroy(&(p->elementTypes)); | |
5747 | hashTableDestroy(&(p->attributeIds)); | |
5748 | hashTableDestroy(&(p->prefixes)); | |
5749 | poolDestroy(&(p->pool)); | |
5750 | poolDestroy(&(p->entityValuePool)); | |
5751 | if (isDocEntity) { | |
5752 | ms->free_fcn(p->scaffIndex); | |
5753 | ms->free_fcn(p->scaffold); | |
5754 | } | |
5755 | ms->free_fcn(p); | |
5756 | } | |
5757 | ||
5758 | /* Do a deep copy of the DTD. Return 0 for out of memory, non-zero otherwise. | |
5759 | The new DTD has already been initialized. | |
5760 | */ | |
5761 | static int | |
5762 | dtdCopy(XML_Parser oldParser, DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms) | |
5763 | { | |
5764 | HASH_TABLE_ITER iter; | |
5765 | ||
5766 | /* Copy the prefix table. */ | |
5767 | ||
5768 | hashTableIterInit(&iter, &(oldDtd->prefixes)); | |
5769 | for (;;) { | |
5770 | const XML_Char *name; | |
5771 | const PREFIX *oldP = (PREFIX *)hashTableIterNext(&iter); | |
5772 | if (!oldP) | |
5773 | break; | |
5774 | name = poolCopyString(&(newDtd->pool), oldP->name); | |
5775 | if (!name) | |
5776 | return 0; | |
5777 | if (!lookup(oldParser, &(newDtd->prefixes), name, sizeof(PREFIX))) | |
5778 | return 0; | |
5779 | } | |
5780 | ||
5781 | hashTableIterInit(&iter, &(oldDtd->attributeIds)); | |
5782 | ||
5783 | /* Copy the attribute id table. */ | |
5784 | ||
5785 | for (;;) { | |
5786 | ATTRIBUTE_ID *newA; | |
5787 | const XML_Char *name; | |
5788 | const ATTRIBUTE_ID *oldA = (ATTRIBUTE_ID *)hashTableIterNext(&iter); | |
5789 | ||
5790 | if (!oldA) | |
5791 | break; | |
5792 | /* Remember to allocate the scratch byte before the name. */ | |
5793 | if (!poolAppendChar(&(newDtd->pool), XML_T('\0'))) | |
5794 | return 0; | |
5795 | name = poolCopyString(&(newDtd->pool), oldA->name); | |
5796 | if (!name) | |
5797 | return 0; | |
5798 | ++name; | |
5799 | newA = (ATTRIBUTE_ID *)lookup(oldParser, &(newDtd->attributeIds), name, | |
5800 | sizeof(ATTRIBUTE_ID)); | |
5801 | if (!newA) | |
5802 | return 0; | |
5803 | newA->maybeTokenized = oldA->maybeTokenized; | |
5804 | if (oldA->prefix) { | |
5805 | newA->xmlns = oldA->xmlns; | |
5806 | if (oldA->prefix == &oldDtd->defaultPrefix) | |
5807 | newA->prefix = &newDtd->defaultPrefix; | |
5808 | else | |
5809 | newA->prefix = (PREFIX *)lookup(oldParser, &(newDtd->prefixes), | |
5810 | oldA->prefix->name, 0); | |
5811 | } | |
5812 | } | |
5813 | ||
5814 | /* Copy the element type table. */ | |
5815 | ||
5816 | hashTableIterInit(&iter, &(oldDtd->elementTypes)); | |
5817 | ||
5818 | for (;;) { | |
5819 | int i; | |
5820 | ELEMENT_TYPE *newE; | |
5821 | const XML_Char *name; | |
5822 | const ELEMENT_TYPE *oldE = (ELEMENT_TYPE *)hashTableIterNext(&iter); | |
5823 | if (!oldE) | |
5824 | break; | |
5825 | name = poolCopyString(&(newDtd->pool), oldE->name); | |
5826 | if (!name) | |
5827 | return 0; | |
5828 | newE = (ELEMENT_TYPE *)lookup(oldParser, &(newDtd->elementTypes), name, | |
5829 | sizeof(ELEMENT_TYPE)); | |
5830 | if (!newE) | |
5831 | return 0; | |
5832 | if (oldE->nDefaultAtts) { | |
5833 | newE->defaultAtts = (DEFAULT_ATTRIBUTE *) | |
5834 | ms->malloc_fcn(oldE->nDefaultAtts * sizeof(DEFAULT_ATTRIBUTE)); | |
5835 | if (!newE->defaultAtts) { | |
5836 | ms->free_fcn(newE); | |
5837 | return 0; | |
5838 | } | |
5839 | } | |
5840 | if (oldE->idAtt) | |
5841 | newE->idAtt = (ATTRIBUTE_ID *) | |
5842 | lookup(oldParser, &(newDtd->attributeIds), oldE->idAtt->name, 0); | |
5843 | newE->allocDefaultAtts = newE->nDefaultAtts = oldE->nDefaultAtts; | |
5844 | if (oldE->prefix) | |
5845 | newE->prefix = (PREFIX *)lookup(oldParser, &(newDtd->prefixes), | |
5846 | oldE->prefix->name, 0); | |
5847 | for (i = 0; i < newE->nDefaultAtts; i++) { | |
5848 | newE->defaultAtts[i].id = (ATTRIBUTE_ID *) | |
5849 | lookup(oldParser, &(newDtd->attributeIds), oldE->defaultAtts[i].id->name, 0); | |
5850 | newE->defaultAtts[i].isCdata = oldE->defaultAtts[i].isCdata; | |
5851 | if (oldE->defaultAtts[i].value) { | |
5852 | newE->defaultAtts[i].value | |
5853 | = poolCopyString(&(newDtd->pool), oldE->defaultAtts[i].value); | |
5854 | if (!newE->defaultAtts[i].value) | |
5855 | return 0; | |
5856 | } | |
5857 | else | |
5858 | newE->defaultAtts[i].value = NULL; | |
5859 | } | |
5860 | } | |
5861 | ||
5862 | /* Copy the entity tables. */ | |
5863 | if (!copyEntityTable(oldParser, | |
5864 | &(newDtd->generalEntities), | |
5865 | &(newDtd->pool), | |
5866 | &(oldDtd->generalEntities))) | |
5867 | return 0; | |
5868 | ||
5869 | #ifdef XML_DTD | |
5870 | if (!copyEntityTable(oldParser, | |
5871 | &(newDtd->paramEntities), | |
5872 | &(newDtd->pool), | |
5873 | &(oldDtd->paramEntities))) | |
5874 | return 0; | |
5875 | newDtd->paramEntityRead = oldDtd->paramEntityRead; | |
5876 | #endif /* XML_DTD */ | |
5877 | ||
5878 | newDtd->keepProcessing = oldDtd->keepProcessing; | |
5879 | newDtd->hasParamEntityRefs = oldDtd->hasParamEntityRefs; | |
5880 | newDtd->standalone = oldDtd->standalone; | |
5881 | ||
5882 | /* Don't want deep copying for scaffolding */ | |
5883 | newDtd->in_eldecl = oldDtd->in_eldecl; | |
5884 | newDtd->scaffold = oldDtd->scaffold; | |
5885 | newDtd->contentStringLen = oldDtd->contentStringLen; | |
5886 | newDtd->scaffSize = oldDtd->scaffSize; | |
5887 | newDtd->scaffLevel = oldDtd->scaffLevel; | |
5888 | newDtd->scaffIndex = oldDtd->scaffIndex; | |
5889 | ||
5890 | return 1; | |
5891 | } /* End dtdCopy */ | |
5892 | ||
5893 | static int | |
5894 | copyEntityTable(XML_Parser oldParser, | |
5895 | HASH_TABLE *newTable, | |
5896 | STRING_POOL *newPool, | |
5897 | const HASH_TABLE *oldTable) | |
5898 | { | |
5899 | HASH_TABLE_ITER iter; | |
5900 | const XML_Char *cachedOldBase = NULL; | |
5901 | const XML_Char *cachedNewBase = NULL; | |
5902 | ||
5903 | hashTableIterInit(&iter, oldTable); | |
5904 | ||
5905 | for (;;) { | |
5906 | ENTITY *newE; | |
5907 | const XML_Char *name; | |
5908 | const ENTITY *oldE = (ENTITY *)hashTableIterNext(&iter); | |
5909 | if (!oldE) | |
5910 | break; | |
5911 | name = poolCopyString(newPool, oldE->name); | |
5912 | if (!name) | |
5913 | return 0; | |
5914 | newE = (ENTITY *)lookup(oldParser, newTable, name, sizeof(ENTITY)); | |
5915 | if (!newE) | |
5916 | return 0; | |
5917 | if (oldE->systemId) { | |
5918 | const XML_Char *tem = poolCopyString(newPool, oldE->systemId); | |
5919 | if (!tem) | |
5920 | return 0; | |
5921 | newE->systemId = tem; | |
5922 | if (oldE->base) { | |
5923 | if (oldE->base == cachedOldBase) | |
5924 | newE->base = cachedNewBase; | |
5925 | else { | |
5926 | cachedOldBase = oldE->base; | |
5927 | tem = poolCopyString(newPool, cachedOldBase); | |
5928 | if (!tem) | |
5929 | return 0; | |
5930 | cachedNewBase = newE->base = tem; | |
5931 | } | |
5932 | } | |
5933 | if (oldE->publicId) { | |
5934 | tem = poolCopyString(newPool, oldE->publicId); | |
5935 | if (!tem) | |
5936 | return 0; | |
5937 | newE->publicId = tem; | |
5938 | } | |
5939 | } | |
5940 | else { | |
5941 | const XML_Char *tem = poolCopyStringN(newPool, oldE->textPtr, | |
5942 | oldE->textLen); | |
5943 | if (!tem) | |
5944 | return 0; | |
5945 | newE->textPtr = tem; | |
5946 | newE->textLen = oldE->textLen; | |
5947 | } | |
5948 | if (oldE->notation) { | |
5949 | const XML_Char *tem = poolCopyString(newPool, oldE->notation); | |
5950 | if (!tem) | |
5951 | return 0; | |
5952 | newE->notation = tem; | |
5953 | } | |
5954 | newE->is_param = oldE->is_param; | |
5955 | newE->is_internal = oldE->is_internal; | |
5956 | } | |
5957 | return 1; | |
5958 | } | |
5959 | ||
5960 | #define INIT_POWER 6 | |
5961 | ||
5962 | static XML_Bool FASTCALL | |
5963 | keyeq(KEY s1, KEY s2) | |
5964 | { | |
5965 | for (; *s1 == *s2; s1++, s2++) | |
5966 | if (*s1 == 0) | |
5967 | return XML_TRUE; | |
5968 | return XML_FALSE; | |
5969 | } | |
5970 | ||
5971 | static unsigned long FASTCALL | |
5972 | hash(XML_Parser parser, KEY s) | |
5973 | { | |
5974 | unsigned long h = hash_secret_salt; | |
5975 | while (*s) | |
5976 | h = CHAR_HASH(h, *s++); | |
5977 | return h; | |
5978 | } | |
5979 | ||
5980 | static NAMED * | |
5981 | lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize) | |
5982 | { | |
5983 | size_t i; | |
5984 | if (table->size == 0) { | |
5985 | size_t tsize; | |
5986 | if (!createSize) | |
5987 | return NULL; | |
5988 | table->power = INIT_POWER; | |
5989 | /* table->size is a power of 2 */ | |
5990 | table->size = (size_t)1 << INIT_POWER; | |
5991 | tsize = table->size * sizeof(NAMED *); | |
5992 | table->v = (NAMED **)table->mem->malloc_fcn(tsize); | |
5993 | if (!table->v) { | |
5994 | table->size = 0; | |
5995 | return NULL; | |
5996 | } | |
5997 | memset(table->v, 0, tsize); | |
5998 | i = hash(parser, name) & ((unsigned long)table->size - 1); | |
5999 | } | |
6000 | else { | |
6001 | unsigned long h = hash(parser, name); | |
6002 | unsigned long mask = (unsigned long)table->size - 1; | |
6003 | unsigned char step = 0; | |
6004 | i = h & mask; | |
6005 | while (table->v[i]) { | |
6006 | if (keyeq(name, table->v[i]->name)) | |
6007 | return table->v[i]; | |
6008 | if (!step) | |
6009 | step = PROBE_STEP(h, mask, table->power); | |
6010 | i < step ? (i += table->size - step) : (i -= step); | |
6011 | } | |
6012 | if (!createSize) | |
6013 | return NULL; | |
6014 | ||
6015 | /* check for overflow (table is half full) */ | |
6016 | if (table->used >> (table->power - 1)) { | |
6017 | unsigned char newPower = table->power + 1; | |
6018 | size_t newSize = (size_t)1 << newPower; | |
6019 | unsigned long newMask = (unsigned long)newSize - 1; | |
6020 | size_t tsize = newSize * sizeof(NAMED *); | |
6021 | NAMED **newV = (NAMED **)table->mem->malloc_fcn(tsize); | |
6022 | if (!newV) | |
6023 | return NULL; | |
6024 | memset(newV, 0, tsize); | |
6025 | for (i = 0; i < table->size; i++) | |
6026 | if (table->v[i]) { | |
6027 | unsigned long newHash = hash(parser, table->v[i]->name); | |
6028 | size_t j = newHash & newMask; | |
6029 | step = 0; | |
6030 | while (newV[j]) { | |
6031 | if (!step) | |
6032 | step = PROBE_STEP(newHash, newMask, newPower); | |
6033 | j < step ? (j += newSize - step) : (j -= step); | |
6034 | } | |
6035 | newV[j] = table->v[i]; | |
6036 | } | |
6037 | table->mem->free_fcn(table->v); | |
6038 | table->v = newV; | |
6039 | table->power = newPower; | |
6040 | table->size = newSize; | |
6041 | i = h & newMask; | |
6042 | step = 0; | |
6043 | while (table->v[i]) { | |
6044 | if (!step) | |
6045 | step = PROBE_STEP(h, newMask, newPower); | |
6046 | i < step ? (i += newSize - step) : (i -= step); | |
6047 | } | |
6048 | } | |
6049 | } | |
6050 | table->v[i] = (NAMED *)table->mem->malloc_fcn(createSize); | |
6051 | if (!table->v[i]) | |
6052 | return NULL; | |
6053 | memset(table->v[i], 0, createSize); | |
6054 | table->v[i]->name = name; | |
6055 | (table->used)++; | |
6056 | return table->v[i]; | |
6057 | } | |
6058 | ||
6059 | static void FASTCALL | |
6060 | hashTableClear(HASH_TABLE *table) | |
6061 | { | |
6062 | size_t i; | |
6063 | for (i = 0; i < table->size; i++) { | |
6064 | table->mem->free_fcn(table->v[i]); | |
6065 | table->v[i] = NULL; | |
6066 | } | |
6067 | table->used = 0; | |
6068 | } | |
6069 | ||
6070 | static void FASTCALL | |
6071 | hashTableDestroy(HASH_TABLE *table) | |
6072 | { | |
6073 | size_t i; | |
6074 | for (i = 0; i < table->size; i++) | |
6075 | table->mem->free_fcn(table->v[i]); | |
6076 | table->mem->free_fcn(table->v); | |
6077 | } | |
6078 | ||
6079 | static void FASTCALL | |
6080 | hashTableInit(HASH_TABLE *p, const XML_Memory_Handling_Suite *ms) | |
6081 | { | |
6082 | p->power = 0; | |
6083 | p->size = 0; | |
6084 | p->used = 0; | |
6085 | p->v = NULL; | |
6086 | p->mem = ms; | |
6087 | } | |
6088 | ||
6089 | static void FASTCALL | |
6090 | hashTableIterInit(HASH_TABLE_ITER *iter, const HASH_TABLE *table) | |
6091 | { | |
6092 | iter->p = table->v; | |
6093 | iter->end = iter->p + table->size; | |
6094 | } | |
6095 | ||
6096 | static NAMED * FASTCALL | |
6097 | hashTableIterNext(HASH_TABLE_ITER *iter) | |
6098 | { | |
6099 | while (iter->p != iter->end) { | |
6100 | NAMED *tem = *(iter->p)++; | |
6101 | if (tem) | |
6102 | return tem; | |
6103 | } | |
6104 | return NULL; | |
6105 | } | |
6106 | ||
6107 | static void FASTCALL | |
6108 | poolInit(STRING_POOL *pool, const XML_Memory_Handling_Suite *ms) | |
6109 | { | |
6110 | pool->blocks = NULL; | |
6111 | pool->freeBlocks = NULL; | |
6112 | pool->start = NULL; | |
6113 | pool->ptr = NULL; | |
6114 | pool->end = NULL; | |
6115 | pool->mem = ms; | |
6116 | } | |
6117 | ||
6118 | static void FASTCALL | |
6119 | poolClear(STRING_POOL *pool) | |
6120 | { | |
6121 | if (!pool->freeBlocks) | |
6122 | pool->freeBlocks = pool->blocks; | |
6123 | else { | |
6124 | BLOCK *p = pool->blocks; | |
6125 | while (p) { | |
6126 | BLOCK *tem = p->next; | |
6127 | p->next = pool->freeBlocks; | |
6128 | pool->freeBlocks = p; | |
6129 | p = tem; | |
6130 | } | |
6131 | } | |
6132 | pool->blocks = NULL; | |
6133 | pool->start = NULL; | |
6134 | pool->ptr = NULL; | |
6135 | pool->end = NULL; | |
6136 | } | |
6137 | ||
6138 | static void FASTCALL | |
6139 | poolDestroy(STRING_POOL *pool) | |
6140 | { | |
6141 | BLOCK *p = pool->blocks; | |
6142 | while (p) { | |
6143 | BLOCK *tem = p->next; | |
6144 | pool->mem->free_fcn(p); | |
6145 | p = tem; | |
6146 | } | |
6147 | p = pool->freeBlocks; | |
6148 | while (p) { | |
6149 | BLOCK *tem = p->next; | |
6150 | pool->mem->free_fcn(p); | |
6151 | p = tem; | |
6152 | } | |
6153 | } | |
6154 | ||
6155 | static XML_Char * | |
6156 | poolAppend(STRING_POOL *pool, const ENCODING *enc, | |
6157 | const char *ptr, const char *end) | |
6158 | { | |
6159 | if (!pool->ptr && !poolGrow(pool)) | |
6160 | return NULL; | |
6161 | for (;;) { | |
6162 | XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end); | |
6163 | if (ptr == end) | |
6164 | break; | |
6165 | if (!poolGrow(pool)) | |
6166 | return NULL; | |
6167 | } | |
6168 | return pool->start; | |
6169 | } | |
6170 | ||
6171 | static const XML_Char * FASTCALL | |
6172 | poolCopyString(STRING_POOL *pool, const XML_Char *s) | |
6173 | { | |
6174 | do { | |
6175 | if (!poolAppendChar(pool, *s)) | |
6176 | return NULL; | |
6177 | } while (*s++); | |
6178 | s = pool->start; | |
6179 | poolFinish(pool); | |
6180 | return s; | |
6181 | } | |
6182 | ||
6183 | static const XML_Char * | |
6184 | poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n) | |
6185 | { | |
6186 | if (!pool->ptr && !poolGrow(pool)) | |
6187 | return NULL; | |
6188 | for (; n > 0; --n, s++) { | |
6189 | if (!poolAppendChar(pool, *s)) | |
6190 | return NULL; | |
6191 | } | |
6192 | s = pool->start; | |
6193 | poolFinish(pool); | |
6194 | return s; | |
6195 | } | |
6196 | ||
6197 | static const XML_Char * FASTCALL | |
6198 | poolAppendString(STRING_POOL *pool, const XML_Char *s) | |
6199 | { | |
6200 | while (*s) { | |
6201 | if (!poolAppendChar(pool, *s)) | |
6202 | return NULL; | |
6203 | s++; | |
6204 | } | |
6205 | return pool->start; | |
6206 | } | |
6207 | ||
6208 | static XML_Char * | |
6209 | poolStoreString(STRING_POOL *pool, const ENCODING *enc, | |
6210 | const char *ptr, const char *end) | |
6211 | { | |
6212 | if (!poolAppend(pool, enc, ptr, end)) | |
6213 | return NULL; | |
6214 | if (pool->ptr == pool->end && !poolGrow(pool)) | |
6215 | return NULL; | |
6216 | *(pool->ptr)++ = 0; | |
6217 | return pool->start; | |
6218 | } | |
6219 | ||
6220 | static XML_Bool FASTCALL | |
6221 | poolGrow(STRING_POOL *pool) | |
6222 | { | |
6223 | if (pool->freeBlocks) { | |
6224 | if (pool->start == 0) { | |
6225 | pool->blocks = pool->freeBlocks; | |
6226 | pool->freeBlocks = pool->freeBlocks->next; | |
6227 | pool->blocks->next = NULL; | |
6228 | pool->start = pool->blocks->s; | |
6229 | pool->end = pool->start + pool->blocks->size; | |
6230 | pool->ptr = pool->start; | |
6231 | return XML_TRUE; | |
6232 | } | |
6233 | if (pool->end - pool->start < pool->freeBlocks->size) { | |
6234 | BLOCK *tem = pool->freeBlocks->next; | |
6235 | pool->freeBlocks->next = pool->blocks; | |
6236 | pool->blocks = pool->freeBlocks; | |
6237 | pool->freeBlocks = tem; | |
6238 | memcpy(pool->blocks->s, pool->start, | |
6239 | (pool->end - pool->start) * sizeof(XML_Char)); | |
6240 | pool->ptr = pool->blocks->s + (pool->ptr - pool->start); | |
6241 | pool->start = pool->blocks->s; | |
6242 | pool->end = pool->start + pool->blocks->size; | |
6243 | return XML_TRUE; | |
6244 | } | |
6245 | } | |
6246 | if (pool->blocks && pool->start == pool->blocks->s) { | |
6247 | int blockSize = (int)(pool->end - pool->start)*2; | |
6248 | BLOCK *temp = (BLOCK *) | |
6249 | pool->mem->realloc_fcn(pool->blocks, | |
6250 | (offsetof(BLOCK, s) | |
6251 | + blockSize * sizeof(XML_Char))); | |
6252 | if (temp == NULL) | |
6253 | return XML_FALSE; | |
6254 | pool->blocks = temp; | |
6255 | pool->blocks->size = blockSize; | |
6256 | pool->ptr = pool->blocks->s + (pool->ptr - pool->start); | |
6257 | pool->start = pool->blocks->s; | |
6258 | pool->end = pool->start + blockSize; | |
6259 | } | |
6260 | else { | |
6261 | BLOCK *tem; | |
6262 | int blockSize = (int)(pool->end - pool->start); | |
6263 | if (blockSize < INIT_BLOCK_SIZE) | |
6264 | blockSize = INIT_BLOCK_SIZE; | |
6265 | else | |
6266 | blockSize *= 2; | |
6267 | tem = (BLOCK *)pool->mem->malloc_fcn(offsetof(BLOCK, s) | |
6268 | + blockSize * sizeof(XML_Char)); | |
6269 | if (!tem) | |
6270 | return XML_FALSE; | |
6271 | tem->size = blockSize; | |
6272 | tem->next = pool->blocks; | |
6273 | pool->blocks = tem; | |
6274 | if (pool->ptr != pool->start) | |
6275 | memcpy(tem->s, pool->start, | |
6276 | (pool->ptr - pool->start) * sizeof(XML_Char)); | |
6277 | pool->ptr = tem->s + (pool->ptr - pool->start); | |
6278 | pool->start = tem->s; | |
6279 | pool->end = tem->s + blockSize; | |
6280 | } | |
6281 | return XML_TRUE; | |
6282 | } | |
6283 | ||
6284 | static int FASTCALL | |
6285 | nextScaffoldPart(XML_Parser parser) | |
6286 | { | |
6287 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
6288 | CONTENT_SCAFFOLD * me; | |
6289 | int next; | |
6290 | ||
6291 | if (!dtd->scaffIndex) { | |
6292 | dtd->scaffIndex = (int *)MALLOC(groupSize * sizeof(int)); | |
6293 | if (!dtd->scaffIndex) | |
6294 | return -1; | |
6295 | dtd->scaffIndex[0] = 0; | |
6296 | } | |
6297 | ||
6298 | if (dtd->scaffCount >= dtd->scaffSize) { | |
6299 | CONTENT_SCAFFOLD *temp; | |
6300 | if (dtd->scaffold) { | |
6301 | temp = (CONTENT_SCAFFOLD *) | |
6302 | REALLOC(dtd->scaffold, dtd->scaffSize * 2 * sizeof(CONTENT_SCAFFOLD)); | |
6303 | if (temp == NULL) | |
6304 | return -1; | |
6305 | dtd->scaffSize *= 2; | |
6306 | } | |
6307 | else { | |
6308 | temp = (CONTENT_SCAFFOLD *)MALLOC(INIT_SCAFFOLD_ELEMENTS | |
6309 | * sizeof(CONTENT_SCAFFOLD)); | |
6310 | if (temp == NULL) | |
6311 | return -1; | |
6312 | dtd->scaffSize = INIT_SCAFFOLD_ELEMENTS; | |
6313 | } | |
6314 | dtd->scaffold = temp; | |
6315 | } | |
6316 | next = dtd->scaffCount++; | |
6317 | me = &dtd->scaffold[next]; | |
6318 | if (dtd->scaffLevel) { | |
6319 | CONTENT_SCAFFOLD *parent = &dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel-1]]; | |
6320 | if (parent->lastchild) { | |
6321 | dtd->scaffold[parent->lastchild].nextsib = next; | |
6322 | } | |
6323 | if (!parent->childcnt) | |
6324 | parent->firstchild = next; | |
6325 | parent->lastchild = next; | |
6326 | parent->childcnt++; | |
6327 | } | |
6328 | me->firstchild = me->lastchild = me->childcnt = me->nextsib = 0; | |
6329 | return next; | |
6330 | } | |
6331 | ||
6332 | static void | |
6333 | build_node(XML_Parser parser, | |
6334 | int src_node, | |
6335 | XML_Content *dest, | |
6336 | XML_Content **contpos, | |
6337 | XML_Char **strpos) | |
6338 | { | |
6339 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
6340 | dest->type = dtd->scaffold[src_node].type; | |
6341 | dest->quant = dtd->scaffold[src_node].quant; | |
6342 | if (dest->type == XML_CTYPE_NAME) { | |
6343 | const XML_Char *src; | |
6344 | dest->name = *strpos; | |
6345 | src = dtd->scaffold[src_node].name; | |
6346 | for (;;) { | |
6347 | *(*strpos)++ = *src; | |
6348 | if (!*src) | |
6349 | break; | |
6350 | src++; | |
6351 | } | |
6352 | dest->numchildren = 0; | |
6353 | dest->children = NULL; | |
6354 | } | |
6355 | else { | |
6356 | unsigned int i; | |
6357 | int cn; | |
6358 | dest->numchildren = dtd->scaffold[src_node].childcnt; | |
6359 | dest->children = *contpos; | |
6360 | *contpos += dest->numchildren; | |
6361 | for (i = 0, cn = dtd->scaffold[src_node].firstchild; | |
6362 | i < dest->numchildren; | |
6363 | i++, cn = dtd->scaffold[cn].nextsib) { | |
6364 | build_node(parser, cn, &(dest->children[i]), contpos, strpos); | |
6365 | } | |
6366 | dest->name = NULL; | |
6367 | } | |
6368 | } | |
6369 | ||
6370 | static XML_Content * | |
6371 | build_model (XML_Parser parser) | |
6372 | { | |
6373 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
6374 | XML_Content *ret; | |
6375 | XML_Content *cpos; | |
6376 | XML_Char * str; | |
6377 | int allocsize = (dtd->scaffCount * sizeof(XML_Content) | |
6378 | + (dtd->contentStringLen * sizeof(XML_Char))); | |
6379 | ||
6380 | ret = (XML_Content *)MALLOC(allocsize); | |
6381 | if (!ret) | |
6382 | return NULL; | |
6383 | ||
6384 | str = (XML_Char *) (&ret[dtd->scaffCount]); | |
6385 | cpos = &ret[1]; | |
6386 | ||
6387 | build_node(parser, 0, ret, &cpos, &str); | |
6388 | return ret; | |
6389 | } | |
6390 | ||
6391 | static ELEMENT_TYPE * | |
6392 | getElementType(XML_Parser parser, | |
6393 | const ENCODING *enc, | |
6394 | const char *ptr, | |
6395 | const char *end) | |
6396 | { | |
6397 | DTD * const dtd = _dtd; /* save one level of indirection */ | |
6398 | const XML_Char *name = poolStoreString(&dtd->pool, enc, ptr, end); | |
6399 | ELEMENT_TYPE *ret; | |
6400 | ||
6401 | if (!name) | |
6402 | return NULL; | |
6403 | ret = (ELEMENT_TYPE *) lookup(parser, &dtd->elementTypes, name, sizeof(ELEMENT_TYPE)); | |
6404 | if (!ret) | |
6405 | return NULL; | |
6406 | if (ret->name != name) | |
6407 | poolDiscard(&dtd->pool); | |
6408 | else { | |
6409 | poolFinish(&dtd->pool); | |
6410 | if (!setElementTypePrefix(parser, ret)) | |
6411 | return NULL; | |
6412 | } | |
6413 | return ret; | |
6414 | } |