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
9 #define XML_BUILDING_EXPAT 1
10
11 #ifdef COMPILED_FROM_DSP
12 #include "winconfig.h"
13 #elif defined(MACOS_CLASSIC)
14 #include "macconfig.h"
15 #elif defined(__amigaos4__)
16 #include "amigaconfig.h"
17 #elif defined(HAVE_EXPAT_CONFIG_H)
18 #include <expat_config.h>
19 #endif /* ndef COMPILED_FROM_DSP */
20
21 #include "expat.h"
22
23 #ifdef XML_UNICODE
24 #define XML_ENCODE_MAX XML_UTF16_ENCODE_MAX
25 #define XmlConvert XmlUtf16Convert
26 #define XmlGetInternalEncoding XmlGetUtf16InternalEncoding
27 #define XmlGetInternalEncodingNS XmlGetUtf16InternalEncodingNS
28 #define XmlEncode XmlUtf16Encode
29 #define MUST_CONVERT(enc, s) (!(enc)->isUtf16 || (((unsigned long)s) & 1))
30 typedef unsigned short ICHAR;
31 #else
32 #define XML_ENCODE_MAX XML_UTF8_ENCODE_MAX
33 #define XmlConvert XmlUtf8Convert
34 #define XmlGetInternalEncoding XmlGetUtf8InternalEncoding
35 #define XmlGetInternalEncodingNS XmlGetUtf8InternalEncodingNS
36 #define XmlEncode XmlUtf8Encode
37 #define MUST_CONVERT(enc, s) (!(enc)->isUtf8)
38 typedef char ICHAR;
39 #endif
40
41
42 #ifndef XML_NS
43
44 #define XmlInitEncodingNS XmlInitEncoding
45 #define XmlInitUnknownEncodingNS XmlInitUnknownEncoding
46 #undef XmlGetInternalEncodingNS
47 #define XmlGetInternalEncodingNS XmlGetInternalEncoding
48 #define XmlParseXmlDeclNS XmlParseXmlDecl
49
50 #endif
51
52 #ifdef XML_UNICODE
53
54 #ifdef XML_UNICODE_WCHAR_T
55 #define XML_T(x) (const wchar_t)x
56 #define XML_L(x) L ## x
57 #else
58 #define XML_T(x) (const unsigned short)x
59 #define XML_L(x) x
60 #endif
61
62 #else
63
64 #define XML_T(x) x
65 #define XML_L(x) x
66
67 #endif
68
69 /* Round up n to be a multiple of sz, where sz is a power of 2. */
70 #define ROUND_UP(n, sz) (((n) + ((sz) - 1)) & ~((sz) - 1))
71
72 /* Handle the case where memmove() doesn't exist. */
73 #ifndef HAVE_MEMMOVE
74 #ifdef HAVE_BCOPY
75 #define memmove(d,s,l) bcopy((s),(d),(l))
76 #else
77 #error memmove does not exist on this platform, nor is a substitute available
78 #endif /* HAVE_BCOPY */
79 #endif /* HAVE_MEMMOVE */
80
81 #include "internal.h"
82 #include "xmltok.h"
83 #include "xmlrole.h"
84
85 typedef const XML_Char *KEY;
86
87 typedef struct {
88 KEY name;
89 } NAMED;
90
91 typedef struct {
92 NAMED **v;
93 unsigned char power;
94 size_t size;
95 size_t used;
96 const XML_Memory_Handling_Suite *mem;
97 } HASH_TABLE;
98
99 /* Basic character hash algorithm, taken from Python's string hash:
100 h = h * 1000003 ^ character, the constant being a prime number.
101
102 */
103 #ifdef XML_UNICODE
104 #define CHAR_HASH(h, c) \
105 (((h) * 0xF4243) ^ (unsigned short)(c))
106 #else
107 #define CHAR_HASH(h, c) \
108 (((h) * 0xF4243) ^ (unsigned char)(c))
109 #endif
110
111 /* For probing (after a collision) we need a step size relative prime
112 to the hash table size, which is a power of 2. We use double-hashing,
113 since we can calculate a second hash value cheaply by taking those bits
114 of the first hash value that were discarded (masked out) when the table
115 index was calculated: index = hash & mask, where mask = table->size - 1.
116 We limit the maximum step size to table->size / 4 (mask >> 2) and make
117 it odd, since odd numbers are always relative prime to a power of 2.
118 */
119 #define SECOND_HASH(hash, mask, power) \
120 ((((hash) & ~(mask)) >> ((power) - 1)) & ((mask) >> 2))
121 #define PROBE_STEP(hash, mask, power) \
122 ((unsigned char)((SECOND_HASH(hash, mask, power)) | 1))
123
124 typedef struct {
125 NAMED **p;
126 NAMED **end;
127 } HASH_TABLE_ITER;
128
129 #define INIT_TAG_BUF_SIZE 32 /* must be a multiple of sizeof(XML_Char) */
130 #define INIT_DATA_BUF_SIZE 1024
131 #define INIT_ATTS_SIZE 16
132 #define INIT_ATTS_VERSION 0xFFFFFFFF
133 #define INIT_BLOCK_SIZE 1024
134 #define INIT_BUFFER_SIZE 1024
135
136 #define EXPAND_SPARE 24
137
138 typedef struct binding {
139 struct prefix *prefix;
140 struct binding *nextTagBinding;
141 struct binding *prevPrefixBinding;
142 const struct attribute_id *attId;
143 XML_Char *uri;
144 int uriLen;
145 int uriAlloc;
146 } BINDING;
147
148 typedef struct prefix {
149 const XML_Char *name;
150 BINDING *binding;
151 } PREFIX;
152
153 typedef struct {
154 const XML_Char *str;
155 const XML_Char *localPart;
156 const XML_Char *prefix;
157 int strLen;
158 int uriLen;
159 int prefixLen;
160 } TAG_NAME;
161
162 /* TAG represents an open element.
163 The name of the element is stored in both the document and API
164 encodings. The memory buffer 'buf' is a separately-allocated
165 memory area which stores the name. During the XML_Parse()/
166 XMLParseBuffer() when the element is open, the memory for the 'raw'
167 version of the name (in the document encoding) is shared with the
168 document buffer. If the element is open across calls to
169 XML_Parse()/XML_ParseBuffer(), the buffer is re-allocated to
170 contain the 'raw' name as well.
171
172 A parser re-uses these structures, maintaining a list of allocated
173 TAG objects in a free list.
174 */
175 typedef struct tag {
176 struct tag *parent; /* parent of this element */
177 const char *rawName; /* tagName in the original encoding */
178 int rawNameLength;
179 TAG_NAME name; /* tagName in the API encoding */
180 char *buf; /* buffer for name components */
181 char *bufEnd; /* end of the buffer */
182 BINDING *bindings;
183 } TAG;
184
185 typedef struct {
186 const XML_Char *name;
187 const XML_Char *textPtr;
188 int textLen; /* length in XML_Chars */
189 int processed; /* # of processed bytes - when suspended */
190 const XML_Char *systemId;
191 const XML_Char *base;
192 const XML_Char *publicId;
193 const XML_Char *notation;
194 XML_Bool open;
195 XML_Bool is_param;
196 XML_Bool is_internal; /* true if declared in internal subset outside PE */
197 } ENTITY;
198
199 typedef struct {
200 enum XML_Content_Type type;
201 enum XML_Content_Quant quant;
202 const XML_Char * name;
203 int firstchild;
204 int lastchild;
205 int childcnt;
206 int nextsib;
207 } CONTENT_SCAFFOLD;
208
209 #define INIT_SCAFFOLD_ELEMENTS 32
210
211 typedef struct block {
212 struct block *next;
213 int size;
214 XML_Char s[1];
215 } BLOCK;
216
217 typedef struct {
218 BLOCK *blocks;
219 BLOCK *freeBlocks;
220 const XML_Char *end;
221 XML_Char *ptr;
222 XML_Char *start;
223 const XML_Memory_Handling_Suite *mem;
224 } STRING_POOL;
225
226 /* The XML_Char before the name is used to determine whether
227 an attribute has been specified. */
228 typedef struct attribute_id {
229 XML_Char *name;
230 PREFIX *prefix;
231 XML_Bool maybeTokenized;
232 XML_Bool xmlns;
233 } ATTRIBUTE_ID;
234
235 typedef struct {
236 const ATTRIBUTE_ID *id;
237 XML_Bool isCdata;
238 const XML_Char *value;
239 } DEFAULT_ATTRIBUTE;
240
241 typedef struct {
242 unsigned long version;
243 unsigned long hash;
244 const XML_Char *uriName;
245 } NS_ATT;
246
247 typedef struct {
248 const XML_Char *name;
249 PREFIX *prefix;
250 const ATTRIBUTE_ID *idAtt;
251 int nDefaultAtts;
252 int allocDefaultAtts;
253 DEFAULT_ATTRIBUTE *defaultAtts;
254 } ELEMENT_TYPE;
255
256 typedef struct {
257 HASH_TABLE generalEntities;
258 HASH_TABLE elementTypes;
259 HASH_TABLE attributeIds;
260 HASH_TABLE prefixes;
261 STRING_POOL pool;
262 STRING_POOL entityValuePool;
263 /* false once a parameter entity reference has been skipped */
264 XML_Bool keepProcessing;
265 /* true once an internal or external PE reference has been encountered;
266 this includes the reference to an external subset */
267 XML_Bool hasParamEntityRefs;
268 XML_Bool standalone;
269 #ifdef XML_DTD
270 /* indicates if external PE has been read */
271 XML_Bool paramEntityRead;
272 HASH_TABLE paramEntities;
273 #endif /* XML_DTD */
274 PREFIX defaultPrefix;
275 /* === scaffolding for building content model === */
276 XML_Bool in_eldecl;
277 CONTENT_SCAFFOLD *scaffold;
278 unsigned contentStringLen;
279 unsigned scaffSize;
280 unsigned scaffCount;
281 int scaffLevel;
282 int *scaffIndex;
283 } DTD;
284
285 typedef struct open_internal_entity {
286 const char *internalEventPtr;
287 const char *internalEventEndPtr;
288 struct open_internal_entity *next;
289 ENTITY *entity;
290 int startTagLevel;
291 XML_Bool betweenDecl; /* WFC: PE Between Declarations */
292 } OPEN_INTERNAL_ENTITY;
293
294 typedef enum XML_Error PTRCALL Processor(XML_Parser parser,
295 const char *start,
296 const char *end,
297 const char **endPtr);
298
299 static Processor prologProcessor;
300 static Processor prologInitProcessor;
301 static Processor contentProcessor;
302 static Processor cdataSectionProcessor;
303 #ifdef XML_DTD
304 static Processor ignoreSectionProcessor;
305 static Processor externalParEntProcessor;
306 static Processor externalParEntInitProcessor;
307 static Processor entityValueProcessor;
308 static Processor entityValueInitProcessor;
309 #endif /* XML_DTD */
310 static Processor epilogProcessor;
311 static Processor errorProcessor;
312 static Processor externalEntityInitProcessor;
313 static Processor externalEntityInitProcessor2;
314 static Processor externalEntityInitProcessor3;
315 static Processor externalEntityContentProcessor;
316 static Processor internalEntityProcessor;
317
318 static enum XML_Error
319 handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName);
320 static enum XML_Error
321 processXmlDecl(XML_Parser parser, int isGeneralTextEntity,
322 const char *s, const char *next);
323 static enum XML_Error
324 initializeEncoding(XML_Parser parser);
325 static enum XML_Error
326 doProlog(XML_Parser parser, const ENCODING *enc, const char *s,
327 const char *end, int tok, const char *next, const char **nextPtr,
328 XML_Bool haveMore);
329 static enum XML_Error
330 processInternalEntity(XML_Parser parser, ENTITY *entity,
331 XML_Bool betweenDecl);
332 static enum XML_Error
333 doContent(XML_Parser parser, int startTagLevel, const ENCODING *enc,
334 const char *start, const char *end, const char **endPtr,
335 XML_Bool haveMore);
336 static enum XML_Error
337 doCdataSection(XML_Parser parser, const ENCODING *, const char **startPtr,
338 const char *end, const char **nextPtr, XML_Bool haveMore);
339 #ifdef XML_DTD
340 static enum XML_Error
341 doIgnoreSection(XML_Parser parser, const ENCODING *, const char **startPtr,
342 const char *end, const char **nextPtr, XML_Bool haveMore);
343 #endif /* XML_DTD */
344
345 static enum XML_Error
346 storeAtts(XML_Parser parser, const ENCODING *, const char *s,
347 TAG_NAME *tagNamePtr, BINDING **bindingsPtr);
348 static enum XML_Error
349 addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId,
350 const XML_Char *uri, BINDING **bindingsPtr);
351 static int
352 defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *, XML_Bool isCdata,
353 XML_Bool isId, const XML_Char *dfltValue, XML_Parser parser);
354 static enum XML_Error
355 storeAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata,
356 const char *, const char *, STRING_POOL *);
357 static enum XML_Error
358 appendAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata,
359 const char *, const char *, STRING_POOL *);
360 static ATTRIBUTE_ID *
361 getAttributeId(XML_Parser parser, const ENCODING *enc, const char *start,
362 const char *end);
363 static int
364 setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *);
365 static enum XML_Error
366 storeEntityValue(XML_Parser parser, const ENCODING *enc, const char *start,
367 const char *end);
368 static int
369 reportProcessingInstruction(XML_Parser parser, const ENCODING *enc,
370 const char *start, const char *end);
371 static int
372 reportComment(XML_Parser parser, const ENCODING *enc, const char *start,
373 const char *end);
374 static void
375 reportDefault(XML_Parser parser, const ENCODING *enc, const char *start,
376 const char *end);
377
378 static const XML_Char * getContext(XML_Parser parser);
379 static XML_Bool
380 setContext(XML_Parser parser, const XML_Char *context);
381
382 static void FASTCALL normalizePublicId(XML_Char *s);
383
384 static DTD * dtdCreate(const XML_Memory_Handling_Suite *ms);
385 /* do not call if parentParser != NULL */
386 static void dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms);
387 static void
388 dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms);
389 static int
390 dtdCopy(DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms);
391 static int
392 copyEntityTable(HASH_TABLE *, STRING_POOL *, const HASH_TABLE *);
393
394 static NAMED *
395 lookup(HASH_TABLE *table, KEY name, size_t createSize);
396 static void FASTCALL
397 hashTableInit(HASH_TABLE *, const XML_Memory_Handling_Suite *ms);
398 static void FASTCALL hashTableClear(HASH_TABLE *);
399 static void FASTCALL hashTableDestroy(HASH_TABLE *);
400 static void FASTCALL
401 hashTableIterInit(HASH_TABLE_ITER *, const HASH_TABLE *);
402 static NAMED * FASTCALL hashTableIterNext(HASH_TABLE_ITER *);
403
404 static void FASTCALL
405 poolInit(STRING_POOL *, const XML_Memory_Handling_Suite *ms);
406 static void FASTCALL poolClear(STRING_POOL *);
407 static void FASTCALL poolDestroy(STRING_POOL *);
408 static XML_Char *
409 poolAppend(STRING_POOL *pool, const ENCODING *enc,
410 const char *ptr, const char *end);
411 static XML_Char *
412 poolStoreString(STRING_POOL *pool, const ENCODING *enc,
413 const char *ptr, const char *end);
414 static XML_Bool FASTCALL poolGrow(STRING_POOL *pool);
415 static const XML_Char * FASTCALL
416 poolCopyString(STRING_POOL *pool, const XML_Char *s);
417 static const XML_Char *
418 poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n);
419 static const XML_Char * FASTCALL
420 poolAppendString(STRING_POOL *pool, const XML_Char *s);
421
422 static int FASTCALL nextScaffoldPart(XML_Parser parser);
423 static XML_Content * build_model(XML_Parser parser);
424 static ELEMENT_TYPE *
425 getElementType(XML_Parser parser, const ENCODING *enc,
426 const char *ptr, const char *end);
427
428 static XML_Parser
429 parserCreate(const XML_Char *encodingName,
430 const XML_Memory_Handling_Suite *memsuite,
431 const XML_Char *nameSep,
432 DTD *dtd);
433 static void
434 parserInit(XML_Parser parser, const XML_Char *encodingName);
435
436 #define poolStart(pool) ((pool)->start)
437 #define poolEnd(pool) ((pool)->ptr)
438 #define poolLength(pool) ((pool)->ptr - (pool)->start)
439 #define poolChop(pool) ((void)--(pool->ptr))
440 #define poolLastChar(pool) (((pool)->ptr)[-1])
441 #define poolDiscard(pool) ((pool)->ptr = (pool)->start)
442 #define poolFinish(pool) ((pool)->start = (pool)->ptr)
443 #define poolAppendChar(pool, c) \
444 (((pool)->ptr == (pool)->end && !poolGrow(pool)) \
445 ? 0 \
446 : ((*((pool)->ptr)++ = c), 1))
447
448 struct XML_ParserStruct {
449 /* The first member must be userData so that the XML_GetUserData
450 macro works. */
451 void *m_userData;
452 void *m_handlerArg;
453 char *m_buffer;
454 const XML_Memory_Handling_Suite m_mem;
455 /* first character to be parsed */
456 const char *m_bufferPtr;
457 /* past last character to be parsed */
458 char *m_bufferEnd;
459 /* allocated end of buffer */
460 const char *m_bufferLim;
461 XML_Index m_parseEndByteIndex;
462 const char *m_parseEndPtr;
463 XML_Char *m_dataBuf;
464 XML_Char *m_dataBufEnd;
465 XML_StartElementHandler m_startElementHandler;
466 XML_EndElementHandler m_endElementHandler;
467 XML_CharacterDataHandler m_characterDataHandler;
468 XML_ProcessingInstructionHandler m_processingInstructionHandler;
469 XML_CommentHandler m_commentHandler;
470 XML_StartCdataSectionHandler m_startCdataSectionHandler;
471 XML_EndCdataSectionHandler m_endCdataSectionHandler;
472 XML_DefaultHandler m_defaultHandler;
473 XML_StartDoctypeDeclHandler m_startDoctypeDeclHandler;
474 XML_EndDoctypeDeclHandler m_endDoctypeDeclHandler;
475 XML_UnparsedEntityDeclHandler m_unparsedEntityDeclHandler;
476 XML_NotationDeclHandler m_notationDeclHandler;
477 XML_StartNamespaceDeclHandler m_startNamespaceDeclHandler;
478 XML_EndNamespaceDeclHandler m_endNamespaceDeclHandler;
479 XML_NotStandaloneHandler m_notStandaloneHandler;
480 XML_ExternalEntityRefHandler m_externalEntityRefHandler;
481 XML_Parser m_externalEntityRefHandlerArg;
482 XML_SkippedEntityHandler m_skippedEntityHandler;
483 XML_UnknownEncodingHandler m_unknownEncodingHandler;
484 XML_ElementDeclHandler m_elementDeclHandler;
485 XML_AttlistDeclHandler m_attlistDeclHandler;
486 XML_EntityDeclHandler m_entityDeclHandler;
487 XML_XmlDeclHandler m_xmlDeclHandler;
488 const ENCODING *m_encoding;
489 INIT_ENCODING m_initEncoding;
490 const ENCODING *m_internalEncoding;
491 const XML_Char *m_protocolEncodingName;
492 XML_Bool m_ns;
493 XML_Bool m_ns_triplets;
494 void *m_unknownEncodingMem;
495 void *m_unknownEncodingData;
496 void *m_unknownEncodingHandlerData;
497 void (XMLCALL *m_unknownEncodingRelease)(void *);
498 PROLOG_STATE m_prologState;
499 Processor *m_processor;
500 enum XML_Error m_errorCode;
501 const char *m_eventPtr;
502 const char *m_eventEndPtr;
503 const char *m_positionPtr;
504 OPEN_INTERNAL_ENTITY *m_openInternalEntities;
505 OPEN_INTERNAL_ENTITY *m_freeInternalEntities;
506 XML_Bool m_defaultExpandInternalEntities;
507 int m_tagLevel;
508 ENTITY *m_declEntity;
509 const XML_Char *m_doctypeName;
510 const XML_Char *m_doctypeSysid;
511 const XML_Char *m_doctypePubid;
512 const XML_Char *m_declAttributeType;
513 const XML_Char *m_declNotationName;
514 const XML_Char *m_declNotationPublicId;
515 ELEMENT_TYPE *m_declElementType;
516 ATTRIBUTE_ID *m_declAttributeId;
517 XML_Bool m_declAttributeIsCdata;
518 XML_Bool m_declAttributeIsId;
519 DTD *m_dtd;
520 const XML_Char *m_curBase;
521 TAG *m_tagStack;
522 TAG *m_freeTagList;
523 BINDING *m_inheritedBindings;
524 BINDING *m_freeBindingList;
525 int m_attsSize;
526 int m_nSpecifiedAtts;
527 int m_idAttIndex;
528 ATTRIBUTE *m_atts;
529 NS_ATT *m_nsAtts;
530 unsigned long m_nsAttsVersion;
531 unsigned char m_nsAttsPower;
532 POSITION m_position;
533 STRING_POOL m_tempPool;
534 STRING_POOL m_temp2Pool;
535 char *m_groupConnector;
536 unsigned int m_groupSize;
537 XML_Char m_namespaceSeparator;
538 XML_Parser m_parentParser;
539 XML_ParsingStatus m_parsingStatus;
540 #ifdef XML_DTD
541 XML_Bool m_isParamEntity;
542 XML_Bool m_useForeignDTD;
543 enum XML_ParamEntityParsing m_paramEntityParsing;
544 #endif
545 };
546
547 #define MALLOC(s) (parser->m_mem.malloc_fcn((s)))
548 #define REALLOC(p,s) (parser->m_mem.realloc_fcn((p),(s)))
549 #define FREE(p) (parser->m_mem.free_fcn((p)))
550
551 #define userData (parser->m_userData)
552 #define handlerArg (parser->m_handlerArg)
553 #define startElementHandler (parser->m_startElementHandler)
554 #define endElementHandler (parser->m_endElementHandler)
555 #define characterDataHandler (parser->m_characterDataHandler)
556 #define processingInstructionHandler \
557 (parser->m_processingInstructionHandler)
558 #define commentHandler (parser->m_commentHandler)
559 #define startCdataSectionHandler \
560 (parser->m_startCdataSectionHandler)
561 #define endCdataSectionHandler (parser->m_endCdataSectionHandler)
562 #define defaultHandler (parser->m_defaultHandler)
563 #define startDoctypeDeclHandler (parser->m_startDoctypeDeclHandler)
564 #define endDoctypeDeclHandler (parser->m_endDoctypeDeclHandler)
565 #define unparsedEntityDeclHandler \
566 (parser->m_unparsedEntityDeclHandler)
567 #define notationDeclHandler (parser->m_notationDeclHandler)
568 #define startNamespaceDeclHandler \
569 (parser->m_startNamespaceDeclHandler)
570 #define endNamespaceDeclHandler (parser->m_endNamespaceDeclHandler)
571 #define notStandaloneHandler (parser->m_notStandaloneHandler)
572 #define externalEntityRefHandler \
573 (parser->m_externalEntityRefHandler)
574 #define externalEntityRefHandlerArg \
575 (parser->m_externalEntityRefHandlerArg)
576 #define internalEntityRefHandler \
577 (parser->m_internalEntityRefHandler)
578 #define skippedEntityHandler (parser->m_skippedEntityHandler)
579 #define unknownEncodingHandler (parser->m_unknownEncodingHandler)
580 #define elementDeclHandler (parser->m_elementDeclHandler)
581 #define attlistDeclHandler (parser->m_attlistDeclHandler)
582 #define entityDeclHandler (parser->m_entityDeclHandler)
583 #define xmlDeclHandler (parser->m_xmlDeclHandler)
584 #define encoding (parser->m_encoding)
585 #define initEncoding (parser->m_initEncoding)
586 #define internalEncoding (parser->m_internalEncoding)
587 #define unknownEncodingMem (parser->m_unknownEncodingMem)
588 #define unknownEncodingData (parser->m_unknownEncodingData)
589 #define unknownEncodingHandlerData \
590 (parser->m_unknownEncodingHandlerData)
591 #define unknownEncodingRelease (parser->m_unknownEncodingRelease)
592 #define protocolEncodingName (parser->m_protocolEncodingName)
593 #define ns (parser->m_ns)
594 #define ns_triplets (parser->m_ns_triplets)
595 #define prologState (parser->m_prologState)
596 #define processor (parser->m_processor)
597 #define errorCode (parser->m_errorCode)
598 #define eventPtr (parser->m_eventPtr)
599 #define eventEndPtr (parser->m_eventEndPtr)
600 #define positionPtr (parser->m_positionPtr)
601 #define position (parser->m_position)
602 #define openInternalEntities (parser->m_openInternalEntities)
603 #define freeInternalEntities (parser->m_freeInternalEntities)
604 #define defaultExpandInternalEntities \
605 (parser->m_defaultExpandInternalEntities)
606 #define tagLevel (parser->m_tagLevel)
607 #define buffer (parser->m_buffer)
608 #define bufferPtr (parser->m_bufferPtr)
609 #define bufferEnd (parser->m_bufferEnd)
610 #define parseEndByteIndex (parser->m_parseEndByteIndex)
611 #define parseEndPtr (parser->m_parseEndPtr)
612 #define bufferLim (parser->m_bufferLim)
613 #define dataBuf (parser->m_dataBuf)
614 #define dataBufEnd (parser->m_dataBufEnd)
615 #define _dtd (parser->m_dtd)
616 #define curBase (parser->m_curBase)
617 #define declEntity (parser->m_declEntity)
618 #define doctypeName (parser->m_doctypeName)
619 #define doctypeSysid (parser->m_doctypeSysid)
620 #define doctypePubid (parser->m_doctypePubid)
621 #define declAttributeType (parser->m_declAttributeType)
622 #define declNotationName (parser->m_declNotationName)
623 #define declNotationPublicId (parser->m_declNotationPublicId)
624 #define declElementType (parser->m_declElementType)
625 #define declAttributeId (parser->m_declAttributeId)
626 #define declAttributeIsCdata (parser->m_declAttributeIsCdata)
627 #define declAttributeIsId (parser->m_declAttributeIsId)
628 #define freeTagList (parser->m_freeTagList)
629 #define freeBindingList (parser->m_freeBindingList)
630 #define inheritedBindings (parser->m_inheritedBindings)
631 #define tagStack (parser->m_tagStack)
632 #define atts (parser->m_atts)
633 #define attsSize (parser->m_attsSize)
634 #define nSpecifiedAtts (parser->m_nSpecifiedAtts)
635 #define idAttIndex (parser->m_idAttIndex)
636 #define nsAtts (parser->m_nsAtts)
637 #define nsAttsVersion (parser->m_nsAttsVersion)
638 #define nsAttsPower (parser->m_nsAttsPower)
639 #define tempPool (parser->m_tempPool)
640 #define temp2Pool (parser->m_temp2Pool)
641 #define groupConnector (parser->m_groupConnector)
642 #define groupSize (parser->m_groupSize)
643 #define namespaceSeparator (parser->m_namespaceSeparator)
644 #define parentParser (parser->m_parentParser)
645 #define ps_parsing (parser->m_parsingStatus.parsing)
646 #define ps_finalBuffer (parser->m_parsingStatus.finalBuffer)
647 #ifdef XML_DTD
648 #define isParamEntity (parser->m_isParamEntity)
649 #define useForeignDTD (parser->m_useForeignDTD)
650 #define paramEntityParsing (parser->m_paramEntityParsing)
651 #endif /* XML_DTD */
652
653 XML_Parser XMLCALL
XML_ParserCreate(const XML_Char * encodingName)654 XML_ParserCreate(const XML_Char *encodingName)
655 {
656 return XML_ParserCreate_MM(encodingName, NULL, NULL);
657 }
658
659 XML_Parser XMLCALL
XML_ParserCreateNS(const XML_Char * encodingName,XML_Char nsSep)660 XML_ParserCreateNS(const XML_Char *encodingName, XML_Char nsSep)
661 {
662 XML_Char tmp[2];
663 *tmp = nsSep;
664 return XML_ParserCreate_MM(encodingName, NULL, tmp);
665 }
666
667 static const XML_Char implicitContext[] = {
668 'x', 'm', 'l', '=', 'h', 't', 't', 'p', ':', '/', '/',
669 'w', 'w', 'w', '.', 'w', '3', '.', 'o', 'r', 'g', '/',
670 'X', 'M', 'L', '/', '1', '9', '9', '8', '/',
671 'n', 'a', 'm', 'e', 's', 'p', 'a', 'c', 'e', '\0'
672 };
673
674 XML_Parser XMLCALL
XML_ParserCreate_MM(const XML_Char * encodingName,const XML_Memory_Handling_Suite * memsuite,const XML_Char * nameSep)675 XML_ParserCreate_MM(const XML_Char *encodingName,
676 const XML_Memory_Handling_Suite *memsuite,
677 const XML_Char *nameSep)
678 {
679 XML_Parser parser = parserCreate(encodingName, memsuite, nameSep, NULL);
680 if (parser != NULL && ns) {
681 /* implicit context only set for root parser, since child
682 parsers (i.e. external entity parsers) will inherit it
683 */
684 if (!setContext(parser, implicitContext)) {
685 XML_ParserFree(parser);
686 return NULL;
687 }
688 }
689 return parser;
690 }
691
692 static XML_Parser
parserCreate(const XML_Char * encodingName,const XML_Memory_Handling_Suite * memsuite,const XML_Char * nameSep,DTD * dtd)693 parserCreate(const XML_Char *encodingName,
694 const XML_Memory_Handling_Suite *memsuite,
695 const XML_Char *nameSep,
696 DTD *dtd)
697 {
698 XML_Parser parser;
699
700 if (memsuite) {
701 XML_Memory_Handling_Suite *mtemp;
702 parser = (XML_Parser)
703 memsuite->malloc_fcn(sizeof(struct XML_ParserStruct));
704 if (parser != NULL) {
705 mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem);
706 mtemp->malloc_fcn = memsuite->malloc_fcn;
707 mtemp->realloc_fcn = memsuite->realloc_fcn;
708 mtemp->free_fcn = memsuite->free_fcn;
709 }
710 }
711 else {
712 XML_Memory_Handling_Suite *mtemp;
713 parser = (XML_Parser)malloc(sizeof(struct XML_ParserStruct));
714 if (parser != NULL) {
715 mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem);
716 mtemp->malloc_fcn = malloc;
717 mtemp->realloc_fcn = realloc;
718 mtemp->free_fcn = free;
719 }
720 }
721
722 if (!parser)
723 return parser;
724
725 buffer = NULL;
726 bufferLim = NULL;
727
728 attsSize = INIT_ATTS_SIZE;
729 atts = (ATTRIBUTE *)MALLOC(attsSize * sizeof(ATTRIBUTE));
730 if (atts == NULL) {
731 FREE(parser);
732 return NULL;
733 }
734 dataBuf = (XML_Char *)MALLOC(INIT_DATA_BUF_SIZE * sizeof(XML_Char));
735 if (dataBuf == NULL) {
736 FREE(atts);
737 FREE(parser);
738 return NULL;
739 }
740 dataBufEnd = dataBuf + INIT_DATA_BUF_SIZE;
741
742 if (dtd)
743 _dtd = dtd;
744 else {
745 _dtd = dtdCreate(&parser->m_mem);
746 if (_dtd == NULL) {
747 FREE(dataBuf);
748 FREE(atts);
749 FREE(parser);
750 return NULL;
751 }
752 }
753
754 freeBindingList = NULL;
755 freeTagList = NULL;
756 freeInternalEntities = NULL;
757
758 groupSize = 0;
759 groupConnector = NULL;
760
761 unknownEncodingHandler = NULL;
762 unknownEncodingHandlerData = NULL;
763
764 namespaceSeparator = '!';
765 ns = XML_FALSE;
766 ns_triplets = XML_FALSE;
767
768 nsAtts = NULL;
769 nsAttsVersion = 0;
770 nsAttsPower = 0;
771
772 poolInit(&tempPool, &(parser->m_mem));
773 poolInit(&temp2Pool, &(parser->m_mem));
774 parserInit(parser, encodingName);
775
776 if (encodingName && !protocolEncodingName) {
777 XML_ParserFree(parser);
778 return NULL;
779 }
780
781 if (nameSep) {
782 ns = XML_TRUE;
783 internalEncoding = XmlGetInternalEncodingNS();
784 namespaceSeparator = *nameSep;
785 }
786 else {
787 internalEncoding = XmlGetInternalEncoding();
788 }
789
790 return parser;
791 }
792
793 static void
parserInit(XML_Parser parser,const XML_Char * encodingName)794 parserInit(XML_Parser parser, const XML_Char *encodingName)
795 {
796 processor = prologInitProcessor;
797 XmlPrologStateInit(&prologState);
798 protocolEncodingName = (encodingName != NULL
799 ? poolCopyString(&tempPool, encodingName)
800 : NULL);
801 curBase = NULL;
802 XmlInitEncoding(&initEncoding, &encoding, 0);
803 userData = NULL;
804 handlerArg = NULL;
805 startElementHandler = NULL;
806 endElementHandler = NULL;
807 characterDataHandler = NULL;
808 processingInstructionHandler = NULL;
809 commentHandler = NULL;
810 startCdataSectionHandler = NULL;
811 endCdataSectionHandler = NULL;
812 defaultHandler = NULL;
813 startDoctypeDeclHandler = NULL;
814 endDoctypeDeclHandler = NULL;
815 unparsedEntityDeclHandler = NULL;
816 notationDeclHandler = NULL;
817 startNamespaceDeclHandler = NULL;
818 endNamespaceDeclHandler = NULL;
819 notStandaloneHandler = NULL;
820 externalEntityRefHandler = NULL;
821 externalEntityRefHandlerArg = parser;
822 skippedEntityHandler = NULL;
823 elementDeclHandler = NULL;
824 attlistDeclHandler = NULL;
825 entityDeclHandler = NULL;
826 xmlDeclHandler = NULL;
827 bufferPtr = buffer;
828 bufferEnd = buffer;
829 parseEndByteIndex = 0;
830 parseEndPtr = NULL;
831 declElementType = NULL;
832 declAttributeId = NULL;
833 declEntity = NULL;
834 doctypeName = NULL;
835 doctypeSysid = NULL;
836 doctypePubid = NULL;
837 declAttributeType = NULL;
838 declNotationName = NULL;
839 declNotationPublicId = NULL;
840 declAttributeIsCdata = XML_FALSE;
841 declAttributeIsId = XML_FALSE;
842 memset(&position, 0, sizeof(POSITION));
843 errorCode = XML_ERROR_NONE;
844 eventPtr = NULL;
845 eventEndPtr = NULL;
846 positionPtr = NULL;
847 openInternalEntities = NULL;
848 defaultExpandInternalEntities = XML_TRUE;
849 tagLevel = 0;
850 tagStack = NULL;
851 inheritedBindings = NULL;
852 nSpecifiedAtts = 0;
853 unknownEncodingMem = NULL;
854 unknownEncodingRelease = NULL;
855 unknownEncodingData = NULL;
856 parentParser = NULL;
857 ps_parsing = XML_INITIALIZED;
858 #ifdef XML_DTD
859 isParamEntity = XML_FALSE;
860 useForeignDTD = XML_FALSE;
861 paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
862 #endif
863 }
864
865 /* moves list of bindings to freeBindingList */
866 static void FASTCALL
moveToFreeBindingList(XML_Parser parser,BINDING * bindings)867 moveToFreeBindingList(XML_Parser parser, BINDING *bindings)
868 {
869 while (bindings) {
870 BINDING *b = bindings;
871 bindings = bindings->nextTagBinding;
872 b->nextTagBinding = freeBindingList;
873 freeBindingList = b;
874 }
875 }
876
877 XML_Bool XMLCALL
XML_ParserReset(XML_Parser parser,const XML_Char * encodingName)878 XML_ParserReset(XML_Parser parser, const XML_Char *encodingName)
879 {
880 TAG *tStk;
881 OPEN_INTERNAL_ENTITY *openEntityList;
882 if (parentParser)
883 return XML_FALSE;
884 /* move tagStack to freeTagList */
885 tStk = tagStack;
886 while (tStk) {
887 TAG *tag = tStk;
888 tStk = tStk->parent;
889 tag->parent = freeTagList;
890 moveToFreeBindingList(parser, tag->bindings);
891 tag->bindings = NULL;
892 freeTagList = tag;
893 }
894 /* move openInternalEntities to freeInternalEntities */
895 openEntityList = openInternalEntities;
896 while (openEntityList) {
897 OPEN_INTERNAL_ENTITY *openEntity = openEntityList;
898 openEntityList = openEntity->next;
899 openEntity->next = freeInternalEntities;
900 freeInternalEntities = openEntity;
901 }
902 moveToFreeBindingList(parser, inheritedBindings);
903 FREE(unknownEncodingMem);
904 if (unknownEncodingRelease)
905 unknownEncodingRelease(unknownEncodingData);
906 poolClear(&tempPool);
907 poolClear(&temp2Pool);
908 parserInit(parser, encodingName);
909 dtdReset(_dtd, &parser->m_mem);
910 return setContext(parser, implicitContext);
911 }
912
913 enum XML_Status XMLCALL
XML_SetEncoding(XML_Parser parser,const XML_Char * encodingName)914 XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName)
915 {
916 /* Block after XML_Parse()/XML_ParseBuffer() has been called.
917 XXX There's no way for the caller to determine which of the
918 XXX possible error cases caused the XML_STATUS_ERROR return.
919 */
920 if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
921 return XML_STATUS_ERROR;
922 if (encodingName == NULL)
923 protocolEncodingName = NULL;
924 else {
925 protocolEncodingName = poolCopyString(&tempPool, encodingName);
926 if (!protocolEncodingName)
927 return XML_STATUS_ERROR;
928 }
929 return XML_STATUS_OK;
930 }
931
932 XML_Parser XMLCALL
XML_ExternalEntityParserCreate(XML_Parser oldParser,const XML_Char * context,const XML_Char * encodingName)933 XML_ExternalEntityParserCreate(XML_Parser oldParser,
934 const XML_Char *context,
935 const XML_Char *encodingName)
936 {
937 XML_Parser parser = oldParser;
938 DTD *newDtd = NULL;
939 DTD *oldDtd = _dtd;
940 XML_StartElementHandler oldStartElementHandler = startElementHandler;
941 XML_EndElementHandler oldEndElementHandler = endElementHandler;
942 XML_CharacterDataHandler oldCharacterDataHandler = characterDataHandler;
943 XML_ProcessingInstructionHandler oldProcessingInstructionHandler
944 = processingInstructionHandler;
945 XML_CommentHandler oldCommentHandler = commentHandler;
946 XML_StartCdataSectionHandler oldStartCdataSectionHandler
947 = startCdataSectionHandler;
948 XML_EndCdataSectionHandler oldEndCdataSectionHandler
949 = endCdataSectionHandler;
950 XML_DefaultHandler oldDefaultHandler = defaultHandler;
951 XML_UnparsedEntityDeclHandler oldUnparsedEntityDeclHandler
952 = unparsedEntityDeclHandler;
953 XML_NotationDeclHandler oldNotationDeclHandler = notationDeclHandler;
954 XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler
955 = startNamespaceDeclHandler;
956 XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler
957 = endNamespaceDeclHandler;
958 XML_NotStandaloneHandler oldNotStandaloneHandler = notStandaloneHandler;
959 XML_ExternalEntityRefHandler oldExternalEntityRefHandler
960 = externalEntityRefHandler;
961 XML_SkippedEntityHandler oldSkippedEntityHandler = skippedEntityHandler;
962 XML_UnknownEncodingHandler oldUnknownEncodingHandler
963 = unknownEncodingHandler;
964 XML_ElementDeclHandler oldElementDeclHandler = elementDeclHandler;
965 XML_AttlistDeclHandler oldAttlistDeclHandler = attlistDeclHandler;
966 XML_EntityDeclHandler oldEntityDeclHandler = entityDeclHandler;
967 XML_XmlDeclHandler oldXmlDeclHandler = xmlDeclHandler;
968 ELEMENT_TYPE * oldDeclElementType = declElementType;
969
970 void *oldUserData = userData;
971 void *oldHandlerArg = handlerArg;
972 XML_Bool oldDefaultExpandInternalEntities = defaultExpandInternalEntities;
973 XML_Parser oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg;
974 #ifdef XML_DTD
975 enum XML_ParamEntityParsing oldParamEntityParsing = paramEntityParsing;
976 int oldInEntityValue = prologState.inEntityValue;
977 #endif
978 XML_Bool oldns_triplets = ns_triplets;
979
980 #ifdef XML_DTD
981 if (!context)
982 newDtd = oldDtd;
983 #endif /* XML_DTD */
984
985 /* Note that the magical uses of the pre-processor to make field
986 access look more like C++ require that `parser' be overwritten
987 here. This makes this function more painful to follow than it
988 would be otherwise.
989 */
990 if (ns) {
991 XML_Char tmp[2];
992 *tmp = namespaceSeparator;
993 parser = parserCreate(encodingName, &parser->m_mem, tmp, newDtd);
994 }
995 else {
996 parser = parserCreate(encodingName, &parser->m_mem, NULL, newDtd);
997 }
998
999 if (!parser)
1000 return NULL;
1001
1002 startElementHandler = oldStartElementHandler;
1003 endElementHandler = oldEndElementHandler;
1004 characterDataHandler = oldCharacterDataHandler;
1005 processingInstructionHandler = oldProcessingInstructionHandler;
1006 commentHandler = oldCommentHandler;
1007 startCdataSectionHandler = oldStartCdataSectionHandler;
1008 endCdataSectionHandler = oldEndCdataSectionHandler;
1009 defaultHandler = oldDefaultHandler;
1010 unparsedEntityDeclHandler = oldUnparsedEntityDeclHandler;
1011 notationDeclHandler = oldNotationDeclHandler;
1012 startNamespaceDeclHandler = oldStartNamespaceDeclHandler;
1013 endNamespaceDeclHandler = oldEndNamespaceDeclHandler;
1014 notStandaloneHandler = oldNotStandaloneHandler;
1015 externalEntityRefHandler = oldExternalEntityRefHandler;
1016 skippedEntityHandler = oldSkippedEntityHandler;
1017 unknownEncodingHandler = oldUnknownEncodingHandler;
1018 elementDeclHandler = oldElementDeclHandler;
1019 attlistDeclHandler = oldAttlistDeclHandler;
1020 entityDeclHandler = oldEntityDeclHandler;
1021 xmlDeclHandler = oldXmlDeclHandler;
1022 declElementType = oldDeclElementType;
1023 userData = oldUserData;
1024 if (oldUserData == oldHandlerArg)
1025 handlerArg = userData;
1026 else
1027 handlerArg = parser;
1028 if (oldExternalEntityRefHandlerArg != oldParser)
1029 externalEntityRefHandlerArg = oldExternalEntityRefHandlerArg;
1030 defaultExpandInternalEntities = oldDefaultExpandInternalEntities;
1031 ns_triplets = oldns_triplets;
1032 parentParser = oldParser;
1033 #ifdef XML_DTD
1034 paramEntityParsing = oldParamEntityParsing;
1035 prologState.inEntityValue = oldInEntityValue;
1036 if (context) {
1037 #endif /* XML_DTD */
1038 if (!dtdCopy(_dtd, oldDtd, &parser->m_mem)
1039 || !setContext(parser, context)) {
1040 XML_ParserFree(parser);
1041 return NULL;
1042 }
1043 processor = externalEntityInitProcessor;
1044 #ifdef XML_DTD
1045 }
1046 else {
1047 /* The DTD instance referenced by _dtd is shared between the document's
1048 root parser and external PE parsers, therefore one does not need to
1049 call setContext. In addition, one also *must* not call setContext,
1050 because this would overwrite existing prefix->binding pointers in
1051 _dtd with ones that get destroyed with the external PE parser.
1052 This would leave those prefixes with dangling pointers.
1053 */
1054 isParamEntity = XML_TRUE;
1055 XmlPrologStateInitExternalEntity(&prologState);
1056 processor = externalParEntInitProcessor;
1057 }
1058 #endif /* XML_DTD */
1059 return parser;
1060 }
1061
1062 static void FASTCALL
destroyBindings(BINDING * bindings,XML_Parser parser)1063 destroyBindings(BINDING *bindings, XML_Parser parser)
1064 {
1065 for (;;) {
1066 BINDING *b = bindings;
1067 if (!b)
1068 break;
1069 bindings = b->nextTagBinding;
1070 FREE(b->uri);
1071 FREE(b);
1072 }
1073 }
1074
1075 void XMLCALL
XML_ParserFree(XML_Parser parser)1076 XML_ParserFree(XML_Parser parser)
1077 {
1078 TAG *tagList;
1079 OPEN_INTERNAL_ENTITY *entityList;
1080 if (parser == NULL)
1081 return;
1082 /* free tagStack and freeTagList */
1083 tagList = tagStack;
1084 for (;;) {
1085 TAG *p;
1086 if (tagList == NULL) {
1087 if (freeTagList == NULL)
1088 break;
1089 tagList = freeTagList;
1090 freeTagList = NULL;
1091 }
1092 p = tagList;
1093 tagList = tagList->parent;
1094 FREE(p->buf);
1095 destroyBindings(p->bindings, parser);
1096 FREE(p);
1097 }
1098 /* free openInternalEntities and freeInternalEntities */
1099 entityList = openInternalEntities;
1100 for (;;) {
1101 OPEN_INTERNAL_ENTITY *openEntity;
1102 if (entityList == NULL) {
1103 if (freeInternalEntities == NULL)
1104 break;
1105 entityList = freeInternalEntities;
1106 freeInternalEntities = NULL;
1107 }
1108 openEntity = entityList;
1109 entityList = entityList->next;
1110 FREE(openEntity);
1111 }
1112
1113 destroyBindings(freeBindingList, parser);
1114 destroyBindings(inheritedBindings, parser);
1115 poolDestroy(&tempPool);
1116 poolDestroy(&temp2Pool);
1117 #ifdef XML_DTD
1118 /* external parameter entity parsers share the DTD structure
1119 parser->m_dtd with the root parser, so we must not destroy it
1120 */
1121 if (!isParamEntity && _dtd)
1122 #else
1123 if (_dtd)
1124 #endif /* XML_DTD */
1125 dtdDestroy(_dtd, (XML_Bool)!parentParser, &parser->m_mem);
1126 FREE((void *)atts);
1127 FREE(groupConnector);
1128 FREE(buffer);
1129 FREE(dataBuf);
1130 FREE(nsAtts);
1131 FREE(unknownEncodingMem);
1132 if (unknownEncodingRelease)
1133 unknownEncodingRelease(unknownEncodingData);
1134 FREE(parser);
1135 }
1136
1137 void XMLCALL
XML_UseParserAsHandlerArg(XML_Parser parser)1138 XML_UseParserAsHandlerArg(XML_Parser parser)
1139 {
1140 handlerArg = parser;
1141 }
1142
1143 enum XML_Error XMLCALL
XML_UseForeignDTD(XML_Parser parser,XML_Bool useDTD)1144 XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD)
1145 {
1146 #ifdef XML_DTD
1147 /* block after XML_Parse()/XML_ParseBuffer() has been called */
1148 if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
1149 return XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING;
1150 useForeignDTD = useDTD;
1151 return XML_ERROR_NONE;
1152 #else
1153 return XML_ERROR_FEATURE_REQUIRES_XML_DTD;
1154 #endif
1155 }
1156
1157 void XMLCALL
XML_SetReturnNSTriplet(XML_Parser parser,int do_nst)1158 XML_SetReturnNSTriplet(XML_Parser parser, int do_nst)
1159 {
1160 /* block after XML_Parse()/XML_ParseBuffer() has been called */
1161 if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
1162 return;
1163 ns_triplets = do_nst ? XML_TRUE : XML_FALSE;
1164 }
1165
1166 void XMLCALL
XML_SetUserData(XML_Parser parser,void * p)1167 XML_SetUserData(XML_Parser parser, void *p)
1168 {
1169 if (handlerArg == userData)
1170 handlerArg = userData = p;
1171 else
1172 userData = p;
1173 }
1174
1175 enum XML_Status XMLCALL
XML_SetBase(XML_Parser parser,const XML_Char * p)1176 XML_SetBase(XML_Parser parser, const XML_Char *p)
1177 {
1178 if (p) {
1179 p = poolCopyString(&_dtd->pool, p);
1180 if (!p)
1181 return XML_STATUS_ERROR;
1182 curBase = p;
1183 }
1184 else
1185 curBase = NULL;
1186 return XML_STATUS_OK;
1187 }
1188
1189 const XML_Char * XMLCALL
XML_GetBase(XML_Parser parser)1190 XML_GetBase(XML_Parser parser)
1191 {
1192 return curBase;
1193 }
1194
1195 int XMLCALL
XML_GetSpecifiedAttributeCount(XML_Parser parser)1196 XML_GetSpecifiedAttributeCount(XML_Parser parser)
1197 {
1198 return nSpecifiedAtts;
1199 }
1200
1201 int XMLCALL
XML_GetIdAttributeIndex(XML_Parser parser)1202 XML_GetIdAttributeIndex(XML_Parser parser)
1203 {
1204 return idAttIndex;
1205 }
1206
1207 void XMLCALL
XML_SetElementHandler(XML_Parser parser,XML_StartElementHandler start,XML_EndElementHandler end)1208 XML_SetElementHandler(XML_Parser parser,
1209 XML_StartElementHandler start,
1210 XML_EndElementHandler end)
1211 {
1212 startElementHandler = start;
1213 endElementHandler = end;
1214 }
1215
1216 void XMLCALL
XML_SetStartElementHandler(XML_Parser parser,XML_StartElementHandler start)1217 XML_SetStartElementHandler(XML_Parser parser,
1218 XML_StartElementHandler start) {
1219 startElementHandler = start;
1220 }
1221
1222 void XMLCALL
XML_SetEndElementHandler(XML_Parser parser,XML_EndElementHandler end)1223 XML_SetEndElementHandler(XML_Parser parser,
1224 XML_EndElementHandler end) {
1225 endElementHandler = end;
1226 }
1227
1228 void XMLCALL
XML_SetCharacterDataHandler(XML_Parser parser,XML_CharacterDataHandler handler)1229 XML_SetCharacterDataHandler(XML_Parser parser,
1230 XML_CharacterDataHandler handler)
1231 {
1232 characterDataHandler = handler;
1233 }
1234
1235 void XMLCALL
XML_SetProcessingInstructionHandler(XML_Parser parser,XML_ProcessingInstructionHandler handler)1236 XML_SetProcessingInstructionHandler(XML_Parser parser,
1237 XML_ProcessingInstructionHandler handler)
1238 {
1239 processingInstructionHandler = handler;
1240 }
1241
1242 void XMLCALL
XML_SetCommentHandler(XML_Parser parser,XML_CommentHandler handler)1243 XML_SetCommentHandler(XML_Parser parser,
1244 XML_CommentHandler handler)
1245 {
1246 commentHandler = handler;
1247 }
1248
1249 void XMLCALL
XML_SetCdataSectionHandler(XML_Parser parser,XML_StartCdataSectionHandler start,XML_EndCdataSectionHandler end)1250 XML_SetCdataSectionHandler(XML_Parser parser,
1251 XML_StartCdataSectionHandler start,
1252 XML_EndCdataSectionHandler end)
1253 {
1254 startCdataSectionHandler = start;
1255 endCdataSectionHandler = end;
1256 }
1257
1258 void XMLCALL
XML_SetStartCdataSectionHandler(XML_Parser parser,XML_StartCdataSectionHandler start)1259 XML_SetStartCdataSectionHandler(XML_Parser parser,
1260 XML_StartCdataSectionHandler start) {
1261 startCdataSectionHandler = start;
1262 }
1263
1264 void XMLCALL
XML_SetEndCdataSectionHandler(XML_Parser parser,XML_EndCdataSectionHandler end)1265 XML_SetEndCdataSectionHandler(XML_Parser parser,
1266 XML_EndCdataSectionHandler end) {
1267 endCdataSectionHandler = end;
1268 }
1269
1270 void XMLCALL
XML_SetDefaultHandler(XML_Parser parser,XML_DefaultHandler handler)1271 XML_SetDefaultHandler(XML_Parser parser,
1272 XML_DefaultHandler handler)
1273 {
1274 defaultHandler = handler;
1275 defaultExpandInternalEntities = XML_FALSE;
1276 }
1277
1278 void XMLCALL
XML_SetDefaultHandlerExpand(XML_Parser parser,XML_DefaultHandler handler)1279 XML_SetDefaultHandlerExpand(XML_Parser parser,
1280 XML_DefaultHandler handler)
1281 {
1282 defaultHandler = handler;
1283 defaultExpandInternalEntities = XML_TRUE;
1284 }
1285
1286 void XMLCALL
XML_SetDoctypeDeclHandler(XML_Parser parser,XML_StartDoctypeDeclHandler start,XML_EndDoctypeDeclHandler end)1287 XML_SetDoctypeDeclHandler(XML_Parser parser,
1288 XML_StartDoctypeDeclHandler start,
1289 XML_EndDoctypeDeclHandler end)
1290 {
1291 startDoctypeDeclHandler = start;
1292 endDoctypeDeclHandler = end;
1293 }
1294
1295 void XMLCALL
XML_SetStartDoctypeDeclHandler(XML_Parser parser,XML_StartDoctypeDeclHandler start)1296 XML_SetStartDoctypeDeclHandler(XML_Parser parser,
1297 XML_StartDoctypeDeclHandler start) {
1298 startDoctypeDeclHandler = start;
1299 }
1300
1301 void XMLCALL
XML_SetEndDoctypeDeclHandler(XML_Parser parser,XML_EndDoctypeDeclHandler end)1302 XML_SetEndDoctypeDeclHandler(XML_Parser parser,
1303 XML_EndDoctypeDeclHandler end) {
1304 endDoctypeDeclHandler = end;
1305 }
1306
1307 void XMLCALL
XML_SetUnparsedEntityDeclHandler(XML_Parser parser,XML_UnparsedEntityDeclHandler handler)1308 XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
1309 XML_UnparsedEntityDeclHandler handler)
1310 {
1311 unparsedEntityDeclHandler = handler;
1312 }
1313
1314 void XMLCALL
XML_SetNotationDeclHandler(XML_Parser parser,XML_NotationDeclHandler handler)1315 XML_SetNotationDeclHandler(XML_Parser parser,
1316 XML_NotationDeclHandler handler)
1317 {
1318 notationDeclHandler = handler;
1319 }
1320
1321 void XMLCALL
XML_SetNamespaceDeclHandler(XML_Parser parser,XML_StartNamespaceDeclHandler start,XML_EndNamespaceDeclHandler end)1322 XML_SetNamespaceDeclHandler(XML_Parser parser,
1323 XML_StartNamespaceDeclHandler start,
1324 XML_EndNamespaceDeclHandler end)
1325 {
1326 startNamespaceDeclHandler = start;
1327 endNamespaceDeclHandler = end;
1328 }
1329
1330 void XMLCALL
XML_SetStartNamespaceDeclHandler(XML_Parser parser,XML_StartNamespaceDeclHandler start)1331 XML_SetStartNamespaceDeclHandler(XML_Parser parser,
1332 XML_StartNamespaceDeclHandler start) {
1333 startNamespaceDeclHandler = start;
1334 }
1335
1336 void XMLCALL
XML_SetEndNamespaceDeclHandler(XML_Parser parser,XML_EndNamespaceDeclHandler end)1337 XML_SetEndNamespaceDeclHandler(XML_Parser parser,
1338 XML_EndNamespaceDeclHandler end) {
1339 endNamespaceDeclHandler = end;
1340 }
1341
1342 void XMLCALL
XML_SetNotStandaloneHandler(XML_Parser parser,XML_NotStandaloneHandler handler)1343 XML_SetNotStandaloneHandler(XML_Parser parser,
1344 XML_NotStandaloneHandler handler)
1345 {
1346 notStandaloneHandler = handler;
1347 }
1348
1349 void XMLCALL
XML_SetExternalEntityRefHandler(XML_Parser parser,XML_ExternalEntityRefHandler handler)1350 XML_SetExternalEntityRefHandler(XML_Parser parser,
1351 XML_ExternalEntityRefHandler handler)
1352 {
1353 externalEntityRefHandler = handler;
1354 }
1355
1356 void XMLCALL
XML_SetExternalEntityRefHandlerArg(XML_Parser parser,void * arg)1357 XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg)
1358 {
1359 if (arg)
1360 externalEntityRefHandlerArg = (XML_Parser)arg;
1361 else
1362 externalEntityRefHandlerArg = parser;
1363 }
1364
1365 void XMLCALL
XML_SetSkippedEntityHandler(XML_Parser parser,XML_SkippedEntityHandler handler)1366 XML_SetSkippedEntityHandler(XML_Parser parser,
1367 XML_SkippedEntityHandler handler)
1368 {
1369 skippedEntityHandler = handler;
1370 }
1371
1372 void XMLCALL
XML_SetUnknownEncodingHandler(XML_Parser parser,XML_UnknownEncodingHandler handler,void * data)1373 XML_SetUnknownEncodingHandler(XML_Parser parser,
1374 XML_UnknownEncodingHandler handler,
1375 void *data)
1376 {
1377 unknownEncodingHandler = handler;
1378 unknownEncodingHandlerData = data;
1379 }
1380
1381 void XMLCALL
XML_SetElementDeclHandler(XML_Parser parser,XML_ElementDeclHandler eldecl)1382 XML_SetElementDeclHandler(XML_Parser parser,
1383 XML_ElementDeclHandler eldecl)
1384 {
1385 elementDeclHandler = eldecl;
1386 }
1387
1388 void XMLCALL
XML_SetAttlistDeclHandler(XML_Parser parser,XML_AttlistDeclHandler attdecl)1389 XML_SetAttlistDeclHandler(XML_Parser parser,
1390 XML_AttlistDeclHandler attdecl)
1391 {
1392 attlistDeclHandler = attdecl;
1393 }
1394
1395 void XMLCALL
XML_SetEntityDeclHandler(XML_Parser parser,XML_EntityDeclHandler handler)1396 XML_SetEntityDeclHandler(XML_Parser parser,
1397 XML_EntityDeclHandler handler)
1398 {
1399 entityDeclHandler = handler;
1400 }
1401
1402 void XMLCALL
XML_SetXmlDeclHandler(XML_Parser parser,XML_XmlDeclHandler handler)1403 XML_SetXmlDeclHandler(XML_Parser parser,
1404 XML_XmlDeclHandler handler) {
1405 xmlDeclHandler = handler;
1406 }
1407
1408 int XMLCALL
XML_SetParamEntityParsing(XML_Parser parser,enum XML_ParamEntityParsing peParsing)1409 XML_SetParamEntityParsing(XML_Parser parser,
1410 enum XML_ParamEntityParsing peParsing)
1411 {
1412 /* block after XML_Parse()/XML_ParseBuffer() has been called */
1413 if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
1414 return 0;
1415 #ifdef XML_DTD
1416 paramEntityParsing = peParsing;
1417 return 1;
1418 #else
1419 return peParsing == XML_PARAM_ENTITY_PARSING_NEVER;
1420 #endif
1421 }
1422
1423 enum XML_Status XMLCALL
XML_Parse(XML_Parser parser,const char * s,int len,int isFinal)1424 XML_Parse(XML_Parser parser, const char *s, int len, int isFinal)
1425 {
1426 switch (ps_parsing) {
1427 case XML_SUSPENDED:
1428 errorCode = XML_ERROR_SUSPENDED;
1429 return XML_STATUS_ERROR;
1430 case XML_FINISHED:
1431 errorCode = XML_ERROR_FINISHED;
1432 return XML_STATUS_ERROR;
1433 default:
1434 ps_parsing = XML_PARSING;
1435 }
1436
1437 if (len == 0) {
1438 ps_finalBuffer = (XML_Bool)isFinal;
1439 if (!isFinal)
1440 return XML_STATUS_OK;
1441 positionPtr = bufferPtr;
1442 parseEndPtr = bufferEnd;
1443
1444 /* If data are left over from last buffer, and we now know that these
1445 data are the final chunk of input, then we have to check them again
1446 to detect errors based on that fact.
1447 */
1448 errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr);
1449
1450 if (errorCode == XML_ERROR_NONE) {
1451 switch (ps_parsing) {
1452 case XML_SUSPENDED:
1453 XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
1454 positionPtr = bufferPtr;
1455 return XML_STATUS_SUSPENDED;
1456 case XML_INITIALIZED:
1457 case XML_PARSING:
1458 ps_parsing = XML_FINISHED;
1459 /* fall through */
1460 default:
1461 return XML_STATUS_OK;
1462 }
1463 }
1464 eventEndPtr = eventPtr;
1465 processor = errorProcessor;
1466 return XML_STATUS_ERROR;
1467 }
1468 #ifndef XML_CONTEXT_BYTES
1469 else if (bufferPtr == bufferEnd) {
1470 const char *end;
1471 int nLeftOver;
1472 enum XML_Error result;
1473 parseEndByteIndex += len;
1474 positionPtr = s;
1475 ps_finalBuffer = (XML_Bool)isFinal;
1476
1477 errorCode = processor(parser, s, parseEndPtr = s + len, &end);
1478
1479 if (errorCode != XML_ERROR_NONE) {
1480 eventEndPtr = eventPtr;
1481 processor = errorProcessor;
1482 return XML_STATUS_ERROR;
1483 }
1484 else {
1485 switch (ps_parsing) {
1486 case XML_SUSPENDED:
1487 result = XML_STATUS_SUSPENDED;
1488 break;
1489 case XML_INITIALIZED:
1490 case XML_PARSING:
1491 result = XML_STATUS_OK;
1492 if (isFinal) {
1493 ps_parsing = XML_FINISHED;
1494 return result;
1495 }
1496 }
1497 }
1498
1499 XmlUpdatePosition(encoding, positionPtr, end, &position);
1500 nLeftOver = s + len - end;
1501 if (nLeftOver) {
1502 if (buffer == NULL || nLeftOver > bufferLim - buffer) {
1503 /* FIXME avoid integer overflow */
1504 char *temp;
1505 temp = (buffer == NULL
1506 ? (char *)MALLOC(len * 2)
1507 : (char *)REALLOC(buffer, len * 2));
1508 if (temp == NULL) {
1509 errorCode = XML_ERROR_NO_MEMORY;
1510 return XML_STATUS_ERROR;
1511 }
1512 buffer = temp;
1513 if (!buffer) {
1514 errorCode = XML_ERROR_NO_MEMORY;
1515 eventPtr = eventEndPtr = NULL;
1516 processor = errorProcessor;
1517 return XML_STATUS_ERROR;
1518 }
1519 bufferLim = buffer + len * 2;
1520 }
1521 memcpy(buffer, end, nLeftOver);
1522 }
1523 bufferPtr = buffer;
1524 bufferEnd = buffer + nLeftOver;
1525 positionPtr = bufferPtr;
1526 parseEndPtr = bufferEnd;
1527 eventPtr = bufferPtr;
1528 eventEndPtr = bufferPtr;
1529 return result;
1530 }
1531 #endif /* not defined XML_CONTEXT_BYTES */
1532 else {
1533 void *buff = XML_GetBuffer(parser, len);
1534 if (buff == NULL)
1535 return XML_STATUS_ERROR;
1536 else {
1537 memcpy(buff, s, len);
1538 return XML_ParseBuffer(parser, len, isFinal);
1539 }
1540 }
1541 }
1542
1543 enum XML_Status XMLCALL
XML_ParseBuffer(XML_Parser parser,int len,int isFinal)1544 XML_ParseBuffer(XML_Parser parser, int len, int isFinal)
1545 {
1546 const char *start;
1547 enum XML_Status result = XML_STATUS_OK;
1548
1549 switch (ps_parsing) {
1550 case XML_SUSPENDED:
1551 errorCode = XML_ERROR_SUSPENDED;
1552 return XML_STATUS_ERROR;
1553 case XML_FINISHED:
1554 errorCode = XML_ERROR_FINISHED;
1555 return XML_STATUS_ERROR;
1556 default:
1557 ps_parsing = XML_PARSING;
1558 }
1559
1560 start = bufferPtr;
1561 positionPtr = start;
1562 bufferEnd += len;
1563 parseEndPtr = bufferEnd;
1564 parseEndByteIndex += len;
1565 ps_finalBuffer = (XML_Bool)isFinal;
1566
1567 errorCode = processor(parser, start, parseEndPtr, &bufferPtr);
1568
1569 if (errorCode != XML_ERROR_NONE) {
1570 eventEndPtr = eventPtr;
1571 processor = errorProcessor;
1572 return XML_STATUS_ERROR;
1573 }
1574 else {
1575 switch (ps_parsing) {
1576 case XML_SUSPENDED:
1577 result = XML_STATUS_SUSPENDED;
1578 break;
1579 case XML_INITIALIZED:
1580 case XML_PARSING:
1581 if (isFinal) {
1582 ps_parsing = XML_FINISHED;
1583 return result;
1584 }
1585 default: ; /* should not happen */
1586 }
1587 }
1588
1589 XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
1590 positionPtr = bufferPtr;
1591 return result;
1592 }
1593
1594 void * XMLCALL
XML_GetBuffer(XML_Parser parser,int len)1595 XML_GetBuffer(XML_Parser parser, int len)
1596 {
1597 switch (ps_parsing) {
1598 case XML_SUSPENDED:
1599 errorCode = XML_ERROR_SUSPENDED;
1600 return NULL;
1601 case XML_FINISHED:
1602 errorCode = XML_ERROR_FINISHED;
1603 return NULL;
1604 default: ;
1605 }
1606
1607 if (len > bufferLim - bufferEnd) {
1608 /* FIXME avoid integer overflow */
1609 int neededSize = len + (int)(bufferEnd - bufferPtr);
1610 #ifdef XML_CONTEXT_BYTES
1611 int keep = (int)(bufferPtr - buffer);
1612
1613 if (keep > XML_CONTEXT_BYTES)
1614 keep = XML_CONTEXT_BYTES;
1615 neededSize += keep;
1616 #endif /* defined XML_CONTEXT_BYTES */
1617 if (neededSize <= bufferLim - buffer) {
1618 #ifdef XML_CONTEXT_BYTES
1619 if (keep < bufferPtr - buffer) {
1620 int offset = (int)(bufferPtr - buffer) - keep;
1621 memmove(buffer, &buffer[offset], bufferEnd - bufferPtr + keep);
1622 bufferEnd -= offset;
1623 bufferPtr -= offset;
1624 }
1625 #else
1626 memmove(buffer, bufferPtr, bufferEnd - bufferPtr);
1627 bufferEnd = buffer + (bufferEnd - bufferPtr);
1628 bufferPtr = buffer;
1629 #endif /* not defined XML_CONTEXT_BYTES */
1630 }
1631 else {
1632 char *newBuf;
1633 int bufferSize = (int)(bufferLim - bufferPtr);
1634 if (bufferSize == 0)
1635 bufferSize = INIT_BUFFER_SIZE;
1636 do {
1637 bufferSize *= 2;
1638 } while (bufferSize < neededSize);
1639 newBuf = (char *)MALLOC(bufferSize);
1640 if (newBuf == 0) {
1641 errorCode = XML_ERROR_NO_MEMORY;
1642 return NULL;
1643 }
1644 bufferLim = newBuf + bufferSize;
1645 #ifdef XML_CONTEXT_BYTES
1646 if (bufferPtr) {
1647 int keep = (int)(bufferPtr - buffer);
1648 if (keep > XML_CONTEXT_BYTES)
1649 keep = XML_CONTEXT_BYTES;
1650 memcpy(newBuf, &bufferPtr[-keep], bufferEnd - bufferPtr + keep);
1651 FREE(buffer);
1652 buffer = newBuf;
1653 bufferEnd = buffer + (bufferEnd - bufferPtr) + keep;
1654 bufferPtr = buffer + keep;
1655 }
1656 else {
1657 bufferEnd = newBuf + (bufferEnd - bufferPtr);
1658 bufferPtr = buffer = newBuf;
1659 }
1660 #else
1661 if (bufferPtr) {
1662 memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr);
1663 FREE(buffer);
1664 }
1665 bufferEnd = newBuf + (bufferEnd - bufferPtr);
1666 bufferPtr = buffer = newBuf;
1667 #endif /* not defined XML_CONTEXT_BYTES */
1668 }
1669 }
1670 return bufferEnd;
1671 }
1672
1673 enum XML_Status XMLCALL
XML_StopParser(XML_Parser parser,XML_Bool resumable)1674 XML_StopParser(XML_Parser parser, XML_Bool resumable)
1675 {
1676 switch (ps_parsing) {
1677 case XML_SUSPENDED:
1678 if (resumable) {
1679 errorCode = XML_ERROR_SUSPENDED;
1680 return XML_STATUS_ERROR;
1681 }
1682 ps_parsing = XML_FINISHED;
1683 break;
1684 case XML_FINISHED:
1685 errorCode = XML_ERROR_FINISHED;
1686 return XML_STATUS_ERROR;
1687 default:
1688 if (resumable) {
1689 #ifdef XML_DTD
1690 if (isParamEntity) {
1691 errorCode = XML_ERROR_SUSPEND_PE;
1692 return XML_STATUS_ERROR;
1693 }
1694 #endif
1695 ps_parsing = XML_SUSPENDED;
1696 }
1697 else
1698 ps_parsing = XML_FINISHED;
1699 }
1700 return XML_STATUS_OK;
1701 }
1702
1703 enum XML_Status XMLCALL
XML_ResumeParser(XML_Parser parser)1704 XML_ResumeParser(XML_Parser parser)
1705 {
1706 enum XML_Status result = XML_STATUS_OK;
1707
1708 if (ps_parsing != XML_SUSPENDED) {
1709 errorCode = XML_ERROR_NOT_SUSPENDED;
1710 return XML_STATUS_ERROR;
1711 }
1712 ps_parsing = XML_PARSING;
1713
1714 errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr);
1715
1716 if (errorCode != XML_ERROR_NONE) {
1717 eventEndPtr = eventPtr;
1718 processor = errorProcessor;
1719 return XML_STATUS_ERROR;
1720 }
1721 else {
1722 switch (ps_parsing) {
1723 case XML_SUSPENDED:
1724 result = XML_STATUS_SUSPENDED;
1725 break;
1726 case XML_INITIALIZED:
1727 case XML_PARSING:
1728 if (ps_finalBuffer) {
1729 ps_parsing = XML_FINISHED;
1730 return result;
1731 }
1732 default: ;
1733 }
1734 }
1735
1736 XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
1737 positionPtr = bufferPtr;
1738 return result;
1739 }
1740
1741 void XMLCALL
XML_GetParsingStatus(XML_Parser parser,XML_ParsingStatus * status)1742 XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status)
1743 {
1744 assert(status != NULL);
1745 *status = parser->m_parsingStatus;
1746 }
1747
1748 enum XML_Error XMLCALL
XML_GetErrorCode(XML_Parser parser)1749 XML_GetErrorCode(XML_Parser parser)
1750 {
1751 return errorCode;
1752 }
1753
1754 XML_Index XMLCALL
XML_GetCurrentByteIndex(XML_Parser parser)1755 XML_GetCurrentByteIndex(XML_Parser parser)
1756 {
1757 if (eventPtr)
1758 return parseEndByteIndex - (parseEndPtr - eventPtr);
1759 return -1;
1760 }
1761
1762 int XMLCALL
XML_GetCurrentByteCount(XML_Parser parser)1763 XML_GetCurrentByteCount(XML_Parser parser)
1764 {
1765 if (eventEndPtr && eventPtr)
1766 return (int)(eventEndPtr - eventPtr);
1767 return 0;
1768 }
1769
1770 const char * XMLCALL
XML_GetInputContext(XML_Parser parser,int * offset,int * size)1771 XML_GetInputContext(XML_Parser parser, int *offset, int *size)
1772 {
1773 #ifdef XML_CONTEXT_BYTES
1774 if (eventPtr && buffer) {
1775 *offset = (int)(eventPtr - buffer);
1776 *size = (int)(bufferEnd - buffer);
1777 return buffer;
1778 }
1779 #endif /* defined XML_CONTEXT_BYTES */
1780 return (char *) 0;
1781 }
1782
1783 XML_Size XMLCALL
XML_GetCurrentLineNumber(XML_Parser parser)1784 XML_GetCurrentLineNumber(XML_Parser parser)
1785 {
1786 if (eventPtr && eventPtr >= positionPtr) {
1787 XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
1788 positionPtr = eventPtr;
1789 }
1790 return position.lineNumber + 1;
1791 }
1792
1793 XML_Size XMLCALL
XML_GetCurrentColumnNumber(XML_Parser parser)1794 XML_GetCurrentColumnNumber(XML_Parser parser)
1795 {
1796 if (eventPtr && eventPtr >= positionPtr) {
1797 XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
1798 positionPtr = eventPtr;
1799 }
1800 return position.columnNumber;
1801 }
1802
1803 void XMLCALL
XML_FreeContentModel(XML_Parser parser,XML_Content * model)1804 XML_FreeContentModel(XML_Parser parser, XML_Content *model)
1805 {
1806 FREE(model);
1807 }
1808
1809 void * XMLCALL
XML_MemMalloc(XML_Parser parser,size_t size)1810 XML_MemMalloc(XML_Parser parser, size_t size)
1811 {
1812 return MALLOC(size);
1813 }
1814
1815 void * XMLCALL
XML_MemRealloc(XML_Parser parser,void * ptr,size_t size)1816 XML_MemRealloc(XML_Parser parser, void *ptr, size_t size)
1817 {
1818 return REALLOC(ptr, size);
1819 }
1820
1821 void XMLCALL
XML_MemFree(XML_Parser parser,void * ptr)1822 XML_MemFree(XML_Parser parser, void *ptr)
1823 {
1824 FREE(ptr);
1825 }
1826
1827 void XMLCALL
XML_DefaultCurrent(XML_Parser parser)1828 XML_DefaultCurrent(XML_Parser parser)
1829 {
1830 if (defaultHandler) {
1831 if (openInternalEntities)
1832 reportDefault(parser,
1833 internalEncoding,
1834 openInternalEntities->internalEventPtr,
1835 openInternalEntities->internalEventEndPtr);
1836 else
1837 reportDefault(parser, encoding, eventPtr, eventEndPtr);
1838 }
1839 }
1840
1841 const XML_LChar * XMLCALL
XML_ErrorString(enum XML_Error code)1842 XML_ErrorString(enum XML_Error code)
1843 {
1844 static const XML_LChar* const message[] = {
1845 0,
1846 XML_L("out of memory"),
1847 XML_L("syntax error"),
1848 XML_L("no element found"),
1849 XML_L("not well-formed (invalid token)"),
1850 XML_L("unclosed token"),
1851 XML_L("partial character"),
1852 XML_L("mismatched tag"),
1853 XML_L("duplicate attribute"),
1854 XML_L("junk after document element"),
1855 XML_L("illegal parameter entity reference"),
1856 XML_L("undefined entity"),
1857 XML_L("recursive entity reference"),
1858 XML_L("asynchronous entity"),
1859 XML_L("reference to invalid character number"),
1860 XML_L("reference to binary entity"),
1861 XML_L("reference to external entity in attribute"),
1862 XML_L("XML or text declaration not at start of entity"),
1863 XML_L("unknown encoding"),
1864 XML_L("encoding specified in XML declaration is incorrect"),
1865 XML_L("unclosed CDATA section"),
1866 XML_L("error in processing external entity reference"),
1867 XML_L("document is not standalone"),
1868 XML_L("unexpected parser state - please send a bug report"),
1869 XML_L("entity declared in parameter entity"),
1870 XML_L("requested feature requires XML_DTD support in Expat"),
1871 XML_L("cannot change setting once parsing has begun"),
1872 XML_L("unbound prefix"),
1873 XML_L("must not undeclare prefix"),
1874 XML_L("incomplete markup in parameter entity"),
1875 XML_L("XML declaration not well-formed"),
1876 XML_L("text declaration not well-formed"),
1877 XML_L("illegal character(s) in public id"),
1878 XML_L("parser suspended"),
1879 XML_L("parser not suspended"),
1880 XML_L("parsing aborted"),
1881 XML_L("parsing finished"),
1882 XML_L("cannot suspend in external parameter entity"),
1883 XML_L("reserved prefix (xml) must not be undeclared or bound to another namespace name"),
1884 XML_L("reserved prefix (xmlns) must not be declared or undeclared"),
1885 XML_L("prefix must not be bound to one of the reserved namespace names")
1886 };
1887 if (code > 0 && code < sizeof(message)/sizeof(message[0]))
1888 return message[code];
1889 return NULL;
1890 }
1891
1892 const XML_LChar * XMLCALL
XML_ExpatVersion(void)1893 XML_ExpatVersion(void) {
1894
1895 /* V1 is used to string-ize the version number. However, it would
1896 string-ize the actual version macro *names* unless we get them
1897 substituted before being passed to V1. CPP is defined to expand
1898 a macro, then rescan for more expansions. Thus, we use V2 to expand
1899 the version macros, then CPP will expand the resulting V1() macro
1900 with the correct numerals. */
1901 /* ### I'm assuming cpp is portable in this respect... */
1902
1903 #define V1(a,b,c) XML_L(#a)XML_L(".")XML_L(#b)XML_L(".")XML_L(#c)
1904 #define V2(a,b,c) XML_L("expat_")V1(a,b,c)
1905
1906 return V2(XML_MAJOR_VERSION, XML_MINOR_VERSION, XML_MICRO_VERSION);
1907
1908 #undef V1
1909 #undef V2
1910 }
1911
1912 XML_Expat_Version XMLCALL
XML_ExpatVersionInfo(void)1913 XML_ExpatVersionInfo(void)
1914 {
1915 XML_Expat_Version version;
1916
1917 version.major = XML_MAJOR_VERSION;
1918 version.minor = XML_MINOR_VERSION;
1919 version.micro = XML_MICRO_VERSION;
1920
1921 return version;
1922 }
1923
1924 const XML_Feature * XMLCALL
XML_GetFeatureList(void)1925 XML_GetFeatureList(void)
1926 {
1927 static const XML_Feature features[] = {
1928 {XML_FEATURE_SIZEOF_XML_CHAR, XML_L("sizeof(XML_Char)"),
1929 sizeof(XML_Char)},
1930 {XML_FEATURE_SIZEOF_XML_LCHAR, XML_L("sizeof(XML_LChar)"),
1931 sizeof(XML_LChar)},
1932 #ifdef XML_UNICODE
1933 {XML_FEATURE_UNICODE, XML_L("XML_UNICODE"), 0},
1934 #endif
1935 #ifdef XML_UNICODE_WCHAR_T
1936 {XML_FEATURE_UNICODE_WCHAR_T, XML_L("XML_UNICODE_WCHAR_T"), 0},
1937 #endif
1938 #ifdef XML_DTD
1939 {XML_FEATURE_DTD, XML_L("XML_DTD"), 0},
1940 #endif
1941 #ifdef XML_CONTEXT_BYTES
1942 {XML_FEATURE_CONTEXT_BYTES, XML_L("XML_CONTEXT_BYTES"),
1943 XML_CONTEXT_BYTES},
1944 #endif
1945 #ifdef XML_MIN_SIZE
1946 {XML_FEATURE_MIN_SIZE, XML_L("XML_MIN_SIZE"), 0},
1947 #endif
1948 #ifdef XML_NS
1949 {XML_FEATURE_NS, XML_L("XML_NS"), 0},
1950 #endif
1951 {XML_FEATURE_END, NULL, 0}
1952 };
1953
1954 return features;
1955 }
1956
1957 /* Initially tag->rawName always points into the parse buffer;
1958 for those TAG instances opened while the current parse buffer was
1959 processed, and not yet closed, we need to store tag->rawName in a more
1960 permanent location, since the parse buffer is about to be discarded.
1961 */
1962 static XML_Bool
storeRawNames(XML_Parser parser)1963 storeRawNames(XML_Parser parser)
1964 {
1965 TAG *tag = tagStack;
1966 while (tag) {
1967 int bufSize;
1968 int nameLen = sizeof(XML_Char) * (tag->name.strLen + 1);
1969 char *rawNameBuf = tag->buf + nameLen;
1970 /* Stop if already stored. Since tagStack is a stack, we can stop
1971 at the first entry that has already been copied; everything
1972 below it in the stack is already been accounted for in a
1973 previous call to this function.
1974 */
1975 if (tag->rawName == rawNameBuf)
1976 break;
1977 /* For re-use purposes we need to ensure that the
1978 size of tag->buf is a multiple of sizeof(XML_Char).
1979 */
1980 bufSize = nameLen + ROUND_UP(tag->rawNameLength, sizeof(XML_Char));
1981 if (bufSize > tag->bufEnd - tag->buf) {
1982 char *temp = (char *)REALLOC(tag->buf, bufSize);
1983 if (temp == NULL)
1984 return XML_FALSE;
1985 /* if tag->name.str points to tag->buf (only when namespace
1986 processing is off) then we have to update it
1987 */
1988 if (tag->name.str == (XML_Char *)tag->buf)
1989 tag->name.str = (XML_Char *)temp;
1990 /* if tag->name.localPart is set (when namespace processing is on)
1991 then update it as well, since it will always point into tag->buf
1992 */
1993 if (tag->name.localPart)
1994 tag->name.localPart = (XML_Char *)temp + (tag->name.localPart -
1995 (XML_Char *)tag->buf);
1996 tag->buf = temp;
1997 tag->bufEnd = temp + bufSize;
1998 rawNameBuf = temp + nameLen;
1999 }
2000 memcpy(rawNameBuf, tag->rawName, tag->rawNameLength);
2001 tag->rawName = rawNameBuf;
2002 tag = tag->parent;
2003 }
2004 return XML_TRUE;
2005 }
2006
2007 static enum XML_Error PTRCALL
contentProcessor(XML_Parser parser,const char * start,const char * end,const char ** endPtr)2008 contentProcessor(XML_Parser parser,
2009 const char *start,
2010 const char *end,
2011 const char **endPtr)
2012 {
2013 enum XML_Error result = doContent(parser, 0, encoding, start, end,
2014 endPtr, (XML_Bool)!ps_finalBuffer);
2015 if (result == XML_ERROR_NONE) {
2016 if (!storeRawNames(parser))
2017 return XML_ERROR_NO_MEMORY;
2018 }
2019 return result;
2020 }
2021
2022 static enum XML_Error PTRCALL
externalEntityInitProcessor(XML_Parser parser,const char * start,const char * end,const char ** endPtr)2023 externalEntityInitProcessor(XML_Parser parser,
2024 const char *start,
2025 const char *end,
2026 const char **endPtr)
2027 {
2028 enum XML_Error result = initializeEncoding(parser);
2029 if (result != XML_ERROR_NONE)
2030 return result;
2031 processor = externalEntityInitProcessor2;
2032 return externalEntityInitProcessor2(parser, start, end, endPtr);
2033 }
2034
2035 static enum XML_Error PTRCALL
externalEntityInitProcessor2(XML_Parser parser,const char * start,const char * end,const char ** endPtr)2036 externalEntityInitProcessor2(XML_Parser parser,
2037 const char *start,
2038 const char *end,
2039 const char **endPtr)
2040 {
2041 const char *next = start; /* XmlContentTok doesn't always set the last arg */
2042 int tok = XmlContentTok(encoding, start, end, &next);
2043 switch (tok) {
2044 case XML_TOK_BOM:
2045 /* If we are at the end of the buffer, this would cause the next stage,
2046 i.e. externalEntityInitProcessor3, to pass control directly to
2047 doContent (by detecting XML_TOK_NONE) without processing any xml text
2048 declaration - causing the error XML_ERROR_MISPLACED_XML_PI in doContent.
2049 */
2050 if (next == end && !ps_finalBuffer) {
2051 *endPtr = next;
2052 return XML_ERROR_NONE;
2053 }
2054 start = next;
2055 break;
2056 case XML_TOK_PARTIAL:
2057 if (!ps_finalBuffer) {
2058 *endPtr = start;
2059 return XML_ERROR_NONE;
2060 }
2061 eventPtr = start;
2062 return XML_ERROR_UNCLOSED_TOKEN;
2063 case XML_TOK_PARTIAL_CHAR:
2064 if (!ps_finalBuffer) {
2065 *endPtr = start;
2066 return XML_ERROR_NONE;
2067 }
2068 eventPtr = start;
2069 return XML_ERROR_PARTIAL_CHAR;
2070 }
2071 processor = externalEntityInitProcessor3;
2072 return externalEntityInitProcessor3(parser, start, end, endPtr);
2073 }
2074
2075 static enum XML_Error PTRCALL
externalEntityInitProcessor3(XML_Parser parser,const char * start,const char * end,const char ** endPtr)2076 externalEntityInitProcessor3(XML_Parser parser,
2077 const char *start,
2078 const char *end,
2079 const char **endPtr)
2080 {
2081 int tok;
2082 const char *next = start; /* XmlContentTok doesn't always set the last arg */
2083 eventPtr = start;
2084 tok = XmlContentTok(encoding, start, end, &next);
2085 eventEndPtr = next;
2086
2087 switch (tok) {
2088 case XML_TOK_XML_DECL:
2089 {
2090 enum XML_Error result;
2091 result = processXmlDecl(parser, 1, start, next);
2092 if (result != XML_ERROR_NONE)
2093 return result;
2094 switch (ps_parsing) {
2095 case XML_SUSPENDED:
2096 *endPtr = next;
2097 return XML_ERROR_NONE;
2098 case XML_FINISHED:
2099 return XML_ERROR_ABORTED;
2100 default:
2101 start = next;
2102 }
2103 }
2104 break;
2105 case XML_TOK_PARTIAL:
2106 if (!ps_finalBuffer) {
2107 *endPtr = start;
2108 return XML_ERROR_NONE;
2109 }
2110 return XML_ERROR_UNCLOSED_TOKEN;
2111 case XML_TOK_PARTIAL_CHAR:
2112 if (!ps_finalBuffer) {
2113 *endPtr = start;
2114 return XML_ERROR_NONE;
2115 }
2116 return XML_ERROR_PARTIAL_CHAR;
2117 }
2118 processor = externalEntityContentProcessor;
2119 tagLevel = 1;
2120 return externalEntityContentProcessor(parser, start, end, endPtr);
2121 }
2122
2123 static enum XML_Error PTRCALL
externalEntityContentProcessor(XML_Parser parser,const char * start,const char * end,const char ** endPtr)2124 externalEntityContentProcessor(XML_Parser parser,
2125 const char *start,
2126 const char *end,
2127 const char **endPtr)
2128 {
2129 enum XML_Error result = doContent(parser, 1, encoding, start, end,
2130 endPtr, (XML_Bool)!ps_finalBuffer);
2131 if (result == XML_ERROR_NONE) {
2132 if (!storeRawNames(parser))
2133 return XML_ERROR_NO_MEMORY;
2134 }
2135 return result;
2136 }
2137
2138 static enum XML_Error
doContent(XML_Parser parser,int startTagLevel,const ENCODING * enc,const char * s,const char * end,const char ** nextPtr,XML_Bool haveMore)2139 doContent(XML_Parser parser,
2140 int startTagLevel,
2141 const ENCODING *enc,
2142 const char *s,
2143 const char *end,
2144 const char **nextPtr,
2145 XML_Bool haveMore)
2146 {
2147 /* save one level of indirection */
2148 DTD * const dtd = _dtd;
2149
2150 const char **eventPP;
2151 const char **eventEndPP;
2152 if (enc == encoding) {
2153 eventPP = &eventPtr;
2154 eventEndPP = &eventEndPtr;
2155 }
2156 else {
2157 eventPP = &(openInternalEntities->internalEventPtr);
2158 eventEndPP = &(openInternalEntities->internalEventEndPtr);
2159 }
2160 *eventPP = s;
2161
2162 for (;;) {
2163 const char *next = s; /* XmlContentTok doesn't always set the last arg */
2164 int tok = XmlContentTok(enc, s, end, &next);
2165 *eventEndPP = next;
2166 switch (tok) {
2167 case XML_TOK_TRAILING_CR:
2168 if (haveMore) {
2169 *nextPtr = s;
2170 return XML_ERROR_NONE;
2171 }
2172 *eventEndPP = end;
2173 if (characterDataHandler) {
2174 XML_Char c = 0xA;
2175 characterDataHandler(handlerArg, &c, 1);
2176 }
2177 else if (defaultHandler)
2178 reportDefault(parser, enc, s, end);
2179 /* We are at the end of the final buffer, should we check for
2180 XML_SUSPENDED, XML_FINISHED?
2181 */
2182 if (startTagLevel == 0)
2183 return XML_ERROR_NO_ELEMENTS;
2184 if (tagLevel != startTagLevel)
2185 return XML_ERROR_ASYNC_ENTITY;
2186 *nextPtr = end;
2187 return XML_ERROR_NONE;
2188 case XML_TOK_NONE:
2189 if (haveMore) {
2190 *nextPtr = s;
2191 return XML_ERROR_NONE;
2192 }
2193 if (startTagLevel > 0) {
2194 if (tagLevel != startTagLevel)
2195 return XML_ERROR_ASYNC_ENTITY;
2196 *nextPtr = s;
2197 return XML_ERROR_NONE;
2198 }
2199 return XML_ERROR_NO_ELEMENTS;
2200 case XML_TOK_INVALID:
2201 *eventPP = next;
2202 return XML_ERROR_INVALID_TOKEN;
2203 case XML_TOK_PARTIAL:
2204 if (haveMore) {
2205 *nextPtr = s;
2206 return XML_ERROR_NONE;
2207 }
2208 return XML_ERROR_UNCLOSED_TOKEN;
2209 case XML_TOK_PARTIAL_CHAR:
2210 if (haveMore) {
2211 *nextPtr = s;
2212 return XML_ERROR_NONE;
2213 }
2214 return XML_ERROR_PARTIAL_CHAR;
2215 case XML_TOK_ENTITY_REF:
2216 {
2217 const XML_Char *name;
2218 ENTITY *entity;
2219 XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc,
2220 s + enc->minBytesPerChar,
2221 next - enc->minBytesPerChar);
2222 if (ch) {
2223 if (characterDataHandler)
2224 characterDataHandler(handlerArg, &ch, 1);
2225 else if (defaultHandler)
2226 reportDefault(parser, enc, s, next);
2227 break;
2228 }
2229 name = poolStoreString(&dtd->pool, enc,
2230 s + enc->minBytesPerChar,
2231 next - enc->minBytesPerChar);
2232 if (!name)
2233 return XML_ERROR_NO_MEMORY;
2234 entity = (ENTITY *)lookup(&dtd->generalEntities, name, 0);
2235 poolDiscard(&dtd->pool);
2236 /* First, determine if a check for an existing declaration is needed;
2237 if yes, check that the entity exists, and that it is internal,
2238 otherwise call the skipped entity or default handler.
2239 */
2240 if (!dtd->hasParamEntityRefs || dtd->standalone) {
2241 if (!entity)
2242 return XML_ERROR_UNDEFINED_ENTITY;
2243 else if (!entity->is_internal)
2244 return XML_ERROR_ENTITY_DECLARED_IN_PE;
2245 }
2246 else if (!entity) {
2247 if (skippedEntityHandler)
2248 skippedEntityHandler(handlerArg, name, 0);
2249 else if (defaultHandler)
2250 reportDefault(parser, enc, s, next);
2251 break;
2252 }
2253 if (entity->open)
2254 return XML_ERROR_RECURSIVE_ENTITY_REF;
2255 if (entity->notation)
2256 return XML_ERROR_BINARY_ENTITY_REF;
2257 if (entity->textPtr) {
2258 enum XML_Error result;
2259 if (!defaultExpandInternalEntities) {
2260 if (skippedEntityHandler)
2261 skippedEntityHandler(handlerArg, entity->name, 0);
2262 else if (defaultHandler)
2263 reportDefault(parser, enc, s, next);
2264 break;
2265 }
2266 result = processInternalEntity(parser, entity, XML_FALSE);
2267 if (result != XML_ERROR_NONE)
2268 return result;
2269 }
2270 else if (externalEntityRefHandler) {
2271 const XML_Char *context;
2272 entity->open = XML_TRUE;
2273 context = getContext(parser);
2274 entity->open = XML_FALSE;
2275 if (!context)
2276 return XML_ERROR_NO_MEMORY;
2277 if (!externalEntityRefHandler(externalEntityRefHandlerArg,
2278 context,
2279 entity->base,
2280 entity->systemId,
2281 entity->publicId))
2282 return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
2283 poolDiscard(&tempPool);
2284 }
2285 else if (defaultHandler)
2286 reportDefault(parser, enc, s, next);
2287 break;
2288 }
2289 case XML_TOK_START_TAG_NO_ATTS:
2290 /* fall through */
2291 case XML_TOK_START_TAG_WITH_ATTS:
2292 {
2293 TAG *tag;
2294 enum XML_Error result;
2295 XML_Char *toPtr;
2296 if (freeTagList) {
2297 tag = freeTagList;
2298 freeTagList = freeTagList->parent;
2299 }
2300 else {
2301 tag = (TAG *)MALLOC(sizeof(TAG));
2302 if (!tag)
2303 return XML_ERROR_NO_MEMORY;
2304 tag->buf = (char *)MALLOC(INIT_TAG_BUF_SIZE);
2305 if (!tag->buf) {
2306 FREE(tag);
2307 return XML_ERROR_NO_MEMORY;
2308 }
2309 tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE;
2310 }
2311 tag->bindings = NULL;
2312 tag->parent = tagStack;
2313 tagStack = tag;
2314 tag->name.localPart = NULL;
2315 tag->name.prefix = NULL;
2316 tag->rawName = s + enc->minBytesPerChar;
2317 tag->rawNameLength = XmlNameLength(enc, tag->rawName);
2318 ++tagLevel;
2319 {
2320 const char *rawNameEnd = tag->rawName + tag->rawNameLength;
2321 const char *fromPtr = tag->rawName;
2322 toPtr = (XML_Char *)tag->buf;
2323 for (;;) {
2324 int bufSize;
2325 int convLen;
2326 XmlConvert(enc,
2327 &fromPtr, rawNameEnd,
2328 (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1);
2329 convLen = (int)(toPtr - (XML_Char *)tag->buf);
2330 if (fromPtr == rawNameEnd) {
2331 tag->name.strLen = convLen;
2332 break;
2333 }
2334 bufSize = (int)(tag->bufEnd - tag->buf) << 1;
2335 {
2336 char *temp = (char *)REALLOC(tag->buf, bufSize);
2337 if (temp == NULL)
2338 return XML_ERROR_NO_MEMORY;
2339 tag->buf = temp;
2340 tag->bufEnd = temp + bufSize;
2341 toPtr = (XML_Char *)temp + convLen;
2342 }
2343 }
2344 }
2345 tag->name.str = (XML_Char *)tag->buf;
2346 *toPtr = XML_T('\0');
2347 result = storeAtts(parser, enc, s, &(tag->name), &(tag->bindings));
2348 if (result)
2349 return result;
2350 if (startElementHandler)
2351 startElementHandler(handlerArg, tag->name.str,
2352 (const XML_Char **)atts);
2353 else if (defaultHandler)
2354 reportDefault(parser, enc, s, next);
2355 poolClear(&tempPool);
2356 break;
2357 }
2358 case XML_TOK_EMPTY_ELEMENT_NO_ATTS:
2359 /* fall through */
2360 case XML_TOK_EMPTY_ELEMENT_WITH_ATTS:
2361 {
2362 const char *rawName = s + enc->minBytesPerChar;
2363 enum XML_Error result;
2364 BINDING *bindings = NULL;
2365 XML_Bool noElmHandlers = XML_TRUE;
2366 TAG_NAME name;
2367 name.str = poolStoreString(&tempPool, enc, rawName,
2368 rawName + XmlNameLength(enc, rawName));
2369 if (!name.str)
2370 return XML_ERROR_NO_MEMORY;
2371 poolFinish(&tempPool);
2372 result = storeAtts(parser, enc, s, &name, &bindings);
2373 if (result)
2374 return result;
2375 poolFinish(&tempPool);
2376 if (startElementHandler) {
2377 startElementHandler(handlerArg, name.str, (const XML_Char **)atts);
2378 noElmHandlers = XML_FALSE;
2379 }
2380 if (endElementHandler) {
2381 if (startElementHandler)
2382 *eventPP = *eventEndPP;
2383 endElementHandler(handlerArg, name.str);
2384 noElmHandlers = XML_FALSE;
2385 }
2386 if (noElmHandlers && defaultHandler)
2387 reportDefault(parser, enc, s, next);
2388 poolClear(&tempPool);
2389 while (bindings) {
2390 BINDING *b = bindings;
2391 if (endNamespaceDeclHandler)
2392 endNamespaceDeclHandler(handlerArg, b->prefix->name);
2393 bindings = bindings->nextTagBinding;
2394 b->nextTagBinding = freeBindingList;
2395 freeBindingList = b;
2396 b->prefix->binding = b->prevPrefixBinding;
2397 }
2398 }
2399 if (tagLevel == 0)
2400 return epilogProcessor(parser, next, end, nextPtr);
2401 break;
2402 case XML_TOK_END_TAG:
2403 if (tagLevel == startTagLevel)
2404 return XML_ERROR_ASYNC_ENTITY;
2405 else {
2406 int len;
2407 const char *rawName;
2408 TAG *tag = tagStack;
2409 tagStack = tag->parent;
2410 tag->parent = freeTagList;
2411 freeTagList = tag;
2412 rawName = s + enc->minBytesPerChar*2;
2413 len = XmlNameLength(enc, rawName);
2414 if (len != tag->rawNameLength
2415 || memcmp(tag->rawName, rawName, len) != 0) {
2416 *eventPP = rawName;
2417 return XML_ERROR_TAG_MISMATCH;
2418 }
2419 --tagLevel;
2420 if (endElementHandler) {
2421 const XML_Char *localPart;
2422 const XML_Char *prefix;
2423 XML_Char *uri;
2424 localPart = tag->name.localPart;
2425 if (ns && localPart) {
2426 /* localPart and prefix may have been overwritten in
2427 tag->name.str, since this points to the binding->uri
2428 buffer which gets re-used; so we have to add them again
2429 */
2430 uri = (XML_Char *)tag->name.str + tag->name.uriLen;
2431 /* don't need to check for space - already done in storeAtts() */
2432 while (*localPart) *uri++ = *localPart++;
2433 prefix = (XML_Char *)tag->name.prefix;
2434 if (ns_triplets && prefix) {
2435 *uri++ = namespaceSeparator;
2436 while (*prefix) *uri++ = *prefix++;
2437 }
2438 *uri = XML_T('\0');
2439 }
2440 endElementHandler(handlerArg, tag->name.str);
2441 }
2442 else if (defaultHandler)
2443 reportDefault(parser, enc, s, next);
2444 while (tag->bindings) {
2445 BINDING *b = tag->bindings;
2446 if (endNamespaceDeclHandler)
2447 endNamespaceDeclHandler(handlerArg, b->prefix->name);
2448 tag->bindings = tag->bindings->nextTagBinding;
2449 b->nextTagBinding = freeBindingList;
2450 freeBindingList = b;
2451 b->prefix->binding = b->prevPrefixBinding;
2452 }
2453 if (tagLevel == 0)
2454 return epilogProcessor(parser, next, end, nextPtr);
2455 }
2456 break;
2457 case XML_TOK_CHAR_REF:
2458 {
2459 int n = XmlCharRefNumber(enc, s);
2460 if (n < 0)
2461 return XML_ERROR_BAD_CHAR_REF;
2462 if (characterDataHandler) {
2463 XML_Char buf[XML_ENCODE_MAX];
2464 characterDataHandler(handlerArg, buf, XmlEncode(n, (ICHAR *)buf));
2465 }
2466 else if (defaultHandler)
2467 reportDefault(parser, enc, s, next);
2468 }
2469 break;
2470 case XML_TOK_XML_DECL:
2471 return XML_ERROR_MISPLACED_XML_PI;
2472 case XML_TOK_DATA_NEWLINE:
2473 if (characterDataHandler) {
2474 XML_Char c = 0xA;
2475 characterDataHandler(handlerArg, &c, 1);
2476 }
2477 else if (defaultHandler)
2478 reportDefault(parser, enc, s, next);
2479 break;
2480 case XML_TOK_CDATA_SECT_OPEN:
2481 {
2482 enum XML_Error result;
2483 if (startCdataSectionHandler)
2484 startCdataSectionHandler(handlerArg);
2485 #if 0
2486 /* Suppose you doing a transformation on a document that involves
2487 changing only the character data. You set up a defaultHandler
2488 and a characterDataHandler. The defaultHandler simply copies
2489 characters through. The characterDataHandler does the
2490 transformation and writes the characters out escaping them as
2491 necessary. This case will fail to work if we leave out the
2492 following two lines (because & and < inside CDATA sections will
2493 be incorrectly escaped).
2494
2495 However, now we have a start/endCdataSectionHandler, so it seems
2496 easier to let the user deal with this.
2497 */
2498 else if (characterDataHandler)
2499 characterDataHandler(handlerArg, dataBuf, 0);
2500 #endif
2501 else if (defaultHandler)
2502 reportDefault(parser, enc, s, next);
2503 result = doCdataSection(parser, enc, &next, end, nextPtr, haveMore);
2504 if (result != XML_ERROR_NONE)
2505 return result;
2506 else if (!next) {
2507 processor = cdataSectionProcessor;
2508 return result;
2509 }
2510 }
2511 break;
2512 case XML_TOK_TRAILING_RSQB:
2513 if (haveMore) {
2514 *nextPtr = s;
2515 return XML_ERROR_NONE;
2516 }
2517 if (characterDataHandler) {
2518 if (MUST_CONVERT(enc, s)) {
2519 ICHAR *dataPtr = (ICHAR *)dataBuf;
2520 XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
2521 characterDataHandler(handlerArg, dataBuf,
2522 (int)(dataPtr - (ICHAR *)dataBuf));
2523 }
2524 else
2525 characterDataHandler(handlerArg,
2526 (XML_Char *)s,
2527 (int)((XML_Char *)end - (XML_Char *)s));
2528 }
2529 else if (defaultHandler)
2530 reportDefault(parser, enc, s, end);
2531 /* We are at the end of the final buffer, should we check for
2532 XML_SUSPENDED, XML_FINISHED?
2533 */
2534 if (startTagLevel == 0) {
2535 *eventPP = end;
2536 return XML_ERROR_NO_ELEMENTS;
2537 }
2538 if (tagLevel != startTagLevel) {
2539 *eventPP = end;
2540 return XML_ERROR_ASYNC_ENTITY;
2541 }
2542 *nextPtr = end;
2543 return XML_ERROR_NONE;
2544 case XML_TOK_DATA_CHARS:
2545 if (characterDataHandler) {
2546 if (MUST_CONVERT(enc, s)) {
2547 for (;;) {
2548 ICHAR *dataPtr = (ICHAR *)dataBuf;
2549 XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
2550 *eventEndPP = s;
2551 characterDataHandler(handlerArg, dataBuf,
2552 (int)(dataPtr - (ICHAR *)dataBuf));
2553 if (s == next)
2554 break;
2555 *eventPP = s;
2556 }
2557 }
2558 else
2559 characterDataHandler(handlerArg,
2560 (XML_Char *)s,
2561 (int)((XML_Char *)next - (XML_Char *)s));
2562 }
2563 else if (defaultHandler)
2564 reportDefault(parser, enc, s, next);
2565 break;
2566 case XML_TOK_PI:
2567 if (!reportProcessingInstruction(parser, enc, s, next))
2568 return XML_ERROR_NO_MEMORY;
2569 break;
2570 case XML_TOK_COMMENT:
2571 if (!reportComment(parser, enc, s, next))
2572 return XML_ERROR_NO_MEMORY;
2573 break;
2574 default:
2575 if (defaultHandler)
2576 reportDefault(parser, enc, s, next);
2577 break;
2578 }
2579 *eventPP = s = next;
2580 switch (ps_parsing) {
2581 case XML_SUSPENDED:
2582 *nextPtr = next;
2583 return XML_ERROR_NONE;
2584 case XML_FINISHED:
2585 return XML_ERROR_ABORTED;
2586 default: ;
2587 }
2588 }
2589 /* not reached */
2590 }
2591
2592 /* Precondition: all arguments must be non-NULL;
2593 Purpose:
2594 - normalize attributes
2595 - check attributes for well-formedness
2596 - generate namespace aware attribute names (URI, prefix)
2597 - build list of attributes for startElementHandler
2598 - default attributes
2599 - process namespace declarations (check and report them)
2600 - generate namespace aware element name (URI, prefix)
2601 */
2602 static enum XML_Error
storeAtts(XML_Parser parser,const ENCODING * enc,const char * attStr,TAG_NAME * tagNamePtr,BINDING ** bindingsPtr)2603 storeAtts(XML_Parser parser, const ENCODING *enc,
2604 const char *attStr, TAG_NAME *tagNamePtr,
2605 BINDING **bindingsPtr)
2606 {
2607 DTD * const dtd = _dtd; /* save one level of indirection */
2608 ELEMENT_TYPE *elementType;
2609 int nDefaultAtts;
2610 const XML_Char **appAtts; /* the attribute list for the application */
2611 int attIndex = 0;
2612 int prefixLen;
2613 int i;
2614 int n;
2615 XML_Char *uri;
2616 int nPrefixes = 0;
2617 BINDING *binding;
2618 const XML_Char *localPart;
2619
2620 /* lookup the element type name */
2621 elementType = (ELEMENT_TYPE *)lookup(&dtd->elementTypes, tagNamePtr->str,0);
2622 if (!elementType) {
2623 const XML_Char *name = poolCopyString(&dtd->pool, tagNamePtr->str);
2624 if (!name)
2625 return XML_ERROR_NO_MEMORY;
2626 elementType = (ELEMENT_TYPE *)lookup(&dtd->elementTypes, name,
2627 sizeof(ELEMENT_TYPE));
2628 if (!elementType)
2629 return XML_ERROR_NO_MEMORY;
2630 if (ns && !setElementTypePrefix(parser, elementType))
2631 return XML_ERROR_NO_MEMORY;
2632 }
2633 nDefaultAtts = elementType->nDefaultAtts;
2634
2635 /* get the attributes from the tokenizer */
2636 n = XmlGetAttributes(enc, attStr, attsSize, atts);
2637 if (n + nDefaultAtts > attsSize) {
2638 int oldAttsSize = attsSize;
2639 ATTRIBUTE *temp;
2640 attsSize = n + nDefaultAtts + INIT_ATTS_SIZE;
2641 temp = (ATTRIBUTE *)REALLOC((void *)atts, attsSize * sizeof(ATTRIBUTE));
2642 if (temp == NULL)
2643 return XML_ERROR_NO_MEMORY;
2644 atts = temp;
2645 if (n > oldAttsSize)
2646 XmlGetAttributes(enc, attStr, n, atts);
2647 }
2648
2649 appAtts = (const XML_Char **)atts;
2650 for (i = 0; i < n; i++) {
2651 /* add the name and value to the attribute list */
2652 ATTRIBUTE_ID *attId = getAttributeId(parser, enc, atts[i].name,
2653 atts[i].name
2654 + XmlNameLength(enc, atts[i].name));
2655 if (!attId)
2656 return XML_ERROR_NO_MEMORY;
2657 /* Detect duplicate attributes by their QNames. This does not work when
2658 namespace processing is turned on and different prefixes for the same
2659 namespace are used. For this case we have a check further down.
2660 */
2661 if ((attId->name)[-1]) {
2662 if (enc == encoding)
2663 eventPtr = atts[i].name;
2664 return XML_ERROR_DUPLICATE_ATTRIBUTE;
2665 }
2666 (attId->name)[-1] = 1;
2667 appAtts[attIndex++] = attId->name;
2668 if (!atts[i].normalized) {
2669 enum XML_Error result;
2670 XML_Bool isCdata = XML_TRUE;
2671
2672 /* figure out whether declared as other than CDATA */
2673 if (attId->maybeTokenized) {
2674 int j;
2675 for (j = 0; j < nDefaultAtts; j++) {
2676 if (attId == elementType->defaultAtts[j].id) {
2677 isCdata = elementType->defaultAtts[j].isCdata;
2678 break;
2679 }
2680 }
2681 }
2682
2683 /* normalize the attribute value */
2684 result = storeAttributeValue(parser, enc, isCdata,
2685 atts[i].valuePtr, atts[i].valueEnd,
2686 &tempPool);
2687 if (result)
2688 return result;
2689 appAtts[attIndex] = poolStart(&tempPool);
2690 poolFinish(&tempPool);
2691 }
2692 else {
2693 /* the value did not need normalizing */
2694 appAtts[attIndex] = poolStoreString(&tempPool, enc, atts[i].valuePtr,
2695 atts[i].valueEnd);
2696 if (appAtts[attIndex] == 0)
2697 return XML_ERROR_NO_MEMORY;
2698 poolFinish(&tempPool);
2699 }
2700 /* handle prefixed attribute names */
2701 if (attId->prefix) {
2702 if (attId->xmlns) {
2703 /* deal with namespace declarations here */
2704 enum XML_Error result = addBinding(parser, attId->prefix, attId,
2705 appAtts[attIndex], bindingsPtr);
2706 if (result)
2707 return result;
2708 --attIndex;
2709 }
2710 else {
2711 /* deal with other prefixed names later */
2712 attIndex++;
2713 nPrefixes++;
2714 (attId->name)[-1] = 2;
2715 }
2716 }
2717 else
2718 attIndex++;
2719 }
2720
2721 /* set-up for XML_GetSpecifiedAttributeCount and XML_GetIdAttributeIndex */
2722 nSpecifiedAtts = attIndex;
2723 if (elementType->idAtt && (elementType->idAtt->name)[-1]) {
2724 for (i = 0; i < attIndex; i += 2)
2725 if (appAtts[i] == elementType->idAtt->name) {
2726 idAttIndex = i;
2727 break;
2728 }
2729 }
2730 else
2731 idAttIndex = -1;
2732
2733 /* do attribute defaulting */
2734 for (i = 0; i < nDefaultAtts; i++) {
2735 const DEFAULT_ATTRIBUTE *da = elementType->defaultAtts + i;
2736 if (!(da->id->name)[-1] && da->value) {
2737 if (da->id->prefix) {
2738 if (da->id->xmlns) {
2739 enum XML_Error result = addBinding(parser, da->id->prefix, da->id,
2740 da->value, bindingsPtr);
2741 if (result)
2742 return result;
2743 }
2744 else {
2745 (da->id->name)[-1] = 2;
2746 nPrefixes++;
2747 appAtts[attIndex++] = da->id->name;
2748 appAtts[attIndex++] = da->value;
2749 }
2750 }
2751 else {
2752 (da->id->name)[-1] = 1;
2753 appAtts[attIndex++] = da->id->name;
2754 appAtts[attIndex++] = da->value;
2755 }
2756 }
2757 }
2758 appAtts[attIndex] = 0;
2759
2760 /* expand prefixed attribute names, check for duplicates,
2761 and clear flags that say whether attributes were specified */
2762 i = 0;
2763 if (nPrefixes) {
2764 int j; /* hash table index */
2765 unsigned long version = nsAttsVersion;
2766 int nsAttsSize = (int)1 << nsAttsPower;
2767 /* size of hash table must be at least 2 * (# of prefixed attributes) */
2768 if ((nPrefixes << 1) >> nsAttsPower) { /* true for nsAttsPower = 0 */
2769 NS_ATT *temp;
2770 /* hash table size must also be a power of 2 and >= 8 */
2771 while (nPrefixes >> nsAttsPower++);
2772 if (nsAttsPower < 3)
2773 nsAttsPower = 3;
2774 nsAttsSize = (int)1 << nsAttsPower;
2775 temp = (NS_ATT *)REALLOC(nsAtts, nsAttsSize * sizeof(NS_ATT));
2776 if (!temp)
2777 return XML_ERROR_NO_MEMORY;
2778 nsAtts = temp;
2779 version = 0; /* force re-initialization of nsAtts hash table */
2780 }
2781 /* using a version flag saves us from initializing nsAtts every time */
2782 if (!version) { /* initialize version flags when version wraps around */
2783 version = INIT_ATTS_VERSION;
2784 for (j = nsAttsSize; j != 0; )
2785 nsAtts[--j].version = version;
2786 }
2787 nsAttsVersion = --version;
2788
2789 /* expand prefixed names and check for duplicates */
2790 for (; i < attIndex; i += 2) {
2791 const XML_Char *s = appAtts[i];
2792 if (s[-1] == 2) { /* prefixed */
2793 ATTRIBUTE_ID *id;
2794 const BINDING *b;
2795 unsigned long uriHash = 0;
2796 ((XML_Char *)s)[-1] = 0; /* clear flag */
2797 id = (ATTRIBUTE_ID *)lookup(&dtd->attributeIds, s, 0);
2798 b = id->prefix->binding;
2799 if (!b)
2800 return XML_ERROR_UNBOUND_PREFIX;
2801
2802 /* as we expand the name we also calculate its hash value */
2803 for (j = 0; j < b->uriLen; j++) {
2804 const XML_Char c = b->uri[j];
2805 if (!poolAppendChar(&tempPool, c))
2806 return XML_ERROR_NO_MEMORY;
2807 uriHash = CHAR_HASH(uriHash, c);
2808 }
2809 while (*s++ != XML_T(':'))
2810 ;
2811 do { /* copies null terminator */
2812 const XML_Char c = *s;
2813 if (!poolAppendChar(&tempPool, *s))
2814 return XML_ERROR_NO_MEMORY;
2815 uriHash = CHAR_HASH(uriHash, c);
2816 } while (*s++);
2817
2818 { /* Check hash table for duplicate of expanded name (uriName).
2819 Derived from code in lookup(HASH_TABLE *table, ...).
2820 */
2821 unsigned char step = 0;
2822 unsigned long mask = nsAttsSize - 1;
2823 j = uriHash & mask; /* index into hash table */
2824 while (nsAtts[j].version == version) {
2825 /* for speed we compare stored hash values first */
2826 if (uriHash == nsAtts[j].hash) {
2827 const XML_Char *s1 = poolStart(&tempPool);
2828 const XML_Char *s2 = nsAtts[j].uriName;
2829 /* s1 is null terminated, but not s2 */
2830 for (; *s1 == *s2 && *s1 != 0; s1++, s2++);
2831 if (*s1 == 0)
2832 return XML_ERROR_DUPLICATE_ATTRIBUTE;
2833 }
2834 if (!step)
2835 step = PROBE_STEP(uriHash, mask, nsAttsPower);
2836 j < step ? (j += nsAttsSize - step) : (j -= step);
2837 }
2838 }
2839
2840 if (ns_triplets) { /* append namespace separator and prefix */
2841 tempPool.ptr[-1] = namespaceSeparator;
2842 s = b->prefix->name;
2843 do {
2844 if (!poolAppendChar(&tempPool, *s))
2845 return XML_ERROR_NO_MEMORY;
2846 } while (*s++);
2847 }
2848
2849 /* store expanded name in attribute list */
2850 s = poolStart(&tempPool);
2851 poolFinish(&tempPool);
2852 appAtts[i] = s;
2853
2854 /* fill empty slot with new version, uriName and hash value */
2855 nsAtts[j].version = version;
2856 nsAtts[j].hash = uriHash;
2857 nsAtts[j].uriName = s;
2858
2859 if (!--nPrefixes) {
2860 i += 2;
2861 break;
2862 }
2863 }
2864 else /* not prefixed */
2865 ((XML_Char *)s)[-1] = 0; /* clear flag */
2866 }
2867 }
2868 /* clear flags for the remaining attributes */
2869 for (; i < attIndex; i += 2)
2870 ((XML_Char *)(appAtts[i]))[-1] = 0;
2871 for (binding = *bindingsPtr; binding; binding = binding->nextTagBinding)
2872 binding->attId->name[-1] = 0;
2873
2874 if (!ns)
2875 return XML_ERROR_NONE;
2876
2877 /* expand the element type name */
2878 if (elementType->prefix) {
2879 binding = elementType->prefix->binding;
2880 if (!binding)
2881 return XML_ERROR_UNBOUND_PREFIX;
2882 localPart = tagNamePtr->str;
2883 while (*localPart++ != XML_T(':'))
2884 ;
2885 }
2886 else if (dtd->defaultPrefix.binding) {
2887 binding = dtd->defaultPrefix.binding;
2888 localPart = tagNamePtr->str;
2889 }
2890 else
2891 return XML_ERROR_NONE;
2892 prefixLen = 0;
2893 if (ns_triplets && binding->prefix->name) {
2894 for (; binding->prefix->name[prefixLen++];)
2895 ; /* prefixLen includes null terminator */
2896 }
2897 tagNamePtr->localPart = localPart;
2898 tagNamePtr->uriLen = binding->uriLen;
2899 tagNamePtr->prefix = binding->prefix->name;
2900 tagNamePtr->prefixLen = prefixLen;
2901 for (i = 0; localPart[i++];)
2902 ; /* i includes null terminator */
2903 n = i + binding->uriLen + prefixLen;
2904 if (n > binding->uriAlloc) {
2905 TAG *p;
2906 uri = (XML_Char *)MALLOC((n + EXPAND_SPARE) * sizeof(XML_Char));
2907 if (!uri)
2908 return XML_ERROR_NO_MEMORY;
2909 binding->uriAlloc = n + EXPAND_SPARE;
2910 memcpy(uri, binding->uri, binding->uriLen * sizeof(XML_Char));
2911 for (p = tagStack; p; p = p->parent)
2912 if (p->name.str == binding->uri)
2913 p->name.str = uri;
2914 FREE(binding->uri);
2915 binding->uri = uri;
2916 }
2917 /* if namespaceSeparator != '\0' then uri includes it already */
2918 uri = binding->uri + binding->uriLen;
2919 memcpy(uri, localPart, i * sizeof(XML_Char));
2920 /* we always have a namespace separator between localPart and prefix */
2921 if (prefixLen) {
2922 uri += i - 1;
2923 *uri = namespaceSeparator; /* replace null terminator */
2924 memcpy(uri + 1, binding->prefix->name, prefixLen * sizeof(XML_Char));
2925 }
2926 tagNamePtr->str = binding->uri;
2927 return XML_ERROR_NONE;
2928 }
2929
2930 /* addBinding() overwrites the value of prefix->binding without checking.
2931 Therefore one must keep track of the old value outside of addBinding().
2932 */
2933 static enum XML_Error
addBinding(XML_Parser parser,PREFIX * prefix,const ATTRIBUTE_ID * attId,const XML_Char * uri,BINDING ** bindingsPtr)2934 addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId,
2935 const XML_Char *uri, BINDING **bindingsPtr)
2936 {
2937 static const XML_Char xmlNamespace[] = {
2938 'h', 't', 't', 'p', ':', '/', '/',
2939 'w', 'w', 'w', '.', 'w', '3', '.', 'o', 'r', 'g', '/',
2940 'X', 'M', 'L', '/', '1', '9', '9', '8', '/',
2941 'n', 'a', 'm', 'e', 's', 'p', 'a', 'c', 'e', '\0'
2942 };
2943 static const int xmlLen =
2944 (int)sizeof(xmlNamespace)/sizeof(XML_Char) - 1;
2945 static const XML_Char xmlnsNamespace[] = {
2946 'h', 't', 't', 'p', ':', '/', '/',
2947 'w', 'w', 'w', '.', 'w', '3', '.', 'o', 'r', 'g', '/',
2948 '2', '0', '0', '0', '/', 'x', 'm', 'l', 'n', 's', '/', '\0'
2949 };
2950 static const int xmlnsLen =
2951 (int)sizeof(xmlnsNamespace)/sizeof(XML_Char) - 1;
2952
2953 XML_Bool mustBeXML = XML_FALSE;
2954 XML_Bool isXML = XML_TRUE;
2955 XML_Bool isXMLNS = XML_TRUE;
2956
2957 BINDING *b;
2958 int len;
2959
2960 /* empty URI is only valid for default namespace per XML NS 1.0 (not 1.1) */
2961 if (*uri == XML_T('\0') && prefix->name)
2962 return XML_ERROR_UNDECLARING_PREFIX;
2963
2964 if (prefix->name
2965 && prefix->name[0] == XML_T('x')
2966 && prefix->name[1] == XML_T('m')
2967 && prefix->name[2] == XML_T('l')) {
2968
2969 /* Not allowed to bind xmlns */
2970 if (prefix->name[3] == XML_T('n')
2971 && prefix->name[4] == XML_T('s')
2972 && prefix->name[5] == XML_T('\0'))
2973 return XML_ERROR_RESERVED_PREFIX_XMLNS;
2974
2975 if (prefix->name[3] == XML_T('\0'))
2976 mustBeXML = XML_TRUE;
2977 }
2978
2979 for (len = 0; uri[len]; len++) {
2980 if (isXML && (len > xmlLen || uri[len] != xmlNamespace[len]))
2981 isXML = XML_FALSE;
2982
2983 if (!mustBeXML && isXMLNS
2984 && (len > xmlnsLen || uri[len] != xmlnsNamespace[len]))
2985 isXMLNS = XML_FALSE;
2986 }
2987 isXML = isXML && len == xmlLen;
2988 isXMLNS = isXMLNS && len == xmlnsLen;
2989
2990 if (mustBeXML != isXML)
2991 return mustBeXML ? XML_ERROR_RESERVED_PREFIX_XML
2992 : XML_ERROR_RESERVED_NAMESPACE_URI;
2993
2994 if (isXMLNS)
2995 return XML_ERROR_RESERVED_NAMESPACE_URI;
2996
2997 if (namespaceSeparator)
2998 len++;
2999 if (freeBindingList) {
3000 b = freeBindingList;
3001 if (len > b->uriAlloc) {
3002 XML_Char *temp = (XML_Char *)REALLOC(b->uri,
3003 sizeof(XML_Char) * (len + EXPAND_SPARE));
3004 if (temp == NULL)
3005 return XML_ERROR_NO_MEMORY;
3006 b->uri = temp;
3007 b->uriAlloc = len + EXPAND_SPARE;
3008 }
3009 freeBindingList = b->nextTagBinding;
3010 }
3011 else {
3012 b = (BINDING *)MALLOC(sizeof(BINDING));
3013 if (!b)
3014 return XML_ERROR_NO_MEMORY;
3015 b->uri = (XML_Char *)MALLOC(sizeof(XML_Char) * (len + EXPAND_SPARE));
3016 if (!b->uri) {
3017 FREE(b);
3018 return XML_ERROR_NO_MEMORY;
3019 }
3020 b->uriAlloc = len + EXPAND_SPARE;
3021 }
3022 b->uriLen = len;
3023 memcpy(b->uri, uri, len * sizeof(XML_Char));
3024 if (namespaceSeparator)
3025 b->uri[len - 1] = namespaceSeparator;
3026 b->prefix = prefix;
3027 b->attId = attId;
3028 b->prevPrefixBinding = prefix->binding;
3029 /* NULL binding when default namespace undeclared */
3030 if (*uri == XML_T('\0') && prefix == &_dtd->defaultPrefix)
3031 prefix->binding = NULL;
3032 else
3033 prefix->binding = b;
3034 b->nextTagBinding = *bindingsPtr;
3035 *bindingsPtr = b;
3036 /* if attId == NULL then we are not starting a namespace scope */
3037 if (attId && startNamespaceDeclHandler)
3038 startNamespaceDeclHandler(handlerArg, prefix->name,
3039 prefix->binding ? uri : 0);
3040 return XML_ERROR_NONE;
3041 }
3042
3043 /* The idea here is to avoid using stack for each CDATA section when
3044 the whole file is parsed with one call.
3045 */
3046 static enum XML_Error PTRCALL
cdataSectionProcessor(XML_Parser parser,const char * start,const char * end,const char ** endPtr)3047 cdataSectionProcessor(XML_Parser parser,
3048 const char *start,
3049 const char *end,
3050 const char **endPtr)
3051 {
3052 enum XML_Error result = doCdataSection(parser, encoding, &start, end,
3053 endPtr, (XML_Bool)!ps_finalBuffer);
3054 if (result != XML_ERROR_NONE)
3055 return result;
3056 if (start) {
3057 if (parentParser) { /* we are parsing an external entity */
3058 processor = externalEntityContentProcessor;
3059 return externalEntityContentProcessor(parser, start, end, endPtr);
3060 }
3061 else {
3062 processor = contentProcessor;
3063 return contentProcessor(parser, start, end, endPtr);
3064 }
3065 }
3066 return result;
3067 }
3068
3069 /* startPtr gets set to non-null if the section is closed, and to null if
3070 the section is not yet closed.
3071 */
3072 static enum XML_Error
doCdataSection(XML_Parser parser,const ENCODING * enc,const char ** startPtr,const char * end,const char ** nextPtr,XML_Bool haveMore)3073 doCdataSection(XML_Parser parser,
3074 const ENCODING *enc,
3075 const char **startPtr,
3076 const char *end,
3077 const char **nextPtr,
3078 XML_Bool haveMore)
3079 {
3080 const char *s = *startPtr;
3081 const char **eventPP;
3082 const char **eventEndPP;
3083 if (enc == encoding) {
3084 eventPP = &eventPtr;
3085 *eventPP = s;
3086 eventEndPP = &eventEndPtr;
3087 }
3088 else {
3089 eventPP = &(openInternalEntities->internalEventPtr);
3090 eventEndPP = &(openInternalEntities->internalEventEndPtr);
3091 }
3092 *eventPP = s;
3093 *startPtr = NULL;
3094
3095 for (;;) {
3096 const char *next;
3097 int tok = XmlCdataSectionTok(enc, s, end, &next);
3098 *eventEndPP = next;
3099 switch (tok) {
3100 case XML_TOK_CDATA_SECT_CLOSE:
3101 if (endCdataSectionHandler)
3102 endCdataSectionHandler(handlerArg);
3103 #if 0
3104 /* see comment under XML_TOK_CDATA_SECT_OPEN */
3105 else if (characterDataHandler)
3106 characterDataHandler(handlerArg, dataBuf, 0);
3107 #endif
3108 else if (defaultHandler)
3109 reportDefault(parser, enc, s, next);
3110 *startPtr = next;
3111 *nextPtr = next;
3112 if (ps_parsing == XML_FINISHED)
3113 return XML_ERROR_ABORTED;
3114 else
3115 return XML_ERROR_NONE;
3116 case XML_TOK_DATA_NEWLINE:
3117 if (characterDataHandler) {
3118 XML_Char c = 0xA;
3119 characterDataHandler(handlerArg, &c, 1);
3120 }
3121 else if (defaultHandler)
3122 reportDefault(parser, enc, s, next);
3123 break;
3124 case XML_TOK_DATA_CHARS:
3125 if (characterDataHandler) {
3126 if (MUST_CONVERT(enc, s)) {
3127 for (;;) {
3128 ICHAR *dataPtr = (ICHAR *)dataBuf;
3129 XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
3130 *eventEndPP = next;
3131 characterDataHandler(handlerArg, dataBuf,
3132 (int)(dataPtr - (ICHAR *)dataBuf));
3133 if (s == next)
3134 break;
3135 *eventPP = s;
3136 }
3137 }
3138 else
3139 characterDataHandler(handlerArg,
3140 (XML_Char *)s,
3141 (int)((XML_Char *)next - (XML_Char *)s));
3142 }
3143 else if (defaultHandler)
3144 reportDefault(parser, enc, s, next);
3145 break;
3146 case XML_TOK_INVALID:
3147 *eventPP = next;
3148 return XML_ERROR_INVALID_TOKEN;
3149 case XML_TOK_PARTIAL_CHAR:
3150 if (haveMore) {
3151 *nextPtr = s;
3152 return XML_ERROR_NONE;
3153 }
3154 return XML_ERROR_PARTIAL_CHAR;
3155 case XML_TOK_PARTIAL:
3156 case XML_TOK_NONE:
3157 if (haveMore) {
3158 *nextPtr = s;
3159 return XML_ERROR_NONE;
3160 }
3161 return XML_ERROR_UNCLOSED_CDATA_SECTION;
3162 default:
3163 *eventPP = next;
3164 return XML_ERROR_UNEXPECTED_STATE;
3165 }
3166
3167 *eventPP = s = next;
3168 switch (ps_parsing) {
3169 case XML_SUSPENDED:
3170 *nextPtr = next;
3171 return XML_ERROR_NONE;
3172 case XML_FINISHED:
3173 return XML_ERROR_ABORTED;
3174 default: ;
3175 }
3176 }
3177 /* not reached */
3178 }
3179
3180 #ifdef XML_DTD
3181
3182 /* The idea here is to avoid using stack for each IGNORE section when
3183 the whole file is parsed with one call.
3184 */
3185 static enum XML_Error PTRCALL
ignoreSectionProcessor(XML_Parser parser,const char * start,const char * end,const char ** endPtr)3186 ignoreSectionProcessor(XML_Parser parser,
3187 const char *start,
3188 const char *end,
3189 const char **endPtr)
3190 {
3191 enum XML_Error result = doIgnoreSection(parser, encoding, &start, end,
3192 endPtr, (XML_Bool)!ps_finalBuffer);
3193 if (result != XML_ERROR_NONE)
3194 return result;
3195 if (start) {
3196 processor = prologProcessor;
3197 return prologProcessor(parser, start, end, endPtr);
3198 }
3199 return result;
3200 }
3201
3202 /* startPtr gets set to non-null is the section is closed, and to null
3203 if the section is not yet closed.
3204 */
3205 static enum XML_Error
doIgnoreSection(XML_Parser parser,const ENCODING * enc,const char ** startPtr,const char * end,const char ** nextPtr,XML_Bool haveMore)3206 doIgnoreSection(XML_Parser parser,
3207 const ENCODING *enc,
3208 const char **startPtr,
3209 const char *end,
3210 const char **nextPtr,
3211 XML_Bool haveMore)
3212 {
3213 const char *next;
3214 int tok;
3215 const char *s = *startPtr;
3216 const char **eventPP;
3217 const char **eventEndPP;
3218 if (enc == encoding) {
3219 eventPP = &eventPtr;
3220 *eventPP = s;
3221 eventEndPP = &eventEndPtr;
3222 }
3223 else {
3224 eventPP = &(openInternalEntities->internalEventPtr);
3225 eventEndPP = &(openInternalEntities->internalEventEndPtr);
3226 }
3227 *eventPP = s;
3228 *startPtr = NULL;
3229 tok = XmlIgnoreSectionTok(enc, s, end, &next);
3230 *eventEndPP = next;
3231 switch (tok) {
3232 case XML_TOK_IGNORE_SECT:
3233 if (defaultHandler)
3234 reportDefault(parser, enc, s, next);
3235 *startPtr = next;
3236 *nextPtr = next;
3237 if (ps_parsing == XML_FINISHED)
3238 return XML_ERROR_ABORTED;
3239 else
3240 return XML_ERROR_NONE;
3241 case XML_TOK_INVALID:
3242 *eventPP = next;
3243 return XML_ERROR_INVALID_TOKEN;
3244 case XML_TOK_PARTIAL_CHAR:
3245 if (haveMore) {
3246 *nextPtr = s;
3247 return XML_ERROR_NONE;
3248 }
3249 return XML_ERROR_PARTIAL_CHAR;
3250 case XML_TOK_PARTIAL:
3251 case XML_TOK_NONE:
3252 if (haveMore) {
3253 *nextPtr = s;
3254 return XML_ERROR_NONE;
3255 }
3256 return XML_ERROR_SYNTAX; /* XML_ERROR_UNCLOSED_IGNORE_SECTION */
3257 default:
3258 *eventPP = next;
3259 return XML_ERROR_UNEXPECTED_STATE;
3260 }
3261 /* not reached */
3262 }
3263
3264 #endif /* XML_DTD */
3265
3266 static enum XML_Error
initializeEncoding(XML_Parser parser)3267 initializeEncoding(XML_Parser parser)
3268 {
3269 const char *s;
3270 #ifdef XML_UNICODE
3271 char encodingBuf[128];
3272 if (!protocolEncodingName)
3273 s = NULL;
3274 else {
3275 int i;
3276 for (i = 0; protocolEncodingName[i]; i++) {
3277 if (i == sizeof(encodingBuf) - 1
3278 || (protocolEncodingName[i] & ~0x7f) != 0) {
3279 encodingBuf[0] = '\0';
3280 break;
3281 }
3282 encodingBuf[i] = (char)protocolEncodingName[i];
3283 }
3284 encodingBuf[i] = '\0';
3285 s = encodingBuf;
3286 }
3287 #else
3288 s = protocolEncodingName;
3289 #endif
3290 if ((ns ? XmlInitEncodingNS : XmlInitEncoding)(&initEncoding, &encoding, s))
3291 return XML_ERROR_NONE;
3292 return handleUnknownEncoding(parser, protocolEncodingName);
3293 }
3294
3295 static enum XML_Error
processXmlDecl(XML_Parser parser,int isGeneralTextEntity,const char * s,const char * next)3296 processXmlDecl(XML_Parser parser, int isGeneralTextEntity,
3297 const char *s, const char *next)
3298 {
3299 const char *encodingName = NULL;
3300 const XML_Char *storedEncName = NULL;
3301 const ENCODING *newEncoding = NULL;
3302 const char *version = NULL;
3303 const char *versionend;
3304 const XML_Char *storedversion = NULL;
3305 int standalone = -1;
3306 if (!(ns
3307 ? XmlParseXmlDeclNS
3308 : XmlParseXmlDecl)(isGeneralTextEntity,
3309 encoding,
3310 s,
3311 next,
3312 &eventPtr,
3313 &version,
3314 &versionend,
3315 &encodingName,
3316 &newEncoding,
3317 &standalone)) {
3318 if (isGeneralTextEntity)
3319 return XML_ERROR_TEXT_DECL;
3320 else
3321 return XML_ERROR_XML_DECL;
3322 }
3323 if (!isGeneralTextEntity && standalone == 1) {
3324 _dtd->standalone = XML_TRUE;
3325 #ifdef XML_DTD
3326 if (paramEntityParsing == XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE)
3327 paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
3328 #endif /* XML_DTD */
3329 }
3330 if (xmlDeclHandler) {
3331 if (encodingName != NULL) {
3332 storedEncName = poolStoreString(&temp2Pool,
3333 encoding,
3334 encodingName,
3335 encodingName
3336 + XmlNameLength(encoding, encodingName));
3337 if (!storedEncName)
3338 return XML_ERROR_NO_MEMORY;
3339 poolFinish(&temp2Pool);
3340 }
3341 if (version) {
3342 storedversion = poolStoreString(&temp2Pool,
3343 encoding,
3344 version,
3345 versionend - encoding->minBytesPerChar);
3346 if (!storedversion)
3347 return XML_ERROR_NO_MEMORY;
3348 }
3349 xmlDeclHandler(handlerArg, storedversion, storedEncName, standalone);
3350 }
3351 else if (defaultHandler)
3352 reportDefault(parser, encoding, s, next);
3353 if (protocolEncodingName == NULL) {
3354 if (newEncoding) {
3355 if (newEncoding->minBytesPerChar != encoding->minBytesPerChar) {
3356 eventPtr = encodingName;
3357 return XML_ERROR_INCORRECT_ENCODING;
3358 }
3359 encoding = newEncoding;
3360 }
3361 else if (encodingName) {
3362 enum XML_Error result;
3363 if (!storedEncName) {
3364 storedEncName = poolStoreString(
3365 &temp2Pool, encoding, encodingName,
3366 encodingName + XmlNameLength(encoding, encodingName));
3367 if (!storedEncName)
3368 return XML_ERROR_NO_MEMORY;
3369 }
3370 result = handleUnknownEncoding(parser, storedEncName);
3371 poolClear(&temp2Pool);
3372 if (result == XML_ERROR_UNKNOWN_ENCODING)
3373 eventPtr = encodingName;
3374 return result;
3375 }
3376 }
3377
3378 if (storedEncName || storedversion)
3379 poolClear(&temp2Pool);
3380
3381 return XML_ERROR_NONE;
3382 }
3383
3384 static enum XML_Error
handleUnknownEncoding(XML_Parser parser,const XML_Char * encodingName)3385 handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName)
3386 {
3387 if (unknownEncodingHandler) {
3388 XML_Encoding info;
3389 int i;
3390 for (i = 0; i < 256; i++)
3391 info.map[i] = -1;
3392 info.convert = NULL;
3393 info.data = NULL;
3394 info.release = NULL;
3395 if (unknownEncodingHandler(unknownEncodingHandlerData, encodingName,
3396 &info)) {
3397 ENCODING *enc;
3398 unknownEncodingMem = MALLOC(XmlSizeOfUnknownEncoding());
3399 if (!unknownEncodingMem) {
3400 if (info.release)
3401 info.release(info.data);
3402 return XML_ERROR_NO_MEMORY;
3403 }
3404 enc = (ns
3405 ? XmlInitUnknownEncodingNS
3406 : XmlInitUnknownEncoding)(unknownEncodingMem,
3407 info.map,
3408 info.convert,
3409 info.data);
3410 if (enc) {
3411 unknownEncodingData = info.data;
3412 unknownEncodingRelease = info.release;
3413 encoding = enc;
3414 return XML_ERROR_NONE;
3415 }
3416 }
3417 if (info.release != NULL)
3418 info.release(info.data);
3419 }
3420 return XML_ERROR_UNKNOWN_ENCODING;
3421 }
3422
3423 static enum XML_Error PTRCALL
prologInitProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)3424 prologInitProcessor(XML_Parser parser,
3425 const char *s,
3426 const char *end,
3427 const char **nextPtr)
3428 {
3429 enum XML_Error result = initializeEncoding(parser);
3430 if (result != XML_ERROR_NONE)
3431 return result;
3432 processor = prologProcessor;
3433 return prologProcessor(parser, s, end, nextPtr);
3434 }
3435
3436 #ifdef XML_DTD
3437
3438 static enum XML_Error PTRCALL
externalParEntInitProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)3439 externalParEntInitProcessor(XML_Parser parser,
3440 const char *s,
3441 const char *end,
3442 const char **nextPtr)
3443 {
3444 enum XML_Error result = initializeEncoding(parser);
3445 if (result != XML_ERROR_NONE)
3446 return result;
3447
3448 /* we know now that XML_Parse(Buffer) has been called,
3449 so we consider the external parameter entity read */
3450 _dtd->paramEntityRead = XML_TRUE;
3451
3452 if (prologState.inEntityValue) {
3453 processor = entityValueInitProcessor;
3454 return entityValueInitProcessor(parser, s, end, nextPtr);
3455 }
3456 else {
3457 processor = externalParEntProcessor;
3458 return externalParEntProcessor(parser, s, end, nextPtr);
3459 }
3460 }
3461
3462 static enum XML_Error PTRCALL
entityValueInitProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)3463 entityValueInitProcessor(XML_Parser parser,
3464 const char *s,
3465 const char *end,
3466 const char **nextPtr)
3467 {
3468 int tok;
3469 const char *start = s;
3470 const char *next = start;
3471 eventPtr = start;
3472
3473 for (;;) {
3474 tok = XmlPrologTok(encoding, start, end, &next);
3475 eventEndPtr = next;
3476 if (tok <= 0) {
3477 if (!ps_finalBuffer && tok != XML_TOK_INVALID) {
3478 *nextPtr = s;
3479 return XML_ERROR_NONE;
3480 }
3481 switch (tok) {
3482 case XML_TOK_INVALID:
3483 return XML_ERROR_INVALID_TOKEN;
3484 case XML_TOK_PARTIAL:
3485 return XML_ERROR_UNCLOSED_TOKEN;
3486 case XML_TOK_PARTIAL_CHAR:
3487 return XML_ERROR_PARTIAL_CHAR;
3488 case XML_TOK_NONE: /* start == end */
3489 default:
3490 break;
3491 }
3492 /* found end of entity value - can store it now */
3493 return storeEntityValue(parser, encoding, s, end);
3494 }
3495 else if (tok == XML_TOK_XML_DECL) {
3496 enum XML_Error result;
3497 result = processXmlDecl(parser, 0, start, next);
3498 if (result != XML_ERROR_NONE)
3499 return result;
3500 switch (ps_parsing) {
3501 case XML_SUSPENDED:
3502 *nextPtr = next;
3503 return XML_ERROR_NONE;
3504 case XML_FINISHED:
3505 return XML_ERROR_ABORTED;
3506 default:
3507 *nextPtr = next;
3508 }
3509 /* stop scanning for text declaration - we found one */
3510 processor = entityValueProcessor;
3511 return entityValueProcessor(parser, next, end, nextPtr);
3512 }
3513 /* If we are at the end of the buffer, this would cause XmlPrologTok to
3514 return XML_TOK_NONE on the next call, which would then cause the
3515 function to exit with *nextPtr set to s - that is what we want for other
3516 tokens, but not for the BOM - we would rather like to skip it;
3517 then, when this routine is entered the next time, XmlPrologTok will
3518 return XML_TOK_INVALID, since the BOM is still in the buffer
3519 */
3520 else if (tok == XML_TOK_BOM && next == end && !ps_finalBuffer) {
3521 *nextPtr = next;
3522 return XML_ERROR_NONE;
3523 }
3524 start = next;
3525 eventPtr = start;
3526 }
3527 }
3528
3529 static enum XML_Error PTRCALL
externalParEntProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)3530 externalParEntProcessor(XML_Parser parser,
3531 const char *s,
3532 const char *end,
3533 const char **nextPtr)
3534 {
3535 const char *next = s;
3536 int tok;
3537
3538 tok = XmlPrologTok(encoding, s, end, &next);
3539 if (tok <= 0) {
3540 if (!ps_finalBuffer && tok != XML_TOK_INVALID) {
3541 *nextPtr = s;
3542 return XML_ERROR_NONE;
3543 }
3544 switch (tok) {
3545 case XML_TOK_INVALID:
3546 return XML_ERROR_INVALID_TOKEN;
3547 case XML_TOK_PARTIAL:
3548 return XML_ERROR_UNCLOSED_TOKEN;
3549 case XML_TOK_PARTIAL_CHAR:
3550 return XML_ERROR_PARTIAL_CHAR;
3551 case XML_TOK_NONE: /* start == end */
3552 default:
3553 break;
3554 }
3555 }
3556 /* This would cause the next stage, i.e. doProlog to be passed XML_TOK_BOM.
3557 However, when parsing an external subset, doProlog will not accept a BOM
3558 as valid, and report a syntax error, so we have to skip the BOM
3559 */
3560 else if (tok == XML_TOK_BOM) {
3561 s = next;
3562 tok = XmlPrologTok(encoding, s, end, &next);
3563 }
3564
3565 processor = prologProcessor;
3566 return doProlog(parser, encoding, s, end, tok, next,
3567 nextPtr, (XML_Bool)!ps_finalBuffer);
3568 }
3569
3570 static enum XML_Error PTRCALL
entityValueProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)3571 entityValueProcessor(XML_Parser parser,
3572 const char *s,
3573 const char *end,
3574 const char **nextPtr)
3575 {
3576 const char *start = s;
3577 const char *next = s;
3578 const ENCODING *enc = encoding;
3579 int tok;
3580
3581 for (;;) {
3582 tok = XmlPrologTok(enc, start, end, &next);
3583 if (tok <= 0) {
3584 if (!ps_finalBuffer && tok != XML_TOK_INVALID) {
3585 *nextPtr = s;
3586 return XML_ERROR_NONE;
3587 }
3588 switch (tok) {
3589 case XML_TOK_INVALID:
3590 return XML_ERROR_INVALID_TOKEN;
3591 case XML_TOK_PARTIAL:
3592 return XML_ERROR_UNCLOSED_TOKEN;
3593 case XML_TOK_PARTIAL_CHAR:
3594 return XML_ERROR_PARTIAL_CHAR;
3595 case XML_TOK_NONE: /* start == end */
3596 default:
3597 break;
3598 }
3599 /* found end of entity value - can store it now */
3600 return storeEntityValue(parser, enc, s, end);
3601 }
3602 start = next;
3603 }
3604 }
3605
3606 #endif /* XML_DTD */
3607
3608 static enum XML_Error PTRCALL
prologProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)3609 prologProcessor(XML_Parser parser,
3610 const char *s,
3611 const char *end,
3612 const char **nextPtr)
3613 {
3614 const char *next = s;
3615 int tok = XmlPrologTok(encoding, s, end, &next);
3616 return doProlog(parser, encoding, s, end, tok, next,
3617 nextPtr, (XML_Bool)!ps_finalBuffer);
3618 }
3619
3620 static enum XML_Error
doProlog(XML_Parser parser,const ENCODING * enc,const char * s,const char * end,int tok,const char * next,const char ** nextPtr,XML_Bool haveMore)3621 doProlog(XML_Parser parser,
3622 const ENCODING *enc,
3623 const char *s,
3624 const char *end,
3625 int tok,
3626 const char *next,
3627 const char **nextPtr,
3628 XML_Bool haveMore)
3629 {
3630 #ifdef XML_DTD
3631 static const XML_Char externalSubsetName[] = { '#' , '\0' };
3632 #endif /* XML_DTD */
3633 static const XML_Char atypeCDATA[] = { 'C', 'D', 'A', 'T', 'A', '\0' };
3634 static const XML_Char atypeID[] = { 'I', 'D', '\0' };
3635 static const XML_Char atypeIDREF[] = { 'I', 'D', 'R', 'E', 'F', '\0' };
3636 static const XML_Char atypeIDREFS[] = { 'I', 'D', 'R', 'E', 'F', 'S', '\0' };
3637 static const XML_Char atypeENTITY[] = { 'E', 'N', 'T', 'I', 'T', 'Y', '\0' };
3638 static const XML_Char atypeENTITIES[] =
3639 { 'E', 'N', 'T', 'I', 'T', 'I', 'E', 'S', '\0' };
3640 static const XML_Char atypeNMTOKEN[] = {
3641 'N', 'M', 'T', 'O', 'K', 'E', 'N', '\0' };
3642 static const XML_Char atypeNMTOKENS[] = {
3643 'N', 'M', 'T', 'O', 'K', 'E', 'N', 'S', '\0' };
3644 static const XML_Char notationPrefix[] = {
3645 'N', 'O', 'T', 'A', 'T', 'I', 'O', 'N', '(', '\0' };
3646 static const XML_Char enumValueSep[] = { '|', '\0' };
3647 static const XML_Char enumValueStart[] = { '(', '\0' };
3648
3649 /* save one level of indirection */
3650 DTD * const dtd = _dtd;
3651
3652 const char **eventPP;
3653 const char **eventEndPP;
3654 enum XML_Content_Quant quant;
3655
3656 if (enc == encoding) {
3657 eventPP = &eventPtr;
3658 eventEndPP = &eventEndPtr;
3659 }
3660 else {
3661 eventPP = &(openInternalEntities->internalEventPtr);
3662 eventEndPP = &(openInternalEntities->internalEventEndPtr);
3663 }
3664
3665 for (;;) {
3666 int role;
3667 XML_Bool handleDefault = XML_TRUE;
3668 *eventPP = s;
3669 *eventEndPP = next;
3670 if (tok <= 0) {
3671 if (haveMore && tok != XML_TOK_INVALID) {
3672 *nextPtr = s;
3673 return XML_ERROR_NONE;
3674 }
3675 switch (tok) {
3676 case XML_TOK_INVALID:
3677 *eventPP = next;
3678 return XML_ERROR_INVALID_TOKEN;
3679 case XML_TOK_PARTIAL:
3680 return XML_ERROR_UNCLOSED_TOKEN;
3681 case XML_TOK_PARTIAL_CHAR:
3682 return XML_ERROR_PARTIAL_CHAR;
3683 case XML_TOK_NONE:
3684 #ifdef XML_DTD
3685 /* for internal PE NOT referenced between declarations */
3686 if (enc != encoding && !openInternalEntities->betweenDecl) {
3687 *nextPtr = s;
3688 return XML_ERROR_NONE;
3689 }
3690 /* WFC: PE Between Declarations - must check that PE contains
3691 complete markup, not only for external PEs, but also for
3692 internal PEs if the reference occurs between declarations.
3693 */
3694 if (isParamEntity || enc != encoding) {
3695 if (XmlTokenRole(&prologState, XML_TOK_NONE, end, end, enc)
3696 == XML_ROLE_ERROR)
3697 return XML_ERROR_INCOMPLETE_PE;
3698 *nextPtr = s;
3699 return XML_ERROR_NONE;
3700 }
3701 #endif /* XML_DTD */
3702 return XML_ERROR_NO_ELEMENTS;
3703 default:
3704 tok = -tok;
3705 next = end;
3706 break;
3707 }
3708 }
3709 role = XmlTokenRole(&prologState, tok, s, next, enc);
3710 switch (role) {
3711 case XML_ROLE_XML_DECL:
3712 {
3713 enum XML_Error result = processXmlDecl(parser, 0, s, next);
3714 if (result != XML_ERROR_NONE)
3715 return result;
3716 enc = encoding;
3717 handleDefault = XML_FALSE;
3718 }
3719 break;
3720 case XML_ROLE_DOCTYPE_NAME:
3721 if (startDoctypeDeclHandler) {
3722 doctypeName = poolStoreString(&tempPool, enc, s, next);
3723 if (!doctypeName)
3724 return XML_ERROR_NO_MEMORY;
3725 poolFinish(&tempPool);
3726 doctypePubid = NULL;
3727 handleDefault = XML_FALSE;
3728 }
3729 doctypeSysid = NULL; /* always initialize to NULL */
3730 break;
3731 case XML_ROLE_DOCTYPE_INTERNAL_SUBSET:
3732 if (startDoctypeDeclHandler) {
3733 startDoctypeDeclHandler(handlerArg, doctypeName, doctypeSysid,
3734 doctypePubid, 1);
3735 doctypeName = NULL;
3736 poolClear(&tempPool);
3737 handleDefault = XML_FALSE;
3738 }
3739 break;
3740 #ifdef XML_DTD
3741 case XML_ROLE_TEXT_DECL:
3742 {
3743 enum XML_Error result = processXmlDecl(parser, 1, s, next);
3744 if (result != XML_ERROR_NONE)
3745 return result;
3746 enc = encoding;
3747 handleDefault = XML_FALSE;
3748 }
3749 break;
3750 #endif /* XML_DTD */
3751 case XML_ROLE_DOCTYPE_PUBLIC_ID:
3752 #ifdef XML_DTD
3753 useForeignDTD = XML_FALSE;
3754 declEntity = (ENTITY *)lookup(&dtd->paramEntities,
3755 externalSubsetName,
3756 sizeof(ENTITY));
3757 if (!declEntity)
3758 return XML_ERROR_NO_MEMORY;
3759 #endif /* XML_DTD */
3760 dtd->hasParamEntityRefs = XML_TRUE;
3761 if (startDoctypeDeclHandler) {
3762 if (!XmlIsPublicId(enc, s, next, eventPP))
3763 return XML_ERROR_PUBLICID;
3764 doctypePubid = poolStoreString(&tempPool, enc,
3765 s + enc->minBytesPerChar,
3766 next - enc->minBytesPerChar);
3767 if (!doctypePubid)
3768 return XML_ERROR_NO_MEMORY;
3769 normalizePublicId((XML_Char *)doctypePubid);
3770 poolFinish(&tempPool);
3771 handleDefault = XML_FALSE;
3772 goto alreadyChecked;
3773 }
3774 /* fall through */
3775 case XML_ROLE_ENTITY_PUBLIC_ID:
3776 if (!XmlIsPublicId(enc, s, next, eventPP))
3777 return XML_ERROR_PUBLICID;
3778 alreadyChecked:
3779 if (dtd->keepProcessing && declEntity) {
3780 XML_Char *tem = poolStoreString(&dtd->pool,
3781 enc,
3782 s + enc->minBytesPerChar,
3783 next - enc->minBytesPerChar);
3784 if (!tem)
3785 return XML_ERROR_NO_MEMORY;
3786 normalizePublicId(tem);
3787 declEntity->publicId = tem;
3788 poolFinish(&dtd->pool);
3789 if (entityDeclHandler)
3790 handleDefault = XML_FALSE;
3791 }
3792 break;
3793 case XML_ROLE_DOCTYPE_CLOSE:
3794 if (doctypeName) {
3795 startDoctypeDeclHandler(handlerArg, doctypeName,
3796 doctypeSysid, doctypePubid, 0);
3797 poolClear(&tempPool);
3798 handleDefault = XML_FALSE;
3799 }
3800 /* doctypeSysid will be non-NULL in the case of a previous
3801 XML_ROLE_DOCTYPE_SYSTEM_ID, even if startDoctypeDeclHandler
3802 was not set, indicating an external subset
3803 */
3804 #ifdef XML_DTD
3805 if (doctypeSysid || useForeignDTD) {
3806 XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs;
3807 dtd->hasParamEntityRefs = XML_TRUE;
3808 if (paramEntityParsing && externalEntityRefHandler) {
3809 ENTITY *entity = (ENTITY *)lookup(&dtd->paramEntities,
3810 externalSubsetName,
3811 sizeof(ENTITY));
3812 if (!entity)
3813 return XML_ERROR_NO_MEMORY;
3814 if (useForeignDTD)
3815 entity->base = curBase;
3816 dtd->paramEntityRead = XML_FALSE;
3817 if (!externalEntityRefHandler(externalEntityRefHandlerArg,
3818 0,
3819 entity->base,
3820 entity->systemId,
3821 entity->publicId))
3822 return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
3823 if (dtd->paramEntityRead) {
3824 if (!dtd->standalone &&
3825 notStandaloneHandler &&
3826 !notStandaloneHandler(handlerArg))
3827 return XML_ERROR_NOT_STANDALONE;
3828 }
3829 /* if we didn't read the foreign DTD then this means that there
3830 is no external subset and we must reset dtd->hasParamEntityRefs
3831 */
3832 else if (!doctypeSysid)
3833 dtd->hasParamEntityRefs = hadParamEntityRefs;
3834 /* end of DTD - no need to update dtd->keepProcessing */
3835 }
3836 useForeignDTD = XML_FALSE;
3837 }
3838 #endif /* XML_DTD */
3839 if (endDoctypeDeclHandler) {
3840 endDoctypeDeclHandler(handlerArg);
3841 handleDefault = XML_FALSE;
3842 }
3843 break;
3844 case XML_ROLE_INSTANCE_START:
3845 #ifdef XML_DTD
3846 /* if there is no DOCTYPE declaration then now is the
3847 last chance to read the foreign DTD
3848 */
3849 if (useForeignDTD) {
3850 XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs;
3851 dtd->hasParamEntityRefs = XML_TRUE;
3852 if (paramEntityParsing && externalEntityRefHandler) {
3853 ENTITY *entity = (ENTITY *)lookup(&dtd->paramEntities,
3854 externalSubsetName,
3855 sizeof(ENTITY));
3856 if (!entity)
3857 return XML_ERROR_NO_MEMORY;
3858 entity->base = curBase;
3859 dtd->paramEntityRead = XML_FALSE;
3860 if (!externalEntityRefHandler(externalEntityRefHandlerArg,
3861 0,
3862 entity->base,
3863 entity->systemId,
3864 entity->publicId))
3865 return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
3866 if (dtd->paramEntityRead) {
3867 if (!dtd->standalone &&
3868 notStandaloneHandler &&
3869 !notStandaloneHandler(handlerArg))
3870 return XML_ERROR_NOT_STANDALONE;
3871 }
3872 /* if we didn't read the foreign DTD then this means that there
3873 is no external subset and we must reset dtd->hasParamEntityRefs
3874 */
3875 else
3876 dtd->hasParamEntityRefs = hadParamEntityRefs;
3877 /* end of DTD - no need to update dtd->keepProcessing */
3878 }
3879 }
3880 #endif /* XML_DTD */
3881 processor = contentProcessor;
3882 return contentProcessor(parser, s, end, nextPtr);
3883 case XML_ROLE_ATTLIST_ELEMENT_NAME:
3884 declElementType = getElementType(parser, enc, s, next);
3885 if (!declElementType)
3886 return XML_ERROR_NO_MEMORY;
3887 goto checkAttListDeclHandler;
3888 case XML_ROLE_ATTRIBUTE_NAME:
3889 declAttributeId = getAttributeId(parser, enc, s, next);
3890 if (!declAttributeId)
3891 return XML_ERROR_NO_MEMORY;
3892 declAttributeIsCdata = XML_FALSE;
3893 declAttributeType = NULL;
3894 declAttributeIsId = XML_FALSE;
3895 goto checkAttListDeclHandler;
3896 case XML_ROLE_ATTRIBUTE_TYPE_CDATA:
3897 declAttributeIsCdata = XML_TRUE;
3898 declAttributeType = atypeCDATA;
3899 goto checkAttListDeclHandler;
3900 case XML_ROLE_ATTRIBUTE_TYPE_ID:
3901 declAttributeIsId = XML_TRUE;
3902 declAttributeType = atypeID;
3903 goto checkAttListDeclHandler;
3904 case XML_ROLE_ATTRIBUTE_TYPE_IDREF:
3905 declAttributeType = atypeIDREF;
3906 goto checkAttListDeclHandler;
3907 case XML_ROLE_ATTRIBUTE_TYPE_IDREFS:
3908 declAttributeType = atypeIDREFS;
3909 goto checkAttListDeclHandler;
3910 case XML_ROLE_ATTRIBUTE_TYPE_ENTITY:
3911 declAttributeType = atypeENTITY;
3912 goto checkAttListDeclHandler;
3913 case XML_ROLE_ATTRIBUTE_TYPE_ENTITIES:
3914 declAttributeType = atypeENTITIES;
3915 goto checkAttListDeclHandler;
3916 case XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN:
3917 declAttributeType = atypeNMTOKEN;
3918 goto checkAttListDeclHandler;
3919 case XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS:
3920 declAttributeType = atypeNMTOKENS;
3921 checkAttListDeclHandler:
3922 if (dtd->keepProcessing && attlistDeclHandler)
3923 handleDefault = XML_FALSE;
3924 break;
3925 case XML_ROLE_ATTRIBUTE_ENUM_VALUE:
3926 case XML_ROLE_ATTRIBUTE_NOTATION_VALUE:
3927 if (dtd->keepProcessing && attlistDeclHandler) {
3928 const XML_Char *prefix;
3929 if (declAttributeType) {
3930 prefix = enumValueSep;
3931 }
3932 else {
3933 prefix = (role == XML_ROLE_ATTRIBUTE_NOTATION_VALUE
3934 ? notationPrefix
3935 : enumValueStart);
3936 }
3937 if (!poolAppendString(&tempPool, prefix))
3938 return XML_ERROR_NO_MEMORY;
3939 if (!poolAppend(&tempPool, enc, s, next))
3940 return XML_ERROR_NO_MEMORY;
3941 declAttributeType = tempPool.start;
3942 handleDefault = XML_FALSE;
3943 }
3944 break;
3945 case XML_ROLE_IMPLIED_ATTRIBUTE_VALUE:
3946 case XML_ROLE_REQUIRED_ATTRIBUTE_VALUE:
3947 if (dtd->keepProcessing) {
3948 if (!defineAttribute(declElementType, declAttributeId,
3949 declAttributeIsCdata, declAttributeIsId,
3950 0, parser))
3951 return XML_ERROR_NO_MEMORY;
3952 if (attlistDeclHandler && declAttributeType) {
3953 if (*declAttributeType == XML_T('(')
3954 || (*declAttributeType == XML_T('N')
3955 && declAttributeType[1] == XML_T('O'))) {
3956 /* Enumerated or Notation type */
3957 if (!poolAppendChar(&tempPool, XML_T(')'))
3958 || !poolAppendChar(&tempPool, XML_T('\0')))
3959 return XML_ERROR_NO_MEMORY;
3960 declAttributeType = tempPool.start;
3961 poolFinish(&tempPool);
3962 }
3963 *eventEndPP = s;
3964 attlistDeclHandler(handlerArg, declElementType->name,
3965 declAttributeId->name, declAttributeType,
3966 0, role == XML_ROLE_REQUIRED_ATTRIBUTE_VALUE);
3967 poolClear(&tempPool);
3968 handleDefault = XML_FALSE;
3969 }
3970 }
3971 break;
3972 case XML_ROLE_DEFAULT_ATTRIBUTE_VALUE:
3973 case XML_ROLE_FIXED_ATTRIBUTE_VALUE:
3974 if (dtd->keepProcessing) {
3975 const XML_Char *attVal;
3976 enum XML_Error result =
3977 storeAttributeValue(parser, enc, declAttributeIsCdata,
3978 s + enc->minBytesPerChar,
3979 next - enc->minBytesPerChar,
3980 &dtd->pool);
3981 if (result)
3982 return result;
3983 attVal = poolStart(&dtd->pool);
3984 poolFinish(&dtd->pool);
3985 /* ID attributes aren't allowed to have a default */
3986 if (!defineAttribute(declElementType, declAttributeId,
3987 declAttributeIsCdata, XML_FALSE, attVal, parser))
3988 return XML_ERROR_NO_MEMORY;
3989 if (attlistDeclHandler && declAttributeType) {
3990 if (*declAttributeType == XML_T('(')
3991 || (*declAttributeType == XML_T('N')
3992 && declAttributeType[1] == XML_T('O'))) {
3993 /* Enumerated or Notation type */
3994 if (!poolAppendChar(&tempPool, XML_T(')'))
3995 || !poolAppendChar(&tempPool, XML_T('\0')))
3996 return XML_ERROR_NO_MEMORY;
3997 declAttributeType = tempPool.start;
3998 poolFinish(&tempPool);
3999 }
4000 *eventEndPP = s;
4001 attlistDeclHandler(handlerArg, declElementType->name,
4002 declAttributeId->name, declAttributeType,
4003 attVal,
4004 role == XML_ROLE_FIXED_ATTRIBUTE_VALUE);
4005 poolClear(&tempPool);
4006 handleDefault = XML_FALSE;
4007 }
4008 }
4009 break;
4010 case XML_ROLE_ENTITY_VALUE:
4011 if (dtd->keepProcessing) {
4012 enum XML_Error result = storeEntityValue(parser, enc,
4013 s + enc->minBytesPerChar,
4014 next - enc->minBytesPerChar);
4015 if (declEntity) {
4016 declEntity->textPtr = poolStart(&dtd->entityValuePool);
4017 declEntity->textLen = (int)(poolLength(&dtd->entityValuePool));
4018 poolFinish(&dtd->entityValuePool);
4019 if (entityDeclHandler) {
4020 *eventEndPP = s;
4021 entityDeclHandler(handlerArg,
4022 declEntity->name,
4023 declEntity->is_param,
4024 declEntity->textPtr,
4025 declEntity->textLen,
4026 curBase, 0, 0, 0);
4027 handleDefault = XML_FALSE;
4028 }
4029 }
4030 else
4031 poolDiscard(&dtd->entityValuePool);
4032 if (result != XML_ERROR_NONE)
4033 return result;
4034 }
4035 break;
4036 case XML_ROLE_DOCTYPE_SYSTEM_ID:
4037 #ifdef XML_DTD
4038 useForeignDTD = XML_FALSE;
4039 #endif /* XML_DTD */
4040 dtd->hasParamEntityRefs = XML_TRUE;
4041 if (startDoctypeDeclHandler) {
4042 doctypeSysid = poolStoreString(&tempPool, enc,
4043 s + enc->minBytesPerChar,
4044 next - enc->minBytesPerChar);
4045 if (doctypeSysid == NULL)
4046 return XML_ERROR_NO_MEMORY;
4047 poolFinish(&tempPool);
4048 handleDefault = XML_FALSE;
4049 }
4050 #ifdef XML_DTD
4051 else
4052 /* use externalSubsetName to make doctypeSysid non-NULL
4053 for the case where no startDoctypeDeclHandler is set */
4054 doctypeSysid = externalSubsetName;
4055 #endif /* XML_DTD */
4056 if (!dtd->standalone
4057 #ifdef XML_DTD
4058 && !paramEntityParsing
4059 #endif /* XML_DTD */
4060 && notStandaloneHandler
4061 && !notStandaloneHandler(handlerArg))
4062 return XML_ERROR_NOT_STANDALONE;
4063 #ifndef XML_DTD
4064 break;
4065 #else /* XML_DTD */
4066 if (!declEntity) {
4067 declEntity = (ENTITY *)lookup(&dtd->paramEntities,
4068 externalSubsetName,
4069 sizeof(ENTITY));
4070 if (!declEntity)
4071 return XML_ERROR_NO_MEMORY;
4072 declEntity->publicId = NULL;
4073 }
4074 /* fall through */
4075 #endif /* XML_DTD */
4076 case XML_ROLE_ENTITY_SYSTEM_ID:
4077 if (dtd->keepProcessing && declEntity) {
4078 declEntity->systemId = poolStoreString(&dtd->pool, enc,
4079 s + enc->minBytesPerChar,
4080 next - enc->minBytesPerChar);
4081 if (!declEntity->systemId)
4082 return XML_ERROR_NO_MEMORY;
4083 declEntity->base = curBase;
4084 poolFinish(&dtd->pool);
4085 if (entityDeclHandler)
4086 handleDefault = XML_FALSE;
4087 }
4088 break;
4089 case XML_ROLE_ENTITY_COMPLETE:
4090 if (dtd->keepProcessing && declEntity && entityDeclHandler) {
4091 *eventEndPP = s;
4092 entityDeclHandler(handlerArg,
4093 declEntity->name,
4094 declEntity->is_param,
4095 0,0,
4096 declEntity->base,
4097 declEntity->systemId,
4098 declEntity->publicId,
4099 0);
4100 handleDefault = XML_FALSE;
4101 }
4102 break;
4103 case XML_ROLE_ENTITY_NOTATION_NAME:
4104 if (dtd->keepProcessing && declEntity) {
4105 declEntity->notation = poolStoreString(&dtd->pool, enc, s, next);
4106 if (!declEntity->notation)
4107 return XML_ERROR_NO_MEMORY;
4108 poolFinish(&dtd->pool);
4109 if (unparsedEntityDeclHandler) {
4110 *eventEndPP = s;
4111 unparsedEntityDeclHandler(handlerArg,
4112 declEntity->name,
4113 declEntity->base,
4114 declEntity->systemId,
4115 declEntity->publicId,
4116 declEntity->notation);
4117 handleDefault = XML_FALSE;
4118 }
4119 else if (entityDeclHandler) {
4120 *eventEndPP = s;
4121 entityDeclHandler(handlerArg,
4122 declEntity->name,
4123 0,0,0,
4124 declEntity->base,
4125 declEntity->systemId,
4126 declEntity->publicId,
4127 declEntity->notation);
4128 handleDefault = XML_FALSE;
4129 }
4130 }
4131 break;
4132 case XML_ROLE_GENERAL_ENTITY_NAME:
4133 {
4134 if (XmlPredefinedEntityName(enc, s, next)) {
4135 declEntity = NULL;
4136 break;
4137 }
4138 if (dtd->keepProcessing) {
4139 const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next);
4140 if (!name)
4141 return XML_ERROR_NO_MEMORY;
4142 declEntity = (ENTITY *)lookup(&dtd->generalEntities, name,
4143 sizeof(ENTITY));
4144 if (!declEntity)
4145 return XML_ERROR_NO_MEMORY;
4146 if (declEntity->name != name) {
4147 poolDiscard(&dtd->pool);
4148 declEntity = NULL;
4149 }
4150 else {
4151 poolFinish(&dtd->pool);
4152 declEntity->publicId = NULL;
4153 declEntity->is_param = XML_FALSE;
4154 /* if we have a parent parser or are reading an internal parameter
4155 entity, then the entity declaration is not considered "internal"
4156 */
4157 declEntity->is_internal = !(parentParser || openInternalEntities);
4158 if (entityDeclHandler)
4159 handleDefault = XML_FALSE;
4160 }
4161 }
4162 else {
4163 poolDiscard(&dtd->pool);
4164 declEntity = NULL;
4165 }
4166 }
4167 break;
4168 case XML_ROLE_PARAM_ENTITY_NAME:
4169 #ifdef XML_DTD
4170 if (dtd->keepProcessing) {
4171 const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next);
4172 if (!name)
4173 return XML_ERROR_NO_MEMORY;
4174 declEntity = (ENTITY *)lookup(&dtd->paramEntities,
4175 name, sizeof(ENTITY));
4176 if (!declEntity)
4177 return XML_ERROR_NO_MEMORY;
4178 if (declEntity->name != name) {
4179 poolDiscard(&dtd->pool);
4180 declEntity = NULL;
4181 }
4182 else {
4183 poolFinish(&dtd->pool);
4184 declEntity->publicId = NULL;
4185 declEntity->is_param = XML_TRUE;
4186 /* if we have a parent parser or are reading an internal parameter
4187 entity, then the entity declaration is not considered "internal"
4188 */
4189 declEntity->is_internal = !(parentParser || openInternalEntities);
4190 if (entityDeclHandler)
4191 handleDefault = XML_FALSE;
4192 }
4193 }
4194 else {
4195 poolDiscard(&dtd->pool);
4196 declEntity = NULL;
4197 }
4198 #else /* not XML_DTD */
4199 declEntity = NULL;
4200 #endif /* XML_DTD */
4201 break;
4202 case XML_ROLE_NOTATION_NAME:
4203 declNotationPublicId = NULL;
4204 declNotationName = NULL;
4205 if (notationDeclHandler) {
4206 declNotationName = poolStoreString(&tempPool, enc, s, next);
4207 if (!declNotationName)
4208 return XML_ERROR_NO_MEMORY;
4209 poolFinish(&tempPool);
4210 handleDefault = XML_FALSE;
4211 }
4212 break;
4213 case XML_ROLE_NOTATION_PUBLIC_ID:
4214 if (!XmlIsPublicId(enc, s, next, eventPP))
4215 return XML_ERROR_PUBLICID;
4216 if (declNotationName) { /* means notationDeclHandler != NULL */
4217 XML_Char *tem = poolStoreString(&tempPool,
4218 enc,
4219 s + enc->minBytesPerChar,
4220 next - enc->minBytesPerChar);
4221 if (!tem)
4222 return XML_ERROR_NO_MEMORY;
4223 normalizePublicId(tem);
4224 declNotationPublicId = tem;
4225 poolFinish(&tempPool);
4226 handleDefault = XML_FALSE;
4227 }
4228 break;
4229 case XML_ROLE_NOTATION_SYSTEM_ID:
4230 if (declNotationName && notationDeclHandler) {
4231 const XML_Char *systemId
4232 = poolStoreString(&tempPool, enc,
4233 s + enc->minBytesPerChar,
4234 next - enc->minBytesPerChar);
4235 if (!systemId)
4236 return XML_ERROR_NO_MEMORY;
4237 *eventEndPP = s;
4238 notationDeclHandler(handlerArg,
4239 declNotationName,
4240 curBase,
4241 systemId,
4242 declNotationPublicId);
4243 handleDefault = XML_FALSE;
4244 }
4245 poolClear(&tempPool);
4246 break;
4247 case XML_ROLE_NOTATION_NO_SYSTEM_ID:
4248 if (declNotationPublicId && notationDeclHandler) {
4249 *eventEndPP = s;
4250 notationDeclHandler(handlerArg,
4251 declNotationName,
4252 curBase,
4253 0,
4254 declNotationPublicId);
4255 handleDefault = XML_FALSE;
4256 }
4257 poolClear(&tempPool);
4258 break;
4259 case XML_ROLE_ERROR:
4260 switch (tok) {
4261 case XML_TOK_PARAM_ENTITY_REF:
4262 /* PE references in internal subset are
4263 not allowed within declarations. */
4264 return XML_ERROR_PARAM_ENTITY_REF;
4265 case XML_TOK_XML_DECL:
4266 return XML_ERROR_MISPLACED_XML_PI;
4267 default:
4268 return XML_ERROR_SYNTAX;
4269 }
4270 #ifdef XML_DTD
4271 case XML_ROLE_IGNORE_SECT:
4272 {
4273 enum XML_Error result;
4274 if (defaultHandler)
4275 reportDefault(parser, enc, s, next);
4276 handleDefault = XML_FALSE;
4277 result = doIgnoreSection(parser, enc, &next, end, nextPtr, haveMore);
4278 if (result != XML_ERROR_NONE)
4279 return result;
4280 else if (!next) {
4281 processor = ignoreSectionProcessor;
4282 return result;
4283 }
4284 }
4285 break;
4286 #endif /* XML_DTD */
4287 case XML_ROLE_GROUP_OPEN:
4288 if (prologState.level >= groupSize) {
4289 if (groupSize) {
4290 char *temp = (char *)REALLOC(groupConnector, groupSize *= 2);
4291 if (temp == NULL)
4292 return XML_ERROR_NO_MEMORY;
4293 groupConnector = temp;
4294 if (dtd->scaffIndex) {
4295 int *temp = (int *)REALLOC(dtd->scaffIndex,
4296 groupSize * sizeof(int));
4297 if (temp == NULL)
4298 return XML_ERROR_NO_MEMORY;
4299 dtd->scaffIndex = temp;
4300 }
4301 }
4302 else {
4303 groupConnector = (char *)MALLOC(groupSize = 32);
4304 if (!groupConnector)
4305 return XML_ERROR_NO_MEMORY;
4306 }
4307 }
4308 groupConnector[prologState.level] = 0;
4309 if (dtd->in_eldecl) {
4310 int myindex = nextScaffoldPart(parser);
4311 if (myindex < 0)
4312 return XML_ERROR_NO_MEMORY;
4313 dtd->scaffIndex[dtd->scaffLevel] = myindex;
4314 dtd->scaffLevel++;
4315 dtd->scaffold[myindex].type = XML_CTYPE_SEQ;
4316 if (elementDeclHandler)
4317 handleDefault = XML_FALSE;
4318 }
4319 break;
4320 case XML_ROLE_GROUP_SEQUENCE:
4321 if (groupConnector[prologState.level] == '|')
4322 return XML_ERROR_SYNTAX;
4323 groupConnector[prologState.level] = ',';
4324 if (dtd->in_eldecl && elementDeclHandler)
4325 handleDefault = XML_FALSE;
4326 break;
4327 case XML_ROLE_GROUP_CHOICE:
4328 if (groupConnector[prologState.level] == ',')
4329 return XML_ERROR_SYNTAX;
4330 if (dtd->in_eldecl
4331 && !groupConnector[prologState.level]
4332 && (dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type
4333 != XML_CTYPE_MIXED)
4334 ) {
4335 dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type
4336 = XML_CTYPE_CHOICE;
4337 if (elementDeclHandler)
4338 handleDefault = XML_FALSE;
4339 }
4340 groupConnector[prologState.level] = '|';
4341 break;
4342 case XML_ROLE_PARAM_ENTITY_REF:
4343 #ifdef XML_DTD
4344 case XML_ROLE_INNER_PARAM_ENTITY_REF:
4345 dtd->hasParamEntityRefs = XML_TRUE;
4346 if (!paramEntityParsing)
4347 dtd->keepProcessing = dtd->standalone;
4348 else {
4349 const XML_Char *name;
4350 ENTITY *entity;
4351 name = poolStoreString(&dtd->pool, enc,
4352 s + enc->minBytesPerChar,
4353 next - enc->minBytesPerChar);
4354 if (!name)
4355 return XML_ERROR_NO_MEMORY;
4356 entity = (ENTITY *)lookup(&dtd->paramEntities, name, 0);
4357 poolDiscard(&dtd->pool);
4358 /* first, determine if a check for an existing declaration is needed;
4359 if yes, check that the entity exists, and that it is internal,
4360 otherwise call the skipped entity handler
4361 */
4362 if (prologState.documentEntity &&
4363 (dtd->standalone
4364 ? !openInternalEntities
4365 : !dtd->hasParamEntityRefs)) {
4366 if (!entity)
4367 return XML_ERROR_UNDEFINED_ENTITY;
4368 else if (!entity->is_internal)
4369 return XML_ERROR_ENTITY_DECLARED_IN_PE;
4370 }
4371 else if (!entity) {
4372 dtd->keepProcessing = dtd->standalone;
4373 /* cannot report skipped entities in declarations */
4374 if ((role == XML_ROLE_PARAM_ENTITY_REF) && skippedEntityHandler) {
4375 skippedEntityHandler(handlerArg, name, 1);
4376 handleDefault = XML_FALSE;
4377 }
4378 break;
4379 }
4380 if (entity->open)
4381 return XML_ERROR_RECURSIVE_ENTITY_REF;
4382 if (entity->textPtr) {
4383 enum XML_Error result;
4384 XML_Bool betweenDecl =
4385 (role == XML_ROLE_PARAM_ENTITY_REF ? XML_TRUE : XML_FALSE);
4386 result = processInternalEntity(parser, entity, betweenDecl);
4387 if (result != XML_ERROR_NONE)
4388 return result;
4389 handleDefault = XML_FALSE;
4390 break;
4391 }
4392 if (externalEntityRefHandler) {
4393 dtd->paramEntityRead = XML_FALSE;
4394 entity->open = XML_TRUE;
4395 if (!externalEntityRefHandler(externalEntityRefHandlerArg,
4396 0,
4397 entity->base,
4398 entity->systemId,
4399 entity->publicId)) {
4400 entity->open = XML_FALSE;
4401 return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
4402 }
4403 entity->open = XML_FALSE;
4404 handleDefault = XML_FALSE;
4405 if (!dtd->paramEntityRead) {
4406 dtd->keepProcessing = dtd->standalone;
4407 break;
4408 }
4409 }
4410 else {
4411 dtd->keepProcessing = dtd->standalone;
4412 break;
4413 }
4414 }
4415 #endif /* XML_DTD */
4416 if (!dtd->standalone &&
4417 notStandaloneHandler &&
4418 !notStandaloneHandler(handlerArg))
4419 return XML_ERROR_NOT_STANDALONE;
4420 break;
4421
4422 /* Element declaration stuff */
4423
4424 case XML_ROLE_ELEMENT_NAME:
4425 if (elementDeclHandler) {
4426 declElementType = getElementType(parser, enc, s, next);
4427 if (!declElementType)
4428 return XML_ERROR_NO_MEMORY;
4429 dtd->scaffLevel = 0;
4430 dtd->scaffCount = 0;
4431 dtd->in_eldecl = XML_TRUE;
4432 handleDefault = XML_FALSE;
4433 }
4434 break;
4435
4436 case XML_ROLE_CONTENT_ANY:
4437 case XML_ROLE_CONTENT_EMPTY:
4438 if (dtd->in_eldecl) {
4439 if (elementDeclHandler) {
4440 XML_Content * content = (XML_Content *) MALLOC(sizeof(XML_Content));
4441 if (!content)
4442 return XML_ERROR_NO_MEMORY;
4443 content->quant = XML_CQUANT_NONE;
4444 content->name = NULL;
4445 content->numchildren = 0;
4446 content->children = NULL;
4447 content->type = ((role == XML_ROLE_CONTENT_ANY) ?
4448 XML_CTYPE_ANY :
4449 XML_CTYPE_EMPTY);
4450 *eventEndPP = s;
4451 elementDeclHandler(handlerArg, declElementType->name, content);
4452 handleDefault = XML_FALSE;
4453 }
4454 dtd->in_eldecl = XML_FALSE;
4455 }
4456 break;
4457
4458 case XML_ROLE_CONTENT_PCDATA:
4459 if (dtd->in_eldecl) {
4460 dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type
4461 = XML_CTYPE_MIXED;
4462 if (elementDeclHandler)
4463 handleDefault = XML_FALSE;
4464 }
4465 break;
4466
4467 case XML_ROLE_CONTENT_ELEMENT:
4468 quant = XML_CQUANT_NONE;
4469 goto elementContent;
4470 case XML_ROLE_CONTENT_ELEMENT_OPT:
4471 quant = XML_CQUANT_OPT;
4472 goto elementContent;
4473 case XML_ROLE_CONTENT_ELEMENT_REP:
4474 quant = XML_CQUANT_REP;
4475 goto elementContent;
4476 case XML_ROLE_CONTENT_ELEMENT_PLUS:
4477 quant = XML_CQUANT_PLUS;
4478 elementContent:
4479 if (dtd->in_eldecl) {
4480 ELEMENT_TYPE *el;
4481 const XML_Char *name;
4482 int nameLen;
4483 const char *nxt = (quant == XML_CQUANT_NONE
4484 ? next
4485 : next - enc->minBytesPerChar);
4486 int myindex = nextScaffoldPart(parser);
4487 if (myindex < 0)
4488 return XML_ERROR_NO_MEMORY;
4489 dtd->scaffold[myindex].type = XML_CTYPE_NAME;
4490 dtd->scaffold[myindex].quant = quant;
4491 el = getElementType(parser, enc, s, nxt);
4492 if (!el)
4493 return XML_ERROR_NO_MEMORY;
4494 name = el->name;
4495 dtd->scaffold[myindex].name = name;
4496 nameLen = 0;
4497 for (; name[nameLen++]; );
4498 dtd->contentStringLen += nameLen;
4499 if (elementDeclHandler)
4500 handleDefault = XML_FALSE;
4501 }
4502 break;
4503
4504 case XML_ROLE_GROUP_CLOSE:
4505 quant = XML_CQUANT_NONE;
4506 goto closeGroup;
4507 case XML_ROLE_GROUP_CLOSE_OPT:
4508 quant = XML_CQUANT_OPT;
4509 goto closeGroup;
4510 case XML_ROLE_GROUP_CLOSE_REP:
4511 quant = XML_CQUANT_REP;
4512 goto closeGroup;
4513 case XML_ROLE_GROUP_CLOSE_PLUS:
4514 quant = XML_CQUANT_PLUS;
4515 closeGroup:
4516 if (dtd->in_eldecl) {
4517 if (elementDeclHandler)
4518 handleDefault = XML_FALSE;
4519 dtd->scaffLevel--;
4520 dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel]].quant = quant;
4521 if (dtd->scaffLevel == 0) {
4522 if (!handleDefault) {
4523 XML_Content *model = build_model(parser);
4524 if (!model)
4525 return XML_ERROR_NO_MEMORY;
4526 *eventEndPP = s;
4527 elementDeclHandler(handlerArg, declElementType->name, model);
4528 }
4529 dtd->in_eldecl = XML_FALSE;
4530 dtd->contentStringLen = 0;
4531 }
4532 }
4533 break;
4534 /* End element declaration stuff */
4535
4536 case XML_ROLE_PI:
4537 if (!reportProcessingInstruction(parser, enc, s, next))
4538 return XML_ERROR_NO_MEMORY;
4539 handleDefault = XML_FALSE;
4540 break;
4541 case XML_ROLE_COMMENT:
4542 if (!reportComment(parser, enc, s, next))
4543 return XML_ERROR_NO_MEMORY;
4544 handleDefault = XML_FALSE;
4545 break;
4546 case XML_ROLE_NONE:
4547 switch (tok) {
4548 case XML_TOK_BOM:
4549 handleDefault = XML_FALSE;
4550 break;
4551 }
4552 break;
4553 case XML_ROLE_DOCTYPE_NONE:
4554 if (startDoctypeDeclHandler)
4555 handleDefault = XML_FALSE;
4556 break;
4557 case XML_ROLE_ENTITY_NONE:
4558 if (dtd->keepProcessing && entityDeclHandler)
4559 handleDefault = XML_FALSE;
4560 break;
4561 case XML_ROLE_NOTATION_NONE:
4562 if (notationDeclHandler)
4563 handleDefault = XML_FALSE;
4564 break;
4565 case XML_ROLE_ATTLIST_NONE:
4566 if (dtd->keepProcessing && attlistDeclHandler)
4567 handleDefault = XML_FALSE;
4568 break;
4569 case XML_ROLE_ELEMENT_NONE:
4570 if (elementDeclHandler)
4571 handleDefault = XML_FALSE;
4572 break;
4573 } /* end of big switch */
4574
4575 if (handleDefault && defaultHandler)
4576 reportDefault(parser, enc, s, next);
4577
4578 switch (ps_parsing) {
4579 case XML_SUSPENDED:
4580 *nextPtr = next;
4581 return XML_ERROR_NONE;
4582 case XML_FINISHED:
4583 return XML_ERROR_ABORTED;
4584 default:
4585 s = next;
4586 tok = XmlPrologTok(enc, s, end, &next);
4587 }
4588 }
4589 /* not reached */
4590 }
4591
4592 static enum XML_Error PTRCALL
epilogProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)4593 epilogProcessor(XML_Parser parser,
4594 const char *s,
4595 const char *end,
4596 const char **nextPtr)
4597 {
4598 processor = epilogProcessor;
4599 eventPtr = s;
4600 for (;;) {
4601 const char *next = NULL;
4602 int tok = XmlPrologTok(encoding, s, end, &next);
4603 eventEndPtr = next;
4604 switch (tok) {
4605 /* report partial linebreak - it might be the last token */
4606 case -XML_TOK_PROLOG_S:
4607 if (defaultHandler) {
4608 reportDefault(parser, encoding, s, next);
4609 if (ps_parsing == XML_FINISHED)
4610 return XML_ERROR_ABORTED;
4611 }
4612 *nextPtr = next;
4613 return XML_ERROR_NONE;
4614 case XML_TOK_NONE:
4615 *nextPtr = s;
4616 return XML_ERROR_NONE;
4617 case XML_TOK_PROLOG_S:
4618 if (defaultHandler)
4619 reportDefault(parser, encoding, s, next);
4620 break;
4621 case XML_TOK_PI:
4622 if (!reportProcessingInstruction(parser, encoding, s, next))
4623 return XML_ERROR_NO_MEMORY;
4624 break;
4625 case XML_TOK_COMMENT:
4626 if (!reportComment(parser, encoding, s, next))
4627 return XML_ERROR_NO_MEMORY;
4628 break;
4629 case XML_TOK_INVALID:
4630 eventPtr = next;
4631 return XML_ERROR_INVALID_TOKEN;
4632 case XML_TOK_PARTIAL:
4633 if (!ps_finalBuffer) {
4634 *nextPtr = s;
4635 return XML_ERROR_NONE;
4636 }
4637 return XML_ERROR_UNCLOSED_TOKEN;
4638 case XML_TOK_PARTIAL_CHAR:
4639 if (!ps_finalBuffer) {
4640 *nextPtr = s;
4641 return XML_ERROR_NONE;
4642 }
4643 return XML_ERROR_PARTIAL_CHAR;
4644 default:
4645 return XML_ERROR_JUNK_AFTER_DOC_ELEMENT;
4646 }
4647 eventPtr = s = next;
4648 switch (ps_parsing) {
4649 case XML_SUSPENDED:
4650 *nextPtr = next;
4651 return XML_ERROR_NONE;
4652 case XML_FINISHED:
4653 return XML_ERROR_ABORTED;
4654 default: ;
4655 }
4656 }
4657 }
4658
4659 static enum XML_Error
processInternalEntity(XML_Parser parser,ENTITY * entity,XML_Bool betweenDecl)4660 processInternalEntity(XML_Parser parser, ENTITY *entity,
4661 XML_Bool betweenDecl)
4662 {
4663 const char *textStart, *textEnd;
4664 const char *next;
4665 enum XML_Error result;
4666 OPEN_INTERNAL_ENTITY *openEntity;
4667
4668 if (freeInternalEntities) {
4669 openEntity = freeInternalEntities;
4670 freeInternalEntities = openEntity->next;
4671 }
4672 else {
4673 openEntity = (OPEN_INTERNAL_ENTITY *)MALLOC(sizeof(OPEN_INTERNAL_ENTITY));
4674 if (!openEntity)
4675 return XML_ERROR_NO_MEMORY;
4676 }
4677 entity->open = XML_TRUE;
4678 entity->processed = 0;
4679 openEntity->next = openInternalEntities;
4680 openInternalEntities = openEntity;
4681 openEntity->entity = entity;
4682 openEntity->startTagLevel = tagLevel;
4683 openEntity->betweenDecl = betweenDecl;
4684 openEntity->internalEventPtr = NULL;
4685 openEntity->internalEventEndPtr = NULL;
4686 textStart = (char *)entity->textPtr;
4687 textEnd = (char *)(entity->textPtr + entity->textLen);
4688
4689 #ifdef XML_DTD
4690 if (entity->is_param) {
4691 int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next);
4692 result = doProlog(parser, internalEncoding, textStart, textEnd, tok,
4693 next, &next, XML_FALSE);
4694 }
4695 else
4696 #endif /* XML_DTD */
4697 result = doContent(parser, tagLevel, internalEncoding, textStart,
4698 textEnd, &next, XML_FALSE);
4699
4700 if (result == XML_ERROR_NONE) {
4701 if (textEnd != next && ps_parsing == XML_SUSPENDED) {
4702 entity->processed = (int)(next - textStart);
4703 processor = internalEntityProcessor;
4704 }
4705 else {
4706 entity->open = XML_FALSE;
4707 openInternalEntities = openEntity->next;
4708 /* put openEntity back in list of free instances */
4709 openEntity->next = freeInternalEntities;
4710 freeInternalEntities = openEntity;
4711 }
4712 }
4713 return result;
4714 }
4715
4716 static enum XML_Error PTRCALL
internalEntityProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)4717 internalEntityProcessor(XML_Parser parser,
4718 const char *s,
4719 const char *end,
4720 const char **nextPtr)
4721 {
4722 ENTITY *entity;
4723 const char *textStart, *textEnd;
4724 const char *next;
4725 enum XML_Error result;
4726 OPEN_INTERNAL_ENTITY *openEntity = openInternalEntities;
4727 if (!openEntity)
4728 return XML_ERROR_UNEXPECTED_STATE;
4729
4730 entity = openEntity->entity;
4731 textStart = ((char *)entity->textPtr) + entity->processed;
4732 textEnd = (char *)(entity->textPtr + entity->textLen);
4733
4734 #ifdef XML_DTD
4735 if (entity->is_param) {
4736 int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next);
4737 result = doProlog(parser, internalEncoding, textStart, textEnd, tok,
4738 next, &next, XML_FALSE);
4739 }
4740 else
4741 #endif /* XML_DTD */
4742 result = doContent(parser, openEntity->startTagLevel, internalEncoding,
4743 textStart, textEnd, &next, XML_FALSE);
4744
4745 if (result != XML_ERROR_NONE)
4746 return result;
4747 else if (textEnd != next && ps_parsing == XML_SUSPENDED) {
4748 entity->processed = (int)(next - (char *)entity->textPtr);
4749 return result;
4750 }
4751 else {
4752 entity->open = XML_FALSE;
4753 openInternalEntities = openEntity->next;
4754 /* put openEntity back in list of free instances */
4755 openEntity->next = freeInternalEntities;
4756 freeInternalEntities = openEntity;
4757 }
4758
4759 #ifdef XML_DTD
4760 if (entity->is_param) {
4761 int tok;
4762 processor = prologProcessor;
4763 tok = XmlPrologTok(encoding, s, end, &next);
4764 return doProlog(parser, encoding, s, end, tok, next, nextPtr,
4765 (XML_Bool)!ps_finalBuffer);
4766 }
4767 else
4768 #endif /* XML_DTD */
4769 {
4770 processor = contentProcessor;
4771 /* see externalEntityContentProcessor vs contentProcessor */
4772 return doContent(parser, parentParser ? 1 : 0, encoding, s, end,
4773 nextPtr, (XML_Bool)!ps_finalBuffer);
4774 }
4775 }
4776
4777 static enum XML_Error PTRCALL
errorProcessor(XML_Parser parser,const char * s,const char * end,const char ** nextPtr)4778 errorProcessor(XML_Parser parser,
4779 const char *s,
4780 const char *end,
4781 const char **nextPtr)
4782 {
4783 return errorCode;
4784 }
4785
4786 static enum XML_Error
storeAttributeValue(XML_Parser parser,const ENCODING * enc,XML_Bool isCdata,const char * ptr,const char * end,STRING_POOL * pool)4787 storeAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata,
4788 const char *ptr, const char *end,
4789 STRING_POOL *pool)
4790 {
4791 enum XML_Error result = appendAttributeValue(parser, enc, isCdata, ptr,
4792 end, pool);
4793 if (result)
4794 return result;
4795 if (!isCdata && poolLength(pool) && poolLastChar(pool) == 0x20)
4796 poolChop(pool);
4797 if (!poolAppendChar(pool, XML_T('\0')))
4798 return XML_ERROR_NO_MEMORY;
4799 return XML_ERROR_NONE;
4800 }
4801
4802 static enum XML_Error
appendAttributeValue(XML_Parser parser,const ENCODING * enc,XML_Bool isCdata,const char * ptr,const char * end,STRING_POOL * pool)4803 appendAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata,
4804 const char *ptr, const char *end,
4805 STRING_POOL *pool)
4806 {
4807 DTD * const dtd = _dtd; /* save one level of indirection */
4808 for (;;) {
4809 const char *next;
4810 int tok = XmlAttributeValueTok(enc, ptr, end, &next);
4811 switch (tok) {
4812 case XML_TOK_NONE:
4813 return XML_ERROR_NONE;
4814 case XML_TOK_INVALID:
4815 if (enc == encoding)
4816 eventPtr = next;
4817 return XML_ERROR_INVALID_TOKEN;
4818 case XML_TOK_PARTIAL:
4819 if (enc == encoding)
4820 eventPtr = ptr;
4821 return XML_ERROR_INVALID_TOKEN;
4822 case XML_TOK_CHAR_REF:
4823 {
4824 XML_Char buf[XML_ENCODE_MAX];
4825 int i;
4826 int n = XmlCharRefNumber(enc, ptr);
4827 if (n < 0) {
4828 if (enc == encoding)
4829 eventPtr = ptr;
4830 return XML_ERROR_BAD_CHAR_REF;
4831 }
4832 if (!isCdata
4833 && n == 0x20 /* space */
4834 && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20))
4835 break;
4836 n = XmlEncode(n, (ICHAR *)buf);
4837 if (!n) {
4838 if (enc == encoding)
4839 eventPtr = ptr;
4840 return XML_ERROR_BAD_CHAR_REF;
4841 }
4842 for (i = 0; i < n; i++) {
4843 if (!poolAppendChar(pool, buf[i]))
4844 return XML_ERROR_NO_MEMORY;
4845 }
4846 }
4847 break;
4848 case XML_TOK_DATA_CHARS:
4849 if (!poolAppend(pool, enc, ptr, next))
4850 return XML_ERROR_NO_MEMORY;
4851 break;
4852 case XML_TOK_TRAILING_CR:
4853 next = ptr + enc->minBytesPerChar;
4854 /* fall through */
4855 case XML_TOK_ATTRIBUTE_VALUE_S:
4856 case XML_TOK_DATA_NEWLINE:
4857 if (!isCdata && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20))
4858 break;
4859 if (!poolAppendChar(pool, 0x20))
4860 return XML_ERROR_NO_MEMORY;
4861 break;
4862 case XML_TOK_ENTITY_REF:
4863 {
4864 const XML_Char *name;
4865 ENTITY *entity;
4866 char checkEntityDecl;
4867 XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc,
4868 ptr + enc->minBytesPerChar,
4869 next - enc->minBytesPerChar);
4870 if (ch) {
4871 if (!poolAppendChar(pool, ch))
4872 return XML_ERROR_NO_MEMORY;
4873 break;
4874 }
4875 name = poolStoreString(&temp2Pool, enc,
4876 ptr + enc->minBytesPerChar,
4877 next - enc->minBytesPerChar);
4878 if (!name)
4879 return XML_ERROR_NO_MEMORY;
4880 entity = (ENTITY *)lookup(&dtd->generalEntities, name, 0);
4881 poolDiscard(&temp2Pool);
4882 /* First, determine if a check for an existing declaration is needed;
4883 if yes, check that the entity exists, and that it is internal.
4884 */
4885 if (pool == &dtd->pool) /* are we called from prolog? */
4886 checkEntityDecl =
4887 #ifdef XML_DTD
4888 prologState.documentEntity &&
4889 #endif /* XML_DTD */
4890 (dtd->standalone
4891 ? !openInternalEntities
4892 : !dtd->hasParamEntityRefs);
4893 else /* if (pool == &tempPool): we are called from content */
4894 checkEntityDecl = !dtd->hasParamEntityRefs || dtd->standalone;
4895 if (checkEntityDecl) {
4896 if (!entity)
4897 return XML_ERROR_UNDEFINED_ENTITY;
4898 else if (!entity->is_internal)
4899 return XML_ERROR_ENTITY_DECLARED_IN_PE;
4900 }
4901 else if (!entity) {
4902 /* Cannot report skipped entity here - see comments on
4903 skippedEntityHandler.
4904 if (skippedEntityHandler)
4905 skippedEntityHandler(handlerArg, name, 0);
4906 */
4907 /* Cannot call the default handler because this would be
4908 out of sync with the call to the startElementHandler.
4909 if ((pool == &tempPool) && defaultHandler)
4910 reportDefault(parser, enc, ptr, next);
4911 */
4912 break;
4913 }
4914 if (entity->open) {
4915 if (enc == encoding)
4916 eventPtr = ptr;
4917 return XML_ERROR_RECURSIVE_ENTITY_REF;
4918 }
4919 if (entity->notation) {
4920 if (enc == encoding)
4921 eventPtr = ptr;
4922 return XML_ERROR_BINARY_ENTITY_REF;
4923 }
4924 if (!entity->textPtr) {
4925 if (enc == encoding)
4926 eventPtr = ptr;
4927 return XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF;
4928 }
4929 else {
4930 enum XML_Error result;
4931 const XML_Char *textEnd = entity->textPtr + entity->textLen;
4932 entity->open = XML_TRUE;
4933 result = appendAttributeValue(parser, internalEncoding, isCdata,
4934 (char *)entity->textPtr,
4935 (char *)textEnd, pool);
4936 entity->open = XML_FALSE;
4937 if (result)
4938 return result;
4939 }
4940 }
4941 break;
4942 default:
4943 if (enc == encoding)
4944 eventPtr = ptr;
4945 return XML_ERROR_UNEXPECTED_STATE;
4946 }
4947 ptr = next;
4948 }
4949 /* not reached */
4950 }
4951
4952 static enum XML_Error
storeEntityValue(XML_Parser parser,const ENCODING * enc,const char * entityTextPtr,const char * entityTextEnd)4953 storeEntityValue(XML_Parser parser,
4954 const ENCODING *enc,
4955 const char *entityTextPtr,
4956 const char *entityTextEnd)
4957 {
4958 DTD * const dtd = _dtd; /* save one level of indirection */
4959 STRING_POOL *pool = &(dtd->entityValuePool);
4960 enum XML_Error result = XML_ERROR_NONE;
4961 #ifdef XML_DTD
4962 int oldInEntityValue = prologState.inEntityValue;
4963 prologState.inEntityValue = 1;
4964 #endif /* XML_DTD */
4965 /* never return Null for the value argument in EntityDeclHandler,
4966 since this would indicate an external entity; therefore we
4967 have to make sure that entityValuePool.start is not null */
4968 if (!pool->blocks) {
4969 if (!poolGrow(pool))
4970 return XML_ERROR_NO_MEMORY;
4971 }
4972
4973 for (;;) {
4974 const char *next;
4975 int tok = XmlEntityValueTok(enc, entityTextPtr, entityTextEnd, &next);
4976 switch (tok) {
4977 case XML_TOK_PARAM_ENTITY_REF:
4978 #ifdef XML_DTD
4979 if (isParamEntity || enc != encoding) {
4980 const XML_Char *name;
4981 ENTITY *entity;
4982 name = poolStoreString(&tempPool, enc,
4983 entityTextPtr + enc->minBytesPerChar,
4984 next - enc->minBytesPerChar);
4985 if (!name) {
4986 result = XML_ERROR_NO_MEMORY;
4987 goto endEntityValue;
4988 }
4989 entity = (ENTITY *)lookup(&dtd->paramEntities, name, 0);
4990 poolDiscard(&tempPool);
4991 if (!entity) {
4992 /* not a well-formedness error - see XML 1.0: WFC Entity Declared */
4993 /* cannot report skipped entity here - see comments on
4994 skippedEntityHandler
4995 if (skippedEntityHandler)
4996 skippedEntityHandler(handlerArg, name, 0);
4997 */
4998 dtd->keepProcessing = dtd->standalone;
4999 goto endEntityValue;
5000 }
5001 if (entity->open) {
5002 if (enc == encoding)
5003 eventPtr = entityTextPtr;
5004 result = XML_ERROR_RECURSIVE_ENTITY_REF;
5005 goto endEntityValue;
5006 }
5007 if (entity->systemId) {
5008 if (externalEntityRefHandler) {
5009 dtd->paramEntityRead = XML_FALSE;
5010 entity->open = XML_TRUE;
5011 if (!externalEntityRefHandler(externalEntityRefHandlerArg,
5012 0,
5013 entity->base,
5014 entity->systemId,
5015 entity->publicId)) {
5016 entity->open = XML_FALSE;
5017 result = XML_ERROR_EXTERNAL_ENTITY_HANDLING;
5018 goto endEntityValue;
5019 }
5020 entity->open = XML_FALSE;
5021 if (!dtd->paramEntityRead)
5022 dtd->keepProcessing = dtd->standalone;
5023 }
5024 else
5025 dtd->keepProcessing = dtd->standalone;
5026 }
5027 else {
5028 entity->open = XML_TRUE;
5029 result = storeEntityValue(parser,
5030 internalEncoding,
5031 (char *)entity->textPtr,
5032 (char *)(entity->textPtr
5033 + entity->textLen));
5034 entity->open = XML_FALSE;
5035 if (result)
5036 goto endEntityValue;
5037 }
5038 break;
5039 }
5040 #endif /* XML_DTD */
5041 /* In the internal subset, PE references are not legal
5042 within markup declarations, e.g entity values in this case. */
5043 eventPtr = entityTextPtr;
5044 result = XML_ERROR_PARAM_ENTITY_REF;
5045 goto endEntityValue;
5046 case XML_TOK_NONE:
5047 result = XML_ERROR_NONE;
5048 goto endEntityValue;
5049 case XML_TOK_ENTITY_REF:
5050 case XML_TOK_DATA_CHARS:
5051 if (!poolAppend(pool, enc, entityTextPtr, next)) {
5052 result = XML_ERROR_NO_MEMORY;
5053 goto endEntityValue;
5054 }
5055 break;
5056 case XML_TOK_TRAILING_CR:
5057 next = entityTextPtr + enc->minBytesPerChar;
5058 /* fall through */
5059 case XML_TOK_DATA_NEWLINE:
5060 if (pool->end == pool->ptr && !poolGrow(pool)) {
5061 result = XML_ERROR_NO_MEMORY;
5062 goto endEntityValue;
5063 }
5064 *(pool->ptr)++ = 0xA;
5065 break;
5066 case XML_TOK_CHAR_REF:
5067 {
5068 XML_Char buf[XML_ENCODE_MAX];
5069 int i;
5070 int n = XmlCharRefNumber(enc, entityTextPtr);
5071 if (n < 0) {
5072 if (enc == encoding)
5073 eventPtr = entityTextPtr;
5074 result = XML_ERROR_BAD_CHAR_REF;
5075 goto endEntityValue;
5076 }
5077 n = XmlEncode(n, (ICHAR *)buf);
5078 if (!n) {
5079 if (enc == encoding)
5080 eventPtr = entityTextPtr;
5081 result = XML_ERROR_BAD_CHAR_REF;
5082 goto endEntityValue;
5083 }
5084 for (i = 0; i < n; i++) {
5085 if (pool->end == pool->ptr && !poolGrow(pool)) {
5086 result = XML_ERROR_NO_MEMORY;
5087 goto endEntityValue;
5088 }
5089 *(pool->ptr)++ = buf[i];
5090 }
5091 }
5092 break;
5093 case XML_TOK_PARTIAL:
5094 if (enc == encoding)
5095 eventPtr = entityTextPtr;
5096 result = XML_ERROR_INVALID_TOKEN;
5097 goto endEntityValue;
5098 case XML_TOK_INVALID:
5099 if (enc == encoding)
5100 eventPtr = next;
5101 result = XML_ERROR_INVALID_TOKEN;
5102 goto endEntityValue;
5103 default:
5104 if (enc == encoding)
5105 eventPtr = entityTextPtr;
5106 result = XML_ERROR_UNEXPECTED_STATE;
5107 goto endEntityValue;
5108 }
5109 entityTextPtr = next;
5110 }
5111 endEntityValue:
5112 #ifdef XML_DTD
5113 prologState.inEntityValue = oldInEntityValue;
5114 #endif /* XML_DTD */
5115 return result;
5116 }
5117
5118 static void FASTCALL
normalizeLines(XML_Char * s)5119 normalizeLines(XML_Char *s)
5120 {
5121 XML_Char *p;
5122 for (;; s++) {
5123 if (*s == XML_T('\0'))
5124 return;
5125 if (*s == 0xD)
5126 break;
5127 }
5128 p = s;
5129 do {
5130 if (*s == 0xD) {
5131 *p++ = 0xA;
5132 if (*++s == 0xA)
5133 s++;
5134 }
5135 else
5136 *p++ = *s++;
5137 } while (*s);
5138 *p = XML_T('\0');
5139 }
5140
5141 static int
reportProcessingInstruction(XML_Parser parser,const ENCODING * enc,const char * start,const char * end)5142 reportProcessingInstruction(XML_Parser parser, const ENCODING *enc,
5143 const char *start, const char *end)
5144 {
5145 const XML_Char *target;
5146 XML_Char *data;
5147 const char *tem;
5148 if (!processingInstructionHandler) {
5149 if (defaultHandler)
5150 reportDefault(parser, enc, start, end);
5151 return 1;
5152 }
5153 start += enc->minBytesPerChar * 2;
5154 tem = start + XmlNameLength(enc, start);
5155 target = poolStoreString(&tempPool, enc, start, tem);
5156 if (!target)
5157 return 0;
5158 poolFinish(&tempPool);
5159 data = poolStoreString(&tempPool, enc,
5160 XmlSkipS(enc, tem),
5161 end - enc->minBytesPerChar*2);
5162 if (!data)
5163 return 0;
5164 normalizeLines(data);
5165 processingInstructionHandler(handlerArg, target, data);
5166 poolClear(&tempPool);
5167 return 1;
5168 }
5169
5170 static int
reportComment(XML_Parser parser,const ENCODING * enc,const char * start,const char * end)5171 reportComment(XML_Parser parser, const ENCODING *enc,
5172 const char *start, const char *end)
5173 {
5174 XML_Char *data;
5175 if (!commentHandler) {
5176 if (defaultHandler)
5177 reportDefault(parser, enc, start, end);
5178 return 1;
5179 }
5180 data = poolStoreString(&tempPool,
5181 enc,
5182 start + enc->minBytesPerChar * 4,
5183 end - enc->minBytesPerChar * 3);
5184 if (!data)
5185 return 0;
5186 normalizeLines(data);
5187 commentHandler(handlerArg, data);
5188 poolClear(&tempPool);
5189 return 1;
5190 }
5191
5192 static void
reportDefault(XML_Parser parser,const ENCODING * enc,const char * s,const char * end)5193 reportDefault(XML_Parser parser, const ENCODING *enc,
5194 const char *s, const char *end)
5195 {
5196 if (MUST_CONVERT(enc, s)) {
5197 const char **eventPP;
5198 const char **eventEndPP;
5199 if (enc == encoding) {
5200 eventPP = &eventPtr;
5201 eventEndPP = &eventEndPtr;
5202 }
5203 else {
5204 eventPP = &(openInternalEntities->internalEventPtr);
5205 eventEndPP = &(openInternalEntities->internalEventEndPtr);
5206 }
5207 do {
5208 ICHAR *dataPtr = (ICHAR *)dataBuf;
5209 XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
5210 *eventEndPP = s;
5211 defaultHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf));
5212 *eventPP = s;
5213 } while (s != end);
5214 }
5215 else
5216 defaultHandler(handlerArg, (XML_Char *)s, (int)((XML_Char *)end - (XML_Char *)s));
5217 }
5218
5219
5220 static int
defineAttribute(ELEMENT_TYPE * type,ATTRIBUTE_ID * attId,XML_Bool isCdata,XML_Bool isId,const XML_Char * value,XML_Parser parser)5221 defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *attId, XML_Bool isCdata,
5222 XML_Bool isId, const XML_Char *value, XML_Parser parser)
5223 {
5224 DEFAULT_ATTRIBUTE *att;
5225 if (value || isId) {
5226 /* The handling of default attributes gets messed up if we have
5227 a default which duplicates a non-default. */
5228 int i;
5229 for (i = 0; i < type->nDefaultAtts; i++)
5230 if (attId == type->defaultAtts[i].id)
5231 return 1;
5232 if (isId && !type->idAtt && !attId->xmlns)
5233 type->idAtt = attId;
5234 }
5235 if (type->nDefaultAtts == type->allocDefaultAtts) {
5236 if (type->allocDefaultAtts == 0) {
5237 type->allocDefaultAtts = 8;
5238 type->defaultAtts = (DEFAULT_ATTRIBUTE *)MALLOC(type->allocDefaultAtts
5239 * sizeof(DEFAULT_ATTRIBUTE));
5240 if (!type->defaultAtts)
5241 return 0;
5242 }
5243 else {
5244 DEFAULT_ATTRIBUTE *temp;
5245 int count = type->allocDefaultAtts * 2;
5246 temp = (DEFAULT_ATTRIBUTE *)
5247 REALLOC(type->defaultAtts, (count * sizeof(DEFAULT_ATTRIBUTE)));
5248 if (temp == NULL)
5249 return 0;
5250 type->allocDefaultAtts = count;
5251 type->defaultAtts = temp;
5252 }
5253 }
5254 att = type->defaultAtts + type->nDefaultAtts;
5255 att->id = attId;
5256 att->value = value;
5257 att->isCdata = isCdata;
5258 if (!isCdata)
5259 attId->maybeTokenized = XML_TRUE;
5260 type->nDefaultAtts += 1;
5261 return 1;
5262 }
5263
5264 static int
setElementTypePrefix(XML_Parser parser,ELEMENT_TYPE * elementType)5265 setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *elementType)
5266 {
5267 DTD * const dtd = _dtd; /* save one level of indirection */
5268 const XML_Char *name;
5269 for (name = elementType->name; *name; name++) {
5270 if (*name == XML_T(':')) {
5271 PREFIX *prefix;
5272 const XML_Char *s;
5273 for (s = elementType->name; s != name; s++) {
5274 if (!poolAppendChar(&dtd->pool, *s))
5275 return 0;
5276 }
5277 if (!poolAppendChar(&dtd->pool, XML_T('\0')))
5278 return 0;
5279 prefix = (PREFIX *)lookup(&dtd->prefixes, poolStart(&dtd->pool),
5280 sizeof(PREFIX));
5281 if (!prefix)
5282 return 0;
5283 if (prefix->name == poolStart(&dtd->pool))
5284 poolFinish(&dtd->pool);
5285 else
5286 poolDiscard(&dtd->pool);
5287 elementType->prefix = prefix;
5288
5289 }
5290 }
5291 return 1;
5292 }
5293
5294 static ATTRIBUTE_ID *
getAttributeId(XML_Parser parser,const ENCODING * enc,const char * start,const char * end)5295 getAttributeId(XML_Parser parser, const ENCODING *enc,
5296 const char *start, const char *end)
5297 {
5298 DTD * const dtd = _dtd; /* save one level of indirection */
5299 ATTRIBUTE_ID *id;
5300 const XML_Char *name;
5301 if (!poolAppendChar(&dtd->pool, XML_T('\0')))
5302 return NULL;
5303 name = poolStoreString(&dtd->pool, enc, start, end);
5304 if (!name)
5305 return NULL;
5306 /* skip quotation mark - its storage will be re-used (like in name[-1]) */
5307 ++name;
5308 id = (ATTRIBUTE_ID *)lookup(&dtd->attributeIds, name, sizeof(ATTRIBUTE_ID));
5309 if (!id)
5310 return NULL;
5311 if (id->name != name)
5312 poolDiscard(&dtd->pool);
5313 else {
5314 poolFinish(&dtd->pool);
5315 if (!ns)
5316 ;
5317 else if (name[0] == XML_T('x')
5318 && name[1] == XML_T('m')
5319 && name[2] == XML_T('l')
5320 && name[3] == XML_T('n')
5321 && name[4] == XML_T('s')
5322 && (name[5] == XML_T('\0') || name[5] == XML_T(':'))) {
5323 if (name[5] == XML_T('\0'))
5324 id->prefix = &dtd->defaultPrefix;
5325 else
5326 id->prefix = (PREFIX *)lookup(&dtd->prefixes, name + 6, sizeof(PREFIX));
5327 id->xmlns = XML_TRUE;
5328 }
5329 else {
5330 int i;
5331 for (i = 0; name[i]; i++) {
5332 /* attributes without prefix are *not* in the default namespace */
5333 if (name[i] == XML_T(':')) {
5334 int j;
5335 for (j = 0; j < i; j++) {
5336 if (!poolAppendChar(&dtd->pool, name[j]))
5337 return NULL;
5338 }
5339 if (!poolAppendChar(&dtd->pool, XML_T('\0')))
5340 return NULL;
5341 id->prefix = (PREFIX *)lookup(&dtd->prefixes, poolStart(&dtd->pool),
5342 sizeof(PREFIX));
5343 if (id->prefix->name == poolStart(&dtd->pool))
5344 poolFinish(&dtd->pool);
5345 else
5346 poolDiscard(&dtd->pool);
5347 break;
5348 }
5349 }
5350 }
5351 }
5352 return id;
5353 }
5354
5355 #define CONTEXT_SEP XML_T('\f')
5356
5357 static const XML_Char *
getContext(XML_Parser parser)5358 getContext(XML_Parser parser)
5359 {
5360 DTD * const dtd = _dtd; /* save one level of indirection */
5361 HASH_TABLE_ITER iter;
5362 XML_Bool needSep = XML_FALSE;
5363
5364 if (dtd->defaultPrefix.binding) {
5365 int i;
5366 int len;
5367 if (!poolAppendChar(&tempPool, XML_T('=')))
5368 return NULL;
5369 len = dtd->defaultPrefix.binding->uriLen;
5370 if (namespaceSeparator)
5371 len--;
5372 for (i = 0; i < len; i++)
5373 if (!poolAppendChar(&tempPool, dtd->defaultPrefix.binding->uri[i]))
5374 return NULL;
5375 needSep = XML_TRUE;
5376 }
5377
5378 hashTableIterInit(&iter, &(dtd->prefixes));
5379 for (;;) {
5380 int i;
5381 int len;
5382 const XML_Char *s;
5383 PREFIX *prefix = (PREFIX *)hashTableIterNext(&iter);
5384 if (!prefix)
5385 break;
5386 if (!prefix->binding)
5387 continue;
5388 if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP))
5389 return NULL;
5390 for (s = prefix->name; *s; s++)
5391 if (!poolAppendChar(&tempPool, *s))
5392 return NULL;
5393 if (!poolAppendChar(&tempPool, XML_T('=')))
5394 return NULL;
5395 len = prefix->binding->uriLen;
5396 if (namespaceSeparator)
5397 len--;
5398 for (i = 0; i < len; i++)
5399 if (!poolAppendChar(&tempPool, prefix->binding->uri[i]))
5400 return NULL;
5401 needSep = XML_TRUE;
5402 }
5403
5404
5405 hashTableIterInit(&iter, &(dtd->generalEntities));
5406 for (;;) {
5407 const XML_Char *s;
5408 ENTITY *e = (ENTITY *)hashTableIterNext(&iter);
5409 if (!e)
5410 break;
5411 if (!e->open)
5412 continue;
5413 if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP))
5414 return NULL;
5415 for (s = e->name; *s; s++)
5416 if (!poolAppendChar(&tempPool, *s))
5417 return 0;
5418 needSep = XML_TRUE;
5419 }
5420
5421 if (!poolAppendChar(&tempPool, XML_T('\0')))
5422 return NULL;
5423 return tempPool.start;
5424 }
5425
5426 static XML_Bool
setContext(XML_Parser parser,const XML_Char * context)5427 setContext(XML_Parser parser, const XML_Char *context)
5428 {
5429 DTD * const dtd = _dtd; /* save one level of indirection */
5430 const XML_Char *s = context;
5431
5432 while (*context != XML_T('\0')) {
5433 if (*s == CONTEXT_SEP || *s == XML_T('\0')) {
5434 ENTITY *e;
5435 if (!poolAppendChar(&tempPool, XML_T('\0')))
5436 return XML_FALSE;
5437 e = (ENTITY *)lookup(&dtd->generalEntities, poolStart(&tempPool), 0);
5438 if (e)
5439 e->open = XML_TRUE;
5440 if (*s != XML_T('\0'))
5441 s++;
5442 context = s;
5443 poolDiscard(&tempPool);
5444 }
5445 else if (*s == XML_T('=')) {
5446 PREFIX *prefix;
5447 if (poolLength(&tempPool) == 0)
5448 prefix = &dtd->defaultPrefix;
5449 else {
5450 if (!poolAppendChar(&tempPool, XML_T('\0')))
5451 return XML_FALSE;
5452 prefix = (PREFIX *)lookup(&dtd->prefixes, poolStart(&tempPool),
5453 sizeof(PREFIX));
5454 if (!prefix)
5455 return XML_FALSE;
5456 if (prefix->name == poolStart(&tempPool)) {
5457 prefix->name = poolCopyString(&dtd->pool, prefix->name);
5458 if (!prefix->name)
5459 return XML_FALSE;
5460 }
5461 poolDiscard(&tempPool);
5462 }
5463 for (context = s + 1;
5464 *context != CONTEXT_SEP && *context != XML_T('\0');
5465 context++)
5466 if (!poolAppendChar(&tempPool, *context))
5467 return XML_FALSE;
5468 if (!poolAppendChar(&tempPool, XML_T('\0')))
5469 return XML_FALSE;
5470 if (addBinding(parser, prefix, NULL, poolStart(&tempPool),
5471 &inheritedBindings) != XML_ERROR_NONE)
5472 return XML_FALSE;
5473 poolDiscard(&tempPool);
5474 if (*context != XML_T('\0'))
5475 ++context;
5476 s = context;
5477 }
5478 else {
5479 if (!poolAppendChar(&tempPool, *s))
5480 return XML_FALSE;
5481 s++;
5482 }
5483 }
5484 return XML_TRUE;
5485 }
5486
5487 static void FASTCALL
normalizePublicId(XML_Char * publicId)5488 normalizePublicId(XML_Char *publicId)
5489 {
5490 XML_Char *p = publicId;
5491 XML_Char *s;
5492 for (s = publicId; *s; s++) {
5493 switch (*s) {
5494 case 0x20:
5495 case 0xD:
5496 case 0xA:
5497 if (p != publicId && p[-1] != 0x20)
5498 *p++ = 0x20;
5499 break;
5500 default:
5501 *p++ = *s;
5502 }
5503 }
5504 if (p != publicId && p[-1] == 0x20)
5505 --p;
5506 *p = XML_T('\0');
5507 }
5508
5509 static DTD *
dtdCreate(const XML_Memory_Handling_Suite * ms)5510 dtdCreate(const XML_Memory_Handling_Suite *ms)
5511 {
5512 DTD *p = (DTD *)ms->malloc_fcn(sizeof(DTD));
5513 if (p == NULL)
5514 return p;
5515 poolInit(&(p->pool), ms);
5516 poolInit(&(p->entityValuePool), ms);
5517 hashTableInit(&(p->generalEntities), ms);
5518 hashTableInit(&(p->elementTypes), ms);
5519 hashTableInit(&(p->attributeIds), ms);
5520 hashTableInit(&(p->prefixes), ms);
5521 #ifdef XML_DTD
5522 p->paramEntityRead = XML_FALSE;
5523 hashTableInit(&(p->paramEntities), ms);
5524 #endif /* XML_DTD */
5525 p->defaultPrefix.name = NULL;
5526 p->defaultPrefix.binding = NULL;
5527
5528 p->in_eldecl = XML_FALSE;
5529 p->scaffIndex = NULL;
5530 p->scaffold = NULL;
5531 p->scaffLevel = 0;
5532 p->scaffSize = 0;
5533 p->scaffCount = 0;
5534 p->contentStringLen = 0;
5535
5536 p->keepProcessing = XML_TRUE;
5537 p->hasParamEntityRefs = XML_FALSE;
5538 p->standalone = XML_FALSE;
5539 return p;
5540 }
5541
5542 static void
dtdReset(DTD * p,const XML_Memory_Handling_Suite * ms)5543 dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms)
5544 {
5545 HASH_TABLE_ITER iter;
5546 hashTableIterInit(&iter, &(p->elementTypes));
5547 for (;;) {
5548 ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter);
5549 if (!e)
5550 break;
5551 if (e->allocDefaultAtts != 0)
5552 ms->free_fcn(e->defaultAtts);
5553 }
5554 hashTableClear(&(p->generalEntities));
5555 #ifdef XML_DTD
5556 p->paramEntityRead = XML_FALSE;
5557 hashTableClear(&(p->paramEntities));
5558 #endif /* XML_DTD */
5559 hashTableClear(&(p->elementTypes));
5560 hashTableClear(&(p->attributeIds));
5561 hashTableClear(&(p->prefixes));
5562 poolClear(&(p->pool));
5563 poolClear(&(p->entityValuePool));
5564 p->defaultPrefix.name = NULL;
5565 p->defaultPrefix.binding = NULL;
5566
5567 p->in_eldecl = XML_FALSE;
5568
5569 ms->free_fcn(p->scaffIndex);
5570 p->scaffIndex = NULL;
5571 ms->free_fcn(p->scaffold);
5572 p->scaffold = NULL;
5573
5574 p->scaffLevel = 0;
5575 p->scaffSize = 0;
5576 p->scaffCount = 0;
5577 p->contentStringLen = 0;
5578
5579 p->keepProcessing = XML_TRUE;
5580 p->hasParamEntityRefs = XML_FALSE;
5581 p->standalone = XML_FALSE;
5582 }
5583
5584 static void
dtdDestroy(DTD * p,XML_Bool isDocEntity,const XML_Memory_Handling_Suite * ms)5585 dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms)
5586 {
5587 HASH_TABLE_ITER iter;
5588 hashTableIterInit(&iter, &(p->elementTypes));
5589 for (;;) {
5590 ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter);
5591 if (!e)
5592 break;
5593 if (e->allocDefaultAtts != 0)
5594 ms->free_fcn(e->defaultAtts);
5595 }
5596 hashTableDestroy(&(p->generalEntities));
5597 #ifdef XML_DTD
5598 hashTableDestroy(&(p->paramEntities));
5599 #endif /* XML_DTD */
5600 hashTableDestroy(&(p->elementTypes));
5601 hashTableDestroy(&(p->attributeIds));
5602 hashTableDestroy(&(p->prefixes));
5603 poolDestroy(&(p->pool));
5604 poolDestroy(&(p->entityValuePool));
5605 if (isDocEntity) {
5606 ms->free_fcn(p->scaffIndex);
5607 ms->free_fcn(p->scaffold);
5608 }
5609 ms->free_fcn(p);
5610 }
5611
5612 /* Do a deep copy of the DTD. Return 0 for out of memory, non-zero otherwise.
5613 The new DTD has already been initialized.
5614 */
5615 static int
dtdCopy(DTD * newDtd,const DTD * oldDtd,const XML_Memory_Handling_Suite * ms)5616 dtdCopy(DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms)
5617 {
5618 HASH_TABLE_ITER iter;
5619
5620 /* Copy the prefix table. */
5621
5622 hashTableIterInit(&iter, &(oldDtd->prefixes));
5623 for (;;) {
5624 const XML_Char *name;
5625 const PREFIX *oldP = (PREFIX *)hashTableIterNext(&iter);
5626 if (!oldP)
5627 break;
5628 name = poolCopyString(&(newDtd->pool), oldP->name);
5629 if (!name)
5630 return 0;
5631 if (!lookup(&(newDtd->prefixes), name, sizeof(PREFIX)))
5632 return 0;
5633 }
5634
5635 hashTableIterInit(&iter, &(oldDtd->attributeIds));
5636
5637 /* Copy the attribute id table. */
5638
5639 for (;;) {
5640 ATTRIBUTE_ID *newA;
5641 const XML_Char *name;
5642 const ATTRIBUTE_ID *oldA = (ATTRIBUTE_ID *)hashTableIterNext(&iter);
5643
5644 if (!oldA)
5645 break;
5646 /* Remember to allocate the scratch byte before the name. */
5647 if (!poolAppendChar(&(newDtd->pool), XML_T('\0')))
5648 return 0;
5649 name = poolCopyString(&(newDtd->pool), oldA->name);
5650 if (!name)
5651 return 0;
5652 ++name;
5653 newA = (ATTRIBUTE_ID *)lookup(&(newDtd->attributeIds), name,
5654 sizeof(ATTRIBUTE_ID));
5655 if (!newA)
5656 return 0;
5657 newA->maybeTokenized = oldA->maybeTokenized;
5658 if (oldA->prefix) {
5659 newA->xmlns = oldA->xmlns;
5660 if (oldA->prefix == &oldDtd->defaultPrefix)
5661 newA->prefix = &newDtd->defaultPrefix;
5662 else
5663 newA->prefix = (PREFIX *)lookup(&(newDtd->prefixes),
5664 oldA->prefix->name, 0);
5665 }
5666 }
5667
5668 /* Copy the element type table. */
5669
5670 hashTableIterInit(&iter, &(oldDtd->elementTypes));
5671
5672 for (;;) {
5673 int i;
5674 ELEMENT_TYPE *newE;
5675 const XML_Char *name;
5676 const ELEMENT_TYPE *oldE = (ELEMENT_TYPE *)hashTableIterNext(&iter);
5677 if (!oldE)
5678 break;
5679 name = poolCopyString(&(newDtd->pool), oldE->name);
5680 if (!name)
5681 return 0;
5682 newE = (ELEMENT_TYPE *)lookup(&(newDtd->elementTypes), name,
5683 sizeof(ELEMENT_TYPE));
5684 if (!newE)
5685 return 0;
5686 if (oldE->nDefaultAtts) {
5687 newE->defaultAtts = (DEFAULT_ATTRIBUTE *)
5688 ms->malloc_fcn(oldE->nDefaultAtts * sizeof(DEFAULT_ATTRIBUTE));
5689 if (!newE->defaultAtts) {
5690 ms->free_fcn(newE);
5691 return 0;
5692 }
5693 }
5694 if (oldE->idAtt)
5695 newE->idAtt = (ATTRIBUTE_ID *)
5696 lookup(&(newDtd->attributeIds), oldE->idAtt->name, 0);
5697 newE->allocDefaultAtts = newE->nDefaultAtts = oldE->nDefaultAtts;
5698 if (oldE->prefix)
5699 newE->prefix = (PREFIX *)lookup(&(newDtd->prefixes),
5700 oldE->prefix->name, 0);
5701 for (i = 0; i < newE->nDefaultAtts; i++) {
5702 newE->defaultAtts[i].id = (ATTRIBUTE_ID *)
5703 lookup(&(newDtd->attributeIds), oldE->defaultAtts[i].id->name, 0);
5704 newE->defaultAtts[i].isCdata = oldE->defaultAtts[i].isCdata;
5705 if (oldE->defaultAtts[i].value) {
5706 newE->defaultAtts[i].value
5707 = poolCopyString(&(newDtd->pool), oldE->defaultAtts[i].value);
5708 if (!newE->defaultAtts[i].value)
5709 return 0;
5710 }
5711 else
5712 newE->defaultAtts[i].value = NULL;
5713 }
5714 }
5715
5716 /* Copy the entity tables. */
5717 if (!copyEntityTable(&(newDtd->generalEntities),
5718 &(newDtd->pool),
5719 &(oldDtd->generalEntities)))
5720 return 0;
5721
5722 #ifdef XML_DTD
5723 if (!copyEntityTable(&(newDtd->paramEntities),
5724 &(newDtd->pool),
5725 &(oldDtd->paramEntities)))
5726 return 0;
5727 newDtd->paramEntityRead = oldDtd->paramEntityRead;
5728 #endif /* XML_DTD */
5729
5730 newDtd->keepProcessing = oldDtd->keepProcessing;
5731 newDtd->hasParamEntityRefs = oldDtd->hasParamEntityRefs;
5732 newDtd->standalone = oldDtd->standalone;
5733
5734 /* Don't want deep copying for scaffolding */
5735 newDtd->in_eldecl = oldDtd->in_eldecl;
5736 newDtd->scaffold = oldDtd->scaffold;
5737 newDtd->contentStringLen = oldDtd->contentStringLen;
5738 newDtd->scaffSize = oldDtd->scaffSize;
5739 newDtd->scaffLevel = oldDtd->scaffLevel;
5740 newDtd->scaffIndex = oldDtd->scaffIndex;
5741
5742 return 1;
5743 } /* End dtdCopy */
5744
5745 static int
copyEntityTable(HASH_TABLE * newTable,STRING_POOL * newPool,const HASH_TABLE * oldTable)5746 copyEntityTable(HASH_TABLE *newTable,
5747 STRING_POOL *newPool,
5748 const HASH_TABLE *oldTable)
5749 {
5750 HASH_TABLE_ITER iter;
5751 const XML_Char *cachedOldBase = NULL;
5752 const XML_Char *cachedNewBase = NULL;
5753
5754 hashTableIterInit(&iter, oldTable);
5755
5756 for (;;) {
5757 ENTITY *newE;
5758 const XML_Char *name;
5759 const ENTITY *oldE = (ENTITY *)hashTableIterNext(&iter);
5760 if (!oldE)
5761 break;
5762 name = poolCopyString(newPool, oldE->name);
5763 if (!name)
5764 return 0;
5765 newE = (ENTITY *)lookup(newTable, name, sizeof(ENTITY));
5766 if (!newE)
5767 return 0;
5768 if (oldE->systemId) {
5769 const XML_Char *tem = poolCopyString(newPool, oldE->systemId);
5770 if (!tem)
5771 return 0;
5772 newE->systemId = tem;
5773 if (oldE->base) {
5774 if (oldE->base == cachedOldBase)
5775 newE->base = cachedNewBase;
5776 else {
5777 cachedOldBase = oldE->base;
5778 tem = poolCopyString(newPool, cachedOldBase);
5779 if (!tem)
5780 return 0;
5781 cachedNewBase = newE->base = tem;
5782 }
5783 }
5784 if (oldE->publicId) {
5785 tem = poolCopyString(newPool, oldE->publicId);
5786 if (!tem)
5787 return 0;
5788 newE->publicId = tem;
5789 }
5790 }
5791 else {
5792 const XML_Char *tem = poolCopyStringN(newPool, oldE->textPtr,
5793 oldE->textLen);
5794 if (!tem)
5795 return 0;
5796 newE->textPtr = tem;
5797 newE->textLen = oldE->textLen;
5798 }
5799 if (oldE->notation) {
5800 const XML_Char *tem = poolCopyString(newPool, oldE->notation);
5801 if (!tem)
5802 return 0;
5803 newE->notation = tem;
5804 }
5805 newE->is_param = oldE->is_param;
5806 newE->is_internal = oldE->is_internal;
5807 }
5808 return 1;
5809 }
5810
5811 #define INIT_POWER 6
5812
5813 static XML_Bool FASTCALL
keyeq(KEY s1,KEY s2)5814 keyeq(KEY s1, KEY s2)
5815 {
5816 for (; *s1 == *s2; s1++, s2++)
5817 if (*s1 == 0)
5818 return XML_TRUE;
5819 return XML_FALSE;
5820 }
5821
5822 static unsigned long FASTCALL
hash(KEY s)5823 hash(KEY s)
5824 {
5825 unsigned long h = 0;
5826 while (*s)
5827 h = CHAR_HASH(h, *s++);
5828 return h;
5829 }
5830
5831 static NAMED *
lookup(HASH_TABLE * table,KEY name,size_t createSize)5832 lookup(HASH_TABLE *table, KEY name, size_t createSize)
5833 {
5834 size_t i;
5835 if (table->size == 0) {
5836 size_t tsize;
5837 if (!createSize)
5838 return NULL;
5839 table->power = INIT_POWER;
5840 /* table->size is a power of 2 */
5841 table->size = (size_t)1 << INIT_POWER;
5842 tsize = table->size * sizeof(NAMED *);
5843 table->v = (NAMED **)table->mem->malloc_fcn(tsize);
5844 if (!table->v) {
5845 table->size = 0;
5846 return NULL;
5847 }
5848 memset(table->v, 0, tsize);
5849 i = hash(name) & ((unsigned long)table->size - 1);
5850 }
5851 else {
5852 unsigned long h = hash(name);
5853 unsigned long mask = (unsigned long)table->size - 1;
5854 unsigned char step = 0;
5855 i = h & mask;
5856 while (table->v[i]) {
5857 if (keyeq(name, table->v[i]->name))
5858 return table->v[i];
5859 if (!step)
5860 step = PROBE_STEP(h, mask, table->power);
5861 i < step ? (i += table->size - step) : (i -= step);
5862 }
5863 if (!createSize)
5864 return NULL;
5865
5866 /* check for overflow (table is half full) */
5867 if (table->used >> (table->power - 1)) {
5868 unsigned char newPower = table->power + 1;
5869 size_t newSize = (size_t)1 << newPower;
5870 unsigned long newMask = (unsigned long)newSize - 1;
5871 size_t tsize = newSize * sizeof(NAMED *);
5872 NAMED **newV = (NAMED **)table->mem->malloc_fcn(tsize);
5873 if (!newV)
5874 return NULL;
5875 memset(newV, 0, tsize);
5876 for (i = 0; i < table->size; i++)
5877 if (table->v[i]) {
5878 unsigned long newHash = hash(table->v[i]->name);
5879 size_t j = newHash & newMask;
5880 step = 0;
5881 while (newV[j]) {
5882 if (!step)
5883 step = PROBE_STEP(newHash, newMask, newPower);
5884 j < step ? (j += newSize - step) : (j -= step);
5885 }
5886 newV[j] = table->v[i];
5887 }
5888 table->mem->free_fcn(table->v);
5889 table->v = newV;
5890 table->power = newPower;
5891 table->size = newSize;
5892 i = h & newMask;
5893 step = 0;
5894 while (table->v[i]) {
5895 if (!step)
5896 step = PROBE_STEP(h, newMask, newPower);
5897 i < step ? (i += newSize - step) : (i -= step);
5898 }
5899 }
5900 }
5901 table->v[i] = (NAMED *)table->mem->malloc_fcn(createSize);
5902 if (!table->v[i])
5903 return NULL;
5904 memset(table->v[i], 0, createSize);
5905 table->v[i]->name = name;
5906 (table->used)++;
5907 return table->v[i];
5908 }
5909
5910 static void FASTCALL
hashTableClear(HASH_TABLE * table)5911 hashTableClear(HASH_TABLE *table)
5912 {
5913 size_t i;
5914 for (i = 0; i < table->size; i++) {
5915 table->mem->free_fcn(table->v[i]);
5916 table->v[i] = NULL;
5917 }
5918 table->used = 0;
5919 }
5920
5921 static void FASTCALL
hashTableDestroy(HASH_TABLE * table)5922 hashTableDestroy(HASH_TABLE *table)
5923 {
5924 size_t i;
5925 for (i = 0; i < table->size; i++)
5926 table->mem->free_fcn(table->v[i]);
5927 table->mem->free_fcn(table->v);
5928 }
5929
5930 static void FASTCALL
hashTableInit(HASH_TABLE * p,const XML_Memory_Handling_Suite * ms)5931 hashTableInit(HASH_TABLE *p, const XML_Memory_Handling_Suite *ms)
5932 {
5933 p->power = 0;
5934 p->size = 0;
5935 p->used = 0;
5936 p->v = NULL;
5937 p->mem = ms;
5938 }
5939
5940 static void FASTCALL
hashTableIterInit(HASH_TABLE_ITER * iter,const HASH_TABLE * table)5941 hashTableIterInit(HASH_TABLE_ITER *iter, const HASH_TABLE *table)
5942 {
5943 iter->p = table->v;
5944 iter->end = iter->p + table->size;
5945 }
5946
5947 static NAMED * FASTCALL
hashTableIterNext(HASH_TABLE_ITER * iter)5948 hashTableIterNext(HASH_TABLE_ITER *iter)
5949 {
5950 while (iter->p != iter->end) {
5951 NAMED *tem = *(iter->p)++;
5952 if (tem)
5953 return tem;
5954 }
5955 return NULL;
5956 }
5957
5958 static void FASTCALL
poolInit(STRING_POOL * pool,const XML_Memory_Handling_Suite * ms)5959 poolInit(STRING_POOL *pool, const XML_Memory_Handling_Suite *ms)
5960 {
5961 pool->blocks = NULL;
5962 pool->freeBlocks = NULL;
5963 pool->start = NULL;
5964 pool->ptr = NULL;
5965 pool->end = NULL;
5966 pool->mem = ms;
5967 }
5968
5969 static void FASTCALL
poolClear(STRING_POOL * pool)5970 poolClear(STRING_POOL *pool)
5971 {
5972 if (!pool->freeBlocks)
5973 pool->freeBlocks = pool->blocks;
5974 else {
5975 BLOCK *p = pool->blocks;
5976 while (p) {
5977 BLOCK *tem = p->next;
5978 p->next = pool->freeBlocks;
5979 pool->freeBlocks = p;
5980 p = tem;
5981 }
5982 }
5983 pool->blocks = NULL;
5984 pool->start = NULL;
5985 pool->ptr = NULL;
5986 pool->end = NULL;
5987 }
5988
5989 static void FASTCALL
poolDestroy(STRING_POOL * pool)5990 poolDestroy(STRING_POOL *pool)
5991 {
5992 BLOCK *p = pool->blocks;
5993 while (p) {
5994 BLOCK *tem = p->next;
5995 pool->mem->free_fcn(p);
5996 p = tem;
5997 }
5998 p = pool->freeBlocks;
5999 while (p) {
6000 BLOCK *tem = p->next;
6001 pool->mem->free_fcn(p);
6002 p = tem;
6003 }
6004 }
6005
6006 static XML_Char *
poolAppend(STRING_POOL * pool,const ENCODING * enc,const char * ptr,const char * end)6007 poolAppend(STRING_POOL *pool, const ENCODING *enc,
6008 const char *ptr, const char *end)
6009 {
6010 if (!pool->ptr && !poolGrow(pool))
6011 return NULL;
6012 for (;;) {
6013 XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end);
6014 if (ptr == end)
6015 break;
6016 if (!poolGrow(pool))
6017 return NULL;
6018 }
6019 return pool->start;
6020 }
6021
6022 static const XML_Char * FASTCALL
poolCopyString(STRING_POOL * pool,const XML_Char * s)6023 poolCopyString(STRING_POOL *pool, const XML_Char *s)
6024 {
6025 do {
6026 if (!poolAppendChar(pool, *s))
6027 return NULL;
6028 } while (*s++);
6029 s = pool->start;
6030 poolFinish(pool);
6031 return s;
6032 }
6033
6034 static const XML_Char *
poolCopyStringN(STRING_POOL * pool,const XML_Char * s,int n)6035 poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n)
6036 {
6037 if (!pool->ptr && !poolGrow(pool))
6038 return NULL;
6039 for (; n > 0; --n, s++) {
6040 if (!poolAppendChar(pool, *s))
6041 return NULL;
6042 }
6043 s = pool->start;
6044 poolFinish(pool);
6045 return s;
6046 }
6047
6048 static const XML_Char * FASTCALL
poolAppendString(STRING_POOL * pool,const XML_Char * s)6049 poolAppendString(STRING_POOL *pool, const XML_Char *s)
6050 {
6051 while (*s) {
6052 if (!poolAppendChar(pool, *s))
6053 return NULL;
6054 s++;
6055 }
6056 return pool->start;
6057 }
6058
6059 static XML_Char *
poolStoreString(STRING_POOL * pool,const ENCODING * enc,const char * ptr,const char * end)6060 poolStoreString(STRING_POOL *pool, const ENCODING *enc,
6061 const char *ptr, const char *end)
6062 {
6063 if (!poolAppend(pool, enc, ptr, end))
6064 return NULL;
6065 if (pool->ptr == pool->end && !poolGrow(pool))
6066 return NULL;
6067 *(pool->ptr)++ = 0;
6068 return pool->start;
6069 }
6070
6071 static XML_Bool FASTCALL
poolGrow(STRING_POOL * pool)6072 poolGrow(STRING_POOL *pool)
6073 {
6074 if (pool->freeBlocks) {
6075 if (pool->start == 0) {
6076 pool->blocks = pool->freeBlocks;
6077 pool->freeBlocks = pool->freeBlocks->next;
6078 pool->blocks->next = NULL;
6079 pool->start = pool->blocks->s;
6080 pool->end = pool->start + pool->blocks->size;
6081 pool->ptr = pool->start;
6082 return XML_TRUE;
6083 }
6084 if (pool->end - pool->start < pool->freeBlocks->size) {
6085 BLOCK *tem = pool->freeBlocks->next;
6086 pool->freeBlocks->next = pool->blocks;
6087 pool->blocks = pool->freeBlocks;
6088 pool->freeBlocks = tem;
6089 memcpy(pool->blocks->s, pool->start,
6090 (pool->end - pool->start) * sizeof(XML_Char));
6091 pool->ptr = pool->blocks->s + (pool->ptr - pool->start);
6092 pool->start = pool->blocks->s;
6093 pool->end = pool->start + pool->blocks->size;
6094 return XML_TRUE;
6095 }
6096 }
6097 if (pool->blocks && pool->start == pool->blocks->s) {
6098 int blockSize = (int)(pool->end - pool->start)*2;
6099 pool->blocks = (BLOCK *)
6100 pool->mem->realloc_fcn(pool->blocks,
6101 (offsetof(BLOCK, s)
6102 + blockSize * sizeof(XML_Char)));
6103 if (pool->blocks == NULL)
6104 return XML_FALSE;
6105 pool->blocks->size = blockSize;
6106 pool->ptr = pool->blocks->s + (pool->ptr - pool->start);
6107 pool->start = pool->blocks->s;
6108 pool->end = pool->start + blockSize;
6109 }
6110 else {
6111 BLOCK *tem;
6112 int blockSize = (int)(pool->end - pool->start);
6113 if (blockSize < INIT_BLOCK_SIZE)
6114 blockSize = INIT_BLOCK_SIZE;
6115 else
6116 blockSize *= 2;
6117 tem = (BLOCK *)pool->mem->malloc_fcn(offsetof(BLOCK, s)
6118 + blockSize * sizeof(XML_Char));
6119 if (!tem)
6120 return XML_FALSE;
6121 tem->size = blockSize;
6122 tem->next = pool->blocks;
6123 pool->blocks = tem;
6124 if (pool->ptr != pool->start)
6125 memcpy(tem->s, pool->start,
6126 (pool->ptr - pool->start) * sizeof(XML_Char));
6127 pool->ptr = tem->s + (pool->ptr - pool->start);
6128 pool->start = tem->s;
6129 pool->end = tem->s + blockSize;
6130 }
6131 return XML_TRUE;
6132 }
6133
6134 static int FASTCALL
nextScaffoldPart(XML_Parser parser)6135 nextScaffoldPart(XML_Parser parser)
6136 {
6137 DTD * const dtd = _dtd; /* save one level of indirection */
6138 CONTENT_SCAFFOLD * me;
6139 int next;
6140
6141 if (!dtd->scaffIndex) {
6142 dtd->scaffIndex = (int *)MALLOC(groupSize * sizeof(int));
6143 if (!dtd->scaffIndex)
6144 return -1;
6145 dtd->scaffIndex[0] = 0;
6146 }
6147
6148 if (dtd->scaffCount >= dtd->scaffSize) {
6149 CONTENT_SCAFFOLD *temp;
6150 if (dtd->scaffold) {
6151 temp = (CONTENT_SCAFFOLD *)
6152 REALLOC(dtd->scaffold, dtd->scaffSize * 2 * sizeof(CONTENT_SCAFFOLD));
6153 if (temp == NULL)
6154 return -1;
6155 dtd->scaffSize *= 2;
6156 }
6157 else {
6158 temp = (CONTENT_SCAFFOLD *)MALLOC(INIT_SCAFFOLD_ELEMENTS
6159 * sizeof(CONTENT_SCAFFOLD));
6160 if (temp == NULL)
6161 return -1;
6162 dtd->scaffSize = INIT_SCAFFOLD_ELEMENTS;
6163 }
6164 dtd->scaffold = temp;
6165 }
6166 next = dtd->scaffCount++;
6167 me = &dtd->scaffold[next];
6168 if (dtd->scaffLevel) {
6169 CONTENT_SCAFFOLD *parent = &dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel-1]];
6170 if (parent->lastchild) {
6171 dtd->scaffold[parent->lastchild].nextsib = next;
6172 }
6173 if (!parent->childcnt)
6174 parent->firstchild = next;
6175 parent->lastchild = next;
6176 parent->childcnt++;
6177 }
6178 me->firstchild = me->lastchild = me->childcnt = me->nextsib = 0;
6179 return next;
6180 }
6181
6182 static void
build_node(XML_Parser parser,int src_node,XML_Content * dest,XML_Content ** contpos,XML_Char ** strpos)6183 build_node(XML_Parser parser,
6184 int src_node,
6185 XML_Content *dest,
6186 XML_Content **contpos,
6187 XML_Char **strpos)
6188 {
6189 DTD * const dtd = _dtd; /* save one level of indirection */
6190 dest->type = dtd->scaffold[src_node].type;
6191 dest->quant = dtd->scaffold[src_node].quant;
6192 if (dest->type == XML_CTYPE_NAME) {
6193 const XML_Char *src;
6194 dest->name = *strpos;
6195 src = dtd->scaffold[src_node].name;
6196 for (;;) {
6197 *(*strpos)++ = *src;
6198 if (!*src)
6199 break;
6200 src++;
6201 }
6202 dest->numchildren = 0;
6203 dest->children = NULL;
6204 }
6205 else {
6206 unsigned int i;
6207 int cn;
6208 dest->numchildren = dtd->scaffold[src_node].childcnt;
6209 dest->children = *contpos;
6210 *contpos += dest->numchildren;
6211 for (i = 0, cn = dtd->scaffold[src_node].firstchild;
6212 i < dest->numchildren;
6213 i++, cn = dtd->scaffold[cn].nextsib) {
6214 build_node(parser, cn, &(dest->children[i]), contpos, strpos);
6215 }
6216 dest->name = NULL;
6217 }
6218 }
6219
6220 static XML_Content *
build_model(XML_Parser parser)6221 build_model (XML_Parser parser)
6222 {
6223 DTD * const dtd = _dtd; /* save one level of indirection */
6224 XML_Content *ret;
6225 XML_Content *cpos;
6226 XML_Char * str;
6227 int allocsize = (dtd->scaffCount * sizeof(XML_Content)
6228 + (dtd->contentStringLen * sizeof(XML_Char)));
6229
6230 ret = (XML_Content *)MALLOC(allocsize);
6231 if (!ret)
6232 return NULL;
6233
6234 str = (XML_Char *) (&ret[dtd->scaffCount]);
6235 cpos = &ret[1];
6236
6237 build_node(parser, 0, ret, &cpos, &str);
6238 return ret;
6239 }
6240
6241 static ELEMENT_TYPE *
getElementType(XML_Parser parser,const ENCODING * enc,const char * ptr,const char * end)6242 getElementType(XML_Parser parser,
6243 const ENCODING *enc,
6244 const char *ptr,
6245 const char *end)
6246 {
6247 DTD * const dtd = _dtd; /* save one level of indirection */
6248 const XML_Char *name = poolStoreString(&dtd->pool, enc, ptr, end);
6249 ELEMENT_TYPE *ret;
6250
6251 if (!name)
6252 return NULL;
6253 ret = (ELEMENT_TYPE *) lookup(&dtd->elementTypes, name, sizeof(ELEMENT_TYPE));
6254 if (!ret)
6255 return NULL;
6256 if (ret->name != name)
6257 poolDiscard(&dtd->pool);
6258 else {
6259 poolFinish(&dtd->pool);
6260 if (!setElementTypePrefix(parser, ret))
6261 return NULL;
6262 }
6263 return ret;
6264 }
6265